Tongxinluo reduces brain edema and inhibits post-ischemic inflammation after middle cerebral artery occlusion in rats

Author’s Accepted Manuscript Tongxinluo reduces brain edema and inhibits postischemic inflammation after middle cerebral artery occlusion in rats Min Cai, Zhonghai Yu, Lili Wang, Xiaoling Song, Jingsi Zhang, Zhennian Zhang, Wen Zhang, Wenwei Li, Dingfang Cai, Jun Xiang www.elsevier.com/locate/jep

PII: DOI: Reference:

S0378-8741(16)30022-8 http://dx.doi.org/10.1016/j.jep.2016.01.026 JEP9937

To appear in: Journal of Ethnopharmacology Received date: 16 September 2015 Revised date: 31 December 2015 Accepted date: 18 January 2016 Cite this article as: Min Cai, Zhonghai Yu, Lili Wang, Xiaoling Song, Jingsi Zhang, Zhennian Zhang, Wen Zhang, Wenwei Li, Dingfang Cai and Jun Xiang, Tongxinluo reduces brain edema and inhibits post-ischemic inflammation after middle cerebral artery occlusion in rats, Journal of Ethnopharmacology, http://dx.doi.org/10.1016/j.jep.2016.01.026 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Tongxinluo Reduces Brain Edema and Inhibits Post-ischemic Inflammation after Middle Cerebral Artery Occlusion in Rats Min Cai1, Zhonghai Yu1*, Lili Wang2, Xiaoling Song1, Jingsi Zhang1, Zhennian Zhang1, Wen Zhang1, Wenwei Li1 , Dingfang Cai 1, Jun Xiang1# 1Department of Integrative Medicine, Zhongshan Hospital, and Laboratory of Neurology, Institute of Integrative Medicine, Fudan University, Shanghai 200032, China 2Department of Diagnostic Radiology, Xiehe Hospital, Fujian Medical University, Fujian 350001, China *co-first author #co-corresponding author Corresponding authors at: Department of Integrative Medicine, Zhongshan Hospital, and Laboratory of Neurology, Institute of Integrative Medicine, Fudan University, Shanghai 200032, China˗180 Fenglin Road, Shanghai 200032, China. Tel:+86 21 64047903; Fax: +86 21 64047903. Email address: [email protected] (Dingfang Cai), [email protected] (Jun Xiang)

Abstract Ethnopharmacological relevance—Tongxinluo (TXL), a widely used traditional Chinese medicine, has been proved multiple therapeutic effects in cerebral ischemic infraction. The purpose of this study was to investigate the protective effects of TXL on the brain edema and post-ischemic inflammatory response. Materials and Methods—Middle cerebral artery occlusion in the rat was used as the ischemia model. Rats were treated with TXL. In the first stage, the best dosage was chosen based on functional assessment and infarct size. In the second stage, rats were randomly divided into 5 groups: sham control (sham), ischemia and reperfusion (IR) 24h, TXL24h, I/R72h, TXL72h. TXL(1.6 g/kg/day) administration was preperformed for 3 days in TXL groups, and was post-performed for 24 hours (TXL24h group) or 72 hours (TXL72h group). Brain edema was measured by water content, MRI and AQP4 expression. Iba1, HMGB1, TLR4, NF-κB expression were examined by immunofluorescence staining or Western blot. TNF-α was determined by enzymelinked immunosorbent assay. Results—High dose (1.6g/kg/day) of TXL remarkably reduced neurological deficit scores and cerebral infarct area. Compared with those results of I/R24h group, prepost treatment with TXL for 3 days decreased brain water content, down-regulated

AQP4 expression, lowered relative signal intensity of T2WI, reduced lesion volume ratio, and inhibited the activation of microglia, HMGB1, TLR4, NF-κB and TNF-α. Conclusions—These results indicated that the TXL pre-post treatment for 3 days could be an effective therapy for brain ischemia by inhibiting the development of brain edema and post-ischemic inflammation. Key Words: Tongxinluo; brain edema; inflammation; HMGB1; MRI Chemical compounds (standard reference compounds) studied in this article: Ginsenoside Rg1 (PubChem CID: 441923); Hirudin (PubChem CID: 72941487); Buthotoxin (PubChem CID: 20054854); Paeoniflorin (PubChem CID: 442534); Chitin(PubChem CID: 6857375); Santalol, alpha- and beta- (PubChem CID: 24832101); beta-Boswellic acid (PubChem CID: 168928); Isoflavone (PubChem CID: 72304); Jujuboside A (PubChem CID: 171446).

1. Introduction Two of the major mechanisms exacerbating ischemic stroke are extensive brain edema formation and post-ischemic inflammation. The detrimental edema reduces focal blood flow, causing cell necrosis and apoptosis. Cerebral inflammation exaggerates vascular dysfunction, inducing further neuronal cell death. A promising focus of relieving brain edema is to modulate the expression of aquaporin4 (AQP4). AQP4 is a bidirectional water channel, which expresses strongly in astroglia and cerebrospinal fluid–brain interfaces (Amiry et al., 2003; Kovalenko, et al., 2006). Magnetic resonance imaging (MRI) is the best non-invasive way to measure ischemic lesion. Both diffusion-weighted imaging (DWI) and T2-weighted imaging (T2WI) reveal a unique alteration in brain water dynamics, therefore detect brain edema (Tuor et al., 1998; Aden et al., 2002). High mobility group box 1 (HMGB1), locating in the nucleus of brain cells normally and released from necrotic neurons at the hyperacute phase, is one of the major inflammatory mediators and damage-associated molecular patterns (DAMPs) in ischemic brain (Shichita et al., 2014). Extracellular HMGB-1 activates its receptor Tolllike receptor 4 (TLR4), thereafter promotes the activation of the transcription factor nuclear factor-κB (NF-κB). NF-κB, a central mediator of inflammatory processes, targets several proinflammatory genes, including tumor necrosis factor-α (TNF-α), interleukin-1 (IL-1), interleukin-6 (IL-6), intercellular adhesion molecule 1(ICAM-1), inducible nitric oxide synthase (iNOS), and matrix metallopeptidase 9(MMP9) (Ridder and Schwaninger, 2009). Tongxinluo (TXL), a Chinese herbal medicine compound, has been approved by the State Food and Drug Administration of China in 1996 for the treatment of angina pectoris and ischemic stroke. It has multiple therapeutic effects in cerebral ischemic

