|Home | About | Journals | Submit | Contact Us | Français|
CD4+ interleukin 17 (IL-17)-producing helper T cells (TH17 cells) are instrumental in the immune response to pathogens. However, an overactive TH17 response results in tissue inflammation and autoimmunity, and therefore it is important to identify the molecular mechanisms that control the development of TH17 cells. IL-2 suppresses such development, but how IL-2 production is actively suppressed during TH7 differentiation is not understood. Here we report that under TH17-polarizing conditions, the transcription factors STAT3 and AhR upregulated the expression of Aiolos, a member of the Ikaros family of transcription factors. Using Aiolos-deficient mice, we demonstrated that Aiolos silenced the Il2 locus, promoting TH17 differentiation in vitro and in vivo. Thus, we have identified a module in the transcriptional program of TH17 cells that actively limits IL-2 production and promotes their differentiation.
Cells of the TH17 subset of helper T cells are characterized by the production of interleukin 17 (IL-17), IL-17F, IL-21 and IL-22 and have emerged as a subset of effector CD4+ T cells with an important role in the control of specific pathogens as well as in the development of autoimmune diseases1. The activation of T cells in the presence of IL-6 (ref. 2) or IL-21 (ref. 3) and transforming growth factor-β1 (TGF-β1) promotes TH17 differentiation by transcription factor STAT3–dependent mechanisms4, whereas IL-21 (ref. 3) and IL-23 (ref. 5) expand TH17 cell populations and stabilize the phenotype of TH17 cells. The signals initiated in T cells by cytokine receptors induce and activate specific transcription factors that control the transcriptional program of TH17 cells. TH17 differentiation is driven by the transcription factors RORγt6 and RORα7. Other transcription factors, such as c-Maf and AhR, also participate in TH17 differentiation. AhR is known to control the expression of IL-21 and IL-22 and has an important role in TH17 differentiation in vivo and in vitro8–10.
Dysregulated population expansion of TH17 cells has been linked to the development of tissue inflammation and autoimmunity; thus, several mechanisms are in place to control the generation and activity of TH17 cells1. Although IL-2 and IL-21 belong to the same family of cytokines, they have opposing effects on TH17 differentiation: IL-21 acts as a growth factor and supports the population expansion of TH17 cells, whereas IL-2 inhibits TH17 differentiation11–15. IL-2 limits TH17 differentiation in vitro and in vivo by interfering with IL-6-dependent signaling events. IL-2 downregulates expression of the IL-6 receptor13 and also triggers the replacement of STAT3 with STAT5 on target DNA-binding sites in the Il17a-Il17f locus and in other genes required for TH17 differentiation12,15 and thus it interferes with the TH17 transcriptional program. Therefore, for TH17 differentiation to proceed unabated, IL-2 expression must be actively downregulated. The uptake of IL-2 by regulatory T cells (Treg cells) that express the transcription factor Foxp3 promotes TH17 differentiation in vivo and in vitro11,14. It is conceivable that in addition to the removal of IL-2 by cell-extrinsic mechanisms, cell-intrinsic mechanisms that actively limit the production of IL-2 are in place to promote TH17 differentiation.
Here we report significant upregulation of Aiolos, a member of the Ikaros family of transcription factors, during TH17 differentiation by a mechanism that involved STAT3 and AhR. Moreover, Aiolos bound to and silenced the Il2 locus, suppressing the production of IL-2 and promoting TH17 differentiation in vitro and in vivo. Thus, our results identify a transcriptional module that actively limits the production of IL-2 during TH17 differentiation to promote the development of TH17 cells.
Our analysis of the transcriptional programs of CD4+ T cell subsets showed that TH17 cells expressed Aiolos (encoded by Ikzf3; data not shown). We found that naive CD4+CD25−CD62L+CD44lo T cells differentiated in vitro into T helper type 1 (TH1) or TH2 cells had modest expression of Ikzf3, but Ikzf3 expression was substantially upregulated in TH17 cells differentiated with TGF-β1 plus IL-6 (Fig. 1a). We did not detect upregulation of genes encoding other members of the Ikaros family of transcription factors, including Ikaros itself, Helios, Eos and Pegasus16, in differentiating TH17 cells (Fig. 1b).
