Neurovascular and Neuronal Protection by E64d After Focal Cerebral Ischemia in Rats
Tamiji Tsubokawa,1 Mitsuo Yamaguchi-Okada,1 John W. Calvert,1
Ihsan Solaroglu,1 Norihito Shimamura,1 Kenichiro Yata,1 and John H. Zhang1,3*
1Department of Physiology and Pharmacology, Loma Linda University School of Medicine, Loma Linda, California
2Division of Neurosurgery, Loma Linda University School of Medicine, Loma Linda, California
3Department of Anesthesiology, Loma Linda University School of Medicine, Loma Linda, California

Calpains and cathepsins are two families of proteases that play an important role in ischemic cell death. In this study, we investigated the effect of E64d, a l-cal- pain and cathepsin B inhibitor, in the prevention of neu- ronal and endothelial apoptotic cell death after focal
cerebral ischemia in rats. Rats underwent 2 hr of tran- sient focal ischemia from middle cerebral artery occlu- sion (MCAO) and were sacrificed 24 hr later. E64d (5 mg/ kg intraperitoneally) was administered 30 min before MCAO. Assessment included neurological function, in- farction volume, brain water content, blood–brain barrier permeability, histology, and immunohistochemistry. The E64d-treated rats had significant brain protection against ischemic damage. We observed a reduction of infarction volume, brain edema, and improved neurological scores in E64d-treated rats compared with the nontreated con- trol. Furthermore, there was a remarkable reduction in both proteases and caspase-3 activation and apoptotic changes in both neurons and endothelial cells in E64d- treated rats. These results suggest that E64d protects the brain against ischemic/reperfusion injury by attenuating neuronal and endothelial apoptosis. VVC 2006 Wiley-Liss, Inc.

Key words: apoptosis; blood–brain barrier; calpain; cathepsin; ischemia-reperfusion

The pathogenesis of stroke is complex, and various pathophysiological pathways, including cysteine pro- teases, have been implicated in cell death after ischemic insult to the central nervous system (CNS; Bartus et al., 1994; Nitatori et al., 1995; Seyfried et al., 1997, 2001; Markgraf et al., 1998; Rami et al., 2000; Rami, 2003; Benchoua et al., 2004). Cathepsins belong to the cyste- ine protease family and are localized predominantly in the lysosomes. Cathepsins B and L are major lysosomal cysteine proteases in the CNS (Yamashima, 2000). Like cathepsins, calpains are also intracellular cysteine pro- teases, the two forms of which are known as l- and m- calpains (Yamashima, 2000). The release of cathepsins af- ter ischemic insult from the damaged lysosomes into the cell cytoplasm and the activation of calpains have been shown to trigger apoptosis in neurons (Seyfried et al., 1997, 2001; Stoka et al., 2001; Benchoua et al., 2004).
Recent studies showed that inhibition of calpains or cathepsins could prevent apoptotic cell death in neurons and improve neurological deficits after focal cerebral is- chemia (Bartus et al., 1994; Seyfried et al., 1997, 2001; Markgraf et al., 1998; Benchoua et al., 2004).
Cysteine protease inhibitors such as AK-295 or CA-074 were designed to provide inhibition of cathe- psins or calpains alone (Bartus et al., 1994; Benchoua et al., 2004). E64d targets both l-calpain and cathepsin B, which could provide an advantage in the prevention of neuronal apoptosis. However, apoptosis is also impli- cated in endothelial cell death after focal cerebral ische- mia, and this may lead to increased blood–brain barrier (BBB) permeability, inflammatory infiltration, brain edema, and infarction volume (del Zoppo and Mabuchi, 2003; Zhang et al., 2005).
To our knowledge, a potential role for l-calpain/
cathepsin B inhibitors in the prevention of endothelial ap- optosis, and thereby neurovascular protection, has not been studied before. Therefore, the present study was conducted to determine whether E64d pretreatment pro- tects against vascular endothelial cell and neuronal apopto- tic cell death in a rat model of transient focal ischemia.

MATERIALS AND METHODS
Experimental Animals
The experimental protocol was approved by the Local Animal Care and Use Committee of Loma Linda University. Sixty-nine adult male Sprague-Dawley rats weighing between
280 and 340 g were divided randomly into three groups: sham, DMSO, and E64d.

