针对肿瘤微环境的单细胞空间CRISPR筛选

The tumor microenvironment (TME) comprises diverse cell types, including mesenchymal, endothelial, adipocyte, stromal, and immune cells. Among them, cancer-associated fibroblasts (CAFs) represent one of the most abundant and functionally active components. CAFs play a crucial role in tumor progression through bidirectional communication with cancer cells. They originate from various sources, such as normal fibroblasts, endothelial cells, and mesenchymal cells, and once activated, promote tumor cell invasion and metastasis. Recent studies have revealed that CAFs consist of multiple subpopulations with distinct phenotypic and functional properties. This heterogeneity of CAFs has provided new insights into tumor biology and has become a key focus in the development of novel targeted therapeutic strategies across different cancer types. CRISPR/Cas9 is a powerful genome-editing tool widely used for knock-out (KO) gene studies to investigate gene function. To overcome the limitations of conventional phenotyping and bulk analysis, a barcoding system known as Perturb-map was developed by the Brown laboratory that allows KO of genes in tumor cells and facilitates identification of effects of the KO on the TME. Perturb-map utilizes triplet combinations of linear epitope protein barcodes (Pro-codes) that enable the identification of cells expressing distinct CRISPR guide RNAs (gRNAs). In this study, we applied the Perturb-map approach to perform parallel CRISPR KO of 34 genes closely associated with CAF function in the TME in a syngeneic mouse breast cancer model. This approach allowed us to simultaneously assess the functional roles of multiple TME-related genes in tumor cells, providing a comprehensive understanding of their contributions to tumor progression. Pro-code-expressing tumors were analyzed using cyclic immunofluorescence (CycIF), a highly multiplexed proteomics imaging platform that enables spatial and single-cell level analysis. We developed, validated, and applied mouse antibody panels targeting approximately 100 proteins, allowing comprehensive profiling of tumor heterogeneity, cellular states, fibroblast, and immune cell activities. Through this approach, we identified potential therapeutic targets in tumor cells that confer growth advantages and contribute to remodeling of the TME. Furthermore, we identified effects of the KOs on CAF subtypes with distinct functional states. CD274 and IL11Ralpha1 KO tumors exhibited accelerated tumor growth accompanied by an increased abundance of myofibroblastic CAFs (myCAFs). In contrast, Snai2 KO tumors showed a marked enrichment of inflammatory CAFs (iCAFs). Our findings demonstrate the power of integrating functional genomics with high-dimensional proteomics to characterize TME dynamics at single-cell resolution.