Experimental Hematology 34 (2006) 289–295

Analysis of the effect of different NKT cell subpopulationson the activation of CD4 and CD8 T cells, NK cells, and B cells

Henry Lina, Mie Niedab, Vladislav Rozenkova, and Andrew J. Nicola

aDepartment of Medicine, University of Queensland, Brisbane, Australia; bSchool of Medicine, Yokohama City University, Yokohama, Japan

(Received 14 October 2005; revised 15 December 2005; accepted 16 December 2005)

Objective. NKT cells have diverse immune regulatory functions including activation of cellsinvolved in Th1- and Th2-type immune activities. Most previous studies have investigatedthe functions of NKT cells as a single family but more recent evidence indicates the distinctfunctional properties of NKT cell subpopulation. This study aims to determine whetherNKT cell subpopulations have different stimulatory activities on other immune cells thatmay affect the outcome of NKT cell–based immunotherapy.

Methods. NKT cells and NKT cell subpopulations (CD4+CD82, CD42CD8+, CD42CD82)were cocultured with PBMC and their activities on immune cells including CD4+ and CD8+

T cells, NK cells, and B cells were assessed by flow cytometry. The production of cytokinesin culture was measured by enzyme-linked immunsorbent assay.

Results. The CD4+CD82 NKT cells demonstrated substantially greater stimulatory activitieson CD4+ T cells, NK cells, and B cells than other NKT cell subsets. The CD42CD8+ NKT cellsshowed the greatest activity on CD8+ T cells, and were the only NKT cell subset that activatedthese immune cells. The CD42CD82 NKT cells showed moderate stimulatory activity onCD4+ T cells and the least activity on other immune cells.

Conclusion. The results here suggest that NKT cell subpopulations differ in their abilities tostimulate other immune cells. This highlights the potential importance of manipulatingspecific NKT cell subpopulations for particular therapeutic situations and of evaluatingsubpopulations, rather than NKT cells as a group, during investigation of a possible role ofNKT cells in various disease settings. � 2006 International Society for Experimental Hem-atology. Published by Elsevier Inc.

Natural killer T (NKT) cells are a unique lymphocyte line-age characterized by the expression of an invariant T cellreceptor (Va24JaQ paired with Vb11 in human) that recog-nizes glycolipids presented by CD1d molecules. Uponstimulation, NKT cells as a family have the ability to pro-duce large amounts of both T-helper (Th)-1 and Th2-typecytokines [1]. The production of these cytokines by NKTcells have been demonstrated to play a significant role inthe modulation of other immune effector cells includingCD4D and CD8D T cells, NK cells, and B cells involvedin Th1- or Th2-type immune activities [2–8]. One of the in-teresting properties of the NKT cell population is its appar-ently paradoxical functions in different diseases including

Offprint requests to: Henry Lin, Ph.D., Department of Medicine, Univer-

sity of Queensland, Centre for Immune and Targeted Therapy, Greenslopes

Private Hospital, Newdegate St., Greenslopes, Australia; E-mail: hlin@

soms.uq.edu.au

0301-472X/06 $–see front matter. Copyright � 2006 International Society fo

doi: 10.1016/j.exphem.2005.12.008

malignancies and autoimmune dysfunctions. NKT cell ac-tivities can be either beneficial [9–13] or detrimental [14–16] in different settings and this is likely to be related tothe balance between Th1- or Th2-biased immune activitiesmediated by NKT cells.

The diverse immune activities of NKT cells could resultfrom differences between the regulation or tissue distribu-tion (including a response to disease) of the differentNKT cell subpopulations, characterized according to theirexpression of CD4 or CD8 surface molecules. Recent stud-ies have highlighted the distinct Th1- and Th2-type cyto-kine profiles of NKT cell subpopulations [17–21]. TheCD4DCD82 NKT cells (CD4 NKT cells) produce bothTh1- and Th2-type cytokines [18–21] and the CD42CD8D

(CD8 NKT cells) and CD42CD82 NKT cells (double-negative, DN NKT cells) produce predominantly Th1-type cytokines [17,19–21]. The different cytokine profilesof NKT cell subpopulations suggest that these cells are

r Experimental Hematology. Published by Elsevier Inc.

