PU.1 Regulates TCR Expression by Modulating GATA-3 Activity

The Journal of Immunology

PU.1 Regulates TCR Expression by Modulating GATA-3 Activity1 Hua-Chen Chang,* Ling Han,* Rukhsana Jabeen,* Sebastian Carotta,‡ Stephen L. Nutt,‡ and Mark H. Kaplan2*† The Ets transcription factor PU.1 is a master regulator for the development of multiple lineages during hematopoiesis. The expression pattern of PU.1 is dynamically regulated during early T lineage development in the thymus. We previously revealed that PU.1 delineates heterogeneity of effector Th2 populations. In this study, we further define the function of PU.1 on the Th2 phenotype using mice that specifically lack PU.1 in T cells using an lck-Cre transgene with a conditional Sfpi1 allele (Sfpi1lckⴚ/ⴚ). Although deletion of PU.1 by the lck-Cre transgene does not affect T cell development, Sfpi1lckⴚ/ⴚ T cells have a lower activation threshold than wild-type T cells. When TCR engagement is limiting, Sfpi1lckⴚ/ⴚ T cells cultured in Th2 polarizing conditions secrete higher levels of Th2 cytokines and have greater cytokine homogeneity than wild-type cells. We show that PU.1 modulates the levels of TCR expression in CD4ⴙ T cells by regulating the DNA-binding activity of GATA-3 and limiting GATA-3 regulation of TCR gene expression. GATA-3-dependent regulation of TCR expression is also observed in Th1 and Th2 cells. In CD4ⴙ T cells, PU.1 expression segregates into subpopulations of cells that have lower levels of surface TCR, suggesting that PU.1 contributes to the heterogeneity of TCR expression. Thus, we have identified a mechanism whereby increased GATA-3 function in the absence of the antagonizing activity of PU.1 leads to increased TCR expression, a reduced activation threshold, and increased homogeneity in Th2 populations. The Journal of Immunology, 2009, 183: 4887– 4894.

T

helper cells have the potential to be programmed and differentiated into various subsets of Th cells that are distinguished by their secreted cytokines. The development of the Th2 subset is promoted by the IL-4-STAT6 signaling pathway and the induction of GATA-3 (1, 2). GATA-3 is a transcription factor expressed in T cells that is selectively induced during Th2 development, whereas expression is reduced in Th1 cells (3). GATA-3 functions both by directly binding to Th2 cytokine loci and chromatin remodeling of Th2 cytokine loci allowing other factors to bind more effectively (4, 5). The central role of GATA-3 has been demonstrated using cells and mice with a conditional deletion of GATA-3 (6, 7) and by ectopic expression of GATA-3, which induces Th2 cytokine production in Th1 cells and STAT6deficient T cells (5, 8). Interfering with GATA-3 function provides an additional mechanism of regulating the Th2 phenotype without altering GATA-3 expression levels. We have previously demonstrated that PU.1 interacts directly with GATA-3, which prevents the binding of GATA-3 to DNA and therefore antagonizes its function (9). PU.1 is an ETS family transcription factor that regulates various genes, including cytokine receptors required for the development of lymphoid and myeloid lineages (10). In Th2 cells, PU.1 is expressed

*Department of Pediatrics, Herman B Wells Center for Pediatric Research; †Department of Microbiology and Immunology, Indiana University School of Medicine, Indianapolis, IN 46202; and ‡The Walter and Eliza Hall Institute of Medical Research, Victoria, Australia Received for publication February 3, 2009. Accepted for publication August 10, 2009. 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 with 18 U.S.C. Section 1734 solely to indicate this fact. 1

This work was supported by U.S. Public Health Service Award AI057459 (to M.H.K.) from the National Institutes of Health.

2

Address correspondence and reprint requests to Dr. Mark H. Kaplan, Department of Pediatrics, Indiana University School of Medicine, HB Wells Center for Pediatric Research, 702 Barnhill Drive RI2600, Indianapolis, IN 46202. E-mail address: [email protected]

www.jimmunol.org/cgi/doi/10.4049/jimmunol.0900363

in subpopulations that express low levels of IL-4 and effectively regulates the heterogeneity of the populations of Th2 cells (9). Decreasing PU.1 expression using a short hairpin RNA (shRNA)3 increases the homogeneity of Th2 cytokine production (9). Although the cytokine milieu is a decisive component for T cell fate following T cell activation, additional factors contribute to the strength of the TCR signal and the degree of Th polarization (11). Among the factors that control the TCR signaling threshold are transcription factors such as Ikaros (12), costimulatory molecules such as B-7 family members and LFA-1 (13, 14), Lck activity (15), and microRNAs that target multiple phosphatases to modulate TCR signaling (16). The strength of T cell activation is also determined by the expression of TCR and the amount of engagement of the complex (14). The ability of specific factors to modulate the expression of TCR genes in this context has not been well studied. In this study, we found that T cells lacking PU.1 expression from mice with a conditional Sfpi1 allele had increased surface TCR expression on CD4⫹ T cells, resulting in increased T cell activation. Because GATA-3 was originally identified as a regulator for TCR␣ expression (17), we hypothesized that PU.1 modulates the function of GATA-3, which regulates the levels of TCR expression. This regulation may play a critical role in setting the T cell activation threshold in a heterogeneous population of T cells.

Materials and Methods Mice Wild-type C57BL/6 female mice were purchased from Harlan Bioscience. Mice with conditional deletion of the PU.1 gene (Sfpi1lck⫺/⫺) on the C57BL/6 background were previously described (18) and mated to mice

3

Abbreviations used in this paper: shRNA, short hairpin RNA; ChIP, chromatin immunoprecipitation assay; DAPA, DNA affinity precipitation assay; DN, double negative; E, enhancer; hCD4, human CD4; MFI, mean fluorescent intensity; siRNA, small interfering RNA. Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00

4888

PU.1 MODULATES GATA-3 ACTIVITY AND TCR EXPRESSION

carrying a Cre transgene under control of an Lck promoter (B6(CBA)Tg(Lck-cre)I540Jxm/J). Examination of allele deletion was performed as previously described (18). Mice were maintained in pathogen-free conditions and all studies were approved by the Indiana University School of Medicine Animal Care and Use Committee.

individually, using quantitative PCR in a SYBR Green Fast reaction (ABI 7500 Fast). The amount of PCR product was determined as percent of input relative to a standard curve of input chromatin. The ChIP results were calculated by subtracting the amount of DNA in the normal mouse IgG control from that in the anti-GATA-3 sample.

