The signature of a T-cell response to KSHV persists across space and time in individuals with epidemic and endemic KS from Uganda

Abstract

Kaposi sarcoma-associated herpesvirus (KSHV) is the etiologic agent of Kaposi sarcoma (KS), which causes significant morbidity and mortality worldwide, particularly in people living with HIV (PLWH) and in sub-Saharan Africa where KSHV seroprevalence is high. Postulating that T-cells specific for KSHV and HIV would be attracted to KS tumors, we performed transcriptional profiling and T-cell receptor (TCR) repertoire analysis of tumor biopsies from 144 Ugandan adults with KS, 106 of whom were also living with HIV. We show that CD8+ T-cells and M2-polarized macrophages are the most common immune cells in KS tumors. The TCR repertoire of T-cells associated with KS tumors is shared across spatially and temporally distinct tumors from the same individual. Clusters of T-cells with predicted shared specificity for uncharacterized antigens, potentially encoded by KSHV or HIV, comprise [~]25% of the T-cells in KS tumors. Single-cell RNA-sequencing of blood from a subset of 9 adults captured 4,283 unique {beta} TCRs carried in 14,698 putative KSHV- or HIV-specific T-cells, which carried an antigen-experienced effector phenotype. T-cells engineered to express a representative sample of these TCRs showed high-avidity recognition of KSHV- or HIV-encoded antigens. These results suggest that a polyspecific, high-avidity KSHV- and HIV-specific T-cell response, potentially inhibited by M2 macrophages, migrates to and localizes with KS tumors. Further analysis of KSHV- and HIV-specific T-cells in KS tumors will provide insight into the pathogenesis of KS and could guide the development of specific immune therapy based on adoptive transfer or vaccination. Author SummaryIn this work, we set out to examine Kaposi Sarcoma (KS) tumor tissue, as well as peripheral blood cells from individuals with KS, both living with and without HIV. Our goal was to identify T-cells that specifically recognize antigens encoded by Kaposi sarcoma-associated herpesvirus (KSHV) or HIV. By analyzing the T-cell repertoire in KS tumor biopsies from people in Uganda with different types of KS, we uncovered clusters of T-cells with previously unknown ability to recognize these viruses. Through single-cell sequencing of peripheral blood cells, we also observed that many of these T-cells had cell-killing properties. Notably, they often coexisted with a subset of macrophages with immunosuppressive properties, which we suspect may be suppressing the function of virus-targeting T-cells. Our findings suggest that additional studies of these virus-targeting T-cells and their interaction with immunosuppressive macrophages could significantly advance the development of effective therapeutics against KS.