The phosphokinase activity of IRE1ɑ prevents the oxidative stress injury through miR‐25/Nox4 pathway after ICH

Abstract Background Endoplasmic reticulum (ER) stress and oxidative stress are the major pathologies encountered after intracerebral hemorrhage (ICH). Inositol‐requiring enzyme‐1 alpha (IRE1α) is the most evolutionarily conserved ER stress sensor, which plays a role in monitoring and responding to the accumulation of unfolded/misfolded proteins in the ER lumen. Recent studies have shown that ER stress is profoundly related to oxidative stress in physiological or pathological conditions. The purpose of this study was to investigate the role of IRE1α in oxidative stress and the potential mechanism. Methods A mouse model of ICH was established by autologous blood injection. The IRE1α phosphokinase inhibitor KIRA6 was administrated intranasally at 1 h after ICH, antagomiR‐25 and agomiR‐25 were injected intraventricularly at 24 h before ICH. Western blot analysis, RT‐qPCR, immunofluorescence staining, hematoma volume, neurobehavioral tests, dihydroethidium (DHE) staining, H2O2 content, brain water content, body weight, Hematoxylin and Eosin (HE) staining, Nissl staining, Morris Water Maze (MWM) and Elevated Plus Maze (EPM) were performed. Results Endogenous phosphorylated IRE1α (p‐IRE1α), miR‐25‐3p, and Nox4 were increased in the ICH model. Administration of KIRA6 downregulated miR‐25‐3p expression, upregulated Nox4 expression, promoted the level of oxidative stress, increased hematoma volume, exacerbated brain edema and neurological deficits, reduced body weight, aggravated spatial learning and memory deficits, and increased anxiety levels. Then antagomiR‐25 further upregulated the expression of Nox4, promoted the level of oxidative stress, increased hematoma volume, exacerbated brain edema and neurological deficits, whereas agomiR‐25 reversed the effects promoted by KIRA6. Conclusion The IRE1α phosphokinase activity is involved in the oxidative stress response through miR‐25/Nox4 pathway in the mouse ICH brain.

Intracerebral hemorrhage (ICH) is a subtype of stroke, which accounts for 10%-15% of all strokes characterized by high mortality and morbidity. 1,2A series of pathophysiological processes, such as brain edema, blood-brain barrier disruption, neuroinflammation, mitochondrial dysfunction, neuronal apoptosis, etc., are triggered by the initial mass mechanical effect of intracerebral hematoma and intrahematoma hemolytic products in ICH brain. 3,4Recently, myriad studies started to focus on the role of various stress events in adaptive response and adverse reactions in neurological diseases.It has been established that endoplasmic reticulum stress (ER stress) and oxidative stress are involved in the complicated pathophysiological processes after ICH insult. 5,6However, the complicated interaction between the two stress responses in ICH brain remains to be revealed.
The endoplasmic reticulum (ER) is a multifunctional organelle and is mainly responsible for proteins synthesis and folding as well as calcium storage.Some unbalanced factors, such as ischemia, hypoxia, and pH changes, will disturb the homeostasis of the ER and lead to unfolded or misfolded proteins accumulating in the ER lumen, resulting in ER stress. 7,8In adaptive ER stress conditions, the unfolded protein response (UPR), a set of tightly controlled regulatory programs, is activated to reconstruct the protein-folding homeostasis and to orchestrate the recuperation of ER function. 7However, under overwhelming or prolonged ER stress, UPR fails to restore the normal function of ER, and the apoptotic cascade will be induced in cells. 9UPR is activated through three ER transmembrane sensors, protein kinase RNA (PKR)-like ER kinase (PERK), activating transcription factor 6 (ATF6), and inositol-requiring protein 1 (IRE1). 10,11Among them, IRE1 is the most conserved and the oldest branch one with two isoforms, IRE1α and IRE1β.As an endoplasmic reticulum transmembrane protein, IRE1α is expressed ubiquitously with the function of both a kinase and a ribonuclease (RNase).The mRNA encoding the transcription for X-box binding protein-1 (XBP1) could be cleaved by ribonuclease activity, resulting in the production of spliced XBP1 (XBP1s).XBP1s improves protein-folding capacity and misfolded protein degradation in the ER lumen. 12In addition to initiating the ER stress survival pathway, sustained activation of IRE1α can induce downstream apoptotic signaling molecules like caspase-2 and CHOP/GADD153 to cause cell apoptosis. 13,14For this reason, the IRE1α activity is a key crossroad for the UPR signaling pathway switching from an adaptive response to negative consequences.
In addition to those coding RNAs such as XBP1 mRNA, the IRE1α RNase can also lead to the cleavage of a variety of noncoding RNAs including miRNAs. 157][18] It was found that miR- 25-3p   is involved in many pathological and physiological processes of the nervous system, including regulating neuroinflammation, neural stem cell proliferation, etc. 19,20 Recently, the role of miR-25-3p in regulating oxidative stress has attracted widespread attention from researchers.For example, miR-25-3p took antioxidant effects on primarily cultured neurons through targeting OXSR1, 21 and cardiomyocytes could be protected by miR-25-3p against oxidative damage via downregulating mitochondrial calcium uniporter. 22idative stress is a series of responses to reactive oxygen species (ROS) increasing in the body caused by the excessive ROS generation or the weaken of ROS scavenging system.Nicotinamide adenine dinucleotide phosphate oxidase 4 (Nox4) is a subtype of the Nox family, which is one of the main sources of ROS in mammals. 23r searching results in the miRNA target prediction software showed that miR-25-3p contains nucleotide sequences complementary to Nox4 mRNA 3'-UTR sequences (see Results section), which indicates that miR-25-3p is likely a regulator of Nox4 expression.At the same time, some studies confirmed that miR-25-3p is involved in the regulation of Nox4 expression in hydroquinone-treated leukemia cells and diabetic peripheral neuropathy. 24,25In view of the above reasons, the role of miR-25-3p on regulating oxidative stress may be realized via Nox4 in ICH brain, which is waiting for being revealed.
Given that a large body of miRNAs are cleaved by activated IRE1α, miR-25-3p is likely one of the cleavage substracts of IRE1α RNase activity.Considering the role of miR-25-3p on oxidative stress, miR-25-3p is probably an important factor connecting with IRE1α activation during ER stress and Nox4 activation during oxidative stress following ICH brain injury.We will explore whether could IRE1α activation induce oxidative stress via miR-25-3p/Nox4 pathway after ICH insult in the present study.

