Interleukin 7 reduces the levels of spontaneous apoptosis in CD4⁺ and CD8⁺ T cells from HIV-1-infected individuals

Abstract

Apoptosis has been suggested as one of the major mechanisms of CD4⁺ T cell depletion during the course of HIV type 1 (HIV-1) infection. Here, we show that interleukin 7 (IL-7), a nonredundant cytokine that plays essential roles in the generation and homeostasis of the T cell compartment of the immune system, exerts strong antiapoptotic effects ex vivo on both CD4⁺ and CD8⁺ T cells derived from HIV-1-infected subjects. The level of IL-7-mediated reduction of apoptosis was inversely correlated with the number of circulating CD4⁺ T cells, indicating a higher sensitivity to IL-7 effects in patients with more advanced disease. The antiapoptotic effect of IL-7 was uncoupled from the induction of cellular proliferation or endogenous HIV-1 replication. These results provide a further rationale for consideration of IL-7 as an agent of immune reconstitution in HIV-1 infection.

Infection with HIV type 1 (HIV-1) causes a progressive depletion of CD4⁺ T cells that ultimately leads to AIDS (1). Several mechanisms have been implicated in the loss of CD4⁺ T cells, including direct virus-induced cytolysis, defective T cell regeneration, anergy, and apoptosis (2). Both CD4⁺ and CD8⁺ T cells derived from HIV-1-infected individuals show an increased propensity to undergo spontaneous apoptosis both in vivo (3–5) and after short-term ex vivo culture (6). Remarkably, the vast majority of the cells that undergo apoptosis are uninfected (5). Similar observations were made in macaques infected with simian immunodeficiency viruses (SIV) (7, 8). The level of apoptosis in HIV-1-infected individuals was shown to correlate with the levels of circulating CD4⁺ T lymphocytes and stage of disease (9, 10), reinforcing the concept that apoptosis may be one of the primary mechanisms of CD4 depletion in HIV infection. Various mechanisms have been invoked to account for the spontaneous apoptosis observed during the course of HIV-1 infection, including sustained immune activation associated with dysregulated cytokine production (10, 11), inappropriate signaling mediated by HIV-1 envelope binding to CD4 (12) or coreceptors (13), defective antigen presentation (14), activation of death receptors (15, 16), and loss of extracellular survival signals (17).

The increase in apoptosis levels seen in HIV-1-infected individuals implies that therapeutic interventions aimed at reducing apoptosis also may decrease the rate of T cell depletion. For example, interleukin 2 (IL-2), a cytokine that induces a marked increase in the number of total and naïve CD4⁺ T cells in HIV-1-infected patients (18), was shown to reduce the levels of T cell apoptosis both in vivo (19) and ex vivo (20); the long-term benefits of this therapeutic regimen are being determined in clinical trials. Another cytokine that is under evaluation as a potential agent of immune reconstitution in HIV-1 infection is IL-7, a key factor in the generation, activation, and homeostasis of the T cell compartment of the immune system (21). The prosurvival effects of IL-7 are mediated primarily by up-regulation of Bcl-2 synthesis and inactivation of proapoptotic proteins like Bad and Bax (22). Conflicting evidence has been reported on the antiapoptotic effects of IL-7 on T cells from HIV-1-infected individuals studied after short-term ex vivo culture. Zaunders et al. (23) found no effects of IL-7 on peripheral blood mononuclear cells (PBMC) from patients with acute primary HIV-1 infection, whereas others have reported a limited antiapoptotic effect in patients with chronic HIV-1 infection, which was less pronounced than in uninfected individuals (24, 25). Likewise, a limited, but not statistically significant, effect of IL-7 was reported on CD4⁺ and CD8⁺ T cells from macaques chronically infected with SIV (8). Another controversial issue is the putative ability of IL-7 to induce HIV-1 replication. At doses that promote T cell proliferation, IL-7 has been reported to induce the transcription of latent HIV-1 in resting peripheral blood CD4⁺ T cells derived from infected patients on antiretroviral therapy cultured for several weeks ex vivo (26). However, in vivo administration of IL-7 to SIV-infected macaques did not cause a detectable increase in viral load both in lymph nodes and in peripheral blood, despite a widespread induction of T cell proliferation (27, 28).

