Skin and Peripheral Lymph Node Invariant NKT Cells Are
Mainly Retinoic Acid Receptor-Related Orphan Receptor ?t?
and Respond Preferentially under Inflammatory Conditions1
Jean-Marc Doisne,* Chantal Becourt,* Latiffa Amniai,* Nadia Duarte,*
Jean-Benoît Le Luduec,†Ge ´rard Eberl,‡and Kamel Benlagha2*
Lymph nodes (LNs) have been long considered as comprising few invariant NKT (iNKT) cells, and these cells have not been
studied extensively. In this study, we unravel the existence of stable rather than transitional LN-resident NK1.1?iNKT cell
populations. We found the one resident in peripheral LNs (PLNs) to comprise a major IL-17-producing population and to express
the retinoic acid receptor-related orphan receptor ?t (ROR?t). These cells respond to their ligand ?-galactosylceramide (?-
GalCer) in vivo by expanding dramatically in the presence of LPS, providing insight into how this rare population could have an
impact in immune responses to infection. PLN-resident ROR?t?NK1.1?iNKT cells express concomitantly CCR6, the integrin
?-chain ?E(CD103), and IL-1R type I (CD121a), indicating that they might play a role in inflamed epithelia. Accordingly, skin
epithelia comprise a major ROR?t?CCR6?CD103?CD121a?NK1.1?cell population, reflecting iNKT cell composition in PLNs.
Importantly, both skin and draining PLN ROR?t?iNKT cells respond preferentially to inflammatory signals and independently
of IL-6, indicating that they could play a nonredundant role during inflammation. Overall, our study indicates that ROR?t?iNKT
cells could play a major role in the skin during immune responses to infection and autoimmunity. The Journal of Immunology,
2009, 183: 2142–2149.
TCR composed of an invariant V?14J?18 chain combined with
one of three V? segments (V?8, 7, or 2) (1). They are capable of
recognizing CD1d-presenting self-glycolipids, such as isoglobotri-
hexosylceramide or exogenous glycolipids found in certain bacte-
ria, as well as the synthetic ?-galactosylceramide (?-GalCer), orig-
inally isolated from marine sponges.
Mouse iNKT cells can be identified using CD1d tetramers
loaded with ?-GalCer regardless of their expression of differenti-
ation markers such as NK1.1, the molecule most commonly asso-
ciated with the NK lineage (2, 3). The usage of CD1d tetramers
allows for the identification of iNKT cells that are devoid of most
NK-associated markers, including NK1.1 (4, 5). Because NK1.1?
ouse V?14 NKT (invariant NKT (iNKT))3cells are a
small subset of CD4?or CD4?CD8?double-negative
(DN) ?? T cell population expressing a semi-invariant
appear before NK1.1?iNKT cells during ontogeny, in the thymus,
spleen, and liver, and because NK1.1?iNKT cells accumulate
progressively and dominate the iNKT cell population in these or-
gans in older mice, it has been accepted that the NK1.1?iNKT
cells are immature and that their ultimate fate is to become
NK1.1?iNKT cells. Therefore, studies aimed at understanding
iNKT cell functions have focused mostly on the NK1.1?iNKT
In response to TCR ligation, iNKT cells promptly produce large
amounts of both IFN-? and IL-4. They have been implicated in the
regulation of immune responses associated with a broad range of
diseases and have been shown to promote inflammatory Th1 and
immunomodulatory Th2 responses (6). Several hypotheses have
been proposed to reconcile such a diversity of functions. iNKT cell
subsets differing in respect to their ability to secrete Th1- vs Th2-
type cytokines was suggested, among other possibilities.
Numerous studies aimed at identifying distinct functional sub-
sets among iNKT cells have been performed. In humans, these
studies showed that DN iNKT cells produce predominantly Th1
cytokines, whereas CD4?iNKT cells are the exclusive producers
of Th2-type cytokines upon primary stimulation (7). In mice, there
is no evidence for a disparity in cytokine secretion between CD4?
and DN iNKT cells, although one study did show distinct func-
tional properties between liver CD4?and DN iNKT, the latter
being more efficient at rejecting tumors (8). Other studies pointed
out phenotypic distinctions among iNKT cells based on their ex-
pression of chemokine receptors and adhesion molecules, depend-
ing on the organ where they reside: CXCR5 in the spleen (9), and
CXCR6 and LFA-1 in the liver (10, 11).
Several lines of research have converged to the recent discovery
of iNKT-expressing retinoic acid receptor-related orphan receptor
?t (ROR?t) and secreting IL-17. However, neither of these studies
shows a phenotype that can distinguish between IL-17?and IL-
17?NK1.1?iNKT cells (12–15). Furthermore, whether these
*INSERM Unite ´ 561/Groupe AVENIR, Ho ˆpital Cochin St. Vincent de Paul, Univer-
site ´ Descartes, Paris, France;†INSERM Unite ´ 851, Institut Fe ´de ´ratif de Recherche
128, Lyon, France; and‡Institut Pasteur, Laboratory of Lymphoid Tissue Develop-
ment, Centre National de la Recherche Scientifique Unite ´ de Recherche Associe ´e
1961, Paris, France
Received for publication April 2, 2009. Accepted for publication May 30, 2009.
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.
1This work was supported by INSERM “Groupe AVENIR” Grant R04193KS, Fon-
dation pour la Recherche Me ´dicale Grant INE20051105133, and Agence Nationale
pour la Recherche Grant R07119KS, awarded to K.B.
2Address correspondence and reprint requests to Dr. Kamel Benlagha, INSERM
Unite ´ 561, 82 Avenue Denfert Rochereau, 75014 Paris, France. E-mail address:
3Abbreviations used in this paper: iNKT, invariant NKT; 7-AAD, 7-aminoactino-
mycin D; ?-GalCer, ?-galactosylceramide; BM-DC, bone marrow-derived dendritic
cell; DC, dendritic cell; DN, double negative; LN, lymph node; PLN, peripheral LN;
ROR?t, retinoic acid receptor-related orphan receptor ?t; TPA, 12-O-tetradecanoyl-
phorbol-13-acetate; WT, wild type.
Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00
The Journal of Immunology
cells, representing minor iNKT cell populations in the liver and
spleen, secrete exclusively IL-17 was not reported. Finally, McNab
et al. (16) suggested the existence of mature, stable NK1.1?iNKT
cells in the spleen. However, a detailed functional and phenotyp-
ical characterization of this subset is still lacking.
