Therapeutic impact of eicosapentaenoic acid on ischemic brain damage following transient focal cerebral ischemia in rats

brain research 1519 (2013) 95–104

Available online at www.sciencedirect.com

www.elsevier.com/locate/brainres

Research Report

Therapeutic impact of eicosapentaenoic acid on ischemic brain damage following transient focal cerebral ischemia in rats Masayuki Uedan, Toshiki Inaba, Chikako Nito, Nobuo Kamiya, Yasuo Katayama Department of Neurology, Nippon Medical School, 1-1-5 Sendagi, Bunkyo-ku, Tokyo 113-8603, Japan

art i cle i nfo

ab st rac t

Article history:

Long-chain n-3 polyunsaturated fatty acids, such as eicosapentaenoic acid (EPA), have been

Accepted 24 April 2013

shown to reduce ischemic neuronal injury. We investigated the effects of ethyl-EPA (EPA-E) on

Available online 2 May 2013

ischemic brain damage using a rat transient focal cerebral ischemia model. Male Sprague–

Keywords:

Dawley rats (n¼ 105) were subjected to 90 min of focal cerebral ischemia. EPA-E (100 mg/kg/

Eicosapentaenoic acid

day) or vehicle was administered once a day for 3, 5 or 7 days prior to ischemia. Different

Oxidative stress

withdrawal intervals of 3, 5, and 7 days prior to ischemia following 7-day pretreatment with

Rho-kinase

EPA-E or vehicle were also examined. In addition, post-ischemic administration of EPA-E was

Focal cerebral ischemia

investigated. Pretreatment with EPA-E for 7 and 5 days, but not 3 days, showed significant infarct volume reduction and neurological improvements when compared with vehicle pretreatment. In addition, withdrawal of EPA-E administration for 3 days, but not 5 and 7 days, also demonstrated significant infarct volume reduction and neurological improvements when compared with vehicle treatment. Post-ischemic treatment of EPA-E did not show any neuroprotection. Immunohistochemistry revealed that 7-day pretreatment with EPA-E significantly reduced cortical expression of 8-hydroxydeoxyguanosine (maker for oxidative DNA damage), 4-hydroxy-2-nonenal (maker for lipid peroxidation), phosphorylated adducin (marker for Rho-kinase activation) and von Willebrand factor (endothelial marker) when compared with vehicle pretreatment. In addition, phosphorylated adducin expression co-localized with von Willebrand factor immunoreactivity. The present study established the neuroprotective effect of EPA-E on ischemic brain damage following transient focal cerebral ischemia in rats, which may be involved in the suppression of oxidative stress and endothelial Rho-kinase activation. & 2013 Elsevier B.V. All rights reserved.

1.

Introduction

Long-chain n-3 polyunsaturated fatty acids (PUFAs), such as eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA),

n

Corresponding author. Fax: +81 3 3822 4865. E-mail addresses: [email protected], [email protected] (M. Ueda).

0006-8993/$ - see front matter & 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.brainres.2013.04.046

which are derived from marine products, have attracted considerable attention. Early epidemiological studies in Greenland discovered that the Inuit people showed a lower incidence of myocardial infarction than Danish people (Bang et al., 1971),

96

brain research 1519 (2013) 95–104

and EPA from a fish-rich diet was considered to be responsible for preventing thrombosis and atherosclerosis in the Inuit (Dyerberg et al., 1978). Furthermore, a recent meta-analysis of cohort studies disclosed an inverse relationship between fish consumption and stroke risk (He et al., 2004). A large amount of highly purified EPA can be prepared in an ethyl-ester form (EPAE) with an approximately 90% purity (Mizuguchi et al., 1993), and EPA-E has been clinically used as a lipid-lowering medication in Japan. In addition, a recent clinical controlled trial of the Japan EPA lipid intervention study (JELIS) has shown that EPA-E reduces stroke recurrence in Japanese hypercholesterolemic patients (Tanaka et al., 2008). We have previously shown that long-term administration of EPA-E (100 mg/kg/day) ameliorated the age-related decline of cerebral blood flow (CBF) in stroke-prone spontaneously hypertensive rats (Katayama et al., 1997). Furthermore, we have also reported that post-ischemic delayed administration of EPA-E (100 mg/kg/day) for 4 weeks increased local CBF within the peri-infarct areas in a rat chronic cerebral infarction model (Katsumata et al., 1999). In addition, several investigations have described the neuroprotective effects of EPA on forebrain ischemia (Okabe et al., 2011; Bas et al., 2007; Ozen et al., 2008; Ajami et al., 2011). However, the protective effects of EPA on acute focal cerebral ischemia remain unclear. Cerebral ischemia consists of complex pathological processes, and various factors can exacerbate ischemic brain damage. Oxidative stress is known to be involved in ischemiareperfusion injury owing to an increase in reactive oxygen species (ROS), which result in DNA damage and lipid peroxidation (Floyd and Carney, 1992; Halliwell, 1992). In addition, Rhokinase activity is associated with various cerebral vascular diseases, including ischemic brain injury (Chrissobolis and Sobey, 2006). A recent investigation revealed that endothelial Rho-kinase activation following cerebral ischemia played an important role for infarct expansion (Yagita et al., 2007). Furthermore, ROS-induced brain endothelial dysfunction involves the Rho-kinase signaling pathway (Kahles et al., 2007). Therefore oxidative stress and subsequent endothelial Rho-kinase activation may be implicated in ischemic brain injury. The present study aimed to examine the effects of preischemic and post-ischemic treatments of EPA-E on ischemic brain damage using a rat transient focal cerebral ischemia model. Furthermore we sought to determine whether EPA-E could suppress oxidative DNA damage, lipid peroxidation and endothelial Rho-kinase activation following focal cerebral ischemia.

2.

Results

2.1.

Plasma fatty acids

Plasma EPA levels were significantly elevated after 7-day pretreatment with EPA-E compared with vehicle treatment (p¼ 0.0019), although plasma arachidonic acid (AA) levels were not different between the groups (p¼0.9701) (Table 1). In experiment 2, plasma EPA concentrations were significantly higher even with withdrawal of EPA-E administration for 3 days compared with vehicle administration (p¼0.0022);

however, EPA-E withdrawal for more than 5 days showed no differences in plasma EPA levels between the groups (Table 1). Plasma AA levels were not different between the groups at any withdrawal interval. In experiment 3, plasma EPA levels were significantly elevated following high-dose EPA-E administration (600 mg/kg/day) compared with the other treatments (Table 2). In contrast, plasma AA levels were significantly decreased following high-dose EPA-E administration compared with the other treatments (Table 2). There were no statistical differences in plasma EPA and AA levels between animals treated with 100 mg/kg/day of EPA-E and vehicle (Table 2).

