CD8 T cells that recognize and kill cells expressing HIV antigens are important in containing early HIV infection [1-3]. Long-term non-progressors also tend to have intact cytotoxic CD8 T cell responses; however, activity is lost during disease progression [4-8].
Significant prognostic phenotypic changes in CD8 T cell subsets occur during HIV disease progression including increased numbers of CD8 T cells expressing CD38 and lacking CD28 [8-13]. CD8 T cells expressing CD38 and DR have immediate effector cytotoxic T lymphocyte (CTL) activity without stimulation in vitro in about 50% of asymptomatic patients . CD8+CD28- cells demonstrate HIV-specific immediate CTL effector activity, suggesting that the CD38 positive and CD28 negative populations overlap . Detection of CD8 HIV-specific binding cells by flow cytometry using tetrameric MHC-peptide molecules shows that the CD8+CD38+ population contains 100-400 times more than that measured by conventional precursor CTL (pCTL) assays [14,15].
Limiting dilution assays provide different information than HIV binding assays or immediate CTL effector assays, because they determine whether precursors with potential to respond can develop into effector CTL [16,17]. Newly made effector cells, detected by binding or immediate effector CTL assays, may not have the ability to proliferate in response to specific HIV antigens.
We hypothesized that the loss of CD8 function in HIV disease was related to loss of the important costimulatory molecule, CD28. CD28 negative cells are unresponsive to stimuli in vitro and are apoptotic . Most CD28-CD8+ T cells in controls coexpress the carbohydrate molecule, CD57. HIV-positive patients, however, have elevated numbers of CD28-CD8+ T cells that are both CD57 negative and CD57 positive . The effector or memory CTL capabilities of these two distinct CD28 negative subpopulations is unknown. To test the hypothesis that CD28 loss reduced precursor CTL, sorted subpopulations of CD8 T cells from asymptomatic HIV-infected patients were examined.
Materials and methods
Five out of the six HIV positive patients studied were long-term non-progressors with infection of more than 12 years‚ duration (Table 1); patient No.1 was recently infected (2years previously). All gave informed consent. Viremia was detected by an AIDS Clinical Trial Group protocol  sensitive to 500 copies/ml. Whole blood immunophenotyping was performed as described previously . Four of the patients have been studied since 1990; only patient No.6 had some loss of CD4 T cells. The other three patients were stable over time. Only patient No.6 and the newly infected patient No.1 were on protease inhibitors. The other four patients were on combinations of other drugs except patient No.2 who was untreated. Three out of the six had CD4:CD8 ratios >1 (patient Nos 1-3); the other three patients had CD4:CD8 ratios <1 (patient Nos 4-6). Three of the six patients had undetectable viral loads and the amount of HIV in the other three was below 5000 copies/ml. The patients were asymptomatic when studied, although CD4 cell counts ranged from 372 to 1862¥106/l. Patient No.6 has since had some opportunistic pulmonary infections and has failed one protease inhibitor and switched to another.
Cell lines and culture medium
Epstein-Barr virus (EBV)-transformed B lymphoblastoid cell lines (B-LCL) were established as described . RPMI 1640 was used for all cell cultures supplemented with 10% heat-inactivated fetal calf serum, L-glutamine (2mM), non-essential amino acids (0.1mM), sodium pyruvate (1mM), and antibiotic- antimycotic mixture. (GIBCO BRL Life Technologies, Gaithersburg, Maryland, USA).
Recombinant vaccinia viruses
Recombinant vaccinia viruses vVK1 (expressing HIV-1HXB1 Gag-Pol), vPE16 (expressing HIV-1BH10 Env) and WR (a control for the vaccinia recombinants) were provided by Dr. B. Moss (NIH, Bethesda, MD) and grown and titered on BCS-1 cells.
Autologous peripheral blood mononuclear cells (PBMC) were infected with recombinant vVK1 vaccinia virus at a multiplicity of infection of 3-5 for 14h followed by irradiation (2400 rad) and were added at 20000 cells/well as antigen presenting cells (APC), similar to another report . For allostimulation, irradiated MHC class I mismatched EBV B cell lines were used as APC at 7500 cells/well.
