Morin protects the blood–brain barrier integrity against cerebral ischemia reperfusion through anti-inflammatory actions in rats

This study aimed to investigate the effects of morin on cerebral damage and blood–brain barrier (BBB) integrity in a middle cerebral artery occlusion (MCAO) and reperfusion model. Wistar rats were exposed to MCAO for 2 h, followed by reperfusion. Thirty mg/kg of morin was administered via intraperitoneal injection at the different time points: before ischemia, during ischemia, and at reperfusion. The rats were divided into five groups, including sham, vehicle, and three groups of morin. Twenty-four hours after reperfusion, the rats were tested for neurological deficits, and the brains were harvested to assess brain damage. In addition, brains were harvested 72 h to determine BBB disruption. We found that morin significantly reduced reactive oxygen species production and lipid peroxidation. It also decreased inflammation via reducing the expression of Toll-like receptor 4, nuclear factor kappa-beta. Morin ameliorated cerebral damage and reduced apoptosis through decreasing the cerebral infarct size, including apoptotic cell death. Moreover, morin decreased the BBB damage via reducing Evans blue extravasation, neutrophil infiltration, and increasing tight junction protein expression. Therefore, morin protected against cerebral and BBB damage by attenuating oxidative stress, inflammation, and apoptosis in MCAO and reperfusion models.


Results
Physiological parameters and regional cerebral blood flow (rCBF) monitoring during McAo. We examined parameters during pre-ischemia, ischemia, and reperfusion. Morin treatment did not affect the rat's body temperature, oxygen saturation, body weight, or heart rate compared with vehicle-treated rats ( Fig. 1A-D). In addition, we measured rCBF (Fig. 1E). There were no changes in the sham group during the operation, but the rCBF in the MCAO rats decreased immediately to less than 25% of the baseline after occlusion (Supplementary file 1). This result showed that the MCAO model was successful. Representative heart rate, oxygen saturation, body temperature, and body weight data. (D) Regional cerebral blood flow monitoring before ischemia, during ischemia, and after reperfusion. The data are presented as the mean ± standard error of the mean (SEM) from three independent experiments (***p < 0.001 compared with the sham group; ### p < 0.001 compared with the vehicle group; n = 6 per group).

Morin suppressed TLR4 and inflammatory factor expression in cerebral I/R model rats.
To investigate the mechanism underlying the ability of morin to reduce inflammation mediated by NF-κB activation in I/R rats, we first examined the expression levels of inflammatory cytokines, including TNF-α and IL-1β, by Western blotting analysis (Fig. 5A, D, E). TNF-α and IL-1β expression in I/R rats increased compared to the control group. Treatment with morin before or during reperfusion significantly reduced the expression of both  www.nature.com/scientificreports/ proteins compared to the I/R group treated with vehicle. The iNOS expression was significantly increased in vehicle-treated rats. However, morin administration in all groups tended to decrease, but with no significant differences when compared to I/R group (Fig. 5A,F). Next, we investigated the levels of pNF-κB by Western blotting. I/R rats exhibited significantly elevated pNF-κB expression (Fig. 5A,C). Treatment with morin before and during reperfusion significant decreased pNF-κB expression to the group treated with vehicle ( Fig. 5A,C). These results indicated that morin attenuates I/R-induced NF-κB pathway activation by inhibiting proteins related to the NF-κB pathway and thereby suppressing the expressions of TNF-α and IL-1β. We further investigated the role of morin in neuroinflammation in I/R injury to determine whether it was mediated by the TLR4 pathway. As shown in Fig. 5A,B, the I/R group presented significantly increased TLR4 expression compared to the control group. Treatment with morin during reperfusion significantly decreased TLR4 expression compared with the I/R group. However, TLR4 expression in rats treated with morin before occlusion or before reperfusion tended to decrease, but with no significant differences when compared to I/R group.
In addition, NF-ĸB activation also produces MMP-9, which is a modulator of inflammation that causes neutrophil infiltration and blood-brain barrier (BBB) breakdown 20,21 . We found that morin treatment significantly decreased MMP-9 expression compared to vehicle-treated rats (p < 0.05 and p < 0.01; Fig. 5A,G).

Morin reduced myeloperoxidase (MPO) activity and neutrophil infiltration following cerebral I/R injury.
We investigated inflammation at 72 h after reperfusion by examining neutrophil infiltration; we performed an MPO activity assay. MPO is a peroxidase enzyme that is most abundantly expressed in neutrophil granulocytes. We found that MPO activity in the vehicle group was significantly increased, but morin treatment significantly decreased this activity in all groups compared with the vehicle (p < 0.01 and p < 0.05; Fig. 6).