290 H. Lin et al. / Experimental Hematology 34 (2006) 289–295

likely to have different actions on different immune effectorcells important in Th1- or Th2-type immune activities. Thismay contribute to the reported paradoxical functions ofNKT cells.

Interest in the potential to manipulate NKT cells fortherapeutic purposes increased since the demonstrationthat NKT cells are stimulated, markedly expanded, andfunctionally altered, for example resulting in secretion oflarge amounts of both Th1- and Th2-type cytokines [1,9],in response to the glycolipid, a-galactosylceramide (a-GalCer) [22–25]. The development of multiple glycolipidanalogues with potentially different capacities to stimulateNKT cell subsets strengthens the prospect of manipulatingNKT cells for therapeutic benefit [26,27]. Depending on theextent to which NKT cell subpopulations differ, it is likelythat nonselective activation of all NKT cells could result inunwanted immunological outcomes for the intended thera-peutic use. For example, CD4 NKT cells have been shownto suppress anti-tumor responses [15,16] and activation ofthis NKT cell subset could be detrimental for anti-tumortherapy. We have previously shown that NKT cell subpop-ulations can be expanded in vitro and that subpopulations ofNKT cells can be manipulated by altering the cytokine en-vironment upon which NKT cells are stimulated [28]. Agreater understanding of the differences between the effectsof NKT cell subpopulations on other immune cells is nec-essary to optimize the potential for therapeutic benefit to bederived from NKT cell manipulation. In this study, we com-pared the stimulatory activities of NKT cell subpopulationson CD4D and CD8D T cells, NK cells, and B cells involvedin different Th1- or Th2-type immune activities.

Materials and methods

Study populationPeripheral blood was collected with informed consent fromhealthy donors (n 5 3) by the Australian Red Cross Blood Service.Blood from one donor was used to generate allogenic monocyte-derived dendritic cells (Mo-DC). In order to obtain sufficient NKTcells, donors were selectively chosen based on their having periph-eral blood NKT cell levels greater than 0.03% of all T cells. Thestudy was approved by the Medical Research Ethics Committee ofthe University of Queensland.

Antibodies used in flow cytometryThe fluorochrome-conjugated monoclonal antibodies used in flowcytometry included: fluorescein isothiocyanate (FITC)-labeled an-tibodies against Va24, CD3, CD8, CD19, CD1d 42.1, interferon-g(IFN-g); phycoerythrin (PE)-labeled antibodies against Vb11,CD3, CD4, CD40, CD56, CD86, IL-4; phycoerythrin cyanine 5(PC5)-labeled antibodies against CD3, CD8, CD14, CD25; Texasred ethyl cysteinate dimer (ECD)-labeled antibodies against CD4,CD69; phycoerythrin cyanine 7 (PC7)-labeled CD3; and the rele-vant isotype controls (Beckman Coulter, USA). The CD69 andCD25 antibodies were used as early and late activation markers,

respectively, to determine the activation status of the stimulatedcell populations.

Isolation of peripheral bloodmononuclear cells and generation of NKT cell linesPeripheral blood mononuclear cells (PBMC) were obtained bydensity-gradient centrifugation using Ficoll Paque Plus (AmershamBiosciences, Sweden). For expansion of NKT cells, PBMC werecultured in complete media containing AIM-V medium (GibcoBRL, Australia) supplemented with 10% fetal calf serum (FCS)(JRH Biosciences, Australia), 100 ng/mL of a-GalCer (Pharmaceu-tical Division, Kirin Brewery, Japan), and 10 ng/mL each of inter-leukin (IL)-7 and IL-15 (both from R&D Systems, USA) for 7days. On day 7, purified NKT cells were obtained by magnetic-ac-tivated cell sorting (MACS) of Va24D cells using Va24 monoclonalantibodies (Beckman Coulter, USA) and rat anti-mouse IgG1microbeads (Miltenyi Biotec, Germany). The purity of NKT cellsobtained after magnetic sorting was confirmed using antibodiesagainst Va24, Vb11, and CD3. The purified NKT cells were stimu-lated with allogenic Mo-DC at a NKT:DC ratio of 10:1 for a further 7days and then cryopreserved until functional assessment. AllogenicMo-DC were obtained from PBMC of a normal donor by 1-houradherence in a tissue culture flask followed by culture of the adheredmonocytes in complete media with 500 U/mL of recombinanthuman IL-4 (R&D Systems, USA) and 400 U/mL of recombinanthuman granulocyte-macrophage colony-stimulating factor (GM-CSF) (Schering-Plough, Australia) for 5 days. On day 4, 100 ng/mL of a-GalCer was added to the culture to prime Mo-DC forspecific stimulation of NKT cells. The Mo-DC were confirmed aslineage-negative cells using CD3, CD19, and CD14 antibodies.The phenotype of the Mo-DC was assessed using antibodies againstCD1d (a gift from Dr. Steve Porcelli), CD40, and CD86. The allo-genic Mo-DC were irradiated at 30 Gy before use.

