1. Vol 441|11 May 2006|doi:10.1038/nature04754
LETTERS
Transforming growth factor-b induces development
of the TH17 lineage
Paul R. Mangan1,2, Laurie E. Harrington1, Darrell B. O’Quinn1, Whitney S. Helms1, Daniel C. Bullard3,
Charles O. Elson4, Robin D. Hatton1, Sharon M. Wahl5, Trenton R. Schoeb3 & Casey T. Weaver1,2
A new lineage of effector CD41 T cells characterized by pro- development under IFN-g-null conditions. Naive Ifng 2/2 T cells
duction of interleukin (IL)-17, the T-helper-17 (TH17) lineage, were activated by Ifng 2/2 splenocytes as before under TH2-neutra-
was recently described based on developmental and functional lizing conditions (Fig. 1b). Addition of IL-23 alone induced few
features distinct from those of classical TH1 and TH2 lineages1,2. IL-17þ cells, indicating that in the absence of IFN-g, IL-23 alone is
Like TH1 and TH2, TH17 cells almost certainly evolved to provide insufficient to promote robust TH17 development. Addition of
adaptive immunity tailored to specific classes of pathogens3, such TGF-b1 promoted a substantially increased fraction of IL-17þ cells,
as extracellular bacteria4. Aberrant TH17 responses have been which was only modestly augmented by co-addition of exogenous
implicated in a growing list of autoimmune disorders5–7. TH17 IL-23. Therefore, TGF-b1 can act independently of IFN-g blockade
development has been linked to IL-23, an IL-12 cytokine family to promote TH17 development. Notably, reconstitution of IFN-g-
member that shares with IL-12 a common subunit, IL-12p40 deficient cultures with high levels of exogenous IFN-g strongly
(ref. 8). The IL-23 and IL-12 receptors also share a subunit, inhibited TH17 development despite abundant TGF-b1 and IL-23.
IL-12Rb1, that pairs with unique, inducible components, IL-23R Similar results were found for IL-4 (Supplementary Fig. S1). Thus,
and IL-12Rb2, to confer receptor responsiveness9. Here we identify TGF-b, IFN-g and IL-4 act antagonistically to specify TH17, TH1 or
transforming growth factor-b (TGF-b) as a cytokine critical for TH2 development, respectively.
commitment to TH17 development. TGF-b acts to upregulate Although the foregoing experiments identified an essential role for
IL-23R expression, thereby conferring responsiveness to IL-23. TGF-b1 in TH17 development, they did not exclude the possibility
Although dispensable for the development of IL-17-producing that TGF-b1 acts together with endogenous IL-23. We therefore used
T cells in vitro and in vivo, IL-23 is required for host protection splenocytes from IL-12p40-deficient (Il12b 2/2; hereafter called
against a bacterial pathogen, Citrobacter rodentium. The action of p40 2/2) mice as a source of IL-23 (and IL-12)-deficient antigen-
TGF-b on naive T cells is antagonized by interferon-g and IL-4, presenting cells (APCs) with which to examine TH17 development
thus providing a mechanism for divergence of the TH1, TH2 and under defined conditions of IL-23 availability. Without IL-23 and
TH17 lineages. exogenous TGF-b1, few IL-17-producing T cells were generated, and
Interferon-g (IFN-g) potently inhibits TH17 development1,2. IL-23 alone did not restore TH17 development (Fig. 1c). Surprisingly,
Given the suppressive actions of TGF-b on IFN-g production10–12, addition of TGF-b1 was sufficient to induce robust TH17 development
we speculated that TGF-b might contribute to TH17 development by in the absence of IL-23, and development of IL-17-producing T cells
limiting inhibitory actions of IFN-g. Naive CD4þ T cells were was only modestly enhanced by co-addition of IL-23. Under more
therefore activated under TH2-neutralizing conditions and con- stringent conditions of IFN-g signalling deficiency, in which IFN-g
trolled availability of IL-23 and IFN-g, with or without exogenous receptor-1-deficient (Ifngr 2/2) T cells were used, more striking
TGF-b1, and cytokine phenotypes were examined (Fig. 1). Addition TGF-b1-dependent, IL-23-independent TH17 development was
of IL-23 did not substantially enhance development of IL-17þ cells observed (Supplementary Fig. S2). Thus, TGF-b1 acts independently
unless endogenous IFN-g was neutralized (Fig. 1a, top panel). of IL-23 to induce TH17 lineage commitment.
