The ability of an AIDS virus to escape from immune containment by selective mutation away from recognition by CTL was explored in simian immunodeficiency virus of macaques (SIVmac)-infected rhesus monkeys. CTL recognition of a previously defined common viral mutation in an immunodominant SIVmac Gag epitope was evaluated. CTL were assessed for their ability to recognize a SIVmac Gag protein with a single residue 2 (T → A) replacement in the minimal epitope peptide bound by the MHC class I molecule Mamu-A*01. SIVmac Gag-specific CTL lysed Mamu-A*01+ target cells infected with recombinant vaccinia virus expressing the wild-type but not the mutant Gag protein. In addition, CTL recognized the mutant epitope peptide less efficiently than the wild-type virus peptide. In studies to determine the mechanism by which the mutant virus evaded CTL recognition, this peptide was shown to bind Mamu-A*01 in a manner that was indistinguishable from the wild-type peptide. However, experiments in which an increasing duration of delay was introduced between peptide sensitization of target cells and the assessment of these cells as targets in killing assays suggest that the mutant peptide with a T → A replacement had a higher off-rate from Mamu-A*01 than the wild-type peptide did. Therefore, these findings suggest that AIDS viruses can evade virus-specific CTL responses through the accelerated dissociation of mutant peptide from MHC class I.

Accumulating evidence suggests that CTL play an important role in the immune containment of HIV replication. It has been shown that the emergence of virus-specific CTL is associated with a reduction of viremia during the acute infection of individuals with AIDS viruses (1, 2, 3, 4). Moreover, high-frequency virus-specific CTL responses appear to contribute to a decrease in virus load or to a delay in disease progression in chronically HIV-1-infected persons (5, 6, 7, 8). Nevertheless, viral replication is never fully controlled in most if not all infections. It has been proposed that this incomplete immune control of AIDS virus replication in infected individuals may be explained in part by the emergence of viral mutants capable of evading CTL recognition. To date, HIV-1 mutants that can escape from CTL recognition have been found in only a limited number of infected persons (9, 10, 11, 12, 13, 14, 15, 16, 17, 18). The difficulty in identifying such escape mutants may be attributed to the small number of well-defined dominant epitope-specific CTL responses in HIV-1-infected persons. Further studies are needed to characterize the mechanisms by which HIV-1 might escape from CTL and the role of viral escape mutants in HIV-1 disease progression.

The SIV-infected rhesus monkey provides an ideal model system in which to explore AIDS virus mutation to evade cell-mediated immune responses. Monkeys can be selected for study on the basis of their MHC class I alleles that will develop predictable, well-defined dominant SIV epitope-specific CTL responses. Moreover, these animals can be inoculated with a pathogenic virus isolate that has been molecularly characterized, facilitating a precise definition of the virus that initiates the infection. We previously addressed the possibility that SIVmac might escape by mutation from a CTL response in chronically infected rhesus monkeys. This study was done in monkeys expressing the MHC class I gene Mamu-A*01 that develop an immunodominant SIVmac Gag epitope-specific CTL response after simian immunodeficiency virus of macaques (SIVmac)4 infection (19). In that study, mutant virus encoding a change from threonine (T) to alanine (A) in the immunodominant Gag epitope (residues 181–189 of the Gag protein) emerged in two of three evaluated Mamu-A*01+ monkeys infected with SIVmac (20). Experiments in which 12 amino acid peptides (p11C, p11C/2A) containing the immunodominant epitope (20) were used to sensitize target cells for lysis did not suggest that this T → A replacement would result in escape from the epitope-specific CTL recognition (20). However, it still remained possible that this mutant protein may not be properly processed in the intracellular MHC class I pathway for recognition by the epitope-specific CTL or that the mutant peptide may not form an optimally stable MHC class I peptide complex. Therefore, we undertook further studies to assess the possibility that SIVmac isolates with this T → A replacement might escape from the immunodominant CTL response in Mamu-A*01+ rhesus monkeys.

Rhesus monkeys (Macaca mulatta) were used in these studies. These animals were maintained in accordance with the guidelines of the Committee on Animals for Harvard Medical School and the Guide for the Care and Use of Laboratory Animals (National Academy Press, 1996). All monkeys were inoculated i.v. with SIVmac 251 strain, as described previously (20). During the period of the studies, the monkeys were infected with SIVmac for 2–5 years but showed no evidence of an AIDS-like syndrome.

