- Email: [email protected]
Early appearance of activated matrix metalloproteinase-9 and blood–brain barrier disruption in mice after focal cerebral ischemia and reperfusion
Brain Research 842 Ž1999. 92–100 www.elsevier.comrlocaterbres
Research report
Early appearance of activated matrix metalloproteinase-9 and blood–brain barrier disruption in mice after focal cerebral ischemia and reperfusion Miki Fujimura, Yvan Gasche, Yuiko Morita-Fujimura, Justin Massengale, Makoto Kawase, Pak H. Chan ) Department of Neurosurgery, Department of Neurology and Neurological Sciences, and Program in Neurosciences, Stanford UniÕersity School of Medicine, Palo Alto, CA, USA Accepted 7 July 1999
Abstract Blood–brain barrier ŽBBB. disruption is thought to play a critical role in the pathophysiology of ischemiarreperfusion. Matrix metalloproteinases ŽMMPs. are a family of proteolytic enzymes that can degrade all the components of the extracellular matrix when they are activated. Gelatinase A ŽMMP-2. and gelatinase B ŽMMP-9. are able to digest the endothelial basal lamina, which plays a major role in maintaining BBB impermeability. The present study examined the expression and activation of gelatinases before and after transient focal cerebral ischemia ŽFCI. in mice. Adult male CD1 mice were subjected to 60 min FCI and reperfusion. Zymography was performed from 1 to 23 h after reperfusion using the protein extraction method with detergent extraction and affinity-support purification. MMP-9 expression was also examined by both immunohistochemistry and Western blot analysis, and tissue inhibitors to metalloproteinase-1 was measured by reverse zymography. The BBB opening was evaluated by the Evans blue extravasation method. The 88-kDa activated MMP-9 was absent from the control specimens, while it appeared 3 h after transient ischemia by zymography. At this time point, the BBB permeability alteration was detected in the ischemic brain. Both pro-MMP-9 Ž96 kDa. and pro-MMP-2 Ž72 kDa. were seen in the control specimens, and were markedly increased after FCI. A significant induction of MMP-9 was confirmed by both immunohistochemistry and Western blot analysis. The early appearance of activated MMP-9, associated with evidence of BBB permeability alteration, suggests that activation of MMP-9 contributes to the early formation of vasogenic edema after transient FCI. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Matrix metalloproteinase; Cerebral ischemia; Reperfusion injury; Blood–brain barrier; Brain edema
1. Introduction Matrix metalloproteinases ŽMMPs. are a family of zinc-binding proteolytic enzymes which are capable of degrading components of the extracellular matrix in a variety of physiological and pathophysiological conditions. Among MMPs, gelatinase A ŽMMP-2. and gelatinase B ŽMMP-9. are able to digest the endothelial basal lamina, which plays a major role in maintaining blood–brain barrier ŽBBB. impermeability, by regulating tight junctions leading to the opening of the BBB. In the pathological condition of ischemiarreperfusion, digestion of the endothelial basal lamina is reported to occur as early as 2 h after ischemia w14x, which may result in BBB permeability ) Corresponding author. Neurosurgical Laboratories, Stanford University, 701B Welch Road, a148, Palo Alto, CA 94304, USA. Fax: q1650-498-4550; E-mail: [email protected]
a few hours after ischemia w4,18x. Recently, MMP-9 andror MMP-2 have been implicated in focal cerebral ischemia ŽFCI. w9,23,24,27x, as well as other central nervous system disorders such as multiple sclerosis w25x, intracerebral hemorrhage w26x and cerebral aneurysm w5,15x. A previous report by Rosenberg and his colleagues on FCI showed the marked increase of pro-MMP-9 after both permanent and transient ischemia, suggesting the possible involvement of MMP-9 in FCI w24,27x. However, it has still not been established whether the 88-kDa activated MMP-9, which is considered to be the fully active form to digest basal lamina w32x, could appear after transient FCI. In the present study, we sought to clarify this point by measuring gelatinase activity in the brain before and after transient FCI by using the protein extraction method with detergent extraction and affinity-support purification, which allowed us to recover gelatinase activity over a 4- to 10-fold range using zymography w36x. The results showed
0006-8993r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 6 - 8 9 9 3 Ž 9 9 . 0 1 8 4 3 - 0
M. Fujimura et al.r Brain Research 842 (1999) 92–100
93
that 88-kDa activated MMP-9 is absent from the control specimens, while it appeared 3 h after 60 min of transient ischemia, as shown by zymography and type IV collagenase activity assay. At this time point, BBB permeability alteration was detected in the ischemic brain by the Evans blue extravasation method. Both pro-MMP-9 Ž96 kDa. and pro-MMP-2 Ž72 kDa. were detected in the control specimens and increased after transient FCI. A significant induction of MMP-9 was confirmed by both immunohistochemistry and Western blot analysis after transient FCI. These results, which are consistent with a previous report showing the prevention of BBB impermeability with the MMP-9 inhibitor BB-1101 3 h after reperfusion w27x, suggest that the activation of MMP-9 during reperfusion may contribute to BBB disruption and the formation of vasogenic edema after transient FCI.
mM CaCl 2 , 0.05% BRIJ-35, 0.02% NaN3 , 1% Triton X-100.. Aliquots Ž10 ml. of the homogenates were saved for total protein measurements ŽBCA Kit, Pierce, Rockford, IL.. The homogenates were centrifuged at 12,000 = g for 5 min. The supernatants were recovered and incubated for 60 min with gelatin-sepharose 4B ŽPharmacia Biotech, Uppsala, Sweden., with constant shaking. After incubation, the samples were centrifuged at 500 = g for 2 min. The pellets were washed with a buffer containing 50 mM Tris–HCl, pH 7.6, 150 mM NaCl, 5 mM CaCl 2 , 0.05% BRIJ-35, 0.02% NaN3 . After a second centrifugation, the pellets were resuspended in elution buffer Ž50 mM Tris– HCl, pH 7.6, 150 mM NaCl, 5 mM CaCl 2 , 0.05% BRIJ-35, 0.02% NaN3 , dimethylsulfoxide 10%. for 30 min. Then the samples were subjected to zymographic measurements of gelatinase activity, to type IV collagenolytic activity assay, and to Western blot analysis.
