Fig. 1: Stable and distinct CD8⁺ PD-1⁺ and CD57⁺ Tex populations share increased chromatin accessibility of exhaustion and cytotoxicity genes.

A Schematic depicting the number of alefacept-treated donors and timepoints that were sampled for ATAC-seq (n = 4 R, R, Visit 0 = baseline, Visit 30 = 104 wk post-treatment) and the cell sorting scheme used to isolate nonnaïve CD8⁺ T cell populations into either TIGIT^-KLRG1^- double negative (DN), or TIGIT⁺KLRG1⁺ (DP) CD57⁺ Tex or PD-1⁺ Tex (24 samples total; PD-1⁺ Tex sorted as TIGIT⁺KLRG1⁺ CD57^- population based on previous work¹⁰, see methods). B Heatmap clustering of all consensus sites per sample (columns). Color blocks indicate sample cell population, donor, and timepoint. C Significance (-log10 FDR-adjusted p value) and log2 fold change of all differentially accessible peaks (FDR p value ≤ 0.05) between CD57⁺ Tex vs DN and PD-1⁺ Tex vs DN. Key genes related to exhaustion (bold) and cytotoxicity (italicized) are labeled. Where multiple peaks were annotated to the same gene, only the peak with the greatest log2 fold change was plotted. Shared genes across comparisons are highlighted in purple. NS not significant. Labeled genes that are below the cutoffs are still regulated in the same direction (i.e., the same side of the volcano), but their regulation is weaker. D Observed gene count and FDR-adjusted p value of KEGG pathways with shared enrichment in both CD57⁺ Tex and PD-1⁺ Tex populations versus DN samples.