Abstract
BACKGROUND: Targeted therapies have improved the clinical outcomes of most patients with cancer. However, the heterogeneity of gastric cancer remains a major hurdle for precision treatment. Further investigations into tumor microenvironment heterogeneity are required to resolve these problems. METHODS: In this study, bioinformatic analyses, including metabolism analysis, pathway enrichment, differentiation trajectory inference, regulatory network construction, and survival analysis, were applied to gain a comprehensive understanding of tumor microenvironment biology within gastric cancer using single-cell RNA-seq and public datasets and experiments were carried out to confirm the conclusions of these analyses. RESULTS: We profiled heterogeneous single-cell atlases and identified eight cell populations with differential expression patterns. We identified two cancer-associated fibroblasts (CAFs) subtypes, with particular emphasis on the role of inflammatory cancer-associated fibroblasts (iCAFs) in EMT and lipid metabolic crosstalk within the tumor microenvironment. Notably, we detected two differentiation states of iCAFs that existed in different tissues with discrepant expression of genes involved in immuno-inflammation or ECM remodeling. Moreover, investigation of tumor-infiltrating myeloid cells has revealed the functional diversity of myeloid cell lineages in gastric cancer. Of which a proliferative cell lineage named C1QC(+)MKI67(+)TAMs was recognized with high immunosuppressive capacities, suggesting it has immune suppression and cell proliferation functions in the tumor niche. Finally, we explored regulatory networks based on ligand-receptor pairs and found crucial pro-tumor crosstalk between CAFs and myeloid cells in the tumor microenvironment (TME). CONCLUSION: These findings provide insights for future cancer treatments and drug discovery.