Previous Article | Next Article ![]()
Journal of Virology, April 2007, p. 3465-3476, Vol. 81, No. 7
0022-538X/07/$08.00+0 doi:10.1128/JVI.02392-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Wisconsin National Primate Research Center, Madison, Wisconsin 53715,1 Department of Pathology and Laboratory Medicine, University of Wisconsin Medical School, Madison, Wisconsin 537062
Received 31 October 2006/ Accepted 16 January 2007
|
|
|---|
|
|
|---|
Although available data remain inconclusive, they strongly suggest that some ECs make successful natural immune responses to HIV. Vigorous HIV-specific CD4 (6, 47) and CD8 (15, 16) cellular immune responses in some controllers have been observed, suggesting that these responses actively contain infection. EC cohorts show enrichment with HLA alleles such as HLA-B27 and HLA-B57, offering indirect evidence that particular CD8+ T-cell responses are important in controlling infection (7, 8, 19-21, 24, 39, 42, 48, 53). Viral escape from an immunodominant epitope bound by HLA-B27 has been associated with eventual disease progression in individuals who had initially controlled infection (5, 10, 12, 13, 22). Furthermore, recent studies indicate that escape from CD8+-T-cell responses exacts a cost to viral fitness, since transmitted escape variants are lost in the absence of the selecting HLA allele (2, 26). Unfortunately, it has been difficult to perform conclusive immunological studies with human ECs. Immune responses are often undetectable in these patients, perhaps because viral replication is so low. Moreover, because the sequence of the infecting virus is unknown, it is difficult to synthesize peptides corresponding to the patients' autologous virus, further hindering our ability to define the entire repertoire of immune responses in these unusual individuals.
In examining set-point virus loads in a cohort of 181 Indian-origin rhesus macaques infected with the cloned virus SIVmac239, we discovered a group of animals with robust and durable control of viremia during chronic infection. Set-point viremia in these macaque ECs had remained below 500 vRNA copy eq/ml of plasma for 1 to 5 years. This EC cohort showed enrichment with certain major histocompatibility complex (MHC) class I alleles, particularly Mamu-B*17 (55). Because of this association between MHC class I alleles and elite control of simian immunodeficiency virus (SIV) infection, we hypothesized that specific CD8+ T-cell responses made by ECs played a crucial role in their effective control of highly pathogenic SIV. We further hypothesized that particular CD8+ T-cell responses were actively suppressing viral replication in these animals, even though viremia had been barely detectable for up to 5 years. Here we depleted six EC macaques of CD8+ cells and monitored viral replication and subsequent changes in adaptive immune responses. Because this treatment also depleted the animals of natural killer (NK) cells, we compared the abilities of NK and CD8+ T cells to suppress SIV replication in a novel ex vivo assay.
|
|
|---|
CD8+ T-cell and NK cell quantification. The kinetics of the depletion of the animals of CD8+ cell populations and the reemergence of these populations were monitored by staining Ficoll density gradient-purified peripheral blood mononuclear cells (PBMC) with fluorescently labeled antibodies specific for CD8 (clone DK25, labeled with phycoerythrin [PE] or allophycocyanin, from DakoCytomation, Carpinteria, CA), CD16 (clone 3G8, labeled with Pacific blue, from BD-Pharmingen, San Jose, CA), and CD3 (clone SP34-2, labeled with Alexa 700 or fluorescein isothiocyanate [FITC], from BD-Pharmingen). Briefly, 500,000 PBMC were incubated in 100 µl of complete tissue culture medium in the presence of the antibodies for 30 min at room temperature. Fluorochrome-stained samples were washed twice, fixed, and run on a BD-LSR-II flow cytometer (Becton Dickinson, San Jose, CA) using FACSDiva software. Data were analyzed by using FlowJo 6.1 software (Treestar, Ashland, OR). Absolute counts were calculated by multiplying the frequency of CD8+ T cells or CD3+, CD8+, CD16+ NK cells within the lymphocyte gate with the lymphocyte counts per microliter of blood obtained from matching complete blood count results.