We also detected Ikzf3 expression in Foxp3+ Treg cells differentiated in vitro and T regulatory type 1 cells (Tr1 cells) induced with IL-27 (Supplementary Fig. 1). The role of transcription factors of the Ikaros family in the differentiation of Foxp3+ Treg cells and Tr1 cells has been investigated17–20; thus, in this study we focused on the role of Aiolos in the differentiation of TH17 cells. We first investigated the kinetics of Ikzf3 expression under TH17-polarizing conditions. Ikzf3 expression was substantially upregulated 6 h after activation in the presence of TGF-β1 and IL-6, and its expression remained very high throughout the TH17 differentiation (Fig. 1c). Upregulation of Ikzf3 expression preceded the induction of Il17a (Fig. 1c). The activation of T cells in the absence of polarizing cytokines (TH0 conditions) did not result in substantial upregulation of Ikzf3 expression (Fig. 1c). Together these data demonstrated that activation of naive CD4+ T cells under TH17-polarizing conditions resulted in the upregulation of Ikzf3 expression.
To determine if Aiolos has a role in the differentiation of TH17 cells, we investigated the effect of loss of Aiolos on TH17 differentiation using naive T cells from wild-type and Aiolos-deficient mice21. Naive Aiolos-deficient CD4+ T cells showed significant impairment in their differentiation into TH17 cells, as shown by their lower expression of Il17a and other genes encoding molecules linked to the TH17 lineage, such as Rorc, Rora, Maf, Ahr, Il17f and Il21, and diminished secretion of IL-17 into culture supernatants (Fig. 2a–c). Retrovirus-mediated overexpression of Ikzf3 in T cells activated under non–TH17-polarizing conditions did not result in upregulation of the expression of Rorc or Il17a (Supplementary Fig. 2), which suggested that Aiolos participated in but was not sufficient to induce TH17 differentiation. Conversely, Aiolos-deficient CD4+ T cells produced more interferon-γ (IFN-γ) than wild-type cells did when activated under TH1-polarizing conditions (Fig. 2b).
To characterize the relevance of Aiolos to TH17 differentiation during the course of an immune response, we studied the in vivo population expansion of TH17 cells after immunization. We immunized naive wild-type and Aiolos-deficient mice with myelin oligodendrocyte peptide (amino acids 35–55 (MOG(35–55)) emulsified in complete Freund’s adjuvant and, 7 d later, assessed the ability of lymph node cells to proliferate in response to MOG(35–55) and produce IFN-γ and IL-17. Aiolos-deficient mice had a slightly lower recall proliferative response to MOG(35–55) (Fig. 2d) and a significantly lower frequency of TH17 cells immediately after isolation, concomitant with a greater proportion of TH1 cells (Fig. 2e).
To analyze the pathogenicity of Aiolos-deficient TH17 cells, we used a model of passive transfer of experimental autoimmune encephalomyelitis (EAE). We immunized wild-type and Aiolos-deficient mice with MOG(35–55) and reactivated lymphocytes from the mice in vitro with MOG(35–55) in the presence of IL-12 or IL-23 to favor the population expansion of pathogenic TH1 cells or TH17 cells, respectively. Aiolos-deficient T cells reactivated under TH17-favoring conditions had significantly less ability to induce EAE after transfer into host mice deficient in recombination-activating gene 2 (Rag2−/− mice; Fig. 2f). Conversely, Aiolos-deficient T cells reactivated under TH1-favoring conditions showed significantly more encephalitogenic activity (Fig. 2g). Thus, Aiolos controlled the differentiation of encephalitogenic T cells. Together these data indicated that Aiolos participated in the differentiation of TH17 cells and limited the generation of TH1 cells in vitro and in vivo.
IL-2 limits the differentiation of TH17 cells in vitro and in vivo11–15. Several members of the Ikaros family have been shown to control the production of IL-2 in mouse T cells18 and human T cells17,22. Thus, we examined the expression of Il2 during TH17 differentiation. After an initial peak, the expression of Il2 was silenced in TH17 cells but not in TH0 cells (Fig. 3a). The arrest in Il2 transcription in TH17 cells coincided with the upregulation of Il17a (Fig. 3a), Rorc and Rora (Supplementary Fig. 3). Conversely, the continued production of Il2 in TH0 cells was concomitant with the increase in expression of Ifng (Fig. 3a).