The plasmid pM40K containing the entire coding region of SIVmac 251 gag was engineered using site-directed mutagenesis to encode the single T → A replacement in the epitope-coding region (21, 22). This was done following the instructions of the QuickChange site-directed mutagenesis kit (Stratagene, La Jolla, CA). The sequence of oligonucleotides used to generate this change was 5′- CAC TGT CAG AAG GTT GCG CCC CCT ATG ACA TTA ACT-3′. The mutated base G and the other two bases, GCC, encoded the change in amino acid from threonine to alanine. The sequence of plasmid DNA containing the desired substitution was confirmed by sequencing. A recombinant vaccinia virus expressing this T → A replacement of SIVmac was then created through homologous recombination using a host range selection system and the plasmid pM40K containing the sequence encoding the single T → A replacement in the epitope (23). The selected recombinant vaccinia viruses were amplifed and titrated using RK13 cells. vGag/182A-2 and vGag/182A-3 were viruses generated from two clones of the corresponding mutated PK40 plasmids. As controls, recombinant vaccinia viruses were also generated that expressed the wild-type SIVmac gag or another SIVmac gag insert encoding a single amino acid replacement 21 aa N-terminal to the epitope-coding sequence (vGag/161V). These three recombinant vaccinia viruses were assessed for the expression of SIVmac Gag after infection of B lymphoblastoid cell line (B-LCL) expressing the Mamu-A*01 gene. In the experiments using effector cells from monkeys 597 and 403, the target B-LCL were infected overnight with the recombinant vaccinia viruses. The infected B-LCL were then divided into two aliquots, one for CTL assays and the other for 35S-translabeling and SDS-PAGE characterization. The recombinant vaccinia virus-infected B-LCL were assessed for viability by trypan blue exclusion before CTL assays. Furthermore, immune fluorescence analysis showed that SIVmac Gag protein was expressed in the B-LCL targets infected with the recombinant viruses.

Rhesus monkey B-LCL were generated by incubating 105 Ficoll-diatrizoate-isolated PBL in 100 μl of culture medium with 100 μl of S594 supernatant. S594 is a cell line productively infected with the transforming baboon Herpesvirus papio (20). The B-LCL were transformed and maintained in RPMI 1640 (Life Technologies, Grand Island, NY) supplemented with l-glutamine (2 mM), penicillin (50 U/ml), streptomycin (50 μg/ml), and 10% FCS (HyClone, Logan, UT).

Rhesus monkey B-LCL immortalized with Herpesvirus papio served as target cells. The B-LCL were incubated at a cell concentration of 1 × 106/ml with recombinant vaccinia viruses carrying the SIVmac gag, with a control (equine herpesvirus gH) gene at 10 PFU, or with decreasing concentrations of synthetic peptides for 8 h at 37°C in a 5% CO2 humidified atmosphere (20). CTL derived from PBL of SIVmac-infected monkeys were used as effector cells in a standard 51Cr-release assay performed in U-bottom 96-well microtiter plates. 51Cr-labeled target cells were incubated for 5 h with effector cells at different E:T ratios. Spontaneous release varied from 10 to 20%. Specific release was calculated as [(experimental release − spontaneous release)/(100% release − spontaneous release)] × 100.

Peptide binding to a C1R cell line stably transfected with Mamu-A*01 was assessed as previously described (24). Cells were incubated overnight at 26°C in the presence of 3 μg/ml human β2-microglobulin (β2m). The next day cells were split into separate aliquots and incubated with 100,000 cpm of the iodinated reference peptide ATPYDINQM and different concentrations of the test peptides CTPYDINQM or CAPYDINQM at 20°C for 4 h. Cell pellets were then spun down, washed, and the incorporated 125I was measured by resuspension in Optiphase and counting in a gamma scintillation counter. Percent inhibition of binding was calculated as [1 − (incorporated cpm in the presence of competitor peptide)/(total incorporated cpm in the absence of competitor peptide) × 100].

Different concentrations (100 μM, 10 μM, and 1 μM) of the wild-type peptide p11C,C-M or the mutant peptide p11C,C-M/2A were added to fixed quantities of rhesus monkey Mamu-A*01 heavy chain and human β2m as described previously (25). The same quantity of the 150-kDa protein alcohol dehydrogenase was added to each reaction as a control standard for determining the relative folding efficiency. Formation of the folded 43-kDa Mamu-A*01/peptide/β2m complex was monitored by gel filtration on a TSK SWxl 3000 column (Tosohaas, Montgomeryville, PA)

The kinetics of target cell-peptide sensitization for CTL-mediated lysis was determined using a peptide-MHC class I stability assay as described by Goulder et al. (10). Mamu-A*01+ target cells were pulsed with 20 μM of the 9 aa wild-type or mutant peptide and washed twice with FCS-free RPMI 1640 medium. In the CTL experiments other than those shown in Fig. 6, peptide sensitization and chromium labeling of target cells were done simultaneously, with an 8-h incubation before the addition of effector cells. In optimizing CTL peptide-MHC class I stability assays, we found that the specific lysis of targets incubated for 8 h with peptides was similar to the specific lysis of target cells incubated for 2 h. Thus, the peptide-loaded target cells were incubated for 0, 2, 16, or 18 h before addition of CTL effector cells. The impact of this period of incubation on CTL lysis was evaluated for the mutant peptide in comparison with the wild-type peptide. An inverse correlation between target cell lysis and the incubation time for a given peptide indicates an unstable peptide-Mamu-A*01 complex.

