|Home | About | Journals | Submit | Contact Us | Français|
The human immunodeficiency virus (HIV)-mediated immune response may be beneficial or harmful, depending on the balance between expansion of HIV-specific T cells and the level of generalized immune activation. The current study utilizes multicolor cytokine flow cytometry to study HIV-specific T cells and T-cell activation in 179 chronically infected individuals at various stages of HIV disease, including those with low-level viremia in the absence of therapy (“controllers”), low-level drug-resistant viremia in the presence of therapy (partial controllers on antiretroviral therapy [PCAT]), and high-level viremia (“noncontrollers”). Compared to noncontrollers, controllers exhibited higher frequencies of HIV-specific interleukin-2-positive gamma interferon-positive (IL-2+ IFN-γ+) CD4+ T cells. The presence of HIV-specific CD4+ IL-2+ T cells was associated with low levels of proliferating T cells within the less-differentiated T-cell subpopulations (defined by CD45RA, CCR7, CD27, and CD28). Despite prior history of progressive disease, PCAT patients exhibited many immunologic characteristics seen in controllers, including high frequencies of IL-2+ IFN-γ+ CD4+ T cells. Measures of immune activation were lower in all CD8+ T-cell subsets in controllers and PCAT compared to noncontrollers. Thus, control of HIV replication is associated with high levels of HIV-specific IL-2+ and IFN-γ+ CD4+ T cells and low levels of T-cell activation. This immunologic state is one where the host responds to HIV by expanding but not exhausting HIV-specific T cells while maintaining a relatively quiescent immune system. Despite a history of advanced HIV disease, a subset of individuals with multidrug-resistant HIV exhibit an immunologic profile comparable to that of controllers, suggesting that functional immunity can be reconstituted with partially suppressive highly active antiretroviral therapy.
In the absence of antiretroviral treatment, many human immunodeficiency virus (HIV)-infected individuals achieve a steady-state level of viremia that remains stable over a period of years. Although the degree to which a host adaptive immune response contributes to this steady state remains controversial, a number of reports suggest an important role for HIV-specific CD4+ and CD8+ T-cell responses (1, 6, 22, 23, 25, 30, 31, 34, 43). However, even in untreated individuals with strong and broad HIV-specific T-cell responses, increased viral replication and accelerated CD4+ T-cell loss eventually occur. Several mechanisms likely contribute to this failure of the adaptive immune system to control viral replication on a durable basis, including abnormal signaling of CD8+ T cells through costimulatory molecules, decreased stores of perforin, impaired antigen presentation, abnormal T-cell differentiation, the emergence of immunologic escape mutations, and/or impaired CD4+ T-cell help (3, 13, 24, 29, 48, 49).
Although the capacity of HIV to stimulate the immune system likely maintains an adaptive immune response, it is also increasingly clear that the proinflammatory aspects of HIV replication can be harmful. HIV infection results in increased immune activation (32), which directly contributes to progressive CD4+ T-cell depletion, perhaps as a consequence of immunologic exhaustion and/or accelerated “aging” of the immune system (2, 21).
Antiretroviral therapy dramatically affects the complex relationship that exists between viral replication and the HIV-specific immune response. Complete suppression of viral replication results in rapid decline in the frequency of circulating, HIV-specific T cells, presumably as a consequence of reduced antigenic stimulation. In contrast, partial viral suppression with combination antiretroviral therapy is associated with high levels of circulating HIV-specific CD4+ and CD8+ T cells, particularly in those individuals who are able to maintain plasma HIV RNA levels in the low to moderate range (i.e., 75 to 10,000 copies RNA/ml) (12, 47). These individuals maintain a steady-state plasma HIV RNA level well below pretreatment levels and harbor virus that is genotypically and phenotypically resistant to therapy. Accordingly, we and others have argued that the generation of effective HIV-specific immune responses contributes in part to the durable partial control of the drug-resistant variant (12, 35, 47).
To better evaluate the immunologic factors associated with the control of drug-resistant HIV in these individuals, we have identified a group of antiretroviral drug-treated individuals who have low to moderate levels of drug-resistant HIV and have chosen to remain on a stable antiretroviral regimen. We refer to these individuals here as “partial controllers on antiretroviral therapy” (PCAT) and define them as individuals whose steady-state plasma HIV RNA levels on treatment were between 500 and 10,000 copies RNA/ml. Like the well-described cohorts of long-term nonprogressors and individuals who control HIV replication after treatment interruptions (17, 31, 37, 38), we believe that these individuals may provide valuable insights into the potential correlates of immune protection against HIV disease progression (12, 35, 47).
