Introduction

NKT cells are a heterogeneous population of lymphocytes loosely defined as cells that express a TCR in addition to NK-cell markers. As such, NKT cells encompass a number of cellular subpopulations¹.

For the purposes of the present review, NKT cells are defined as a population of cells that are CD1d-restricted, either through an invariant TCR², or through a diverse TCR^{3, 4}. In mice, invariant NKT (iNKT) cells express the V14/J18 TCR -chain in association with V8.2, V7 or V2, and can be further subdivided into CD4⁺ or double-negative (DN) populations. Human iNKT cells are similarly defined by an invariant TCR comprising the -chain V24/JQ in association with the -chain V11. Human iNKT can be further subdivided into four categories: CD4⁺, CD8⁺, CD8⁺ or DN. Most, but not all iNKT cells express the CD161 antigen, defined as NK1.1 in mice and NKR-P1A in humans.

CD1d-dependent T cells represent an additional population of NKT cells found in both mice and humans. In comparison to CD1d-restricted NKT cells, this population expresses a more diverse TCR repertoire, with a preference for V3.2-J9 or V8 together with V8 (in mice). Murine CD1d-dependent T cells can be CD4⁺ or DN^{5, 6}. CD1d-dependent T cells with a diverse TCR repertoire have also been identified in humans^{5, 6}.

CD1d-restricted NKT cells are defined based on the recognition of antigens presented by the non-classical MHC class I-like molecule CD1d. Human and murine NKT cells are functionally and phenotypically homologous to the extent that mouse CD1d-restricted NKT cells can recognize human CD1d, and vice versa⁷. Because of the impressive functional similarities between the human and mouse immune systems, murine models have been widely exploited to define the role of NKT cells in various biological situations. Here, we summarize the recent publications dealing with the functions and relevance of NKT cells in the context of viral infections, both in murine models and in humans.

Immunoregulatory role of NKT cells

NKT cells have attracted a great deal of attention because of their potential to link the innate and adaptive arms of the immune system. Characteristically, NKT cells respond very rapidly to stimuli and are then able to activate a number of immune effectors. Diverse immunoregulatory functions have been ascribed to NKT cells, including the ability to prevent tumour development and metastatic spread^{8, 9, 10, 11, 12}, and the capacity to regulate the development and severity of autoimmune and allergic diseases^{13, 14, 15, 16, 17}. In addition, NKT cells have been implicated as important mediators of the immune responses conferring protection against viruses, bacteria and parasites.

Because the identity of the natural ligands for CD1d remains unclear, a marine sponge-derived glycolipid, -galactosylceramide (GalCer), has been used to activate iNKT cells selectively in a CD1d-dependent manner^{18, 19, 20}. GalCer has been instrumental in elucidating the mechanisms of antigen-specific activation of iNKT cells and in delineating the functional capacities of these cells. Presentation of GalCer by CD1d-expressing APC, such as dendritic cells (DC), results in rapid activation of iNKT cells. GalCer-activated iNKT cells can then exert their immunological functions either directly or indirectly. iNKT cells can be directly cytolytic through mechanisms believed to involve the perforin/granzymes pathway as well as the Fas/FasL pathway^{21, 22}. Indirectly, NKT cells exert their activities by the rapid release of abundant quantities of cytokines, including IFN- and IL-4^{18, 23, 24}. It is these iNKT-cell-derived cytokines that can then affect innate immunity through the recruitment and activation of immune effectors, including NK cells^{21, 25, 26}, macrophages²⁷ and DC²⁸. In addition, bystander-activation of conventional T cells^{29, 30, 31} and B cells³² represents a way for NKT cells to modulate adaptive immune responses.

Because of their ability to secrete IFN- and IL-4, iNKT cells activated with GalCer can drive immune responses with either a Th1- or a Th2-bias. NKT-cell-derived Th1 cytokines, such as IFN-, are important in conferring protection against microbial pathogens, like malaria^{33, 34}, and have been shown to play a crucial role in interfering with the initiation, growth and metastatic spread of tumours^{8, 9, 10, 11, 35}. NKT-cell-derived Th2 cytokines, such as IL-4, can down-regulate immune responses and have been shown to contribute to protection against the development of autoimmune diseases. For example, repeated GalCer administration enhances Th2-mediated immune responses^{29, 32, 36}, which can protect against the development of experimental allergic encephalomyelitis (EAE)¹⁵ and type I diabetes^{13, 14}.

Although much remains to be understood about the molecular mechanisms by which NKT cells mediate their various functions, the molecular nature of some of the responses induced by GalCer has been elucidated. In addition to the rapid release of both IFN- and IL-4, GalCer-activated NKT cells up-regulate CD40 ligand (CD40L) and modulate the activities of DC by CD40-CD40L interactions. These interactions result in the secretion of IL-12 from DC^{37, 38}. Thus, GalCer-activated NKT cells can influence the fate of DC and can induce them to mature into Th1-promoting APC. The IL-12 secreted by DC can further activate iNKT cells, but does not appear to have an effect on NK cells or T cells^{38, 39}. IL-12 activation of NKT cells induces a second burst of IFN-, which, together with IL-12, results in the indirect activation of NK cells. Activation of NK cells is accompanied by the release of copious quantities of IFN-, and it is the NK-cell-derived cytokine that ultimately constitutes the bulk of the systemic IFN- detected after activation of NKT cells by GalCer. In this system, IL-12 appears to be critical in driving a Th1 immune response. The production of IL-4 by GalCer-activated NKT cells is independent of IL-12³⁷.

Role of CD1d-restricted NKT cells in viral immunity

The role of NKT cells in immunity to viral infections has not been fully elucidated; however, it is clear that the capacity to participate in early immune responses and to modulate both innate and adaptive immunity confers NKT cells the potential to mediate activities important in the control of pathogens and the subsequent clearance of infections. What is currently understood about the participation of NKT cells in viral immunity will be discussed in detail in the following sections of this review. The involvement of NKT cells in viral infection in mouse models and in human situations is summarized in Table 1^{40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57}.

Table 1 - Involvement of NKT cells in viral immune-surveillance.

Full table

Establishing the role of NKT cells in viral immunity and defining the molecular basis of NKT-cell-mediated effects

The potential involvement of NKT cells in viral immunity has been addressed by examining the outcomes of infection and pathogenesis in either CD1d^-/- and/or J18^-/- gene-targeted mice. Before we discuss some of the actual research findings, it is important to note that a distinction needs to be made with regards to the defect in NKT-cell responses in these two model systems. In CD1d^-/- mice, all CD1d-restricted NKT-cell responses are defective, as positive selection of both the iNKT and the CD1d-restricted T cells requires expression of the CD1d molecule during thymic development^{58, 59}. The activities of cells other than NKT cells can also be affected, as CD1d is expressed on a number of cell types, including DC. In contrast, J18^-/- mice selectively lack the iNKT-cell population that expresses an invariant V14/J18 TCR chain³⁹.

