Single-cell transcriptomics reveals the landscape of intra-tumoral heterogeneity and sternness-related subpopulations in liver cancer

Hepatocellular carcinoma (HCC) is heterogeneous, rendering its current curative treatments ineffective. The emergence of single-cell genomics represents a powerful strategy in delineating the complex molecular landscapes of cancers. In this study, we demonstrated the feasibility and merit of using single-cell RNA sequencing to dissect the infra-tumoral heterogeneity and analyze the single-cell transcriptomic landscape to detect rare cell subpopulations of significance. Exploration of the inter-relationship among liver cancer stem cell markers showed two distinct major cell populations according to EPCAM expression, and the EPCAM(+) cells had upregulated expression of multiple oncogenes. We also identified a CD24(+) /CD44(+)-enriched cell subpopulation within the EPCAM(+) cells which had specific signature genes and might indicate a novel sternness-related cell subclone in HCC. Notably, knockdown of signature gene CTSE for CD24(+)/CD44(+) cells significantly reduced self renewal ability on HCC cells in vitro and the sternness-related role of CTSE was further confirmed by in vivo tumorigenicity assays in nude mice. In summary, single-cell genomics is a useful tool to delineate HCC intratumoral heterogeneity at better resolution. It can identify rare but important cell subpopulations, and may guide better precision medicine in the long run.