|Home | About | Journals | Submit | Contact Us | Français|
Antigen-specific γδ T cells may play a role in anti-mycobacterial immunity. Studies done in humans and animal models have demonstrated complex patterns of γδ T cell immune responses during early mycobacterial infections and chronic tuberculosis. Recent studies have also shown a clinical correlation between major recall expansion of antigen-specific γδ T cells and immunity against fatal early mycobacterial diseases. Multiple host and microbial factors can regulate diverse immune responses of phosphoantigen-specific γδ T cells during mycobacterial infections.
While CD4 T cells have been shown to be important for immunity against tuberculosis, the role of γδ T cells in anti-mycobacterial immune responses remains poorly characterized. γδ T cells have long been considered innate-like immune cells. However, adaptive immune responses of antigen-specific γδ T cells during mycobacterial infections have recently been demonstrated in humans and monkey models. It is important to note that phosphoantigen remains the sole mycobacterial antigen recognized by a primate γδ T cell subset that expresses Vγ2 and Vδ2 TCR (Vγ2Vδ2 T cells), and that rodent γδ T cells have not been shown to recognize phosphoantigen or other mycobacterial antigens. Phosphoantigen-specific Vγ2Vδ2 T cells constitute 60–95% of total circulating human γδ T cells, implicating important and broad function of these γδ T cells in the development of immune responses against mycobacterial infections. This article reviews recent human and animal studies of γδ T cell immune responses during mycobacterial infections.
Mycobacterial infections may regulate both innate and adaptive γδ T cell immune responses. Such regulated responses of γδ T cells may be particularly evident during the early infection due to a burst of bacterial replication. While adaptive γδ T cell immune responses are regulated by TCR-dependent clonal expansion and immune memory in response to specific antigen stimulation, regulatory factors driving TCR-independent activation/expansion of γδ T cells remains poorly characterized. Interestingly, despite the absence of defined microbial antigens recognized by murine γδ TCR, mouse γδ T cells can expand during early infection with mycobacteria and other pathogens [3,6,7] (Table 1). The expansion usually peaks a week after infection and lasts for a short time. Similarly, early Mycobacterium bovis infection in ruminants can drive an increase in numbers of activated γδ T cells in both the lung and blood, although there is no apparent expansion of total circulating γδ T cells . Whether an activation/expansion of murine and bovine γδ T cells represents antigen-specific or innate-like responses remain to be determined  (Table 1). Recent identification of mycolylarabinogalactan peptidoglycan recognized by bovine γδ T cells suggests that ruminant γδ T cell responses may be induced by specific antigens in early M. bovis infection , although bovine TCR–phosphoantigen interaction has not been defined. It might also be possible that the activation of murine and bovine γδ T cells in early infection occurs as a result of TCR-independent stimulation events driven by inflammatory cytokines and other regulatory factors. This presumption is indeed supported by a recent observation that human Vδ1 T cells, which do not recognize mycobacterial phosphoantigen, can be nonspecifically activated by inflammatory cytokines in the context of the Vδ1 TCR. Other factors such as dendritic cells, adhesion (LFA3/CD2, LFA1/ICAM1), and costimulatory (MHC class I-related chain B molecule/NK-activating receptor G2D) molecules also contribute to this antigen-nonspecific activation of human Vδ1 T cells .
Adaptive immune responses of γδ T cells during mycobacterial infections have been demonstrated in macaque animal models (Table 1). While Vγ2Vδ2 T cells recognize mycobacterial phosphoantigen, Vγ2Vδ2 TCR in these γδ T cells clearly plays a role in driving clonal expansion and adaptive immune responses during mycobacterial infections. Primary M. bovis BCG infection induces major clonal expansion of phosphoantigen-specific Vγ2Vδ2 T cells [14,15]. Surprisingly, these antigen-specific γδ T cells can have immune memory and mount rapid, prolonged and high-magnitude recall expansion of Vγ2Vδ2 T cell subpopulation after BCG re-infection or Mycobacterium tuberculosis infection in monkeys previously exposed to BCG [14,15]. The memory immune responses of antigen-specific Vγ2Vδ2 T cells are also seen in human studies [17,18]. The memory-type responses can be detected in BCG-vaccinated humans, whereas tea antigen recognized by γδ T cells mediate both the innate-like and memory responses of Vγ2Vδ2 T cells in tea-drinking humans [17,18].