2.2.

Magnetic resonance imaging (MRI)

Fig. 1A displays representative CBF and apparent diffusion coefficient (ADC) images during ischemia in animals subjected to focal cerebral ischemia following 7-day pretreatment with EPA-E or vehicle. Decreased CBF area in EPA-E and vehicle treated groups was 50.079.1 mm2 and 48.174.0 mm2, respectively. Reduced ADC area in EPA-E and vehicle treated groups was 12.777.4 mm2 and 30.276.0 mm2, respectively. The decreased ADC area was significantly smaller in the EPAE treated group when compared with the vehicle-treated group (p¼ 0.0034), although reduced CBF areas were not different between the groups (p ¼0.6956) (Fig. 1A).

2.3.

Infarct volumes

Typical 2,3,5-triphenyltetrazolium chloride (TTC)-stained brain sections, obtained from the present study, are shown in Fig. 1B–D. Fig. 1B displays representative TTC-stained brain sections from experiment 1. In the 3-day pretreatment group, the cortical and striatal infarct volumes (EPA-E versus vehicle) were 180.0745.8 versus 162.8730.3 mm3 (p¼0.5046) and 60.8712.7 versus 53.8714.1 mm3 (p¼ 0.4339), respectively. In the 5-day pretreatment group, the cortical and striatal infarct volumes were 51.8753.5 versus 128.9741.9 mm3 (p¼ 0.0349) and 58.576.2 versus 65.9710.9 mm3 (p¼0.2277), respectively. In the 7-day pretreatment group, the cortical and striatal infarct volumes at 24 h after reperfusion were 80.6729.9 versus 145.6733.9 mm3 (p¼0.0123) and 43.277.2 versus 63.776.3 mm3 (p¼ 0.0014), respectively. In addition, the cortical and striatal infarct volumes at 72 h after reperfusion following 7-day pretreatment were 52.4747.4 versus 159.7742.9 mm3 (p¼0.0056) and 36.078.9 versus 58.774.2 mm3 (p¼ 0.0009), respectively. Pretreatment with EPA-E for 7 and 5 days, but not 3 days, showed significant reduction in cortical infarct volumes, compared with the vehicle pretreatment (Fig. 2). Furthermore, 7-day pretreatment with EPA-E resulted in significant infarct volume reduction even in the striatum compared with vehicle pretreatment (Fig. 2). The protective effects, observed in animals with 7-day pretreatment of EPA-E, persisted for 72 h after reperfusion (Fig. 2). Fig. 1C shows representative TTC-stained brain sections from experiment 2. In the 3-day withdrawal group, the cortical and striatal infarct volumes (EPA-E versus vehicle) were 28.6710.9 versus 135.2749.4 mm3 (p¼0.0015) and 57.777.7 versus 62.1710.1 mm3 (p¼0.4683) at 24 h after reperfusion, respectively. In the 5-day withdrawal, the cortical and striatal

97

brain research 1519 (2013) 95–104

Table 1 – Plasma levels of eicosapentaenoic acid and arachidonic acid.

7-day pretreatment 3-day withdrawal 5-day withdrawal 7-day withdrawal

EPA (μg/mL) AA (lg/mL) EPA (lg/mL) AA (μg/mL) EPA (μg/mL) AA (μg/mL) EPA (μg/mL) AA (μg/mL)

EPA-E

Vehicle

p-value

15.172.8 * 222.3752.4 16.172.6 * 289.7746.1 15.974.2 260.2716.4 11.875.6 238.8768.2

8.072.1 223.8768.1 6.474.2 267.1759.5 12.072.0 288.1764.4 11.074.3 288.1764.3

0.0019 0.9701 0.0022 0.5217 0.0978 0.3755 0.8052 0.2731

EPA-E and Vehicle indicate treatment with EPA-E and vehicle, respectively. The “7-day pretreatment” indicates drug administration by gavage once a day for 7 days prior to ischemia. The “withdrawal” indicates drug withdrawal for 3, 5 or 7 days prior to ischemia following 7-day pretreatment of either EPA-E or vehicle. EPA and AA indicate eicosapentaenoic acid and arachidonic acid, respectively. n indicates significant difference from vehicle treatment by analysis of variance (ANOVA) with Scheffe's post-hoc test.

Table 2 – Plasma levels of eicosapentaenoic acid and arachidonic acid following 3-day post-ischemic treatment. EPA-E (600 mg) EPA (lg/mL) AA (lg/mL)

,

56.5717.4 * ** 147.5735.2 *,**

EPA-E (100 mg)

Vehicle

20.873.1 238.1734.3

26.073.5 230.8722.3

EPA-E (600 mg), EPA-E (100 mg) and Vehicle indicates post-ischemic treatment with 600 mg/kg/day of EPA-E, 100 mg/kg/day of EPA-E and vehicle for 3 days, respectively. EPA and AA indicate eicosapentaenoic acid and arachidonic acid, respectively. n indicates significant difference from vehicle treatment by analysis of variance (ANOVA) with Scheffe's post-hoc test (po0.05). nn indicates significant difference from EPA-E (100 mg/kg/day) treatment by analysis of variance (ANOVA) with Scheffe's post-hoc test (po0.05).

infarct volumes were 86.7746.8 versus 130.4737.2 mm3 (p¼0.1409) and 63.7718.7 versus 54.578.9 mm3 (p¼0.3502), respectively. In the 7-day withdrawal, the cortical and striatal infarct volumes were 130.3749.5 versus 145.8719.5 mm3 (p¼0.5346) and 66.779.7 versus 54.9714.4 mm3 (p¼0.1670), respectively. Moreover, the cortical and striatal infarct volumes at 72 h after reperfusion in the 3-day withdrawal group were 80.9749.8 versus 158.8722.8 mm3 (p¼ 0.013) and 67.578.1 versus 66.0716.5 mm3 (p¼ 0.8634), respectively. The withdrawal of EPA-E administration for 3 days, but not 5 and 7 days, also demonstrated significant reduction in cortical infarct volumes when compared with the vehicle treatment (Fig. 3A). Fig. 1D shows representative TTC-stained brain sections from experiment 3. The cortical infarct volumes in animals with post-ischemic treatment with vehicle, EPA-E (100 mg/kg/ day) and EPA-E (600 mg/kg/day) were 160.5741.6, 133.0736.1 and 141.4710.7 mm3, respectively. The striatal infarct volumes in animals with post-ischemic treatment of vehicle, EPA-E (100 mg/kg/day) and EPA-E (600 mg/kg/day) were 57.373.6, 57.375.3 and 61.476.7 mm3, respectively. The post-ischemic administration of EPA-E did not show infarct volume reduction even following high-dose treatment (Fig. 3B).

