HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hameg, A. Articles by Herbelin, A. Search for Related Content PubMed PubMed Citation Articles by Hameg, A. Articles by Herbelin, A.

A Subset of NKT Cells That Lacks the NK1.1 Marker, Expresses CD1d Molecules, and Autopresents the -Galactosylceramide Antigen¹

Agathe Hameg^*, Irina Apostolou, Maria Leite-de-Moraes, Jean-Marc Gombert^§, Corinne Garcia^¶, Yasuhiko Koezuka^||, Jean-François Bach^* and André Herbelin^2,*

^* Institut National de La Santé et de la Recherche Médicale (INSERM) Unité 25 and Centre Claude Bernard, Hôpital Necker, Paris, France; INSERM Unité 277, Institut Pasteur, Paris, France; Centre National de la Recherche Scientifique, Unité Mixte de Recherche 8603, Université René Descartes, Hôpital Necker, Paris, France; ^§ Laboratoire d’Immunologie-Immunopathologie, Centre Hospitalier Universitaire, Poitiers, France; ^¶ INSERM Unité 373, Institut Necker, Paris, France; and ^|| Pharmaceutical Research Laboratory, Kirin Brewery Company, Gunma, Japan

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
In the present report, we characterize a novel T cell subset that shares with the NKT cell lineage both CD1d-restriction and high reactivity in vivo and in vitro to the -galactosylceramide (-GalCer) glycolipid. These cells preferentially use the canonical V14-J281 TCR--chain and Vß8 TCR-ß segments, and are stimulated by -GalCer in a CD1d-dependent fashion. However, in contrast to classical NKT cells, they lack the NK1.1 marker and express high surface levels of CD1d molecules. In addition, this NK1.1^- CD1d^high T subset, further referred to as CD1d^high NKT cells, can be distinguished by its unique functional features. Although NK1.1⁺ NKT cells require exogenous CD1d-presenting cells to make them responsive to -GalCer, CD1d^high NKT cells can engage their own surface CD1d in an autocrine and/or paracrine manner. Furthermore, in response to -GalCer, CD1d^high NKT cells produce high amounts of IL-4 and moderate amounts of IFN-, a cytokine profile more consistent with a Th2-like phenotype rather than the Th0-like phenotype typical of NK1.1⁺ NKT cells. Our work reveals a far greater level of complexity within the NKT cell population than previously recognized and provides the first evidence for T cells that can be activated upon TCR ligation by CD1d-restricted recognition of their ligand in the absence of conventional APCs.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
CD1 molecules represent a family of conserved nonpolymorphic cell surface glycoproteins related to the MHC class I and class II molecules, but encoded by genes outside the MHC (1). In accordance with their MHC Ag-presenting molecule-like structure, and their expression by APCs, they have been described as targets for T cell recognition (1, 2, 3). CD1d-reactive T cells have been initially identified in the mouse. They are autoreactive in vitro to CD1d-expressing cells (1, 4) and display a large specificity pattern toward various CD1d-positive cell types and transfectants (5, 6). Among them, NKT cells have been a recent focus of interest (2, 7). In the thymus, the NKT cell subset, which includes both double-negative (DN)³ and CD4⁺ T cells, is characterized by the expression of NK lineage-specific receptors, namely NK1.1 (CD161), and the preferential usage of the invariant V14-J281 TCR--chain paired with Vß8, Vß7, or Vß2 TCR-ß segments (2, 8). These cells can also be distinguished by their intermediate level of TCR expression and their activated/memory-like phenotype (2, 8). A human counterpart of the mouse NKT cells has been identified. It expresses a restricted repertoire homologous to the one expressed by mouse NKT cells and is reactive to CD1d (9, 10). The unique capacity of NKT cells to promptly release IFN- and IL-4 upon TCR engagement is thought to be the basis of their various regulatory functions during the effector phase of immune responses, including regression of tumor metastases (11, 12, 13), regulation of autoimmune diseases (14, 15, 16), and the protection against bacterial or parasitic infections (17, 18, 19).

Although the early cytokine response in the spleen of mice having received an i.v. injection of anti-CD3 mAb can be ascribed to NKT cells (20, 21), little is known about their physiological stimuli. Recently, -galactosylceramide (-GalCer), originally isolated from marine sponge (22), was found to activate most V14-J281⁺/Vß8⁺ NKT cells in a CD1d-dependent fashion (23, 24, 25), suggesting that these cells might recognize unidentified self-glycolipids (24, 25). The responsiveness to -GalCer seems to be restricted mainly to V14-J281⁺ NKT cells because most V14-J281⁺ hybridomas are activated by -GalCer, conversely to CD1d-autoreactive hybridomas that do not use the V14 TCR (24, 25).

We have recently reported that CD1d-restricted T lymphocytes that share with NK1.1⁺ T cells the capacity to produce large amounts of IL-4 and IFN- upon TCR cross-linking but do not express the NK1.1 molecule can be identified ex vivo (26). In keeping with these findings, other studies have provided evidence for class-I-restricted and/or CD1d-reactive cells among memory/activated CD4⁺ NK1.1^- T lymphocytes (4, 27). Analyses of the CD1d-restricted CD4⁺ compartment in mice lacking MHC class II molecules (I-Aß^-/- C57BL/6 mice) have shown that it can be subdivided into NK1.1⁺ and NK1.1^- cell populations, whose TCR repertoire was found to be skewed toward Vß8 segments (4, 26, 27, 28). Moreover, Sköld et al. (28) demonstrated that some of these cells express the V14-J281 rearrangement typical of NK1.1⁺ T cells.

To date, -GalCer reactivity during primary cultures has been investigated exclusively in the NK1.1⁺ T cell subset (24). In the present study, we addressed the question whether the CD1d-restricted NK1.1^- T cells were also responsive to this glycolipid Ag. To this end, we isolated NK1.1^- CD4⁺ T splenocytes from I-Aß^-/- C57BL/6 mice in which CD1d-restricted CD4⁺ T cells are enriched (4) and identified a subset of CD1d-restricted NK1.1^- CD4⁺ T cells that, like classical NKT cells, produce high amounts of IL-4 upon stimulation by -GalCer in a CD1d-dependent fashion. We found that, in contrast to the classical NKT population, CD1d-restricted NK1.1^- CD4⁺ T cells display high levels of CD1d molecules on their surface and possess the remarkable capacity of autopresenting the -GalCer Ag. The implications of this property for the regulatory functions of NKT cells during the immune responses will be discussed.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Mice

Wild-type and mutant I-Aß^-/- C57BL/6 mice were bred and maintained in our animal facilities under specific pathogen-free conditions. Mutant CD1d^-/- C57BL/6 mice were kindly provided by L. Van Kaer (Howard Hughes Medical Institute, Vanderbilt University School of Medicine, Nashville, TN) (29).