| Experimental animals
A total of 287 Kunming mice (8-10 weeks, weight 25-30 g) were purchased from the Animal Center of Chongqing Medical University.Among them, 15 mice were excluded due to death during surgery.Table 1 displays the utilization of our different mouse groups and the mortality rate among the mice.All mice were provided with standard rodent food and water and accommodated in an air-conditioned environment with a 12 h light/dark cycle.All procedures were approved by the Animal Ethics Committee of Chongqing Medical University.

| Experimental design
Four separate experiments were designed in the following way (Figure 1).

| Experiment 1
In order to determine the time course of endogenous IRE1α, p-IRE1α, miR-25-3p, and Nox4 expression levels in the right cerebral hemisphere after ICH, reverse transcription quantitative real-time polymerase chain reaction (RT-qPCR) and western blot analysis were performed.According to different time points after ICH, the mice were distributed randomly and equally into six groups: the sham group and the ICH groups (6 h, 12 h, 1 d, 3 d, 7 d).Furthermore, immunofluorescence double staining was used to determine the cellular localization of p-IRE1α, respectively, with the marker of astrocytes (glial fibrillary acidic protein, GFAP), the marker of neurons (neuronal specific nuclear protein, NeuN) and the marker of microglia (ionized calcium-binding adaptor molecule 1, Iba-1).The mice were euthanized at 12 h post-ICH and brain samples were collected for immunofluorescence staining.
Western blot analysis was performed at 12 h after ICH.

| Experiment 3
In order to determine the effects of KIRA6 on the level of miR-

| Experiment 4
In order to determine whether miR-25-3p is involved in the relationship between IRE1α and oxidative stress, the antago-miR-NC/antagomiR-

| ICH model establishment
7][28] The mice were anesthetized and secured in a stereotactic apparatus.A longitudinal incision was made to expose the skull.The stereoscopic coordinates were 2.3 mm lateral and 0.2 mm anterior to the bregma.A total of 30 μL autologous blood drawn from the tail was injected into the hole by a microinjection syringe.The blood injection rate during the whole injection process was 2 μL/min.After injection, the needle was retained for 10 min and slowly withdrawn at a rate of 1 mm/min.The tiny hole was then plugged with bone wax and the wound is stitched with sutures.
For sham groups, the operation procedures were performed with needle insertion only.After surgery, rectal temperature was maintained at 37.0 ± 0.5°C through the use of a temperature-controlled heating pad until the mice regained consciousness.Subsequently, the mice were housed in a temperature-controlled environment at a constant room temperature of 22°C.Mice body weight were monitored at three time points: 1 hour pre-ICH, 1 hour post-ICH, and 1 day post-ICH.