2.4.

Neurological scores

Following 3-day pretreatment, the hemiparesis and posture scores (median, EPA-E versus vehicle) at 24 h after reperfusion were 2 versus 3 (p¼ 0.2207) and 2 versus 3 (p¼0.0719),

respectively. In the 5-day pretreatment group, the hemiparesis and posture scores at 24 h after reperfusion were 1 versus 3 (p¼0.0086) and 1 versus 2 (p¼0446), respectively. In the 7-day pretreatment group, the hemiparesis and posture scores at 24 h after reperfusion were 1 versus 3 (p¼ 0.0201) and 1 versus 2 (p¼0.0312), respectively. In addition, the hemiparesis and posture scores at 72 h after reperfusion were 1 versus 3 (p¼0.0539) and 1 versus 2 (p¼0.0109), respectively. Pretreatment with EPA-E for 7 and 5 days, but not 3 days, showed significant improvement both in the hemiparesis and posture scores when compared with the vehicle pretreatment (Fig. 2). The neurological improvement in the posture score, observed in animals with 7-day pretreatment of EPA-E, persisted for 72 h after reperfusion (Fig. 2). In the 3-day withdrawal group, the hemiparesis and posture scores (median, EPA-E versus vehicle) at 24 h after reperfusion were 1 versus 2 (p¼ 0.0374) and 1 versus 3 (p¼0.0177), respectively. In the 5-day withdrawal, the hemiparesis and posture scores at 24 h after reperfusion were 2 versus 2 (p¼ 0.9999) and 2 versus 2 (p¼ 0.3390), respectively. In the 7-day withdrawal, the hemiparesis and posture scores at 24 h after reperfusion were 2 versus 3 (p¼ 0.4189) and 2 versus 3 (p¼0.5485), respectively. In the 3-day withdrawal group, the hemiparesis and posture scores at 72 h after reperfusion were 2 versus 3 (p¼ 0.0283) and 1 versus 2 (p¼0.0472), respectively. The withdrawal of EPA-E administration for 3 days, but not 5 and 7 days, also demonstrated significant improvement both in the hemiparesis and posture scores when compared with the vehicle treatment (Fig. 3A).

98

brain research 1519 (2013) 95–104

Withdrawal interval Pretreatment interval

MRI images during ischemia Vehicle (24 h)

CBF

ADC

3-day (24 h)

EPA-E

Vehicle

EPA-E Vehicle

3-day (24 h)

5-day (24 h)

7-day (24 h)

3-day (72 h) 5-day (24 h)

Post-ischemic treatment

60 50

7-day (24 h)

40 30 20

EPA-E 100mg (72 h)

*

10 0

CBF

Vehicle (72 h)

ADC

7-day (72 h)

EPA-E 600mg (72 h)

Fig. 1 – Magnetic resonance imaging and TTC-stained brain sections. (A): MRI during ischemia. Representative intra-ischemic cerebral blood flow (CBF) and apparent diffusion coefficient (ADC) images at the level of bregma in animals subjected to 90 min of focal cerebral ischemia following 7-day pretreatment with EPA-E and vehicle (upper panel). Note the obvious small low ADC area in EPA-E treated animals. The decreased ADC area was significantly smaller in the EPA-E treated group when compared with the vehicle-treated group, although reduced CBF areas were not different between between groups (lower panel). * indicates a significant difference from vehicle treatment by ANOVA with Scheffe's post-hoc test. (B): Representative TTC-stained brain sections in experiment 1. Animals were subjected to 90 min of focal cerebral ischemia, followed by 24 h (24 h) or 72 h (72 h) reperfusion. Pretreatment intervals for EPA-E (100 mg/kg/day) were 3 (3-day), 5 (5-day) and 7 days (7-day). Note the small infarction at 24 h after reperfusion in animals with 5-day and 7-day pretreatments of EPA-E, as well as the small infarction at 72 h after reperfusion in animals with 7-day pretreatments of EPA-E. (C): Representative TTC-stained brain sections in experiment 2. Animals were subjected to 90 min of focal cerebral ischemia, followed by 24 h (24 h) or 72 h (72 h) of reperfusion. Withdrawal intervals were 3 (3-day), 5 (5day) and 7 days (7-day) prior to ischemia following 7-day pretreatment with EPA-E (100 mg/kg/day) or vehicle. Note the small infarction at 24 h after reperfusion in animals with 3-day withdrawal of EPA-E, as well as the small infarction at 72 h after reperfusion in animals with 3-day withdrawal of EPA-E. (D): Representative TTC-stained brain sections in experiment 3. Animals were subjected to 90 min of focal cerebral ischemia, followed by 72 h (72 h) of reperfusion. Post-ischemic administration of EPA-E (100 or 600 mg/kg/day) or vehicle was carried out for 3 days. Note no obvious differences among the animals.

The hemiparesis scores (median) at 72 h after reperfusion in animals with post-ischemic treatment with vehicle, EPA-E (100 mg/kg/day) and EPA-E (600 mg/kg/day) were 2, 1 and 2, respectively (p¼ 0.5344). The posture scores at 72 h after reperfusion in animals with post-ischemic treatment with vehicle, EPA-E (100 mg/kg/day) and EPA-E (600 mg/kg/day) were 3, 2 and 2, respectively (p¼ 0.1112). The post-ischemic administration of EPA-E showed no neurological score improvement even following high-dose treatment (Fig. 3B).

2.5.

Immunohistochemical evaluation

Representative microphotographs of the immunohistochemistry in the cortical regions are shown in Fig. 4A. The vehicle treated animals displayed obvious cortical 8-hydroxydeoxyguanosine (8-OHdG), 4-hydroxy-2-nonenal (4-HNE), phosphorylated adducin (p-adducin) and von Willebrand factor (vWF) expression when

compared with EPA-E treated animals. Double immunohistochemistry revealed that p-adducin expression co-localized with vWF-positive vessels (Fig. 4B). The numbers of 8-OHdG-positive cells in the cortex and striatum (EPA-E versus vehicle) were 16.6715.4 versus 205.8728.4/1.33 mm2 (po0.0001) and 225.677.7 versus 240.6734.4 mm3 (p¼ 0.4046), respectively. The numbers of 4-HNE-positive cells in the cortex and striatum were 17.4711.8 versus 185.0716.2/1.33 mm2 (po0.0001) and 194.0727.9 versus 219.0741.2 mm3 (p¼0.2936), respectively. The numbers of padducin-positive cells in the cortex and striatum were 89.4714.1 versus 277.6784.2/1.33 mm2 (p¼ 0.0012) and 253.0762.8 versus 271.6765.6 mm3 (p¼0.6592), respectively. The numbers of vWF-positive cells in the cortex and striatum were 170.0745.6 versus 329.2768.8/1.33 mm2 (p¼ 0.0026) and 395.27126.3 versus 382.07173.3 mm3 (p¼ 0.8939), respectively. EPA-E treated animals showed a significant decrease in the numbers of 8-OHdG-, 4-HNE-, p-adducin- and vWF-positive cells

