HIV-1 replication in vivo is in part controlled by HIV-specific CD8 T lymphocytes [1,2] although the mechanism by which this control is exerted is poorly understood. A greater functionality of CD8 T cells is found in HIV nonprogressors and similar responses are responsible for resolution of chronic viral infections that are controlled through cell-mediated immunity . The success of CD8 T cell immune responses appears to rely on the type of CD8 cell response rather than the quantity of specific CD8 T cells . Indirect although compelling evidence for the role of CD8 T cells in controlling HIV replication arises from observations of high frequencies of amino acid sequence changes in cytotoxic T cell targets. Mutations occur specifically in major histocompatibility class I recognition sequences matching the human leukocyte antigen type of the infected individual; observation of rapid reversion to ‘wild type’ on transmission further indicates that many of the immune-driven mutations incur substantial fitness costs to the virus [5–8].
A number of studies have demonstrated that CD8 T cells can be infected with HIV-1 in vivo [9–15]. Infection of CD8 T cells occurs following activation of the cells and subsequent expression of CD4 [16,17], and infection of CD8+CD4+ T cells [CD8 double positive T cells (CD8 DP T cells)] both in vivo and in vitro is now well documented [15–18]. The generalized immune activation seen during HIV-1 infection leads to increased numbers of cells with an activated phenotype, a state that contributes to pathogenesis . In turn, lymphoid tissues are the main sites of residence for latent, yet replication-competent infected cells , a region where activated CD8 cells are attracted during chronic infection . While HIV-1 infection of CD8 DP T cells during primary infection has been shown , the frequency of infected cells increases during AIDS [9,15]. We have recently demonstrated that CD8-derived HIV-1 virions in plasma of infected individuals form a major component of total viral load , indicating that infection of CD8 T cells is numerically significant and contributes substantially to the replicating pool of HIV-1 in vivo.
HAART suppresses viral replication in the vast majority of individuals, with the rapid decrease in plasma HIV-1 resulting from clearance of activated, productively infected cells . Nonetheless, low viral replication has been shown to occur and can be detected in peripheral blood , the kinetics of which correlate with replication within lymphoid tissue  and contribute to slower clearance of infected cells . The exceedingly long half-life of latently infected, replication-competent cells  prevents the complete clearance of HIV-1 by HAART. Indeed, the half-life of HIV-1 proviral DNA sequences detectable in peripheral blood mononuclear cells (PBMC) (4–6 months) is similar to that of memory CD4 T cells , the apparent long-term reservoir .
To investigate the potential for infected CD8 T cells to contribute to long-term persistence of HIV-1 during suppression of viral replication, the rates of HIV-1 DNA clearance from both CD4 T cells and CD8 DP T cells were determined following a period of HAART (200–450 days).
HIV-positive study subjects about to undergo a course of HAART for the first time or following a period of treatment interruption were recruited from the Western General Hospital, Edinburgh [Table 1]. A pretreatment blood sample (20–50 ml) was taken immediately prior to the onset of therapy (median 4 days; range, 0–19) and a second sample 200–400 days after starting therapy. Details of drug regimens are shown in Table 2.
Lymphocyte isolation and proviral load estimation
Lymphocyte subsets [CD4: CD4+CD8β-; CD8 single positive (SP): CD4−CD8β+; CD8 DP: CD4+CD8β+] were isolated from each blood sample using a combination of magnetic cell enrichment and fluorescence activated cell sorting (FACS). Lymphocytes were purified from PBMC by Histopaque (Sigma, Poole, UK) density centrifugation followed by magnetic cell sorting using an autoMACS Separator (Miltenyi Biotec, Gladbach, Germany). CD8 T cells were purified by positive selection using CD8 MicroBeads (Miltenyi Biotec) and the negative fraction retained and used for purification of CD3 T cells by positive selection using CD3 MicroBeads (Miltenyi Biotec). FACS analysis was performed as described elsewhere  using CD8 cells for FACS purification of CD8 subsets and CD3 cells for FACS purification of CD4 T cells. HIV proviral loads were determined by limiting dilution PCR [15,31]. This assay is consistently sensitive to a single copy in our laboratory using DNA standards obtained from the National Institute for Biological Standards (Potters Bar, UK) and has previously been shown to estimate proviral loads that are significantly correlated with those estimated by real-time PCR . Proviral loads of CD8 T cell populations were adjusted according to the potential level of CD4 T cell contamination . Where both CD8 DP and CD8 SP T cell purity data were unavailable, the 75th percentile of all available CD4 cell contamination data was used (0.11%). Where cell numbers were limiting or provirus was detected from a single PCR, the standard error of the proviral load was not calculated.
