- Ti-Treg cell:
TGF-β-induced regulatory T cell
Regulatory T (Treg) cells play a key role in the maintenance of the immune system homeostasis. Treg cells can be generated in the periphery under control of TGF-β, a cytokine involved in the negative control of the immune system. However, TGF-β cooperates with IL-6 in the generation of Th17 cells, a novel class of effector cells involved in numerous inflammatory diseases, including colitis. Therefore, TGF-β emerges as a mediator of both anti-inflammatory and pro-inflammatory processes, depending on the local cytokine milieu. Here we demonstrate that IL-21, a type-1 cytokine produced by T cells and involved in the pathogenesis of immune-mediated diseases, prevents the TGF-β-dependent expression of FoxP3, the master regulator of Treg cell commitment, and the induction of suppressive capacity in naive CD4+ T cells, while promoting the differentiation of Th17 cells. In vivo, CD4+ naive T cells activated in the presence of TGF-β and IL-21 failed to suppress colitis while inducing an inflammatory response characterized by high levels of IL-17 and RORγt, the transcription factor expressed by Th17 cells. Therefore, IL-21 emerges as a key modulator of TGF-β signaling, leading to the reciprocal differentiation of Treg and Th17 cells.
Maintenance of immune homeostasis depends on the tightly controlled balance between activating and suppressing molecular events. Thus, defective counter-regulation and/or excessive activation of effector cells may each promote tissue-damaging immune responses 1–3.
Interleukin (IL)-21 is a T cell-derived cytokine that signals through a receptor composed of a specific subunit, termed IL-21R, and the common γ-chain subunit, shared with IL-2, IL-4, IL-7, IL-9, IL-13, and IL-15 receptors. IL-21R is highly expressed on T and B lymphocytes and NK cells, all of which functionally respond to IL-21 4. Previous studies have shown that IL-21 is important in the regulation of immunoglobulin synthesis by plasma cells, cytotoxic activity of NK and CD8+ T cells, and cytokine production by CD4+ T cells. Enhanced expression of IL-21 and/or IL-21R has been documented in various diseases associated with a defective capacity of counter-regulatory mechanisms to dampen T cell-mediated inflammatory responses 5, 6. IL-21 also appears to positively regulate the homeostatic expansion of autoreactive T cells in murine models of diabetes 2. Overall, these findings suggest that IL-21 may play an important role in the induction and/or perpetuation of T cell-dependent inflammatory processes.
CD4+CD25+ regulatory T (Treg) cells are a specialized class of lymphocytes characterized by the expression of the forkhead transcription factor FoxP3, the master regulator of the regulatory cell commitment 7. Naturally occurring Treg cells are produced by the thymus, but Treg cells can also be induced in the periphery by TGF-β-mediated conversion of CD4+CD25– T cells into FoxP3-positive CD4+CD25+ T cells (now termed Ti-Treg cells) 8, 9. Functionally, Treg cells can suppress T cell activation and proliferation in vitro and in vivo10. A reduced number and/or activity of Treg cells is associated with the development and/or progression of various inflammatory diseases 7. Recent evidence has also suggested that inflammatory stimuli can suppress either the generation or activity of Ti-Treg cells 11, 12. Here, we demonstrate that the pro-inflammatory cytokine IL-21 acts as an important immunomodulator by preventing the peripheral generation of Ti-Treg cells and promoting the differentiation of Th17 cells, a novel class of effector T cells characterized by the expression of the transcription factor RORγt and secretion of IL-17, IL-6 and TNF-α. We also demonstrate that Ti-Treg cells generated in the presence of IL-21 fail to suppress colitis in a murine adoptive transfer model while promoting an immune response characterized by high levels of IL-17.