Morin treatment decreased BBB leakage via improving tight junctions (TJs) of the BBB.
To investigate the effects of morin on BBB leakage in cerebral I/R models, we performed the Evans blue extravasation assay 72 h after reperfusion. The BBB leakage was represented by the seepage of the Evans blue dye as dark blue color in the brain tissue. The severity of the BBB leakage is presented as OD 620nm /g. In the morintreated groups, there was significantly reduced Evans blue color extravasation compared with the vehicle group (p < 0.001 and p < 0.01; Fig. 7A). As expected, the sham group had no Evans blue dye leakage.
We next detected TJ protein expression, including occluding and claudin, at 72 h after reperfusion via western blotting. We found that the vehicle group exhibited significantly reduced expression of both TJ proteins when compared to the sham group. All morin-treated groups presented significantly increased occludin and claudin expression compared with the vehicle group (p < 0.05 and p < 0.001; Fig. 7B). In addition, we used transmission electron microscopy (TEM) to examine the effects of morin on the ultrastructure of the BBB in cerebral I/R rats at 72 h after reperfusion. Our results showed that cerebral I/R caused ultrastructural changes in the BBB as well www.nature.com/scientificreports/ as astrocyte swelling. Moreover, there were alterations in the vascular lumen and the numbers of microvilli in the vehicle group, but these changes were attenuated in the morin-treated groups (Fig. 7C).
Morin decreased apoptosis and reduced morphological changes. To investigate the effects of morin on apoptosis, we measured caspase-3 expression via western blot (Fig. 8A). Caspase-3 expression in the  www.nature.com/scientificreports/ vehicle group was significantly increased when compared with the sham group. Morin treatment during reperfusion significantly decreased caspase-3 expression compared to the vehicle-treated rats (p < 0.01; Fig. 8B). We examined caspase-3 activity with a caspase-3 assay kit. This activity was significantly increased in the vehi- www.nature.com/scientificreports/ cle compared to the sham group, but morin-treated groups exhibited significantly decreased caspase-3 activity when compared with the vehicle-treated group (Fig. 8C). This result correlated with the caspase-3 expression data. We next determined the number of apoptotic cell deaths in the penumbra area in the cerebral cortex via terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining. TUNEL-positive cells appeared as dark brown particles, which are indicative of apoptotic cells (Fig. 8D). There were no TUNEL-positive cells in the sham group. Our result showed that the apoptotic index was significantly decreased in the morin-treated compared with the vehicle group (p < 0.05; Fig. 8E).
After 24 h of reperfusion, we assessed morphological changes in the cerebral cortex and striatum with hematoxylin and eosin (H&E) staining. Neuronal cells with pyknotic nucleus and vacuoles around their nucleus were widely represented in the vehicle group. There were no changes in the sham group, and the morin-treated groups presented reduced changes in neuronal cells (Fig. 8F). The result showed that the percentage of pyknotic nucleus was significantly decreased in the morin-treated compared with the vehicle group (p < 0.001; Fig. 8G).

Methods
Antibodies and reagents. Anti-TLR4, anti-TNF-α, and anti-IL-1β were purchased from Abcam (Abcam, MA, USA). Anti-MMP-9 was purchased from Cell Signaling (Danvers, MA, USA). Anti-NF-κB, anti-iNOS, anti-caspase-3, anti-occludin, anti-claudin-5, anti-GFAP, anti-Iba1, and anti-actin were purchased from Millipore (Millipore, MA, USA). All other reagents were obtained from Sigma-Aldrich (St. Louis, MO).  Animals and drug administration. We obtained male Wistar rats weighing 300-350 g from the National Laboratory Animal Center, Mahidol University, Salaya, Nakorn Pathom, Thailand. All animals were kept on a 12/12 h light/dark cycle with a constant ambient temperature (25 ± 1 °C). All animals were given a standard pellet rat diet and water ad libitum. All experiments in this study were approved by the Institutional Animal Care and Use Committee at the Faculty of Medicine, Chiang Mai University (Permit number: 33/2559), in compliance with National Institutes of Health (NIH) guidelines. Briefly, we randomly divided the 120 rats into five groups (n = 24) as follows: (1)