Cell sorting analysis
To assess the number of activated cells of each phenotype, numbers of large and small CD4 and CD8 T cells pre-HAART and during HAART were determined by FACS analysis using postsorting, as described above, with the addition of tubulin phycoerythrin–cyanine-7 (clone TB1A337.3; Beckman Coulter, High Wycombe, UK) staining to exclude dead cells.
All statistical analyses were performed using SPSS version 14.0 (SPSS, Chicago, Illinois, USA). Comparison of paired sample means was performed using the Wilcoxon signed rank test. Nonparametric correlations were estimated using Spearman's rank correlation coefficient (rs).
To assess the number of activated cells of each phenotype, numbers of large and small CD4 and CD8 T cells pre-HAART and during HAART were determined by FACS analysis (Fig. 1a–d). The potential level of CD4 T cell contamination (Table 3) was used to adjust the proviral loads of CD8 T cell populations before and after HAART.
Pre-HAART proviral loads
Overall, 20 of the 30 study subjects (67%) had CD8 DP T cells that were HIV provirus-positive before HAART. Although 13 (43%) had detectable infected CD8 SP T cells, proviral loads were exceptionally low in this subset, approaching the sensitivity of the assay [mean, 2 copies/106 cells; 95% confidence interval (CI), 1–3]. The mean CD4 T cell proviral load for the 30 subjects (4098 copies/106 cells; 95% CI, 1575–6620) was significantly higher than that of CD8 DP T cells (258 copies/106 cells; 95% CI, 71–446) (Z = −4.659, P < 0.001).
There was a significant negative correlation between both CD4 and CD8 DP T cell pre-HAART proviral loads and CD4 cell count in the 30 subjects (CD4 cells: rs = −0.406, P = 0.026; CD8 DP: rs = −0.467, P = 0.009; Fig. 2a,b). The significant relationship between CD8 DP proviral load and CD4 cell count was not affected by the removal of PCR-negative samples (n = 20; rs = −0.495; P = 0.026). CD4 T cell proviral load significantly correlated with plasma viral load (n = 29; rs = 0.384; P = 0.040) (Fig. 2c), whereas no such relationship was true for CD8 DP T cell proviral load with (n = 29; rs = 0.090; P = 0.642) and without (n = 20; rs = −0.008; P = 0.974) the inclusion of PCR-negative samples (Fig. 2d).
HIV-1 DNA decay
HIV-positive CD4 T cells were detected during HAART in all 12 study subjects tested (Fig. 3a). Samples from three subjects yielded insufficient cells to estimate CD8 DP T cell proviral loads precisely during HAART and were excluded from the analysis (Table 1). In the remaining nine subjects, four had CD8 DP T cells in which HIV-1 proviral DNA was detectable (Fig. 3b). Although the lower limit of detection for the other five were low (< 3, < 11, < 15, < 22 and < 43 copies/106 cells), this does not preclude a level of infection below assay sensitivity. Where indicated, values halfway between the negative cut-off and zero have, therefore, been used for these PCR-negative CD8 DP T cell samples. The ratio of CD4 T cell proviral load to CD8 DP T cell proviral load pre-HAART and the corresponding ratio during HAART were strongly correlated (n = 9; rs = 0.891; P = 0.001), whereas there was no significant correlation between paired CD4 and CD8 DP T cell proviral loads pre-HAART (n = 30; rs = 0.251; P = 0.181) and during HAART (n = 9; rs = 0.033; P = 0.932). Only one study subject (WT02) had CD8 SP T cells that were HIV positive during HAART (7 copies/106 cells) although CD8 DP T cells from this sample had a considerably higher proviral load (182 copies/106 cells).
The median percentage reduction in proviral load was similar between CD4 T cells (64%; n = 12; 95% CI, 56–82) and CD8 DP T cells (82%; n = 9; 95% CI, 68–90). Moreover, there were no significant differences between the values for all eight paired samples (Z = −1.362; P = 0.173) and for just the four paired samples where CD8 DP T cell proviral load was positive during HAART (Z = −0.730; P = 0.465). Both T cell subsets showed a significant reduction in mean proviral load following therapy (CD4 T cells: n = 12; Z = −3.621; P < 0.001; CD8 DP T cells: n = 9; Z = −2.666; P = 0.008), irrespective of whether the detection limit or half its value was used as the proviral load for PCR-negative CD8 DP T cells (statistics identical).
