Natural T Cell Immunity to Intracellular Pathogens and Nonpeptidic Immunoregulatory Drugs

Innate functions and immunobiology of NT cells

T lymphocytes contribute to the immune response against intracellular pathogens by killing infected cells and helping B lymphocytes to produce antibodies. Most T cells recognize processed antigens in a form of short peptides bound to major histocompatibility complex (MHC) molecules on the antigen-presenting cell (APC) surface [1,2]. Alternatively, the recognition of superantigens is based on their interactions with particular V-regions of single T-cell receptors (TCR) chains, regardless of partner chains or the V-J junctional sequences [3-5]. Differently, NK lymphocytes can ‘see’ and kill target cells deficient in the expression of one or more MHC class I molecules. The NK function is negatively and positively regulated through a variety of receptors most of which are known to interact with MHC class I molecules (reviewed in [6-10]). The killer Ig-like receptor (KIR) family on human NK cells include receptors with well defined specificities for HLA-B and HLA-C alleles. Moreover, NK cells recognize several HLA class Ib molecules employing both immunoglobulin-like (Ig-like) and C-type lectin receptors. A dynamic and coordinated balance between activating and inhibitory receptors controls NK cell functions and influences the selective recognition of virus-infected, tumor or allogeneic cells.

In addition to the classical NK and T lymphocytes, our attention is drawn to a diverse group of cells that express both NK-like and T-cell like recognition structures, but do not appear to show the massive clonal expansion and immunological memory of classical CD4⁺ and CD8⁺ T cells. These T cell subsets may be collectively termed “natural T” (NT) cells, because their effector functions do not appear to require prior exposure to their specific antigen(s). NT cells have a highly restricted TCR repertoire and are frequently double negative (DN) for CD4 and CD8, Fig. (1). Both in humans and rodents, ab NT cells express NK-like markers belonging to the NKRP1 family and recognize glycolipid antigens presented by CD1 molecules [11-13] (Table 1). The ability to present glycolipid antigens distinguishes the CD1 molecules from other antigen-presenting molecules. Structural studies have shown that mouse CD1 has an hydrophobic groove that could accomodate the acyl chains of lipid antigens, and glycophosphatidylinositol (GPI) has been identified as a major natural ligand [14,15]. Human CD1 proteins can be divided into two separate groups – Group 1 comprising CD1a, CD1b and CD1c and Group 2 containing the more divergent CD1d [16,17]. The group 1 CD1 proteins were mainly expressed on many specialized antigen presenting cells (APC) including Langerhans cells in the epidermis and dendritic cells in different organs [18-20], whereas the Group 2 CD1 molecules are expressed on the gastrointestinal epithelium and hematopoietic cells [21,22]. The first type of CD1-dependent NT-cell recognition described was a direct reactivity to CD1 proteins, i.e. indicating a form of autoreactivity that is present within the normal immune system. NT lymphocytes express pratically monomorphic TCR a chains (Va14, Ja281 in mice; Va24, JaQ in humans). Among these TCRs, there is a very high representation of Vb11 which is the human homolog of mouse Vb8 [23-27]. The directly CD1-reactive NT cells are often DN or express only the CD8a subunit, suggesting that these cells belong to a different developmental lineage than the majority of MHC-restricted T cells. CD1-reactive DN NT cells can express either ab or gd TCR and are present in the circulation of all normal people (Homo sapiens) [26,28]. Interestingly, the autoreactivity of mouse

Figure 1. Patterns of self/non-self recognition by lymphocytes. ab TCR of CD4⁺ and CD8⁺ T cells recognize processed antigens in a form of short peptides bound to classical MHC molecules (class I: HLA-A,-B,-C; class II: HLA-DR,-DQ,-DP) on the APC surface (panel A). NK cells recognize classical (HLA-A, -B, -C) and nonclassical (HLA-E, -G) MHC class I molecules through NK-R (panel B). Va24Vb11, Vg9Vd2, VgxVd1 TCR define a class of natural T lymphocytes recognizing nonpeptidic antigens and/or MHC class I or CD1 molecules (panel C).

NT cells for CD1 is greatly augmented by the addition of the glycosfingolipid a-galactosilceramide (a-GalCer) [29,30]. Using Ja281-deficient mice, it was shown that the antitumor activity of a-GalCer is due to the CD1-mediated stimulation of NT cells [31]. Also, CD1d-specific human NT cells can be stimulated by a-GalCer. NT cells can be rapidly expanded from human peripheral blood lymphocytes in the presence of a-GalCer and IL-2 [32]. NT cells that recognize antigens of mycobacteria in a MHC-independent but CD1-restricted fashion have been found to express different ab TCR and the DN or CD8ab⁺ phenotype [33-35]. These cells are somewhat more specific in their recognition characteristics since they express different ab TCR rearrangements and are restricted by a single form of CD1 (including CD1a, -b, and –c). This may allow us to permit the distintion between antigens derived from closely related mycobacterial species [36]. Human CD1b-restricted NT cells recognize lipid and glycolipid components of the bacterial cell wall such as mycolic acids, lipoarabinomannan (LAM), phosphoinositide mannosides (PIMs), and glucose monomycolate (G-MM) [33-35,37]. It is noteworthy that the CD1-restricted NT cells specific for mycobacterial antigens produce proinflammatory cytokines upon activation [34]. This is compatible with their possible role in the control of intracellular bacterial infections. Moreover, CD1-restricted NT cells specific for the monosialoganglioside (GM1) were also described in patients with multiple sclerosis [38], suggesting that the antigens capable of stimulating autoreactive NT cells may include glycolipids.

Figure 2. Expression of NK markers and TCR by natural gd T lymphocytes. The percentage of NK (CD16⁺), T (CD3⁺) and NK-T cells (CD3⁺CD16⁺) among total lymphocytes was analysed by flow cytometry (panel A) using specific monoclonal antibodies: anti-Vd1-FITC (IgG1, clone 005768, Endogen, Woburn, MA), anti-Vd2-PE (IgG1, clone B6.1, Pharmingen, San Diego, CA), anti-CD3-Cy5 (IgG1, clone UCHT-1, Immunotech, Marseille, France) and CD56-APC (IgG1, cone B159, Pharmingen, San Diego, CA). Staining with anti-Vd1 and anti-Vd2 mAbs was performed and the percentages of Vd1⁺ and Vd2 T cells were calculated among CD3⁺ CD16^- T lymphocytes gated in region 2 (R2) and CD3⁺ CD16⁺ NK-T cells (R3) (panel B and C respectively).

