Persistent Apoptosis in HIV-1-Infected Individuals Receiving Potent Antiretroviral Therapy Is Associated With Poor Recovery of CD4 T Lymphocytes

Hansjee, Natasha PhD*; Kaufmann, Gilbert R. MD†‡; Strub, Christoph MD†; Weber, Rainer MD§; Battegay, Manuel MD†; Erb, Peter PhD*; and the Swiss HIV Cohort Study

JAIDS Journal of Acquired Immune Deficiency Syndromes:
Basic Science

CD4 T-cell depletion in HIV-1 infection is partly the result of T-cell apoptosis. Spontaneous apoptosis (SA) and apoptosis markers Fas-associated death-domain-like IL-1 beta converting enzyme (FLICE)-like inhibitory protein (FLIP), Bcl-2, TRAIL (tumor necrosis factor–related apoptosis-inducing ligand), TRAIL receptor 1, and Fas were determined in 55 HIV-1 infected persons treated with highly active antiretroviral therapy (HAART) for 48 months. Despite suppressive HAART, SA remained elevated. Increased SA of peripheral blood mononuclear cells (PBMCs) and CD8 T lymphocytes and increased TRAIL receptor 1 expression strongly predicted a poorer recovery of CD4 T-cell count. HAART did not significantly alter anti-or proapoptotic markers in cultured PBMCs and T lymphocytes. The significant relationship between residual T-lymphocyte apoptosis and CD4 T-cell recovery suggests that persistent apoptosis may impede immune restoration.

Author Information

From the *Institute for Medical Microbiology, University of Basel, †Division of Infectious Diseases, Department of Internal Medicine, University Hospital Basel, ‡Medical Outpatient Department, University Hospital Basel, and §Division of Infectious Diseases, University Hospital Zurich, Switzerland.

Received for publication November 19, 2003; accepted February 13, 2004.

Drs. Hansjee and Kaufmann contributed equally to this study.

Reprints: Gilbert Kaufmann, Division of Infectious Diseases, University Hospital Basel, Petersgraben 4, CH-4031 Basel, Switzerland (e-mail:

Article Outline

The mechanisms involved in the progressive decline of CD4 T-lymphocyte numbers in untreated HIV-1-infected patients are the focus of intensive research but remain inconclusive. Viral infection and subsequent cytopathic effects or enhanced apoptosis of CD4 T lymphocytes have been proposed to be the major reasons for the progressive loss of these cells. 1 However, it is difficult to estimate the contribution of apoptosis to T-cell depletion. Although apoptosis affects more CD8 than CD4 T lymphocytes, only CD4 T lymphocytes decline during early stages of HIV-1 infection. 2

Apoptosis is a tightly regulated process, induced by death receptor signaling after interaction with death receptor–specific ligands. At least 4 cellular receptors are known, including Fas, tumor necrosis factor (TNF) receptor, and TNF-related apoptosis-inducing ligand (TRAIL) receptors R1 and R2. 3 Apoptosis is inhibited by Bcl-2 and the FLICE-like inhibitory protein (c-FLIP), 4 the latter preventing caspase-8 activation. T lymphocytes of untreated HIV-1-infected patients are highly susceptible to spontaneous (SA) and activation-induced apoptosis. 5 The level of SA is an important predictor of the rate of disease progression. 6 In treated HIV-1 infection, short- and medium-term observations suggest that higher levels of SA are associated with poorer recovery of CD4 T lymphocytes. 7,8

In this investigation, we evaluated the long-term effect of highly active antiretroviral therapy (HAART) on the level of SA of peripheral blood mononuclear cells (PBMCs), CD4, and CD8 T lymphocytes and studied the relationship of SA and the recovery of CD4 T lymphocytes.

Back to Top | Article Outline



From March 2001 until July 2001, 73 individuals participating in the Swiss HIV Cohort Study (Basel and Zurich HIV Units) were consecutively enrolled in this investigation. The mean age of the study population was 41.1 ± 8.7 years. The duration of HIV-1 infection was 7.7 ± 4.8 years. Further baseline characteristics are shown in Table 1.