There are few reports concerning lymph node (LN)-resident
iNKT cells because of their rarity. Early reports using CD1d tet-
ramers indicated that they have a high frequency of NK1.1?iNKT
cells (3). Other studies indicated phenotypic and functional differ-
ences between iNKT cells colonizing mediastinal and mesenteric
(splanchnic) and peripheral LNs (PLNs) (17); however, these stud-
ies focused on NK1.1?iNKT in V?14 transgenic animals.
In this study, we uncovered LN NK1.1?iNKT cell populations
representing mature iNKT cells at a final stage of their develop-
ment. Contrary to previously described organ-resident iNKT cells,
which are mostly NK1.1?, NK1.1?iNKT cells represent the ma-
jority of CD4?and DN iNKT cells in LNs. The latter, mainly
located in PLNs, produce IL-17, but not IFN-? or IL-4, and ex-
press the nuclear receptor and the transcription factor ROR?t. We
also found that most of skin-resident iNKT, like PLN iNKT cells,
express CCR6, CD103, and CD121a in addition to producing IL-
17, and that both populations respond in vivo preferentially to
inflammatory signals and could be involved in immune responses
to infection and autoimmunity.
Materials and Methods
C57BL/6 mice were purchased from Janvier. IL-6?/?mice were purchased
from The Jackson Laboratory. Rorc(?t)-GfpTGmice were generated, as
previously described (18), and backcrossed nine times to C57BL/6. All
mice were maintained under specific pathogen-free conditions at our ani-
mal facility, and experimental studies were in accordance with the Insti-
tutional Animal Care and Use Guidelines.
In vitro stimulation
Cell suspensions were prepared from thymus, liver, spleen, and LNs and
enriched in iNKT cells by depletion of CD8?and CD19?cells using
mouse depletion dynabeads (Invitrogen) and following the manufacturer’s
instructions. Ear skin was directly incubated in RPMI 1640 medium con-
taining collagenase type VIII (Sigma-Aldrich) and DNase I (Sigma-
Aldrich), and cell suspensions were obtained, as described previously (18).
Cells were incubated with 0.1 ng/ml ?-GalCer-preloaded bone marrow-
derived dendritic cells (BM-DCs), obtained after culture with GM-CSF
(PeproTech) at 2 ng/ml for 6 days, or PMA (5 ng/ml)/ionomycin (500
ng/ml), for 4 h in the presence of 5 ?g/ml brefeldin A (all purchased from
Sigma-Aldrich). IL-17 content in culture supernatants of sorted LN sub-
sets, stimulated with BM-DCs in the presence of 100 ng/ml ?-GalCer, was
measured by ELISA using the DuoSet ELISA kit from R&D Systems and
following the manufacturer’s instructions.
Abs and flow cytometry
PE-Texas Red- and PerCP-Cy5.5-conjugated mAbs against B220 clone
RA3-6B2; PerCP-Cy5.5- and Alexa Fluor 700-conjugated mAbs against
CD4 clone RM4-5; PerCP-Cy5.5-conjugated mAb against NK-1.1 clone
PK136; FITC- and PE- conjugated mAb against Ki67 clone B56; PE-con-
jugated mAb against CD121a clone 35F5; Alexa Fluor 647-conjugated
mAb against CCR6 clone 140706; nonconjugated mAbs against CD8?
clone 53-6.7, and CD19 clone 1D3; and 7-aminoactinomycin D (7-AAD)
were obtained from BD Biosciences. Alexa Fluor 488-conjugated mAb
against IL-17 clone eBio17B7; PE-conjugated mAbs against TCR? clone
H57-597, IL-4 clone 11B11, IFN-? clone XMG1.2, and ROR?t clone
AFKJS-9; PE-Cy5-conjugated mAb against heat-stable Ag clone M1/39;
PE-Cy7-conjugated mAb against NK-1.1 clone PK136; FITC- and PE-
conjugated mAbs against CD103 clone 2E7; and Pacific Blue-conjugated
mAb against CD4 clone RM4-5 and CD45 clone 30-F11 were purchased
from eBiosciences. PE-Cy7-conjugated mAb against heat-stable Ag clone
M1/69 and Alexa Fluor 700-conjugated mAb against TCR? clone H57-597
were purchased from Biolegend. FITC- and PE-conjugated mAbs against
CCR6 clone 140706 were from R&D Systems. Pacific Orange-conjugated
mAb against CD4 clone RM4-5 was obtained from Invitrogen. CD1d-?-
GalCer or -PBS57 tetramers were produced with streptavidin-allophyco-
cyanin (BD Biosciences) or -PE-Cy7 (eBiosciences) and used for staining,
as described previously (2). BD Fixation/Permeabilization Kit (cytokine
detection) and Foxp3 Staining Buffer Set from eBiosciences (transcription
factors and Ki67 detection) were used for intracellular staining, according
to manufacturer’s instructions. Flow cytometry was performed with
FACSAria (BD Biosciences), and data were analyzed with FACSDiva soft-
ware v6.1.2 (BD Biosciences) and FlowJo 7.2.5 (Tree Star).
In vivo stimulation
Four-month-old C57BL/6 mice were injected with 100 ?l of PBS contain-
ing LPS (25 ?g), or LPS and ?-GalCer (0.05 ?g), or one million BM-DCs,
activated or not with LPS (Sigma-Aldrich) at 5 ?g/ml, and preloaded or not
with 10 ng/ml ?-GalCer. Contralateral footpads were injected with 100 ?l
Induction of skin inflammation
Induction of inflammation with 12-O-tetradecanoylphorbol-13-acetate
(TPA; Sigma-Aldrich) started at day 0 with application of 20 ?l of 0.01%
TPA suspended in acetone (10 ?l in each side of the ear) and repeated
every 2 days, as described previously (19, 20). A control group treated with
acetone was included in each experiment.
