Integration of bulk RNA-seq data with single-cell RNA-seq data to identify urinary biomarkers derived from prostate tumor tissues

Background Prostate cancer (PCa) tissue biopsy has potential complications, and various methods guide biopsy decision. However, these methods have limitations. Urine-derived RNA shows promise for predicting and distinguishing PCa or clinically significant PCa (csPCa). The study aims to screen for prostate tumor cell-derived mRNAs in urine samples. Methods Strand-specific capture RNA sequencing (RNA-seq) was performed on five paired tumor and adjacent normal prostate tissues from PCa patients, as well as ten urine samples derived from PCa patients and age-matched cancer-free healthy men. By integrating these transcriptomic data with single-cell RNA sequencing (scRNA-seq) datasets of three normal prostate tissues and 13 PCa tissues, we systematically screened and identified potential urinary biomarkers for PCa diagnosis. To further explore the diagnostic efficacy of the screened urinary biomarkers for csPCa, quantitative real-time polymerase chain reaction (qPCR) was conducted on 63 urine samples, including 33 samples from PCa patients and 30 samples from age-matched cancer-free healthy men. Additionally, in vitro gene knockdown experiments were implemented to preliminarily explore the biological functions of the identified urinary biomarkers. Results Integrated analysis of strand-specific capture bulk RNA-seq and single-cell RNA-seq datasets identified urine-derived MARCKSL1 as a highly specific candidate biomarker for PCa diagnosis. In vitro functional assays demonstrated that genetic knockdown of MARCKSL1 robustly suppresses the migratory and invasive capacities of PCa cells. Furthermore, qPCR profiling verified that urine-derived MARCKSL1 exhibits favorable diagnostic performance, allowing accurate stratification of patients with PCa and csPCa. Conclusions Urine-derived MARCKSL1 has the potential to facilitate pre-biopsy diagnosis and risk stratification of PCa and csPCa. Our findings may advance the development of non-invasive tools for pre-biopsy PCa risk evaluation and csPCa discrimination.