Introduction
TH-17 cells, T helper cells that produce interleukin 17 (IL-17) and other proinflammatory cytokines, have been shown to have key functions in several mouse autoimmune disease models and are thought to be similarly involved in human disease1, 2, 3. In healthy humans, IL-17-secreting cells are present in the CD45RO+CCR6+ T cell populations of peripheral blood4, 5 and gut5. TH-17 cells or their products are associated with the pathology of many human inflammatory and autoimmune disorders. IL-17 and CD4+ TH-17 cells are found in lesions from multiple sclerosis patients6, 7, 8, where they seem to contribute to disruption of the blood-brain barrier9. IL-17, produced by CD4+ T cells of rheumatoid synovium10, is believed to contribute to the inflammation of rheumatoid arthritis11, 12. In psoriasis, products associated with TH-17 cells, including IL-17, IL-17F, IL-22 and the chemokine receptor CCR6, are induced13, 14, 15. TH-17 cells are detected and IL-17 is induced in gut mucosa from patients with Crohn's disease and ulcerative colitis13, 16. IL-23, produced by intestinal dendritic cells17, contributes substantially to the differentiation of TH-17 cells18, and polymorphisms in the gene encoding the IL-23 receptor (IL23R) are associated with Crohn's disease, which further links the TH-17 cell pathway to such pathogenesis19.
The mechanisms leading to the differentiation of TH-17 cells have been well established in mice but are still poorly understood in humans. In mice, the differentiation of TH-17 cells that secrete IL-17 (also called IL-17A) and IL-17F requires expression of the transcription factors ROR
t (an orphan nuclear hormone receptor), STAT3 and IRF4 (ref. 20). RORt is sufficient to direct IL-17 expression in activated mouse T cells21 and is thus considered the effector transcription factor that establishes the TH-17 differentiation lineage. The conditions that induce the differentiation of TH-17 cells from naive mouse T cells, including RORt expression, have been established. TGF-
(A002271) and IL-6 (refs. 22,23,24) or TGF- and IL-21 (refs. 25,26,27) are sufficient to initiate the expression of IL-17 and IL-17F. Expression of IL-22, considered to be another TH-17 cytokine, is induced by IL-6 and inhibited by high concentrations of TGF-14. IL-23 is required in vivo for the generation of pathogenic TH-17 cells but is not required in vitro for the induction of IL-17, IL-17F or IL-22 (ref. 18).
In contrast to mouse T cells, human T cells with a naive surface phenotype fail to produce IL-17 in the presence of TGF- and IL-6 (refs. 28,29,30,31). However, increased IL-17 expression has been reported in response to IL-1 alone29 or with IL-23 (ref. 15), although other studies have not noted such a response30. The dissimilarity in differentiation requirements for mouse and human TH-17 cells may result from divergent differentiation processes, although it is possible that T cells purified from human adult peripheral blood on the basis of CD45RA expression alone may not be equivalent to naive mouse T cells32, 33, 34.
To avoid having antigen-experienced cells and serum-derived TGF- in TH-17 differentiation cultures, here we used human cord blood CD4+ T cells and serum-free medium. In these conditions, IL-17 and other TH-17 gene products were induced only when TGF- was added to the culture medium. In contrast to requirements in mouse T cell cultures, we did not find differentiation of human TH-17 cells when IL-6 or IL-21 was combined with TGF-. Instead, IL-1 and IL-6, IL-21 or IL-23 were required along with TGF- for the induction of IL-17 expression. As in the differentiation of mouse TH-17 cells, RORt was induced in human cells by TGF- and was required and sufficient for expression of TH-17 cell products, which indicated that the basic mechanism of the differentiation of TH-17 cells is evolutionarily conserved.
Results
ROR
t-dependent IL-17 expression in memory T helper cells
To evaluate the effect of IL-1 on IL-17 production, we sorted naive CD45RO-
CD25-
and TH-17 cell–containing memory CD45RO+CD25-
CCR6+ CD4+ T cell populations from human adult peripheral blood and cultured them in serum-containing medium in the presence or absence of IL-1. IL-1 induced a twofold increase in IL-17 production in CCR6+ memory cells but had no effect on CD45RO-
cells (Fig. 1a). We obtained similar results with sorted CD45RA+ cells (data not shown). With the goal of identifying requirements for the differentiation of TH-17 cells in humans, we initially sought to evaluate whether RORt was necessary in precommitted TH-17 cells to maintain effector function. To ablate RORt expression, we used two short hairpin RNA (shRNA) molecules that showed potent ability to 'knock down' RORt after transient transfection (data not shown). We transduced sorted CD45RO+CCR6+ memory CD4+ T cells isolated from adult blood with the shRNA vectors. After 6 d, shRNA-1 and shRNA-2 decreased expression of RORC mRNA (encoding RORt) by 50% and 90%, respectively (Fig. 1b). Correspondingly, we noted on average a decrease of 50% and 70% in the number of IL-17+ cells with shRNA-1 and shRNA-2, respectively (Fig. 1c). The proportion of interferon- (IFN-)–positive cells remained high in all samples. Thus, RORt was required for maintenance of IL-17 expression in differentiated T cells.
