Share this article on:

Lack of Epstein-Barr virus- and HIV-specific CD27− CD8+ T cells is associated with progression to viral disease in HIV-infection

van Baarle, Debbiea,b,*; Kostense, Stefana,*; Hovenkamp, Egberta; Ogg, Grahamc; Nanlohy, Neninga; Callan, Margaret FCc; Dukers, Nicole HTMd; McMichael, Andrew Jc; van Oers, Marinus HJb; Miedema, Franka,e

Basic Science

Objective: Despite readily detectable virus-specific CD8+ T cells in most HIV-infected patients, immune surveillance is eventually lost, leading to progression to AIDS. To investigate the underlying mechanism of this loss of immune control phenotypic analysis of HIV- and Epstein–Barr virus (EBV)-specific CD8+ T cells was performed.

Design: In three clinically distinct groups, long-term asymptomatics, progressors to opportunistic infections and progressors to EBV-associated non-Hodgkin lymphoma's (NHL), both number and phenotype of virus-specific CD8+ T cells was studied longitudinally.

Methods: The number of HIV- and EBV-specific T cells were determined using HLA-peptide tetrameric complexes. The phenotype of these virus-specific T cells was investigated by costaining with CD27 and CD45RO and thereby identifying specific subsets of CD8+ T cells.

Results: Individuals co-infected with HIV and EBV persistently had low numbers of HIV-specific CD27− T cells, in contrast to rising numbers of EBV-specific CD27− CD8+ T cells. However, HIV-infected individuals developing EBV-associated AIDS-related NHL had very low numbers of EBV-specific CD27− CD8+ T cells. Higher numbers of HIV-specific CD27− CD8+ T cells were associated with delayed disease progression. Virus-specific CD27− T cells, compared with CD27+ T cells showed elevated interferon-gamma production in response to viral peptides in vitro, indicative for strong effector function.

Conclusions: Taken together, our data indicate that virus-specific CD27− T cells may be important effector T cells in controlling chronic viral infections in humans and that lack of differentiation into CD27− effector T cells may lead to progression of viral disease.

From the aDepartment of Clinical Viro-Immunology, CLB/Sanquin & Landsteiner Laboratory of the Academic Medical Center, University of Amsterdam, the bDepartment of Hematology, Academic Medical Center, Amsterdam, The Netherlands, the cMolecular Immunology Group, Institute of Molecular Medicine, Oxford, UK, the dDepartment of Public Health, Municipal Health Service, Amsterdam and the eDepartment of Human Retrovirology, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands.

Note: *These authors contributed equally to the presented work.

Correspondence to D van Baarle, Dept. Clinical Viro-Immunology, CLB, Plesmanlaan 125, 1066 CX, Amsterdam, The Netherlands. Tel: +31 20 5123314; fax: +31 20 5123310; email:

Received: 5 October 2001; revised: 24 May 2002; accepted: 10 June 2002.

Back to Top | Article Outline


CD8+ MHC class I-restricted cytotoxic T-lymphocyte (CTL) responses are essential in controlling virus infections [1–3]. After exposure to virus, CD8+ T cells are activated to kill virus-infected cells or inhibit viral replication, which is achieved through diverse effector functions, such as cytolysis and cytokine release, most notably interferon gamma (IFNγ) [4,5].

In human viral infections such as HIV (the causative agent of AIDS), Epstein–Barr virus (EBV, a widespread persistent gamma herpesvirus), and cytomegalovirus (CMV, a widespread beta-herpesvirus), high levels of antigen-specific CD8+ T cells are found [6–9]. Most chronic infections are adequately controlled by virus-specific T cells. Although apparently HIV-specific T cells initially control viral replication, HIV-specific T-cell responses are lost in individuals progressing to AIDS [10]. In addition, HIV-infected individuals are at high risk for EBV-positive AIDS-related non- Hodgkin's lymphomas (AIDS-NHL) [11], which are thought to develop because of progressive loss of EBV-specific T-cell immunity [12].

Recently primed virus-specific T cells are mostly of the CD45RO+CD27+ phenotype with cytolytic activity only after restimulation [13]. These CD27+ T cells irreversibly switch off CD27 expression when stimulated for prolonged periods [14] and eventually revert to CD45RA+ T cells [15–18]. The readily detectable effector functions of CD27− T cells, like high granzyme B expression and direct cytolytic activity, would predict these cells to be critical in control of virus replication [19,20].

Significant numbers of CMV- [20,21], EBV [22] and hepatitis C virus-specific T cells [23] were reported to be of the CD27− phenotype. In contrast, in HIV-infected individuals the majority of HIV-specific T cells are of the CD27+ memory phenotype, whereas only minor populations of CD27− cells are found [7,21,24]. To define the role of CD8+ CD27−, so-called effector T cells in human viral infections in relation to clinical outcome, we investigated the differentiation status of HIV- and EBV-specific T cells in the course of HIV-1 infection. We studied the kinetics of virus-specific CD8+ T lymphocytes using MHC class I-peptide tetrameric complexes [25], and assessed their differentiation status applying the CD45RO and CD27 surface markers in different groups of HIV-infected individuals in the course of infection. In addition, we studied the functional consequences of differentiation towards the CD27− phenotype in these individuals by measuring IFNγ production in virus-specific T cells. Furthermore, in a cross-sectional analysis, the impact of higher percentages of HIV-specific CD27− T cells was examined in relation to clinical outcome and HIV RNA viral load.

