Some pathways of T cell differentiation are associated with characteristic patterns of chemokine receptor expression. A new lineage of effector/memory CD4+ T cells has been identified whose signature products are IL-17 cytokines and whose differentiation requires the nuclear receptor, RORγt. These Th17 cells are critical effectors in mouse models of autoimmune disease. We have analyzed the association between chemokine receptor expression and IL-17 production for human T cells. Activating cord blood (naive) CD4+ T cells under conditions driving Th17 differentiation led to preferential induction of CCR6, CCR9, and CXCR6. Despite these data, we found no strong correlation between the production of IL-17 and expression of CCR9 or CXCR6. By contrast, our analyses revealed that virtually all IL-17-producing CD4+ T cells, either made in our in vitro cultures or found in peripheral blood, expressed CCR6, a receptor found on ∼50% of CD4+ memory PBL. Compared with CD4+CD45RO+CCR6− cells, CD4+CD45RO+CCR6+ cells contained at least 100-fold more IL-17A mRNA and secreted 100-fold more IL-17 protein. The CCR6+ cells showed a similar enrichment in mRNA for RORγt. CCR6 was likewise expressed on all IL-17-producing CD8+ PBL. CCR6 has been associated with the trafficking of T, B, and dendritic cells to epithelial sites, but has not been linked to a specific T cell phenotype. Our data reveal a fundamental feature of IL-17-producing human T cells and a novel role for CCR6, suggesting both new directions for investigating IL-17-related immune responses and possible targets for preventing inflammatory injury.
Antigen activation drives naive T cells down any of a number of pathways of differentiation to yield highly heterogeneous populations of effector/memory cells. Heterogeneity is a particularly prominent feature of the effector/memory CD4+ T cell population, which includes subsets capable of producing polarized patterns of cytokines that serve specialized functions and have profound effects on the quality of the immune response (1). The varied effector/memory phenotypes are the consequences of changes in gene activities and related remodeling of chromatin, processes for which a number of transcription factors have been identified as master regulators (2). Among the genes induced as part of the programs of differentiation are those encoding adhesion molecules and chemoattractant receptors, which are responsible for the alterations in patterns of migration for effector/memory vs naive cells (3, 4).
The complexity of the T cell effector/memory population is reflected in the expanded repertoire of chemokine receptors, from two or three species on naive CD4+ or CD8+ T cells, respectively, to at least fifteen species of chemokine receptors expressed in various combinations on subsets of effector/memory cells. Although the mechanisms underlying the regulation of chemokine receptor expression on T cell subsets are poorly understood, some patterns of receptor expression have been associated with well established phenotypes and pathways of differentiation. For example, CXCR3 is expressed preferentially on human Th1 cells (5, 6, 7, 8), and its expression is driven by T-bet (9, 10), the transcription factor that is critical for Th1 differentiation (2). Conversely, CCR4 is expressed preferentially on Th2 cells (5, 6, 7, 8), and its expression can be driven by GATA-3 (9), the transcription factor that is critical for Th2 differentiation (2). It is likely that these receptors are components of positive feedback loops to reinforce type 1 or type 2 responses in tissues, because CXCR3 ligands are induced by IFN-γ (11), the signature cytokine of Th1 cells, and CCR4 ligands are induced by IL-4 (12), the signature cytokine of Th2 cells.
Over the last several years, IL-17 and IL-17-producing effector/memory T cells have been recognized as critical for producing tissue damage in mouse models of autoimmune disease, as well as in host defense against bacterial infection (reviewed in Ref. 13 , 14). Much of the inflammatory damage previously ascribed to the type 1 response is now understood to depend on IL-17 and on IL-23, the cytokine important for supporting the Th17 response in vivo (15, 16). Recently, it has been recognized that IL-17-producing CD4+ T cells (Th17 cells) form a lineage separate from Th1 and Th2 cells, and that, in fact, the production of Th17 cells is inhibited by factors that support Th1/Th2 differentiation (17, 18). Differentiation down the Th17 pathway depends on the orphan nuclear receptor RORγt (19), which likely serves as a master transcriptional regulator for Th17 cells.
