|Home | About | Journals | Submit | Contact Us | Français|
The expression of particular major histocompatibility complex (MHC) class I alleles can influence the rate of disease progression following lentiviral infections. This effect is a presumed consequence of potent cytotoxic T-lymphocyte (CTL) responses that are restricted by these MHC class I molecules. The present studies have examined the impact of the MHC class I allele Mamu-A*01 on simian/human immunodeficiency virus 89.6P (SHIV-89.6P) infection in unvaccinated and vaccinated rhesus monkeys by exploring the contribution of dominant-epitope specific CTL in this setting. Expression of Mamu-A*01 in immunologically naive monkeys was not associated with improved control of viral replication, CD4+ T-lymphocyte loss, or survival. In contrast, Mamu-A*01+ monkeys that had received heterologous prime/boost immunizations prior to challenge maintained higher CD4+ T-lymphocyte levels and better control of SHIV-89.6P replication than Mamu-A*01− monkeys. This protection was associated with the evolution of high-frequency anamnestic CTL responses specific for a dominant Mamu-A*01-restricted Gag epitope following infection. These data indicate that specific MHC class I alleles can confer protection in the setting of a pathogenic SHIV infection by their ability to elicit memory CTL following vaccination.
The expression of certain major histocompatibility complex (MHC) class I alleles influences disease progression following human immunodeficiency virus type 1 (HIV-1) and simian immunodeficiency virus (SIV) infection. Expression of HLA-B*35 and HLA-Cw*04 has been linked to rapid disease progression (4), whereas expression of HLA-B*27 and HLA-B*57 is associated with slow disease progression in HIV-1-infected humans (8, 14). Similarly, expression of the MHC class I alleles Mamu-A*01 and Mamu-B*17 in Indian origin rhesus monkeys has been associated with slow disease progression following infection with the pathogenic virus SIVmac251, SIVsmE660, or SIVmac239 (15, 16, 20, 21). It is presumed that the improved survival of these virus-infected humans and monkeys reflects the contribution of MHC class I-restricted epitope-specific cytotoxic T-lymphocyte (CTL) responses. Immunodominant Gag-specific CTL responses restricted by HLA-B*27 and HLA-B*57 have been demonstrated for individuals chronically infected with HIV-1 who have slowly progressive disease (6, 7, 18). Further, Mamu-A*01- and Mamu-B*17-restricted dominant-epitope-specific CTL responses in rhesus monkeys have been associated with slow disease progression following SIVmac239 infection (20).
Virus-specific CTLs also play a central role in containing viral replication in vaccinated monkeys that are subsequently challenged with primate immunodeficiency viruses. Vaccine-elicited memory CTLs have been shown to expand in monkeys following viral infection and to contribute to protection against the progression of clinical disease (3, 10, 21, 23, 27). This vaccine-generated immunity controls the extent of viral replication and diminishes CD4+ T-lymphocyte loss in infected animals. Whether MHC class I haplotype can influence this vaccine-mediated protection has not been systematically explored.
Chimeric simian/human immunodeficiency viruses (SHIVs) are widely used as challenge viruses in nonhuman primate studies of HIV-1 vaccine strategies (1-3, 5, 10, 13, 23, 27). Infection of rhesus monkeys with the CXCR4-tropic, highly pathogenic simian/human immunodeficiency virus 89.6P (SHIV-89.6P) virus is associated with acute loss of CD4+ T lymphocytes, high levels of persistent viral replication, and rapid progression to AIDS and death (22). Conflicting data have been reported concerning the impact of particular MHC class I alleles on disease progression in SHIV-89.6P-infected rhesus monkeys. Some investigators have reported that Mamu-A*01 expression is associated with a more benign clinical course (29), while others have observed no impact on disease evolution in previously naive animals (21).
In the present study, we analyzed virus replication, CD4+ T-lymphocyte loss, and Gag-specific cellular immune responses in unvaccinated and vaccinated Mamu-A*01+ and Mamu-A*01− rhesus monkeys following SHIV-89.6P infection. Expression of Mamu-A*01 did not confer a clinical benefit on SHIV-89.6P-induced disease in previously naive monkeys. In contrast, expression of this MHC class I allele was associated with improved control of SHIV-89.6P infection in monkeys immunized prior to infection. This viral control correlated with the development of a high-frequency dominant-epitope Gag-specific CTL response following viral challenge.