Tetramer staining. Epitope-specific CD8+-T-cell populations were enumerated by staining PBMC with Mamu-A*01, -A*02, and -B*17 tetrameric complexes loaded with synthetic epitope peptides largely as described previously (45). Briefly, 0.5 to 2 million PBMC were suspended in 100 µl of complete culture medium and incubated for 1 h at 37°C with tetramers labeled with PE. Next, antibodies recognizing CD3 (FITC-labeled SP34-2) and CD8 (clone SK1 labeled with peridin chlorophyll protein; BD Biosciences, San Jose, CA) were added, and samples were incubated for an additional 40 min at room temperature. Cells were then washed twice with fluorescence-activated cell sorter (FACS) buffer (phosphate-buffered saline containing 2% fetal bovine serum) and fixed in 1% paraformaldehyde. Data were acquired on a FACSCalibur flow cytometer (Becton Dickinson) and analyzed using FlowJo software (Treestar).
ICS.
Intracellular cytokine-staining (ICS) assays were performed as described in detail previously (54). Briefly, 500,000 PBMC were incubated for 1.5 h at 37°C in 100 µl of complete medium with anti-CD28 (clone L293; BD-Pharmingen), anti-CD49d (clone 9F10; BD-Pharmingen), and synthetic peptides (pools of 10- to 15-mers or single peptides representing minimal optimal CD8+ T-cell epitopes) based on the SIVmac239 protein sequence. Then 10 µg of brefeldin A per ml was added to prevent protein transport from the Golgi apparatus, and the cells were incubated a further 5 h at 37°C. Cells were then washed and stained for surface expression of CD4 (clone SK3 labeled with allophycocyanin; BD Biosciences) and CD8 (clone SK1 labeled with peridin chlorophyll protein; BD Biosciences) and fixed overnight in 1% paraformaldehyde at 4°C. The following day, cells were permeabilized in FACS buffer containing 0.1% saponin and stained for expression of the cytokines gamma interferon (IFN-
; clone 4S.B3 labeled with FITC [BD-Pharmingen]) and interleukin-2 (IL-2; clone MQ1-17H12 labeled with PE [BD-Pharmingen]). Antibodies against these cytokines were titrated for optimal staining. After intracellular staining, PBMC were fixed in 1% paraformaldehyde for 2 h at 4°C. Events were then collected on a FACSCalibur flow cytometer and analyzed with FlowJo.
Quantification of viral RNA in plasma. vRNA was detected in EDTA-anticoagulated plasma by quantitative reverse transcription-PCR by using a modification of a previously published protocol (9). Viral RNA was isolated from plasma as described previously (9). Viral RNA was reverse transcribed and quantified using a one-step quantitative reverse transcription-PCR kit (Invitrogen, Carlsbad, CA) with the LightCycler 1.2 (Roche, Indianapolis, IN). The final reaction mixtures (20-µl total volume) contained 0.2 mM (each) deoxynucleoside triphosphates, 3 mM MgSO4, 0.015% bovine serum albumin, 150 ng of random hexamer primers (Promega, Madison, WI), 0.8 µl of SuperScript III reverse transcriptase and Platinum Taq DNA polymerase in a single enzyme mix, 600 nM (each) amplification primers (forward [SIV1552], 5'-GTCTGCGTCATCTGGTGCATTC-3', and reverse [SIV1635], 5'-CACTAGCTGTCTCTGCACTATGTGTTTTG-3'), and 100 nM probe (5'-6-carboxyfluorescein-CTTCCTCAGTGTGTTTCACTTTCTCTTCTGCG-3'). The reverse transcription reaction was performed at 37°C for 15 min and then at 50°C for 30 min. An activation temperature of 95°C for 2 min was followed by 50 amplification cycles of 95°C for 2 min and 62°C for 1 min with ramp times set to 3°C/s. Synthetic 10-fold serial dilutions of an SIV gag in vitro transcript served as an internal standard curve for each run. Copy numbers for samples were determined by interpolation onto the standard by curve using the LightCycler software version 4.0.