The mean rate of HIV-1 DNA decay per month was very similar for both T cell subsets (CD4 cells: 0.071 log10 copies/106 cells; 95% CI, 0.035–0.107; CD8 DP: 0.077 log10 copies/106 cells; 95% CI, 0.047–0.108; Fig. 4). Although the true value for CD8 DP T cells may differ slightly from this value owing to the use of lower detection limits in its estimation, the mean rate of decay of HIV-1 proviral load for the four CD8 DP T cell PCR-positive samples (0.078 log10 copies/106 cells; 95% CI, 0.011–0.144) was almost identical to that estimated including the ‘negative’ samples. Individual values of CD8 DP proviral decay showed a significant negative correlation (albeit marginal) with the time to complete plasma virus suppression (< 50 copies/ml) (n = 9; rs = −0.667; P = 0.050) whereas no such relationship was true for CD4 T cell decay (n = 12; rs = −0.168; P = 0.602).
Frequencies of CD4+CD8+ T cells in peripheral blood
Before the onset of HAART, CD8 DP cells represented an average of 2.18% of CD8 lymphocytes (n = 29; 95% CI, 1.75–2.60). After HAART, this value had fallen significantly (n = 11; Z = −2.934; P = 0.003) to 0.64% (n = 12; 95% CI, 0.33–0.94) (Fig. 3c).
Decline in large T cells during HAART
CD4 and CD8 DP T cells showed significant overall reduction in the numbers of large cells following HAART (CD8 DP cells: n = 8; Z = −2.240; P = 0.025; CD4 cells: n = 9; Z = −2.192; P = 0.028; Fig. 5). No significant correlation was found between the percentage reduction in HIV-1 proviral load and the reduction in the percentage of large cells of the same subset (CD4: n = 9; rs = −0.317; P = 0.406; CD8 DP: n = 6; rs = −0.314; P = 0.554).
Although the occurrence and mechanisms of infection of CD8 DP T cells in vivo and in vitro are well established, its significance to HIV-1 pathogenesis remains uncertain [9–15]. While the frequency of CD8 T cell infection increases with disease progression [9,15], total frequencies of HIV-infected CD8 DP T cells are often low compared with that of CD4 T cells in peripheral blood. To an extent, it might be argued that activated CD8 DP T cells are underrepresented in PBMC populations because of their propensity to migrate from blood; proviral loads in CD8 T cells in lung have indeed been found to be several times higher than those recovered from PBMC [10,32]. Furthermore, we have recently demonstrated that plasma from HIV-1-infected individuals contains substantial amounts of CD8 cell-derived virus (15% estimated mean contribution to circulating plasma viral load), indicating that productive infection of CD8 T cells is substantial . This allows for a scenario by which HIV-induced destruction of these cells could contribute towards the immune dysfunction apparent during AIDS.
Heterogeneity in both plasma viral load and the number of productively infected cells prior to the start of HAART influences the kinetics of HIV-1 DNA decay. Moreover, decay kinetics are biphasic, with a variable time period for first-phase clearance [33,34], and combined analyses of decay rates may show no significant trend in the temporal reduction of HIV-1 DNA during therapy [35,20]. All study subjects here, however, showed reductions in proviral load for both cell types, and mean proviral loads were significantly reduced after the 6–13 months on therapy. Values derived for individual study subjects likely represent an average rate covering both the initial rapid decay during the first few months of therapy and the slower second phase . Importantly, CD8 T cell subsets were highly pure, with negligible CD4 T cell contamination (Table 3). In addition, only 7 of 17 study subjects tested had HIV-1-positive CD14+CD3− monocytes (with very low proviral loads, average 5 copies/106 cells; results not shown) ensuring that CD8 T cell proviral loads are not obscured by contamination with other infected cell types. Our finding of a significant reduction in large CD4 and CD8 DP T cells during HAART most probably represents clearance of activated, acutely infected cells , as the increased size of T cells has been shown to be a sensitive corollary of activation .