Intracellular cytokine assessment of NKT cell subpopulationsIntracellular IFN-g and IL-4 expression of NKT cell subpopula-tions was assessed by flow cytometry using purified NKT cells.Briefly, NKT cells were stimulated with 20 ng/mL of Phorbol-12-myristate-13-acetate (PMA) and 2 mg/mL of ionomycin (allfrom Sigma) for 4 hours. Brefeldin A (BFA) (Sigma) was addedto the cells at the concentration of 20 mg/mL to prevent cytokineexcretion. Cells were labeled with antibodies against CD4 andCD8 and then fixed and permeabilized using the IntraPrep Perme-abilization Reagent (Beckman Coulter) according to the manufac-turer’s specifications. The permeabilized cells were labeled withantibodies against IFN-g and IL-4 and then analyzed on theflow cytometer (FC500, Beckman Coulter).

Isolation of NKT cell subpopulationsCryopreserved NKT cells were thawed and then stimulated witha-GalCer-pulsed allogenic Mo-DC in the presence of 10 ng/mLof IL-7 and IL-15 for 2 days and used as the stimulator cells.NKT cell subpopulations were isolated from the activated NKTcell population by MACS using the CD4 and CD8 microbeads(Miltenyi Biotec). NKT cells were first stained with CD4 microbe-ads and passed through a LD column (Miltenyi Biotec) to obtainthe CD4D and CD42 NKT cell fractions. The CD42 cells werethen stained with CD8 microbeads and passed through a LD col-umn to obtain the CD42CD8D and CD42CD82 cells. The purityof the CD4D and CD8D cell fractions was further improved by

291H. Lin et al./ Experimental Hematology 34 (2006) 289–295

passing them through MS columns (Miltenyi Biotec). Three NKTcell subpopulations, the CD4DCD82, CD42CD8D, andCD42CD82 NKT cells, were isolated using the above proceduresand the purity of these cells was greater than 95% as determinedby flow cytometry.

Stimulation of CD4D and CD8D T cells,NK cells, and B cells by activated NKT cellsActivated NKT cells were cocultured with autologous PBMC induplicates at a NKT-to-PBMC ratio of 1:5 for 5 days. Culture con-taining PBMC without coculture with NKT cells was used as thenegative control. The autologous PBMC was used as the respondercell population for assessment of CD4D and CD8D T cells, NKcells, and B cells using 5-color flow cytometry (FC500, BeckmanCoulter). Analysis of the activation status of NKT cells and theresponder cells was performed on days 1, 3, and 5 of culture todetermine the period of greatest stimulatory activities. To comparethe stimulatory activities of NKT cell subpopulations on theresponder cells, isolated NKT subpopulations were cultured withautologous PBMC at a NKT-to-PBMC ratio of 1:5 and the activa-tion status of the responder cells was assessed by flow cytometryon day 1 (the period of maximal stimulatory activities determinedfrom the above).

Assessment of cytokine production in culture environmentsCulture supernatants were collected on day 1 of the culture (thedetermined period of maximal stimulatory activities) and then fro-zen at 280�C until cytokine assessment. The production of IFN-gand IL-4 was assessed using the Human IFN-g and IL-4 OptEIAELISA sets (BD Biosciences, USA), according to the manufac-turer’s protocols.