2. Materials and methods
2.3. Zymographic measurement of gelatinase actiÕity
2.1. Focal cerebral ischemia and histological assessment Adult male CD-1 mice Ž35–40 g, n s 33. were subjected to transient FCI by intraluminal middle cerebral artery ŽMCA. blockade with a nylon suture as described w33x. The mice were anesthetized with 2.0% isoflurane in 30% oxygen and 70% nitrous oxide using a face mask. The rectal temperature was controlled at 378C with a homeothermic blanket. Cannulation of a femoral artery allowed the monitoring of blood pressure and arterial blood gases, samples for analysis being taken immediately after cannulation, 10 min after occlusion and 10 min after reperfusion. After the midline skin incision, the left external carotid artery was exposed and its branches were electrocoagulated. A 11.0-mm 5-0 surgical monofilament nylon suture, blunted at the end, was introduced into the left internal carotid artery through the external carotid artery stump. After 60 min of MCA occlusion, blood flow was restored by the withdrawal of the nylon suture. The experimental animals were killed 1, 3 and 23 h after 60 min of MCA occlusion. The brains were removed, rapidly frozen in y208C 2-methylbutane, and stored at y808C. They were sectioned with a cryostat into a thickness of 20 mm from the anterior side to the posterior side and stained with cresyl violet. 2.2. Preparation of tissue extracts To analyze the gelatinase activity and MMP protein expression patterns in control and ischemic brain tissue, protein extraction of the tissue was performed as previously described w36x. Approximately 100 mg of both ischemic brain and homologous tissue from the contralateral side were extracted and rapidly frozen in powdered dry ice after 1, 3 and 23 h of reperfusion. The brain tissue was homogenized with 2 ml Teflon glass homogenizers in a lysis buffer Ž50 mM Tris–HCl, pH 7.6, 150 mM NaCl, 5
Zymographies were performed according to a previously reported method with minor modifications w17x. Ten microliters of a non-reducing sample buffer Ž0.4 M Tris, pH 6.8, 5% sodium dodecyl sulfate ŽSDS., 20% glycerol, 0.05% bromophenol blue. were added to the sample volume which was corrected for protein concentration. Samples were then loaded onto 8% SDS-polyacrylamide gel electrophoresis ŽPAGE. gels in which 0.15% porcine skin gelatin ŽSigma, St. Louis, MO. was co-polymerized. The gels were run at 40 mA for 3 h. After migration, the gels were incubated with 2.5% Triton-X 100 twice for 1 h at room temperature, washed for 10 min in a buffer containing 50 mM Tris, pH 7.5, 200 mM NaCl, 5 mM CaCl 2 ŽTNCA. and further incubated for 16 h in TNCA, in a water bath at 378C. The gels were then stained for 90 min in coomassie blue Ž1% coomassie brilliant blue, 30% methanol, 10% acetic acid. and destained in 30% methanolr10% acetic acid, four times for 5, 15, 30 and 60 min, respectively. White bands on a blue background indicated zones of digestion corresponding to the presence of different MMPs. The 96-kDa band corresponded to proMMP-9, the 88-kDa band to MMP-9, and the 72-kDa band corresponded to proMMP-2. Twenty microliters of MMP-9 and MMP-2 standards ŽOncogene Research, Cambridge, MA. were loaded into each gel to allow comparison between the different gels. The bands were scanned using a densitometer ŽGS-700, Bio-Rad Laboratories, Richmond, CA. and quantification was performed using the Multi-Analyst 1.0 software ŽBio-Rad.. 2.4. Type IV collagenase actiÕity assay Type IV collagenase activity assay was performed using a commercially available kit ŽChemicon International, Temecula, CA.. According to the manufacturer protocol, samples were incubated with TNCA and FITC labeled type IV collagen solution, first for 20 min at room temperature and then for 2 h at 428C in a water bath. Samples
94
M. Fujimura et al.r Brain Research 842 (1999) 92–100
were then placed on ice for 5 min. Three hundred microliters of cold enzyme stop reagentrextraction solution were added to all samples which were further stored for 35 min on ice. Residual undigested collagen was precipitated by centrifugation Ž10 min at 12,200 = g .. A standard curve was made using MMP-9 standard solution provided by the manufacturer. Proteolytic activity of ischemic samples was measured in the presence or absence of ethylenediaminetetraacetic acid ŽEDTA, 0.5 mM., which is known to inhibit MMPs w21x. Finally, fluorescence intensity of sample supernatants was measured with a fluorimeter ŽMolecular Devices, Sunnyvale, CA. at 520 nmr495 nm. 2.5. Immunohistochemistry of MMP-9 To assess the anatomical distribution of MMP-9 after ischemiarreperfusion, immunohistochemistry of MMP-9 was performed. Anesthetized animals were perfused with 10 Urml heparin and subsequently with 4% formaldehyde in 0.1 M phosphate-buffered saline ŽPBS. ŽpH 7.4. 1, 3 and 23 h after reperfusion. The brains were removed, postfixed for 12 h in 4% formaldehyde, sectioned at 50 mm on a vibratome, and processed for immunohistochemistry. The sections were incubated with a blocking solution as described w11x and reacted with anti-MMP-9 polyclonal antibody ŽAnawa, Wangen, Switzerland. at a dilution of 1:50. Immunohistochemistry was performed using the avidin–biotin technique as described w11x, and then the nuclei were counterstained with methyl green solution for 10 min. As a negative control, sections were incubated without primary antibodies. For histological assessment, alternate slices from each brain section were stained with cresyl violet. 2.6. Western blot analysis To further investigate the protein expression of MMP-9 in the samples from the control and ischemic brains, Western blot analysis was performed as previously described w11x. Equal amounts of the protein were loaded per lane. The primary antibodies were used at a 1:500 dilution of polyclonal antibody against MMP-9 ŽAnawa.. Western blots were performed with horseradish peroxidase-conjugated anti-rabbit immunoglobulin G using the Boehringer Mannheim ŽIndianapolis, IN. chemiluminescent system, and bands were visualized by enhanced chemiluminescence reagent ŽECL plus Western blot analysis system, Amersham Life Science, Arlington Heights, IL.. Recombinant human proMMP-9 and MMP-9 were used as controls.
20% glycerol, 0.05% bromophenol blue. were added to the samples Žvolume corrected for protein concentration.. The samples were then loaded onto 15% SDS-PAGE gels in which 2.2 mgrml porcine skin gelatin ŽSigma. and 150 ngrml of proMMP-9 were co-polymerized. The gels were run at 40 mA for 3 h. After migration, the gels were incubated with 2.5% Triton-X 100 twice for 1 h at room temperature, washed for 10 min in TNCA, and further incubated for 16 h in TNCA in a water bath at 378C. The gels were stained for 30 min in coomassie blue Ž1% coomassie brilliant blue, 30% methanol, 10% acetic acid. and destained in 30% methanolr10% acetic acid, four times for 5, 15, 30 and 60 min, respectively. Tissue inhibitor of metalloproteinase-1 ŽTIMP-1. was identified as a 28-kDa blue band; 20 ml of TIMP-1 ŽOncogene Research. were loaded as a standard. The bands were scanned using a densitometer ŽGS-700, Bio-Rad. and quantification was performed using the Multi-Analyst 1.0 software ŽBio-Rad.. 2.8. EÕans blue extraÕasation In situ detection of Evans blue leakage was based on the previously described method w18x with modification. One hour before the transcardiac perfusion, 0.1 ml of 4% Evans blue ŽSigma. in 0.9% saline was intravenously injected. The animals were perfused with 10 Urml heparin Ž200 ml. and subsequently with 4% formaldehyde in 0.1 M PBS ŽpH 7.4. 1 and 3 h after reperfusion. The brains were sectioned with a vibratome, mounted on slides, and counterstained with Hoechst 33258 ŽMolecular Probes, Eugene, OR.. BBB disruption was assessed fluoromicroscopically at excitations 355 nm and emission ) 415 nm for Evans blue detection. To analyze the fluorescence signal of Evans blue, we scanned photomicrographs Ž=630. with a GS-700 imaging densitometer ŽBio-Rad. and then measured the red signal intensity in 15 individual regions from three animals in each group using Muti-Analyst software ŽBio-Rad.. 2.9. Statistical analysis Data are presented as the mean " S.D. Statistical comparisons were made using unpaired ANOVA with STAT VIEW software, version 4.5 ŽAbacus Concepts, Berkeley, CA.. P - 0.05 was considered to be statistically significant.
3. Results 3.1. Physiological data and cerebral infarction
2.7. ReÕerse zymography Reverse zymographies were performed as previously described w20x with modifications. Ten microliters of a non-reducing sample buffer Ž0.4 M Tris, pH 6.8, 5% SDS,
Physiological parameters showed no significant differences in mean arterial blood pressure ŽMABP. and arterial blood gas analysis between each time point. The preischemic physiological variables were as follows: MABP,
M. Fujimura et al.r Brain Research 842 (1999) 92–100
95
Fig. 1. Zymographic measurement of gelatinase activities before and after transient ischemia. Pro-MMP-9 Ž96 kDa. and pro-MMP-2 were present in the non-ischemic brain and increased after ischemiarreperfusion. Activated MMP-9 Ž88 kDa. was absent in the control specimens, although it appeared at 3 h and further increased 23 h after reperfusion. Activated MMP-2 was absent through all the time points. n s non-ischemic control; c s contralateral brain.