IL-21 suppresses TGF-β-mediated induction of FoxP3 in CD4+CD25– cells
We and others have previously demonstrated that polyclonally activated CD4+CD25– T cells differentiate into FoxP3-expressing cells in the presence of TGF-β 8, 9. To assess whether IL-21 may interfere with this process, CD4+CD25– T cells were sorted from BALB/c splenocytes and polyclonally activated in the presence of TGF-β with or without recombinant IL-21. As expected, TGF-β markedly enhanced FoxP3 expression, and no change in FoxP3 RNA expression was seen in cells activated in the presence of IL-21 alone. In contrast, IL-21 dose-dependently reduced the TGF-β-mediated FoxP3 induction (Fig. 1A, B). A marked reduction in the percentage of FoxP3-positive cells was also seen in IL-21-treated Ti-Treg cell cultures (Fig. 1C).
IL-21 is known to negatively modulate the in vitro activation of antigen-presenting cells (APC) 13, so we next investigated whether IL-21 could suppress FoxP3 expression in a more physiologic setting, where Ti-Treg cells were generated in the presence of APC with or without IL-21. The results in Fig. 1D show that IL-21 was also able to block the induction of Ti-Treg cells in this experimental setting.
We then examined whether IL-21 reverses the inhibitory effect of TGF-β on T cell cytokine production. CD4+CD25– T cells were activated in the presence or absence of TGF-β with or without IL-21. As expected, TGF-β significantly inhibited the secretion of TNF-α, IL-6, and IFN-γ, but this was not counteracted by IL-21. Moreover, in the same cell cultures, the ability of TGF-β to inhibit the expression of CD45RB and CD69 was not altered by IL-21 (not shown). The increase in membrane CD44 and CD62L on T cells in TGF-β-treated cultures was also not affected by IL-21 (not shown). Overall, these findings indicate that IL-21 selectively inhibits the TGF-β-mediated induction of FoxP3, rather than blocking the activity of this cytokine.
At what stage does IL-21 inhibit Treg cell induction?
IL-21 is a growth factor for CD4+ effector T cells. Therefore, it was possible that the reduced expression of FoxP3 in IL-21-treated cultures of CD4+CD25– T cells activated with anti-CD3 Ab, anti-CD28 Ab, and TGF-β was because of preferential expansion of FoxP3-negative T cells. We first examined FoxP3 in cultures of CD4+CD25– T cells activated in the presence or absence of IL-21 for different times. As shown in Fig. 2A, the inhibition of FoxP3 RNA in IL-21-treated cultures occurred early (i.e. 12 and 24 h) and persisted. Second, we evaluated whether IL-21 exerts its effects in already established cultures of CD4+CD25– T cells activated with anti-CD3 Ab, anti-CD28 Ab and TGF-β. IL-21 was added 0, 12, and 48 h after the anti-CD3 + CD28 Ab activation of T cells in the presence of TGF-β. Importantly, IL-21 caused a reduction of FoxP3 RNA when added simultaneously with TGF-β as well as 12 and 48 h after the cell activation (Fig. 2B). These effects were paralleled by a reduction in the percentage of FoxP3-positive cells even when IL-21 was added 48 h after the activation of cells in the presence of TGF-β (Fig. 2C). Finally, we examined whether IL-21 differently modulated the proliferation of FoxP3-positive and -negative T cells. Freshly isolated CD4+CD25– T cells were stained with CFSE and activated in the presence of TGF-β, with or without IL-21. After 5 days, cells were stained for FoxP3 and the number of divisions was evaluated in both the FoxP3-positive and FoxP3-negative T cell fractions. While the proliferating fraction of FoxP3-positive cells was slightly enhanced by IL-21, there was a marked proliferation (more than twofold) of FoxP3-negative cells in response to IL-21 (Fig. 3A, B). Overall, these data suggest that the relative reduction of the fraction of FoxP3-expressing cells in cultures added with IL-21 (Fig. 3A, B) is due to both the suppression of FoxP3 gene transcription and the preferential expansion of FoxP3-negative T cells.
IL-21 modulates the apoptosis of different cell types 14, 15 and promotes the survival of T cells by activating the phosphatidylinositol 3-kinase pathway 16. However, when IL-21 was added to CD4+CD25– T cells activated in the presence of TGF-β, no significant change in the rate of T cell apoptosis was seen both at 24 and 48 h (not shown).