Identification of LN-resident NK1.1?iNKT cells
Previous studies showed that most iNKT cells exported from the
thymus are NK1.1?and that they acquire NK1.1 in the periphery
(4, 5). Thus, it has been believed that peripheral NK1.1?iNKT
cells in normal mice represent recent thymic emigrants at an im-
mature transitional stage of their development and that NK1.1 ex-
pression is a marker of maturation. We monitored the kinetics of
the natural NK1.1?to NK1.1?transition during iNKT cell mat-
uration, by tracking iNKT cells with CD1d tetramers loaded with
?-GalCer, and found that this maturation takes place in the thy-
mus, liver, and spleen, leading to an accumulation of NK1.1?
iNKT cells as mice age (Fig. 1). However, whereas NK1.1?iNKT
cell proportions plateau at ?2 mo of age in the thymus and liver,
accumulation of these cells occurs with slower kinetics in the
spleen, where 4 mo are needed. Moreover, the frequency of steady-
state splenic NK1.1?iNKT cell populations is higher than those
found in the thymus or liver. Analysis of pooled cells from max-
illary, axillary, inguinal, and mesenteric LNs (pooled LNs) re-
vealed that few NK1.1?iNKT cells progress to the NK1.1?stage,
regardless of mouse age, with equilibrium established as early as
1 mo of age (Fig. 1 and Table S1).4The absence of NK1.1?iNKT
cells in LNs might reflect a suboptimal environment that does not
permit immature NK1.1?iNKT cells to reach the NK1.1?stage
and/or favor NK1.1?iNKT cell maturation and maintenance. Ac-
cordingly, it was reported that the stimulating capabilities of APCs
in mesenteric LNs are distinct from those of splenic APCs (17).
The frequency of NK1.1?iNKT cells in LNs might also reflect the
number of immature vs mature NK1.1?iNKT cells that reach
distinct LNs. Regardless, these results indicate that the NK1.1?to
NK1.1?transition does not occur in the LN, revealing a stable,
nontransitional NK1.1?iNKT cell subset, possibly representing a
mature iNKT cell population. Accordingly, IL-2R? (CD122),
moderately and highly expressed on liver and spleen NK1.1?and
NK1.1?iNKT cells, respectively, is not expressed on LN NK1.1?
iNKT cells (Fig. 2). This supports the probability that they repre-
sent a distinct iNKT cell population that is or becomes unrespon-
sive to IL-15, which is described to be crucial for NK1.1?to
Phenotypic characterization of LN-resident NK1.1?iNKT cells
Based on our previous finding showing that equilibrium between
NK1.1?and NK1.1?iNKT cells is established in all organs tested
4The online version of this article contains supplemental material.
2143The Journal of Immunology
between 2 and 4 mo of age, we decided to characterize phenotyp-
ically NK1.1?iNKT cells using 4-mo-old mice to allow unam-
biguous distinction between transitional and resident NK1.1?
iNKT cells. Separate phenotypic analysis of peripheral (axillary,
maxillary, popliteal, and inguinal) and mesenteric LNs showed, as
previously observed with pooled LNs, that the majority of LN
iNKT cells draining different areas are NK1.1?(Fig. 3 and Table
S1). We also found a higher frequency of DN than CD4 NK1.1?
iNKT cells in PLNs, whereas the opposite picture was observed in
mesenteric LNs. In thymic, splenic, and liver NK1.1?iNKT cells,
the frequency of CD4?NK1.1?iNKT cells was found to be
higher in all of these organs (data not shown). In LNs, it is possible
that some subsets expand, home, or are retained preferentially in
some LNs, but not others. Other factors or receptors, yet to be
defined, might explain these differences in subset composition.
More importantly, we found that ?70–80% of NK1.1?PLN
iNKT cells coexpress CCR6, CD103, and CD121a and ?10% of
CCR6?CD103?CD121a?cells express CD4 (Fig. 3A and Table
S1). This population is minor in mesenteric LNs (Fig. 3B and
Table S1). These results indicate that NK1.1?iNKT cells are com-
posed of two populations, as follows: the first expresses CD4 and
is essentially located in mesenteric LNs; the second is DN and
mainly located in PLNs, expressing simultaneously CCR6,
CD103, and CD121a.
Functional characterization of LN-resident NK1.1?iNKT cells
NK1.1?iNKT cells were shown to produce IL-17 in the thymus,
liver, spleen, and lung (12–15). To determine whether those in the
LNs produce IL-17, we stimulated LN iNKT cells in vitro, with
?-GalCer-loaded BM-DCs. The concentration was optimized to
avoid NK1.1 and TCR down-modulation (Fig. S2), which is re-
ported to occur after strong TCR engagement. We found that IL-17
was produced by NK1.1?iNKT cells in all LNs tested, with the
highest frequency of IL-17?cells in PLNs (Fig. 4A). We also
found that liver, spleen, and thymic NK1.1?iNKT cells produce
IL-17, thus confirming previous studies (Fig. 4A). However, the
frequencies of these cells do not exceed 1% of total iNKT cells.
Moreover, we found in PLNs that all IL-17?iNKT cells express
concomitantly CCR6, CD103, and CD121a, and ?10% of these
cells express CD4 (Fig. 4B and Table S1). In mesenteric LNs,
containing low frequencies of CCR6?CD103?CD121a?cells, IL-
17?iNKT cells are not clearly defined by their coexpression of the
three surface markers, indicating a phenotypic heterogeneity
within IL-17?iNKT cells (Fig. 4B and Table S1). This heteroge-
neity is also seen in liver, spleen, and thymus IL-17?iNKT cells
(data not shown). We confirmed the secretion of IL-17 by quan-
tifying this cytokine in the culture supernatant of sorted PLN
CCR6?CD103?CD121a?NK1.1?iNKT cells stimulated with
?-GalCer-loaded BM-DCs (Fig. 4C).
In mice, Th17 cells were shown to produce IL-17, but not IFN-?
and IL-4, and express the transcription factor ROR?t, important
for IL-17 production, but not T-bet and GATA-3, respectively,
important for IFN-? and IL-4 production (22). To check to what
extent LN IL-17?iNKT cells are related to Th17 cells, we ana-
lyzed PLN iNKT cells in Rorc(?t)-GfpTGmice, in which GFP?
cells are ROR?t?. We found that GFP?, but not GFP?iNKT cells
produce IL-17 after stimulation (Fig. 4D). In addition, we found
that GFP?iNKT cells in PLNs are CCR6?CD103?CD121a?
NK1.1?and are mainly DN, a phenotype identical with PLN IL-
17?iNKT cells from wild-type (WT) mice, confirming that IL-
17?and ROR?t?iNKT cells represent the same population (Fig.