-GalCer

-GalCer [(2S, 3S, 4R)-1-O-(-D-galactopyranosyl)-2-(N-hexacosanoylamino)-1,3,4-octadecanetriol] (22) was synthesized by Pharmaceutical Research Laboratories, Kirin Brewery (Gunma, Japan). The preparation had a single dominant peak of the expected m.w. by electrospray mass spectrometry, ruling out the presence of significant amounts of degradation products. The stock solution was dissolved at 10 µg/ml in 10% DMSO, and diluted in culture medium to a final concentration of 100 ng/ml. A 10% DMSO vehicle solution diluted to a 0.1% final concentration was routinely used as control.

In vivo treatment with -GalCer

Mice received a single injection of 2–4 µg of -GalCer, (1–2 µg i.v. plus 1–2 µg i.p.), diluted in NaCl. Control mice were injected with an identical volume of vehicle solution alone (0.025% polysorbate). Mice were sacrificed 2 h after injection.

Abs and flow cytometry analysis

FITC- and PE-anti-CD1d (clone 1B1), PE-anti-CD24 (clone M1/69), FITC-anti-CD44 (clone 1 M7.8), PE-anti-NK1.1 (clone PK136), FITC- or APC- anti-TCR-ß (clone H57-597), and APC-anti-CD4 (clone RM4.5) mAbs, PE-anti-IL-4 (clone 11B11) and its PE-isotype control (clone R3-34) mAbs were obtained from PharMingen (San Diego, CA). PE-anti-IFN- (clone XMG1.2) mAb and its PE-isotype IgG1 control mAb were obtained from Caltag Laboratories (Burlingame, CA). Anti-CD62-L (clone MEL-14) and anti-CD1d (clone 20H2) mAbs, kindly provided by F. Lepault (Centre National de la Recherche Scientifique, Unité Mixte de Recherche 8603, Institut Necker, Paris, France) and Dr. A. Bendelac (Department of Molecular Biology, Princeton University, Princeton, NJ), respectively, were purified and biotinylated in the laboratory. Anti-CD8 (clone 53.6.7), anti-TCR-ß (clone H57-597), anti-Vß8 (clone F23.1), and anti-CD24 (anti-heat-stable Ag (HSA), clone J11d) were purified and/or fluoresceinated or biotinylated in our laboratory. For cell surface marker analysis, four-color staining was performed as described (26). After incubation with the appropriate FITC-labeled, PE-labeled, and biotinylated mAbs, cells were incubated with the appropriate APC-labeled mAb plus streptavidin-peridinin chlorophyll protein or streptavidin-Cy-Chrome (PharMingen). Control staining with irrelevant Abs was always performed in parallel. For intracellular cytokine staining, 5 x 10⁵ cells were incubated with FITC-anti-CD1d and biotinylated anti-NK1.1 mAbs. After extensive washing, cells were incubated with APC-anti-TCR-ß mAb and streptavidin-Cy-Chrome. They were fixed in 4% paraformaldehyde for 5 min at room temperature, washed with PBS containing 1% BSA and 0.5% saponin (Sigma, St. Louis, MO), and incubated with either PE-anti-IL-4 or PE-anti-IFN- (or with the appropriate PE-labeled-isotype control) for 30 min at room temperature. The cells were washed again with PBS/BSA/saponin and then with PBS/BSA without saponin to allow membrane closure. A FACScalibur cytometer (Becton Dickinson, Mountain View, CA) was used and a minimum of 5 x 10⁴ events gated from viable cells were acquired with CellQuest software. Results were analyzed using Mac CellQuest software. Each analytical gate contained at least 2 x 10³ events.

Cell preparation

Thymi and spleens were carefully removed from exsanguinated mice. Great care was taken to avoid contamination of thymi by cells from blood or thoracic lymph nodes. DN and CD4⁺ mature thymocytes were enriched by treating freshly isolated thymocytes with the IgM Abs Y169 (rat anti-mouse CD8) and J11d (rat anti-mouse heat stable Ag, anti-HSA) plus complement killing (Low-Tox rabbit Complement, Cedarlane, Ontario, Canada) at 37°C for 40 min. In these conditions, mature CD8⁺, CD4⁺CD8⁺, and most immature CD4^-CD8^- thymocytes died and were further eliminated by centrifugation on a density gradient (J. Prep.; Techgen, Les Ulis, France). Purity of the HSA^-CD8^- thymocyte preparation was checked by staining the cells with PE-anti-HSA (clone M.1/69) and FITC-anti-CD8 (clone 53.6.7). In either case, the preparation contained >95% HSA^-CD8^- cells upon reanalysis. Spleen cell suspensions were prepared using a homogenizer, and RBC were lysed in an hemolysis buffer (NH₄Cl, KHCO₃, EDTA). Splenocytes were enriched for CD4⁺ T cells using anti-CD4-coated magnetic beads (Miltenyi Biotech, Bergisch-Gladbach, Germany), as reported (26). Purity of enriched-CD4⁺ cell fractions was always greater than 92%, except for I-Aß^-/- mice (75–85%). For cell sorting, MACS enriched-CD4⁺ splenocytes were stained with FITC-anti-CD44, PE-anti-NK1.1, biotinylated-anti-CD62L plus SAv-Red613 (Life Technologies, Grand Island, NY), and APC-anti-CD4. NK1.1⁺ and NK1.1^- populations were then sorted (among gated CD44^high CD62L^- CD4⁺ cells) using a FACSvantage cell sorter (Becton Dickinson). In some experiments, MACS-enriched CD4⁺ splenocytes were stained with FITC-anti-CD1d, PE-anti-NK1.1, and APC-anti-CD4. In these conditions, CD1d^low NK1.1⁺, CD1d^low NK1.1^- and CD1d^high NK1.1^- populations were sorted among gated CD4⁺ cells. Purity of each enriched cell fraction was 90% upon reanalysis (see inset of Fig. 3A).

View larger version (25K): [in this window] [in a new window]   FIGURE 3. A subpopulation of nonconventional NK1.1- T cells expresses higher surface levels of CD1d than NK1.1+ T cells. A, Comparison of CD1d expression on NK1.1+ T cells, nonconventional NK1.1- T cells (TCR-ß+ NK1.1+ and TCR-ß+ NK1.1- cells, respectively, gated from enriched CD4+ splenocytes from I-Aß-/- mice), splenic B cells (TCR-ß- B220+ cells gated from splenocytes from I-Aß-/- mice) and mainstream splenic T cells (TCR-ß+ cells gated from splenocytes from CD1d-/- mice). B, CD1d and NK1.1 expression on enriched CD4+ splenocytes (gated TCR-ß+ cells) from I-Aß-/- mice. Percentages of each population are indicated in the quadrants. On the basis of six separate experiments performed in 8- to 14-wk-old class II-deficient mice, CD1dhigh NK1.1- T cells were found to represent around 15–25% of splenic CD4+ T cells, i.e., a proportion 2-fold lower than that of NK1.1+ T cells (30–45%).