Ex vivo suppression assay.
PBMC were isolated by centrifugation over a Ficoll density gradient. To generate target cells, a portion of the PBMC were depleted of CD8+ cells by using anti-CD8 nonhuman primate microbeads on an AutoMACS bead separation unit (Miltenyi, Auburn, CA) according to the manufacturer's protocol. Depletions were >99% effective. We stimulated the CD8 fraction with 5 µg/ml concanavalin A (Sigma) for 18 to 24 h, after which cells were washed and incubated in R15-50 (RPMI medium-15% fetal calf serum-1% L-glutamine-50 U/ml interleukin-2 [National Institutes of Health AIDS Reference Reagent Program, Germantown, MD]) for an additional day. Target cells were then incubated with clonal SIVmac239 at a multiplicity of infection of 0.00005 for 4 h, washed twice, and added to coculture assay mixtures. To generate effector cells, PBMC derived from the same blood draw were cultured in R15-50 for 2 days. These effectors were labeled with anti-CD8 (clone SK1 labeled with FITC; BD Biosciences) and anti-CD3 (clone SP34-2 labeled with PE-Cy7; BD Biosciences) antibodies for live-cell sorting using a MoFlo cell sorter (DakoCytomation, Fort Collins, CO). Effector cells were either mock sorted (antibody labeled and run through the cell sorter), depleted of NK cells (CD3, CD8+), depleted of CTL (CD3+, CD8+), or depleted of all NK and CTL to >99% purity. After sorting, the viability and quantity of effector cells were confirmed by trypan dye exclusion. For the suppression assay, 150,000 target cells were mixed with autologous effectors at an effector/target cell ratio of
3:1. The absolute number of effector cells was adjusted to ensure that the number of non-CD8+ effector cells remained constant for each condition, and each population was tested in duplicate for each experiment. Cells were cultured in 2 ml of R15-50 in 24-well plates. Culture medium (0.5 ml) was removed on days 3, 5, and 7 for quantitation of viral RNA and replaced with fresh medium. On day 7, cells were removed from culture and stained for surface expression of CD3, CD4, and CD8 and intracellular Gag p27 (clone 55-2F12; National Institutes of Health AIDS Research and Reference Reagent Program) by using Fix and Perm (CALTAG, Burlingame, CA). Events were collected on a FACSCalibur flow cytometer and analyzed with FlowJo.
Viral genome sequencing. Complete SIV genomes were sequenced as described in a previous study (43). Eighteen primer pairs were used in a single-step reverse transcription-PCR to produce overlapping amplicons, which were sequenced on a 3730 DNA analyzer (Applied Biosystems, Foster City, CA). Sequences were assembled using CodonCode aligner (CodonCode, Deadham, MA). DNA sequences were conceptually translated by using the MacVector 8 trial version.