Currently, mouse models of viral infection have defined a role for different populations of NKT cells in immunity to HSV type 1 (HSV-1), HSV-2, encephalomyocarditis virus (ECMV-D) and respiratory syncytial virus (RSV).

The role of invariant V14/J18 NKT cells as well as CD1d-mediated responses have been defined in the context of the immune responses elicited in response to HSV-1 infection. In comparison to wild-type C57BL/6 (B6) mice, both B6.CD1d^-/- and B6.J18^-/- mice show increased susceptibility to HSV-1 infection, characterized by enhanced spread of virus to the nervous system, defective clearance of virus from the skin and nervous system and exacerbated disease symptoms⁴⁰. The mechanisms by which CD1d and iNKT cells participate in the control of HSV-1 infection remain to be elucidated, but it has been suggested that the lack of NKT-cell responses impairs the development of virus-specific adaptive immune responses⁴⁰.

A role for CD1d-restricted T cells, iNKT cells and/or the CD1d molecule in immunity against HSV-2 was investigated by examining viral immunopathology in CD1d-deficient mice. On intravaginal HSV-2 infection, B6.CD1d^-/- mice exhibited increased virus shedding, reduced survival and exacerbated vaginal pathology. In this study, a series of gene-targeted mice was used to determine which components of innate immunity are required to control infection in the genital epithelium. NK cells, CD1d-restricted NKT cells, IL-15 and IFN- were shown to play a role. Although the contribution of cytolysis as opposed to production of IFN- was not defined, NK and NKT cells were shown to be the principal source of IFN- in the vagina early after HSV-2 infection⁴¹. Indeed, the worst disease outcome was observed in IL-15^-/- mice, which lack NK and NKT cells, suggesting that activation of both NK and NKT cells might be required for optimal control of HSV-2.

A protective role for CD1d-restricted T cells, but not iNKT cells, has been demonstrated in infection with the picornavirus ECMV-D, a pathogen that causes paralysis, diabetes and myocarditis. The severity of the disease caused by ECMV-D infection is genetically determined, with B6 mice showing reduced susceptibility compared with BALB/c mice. NK cells, T cells and macrophages have been implicated in the control of this infection, with both IL-12 and IFN- playing critical roles⁶⁰. Studies by Exley et al. found BALB/c.CD1d^-/- mice to be more susceptible to ECMV-D-induced disease than wild-type controls; however, susceptibility was not increased in BALB/c.J18^-/- mice⁴². The defect observed in the CD1d-deficient mice was attributed to a loss of IL-12 production from APC and a subsequent defect in the activation of NK cells, with the consequential impairment in production of IFN-⁴³. In support of this hypothesis, treatment of CD1d-deficient mice with exogenous IL-12 prior to ECMV-D infection conferred complete protection against disease. Thus, in ECMV-D infection, it is likely that rapidly activated CD1d-restriced T cells act primarily to stimulate APC, and thus induce the production of IL-12. Because the IL-12-conferred protection correlated to increased IFN- production, and given that NK cells are important in mediating resistance to ECMV-D-induced disease, Exley et al. proposed a model where IFN- from NKT cells and IL-12 from APC activate bystander NK cells, leading to a further burst of IFN- and activation of adaptive immune responses⁴³.

In many instances, resistance to viral infection requires the concerted efforts of innate and adaptive immune responses. Thus, the protection conferred by NKT cells could be a result of the fact that the cytokines produced by the NKT cells are not only critical in activating early innate immune responses, but also favour the development of the classical virus-specific T-cell responses that are ultimately responsible for clearing the infection. This mechanism has been proposed to account for the protective effects of NKT cells in infection with ECMV-D⁴³. Evidence in support of a role of NKT cells in the induction of antiviral CD8⁺ T-cell responses comes from studies of RSV infection. In this model, the relevance of NKT cells was found to depend on the genetic background of the host. RSV infection is complex, and although viral clearance requires the induction of effective NK and CD8⁺ T-cell responses, exaggerated T-cell activation participates in immunopathology and can result in more severe disease. In CD1d-deficient B6 mice, viral clearance was delayed in comparison to control wild-type B6 mice. However, viral clearance was not compromised in CD1d-deficient BALB/c mice. CD8⁺ T-cell numbers were reduced in B6.CD1d^-/- as well as BALB/c.CD1d^-/- mice, whereas impaired IFN- production was observed in BALB/c.CD1d^-/- mice, but not in CD1d-deficient B6 mice. In B6.CD1d^-/- mice, an increase in NK-cell recruitment/proliferation was observed following RSV infection, suggesting that effective NK-cell responses in these mice could compensate for the lack of CD1d expression and allow a normal IFN- response to be elicited⁴⁴. Because disease symptoms in the RSV model are caused by the immune responses generated against the virus, rather than by cytopathic effects of the virus itself, disease was more severe in control wild-type BALB/c mice than in CD1d-deficient animals⁴⁴.

The use of gene-targeted CD1d^-/- and J18^-/- mice has excluded a role for NKT cells in the control of infection with two viruses that are natural pathogens of mice, murine CMV (MCMV) and lymphocytic choriomeningitis virus (LCMV). NKT cells were excluded as a major contributor in the early control of MCMV infection, because B6.J18^-/- mice exhibited viral titres equivalent to those observed in control wild-type mice during the early acute stages of infection⁴⁵. Similarly, studies in CD1d^-/- mice ruled out a role for NKT cells in the control of LCMV. Viral titres in the spleen and blood were equivalent in the gene-targeted and wild-type mice, and importantly, NK and virus-specific cytotoxic T-cell (CTL) responses were generated effectively in the CD1d-deficient mice⁴⁶. A rapid but transient loss in NKT cells has been reported to accompany infection with LCMV, with the numbers of NKT cells in the liver, spleen and peritoneum decreasing within 2-3 days of infection and returning to normal levels within 1-2 weeks⁴⁷. A similar loss of NKT cells was observed after treatment with poly-polyinosinic-polycytidylic acid (IC), which, like LCMV infection, induces the secretion of type I IFN. The detection of LCMV transcripts in NKT cells purified from infected mice provides evidence that these cells could be infected by the virus. These findings, together with the fact that NKT cells appear to be lost in situations where IFN- are secreted⁴⁸, suggests that the reduction in NKT-cell percentages observed after LCMV infection could be attributed to the cytotoxic effects of IFN- resulting in death of infected NKT cells. In support of this hypothesis, an increase in caspase 3 activation was reported in liver NKT cells isolated from LCMV-infected mice⁴⁷. The loss of NKT cells in response to poly-IC treatment, however, would suggest that if the loss of these cells is mediated by IFN-, additional recognition of the targets is not required.