CD4⫹ T activation

Gata3 small interfering RNA (siRNA)

CD4⫹ T cells were purified from spleens and lymph nodes by positive selection using anti-CD4 magnetic beads (Miltenyi Biotech). CD4⫹ T cells (1 ⫻ 106 cells/ml complete RPMI 1640 medium) were cultured in a 6-well plate with coated anti-CD3 (clone 145–2C11) ranging from 4 to 0.06 ␮g/ml and with soluble anti-CD28 (clone 37.51) at 0.5 ␮g/ml for 2 days. The supernatant was collected for determining the levels of IL-2 production, and the cells were analyzed for expression of activation markers using flow cytometry.

Total CD4⫹ T cells (5 ⫻ 106) from wild-type and Sfpi1lck⫺/⫺ mice were transiently transfected with predesigned Gata3 siRNA (0.6 ␮M; Santa Cruz Biotechnology) and scrambled control (0.3 ␮M; Dharmacon) using the Amaxa Nucleofector (Lonza). After nucleofection, cells were rested in 5 ml complete medium in 5% CO2 incubator for 4 h. Cells (1 ⫻ 106/ml) were cultured with plate-bound anti-CD3 at 1 ␮g/ml and soluble anti-CD28 at 0.5 ␮g/ml for 2 days. IL-2 levels were evaluated from supernatants using ELISA. GATA-3 protein levels were determined using Western blot.

Th cell differentiation

Statistics

CD4⫹ T cells were purified from spleens and lymph nodes as described above. CD4⫹ T cells (1 ⫻ 106 cells/ml complete RPMI 1640 medium) were cultured in a 6-well plate with coated anti-CD3 (2 ␮g/ml) and soluble anti-CD28 (0.5 ␮g/ml) mAbs under Th1 (IL-12 at 5 ng/ml (PeproTech) and anti-IL-4 (11B11) mAb at 10 ␮g/ml or Th2 (IL-4 at 10 ng/ml (PeproTech) and anti-IFN-␥ (XMG) mAb at 10 ␮g/ml differentiation conditions. After 3 days of culture, 8 ml complete RPMI 1640 medium was added to each well. After 5– 6 days of culture, cell counts were determined using a hemacytometer. Differentiated cells were restimulated with plate-bound antiCD3 for 1 day and the cell-free supernatant was collected after centrifugation and stored at ⫺20°C until use. The levels of IL-2, IL-4, IL-5, IL-10, IL-13, and IFN-␥ produced were determined using ELISA analysis (9). Total protein extracts (100 ␮g) of Th2 cells differentiated from both C57BL/6 and Sfpi1lck⫺/⫺ mice were used for DNA affinity precipitation assay (DAPA) as described previously (9). Cells were lysed with cell lysis buffer (19) and total protein extracts were separated on SDS-PAGE gel followed by immunoblot analysis with Abs against GATA-3 (R&D Systems) and GAPDH (Cell Signaling).

Flow cytometry and cell sorting Cells from spleen and thymus of both C57BL/6 and Sfpi1lck⫺/⫺ mice were stained with anti-CD4 FITC and anti-CD8 PE Abs (BD Pharmingen). Double-negative (DN) thymocytes were sorted as CD4⫺CD8⫺ and divided into populations based on CD25 and CD44 expression. The expression of activation markers were analyzed by staining the cells with anti-CD25 FITC and anti-CD69 PerCP Abs (BD Pharmingen). TCR expression was measured by staining the cells with anti-TCR␤ FITC (clone H57– 597, BD Biosciences) and anti-CD3␧ PE (BD Biosciences). Cytokine intracellular staining was performed in differentiated Th2 cells that had been incubated in monensin for the last 2 hours of a 5-hour anti-CD3 stimulation (4 ␮g/ml). Stained cells were analyzed with the BD FACSCalibur flow cytometer instrument. Total CD4⫹ T cells, expressing high (top 15%) or low (bottom 15%) TCR␤ levels, were purified by flow sorting using the BD FACSVantage SE.

Retroviral transduction A bicistronic retroviral vector encoding mouse GATA-3 and human CD4 (hCD4) or hCD4 alone was described previously (9). Production of the retroviral supernatant followed by transduction into one-week-old differentiated Th1 cells was described previously (20). In transduced cells, expression of TCR␤ or CD3␧ was evaluated by flow cytometry on 5000 hCD4 positive cells. Transduction of differentiating Th2 cells with PU.1expressing retrovirus was performed as previously described (9).

Chromatin immunoprecipitation assay (ChIP) The ChIP experiment was performed using total CD4⫹ T cells or differentiated Th2 cells from wild-type and Sfpi1lck⫺/⫺ mice following the protocol as previously described (21). The resulting chromatin (2 ⫻ 106 cells per immunoprecipitation reaction) was immunoprecipitated using antiGATA-3 (HG3–31) AC agarose (Santa Cruz Biotechnology) or normal mouse IgG-AC agarose, followed by washing, de-cross-linking, and DNA purification. Purified DNA was resuspended in double-distilled H2O. The enriched DNA was analyzed for TCR enhancer (E) ␤ (forward primer: 5⬘-TGT AGG ACC TGG TAA ATG TCA AAC-3⬘ and reverse primer: 5⬘-GGA AGG GGT GGA AGC ATC TC-3⬘) (22), TCR E␣ (forward primer: 5-GCCAGAAGTAGAACAGGAAATGGA-3 and reverse primer: 5-GGGACCTGTTTGCCCATGT-3) (23), and Il4 Va (forward primer: 5-GAC TGA GAA CCC AAC AGA GAT GC-3 and reverse primer: 5-GCC TTG GTT CCA TAT ATT CAT GAC T-3) (24)

Data are presented as mean ⫾ SD. Statistical significance was evaluated with an independent Student’s t test using SPSS 16.0 program (SPSS), and p ⬍ 0.05 was considered significant.