71.0 " 3.8 mmHg; PaO 2 , 146.4 " 9.9 mmHg; PaCO 2 , 28.9 " 7.8 mmHg; pH, 7.35 " 0.04 Žvariables are mean " S.D., n s 4.. There was no deviation from these values over the period of assessment. An ischemic lesion of the core of the caudate putamen was visible as a pale, slightly stained area in the ischemic hemisphere as early as 1 h after reperfusion, and extended to the entire MCA territory at 3 h by cresyl violet staining Ždata not shown.. The time-dependent increase of infarction in mouse brain using the intraluminal suture blockade is consistent with previous reports that used the same focal stroke model in mice w18,33x. 3.2. Appearance of actiÕated MMP-9 3 h after transient MCA occlusion Both the proform and activated form of the metalloproteinases were evaluated using the zymography technique, in which pro-MMP-9 and MMP-9 appear as characteristic bands of 96 and 88 kDa, while both pro-MMP-2 and MMP-2 were evident as 72 and 62 kDa bands, respectively. The low levels of pro-MMP-9 Žmean optical density; O.D.s 110.7 " 28.7, n s 17. and pro-MMP-2 Žmean O.D.s 88.1 " 28.5, n s 17. were detected in control samples. One hour after reperfusion, moderate increase of pro-MMP-9 was seen in the ischemic samples ŽO.D.s 165.1 " 36.6, n s 3, P - 0.01 compared to normal control. and was sustained 3 h ŽO.D.s 218.9 " 142.2, n s 4, P - 0.01. and 23 h Ž275.9 " 173.4, n s 5, P - 0.001.
after reperfusion ŽFig. 1.. The activated form of MMP-9 was absent in the control samples, whereas it appeared as early as 3 h Žmean O.D.s 18.4 " 15.9, n s 4, P - 0.0001 compared to normal control. and further increased 23 h Žmean O.D.s 44.8 " 26.3, n s 5, P - 0.0001. after reperfusion ŽFig. 1.. pro-MMP-2 showed a significant increase 23 h after reperfusion Ž170.8 " 55.9, n s 5, P - 0.001. but not at 1 and 3 h. Activated MMP-2 was not detected at any time points in the current study. The results of zymography is shown in Table 1. In order to further evaluate the actual proteolytic activity of our extracted samples, we performed a type IV collagenase activity assay which showed a significant increase of proteolytic activity in samples taken from ischemic brain regions of mice subjected to 1 h of ischemia followed by 3 h of reperfusion as compared to control animals ŽFig. 2.. This activity was inhibited by EDTA, confirming the involvement of MMPs w21x. Since no active form of MMP-2 was observed at any time point of our study on our zymographies, we can assume that the diges-
Table 1 Summary of gelatinase activities measured by zymography after transient focal ischemia
Normal control 1h 3h 23 h U
: P - 0.01, control.
UU
pro-MMP-9
MMP-9
pro-MMP-2
110.7"28.7 165.1"36.6U 218.9"142.2U 275.9"173.4UU
0 0 18.4"15.9UUU 44.8"26.3UUU
88.1"28.5 76.3"25.4 99.0"31.4 170.8"55.9UU
: P - 0.001,
UUU
: P - 0.0001, compared to normal
Fig. 2. Graph representing type IV collagenase activity observed in extracted samples from non-ischemic and ischemic brain. Values are expressed in units per milliliter ŽUrml.. Collagenolytic activity was significantly increased 3 h after transient ischemia compared to control conditions ŽU P - 0.005.. This activity was inhibited by EDTA, confirming the involvement of MMPs w21x.
M. Fujimura et al.r Brain Research 842 (1999) 92–100
96
Table 2 Summary of MMP-9 expression after 60 min of MCA occlusion Region
1h
3h
23 h
Cortex Ipsilateral Contralateral
qU y
qq y
qq y
Caudate putamen Ipsilateral Contralateral
q† y
y y
y y
Cortical penumbra Ipsilateral Contralateral
q y
qq y
qq y
q sWeakly detectable expression; qq s prominent expression; y s no expression. U One of the three mice showed qMMP-9 expression. † One of the three mice showed yMMP-9 expression.
tion of type IV collagen revealed in the assay was due to MMP-9. To assess the anatomical distribution of MMP-9
Fig. 4. Western blot analysis of MMP-9 protein after transient ischemia. pro-MMP-9 was evident as 96-kDa characteristic bands in both the non-ischemic and ischemic brain 23 h after reperfusion. Activated MMP-9, as shown by the 88-kDa band, was absent in the non-ischemic brain, but was detected in the ischemic brain. A similar result was obtained from another independent study. c sControl.
after ischemia, immunohistochemistry of MMP-9 was undertaken at 1, 3 and 23 h after reperfusion ŽTable 2.. As shown in Fig. 3, the MMP-9 protein was barely recognized in the normal brain without ischemia ŽFig. 3A,B.. One hour after reperfusion, a slight increase of MMP-9 immunoreactivity was observed in the MCA territory cortex ŽFig. 3C, arrowheads. and a lesser amount in the caudate putamen ŽFig. 3D.. MMP-9 expression was
Fig. 3. Immunohistochemistry of MMP-9 after transient ischemia. MMP-9 was not seen in the non-ischemic brain, including the MCA territory cortex ŽA. and caudate putamen ŽB., while it did appear in the MCA territory cortex ŽC. and to a lesser degree in the caudate putamen ŽD. 1 h after reperfusion. At 3 h, immunoreactivity increased in the ischemic cortex ŽE, arrowheads., while it decreased in the caudate putamen ŽF. compared with 1 h after reperfusion. Twenty-three hours after reperfusion, MMP-9 was observed in the cortex ŽG, arrowheads., however it was barely discernable in the caudate putamen ŽH.. Immunoreactive cells showed heterogenous morphology ŽC, E, G, arrowheads.. An alternative slice incubated without a primary antibody did not show any immunoreactivity ŽI.. The results shown are representative of two independent studies. Scale bar s 20 mm.
M. Fujimura et al.r Brain Research 842 (1999) 92–100
Fig. 5. Reverse zymography after transient ischemia. A similar amount of TIMP-1 was evident as 28-kDa characteristic bands in the ischemic and non-ischemic brain at both 3 and 23 h after reperfusion. The results shown were derived from two samples from each time point. c s Contralateral samples; TIMP-1s control TIMP-1 ŽOncogene Research..
further increased in the MCA territory cortex 3 h after reperfusion ŽFig. 3E, arrowheads., while it was barely detected in the caudate putamen ŽFig. 3F. compared with 1 h after reperfusion. Twenty-three hours after reperfusion, MMP-9 was detected in the MCA territory cortex ŽFig. 3G, arrowheads. but not in the caudate putamen ŽFig. 3H.. Immunoreactive cells showed a relatively heterogenous morphology ŽFig. 3, arrowheads.. There was no immunoreactivity in the control specimens, which were treated without a primary antibody ŽFig. 3I.. To further confirm
97
that the bands were from both the proform and activated form of MMP-9, Western blot analysis was performed using a polyclonal antibody against MMP-9 ŽAnawa.. As shown in Fig. 4, MMP-9 immunoreactivity was evident as bands of molecular mass 96 kDa of the pro-MMP-9 in the non-ischemic sample, and was increased in the sample from the ischemic brain 23 h after reperfusion. The activated MMP-9, as characterized by the 88 kDa band, was absent in the control brain, although it appeared 23 h after reperfusion. These data not only confirm the specificity of the antibody for MMP-9 used in this study, but also suggest that MMP-9 increased after transient FCI. 3.3. TIMP-1 expression after ischemiar reperfusion To evaluate the balance between MMP-9 expression and its endogenous inhibitor, the TIMP-1 level was measured using the reverse zymography technique. A characteristic band of 28 kDa molecular weight was evident in the control sample from the contralateral hemisphere, and was not modified in the ischemic sample ŽFig. 5.. This suggests that the increased expression of MMP-9 was not due to the reduction of TIMP-1.