Defining the capacity of activated NKT cells to improve antiviral immunity

Although the mechanisms by which NKT cells induce antiviral responses are not fully characterized, the experimental evidence summarized above implicates NKT cells in activation of innate immunity and induction of the adaptive Th1-biased immune responses required for the effective control of infection with several viral pathogens. In these models, the antiviral effects of NKT cells have been partially defined and mostly attributed to the release of large quantities of IFN-. Indeed, the importance of NKT cells might depend on their ability to be activated early during viral infection, and to produce cytokines such as IFN- quickly, thus initiating the activation of NK cells, which then exert the primary antiviral effects, either directly by cytolysis or indirectly through the further release of cytokines.

Despite the fact that NKT cells might not play a significant role in the control of some viral infections, their ability to release copious quantities of immunoregulatory cytokines quickly on activation makes them a target worth exploiting to improve antiviral immune responses. GalCer effectively activates NKT cells. Administration of GalCer during viral infection has proved that specific CD1d-dependent activation of NKT cells has beneficial therapeutic effects. Indeed, even in infection models⁴⁵ where NKT cells have been determined not to play a natural role in antiviral immunity, activation of NKT cells with GalCer has resulted in significant therapeutic responses. For example, despite the fact that NKT cells⁴⁵ are not required for the early control of MCMV infection, administration of GalCer results in markedly decreased viral replication in visceral organs. The therapeutic effects of GalCer have been observed⁴⁵ in MCMV-susceptible BALB/c mice as well as in MCMV-resistant B6 mice infected with high doses of virus. A series of analyses in gene-targeted mice or mice specifically depleted of various cellular subsets by antibody immunotherapy clearly demonstrated that the antiviral effects of GalCer in MCMV infection required NKT cells, but were ultimately mediated through the activation of bystander NK cells and involved both perforin and IFN-⁴⁵. Thus, the therapeutic effects of GalCer were absent in gene-targeted CD1d-deficient or J18^-/- mice, as well as in mice possessing an intact NKT population but depleted of NK cells by treatment with the anti-asialoGM1 antibody prior to GalCer administration⁴⁵. Gene-targeted mice lacking perforin and/or IFN-, the two main molecular effector mechanisms used by NK cells, clearly demonstrated that both perforin and IFN- participate in the antiviral effectiveness of GalCer, with the therapeutic effects being compromised in mice lacking one or both of these effector systems⁴⁵.

Therapeutic effects of GalCer have been described in other viral infection models. A single treatment with GalCer during ECMV-D infection protects against virally induced diabetes and myocarditis, and results in dramatically less severe paralysis⁴². The role of CD1d-reactive NKT cells was clearly established by the finding that the ameliorating effects of GalCer are lost in CD1d- or J18-deficient mice. The protective effects of GalCer are speculated to require a combination of effects, initiated by the release of IFN-, and involving the subsequent activation of DC, macrophages, NK cells and T cells. The relevance of each component has not been established experimentally⁴².

In RSV infection, GalCer administration results in a significant reduction in illness; however, in what appears to be a paradox, it also leads to a significant delay in viral clearance. Although GalCer activation induces the release of IFN- and the infiltration of lungs with greater numbers of CD8⁺ T cells and NK cells, the concomitant increase in the production of IL-4 appears to curb the antiviral effectiveness of the CD8⁺ T cells⁴⁴. In this study, RSV infected mice were given multiple (3) injections of GalCer, which might have resulted in the preferential secretion of large quantities of IL-4 and the resultant impairment in CTL function^{18, 29, 32, 61, 62}.

Mouse models can be developed to permit the analysis of human pathogens that do not have a mouse equivalent. One such pathogen is hepatitis B virus (HBV). The transgenic HBV model, generated by Chisari and Guidotti et al. provides an elegant system to study the pathogenesis and immunobiology of this important human pathogen⁶³. In this model, non-cytopathic HBV replicates effectively in the liver. Despite the fact that the natural role of NKT cells in HBV infection has not been established, GalCer therapy in transgenic mice leads to rapid production of IFN- by intrahepatic NKT cells, and results in a marked decrease in viral replication⁴⁸. Although the effects of GalCer are associated with the ability of NKT-cell-derived IFN- to recruit activated NK cells and T cells into the liver, it has been shown that IFN- released by NKT cells, and perhaps resident hepatic NK cells, is sufficient to mediate the antiviral response of GalCer in the liver⁴⁹. Indeed, depletion of CD4⁺ and CD8⁺ T cells prior to treatment with GalCer demonstrates that therapeutic effectiveness is independent of the activation of conventional T cells⁴⁸. In addition to defining the critical contribution of IFN- in mediating the antiviral effects induced by GalCer, these studies established a role for IFN-. Increased levels of IFN- were detected in the liver 24 h after GalCer treatment, and the therapeutic effects of GalCer were diminished in mice lacking the IFN- receptor⁴⁸. The ability of GalCer to induce the release of IFN- in the liver was unappreciated prior to the findings reported in this study⁴⁸.

The studies summarized above provide unequivocal evidence for the role played by activated NKT cells in limiting viral infections. The evidence for the participation of NKT cells, either in a natural state or postactivation, in exacerbating the disease that accompanies some viral infections will be discussed in the following section.

Defining the involvement of NKT cells in viral immunopathology

Although many studies have attributed beneficial antiviral responses to NKT cells, these cells have also been implicated in detrimental immune responses that lead to immunopathology and disease. An example of the contribution of NKT cells to the immunopathology of viral infection has been provided by studies of RSV infection. In this system, CD1d-deficient mice demonstrate diminished disease after infection with RSV. As discussed earlier, RSV is a complex model where the disease symptoms that accompany the infection are caused by the immune response to the pathogen, rather than the cytopathogenicity of the virus itself. In RSV infection, the presence of NKT cells correlates to the recruitment/proliferation of RSV-specific CD8⁺ T cells infiltrating the lungs and the enhanced production of IFN-. Although these responses are required to reduce viral replication, they are also the cause of the illness that occurs post-RSV infection; thus, disease severity is actually reduced in CD1d-deficient mice⁴⁴. Given the role of activated CTL in the immunopathology of RSV, it is then not surprising that reducing the efficacy of these effectors would ameliorate disease symptoms. Indeed, activating NKT cells with GalCer to skew the T-cell response towards a Th2-bias because of excessive IL-4 production results in reduced immunopathology, despite the concomitant delay in viral clearance⁴⁴.

Another infection in which NKT cells have been shown to have detrimental effects with respect to disease induction is infection with Coxsackie virus B3 (CVB3). This virus causes severe myocarditis. Although the mechanisms underlying this disease remain the subject of debate, the immunopathological role of CD8⁺ T cells has been well documented. So, as for RSV infection, CD8⁺ T cells are required to control viral replication, but are also causative of the inflammation that results in myocardial disease⁵¹. A study by Huber et al. showed that CVB3 causes severe myocarditis in BALB/c and BALB/c.J18^-/- mice, but that disease is reduced in BALB/c.CD1d^-/- mice⁵². No differences were observed among the three strains with respect to titres of CVB3 in the heart. Thus, despite the fact that CD1d expression and CD1d-restricted T cells appear not to be involved in controlling viral replication, they are involved in causing myocarditis. It has been shown that expression of CD1d increases early during the course of CVB3 infection, both in vitro and in vivo. Given that expression of CD1d correlates to the sensitivity of myocytes to T-cell-mediated lysis in vitro, it has been postulated that CD1d could allow the recognition and elimination of CVB3-infected myocytes by these effector T cells⁵². Alternatively, or in addition, activation of T cells to produce pro-inflammatory cytokines, and thus to activate additional adaptive T-cell responses, might cause the immunopathology that accompanies CVB3 infection.