Figure 1: ROR
t is necessary and sufficient for IL-17 expression in human CD4+ T cells.
![Figure 1 : ROR|[gamma]|t is necessary and sufficient for IL-17 expression in human CD4+ T cells.](/ni/journal/v9/n6/thumbs/ni.1610-F1.jpg)
(a) Flow cytometry of the production of IL-17 and IFN- by sorted CD45RO-
and CD45RO+CCR6+ cell populations activated and expanded in the presence of IL-2 with or without IL-1, analyzed on day 6. (b,c) RT-PCR analysis of RORC mRNA (b) and flow cytometry of the intracellular production of IL-17 and IFN- (c) in sorted CD45RO+CCR6+ cells transduced with empty vector or vector encoding RORt-specific shRNA (shRNA-1 and shRNA-2), analyzed on day 6. (d) Flow cytometry of the intracellular production of IL-17 and IFN- by naive cord blood CD4+ T cell populations activated, transduced by IRES-HSA (Control) or RORt-IRES-HSA (RORt), then expanded in the presence of IL-2 and analyzed on day 6. (e) RT-PCR analysis of IL17, IL17F and IL26 mRNA in naive cord blood CD4+ T cells transduced with IRES-GFP or RORt-IRES-GFP, sorted for GFP expression and analyzed on day 6. (f) Flow cytometry of the cell surface expression of CCR6 by naive cord blood CD4+ T cells transduced with IRES-HSA, RORt-IRES-HSA, GATA-3–IRES-HSA or T-bet–IRES-HSA, assessed on day 12. Numbers in quadrants (a,d,f) or beside outlined areas (c) indicate percent cells in each; expression in b,e is presented relative to the expression of ACTB. Data are representative of three (a,b,d) or four (c) independent experiments.
RORt-induced cytokines from cord blood cells
We next sought to determine whether overexpression of RORt, which would bypass any requirement of its induction by cytokines, would be sufficient to achieve IL-17 expression in naive CD4+ human T cells. We isolated CD4+ T cells from human cord blood to ensure a naive phenotype, activated them with antibody to CD3 (anti-CD3) and anti-CD28 and transduced them with control lentivirus or lentivirus encoding human RORt. IL-17 expression was readily detected, peaking at day 6 in cells transduced with the RORt vector (Fig. 1d and data not shown). The proportion of IFN--expressing cells was substantially decreased by expression of RORt (Fig. 1d). ROR
d and ROR, two other ROR family members, also induced IL-17 expression when overexpressed in primary human T cells (Supplementary Fig. 1 online).
We did not detect IL-22 protein induction with RORt overexpression (data not shown). This was unexpected, because Il22 mRNA is strongly upregulated by RORt in mouse CD4+ T cells (L. Zhou and D.R.L., unpublished data). This apparent discrepancy between mouse and human led us to investigate the gene encoding IL-22 in various species (Supplementary Fig. 2a online). In human, IL22 is located in the same locus as IFNG and IL26. In mouse, there is no Il26, and Il22 is located in the same locus as Ifng and Iltifb, an Il22 duplication. Like IL-22, IL-26 is an IL-10-related cytokine and is found in memory CD4+ T cells expressing IL-17 (ref. 15). Quantitative PCR analysis of RORt-expressing cells indicated that IL26 was induced along with IL17 and IL17F by RORt (Fig. 1e). A gene encoding IL-26 is also found in the genome of preplacental vertebrates, including zebrafish35, but not in the genome of rat and mouse (Supplementary Fig. 2b), which indicates that it was lost in a common ancestor of these rodents.