| Intranasal administration
KIRA6 was dissolved in dimethyl sulfoxide (DMSO) as the stock solution.After being anesthetized, the mice were placed in a supine position.A total volume of 5 μL KIRA6 (diluted to the final dose) or vehicle was divided into four drops and applied into both sides of the nostril alternatively.Each side nostril was administered 1.25 μL KIRA6 or vehicle each time with 2 min intervals.

| Intracerebroventricular injection
To investigate the effects of miR-25-3p on oxidative stress after ICH, antagomiR-NC/antagomiR-25 (0.5 nmol/mice, B05002, GenePharma) or agomiR-NC/agomiR-25 (0.5 nmol/mice, B06002, GenePharma) were administrated into the lateral ventricle at 24 h before ICH.Briefly, a tiny hole was made in the parietal region (1.7 mm deep, 1.5 mm posterior, and 1.5 mm lateral) as previously described, 28  a flow rate of 2 μL/min.After injection, the needle was retained for 10 min and slowly withdrawn at a rate of 1 mm/min.The tiny hole was then plugged with bone wax and the wound is stitched with sutures.

| Neurological score assessment
After the establishment of the mouse model a series of behavioral tests, including the composite Garcia, corner turn test, MWM, and EPM were performed.The composite Garcia test and the corner turn test were used to assess neurological function at 1 day post-ICH.In the assessment process, we adhere to the principle of blind evaluation.The composite Garcia test consists of seven elements, which are spontaneous activity, climbing, vibrissae proprioception, lateral turning, symmetry of limb movement, forelimb walking, and axial sensation.In the corner turn test, mice were placed in a 30° corner with their heads toward the corner and were allowed to take either a right or left turn to exit the corner, the right or left turn times were recorded.These trials were performed 10 times per day, with intervals of at least 30 s between each test.
The number of (right turns/all turns 100%) was used to calculate the score. 29,30ter the establishment of the mouse model for 4 weeks, the MWM was used to test the spatial learning and memory abilities of rodents. 31Prior to the commencement of the tests, the mice Prior to each test, the subjects were cleaned with 75% alcohol to remove any olfactory cues.

| Reverse transcription quantitative real-time polymerase chain reaction (RT-qPCR)
Total mRNA and miRNAs were extracted from the right cerebral hemisphere using the RNAiso Plus reagent (TaKaRa, Japan).RT-qPCR detection was performed using the PrimeScriptTM RT Reagent kit and the Green® Premix Ex TaqTM II (Tli RNaseH Plus) kit (Takara, Japan).Ultimately, miR-25-3p and Nox4 mRNA expression were calculated by the 2 −ΔΔct method.The sequence of RT-PCR primers is listed in Table 2.

| Preparation of frozen sections
As previously described, 33 the brain samples were perfused with 4% paraformaldehyde and then dehydrated in gradients of 10%, 20%, and 30% sucrose solution.The samples were divided into 10 μm thick coronal sections, and then we collected consecutive sections within a range of 200 μm both anterior and posterior to the injection site.The well-prepared frozen sections are stored in a −20°C refrigerator for immunofluorescent staining, HE staining, and Nissl staining.

| DHE Staining
The DHE staining method was used for the detection of ROS content in brain tissue.DHE (10 μmol/L) was injected intraperitoneally into the ICH model for 30 min before the mice were sacrificed. 34The brains were then sliced and placed on glass slides.The sections were incubated with DHE (S0063, Beyotime) in a dark container at 37°C for 30 min.Following rinsing of the excess reagent and staining of the nuclei, the fluorescent pictures were taken under a fluorescence microscope.To quantify the immunostaining intensity, we utilized the Integrated Density analysis function of Image J software (ImageJ, Germany).

| HE staining
The stored frozen sections were rewarmed for 5 min, washed with water, stained with hematoxylin for 5 min, fractionated for 10 s, and stained with eosin for 1 min.The sections were then rinsed in 75% ethanol, 85% ethanol, 95% ethanol, and anhydrous ethanol for 2-3 s, transparented in xylene, sealed with neutral gum, and observed under a microscope.

| Nissl staining
The stored frozen sections were rewarmed and washed in water for 2-4 s.The sections were then stained with a 1% aqueous toluidine blue solution for 5 min, washed in water, and rinsed in 70% ethanol, 95% ethanol, and anhydrous ethanol, respectively, transparented in xylene, sealed with neutral gum, and observed under a microscope.
And then, we employed the automated cell counting function of Image J software (Image J,Germany) to calculate the number of surviving neurons.