In support for a role of NKT cells in immunopathology, a study by Baron et al. has shown that acute hepatitis develops in HBV-transgenic mice after adoptive transfer of NKT cells⁵⁰. Transfer of naïve wild-type splenocytes into HBV-transgenic recombinase-activating gene deficient (RAG)1^-/- or TCR C recipients leads to the rapid development of acute hepatitis. The cells causing the hepatitis are prevalent in the liver, as 100-fold fewer liver lymphocytes are sufficient to induce disease. A combination of sorting and depletion experiments demonstrated that a population of CD1d-restricted NKT cells specifically mediate the liver injury observed post-transfer. Interestingly, these cells are not J18⁺ iNKT cells and fail to recognize GalCer presented by CD1d⁵⁰. Why NKT cells become activated post-transfer to recipient mice and which mechanisms they then use to induce hepatitis remain unclear.

The role of NKT cells in immunopathology during viral infection appears to be restricted. Although this could indeed be the case, it is possible that NKT cells could be involved in causing disease more often, and that their role has remained somewhat unrecognized because many analyses of viral infection focus on viral clearance, rather than more broadly addressing viral disease. Furthermore, at least in the case of myocarditis, the analysis is somewhat simplified by the fact that the pathology is restricted to the myocardium, rather than affecting a number of the host's organs.

Defining the involvement of NKT cells in human viral infections

Little is known about the relevance of human NKT cells during viral infection, and only few studies have attempted to investigate what happens to these cells during the course of a viral infection in human hosts.

Hepatitis C virus (HCV) can establish a persistent infection in the liver. The frequency of invariant V24/V11 NKT cells in peripheral blood mononuclear cell preparations has been found to be significantly lower in HCV-seropositive and viraemic (PCR-positive) individuals compared with seropositive, non-viraemic (PCR-negative) or uninfected individuals⁵³. The reduced numbers of NKT cells during active HCV infection might be a result of (i) TCR down-regulation⁶⁴; (ii) activation-induced cell death; or (iii) compartmentalization of NKT cells in peripheral organs, such as the liver. Hepatic V24/V11 NKT cells from PCR-positive viraemic individuals are highly activated as compared with cells obtained from PCR-negative patients, indicating that NKT cells can become activated in the hepatic compartment⁵³. An independent study, focusing on the fate of non-invariant CD1d-reactive T cells in HCV-infected livers, proposed that the hepatic T-cell pool from chronically HCV-infected patients contains a sizeable Th1-like population of CD1d-reactive T cells with non-invariant TCR repertoires⁵⁴. Primary intrahepatic lymphocyte lines derived from patients with chronic hepatitis express high levels of CD161 and CD1d, but invariant TCR repertoires are rarely observed. Furthermore, the CD1d-reactive lines are Th1-biased and exhibit CD1d-dependent cytotoxicity⁵⁴.

Human NKT cells could be involved in the initial stages of HIV-1 infection and in the spread of the virus throughout the host. HIV entry into target cells requires concomitant expression of CD4 and chemokine receptors. CD4⁺ NKT cells are suitable targets for HIV-1 infection because they express CD4 as well as high levels of the chemokine receptor CCR5. Indeed, these cells are highly susceptible to infection with R5-tropic HIV-1 strains⁵⁵. In addition, NKT cells are selectively depleted during HIV infection, and circulating CD4⁺ V24⁺ V11⁺ NKT cells are found at reduced frequencies in HIV-infected patients^{55, 56, 57}. In some HIV-infected individuals, the number of CD4^- NKT cells also diminishes during infection. The major loss in NKT cells after HIV infection occurs within the first year of seroconversion⁵⁷ and could be a result of cytopathic infection and/or a continuous process of activation-induced cell death. The effect that the loss of NKT cells has on the course of HIV-1 infection is unclear, although it possibly contributes to the development of the immunodeficiency state. A non-human primate model is being used to further investigate the role of NKT cells during the course of HIV infection⁶⁵.

The relevance of NKT-cell responses in controlling herpesvirus infection in humans is provided by a recent study reporting the development of severe disease after vaccination against Varicella zoster virus in a subject showing a significant deficiency in NKT cells, especially the V24⁺ NKT subset⁶⁶.

How are NKT cells activated during a viral infection and how is the immune response skewed to a Th1 phenotype?

The collective findings from the viral infection models examined to date highlight the diverse roles NKT cells can play in viral immunity (Table 1). Indeed, the role of NKT cells varies from being fully dispensable, to being critical inducers of antiviral immunity, to being causative of the immunopathology that results in the disease observed during an infection.

Although the available data do not elucidate the exact role of NKT cells in the context of viral clearance, in the instances when these cells are required, they have been shown to enhance innate immunity as well as aiding in the development of effective Th1-adaptive immune responses. In the context of many viral infections, the generation of Th1-biased immune responses is considered critical to effectively controlling viral replication. Although NKT cells might affect the development of such responses, the mechanisms responsible for skewing immunity towards a Th1 phenotype are at present undefined. It is possible that the Th1/Th2 phenotype of adaptive immune responses is influenced by interactions between NKT cells and DC (Figure 1). The complex interactions that occur between NKT cells and immature CD1d-expressing DC could result in maturation of DC with different effector phenotypes (i.e. IL-12-producing DC vs IL-10-producing DC). The strength of the TCR signal is thought to define the phenotype of the DC interacting with NKT cells. Thus, a strong TCR signal, such as that provided by GalCer, will up-regulate CD40L on NKT cells, which, in the presence of IFN-, will induce the maturation of DC able to secrete IL-12^{38, 67}. In contrast, the presentation of self-ligands to autoreactive NKT cells delivers a weak TCR signal that does not induce the up-regulation of CD40L. In situations where this is combined with the presence of large quantities of IL-4, maturation of IL-10-secreting DC is favoured⁶⁷. Importantly, a weak TCR signal could be amplified by a cytokine signal (e.g. IL-12), thus eliciting Th1-biased immune responses (Figure 1).

Figure 1.