FIGURE 6.

CTL peptide/MHC stability assays showed that the T → A replacement accelerated the dissociation of the mutant peptide from Mamu-A*01. Mamu-A*01+ target cells were pulsed with 20 μM of the wild-type (▨) and mutant (▪) 9-aa peptides, washed twice to remove unloaded peptides, and then incubated at 37°C for increased numbers of hours before the addition of CTL effector cells. The loss of CTL recognition of target cells as the duration of time after peptide pulsing increased indicated the lack of stability of the peptide-Mamu-A*01 complex. Data were derived from up to three repeated experiments, with error bars indicating the variability among the CTL assays.

FIGURE 6.

CTL peptide/MHC stability assays showed that the T → A replacement accelerated the dissociation of the mutant peptide from Mamu-A*01. Mamu-A*01+ target cells were pulsed with 20 μM of the wild-type (▨) and mutant (▪) 9-aa peptides, washed twice to remove unloaded peptides, and then incubated at 37°C for increased numbers of hours before the addition of CTL effector cells. The loss of CTL recognition of target cells as the duration of time after peptide pulsing increased indicated the lack of stability of the peptide-Mamu-A*01 complex. Data were derived from up to three repeated experiments, with error bars indicating the variability among the CTL assays.

Close modal

The circulating virions and proviruses in six Mamu-A*01+ monkeys were assessed for the mutation encoding the T → A replacement in the epitope corresponding to p11C,C-M. To this end, plasma SIV RNA was isolated and reverse-transcribed to cDNA (20); proviral DNA was extracted from PBL of infected monkeys as described previously (20). The SIV cDNA and proviral DNA were then characterized for mutations in the epitope-coding region using PCR-based cloning and sequencing strategy (20). Up to 80 clones generated by PCR from each SIV cDNA or DNA sample were analyzed. Sequence analyses showed that the mutation encoding the T → A replacement in the epitope could be identified both in the plasma SIV RNA and proviral DNA in PBL of the infected monkeys (data not shown).

To determine whether virally expressed Gag containing the T → A mutation would be endogenously processed and recognized by epitope-specific CTL, we generated a recombinant vaccinia-SIVmac gag construct encoding this mutant Gag sequence. As controls, recombinant vaccinia viruses were also generated that expressed the wild-type SIVmac gag or a control mutant SIVmac gag encoding a single amino acid replacement 21 aa N-terminal to the epitope-coding sequence (vGag/161V). These three recombinant vaccinia viruses expressed SIVmac Gag in infected Mamu-A*01+ B-LCL (Fig. 1). B-LCL infected with these recombinant vaccinia viruses were then assessed for recognition by Gag epitope-specific CTL (Fig. 2). Epitope-specific effector cells generated from Mamu-A*01+ monkeys infected with SIVmac lysed B-LCL infected with vaccinia virus expressing the wild-type and control mutant SIVmac Gag but not the T → A mutant Gag (Fig. 2). These results indicated that the mutant Gag containing the T → A replacement expressed in a recombinant vaccinia virus failed to be recognized by the CTL and suggested that processing or presentation of the mutant peptide was impaired.

FIGURE 1.

SDS-PAGE analysis of expression of SIVmac Gag proteins by recombinant vaccinia viruses. B-LCL infected overnight with recombinant vaccinia viruses were 35S-translabeled for 6 h and assessed for expression of SIVmac Gag protein by precipitating lysates with sera from two SIVmac-infected monkeys. The size of the precursor protein of SIVmac Gag (55 Kd) is indicated by an arrow. v249 is a recombinant vaccinia virus that expresses the control equine herpesvirus gH gene vGag/161V (simplified as v161V in Fig. 2) and expresses the full-length SIVmac Gag protein containing a single replacement (V) 21 aa N-terminal to the epitope. vGag/182A (simplified as v182A in Fig. 2) expresses the full-length SIVmac Gag protein containing a T → A replacement at the second position of the epitope (vGag/182A-2 and vGag/182A-3 were viruses generated from two clones of the corresponding mutated PK40 plasmids). vGag expresses the full-length wild-type SIVmac Gag protein.

FIGURE 1.

SDS-PAGE analysis of expression of SIVmac Gag proteins by recombinant vaccinia viruses. B-LCL infected overnight with recombinant vaccinia viruses were 35S-translabeled for 6 h and assessed for expression of SIVmac Gag protein by precipitating lysates with sera from two SIVmac-infected monkeys. The size of the precursor protein of SIVmac Gag (55 Kd) is indicated by an arrow. v249 is a recombinant vaccinia virus that expresses the control equine herpesvirus gH gene vGag/161V (simplified as v161V in Fig. 2) and expresses the full-length SIVmac Gag protein containing a single replacement (V) 21 aa N-terminal to the epitope. vGag/182A (simplified as v182A in Fig. 2) expresses the full-length SIVmac Gag protein containing a T → A replacement at the second position of the epitope (vGag/182A-2 and vGag/182A-3 were viruses generated from two clones of the corresponding mutated PK40 plasmids). vGag expresses the full-length wild-type SIVmac Gag protein.