99

brain research 1519 (2013) 95–104

3-day Pretreatment 24 hours after reperfusion

5-day Pretreatment 24 hours after reperfusion

7-day Pretreatment 24 hours after reperfusion

7-day Pretreatment 72 hours after reperfusion

250 200 150

*

*

100

* *

*

**

**

50 0

3

**

2

**

**

1 0

EPA-E

Vehicle

Fig. 2 – Infarct volumes and neurological scores in experiment 1. Pretreatment with EPA-E for 7 and 5 days, but not 3 days, showed significant reduction in cortical infarct volumes, as well as significantly improved neurological scores when compared with vehicle pretreatment. Furthermore, 7-day pretreatment with EPA-E resulted in significant infarct volume reduction even in the striatum when compared with vehicle pretreatment. The protective effects observed in animals with 7-day pretreatment of EPA-E persisted for 72 h after reperfusion. * indicates significant difference from vehicle treated animals by ANOVA with Scheffe's post-hoc test (po0.05). ** indicates significant difference from vehicle treated animals by Mann–Whitney U test. Box plots indicate the median and interquartile range, and whiskers indicate the 5th and 95th percentiles (po0.05).

3-day Withdrawal 24 hours after reperfusion

5-day Withdrawal 24 hours after reperfusion

7-day Withdrawal 24 hours after reperfusion

3-day Withdrawal 72 hours after reperfusion

250 200 150

*

100 50

*

0

3 2

**

**

**

**

1

EPA-E Vehicle

0

3-day Post-ischemic treatment / 72 hours after reperfusion 250 200

3

150

2

100

1

50

0

0

EPA(100mg/kg) EPA(600mg/kg) Vehicle

Fig. 3 – Infarct volumes and neurological scores in experiment 2 and experiment 3. (A): The withdrawal of EPA-E administration for 3 days, but not 5 and 7 days, showed a significant reduction in cortical infarct volumes and significant improvement of neurological scores when compared with the vehicle treatment. (B): The post-ischemic administration of EPA-E (100 mg and 600 mg/kg/day) did not show any neuroprotection, including infarct volume reduction and neurological score improvement, even in the high-dose treatment group. * indicates significant difference from vehicle treated animals by ANOVA with Scheffe's post-hoc test (po0.05). ** indicates significant difference from vehicle treated animals by Mann–Whitney U test. Box plots indicate the median and interquartile range, and whiskers indicate the 5th and 95th percentiles (po0.05).

100

brain research 1519 (2013) 95–104

Double immunohistochemistry

EPA-E

Vehicle p-adducin p-adducin

4-HNE

8-OHdG

8-OHdG

Merge

vWF

p-adducin

4-HNE

300

600

250

500

200

400

150

300

100

200

50

100

*

vWF

*

vWF

* *

0

0

Cortex

Striatum

Cortex

Striatum

Cortex

Striatum

Cortex Vehicle

Striatum EPA-E

Fig. 4 – Immunohistochemical analyses. (A): Representative microphotographs of immunohistochemistry. Note the increased cortical expressions of 8-hydroxydeoxyguanosine (8-OHdG), 4-hydroxy-2-nonenal (4-HNE), phosphorylated adducin (p-adducin) and von Willebrand factor (vWF) in the vehicle treated animal, while relatively weak immunohistochemical expressions was observed in EPA-E treated animals. Bars indicate 50 μm. (B): Double immunohistochemistry. Green fluorescent signals indicate p-adducin expression (left panel), and red fluorescent signals indicate vWF expression (middle panel). Yellow fluorescent signals (arrows) in the merged image indicate co-localization with p-adducin and vWF expressions (right panel). Bars indicate 50 μm. (C): Immunohistochemical cell counts. EPA-E treated animals showed a significant decrease in the number of 8-OHdG-, 4-HNE-, p-adducin- and vWF-positive cells in the cortical ischemic boundary area, but not in the striatum when compared with vehicle treated animals. * indicates significant difference from vehicle treated animals by ANOVA with Scheffe's post-hoc test (po0.05).

in the cortical ischemic boundary area, but not in the striatum when compared with vehicle treated animals (Fig. 4C).

3.

Discussion

Several investigators have studied the effects of n-3 PUFAs on cerebral ischemia (Okabe et al., 2011; Bas et al., 2007; Ozen et al., 2008; Ajami et al., 2011; Cao et al., 2004; Pan et al., 2009). The examined n-3 PUFAs in each study were EPA, DHA or a mixture of EPA/DHA. Okabe and colleagues reported that EPA inhibited inflammatory responses and oxidative damage following transient forebrain ischemia in gerbils (Okabe et al., 2011), and Bas and colleagues studied the effects of fish oil containing both EPA and DHA on hippocampal neurons and the antioxidant enzymatic activities in a rat transient forebrain ischemia model (Bas et al., 2007). Similar results to Bas were observed in the prefrontal cortex by Ozen and colleagues (Ozen et al., 2008). Ajami and colleagues reported that combination treatment with EPA and DHA modulated hippocampal Bcl-2/Bax expression following transient forebrain ischemia in rats (Ajami et al., 2011), and Pan and colleagues examined the effects of DHA on rat focal cerebral ischemia (Pan et al., 2009). They found increased

antioxidant enzymatic activities following chronic daily DHA administration as well as upregulation of extracellularsignal-regulated kinase and Bcl-2 (Pan et al., 2009). However, details on the protective effects of monotherapy with highly purified EPA-E on transient focal ischemia have not been investigated. In the present study, we showed that 7-day pretreatment with EPA-E increased plasma EPA concentrations, reduced infarct volumes and improved neurological scores in a rat transient focal ischemia model, and that 5-day, but not 3-day, pretreatment with EPA-E was also neuroprotective. In addition, the impact of withdrawal intervals on the neuroprotective effects of 7-day pretreatment with EPA-E was investigated in experiment 2. The protective effects of 7-day pretreatment with EPA-E still remained after 3-day withdrawal together with increased plasma EPA concentrations, while the effects disappeared after withdrawal for 5 days or more, in which plasma EPA concentrations were not high anymore. In contrast, post-ischemic administration of EPA-E was not neuroprotective even at higher doses (600 mg/kg/ day), regardless of marked increase in plasma EPA concentration together with decrease in plasma AA levels at 3 days after administration. Therefore, continuous EPA-E administration to some extent may be necessary to acquire

