Dexmedetomidine‐up‐regulated microRNA‐381 exerts anti‐inflammatory effects in rats with cerebral ischaemic injury via the transcriptional factor IRF4

Abstract Dexmedetomidine (Dex) possesses analgesic and anaesthetic values and reported being used in cerebral ischaemic injury therapeutics. Accumulating studies have determined the effect of microRNAs (miRNAs) on the cerebral ischaemic injury. Thus, the present study aimed to unravel the molecular mechanism of miR‐381 and Dex in cerebral ischaemic injury. For this purpose, the cerebral ischaemic injury rat model was established by induction of middle cerebral artery occlusion (MCAO) and expression of miR‐381 and IRF4 was determined. Thereafter, MCAO rats were treated with Dex, miR‐381 mimic, miR‐381 inhibitor and oe‐IRF4 respectively, followed by evaluation of neurological function. Furthermore, neuron cells were isolated from the hippocampus of rats and subjected to oxygen‐glucose deprivation (OGD). Then, OGD‐treated neuron cells and primary neuron cells were examined by gain‐ and loss‐of‐function assay. Neuron cell apoptosis was detected using TUNEL staining and flow cytometry. The correlation between interferon regulatory factor 4 (IRF4) and interleukin (IL)‐9 was detected. Our results showed down‐regulated miR‐38 and up‐regulated IRF4 in MCAO rats. Besides, IRF4 was targeted by miR‐381 in neuron cells. Dex and overexpressed miR‐381, or silenced IRF4 improved the neurological function and inhibited neuron cell apoptosis in MCAO rats. Additionally, in MCAO rats, Dex was found to increase the miR‐381 expression and reduced IRF4 expression to decrease the IL‐9 expression, which suppressed the inflammatory response and cell apoptosis both in vivo and in vitro. Importantly, our study demonstrated that Dex elevated the expression of miR‐381, which ultimately results in the inhibition of inflammation response in rats with cerebral ischaemic injury.


| INTRODUC TI ON
The cerebral ischaemic injury, often induced by ischaemic stroke, is associated with long-term disability. 1 However, in more serious conditions, of cerebral ischaemic injury, inflammation, ischaemia and glial cell dysfunction contribute to persistent brain injury, 2 thus indicating that dysfunction of glial cells and damage of the blood-brain barrier are involved in ischaemic brain injury. 3 Nonetheless, cerebral ischaemic injury has been attributed to the devastating complication of neurological and cardiovascular surgeries, 4 while the neurodegenerative disorders caused by brain injury significantly impair the memory and learning ability, limb use, and other neurological performances. 5 Moreover, dexmedetomidine (Dex) is an α2-adrenergic receptor agonist which possesses analgesic and sedative properties. 6 Intriguingly, Dex has been demonstrated to enhance the outcomes of the cardiac and neurological surgeries and to relieve the patient pain. 7 Importantly, Dex has been reported to exhibit neuroprotective against cerebral ischaemic or reperfusion injury in rats. 8 Besides, other studies have illustrated the protective effect of Dex on the ischaemic or reperfusion injury of various organs including the heart and kidney. 8 On the other hand, recent studies have indicated the prognostic value of circulating miR-19a and suggested it as a promising molecular target for early diagnosis and prognosis of acute myocardial infarction. 9 Though microRNAs (miRNAs), that is miR-129-5p 10 and miR-29b, 11 have indicated being involved in the neuroprotective effect of Dex.
Yet, scarce studies investigated the relation between Dex and miR-381 in cerebral ischaemic injury. Thus, the present study attempted to unravel the mechanism through which miR-381 enhances the neuroprotective effects of Dex in cerebral ischaemic injury.
Importantly, miR-381 has been demonstrated to possess a crucial role in the proliferation and differentiation of neural stem cells. 12 Further studies have proved that miR-381 restores the injury of cerebral ischaemia in rats and exerts neuroprotective function. 13 Moreover, the expression of mmu-miR-381 could be elevated after Dex treatment in lung injury. 14 Noticeably, a close relationship has been reported between miRNA and interferon regulatory factor (IRF) in the brain. For instance, miR-301 associated with IRF1 was indicated being implicated in the neuronal innate immune response. 15 Additionally, further studies have illustrated IRF4 as a transcriptional factor correlated with the inflammatory response in the brain and suggested it as a target for the injury of neonatal ischaemic brain. 16 Thus, in the present study, we hypothesized that miR-381 and IRF4 might have an association with a Dex-mediated protective effect against cerebral ischaemic injury in rats. Hence, our study aimed to investigate the impacts of Dex-regulated miR-381 and the relevant regulatory mechanism in cerebral ischaemic injury.