|
|
|---|
500 vRNA copy eq/ml of plasma) during chronic infection. Six ECs were used in the present study (Table 1); four of these expressed Mamu-B*17, and two did not. Data on the epitopic breadths of cellular immune responses to SIVmac239 during acute infection were available for four of these animals (data summarized in Table 2); we have reported on the acute-phase responses of animals 95096 and AJ11 previously (37, 44). Animals AJ11 and 01064 made particularly broad responses, with PBMC from each animal recognizing more than 20 separate 15-mer pools (Table 2). Animals 95096 and 95071 were tested by ICS for IFN-
(data not shown). These experiments revealed that CD8+ lymphocyte responses in animal 95096 were dominated by those directed to Tat pool A, which contained the Mamu-A*01-restricted Tat SL8 epitope, and responses in animal 95071 were dominated by those directed to Vpr pool B, which did not contain a known minimal optimal epitope. Acute-phase cellular immune responses in AJ11 and 01064 were assessed by using an IFN-
enzyme-linked immunospot assay. AJ11's acute-phase repertoire was dominated by a population responding to Nef pool C, which did not contain a previously described minimal optimal epitope. 01064 had essentially codominant populations recognizing Nef pool D, which contained the Mamu-A*02-restricted epitope Nef YY9-159, and Vif pool E, which contained no known epitopes. Although they were gathered from different experiments, the available data indicate that ECs targeted multiple SIV antigens during the acute phase of infection, before viremia was controlled (Table 2). |
View this table: [in a new window] |
TABLE 1. Animals used in this study
|
|
View this table: [in a new window] |
TABLE 2. Comparison of epitopic breadths and immunodominant populations detected by peptide-specific IFN- secretion in the acute phase, at the chronic-phase predepletion baseline, and postdepletion
|
, intravenously at 50 mg/kg on day 0. Circulating CD8+ T cells were almost completely removed from peripheral blood for approximately 3 weeks (Fig. 1a). CD8
is also present on NK cells (CD3, CD8+, CD16+), of which animals were depleted with the same kinetics as CD8+ T cells (Fig. 1b). While there was some variation among animals, the numbers of both CD8+ T cells and NK cells appeared to return quickly to predepletion levels.
![]() View larger version (16K): [in a new window] |
FIG. 1. Dynamics of CD8+ T cells and SIV replication in ECs treated with monoclonal antibody cM-T807. Animals were given 50 mg/kg of antibody on day 0. (a) CD3+, CD8+ T-cell dynamics. (b) CD3, CD8+, CD16+ NK cell dynamics. The NK cell count for animal 95071 at the predepletion time point is unavailable due to a technical error. (c) Virus replication kinetics. Set point indicates the geometric mean virus load for each animal from 12 weeks postinfection until CD8+ cell depletion. Black lines and filled symbols indicate ECs expressing the MHC class I allele Mamu-B*17. ECs that did not express this allele are represented with gray lines and open symbols.
|
Viremia was controlled when CD8+ cells repopulated the periphery. Peripheral CD8+ cells reappeared coincident with a reassertion of control over viral replication in all six ECs. Previous studies with normal-progressor macaques had shown that transient depletion of CD8+ lymphocytes resulted in increased viral burdens, but they had not evaluated total SIV-specific responses by the rebounding CD8+ population (18, 27, 28, 36, 50, 51). We reasoned that analyzing the antigen specificity of returning CD8+ T cells could identify T-cell correlates of control in these animals. One of the hallmarks of ECs is the ability of their antigen-specific CD8+ T cells to proliferate (38), so we hypothesized that populations of SIV-specific CD8+ T cells would undergo a generalized expansion upon return to the periphery. At 21 days after depletion, when CD8+ cells first returned, we used Mamu-A*01, -A*02, and -B*17 tetramers loaded with commonly recognized epitopes to measure the antigen-specific CD8+-T-cell response.
To our surprise, the reassertion of control of viral replication was associated not with generalized proliferation of all previously detected SIV-specific CD8+ T-cell populations but rather with differential increases in the frequencies of particular CD8+ T-cell specificities. The populations of only a few specificities expanded in most animals, and previously subdominant populations tended to show a much greater degree of increase than immunodominant ones. Remarkably, we could detect only a single CD8+ T-cell response in the Mamu-B*17-positive macaque AJ11. This response was directed against the Nef MW9 epitope (5.67% of CD3+, CD8+ T cells) (Fig. 2a), recognized by a subdominant population of CD8+ T cells during acute infection (37). The frequency of Nef MW9-specific CD8+ T cells increased 22-fold after depletion.