Fifty-five of these individuals were treated with HAART for a median time of 48 months (interquartile range: 38–57), 35 of whom (63.6%) suppressed viral load to levels <1000 copies/mL throughout the observation period. Twenty subjects (36.4%) experienced virologic failure, as defined by plasma HIV-1 RNA levels increasing at least once to >1000 copies/mL. Eighteen untreated HIV-1-infected individuals and 19 healthy HIV-1-uninfected blood donors served as control groups.

Back to Top | Article Outline
Laboratory Methods

SA assays were performed as reported previously. 9 In brief, PBMCs were isolated from fresh blood samples by centrifugation on a Ficoll-Paque density gradient (Pharmacia, Upsala, Sweden). To induce SA, 106/mL PBMCs were cultured for 7 days in complete medium without stimulation at 37°C and 5% CO2 atmosphere. The proportions of apoptotic CD3+, CD8+, and CD3+/CD8 (CD4+) PBMCs were determined, labeling cells with 7-aminoactinomycin D (Molecular Probes, Lucerne, Switzerland). The expression of apoptotic and anti-apoptotic markers was determined by double-staining incubated cells with monoclonal antibodies CD3 and either Fas-FITC (fluorescein isothiocyanate) (Serotec, Raleigh, NC), Bcl-2-FITC (DAKO, Glostrup, Denmark), TRAIL-FITC, TRAIL-R1-FITC, or FLIP-FITC. The latter 3 antibodies were produced and evaluated in our own laboratory. 10

Plasma HIV-1 RNA was quantified using a commercial reverse transcriptase polymerase chain reaction (Amplicor, Roche Diagnostic Systems, Basel, Switzerland).

Back to Top | Article Outline
Statistical Analysis

The results were analyzed according to HIV-1 serostatus, antiretroviral therapy, and virologic response to therapy. SA and the expression of apoptosis markers were compared among groups using a Mann-Whitney U or Kruskal-Wallis test as appropriate. The relationships between CD4 T-cell count, CD4 T-cell increase, plasma HIV-1 RNA, and apoptosis markers were analyzed with a Spearman rank correlation or Kruskal-Wallis test. A 2-sided P value <0.05 was considered statistically significant. The statistical analysis was performed with SPSS V.11.0 (SPSS, Inc., Chicago, IL).

Back to Top | Article Outline


Elevated Spontaneous Apoptosis in Untreated HIV-1-Infected Individuals

The median proportions of PBMCs and T lymphocytes undergoing SA were significantly higher in untreated HIV-1-infected than in HIV-1-uninfected individuals (39.8 vs. 20.5%; P = 0.002 and 41.4 vs. 24.3%; P = 0.002;Fig. 1A). CD8 T lymphocytes (65.4% vs. 30.0%; P = 0.011) as well as CD4 T lymphocytes (32.6% vs. 19.0%; P = 0.025) contributed to the larger proportion of apoptotic T lymphocytes in HIV-1-infected subjects. The magnitude of SA of PBMCs and T lymphocytes was similar in male and female HIV-1-infected individuals (Fig. 1B and C), but there was a nonsignificant trend toward larger SA of PBMCs and T lymphocytes in individuals with higher plasma HIV-1 RNA levels. In addition, higher age was associated with higher SA levels of PBMCs (P = 0.049), but not of T lymphocytes. The duration of HIV-1 infection was not associated with SA levels.

Back to Top | Article Outline
Increased Levels of Spontaneous Apoptosis Despite Long-Term Suppression of Viral Replication

Treated persons received HAART for a median time of 48 months (IQR: 38–57). Median CD4 and CD8 T-cell counts increased from 166 (IQR: 93–365) to 407 (260–639) cells/μL and from 695 (494–1115) to 977 (731–1173) cells/μL, respectively. Plasma HIV-1 RNA declined from 4.4 (3.1–5.0) to 1.8 (1.3–3.3) log10 copies/mL.

Median SA levels of PBMCs and CD4 and CD8 T lymphocytes were lower in treated than in untreated individuals (27.7 vs. 39.8%, 21.2 vs. 32.6%, 50.6 vs. 65.4%;Fig. 1A). However, the reductions of SA were statistically not significant. A total of 45.7% of individuals with viral loads <1000 copies/mL showed levels of apoptotic PBMCs >75th percentile of HIV-1-uninfected subjects and in 50% of individuals with plasma HIV-1 RNA <1000 copies/mL, T-lymphocyte apoptosis remained elevated >75th percentile. The magnitude of SA did not depend on the duration of antiretroviral therapy.