| Ethics statement
The study was conducted with the approval of the Animal Ethics Committee of Guizhou Provincial People's Hospital. All rats were treated by following the Guidelines for the Care and Use of Laboratory Animals proposed by the National Institutes of Health.

| Bioinformatics analysis
The downstream target genes of hsa-miR-381 were predicted by online websites StarBase, 17 mirDIP, 18 miRWalk 19 and TargetScan. 20 Human transcription factor names obtained from the Cistrome database. 21 The target genes of rno-miR-381 were obtained by miRDB (Target score ≥ 60), 22 miRWalk (Top 500 with the accessibility of binding relation) and TargetScan (Cumulative weighted context ++ score ≤ −0.05). In order to obtain more reliable target genes, Venn mapping was carried out to obtain the common genes predicted by online tools.

| Establishment of rat models
A total of 120 specific pathogen-free Sprague-Dawley rats (males; aged 3 months) were used in this study. Among them, 12 rats underwent sham-operation and 108 rats were used for model establishment. The MCAO model was established by following the previously reported procedures at room temperature (20 ± 5°C). 13,23 The fivepoint scale (Longa scoring system) introduced in previous work was utilized for the evaluation of the modelling, 24 that is 0 point: no neurological deficit; one point: unable to fully extend left forepaw; two points: circling to the left; three points: difficult to walk and leaning to the opposite side; and four points: unable to walk spontaneously and almost lost consciousness. The rats with score 0 and 4 were excluded from this study, while rats with score 1-3 were included.
The ischaemic part of the brain tissues was white, while the normal brain tissues were red or purple. The area of cerebral infarction was calculated.
After MCAO modelling, five rats were randomly selected and the brain tissues were removed. The area of cerebral infarction was detected by TTC staining to evaluate the success of MCAO modelling. TTC staining and infarct size calculation showed that MCAO rat model establishment was successful ( Figure S1A, B).

| Primary hippocampal neuron cell culture and oxygen-glucose deprivation (OGD) treatment
The primary neuron cells were isolated from the post-natal mice (n = 3) by following the instructions of Papain dissociation kit (#3150; Worthington Chemicals), followed by isolation of mixed neuron cells from the rat hippocampal tissues according to the protocols of the manufacturer. Then, cells were cultured and grown to confluence according to the detailed procedure established before. 26 Subsequently, the primary neuron cells for OGD treatment were cultured in serum-and glucose-free medium and then transferred into a box with 95% N 2 and 5% CO 2 for 6-hours incubation, while the normal control was placed in norm-oxygenated DMEM containing glucose.

| Cell grouping
Lentivirus vector LV5-GFP (#25999; Addgene Inc) was adopted for overexpression, while pSIH1-H1-copGFP (LV601B-1; System Biosciences) was utilized for silencing of cells. The cells were transfected with short hairpin (sh)-RNA targeting IRF4, sh-NC, miR-381 inhibitor, and the corresponding inhibitor NC separately or combinedly constructed by Shanghai GenePharma Co., Ltd.. Afterwards, the packaged virus and targeted vectors were cotransfected into HEK293T cells and incubated for 48 hours followed by the collection of supernatants. Then, viral particle was filtered by the centrifugation of supernatant and virus titre was determined. Following after, the HEK293T cell was cultured in RPMI-1640 medium containing 10% FBS and subcultured after every 2 days, and the virus in the exponential growth period was obtained. Meanwhile, control cells were treated or untreated with sh-NC lentivirus, sh-IRF4 lentivirus, oe-NC lentivirus, oe-IRF4 lentivirus or Dex, while OGD-treated cells were treated or untreated with Dex, oe-NC lentivirus + Dex, oe-IRF4 lentivirus + Dex, inhibitor NC lentivirus + Dex or miR-381 inhibitor lentivirus + Dex. Subsequently, cells were cultured at 37°C in 5% CO 2 for 48 hours.

| Modified neurological severity score (mNSS)
At 24 hours after MCAO modelling, mNSS was applied for the evaluation of the neurological function of rats at different timepoints according to the previously established procedure. 27 Tests included balance-beam, walking, abnormal behaviour, sensory, tail suspension tests and loss of reflex, respectively. The evaluation criteria were light functional deficit (1-6 points), moderate functional deficit (7-12 points) and severe functional deficit (13-18 points).

| Cardiac perfusion and fixation of the brain samples
The state of consciousness, general behaviour and physical activity of the rats were observed before the experiment, after the anaesthesia, after the surgery, during and after the modelling, respectively. After 24 hours of mNSS, 12 rats in each group were randomly selected for anaesthesia followed by fixing on the surgical plate to open the thoracic cavity. Then, 20-mL sterile syringe was carefully inserted into the left ventricle, while the right atrial appendage was cut and injected with sterile normal saline until the outflow liquid became clear. Thereafter, 4% paraformaldehyde (about 10-20 mL) was added. When the liver and limbs of rats turned white and stiff, the brain tissues were extracted from rats, observed by naked eyes, fixed in paraformaldehyde, and slices into tissue sections.

| Terminal deoxynucleotidyl transferase dUTP nick end labelling (TUNEL) staining
The brain tissues were dehydrated by gradient alcohol, cleared with xylene, embedded in paraffin and then followed by preparation of tissue sections. The tissue sections were then dewaxed by xylene for 10 minutes, soaked in 100%-75% gradient alcohol, and distilled water, each for

| Immunohistochemistry
Immunohistochemistry staining was performed 2 or 4 weeks after MCAO. The slices were pretreated with citric acid buffer for 5 minutes, then incubated with 5% normal goat serum for 1 hour, incu-

| Enzyme-linked immunosorbent assay (ELISA)
The IL-9 content of the brain tissues in the experimental group and the control group was measured by an ELISA kit (R&D Systems). The 10% tissue homogenate (1000 μL) was centrifuged at 4000 rpm/ min for 10 min at 4°C followed by the collection of the supernatant.
Then, the standard samples were added with 2 mL of distilled water to prepare a 20 ng/mL standard sample solution. For this purpose, eight standard tubes were set, in which the first tube was added with a 900 μL diluted sample solution, while the rest of the tubes were added with a 500 μL sample solution. The content in each tube was repeatedly diluted with the eighth tube set as a blank control.
Then, each well was added with 100 μL standard or test samples and placed on the reaction place at 37°C for 120 minutes. Each sample was detected following the instructions of the ELISA kit. The corresponding IL-9 content was determined on the curve based on the sample optical density (OD) value.

| Reverse transcription-quantitative polymerase chain reaction (RT-qPCR)
The total RNA was extracted by the TRIzol method, while the purity and concentration of the sample RNA were determined by a microplate reader (DNM-9606; Beijing Pulang New Technology Co., Ltd).
The total RNA was extracted using the RNeasy Mini Kit (Qiagen).  Table 1. The expression levels of mRNA in brain tissues were calculated by the 2 -ΔΔCt method. 30  by the collection of the supernatant. Hence, the luciferase activity was detected using a Dual-Luciferase Reporter Assay System (E1910; Promega).

| Western blot analysis
Total protein was extracted from 10% brain tissues homogenate in the experimental group and the control group, respectively, and then centrifuged at 4°C and 500 rpm for 15 minutes, followed by the collection of the supernatant. The protein concentration was determined by the Brad-ford method. Briefly, 30 μg sample protein was mixed well with immobilized pH gradient strip solution, while 10% sodium dodecyl sulphate-polyacrylamide gel electrophoresis gel was introduced for 2-hour electrophoresis.
Then, the protein was transferred to the polyvinylidene fluoride

| Flow cytometry
Flow cytometry was performed to detect the apoptotic rate of cells.
Specifically, the brain tissues were cut, ground, trypsinized and rinsed with phosphate buffer saline. The 1 × 10 3 cells/mL single cell suspension was prepared and stained with Annexin V-propidium iodide (PI). After that, a flow cytometer (Gallios; Beckman Coulter Inc) was used to read the fluorescence at the excitation wavelength of 488 nm to detect cell cycle progression.

| Statistical analysis
The SPSS 21.0 (IBM SPSS Statistics) was employed for statistical analysis. The measurement data were exhibited as the mean ± SD.
Data comparisons among multiple groups were performed using one-way analysis of variance (ANOVA), and the paired mean values among groups were compared with Tukey's post hoc test. Data among multiple groups at different time-points were compared by repeated-measures ANOVA followed by Bonferroni's post hoc test.
Non-parametric test (Mann-Whitney U test or Kruskal-Wallis test) was used to compare the neurological scores. The differences were statistically significant at P < .05.

| Dex ameliorates cerebral ischaemic injury in rats by activating miR-381
To investigate whether Dex could regulate cerebral ischaemic injury, the MCAO rat model was developed and injected with Dex  Figure 1L, M). Hence, the above-mentioned results suggested that Dex was able to up-regulate miR-381 expression, thereby improving the cerebral ischaemic injury.  Figure S3A; Table S1) were obtained. The target genes of rat miR-381 were predicted by the miRDB, miRWalk and TargetScan database, obtaining 403, 429 and 2225 downstream target genes, while fifteen target genes ( Figure S3B; Table S2) were obtained by the intersection of the predicted results from the above three databases using the Venn map.

| miR-381 could bind to IRF4 to improve cerebral ischaemic injury in rats
Accordingly, our results indicated only the presence of transcription factor IRF4 in prediction results of both human and rat miR-381. The binding sites between miR-381 and IRF4 3'UTR are shown in Figure   S3C.
Subsequently, the targeting relationship between miR-381 and IRF4 was verified by the dual-luciferase reporter gene assay and the results indicated that the MUT-IRF4 was the site of mutant S447 (S447A) and S448 (S448A), consistent with the previously reported work. 31 Additionally, our results in (Figure 2A

| Dex down-regulates IRF4-interleukin (IL)-9 by activating miR-381 in neuron cells
Previous study has shown that the transcription factor IRF4 can promote the expression of IL-9, 32    with Dex + sh-IRF4 was remarkably lowered ( Figure 3H). Besides, our results from ELISA and Western blot analysis presented that IL-9 expression was significantly decreased in cells after Dex or sh-IRF4 treatment ( Figure 3I, J), thus, suggesting that IL-9 expression down-regulated by Dex was dependent on IRF4. Collectively, these findings suggested that Dex could inhibit the IRF4-IL-9 by up-regulating the miR-381 in neuron cells.

| Dex ameliorates cerebral ischaemic injury by reducing the inflammatory response induced by IRF4-IL9
After

| D ISCUSS I ON
Insufficient flow of blood is considered as the major cause of cerebral ischaemic injury. 33 While previously reported studies have revealed the multiple functions of Dex treatment, for instance, Dex could effectively improve the outcome of cardiac surgery. 34 Intriguingly, Dex has been indicated to possess a neuroprotective role in rat cerebral ischaemic injury. 35 The fact that accumulating studies have reported the beneficial effects of Dex on the cerebral ischaemic injury, yet to date, its underlying molecular mechanisms remained elusive.
Thus, the present study unravelled that Dex could potentially induce highly expressed miR-381, which further results in the inhibition of inflammation response in rats with cerebral ischaemic injury.   Moreover, Dex has also been indicated to improve the long-term memory and learning ability in rats with the hypoxia/reoxygenation-induced cerebral injury, 6 while another study found that Dex could repair the brain ischaemic injury by promoting the extracellular signal-regulated kinase-related signalling pathway. 37 Besides, it has been shown that the Dex administration elevated the expression of five miRNAs, including miR-702-3p, miR-7a-2-3p, miR-3596d, miR-496-5p and miR-434-3p in heart of rats, thus indicating the correlation between Dex and miRNAs. 38 Additionally, in lung injury, Dex could elevate the expression of miR-381 in modelled mice. 14 While miR-381 has reported to exhibiting protective effects on peripheral neuropathy. 39 Thus, the repairing effect of miR-381 in cerebral ischaemic injury has been validated before. 13

F I G U R E 4
Dex reduces the inflammatory response by down-regulating the IRF4-IL-9 to alleviate cerebral ischaemic injury. A-C, The mRNA expression of IRF4 and IL-9 in OGD-treated neuron cells treated with Dex and oe-IRF4 detected by RT-qPCR, ELISA and Western blot analysis, respectively. D, Apoptosis of OGD-treated neuron cells treated with Dex and oe-IRF4 measured by flow cytometry. E-G, The mRNA expression and protein expression of IRF4 in brain tissues of MCAO rats treated with Dex and oe-IRF4 assessed by RT-qPCR, ELISA and Western blot analysis, respectively. H-J, The expression of IRF4 in brain tissues of MCAO rats observed by RT-qPCR, ELISA and Western blot analysis. K, mNSS evaluation on the neurological function of MCAO rats. L, M, Cell apoptosis in brain tissues of MCAO rats detected by TUNEL staining (scale bar = 100 μm). N, Cell apoptosis in brain tissues of MCAO rats using flow cytometry. O, BrdU-positive cell rate was detected by BrdU. P, Ki-67 staining to detect the number of Ki-67-positive expression. *P < .05. The above-mentioned measurement data were expressed as the mean ± SD, and compared using one-way ANOVA and Tukey's post hoc test. Non-parametric test (Kruskal-Wallis test) was used to compare the neurological scores. n = 12. Each cellular experiment repeated three times our results from Western blot analysis and RT-qPCR determined the elevated expression of IRF4 in MCAO rats; however, it was declined after the addition of miR-381 mimic, thus, reflecting that IRF4 was negatively correlated with miR-381. Accordingly, a previous study has reported the overexpressed miRNA, namely, miR-30b/d/e could notably inhibit the expression of IRF4 in plasma cell differentiation. 42 Besides, miR-125b has been reported to represses the IRF4 expression and results in the induction of B cell leukaemia and myeloid. 43 Furthermore, our study revealed that IRF4 overexpression abolished the inhibitory effect of miR-381 mimic on cell apoptosis and neurological function damage in MCAO rats. Additionally, it has been reported that IRF4-positive is highly expressed in the ischaemic brain, which promoted the expression of inflammatory factor interleukin-17. 44 Finally, the most important finding of our study depicted that Dex reduced the inflammatory response by down-regulating the IRF4-IL-9 via miR-381 to alleviate cerebral ischaemic injury.
Though previously reported study has shown the correlation between cerebral ischaemic injury, inflammation, cell apoptosis, and oxidation. 45 Noticeably, accumulating studies have shown the role of IL-9 in various types of inflammatory processes. 46 Thus, IL-9 has been verified as a potential pathogenic factor in ischaemic stroke. 47 Besides, it has been also indicated that IRF4 promotes the expression of IL-9 in a dose-dependent way in human and mouse T helper nine cells. 32 However, the participation of Dex could suppress the inflammatory response induced by lipopolysaccharide, 48 while the anti-inflammation effect of miR-381 on macrophages has also been reported. 49

| CON CLUS ION
In summary, our study suggested the protective effects of Dex exerts on cerebral ischaemic injury in vivo, which may account for its involvement in the regulation of inflammatory response by inhibiting IRF4-IL-9 via miR-381 inhibitor ( Figure 5). Intriguingly, the data presented in this study could be of clinical importance, thus, suggesting the therapeutic application of Dex as a neuroprotective agent.
Nevertheless, the limitation remains. Additionally, the dosage of Dex used in the present study is applicable for rats only and cannot be used for the clinical practice. Therefore, further characterization of Dex-mediated mechanisms is a prerequisite in future research.

ACK N OWLED G EM ENTS
We would like to acknowledge the reviewers for their helpful comments on this study.

CO N FLI C T O F I NTE R E S T
The authors declare no conflicts of interest. Funding acquisition (equal); Writing-review & editing (equal).

DATA AVA I L A B I L I T Y S TAT E M E N T
Research data are not shared.