IL-21 prevents the TGF-β-dependent induction of functional Treg cells
CD4+CD25– cells pre-activated in the presence of TGF-β with or without IL-21 were used in a conventional co-culture assay, and their capacity to suppress the proliferation of freshly isolated CFSE-labeled CD4+CD25– T (responder) cells was evaluated. Ti-Treg cells and naturally occurring Treg cells (CD4+CD25+) were able to suppress responder cell proliferation at all the responder/suppressor ratios. By contrast, T cells generated in the presence of TGF-β and IL-21 failed to exert any suppressive effect on the growth of responder cells (Fig. 4A). The analysis of the proliferation index in the different co-cultures showed a significant loss of the suppressive capacity of T cells generated in the presence of TGF-β and IL-21, in comparison to both Ti-Treg cells generated in the absence of IL-21 and naturally occurring CD4+CD25+ Treg cells (Fig. 4B). These data indicate that the IL-21-mediated suppression of the TGF-β-dependent induction of FoxP3 in CD4+CD25– T cells parallels the loss of their suppressive function in vitro.
CD4+CD25– T cells treated with TGF-β and IL-21 fail to suppress colitis
We have previously demonstrated that in vitro generated Ti-Treg cells can prevent colitis in immunodeficient mice reconstituted with naive CD4+ T cells 10. In order to assess the in vivo effect of CD4+CD25– T cells activated in the presence of TGF-β and IL-21, SCID mice were reconstituted with freshly isolated CD4+CD25– T cells alone or together with an equal number of Ti-Treg cells or CD4+CD25– T cells pre-activated in the presence of TGF-β and IL-21. Mice reconstituted with CD4+CD25– T cells alone showed a progressive weight loss by day 10, and this weight loss was prevented by simultaneous transfer of Ti-Treg cells. When CD4+CD25– T cells were co-transferred with T cells generated in the presence of TGF-β and IL-21, no prevention of weight loss was seen (Fig. 5A). Endoscopy performed 3 wk after the reconstitution revealed signs of mild colitis in animals receiving CD4+CD25– T cells alone, which were completely suppressed by co-transfer of Ti-Treg cells. In contrast, mice reconstituted with CD4+CD25– T cells generated in the presence of TGF-β and IL-21 showed macroscopic signs of frank colitis as indicated by the presence of loose stools, loss of bowel wall translucency, and friable mucosa (Fig. 5B). Histology of the colons collected at day 25 after reconstitution showed a strong mucosal hypertrophy with loss of goblet cells, and enhanced infiltration of the mucosa with mononuclear cells in both mice receiving CD4+CD25– T cells alone or in combination with T cells generated in the presence of TGF-β and IL-21 (Fig. 5C). Analysis of cytokine RNA expression revealed that IFN-γ, TNF-α, IL-21 and IL-17 were significantly increased in the colon of mice receiving CD4+CD25– T cells pre-activated in the presence of TGF-β and IL-21 in comparison to those reconstituted with either CD4+CD25– T cells alone or with the addition of Ti-Treg cells (Fig. 5D). Animals receiving T cells generated in the presence of TGF-β and IL-21 showed reduced colonic (Fig. 5D), spleen and MLN (Fig. 5E) FoxP3 expression as compared with mice receiving Ti-Treg cells. Such a reduction of FoxP3 expression was associated with an increase of RORγt expression in the colonic mucosa (Fig. 5D), thus suggesting the accumulation of Th17-like cells at this level in animals receiving T cells generated in the presence of TGF-β and IL-21.
IL-21 autocrine secretion regulates the Ti-Treg and Th17 cell balance
Since mice receiving CD4+CD25– T cells pre-activated in the presence of IL-21 showed high expression of markers associated with Th17 cells, we investigated whether IL-21, in the presence of TGF-β, is able to redirect the genetic program leading to the generation of regulatory cells towards a Th17 phenotype. Indeed, activation of T cells in the presence of TGF-β and IL-21 resulted in a marked increase in the expression of IL-17 and RORγt as compared with cells activated in the presence of TGF-β alone (Fig. 6A). These data indicate that IL-21 acts as a switch between the genetic programs leading either to the generation of Ti-Treg or Th17 cells.
