INFECTION AND IMMUNITY, Oct. 2006, p. 5903–5913 Vol. 74, No. 100019-9567/06/$08.00�0 doi:10.1128/IAI.00311-06Copyright © 2006, American Society for Microbiology. All Rights Reserved.

Rapid Development of a Gamma Interferon-SecretingGlycolipid/CD1d-Specific V�14� NK1.1� T-Cell

Subset after Bacterial InfectionMasashi Emoto,1,2* Izumi Yoshizawa,1 Yoshiko Emoto,1,2 Mamiko Miamoto,1

Robert Hurwitz,3 and Stefan H. E. Kaufmann1

Department of Immunology, Max Planck Institute for Infection Biology, D-10117 Berlin, Germany1; Laboratory of Immunology,Department of Laboratory Sciences, Gunma University School of Health Sciences, Maebashi, Gunma 371-8511, Japan2; and

Central Support Unit Biochemistry, Max Planck Institute for Infection Biology, D-10117 Berlin, Germany3

Received 24 February 2006/Returned for modification 1 May 2006/Accepted 16 July 2006

The phenotypic and functional changes of glycolipid presented by CD1d(glycolipid/CD1d) specific V�14� T cellsin the liver of mice at early stages of bacterial infection were investigated. After Listeria monocytogenes infection orinterleukin-12 (IL-12) treatment, �-galactosylceramide/CD1d tetramer-reactive (�-GalCer/CD1d�) T cells coex-pressing natural killer (NK) 1.1 marker became undetectable and, concomitantly, cells lacking NK1.1 emerged inboth euthymic and thymectomized animals. Depletion of the NK1.1� subpopulation prevented the emergence of�-GalCer/CD1d� NK1.1� T cells. Before infection, NK1.1�, rather than NK1.1�, �-GalCer/CD1d� T cells coex-pressing CD4 were responsible for IL-4 production, whereas gamma interferon (IFN-�) was produced by cellsregardless of NK1.1 or CD4 expression. After infection, IL-4-secreting cells became undetectable among �-GalCer/CD1d� T cells, but considerable numbers of IFN-�-secreting cells were found among NK1.1�, but not NK1.1�, cellslacking CD4. Thus, NK1.1 surface expression and functional activities of V�14� T cells underwent dramaticchanges at early stages of listeriosis, and these alterations progressed in a thymus-independent manner. In mutantmice lacking all �-GalCer/CD1d� T cells listeriosis was ameliorated, suggesting that the subtle contribution of theNK1.1� T-cell subset to antibacterial protection is covered by more profound detrimental effects of the NK1.1�

T-cell subset.

Natural killer (NK) T cells represent a unique T-cell popu-lation, which shares characteristic features with NK cells. No-tably, both cell types surface express type II C-type lectin,NKR-P1B, and C (NK1.1) (3). In the mouse, the majority ofNKT cells express an invariant T-cell receptor (TCR) � chainencoded by V�14 gene segments paired with J�18 and a highlybiased TCRV� toward V�8.2, V�7, and V�2 (3). The devel-opment of V�14� NKT cells depends on CD1d, which is sur-face expressed together with �2-microglobulin (�2m) (3, 7, 39,48). The �-galactosylceramide (�-GalCer), which is derivedfrom a marine sponge, is recognized by all V�14� NKT cells inthe context of CD1d (29), and microbial ligands have recentlybeen identified (19, 31, 38). The V�14� NKT cells are abun-dant in the liver, where the majority of cells express CD4 andfew cells lack both CD4 and CD8 (12). Liver V�14� NKT cellsrapidly secrete high concentrations of both gamma interferon(IFN-�) and interleukin-4 (IL-4) upon TCR ligation (12, 15,16, 17).

Listeria monocytogenes is a gram-positive facultative intra-cellular bacterium that preferentially replicates in macro-phages and liver parenchymal cells (28). Type I cytokines,notably IL-12 and IFN-�, play a pivotal role in protectionagainst experimental listeriosis of mice (1, 2, 26, 27, 41, 45, 54,

55, 57), whereas type II cytokines such as IL-4 exacerbatedisease (22, 50, 53, 56). After systemic infection, the vast ma-jority of L. monocytogenes organisms are rapidly trapped in theliver (34). Hence, immunocompetent cells, which reside in theliver, are critical for the control of infection (20, 28). Althoughsterile eradication of this pathogen is ultimately achieved byconventional T cells (28), IFN-�-secreting NK1.1� cells seemto participate in protection against L. monocytogenes infection(1, 2, 26, 28, 41, 45, 47, 55).

We have previously shown that cells stained with monoclonalantibodies (MAbs) to CD4 and NK1.1 (CD4� NKT cells) be-come undetectable in the liver of mice after L. monocytogenesinfection (17, 18), which could be due to downmodulation of theNK1.1 marker upon activation (6, 44). In the present study, weassessed NK1.1 expression on V�14� NKT cells in the liver ofmice at early stages of listeriosis by using �-GalCer-loaded CD1d(�-GalCer/CD1d) tetramers. We found that during listeriosis, an�-GalCer/CD1d tetramer-reactive (�-GalCer/CD1d�) NK1.1�

T-cell population developed from an NK1.1� subpopulation in athymus-independent manner. These cells secreted IFN-� but notIL-4. We assume that this �-GalCer/CD1d� NK1.1� subset con-tributes to early antilisterial resistance, thus bridging the gap be-tween early resistance mediated by professional phagocytes andsubsequent acquired immunity mediated by conventional T cells.However, listeriosis was ameliorated in mice lacking �-GalCer/CD1d� T cells. It is possible that the NK1.1� subset, whichproduces IL-4 in addition to IFN-�, is a detriment to the infectedhost and covers protective effects of the NK1.1� subset, whichexclusively produces IFN-�.

* Corresponding author. Mailing address: Laboratory of Immunol-ogy, Department of Laboratory Sciences, Gunma University School ofHealth Sciences, 3-39-22 Showa-machi, Maebashi, Gunma 371-8511,Japan. Phone: 81-27-220-8935. Fax: 81-27-220-8935. E-mail: memoto@health.gunma-u.ac.jp.

5903

on July 12, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

MATERIALS AND METHODS

Mice. Female adult thymectomized (ATX, 8 weeks after birth) C57BL/6 micewere purchased from the Jackson Laboratory (Bar Harbor, ME). Breeding pairsof J�18�/� (29) and C57BL/6 V�a (V�a) mice were kindly provided by M.Taniguchi (RIKEN Research Center for Allergy and Immunology, Yokohama,Japan) and A. M. Livingstone (Imperial College of Science, Technology, andMedicine, London, Great Britain), respectively. These mutants backcrossed ontoC57BL/6 mice (J�18�/� and J�18�/�, 8th generation; V�a, 18th generation),and C57BL/6 mice were maintained under specific-pathogen-free conditions,and weight-matched female mice were used at 8 to 12 weeks of age.

Antibodies. MAbs to TCR�/� (H57-597), TCR�/� (GL3), CD3ε (145-2C11),NK1.1 (PK136), CD4 (YTS.191.1), CD8� (YTS169.4), Fc� receptor (Fc�R)(2.4G2), IL-12 (p40/p70) (C17.8), IL-4 (11B11, BVD6-24G2), and IFN-� (R4-6A2, XMG1.2) were purified from hybridoma culture supernatants. MAbs toIFN-� (XMG1.2) and IL-4 (BVD6-24G2) were biotinylated, and MAbs toTCR�/� and CD3ε were conjugated with fluorescein isothiocyanate (FITC) byconventional methods. FITC-conjugated MAbs to CD11a (M17/4), CD54(H9.2B8), CD25 (7D4), CD122 (TM-�1), CD49d (R1-2), CD69 (H1.2F3), CD4(H129.19), NK1.1 (PK136), and mouse immunoglobulin G2a (IgG2a; R19-15);biotinylated MAbs to NK1.1 (PK136), rat IgG2b (G15-337), and mouse IgG2a

(R19-15); and biotinylated mouse IgG2a (G155-178) were purchased from BDPharMingen (Hamburg, Germany). Rabbit anti-asialo GM1 (ASGM1) antibodyand rabbit IgG were obtained from Wako Chemicals (Neuss, Germany) andSigma-Aldrich (Schnelldorf, Germany), respectively.

