ICOS-Mediated Costimulation on Th2 Differentiation Is Achieved by the Enhancement of IL-4 Receptor-Mediated Signaling
http://www.100md.com
免疫学杂志 2005年第4期
Abstract
ICOS is the third member of the CD28 family molecules and plays a critical role in many T cell-dependent immune responses. Although accumulated data suggest that ICOS costimulatory signals play an important role in Th2-mediated immune responses, the molecular basis for this selective differentiation mechanism is largely unknown. To clarify this mechanism, we used DO11.10 TCR transgenic ICOS–/– mice and evaluated the nature of ICOS costimulatory signals during the process of Ag-specific activation and differentiation of naive CD4+ T cells. Results obtained from these experiments demonstrated that Ag stimulation of naive CD4+ T cells in the absence of an ICOS signal resulted in impaired Th2 development. Unlike previous reports, we found that primary IL-4 production by these T cells was intact and that IL-4R sensitivity of these T cells was reduced as evidenced by a profound defect in IL-4-induced Stat6 phosphorylation and the early induction of GATA-3. The fact that ICOS ligation of wild-type T cells significantly enhanced IL-4-induced Stat6 phosphorylation and primary GATA-3 induction, but not IL-4 transcription, of naive CD4+ T cells was consistent with the results obtained from ICOS–/– T cell experiments. These observations led us to propose that the predominant effect of ICOS-mediated costimulation on Th2 differentiation is achieved by the enhancement of IL-4R-mediated signaling.
Introduction
Helper T cell differentiation is controlled by several signaling pathways such as cytokine signals, TCR signals, and costimulatory signals (1). The IL-4/IL-4R/Stat6 pathway is critical for Th2 differentiation of Ag-stimulated naive CD4+ T cells by enhancing the expression of GATA-3, a master regulator of Th2 differentiation (2, 3, 4, 5). It has been reported that NFAT, MAPK, and/or NF-B pathways, which are downstream of TCR and costimulatory signals, are involved in modulating Th2 differentiation by affecting IL-4 transcription, IL-4R sensitivity, and/or GATA-3 expression (6, 7, 8, 9).
Costimulatory signals, in addition to TCR-mediated signals, are necessary for full activation and play an important role in the regulation of T cell responses (10, 11). The CD28 family molecules, CD28 and its homolog CTLA-4, are the most extensively characterized costimulatory molecules that positively or negatively regulate T cell responses, respectively (10, 11). More recently, the third member of CD28 family, ICOS, was identified (12, 13, 14). ICOS expressed at very low levels on naive T cells, is up-regulated after T cell activation, and is retained on effector/memory T cells. In particular, ICOS expression is skewed in Th cell subsets, namely, higher in Th2 than in Th1 (15). ICOS ligation has been shown to enhance TCR-mediated T cell proliferation, indicating that ICOS provides CD28-like positive costimulatory signals to T cells. In fact, like CD28, ICOS-mediated signals delivered by stimulation with anti-ICOS mAb or an Ig fusion protein of ICOS ligand, B7h (B7h-Ig), can enhance secretion of IL-4, IL-5, IL-10, IFN-, and TNF-. However, in contrast to CD28, ICOS does not costimulate IL-2 secretion (12, 13).
The important role of ICOS in various immune responses has been well documented (11). In particular, much attention has been drawn to the role of ICOS in Th cell differentiation. A number of studies demonstrated that blocking of ICOS-mediated costimulation with Abs, or with an ICOS-Ig fusion protein, resulted in the preferential inhibition of Th2-mediated immune responses (15, 16, 17, 18), although some Th1 responses were found to be affected (19, 20). Several ICOS-deficient mouse lines have been independently generated and examined for their immune function (21, 22, 23). It has been shown that these mice have defects in germinal center formation and Ab class switching (22, 23), and showed exacerbation of experimental autoimmune encephalomyelitis (21). Furthermore, ICOS- and B7h-deficient mice have also been shown to have a deficiency in in vivo generation of Ag-specific Th2 development (23, 24). These findings strongly suggested that ICOS plays a predominant role in Th2 differentiation, and/or effector function. However, the molecular mechanisms by which ICOS costimulatory signals enhance Th2 differentiation remain undefined.
In this study, we focused on the role of the ICOS costimulatory signal on Th2 differentiation of naive CD4+ T cells. We found that ICOS ligation by Ab cross-linking selectively enhanced the generation of IL-4-producing CD4+ T cells. Interestingly, it was found that ICOS ligation did not costimulate IL-4 transcription of T cells in the early phase of T cell activation. Instead, phosphorylation of Stat6 and expression of GATA-3 were dramatically enhanced, suggesting that the ICOS signal enhances IL-4R signaling. To evaluate the role of ICOS costimulatory signals during Ag-specific activation and differentiation of naive CD4+ T cells under physiological conditions, we examined the response to OVA peptide of naive T cells from ICOS–/– mice that have DO11.10 TCR transgenic (Tg)3 BALB/c background genes. We found that primary IL-4 production by these ICOS–/– T cells was intact and equivalent to ICOS+/+ T cells; however, they showed reduced IL-4R sensitivity as evidenced by a profound defect in IL-4-induced Stat6 phosphorylation and the early induction of GATA-3. These data provide evidence that a predominant effect of ICOS-mediated costimulation on Th2 differentiation may be achieved by the enhancement of IL-4R-mediated signaling.
Materials and Methods
Mice
ICOS–/– mice were generated as previously described (25) and provided by JT Pharmaceutical Frontier Research Laboratory. ICOS–/– mice were backcrossed onto BALB/c or C57BL/6 mice over six to eight times. ICOS–/– DO11.10 TCR Tg mice were obtained by intercrossing BALB/c background ICOS–/– mice and BALB/c DO11.10 TCR Tg mice, which were originally distributed by Dr. D. Loh (Roche, Nutley, NJ). Wild-type BALB/c and C57BL/6 mice were obtained from Sankyo Laboratory. Mice were maintained in our animal facility in specific pathogen-free conditions. For entire experiments, 8- to 10-wk-old mice were used. The experiment herein were conducted according to the principles set forth in Ref.26 .
Abs and reagents
Anti-mouse-ICOS mAb (B10.5) was produced as previously described (16) and provided by JT Pharmaceutical Frontier Research Laboratory. Purified anti-CD3 (2C11), anti-TCR (H57), anti-CD28 (PV-1), and rat anti-NP-IgG2a Abs, and nonpurified anti-CD8 Ab (53-6-7) were prepared in our laboratory from hybridoma culture medium. FITC-labeled anti-CD44 and IFN- Abs, PE-labeled anti-IL-4R and IL-4 Abs, allophycocyanin-labeled anti-CD4 Ab, biotin-labeled anti-IL-4, IL-5, and IFN- Abs, and purified anti-IL-4, IL-5, IFN-, and CD132 (c) Abs were purchased from BD Pharmingen. Biotin-labeled anti-IL-13, purified anti-IL-13, and anti-Stat6 Abs were purchased from R&D Systems. Anti-GATA-3, T-bet, NFATc1, NF-B1, c-Rel, RelA, and -actin Abs were purchased from Santa Cruz. Anti-phospho-Stat-6 Ab was purchased from Cell Signaling.
Preparation of Ab-conjugated beads
Latex beads were purchased from Interfacial Dynamics Corporation. Briefly, Latex beads (5 x 107) were washed twice with PBS and were incubated with anti-CD3 mAb (2C11; 1 μg/ml) plus one of following Abs: anti-CD28 (PV-1; 9 μg/ml), anti-ICOS (B10.5; 9 μg/ml), or anti-NP-IgG2a (isotype control; 9 μg/ml) for 1.5 h at 37°C. Beads were then washed with PBS, and blocked with 10% FCS/RPMI 1640 medium.
Cell purification and stimulation
Naive (CD44low) CD4+ T cells were purified by cell sorter. Briefly, to enrich bulk CD4+ T cells, single-cell suspensions of mice splenocytes were treated with anti-CD8 (53-6-7), and incubated on anti-mouse Ig-coated dish to eliminate CD8+ T cells and B cells. The cells were recovered and stained with anti-CD44-FITC and anti-CD4-allophycocyanin, and CD4+CD44low cell population was collected with FACSVantage (BD Pharmingen) as naive CD4+ T cells. Purity of sorted population was constantly >95%. Purified DO11.10 CD4+CD44low T cells (1 x 105) were stimulated with OVA323–339 peptide (0.1 μM) in the presence of mitomycin C-treated T-depleted normal BALB/c spleen cells (3 x 105) in 96-well plate. Naive CD4+ T cells (1 x 105) purified from normal mice were stimulated with Ab-conjugated beads (1 x 105) in 96-well culture plate. For cell culture, RPMI 1640 medium (Sigma-Aldrich) containing 10% FCS, 2ME, L-glutamine, penicillin/streptomycin, and 0.1 mM HEPES was used.
ELISA
Primary Ab was coated on 96-well plate at 37°C for 1 h, and the plate was blocked with 1% BSA in PBS. Culture supernatants was then applied on the plate and incubated for 2 h at 37°C. The plate was washed and reacted with biotinylated secondary Ab for 1 h, followed by reacting with HRP-coupled streptavidin (Sigma-Aldrich) for 30 min. Substrate (ABTS plus H2O2) was applied and OD405 was measured by Microplate Reader model 3550 (Bio-Rad).
Flow cytometry
Briefly, cells were washed with wash buffer (PBS containing 1% BSA and 0.05% NaN3), treated with anti-FcR (2.4G2), stained with anti-IL-4R-PE or anti-CD132 (c) followed by anti-rat-PE Abs. Electrical data was acquired on a FACSCalibur and analyzed by CellQuest software (BD Pharmingen).
Intracellular cytokine staining
Cultured Th cells were recovered, washed, and stimulated with plate-bound anti-TCR mAb (H57) for 6 h in the presence of monensin (2 μM). The cells were washed, fixed with 4% paraformaldehyde for 10 min, and treated with permeabilization buffer (50 mM NaCl, 5 mM EDTA, 0.02% NaN3, 0.5% Triton X) for 10 min. IL-4 and IFN- were detected with anti-IL-4-PE and anti-IFN--FITC Abs by flow cytometry.
Western blot analysis
To prepare total cell lysate, naive or activated CD4+ T cells were treated with lysis buffer (500 mM NaCl, 5 mM EDTA (pH 8.0), 100 mM Tris-HCl (pH 8.0), 1.25% Nonidet P-40) for 30 min on ice, and centrifuged at 15,000 rpm for 10 min. The supernatants were used for experiments. To prepare nuclear protein, cells were washed with PBS, and suspended in lysis buffer (10 mM HEPES (pH7.5), 10 mM KCl, 0.1 mM EDTA, 0.1% Nonidet P-40), vortexed vigorously, and centrifuged at 5,000 rpm for 2 min. Pellet were then suspended in extraction buffer (20 mM HEPES (pH 7.9), 0.4 M NaCl, 1 mM EDTA), incubated for 30 min on ice, and centrifuged at 15,000 rpm for 10 min. The supernatants were used as nuclear protein extract. Protein concentration of each sample was determined by BCA protein assay reagent (Pierce). Same amount of protein samples was resolved by SDS-PAGE with 12% acrylamide gel, transferred to polyvinylidene difluoride membrane, probed with specific primary Ab, followed by reacting with HRP-conjugated secondary Ab, and developed by ECL detection system (PerkinElmer).