There is a strong connection between the Th17 response and the chemokine system, because a major aspect of IL-17-driven inflammation is neutrophil recruitment, for which the induction of CXC chemokines is an important component (13). An unanswered question, however, is whether or not there is a pattern of chemokine receptor expression/chemokine responsiveness that is characteristic of IL-17-producing T cells. This question is relevant for understanding the mechanisms of IL-17-induced inflammation because IL-17 and/or IL-17-producing T cells are found in the affected tissues and it is presumed that the cytokine’s presence at the site is critical for producing tissue damage (16, 20, 21, 22, 23). It is possible, therefore, that blocking the entry of Th17 cells into tissue – for example, by inhibiting one or more chemokine receptors – might be an effective way of ameliorating inflammatory damage. In the data presented below, we demonstrate preferential induction and function of chemokine receptors CCR6, CCR9, and CXCR6 on cord blood CD4+ T cells activated under conditions yielding IL-17-producing cells vs conditions yielding nonpolarized, Th1, or Th2 cells. Although we could not demonstrate a correlation between a cell’s ability to produce IL-17A and its expression of CCR9 or CXCR6, we found that virtually all IL-17-producing T cells, either made in our in vitro cultures or found in peripheral blood, express CCR6.
CCR6 is expressed on B cells and on subsets of T cells and dendritic cells and has only one chemokine ligand, CCL20/MIP-3α, whose only receptor is CCR6 (reviewed in Ref. 24). CCR6 has been reported to have roles in host defense and inflammation particularly at epithelial sites (25, 26, 27, 28, 29, 30, 31, 32, 33). Our data suggest that up-regulation and stable expression of CCR6 is a fundamental feature of Th17 differentiation and raise the possibility that blocking CCR6 might be of benefit in treating tissue damage mediated by T cells that produce IL-17.
Materials and Methods
Anti-CD3 (clone OKT3) was obtained from Ortho Biotech, and anti-CD28 (clone CK248 (34
In vitro activation of naive cord blood CD4+ T lymphocytes
Human cord blood was obtained from term placentas following the delivery of healthy newborns at Shady Grove Adventist Hospital (Gaithersburg, MD) as approved by that institution’s review board. Naive CD4+ T cells were isolated from cord blood with the RosetteSep/human CD4+ T cell reagent (StemCell Technologies). The purity of CD4+ T cells was >95%. Cells were cultured in RPMI 1640 supplemented with 2 mM l-glutamine, penicillin, and streptomycin, and 10% FBS. CD4+ T cells were stimulated with plate-bound anti-CD3 (10 μg/ml), and soluble anti-CD28 (1 μg/ml). In addition, we used the following combinations of cytokines and Abs: for nonpolarizing conditions: IL-2 (200 IU/ml), anti-IL-4 (0.4 μg/ml), anti-IL-12 (2 μg/ml), anti-IFN-γ (8 μg/ml), and TGF-β1.2 (10 ng/ml); for Th1 conditions: IL-2 (200 IU/ml), IL-12 (2 ng/ml) and anti-IL-4 (0.4 μg/ml); for Th2 conditions: IL-2 (200 IU/ml), IL-4 (4 ng/ml), anti-IL-12 (2 μg/ml), and anti-IFN-γ (8 μg/ml); and for Th17 conditions: IL-23 (100 ng/ml), IL-6 (20 ng/ml), IL-1β (10 ng/ml), TNF-α (10 ng/ml), TGF-β 1.2 (10 ng/ml), anti-IL-4 (0.4 μg/ml), anti-IL-12 (2 μg/ml), and anti-IFN-γ (8 μg/ml). After 4 days, proliferating cells were washed in PBS and expanded under the same conditions in the absence of anti-CD3 and anti-CD28.
Staining for cell surface receptors and intracellular cytokines
For phenotype analysis of cell surface proteins, cells were stained with mAbs for 15 min at room temperature in FACS buffer (HBSS containing 1% FBS), washed twice with same buffer and fixed in 1% paraformaldehyde. For intracellular cytokines, cells were stimulated with Leukocyte Activation Cocktail, with GolgiPlus (BD Pharmingen) for 5 h at 37°C before being stained with fluorophore-conjugated anti-IFN-γ and anti-IL-17A using Cytofix/CytoPerm Plus kit (BD Pharmingen) and analyzed on a FACSCalibur cytometer with CellQuest software or on an LSR II cytometer with FACSDiva software (BD Biosciences). Flow cytometry data were analyzed using FlowJo software (Tree Star). For quantifying stained cells in dot plots, quadrants were drawn based on the signals using isotype controls (data not shown).