Results

Normal development of CD4⫹ T lymphocytes from Sfpi1lck⫺/⫺ mice We have previously demonstrated the function of PU.1 in Th2 cells using shRNA in cultured cells (9). To further analyze the effects of PU.1 deficiency on T cell phenotypes, we used mice that carry conditional alleles of the Sfpi1 gene on the C57BL/6 background and crossed to lck-Cre transgenic mice (denoted as Sfpi1lck⫺/⫺) (18). The lck-Cre transgene mediates deletion of the floxed Cre allele beginning in the DN3 stage of thymocyte development (Fig. 1A). In naive CD4 T cells isolated from spleen the nondeleted allele was not detected (Fig. 1A). Importantly, deletion of the Sfpi1 allele is not complete until after Sfpi1 expression is extinguished from developing thymocytes (Fig. 1B) as previously described (18, 25). Thus, as expected, the T cell populations in Sfpi1lck⫺/⫺ mice as analyzed by flow cytometry were indistinguishable from wild-type mice. Flow cytometry of thymocyte populations showed similar profiles ( p ⬎ 0.05) of CD4⫹ and CD8⫹ T cells between wild-type and Sfpi1lck⫺/⫺ mice (Fig. 1). The proportions of mature CD4⫹ and CD8⫹ T cells in the spleens of the Sfpi1lck⫺/⫺ mice were comparable to spleens from wild-type mice ( p ⬎ 0.05; Fig. 1C). In addition, we did not observe any significant difference in the percentage of other populations examined, including T cells (CD3⫹), B cells (B220⫹), NK and NKT cells (CD3 ⫻ NK1.1), naive and memory CD4⫹ T cells (CD4 ⫻ CD62L ⫻ CD44), or natural T regulatory cells (CD4⫹ CD25⫹) (data not shown). The total numbers of thymocytes (11.5 ⫾ 4 ⫻ 107 in Sfpi1lck⫺/⫺ mice vs 13.3 ⫾ 6 ⫻ 107 in wild-type mice, n ⫽ 3) and splenocytes (8 ⫾ 6 ⫻ 107 in Sfpi1lck⫺/⫺ mice vs 4 ⫾ 2 ⫻ 107 in wild-type mice, n ⫽ 4) in Sfpi1lck⫺/⫺ mice did not differ ( p ⬎ 0.05) from those of wild-type controls. PU.1 regulates T cell signaling threshold and Th2 cytokine production We next wanted to examine the sensitivity of CD4⫹ T cells in response to different strengths of stimuli using increasing concentrations of plate-bound anti-CD3␧ with a constant concentration of soluble anti-CD28. CD4⫹ T cells from Sfpi1lck⫺/⫺ mice produced significantly higher levels of IL-2 ( p ⬍ 0.05) compared with cells from wild-type control mice, even at the lowest dose of anti-CD3 at 0.06 ␮g/ml following 48 h of stimulation (Fig. 2A). Similar results are observed using conditional PU.1-deficient T cells from mice expressing Cre from a CD4 promoter (data not shown), suggesting that increased IL-2 production is not due to an early thymic defect or a specific effect of the lck-Cre transgene. Although the


4889

FIGURE 1. T cell development in the absence of PU.1. A, Thymocytes or splenocytes were sorted into the indicated populations and genomic DNA was analyzed for the presence of the wild-type (WT; ⫹), floxed (fl), or deleted (⌬) allele. Results are representative of two experiments. B, Thymocytes from Sfpi1lck⫺/⫺ and wild-type control mice were sorted into DN subpopulations and RNA was analyzed for expression of Sfpi1. Results are averages ⫾ SD of two independent experiments. C, Flow cytometry analysis of cells in thymus and spleens from Sfpi1lck⫺/⫺ and wild-type control mice. The CD4⫹ or CD8⫹ T cell percentage is presented as mean ⫾ SD of six mice from spleens and three mice from thymus. CD4⫹ T cell development was normal in Sfpi1lck⫺/⫺ mice as shown by similar (p ⬎ 0.05) percentage of cells staining with anti-CD4 and/or antiCD8. DP, Double positive.

percentage of cells expressing CD25 was similar between PU.1deficient and wild-type cells 48 h after anti-CD3 stimulation, the expression of CD25 presented as mean fluorescent intensity (MFI) was significantly ( p ⬍ 0.05) higher in CD4⫹ T cells lacking PU.1 compared with those from wild-type cells (Fig. 2B and data not shown). Because PU.1 appeared to modulate the activation threshold of T cells and because we have previously shown PU.1 to regulate Th2 cytokine production, we next evaluated the phenotype of Th2 populations in the absence of PU.1. Given the ability of PU.1 deficiency to increase the sensitivity of T activation, we compared

FIGURE 2. CD4⫹ T cells from Sfpi1lck⫺/⫺ mice are more sensitive in response to stimuli. CD4⫹ T cells isolated from Sfpi1lck⫺/⫺ and wild-type (WT) control mice were stimulated with the indicated concentrations of plate-bound anti-CD3␧ and a constant concentration of soluble anti-CD28 (0.5 ␮g/ml) for 2 days. A, Supernatants were collected for determining the levels of IL-2 production using ELISA. B, The expression levels (MFI) of CD25 were evaluated using flow cytometry. Values presented are mean ⫾ SD of three mice from each group. ⴱ, Significant difference between wildtype control and Sfpi1lck⫺/⫺ samples (p ⬍ 0.05). Results shown are representative of two independent experiments.