Addition of TGF-b1 alone reduced the fraction of IFN-gþ T cells The IL-23 receptor is a heterodimer of IL-12Rb1, which is con-
by more than twofold and induced the development of a small, but stitutively expressed by naive T cells, and IL-23R, which is not9. In
appreciable, fraction of IL-17þ cells (Fig. 1a, bottom panel). In the view of the foregoing results, we speculated that TGF-b might act
presence of exogenous IL-23, TGF-b1 suppressed IFN-g induction proximally in TH17 development to induce IL-23R upregulation,
similarly to that in the absence of added IL-23, while modestly analogous to the induction of IL-12Rb2 by IFN-g during TH1
increasing IL-17þ cells. Under conditions of IL-23 addition and development13. We therefore compared expression of IL-12 and
IFN-g neutralization, exogenous TGF-b1 induced further suppres- IL-23 receptors during TH1 or TH17 development. Naive T cells
sion of IFN-gþ cells compared to that of IL-23 addition alone and, from Ifng 2/2 mice were activated under neutral cytokine conditions,
notably, induced a markedly increased fraction of IL-17þ cells. or under T H17- or T H1-polarizing conditions, and relative
Importantly, a similar induction of IL-17þ cells was found irrespec- expression of IL-12Rb1, IL-12Rb2 and IL-23R messenger RNA
tive of exogenous IL-23 addition, suggesting that endogenous levels determined (Fig. 1d). Compared to neutral conditions, TH1 polar-
of IL-23 were either adequate or non-contributory. Collectively, ization induced IL-12Rb2 expression while inhibiting IL-23R
these data indicate that in addition to its inhibitory effect on TH1 expression. In contrast, addition of TGF-b1 induced IL-23R
development, TGF-b1 promotes development of TH17 cells. expression, irrespective of IL-23 addition. Thus, TGF-b1 and IFN-g
To determine whether the augmenting effects of TGF-b1 were due differentially induce mRNA for IL-23 and IL-12 receptors, respectively.
to the enhanced suppression of IFN-g or also to IFN-g-independent TGF-b1 thereby acts proximally in TH17 development to confer IL-23
mechanisms, we examined the effects of exogenous TGF-b1 on TH17 responsiveness.
1
Departments of 1Pathology, 2Microbiology, 3Genomics and 4Medicine, University of Alabama at Birmingham, Birmingham, Alabama 35294-2170, USA. 5Cellular Immunology
Section, Oral Infection and Immunity Branch, National Institute of Dental and Craniofacial Disease, National Institutes of Health, Bethesda, Maryland 20892-4352, USA.