| H 2 O 2 content
The mice were deeply anesthetized and transcardially perfused with PBS.The right cerebral hemisphere was then removed and mixed with 0.9% NaCl at a ratio of 1:

| Hematoma volume
According to the manufacturer's instructions, the free hemoglobin assay reagent (A071-1-1, Jiancheng) was added to the prepared tissue supernatants and then incubated at 37°C.The absorbance was measured at 510 nm using a spectrophotometer.

| Brain water content
BWC was determined at 1 day after ICH based on the dry-wet weight method as previously described.Briefly, the brain samples were collected and divided into five sections: ipsilateral and The sequence of primers used in RT-qPCR.

| Statistical analysis
Statistical analyses were performed using the GraphPad Prism software (GraphPad Software, USA).Data were presented as means ± SD.The Shapiro-Wilk test was assessed the normal distribution of data within each group.Homogeneity of variance was evaluated using F test for unpaired t-test analysis between two groups.
Homogeneity of variance was evaluated using Brown-Forsythe test for one-way ANOVA between three and more groups, followed by multiple comparisons using Tukey's post hoc test.p < 0.05 was regarded statistically significant.As presented in Figure 3C,F, the fluorescence intensity of p-IRE1α increased in microglia at 12 h after ICH (p < 0.05).These results indicated that p-IRE1α was extensively expressed in astrocytes, neurons, and microglia, and its expression increased in various cell types after ICH.

| KIRA6 dose-dependently inhibited p-IRE1α/IRE1α protein expression, increased hematoma volume and neurological function after ICH
The optimal effective dose of KIRA6 in the ICH mouse model was (p < 0.05, Figure 4A,B).
The hematoma volume data revealed that the hematoma volume in the ICH + vehicle group were higher than that in the sham group.Three doses of KIRA6 could effectively increase the hematoma volume compared with the ICH + vehicle group, and the increasing effect was KIRA6 dose-dependent (p < 0.05, Figure 4C,D).
The Garcia score decreased in the ICH + vehicle group compared with the sham group.KIRA6 could further decrease the score compared with the ICH + vehicle group, and the decreasing effect was dose-dependent on KIRA6.(p < 0.05, Figure 4E).
The number of right turns increased after ICH compared with the sham group.KIRA6 could further increase the number compared with the ICH + vehicle group, and the increasing effect was dose-dependent on KIRA6.(p < 0.05, Figure 4F).
Considering alleviating the additional distress to mice, the low dose (5 mg/kg) of KIRA6 was selected for further experiments in this study.

| Administration of KIRA6 downregulated the expression level of miR-25-3p after ICH
To verify the relation between IRE1α and miR-25-3p in the ICH mouse model, RT-qPCR was performed following KIRA6 administration.The results demonstrated that miR-25-3p expression was increased after ICH compared with the sham group, whereas the expression level of miR-25-3p was suppressed by KIRA6 administration (p < 0.05, Figure 5A).Meanwhile, the results demonstrated that the IRE1α activity and miR-25-3p expression level changed toward the same direction after ICH and IRE1α inhibition, which indicated that miR-25-3p is not the cleavage substrate of IRE1α RNase.This conclusion is not consistent with our initial deduction that IRE1α activation may lead to the decay of miR-25-3p.

| Administration of KIRA6 upregulated Nox4 protein expression, oxidative stress, brain injury, reduced body weight, exacerbated spatial learning and memory deficits, and increased anxiety levels after ICH
To determine whether the administration of KIRA6 has an effect on Nox4 and Nrf2 expression levels, western blot was performed.
It was found that the expression level of Nox4 did not change markedly after ICH compared with the sham group, while administration of KIRA6 increased the level of Nox4 expression.Nrf2 is a key transcription factor in the biological antioxidant defense system.[37] The expression level of Nrf2 was higher after ICH compared with the sham group, while administration of KIRA6 suppressed the expression level of Nrf2.(p < 0.05, Figure 5B-D).
As one of the downstream products of Nox4, H Brain water content was higher in the ICH + vehicle group compared with the sham group, and it continued to increase after administration of KIRA6 (p < 0.05, Figure 5H).On the contrary, a part of neurons were slightly atrophy and degeneration at 12 h after ICH, and KIRA6 aggravated the pathological changes.
Nissl staining of the brain tissue showed that there was a large number of neurons in the sham group, with clear nuclei and lightly staining.The number of surviving neurons at 12 h after ICH was slightly decreased compared with the sham group, but this difference was not statistically significant.However, the administration of KIRA6 significantly decreased the number of surviving neurons (p < 0.05, Figure 6B-E).
In order to assess the spatial learning ability, the Morris water maze test was used.During the spatial acquisition phase, as shown in Figure 7A-C, the swimming paths of the sham group, ICH + vehicle group and ICH + KIRA6 group on the fifth day were shown.
The results showed no differences in spatial bias or basal swimming speed (Figure 7D).We also measured the escape latency, which is the time spent by the mice to find the hidden platform.The escape latency of the mice in all three groups decreased over the five consecutive training days, but the escape latency of the ICH + vehicle was still significantly higher compared to the sham group at Day 5, and ICH + KIRA6 group was also significantly higher compared to the ICH + vehicle group (Figure 7E, p < 0.05).In the probe test, when the platform was removed, the time spent in the platform zone decreased for each group compared to the sham group (p < 0.05), In order to assess the impact of KIRA6 on anxiety levels in mice after cerebral hemorrhage, the elevated plus maze test was utilized.
We recorded the number of entries into the open arms and the per-