In the present study, we investigated the protective effects of IL-7 against spontaneous T cell apoptosis in a cohort of HIV-1-infected subjects at different stages of disease. An extensive time-course analysis over 7 days of ex vivo culture permitted us to document a potent antiapoptotic effect of IL-7 on both CD4⁺ and CD8⁺ T cells from HIV-1-infected individuals. This effect was not associated with T cell proliferation or reactivation of latent provirus.

Results

IL-7 Reduces Spontaneous Apoptosis in PBMC from HIV-1-Infected Subjects.

The effects of IL-7 initially were evaluated in unfractionated PBMC from a cohort of HIV-1-infected subjects (n = 24) at different stages of disease as reflected by levels of peripheral CD4⁺ T cell counts and HIV-1 plasma viremia [nos. 1–24, supporting information (SI) Table 2]. The cells were cultured ex vivo in the presence or absence of recombinant human IL-7; IL-2 was tested in parallel as a positive control for the reduction of apoptosis. Apoptosis was evaluated on freshly isolated cells (baseline) and then every day for 7 days by using two unrelated techniques: annexin V binding and caspase 3 activation. To determine most accurately the effect of IL-7 on apoptosis over the 7-day period, we calculated the mean of all daily apoptosis measurements for each patient, and then from those data we determined the mean value for the 24 patients.

The proportion of annexin V-binding cells at baseline was similar in HIV-1-infected patients and controls (P = 0.61); however, it significantly increased upon ex vivo culture in HIV-1-infected subjects (P < 0.0001), whereas the increase was only marginal in seronegative controls (P = 0.52) (Fig. 1A). The addition of exogenous IL-7 induced a dramatic reduction in the levels of spontaneous apoptosis in cells from all of the HIV-1-infected patients tested. The mean level of annexin V binding over the 7 days of culture was significantly lower in cells treated with IL-7 at 5 ng/ml than in untreated homologous cells (P < 0.0001) (Fig. 1A). At each time point between days 1 and 6, annexin V binding was reduced significantly in IL-7-treated cells compared with controls; the difference was no longer significant at day 7 (Fig. 1B). In PBMC from selected patients, IL-7 was titrated over a wide dose range (0.04–50 ng/ml). Reduction of apoptosis consistently was seen at doses higher than 0.5 ng/ml (data not shown).

Fig. 1.

Protective effects of IL-7 against spontaneous apoptosis in ex vivo-cultured PBMC from HIV-1-infected patients and uninfected controls. (A) Mean levels of annexin V binding at baseline (day 0) and over 7 days of ex vivo culture in the presence or absence of IL-7 at 5 ng/ml. C, untreated controls. (B) Time-course analysis of IL-7-mediated reduction of annexin V binding over 7 days of ex vivo culture in PBMC from HIV-1-infected patients. Reduction of apoptosis by IL-7 was calculated by subtraction of the percentage of annexin V⁺ cells in IL-7-treated cultures from the percentage of annexin V⁺ cells in untreated controls. ∗, P < 0.0001 for the comparison between IL-7-treated and untreated cultures; †, P < 0.005; ‡, not significant. The comparisons were performed by paired Student's t test. (C) Time-course analysis of annexin V binding and caspase 3 activation in PBMC from a representative HIV-1-infected patient cultured in the presence or absence of IL-7 at 5 ng/ml.

The antiapoptotic effect of IL-7 also was investigated by measuring the activation of caspase 3. Overall, the mean levels of caspase 3 activation during the 7 days of culture were significantly lower in IL-7-treated than in untreated cells (n = 7; P = 0.01). The levels of caspase 3 activation paralleled those of annexin V binding in the presence and absence of IL-7 (Fig. 1C). In line with previous observations (29), the reduction of apoptosis mediated by IL-7 was associated with an increase in intracellular levels of Bcl-2 (data not shown).

