FIG 8.

The T_EM subset shows the greatest contribution to latency reversal from different classes of LRAs. (a) The effect of LRAs on the distribution of the memory CD4⁺ T cell subsets was assessed and is represented by the percentage of cells in each subset in the CD45RA⁻ population (bars indicate SEM; asterisks indicate a P value of <0.05 in a comparison of the drug with unstimulated controls in each memory subset, tested by a paired Wilcoxon rank sum test). (b) The contribution of the T_CM, T_TM, and T_EM subsets to the overall CD4⁻ Gag⁺ signal for each LRA was determined. Each pie slice was calculated using the frequency of cells in each memory subset from the CD4⁻ Gag⁺ population. Frequency is indicated within each pie slice. n = 10. (c) Correlation between the percent change in CD4⁻ Gag⁺ cells after LRA stimulation and the percent change in cells in the T_EM, T_TM, and T_CM subsets after exposure to LRAs. Each circle represents an independent donor after administration of anti-CD3 CD28 (yellow), bryostatin (brown), disulfiram (gray), IL-15 (orange), panobinostat (green), romidepsin (blue), or SAHA (olive). P values of Spearman’s correlation test are indicated. (d) Genes of the HDAC gene sets expressed at day 14 in LARA in vitro culture in T_CM and T_EM cells correlated with the percentage of CD4⁻ Gag⁺ cells measured postreactivation in T_CM and T_EM cells by anti-CD3 and CD28 antibody stimulation. The enrichment of gene sets was tested by GSEA using the PID HDAC pathway gene sets and custom gene sets as the database (enrichment P value < 0.05). The genes are represented on the x axis, and the Pearson correlation coefficient of each gene to the HIV msRNA is represented on the y axis. The size of the dots represents the log₂ fold change of the gene in T_EM cells compared to that in T_CM cells.