| Nox4 gene was a direct target gene of miR-25-3p
To investigate whether Nox4 is a target gene of miR-25-3p, two online target prediction software, Targetscan and miRDB, were searched.It was found that miR-25-3p contains nucleotide sequences complementary to the 3′-UTR sequence of Nox4 mRNA (Figure 8A).

| DISCUSS ION
IRE1α is an ER transmembrane protein, which contains an N-terminal ER luminal domain responsible for stress sensing and C-terminal kinase and endoribonuclease domain in the cytosol. 38The accumulation of unfolded/misfolded proteins triggers the inactive mono- and microRNAs. 38The spatial and temporal expression patterns of IRE1α in ICH mouse brain were detected in the current study.The results showed that IRE1α is widely expressed on astrocytes, neurons, and microglia in various brain regions, which is consistent with other previous research results that IRE1α is expressed extensively in various mammalian cells. 38The time course results presented the ratio of p-IRE1α/ IRE1α relative expression level increased in a timedependent manner, which indicated that the phosphokinase activity is activated in ICH mouse brain.
We initially speculated before the experiment that KIRA6, the Although increasingly studies have paid attention to the regulation role of miR-25-3p on the extent of oxidative stress, 24,39,40 there ) and stable nonradical oxidants such as H 2 O 2.

23,41
The brain is highly vulnerable to oxidative injury due to the high shown that ER stress can affect the production of ROS through a series of complex ways. 43,44For example, one of the ER stress markers, C/EBP homologous protein (CHOP), could induce the activation of endoplasmic reticulum oxidoreductase (ERO1α). 45,46en the activated ERO1α promotes the ROS production in ER by the transfer of electrons from the enzyme protein disulfide isomerase (PDI) to molecular oxygen, 43,47 and induces Ca 2+ release by stimulating various receptors such as inositol-1,4,5-trisphosphate receptor (IP3R) or RyRs. 48,49At last, the mitochondrial metabolism and production of ROS will be stimulated by the ROS produced from the ER and the released Ca 2+ . 50,51Currently, the conclusion about whether ER stress promotes or inhibits ROS production in oxidative stress injury is controversial in numerous research.3][54] The contradictions in these conclusions may be in that some studies just focused on one or some specific pathways among those numerous pathways activated in ER stress, some of which play protective roles while some others aggravate oxidative stress.So the crosstalks between ER stress and oxidative stress after ICH are intricate and requires 25-3p, Nox4, and Nrf2 expression, oxidative stress, neurobehavior function, brain edema, body weight after ICH, mice were distributed randomly into three groups: sham, ICH + vehicle, ICH + KIRA6 (5 mg/kg).RT-qPCR, Western blot, dihydroethidium (DHE) staining, Hematoxylin and Eosin (HE) staining, Nissl staining, and hydrogen peroxide (H 2 O 2 ) content analysis were performed at 12 h after ICH.Brain water content (BWC) and body weight were estimated at 1 day post-ICH.Morris Water Maze (MWM) and Elevated Plus Maze (EPM) were performed at 4 weeks after ICH.