A PCR-based assay was used to select adult rhesus monkeys (Macaca mulatta) that expressed the Mamu-A*01 MHC class I allele. Briefly, DNA was extracted from peripheral blood mononuclear cells (PBMC) by using a QIAmp blood kit (QIAGEN, Chatsworth, Calif.). PCR amplification was then performed using Mamu-A*01-specific primers (forward, 5′ GAC AGC GAC GCC GCG AGC CAA 3′; reverse, 5′ CGC TGC AGC GTC TCC TTC CCC 3′). Two additional primers specific for a conserved MHC class II sequence were also included as internal positive controls (forward, 5′ GCC TCG AGT GTC CCC CCA GCA CGT TTC 3′; reverse, 5′ GCA AGC TTT CAC CTC GCC GCT G 3′). Electrophoresis on a 2% agarose gel yielded a 685-bp band in Mamu-A*01-positive samples and a 260-bp control band in all samples. Verification of positive samples was achieved by complete DNA sequence analysis and comparison with the published Mamu-A*01 sequence. Monkeys expressing the Mamu-A*02 MHC class I allele were identified using Mamu-A*02 specific primers (forward, 5′ GTG GGT GGA GCA GGA GGG TCC A 3′; reverse, 5′ CAG CAC CTC AGG GTG GCC TCT 3′) as described above. Monkeys were housed in facilities accredited by the Association for the Assessment and Accreditation of Laboratory Animal Care in accordance with the guidelines of the Institutional Animal Care and Use Committee for Harvard Medical School and the Guide for the Care and Use of Laboratory Animals.
Two cohorts each of unvaccinated and vaccinated monkeys were analyzed in this study. The first evaluated group was comprised of 12 Mamu-A*01+ and 16 Mamu-A*01− rhesus monkeys that received DNA prime/recombinant poxvirus boost immunizations and 9 Mamu-A*01+ and 19 Mamu-A*01− sham-immunized control animals (25). Plasmid DNA vaccines were administered by intramuscular injection at 0, 4, and 8 weeks by the use of a needle-free Biojector system and a no. 3 syringe. For each DNA immunization, monkeys received 5 mg of SIVmac239 gag plasmid and 5 mg of HIV-1 89.6P env plasmid (10 mg total). These monkeys also received 5 mg of IL-2/Ig plasmid on day 2 after DNA vaccination. On week 42 of the study, the treated monkeys were divided into four groups (each having three Mamu-A*01+ and four Mamu-A*01− animals) and received boost immunizations with plasmid DNA or one of the three recombinant poxvirus vectors expressing immunogens matching those used for the DNA prime. A total of 2 × 109 PFU of recombinant modified vaccinia virus Ankara, recombinant fowlpox virus, or recombinant vaccinia virus were delivered by both intradermal and intramuscular injections (109 PFU expressing SIVmac239 Gag and 109 PFU expressing HIV-1 89.6P Env). Control monkeys (9 Mamu-A*01+ and 19 Mamu-A*01−) were immunized with 10 mg of sham DNA at weeks 0, 4, and 8 and boosted with sham DNA or 2 × 109 PFU of empty fowlpox virus, modified vaccinia virus Ankara, or vaccinia virus vectors at week 42. At week 60, all monkeys received an intravenous challenge with 50 50% monkey infective doses (MID50) of SHIV-89.6P.