Although the kinetics of HIV-1 clearance from CD4 T cells by HAART has been studied [28,36,37], this is the first study of the rate of clearance of HIV-1-infected CD8 T cells. HAART has previously been shown to reduce numbers of activated CD8 T cells [38,39], an observation confirmed here by the significant reduction in numbers of CD8 DP T cells (as well as CD4 T cells) in peripheral blood following HAART. The action of HAART on HIV-1 replication appears to be similar for both CD4 T cells and CD8 DP T cells. HIV-1-infected CD8 DP T cells were cleared at a similar rate to that of infected CD4 T cells, while the relative level of infection of CD8 DP T cells to that of CD4 T cells remained unaltered. Despite this similarity, it is interesting to note that the time taken to suppress viral replication below detectable levels (< 50 copies/ml in peripheral blood) correlated significantly with the rate of decay of HIV-1-infected CD8 DP T cells, but not with the decay of infected CD4 T cells. This indicates that it is the removal of HIV-1 infection from the CD8 T cell compartment which is crucial to the speed of successful suppression of viral replication. The small sample size of this study, coupled with the similar structure of most HAART regimens, prevents any further analysis based upon specific drugs/regimens.
HAART undoubtedly reduces the activation status of the immune system, observed here by a decrease in the numbers of circulating CD8 DP T cells. Moreover, it appears that HAART leads to a preferential clearance of HIV-1-infected large, activated T cells (both CD4 and CD8). Both of these mechanisms contribute to a reduction in the CD8-specific HIV-1 reservoir. Nonetheless, prior to HAART, activated CD8 T cells may enter the resting pool . Although the majority of effector T cells undergo apoptosis, a minority (∼10%) persist to form memory cells, many of which can be maintained for long periods of time . This situation provides a means for the establishment of a CD8-based HIV-1 reservoir that is not readily cleared by therapy. The lack of complete clearance of HIV-1-infected CD8 DP T cells found here provides further support for a pathway that creates an ideal situation for HIV-1 to establish a CD8 T cell reservoir analogous to the well-characterized reservoir in CD4 memory T cells. CD8 T cells should, therefore, be considered a potentially significant long-term reservoir for HIV-1, particularly given their significant contribution to the replication population of HIV-1 in untreated individuals .
The authors would like to thank the staff of the Regional Infectious Diseases Unit, Western General Hospital, Edinburgh for their invaluable help in collection of blood samples and provision of clinical data. We also thank Alison Hardie for assistance with the processing of samples, Shonna Johnston for FACS sorting of cells and Fraser Lewis for statistical advice.
Sponsorship: This study was funded by the Wellcome Trust.
1. Borrow P, Lewicki H, Hahn BH, Shaw GM, Oldstone MBA. Virus-specific CD8+ cytotoxic T-lymphocyte activity associated with control of viremia in primary human immunodeficiency virus type 1 infection. J Virol 1994; 68:6103–6110.
2. Koup RA, Safrit JT, Cao Y, Andrews CA, McLeod G, Borkowsky W, et al. Temporal association of cellular immune responses with the initial control of viremia in primary human immunodeficiency virus type 1 syndrome. J Virol 1994; 68:4650–4655.
3. Betts MR, Nason MC, West SM, De Rosa SC, Migueles SA, Abraham J, et al. HIV nonprogressors preferentially maintain high functional HIV-specific CD8+ T cells. Immunobiology 2006; 107:4781–4789.
4. Pantaleo G, Koup RA. Correlates of immune protection in HIV-1 infection: what we know, what we don't know, what we should know. Nat Med 2004; 10:806–810.
5. Phillips RE, Rowland Jones S, Nixon DF, Gotch FM, Edwards JP, Ogunlesi AO, et al. Human immunodeficiency virus genetic variation that can escape cytotoxic T cell recognition. Nature 1991; 354:453–459.
6. Goulder PJ, Brander C, Tang Y, Tremblay C, Colbert RA, Addo MM, et al. Evolution and transmission of stable CTL escape mutations in HIV infection. Nature 2001; 412:334–338.
7. Friedrich TC, Dodds EJ, Yant LJ, Vojnov L, Rudersdorf R, Cullen C, et al. Reversion of CTL escape-variant immunodeficiency viruses in vivo. Nat Med 2004; 10:275–281.