| 5 of 20 LIAO
and the antagomiR-NC/antagomiR-25 or agomiR-NC/ agomiR-25 (10 μL) solution was injected into the lateral ventricle at F I G U R E 1 Four-part experimental design.et al.
are acclimated in a habituation room to familiarize themselves with the environment for 1 h.It consists of a circular container with a diameter of 120 cm and a depth of 50 cm.The container is filled with nontoxic brown dye-stained circulating water at a constant temperature, maintained at room temperature (20 ± 1°C).The maze is designated with two imaginary axes, where each line divides the maze into four equal quadrants vertically to the other line.The endpoints of each line divide the maze into four reference points: east, south, west, and north.A platform is located in the center of one of the quadrants, submerged 0.5 cm below the water surface.Each mouse is gently released into the maze from the expected starting position, facing away from the container walls.To promote the development of spatial memory, the starting point is different each day.After a 24 h open field test, each mouse undergoes four consecutive training trials per day, with a 20 min interval, lasting for 5 days, in order to locate the hidden platform.If a mouse reaches the hidden platform and stays there for at least 2 s, or if it fails to find the platform within 60 s, the training trial ends.Mice that fail to find the platform within the designated time are guided to the platform and allowed to stay there for 10 s.After the session, the mice are removed, dried, and returned to their home cages.Within 24 h after the last training trial, the platform is moved for a probe test.The mice are released in a new position to ensure the use of spatial memory rather than a specific swimming path, and they are allowed to freely swim for 60 s.The time spent in the target quadrant and the number of crossings over the platform are recorded as variables for learning and memory abilities.The escape latency and first entry into the target (platform location area), target crossings, swim speed (cm/s), and time spent in the target quadrant during the probe test are recorded using an automated tracking system during the initial training.After the establishment of the mouse model for 4 weeks, the EPM test was commonly used to assess the locomotor activity, exploration, and anxiety-like behavior in rodents.32Prior to the commencement of the tests, the mice are acclimated in a habituation room to familiarize themselves with the environment for 1 h.The EPM is made of opaque blue plastic material and consists of two open arms (30 × 6 cm) opposite to each other and perpendicular to two enclosed arms (30 × 6 cm), elevated 30 cm above the floor.Each mouse was placed near the center area, avoiding visibility of the enclosed arms, and allowed to freely explore for 5 min.A video tracking system automatically recorded the mouse's movement speed and percentage of time spent on the open and enclosed arms, as well as the number of entries into the center area during the 5 min session.

3 | RE SULTS 3 . 1 |
Figure 3B,E, the fluorescence intensity of p-IRE1α increased in neurons in ICH-12 h group compared with that in sham group (p < 0.05).

2 O 2
Figure 5E-G, the intensity of DHE fluorescence and H 2 O 2 content was slightly increased in the ICH + vehicle group compared with the sham group, but the difference between the two groups was not statistically significant.KIRA6 dramatically enhanced the fluorescence intensity of DHE and H 2 O 2 contents compared with the ICH + vehicle group (p < 0.05).

F I G U R E 2
The expression levels of p-IRE1α, IRE1α, Nox4, and miR-25-3p at different time points after ICH.(A) The protein bands of pIRE1α, IRE1α, and Nox4.(B,C) Quantification of the p-IRE1α/IRE1α and Nox4 protein levels at various time points (n = 4).(D,E) Quantification of the Nox4 mRNA and miR-25-3p expression levels at various time points (n = 4).Data are the mean ± SD. *p < 0.05 vs. sham group.F I G U R E 3 (A-C) Representative immunofluorescent images showing the co-localization of p-IRE1α with GFAP, NeuN, and Iba-1, respectively, around the hematoma in sham and ICH-12 h group (n = 4).The dotted line indicates the approximate boundary of the hematoma.The arrows point to co-expression cells.The details are zoomed into the rightmost column.Scale bar = 20 μm.(D-F) Quantification of the intensity of p-IRE1α (green fluorescence) in astrocytes/neurons/microglia in sham and ICH-12 h group by Image J software (n = 4).Data are presented as mean ± SD. *p < 0.05 vs. sham group.We found that the body weight of mice slightly decreased in the ICH + vehicle group and the ICH + KIRA6 group compared with the Sham group 1 h after ICH, but the difference between the three groups was not statistically significant.However, 1 day after ICH, the body weight of mice in the ICH + vehicle group was significantly lower compared to the Sham group (p < 0.05, Figure 5I), and KIRA6 further reduced the body weight compared to the ICH + vehicle group (p < 0.05, Figure 5I).To assess the pathological changes and neuronal damage in the brain tissue in ICH mice, HE staining and Nissl staining were performed.As shown in Figure 6A, HE staining of the brain tissue showed that neurons have clear nuclei and distinct nucleoli in the sham group.