The second group was comprised of 8 Mamu-A*01+ and 16 Mamu-A*01− (of which 7 were Mamu-A*02+) rhesus monkeys that received DNA prime/recombinant replication-defective adenovirus (rAd) boost immunizations and 2 Mamu-A*01+ and 4 Mamu-A*01− sham-immunized control animals (26). Plasmid DNA vaccines were administered by intramuscular injection using a needle-free Biojector system and a no. 3 syringe (Bioject, Portland, Oreg.) at 0, 4, and 8 weeks. For each DNA immunization, monkeys received 4.5 mg of SIVmac239 gag/pol/nef plasmid and 4.5 mg of HIV-1 env plasmid(s) (9 mg total). For the HIV-1 Env component of the vaccine, four groups of monkeys (each having two Mamu-A*01+ and four Mamu-A*01− animals) received either 4.5 mg of HXBc2/BaL clade B env, 1.5 mg of HXBc2/BaL clade B env plus 3.0 mg of sham plasmid, 4.5 mg of clade C env, or a mixture of 1.5 mg each of clade A, B, and C env plasmids. Plasmid DNAs were divided into two aliquots of 0.5 ml each and delivered into each quadriceps muscle. At week 26, monkeys received a boost immunization with rAd vectors expressing immunogens matching those used for the DNA prime. A total of 2 × 1012 rAd particles (1 × 1012 rAd-SIVmac239 gag/pol and 1 × 1012 rAd-HIV-1 env[s]) were delivered by intramuscular injection as described above. Control monkeys (two Mamu-A*01+ and four Mamu-A*01−) were immunized with sham DNA at weeks 0, 4, and 8 and boosted with sham rAd vectors at week 26. At week 42, all monkeys received an intravenous challenge with 50 MID50 of SHIV-89.6P. The same stock of SHIV-89.6P challenge virus was used for both studies. To investigate the effect of Mamu-A*01 expression on immunologically naive rhesus monkeys infected with SHIV-89.6P, unvaccinated Mamu-A*01+ and Mamu-A*01− monkeys from both studies were grouped together for analysis. To investigate the effect of Mamu-A*01 expression on vaccinated monkeys infected with SHIV-89.6P, the groups of Mamu-A*01+ and Mamu-A*01− monkeys receiving either DNA prime/rAd boost or DNA prime/poxvirus boost immunizations were analyzed separately.
Plasma viral RNA levels were measured by an ultrasensitive branched DNA amplification assay (Bayer Diagnostics, Berkeley, Calif.) with a lower detection limit of 125 copies per ml. The percentage of CD4+ T lymphocytes in the peripheral blood of infected monkeys was determined by monoclonal antibody staining and flow cytometric analysis. Briefly, freshly isolated peripheral blood lymphocytes (PBL) were stained with anti-CD3 allophycocyanin (FN18), anti-CD4 phycoerythrin (19Thy5D7), and anti-CD8 fluorescein isothiocyanate (SK1; BD Biosciences, Mountain View, Calif.). Samples were acquired using a FACSCalibur flow cytometer, and data were analyzed using CellQuest software (BD Biosciences). For the control animals, RNA levels and CD4+ T-lymphocyte percentages were compared for each available time point up to day 168. The average set point viral RNA levels and CD4+ percentages between days 42 and 168 were also compared. The reason for choosing day 168 as the last observation boundary of these comparisons and averages was that the majority of unvaccinated control animals in both groups remained alive through this time point. The peak viral loads (which occurred between days 14 and 16 following challenge) were also compared for the two groups of monkeys. These same comparisons were done for the vaccinated Mamu-A*01+ and Mamu-A*01− monkeys, with the exception that measurements were compared at all time points up to day 300 following challenge and that the averages were calculated for the period of day 42 to day 300. The reason for choosing day 300 for the vaccinated groups of monkeys was that all animals were monitored for at least this long for both viral load and CD4+ percentage.
Multiscreen 96-well plates (Millipore, Bedford, Mass.) were coated overnight (100 μl/well) at 4°C with mouse anti-human gamma interferon (IFN-γ) monoclonal antibody (B27; BD PharMingen, San Diego, Calif.) at 10 μg/ml in endotoxin-free Dulbecco's phosphate-buffered saline (PBS) (D-PBS; Life Technologies, Gaithersburg, Md.). Plates were washed three times with D-PBS containing 0.25% Tween 20, blocked for 2 h at 37°C with 100 μl of D-PBS containing 5% fetal bovine serum/well, and rinsed with RPMI medium containing 10% fetal bovine serum to remove the Tween 20. PBL were plated in triplicate at 2 × 105/well in a 100-μl final volume with medium alone, Gag peptide pool, 89.6P Env peptide pool, or individual Gag-derived epitope peptides. The peptide pools covered the entire SIVmac239 Gag or HIV-1 89.6P Env proteins and were comprised of 15 amino acid peptides overlapping by 11 amino acids or of 20 amino acid peptides overlapping by 10 amino acids, respectively. Each peptide in the pool was present at a concentration of 1 μg/ml. The Gag-derived epitope peptides p11C (CTPYDINQM) and p17G (GSENLKSLY) were used at 1 μg/ml for measuring antigen-specific CTL responses in Mamu-A*01+ and Mamu-A*02+ rhesus monkeys, respectively. Following an 18-h incubation at 37°C, the plates were washed nine times with D-PBS containing 0.25% Tween 20 and once with distilled water. The plates were then incubated with 2 μg of biotinylated rabbit anti-human IFN-γ antibody (Biosource, Camarillo, Calif.)/ml for 2 h at room temperature, washed six times with Coulter Wash (Beckman Coulter, Miami, Fla.), and incubated for 2 h with a 1:500 dilution of streptavidin-AP (Southern Biotechnology, Birmingham, Ala.). Following five washes with Coulter Wash and one with D-PBS, the plates were developed with nitroblue tetrazolium-5-bromo-4-chloro-3 indolyl phosphate chromogen (Pierce, Rockford, Ill.), the enzymatic reaction was stopped by washing with tap water, and the plates were air dried and read using an enzyme-linked immunospot (ELISPOT) reader (Hitech Instruments, Edgement, Penn.). The mean number of spots from triplicate wells was calculated for each animal and adjusted to represent the mean number of spots per 106 PBMC. Negative-control wells with no peptide antigen consistently had less than 50 spots per 106 PBMC.