Transient Middle Cerebral Artery Occlusion
Rats were weighted prior to surgery. Anesthesia was induced with ketamine (80 mg/kg i.p.), followed by atropine

*Correspondence to: John H. Zhang, MD, PhD, Division of Neurosurgery, Loma Linda University Medical Center, 11234 Anderson Street, Room 2562B, Loma Linda, CA 92354. E-mail: [email protected]
Received 6 March 2006; Revised 18 April 2006; Accepted 7 May 2006
Published online 26 June 2006 in Wiley InterScience (www. interscience.wiley.com). DOI: 10.1002/jnr.20977

‘ 2006 Wiley-Liss, Inc.

E64d and Cerebral Ischemia 833

(0.1 mg/kg). A heating pad and a heating lamp were used to maintain the rectal temperature between 36.58C 6 0.58C. Rats were intubated and respiration was maintained with a small ani- mal respirator (Harvard Apparatus, MA). Rats were subjected to middle cerebral artery occlusion (MCAO) as described by Yin et al. (2003), with some modifications. Briefly, with the use of an operating microscope, the left common carotid artery, internal carotid artery (ICA), and external carotid artery (ECA) were surgically exposed. The ECA was then isolated and coa- gulated. A 4-0 silicon-coated nylon suture (Doccol Co., Albu- querque, NM) was inserted into the ICA through the ECA stump and gently advanced to occlude the MCA. The mean arterial blood pressure, heart rate, arterial blood gases, and blood glucose levels before, during, and after ischemia were analyzed. After 2 hr of MCA occlusion, the suture was removed to restore blood flow, the neck incision was closed, and the rats were allowed to recover. After the experiment, the animals were housed individually until sacrifice. All animals had free access to food and water.

Treatment Methods
The treatment group was injected with E64d [(l-3- trans-carboxyrane2), 5 mg/kg i.p.; Biomol Inc.] 30 min before focal ischemia. E64d was diluted with 1% dimethyl sulfoxide (DMSO) to a concentration of 15 mg/ml. In the DMSO group, the rats were treated with the same volume of DMSO that was delivered i.p. 30 min before focal ischemia.

Neurological Score
The neurological scores were evaluated 24 hr after MCAO by using the scoring system described by Garcia et al. (1995).

2,3,5-Triphenyltetrazolium Chloride Staining and Evaluation of Infarction Volume
¼
¼ ¼
Samples from sham (n 3), MCAO (n 6), and E64d (n 6) groups were used for evaluation of infarction volume. Twenty-four hours after MCAO, rats were deeply anesthe- tized with ketamine and perfused transcardially with ice-cold phosphate-buffered saline (PBS). Then, rats were decapitated, after which the brains were rapidly removed. All brains were carefully evaluated for macroscopic hemorrhagic changes before 2,3,5-triphenyltetrazolium chloride (TTC) staining. The tissue was sliced into 2-mm-thick coronal sections, and the slices were stained in TTC (Sigma, Germany) for 30 min at 378C in the dark (Yin et al., 2003). When the stain had developed, the tissue blocks were moved into 10% formalin
overnight. Seven coronal sections per animal were then pho- tographed. TTC stains both neuronal and glial cells with a deep red pigment. In areas where neuronal loss occurs, TTC does not stain and tissue remains white. Hence, the areas of unstained tissue (the infarcted areas) were demarcated and were analyzed in Image J (NIH) version 1.32.

Nissl Staining
¼
¼ ¼
Samples from sham (n 2), MCAO (n 4), and E64d (n 4) groups were used for histological assessment. Twenty- four hours after the MCAO, rats were anesthetized and trans- cardially perfused with ice-cold PBS followed by 10% parafor-
maldehyde as described previously (Yin et al., 2003). The brains were quickly removed and postfixed in 10% parafor- maldehyde and 30% sucrose for 3 days. The brains were cryo- protected and then rapidly frozen by 2-meyhtlbutane chilled in liquid nitrogen. Coronal tissue sections 10 lm thick were cut with the aid of a cryostat (Leica LM3050S).
For Nissl staining, slices were hydrated in 0.1% cresyl violet for 3 min. Then they were dehydrated in ethanol and cleaned with xylenes. The slides were next examined via light microscopy, and pictures were taken with a digital camera (Olympus BX51).

Brain Water Content
¼
Samples from three each groups (n 6 for each group) were used for evaluation of brain edema. Twenty-four hours after MCAO, brains were removed and immediately separated into left and right hemispheres. The samples were weighted after removal (wet weight) and again after drying in the oven at 1058C for 24 hr as described by Xi et al. (2002). The water content was calculated as [(wet weight – dry weight)/wet weight] 3 100%.

