Identification of novel mutations of ovarian cancer-related genes from RNA-sequencing data for Japanese epithelial ovarian cancer patients

Abstract

Ovarian cancer has the highest mortality rate among gynecological cancers. Gene mutations are involved in the carcinogenesis, metastasis, and therapeutic response in ovarian cancer. However, the variety and proportion of gene mutation is not fully analyzed in Japanese ovarian cancer patients, especially, in those with recurrent tumors. In the present study, RNAsequencing was performed for 32 clinical ovarian specimens obtained from 24 Japanese patients (24 primary cancer specimens and 8 recurrent specimens paired with corresponding primary cancer specimens). Mutations in 24 primary specimens were analyzed by comparing the sequence data mapped on RefSeq genes with those in the public online databases BRCA Exchange, COSMIC, ClinVar, and cBioportal. Mutations were observed in TP53 in 16 specimens (67%), BRCA1 in 9 (38%), BRCA2 in 13 (54%), ARID1A in 3 (13%), PIK3CA in 2 (8%), KRAS in 1 (4%), PTEN in 1 (4%), and CTNNB1 in 1 (4%), excluding synonymous mutations. Among those identified muations, 13 of 14 mutations in TP53, 10 of 11 mutations of BRCA1, 10 of 23 mutation positions of BRCA2, none of 7 mutations of ARID1A, 1 mutation of PIK3CA, and 1 mutation of CTNNB1 were consistent with those reported in the public online databases; however, the other mutations identified were novel. Comparison between matched-paired specimens of primary and recurrent tumors revealed the changes of mutational status in expressed RNAs. RNA-sequencing-based mutation analysis will be useful to reveal ethnic differences of gene mutations in ovarian cancer and to understand the contribution of gene mutations to recurrence.

Long Intergenic Noncoding RNA OIN1 Promotes Ovarian Cancer Growth by Modulating Apoptosis-Related Gene Expression

“…In the present study, we aimed to identify functional lincRNAs predominantly expressed in ovarian cancer. We screened our RNA-seq data analyzed by NONCODE database [ 18 ] obtained from clinical specimens of normal ovarian tissues ( n = 6) and ovarian cancers ( n = 15) [ 15 , 16 ]. Screening NONCODE v4 transcripts abundantly expressed in ovarian cancer compared with normal ovarian tissues by ≥10-folds at an FDR q -value threshold <0.05, we identified 10 particular lincRNAs ( Table 1 ), including known oncogenic RNAs: competing endogenous lncRNA 2 for microRNA let-7b ( CERNA2 )/ human ovarian cancer-specific transcript 2 (HOST2) [ 19 ], urothelial cancer associated 1 ( UCA1 ) [ 14 , 20 , 21 , 22 ], and long intergenic non-protein coding RNA 958 (LINC00958) [ 23 ].…”

“…We next screened OIN1 -associated genes whose expression levels exhibited positive or negative correlations with OIN1 in the RNA-seq data from ovarian cancer specimens [ 15 , 16 ]. We found that 323 and 312 genes had correlation coefficients of ≥0.6 and ≤−0.45 with OIN1 , respectively, and analyzed enriched biological pathways in these identified genes using the DAVID Bioinformatics Resources 6.8 [ 24 ] ( Tables S1 and S2 ).…”

“…Ovarian cancer and normal tissue specimens were obtained from patients who underwent surgery for primary ovarian tumor. Detailed information for the clinical specimens and RNA-seq methods were described previously [ 15 , 16 ]. Expression values of RNA-seq were quantified as RPKM based on RefSeq and NONCODE ( [accessed on 22 January 2021]) gene models [ 18 ].…”

“…( A ) OIN1 RNA (registered as NONHSAT013448 in NONCODE) is overexpressed in ovarian cancer. RPKM (reads per kilobase of transcript length per million mapped reads) values of OIN1 RNA were estimated by RNA-seq analysis using ovarian cancer ( n = 15) and normal tissues ( n = 6) [ 15 , 16 ]. Data are presented by box plots.…”

“…We previously performed RNA-sequencing (RNA-seq) analysis using clinical ovarian tissues and cancer specimens and identified candidate genes associated with the pathophysiology of ovarian cancer [ 15 , 16 ]. We also identified novel mutations in known ovarian cancer-associated genes, such as TP53 , breast cancer 2 ( BRCA2 ), and AT-rich interaction domain 1A ( ARID1A ) [ 16 ]. These results indicate that RNA-seq analysis using clinical samples is useful for the characterization of ovarian cancer gene signature and the screen of new therapeutic targets.…”