Adaptive immune responses of mycobacterial phosphoantigen-specific γδ T cells may contribute to immune control of bacterial burdens in early mycobacterial infections. Major expansions of Vγ2Vδ2 T cells are associated with the decline of bacterial burdens and resolution of active BCG infection in monkeys [14,15]. In addition, a rapid recall expansion of Vγ2Vδ2 T cells in the pulmonary compartment is associated with low levels of bacterial burdens and immunity against acutely fatal tuberculosis after M. tuberculosis challenge of BCG-vaccinated monkeys by aerosol . The association of Vγ2Vδ2 T cells with anti-mycobacterial immunity is also seen in the macaque models of AIDS-related tuberculosis-like disease . Nevertheless, a role of murine γδ T cells in immunity against mycobacterial infections is controversial [21,23–25]. It is argued that an absence of γδ T cell-mediated immunity in mice may be attributed to the fact that murine γδ T cells do not recognize phosphoantigen or other mycobacterial antigens.
The capacity of antigen-specific Vγ2Vδ2 T cells to mount adaptive immune response is consistent with what is found in the studies of memory/effector phenotypes of these cells in humans. It has been shown that the surface expression of CD45RA and CD27 antigens defines four subsets of Vδ2 T cells: naïve CD45RA+CD27+; memory CD45RA−CD27+; effector memory CD45RA−CD27− and terminally differentiated CD45RA+CD27− cells . Naïve CD45RA+CD27+ and memory CD45RA−CD27+ Vδ2 T cells express lymph node-homing receptors and predominate in lymph nodes; whereas effector memory CD45RA−CD27− and terminally differentiated CD45RA+CD27− Vδ2 T cells are mainly present in the circulation and inflamed tissues, respectively . Interestingly, within effector memory subset, Vδ2 T cells can be functionally grouped into IFNγ/TNFα-producing effectors and cytotoxic effectors. The IFNγ/TNFα-producing effectors do not express CD16 but express high levels of chemokine receptors and produce large amounts of IFNγ/TNFα in response to phosphoantigens. In contrast, cytotoxic effectors are CD16+ with NK-like phenotypes and highly active against tumor cell lines, but they are refractory to phosphoantigen-mediated production of IFNγ and TNFα . Consistently, macaque Vγ2Vδ2 T cells that do not express CD45RA or CD27 are predominated in the blood circulation and spleen but not in the lymph nodes after mycobacterial infection (data not shown). Clonotypic TCR analyses indicate that the γδ T cells with those memory phenotypes have undergone recall expansion after mycobacterial infections (data not shown and ). The data suggest that mycobacterial infection regulates both memory/effector phenotypes and adaptive immune function of phosphoantigen-specific Vγ2Vδ2 T cells.
While virulence or pathogenicity of mycobacterial strains can impact on the immune system and consequences of infection, infection dose and route have not been assessed for their capacity to regulate immune responses of antigen-specific γδ T cells in humans. Recent studies in monkey models of mycobacterial infections provide evidence that mycobacterial infection dose and route play a role in regulating γδ T cell immune responses. It has been shown that the systemic BCG infection introduced by intravenous inoculation with 108 BCG organisms results in up to 5-fold expansion of Vγ2Vδ2 T cells in the blood . In contrast, a systemic BCG infection generated by 103 BCG organisms or a pulmonary infection by bronchoscope-guided inoculation induces only subtle changes in numbers of circulating Vγ2Vδ2 T cells. On the other hand, pulmonary exposure to BCG through the bronchial route induces detectable expansions of CD4+, CD8+, and Vγ2Vδ2 T cells only in the lung but not in the blood. Similarly, a progressive increase in pulmonary M. tuberculosis burden drives a linear expansion of Vγ2Vδ2 T cells in the lung but not in the blood of naïve monkeys after M. tuberculosis infection by aerosol (data not shown and ). In humans, systemic infections, which may represent high microbial burdens, clearly drive major clonal expansions of Vγ2Vδ2 T cells. Disseminated M. avium infection can stimulate Vγ2Vδ2 T cell expansion to the extent of 40% of total circulating T cells . Systemic infections with other phosphoantigen-producing pathogens such as Malaria, H. influenzae, N. meninggitidis, S. pneumoniae, and F. tularensis also coincide with a profound expansion of Vγ2Vδ2 T cells . The infection dose- and route-dependent responses of Vγ2Vδ2 T cells may help to explain why some individuals with pulmonary tuberculosis exhibit no detectable expansion of Vγ2Vδ2 T cells in the blood circulation [15,29].