231
© 2006 Nature Publishing Group

2. LETTERS NATURE|Vol 441|11 May 2006
The foregoing results indicated a central role for TGF-b in the for which an intact IL-23–IL-17 axis seems to be essential for host
initiation of TH17 differentiation and placed TGF-b signalling protection (Fig. 2 and D.B.O. et al., manuscript in preparation). In
proximal to IL-23 receptor expression and signalling in the TH17 immunocompetent mice, C. rodentium induces a transient, distal
developmental programme. To confirm and extend these findings colitis with resolution of lesions and clearance of the bacteria after
in vivo, we used a natural rodent pathogen, Citrobacter rodentium14, 14–21 days, after induction of a systemic, CD4þ T-cell-dependent
IgG response15–17. Although previously associated with TH1 adaptive
immunity18, oral challenge with this organism induced a potent TH17
response that was associated with host protection. At the peak of
inflammation, 8 days after inoculation (Fig. 2a), a large fraction of
colonic CD4þ T cells expressed IL-17, compared with a lower
frequency of cells that expressed IFN-g (Fig. 2b). Mice deficient in
IL-23 (p19 2/2) failed to clear the infection, and uniformly suc-
cumbed at a rate comparable to mice deficient for both IL-12 and
IL-23 (p40 2/2) (Fig. 2c). Although uninfected wild-type and IL-23-
deficient mice had similar baseline histological features, the latter
developed significantly less colonic inflammation after infection
(Fig. 2d), despite impaired bacterial clearance. The induction of an
IL-17 response was unimpaired in all examined tissues of infected
Figure 1 | TGF-b1 is required for TH17 commitment, independently of
IL-23. a, Naive CD4þ T cells were isolated from DO11.10 TCR transgenic
mice and activated with OVAp under the indicated conditions (see
Supplementary Information for details). b, Naive CD4þ T cells from Ifng 2/2 Figure 2 | In vivo development of an IL-17 effector response to a
mice were cultured with irradiated Ifng 2/2 splenocytes, anti-CD3 TH17-dependent pathogen is IL-23-independent, but protection is IL-23-
monoclonal antibody and anti-IL-4, with IL-23 and/or TGF-b1, as indicated. dependent. a, Histopathology of distal colon of B6 (wild-type) mice
IFN-g was included in cultures where indicated. c, Naive CD4þ T cells were inoculated orally with 1–2 £ 109 CFU Citrobacter rodentium and analysed
isolated from wild-type (B6) mice and cultured with p40 2/2 splenocytes and before colonization (d0), or at the peak of infection (d8). b, Cytokine
anti-CD3, anti-IL-4 and anti-IFN-g. Cultures were supplemented with phenotype of lamina propria lymphocytes isolated from B6 mice sham-
nothing or IL-23 and TGF-b1 added alone or in combination. In all cases infected (2) or infected (þ) with C. rodentium (8 days after inoculation) as
(a–c), T cells were recovered after 6 days and re-stimulated with PMA plus in a. Recovered cells were processed for FACS as in Fig. 1. c, Survival analysis
ionomycin for 5 h with monensin block before intracellular cytokine of wild-type (n ¼ 5), p40 2/2 (n ¼ 4) and p19 2/2 (n ¼ 8) mice after
staining for IL-17 and IFN-g, and analysis by flow cytometry. Plots are gated infection with C. rodentium as in a. d, Histopathology of day 8 colonic
on CD4þ cells and the quadrant percentiles of cells staining positively for the tissues from sham-infected (2) or infected (þ) mice of the indicated
indicated cytokines are shown. d, TGF-b and IFN-g reciprocally induce genotypes. The arrow denotes bacteria adherent to colonic epithelial cells.
expression of the IL-23 and IL-12 receptors. Naive CD4þ T cells isolated e, Intracellular cytokine analysis of lamina propria CD4þ T cells isolated
from Ifng 2/2 mice were activated with Ifng 2/2 splenocytes and anti-CD3 from wild-type or p19 2/2 mice 8 days after sham infection (2) or infection
plus anti-IL-4. Cultures were supplemented with TGF-b1 alone, TGF-b1 (þ) with C. rodentium as in a. Cytokine staining and analysis as in Fig. 1.
plus IL-23, or IFN-g plus IL-12, as indicated. After 6 days, T cells were f, Severe ulcerative, haemorrhagic colitis induced by anti-TGF-b treatment.
collected and processed for mRNA quantification by real-time RT–PCR. Histopathology of day 8 colonic tissue from C. rodentium-infected p19 2/2
Data shown are fold differences relative to T cells from a culture that was mice treated with anti-TGF-b as previously reported26. The arrows denote
differentiated under non-polarizing (neutral) conditions (that is, was not bacteria invading ulcerated colonic epithelium. All sections (except inset)
supplemented with cytokines). were photographed at the same magnification (£20).