S3A). We also found that LN GFP?NK1.1?iNKT cells from
Rorc(?t)-GfpTGmice (Fig. S3B) and LN IL-17?NK1.1?iNKT
cells from WT mice (Fig. 4D) do not produce IFN-?, whereas a
minute fraction of cells has the potential to secrete IL-4. We con-
firmed the presence of IL-4, but not IFN-?, in the culture super-
natant of sorted PLN CCR6?CD103?CD121a?NK1.1?iNKT
cells after stimulation with ?-GalCer-loaded BM-DCs (data not
shown). We found that LN IL-17?iNKT cells do not express the
transcription factor T-bet (Fig. S4), and that these cells are not
related to T regulatory cells, because they do not express Foxp3
(Fig. S4), GITR, or CTLA-4, all associated with this subset (data
C57BL/6 mice (age 1–8 mo) were stained with CD1d tetramers and anti-NK1.1 mAb. The percentages of NK1.1?(f) or NK1.1?(E) tetramer?(Tet?)
iNKT cells are shown. LN cell analyses were performed on pooled maxillary, axillary, inguinal, and mesenteric LNs. B, Shown are representative dot plots
of NK1.1 expression on iNKT cells in 4-mo-old mice. Data are representative of at least three individual experiments, with three mice pooled in each
Kinetics of iNKT cell maturation and identification of mature LN NK.1.1?iNKT cells. A, Thymus, liver, spleen, and LN iNKT cells from
Liver, spleen, and LN cells were stained with CD1d
tetramers and mAbs directed against NK1.1 and
CD122, and shown is IL-2R? expression, as mentioned.
Data are representative of three experiments.
IL-2R? expression among iNKT cells.
2144SKIN AND LN IL-17?NK1.1?iNKT CELLS
Altogether, these results, supported by Rorc(?t)-GfpTGmice, show
that the concomitant expression of CCR6, CD103, and CD121a in
PLNs is strongly associated with IL-17 production by iNKT cells.
IL-17?CCR6?CD103?CD121a?NK1.1?iNKT cells respond
and expand preferentially in vivo to ?-GalCer in the presence
Studies performed by Parekh et al. (23) have shown that iNKT
cells are capable of substantial in vivo expansion in response to
?-GalCer treatment. iNKT cell expansion was observed in many
organs and was maximal ?3 days after injection. This expansion
was accompanied by secretion of IL-4 and IFN-? as early as 2 h
after ?-GalCer treatment. Two hours after i.p. injection of ?-Gal-
Cer, we did not detect IL-17 production by PLN iNKT cells,
whereas more than half of splenic iNKT cells produce IFN-? and
IL-4 and less than 1% produce IL-17 (Fig. S5). Also, popliteal LN
iNKT cells did not produce IL-17 or expand 2 h after ?-GalCer
injection in the footpad (data not shown). We thus decided to an-
alyze iNKT cell response at different time points after ?-GalCer or
?-GalCer-loaded BM-DC injection in the footpad. We observed an
optimal amplification of popliteal LN iNKT cells 3 days postin-
jection and found that expanded cells do not produce IL-17 ex vivo
at this time point or any other time tested between 1 and 5 days
(Fig. 5A and data not shown). In vitro stimulation of expanded
popliteal LN iNKT cells showed that these cells keep their poten-
tial to secrete IL-17 and indicated that popliteal LN IL-17?iNKT
cells did not expand preferentially because their frequency re-
mained unchanged 3 days postinjection (Fig. 5A). To mimic what
major iNKT cell population in PLNs. Inguinal (A; representative of PLNs) and
mesenteric (B) LNs from 4-mo-old C57BL/6 mice were stained with CD1d
tetramers and mAbs directed against NK1.1, CCR6, CD103, CD121a, and
CD4. Note that Tet?iNKT are mainly NK1.1?(upper left, A and B), with DN
and CD4?being the main populations in peripheral and mesenteric LNs, re-
spectively (upper right, A and B), and that CCR6?CD103?cells are the major
population among peripheral, but not mesenteric Tet?NK1.1?iNKT cells
(lower left dot, A and B), with CCR6?CD103?, but not CCR6?CD103?cells
expressing CD121a and being mainly CD4?(lower right, A and B). Data are
representative of at least 10 individual experiments.
CCR6?CD103?CD121a?NK.1.1?iNKT cells represent a
thymus, liver, and spleen were enriched in iNKT cells and stimulated with ?-GalCer-loaded BM-DCs. Tet?iNKT cells were then analyzed for NK1.1 and IL-17
expression. Note that Tet?IL-17?iNKT cells are NK1.1?, with the highest frequency of IL-17?cells being in PLNs. Data are representative of at least 10
individual experiments, with 3 mice pooled in each experiment. Numbers represent percentages. B, LN cells were stimulated as in A, and shown is CCR6, CD103,
CD121a, and CD4 expression on Tet?IL-17?NK1.1?iNKT cells. Data are representative of at least four individual experiments, with pooled PLN cells from
20 mice in each experiment. C, Sorted Tet?CCR6?CD103?CD121a?NK1.1?(TP), CCR6?CD103?CD121a?NK1.1?(TN), and NK1.1?LN iNKT cells, as
shown in histograms for postsort check (purity superior to 95%), were stimulated for 48 h with BM-DCs in the presence of 100 ng/ml ?-GalCer, and IL-17 content
in the culture supernatant was measured by ELISA. Data are representative of at least three individual experiments, with pooled PLN cells from at least 20 mice
in each experiment. D, PLN iNKT cells from C57BL/6 (upper and middle dot plots) or Rorc(?t)-GfpTG(lower dot plots) mice were stimulated with PMA/
ionomycin, and Tet?NK1.1?iNKT cells were checked for IL-17, IL-4, and IFN-? production. Data are representative of at least two individual experiments, with
3 pooled mice in each experiment.
CCR6, CD103, and CD121a define IL-17?iNKT cells in PLNs and express ROR?t. A, Inguinal (representative of PLNs) and mesenteric LNs,
2145The Journal of Immunology
would occur in physiopathological conditions after migration of
activated dendritic cells (DCs) or Langerhans cells to LNs under
inflammatory conditions, we injected LPS-activated BM-DCs
loaded with ?-GalCer. Three days later, the overall popliteal LN
iNKT cells expanded, but importantly, we found an increased fre-
quency of popliteal LN IL-17?iNKT cells, indicating that they
expanded preferentially under these stimulation conditions (Fig.