Despite the consistency of the antiapoptotic effect of IL-7, its kinetics and magnitude showed a marked variability among patients. The peak reduction in the proportion of apoptotic cells in different patients ranged from 5.1% to 28.4% (mean, 14.1 ± 5.7%) and occurred at different time points during ex vivo culture (range, 1–6 days; mean, 3.9 ± 1.5). In some patients, the effect of IL-7 increased over time (positive slope), whereas in others, it remained constant or even decreased (negative slope). Z score values comparing each patient's slope to the overall slope confirmed that such differences were real (P < 0.0001). The fact that patients had relatively similar baseline values but different trajectories over time explains the increased variability in the last days of culture (Fig. 1B; also confirmed by Pitman's test for paired variances).

As expected, exposure of PBMC from HIV-1-infected individuals to IL-2 (100 units/ml) effectively reduced the levels of spontaneous apoptosis with peak reductions ranging from −0.9% to 31.0% (mean, 16.6 ± 7.1%); the peak effect of IL-2 tended to occur slightly earlier than that of IL-7 (range, 1–4 days; mean, 2.3 ± 1.1) (data not shown).

Limited Antiapoptotic Effects of IL-7 on PBMC from HIV-Seronegative Individuals.

Treatment with IL-7 had limited effects in PBMC from a group of age-matched HIV-uninfected subjects (n = 14). The mean levels of spontaneous apoptosis during the 7-day culture period were not significantly different in IL-7-treated and untreated cultures (Fig. 1A), and comparisons at each time point showed a significant apoptosis reduction only at day 2 (P = 0.002). Thus, the mean effect of IL-7 over 7 days of culture was significantly different in cells from HIV-1-infected and uninfected subjects (mean, 8.2 ± 5.4% vs. 0.6 ± 3.3%; P < 0.0001) (Fig. 1A); a similar difference was seen considering the peak levels of apoptosis reduction (mean, 14.1 ± 5.7% vs. 5.3 ± 4.1; P < 0.0001). Analogous results were obtained in IL-2-treated cells with regard to both the mean (−2.0 ± 5.2% vs. 7.6 ± 8.9%; P = 0.002) and peak (6.5 ± 8.9% vs. 16.6 ± 7.1%; P = 0.01) levels of reduction of apoptosis (data not shown).

The Sensitivity to the Antiapoptotic Effect of IL-7 Correlates with the CD4⁺ T Cell Count.

The propensity to undergo spontaneous apoptosis in HIV-1-infected subjects previously was shown to correlate directly with the plasma viral load and inversely with the levels of circulating CD4⁺ T cells (30, 31). Thus, we evaluated the potential correlations between the sensitivity to the antiapoptotic effects of IL-7 and several demographic, clinical, and immunologic parameters at the time of sampling (see SI Tables 2 and 3). No significant correlations were observed between the mean levels of apoptosis reduction by IL-7 over the first 6 days in culture and the plasma HIV-1 load, CD8 count, and age (Fig. 2Left) as well as the expression of CD8⁺ T cell naïve/memory markers, CD8⁺ T cell activation markers (HLA-DR⁺, CD38⁺, CD38⁺HLA-DR⁺, CD25⁺), selected CD4⁺ T cell activation markers (HLA-DR⁺, CD38⁺HLA-DR⁺, CD25⁺), and treatment status (data not shown). In contrast, a significant inverse correlation was observed with the circulating CD4 count (R² = 0.266, P = 0.0099) (Fig. 2 Left). For selected parameters, these findings were confirmed by comparing the levels of reduction of apoptosis between different groups of patients defined by specific cut-off values (Fig. 2 Right). Moreover, the mean levels of reduction of apoptosis by IL-7 over the first 6 days in culture were inversely correlated with the absolute number of memory CD4⁺ T cells (R² = 0.229, P = 0.018), as well as with the proportion of CD4⁺ T cells expressing the activation marker CD38 (R² = 0.221, P = 0.020) (data not shown). The CD4 counts also were predictive of the slopes of IL-7-mediated reduction of apoptosis over time, with a greater increase in IL-7 effectiveness over time in patients with lower CD4 counts (see SI Fig. 5).

Fig. 2.

Correlation between the mean levels of IL-7-mediated reduction of apoptosis in PBMC from 24 HIV-1-infected patients over the first 6 days in culture and various demographic, clinical, and immunologic parameters. (Left) Scatterplots with regression lines relating differences in apoptosis to covariates. Also shown are R² values (squares of the Pearson's correlation coefficients) and P values. (Right) Comparisons between different groups of patients defined by specific cut-off values for each parameter. Apoptosis was measured by annexin V binding.