Tetrameric complexes of Mamu-A*01/p11C and Mamu-A*02/p17G were prepared as previously described (9, 17). PBL were isolated from whole-blood specimens by Ficoll-Paque (Amersham-Pharmacia Biotech, Uppsala, Sweden) density gradient centrifugation. Phycoerythrin-coupled tetrameric complexes were used in combination with anti-CD3-allophycocyanin and anti-CD8-fluorescein isothiocyanate to stain 2 × 105 freshly isolated PBL. Samples were acquired using a FACSCalibur flow cytometer, and data were analyzed using CellQuest software.
The exact Wilcoxon rank sum test was used to compare CD4+ T lymphocytes, plasma viral RNA levels, and Gag-specific ELISPOT and tetramer responses between Mamu-A*01+ and Mamu-A*01− rhesus monkeys in the unvaccinated and vaccinated groups. Viral loads below 125 (the lower detection limit of the assay) were treated as though they were 125. All tests were two sided. The power statements for these tests were based on the observed standard deviations and the fact that the asymptotic efficiency of the t test relative to that of the Wilcoxon rank sum test is no greater than 1.16 for any continuous distribution. There were no corrections for multiple comparisons. The comparison of times to progression for control monkeys used the log rank statistic; the power statement for this test is an approximation based on the assumption that both distributions of times are exponential with a guarantee time of 100 days.
We first sought to determine whether expression of the MHC class I allele Mamu-A*01 in immunologically naive rhesus monkeys influenced viral replication and the rate of disease progression following infection with SHIV-89.6P. A total of 11 Mamu-A*01+ and 23 Mamu-A*01− rhesus monkeys that served as control animals in two vaccine studies were evaluated. These unvaccinated monkeys were challenged intravenously with 50 MID50 of SHIV-89.6P. Plasma viral RNA levels peaked between days 14 and 16 after challenge for both the Mamu-A*01+ and Mamu-A*01− monkeys, with similar mean log10 values of 8.1 copies of viral RNA per ml measured in each group of animals (Fig. (Fig.1,1, upper panels, left versus right). Long-term containment of viral replication was evaluated in these monkeys by measuring plasma viral RNA levels between days 42 and 168 following challenge. A set point viral RNA level for each monkey was calculated as the mean of these values; data for each animal included at least seven time points within this time interval. No significant difference in viral replication during this chronic phase of infection between these groups of animals was detected, with mean log10 set point plasma viral RNA levels of 5.7 and 6.1 copies per ml measured for Mamu-A*01+ and Mamu-A*01− monkeys, respectively (P = 0.56).
Because SHIV-89.6P infection of rhesus monkeys causes a dramatic and persistent decline in their peripheral blood CD4+ T lymphocyte levels, we assessed whether a difference in CD4+ T-lymphocyte depletion could be detected between the Mamu-A*01+ and Mamu-A*01− monkeys. Most monkeys in both groups of animals exhibited a profound loss of peripheral blood CD4+ T lymphocytes within the first 21 days of infection (Fig. (Fig.1,1, lower panels, left versus right). No significant difference in the percentages of peripheral blood CD4+ T cells was detected between the Mamu-A*01+ and Mamu-A*01− monkeys between days 42 and 168 postchallenge (P = 0.34). Together with the viral load analysis, these results suggest that expression of Mamu-A*01 in immunologically naive rhesus monkeys does not affect acute or chronic SHIV-89.6P replication or the associated loss of CD4+ T lymphocytes.
