High-throughput RNA sequencing for screening the circular RNA expression profiling associated with chemo-resistance inosteosarcoma

Abstract

Objective To screen the expression profile of circular RNA associated with chemo-resistance in osteosarcoma, and to analyze and identify its possible molecular functions. Methods Three pairs of matched drug-resistant and sensitive human osteosarcoma cell lines (MG63/DXR vs MG63, U2OS/DXR vs U2OS, KHOS/DXR vs KHOS) were first tested by CCK-8 assay for three commonly used osteosarcoma chemotherapy drugs (doxorubicin, cisplatin, and methotrexate); then, using next-generation high-throughput RNA sequencing technology, comparative analysis of circRNA expression was performed in three pairs of matched multidrug-resistant and sensitive human osteosarcoma cell lines; followed by real-time quantitative PCR (qRT-PCR) to confirm the reliability and accuracy of sequencing data in chemotherapy-resistant osteosarcoma cell lines and tissues. In addition, a variety of bioinformatics analyses, including GO, KEGG pathway analysis and the construction of circRNA-miRNA networks, were performed to predict potential molecular functions of differentially expressed circRNAs and to construct relevant regulatory pathways or networks. Results Three osteosarcoma-resistant cells (MG63/DXR, U2OS/DXR, KHOS/DXR) were significantly resistant to the three common osteosarcoma chemotherapy drugs compared with control cells (MG63, U2OS, KHOS), which laid a solid foundation for the further experiments. RNA sequencing revealed a total of 80 circRNAs differentially expressed between the two groups. Compared with drug-sensitive osteosarcoma cells, 57 circRNAs were upregulated and 23 circRNAs weredownregulated in the drug-resistant osteosarcoma cell lines (fold change≥ 2 or ≤0.5, P<0.05). Tenrandomly selected circRNAs were verified by qRT-PCR and the results showed that 9 of the 10 circRNAswere consistent with the sequencing results. In addition, KEGG pathway analysis showed that 56 pathways were significantly enriched in differentially expressed circRNAs, including Glycolysis/gluconeogenesis, ABC transporter, VEGF signaling pathway, and so on. Moreover, the differently expressedcircRNA-hsa_circ_0004674 with the highest fold change was highly expressedin the chemotherapy-resistant osteosarcoma cells and tissues, associated with poor prognosis in osteosarcoma patients. Some potential endogenous competitive RNA (ceRNA) regulatory pathways associated with hsa_circ_0004674, such as hsa_circ_0004674-miR-490-3p-ABCC2 and hsa_circ_0004674- miR-1254- EGFR, were constructed by use of the authoritative databases (target scan and miRanda) and literature searching and the miRNA response element sequences (MREs) between miRNAs that have a potential ceRNA regulatory relationship with hsa_circ_0004674 were alsopredicted. Conclusion CircRNA is closely related to tumor progression and may play a role in osteosarcoma chemo-resistance. Besides, hsa_circ_0004674 may be a potential candidate for reversing drug resistance of osteosarcoma. Key words: Sarcoma; RNA; Drug resistance, neoplasm; MicroRNAs; High-throughput screening assays