Close modal
FIGURE 2.

CTL lyse B-LCL infected with a recombinant vaccinia virus expressing a wild-type but not a mutant SIVmac gag encoding a T → A replacement in a dominant epitope. A, Mamu-A*01+ B-LCL infected with 10 PFU of the noted recombinant vaccinia viruses were assessed for lysis using varied E:T ratios as indicated. The effector cells were PBL of SIVmac-infected Mamu-A*01+ rhesus monkeys stimulated in vitro with Con A for 3 days and then expanded in IL-2-containing medium for an additional 3 days. B, Mamu-A*01+ B-LCL infected with various PFU of recombinant vaccinia viruses were assessed for specific lysis by CTL using an 80:1 E:T ratio. Effector cells were prepared as described above.

FIGURE 2.

CTL lyse B-LCL infected with a recombinant vaccinia virus expressing a wild-type but not a mutant SIVmac gag encoding a T → A replacement in a dominant epitope. A, Mamu-A*01+ B-LCL infected with 10 PFU of the noted recombinant vaccinia viruses were assessed for lysis using varied E:T ratios as indicated. The effector cells were PBL of SIVmac-infected Mamu-A*01+ rhesus monkeys stimulated in vitro with Con A for 3 days and then expanded in IL-2-containing medium for an additional 3 days. B, Mamu-A*01+ B-LCL infected with various PFU of recombinant vaccinia viruses were assessed for specific lysis by CTL using an 80:1 E:T ratio. Effector cells were prepared as described above.

Close modal

We then sought to determine whether a peptide containing the T → A replacement could form a complex with Mamu-A*01 that would be recognized by CTL. In experiments using 12 aa peptides containing the epitope, differential CTL recognition of the mutant peptide with the T → A replacement and the wild-type peptide was not clear-cut. The concentration of mutant peptide required for sensitizing a target cell for CTL recognition appeared to be 10 times higher than the concentration needed for sensitizing targets with wild-type peptide (Refs. 20, 24 and data not shown). However, the ability of the mutant peptide to evade CTL recognition was evident when two 9-aa peptides, which corresponded to the wild-type (p11C,C-M; CTPYDINQM) and the mutant (p11C,C-M/2A; CAPYDINQM) viral sequences, were used in CTL assays. Although target cells were sensitized for lysis by the wild-type p11C,C-M at peptide concentrations as low as 1 ng/ml, 1000-fold higher concentrations of mutant p11C,C-M/2A were required to sensitize target cells for comparable levels of epitope-specific lysis (Fig. 3). Thus, the results of experiments using optimal epitope peptides support the observation made in the study of CTL recognition of the mutant Gag in the vaccinia expression system. The mutant Gag containing a T → A replacement at the second position of the p11C,C-M epitope appears capable of escaping from CTL recognition.

FIGURE 3.

Epitope-specific CTL inefficiently lyse Mamu-A*01+ B-LCL pulsed with mutant p11C,C-M/2A. Peptide titration in sensitizing B-LCL for CTL lysis showed a marked decrease in CTL recognition of 9-aa mutant peptide. The E:T ratio used in the study of cells from monkey 403 was 50:1; the ratio for the study using cells from monkey 297 was 100:1. The sequences of wild-type and mutant peptides are as follows: P11C,C-M, CTPYDINQM; P11C,C-M/2A, CAPYDINQM; P11B (the control peptide), QALSEGCTPYDI.

FIGURE 3.

Epitope-specific CTL inefficiently lyse Mamu-A*01+ B-LCL pulsed with mutant p11C,C-M/2A. Peptide titration in sensitizing B-LCL for CTL lysis showed a marked decrease in CTL recognition of 9-aa mutant peptide. The E:T ratio used in the study of cells from monkey 403 was 50:1; the ratio for the study using cells from monkey 297 was 100:1. The sequences of wild-type and mutant peptides are as follows: P11C,C-M, CTPYDINQM; P11C,C-M/2A, CAPYDINQM; P11B (the control peptide), QALSEGCTPYDI.

Close modal

We then sought to characterize the mechanism by which this mutant viral epitope escapes CTL recognition. We first determined whether the absence of CTL recognition of the T → A mutant Gag could be attributed to an inability of the viral peptide to bind to the Mamu-A*01 molecule. Peptide-MHC class I binding studies were conducted using the two 9-aa peptides, which corresponded to the wild-type and the mutant sequences of the epitope (20). These two peptides were assessed for their relative ability to compete with an iodinated index peptide for binding to a Mamu-A*01+ cell line. Interestingly, mutant p11C,C-M/2A was still able to bind quite efficiently to Mamu-A*01 expressed on the surface of cells. Its binding capacity may at most have been only slightly lower than that of the wild-type p11C,C-M (Fig. 4).