Expression of Mamu-A*01 also did not influence the rate of disease progression in these animals following challenge. Mamu-A*01+ and Mamu-A*01− rhesus monkeys had similar survival curves following SHIV-89.6P infection (Fig. (Fig.2).2). The median times to death, 196 days for Mamu-A*01+ monkeys and 273 days for Mamu-A*01− monkeys, were not significantly different (P = 0.43, two-sided log rank test). By day 300 postinfection, only three Mamu-A*01+ monkeys (27%) and eight Mamu-A*01− monkeys (35%) remained alive. Thus, expression of Mamu-A*01 in immunologically naive rhesus monkeys does not appear to confer a survival advantage in the setting of intravenous SHIV-89.6P infection.
Immunodominant Mamu-A*01-restricted CTL responses specific for the SIV Gag-derived p11C epitope are associated with decreased viral replication following SIV infection in rhesus monkeys (20). We therefore examined whether the failure to control viral replication in SHIV-89.6P-infected naive monkeys reflected an inability to generate the dominant Gag-specific CTL response in these Mamu-A*01+ animals. The Gag-specific cellular immune responses in the SHIV-89.6P-infected Mamu-A*01+ and Mamu-A*01− monkeys were assessed by pooled peptide IFN-γ ELISPOT assays using freshly isolated PBL. The majority of the Mamu-A*01+ and Mamu-A*01− monkeys demonstrated low-frequency or undetectable Gag-specific cellular responses on days 14 and 28 postinfection (Fig. (Fig.3).3). These data indicate that both naive Mamu-A*01+ and Mamu-A*01− monkeys infected with SHIV-89.6P fail to develop robust Gag-specific cellular immune responses.
We next examined whether expression of Mamu-A*01 in previously vaccinated rhesus monkeys conferred clinical protection following SHIV-89.6P challenge. Data on Mamu-A*01+ and Mamu-A*01− monkeys from two independent vaccine studies were analyzed. In the first study, 12 Mamu-A*01+ and 16 Mamu-A*01− monkeys received DNA prime and DNA or recombinant poxvirus boost immunizations with vaccine vectors encoding SIVmac239 Gag and HIV-1 89.6P Env immunogens (25). At 18 weeks following the last immunization, all monkeys received an intravenous inoculation of 50 MID50 of SHIV-89.6P, the same challenge dose used for unvaccinated animals. Viral replication did not differ significantly between the vaccinated Mamu-A*01+ and Mamu-A*01− monkeys during acute infection (Fig. (Fig.4A).4A). Peak plasma viral loads were detected on day 14 postchallenge in both groups of animals, with mean log10 values of 6.8 and 7.2 copies of viral RNA per ml measured in the Mamu-A*01+ and Mamu-A*01− monkeys, respectively (P = 0.73). Long-term containment of viral replication in these groups of monkeys was evaluated by measuring plasma viral RNA levels between days 42 and 300 following challenge. Mamu-A*01+ monkeys exhibited a trend toward lower set point plasma viral RNA levels during this time period compared with Mamu-A*01− monkeys (P = 0.12). This difference appeared to become more marked at the later time points. All DNA prime- and DNA or recombinant poxvirus boost-vaccinated monkeys demonstrated blunting of peripheral blood CD4+ T-lymphocyte loss compared to unvaccinated control monkeys (Fig. (Fig.4B4B versus Fig. Fig.11 [lower panels]). Levels of CD4+ T lymphocytes during chronic infection were higher in Mamu-A*01+ than in Mamu-A*01− monkeys, although this difference did not approach statistical significance (P = 0.50).