Surgical preparation of McAo model in rats.
We anesthetized rats using zoletil (30 mg/kg) and xylazine (10 mg/kg) i.p. We conducted MCAO on the rats with the use of an intraluminal monofilament technique, as previously described 23 In brief, we identified the right common carotid artery (CCA) and external carotid artery (ECA). We introduced 4-0 filament (Doccol Corp., Sharon, MA, USA) into the internal carotid artery and advanced it until we felt a slight resistance. Importantly, we measured the sudden drop in cortical perfusion (< 25% of the baseline value) by laser Doppler flowmetry (AD instruments, Dunedin, New Zealand) 24 . After 120 min of MCAO, we carefully removed the filament to permit MCA reperfusion. After surgery, we transferred the rats to a room temperature environment (25 ± 1 °C) until sacrifice.
neurological assessment. Twenty-four h after reperfusion, we assessed 6 rats from each group for neurological deficits, according to a 5-point scale system, as previously described 23 . The scale is as follows: 0 = no neurological deficits; 1 = failure to extend contralateral forepaw fully; 2 = circling to the ipsilateral side when held by the tail; 3 = falling to the contralateral side; and 4 = did not walk spontaneously and has a depressed level of consciousness.
Analysis of infarct volume. Three whole brains from each group were harvested 24 h after reperfusion, and the brain tissues were processed to six 2-mm-thick slices. We incubated the slices in 1% 2,3,5-triphenyltetrazolium chloride (TTC) at 37 °C for 20 min and then fixed them in 4% buffered formaldehyde solution overnight. We determined the infarct volume by measuring the area of the ischemic lesion in each section by using ImageJ software. The infarct volume (%) was calculated as follows: 100% × [(contralateral hemisphere volume − noninfarct ipsilateral hemisphere volume)/contralateral hemisphere volume] 25 . tUneL assay. We detected apoptotic cell death by TUNEL staining on the basis of DNA fragmentation. We sacrificed rats from each group 24 h after reperfusion and fixed the brains in 4% paraformaldehyde (PFA). Next, we processed the brain tissues into 4-μm-thick slices. We deparaffinized the sections and then rehydrated them. We performed TUNEL staining according to the manufacturer's instruction for the TUNEL assay kit ( investigation of BBB disruption. We investigated BBB leakage by Evans blue injection, as previously described 20 . Briefly, at 72 h after reperfusion, rats from each group were anesthetized and administered 2% Evans blue solution (4 ml/kg) via intravenous injection into the jugular vein. Thirty min after EV injection, the rats were perfused intracardially with cold phosphate-buffered saline (PBS, pH 7.4). Then, we harvested the brains, homogenized them in DMSO, and incubated the homogenates at 50 °C. We centrifuged the samples at 12,000g at 4 °C for 30 min and collected the supernatants. We measured the absorbance at 620 nm by using a spectrophotometer (BioTek Instruments Inc, Winooski, VT, USA).

Histology analysis.
Whole brains from each group were harvested and fixed for 48 h in 4% PFA. Next, the brain tissues were commonly embedded in paraffin, processed to 4-μm-thick slices, and stained with H&E. We observed morphological changes in the cerebral cortex and the striatum at the same level in each group were observed using a light microscope (Olympus AX70, Japan investigation of RoS production. We used oxidation-sensitive 2ʹ,7ʹ-dichlorofluorescein diacetate (DCFH-DA) dye to investigate intracellular ROS. Twenty-four h after reperfusion, we harvested the cerebral penumbra from each group and processed them to 2-mm-thick slices at the same level. We added total lysis buffer with protease inhibitor cocktail, including 10 mM HEPES (pH 7.9), 1.5 mM MgCl 2 and 10 mM KCl. We homogenized the samples, centrifuged them at 12,000 rpm for 10 min at 4 °C, and collected the supernatants. We placed the supernatants were placed in a 96-well plate and mixed them with 10 µl of H 2 DCF-DC solutions, followed by incubation in the dark for 25 min. We measured the samples with a microplate reader (DTX800, Beckman Coulter, Austria) at the excitation wavelength 480 nm and the emission wavelength 530 nm.
investigation of lipid peroxidation (MDA assay). Twenty-four h after reperfusion, we determined the MDA level using a calorimetric assay. MDA is the end-product of lipid hydroperoxide decomposition. In brief, we harvested the cerebral penumbra from each group and homogenized them in lysis buffer. The samples were mixed with 10 µl of butylated hydroxytoluene, 250 µl of 1 M phosphoric acid, and added 250 µl of 2-thiobarbituric acid. All reactions were mixed and incubated at 60 °C for 1 h. Then, the samples were centrifuged at 12,000 rpm for 5 min. The supernatants were transferred into a 96-well plate for measuring the absorbance by a microplate reader (BioTek Instruments Inc, Winooski, VT, USA) at 532 nm.
Mpo activity assay. MPO is a peroxidase enzyme that is most abundantly expressed in neutrophil granulocytes. Evaluating MPO activity is crucial to understanding its effects on inflammation. Briefly, the brain tissues of each group were collected from the ipsilateral hemisphere at 72 h after reperfusion. The MPO activity was measured following the myeloperoxidase (MPO activity assay kit) (Abcam, MA, USA).