infraction, including lowering cerebral infarct volume, protecting tight junction NFκB, attenuating neuronal loss, and enhancing neurogenesis as well as angiogenesis (Li et al., 2015; Cheng et al., 2014; Chen et al., 2014; Wang et al., 2014). It protects the brain from BBB disruption via reducing cytokines IL-6,IL-1β and TNF-α mRNA expression (Liu et al., 2013), which are NF-κB target genes. We hypothesized that Tongxinluo reduced brain edema by down-regulating AQP4 expression, and inhibited post-ischemic inflammation through HMGB1/TLR4induced NF-κB activation pathway.

2. Materials and Methods 2.1 Preparation of Tongxinluo TXL powder, in the form of dried superfine powder (≤10μ m), was provided by Shijiazhuang Yiling Pharmaceutical Incorporated Company (Shijiazhuang, China). It contains 12 components (Table 1), and was authenticated and standardized on the basis of marker compounds in the Chinese Pharmacopoeia (Committee, 2005). The powder was finally dissolved in normal saline and stored at 4ć until being used.

Table 1 12 Ingredients of Tongxinluo. Ingrediens (La tin name)

Family

Voucher specimen number

Part use d

Processin g

Amount used (%)

Araliaceae

11,001

Root and rhizome

Extractio n

1.677

Rhamnaceae

11,002

Seed

Extractio n

1.173

Paeonia lactiflora Pall.

Ranunculaceae

11,003

Root

Extractio n

1.558

Santalum album L.

Santalaceae

11,004

Heartwo od of st em

Extractio n

0.354

Plants Panax ginseng C.A.Mey. Ziziphus jujuba Mill. Var. spinosa (Bunge) Hu H.F.Chou

Dalbergia odorifera T.Chen

Leguminosae

11,005

Extractio n

4.000

11,006

Heartwo od of st em and root Resin

Boswellia carteri Birdw

Burseraceae

Farina

5.927

Borneolum syntheticum

Dipterocarpaceae

11,007

Resin

Artificial

3.626

Scolopendra subspinipes mutilans L. Koch

Psittacidae

12,001

Dried bo dy

Farina

3.623

Buthus martensii Karsch

Buthidae

12,002

Dried bo dy

Farina

18.111

Steleophage plancyi (Boleny)

Corydiidae

12,003

Female dried bo dy

Mircro-or yzae fari na

18.111

Hirudo nipponica Whitman

Hirudinidae

12,004

Dried bo dy

Farina

27.330

Cryptotympana pustulata Fabricius

Cicadidae

12,005

Skin

Farina

18.111

Insects

2.2

Animals

One hundred and sixty-three adult male Sprague-Dawley rats (Experimental Animal Center, Fudan University, China), weighing 230–280 g, were performed on for the experiment. Rats were maintained on a 12:12 h light: dark cycle, with free access to food and water and a constant room temperature of 21ć. The experimental protocols and animal handling procedures were approved by the Animal Care and Use Committee (ACUC) of Fudan University and consistent with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Six rats died from intracranial hemorrhage and two died from anesthesia.

2.3

Transient Focal Cerebral Ischemia and Reperfusion

The transient focal ischemia was induced by left middle cerebral artery occlusion (MCAO) as described (Yang et al., 1994; Lan et al., 2013). Rats were anesthetized with 10% chloral hydrate (350 mg/kg, intraperitoneal (i.p.) injection). After the skin incision, the left common carotid artery (CCA) was exposed and carefully separated from nerves and tissue. The external carotid artery (ECA) was dissected from internal carotid artery (ICA). After clipping the ECA, a nylon monofilament was inserted through an incision at the far-end of ECA, approximately 1cm away from the bifurcation. The microvascular clip was removed. Resistance could be felt after the monofilament was inserted into the ECA stump for about 18-20mm, indicating the blunt tip had passed the MCA origin. During the course of the surgery, the rectal temperature, blood gases, and cardiovascular rate of each rat were monitored. Reperfusion was initiated by withdrawal of the monofilament after 90min of ischemia. 2.4

Groups

The study was designed for 2 stages. In the first stage, the best dosage was detected. Rats were randomly divided into 5 groups: sham group (sham), ischemia and reperfusion (IR), and IR+TLX-L(Low), TXL-M(Medium), and TXL-H(High), 6 rats for each group. According to the recommendation by Yiling Pharmaceutical Incorporated Company, we selected 0.4, 0.8, and 1.6 g/kg/day as the low, medium, and high dosages respectively. The rats in the TXL treated groups were orally administered the corresponding doses of TXL, while other rats were given the same volume of normal saline. TXL administration was performed twice a day at 8:00 and 18:00 for 3 days before IR. Neurological deficits and the percentage of cerebral infarct area were evaluated in order to pick up the best dosage. In the second stage, rats were randomly divided into 5 groups: sham control (sham), ischemia and reperfusion (IR) 24h, TXL24h, I/R72h, TXL72h. TXL(1.6 g/kg/day) administration was pre-performed for 3 days in TXL groups, and was postperformed for 24 hours (TXL24h group) or 72 hours (TXL72h) before examinations. Rats in sham and IR groups were give the same volume of normal saline. Rats in the sham group were subjected to the same operation, without monofilament inserted. 2.5 Functional Assessment 2.5.1 Neurological Deficit Scores Neurological examinations were performed on 0, 1, 3, 5, and 7 days after reperfusion in the first stage of this study. A 5-point scale was applied to assess the neurological deficits (Longa et al., 1989): 0 = no deficit; 1 = failure to extend right forepaw; 2 = circling to the right; 3 = falling to the right; 4 = no spontaneous walking with a depressed level of consciousness. In the study, 4 rats that underwent MCAO without any detectable neurological deficits were excluded from the following