Human TH-17 cells are found exclusively in the CD45RO+CCR6+ compartment in adult blood. However, this compartment also contains IFN-+IL-17-
cells and IFN-+IL-17+ cells. To determine which transcription factor induces CCR6 expression in CD4+ T cells, we transduced cord blood CD4+ T cells with vectors encoding RORt or the transcription factors involved in specification of the T helper type 1 and type 2 lineages (T-bet and GATA3, respectively), or a control empty vector. CCR6 was induced in RORt-expressing cells but not in cells transduced with vector encoding GATA3 or T-bet, and it was not induced in trans in cultures of RORt-expressing cells (Fig. 1f). Expression of CCR2 and CCR4, also suggested to be TH-17 cell markers4, 36, was not altered by overexpression of RORt (data not shown).
Antagonistic effects of TGF-
on RORt function
We next sought to determine how cytokines known to affect TH-17 cells in mice or humans would affect IL-17 expression after RORt overexpression, which circumvents the potential effect of the cytokines on RORt expression itself. We transduced cord blood CD4+ T cells with RORt alone or in the presence of IL-1, IL-6 or IL-21 in combination with various concentrations of TGF- (Fig. 2a). The addition of IL-1, IL-6 and IL-21 increased the proportion of IL-17-producing cells obtained after forced expression of RORt by about twofold. However, TGF- potently suppressed IL-17 production. Notably, the addition of IL-1, IL-6 or IL-21 partially relieved the suppression induced by TGF-. We then sought to determine whether these cytokines influenced expression of endogenous RORt. We sorted CD45RO-
naive CD4+ T cells as well as subsets of memory CD4+ T cells from adult peripheral blood on the basis of CCR6 and CCR4 expression to compare expression of RORC mRNA. RORC expression was enriched in CCR6+ cells (Fig. 2b). We screened various cytokines for their ability to induce RORC expression in CD45RO-
adult naive CD4+ T cells. Unexpectedly, TGF- alone induced RORC expression in a dose-dependent way, but none of the other cytokines had such an effect (Fig. 2b,c). However, treatment with TGF-, alone or with IL-1, IL-6 or IL-21, was insufficient to induce IL-17 expression, as detected by intracellular staining in these conditions (data not shown).
Figure 2: TGF-
induces RORt and inhibits its activity, but this inhibition is relieved by inflammatory cytokines.
![Figure 2 : TGF-|[beta]| induces ROR|[gamma]|t and inhibits its activity, but this inhibition is relieved by inflammatory cytokines.](/ni/journal/v9/n6/thumbs/ni.1610-F2.jpg)
(a) Flow cytometry of intracellular staining for IL-17 in naive cord blood CD4+ T cells transduced with RORt-IRES-HSA alone (–) or with increasing concentrations of TGF- plus IL-1, IL-6 or IL-21, assessed on day 6. (b,c) RT-PCR analysis of RORC mRNA in freshly sorted CCR4-
CCR6-
, CCR4+CCR6-
, CCR4-
CCR6+ and CCR4+CCR6+ adult memory CD4+ T cells and in naive CD4+ T cells cultivated for 3 d in the presence of various cytokines (b) and in naive cord blood CD4+ T cells cultivated with various concentrations of TGF- (c). Expression is presented relative to ACTB expression. (d) Flow cytometry of the intracellular expression of IL-17 and Foxp3 in naive cord blood CD4+ T cells transduced with RORt-IRES-HSA in serum-containing medium (FBS) with or without anti-TGF- or in serum-free medium, analyzed on day 6. Results are from a donor with low IL-17 expression after RORt transduction in the presence of serum. Numbers in quadrants (a,d) indicate percent cells in each. Data are representative of at least three independent experiments.
The observation that cultures containing IL-1, IL-6 and IL-21 had higher IL-17 expression after RORt transduction (Fig. 2a) suggested that an endogenous source of TGF- existed in our culture conditions and that the addition of the other cytokines relieved its effect in a way similar to their effects after the addition of exogenous TGF-. Indeed, TGF- is found in human and bovine serum37, 38, and serum TGF- is sufficient to induce expression of the transcription factor Foxp3 in naive human CD4+ T cells39. We thus sought to determine whether IL-17 expression was higher in serum-free conditions. We transduced cord blood CD4+ T cells with RORt in serum-containing (10% FBS) and serum-free RPMI media. We noted substantially more IL-17 production in serum-free medium, which was most prominent when T cells had relatively little IL-17 after RORt transduction in the presence of serum (Fig. 2d). Concurrently, Foxp3 expression was induced in serum-containing but not serum-free medium. After the addition of a neutralizing antibody to TGF- in serum-containing medium, Foxp3 induction was almost completely abolished, whereas IL-17 expression was increased, but not to the extent noted in serum-free medium. A higher concentration of neutralizing antibody did not 'improve' IL-17 expression (Supplementary Fig. 3 online). These observations indicated that TGF- present in serum inhibited to some extent the IL-17 expression induced by RORt and that other unidentified compounds in fetal bovine serum also possibly counteracted the TH-17 differentiation.
