Table 1.
Clinical characteristics at study enrollment.
Table 2.
HLA allele genotyping.
Fig 1.
HIV-1 persistent controllers display low levels of CD8 + T cell activation markers.
PBMCs were isolated from blood samples of PWoH (n = 5), VC (n = 5) and SP ART- (n = 4) and flow cytometry was performed gating for live (Zombie Yellow-), CD8+ T cells (CD3+CD8+). A) Representative gating of CD8+ T cells analyzed for the expression of CD38 (x axis) and HLA-DR (y axis). Percentage of live CD8+ T cells that are B) CD38+, C) HLA-DR+, D) CD38-HLA-DR-, E) CD38+HLA-DR-, F) CD38-HLA-DR+, G) CD38+HLA-DR+. Ordinary one-way ANOVA with uncorrected Fisher’s LSD was performed to determine statistical significance. * p ≤ 0.05, ** p ≤ 0.01.
Fig 2.
HIV-1 persistent controllers display low levels of CD8+ T cell exhaustion markers PD-1 and Tim-3.
PBMCs were isolated from blood samples of PWoH (n = 5), VC (n = 8) and SP ART- (n = 4) and flow cytometry was performed gating for live (Zombie Yellow-), CD8+ T cells (CD3+CD8+). A) Representative gating of CD8+ T cells analyzed for the expression of PD-1 (x axis) and Tim-3 (y axis). Percentage of live CD8+ T cells that are B) PD-1+, C) Tim-3+, D) PD-1-Tim-3-, E) PD-1-Tim-3+, F) PD-1+Tim-3-, and G) PD-1+Tim-3+. Ordinary one-way ANOVA with uncorrected Fisher’s LSD was performed to determine statistical significance. * p ≤ 0.05, ** p ≤ 0.01.
Fig 3.
Exhaustion, but not activation markers are increased after HIV-1 transient controllers lose control.
PBMCs were isolated from blood samples of VC – control (n = 5−8) or TC – post control (n = 3) and flow cytometry was performed gating for live (Zombie Yellow-), CD8+ T cells (CD3+CD8+). Graphed are the percentages of live CD8+ T cells that are A) CD38+, B) HLA-DR+, C) CD38-HLA-DR-, D) CD38+HLA-DR-, E) CD38-HLA-DR+, F) CD38+HLA-DR+, G) PD-1+, H) Tim-3+, I) PD-1-Tim-3-, J) PD-1+Tim-3-, K) PD-1-Tim-3+, and L) PD-1+Tim-3+. Unpaired two-tailed T tests were performed for each panel to determine statistical significance. *** p ≤ 0.0005, **** p ≤ 0.0001.
Fig 4.
Clustering and tSNE dimensionality reduction of flow cytometry profiles of CD8+ T cells across participant groups.
Flow cytometry was performed on PBMC from PWoH, PC and TC during successful control (VC – control), TC post control (TC – post control), SP ART-, and SP ART + . A) Cells from all participant groups identified as live (Zombie Yellow-) T cells (CD3+) were clustered using the FlowSOM plugin in FlowJo, labeled 1-10, color coded, and displayed as a heatmap. B) The relative contribution of each cluster to the overall makeup of a given participant group was plotted as a heatmap showing the percentage of cells in each participant group from a given cluster. C) tSNE analysis and plotting was performed on the same live T cell populations using the associated function in FlowJo v10. Final plot was colored according to the clusters identified in panel A. Cells from study participants corresponding to D) PWoH, E) SP ART-, F) VC – control G) TC – post control, and H) SP ART+ were mapped back onto the tSNE plot. Circled populations are those unique to either VC – control, TC – post control, or SP ART+ and are labelled with an identifying number and “*” that corresponds to the cluster (Fig 4A) of which they are a sub-population.
Fig 5.
Expression of individual markers by unique populations as defined by tSNE.
The intensity of staining for each individual marker was mapped back onto the tSNE plot generated in Fig 4. Heatmap indicates the relative expression levels of each marker from low (blue) to high (red). The circled and labeled populations are those that were unique to VC – control (1* and 8*), TC – post control (4* and 5*), and SP ART+ (4*) and are identified according to the cluster they were assigned to in Fig 4.
Fig 6.
CD8+ T cell exhaustion levels remain altered over time on ART in HIV-1 transient controllers.
PBMCs were isolated from blood samples of PC/TC during successful control (VC – control) (n = 5-8), TC – post control < 180d ART (TC – post control, P1) (n = 1-3), TC – post control 181-365d ART (TC – post control, P2) (n = 3), TC – post control > 365d ART (TC – post control, P3) (n = 1-3), or SP ART+ (n = 4) and flow cytometry was performed gating for live (Zombie Yellow-), CD8+ T cells (CD3+CD8+). Graphed are the percentages of live CD8+ T cells that are A) PD-1-Tim-3-, B) PD-1+Tim-3-, C) PD-1-Tim-3+, D) PD-1+Tim-3+, E) PD-1-Tim-3-, F) PD-1+Tim-3-, G) PD-1-Tim-3+, and H) PD-1+Tim-3+. A-D) TC – post control represent TC who lost control in vivo and ex vivo. E-F) TC – post control represent TC who lost control in vivo but maintained ex vivo control. Ordinary one-way ANOVA with Tukey’s multiple comparisons test was performed to determine statistical significance. * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.0005, **** p ≤ 0.0001.
Fig 7.
Combination immune checkpoint blockade restores ex vivo control of viral replication in HIV-1 transient controllers – post control and CD8+ T cell exhaustion profiles are associated with cICB responsiveness.
A) CD4+ and CD8+ T cells were isolated from PBMC of TC – post control (n = 2) and EC (n = 1). CD4+ and CD8+ T cells were activated by PHA-P for 48 h and cultured alone or in a 1:1 co-culture and infected with HIV-1 NL4−3 after pre-treatment with ICB antibodies. Cultures were maintained in the given concentration of B) anti-PD-1 antibody alone, C) anti-Tim-3 antibody alone, or D) anti-Tim-3 antibody with 50 ng/mL anti-PD-1. Supernatant levels of HIV-1 p24 were determined by ELISA on days 2, 4 and 6 post infection. Ability of 50 ng/mL anti-PD-1 and 50 ng/mL anti-Tim-3 cICB to restore suppression of viral replication is shown for E) purified CD4+/CD8+ T cell co-cultures and F) total PBMCs and was determined by computing the percentage of viral suppression at peak p24 levels as compared to cICB untreated cultures. TC – post control are indicated as black symbols and the EC is indicated as a red symbol. PBMCs from blood draws performed at multiple time points following the beginning of ART in TC – post control (VQY4910 and MPY1313) and the EC (EXT1011) were examined by flow cytometry for the percentage of G) PD-1+Tim-3-, H) PD-1-Tim-3+ and I) PD-1+Tim-3+ CD8+ T cells. E, F) Paired two-tailed T tests were performed to determine statistical significance. G-I) Ordinary one-way ANOVA with Tukey’s multiple comparisons test was performed to determine statistical significance. ** p ≤ 0.01, **** p ≤ 0.0001.
Table 3.
Clinical characteristics of cICB-responsive and non-responsive participants.