Role of CD1d-restricted NKT cells in viral immune-surveillance. (a) The interplay between NKT cells and dendritic cells (DC) defines the nature of early immune responses. NKT cells receiving an activation signal through TCR ligation will rapidly release both Th1 and Th2 cytokines. DC can receive maturation signals resulting in the release of IL-12, which principally feeds back to NKT cells to induce a second burst of IFN-. Signals inducing the production of IL-12 can be complex and possibly require engagement of multiple costimulatory molecules. Weak signals can favour the generation of tolerogenic immune responses. Weak TCR signals can be amplified by a concomitant cytokine signal, such as that provided by IL-12. In addition to producing cytokines, activated NKT cells have the potential to be directly cytolytic. (b) Cytokines released by NKT cells (IFN-, IL-4) and DC (IL-12, IFN-, IL-10 and possibly TGF-) indirectly activate cellular components of the innate and adaptive immune system. A fine balance in the type of responses elicited will determine whether the outcome of NK-T-cell activation is beneficial and improves viral clearance and/or disease, or detrimental and the cause of immunopathology.

Full figure and legend (59K)

Experimental evidence suggests that exposure to pathogen-derived antigens is not obligatory for NKT-cell selection and activation. Self-antigens are sufficient for the positive selection and activation of NKT cells. Indeed, NKT cells develop normally, express activation markers (CD69 and CD44) and have the memory phenotype of conventional T cells in germ-free mice⁶⁸. The self-reactivity of NKT cells could also influence their effector functions. Weak responsiveness of NKT cells to self-ligands can be amplified by IL-12 secreted by DC in response to activation of Toll-like receptors (TLR) by microbial products. NKT cells receiving these two activating signals (IL-12 from DC and signalling from self-ligands through CD1d presentation) are then capable of producing large quantities of IFN-⁶⁹. It has recently become clear that viral components recognized by TLR induce DC activation and inflammatory cytokine production^{70, 71, 72}. Thus, in response to viral infection, strong NKT-cell activation involving a cytokine-mediated signal is likely to occur (Figure 1).

At present, the identity of the ligands involved in the physiological activation of NKT cells during viral infections remains unknown. Possible ligands include virus-derived hydrophobic peptide antigens or self-antigens derived from glycolipids or polypeptides produced by infected cells in response to stress. There is little experimental evidence for the specific recognition by NKT cells of pathogen-derived glycolipids presented in the context of CD1d. However, it is known that CD1d molecules can present endogenous cellular lipid ligands, including phosphatidyl inositol (PI) and phosphatidyl ethanolamine, which can be recognized by self-reactive CD1d-restricted NKT cells^{7, 18, 19, 61, 73, 74}. The CD1d molecule can also potentially present exogenous pathogen-derived lipids that are processed in the endosomal compartment. Indeed, loading of CD1d with self-ligands such as PI could be involved in this process. PI possibly functions as a non-antigenic lipid chaperone, loaded onto CD1d molecules in the endoplasmic reticulum or in early compartments of the secretory pathway, to protect the CD1d groove until CD1d molecules encounter antigenic lipids in endosomal compartments⁷⁵. CD1d is sorted to late endosomes by adaptor-protein complexes that recognize a tyrosine motif in the cytoplasmic tail of this molecule^{76, 77, 78, 79}. Trafficking of CD1d to endosomal compartments is integral to the ability to present antigen to NKT cells, and cells expressing tail-deleted CD1d molecules do not activate NKT cells^{19, 79}. Activation of CD1d-restricted T cells, however, does not require trafficking of CD1d to endosomes, and cells expressing tail-deleted CD1d can effectively stimulate mouse T-cell hybridomas and positively select CD1d-restricted T cells^{79, 80}. The different CD1d trafficking required for the activation of NKT and CD1d-dependent T cells suggests that these two populations might recognize different antigens. Importantly, activation of different NKT-cell subsets could depend on where viral antigens come into contact with CD1d molecules.

The antiviral activity of GalCer has been well defined and the synthetic ligand has helped to elucidate the mechanisms of iNKT-cell activation. Despite this, the immune responses elicited by GalCer might not exactly reflect those that are relevant during a natural viral infection. First, both iNKT cells (activated by GalCer) and CD1d-restricted T cells (not activated by GalCer) have been implicated in controlling viral replication. Second, GalCer-mediated activation of iNKT cells induces the production of Th2-type cytokines, such as IL-4 and IL-13. Indeed, multiple administrations of GalCer lead to increased release of IL-4 from iNKT cells and the establishment of predominantly Th2-biased immune responses^{29, 32, 62}. Such responses, generally unfavourable in controlling primary viral infections, are, for example, induced by GalCer therapy in RSV infection. Thus, future exploitation of NKT-cell activation to improve antiviral immunity should focus on the development of synthetic ligands that selectively stimulate NKT cells to induce Th1- or Th2-biased immune responses. An GalCer analogue has been identified that preferentially induces NKT cells to secrete IL-4. This compound has been shown to suppress the development of EAE by inducing a strong Th2 polarization⁸¹.

Selective activation of specific immune responses could be possible if subsets of NKT cells with different functionalities exist. Functionally distinct NKT-cell subsets could also explain the apparently contradictory activities ascribed to NKT cells. The existence of NKT1 and NKT2 subsets in mice has been suggested⁸². The NKT1 subset is proposed to comprise non-CD1d-restricted NKT cells using a diverse TCR, as well as CD1d-restricted T cells; these cells release IFN-, but not IL-4. The NKT2 subset is thought to include mainly iNKT cells capable of releasing IL-4, as well as IFN-⁸². Similarly, human V24 iNKT cells have been subdivided into two subsets with differing cytokine profiles, and hence potentially different functionalities. In humans, CD4⁺ NKT cells are the main producers of IL-4 when analysed ex vivo^{83, 84}.

Concluding remarks

In conclusion, NKT cells have been implicated in antiviral immunity with either beneficial or detrimental effects. Similarly, activation of NKT cells with GalCer can have both therapeutic and damaging effects. Although the mechanisms by which NKT cells exert their functions in viral immunity need to be more clearly defined, the studies performed to date underscore the importance of accurately dissecting the functionality of NKT cells in models of viral infection before these cells can be exploited in therapeutic settings.