Results

Response of NKT cells to stimulation with a-GalCerStimulation of PBMC with a-GalCer and cytokines IL-7and IL-15 resulted in an average 200-fold proliferation ofNKT cells (n 5 3) after 7 days. Further expansion ofNKT cells was achieved using a-GalCer-pulsed allogenicMo-DC and a minimum of 1 3 108 NKT cells was ob-tained. Analysis of cytokine expression profiles of NKT

cell subpopulations showed that all NKT cell subpopula-tions expressed IFN-g but that the CD4 NKT cell popula-tion was the only subset that expressed significant levelsof IL-4 (Fig. 1).

Dynamics of CD4D and CD8D T-cell, NK-cell,and B-cell activation following stimulation by NKT cellsAssessment of the activation dynamics of various immunecells stimulated by NKT cells was performed to determinethe period of greatest stimulatory activities. It was observedthat the expression of CD69 and CD25 on CD4D T cellspeaked on day 1 following NKT cell stimulation and thendeclined throughout the remaining culture period(Fig. 2A). The peak of CD8D T-cell activation, character-ized by the expression of CD25, occurred between day 1and day 3 (Fig. 2B). In contrast, the CD69 expression onCD8D T cells did not increase. The greatest activationof NK cells, characterized by upregulated expression ofCD69, occurred on day 1; however the expression ofCD25 showed no significant change from the day-0 baselinelevel (Fig. 2C). Activation of B cells, observed by the upre-gulated expression of CD69 and CD25, peaked on day 1following NKT cell stimulation and then gradually declined(Fig. 2D). The expression of CD69 and CD25 on all re-sponder cells was higher than the negative control afterstimulation with NKT cells throughout the culture period.Collectively, these results show that NKT cells have theability to stimulate all responder cells including CD4D

and CD8D T cells, NK cells, and B cells. In most cases,the peak of activation of these immune cells occurred onday 1 following stimulation with NKT cells. Based on theseobservations, comparison of stimulatory activities betweenNKT cell subpopulations was performed on day 1 follow-ing culture.

Activation of CD4D T cells by NKT cell subpopulationsSubstantial increases in CD69 and CD25 expression onCD4 T cells were only induced by CD4 and DN NKT cells(Fig. 3). In contrast, CD8 NKT cells did not increase the

Figure 1. Representative plots showing the cytokine profiles of the cultured NKT cells after stimulation of the magnetically isolated Va24D cells. Va24D

NKT cells were labeled with CD4, CD8, intracellular IFN-g, and IL-4 monoclonal antibodies. NKT cell subpopulations were identified by the expression of

CD4 and CD8 surface molecules. IFN-g and IL-4 expression of (A) CD4 NKT, (B) CD8 NKT, and (C) DN NKT cells are shown. CD4 NKT cells produce

both IFN-g and IL-4 while CD8 and DN NKT cells produce predominantly IFN-g.

292 H. Lin et al. / Experimental Hematology 34 (2006) 289–295

Figure 2. Activation status of various immune cells following stimulation with NKT cells. Dynamics of CD4 (A) and CD8 T-cell (B), NK-cell (C), and

B-cell (D) activation characterized by the mean expression of CD69 and CD25 from day 0 to day 5 (n 5 3).

expression of either CD69 or CD25 on CD4 T cells. CD4NKT cells demonstrated a substantially greater capacityto upregulate CD69 and CD25 expression on CD4 T cellscompared with other NKT cell subsets.

Activation of CD8D T cells by NKT cell subpopulationsSince stimulation of CD8D T cells using whole NKT cellsonly increased the expression of CD25 but not CD69(Fig. 2B), the stimulatory activities of NKT cell subpopula-tions on CD8D T cells were compared by evaluating the ex-pression of CD25. It was found that the CD8 NKT cell wasthe only NKT subset that induced substantial upregulationof CD25 on CD8D T cells (Fig. 4).

Activation of NK cells by NKT cell subpopulationsSince stimulation of NK cells using whole NKT cells onlyincreased the expression of CD25 but not CD69 (Fig. 2C),

Figure 3. Activation status of CD4 T cells following 24 hours stimulation

with NKT cell subpopulations. The graph shows the mean 6 SEM of the

CD69 and CD25 expression on CD4 T cells in the different culture condi-

tions specified (n 5 3).

the stimulatory activities of NKT cell subpopulations onNK cells were compared by evaluating the expression ofCD69. All NKT cell subpopulations increased the expres-sion of CD69 on NK cells (Fig. 5). CD4 NKT cells showedsubstantially greater capacity to upregulate NK cell CD69expression than the CD8 and DN NKT cells.