the differentiation of wild-type and Sfpi1lck⫺/⫺ CD4⫹ T cells under optimal (2 ␮g/ml) or suboptimal (0.125 ␮g/ml) concentrations of anti-CD3 in combination with constant doses of anti-CD28, IL-4, and anti-IFN-␥ for Th2 differentiation. After 5– 6 days of culture, the numbers of Th2 cells from Sfpi1lck⫺/⫺ mice were almost 2-fold higher than those from wild-type mice following suboptimal stimulation (Fig. 3A). However, the cell numbers did not differ significantly ( p ⬎ 0.05) under optimal concentrations of anti-CD3 (Fig. 3A). The requirement for lower CD3 stimulation to observe significant differences between wild-type and Sfpi1lck⫺/⫺ T cells may be due to higher expression of Sfpi1 in cells cultured with lower doses of anti-CD3 (Fig. 3B). Th2 cells (1 ⫻ 106 cells/ ml) differentiated under either condition were restimulated with anti-CD3 (2 ␮g/ml) for 24 h. The levels of secreted cytokines in the supernatant were significantly ( p ⬍ 0.05) higher in Th2 cultures from Sfpi1lck⫺/⫺ mice compared with the wild-type control under the suboptimal concentrations of anti-CD3 (Fig. 3C). Although the levels of Th2 cytokine production under optimal anti-CD3 stimulation showed a similar trend of Th2 cytokines production, only IL-5 production was significantly different from wild-type cultures. The levels of IFN-␥ in Th2 cultures were similarly ( p ⬎ 0.05) low between Sfpi1lck⫺/⫺ and wild-type control cultures stimulated with either optimal or suboptimal anti-CD3 (Fig. 3C). As stated above, greater differences between wild-type and Sfpi1lck⫺/⫺ Th2 cells when cultured under the suboptimal conditions may be due to increased expression of PU.1 under these conditions (Fig. 3C). We have previously shown that reduction of PU.1 expression by shRNA increases the homogeneity of Th2 cytokine production, as evidenced by more cells simultaneously producing two Th2 cytokines (9). We predicted that Th2 cells differentiated from Sfpi1lck⫺/⫺ mice would be more homogenous in their cytokine production than wild-type cells. Indeed, intracellular cytokine staining revealed significantly ( p ⬍ 0.05) more IL-4/IL-5 and IL5/IL-10 double-positive-producing Th2 cells in the absence of PU.1 than those cells from wild-type control mice under the suboptimal anti-CD3 (0.125 ␮g/ml) stimulation (Fig. 4A). There were

4890


FIGURE 3. Cytokine production from wild-type (WT) and PU.1-deficient Th2 cultures. Total CD4⫹ T cells isolated from both Sfpi1lck⫺/⫺ and wild-type control mice were differentiated into Th2 cells in the presence of either 0.125 or 2 ␮g/ml anti-CD3␧. A, The numbers of differentiated cells were determined by hemocytometer with trypan blue staining. B, Sfpi1 mRNA was determined in wild-type Th2 cells differentiated with the anti-CD3 dose indicated using quantitative PCR. C, Differentiated Th2 cells were restimulated with plate-bound anti-CD3␧ (2 ␮g/ml) for 24 h. The supernatant was collected for cytokine ELISA. Data presented are mean ⫾ SD of three mice from each group. ⴱ, Significant difference between wild-type control and Sfpi1lck⫺/⫺ samples (p ⬍ 0.05). Results shown are representative from five independent experiments. NS, Not significant.

also significant increases ( p ⬍ 0.05) in IL-5-single positive populations (Fig. 4A). However, there were similar ( p ⬎ 0.05) frequencies of double-positive cytokine-producing cells when cultured under optimal anti-CD3 (2 ␮g/ml) stimulation conditions (Fig.

FIGURE 4. Intracellular staining of cytokine production in wild-type (WT) and PU.1-deficient Th2 cells. Th2 cells were differentiated with antiCD3 at 0.125 ␮g/ml (A) or 2 ␮g/ml (B), restimulated with plate-bound anti-CD3␧ (4 ␮g/ml), and treated with monensin for the last 2 h of a 5-h incubation. Intracellular cytokine production was determined by staining cells with fluorochromes-conjugated Abs including IL-4, IL-5, and IL-10. Results are presented as mean ⫾ SD of three mice from each group. ⴱ, Significant difference between wild-type control and Sfpi1lck⫺/⫺ samples (p ⬍ 0.05). Results shown are a representative from five independent experiments.

4B). In agreement with the ELISA results (Fig. 3B), IL-5 singlepositive populations were also significantly higher under optimal conditions (Fig. 4B). PU.1 interferes with GATA-3-dependent TCR expression Because CD4⫹ T cells are more sensitive to stimuli in the absence of PU.1 expression, we hypothesized that CD4⫹ T cells


4891

FIGURE 5. TCR expression on wild-type (WT) and PU.1-deficient CD4⫹ T cells. A–C, CD4⫹ splenocytes (A), CD4 single-positive thymocytes (B), or CD4⫹ splenocytes gated on naive (CD62Lhigh CD44low) or memory (CD62Llow CD44high) populations (C) isolated from Sfpi1lck⫺/⫺ and wild-type mice were evaluated for the surface expression of CD3␧ and TCR␤ as indicated. The expression levels are presented as mean MFI ⫾ SD of three mice from each group. D, CD4 T cells from isolated from Sfpi1lck⫺/⫺ and wild-type splenocytes were stimulated with PMA (10 or 50 ng/ml as indicated) and ionomycin (500 ng/ml) for 48 h before IL-2 concentration in supernatants was determined using ELISA. E, CD4⫹CD62L⫹ (naive) T cells were isolated using magnetic selection. Total protein extracts were subjected to Western blotting analysis for GATA-3 with GAPDH as a loading control. The Western blot band intensity was analyzed by densitometry and normalized to GAPDH as the ratio of GATA-3 and GAPDH. Data are presented as the mean ratio ⫾ SD of three mice from each group. NS, not significantly different (p ⬎ 0.05). F, Isolated total CD4⫹ T cells were pooled from three wild-type and three Sfpi1lck⫺/⫺ mice for ChIP assay to evaluate the binding of GATA-3 to the TCR E␤ region. The DNA amplified by each primer set using real-time PCR is presented as the mean percentage of input ⫾ SD of two duplicates. G, CD4 T cells isolated from Sfpi1lck⫺/⫺ and wild-type splenocytes were transfected with Gata3-specific siRNA and stimulated with anti-CD3 (1 ␮g/ml) for 48 h before determination of IL-2 concentrations in supernatants using ELISA. Inset, Immunoblot of GATA-3 and GAPDH in control (Ctrl) and Gata3 siRNA transfected cells. Numbers indicate the relative level of GATA-3 protein and are representative of two experiments. H, DAPA was performed with total protein extracts from differentiated Th2 cells from both wild-type and Sfpi1lck⫺/⫺ mice. Results shown are a representative from three independent experiments. I, ChIP assay for GATA-3 binding to the regions of Tcr E␤ and Tcr E␣, and to the Il4 Va regions (24) was conducted with chromatin from Th2 cultures differentiated from wild-type and Sfpi1lck⫺/⫺ mice. Data are average of two independent experiments. For all panels, ⴱ indicates a significant difference between wild-type control and Sfpi1lck⫺/⫺ samples (p ⬍ 0.05).

