circ_014260/miR-384/THBS1 aggravates spinal cord injury in rats by promoting neuronal apoptosis and endoplasmic reticulum stress

Abstract

Objective: To explore the mechanism of circ_014260 regulating neuronal apoptosis, oxidative stress, and endoplasmic reticulum stress in rats with spinal cord injury (SCI) via miR-384/THBS1 axis.

Methods: T9-L10 spinal cord segments of Sprague Dawley rats were subjected to compression or contusion injuries after T10 laminectomy to establish rat models of SCI, which were then divided into SCI group, si-circ group and oe-circ group according to the transfection. There was another sham operation group which received no treatment. There were 10 rats in each group. The Basso-Beattie-Bresnahan scale and HE staining were used to evaluate the changes in neuronal motor function in rats with SCI. TUNEL staining was used to determine the neuronal apoptosis. Flow cytometry was used to measure the changes in H₂O₂-induced apoptosis of primary neurons. The activities of myeloperoxidase, malondialdehyde, superoxide dismutase and catalase were measured to evaluate the level of oxidative stress. Western blot was used to measure the expressions of CHOP and CRP78 (which are related to endoplasmic reticulum stress). Expression of circ_014260, miR-384 and THBS1 in tissues and cells was measured by qRT-PCR. RNase R restriction enzyme digestion and chromatin fractionation were used to identify the nature of circ_014260. Dual-luciferase reporter assay and RNA immunoprecipitation were used to verify the targeted binding relationship between circ_014260 and miR-384, as well as between miR-384 and THBS1.

Results: Compared with the sham operation group or the untreated rat primary neurons (control group), increased expression of circ_014260 and THBS1 as well as decreased expression of miR-384 were observed in the spinal cord tissue from rats with SCI and in H₂O₂-treated primary neurons (all P<0.05). The results of both in vivo and in vitro experiments showed that knocking down circ_014260 could reduce neuronal apoptosis and inhibit oxidative stress and endoplasmic reticulum stress in rats with SCI (all P<0.05). Circ_014260 targetedly inhibited miR-384 to up-regulate the expression of THBS1. Both miR-384 inhibitor and THBS1 overexpression vector partially reversed the alleviated neuronal damage by knocking down circ_014260 (both P<0.05).

Conclusion: Circ_014260 promotes neuronal damage in rats with SCI by inhibiting miR-384 to up-regulate the expression of THBS1. Thus, circ_014260 could possibly be a new molecular target of SCI.

Keywords: THBS1; circ_014260; miR-384; spinal cord injury.

Figure 1

Results of bioinformatics analyses. A: The heat map of differentially expressed miRNAs in the GSE114426 data set; B: The heat map of the top 5 circRNAs according to the log2FC value; C: circ_014260 is the intersection of the downstream target miRNAs of StarBase and TSCD; D: Target binding site of circ_014260 and miR-384; E: The intersection of miR-384 downstream target genes in StarBase, miRDB and TargetScan; F: Results of Gene Ontology enrichment analysis; G: The targeted binding site of miR-384 and THBS1.

Figure 2

Identification of neurons by tubulin staining.

Figure 3

Circ_014260 and THBS1 were up-regulated and miR-384 was down-regulated in SCI. A: Expression of circ_014260 in sham operation group and SCI group measured by qRT-PCR (n=10); B: Expression of miR-384 in the sham operation group and SCI group measured by qRT-PCR (n=10); C: Expression of THBS1 in the sham operation group and SCI group measured by qRT-PCR (n=10); D: Expression of circ_014260 in untreated and H₂O₂-treated rat neurons measured by qRT-PCR (n=3); E: Expression of miR-384 in untreated and H₂O₂-treated rat neurons measured by qRT-PCR (n=3); F: Expression of THBS1 in untreated and H₂O₂-treated rat neurons measured by qRT-PCR (n=3). Compared with sham operation group, &P<0.05; compared with control group, #P<0.05. SCI: spinal cord injury.