F I G U R E 5
Effects of intranasal administration of KIRA6 on the expression levels of miR-25-3p and Nox4, oxidative stress, and brain edema.(A) RT-qPCR analysis of miR-25-3p expression after KIRA6 administration (n = 4).Values are as mean ± SD. *p < 0.05 vs. sham group; # p < 0.05 vs. ICH + vehicle group.(B) Representative western blot bands of Nox4 and Nrf2.(C) Quantitative analysis of the relative expression levels of Nox4 (n = 4).*p < 0.05 vs. ICH + vehicle group.(D) Quantification of the Nrf2 protein levels (n = 4).*p < 0.05 vs. sham group; # p < 0.05 vs. ICH + vehicle group.(E) Representative DHE staining images around the hematoma.The boundary of the hematoma is indicated by the dotted line.Scale bar = 100 μm.(F) Quantification of the intensity of DHE (red fluorescence) by Image J software (n = 4).(G) The measurement of H 2 O 2 contents (n = 4).Values are presented as mean ± SD. *p < 0.05 vs. ICH + vehicle group.(H) The analysis of brain edema in sham, ICH + vehicle, and ICH + KIRA6 groups (n = 4).(I) The bodies weight was evaluated in the groups of Sham, ICH + vehicle, and ICH + KIRA6 (n = 8).Data are the mean ± SD. *p < 0.05 vs. sham group; # p < 0.05 vs. ICH + vehicle group.but the decrease in the ICH + vehicle group was not statistically significant (Figure 7F).Additionally, compared to the sham group and ICH + vehicle group, the ICH + KIRA6 group spent significantly less time in the target quadrant (p < 0.05, 7G).These results indicate that KIRA6 exacerbates spatial learning and memory impairments in mice after cerebral hemorrhage.

F I G U R E 6
Figure7H-I).

F I G U R E 7
Effects of intranasal administration of KIRA6 on spatial learning and memory abilities and anxiety levels.(A-C) A representative swimming path diagram of each group at Day 5 (n = 8).The green color represents the starting point, and the red color represents the endpoint.(D) The Morris speed at Day 1 in the groups of Sham, ICH + vehicle, and ICH + KIRA6 (n = 8).(E) Escape latency to the platform from day one to day five of each group (n = 8).Values are presented as mean ± SD. *p < 0.05 vs. Sham group; # p < 0.05 vs. ICH + vehicle group.(F) Number of times entering the platform during the probe test (n = 8).Values are presented as mean ± SD. *p < 0.05 vs. ICH + vehicle.(G) Duration spent in the targeted quadrant during the probe test (n = 8).(H) Number of times entering the open arm (n = 8).(I) Percentage of time spent in open arms (n = 8).(G-I) Values are presented as mean ± SD. *p < 0.05 vs. Sham group; # p < 0.05 vs. ICH + vehicle group.