investigations and analysis to exclude operative failures. 2.5.2 Rota-rod Test The rota-rod test was performed according to Hunter (Hunter et al., 2000). Animals were habituated to the rota-rod, trained to stay on the rotating drum (speed 4 r.p.m.) for 90 s, which was to provide a pre-operative baseline. Then animals were placed on the rotarod set in the accelerating mode (from 4 to 40 r.p.m. over 5 min) in the testing sessions. The latency for the animal to fall off the drum was recorded (max 5 min). Testing took place on 0, 1, 3, 5, and 7

days after reperfusion. 2.6

Quantification of Ischemic Infarct Area

In the first stage, the cerebral ischemic infarct areas were evaluated by 2,3,5triphenyltetrazolium chloride (TTC; Sigma-Aldrich, St. Louis, MO, USA) staining at 24 hours after reperfusion. Rats were sacrificed under deeply anesthesia, their brains were quickly removed, and placed at −20ć for 20min. Brains were sectioned into 5 coronal slices, 2mm thickness each slice, and stained with 1% TTC solution for 20min, in the dark, at 37ć. Finally, the tissues were fixed in 4% paraformaldehyde (in 0.1M phosphate buffer, pH7.4). The quantification of ischemic infarct area was calculated with microscope image analysis software (Image-Pro Plus, USA) 24 hours later, according to the formula: [contralateral hemisphere area−(ipsilateral hemisphere area–infarct area) ]/(2×contralateral hemisphere area)× 100%. 2.7 Histologic Staining At 24/72 hours after reperfusion, the rats were deeply anesthetized and perfused transcardially with 200mL normal saline followed by 150mL 4% paraformaldehyde (0.1Mphosphate buffered saline, pH 7.4). The brains were removed and fixed in the 4% paraformaldehyde for 24 hours at 4ć. The brains were dehydrated in graded ethanol and xylene, and were embedded in paraffin. The paraffin-embedded brains were sectioned into slices of 4 m on a rotary microtome. H&E staining was performed to assess brain injury. Briefly, slices were dewaxed, dehydrated, and stained with hematoxy and eosin. Then the slices were rinsed in graded ethanol and xylene. After coverslipped under permount, the photomicrographs of slices were captured with a light microscope (Olympus/BX51, Tokyo, Japan). 2.8 Brain Water Content Measurement Brain water content was measured with the dry-wet weight method. At 24/72 hours after reperfusion, the brains were removed after rats were sacrificed under deeply anesthesia, and were separated into ischemic and non-ischemic hemispheres. Both hemispheres were immediately weighed to get the wet weight (WW). The dry weight (DW) was obtained after the tissues were placed in an oven at 100ć for 24 hours . The brain water content was assessed with the following formula: 100%

×(WW−DW)/WW. 2.9 MRI Examination MRI experiments were performed with a 3.0-T MR scanner system (MAGNETOM Verio, Siemens Healthcare, Germany). The rat was positioned supine with the head inside a 4- channel surface coil designed for the rat. To monitor infarct evolution, diffusion-weighted images and T2-weighted images (DWI and T2WI) were acquired at each imaging time point. For spin-echo echo-planar imaging (SE-EPI) DWI sequence, the three scan trace diffusion mode and b values of 0, 500, 1000 s/mm2 were used respectively, and maps of the apparent diffusion coefficient (ADC) were calculated automatically. The images were acquired with a 80X80 matrix, field of view (FOV) of 100 mm, repetition time (TR) of 4642ms, echo time (TE) of 86 ms, 4 average, Bandwidth of 1078Hz/Px, 20 coronal slices, slice thickness(ST) of 1 mm. The parameters for the T2-TSE sequence were: matrix 192x192, FOV 100mm, TR 3500ms, TE 86ms, flip angle 120°, 2 average, number of slices 20 and slice thick 1mm. A region of interest (ROI), completely covering the hyper-intense region on the ipsilateral brain and the symmetrical region on the contralateral brain, was used to monitor the signal intensity changes on both T2WI and the ADC maps. All values of signal intensities were hemispheric ratios (ratio of ipsilateral to contralateral signal intensities). The volume of infarction was calculated as 1mm (thickness of the slice) × (sum of the infarction area in all brain slides [mm2]) (Wang et al., 1997). Lesion volume ratio=ipsilateral lesion volume/(2×contralateral lesion volume)×100%. 2.10