NT cells expressing the gd TCR share many common characteristic with ab NT cells, indicating that the function of NT cells is independent of the ab or gd TCR rearrangements. T cell subsets that belong to the NT category appear to share several characteristic properties: 1) the expression of members of the NK-receptor family, 2) the expression of antigen receptors of limited diversity, 3) the lack of restriction by classical MHC class I and class II molecules (but they are sometimes restricted by CD1), and 4) the rapid release of large amounts of cytokines following their activation. Approximately 3-6% of human peripheral blood lymphocytes express the gd TCR [39] associated with the CD16 NK marker, further confirming their NT nature as shown in Fig. (2). The majority of these cells express the Vd2 TCR variable segment associated with the Vg9 segment [40,41]. The Vg9Vd2 subset is rare in the adult thymus but increases with age in the blood. This may be the result of positive selection in the periphery subsequent to sustained antigenic stimulation [41]. Vg9Vd2 NT lymphocytes recognize phosphorylated nonpeptidic microbial metabolites [42-46] and alkylamines [47] (Table 1). This recognition of nonpepdidic compounds by gd NT cells requires neither antigen uptake or processing, nor MHC class I or class II expression [43] allowing for a very rapid response to microbial immune challenges. An important mechanism that regulates Vg9Vd2 NT-cell responses to nonpeptidic compounds involves the NK-receptors (NK-R) (reviewed in [48-50]). One of the features is that in peripheral Vg9Vd2 NT cells, a prototype NK-R (the CD94/NKG2A complex) is translocated to the cell membrane after the exposure to nonpeptidic phosphoantigens. This results in a block of Vg9Vd2 NT cell activation [51]. A similar mechanism could be responsible for the Vg9Vd2 functional deficit observed in some HIV-infected persons [52,53]. It is noteworthy that these inhibitory receptors may also influence gd responses in anti-viral reactivity, tumor immunity and autoimunity [48,54]. The Vg9Vd2 subset appear to recognize ligands uniquely generated by abnormal metabolic routes [55]. Thus, the relevant pathophysiological situations are targeted by discriminating between concentration thresholds of nonself-nonsafe (or self-nonsafe) from self-safe molecules [56]. Also, gd NT cells have been shown to be a major source of cytokines such as IFNg, TNF-a [57], and b-chemokines (MIP-1a, MIP-1b, and RANTES) involved in the recruitment of cells of the monocyte/macrophage lineage during an inflammatory reaction and they may play an important role in HIV-infection [58,59].

Vd1 NT cells represent a minor lymphocyte subpopulation in the peripheral blood usually expressing a naive phenotype in healthy donors [60]. However, Vd1⁺ cells are the predominant gd T cell population in the postnatal thymus [61] and represent a major T cell population in the skin, intestinal and pulmonary epithelium [62]. Up to 1% of the epidermal cells in the skin of mice are gd T cells, called dendritic epidermal T cells (DETC) which fail to express both CD4 and CD8 and may originate from extrathymic differentiation [63]. These cells could be uniquely directed to combact antigens that enter through the skin. A second population of murine gd T cells has been identified in the intestinal epithelium of the mouse; these cells are called intestinal epithelial lymphocytes (IEL). Each subset was shown to display a limited TCR V gene usage (Vg3 for DECT and Vd5 for IEL). Unlike ab T cells, which re-circulate extensively, gd T cells in these epithelial tissues seems to remain immobile. The selective expression of TCR V gene segments in different epithelial tissues may reflect the possibility that these T cells are specialized in responding to certain types of antigens expressed at these sites. Similarly, Vd1 T cells represent the major T cell subpopulation in the human intestine [64-69], suggesting their possible role as first defense line against the invading microbes entering the gastrointestinal tract. Their Vd1-encoded receptor chain is typically co-expressed with Vg-encoded chains distinct from Vg9 [70]. The ligands recognized by Vd1 T lymphocytes are stress antigens of cellular origin (Table 1). Vd1⁺ NT cells recognize and interact with the MICA and MICB glycoproteins that are expressed mainly in gastrointestinal and cortical thymic cortical epithelium and share a low homology with MHC class I molecules [71]. The MIC genes have heat-shock response elements in their promoters, and the expression of their products may be (in contrast to the products of MHC class I genes) independent of peptide binding, since MICA does not associate with b2-microglobulin and can be expressed in TAP-negative cells [72]. The stress-induced expression of MICA and the closely related MICB, and their recognition by polyclonal Vd1 NT cells through TCR or the NKG2-D natural killer receptor may serve as an immune surveillance mechanism for detecting damaged, infected, or transformed intestinal epithelial cells. Moreover, tissue Vd1 NT lymphocytes were recently shown to recognize nonpolymorphic CD1c molecules [73]. Specifically, Vd1 NT cells were found to proliferate and release Th1 cytokines in response to CD1⁺ presenting cells and to lyse CD1c⁺ targets. The recognition of CD1c is TCR mediated and dependent on the Vd1 TCR expression. The gd NT cell lines and clones show a highly specific reactivity to CD1c proteins that is not dependent on the presence of exogenous antigen.

An increasing amount of data indicates that subsets of gd and ab NT cells are the key cellular elements of innate immunity and possess sophisticated recognition receptor systems. A variety of defined antigens influences NT cell activities both in vitro and in vivo (Table 1). NT cells may play a decisive role in establishing the “immunological difference” between dangerous invading pathogenic agents and harmless self-molecules. A striking characteristic of NT cells is their ability to produce high levels of IL-4, IFN-g and other cytokines/chemokines within a few hours of in vivo activation. Consequently, NT cells may have a critical influence on adaptive immune responses (reviewed in [74]).

Protective and pathogenic roles of NT cells in AIDS

In most infected persons, the chronic HIV infection leads to the degeneration of the immune and central nervous systems known as the acquired immunodeficiency syndrome (AIDS) [75-79]. The progressive course of the HIV disease can be substantially modified by the so-called highly active antiretroviral therapy (HAART) [80-82] In lymphoid tissues damaged by HIV infection, a CD4⁺ T-cell re-population occurs during the implementation of HAART. This treatment can induce sustained recovery of CD4 T cell reactivity against opportunistic pathogens, but does not eliminate HIV from the body, most likely due to virus escape tactics [83-89]. It has been suggested that cell-mediated immunity plays a central role in the HIV immunosurveillance mechanism(s) [90,91]. In particular, the role of expanded cytotoxic T lymphocytes (CTL) may be important in immunodeficiency virus infections (reviewed in [92,93]). In addition to classical adaptive ab CTL responses, activities of other cytotoxic lymphoid cells that are typically associated with innate immunity – i.e., ab and gd NT cells– may be positively or negatively involved in the course of HIV infection.