Bacteria and infection. L. monocytogenes (strain EGD) recovered from in-fected liver were grown in tryptic soy broth (Difco Laboratories, Detroit, MI) at37°C for 18 h, and aliquots were frozen at �80°C until used. The final concen-tration of viable bacteria was enumerated by plate counts on tryptic soy agar(Difco). Mice were infected intravenously with 2 � 103 L. monocytogenes.

In vivo treatment. Mice were treated intraperitoneally (i.p.) with 1 �g ofrecombinant IL-12 (rIL-12; Genzyme, Alzenau, Germany). To deplete NK1.1�

cells, mice were treated i.p. with anti-NK1.1 MAb (300 �g) 4 and 2 days beforeinfection. For depletion of NK cells, mice were treated i.p. with anti-ASGM1antibody (5 mg) 3 days before infection. To deplete CD4� cells, mice weretreated i.p. with anti-CD4 MAb (300 �g) 4 and 2 days before infection. Depletionof NK1.1� cells (95%), NK cells (95%), and CD4� cells (98%) was verifiedby immunohistochemistry and/or flow cytometry. For endogenous IL-12 neutral-ization, mice were treated i.p. with anti-IL-12 MAb (500 �g) 2 h before infection.Isotype-matched MAbs purified by the same procedure as for specific MAbs orphosphate-buffered saline (PBS) used for MAb purification were used as con-trols. Rabbit IgG served as a control for anti-ASGM1 antibody.

FIG. 1. Appearance of �-GalCer/CD1d� T cells in the liver ofC57BL/6 mice in response to L. monocytogenes infection or rIL-12treatment. Mice were infected with L. monocytogenes (A, B, and C) ortreated with rIL-12 (D) on day 0, and HLs were prepared on the daysindicated in the figure. Cells were stained with FITC-conjugated anti-TCR�/� MAb, biotinylated anti-NK1.1 MAb, and PE-labeled �-Gal-Cer/CD1d tetramers, followed by SA-conjugated CyChrome. (A) Dataare expressed as dot plots after gating on lymphoid cells. The numbersin the dot plots represent the percentages of each cell populationwithin the square. Representative staining patterns from three to sixmice at each time point are shown. (B) Data represent recovery num-bers of HLs and are expressed as means of three to six mice at eachtime point. Vertical lines represent error bars. (C and D) Data repre-sent absolute numbers of �-GalCer/CD1d� NK1.1� T cells (�),�-GalCer/CD1d� NK1.1� T cells (‚), and total �-GalCer/CD1d� Tcells (E) and are expressed as means of three to six mice at each timepoint. �, P 0.05; ��, P 0.01 (before infection versus after infection).

5904 EMOTO ET AL. INFECT. IMMUN.

on July 12, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

Mouse CD1d/�2m tetramer. Mouse CD1d/�2m tetramers were prepared byusing the baculovirus expression system using a CD1d construct with a BirAbiotinylation site, followed by a His6 tag as described previously (37). In brief, Sf9insect cell line (BD Biosciences, Heidelberg, Germany) was infected with mouseCD1d/�2m-expressing virus stock kindly provided by M. Kronenberg (37) at amultiplicity of infection of 0.1 for expanding the viral stock. Culture supernatantswere harvested on day 4 postinfection (p.i.) and used at high multiplicity ofinfection of 1 to 5. These large-scale cultures were performed in Sf-900 IIserum-free medium (Gibco/Invitrogen Corp., Karlsruhe, Germany) and har-vested on day 3 or 4 by centrifugation. Supernatants were concentrated bypassing through a hollow-fiber tangential flow module (MiniKros 1,100 cm2;Spectrum; MembraPure, Boddenheim, Germany). The CD1d molecules werepurified by metal-chelating chromatography (Chelating Sepharose Fast Flow;Amersham Pharmacia Biotech, Uppsla, Sweden) on an NTA-Sepharose column

(Amersham Pharmacia Biotech) charged with cobalt chloride (Roth, Karsruhe,Germany). Protein was eluted with 200 mM imidazole (Merck, Darmstadt,Germany), and pooled fractions were concentrated to 0.5 ml by ultrafiltration(Ultrafree units; Millipore, Bedford, MA). Purity and protein amount wereassessed by using sodium dodecyl sulfate-polyacrylamide gel electrophoresis andBradford reagent (Bio-Rad, Munich, Germany), respectively. Purified CD1dprotein was subsequently biotinylated with BirA enzyme (Avidity, Denver, CO)according to the manufacturer’s instructions. Biotinylated CD1d proteins werepurified by gel filtration (Superdex 200 HR10/30; Amersham Pharmacia Bio-tech). Soluble biotinylated CD1d/�2m protein was loaded with or without �-Gal-Cer (kindly provided by Kirin Brewery Co., Ltd., Tokyo, Japan), which wasdissolved in PBS containing 0.5% Tween 20 at 220 �g/ml at a molar ratio of 1:3(protein-lipid) overnight at room temperature. For tetramerization, streptavidin(SA)-conjugated phycoerythrin (PE) (MobiTec, Gottingen, Germany) wasadded to �-GalCer/CD1d/�2m monomers at a 1:4 (monomer–SA-conjugatedPE) molar ratio. The purification of PE-labeled CD1d/�2m tetramers loadedwith or without �-GalCer was performed by gel filtration (Superdex 200 HR10/30; Amersham Pharmacia Biotek).

Cell preparation and flow cytometry. Mice were killed by cervical dislocationand organs were collected. Hepatic leukocytes (HLs) were prepared as describedpreviously (15). Bone marrow (BM) plugs in the femurs were eluted by flushingwith RPMI 1640 containing 10% fetal calf serum (FCS). T cells were enriched bypassage through a nylon wool column. Splenocytes were prepared by conven-tional methods. For staining with MAbs, cells were incubated with anti-Fc�RMAb and then stained with conjugated MAbs at 4°C for 15 min. BiotinylatedMAbs were visualized by using SA-conjugated CyChrome (BD PharMingen).Stained cells were washed with PBS containing 0.1% bovine serum albumin(Serva, Heidelberg, Germany) and 0.1% sodium azide (Merck, Darmstadt, Ger-many), fixed with 1% paraformaldehyde (Merck), and acquired by FACScan orFACSCalibur (BD Biosciences, Mountain View, CA), and lymphoid cells wereanalyzed with CellQuest software. For staining with �-GalCer/CD1d tetramer,cells were stained with PE-labeled �-GalCer/CD1d tetramers for 15 min at roomtemperature after blocking.

Determination of CFU. Mice were killed by cervical dislocation on day 4 p.i.The liver was perfused with 10 ml of sterile PBS to wash out bacteria in the bloodvessels, and CFU counts in the liver were determined by plating serial dilutionsof liver homogenates on tryptic soy agar plates.

Cell sorting. �-GalCer/CD1d� cells were positively sorted by magnetic cellsorter (Miltenyi Biotec, Bergisch Gladbach, Germany) according to the manu-facturer’s instructions. In brief, HLs were stained with PE-labeled �-GalCer/CD1d tetramers for 15 min at room temperature after blocking and subsequentlyincubated with anti-PE microbeads (Miltenyi Biotec) at 6°C for 15 min. Cellsuspensions were applied to an MS�/RS� column (Miltenyi Biotec), and then�-GalCer/CD1d� cells were enriched. The purity of �-GalCer/CD1d� cells wasconsistently 95%.