Back to Top | Article Outline

Materials and methods

Study population

This study was performed on samples from participants of the Amsterdam Cohort studies on AIDS and HIV-1 infection and was approved by the ethical committee of the Academic Medical Center. Peripheral blood mononuclear cells (PBMC) were cryopreserved according to a standard computerized freezing protocol. HIV-seropositive male individuals were selected according to duration of follow-up, availability of samples and HLA-type (HLA-A2 and/or B8).

In a cross-sectional analysis, 33 HIV-1-infected individuals with a broad range of viral load were studied for the presence and phenotype of HIV-specific CD8+ T cells. Furthermore, we analysed longitudinal PBMC samples from nine HIV-1-infected individuals for HIV-specific CD8+ T cells and 11 HIV-1-infected individuals for EBV-specific CD8+ T cells (all during the natural course of HIV-infection), including four individuals who were studied for both EBV- and HIV-specific T cells. Five of the 11 individuals were selected because they progressed to AIDS-related diffuse large cell NHL, an EBV-related disease and therefore particularly interesting to study in the context of EBV-specific immunity. In addition, three individuals were selected who progressed to AIDS (classification of the Centers for Disease Control 1993) with opportunistic infections (AIDS-OI) and three who remained long-term asymptomatic (LTA) with CD4+ T-cell counts > 500 × 106 cells/l during more than 8 years of asymptomatic follow-up. The earliest time point studied was between 1 and 2 years after HIV-seroconversion or study entry. The latest time point studied for progressors was around AIDS-diagnosis and for LTA on average between 6 and 8 years after HIV-seroconversion or study entry. No time points were included at which the individuals were on anti-retroviral therapy. Clinical and laboratory findings of these HIV-1 infected individuals have been in part published elsewhere [7,26] and are summarized in Table 1.

Table 1

Table 1

Back to Top | Article Outline

Flow cytometry and tetramer staining

MHC class I tetramers complexed with EBV and HIV-peptides were produced as previously described [25]. The peptides used were two immunodominant epitopes from EBV lytic cycle proteins, the HLA A2-restricted epitope GLCTLVAML (A2-GLC) from BMLF-1 and the HLA B8-restricted epitope RAKFKQLL (B8-RAK) from BZLF-1, and one immunodominant epitope from the EBV latent antigen EBNA-3A, the HLA B8-restricted epitope FLRGRAYGL (B8-FLR). From HIV p17Gag and Pol proteins, SLYNTVATL and ILKEPVHGV were complexed with HLA-A2; p24Gag and Nef peptides EIYKRWII and FLKEKGGL were refolded in HLA-B8 proteins.

Four colour fluorescence analysis was performed. Briefly, PBMC were thawed and 1.5 × 106 cells were stained with appropriate MHC class I tetramers and fluorochrome-conjugated antibodies anti-CD8, anti-CD45RO, anti-CD27, for subset distribution analysis; anti-CD8, CD27 and anti-CD28 to further subdivide the CD27− population. To determine granzyme B (GB) content, GB antibodies were used (CLB). After staining, the cells were washed with Phosphate-buffered saline (PBS) containing bovine serum albumin (BSA) and fixed in PBS/1% paraformaldehyde (PFA), and at least 250000 events were acquired using a FACSCalibur flow cytometer (Becton Dickinson, San José, California, USA). Lymphocytes were gated by forward and sideward scatter. Data were analysed using the software program CELL Quest (Becton Dickinson).

Back to Top | Article Outline

Intracellular IFNγ staining after antigen-specific stimulation

Two million PBMC/ml were stimulated with 1μg EBV-peptide (RAK), used in the tetrameric complexes, or PMA/ionomycin (positive control) or not stimulated (medium alone as negative control) at 37°C for 4 h in the presence of 3 μM monensin [27]. After incubation, cells were washed and stained in PBS supplemented with 0.5% (v/v) BSA for 15 min with HLA-B8-RAK tetramers (PE) in case no intracellular IFNγ was stained. For phenotypic analysis, cells were also stained with PerCP conjugated monoclonal antibody (MAb) CD8 (Becton Dickinson), anti-CD27 (fluorescein isiothiocyanate, CLB) and anti-CD45RO (APC) (Becton Dickinson). After membrane staining, cells were washed with PBA and fixed with 4% paraformaldehyde, permeabilized (Permeabilization kit; Becton Dickinson) and stained intracellularly with IFNγ-PE (Becton Dickinson) for 30 min at 4°C. At least 200 000 events in the lymphogate were acquired using a FACSCalibur flow cytometer (Becton Dickinson). Negative control stimulation did not induce IFNγ production and PMA/ionomycin stimulation resulted in high levels of IFNγ production in both CD8+ and CD4+ T cells (data not shown) [27].

