Interleukin 7 Increases Human Immunodeficiency Virus Type 1 LAI-Mediated Fas-Induced T-Cell Death

Abstract

Fas-mediated T-cell death is known to occur during human immunodeficiency virus (HIV) infection. In this study, we found that HIV type 1 LAI (HIV-1_LAI) primes CD8⁺ T cells from healthy donors for apoptosis, which occurs after Fas ligation. This effect is counteracted by a broad caspase inhibitor (zVAD-fmk). Fas-mediated cell death does not depend on CD8⁺ T-cell infection, because it occurred in the presence of reverse transcriptase inhibitors. However, purified CD8⁺ T cells are sensitive to Fas only in the presence of soluble CD4. Finally, we found that interleukin 7 (IL-7) increases Fas-mediated CD4⁺ and CD8⁺ T-cell death induced by HIV-1_LAI. Since high levels of IL-7 are a marker of poor prognosis during HIV infection, our data suggest that enhancement of Fas-mediated T-cell death by HIV-1_LAI and IL-7 is one of the mechanisms involved in progression to AIDS.

During human immunodeficiency virus type 1 (HIV-1) and pathogenic simian immunodeficiency virus infections, T cells exhibit increased spontaneous apoptosis and activation-induced cell death in vitro (2, 3, 10, 16, 26, 37) and in vivo (14, 29, 32). T cells from HIV-1-infected persons show enhanced cell surface expression of Fas and increased sensitivity to Fas-mediated cell death induced either by an agonistic anti-Fas antibody or by a soluble Fas ligand (4, 11, 13, 17, 22, 31). Such cell death has been reported for CD4⁺ and CD8⁺ T cells. Whereas CD4⁺ T-cell loss is a continuous phenomenon during natural HIV infection, CD8⁺ T cells are progressively depleted during the late stages of the infection, especially when X4 HIV strains are present (39). Since a large number of investigators have suggested that CD8-mediated immune responses play a key role in controlling HIV infections, understanding the mechanisms involved in CD8⁺ T-cell depletion is of major interest. Several reports have mentioned that HIV-1 may cause CD8⁺ T-cell death in vitro (9, 18, 46). Vlahakis et al. (46) showed, using gp120, the direct killing of CD8⁺ T cells that could not be blocked by anti-Fas agonists. Esser et al. (9), using whole inactivated viruses, reported an increase in both Fas expression and cell death after 10 days of culture but did not study Fas sensitivity.

We have previously reported that the incubation of quiescent peripheral blood mononuclear cells (PBMC) from healthy donors in vitro with HIV_LAI for 5 days in medium alone, in the absence of any additional T-cell stimuli, leads to the rapid death of CD4⁺ T cells in response to subsequent antibody-mediated Fas ligation (12, 38). Here we found that HIV-1_LAI also primes unstimulated CD8⁺ T cells from healthy donors to undergo cell death in response to Fas ligation (Fig. 1A and B). A similar percentage of cells died in response to Fas ligation when PBMC were treated with a combination of the antiretroviral drugs (reverse transcriptase inhibitors) dideoxyinosine (ddI) and zidovudine (AZT) before and during incubation with HIV-1_LAI. This indicates that the priming of CD8⁺ T cells for Fas-mediated death does not require productive HIV-1 infection. To confirm that CD8⁺ T cells were not infected with HIV, we performed three-color flow cytometry, using an anti-p24 monoclonal antibody (MAb) (KC57-fluorescein isothiocyanate [FITC]; Beckman Coulter). Activated CD4⁺ T cells (Fig. 1C), but not unstimulated or activated CD8⁺ T cells, were positive for p24. In contrast to what was observed with CD4⁺ T cells (12), the proportion of CD8⁺ T cells expressing Fas, as determined by flow cytometry, did not change even when the HIV inoculum was increased (Fig. 1D). This suggests that HIV_LAI sensitizes CD8⁺ T cells to Fas-mediated death independently of its effect on Fas expression. We also found that CD8⁺ T-cell death in response to Fas ligation was dependent on caspase activation. Indeed, preincubation with the broad caspase inhibitor zVAD-fmk for 1 h prior to Fas antibody treatment significantly reduced, in a dose-dependent manner, Fas-mediated CD8⁺ T cell-death (Fig. 1E); this effect lasted for at least 36 h.

FIG. 1.