ELISPOT assay. IFN-� and IL-4 production was measured by the enzyme-linked immunospot (ELISPOT) method as described previously (15, 17) with aslight modification. Briefly, cells were cultured for 18 h in the presence orabsence of phorbol myristate acetate (PMA; 10 ng/ml; Sigma-Aldrich) and iono-mycin (1 �g/ml) in ELISPOT plates (Millipore, Eschborn, Germany) precoatedwith anti-IFN-� MAb (R4-6A2) or anti-IL-4 MAb (11B11). After being washed,the plates were incubated with biotinylated anti-IFN-� MAb (XMG1.2) or bio-tinylated anti-IL-4 MAb (BVD6-24G2), respectively. For developing spots, SA-conjugated alkaline phosphatase (Dianova, Hamburg, Germany) and BCIP (5-bromo-4-chloro-3-indolylphosphate)–nitroblue tetrazolium tablets (Sigma-Aldrich) were used. The numbers of cytokine-secreting cells were estimated bycounting spots using a dissecting microscope.

Statistical analysis. Statistical significance was determined by using a Studentt test, and P values of 0.05 were regarded as significant.

RESULTS

Differential appearance of NK1.1� and NK1.1� subsets of�-GalCer/CD1d� T cells in the liver after L. monocytogenesinfection. C57BL/6 mice were infected with L. monocytogenes,and the kinetics of V�14� T cells in the liver were monitoredusing �-GalCer/CD1d tetramers. High proportions of �-Gal-Cer/CD1d� T cells, as well as NK1.1� TCR�/� cells, weredetected in the liver before infection (Fig. 1A). Consistent withrecent findings (4), the majority of �-GalCer/CD1d� T cells

FIG. 2. Influence of endogenous IL-12 neutralization on emer-gence of �-GalCer/CD1d� NK1.1� T cells after L. monocytogenesinfection and on bacterial growth in the liver. Mice were treated withanti-IL-12 MAb or PBS 2 h before L. monocytogenes infection.(A) HLs were prepared on day 4 p.i., and cells were stained withPE-labeled �-GalCer/CD1d tetramers and biotinylated anti-NK1.1MAb, followed by SA-conjugated CyChrome. The data are expressedas dot plots after gating on lymphoid cells. The numbers in the dotplots represent percentages of each cell population within the square.Representative staining patterns from three mice in each group areshown. (B) CFU in the liver were determined on day 4 p.i. The data arefrom five mice, and each symbol represents CFU in an individualanimal. Horizontal bars represent mean CFU. Because no significantdifference was found between rat IgG2a (isotype-matched MAb foranti-IL-12 MAb)-treated and PBS-treated group in another experi-ment, PBS was used as a control. *, P 0.05 (PBS-treated groupversus the anti-IL-12 MAb-treated group).

VOL. 74, 2006 CHANGES IN V�14� T CELLS DURING BACTERIAL INFECTION 5905

on July 12, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

coexpressed NK1.1, and a small population lacked this marker.The �-GalCer/CD1d� T cells were undetectable in J�18�/�

mice, verifying that V�14� T cells are specifically detected by�-GalCer/CD1d tetramers (data not shown). The proportions

of �-GalCer/CD1d� T cells, as well as NK1.1� TCR�/� cells,were markedly reduced by day 2 p.i. and were still low on day4 p.i. (Fig. 1A). At these time points, only a few �-GalCer/CD1d� T cells coexpressed the NK1.1 marker.

The numbers of HLs gradually increased after infection(Fig. 1B). The absolute numbers of �-GalCer/CD1d� NK1.1�

T cells were markedly diminished by day 2 p.i., whereas thenumbers of �-GalCer/CD1d� NK1.1� T cells increased sub-sequently, and the latter were dominant among the �-GalCer/CD1d� T-cell population on days 2 and 4 p.i. (Fig. 1C). Noinverse relationship was found in the appearance of NK1.1�

and NK1.1� �-GalCer/CD1d� T cells after L. monocytogenesinfection. Similar results were obtained in V�a mice (data notshown). Thus, the NK1.1� and NK1.1� subsets of �-GalCer/CD1d� T cells showed differential kinetics during listeriosis.

Selective expansion of the �-GalCer/CD1d� NK1.1� T-cellsubset after L. monocytogenes infection involves IL-12. IL-12has been shown to contribute to the compression of the NKTcell population coexpressing CD4 in the livers of L. monocy-togenes-infected mice (18). Here we determined whether IL-12

FIG. 3. Influence of anti-NK1.1 MAb or anti-ASGM1 antibody treat-ment on �-GalCer/CD1d� T cells in the L. monocytogenes-infected liverof C57BL/6 mice. Mice were treated with anti-NK1.1 MAb or PBS ondays �4 and �2 (A, B, C, and D), or with anti-ASGM1 antibody or rabbitIgG on day �3 (E and F), infected with L. monocytogenes on day 0, andthe HLs were prepared on day 0 (A, B, C, and E) and/or day 4 (A, C, D,E, and F). Cells were stained with FITC-conjugated anti-CD3ε MAb,biotinylated anti-NK1.1 MAb and PE-labeled �-GalCer/CD1d tetramers,followed by SA-conjugated CyChrome (A, D, E, and F), biotinylatedmouse IgG2a, and PE-labeled �-GalCer/CD1d tetramers, followed bySA-conjugated CyChrome (B), or FITC-conjugated anti-NK1.1 MAb andPE-labeled �-GalCer/CD1d tetramers (C). (A) Data are expressed as dotplots after gating on lymphoid cells. The numbers in the dot plots repre-sent percentages of each cell population within the square. Representativestaining patterns from eight mice in each group are shown. Because nosignificant difference was found between the anti-human transferrin re-ceptor MAb (isotype-matched MAb for anti-NK1.1 MAb)-treated andPBS-treated groups in another experiment, PBS was used as a control.(B) The data are expressed as dot plots after gating on lymphoid cells. Thenumbers in the dot plots represent percentages of the cell populationwithin the square. Representative staining pattern from two mice isshown. (C) The data are expressed as dot plots after gating on lymphoidcells. The numbers in dot plots represent the percentages of cell popula-tions within the square. Representative staining patterns from two mice ineach group are shown. (D) The data represent absolute numbers ofNK1.1� and NK1.1� �-GalCer/CD1d� T cells. Each symbol representsan individual animal, and horizontal bars represent the means of eightmice. �, P 0.01 (anti-NK1.1 MAb-treated group versus the PBS-treatedgroup). (E) The data are expressed as dot plots after gating on lymphoidcells. The numbers in the dot plots represent percentages of each cellpopulation within the square. Representative staining patterns from fivemice per group are shown. As a control for polyclonal anti-ASGM1antibody (rabbit IgG), rabbit IgG was used. (F) The data represent theabsolute numbers of NK1.1� and NK1.1� �-GalCer/CD1d� T cells. Eachsymbol represents an individual animal, and horizontal bars represent themeans of five mice. �, P 0.01 (anti-ASGM1 antibody-treated groupversus the rabbit IgG-treated group).

5906 EMOTO ET AL. INFECT. IMMUN.

on July 12, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

is involved in the expansion of the �-GalCer/CD1d� NK1.1�

T-cell subset. Consistent with previous findings (44), the abso-lute numbers of the �-GalCer/CD1d� NK1.1� T-cell subsetwere markedly diminished by day 2 after rIL-12 treatment,whereas the NK1.1� T-cell subset was virtually unaffected atthis time point and expanded thereafter (Fig. 1D). Consistentwith this, anti-IL-12 MAb treatment reversed the reduction ofthe NK1.1� T-cell subset and the subsequent increase of the

NK1.1� T-cell subset of the �-GalCer/CD1d� population dur-ing listeriosis (Fig. 2A). Similar results were obtained in V�a

mice (data not shown). Consistent with previous findings (54),the CFU counts in the liver were significantly, althoughslightly, reduced by anti-IL-12 MAb treatment (Fig. 2B).Hence, IL-12 plays a critical role both in the reduction of theNK1.1� subset and in the subsequent increase of the NK1.1�

subset of the �-GalCer/CD1d� T-cell population.