Given the inhibitory effects of IL-2 on TH17 differentiation and the reported inhibitory effects of Ikaros proteins on the production of IL-2 by T cells17,18,22, we determined whether Aiolos promoted TH17 differentiation by inhibiting the production of IL-2. Aiolos-deficient T cells activated under TH17-polarizing conditions produced more IL-2 than did wild-type T cells activated in the same way (Fig. 3b–d), which correlated with less production of IL-17 by Aiolos-deficient cells (Fig. 3d). IL-2 induces expression of the α-chain of the IL-2 receptor (CD25), presumably as a part of a positive feedback loop23. Accordingly, we found upregulation of Il2ra in Aiolos-deficient T cells activated under TH17-polarizing conditions (Supplementary Fig. 3b,c). Thus, Aiolos-deficient T cells produced more IL-2 when activated under TH17-polarizing conditions, and the autocrine effects of IL-2 upregulated expression of the high-affinity IL-2 receptor, which made these cells more sensitive to the biological effects of IL-2. We also observed more production of IL-2 (Fig. 3c) and a greater frequency of IL-2+ T cells (Supplementary Fig. 4a,b) among naive Aiolos-deficient CD4+ T cells activated under TH1-polarizing conditions than among wild-type CD4+ T cells activated under TH1-polarizing conditions. In addition, Aiolos-deficient CD4+ T cells activated under TH1-polarizing conditions had higher expression of Tbx21 (which encodes T-bet) and CD25 than wild-type cells activated under TH1-polarizing conditions had (Supplementary Fig. 4c–e). Together these data suggested that Aiolos negatively regulated the production of IL-2.
To determine whether the dysregulated production of IL-2 in Aiolos deficient CD4+ T cells under TH17-polarizing conditions was responsible for the diminished differentiation of these cells along the TH17 pathway, we investigated the differentiation of wild-type and Aiolos-deficient naive T cells under TH17 conditions in the presence of blocking antibody to IL-2 or isotype-matched control antibody. The antibody to IL-2 abrogated the deficit in the production of TH17 cells by Aiolos-deficient T cells (Fig. 3e), whereas a blocking antibody to IFN-γ had no substantial effect on the TH17 differentiation of Aiolos-deficient cells (Supplementary Fig. 3d). Those observations were in agreement with the lack of upregulation of Tbx21 in Aiolos-deficient T cells activated under TH17-polarizing conditions (Fig. 2c) and demonstrated that the diminished differentiation of Aiolos-deficient T cells into TH17 cells in vitro did not result from the inhibitory effects of IFN-γ on TH17 differentiation24,25. Antibody to IL-2 abrogated the enhanced production of IFN-γ and the upregulation of Tbx21 and Il2ra in Aiolos-deficient T cells activated under TH1-polarizing conditions (Supplementary Fig. 4f–h). Together these data demonstrated that Aiolos promoted TH17 differentiation by limiting the production of IL-2.
Members of the Ikaros family of transcription factors regulate IL-2 expression in mouse T cells18 and human T cells17,22 by interacting with the locus encoding IL-2 and triggering epigenetic changes that limit its accessibility to other transcription factors. Thus, we searched the mouse Il2 promoter for Aiolos-binding sites and found a putative Aiolos-binding element 13 base pairs upstream of the transcription start site in the Il2 promoter (Fig. 4a). To determine whether Aiolos bound to its putative target sequence in TH17 cells, we used electrophoretic mobility-shift assay (EMSA) with a double-stranded oligonucleotide composed of the Aiolos-binding element in the Il2 promoter to probe nuclear extracts from TH17, TH1 or TH0 cells differentiated in vitro. Nuclear extracts of TH17 cells bound the putative Aiolos-binding site, but those of TH0 or TH1 cells did not (Fig. 4b). The interaction between the radiolabeled oligonucleotide and the nuclear extracts of TH17 cells was lower after preincubation with antibody to Aiolos but not after preincubation with control antibody to tubulin (Fig. 4b). Antibodies to proteins of the Ikaros family disrupt the oligomeric form responsible for DNA binding in EMSA26, which suggests that Aiolos in the nuclear extracts of TH17 cells interacted with its consensus binding element in the Il2 promoter. In addition, chromatin immuno-precipitation (ChIP) assays showed that Aiolos interacted with its binding site in the Il2 promoter in TH17 cells but not in TH1 or TH0 cells (Fig. 4c). Thus, Aiolos interacted with the Il2 promoter during TH17 differentiation.
Transcription factors of the Ikaros family control the epigenetic status of their target genes18,22,27. Chromatin modifications have an important role in the regulation of gene expression; the presence of acetylated histone H3 (H3ac) and acetylated histone H4 (H4ac), and of histone H3 trimethylated at Lys4 (H3K4me3), has been linked to the active transcription of genes, whereas the presence of H3K9me3 or H3K27me3 is associated with the repression of gene expression14. Thus, to investigate the molecular basis of the suppression of Il2 expression by Aiolos in TH17 cells, we analyzed the abundance of H3K4me3, H3K9me3, H3K27me3, H3ac and H4ac in the Il2 promoter of wild-type and Aiolos-deficient T cells polarized in vitro into TH17 cells. Aiolos-deficient TH17 cells had less H3K9me3 and H3K27me3 (Fig. 4d) but more H3K4me3 and more H3ac and H4ac (Fig. 4e) than did wild-type TH17 cells. Thus, in Aiolos-deficient TH17 cells, the Il2 promoter was characterized by epigenetic modifications associated with active gene expression.