Western Blotting Analysis

Western blotting was used to assess the expression levels of HMGB-1, TLR4, and NF-κB 24/72 hours after reperfusion. The ischemic cortex and striatum tissues were prepared with protease inhibitors (Beyotime, Haimen, Jiangsu, China) in lysis buffer and centrifuged at 13,000 ×g for 5min. The supernatant was collected. Protein concentrations were detected using a BCA kit (Beyotime). Protein solution, weighted 50mg, was separated by polyacrylamide gel electrophoresis with different concentrations. The gel was then transferred to polyvinylidene fluoride membranes (Millipore, Bedford, MA, USA), blocked for 2 hours in a 5% solution of skim milk, which was prepared with Tris-buffered saline plus 0.1% Tween-20 (TBST). The membranes were incubated with primary antibodies, monoclonal rabbit anti-HMGB1, -TLR4, and – NF-κB (all diluted 1 : 1000; Abcam, HK, China) respectively at 4 ć overnight. The membranes were washed with TBST three times, 10 minutes each time, and then were incubated with the secondary antibody conjugated with horseradish-peroxidase (Beyotime). The targeted antigens were detected by standard chemical luminescence methods (Beyotime) with Fluor Chem FC2 gel imaging system (Alpha Innotech, Santa Clara, CA, USA). The expression of HMGB-1 and TLR4 were determined by using the GADPH protein as a loading control, while NFκB was determined by Histone H3 protein. Western blots were duplicated with three independent sets. Band intensities were measured with Quantity One software (Bio-

Rad Laboratories, Hercules, CA, USA). 2.11 Immunohistochemistry Staining Immunohistochemical staining was used to evaluate whether treatment with TXL changes the expression of AQP4 and HMGB1 after cerebral ischemia and reperfusion. Rat brains were removed after transcardially perfused with 200mL normal saline followed by 150mL 4% paraformaldehyde, fixed in 4% paraformaldehyde for 24 hours, and then immersed in 30% sucrose solution with phosphate buffer saline (PBS, pH 7.4) for 24 hours. Coronal sections , 4 m thick, in the infarct region with both cortex and striatum were obtained. For immunohistochemistry, sections were deparaffinized and incubated with 0.3% H2O2 in PBS. After blocking with 5% bovine serum albumin(BSA) serum, the sections were incubated with anti-AQP4 antibody (diluted 1 : 100; Abcam), anti-Iba1 antibody(diluted 1 : 100; Abcam) or antiHMGB1 antibody(diluted 1:100, Abcam) at 4ć overnight. Those sections covered with anti-Iba1 or anti-HMGB1 were incubated with fluorescent-labeled secondary antibody FICT-conjugated anti-rat IgG (1:100, Beyotime) or Cy3-conjugated antimouse IgG (1:100, Beyotime), respectively, at 37ć for 30 minutes, followed by DAPI (Beyotime) for 10 minutes after washed in PBS. The sections with anti-AQP4 were covered with the secondary antibody conjugated with horseradish-peroxidase (Beyotime) for 1 hour at 37ć. Negative control sections were incubated with 0.01Mphosphate-buffered saline as a substitute for the primary antibody. The later sections were visualized using 3,3-diaminobenzidine tetrahydrochloride (DAB kit; Beyotime). All sections were photographed and observed with a light microscope (Olympus/BX51, Tokyo, Japan). 2.12

Enzyme-Linked Immunosorbent Assay

TNF-¢ levels in cytosolic brain fractions were analyzed by enzyme- linked immunosorbent assay (ELISA) (Elabscience).

2.13 Statistical Analysis Data analysis was performed with SPSS version 17.0 (SPSS, Chicago, IL, USA). All variables were expressed as means ± standard error of the mean (SEM). KruskalWallis test followed by Mann-Whitney test were used to analyze data of neurological deficit scores, while others were analyzed by one-way ANOVA followed by Tukey’s multiple comparison test. Differences were considered statistically significant when P < 0.05.

3. Results

3.1

TXL reduced cerebral infarction

High dose (1.6g/kg/day) of TXL remarkably reduced neurological deficit scores on the first and third day after reperfusion (P<0.05) , while it improved the Rota-rod test results on the third day after reperfusion(P<0.05) (Fig. 1A). TTC results indicated that both medium dose and high dose of TXL remarkably reduced cerebral infarct area (P<0.01) (Fig. 1B and 1C), which meant the effect of reducing infarct area did not increase with the concentration of drug. According to the above results, high dose was chosen to be the appropriate therapeutic one applied in the subsequent experiments. H&E staining in I/R24h group showed that large areas of necrosis, accompanied by obvious inflammatory reaction occurred in the lesion part. Parts of cells dissolved while other parts survived. These cells had different forms, with reddyed sarcoplasmic, pyknotic or swelling nucleus, congestive and edematous stromal vascular, lots of neutrophils and mononuclear phagocytes infiltrated in. Brain tissues in TXL72h group had the mildest necrosis and inflammation, while angiogenesis occurred (Fig. 1D).

A A.

B.

C.

D.

Cortex

Striatum

Sham

I/R24h

I/R72h

TXL24h

TXL72h

Figure 1. Effects of TXL on neurological function and cerebral infarct area. (A) The neurological deficits test and Rota-rod test. (B) The quantitative analysis of cerebral infarct area according to TTC results. (C) Images of TTC-stained brain slices. (D) HE staining of ischemic cortex and striatum. Scale bar=200­m. The arrows indicate cell loss. Data are presented as meanfSEM, n=6 rats for each group in neurological deficits test, n=5 rats for each group in other tests. * P0.05, ** P0.01 compared with the I/R (ischemia/reperfusion) group. 3.2 TXL alleviated brain edema

Brain water content of I/R24h group was higher than sham group (P<0.05), while that of TXL72h group lowered significantly (P<0.05) (Fig. 2A). AQP4 is the most abundant channel in the brain, and up-regulated in transient cerebral ischemia (Vella et al., 2015). In the TXL72h group, the immunoreactivity of AQP4 in the cerebral cortex and striatum was significantly increased compared with I/R 24h group (P<0.05) (Fig. 2B). The results of AQP4 immunohistochemistry were quantitatively determined by counting the AQP4-positive blood vessels. The treatment with TXL for 72h remarkably inhibited the expression of AQP4 both in the cerebral cortex and striatum (Fig. 2C).