References

1. Kronenberg M, Gapin L. The unconventional lifestyle of NKT cells. Nat. Rev. Immunol. 2002; 2: 557–68. | Article | PubMed | ISI | ChemPort |
2. Bendelac A, Lantz O, Quimby ME, Yewdell JW, Bennink JR, Brutkiewicz RR. CD1 recognition by mouse NK1⁺ T lymphocytes. Science 1995; 268: 863–5. | Article | PubMed | ISI | ChemPort |
3. Behar SM, Podrebarac TA, Roy CJ, Wang CR, Brenner MB. Diverse TCRs recognize murine CD1. J. Immunol. 1999; 162: 161–7. | PubMed | ISI | ChemPort |
4. Park SH, Weiss A, Benlagha K, Kyin T, Teyton L, Bendelac A. The mouse CD1d-restricted repertoire is dominated by a few autoreactive T cell receptor families. J. Exp. Med. 2001; 193: 893–904. | Article | PubMed | ISI | ChemPort |
5. Cardell S, Tangri S, Chan S, Kronenberg M, Benoist C, Mathis D. CD1-restricted CD4⁺ T cells in major histocompatibility complex class II-deficient mice. J. Exp. Med. 1995; 182: 993–1004. | Article | PubMed | ISI | ChemPort |
6. Behar SM, Cardell S. Diverse CD1d-restricted T cells: diverse phenotypes, and diverse functions. Semin. Immunol. 2000; 12: 551–60. | Article | PubMed | ISI | ChemPort |
7. Brossay L, Chioda M, Burdin N et al. CD1d-mediated recognition of an -galactosylceramide by natural killer T cells is highly conserved through mammalian evolution. J. Exp. Med. 1998; 188: 1521–8. | Article | PubMed | ISI | ChemPort |
8. Kawano T, Cui J, Koezuka Y et al. Natural killer-like nonspecific tumor cell lysis mediated by specific ligand-activated V14 NKT cells. Proc. Natl Acad. Sci. USA 1998; 95: 5690–93. | Article | PubMed | ChemPort |
9. Nakagawa R, Motoki K, Nakamura H et al. Antitumor activity of -galactosylceramide, KRN7000, in mice with EL-4 hepatic metastasis and its cytokine production. Oncol. Res. 1998; 10: 561–8. | PubMed | ChemPort |
10. Toura I, Kawano T, Akutsu Y, Nakayama T, Ochiai T, Taniguchi M. Inhibition of experimental tumor metastasis by dendritic cells pulsed with -galactosylceramide. J. Immunol. 1999; 163: 2387–91. | PubMed | ISI | ChemPort |
11. Smyth MJ, Crowe NY, Pellicci DG et al. Sequential production of interferon- by NK1.1⁺ T cells and natural killer cells is essential for the antimetastatic effect of -galactosylceramide. Blood 2002; 99: 1259–66. | Article | PubMed | ISI | ChemPort |
12. Smyth MJ, Crowe NY, Hayakawa Y, Takeda K, Yagita H, Godfrey DI. NKT cells - conductors of tumor immunity? Curr. Opin. Immunol. 2002; 14: 165–71. | Article | PubMed | ISI | ChemPort |
13. Hong S, Wilson MT, Serizawa I et al. The natural killer T-cell ligand -galactosylceramide prevents autoimmune diabetes in non-obese diabetic mice. Nat. Med. 2001; 7: 1052–6. | Article | PubMed | ISI | ChemPort |
14. Sharif S, Arreaza GA, Zucker P et al. Activation of natural killer T cells by -galactosylceramide treatment prevents the onset and recurrence of autoimmune Type 1 diabetes. Nat. Med. 2001; 7: 1057–62. | Article | PubMed | ISI | ChemPort |
15. Singh AK, Wilson MT, Hong S et al. Natural killer T cell activation protects mice against experimental autoimmune encephalomyelitis. J. Exp. Med. 2001; 194: 1801–11. | Article | PubMed | ISI | ChemPort |
16. Yang JQ, Saxena V, Xu H, van Kaer L, Wang CR, Singh RR. Repeated -galactosylceramide administration results in expansion of NK T cells and alleviates inflammatory dermatitis in MRL-lpr/lpr mice. J. Immunol. 2003; 171: 4439–46. | PubMed | ISI | ChemPort |
17. Lisbonne M, Diem S, de Castro Keller A et al. Invariant V14 NKT cells are required for allergen-induced airway inflammation and hyperreactivity in an experimental asthma model. J. Immunol. 2003; 171: 1637–41. | PubMed | ChemPort |
18. Kawano T, Cui J, Koezuka Y et al. CD1d-restricted and TCR-mediated activation of V14 NKT cells by glycosylceramides. Science 1997; 278: 1626–9. | Article | PubMed | ISI | ChemPort |
19. Spada FM, Koezuka Y, Porcelli SA. CD1d-restricted recognition of synthetic glycolipid antigens by human natural killer T cells. J. Exp. Med. 1998; 188: 1529–34. | Article | PubMed | ISI | ChemPort |
20. Hayakawa Y, Godfrey DI, Smyth MJ. -galactosylceramide: potential immunomodulatory activity and future application. Curr. Med. Chem. 2004; 11: 241–52. | Article | PubMed | ISI | ChemPort |
21. 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: 3114–22. | PubMed | ISI | ChemPort |
22. Kawano T, Nakayama T, Kamada N et al. Antitumor cytotoxicity mediated by ligand-activated human V24 NKT cells. Cancer Res. 1999; 59: 5102–5. | PubMed | ISI | ChemPort |
23. Bendelac A, Rivera MN, Park SH, Roark JH. Mouse CD1-specific NK1 T cells: development, specificity, and function. Annu. Rev. Immunol. 1997; 15: 535–62. | Article | PubMed | ISI | ChemPort |
24. Godfrey DI, Hammond KJ, Poulton LD, Smyth MJ, Baxter AG. NKT cells: facts, functions and fallacies. Immunol. Today 2000; 21: 573–83. | Article | PubMed | ISI | ChemPort |
25. Carnaud C, Lee D, Donnars O et al. Cross-talk between cells of the innate immune system: NKT cells rapidly activate NK cells. J. Immunol. 1999; 163: 4647–50. | PubMed | ISI | ChemPort |
26. Eberl G, MacDonald HR. Selective induction of NK cell proliferation and cytotoxicity by activated NKT cells. Eur. J. Immunol. 2000; 30: 985–92. | Article | PubMed | ISI | ChemPort |
27. Nakagawa R, Serizawa I, Motoki K et al. Antitumor activity of -galactosylceramide, KRN7000, in mice with the melanoma B16 hepatic metastasis and immunohistological study of tumor infiltrating cells. Oncol. Res. 2000; 12: 51–8. | PubMed | ISI | ChemPort |
28. Naumov YN, Bahjat KS, Gausling R et al. Activation of CD1d-restricted T cells protects NOD mice from developing diabetes by regulating dendritic cell subsets. Proc. Natl Acad. Sci. USA 2001; 98: 13 838–43.
29. Singh N, Hong S, Scherer DC et al. Activation of NK T cells by CD1d and -galactosylceramide directs conventional T cells to the acquisition of a Th2 phenotype. J. Immunol. 1999; 163: 2373–7. | PubMed | ISI | ChemPort |
30. 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 differentiation of antigen-specific cytotoxic T lymphocytes. Int. Immunol. 2000; 12: 987–94. | Article | PubMed | ISI | ChemPort |