![]() View larger version (26K): [in a new window] |
FIG. 2. Dynamics of CD8+ T-cell responses enumerated by tetramer staining. PBMC were stained with tetramers loaded with the indicated peptides before (white bars) and 21 days after (gray bars) depletion, at which time CD8 cell counts had returned nearly to normal. Numbers above bars indicate changes (n-fold) in the magnitude of responses between baseline and day 21. Tetramers were available for the MHC class I molecules Mamu-A*01, Mamu-A*02, and Mamu-B*17. Animal 00078 did not express any of the corresponding alleles, so data for this animal are not shown.
|
10-fold following depletion in this animal. Similarly, a strong immunodominant response to the Mamu-B*17-restricted epitope Nef IW9 occurred in animal 95071 before depletion (9.1% of CD3+, CD8+ T cells) (Fig. 2c). As CD8+ T cells rebounded following depletion, the Nef IW9-specific population was still dominant (10.6% of CD3+, CD8+ T cells; 1.2-fold increase), but there was an eightfold expansion of cells specific for a different Mamu-B*17-restricted epitope, Vif HW8 (1.0 to 8.2%) (Fig. 2c). The Mamu-B*17-positive animal 98016 made strong responses against the Vif HW8 and Nef IW9 epitopes after depletion (41- and 2,436-fold increases, respectively), although these responses were barely detectable prior to depletion (Fig. 2d). Animal 01064 expressed Mamu-A*02, and not Mamu-B*17, but its SIV-specific CD8+ T-cell response showed a similar pattern. Before depletion, it made a dominant response to the Mamu-A*02-restricted Gag GY9 epitope. This response expanded fourfold after depletion, while the previously subdominant population of CD8+ T cells recognizing the Nef YY9 epitope expanded 15-fold, becoming the newly dominant SIV-specific population (Fig. 2e).
Control of viral replication was correlated with previously subdominant populations of epitope-specific CD8+ T cells and broad CD4+ T-cell responses against Gag.
Our analysis at 21 days after depletion was limited to tetramer staining due to constraints on blood draw volume with rhesus macaques. Therefore, we obtained maximal blood draw volumes the following week to measure the cellular immune responses to all nine viral gene products. At 28 days postdepletion, we performed ICS for IFN-
and IL-2 by using previously identified minimal optimal epitope peptides (3, 31, 40, 54) as well as pools of 15-mer peptides spanning the entire SIV proteome. This approach allowed us to detect responses to both known and previously undescribed epitopes recognized by CD4+ and CD8+ T cells.
For animals AJ11 and 95096, ICS revealed that only a few SIV-specific CD8+ T-cell responses were present (Fig. 3a and b). In accordance with the tetramer result, ICS assays detected only a single CD8+ T-cell response in AJ11, directed against the Nef MW9 epitope recognized by a previously subdominant T-cell population. ICS also detected three CD8+ T-cell responses to previously described epitopes in 95096, in accordance with the results of tetramer staining. However, the Mamu-B*17-restriced epitope Nef IW9 failed to elicit IFN-
or IL-2 responses in 95096, although tetramer stains 1 week earlier detected a population of T cells that bound this peptide (0.99% of CD3+, CD8+ T cells) (Fig. 2b). 95096 also made CD8+ T-cell responses to three peptide pools, one each from Gag, Vif, and Nef, which could not be accounted for by the responses to known minimal optimal epitopes. These responses are therefore likely directed against previously undescribed epitopes (Fig. 3b). ICS additionally revealed a robust antiviral CD4+ cell response in both 95096 and AJ11, directed primarily against Gag (Fig. 4a and b).
![]() View larger version (22K): [in a new window] |
FIG. 3. Expansion of a subset of CD8+ T-cell responses after the depletion of Mamu-B*17-positive and -negative ECs of CD8+ cells. PBMC were stimulated with synthetic peptides representing minimal optimal Mamu-A*01-, Mamu-A*02-, or Mamu-B*17-restricted epitopes or pools of 15-mer peptides representing the entire SIVmac239 proteome and stained for intracellular accumulation of IL-2 and IFN- 1 month prior to (white bars) and 28 days after (gray bars) CD8 cell depletion. Responses to individual minimal optimal peptides are indicated according to the presenting MHC class I molecule. In each case, responses stimulated by the minimal optimal peptide were also detected using the corresponding 15-mer pool (data not shown); responses to these pools were subtracted from the data presented here. Responses to 15-mer pools that did not contain previously identified minimal optimal epitope sequences are summed for each protein and indicated as "unknown." Due to the limited availability of PBMC prior to depletion, cells from animals 95096 and AJ11 were stimulated with epitope peptides only.