Fig. 6. Evaluation of BBB permeability by the Evans blue extravasation method after transient ischemia. No interstitial expression of Evans blue was detected in the non-ischemic brain ŽA, C. or in the ischemic brain 1 h after reperfusion ŽB., whereas marked extravasation of Evans blue Žred in color. was seen in the ischemic brain 3 h after reperfusion ŽD.. The results shown were derived from two samples from each time point. Scale bar s 100 mm.
98
M. Fujimura et al.r Brain Research 842 (1999) 92–100
3.4. Vasogenic edema after transient MCA occlusion BBB permeability was evaluated in the ischemic hemisphere as well as in the contralateral side 1 and 3 h after 60 min of MCA occlusion. No Evans blue was detected in the non-ischemic hemispheres ŽFig. 6A,C. Žmean O.D.s 0.315 " 0.081; mean " S.D.. and was not increased in the ischemic hemisphere 1 h after ischemia ŽFig. 6B. Žmean O.D.s 0.470 " 0.268.. Four hours after ischemia, the Evans blue signal was significantly increased in the ischemic hemisphere ŽFig. 6D. compared to the non-ischemic hemisphere Žmean O.D.s 4.251 " 0.899, P - 0.0001..
4. Discussion The current study provides evidence that activated MMP-9 appears as early as 3 h after 60 min of transient FCI, and that a significant increase of pro-MMP-9 occurred in a time dependent manner during reperfusion. These conclusions are derived from the following observations. First, a constitutive expression of pro-MMP-9 in the control specimens was evident as a 96-kDa characteristic band by both zymography ŽFig. 1. and Western blot analysis ŽFig. 4., and markedly increased during reperfusion after FCI ŽFig. 1.. Second, activated MMP-9, characterized as an 88 kDa band, was absent in the control specimens, whereas it appeared as early as 3 h and further increased at 23 h after reperfusion ŽFig. 1.. Moreover, the activated nature of this 88 kDa MMP-9 was supported by the presence of a significant increase of type IV collagenase activity in our extracted samples obtained from ischemic brains 3 h after reperfusion ŽFig. 2.. Up-regulation of pro-MMP-9 was first reported by Rosenberg and his colleagues using a permanent FCI model of spontaneous hypertensive rats w24x, in which a marked increase of pro-MMP-9 was shown 12 h after permanent ischemia, but no activated MMP-9 was found after ischemia. In their study, pro-MMP-9 was not detected in the control specimens, while it appeared 12 h after the initiation of ischemia w24x. However, the activated forms of MMP-9 and MMP-2 were not detected w24x. As for transient FCI, a more recent study by Rosenberg et al. w27x successfully showed the significant increase of pro-MMP-9 48 h after 2 h of MCA occlusion; however, the activated form of MMP-9 was not detected. The discrepancy between our results in the current study and these previous studies could be due to the following reasons: Ž1. we employed the protein extraction method with detergent extraction and affinity-support purification, which was recently reported to be capable of recovering gelatinase activity over a 4- to 10-fold range using zymography w36x. The use of this method could contribute to the early detection of activated MMP-9 Ž88
kDa. as well as to the detection of pro-MMP-9 Ž96 kDa. in the control specimens; andror Ž2. since we used the mouse model of 1 h focal ischemia and subsequent reperfusion w33x, a difference in the species used in this study could have affected the results of MMP-9 induction andror activation. Reperfusion injury is thought to play a critical role in cerebral infarction and BBB disruption w34x. Although reperfusion of ischemic tissue with thrombolytic agents after a short period of ischemia is considered to reduce infarct volume w1x, reperfusion at a later period exacerbates ischemic damage. In the present study, we used the transient FCI model of 60 min of MCA occlusion and subsequent reperfusion, which consistently produced infarction and brain edema in the MCA territory area on the ischemic side w18,33x. The present study demonstrated the evidence of BBB disruption by the Evans blue extravasation method 3 h but not 1 h after reperfusion ŽFig. 6D., as confirmed by the semi-quantitative analysis of Evans blue extravasation using the densitometer. These data are consistent with a previous report showing the increase of brain sucrose uptake 4 h after 2 h of MCA occlusion w27x. At the same time point as Evans blue leakage, we detected the appearance of activated MMP-9 ŽFig. 1., which may suggest the possible involvement of MMP-9 activation in early BBB dysfunction. In fact, it was reported that the MMP inhibitor BB-1101 reduced the BBB opening 3 h after transient FCI w27x, which may also support the idea that activation of MMP-9 contributes to BBB dysfunction at the early stage of transient FCI. Reactive oxygen species ŽROS. are produced during reperfusion, and have been implicated in BBB disruption and the expansion of infarction after FCI. Based on this, we have reported that antioxidant enzymes, such as superoxide dismutase ŽSOD.-1, play a protective role in transient focal ischemia w6,7,16,18,33x, suggesting that free radical production during reperfusion exacerbates cerebral infarction. SOD-1 transgenic mice show a significant reduction in infarct volume after transient focal ischemia w16,33x, whereas infarction is increased in SOD-1 knockouts after transient ischemia. Although the molecular cascade between free radical production and the exacerbation of ischemiarreperfusion injury is still controversial, we may find some clue in recent in vitro studies. It was reported that inflammatory molecules such as intercellular adhesion molecule-1 ŽICAM-1. w2x, and MMPs w22,29x could be induced in cultured endothelial cells after hypoxiarreoxygenation. Also, in vivo studies have shown that ICAM-1 is markedly induced after ischemiarreperfusion, and that ICAM-1 knockouts show a smaller infarction after transient ischemia w10,30x, which suggests the deleterious role of this molecule in ischemiarreperfusion injury. In fact, antibodies against ICAM-1 reduced infarction after transient FCI w8x, but not after permanent FCI w35x. In the present study, no activated MMP-9 was detected in the control specimens, while it did appear at 3 h
M. Fujimura et al.r Brain Research 842 (1999) 92–100
and was further increased 23 h after 60 min of transient FCI ŽFig. 1.. Furthermore, a time dependent increase of pro-MMP-9 was seen during reperfusion after ischemia. Although further examination is necessary to elucidate the direct relationship between free radical production during reperfusion and MMP-9 induction, our data suggest that reperfusion after FCI may contribute to the activation of MMP-9 as well as the increase of pro-MMP-9. The exact mechanisms involved in the induction and activation of MMPs after ischemiarreperfusion are not well defined. Since MMP-9 has an activator protein-1 ŽAP-1. binding site in its promoter region, it is conceivable that the induction of pro-MMP-9 is associated with the alteration of AP-1 binding activity during reperfusion. In fact, a significant increase in AP-1 binding activity was detected after 1 h of reperfusion following 90 min of MCA occlusion in rats w28x, which is compatible with our results showing pro-MMP-9 induction 3 h after reperfusion. It is also conceivable that free radical production during reperfusion affects the activation of MMP-9, since it has been suggested that ROS regulate the activity of MMPs in vitro w22,29x. Based on this, we are currently undertaking experiments to elucidate whether the induction and activation of MMP-9 is affected in transgenic mice that overexpresss the anti-oxidant enzyme, SOD-1. In the present study, different patterns of MMP-9 expression were seen between each time point by immunohistochemistry ŽFig. 3.. One hour after reperfusion, an increase in MMP-9 expression was seen in the entire area of the ischemic brain ŽFig. 3C,D., while at later time points such as 3 and 23 h, more MMP-9 expression was detected in the MCA territory cortex ŽFig. 3E,G. compared with the ischemic core of the caudate putamen ŽFig. 3F,H.. It is unclear why there was a regional predominance in the cortex rather than in the ischemic core of the caudate putamen at the later time points. Since the core of the caudate putamen is assumed to be undergoing more severe ischemia than the MCA territory cortex that has collateral circulation such as leptomeningeal anastomosis, it is conceivable that severe ischemia in the ischemic core during MCA occlusion may contribute to less MMP-9 expression in the caudate putamen at the later time points. As for MMP-9 cell populations, they could be induced in a variety of cells in the central nervous system, including endothelial cells w19x, astrocytes w12x, oligodendrocytes w31x, microglia w13x, and neurons w3x. Also in the model of permanent FCI in rats, early MMP-9 expression on endothelial cells as well as on neutrophils was reported, while there was no alteration of TIMP-1 after ischemia w23x. In the present study, immunohistochemistry revealed the heterogeneous morphology of the immunopositive cells in the area of the MCA territory on the ischemic side ŽFig. 3.. They are likely to involve neuronal ŽFig. 3E, arrowheads. and endothelial cell ŽFig. 3G. populations. A definition of the cell populations of these immunopositive cells remains to be elucidated in a future study. Double staining
99
of MMP-9 and various cell markers would be useful to clarify this issue. In conclusion, we have shown that MMP-9 was activated as early as 3 h after transient FCI, and that proMMP-9 also increased in a time dependent manner after transient ischemia. Although further examination is necessary to elucidate the relationship between the activation of MMP-9 and ischemiarreperfusion injury, our results suggest the possibility that early activation of MMP-9 may contribute to BBB dysfunction and subsequent formation of vasogenic edema after transient FCI. It is suggested that pharmacological or therapeutic intervention should be considered in targeting the activation andror induction of MMP-9 in FCI.