brain research 1519 (2013) 95–104

neuroprotection against cerebral ischemia, probably requiring high EPA concentrations at the ischemic insult. Philbrick and colleagues demonstrated that cellular membrane phospholipids of the brain became enriched with EPA following EPA-rich feeding (Philbrick et al., 1987). The biochemical EPA accumulation in cellular membrane phospholipids results in partial replacement of endogenous AA available for production of thromboxane-A2, a strong inducer of vasoconstriction and platelet aggregation (Holub, 1988). Moreover, EPA is converted into thromboxane-A3, which does not promote platelet aggregation, instead of thromboxane-A2 [27]. Therefore, EPA-rich conditions may be favorable for reducing ischemic brain damage. We showed that EPA-E pretreatment suppressed 8-OHdG and 4-HNE expression in the cortical ischemic boundary zone following transient focal cerebral ischemia. In a rat middle cerebral artery (MCA) occlusion model, burst-like ROS production occurs after reperfusion, and ROS formation is also increased during ischemia (Peters et al., 1998). The generated free radicals during cerebral ischemia-reperfusion are known to contribute to neuronal cell death (Kitagawa et al., 1990). 8OHdG is a major product of free radical-induced oxidative DNA damage, and 4-HNE is a major lipid peroxidation product generated by free radical injury on n-6 PUFAs, such as AA, in the cellular membrane phospholipids (Dalle-Donne et al., 2006). The toxic 4-HNE is also known to be an important mediator for apoptosis (Dalle-Donne et al., 2006). Recent investigations disclosed an accumulation of 8-OHdG and 4HNE within the penumbral tissues in a mouse transient focal ischemia model (Zhang et al., 2005), consistent with the present observation. Previous studies have shown that treatment with n-3 PUFAs enhance antioxidant enzymes following transient cerebral ischemia (Okabe et al., 2011; Bas et al., 2007; Ozen et al., 2008; Cao et al., 2004; Pan et al., 2009). Furthermore, EPA-rich conditions itself, which lead to a relative decrease in AA in membrane phospholipids, may also reduce 4-HNE production, because 4-HNE is generated from AA (Dalle-Donne et al., 2006). We also demonstrated that EPA-E pretreatment suppressed p-adducin and vWF expression in the cortical ischemic boundary zone following transient focal cerebral ischemia. Because endothelial injury results in increased vWF expression (Lowenstein et al., 2005), the decreased immunoreactivity of vWF in EPA-E treated animals may suggest a protective affect of EPA-E in endothelial damage following ischemia. The adducin is a specific substrate of Rho-kinase, and its phosphorylation is associated with Rhokinase activation (Kimura et al., 1998). Hypoxia-induced decrease in endothelial nitric oxide synthase is mediated by Rho-kinase (Takemoto et al., 2002), and EPA-E enhances endothelial nitric oxide production in vitro (Okuda et al., 1997). In addition, EPA reportedly inhibits Rho-kinase mediated Ca2+ sensitization of vascular smooth muscle contraction, which is involved in cerebral vasospasm after subarachnoid hemorrhage (Shirao et al., 2008). Furthermore, AA is known to induce Rho-kinase activation (Feng et al., 1999). Therefore, the present data indicated that EPA-rich conditions suppressed oxidative stress, including oxidative DNA damage and lipid peroxidation, during ischemia-reperfusion, and that EPA-E pretreatment inhibited endothelial damage

101

and subsequent Rho-kinase activation following transient ischemia. The present MRI analyses disclosed intra-ischemic neuroprotection by EPA-E. We have previously confirmed, using an autoradiography technique, that long-term EPA-E administration did not change CBF in the normal hemisphere in a rat chronic cerebral infarction model (Katsumata et al., 1999). Thus the EPA-E and vehicle treated animals in the present study were, at least, estimated to show similar CBF until ischemia induction. Precisely, the present data did not clarify whether EPA-E pretreatment could modulate CBF during ischemia, because animals with unsuccessful MCA occlusion were excluded from the study based on the CBF images. Surprisingly, the EPA-E treated animals showed significantly less ADC reduction than the vehicle treated animals under similar CBF reduction between the groups, indicating the presence of neuroprotective mechanisms, which might be independent of CBF, even before reperfusion. Although ischemia-induced ROS formation explosively increases after reperfusion, ROS is also produced during ischemia (Peters et al., 1998). Therefore, pretreatment with EPA-E might exert antioxidant activity during ischemia, leading to intraischemic neuroprotection. However, additional factors contributing to the neuroprotection of EPA-E pretreatment should be elucidated, and further investigations are needed to determine the precise mechanisms of the effects of EPA. In conclusion, the present study established the neuroprotective effects of EPA-E on ischemic brain damage following transient focal cerebral ischemia, which might be involved in the suppression of oxidative stress and endothelial Rho-kinase activation.

4.

Experimental procedure

All experimental protocols were carried out in accordance with the institutional guidelines of Nippon Medical School for animal care and use.

4.1.

Focal cerebral ischemia

In the present study, 105 male Sprague–Dawley rats, weighing 250–300 g, were used. Anesthesia was initially induced with 5% (v/v) halothane, and then maintained with 1% (v/v) halothane in a mixture of 70% (v/v) N2O and 30% (v/v) O2 under spontaneous breathing, during the surgical procedure. Animals were subjected to 90 min of focal cerebral ischemia using an intraluminal suture technique following an overnight fast, as described previously (Nito et al., 2011). Briefly, the left common, internal and external carotid arteries were carefully exposed through a midline cervical incision, and the common and external carotid arteries were doubly ligated using 4-0 silk sutures. Focal cerebral ischemia was induced by inserting a silicone rubber-coated 4-0 nylon thread with a round-tip through the left internal carotid artery (approximately 18 mm from the bifurcation) for 90 min to occlude the origin of the MCA, and reperfusion was achieved by withdrawing the thread. Rectal temperature was maintained at 3770.5 1C using a heating pad and lamp during ischemia and up to 2 h after reperfusion.