|
![]() View larger version (28K): [in a new window] |
FIG. 4. Expansion of SIV-specific CD4+ T-cell responses following the depletion of ECs of CD8+ cells. PBMC were stimulated with pools of 10 to 50 overlapping 15-mer peptides spanning the entire SIVmac239 proteome 1 month before (white bars) and 28 days after (gray bars) the depletion of ECs of peripheral CD8+ cells. Each column represents the frequency of CD4+, IFN- + lymphocytes responding to one peptide pool. CD4+ T-cell response data for animals 95096 and AJ11 prior to depletion are not available due to the limited availability of PBMC.
|
+ T cells). Other responses were weak in comparison (Fig. 3c). Although this animal's response to Nef IW9 (10.6% of CD3+, CD8+ T cells) (Fig. 2c) appeared to be dominant in tetramer assays, CD8+, IFN-
+ T-cell responses to Nef IW9 were far more modest (0.55% of lymphocytes expressing CD8 and IFN-
), similar to our observations with animal 95096. 95071 also made extremely strong CD4+ T-cell responses to SIV, even during the CD8+ cell depletion phase when viral replication was high (data not shown). These responses expanded upon recovery of CD8+ cells and were dominated by responses to Rev pool A, with a substantial response to Gag pool B (Fig. 4c).
Animal 98016 also had broad CD8+ and CD4+ T-cell responses following depletion (19 epitopes were recognized by CD8+ and CD4+ T-cell subsets), although they were smaller in magnitude than those in 95071. There was a strong response to Vif HW8 (2.3%) (Fig. 3d), although Nef IW9-specific CD8+ T cells had shown higher frequencies in tetramer assays. The strongest CD8+ T-cell response in this animal was directed against Nef pool D (3.3% of lymphocytes expressing CD8 and IFN-
), while the response by the highest-frequency CD4+ T cells recognized Gag pool I (0.18% of lymphocytes expressing CD4 and IFN-
) (Fig. 4d).
Animal 00078, which did not express Mamu-A*01, -A*02, or -B*17, made at least 13 distinct CD8+ T-cell responses, dominated by IFN-
+ cells responding to pools of peptides from Vif (Vif pool E, 0.72%) (Fig. 3e) and Nef (Nef pool D, 0.26%). 00078 made at least 20 CD4+ T-cell responses, with high-frequency responses to pools of peptides from Gag (Gag pool B, 0.22%; Gag pool G, 0.2%) (Fig. 4e) and Nef (Nef pool F, 0.14%). The majority of these responding CD4+ cells were positive for both IFN-
and IL-2 (data not shown). Similarly, animal 01064, which expresses Mamu-A*02 but not Mamu-B*17, made at least 18 distinct CD8+-T-cell responses, the largest targeting the Nef YY9 epitope (1% of lymphocytes expressing CD8 and IFN-
) (Fig. 3f). Other strong responses were directed against Rev peptide pools B and C (total of 0.8% of lymphocytes expressing CD8 and IFN-
) and Vif pool E (0.34% of lymphocytes expressing CD8 and IFN-
). Interestingly, although we detected a strong response to the previously dominant Gag GY9 epitope in tetramer assays, this epitope never stimulated an IFN-
response in 01064 following depletion (Fig. 3f and data not shown). CD4+ cells from 01064 recognized nine different peptide pools after depletion, with the strongest response directed against Gag pool B (0.13% of lymphocytes expressing CD4 and IFN-
) (Fig. 4f).