5A). The increase in popliteal LN IL-17?iNKT cell number is due
to cell proliferation of LN-resident cells rather than to enhanced
iNKT cell recruitment because they highly express Ki67, a nuclear
cell proliferation-associated Ag expressed in all active stages of
the cell cycle (Fig. 5B). We also observed a preferential expansion
of popliteal LN IL-17?iNKT cells after coinjection of free ?-Gal-
Cer and LPS, indicating that LPS-responsive CD1d-expressing
cells present at the site of injection or in the draining LN could
recapitulate the preferential amplification of popliteal LN IL-17?
iNKT cells observed in response to ?-GalCer presented by LPS-
activated BM-DCs (Fig. 5A). Overall, our results indicate that rest-
ing NK1.1?iNKT cells respond to antigenic challenge in vivo by
dramatically expanding while keeping their potential to produce
IL-17. Importantly, IL-17?iNKT cells respond preferentially un-
der inflammatory conditions, providing insight into how this rare
population could have an impact in immune responses to infection.
Skin-resident iNKT cells are mainly ROR?t?
CCR6?CD103?CD121a?NK1.1?iNKT and respond
preferentially under inflammatory conditions
We analyzed iNKT cell composition in the skin and gut, organs
connected to PLN and mesenteric LNs, respectively. We found
that skin, but not gut (intraepithelial lymphocytes, lamina propria
lymphocytes, and Peyer’s patches), contains a dominant PLN-like
NK1.1?CD4?iNKT cell population (Fig. 6A). Moreover, we
found a high frequency of iNKT cells expressing CCR6, CD103,
and CD121a, and most iNKT cells express CD103 (Fig. 6A). Skin-
resident CCR6?CD103?CD121a?NK1.1?iNKT cells produce
IL-17 after stimulation with PMA/ionomycin (Fig. 6B). The fre-
quency of IL-17?-producing cells is, however, lower than that
observed in draining maxillary LNs (Fig. 6B and Table S1). This
is probably not due to a differential expression of ROR?t, impor-
tant for IL-17 production, because we found that skin and PLN
iNKT cells from Rorc(?t)-GfpTGexpress the same level of ROR?t
(data not shown). Importantly, the frequency of IL-17-producing
iNKT cells increased after induction of ear skin inflammation with
topical application of TPA (Fig. 6B). TPA application is also ac-
companied by an increase in ear thickness and in number of iNKT
cells, with a peak at day 8, preceded by increased iNKT cell num-
bers in the draining maxillary LNs (Fig. S6). Skin and maxillary
LN iNKT cells proliferate after TPA application, as assessed with
Ki67 staining (Fig. 6C). Importantly, we observed a preferential
expansion of ROR?t?among iNKT cells (Fig. 6C).
Overall, our results indicate that skin and PLN iNKT cells have
similar phenotypical and functional features, and that both popu-
lations respond preferentially under inflammatory conditions.
IL-6-independent generation and response of ROR?t?
It has been shown previously that IL-6 is not required for IL-17
production from spleen iNKT cells (15). To address the question of
whether PLN and skin iNKT cell generation and function are al-
tered in the absence of IL-6, we analyzed the phenotype and the
function of both populations in IL-6?/?mice. We found that iNKT
cells are present in the skin and PLN of IL-6?/?mice with similar
composition and absolute numbers to what we observed in WT
animals, although iNKT cell frequencies in IL-6?/?mice were
higher (Fig. 7 and data not shown). Moreover, the frequency of
IL-17?iNKT cells, detected after stimulation with PMA/ionomy-
cin, is unaltered in the mutant mice (data not shown). This indi-
cates that skin and PLN IL-17?iNKT cells are normally generated
in the absence of IL-6 and with no intrinsic defect in their capa-
bility to produce IL-17. To address whether their function could be
altered in vivo in the absence of IL-6, we analyzed both popula-
tions after ear inflammation with TPA. We found that skin and
PLN iNKT cells expanded, as assessed by Ki67 staining (data not
shown), and their absolute numbers exceeded what we observed in
WT animals (Fig. 7). This increase in iNKT cell numbers in the
absence of IL-6 might be due to the lower frequency of cell types
that compete with iNKT cells for growth and differentiating fac-
tors. In addition, skin and PLN iNKT cells produce IL-17 in the
absence of IL-6, with a frequency of IL-17?-producing cells
similar to the one observed in WT animals (Fig. 7). Overall, our
and ?-GalCer, one million BM-DCs (DC), DCs preloaded with ?-GalCer (DC?-GalCer), DCs stimulated with LPS (DCLPS), and DCs activated with LPS
and preloaded with ?-GalCer (DC?-GalCer/LPS). Contralateral footpads were injected with 100 ?l of PBS. Three days after injection, cells from popliteal LNs
were counted, assessed for their frequency of Tet?iNKT cells, and, in parallel, stimulated with PMA/ionomycin. Four hours later, cells were subjected
to intracellular staining to assess IL-17 production and Ki67 expression. A, Shown are the fold increase in total cell numbers, Tet?, and IL-17?/Tet?iNKT
cells in popliteal LNs after different condition injections as compared with contralateral LNs after PBS injection. B, Shown are representative dot plots of
IL-17 production and Ki67 expression by Tet?iNKT cells. Numbers represent percentages. Data are representative of five experiments, with three mice
pooled in each experiment. Values of p by unpaired Student’s t test are shown.
In vivo expansion of PLN iNKT cells. Mice were injected in the footpad with 100 ?l of PBS containing one of the following: LPS, LPS
2146SKIN AND LN IL-17?NK1.1?iNKT CELLS
results indicate that neither the generation nor the function of
PLN and skin iNKT cells is affected in the absence of IL-6, and
that iNKT cells could play crucial and nonredundant roles dur-
We show the existence of a stable rather than transitional NK1.1?