Activation of B cells by NKT cell subpopulationsAll NKT cell subpopulations stimulated a substantial in-crease in the expression of CD69 and CD25 on B cells(Fig. 6). The ability of CD4 NKT cells to upregualte B-cellCD69 expression was substantially greater than DN NKTcells and to a less extent also when compared with CD8NKT cells. In addition, the CD4 NKT cells demonstratedsubstantially greater capacity to upregulate B-cell CD25expression compared with both CD8 and DN NKT cells.

Figure 4. Activation status of CD8 T cells following 24 hours stimulation

with NKT cell subpopulations. The graph shows the mean 6 SEM of the

CD25 expression on CD8 T cells in the different culture conditions spec-

ified (n 5 3).

293H. Lin et al./ Experimental Hematology 34 (2006) 289–295

Th1- and Th2-type cytokine productionafter stimulation of PBMC with NKT cellsProduction of IFN-g and IL-4 was detected in the culturesupernatant after coculture of PBMC with activated NKTcells (Fig. 7). In culture conditions involving isolatedNKT cell subpopulations, the greatest production of IFN-gand IL-4 was observed in cultures containing the CD4NKT cells. The production of IL-4 was significantly lowerin cultures containing the CD8 and DN NKT cell subsets.

DiscussionThe data described here confirm that activated human NKTcells rapidly produce secondary activation of other immunecells and further demonstrate that NKT cell subpopulations(CD4, CD8, and DN NKT cells) differ in their ability tostimulate different immune cells (CD4 T cells, CD8 T cells,NK cells, and B cells). This is the first study comparing thestimulatory activities of NKT cell subpopulations on otherimmune cells involved in both Th1- and Th2-type immu-nity. The capacity of NKT cell subpopulations to activateimmune cells was consistent with the cytokines producedin culture; however, additional pathways may also be in-volved. In most cases, NKT cell–mediated activation of im-mune cells was rapid and transient, with a peak of activitywithin 1 day and a decline in activity thereafter. All NKTcell subpopulations were able to activate immune cellswithin 24 hours of culture. The rapid but short-lived in vitroactivation of immune cells observed in this study follows

Figure 5. Activation status of NK cells following 24 hours stimulation

with NKT cell subpopulations. The graph shows the mean 6 SEM of

the CD69 expression on NK cells in the different culture conditions spec-

ified (n 5 3).

Figure 6. Activation status of B cells following 24 hours stimulation with

NKT cell subpopulations. The graph shows the mean 6 SEM of the CD69

and CD25 expression on B cells in the different culture conditions speci-

fied (n 5 3).

a time course very similar to what we have observed in clin-ical trials of a-GalCer-pulsed DC [29]. Secondary activa-tion of NK and T cells was maximal on the first dayfollowing administration of a-GalCer-pulsed DC andpeak production of cytokines, such as IFN-g, occurredwithin 24 hours with very rapid return to baseline levels af-ter the peak. The data presented here provides evidence thatall NKT cell subpopulations may contribute to these clini-cal observations, but that individual subpopulations arelikely to contribute to different extents. This data cannotreadily be obtained in clinical studies in which all NKTcell subpopulations are present and activated. An importantdifference between our observations and those that occur inphysiological settings is that we used a fixed ratio of NKTcells: responder cells. In vivo, there are likely to be differ-ences in the relative numbers of NKT cell subpopulationsand these differences may be situation dependent. For ex-ample, in the human liver, DN NKT cells were found atgreater frequency than CD4 NKT cells [30]. The potentialdifferences in the proportion of NKT cell subpopulationsin different tissue sites provide the possibility to further am-plify skewing of immune response by one or another NKTcell subpopulation.

We have observed that CD8D T-cell activation stimu-lated by NKT cells was characterized by upregulation ofCD25 but not CD69. The expression of CD25 on T cellsis associated with cell proliferation [31]. Our clinical trialinvolving administration of a-GalCer-pulsed DC has shownthat the number of CD8D T cells increases shortly aftervaccination ([29] and unpublished observations). The datahere suggest that the increase of CD8D T cell number couldinvolve upregulation of CD25 induced by NKT cells. Wedid not observe upregulation of CD69 on CD8D T cellsin response to NKT cell stimulation in contrast to the pre-vious murine data [32,33]. It is possible that the increase inCD69 expression on CD8 T cells was very short lived afterNKT cell stimulation as observed in mice [32,33] and thatour detection time points were too late.