Chronically active tuberculosis appears to down-regulate Vγ2Vδ2 T cell responses rather than drive major expansion of these cells. While an expansion of antigen-specific Vγ2Vδ2 T cells is seen in the lung of monkeys after M. tuberculosis infection by aerosol , γδ T cell responses in humans during early M. tuberculosis infection remains largely inconclusive . The absence of definitive information regarding an early human Vγ2Vδ2 T cell expansion may be attributed to a clinical difficulty for early sampling, since tuberculous patients would not visit clinics until they had onset of tuberculosis after infection. Studies done at a chronic phase of tuberculosis have reported a decrease in percentage of Vγ2Vδ2+ T cells in both the blood and BAL fluid of patients with pulmonary tuberculosis when compared with normal individuals [30,31]. It has also been shown that patients with active pulmonary tuberculosis exhibit an impaired ability of Vγ2Vδ2+ T cells to produce cytokines or proliferate in response to in vitro stimulation with phosphoantigens [30,32]. The down-regulation or impaired function of Vγ2Vδ2 T cells appear to be consistent with an absence or anergy of CD4+ T cell responses and PPD skin test reactivity identified in patients with active tuberculosis [33,34].
The reduced number and function of Vγ2Vδ2 T cells seen in tuberculous patients may be regulated through complex immune mechanisms (Fig. 1). Irreversible one-way pulmonary migration of activated Vγ2Vδ2 T cells may occur in tuberculosis and contribute to reduced numbers of these cells in the circulation. In fact, it has been shown that pulmonary M. tuberculosis leads to an increase in numbers of macaque Vγ2Vδ2 T cells in the lung and an associated decrease in circulating Vγ2Vδ2 T cells (Dan et al., data not shown and ). Since activated Vγ2Vδ2 T cells express high levels of chemokine receptors [20,35,36], the chemokine/receptor-mediated chemostasis may facilitate the pulmonary migration of Vγ2Vδ2 T cells. Given the possibility that adult pulmonary tuberculosis usually occurs as a result of a loss of immunity against latent infection or re-infection, there may be a fundamental immune suppression or defect of immune cells including Vγ2Vδ2 T cells in patients with tuberculosis. Such an immune suppression or defect may help to explain the reduced level of Vγ2Vδ2 T cell proliferation in response to in vitro antigen stimulation [32,37]. Changes in cytokine environments during tuberculosis may also contribute to the down-regulation of Vγ2Vδ2 T cells. Furthermore, the defect in effector function of Vγ2Vδ2 T cells in tuberculous patients may result from chronic activation and exhaustion of these cells. This possibility is supported by a recent finding that patients with active tuberculosis display a decrease in numbers of CD45RA−CD27− Vγ2Vδ2 T cells and a loss of in vitro effector function in responses to phosphoantigens, suggesting that these cells are pre-terminally activated γδ T cells [32,37]. Further studies are needed to elucidate the precise mechanisms by which tuberculosis induces down-regulation or defect of Vγ2Vδ2 T cells.
The possibility that CD4 T effector cells and HIV can function as important host and microbial factors for immune regulation of antigen-specific Vγ2Vδ2 T cell responses during active M. tuberculosis infection has not been studied in humans. Since the hallmark of HIV infection is CD4 T cell depletion and global immune suppression, it is important to determine the extent to which HIV inhibits adaptive immune responses of Vγ2Vδ2 T cells during active mycobacterial coinfection of HIV-infected humans. It is also important to determine whether immune reconstitution during anti-retroviral treatment can restore potentially inhibited responses of antigen-specific Vγ2Vδ2 T cell responses during active mycobacterial coinfection of HIV-infected individuals. SIVmac-infected monkeys provide a good system in which to address these questions. While SIVmac-infected macaques with high viral loads exhibit inhibited responses of Vγ2Vδ2 T cells in the blood and lungs after BCG infection and reinfection, CD4 T cells are needed for the in vitro restoration of the ability of Vγ2Vδ2 T cells to proliferate in response to phosphoantigen . Tenofovir or tenofovir + indinavir antiretroviral treatment of SIVmac-infected monkeys can restore the capacity of Vγ2Vδ2 T cells to undergo major expansions and pulmonary migration in active BCG re-infection. Importantly, the restored expansion of Vγ2Vδ2 T cells coincides with increases in numbers of PPD-specific IFNγ-producing CD4 T cells and increases in the magnitude of their proliferative responses (Fig. 2 and ). Such a restored Vγ2Vδ2 T cell response during antiretroviral treatment of SIVmac-infected monkeys is mounted on a polyclonal γδ T cell repertoire background. These results suggest that the development of adaptive immune responses of phosphoantigen-specific Vγ2Vδ2 T cells in AIDS requires control of viral infection and immune competence of peptide-specific CD4 T cells during early mycobacterial infection. The findings in SIV-infected monkeys are indeed supported by in vitro studies in HIV-infected humans who received antiretroviral drugs. Effective antiretroviral treatment in HIV-infected humans has been shown to restore the TCR repertoire  and the ability of Vγ2Vδ2+ T cells to produce IFN-γ in response to in vitro stimulation with phosphoantigen [40,41]. The monkey and human studies suggest that CD4 T helper function and HIV coinfection are important host or microbial factors that impact on the development of Vγ2Vδ2+ T cell immune responses during active mycobacterial infections.
Given the possibility that effective T cell immune responses during early mycobacterial infection impact on initial bacterial burdens and subsequent disease outcomes, restored immune responses of T cells including γδ T cells may contribute to immunity against AIDS-related tuberculosis after mycobacterial coinfection or re-infection. Interestingly, restored immune responses of antigen-specific Vγ2Vδ2 T cells and CD4 T cells during antiretroviral treatment of SIV/BCG-coinfected monkeys are associated with immunity against fatal SIV-related tuberculosis-like diseases (Shen et al., data not shown). In contrast, inhibited responses of Vγ2Vδ2 T cells and CD4 T cells during active BCG coinfection coincide with the development of the fatal SIV-related tuberculosis-like disease in early SIVmac-infected monkeys or in naïve monkeys simultaneously coinfected with SIV and BCG [38,42]. The finding that AIDS-related inhibition of Vγ2Vδ2 T cell responses in active mycobacterial coinfection is restorable during antiretroviral treatment may have potential clinical implications. Vγ2Vδ2 T cells may emerge as important anti-microbial effector cells in AIDS patients whose CD4 T cell counts remains low during HAART. Since many microbes produce phosphoantigen, Vγ2Vδ2 T cells may confer broad immunity against opportunistic infections in AIDS patients.
Progress has been made about the nature and regulation of antigen-specific γδ T cell responses in mycobacterial infections. The unique ability of Vγ2Vδ2 T cells to mount both innate-like and adaptive immune responses may allow these γδ T cells to respond rapidly to phosphoantigen-producing mycobacteria and other pathogens. Further studies are clearly needed to define a precise role of Vγ2Vδ2 T cells in immunity to mycobacterial infections as well as understand immuno-pathogenesis of these antigen-specific γδ T cells in tuberculosis.
The author would like to thank members at the Chen’s laboratory for discussions and suggestions. The studies in nonhuman primates were supported by NIH RO1 grants RR13601 (to ZWC) and HL64560 (to ZWC) and by the Pittsfield Anti-Tuberculosis Association (to ZWC).