Evans Blue Dye Extravasation
¼
Samples from the three groups (n 5 for each group) were used for evaluation of experiments. Disruption of the BBB was analyzed 24 hr after the MCAO by using Evans blue (EB) dye as previously reported by Park et al. (2004), with some modifications. Briefly, EB dye (4%; 2.5 ml/kg) was injected over 2 min into the left femoral vein and allowed to circulate for 60 min. Rats were deeply anesthetized and transcardially perfused with PBS until a colorless perfusion fluid was obtained from the right atrium. The amount of extravasated EB in the brain was determined by spectroflurophotometry. Measurements were con- ducted at an excitation wavelength of 610 nm, an emission wavelength of 680 nm, and a bandwidth of 10 nm. The per- centage of EB was calculated as [(left hemisphere (ischemic side)/right hemisphere (contra lateral)] 3 100%.

Immunohistochemistry
Immunostaining was carried out with the avidin-biotin- peroxidase complex (ABC) method. Rats were sacrificed 24 hr after MCAO. The slides were prepared by same methods as for Nissl staining.
Brain immunolocalization of IgG was conducted accord- ing to the protocols described previously (Park et al., 2004). Briefly, sections were incubated with biotinylated anti-rat IgG antibody 1:100 (Santa Cruz Biotechnology, Santa Cruz, CA) and then treated with an ABC staining kit (Santa Cruz Bio- technology).
For immunohistochemistry, brain sections were incu- bated overnight at 48C with primary antibody. The following antibodies (Santa Cruz Biotechnology) were used: anticathep- sin B, anti-l-calpain, and anticleaved caspase-3 (Cell Signal- ing, Beverly, MA) at a concentration 1:100 each. After, sec- tions were treated with ABC staining kit (Santa Cruz Biotech- nology). TUNEL staining was performed according to methods (Roche Diagnostics, Indianapolis, IN) described pre- viously (Park et al., 2004).

Journal of Neuroscience Research DOI 10.1002/jnr

834 Tsubokawa et al.

Immunofluorescent labeling was performed as described previously (Yin et al., 2003). Briefly, brain sections were incu- bated with primary antibodies, namely, cathepsin B (Santa Cruz Biotechnology; 1:200), l-calpain (Santa Cruz Biotechnology; 1:200), cleaved caspase-3 (Cell Signaling; 1:200), vWF (BD Biosciences, San Jose, CA; 1:200), CD31 (Chemicon, Teme- cula, CA; 1:200), and NeuN (Chemicon; 1:200) at 48C for 24 hr. After rinsing with PBS, the sections were incubated for
1 hr in fluorescein isothiocyanate (FITC)-, Texas red-conju- gated secondary antibodies (Jackson Immunoresearch, West Grove, PA; 1:100). The sections were then visualized by using a fluorescent microscope (Olympus), and photographs were taken (MagnaFire SP 2.1B software). For negative controls, ei- ther the primary or the secondary antibodies were omitted, and the same staining procedures were followed.

Western Blot
¼
Animals were sacrificed at 24 hr after MCAO for West- ern blotting (n 4, for each group). Brains were stored at
–808C until analysis. Western blot analysis was performed as described previously (Calvert et al., 2003).
Whole-cell lysates were obtained by gently homogeniz- ing the brain sample with a homogenizer in 5 volume of buffer A [20 mM HEPES, 1.5 mM MgCl2, 10 mM KCl, 1 mM
EDTA, 1 mM EGTA, 250 mM sucrose, 0.1 mM phenylmeth- ylsulfonyl fluoride (PMSF), 1 mM dithiothreitol (DTT), and proteinase inhibitor cocktail tablets, pH 7.9]. Samples were fur- ther centrifuged at 14,000g at 48C for 15 min to separate the sample into supernatant A and pellet A. Supernatant A was used the cytosolic and mitochondrial fraction after resuspension in buffer A. Protein concentration were then determined by the DC protein assay.
Equal amounts of protein (100 lg) were loaded in each
lane of polyacrylamide-sodium dodecyl sulfate (SDS) gels. The gels were electrophoresed, followed by a transfer of the pro- tein to a nitrocellulose membrane. The membrane was then blocked with a blocking solution and probed with primary antibodies overnight at 48C.
The primary antibody (and concentration) used is rabbit
polyclonal IgG cleaved caspase-3 antibody (BD Biosciences; 1:1,000). Immunoblots were next processed with secondary antibody (Santa Cruz Biotechnology; 1:2,000) for 1 hr at room temperature. Immunoblots were then probed with an ECL Plus chemiluminescence reagent kit (Amersham Bio- sciences, Arlington Heights, IL) to visualize signal, followed by exposure to X-ray film. Films were scanned and optical density was determined with Image J (NIH). Actin (Santa Cruz Biotechnology; 1:2,000) was blotted on the same mem- brane as loading control.