The present study was undertaken to more fully characterize the factors associated with the control of HIV replication in the presence and absence of antiretroviral therapy. The immunologic characteristics of PCAT patients were compared to those observed in a group of patients who durably controlled HIV replication in the absence of therapy (“controllers”). We also studied individuals on antiretroviral therapy with complete virologic suppression (“virologic responders”) and individuals with persistent plasma HIV RNA levels above 10,000 copies/ml (“noncontrollers”). Our primary objective was to characterize the immunologic correlates of virologic control in the setting of chronic HIV infection, focusing on the level of HIV-specific cytokine-producing cells, the differentiation state of HIV-specific T cells, and the levels of T-cell proliferation and activation across all T-cell subpopulations.
Blood samples were obtained from HIV-infected adults enrolled in the University of California, San Francisco SCOPE cohort. SCOPE is an ongoing prospective cohort study aimed at investigating the long-term clinical and immunological consequences associated with HIV infections and their treatment (18). Subjects in SCOPE are seen at 4-month intervals, at which time they complete interviewer and self-administered questionnaires and undergo routine real-time laboratory evaluation, including HIV plasma RNA level and CD4+ T-cell count.
From this cohort, we selected participants meeting criteria for one of four treatment groups: (i) PCAT, defined as antiretroviral drug-treated individuals with multidrug-resistant HIV and a steady-state plasma HIV RNA level between 500 and 10,000 copies/ml, (ii) virologic “controllers,” defined as antiretroviral drug-untreated individuals who have a steady-state plasma HIV RNA levels below 10,000 copies RNA/ml, (iii) virologic “noncontrollers,” defined as both antiretroviral drug-treated and untreated individuals with plasma HIV RNA levels of >10,000 copies/ml, and (iv) virologic “responders,” defined as antiretroviral drug-treated individuals with undetectable plasma HIV RNA levels.
The proportion of Gag-specific CD4+ and CD8+ T cells that express gamma interferon (IFN-γ) and/or interleukin-2 (IL-2) was measured using cytokine flow cytometry (CFC) (26). Freshly collected whole blood was stimulated with peptides spanning the entire p55 Gag sequence (15 amino acid peptides overlapping by 11 amino acids) (BD Biosciences, San Jose, CA) in the presence of brefeldin A for 6 h. Nonstimulated cells and superantigen staphylococcal enterotoxin B (Sigma Aldrich)-stimulated cells were used as negative and positive controls, respectively. Activated cells were fixed and permeabilized before incubation with anti-IFN-γ-fluorescein isothiocyanate (FITC), anti-IL-2- phycoerythrin (PE), anti-CD4-PE-Cy5, and anti-CD3-allophycocyanin (APC). The fraction of cytokine-secreting CD4+ and CD8+ T lymphocytes was determined by flow cytometry using a Becton Dickinson FACSCalibur. CD4+ T cells were identified as mononuclear cells on the basis of forward and side scatter that were CD3+ CD4+. CD8+ T cells were identified as mononuclear cells that were CD3+ CD4−.
A total of 1 × 106 to 2 × 106 cryopreserved peripheral blood mononuclear cells (PBMCs) were stimulated with 5 μg/ml p55 Gag peptide mix, 5 μg staphylococcal enterotoxin B, or with RPMI 1640 supplemented with 15% fetal calf serum. Cells were then incubated for 16 h at 37°C in a final volume of 200 μl, with brefeldin A added at a final concentration of 5 μg/ml 1 hour after the beginning of incubation. Cells were washed in phosphate-buffered saline (PBS) containing 2 mM EDTA and then with PBS containing 1% bovine serum albumin (Sigma-Aldrich). Cells were then stained with a panel of fluorescently labeled antibodies against the following cell surface markers: anti-CD45RA-PE (BD Biosciences), anti-CCR7-PE-Cy7 (BD Biosciences), anti-CD28-APC (BD Biosciences), anti-CD27-APC-Cy7 (eBiosciences, San Diego, CA), and anti-CD8-Cascade Blue (Nicole Blumgarth, University of California, San Diego). For discrimination of dead cells, 5 μg/ml ethidium monoazide (EMA; Molecular Probes, Eugene, OR) was included in the cocktail of antibodies (33). Cells were exposed to a 40-W fluorescent light bulb for 5 min prior to being washed with PBS containing 1% bovine serum albumin, fixed in 1% paraformaldehyde, and permeabilized in fluorescence-activated cell sorter permeabilizing solution (BD Biosciences) for 10 min prior to being stained with antibodies anti-CD3-PE-Texas Red (Beckman Coulter, Fullerton, CA) and either anti-IFN-γ-FITC (BD Biosciences), anti-IL-2-FITC (BD Biosciences), or anti-Ki67-FITC (BD Biosciences).