RT-PCR
Total RNA was extracted using TRIzol reagent (Sigma-Aldrich), and mRNA was reverse-transcribed to cDNA using oligo(dT) primer and SuperScript II (Invitrogen Life Technologies). The following primer sets were used for RT-PCR analysis: IL-2 (forward (F): GAGTCAAATCCAGAACATGCC; reverse (R): TCCACTTCAAGCTCTACAG); IL-4 (F: GAATGTACCAGGAGCCATATC; R: CTCAGTACTACGAGTAATCCA); IFN- (F: AACGCTACACACTGCATCTTGG; R: GACTTCAAAGAGTCTGAGG); IL-10 (F: CGGGAAGACAATAACTG; R: CATTTCCGATAAGGCTTGG); GATA-3 (F: GCAGAACCGGCCCCTTATCAA; R: CATGGGCGTCGGTGTGGTCAG); SOCS-1 (F: CTTAACCCGGTACTCCGTGA; R: GAGGTCTCCAGCCAGAAGTG); SOCS-3 (F: AGCAAGTTTCCCGCCGCC; R: GCAGCGAAAAGCTGCCCC); SOCS-5 (F: CGGCCTTCAGAGGCGAGA; R: AGTGTGGGCTCACAGGGG); CIS (F: CCCGGCAGCACCTACAGA; R: CACGGGGATAGGCAGCAC); HPRT (F: GTTGGATACAGGCCAGACTTTGTTG; R: GATTCAACTTGCGCTCATCTTAGG).
Results
ICOS signaling does not influence IL-4 transcription of naive CD4+ T cells, but enhances IL-4R signaling in Th2 differentiation
The effect of ICOS ligation on Th cell differentiation was evaluated by stimulation of naive CD4+ T cells with beads containing an anti-ICOS mAb in the presence of suboptimal anti-CD3 and anti-CD28 mAb. As shown in Fig. 1A, anti-CD3/CD28/ICOS stimulation significantly enhanced the generation of IL-4-producing T cells compared with anti-CD3/CD28 stimulation in both BALB/c- and C57BL/6-derived naive CD4+ T cells. These results indicated that the ICOS-mediated costimulatory signal preferentially acts on Th2 differentiation.
FIGURE 1. ICOS cross-linking enhances Th2 differentiation by increasing IL-4R sensitivity. A, ICOS cross-linking by anti-ICOS mAb enhances Th2 differentiation. Purified CD44low naive CD4+ T cells from BALB/c or C57BL/6 mice were stimulated with the indicated Ab-conjugated beads. After 6-day stimulation, Th differentiation was analyzed by intracellular cytokine staining. B, ICOS cross-linking does not affect primary IL-4 transcription. Purified CD44low naive CD4+ T cells from BALB/c mice were stimulated as indicated for 24–48 h. IL-4 mRNA levels were determined by RT-PCR. C, ICOS cross-linking enhances IL-4-induced Stat6 phosphorylation. Purified CD44low naive CD4+ T cells from BALB/c mice were stimulated for 36 h in the presence of anti-IL-4 and anti-IL-12 Abs, washed, and then restimulated with exogenous IL-4 (1.0 ng/ml) for 20 min. Total cell lysates were prepared, and Stat6 phosphorylation was determined by Western blotting. D, ICOS cross-linking enhances GATA-3 expression. Purified CD44low naive CD4+ T cells from BALB/c mice were stimulated as indicated for 48 h. Total cell lysates were subjected to Western blot analysis for GATA-3 expression. A–D, Data show representative results from one of three independent experiments.
This preferential Th2 differentiation by ICOS ligation could be achieved by two distinct mechanisms. First, the ICOS-mediated costimulatory signal could augment the initial production of IL-4 by T cells, and second, it could enhance IL-4R sensitivity, both of which would accelerate IL-4R signaling. To evaluate the first possibility, IL-4 transcription of purified naive CD4+ T cells was examined at 24–48 h after coculture with beads bound by Abs specific for CD3, CD28, and/or ICOS. As shown in Fig. 1B, anti-CD3/CD28/ICOS cross-linking had no apparent effect on IL-4 transcription of CD4+ T cells compared with anti-CD3/CD28 stimulation, indicating that the ICOS signal had no costimulatory effect on the initial expression of IL-4 by T cells. To evaluate the second possibility, we tested IL-4R signaling after stimulation with ICOS by analyzing the signaling pathway downstream of the IL-4R. It is well known that Stat6 plays a critical role in the IL-4-mediated signal, and that its phosphorylation reflects the strength of IL-4R signaling (2, 3, 27). Thus, the effect of ICOS cross-linking on the capability of IL-4 to induce Stat6 phosphorylation was tested. Naive CD4+ T cells were stimulated with Ab-conjugated beads for 36 h and restimulated with rIL-4 for 20 min, and their Stat6 phosphorylation level was determined by Western blot analysis. As shown in Fig. 1C, although CD28 ligation alone induced a weak IL-4-dependent Stat6 phosphorylation signal, the additional ligation of ICOS dramatically enhanced this effect. These results suggest that ICOS ligation enhances IL-4R signaling.
After stimulation by IL-4, the phosphorylated Stat6 dimerizes and translocates to the nucleus where it activates a number of IL-4 target genes, including GATA-3. To confirm the enhancing effect of ICOS ligation on the IL-4R signal, we examined GATA-3 production after stimulation with Ab-conjugated beads (Fig. 1D). In the presence of suboptimal CD3 and CD28 stimulation, ICOS cross-linking dramatically enhanced GATA-3 production. Although we constantly observed weak GATA-3 induction by CD3 plus ICOS stimulation, this phenomenon was not likely to be a direct effect of ICOS signaling because anti-IL-4 mAb could abrogate GATA-3 induction in this experimental condition (data not shown).
The enhancement of downstream signaling of the IL-4R and no observable effects on IL-4 transcription are consistent with the idea that the ICOS-mediated Th2 differentiation of naive CD4+ T cells is caused by the enhancement of IL-4R sensitivity but not by IL-4 production.
Ag stimulation in the absence of ICOS results in impaired Th2 differentiation
To characterize the physiological role of ICOS in in vivo immune responses, we used ICOS–/– mice (25). Several groups have reported that ICOS deficiency resulted in attenuation of germinal center formation and Ab class switching (22, 23). Consistent with their reports, our ICOS–/– mice showed a defect in IgG1 Ab production and germinal center formation in lymph nodes and spleens against the T-dependent Ags (data not shown). These results are consistent with the notion that ICOS–/– mice have a defect in the differentiation and effector function of Th2 immune responses.
To evaluate the molecular mechanisms of preferential Th2 differentiation by ICOS under physiological conditions, we crossed ICOS–/– mice with DO11.10 TCR Tg mice and examined the responses of their naive CD4+ T cells to OVA peptide presented by APC.
Highly purified naive CD4+ T cells (CD44low) from DO11.10 ICOS–/– or ICOS+/+ mice were stimulated with OVA peptide in the presence of mitomycin C-treated normal BALB/c T-depleted spleen cells. At 48 h after Ag stimulation, cultured cells were harvested, washed, and recultured for additional 5 days in the presence of rIL-2 to generate Th cells. Live cells were restimulated with anti-TCR Ab, and their cytokine contents were analyzed by flow cytometry. As shown in Fig. 2A, the IL-4-producing T cell population in cultured ICOS–/– T cells was significantly smaller than that in ICOS+/+ T cells. In repeated experiments, the IL-4-producing population in ICOS–/– T cells was consistently reduced by 30–50% compared with ICOS+/+ T cells, indicating that ICOS–/– CD4+ T cells show impaired Th2 development after stimulation with Ag.
FIGURE 2. Impaired Th2 differentiation in Ag-stimulated ICOS–/– naive CD4+ T cells. DO11.10 naive CD4+ T cells from ICOS+/+ or ICOS–/– mice were stimulated with OVA peptide (0.1 μM) for 48 h. Cells were then recovered, washed, and recultured with rIL-2 (30 U/ml) for an additional 5 days. A, Th cells were restimulated with plate-bound anti-TCR for 6 h in the presence of monensin, and Th differentiation pattern was analyzed by intracellular cytokine staining. Data show representative one of three independent experiments. B, Th cells were stimulated with plate-bound anti-TCR for 24 h. IL-4, IL-5, IL-13, and IFN- concentrations in the culture supernatants were determined by ELISA. C, Th cells were restimulated as described in B, cell lysates were prepared, and protein expression levels of GATA-3, c-Maf, and T-bet were analyzed by Western blotting. B and C, Data show representative results from one of two independent experiments.
Next, we studied cytokine production by these T cells. By measuring cytokines in culture supernatants, we found that ICOS–/– T cells produce significantly less Th2 cytokines, IL-4, IL-5, and IL-13, than ICOS+/+ T cells, whereas IFN- production was not significantly different between ICOS+/+ and ICOS–/– T cells (Fig. 2B). Finally, we tested the production of transcription factors that are specific for Th subsets. As shown in Fig. 2C, the production of Th2-specific transcription factors, GATA-3 (4) and c-Maf (28), were reduced in ICOS–/– T cells, but the production level of Th1-specific transcription factor T-bet (29) was slightly higher in ICOS–/– T cells. These results confirmed the conclusion that ICOS–/– T cells have a defect in Th2 differentiation by Ag stimulation.
ICOS deficiency does not affect primary IL-4 production of naive CD4+ T cells
To test whether the ICOS-mediated Th2 differentiation of naive CD4+ T cells is the result of the enhancement of IL-4R sensitivity, but not IL-4 production under physiological conditions with Ag stimulation, we studied ICOS–/– DO11.10 TCR Tg T cells for IL-4 production and IL-4R sensitivity.
Highly purified DO11.10 naive CD4+ T cells from ICOS+/+ or ICOS–/– mice were stimulated with OVA peptide, and their IL-4 and IFN- production for the first 48 h after Ag stimulation was determined by ELISA. Although in each experiment the amount of IL-4 detected in the culture medium varied, there was no significant difference in IL-4 and IFN- production between ICOS+/+ and ICOS–/– T cells (Fig. 3A). We also tested IL-4 production from these cultures at different time points and found that IL-4 production by ICOS–/– T cells was not different from that of ICOS+/+ cells (Fig. 3B). Furthermore, cytokine gene transcription by these two T cell populations at 24 h after stimulation did not show differences when evaluated by semiquantitative RT-PCR (Fig. 3C) or quantitative real-time PCR (data not shown), indicating that cytokine transcription, including IL-4, by Ag-stimulated ICOS–/– T cells is normal. Similar results were also observed when non-TCR Tg T cells, isolated from BALB/c or C57BL/6 background ICOS–/– mice, were stimulated with anti-CD3 plus APC (data not shown). These results collectively indicated that impaired Th2 differentiation in ICOS–/– CD4+ T cells is not due to a defect in primary IL-4 production after Ag stimulation.
FIGURE 3. Primary IL-4 production is normal in Ag-stimulated ICOS–/– naive CD4+ T cells. A, DO11.10 naive CD4+ T cells from ICOS+/+ or ICOS–/– mice were stimulated with OVA peptide (0.1 μM) for 48 h. IL-4 and IFN- concentrations in culture supernatants were determined by ELISA. Data show results of five independent experiments. B, DO11.10 naive CD4+ T cells were stimulated as described in A. IL-4 concentrations of culture supernatants at the indicated time points were determined by ELISA. Data show representative one from two independent experiments. C, DO11.10 naive CD4+ T cells were stimulated as described in A. At 24 h after Ag stimulation, mRNA was extracted from isolated T cells, and the converted cDNAs were subjected to semiquantitative RT-PCR for the indicated cytokine gene products. Data show representative results from one of two independent experiments.
Next, we evaluated whether exogenous IL-4 could compensate for this impairment of Th2 differentiation in ICOS–/– CD4+ T cells. Highly purified DO11.10 naive CD4+ T cells from ICOS–/– or ICOS+/+ mice were stimulated with OVA peptide in the presence or absence of rIL-4 (10 ng/ml) for 48 h. After extensive washes and 5 additional days of cultivation in the presence of rIL-2 (30 U/ml), live cells were tested for IL-4 and IFN- by intercellular cytokine staining (Fig. 4A). Under these conditions, IL-4Rs on ICOS+/+ DO11.10 T cells might be presumably fully saturated with endogenous IL-4, indicating that exogenous IL-4 did not affect the frequency of IL-4-producing cells. Consistent with previous experiments, ICOS–/– DO11.10 T cells showed reduced numbers of IL-4-producing T cells, suggesting that this defective Th2 differentiation could not be compensated by the addition of exogenous IL-4.