Back to Top | Article Outline
Phenotype of Cultured Cells

Bcl-2 expression of PBMCs was higher in HIV-1-uninfected than in untreated HIV-1-infected persons (98.1 vs. 94.2%; P = 0.006;Fig. 2A), but Bcl-2 expression of T lymphocytes was similar (77.0 vs. 79.0%, P = 0.953). The same trend was observed regarding c-FLIP expression, but differences were statistically not significant (67.0 vs. 56.5%; 57.0 vs. 49.0%;Fig. 2B).

HIV-infected patients showed a trend toward a larger proportion of TRAIL-expressing PBMCs (14.9 vs. 11.3%; P = 0.095) and T lymphocytes (9.6 vs. 6.2%; P = 0.073;Fig. 2C), but TRAIL receptor 1 (TR1) expression of PBMCs and T lymphocytes was similar as in HIV-1-uninfected subjects (14.5 vs. 13.3%, P = 0.905; 10.2 vs. 11.5%, P = 0.556;Fig. 2D). In addition, Fas expression of PBMCs and T lymphocytes was comparable (29.2 vs. 29.1%, P = 0.929; 22.7 vs. 22.9%, P = 0.561;Fig. 2E).

Back to Top | Article Outline
Phenotype of Cultured Cells in Individuals Receiving HAART

The median proportions of PBMCs and T lymphocytes expressing c-FLIP were larger in successfully treated than in untreated HIV-1 infection (86.0 vs. 56.5% and 68.5 vs. 49.0%;Fig. 2B), but individual variability was large and differences were statistically not significant. Bcl-2 showed similar expression in PBMCs and T lymphocytes in successfully treated as in untreated HIV-1-infected subjects (94.4 vs. 94.2%; P = 0.343; 77.5 vs. 79.0%, P = 0.462;Fig. 2A).

The median percentages of PBMCs and T lymphocytes expressing TRAIL (22.0 vs. 14.9% and 12.0 vs. 9.6%;Fig. 1D) and TRAIL receptor 1 (20.2 vs. 13.3% and 14.6 vs. 11.5%;Fig. 1E) were larger in successfully treated than in untreated HIV-1-infected individuals, but differences were statistically not significant. Similarly, the slightly downregulated Fas expression of PBMCs and T lymphocytes in treated individuals did not significantly differ from that of untreated patients (26.2 vs. 29.1%, P = 0.868 and 17.8 vs. 22.9%, P = 0.766;Fig. 2E).

Back to Top | Article Outline
Spontaneous Apoptosis Is Associated With the Recovery of CD4 T Lymphocytes in Treated HIV-1 Infection

In individuals receiving HAART, an inverse relationship was observed between the absolute CD4 T-cell count after HAART and the percentage of apoptotic PBMCs (r = −0.468; P < 0.001;Fig. 3A). In addition, the absolute increase in CD4 T-cell count was inversely associated with the percentage of apoptotic PBMCs (r = −0.429; P = 0.001;Fig. 3A), T lymphocytes (r = −0.264; P = 0.059), and CD8 T lymphocytes (r = −0.373; P = 0.010). Stratifying successfully treated individuals into 3 equally sized groups according to SA levels of PBMCs further confirmed the strong inverse relationship between SA and recovery of CD4 T lymphocytes (Fig. 3B). Even after adjustment for baseline CD4 T-cell count and the time on HAART, the relationship was highly significant (r = −0.553, P = 0.001). Of note, all 3 patient groups showed very similar virologic responses (Fig. 3C).

TR1 expression in T lymphocytes was the only apoptosis marker that was significantly associated with the absolute CD4 T-cell count after HAART (r = −0.390, P = 0.006) and the CD4 T-cell increase during the observation period (r = −0.387, P = 0.007).

Back to Top | Article Outline


Spontaneous apoptosis and the phenotype of cultured PBMCs and T lymphocytes were analyzed in 55 subjects receiving HAART for a substantial period of 4 years. We observed a reduction of SA after long-term antiretroviral therapy. 11 The median proportion of apoptotic CD4 T lymphocytes almost reached the level of HIV-1-uninfected individuals, but apoptosis of PBMCs and CD8 T lymphocytes remained elevated despite excellent virologic responses in 64% of these individuals. Our results therefore confirm and extend observations reported by De Oliveira et al, 8 who found persistently elevated levels of T-cell apoptosis in 60% of individuals receiving HAART for up to 2 years.