Caspase-3 activity assay.
The assay is based on detection of cleavage of substrate DEVD-AFC, which is cleaved by caspase-3. Twenty-four h after reperfusion, we collected the cerebral penumbra from each group and homogenized the samples in lysis buffer. Then, we measured caspase-3 activity using an assay kit (Abcam, MA, USA).

Western blot analysis.
Twenty-four or seventy-two hours after reperfusion, cerebral penumbra of each group was harvested and stored them at − 80 °C until use. To extract total proteins, we homogenized the brains and determined the total protein concentrations using the Bradford protein assay (Bio-Rad Laboratories, Hercules, CA, USA) with bovine serum albumin (BSA) as the standard. We applied 25 µg of total protein in each sample were applied into the lanes of a 10-15% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels. Then, we transferred isolated proteins to polyvinylidene difluoride (PVDF) membrane (Immobilon-P, Millipore, Bedford, MA, USA) and blocked them in a fresh blocking buffer (containing 5% skim milk in 0.1% Tween-20 in Tris-buffered saline, pH 7.4) for 3 h at room temperature. Subsequently, the membranes were incubated with the primary antibodies (anti-TLR4, anti-NF-κB, anti-TNF-α, anti-IL-1β, anti-iNOS, anti-MMP-9, anti-caspase-3, anti-GFAP, anti-Iba1, anti-occludin, or anti-claudin) overnight at 4 °C. Next, the membranes were washed with Tris-buffer saline and Tween-20 (TBST) and incubated them with anti-rabbit IgG or anti-mouse peroxidase-conjugated secondary antibody. The membranes were incubated with Immobilon Western (Millipore, MA, USA) and exposed them to X-ray film. Densitometric analysis was performed by a scanning films with a densitometer the results were normalized using β actin by image J analysis.
Statistical analysis. The data are presented as mean ± standard error of the mean (SEM). Statistical differences between the two groups were determined by using Student's t-test. Other data were analyzed using a one-way analysis of variance (ANOVA), followed by Dunnett's post hoc test. However, the data with non-normal distribution were analyzed using Kruskal-Wallis test. A p value < 0.05 was considered a statistically significant difference between experimental and control groups.