Statistical Analaysis
Data are presented as means 6 SEM. Statistical differen- ces between the various groups were assessed by one-way anal- ysis of variance (ANOVA), followed by the post hoc test. Comparisons between the two groups were assessed by unpaired t-test. P < 0.05 was considered statistically significant.

Fig. 1. E64d decreases infarction volume and improves neurological outcome after transient focal brain ischemia in rat. A: Representative samples of TTC staining from the DMSO and E64d groups. Arrows show protected areas. Nissl staining of the cortex in sections of the brain from DMSO-treated (A: A,C) and E64d-treated (A: B,D) rats. Ipsilateral cortex (A: A,B) and contralateral areas (A: C,D). E64d treatment prevented cell death in ipsilateral cortex after transient focal brain ischemia. B: Quantification of infarction volume by TTC staining 24 hr after MCAO (n 6, for each). Treatment with E64d significantly decreased infarction volume compared with the DMSO group (*P < 0.05). C: Neurological score 24 hr after MCAO. The E64d group had significantly better neurological scores than the DMSO group (*P < 0.001 vs. sham and #P < 0.05 vs. DMSO).
¼
Scale bars ¼ 20 lm.

Journal of Neuroscience Research DOI 10.1002/jnr

E64d and Cerebral Ischemia 835

RESULTS
Physiological Data
No statistical differences were observed between the dimethyl sulfoxide (DMSO) and the E64d groups with regard to mean arterial blood pressure, heart rate, arterial blood gases, or glucose levels before, during, or after ischemia (data not shown). No statistical differences were observed between the DMSO and E64d groups for mean body weight at 24 hr after MCAO.
Effect of E64d on Improvement of Neurological Score and on Reduction of Infarction Volume
Representative samples of TTC-stained and Nissl- stained sections from the DMSO and E64d groups are shown in Figure 1A. Cortical infarct areas were significantly reduced by treatment with E64d compared with the DMSO group. Nissl staining showed that neuronal cells were normal in the contralateral hemisphere of both groups. Neuronal cells are substantially protected from transient ischemic injury in ipsilateral cortex by E64d pretreatment.
The mean infarct volume was also significantly decreased in the E64d group compared with the DMSO group (211.9 6 19.4 vs. 342.0 6 19.6 mm3, respec- tively; P < 0.05; Fig. 1B). E64d pretreatment signifi- cantly reduced infarct volume compared with the DMSO group.
The neurological scores revealed a significant dif- ference between groups at 24 hr after MCAO (P < 0.05). The neurological scores in the DMSO and E64d groups were 10.4 6 0.3 and 12.2 6 0.4, respectively (Fig. 1C). E64d treatment was found to increase the neurological scores significantly (P < 0.05).
BBB Disruption and Immunohistochemistry of the Intraparenchymal Vessels
IgG staining demonstrated BBB leakage in the MCA area, represented by IgG passing through the dis- rupted BBB and penetrating into the brain parenchyma. In the E64d group, IgG staining was observed to a lesser extent compared with the DMSO group (Fig. 2A). A sig- nificant increase in brain water content was revealed in rats at 24 hr after MCAO compared with sham group rats in the left hemisphere. The mean water content of the left hemisphere was 80.5% 6 0.6% in the E64d group and 83.8% 6 0.5% in the DMSO group (P < 0.05; Fig. 2B). No statistical difference among all groups in water content of right hemisphere was observed (Fig. 2B).
There was a significant increase in the permeability of the BBB in the DMSO group compared with the E64d group at 24 hr after MCAO (P < 0.05). The mean EB ratios of sham group, DMSO group, and E64d group were 1.04 6 0.01, 2.36 6 0.25, and 1.59 6 0.12, respectively (Fig. 2C). The EB ratio increased dra- matically in the DMSO group compared to the E64d group. However, E64d significantly reduced EB ratio following injury (P < 0.05).
An extensive activation of calpain and cathepsin B
was observed in the endothelium of intraparenchymal
vessels in samples from the DMSO group at 24 hr after MCAO as shown by triple-fluorescence staining (Fig. 2D). Moreover, increased cleaved caspase-3 expression was observed in the endothelium in samples from the DMSO group, indicating apoptotic endothelial cells at the same time point (Fig. 2E). However, E64d inhibited cleaved caspase-3 expression in endothelium of micro- vessels (Fig. 2E).