In the second vaccine study, 8 Mamu-A*01+ and 16 Mamu-A*01− monkeys received DNA prime/recombinant adenovirus boost immunizations with vectors encoding SIVmac239 Gag-Pol-Nef polyprotein and HIV-1 Env immunogens (26). At 16 weeks after the final immunization, all monkeys were challenged intravenously with 50 MID50 of SHIV-89.6P. No significant differences in peak plasma viral RNA levels were detected between Mamu-A*01+ and Mamu-A*01− monkeys (Fig. (Fig.5A)5A) (P = 0.54). Set point plasma viral RNA levels between days 42 and 300 postchallenge, however, were significantly lower in the Mamu-A*01+ monkeys than in the Mamu-A*01− monkeys (mean values of 3.3 and 4.8 log10 copies per ml, respectively; P = 0.01). Furthermore, vaccinated Mamu-A*01+ monkeys demonstrated a trend toward greater preservation of peripheral blood CD4+ T lymphocytes during this time period than did the vaccinated Mamu-A*01− monkeys (Fig. (Fig.5B)5B) (P = 0.08). The data from these two studies therefore suggest that in the setting of prior vaccination, Mamu-A*01+ monkeys demonstrate significantly better control of chronic viral replication and a trend toward better preservation of CD4+ T lymphocytes than do Mamu-A*01− monkeys. Nevertheless, all experimentally vaccinated monkeys from these two vaccine studies, Mamu-A*01+ and Mamu-A*01−, remained alive through the period of evaluation, 300 days postchallenge. Thus, differences in the rates of disease progression in vaccinated, SHIV-89.6P-infected Mamu-A*01+ and Mamu-A*01− monkeys could not be determined.
Since cellular immune responses contribute to primate lentivirus containment, we sought to determine whether differences in virus-specific CD8+ T-lymphocyte responses might explain the improved control of SHIV-89.6P replication in the vaccinated Mamu-A*01+ monkeys. We first examined whether the Mamu-A*01+ monkeys demonstrated higher-frequency Gag-specific CTL responses than the Mamu-A*01− monkeys following immunization and/or challenge. Freshly isolated PBL from monkeys receiving DNA prime and DNA or recombinant poxvirus boost or DNA prime/rAd boost vaccinations were assessed for Gag-specific cellular immune responses by pooled peptide IFN-γ ELISPOT assay. PBL of both the Mamu-A*01+ and Mamu-A*01− monkeys immunized with plasmid DNA and recombinant poxvirus vectors demonstrated robust Gag-specific cellular immune responses (Fig. (Fig.6A).6A). In fact, no significant differences between these cohorts of monkeys were observed in the numbers of spot-forming cells (SFC) generated following either the DNA prime or recombinant poxvirus boost immunizations (P = 0.19 and 0.14, respectively). Further, the magnitudes of the PBL Gag-specific SFC responses of these groups of monkeys were indistinguishable on the day of SHIV-89.6 challenge (P = 0.20).
Following challenge, however, PBL from the Mamu-A*01+ monkeys demonstrated higher-frequency Gag-specific cellular immune responses than did PBL of the Mamu-A*01− monkeys. The mean SFC responses of PBL of the Mamu-A*01+ and Mamu-A*01− monkeys on day 14 postchallenge were 6,167 ± 792 (standard error of the mean [SEM]) and 3,443 ± 550, respectively (P = 0.004); and on day 28 postchallenge these immune responses were 3,496 ± 240 and 1,553 ± 327, respectively (P = 0.0003). Similar results were observed in monkeys receiving DNA prime/rAd boost immunizations (Fig. (Fig.6B).6B). The magnitudes of the peak Gag-specific SFC responses were similar in Mamu-A*01+ and Mamu-A*01− monkeys following both the DNA prime (P = 0.13) and the rAd boost (P = 0.08) immunizations; however, the frequency of Gag-specific responses was higher in Mamu-A*01+ monkeys than in Mamu-A*01− monkeys on the day of challenge (P = 0.003), 16 weeks following the rAd boost. PBL of these Mamu-A*01+ monkeys further demonstrated higher-frequency Gag-specific cellular immune responses immediately following SHIV-89.6P challenge, with mean SFC responses from Mamu-A*01+ and Mamu-A*01− monkeys measuring 4,331 ± 574 and 1,843 ± 417, respectively, on day 14 (P = 0.007), and 3,253 ± 285 and 1,709 ± 220 on day 28 following challenge (P = 0.002). Interestingly, while PBL of both cohorts of vaccinated Mamu-A*01+ monkeys exhibited higher-frequency Gag-specific cellular immune responses than PBL of vaccinated Mamu-A*01− monkeys following SHIV-89.6P infection, this difference in cellular immune responses did not extend to all viral antigens. In fact, cellular immune responses to 89.6P Env were equivalent in PBL of vaccinated Mamu-A*01+ and Mamu-A*01− monkeys on days 14 and 28 following challenge in both vaccine studies (Fig. (Fig.77).