To further investigate the role of IL-21 in the generation of Ti-Treg cells, we first analyzed the induction of Ti-Treg cells in CD4+CD25– cells isolated either from IL-21 knockout mice and wild-type littermates. As shown in Fig. 6B (left panel), FoxP3 expression did not differ in IL-21–/– and wild-type T cells activated for 5 days in the presence of TGF-β alone, indicating that the endogenous expression of IL-21 is not sufficient to counteract the induction of Ti-Treg cells. This could rely on the fact that TGF-β inhibits the expression of endogenous IL-21 in T cells. Indeed the expression of IL-21 induced by activation of wild-type T cells was inhibited by the addition TGF-β to the culture medium (Fig. 6B, right panel). However, in the presence of limiting doses of exogenous IL-21, FoxP3 suppression was more profound in wild-type than in IL-21–/– cells (Fig. 6B, left panel), and this effect was associated with a dose-dependent recovery of the capacity to express endogenous IL-21 by wild-type cells (Fig. 6B, right panel). These data indicate that exogenous IL-21 induces its own expression in TGF-β-stimulated cells, which substantially contributes to the suppression of Ti-Treg cell induction. According to the essential role of endogenous IL-21 in the generation of Th17 cells, IL-21–/– cells expressed less RORγt when stimulated with TGF-β and exogenous IL-21 (Fig. 6C). Collectively, these data indicate that the induction of the autocrine expression of IL-21 plays an important role in the suppression of Treg cell induction and the generation of Th17 cells.
TGF-β is an important immunosuppressive cytokine, as substantiated by the demonstration that TGF-β-deficient animals develop a multi-organ autoimmune disease leading to death shortly upon birth 17. The counter-regulatory properties of TGF-β are dependent both on a direct suppression of immune cell activation and the induction of Treg cells. We and others have previously demonstrated that TGF-β is able to promote the differentiation of naive T cells into regulatory cells in the periphery (Ti-Treg cells) 8, 9. However, in the presence of pro-inflammatory cytokines such as IL-6, TGF-β favors the differentiation of a novel class of CD4+ effector T cells 18, 19. These cells, termed Th17 cells, are characterized by the expression of the transcription factor RORγt and production of high levels of IL-17, and they have been implicated in the pathogenesis of several immune diseases, including colitis. In this study we investigated the effect of IL-21, a CD4+ T cell-derived cytokine produced in excess in chronic immune-inflammatory diseases on the induction of Ti-Treg cells. We show that IL-21 is able to prevent the induction of Ti-Treg cells as evidenced by the reduced expression of the transcription factor FoxP3. However, in the presence of IL-21, there was an enhanced proliferation of FoxP3-negative cells; it is therefore plausible that the reduced expression of FoxP3 in these cell cultures is imputable to a preferential expansion of FoxP3-negative cells. However, a reduced induction of FoxP3 was seen at early time points (i.e. 12 h) of culture, raising the possibility that IL-21 may directly inhibit FoxP3 gene transcription. A further possibility is that IL-21 might reduce the number of FoxP3-positive cells by promoting the transdifferentiation of Treg cells into Th17 cells 17.
Functionally, CD4+CD25– T cells pre-activated in the presence of TGF-β and IL-21 failed to suppress the proliferation of target T cells in vitro, thus confirming that suppression of FoxP3 correlates with the loss of Ti-Treg cell-suppressive capacity. Consistently, colitis induced by the adoptive transfer of CD4+CD25– cells in SCID mice was suppressed by Ti-Treg cells generated in the absence of IL-21, but not by T cells generated in the presence of TGF-β and IL-21. By contrast, these later cell types caused an even more severe colitis than that observed in recipients that received no Ti-Treg cells. Such colitis was characterized by a reduced number of FoxP3-expressing cells in the gut mucosa as well as in the spleen and MLN, indicating that IL-21-induced FoxP3 suppression is not transient but persists once cells are transferred in vivo. Moreover, mice receiving T cells generated in the presence of TGF-β and IL-21 exhibited high levels of IL-17 and RORγt in the colon. This is consistent with the recent observation that IL-21 in the presence of TGF-β induces Th17 cells 20, 21. Indeed, in our cell cultures, IL-21-mediated suppression of FoxP3 was accompanied by an increase of both IL-17 and RORγt, indicating that IL-21 can induce Th17 cells in the presence of TGF-β and suppresses the generation of Ti-Treg cells.