FIG. 3—Continued.

VOL. 74, 2006 CHANGES IN V�14� T CELLS DURING BACTERIAL INFECTION 5907

on July 12, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

The NK1.1� T-cell subset is derived from the NK1.1� subsetof �-GalCer/CD1d� T-cell population. To determine whetherthe appearance of the NK1.1� T-cell subset during listeriosiswas due to downmodulation of the NK1.1 marker, the conse-quences of NK1.1� cell depletion on the emergence of �-Gal-Cer/CD1d� NK1.1� T cells were assessed. In addition to NK(CD3�NK1.1�) cells, total NKT (CD3�NK1.1�) cells, as wellas �-GalCer/CD1d� NK1.1� T cells, became virtually unde-tectable in the liver after anti-NK1.1 MAb treatment, whereasthe proportion of �-GalCer/CD1d� NK1.1� T cells increased(Fig. 3A). Labeling with FITC-conjugated anti-mouse IgG2aMAb (secondary MAb for detection of anti-NK1.1 MAb) didnot stain cells from MAb-treated mice, excluding the residualcoating of the cells with MAb (data not shown). It is likely thatsome cells, which were slightly stained with anti-NK1.1 MAb,still remained in the liver even after anti-NK1.1 MAb treat-ment. This was probably caused by a biotin-mediated nonspe-cific reaction, because (i) similar staining patterns were ob-tained when cells were stained with biotinylated mouse IgG2a(isotype-matched MAb for anti-NK1.1 MAb; Fig. 3B) and (ii)NK1.1 staining was completely negative when FITC-conju-gated MAb to NK1.1 was used (Fig. 3C). The increase in boththe proportion and absolute numbers of �-GalCer/CD1d�

NK1.1� T cells after L. monocytogenes infection was preventedby NK1.1� cell depletion (Fig. 3A and D), suggesting thatNK1.1� cells are critical for the development of �-GalCer/CD1d� NK1.1� T cells in the livers of L. monocytogenes-infected mice.

Because some T cells bearing CD8� and/or TCR�/� alsocoexpress NK1.1 (11, 13), we determined the involvement ofthese cells in the expansion of the �-GalCer/CD1d� NK1.1�

T-cell subset. No measurable alterations were found in thenumbers of �-GalCer/CD1d� NK1.1� T cells in the liver of L.monocytogenes-infected mice after CD8�� or TCR�/�� celldepletion (data not shown). These results exclude the involve-ment of NK1.1� cells expressing CD8� and/or TCR�/� in theexpansion of the �-GalCer/CD1d� NK1.1� T-cell population.

Because both NKT cells and NK cells express NK1.1, weexamined whether NK cells participate in the increase of the�-GalCer/CD1d� NK1.1� T-cell population. C57BL/6 micewere treated with anti-ASGM1 antibody and infected with L.monocytogenes, and the appearance of �-GalCer/CD1d�

NK1.1� T cells was assessed. Consistent with previous findings(14, 40), NK (CD3� NK1.1�) cells became undetectable in theliver after anti-ASGM1 antibody treatment, whereas �-Gal-Cer/CD1d� T cells were virtually unaffected (Fig. 3E). In con-trast to anti-NK1.1 MAb treatment, both the proportion andthe absolute number of �-GalCer/CD1d� NK1.1� T cells dur-ing listeriosis were increased by anti-ASGM1 antibody treat-ment (Fig. 3E and F). These results indicate that NK cellsprevent the development of �-GalCer/CD1d� NK1.1� T cellsin the livers of L. monocytogenes-infected mice. We concludethat the NK1.1� T-cell subset, which appears in the L. mono-cytogenes-infected liver, is primarily derived from the NK1.1�

subpopulation of �-GalCer/CD1d� T cells.�-GalCer/CD1d� NK1.1� T cells in the thymus are not

affected by L. monocytogenes infection. Mice were treated withanti-NK1.1 MAb, and the numbers of �-GalCer/CD1d�

NK1.1� T cells in the thymus and BM were determined. Asmall but distinct population of �-GalCer/CD1d� T cells was

detected in the BM, and the majority of these coexpressedNK1.1 (Fig. 4A). In contrast, virtually all �-GalCer/CD1d� Tcells in the thymus coexpressed NK1.1 on their cell surface.Similar to the liver, �-GalCer/CD1d� NK1.1� T cells in BMand thymus became virtually undetectable after anti-NK1.1MAb treatment (Fig. 4A). This was also true in other organs,including the spleen (data not shown). In parallel to the com-pression of the NK1.1� subset, the proportion of the NK1.1�

subset of �-GalCer/CD1d� T cells increased in both organsafter anti-NK1.1 MAb treatment (Fig. 4A), although the ab-solute numbers remained virtually unchanged (data notshown). Slight staining with anti-NK1.1 MAb was due to abiotin-mediated nonspecific reaction (data not shown). Thus,anti-NK1.1 MAb treatment efficiently depleted �-GalCer/CD1d� NK1.1� T cells from the lymphoid organs.

Similar to the liver, �-GalCer/CD1d� NK1.1� T cells be-came virtually undetectable in the BM of L. monocytogenes-infected mice (Fig. 4B). In contrast, in the thymus of L. mono-cytogenes-infected mice considerable numbers of the NK1.1�

subset persisted, and the subset of �-GalCer/CD1d� T cellsdid not enlarge. This discrepancy is probably due to differentialIL-12 levels in different organs: high frequencies of IL-12-producing cells were detected in BM and liver during listerio-sis, whereas IL-12 producers were virtually absent from thethymus (data not shown). Thus, thymic �-GalCer/CD1d�

NK1.1� T cells were apparently unaffected by L. monocyto-genes infection.

Thymus-independent emergence of �-GalCer/CD1d� NK1.1� Tcells. ATX mice were infected with L. monocytogenes, and thenumbers of liver V�14� T cells were determined by using�-GalCer/CD1d tetramers. Similar to euthymic mice, highnumbers of �-GalCer/CD1d� T cells were detected in the liverbefore infection, and the majority of these cells coexpressedNK1.1 (Fig. 5). Absolute numbers of the NK1.1� subset weremarkedly diminished by day 4 p.i., whereas the NK1.1� subsetof �-GalCer/CD1d� T cells was numerically increased (Fig. 5).Thus, numerical alterations in the NK1.1� and NK1.1� subsetsof the �-GalCer/CD1d� T-cell populations, in response to L.monocytogenes infection, were thymus independent.

Phenotypic changes of �-GalCer/CD1d� T cells during lis-teriosis. We compared the surface expression of various markerson �-GalCer/CD1d� T cells before and after L. monocytogenesinfection. In uninfected mice, the vast majority of �-GalCer/CD1d� T cells expressed CD122 and high levels of CD11a, andthis expression pattern remained unchanged during infection(data not shown). In contrast, CD69, CD54, CD25, and CD49dwere upregulated after infection (Fig. 6). These results suggestthat �-GalCer/CD1d� NK1.1� T cells are highly activated.

Similarly, we compared CD4 expression on �-GalCer/CD1d� T cells in the liver during listeriosis. Before infection,the majority (�80%) of �-GalCer/CD1d� T cells coexpressedCD4 and a minority (�20%) lacked this marker (Fig. 7A). Incontrast, the percentages of CD4� and CD4� cells among�-GalCer/CD1d� T cells shifted to 65 and 35%, respectively,on day 4 p.i. Hence, L. monocytogenes infection primarily com-pressed the CD4� rather than CD4� CD8� (CD4�8�) �-Gal-Cer/CD1d� T-cell population.