Back to Top | Article Outline

ELIspot assay for single cell IFNγ-release

IFNγ producing antigen-specific T cells were enumerated using IFNγ-specific ELIspot assays as previously described [28] using 96-well nylon-backed plates (Nunc, Roskilde, Denmark) and MAb from MABTECH (Stockholm, Sweden). PBMC were incubated overnight at 37°C in triplicate wells at 1 × 105 cells/well in case of HLA B8-restriced responses or 2 × 105 cells/well in case of HLA A2-restricted responses in the absence or presence of 2 μM peptide. As a positive control to test the capacity of PBMC to produce IFNγ in general, phytohemoagglutinin (PHA) (Murex Diagnostics, Dartford, UK) was added. Individual cytokine-producing cells were detected as dark purple spots which were counted after computerized visualization by a scanner (Hewlett Packard Company, Baise, Idaho, USA). The number of specific T-cell responders per 106 PBMC was calculated after subtracting negative control values. Because the percentage of dead cells and the percentage of CD8+ T cells was assessed in the same samples, the number of specific T-cell responders/106 living CD8+ T cells could be calculated.

Back to Top | Article Outline

Statistical analysis

For group comparisons Mann–Whitney tests were performed, using the software program SPSS 7.5 (SPSS Inc., Chicago, Illinois, USA). Cross-sectional data were analysed by Spearman correlation tests and analysis of variance stepwise regression analysis. Repeated measurements analyses were performed on longitudinal data after cube root transformation of all variables to obtain a normal distribution, which was necessary for the analyses. In the analyses, we corrected for dependency between observations within a person assuming a compound symmetry (CS) structure using the software program SAS/STAT (SAS Institute Inc. Cary, North Carolina, USA).

Back to Top | Article Outline


Stable numbers of HIV- and EBV-specific CD8+ T cells in the course of HIV-1 infection

To study HIV- and EBV-specific T cells in the course of HIV-infection, PBMC from HIV-infected individuals collected at several time points in the course of HIV-infection were stained using HLA-HIV-peptide and HLA-EBV-peptide tetrameric complexes (Fig. 1b, c; left panel). As shown in Figure 2, high numbers of HIV- (range 0.1–2.7% of CD8+ T cells) and EBV-specific T cells (range 0.1–3.7% of CD8+ T cells) were present. Overall, no change in the number of both HIV- (P = 0.97, Wilcoxon test) and EBV-specific T cells (P = 0.97, Wilcoxon test) was observed in the course of HIV-infection (Fig. 2), suggesting that the well-established immunodominant HIV- and EBV-peptides used in the tetrameric complexes were still recognized by specific CD8+ T cells after several years of HIV-positive follow-up.

Fig. 1.

Fig. 1.

Fig. 2.

Fig. 2.

Back to Top | Article Outline

Lack of differentiation of HIV-specific T cells to CD27− T cells, but not EBV-specific T cells, in the course of HIV-infection

To investigate the differentiation stage of HIV- versus EBV-specific T cells in individuals who experienced relatively long periods of high HIV-viraemia, we applied the CD27 and CD45RO T-cell antigen expression patterns, revealing distinct subsets of T cells (Fig. 1a). Many HIV-specific CD8+ T cells in a long-term asymptomatic (LTA) HIV carrier were directed against Gag. (Fig. 1c; left panel) Despite persistent active viral-replication (+/– 100 000 viral RNA copies/ml) which one would expect to drive HIV-specific T cells to the CD27− phenotype [14,29], costaining with CD8, CD45RO and CD27 revealed that almost all the HIV (B8-Gag)-specific CD8+ T cells were of the CD27+ memory (CD45RO+CD27+) phenotype (Fig. 1c; right panel). Most EBV-specific CD8+ T cells in a HLA-B8+ long-term HIV-EBV carrier were directed against the lytic epitope RAK and the majority was of the CD27+ memory phenotype (Fig. 1b; left panel). However, in contrast to HIV-specific T cells, a substantial proportion of the EBV-specific CD8+ T cells could also be found in the CD45RO+CD27− subpopulation. (Fig. 1b; right panel)

To investigate the development of HIV- and EBV-specific CD8+ T cells over time, a longitudinal study was performed in HIV-infected individuals. As shown in Fig. 3a for one long-term asymptomatic individual, early in HIV infection (within 3 years after study entry or HIV-seroconversion) the majority of the HIV- (left panel) and EBV-specific CD8+ T cells (middle panel) had a CD45RO+CD27+ phenotype. In contrast to the HIV-specific T cells, a substantial proportion (30–50%) of the EBV-specific T cells was of the CD27− phenotype already early in HIV- infection. In the course of HIV-1 infection, the HIV-specific CD8+ T cells remained of the memory phenotype, whereas the proportion CD27− EBV-specific T cells increased.

Fig. 3.

Fig. 3.