Acknowledgements The authors are grateful to Ms. Cheryl Christensen for her editorial assistance and Ms. Liza Reola, Mr. Bernard Calagui and Ms. Jane O. Kim for their technical assistance. This study was supported by National Institutes of Health grants NS25372, NS14543, NS36147, NS38653 and NO1 NS82386. P.H.C. is a recipient of the Jacob Javits Neuroscience Investigator Award.
References w1x The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group, Tissue plasminogen activator for acute ischemic stroke, N. Engl. J. Med. 333 Ž1995. 1581–1587. w2x T. Aoki, Y. Suzuki, K. Nishio, K. Suzuki, A. Miyata, Y. Oyamada, M. Mori, H. Fujita, K. Yamaguchi, Effect of antioxidants on hyperoxia-induced ICAM-1 expression in human endothelial cells, Adv. Exp. Med. Biol. 411 Ž1997. 503–511. w3x J.R. Backstrom, G.P. Lim, M.J. Cullen, Z.A. Tokes, Matrix metalloproteinase-9 ŽMMP-9. is synthesized in neurons of the human hippocampus and is capable of degrading the amyloid-beta peptide Ž1–40., J. Neurosci. 16 Ž1996. 7910–7919. w4x L. Belayev, R. Busto, W. Zhao, M.D. Ginsberg, Quantitative evaluation of blood–brain barrier permeability following middle cerebral artery occlusion in rats, Brain Res. 739 Ž1996. 88–96. w5x G. Bruno, R. Todor, I. Lewis, D. Chyatte, Vascular extracellular matrix remodeling in cerebral aneurysms, J. Neurosurg. 89 Ž1998. 431–440. w6x P.H. Chan, C.J. Epstein, Y. Li, T.T. Huang, E. Carlson, H. Kinouchi, G. Yang, H. Kamii, S. Mikawa, T. Kondo et al., Transgenic mice and knockout mutants in the study of oxidative stress in brain injury, J. Neurotrauma 12 Ž1995. 815–824. w7x P.H. Chan, Role of oxidants in ischemic brain damage, Stroke 27 Ž1996. 1124–1129. w8x M. Chopp, Y. Li, N. Jiang, R.L. Zhang, J. Prostak, Antibodies against adhesion molecules reduce apoptosis after transient middle cerebral artery occlusion in rat brain, J. Cereb. Blood Flow Metab. 16 Ž1996. 578–584. w9x A.W. Clark, C.A. Krekoski, S.S. Bou, K.R. Chapman, D.R. Edwards, Increased gelatinase A ŽMMP-2. and gelatinase B ŽMMP-9. activities in human brain after focal ischemia, Neurosci. Lett. 238 Ž1997. 53–56.
100
M. Fujimura et al.r Brain Research 842 (1999) 92–100
w10x E.S. Connolly Jr., C.J. Winfree, T.A. Springer, Y. Naka, H. Liao, S.D. Yan, D.M. Stern, R.A. Solomon, J.C. Gutierrez-Ramos, D.J. Pinsky, Cerebral protection in homozygous null ICAM-1 mice after middle cerebral artery occlusion. Role of neutrophil adhesion in the pathogenesis of stroke, J. Clin. Invest. 97 Ž1996. 209–216. w11x M. Fujimura, Y. Morita-Fujimura, K. Murakami, M. Kawase, P.H. Chan, Cytosolic redistribution of cytochrome c after transient focal cerebral ischemia in rats, J. Cereb. Blood Flow Metab. 18 Ž1998. 1239–1247. w12x P.E. Gottschall, X. Yu, Cytokines regulate gelatinase A and B Žmatrix metalloproteinase 2 and 9. activity in cultured rat astrocytes, J. Neurochem. 64 Ž1995. 1513–1520. w13x P.E. Gottschall, X. Yu, B. Bing, Increased production of gelatinase B Žmatrix metalloproteinase-9. and interleukin-6 by activated rat microglia in culture, J. Neurosci. Res. 42 Ž1995. 335–342. w14x G.F. Hamann, Y. Okada, R. Fitridge, G.J. del Zoppo, Microvascular basal lamina antigens disappear during cerebral ischemia and reperfusion, Stroke 26 Ž1995. 2120–2126. w15x S.C. Kim, M. Singh, J. Huang, C.J. Prestigiacomo, C.J. Winfree, R.A. Solomon, E.S. Connolly Jr., Matrix metalloproteinase-9 in cerebral aneurysms, Neurosurgery 41 Ž1997. 642–666. w16x H. Kinouchi, C.J. Epstein, T. Mizui, E. Carlson, S.F. Chen, P.H. Chan, Attenuation of focal cerebral ischemic injury in transgenic mice overexpressing CuZn superoxide dismutase, Proc. Natl. Acad. Sci. U.S.A. 88 Ž1991. 11158–11162. w17x D.E. Kleiner, W.G. Stetler-Stevenson, Quantitative zymography: detection of picogram quantities of gelatinases, Anal. Biochem. 218 Ž1994. 325–329. w18x T. Kondo, A.G. Reaume, T.T. Huang, E. Carlson, K. Murakami, S.F. Chen, E.K. Hoffman, R.W. Scott, C.J. Epstein, P.H. Chan, Reduction of CuZn-superoxide dismutase activity exacerbates neuronal cell injury and edema formation after transient focal cerebral ischemia, J. Neurosci. 17 Ž1997. 4180–4189. w19x M. Nguyen, J. Arkell, C.J. Jackson, Active and tissue inhibitor of matrix metalloproteinase-free gelatinase B accumulates within human microvascular endothelial vesicles, J. Biol. Chem. 273 Ž1998. 5400–5404. w20x G.W. Oliver, J.D. Leferson, W.G. Stetler-Stevenson, D.E. Kleiner, Quantitative reverse zymography: analysis of picogram amounts of metalloproteinase inhibitors using gelatinase A and B reverse zymograms, Anal. Biochem. 244 Ž1997. 161–166. w21x A.R. Quesada, M.M. Barbacid, E. Mira, P. Fernandez-Resa, G. Marquez, M. Aracil, Evaluation of fluorometric and zymographic methods as activity assays for stromelysins and gelatinases, Clin. Exp. Metastasis 15 Ž1997. 26–32. w22x S. Rajagopalan, X.P. Meng, S. Ramasamy, D.G. Harrison, Z.S. Galis, Reactive oxygen species produced by macrophage-derived foam cells regulate the activity of vascular matrix metalloproteinases in vitro. Implications for atherosclerotic plaque stability, J. Clin. Invest. 98 Ž1996. 2572–2579.
