Identification of key microRNAs and the underlying molecular mechanism in spinal cord ischemia-reperfusion injury in rats

Identification of the key differentially expressed miRNAs in SCII

The existing studies on the miRNA expressions in SCII were located on PubMed with keywords such as miRNA and spinal cord ischemic reperfusion injury on 30 November 2020. The species involved in the related studies were limited to rats. The information on SCII and the miRNA sequencing samples and the detection methods for the differentially expressed miRNAs (DEmiRNAs) were extracted from the relevant studies. The upregulated and downregulated DEmiRNAs in SCII compared with the sham groups were also extracted. The DEmiRNAs occurrence of each dataset was calculated, and DEmiRNAs that appeared in at least two datasets were defined as key DEmiRNAs.

Predicting the target genes of DEmiRNAs in SCII

The target genes of DEmiRNAs were predicted on miDRB with the gene target score set to over 80.

Gene ontology and KEGG enrichment analysis on the target genes of DEmiRNAs in SCII

Information on the target genes of DEmiRNAs was obtained from the DAVID database for KEGG pathway analysis and GO enrichment analysis (Guo et al., 2019). The cut-off criteria for the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment and the gene ontology (GO) term enrichment were statistical significance (p < 0.05). According to the relations between the genes and the statistically significant biological processes as well as the relations among miR-3568, genes, and the biological processes, a miR-3568-biological processes-gene network was built.

Transcription factor analysis of the miRNA-regulated target genes in SCII

The target genes of DEmiRNAs in SCII were submitted to the TFactS database (http://www.tfacts.org/), and the transcription factors regulating the target genes of DEmiRNAs were predicted by the false discovery rate, E value, q value, and p value. To obtain reliable transcription factors, the false discovery rate, E value, q value, and p value should all be lower than 0.05. The transcription factors of the target genes of DEmiRNAs were counted, respectively. The unique and common transcription factors were compared.

Rat model

Male Sprague Dawley (SD) rats, 8 weeks, weighing 200–250 g, used in the SCII model were purchased from Liaoning Changsheng Biotechnology Co., Ltd. This study was approved by the Ethics Committee of China Medical University (CMU2020266). SCII was induced in the rats via a cross-clamped aortic arch (Li et al., 2014a, 2014b). Briefly, upon anesthesia by intraperitoneally injecting 4% sodium pentobarbital (50 mg/kg; Beyotime Biotechnology, China), endotracheal intubation (24 g trocar sleeve) and lung ventilation was accomplished with a small-animal ventilator (Harvard Apparatus, Holliston, MA, USA; tidal 15 mL/kg, breathing frequency 80–100 times/min, breathing ratio 1:1). Body temperatures were kept at 37.5 ± 0.5 °C and monitored with a rectal probe. Under aseptic conditions, the left common carotid artery was exposed in the middle of neck. Then, the aortic arch was uncovered through a cervicothoracic incision. Under direct vision, the aortic arch was cross-clamped for 14 min between the left carotid artery and the left subclavian artery to induce ischemia. A catheter (24 g trocar sleeve) was inserted into the femoral artery for blood pressure measurements. After ischemia confirmation (90% reduction of the flow assessed at the femoral artery using a laser Doppler blood flow monitor (Moor Instruments, Axminster, Devon, UK)), the clamping was removed, followed by 24-h reperfusion. This procedure was performed on the sham animals, except for blockade.

Interventions

Thirty-two male SD rats were assigned randomly to four groups: (1) sham group; (2) SCII group; (3) SCII + anti-miRNA oligonucleotides (AMOs) of miR-3568 group; and (5) SCII + NC-miR-3568 group. Rats were intrathecally injected with AMOs (Li et al., 2015b). A synthetic miR-3568 AMO (AMO-3568) (5′-CUGCUUCUGCACGGGAAGAACA-3′), and negative control were purchased from Jima Inc (China). Rats were intrathecally injected with liposome complexes of the oligonucleotides (50 mg/kg) and Lipofectamine^® 2000 (Invitrogen, Carlsbad, CA, USA). Rats were injected once a day for three consecutive days before the surgical procedure.

Quantitative reverse transcription-polymerase chain reaction (qRT-PCR)

Rats were euthanized by sevoflurane overdose at 24 h after SCII in accordance with the established protocol by the Experimental Animal Center of China Medical University. Segments L4-L6 of the spinal cord were collected to extract total RNA with the Trizol reagent (Takara, Otsu, Japan). The RNA was reverse-transcribed into cDNA using a Prime-Script RT reagent Kit with gDNA Eraser (Takara, Otsu, Japan) (Jia et al., 2019). The levels of miRNA were measured using an SYBR Premix qRT-PCR (Takara, Otsu, Japan) on the Applied Biosystems 7500 Real Time PCR system (Takara) with U6 as an internal control. The primer sequences are shown in Table 1. The 2^−ΔΔCt method was used to calculate the data.

Neurological evaluation

At 24 h after SCII, the hind limb functions were evaluated based on the Tarlov scores: 0 = no voluntary hind limb function, 1 = poor hind limb motor function with perceptible movement, 2 = joint motion present with no ability to stand, 3 = stands and walks, and 4 = normal hind limb function (Fang et al., 2015; Li et al., 2016a; Tarlov, 1972).

Western blotting

The expression levels of GATA6, GATA4, RBPJ, BCL-2, Bax, and cleaved caspase-3 in spinal cord tissues were measured with Western Blotting. Total proteins were extracted from the L4-L6 segments of the spinal cords with RIPA buffer (KangChen, China). The antibodies used were rabbit polyclonal anti-GATA6 (Affinity, AF5270, China), rabbit polyclonal anti-GATA4 (Affinity, AF5245, China), rabbit polyclonal anti-RBPJ (Affinity, DF7453, China), rabbit monoclonal anti-Bax (#2772; Cell Signaling Technology, Danvers, MA, USA), rabbit polyclonal anti-Bcl-2 (26593-1-AP, ProteinTech, Rosemont, IL, USA), rabbit polyclonal anti- cleaved caspase-3 (ab2302; Abcam, Cambridge, MA, USA), rabbit anti-GAPDH (Boster, A00227, China), and HRP-conjugated secondary antibodies (Beyotime, China).

Statistical analysis

SPSS 15.0 (IBM, Armonk, NY, USA) was used for statistical analyses. The results were expressed as mean ± standard deviation. Student’s t-test, one-way ANOVA followed by the Tukey’s test, or two-way repeated-measures ANOVA with the appropriate post hoc analysis were adopted to calculate the significant difference. p < 0.05 was defined as significant.

Searching and identification of key DEmiRNAs in SCII

From the existing SCII miRNA expression profiling in rats, 23 independent miRNA expression datasets were obtained from PubMed, which provided the DEmiRNAs in the spinal cord tissues of SCII rats compared with the sham rats. SCII miRNA datasets were named based on the corresponding authors and year of publication for further study. Basic characteristics of SCII DEmiRNAs datasets were displayed in Table 2.

The number of DEmiRNAs in each of the 23 SCII miRNA expression datasets was different (Fig. 2). A total of 151 DEmiRNAs were identified in the 23 SCII miRNA expression datasets. Several studies identified more DEmiRNAs than others, such as HJR2013, LXQ2015, and ZGL2020. The number of DEmiRNAs in dataset ZGL2020 was the largest (12 and 13 upregulated DEmiRNAs; 1 and 27 downregulated DEmiRNAs). Due to the differences in DEmiRNAs among the datasets, a systematic analysis of the SCII miRNA differential expression datasets was conducted to identify the key DEmiRNAs and the related potential biological functions in SCII.

DEmiRNAs appearing in at least two datasets were defined as key DEmiRNAs, and a total of 19 key DEmiRNAs were identified (Figs. 2 and 3). SCII induced 9 upregulated expressions, namely, miR-144-3p, miR-3568, miR-204, miR-30c, miR-200b, miR-463, miR-760-5p, miR-155-3p, and miR-34c-3p, and 10 downregulated expressions, namely, miR-125b-2-3p, miR-21-5p, miR-199a-3p, miR-352, miR-743b-3p, miR-28-5p, miR-291a-3p, miR-702-3p, miR-129-1-3p, and miR-136. Among the 19 key DEmiRNAs, the roles of five key DEmiRNAs in SCII, including miR-204, miR-30c, miR-21-5p, miR-155-3p, and miR-136, have been investigated. However, the roles of the remaining key DEmiRNAs have not been explored.

Although appeared in three independent datasets, LJA2016, ZGL2020, and HF2020, miR-22-3p showed inconsistent expression trends.

The expression of miR-760-5p has been validated via RT-PCR in HJR2013, and the expression of miR-200b has been validated via RT-PCR in LXQ2015. RT-PCR was adopted to explore the expression of the key DEmiRNAs that have not been studied or validated via RT-PCR. The results revealed that the expression levels of miR-144-3p, miR-3568, miR-34c-3p, and miR-463 increased significantly at 24 h following SCII (Fig. 3A), while the expression levels of miR-291a-3p, miR-702-3p, miR-28-5p, miR-199a-3p, miR-352, miR-743b-3p, miR-125b-2-3p, and miR-129-1-3p decreased significantly at 24 h following SCII (Fig. 3B).

KEGG and GO enrichment analysis of the target genes of DEmiRNAs in SCII

The target genes of the 19 key DEmiRNAs were predicted on miDRB. The minimum target score was set to 80, and the number of the target gene of the 19 key DEmiRNAs were obtained (Table 3).

According to the KEGG enrichment analysis of the target genes of the upregulated and downregulated DEmiRNAs, the involved pathways in the upregulated DEmiRNAs are cGMP-PKG and cAMP signaling pathway and that in the downregulated DEmiRNAs are Chemokine and MAPK signaling pathway. As shown in Figs. 4A–4B, GO enrichment analysis results indicated that target genes of the upregulated DEmiRNAs were markedly enriched in biological processes such as brain development and positive regulation of transcription from RNA polymerase II promoter. Target genes of the downregulated DEmiRNAs were mainly enriched in biological processes such as intracellular signal transduction and negative regulation of cell proliferation, as shown in Figs. 4C–4D.

Transcription factor analysis of the key DEmiRNAs target genes in SCII

The transcription factors corresponding to the key DEmiRNAs regulated target genes were analyzed. For the upregulated DEmiRNAs target genes, 11 transcription factor genes with 178 interactions were obtained. For the downregulated DEmiRNAs target genes, four transcription factor genes with 37 interactions were formed. Among the 11 transcription factor genes that regulated the key DEmiRNAs regulated target genes, four transcription factor genes could regulate the target genes of the upregulated or downregulated DEmiRNAs (Fig. 5A). The four transcription factors, including SP1, GLI1, GLI2, and FOXO3, had significant regulatory effects on the target genes of the key DEmiRNAs (Fig. 5B).

Verification of miR-3568’s involvement in SCII

For the 19 key DEmiRNAs not studied in SCII, miR-3568 was especially interesting. A previous study found that miR-3568 was upregulated in liver and serum in rats with alcoholic steatohepatitis and associated with MAPK signaling pathway (Chen et al., 2013). The expression of miR-3568 also increased in matrix vesicles (MV) compared with vascular smooth muscle cell (VSMC) in the rats with chronic kidney disease, indicating the role of miR-3568 in vascular calcification and/or MV formation (Chaturvedi et al., 2015). A recent study revealed that miR-3568 expression in simulated IRI-induced H9C2 cardiomyocytes increased in a time-dependent manner, which promotes simulated IRI-induced apoptosis in H9C2 cardiomyocytes through targeting TRIM62 (Li et al., 2020b). Although miR-3568 was upregulated after SCII in rats based on microRNA microarrays results (Chen et al., 2020b; Li et al., 2015b), the expression and potential function of miR-3568 in SCII has not been further explored.

Enriched biological processes for the target genes of miR-3568 were obtained through GO analysis. According to the relations between genes and statistically significant biological processes as well as the relations among miR-3568, genes, and biological processes, a miR-3568-biological processes-gene network was constructed. GO analysis showed that miR-3568 target genes were enriched notably in several biological processes (p < 0.05), and GATA6, GATA4, and RBPJ were enriched in several biological functions (Fig. 6).

In addition, the SCII induced severe neurological deficits of lower extremities, while miR-3568 AMO improved neurological function (Fig. 7A). Cleaved caspase-3 was markedly upregulated in SCII compared with the sham group, and miR-3568 AMO reduced cleaved caspase-3 expression. Bax expression levels showed similar trends as cleaved caspase-3. Bcl-2 expression levels significantly decreased after SCII, and miR-3568 AMO increased Bcl-2 expression (Figs. 7B–7E). Moreover, the expression of GATA6, GATA4, and RBPJ decreased after SCII. Intrathecal injection with miR-3568 AMO attenuated this upregulation. The results were shown in Figs. 8A–8D.