from Sfpi1lck⫺/⫺ mice had higher levels of TCR expression, which contributed to increased sensitivity to anti-CD3 stimulation. We examined the levels of TCR expression from freshly isolated splenocytes. Flow cytometry revealed that both CD3␧ and TCR␤ (clone H57–597) expression were significantly ( p ⬍ 0.05) higher in T cells isolated from Sfpi1lck⫺/⫺ mice than those from wild-type mice (Fig. 5A). We also observed that individual TCR V␤ regions (mouse V␤ TCR screening panel; BD Biosciences) on T cells from Sfpi1lck⫺/⫺ mice had higher expression levels than wild-type cells (data not shown). CD3 and TCR␤ expression were also higher on CD4 single-positive thymocytes from Sfpi1lck⫺/⫺ mice than wild-type cells and, in

spleen, CD3 and TCR␤ expression were higher on both naive and memory phenotype CD4 T cells (Fig. 5, B and C, and data not shown). As increased TCR expression might explain the increased sensitivity to Ag receptor stimulation (Fig. 2), we next tested whether IL-2 production is different when bypassing the TCR complex by stimulating wild-type and Sfpi1lck⫺/⫺ T cells with PMA and ionomycin. IL-2 production was similar between wild type and Sfpi1lck⫺/⫺ T cells stimulated with PMA and ionomycin, suggesting that the increased sensitivity of Sfpi1lck⫺/⫺ T cells to anti-CD3 stimulation is due to altered Ag receptor expression and not other alterations in downstream signaling factors (Fig. 5D).

4892


GATA-3 was originally identified as an important transcription factor for regulation of TCR␣ gene expression (17). In addition, PU.1 antagonizes GATA-3 function by direct interaction and prevents the binding of GATA-3 to target genes (9). The average level of GATA-3 protein expression in naive CD4 T cells from Sfpi1lck⫺/⫺ mice was not significantly ( p ⬎ 0.05) different from wild-type controls (Fig. 5E). However, the binding of GATA-3 to the TCR E␤ region in total CD4⫹ T cells, as evaluated by ChIP assay, was increased 95% in T cells from Sfpi1lck⫺/⫺ mice compared with wild-type cells (Fig. 5F). To determine whether GATA-3 was required for increased IL-2 production from Sfpi1lck⫺/⫺ T cells, activated wild-type and Sfpi1lck⫺/⫺ T cells were transfected with control or Gata3-specific siRNA. Diminishing GATA-3 expression decreased IL-2 production from wild-type cells and also reduced IL-2 production from Sfpi1lck⫺/⫺ T cells to amounts near control wild-type T cells (Fig. 5G). Thus, GATA-3 is required for the increased sensitivity of Sfpi1lck⫺/⫺ T cells to Ag receptor stimulation. Because we observed higher Th2 cytokine production in cultures differentiated from Sfpi1lck⫺/⫺ mice (Figs. 3 and 4), whereas the levels of GATA-3 protein were similarly expressed in differentiated wild-type and PU.1-deficient Th2 cultures (Fig. 5H), we hypothesized that the DNA binding activity of GATA-3 was enhanced in Th2 cells lacking PU.1. To determine whether there was an increase in total GATA-3 DNA binding activity, we tested GATA-3 DNA binding activity using DAPA. DAPA experiments revealed that 60% more GATA-3 was capable of binding DNA from PU.1-deficient extracts than from wild-type extracts (Fig. 5H). Similarly, increased binding activity of GATA-3 to TCR E␣ and E␤ regions was observed using ChIP assay with Th2 cells differentiated from Sfpi1lck⫺/⫺ mice compared with wild-type controls (Fig. 5I). As we previously observed in IL-4low cells (9), PU.1 expression limited GATA-3 binding to the Il4 Va enhancer (Fig. 5I). Thus, PU.1 limits GATA-3 DNA binding to Tcr and Il4 loci. Our results suggest that the increased DNA binding activity of GATA-3 in Sfpi1lck⫺/⫺ mice contributes to higher levels of TCR expression. This further suggests that Th2 cells, which have higher expression of GATA-3 than Th1 cells, might also express higher TCR levels. Following differentiation, Th2 cells express almost 20-fold higher levels of GATA-3 compared with Th1 cells, with a reciprocal pattern of T-bet expression (Fig. 6A). Flow cytometry analysis revealed that TCR expression was higher in differentiated Th2 cultures than in Th1 cells (Fig. 6B). In addition, levels of Tcra mRNA were higher in Th2 cells than Th1 cells (Fig. 6C). We hypothesized that altering the levels of GATA-3 activity would affect TCR expression. We first tested whether ectopic expression of GATA-3 in Th1 cells would increase TCR expression. After 6 days of differentiation, Th1 cells from wild-type mice were transduced with retroviruses encoding GATA-3 and human CD4 (hCD4) or hCD4 alone. We have previously shown that GATA-3 transduction into differentiated Th1 does not eliminate the Th1 phenotype (20). Three days after transduction, expression of TCR␤ and CD3␧ were evaluated on hCD4-positive cells. Flow cytometry showed that ectopic expression of GATA-3 in differentiated Th1 cells increased TCR expression (Fig. 6D). We then tested the converse; whether increased PU.1 expression would decrease TCR expression. Differentiating Th2 cells were transduced with control or PU.1-expressing retroviruses and sorted for enhanced GFP expression (9). Ectopic PU.1 expression decreased Tcra mRNA levels compared with control transduced cells (Fig. 6E). These results further supported that higher levels of GATA-3 activity contributed to increased TCR expression. T cells exhibit heterogeneous TCR expression as indicated by the distribution of TCR staining in flow cytometry (Fig. 7A). We

