Combining Bioinformatics and Experiments to Identify and Verify Key Genes with Prognostic Values in Endometrial Carcinoma

Endometrial carcinoma(EC) is the most common cancer of female reproductive system, thus requiring for new effective biomarkers which could predict the onset of EC and poor prognosis. Our study integrated two GEO datasets(i.e.GSE63678, GSE17025) and TCGA(The Cancer Genome Atlas ) UCEC data to screen out 344 common differentially expressed genes(DEGs), which were further analyzed by GO(gene ontology) functions and KEGG(Kyoto Encyclopedia of Gene and Genome) pathways. KEGG analysis results showed these DEGs were mainly enriched in cell cycle, oocyte meiosis, cellular senescence, carbon metabolism and p53 signaling pathway. Top 20 hub genes with higher degree were selected from PPI(protein-protein interaction) network and 15 of them were associated with the prognosis of EC, that is, CCNB2, CDC20, BUB1B, UBE2C, AURKB, FOXM1, NCAPG, RRM2, TPX2, DLGAP5, CDCA8, CDC45, MKI67, BUB1, KIF2C. UBE2C(Ubiquitin Conjugating Enzyme E2 C) was chosen for further validation in TCGA cohort on mRNA level and in our patient samples on protein level by immunohistochemistry. UBE2C was significantly highly expressed in endometrial carcinoma, and its expression level was associated with advanced FIGO staging and poor prognosis. Cox risk model demonstrated high UBE2C expression was an independent risk factor. Somatic mutations, elevated copy number, DNA hypomethylation all contributed to its overexpression. Therefore, by combination of bioinformatics and experiment, our study provided a unique insight into the pathogenesis and molecular mechanisms underlying EC and discovered new biomarkers for early diagnosis and prognostic prediction. UBE2C could serve as a potential marker to predict poor prognosis and as a therapeutic target.