FIGURE 4. Exogenous IL-4 is insufficient to compensate ICOS deficiency for Th2 development. A, DO11.10 naive CD4+ T cells from ICOS+/+ or ICOS–/– mice were stimulated with OVA peptide (0.1 μM) in the presence of rIL-4 (10 ng/ml). At 48 h after Ag stimulation, cells were washed and recultured in exogenous IL-2 (30 U/ml) for an additional 5 days. At day 7, Th differentiation was analyzed by intracellular cytokine staining. B, DO11.10 naive CD4+ T cells were isolated and stimulated as described in A with the exception that rIL-2 was not added to the reculture medium. Data show representative one of two independent experiments. C, DO11.10 naive CD4+ T cells from ICOS+/+ or ICOS–/– mice were stimulated with OVA peptide (0.1 μM) for 48 h. IL-4R and c expression was analyzed by flow cytometry. Data show representative results from one of three independent experiments.
For further analysis of IL-4R signaling in ICOS–/– CD4+ T cells, naive CD4+ T cells were stimulated with OVA peptide in the presence of serial dilutions of exogenous IL-4 for 48 h, washed, and cultured for a further 5 days without rIL-2. At day 7, Th differentiation was evaluated by intracellular cytokine staining of IL-4 and IFN-. Under these culture conditions, ICOS+/+ DO.11.10 CD4+ T cells did not differentiate into Th2 cells without additional IL-4. As shown in Fig. 4B, addition of exogenous IL-4 resulted in the generation of IL-4-producing ICOS+/+ T cells in a dose-dependent manner. In contrast, the effect of exogenous IL-4 on Th2 differentiation of ICOS–/– T cells was weaker, and even at very high concentrations of exogenous IL-4 (10 ng/ml), the defect of ICOS–/– T cells was not compensated.
Two possible explanations for the low responsiveness of ICOS–/– T cells to exogenous IL-4 may be considered: first, ICOS–/– T cells may have defective IL-4R expression, or second, their intracellular IL-4R signaling is impaired. To evaluate the former, we examined the expression of IL-4R -chain and c chain and found comparable expression of those on ICOS–/– and ICOS+/+ T cells (Fig. 4C). These results support the hypothesis that the defect to Th2 differentiation of ICOS–/– CD4+ T cells may be due to impaired IL-4R-mediated signaling triggered by Ag stimulation.
Alteration of IL-4R-mediated signaling in ICOS–/– naive CD4+ T cell occurs at Ag recognition
To define the mechanism underlying the impaired IL-4R signal of ICOS–/– CD4+ T cells at the level of Ag recognition, we measured Stat6 phosphorylation of ICOS–/– and ICOS+/+ CD4+ T cells in response to exogenous IL-4. To this end, naive DO11.10 CD4+ T cells from ICOS–/– and ICOS+/+ mice were stimulated with OVA peptide for 36 h. T cells were then isolated and restimulated with rIL-4, and phosphorylation levels of Stat6 were determined by Western blot analysis. As shown in Fig. 5A, phosphorylation of Stat6 in ICOS–/– T cells was significantly lower than that in ICOS+/+ T cells. To rule out the possibility that ICOS–/– naive CD4+ T cells had an intrinsic defect for Stat6 phosphorylation, ICOS–/– and ICOS+/+ naive T cells were stimulated with exogenous IL-4 without Ag stimulation. As shown in Fig. 5B, Stat6 phosphorylation in these two T cell populations to exogenous IL-4 alone did not differ. These results indicate that ICOS–/– T cells have a defect in IL-4R signaling at the level of Ag recognition, and support the hypothesis that the ICOS-mediated costimulatory signal provided by APC at the time of initial Ag stimulation augments the receptor sensitivity to IL-4.
FIGURE 5. Stat6 activation is impaired in primed ICOS–/– CD4+ T cells. A, DO11.10 naive CD4+ T cells from ICOS+/+ or ICOS–/– mice were stimulated with OVA peptide for 36 h in the presence of anti-IL-4 and anti-IL-12 Abs. T cells were then isolated and stimulated with rIL-4 (1.0 ng/ml) for 20 min. Total cell lysates were prepared, and Stat6 phosphorylation was analyzed by Western blotting. B, Naive CD4+ T cells from ICOS+/+ or ICOS–/– mice were stimulated with rIL-4 (1.0 ng/ml) without Ag stimulation. Stat6 phosphorylation was determined by Western blotting. C, DO11.10 naive CD4+ T cells were stimulated as described in A. At 24 h after stimulation, mRNA was extracted from isolated T cells, and the converted cDNAs were subjected to RT-PCR analysis for each SOCS gene expression.
Impaired IL-4R signal of ICOS–/– T cells is not regulated by suppressors of cytokine signaling (SOCS)
Recently, the importance of SOCS family proteins in Th cell differentiation by negatively regulating cytokine receptor signaling pathways has been reported (30). For example, SOCS-3 is selectively expressed in Th2 cells and enhances Th2-mediated immune responses, presumably by binding specifically to the cytoplasmic region of the IL-12R 2 and inhibiting IL-12-mediated Stat4 activation (31). Thus, it is possible that the absence of an ICOS costimulatory signal results in the down-regulation of SOCS-3, resulting in the enhancement of Th1 differentiation that antagonizes Th2 differentiation. Alternatively, because it was reported that SOCS-1 and/or SOCS-5 inhibit Stat6 phosphorylation and Th2 differentiation (32, 33), it is conceivable that abrogation of the ICOS signal increases SOCS-1 and/or SOCS-5 expression, resulting in inhibition of Stat6 phosphorylation and Th2 differentiation. To evaluate these possibilities, the mRNA expression level of SOCS genes after Ag stimulation was analyzed. At 24 h after Ag stimulation, mRNA was extracted, converted to cDNA, and subjected to RT-PCR analysis. As shown in Fig. 5C, there was no significant difference between ICOS+/+ and ICOS–/– CD4+ T cells in the expression of any SOCS genes. Thus, SOCS proteins are unlikely to be involved in this phenomenon.
Involvement of transcription factors in impaired Th2 differentiation of ICOS–/– naive CD4+ T cells
In early activation phase, several transcription factors, such as GATA-3 (4, 34), NF-B (9), and NFATc1 (6), are thought to be involved in the Th2 differentiation process. In particular, GATA-3 is recognized as a master regulator of Th2 differentiation (4) and its expression is known to be largely dependent upon the Stat6 pathway in primary responses (34, 35). Thus, we measured GATA-3 mRNA levels at 24 h after Ag stimulation from ICOS–/– and ICOS+/+ naive CD4+ T cells. As shown in Fig. 6A, GATA-3 transcription by ICOS–/– T cells was significantly lower than ICOS+/+ T cells. The presence of GATA-3 protein in the nucleus of these T cell populations was also tested at 48 h after Ag stimulation, and we found it to be reduced in ICOS–/– T cells (Fig. 6B). These results indicated that ICOS–/– T cells have a defect in the early induction of GATA-3 expression.
FIGURE 6. Early GATA-3 expression is impaired in ICOS–/– naive CD4+ T cells. A, DO11.10 naive CD4+ T cells from ICOS+/+ or ICOS–/– mice were stimulated with OVA peptide (0.1 μM). At 24 h after Ag stimulation, mRNA was extracted from isolated T cells, and the converted cDNAs were subjected to semiquantitative RT-PCR analysis (*, p = 0.014). B, DO11.10 naive CD4+ T cells were isolated and stimulated as described in A. After 24–48 h of Ag stimulation, nuclear protein was extracted from isolated T cells and subjected to Western blot analysis for translocation of indicated transcription factors. Data show representative results from one of two independent experiments.
Optimal GATA-3 induction in naive CD4+ T cells requires integration of both TCR-mediated and IL-4R-mediated signals. Recently, the importance of NF-B for optimal GATA-3 induction was reported, because GATA-3 expression and Th2 differentiation were specifically abrogated in p50–/– T cells and in cells where nuclear translocation of NF-B was inhibited (9). To test whether impaired GATA-3 induction in ICOS–/– T cells is the result of defective NF-B activation, RelA (p65), NF-B1 (p50), and c-Rel were analyzed at 24–48 h after Ag stimulation. As shown in Fig. 6B, their translocation level in ICOS–/– T cells did not differ from that of ICOS+/+ T cells, indicating that impaired GATA-3 induction in ICOS–/– T cells is not due to a defective NF-B activation, but a defective IL-4R/Stat6 pathway. It has been demonstrated that NFAT proteins regulate the promoters of multiple cytokine genes expressed in T cells. Among four NFAT family members, NFATc1 is considered to be a direct transcriptional activator of the IL-4 gene (6). Similar to NF-B, NFATc1 translocation in ICOS–/– T cells did not differ from ICOS+/+ T cells (Fig. 6B). Expression patterns of transcription factors in ICOS–/– and ICOS+/+ T cells after Ag stimulation are consistent with the conclusion that the defect of ICOS–/– T cells in Th2 differentiation may not be the result of IL-4 production, but of IL-4R-mediated signal transduction.
Discussion
In this report, we demonstrate that DO11.10 TCR Tg ICOS–/– mice have a defect in CD4+ T cells for Th2 differentiation by showing a reduced frequency of IL-4-producing cells (Fig. 2A), a reduction of total Th2-type cytokine production (IL-4, IL-5, and IL-13) (B), and impaired expression of Th2-specific proteins, such as GATA-3 and c-Maf (C), after Ag stimulation. Our results are consistent with a previous report that showed Ag-stimulated DO11.10 T cells were abrogated for the development of Th2 subsets by blocking ICOS-B7h interaction with ICOS-Ig treatment (36). Interestingly, despite the reduced frequency of IL-4-producing cells, IL-4 transcription and production levels of ICOS–/– CD4+ T cells in the early phase of Ag stimulation did not differ from that of ICOS+/+ CD4+ T cells (Fig. 3). This phenomenon is not specific for this particular gene-targeting strain, because ICOS ligation of naive CD4+ T cells from normal mice did not costimulate IL-4 transcription of these T cells in early phase of T cell activation (Fig. 1B). Because IL-4 is a key cytokine for the differentiation of Th2 cells, we considered the possibility that ICOS costimulation may influence IL-4R signaling. Indeed, it has been reported that the CD28 costimulatory signal affects IL-4 and IL-12 receptor reactivity (37, 38). To this end, we studied the role of ICOS ligation on IL-4R sensitivity using normal as well as DO11.10 TCR Tg ICOS–/– mice. We found that Stat6 phosphorylation and GATA-3 expression of normal naive CD4+ T cells were significantly enhanced by anti-ICOS mAb cross-linking (Fig. 1, C and D). Furthermore, the response of naive T cells from DO11.10 TCR Tg ICOS–/– mice to exogenous IL-4 showed a profound defect in the generation of IL-4-producing cells (Fig. 4, A and B), and IL-4-dependent Stat6 phosphorylation (Fig. 5A). In addition, GATA-3 expression in early T cell activation with OVA stimulation was impaired in naive ICOS–/– T cells (Fig. 6). There results suggested that ICOS costimulation accelerates Th2-meditated immune responses by positively controlling IL-4R signaling.
Recently, Nurieva et al. (39) reported that the defect of Th2 differentiation of ICOS–/– T cells was due to the reduction of primary IL-4 production. They showed that ICOS–/– CD4+ T cells exhibited impaired NFAT nuclear translocation in the early phase, and reduced expression of the c-Maf transcription factor, but not GATA-3, in the effector phase. Based on these results, they hypothesized that the ICOS costimulatory signal enhances NFATc1 expression and the production of IL-4 during early T cell differentiation, which regulates the expression of the IL-4-specific transcription factor c-Maf in the effector phase. However, we observed neither a defect of IL-4 production (Fig. 3) nor NFATc1 (Fig. 6B) nuclear translocation of ICOS–/– CD4+ T cells in early phase of Ag stimulation. In addition, effector T cells from ICOS–/– mice expressed significantly lower levels of GATA-3 as well as c-Maf (Fig. 2C).
