Integrative analysis identifies a glycosylation-related lncRNA signature associated with prognosis in kidney renal clear cell carcinoma.

Abstract

BACKGROUND: Glycosylation and long non-coding RNAs (lncRNAs) play critical roles in tumor progression. However, the prognostic significance of glycosylation-related lncRNAs (GRLncs) in kidney renal clear cell carcinoma (KIRC) remains largely unclear. This study aimed to identify prognostic GRLncs and construct a predictive model for KIRC prognosis. METHODS: Transcriptomic and clinical data of KIRC patients were analyzed to identify GRLncs associated with overall survival (OS). A prognostic model was constructed based on selected GRLncs, and its predictive performance was evaluated using Kaplan-Meier (KM) survival analysis, receiver operating characteristic (ROC) curves, and univariate and multivariate Cox regression analyses. Patients were stratified into high- and low-risk groups according to the median risk score, and internal validation was performed using training and testing cohorts to assess the stability of the model. Tumor microenvironment characteristics, immune checkpoint expression, immunotherapy response, and drug sensitivity were further analyzed. In addition, the expression of three signature lncRNAs was validated by real-time quantitative polymerase chain reaction (RT-qPCR) in 10 paired KIRC tumor and adjacent normal tissues. Functional roles of selected lncRNAs were investigated using antisense oligonucleotides (ASOs)-mediated knockdown in KIRC cell lines, followed by Cell Counting Kit 8 (CCK-8), 5-ethynyl-2'-deoxyuridine (EdU) incorporation, colony formation, and migration assays. RESULTS: Five GRLncs (AC093278.2, EPB41L4A-DT, DLGAP1-AS2, AC084876.1, and AC005261.3) were identified and used to construct a prognostic model. AC093278.2 and EPB41L4A-DT were protective factors, whereas DLGAP1-AS2, AC084876.1, and AC005261.3 were risk factors. KM analysis on GRLncs-based risk score stratification revealed patients in the high-risk group had significantly poorer OS than those in the low-risk group. ROC analysis and Cox regression demonstrated that the GRLnc-based risk score served as an independent predictor of KIRC prognosis and exhibited favorable predictive performance compared with conventional clinical variables. High- and low-risk groups also exhibited distinct immune microenvironment characteristics, immune checkpoint expression patterns, and predicted drug sensitivities. RT-qPCR detected significant downregulation of protective factor-EPB41L4A-DT in KIRC tissues, while risk factors-DLGAP1-AS2 and AC084876.1 showed expression trends consistent with their predicted risk attributes. Functional experiments further revealed that knockdown of DLGAP1-AS2 and AC084876.1 suppressed proliferation and migration of KIRC cells, whereas knockdown of EPB41L4A-DT promoted these processes, supporting the biological relevance of these three signature lncRNAs. CONCLUSIONS: This study establishes a novel prognostic model based on five GRLncs that showed promising performance in The Cancer Genome Atlas (TCGA)-based analyses of KIRC. The combined clinical expression analysis and functional validation of three constituent GRLncs (DLGAP1-AS2, EPB41L4A-DT, and AC084876.1) supports the biological plausibility of the model and suggest that GRLncs may serve as potential prognostic biomarkers and therapeutic targets for KIRC.