FIGURE 6. PU.1 and GATA-3 alter TCR expression in Th1 and Th2 cells. CD4⫹ T cells isolated from both Sfpi1lck⫺/⫺ and wild-type control mice were differentiated into Th1 and Th2 cells. A, Expression levels of Gata3 and Tbx21 were determined by real-time PCR and presented as mean relative gene expression ⫾ SD of two duplicates. B, TCR␤ surface expression on wild-type Th1 and Th2 cells was evaluated using FACS. C, Expression levels of Tcra in wild-type Th1 and Th2 cultures were determined by real-time PCR and are presented as mean relative gene expression ⫾ SD of two duplicates. D, After 5– 6 days of differentiation, Th1 cells from wild-type mice were transduced with retroviruses encoding GATA-3 and hCD4 or hCD4 alone. Three days after transduction, expression of TCR␤ and CD3␧ was evaluated on 5000 hCD4-positive cells by flow cytometry. E, Wild-type CD4 T cells differentiated under Th2 conditions were transduced with control or PU.1-expressing bicistronic EGFP retrovirus. EGFP-positive cells were sorted and expression of Tcra was examined using real-time PCR. MIEG, Empty vector control.

purified the cell populations expressing high (MFI ⫽ 100) and low (MFI ⫽ 25) levels of TCR␤ for further analysis. After flow sorting from wild-type control mice, TCR-low and TCR-high populations were stimulated with increasing doses of anti-CD3. Both TCR-low and TCR-high cells were activated upon stimulation because CD69 was expressed on both populations with a slightly higher intensity in TCR-high populations (Fig. 7B). However, TCR-high populations were more sensitive to increasing concentrations of anti-CD3 stimulation, producing much higher levels of IL-2 than similarly stimulated TCR-low cells (Fig. 7C). Although TCR-high and TCR-low populations expressed similar levels of GATA-3 (Fig. 7D), TCR-low CD4 T cell populations had almost 100-fold higher expression of PU.1 than TCR-high cells (Fig. 7E). Similarly, in Th2 cells, TCR-low cells had higher expression of Sfpi1 and lower expression of IL-4 (Fig. 7, E and F), consistent with our previous observations (9). Together, these data indicate that PU.1 interferes with GATA-3-dependent regulation of TCR expression.

Discussion We have previously demonstrated that PU.1 expression affects the heterogeneity of Th2 phenotypes by antagonizing GATA-3-DNA binding activity and therefore regulates the levels of Th2 cytokine expression (9). In this study, we further evaluate the function of


4893

FIGURE 7. Functional heterogeneity of TCR high and low populations. A, Cell populations expressing high (MFI ⫽ 100) and low (MFI ⫽ 25) levels of TCR␤ were collected by flow sorting from wild-type mice. B, TCR-low and TCR-high cells were activated with the indicated concentrations of plate-bound anti-CD3␧ and a constant concentration of soluble anti-CD28 (0.5 ␮g/ml) for 2 days before the activation marker CD69 was evaluated by flow cytometry. C, The production of IL-2 cytokine after 2 days of stimulation was evaluated using ELISA. Values are presented as the mean ⫾ SD of two independent experiments. D, Total protein extracts from TCR-high and TCR-low populations after sorting were subjected to Western blot analysis for GATA-3 protein levels with GAPDH as a loading control. E, Gene expression of Sfpi1 from TCR-low and TCR-high populations after sorting of total CD4 T or Th2 cells were analyzed by real-time PCR and are presented as mean relative gene expression ⫾ SD of two duplicates. Data are a representative of two independent experiments. F, Th2 cells, sorted as in E, were stimulated for 24 h with anti-CD3 before IL-4 concentration in supernatant was determined using ELISA. Data are representative of two independent experiments.

PU.1 in Th2 cells using mice with a conditional deletion of the PU.1 gene. Results indicate that PU.1 modulates the levels of TCR expression in CD4⫹ T cells by regulating the DNA-binding activity of GATA-3. In the absence of PU.1 in Sfpi1lck⫺/⫺ T cells, increased GATA-3 function resulted in increased TCR expression, lowering the activation threshold and making CD4⫹ T cells more sensitive to anti-CD3 engagement and activation. The increased activation threshold is dependent upon TCR ligation because IL-2 production was similar between wild-type and Sfpi1lck⫺/⫺ T cells stimulated with PMA/ionomycin. GATA-3 is required for this effect and Gata3-specific siRNA reduced IL-2 production from PU.1-deficient T cells. The development of hematopoietic cells is directed by the activity of lineage-specific transcription factors that are expressed at different concentrations during a particular window of time. During T cell development, the level of PU.1 expression is greatly reduced from the DN2 to the DN3 stages of pro-T development (Fig. 1B), which is required for T lineage commitment in the pre-T stage. Enforced expression of PU.1 in fetal hematopoietic progenitors arrests T cells at the pro-T stage (25). In Sfpi1lck⫺/⫺ mice, ablation of PU.1 expression begins at the DN3 stage (Refs. 26 and 27, and Fig. 1A), making deletion unlikely to interfere with normal T cell development. Deletion of the allele was highly efficient because we did not detect the nondeleted allele in naive CD4 T cells (Fig. 1A). As such, we did not observe any defects in T cell development in Sfpi1lck⫺/⫺ mice (Fig. 1). In addition, GATA-3 expression is similar in naive CD4⫹ T cells isolated from wild-type and Sfpi1lck⫺/⫺ mice (Fig. 5E), suggesting that lacking PU.1 expression from DN3 on does not perturb the balance of the transcriptional environment during the T cell development. Regulation of TCR expression by GATA-3 has been previously described. GATA-3 is required at multiple stages of T cell development, including the early DN stages, ␤-selection between DN3 and DN4, and positive selection of CD4 single-positive thymo-