Preferential IFN-� production by CD4�8� �-GalCer/CD1d� T cells lacking CD4, CD8, and NK1.1 during listerio-sis. We previously reported that IL-4-secreting cells in the liver

5908 EMOTO ET AL. INFECT. IMMUN.

on July 12, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

become virtually undetectable in parallel with the disappear-ance of CD4� NK1.1� T cells after L. monocytogenes infection(17, 18). Because �-GalCer/CD1d� NK1.1� T cells, whichdeveloped in the liver after L. monocytogenes infection, arephenotypically different from the NK1.1� subpopulation thatresides in naive mice, we wondered whether the NK1.1� andNK1.1� subpopulations differ functionally. To address this is-sue, the �-GalCer/CD1d� T cells were purified from the liverbefore and after infection, and IFN-� and IL-4 secretion weredetermined by ELISPOT assay. Substantial numbers of IFN-�and IL-4 producers were detected among �-GalCer/CD1d� Tcells from uninfected animals after in vitro stimulation withPMA and ionomycin (Fig. 7B). Considerable numbers ofIFN-� producers were found among �-GalCer/CD1d� T cells

from L. monocytogenes-infected mice, whereas IL-4-producingcells became virtually undetectable. These results suggest that�-GalCer/CD1d� NK1.1� T cells, which emerged in the L.monocytogenes-infected liver, fail to produce IL-4 and henceexpress a Th1-like phenotype.

The �-GalCer/CD1d� T cells segregate into several subsetson the basis of CD4/NK1.1 surface expression. Moreover, theNK1.1� and NK1.1� subsets are differentially influenced by L.monocytogenes infection. Finally, the CD4� and CD4�8�

�-GalCer/CD1d� T-cell populations are differentially influ-enced by L. monocytogenes infection. Hence, we wonderedwhether the functional activities of V�14� T cells coexpressingCD4 and/or NK1.1 differ from those lacking CD4 and/orNK1.1. To address these issues, mice were treated with anti-

FIG. 4. Influence of NK1.1� cell depletion or L. monocytogenes infection on �-GalCer/CD1d� T cells in the BM and thymus of C57BL/6 mice.(A) Mice were treated with anti-NK1.1 MAb or PBS on days 0 and 2, and BM cells and thymocytes were prepared on day 4. Cells were stainedand analyzed as in Fig. 3. Representative staining patterns from eight mice in each group are shown. Because no significant difference was foundbetween anti-human transferrin receptor MAb (isotype-matched MAb for anti-NK1.1 MAb)-treated and PBS-treated groups in another experi-ment, PBS was used as a control. (B) Mice were infected with L. monocytogenes, and BM cells and thymocytes were prepared on days 0 and 2 p.i.Cells were stained and analyzed as in Fig. 1. Representative staining patterns from three to six mice in each group are shown.

VOL. 74, 2006 CHANGES IN V�14� T CELLS DURING BACTERIAL INFECTION 5909

on July 12, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

CD4 or anti-NK1.1 MAb and infected with L. monocytogenes,and the frequencies of IL-4 and IFN-� producers among�-GalCer/CD1d� T cells were determined by ELISPOT assay.NK1.1� cells and CD4� cells became virtually undetectable inthe liver after anti-NK1.1 Mab and anti-CD4 MAb treatment,respectively (data not shown). Labeling with FITC-conjugatedanti-mouse IgG2a MAb (secondary MAb for detection of anti-NK1.1 MAb) or FITC-conjugated anti-rat IgG2b MAb (sec-ondary MAb for detection of anti-CD4 MAb) did not staincells from MAb-treated mice, excluding the residual coating ofthe cells with MAbs (data not shown). In uninfected mice,CD4� or NK1.1� cell depletion markedly reduced the num-bers of IL-4-producing �-GalCer/CD1d� T cells, whereas thenumbers of IFN-� producers were comparable in both groups(Fig. 7B). In contrast, after infection, the frequencies of IFN-�-producing �-GalCer/CD1d� T cells were 3.5-fold higher inCD4� cell-depleted mice compared to nondepleted mice.These results suggest that �-GalCer/CD1d� NK1.1� T cells inthe L. monocytogenes-infected liver, which produce IFN-�, arepreferentially CD4�8�.

J�18�/� mice are more resistant than control animals to L.monocytogenes infection. IFN-� is essential for protectionagainst L. monocytogenes infection (1, 2, 26, 27, 41, 45, 54, 55,

FIG. 5. Appearance of �-GalCer/CD1d� T cells in the L. monocy-togenes-infected liver of ATX mice. Mice were infected with L. mono-cytogenes on day 0, and HLs were prepared on days 0 and 4 p.i. Cellswere stained as described in Fig. 1. The data represent absolute num-bers of �-GalCer/CD1d� NK1.1� T cells (f) and �-GalCer/CD1d�

NK1.1� T cells (�) and are expressed as the means of three mice ateach time point. �, P 0.01 (day 0 versus day 4).

FIG. 6. Cell surface phenotype of �-GalCer/CD1d� T cells in the liver of C57BL/6 mice before and after L. monocytogenes infection. Mice wereinfected with L. monocytogenes, and HLs were prepared on days 0 and 4 p.i. Cells were stained with PE-labeled �-GalCer/CD1d tetramers andthe FITC-conjugated or biotinylated MAbs indicated in the figure, followed by SA-conjugated CyChrome. The data are expressed as histogramsafter gating on �-GalCer/CD1d� T cells. Representative staining patterns from three to six mice are shown.

FIG. 7. CD4 surface expression on �-GalCer/CD1d� T cells in theliver of C57BL/6 mice before and after L. monocytogenes infection andinfluence of CD4� or NK1.1� cell depletion on IFN-� and IL-4 produc-tion by �-GalCer/CD1d� T cells. (A) Mice were infected with L. mono-cytogenes, and HLs were prepared on days 0 and 4 p.i. Cells were stainedwith FITC-conjugated anti-CD4 MAb, biotinylated anti-NK1.1 MAb, andPE-labeled �-GalCer/CD1d tetramers, followed by SA-conjugated Cy-Chrome. The data of NK1.1 versus �-GalCer/CD1d tetramers are ex-pressed as dot plots after gating on lymphoid cells. Profiles of CD4 areexpressed as histograms after gating on �-GalCer/CD1d� T cells. Repre-sentative staining patterns from five mice per group are shown. (B) Micewere treated with anti-CD4 MAb, anti-NK1.1 MAb, or PBS on days �4and �2 and infected with L. monocytogenes on day 0, and HLs wereprepared on day 0 and/or day 4. Cells were stained with PE-labeled�-GalCer/CD1d tetramers, followed by anti-PE microbeads. �-GalCer/CD1d� cells were positively sorted by magnetic cell sorter and cultured inthe presence or absence of PMA and ionomycin on ELISPOT platescoated with either anti-IFN-� MAb or anti-IL-4 MAb. The data areexpressed as means of duplicate cultures (standard deviation of 10%).Experiments were performed three times with comparable results. �, Inthe presence of PMA and ionomycin; �, in the absence of PMA andionomycin. Because virtually all �-GalCer/CD1d� T cells in the liver onday 4 p.i. lacked NK1.1, anti-NK1.1 MAb treatment was not performed ininfected mice. Because no significant difference was found in the anti-human transferrin receptor MAb (isotype-matched MAb for anti-NK1.1MAb)-treated, rat IgG2b MAb (isotype-matched MAb for anti-CD4MAb)-treated, and PBS-treated groups in another experiment, PBS wasused as a control.

5910 EMOTO ET AL. INFECT. IMMUN.

on July 12, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

57), whereas IL-4 exacerbates the disease (22, 50, 53, 56).Because high numbers of IFN-�-secreting cells were detectedin the livers of L. monocytogenes-infected mice, we determinedwhether V�14� T cells participate in resistance against L.monocytogenes infection. To this end, J�18�/� and J�18�/�

mice were infected with L. monocytogenes, and bacterial bur-dens were determined on day 4 p.i. Listerial CFU in both liversand spleens were significantly lower in J�18�/� mice than inJ�18�/� mice (Fig. 8). Thus, J�18�/� mice were more resistantto L. monocytogenes infection than heterozygous littermates.These findings suggest that �-GalCer/CD1d� T cells comprisea heterogeneous population, which in its entirety does notcontribute to antilisterial protection and may even exacerbatedisease. We consider it likely that the IFN-�-producingNK1.1� subset of �-GalCer/CD1d� T cells ameliorates, whilethe IL-4-producing NK1.1� subset exacerbates, disease.