To determine the functional consequences of the interaction of Aiolos with the Il2 promoter, we transfected reporter constructs containing the firefly luciferase gene under the control of the IL2 promoter into human Jurkat T cells. Transfection of this luciferase reporter together with a construct encoding Aiolos resulted in substantial and dose-dependent repression of the transcriptional activity of the IL2 promoter (Fig. 4f). Transfection of expression vectors for Aiolos into Jurkat cells also induced inhibition of expression of endogenous IL2 (data not shown). Together these data demonstrated that Aiolos induced epigenetic modifications that silenced Il2 expression in TH17 cells.
Activation of STAT3 initiates TH17 differentiation in response to polarizing cytokines4. Detailed kinetic analysis of gene expression showed that upregulation of Stat3, Rorc and Ahr preceded the expression of Ikzf3 during TH17 differentiation (Fig. 5a). We detected phosphorylation of STAT3, which is known to control its own transcription as well as the expression of Rorc and Ahr in TH17 cells4,28,29, as early as 5 min after the initiation of TH17 differentiation (Supplementary Fig. 5). Genome-wide ChIP analyses have reported interaction of STAT proteins with the Ikzf3 locus30. Moreover, the transcription factor AhR is important in the development of TH17 cells8,10. Thus, to investigate the mechanisms that control Aiolos expression in TH17 cells, we searched for STAT3- and AhR-binding sites in the mouse Ikzf3 promoter and found three putative STAT3-responsive elements and three putative AhR-binding sites (Fig. 5b). Whole-genome ChIP analysis of RORγt followed by sequencing did not show enrichment in the binding of RORγt to the Ikzf3 promoter (data not shown).
ChIP assays showed interaction of AhR (Fig. 5c) and STAT3 (Fig. 5d) with their binding sites in the Ikzf3 promoter in TH17 cells but not in TH0 cells (used as a control) or when we used isotype-matched control antibodies. To determine the functional consequences of the interaction of AhR and STAT3 with the Ikzf3 promoter in TH17 cells, we assessed at the effect of AhR and STAT3 on transactivation of the IKZF3 promoter in luciferase reporter assays. Transfection of a reporter containing the firefly luciferase gene under the control of the IKZF3 promoter together with a construct encoding AhR or constitutively activated STAT3 into Jurkat T cells resulted in substantial transactivation of the IKZF3 promoter (Fig. 5e).
Retrovirus-mediated overexpression of AhR in activated T cells resulted in small but significant upregulation of Ikzf3 expression and concomitantly lower Il2 expression (P < 0.05; Supplementary Fig. 6), which suggested that additional signaling pathways were needed to make the Ikzf3 locus accessible to AhR during TH17 differentiation. To further investigate the role of AhR in the control of Ikzf3 expression, we used the AhR-specific ligand FICZ, which activates AhR in TH17 cells10,31. Activation of AhR with FICZ during TH17 differentiation resulted in significant upregulation of Ikzf3 expression (Fig. 5f). Conversely, inhibition of AhR signaling with the specific inhibitor CH-223191 led to significantly lower Ikzf3 expression during TH17 differentiation (Fig. 5f). We obtained similar results with Jurkat cells activated in the presence of FICZ or CH-223191 (data not shown). Together these data demonstrated that STAT3 and AhR regulated Ikzf3 expression in TH17 cells.
IL-2 has an important role in the phenotype and abundance of memory T cells32. Because our data showed that Aiolos limited the generation of TH17 cells in vivo by silencing Il2 expression, we investigated the role of Aiolos in the generation of TH17 cells under physiological conditions. We found a significantly greater frequency and quantity of CD62LloCD44hi CD4+ memory T cells in Aiolos-deficient mice than in wild-type mice (Fig. 6a–c), which suggested that the population expansion and/or survival of memory T cells was enhanced in Aiolos-deficient mice. Moreover, Aiolos-deficient mice had a significantly lower frequency of IL-17+ CD4+ memory T cells (Fig. 6d,e) and lower RORγt expression in those cells (Fig. 6f) than did wild-type mice, concomitant with a greater frequency of IFN-γ-producing memory TH1 cells and higher expression of Tbx21 in Aiolos-deficient mice than in wild-type mice (Fig. 6d–f). Thus, Aiolos limited the primary differentiation of TH17 cells and their incorporation into the pool of CD4+ memory T cells under physiological conditions.