232
© 2006 Nature Publishing Group

3. NATURE|Vol 441|11 May 2006 LETTERS
IL-23-deficient mice, and dominated the IFN-g response before and and data not shown), we surveyed T cells of the gut, as well as
after infection (Fig. 2e, and data not shown). Thus, despite an peripheral lymphoid tissues, for IL-17 expression (Fig. 3a). Com-
impaired inflammatory response, and deficiencies in bacterial clear- pared to age- and sex-matched wild-type controls, we found that
ance and host protection, IL-23-deficient mice were nevertheless TH17 cells were profoundly diminished or absent in all tissue sites
competent to develop a vigorous effector IL-17 response. Therefore, of TGF-b1-deficient mice; mice hemizygous for TGF-b1 deficiency
whereas IL-23 is dispensable for the differentiation of IL-17- (Tgfb þ/2) were intermediate. Interestingly, deficiency of IL-17þ cells
competent T cells, it is indispensable for a fully effective, protective was associated with significant decreases in basal circulating levels of
TH17 response. IL-17 (Fig. 3b), consistent with the predominance of T cells as a
To determine whether TGF-b deficiency impaired TH17 devel- source of IL-17 and a major role for the TGF-b1 isoform in
opment in vivo, we initially examined the effect of anti-TGF-b controlling homeostatic levels of IL-17 production. Furthermore,
neutralizing antibody on the course of C. rodentium infection. there was a striking, inverse correlation of IFN-g-producing cells
IL-23-deficient mice (p19 2/2) treated with anti-TGF-b developed with TGF-b1 deficiency, in accord with the spontaneous auto-
severe ulcerative and haemorrhagic intestinal lesions with gross inflammatory syndrome that these mice develop19,20. Notably,
bacterial invasion, neither of which was found in IL-23-deficient although fewer naive precursors were present in Tgfb1 2/2 mice,
mice treated with an isotype control, or infected wild-type mice there was no intrinsic defect in the ability of these cells to undergo
(Fig. 2f). Although this suggested a critical role for TGF-b in TH17 development, provided that the high endogenous IFN-g levels
protection against C. rodentium, a specific link with impaired TH17 were at least partially neutralized and exogenous TGF-b1 was
development could not be made due to the rapid morbidity and provided (Supplementary Fig. S3). Collectively, these data support
mortality associated with this treatment (data not shown). We a critical function for TGF-b1 in the development of TH17 cells
therefore examined development of TH17 cells in TGF-b1-deficient in vivo.
mice (Tgfb1 2/2) in a non-infectious setting. In view of our finding TGF-b1 has been associated with immunosuppression through its
that TH17 cells were normally enriched in intestinal tissues (Fig. 2, inhibitory effect on effector T-cell development (for example, TH1)
and its role as an immunosuppressant cytokine produced by some
regulatory T (Treg) cells. TGF-b1 also directs the development of Treg
cells that express the transcription factor Foxp3 (refs 21, 22). We
therefore examined whether TGF-b induced expression of Foxp3
together with IL-17. Naive T cells were activated in the presence
of exogenous TGF-b1 and evaluated for intracellular expression of
IL-17 and Foxp3 (Fig. 4). We found that distinct subpopulations of
IL-17- and Foxp3-expressing T cells developed—albeit with a marked
predominance of IL-17-producing T cells—under the conditions
examined. As in previous experiments, the effects of exogenous IL-23
were modest in comparison to the effects of TGF-b1 alone. In
accordance with a recent report23, we found that the frequency of
Figure 4 | TGF-b1 induces IL-17 and Foxp3 expression by distinct CD41
Figure 3 | Development of TH17 cells is impaired in TGF-b1-deficient mice. subpopulations. a, Naive CD4þ T cells purified from Ifng 2/2 mice by FACS
