Fig 2. Recognition of latently-infected primary CD4⁺ T-cells by virus-specific CD8⁺ T-cell clones following exposure to candidate LRAs.

HIV-, CMV- and HERV-K-specific CD8⁺ T-cell clones were derived from the HIV-infected participant OM9. Latently-infected and productively-infected target cells were prepared and characterized as in Fig 1A. The designated CD8⁺ T-cell clones were co-cultured with the indicated target CD4⁺ T-cells (autologous) immediately after the depletion of activated cells. Cells were stained and fixed after 16 hour co-cultures. A. Shown are flow cytometry data gated on CD3⁺CD8⁺ lymphocytes and depicting CD107a (degranulation)–y-axis, by IFN-γ –x-axis. Latently-infected cells did not induce CD107a exposure. B. Summary flow cytometry data of the same experiment depicted in A. P-values were calculated by ANOVA with Holm-Sidak’s multiple comparison test (comparing each condition with latent-mock). All conditions/replicates tested are shown. Latent-mock and latent-HIV conditions of MHC-I mismatch were omitted from the HIV-Nef-spec CD8⁺ T-cells due to insufficient cell numbers, as were replicates of MHC-I mismatch conditions for the HERV-K-Env-specific CD8⁺ T-cell clone. CD107a exposure by an HIV-specific CD8⁺ T-cell clone was only induced by productive HIV infection. C. In a separate experiment, latently-infected target cells were prepared in the same manner as A, rested for 72 hours, and then combined with an autologous HIV-Gag-specific CD8⁺ T-cell clone (upper panels) or an autologous CMV-pp65-specific CD8⁺ T-cell clone (lower panels) for a 24 hour co-culture period. Candidate latency-reversing drugs were added as indicated above the corresponding panels, and left in for the duration of co-cultures. Shown are flow cytometry data gated on CD3⁺CD8⁺ lymphocytes and depicting CD137 (4-1BB)–y-axis by CD8 x-axis. CD137 expression by an HIV-specific CD8⁺ T-cell clone was induced following treatment with IL-15, IL-2, and prostratin, but not SAHA.