Finally, we demonstrate that endogenous IL-21 is required to obtain an optimal IL-21-mediated Ti-Treg cell suppression and Th17 cell induction. Indeed, we observed that exogenous IL-21 re-establishes the expression of IL-21 in TGF-β-stimulated cells. Moreover, analysis of IL-21–/– and wild-type T cells showed that endogenous IL-21 substantially contributes to the suppression of FoxP3 and the generation of Th17 cells. These data indicate that IL-21 generates a self-sustaining autocrine loop in the TGF-β stimulated T cells which play a key role in the balance between Ti-Treg and Th17 cell generation.
In summary, IL-21 emerges as a key cytokine for the maintenance of the mucosal immune system homeostasis by fine-tuning either the generation of tolerogenic Ti-Treg cells or that of pro-inflammatory Th17 cells. Our data also well fit with previous studies showing the ability of IL-21 to counteract the activity of Treg cells in other cell systems, and with the demonstration that IL-21 is up-regulated in diseases characterized by defects of counter-regulatory mechanisms. Moreover, there is evidence that blockade of IL-21 is useful to limit the severity of inflammation in experimental models of immune-mediated diseases, which may be due, at least in part, to an enhanced de novo generation of Ti-Treg cells 22, 23. In the context of gut disease, which we have used here to show the in vivo significance of IL-21 inhibition of Treg cell development and function, our results reinforce a new notion which is that the host deliberately inhibits immune suppression in inflammation. Crohn's disease in humans is thought to be due to a disruption in mucosal homeostasis so that the host mounts a vigorous Th1 response to the normal flora, similar to an anti-pathogen response. It makes sense in terms of protective immunity to allow pro-inflammatory pathways to predominate until the pathogen is eliminated, or perhaps to dampen anti-inflammatory pathways so that the immune response does not kill the host. Our results here are quite consistent with a cellular corollary of this phenomenon, where IL-21, expressed at increased levels in inflammatory bowel disease, prevents the induction and expression of Treg cell activity, thus allowing ongoing inflammation.
Material and methods
BALB/c and SCID mice at 6–8 wk of age were obtained from the animal facility of the University of Rome “Tor Vergata” (Italy). All animal experiments were performed in accordance with the local institutional guidelines. IL-21 knockout mice (B6;129S5-il21tmLex) were purchased from the University of California, Davis, CA, USA and hosted in our animal facility under SPF conditions.
Isolation and culture of primary cells
CD4+CD25– T cells were isolated from murine splenocytes by CD4-MACS® negative selection followed by CD25-MACS® positive selection (Miltenyi Biotech, Bergisch Gladbach, Germany). CD4+CD25– T cell preparations used in all the experiments were at least 97% pure as shown by flow cytometry. Treg cells were induced as previously described 8. In brief, CD4+CD25– T cells were cultured in serum-free medium X-Vivo15 (BioWhittaker, Heidelberg, Germany) in the presence of plate-bound anti-CD3 Ab (1 µg/mL; clone 145–2C11), soluble anti-CD28 Ab (2 µg/mL; clone 37.51) and recombinant TGF-β1 (5 ng/mL; Sigma-Aldrich, Milan, Italy), in the presence or absence of IL-21 (R&D Systems, Minneapolis, MN).
Flow cytometric analysis
Cells were stained for CD4, CD44, CD62L, CD69 (Immunotools, Friesoythe, Germany), CD45RB (BD, Heidelberg, Germany), and FoxP3 (eBioscience, San Diego, CA), according to the manufacturers’ instructions. Cells were analyzed by means of a FASCalibur (BD Bioscience, San Jose, CA).