To assess cytokine production and proliferation, cells in RPMI 1640 Glutamax culture medium (Life Technologies) supplemented with 10% FCS (Techgen), 2-ME 0.05 mM, penicillin (100 IU/ml), and streptomycin (100 µg/ml) were cultured (2–5 x 10⁴/well; 200 µl final volume) in 96-well round bottom microplates (Nunc, Roskilde, Denmark) with immobilized anti-TCR-ß (clone H57-597) or soluble -GalCer (100 ng/ml) for 60 h in the presence or absence of APCs (autologous or heterologous unseparated spleen cells at 2 x 10⁵/well -irradiated at 2500 rad). The blocking anti-CD1d mAb 20H2 (10 µg/ml) was added simultaneously to the responder spleen cells. For control cultures, the responder cells were omitted or vehicle alone was used instead of -GalCer. No significant proliferation or cytokine production was detected in control wells (Fig. 1A). The supernatants were harvested 60 h later and stored at -70°C until IL-4 and IFN- assays and wells were replenished with medium and pulsed for 7–8 h with 1 µCi of [³H]TdR (5 Ci/mM; Amersham, Buckinghamshire, U.K.). Cells were then harvested and thymidine uptake was measured. Enriched-CD4⁺ spleen cells (1.25 x 10⁵/well; 0.5 ml final volume) obtained from -GalCer- or vehicle-treated mice were cultured in 24-well plates (Falcon, Becton Dickinson) without additional stimulus. Supernatants were harvested after 2 h and stored at -70°C until cytokine assays were performed.

View larger version (30K): [in this window] [in a new window]   FIGURE 1. NK1.1- CD44high CD4+ T spleen cells from class II-deficient mice produce cytokines and proliferate in response to -GalCer in a CD1d-dependent manner. NK1.1- and NK1.1+ CD44high CD4+ spleen cells (5 x 104/well) from I-Aß-/- mice were cultured in the following conditions: A, with or without -GalCer in the presence of APCs isolated from I-Aß-/- mice (CD1d+/+ APCs). Addition of -GalCer causes a strong induction of both proliferation and cytokine production in NK1.1- cultures (70-, 150-, and 100-fold more than the controls for proliferation, IL-4 and IFN-, respectively); B, with -GalCer in the presence of APCs isolated from CD1d-/- mice (CD1d-/- APCs) or in the absence of APCs; C, with -GalCer alone or in combination with anti-CD1d mAb, in the presence of CD1d+/+ APCs and without APCs for NK1.1+ and NK1.1- cells, respectively. Supernatants were harvested 60 h later and assayed for IL-4 and IFN- by specific ELISAs. For proliferation assays, wells were replenished with medium and pulsed for 7–8 h with 1 µCi of [3H]TdR. They were then harvested and thymidine uptake was assessed. The data are means ± SEM from two to three experiments in triplicate. When SEM values are not indicated, data are the arithmetic means of triplicates from one typical experiment of two to five.

Analysis of V14-C and V14-J281 mRNA expression

Total RNA was extracted with RNABle reagent (Eurobio, Paris, France), followed by ethanol precipitation and resuspension of pellets in 20 µl distilled water. Reverse transcription was performed with random hexamers and oligo(dT) priming using standard methods. Quantitative kinetic ELISA PCR was conducted as described using oligonucleotides specific for C, V14, and J281 (30). In this method, for two samples containing the same amount of C, a shift of n cycles along the x-axis of the two amplification curves represents an 1.8ⁿ-fold difference in the two samples studied. Immunoscope analysis of the V14C rearrangement diversity was performed as described elsewhere (19, 31). Briefly, the indicated cDNAs were subjected to a 40-cycle PCR using the V14- and C-specific unlabeled primers. The resulting PCR products were then used in run-off experiments using the C-specific nested fluorescent primer. The length of the CDR3 region and the intensity of each band from run-off products were determined on an automated sequencer (Perkin-Elmer, Norwalk, CT) and analyzed using the immunoscope software.

Oligonucleotides and probes

For quantitative kinetic ELISA PCR, all primers and probes were obtained from Genosys Biotechnologies (Cambridgeshire, U.K.). The following primer specific sequences were used: biotinylated-V14 primer, GGGAGATACTCAGCAACTCTGG; J281 primer, TCCCAGCTCCAAAATGCAGCC; biotinylated-3'C, CTCGGTCAACGTGGCATCACA; and 5'C, CCCTCTGCCTGTTCACCGACTT. The following fluoresceinated probe specific sequences were used for chemiluminescence detection of the amplified products: V14 probe, CAGCAGGGTGGCTGTGAT; C probe, GAGACCAACGCCACCTAC. For RT-PCR-based immunoscope technique, the following primers were used: V14, CTAAGCACAGCACGCTGCACA; C, TGGCGTTGGTCTCTTTGAAG; and labeled C, FAM-ACACAGGAGGTTCTGGGTTC.

Cytokine assays by specific ELISA

IL-4 and IFN- contents in supernatants were quantified using standard sandwich ELISAs (26) with a coated capture mAb (11B11 and R46A2 mAbs for IL-4 and IFN- assays, respectively) and a biotinylated detection mAb (BVD6 and AN18 mAbs, for IL-4 and IFN- assays, respectively). Conjugation of streptavidin-peroxidase (Amdex, Amersham, Les Ulis, France) was revealed using o-phenylenediamine and hydrogen peroxide (Sigma). IFN- and IL-4 concentrations are expressed in ng/ml, as calculated from calibration curves from serial dilutions of mouse recombinant standards (R&D Systems, Abingdon, U.K.) in each assay. The sensitivity of both IL-4 and IFN- assays was 20–40 pg/ml.

Statistical analysis

Data were expressed as means ± SEM, and statistical significance was evaluated by Student’s t test. Values of p < 0.05 were considered statistically significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
The splenic NK1.1^- CD4⁺ T cell compartment from I-Aß^-/- C57BL/6 mice contains cells that are responsive to -GalCer

CD1d-restricted T cells that are responsible for rapid IL-4 production in vitro upon TCR engagement comprise NK1.1⁺ and NK1.1^- T cells (26). We examined the latter subset for its responsiveness to -GalCer. In 8- to 11-wk-old I-Aß^-/- C57BL/6 mice (further referred to as class II-deficient mice), the large majority (65–70%) of the residual, and thus nonconventional, CD4⁺ T cells (4–5% of splenocytes) are of memory phenotype. These cells comprise NK1.1⁺ (35–45%) and NK1.1^- (55–65%) subsets. Fig. 1A compares cytokine production and proliferation of freshly sorted NK1.1⁺ and NK1.1^- CD44^high CD4⁺ T cells in response to -GalCer. Because CD1d-expressing APCs are required for -GalCer presentation to NK1.1⁺ T cells (23, 24), -irradiated autologous spleen cell suspensions were added to each subset. Both NK1.1^- and NK1.1⁺ T cells responded to -GalCer by a strong proliferation and cytokine production (Fig. 1A). Interestingly, the NK1.1^- T subset was repeatedly found to be the more potent IL-4 producer.