As a measure of the extent to which virus-specific T cells had differentiated to the CD27− stage, the ratio of CD27−/CD27+ memory virus-specific tetramer-binding T cells was determined for all individuals. (Fig. 3c) By doing so the fraction of naive cells, which can vary between individuals, was excluded. Early in HIV infection (within 3 years after study entry or HIV-seroconversion) the majority of the HIV- (n = 9, left panel) and EBV-specific CD8+ T cells (n = 11, right panel) had a low ratio of CD27− /CD27+ T cells, due to the low percentage of CD27− T cells. (Fig. 3c) In the course of HIV-1 infection, both the percentage of CD27− (data not shown) as well as the CD27−/CD27+ ratio of HIV-specific CD8+ T cells remained stable, whereas in most individuals, the percentage of CD27− and the CD27− /CD27+ ratio of EBV-specific T cells increased, indicating that the majority of EBV-specific CD8+ T cells had differentiated to and increasingly accumulated at the CD27−stage (Fig. 3c). Late in HIV-1 infection, between 6 and 12 years after study entry or HIV-seroconversion, both the percentage of CD27− as well as the ratio CD27−/ CD27+ was significantly higher (median = 1.1) for EBV-specific CD8+ T cells than for HIV-specific CD8+ T cells (median = 0.3) (P < 0.014, Mann–Whitney test). In addition, co-staining with CD28 revealed that both CD27− and CD27+ EBV- and HIV-specific T cells were CD28− (data not shown).

Back to Top | Article Outline

Lack of differentiation of EBV-specific T cells into CD27− T cells in AIDS-NHL patients

Although in HIV-infected individuals the median ratio of EBV-specific CD27−/CD27+ was increased, in a small group of individuals the EBV-specific CD8+ T cells maintained CD27 expression throughout follow up. To study whether this could be related to clinical outcome, we compared the ratio CD27−/CD27+ for AIDS-NHL patients (n = 5), who were likely to have defective EBV-specific immunity [12], with HIV-1-infected individuals who either progressed to AIDS with opportunistic infections (PROG/AIDS-OI, n = 3) or remained clinically stable (LTA, n = 3). As shown in Figure 3d (and Fig. 3b), in AIDS-NHL patients EBV-specific T cells predominantly were of the CD27+ phenotype (low CD27−/CD27+ ratio), whereas an increase in the CD27−/CD27+ ratio was observed for all EBV-epitope-specific T cells in the LTA group, due to a higher percentage of CD27− T cells. Later in HIV-infection, both the percentage of CD27− and the CD27−/CD27+ ratio in LTA was significantly higher (median = 1.48) than in AIDS-NHL patients (median = 0.43) (P < 0.034, Mann–Whitney test). Progressors to opportunistic infections (AIDS-OI) also showed a trend towards higher percentages of EBV-specific CD27− T cells (n = 3) (median ratio CD27−/CD27+ = 0.80). The maintenance of the CD27-expression in AIDS-NHL patients occurred despite high levels of EBV load in the PBMC, which was comparable with the EBV load in LTA and other progressors to AIDS (Table 1).

Thus, in AIDS-NHL patients, who are known to have failing immune control of EBV infection, we observed the same lack of differentiation to the CD27− effector phenotype for EBV-specific T cells as for HIV-specific T cells, which also fail to control HIV-infection.

Back to Top | Article Outline

Abundance of HIV-specific CD27− T cells is associated with slower progression to AIDS

Although in the majority of HIV-infected persons most HIV-specific T cells expressed CD27, in some HIV-infected individuals a higher percentage of CD27− within HIV-specific T cells was observed. To investigate the effect of the percentage of CD27− within HIV-specific T cells on disease progression we correlated the percentage of HIV-tetramer+ T cells lacking CD27 expression with viral load and months of AIDS-free follow up in HLA-A2 or HLA-B8 positive participants of the Amsterdam Cohort. In samples drawn between 1 and 3 years after seroconversion (n = 11) an inverse correlation between HIV viral load and months of AIDS-free survival was found (Spearman, R = 0.78, P = 0.005) (Fig. 4a) as described before [30]. For the same samples an inverse correlation was observed between the percentage of CD27− within HIV-specific T cells and viral load (R = 0.79, P = 0.004). (Fig. 4b) Interestingly, high percentages of HIV-specific CD27− cells detected between 1 and 3 years after seroconversion (n = 11) and in samples randomly drawn during HIV-infection (n = 33) correlated with delayed disease progression (Spearman: R = 0.77, P = 0.005 and R = 0.46, P = 0.006, respectively (Fig. 4c and d, respectively), whereas the total number of tetramer+ T cells did not correlate with protection from disease (R = 0.22, P = 0.49)(data not shown). In a multivariate stepwise analysis, the percentage of CD27− within HIV-specific T cells was predictive for progression rate independent of viral load (CD27−: β = 0.82, P = 0.002; viral load: β = 0.112, P = 0.75).

Fig. 4.

Fig. 4.

These data suggest that CD27− CD8+ T cells may be more efficient than HIV-specific T cells with other phenotypes, in particular CD27+ memory T cells, in controlling HIV-infection and delaying disease progression.

