Abstract
BACKGROUND: Gastric cancer (GC) remains a leading cause of cancer mortality. E3 ubiquitin ligases, as central regulators of protein stability and signaling within the ubiquitin-proteasome system, have been implicated in tumor progression, but their functional roles in GC are not well established. METHODS: We integrated bioinformatics analysis of TCGA and GEO datasets, in vitro experiments (including cell proliferation, migration, and apoptosis assays), and computational modeling to identify key prognostic factors in GC. RESULTS: We established two molecular subtypes (E3GC1/E3GC2) with distinct clinical outcomes and developed a 10-gene prognostic signature. The model showed moderate predictive accuracy (AUC: 0.61-0.71) and was validated externally. MMRN1 was upregulated in GC cells and its knockdown significantly inhibited malignant phenotypes. Critically, drug sensitivity analysis revealed high-risk patients were more sensitive to proteasome inhibitors (bortezomib), while low-risk patients responded better to taxane-based chemotherapy (docetaxel). Molecular docking predicted a high-confidence interaction between MMRN1 and NEDD4L, suggesting potential ubiquitination regulation. CONCLUSIONS: MMRN1 drives GC cell proliferation and migration in vitro and may be regulated by NEDD4L-mediated ubiquitination. Our study provides a foundation for E3 ligase-based patient stratification and personalized therapy selection in GC. While this study provides comprehensive multi-omics evidence supporting the role of MMRN1 in GC progression, its clinical translation is limited by the lack of in vivo validation and direct experimental evidence of NEDD4L-MMRN1 physical interaction. Further studies using animal models and clinical specimens are warranted to confirm these findings.