Construction of a novel choline metabolism-related signature

IntroductionColon adenocarcinoma (COAD) is a lethal cancer and is the second leading cause of cancer deaths worldwide (1). Currently, the standard treatment modalities for COAD mainly include surgery, chemotherapy, radiotherapy, targeted drugs and immunotherapy. Despite significant advances in these treatment modalities, the 5-year survival of COAD patients remains unsatisfactory (2). Therefore, finding a novel and reliable biomarker to improve the diagnostic accuracy and treatment effect is of great significance for improving the prognosis of COAD patients.In recent years, increasing evidence has shown that abnormal choline metabolism is accompanied by elevations in phosphocholine (PCho) and glycerophosphocholine (GPC) and may serve as a marker of metabolic reprogramming in cancer (3). Due to the active proliferation of malignant tumor cells, a large amount of choline needs to be converted into phosphatidylcholine for the synthesis of cell membranes. A study showed that an abnormal choline metabolite profile of cancer can be used as an adjunct to current methods of diagnosing primary and other tumors (4). For example, choline metabolite profiling can be used as an aid in the diagnosis of brain malignancies and prostate and breast cancers (5–7). However, the gene changes and specific mechanisms related to choline metabolism in COAD remain to be elucidated.In this study, we applied The Cancer Genome Atlas (TCGA) dataset as a training cohort to construct a choline metabolism-related signature, which was composed of two genes (CHKB and PEMT). Subsequently, we validated the prognostic value of the signature in a Gene Expression Omnibus (GEO) dataset (GSE17536). Next, we further analyzed the correlation of the choline metabolism-related signature via nomogram construction with the immune microenvironment, chemotherapy response and somatic mutations of patients. In addition, we analyzed the distribution and expression of two key genes, CHKB and PEMT, in immune cells based on single-cell RNA sequencing (scRNA-seq). We also predict potential drugs for COAD treatment based on molecular docking. Finally, we verified the expression of CHKB and PEMT in COAD and normal samples by IHC, Western blot, and qRT-PCR experiments. In short, the choline metabolism-related signature is expected to be a potential biomarker for COAD, which may provide new perspectives for the diagnosis and treatment of this disease.Materials and methodsData source and processingThe clinical information, transcriptome expression data, and gene mutation profiles of COAD patients were obtained from the TCGA database (https://cancergenome.nih.gov/). Patients with incomplete survival information were excluded. We finally included 452 COAD patients as the TCGA-COAD cohort (Supplementary Table 1). The GSE17536 dataset containing corresponding RNA expression data and clinical information was downloaded from the GEO database (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE17536). Seventy-six choline metabolism-related genes were obtained from the KEGG, AmiGO2 and Reactome Pathway Databases (Supplementary Table 2). A total of thirty differentially expressed genes (DEGs) between COAD tissue and normal tissue were screened (|log2 (fold change) |>0.5, p
Read More