iNKT cell population resident in LN. The absence of IL-2R? ex-
pression at the surface of these cells, crucial for iNKT cell matu-
ration (24), indicates that they are not equipped to transit to the
NK1.1?stage like previously described for conventional iNKT
cells, and thus represent a distinct iNKT cell population. Impor-
tantly, we found that PLNs comprise a major DN ROR?t?IL-17-
secreting subset defined by the expression of CCR6, CD103, and
CD121a. We also unravel the existence of a phenotypic and func-
tionally equivalent population in the skin, and found that both PLN
and skin iNKT cell populations respond preferentially to inflam-
matory signals, independently of IL-6. CD103 (associated with ?7)
was initially described as a marker for intraepithelial T cells re-
siding in the gut wall and skin, and important for their retention
(25). We found CD103 to be expressed on most skin-resident
iNKT cells, and thus might play a role in their retention. CD103
expressed on PLN IL-17?NK1.1?iNKT cells is most likely not
related to LN homing, especially because E-cadherin, recognized
by CD103 (26), is not expressed on endothelium or LNs. Our
results indicate the lack of CD62L and CCR7 expression (data not
shown), which are key homing receptors mediating lymphocyte
entry into LNs, and suggest that other receptors, yet to be deter-
mined, are involved in this process or that they are highly mobile
in lymph. The presence of CD103 on IL-17?iNKT cells might
indicate their previous maturation in a TGF-?-rich milieu. Accord-
ingly, TGF-? has been proposed as a key factor in the development
of iNKT cells (27). Early data showed that CD103 enhanced CD3-
induced activation (28); its costimulatory function could point to-
ward a role in the development of this subset. CCR6 was mainly
shown to be expressed on subsets of T cells, DCs, and Langerhans
cells, and required for the trafficking of these cells via CCL20
expressed by epithelial cells (29). The expression of CCR6 on
Th17 cells was reported on human and mouse T cells (30, 31), and
has been shown to be involved in the recruitment of pathogenic T
cells in different autoimmune diseases (29). Our in vitro chemo-
taxis experiments show that PLN iNKT cells are attracted by
CCL20 (data not shown). We could hypothesize that during in-
flammatory conditions, CCR6?IL-17?NK1.1?iNKT cells could
be attracted by CCL20 produced by inflamed epithelial cells. The
presence of CD103 on IL-17?NK1.1?iNKT cells predisposes
them for retention within epithelial sites, allowing them to partic-
ipate in local responses. Accordingly, CD103 was shown to be
expressed on highly potent regulatory CD25?and CD25?T cell
subsets in mice (32), and data from CD103-deficient animals sug-
gest that this molecule might indeed be involved in the control of
autoimmunity in the skin (33). Thus, the chemokine receptor
CCR6 and the integrin CD103 can be regarded as novel markers
for an IL-17?NK1.1?iNKT cell subset specialized in exerting
function within the skin. The functions of CCR6 and CD103 and
their ligands in controlling in vivo migration and function of IL-
17?NK1.1?iNKT cells remain to be established by analysis of
knockout animals. In addition to the migratory potential of PLN
iNKT cells to inflamed skin, our results indicate that iNKT cells
are already present in this site. These cells are less potent in pro-
ducing IL-17 than their PLN counterpart in normal conditions, but
this potential increases after inflammation. We also show that PLN
iNKT cells respond to their ligand ?-GalCer preferentially in the
presence of LPS, indicating that they might be involved in the
response to inflammation upon bacterial infection. The mecha-
nisms underlying the preferential response of IL-17?iNKT cells in
the presence of inflammatory signals are under investigation. We
could speculate that, once activated, IL-17?iNKT cells in general
will expand and respond specifically to cytokines such as IL-21 or
IL-23. Accordingly, splenic IL-17?iNKT cells have been shown
to constitutively express IL-23R and to respond to IL-23 (15), and
iNKT cells were also shown to respond to IL-21 (34). The specific
expression of CD121a in iNKT cells could play a crucial role in
this preferential response. Accordingly, an in vitro study showed
that the presence of IL-1 expanded selectively Foxp3?NK1.1?
CD1d tetramer-positive cells enriched in IL-17 producers in a
CD4?CD25?T cell culture (35). Also, IL-1, which is a major
cytokine connected to inflammation, was shown recently to be
ident and respond preferentially under inflammation conditions. A, Ear cell
suspensions from 4-mo-old C57BL/6 mice were stained with CD1d tet-
ramers and mAbs directed against NK1.1, CCR6, CD103, CD121a, and
CD4. Dot plots are gated on live (7-AAD?) CD45?. Data are representa-
tive of four individual experiments, with pooled ear cells from at least 20
mice in each experiment. B, Histograms (upper) show frequency of IL-17-
producing cells among iNKT cells after PMA/ionomycin stimulation in
normal and TPA-inflamed skin. Lower are representative histograms. Data
are representative of four individual experiments, with pooled ear cell sus-
pensions from at least 10 mice in each experiment. C, Histograms show
frequency of Ki67?iNKT cells among ROR?t?or ROR?t?subsets in
normal or TPA-inflamed skin ear (upper) and in maxillary LNs (lower).
Data are representative of four individual experiments, with pooled ear
cells from at least 10 mice in each experiment. Values of p by unpaired
Student’s t test are shown.
IL-17?CCR6?CD103?CD121a?iNKT cells are skin res-
2147 The Journal of Immunology
required to prime human CCR6?Th17 cells to enable IL-23-in-
duced cytokine release (36). Our own results show that ear inflam-
mation after TPA application is accompanied by 5-fold increase in
IL-1 transcript (data not shown). Interestingly, although IL-6 was
shown to be a major factor for Th17 cell differentiation, our results
indicate that IL-17?iNKT cell generation and function occur in
the absence of this proinflammatory cytokine. Thus, IL-1 and other
factors yet to be determined are likely to play key roles in iNKT
cell responses in the absence of IL-6 and indicate that iNKT cells
could play a nonredundant role during inflammation.
IL-17?iNKT cells have been described in other studies in the
thymus, liver, spleen, and lung, but not in LNs or skin (13, 15). In
these studies, it was suggested that all NK1.1?iNKT cells have the
potential to secrete IL-17. Our study shows heterogeneity among
NK1.1?iNKT cells, as exemplified by the few IL-17-secreting
cells found in mesenteric LNs. Also, in these studies, tissue-resi-
dent IL-17?NK1.1?iNKT cells represented a minor subset com-
pared with IL-4- and IFN-?-producing NK1.1?iNKT cells. We
found this composition inverted in PLNs and skin, because
NK1.1?IL-17?cells represent the major population among iNKT
cells. This composition heterogeneity could explain the diversity
of iNKT cell function depending on the anatomic location. This
physical separation of iNKT cells with antagonistic functions
could solve part of the paradox of how iNKT cells with different
potentials can respond to similar ligands. Finally, previous studies
did not connect IL-17?iNKT cells with specific phenotypic char-
acteristics, such as CCR6, CD103, and CD121a, which, in the case
of PLN and skin IL-17?iNKT cells, are most likely important for
exerting their function. More work must be done to show in vivo
evidence for a role of IL-17?iNKT cells and other NK1.1?iNKT
subsets in relevant pathological situations.