8. Leslie AJ, Pfafferott KJ, Chetty P, Draenert R, Addo MM, Feeney M, et al. HIV evolution: CTL escape mutation and reversion after transmission. Nat Med 2004; 10:282–289.
9. Livingstone WJ, Moore M, Innes D, Bell JE, Simmonds P, and the Edinburgh Heterosexual Transmission Study Group. Frequent infection of peripheral blood CD8-positive T-lymphocytes with HIV-1. Lancet 1996; 348:649–654.
10. Semenzato G, Agostini C, Chieco-Bianchi L, De Rossi A. HIV load in highly purified CD8+ T cells retrieved from pulmonary and blood compartments. J Leukoc Biol 1998; 64:298–301.
11. Imlach S, McBreen S, Shirafuji T, Leen C, Bell JE, Simmonds P. Activated peripheral CD8 lymphocytes express CD4 in vivo and are targets for infection by human immunodeficiency virus type 1. J Virol 2001; 75:11555–11564.
12. McBreen S, Imlach S, Shirafuji T, Scott GR, Leen C, Bell JE, et al. Infection of the CD45RA+ (naive) subset of peripheral CD8+ lymphocytes by human immunodeficiency virus type 1 in vivo. J Virol 2001; 75:4091–4102.
13. Saha K, Zhang J, Gupta A, Dave R, Yimen M, Zerhouni B. Isolation of primary HIV-1 that target CD8+ T lymphocytes using CD8 as a receptor. Nat Med 2001; 7:65–72.
14. Brenchley JM, Hill BJ, Ambrozak DR, Price DA, Guenaga FJ, Casazza JP, et al. T-cell subsets that harbor human immunodeficiency virus in vivo: implications for HIV pathogenesis. J Virol 2004; 78:1160–1168.
15. Cochrane A, Imlach S, Leen C, Scott G, Kennedy D, Simmonds P. High levels of human immunodeficiency virus infection of CD8 lymphocytes expressing CD4 in vivo. J Virol 2004; 78:9862–9871.
16. Kitchen SG, Korin YD, Roth MD, Landay A, Zack JA. Costimulation of naive CD8+ lymphocytes induces CD4 expression and allows human immunodeficiency virus type 1 infection. J Virol 1998; 72:9054–9060.
17. Flamand L, Crowley RW, Lusso P, Colombini-Hatch S, Margolis DM, Gallo RC. Activation of CD8+ T lymphocytes through the T cell receptor turns on CD4 gene expression: implications for HIV pathogenesis. Proc Natl Acad Sci USA 1998; 95:3111–3116.
18. Kitchen SG, Jones NR, LaForge S, Whitmire JK, Vu B-A, Galic Z, et al. CD4 on CD8+ T cells directly enhances effector function and is a target for HIV infection. Proc Natl Acad Sci USA 2004; 101:8727–8732.
19. Grossman Z, Meier-Schellersheim M, Paul WE, Picker LJ. Pathogenesis of HIV infection: what the virus spares is as important as what it destroys. Nat Med 2006; 12:289–295.
20. Poles MA, Boscardin JW, Elliot J, Taing P, Fuerst MMP, McGowan I, et al. Lack of decay of HIV-1 in gut-associated lymphoid tissue reservoirs in maximally suppressed individuals. J Acquir Imuune Defic Syndr 2006; 43:65–68.
21. Biancotto A, Grivel J-C, Iglehart SJ, Vanpouille C, Lisco A, Sieg SF, et al. Abnormal activation and cytokine spectra in lymph nodes of persons chronically infected with HIV-1. Blood 2007; 109:4272–4279.
22. Cochrane A, Hughes GJ, Seaton RA, Simmonds P. First evidence of HIV infection of CD8 lymphocytes expressing CD4 during primary HIV-1 infection. AIDS 2005; 19:1237–1239.
23. Hughes GJ, Willey S, Cochrane A, Leen C, Bell JE, Simmonds P. Virus immunocapture provides evidence for CD8 lymphocyte derived HIV-1 in vivo. AIDS 2007; 21:1507–1513.
24. Perelson AS, Essunger P, Cao Y, Vesanen M, Hurley A, et al. Decay characteristics of HIV-1 infected compartments during combination therapy. Nature 1997; 387:188–191.
25. Havlir DV, Strain MC, Clerici M, Ignacio C, Trabattoni D, Ferrante P, et al. Productive infection maintains a dynamic steady state of residual viremia in human immunodeficiency virus type 1-infected persons treated with suppressive antiretroviral therapy for five years. J Virol 2003; 77:11212–11219.