IL-7 Exerts Antiapoptotic Effects on both CD4⁺ and CD8⁺ T Cells from HIV-1-Infected Subjects.

As a preliminary approach to elucidate whether IL-7 can protect from apoptosis both CD4⁺ and CD8⁺ T cells derived from HIV-1-infected individuals, we analyzed separately the two subpopulations by multiple color cytofluorimetry within unfractionated PBMC cultures. The results indicated that IL-7 treatment reduced the level of annexin V binding in both CD3⁺CD4⁺- and CD3⁺CD8⁺-gated cells (data not shown). To address this issue more formally, we enriched CD4⁺ and CD8⁺ T cell subpopulations by negative selection from PBMC of five HIV-1-infected individuals, and then we cultured the cells for 7 days in the presence or absence of IL-7 at 5 ng/ml. IL-7 had strong antiapoptotic effects on both CD4⁺ and CD8⁺ T cells from all of the HIV-1-infected individuals analyzed (Fig. 3): the peak levels of reduction of apoptosis measured by annexin V binding ranged from 10.2% to 40.5% for CD4⁺ T cells (mean, 20.8 ± 13.1%; P = 0.02 for the comparison between IL-7-treated and untreated cells) and from 15.5% to 52.8% for CD8⁺ T cells (mean, 27.2 ± 16.2%; P = 0.02). Likewise, there was a marked reduction in the mean levels of apoptosis over the first 6 days of culture in both CD4⁺ T cells (11.7 ± 6.6%; P = 0.02) and CD8⁺ T cells (14.6 ± 9.1%; P = 0.02). Similar data (not shown) were obtained when caspase 3 activation was used as an indicator of apoptosis. Representative time courses of annexin V binding and caspase 3 activation in purified CD4⁺ and CD8⁺ T cells in the presence or absence of IL-7 are presented in SI Fig. 6.

Fig. 3.

Protective effects of IL-7 against spontaneous apoptosis in ex vivo-cultured purified populations of CD4⁺ and CD8⁺ T cells from 5 HIV-1-infected patients. Mean levels of annexin V binding over 6 days of culture in the presence or absence of IL-7 (5 ng/ml) in purified CD4⁺ (A) or CD8⁺ (B) T cell populations.

When naïve and memory CD4⁺ and CD8⁺ T cells were analyzed separately either on unfractionated PBMC cultures or on purified CD4⁺ and CD8⁺ T cells by multiple color cytofluorimetry, IL-7 was shown to reduce apoptosis in both the naïve and memory subsets (SI Fig. 7). As expected, treatment with IL-2, used as a positive control, reduced the levels of spontaneous apoptosis in both CD4⁺ and CD8⁺ T cells from HIV-1-infected individuals (data not shown).

The Antiapoptotic Effect of IL-7 Is Uncoupled from the Induction of Cellular Proliferation.

Because IL-7 exerts concentration-dependent proliferative effects on CD4⁺ and CD8⁺ T cells (32), we measured the fraction of cells undergoing cycle progression and proliferation in IL-7-treated cells from 10 selected patients by using three different methods: expression of the nuclear antigen Ki67, dilution of the vital dye carboxyfluorescein diacetate succinimidyl ester (CFSE), and absolute cell counting. Although IL-7 at 5 ng/ml induced cellular proliferation in PBMC from all of the patients analyzed, we consistently observed a temporal dissociation with the antiapoptotic effect. As illustrated in Fig. 4, IL-7 induced an antiapoptotic effect from the earliest days of culture; in contrast, Ki67 expression appeared at a later time (day 5) and was seen in a high proportion of cells only at 5 ng/ml of IL-7. At lower IL-7 concentrations, Ki67 was expressed in <10% of the cells, with negligible levels in cells treated with 0.6 ng/ml. Analogous results were obtained by analysis of CFSE dilution over time (SI Fig. 8), as well as by absolute cell counting (data not shown). Unlike IL-7, IL-2 at a concentration of 100 units/ml induced an early proliferative effect that first was detectable at day 3 of culture and peaked at day 6 (Fig. 4). However, even with IL-2, we observed a temporal dissociation between reduction in apoptosis and Ki67 expression because its antiapoptotic effect was already evident at day 1 of culture and peaked at day 2 (Fig. 4).