EC macaques showed different patterns of epitope recognition after CD8 depletion treatment. The epitopic breadths of responses in animals 95071 and 98016 increased over those seen at baseline, while the numbers of antigens targeted by other animals were similar pre- and postdepletion. Despite these differences, there were nonetheless clear similarities in the total cellular immune responses as viremia declined. We observed differential increases in the frequencies of SIV-specific CD8+ T-cell populations, with previously subdominant responses increasing to a much greater degree than dominant ones. Furthermore, expanding CD8+ T-cell populations in each EC preferentially targeted epitopes or pools from Nef and Vif. ECs also each had strong and broad CD4+ T-cell responses. Strikingly, while the dominant CD8+ T-cell responses were different in each EC, three of six ECs shared a dominant CD4+ T-cell response to Gag pool B, while the other three ECs made strong but subdominant responses to Gag pool B peptides (Fig. 4).
Both NK and CD8+ T cells from ECs suppressed SIV replication in a novel ex vivo assay. Since the cM-T807 antibody depleted animals of peripheral NK cells as well as CD8+ T cells, it is possible that the rebound in viremia we observed was due simply to the removal of NK cells, with CD8+ T-cell populations expanding in response to the antigens but not actively controlling viral replication. Following the animals' return to set-point viremia, we sought to determine the relative contributions of NK and CD8+ T cells in suppressing viral replication by using a novel ex vivo assay. We cocultured animals' PBMC with activated, autologous target cells superinfected with clonal SIVmac239. These unfractionated PBMC suppressed viral replication (Fig. 5a), so we reasoned that the depletion of effector PBMC of NK and/or CD8+ T cells should inhibit suppression, analogous to in vivo CD8 depletion. Indeed, as we saw in vivo, the removal of all CD8+ cells from the effector PBMC resulted in levels of viral replication equal to that seen when no effectors were added to the cultures (Fig. 5d). Removing either NK or CD8+ T-cell populations from the cultures also led to increased levels of viral replication. The relative contribution of each subset was variable among the animals tested (Fig. 5b and c), though consistent for each animal across multiple experiments at different times postdepletion. These data indicate that both CD8+ T cells and NK cells from these animals have antiviral functions and likely contribute to maintaining elite control of SIV.
![]() View larger version (29K): [in a new window] |
FIG. 5. Ex vivo assay for functions of CD8+ lymphocyte subsets. Representative results from the coculture of infected, autologous target cells with the indicated effector cell populations are shown. Total PBMC, PBMC depleted of NK cells, PBMC depleted of CD8+ T cells (CD8T), or PBMC depleted of all CD8+ cells were used as effectors at a ratio of approximately 3:1 in a 7-day assay. Cells were stained for p27 Gag expression; the percentage of CD8 cells positive for Gag expression is indicated in each panel. Quantitation of SIV vRNA in supernatants confirmed these results (data not shown).
|
in response to this peptide was far lower (Fig. 3). It is therefore possible that the variant sequence is able to drive the expansion of populations of CD8+ T cells that bind the tetramer but do not produce cytokines in response to the IW9 peptide. Animal 95096's virus harbored a glutamine-for-histidine substitution at position 2 in the Env FW9 epitope, a substitution not seen in Mamu-B*17-positive progressors (Fig. 6 and data not shown). Nonetheless, the population of Env FW9-specific CD8+ T cells expanded in 95096 after depletion, though the percentage of cells secreting IFN-
(0.2%) (Fig. 3b) was lower than that detected by the FW9 tetramer (0.94%) (Fig. 2b). By contrast, virus from animal 95071 showed mixed-base heterogeneity in this epitope at position 8, a mutation also observed in virus from Mamu-B*17-positive progressors (Fig. 6 and data not shown). This may be a more effective escape mutation, since both tetramer and ICS assays detected Env FW9-specific CD8+ T cells only at very low frequencies in 95071.