A.

B.

C.

Cortex

Striatum

Sham

I/R24h

I/R72h

TXL24h

Figure 2. Brain water content and AQP4 immunohistochemistry in the ischemic cortex

TXL72h

as well as striatum. (A) The quantitative analysis of brain water content. (B) Semiquantitative analysis of AQP4 positive cells percentage. (C) Images of AQP4 immunohistochemistry in different groups. The arrows indicate AQP4 positive cells. Scale bar=200­m. Data are presented as meanfSEM, n=5 rats for each group. * P0.05, **0.01 compared with the I/R24h group. 3.3 MRI Studies The representative maps are shown in Figure 3A. Rats in sham group showed ADC and T2 values between ipsilateral and contralateral brain had no big differences. Relative signal intensity of T2WI(rT2WI) (ipsilateral/contralateral) increased significantly in I/R24h group compared to the sham group (P<0.05), and they declined remarkably in TXL72h group (P<0.05). However, rADC had no significant changes in the 5 groups (Fig. 3B) (P>0.05). Lesion volume ratio assessed by T2WI is similar to that by DWI. In TXL72h group, lesion volume ratio lowered compared to I/R24h group (P<0.05). (Fig. 3C).

A.

T 2 W I

DWI

ADC

Sham

I/R24h

I/R72h

TXL24h

TXL72h

B.

C.

Figure 3. Results of MRI study. (A) Characteristic MRI (T2WI, DWI and ADC) maps of different groups. (B) Relative ADC values and signal intensities on T2weighted images. All values are ratios of ipsilateral against contralateral measurements. (C) Lesion volume ratio assessed by T2WI. Data are presented as mean±SEM, n=5 rats for each group. * P<0.05, ** <0.01 compared with the I/R24h group. 3.4 TXL inhibited post-ischemic inflammation Iba1 expression increased in the I/R24h group, and decreased partially in the I/R72h group. TXL pre-post treatment for 72 hours inhibited it (Fig. 4A). The expression of HMGB1, TLR4 and NF-κB in the ischemia core area of both cortex and striatum were examined using Western blot (Fig. 4B). Quantitative results of the bands showed 24 hours after reperfusion, HMGB1 decreased significantly both in cortex and striatum, compared to sham group (P<0.01), while TXL pre-post treatment for 72 hours of ischemia/reperfusion inhibited the reduction (P<0.01) (Fig. 4C), Which means TXL inhibited HMGB1 from being released to extracellular. On the contrary, the expressions of TLR4 and NF-κB significantly increased in the ischemic cortex and striatum in I/R24h group(P<0.05ˈP<0.01, respectively), and were downregulated in TLX72h group (P<0.05ˈP<0.01, respectively) (Fig. 4D and 4E). Results of HMGB1 expression examined using immunofluorescence staining were consistent with that of Western blot (Fig. 4G). Quantity of TNF-α protein measured by ELISA of I/R24 group increased significantly both in cortex and striatum, compared to sham group (P<0.01), while that of TXL72h group decreased (P<0.01).

A. Sham

I/R24h

I/R72h

TXL24h

TXL72h

Iba1

DAPI

Merge

B.

TXL72h

TXL24h TXL24h

F.

TXL24h

Striatum Cortex

I/R72h

E.

I/R72h

D.

I/R24h Sham

TXL24h TXL24h I/R72h I/R24h

Sham

C.

I/R24h

Sham

TXL72h

TXL24h I/R72h I/R24h

Sham



Striatum Cortex

HMGB-1

TLR4

GAPDH

NF-κB

Histone-H3

G.

HMGB1

DAPI

Merge

Sham

I/R72h

TXL72h

Figure 4. Detection of Iba1, HMGB1, TLR4, or NF-κB in the ischemic core area of cortex as well as striatum using immunofluorescence staining or Western blotting. Detection of TNF-α using ELISA kit. Representative images of Iba1(Fig. 4A), scale bar=100μm. The arrows indicate Iba1 positive microglia. The bands of HMGB1, TLR4, NF-κB, and GAPDH, Histone H3 as internal reference (Fig. 4B). Quantitative results of the bands for HMGB1 relative to GAPDH (Fig. 4C). Quantitative results of the bands for TLR4 relative to GAPDH (Fig. 4D). Quantitative results of the bands for NF-κB relative to Histone H3 (Fig. 4E). TNF-α protein determined by ELISA (Fig. 4F). Representative images of HMGB1 by immunofluorescence staining (Fig. 4G), scale bar=200μm. Data are presented as meanfSEM, n=5 rats for each group. * P0.05, ** 0.01 compared with the I/R24h group.

4. Discussion In the present study, we provided evidences for the therapeutic effects of Tongxinluo on brain edema and post-ischemic inflammation induced by brain ischemia and reperfusion. High dose of Tongxinluo pre-post for 3 days treatment decreased neurologic deficits, infarct volume and brain water content, down-regulated AQP4 expression, and inhibited activation of HMGB1, TLR4, NF-κB.