Mak et al. (24) have recently reported that ICOS ligand-deficient (ICOSL–/–) mice have a similar defect to ICOS–/– mice in the development of Th2-mediated responses. They showed that the frequency of IL-4-producing cells in KLH-primed ICOSL–/– CD4+ T cells after Ag restimulation in vitro was significantly lower than that of ICOS+/+ T cells. However, IL-4 production of CD4+ T cells on a per-cell basis (fluorescence intensity) was not different between ICOSL+/+ and ICOSL–/– mice. We (Figs. 2A and 4A) as well as Tafuri et al. (23) have also shown that fluorescence intensity of IL-4 staining of Ag-responding ICOS–/– T cells was not significantly different from ICOS+/+ T cells. All these results support the notion that the ICOS-mediated costimulatory signal enhances the frequency of IL-4-positive Th2 cells, but not their capability for IL-4 production.
The reason why ICOS–/– naive T cells, described in the report by Nurieva et al. (39), were found to have a defect in primary IL-4 production is unclear. However, there are differences in their experimental systems that could explain the discrepancy with findings. First, whereas we study T cell responses against peptide Ags presented by normal APC, Nurieva et al. use anti-CD3 cross-linking for TCR stimulation and coculture APC as a source of costimulatory signals. It is conceivable that the first signal provided by TCR ligation with peptide/MHC and Ab is different in terms of quality as well as quantity. Second, we used Th2-dominant BALB/c background mice, whereas they used Th1-dominant C57BL/6 background mice. Third, T cells from their ICOS–/– mice have defects in IL-2 production and proliferative responses, whereas our mice do not have such defects. More precisely, IL-2 gene transcription (Fig. 3C) as well as IL-2 production (data not shown) by our ICOS–/– T cells after Ag stimulation was no different from those of ICOS+/+ T cells. Importantly, the ICOS–/– mice, which were generated independently by two other groups, showed similar phenotype to ours. Namely, Tafuri et al. (23) reported that T cells from their ICOS–/– mice proliferated normally to Con A, as well as anti-CD3 plus anti-CD28 stimulation. McAdam et al. (22) also showed that proliferation and IL-2 production of ICOS–/– CD4+ T cells against anti-CD3 plus APC stimulation and KLH stimulation was normal.
The molecular mechanism by which the ICOS-mediated signal enhances IL-4R activity is currently not known. Several studies indicate that the TCR- and/or costimulatory receptor-mediated signal modulates cytokine receptor reactivity during the early activation phase in a cytokine receptor expression level-independent manner (8, 37, 38, 40, 41). Zhu et al. (40) reported TCR engagement induced transient hyporesponsiveness of IL-4R signaling, and they demonstrated involvement of the ERK/MAPK and calcium pathways in this phenomenon. Thus, it is possible that decreased IL-4R responsiveness of ICOS–/– T cells may result from a perturbation in these signaling pathways. Because we show that NFATc1 nuclear translocation is comparable between ICOS+/+ and ICOS–/– T cells (Fig. 6B), the calcium pathway may be intact in ICOS–/– T cells. It will be of interest to determine whether ICOS deficiency affects the MAPK pathway.
It is well-established that the PI3K pathway plays an important role for CD28-mediated costimulation. We have previously shown that CD28 cross-linking enhances the IL-4R sensitivity (37). Because ICOS cytoplasmic region also has a PI3K binding motif, it is conceivable that PI3K pathway may be involved in up-regulation of IL-4R sensitivity by ICOS ligation. In fact, a number of studies also predict the role of PI3K pathway in Th differentiation. Suzuki et al. (42) showed that mice lacking PTEN, which is a phosphatase for PIP3, exhibit enhancement of PI3K signal pathway and increase in both Th1 and Th2 cytokine production. Arimura et al. (43) demonstrated that overexpression of Akt, a major downstream target of PI3K, resulted in amplification of both Th1 and Th2 development. Furthermore, Schaeffer et al. (44) have reported that Tec family tyrosine kinase Itk, which is also regulated by PI3K, has an important role for Th2 development. Using ICOS mutant mice, in which ICOS does not bind to PI3K, we are currently studying a specific role of ICOS-PI3K pathway for IL-4R signaling in Ag-specific T cell responses.
Another considerable possibility is that, similar to the IL-2R complex (45), the IL-4R may also be reorganized within the cell membrane upon T cell activation, and ICOS might have an important role for the formation of this functional receptor complex. Additional experiments are required to elucidate the molecular mechanism underlying this phenomenon.
From our results, it is expected that ICOS blockade effectively prevents the onset of Th2-dysregulated immune responses, such as allergy. In contrast, several reports indicated that the administration of ICOS-B7h blocking agents at an early phase of some immune responses exacerbates the pathological symptoms, and this may be due to augmenting Th1 responses (16, 20). Thus, although recent studies predict ICOS to be an attractive therapeutic target, it is important to understand the precise role of ICOS throughout the entire immune response. We believe our study will contribute to a better understanding of ways to approach ICOS-targeting therapeutic strategies.
Acknowledgments
We thank Dr. Motoko Kotani and Dr. Shuhei Ogawa for their discussion, and a member of Science Service, Inc., for care of experimental animals. We also gratefully acknowledge JT Pharmaceutical Frontier Research Laboratories for providing the ICOS–/– mice and anti-ICOS mAb.
Footnotes
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 a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology, Japan.
2 Address correspondence and reprint requests to Dr. Ryo Abe, Research Institute for Biological Sciences, Tokyo University of Sciences, 2669 Yamazaki, Noda, Chiba 278-0022, Japan. E-mail address: rabe@rs.noda.tus.ac.jp
3 Abbreviations used in this paper: Tg, transgenic; F, forward; R, reverse; SOCS, suppressor of cytokine signaling; ICOSL, ICOS ligand.
Received for publication September 22, 2004. Accepted for publication November 19, 2004.
References
Murphy, K. M., S. L. Reiner. 2002. The lineage decisions of helper T cells. Nat. Rev. Immunol. 2:933.
Kaplan, M. H., U. Schindler, S. T. Smiley, M. J. Grusby. 1996. Stat6 is required for mediating responses to IL-4 and for development of Th2 cells. Immunity 4:313.
Takeda, K., T. Tanaka, W. Shi, M. Matsumoto, M. Minami, S. Kashiwamura, K. Nakanishi, N. Yoshida, T. Kishimoto, S. Akira. 1996. Essential role of Stat6 in IL-4 signalling. Nature 380:627.
Zheng, W., R. A. Flavell. 1997. The transcription factor GATA-3 is necessary and sufficient for Th2 cytokine gene expression in CD4 T cells. Cell 89:587.
Kurata, H., H. J. Lee, A. O’Garra, N. Arai. 1999. Ectopic expression of activated Stat6 induces the expression of Th2-specific cytokines and transcription factors in developing Th1 cells. Immunity 11:677.
Yoshida, H., H. Nishina, H. Takimoto, L. E. Marengere, A. C. Wakeham, D. Bouchard, Y. Y. Kong, T. Ohteki, A. Shahinian, M. Bachmann, et al 1998. The transcription factor NF-ATc1 regulates lymphocyte proliferation and Th2 cytokine production. Immunity 8:115.
Hodge, M. R., A. M. Ranger, F. Charles de la Brousse, T. Hoey, M. J. Grusby, L. H. Glimcher. 1996. Hyperproliferation and dysregulation of IL-4 expression in NF-ATp-deficient mice. Immunity 4:397.
Yamashita, M., M. Kimura, M. Kubo, C. Shimizu, T. Tada, R. M. Perlmutter, T. Nakayama. 1999. T cell antigen receptor-mediated activation of the Ras/mitogen-activated protein kinase pathway controls interleukin 4 receptor function and type-2 helper T cell differentiation. Proc. Natl. Acad. Sci. USA 96:1024.
Das, J., C. H. Chen, L. Yang, L. Cohn, P. Ray, A. Ray. 2001. A critical role for NF-B in GATA3 expression and TH2 differentiation in allergic airway inflammation. Nat. Immunol. 2:45.
Carreno, B. M., M. Collins. 2002. The B7 family of ligands and its receptors: new pathways for costimulation and inhibition of immune responses. Annu. Rev. Immunol. 20:29.
Sharpe, A. H., G. J. Freeman. 2002. The B7-CD28 superfamily. Nat. Rev. Immunol. 2:116
Hutloff, A., A. M. Dittrich, K. C. Beier, B. Eljaschewitsch, R. Kraft, I. Anagnostopoulos, R. A. Kroczek. 1999. ICOS is an inducible T-cell co-stimulator structurally and functionally related to CD28. Nature 397:263.
Yoshinaga, S. K., J. S. Whoriskey, S. D. Khare, U. Sarmiento, J. Guo, T. Horan, G. Shih, M. Zhang, M. A. Coccia, T. Kohno, et al 1999. T-cell co-stimulation through B7RP-1 and ICOS. Nature 402:827.
Tamatani, T., K. Tezuka, N. Hanzawa-Higuchi. 2000. AILIM/ICOS: a novel lymphocyte adhesion molecule. Int. Immunol. 12:51.
Coyle, A. J., S. Lehar, C. Lloyd, J. Tian, T. Delaney, S. Manning, T. Nguyen, T. Burwell, H. Schneider, J. A. Gonzalo, et al 2000. The CD28-related molecule ICOS is required for effective T cell-dependent immune responses. Immunity 13:95.
Ogawa, S., G. Nagamatsu, M. Watanabe, S. Watanabe, T. Hayashi, S. Horita, K. Nitta, H. Nihei, K. Tezuka, R. Abe. 2001. Opposing effects of anti-activation-inducible lymphocyte-immunomodulatory molecule/inducible costimulator antibody on the development of acute versus chronic graft-versus-host disease. J. Immunol. 167:5741.
Gonzalo, J. A., J. Tian, T. Delaney, J. Corcoran, J. B. Rottman, J. Lora, A. Al-garawi, R. Kroczek, J. C. Gutierrez-Ramos, A. J. Coyle. 2001. ICOS is critical for T helper cell-mediated lung mucosal inflammatory responses. Nat. Immunol. 2:597.
Miyahira, Y., H. Akiba, S. H. Ogawa, T. Ishi, S. Watanabe, S. Kobayashi, T. Takeuchi, T. Aoki, K. Tezuka, R. Abe, et al 2003. Involvement of ICOS-B7RP-1 costimulatory pathway in the regulation of immune responses to Leishmania major and Nippostrongylus brasiliensis infections. Immunol. Lett. 89:193.
Ozkaynak, E., W. Gao, N. Shemmeri, C. Wang, J. C. Gutierrez-Ramos, J. Amaral, S. Qin, J. B. Rottman, A. J. Coyle, W. W. Hancock. 2001. Importance of ICOS-B7RP-1 costimulation in acute and chronic allograft rejection. Nat. Immunol. 2:591.
Rottman, J. B., T. Smith, J. R. Tonra, K. Ganley, T. Bloom, R. Silva, B. Pierce, J. C. Gutierrez-Ramos, E. Ozkaynak, A. J. Coyle. 2001. The costimulatory molecule ICOS plays an important role in the immunopathogenesis of EAE. Nat. Immunol. 2:605.
Dong, C., A. E. Juedes, U. A. Temann, S. Shresta, J. P. Allison, N. H. Ruddle, R. A. Flavell. 2001. ICOS co-stimulatory receptor is essential for T-cell activation and function. Nature 409:97.
McAdam, A. J., R. J. Greenwald, M. A. Levin, T. Chernova, N. Malenkovich, V. Ling, G. J. Freeman, A. H. Sharpe. 2001. ICOS is critical for CD40-mediated antibody class switching. Nature 409:102.
Tafuri, A., A. Shahinian, F. Bladt, S. K. Yoshinaga, M. Jordana, A. Wakeham, L. M. Boucher, D. Bouchard, V. S. Chan, G. Duncan, et al 2001. ICOS is essential for effective T-helper-cell responses. Nature 409:105.