a, CD4þ T cells were purified from the indicated tissues of Tgfb1 2/2 mice sorting were co-cultured with irradiated Ifng 2/2 splenic feeders and
(2/2), and age-matched hemizygous (þ/2) and wild-type (þ/þ) activated with anti-CD3 under TH2-neutralizing conditions. IL-23 or
littermates. Isolated cells were stimulated immediately after isolation with TGF-b1 was added alone or in combination. b, Naive CD4þ T cells were
PMA and ionomycin for 5 h before intracellular cytokine staining for IL-17 prepared from wild-type B6 mice and cultured with wild-type splenocytes,
and IFN-g. Stained T cells were acquired and analysed by flow cytometry, as anti-CD3, TGF-b1, anti-IFN-g and anti-IL-4 as in a. After 6 days in culture,
in Fig. 1. MLN, mesenteric lymph nodes; LPL, lamina propria lymphocytes. T cells (from a and b) were re-stimulated with PMA and ionomycin before
b, Serum from age-matched Tgfb1 2/2 mice (2/2), and hemizygous (þ/2) intracellular cytokine staining for IL-17 and Foxp3. Stained T cells were
and wild-type (þ/þ) littermates, was collected and analysed for IL-17A by acquired and analysed by flow cytometry. Plots are gated on CD4þ cells and
enzyme-linked immunosorbent assay. Data are the mean ^ s.e.m. of the quadrant percentiles are given for cells staining positively for IL-17 or
triplicate determinations from 4–8 mice. Foxp3.
233
© 2006 Nature Publishing Group

4. LETTERS NATURE|Vol 441|11 May 2006
Foxp3þ cells generated in the presence of TGF-b was inversely related 4. Happel, K. I. et al. Divergent roles of IL-23 and IL-12 in host defense against
Klebsiella pneumoniae. J. Exp. Med. 202, 761– -769 (2005).
to levels of IL-6, such that Foxp3þ cells were nearly extinguished
5. Cua, D. J. et al. Interleukin-23 rather than interleukin-12 is the critical cytokine
in the presence of exogenous IL-6 (Supplementary Fig. S4). Thus, for autoimmune inflammation of the brain. Nature 421, 744– -748 (2003).
TGF-b seems to have a dual role in T-cell differentiation by directing 6. Murphy, C. A. et al. Divergent pro- and antiinflammatory roles for IL-23 and
distinct subpopulations of Foxp3þ Treg cells and TH17 cells, con- IL-12 in joint autoimmune inflammation. J. Exp. Med. 198, 1951– -1957 (2003).
tingent upon the inflammatory cytokine environment, perhaps 7. Langrish, C. L. et al. IL-23 drives a pathogenic T cell population that induces
autoimmune inflammation. J. Exp. Med. 201, 233– -240 (2005).
providing a mechanism by which Treg cells are poised to terminate 8. Oppmann, B. et al. Novel p19 protein engages IL-12p40 to form a cytokine,
TH17 responses after antigen clearance. IL-23, with biological activities similar as well as distinct from IL-12. Immunity
Our findings, and complementary findings published during the 13, 715– -725 (2000).
revision of our paper23, define a role for TGF-b in TH17 lineage 9. Parham, C. et al. A receptor for the heterodimeric cytokine IL-23 is composed
of IL-12Rb1 and a novel cytokine receptor subunit, IL-23R. J. Immunol. 168,
commitment, thereby linking this pleiotropic cytokine to adaptive 5699– -5708 (2002).
immunity in a way that has important implications for mechanisms 10. Laouar, Y., Sutterwala, F. S., Gorelik, L. & Flavell, R. A. Transforming growth
of host defence, immune homeostasis and autoimmunity. Although factor-b controls T helper type 1 cell development through regulation of natural
TGF-b has heretofore been associated primarily with immuno- killer cell interferon-g. Nature Immunol. 6, 600– -607 (2005).
suppressive functions in T-cell immunity—either through the pro- 11. Lin, J. T., Martin, S. L., Xia, L. & Gorham, J. D. TGF-b1 uses distinct mechanisms
þ
to inhibit IFN-g expression in CD4 T cells at priming and at recall: differential
motion of Treg development and function or inhibition of TH1 and involvement of Stat4 and T-bet. J. Immunol. 174, 5950– -5958 (2005).
