Fig. 2: Metabolic profiling of colorectal cell lines with differential metastatic potential by mass spectrometry-based untargeted metabolomic analysis.

a Schematic pipeline of metabolic screening (upper) and steps for refining the potential metabolites (lower). b OPLS-DA plots of individual cells of SW480/SW620 and HT-29/COLO 205 in negative and positive ion modes (R2X(cum) = 0.705, R2Y(cum) = 0.952, Q2(cum) = 0.725 and R2X(cum) = 0.679, R2Y(cum) = 0.994, Q2(cum) = 0.987 in negative ion mode for SW480/SW620 and HT-29/COLO 205; R2X(cum) = 0.426, R2Y(cum) = 0.869, Q2(cum) = 0.785 and R2X(cum) = 0.632, R2Y(cum) = 0.996, Q2(cum) = 0.961 in positive ion mode for SW480/SW620 and HT-29/COLO 205). c Volcano plots of metabolite abundance in SW480/SW620 cells and HT-29/COLO 205 cells. The red and cyan dots represent significantly increased/decreased metabolic features (|Log₂FC | >1, p < 0.05, VIP > 1). d The intersection of the metabolites with significantly differential abundance obtained from the two pairs of cell lines (SW480 vs. SW620, HT-29 vs. COLO 205). e Heatmap of the relative abundance (Log₁₀ transformation) of the shared metabolites with differential abundance in the two pairs of cells (n = 9). f Z-score plot of 14 representative metabolites with differential abundance in SW480/SW620 cells and HT-29/COLO 205 cells (n = 9 independent experiments). Data are presented as median with interquartile range and points are colored by assigned cell type. g Metabolic network of representative altered metabolites. Some graphical elements in (a) and (g) were created with BioRender.com (accessed on 24 March 2023). UPLC-HRMS ultra-performance liquid chromatography-tandem high-resolution mass spectrometry, FC fold change, VIP variable important in projection, KEGG Kyoto Encyclopedia of Genes and Genomes, MSEA metabolite set enrichment analysis, MRM multiple reaction monitoring.