Stimulatory activities of CD4 NKT cellsIt was observed that CD4 NKT cells induced the greatestactivation of NK cells, but have no effect on CD8D T cells,both of which are involved in Th1-type immunity. The CD4NKT cell subset also induced the greatest activation of Bcells, which is involved in Th2-type immunity, and CD4D

T cells, which could be involved in either Th1- or Th2-type immunity. The observed stimulatory activities ofCD4 NKT cells on NK cells, B cells, and CD4D T cellscould be associated with the IFN-g and IL-4 productionobserved in culture. This is supported by previous studiesshowing that NKT cell–mediated activation of NK cellsand B cells involves IFN-g [2,32] and IL-4 respectively[8]. Activation of CD4D T cells by NKT cells could involveeither IFN-g or IL-4 depending on the specific situations[6,7]. Collectively, these data provide direct evidence on

294 H. Lin et al. / Experimental Hematology 34 (2006) 289–295

Figure 7. Cytokine production after 24 hours stimulation of PBMC with NKT cells. Mean 6 SEM of the (A) production of IFN-g and (B) production of IL-4

detected in the different culture conditions specified (n 5 3).

the role of CD4 NKT cells in both Th1- and Th2-type im-munity through activation of different immune cells. Thisprovides important functional data which extend previoussubpopulation data that was largely based on cytokine ex-pression or surface marker expression. What remain to beinvestigated are the factors that determine how the CD4NKT cells participate in different situations depending onwhether a Th1- or Th2-biased immune activity (e.g., duringdifferent disease conditions) is desirable.

Stimulatory activities of CD8 NKT cellsOnly CD8 NKT cells induced activation of all immune cellsinvolved in Th1-type immunity (NK cells and CD8D Tcells) included in this study. The effect of this NKT cellsubset on immune cells involved in Th2-type immunitywas modest, involving some activation of B cells but no ef-fect on CD4D T cells. Activation of NK cells by CD8 NKTcells was consistent with the Th1-biased cytokine produc-tion (IFN-g) observed in culture. CD8 NKT cells had sub-stantially greater stimulatory activities on CD8 T cellscompared to other NKT cell subsets, while producing lessIFN-g. This suggests that the observed differential stimula-tory activity of CD8 NKT cells involves other unidentifiedpathways or mechanisms independent of IFN-g. In contrastto previous data suggesting that NKT-cell activation of Bcells is IL-4 dependent [8], we have observed stimulatoryactivities of B cells in culture containing CD8 NKT cellsin which levels of IL-4 were undetectable but with highlevels of IFN-g. These observations suggest that the stimu-latory activities of NKT cells on B cells can also be IL-4independent and may involve IFN-g. This possibility is

supported by a previous study demonstrating, in the murinesystem, the role of IFN-g on activation of B cells [34].

Stimulatory activities of DN NKT cellsOf the three NKT cell subpopulations examined, the DNNKT cells induced the least activation of NK cells andhad no effect on CD8D T cells. This NKT cell subsetalso showed lowest stimulatory activity on B cells, butmoderate stimulatory activity on CD4D T cells. The effectof DN NKT cells on the functions of CD4D T cells is un-clear. As DN NKT cells have a Th1-biased cytokine pro-duction profile, this NKT cell subset presumably supportthe differentiation of the CD4D T cells into Th1-type cells.In the culture environment, the DN NKT cells induced sim-ilar cytokine production patterns to CD8 NKT cells (highIFN-g production and undetectable levels of IL-4). How-ever, the stimulatory capacities of these NKT cell subsetson other immune cells are strikingly different. This impliesthat the commonly accepted view that DN and CD8 NKTcells are functionally alike, based on similar cytokine pro-duction profiles, may be overly simplistic. The differencesbetween NKT cell subpopulations, other than just cytokineproduction profiles, need to be further investigated to betterdelineate the functional differences between these cells andthe mechanisms controlling these differences.