Purification and sorting of lymphocyte subsets
For cells polarized in vitro under Th17 conditions, CD4+CCR6− and CD4+CCR6+ subsets were purified by cell sorting. For lymphocyte subsets analyzed directly after isolation from peripheral blood, we started with elutriated lymphocytes obtained from healthy donors by the Department of Transfusion Medicine, Clinical Center, National Institutes of Health (Bethesda, MD) under an institutional review board-approved protocol. CD4+ and CD8+ T cells were purified with the RosetteSep reagents (StemCell Technologies). Subsets were sorted after staining with fluorophore-conjugated anti-CD4, anti-CD45RO, anti-CCR6, anti-CCR9 and anti-CXCR6. Cell sorting was done using a FACS Aria (BD Biosciences), and the purity of sorted populations was 95–99%.
Isolation of RNA and real-time fluorogenic RT-PCR
Chemotaxis assays on cultured CD4+ T cells were performed using Transwells (Costar). Following culturing for 6 days, aliquots of one million cells were placed in inserts on membranes containing 5 μm pores over wells with or without chemokines. Following incubation at 37°C in 5% CO2 for 3 h, cells in the lower wells were collected and counted using a Vi-CELL analyzer (Beckman Coulter).
Cytokine production and ELISA
Subsets of sorted cells were stimulated with 20 ng/ml PMA, and 1 μM ionomycin for 6 h at 37°C before supernatants was collected and total IL-17 and TNF-α production was determined using ELISA by Searchlight Technology, Pierce Biotechnology.
Statistical significance was calculated by Student’s t test using Prism software (GraphPad). All p values <0.05 were considered significant.
Th17 differentiation of human CD4+ T cells from cord blood preferentially up-regulates CCR6, CCR9, and CXCR6
We activated human cord blood CD4+ T cells using plate-bound anti-CD3 and soluble anti-CD28 under conditions for producing Th1, Th2, and nonpolarized cells, and in addition under conditions adapted from protocols for producing mouse Th17 cells (35). Results of intracellular staining for IFN-γ and IL-17A after 6 days of culture (and activation with PMA and ionomycin) demonstrated the expected results for the nonpolarizing, Th1, and Th2 conditions, and a small percentage of cells staining for IL-17A under the “Th17” conditions (Fig. 1⇓A). Despite the small number of IL-17A+ cells by staining, cells cultured under Th17 conditions showed levels of IL-17A mRNA >1,000-fold higher than cells cultured under the other conditions (Fig. 1⇓B). In addition, the Th17 conditions produced dramatic induction of mRNA for RORγt (Fig. 1⇓C). Together, the data suggest that some cells cultured under the Th17 conditions did, in fact, differentiate along a Th17 pathway, likely in larger numbers than could be identified by staining for IL-17A.
Using cord blood CD4+ T cells cultured for 6 days as in Fig. 1⇑, we analyzed chemokine receptor expression by flow cytometry (Fig. 2⇓A). Levels of expression on cells cultured under Th1 vs Th2 conditions were as expected for CCR4, CCR5 and CXCR3, receptors that are known to be expressed differently on Th1 vs Th2 cells (see above). Three receptors, CCR6, CCR9, and CXCR6 showed a reproducible pattern of particular interest, being expressed on >10% of “Th17” cells and, most importantly, being expressed preferentially on cells cultured under Th17 vs the three other conditions.