Immunohistochemistry of the MCA Branch and Hemorrhagic Transformation
In the DMSO group, the macroscopic hemorrhagic transformation ratio was 15.4% (4 of 26 rats) at 24 hr af- ter the MCAO. In contrast, the hemorrhagic transforma- tion ratio was lower in the E64d group, with a ratio of 3.8% (1 of 26 rats).
Immunohistochemical analysis of the MCA branch samples from the DMSO group revealed extensive acti- vation of l-calpain and cathepsin B in the endothelial layer of vascular structure at 24 hr after MCAO (Fig. 3). There were also extensive foci of cleaved caspase-3- and TUNEL-positive cells in the branch of MCA samples from the DMSO group (Fig. 3), suggesting that reperfu- sion injury produced vascular apoptotic cell death after focal cerebral ischemia. However, E64d prevented l-cal- pain and cathepsin B activation and apoptotic cell death in the vessel wall (Fig. 3).
Immunohistochemistry of the Ipsilateral Cortex
The comparison of immunohistological findings in samples from the DMSO and E64d groups is shown in Figure 4A. Immunohistochemistry of the ipsilateral cor- tex at 24 hr after MCAO showed the presence of exten- sive foci of l-calpain- and cathepsin B-staining cells in samples from the DMSO group (Fig. 4A). Moreover, l- calpain and cathepsin B activation colocalized in the neurons, as shown by triple-fluorescence staining in white color (Fig. 4B). However, E64d treatment pre- vented the activation of l-calpain and cathepsin B in the ipsilateral cortex and reduced the number of protease- expressing neurons (Fig. 4A,B).
TUNEL and cleaved caspase-3 staining showed exten- sive foci of positive cells in samples from the DMSO group (Fig. 4A). An increased number of cleaved caspase-3-stained neurons was observed in the DMSO group compared with the E64d group, showing apoptotic neuronal cell death by double-fluorescence staining in yellow color (Fig. 4A,C). Remarkable reduction of TUNEL and cleaved caspase-3 immunoreactivity was observed in ipsilateral cortex area in samples from animals treated with E64d (Fig. 4A).
Cleaved Caspase-3 Protein Expression
The DMSO rats showed a massive activation of caspase-3 in the ischemic cortex 24 hr after MCAO. With E64d treatment, caspase-3 activation was drastically reduced at 24 hr (Fig. 4D). The ratio of cleaved caspase- 3/actin was significantly increased in the DMSO group compared with the E64d group (*P < 0.05; Fig. 4D).

Journal of Neuroscience Research DOI 10.1002/jnr

836 Tsubokawa et al.

Fig. 2. E64d prevented BBB disruption after transient focal brain is- chemia in rats. A: IgG staining in sections of the rat brain for sham, DMSO, and E64d groups, respectively. Negative staining in the sham group, with extensive, strong staining seen in DMSO group and con- fined staining observed in E64d group. Arrows show IgG passing through penetrating into the brain parenchyma. B: Quantification of brain water content in the ipsilateral (left columns) and contralateral (right columns) brain hemisphere 24 hr after MCAO. Compared with that in sham-operated animals, brain water content was markedly increased in the DMSO group. E64d significantly decreased ipsilateral hemisphere water content compared with DMSO (*P < 0.05 vs. sham
and #P < 0.05 vs. DMSO). There was no statistical difference among
¼
the groups for contralateral hemisphere water content (n 6, for each group). C: Quantification of EB content of ipsilateral brain hemisphere
24 hr after MCAO. The data are expressed as ratio of ipsilateral/con- tralateral hemisphere EB content. The EB ratio significantly increased
¼
24 hr after the MCAO (*P < 0.05 vs. sham). E64d prevented an increase in EB ratio (#P < 0.05 vs. DMSO; n 5, for each group). D: Triple-immunofluorescence staining was performed for CD31 (green), l-calpain (red), and cathepsin B (blue) at 24 hr after the MCAO. Arrows indicate extensive protease activation colocalized in
the endothelium in samples from DMSO group (merge, white). E: Double-immunofluorescence staining was performed for vWF (red) and cleaved caspase-3 (green) at 24 hr after MCAO. Arrows indicate microvascular endothelial apoptosis in samples from DMSO group (merge, yellow). E64d treatment prevented caspase-3 activation in the endothelium of microvessels. Scale bars ¼ 20 lm.