We have provided the first in vitro functional data ad-dressing the different stimulatory activities of NKT cellsubpopulations on immune cells involved in both Th1-and Th2-type immune activities. The data described herehighlight the need to evaluate NKT cell subpopulations,rather than as a group, in investigation of the role of

295H. Lin et al./ Experimental Hematology 34 (2006) 289–295

NKT cells in health and disease. There are also importantimplications for the design of clinical studies aiming tomodify NKT cells for therapeutic benefits. Clinical trialsto investigate the therapeutic potential of NKT cells shouldprobably utilize strategies that activate or bias towardspecific NKT cell subpopulations rather than the NKTcell population as a whole.

References1. Yang OO, Racke FK, Nguyen PT, et al. CD1d on myeloid dendritic

cells stimulates cytokine secretion from and cytolytic activity of

Va24JaQ T cells: A feedback mechanism for immune regulation.

J Immunol. 2000;165:3756.

2. Smyth MJ, Crowe NY, Pellicci DG, et al. Sequential production of

interferon-g by NK1.1D T cells and natural killer cells is essential

for the antimetastatic effect of a-galactosylceramide. Blood. 2002;

99:1259.

3. Hayakawa Y, Takeda K, Yagita H, et al. Critical contribution of IFN-g

and NK cells, but not perforin-mediated cytotoxicity, to anti-meta-

static effect of a-galactosylceramide. Eur J Immunol. 2001;31:1720.

4. Metelitsa LS, Naidenko OV, Kant A, et al. Human NKT cells mediate

antitumor cytotoxicity directly by recognizing target cell CD1d with

bound ligand or indirectly by producing IL-2 to activate NK cells.

J Immunol. 2001;167:3114.

5. Yang YF, Tomura M, Ono S, Hamaoka T, Fujiwara H. Requirement

for IFN-g in IL-12 production induced by collaboration between

va14D NKT cells and antigen-presenting cells. Int Immunol. 2000;

12:1669.

6. Eberl G, Brawand P, MacDonald HR. Selective bystander proliferation

of memory CD4D and CD8D T cells upon NK T or T cell activation.

J Immunol. 2000;165:4305.

7. Singh N, Hong S, Scherer DC, et al. Cutting edge: activation of NK T

cells by CD1d and a-galactosylceramide directs conventional T cells

to the acquisition of a Th2 phenotype. J Immunol. 1999;163:2373.

8. Kitamura H, Ohta A, Sekimoto M, et al. a-galactosylceramide induces

early B-cell activation through IL-4 production by NKT cells. Cell

Immunol. 2000;199:37.

9. Kikuchi A, Nieda M, Schmidt C, et al. In vitro anti-tumour activity of

a-galactosylceramide-stimulated human invariant Va24D NKT cells

against melanoma. Br J Cancer. 2001;85:741.

10. Duarte N, Stenstrom M, Campino S, et al. Prevention of diabetes in

nonobese diabetic mice mediated by CD1d-restricted nonclassical

NKT cells. J Immunol. 2004;173:3112.

11. Sharif S, Arreaza GA, Zucker P, Mi QS, Delovitch TL. Regulation of

autoimmune disease by natural killer T cells. J Mol Med. 2002;80:290.

12. Kawano T, Cui J, Koezuka Y, et al. Natural killer–like nonspecific

tumor cell lysis mediated by specific ligand-activated Va14 NKT

cells. Proc Natl Acad Sci U S A. 1998;95:5690.

13. Kawano T, Nakayama T, Kamada N, et al. Antitumor cytotoxicity

mediated by ligand-activated human Va24 NKT cells. Cancer Res.

1999;59:5102.

14. Kim HY, Kim HJ, Min HS, et al. NKT cells promote antibody-induced

joint inflammation by suppressing transforming growth factor b1

production. J Exp Med. 2005;201:41.

15. Terabe M, Matsui S, Noben-Trauth N, et al. NKT cell–mediated re-

pression of tumor immunosurveillance by IL-13 and the IL-4R-

STAT6 pathway. Nat Immunol. 2000;1:515.

16. Moodycliffe AM, Nghiem D, Clydesdale G, Ullrich SE. Immune sup-

pression and skin cancer development: regulation by NKT cells. Nat

Immunol. 2000;1:521.