CD1-restricted MTB-reactive NT cells can be expanded from HIV-infected patients. These NT cells produce IFN-g in response to macrophages infected with live MTB, and show a marked CTL activity against macrophages infected with live MTB or pulsed with MTB antigens [94]. Significant alterations of peripheral blood gd NT cell distribution and function have been previously reported in HIV-infected persons, including a polyclonal decrease of Vg9Vd2 NT-absolute cell numbers [52,60,95-98] This Vd2 NT-cell depletion is more evident in HIV patients with opportunistic and other co-infections (OIC) [99]). The remaining Vg9Vd2 NT cells are frequently unable to proliferate [53,98,100] to their constitutive antigens [52] and/or to express the IL-2 receptor [52]). In normal conditions, Vg9Vd2 NT cells respond to antigen challenge by secreting large amounts of TNF-a and IFN-g [101] that contribute to the activation of both specific and aspecific immune responses [49]. In addition, gd NT cells are important IFN-g producers during tuberculosis infection, which is exacerbated in knocked-out mice lacking gd NT cells [102]. Recently, we recently observed a significant reduction of IFN-g production by Vd2 NT cells from HIV patients with OIC [99]. Thus, the most important consequence of the Vd2 deletion and anergy in HIV-infected persons may be an increased susceptibility to opportunistic pathogens. However, the mechanisms of Vg9Vd2 NT cell anergy are still poorly understood.

HIV has evolved a clever mechanism of avoiding the host’s NK and NT response [89]. Human cells are protected from NK-R recognition and killing primarily by HLA-C and HLA-E that inhibit the NK cell-mediated lysis by interacting with CD94/NKG2A receptors. HIV selectively downregulates HLA-A and HLA-B (minimizing detection by ab CTLs), but does not substantially affect the expression of HLA-C and HLA-E (protecting itself from the NK- and NT-cell attack). Although the MHC class Ia expression is significantly reduced by in vitro HIV-1 infection, the expression of HLA-E is rapidly up-regulated (Di Pietro F. et al., manuscript in preparation). The MHC class Ia MHC downregulation is mediated mainly by the HIV Nef activity. In contrast, using a Nef-deleted HIV strain we recently observed that HLA-E (class Ib molecule) expression is independent of the Nef protein. Intracellular staining for the p24 antigen revealed that HLA-E up-regulation is restricted to productively infected cells. Interestingly, HLA-E expression was increased on the surface of CD4 T cells from HIV-infected persons and related to disease progression. Thus, the induction of HLA-E expression may contribute to the evasion of NK/NT cell lysis and represents an additional escape route for HIV. Moreover, activated lymphocytes are potent stimulators of autoreactive cytotoxic Vd2 NT cells [103], and HIV infection is known to induce a persistent activation of the immune system. Since Vg9Vd2 NT cell activation is followed by the programmed cell death of the cytotoxic Vg9Vd2 NT cell effectors [104], the persistent stimulation of Vd2 NT cells in vivo during chronic HIV infection may be responsible for the observed Vd2 NT cells deletion/anergy. Indeed, preliminary data show differences between anergic and responder HIV patients at the level of Vg9 CDR3 sequences (C.D.Pauza, personal communication), known to be involved in the recognition of nonpeptidic antigens [105]. Thus, an altered Vg9 repertoire may also contribute to the HIV-associated gd NT-cell anergy. Finally, cytokine dysregulations in HIV infections may additionally contribute to the HIV-related gd NT cell anergy. Since IL-15 acts in synergy with IL-12 and promotes IFN-g production by gd NT cells [106], the Vd2 NT cell anergy may be due to a defective IL-15 and/or IL-12 production by accessory cells. It has been previously shown that the gd NT-cell anergy in HIV infected patients may depend on CD4-T cell functions. Specifically, allogeneic CD4 T cells from healthy donors were shown to restore the gd NT cell reactivity in HIV-infected persons [98]. Accordingly, we observed that in vitro CD45RA depletion results in gd NT cell hyporesponsiveness, suggesting that IL-2 may be provided by CD45RA helper cells for the gd NT cell activation [99]. Thus, the observed gd NT-cell restoration after HAART patients may be consequent to the recovery of helper functions mediated by naive T cells.

In spite of their lack of specificity in virus recognition, Vg9Vd2 NT cells could exert an important anti-viral function because of their ability to respond quickly, polyclonally and to differentiate into cytotoxic effectors [48,107-109]. Recognition of in vitro HIV-infected cells by peripheral Vg9Vd2 NT cells induces an in vitro expansion (comparable to that induced by mycobacterial nonpeptidic ligands) and a potent cytotoxic activity [48]. The lysis of virus-infected cells involves the CD94/NKG2 receptor [48] and is calcium-dependent [110], suggesting that the apoptotic death of target cells is induced more likely through the perforin/granzyme pathway rather than through the Fas/FasL pathway. Interestingly, the ability of PBMCs to produce high levels of CC-chemokines in vitro has been observed in exposed uninfected individuals and in HIV⁺ long term nonprogressors. We have shown that gd NT lymphocytes from healthy donors are able to produce high levels of the CC-chemokines MIP-1a, MIP-1b and RANTES [58,59] especially under conditions designed to mimic the in vivo activation of gd NT lymphocytes (such as the natural response to nonpeptidic microbial phosphoantigens). These observations suggest that gd NT lymphocytes may be one of the most important lymphocytic sources of CC-chemokines, acting polyclonally at the very early stages of microbial infections. This is supported by our finding that phosphoantigen-activated gd NT lymphocytes were able to suppress HIV replication in autologous cultures at efficiencies comparable to those of CD8⁺ T cells [58,111,112]. The observation that phosphoantigen-activated gd NT cells secrete high levels of CC-chemokines in vitro without any obvious requirement for clonal expansion and in a MHC-unrestricted manner, make gd T cells suitable targets for novel immunotherapeutic strategies in AIDS.

Increases in relative and absolute numbers of Vd1⁺ NT cells in the peripheral blood have been described in HIV-1-seropositive persons [60,95,96,113-116]. These increases do not represent a clonal expansion in response to HIV since they are a) independent of particular g-chain expression, b) not correlated with a CDR3-dependent Vd1 selection [60,116], and c) not associated with any particular junctional motifs [60]. They are also not correlated with high levels of HIV-1 antigenemia [96]. It was suggested that the peripheral expansion of Vd1 NT cells may be a result of cell trafficking from various tissues into circulation under the influence of cytokine changes occurring during HIV infections [60]. More recently, the gd NT-cell increases were associated with Mycobacterium avium infections (but not MTB) in AIDS patients [117]. In addition, recent data suggest that Vd1⁺ NT cells may play an important role in the immune response of immunosuppressed patients to cytomegalovirus [118]. Altogether, NT cell subsets are uniquely involved in the course of HIV infection and may be utilized to modulate antiviral immunity and to improve that host’s reactivity against OIC.