![]() View larger version (32K): [in a new window] |
FIG. 6. Sequence variation in epitopes restricted by Mamu-A*01, Mamu-A*02, and Mamu-B*17 in recrudescent virus in ECs. For each animal, locations of epitopes recognized upon the return of CD8+ T cells after depletion are mapped onto the viral proteome. Solid lines represent responses to individual minimal optimal epitopes detected by both tetramers and IFN- secretion. Vertical boxes, e.g., that corresponding to animal 95071 for Nef YY9-159, indicate responses detected by tetramers but not by IFN- secretion. Variation in known epitopes is represented by circles.
|
responses to the Gag GY9 peptide (Fig. 3d and e). Virus from animals 95096 and AJ11 had variations in the Mamu-B*17-restricted Env LF11, and we detected no responses to this epitope (Fig. 6). Animal 95096's virus had variations in the Mamu-B*17-restricted Pol MF8 and Mamu-A*01-restricted Tat SL8, to which no responses were detected by tetramers or ICS (Fig. 6). However, escape mutations could not always explain the lack of certain responses. No variation was detected in Vif HW8 in the four Mamu-B*17-positive ECs. Animals 95071 and 98016 made extremely strong responses to this epitope after depletion, but no response was detected in 95096 or AJ11. |
|
|---|
Although each animal in this study had a unique combination of cellular immune responses and viral sequences, we can generalize at least three important observations. First, in all the ECs, elite control was maintained even as the epitopic breadths of the cellular immune responses narrowed in chronic infection. This phenomenon was most striking for animal AJ11, whose CD8+ T cells recognized at least 27 different epitopes during the acute phase (Table 2) (37) but who apparently maintained control with only one CD8+ T-cell response by the time of CD8+ cell depletion 3 years later. This single CD8+ T-cell response was directed against epitope Nef MW9, recognized by a subdominant CD8+ T-cell population during acute infection. Second, the Mamu-B*17-positive ECs appeared to control SIV replication despite variations in the epitope Nef IW9, recognized by immunodominant CD8+ T-cell populations. This control is analogous to the control of HIV replication in HLA-B57-positive subjects, which appears to endure even after the escape of the epitope corresponding to the initially immunodominant response (26). Finally, previously subdominant CD8+ T-cell populations expanded most dramatically in each animal as control was reasserted. Although this expansion alone does not directly implicate subdominant CD8+ T-cell populations in the control of infection, it is striking that these populations proliferated to a greater degree than dominant ones, and it is tempting to speculate that subdominant populations with proliferative capacity may play an important role in maintaining control of viremia in the chronic phase. In support of this idea, a recent study of HIV-infected subjects indicates that control of viremia associated with HLA-B*1503 may depend on subdominant responses, not immunodominant ones (11).
CD8+ T-cell efficacy is likely determined by a variety of factors that combine to make each epitope-specific response unique. One challenge to HIV vaccine design is to identify these factors and elucidate the ways in which they can combine to determine the potency of immune responses. T-cell receptor avidities and clonotypes, cytokine secretion profiles, the kinetics of expression of target proteins during the virus life cycle, and evolutionary constraints on viral epitope regions all likely influence this potency. Our data indicate that particular CD8+ T-cell specificities, together with robust CD4+ T-cell responses, can control a highly pathogenic AIDS virus. Perhaps most encouragingly, they suggest that this control can be achieved with a small number of particularly effective recall responses. Our results also suggest that effective subdominant CD8+ T-cell responses may be important for maintaining control of chronic immunodeficiency virus replication. Current vaccine modalities are designed to deliver large antigens, which may induce potent immunodominant responses at the expense of subdominant ones. Our data suggest the possibility that AIDS vaccines which allow for the development of broad subdominant responses in addition to immunodominant ones may have broader and more durable efficacy than present approaches.
This research was supported by National Institutes of Health grants R01 AI052056 and R01 AI049120 to D.I.W. and by National Center of Research Resources grant P51 RR 000167 to WNPRC. Animals in this study were housed at a facility constructed with support from Research Facilities Improvement Program grant numbers RR15459-01 and RR020141-01.
We have no competing financial interests.
Published ahead of print on 24 January 2007. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»