Nonconventional splenic NK1.1^- CD4⁺ T cells do not require exogenous CD1d-expressing cells to respond to -GalCer in a CD1d-dependent manner

Unexpectedly, the cytokine production and proliferation of NK1.1^- CD44^high CD4⁺ T cells from class II-deficient mice was affected neither by the absence of CD1d^+/+ APCs nor by the addition of CD1d^-/- APCs (Fig. 1B). Hence, their stimulation by -GalCer does not depend on the presence of CD1d-expressing cells. To determine whether CD1d was involved, we conducted Ab-blocking studies. IL-4 and IFN- secretion as well as proliferation were abrogated by the simultaneous addition of -GalCer and anti-CD1d mAb to nonconventional NK1.1^- T cells (Fig. 1C). -GalCer-induced activation of the NK1.1⁺ subset was also inhibited when CD1d mAb was added together with APCs (Fig. 1C). The specificity of the treatment with anti-CD1d mAb was assessed by the unaltered cytokine production of NK1.1⁺ T cells cultured on anti-TCR mAb-coated plates in the presence of anti-CD1d mAb (data not shown), which ruled out a nonspecific cytotoxic effect of the mAb on NK1.1^- T cell stimulation.

Evidence for the existence of -GalCer-reactive CD1d-restricted NK1.1^- CD4⁺ splenic T cells in wild-type mice

Do the above -GalCer- reactive CD44^high NK1.1^- CD4⁺ splenic T cells also exist in wild-type mice? Fig. 2A shows that CD44^high NK1.1^- CD4⁺ splenic T cells purified from wild-type mice produce significant amounts of both IL-4 and IFN- in response to -GalCer in vitro, even when addition of autologous APCs was omited. In contrast, similar addition of -GalCer to cultures of CD44^high NK1.1^- CD4⁺ cells from CD1d-deficient mice did not induce cytokine production above the background level obtained without -GalCer. Moreover, addition of heterologous CD1d^+/+ APCs did not reverse the defect in IL-4 production by CD4⁺ splenocytes from CD1d-deficient mice (Fig. 2B). Hence, these results clearly demonstrate that the splenic -GalCer-reactive NK1.1^- CD4⁺ T cell subset is present in wild-type mice but not in CD1d-deficient mice, which likely reflect a failure to positively select CD1d-restricted T cells.

View larger version (20K): [in this window] [in a new window]   FIGURE 2. -GalCer-reactive NK1.1- T cells are present in wild-type mice but not in CD1d-deficient mice. Splenic CD4+ T subsets isolated from wild-type or CD1d-/- mice were cultured with soluble -GalCer. A, CD44high NK1.1- T cells (5 x 104 cells/well) were cultured in the absence of APCs. B, CD4+ T cells were cultured (2 x 105 cells/well) alone or in the presence of APCs isolated from CD1d+/+ mice. Supernatants were harvested 60 h later and assayed for IL-4 and IFN- by specific ELISAs. The data are means ± SEM from two to three experiments in triplicate. When SEM values are not indicated, data are the arithmetic means of triplicates from one typical experiment.

The -GalCer- and CD1d-mediated stimulation of NK1.1^- T cells in the absence of CD1d-expressing APCs could be due to the engagement of endogenous CD1d molecules displayed on the surface of this T cell subset. Indeed, relative to splenic T cells from CD1d-deficient mice, a significant fraction of nonconventional splenic NK1.1^- CD4⁺ T cells from I-Aß^-/- mice expressed CD1d molecules at intermediate levels ranging between NK1.1⁺ T cells and B cells, as assessed by flow cytometry (Fig. 3A). Fig. 3B shows that the large majority of nonconventional CD4⁺ T cells that express CD1d were confined to the NK1.1^- subset. Indeed, while about 50% of nonconventional NK1.1^- CD4⁺ T cells exhibited significant levels of CD1d, this percentage dropped to 10–15% in their NK1.1⁺ counterpart (Fig. 3B). We will further refer to nonconventional NK1.1^- CD4⁺ T cells expressing CD1d as CD1d^high NK1.1^- T cells.

Nonconventional CD1d^high NK1.1^- T cells autopresent the -GalCer Ag

To investigate whether CD1d expression participated in the reactivity of nonconventional NK1.1^- T cells to -GalCer, we tested purified CD1d^low NK1.1^-, CD1d^high NK1.1^-, and CD1d^low NK1.1⁺ T cells (Fig. 4A) for their responsiveness to this glycolipid Ag. The latter two cell subsets could be stimulated by -GalCer in a CD1d-dependent manner, as established by anti-CD1d mAb blocking experiments (Fig. 4B). However, conversely to CD1d^low NK1.1⁺ T cells, the CD1d^high NK1.1^- T population did not require CD1d-expressing APCs to produce IL-4 and IFN- in response to -GalCer. Furthermore, it was biased toward IL-4 rather than IFN- production in response to an identical in vitro stimulation by -GalCer, unlike the CD1d^low NK1.1^- T subset that presents a cytokine profile similar to that of its NK1.1⁺ counterpart (Fig. 4B).

View larger version (28K): [in this window] [in a new window]   FIGURE 4. Nonconventional CD1dhigh NK1.1- T cells but not NK1.1+ T cells produce cytokines in response to -GalCer in an APC-independent manner. A, Relative CD1d expression on purified CD1dlow NK1.1+ CD4+, CD1dhigh NK1.1- CD4+, and CD1dlow NK1.1- CD4+ splenic cells from I-Aß-/- mice. The degree of purity of sorted populations is indicated as percentage in the respective dot plots (see insets). B, Purified splenic populations (2 x 104 cells/well) were cultured with -GalCer with or without CD1d+/+ APCs. Anti-CD1d mAb was added to the cultures of CD1dlow NK1.1+ T cells in the presence of CD1d+/+ APCs and of CD1dhigh NK1.1- T cells in the absence of APCs. Supernatants were harvested after 60 h and assayed for IL-4 and IFN- by specific ELISAs. The data are expressed as means ± SEM from four to six experiments. When SEM values are not indicated, data are the arithmetic means of triplicates from one typical experiment of three. Differences between CD1dlow NK1.1+ and CD1dhigh NK1.1- T cells are statistically significant (p < 0.03 and p < 0.02 for IL-4 and IFN-, respectively).

Nonconventional CD1d^high NK1.1^- T cells contribute to the early burst of IL-4 and IFN- induced by -GalCer in vivo

A hallmark of NK1.1⁺ T cells is their extraordinary ability to synthesize and secrete large amounts of cytokines, especially IL-4, only 2 h after anti-CD3 injection (20, 21). We thus examined whether -GalCer had similar effects on nonconventional CD4⁺ T cells when administered in vivo. Fig. 5A shows that 2 h after -GalCer but not vehicle injection, nonconventional CD4⁺ splenocytes produce significant amounts of both IFN- and IL-4. Intracytoplasmic cytokine staining experiments revealed the contribution of both NK1.1⁺ and CD1d^high NK1.1^- T cell subsets to this early cytokine burst because a significant increase of the frequency of IL-4- and IFN-- producing cells occurred in both subsets (Fig. 5B). The frequency and absolute number of IL-4- and IFN--producing NK1.1⁺ and CD1d^high NK1.1^- T cells in both mutant I-Aß^-/- and littermate mice are recapitulated in Table I. Because of expression of CD1d by a large fraction of mainstream CD4⁺ T cells, frequencies of CD1d^high NK1.1^- T cells secreting IL-4 and IFN- were consistently reduced in wild-type mice as compared with the ones found in I-Aß^-/- mice. Despite this, in wild-type as in I-Aß^-/- mice, the frequency of IL-4- and IFN-- producing CD1d^high NK1.1^- T cells is of about one-half and one-third that of CD1d^low NK1.1⁺ T cells, respectively, in response to -GalCer. Because splenic NK1.1⁺ T cells are present at the same frequency in wild-type and I-Aß^-/- mice, it can be concluded that in the two types of mice the -GalCer-reactive subset within the CD1d^high NK1.1^- population is represented in equivalent number of cells and contributes to the early burst of cytokines (Table I).