HIV-1_LAI mediates priming of quiescent CD8⁺ T cells for Fas-induced cell death. (A) PBMC were incubated for 5 days in the absence (mock, medium) or presence of HIV-1_LAI (MOI of 0.01). On day 5, cells were incubated for a further 18 h in the absence or presence of an agonistic anti-Fas IgM MAb (1 μg/ml). The percentage of dying CD8⁺ T cells was assessed by using two-color flow cytometry, using PC5-labeled anti-CD8 MAb (Beckman Coulter) and FITC-labeled annexin V (Beckman Coulter) (12). The percentages of living (CD8⁺ annexin⁻) and dying (CD8⁺ annexin⁺) CD8⁺ T cells are indicated on the dot plot. (B) PBMC were incubated as in (A) for 5 days in the absence (−) or presence (+) of HIV-1_LAI (MOI of 0.01) and in the absence (−) or presence (+) of a combination of the reverse transcriptase inhibitors ddI (0.5 μM) and AZT (1 μM), added at two time points (before addition of HIV-1_LAI and 2 days later), and further incubated for 18 h in the absence (−) or presence (+) of an agonistic Fas IgM MAb (1 μg/ml). The percentage of dying CD8⁺ T cells was calculated as follows: [CD8⁺ annexin⁺ cells/(CD8⁺ annexin⁻ cells + CD8⁺ annexin⁺ cells)] × 100. (C) PBMC were incubated with HIV-1_LAI (MOI of 0.01) and either stimulated with phytohemagglutinin (10 μg/ml) and IL-2 (10 U/ml) (Chiron Corporation) (PHA/IL-2) or left unstimulated (MEDIUM). Intracellular anti-p24 staining (FITC-labeled KC57 MAb; Beckman Coulter) was gated on CD4⁺ and CD8⁺ T cells. Three-color flow cytometry was performed, using PC5-labeled anti-CD8 and APC-labeled anti-CD4. Histograms show the percentage of p24+ cells in each population. (D) PBMC from healthy donors were incubated as in (A) in the absence (None) or presence of HIV-1_LAI at different MOIs (0.001, 0.01, or 0.1) for 5 days in medium alone. Fas expression on CD8⁺ T cells was analyzed by two-color flow cytometry using PC5-labeled anti-CD8 and FITC-labeled anti-Fas MAb (UB2; Beckman Coulter). (E) PBMC incubated with HIV_LAI (MOI of 0.01) as in (A) were further incubated in the absence (−) or presence (+) of agonistic anti-Fas MAb and in the absence (−) or presence (+) of the broad caspase inhibitor zVAD-fmk (20 μM). The percentage of dying CD8⁺ T cells was assessed as in (B). Statistical significance was assessed by the Student t test (*, P < 0.05; NS, no significant difference).

Other cell populations, namely macrophages or CD4⁺ T cells, seem to be required for HIV-mediated CD8⁺ T cell death (18, 31, 44). We next explored whether purified CD8⁺ T cells are sensitive to Fas ligation. Unlike PBMC, purified CD8⁺ T cells—obtained by negative selection using anti-CD19, -CD56, -CD4, and -CD14 MAbs (Beckman Coulter) and magnetic beads coated with antimouse immunoglobulin G (IgG) (Dynals)—incubated with HIV-1_LAI (multiplicity of infection [MOI] of 0.01) were not sensitized to Fas-mediated cell death (Fig. 2B). This suggests that the effect of HIV-1_LAI on CD8⁺ T cells depends on the presence of an additional factor(s). It has been reported that a low dose (0.1 μg/ml) of soluble CD4 alters the conformation of the gp120 envelope glycoprotein, facilitating its interaction with its chemokine receptor and subsequently the entry of virus, even in the absence of cell surface CD4 molecules (43). Thus, we incubated purified CD8⁺ T cells with different concentrations of human soluble recombinant CD4 (sCD4). Preincubation of HIV-1_LAI for 1 h with 0.1 μg of sCD4/ml primed CD8⁺ T cells for Fas-mediated cell death, whereas 10 μg/ml had no effect (Fig. 2B). sCD4 was initially reported to block viral entry when used at a concentration of 10 μg/ml (8). Fas sensitization mediated by a low dose of soluble CD4 is not associated with viral infection (Fig. 2A). As we observed that sCD4 facilitates Fas-mediated CD8⁺ T-cell death, we then assessed whether engagement of CXCR4 sensitizes CD8⁺ T cells to undergo apoptosis following Fas ligation. Using a specific anti-CXCR4 MAb (12G5; R&D System), we observed Fas-mediated CD8⁺ T-cell death (Fig. 2C). These results indicate that after 5 days in culture, whole particles of HIV-1_LAI are able to sensitize not only CD4⁺ T cells to undergo caspase-dependent death following Fas ligation, as we previously described (12, 38), but also CD8⁺ T cells after the interaction of CXCR4 with Env.

FIG. 2.

HIV-1_LAI requires soluble CD4 for Fas sensitization of quiescent purified CD8⁺ T cells. (A) CD8⁺ T cells were purified as previously reported (12) and incubated for 5 days in the absence or presence of HIV-1_LAI (MOI of 0.01). HIV-1_LAI was also pretreated for 1 h with either 0.1 μg (HIV-1_LAI/sCD4 0.1) or 10 μg (HIV-1_LAI/sCD4 10) of soluble recombinant CD4/ml. On day 5, flow cytometry was performed with FITC-labeled KC57. Histograms show the percentages of p24+ cells among CD8⁺ T cells. (B) Purified unstimulated CD8⁺ T cells were incubated (+) or not (−) with HIV-1_LAI (MOI of 0.01) with ddI (0.5 μM) and AZT (1 μM). HIV-1_LAI was pretreated (+) or not (−) for 1 h with either 0.1 μg (sCD4 0.1) or 10 μg (sCD4 10) of soluble recombinant CD4/ml. On day 5, an agonistic anti-Fas IgM MAb (1 μg/ml) was added as for Fig. 1A. Dying cells were quantified by flow cytometry using FITC-labeled annexin V. Note the increase in cell death in comparison to results with medium alone. (C) Purified unstimulated CD8⁺ T cells were incubated either with a plate-immobilized CXCR4-specific MAb (10 μg, 12G5; R&D System) or an isotype control (Ig). On day 5, the cells were incubated with an agonistic anti-Fas IgM MAb (1 μg/ml). Note the increase in cell death in comparison to results with medium alone. All results are means ± standard deviations for three independent experiments, each performed in triplicate. Statistical significance was assessed by the Student t test (*, P < 0.05).

