HIV-1 infection is characterized by a gradual decline in CD4 T cell numbers and by chronic hyperactivation of the immune system. In fact, elevated immune activation is related to the outcome of HIV-1 infection. Using proportional hazards models, various parameters for immune activation have additive or stronger prognostic value in predicting progression to AIDS than CD4 T cell numbers or plasma HIV-1 RNA alone [1–9]. Most studies have focused on the predictive value of such markers for staging and monitoring of treatment of HIV-1-infected individuals. The mechanism by which chronic activation of the immune system may influence HIV-1 pathogenesis remains, however, uncertain.
We have postulated that persistent hyperactivation of the immune system may cause depletion of CD4 T cells because it leads to erosion of the naive T cell pool . We further elaborated on this hypothesis taking advantage of the unique material of the Amsterdam Cohort Studies on HIV-1 infection and AIDS among homosexual men (ACS), which includes frozen peripheral blood mononuclear cells (PBMC) collected before HIV-1 seroconversion. In this prospective cohort study consisting of patients with a known seroconversion interval, the level of CD4 and CD8 T cell activation at several time points following infection and, importantly, before seroconversion was determined using CD38, HLA-DR, CD70 and Ki67. CD38 and HLA-DR (MHC class II antigen) are two well-known antigens whose expression is upregulated on activated T cells. The tumor necrosis factor (TNF) receptor family member CD70 is expressed on activated T cells and controls the magnitude and duration of T cell responses . HIV-1-infected individuals have upregulated CD70 expression on T cells, which may reflect and maintain increased levels of immune activation . Ki67 is a nuclear antigen that is expressed exclusively by cells that are in cell cycle  and is, therefore, frequently used as a surrogate marker for T cell proliferation; however, expression of this antigen does not differentiate between cells that are undergoing proliferation or cells that are arrested in cell cycle. Nevertheless, combination of ex vivo Ki67 analysis with in vivo labelling of dividing T cells using 5-bromo-2-deoxyuridine or deuterated glucose recently confirmed that Ki67 can be used as a surrogate marker for T cell proliferation, even in the setting of HIV-1 infection [14,15]. The predictive value of immune activation and CD4 T cell numbers is examined using Cox proportional hazards analyses at time points after and before HIV-1 seroconversion.
Out of a total of 149 participants in the ACS, 102 individuals (68%) were selected who had a documented time of seroconversion and cryopreserved PBMC available from before seroconversion (more than 1 year before seroconversion; n = 56) and from 1 and/or 5 years after seroconversion (n = 97 and n = 64, respectively). Details on inclusion criteria, enrollment and baseline characteristics of ACS participants can be obtained at http://www.amsterdamcohortstudies.org. In the present study, follow-up was censored at 1 June 2000, resulting in a maximum follow-up from seroconversion of 15.1 years. Median follow-up to AIDS diagnosis or censoring was 6.3 years [interquartile range (IQR), 4.1–8.5], 4.2 years (IQR, 1.8–7.2) and 2.1 years (IQR, 0.4–5.0), from seroconversion, from 1 and from 5 years postseroconversion, respectively. Of the 102 participants, 38 progressed to AIDS during follow-up (Centers for Disease Control and Prevention case definition 1987), 18 within 5 years of seroconversion. Twenty-two patients started highly active antiretroviral therapy (HAART) and this was corrected for in the analyses. On study entry, patients gave institutional review board-approved informed consent.
Plasma HIV-1 RNA and T cell numbers
At every visit, plasma HIV-1 RNA was determined with Roche Amplicor Monitor Standard Assay (Roche Diagnostics, Branchburg, New Jersey, USA) or NASBA HIV-1 RNA QT (Organon Teknika, Boxtel, the Netherlands). Naive (CD27+CD45RO−), CD27 memory (CD27+CD45RO+), CD27− memory (CD27− CD45RO+) and effector (CD27−CD45RO−) CD4 and CD8 T cells were defined as described previously  and their numbers assessed by flow cytometry.