Journal of Neuroscience Research DOI 10.1002/jnr

E64d and Cerebral Ischemia 837

Fig. 3. Immunostaining for l-calpain, cathepsin B, and cleaved cas- pase-3 and TUNEL staining in the wall of MCA branch of the ipsi- lateral hemisphere at 24 hr after MCAO in the DMSO group (A–H) and E64d group (I–P). Activation of l-calpain and cathepsin B is seen 24 hr after MCAO in the endothelial layer of vessel wall (E,F; arrows). No activation of l-calpain and cathepsin B is seen in the E64d group in the endothelium at 24 hr after MCAO (M,N). TUNEL staining and immunostaining of cleaved caspase-3 in the en-
dothelium (arrows) at 24 hr after MCAO in the DMSO group (G,H) and the E64d group (O,P). Twenty-four hours after the MCAO, increased cleaved caspase-3 immunostaining and TUNEL staining are seen in the endothelial layer of vessel in the DMSO group (G,H). E64d treatment prevented apoptotic endothelial cell death in the ischemic cortex at 24 hr after reperfusion (O,P). Scale bars 20 lm in D (applies to A–D); 20 lm in L (applies to I–L); 10 lm in H (applies to E–H); 10 lm in P (applies to M–P).

¼
DISCUSSION
Although the precise physiological functions of cys- teine proteases have not yet been clearly identified, there is evidence that under pathological conditions they are involved in both apoptotic and necrotic cell death in the CNS. Hence, they are potential targets for therapeutic intervention after ischemic insult (Bartus et al., 1994; Nitatori et al., 1995; Seyfried et al., 1997, 2001; Rami et al., 2000; Yamashima, 2000; Rami, 2003; Benchoua
et al., 2004).
Although cathepsins have been associated mainly with necrotic cell death (Yamashima, 2000), recent evi-
dence has shown their critical involvement in apoptosis, through alterations of mitochondria homeostasis and activation of proapoptotic members of the Bcl-2 family (Guicciardi et al., 2000; Stoka et al., 2001; Boya et al., 2003). Calpain has also been shown to induce apoptosis by activating caspase-12 and by inactivating Bcl-xL (Nakagawa and Yuan, 2000). Rami (2003) reported that calpain translocates into the nucleus under ischemic con- ditions, resulting in apoptotic cell death by activating p53. However, the release of cathepsin B from lyso- somes and its subsequent activation induce cytochrome c release from mitochondria and the activation of caspases,

Journal of Neuroscience Research DOI 10.1002/jnr

838 Tsubokawa et al.

Figure 4.

Journal of Neuroscience Research DOI 10.1002/jnr

E64d and Cerebral Ischemia 839

thereby leading to apoptosis (Guicciardi et al., 2000; Boya et al., 2003). In the present study, ischemic injury to the brain resulted in neuronal activation of both l- calpain and cathepsin B in the cortex and hence increased numbers of apoptotic neurons. Consistently with previous reports, the pharmacological inhibition of calpains and cathepsins reduced the caspase-3 activation and neuronal apoptosis, which may account for the sig- nificantly decreased infarct volume (Rami et al., 2000; Benchoua et al., 2004). As with other protease inhibi- tors, E64d not only inhibited proteases but also inhibited caspase activation. Because cathepsins and calpains appear to have an important role in cell death, we speculate that inhibition of both proteases with a single pharmacologi- cal agent could provide neuroprotection.
However, in addition to the neuronal apoptosis, ap- optotic cell death in neurovascular structures has also been implicated in the pathophysiology of ischemic stroke. It is well established that disrupted endothelium leads to a breakdown of the BBB, hemorrhage, and edema, which increases cerebral infarction volume after ischemic insult (del Zoppo and Mabuchi, 2003; Zhang et al., 2005). Hence, neurovascular protection is an important target in preventing or reducing neuronal cell death. In the present study, there was a significant increase in EB extravasation and brain edema, both of which indicate increased per- meability of BBB after MCAO. It is well known that en- dothelium in intraparenchymal vessels plays an important role in preservation of the BBB integrity. Recently, Li and Pober (2005) found that cathepsin-mediated death contributes to human endothelial injury. Results from the present study clearly showed that both l-calpain and ca- thepsin B activation occurs in the endothelium of intra-

