Fig 1.
Rationale for hypothesizing preferential T-cell recognition of early versus late HIV gene products in ART-treated individuals.
Fig 2.
Comparing recognition of latently-infected and partially reactivated HIV-infected cells by Nef- and Gag-specific CD8+ T-cells.
A. Timeline for generation of primary cell latency model. B. Characterization of latently-infected cells from two ART-treated participants showing a lack of Gag expressing cells post-sorting on Day 17, and reactivation of Gag expression from these populations following 48 hours of stimulation with anti-CD3/anti-CD28. D. Latently-infected cells from Day 17 (post-sort) were used for CD8+ T-cell recognition assays. Flow cytometry data from 10 hours of stimulation with bryostatin, or unstimulated controls, show a lack of detectable induction of Gag expression at this early time point. CD4 downregulation is observed as a result of bryostatin stimulation. Note that these are the same cells that show Gag expression following 48 hours of stimulation in B. D—F. Cells from this 10 hour stimulation time point were co-cultured with autologous HIV-Gag-specific (D), CMV-pp65-specific (neg control, E), or HIV-Nef-specific (F) CD8+ T-cell clones. Recognition of target cells by CD8+ T-cell clones was measured by degranulation (CD107a exposure). The results show recognition of latently-infected and partially reactivated cells by Nef-specific CD8+ T-cell clones, contemporaneous with a lack of detectable Gag expression and lack of recognition by Gag-specific CD8+ T-cells. Mean ± SD values are shown and p values were calculated by Student’s T test.
Fig 3.
Magnitudes of T-cell responses to peptide pools, and correlations with virologic parameters.
A. Peptide pools were tested in duplicate. CMV pp65 and EBV BZLF1 peptides are included as controls as responses to these other viruses are unlikely to be related to HIV reservoirs. Each data point represents the mean number of spots per 106 PBMCs following background subtraction. Vertical lines and error bars represent mean and standard deviation for each peptide pool. B. Correlations between IFN-γ -producing HIV-specific T-cell responses (as displayed in A) and frequencies of HIV-infected cells as measured by cell associated HIV DNA, with each dot representing a single individual. Spearman’s r and p values are given for each peptide pool. C. Tests for correlations between T-cell responses and additional virologic parameters (cell-associated and cell-free HIV RNA). D. The ELISPOT data set depicted in B were split based on duration of ART (4 years, or >4 years). The results are consistent with the correlation between Nef/Tat/Rev-specific T-cell responses and HIV DNA emerging after years of ART. E. Correlations between IFN-γ -producing HIV-specific T-cell responses and frequencies of HIV-infected cells as measured by cell associated HIV DNA in the Toronto cohort.
Table 1.
Clinical characteristics of study participants.
Fig 4.
Magnitudes and breadths of Nef/Tat/Rev-specific T-cell responses, and correlations with virologic parameters.
T-cell responses were measured by IFN-γ ELISPOT using a peptide matrix pool strategy (see Methods). A. Total magnitudes of responses to each gene product, calculated as sums of responses to individual matrix pools following background subtraction. Vertical lines and error bars represent mean and standard deviation for each gene product. B. Correlations between T-cell responses (as in A) and frequencies of HIV-infected cells as measured by cell associated HIV DNA. C. The total numbers of Nef epitopes recognized by each individual are shown plotted against frequencies of HIV-infected cells. D. Tests for correlations between breadths of T-cell responses (as in C) and frequencies of HIV-infected cells.
Fig 5.
Polyfunctionality of HIV-specific T-cell responses, and tests for correlations with virologic parameters.
A. Shown are frequencies of CD8+ and CD4+ T-cells that responded with the indicated effector functions following stimulation with HIV and CMV peptide pools. Each dot represents the background-subtracted mean response from a given individual (tested in duplicate). Vertical lines and error bars represent mean and standard deviation for each gene product. B. Tests for correlations between responses depicted in A and virologic parameters. Significant associations (p < 0.05) are indicated in bold.