Cytokines required for human TH-17 cell polarization
The findings reported above prompted us to evaluate whether TGF-, IL-1, IL-6 and IL-21 would induce IL-17 production in serum-free medium. We activated naive cord blood CD4+ T cells with beads coated with anti-CD3 and anti-CD28 in serum-free medium in the presence of anti-IL-4 and anti-IFN- alone or with various combinations of cytokines. These included increasing concentrations of TGF- with no added cytokine, or with IL-1, IL-6 or IL-21, with or without IL-23 (Fig. 3a). After 2 weeks of culture, we detected IL-17 expression by intracellular staining only in cells cultivated with a combination of TGF-, IL-1 and IL-23. Although IL-2 inhibits IL-17 expression in mice40, IL-1 relieves this effect41. We therefore tested the effect of IL-2 and neutralizing anti-IL-2 in our culture conditions. In the presence of TGF-, IL-1 and IL-23, IL-17 expression was higher at day 6 when IL-2 was included (Fig. 3b). The addition of an IL-2-blocking antibody prevented cell proliferation, and IL-17 expression could not be detected. Thus, IL-2 seemed to have a positive effect on IL-17 expression in human CD4+ T cell culture. In mice, IL-23, IL-21 and IL-6 share the ability to activate STAT3 if their cognate receptors are expressed. Although IL-6 and IL-21 with TGF- alone did not induce IL-17 in human cord blood T cells, they did induce IL-17 when IL-1 was included, albeit not as strongly as the combination of TGF-, IL-23 and IL-1 did (Fig. 3c). In mice, IL-21 is synthesized in response to IL-6 by TH-17 cells and functions in an autocrine way to induce TH-17 differentiation25, 26, 27. We therefore sought to determine whether IL-21 or IL-6 was also required for human IL-17 induction. In human cells, IL-6 and IL-21 were not induced by IL-1, IL-23 and TGF- (Supplementary Fig. 4a,b online). Furthermore, the addition of neutralizing anti-IL-6 or soluble IL-21 receptor in amounts that inhibited STAT3 phosphorylation had no effect on IL-17 expression (Supplementary Fig. 4c and data not shown). Thus, in contrast to results obtained with naive mouse CD4+ T cells, IL-23 participated in the induction of IL-17 in human T cells in the absence of IL-6 and IL-21.
Figure 3: TGF-, IL-1 and IL-6, IL-21 or IL-23 are required for human TH-17 polarization in serum-free conditions.
![Figure 3 : TGF-|[beta]|, IL-1|[beta]| and IL-6, IL-21 or IL-23 are required for human TH-17 polarization in serum-free conditions.](/ni/journal/v9/n6/thumbs/ni.1610-F3.jpg)
(a) Intracellular staining of IL-17 in naive cord blood CD4+ T cells activated without cytokines (none) or with IL-1, IL-6 or IL-21 with or without IL-23, alone (0) or with increasing concentrations of TGF- (wedges indicate tenfold increments, with a maximum of 10 ng/ml), with IL-2 added on day 3, analyzed on day 14. (b) Flow cytometry of IL-17 expression by naive cord blood CD4+ T cells activated with or without a combination of IL-1, IL-23 and TGF- and with or without IL-2 or neutralizing anti-IL-2 (-IL-2), analyzed on day 6. (c) Intracellular staining for IL-17 in naive cord blood CD4+ T cells cultivated with IL-2 and increasing concentrations of TGF- in the presence of various cytokines (below graph), analyzed on day 6. (d) Flow cytometry assessing the time course of the production of IL-17 and IFN- in naive cord blood CD4+ T cells polarized in the presence of IL-2, IL-1 and IL-23 and increasing concentrations of TGF-. (e) Intracellular staining for IL-17 in naive cord blood CD4+ T cells from various donors (n = 11) in the presence of IL-2 alone or with IL-1, IL-23 and TGF-, assessed on day 6 of culture. Each symbol indicates a separate donor. (f) Intracellular staining for IL-17 and IL-22 in naive cord blood CD4+ T cells cultivated with increasing concentrations of TGF- and IL-2 alone or with IL-23, IL-1 and IL-2, analyzed on day 6. Data are representative of at least three independent donors.