PBMC were isolated from heparinized blood by Ficoll-Hypaque gradients. CD8 T cells were enriched by negative selection using monoclonal antibodies to CD4, CD19, CD14 (Becton Dickinson, San Jose, California, USA) and CD16 (PharMingen, San Diego, California, USA), and removal of positive cells by anti-mouse Ig magnetic beads (Immunotech, Westbrook, Maine, USA). Purity of CD8 T cells was 80-90%, but 95% CD3 positive. PBMC (or CD8) were seeded at various concentrations in 96-well plates in a final volume of 200μl per well with recombinant human interleukin-2 at 20U/ml (R&D systems, Minneapolis, Minnesota, USA). The cultures were fed every 3-5days with medium plus interleukin-2. Approximately 2weeks later, each well was split into three wells for the CTL assay. All assays had evidence of growth in individual wells.
For sorting experiments, CD8 T cells were prepared by negative selection and stained with CD28 and CD57 antibodies (Becton Dickinson). Previous experiments revealed that the CD28 antibody had no effect on the ability to proliferate  however, CD8 could not be used as a selection agent because of interference with CTL development . The CD8 T cells were sorted into three subpopulations: CD28+CD57-, CD28- CD57-, and CD28-CD57+. Sort purity was checked by reanalysis and in most cases was above 90%. The contaminants for both the sorted CD28+CD57- and CD28-CD57+ cells were always CD28-CD57- cells, which had fewer precursors in most cases. Sorted subpopulations were plated as before for limiting dilution analysis.
Target B-LCL infected with recombinant vaccinia viruses were prepared by infecting B-LCL as before; they were then labeled with Na2 51CrSO4for 1h and washed three times. Cytolytic activity was determined by 51C release using U-bottomed 96-well plates containing 5000 target cells per well. Plates were incubated at 37°C for 4-5h. Supernatants were counted on a gamma counter, and percent specific lysis was determined from the formula:
L=10¥experimental release-spontaneous release
maximum release-spontaneous release
Maximum release was determined by lysis of targets in 2% Triton X-100. Spontaneous release was <15% of maximal release for all assays.
For allostimulation, the effector cells from limiting dilution cultures were tested on 51Cr-labeled EBV cell lines including the original stimulating cells, autologous targets and mismatched target cells.
Each well was regarded as positive for cytotoxicity when the 51Cr release exceeded a threshold of 10% specific lysis. The numbers of negative wells were calculated and precursor frequencies estimated using Poisson statistics. Reproducibility of the assays was within 10%.
HIV specific precursors
To examine whether HIV-specific precursors were confined to a specific CD8 subpopulation based on expression of CD28, PBMC were isolated, CD8 T cells purified by negative selection, stained for CD28 and CD57 and sorted into subpopulations. Table 1 shows the number of CD8 T cell precursors in the unsorted CD8 population and the proportion of total precursors in each sorted subpopulation based on the percentage of the subpopulation in the total CD8 population. This adjustment indicates the capability of a given CD8 population to respond to HIV in vivo. Fig. 1 shows the HIV-specific pCTL precursor frequency curves from patient No.5 with controls. The CD28+CD57- and CD28-CD57+ subpopulations but not the CD28-CD57- subpopulation had the greatest number of precursors. Only patient No.6 had high levels of pCTL in both CD28-negative subpopulations. Sorted CD8 subsets from uninfected controls had alloreactivity, but no HIV-specific activity (data not shown).
Because CD28-negative cells have reduced proliferative capacity in vitro, it was determined whether the failure to detect precursors was due to failure to proliferate. All of the sorted populations responded to allostimulation and at about the same frequencies (>10000/106 CD8 cells). The cells were thus able to respond to stimuli but did not contain HIV-specific CTL precursors (data not shown). Moreover, the the CD4:CD8 ratio, a marker of disease progression, was positively correlated with the fraction of HIV-specific pCTL that were CD28+CD57- (r2, 0.68). In two individuals with CTL precursors who had CD4:CD8 >1, HIV- specific pCTL of the CD28+CD57- subset predominated, whereas in the three patients with CD4:CD8 <1, the CD28-CD57+ subset had the most CTL precursors.
These results show that loss of CD28 on CD8 T cells does not result in obligatory loss of HIV-specific pCTL. The finding that pCTL exist in the CD28-CD57+ population was unexpected, as these cells are unresponsive to stimuli in vitro  and are apoptotic in controls and in HIV positive patients [18,23]. CD28 negative cells in HIV positive patients were reported as effector, not precursor CTL, although no distinction among CD28 negative subsets was made . We suggest that the CD28-CD57- subset might contain immediate effectors because this population did not have as many pCTL. The major CD28 negative population in controls, which are CD57 positive, can be cloned in vitro, with limited division potential .