17. Takahashi T, Chiba S, Nieda M, et al. Cutting edge: analysis of human

V a 24D CD8D NK T cells activated by a-galactosylceramide-pulsed

monocyte-derived dendritic cells. J Immunol. 2002;168:3140.

18. Takahashi T, Nieda M, Koezuka Y, et al. Analysis of human V a 24D

CD4D NKT cells activated by a-glycosylceramide-pulsed monocyte-

derived dendritic cells. J Immunol. 2000;164:4458.

19. Gumperz JE, Miyake S, Yamamura T, Brenner MB. Functionally

distinct subsets of CD1d-restricted natural killer T cells revealed by

CD1d tetramer staining. J Exp Med. 2002;195:625.

20. Kim CH, Butcher EC, Johnston B. Distinct subsets of human Va24-

invariant NKT cells: cytokine responses and chemokine receptor

expression. Trends Immunol. 2002;23:516.

21. Lee PT, Benlagha K, Teyton L, Bendelac A. Distinct functional line-

ages of human Va24 natural killer T cells. J Exp Med. 2002;195:637.

22. Nieda M, Nicol A, Koezuka Y, et al. Activation of human Va24NKT

cells by a-glycosylceramide in a CD1d-restricted and Va24TCR-

mediated manner. Hum Immunol. 1999;60:10.

23. van der Vliet HJ, Nishi N, Koezuka Y, et al. Potent expansion of

human natural killer T cells using a-galactosylceramide

(KRN7000)-loaded monocyte-derived dendritic cells, cultured in the

presence of IL-7 and IL-15. J Immunol Methods. 2001;247:61.

24. Nishi N, van der Vliet HJ, Koezuka Y, et al. Synergistic effect of

KRN7000 with interleukin-15, -7, and -2 on the expansion of

human V a 24DV b 11D T cells in vitro. Hum Immunol. 2000;61:357.

25. Brossay L, Chioda M, Burdin N, et al. CD1d-mediated recognition of

an a-galactosylceramide by natural killer T cells is highly conserved

through mammalian evolution. J Exp Med. 1998;188:1521.

26. Yu KO, Im JS, Molano A, et al. Modulation of CD1d-restricted NKT

cell responses by using N-acyl variants of a-galactosylceramides. Proc

Natl Acad Sci U S A. 2005;102:3383.

27. Parekh VV, Singh AK, Wilson MT, et al. Quantitative and qualitative

differences in the in vivo response of NKT cells to distinct a- and b-

anomeric glycolipids. J Immunol. 2004;173:3693.

28. Lin H, Nieda M, Nicol AJ. Differential proliferative response of NKT

cell subpopulations to in vitro stimulation in presence of different

cytokines. Eur J Immunol. 2004;34:2664.

29. Nieda M, Okai M, Tazbirkova A, et al. Therapeutic activation of

Va24DVb11D NKT cells in human subjects results in highly

coordinated secondary activation of acquired and innate immunity.

Blood. 2004;103:383.

30. Kenna T, Golden-Mason L, Porcelli SA, et al. NKT cells from normal

and tumor-bearing human livers are phenotypically and functionally

distinct from murine NKT cells. J Immunol. 2003;171:1775.

31. Nakajima H, Liu XW, Wynshaw-Boris A, et al. An indirect effect of

Stat5a in IL-2-induced proliferation: a critical role for Stat5a in IL-

2-mediated IL-2 receptor a chain induction. Immunity. 1997;7:691.

32. Carnaud C, Lee D, Donnars O, et al. Cutting edge: Cross-talk between

cells of the innate immune system: NKT cells rapidly activate NK

cells. J Immunol. 1999;163:4647.

33. Nishimura T, Kitamura H, Iwakabe K, et al. The interface between

innate and acquired immunity: glycolipid antigen presentation by

CD1d-expressing dendritic cells to NKT cells induces the differentia-

tion of antigen-specific cytotoxic T lymphocytes. Int Immunol. 2000;

12:987.

34. Johnson-Leger C, Hasbold J, Holman M, Klaus GG. The effects of

IFN-g on CD40-mediated activation of B cells from X-linked immu-

nodeficient or normal mice. J Immunol. 1997;159:1150.