Brain edema is an important factor of the high morbidity and mortality of large territory ischemic strokes. Brain swelling occurs due to the accumulation of excess water in the brain parenchyma (Kimelberg, 1995). Usually brain edema has two types: cytotoxic and vasogenic. Cytotoxic edema is intracellular accumulation of water due to energy failure and inability of cells to regulate their volumes in early cerebral ischemia (starting from within 1 hour after ischemia) (Kimelberg, 1995). Vasogenic edema mainly involves disruption of the blood–brain barrier, usually from no early than 3 hours, and developing for 24-72 hours in rats. The role of APQ4 differs between these two forms. It plays a major role in cytotoxic brain (Manley et al.,2000), while in vasogenic brain edema, AQP4 increases the edema fluid elimination rate (Papadopoulos et al., 2004). Our study showed AQP4 expression increased 24 hours after reperfusion, decreased slightly at 72 hours. On the whole, its expression level was consistent with brain edema, especially in striatum. The pre-post treatment for 72 hours with TXL remarkably inhibited the expression of AQP4 both in the cerebral cortex and striatum . Both T1WI and T2WI can identify vasogenic edema at later times during stroke (Quast et al., 1993), 90% of infarctions are visible on T2WI at 24 hours, compared to 50% on T1WI. Therefore, T2WI is called the ‘gold standard’ for imaging cerebral infarction (Yuh et al., 1991). T2- weighted MRI is now frequently used to determine the presence or absence of edema, especially vasogenic edema induced by reperfusion (Jura’nek and Baciak, 2009). However, T1 and T2 relaxation times are generally unaffected early after stroke onset and only begin to change with the advent of vasogenic edema (typically > 6 hours) (Gonzalez et al., 1999; Mohr et al., 1995). Diffusion-weighting imaging (DWI), which detects the movement of water molecules, has been proven to be a powerful tool for early detection (within minutes after stroke onset) of ischemic brain (Moseley et al., 1990). The hyperintense regions on DWI correspond to tissue experiencing cerebral ischemia (Beaulier et al., 1999). Under normal conditions, water molecules diffuse randomly in tissue. When ischemia occurs, water motion is restricted, which manifests as hyperintensity on DWI. The degrees of diffusion can be expressed quantitatively using the apparent diffusion coefficient (ADC) (Hsiao et al., 2013). ADC reflects the diffusion speed of water molecules. The quickly diffusing water molecules can be revealed with a larger ADC value with lower signals of DWI images. Therefore the changes of ADC value reflect the status of water molecule diffusion in post brain infarction periods (Ding et al., 2007). The temporal evolution of the MR parameters can be roughly separated into 3 phases: (1) normal transverse relaxation time (T2), reduced ADC; (2) elevated and increasing T2, low but increasing ADC; (3) high but decreasing T2, elevated and increasing ADC (Liu et al., 2007). In the present study, relative signal intensity of T2WI (rT2WI) and DWI (rDWI) increased sharply in I/R24h group, and partially declined in I/R72h group. TXL24h treatment could lowered rT2WI and rDWI to a certain extent, but had no significant difference from I/R24h group (P>0.05), while they were remarkably decreased in

TXL72h group compared to I/R24h group. However, rADC showed no significant difference between these 5 groups. The results were consisting with transient MCAO for 2.5 hours (Neumann et al., 2000). ADC value of ischemic area decrease sharply in hyperacute stage, reaches nadir at about 24h (Neumann et al., 2000), thereafter elevated to normal range, showing the “pseudonormalization”. The significantly elevated ADC values after occlusion may indicate total lysis of cells, resulting in a disruption of diffusion barriers. Therefore, rADC was not the best parameter for edema estimation. A limitation of this study was that the change of rADC value at different time points after reperfusion was not detected. Microglia in the brain has a prominent role in initiating and sustaining postischemic inflammation. The present study showed TXL pre-post treatment for 3days effectively inhibited microglia activation, therefore reduced post-ischemic inflammation. HMGB1 exists in nuclei under normal circumstances, regulating chromatin structure and gene transcription, whereas cytosolic HMGB1 is involved in inflammasome activation and autophagy. During ischemic injury, HMGB1 is related to molecular mechanisms of tissue damage. Circulating HMGB1 levels increase in patients with cerebral ischemia within hours after symptom onset (Goldstein et al., 2006). In rodent models, nuclear HMGB1 translocates from the neuronal cell nuclei into the cytoplasm within 1 hour after the onset of MCAO, and is immediately released into the extracellular space (Kim et al., 2006; kim et al., 2008; Qiu et al., 2008). In the case of neurons, the time-dependent translocation of HMGB1 can be summarized into 3 steps: (1) its redistribution inside the nucleus; (2) the translocation of HMGB1 from the nucleus into the cytosolic compartment; and (3) the release of HMGB1 into the extracellular space. The typical translocation of HMGB1 from the nucleus to the cytoplasm can be observed at 2-4 hours after reperfusion, and peaks at 24 hours after cerebral ischemia, lasts for 14 days, though gradually decreases at 48 hours and 72 hours (Zhang et al., 2011). Extracellular HMGB1 can bind to related cell signaling transduction receptors, such as the receptor for advanced glycation end products (RAGE), Toll-like receptor (TLR) 2, TLR4 and TLR9 (Tsung et al., 2014). Ischemia is a potent trigger for the regulated release of HMGB1, and TLR4 has been shown to be one pathway for HMGB1 in a potential feed-forward mechanism, resulting in the nuclear factor κB (NF-κB) translocating from cytosol to nucleus, which up-regulates leukocyte adhesion molecules and the production of proinflammatory cytokines like TNF-α in both hematopoietic and endothelial cells (Tsung et al., 2007). Neutralizing HMGB1 anti-bodies treatment has a neuroprotective effect on animals subjected to MCAO, reduces the magnitude of cell death and protects the blood-brain barrier (Liu et al., 2007). In the present study, a large number of HMGB1 could be seen in nucleus of brain tissue in sham group. The number decreased sharply 24 hours after reperfusion, which

indicated HMGB1 translocated from nuclei to extracellular space and activated TLR4. Activated TLR4 induced NF-κB moving from cytosol to nucleus, which was detected by comparing with nuclear reference Histone H3. This proinflammatory pathway could be inhibited with TXL pre-post treatment for 3days by hampering HMGB1 from activating. In conclusion, the present study demonstrated that TXL pre-post treatment for 3 reduced lesion area, alleviated brain edema, and inhibited post-ischemic inflammation. As both brain edema and HMGB1 have close relations to BBB, there may be some internal contact between them. Further studies should clarify the correlation between brain edema and HMGB1/TLR4-induced NF-κB pathway in ischemia/reperfusion rats.