w23x A.M. Romanic, R.F. White, A.J. Arleth, E.H. Ohlstein, F.C. Barone, Matrix metalloproteinase expression increases after cerebral focal ischemia in rats: inhibition of matrix metalloproteinase-9 reduces infarct size, Stroke 29 Ž1998. 1020–1030. w24x G.A. Rosenberg, M. Navratil, F. Barone, G. Feuerstein, Proteolytic cascade enzymes increase in focal cerebral ischemia in rat, J. Cereb. Blood Flow Metab. 16 Ž1996. 360–366. w25x G.A. Rosenberg, J.E. Dencoff, N. Correa Jr., M. Reiners, C.C. Ford, Effect of steroids on CSF matrix metalloproteinases in multiple sclerosis: relation to blood–brain barrier injury, Neurology 46 Ž1996. 1626–1632. w26x G.A. Rosenberg, M. Navratil, Metalloproteinase inhibition blocks edema in intracerebral hemorrhage in the rat, Neurology 48 Ž1997. 921–926. w27x G.A. Rosenberg, E.Y. Estrada, J.E. Dencoff, Matrix metalloproteinases and TIMPs are associated with blood–brain-barrier opening after reperfusion in rat brain, Stroke 29 Ž1998. 2189–2195. w28x A. Salminen, P.K. Liu, C.Y. Hsu, Alteration of transcription factor binding activities in the ischemic rat brain, Biochem. Biophys. Res. Commun. 212 Ž1995. 939–944. w29x M. Shingu, T. Isayama, C. Yasutake, T. Naono, M. Nobunaga, K. Tomari, K. Horie, Y. Goto, Role of oxygen radicals and IL-6 in IL-1-dependent cartilage matrix degradation, Inflammation 18 Ž1994. 613–623. w30x S.G. Soriano, S.A. Lipton, Y.F. Wang, M. Xiao, T.A. Springer, J.C. Gutierrez-Ramos, P.R. Hickey, Intercellular adhesion molecule-1deficient mice are less susceptible to cerebral ischemia-reperfusion injury, Ann. Neurol. 39 Ž1996. 618–624. w31x J.H. Uhm, N.P. Dooley, L.Y. Oh, V.W. Yong, Oligodendrocytes utilize a matrix metalloproteinase, MMP-9, to extend processes along an astrocyte extracellular matrix, Glia 22 Ž1998. 53–63. w32x H.E. Van Wart, H. Birkedal-Hansen, The cysteine switch: a principle of regulation of metalloproteinase activity with potential applicability to the entire matrix metalloproteinase gene family, Proc. Natl. Acad. Sci. U.S.A. 87 Ž1990. 5578–5582. w33x G. Yang, P.H. Chan, J. Chen, E. Carlson, S.F. Chen, P. Weinstein, C.J. Epstein, H. Kamii, Human copper–zinc superoxide dismutase transgenic mice are highly resistant to reperfusion injury after focal cerebral ischemia, Stroke 25 Ž1994. 165–170. w34x G.Y. Yang, A.L. Betz, Reperfusion-induced injury to the blood–brain barrier after middle cerebral artery occlusion in rats, Stroke 25 Ž1994. 1658–1664, Discussion 1664-1655. w35x R.L. Zhang, M. Chopp, N. Jiang, W.X. Tang, J. Prostak, A.M. Manning, D.C. Anderson, Anti-intercellular adhesion molecule-1 antibody reduces ischemic cell damage after transient but not permanent middle cerebral artery occlusion in the Wistar rat, Stroke 26 Ž1995. 1438–1442. w36x J.W. Zhang, P.E. Gottschall, Zymographic measurement of gelatinase activity in brain tissue after detergent extraction and affinitysupport purification, J. Neurosci. Methods 76 Ž1997. 15–20.
Research report
Early appearance of activated matrix metalloproteinase-9 and blood–brain barrier disruption in mice after focal cerebral ischemia and reperfusion Miki Fujimura, Yvan Gasche, Yuiko Morita-Fujimura, Justin Massengale, Makoto Kawase, Pak H. Chan ) Department of Neurosurgery, Department of Neurology and Neurological Sciences, and Program in Neurosciences, Stanford UniÕersity School of Medicine, Palo Alto, CA, USA Accepted 7 July 1999
Abstract Blood–brain barrier ŽBBB. disruption is thought to play a critical role in the pathophysiology of ischemiarreperfusion. Matrix metalloproteinases ŽMMPs. are a family of proteolytic enzymes that can degrade all the components of the extracellular matrix when they are activated. Gelatinase A ŽMMP-2. and gelatinase B ŽMMP-9. are able to digest the endothelial basal lamina, which plays a major role in maintaining BBB impermeability. The present study examined the expression and activation of gelatinases before and after transient focal cerebral ischemia ŽFCI. in mice. Adult male CD1 mice were subjected to 60 min FCI and reperfusion. Zymography was performed from 1 to 23 h after reperfusion using the protein extraction method with detergent extraction and affinity-support purification. MMP-9 expression was also examined by both immunohistochemistry and Western blot analysis, and tissue inhibitors to metalloproteinase-1 was measured by reverse zymography. The BBB opening was evaluated by the Evans blue extravasation method. The 88-kDa activated MMP-9 was absent from the control specimens, while it appeared 3 h after transient ischemia by zymography. At this time point, the BBB permeability alteration was detected in the ischemic brain. Both pro-MMP-9 Ž96 kDa. and pro-MMP-2 Ž72 kDa. were seen in the control specimens, and were markedly increased after FCI. A significant induction of MMP-9 was confirmed by both immunohistochemistry and Western blot analysis. The early appearance of activated MMP-9, associated with evidence of BBB permeability alteration, suggests that activation of MMP-9 contributes to the early formation of vasogenic edema after transient FCI. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Matrix metalloproteinase; Cerebral ischemia; Reperfusion injury; Blood–brain barrier; Brain edema
1. Introduction Matrix metalloproteinases ŽMMPs. are a family of zinc-binding proteolytic enzymes which are capable of degrading components of the extracellular matrix in a variety of physiological and pathophysiological conditions. Among MMPs, gelatinase A ŽMMP-2. and gelatinase B ŽMMP-9. are able to digest the endothelial basal lamina, which plays a major role in maintaining blood–brain barrier ŽBBB. impermeability, by regulating tight junctions leading to the opening of the BBB. In the pathological condition of ischemiarreperfusion, digestion of the endothelial basal lamina is reported to occur as early as 2 h after ischemia w14x, which may result in BBB permeability ) Corresponding author. Neurosurgical Laboratories, Stanford University, 701B Welch Road, a148, Palo Alto, CA 94304, USA. Fax: q1650-498-4550; E-mail: [email protected]
a few hours after ischemia w4,18x. Recently, MMP-9 andror MMP-2 have been implicated in focal cerebral ischemia ŽFCI. w9,23,24,27x, as well as other central nervous system disorders such as multiple sclerosis w25x, intracerebral hemorrhage w26x and cerebral aneurysm w5,15x. A previous report by Rosenberg and his colleagues on FCI showed the marked increase of pro-MMP-9 after both permanent and transient ischemia, suggesting the possible involvement of MMP-9 in FCI w24,27x. However, it has still not been established whether the 88-kDa activated MMP-9, which is considered to be the fully active form to digest basal lamina w32x, could appear after transient FCI. In the present study, we sought to clarify this point by measuring gelatinase activity in the brain before and after transient FCI by using the protein extraction method with detergent extraction and affinity-support purification, which allowed us to recover gelatinase activity over a 4- to 10-fold range using zymography w36x. The results showed
0006-8993r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 6 - 8 9 9 3 Ž 9 9 . 0 1 8 4 3 - 0
M. Fujimura et al.r Brain Research 842 (1999) 92–100
93
that 88-kDa activated MMP-9 is absent from the control specimens, while it appeared 3 h after 60 min of transient ischemia, as shown by zymography and type IV collagenase activity assay. At this time point, BBB permeability alteration was detected in the ischemic brain by the Evans blue extravasation method. Both pro-MMP-9 Ž96 kDa. and pro-MMP-2 Ž72 kDa. were detected in the control specimens and increased after transient FCI. A significant induction of MMP-9 was confirmed by both immunohistochemistry and Western blot analysis after transient FCI. These results, which are consistent with a previous report showing the prevention of BBB impermeability with the MMP-9 inhibitor BB-1101 3 h after reperfusion w27x, suggest that the activation of MMP-9 during reperfusion may contribute to BBB disruption and the formation of vasogenic edema after transient FCI.