31. Gonzalez-Aseguinolaza G, Van Kaer L, Bergmann CC et al. Natural killer T cell ligand -galactosylceramide enhances protective immunity induced by malaria vaccines. J. Exp. Med. 2002; 195: 617–24. | Article | PubMed | ISI | ChemPort |
32. Burdin N, Brossay L, Kronenberg M. Immunization with -galactosylceramide polarizes CD1-reactive NK T cells towards Th2 cytokine synthesis. Eur. J. Immunol. 1999; 29: 2014–25. | Article | PubMed | ISI | ChemPort |
33. Gonzalez-Aseguinolaza G, de Oliveira C, Tomaska M et al. -galactosylceramide-activated V14 natural killer T cells mediate protection against murine malaria. Proc. Natl Acad. Sci. USA 2000; 97: 8461–6. | Article | PubMed | ChemPort |
34. Hansen DS, Siomos MA, Buckingham L, Scalzo AA, Schofield L. Regulation of murine cerebral malaria pathogenesis by CD1d-restricted NKT cells and the natural killer complex. Immunity 2003; 18: 391–402. | Article | PubMed | ISI | ChemPort |
35. Hayakawa Y, Rovero S, Forni G, Smyth MJ. -galactosylceramide (KRN7000) suppression of chemical- and oncogene-dependent carcinogenesis. Proc. Natl Acad. Sci. USA 2003; 100: 9464–9. | Article | PubMed | ChemPort |
36. Kitamura H, Ohta A, Sekimoto M et al. -galactosylceramide induces early B-cell activation through IL-4 production by NKT cells. Cell Immunol. 2000; 199: 37–42. | Article | PubMed | ISI | ChemPort |
37. Tomura M, Yu WG, Ahn HJ et al. A novel function of V14⁺CD4⁺ NKT cells: stimulation of IL-12 production by antigen-presenting cells in the innate immune system. J. Immunol. 1999; 163: 93–101. | PubMed | ISI | ChemPort |
38. Kitamura H, Iwakabe K, Yahata T et al. The natural killer T (NKT) cell ligand -galactosylceramide demonstrates its immunopotentiating effect by inducing interleukin (IL)-12 production by dendritic cells and IL-12 receptor expression on NKT cells. J. Exp. Med. 1999; 189: 1121–8. | Article | PubMed | ISI | ChemPort |
39. Cui J, Shin T, Kawano T et al. Requirement for V14 NKT cells in IL-12-mediated rejection of tumors. Science 1997; 278: 1623–6. | Article | PubMed | ISI | ChemPort |
40. Grubor-Bauk B, Simmons A, Mayrhofer G, Speck PG. Impaired clearance of herpes simplex virus type 1 from mice lacking CD1d or NKT cells expressing the semivariant V14-J281 TCR. J. Immunol. 2003; 170: 1430–34. | PubMed | ISI | ChemPort |
41. Ashkar AA, Rosenthal KL. Interleukin-15 and natural killer and NKT cells play a critical role in innate protection against genital herpes simplex virus type 2 infection. J. Virol. 2003; 77: 10 168–71.
42. Exley MA, Bigley NJ, Cheng O et al. CD1d-reactive T-cell activation leads to amelioration of disease caused by diabetogenic encephalomyocarditis virus. J. Leukoc. Biol. 2001; 69: 713–18. | PubMed | ISI | ChemPort |
43. Exley MA, Bigley NJ, Cheng O et al. Innate immune response to encephalomyocarditis virus infection mediated by CD1d. Immunology 2003; 110: 519–26. | Article | PubMed | ChemPort |
44. Johnson TR, Hong S, Van Kaer L, Koezuka Y, Graham BS. NK T cells contribute to expansion of CD8⁺ T cells and amplification of antiviral immune responses to respiratory syncytial virus. J. Virol. 2002; 76: 4294–303. | Article | PubMed | ISI | ChemPort |
45. van Dommelen SL, Tabarias HA, Smyth MJ, Degli-Esposti MA. Activation of natural killer (NK) T cells during murine cytomegalovirus infection enhances the antiviral response mediated by NK cells. J. Virol. 2003; 77: 1877–84. | Article | PubMed | ChemPort |
46. Spence PM, Sriram V, Van Kaer L, Hobbs JA, Brutkiewicz RR. Generation of cellular immunity to lymphocytic choriomeningitis virus is independent of CD1d1 expression. Immunology 2001; 104: 168–74. | Article | PubMed | ChemPort |
47. Hobbs JA, Cho S, Roberts TJ et al. Selective loss of natural killer T cells by apoptosis following infection with lymphocytic choriomeningitis virus. J. Virol. 2001; 75: 10 746–54.
48. Kakimi K, Guidotti LG, Koezuka Y, Chisari FV. Natural killer T cell activation inhibits hepatitis B virus replication in vivo. J. Exp. Med. 2000; 192: 921–30. | Article | PubMed | ISI | ChemPort |
49. Kakimi K, Lane TE, Chisari FV, Guidotti LG. Inhibition of hepatitis B virus replication by activated NKT cells does not require inflammatory cell recruitment to the liver. J. Immunol. 2001; 167: 6701–5. | PubMed | ISI | ChemPort |
50. Baron JL, Gardiner L, Nishimura S, Shinkai K, Locksley R, Ganem D. Activation of a nonclassical NKT cell subset in a transgenic mouse model of hepatitis B virus infection. Immunity 2002; 16: 583–94. | Article | PubMed | ISI | ChemPort |
51. Henke A, Huber S, Stelzner A, Whitton JL. The role of CD8⁺ T lymphocytes in coxsackievirus B3-induced myocarditis. J. Virol. 1995; 69: 6720–28. | PubMed | ISI | ChemPort |
52. Huber S, Sartini D, Exley M. Role of CD1d in coxsackievirus B3-induced myocarditis. J. Immunol. 2003; 170: 3147–53. | PubMed | ISI | ChemPort |
53. Lucas M, Gadola S, Meier U et al. Frequency and phenotype of circulating V24/V11 double-positive natural killer T cells during hepatitis C virus infection. J. Virol. 2003; 77: 2251–7. | Article | PubMed | ChemPort |
54. Exley MA, He Q, Cheng O et al. Compartmentalization of Th1-like noninvariant CD1d-reactive T cells in hepatitis C virus-infected liver. J. Immunol. 2002; 168: 1519–23. | PubMed | ISI | ChemPort |
55. Motsinger A, Haas DW, Stanic AK, Van Kaer L, Joyce S, Unutmaz D. CD1d-restricted human natural killer T cells are highly susceptible to human immunodeficiency virus 1 infection. J. Exp. Med. 2002; 195: 869–79. | Article | PubMed | ChemPort |
56. Sandberg JK, Fast NM, Palacios EH et al. Selective loss of innate CD4⁺ V24 natural killer T cells in human immunodeficiency virus infection. J. Virol. 2002; 76: 7528–34. | Article | PubMed | ChemPort |
57. van der Vliet HJ, von Blomberg BM, Hazenberg MD et al. Selective decrease in circulating V24⁺ V11⁺ NKT cells during HIV type 1 infection. J. Immunol. 2002; 168: 1490–95. | PubMed | ChemPort |
58. Chen YH, Chiu NM, Mandal M, Wang N, Wang CR. Impaired NK1⁺ T cell development and early IL-4 production in CD1-deficient mice. Immunity 1997; 6: 459–67. | Article | PubMed | ISI | ChemPort |