View larger version (26K): [in this window] [in a new window]   FIGURE 5. Nonconventional CD1dhigh NK1.1- T cells contribute to the early cytokine burst induced by -GalCer in vivo. I-Aß-/- mice were treated with -GalCer or vehicle and sacrificed 2 h later. Data are representative of one experiment of three. A, Enriched CD4+ splenocytes (1.25 x 105/well) were cultured without additional stimuli. Supernatants were harvested 2 h later and IL-4 and IFN- were quantified by specific ELISAs. B, Intracytoplasmic IL-4 and IFN- staining in CD1dhigh NK1.1- and NK1.1+ TCR-ß+ cells gated from enriched CD4+ splenocytes. Histogram profiles obtained with labeled anti-cytokine mAbs for vehicle-treated mice and isotype controls for -GalCer-treated mice were identical. Means of three experiments for IL-4: 44.4% ± 8.9 and 27.7% ± 2.2 for NK1.1+ and CD1dhigh NK1.1- T cells, respectively; and for IFN-: 37.9% ± 5.6 and 21.9% ± 3.2 for NK1.1+ and CD1dhigh NK1.1- T cells, respectively.

Like NK1.1⁺ T cells, CD1d^high NK1.1^- T cells exhibit a memory/activated phenotype, i.e., CD44^high CD62L^- CD69⁺ and express lower TCR-ß levels than conventional CD4⁺ T cells, although not as low as those expressed by NK1.1⁺ T cells (Fig. 6, Table II, and data not shown). Furthermore, they share with NK1.1⁺ T cells the constitutive expression of CD122 molecules (the ß-chain of the IL-2R) (Fig. 6 and Table II). Unlike conventional CD4⁺ T cells, TCR-ß-chains expressed by NK1.1⁺ and to a lesser extent by NK1.1^- T cells were biased toward a preferential usage of Vß8 segments (Fig. 6 and Table II). Hence, CD1d^high NK1.1^- T cells share more phenotypical features with NK1.1⁺ T cells than with conventional T cells.

View larger version (23K): [in this window] [in a new window]   FIGURE 6. Nonconventional CD1dhigh NK1.1- T cells share more phenotypical features with NK1.1+ T cells than with conventional T cells. CD44, CD122, CD69, TCR-ß, and Vß8 expression on nonconventional CD4+ T subsets (CD1dhigh NK1.1- TCR-ß+ cells and CD1dlow NK1.1+ TCR-ß+ cells gated from enriched CD4+ splenocytes from I-Aß-/- mice) as compared with the conventional CD4+ T cell subset (NK1.1- TCR-ß+ cells gated from enriched CD4+ splenocytes from wild-type mice). CD1dhigh NKT cells are phenotypically similar to NK1.1+ NKT cells, although more heterogenous with regard to the different markers studied. Data are representative of one experiment of three.

Nonconventional CD1d^high NK1.1^- T cells express the canonical V14-J281 TCR--chain that defines NK1.1⁺ T cells

We next examined CD1d^high NK1.1^- CD4⁺ T cells from class II-deficient mice for the expression of specific transcripts of the canonical V14-J281 chain that defines NK1.1⁺ T cells. Quantitative kinetic ELISA PCR analysis revealed a significant expression of V14-J281 mRNA as compared with the background levels in CD4⁺ NK1.1^- T cells from CD1d^-/- mice, known to be deficient in NKT cells (Fig. 7A). Direct comparison between NK1.1⁺ T cells and nonconventional CD1d^high NK1.1^- T cells is shown in Fig. 7B. Clearly, the specific transcripts for the V14-J281 chain were detected in the two cell types. Hence, we analyzed the diversity of the V14 chains expressed by the two samples using the highly sensitive RT-PCR-based immunoscope technique. CD4⁺ NK1.1^- T cells from CD1d^-/- mice displayed a V14C immunoscope profile composed of several peaks, indicative of the polyclonality of V14 chains expressed by this T cell subset (Fig. 7C). In contrast, V14C rearrangements present in both CD1d^low NK1.1⁺ and CD1d^high NK1.1^- T cell populations yielded a single peak corresponding to rearrangements within a 10-amino acid length CDR3 region (Fig. 7C). Direct sequencing of the latter PCR products revealed that they resulted from the amplification of the semiinvariant V14-J281 rearrangement characteristic of NKT cells (data not shown).

View larger version (27K): [in this window] [in a new window]   FIGURE 7. Both CD1dlow NK1.1+ and nonconventional CD1dhigh NK1.1- T subsets express the canonical semi-invariant V14-J281 chain. Total RNA was extracted from the following purified cell subsets: CD1dlow NK1.1+ CD4+ and CD1dhigh NK1.1- CD4+ splenic cells (1–2 x 105/sample) from I-Aß-/- mice and NK1.1- CD4+ splenic cells (4 x 105/sample) from CD1d-/- mice. A and B, After reverse transcription, V14-J281 and C amplicons were quantified using a quantitative kinetic ELISA PCR. The data are representative of one experiment of three. C, Immunoscope analysis of the diversity of V14C. x-axis, amino acid length of the CDR3 region of V14-C-specific cDNAs. y-axis, relative fluorescence intensity.

The thymus is essential for V14⁺ NK1.1⁺ T cell generation (32, 33). To verify the presence of nonconventional CD1d^high NK1.1^- T cells in this organ, we analyzed mature (HSA^-) CD8^- thymocytes that concentrate both CD4⁺ and DN contingents of NK1.1⁺ T cells (16, 34). In young 4- to 5-wk-old class II-deficient mice, CD1d^high NK1.1^- T cells accounted for 15–20% of HSA^- CD8^- thymocytes (Fig. 8A) and comprised both CD4⁺ and DN T cells (Fig. 8B). The phenotype of thymic CD1d^high NK1.1^- T cells resembled their splenic counterpart (Fig. 8C), but differed from conventional T cells by a lower TCR expression, a TCR repertoire highly skewed toward Vß8 segment usage (36% of Vß8⁺ in class II-deficient mice vs 21.5% in wild-type mice), an activated/memory-like phenotype (CD44^high), and the expression of CD122 (50% of CD122⁺ in class II-deficient mice vs 4–7% in wild-type mice). The phenotype of thymic CD1d^high NK1.1^- T cells was similar to that of thymic NK1.1⁺ T cells with the exception of their Ly-6C expression (Fig. 8C).