Fig. 4.

Temporal dissociation between the antiapoptotic and proliferative effects of IL-7 and IL-2. Time-course analysis of annexin V-binding reduction (A) and Ki67 expression (B) in PBMC from a representative HIV-1-infected patient (no. 14) cultured for 6 days in the presence or absence of IL-7 (0.6, 1.2, 2.5, and 5 ng/ml) or IL-2 (100 units/ml).

IL-7 Does Not Induce Endogenous HIV-1 Replication in Purified CD4⁺ T Cells from HIV-1-Infected Individuals.

Ex vivo treatment of purified resting CD4⁺ T cells from a fraction of HIV-1-infected patients with high doses of IL-7 has been reported to induce the activation of latent provirus after several weeks in culture (26, 33). Thus, we measured the levels of endogenous HIV-1 replication over time in purified CD4⁺ T cells derived from seven HIV-1-infected individuals, cultured in the presence or absence of IL-7 at 5 ng/ml or IL-2 at 100 units/ml. Table 1 shows that supernatants from five of seven CD4⁺ T cell cultures treated with IL-7 remained negative for HIV-1 p24 throughout the culture period, whereas the remaining two showed only minimal levels of p24 release (0.01 ng/ml). Of note, similar levels of p24 also were seen in IL-7-untreated cultures at approximately the same time points, suggesting that HIV-1 replication was unlikely a result of IL-7 treatment and actually represented spontaneous release of virus. Treatment with IL-2 at 100 units/ml resulted in the appearance of low levels of p24 (0.01–0.142 ng/ml) in CD4⁺ T cell cultures from four of the seven patients, including one (no. 29) in which IL-7 did not induce any detectable viral replication, whereas IL-2 yielded the highest levels of p24 release. Analogous data (not shown) were obtained by using unfractionated PBMC followed for up to 21 days after establishment in culture.

Table 1.

HIV-1 replication in purified CD4⁺ T cells derived from selected HIV-1-infected subjects, cultured ex vivo for 6 days in the presence of IL-7 (5 ng/ml) or IL-2 (100 units/ml)

Patient no.	Days in culture	HIV-1 p24 release (ng/ml)
		Control	IL-7	IL-2
5	2	0	0 (7.1)	0 (7.5)
	3	0	0 (9.2)	0 (9.3)
	4	0	0 (9.7)	0 (9.2)
	5	0	0 (11.6)	0 (7.5)
	6	0	0 (9.3)	0 (2.9)
10	2	−	−	−
	3	0	0 (−)	0 (−)
	4	−	−	−
	5	−	−	−
	6	0	0 (9.8)	0 (9.2)
25	2	0	0 (0.2)	0 (−4.2)
	3	0	0 (2.1)	0 (3.1)
	4	0	0 (14.5)	0 (8.5)
	5	0	0 (19.7)	0.05 (−)
	6	0	0 (27.9)	0 (19.0)
26	2	0	0 (1.9)	0.04 (1.0)
	3	0	0 (11.0)	0.04 (7.6)
	4	0	0 (8.4)	0 (6.9)
	5	0	0 (9.8)	0 (7.9)
	6	0	0 (13.9)	0 (15.5)
27	2	0	0 (18.3)	0 (7.3)
	3	0	0 (19.1)	0 (31.5)
	4	0	0 (29.9)	0 (39.7)
	5	0.01	0.01 (23.9)	0 (31.0)
	6	0.01	0 (40.5)	0 (49.7)
28	2	0	0.01 (9.2)	0.01 (6.3)
	3	0.01	0.01 (10.2)	0.01 (10.8)
	4	0.01	0.01 (6.3)	0.01 (5.9)
	5	0	0.01 (8.1)	0.01 (10.3)
	6	0.01	0.01 (8.5)	0.01 (11.4)
29	2	0	0 (2.5)	0 (−0.41)
	3	−	−	−
	4	0	0 (−)	0 (−)
	5	−	−	−
	6	0	0 (0)	0.142 (7.7)

Reduction of apoptosis at the same time point is shown in parentheses. −, not tested.