Mak, T. W., A. Shahinian, S. K. Yoshinaga, A. Wakeham, L. M. Boucher, M. Pintilie, G. Duncan, B. U. Gajewska, M. Gronski, U. Eriksson, et al 2003. Costimulation through the inducible costimulator ligand is essential for both T helper and B cell functions in T cell-dependent B cell responses. Nat. Immunol. 4:765.(Masashi Watanabe, Shiho W)
ICOS is the third member of the CD28 family molecules and plays a critical role in many T cell-dependent immune responses. Although accumulated data suggest that ICOS costimulatory signals play an important role in Th2-mediated immune responses, the molecular basis for this selective differentiation mechanism is largely unknown. To clarify this mechanism, we used DO11.10 TCR transgenic ICOS–/– mice and evaluated the nature of ICOS costimulatory signals during the process of Ag-specific activation and differentiation of naive CD4+ T cells. Results obtained from these experiments demonstrated that Ag stimulation of naive CD4+ T cells in the absence of an ICOS signal resulted in impaired Th2 development. Unlike previous reports, we found that primary IL-4 production by these T cells was intact and that IL-4R sensitivity of these T cells was reduced as evidenced by a profound defect in IL-4-induced Stat6 phosphorylation and the early induction of GATA-3. The fact that ICOS ligation of wild-type T cells significantly enhanced IL-4-induced Stat6 phosphorylation and primary GATA-3 induction, but not IL-4 transcription, of naive CD4+ T cells was consistent with the results obtained from ICOS–/– T cell experiments. These observations led us to propose that the predominant effect of ICOS-mediated costimulation on Th2 differentiation is achieved by the enhancement of IL-4R-mediated signaling.
Introduction
Helper T cell differentiation is controlled by several signaling pathways such as cytokine signals, TCR signals, and costimulatory signals (1). The IL-4/IL-4R/Stat6 pathway is critical for Th2 differentiation of Ag-stimulated naive CD4+ T cells by enhancing the expression of GATA-3, a master regulator of Th2 differentiation (2, 3, 4, 5). It has been reported that NFAT, MAPK, and/or NF-B pathways, which are downstream of TCR and costimulatory signals, are involved in modulating Th2 differentiation by affecting IL-4 transcription, IL-4R sensitivity, and/or GATA-3 expression (6, 7, 8, 9).
Costimulatory signals, in addition to TCR-mediated signals, are necessary for full activation and play an important role in the regulation of T cell responses (10, 11). The CD28 family molecules, CD28 and its homolog CTLA-4, are the most extensively characterized costimulatory molecules that positively or negatively regulate T cell responses, respectively (10, 11). More recently, the third member of CD28 family, ICOS, was identified (12, 13, 14). ICOS expressed at very low levels on naive T cells, is up-regulated after T cell activation, and is retained on effector/memory T cells. In particular, ICOS expression is skewed in Th cell subsets, namely, higher in Th2 than in Th1 (15). ICOS ligation has been shown to enhance TCR-mediated T cell proliferation, indicating that ICOS provides CD28-like positive costimulatory signals to T cells. In fact, like CD28, ICOS-mediated signals delivered by stimulation with anti-ICOS mAb or an Ig fusion protein of ICOS ligand, B7h (B7h-Ig), can enhance secretion of IL-4, IL-5, IL-10, IFN-, and TNF-. However, in contrast to CD28, ICOS does not costimulate IL-2 secretion (12, 13).
The important role of ICOS in various immune responses has been well documented (11). In particular, much attention has been drawn to the role of ICOS in Th cell differentiation. A number of studies demonstrated that blocking of ICOS-mediated costimulation with Abs, or with an ICOS-Ig fusion protein, resulted in the preferential inhibition of Th2-mediated immune responses (15, 16, 17, 18), although some Th1 responses were found to be affected (19, 20). Several ICOS-deficient mouse lines have been independently generated and examined for their immune function (21, 22, 23). It has been shown that these mice have defects in germinal center formation and Ab class switching (22, 23), and showed exacerbation of experimental autoimmune encephalomyelitis (21). Furthermore, ICOS- and B7h-deficient mice have also been shown to have a deficiency in in vivo generation of Ag-specific Th2 development (23, 24). These findings strongly suggested that ICOS plays a predominant role in Th2 differentiation, and/or effector function. However, the molecular mechanisms by which ICOS costimulatory signals enhance Th2 differentiation remain undefined.
In this study, we focused on the role of the ICOS costimulatory signal on Th2 differentiation of naive CD4+ T cells. We found that ICOS ligation by Ab cross-linking selectively enhanced the generation of IL-4-producing CD4+ T cells. Interestingly, it was found that ICOS ligation did not costimulate IL-4 transcription of T cells in the early phase of T cell activation. Instead, phosphorylation of Stat6 and expression of GATA-3 were dramatically enhanced, suggesting that the ICOS signal enhances IL-4R signaling. To evaluate the role of ICOS costimulatory signals during Ag-specific activation and differentiation of naive CD4+ T cells under physiological conditions, we examined the response to OVA peptide of naive T cells from ICOS–/– mice that have DO11.10 TCR transgenic (Tg)3 BALB/c background genes. We found that primary IL-4 production by these ICOS–/– T cells was intact and equivalent to ICOS+/+ T cells; however, they showed reduced IL-4R sensitivity as evidenced by a profound defect in IL-4-induced Stat6 phosphorylation and the early induction of GATA-3. These data provide evidence that a predominant effect of ICOS-mediated costimulation on Th2 differentiation may be achieved by the enhancement of IL-4R-mediated signaling.
Materials and Methods
Mice
ICOS–/– mice were generated as previously described (25) and provided by JT Pharmaceutical Frontier Research Laboratory. ICOS–/– mice were backcrossed onto BALB/c or C57BL/6 mice over six to eight times. ICOS–/– DO11.10 TCR Tg mice were obtained by intercrossing BALB/c background ICOS–/– mice and BALB/c DO11.10 TCR Tg mice, which were originally distributed by Dr. D. Loh (Roche, Nutley, NJ). Wild-type BALB/c and C57BL/6 mice were obtained from Sankyo Laboratory. Mice were maintained in our animal facility in specific pathogen-free conditions. For entire experiments, 8- to 10-wk-old mice were used. The experiment herein were conducted according to the principles set forth in Ref.26 .
Abs and reagents
Anti-mouse-ICOS mAb (B10.5) was produced as previously described (16) and provided by JT Pharmaceutical Frontier Research Laboratory. Purified anti-CD3 (2C11), anti-TCR (H57), anti-CD28 (PV-1), and rat anti-NP-IgG2a Abs, and nonpurified anti-CD8 Ab (53-6-7) were prepared in our laboratory from hybridoma culture medium. FITC-labeled anti-CD44 and IFN- Abs, PE-labeled anti-IL-4R and IL-4 Abs, allophycocyanin-labeled anti-CD4 Ab, biotin-labeled anti-IL-4, IL-5, and IFN- Abs, and purified anti-IL-4, IL-5, IFN-, and CD132 (c) Abs were purchased from BD Pharmingen. Biotin-labeled anti-IL-13, purified anti-IL-13, and anti-Stat6 Abs were purchased from R&D Systems. Anti-GATA-3, T-bet, NFATc1, NF-B1, c-Rel, RelA, and -actin Abs were purchased from Santa Cruz. Anti-phospho-Stat-6 Ab was purchased from Cell Signaling.
Preparation of Ab-conjugated beads
Latex beads were purchased from Interfacial Dynamics Corporation. Briefly, Latex beads (5 x 107) were washed twice with PBS and were incubated with anti-CD3 mAb (2C11; 1 μg/ml) plus one of following Abs: anti-CD28 (PV-1; 9 μg/ml), anti-ICOS (B10.5; 9 μg/ml), or anti-NP-IgG2a (isotype control; 9 μg/ml) for 1.5 h at 37°C. Beads were then washed with PBS, and blocked with 10% FCS/RPMI 1640 medium.
Cell purification and stimulation
Naive (CD44low) CD4+ T cells were purified by cell sorter. Briefly, to enrich bulk CD4+ T cells, single-cell suspensions of mice splenocytes were treated with anti-CD8 (53-6-7), and incubated on anti-mouse Ig-coated dish to eliminate CD8+ T cells and B cells. The cells were recovered and stained with anti-CD44-FITC and anti-CD4-allophycocyanin, and CD4+CD44low cell population was collected with FACSVantage (BD Pharmingen) as naive CD4+ T cells. Purity of sorted population was constantly >95%. Purified DO11.10 CD4+CD44low T cells (1 x 105) were stimulated with OVA323–339 peptide (0.1 μM) in the presence of mitomycin C-treated T-depleted normal BALB/c spleen cells (3 x 105) in 96-well plate. Naive CD4+ T cells (1 x 105) purified from normal mice were stimulated with Ab-conjugated beads (1 x 105) in 96-well culture plate. For cell culture, RPMI 1640 medium (Sigma-Aldrich) containing 10% FCS, 2ME, L-glutamine, penicillin/streptomycin, and 0.1 mM HEPES was used.
ELISA
Primary Ab was coated on 96-well plate at 37°C for 1 h, and the plate was blocked with 1% BSA in PBS. Culture supernatants was then applied on the plate and incubated for 2 h at 37°C. The plate was washed and reacted with biotinylated secondary Ab for 1 h, followed by reacting with HRP-coupled streptavidin (Sigma-Aldrich) for 30 min. Substrate (ABTS plus H2O2) was applied and OD405 was measured by Microplate Reader model 3550 (Bio-Rad).
Flow cytometry
Briefly, cells were washed with wash buffer (PBS containing 1% BSA and 0.05% NaN3), treated with anti-FcR (2.4G2), stained with anti-IL-4R-PE or anti-CD132 (c) followed by anti-rat-PE Abs. Electrical data was acquired on a FACSCalibur and analyzed by CellQuest software (BD Pharmingen).
Intracellular cytokine staining
Cultured Th cells were recovered, washed, and stimulated with plate-bound anti-TCR mAb (H57) for 6 h in the presence of monensin (2 μM). The cells were washed, fixed with 4% paraformaldehyde for 10 min, and treated with permeabilization buffer (50 mM NaCl, 5 mM EDTA, 0.02% NaN3, 0.5% Triton X) for 10 min. IL-4 and IFN- were detected with anti-IL-4-PE and anti-IFN--FITC Abs by flow cytometry.
Western blot analysis
To prepare total cell lysate, naive or activated CD4+ T cells were treated with lysis buffer (500 mM NaCl, 5 mM EDTA (pH 8.0), 100 mM Tris-HCl (pH 8.0), 1.25% Nonidet P-40) for 30 min on ice, and centrifuged at 15,000 rpm for 10 min. The supernatants were used for experiments. To prepare nuclear protein, cells were washed with PBS, and suspended in lysis buffer (10 mM HEPES (pH7.5), 10 mM KCl, 0.1 mM EDTA, 0.1% Nonidet P-40), vortexed vigorously, and centrifuged at 5,000 rpm for 2 min. Pellet were then suspended in extraction buffer (20 mM HEPES (pH 7.9), 0.4 M NaCl, 1 mM EDTA), incubated for 30 min on ice, and centrifuged at 15,000 rpm for 10 min. The supernatants were used as nuclear protein extract. Protein concentration of each sample was determined by BCA protein assay reagent (Pierce). Same amount of protein samples was resolved by SDS-PAGE with 12% acrylamide gel, transferred to polyvinylidene difluoride membrane, probed with specific primary Ab, followed by reacting with HRP-conjugated secondary Ab, and developed by ECL detection system (PerkinElmer).
RT-PCR
Total RNA was extracted using TRIzol reagent (Sigma-Aldrich), and mRNA was reverse-transcribed to cDNA using oligo(dT) primer and SuperScript II (Invitrogen Life Technologies). The following primer sets were used for RT-PCR analysis: IL-2 (forward (F): GAGTCAAATCCAGAACATGCC; reverse (R): TCCACTTCAAGCTCTACAG); IL-4 (F: GAATGTACCAGGAGCCATATC; R: CTCAGTACTACGAGTAATCCA); IFN- (F: AACGCTACACACTGCATCTTGG; R: GACTTCAAAGAGTCTGAGG); IL-10 (F: CGGGAAGACAATAACTG; R: CATTTCCGATAAGGCTTGG); GATA-3 (F: GCAGAACCGGCCCCTTATCAA; R: CATGGGCGTCGGTGTGGTCAG); SOCS-1 (F: CTTAACCCGGTACTCCGTGA; R: GAGGTCTCCAGCCAGAAGTG); SOCS-3 (F: AGCAAGTTTCCCGCCGCC; R: GCAGCGAAAAGCTGCCCC); SOCS-5 (F: CGGCCTTCAGAGGCGAGA; R: AGTGTGGGCTCACAGGGG); CIS (F: CCCGGCAGCACCTACAGA; R: CACGGGGATAGGCAGCAC); HPRT (F: GTTGGATACAGGCCAGACTTTGTTG; R: GATTCAACTTGCGCTCATCTTAGG).