Chronic stimulation in response to HIV activates CD8 T cells (CD38+DR+) and a greater number of pCTL have been detected in those patients with more activation [9,25]. However, those with normal numbers of CD4 T cells had fewer pCTL. Our data suggest that CD8 pCTL expressing CD57 develop during infection and may indicate an ongoing response to virus. Because CD28 T cells contain predominantly naive CD8 T cells and the number of CD28 cells diminishes during infection, CTL responses to new HIV epitopes may become more limited because of a reduction of T cell receptor diversity. This might be more important in adults because of a more limited capacity to replenish the naive CD8 subset. Another report shows that both CD28 and CD28-CD8+ T cells in HIV-infected patients have skewed Vß distributions and that the CD28 negative subset is probably derived from the CD28 positive subset . In our patients, there were clearly two CD28 negative populations, one expressing CD57 which contained HIV-specific pCTL and another CD28 negative population that was CD57 negative, which did not contain as many pCTL precursors. This CD28-CD57- population may be similar to CD8 pCTL found in CD28-/- and B7.1-/- knockout mice . Such mice develop potent primary CTL responses, but long-lived memory CTL are reduced.
Requirements for the generation and maintenance of CD28-CD57+ cells are not understood. In controls, the CD28-CD57+ population predominates and is characterized by shortened telomeres and oligoclonal Vß distribution suggesting antigen-driven responses [28,29]. In a recent paper, CD8 T cells were separated based on CD45Ra and CD27. HIV-specific pCTL in the CD45Ra-CD27+ subset of one subject were classified as memory CTL. However, this population expressed CD28 but not CD57 . In our patients with low CD4:CD8 ratios, the predominant pCTL clearly did not express CD28 and most were CD57 positive. Thus the CD8 T cell subset studied by previous investigators is probably CD28+CD57- and the patient probably had a normal CD4:CD8 ratio.
The flow cytometric assays measuring HIV-specific binding T cells are important for analysis of CD8 T cell function in HIV infection. However, there are differences between antigen binding, effector CTL, and pCTL, which these assays do not distinguish. The fact that pCTL specific for HIV can be recovered in the CD28-CD57+ population suggests that this may be an important cell population in vaccine and therapy studies.
The authors thank the Baylor College of Medicine flow facility for the sorting experiments, J. Masterson for vaccinia stocks and A. Wirt and S. Esquivel for secretarial help.
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
2.Legrand E, Pellegrin I, Neau D, et al. Course of specific T lymphocyte cytotoxicity, plasma and cellular viral loads, and neutralizing antibody titers in 17 recently seroconverted HIV type 1-infected patients. AIDS Res Hum Retroviruses
3.Price DA, Goulder PJR, Klenerman P, et al. Positive selection of HIV-1 cytotoxic T lymphocyte escape variants during primary infection. Proc Natl Acad Sci USA
4.Carmichael A, Jin X, Sissons P, Borysiewicz, L. Quantitative analysis of the human immunodeficiency virus type 1 (HIV-1)-specific cytotoxic T lymphocyte (CTL) response at different stages of HIV-1 and Epstein-Barr virus in late disease. J Exp Med
5.Rinaldo C, Huang XL, Fan Z, et al. High levels of anti-human immunodeficiency virus type 1 (HIV-1) memory cytotoxic T-lymphocyte activity and low viral load are associated with lack of disease in HIV-1-infected long-term nonprogressors. J Virol
6.Greenough TC, Brettler DB, Somasundaran M, Panicali DL, Sullivan JL. Human immunodeficiency virus type 1-specific cytotoxic T lymphocytes (CTL), virus load, and CD4 T cell loss: evidence supporting a protective role for CTL in vivo. J Infect Dis