It is generally accepted that the cytokine environment can modulate Fas-mediated cell death (11, 22, 30). In humans, interleukin 7 (IL-7) is involved in normal T lymphopoiesis but also acts as a survival factor for peripheral T lymphocytes (15, 28). IL-7 and IL-2 prevent activated T-cell death in vitro via a signal thought to be mediated by their common gamma chain receptor (45) inducing Bcl-2 expression (1). IL-2 and IL-7 are candidates or are already used for immunotherapy during HIV infection (5, 25, 36). IL-7 is a key cytokine in HIV pathogenesis and is associated with an unfavorable prognosis (27, 33, 34). Serum IL-7 concentrations are inversely correlated with CD4⁺ and CD8⁺ T-cell counts during HIV infection (6, 27, 34). Whether increased levels of IL-7 are the cause or the consequence of CD4⁺ depletion is not clear. In vitro treatment of naive T cells with IL-7 may favor the replication of X4 and R5 HIV strains (7, 41, 42) and the emergence of X4 variants (27, 40). Despite these harmful in vitro effects, exogenous IL-7 has been proposed as a therapeutic adjuvant in HIV infection to stimulate T-cell renewal (23), and IL-7 is considered a good candidate for immunotherapy for AIDS (35).

We then compared the effects of these cytokines on Fas priming of CD4⁺ and CD8⁺ T cells. IL-2 or IL-7 was added to each medium on day 1, and PBMC were incubated with HIV-1_LAI as described above in the presence of antiretroviral drugs to prevent cytokine-mediated viral replication. Our data revealed that IL-7 enhances HIV-1_LAI-induced Fas sensitization of both CD4⁺ and CD8⁺ T cells (Fig. 3A). In contrast, IL-2 reduced HIV-1_LAI-induced Fas sensitization of CD4⁺ T-cell death, whereas it had no effect on CD8⁺ T-cell death (Fig. 3A). We assessed the expression of Fas and CXCR4 by flow cytometric analysis of purified CD4⁺ and CD8⁺ T cells incubated with either IL-2 or IL-7. Our data revealed that differences in Fas sensitivity in response to IL-2 and IL-7 are related to a difference in CXCR4 but not in Fas expression (Fig. 3B). Previous studies have also shown that IL-7 sensitizes pro-B cells (24) and recent thymic emigrants (21) for Fas-mediated cell death. While IL-7 may inhibit spontaneous apoptosis by increasing the level of the antiapoptotic Bcl-2, Bcl-2 has a poor preventive effect on Fas-mediated T cell death (19, 20). Therefore, the fact that IL-7 is capable of increasing HIV_LAI-mediated Fas-induced T-cell death is not a new assertion and is not in contradiction with previous published reports on the antiapoptotic effects of this cytokine.

FIG. 3.

IL-7 but not IL-2 increases HIV-1_LAI-mediated priming of quiescent T cells for Fas-induced cell death. (A) PBMC were incubated as for Fig. 1 in the presence of HIV-1_LAI (MOI of 0.01) and with ddI (0.5 μM) and AZT (1 μM). IL-2 or IL-7 (20 ng/ml) was added to each medium, and on day 5, PBMC were further incubated with the agonistic anti-Fas IgM MAb (1 μg/ml). The percentages of dying CD4⁺ and CD8⁺ T cells were assessed by flow cytometry, using FITC-labeled annexin V and PC5-labeled anti-CD8 or CD4 MAb after 18 h and calculated as for Fig. 1B. All results are means ± standard deviations for three independent experiments, each performed in triplicate. (B) Purified CD4⁺ and CD8⁺ T cells were treated with either IL-2 (black lines) or IL-7 (gray lines) (20 ng/ml). The cells were analyzed after 5 days in culture. The expression of Fas and CXCR4 was assessed by flow cytometry, using FITC-labeled Fas MAb (Beckman Coulter) and phycoerythrin-labeled CXCR4 MAb (R&D System). An isotype control is also shown (dotted lines). Histograms are representative of three experiments that gave similar results. Statistical significance was assessed by the Student t test (*, P < 0.05).

In conclusion, our results demonstrate the role of Fas sensitization of uninfected CD8⁺ T lymphocytes during HIV infection. We also provide new data suggesting that IL-7 has a deleterious effect on HIV pathogenesis and that exogenous IL-7 must be used with caution during HIV infection.