The surface staining correlated with receptor-mediated chemotaxis, as shown in Fig. 2⇑B. Migration to CCL20, CCL25, and CXCL16, chemokines specific for CCR6, CCR9, and CXCR6, respectively, was greatest for cells cultured under Th17 conditions. No differences in migration of cells were seen in the absence of chemokine, and chemotaxis to the CXCR3 ligand, CXCL9, was greatest for “Th1” and lowest for “Th2” cells, also consistent with the results in Fig. 2⇑A. CCR9 is a receptor expressed both on early thymocytes, and of greater relevance here, on memory/effector T cells that are activated by gut-associated dendritic cells and imprinted to traffic to the small bowel (36). CXCR6 is a receptor expressed preferentially on Th1 cells, and although, like CCR9, CXCR6 is found on only a small percentage of T cells in the peripheral blood (see Fig. 4⇓E), CXCR6 is expressed on significant numbers of T cells at inflammatory sites (37). CCR6 will be described in greater detail below.
Cultured CD4+ T cells producing IL-17A are found only within the CCR6+ subset
We investigated the relationships among chemokine receptor expression and the ability to produce IL-17 by costaining for chemokine receptors and IL-17A after activating the Th17-cultured cells with PMA and ionomycin. Although results from costaining were inconclusive due to variable loss of chemokine receptor staining post activation and the relatively low numbers of IL-17A+ cells, these experiments suggested a correlation between CCR6 and the ability to produce IL-17A (data not shown). To evaluate the CCR6-expressing cells further, we sorted the cells polarized under Th17 conditions into CD4+CCR6− and CD4+CCR6+ subsets. Purified cells were activated immediately and analyzed for production of IL-17A and IFN-γ by intracellular staining. As shown in Fig. 3⇓A, essentially all the IL-17A+ cells were present in CCR6+ subset. We also analyzed the purified subsets for mRNA for IL-17A and RORγt as shown in Fig. 3⇓B and C, respectively. There were >500-fold enrichments for these mRNAs in the CCR6+ vs CCR6− subsets.
Peripheral blood T cells producing IL-17 are found only within the CCR6+ subset
We next analyzed subsets of CD4+ T cells from peripheral blood based on the expression of CCR6, CCR9, and CXCR6. We purified CD4+ T cells into CD45RO− and CD45RO+ samples, and further purified CD45RO+ cells into subsets either positive or negative for the relevant chemokine receptors. Purified cells were activated immediately ex vivo and analyzed for mRNAs and/or for protein production as shown in Fig. 4⇓. Virtually all the mRNA for IL-17A was found within the CD45RO+ cells, and virtually all the signal from these cells was found within the CCR6+ subset, giving a ∼100-fold difference between the CD45RO+CCR6+ vs CD45RO+CCR6− cells (Fig. 4⇓A), where differences are necessarily limited by the purities of the sorted samples. Small degrees of enrichment for IL-17A mRNA were also seen for the CCR9+ vs CCR9− and CXCR6+ vs CXCR6− CD45RO+ cells, but the fold differences were minimal as compared with the results for cells sorted based on CCR6. A high level of enrichment was also found in CD45RO+CCR6+ vs CD45RO+CCR6− cells for expression of the mRNA encoding RORγt (Fig. 4⇓B).
We analyzed the purified subsets for production of IL-17A and IFN-γ by intracellular staining following activation as shown in Fig. 4⇑C. In parallel with the results for mRNA, essentially all the IL-17A+ cells were in the CD45RO+CCR6+ subset. For this donor, no enrichment in IL-17A+ cells was seen in the CCR9+ vs CCR9− subsets, whereas there was a two-fold increase in IL-17A+ cells in the CXCR6+ vs CXCR6− subsets. Analysis of supernatants collected from the purified subsets after six hours of stimulation showed a pattern similar to the results from intracellular staining. CD45RO+CCR6+ cells secreted >100-fold the IL-17 produced by the CD45RO+CCR6− cells (Fig. 4⇑D). In addition, TNF-α, which is produced by Th17 cells (14), was found at levels ∼6-fold higher in supernatants from CD45RO+CCR6+ vs CD45RO+CCR6− cells (data not shown).