Immune activation parameters and T cell proliferation
Immune activation status was analysed by measuring CD4 and CD8 T cell expression of CD38, HLA-DR and CD70. Briefly, PBMC were thawed and two samples of 0.5 × 106 cells were incubated with monoclonal antibodies: the first with CD4–APC (allophycocyanin) (Becton Dickinson, San Jose, California, USA), CD8–PerCP (peridinin–chlorophyll a complex protein) (Becton Dickinson), CD38–PE (phycoerythrin) (CLB, Amsterdam, the Netherlands) and HLA-DR–FITC (fluoroscein isothiocyanate) (CLB) and the second with CD4–FITC (Becton Dickinson), CD8–PerCP (Becton Dickinson), biotinylated CD70 (a kind gift of Dr R. A. W. van Lier, CLB) and streptavidin–APC (Becton Dickinson). Both were fixed using Cellfix (Becton Dickinson). Total and naive, memory and effector CD4 and CD8 T cell proliferation was analysed by measuring Ki67 expression. For this, two samples of at least 1 × 106 cells each were incubated with CD4–PerCP or CD8–PerCP (Becton Dickinson), CD45RO-PE (Becton Dickinson), biotinylated CD27 (CLB), and streptavidin–APC (Becton Dickinson). Lymphocytes were fixed (FACS Lysing Solution; Becton Dickinson), permeabilized (FACS Permeabilization Buffer; Becton Dickinson), stained intracellularly (Ki67–FITC monoclonal antibody; Immunotech, Marseille, France) and fixed with Cellfix (Becton Dickinson). Ki67, CD38, HLA-DR and CD70 expression on CD4 and CD8 T cells was analysed on a FACSCalibur (Becton Dickinson) with Cellquest software.
For analysis of the development of CD4 and CD8 T cell numbers, plasma HIV-1 RNA, expression of activation markers and T cell division, patients were divided into two groups: patients who progressed to AIDS within 7 years after seroconversion (fast progressors) and patients who did not progress to AIDS during that period (slow progressors). Differences between both groups at each time point (i.e. preseroconversion, 1 and 5 years after seroconversion) and between time points in each group were analysed using the non-parametric Mann–Whitney U test and Wilcoxon test for paired samples, respectively. Correlations between parameters were estimated with non-parametric Spearman's correlation coefficients (r). Patients receiving HAART were excluded from these analyses.
The predictive values of plasma HIV-1 RNA, CD4 and CD8 T cell numbers, activation marker expression and Ki67 expression for progression to AIDS were studied using Kaplan–Meier survival and Cox proportional hazards analyses [17,18]. Each individual immunological marker was tested in a separate model. Univariate and multivariate relative hazards (RH) and confidence intervals (CI) were calculated for markers measured at each time point (preseroconversion, 1 and 5 years after seroconversion; baseline analyses). In multivariate analyses, each model was corrected for known prognostic parameters: plasma HIV-1 RNA at 1 year, and plasma HIV-1 RNA and CD4 T cell count at 5 years after seroconversion [19,20]. Significance was defined by log-rank and log-likelihood tests, respectively. In addition, uni- and multivariate time-dependent Cox analyses were performed using postseroconversion data only. For all analyses, patients were divided into three groups by CD4 cell count: > 500, 300–500 and < 300 × 106 cells/l. Plasma HIV-1 RNA was defined by the 33rd and 67th percentile of each time point (1 year after seroconversion: 7.4 × 103 and 14.9 × 103 copies/ml; 5 years after seroconversion 13.1 × 103 and 26.2 × 103 copies/ml). Patients were divided into two groups based on activation marker and Ki67 expression, depending on the median value of each parameter per time point in the time-independent analyses and on the median value of all postseroconversion measurements for time- dependent analyses. Seven patients received HAART at baseline (three at 1 year and four at 5 years after seroconversion) and 15 patients started HAART during subsequent follow-up. Correction for the use of antiretroviral drugs was done by excluding those patients that received HAART at baseline, and by censoring the others at the moment they started HAART.