TH1 cells33 and TH17 cells34 are polarized in the gut in response to signals provided by the commensal flora. IFN-γ-producing TH1 cells have a pathogenic role in the induction of colitis after transfer into Rag2−/− mice, but TH17 cells do not35. IL-2 limits the differentiation of TH17 cells in the gut12,15; thus, we investigated the role of Aiolos in the differentiation of gut TH17 cells. We transferred naive wild-type or Aiolos-deficient CD4+ T cells into Rag2−/− mice and analyzed the production of IL-17 and IFN-γ in recipient mice 1 month after that transfer. T cells isolated from the mesenteric lymph nodes of mice reconstituted with Aiolos-deficient T cells produced less IL-17 than did those from mice reconstituted with wild-type cells (Fig. 7a). Consistent with our in vitro data, the diminished production of IL-17 was concomitant with enhanced production of IFN-γ and IL-2 (Fig. 7a), which led to a significantly higher ratio of IFN-γ+ cells to IL-17+ cells among CD4+ T cells (Fig. 7a,b).
To assess the relevance of silencing of the Il2 locus by Aiolos in the context of inflammation, we studied the development of colitis in Rag2−/− mice given adoptive transfer of CD4+CD25−CD45RBhi T cells. In this model, the transferred cells differentiate into TH1 cells and TH17 cells and drive the development of inflammatory colitis. IL-2 is colitogenic in this experimental model36, and increases in the ratio of TH1 cells to TH17 cells37 have been linked to a more-severe wasting disease. We adoptively transferred purified CD4+CD25−CD45RBhi T cells from wild-type or Aiolos-deficient mice into Rag2−/− mice. Recipients of Aiolos-deficient T cells developed a more-aggressive wasting disease than that of recipients of wild-type T cells, as shown by their accelerated weight loss (Fig. 7c,d), shorter colons (Fig. 7e) and inflammatory histological scores (Fig. 7f). The greater colitogenic activity of Aiolos-deficient T cells was linked to more production of IFN-γ and a higher ratio of IFN-γ+ cells to IL-17+ cells in the CD3+CD4+ T cell compartment of the lamina propria (Fig. 7g,h) and less production of IL-17 by gut T cells (Fig. 7g). We found no significant difference between recipients of wild-type or Aiolos-deficient T cells in their production of IL-22 or in the Foxp3+ Treg cell or IL-10+ T cell compartments (Supplementary Fig. 7a–c). To determine whether the differences in disease induction were dose dependent, we transferred fewer purified CD4+ CD25− CD45RBhi T cells from wild-type or Aiolos-deficient mice (1 × 105 cells per mouse) into Rag2−/− mice. Even under these experimental conditions, in which the recipient mice required more time to develop wasting disease, the Aiolos-deficient T cells induced a more-severe disease than did the wild-type T cells. Together these data showed that Aiolos controlled the ratio of TH1 cells to TH17 cells in vivo and, consequently, the development of tissue inflammation.
IL-2 has been identified as a negative regulator of TH17 differentiation in vivo and in vitro12. In this study, we have identified the transcription factor Aiolos as an active inhibitor of IL-2 production and thus a promoting factor for TH17 differentiation. We found that under TH17-polarizing conditions, AhR together with STAT3 induced Aiolos expression. Aiolos then bound to the Il2 promoter and induced chromatin modifications that resulted in Il2 silencing. Aiolos-deficient naive CD4+ T cells produced more IL-2 and showed impaired differentiation into TH17 cells, which was reversed by blockade of IL-2 function. Thus, Aiolos promoted TH17 differentiation by actively silencing Il2 transcription under TH17-polarizing conditions. In conclusion, we have identified a module in the transcriptional program of TH17 cells that limits the autocrine inhibitory effects of IL-2 and thereby promotes TH17 differentiation.
In addition to finding Aiolos expression in TH17 cells, we detected Aiolos expression in Foxp3+ Treg cells and IL-27-induced Tr1 cells. Because IL-27 suppresses IL-2 production in Tr1 cells38, our data would suggest that this is most probably achieved by the induction of Aiolos in these cells. Similarly, there is strict control of the production of IL-2 by Treg cells such that they are entirely dependent on exogenous IL-2 for population expansion. Aiolos17 and Eos18, another member of the Ikaros family, participate in the control of endogenous IL-2 production in these cells.