Furthermore, we found a strong inverse relationship between SA rates of PBMCs and CD8 T lymphocytes and the recovery of CD4 T lymphocytes. Roger et al 7 observed an inverse relationship between the magnitude of apoptosis of CD4 T lymphocytes and the short-term increase of these cells in the first 3 months after initiation of HAART. De Oliveira et al 8 reported a stronger inverse relationship between CD8 than CD4 T-lymphocyte apoptosis and CD4 T-cell recovery. In both studies, these associations were independent of viral load. Based on the physiological involution and reduced function of the thymus in adulthood, it has been suggested that the major reason for the incomplete replenishment of CD4 T lymphocytes in treated HIV-1 infection may be the limited regenerative capacity of the immune system. Our data provide evidence that persistently elevated apoptosis levels are associated with poor immune reconstitution as well. However, it remains to be determined whether poor CD4 T-cell recovery is the direct consequence of increased cell death. The downregulation of Bcl-2 and increased T-cell apoptosis in HIV-1 infection is closely related to T-lymphocyte activation. 2 Persistent SA may be a surrogate marker of increased T-cell activation, which would explain why the relationship of SA and increased CD4 T-cell recovery was stronger for CD8 as for CD4 T lymphocytes. In fact, CD4 T-cell activation has recently been reported to be associated with poor CD4 T-cell recovery. 12

Interestingly, TRAIL receptor 1 expression in T lymphocytes was strongly associated with CD4 T-cell recovery. T lymphocytes of HIV-1-infected subjects appear to be more susceptible to TRAIL-mediated cell death. 13 In addition, TRAIL-positive macrophages have recently been implicated to act as killer cells of bystander T lymphocytes. 14

We observed a tendency toward a decreased expression of antiapoptotic markers c-FLIP and Bcl-2 in the HIV-1-infected population, whereas proapoptotic markers were very similarly expressed in HIV-1-infected and uninfected individuals. HAART had only a small effect and did not induce significant changes of anti- and proapoptotic markers. Nevertheless, the median expression of antiapoptotic proteins in treated individuals was shifted toward slightly higher values. These results are partly in contrast with the literature. Fas ligand has been reported to be upregulated in HIV-1-infected T lymphocytes, 15 and the proportion of Fas-expressing cells appears to increase in parallel with HIV-1 disease progression. 16 The discrepant findings may be explained by differences in laboratory methods. In our study, PBMCs were cultured for 7 days without stimulation, which was significantly longer than in other studies. 7 The longer duration of cell culture may have altered the expression of apoptosis markers.

We did not find a significant relationship between plasma HIV-1 RNA levels and the rate of SA, which is in agreement with other studies. 7 It may indicate that SA is mediated rather by bystander mechanisms such as the increased secretion of cytokines TNF-α, lymphotoxin, or transforming growth factor-β 17 or the cross-linking of CD4 molecules 18 than by the virus itself.

In summary, we have observed elevated levels of SA of PBMCs and CD8 T lymphocytes in HIV-1-infected patients who have been treated successfully with HAART for a median time of 4 years. Increased apoptosis levels were associated with poorer CD4 T-cell recovery and may represent a further explanation for discordant virologic and immunologic responses.

Back to Top | Article Outline


The members of the Swiss HIV Cohort Study are M. Battegay, M.-C. Bernard, E. Bernasconi, H. Bucher, Ph. Bürgisser, M. Egger, P. Erb, W. Fierz, M. Flepp (Chairman of the Clinical and Laboratory Committee), P. Francioli (President of the SHCS, Centre Hospitalier Universitaire Vaudois, CH-1011- Lausanne), H.J. Furrer, M. Gorgievski, H. Günthard, P. Grob, B. Hirschel, C. Kind, Th. Klimkait, B. Ledergerber, U. Lauper, M. Opravil, F. Paccaud, G. Pantaleo, L. Perrin, J.-C. Piffaretti, M. Rickenbach (Head of Data Center), C. Rudin (Chairman of the Mother and Child Substudy), J. Schupbach, A. Telenti, P. Vernazza (Chairman of the Scientific Board), Th. Wagels, and R. Weber.

