Single-cell transcriptome analysis reveals heterogeneity and convergence of the tumor microenvironment in colorectal cancer

IntroductionColorectal cancer (CRC) ranks second for mortality and third for morbidity among the most commonly diagnosed cancers worldwide. We aimed to investigate the heterogeneity and convergence of tumor microenvironment (TME) in CRC. MethodsWe analyzed the single-cell RNA sequencing data obtained from the Gene Expression Omnibus (GEO) database and identified 8 major cell types and 25 subgroups derived from tumor, para-tumor and peripheral blood. ResultsIn this study, we found that there were significant differences in metabolic patterns, immunophenotypes and transcription factor (TF) regulatory patterns among different subgroups of each major cell type. However, subgroups manifested similar lipid metabolic patterns, immunosuppressive functions and TFs module at the end of the differentiation trajectory in CD8+ T cells, myeloid cells and Fibroblasts. Meanwhile, TFs regulated lipid metabolism and immunosuppressive ligand-receptor pairs were detected by tracing the differentiation trajectory. Based on the cell subgroup fractions calculated by CIBERSORTx and bulk RNA-sequencing data from The Cancer Genome Atlas (TCGA), we constructed an immune risk model and clinical risk model of CRC which presented excellent prognostic value. ConclusionThis study identified that the differentiation was accompanied by remodeling of lipid metabolism and suppression of immune function, which suggest that lipid remodeling may be an important trigger of immunosuppression. More importantly, our work provides a new perspective for understanding the heterogeneity and convergence of the TME and will aid the development of prognosis and immunotherapies of CRC patients.

Automated summary

This study analyzed single-cell RNA sequencing data to investigate the heterogeneity and convergence of the tumor microenvironment (TME) in colorectal cancer (CRC). Significant differences were found in metabolic patterns, immunophenotypes, and transcription factor regulatory patterns among different subgroups of major cell types. The study also identified similarities in lipid metabolic patterns, immunosuppressive functions, and transcription factor modules at the end of the differentiation trajectory in CD8+ T cells, myeloid cells, and fibroblasts. Additionally, an immune risk model and clinical risk model of CRC were constructed, which showed excellent prognostic value. The findings suggest that lipid remodeling may contribute to immunosuppression and provide new insights into the heterogeneity and convergence of the TME for the development of prognosis and immunotherapies in CRC patients.