Regulation of TH-17 'signature' genes by TGF-, IL-1 and IL-23
On the basis of the results presented above, we next used a combination of TGF-, IL-1, IL-23 and IL-2 for TH-17 polarization of human cord blood naive CD4+ T cells. We detected IL-17+ cells as early as day 3, and they increased up to day 6 in culture (Fig. 3d). In several cord blood samples, the proportion of IL-17+ cells obtained in these conditions ranged from 0.5% to 11% (Fig. 3e). As in mice42, IL-17 induction was inhibited by the addition of retinoic acid (Supplementary Fig. 5 online). We also evaluated IL-22 in TH-17 differentiation cultures of cord blood cells. A substantial proportion of naive cord blood CD4+ T cells spontaneously expressed IL-22 after 6 d of culture (Fig. 3f); as the concentration of TGF- was increased, IL-22 expression was progressively inhibited.
Because RORt induced IL26 expression, we sought to determine whether IL26 could be similarly induced in human cord blood cells cultured in TH-17 differentiation conditions. In such conditions, IL17 mRNA expression was maximal in the presence of IL-1, IL-23 and TGF-, consistent with the intracellular staining (Fig. 4a). We also detected IL26 expression and found that IL26 mRNA increased with the dose of TGF- (Fig. 4b). Some expression of IL17F was induced by IL-1 alone, but IL-23 alone had no effect (Fig. 4c). However, increasing concentrations of TGF- acted in synergy with IL-23 and IL-1 to induce maximum IL17F expression. RORC expression was gradually induced with increasing concentrations of TGF- and was further enhanced by the addition of both IL-1 and IL-23 but not by either cytokine alone (Fig. 4d). In the same conditions, RORA expression was slightly induced by TGF-, and there was no further effect after the addition of IL-1 and IL-23 (Supplementary Fig. 6a online). That observation is in agreement with the slight enrichment for RORA mRNA in memory CCR6+ cells relative to that in CCR6-
cells (Supplementary Fig. 6b).
Figure 4: Induction of IL26, IL17F, IL17, RORC and IL23R mRNA during human TH-17 differentiation.
(a–c) RT-PCR analysis of IL17 mRNA (a), IL26 mRNA (b) and IL17F mRNA (c) in naive cord blood CD4+ T cells cultivated with increasing concentrations of TGF- and with IL-2 alone (–), IL-2 plus IL-1 or IL-23, or IL-2 plus IL-1 and IL-23, analyzed on day 6 after restimulation with phorbol 12-myristate 13-acetate and ionomycin. (d,e) RT-PCR analysis of RORC mRNA (d) and IL23R mRNA (e) in naive cord blood CD4+ T cells cultivated with increasing concentrations of TGF- and with IL-2 alone (–), IL-2 plus IL-1 or IL-23, or IL-2 plus IL-1 and IL-23, analyzed on day 6. Expression is presented relative to ACTB expression. Data are representative of three independent donors.
In mice, Il23r is induced by IL-6 or IL-21 but is inhibited by high concentrations of TGF-43. In human cells, IL23R expression was induced to some extent by IL-23 alone but not by IL-1, consistent with a published report28 (Fig. 4e). However, IL23R expression reached a maximum in the presence of IL-1 and IL-23 with increasing concentrations of TGF-. This indicated that in the presence of TGF- and IL-1, IL-23 induced expression of its own receptor through a positive feedback loop, leading to maximum expression and induction of RORC, IL17 and IL17F. In addition, CCR6 cell surface expression was induced by TGF- alone (Fig. 5a). In TH-17 differentiation conditions, we detected IL-17 only in CCR6+ cells (Fig. 5b), consistent with what we noted for memory TH-17 cells. TGF- also induced Foxp3 expression in a dose-dependent way (Fig. 5c). The addition of IL-23 as well as of IL-6 and IL-21 suppressed the induction of Foxp3, but the addition of IL-1 did not (Fig. 5c and data not shown). Thus, regulation of Foxp3 expression during TH-17 differentiation was similar in mice and humans.
Figure 5: Expression of CCR6 and Foxp3 during human TH-17 differentiation.
(a) Surface staining of CCR6 on naive cord blood CD4+ T cells cultivated with increasing concentrations of TGF- and with IL-2 alone (–), IL-2 plus IL-1 or IL-23, or IL-2 plus IL-1 and IL-23, analyzed on day 6. (b) Flow cytometry of intracellular IL-17 in naive cord blood CD4+ T cells cultivated for 6 d in IL-2, IL-23, IL-1 and TGF- and sorted as CCR6+ and CCR6-
cells. (c) Flow cytometry of the expression of Foxp3 and IL-17 by naive cord blood CD4+ T cells cultivated as described in a and analyzed on day 6. Data are representative of four independent experiments.