Results
ICOS signaling does not influence IL-4 transcription of naive CD4+ T cells, but enhances IL-4R signaling in Th2 differentiation
The effect of ICOS ligation on Th cell differentiation was evaluated by stimulation of naive CD4+ T cells with beads containing an anti-ICOS mAb in the presence of suboptimal anti-CD3 and anti-CD28 mAb. As shown in Fig. 1A, anti-CD3/CD28/ICOS stimulation significantly enhanced the generation of IL-4-producing T cells compared with anti-CD3/CD28 stimulation in both BALB/c- and C57BL/6-derived naive CD4+ T cells. These results indicated that the ICOS-mediated costimulatory signal preferentially acts on Th2 differentiation.
FIGURE 1. ICOS cross-linking enhances Th2 differentiation by increasing IL-4R sensitivity. A, ICOS cross-linking by anti-ICOS mAb enhances Th2 differentiation. Purified CD44low naive CD4+ T cells from BALB/c or C57BL/6 mice were stimulated with the indicated Ab-conjugated beads. After 6-day stimulation, Th differentiation was analyzed by intracellular cytokine staining. B, ICOS cross-linking does not affect primary IL-4 transcription. Purified CD44low naive CD4+ T cells from BALB/c mice were stimulated as indicated for 24–48 h. IL-4 mRNA levels were determined by RT-PCR. C, ICOS cross-linking enhances IL-4-induced Stat6 phosphorylation. Purified CD44low naive CD4+ T cells from BALB/c mice were stimulated for 36 h in the presence of anti-IL-4 and anti-IL-12 Abs, washed, and then restimulated with exogenous IL-4 (1.0 ng/ml) for 20 min. Total cell lysates were prepared, and Stat6 phosphorylation was determined by Western blotting. D, ICOS cross-linking enhances GATA-3 expression. Purified CD44low naive CD4+ T cells from BALB/c mice were stimulated as indicated for 48 h. Total cell lysates were subjected to Western blot analysis for GATA-3 expression. A–D, Data show representative results from one of three independent experiments.
This preferential Th2 differentiation by ICOS ligation could be achieved by two distinct mechanisms. First, the ICOS-mediated costimulatory signal could augment the initial production of IL-4 by T cells, and second, it could enhance IL-4R sensitivity, both of which would accelerate IL-4R signaling. To evaluate the first possibility, IL-4 transcription of purified naive CD4+ T cells was examined at 24–48 h after coculture with beads bound by Abs specific for CD3, CD28, and/or ICOS. As shown in Fig. 1B, anti-CD3/CD28/ICOS cross-linking had no apparent effect on IL-4 transcription of CD4+ T cells compared with anti-CD3/CD28 stimulation, indicating that the ICOS signal had no costimulatory effect on the initial expression of IL-4 by T cells. To evaluate the second possibility, we tested IL-4R signaling after stimulation with ICOS by analyzing the signaling pathway downstream of the IL-4R. It is well known that Stat6 plays a critical role in the IL-4-mediated signal, and that its phosphorylation reflects the strength of IL-4R signaling (2, 3, 27). Thus, the effect of ICOS cross-linking on the capability of IL-4 to induce Stat6 phosphorylation was tested. Naive CD4+ T cells were stimulated with Ab-conjugated beads for 36 h and restimulated with rIL-4 for 20 min, and their Stat6 phosphorylation level was determined by Western blot analysis. As shown in Fig. 1C, although CD28 ligation alone induced a weak IL-4-dependent Stat6 phosphorylation signal, the additional ligation of ICOS dramatically enhanced this effect. These results suggest that ICOS ligation enhances IL-4R signaling.
After stimulation by IL-4, the phosphorylated Stat6 dimerizes and translocates to the nucleus where it activates a number of IL-4 target genes, including GATA-3. To confirm the enhancing effect of ICOS ligation on the IL-4R signal, we examined GATA-3 production after stimulation with Ab-conjugated beads (Fig. 1D). In the presence of suboptimal CD3 and CD28 stimulation, ICOS cross-linking dramatically enhanced GATA-3 production. Although we constantly observed weak GATA-3 induction by CD3 plus ICOS stimulation, this phenomenon was not likely to be a direct effect of ICOS signaling because anti-IL-4 mAb could abrogate GATA-3 induction in this experimental condition (data not shown).
The enhancement of downstream signaling of the IL-4R and no observable effects on IL-4 transcription are consistent with the idea that the ICOS-mediated Th2 differentiation of naive CD4+ T cells is caused by the enhancement of IL-4R sensitivity but not by IL-4 production.
Ag stimulation in the absence of ICOS results in impaired Th2 differentiation
To characterize the physiological role of ICOS in in vivo immune responses, we used ICOS–/– mice (25). Several groups have reported that ICOS deficiency resulted in attenuation of germinal center formation and Ab class switching (22, 23). Consistent with their reports, our ICOS–/– mice showed a defect in IgG1 Ab production and germinal center formation in lymph nodes and spleens against the T-dependent Ags (data not shown). These results are consistent with the notion that ICOS–/– mice have a defect in the differentiation and effector function of Th2 immune responses.
To evaluate the molecular mechanisms of preferential Th2 differentiation by ICOS under physiological conditions, we crossed ICOS–/– mice with DO11.10 TCR Tg mice and examined the responses of their naive CD4+ T cells to OVA peptide presented by APC.
Highly purified naive CD4+ T cells (CD44low) from DO11.10 ICOS–/– or ICOS+/+ mice were stimulated with OVA peptide in the presence of mitomycin C-treated normal BALB/c T-depleted spleen cells. At 48 h after Ag stimulation, cultured cells were harvested, washed, and recultured for additional 5 days in the presence of rIL-2 to generate Th cells. Live cells were restimulated with anti-TCR Ab, and their cytokine contents were analyzed by flow cytometry. As shown in Fig. 2A, the IL-4-producing T cell population in cultured ICOS–/– T cells was significantly smaller than that in ICOS+/+ T cells. In repeated experiments, the IL-4-producing population in ICOS–/– T cells was consistently reduced by 30–50% compared with ICOS+/+ T cells, indicating that ICOS–/– CD4+ T cells show impaired Th2 development after stimulation with Ag.
FIGURE 2. Impaired Th2 differentiation in Ag-stimulated ICOS–/– naive CD4+ T cells. DO11.10 naive CD4+ T cells from ICOS+/+ or ICOS–/– mice were stimulated with OVA peptide (0.1 μM) for 48 h. Cells were then recovered, washed, and recultured with rIL-2 (30 U/ml) for an additional 5 days. A, Th cells were restimulated with plate-bound anti-TCR for 6 h in the presence of monensin, and Th differentiation pattern was analyzed by intracellular cytokine staining. Data show representative one of three independent experiments. B, Th cells were stimulated with plate-bound anti-TCR for 24 h. IL-4, IL-5, IL-13, and IFN- concentrations in the culture supernatants were determined by ELISA. C, Th cells were restimulated as described in B, cell lysates were prepared, and protein expression levels of GATA-3, c-Maf, and T-bet were analyzed by Western blotting. B and C, Data show representative results from one of two independent experiments.
Next, we studied cytokine production by these T cells. By measuring cytokines in culture supernatants, we found that ICOS–/– T cells produce significantly less Th2 cytokines, IL-4, IL-5, and IL-13, than ICOS+/+ T cells, whereas IFN- production was not significantly different between ICOS+/+ and ICOS–/– T cells (Fig. 2B). Finally, we tested the production of transcription factors that are specific for Th subsets. As shown in Fig. 2C, the production of Th2-specific transcription factors, GATA-3 (4) and c-Maf (28), were reduced in ICOS–/– T cells, but the production level of Th1-specific transcription factor T-bet (29) was slightly higher in ICOS–/– T cells. These results confirmed the conclusion that ICOS–/– T cells have a defect in Th2 differentiation by Ag stimulation.
ICOS deficiency does not affect primary IL-4 production of naive CD4+ T cells
To test whether the ICOS-mediated Th2 differentiation of naive CD4+ T cells is the result of the enhancement of IL-4R sensitivity, but not IL-4 production under physiological conditions with Ag stimulation, we studied ICOS–/– DO11.10 TCR Tg T cells for IL-4 production and IL-4R sensitivity.
Highly purified DO11.10 naive CD4+ T cells from ICOS+/+ or ICOS–/– mice were stimulated with OVA peptide, and their IL-4 and IFN- production for the first 48 h after Ag stimulation was determined by ELISA. Although in each experiment the amount of IL-4 detected in the culture medium varied, there was no significant difference in IL-4 and IFN- production between ICOS+/+ and ICOS–/– T cells (Fig. 3A). We also tested IL-4 production from these cultures at different time points and found that IL-4 production by ICOS–/– T cells was not different from that of ICOS+/+ cells (Fig. 3B). Furthermore, cytokine gene transcription by these two T cell populations at 24 h after stimulation did not show differences when evaluated by semiquantitative RT-PCR (Fig. 3C) or quantitative real-time PCR (data not shown), indicating that cytokine transcription, including IL-4, by Ag-stimulated ICOS–/– T cells is normal. Similar results were also observed when non-TCR Tg T cells, isolated from BALB/c or C57BL/6 background ICOS–/– mice, were stimulated with anti-CD3 plus APC (data not shown). These results collectively indicated that impaired Th2 differentiation in ICOS–/– CD4+ T cells is not due to a defect in primary IL-4 production after Ag stimulation.
FIGURE 3. Primary IL-4 production is normal in Ag-stimulated ICOS–/– naive CD4+ T cells. A, DO11.10 naive CD4+ T cells from ICOS+/+ or ICOS–/– mice were stimulated with OVA peptide (0.1 μM) for 48 h. IL-4 and IFN- concentrations in culture supernatants were determined by ELISA. Data show results of five independent experiments. B, DO11.10 naive CD4+ T cells were stimulated as described in A. IL-4 concentrations of culture supernatants at the indicated time points were determined by ELISA. Data show representative one from two independent experiments. C, DO11.10 naive CD4+ T cells were stimulated as described in A. At 24 h after Ag stimulation, mRNA was extracted from isolated T cells, and the converted cDNAs were subjected to semiquantitative RT-PCR for the indicated cytokine gene products. Data show representative results from one of two independent experiments.
Next, we evaluated whether exogenous IL-4 could compensate for this impairment of Th2 differentiation in ICOS–/– CD4+ T cells. Highly purified DO11.10 naive CD4+ T cells from ICOS–/– or ICOS+/+ mice were stimulated with OVA peptide in the presence or absence of rIL-4 (10 ng/ml) for 48 h. After extensive washes and 5 additional days of cultivation in the presence of rIL-2 (30 U/ml), live cells were tested for IL-4 and IFN- by intercellular cytokine staining (Fig. 4A). Under these conditions, IL-4Rs on ICOS+/+ DO11.10 T cells might be presumably fully saturated with endogenous IL-4, indicating that exogenous IL-4 did not affect the frequency of IL-4-producing cells. Consistent with previous experiments, ICOS–/– DO11.10 T cells showed reduced numbers of IL-4-producing T cells, suggesting that this defective Th2 differentiation could not be compensated by the addition of exogenous IL-4.