DISCUSSION

Our findings reveal the disappearance of the IL-4-producingNK1.1� subset of �-GalCer/CD1d� T cells and the appear-ance of IFN-�-producing NK1.1� subset in the livers of L.monocytogenes-infected mice. Thus, NK1.1 surface expressionand functional activities of glycolipid presented by CD1d-spe-cific V�14� T cells are markedly and rapidly influenced bylisterial infection.

Similar kinetics of �-GalCer/CD1d� T cells were seen inV�a mice, in which V�14� T cells express TCRV�7 orTCRV�2, but not TCRV�8 (42). Moreover, most �-GalCer/CD1d� NK1.1� T cells in the L. monocytogenes-infected liverexpressed TCRV�8, and only few expressed TCRV�7 andTCRV�2 (I. Yoshizawa and M. Emoto, unpublished results).We conclude that the compression of the �-GalCer/CD1d�

NK1.1� T-cell population and the subsequent expansion ofNK1.1� subpopulations occur independently of TCRV� usage.

IL-15 plays a central role in the proliferation of both NKcells and NKT cells (30, 33, 43). Recently, it was shown that the

depletion of NK cells by anti-ASGM1 antibody treatment pro-motes the expansion of thymus-derived V�14� T cells in theliver due to increased local concentrations of IL-15 (35). Be-cause NK cells are numerically increased in the liver of miceduring L. monocytogenes infection (53), we consider it likelythat similar mechanisms underlie the numerical increase of�-GalCer/CD1d� NK1.1� T cells after NK cell depletion.

The NK1.1� subset in the periphery was recently found to bederived from a thymic NK1.1� subpopulation of �-GalCer/CD1d� T cells (5). In that study, NK1.1 surface expression onV�14� T cells was acquired in the periphery after the cells hadleft the thymus during ontogeny. In our study, however, theNK1.1� subset that emerged in the L. monocytogenes-infectedliver developed from the NK1.1� subpopulation of �-GalCer/CD1d� T cells. Hence, the accumulation of V�14� T cells inthe liver is probably differentially regulated under physio-logical and under inflammatory conditions (10).

Expansion of V�14� T cells in response to �-GalCer in vivohas been reported (8, 25, 58). Because the numbers of NK cellswere increased after L. monocytogenes infection (53) and theexpansion of V�14� T cells was severely impaired in the pres-ence of NK cells (35, 46), it is unlikely that the �-GalCer/CD1d� NK1.1� T-cell population was expanded in situ. Anincrease of the �-GalCer/CD1d� NK1.1� T-cell populationduring listeriosis depended on IL-12. Hence, it is likely thatstimulation by a specific antigen (�-GalCer) and by the cyto-kine IL-12 have differential outcomes. Although we cannotformally exclude in situ expansion of V�14� T cells, we con-sider the following scenario most likely. After having left theBM, �-GalCer/CD1d� NK1.1� T cells encounter endogenousIL-12 at sites of inflammation, resulting in loss of the NK1.1surface marker.

CD4� NKT cells differ from CD4�8� NKT cells in their cyto-kine production profile (9, 21, 23, 24, 32, 52). Generally, CD4�

rather than CD4�8� NKT cells are responsible for IL-4 produc-tion, albeit at various levels in different organs. In the liver, CD4�

FIG. 7—Continued.

VOL. 74, 2006 CHANGES IN V�14� T CELLS DURING BACTERIAL INFECTION 5911

on July 12, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

cells dominate over CD4�8� cells among V�14� T cells (12, 15).Moreover, IL-4-producing cells among HLs are markedly re-duced by NK1.1� cell depletion (15). Finally, large numbers ofIL-4 producers are detected among purified liver CD4� NK1.1�

cells after TCR ligation (17). In our experiments, the numbers ofIL-4-producing �-GalCer/CD1d� T cells from CD4� or NK1.1�

cell-depleted mice were minute. Thus, it appears that under phys-iological conditions the CD4�NK1.1� subset is mainly responsi-ble for IL-4 production in the liver, although �-GalCer/CD1d� Tcells express both IL-4 and IFN-� mRNA upon stimulation (36,49). Differential stimulation, i.e., specific antigen versus IL-12, isprobably responsible for distinct cytokine production with inflam-mation driving IFN-� production by V�14� T cells (17, 18; thepresent study).

At first sight, the finding that listeriosis in V�14� T-cell-defi-cient mice was ameliorated could be taken as an argument againsta protective role of the V�14� T cells in listeriosis. However, asshown here, two subsets of V�14� T cells exist. (i) The first is theCD4�NK1.1� subset, which produces IL-4 and hence should beof no benefit or even be a detriment in listeriosis. At early stagesof infection, exacerbation by this T-cell population seems to dom-inate because depletion of the total V�14� T-cell populationameliorated listeriosis. This notion is consistent with previousfindings showing that anti-CD1 MAb treatment improves listeri-osis (51). (ii) The CD4� NK1.1� subset produces IFN-�, suggest-ing its beneficial role in listeriosis. The contribution of theNK1.1� subset to resistance, however, occurs later and seemssupportive but not essential, because conventional T cells canassume the burden of protection. These more subtle effects ofV�14� NK1.1� T cells could not be visualized in mouse mutantsdevoid of the entire V�14� T-cell population because geneknockout mutants only allow identification of essential pheno-typic features of gene deletion. However, recent studies revealeda contribution of V�14� T cells in protection against entericlisteriosis, suggesting a prevailing role of protective V�14� T cellsin mucosal defense (47).

In conclusion, we show here the influences of bacterial infec-tion on NK1.1 surface expression by V�14� T cells. AlthoughV�14� T cells produce both IFN-� and IL-4 in naive mice, themajority of this cell population produces IFN-�, but not IL-4,during listeriosis due to abundant IL-12 production. Therefore, itis tempting to assume that distinct V�14� T-cell populations play

unique roles in infection. Of these, the NK1.1� subset seemsineffectual or even harmful, whereas the NK1.1� subset appearsto contribute to antilisterial protection by means of IFN-�, thusprobably bridging the gap between early resistance and subse-quent acquired immunity.

ACKNOWLEDGMENTS

This study was supported by grants from the German Science Foun-dation (SFB 421); a Grant-in-Aid for Scientific Research (grant17590383) from the Japan Society for the Promotion of Science; TheWaksman Foundation of Japan, Inc.; and the Japan Research Foun-dation for Clinical Pharmacology.

We are grateful to M. Taniguchi, A. M. Livingstone, M. Kronenberg,and the Kirin Brewery Co., Ltd., for providing J�18�/� mice, V�a

mice, mouse CD1d/�2m-expressing baculovirus, and �-GalCer, re-spectively. We thank M. Staber for MAb purification and DanielaGroine-Triebkorn for the screening of mice.

REFERENCES

1. Bancroft, G. J., R. D. Schreiber, and E. R. Unanue. 1991. Natural immunity:a T-cell-independent pathway of macrophage activation, defined in the scidmouse. Immunol. Rev. 124:5–24.

2. Bancroft, G. J., R. D. Schreiber, G. C. Bosma, M. J. Bosma, and E. R.Unanue. 1987. A T cell-independent mechanism of macrophage activationby interferon-�. J. Immunol. 139:1104–1107.

3. Bendelac, A., M. N. Rivera, S. Park, and J. H. Roark. 1997. Mouse CD1-specific NK1 T cells: development, specificity, and function. Annu. Rev.Immunol. 15:535–562.

4. Benlagha, K., A. Weiss, A. Beavis, L. Teyton, and A. Bendelac. 2000. In vivoidentification of glycolipid antigen-specific T cells using fluorescent CD1dtetramer. J. Exp. Med. 191:1895–1903.