Data were collected on a FACSDiVa (BD Biosciences) flow cytometer and analyzed using FlowJo software (Tree Star, San Carlos, CA). Seven HIV-seronegative donors were screened for T-cell responses against HIV Gag peptides. These individuals had less than 0.03% of CD4+ or CD8+ T cells responding to HIV peptide pools (data not shown), and this threshold was used to define a positive cytokine response.
For each sample, staining by EMA was used to eliminate dead/dying cells from further analysis. EMAneg T cells were defined by CD3+ staining and then by forward and side scatter (Fig. (Fig.1A).1A). CD3+ CD8− cells and CD3+ CD8+ cells represented CD4+ and CD8+ T cells, respectively. The percentage of CD4+ and CD8+ cells producing cytokine was then determined after subtracting the cytokine-positive events seen in unstimulated controls (Fig. (Fig.1B).1B). The CD4+ and CD8+ T-cell populations were each divided into subpopulations based on their expression of CD45RA, CD27, CD28, and CCR7 (Fig. (Fig.1C).1C). This allowed for the potential recognition of 16 phenotypic subpopulations of CD4+ and CD8+ T cells. By using the same gates on the cytokine-producing cells, 16 phenotypes of HIV-specific CD4+ and CD8+ T cells were also defined.
Immunophenotyping was performed using cryopreserved PBMCs by seven-color multiparameter flow cytometry. Cells were thawed, immediately labeled by Indo-1 AM (Molecular probes, Eugene, OR) used here as a live/dead marker and subsequently stained for anti-CD38-FITC (BD Biosciences, San Jose, CA), anti-major histocompatibility complex class I antigen-PE (Dako, Carpenteria, CA), anti-CD45RA-ECD (Beckman Coulter, Fullerton, CA), anti-CD27-APC (eBioscience, San Diego, CA), anti-CD8- PE-Cy5.5 (Caltag, Burlingame, CA), anti-CD4-PE-Cy7 (BD Biosciences), and anti-CD3-APC-Cy7 (BD Biosciences). T-cell analysis was performed on gated populations of live cells as defined by Indo-1+ blue, then on a gated population of lymphocytes as defined from forward versus side scatter two-dimensional contour plots, and finally on a gated population of T cells as defined from anti-CD3+ versus Indo-1+ violet contour plots. Importantly, samples were all processed, labeled, stained, and acquired—at the same time—on a FACSDiVa flow cytometer, with an average of 250,000 ungated Indo-1+ PBMCs (live cells) recorded for each acquisition. T-cell subpopulations of memory, effector, and terminally differentiated effector CD4+ or CD8+ T cells were gated, respectively, on CD45RA-CD27+, CD45RA-CD27−, and CD45RA+ CD27− subpopulations of T cells. Naïve T cells were gated on CD45RAbright CD27+ cells. FLOW acquisition parameters for voltage gain and percent compensation were determined such that more than 99% of unstained lymphocytes exhibited fluorescence values of less than 10. This approach is similar to the Quantibrite approach (21); however, we did not convert the relative fluorescence units to number of CD38 molecules per cell.
Nonparametric tests were used for all analyses. Differences in variables between any two patient groups were analyzed using the Mann-Whitney U test. Spearman's rank correlation was used to determine correlations between variables.
To determine whether control of HIV replication is associated with a particular pattern of T-cell phenotype and/or function, three levels of analyses were applied to cells obtained from our patient cohort. In the first analysis (four-color flow cytometry), we used freshly collected blood from sequentially identified subjects to determine the proportion of Gag-specific CD4+ and CD8+ T cells. In the second analysis (eight-color flow cytometry), we used cryopreserved PBMCs to define the immunophenotypic profile and proliferation status of HIV-specific T cells. In the final analysis, we again used cryopreserved PBMCs to define the activation status of T-cell subpopulations. The same cohort was used for each analysis; however, because our initial study used fresh blood while the subsequent studies used archived PBMCs, not all subjects contributed to each analysis.