Discussion

Although the introduction of effective antiretroviral therapy (ART) has resulted in a dramatic decline in the morbidity and mortality among HIV-1-infected individuals, phenotypic and functional immunologic abnormalities may persist even in patients with sustained suppression of viremia for several years (34). Thence, innovative approaches aimed at improving immune reconstitution have been proposed as a complement to ART. In particular, cytokines that play an essential role in T cell homeostasis and proliferation, such as IL-2 and IL-7, currently are under clinical investigation (18). In this study, we demonstrated that IL-7 exerts protective effects against the spontaneous apoptosis of both CD4⁺ and CD8⁺ T cells derived from HIV-1-infected individuals. In agreement with previous reports (6, 9, 35), we found that the level of spontaneous apoptosis was significantly higher in HIV-1-infected subjects than in seronegative controls and that, among the former, it correlated with the degree of CD4⁺ T cell depletion in vivo, reinforcing the concept that apoptosis may be a key mechanism of T cell destruction during HIV-1 infection. Thus, the antiapoptotic activity of IL-7 provides an additional rationale for consideration of this cytokine as a potential immunotherapeutic agent in the treatment of HIV-1-infected individuals.

An unexpected finding emerging from our study is that the reduction of apoptosis mediated by IL-7 in HIV-1-infected individuals was inversely correlated with the level of circulating CD4⁺ T cells. Thus, despite a more severe immune dysfunction, patients with lower CD4⁺ T cell counts showed a higher sensitivity to IL-7, whereas those with higher CD4⁺ T cell counts, whose levels of spontaneous apoptosis generally are lower, exhibited a lesser sensitivity. In line with this trend, the effect of IL-7 on spontaneous apoptosis was even lower, failing to reach statistical significance, on cells derived from HIV-1-seronegative individuals. These findings suggest that, under physiological conditions, there is a basal level of spontaneous apoptosis that largely is insensitive to IL-7. With the progression of HIV-1 disease, the proportion of cells that are prone to spontaneous apoptosis increases, and a large fraction of these cells is sensitive to the antiapoptotic effects of IL-7. The ex vivo effects of IL-7 in chronically HIV-1-infected individuals previously have been investigated in only two studies, which reported a limited, if any, reduction of spontaneous apoptosis, whereas a more pronounced effect was seen with cells derived from HIV-1-seronegative individuals (24, 25). However, these studies evaluated apoptosis only at a single time point after several days (6 or 7 days) of ex vivo culture. By contrast, the present study was designed to provide a detailed longitudinal representation of apoptosis levels, with daily measurements for 7 days of ex vivo culture in the presence or absence of IL-7. This extensive time-course analysis permitted us to document variable kinetics of IL-7-mediated reduction of apoptosis in different patients, with some patients showing an increasing effect over time and others manifesting an effect that remained constant or even decreased over time. Thus, there was an increasing variability of the IL-7 effect at the late time points of culture, as attested by a reduced statistical significance at days 5 and 6 and a loss of significance at day 7. These observations suggest that a single determination is unlikely to accurately represent the effects of IL-7 on spontaneous apoptosis.

The role played by IL-7 in vivo in the course of HIV-1 infection still is poorly understood. Increased plasma levels of IL-7 were documented in HIV-1-infected individuals and inversely correlated with CD4 counts (30, 31, 36). In parallel, the expression of the α-chain of the IL-7 receptor (CD127) was shown to be downmodulated (24, 36, 37). Because the levels of IL-7 decreased when the CD4⁺ T cell counts were restored during therapy, the increases in IL-7 levels have been interpreted as a homeostatic response to lymphopenia (31). However, the fact that in the absence of ART the number of circulating CD4⁺ T cells remains low suggests that IL-7 per se is unable to reverse the loss of CD4⁺ T cells resulting from the effects of HIV viremia. Different mechanisms may account for this inability. The rate of CD4⁺ T cell destruction in patients with advanced HIV-1 disease may exceed the regenerative capacity of IL-7 or the pool of progenitor T cells may be irreversibly damaged or depleted. Alternatively, the IL-7/IL-7 receptor axis may be impaired (37). The marked sensitivity to the effects of IL-7 that we documented in patients with low CD4⁺ T cell counts indicates that circulating T cells from these patients did express a functional IL-7 receptor. Interestingly, however, kinetic analysis demonstrated that the ex vivo effect of IL-7 in these patients was delayed, suggesting that their T cells might be in a state of IL-7 refractoriness in vivo. Responsiveness to IL-7 then would gradually be restored upon ex vivo culturing presumably because of the removal of putative inhibitory factor(s).