SCID mice were administered intraperitoneally with freshly isolated CD4+CD25– T cells from BALB/c mice (4 × 105) in the presence or absence of CD4+CD25– T cells (4 × 105) stimulated for 5 days with anti-CD3 and anti-CD28 Ab, recombinant TGF-β1, and/or IL-21. Weight changes were monitored daily. For monitoring of colitis, the Coloview high-resolution mouse video endoscopic system (Karl-Storz, Tuttlingen, Germany) was used as described 24.
Histological analysis of colon cross-sections
Sections were obtained from the colons and stained with hematoxylin and eosin (H&E). The degree of inflammation on microscopic cross-sections of the colon was graded by the same pathologist in a blinded fashion.
TNF-α, IL-6, and IFN-γ levels in the cell culture supernatants were quantified by means of a cytofluorimetry-based ELISA system (Flowcytomix; Bender Medsystems, Vienna, Austria) according to the manufacturer's protocol.
Real-time quantitative PCR
Total RNA was isolated using the peqGOLD Total RNA Kit (Peqlab, Erlangen, Germany). Reverse transcription into complementary DNA (cDNA) was performed with the M-MLV reverse transcriptase (Invitrogen, Karlsruhe, Germany). PCR was performed using a SYBR Green-based PCR using iQ SYBR mix (Bio-Rad, Hercules, CA). Specific primers were designed on the cDNA sequence: FoxP3 sense: 5′-GGT ACAC CCA GGA AAG ACA G-3′; FoxP3 antisense: 5′-ATC CAG GAG ATG ATC TGC TTG-3′; β-actin sense: 5′-AAG ATG ACC CAG ATCATG TTT GAG ACC-3′; β-actin antisense: 5′-agc cag gtc cag acg cag gat-3′; CD3 sense: 5′-GTA TCA CTC TGG GCT TGC TG-3′; CD3 antisense: 5′-TGG GCT CAT AGT CTG GGT TG-3′; TNF-α sense: 5′-ACC CTC ACA CTC AGA TCA TC-3′; TNF-α antisense: 5′-GAG TAG ACA AGG TAC AAC CC-3′; IFN-γ sense: 5′-CAA TAG ACG CTA CAC ACT GC-3′; IFN-γ antisense: 5′-CCA CAT CTA TGC CAC TTG AG-3′; IL-17A sense: 5′-CTC AGA CTA CCT CAA CCG TTC-3′; IL-17A antisense: 5′-TTC AGG ACC AGG ATC CT TGC-3′. IL-21 expression PCR was analyzed using a TaqMan-based assay (Applied Biosystems, Darmstadt, Germany) and the iQ Supermix (Bio-Rad). RORγt expression was analyzed as described 25.
Gene expression was calculated relative to the housekeeping gene β-actin using the ΔΔCt algorithm.
CD4+CD25– T cells were labeled with Celltrace CFSE (Invitrogen, Karlsruhe, Germany) according to the manufacturer's instructions. In some experiments, proliferation of CD4–CD25– T cells activated in the presence of TGF-β with or without IL-21 was evaluated. In other experiments CFSE-labeled CD4+CD25– T responder cells were co-cultured at different ratios with cells pre-activated for 5 days with TGF-β alone or with the addition of IL-21. Mitomycin-C (Sigma-Aldrich, St. Louis, MO)-treated autologous CD90-negative splenocytes were used as APC. Proliferation of responder cells was evaluated by flow cytometry after 3 days of culture. Quantification of cells undergoing divisions was evaluated by determining the proliferation index obtained with the proliferation module of the ModFit LT Macintosh software.
Student's t-test was used to calculate statistical significance in a particular measurement of a continuous variable between groups.
This work received support from the “Fondazione Umberto di Mario”, Rome, the Broad Medical Research Program Foundation (Grant No. IBD-0154R), and Giuliani SpA, Milan, Italy.Conflict of interest: These authors declare that there is no financial or commercial conflicts of interest.