Discussion
On the basis of studies using both in vitro culture systems and genetic approaches, it is now clear that TGF- acts in concert with the proinflammatory cytokines IL-6, IL-21 and IL-23 to induce the differentiation of TH-17 cells in mice20. Phosphorylation of STAT3 after engagement of receptors for inflammatory cytokines27, 44 and induction of RORt expression are essential for mouse TH-17 differentiation21. The requirement for TGF- in TH-17 differentiation was initially unexpected, as it is known to act as an anti-inflammatory cytokine, at least in part through its induction and maintenance of regulatory T cells22. The function of TGF- may be dependent on context and thresholds, favoring differentiation of TH-17 cells at low concentrations in the presence of inflammatory cytokines and differentiation of regulatory T cells at high concentrations43.
Given the pivotal function of TGF- in controlling the balance between TH-17 cells and regulatory T cells in mice, it was unclear why it has been found to be inhibitory in the induction of IL-17 in human CD4+ T cells with a naive surface phenotype15, 29. We used serum-free medium to show that TGF- is essential in the differentiation of naive human CD4+ T cells toward the TH-17 lineage, similar to what has been noted in mice. In human T cells, TGF- induced RORt expression yet paradoxically inhibited its transcriptional activity, thus preventing expression of IL-17. A combination of IL-1 and IL-6, IL-21 or IL-23 relieved that inhibition and also contributed to RORt expression, leading to induction of IL-17. Thus, as in mice, TGF- was required for IL-17 expression in human T cells, and additional transcription factors induced by stimulation of the T cell antigen receptor, IL-1 and IL-6, IL-21 or IL-23, may be involved in inducing IL-17 expression20.
We have shown a requirement for IL-23 in the in vitro differentiation of human TH-17 cells, which contrasts with what has been noted in mice, in which IL-23 is required only in vivo18. However, with low concentrations of TGF- in mouse T cell culture, we found a positive effect of IL-23 on IL-17 production43. Therefore, the discrepancy between the mouse and human systems in terms of IL-23 may be due to different culture conditions or different sensitivities to TGF-. The inflammatory cytokines IL-6, IL-21 and IL-23 share signaling pathways by activating both STAT1 and STAT3 (refs. 45,46,47,48,49), whereas IL-1 is thought to activate the kinases IRAK1 and IRAK2 through recruitment of the adaptor MyD88 (refs. 50,51). Thus, STAT3 is probably a common denominator in the induction of expression of RORt and IL-17 in both species20. The IL-1 pathway is important for the in vivo induction of TH-17 cells in mice but does not seem to be required for polarization in vitro in the presence of serum52, 53. It remains to be determined whether an unrecognized requirement 'downstream' of the IL-1 receptor is also needed during mouse TH-17 differentiation in vitro.
Published observations showing an inhibitory activity for TGF- in the differentiation of human TH-17 cells have probably been confounded by the use of serum and suboptimal purification of naive cells28, 29. Indeed, TGF- has long been recognized as a 'switch' cytokine that is highly context and concentration dependent54. We identified an essential effect of exogenous TGF- on TH-17 differentiation by using serum-free medium, which indicated that TGF- present in serum might have masked this effect. However, neutralization of TGF- did not completely abolish the inhibitory effect of serum on RORt-directed IL-17 expression, and TGF- did not completely inhibit IL-17 expression after RORt overexpression in serum-free conditions (data not shown). This indicated that as-yet-unidentified inhibitory factors in the serum acted in synergy with TGF- to counteract TH-17 differentiation. In addition to IL17, we have shown that IL17F and IL26 were induced by RORt overexpression and cytokine polarization. IL-26 targets epithelial cells and has been suggested to be involved in mucosal immunity55, which is consistent with its induction in TH-17 cells. Rearrangements of the Il22-Ifng locus seem to have occurred in the mouse-rat lineage, leading to loss of Il26, although IL-26 may be important in host defense and inflammation in humans. Expression of IL-22 in human T cells was inhibited by TGF-, as noted before in mice14. Although all IL-17+ cells expressed CCR6 after TH-17 differentiation, TGF- alone induced CCR6. That is in agreement with the observation that Foxp3+ cells can also express CCR6 (ref. 56). As expected, Foxp3 expression was induced by TGF- alone in serum-free conditions, and IL-6, IL-21 and IL-23 were each able to suppress this induction, as also occurs in mice20. However, the addition of IL-1 was required for the induction of IL-17 through a mechanism that is at present unknown.