The weak relationships of production of IL-17 with expression of CXCR6 (and to an even lesser extent with expression of CCR9) vs the strong association with expression of CCR6 led us to analyze patterns of coexpression for CXCR6 and CCR9 vs CCR6. Staining an aliquot of the cells used for Fig. 4⇑C, as shown in Fig. 4⇑E, revealed frequencies of CCR6+ cells that were equal for CCR9+ vs CCR9− cells and 2- to 3-fold higher for CXCR6+ vs CXCR6− cells – relative frequencies that correlated with the distributions of IL-17A+ cells within these subsets. It is of interest that within the CXCR6+ population, the majority of IL-17A+ cells also express IFN-γ, whereas the opposite is the case within the (CCR6+) population expressing IL-17A as a whole. In fact, the enrichment of IL-17A-producing cells seen in the CXCR6+ vs CXCR6− populations is wholly within the subset expressing IL-17A plus IFN-γ, in line with an overall enrichment in IFN-γ+ cells in CXCR6+ vs CXCR6− populations.
Finally, we analyzed the relationship between CCR6 expression and production of IL-17A by costaining for chemokine receptor and cytokine in unsorted populations of both CD4+ and CD8+ T cells purified from peripheral blood. As for the cells cultured in vitro, activation with PMA/ionomycin led to significant loss of staining for CCR6. Nonetheless, as shown in Fig. 5⇓A, all the IL-17A+ cells remained within the CCR6+ population. Although there are lower percentages CCR6+ (and IL-17A+) cells in the CD8+ vs CD4+ T cell population, just as for CD4+ T cells, the IL-17A-producing CD8+ T cells were also found exclusively within the CCR6+ subset (Fig. 5⇓B). Fig. 5⇓ also demonstrates that no IL-17A can be detected in cells from the peripheral blood in the absence of activation ex vivo.
Our data demonstrate a powerful, one-way association between the ability to produce IL-17 and the expression of CCR6 by human T cells. Virtually all the IL-17-producing cells that we analyzed expressed CCR6 but not the converse. Activation of cord blood CD4+ T cells under conditions that produced Th17 cells preferentially induced CCR6 expression and responsiveness to CCL20 as compared with conditions that produced nonpolarized, Th1, or Th2 cells. Although we used combinations of cytokines for producing Th17 cells that have worked well using cells from mice (35), the percentages of IL-17-staining cells in our cultures were low. Nonetheless, we found dramatic induction of mRNAs for IL-17A and RORγt after activation of cells from the Th17 cultures, and the cells staining for IL-17A were found exclusively in the CCR6-expressing subset, matching our findings with freshly isolated PBL. Taken together, we believe that despite the low percentage of IL-17A-staining cells, our data from the in vitro cultures are informative. Moreover, at the time our manuscript was submitted, as far as we knew, there were no published data on generating Th17 cells from naive human T cells. Very recent work (38, 39) on this problem has yielded conflicting results, so that the critical and optimal factors for the differentiation of human Th17 cells are as yet undefined.
Although the in vitro cultures served as a useful screen, such cultures provide inherently imperfect replicas of processes occurring in vivo. Our most informative results, therefore, came from analysis of CD4+ and CD8+ T cells directly from peripheral blood, which revealed that the IL-17 producing cells were also confined to the CCR6+ subsets. In addition, the CCR6+CD4+ T cells expressed high levels of the mRNA for RORγt, the transcription factor identified as critical for Th17 differentiation (19). All our assays of sorted subsets of CD4+ T cells for IL-17 mRNA and/or protein, and for mRNA for RORγt, showed ratios between the CCR6+ vs CCR6− cells that approached the limits of purity obtained by cell sorting.
Our in vitro data suggested that CCR9 and CXCR6 might also be expressed preferentially on IL-17-producing CD4+ T cells. Consistent with these findings, some CCR9+ lamina propria lymphocytes have been reported to produce IL-17 (40). However, we noted no significant increase in IL-17-producing cells in CCR9+ vs CCR9− populations of PBL. For CD4+CXCR6+ PBL, there was a 2- to 3-fold enrichment in IL-17 production in the receptor+ vs receptor− cells. The results for cells expressing CCR9 or CXCR6 correlated with the frequencies of CCR6+ cells within the various subsets, findings that support the strong association between CCR6 expression and the ability to produce IL-17. Within the IL-17A+ cells, the CXCR6+ subset was enriched for cells coproducing IFN-γ. This paralleled an overall enrichment in IFN-γ+ cells in CXCR6+ vs CXCR6− populations, consistent with published data (37) for CD4+CXCR6+ cells in humans. Although our data suggest that CCR9 and CXCR6 may be induced preferentially on CD4+ T cells activated under Th17-inducing conditions, we found no strong and/or independent association between expression of these receptors and the ability to produce IL-17 by individual cells, either for cells acutely activated, or for resting cells isolated from peripheral blood. In toto, our data demonstrate that IL-17A-producing human T cells – at least those made from cord blood cells in vitro and those isolated from adult peripheral blood – are found wholly within the subset expressing CCR6, and suggest no similar correspondence between IL-17A production and any of the other fourteen chemokine receptors that we analyzed in vitro and/or directly ex vivo.
