Identification of a circRNA-miRNA-mRNA regulatory network for exploring novel therapeutic options for glioma

Yi He; Yihong Chen; Yuxin Tong; Wenyong Long; Qing Liu

doi:10.7717/peerj.11894

Identification of a circRNA-miRNA-mRNA regulatory network for exploring novel therapeutic options for glioma

PeerJ. 2021 Aug 6:9:e11894. doi: 10.7717/peerj.11894. eCollection 2021.

Authors

Yi He^#¹, Yihong Chen^#¹, Yuxin Tong^{2

3}, Wenyong Long¹, Qing Liu¹

Affiliations

¹ Neurosurgery Department, Xiangya Hospital Central South University, Changsha, Hunan, China.
² Department of Ophthalmology, Second Xiangya Hospital, Central South University, Changsha, Hunan, China.
³ Hunan Clinical Research Center of Ophthalmic Disease, Changsha, Hunan, China.

^# Contributed equally.

Abstract

Background: Glioma is the most common brain neoplasm with a poor prognosis. Circular RNA (circRNA) and their associated competing endogenous RNA (ceRNA) network play critical roles in the pathogenesis of glioma. However, the alteration of the circRNA-miRNA-mRNA regulatory network and its correlation with glioma therapy haven't been systematically analyzed.

Methods: With GEO, GEPIA2, circBank, CSCD, CircInteractome, mirWalk 2.0, and mirDIP 4.1, we constructed a circRNA-miRNA-mRNA network in glioma. LASSO regression and multivariate Cox regression analysis established a hub mRNA signature to assess the prognosis. GSVA was used to estimate the immune infiltration level. Potential anti-glioma drugs were forecasted using the cMap database and evaluated with GSEA using GEO data.

Results: A ceRNA network of seven circRNAs (hsa_circ_0030788/0034182/0000227/ 0018086/0000229/0036592/0002765), 15 miRNAs(hsa-miR-1200/1205/1248/ 1303/3925-5p/5693/581/586/599/607/640/647/6867-5p/767-3p/935), and 46 mRNAs (including 11 hub genes of ARHGAP11A, DRP2, HNRNPA3, IGFBP5, IP6K2, KLF10, KPNA4, NRP2, PAIP1, RCN1, and SEMA5A) was constructed. Functional enrichment showed they influenced majority of the hallmarks of tumors. Eleven hub genes were proven to be decent prognostic signatures for glioma in both TCGA and CGGA datasets. Forty-six LASSO regression significant genes were closely related to immune infiltration. Finally, five compounds (fulvestrant, tanespimycin, mifepristone, tretinoin, and harman) were predicted as potential treatments for glioma. Among them, mifepristone and tretinoin were proven to inhibit the cell cycle and DNA repair in glioma.

Conclusion: This study highlights the potential pathogenesis of the circRNA-miRNA-mRNA regulatory network and identifies novel therapeutic options for glioma.

Keywords: ceRNA network; circRNA; Biomarker; Glioma; Novel therapy.

Grants and funding

This work was supported by the National Natural Science Foundation of China (grant number 81802974). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.