Discussion
Cerebral I/R injury leads to the excessive production of ROS 27 , a phenomenon that can activate a cascade of pathophysiological processes, including oxidative stress, inflammation, and neuronal apoptosis [28][29][30] . A previous study demonstrated the anti-inflammatory and anti-oxidant activity of morin in a transient cerebral model. Morin acts as a powerful anti-inflammatory agent by suppressing activation of NF-ĸB and the expression levels of pro-inflammatory cytokines such as TNF-α, IL-1β, and iNOS 19 . However, its benefits toward preventing or ameliorating BBB disruption after cerebral I/R have not been reported. Hence, in the present study, we first investigated the ability of morin to attenuate BBB and cerebral damage following cerebral ischemia/reperfusion injury during the acute phase of rats subjected to MCAO and reperfusion injury. We focused on three different time points: pre-ischemia, during ischemia, and during reperfusion. Our results demonstrated that morin treatment had no effect on body temperature, oxygen saturation, body weight or heart rate in the rats. Further, morin decreased NF-ĸB activation, whereas TNF-α, IL-1β, and iNOS expression decreased. The major findings showed that morin inhibited NF-ĸB activation which, in turn, decreased the expression of proinflammatory www.nature.com/scientificreports/ cytokines. Moreover, morin suppressed MMP-9 expression, which improved BBB disruption by reducing TJ protein degradation and BBB permeability at pre-ischemia, during ischemia, and during reperfusion. Cerebral ischemia leads to cerebral infarction via mechanisms that cause irreversible damages and neuronal cell death 31 . The mechanisms occur extremely rapidly in the ischemic core, and thus this area is difficult to protect. However, preservation of the penumbra by residual or collateral blood flow and salvaging the area can help to prevent the continued growth of the ischemic core or decrease the infarct size. Our results showed that at all time points morin treatment improved the neurological outcome and attenuated the cerebral infarct size. When cerebral perfusion is re-established (reoxygenation), a high oxygen surge occurs in the penumbra. This event leads to marked increase in ROS generation. Many studies have reported that a large quantity of ROS causes subsequent oxidative stress and neuronal cell injury [5][6][7] . Following the oxidative stress of cerebral ischemia, inflammation, BBB disruption, and apoptosis are induced. Therefore, suppression of oxidative stress might prevent neuronal injury in cerebral I/R. A decrease of ROS may reduce neuronal cells death by attenuating lipid peroxidation and oxidative stress. Many studies have reported that morin has strong antioxidant properties. Morin prevents lipid peroxidation, scavenges intracellular ROS, and protects the antioxidant system [32][33][34] . Moreover, a previous study reported that morin plays an antioxidant role in the MCAO model. Our results confirmed that morin-treated groups tends to decrease ROS production and reduce lipid peroxidation, as observed by the MDA level. These outcomes are related to the levels of Iba1 and GFAP expression, markers that represent microglia and astrocytes, respectively. This study showed that microglia and astrocyte were the source of ROS and oxidative stress in cerebral I/R injury.
ROS overexpression also triggers neuronal inflammation, which includes the contribution of TLR4. TLR4 is upregulated after cerebral I/R injury to recognize the intracellular components from neuronal cell death was known as DAMPs 35,36 , and activation of TLR4 leads to neuronal inflammation through an NF-ĸB signaling pathway 8,9 . NF-ĸB activation plays an important role in pro-inflammatory cytokine regulation, including TNFα, IL-1β, and iNOS. Therefore, TLR4 inhibition downregulates activation of the NF-ĸB signaling pathway, in addition to reducing pro-inflammatory cytokines production. Our studies found that morin attenuated TLR4 expression and also decreased NF-ĸB activation, which is closely related to reducing TNF-α, IL-1β, and iNOS expression. Furthermore, MMP-9, a pro-inflammatory cytokine produced by NF-ĸB activation, is also released to induce BBB disruption 20 . MMP-9 directly degrades the extracellular matrix, basal membrane, and TJ proteins, such as occludin and claudin, all of which lead to BBB breakdown 37 . Subsequently, the BBB disruption allows neutrophil infiltration from the blood into the brain at 72 h after cerebral I/R 12 . However, previous research suggested that suppressing MMP-9 expression improves BBB disruption by reducing TJ protein degradation and BBB permeability in a cerebral I/R model 26,38 . Our results revealed that morin administration suppressed MMP-9 expression. In addition, morin decreased the degradation of TJ proteins, like occludin and claudin, and reduced the infiltration of neutrophils, correlated with decreased MPO activity at 72 h after reperfusion. These results, related to the Evans blue assay, showed that the leakage of Evans blue dye in morin-treated groups was significantly decreased compared with the vehicle group. These findings suggest that morin decreases BBB permeability and BBB leakage. Moreover, we further studied the ultrastructural changes of the BBB using TEM. We found that the cerebral microvascular was altered in morin-treated groups. Sham rats showed mostly vascular lumen alterations and the formation of microvilli. These results suggest that morin reduces inflammation and enhances BBB integrity after cerebral ischemia/reperfusion in the MCAO model.
Several death receptor ligands, such as TNF-α and IL-1β, can markedly induce neuronal apoptosis 31,39 . Previous studies demonstrated that apoptosis plays an important role in the pathogenesis of cerebral I/R and leads to cerebral infarction. Decreasing caspase-3 expression reduces apoptotic cells and cerebral infarction after MCAO. A previous study revealed that morin-treated rats subjected to MCAO showed caspase-3 downregulation 19 . In our study, morin treatment at each time point decreased caspase-3 expression and activity. In addition, we noted fewer TUNEL-positive cells in the ischemic brain of morin-treated groups, data that indicate a decrease in apoptotic cells. Our results also showed histological changes in the cerebral cortex and striatum. We found that the number of neuronal cells was decreased in the vehicle group, and neuronal cells exhibited pyknotic nuclei and overall shrinkage. Hence, our data indicated that morin can reduce apoptosis in the acute phase of cerebral ischemia/reperfusion.
In conclusion, our study provided evidence that morin has a neuroprotective effect on cerebral ischemia/ reperfusion. It improved cerebral damage and neurological outcomes by attenuating oxidative stress, inflammation, and apoptosis. Here, we found that morin improved BBB disruption and enhanced BBB integrity.