F I G U R E 9
meric state IRE1α dimerization upon ER stress, which activates the kinase and IRE1α self-phosphorylation. Then the self-phosphorylation induces a conformational change and leads to the activation of IRE1α RNase activities, which take action on the excision of the mRNA encoding the transcription of X-box binding protein-1 (XBP1) or other multiple RNAs, including mRNAs, ribosomal RNAs, Effects of miR-25-3p antagomir or agomir on oxidative stress.(A) Representative DHE staining images around the hematoma in ICH + KIRA6 + antagomiR-NC and ICH + KIRA6 + antagomiR-25 groups.The boundary of the hematoma is indicated by the dotted line.Scale bar = 100 μm.(B) Quantification of the intensity DHE (red fluorescence) by Image J software (n = 4).(C) The measurement of H 2 O 2 contents in ICH + KIRA6 + antagomiR-NC and ICH + KIRA6 + antagomiR-25 groups (n = 4).Values are presented as mean ± SD. *p < 0.05 vs. ICH + KIRA6 + antagomiR-NC group.(D) Representative DHE staining images around the hematoma in ICH + KIRA6 + agomiR-NC and ICH + KIRA6 + agomiR-25 groups.The boundary of the hematoma is indicated by the dotted line.Scale bar = 100 μm.(E) Quantification of the intensity DHE (red fluorescence) by Image J software (n = 4).(F) The measurement of H 2 O 2 contents in ICH + KIRA6 + agomiR-NC and ICH + KIRA6 + agomiR-25 groups (n = 4).Data are the mean ± SD. *p < 0.05 vs. ICH + KIRA6 + agomiR-NC group.
inhibitor of IRE1α phosphokinase, can inhibit the RNase activity by inhibiting the IRE1α self-phosphorylation and block the cleavage of miR-25-3p, so as to upregulate the expression level of miR-25-3p.However, our PCR results here showed that the expression level of miR-25-3p is decreased by KIRA6 instead of being upregulated.This result is conflicting to our expectations and suggests that the kinase and RNase activities of IRE1α are uncoupled completely, in other words, the downstream effects of kinase activation include RNase activation, but are not limited to RNase activation.The downregulation of miR-25-3p by kinase inhibitor KIRA6 in ICH mouse brain should be through a pathway unrelated to RNase activity blockage.In our previous experiments,28 the IRE1α RNase inhibitor STF083010 upregulated the expression of miRNA-125-b-2-3p and promoted the recovery of neurological function in ICH mice brain, which contradicts the results in the current study, that the IRE1α kinase inhibitor KIRA downregulated the expression of miRNA-25-3p and decreased the neurological function score of ICH model mice.The two kinds of IRE1α inhibitors, STF083010 and KIRA6, causing opposite outcomes in miRNAs expression may be due to that miRNA-125-b-2-3p and miRNA-25-3p are controlled by different kinase downstream pathways: miRNA-125-b-2-3p expression is mainly regulated by the RNase activity, so the inhibition of RNase activity by STF083010 will leads to the blockage of miRNAs cleavage and upregulated the expression of miRNA-125-b-2-3p as expected; while miRNA-25-3p is regulated principally by other kinase downstream phosphorylation pathway after kinase inhibitor KIRA6 administration and be F I G U R E 1 0 Effects of miR-25-3p antagomir or agomir on brain edema, hematoma volume, and neurological function.(A) The composite Garcia score were evaluated in the ICH + KIRA6 + antagomiR-NC and ICH + KIRA6 + antagomiR-25 groups (n = 4).(B) The corner turn times were evaluated in ICH + KIRA6 + antagomiR-NC and ICH + KIRA6 + antagomiR-25 groups (n = 4).(C) The analysis of hematoma volume in ICH + KIRA6 + antagomiR-NC and ICH + KIRA6 + antagomiR-25 groups (n = 4).(D) The analysis of brain edema in the ICH + KIRA6 + antagomiR-NC and ICH + KIRA6 + antagomiR-25 groups (n = 4).Values are presented as mean ± SD. *p < 0.05 vs. ICH + KIRA6 + antagomiR-NC group.(E) The analysis of composite Garcia score in ICH + KIRA6 + agomiR-NC and ICH + KIRA6 + agomiR-25 groups (n = 4).(F) The analysis of corner turn times in ICH + KIRA6 + agomiR-NC and ICH + KIRA6 + agomiR-25 groups (n = 4).(G) The analysis of hematoma volume in ICH + KIRA6 + agomiR-NC and ICH + KIRA6 + agomiR-25 groups (n = 4).(H) The analysis of brain edema in the ICH + KIRA6 + agomiR-NC and ICH + KIRA6 + agomiR-25 groups (n = 4).Data are the mean ± SD. *p < 0.05 vs. ICH + KIRA6 + agomiR-NC group.downregulated.Which pathway(s) mediate(s) IRE1α kinase activity regulating on miR-25-3p expression needs to be further explored in future experiments.
little research on the expression pattern of miR-25-3p following ICH and its relationship with oxidative injury in ICH brain.In this experiment, our results showed that the level of miR-25-3p was significantly increased following ICH injury.The fluorescence intensity of DHE staining and the content of H 2 O 2 in ICH brain were increased by antagomiR-25 and decreased by agomiR-25.Accordingly, the neurological function score was reduced by antagomir-25 and increased by agomiR-25.The above results suggest that miR-25-3p can inhibit oxidative stress injury following ICH, and the upregulation of miR-25-3p in the early stage of ICH brain is a protective adaptive response.ROS are a category of oxygen-derived small molecules in the cell and include highly reactive free oxygen radicals such as superoxide anion (O 2 −

O 2
consumption, and excessive ROS production during oxidative stress following ICH could induce brain injury through a variety of different mechanisms, 42 including destroying membrane integrity by promoting lipid peroxidation, inducing autophagy mediated by activation of various autophagy-related protein, leading to neuronal apoptosis by DNA and protein damage, initiating inflammatory response by activation of inflammasomes, and so on.Studies have

F I G U R E 11
The proposed pathway by which IRE1α regulates the oxidative stress response via miR-25-3p/Nox4 in neuron and glia cells after ICH.① ICH induces ER stress and activates the phosphokinase activity of IRE1α.② Self-phosphorylated IRE1ɑ increases the expression of miR-25-3p.③ The complementary combination of miR-25-3p with Nox4 mRNA inhibits the translation of Nox4 mRNA and subsequently downregulated expression of Nox4, which leads to ROS production decrease, especially H 2 O 2 .

E 1 The mortality rate of mice. Groups WB PCR BWC IF staining HE/Nissl staining Hematoma volume DHE staining H 2 O 2 contents MWM/EPM/body weight test Mortality Total
by western blot, hematoma volume assay, and neurobe-