59. MacDonald HR. Development and selection of NKT cells. Curr. Opin. Immunol. 2002; 14: 250–54. | Article | PubMed | ISI | ChemPort |
60. Ozmen L, Aguet M, Trinchieri G, Garotta G. The in vivo antiviral activity of interleukin-12 is mediated by gamma interferon. J. Virol. 1995; 69: 8147–50. | PubMed | ChemPort |
61. Burdin N, Brossay L, Koezuka Y et al. Selective ability of mouse CD1 to present glycolipids: -galactosylceramide specifically stimulates V14⁺ NK T lymphocytes. J. Immunol. 1998; 161: 3271–81. | PubMed | ISI | ChemPort |
62. Matsuda JL, Gapin L, Baron JL et al. Mouse V14i natural killer T cells are resistant to cytokine polarization in vivo. Proc. Natl Acad. Sci. USA 2003; 100: 8395–400. | Article | PubMed | ChemPort |
63. Guidotti LG, Matzke B, Schaller H, Chisari FV. High-level hepatitis B virus replication in transgenic mice. J. Virol. 1995; 69: 6158–69. | PubMed | ISI | ChemPort |
64. Crowe NY, Uldrich AP, Kyparissoudis K et al. Glycolipid antigen drives rapid expansion and sustained cytokine production by NK T cells. J. Immunol. 2003; 171: 4020–27. | PubMed | ISI | ChemPort |
65. Motsinger A, Azimzadeh A, Stanic AK et al. Identification and simian immunodeficiency virus infection of CD1d-restricted macaque natural killer T cells. J. Virol. 2003; 77: 8153–8. | Article | PubMed | ISI | ChemPort |
66. Levy O, Orange JS, Hibberd P et al. Disseminated varicella infection due to the vaccine strain of varicella-zoster virus, in a patient with a novel deficiency in natural killer T cells. J. Infect. Dis. 2003; 188: 948–53. | Article | PubMed |
67. Vincent MS, Leslie DS, Gumperz JE, Xiong X, Grant EP, Brenner MB. CD1-dependent dendritic cell instruction. Nat. Immunol. 2002; 3: 1163–8. | Article | PubMed | ISI | ChemPort |
68. Park SH, Benlagha K, Lee D, Balish E, Bendelac A. Unaltered phenotype, tissue distribution and function of V14⁺ NKT cells in germ-free mice. Eur. J. Immunol. 2000; 30: 620–25. | Article | PubMed | ISI | ChemPort |
69. Brigl M, Bry L, Kent SC, Gumperz JE, Brenner MB. Mechanism of CD1d-restricted natural killer T cell activation during microbial infection. Nat. Immunol. 2003; 4: 1230–37. | Article | PubMed | ISI | ChemPort |
70. Lund J, Sato A, Akira S, Medzhitov R, Iwasaki A. Toll-like receptor 9-mediated recognition of Herpes simplex virus-2 by plasmacytoid dendritic cells. J. Exp. Med. 2003; 198: 513–20. | Article | PubMed | ISI | ChemPort |
71. Lore K, Betts MR, Brenchley JM et al. Toll-like receptor ligands modulate dendritic cells to augment cytomegalovirus- and HIV-1-specific T cell responses. J. Immunol. 2003; 171: 4320–28. | PubMed | ISI | ChemPort |
72. Compton T, Kurt-Jones EA, Boehme KW et al. Human cytomegalovirus activates inflammatory cytokine responses via CD14 and Toll-like receptor 2. J. Virol. 2003; 77: 4588–96. | Article | PubMed | ISI | ChemPort |
73. Brossay L, Naidenko O, Burdin N, Matsuda J, Sakai T, Kronenberg M. Structural requirements for galactosylceramide recognition by CD1-restricted NK T cells. J. Immunol. 1998; 161: 5124–8. | PubMed | ISI | ChemPort |
74. Gumperz JE, Roy C, Makowska A et al. Murine CD1d-restricted T cell recognition of cellular lipids. Immunity 2000; 12: 211–21. | Article | PubMed | ISI | ChemPort |
75. Moody DB, Porcelli SA. Intracellular pathways of CD1 antigen presentation. Nat. Rev. Immunol. 2003; 3: 11–22. | Article | PubMed | ISI | ChemPort |
76. Roberts TJ, Sriram V, Spence PM et al. Recycling CD1d1 molecules present endogenous antigens processed in an endocytic compartment to NKT cells. J. Immunol. 2002; 168: 5409–14. | PubMed | ISI | ChemPort |
77. Jayawardena-Wolf J, Benlagha K, Chiu YH, Mehr R, Bendelac A. CD1d endosomal trafficking is independently regulated by an intrinsic CD1d-encoded tyrosine motif and by the invariant chain. Immunity 2001; 15: 897–908. | Article | PubMed | ISI | ChemPort |
78. Rodionov DG, Nordeng TW, Pedersen K, Balk SP, Bakke O. A critical tyrosine residue in the cytoplasmic tail is important for CD1d internalization but not for its basolateral sorting in MDCK cells. J. Immunol. 1999; 162: 1488–95. | PubMed | ISI | ChemPort |
79. Chiu YH, Jayawardena J, Weiss A et al. Distinct subsets of CD1d-restricted T cells recognize self-antigens loaded in different cellular compartments. J. Exp. Med. 1999; 189: 103–10. | Article | PubMed | ISI | ChemPort |
80. Chiu YH, Park SH, Benlagha K et al. Multiple defects in antigen presentation and T cell development by mice expressing cytoplasmic tail-truncated CD1d. Nat. Immunol. 2002; 3: 55–60. | Article | PubMed | ISI | ChemPort |
81. Miyamoto K, Miyake S, Yamamura T. A synthetic glycolipid prevents autoimmune encephalomyelitis by inducing TH2 bias of natural killer T cells. Nature 2001; 413: 531–4. | Article | PubMed | ISI | ChemPort |
82. Stenstrom M, Skold M, Ericsson A et al. Surface receptors identify mouse NK1.1⁺ T cell subsets distinguished by function and T cell receptor type. Eur. J. Immunol. 2004; 34: 56–65. | Article | PubMed | ChemPort |
83. 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: 625–36. | Article | PubMed | ISI | ChemPort |
84. Lee PT, Benlagha K, Teyton L, Bendelac A. Distinct functional lineages of human V24 natural killer T cells. J. Exp. Med. 2002; 195: 637–41. | Article | PubMed | ISI | ChemPort |

Acknowledgements

The authors are grateful to Dr Mark Smyth (Cancer Immunology Program, Trescowthick Laboratories, Peter MacCallum Cancer Centre, Melbourne, Victoria, Australia) for insightful discussions and for his valuable contribution to the MCMV studies. Research undertaken in our laboratory is supported by the National Health and Medical Research Council of Australia and the Wellcome Trust. MA Degli-Esposti is supported by a Wellcome Trust Overseas Senior Research Fellowship in Biomedical Science. GalCer was kindly provided to us and other scientists in the field by Kirin Brewery (Gunma, Japan). We thank Dr A Scalzo, Dr D Andrews and Dr C Andoniou for critical reading of this manuscript. Our apologies to scientists whose work was referenced only indirectly through reviews.