A total of 179 unique subjects were studied, including 61 antiretroviral-treated “responders” with undetectable HIV RNA levels, 35 antiretroviral untreated “controllers” with low-level viremia, 37 PCAT, and 45 “noncontrollers” with high-level viremia. The median CD4+ T-cell counts at the time of analysis were 487, 595, 286, and 216 cells/mm3, and the median HIV RNA levels were 1.88, 2.23, 3.47, and 4.57 log10 copies RNA/ml (for responders, controllers, PCAT, and noncontrollers, respectively). In general, the PCAT subjects had prior history of advanced disease (pretreatment nadir CD4+ T-cell counts of 90 cells/mm3 and pretreatment HIV RNA level of 5.0 log10 copies RNA/ml). PCAT subjects also had high-level phenotypic resistance to antiretroviral therapy and a low replicative capacity.
Although the ability of CD4+ and CD8+ T cells to produce IFN-γ in response to HIV antigens is a measure of virus-specific T-cell function, there have been conflicting reports regarding the relevance of such responses to protective immunity against HIV (4, 19). We therefore measured the percentage of CD4+ and CD8+ T cells producing IFN-γ, IL-2, or both in response to HIV p55 Gag peptide pools (Fig. (Fig.2).2). For this analysis, we studied 123 individuals (32 controllers, 31 PCAT, 19 noncontrollers, and 41 responders) (Table (Table1).1). Compared to noncontrollers, controllers had a higher percentage of circulating CD4+ T cells that secreted both IL-2 and IFN-γ (mean values of 0.27 and 0.07%, respectively; P = 0.003). There was no difference between the controllers and noncontrollers in terms of Gag-specific CD4+ T cells secreting only IL-2 or only IFN-γ (P = 0.09 and 0.81, respectively).
Having shown that the presence of IFN-γ- and IL-2-producing Gag-specific CD4+ T cells was a strong correlate of HIV control in the absence of therapy, we compared the controllers and noncontrollers to PCAT subjects. Although the PCAT patients had a history of progressive disease, the proportion of CD4+ T cells secreting both IL-2 and IFN-γ in response to Gag was much higher than that observed in noncontrollers (P = 0.02) and comparable to that in controllers (P = 0.55). Among Gag-specific CD8+ T cells, controllers had increased frequencies of IFN-γ+ IL-2+ CD8+ T cells compared to noncontrollers (P = 0.001) and compared to PCAT patients (P = 0.002) (Fig. (Fig.22).
When the controllers, PCAT, and noncontrollers were combined, there was a negative correlation between the frequency of IFN-γ+ IL-2+ CD4+ T cells and plasma HIV RNA levels (rho = −0.42; P < 0.001), as well as between the frequency of IFN-γ+ IL-2+ CD8+ T cells and plasma HIV RNA (rho = −0.43; P < 0.001). The association between plasma HIV RNA levels and the frequency of Gag-specific T cells producing IL-2 or IFN-γ alone was not significant.
In summary, our four-color CFC analysis in a large cohort of chronically infected individuals suggests that the presence of Gag-specific IFN-γ- IL-2-producing CD4+ T cells was the strongest correlate of control in the absence of therapy and that our PCAT group had levels of these cells comparable to that observed in those naturally controlling HIV replication.
To determine whether high levels of Gag-specific CD4+ T cells in our cohort were simply a consequence of low-level viremia, we compared the frequency of Gag-specific CD4+ and CD8+ T cells in patients who were aviremic in the absence of therapy (“elite” suppressors, n = 13) with those who were aviremic as a consequence of antiretroviral treatment (“responders,” n = 41). Those fully controlling HIV replication in the absence of therapy had consistently higher levels of all Gag-specific CD4+ T cells (e.g., P < 0.006 for elite suppressors versus responders with regard to Gag-specific IL-2+ IFN-γ+ CD4+ T cells, IL-2− IFN-γ+ CD4+ T cells, IL-2+ IFN-γ+ CD8+ T cells, and IL-2− IFN-γ+ CD8+ T cells) (Fig. (Fig.2C).2C). These data suggest but do not prove that these cells contribute to virologic control, at least in this subset of “elite” suppressors. However, because of the low level of Gag-specific CD4+ T cells in antiretroviral-treated patients, we can safely assume that the absence of viremia does not in and of itself result in high levels of HIV-specific cytokine responses.