Figure 4

Circ_014260 induced neuronal cell apoptosis, oxidative stress and ERs in SCI. A: Transfection efficiency of si-circ or oe-circ measured by qRT-PCR (n=3); B, C: Changes in neuronal apoptosis detected by flow cytometry (n=3); D: Changes in MPO activity measured by Elisa (n=3); E: Changes in MDA activity measured by Elisa (n=3); F: Changes in SOD activity measured by Elisa (n=3); G: Changes in CAT activity measured by Elisa (n=3); H, I: Protein levels of CHOP and GRP78 measured by Western blot (n=3). Compared with control group, #P<0.05; compared with H₂O₂+si-NC group, ^P<0.05; compared with H₂O₂+oe-NC group, $P<0.05. ERs: endoplasmic reticulum stress, SCI: spinal cord injury, MPO: myeloperoxidase, MDA: malondialdehyde; SOD: superoxide dismutase; CAT: catalase.

Figure 5

Circ_014260 was proved to be a competitive endogenous RNA for miR-384. A: Results of RNase restriction enzyme digestion (n=3); B: Results of chromatin fractionation (n=3); C: Results of Dual-luciferase reporter assay (n=3); D: Results of RNA immunoprecipitation (n=3); E: Level of miR-384 in rat neurons measured by qRT-PCR (n=3). Compared with the Mock group, ■P<0.05; compared with the Cytosol group, □P<0.05; compared with miR-NC group, ♢P<0.05; compared with the IgG group, ♦P<0.05; compared with control group, #P<0.05; compared with H₂O₂+si-NC group, ^P<0.05; compared with H₂O₂+oe-NC group, $P<0.05.

Figure 6

Inhibition of miR-384 aggravated the neuronal damage alleviated by downregulation of circ_014260. A: Level of miR-384 measured by qRT-PCR (n=3); B, C: Changes in neuronal apoptosis detected by flow cytometry (n=3); D: Changes in MPO activity measured by Elisa (n=3); E: Changes in MDA activity measured by Elisa (n=3); F: Changes in SOD activity measured by Elisa (n=3); G: Changes in CAT activity measured by Elisa (n=3); H, I: Protein levels of CHOP and GRP78 measured by Western blot (n=3). Compared with control group, #P<0.05; compared with H₂O₂+si-NC group, ^P<0.05; compared with H₂O₂+si-circ+miR-NC group, △P<0.05. MPO: myeloperoxidase; MDA: malondialdehyde; SOD: superoxide dismutase; CAT: catalase.

Figure 7

THBS1 was the target gene of miR-384. A: Results of Dual-luciferase reporter assay (n=3); B: Results of RNA immunoprecipitation (n=3); C, D: Level of THBS1 measured by qRT-PCR (n=3). Compared with miR-NC group, ♢P<0.05; compared with the IgG group, ♦P<0.05; Compared with control group, #P<0.05; compared with H₂O₂+si-NC group, ^P<0.05; compared with H₂O₂+si-circ+miR-NC group, △P<0.05.

Figure 8

Overexpression of THBS1 partially reversed the neuronal damage reduced by knocking down circ_014260. A: Level of THBS1 measured by qRT-PCR (n=3); B, C: Changes in neuronal apoptosis detected by flow cytometry (n=3); D: Changes in MPO activity measured by Elisa (n=3); E: Changes in MDA activity measured by Elisa (n=3); F: Changes in SOD activity measured by Elisa (n=3); G: Changes in CAT activity measured by Elisa (n=3); H, I: Protein levels of CHOP and GRP78 measured by Western blot (n=3). Compared with control group, #P<0.05; compared with H₂O₂+si-NC group, ^P<0.05; compared with H₂O₂+si-circ+pcDNA3.1-NC group, ▲P<0.05. MPO: myeloperoxidase; MDA: malondialdehyde; SOD: superoxide dismutase; CAT: catalase.

Figure 9

Knockdown of circ_014260 reduced neuronal damage in rats with SCI (n=10). A: Level of circ_014260 measured by qRT-PCR (n=3); B: SCI lesions in rats (HE staining, 400×; n=3); C: Comparison of spinal cavity area in each group (n=3); D: Changes of BBB score in rats with SCI (n=3); E, F: Changes in neuronal apoptosis in spinal cord tissue of rats with SCI (TUNEL staining, 400×; n=3). Compared with sham operation group, &P<0.05; compared with the SCI group, ○P<0.05. SCI: spinal cord injury; BBB: Basso, Beattie & Bresnahan locomotor rating.

Figure 10

Mechanism of circ_014260/miR-384/THBS1 promoting neuronal apoptosis and endoplasmic reticulum stress and aggravating spinal cord nerve injury in rats.