cytes (28). Enforced GATA-3 expression in CD2-GATA-3 transgenic mice resulted in enhanced TCR up-regulation at the doublepositive and CD4 single-positive thymocytes (29, 30). In addition, GATA-3 deficiency in CD4-Cre Gata3fl/fl mice is associated with diminished expression of surface TCR in single-positive CD4 and CD8 thymocytes (6, 29). In this study, we identified PU.1 as a regulator GATA-3 function. PU.1 deficiency results in increased TCR expression and reduced activation thresholds in mature CD4⫹ T cells (Figs. 5 and 7). Moreover, the heterogeneity of TCR expression correlates with PU.1, in that PU.1 expression is higher in TCR-low T cells (Fig. 7). It is possible that PU.1 could also have direct effects on TCR expression. Other ETS family members can activate TCR gene expression when they bind directly to regulatory elements, though is it not clear whether these factors are required for TCR gene expression (31–33). Our data suggests that regulating GATA-3 activity may be an additional mechanism of controlling T cell activation. Transgenic models have demonstrated that Ag dose contributes to the polarization of T cells (11). Similarly, the regulation of TCR expression and activation thresholds by PU.1 contributes to the development of Th2 cells. In Th2 polarizing conditions where TCR engagement was limiting, the numbers of Th2 cells and the cytokine production from Th2 cells were increased in Sfpi1lck⫺/⫺ cultures compared with wild-type cultures under the same conditions. However, when TCR engagement was not limiting, the differences in Th2 cytokine production between wild-type and Sfpi1lck⫺/⫺ cultures were minimized. This is also reflected in Th2 homogeneity, defined as cells producing two or more cytokines simultaneously (9). In conditions of limiting TCR engagement, more Th2 cells are double-cytokine producers from Sfpi1lck⫺/⫺ cultures than from wild-type cultures (Fig. 4A). This suggests that lowering the activation threshold by deleting PU.1 expression increased Th2 cell homogeneity. The differences between wild-type and Sfpi1lck⫺/⫺ cells may be more dramatic in low-dose anti-CD3

4894


cultures because Sfpi1 expression is increased under these conditions, making the effects of deficiency more apparent. Thus, PU.1 may be regulating Th2 heterogeneity through several mechanisms. First, it regulates GATA-3 activity in naive CD4 T cells, setting a threshold of activation that ultimately impacts the differentiation phenotype. The proportion of naive CD4 T cells expressing PU.1 is still unclear. Our previous report demonstrated that PU.1 is expressed in naive cells, and expression is further induced by activation (9). Our data suggests that only a subpopulation of T cells expresses PU.1 and in those cells TCR expression is the lowest (Fig. 7). It is also possible that a gradient of PU.1 expression exists in all naive T cells and that in PU.1-deficient cells, TCR expression of the entire population shifts up, rather than changes of expression in only a subpopulation of cells. At the moment it is hard to distinguish between these possibilities. In the subpopulation of Th2 cells that express PU.1, GATA-3 DNA binding is limited and the activation of Th2 cytokine loci is decreased. Both of these mechanisms may be important in determining the potency of the developing Th2 response and the function of the Th2 cells that develop. T cells are activated upon engagement of the TCR-CD3 complex, which triggers the signaling events of several ITAM motifs on CD3 subunits (34). The number of expressed TCR on the surface determines the magnitude of T cell activation (14). The higher the TCR expression on the surface of T cells, the lower activation threshold they require for activation, and thus are more sensitive to stimulation. In this report, we have identified PU.1 as a regulator of TCR expression by modulating GATA-3 activity. CD4 T cells expressing PU.1 have a higher threshold for activation and a decreased propensity for Th2 development at low Ag dose. That differentiated Th2 cells express higher levels of surface TCR than Th1 cells (Fig. 6B) may contribute to the preferential activation of Th2 cells in response to stimuli that are present at low concentrations, such as allergens. Interventions that could modulate PU.1 expression could increase the activation threshold in allergic patients and might provide an option for controlling Th2 development and function.

Acknowledgments We thank Dr. A. Dent for critical review of this manuscript and the staff of the Indiana University Simon Cancer Center Flow Cytometry Facility.

Disclosures The authors have no financial conflict of interest.

References 1. 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. Immunity 4: 313–319. 2. Farrar, J. D., W. Ouyang, M. Lohning, M. Assenmacher, A. Radbruch, O. Kanagawa, and K. M. Murphy. 2001. An instructive component in T helper cell type 2 (Th2) development mediated by GATA-3. J. Exp. Med. 193: 643– 650. 3. 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 controls Th2-specific expression of the interleukin-5 gene. J. Biol. Chem. 272: 21597–21603. 4. Lee, G. R., P. E. Fields, and R. A. Flavell. 2001. Regulation of IL-4 gene expression by distal regulatory elements and GATA-3 at the chromatin level. Immunity 14: 447– 459. 5. Lee, H. J., N. Takemoto, H. Kurata, Y. Kamogawa, S. Miyatake, A. O’Garra, and N. Arai. 2000. GATA-3 induces T helper cell type 2 (Th2) cytokine expression and chromatin remodeling in committed Th1 cells. J. Exp. Med. 192: 105–115. 6. Pai, S. Y., M. L. Truitt, and I. C. Ho. 2004. GATA-3 deficiency abrogates the development and maintenance of T helper type 2 cells. Proc. Natl. Acad. Sci. USA 101: 1993–1998. 7. Zhu, J., B. Min, J. Hu-Li, C. J. Watson, A. Grinberg, Q. Wang, N. Killeen, J. F. Urban, Jr., L. Guo, and W. E. Paul. 2004. Conditional deletion of Gata3 shows its essential function in T(H)1-T(H)2 responses. Nat. Immunol. 5: 1157–1165. 8. Ouyang, W., M. Lohning, Z. Gao, M. Assenmacher, S. Ranganath, A. Radbruch, and K. M. Murphy. 2000. Stat6-independent GATA-3 autoactivation directs IL4-independent Th2 development and commitment. Immunity 12: 27–37.

