Identify GDPD3 as a key regulator of epithelial-mesenchymal transition and prostate adenocarcinoma progression via...

Prostate adenocarcinoma (PRAD) is a common malignancy with marked clinical heterogeneity, complicating prognosis and disease monitoring. Traditional tools like the Gleason score lack molecular and microenvironmental insights, underscoring the need for biomarker-driven predictive models.

Single-cell RNA-seq data from GEO and bulk RNA-seq data from TCGA were analyzed. scRNA-seq processing used the Seurat package, with cluster-specific genes identified via FindAllMarkers. Differentially expressed genes (DEGs) from bulk data were obtained using limma, and key gene modules were identified through WGCNA. Using univariate Cox regression and LASSO analysis, a prognostic model was developed based on cluster-specific genes, key module genes, and differentially expressed genes. Clinical validation included comparison of tumor and adjacent normal tissues, revealing significantly elevated GDPD3 expression, further confirmed by immunohistochemistry. In vitro knockdown experiments were conducted in DU145 cells to assess GDPD3's role in promoting proliferation, migration, and epithelial-mesenchymal transition (EMT).

In this study, through integrated single-cell sequencing and Bulk-RNA-seq analyses, we established a 21-gene prognostic model. QPCR confirmed significant upregulation of three candidates, including GDPD3, which was also elevatedin malignant tissues. Knockdown of GDPD3 inhibited tumor cell proliferation, invasion, and migration. Mechanistically, GDPD3 regulated the levels of lysophosphatidic acid (LPA), which in turn induced EMT in tumor cells. Inhibition or knockdown of the LPA receptor LPAR1 suppressed EMT. LPA promoted EMT through activation of the AKT signaling pathway, and inhibition of this pathway reversed LPA-induced EMT.

This study underscores key molecular mechanisms underlying prostate cancer progression, with GDPD3 emerging as a potential therapeutic target.