View larger version (42K): [in this window] [in a new window]   FIGURE 8. Nonconventional CD4+ and DN CD1dhigh NK1.1- T cells are found in the thymus and share similar phenotypical features with NK1.1+ thymocytes. Freshly isolated HSA- and CD8-depleted thymocytes were from mutant I-Aß-/- and wild-type mice. A, Distribution of CD1d and NK1.1 surface markers on gated TCR-ß+ thymocytes from I-Aß-/- mice. B, Distribution of CD4+ and DN cells among gated TCR-ß+ CD1dhigh NK1.1- and TCR-ß+ CD1dlow NK1.1+ thymocytes from I-Aß-/- mice. C, CD44, CD122, Ly-6C, TCR-ß, and Vß8 expression on nonconventional cell subsets (gated CD1dhigh NK1.1- TCR-ß+ and CD1dlow NK1.1+ TCR-ß+ thymocytes from I-Aß-/- mice) and on conventional CD4+ T cells (gated NK1.1- TCR-ß+ CD4+ thymocytes from wild-type mice). Data represent a typical experiment of two.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The present study was performed with CD1d^high NKT cells purified from class II-deficient mice. Characterization of this NKT subset in wild-type mice was complicated by the expression of CD1d molecules by some conventional CD4⁺ T cells (data not shown) and by their low frequency among CD4⁺ NK1.1^- T cells. Nevertheless, -GalCer-reactive CD1d^high NKT cells do also exist in wild-type mice because their splenic CD44^high NK1.1^- CD4⁺ T cells produce significant amounts of both IL-4 and IFN- in response to -GalCer, even in the absence of CD1d-expressing cells. In accordance with these data, CD4⁺ T splenocytes from wild-type mice respond to -GalCer in the absence of exogenous APCs. Moreover, Burdin et al. (24) have recently identified a V14-expressing CD1d-autoreactive NKT cell hybridoma derived from wild-type mice that expresses CD1d at high levels and is capable of autopresenting -GalCer. Finally, intracytoplasmic cytokine staining performed following a short-term treatment of mice by -GalCer allowed us to determine that the ratio of -GalCer-responding CD1d^high NKT cells to -GalCer-responding NK1.1⁺ NKT cells is identical in wild-type mice and I-Aß^-/- mice.

The unique functional properties of CD1d^high NKT cells demonstrated herein raise the question of the physiological relevance of either NKT subset in various immune responses, including those previously ascribed to NK1.1⁺ NKT cells. Because the activation of CD1d^high NKT cells by -GalCer in vitro occurs in the absence of APCs it is likely that they are also autonomous in vivo. If so, their independence from costimulatory signals delivered by APCs would endow them with the advantage of reacting to glycolipid Ags more rapidly and more efficiently than their NK1.1⁺ counterpart. Surprisingly and in accordance with this hypothesis, CD1d^high NKT cells, but not NK1.1⁺ NKT cells, were found to express CD40 and CD86, two costimulatory molecules usually expressed by APCs (A. Herbelin and A. Hameg, unpublished observations).

Although -GalCer is a synthetic glycosylceramide, it might be hypothesized that it mimics endogenous Ag(s) expressed on normal/activated, stressed, apoptotic, or tumorized cells (25). CD1d^high NKT cells might thus be envisioned as an early sentinel to survey at any moment the homeostasis or integrity of cells expressing such endogenous Ags. Through the expression of CD1d, they could also exert other putative functions, such as Ag presentation to other CD1d-restricted T cells, whether they belong to the NKT subset or not, thus rendering them capable of participating in their thymic and/or peripheral selection/education process. Because NK1.1⁺ NKT cells can recognize exogenous glycolipids, such as phosphatidylinositolmannosides derived from mycobacterial cell walls (19) and glycosylphosphatidylinositols isolated from the membrane of parasites (35), CD1d^high NKT cells might be able to do the same, thus counteracting systemic infections by responding to soluble parasite or mycobacterial derivatives in an APC-independent manner. Do NK1.1⁺ and CD1d^high NKT cells possess different Ag specificities, thus enabling them to exert a large spectrum of recognition? As far as putative endogenous Ags are concerned, it would be informative to test whether CD1d^high NKT cells, like NK1.1⁺ NKT cells (6, 24, 25) display tissue-specificity. Future identification of natural ligand(s) for these two NKT cell types in mammals will provide further insights into these issues.

Recent studies have suggested that -GalCer might emerge as a useful tool to manipulate the balance of adaptive immune responses by directing conventional T cells to a Th2 differentiation pathway (36, 37). It would be of interest to determine the respective contribution of NK1.1⁺ and CD1d^high NKT cell subsets in the polarization of the immune response. In vivo, we show that both subsets contribute to the burst of IL-4, whereas a clear dichotomy between the two populations in terms of cytokine production occurs in vitro. Indeed, NK1.1⁺ NKT cells produce IL-4 and IFN- at equivalent amounts whereas CD1d^high NKT cells generate much more IL-4 than IFN-. After -GalCer administration, the ratio of IL-4-producing cells to IFN--producing cells was for CD1d^high NKT cells and NK1.1⁺ NKT cells of 1.53 ± 0.12 (mean ± SEM, n = 4) and 1.16 ± 0.07 (mean ± SEM, n = 4) in I-Aß^-/- mice (p < 0.006), respectively, and of 1.71 ± 0.01 (mean ± SEM, n = 2) and 1.16 ± 0.14 (mean ± SEM, n = 2) in wild-type mice, respectively. Taken together, it suggests that NK1.1⁺ NKT cells behave like Th0 cells, whereas CD1d^high NKT cells behave like Th2-cells. The identification of glycolipid Ags stimulating CD1d^high NKT cells but not NK1.1⁺ NKT cells would be useful for studies of autoimmune diseases for which a shift toward a Th1 profile and a deficit in NKT cells have been evidenced (15, 16).

Another important issue of this work concerns the possible lineage relationship between -GalCer-reactive NK1.1⁺ and NK1.1^- T cells. Knowing that NKT cells can upon in vitro activation down-regulate the NK1.1 molecule (38) and up-regulate CD1d in a transitory manner (A. Herbelin and A. Hameg, unpublished observations), it could be argued that CD1d^high NKT cells correspond to activated splenic NK1.1⁺ NKT cells. In naive mice, a low but significant fraction (10–15%) of the NK1.1⁺ pool of cells express high levels of CD1d. Because of the increased size of the latter cells, one can assume that they are blastic at a high frequency (51.3 ± 4.7%, n = 4), which contrasts with the 2-fold reduced frequency of blastic cells among the CD1d^high NKT cells (24.2 ± 5.6%; n = 4). If the CD1d^high NKT cells were exclusively arising from the NK1.1⁺ NKT cells, one would expect to find about 50% of blastic cells within the CD1d^high NKT cell compartment. Thus, although we cannot exclude that a fraction of CD1d^high NKT cells derives from NK1.1⁺ NKT cells having up-regulated CD1d and down-regulated NK1.1 molecules, our data support the idea that a large majority of CD1d^high NKT cells express constitutively CD1d. The presence of CD1d^high NKT cells in thymus and their identical activated/memory phenotype in thymus and spleen are consistent with the notion that their phenotype is not exclusively due to activation and/or education processes in secondary lymphoid organs. These latter observations along with the CD1d restriction of both populations are in accordance with a shared selection pathway whose subtle differences might explain the distinct phenotype. It might indeed be hypothesized that double positive (DP) thymocytes, which express CD1d, lose their expression according to their TCR specificity after interaction with different endogenous Ags. An alternative possibility is that CD1d^low and CD1d^high populations diverge at early stage of differentiation and could exist at DP stages before positive selection occurs.

