AIDS virus specific CD8+ T lymphocytes against an immunodominant cryptic epitope select for viral escape

Abstract

Cryptic major histocompatibility complex class I epitopes have been detected in several pathogens, but their importance in the immune response to AIDS viruses remains unknown. Here, we show that Mamu-B*17(+) simian immunodeficiency virus (SIV)mac239-infected rhesus macaques that spontaneously controlled viral replication consistently made strong CD8(+) T lymphocyte (CD8-TL) responses against a cryptic epitope, RHLAFKCLW (cRW9). Importantly, cRW9-specific CD8-TL selected for viral variation in vivo and effectively suppressed SIV replication in vitro, suggesting that they might play a key role in the SIV-specific response. The discovery of an immunodominant CD8-TL response in elite controller macaques against a cryptic epitope suggests that the AIDS virus-specific cellular immune response is likely far more complex than is generally assumed.

Figure 1.

Location of cRW9 in the SIVmac239 genome. The cRW9 epitope (enclosed in box) is located in the same reading frame (frame 3 in this image) as exon 1 of the rev ORF but downstream of the only known splice donor site (indicated by an arrow).

Figure 2.

Magnitude of the cRW9-specific response in four animals. (a) IFN-γ ELISPOT to determine if SIV-infected macaques (Mamu-B*17⁺ or Mamu-B*17⁻) or SIV-naive animals make cRW9-specific responses. Responses are measured as spot-forming cells (SFC) per million PBMCs. The mean number of spots of the negative control wells was subtracted from each well. Responses were considered positive if the number of spots (per million PBMCs) was >50 (represented by a vertical bar). Error bars represent the mean ± the SE for each measurement. (b) IFN-γ ELISPOT to test four animals in the chronic phase of SIV infection for responses to known SIV epitopes restricted by MHC-I alleles expressed by each animal. For each animal, a gray bar indicates the cRW9 response, while all others are black. On the x axis are the epitopes tested and their restricting alleles. (c) Elite controllers r95071 and r98016 were also tested for response frequency using tetramers of the Mamu-B*17 molecule loaded with the cRW9 peptide. Tetramer⁺ populations are expressed as percentages of CD3⁺ lymphocytes. (d) Animal r95071 was also tested for responses to epitopes restricted by the MHC-I molecules Mamu A*02 and Mamu-B*17. Percentages in the pie chart represent the frequency of each response (by tetramer) relative to other measured responses.

Figure 3.

Sequence variation in cRW9 associated with loss of binding and positive selection. (a) Nucleotide alignments of five animals sequenced at time of death reveals possible escape in the cRW9 epitope. Animal r95003 is marked with an asterisk because this animal is Mamu-B*17 ⁻ yet made characteristic Mamu-B*17 responses (Fig. 2 a and explained in Results and Discussion). (b) Relative binding of the cRW9 and the mutant peptide to the Mamu-B*17 molecule. The tryptophan, W, to arginine, R, mutation at the C terminus of the cRW9 peptide reduces binding to the Mamu-B*17 molecule by 97%. Binding is measured as the concentration of peptide at which binding of a radioactively labeled reference peptide is reduced by 50% (IC₅₀). (c) Means (± SE) for five monkeys of the number of synonymous substitutions per 100 sites (d_S) and of the number of nonsynonymous substitutions per 100 nonsynonymous sites (d_N) in comparisons between time of death and inoculum (SIVmac239) sequences.

Figure 4.

Viral escape in cRW9 leads to loss of recognition. (a) PBMCs from animal r95071 were tested for their ability to recognize (measured by spot-forming cells [SFC] per 10⁶ total cells; y axis) the wild-type cRW9 and the mutant peptides at a titration of peptide concentrations (x axis). Error bars represent the mean ± the SE for each peptide concentration. (b) Frequency of the cRW9 response, measured as a percentage of CD3⁺ CD8⁺ lymphocytes (right y axis, filled box symbols), from animals r98015 and r97044 was measured by staining PBMCs from various time points with Mamu-B*17 tetramers loaded with cRW9 peptide (x axis) to determine the frequency of the cRW9-specific response before, during, and after apparent escape from the cRW9 response. The virus load of the animals is included for reference (left y axis, filled triangles). (c) cRW9-specific CD8-TLs were measured for their ability to recognize Mamu-B*17⁺ target cells infected with wild-type versus mutant SIV-T6913C. Cells were infected using the magnetofection technique (reference 19). After 24 h, recognition was assayed by ICS using IFN-γ and TNF-α production. The far left panel shows the response of the cRW9-specific cell line to MHC-matched B cells pulsed with the cRW9 peptide, and the adjacent panel shows the response to the same B cells with no peptide pulse.

Figure 5.

cRW9-specific CD8-TLs effectively suppress SIV replication in Mamu-B*17⁺ target cells. Quantitative PCR of vRNA on the final day (day 7) of the assay. The ability of a cRW9-specific cell line derived from animal r95071 to suppress wild-type or SIV-T6913C replication in target cells that were either Mamu-B*17⁺ or Mamu-B*17⁻ was measured. (a) The data is represented as fold-reduction between the wells with no CD8-TLs added (no CTL) and those with an E/T ratio of 1:20 (+CTL). (b) The same data as in panel a but represented as the total vRNA copies/ml supernatant on day 7 of the assay. Error bars represent the mean ± the SE for each measurement.