3
Fig. 4. A: Immunohistochemistry of the ipsilateral cortex at 24 hr af- ter MCAO showed the presence of extensive foci of l-calpain, ca- thepsin B, TUNEL, and cleaved caspase-3 staining cells in samples from DMSO group (A: A–D). No activation of l-calpain and ca- thepsin B is seen in the ischemic cortex of the ipsilateral hemisphere in the E64d group at 24 hr after MCAO (A: E,F). E64d reduced the apoptotic neuronal death in the ipsilateral cortex at 24 hr after MCAO (A: G,H). B: Assessment of the association between l-cal- pain and cathepsin B shows colocalization of activated l-calpain and cathepsin B in the ischemic peripheral area at 24 hr after MCAO. Triple-immunofluorescence staining was performed for NeuN (green), l-calpain (red), and cathepsin B (blue). Arrows indicate extensive protease activation colocalized in the neurons. Merge imag- ing is shown in white (arrows). E64d reduced the number of protease expressing neurons. C: Double immunofluorescence staining was performed for NeuN (red) and cleaved caspase-3 (green). At 24 hr, an increased cleaved caspase-3 immunostaining is seen in neurons in the DMSO group. Neurons expressing cleaved caspase-3 appeared yellow (arrows). D: Representative Western blot analysis showing caspase-3 and cleaved caspase-3, as well as b-actin as a loading con- trol, expression in brain tissue at 24 hr after MCAO. Quantification of Western blot analysis showed an increased cleaved caspase-3/b- actin ratio in the DMSO group compared with the E64d group at 24 hr after MCAO (*P < 0.05 vs. DMSO; n ¼ 4, for each group).
Scale bar ¼ 20 lm.
parenchymal vessels, which was accompanied by an endo- thelial cell apoptosis. Inhibition of protease activation by E64d exerted neurovascular protection and significantly reduced BBB permeability. These data suggest that both proteases could play important roles in BBB disruption af- ter MCAO.
However, we also showed endothelial apoptotic cells in the branch of MCA in the ipsilateral cortex 24 hr after MCAO. More importantly, we also demon- strated that l-calpain and cathepsin B activation were localized in the branch of the MCA, suggesting that l- calpain and cathepsin B could be involved in MCA injury. There was an increased hemorrhagic transforma- tion ratio in the DMSO group compared with the E64d group, supporting the hypothesis that activation of both proteases is involved in vascular injury.
E64d, a cell-membrane-permeable drug, is highly
selective for cysteine proteases and four times more potent than the other E64 forms (van Acker et al., 2002; Carragher, 2006). However, the pharmacological prop- erties of E64d, including its penetration of the BBB, have not yet been fully investigated. E64d is a cell-mem- brane-permeable drug and the endothelium is located at the interface between the blood and the vessel wall, so we hypothesized that E64d could penetrate the endothe- lial cell membrane and could inhibit proteases, resulting in decreased BBB leakage, brain edema, and hemor- rhagic transformation. However, the inhibition of neural apoptosis may be related to the preservation of the BBB integrity or the direct effect of E64d on neurons. To our knowledge, this is the first study that shows neuro- vascular protection by E64d pretreatment after transient focal ischemia in the rat.
In conclusion, E64d reduced cerebral infarction
volume and prevented neurovascular injury after stroke. However, further work is required to clarify the rela- tionship among l-calpain, cathepsin B, and caspase-3 activation in neuronal and endothelial cells following cerebral ischemia. The present observations suggest that combination therapy with both l-calpain and cathepsin B inhibitor might be an effective therapeutic regimen for in vivo models of neurodegenerative disease, espe- cially acute cerebral ischemia.

ACKNOWLEDGMENTS
This study was partially supported by NIH grants NS45694, HD43120, and NS43338 to J.H.Z.

REFERENCES
Bartus RT, Hayward NJ, Elliott PJ, Sawyer SD, Baker KL, Dean RL, Akiyama A, Straub JA, Harbeson SL, Li Z. 1994. Calpain inhibitor AK295 protects neurons from focal brain ischemia. Effects of postocclu- sion intra-arterial administration. Stroke 25:2265–2270.
Benchoua A, Braudeau J, Reis A, Couriaud C, Onteniente B. 2004. Activation of proinflammatory caspases by cathepsin B in focal cerebral ischemia. J Cereb Blood Flow Metab 24:1272–1279.
Boya P, Andreau K, Poncet D, Zamzami N, Perfettini JL, Metivier D, Ojcius DM, Jaattela M, Kroemer G. 2003. Lysosomal membrane per-