In conclusion, this work provides evidence for a novel subset of CD1d-restricted and -GalCer-reactive NKT cells and reveals an additional level of complexity within the NKT cell population. Interestingly, two very recent reports (which were published during the preparation of this manuscript) demonstrated the existence of -GalCer-reactive NK1.1^- NKT cells (39, 40). The respective roles of each NKT subset in the regulation of homeostasis and pathophysiological responses need to be further investigated.

	Acknowledgments

 
We thank C. Gouarin for her excellent technical assistance. We thank Dr. L. Van Kaer (Howard Hughes Medical Institut, Vanderbilt University School of Medicine, Nashville, TN) for providing mutant CD1d^-/- mice. We thank Dr. A. Bendelac (Department of Molecular Biology, Princeton University, Princeton, NJ) for generously providing anti-CD1d mAb. We also thank Drs. Michel Dy and Elke Schneider for critical review of the manuscript.

	Footnotes

 
¹ This work was supported by institute funds from Institut National de La Santé et de la Recherche Médicale and the Ligue Nationale contre le Cancer (Axe Immunologie, 2000). A.H. is supported by the Fondation pour la Recherche Medicale.

² Address correspondence and reprint requests to Dr. André Herbelin, Institut National de La Santé et de la Recherche Médicale Unité 25, Hôpital Necker, 161 rue de Sèvres, 75743 Paris Cedex 15, France.

³ Abbreviations used in this paper: DN, double-negative; HSA, heat-stable Ag; -GalCer, -galactosylceramide.