Our data lead us to consider possible relationships between what is known about the function of CCR6 and the biology of IL-17-producing cells. CCR6 was identified initially on T cells, dendritic cells, and B cells and found to have CCL20 as its single chemokine ligand (reviewed in Ref. 24). In mouse models of immune response and disease, CCR6 and/or CCL20 have been reported to have roles in Ab production and pathogen-induced T cell activation in the gut (25, 26, 27); delayed-type and contact hypersensitivity in the skin (28, 29); allergic inflammation in the lung (30); inflammatory bowel disease (31, 32); and graft-vs-host disease (33), models which depend on B cell and/or T cell and/or dendritic cell trafficking at epithelial sites. In humans, CCR6 can function to mediate arrest of T cells on dermal endothelial cells (41) and is highly expressed on T cells resident in both normal (42) and psoriatic (43, 44) skin, and CCR6 and/or CCL20 have been implicated in the pathogenesis of rheumatoid arthritis (45) and inflammatory bowel disease (46). Psoriasis, rheumatoid arthritis, and inflammatory bowel disease are all autoimmune disorders in which Th17 cells are thought to play a critical part (13), and our data suggest that CCL20 and CCR6 may have a role in these disorders by recruiting Th17 cells to target tissues.
Because IL-17-driven inflammation is characterized by neutrophil infiltration, it is of interest that CCR6+ T cells in skin have been reported to produce CXCL8 (44) and that neutrophils have been reported both to produce CCL20 (47) and to express CCR6 following exposure to TNF-α (48). Moreover, IL-17 has been reported to be a potent inducer of CCL20 (45). Together, these data suggest that CCL20 and CCR6 contribute to a positive feedback loop involving T cells, dendritic cells, neutrophils, and epithelial/resident cells that amplifies the Th17 inflammatory response. The CCL20/CCR6 loop would be analogous to those involving CXCR3 and its ligands and CCR4 and its ligands for type 1 and type 2 inflammation, respectively. An intriguing aspect of Th17 differentiation is the requirement for TGF-β, which can also induce the production of CD4+CD25+ regulatory T cells (49), and which is a major mediator of regulatory T cell activity (50). TGF-β has been reported (51) to up-regulate CCR6, and CCR6 has been found (52, 53) on a significant percentage of regulatory T cells, so that the expression of CCR6 may reflect another connection between regulatory T cells and Th17 cells and, by mediating their colocalization, may contribute to the interactions that have been postulated to occur between these subsets (54).
There have been some previous reports describing relationships between the production of IL-17 and the expression of CCR6. Using immunocytochemistry, one group found both CCR6+ and CCR6− IL-17+ cells in cytospin preparations of human PBMC after activation ex vivo and in rheumatoid synovium (55). Finding some CCR6− IL-17+ cells on the activated PBMC may have been due to activation-induced down-regulation of CCR6, as we describe above. A similar phenomenon occurring in rheumatoid synovium might also explain the findings in that tissue, although others have now reported that all Th17 cells in synovial fluid from patients with juvenile idiopathic arthritis express CCR6 (56 , see below).
A second paper analyzed the expression of an array of surface proteins, including chemokine receptors, on TCR transgenic mouse T cells after 3 days of activation with Ag in the presence of IL-12 vs IL-23 (57). In contrast to our results using cells activated/differentiated in vitro, the data in that paper showed no difference in CCR6 expression between cells cultured with the two cytokines. Up-regulation of CCR6 was seen on all cells activated in the presence of either IL-12 (“Th1 conditions”) or IL-23 (“Th17 conditions”). The differences between these data and our results could be due to any of the many experimental differences – in species, activation protocols, and timing of analyses. Moreover, the finding that CCR6 can be up-regulated under Th1 conditions would be consistent with our data that CCR6 can be found not only on IL-17-producing cells, but also on some CD4+ PBL that can produce IFN-γ with or without IL-17A.