FIGURE 4. Exogenous IL-4 is insufficient to compensate ICOS deficiency for Th2 development. A, DO11.10 naive CD4+ T cells from ICOS+/+ or ICOS–/– mice were stimulated with OVA peptide (0.1 μM) in the presence of rIL-4 (10 ng/ml). At 48 h after Ag stimulation, cells were washed and recultured in exogenous IL-2 (30 U/ml) for an additional 5 days. At day 7, Th differentiation was analyzed by intracellular cytokine staining. B, DO11.10 naive CD4+ T cells were isolated and stimulated as described in A with the exception that rIL-2 was not added to the reculture medium. Data show representative one of two independent experiments. C, DO11.10 naive CD4+ T cells from ICOS+/+ or ICOS–/– mice were stimulated with OVA peptide (0.1 μM) for 48 h. IL-4R and c expression was analyzed by flow cytometry. Data show representative results from one of three independent experiments.
For further analysis of IL-4R signaling in ICOS–/– CD4+ T cells, naive CD4+ T cells were stimulated with OVA peptide in the presence of serial dilutions of exogenous IL-4 for 48 h, washed, and cultured for a further 5 days without rIL-2. At day 7, Th differentiation was evaluated by intracellular cytokine staining of IL-4 and IFN-. Under these culture conditions, ICOS+/+ DO.11.10 CD4+ T cells did not differentiate into Th2 cells without additional IL-4. As shown in Fig. 4B, addition of exogenous IL-4 resulted in the generation of IL-4-producing ICOS+/+ T cells in a dose-dependent manner. In contrast, the effect of exogenous IL-4 on Th2 differentiation of ICOS–/– T cells was weaker, and even at very high concentrations of exogenous IL-4 (10 ng/ml), the defect of ICOS–/– T cells was not compensated.
Two possible explanations for the low responsiveness of ICOS–/– T cells to exogenous IL-4 may be considered: first, ICOS–/– T cells may have defective IL-4R expression, or second, their intracellular IL-4R signaling is impaired. To evaluate the former, we examined the expression of IL-4R -chain and c chain and found comparable expression of those on ICOS–/– and ICOS+/+ T cells (Fig. 4C). These results support the hypothesis that the defect to Th2 differentiation of ICOS–/– CD4+ T cells may be due to impaired IL-4R-mediated signaling triggered by Ag stimulation.
Alteration of IL-4R-mediated signaling in ICOS–/– naive CD4+ T cell occurs at Ag recognition
To define the mechanism underlying the impaired IL-4R signal of ICOS–/– CD4+ T cells at the level of Ag recognition, we measured Stat6 phosphorylation of ICOS–/– and ICOS+/+ CD4+ T cells in response to exogenous IL-4. To this end, naive DO11.10 CD4+ T cells from ICOS–/– and ICOS+/+ mice were stimulated with OVA peptide for 36 h. T cells were then isolated and restimulated with rIL-4, and phosphorylation levels of Stat6 were determined by Western blot analysis. As shown in Fig. 5A, phosphorylation of Stat6 in ICOS–/– T cells was significantly lower than that in ICOS+/+ T cells. To rule out the possibility that ICOS–/– naive CD4+ T cells had an intrinsic defect for Stat6 phosphorylation, ICOS–/– and ICOS+/+ naive T cells were stimulated with exogenous IL-4 without Ag stimulation. As shown in Fig. 5B, Stat6 phosphorylation in these two T cell populations to exogenous IL-4 alone did not differ. These results indicate that ICOS–/– T cells have a defect in IL-4R signaling at the level of Ag recognition, and support the hypothesis that the ICOS-mediated costimulatory signal provided by APC at the time of initial Ag stimulation augments the receptor sensitivity to IL-4.
FIGURE 5. Stat6 activation is impaired in primed ICOS–/– CD4+ T cells. A, DO11.10 naive CD4+ T cells from ICOS+/+ or ICOS–/– mice were stimulated with OVA peptide for 36 h in the presence of anti-IL-4 and anti-IL-12 Abs. T cells were then isolated and stimulated with rIL-4 (1.0 ng/ml) for 20 min. Total cell lysates were prepared, and Stat6 phosphorylation was analyzed by Western blotting. B, Naive CD4+ T cells from ICOS+/+ or ICOS–/– mice were stimulated with rIL-4 (1.0 ng/ml) without Ag stimulation. Stat6 phosphorylation was determined by Western blotting. C, DO11.10 naive CD4+ T cells were stimulated as described in A. At 24 h after stimulation, mRNA was extracted from isolated T cells, and the converted cDNAs were subjected to RT-PCR analysis for each SOCS gene expression.
Impaired IL-4R signal of ICOS–/– T cells is not regulated by suppressors of cytokine signaling (SOCS)
Recently, the importance of SOCS family proteins in Th cell differentiation by negatively regulating cytokine receptor signaling pathways has been reported (30). For example, SOCS-3 is selectively expressed in Th2 cells and enhances Th2-mediated immune responses, presumably by binding specifically to the cytoplasmic region of the IL-12R 2 and inhibiting IL-12-mediated Stat4 activation (31). Thus, it is possible that the absence of an ICOS costimulatory signal results in the down-regulation of SOCS-3, resulting in the enhancement of Th1 differentiation that antagonizes Th2 differentiation. Alternatively, because it was reported that SOCS-1 and/or SOCS-5 inhibit Stat6 phosphorylation and Th2 differentiation (32, 33), it is conceivable that abrogation of the ICOS signal increases SOCS-1 and/or SOCS-5 expression, resulting in inhibition of Stat6 phosphorylation and Th2 differentiation. To evaluate these possibilities, the mRNA expression level of SOCS genes after Ag stimulation was analyzed. At 24 h after Ag stimulation, mRNA was extracted, converted to cDNA, and subjected to RT-PCR analysis. As shown in Fig. 5C, there was no significant difference between ICOS+/+ and ICOS–/– CD4+ T cells in the expression of any SOCS genes. Thus, SOCS proteins are unlikely to be involved in this phenomenon.
Involvement of transcription factors in impaired Th2 differentiation of ICOS–/– naive CD4+ T cells
In early activation phase, several transcription factors, such as GATA-3 (4, 34), NF-B (9), and NFATc1 (6), are thought to be involved in the Th2 differentiation process. In particular, GATA-3 is recognized as a master regulator of Th2 differentiation (4) and its expression is known to be largely dependent upon the Stat6 pathway in primary responses (34, 35). Thus, we measured GATA-3 mRNA levels at 24 h after Ag stimulation from ICOS–/– and ICOS+/+ naive CD4+ T cells. As shown in Fig. 6A, GATA-3 transcription by ICOS–/– T cells was significantly lower than ICOS+/+ T cells. The presence of GATA-3 protein in the nucleus of these T cell populations was also tested at 48 h after Ag stimulation, and we found it to be reduced in ICOS–/– T cells (Fig. 6B). These results indicated that ICOS–/– T cells have a defect in the early induction of GATA-3 expression.
FIGURE 6. Early GATA-3 expression is impaired in ICOS–/– naive CD4+ T cells. A, DO11.10 naive CD4+ T cells from ICOS+/+ or ICOS–/– mice were stimulated with OVA peptide (0.1 μM). At 24 h after Ag stimulation, mRNA was extracted from isolated T cells, and the converted cDNAs were subjected to semiquantitative RT-PCR analysis (*, p = 0.014). B, DO11.10 naive CD4+ T cells were isolated and stimulated as described in A. After 24–48 h of Ag stimulation, nuclear protein was extracted from isolated T cells and subjected to Western blot analysis for translocation of indicated transcription factors. Data show representative results from one of two independent experiments.
Optimal GATA-3 induction in naive CD4+ T cells requires integration of both TCR-mediated and IL-4R-mediated signals. Recently, the importance of NF-B for optimal GATA-3 induction was reported, because GATA-3 expression and Th2 differentiation were specifically abrogated in p50–/– T cells and in cells where nuclear translocation of NF-B was inhibited (9). To test whether impaired GATA-3 induction in ICOS–/– T cells is the result of defective NF-B activation, RelA (p65), NF-B1 (p50), and c-Rel were analyzed at 24–48 h after Ag stimulation. As shown in Fig. 6B, their translocation level in ICOS–/– T cells did not differ from that of ICOS+/+ T cells, indicating that impaired GATA-3 induction in ICOS–/– T cells is not due to a defective NF-B activation, but a defective IL-4R/Stat6 pathway. It has been demonstrated that NFAT proteins regulate the promoters of multiple cytokine genes expressed in T cells. Among four NFAT family members, NFATc1 is considered to be a direct transcriptional activator of the IL-4 gene (6). Similar to NF-B, NFATc1 translocation in ICOS–/– T cells did not differ from ICOS+/+ T cells (Fig. 6B). Expression patterns of transcription factors in ICOS–/– and ICOS+/+ T cells after Ag stimulation are consistent with the conclusion that the defect of ICOS–/– T cells in Th2 differentiation may not be the result of IL-4 production, but of IL-4R-mediated signal transduction.
Discussion
In this report, we demonstrate that DO11.10 TCR Tg ICOS–/– mice have a defect in CD4+ T cells for Th2 differentiation by showing a reduced frequency of IL-4-producing cells (Fig. 2A), a reduction of total Th2-type cytokine production (IL-4, IL-5, and IL-13) (B), and impaired expression of Th2-specific proteins, such as GATA-3 and c-Maf (C), after Ag stimulation. Our results are consistent with a previous report that showed Ag-stimulated DO11.10 T cells were abrogated for the development of Th2 subsets by blocking ICOS-B7h interaction with ICOS-Ig treatment (36). Interestingly, despite the reduced frequency of IL-4-producing cells, IL-4 transcription and production levels of ICOS–/– CD4+ T cells in the early phase of Ag stimulation did not differ from that of ICOS+/+ CD4+ T cells (Fig. 3). This phenomenon is not specific for this particular gene-targeting strain, because ICOS ligation of naive CD4+ T cells from normal mice did not costimulate IL-4 transcription of these T cells in early phase of T cell activation (Fig. 1B). Because IL-4 is a key cytokine for the differentiation of Th2 cells, we considered the possibility that ICOS costimulation may influence IL-4R signaling. Indeed, it has been reported that the CD28 costimulatory signal affects IL-4 and IL-12 receptor reactivity (37, 38). To this end, we studied the role of ICOS ligation on IL-4R sensitivity using normal as well as DO11.10 TCR Tg ICOS–/– mice. We found that Stat6 phosphorylation and GATA-3 expression of normal naive CD4+ T cells were significantly enhanced by anti-ICOS mAb cross-linking (Fig. 1, C and D). Furthermore, the response of naive T cells from DO11.10 TCR Tg ICOS–/– mice to exogenous IL-4 showed a profound defect in the generation of IL-4-producing cells (Fig. 4, A and B), and IL-4-dependent Stat6 phosphorylation (Fig. 5A). In addition, GATA-3 expression in early T cell activation with OVA stimulation was impaired in naive ICOS–/– T cells (Fig. 6). There results suggested that ICOS costimulation accelerates Th2-meditated immune responses by positively controlling IL-4R signaling.
Recently, Nurieva et al. (39) reported that the defect of Th2 differentiation of ICOS–/– T cells was due to the reduction of primary IL-4 production. They showed that ICOS–/– CD4+ T cells exhibited impaired NFAT nuclear translocation in the early phase, and reduced expression of the c-Maf transcription factor, but not GATA-3, in the effector phase. Based on these results, they hypothesized that the ICOS costimulatory signal enhances NFATc1 expression and the production of IL-4 during early T cell differentiation, which regulates the expression of the IL-4-specific transcription factor c-Maf in the effector phase. However, we observed neither a defect of IL-4 production (Fig. 3) nor NFATc1 (Fig. 6B) nuclear translocation of ICOS–/– CD4+ T cells in early phase of Ag stimulation. In addition, effector T cells from ICOS–/– mice expressed significantly lower levels of GATA-3 as well as c-Maf (Fig. 2C).