Received for publication May 1, 2000. Accepted for publication August 3, 2000.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Porcelli, S. A., R. L. Modlin. 1999. The CD1 system: antigen-presenting molecules for T cell recognition of lipids and glycolipids. Annu. Rev. Immunol. 17:297.[Medline]
2. Bendelac, A., M. N. Rivera, S. H. Park, J. H. Roark. 1997. Mouse CD1-specific NK1⁺ T cells: development, specificity, and function. Annu. Rev. Immunol. 15:535.[Medline]
3. Maher, J. K., M. Kronenberg. 1997. The role of CD1 molecules in immune responses to infection. Curr. Opin. Immunol. 9:456.[Medline]
4. Cardell, S., S. Tangri, S. Chan, M. Kronenberg, C. Benoist, D. Mathis. 1995. CD1-restricted CD4⁺ T cells in major histocompatibility complex class II-deficient mice. J. Exp. Med. 182:993.[Abstract/Free Full Text]
5. Brossay, L., S. Tangri, M. Bix, S. Cardell, R. Locksley, M. Kronenberg. 1998. Mouse CD1-autoreactive T cells have diverse patterns of reactivity to CD1⁺ targets. J. Immunol. 160:3681.[Abstract/Free Full Text]
6. Chiu, Y. H., J. Jayawardena, A. Weiss, D. Lee, S. H. Park, A. Dautry-Varsat, A. Bendelac. 1999. Distinct subsets of CD1d-restricted T cells recognize self-antigens loaded in different cellular compartments. J. Exp. Med. 189:103.[Abstract/Free Full Text]
7. Bendelac, A., O. Lantz, M. E. Quimby, J. W. Yewdell, J. R. Bennink, R. R. Brutkiewicz. 1995. CD1 recognition by mouse NK1⁺ T lymphocytes. Science 268:863.[Abstract/Free Full Text]
8. Leite-de-Moraes, M. C., M. Dy. 1997. Natural killer T cells: a potent cytokine-producing cell population. Eur. Cytokine Network 8:229.[Medline]
9. Dellabona, P., E. Padovan, G. Casorati, M. Brockhaus, A. Lanzavecchia. 1994. An invariant V 24-J Q/V 11 T cell receptor is expressed in all individuals by clonally expanded CD4–8- T cells. J. Exp. Med. 180:1171.[Abstract/Free Full Text]
10. Exley, M., J. Garcia, S. P. Balk, S. Porcelli. 1997. Requirements for CD1d recognition by human invariant V⁺ CD4^-CD8^- T cells. J. Exp. Med. 186:109.[Abstract/Free Full Text]
11. Takeda, K., S. Seki, K. Ogasawara, R. Anzai, W. Hashimoto, K. Sugiura, M. Takahashi, M. Satoh, K. Kumagai. 1996. Liver NK1.1⁺ CD4⁺ T cells activated by IL-12 as a major effector in inhibition of experimental tumor metastasis. J. Immunol. 156:3366.[Abstract]
12. Cui, J., T. Shin, T. Kawano, H. Sato, E. Kondo, I. Toura, Y. Kaneko, H. Koseki, M. Kanno, M. Taniguchi. 1997. Requirement for V^- 14 NKT cells in IL-12-mediated rejection of tumors. Science 278:1623.[Abstract/Free Full Text]
13. Smyth, M. J., K. Y. Thia, S. E. Street, E. Cretney, J. A. Trapani, M. Taniguchi, T. Kawano, S. B. Pelikan, N. Y. Crowe, D. I. Godfrey. 2000. Differential tumor surveillance by natural killer (NK) and NKT cells. J. Exp. Med. 191:661.[Abstract/Free Full Text]
14. Mieza, M. A., T. Itoh, J. Q. Cui, Y. Makino, T. Kawano, K. Tsuchida, T. Koike, T. Shirai, H. Yagita, A. Matsuzawa, H. Koseki, M. Taniguchi. 1996. Selective reduction of V 14⁺ NK T cells associated with disease development in autoimmune-prone mice. J. Immunol. 156:4035.[Abstract]
15. Wilson, S. B., S. C. Kent, K. T. Patton, T. Orban, R. A. Jackson, M. Exley, S. Porcelli, D. A. Schatz, M. A. Atkinson, S. P. Balk, J. L. Strominger, D. A. Hafler. 1998. Extreme Th1 bias of invariant V 24J Q T cells in type 1 diabetes. Nature 391:177.[Medline]
16. Gombert, J. M., A. Herbelin, E. Tancrede-Bohin, M. Dy, C. Carnaud, J. F. Bach. 1996. Early quantitative and functional deficiency of NK1⁺-like thymocytes in the NOD mouse. Eur. J. Immunol. 26:2989.[Medline]
17. Denkers, E. Y., T. Scharton-Kersten, S. Barbieri, P. Caspar, A. Sher. 1996. A role for CD4⁺ NK1.1⁺ T lymphocytes as major histocompatibility complex class II independent helper cells in the generation of CD8⁺ effector function against intracellular infection. J. Exp. Med. 184:131.[Abstract/Free Full Text]
18. Flesch, I. E., A. Wandersee, S. H. Kaufmann. 1997. IL-4 secretion by CD4⁺ NK1⁺ T cells induces monocyte chemoattractant protein-1 in early listeriosis. J. Immunol. 159:7.[Abstract]
19. Apostolou, I., Y. Takahama, C. Belmant, T. Kawano, M. Huerre, G. Marchal, J. Cui, M. Taniguchi, H. Nakauchi, J. J. Fournie, P. Kourilsky, G. Gachelin. 1999. Murine natural killer T(NKT) cells [correction of natural killer cells] contribute to the granulomatous reaction caused by mycobacterial cell walls [published erratum appears in 1999 Proc. Natl. Acad. Sci. USA 96:7610.]. Proc. Natl. Acad. Sci. USA 96:5141.[Abstract/Free Full Text]
20. Yoshimoto, T., W. E. Paul. 1994. CD4^pos, NK1.1^pos T cells promptly produce interleukin 4 in response to in vivo challenge with anti-CD3. J. Exp. Med. 179:1285.[Abstract/Free Full Text]
21. Leite-de-Moraes, M., A. Herbelin, F. Machavoine, A. Vicari, J. M. Gombert, M. Papiernik, M. Dy. 1995. MHC class I-selected CD4^-CD8^-TCR-ß⁺ T cells are a potential source of IL-4 during primary immune response. J. Immunol. 155:4544.[Abstract]
22. Morita, M., K. Motoki, K. Akimoto, T. Natori, T. Sakai, E. Sawa, K. Yamaji, Y. Koezuka, E. Kobayashi, H. Fukushima. 1995. Structure-activity relationship of -galactosylceramides against B16-bearing mice. J. Med. Chem. 38:2176.[Medline]
23. Kawano, T., J. Cui, Y. Koezuka, I. Toura, Y. Kaneko, K. Motoki, H. Ueno, R. Nakagawa, H. Sato, E. Kondo, H. Koseki, M. Taniguchi. 1997. CD1d-restricted and TCR-mediated activation of NKT cells by glycosylceramides. Science 278:1626.[Abstract/Free Full Text]
24. Burdin, N., L. Brossay, Y. Koezuka, S. T. Smiley, M. J. Grusby, M. Gui, M. Taniguchi, K. Hayakawa, M. Kronenberg. 1998. Selective ability of mouse CD1 to present glycolipids: -galactosylceramide specifically stimulates V14⁺ NK T lymphocytes. J. Immunol. 161:3271.[Abstract/Free Full Text]
25. Gumperz, J. E., C. Roy, A. Makowska, D. Lum, M. Sugita, T. Podrebarac, Y. Koezuka, S. A. Porcelli, S. Cardell, M. B. Brenner, S. M. Behar. 2000. Murine CD1d-restricted T cell recognition of cellular lipids. Immunity 12:211.[Medline]
26. Hameg, A., C. Gouarin, J. M. Gombert, S. Hong, L. Van Kaer, J. F. Bach, A. Herbelin. 1999. IL-7 up-regulates IL-4 production by splenic NK1.1⁺ and NK1.1- MHC class I-like/CD1-dependent CD4⁺ T cells. J. Immunol. 162:7067.[Abstract/Free Full Text]
27. Chen, H., W. E. Paul. 1998. A population of CD62L^low Nk1.1^- CD4⁺ T cells that resembles NK1.1⁺ CD4⁺ T cells. Eur. J. Immunol. 28:3172.[Medline]
28. Skold, M., A. Rytter, F. Ivars, S. Cardell. 1999. Characterization of subpopulations of T-cell receptor intermediate (TCR^int) T cells. Scand. J. Immunol. 49:611.[Medline]
29. Mendiratta, S. K., W. D. Martin, S. Hong, A. Boesteanu, S. Joyce, L. Van Kaer. 1997. CD1d1 mutant mice are deficient in natural T cells that promptly produce IL-4. Immunity 6:469.[Medline]
30. Taoufik, Y., D. Froger, S. Benoliel, C. Wallon, E. Dussaix, J. F. Delfraissy, O. Lantz. 1998. Quantitative ELISA-polymerase chain reaction at saturation using homologous internal DNA standards and chemiluminescence revelation. Eur. Cytokine Network 9:197.[Medline]
31. Pannetier, C., J. P. Levraud, A. Lim, J. Even, P. Kourilsky. 1997. J. R. Oksenznberg, and L. Wang, and J. Y.Y. Jeffery, eds. The T-Cell Receptor: Selected Protocols and Applications 287. Chapman and Hall, New York.
32. Tilloy, F., J. P. Di Santo, A. Bendelac, O. Lantz. 1999. Thymic dependence of invariant V 14⁺ natural killer-T cell development. Eur. J. Immunol. 29:3313.[Medline]
33. Hammond, K., W. Cain, I. van Driel, D. Godfrey. 1998. Three day neonatal thymectomy selectively depletes NK1.1⁺ T cells. Int. Immunol. 10:1491.[Abstract/Free Full Text]
34. Bendelac, A., N. Killeen, D. R. Littman, R. H. Schwartz. 1994. A subset of CD4⁺ thymocytes selected by MHC class I molecules. Science 263:1774.[Abstract/Free Full Text]
35. Schofield, L., M. J. McConville, D. Hansen, A. S. Campbell, B. Fraser-Reid, M. J. Grusby, S. D. Tachado. 1999. CD1d-restricted immunoglobulin G formation to GPI-anchored antigens mediated by NKT cells. Science 283:225.[Abstract/Free Full Text]
36. Burdin, N., L. Brossay, M. Kronenberg. 1999. Immunization with -galactosylceramide polarizes CD1-reactive NK T cells towards Th2 cytokine synthesis. Eur. J. Immunol. 29:2014.[Medline]
37. Singh, N., S. Hong, D. C. Scherer, I. Serizawa, N. Burdin, M. Kronenberg, Y. Koezuka, L. Van Kaer. 1999. Cutting edge: activation of NK T cells by CD1d and -galactosylceramide directs conventional T cells to the acquisition of a Th2 phenotype. J. Immunol. 163:2373.[Abstract/Free Full Text]
38. Chen, H., H. Huang, W. E. Paul. 1997. NK1.1⁺ CD4⁺ T cells lose NK1.1 expression upon in vitro activation. J. Immunol. 158:5112.[Abstract]
39. Benlagha, K., A. Weiss, A. Beavis, L. Teyton, A. Bendelac. 2000. In vivo identification of glycolipid antigen-specific T cells using fluorescent CD1d tetramers. J. Exp. Med. 191:1895.[Abstract/Free Full Text]
40. Matsuda, J. L., O. V. Naidenko, L. Gapin, T. Nakayama, M. Taniguchi, C.-R. Wang, Y. Koezuka, M. Kronenberg. 2000. Tracking the response of natural killer T cells to a glycolipid antigen using cold tetramers. J. Exp. Med 192:741.[Abstract/Free Full Text]

This article has been cited by other articles: