Biomarkers mining for spinal cord injury based on integrated multi-transcriptome expression profile data.

BACKGROUND

This study was aimed to discover more biomarkers associated with spinal cord injury (SCI) by constructing a competing endogenous RNA (ceRNA) network.

METHODS

The transcriptome expression profile data related to SCI (GSE45006 GSE20907) were downloaded from GEO database. The differentially expressed RNAs (DERs), including lncRNAs, miRNAs, and mRNAs, between SCI and control groups were selected, which were then performed function enrichment analyses. Following that, a SCI-related ceRNA regulatory network was constructed. PCA analysis was performed on the genes constituting the ceRNA regulatory network directly related to SCI.

RESULTS

In GSE45006 and GSE20907 datasets, there were respectively 3336 and 1453 DERs. Venn analysis showed that there were 429 DERs which had consistent differential expression direction. RGD1564534-miR-29b-5p relation pair and 103 miRNA-target regulatory pairs were integrated to construct the ceRNA regulatory network. Then a SCI-related ceRNA regulatory network including 8 mRNAs of IFNGR1, STAT2, CYBB, NFATC1, FCGR2B, HMOX1, TLR4, and HK2, a lncRNA of RGD1564534, and a miRNA of miR-29b-5p was constructed. Additionally, two pathways, osteoclast differentiation, and HIF-1 signaling pathway, were involved in this network. PCA indicated the samples before and after injury can be significantly distinguished based on the genes in the ceRNA network.

CONCLUSION

A total of 8 SCI-related mRNAs have been identified in the ceRNA network, including IFNGR1, STAT2, CYBB, NFATC1, FCGR2B, HMOX1, TLR4, and HK2. Moreover, RGD1564534 may serve as ceRNA by competitively binding to miR-29b-5p to regulate the expression of 8 SCI-related mRNAs. Therefore, these genes may serve as key biomarkers of SCI.