Journal of Neuroscience Research DOI 10.1002/jnr

840 Tsubokawa et al.

meabilization induces cell death in a mitochondrion-dependent fashion. J Exp Med 197:1323–1334.
Calvert JW, Zhou C, Nanda A, Zhang JH. 2003. Effect of hyperbaric oxygen on apoptosis in neonatal hypoxia-ischemia rat model. J Appl Physiol 95:2072–2080.
Carragher NO. 2006. Calpain inhibition: a therapeutic strategy targeting multiple disease states. Curr Pharm Des 12:615–638.
del Zoppo GJ, Mabuchi T. 2003. Cerebral microvessel responses to focal ischemia. J Cereb Blood Flow Metab 23:879–894.
Garcia JH, Wagner S, Liu KF, Hu XJ. 1995. Neurological deficit and extent of neuronal necrosis attributable to middle cerebral artery occlu- sion in rats. Statistical validation. Stroke 26:627–634.
Guicciardi ME, Deussing J, Miyoshi H, Bronk SF, Svingen PA, Peters C, Kaufmann SH, Gores GJ. 2000. Cathepsin B contributes to TNF- alpha-mediated hepatocyte apoptosis by promoting mitochondrial release of cytochrome c. J Clin Invest 106:1127–1137.
Li JH, Pober JS. 2005. The cathepsin B death pathway contributes to TNF plus IFN-gamma-mediated human endothelial injury. J Immunol 175:1858–1866.
Markgraf CG, Velayo NL, Johnson MP, McCarty DR, Medhi S, Koehl JR, Chmielewski PA, Linnik MD. 1998. Six-hour window of opportu- nity for calpain inhibition in focal cerebral ischemia in rats. Stroke 29: 152–158.
Nakagawa T, Yuan J. 2000. Cross-talk between two cysteine protease families. Activation of caspase-12 by calpain in apoptosis. J Cell Biol 150:887–894.
Nitatori T, Sato N, Waguri S, Karasawa Y, Araki H, Shibanai K, Komi- nami E, Uchiyama Y. 1995. Delayed neuronal death in the CA1 py- ramidal cell layer of the gerbil hippocampus following transient ische- mia is apoptosis. J Neurosci 15:1001–1011.
Park S, Yamaguchi M, Zhou C, Calvert JW, Tang J, Zhang JH. 2004. Neurovascular protection reduces early brain injury after subarachnoid hemorrhage. Stroke 35:2412–2417.
Rami A. 2003. Ischemic neuronal death in the rat hippocampus: the cal- pain-calpastatin-caspase hypothesis. Neurobiol Dis 13:75–88.
Rami A, Agarwal R, Botez G, Winckler J. 2000. Mu-calpain activation, DNA fragmentation, and synergistic effects of caspase and calpain inhib- itors in protecting hippocampal neurons from ischemic damage. Brain Res 866:299–312.
Seyfried D, Han Y, Zheng Z, Day N, Moin K, Rempel S, Sloane B, Chopp M. 1997. Cathepsin B and middle cerebral artery occlusion in the rat. J Neurosurg 87:716–723.
Seyfried DM, Veyna R, Han Y, Li K, Tang N, Betts RL, Weinsheimer S, Chopp M, Anagli J. 2001. A selective cysteine protease inhibitor is non-toxic and cerebroprotective in rats undergoing transient middle cerebral artery ischemia. Brain Res 901:94–101.
Stoka V, Turk B, Schendel SL, Kim TH, Cirman T, Snipas SJ, Ellerby LM, Bredesen D, Freeze H, Abrahamson M, Bromme D, Krajewski S, Reed JC, Yin XM, Turk V, Salvesen GS. 2001. Lysosomal protease pathways to apoptosis. Cleavage of bid, not pro-caspases, is the most likely route. J Biol Chem 276:3149–3157.
van Acker GJ, Saluja AK, Bhagat L, Singh VP, Song AM, Steer ML. 2002. Cathepsin B inhibition prevents trypsinogen activation and reduces pan- creatitis severity. Am J Physiol Gastrointest Liver Physiol 283:G794–G800.
Xi G, Hua Y, Keep RF, Younger JG, Hoff JT. 2002. Brain edema after intracerebral hemorrhage: the effects of systemic complement depletion. Acta Neurochir Suppl 81:253–256.
Yamashima T. 2000. Implication of cysteine proteases calpain, cathepsin and caspase in ischemic neuronal death of primates. Prog Neurobiol 62: 273–295.
Yin D, Zhou C, Kusaka I, Calvert JW, Parent AD, Nanda A, Zhang JH. 2003. Inhibition of apoptosis by hyperbaric oxygen in a rat focal cere- bral ischemic model. J Cereb Blood Flow Metab 23:855–864.
Zhang Y, Zhang X, Park TS, Gidday JM. 2005. Cerebral endothelial cell apoptosis after ischemia-reperfusion: role of PARP activation and AIF translocation. J Cereb Blood Flow Metab 25:868–877.

Journal of Neuroscience Research DOI 10.1002/jnr