We then assessed the extent to which the differences in the magnitudes of the Gag-specific T-cell responses reflected the contribution of the dominant p11C epitope-specific CTL response. We first evaluated the contribution of p11C epitope-specific CTL to the total Gag-specific cellular immune response measured in individual DNA prime/rAd boost-immunized Mamu-A*01+ monkeys on day 28 postchallenge. As shown in Fig. Fig.88 (left panel), SFC responses to the p11C epitope peptide (hatched bars) were equivalent in magnitude to the responses specific for the entire Gag peptide pool (solid bars) for seven of eight monkeys. These data suggest that CTL specific for the p11C epitope dominate the Gag-specific immune response in the Mamu-A*01+ monkeys. For comparison, we also measured Gag epitope-specific CTL responses for seven monkeys in the Mamu-A*01− group that expressed the MHC class I allele Mamu-A*02. The Gag-derived peptide p17G (also referred to as GY9) has previously been described as a dominant Mamu-A*02-restricted CTL epitope (28). As shown in Fig. Fig.88 (middle panel), SFC responses specific for the p17G peptide (hatched bars) were equivalent in magnitude to responses specific for the entire Gag peptide pool (solid bars) for all seven monkeys. Yet, despite the fact that p17G-specific CTL are dominant Gag-epitope-specific CTL in Mamu-A*02+ rhesus monkeys, cellular immune responses to the Gag antigen were significantly lower in PBL of Mamu-A*02+ monkeys (P = 0.004) and Mamu-A*01−/A*02− monkeys (P = 0.006) (Fig. (Fig.8,8, right panel) than in PBL of the Mamu-A*01+ animals.
Tetramer staining analysis provided further evidence for the contribution of the CD8+ T-cell responses specific for the p11C epitope to viral containment postchallenge in the Mamu-A*01+ monkeys. Tetramer staining was performed on PBL of the vaccinated-challenged monkeys to determine the magnitudes of the p11C-specific CTL responses in the Mamu-A*01+ monkeys and of the p17G-specific CTL responses in Mamu-A*02+ monkeys following both vaccination and challenge (Fig. (Fig.9).9). Mean responses to p11C and p17G were 6.7% ± 1.6% and 1.9% ± 0.7% of CD3+ CD8+ T lymphocytes, respectively, following the rAd boost immunization (P = 0.006), 27.3% ± 6.8% and 7.4% ± 1.7% on day 14 postchallenge (P = 0.029), and 16.5% ± 3.6% and 4.3% ± 1.2% on day 28 postchallenge (P = 0.009). Thus, the tetramer staining data indicated that the p11C-specific CTL responses in PBL of the Mamu-A*01+ monkeys were significantly higher in magnitude than the p17G-specific CTL responses in PBL of the Mamu-A*02+ monkeys, both postvaccination and postchallenge. Together, these data demonstrate that vaccinated Mamu-A*01+ monkeys develop a higher-magnitude response to Gag following SHIV-89.6P challenge than vaccinated Mamu-A*01− monkeys and that this enhanced response reflects the contribution of a single dominant Gag-epitope-specific CTL response.
While we have demonstrated an association between the expression of the MHC class I molecule Mamu-A*01 and delayed disease progression in rhesus monkeys infected with SHIV-89.P, this protective effect was apparent in the setting of prior vaccination but not in that of immunologically naive animals. In the previous studies demonstrating an association between Mamu-A*01 expression and clinical protection in monkeys, immunologically naive monkeys were experimentally infected with the pathogenic virus SIVmac251, SIVsmE660, or SIVmac239 (15, 16, 20, 21). The Mamu-A*01+ monkeys demonstrated lower set point viral loads, a greater preservation of CD4+ T lymphocytes, or increased survival compared to Mamu-A*01− monkeys. Our inability to demonstrate significant clinical differences between naive Mamu-A*01+ and Mamu-A*01− monkeys infected with SHIV-89.6P may be explained by the dramatic loss of CD4+ T lymphocytes that is seen within the first few weeks following SHIV-89.6P infection.
