Type 17 Th (Th17) cells have been identified as a distinct population of CD4+ effector T cells different from Th1 and Th2 cells. While the pre-eminent cytokine of Th1 cells is IFN-γ and that of Th2 cells is IL-4, the distinctive cytokine of Th17 cells is IL-17A. However, although murine and human Th1 and Th2 cells exhibit strong similarities, human and murine Th17 cells seem to differ in several aspects.
Several studies in the past 2 years have described the main features of human Th17 cells, including the secretion of cytokines such as IL-17A, IL-17F, IL-21, IL-22, IL-26, as well as the chemokine CCL20 1–4. Moreover, human Th17 cells express RORC, which is a human ortholog of mouse RORγt, IL-23 receptor (IL-23R) 1, 2 and also IL-1RI (our unpublished data). Regarding chemokine receptors, human Th17 cells express CCR6 and CCR4, but do not express CXCR3, which is instead expressed by Th1 cells 1, 2, a finding previously not reported in the mouse. Interestingly, Candida albicans-specific T cells are CCR6+CCR4+, and produce IL-17A but not IFN-γ, whereas Mycobacterium tuberculosis PPD-specific T cells are CCR6+CXCR3+, and produce IFN-γ but not IL-17A 1. Several studies have also demonstrated the existence of substantial numbers of human CD4+ T cells that are able to produce both IL-17A and IFN-γ 1, 2, 5, 6; the term “Th17/Th1” cells was proposed based on these observations 2. Chemokine receptor expression pattern may be a useful tool to delineate Th17 cells and Th1 cells – for example, data from our laboratory have shown that in peripheral blood T cells, the CCR6+CXCR3− fraction contains mainly Th17 and some Th17/Th1 cells, whereas all the CCR6−CXCR3+ cells are Th1 cells, and the CCR6+CXCR3+ circulating cell fraction contains all three cell types (Th17, Th17/Th1 and Th1) 7. Other features, which may suggest a common developmental origin between human Th17 and Th1 cells, have also been observed. First, human Th17 clones appeared to express the IL-12Rβ2 in addition to the IL-23R, and the Th1-related transcription factor T-bet in addition to RORC 2. More importantly, human Th17 clones could be induced to produce IFN-γ when cultured in the presence of IL-12; consistent with this, IL-12 is also able to up-regulate T-bet expression in human Th17 clones 2.
More recently, based on the results of a microarray analysis performed on human Th17 in comparison with Th1 and Th2 clones derived from the same donor, another unexpected phenotypic property of human Th17 cells was discovered. Virtually, all human Th17 clones expressed CD161 4, the human ortholog of murine NK1.1 8. Moreover, virtually all RORC-, IL-23R- and CCR6-expressing, as well as IL-17A-producing CD4+ T cells were within the CD161+ circulating cell fraction, and also within the CD161+ cell fraction infiltrating the inflamed gut of patients with Crohn's disease or the inflamed skin of patients with psoriasis 4. A summary of the main phenotypic features of human Th17, Th17/Th1, as compared with Th1, cells is depicted in Fig. 1.
Human Th17 cells have a different origin compared with that of mice
The source of human Th17 cells, as well as the cytokines inducing their development, are both contentious issues. Some independent studies clearly demonstrate that TGF-β is required for Th17 initiation and that IL-6 is a critical co-factor for mouse Th17-cell differentiation 9–11. While IL-23 maintains and/or expands Th17 cells, and IL-1β and TNF-α amplifies this process, none of these cytokines could substitute for either TGF-β or IL-6 for Th17 development 9–11. IL-21, another cytokine produced by Th17 cells, has been shown to provide an additional autocrine amplificatory signal 12–14.