We also asked whether there were differences in the differentiation status of cells that responded to HIV antigens among the various groups. Cryopreserved cells were obtained from 19 responders, 12 controllers, 22 PCAT, and 25 noncontrollers (Table (Table2).2). To perform this analysis, cells producing cytokine (IFN-γ or IL-2) in response to HIV were characterized by CD45RA, CCR7, CD27, and CD28, allowing discrimination of 16 phenotypically distinct subpopulations of HIV-specific CD4+ to CD8+ T cells (Fig. (Fig.1).1). Twelve of these phenotypically defined subpopulations are shown in Fig. Fig.3;3; the remaining four are not depicted because of their low frequency (<0.13% and <0.20% in the CD4+ and CD8+ T-cell populations, respectively). Based on prior studies that have examined T-cell function and telomere length, these phenotypically defined subpopulations have been arranged in a presumptive order of differentiation, although this arrangement may not reflect a truly linear progression from memory to effector functions (2, 39, 42).
The most predominant phenotype among both the CD4+ and CD8+ IFN-γ+ T cells was CD27− CD28− CCR7− CD45RA− (Fig. (Fig.3A).3A). We observed no significant difference between controller and PCAT patients with regard to the distribution of phenotypes. Compared to these two groups, virologic noncontrollers did not have decreased percentages of the most differentiated phenotypic subpopulations. Instead, they exhibited a higher proportion of Gag-specific IFN-γ+ T cells within the more differentiated phenotypes (e.g., CD27− CD28− CCR7− CD45RA+/−) and a lower proportion of T cells within the less differentiated phenotype (e.g., CD27+ CD28+ CCR7− CD45RA−) (Fig. (Fig.3A).3A). A similar trend was observed with the CD8+ IFN-γ+ T-cell subpopulation. Virtually all CD8+ IFN-γ+ T cells expressed a CCR7− phenotype. Among the CCR7− subpopulations, noncontrollers had a smaller proportion of less differentiated cells and a greater proportion of more differentiated cells. Thus, for both HIV-specific CD4+ and CD8+ IFN-γ+ cells, there was no evidence of a block in T-cell differentiation in individuals with high levels of viral replication.
The most abundant subpopulation among IL-2-producing CD4+ T cells was CD27+ CD28+ CCR7− CD45RA− (Fig.3B). This subpopulation was also well represented among the IFN-γ-producing T cells, suggesting that these cells reflect the IFN-γ+ IL-2+ CD4+ T cells observed in our cohort of individuals studied using four-color flow cytometry. For Gag-specific IL-2-producing CD4+ and CD8+ T cells, there were no significant differences between noncontrollers, and controllers, or PCAT with regard to the distribution of immunophenotypically defined subpopulations (Fig. (Fig.3B3B).
We next evaluated the association of HIV replication on T-cell proliferation (as defined by high-level Ki67 expression). Our primary hypothesis was that less differentiated T cells would be more quiescent (i.e., not proliferating) in the subset of individuals who were effectively controlling viral replication. As a group, controllers and PCAT subjects had lower levels of Ki67 expression in the total CD4+ T-cell population compared with noncontrollers (data not shown). We next asked whether the ability of CD4+ T cells to produce IL-2 might be related to the level of Ki67 expression in each of the phenotypically defined T-cell subpopulations (Fig. (Fig.4).4). Ki67 expression was quantified and compared between individuals who had detectable HIV-specific CD4+ IL-2 T-cell production and those who did not. Those with HIV-specific CD4+ IL-2 production exhibited decreased levels of Ki67 expression in the less differentiated subpopulations but had comparable Ki67 expression in the more differentiated subpopulations. Since IL-2 production is observed primarily in naïve and less differentiated memory CD4+ T cells (50), the ability to maintain low-level proliferation in these subpopulations may facilitate their maintenance over time, as they are less likely to be recruited into the shorter-lived effector memory/effector cell compartments.