Back to Top | Article Outline

Virus-specific CD8+ CD27− T cells are high IFNγ producers

Although it is known that CD27− CD8+ T cells (total CD8+) express higher levels of granzymes and perforin and exert direct cytolytic activity, we have no understanding of the effect of CD27− differentiation on anti-viral CD8+ T-cell function. Therefore, we tested the relation between the percentage of virus-specific CD27− T cells and – as a read-out for effector function – the number of IFNγ-producing T cells. We measured the number of IFNγ-producing CD8+ T cells after stimulation with EBV-peptides for all individuals. To perform regression analysis we selected those HIV-1-infected individuals in whom a significant accumulation of CD27− within EBV-specific T cells was observed (three progressors to OI and three LTA). Using regression analyses (mixed linear model) the percentage of EBV-specific CD27− T cells measured at all time points (n = 34) was positively correlated with the number of EBV-specific CD8+ T cells producing IFNγ (β = 1.2), which was highly significant in multivariate (P < 0.001) analyses controlling for EBV load and the number of CD4+ T cells. (Fig. 4e)

To investigate in detail whether the overall increase in IFNγ-producing antigen-specific T cells was indeed caused by an increase in the percentage of virus-specific CD27− T cells, we compared simultaneously the distribution over the different T-cell subsets of EBV-specific T cells defined by tetramer-staining and by IFNγ production, using intracellular IFNγ staining after stimulation with the specific EBV peptide. In a long-term asymptomatic HIV-infected individual (LTA 0036) early in HIV-infection (left panels), tetramer+ T cells resided for 58% in the CD45RO+CD27+ memory and for 33% in the CD27− subset (Fig. 5a). In contrast only 28% of the IFNγ+ EBV-specific T cells, which were lower in number than the tetramer+ T cells, resided in the CD45RO+CD27+ memory and 60% in the CD27− subset (Fig. 5c), showing that IFNγ-producing cells were enriched in the CD27− subset compared with tetramer binding cells. Similar distribution patterns were observed for another LTA individual (LTA1160, Fig. 5d). Moreover, at the later time point the percentage of tetramer+ T cells (LTA0036) had not increased in the CD8+ T-cell population, but EBV-specific T cells became two-fold more enriched in the CD27− fraction (right panels (Fig. 5a). Concomittantly, the percentage of IFNγ-producing T cells had doubled from 0.16 to 0.30% and accumulated in the CD27− fraction (Fig. 5c). The other LTA individual (LTA1160) did show an increase in the number of tetramer-binding cells. (Fig. 5d) At the same time the EBV-specific cells became more enriched in the CD27− fraction (right panels (Fig. 5d). Concomittantly, the percentage of IFNγ-producing T cells had increased from 0.14 to 0.47% and accumulated mainly in the CD27− fraction (Fig. 5f).

Fig. 5.

Fig. 5.

Similar results were obtained when HIV-specific CD8+ T cells were studied after initiation of HAART, which led to increased percentages of B8-Nef-specific CD27− T cells and percentage of IFNγ production (data not shown). Thus, the overall increase in IFNγ-producing antigen-specific T cells was indeed related to an increase in the percentage of virus-specific CD27− T cells (Fig. 5)

To confirm that the observed higher percentage of IFNγ in CD27− T cells and the overall increase in IFNγ in the course of HIV-infection could indeed be token as a sign for effector function, we also stained tetramer+ T cells with GB antibodies to determine the GB content of the EBV-specific T cells of these two LTA individuals. As shown in Fig. 5b and e, the number of GB-positive CD8+ tetramer+ T cells increased in the course of infection from 18 to 26% in LTA0036 (Fig. 5b) and from 43 to 66% (Fig. 5e) in LTA1160. Furthermore, CD27− T cells consistently had higher GB content than CD27+ T cells in these two LTA individuals (LTA0036: 30% early and 37% late/LTA1160: 70% early and 74% late in CD27− versus LTA0036: 12% early and 10% late/LTA1160: 36% early and 58% late in CD27+ T cells). (Fig. 5b/e).

Back to Top | Article Outline


In this study, we investigated the differentiation status of HIV- and EBV-specific CD8+ T cells in the course of HIV-1 infection to evaluate the role of virus-specific CD27− CD8+ T cells in human viral infections in relation to clinical outcome. Our data indicate that differentiation of HIV- or EBV-specific T cells in the CD27− stage was associated with slower progression to AIDS or protection against AIDS-NHL, respectively. In AIDS-NHL patients, EBV-specific CD8+ T cells remained of the less differentiated CD27+ phenotype in the course of HIV-1 infection, despite high levels of EBV load in PBMC. For HIV-specific T cells the percentage of CD27− T cells appeared to be predictive of disease progression, independent of HIV serum load. Finally, we observed that virus-specific CD27− T cells showed elevated IFNγ production compared with CD27+ T cells in response to viral peptides in vitro, which is indicative of the strong effector function of these cells.