Changes in CD4 and CD8 T cell numbers in slow and fast progressors
Immunological parameters were compared between patients who progressed to AIDS within 7 years after seroconversion and patients who did not progress to AIDS during that period, using non-parametric Mann–Whitney U test and Wilcoxon test for paired samples. Before seroconversion, slow and fast progressors had similar numbers of CD4 and CD8 T cells (Fig. 1a, upper panels). Following HIV-1 infection, CD4 T cell numbers declined more rapidly in patients who progressed to AIDS (P = 0.004 5 years after seroconversion; Fig. 1a). This was attributable to a more pronounced decrease in the numbers of naive and of CD27 memory CD4 T cells in the fast progressors [P = 0.015 and P = 0.008, respectively; Fig. 1a (lower panels) and data not shown]. CD8 T cell numbers increased following seroconversion, which was related to an expansion of CD27 memory CD8 T cell subsets [Fig. 1a (upper panels) and data not shown]. Interestingly, this increase was only statistically significant in the slow progressors. In addition, fast progressors showed a significant decline in the number of naive CD8 T cells 5 years after seroconversion (P = 0.050), which was not observed in the slow-progressor group (Fig. 1, lower panels). The number of naive CD4 T cells correlated significantly with the number of naive CD8 T cells (r= 0.54, P < 0.001 for postseroconversion measurements; data not shown).
T cell division and activation in slow and fast progressors
Following seroconversion, Ki67, HLA-DR, CD38 and CD70 expression on CD4 and CD8 T cells increased significantly (P < 0.05 compared with preseroconversion values, except for the proportion of CD38+CD4+ T cells, CD70+CD4+ T cells and CD70+CD8+ T cells in fast progressors; Fig. 1b and data not shown). In slow progressors, Ki67 expression on CD4 and CD8 T cells stabilized over time, whereas fast progressors showed a significant increase in the proportion of CD4 and CD8 T cells expressing this antigen (Fig. 1b). Five years after seroconversion, T cell activation marker expression [except for the percentage of CD38+CD4+ T cells (P = 0.062), HLA-DR+CD8+ T cells (P = 0.062) and CD38+HLA-DR+CD8+ T cells (P = 0.306)] and T cell division rates were significantly higher in slow progressors compared with fast progressors (P < 0.05; Fig. 1b and data not shown). Ki67 expression on naive, memory and CD27– effector T cells showed similar dynamics compared to Ki67 expression on total CD4 and CD8 T cells (data not shown).
Correlation of plasma HIV-1 RNA and CD4 T cell numbers with immune activation
In agreement with previous observations , plasma HIV-1 RNA was higher in fast progressors than in slow progressors when measured 1 and 5 years after seroconversion (P < 0.05; data not shown). Expression of activation and proliferation markers on CD4 and CD8 T cells correlated weekly with plasma HIV-1 RNA (all P values < 0.05; r, 0.17–0.37). Significant correlations were found between the number of CD4 T cells and the proportions of Ki67+CD4+ T cells, HLA-DR+CD4+ T cells and HLA-DR+CD38+ CD4+ T cells (P < 0.05; r, −0.506 to −0.599) but not between CD8 T cell numbers and activation marker or Ki67 expression. Of note, the proportion of Ki67+ CD4+ T cells measured 5 years after seroconversion correlated significantly with the decline in naive CD4 T cell numbers between 1 and 5 years after infection (P < 0.001, r= 0.51; data not shown).
Predictive value of Ki67 and T cell activation marker expression
The prognostic value of the proportion of Ki67+ CD4+ and Ki67+CD8+ T cells and Ki67 naive, memory and effector CD4 and CD8 T cell subsets for development of AIDS was studied using Cox proportional hazards analysis. In univariate analyses 1 year after seroconversion, an elevated proportion of proliferating CD27– memory CD4 T cells was associated with a 2.1-fold increased risk (95% CI, 1.1–3.9) to develop AIDS. The RH decreased only marginally when adjusted for the other prognostic marker at this time point, plasma HIV-1 RNA (Table 1). Five years after seroconversion, CD4 and CD8 T cell proliferation or the proportion of Ki67 naive, memory and effector T cell subsets were highly predictive for progression to AIDS (Table 1). When fitted in multivariate models, the RH values decreased but remained significant for most covariates (Table 1).