5. Benlagha, K., T. Kyin, A. Beavis, L. Teyton, and A. Bendelac. 2002. A thymicprecursor to the NK T-cell lineage. Science 296:553–555.

6. Chen, H., H. Huang, and W. E. Paul. 1997. NK1.1�CD4� T cells lose NK1.1expression upon in vitro activation. J. Immunol. 158:5112–5119.

7. Chen, Y., N. M. Chiu, M. Mandal, N. Wang, and C. Wang. 1997. ImpairedNK1� T-cell development and early IL-4 production in CD1-deficient mice.Immunity 6:459–467.

8. Crowe, N. Y., A. P. Uldrich, K. Kyparissoudis, K. J. L. Hammond, Y.Hayakawa, S. Sidobre, R. Keating, M. Kronenberg, M. J. Smyth, and D. I.Godfrey. 2003. Glycolipid antigen drives rapid expansion and sustained cy-tokine production by NK T cells. J. Immunol. 171:4020–4027.

9. Davodeau, F., M. A. Peyrat, R. A. Necker, F. R. Dominici, C. F. Blanchard,J. C. Leget, P. J. Gaschet, Y. P. Costa, A. Y. Jacques, H. A. Godard, A. H. Vie,F. A. Poggi, F. Romagne, and M. Bonneville. 1997. Close phenotypic andfunctional similarities between human and murine �� T cells expressinginvariant TCR �-chains. J. Immunol. 158:5603–5611.

10. Eberl, G., and H. R. MacDonald. 1998. Rapid death and regeneration ofNKT cells in anti-CD3-ε- or IL-12-treated mice: a major role for bonemarrow in NKT cell homeostasis. Immunity 9:345–353.

11. Emoto, M., J. Zerrahn, M. Miyamoto, B. Perarnau, and S. H. E. Kaufmann.2000. Phenotypic characterization of CD8�NKT cells. Eur. J. Immunol.30:2300–2311.

FIG. 8. Bacterial growth in livers of J�18�/� and J�18�/� mice after L. monocytogenes infection. Mice were infected with L. monocytogenes, and CFUcounts in livers and spleens were determined on day 4 p.i. The data are from five mice, and each symbol represents the CFU in an individual animal.Horizontal bars represent mean CFU. �, P 0.02 (J�18�/� versus J�18�/�). Experiments were performed twice with comparable results.

5912 EMOTO ET AL. INFECT. IMMUN.

on July 12, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

12. Emoto, M., and S. H. E. Kaufmann. 2003. Liver NKT cells: an account ofheterogeneity. Trends Immunol. 24:364–369.

13. Emoto, M., M. Miyamoto, Y. Emoto, J. Zerrahn, and S. H. E. Kaufmann.2001. A critical role of T-cell receptor �/� cells in antibacterial protection inmice early in life. Hepatology 33:887–893.

14. Emoto, M., M. Miyamoto, K. Namba, R. Schmits, N. van Rooijen, E. Kita,and S. H. E. Kaufmann. 2000. Participation of leukocyte function-associatedantigen-1 and NK cells in the homing of thymic CD8�NKT cells to the liver.Eur. J. Immunol. 30:3049–3056.

15. Emoto, M., Y. Emoto, and S. H. E. Kaufmann. IL-4 producing CD4�

TCR��int liver lymphocytes: influence of thymus, �2-microglobulin, andNK1.1 expression. 1995. Int. Immunol. 7:1729–1739.

16. Emoto, M., Y. Emoto, I. B. Buchwalow, and S. H. E. Kaufmann. 1999.Induction of IFN-�-producing CD4� natural killer T cells by Mycobacteriumbovis bacillus Calmette Guerin. Eur. J. Immunol. 29:650–659.

17. Emoto, M., Y. Emoto, and S. H. E. Kaufmann. 1995. Interleukin-4-producingCD4� NK1.1� TCR�/�intermediate liver lymphocytes are down-regulated byListeria monocytogenes. Eur. J. Immunol. 25:3321–3325.

18. Emoto, Y., M. Emoto, and S. H. E. Kaufmann,. 1997. Transient control ofinterleukin-4-producing natural killer T cells in the livers of Listeria mono-cytogenes-infected mice by interleukin-12. Infect. Immun. 65:5003–5009.

19. Fischer, K., E. Scotet, M. Niemeyer, H. Koebernick, J. Zerrahn, S. Maillet, R.Hurwitz, M. Kursar, M. Bonneville, S. H. E. Kaufmann, and U. E. Schaible.2004. Mycobacterial phosphatidylinositol mannoside is a natural antigen forCD1d-restricted T cells. Proc. Natl. Acad. Sci. USA 101:10685–10690.

20. Gregory, S. H., and E. J. Wing. 1998. Neutrophil-Kupffer cell interaction inhost defenses to systemic infections. Immunol. Today 19:507–510.

21. Gumperz, J. E., S. Miyake, T. Yamamura, and M. B. Brenner. 2002. Func-tionally distinct subsets of CD1d-restricted natural killer T cells revealed byCD1d tetramer staining. J. Exp. Med. 195:625–636.

22. Haak-Frendscho, M., J. F. Brown, Y. Iizawa, R. D. Wagner, and C. J.Czuprynski. 1992. Administration of anti-IL-4 monoclonal antibody 11B11increases the resistance of mice to Listeria monocytogenes infection. J. Im-munol. 148:3978–3985.

23. Hameg, A., I. Apostolou, M. Leite-De-Moraes, J. M. Gombert, C. Garcia, Y.Koezuka, J. F. Bach, and A. Herbelin. 2000. A subset of NKT cells that lacksthe NK1.1 marker, expresses CD1d molecules, and autopresents the �-galac-tosylceramide antigen. J. Immunol. 165:4917–4926.

24. Hammond, K. J., S. B. Pelikan, N. Y. Crowe, E. Randle-Barrett, T.Nakayama, M. Taniguchi, M. J. Smyth, I. R. van Driel, R. Scollay, A. G.Baxter, and D. I. Godfrey. 1999. NKT cells are phenotypically and function-ally diverse. Eur. J. Immunol. 11:3768–3781.

25. Harada, M., K. Seino, H. Wakao, S. Sakata, Y. Ishizuka, T. Ito, S. Kojo, T.Nakayama, and M. Taniguchi. 2004. Down-regulation of the invariant V�14antigen receptor in NKT cells upon activation. Int. Immunol. 16:241–247.

26. Holmberg, L. A., and K. A. Ault. 1986. Characterization of Listeria monocy-togenes-induced murine natural killer cells. Immunol. Res. 5:50–60.

27. Hsieh, C. S., S. E. Macatonia, C. S. Tripp, S. F. Wolf, A. O’Garra, and K. M.Murphy. 1993. Development of TH1 CD4� T cells through IL-12 producedby Listeria-induced macrophages. Science 260:547–549.

28. Kaufmann, S. H. E. 2003. Immunity to intracellular bacteria, p. 1229–1261.In W. E. Paul (ed.), Fundamental immunology, 5th ed. Lippincott-RavenPublishers, Philadelphia, Pa.

29. Kawano, T., J. Cui, Y. Koezuka, I. Toura, Y. Kaneko, K. Motoki, H. Ueno, R.Nakagawa, H. Sato, E. Kondo, H. Koseki, and M. Taniguchi. 1997. CD1d-restricted and TCR-mediated activation of V�14NKT cells by glycosylcer-amides. Science 278:1626–1629.

30. Kennedy, M. K., M. Glaccum, S. N. Brown, E. A. Butz, J. L. Viney, M.Embers, N. Matsuki, K. Charrier, L. Sedger, C. R. Willis, K. Brasel, P. J.Morrissey, K. Stocking, J. C. Schuh, S. Joyce, and J. J. Peschon. 2000.Reversible defects in natural killer and memory CD8 T-cell lineages ininterleukin 15-deficient mice. J. Exp. Med. 191:771–780.