The observed defective maturation of HIV-specific T cells is compatible with the reported lower perforin levels in CD27+ T cells [21]. However, we show here that this defective maturation does not seem to be specific for HIV-specific T cells, since EBV-specific T cells in AIDS-NHL patients were also shown to be mainly of the CD27+ phenotype. We further show that high numbers of fully differentiated CD8+ T cells specific for HIV or EBV are associated with protection from progression to AIDS or AIDS-NHL, respectively. The accumulation of virus-specific CTL in the CD27− CD8+ population in these clinical conditions suggests that these T cells may be critical for controlling viral infections that are chronically active as reflected by active viral replication. This is compatible with our finding that these T cells have augmented antiviral effector function compared to CD27+ T cells, as reflected by the correlation between the percentage CD27− T cells and high IFNγ production by virus-specific CD8+ T cells. In addition, CD27− T cells have also been shown to contain more GB and perforin and exert a stronger direct cytolytic activity compared to CD27+ cells [19,20]. Moreover, in a previous study we have shown that HIV-peptide-specific T cells which contain higher percentages of CD27− T cells, also have a higher capacity to produce IFNγ after peptide stimulation [27].

Although IFNγ production does not infer direct cytolytic activity, it has been shown that effector functions such as cytolytic capacity and IFNγ production largely coincide [31]. In addition, the observed state of T-cell effector dysfunction in several clinical conditions has been shown to occur both at the level of IFNγ production and cytolytic activity [32,33]. Furthermore, in healthy individuals there is a good correlation between IFNγ-producing T cells and CTLprecursor (CTLp) frequencies [34]. Moreover, our own observations indicate that the number of EBV-specific CTLp [12] and IFNγ-producing T cells correlate as they both decrease in the course of HIV-infection in AIDS-NHL patients [12,26]. Interestingly, parallel analysis of IFNγ-producing EBV-specific T cells and GB-containing EBV-specific T cells showed that both increase in the course of HIV-infection and both are higher in the CD27− population of T cells.

One could argue that the tetramer-staining and IFNγ-producing CD8+ cells after peptide-stimulation might be CD8+ natural killer (NK) cells. However, cells that stain with HLA-peptide tetramers are T-cell receptor (TCR) alpha-beta postive cells (thus CD8+ T lymphocytes) and not NK cells. In addition, cells that respond with IFNγ production upon 8–9 aminoacid long peptides, recognize with their TCR these peptides in the context of MHC class I, an interaction specific for CD8+ T cells and not NK cells. Furthermore, CD8+ NK cells are mostly found not to be CD8 bright but CD8 dim cells. Tetramer-staining shows that the CD8-bright cells stain with the tetramer, and not the CD8 dim cells. Finally, all techniques to analyse virus-specific CD8+ T cells are highly validated using depletion and purification of certain cell populations and co-stainings with other CD-markers.

A large dissociation between tetramer+ and IFNγ+ EBV-specific T cells was observed in this study. The proportion of IFNγ-producing T cells within tetramer+ T cells has been shown to vary both in HIV-infected individuals as well as in healthy individuals and is on average 25% [26]. In HIV-infected individuals, we have shown that this percentage can be somewhat lower early in HIV-infection and further decreases in patients developing EBV-related disease, which is suggestive of dysfunction [26].

Since development of functional CTL is known to be dependent on CD4+ T-cell help [35,36] and HIV-specific CD4+ T cells are believed to be irreversibly lost during acute infection [35], HIV-specific CD8+ T cells may not be able to efficiently differentiate to CD27− T cells despite high HIV viraemia. Indeed, we observed significant numbers of CD27− cells only in LTA individuals that experience a relatively stable clinical course, have modest viral loads and are known to still have measurable HIV-specific CD4+ T-helper activity [37]. In contrast to HIV-specific helper T cells, it is most likely that EBV-specific CD4+ T cells were present before HIV seroconversion, and this CD4+ T-cell help may decrease more gradually. When EBV-specific CD4+ T- cell help is eventually lost, due to immune activation-induced increased EBV reactivation, the functional differentiation to EBV-specific CD27− T cells will also be hampered and the individuals become at high risk of developing NHL.

In addition, as shown previously, CD4+ T-cell numbers indeed correlated with the number of EBV-specific IFNγ-producing T cells, suggesting a role for CD4+ T cells in maintaining functional capacity of CD8+ T cells [26]. Furthermore, the expression of CD27 has been shown to decrease after interaction with its specific ligand, CD70 [38] and CD70 expression is upregulated by T helper 1 cytokines [39]. Therefore, it could be that helper T cells, directly or indirectly, contribute to this differentiation process of CD8+ T cells.

Despite the fact that high percentages of EBV-specific CD27− T cells seem to be important in controlling EBV, healthy EBV carriers are heterogeneous in the expression of T-cell markers such as CD27 and CD28 and usually have low levels of these so-called ‘effector’ T cells. A possible explanation for this phenomenon is that during HIV-infection there is a continuous antigenic trigger caused by a chronic high level of viral load, continuously driving EBV-specific T cells into the CD27− compartment and thereby inducing and sustaining a high percentage of CD27− T cells. In immunocompetent individuals, acute EBV-infection leads to a peak in viral load [40]. The initiation of a CTL response, also inducing CD28− and CD27− T cells [22], leads to a reduction in EBV load and subsequent vanishing of the ‘effector’ T cells.