FIGURE 4.

The wild-type and mutant peptides bind comparably to cell surface-expressed Mamu-A*01. The ability of the native sequence (CTPYDINQM; □), mutant sequence (CAPYDINQM; ⋄), and control sequence (p11B; ○) peptides were assessed for their ability to compete with the iodinated reference peptide (ATPYDINQM) for binding to Mamu-A*01-transfected C1R cells. The IC50 of binding for CTPYDINQM was 5 nm and for CAPYDINQM was 30 nm.

FIGURE 4.

The wild-type and mutant peptides bind comparably to cell surface-expressed Mamu-A*01. The ability of the native sequence (CTPYDINQM; □), mutant sequence (CAPYDINQM; ⋄), and control sequence (p11B; ○) peptides were assessed for their ability to compete with the iodinated reference peptide (ATPYDINQM) for binding to Mamu-A*01-transfected C1R cells. The IC50 of binding for CTPYDINQM was 5 nm and for CAPYDINQM was 30 nm.

Close modal

To confirm the results of this peptide binding experiment, we initiated small-scale in vitro folding reactions to assess the ability of the two 9-aa peptides to induce folding of the Mamu-A*01-β2m complex at limiting peptide concentrations. Formation of the folded 43-kDa Mamu-A*01-peptide-β2m complex was monitored by gel filtration. As shown in Fig. 5, the mutant and wild-type peptides were equally efficient in inducing the formation of a 43-kDa Mamu-A*01-peptide-β2m complex.

FIGURE 5.

Folding of Mamu-A*01 and human β2m around the native p11C,C-M and the mutant p11C,C-M/2A are comparably efficient. Gel filtration profiles of soluble Mamu-A*01 monomers folded with human β2m and either the wild-type p11C,C-M or the mutant p11C, C-M/2A are shown. The profile of the control alcohol dehydrogenase is also shown. The peptide concentrations (100 μM, 10 μM, and 1 μM) used in each reaction are indicated on the left.

FIGURE 5.

Folding of Mamu-A*01 and human β2m around the native p11C,C-M and the mutant p11C,C-M/2A are comparably efficient. Gel filtration profiles of soluble Mamu-A*01 monomers folded with human β2m and either the wild-type p11C,C-M or the mutant p11C, C-M/2A are shown. The profile of the control alcohol dehydrogenase is also shown. The peptide concentrations (100 μM, 10 μM, and 1 μM) used in each reaction are indicated on the left.

Close modal

These results indicated that the mutation containing the T → A replacement did not significantly interfere with peptide-Mamu-A*01 binding. Therefore, these observations suggest that the absence of CTL recognition of the T → A Gag mutant was not due to the inability of the viral peptide to bind to Mamu-A*01 expressed on the cell surface.

Despite the preserved ability of mutant p11C,C-M/2A to bind Mamu-A*01, the T → A replacement might allow the virus to escape CTL recognition by altering the stability of the peptide-MHC complex, increasing the off-rate of the bound peptide. To assess this possibility, we employed a peptide-MHC class I stability assay (Fig. 6). Mamu-A*01+ B-LCL were pulsed with 20 μM of wild-type p11C,C-M or mutant p11C,C-M/2A and were washed with FCS-free medium. The peptide-loaded cells were then incubated for 0, 2, 16, or 18 h before their use as targets for CTL effector cells. When compared with the wild-type peptide, the mutant peptide with the T → A replacement exhibited a decreased ability over time to maintain a stable Mamu-A*01-peptide complex that was recognized by CTL. Epitope-specific CTL recognized wild-type peptide- and mutant peptide-loaded target cells equivalently when peptide-pulsed targets were incubated for 0 or 2 h before the assay. In contrast, 16 and 18 h after peptide pulsing, the mutant peptide-loaded targets were poorly recognized by the epitope-specific CTL, whereas the wild-type peptide-loaded targets remained susceptible to lysis. These results suggest that the T → A replacement in the epitope accelerates the dissociation of the peptide from Mamu-A*01 despite the preserved capacity of the peptide to bind Mamu-A*01.

These experiments to characterize CTL recognition of recombinant vaccinia virus-expressed Gag, the impact of peptide concentration on target cell sensitization for CTL lysis, and the relative efficiency of peptides to maintain stable peptide/MHC class I complex suggest that the T → A replacement in the p11C,C-M epitope confers upon SIVmac the ability to escape from CTL recognition. The results of the experiments employing peptide titration for sensitizing target cells using the optimal 9-aa p11C,C-M/2A peptide contrasted with those seen in the experiments using the 12-aa p11C and p11C/2A peptide-pulsed cells (Ref. 20 and Fig. 3). These differing results indicate that a precise definition of the optimal CTL epitope is crucial for characterizing escape mutants of AIDS viruses. On the other hand, viral escape from CTL recognition by the T → A replacement was most readily appreciated when using target cells prepared by vaccinia virus expression of the mutant Gag. The absence of CTL recognition of mutant peptides generated in the cell through processing of vaccinia virus-expressed Gag may mimic the naturally occurring events of CTL escape because viral peptides must be processed from endogenously synthesized proteins and presented by MHC class I molecules.