Acknowledgments This work was supported by the National Natural Science Foundation of China (Grant no. 81173389 and no. 81202813) and the Key Diseases Construction Project of Shanghai Integrated Medicine (Grant no. zxbz 2012-10). The authors would like to thank all the members of their department for the experimental supports.

References Adén U., Dahlberg V., Fredholm B.B., Lai L.J., Chen Z., Bjelke B., 2002. MRI evaluation and functional assessment of brain injury after hypoxic ischemia in neonatal mice. Stroke 33, 1405-1410. Amiry-Moghaddam M., Otsuka T., Hurn P.D., Traystman R.J., Haug F.M., Froehner S.C., Adams M.E., Neely J.D., Agre P., Ottersen O.P., Bhardwaj A., 2003. An alpha-syntrophin-dependent pool of AQP4 in astroglial end-feet confers bidirectional water flow between blood and brain. Proceedings of the National Academy of Sciences of the United States of America 100, 2106-2111. Beaulieu C., de Crespigny A., Tong D.C., Moseley M.E., Albers G.W., Marks M.P., 1999. Longitudinal magnetic resonance imaging study of perfusion and diffusion in stroke: evolution of lesion volume and correlation with clinical outcome. Annals of Neurology 46, 568-578. Chen L., Wang X., Chen X., Xing S., Zhang J., Li J., Dang C., Liu G., Dang G., Zeng J., 2014. Tongxinluo attenuates neuronal loss and enhances neurogenesis and angiogenesis in the ipsilateral thalamus and improves neurological outcome after focal cortical infarction in hypertensive rats. Restorative Neurology and Neuroscience 32, 533-546.

Cheng X., Luo H., Zhou L., Wang L., Sun J., Huang Y., Luo E., Cai Y., 2014. Neuroprotective effect of the traditional Chinese herbal formula Tongxinluo: a PET imaging study in rats. Neural Regeneration Research 9, 1267-1274. Ding X.Q., Finsterbusch J., Wittkugel O., 2007. Apparent diffusion coefficient, fractional anisotropy and T2 relaxation time measuremeth. Clinical Neuroradiology 17, 230-238. Gonzalez R.G., Schaefer P.W., Buonanno F.S., Schwamm L.H., Budzik R.F., Rordorf G., Wang B., Sorensen A.G., Koroshetz W.J., 1999. Diffusion- weighted MR imaging: Diagnostic accuracy in patients imaged within 6 hours of str-oke symptom onset. Radiology 210, 155-162. Goldstein RS1, Gallowitsch-Puerta M, Yang L, Rosas-Ballina M, Huston JM, Czura CJ, Lee D.C., Ward M.F., Bruchfeld A.N., Wang H., Lesser M.L., Church A.L., Litroff A.H., Sama A.E., Tracey K.J., 2006. Elevated high-mobility group box 1 levels in patients with cerebral and myocardial ischemia. Shock 25, 571-574. Hunter A.J., Hatcher J., Virley D., Nelson P., Irving E., Hadingham S.J., Parsons A.A., 2000. Functional assessments in mice and rats after focal stroke. Neuropharmacology 39, 806-816. Juránek I., Baciak L., 2009. Cerebral hypoxia-ischemia: focus on the use of magnetic resonance imaging and spectroscopy in research on animals. Neurochemistry International 54, 471-480. Kimelberg H.K., 1995. Current concepts of brain edema. Review of laboratory investigations. Journal of Neurosurgery 83, 1051-1059. Kim J.B., Sig Choi J., Yu Y.M., Nam K., Piao C.S., Kim S.W., Lee M.H., Han P.L., Park J.S., Lee J.K., 2006. HMGB1, a novel cytokine-like mediator linking acute neuronal death and delayed neuroinflammation in the post-ischemic brain. Journal of Neuroscience 26, 6413-6421. Kim J.B., Lim C.M., Yu Y.M., Lee J.K., 2008. Induction and subcellular localization of high-mobility group box-1 (HMGB1) in the postischemic rat brain. Journal of Neuroscience Research 86, 1125-1131. Kovalenko T., Osadchenko I., Nikonenko A., Lushnikova I., Voronin K., Nikonenko I., Muller D., Skibo G., 2006. Ischemia-induced modification in hippocampal CA1 stratum radiatum excitatory synapses. Hippocampus 16, 814-825. Lan R., Xiang J., Wang G.H., Li W.W., Zhang W., Xu L.L., Cai D.F., 2013. XiaoXu-Ming Decoction Protects against Blood-Brain Barrier Disruption and