In conclusion, we show that fully differentiated CD8+ T cells that are specific for HIV but also EBV are associated with protection from progression to AIDS or AIDS-NHL, respectively. Thus, our data indicate that impaired maturation of CD8+ T cells is not specific for HIV, but may well be a general phenomenon in other conditions with failing anti-viral or anti-tumour immune control. Since virus-specific CD27− T cells may be critical for the control of chronic active viral infections, our data support the notion that phenotypic analyses of virus or tumour-specific CD8+ T cells based on CD27 expression may be informative for protective immunity to virus-induced disease and tumour development.

Back to Top | Article Outline


This study was part of the Amsterdam Cohort Studies on AIDS and HIV-1 infection, a collaboration of the Municipal Health Service, the Academic Medical Centre and CLB. We thank Dr.M.Th.L. Roos and collaborators for T lymphocyte immunophenotyping and processing of patient samples. We thank Dr. D. Hamann and Dr. R.A.W. van Lier for critical reading of the manuscript.

Sponsorship: This study was financially supported by the Dutch Cancer Society (grant number 96-1168) and the Dutch AIDS Fund (grant number 1007 and 2461).

Back to Top | Article Outline


1.Doherty PC, Christensen JP. Accessing complexity: the dynamics of virus-specific T cell responses. Annu Rev Immunol 2000, 18:561–592.
2.Ahmed R, Gray D. Immunological memory and protective immunity: understanding their relation. Science 1996, 272: 54–60.
3.Schmitz JE, Kuroda MJ, Santra S, et al. Control of viremia in simian immunodeficiency virus infection by CD8+ lymphocytes. Science 1999, 283:857–860.
4.Guidotti LG, Ishikawa T, Hobbs MV, Matzke B, Schreiber R, Chisari FV. Intracellular inactivation of the hepatitis B virus by cytotoxic T lymphocytes. Immunity 1996, 4:25–36.
5.Slifka MK, Whitton JL. Activated and memory CD8+ T cells can be distinguished by their cytokine profiles and phenotypic markers. J Immunol 2000, 164:208–216.
6.Ogg GS, Jin X, Bonhoeffer S, et al. Quantitation of HIV-1-specific cytotoxic T lymphocytes and plasma load of viral RNA. Science 1998, 279:2103–2106.
7.Ogg G, Kostense S, Klein MR, et al. Longitudinal phenotypic analysis of human immunodeficiency virus type 1-specific T lymphocytes: correlation with disease progression. J Virol 1999, 73:9153–9160.
8.Callan MFC, Tan L, Annels N, et al. Direct visualization of antigen-specific CD8+ T cells during the primary immune response to Epstein–Barr virus in vivo . J Exp Med 1999, 187:1395–1402.
9.Jin X, Demoitie MA, Donahoe S, et al. High frequency of cytomegalovirus-specific cytotoxic T-effector cells in HLA-A*0201-positive subjects during multiple viral coinfections. J Infect Dis 2000, 181:165–175.
10.Klein MR, Van Baalen CA, Holwerda AM, et al. Kinetics of Gag-specific CTL responses during the clinical course of HIV-1 infection: a longitudinal analysis of rapid progressors and long-term asymptomatics. J Exp Med 1995, 181:1365–1372.
11.Kersten MJ, Van Gorp J, Pals ST, Boon F, Van Oers MHJ. Expression of Epstein–Barr virus latent genes and adhesion molecules in AIDS-related non-Hodgkin's lymphomas: correlation with histology and CD4-cell number. Leuk Lymphoma 1998, 30: 515–524.
12.Kersten MJ, Klein MR, Holwerda AM, Miedema F, Van Oers MHJ. EBV-specific cytotoxic T cell responses in HIV-1 infection: different kinetics in patients progressiong to opportunistic infection or non-Hodgkin's lymphoma. J Clin Invest 1997, 99:1525–1533.
13.Hamann D, Baars PA, Rep MHG, et al. Phenotypic and functional separation of memory and effector human CD8+ T cells. J Exp Med 1997, 186:1407–1418.
14.Hintzen R, de Jong R, Lens SMA, Brouwer M, Baars P, Van Lier RAW. Regulation of CD27 expression on subsets of mature T lymphocytes. J Immunol 1993, 151:2426–2435.
15.Akbar AN, Terry L, Timms A, Beverley PC, Janossy G. Loss of CD45R and gain of UCHL1 reactivity is a feature of primed T cells. J Immunol 1988, 140:2171–2178.
16.Okumura M, Fujii Y, Inada K, Nakahara K, Matsuda H. Both CD45RA+ and CD45RA- sunpopulations of CD8+ T cells contain cells with high levels of lymphocyte function-associated antigen-1 expression, a phenotype of primed T cells. J Immunol 1993, 150:429–437.
17.Hamann D, Roos MThL, Van Lier RAW. Faces and phases of human CD8+ T cell development. Immunol Today 1999, 20: 177–180.