The present studies suggest that the absence of recognition of the mutant epitope by CTL was a result of the rapid dissociation of the Gag peptide from Mamu-A*01. The impact of the T → A replacement on peptide-MHC complex formation and viral escape from CTL recognition was most evident when the peptide-loaded target cells was incubated for a prolonged period of time before they were employed in the CTL assays. Despite its inability to maintain a stable complex with Mamu-A*01, mutant peptide exogenously maintained its ability to bind to Mamu-A*01. This may explain the finding that CTL recognition of the mutant peptide occurred when target cells were pulsed with high but not low concentrations of this protein fragment. A high concentration of the mutant peptide may maximize its association with MHC class I molecules, overcoming the increased off-rate of MHC-peptide interaction. The finding that the mutant peptide rapidly dissociates from Mamu-A*01 may also explain the complete abrogation of CTL recognition of the target cells infected with a recombinant vaccinia virus expressing the mutant Gag. Mutant peptides derived from the endogenous processing pathway may be particularly prone to dissociate from MHC class I molecules, given that these endogenously generated peptides are produced in small quantities in cells and take part in a number of biologic interactions before expression on the surface of cells. The MHC-mutant peptide complex may not be formed or stabilized on the surface of cells after an endogenous processing and thus may not sensitize target cells for recognition by CTL. Nevertheless, we cannot exclude the possibility that the present results reflect a component of minor differences in the on-rate of wild-type and mutant peptides because the peptide binding assay did suggest a subtle, less favorable binding of the mutant peptide to Mamu-A*01 (Figs. 3 and 4).

Mechanisms underlying viral escape from CTL recognition are likely to be complex, involving peptide processing, binding of peptide to MHC class I molecules, and peptide interactions with the TCR of CTL (26). Although EBV and HSV as well as a mutant of murine leukemia virus have been shown to inhibit peptide processing and presentation for CTL recognition (27, 28, 29), precise mechanisms by which HIV-1 mutants escape CTL recognition have not been formally documented (9, 10, 11, 12, 13, 14, 15, 16, 17, 18). Some studies suggest that HIV-1 mutants may evade CTL recognition through interfering with the peptide binding to MHC class I molecules or through interactions with the TCR of CTL (15, 16, 30). In fact, T cell recognition depends on the functional density of the TCR ligand comprised of MHC-peptide complex. It remains possible that the T → A replacement identified in the epitope corresponding to the P11C,C-M can result in a decrease in the affinity threshold that is required for CTL lysis of target cells. Additional studies are needed to address this possibility.

The results of the present studies suggest that an amino acid substitution in the epitope allows the virus to escape from CTL recognition through an accelerated dissociation of the peptide from MHC class I molecules. This in vitro observation implies that the single mutation encoding the T → A replacement in the CTL epitope may play a role in the persistence of SIVmac infection in infected monkeys. The rapid dissociation of the T-A mutant peptide from MHC class I can certainly confer upon the virus a survival advantage through the escape from CTL recognition. SIVmac may evade an antiviral T cell response through the accumulation of mutations in CTL epitopes in infected monkeys.

We thank Dr. Gail Mazzara (Therion Biologics Corporation) for providing the PK40 plasmid containing the full-length SIVmac gag gene and Dr. Alicia Gomez Yafal (Therion Biologics Corporation) for technical assistance with the host range selection system used in these studies.

1

This work was supported by National Institutes of Health Grants AI33628 (to Z.W.C.), AI20729 (to N.L.L.), and AI26507 (to N.L.L.). G.V. was supported by a fellowship from Deutsche Krebsforschungszentrum (Hiedelberg, Germany).

4

Abbreviations used in this paper: SIVmac, simian immunodeficiency virus of macaques; B-LCL, B lymphoblastoid cell line; β2m, β2-microglobulin.