In univariate time-dependent analyses, the RH for progression to AIDS varied between 1.6 (95% CI, 0.8–3.1) for the proportion of Ki67+CD27+ memory CD8 T cells and 3.7 (95% CI, 1.8–7.5) for the proportion of Ki67+CD27− memory CD4 T cells (Table 1). Multivariate time-dependent analyses of all markers showed decreased RH values such that the predictive value of elevated Ki67 expression on naive CD4 T cells or on CD27− effector CD8 T cells no longer reached statistical significance, whereas Ki67 expression on the other subsets did (Table 1).
One year after seroconversion, only the percentage of CD38+CD4+ T cells or of CD38+HLA-DR+CD8+ T cells had significant predictive value in univariate analyses (Table 1). When measured 5 years after seroconversion, expression of most activation markers was predictive for the development of AIDS (Table 1). However, CD4 T cell count and plasma HIV-1 RNA acted as confounders and only elevated CD70 expression on CD4 T cells remained independently associated with disease progression when fitted in a multivariate model (Table 1). When treated as time-dependent covariates, the proportion of CD38+CD4+ or of CD38+CD8+ T cells were predictive for progression; however, in multivariate time-dependent analyses, only the percentage of CD38+CD4+ T cells maintained statistical significance (Table 1).
Predictive value of plasma HIV-1 RNA and CD4 T cell number adjusted for T cell division
Immune activation and cell proliferation of both CD8 and CD4 T cells are thus important determinants for HIV-1 disease progression. The prognostic value of low CD4 T cell numbers or increased HIV-1 plasma RNA for development of AIDS have been reported previously [19,20,22] and could be confirmed in our study group (data not shown). The predictive value of plasma HIV-1 RNA did not change when corrected for the proportion of Ki67+CD27– memory CD4 T cells 1 year after seroconversion [RH 2.4 for patients with high plasma HIV-1 RNA (> 67th percentile); 95% CI, 1.0–5.3]. Five years after seroconversion, the predictive value of high plasma HIV-1 RNA was no longer significant after adjustment for T cell proliferation (RH 2.7; 95% CI, 0.7–10.9); however, low CD4 T cell numbers (< 300 × 106 cells/l) remained significantly associated with progression to AIDS (RH 5.9; 95% CI, 1.1–32.1). In a stepwise model, only CD4 T cell count and the proportion of Ki67+CD27– memory CD4 T cells were selected as predictors of progression to AIDS at this time point [proportion Ki67+CD27– memory CD4 T cells: RH 6.6 (95% CI, 1.7–25.1); CD4 cell count < 300 × 106 cells/l: RH 5.9 (95% CI, 1.1–32.1)].
Preseroconversion measurements of CD4 T cell number and immune activation
If indeed persistent hyperactivation of CD4 T cells through chronic HIV-1 infection leads to depletion of these cells, one would expect that preseroconversion parameters related to the size or activation level of the CD4 T cell pool would be determinants of HIV-1 disease progression. To test this, preseroconversion CD4 T cell numbers and activation parameters were included in our analysis. Because none of the patients had < 300 × 106 cells/l CD4 T cells before seroconversion, the predictive value of having < 500 × 106 cells/l at this time point was tested. Median time to AIDS for these patients was 7.6 years (95% CI, 5.2–10.0), compared with more than 14.0 years for patients who had > 500 × 106 cells/l CD4 T cells before seroconversion (P = 0.034; Fig. 2a). In addition, having higher proportions of activated CD70+CD4+ T cells before seroconversion was associated with shorter AIDS-free survival following seroconversion [median survival time 7.8 years (95% CI, 6.4–9.3) and > 14.0 years, respectively; P = 0.048; Fig. 2b]. Indeed, low CD4 T cell number and high CD70+CD4+ T cell percentage were significantly associated with an increased risk to develop AIDS following seroconversion in a univariate model [RH 2.7 (95% CI, 1.0–6.8) and 2.5 (95% CI, 1.0–6.3), respectively; Table 2]. In a multivariate model including both parameters, the RH values increased to 3.5 (95% CI, 1.3–9.3) and 3.1 (95% CI, 1.2–8.2), respectively (Table 2).