TH2 development12 —it is now apparent that TGF-b may also 12. Li, M. O., Wan, Y. Y., Sanjabi, S., Robertson, A. K. & Flavell, R. A. Transforming
facilitate pro-inflammatory responses by promoting TH17 develop- growth factor-b regulation of immune responses. Annu. Rev. Immunol. 24,
ment. These data support a model in which early signalling by TGF-b 99– -146 (2006).
13. Berenson, L. S., Ota, N. & Murphy, K. M. Issues in T-helper 1 development—
in an inflammatory context initiates TH17 commitment and upregu- resolved and unresolved. Immunol. Rev. 202, 157– -174 (2004).
lates IL-23R, providing a basis for TGF-b and IL-23 effects in TH17 14. MacDonald, T. T., Frankel, G., Dougan, G., Goncalves, N. S. & Simmons, C. Host
development that parallel those of, and are antagonized by, sequen- defences to Citrobacter rodentium. Int. J. Med. Microbiol. 293, 87– (2003).
-93
tial IFN-g and IL-12 signalling in TH1 development. The reciprocal 15. Simmons, C. P. et al. Central role for B lymphocytes and CD4þ T cells in
and antagonistic actions of TGF-b, IFN-g and IL-4 on TH17, TH1 immunity to infection by the attaching and effacing pathogen Citrobacter
rodentium. Infect. Immun. 71, 5077– -5086 (2003).
and TH2 development—both through direct actions on the devel- 16. Maaser, C. et al. Clearance of Citrobacter rodentium requires B cells but not
oping T cell and indirectly by modulating cytokine production by secretory immunoglobulin A (IgA) or IgM antibodies. Infect. Immun. 72,
innate immune cells—provide an extended mechanism for the 3315– -3324 (2004).
efficient matching of effector T-cell polarization, and thus adaptive 17. Bry, L. & Brenner, M. B. Critical role of T cell-dependent serum antibody, but
not the gut-associated lymphoid tissue, for surviving acute mucosal infection
immunity, to offending pathogens. with Citrobacter rodentium, an attaching and effacing pathogen. J. Immunol. 172,
433– -441 (2004).
METHODS 18. Higgins, L. M., Frankel, G., Douce, G., Dougan, G. & MacDonald, T. T.
CD41 T-cell isolation and culture. CD4þ T cells from the indicated strains of Citrobacter rodentium infection in mice elicits a mucosal Th1 cytokine response
mice (Supplementary Information) were purified from pooled spleen and lymph and lesions similar to those in murine inflammatory bowel disease. Infect.
nodes by magnetic sorting using mouse anti-CD4 beads. Cells were cultured Immun. 67, 3031– -3039 (1999).
with irradiated splenic feeder cells or bone-marrow-derived dendritic cells in 19. Shull, M. M. et al. Targeted disruption of the mouse transforming growth
complete medium as described previously1,24. DO11.10 T-cell receptor (TCR) factor-b1 gene results in multifocal inflammatory disease. Nature 359,
transgenic CD4þ cells were activated with 5 mg ml21 OVA peptide 323–339 693– -699 (1992).
20. Kulkarni, A. B. et al. Transforming growth factor b1 null mutation in mice
(OVAp), whereas non-transgenic cells were stimulated with 2.5 mg ml21 anti-
causes excessive inflammatory response and early death. Proc. Natl Acad. Sci.
CD3. Where indicated, cultures were supplemented with recombinant cytokines USA 90, 770– -774 (1993).
or antibodies (Supplementary Information). Cells were harvested on day 6 for 21. Chen, W. et al. Conversion of peripheral CD4þCD252 naive T cells to
analysis. Colonic lamina propria lymphocytes were obtained by a protocol þ þ
CD4 CD25 regulatory T cells by TGF-b induction of transcription factor
modified from that previously described25. Foxp3. J. Exp. Med. 198, 1875– -1886 (2003).