31. Kinjo, Y., D. Wu, G. Kim, G. W. Xing, M. A. Poles, D. D. Ho, M. Tsuji, K.Kawahara, C. H. Wong, and M. Kronenberg. 2005. Recognition of bacterialglycosphingolipids by natural killer T cells. Nature 434:520–525.

32. Lee, P. T., K. Benlagha, L. Teyton, and A. Bendelac. 2002. Distinct functionallineages of human V�24 natural killer T cells. J. Exp. Med. 195:637–641.

33. Lodolce, J. P., D. L. Boone, S. Chai, R. E. Swain, T. Dassopoulos, S. Trettin,and A. Ma. 1998. IL-15 receptor maintains lymphoid homeostasis by sup-porting lymphocyte homing and proliferation. Immunity 9:669–676.

34. Mackaness, G. P. 1962. Cellular resistance to infection. J. Exp. Med. 116:381–406.

35. 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.

36. Matsuda, J. L., L. Gapin, J. L. Baron, S. Sidobre, D. B. Stetson, M. Mohrs,R. M. Locksley, and M. Kronenberg. 2003. Mouse V�14i natural killer T

cells are resistant to cytokine polarization in vivo. Proc. Natl. Acad. Sci. USA100:8395–8400.

37. Matsuda, J. L., O. V. Naidenko, L. Gapin, T. Nakayama, M. Taniguchi, C. R.Wang, Y. Koezuka, and M. Kronenberg. 2000. Tracking the response ofnatural killer T cells to glycolipid antigens using CD1d tetramer. J. Exp.Med. 192:741–753.

38. Mattner, J., K. L. Debord, N. Ismail, R. D. Goff, C. III Cantu, D. Zhou, P.Saint-Mezard, V. Wang, Y. Gao, N. Yin, K. Hoebe, O. Schneewind, D.Walker, B. Beutler, L. Teyton, P. B. Savage, and A. Bendelac. 2005. Exoge-nous and endogenous glycolipid antigens activate NKT cells during microbialinfections. Nature 434:525–529.

39. Mendiratta, S. K., W. D. Martin, S. Hong, A. Boesteanu, S. Joyce, and L. vanKaer. 1997. CD1d1 mutant mice are deficient in natural T cells that promptlyproduce IL-4. Immunity 6:469–477.

40. Miyamoto, M., M. Emoto, V. Brinkmann, N. van Rooijen, R. Schmits, E.Kita, and S. H. E. Kaufmann. 2000. Contribution of NK cells to the homingof thymic CD4�NKT cells to the liver. J. Immunol. 165:1729–1732.

41. Nakane, A., A. Numata, M. Asano, M. Kohanawa, Y. Chen, and T. Mina-gawa. 1990. Evidence that endogenous � interferon is produced early inListeria monocytogenes infection. Infect. Immun. 58:2386–2388.

42. Ohteki, T., and H. R. MacDonald. 1996. Stringent V� requirement for thedevelopment of NK1.1� T-cell receptor-�/�� cells in mouse liver. J. Exp.Med. 183:1277–1282.

43. Ohteki, T., H. Shirley, H. Suzuki, T. W. Mak, and P. S. Ohashi. 1997. Rolefor IL-15/IL-15R� chain in NK1.1� T-cell receptor-��� cell development.J. Immunol. 159:5931–5935.

44. Park, S., T. Kyin, A. Bendelac, and C. Carnaud. 2003. The contribution ofNKT cells, NK cells, and other �-chain-dependent non-T non-B cells toIL-12-mediated rejection of tumors. J. Immunol. 170:1197–1201.

45. Poston, R. M., and R. J. Kurlander. 1991. Analysis of the time course ofIFN-� mRNA and protein production during primary murine listeriosis. Theimmune phase of bacterial elimination is not temporally linked to IFNproduction in vivo. J. Immunol. 146:4333–4337.

46. Ranson, T., C. A. Vosshenrich, E. Corcuff, O. Richard, V. Laloux, A.Lehuen, and J. P. Di Santo. 2003. IL-15 availability conditions homeostasis ofperipheral natural killer T cells. Proc. Natl. Acad. Sci. USA 100:2663–2668.

47. Ranson, T., S. Bregenholt, A. Lehuen, O. Gaillot, M. C. Leite-de-Moraes, A.Herbelin, P. Berche, and J. P. Di Santo. 2005. Invariant V�14� NKT cellsparticipate in the early response to enteric Listeria monocytogenes infection.J. Immunol. 175:1137–1144.

48. Smiley, S. T., M. H. Kaplan, and M. J. Grusby. 1997. Immunoglobulin Eproduction in the absence of interleukin-4-secreting CD1-dependent cells.Science 275:977–979.

49. Stetson, D. B., M. Mohrs, R. L. Reinhardt, J. L. Baron, Z. E. Wang, L.Gapin, M. Kronenberg, and R. M. Locksley. 2003. Constitutive cytokinemRNAs mark natural killer (NK) and NKT cells poised for rapid effectorfunction. J. Exp. Med. 198:1069–1076.

50. Szalay, G., C. H. Ladel, C. Blum, and S. H. E. Kaufmann. 1996. IL-4neutralization or TNF-� treatment ameliorate disease by an intracellularpathogen in IFN-� receptor-deficient mice. J. Immunol. 157:4746–4750.

51. Szalay, G., C. H. Ladel, C. Blum, L. Brossay, M. Kronenberg, and S. H. E.Kaufmann. 1999. Anti-CD1 monoclonal antibody treatment reverses theproduction patterns of TGF-�2 and Th1 cytokines and ameliorates listeriosisin mice. J. Immunol. 162:6955–6958.

52. Takahashi, T., M. Nieda, Y. Koezuka, A. Nicol, S. A. Porcelli, Y. Ishikawa, K.Tadokoro, H. Hirai, and T. Juji. 2000. Analysis of human V�24�CD4�NKTcells activated by �-glycosylceramide-pulsed monocyte-derived dendriticcells. J. Immunol. 164:4458–4464.

53. Teixeira, H. C., and S. H. E. Kaufmann. 1994. Role of NK1.1� cells inexperimental listeriosis. NK1� cells are early IFN-� producers but impairresistance to Listeria monocytogenes infection. J. Immunol. 152:1873–1882.

54. Tripp, C. S., M. K. Gately, J. Hakimi, P. Ling, and E. R. Unanue. 1994.Neutralization of IL-12 decreases resistance to Listeria in SCID and C.B-17mice. Reversal by IFN-�. J. Immunol. 152:1883–1887.

55. Tripp, C. S., S. F. Wolf, and E. R. Unanue. 1993. Interleukin 12 and tumornecrosis factor � are costimulators of interferon � production by naturalkiller cells in severe combined immunodeficiency mice with listeriosis, andinterleukin 10 is a physiologic antagonist. Proc. Natl. Acad. Sci. USA 90:3725–3779.

56. Wagner, R. D., and C. J. Czuprynski. 1993. Cytokine mRNA expression in liversof mice infected with Listeria monocytogenes. J. Leukoc. Biol. 53:525–531.

57. Wagner, R. D., H. Steinberg, J. F. Brown, and C. J. Czuprynski. 1994.Recombinant interleukin-12 enhances resistance of mice to Listeria mono-cytogenes infection. Microb. Pathog. 17:175–186.

58. Wilson, M. T., C. Johansson, D. Olivares-Villagomez, A. K. Singh, A. K. Stanic, C.Wang, S. Joyce, M. J. Wick, and L. van Kaer. 2003. The response of natural killerT cells to glycolipid antigens is characterized by surface receptor down-modulationand expansion. Proc. Natl. Acad. Sci. USA 100:10913–10918.

Editor: J. L. Flynn

VOL. 74, 2006 CHANGES IN V�14� T CELLS DURING BACTERIAL INFECTION 5913

on July 12, 2020 by guesthttp://iai.asm

.org/D

ownloaded from