Studies of IL-7 administration to uninfected and SIV-infected macaques have demonstrated a proliferative effect on CD4⁺ and CD8⁺ T cells expressing both memory and naïve phenotypes (27, 28, 38). Analogous results were obtained more recently in human cancer patients (39), as well as in HIV-1-infected individuals (M. Lederman, personal communication). In the present study, we observed that IL-7 given ex vivo at concentrations higher than 2.5 ng/ml consistently induced cellular proliferation. However, this effect became detectable only after several days in culture and was temporally dissociated from the effect of IL-7 on apoptosis. Moreover, at concentrations below 1 ng/ml, IL-7 still was effective in reducing apoptosis but failed to induce cellular proliferation. The evidence that the different activities of IL-7 can be dissociated is encouraging with regard to the potential therapeutic use of this cytokine in HIV-1-infected subjects because it mitigates concerns related to the putative risk of enhancing the level of immune activation and viral replication. Previous studies have shown that ex vivo treatment with IL-7 increases the levels of HIV-1 replication in naturally infected CD8-depleted mononuclear cells (33) and may lead to reactivation of latent provirus in purified resting CD4⁺ T lymphocytes isolated from HIV-1-infected subjects receiving ART with undetectable plasma viremia (26). At variance with these observations, however, we detected only minimal levels, if any, of HIV-1 replication in purified CD4⁺ T cells treated ex vivo with IL-7. Our results are in agreement with the findings in SIV-infected macaques treated in vivo with IL-7, which showed no increase in viral load both in the presence (27) and in the absence (28) of concomitant ART.

In conclusion, the results of the present study provide a further rationale for consideration of IL-7 as a potential adjuvant therapy in HIV-1-infected individuals in association with ART. Future studies will elucidate the precise molecular mechanisms underlying the increased propensity to spontaneous apoptosis in HIV-1-infected individuals as well as the antiapoptotic action of IL-7. Further experimental studies in nonhuman primates and clinical studies in HIV-1-infected individuals should allow elucidation as to whether IL-7 administration effectively leads to a reduction in the levels of spontaneous apoptosis in vivo and whether this effect is associated with a stable immune reconstitution in patients with sustained suppression of viremia.

Materials and Methods

Study Subjects.

Twenty-nine HIV-1-infected individuals and 14 HIV-1-seronegative age-matched controls were included in this study. The HIV-1-infected individuals were selected among those attending the National Institute of Allergy and Infectious Diseases (NIAID) Outpatient Clinic to represent different clinical stages. All patients provided informed consent, in accordance with the NIAID Institutional Review Board, and were subjected to leukapheresis according to approved protocols to obtain PBMC.

Isolation of PBMC and Purification of CD4⁺ and CD8⁺ T Cells.

PBMC were isolated by gradient centrifugation from leukapheresis packs by using Lymphocyte Separation Medium (MP Biomedicals, Solon, OH). CD4⁺ and CD8⁺ T cells were purified by negative selection by using Dynabeads goat anti-mouse IgG (Dynal, Carlsbad, CA) and a mixture of purified monoclonal antibodies (mAb) against human CD14 and CD19 (Biodesign International, Saco, ME), CD16 and CD56 (BD PharMingen, San Diego, CA), and either CD8 (BD PharMingen) or CD4 (Biodesign International). The purity of the CD4⁺ and CD8⁺ populations was consistently >95%, as determined by cytofluorimetry.

Cell Cultures.

