Interferon‐γ–Mediated Antiviral Immunity against Orthopoxvirus Infection Is Provided by γδ T Cells

To the Editor —Abate et al. [1] reported increases in the numbers of vaccinia virus (VV)–responding CD4+ and CD8+ T cells in recipients of smallpox vaccine, compared with those in nonvaccinated control subjects. The population of interferon (IFN)–γ–producing γδ T cells was also expanded in vaccinated persons, suggesting that, in addition to CD4+ and CD8+ αβ T cells, γδ T cells might be involved in orthopoxvirus-induced immune responses These data confirm our previous observation that VV-specific CD4+ and CD8+ T cell recall responses are detectable by flow cytometry after recent smallpox vaccination [2]. Recently, we have observed that VV is also able to induce a substantial in vitro expansion of the γδ T cell population in peripheral-blood mononuclear cell samples obtained from nonvaccinated volunteers (2.5–10-fold increases in γδ T cell numbers were observed after 7 days of culture in the presence of VV in 4 of 5 nonvaccinated donors). Previously, substantial numbers of VV-specific γδ T cells have been observed in uninfected mice [3], and γδ T cells from a high proportion of apparently naive cattle could be activated by modified VV Ankara [4]. Cumulatively, these findings suggest that VV may induce γδ T cell activation through innate pathways of immune recognition In our study, the VV-expanded human γδ T cell population displayed the Vγ9Vδ2 T cell receptor (TCR) (the vast majority of Vδ2 chains form the Vγ9JpC1γ/Vδ2DJ1Cδ combination—i.e., the Vγ9Vδ2 TCR) [5]. Similar expansions of the Vγ9Vδ2 T cell subset have been noted in persons receiving live recombinant canarypox virus vaccination [6]. Proliferative, cytotoxic, and cytokine responses of the Vγ9Vδ2 T cell subset are constitutively induced by both nonpeptidic antigens, such as pyrophosphomonoesters and alkylamines, and nitrogen-containing bisphosphonates (N-BPs), such as zoledronic acid (ZA) [7, 8]. Moreover, Vγ9Vδ2 T cells are known to exert broad antiviral activity against different viruses through cytolytic and noncytolytic mechanisms that include the release of soluble γδ factors (GDFs) [9, 10]. Therefore, we tested the possible VV inhibitory activity of GDFs in vitro. Specifically, GDFs released by Vγ9Vδ2 T cells were collected after 24 h of stimulation with either isopentenyl-pyrophosphate (IPP) (one of the well-studied “nonpeptidic antigens”) or ZA. The antiviral activity of the GDFs was assessed using the A549 cell line infected with VV. Because the supernatants of IPP-stimulated γδ T cell cultures usually contain relevant amounts of IFN-γ (range, 50–500 IU/mL), neutralization experiments were performed using an IFN-γ–specific polyclonal rabbit antiserum (final concentration, 10,000 neutralization units/mL). Our data indicate that (1) the GDFs from the IPP- or ZA-stimulated γδ T cells, but not from the control γδ T cells, inhibited VV replication and (2) the neutralizing antibodies against IFN-γ virtually abrogated the virus inhibitory activity of the GDFs present in the supernatants from the stimulated Vγ9Vδ2 T cell cultures (figure 1). These findings suggest that the VV inhibitory activity of GDFs from in vitro–stimulated γδ T cells is largely mediated through IFN-γ, confirming a main role for this cytokine in the γδ T cell response to VV [1]. These data are also in agreement with those from a study showing that augmented γδ T cell responses in recipients of canarypox vaccine are mediated primarily by IFN-γ–producing Vγ9Vδ2 T cells [6]. It is noteworthy that, despite substantial differences between mouse and human γδ T cells, VV-infected mice deficient in γδ T cells have significantly higher VV titers and mortality rates than do normal mice [3] and that rapid and substantial increases in IFN-γ–producing γδ T cells occur in αβ-TCR–deficient mice that survive VV infection [3] Vaccinia virus (VV) inhibitory activity of γδ factors (GDFs). For the experiments shown in panel A, Vδ2 T cell lines (>80% Vδ2+ cells) were obtained by stimulating peripheral-blood mononuclear cells from healthy donors with isopentenyl-pyrophosphate (IPP; 10 μmol/L) and interleukin-2 (100 IU/mL) for 12 days. The Vδ2 T cell lines were then restimulated with either IPP (10 μmol/L) or zoledronic acid (ZA; 2 μmol/L) for 24 h, and the culture supernatants that contained GDFs (i.e., factors released by activated γδ T cells) were diluted 1:2 and added to A549 cell cultures for an additional 24 h. The A549 cell cultures were then washed and infected with VV at an MOI of 0.01. Seventy-two hours after infection, the A549 cell culture supernatants were collected, and the numbers of infectious VV particles were titrated by limiting dilution assay on uninfected target cells and expressed as the TCID50 per milliliter. Panel B shows the results of the cytokine neutralization assay, which was performed using polyclonal rabbit anti–interferon (IFN)–γ antiserum, as described elsewhere [11]. ctrl, control. *P<.05 (statistically significant difference by Student’s t test), for the comparison of GDFs IPP or GDFs ZA with GDFs ctrl (factors released without activating γδ cells) or with VV ctrl (VV-infected control cultures) in panel A, and for the comparison of GDFs IPP with or without anti–IFN-γ antiserum in panel B These observations emphasize the importance of IFN-γ–producing γδ T cells in innate resistance to VV infection. Moreover, it is likely that the increased IFN-γ production by γδ T cells boosts the induction of protective type 1 memory immunity and augments the effectiveness of live vaccines. The human Vγ9Vδ2 T cell subset can be activated in vivo by the administration of nonpeptidic antigens or N-BPs [9, 10]. N-BPs are widely used to treat bone-demineralization disorders and certain cancers [8]. The low toxicity of N-BPs, combined with the current lack of effective smallpox therapies, suggests that the use of N-BP–induced activation of Vγ9Vδ2 T cells for possible postexposure prophylaxis and/or treatment of orthopoxvirus infections should be explored