This study demonstrated for the first time that low preseroconversion numbers of CD4 T cells and increased levels of immune activation were associated with an increased risk to develop AIDS after seroconversion. In addition, the predictive value of immune activation increased over time after HIV-1 seroconversion and involved not only CD8 T cells, as reported previously, but also CD4 T cells; this could only partly be explained by CD4 T cell numbers and plasma HIV-1 RNA. Not all activation markers had equally predictive value, possibly reflecting their different roles in amplifying and maintaining immune responses.
Therefore, our data show that the size of the CD4 T cell pool and the level of immune activation that persists after the phase of acute infection are important determinants for HIV-1 disease progression. In vivo labelling of dividing T cells in HIV-1-infected individuals has demonstrated that increased levels of immune activation involves a subset of CD4 and CD8 T cells that rapidly divide, expand and die, most likely through activation-induced cell death [14,23,24]. Because of this initial expansion, increased death of activated T cells by itself cannot explain T cell depletion [24,25]. However, continuous recruitment of T cells to the pool of rapidly proliferating and dying T cells may cause gradual depletion of quiescent naive T cells because the capacity to replace lost naive T lymphocytes is limited in adults [10,26]. Indeed, patients who progressed to AIDS within 7 years after seroconversion showed a significant decline of both naive CD4 and naive CD8 T cell numbers . The availability of preseroconversion samples allowed the demonstration that both the size of the CD4 T cell pool and the level of immune activation therein before HIV-1 infection were important determinants for subsequent disease progression. Together, these data support our primary hypothesis that continuous immune hyperactivation associated with HIV-1 infection may lead to continuous activation and differentiation of naive T cells, thereby accelerating depletion of the naive T cell pool, which is already small in some patients before HIV-1 infection. The finding that disease progression is more rapid in subjects with increased levels of immune activation before HIV-1 infection is compatible with the poor prognosis reported previously in patients carrying HLA-A1 HLA-B8 HLA-DR3, who are of the high immune-responder phenotype . Interestingly, emergence of CXCR4-using HIV-1 variants, which occurred in 28 patients, was associated with increased levels of immune activation and may in that way accelerate activation-induced erosion of the naive T cell pool .
Whereas HIV-1-induced immune activation involves both CD4 and CD8 T cells, only peripheral blood total CD4 T cell numbers decline gradually; total CD8 T cell numbers typically remain elevated until late into HIV-1 infection. This can only be partly attributed to direct cytopathic effects of HIV, as the number of productively infected CD4 T cells is too low to account for all of the CD4 T cell loss [30–32]. Rather, different expansion rates of CD4 and CD8 T cells in viral infection may be involved. It was recently shown in mice that, following clearance of viral infection, numbers of antigen-specific CD4 memory T cells readily declined whereas the number of CD8 memory and effector T cells remained elevated for life . These data suggest that the lifespan of expanded CD4 and CD8 T cells differ considerably, which may be key to the differential kinetics and death rates of memory and effector CD4 and CD8 T cells in the course of HIV-1 infection and other conditions with persistent immune activation. Indeed, in HIV-1-infected individuals, the CD8 T cell memory/effector compartment was largely expanded, and telomere shortening (another measure of T cell replicative history) was found to be threefold more rapid in CD8 T cells than in CD4 T cells [34,35]. Taken together, HIV-1 infection may lead to continuous activation of naive T cells, thereby slowly depleting the naive CD4 and CD8 T cell pool; however, because of biological differences in primed CD4 and CD8 T cell kinetics, chronic immune hyperactivation may lead to a decline in total CD4 T cell numbers but to expansion of the memory/effector CD8 T cell compartment.
In conclusion, our data show that chronic immune activation and the size of the CD4 T cell pool are critical factors in HIV-1 pathogenesis, even when measured before seroconversion. It may well be that the level of immune activation elicited by HIV-1 determines the recruitment rate of naive T cells to the memory and effector T cell pools, which ultimately leads to erosion of the CD4 and CD8 naive T cell compartment.
The authors thank the participants of the Amsterdam Cohort Studies on HIV-1 infection and AIDS, and Drs R. P. van Rij and A. Krol for helpful discussions.
