Abstract
Ovarian clear cell adenocarcinoma (Ov-CCA) has a higher prevalence in the Japanese ancestry than other populations. The ancestral disparities in Ov-CCA prevalence suggests the presence of Ov-CCA-specific genetic alterations and may provide an opportunity to identify the novel genes associated with Ov-CCA tumorigenesis. Using 94 previously reported genes as the phenotypic trait, we conducted multistep expression quantitative trait loci (eQTL) analysis with the HapMap3 project datasets. Four single-nucleotide polymorphisms (SNPs) (rs4873815, rs12976454, rs11136002, and rs13259097) that had different allele frequencies in the Japanese ancestry and seven genes associated in cis (APBA3, C8orf58, KIAA1967, NAPRT1, RHOBTB2, TNFRSF10B, and ZNF707) were identified. In silico functional annotation analysis and in vitro promoter assay validated the regulatory effect of rs4873815-TT on ZNF707 and rs11136002-TT on TNFRSF10B. Furthermore, ZNF707 was highly expressed in Ov-CCA and had a negative prognostic value in disease recurrence in our sample cohort. This prognostic power was consistently observed in The Cancer Genome Atlas (TCGA) clear cell renal cell carcinoma dataset, suggesting that ZNF707 may have prognostic value in clear cell histology regardless of tissue origin. In conclusion, rs4873815-TT/ZNF707 may have clinical significance in the prognosis and tumorigenesis of Ov-CCA, which may be more relevant to clear cell histology. Besides, this study may underpin the evidence that cis-eQTL analysis based on ancestral disparities can facilitate the discovery of causal genetic alterations in complex diseases, such as cancer.
Highlights
Gene expression levels can be considered as quantitative traits, and specific variants associated with transcript levels are referred to as expression quantitative trait loci. eQTL analysis is a well-established in silico approach that provides evidence of the functional effects of specific loci on gene expression and identifies the novel candidate genes at the risk loci [1,2,3].Identifying causal variants located outside of the protein coding region is an ingenious and perceptive way to discover the genetic landscape of complex diseases, including cancer [4]
We demonstrated that expression of ZNF707, C8orf58, KIAA1967, RHOBTB2, and TNFRSF10B was significantly increased in Ovarian clear cell adenocarcinoma (Ov-CCA) in comparison to high-grade serous ovarian carcinoma (HGSOC) (p < 0.001, p = 0.004, p = 0.017, p = 0.001, and p < 0.001, respectively), whereas expression of APBA3 was significantly lower in Ov-CCA (p < 0.001, Figure 3B)
Multiple in silico and in vitro analyses were conducted to demonstrate the function of the four single-nucleotide polymorphisms (SNPs) and seven associate genes in Ov-CCA. rs4873815-TT, which demonstrated enrichment in the Japanese ancestry compared to European and Chinese ancestries, induced promoter activity on ZNF707 compared to rs4873815-CC
Summary
Identifying causal variants located outside of the protein coding region is an ingenious and perceptive way to discover the genetic landscape of complex diseases, including cancer [4]. Several studies on cancers, including epithelial ovarian carcinoma (EOC), have identified novel genes and proposed molecular pathways through eQTL analysis using the identified risk loci in GWAS [6,7,8]. Using well-known genetic mutations or disease-specific characteristics as the phenotypic trait, several studies with eQTL analysis have already identified specific risk loci and associated novel candidate genes in multiple cancer types, including breast, prostate, lung, colorectal, and high-grade serous ovarian carcinoma (HGSOC) [3,9,10,11]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
