Construction of a circular RNA-based competing endogenous RNA network to screen biomarkers related to intervertebral disc degeneration

Abstract

Background: Intervertebral disc degeneration (IDD) is a leading cause of disability with limited treatment strategies. A better understanding of the mechanism of IDD might enable less invasive and more targeted treatments. This study aimed to identify the circular RNA (circRNA)-microRNA (miRNA)-messenger RNA (mRNA) competing endogenous RNA (ceRNA) regulatory mechanisms in IDD. METHODS : The GSE67567 microarray dataset was downloaded from the Gene Expression Omnibus database. After data preprocessing, differentially expressed circRNAs, miRNAs and mRNAs between IDD and controls were identified. A ceRNA network was constructed on the basis of the interaction between circRNAs and miRNAs, and miRNAs and mRNAs. Pathway enrichment analysis was performed on the mRNAs in the ceRNA network. Then, with 'intervertebral disc degeneration' as keywords, IDD-related Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways were searched for in the Comparative Toxicogenomics Database.

Results: A total of 105 differentially expressed circRNAs, 84 miRNAs and 967 mRNAs were identified. After analysis, 86 circRNA-miRNA, and 126 miRNA-mRNA regulatory relationship pairs were obtained to construct a ceRNA network. The mRNAs were enriched in six KEGG signalling pathways, and four were associated with IDD: the hsa04350: TGF-beta signalling pathway, hsa04068: FoxO signalling pathway, hsa05142: Chagas disease (American trypanosomiasis) and hsa04380: Osteoclast differentiation. An IDD-related ceRNA network was constructed involving four circRNAs, three miRNAs and 11 mRNAs. Auxiliary validation showed that the expression levels of miR-185-5p, miR-486-5p, ACVR1B, FOXO1, SMAD2 and TGFB1 were consistent in different databases.

Conclusions: Our study identified some circRNA-miRNA-mRNA interaction axes potentially associated with the progression of IDD, viz.: circRNA_100086-miR-509-3p-MAPK1, circRNA_000200-miR-185-5p-TGFB1, circRNA_104308-miR-185-5p-TGFB1, circRNA_400090-miR-486-5p-FOXO1 and circRNA_400090-miR-486-5p-SMAD2.

Keywords: Circular RNA; Competing endogenous RNA; Intervertebral disc degeneration; mRNA; microRNA.

Fig. 1

Identification of differentially expressed circRNAs, miRNAs and mRNAs between IDD group and control group. Volcano plots (up panel) and heat maps (low panel) based on significantly differentially expressed circRNA (A), miRNA (B) and mRNA (C). Blue and red dots indicate significantly downregulated and upregulated RNAs, respectively. The black horizontal line indicates FDR = 0.05, and the two vertical dashed lines indicate |log₂FC|= 1

Fig. 2

Construction and functional analysis of co-expression networks. A Co-expression network diagram of circRNAs and mRNAs with significant differences. The square represents the circRNAs, the circle represents the mRNAs and the colour change from blue to red represents the change in |Log₂FC| from low to high. B Bubble diagram of GO biological process and KEGG signalling pathway of mRNAs in the co-expression network. The horizontal axis represents the number of genes, and the vertical axis represents the names of GO and KEGG signalling pathways. The size and colour of the node represent − log₁₀(p value)

Fig. 3

Construction of a ceRNA regulatory network. A Regulatory network diagram of the circRNAs and mRNAs with significant differences. B Regulatory network of miRNA–target. C ceRNA regulatory network. The square represents the circRNAs, the triangle represents the miRNAs, the circle represents the mRNAs and the colour change from blue to red represents the change in |Log₂FC| from low to high

Fig. 4

Intervertebral disc degeneration-related integrated ceRNA regulatory network. The rhombus represents the disease-related pathway, the square represents the circRNAs, the triangle represents the miRNAs, the circle represents the mRNAs and the colour change from blue to red represents the change in |Log₂FC| from low to high

Fig. 5

Validation of differential expression of miRNAs in the ceRNA network in independent dataset. Expression levels of three miRNAs in GSE63492 (A) and GSE19943 (B). The difference of expression level (log2 transformed expression) was compared between control (CTRL) group and IDD group by student’s t test. * represents P < 0.05

Fig. 6

Validation of differential expression of mRNAs in the ceRNA network in independent dataset. Expression levels of 11 mRNAs in GSE56081 (A) and GSE34095 (B). The difference of expression level (log2 transformed expression) was compared between control (CTRL, patients with scoliosis) group and IDD group by student’s t test. * represents P < 0.05 and *** represents P < 0.001