NT cell immunity in HCV immunopatho-genesis

HCV is responsible for chronic liver disease often progressing to cirrhosis and/or hepatocellular carcinoma [119,120]. Unfortunately, the majority of HCV-infected persons develop chronic infection that is difficult to eradicate with the current therapeutic protocols [120]. In contrast, less than 10% of individuals infected with hepatitis B virus will develop the chronic disease [121]. This difference may imply dissimilarities in “survival strategies” of hepatitis viruses and/or distinctions in host immune responses against HCV versus HBV. The progression of chronic C-hepatitis is characterized by a massive lymphocyte infiltrate in the liver [122,123]. The recruitment and migration of T cells to the site of infection is an important event in the course of inflammation. However the majority of intrahepatic lymphocytes (IHL) are not HCV specific. Thus, in the acute phase IHL could be critical for the resolution of the disease whereas in the chronic phase, IHL could be dangerous, damaging infected or non-infected hepatocytes. In the liver of HCV infected patients an increase of Va24 NT cells (up to 20 fold in comparison with the peripheral blood) was observed [124]. The memory/activated phenotype which is associated with the oligoclonal expansion suggests that these NT cells may recognize an endogenous ligand or ubiquitous antigen. The endogenous ligand could normally be present in the liver, but as a consequence of tissue damage it could be processed and presented in the context of the CD1d to the Va24/Vb11 TCR-expressing cells. As other NT cells, this subset is able to rapidly produce high amounts of Th1 cytokines and to kill other cells with a NK-like mechanism [125]. Recent findings have suggested a pathological role for NT cells producing high levels of IFN-g. It has been reported that Va14 NT cells play a critical role in various diseases including Salmonella infections [126], autoimmune diabetes [127], and systemic sclerosis [128]. ConA-induced hepatitis is considered a model of autoimmune hepatits and its characteristic liver injury is associated with lymphocyte infiltrates. The absence of liver injury in athymic nude mice or SCID mice indicates the immunopathological origin of hepatitis [129]. Several observations demonstrate that only Va14⁺ NT cells are required for the development of ConA-induced-hepatitis [130]. For example, the antibody depletion of NK1.1⁺ cells in vivo confers resistance to hepatitis, and this resistance is also present in b2 microglobulin or CD1 KO mice that have dysfunctional NK1.1⁺ NT cells [131,132]. The liver damage seems to be induced by different mechanisms such as Fas-FasL interactions [133,134], the perforin-granzyme system [135] and IFN-g [133,136]. The function of Va24Vb11 NT cells in humans is highly conserved, sharing ligand specificity and CD1d restriction [32,137,138] that could be involved in mediating the liver injury.

In the course of HCV infection, we have observed an increased fraction of NT cells expressing gd TCR in the liver (Agrati et al., submitted). Phenotypic analysies of gd NT cells have demonstrated that this increase is due to CD3⁺ cells expressing Vd1 TCR and results in an inversion of the intrahepatic Vd2 to Vd1 ratio. It is noteworthy that the systemic HIV infection induces Vd1 mobilization from tissues to the peripheral blood, whereas the same Vd1 NT cells are localized preferentially in the HCV-infected liver. The involvement of gd NT lymphocytes in immunosurveillance has been suggested in several infections with herpesviruses including herpes-simplex virus [108] and cytomegalovirus [118]. Several reports have provided strong evidence of anti-inflammatory role of gd NT through the homeostatic regulation of ab T cells [139]. The rapid recruitment of circulating gd NT cells in the liver could be induced by the recognition of virus-induced antigens and may determine the pattern of adaptive responses mediated by the subsequent activation of MHC-restricted ab T lymphocytes. The Vd1 NT cells recruited in the liver display an activated/memory phenotype. The expression of markers specific for memory cells is a common characteristic of all intrahepatic T cells in HCV⁺ patients. These cells tend to home in the region of infection by recognizing the altered vascular endothelium and chemoatractant molecules generated during the inflammatory process [140]. The acquisition of CD45RO marker indicates an antigen mediated stimulation. The Vd1 NT cell subset does not selectively express any particular Vg-chain which may be due to polyclonal superantigenic-like activation. Possible ligands induce MHC class I-related MICA and MICB that are recognized by intestinal epithelial NT cells co-expressing Vd1 and various Vg chains. MICA and MICB are present on epithelial cells and hepatocyte derived cell lines, and their expression may be stress-induced rather than constitutive [71]. The recognition could be mediated by the TCR or the NKG2-D natural killer receptor and could play an important role in the immune surveillance mechanism for the detection of damaged infected or transformed epithelial cells.

Nonpeptidic mycobacterial antigens and NT cells in Tuberculosis

Tuberculosis is still one of the leading causes of morbidity and mortality worldwide. After decades of declining incidence, the number of infected individuals is increasing once again in developed as well as developing countries, with the spread of multidrug-resistant tuberculosis making treatment of the disease increasingly difficult. Anti-MTB immunity depends on the interaction of antigen-specific CD4⁺ ab⁺ T lymphocytes with macrophages [141,142] although several studies indicate that NT lymphocytes are important for MTB immunosurveillance [114,143]. The complex array of cell-surface carbohydrate and lipid components is characteristic of mycobacteria. These nonpeptidic antigens have long been recognized as major targets of the humoral immune response against MTB, but more recently they have also been shown to stimulate several NT cell subsets. The first class is comprised of lipids and lipoglycans that trigger responses of a subpopulation of NT cells capable of recognizing antigens bound to the CD1 protein on the surface of presenting cells. The CD1-restricted NT cells specific for mycobacterial antigens produce proinflammatory cytokines, especially IFN-g [34], suggesting a function similar to CD4 T cells that confer protection against intracellular bacterial pathogens in mice [144]. CD1-restricted CD8⁺ NT cells kill MTB-infected macrophages through a cytotoxic granule-dependent mechanism that has a direct bactericidal effect. This is in contrast to the lysis of MTB-infected targets by DN CD1-restricted NT cells, that is caused by an interaction of the apoptosis-promoting Fas protein with its ligand and does not result in directly killing the released bacteria [145].