IL-2 is a potent endogenous T cell–driven inhibitor of TH17 differentiation, and IL-2 concentrations must be controlled to promote TH17-type immunity12,13,15. In vivo, the availability of IL-2 (and consequently TH17 differentiation) can be regulated by control of the production of IL-2 and/or its consumption. Foxp3+ Treg cells have been shown to promote TH17 differentiation in vivo and in vitro by consuming IL-2 (refs. 11,14). Our data have demonstrated that under TH17-polarizing conditions, the production of IL-2 was limited by a TH17 cell–intrinsic mechanism mediated by Aiolos. Thus, the silencing of Il2 expression by Aiolos complemented the consumption of IL-2 by Foxp3+ Treg cells to promote the differentiation of TH17 cells during the course of an immune response. Moreover, the silencing of Il2 expression in TH17 cells by Aiolos might be complemented by other mechanisms that operate in T cells to control the production of IL-2. One such mechanism is the inhibition of activation of the transcription factor NF-κB and Il2 expression by the transcription factors Foxo1 and Foxo3a, which have high expression in resting T cells and whose expression is partially controlled by STAT3 in activated T cells39.
IL-2 and its receptor have been genetically linked to many human and mouse autoimmune diseases. The development of type 1 diabetes in mice of the nonobese diabetic (NOD) strain shows linkage to the Idd3 locus, which includes Il2 (ref. 40). NOD mice have less IL-2, an observation initially linked to diminished function of Foxp3+ Treg cells40. However, the lower expression of IL-2 in NOD mice is correlated with more circulating TH17 cells41, which suggests that an imbalance in the production of IL-2 and its control over TH17 differentiation, in addition to its effects on Foxp3+ Treg cells, has important consequences for the development of autoimmunity. Thus, in vivo, mechanisms that regulate Il2 expression might influence the development of autoimmunity at different levels. First, they can control the activity40 and homeostasis42 of Foxp3+ Treg cells. Second, they can regulate the generation of TH17 cells and alter the balance between TH1 cells and TH17 cells12,13,15,39 (as shown here). Third, initial alterations in the balance between TH1 cells and TH17 cells might be amplified by the inhibitory effects of IFN-γ and IL-17 on TH17 differentiation24 and TH1 differentiation35, respectively.
IL-21 belongs to the IL-2 family of cytokines; however, IL-2 and IL-21 have opposite effects on TH17 cells: IL-2 inhibits the differentiation of TH17 cells12, whereas IL-21 promotes the generation and population expansion of TH17 cells3. The Il2 and Il21 loci are in close proximity and are inherited as a haplotype on chromosome 3 in mice and chromosome 4 in humans; thus, it has been proposed that they might be regulated by competition for alternative use of regulatory elements43. We did not find binding sites for Ikaros proteins in the Il21 promoter (data not shown). On the surface, such findings might suggest that members of the Ikaros family of transcription factors do not regulate Il21 expression. However, AhR, which has high expression by TH17 cells10,31, transactivates both the Il2 promoter44 and the Il21 promoter45. Therefore, it is possible that by restricting the accessibility of target DNA sequences in the Il2 promoter of TH17 cells, Aiolos directs AhR to induce the expression of Il21 but not of Il2 in TH17 cells.
In conclusion, we found that Aiolos participated in a transcriptional module in TH17 cells that limited the autocrine inhibitory effects of IL-2 on TH17 differentiation. This TH17 cell–intrinsic mechanism of the regulation of IL-2 complemented the consumption of IL-2 by Foxp3+ Treg cells to promote TH17 differentiation in vivo and in vitro. As TH17 cells have an important role in the immune response to infection and autoimmune disorders1, characterization of the molecular mechanisms that control the development of TH17 cells might provide new targets for the therapeutic manipulation of effector T cell responses.
Aiolos-deficient mice21, Rag2−/− mice, RORγt-deficient mice and C57BL/6 mice were kept in a conventional, pathogen-free facility at the Harvard Institutes of Medicine. All experiments were carried out in accordance with guidelines prescribed by the Institutional Animal Care and Use Committee of Harvard Medical School.