1
Borrow, P., H. Lewicki, B. H. Hahn, G. M. Show, M. B. A. Oldstone.
1994
. Virus specific CD8+ cytotoxic T lymphocyte activity associated with control of viremia in primary human immunodeficiency virus type 1 infection.
J. Virol.
68
:
6103
2
Chen, Z. W., Z. Kou, C. Lekutis, L. Shen, D. Zhou, M. Halloran, J. Li, J. Sodroski, D. Lee-Parritz, N. L. Letvin.
1995
. T cell receptor Vβ repertoire in an acute infection of rhesus monkeys with simian immunodeficiency viruses and a chimeric of simian-human immunodeficiency virus.
J. Exp. Med.
182
:
21
3
Koup, R. A., J. T. Safrit, Y. Gao, C. A. Andrews, G. McLeod, W. Borkowsky, C. Farthing, D. D. Ho.
1994
. Temporal association of cellular immune responses with the initial control of viremia in primary human immunodeficiency virus type 1 syndrome.
J. Virol.
68
:
4650
4
Schmitz, J. E., M. J. Kuroda, S. Santra, V. G. Sasserville, M. A. Simon, M. A. Lifton, P. Racz, K. Tenner-Racz, M. Dalesandro, B. J. Scallon, et al
1997
. Control of viremia in simian immunodefefciency virus infection by CD8+ lymphocytes.
Science
283
:
857
5
Carmichael, A., X. Jin, P. Sissons, L. Borysiewicz.
1993
. Quantitative analysis of the human immunodeficiency virus type 1 (HIV-1)-specific cytotoxic T lymphocyte response at different stages of HIV-1 infection: differential CTL responses to HIV-1 and Epstein-Barr virus in late disease.
J. Exp. Med.
177
:
249
6
Klain, M. R., C. A. van Baalen, A. M. Holwerda, S. R. Kerkhof Garde, R. J. Bende, I. P. M. Keet, J. K. M. Eeftinck-Schattenkerk, A. D. M. E. Osterhaus, H. Schuitemaker, F. Miedema.
1995
. Kinetics of gag-specific cytotoxic T lymphocyte responses during the clinical course of HIV-1 infection: a longitudinal analysis of rapid progressors and long-term asymptomatics.
J. Exp. Med.
181
:
1365
7
Ogg, G. S., X. Jin, S. Bonhoeffer, P. R. Dunbar, M. A. Nowak, S. Monard, J. P. Segal, Y. Cao, S. L. Rowland-Jones, V. Cerundolo, et al
1998
. Quantitation of HIV-1-specific cytotoxic T lymphocytes and plasma load of viral RNA.
Science
279
:
2103
8
Rinaldo, C., X. L. Huang, Z. Fan, M. Ding, L. Beltz, A. Logar, D. Panicali, G. Mazzara, J. Liebmann, M. Cottrill, P. Gupta.
1995
. High levels of anti-human immunodeficiency virus type 1 (HIV-1) memory cytotoxic T lymphocyte activity and low viral load are associated with lack of disease in HIV-1-infected long term nonprogressors.
J. Virol.
69
:
5838
9
Borrow, P., H. Lewicki, S. P. Wei, M. S. Horwitz, N. Peffer, H. Meyers, J. A. Nelson, E. Gairin, B. H. Hahn, M. B. A. Oldstone, G. M. Shaw.
1997
. Antiviral pressure exerted by HIV-1-specific cytotoxic T lymphocytes (CTLs) during primary infection demonstrated by rapid selection of CTL escape virus.
Nat. Med.
3
:
205
10
Goulder, P. J. R., R. E. Phillips, R. A. Colbert, S. McAdam, G. Ogg, M. A. Nowak, P. Giangrande, G. Luzzi, B. Morgan, A. Edwards, et al
1997
. Late escape from an immunodominant cytotoxic T lymphocyte response associated with progression to AIDS.
Nat. Med.
3
:
212
11
Goulder, P. J. R., A. K. Swell, D. G. Lalloo, D. A. Price, J. A. Whelan, J. Evens, G. P. Taylor, G. Luzzi, P. Giangrande, R. E. Phillips, A. J. McMichael.
1997
. Patterns of immunodominance in HIV-1-specific cytotoxic T lymphocyte antigen (HLA)-identical siblings with HLA-A*0201 are influenced by epitope mutation.
J. Exp. Med.
185
:
1423
12
Haas, G., U. Plikat, P. Debre, M. Lucchiary, C. Katlama, Y. Dudoit, O. Bonduelle, M. Bauer, H. Ihlenfeldt, G. Jung, et al
1996
. Dynamics of viral variants in HIV-1 Nef and specific cytotoxic T lymphocytes in vivo.
J. Immunol.
157
:
4212
13
Koenig, S., A. J. Conley, Y. A. Brewah, G. M. Jones, S. Leath, L. J. Boots, V. Davey, G. Pantaleo, J. F. Demarest, C. Carter, et al
1995
. Transfer of HIV-1-specific cytotoxic T lymphocytes to an AIDS patient leads to selection for mutant HIV variants and subsequent disease progression.
Nat. Med.
1
:
330
14
