Abstract
PURPOSE: To identify potential biomarkers and the molecular mechanisms associated with repeated implantation failure (RIF), three microarray datasets, GSE71331 (lncRNA + mRNA), GSE111974 (lncRNA + mRNA), and GSE71332 (miRNA), were retrieved from the Gene Expression Omnibus (GEO) database. METHODS: The differentially expressed mRNAs (DEMs), lncRNAs (DElncRNAs), and miRNAs (DEmiRNAs) between normal control samples (C group) and RIF samples (RIF group) were identified, and then a module partition analysis was performed based on weighted correlation network analysis (WGCNA). Following enrichment analysis of the genes, the lncRNA-miRNA-mRNA interactions (ceRNA) were examined. The mRNAs in the ceRNA network were evaluated using the GSE58144 dataset. Finally, the key RNAs were verified using reverse transcription-quantitative polymerase chain reaction (RT-qPCR). RESULTS: Fifty-three DEmiRNAs, 327 DEMs, and 13 DElncRNAs were identified between the C and RIF groups. According to WGCNA, the magenta module was positively correlated with RIF disease status. The lncRNA-mRNA interaction analysis based on genes in the magenta module revealed the intersecting lncRNAs, including peptidylprolyl isomerase E-like pseudogene (PPIEL) and the testis-specific transcript, y-Linked 14 (TTTY14); these lncRNAs are mainly involved in functions, such as plasma membrane organization. The ceRNA network analysis revealed several interactions, such as TTTY14-miR-6088-semaphorin 5 A (SEMA5A). Finally, SEMA5A and the zinc finger protein 555 (ZNF555) were identified to be significantly upregulated in the RIF group compared with those in the C group in the GSE58144 dataset. The RT-qPCR results aligned with the above results. CONCLUSIONS: Overall, TTTY14, ZNF555, SEMA5A, PPIEL, and miR-6088 could serve as novel biomarkers of RIF.