Isolated PBMC or purified CD4⁺ and CD8⁺ T cell populations were cultured for up to 21 days at 37°C in 25-cm² flasks at a density of 10⁶ cells per ml in complete culture medium [RPMI medium 1640 with l-glutamine (Gibco, Carlsbad, CA) supplemented with penicillin-streptomycin at 10,000 units/ml each (Gibco) and 10% FBS (HyClone, Logan, UT)]. Recombinant human IL-7 (Peprotech, Rocky Hill, NJ) at different concentrations (0.04–50 ng/ml) or IL-2 (Roche, Indianapolis, IN) at 100 units/ml was added at the beginning of the culture.

Measurement of Apoptosis.

Analysis of apoptosis was performed every day over the first 7 days of culture by the detection of annexin V binding and caspase 3 activation. Annexin V binding was measured by using the annexin V:phycoerythrin (PE) Apoptosis Detection Kit I (BD PharMingen). Briefly, at each time point, 5 × 10⁵ cells were stained with mAbs anti-CD3-APC, anti-CD4-FITC, and anti-CD8-PerCP (BD PharMingen), and in selected experiments anti-CD45RA-FITC or anti-CD45RO-FITC (BD PharMingen), and then washed and incubated with 2 μl of annexin V-PE or 5 μl of propidium iodide (PI). Cells then were analyzed immediately with a FACScalibur Flow Cytometer (Becton Dickinson, Franklin Lakes, NJ). Both live and apoptotic cells were included in the analysis, and cellular debris was excluded. Apoptosis was demonstrated by transition from a single-positive (annexin V⁺ PI⁻) toward a double-positive phenotype. Reduction of apoptosis by IL-7 or IL-2 was calculated by subtracting the percentage of annexin V⁺ cells in cytokine-treated cultures from the percentage of annexin V⁺ cells in untreated controls. In selected experiments, apoptosis was confirmed by evaluating the activation of caspase 3. After surface staining as described above, the cells were fixed with BD FACS Lysing Solution, permeabilized by using BD FACS Permeabilizing Solution 2, and incubated with a PE-conjugated rabbit mAb against the active form of caspase 3 (BD PharMingen). The cells then were analyzed by flow cytometry. All of the flow data were analyzed by using the Flowjo software (TreeStar, Ashland, OR).

Analysis of Cellular Proliferation.

Cellular proliferation was evaluated by using intranuclear expression of Ki67, dilution of the vital dye CFSE, and absolute cell counting. Ki67 was assessed by using a PE-conjugated anti-Ki67 mAb (BD PharMingen) after fixation and permeabilization of the cells. Labeling with CFSE was performed by using the CellTrace CFSE Cell Proliferation Kit (Invitrogen, Carlsbad, CA), following the provided protocol. Labeled cells then were cultured in complete medium for up to 7 days with or without rhIL-7 or rhIL-2, as described above, and analyzed by cytofluorimetry. Absolute cell counting was obtained by flow cytometry analysis of a fixed volume of cell culture at high flow pressure for 30 sec. To calculate the absolute number of cells in each sample, the number of events acquired was multiplied by the flow rate.

Measurement of HIV-1 Replication.

HIV-1 replication was evaluated by measuring the levels of HIV-1 p24 antigen released into the culture supernatants by using an HIV-1 p24 Antigen capture EIA (Beckman Coulter, Fullerton, CA). To maximize sensitivity, the samples were analyzed undiluted. The plates were read according to a kinetic protocol by using a microplate spectrophotometer (Bio-Rad Instruments, Hercules, CA).

Statistical Analysis.

Statistical analysis was conducted by using the software SAS (version 9.1 for Windows), S-Plus (version 6.2 for Windows), and StatView (version 5.0.1). Paired t tests were used to compare differences in apoptosis between IL-7-treated and untreated cells on each day and averaged over all 7 days. Regression analysis was used to correlate the effect of IL-7, averaged across the first 6 days of culture, to other covariates. Mixed-model methodology was used in analyses comparing slopes relating the IL-7 effect and days. Pitman's test was used to compare pairs of days with respect to the variability of IL-7 effects.

Abbreviations

SIV

simian immunodeficiency virus

PBMC

peripheral blood mononuclear cells

CFSE

carboxyfluorescein diacetate succinimidyl ester

ART

antiretroviral therapy

PE

phycoerythrin.