Sponsorship: This study was financially supported by the Dutch AIDS Foundation and the Dutch Foundation for Scientific Research.
1. Fahey JL, Taylor JMG, Detels R, Hofmann B, Melmed R, Nishanian P, et al.The prognostic value of cellular and serologic markers in infection with human immunodeficiency virus type 1.N Engl J Med
2. Fahey JL, Taylor JGM, Manna B, Nishanian P, Aziz N, Giorgi JV, et al.Prognostic significance of plasma markers of immune activation, HIV viral load and CD4 T-cell measurements.AIDS
3. Carbone J, Gil J, Benito JM, Navarro J, Muñóz-Fernández A, Bartolomé J, et al.Increased levels of activated subsets of CD4 T cells add to the prognostic value of low CD4 T cell counts in a cohort of HIV-infected drug users.AIDS
4. Froebel KS, Raab GM, D'Alessandro C, Armitage MP, Mackenzie KM, Struthers M, et al.A single measurement of CD38CD8 cells in HIV+, long-term surviving injecting drug users distinguishes those who will progress to AIDS from those who will remain stable.Clin Exp Immunol
5. Plaeger S, Bass HZ, Nishanian P, Thomas J, Aziz N, Detels R, et al.The prognostic significance in HIV infection of immune activation represented by cell surface antigen and plasma activation marker changes.Clin Immunol
6. Liu Z, Cumberland WG, Hultin LE, Prince HE, Detels R, Giorgi JV. Elevated CD38 antigen expression on CD8+ T cells is a stronger marker for the risk of chronic HIV disease progression to AIDS and death in the multicenter AIDS cohort study than CD4+cell count, soluble immune activation markers, or combinations of HLA-DR and CD38 expression.J Acquir Immune Defic Syndr
7. Liu Z, Cumberland WG, Hultin LE, Kaplan AH, Detels R, Giorgi JV. CD8+T lymphocyte activation in HIV-1 disease reflects an aspect of pathogenesis distinct from viral burden and immunodeficiency.J Acquir Immune Defic Syndr
8. Giorgi JV, Liu Z, Hultin LE, Cumberland WG, Hennessey K, Detels R. Elevated levels of CD38+ CD8+ T cells in HIV infection add to the prognostic value of low CD4+ T cell levels: results of 6 years of follow-up. The Los Angeles Center, Multicenter AIDS Cohort Study.J Acquir Immune Defic Syndr
9. Liu Z, Hultin LE, Cumberland WG, Hultin P, Schmid I, Matud JL, et al.Elevated relative fluorescence intensity of CD38 antigen expression on CD8+ T cells is a marker of poor prognosis in HIV infection: results of 6 years of follow-up.Cytometry
10. Hazenberg MD, Hamann D, Schuitemaker H, Miedema F. T cell depletion in HIV-1 infection: how CD4+ T cells go out of stock.Nat Immunol
11. Lens SMA, Tesselaar NA, van Oers MHJ, van Lier RAW. Control of lymphocyte function through CD27–CD70 interactions.Semin Immunol
12. Wolthers KC, Otto SA, Lens SMA, Kolbach DN, van Lier RAW, Miedema F, et al.Increased expression of CD80, CD86 and CD70 on T cells from HIV-infected individuals upon activation in vitro: regulation by CD4+ T cells.Eur J Immunol
13. Gerdes J, Lemke H, Baisch H, Wacker H-H, Schwab U, Stein H. Cell cycle analysis of a cell proliferation-associated human nuclear antigen defined by the monoclonal antibody Ki-67.J Immunol
14. Kovacs JA, Lempicki RA, Sidorov IA, Adelsberger JW, Herpin B, Metcalf JA, et al.Identification of dynamically distinct subpopulations of T lymphocytes that are differentially affected by HIV.J Exp Med
15. Mohri H, Perelson AS, Tung K, Ribeiro RM, Ramratnam B, Markowitz M, et al.Increased turnover of T lymphocytes in HIV-1 infection and its reduction by anti-retroviral therapy.J Exp Med
16. Hazenberg MD, Cohen Stuart JWT, Otto SA, Borleffs JCC, Boucher CA, de Boer RJ, et al.T cell division in human immunodeficiency virus (HIV-1)-infection is mainly due to immune activation: a longitudinal analysis in patients before and during highly active anti-retroviral therapy.Blood
17. Norusis MJ SPSS 7.5 for Windows. User's Guide.
Chicago, IL: Statistical Package for Social Sciences Inc.; 1997.
