Exploration of propionate metabolism-related genes to predict prognosis and immunotherapy response in ovarian cancer.

Background: Related studies have shown that propionate metabolism-related genes (PMRGs) were associated with the progress of cancers. However, the roles of PMRGs in ovarian cancer (OC) were unclear.

Methods: In this study, OC-related transcriptome data and clinical information were extracted from The Cancer Genome Atlas (TCGA),Genotype-Tissue Expression (GTEx) and Gene Expression Omnibus (GEO) databases. Firstly, the differentially expressed genes (DEGs) between OC and healthy control (HC) samples were screened by differential expression analysis. Then, the differentially expressed PMRGs (DE-PMRGs) were obtained by intersecting the DEGs with PMRGs. Next, the enrichment analyses of DEGs and DE-PMRGs were conducted to investigate the functions. Moreover, the biomarkers of OC were screened and the risk score was calculated. Then, the nomogram predicting the survival of OC was constructed. Furthermore, the tumor microenvironment analyses and drug sensitivity analysis were proceeded. In addition, the transcription factor (TF)-mRNA regulatory network was constructed to reveal the potential molecular-level regulation of biomarkers. Additionally, the expression levels of biomarkers in IOSE-80, OVCA429, hey and OVCAR-8 were detected through the Quantitative Real-time Polymerase Chain Reaction (qRT-PCR). Immunohistochemistry (IHC) was performed to validate the protein expression of key biomarkers (CETP, ALDH5A1, and PTH) in ovarian cancer tissue microarrays.

Results: Totals of 280 DE-PMRGs were obtained by intersecting the 9,466 DEGs and 531 PMRGs, and these genes were associated with steroid and fatty acid metabolic process. Five biomarkers (ALDH5A1, CETP, GRIA1, PTH, and TPMT) were identified, and the nomogram was constructed with risk score, age and TMB. Among them, GRIA1 was a negative factor, while age and risk score were negatively associated with patients' survival. Noticeable, the tumor purity was low and the level of immune escape was high in OC groups. Besides, AKT.inhibitor.VIII,A.443654,LFM.A13,BMS.509744 and BMS.536924 were positively associated with the risk score. Furthermore, the TF-mRNA regulatory network of OC was constructed, among them, EGR1 was the key TF which could regulate ALDH5A1 and TPMT simultaneously. The qRT-PCR proved the up-regulated expression levels of ALDH5A1, CETP, PTH and TPMT in OVCA429, hey and OVCAR-8. IHC results confirmed significantly higher protein expression of CETP, ALDH5A1, and PTH in ovarian cancer tissues compared to normal controls (p < 0.05), further validating their roles as potential prognostic biomarkers.

Conclusion: This study identified 5 biomarkers associated with the prognosis of OC, which might be helpful in understanding the roles of PMRGs in the development of OC in depth. The IHC validation provided additional evidence at the protein level, reinforcing the clinical relevance of these findings.