102

4.2.

brain research 1519 (2013) 95–104

Drug administration and experimental groups

The highly purified EPA-E (Mochida Pharmaceutical, Tokyo, Japan) was used in the study. In the preliminary study, animals were given a suspension of EPA-E (30, 100, or 300 mg/kg/day) or vehicle by gavage once a day for 7 days, and they were subjected to 90 min of focal cerebral ischemia, followed by 24 h of reperfusion (n¼5, each). Total infarct volumes were significantly reduced in animals treated with medium and high doses of EPA-E (100 and 300 mg/kg/day), but not low doses (30 mg/kg/day) when compared with vehicle-treated animals. There was no difference between the high and medium dose-treated animals (data not shown). Therefore, EPA-E at a dose of 100 mg/kg/day by gavage administration was used as the pretreatment concentration in all subsequent experiments. In experiment 1, rats were administered EPA-E (100 mg/kg/ day) or vehicle by gavage once a day for 3, 5 or 7 days prior to ischemia induction, and then subjected to focal cerebral ischemia, followed by 24 h of reperfusion (n¼5, each). In addition, focal cerebral ischemia was similarly introduced in other rats after 7-day pretreatment with EPA-E (100 mg/kg/ day) or vehicle, followed by 72 h of reperfusion (n ¼5, each). In experiment 2, rats were also subjected to focal cerebral ischemia with different withdrawal intervals of 3, 5 or 7 days after 7-day pretreatment with EPA-E (100 mg/kg/day) or vehicle, followed by 24 or 72 h of reperfusion (n¼ 5, each). In experiment 3, rats were subjected to focal cerebral ischemia, followed by 72 h of reperfusion, and post-ischemic administration of EPA-E (100 or 600 mg/kg/day) or vehicle was carried out once a day for 3 days (n ¼5, each). The first administration was done immediately after reperfusion.

4.3.

MRI experiments

MRI experiments were performed just prior to reperfusion using an 18 cm bore 7T horizontal magnet (Magnex Scientific, Abingdon, UK) with a Varian Unity-INOVA-300 (Varian Inc., Palo Alto, CA, USA) system equipped with an actively shielded gradient. A 6 cm internal diameter quadrature birdcage coil was used in the study. The rat was anesthetized with 1% (v/v) halothane in 70% (v/v) N2O/30% (v/v) O2 mixture and positioned on its back in a self-made stereotaxic holder. The axial position of the rat was adjusted until the image slice was 5 mm caudal to the rhinal fissure in the center of the coil. The continuous arterial spin labeling method (Williams et al., 1992), modified with a centrally encoded variable-tipangle gradient echo technique (Ewing et al., 2003), with a repetition time (TR) of 5 ms and echo time (TE) of 3 ms, was performed to obtain CBF images. A coronal slice at the level of bregma was obtained with a 2 mm thickness, a field of view (FOV) of 50  50 mm2 and a matrix size of 128  128. Inflowing protons of the arterial blood were labeled by adiabatic inversion with an axial gradient of 74.5 kHz/mm and a continuous radio frequency transmission of approximately 0.6 kHz at a frequency offset of 79 kHz, which placed the inversion plane 20 mm below the imaging plane. Forty-eight pairs of images were produced with and without the inversion of inflowing protons, and were summed to improve the signal to noise ratio. Diffusion-weighted imaging (DWI) was also performed using a Stejskal–Tanner spin-echo sequence with a TE of

40 ms and TR of 2000 ms, as described previously (Okubo et al., 2007). A coronal slice at the level of bregma was obtained with a 2 mm thickness, a FOV of 50  50 mm2 and a matrix size of 128  128. Diffusion weighting used b-values of 100, 750 and 1500 s/mm2 to determine the ADC map. Three DWI measurements with an orthogonal diffusion gradient were performed to calculate the isotropic ADC map. Image manipulation was carried out using MR vision software (MR Vision Co., Menro Park, CA, USA) on a Blade1000 workstation (Sun Microsystems, Milpitas, CA, USA). The CBF image during ischemia in each rat was used to confirm successful MCA occlusion, and animals without obvious CBF reduction were excluded from the experiments as previously described (Nito et al., 2011). Moreover, detailed MRI analyses were performed in the animals subjected to focal cerebral ischemia following 7-day pretreatment with EPA-E (100 mg/kg/day) or vehicle (n¼ 5, each). At the coronal level of bregma, the reduced CBF areas were determined as less than 60% of the average of the non-ischemic hemisphere according to a previous report (Meng et al., 2004), and the decreased ADC areas were determined as less than 80% of the average of the non-ischemic hemisphere based on previous investigations (Lo et al., 1997).

4.4. Neurological scores, blood examination and infarct volumes Neurological symptoms in each rat were assessed in a blind fashion, based on hemiparesis and abnormal posture using a neurological deficit score (Yonemori et al., 1998) at 24 or 72 h after reperfusion, as previously described (Nito et al., 2011). The right hindlimb of each rat was gently extended with round tipped forceps to evaluate hemiparesis, and flexor response was scored as 0 (normal), 1 (slight deficit), 2 (moderate deficit), and 3 (severe deficit). Rats were suspended by the tails to assess abnormal posture, and forelimb flexion and body twisting were scored as 0 (normal), 1 (slight twisting), 2 (marked twisting), or 3 (marked twisting and forelimb flexion). Then, animals were decapitated for infarct volume assessment. Blood samples were collected from animals subjected to focal cerebral ischemia following 7 days-pretreatment with EPA-E (100 mg/kg/day) or vehicle, at 24 h after reperfusion, to examine plasma EPA and AA levels (n¼ 5, each). In addition, these plasma levels were similarly determined just prior to decapitation in animals from experiments 2 and 3 (n¼5, each). Infarct volumes were determined using TTC-stained brain sections with 2 mm intervals at 24 or 72 h after reperfusion. The infarct areas in TTC-stained sections were traced using Image J software version 1.31J (NIH, Bethesda, MD, USA) on a Macintosh computer, and the obtained infarct areas from each animal were summed and multiplied by the interval thickness to yield the infarct volume in a blind fashion. Cortical and striatal infarct volumes were separately calculated.

4.5.