mM CaCl 2 , 0.05% BRIJ-35, 0.02% NaN3 , 1% Triton X-100.. Aliquots Ž10 ml. of the homogenates were saved for total protein measurements ŽBCA Kit, Pierce, Rockford, IL.. The homogenates were centrifuged at 12,000 = g for 5 min. The supernatants were recovered and incubated for 60 min with gelatin-sepharose 4B ŽPharmacia Biotech, Uppsala, Sweden., with constant shaking. After incubation, the samples were centrifuged at 500 = g for 2 min. The pellets were washed with a buffer containing 50 mM Tris–HCl, pH 7.6, 150 mM NaCl, 5 mM CaCl 2 , 0.05% BRIJ-35, 0.02% NaN3 . After a second centrifugation, the pellets were resuspended in elution buffer Ž50 mM Tris– HCl, pH 7.6, 150 mM NaCl, 5 mM CaCl 2 , 0.05% BRIJ-35, 0.02% NaN3 , dimethylsulfoxide 10%. for 30 min. Then the samples were subjected to zymographic measurements of gelatinase activity, to type IV collagenolytic activity assay, and to Western blot analysis.
2. Materials and methods
2.3. Zymographic measurement of gelatinase actiÕity
2.1. Focal cerebral ischemia and histological assessment Adult male CD-1 mice Ž35–40 g, n s 33. were subjected to transient FCI by intraluminal middle cerebral artery ŽMCA. blockade with a nylon suture as described w33x. The mice were anesthetized with 2.0% isoflurane in 30% oxygen and 70% nitrous oxide using a face mask. The rectal temperature was controlled at 378C with a homeothermic blanket. Cannulation of a femoral artery allowed the monitoring of blood pressure and arterial blood gases, samples for analysis being taken immediately after cannulation, 10 min after occlusion and 10 min after reperfusion. After the midline skin incision, the left external carotid artery was exposed and its branches were electrocoagulated. A 11.0-mm 5-0 surgical monofilament nylon suture, blunted at the end, was introduced into the left internal carotid artery through the external carotid artery stump. After 60 min of MCA occlusion, blood flow was restored by the withdrawal of the nylon suture. The experimental animals were killed 1, 3 and 23 h after 60 min of MCA occlusion. The brains were removed, rapidly frozen in y208C 2-methylbutane, and stored at y808C. They were sectioned with a cryostat into a thickness of 20 mm from the anterior side to the posterior side and stained with cresyl violet. 2.2. Preparation of tissue extracts To analyze the gelatinase activity and MMP protein expression patterns in control and ischemic brain tissue, protein extraction of the tissue was performed as previously described w36x. Approximately 100 mg of both ischemic brain and homologous tissue from the contralateral side were extracted and rapidly frozen in powdered dry ice after 1, 3 and 23 h of reperfusion. The brain tissue was homogenized with 2 ml Teflon glass homogenizers in a lysis buffer Ž50 mM Tris–HCl, pH 7.6, 150 mM NaCl, 5
Zymographies were performed according to a previously reported method with minor modifications w17x. Ten microliters of a non-reducing sample buffer Ž0.4 M Tris, pH 6.8, 5% sodium dodecyl sulfate ŽSDS., 20% glycerol, 0.05% bromophenol blue. were added to the sample volume which was corrected for protein concentration. Samples were then loaded onto 8% SDS-polyacrylamide gel electrophoresis ŽPAGE. gels in which 0.15% porcine skin gelatin ŽSigma, St. Louis, MO. was co-polymerized. The gels were run at 40 mA for 3 h. After migration, the gels were incubated with 2.5% Triton-X 100 twice for 1 h at room temperature, washed for 10 min in a buffer containing 50 mM Tris, pH 7.5, 200 mM NaCl, 5 mM CaCl 2 ŽTNCA. and further incubated for 16 h in TNCA, in a water bath at 378C. The gels were then stained for 90 min in coomassie blue Ž1% coomassie brilliant blue, 30% methanol, 10% acetic acid. and destained in 30% methanolr10% acetic acid, four times for 5, 15, 30 and 60 min, respectively. White bands on a blue background indicated zones of digestion corresponding to the presence of different MMPs. The 96-kDa band corresponded to proMMP-9, the 88-kDa band to MMP-9, and the 72-kDa band corresponded to proMMP-2. Twenty microliters of MMP-9 and MMP-2 standards ŽOncogene Research, Cambridge, MA. were loaded into each gel to allow comparison between the different gels. The bands were scanned using a densitometer ŽGS-700, Bio-Rad Laboratories, Richmond, CA. and quantification was performed using the Multi-Analyst 1.0 software ŽBio-Rad.. 2.4. Type IV collagenase actiÕity assay Type IV collagenase activity assay was performed using a commercially available kit ŽChemicon International, Temecula, CA.. According to the manufacturer protocol, samples were incubated with TNCA and FITC labeled type IV collagen solution, first for 20 min at room temperature and then for 2 h at 428C in a water bath. Samples
94
M. Fujimura et al.r Brain Research 842 (1999) 92–100
were then placed on ice for 5 min. Three hundred microliters of cold enzyme stop reagentrextraction solution were added to all samples which were further stored for 35 min on ice. Residual undigested collagen was precipitated by centrifugation Ž10 min at 12,200 = g .. A standard curve was made using MMP-9 standard solution provided by the manufacturer. Proteolytic activity of ischemic samples was measured in the presence or absence of ethylenediaminetetraacetic acid ŽEDTA, 0.5 mM., which is known to inhibit MMPs w21x. Finally, fluorescence intensity of sample supernatants was measured with a fluorimeter ŽMolecular Devices, Sunnyvale, CA. at 520 nmr495 nm. 2.5. Immunohistochemistry of MMP-9 To assess the anatomical distribution of MMP-9 after ischemiarreperfusion, immunohistochemistry of MMP-9 was performed. Anesthetized animals were perfused with 10 Urml heparin and subsequently with 4% formaldehyde in 0.1 M phosphate-buffered saline ŽPBS. ŽpH 7.4. 1, 3 and 23 h after reperfusion. The brains were removed, postfixed for 12 h in 4% formaldehyde, sectioned at 50 mm on a vibratome, and processed for immunohistochemistry. The sections were incubated with a blocking solution as described w11x and reacted with anti-MMP-9 polyclonal antibody ŽAnawa, Wangen, Switzerland. at a dilution of 1:50. Immunohistochemistry was performed using the avidin–biotin technique as described w11x, and then the nuclei were counterstained with methyl green solution for 10 min. As a negative control, sections were incubated without primary antibodies. For histological assessment, alternate slices from each brain section were stained with cresyl violet. 2.6. Western blot analysis To further investigate the protein expression of MMP-9 in the samples from the control and ischemic brains, Western blot analysis was performed as previously described w11x. Equal amounts of the protein were loaded per lane. The primary antibodies were used at a 1:500 dilution of polyclonal antibody against MMP-9 ŽAnawa.. Western blots were performed with horseradish peroxidase-conjugated anti-rabbit immunoglobulin G using the Boehringer Mannheim ŽIndianapolis, IN. chemiluminescent system, and bands were visualized by enhanced chemiluminescence reagent ŽECL plus Western blot analysis system, Amersham Life Science, Arlington Heights, IL.. Recombinant human proMMP-9 and MMP-9 were used as controls.