The second class of MTB nonpeptidic antigens is a family of phosphorylated ligands that activate gd NT cells. We have analyzed the gd NT-cell reactivity to nonpeptidic antigens ex vivo in primary MTB-infected children, and in vivo in rhesus monkeys possessing Vg9Vd2 NT cells closely related to their human counterparts [146]. These studies assessed the ability of Vg9Vd2 cells to expand in 14-day cultures with different synthetic [such as ribose-1-phosphate (Rib-1-P), xylose-1-phosphate (Xyl-1-P), dimethylallyl-pyrophosphate (DMAPP), monoethyl-pyrophosphate (MEP), diphosphoglyceric acid (DPG), and isopentenyl-pyrophosphate (IPP)], or natural (TUBAg-1) phosphoantigens (Table 1). The Vg9Vd2 NT-cell responses were highly increased in MTB-infected children in comparison with age-matched controls. The Vg9Vd2 NT-cell subset in tuberculin-positive MTB-infected children responded well to all antigens used and the strongest responses were detected using IPP. In contrast, the IPP-induced expansion of Vg9Vd2 NT cells from healthy, tuberculin-negative children was approximately 10 times lower. A few months after chemotherapy [146], the increased responsiveness of Vg9Vd2 cells sharply declined close to the levels detected in healthy tuberculin-negative children suggesting that persistent exposure to mycobacterial antigens was required for the presence of hyperactivity against phosphoantigens. In contrast, responses of ab T lymphocytes to tuberculin (PPD) and to an immunodominant epitope of the 38 kD MTB protein were increased after chemoterapy [147]. We have confirmed that the reactivity of gd NT cells in the peripheral blood is significantly increased during acute MTB infection also in adult TB patients (Gioia C. et al, manuscript in preparation). This increased reactivity was detected both in the lung and in the peripheral blood. The ability to produce IFNg following IPP stimulation was significantly reduced in both compartements, suggesting a funtional alteration of gd NT cells during active pulmonary tuberculosis. After effective chemotherapy, the IPP-induced NT cell proliferation was diminushed both in the peripheral blood and bronchoalveolar lavages, confirming that this immune response is dependent on the presence of active MTB infection. This functional defect in cytokine synthesis remained present even after 6 monhts of chemotherapy. Thus, the failure to secrete IFN-g upon activation of gd NT cells with nonpeptidic antigens may represent an important component of the anti-MTB immune response that is lost in TB patients and is not restored by chemotherapy. To analyze the phosphoantigen-priming of the Vg9Vd2 NT-cell subset in vivo, rhesus monkeys received intravenous injections of DPG. The DPG treatment resulted in a substantial upregulation of both proliferative and cytokine (IFN-g) responses to isopentenyl pyrophosphate (an increase from 2.2% to 48.9% of IFN-g producing Vg9Vd2 NT cells) in immunized animals [146].

Historically, the main focus of most vaccination strategies has been to boost immune reactivities against protein antigens. However, it is now clear that there is a substantial pool of lymphocytes that do not primarily recognize peptide antigens bound to MHC molecules (Table 1). In vivo exposures to nonpeptidic antigens substantially augment immune functions of genetically unrestricted NT cells. When activated, these cells exert a powerful antibacterial activity and release IFN-g and TNF-a. These properties may be useful in the design and development of novel vaccines and/or adjuvants. The long-lasting memory response of ab T lymphocytes is clinically applicable for evaluating previous exposures to MTB, but provides unsatisfactory information about the presence or absence of productive infection. Since, the gd NT-cell hyperactivity against phosphoantigens requires persistent antigenic exposure, the assessment of Vg9Vd2 NT-cell responses may become a useful tool to monitor a) active MTB infections and b) the efficacy of antimycobacterial therapies.

Emerging/re-emerging infectious diseases and new strategies of immune intervention with NT immunoregulatory drugs

Since the discovery of HIV in 1983 as the etiological agent of AIDS, the epidemic continues to expand, (reviewed in [148]). However, patterns of HIV infection and disease are changing. The availability of new, effective combinations of antiretroviral drugs modifies the course of HIV disease. Despite some advances in treating this disease, the therapies are expensive and accessis only possible in industrialized countries. The treatments themselves are difficult to administer and commonly fail, thus making prevention strategies of paramount importance for controlling the HIV epidemic [148]. While human vaccines targeting ab T cells and/or B cells are common, the possibility of stimulating the gd NT cell subset (which possesses potent antiviral activities and memory phenotypes) has not been fully explored. However, very recent experiments (Poccia et al., manuscript in preparation) have demonstrated a protective effect of activated gd NT cells in SIV infection, substantially improving both the clinical (survival) and virological (levels of viremia) parameters in vivo. These results together with the adjuvant capacity of cytokines produced by activated Vg9Vd2 NT cells suggest that in “combination vaccines” (targeting gd and ab T cells, and B cells) may have superior prophylactic and therapeutic effects in comparison with the classical peptide-polypeptide-based vaccines.

HIV infection markedly increases susceptibility to tuberculosis [149], and tuberculosis in HIV-infected patients accelerates the progression of immunodeficiency [150]. Standard vaccination strategies are particularly problematic because they result in the activation of CD4⁺ cells, which are the major reservoire of HIV. Thus, a better vaccination strategy could be to upregulate immune responses of CD4-negative T cells that are relatively resistant to HIV infection. Antigens recognized by NT cells may be particularly useful for this purpose. First, the reactivity to nonpeptidic antigens is MHC-unrestricted so that carbohydrates and glycolipids are likely to be recognized by most individuals, whereas MHC-restricted peptide antigens are generally recognized only by persons whose cells bear specific peptide-binding MHC molecules. Second, the nonpeptidic antigen stimulation of NT cells activates mainly DN ab T cells, CD8 cells and gd T cells, but not CD4 cells, so that this stimulation is less likely to activate CD4⁺cells and consequently to upregulate HIV replication. Vaccination strategies that augment the response of NT cells potentially represent a novel means of providing protection against tuberculosis in HIV-infected patients, while minimizing the risk of enhancing HIV replication through the stimulation of CD4⁺ T cells.

The emergence of MTB strains that are resistant to existing treatments adds a further alarming dimension to the problems of disease control, and highlights the potential for widespread transmission. Tuberculosis control efforts are hampered by delayed diagnosis, cost of antituberulosis drugs, difficulty in ensuring completion of prolonged therapy, and increasing rates of drug resistance. The enhanced strategy for global control would combine improved diagnosis and treatment with the prevention of the disease by effective vaccination. Vaccinations with the live attenuated vaccine strain, Mycobacterium bovis bacille Calmette Guerin (BCG), are a component of the neonatal immunization schedule recommended by the WHO Expanded Programme for Immunization. Although the BCG vaccination appears to reduce the incidence of childhood forms of tuberculosis, such as the frequently fatal tuberculous meningitis, a wide variation of its effect of BCG on the incidence of predominant pulmonary disease in adults, clearly shows that BCG is far from an ideal tuberculosis vaccine. Recent definitions of a series of MTB-specific but functionally distinct T cell subsets have provided new insights into the immunological processes that regulate MTB infections. For instance, NT cells recognizing lipoglycan components associated with the mycobacterial cell wall may be of particular interest in the development of new MTB vaccine candidates.