Immunohistochemistry

Another set of animals subjected to focal cerebral ischemia, with 7-day pretreatment of EPA-E (100 mg/kg/day) or vehicle, was deeply anesthetized and transcardially perfusion-fixed using 4% (w/v) paraformaldehyde 24 h after reperfusion (n¼ 5, each). The brains were carefully removed, stored in the same

brain research 1519 (2013) 95–104

fixative overnight at 4 1C, and 20 mm-thick coronal frozen sections were cut on a cryostat. The sections were incubated in 0.3% (v/v) H2O2 in methanol for 30 min, followed by 5% (v/v) normal bovine serum in Tris-buffered saline. They were incubated overnight at 4 1C with a mouse monoclonal anti8-OHdG antibody (1:50, Japan Institute for the Control of Aging, Shizuoka, Japan) to assess oxidative DNA damage, a mouse monoclonal anti- 4-HNE antibody (1:50, Japan Institute for the Control of Aging, Shizuoka, Japan) to detect lipid peroxidation, a rabbit polyclonal anti-p-adducin antibodies (1:100, Santa Cruz Biotechnology, Santa Cruz, CA, USA) to examine Rho-kinase activation, or a mouse monoclonal antivWF antibody (1:200, Dako, Glostrup, Denmark) as an endothelial marker. Sections were then incubated with secondary antibodies, followed by incubation with avidin-biotinperoxide complex (Vector Laboratories, Burlingame, CA, USA) for 30 min, and were visualized using diaminobenzidine. Each coronal section at the level of the caudoputamen was examined using a microscope. The numbers of positively stained cells in the cortical ischemic boundary zone and the striatum were counted in four random regions (1.33 mm2) in a blind fashion. Each procedure was followed by several rinses in 100 mM phosphate-buffered saline. Double immunohistochemistry was performed in several sections using Alexa 488 (Molecular Probes, Eugene, OR, USA) and Rhodamine Red-X (Molecular Probes, Eugene, OR, USA) following incubation with the primary antibodies against p-adducin and vWF at the same dilution, to evaluate co-localization of expression. Labeled fluorescent signals were detected using Eclipse E600W fluorescent microscopy (Nikon Corporation, Tokyo, Japan), and the obtained images were processed with Adobe Photoshop version 4.0 software (Adobe system, Mountain View, CA, USA) on a Macintosh computer.

4.6.

Statistical analyses

Statistical analysis was performed using StatView version 5.0 software (SAS Institute, Cary, NC, USA) on a Macintosh computer. An analysis of variance (ANOVA) with Scheffe's post-hoc test was used for comparisons in plasma fatty acid levels, reduced CBF areas, decreased ADC areas, infarct volumes and immunohistochemical cell counts, and data were expressed as the mean7SD. To compare neurological scores, a Mann–Whitney U test or a Kruskal–Wallis test was used, and data were expressed as the median and interquartile range. Statistical significance was set at po0.05.

Contributors MU and YK conceived of the experiments. MU, TI, CN and NK performed the experiments, and analyzed the data with YK. MU wrote the paper. All authors discussed the results and commented on the manuscript.

Acknowledgment EPA-E was a generous gift from Mochida Pharmaceutical Co., Ltd. (Tokyo, Japan).

103

r e f e r e nc e s

Ajami, M., Eghtesadi, S., Razaz, J.M., Kalantari, N., Habibey, R., Nilforoushzadeh, M.A., Zarrindast, M., Pazoki-Toroudi, H., 2011. Expression of Bcl-2 and Bax after hippocampal ischemia in DHA+EPA treated rats. Neurol. Sci. 32, 811–818. Bang, H.O., Dyerberg, J., Nielsen, A.B., 1971. Plasma lipid and lipoprotein pattern in Greenlandic West-coast Eskimos. Lancet 1 (7710), 1143–1145. Bas, O., Songur, A., Sahin, O., Mollaoglu, H., Ozen, O.A., Yaman, M., Eser, O., Fidan, H., Yagmurca, M., 2007. The protective effect of fish n-3 fatty acids on cerebral ischemia in rat hippocampus. Neurochem. Int. 50, 548–554. Cao, D.H., Xu, J.F., Xue, R.H., Zheng, W.F., Liu, Z.L., 2004. Protective effect of chronic ethyl docosahexaenoate administration on brain injury in ischemic gerbils. Pharmacol. Biochem. Behav. 79, 651–659. Chrissobolis, S., Sobey, C.G., 2006. Recent evidence for an involvement of rho-kinase in cerebral vascular disease. Stroke 37, 2174–2180. Dalle-Donne, I., Rossi, R., Colombo, R., Giustarini, D., Milzani, A., 2006. Biomarkers of oxidative damage in human disease. Clin. Chem. 52, 601–623 2006. Dyerberg, J., Bang, H.O., Stoffersen, E., Moncada, S., Vane, J.R., 1978. Eicosapentaenoic acid and prevention of thrombosis and atherosclerosis?. Lancet 2 (8081), 117–119. Ewing, J.R., Wei, L., Knight, R.A., Pawa, S., Nagaraja, T.N., Brusca, T., Divine, G.W., Fenstermacher, J.D., 2003. Direct comparison of local cerebral blood flow rates measured by MRI arterial spin-tagging and quantitative autoradiography in a rat model of experimental cerebral ischemia. J. Cereb. Blood Flow Metab. 23, 198–209. Feng, J., Ito, M., Kureishi, Y., Ichikawa, K., Amano, M., Isaka, N., Okawa, K., Iwamatsu, A., Kaibuchi, K., Hartshorne, D.J., Nakano, T., 1999. Rho-associated kinase of chicken gizzard smooth muscle. J. Biol. Chem. 274, 3744–3752. Floyd, R.A., Carney, J.M., 1992. Free radical damage to protein and DNA: mechanisms involved and relevant observations on brain undergoing oxidative stress. Ann. Neurol. 32 (Suppl), S22–S27. Halliwell, B., 1992. Reactive oxygen species and the central nervous system. J. Neurochem. 59, 1609–1623. He, K., Song, Y., Daviglus, M.L., Liu, K., Van Horn, L., Dyer, A.R., Goldbourt, U., Greenland, P., 2004. Fish consumption and incidence of stroke: a meta-analysis of cohort studies. Stroke 35, 1538–1542. Holub, B.J., 1988. Dietary fish oils containing eicosapentaenoic acid and the prevention of atherosclerosis and thrombosis. CMAJ 139, 377–381. Kahles, T., Luedike, P., Endres, M., Galla, H.J., Steinmetz, H., Busse, R., Neumann-Haefelin, T., Brandes, R.P., 2007. NADPH oxidase plays a central role in blood-brain barrier damage in experimental stroke. Stroke 38, 3000–3006. Katayama, Y., Katsumata, T., Muramatsu, H., Usuda, K., Obo, R., Terashi, A., 1997. Effect of long-term administration of ethyl eicosapentate (EPA-E) on local cerebral blood flow and glucose utilization in stroke-prone spontaneously hypertensive rats (SHRSP). Brain Res. 761, 300–305. Katsumata, T., Katayama, Y., Obo, R., Muramatsu, H., Ohtori, T., Terashi, A., 1999. Delayed administration of ethyl eicosapentate improves local cerebral blood flow and metabolism without affecting infarct volumes in the rat focal ischemic model. Eur. J. Pharmacol. 372, 167–174. Kimura, K., Fukata, Y., Matsuoka, Y., Bennett, V., Matsuura, Y., Okawa, K., Iwamatsu, A., Kaibuchi, K., 1998. Regulation of the association of adducin with actin filaments by Rho-associated kinase (Rho-kinase) and myosin phosphatase. J. Biol. Chem. 273, 5542–5548.