In conclusion, our results represent a functional and phenotyp-
ical characterization of mature, stable NK1.1?iNKT cells, mainly
located in LNs. Our study shows that the ones located in PLN
mainly secrete the proinflammatory cytokine Th17 and are armed
to exert their function within epithelia, such as the skin, where an
equivalent population of iNKT cells was also detected. Impor-
tantly, both populations respond preferentially to inflammation in-
dependently of Th17 differentiating signals. Our study opens a new
area of investigation regarding the role of these cells in immune
responses to infection and autoimmunity.
We thank Albert Bendelac, Paul B. Savage, Maria Leite-de-Moraes, Bruno
Lucas, and Francoise Lepault for providing reagents and mice; Albert Ben-
delac, Luc Teyton, Olivier Lantz, Maria Leite-de-Moraes, and Agnes Le-
huen for reviewing the manuscript; Bertrand Meresse for help with gut
experiments; Julien Marie for his molecular expertise; and Laetitia Breton
for assistance with animal care. Mouse CD1d monomers were obtained
from Luc Teyton and Mitchell Kronenberg, and through the National In-
stitutes of Health tetramer facility.
The authors have no financial conflict of interest.
1. Bendelac, A., P. B. Savage, and L. Teyton. 2007. The biology of NKT cells.
Annu. Rev. Immunol. 25: 297–336.
2. Benlagha, K., A. Weiss, A. Beavis, L. Teyton, and A. Bendelac. 2000. In vivo
identification of glycolipid antigen-specific T cells using fluorescent CD1d tet-
ramers. J. Exp. Med. 191: 1895–1903.
3. Matsuda, J. L., O. V. Naidenko, L. Gapin, T. Nakayama, M. Taniguchi,
C. R. Wang, Y. Koezuka, and M. Kronenberg. 2000. Tracking the response of
natural killer T cells to a glycolipid antigen using CD1d tetramers. J. Exp. Med.
4. Benlagha, K., T. Kyin, A. Beavis, L. Teyton, and A. Bendelac. 2002. A thymic
precursor to the NK T cell lineage. Science 296: 553–555.
5. Pellicci, D. G., K. J. Hammond, A. P. Uldrich, A. G. Baxter, M. J. Smyth, and
D. I. Godfrey. 2002. A natural killer T (NKT) cell developmental pathway in-
volving a thymus-dependent NK1.1?CD4?CD1d-dependent precursor stage.
J. Exp. Med. 195: 835–844.
6. Taniguchi, M., M. Harada, S. Kojo, T. Nakayama, and H. Wakao. 2003. The
regulatory role of V?14 NKT cells in innate and acquired immune response.
Annu. Rev. Immunol. 21: 483–513.
7. Lee, P. T., K. Benlagha, L. Teyton, and A. Bendelac. 2002. Distinct functional
lineages of human V?24 natural killer T cells. J. Exp. Med. 195: 637–641.
8. Crowe, N. Y., J. M. Coquet, S. P. Berzins, K. Kyparissoudis, R. Keating,
D. G. Pellicci, Y. Hayakawa, D. I. Godfrey, and M. J. Smyth. 2005. Differential
mice and IL-6?/?mice were stained with CD1d tetramers and mAb directed against TCR? in normal or TPA-inflamed skin. Representative dot plots are
gated on live (7-AAD?) CD45?cells. Histograms show absolute number (upper) and percentage of IL-17 (lower) among iNKT cells from C57BL/6 mice
and IL-6?/?mice after PMA/ionomycin stimulation. Data are representative of three individual experiments, with pooled ear cell suspensions from at least
10 mice in each experiment. Values of p by unpaired Student’s t test are shown. B, Shown are representative dot plots of maxillary LN cells from 4-mo-old
C57BL/6 mice and IL-6?/?mice stained with CD1d tetramers, and mAb directed against B220 in normal or TPA-inflamed skin. Histograms show absolute
number (upper) and percentage of IL-17 (lower) among iNKT cells from C57BL/6 mice and IL-6?/?mice after PMA/ionomycin stimulation. Data are
representative of three individual experiments, with pooled maxillary LN cells from at least 10 mice in each experiment. Values of p by unpaired Student’s
t test are shown.
Generation and function in vivo of iNKT cells from PLNs and skin are IL-6 independent. A, Ear cell suspensions from 4-mo-old C57BL/6
2148SKIN AND LN IL-17?NK1.1?iNKT CELLS
antitumor immunity mediated by NKT cell subsets in vivo. J. Exp. Med. 202: Download full-text
9. Johnston, B., C. H. Kim, D. Soler, M. Emoto, and E. C. Butcher. 2003. Differ-
ential chemokine responses and homing patterns of murine TCR?? NKT cell
subsets. J. Immunol. 171: 2960–2969.
10. Geissmann, F., T. O. Cameron, S. Sidobre, N. Manlongat, M. Kronenberg,
M. J. Briskin, M. L. Dustin, and D. R. Littman. 2005. Intravascular immune
surveillance by CXCR6?NKT cells patrolling liver sinusoids. PLoS Biol.
11. Ohteki, T., C. Maki, S. Koyasu, T. W. Mak, and P. S. Ohashi. 1999. Cutting edge:
LFA-1 is required for liver NK1.1?TCR???cell development: evidence that
liver NK1.1?TCR???cells originate from multiple pathways. J. Immunol. 162:
12. Coquet, J. M., S. Chakravarti, K. Kyparissoudis, F. W. McNab, L. A. Pitt,
B. S. McKenzie, S. P. Berzins, M. J. Smyth, and D. I. Godfrey. 2008. Diverse
cytokine production by NKT cell subsets and identification of an IL-17-producing
CD4?NK1.1?NKT cell population. Proc. Natl. Acad. Sci. USA 105:
13. Michel, M. L., A. C. Keller, C. Paget, M. Fujio, F. Trottein, P. B. Savage,
C. H. Wong, E. Schneider, M. Dy, and M. C. Leite-de-Moraes. 2007. Identifi-
cation of an IL-17-producing NK1.1negiNKT cell population involved in airway
neutrophilia. J. Exp. Med. 204: 995–1001.
14. Michel, M. L., D. Mendes-da-Cruz, A. C. Keller, M. Lochner, E. Schneider,
M. Dy, G. Eberl, and M. C. Leite-de-Moraes. 2008. Critical role of ROR-?t in a
new thymic pathway leading to IL-17-producing invariant NKT cell differenti-
ation. Proc. Natl. Acad. Sci. USA 105: 19845–19850.