Both HCV and HIV are transmitted through the parenteral route. Similar risk factors contribute to the frequent association of HCV with HIV infection. HCV co-infection is present in 8-23% of HIV infected patients [151], and almost 80% of HIV-positive intravenous drug-user and hemophiliacs are co-infected with HCV. HCV does not to accelerate the progression of HIV infection. However, HIV may accelerate the development of HCV pathology, since the appearance of cirrhosis is more frequent and rapid in HIV co-infected patients. Recent results suggest that NT cells represent a solid fraction of IHL. Specifically, we have observed that Vd1 NT cells form a substantial part of IHL and this subset is abundant both in the liver and in the peripheral blood of HCV-HIV coinfected individuals (Agrati et al., manuscript in preparation). However, the role of NT cells in immunoprotective and/or immunopathological activities in the liver remains to be elucidated.

Figure 3. Activation of NT cells by microbial nonpeptidic compounds and/or stress antigens. The activation of natural T cells is regulated by positive or negative signals mediated by NK-R. Infections by viruses or intracellular bacteria can down regulate MHC class I, causing a loss of NK-R signals. Va24Vb11 TCR may recognize nonpeptidic antigens in the context of CD1, Vg9Vd2 TCR recognizes phosphometabolites (PMs) in a unrestricted manner and VgxVd1 are probably activated through MICA and MICB or CD1/TCR interactions. Viral or parasite infections may induce cellular stress that is able to increase the expression of different molecules such as MICA, MICB or CD1. In addition, the parasite metabolism and the increased host metabolism release high levels of PMs. The activated Va24Vb11 and gd NT lymphocytes show a different patterns of cytokine release (Th0/Th2 and Th0/Th1, respectively).

Recent observations indicate that nonpeptidic immunoregulatory compounds may be effectively used in vivo (Table 1). Since mouse gd NT cells have different ligand specificities than humans ones, the in vivo biological activity of these compounds has been analyzed in vivo in non-human primates. Simian gd NT cells activated in vivo by the immunization with DPG show increased proliferative activity and cytokine production [146]. Moreover, rhesus monkeys with phosphoantigen-specific gd NT cells display potent delayed type hypersensitivity reactions after intradermal injections of relevant phosphoantigens (Poccia et al., manuscript in preparation). Bisphosphonates compounds that are currently used in human trials to inhibit osteoclastic bone resorption, are structurally similar to recently identified nonpeptidic gd NT ligands, such as isopentenyl pyrophosphate (Table 1). Recent clinical reports on these compounds support the possibility of direct in vivo stimulation of Vg9Vd2 NT cells, i.e., patients treated with pamidronate showed substantial increases in the percentage of gd NT cells in their peripheral blood [152]. Pamidronate also induced a significant expansion of Vg9Vd2 NT cells in peripheral blood mononuclear cell cultures from healthy donors. Pamidronate-activated gd NT cells produced Th1 cytokines (IFN-g) and exhibited specific cytotoxicity against lymphoma (Daudi) and myeloma cell lines (RPMI 8226, U266). Pamidronate-treated bone marrow cultures from multiple myeloma patients showed significantly reduced plasma cell survival compared to untreated cultures, indicating that pamidronate-stimulated gd NT cells may have influential effects that might contribute to the antitumor effects of these drugs. [153]. The MHC-unrestricted Vg9Vd2 NT-cell activity against bacterial, viral and tumor antigens [48,107,108,154] provides a functional link between anti-microbial and anti-tumor innate immunity.

Recent evidence indicates that innate immune responses exhibit a much higher degree of specificity and are substantially more refined and complex than ever suspected [155,156]. The key cellular elements of innate immunity include subsets of gd and ab NT cells, both equipped with sophisticated recognition receptor systems that rival those used by the most important cellular entities of adaptive immunity – i.e., by B and ab T cells as summarized in Fig. (3). It is absolutely clear that NT cells have a decisive immunoregulatory influence on the adaptive immune response. It is conceivable that the innate immune system (and in particular its cellular components with discriminating cell-surface receptors) may play an ultimate role in establishing the “immunological difference” between dangerous invading pathogenic agents and harmless self-molecules. The expected advances in understanding the innate immune responses will provide fresh ideas for the treatment of emerging and re-emerging infections.

Aknowledgements

This work was supported by grants from the Current and Goal-oriented Research Projects of the Italian Ministry of Health and by the European Community Contract “QLK2-CT-1999-01093”. M.M. is holder of the UNESCO Chair in Interdisciplinary Biotechnology at the University of Rome “Tor Vergata”. C.A. is a fellow of the Institute for Infectious Diseases "L.Spallanzani". We thank Enitan Akinde for careful reading of the manuscript.

Abbreviations

AIDS = Acquired Immunodeficiency Syndrome

APC = Antigen Presenting cells

CMV = Cytomegalovirus

DN = Double Negative

HAART = Highly Active Anti-Retroviral Theraphy

HCV = Hepatitis-C Virus

HIV = Human Immunodeficiency Virus

IHL = Intrahepatic Lymphocytes

MHC = Major Histocompatibility Complex

MTB = Mycobacterium tuberculosis

NK = Natural Killer

NT = Natural T

OIC = Oppurtunistic Infections and Co- infections

PMs = Phosphometabolites

References:

[1] Babbitt, B. P., Allen, P. M., Matsueda, G., Haber, E. and Unanue, E. R. (1985) Nature, 317, 359-361.

[2] Bjorkman, P. J., Saper, M. A., Samraoui, B., Bennett, W. S., Strominger, J. L. and Wiley, D. C. (1987) Nature, 329, 506-512.

[3] Scherer, M. T., Ignatowicz, L., Winslow, G. M., Kappler, J. W. and Marrack, P. (1993) Annu. Rev. Cell Biol., 9, 101-128.

[4] Kappler, J., White, J., Kozono, H., Clements, J. and Marrack, P. (1994) Proc. Natl. Acad. Sci. U. S. A, 91, 8462-8466.

[5] Webb, S. R. and Gascoigne, N. R. (1994) Curr. Opin. Immunol., 6, 467-475.

[6] Moretta, A., Bottino, C., Millo, R. and Biassoni, R. (1999) Curr. Top. Microbiol. Immunol., 244, 69-84.

[7] Braud, V. M. and McMichael, A. J. (1999) Curr. Top. Microbiol. Immunol., 244, 85-95.

[8] Colonna, M., Nakajima, H. and Cella, M. (1999) J. Leukoc. Biol., 66, 718-722.

[9] Lopez-Botet, M. and Bellon, T. (1999) Curr. Opin. Immunol., 11, 301-307.

[10] Rajagopalan, S. and Long, E. O. (1999) J. Exp. Med., 189, 1093-1100.