7.Bariou C, Genetet J, Fuffault A, Michelet C, Cartier F, Genetet B. Longitudinal study of HIV-specific cytotoxic lymphocytes in HIV type 1-infected patients: relative balance between host immune response and the spread of HIV type 1 infection. AIDS Res Hum Retroviruses
8.Goulder PJR, Phillips RE, Colbert RA, et al. Late escape from an immunodominant cytotoxic T-lymphocyte response associated with progression to AIDS. Nature Med
9.Ho HN, Hultin LE, Mitsuyasu RT, et al. Circulating HIV-specific CD8+ cytotoxic T cells express CD38 and HLA-DR antigens. J Immunol
10.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. J Acquir Immune Defic Syndr
11.Brinchmann JE, Dobloug JH, Heger BH, Haaheim LL, Sannes M, Egeland T. Expression of costimulatory molecule CD28 on T cells in Human immunodeficiency virus type 1 infection: functional and clinical correlations. J Infect Dis
12.Fiorentino S, Dalod M, Olive D, Guillet JG, Gomard E. Predominant involvement of CD8+CD28- lymphocytes in human immunodeficiency virus-specific cytotoxic activity. J Virol
13.Choremi-Papadopoulou H, Giglis V, Gargalianos P, Kordossis T, Iniotaki-Theodoraki A, Kosmidis J. Downregulation of CD28 surface antigen on CD4+ and CD8+ T lymphocytes during HIV-1 infection. J Acqir Immune Defic Synd
14.Altman JD, Moss PAH, Goulder PJR, et al. Phenotypic analysis of antigen-specific T lymphocytes. Science
15.Ogg GS, Jin X, Bonhoeffer S, et al. Quantitation of HIV-1-specific cytotoxic T lymphocytes and plasma load of viral RNA. Science
16.Koup RA, Pikora CA, Luzuriaga K, et al. Limiting dilution analysis of cytotoxic T lymphocytes to human immunodeficiency virus gag antigens in infected persons: in vitro quantitation of effector cell populations with p17 and p24 specificities. J Exp Med
17.Pantaleo G, Koenig S, Baseler M, Lane HC, Fauci AS. Defective clonogenic potential of CD8+ T lymphocytes in patients with AIDS. J Immunol
18.Lewis DE, Ng Tang DS, Adu-Oppong A, Schober W, Rodgers JR. Anergy and apoptosis in CD8+ T cells from HIV-infected persons. J Immunol
19.Lin HJ, Myers LE, Yen-Lieberman B, et al. Multicenter evaluation of quantification methods for plasma human immunodeficiency virus type 1. J Infect Dis
20.Lloyd TE, Yang L Ng Tang D, et al. Regulation of CD28 costimulation in human CD8+ T cells. J Immunol
21.Tosato G. Generation of Epstein-Barr Virus (EBV)-immortalized B cell lines. Curr Prot Immunol
, 1991, p. 7.22.1-7.22.3.
22.Musey L, Hughes J, Schacker T, Shea T, Corey L, McElrath MJ. Cytotoxic T cell responses, viral load, and disease progression in early human immunodeficiency virus type-1 infection. New Engl J Med
23.Boudet F, Lecouer H, Gougeon ML. Apoptosis associated with ex-vivo down-regulation of Bcl-2 and up-regulation of Fas in potential cytotoxic CD8+ T lymphocytes during HIV infection. J Immunol
24.Azuma M, Phillips JH, Lanier LL. CD28- T lymphocytes: antigenic and functional properties. J Immunol
25.Ferbas J, Kaplan AH, Hausner MA, et al
. Virus burden in long-term survivors of human immunodeficiency virus (HIV) infection is a determinant of anti-HIV CD8+ lymphocyte activity. J Infect Dis
26.Mugnaini EN, Spukland A, Egeland T, Sannes N, Brinchmann JE. Demonstration of identical expanded clones within both CD8+CD28+ and CD8+CD28- T cell subsets in HIV type 1-infected individuals. T cell subsets in HIV type 1-infected individuals. Eur J Immunol
27.Shahinian A, Pfeffer K, Lee KP, et al. Differential T cell costimulatory requirements in CD28-deficient mice. Science
28.Morley JK, Batliwalla FM, Hingorani R, Gregersen PK. Oligoclonal CD8+ T cells are preferentially expanded in the CD57+ subset. J Immunol
29.Monteiro J, Batliwalla F, Ostrer J, Gregersen PK. Shortened telomeres in clonally expanded CD28- CD8+ T cells imply a replicative history that is distinct from their CD28+ CD8+ counterparts. J Immunol
30.Hamann D, Baar PA, Rep MHG, et al. Phenotypic and functional separation of memory and effector human CD8+ T cells. J Exp Med