18. Anonymous. EGRET Users's Manual
. Seattle, WA: Statistics and Epidemiology Research, Corp.; 1992.
19. Saksela K, Stevens C, Rubinstein P, Baltimore D. Human immunodeficiency virus type 1 mRNA expression in peripheral blood cells predicts disease progression independently of the numbers of CD4+ lymphocytes.Proc Natl Acad Sci USA
20. Mellors JW, Munoz A, Giorgi JV, Margolick JB, Tassoni CJ, Gupta P, et al.Plasma viral load and CD4+ lymphocytes as prognostic markers of HIV-1 infection.Ann Intern Med
21. Gupta P, Kingsley L, Armstrong J, Ming Ding, Cottrill M, Rinaldo C. Enhanced expression of human immunodeficiency virus type 1 correlates with development of AIDS.Virology
22. de Wolf F, Spijkerman I, Schellekens PThA, Langendam M, Kuiken CL, Bakker M, et al.AIDS prognosis based on HIV-1 RNA, CD4+T cell count and function: Markers with reciprocal predictive value over time after seroconversion.AIDS
23. Meyaard L, Otto SA, Jonker RR, Mijnster MJ, Keet RPM, Miedema F. Programmed death of T cells in HIV-1 infection.Science
24. Grossman Z, Herberman RB, Dimitrov DS. T cell turnover in SIV infection.Science
25. Grossman Z, Meier-Schellersheim M, Sousa AE, Victorino RMM, Paul WE. CD4 T-cell depletion in HIV infection: are we closer to understanding the cause?Nat Med
26. Haase AT. Population biology of HIV-1 infection: viral and CD4+ T cell demographics and dynamics in lymphatic tissues.Annu Rev Immunol
27. Roederer M, Gregson Dubs J, Anderson MT, Raju PA, Herzenberg LA, Herzenberg L. CD8 naive T cell counts decrease progressively in HIV-infected adults.J Clin Invest
28. Steel CM, Ludlam CA, Beatson D, Peutherer JF, Cuthbert RJ, Simmonds P, et al.HLA haplotype A1 B8 DR3 as a risk factor for HIV-related disease.Lancet
29. Hazenberg MD, Otto SA, Hamann D, Roos MT, Schuitemaker H, De Boer RJ, et al.Depletion of naive CD4 T cells by CXCR4− using HIV-1 variants occurs mainly through increased T-cell death and activation.AIDS
30. Embretson J, Zupancic M, Beneke J, Till M, Wolinsky S, Ribas JL, et al.Analysis of human immunodeficiency virus-infected tissues by amplification andin situhybridization reveals latent and permissive infections at single-cell resolution.Proc Natl Acad Sci USA
31. Chun T-W, Carruth L, Finzi D, Shen X, DiGiuseppe JA, Taylor H, et al
. Quantification of latent tissue reservoirs and total body viral load in HIV-1 infection.Nature
32. Clark DR, de Boer RJ, Wolthers KC, Miedema F. T cell dynamics in HIV-1 infection.Adv Immunol
33. Homann D, Teyton L, Oldstone MBA. Differential regulation of anti-viral T-cell immunity results in stable CD8+but declining CD4+ T-cell memory.Nat Med
34. Wolthers KC, Wisman GBA, Otto SA, de Roda Husman AM, Schaft N, de Wolf F, et al.T-cell telomere length in HIV-1 infection: No evidence for increased CD4+ T cell turnover.Science
35. Wolthers KC, Noest AJ, Otto SA, Miedema F, de Boer RJ. Normal telomere lengths in naive and memory CD4+ T cells in HIV-1 infection: a mathematical interpretation.AIDS Res Hum Retroviruses