[11] Porcelli, S. A. and Modlin, R. L. (1999) Annu. Rev. Immunol., 17, 297-329.

[12] Kronenberg, M., Brossay, L., Kurepa, Z. and Forman, J. (1999) Immunol. Today, 20, 515-521.

[13] Couedel, C., Peyrat, M. A., Brossay, L., Koezuka, Y., Porcelli, S. A., Davodeau, F. and Bonneville, M. (1998) Eur. J. Immunol., 28, 4391-4397.

[14] Joyce, S., Woods, A. S., Yewdell, J. W., Bennink, J. R., De Silva, A. D., Boesteanu, A., Balk, S. P., Cotter, R. J. and Brutkiewicz, R. R. (1998) Science, 279, 1541-1544.

[15] Zeng, Z., Castano, A. R., Segelke, B. W., Stura, E. A., Peterson, P. A. and Wilson, I. A. (1997) Science, 277, 339-345.

[16] Calabi, F., Jarvis, J. M., Martin, L. and Milstein, C. (1989) Eur. J. Immunol., 19, 285-292.

[17] Porcelli, S. A. (1995) Adv. Immunol., 59, 1-98.

[18] Fithian, E., Kung, P., Goldstein, G., Rubenfeld, M., Fenoglio, C. and Edelson, R. (1981) Proc. Natl. Acad. Sci. U. S. A, 78, 2541-2544.

[19] Nestle, F. O., Zheng, X. G., Thompson, C. B., Turka, L. A. and Nickoloff, B. J. (1993) J. Immunol., 151, 6535-6545.

[20] Meunier, L., Gonzalez-Ramos, A. and Cooper, K. D. (1993) J. Immunol., 151, 4067-4080.

[21] Bleicher, P. A., Balk, S. P., Hagen, S. J., Blumberg, R. S., Flotte, T. J. and Terhorst, C. (1990) Science, 250, 679-682.

[22] Blumberg, R. S., Colgan, S. P. and Balk, S. P. (1997) Clin. Exp. Immunol., 109, 223-225.

[23] Arase, H., Arase, N., Ogasawara, K., Good, R. A. and Onoe, K. (1992) Proc. Natl. Acad. Sci. U. S. A, 89, 6506-6510.

[24] Porcelli, S., Yockey, C. E., Brenner, M. B. and Balk, S. P. (1993) J. Exp. Med., 178, 1-16.

[25] Lantz, O. and Bendelac, A. (1994) J. Exp. Med., 180, 1097-1106.

[26] Exley, M., Garcia, J., Balk, S. P. and Porcelli, S. (1997) J. Exp. Med., 186, 109-120.

[27] Dellabona, P., Casorati, G., Friedli, B., Angman, L., Sallusto, F., Tunnacliffe, A., Roosneek, E. and Lanzavecchia, A. (1993) J. Exp. Med., 177, 1763-1771.

[28] Beckman, E. M., Melian, A., Behar, S. M., Sieling, P. A., Chatterjee, D., Furlong, S. T., Matsumoto, R., Rosat, J. P., Modlin, R. L. and Porcelli, S. A. (1996) J. Immunol., 157, 2795-2803.

[29] Kawano, T., Cui, J., Koezuka, Y., Toura, I., Kaneko, Y., Motoki, K., Ueno, H., Nakagawa, R., Sato, H., Kondo, E., Koseki, H. and Taniguchi, M. (1997) Science, 278, 1626-1629.

[30] Burdin, N., Brossay, L., Koezuka, Y., Smiley, S. T., Grusby, M. J., Gui, M., Taniguchi, M., Hayakawa, K. and Kronenberg, M. (1998) J. Immunol., 161, 3271-3281.

[31] Kawano, T., Cui, J., Koezuka, Y., Toura, I., Kaneko, Y., Sato, H., Kondo, E., Harada, M., Koseki, H., Nakayama, T., Tanaka, Y. and Taniguchi, M. (1998) Proc. Natl. Acad. Sci. U. S. A, 95, 5690-5693.

[32] Brossay, L., Chioda, M., Burdin, N., Koezuka, Y., Casorati, G., Dellabona, P. and Kronenberg, M. (1998) J. Exp. Med., 188, 1521-1528.

[33] Beckman, E. M., Porcelli, S. A., Morita, C. T., Behar, S. M., Furlong, S. T. and Brenner, M. B. (1994) Nature , 372, 691-694.

[34] Sieling, P. A., Chatterjee, D., Porcelli, S. A., Prigozy, T. I., Mazzaccaro, R. J., Soriano, T., Bloom, B. R., Brenner, M. B., Kronenberg, M. and Brennan, P. J. (1995) Science, 269, 227-230.

[35] Moody, D. B., Reinhold, B. B., Guy, M. R., Beckman, E. M., Frederique, D. E., Furlong, S. T., Ye, S., Reinhold, V. N., Sieling, P. A., Modlin, R. L., Besra, G. S. and Porcelli, S. A. (1997) Science, 278, 283-286.

[36] Porcelli, S. A., Segelke, B. W., Sugita, M., Wilson, I. A. and Brenner, M. B. (1998) Immunol. Today, 19, 362-368.

[37] Moody, D. B., Ulrichs, T., Muhlecker, W., Young, D. C., Gurcha, S. S., Grant, E., Rosat, J. P., Brenner, M. B., Costello, C. E., Besra, G. S. and Porcelli, S. A. (2000) Nature, 404, 884-888.

[38] Shamshiev, A., Donda, A., Carena, I., Mori, L., Kappos, L. and De Libero, G. (1999) Eur. J. Immunol., 29, 1667-1675.

[39] Lanier, L. L., Ruitenberg, J., Bolhuis, R. L., Borst, J., Phillips, J. H. and Testi, R. (1988) Eur. J. Immunol., 18, 1985-1992.

[40] Triebel, F., Faure, F., Mami-Chouaib, F., Jitsukawa, S., Griscelli, A., Genevee, C., Roman-Roman, S. and Hercend, T. (1988) Eur. J. Immunol., 18, 2021-2027.

[41] Parker, C. M., Groh, V., Band, H., Porcelli, S. A., Morita, C., Fabbi, M., Glass, D., Strominger, J. L. and Brenner, M. B. (1990) J. Exp. Med., 171, 1597-1612.

[42] Constant, P., Davodeau, F., Peyrat, M. A., Poquet, Y., Puzo, G., Bonneville, M. and Fournie, J. J. (1994) Science, 264, 267-270.

[43] Tanaka, Y., Morita, C. T., Tanaka, Y., Nieves, E., Brenner, M. B. and Bloom, B. R. (1995) Nature, 375, 155-158.