While this manuscript was under review, several additional publications appeared addressing chemokine receptor expression on human Th17 cells (56, 58, 59). Two of these papers (56, 59) reported similar findings to our own regarding the presence of human Th17 PBL solely within the CCR6+ subset. Also similar to our findings, Annunziato et al. (59) reported high expression of CXCR6 on Th17 clones, although they did not report any preferential expression of CXCR6 on Th17 vs Th1 or Th2 clones. Acosta-Rodriguez et al. (56) subdivided the CCR6+ population into a CCR4+ subset containing IL-17A+IFN-γ− but not IL-17A+IFN-γ+ cells and a CXCR3+ subset containing both IL-17A+IFN-γ− and IL-17A+IFN-γ+ cells. They identified only the IL-17A+IFN-γ− pattern as characteristic of Th17 cells. Questions of nomenclature and biology of IL-17+IFN-γ− vs IL-17+IFN-γ+ cells aside, these data might be interpreted as indicating an important, independent relationship between IL-17 production/the Th17 lineage and expression of CCR4, which the authors did not address directly. In fact, IL-17A+IFN-γ− cells can readily be found in the subset of CD4+ PBL that are CCR6+ and CCR4− (or CCR4+) (S.P.S. and J.M.F., unpublished data).
Finally, the paper by Sato et al. (58) did not examine CCR6, but reported increased production of IL-17 vs IFN-γ in CCR2+CCR5− vs CCR2+CCR5+ CD4+ T cells. Sato et al. (58) did not provide direct data on frequencies of IL-17-producing cells within the various CD4+ subsets. We have found that although, as compared with the CCR2+CCR5+ subset, the CCR2+CCR5− subset does contain a higher percentage of Th17 cells, only a minority of the CCR2+CCR5− cells could be induced to make IL-17A and only a small percentage of the total Th17 cells in peripheral blood were found within the CCR2+CCR5− subset, a subset that typically contains only 1–2% of the CD4+CD45RO+ cells (S.P.S., H.H.Z., and J.M.F., unpublished data). As might be predicted, as compared with the CCR2+CCR5+ subset, the CCR2+CCR5− subset contained a higher percentage of CCR6+ cells (S.P.S. and J.M.F., unpublished data). Taken together, the critical point from all these recent data is the unique relationship between the production of IL-17 and the expression of CCR6, and not the other chemokine receptors analyzed.
This strong association between IL-17 production and the expression of CCR6 on human T cells raises a number of additional questions that suggest new areas for research. Do all CCR6-expressing cells, such as B cells and subsets of dendritic cells, that would be corecruited with IL-17-producing T cells contribute to IL-17-driven immune responses? Are the functions of these CCR6+ cells affected by the products of Th17 cells and visa versa? Is the CCR6 gene a direct target of RORγt? What does the regulation of CCR6 expression tell us about the pathway of Th17 differentiation? Are CCR6/CCL20 critical for host defense in infections where IL-17 is essential? Do CCR6/CCL20 make significant contributions to IL-17-driven tissue inflammation? Answering these questions will enhance our understanding of IL-17-related immune responses and the functions of CCR6 (and CCL20), and potentially identify novel targets for treating autoimmune diseases.
We thank Calvin Eigsti and other members of the Flow Cytometry Section, Research Technology Branch, National Institute of Allergy and Infectious Diseases, National Institutes of Health, for their help with cell sorting.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
↵1 This research was supported by the Intramural Research Program of the National Institutes of Health, National Institute of Allergy and Infectious Diseases.
↵2 Address correspondence and reprint requests to Dr. Joshua M. Farber, Laboratory of Molecular Immunology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Building 10, Room 11N-111, 10 Center Drive, Bethesda, MD 20892. E-mail address:
- Received April 24, 2007.
- Accepted October 29, 2007.
- Copyright © 2008 by The American Association of Immunologists