Background: Colorectal cancer (CRC) is a common global malignancy associated with high invasiveness, high metastasis, and poor prognosis. CRC commonly metastasizes to the liver, where the treatment of metastasis is both difficult and an important topic in current CRC management.Methods: Microarrays data of human CRC with liver metastasis (CRCLM) were downloaded from the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus (GEO) database to identify potential key genes. Differentially expressed (DE) genes (DEGs) and DEmiRNAs of primary CRC tumor tissues and metastatic liver tissues were identified. Microenvironment Cell Populations (MCP)-counter was used to estimate the abundance of immune cells in the tumor micro-environment (TME), and weighted gene correlation network analysis (WGCNA) was used to construct the co-expression network analysis.Gene Ontology and Kyoto Encyclopaedia of Gene and Genome (KEGG) pathway enrichment analyses were conducted, and the protein-protein interaction (PPI) network for the DEGs were constructed and gene modules were screened.Results: Thirty-five pairs of matched colorectal primary cancer and liver metastatic gene expression profiles were screened, and 610 DEGs (265 up-regulated and 345 down-regulated) and 284 DEmiRNAs were identified. The DEGs were mainly enriched in the complement and coagulation cascade pathways and renin secretion. Immune infiltrating cells including neutrophils, monocytic lineage, and cancer-associated fibroblasts (CAFs) differed significantly between primary tumor tissues and metastatic liver tissues. WGCN analysis obtained 12 modules and identified 62 genes with significant interactions which were mainly related to complement and coagulation cascade and the focal adhesion pathway. The best subset regression analysis and backward stepwise regression analysis were performed, and eight genes were determined, including F10, FGG, KNG1, MBL2, PROC, SERPINA1, CAV1, and SPP1. Further analysis showed four genes, including FGG, KNG1, CAV1, and SPP1 were significantly associated with CRCLM.Conclusions: Our study implies complement and coagulation cascade and the focal adhesion pathway play a significant role in the development and progression of CRCLM, and FGG, KNG1, CAV1, and SPP1 may be metastatic markers for its early diagnosis.

Background: Globally, the incidence and mortality of colorectal cancer (CRC) rank amongst the highest of all malignancies. Thus, research aimed at developing new screening strategies and biomarkers for the early detection of CRC is needed. At present, conventional screening methods have limitations; therefore, new testing strategies have been considered. Using metabolomics to explore the molecular changes in CRC tissue is a mainstream method for identifying potential biomarkers and key cancer factors.Methods: In the present study, 27 samples from nine CRC patients were used to analyze the metabolite differences between the tumor, paracancerous, and normal tissues. The metabolite differences in the various stages of CRC (stages IIA, IIB, and IIIC) were analyzed as well. Subsequently, principal component analysis (PCA), permutation, and trend analyses were performed. Weighted gene co-expression and metabolitemetabolite interaction networks were also constructed.Results: A total of 5,834 metabolites were identified among the included samples. Permutation analysis showed a clear separation between the different tissues and different stages. Compared with normal tissues, tumor tissues exhibited 11, 233, and 25 up-regulated metabolites as well as one, 77, and zero down-regulated metabolites in stages IIA, IIB, and IIIC, respectively. Moreover, tumor tissues in stage IIB exhibited more differential metabolites (233 up-regulated and 77 down-regulated). Weighted Gene Correlation Network Analysis (WGCNA) clustered the 5,834 metabolites into 15 different modules, of which four modules were significantly correlated with tissue specificity. Notably, glycerophospholipid metabolism, fatty acid metabolism, and other pathways were enriched in these modules.Conclusions: Fatty acids and glycerophospholipids were significantly related to the development of CRC. This result is of great significance for future targeted screening of CRC biomarkers and further clarifying the nutrient metabolism of cancer cells.