26. Wong JK, Günthard HF, Havlir DV, Zhang Z-Q, Haase AT, Ignacio CC, et al. Reduction of HIV-1 in blood and lymph nodes following potent antiretroviral therapy and the virologic correlates of treatment failure. Proc Nat Acad Sci USA 1997; 94:12574–12579.
27. Ramratnam B, Mittler JE, Zhang L, Boden D, Hurley A, Fang F, et al. The decay of the latent reservoir of replication-competent HIV-1 is inversely correlated with the extent of residual viral replication during prolonged antiretroviral therapy. Nat Med 2000; 6:82–85.
28. Finzi D, Blankson J, Siliciano JD, Margolick JB, Chadwick K, Pierson T, et al. Latent infection of CD4+ T cells provides a mechanism for lifelong persistence of HIV-1, even in patients on effective combination therapy. Nat Med 1999; 5:512–517.
29. Strain MC, Little SJ, Daar ES, Havlir DV, Günthard HF, Lam RY, et al. Effect of treatment, during primary infection, on establishment and clearance of cellular reservoirs of HIV-1. J Infect Dis 2005; 191:1410–1418.
30. Han Y, Wind-Rotolo M, Yang H-C, Siliciano JD, Siliciano RF. Experimental approaches to the study of HIV-1 latency. Nat Rev Microbiol 2007; 5:95–106.
31. Simmonds P, Balfe P, Peutherer JF, Ludlam CA, Bishop JO, Leigh Brown AJ. Human immunodeficiency virus-infected individuals contain provirus in small numbers of peripheral mononuclear cells and at low copy numbers. J Virol 1990; 64:864–872.
32. Semenzato G, Agostini C, Ometto L, Zambello R, Trentin L, Chieco-Bianchi L, et al. CD8+ T lymphocytes in the lung of acquired immunodeficiency syndrome patients harbour human immunodeficiency virus type 1. Blood 1995; 85:2308–2314.
33. Yerly S, Perneger TV, Vora S, Hirschel B, Perrin L. Decay of cell-associated HIV-1 DNA correlates with residual replication in patients treated during acute HIV-1 infection. AIDS 2000; 14:2805–2812.
34. Mourand-Joubert L, Marcellin F, Launay O, Hugon-Guiramond S, Gérard L, Yeni P, et al. Contribution of cellular HIV-1 DNA quantification to the efficacy analysis of antiretroviral therapy. J Acquir Immune Defic Syndr 2005; 38:268–276.
35. Teague KT, Munn L, Zygourakis K, McIntyre BW. Analysis of lymphocyte activation and proliferation by video microscopy and digital imaging. Cytometry 1993; 14:772–782.
36. Chun T-W, Carruth L, Finzi D, Shen X, DiGuiseppe JA, Taylor H, et al. Quantification of latent tissue reservoirs and total body viral load in HIV-1 infection. Nature 1997; 387:183–188.
37. Zhou Y, Zhang H, Siliciano JD, Siliciano RF. Kinetics of human immunodeficiency virus type 1 decay following entry into resting CD4+ T cells. J Virol 2005; 79:2199–2210.
38. Lederman MM, Connick E, Landay A, Kuritzkes DR, Spritzler J, St. Clair M, et al. Immunologic responses associated with 12 weeks of combination antiretroviral therapy consisting of zidovudine, lamivudine, and ritonavir: results of AIDS Clinical Trials Group Protocol 315. J Infect Dis 1998; 178:70–79.
39. Smith K, Aga E, Bosch RJ, Valdez H, Connick E, Landay A, et al. Long-term changes in circulating CD4 T lymphocytes in virologically suppressed patients after 6 years of highly active antiretroviral therapy. AIDS 2004; 18:1953–1956.
40. Doeuk DC, Pickler LJ, Koup RA. T cell dynamics in HIV-1 infection. Annu Rev Immunol 2003; 21:265–304.
41. Miller JD, Masopust D, Wherry EJ, Kaech S, Silvestri G, Ahmed R. Differentiation of CD8 T cells in response to acute and chronic viral infections: implications for HIV vaccine development. Curr Drug Targets Infect Disord 2005; 5:121–129.
© 2008 Lippincott Williams & Wilkins, Inc.