Because a proinflammatory state may inhibit or reduce the capacity of the adaptive immune system to function efficiently (10), we measured T-cell activation in each of our treatment groups (10 in each of our four groups) (Table (Table3).3). Levels of T-cell activation were defined by the median fluorescence intensify of CD38 on CD4+ and CD8+ T-cell subpopulations. Naïve and memory T cells were defined based on the expression of CD45RA and CD27.
CD38 expression was lower in all memory CD8+ T-cell subpopulations (CD45RA− CD27+, CD45RA− CD27−, and CD45RA+ CD27−) in the controllers compared to the noncontrollers (P < 0.01 for each pairwise comparison) (Fig. (Fig.5).5). Similarly, CD38 expression was lower in PCAT subjects compared to the noncontrollers in each of these subpopulations (P < 0.01). The level of CD38 expression was comparable in controllers and PCAT subjects (P > 0.20 for each pairwise comparison). Similar trends were observed in the CD4+ T-cell population (data not shown).
In this study, we investigated HIV-specific immune responses and the differentiation patterns of circulating T cells in individuals with varying levels of viremia on and off antiretroviral therapy. A number of conclusions are evident from this analysis. First, individuals controlling HIV in the absence of therapy (controllers) have high numbers of IFN-γ- and IL-2-producing HIV-specific T cells, low levels of T-cell proliferation, and low levels of T-cell activation. The presence of HIV-specific IL-2-producing CD4+ T cells (most of which express markers associated with T-cell memory) appeared to be the most striking difference when we compared these individuals with those not controlling HIV. These data are consistent with recent observations from other groups (7, 15, 16). Second, individuals maintaining durable partial control of HIV replication on antiretroviral therapy (PCAT) exhibited many of the immunologic characteristics of those controlling HIV in the absence of therapy. Third, individuals with high levels of IL-2-producing cells generally exhibited low levels of T-cell proliferation-associated antigen Ki67 within the less differentiated T-cell subpopulations. Finally, we observed no evidence of a defect in differentiation in individuals with progressive disease. The degree to which cytokine-producing HIV-specific T cells differentiated from memory to effector cells (or late memory cells) was not impaired in those with high-level viral replication. Collectively, these data suggest that virologic control is associated with preservation of both IFN-γ- and IL-2-producing HIV-specific T cells, especially those of a less differentiated phenotype, and with low levels of T-cell proliferation. This immunologic state can be best characterized as one in which the host responds to HIV by expanding but not exhausting HIV-specific T cells while maintaining a relatively quiescent immune system.
One model to explain the relevance of IL-2-producing cells is that this marker may be associated with an improved proliferative capacity of these cells (20, 28, 50). Assuming that in vitro proliferation studies translate to higher levels of HIV-specific proliferation in vivo, this model suggests that IL-2 production facilitates constant replenishment of HIV-specific T cells in the face of chronic antigen stimulation. The preservation of high-level IL-2 production in controllers is consistent with this model.
To further explore the relationship between T-cell proliferation and IL-2 production, we measured Ki67 (a marker of T-cell proliferation) in 16 phenotypically distinct T-cell subpopulations. A careful assessment revealed low levels of Ki67 within the less differentiated CD4+ T-cell subpopulations in individuals with detectable HIV-specific IL-2 production. This finding suggests an alternative, or additional, mechanism by which IL-2 production may prove beneficial: by decreasing levels of T-cell turnover. This decreased level of T-cell proliferation in less differentiated subpopulations may preserve these subpopulations of cells by preventing their differentiation into the late memory/effector compartments. This latter model is supported by murine studies where a deficiency in IL-2 or IL-2Rβ results in rapid and severe autoimmunity, implying that the lack of IL-2 results in unchecked immune activation (40, 41, 46). In IL-2Rβ-deficient animals, such autoimmune manifestations are prevented by CD4+ CD25+ T regulatory cells, suggesting that IL-2 may play an important role in maintaining T regulatory cell numbers in vivo (27). In further support of this model, Sereti et al. have reported that individuals who have received therapeutic IL-2 have decreased levels of CD4+ T-cell proliferation in naïve and recall memory subsets (but not in the effector memory subset) and increased numbers of T regulatory cells (44). This group has also recently reported that individuals who received therapeutic IL-2 had increased numbers of foxP3+ T cells that exerted weak suppression of polyclonal naïve T-cell activation, suggesting that IL-2 administration results in decreased T-cell activation via T regulatory cells (45).
