Dynamic single-cell metabolomics reveals cell-cell interaction between tumor cells and macrophages

Single-cell metabolomics reveals cell heterogeneity and elucidates intracellular molecular mechanisms. However, general concentration measurement of metabolites can only provide a static delineation of metabolomics, lacking the metabolic activity information of biological pathways. Herein, we developed a universal system for dynamic metabolomics by stable isotope tracing at the single-cell level. This system comprises an organic mass cytometry-based single-cell metabolomic platform and an automated metabolite quantification and untargeted metabolic flux data processing platform, providing an integrated workflow for dynamic metabolomics of living single cells. As a proof-of-concept, a total of 40 labeled metabolites were tracked in single breast cancer cells, enabling the global profiling and correlation analysis of metabolic activities across various metabolic pathways, as well as flow analysis of interlaced metabolic networks. The significance of metabolic activity profiling was underscored by a 2-deoxyglucose inhibition model, demonstrating delicate metabolic alteration within single cells which cannot reflected by concentration analysis. Significantly, the system combined with a neural network model successfully enabled the metabolomic profiling of direct co-cultured tumor cells and macrophages. This revealed intricate cell-cell interaction mechanisms within the tumor microenvironment and firstly identified versatile polarization subtypes of tumor-associated macrophages based on their metabolic signatures. Importantly, these findings are in line with the renewed diversity atlas of macrophages from genome features obtained through single-cell RNA-sequencing. The developed system facilitates a comprehensive understanding and opens a new avenue of single-cell metabolomics from both static and dynamic perspectives.