Back to Top | Article Outline


1. Ameisen JC. HIV: setting death in motion. Nature. 1998,395:117,119.
2. Zaunders JJ, Moutouh-de Parseval L, Kitada S, et al. Polyclonal proliferation and apoptosis of CCR5+ T lymphocytes during primary human immunodeficiency virus type 1 infection: regulation by interleukin (IL)-2, IL-15, and Bcl-2. J Infect Dis. 2003;187:1735–1747.
3. Nagata S. Apoptosis by death factor. Cell. 1997;88:355–365.
4. Irmler M, Thome M, Hahne M, et al. Inhibition of death receptor signals by cellular FLIP. Nature. 1997;388:190–195.
5. Laurent-Crawford AG, Krust B, Muller S, et al. The cytopathic effect of HIV is associated with apoptosis. Virology. 1991;185:829–839.
6. Prati E, Gorla R, Malacarne F, et al. Study of spontaneous apoptosis in HIV+ patients: correlation with clinical progression and T cell loss. AIDS Res Hum Retroviruses. 1997;13:1501–1508.
7. Roger PM, Breittmayer JP, Durant J, et al. Early CD4(+) T cell recovery in human immunodeficiency virus-infected patients receiving effective therapy is related to a down-regulation of apoptosis and not to proliferation. J Infect Dis. 2002;185:463–470.
8. de Oliveira Pinto LM, Lecoeur H, Ledru E, et al. Lack of control of T cell apoptosis under HAART: influence of therapy regimen in vivo and in vitro. AIDS. 2002;16:329–339.
9. Regamey N, Harr T, Battegay M, et al. Downregulation of Bcl-2, but not of Bax or Bcl-x, is associated with T lymphocyte apoptosis in HIV infection and restored by antiretroviral therapy or by interleukin 2. AIDS Res Hum Retroviruses. 1999;15:803–810.
10. Bachmann F, Buechner SA, Wernli M, et al. Ultraviolet light downregulates CD95 ligand and TRAIL receptor expression facilitating actinic keratosis and squamous cell carcinoma formation. J Invest Dermatol. 2001;117:59–66.
11. Badley AD, Parato K, Cameron DW, et al. Dynamic correlation of apoptosis and immune activation during treatment of HIV infection. Cell Death Differ. 1999;6:420–432.
12. Hunt PW, Martin JN, Sinclair E, et al. T cell activation is associated with lower CD4+ T cell gains in human immunodeficiency virus-infected patients with sustained viral suppression during antiretroviral therapy. J Infect Dis. 2003;187:1534–1543.
13. Jeremias I, Herr I, Boehler T, et al. TRAIL/Apo-2-ligand-induced apoptosis in human T cells. Eur J Immunol. 1998;28:143–152.
14. Zhang M, Li X, Pang X, et al. Identification of a potential HIV-induced source of bystander-mediated apoptosis in T cells: upregulation of TRAIL in primary human macrophages by HIV-1 tat. J Biomed Sci. 2001;8:290–296.
15. Kaplan D, Sieg S. Role of the Fas/Fas ligand apoptotic pathway in human immunodeficiency virus type 1 disease. J Virol. 1998;72:6279–6282.
16. Estaquier J, Tanaka M, Suda T, et al. Fas-mediated apoptosis of CD4+ and CD8+ T cells from human immunodeficiency virus-infected persons: differential in vitro preventive effect of cytokines and protease antagonists. Blood. 1996;87:4959–4966.
17. Zheng L, Fisher G, Miller RE, et al. Induction of apoptosis in mature T cells by tumour necrosis factor. Nature. 1995;377:348–351.
18. Oyaizu N, McCloskey TW, Coronesi M, et al. Accelerated apoptosis in peripheral blood mononuclear cells (PBMCs) from human immunodeficiency virus type-1 infected patients and in CD4 cross-linked PBMCs from normal individuals. Blood. 1993;82:3392–3400.

HIV infection; T lymphocytes; spontaneous apoptosis; immune reconstitution; antiretroviral therapy

© 2004 Lippincott Williams & Wilkins, Inc.