15. Rachitskaya, A. V., A. M. Hansen, R. Horai, Z. Li, R. Villasmil, D. Luger,
R. B. Nussenblatt, and R. R. Caspi. 2008. Cutting edge: NKT cells constitutively
express IL-23 receptor and ROR?t and rapidly produce IL-17 upon receptor
ligation in an IL-6-independent fashion. J. Immunol. 180: 5167–5171.
16. McNab, F. W., D. G. Pellicci, K. Field, G. Besra, M. J. Smyth, D. I. Godfrey, and
S. P. Berzins. 2007. Peripheral NK1.1 NKT cells are mature and functionally
distinct from their thymic counterparts. J. Immunol. 179: 6630–6637.
17. Laloux, V., L. Beaudoin, C. Ronet, and A. Lehuen. 2002. Phenotypic and func-
tional differences between NKT cells colonizing splanchnic and peripheral lymph
nodes. J. Immunol. 168: 3251–3258.
18. Lochner, M., L. Peduto, M. Cherrier, S. Sawa, F. Langa, R. Varona,
D. Riethmacher, M. Si-Tahar, J. P. Di Santo, and G. Eberl. 2008. In vivo equi-
librium of proinflammatory IL-17?and regulatory IL-10?Foxp3?ROR?t?T
cells. J. Exp. Med. 205: 1381–1393.
19. Hvid, H., I. Teige, P. H. Kvist, L. Svensson, and K. Kemp. 2008. TPA induction
leads to a Th17-like response in transgenic K14/VEGF mice: a novel in vivo
screening model of psoriasis. Int. Immunol. 20: 1097–1106.
20. Stanley, P. L., S. Steiner, M. Havens, and K. M. Tramposch. 1991. Mouse skin
inflammation induced by multiple topical applications of 12-O-tetradecanoyl-
phorbol-13-acetate. Skin Pharmacol. 4: 262–271.
21. Ohteki, T., S. Ho, H. Suzuki, T. W. Mak, and P. S. Ohashi. 1997. Role for
IL-15/IL-15 receptor ?-chain in natural killer 1.1?T cell receptor-???cell de-
velopment. J. Immunol. 159: 5931–5935.
22. Dong, C. 2008. TH17 cells in development: an updated view of their molecular
identity and genetic programming. Nat. Rev. Immunol. 8: 337–348.
23. Parekh, V. V., S. Lalani, and L. Van Kaer. 2007. The in vivo response of invariant
natural killer T cells to glycolipid antigens. Int. Rev. Immunol. 26: 31–48.
24. Matsuda, J. L., L. Gapin, S. Sidobre, W. C. Kieper, J. T. Tan, R. Ceredig,
C. D. Surh, and M. Kronenberg. 2002. Homeostasis of V?14i NKT cells. Nat.
Immunol. 3: 966–974.
25. Cerf-Bensussan, N., A. Jarry, N. Brousse, B. Lisowska-Grospierre, D. Guy-
Grand, and C. Griscelli. 1987. A monoclonal antibody (HML-1) defining a novel
membrane molecule present on human intestinal lymphocytes. Eur. J. Immunol.
26. Cepek, K. L., S. K. Shaw, C. M. Parker, G. J. Russell, J. S. Morrow, D. L. Rimm,
and M. B. Brenner. 1994. Adhesion between epithelial cells and T lymphocytes
mediated by E-cadherin and the ?E?7integrin. Nature 372: 190–193.
27. Marie, J. C., D. Liggitt, and A. Y. Rudensky. 2006. Cellular mechanisms of fatal
early-onset autoimmunity in mice with the T cell-specific targeting of transform-
ing growth factor-? receptor. Immunity 25: 441–454.
28. Sarnacki, S., B. Begue, H. Buc, F. Le Deist, and N. Cerf-Bensussan. 1992. En-
hancement of CD3-induced activation of human intestinal intraepithelial lym-
phocytes by stimulation of the ?7-containing integrin defined by HML-1 mono-
clonal antibody. Eur. J. Immunol. 22: 2887–2892.
29. Schutyser, E., S. Struyf, and J. Van Damme. 2003. The CC chemokine CCL20
and its receptor CCR6. Cytokine Growth Factor Rev. 14: 409–426.
30. Acosta-Rodriguez, E. V., L. Rivino, J. Geginat, D. Jarrossay, M. Gattorno,
A. Lanzavecchia, F. Sallusto, and G. Napolitani. 2007. Surface phenotype and
antigenic specificity of human interleukin 17-producing T helper memory cells.
Nat. Immunol. 8: 639–646.
31. Hirota, K., H. Yoshitomi, M. Hashimoto, S. Maeda, S. Teradaira, N. Sugimoto,
T. Yamaguchi, T. Nomura, H. Ito, T. Nakamura, et al. 2007. Preferential recruit-
ment of CCR6-expressing Th17 cells to inflamed joints via CCL20 in rheumatoid
arthritis and its animal model. J. Exp. Med. 204: 2803–2812.
32. Lehmann, J., J. Huehn, M. de la Rosa, F. Maszyna, U. Kretschmer, V. Krenn,
M. Brunner, A. Scheffold, and A. Hamann. 2002. Expression of the integrin ?E?7
identifies unique subsets of CD25?as well as CD25?regulatory T cells. Proc.
Natl. Acad. Sci. USA 99: 13031–13036.
33. Schon, M. P., M. Schon, H. B. Warren, J. P. Donohue, and C. M. Parker. 2000.
Cutaneous inflammatory disorder in integrin ?E(CD103)-deficient mice. J. Im-
munol. 165: 6583–6589.
34. Coquet, J. M., K. Kyparissoudis, D. G. Pellicci, G. Besra, S. P. Berzins,
M. J. Smyth, and D. I. Godfrey. 2007. IL-21 is produced by NKT cells and
modulates NKT cell activation and cytokine production. J. Immunol. 178:
35. Brinster, C., and E. M. Shevach. 2008. Costimulatory effects of IL-1 on the
expansion/differentiation of CD4?CD25?Foxp3?and CD4?CD25?Foxp3?T
cells. J. Leukocyte Biol. 84: 480–487.
36. Kleinschek, M. A., K. Boniface, S. Sadekova, J. Grein, E. E. Murphy,
S. P. Turner, L. Raskin, B. Desai, W. A. Faubion, R. de Waal Malefyt, et al. 2009.
Circulating and gut-resident human Th17 cells express CD161 and promote in-
testinal inflammation. J. Exp. Med. 206: 525–534.
2149 The Journal of Immunology