By contrast, the origin of human Th17 cells, and in particular the role of TGF-β, in their differentiation, is still subject to intense debate. Acosta Rodriguez et al. 15 showed that human Th17 originate from purified peripheral blood naïve CD4+ T cells following their activation in the presence of IL-1β and IL-6, whereas the addition of TGF-β was in fact inhibitory. Wilson et al. 3 found that naïve T cells could be induced to differentiate into Th17 cells in response to either IL-1β or IL-23, but not in response to TGF-β. Chen et al. 5 also concluded that under the same conditions IL-23, but not TGF-β, was critical for human Th17-cell development, at least with regard to the production of IL-22 by these Th17 cells. Evans et al. 16 reported the induction of Th17-cell differentiation from human circulating naïve CD4+ T cells only following their contact with activated monocytes, whereas no inducing soluble factors were identified. Finally, van Beelen et al. 6 demonstrated that human Th17 cells could be derived only from memory, but not from naïve T cells, and this effect was due to the nucleotide oligomerization domain 2 ligand muramylpeptide, which enhanced IL-23 and IL-1 production by DC. All these studies were performed by using naïve (CD45RA+CD45RO−) CD4+ T cells sorted from adult peripheral blood, and thus raises the question of the true “naiveté” of these cells, and the extent to which the cytokines are acting on truly naïve, or are in fact, acting on activated and/or memory contaminating CD4+ T cells 17. Moreover, in many of these studies, human naive CD4+ T cells were cultured in bovine or human serum, which contain traces of TGF-β, suggesting a possible underestimation of its importance in Th17 differentiation. Subsequently, indeed, three independent studies 18–20 showed that even the development of human Th17 cells requires the activity of TGF-β, as it does in mice, even if the results of these studies were somewhat contradictory. Indeed, Manel et al. 18 found that TGF-β, IL-1β, IL-6 and IL-21 or TGF-β, IL-1β and IL-23 were necessary and sufficient to induce IL-17 expression in naïve umbilical cord blood (UCB) human CD4+ T cells. TGF-β up-regulated RORC expression but simultaneously inhibited its transcriptional activity, thus preventing the IL-17 expression. The inhibition exerted by TGF-β on RORC transcriptional activity was antagonized by the inflammatory cytokines IL-1β, IL-6, IL-21 and IL-23. Volpe et al. 19 found that TGF-β, IL-23, IL-1β and IL-6 were all essential for human Th17 development, but differentially modulated the cytokines produced by Th17 cells. More importantly, the absence of TGF-β induced a shift from a Th17-like to a Th1-like profile. By contrast, Yang et al. 20 found that TGF-β and IL-21 uniquely promoted the differentiation of human naïve CD4+ T cells into Th17 cells, accompanied by the expression of RORC. The results of these studies were enthusiastically welcomed 21, because they seemed to exclude the unfortunate and unhelpful possibility for the biomedical research community that mice may be of limited use as models for the development of Th17 cells in the human immune system 22.
However, our data showing that all human Th17 cells are included within the CD161+CD4+ T-cell fraction 4 has again opened the apparently solved question of the necessity of TGF-β during human Th17-cell development. In agreement with the results reported by van Beelen et al. 6, we were indeed unable to induce under any experimental condition reported, the development of IL-17A-producing cells from naïve CD4+ T cells purified from adult peripheral blood. We found, however, that human IL-17A-producing cells could originate from CD161+CD4+ T-cell precursors, detectable in both human UCB and thymus, if these cells were activated in the presence of both IL-1β and IL-23. Interestingly, the presence of both of these cytokines also induced not only an increase in the expression of RORC and IL-23R, but also of T-bet and IL-12Rβ2 and allowed the development of higher numbers of Th1 cells, suggesting again a possible developmental relationship between human Th17 and Th1 cells. No other cytokine or cytokine combinations (including TGF-β, IL-6 and IL-21) were able to induce IL-17A mRNA expression and IL-17A production. Interestingly, CD161+ naïve CD4+ T could be induced to differentiate also into Th1 or Th2 cells when cultured in the presence of exogenous IL-12 or IL-4, respectively. By contrast, CD161−, naïve CD4+ T cells from both UCB or thymus could be induced to differentiate only into Th1 or Th2 cells under appropriate polarizing conditions (IL-1β+IL-23, or IL-12 alone, for Th1 cells and IL-4 for Th2 cells), but they never differentiated into IL-17A-producing cells 4.
During the initial phase of our studies, cells were cultured in bovine-serum-containing media and thus we could not exclude the possibility of traces of contaminating TGF-β (even though the active role of this contaminant is perhaps negligible as demonstrated by the lack of effect of a neutralizing anti-TGF-β antibody during culture) 23. More importantly, IL-1β and IL-23 were found to be sufficient to induce the development of IL-17A-producing cells even when purified UCB CD161+CD4+ T cells were cultured and activated under serum-free conditions 23. The discrepancy between our study and that of the others showing an absolute requirement for exogenously added TGF-β for human Th17 differentiation in serum-free cultures 18–20 is currently unclear. One major difference between our study and the others is certainly the source of Th17 precursors. In the above-mentioned studies 18–20, purified total naïve CD4+ T cells from either adult or UCB were used, whereas we employed sorted UCB or thymus CD161+CD4+ T cells, based on the observation that IL-17A-producing cells could never have originated under any experimental condition from naïve CD161−CD4+ T cells 4. This different experimental approach may provide one explanation for the discrepancies. In mouse, the in vitro addition of TGF-β is usually thought to be required for the induction of RORγt, which provides the molecular basis for Th17 differentiation. However, when purified CD161+ CD4+ T-cell fraction in either human UCB or newborn's thymus was examined, it was surprisingly found to contain cells already expressing RORC (and IL-23R) mRNA ex vivo4, thus explaining why the addition of TGF-β in vitro was not required. Obviously, the possibility that the presence in the newborn's thymus and in UCB of a small subset of CD4+ T cells that already express RORC and act as precursors of Th17 cells is related to an in vivo activity of TGF-β cannot be excluded, but in any case it does not underscore the concept that the addition in vitro of IL-1β and IL-23, but not of TGF-β, is critical for the development of human Th17 cells. This may be in keeping with the results of a recent study showing that patients with mutations of TGF-B1 or TGF-BR2 did not exhibit any difference in the numbers of IL-17A-producing T cells compared with healthy controls, whereas patients with STAT3 and IL-12Rβ1 mutations have impaired Th17 development 24.