Citrobacter rodentium inoculation. C. rodentium was prepared by incubation 22. Wan, Y. Y. & Flavell, R. A. Identifying Foxp3-expressing suppressor T cells with
with shaking at 37 8C for 6 h in LB broth. After 6 h, the bacterial density was a bicistronic reporter. Proc. Natl Acad. Sci. USA 102, 5126– -5131 (2005).
assessed by absorbance at an optical density of 600 nm and confirmed by plating 23. Veldhoen, M., Hocking, R. J., Atkins, C. J., Locksley, R. M. & Stockinger, B.
of serial dilutions. Inoculation of mice was by oral administration with 1–2 £ 109 TGF-b in the context of an inflammatory cytokine milieu supports de novo
colony forming units (CFU). Tissues were collected for histology and/or differentiation of IL-17-producing T cells. Immunity 24, 179– -189 (2006).
24. Kubo, T. et al. Regulatory T cell suppression and anergy are differentially
cytokine phenotyping at times indicated after inoculation.
regulated by proinflammatory cytokines produced by TLR-activated dendritic
Flow cytometry. Cells were stimulated for 5–6 h with 50 ng ml21 phorbol cells. J. Immunol. 173, 7249– -7258 (2004).
myristate acetate (PMA) and 750 ng ml21 ionomycin, or not at all, and processed 25. Iqbal, N. et al. T helper 1 and T helper 2 cells are pathogenic in an antigen-
for flow cytometry as previously described1 using the indicated antibody specific model of colitis. J. Exp. Med. 195, 71– -84 (2002).
conjugates (see Supplementary Information). Samples were acquired on a 26. Powrie, F., Carlino, J., Leach, M. W., Mauze, S. & Coffman, R. L. A critical role
FACSCalibur flow cytometer and data analysis used CellQuest Pro software for transforming growth factor-b but not interleukin 4 in the suppression of
(BD Biosciences). T helper type 1-mediated colitis by CD45RBlow CD4þ T cells. J. Exp. Med. 183,
RNA isolation, cDNA synthesis and real-time RT–PCR. Recovered T cells were 2669– -2674 (1996).
re-stimulated with PMA and ionomycin for 5 h. Cells were lysed and RNA isolated
Supplementary Information is linked to the online version of the paper at
and processed for real-time, reverse-transcribed PCR (RT–PCR) as previously
www.nature.com/nature.
described1, using primer and probe sequences provided in Supplementary
Information. Acknowledgements The authors thank P. Bucy, D. Chaplin, S. Schoenberger,
A. Zajac and members of the Weaver laboratory for their comments and
Received 6 March; accepted 28 March 2006. suggestions. We thank J. Oliver, M. Blake and C. Song for animal husbandry and
Published online 30 April 2006. phenotyping, and N. LeLievre for editorial assistance. We also thank R. Kastelein
for provision of p19 2/2 mice. This work was supported by grants from the NIH
1. Harrington, L. E. et al. Interleukin 17-producing CD4þ effector T cells develop (to C.T.W., R.D.H. and P.R.M.), Sankyo Co. Ltd (C.T.W.) and a postdoctoral
via a lineage distinct from the T helper type 1 and 2 lineages. Nature Immunol. fellowship from the National Multiple Sclerosis Society (L.E.H.).
6, 1123– -1132 (2005).
2. Park, H. et al. A distinct lineage of CD4 T cells regulates tissue inflammation by Author Information Reprints and permissions information is available at
producing interleukin 17. Nature Immunol. 6, 1133– -1141 (2005). npg.nature.com/reprintsandpermissions. The authors declare no competing
3. Murphy, K. M. & Reiner, S. L. The lineage decisions of helper T cells. Nature financial interests. Correspondence and requests for materials should be
Rev. Immunol. 2, 933– -944 (2002). addressed to C.T.W. (cweaver@uab.edu).
234
© 2006 Nature Publishing Group