Neurological Injury Induced by Cerebral Ischemia and Reperfusion in Rats. Evidence-based Complementary and Alternative Medicine 2013, 629782. Liu K., Mori S., Takahashi H.K., Tomono Y., Wake H., Kanke T., Sato Y., Hiraga N., Adachi N., Yoshino T., Nishibori M., 2007. Anti-high mobility group box 1 monoclonal antibody ameliorates brain infarction induced by transient ischemia in rats. The FASEB Journal 21, 3904-3916. Liu Y., D'Arceuil H.E., Westmoreland S., He J., Duggan M., Gonzalez R.G., Pryor J., de Crespigny A.J., 2007. Serial diffusion tensor MRI after transient and permanent cerebral ischemia in nonhuman primates. Stroke 38, 138-145. Liu Y., Tang G.H., Sun Y.H., Lin X.J., Wei C., Yang G.Y., Liu J.R., 2013. The protective role of Tongxinluo on blood-brain barrier after ischemia-reperfusion brain injury. Journal of Ethnopharmacology 148, 632-639. Li L.M., Zheng B., Zhang R.N., Jin L.S., Zheng C.Y., Wang C., Zhou P.P., Guo Z.W., Ma D., Wen J.K., 2015. Chinese medicine Tongxinluo increases tighjunction protein levels by inducing KLF5 expression in microvascular endothelial cells. Cell Biochemistry and Function 33, 226-234. Longa E.Z., Weinstein P.R., Carlson S., Cummins R., 1989. Reversible middle cerebral artery occlusion without craniectomy in rats. Stroke 20, 84-91. Manley G.T., Fujimura M., Ma T., Noshita N., Filiz F., Bollen A.W., Chan P., Verkman A.S., 2000. Aquaporin-4 deletion in mice reduces brain edema after acute water intoxication and ischemic stroke. Nature Medicine 6, 159-163. Mohr J.P., Biller J., Hilal S.K., Yuh W.T., Tatemichi T.K., Hedges S., Tali E., Nguyen H., Mun I., Adams H.P. Jr, et al., 1995. Magnetic resonance versus computed tomographic imaging in acute stroke. Stroke 26, 807-812. Moseley M.E., Cohen Y., Mintorovitch J., Chileuitt L., Shimizu H., Kucharczyk J., Wendland M.F., Weinstein P.R., 1990. Early detection of regional cerebral ischemia in cats: comparison of diffusion- and T2-weighted MRI and spectroscopy. Magnetic Resonance in Medicine 14, 330-346 Neumann-Haefelin T., Kastrup A., de Crespigny A., Yenari M.A., Ringer T., Sun G.H., Moseley M.E., 2000. Serial MRI after transient focal cerebral ischemia in rats: dynamics of tissue injury, blood-brain barrier damage, and edema formation. Stroke 31, 1965-1972.

Papadopoulos M.C., Manley G.T., Krishna S., Verkman A.S., 2004. Aquaporin-4 facilitates reabsorption of excess fluid in vasogenic brain edema. The FASEB Journal 18, 1291-1293. Qiu J., Nishimura M., Wang Y., Sims J.R., Qiu S., Savitz S.I., Salomone S., Moskowitz M.A., 2008. Early release of HMGB-1 from neurons after the onset of brain ischemia. Journal of Cerebral Blood Flow and Metabolism 28, 927-938. Quast M.J., Huang N.C., Hillman G.R., Kent T.A., 1993. The evolution of acute stroke recorded by multimodal magnetic resonance imaging. Journal of Magnetic Resonance Imaging 11, 465-471. Ridder D.A., Schwaninger M., 2009. NF-kappaB signaling in cerebral ischemia. Neuroscience. 158, 995-1006. Shichita T., Ito M., Yoshimura A., 2014. Post-ischemic inflammation regulates neural damage and protection. Frontiers in Cellular Neuroscience 14, 319. Tsung A., Klune J.R., Zhang X., Jeyabalan G., Cao Z., Peng X., Stolz D.B., Geller D.A., Rosengart M.R., Billiar T.R., 2007. HMGB1 release induced by liver ischemia involves Toll-like receptor 4 dependent reactive oxygen species production and calcium-mediated signaling. Journal of Experimental Medicine 204, 2913-2923. Tsung A., Tohme S., Billiar T.R., 2014. High-mobility group box-1 in sterile inflammation. Journal of Internal Medicine 276, 425-443. Tuor U.I., Kozlowski P., Del Bigio M.R., Ramjiawan B., Su S., Malisza K, Saunders J.K.., 1998. Diffusion- and T2- weighted increases in magnetic resonance images of immature brain during hypoxia–ischemia: transient reversal posthyp-oxia. Experimental Neurology 150, 321-328. Vella J., Zammit C., Di Giovanni G., Muscat R., Valentino M., 2015. The central role of aquaporins in the pathophysiology of ischemic stroke. Frontiers in Cellular Neuroscience 9, 108. Wang B., Yang Q., Bai W.W., Xing Y.F., Lu X.T., Sun Y.Y., Zhao Y.X., 2014. Tongxinluo protects against pressure overload-induced heart failure in mice involving VEGF/Akt/eNOS pathway activation. PLoS One 9, e98047. Wang Y., Lin S.Z., Chiou A.L., Williams L.R., Hoffer B.J., 1997. Glial cell linederived neurotrophic factor protects against ischemia-induced injury in the cerebral cortex. The Journal of Neuroscience 17: 4341-4348.

Wey H.Y., Desai V.R., Duong T.Q., 2013. A review of current imaging methods used in stroke research. Neurological Research 36, 1092-1102. Yang, G., Chan, P.H., Chen, J., Carlson, E., Chen, S.F., Weinstein, P., Epstein, C.J., Kamii, H., 1994. Human copper-zinc superoxide dismutase transgenic mice are highly resistant to reperfusion injury after focal cerebral ischemia. Stroke 25, 165–170. Yuh W.T., Crain M.R., Loes D.J., Greene G.M., Ryals T.J., Sato Y., 1991. MR imaging of cerebral ischemia: findings in the first 24 hours. American Journal of Neuroradiology 12, 621-629. Zhang J., Takahashi H.K., Liu K., Wake H., Liu R., Maruo T., Date I., Yoshino T., Ohtsuka A., Mori S., Nishibori M., 2011. Anti-high mobility group box 1 monoclonal antibody protects the blood-brain barrier from ischemia-induced disruption in rats. Stroke 42, 1420-1428.