The results of our studies also showed that the addition of TGF-β to serum-free cultures containing IL-1β and IL-23 increases the relative proportions of CD161+ T cells that differentiate into Th17 cells, but inhibited both T-bet expression and Th1 development. This observation suggests that Th17 and Th1 cells may have a different susceptibility to the suppressive activity of TGF-β 23. In agreement with this hypothesis, when the proliferation of human Th17 clones and also of circulating Th17 cells in the presence or absence of TGF-β was examined, these cells showed a lower susceptibility to the anti-proliferative effect of TGF-β than Th1 or Th2 clones or circulating IFN-γ-producing (Th1) cells 22. This difference appeared to be due to the reduced apoptotic cell death of Th17 cells in the presence of TGF-β in comparison with Th1 cells, which appeared to be consistent with the observation that human Th17 cells exhibit lower expression of clusterin (TGF-β signalling, pro-apoptotic), and higher Bcl-2 (anti-apoptotic) in comparison with Th1 or Th2 clones 23. Taken together, these findings support the concept that TGF-β does not have a direct and critical effect on the development of human Th17 cells, but it can indirectly favour their development by selectively inhibiting both T-bet expression and the development of Th1 cells (Fig. 2).
Which is the role of CD161 expression by human Th17 cells?
The results of our studies shows that naïve CD161+CD4+ T cells can give rise to Th1, Th2 or Th17 cells under appropriate conditions, whereas CD161− cells can only develop into Th1 or Th2, but not into Th17, cells. This therefore raises the question of whether CD161+ and CD161− naïve CD4+ T cells represent two distinct lineages or whether CD161− originate from the more committed CD161+ subset (Fig. 2). Presently, however, this question remains unanswered.
CD161 (or NKR-P1A) is the human ortholog of the murine NK1.1 8. NK1.1 is not only expressed on almost all NK cells, but also by a subset of T cells, which have been named as invariant NKT (iNKT) cells 25, 26. NKT cells are CD1d-restricted and use an invariant or highly restricted TCR-α-chain 25, 26. However, other cell types that express CD161, but are restricted by conventional MHC molecules instead of CD1d and therefore exhibit non-invariant TCR, have been described in humans and named as NKT-like CD4+ or CD8+ T cells 27, 28. CD161+ cells were found to be a minority (approximately one of five) among human circulating CD4+ T cells and a significant majority of CD161+ cells expressed CD45RO, which suggests a memory T-cell phenotype 29. The percentage of iNKT cells among human circulating CD4+ T cells is lower than 1% 29, which is in stark contrast to mice, in which the majority of circulating NK1.1+ cells are iNKT cells 25. The IL-17-producing CD161+CD4+ T cells described in our study were not CD1d-restricted NKT cells, but MHC class II-restricted NKT-like CD4+ T cells. Thus, our findings are again different from those reported in mice showing that a subset of iNKT cells produce IL-17, but curiously, although most murine iNKT cells express NK1.1, the majority of IL-17-producing iNKT cells are NK1.1-negative 30, 31.
The finding that all human memory/activated Th17 cells and their precursors express CD161 also raises the critical question of its functional significance. One physiological CD161 ligand has been identified as the lectin-like transcript 1 (LLT1 or CLEC2D), which belongs to the C-type lectin domain family 2 (CLEC2) 32. It has been recently shown that CLEC2D is expressed by TLR-activated plasmacytoid DC, monocyte-derived DC and B cells and its interaction with CD161 expressed by NK and CD8+ T cells induce inhibition of both cytolitic activity and cytokine production, respectively, 33. No data are presently available regarding its possible role on Th17 development and function. More recently, however, a second CD161 ligand has been identified and named as proliferation-induced lymphocyte-associated receptor or CLEC2A 34, 35. proliferation-induced lymphocyte-associated receptor signalling through CD161 supports CD3- or antigen-induced T-cell proliferation by increasing the expression of anti-apoptotic Bcl-xL 34. This finding suggests that CD161 expression by human Th17 cells and their precursors may play an important role in the modulation of their expansion. Additional information on these mechanisms may allow for better understanding of the differences between human and murine Th17 cells, as well as to develop novel therapeutic strategies for the regulation of these important effectors of adaptive immunity.
The studies reported in this review have been supported by funds of the Italian Ministry of Education (40%), the Italian Ministry of Health, the Ente Cassa di Risparmio of Florence, EU Projects SENS-IT-IV (FP6-LSBH-CT-2006–018861), INNOCHEM (FP6-LSHB-CT-2005–518167), Agenzia Spaziale Italiana MoMa (WP ref: 1B1241-X5).
Conflict of interest: The authors declare no financial or commercial conflict of interest.