Naive CD4+ T cells (CD4+CD44loCD62LhiCD25−) were purified by flow cytomtery from the spleen and lymph nodes; the purity of isolated T cell populations routinely exceeded 98%. Naive T cells were stimulated with plate-bound anti-CD3 (2 μg/ml; 145-2C11; Biolegend) and anti-CD28 (2 μg/ml; PV-1; Abcam), in the presence of IL-12 (30 ng/ml) and anti-IL-4 (10 μg/ml; C17.8; Biolegend) for the generation of TH1 cells; IL-4 (30 ng/ml) and anti-IFN-γ (10 μg/ml; XMG1.2; Biolegend) for the generation of TH2 cells; or IL-6 (30 ng/ml), TGF-β1 (3 ng/ml) and anti-IL4 and anti-IFN-γ for the generation of TH17 cells. Mouse IL-4, IL-6, IL-12, anti-IFN-γ and anti-IL-2 (404-ML-010, 406-ML-025, 419-ML-050, MAB485 and MAB702, respectively) were from R&D Systems; TGF-β1 (PHG9204) was from Invitrogen. FICZ (6-formylindolo[3,2-b]barbazole) was from Enzo Life Sciences and CH-223191 was from EMD Chemicals.
RNA was extracted with RNAeasy columns (Qiagen), then cDNA was prepared and used as a template for real-time PCR. Expression was normalized to Gapdh. All primers and probes were from Applied Biosystems (assay identifier in parentheses): Gata3 (Mm01337569_m1), Tbx21 (Mm00450960_m1), Rorc (Mm00441144_g1), Ikzf1 (1187884), Ikzf2 (1263584), Ikzf3 (Mm01304161_m1), Ikzf4 (1160002), Ikzf5 (Mm01131648_g1), Ifng (Mm00801778_m1), Il17a (Mm00439619_m1), Il2 (Mm00434256_m1), Ahr (Mm01291777_m1), Stat3 (Mm01219775_m1), Gapdh (Mm99999915_g1), RORC (Hs01076112_m1), IL17A (Hs00936345_m1), AHR (Hs00907316_m1), IKZF3 (Hs00232635_m1), B2M (4326319E) and GAPDH (4310884E).
Cytokines were measured by ELISA31. For intra-cellular cytokine staining, cells were stimulated for 4 h with PMA (phorbol 12-myristate 13-acetate; 50 ng/ml; Sigma), ionomycin (1 μg/ml; Sigma) and monensin (GolgiStop; 1 ml/ml; BD Biosciences). After staining of surface markers, cells were fixed and made permeable according to the manufacturer’s instructions (BD Biosciences). All antibodies to cytokines were from Biolegend.
Lymph-node cells obtained from mice at day 7 after immunization were restimulated for 3 d in the presence of MOG(35–55) and incorporation of [3H]thymidine was measured as described9.
The IL2 and IKZF3 promoter reporters were from SwitchGear Genomics. Vectors encoding constitutively activated STAT3, AhR and Aiolos were from GeneCopoeia or Addgene.
TH0, TH1 and TH17 cells were differentiated for 48 h, then nuclear extracts were prepared and normalized for protein content with the Bradford Assay. For construction of the EMSA probe, two complementary oligonucleotides containing the Aiolos-binding site in the Il2 promoter (5′-AAATTGC CTCCCATGCTGAAGAGCTGCCTA-3′ and 5′-TAGGCAGCTCTTCAGCAT GGGAGGCAATTT-3′) were annealed and radiolabeled with [γ-32P]dATP. Nuclear extracts were incubated for 30 min at 25 °C with radiolabeled DNA probe in binding buffer (10 mM Tris, pH 7.9, 50 mM NaCl, 2 mM MgCl2, 1 mM EDTA, 1 mM DTT, 5% glycerol (vol/vol) and 25 ng/μl poly(dI:dC)). DNA-protein complexes were fractionated by electrophoresis through a 6% nondenaturing polyacrylamide gel. For identification of binding specificity, nuclear extracts were preincubated with anti-Aiolos (8B2; Active Motif).