The relationship between T-cell differentiation and outcome has been controversial, with some studies indicating that HIV induces a “maturation block” and/or depletes HIV-specific T cells before they become fully effective (8, 9). Here, multiple markers of T-cell differentiation detailed the phenotypic characteristics of cytokine-producing HIV-specific T cells. No correlation was found between the level of plasma viremia and the differentiation profiles of HIV-specific CD4+ and CD8+ T cells producing IFN-γ or IL-2. If T-cell differentiation were impaired by HIV, one would expect to see lower percentages of more differentiated HIV-specific cells in those with higher levels of HIV replication. However, the opposite trend was observed. We believe that differentiation occurs as a consequence of antigenic stimulation and that the accumulation of more mature HIV-specific T cells and depletion of less differentiated T cells reflect an immune system that cannot effectively control the virus without exhausting reserves.
Since many individuals are unable to fully control viral replication with combination antiretroviral therapy, strategies aimed at enhancing immunologic control of drug-resistant HIV need to be considered (11, 36). We therefore set out to determine to what degree individuals maintaining control of drug-resistant HIV have an immunologic profile comparable to that observed in individuals who maintain control of HIV replication in the absence of therapy (“controllers”). Assuming that PCAT patients once had an immunologic profile comparable to that of the noncontrollers (a reasonable assumption given their low pretreatment CD4+ T-cell counts and high pretreatment viral loads), our data suggest that partially suppressive therapy in these individuals has resulted in an immunologic profile which shares some similarities to controllers (a group often referred to as long-term nonprogressors).
The primary limitation of this study is its cross-sectional nature. Defining cause and effect in such studies is difficult. For example, it is possible that the preservation of high levels of IL-2+ and IFN-γ+ cells may be a consequence of limited virus replication rather than a cause. However, our observation that Gag-specific T cells are consistently higher in aviremic untreated patients (“elite suppressors”) compared to aviremic treated patients suggests that the presence of these T cells is not a consequence of virus control. A second limitation pertains to the assays used. Our conclusions are based on the enumeration of cytokine-producing cells, whereas other studies have used major histocompatibility complex class I tetramers to quantify the HIV-specific CD8+ T-cell subpopulation (2, 9). There are several reasons why this approach was chosen. First, the use of tetramers limits the analysis to a small subset of HIV-specific cells, possibly those that are not representative of the T-cell response in aggregate (5). Second, recent studies have shown that the majority of tetramer-binding cells produce IFN-γ, increasing the number of HIV-specific cells included in the analysis (3, 14). Third, we used both IFN-γ- and IL-2-producing cells to define HIV specificity, which avoids skewing toward a particular T-cell subpopulation.
In summary, a comprehensive assessment of the immune system in those controlling HIV versus those not controlling HIV reveals a number of consistent trends across a variety of patient groups. Among antiretroviral-treated and untreated patients, durable control of HIV replication is associated with high levels of HIV-specific IL-2+ and IFN-γ+ CD4+ T cells, low levels of T-cell activation, and preservation of an expanded population of HIV-specific T cells with a less differentiated immunophenotype. This immunologic profile suggests that control of HIV in the setting of chronic disease may require durable maintenance of HIV-specific memory T cells and the absence of generalized immune activation. Our finding that such a state can be achieved in some patients with a history of progressive disease suggests that immunodeficiency associated with HIV may be reversible; hence, efforts at reconstituting a functional immune response to HIV should be pursued, particularly for those with limited antiretroviral treatment options. Our data defining correlates of control in the setting of chronic HIV infection may also be relevant to efforts aimed at developing preventative vaccines to control HIV replication in those who become infected (30).
This work was supported in part by grants from the NIAID (AI052745, AI055273, and AI44595), the California AIDS Research Center (CC99-SF and ID01-SF-049), the National Institutes of Health UCSF/Gladstone Institute of Virology & Immunology Center for AIDS Research (P30 MH59037 and P30 AI27763), the Center for AIDS Prevention Studies (P30 MH62246), and the General Clinical Research Center at San Francisco General Hospital (5-MO1-RR00083-37). Douglas F. Nixon and Joseph M. McCune are Elizabeth Glaser Scientists. J.M.M. is also a recipient of the Burroughs Wellcome Fund Clinical Scientist Award in Translational Research and the NIH Director's Pioneer Award.