Mak et al. (24) have recently reported that ICOS ligand-deficient (ICOSL–/–) mice have a similar defect to ICOS–/– mice in the development of Th2-mediated responses. They showed that the frequency of IL-4-producing cells in KLH-primed ICOSL–/– CD4+ T cells after Ag restimulation in vitro was significantly lower than that of ICOS+/+ T cells. However, IL-4 production of CD4+ T cells on a per-cell basis (fluorescence intensity) was not different between ICOSL+/+ and ICOSL–/– mice. We (Figs. 2A and 4A) as well as Tafuri et al. (23) have also shown that fluorescence intensity of IL-4 staining of Ag-responding ICOS–/– T cells was not significantly different from ICOS+/+ T cells. All these results support the notion that the ICOS-mediated costimulatory signal enhances the frequency of IL-4-positive Th2 cells, but not their capability for IL-4 production.
The reason why ICOS–/– naive T cells, described in the report by Nurieva et al. (39), were found to have a defect in primary IL-4 production is unclear. However, there are differences in their experimental systems that could explain the discrepancy with findings. First, whereas we study T cell responses against peptide Ags presented by normal APC, Nurieva et al. use anti-CD3 cross-linking for TCR stimulation and coculture APC as a source of costimulatory signals. It is conceivable that the first signal provided by TCR ligation with peptide/MHC and Ab is different in terms of quality as well as quantity. Second, we used Th2-dominant BALB/c background mice, whereas they used Th1-dominant C57BL/6 background mice. Third, T cells from their ICOS–/– mice have defects in IL-2 production and proliferative responses, whereas our mice do not have such defects. More precisely, IL-2 gene transcription (Fig. 3C) as well as IL-2 production (data not shown) by our ICOS–/– T cells after Ag stimulation was no different from those of ICOS+/+ T cells. Importantly, the ICOS–/– mice, which were generated independently by two other groups, showed similar phenotype to ours. Namely, Tafuri et al. (23) reported that T cells from their ICOS–/– mice proliferated normally to Con A, as well as anti-CD3 plus anti-CD28 stimulation. McAdam et al. (22) also showed that proliferation and IL-2 production of ICOS–/– CD4+ T cells against anti-CD3 plus APC stimulation and KLH stimulation was normal.
The molecular mechanism by which the ICOS-mediated signal enhances IL-4R activity is currently not known. Several studies indicate that the TCR- and/or costimulatory receptor-mediated signal modulates cytokine receptor reactivity during the early activation phase in a cytokine receptor expression level-independent manner (8, 37, 38, 40, 41). Zhu et al. (40) reported TCR engagement induced transient hyporesponsiveness of IL-4R signaling, and they demonstrated involvement of the ERK/MAPK and calcium pathways in this phenomenon. Thus, it is possible that decreased IL-4R responsiveness of ICOS–/– T cells may result from a perturbation in these signaling pathways. Because we show that NFATc1 nuclear translocation is comparable between ICOS+/+ and ICOS–/– T cells (Fig. 6B), the calcium pathway may be intact in ICOS–/– T cells. It will be of interest to determine whether ICOS deficiency affects the MAPK pathway.
It is well-established that the PI3K pathway plays an important role for CD28-mediated costimulation. We have previously shown that CD28 cross-linking enhances the IL-4R sensitivity (37). Because ICOS cytoplasmic region also has a PI3K binding motif, it is conceivable that PI3K pathway may be involved in up-regulation of IL-4R sensitivity by ICOS ligation. In fact, a number of studies also predict the role of PI3K pathway in Th differentiation. Suzuki et al. (42) showed that mice lacking PTEN, which is a phosphatase for PIP3, exhibit enhancement of PI3K signal pathway and increase in both Th1 and Th2 cytokine production. Arimura et al. (43) demonstrated that overexpression of Akt, a major downstream target of PI3K, resulted in amplification of both Th1 and Th2 development. Furthermore, Schaeffer et al. (44) have reported that Tec family tyrosine kinase Itk, which is also regulated by PI3K, has an important role for Th2 development. Using ICOS mutant mice, in which ICOS does not bind to PI3K, we are currently studying a specific role of ICOS-PI3K pathway for IL-4R signaling in Ag-specific T cell responses.
Another considerable possibility is that, similar to the IL-2R complex (45), the IL-4R may also be reorganized within the cell membrane upon T cell activation, and ICOS might have an important role for the formation of this functional receptor complex. Additional experiments are required to elucidate the molecular mechanism underlying this phenomenon.
From our results, it is expected that ICOS blockade effectively prevents the onset of Th2-dysregulated immune responses, such as allergy. In contrast, several reports indicated that the administration of ICOS-B7h blocking agents at an early phase of some immune responses exacerbates the pathological symptoms, and this may be due to augmenting Th1 responses (16, 20). Thus, although recent studies predict ICOS to be an attractive therapeutic target, it is important to understand the precise role of ICOS throughout the entire immune response. We believe our study will contribute to a better understanding of ways to approach ICOS-targeting therapeutic strategies.
Acknowledgments
We thank Dr. Motoko Kotani and Dr. Shuhei Ogawa for their discussion, and a member of Science Service, Inc., for care of experimental animals. We also gratefully acknowledge JT Pharmaceutical Frontier Research Laboratories for providing the ICOS–/– mice and anti-ICOS mAb.
Footnotes
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 a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology, Japan.
2 Address correspondence and reprint requests to Dr. Ryo Abe, Research Institute for Biological Sciences, Tokyo University of Sciences, 2669 Yamazaki, Noda, Chiba 278-0022, Japan. E-mail address: rabe@rs.noda.tus.ac.jp
3 Abbreviations used in this paper: Tg, transgenic; F, forward; R, reverse; SOCS, suppressor of cytokine signaling; ICOSL, ICOS ligand.
Received for publication September 22, 2004. Accepted for publication November 19, 2004.
References
Murphy, K. M., S. L. Reiner. 2002. The lineage decisions of helper T cells. Nat. Rev. Immunol. 2:933.
Kaplan, M. H., U. Schindler, S. T. Smiley, M. J. Grusby. 1996. Stat6 is required for mediating responses to IL-4 and for development of Th2 cells. Immunity 4:313.
Takeda, K., T. Tanaka, W. Shi, M. Matsumoto, M. Minami, S. Kashiwamura, K. Nakanishi, N. Yoshida, T. Kishimoto, S. Akira. 1996. Essential role of Stat6 in IL-4 signalling. Nature 380:627.
Zheng, W., R. A. Flavell. 1997. The transcription factor GATA-3 is necessary and sufficient for Th2 cytokine gene expression in CD4 T cells. Cell 89:587.
Kurata, H., H. J. Lee, A. O’Garra, N. Arai. 1999. Ectopic expression of activated Stat6 induces the expression of Th2-specific cytokines and transcription factors in developing Th1 cells. Immunity 11:677.
Yoshida, H., H. Nishina, H. Takimoto, L. E. Marengere, A. C. Wakeham, D. Bouchard, Y. Y. Kong, T. Ohteki, A. Shahinian, M. Bachmann, et al 1998. The transcription factor NF-ATc1 regulates lymphocyte proliferation and Th2 cytokine production. Immunity 8:115.
Hodge, M. R., A. M. Ranger, F. Charles de la Brousse, T. Hoey, M. J. Grusby, L. H. Glimcher. 1996. Hyperproliferation and dysregulation of IL-4 expression in NF-ATp-deficient mice. Immunity 4:397.
Yamashita, M., M. Kimura, M. Kubo, C. Shimizu, T. Tada, R. M. Perlmutter, T. Nakayama. 1999. T cell antigen receptor-mediated activation of the Ras/mitogen-activated protein kinase pathway controls interleukin 4 receptor function and type-2 helper T cell differentiation. Proc. Natl. Acad. Sci. USA 96:1024.
Das, J., C. H. Chen, L. Yang, L. Cohn, P. Ray, A. Ray. 2001. A critical role for NF-B in GATA3 expression and TH2 differentiation in allergic airway inflammation. Nat. Immunol. 2:45.
Carreno, B. M., M. Collins. 2002. The B7 family of ligands and its receptors: new pathways for costimulation and inhibition of immune responses. Annu. Rev. Immunol. 20:29.
Sharpe, A. H., G. J. Freeman. 2002. The B7-CD28 superfamily. Nat. Rev. Immunol. 2:116
Hutloff, A., A. M. Dittrich, K. C. Beier, B. Eljaschewitsch, R. Kraft, I. Anagnostopoulos, R. A. Kroczek. 1999. ICOS is an inducible T-cell co-stimulator structurally and functionally related to CD28. Nature 397:263.
Yoshinaga, S. K., J. S. Whoriskey, S. D. Khare, U. Sarmiento, J. Guo, T. Horan, G. Shih, M. Zhang, M. A. Coccia, T. Kohno, et al 1999. T-cell co-stimulation through B7RP-1 and ICOS. Nature 402:827.
Tamatani, T., K. Tezuka, N. Hanzawa-Higuchi. 2000. AILIM/ICOS: a novel lymphocyte adhesion molecule. Int. Immunol. 12:51.
Coyle, A. J., S. Lehar, C. Lloyd, J. Tian, T. Delaney, S. Manning, T. Nguyen, T. Burwell, H. Schneider, J. A. Gonzalo, et al 2000. The CD28-related molecule ICOS is required for effective T cell-dependent immune responses. Immunity 13:95.
Ogawa, S., G. Nagamatsu, M. Watanabe, S. Watanabe, T. Hayashi, S. Horita, K. Nitta, H. Nihei, K. Tezuka, R. Abe. 2001. Opposing effects of anti-activation-inducible lymphocyte-immunomodulatory molecule/inducible costimulator antibody on the development of acute versus chronic graft-versus-host disease. J. Immunol. 167:5741.
Gonzalo, J. A., J. Tian, T. Delaney, J. Corcoran, J. B. Rottman, J. Lora, A. Al-garawi, R. Kroczek, J. C. Gutierrez-Ramos, A. J. Coyle. 2001. ICOS is critical for T helper cell-mediated lung mucosal inflammatory responses. Nat. Immunol. 2:597.
Miyahira, Y., H. Akiba, S. H. Ogawa, T. Ishi, S. Watanabe, S. Kobayashi, T. Takeuchi, T. Aoki, K. Tezuka, R. Abe, et al 2003. Involvement of ICOS-B7RP-1 costimulatory pathway in the regulation of immune responses to Leishmania major and Nippostrongylus brasiliensis infections. Immunol. Lett. 89:193.
Ozkaynak, E., W. Gao, N. Shemmeri, C. Wang, J. C. Gutierrez-Ramos, J. Amaral, S. Qin, J. B. Rottman, A. J. Coyle, W. W. Hancock. 2001. Importance of ICOS-B7RP-1 costimulation in acute and chronic allograft rejection. Nat. Immunol. 2:591.
Rottman, J. B., T. Smith, J. R. Tonra, K. Ganley, T. Bloom, R. Silva, B. Pierce, J. C. Gutierrez-Ramos, E. Ozkaynak, A. J. Coyle. 2001. The costimulatory molecule ICOS plays an important role in the immunopathogenesis of EAE. Nat. Immunol. 2:605.
Dong, C., A. E. Juedes, U. A. Temann, S. Shresta, J. P. Allison, N. H. Ruddle, R. A. Flavell. 2001. ICOS co-stimulatory receptor is essential for T-cell activation and function. Nature 409:97.
McAdam, A. J., R. J. Greenwald, M. A. Levin, T. Chernova, N. Malenkovich, V. Ling, G. J. Freeman, A. H. Sharpe. 2001. ICOS is critical for CD40-mediated antibody class switching. Nature 409:102.
Tafuri, A., A. Shahinian, F. Bladt, S. K. Yoshinaga, M. Jordana, A. Wakeham, L. M. Boucher, D. Bouchard, V. S. Chan, G. Duncan, et al 2001. ICOS is essential for effective T-helper-cell responses. Nature 409:105.
Mak, T. W., A. Shahinian, S. K. Yoshinaga, A. Wakeham, L. M. Boucher, M. Pintilie, G. Duncan, B. U. Gajewska, M. Gronski, U. Eriksson, et al 2003. Costimulation through the inducible costimulator ligand is essential for both T helper and B cell functions in T cell-dependent B cell responses. Nat. Immunol. 4:765.(Masashi Watanabe, Shiho W)