McAdam, S., P. Klenerman, L. Tussey, S. Rowland-Jones, D. Lalloo, R. Phillips, A. Edwards, P. Giangrande, A. L. Brown, F. Gotch, A. McMichael.
1995
. Immunogenic HIV variant peptides that bind to HLA-B8 can fail to stimulate cytotoxic T lymphocyte responses.
J. Immunol.
155
:
2729
15
Meier, U., P. Klenerman, P. Griffin, W. James, B. Koppe, B. Larder, A. McMichael, R. Phillips.
1995
. Cytotoxic T lymphocyte lysis inhibited by viable HIV mutants.
Science
270
:
1360
16
Phillip, R. E., S. Rowland-Jones, D. F. Nixon, F. M. Gotch, J. A. Edwards, A. O. Ogunlesi, J. A. Rothbard, C. R. M. Bangham, C. R. Rizza, A. J. McMichael.
1991
. Human immunodeficiency virus genetic variation that can escape cytotoxic T cell recognition.
Nature
354
:
453
17
Price, D. A., P. J. R. Goulder, P. Klenerman, A. K. Sewell, P. J. Easterbrook, M. Troop, C. R. Bangham, R. E. Phillips.
1997
. Positive selection of HIV-1 cytotoxic T lymphocyte escape variants during primary infection.
Proc. Natl. Acad. Sci. USA
94
:
1890
18
Safrit, J. T., C. A. Andrews, T. Zhou, D. D. Ho, R. A. Koup.
1994
. Characterization of human immunodeficiency virus type-1 specific cytotoxic T lymphocyte clones isolated during acute seroconversion: recognition of autologous sequences within a conserved immunodominant epitope.
J. Exp. Med.
179
:
463
19
Miller, M. D., H. Yamamoto, A. L. Hughes, D. I. Watkins, N. L. Letvin.
1991
. Definition of an epitope and MHC class I molecule recognized by gag-specific cytotoxic T lymphocytes in SIVmac-infected rhesus monkeys.
J. Immunol.
147
:
320
20
Chen, Z. W., L. Shen, M. D. Miller, S. H. Ghim, A. L. Hughes, N. L. Letvin.
1992
. Cytotoxic T lymphocytes do not appear to select for mutations in an immunodominant epitope of simian immunodeficienccy virus gag.
J. Immunol.
149
:
4060
21
Braman, J., C. Papworth, A. Greener.
1996
. Site-directed mutagenesis using double-stranded plasmid DNA templates.
Methods Mol. Biol.
57
:
31
22
Kunkel, T. A..
1985
. Rapid and efficient site-specific mutagenesis without phenotypic selection.
Proc. Natl. Acad. Sci. USA
82
:
488
23
Smith, K. A, V. Stallard, J. M. Roos, C. Hart, N. Cormier, L. K. Cohen, B. E. Roberts, L. G. Payne.
1993
. Host range selection of vaccinia recombinants containing insertions of foreign genes into non-coding sequences.
Vaccine
11
:
43
24
Allen, T. M., J. Sidney, M. F. D. Guercia, R. L. Glickman, G. L. Lensmeyer, D. A. Wiebe, R. DeMars, C. D. Pauza, R. P. Johnson, A. Sette, D. I. Watkins.
1998
. Characterization of the peptide binding motif of a rhesus MHC class I molecule (Mamu-A*01) that binds an immunodominant CTL epitope from SIV.
J. Immunol.
160
:
6062
25
Egan, M. A., M. J. Kuroda, G. Voss, J. E. Schmitz, W. A. Charini, M. A. Forman, N. L. Letvin.
1999
. Use of MHC class I/peptide tetramers to quantitate CD8+ CTL specific dominant and non-dominant viral epitopes in simian-human immunodeficiency virus-infected rhesus monkeys.
J. Virol.
73
:
5466
26
Davenport, M. P..
1995
. Antagonists or altruists: do viral mutants modulate T cell responses.
Immunol. Today
16
:
432
27
Hill, A., P. Jugovic, I. York, G. Russ, J. Bennink, J. Yewdell, H. Ploegh, D. Johnson.
1995
. Herpes simplex virus turns off the TAP to evade host immunity.
Nature
375
:
421
28
Levitskaya, J., M. Coram, V. Levitsky, S. Imreh, P. M. Steigerwald-Mullen, G. Klein, M. G. Kurilla, M. G. Masucci.
1995
. Inhibition of antigen processing by the internal repeat region of the Epstein-Barr virus nuclear antigen-1.
Nature
375
:
685
29
Ossendorp, F., M. Eggers, A. Neisig, T. Ruppert, M. Groettrup, A. Sijts, E. Mengede, P. M. Kloetzel, J. Neefjes, U. Koszinowski, C. Melief.
1996
. A single residue exchange within a viral CTL epitope alters proteasome-mediated degradation resulting in lack of antigen presentation.
Immunity
5
:
115
30
Mortara, L., F. Letourneur, H. Gras-Masse, A. Venet, J. Guillet, I. Bourgault-Villada.
1998
. Selection of virus variants and emergence of virus escape mutants after immunization with an epitope vaccine.
J. Virol.
72
:
1403