9. Chang, H. C., S. Zhang, V. T. Thieu, R. B. Slee, H. A. Bruns, R. N. Laribee, M. J. Klemsz, and M. H. Kaplan. 2005. PU.1 expression delineates heterogeneity in primary Th2 cells. Immunity 22: 693–703. 10. Nerlov, C., and T. Graf. 1998. PU.1 induces myeloid lineage commitment in multipotent hematopoietic progenitors. Genes Dev. 12: 2403–2412. 11. Constant, S. L., and K. Bottomly. 1997. Induction of Th1 and Th2 CD4⫹ T cell responses: the alternative approaches. Annu. Rev. Immunol. 15: 297–322. 12. Avitahl, N., S. Winandy, C. Friedrich, B. Jones, Y. Ge, and K. Georgopoulos. 1999. Ikaros sets thresholds for T cell activation and regulates chromosome propagation. Immunity 10: 333–343. 13. Perez, O. D., D. Mitchell, G. C. Jager, S. South, C. Murriel, J. McBride, L. A. Herzenberg, S. Kinoshita, and G. P. Nolan. 2003. Leukocyte functional antigen 1 lowers T cell activation thresholds and signaling through cytohesin-1 and Jun-activating binding protein 1. Nat. Immunol. 4: 1083–1092. 14. Viola, A., and A. Lanzavecchia. 1996. T cell activation determined by T cell receptor number and tunable thresholds. Science 273: 104 –106. 15. Lovatt, M., A. Filby, V. Parravicini, G. Werlen, E. Palmer, and R. Zamoyska. 2006. Lck regulates the threshold of activation in primary T cells, while both Lck and Fyn contribute to the magnitude of the extracellular signal-related kinase response. Mol. Cell. Biol. 26: 8655– 8665. 16. Li, Q. J., J. Chau, P. J. Ebert, G. Sylvester, H. Min, G. Liu, R. Braich, M. Manoharan, J. Soutschek, P. Skare, L. O. Klein, M. M. Davis, and C. Z. Chen. 2007. miR-181a is an intrinsic modulator of T cell sensitivity and selection. Cell 129: 147–161. 17. Ho, I. C., P. Vorhees, N. Marin, B. K. Oakley, S. F. Tsai, S. H. Orkin, and J. M. Leiden. 1991. Human GATA-3: a lineage-restricted transcription factor that regulates the expression of the T cell receptor alpha gene. EMBO J. 10: 1187–1192. 18. Dakic, A., D. Metcalf, L. Di Rago, S. Mifsud, L. Wu, and S. L. Nutt. 2005. PU.1 regulates the commitment of adult hematopoietic progenitors and restricts granulopoiesis. J. Exp. Med. 201: 1487–1502. 19. Chang, H. C., S. Zhang, I. Oldham, L. Naeger, T. Hoey, and M. H. Kaplan. 2003. STAT4 requires the N-terminal domain for efficient phosphorylation. J. Biol. Chem. 278: 32471–32477. 20. Yu, Q., H. C. Chang, A. N. Ahyi, and M. H. Kaplan. 2008. Transcription factordependent chromatin remodeling of Il18r1 during Th1 and Th2 differentiation. J. Immunol. 181: 3346 –3352. 21. Thieu, V. T., Q. Yu, H. C. Chang, N. Yeh, E. T. Nguyen, S. Sehra, and M. H. Kaplan. 2008. Signal transducer and activator of transcription 4 is required for the transcription factor T-bet to promote T helper 1 cell-fate determination. Immunity 29: 679 – 690. 22. Mathieu, N., W. M. Hempel, S. Spicuglia, C. Verthuy, and P. Ferrier. 2000. Chromatin remodeling by the T cell receptor (TCR)-␤ gene enhancer during early T cell development: Implications for the control of TCR-␤ locus recombination. J. Exp. Med. 192: 625– 636. 23. Magdinier, F., T. M. Yusufzai, and G. Felsenfeld. 2004. Both CTCF-dependent and -independent insulators are found between the mouse T cell receptor ␣ and Dad1 genes. J. Biol. Chem. 279: 25381–25389. 24. Avni, O., D. Lee, F. Macian, S. J. Szabo, L. H. Glimcher, and A. Rao. 2002. T(H) cell differentiation is accompanied by dynamic changes in histone acetylation of cytokine genes. Nat. Immunol. 3: 643– 651. 25. Anderson, M. K., A. H. Weiss, G. Hernandez-Hoyos, C. J. Dionne, and E. V. Rothenberg. 2002. Constitutive expression of PU.1 in fetal hematopoietic progenitors blocks T cell development at the pro-T cell stage. Immunity 16: 285–296. 26. Levin, S. D., S. J. Anderson, K. A. Forbush, and R. M. Perlmutter. 1993. A dominant-negative transgene defines a role for p56lck in thymopoiesis. EMBO J. 12: 1671–1680. 27. Molina, T. J., K. Kishihara, D. P. Siderovski, W. van Ewijk, A. Narendran, E. Timms, A. Wakeham, C. J. Paige, K. U. Hartmann, A. Veillette, et al. 1992. Profound block in thymocyte development in mice lacking p56lck. Nature 357: 161–164. 28. Ting, C. N., M. C. Olson, K. P. Barton, and J. M. Leiden. 1996. Transcription factor GATA-3 is required for development of the T-cell lineage. Nature 384: 474 – 478. 29. Ling, K. W., J. P. van Hamburg, M. J. de Bruijn, D. Kurek, G. M. Dingjan, and R. W. Hendriks. 2007. GATA3 controls the expression of CD5 and the T cell receptor during CD4 T cell lineage development. Eur. J. Immunol. 37: 1043–1052. 30. Nawijn, M. C., R. Ferreira, G. M. Dingjan, O. Kahre, D. Drabek, A. Karis, F. Grosveld, and R. W. Hendriks. 2001. Enforced expression of GATA-3 during T cell development inhibits maturation of CD8 single-positive cells and induces thymic lymphoma in transgenic mice. J. Immunol. 167: 715–723. 31. Ho, I. C., N. K. Bhat, L. R. Gottschalk, T. Lindsten, C. B. Thompson, T. S. Papas, and J. M. Leiden. 1990. Sequence-specific binding of human Ets-1 to the T cell receptor ␣ gene enhancer. Science 250: 814 – 818. 32. Muthusamy, N., K. Barton, and J. M. Leiden. 1995. Defective activation and survival of T cells lacking the Ets-1 transcription factor. Nature 377: 639 – 642. 33. Wang, C. Y., B. Petryniak, I. C. Ho, C. B. Thompson, and J. M. Leiden. 1992. Evolutionarily conserved Ets family members display distinct DNA binding specificities. J. Exp. Med. 175: 1391–1399. 34. Holst, J., H. Wang, K. D. Eder, C. J. Workman, K. L. Boyd, Z. Baquet, H. Singh, K. Forbes, A. Chruscinski, R. Smeyne, et al. 2008. Scalable signaling mediated by T cell antigen receptor-CD3 ITAMs ensures effective negative selection and prevents autoimmunity. Nat. Immunol. 9: 658 – 666.