Studies of both humans and monkeys have demonstrated that robust CD4+ T-helper-cell responses generated early after infection can ameliorate disease progression, most likely due to the ability of these cells to potentiate antiviral CD8+ T-cell and B-cell immunity (11, 12, 24). However, monkeys infected with SHIV-89.6P have been shown to develop a rapid and almost complete depletion of CXCR4-expressing naïve and central memory CD4+ T cells (19). Unvaccinated monkeys that are infected with SHIV-89.6P may therefore lose the CD4+ T-cell help needed to mount efficient effector CTL responses, thereby eliminating the contribution of the Mamu-A*01-associated CTL-associated protection. The inability of unvaccinated Mamu-A*01+ monkeys infected with SHIV-89.6P to develop robust Gag-specific cellular immune responses is consistent with this hypothesis (Fig. (Fig.3).3). Whether there are qualitative differences in the CTL populations generated in unvaccinated and vaccinated monkeys following SHIV-89.6P infection remains to be elucidated.
A recent study demonstrated attenuated disease progression in unvaccinated Mamu-A*01+ monkeys infected with SHIV-89.6P (29). Infected monkeys expressing Mamu-A*01 were reported to develop lower viral loads in lymphoid tissues and better preservation of lymph node architecture than monkeys that were Mamu-A*01−. These particular parameters of disease were not assessed in the present report. While Mamu-A*01+ monkeys in that earlier study had longer survival times than Mamu-A*01− monkeys, no significant differences in acute or chronic plasma viral RNA levels or preservation of peripheral blood CD4+ T lymphocytes were observed following infection between these two groups of animals. The discrepancy in the duration of survival of Mamu-A*01+ monkeys between the earlier and present study may reflect other contributing genetic factors.
Following SHIV-89.6P infection, robust cellular immune responses were observed for all monkeys that had previously received experimental vaccines. The populations of virus-specific T cells that expanded in the vaccinated monkeys following infection likely contributed to a partial containment of viral replication and an associated blunting of the CD4+ T-cell loss during the first days after infection. Interestingly, the peak Gag-specific cellular immune responses elicited in Mamu-A*01+ and Mamu-A*01− monkeys following both the DNA prime and recombinant poxvirus or rAd boost immunizations were of similar frequencies. However, in the DNA prime/rAd boost-immunized monkeys, but not in the DNA prime/recombinant poxvirus-boosted monkeys, higher-frequency Gag-specific responses were detected in Mamu-A*01+ than in Mamu-A*01− monkeys on the day of challenge, 16 weeks following the last immunization. Whether rAd and recombinant poxvirus vectors differ in their abilities to generate long-lived CTL populations when utilized as vaccine-boosting modalities warrants evaluation. Following viral challenge, the Mamu-A*01+ monkeys in both vaccine studies generated significantly higher-frequency Gag-specific T-cell responses. Most or all of this differential in T-cell responses could be attributed to p11C-specific CTL. The fact that cellular immune responses to 89.6P Env were of similar magnitudes in vaccinated Mamu-A*01+ and Mamu-A*01− monkeys following infection (Fig. (Fig.7)7) further suggests that the dominant CTL response to the p11C epitope provided the incremental protection observed in Mamu-A*01+ monkeys. Even the immunodominant Gag-specific CTL responses elicited in immunized Mamu-A*02+ monkeys were lower in frequency than the Mamu-A*01−-restricted p11C responses following challenge. These data suggest that Mamu-A*01+ monkeys may have a significant advantage in generating effective populations of T cells that can then expand rapidly following reexposure to antigen.
We are grateful to Michelle Lifton, Darci Gorgone, Kristin Beaudry, Kristi Martin, Margaret Beddall, Ayako Miura, Birgit Korioth-Schmitz, Georgia Krivulka, and Faye Yu for excellent technical assistance, Srini Rao, Jim Treece, Sharon Orndorff, and Debra Weiss for management of nonhuman primate studies, Vi Dang and Alida Ault for conduct of animal studies, and Nancy Miller for helpful conversations.
This work was supported by National Institutes of Health grant AI30033 and National Cancer Institute/SAIC-Frederick contract DD1114.