Abstract
BACKGROUND: Colorectal cancer (CRC) ranks as the second leading cause of cancer-related mortality worldwide. Liquid-liquid phase separation (LLPS) is a phenomenon driven by multivalent weak interactions among biomolecules, such as proteins and nucleic acids, which results in the formation of biomolecular condensates. Abnormal LLPS is closely linked to tumorigenesis and progression. However, a prognostic signature based on LLPS-related genes (LRGs) has rarely been established in CRC. METHODS: We retrieved CRC data from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. LRGs were obtained from the DrLLPS database. A novel signature was developed in the TCGA training cohort using univariate Cox regression, least absolute shrinkage and selection operator (LASSO) Cox regression and multivariate Cox regression. Subsequently, a risk signature termed the LLPS-related risk score (LRS) was introduced, categorizing CRC patients into high-risk and low-risk groups based on their LRS. Furthermore, the efficacy of the risk signature was evaluated using principal component analysis (PCA), Kaplan-Meier survival, receiver operating characteristic (ROC), nomogram, and concordance index (C-index). Additionally, the differences in immune cell infiltration, immune function, the tumor microenvironment (TME), drug sensitivity, and tumor mutation burden (TMB) between these risk subgroups were analyzed. Importantly, experimental investigations were conducted to assess the expression levels and LLPS capabilities of ARF Like GTPase 6 Interacting Protein 4 (ARL6IP4)in CRC. RESULTS: We developed an eight-LRG prognostic risk signature based on LRGs and confirmed its role as an independent prognostic factor for patients with CRC. Additionally, we observed significant differences in immune cell infiltration, immune function, and TME between the low-risk and high-risk groups. Furthermore, the low-risk group demonstrated a lower Tumor Immune Dysfunction and Exclusion (TIDE) score and exhibited greater sensitivity to certain clinical therapeutic agents. Moreover, ARL6IP4 was found to be upregulated in CRC and underwent LLPS in live cells. CONCLUSION: We constructed and validated an eight-LRG risk signature, which can be employed to predict prognosis, characterize the immune landscape, and assess drug sensitivity in CRC.