INTRODUCTION

Castration-resistant prostate cancer (CRPC) poses a major clinical challenge due to its aggressive nature, therapeutic resistance, and frequent progression to bone metastasis. Emerging evidence suggests that E3 ubiquitin ligases play a key role in cancer progression, yet their involvement in CRPC remains poorly understood.

METHODS AND RESULTS

In this study, we employed machine learning (ML) algorithms and integrative bioinformatics approaches to analyze the gene expression profiles of E3 ubiquitin ligases in CRPC patients. Our ML analysis identified FBXO5, TRIM52, PDZRN4, CCNF, TRIM59, and SHPRH as key predictor variables in CRPC. Furthermore, differential expression analysis revealed that FBXO5, TRIM52, CCNF, and PDZRN4 are significantly dysregulated in patient samples exhibiting strong concordance with the ML results. Notably, FBXO5 exhibits robust interactive, survival, and gene expression correlation with BARD1, BRCA1, UHRF1, and SKP2; key regulators of genomic instability and drug resistance in CRPC. Additionally, FBXO5 also exhibits correlative association with CXCR4, a key mediator of prostate cancer bone metastasis. To experimentally validate these in-silico results, we performed functional assays, including cell proliferation, wound healing, and flow cytometry-based cell cycle analysis in FBXO5-knockdown prostate cancer cells. Our results demonstrate that FBXO5 depletion significantly impairs cell proliferation and migration while inducing G1-phase arrest, supporting its oncogenic role in CRPC.

CONCLUSION

Taken together, our findings establish FBXO5 as a critical mediator of CRPC progression and bone metastatic potential, underscoring its importance as a promising therapeutic target for this aggressive disease.