18.Hamann D, Kostense S, Wolthers KC, et al. Evidence that human CD8+ CD45RA+CD27− cells are induced by antigen and evolve through extensive rounds of division. Int Immunol 1999, 11: 1027–1033.
19.Bachmann MF, Barner M, Viola A, Kopf M. Distinct kinetics of cytokine production and cytolysis in effector and memory T cells after viral infection. Eur J Immunol 1999, 29:291–299.
20.Kern F, Khatamzas E, Surel I, et al. Distribution of human CMV-specific memory T cells among the CD8+ subsets defined by CD57, CD27, and CD45 isoforms. Eur J Immunol 1999, 29: 2908–2915.
21.Appay V, Nixon DF, Donahoe SM, et al. HIV-specific CD8(+) T cells produce antiviral cytokines but are impaired in cytolytic function. J Exp Med 2000, 192:63–76.
22.Roos MThL, Van Lier RAW, Hamann D, et al. Changes in the composition of circulating CD8+ T cell subsets during acute Epstein–Barr virus and human immunodeficiency virus infection in man. J Infect Dis 2000, 182:451–458.
23.He XS, Rehermann B, Lopez-Labrador FX, et al. Quantitative analysis of hepatitis C virus-specific CD8+ T cells in peripheral blood and liver using peptide-MHC tetramers. Proc Natl Acad Sci USA 1999, 96:5692–5697.
24.Shankar P, Russo M, Harnisch B, Patterson M, Skolnik P, Lieberman J. Impaired function of circulating HIV-specific CD8+ T cells in chronic human immunodeficiency virus infection. Blood 2000, 96:3094–3101.
25.Altman JD, Moss PAH, Goulder PJR, et al. Phenotypic analysis of antigen-specific T lymphocytes. Science 1996, 274:94–96.
26.van Baarle D, Hovenkamp E, Callan MFC, et al. Dysfunctional Epstein-Barr virus (EBV)-specific CD8+ T lymphocytes and increased EBV load in HIV-1 infected individuals progressing to AIDS-related non-Hodgkin's lymphoma. Blood 2001, 98: 146–155.
27.Kostense S, Ogg GS, Manting EH, et al. High viral burden in the presence of major HIV-specific CD8+ T cell expansions: evidence for impaired T cell effector function. Eur J Immunol 2001, 31:677–686.
28.Lalvani A, Brookes R, Hambleton S, Britton WJ, Hill AVS, McMichael AJ. Rapid effector function in CD8+ memory T cells. J Exp Med 1997, 186:859–865.
29.De Jong R, Brouwer M, Hooibrink B, Van der Pouw Kraan TCTM, Miedema F, Van Lier RAW. The CD27- subset of peripheral blood memory CD4+ lymphocytes contains functionally differentiated T lymphocytes that develop by persistent antigenic stimulation in vivo . Eur J Immunol 1992, 22:993–999.
30.Mellors JW, Kingsley L, Rinaldo CR, Jr., et al. Quantitation of HIV-1 RNA in plasma predicts outcome after seroconversion. Ann Intern Med 1995, 122:573–579.
31.Hislop AD, Gudgeon NH, Callan MFC, et al. EBV-specific CD8+ T cell memory: relationships beteween epitope specificity, cell phenotype, and immediate effector function. J Immunol 2001, 167:2019–2029.
32.Lechner F, Wong DK, Dunbar PR, et al. Analysis of successful immune responses in persons infected with hepatitis C virus. J Exp Med 2000,191:1499–1512.
33.Lee PP, Yee C, Savage PA, et al. Characterization of circulating T cells specific for tumor-associated antigens in melanoma patients. Nat Med 1999, 5:677–685.
34.Tan LC, Gudgeon N, Annels NE, et al. A re-evaluation of the frequency of CD8+ T cells specific for EBV in healthy virus carriers. J Immunol 1999, 162:1827–1835.
35.Rosenberg ES, Billingsley JM, Caliendo AM, et al. Vigorous HIV-1 specific CD4+ T cell responses associated with control of viremia. Science 1997, 278:1447–1450.
36.Kalams SA, Buchbinder SP, Rosenberg ES, et al. Association between virus-specific cytotoxic T-lymphocyte and helper responses in Human Immunodeficiency Virus type 1 infection. J Virol 1999, 73:6715–6720.
37.Pitcher CJ, Quittner C, Peterson DM, et al. HIV-1-specific CD4+ T cells are detectable in most individuals with active HIV-1 infection, but decline with prolonged viral suppression. Nat Med 1999, 5:518–525.
38.Hintzen RQ, Lens SM, Beckmann MP, Goodwin RG, Lynch D, van Lier RA. Characterization of the human CD27 ligand, a novel member of the TNF gene family. J Immunol 1994, 152: 1762–1773.
39.Lens SM, Baars PA, Hooibrink B, van Oers MH, van Lier RA. Antigen-presenting cell-derived signals determine expression levels of CD70 on primed T cells. Immunology 1997, 90:38–45.
40.Thorley-Lawson DA, Babcock GJ. A model for persistent infection with Epstein–Barr virus: the stealth virus of human B cells. Life Sci 1999, 65:1433–1453.

CD27; Epstein–Barr virus; HIV; CD8+ T cells; disease progression

© 2002 Lippincott Williams & Wilkins, Inc.