104

brain research 1519 (2013) 95–104

Kitagawa, K., Matsumoto, M., Oda, T., Niinobe, M., Hata, R., Handa, N., Fukunaga, R., Isaka, Y., Kimura, K., Maeda, H., Mikoshiba, K., Kamada, T., 1990. Free radical generation during brief period of cerebral ischemia may trigger delayed neuronal death. Neuroscience 35, 551–558. Lo, E.H., Pierce, A.R., Mandeville, J.B., Rosen, B.R., 1997. Neuroprotection with NBQX in rat focal cerebral ischemia. Effects on ADC probability distribution functions and diffusion–perfusion relationships. Stroke 28, 439–446. Lowenstein, C.J., Morrell, C.N., Yamakuchi, M., 2005. Regulation of Weibel-Palade body exocytosis. Trends Cardiovasc. Med. 15, 302–308. Meng, X., Fisher, M., Shen, Q., Sotak, C.H., Duong, T.Q., 2004. Characterizing the diffusion/perfusion mismatch in experimental focal cerebral ischemia. Ann. Neurol. 55, 207–212. Mizuguchi, K., Yano, T., Tanaka, Y., Ishibashi, M., Masada, A., Mizota, M., Fukutake, K., Saito, Y., 1993. Mechanism of the lipid-lowering effect of ethyl all-cis-5,8,11,14,17icosapentaenoate. Eur. J. Pharmacol. 231, 121–127. Nito, C., Ueda, M., Inaba, T., Katsura, K., Katayama, Y., 2011. FK506 ameliorates oxidative damage and protects rat brain following transient focal cerebral ischemia. Neurol. Res. 33, 881–889. Okabe, N., Nakamura, T., Toyoshima, T., Miyamoto, O., Lu, F., Itano, T., 2011. Eicosapentaenoic acid prevents memory impairment after ischemia by inhibiting inflammatory response and oxidative damage. J. Stroke Cerebrovasc. Dis. 20, 188–195. Okubo, S., Igarashi, H., Kanamatsu, T., Hasegawa, D., Orima, H., Katayama, Y., 2007. FK-506 extended the therapeutic time window for thrombolysis without increasing the risk of hemorrhagic transformation in an embolic rat stroke model. Brain Res. 1143, 221–227. Okuda, Y., Kawashima, K., Sawada, T., Tsurumaru, K., Asano, M., Suzuki, S., Soma, M., Nakajima, T., Yamashita, K., 1997. Eicosapentaenoic acid enhances nitric oxide production by cultured human endothelial cells. Biochem. Biophys. Res. Commun. 232, 487–491. Ozen, O.A., Cosar, M., Sahin, O., Fidan, H., Eser, O., Mollaoglu, H., Alkoc, O., Yaman, M., Songur, A., 2008. The protective effect of fish n-3 fatty acids on cerebral ischemia in rat prefrontal cortex. Neurol. Sci. 29, 147–152. Pan, H.C., Kao, T.K., Ou, Y.C., Yang, D.Y., Yen, Y.J., Wang, C.C., Chuang, Y.H., Liao, S.L., Raung, S.L., Wu, C.W., Chiang, A.N., Chen,

C.J., 2009. Protective effect of docosahexaenoic acid against brain injury in ischemic rats. J. Nutr. Biochem. 20, 715–725. Peters, O., Back, T., Lindauer, U., Busch, C., Megow, D., Dreier, J., Dirnagl, U., 1998. Increased formation of reactive oxygen species after permanent and reversible middle cerebral artery occlusion in the rat. J. Cereb. Blood Flow Metab. 18, 196–205. Philbrick, D.J., Mahadevappa, V.G., Ackman, R.G., Holub, B.J., 1987. Ingestion of fish oil or a derived n-3 fatty acid concentrate containing eicosapentaenoic acid (EPA) affects fatty acid compositions of individual phospholipids of rat brain, sciatic nerve and retina. J. Nutr. 117, 1663–1670. Shirao, S., Fujisawa, H., Kudo, A., Kurokawa, T., Yoneda, H., Kunitsugu, I., Ogasawara, K., Soma, M., Kobayashi, S., Ogawa, A., Suzuki, M., 2008. Inhibitory effects of eicosapentaenoic acid on chronic cerebral vasospasm after subarachnoid hemorrhage: possible involvement of a sphingosylphosphorylcholine-rhokinase pathway. Cerebrovasc. Dis. 26, 30–37. Takemoto, M., Sun, J., Hiroki, J., Shimokawa, H., Liao, J.K., 2002. Rho-kinase mediates hypoxia-induced downregulation of endothelial nitric oxide synthase. Circulation 106, 57–62. Tanaka, K., Ishikawa, Y., Yokoyama, M., Origasa, H., Matsuzaki, M., Saito, Y., Matsuzawa, Y., Sasaki, J., Oikawa, S., Hishida, H., Itakura, H., Kita, T., Kitabatake, A., Nakaya, N., Sakata, T., Shimada, K., Shirato, K., JELIS Investigators, Japan, 2008. Reduction in the recurrence of stroke by eicosapentaenoic acid for hypercholesterolemic patients: subanalysis of the JELIS trial. Stroke 39, 2052–2058. Williams, D.S., Detre, J.A., Leigh, J.S., Koretsky, A.P., 1992. Magnetic resonance imaging of perfusion using spin inversion of arterial water. Proc. Nat. Acad. Sci. USA 89, 212–216. Yagita, Y., Kitagawa, K., Sasaki, T., Terasaki, Y., Todo, K., OmuraMatsuoka, E., Kaibuchi, K., Hori, M., 2007. Rho-kinase activation in endothelial cells contributes to expansion of infarction after focal cerebral ischemia. J. Neurosci. Res. 85, 2460–2469. Yonemori, F., Yamaguchi, T., Yamada, H., Tamura, A., 1998. Evaluation of a motor deficit after chronic focal cerebral ischemia in rats. J. Cereb. Blood Flow Metab. 18, 1099–1106. Zhang, N., Komine-Kobayashi, M., Tanaka, R., Liu, M., Mizuno, Y., Urabe, T., 2005. Edaravone reduces early accumulation of oxidative products and sequential inflammatory responses after transient focal ischemia in mice brain. Stroke 36, 2220–2225.