In human memory cells, expression of both IL-17 and IFN- is often detected, which raises the issue of how such cells are derived. The polarization conditions described here for naive T cells resulted in the differentiation of only IL-17+IFN--
cells. Although we have not investigated why IFN- was not expressed in these conditions, we believe that the high concentration of TGF- required for TH-17 polarization most probably inhibits IFN- expression. It remains to be determined whether in some conditions IL-17 and IFN- can be expressed together after the differentiation of naive T cells. Functional plasticity has been noted in the differentiation of T helper cells57. It is thus possible that either IL-17 or IFN- is expressed in response to diverse stimuli received by previously differentiated T helper type 1 or TH-17 memory cells, respectively.
RORt is uniquely expressed in mouse T cells that produce IL-17 and is required for the upregulation of this cytokine in T cells both in vivo and in vitro21. Here we have shown that RORt also has a central function in the differentiation of human TH-17 cells. 'Knockdown' of RORt with shRNA in memory CCR6+ cells resulted in much less IL-17 expression, which indicated that RORt was required for the maintenance of cytokine production in TH-17 cells. This result does not rule out the possibility of a small contribution in directing IL-17 expression by the closely related paralog ROR, which has a similar function in mice58. However, we did not find substantial enrichment of RORA mRNA in CCR6+ cells relative to that in CCR6-
cells. Furthermore, RORA mRNA was not strongly induced by a combination of IL-1, IL-23 and TGF- that induced a 50-fold increase in RORC mRNA. However, four differentially spliced isoforms of RORA have been described, and their respective transcriptional regulation has not been determined. Our data do not exclude the possibility of potential post-transcriptional regulation of ROR. Finally, the ability of ROR to induce IL-17 expression needs to be evaluated in relevant cell types, as ROR expression has not been detected in peripheral CD4+ T cells (data not shown).
The IL-23–TH-17 axis has been linked to many human diseases59. Our demonstration of a requirement for IL-23 in the differentiation of TH-17 cells is relevant, given the many polymorphisms in the human IL23R gene reported to be associated with Crohn's disease and psoriasis19, 60, 61, 62, 63, 64. It will be important to elucidate the functions of IL-1, IL-6, IL-21, IL-23 and TGF- in the pathogenesis of human diseases involving TH-17 cells. The participation of TGF- in the induction of TH-17 and regulatory T cells is probably critical in the maintenance of immune system homeostasis, particularly at mucosal surfaces, and imbalance in this system may result in autoimmunity. In this context, our results offer a working model for studying the differentiation of human TH-17 cells and provide new opportunities for manipulating these cells in inflammatory diseases.
Methods
Cell purification.
Blood samples were obtained from the New York Blood Center. Mononuclear cells were prepared from the buffy coats of samples from healthy adult donors or from cord blood on FicollPAQUE gradients. CD4+ T cells were isolated on an autoMACS Pro with bead depletion of CD14+ and CD25+ cells (Miltenyi), followed by positive selection of CD4+ cells. Cord blood CD4+ T cells were over 97% pure and 100% CD45RA+ and were therefore used for initial transduction experiments. Adult CD4+ T cell subsets and naive cord blood CD4+ T cells were further purified as CD3+CD4+CD25- CD45RO- or CD3+CD4+CD25- CD45RO+ and CD3+CD4+CD25- HLA-DR- CD45RA+, respectively, by cell sorting on a FACSAria (BD).
Cell culture and lentiviral transduction.
