Abstract
To identify the cell types that are infected with KSHV and the immune interactions in Kaposi sarcoma (KS) lesions, we performed spatial transcriptomics with seven KS skin tumors. We used a single-cell RNA-sequencing reference data set from healthy skin donors with a method to conduct spatially informed cell-type deconvolution for spatial transcriptomics. This allowed us to predict the relative amounts of each cell type within the patient sample sections. We included custom probes for five KSHV genes that allowed us to measure human and KSHV expression patterns at the same time. We then compared the spatial gene expression data of KS skin samples with six normal skin samples and found higher expression of marker genes corresponding to macrophages/dendritic cells, lymphatic endothelial cells, and vascular endothelial cells in the KS skin lesions when compared to normal skin samples. Our spatial transcriptomic results from thousands of spots across multiple KS tumors indicated a correlation between high levels of STC1 and decreased expression of macrophage markers. Together, these analyses offer potential mechanisms by which KSHV infection may remodel skin tissue, inhibit immune responses against KSHV infection, and confer resistance to anticancer therapies.