Cells were crosslinked with 1% formaldehyde and lysed with 0.35 ml lysis buffer (1% SDS, 10 mM EDTA and 50 mM Tris-HCl, pH 8.1) containing 1× protease inhibitor ‘cocktail’ (Roche Molecular Biochemicals). Chromatin was sheared by sonication and supernatants were collected after centrifugation and diluted in buffer (1% Triton X-100, 2 mM EDTA, 150 mM NaCl nd 20 mM Tris-HCl, pH 8.1). Antibody (5 μg) was prebound for a minimum of 4 h to protein A and protein G Dynal magnetic beads (Invitrogen) and was washed three times with ice-cold 5% BSA in PBS, and then was added to the diluted chromatin and immunoprecipitated overnight (antibodies described below). Magnetic bead–chromatin complexes were then washed six times in RIPA buffer (50 mM HEPES, pH 7.6, 1 mM EDTA, 0.7% sodium deoxycholate, 1% Nonidet-P40 and 0.5 M LiCl), followed by three times with Tris-EDTA buffer. Immunoprecipitated chromatin was then extracted with 1% SDS, 0.1 M NaHCO3 and heated for at least 6 h at 65 °C for reversal of the formaldehyde crosslinking. DNA fragments were purified with a QIAquick DNA purification Kit (Qiagen) and analyzed by SYBR Green real-time PCR (primers described below). The following antibodies were used for ChIP: anti-AhR (N-19; Santa Cruz Biotechnology), anti-Aiolos (ab69930; Abcam), antibody to phosphorylated STAT3 (9134; Cell Signaling Technology), antibody to acetylated H3 (06-599; Millipore), antibody to acetylated H4 (06-866; Millipore), anti-H3K4me3 (ab8580; Abcam), anti-H3K9me3 (ab8898; Abcam) and anti-H3K27me3 (6002; Abcam). The following primer pairs were used: Il2 forward, 5′-TAAGTGTGGGCTAACCCGA-3′, and reverse, 5′-CAAGGAGCACAAGTGTCAATGTGA-3′; ikzf3-1 forward, 5′-GGCTTGACATTCAAAAGTGGGTGCG-3′′, and reverse, 5′-CTGGCT GGCGGCCGCTTTAAAG-3′; ikzf3-2 forward, 5′-TCCTTCCGTGTCAGGCA AATCTGGA-3′, and reverse, 5′-GGGTTCCGCACCCACTTTTGAATGT-3′; ikzf3-3 forward, 5′-GTTCTGCCTTCAGTTCCTCTCCCCC-3′, and reverse, 5′-GAAGGTGTTGGTTCACAGAGGCGTG-3′.
Jurkat T cells were grown in DMEM supplemented with 10% FBS, transfected with TransIT-Jurkat Transfection Reagent (Mirus) and 2–5 μg plasmid and activated with PMA and ionomycin, then luciferase activity was analyzed 48 h after transfection with a dual luciferase assay kit (New England Biolabs).
Naive CD62LhiCD25−CD44loCD4+ T cells were transduced with retroviruses as described45 after activation with plate-bound anti-CD3 and anti-CD28.
Adoptive transfer of cells to induce EAE was done as described45. Wild-type and Aiolos-deficient mice were immunized with 100 μg MOG(35–55) (MEVGWYRSPFSRVVHLYRNGK) in complete Freund’s adjuvant. Draining lymph nodes and spleens were collected 11 d after immunization and were cultured for 3 d with MOG(35–55) (20 ug/ml) and carrier-free recombinant IL-12 (10 ng/ml; R&D Systems) or IL-23 (10 ng/ml; R&D Systems). Subsequently, 20 × 106 cells were transferred intravenously into naive Rag2−/− mice, which were injected intraperitoneally with 200 ng pertussis toxin on days 0 and 2.
CD4+ T cells were purified from the spleen and lymph nodes of wild-type and Aiolos-deficient mice by negative selection with magnetic beads (Miltenyi Biotec), and naive CD4+CD25−CD45RBhi T cells were purified by sorting with flow cytometry. CD4+CD45RBhiCD25 naive T cells (2 × 105) were injected into age- and sex-matched Rag2−/− mice and weight loss was monitored.
Prism software (GraphPad Software) was used for statistical analyses. P values of less than 0.05 were considered significant.
We thank D. Kozoriz for cell sorting, and B. Waksman and I.R. Cohen for discussions. Supported by the US National Institutes of Health (P01AI073748, P01NS038037, P01AI056299 and R01NS030843 to V.K.K.; and AI075285 and AI093903 to F.J.Q.), the National Multiple Sclerosis Society (RG4111A1 to F.J.Q.), the Harvard Medical School Office for Diversity and Community Partnership (F.J.Q.) and the National Multiple Sclerosis Society (FG1850-A-1 to H.J.).
Note: Supplementary information is available in the online version of the paper.
AUTHOR CONTRIBUTIONSH.J., F.J.Q., M.N. and M.R. did in vitro and in vivo experiments; M.N., A.Y., E.J.B., D.K., C.Z. and S.X. did in vitro experiments; K.G. and J.S. provided reagents; F.J.Q. wrote the manuscript; and V.K.K. and F.J.Q. supervised the study and edited the manuscript.
COMPETING FINANCIAL INTERESTS
The authors declare no competing financial interests.
Reprints and permissions information is available online at http://www.nature.com/reprints/index.html.