Cells were cultivated in an incubator at 37 °C and 5% CO2 in RPMI 1640 medium (Invitrogen) supplemented with 10% (vol/vol) FBS (Hyclone), penicillin-streptomycin, 2 mM glutamine, 10 mM HEPES, 1 mM pyruvate and 0.1 mM nonessential amino acids or in XVIVO-20 serum-free medium (Lonza) supplemented with penicillin-streptomycin. CD4+ T cells were stimulated by the addition of anti–mouse immunoglobulin G magnetic beads (Pierce) previously coated with purified anti-CD3 and anti-CD28 at a final concentration of one bead per cell and a concentration of 1 
g/ml of each antibody. For transduction experiments, cells were seeded on day 0 at a density of 1 
106 cells per ml in 24-well plates with polybrene (10 g/ml), IL-2 (10 U/ml) and beads coated with anti-CD3 and anti-CD28. Lentiviral supernatants were added at a multiplicity of infection of 1–10. Cells were washed on day 1 and were split as needed in the presence of IL-2. For shRNA experiments, puromycin (2 g/ml) was added on day 2. For polarization experiments, cells were seeded at a density of 5 105 to 5 106 cells per ml in U-bottomed 96-well plates with beads coated with anti-CD3 and anti-CD28. IL-2 (10 U/ml) was added on day 0 or 3. The medium was replaced on day 3 and cells were split in the presence of IL-2. For long-term experiments, cells were split as needed. In some cases, IL-1 (10 ng/ml; eBioscience), IL-6 (10 ng/ml; eBioscience), IL-21 (10 ng/ml; Cell Sciences), IL-23 (10 ng/ml; eBioscience), tumor necrosis factor (10 ng/ml; eBioscience), IL-4 (10 ng/ml; eBioscience), various concentrations of TGF-1 (PeproTech), IL-21-neutralizing soluble IL-21 receptor (R&D Systems) and neutralizing antibody to IL-2, IL-4, IL-6, IFN- or TGF- (1 g/ml except where noted otherwise; Supplementary Table 1 online) were added at day 0 and were maintained throughout the experiment. Cells were collected on day 6 for intracellular staining and real-time PCR analysis except where specified otherwise.
Surface and intracellular staining.
For intracellular cytokine staining, cells were incubated for 5 h with phorbol ester (50 ng/ml; Sigma), ionomycin (500 ng/ml; Sigma) and GolgiStop (BD). Surfaces were stained by incubation for 15 min on ice with the corresponding fluorescence-labeled antibodies (Supplementary Table 1). The Cytofix/Cytoperm buffer set (BD) was used for intracellular staining. Cells were fixed and made permeable for 30 min at 21 °C and were stained for 30 min at 21 °C in permeabilization buffer. An LSR II (BD Biosciences) and FlowJo software (Tree Star) were used for flow cytometry. Foxp3 staining buffers (eBioscience) were used for Foxp3 staining. Antibody FJK-16s (originally reported as an antibody to mouse and rat Foxp3) was used; it robustly stained endogenous and overexpressed full-length human Foxp3 (data not shown).
Plasmids and lentiviral production.
The gene encoding human RORt was cloned from human thymus. A double Flag tag was added to the amino terminus. The gene encoding human ROR isoform d was cloned from peripheral CD4+ T cells. The gene encoding human ROR was cloned from U937 cells. The resultant cDNA was cloned into a vector consisting of human immunodeficiency virus–derived vector (HDV), an internal ribosome entry site (IRES) and a tag of heat-stable antigen (HSA) or green fluorescent protein (GFP): HDV-IRES-HSA or HDV-IRES-GFP65. Lentiviral vectors expressing human GATA-3 and T-bet have been described57. Vectors expressing shRNA were from OpenBiosystems; shRNA-1 (TRCN33657) target sequence, TCTGCAAGACTCATCGCCAAA), shRNA-2 (TRCN33658) target sequence, CGAGGATGAGATTGCCCTCTA), and pLKO.1puro, as control. Viral supernatants were produced by transient transfection of HEK293T cells with a combination of an shRNA vector, VSV-G expression plasmid and the packaging plasmid pCMV
R8.9. Viral particles were concentrated by ultracentrifugation for 2 h at 4 °C and 25,000 r.p.m (SW28 rotor), then were resuspended in PBS containing 1% (wt/vol) BSA, divided into aliquots and frozen.
Real-time PCR.
RNA was extracted with TRIzol (Invitrogen) and cDNA was synthesized with Superscript II (Invitrogen) and random primers. The iQ SYBR Green Supermix (Bio-Rad) or QuantiTect Multiplex PCR mix (Qiagen) and the iCycler Sequence Detection System (Bio-Rad) were used for analysis of cDNA by real-time quantitative PCR in triplicate (primer sets, Supplementary Table 2 online). The starting quantity of the initial cDNA sample was calculated from primer-specific standard curves with iCycler Data Analysis Software. The expression of each gene was normalized to the expression of ACTB (encoding -actin) by the standard curve method. 'Fold changes' were calculated by normalization to the first sample of each set. Error bars were calculated based on triplicate measurements for each gene.
Accession code.
UCSD-Nature Signaling Gateway (http://www.signaling-gateway.org): A002271.
Note: Supplementary information is available on the Nature Immunology website.
Author contributions
N.M., D.U. and D.R.L. designed experiments; N.M. did all experiments and N.M. and D.R.L. wrote the manuscript.