[44] Tanaka, Y., Sano, S., Nieves, E., De Libero, G., Rosa, D., Modlin, R. L., Brenner, M. B., Bloom, B. R. and Morita, C. T. (1994) Proc. Natl. Acad. Sci. U. S. A, 91, 8175-8179.

[45] Burk, M. R., Mori, L. and De Libero, G. (1995) Eur. J. Immunol., 25, 2052-2058.

[46] Bukowski, J. F., Morita, C. T., Tanaka, Y., Bloom, B. R., Brenner, M. B. and Band, H. (1995) J. Immunol., 154, 998-1006.

[47] Bukowski, J. F., Morita, C. T. and Brenner, M. B. (1999) Immunity., 11, 57-65.

[48] Poccia, F., Cipriani, B., Vendetti, S., Colizzi, V., Poquet, Y., Battistini, L., Lopez-Botet, M., Fournie, J. J. and Gougeon, M. L. (1997) J. Immunol., 159, 6009-6017.

[49] Poccia, F., Gougeon, M. L., Bonneville, M., Lopez-Botet, M., Moretta, A., Battistini, L., Wallace, M., Colizzi, V. and Malkovsky, M. (1998) Immunol. Today, 19, 253-256.

[50] Gougeon, M. L., Boullier, S., Colizzi, V. and Poccia, F. (1999) Microbes. Infect., 1, 219-226.

[51] Boullier, S., Poquet, Y., Halary, F., Bonneville, M., Fournie, J. J. and Gougeon, M. L. (1998) Eur. J. Immunol., 28, 3399-3410.

[52] Poccia, F., Boullier, S., Lecoeur, H., Cochet, M., Poquet, Y., Colizzi, V., Fournie, J. J. and Gougeon, M. L. (1996) J. Immunol., 157, 449-461.

[53] Wallace, M., Scharko, A. M., Pauza, C. D., Fisch, P., Imaoka, K., Kawabata, S., Fujihashi, K., Kiyono, H., Tanaka, Y., Bloom, B. R. and Malkovsky, M. (1997) Mol. Med., 3, 60-71.

[54] Fisch, P., Moris, A., Rammensee, H. G. and Handgretinger, R. (2000) Immunol. Today, 21, 187-191.

[55] Belmant, C., Espinosa, E., Poupot, R., Peyrat, M. A., Guiraud, M., Poquet, Y., Bonneville, M. and Fournie, J. J. (1999) J. Biol. Chem., 274, 32079-32084.

[56] Sicard, H. and Fournie, J. J. (2000) Infect. Immun., 68, 4375-4377.

[57] Lang, F., Peyrat, M. A., Constant, P., Davodeau, F., David-Ameline, J., Poquet, Y., Vie, H., Fournie, J. J. and Bonneville, M. (1995) J. Immunol., 154, 5986-5994.

[58] Poccia, F., Battistini, L., Cipriani, B., Mancino, G., Martini, F., Gougeon, M. L. and Colizzi, V. (1999) J. Infect. Dis., 180, 858-861.

[59] Cipriani, B., Borsellino, G., Poccia, F., Placido, R., Tramonti, D., Bach, S., Battistini, L. and Brosnan, C. F. (2000) Blood, 95, 39-47.

[60] Boullier, S., Cochet, M., Poccia, F. and Gougeon, M. L. (1995) J. Immunol., 154, 1418-1431.

[61] Casorati, G., De Libero, G., Lanzavecchia, A. and Migone, N. (1989) J. Exp. Med., 170, 1521-1535.

[62] Band, H., Porcelli, S. A., Panchamoorthy, G., McLean, J., Morita, C. T., Ishikawa, S., Modlin, R. L. and Brenner, M. B. (1991) Curr. Top. Microbiol. Immunol., 173 , 229-234.

[63] Poccia, F., Cicconi, R., Frasca, D., Mancini, C., Colizzi, V. and Doria, G. (1998) Int. Immunol., 10, 547-551.

[64] Spencer, J., Isaacson, P. G., Diss, T. C. and MacDonald, T. T. (1989) Eur. J. Immunol., 19, 1335-1338.

[65] Halstensen, T. S., Scott, H. and Brandtzaeg, P. (1989) Scand. J. Immunol., 30, 665-672.

[66] Deusch, K., Luling, F., Reich, K., Classen, M., Wagner, H. and Pfeffer, K. (1991) Eur. J. Immunol., 21, 1053-1059.

[67] Ullrich, R., Schieferdecker, H. L., Jahn, H. U. and Zeitz, M. (1991) Immunol. Res., 10, 306-309.

[68] De Libero, G., Rocci, M. P., Casorati, G., Giachino, C., Oderda, G., Tavassoli, K. and Migone, N. (1993) Eur. J. Immunol., 23, 499-504.

[69] Peyrat, M. A., Davodeau, F., Houde, I., Romagne, F., Necker, A., Leget, C., Cervoni, J. P., Cerf-Bensussan, N., Vie, H. and Bonneville, M. (1995) J. Immunol., 155, 3060-3067.

[70] Triebel, F. and Hercend, T. (1989) Immunol. Today, 10, 186-188.

[71] Groh, V., Steinle, A., Bauer, S. and Spies, T. (1998) Science, 279, 1737-1740.

[72] Groh, V., Bahram, S., Bauer, S., Herman, A., Beauchamp, M. and Spies, T. (1996) Proc. Natl. Acad. Sci. U. S. A, 93, 12445-12450.

[73] Spada, F. M., Grant, E. P., Peters, P. J., Sugita, M., Melian, A., Leslie, D. S., Lee, H. K., van Donselaar, E., Hanson, D. A., Krensky, A. M., Majdic, O., Porcelli, S. A., Morita, C. T. and Brenner, M. B. (2000) J. Exp. Med., 191, 937-948.

[74] Born, W., Cady, C., Jones-Carson, J., Mukasa, A., Lahn, M. and O'Brien, R. (1999) Adv. Immunol., 71, 77-144.

[75] Dalgleish, A. G. and Malkovsky, M. (1988) Br. J. Surg., 75, 171-172.

[76] Letvin, N. L. and King, N. W. (1990) J. Acquir. Immune. Defic. Syndr., 3, 1023-1040.

[77] Dalgleish, A. and Malkovsky, M. (1988) Adv. Cancer Res., 51, 307-360.

[78] Fauci, A. S., Schnittman, S. M., Poli, G., Koenig, S. and Pantaleo, G. (1991) Ann. Intern. Med., 114, 678-693.

[79] Buchbinder, S. P., Katz, M. H., Hessol, N. A., O'Malley, P. M. and Holmberg, S. D. (1994) AIDS, 8, 1123-1128.

[80] Ho, D. D., Neumann, A. U., Perelson, A. S., Chen, W., Leonard, J. M. and Markowitz, M. (1995) Nature, 373, 123-126.