20% glycerol, 0.05% bromophenol blue. were added to the samples Žvolume corrected for protein concentration.. The samples were then loaded onto 15% SDS-PAGE gels in which 2.2 mgrml porcine skin gelatin ŽSigma. and 150 ngrml of proMMP-9 were co-polymerized. The gels were run at 40 mA for 3 h. After migration, the gels were incubated with 2.5% Triton-X 100 twice for 1 h at room temperature, washed for 10 min in TNCA, and further incubated for 16 h in TNCA in a water bath at 378C. The gels were stained for 30 min in coomassie blue Ž1% coomassie brilliant blue, 30% methanol, 10% acetic acid. and destained in 30% methanolr10% acetic acid, four times for 5, 15, 30 and 60 min, respectively. Tissue inhibitor of metalloproteinase-1 ŽTIMP-1. was identified as a 28-kDa blue band; 20 ml of TIMP-1 ŽOncogene Research. were loaded as a standard. The bands were scanned using a densitometer ŽGS-700, Bio-Rad. and quantification was performed using the Multi-Analyst 1.0 software ŽBio-Rad.. 2.8. EÕans blue extraÕasation In situ detection of Evans blue leakage was based on the previously described method w18x with modification. One hour before the transcardiac perfusion, 0.1 ml of 4% Evans blue ŽSigma. in 0.9% saline was intravenously injected. The animals were perfused with 10 Urml heparin Ž200 ml. and subsequently with 4% formaldehyde in 0.1 M PBS ŽpH 7.4. 1 and 3 h after reperfusion. The brains were sectioned with a vibratome, mounted on slides, and counterstained with Hoechst 33258 ŽMolecular Probes, Eugene, OR.. BBB disruption was assessed fluoromicroscopically at excitations 355 nm and emission ) 415 nm for Evans blue detection. To analyze the fluorescence signal of Evans blue, we scanned photomicrographs Ž=630. with a GS-700 imaging densitometer ŽBio-Rad. and then measured the red signal intensity in 15 individual regions from three animals in each group using Muti-Analyst software ŽBio-Rad.. 2.9. Statistical analysis Data are presented as the mean " S.D. Statistical comparisons were made using unpaired ANOVA with STAT VIEW software, version 4.5 ŽAbacus Concepts, Berkeley, CA.. P - 0.05 was considered to be statistically significant.
3. Results 3.1. Physiological data and cerebral infarction
2.7. ReÕerse zymography Reverse zymographies were performed as previously described w20x with modifications. Ten microliters of a non-reducing sample buffer Ž0.4 M Tris, pH 6.8, 5% SDS,
Physiological parameters showed no significant differences in mean arterial blood pressure ŽMABP. and arterial blood gas analysis between each time point. The preischemic physiological variables were as follows: MABP,
M. Fujimura et al.r Brain Research 842 (1999) 92–100
95
Fig. 1. Zymographic measurement of gelatinase activities before and after transient ischemia. Pro-MMP-9 Ž96 kDa. and pro-MMP-2 were present in the non-ischemic brain and increased after ischemiarreperfusion. Activated MMP-9 Ž88 kDa. was absent in the control specimens, although it appeared at 3 h and further increased 23 h after reperfusion. Activated MMP-2 was absent through all the time points. n s non-ischemic control; c s contralateral brain.
71.0 " 3.8 mmHg; PaO 2 , 146.4 " 9.9 mmHg; PaCO 2 , 28.9 " 7.8 mmHg; pH, 7.35 " 0.04 Žvariables are mean " S.D., n s 4.. There was no deviation from these values over the period of assessment. An ischemic lesion of the core of the caudate putamen was visible as a pale, slightly stained area in the ischemic hemisphere as early as 1 h after reperfusion, and extended to the entire MCA territory at 3 h by cresyl violet staining Ždata not shown.. The time-dependent increase of infarction in mouse brain using the intraluminal suture blockade is consistent with previous reports that used the same focal stroke model in mice w18,33x. 3.2. Appearance of actiÕated MMP-9 3 h after transient MCA occlusion Both the proform and activated form of the metalloproteinases were evaluated using the zymography technique, in which pro-MMP-9 and MMP-9 appear as characteristic bands of 96 and 88 kDa, while both pro-MMP-2 and MMP-2 were evident as 72 and 62 kDa bands, respectively. The low levels of pro-MMP-9 Žmean optical density; O.D.s 110.7 " 28.7, n s 17. and pro-MMP-2 Žmean O.D.s 88.1 " 28.5, n s 17. were detected in control samples. One hour after reperfusion, moderate increase of pro-MMP-9 was seen in the ischemic samples ŽO.D.s 165.1 " 36.6, n s 3, P - 0.01 compared to normal control. and was sustained 3 h ŽO.D.s 218.9 " 142.2, n s 4, P - 0.01. and 23 h Ž275.9 " 173.4, n s 5, P - 0.001.
after reperfusion ŽFig. 1.. The activated form of MMP-9 was absent in the control samples, whereas it appeared as early as 3 h Žmean O.D.s 18.4 " 15.9, n s 4, P - 0.0001 compared to normal control. and further increased 23 h Žmean O.D.s 44.8 " 26.3, n s 5, P - 0.0001. after reperfusion ŽFig. 1.. pro-MMP-2 showed a significant increase 23 h after reperfusion Ž170.8 " 55.9, n s 5, P - 0.001. but not at 1 and 3 h. Activated MMP-2 was not detected at any time points in the current study. The results of zymography is shown in Table 1. In order to further evaluate the actual proteolytic activity of our extracted samples, we performed a type IV collagenase activity assay which showed a significant increase of proteolytic activity in samples taken from ischemic brain regions of mice subjected to 1 h of ischemia followed by 3 h of reperfusion as compared to control animals ŽFig. 2.. This activity was inhibited by EDTA, confirming the involvement of MMPs w21x. Since no active form of MMP-2 was observed at any time point of our study on our zymographies, we can assume that the diges-
Table 1 Summary of gelatinase activities measured by zymography after transient focal ischemia
Normal control 1h 3h 23 h U
: P - 0.01, control.
UU
pro-MMP-9
MMP-9
pro-MMP-2
110.7"28.7 165.1"36.6U 218.9"142.2U 275.9"173.4UU
0 0 18.4"15.9UUU 44.8"26.3UUU
88.1"28.5 76.3"25.4 99.0"31.4 170.8"55.9UU
: P - 0.001,
UUU
: P - 0.0001, compared to normal
Fig. 2. Graph representing type IV collagenase activity observed in extracted samples from non-ischemic and ischemic brain. Values are expressed in units per milliliter ŽUrml.. Collagenolytic activity was significantly increased 3 h after transient ischemia compared to control conditions ŽU P - 0.005.. This activity was inhibited by EDTA, confirming the involvement of MMPs w21x.
M. Fujimura et al.r Brain Research 842 (1999) 92–100
96
Table 2 Summary of MMP-9 expression after 60 min of MCA occlusion Region
1h
3h
23 h
Cortex Ipsilateral Contralateral
qU y
qq y
qq y
Caudate putamen Ipsilateral Contralateral
q† y
y y
y y
Cortical penumbra Ipsilateral Contralateral
q y
qq y
qq y
q sWeakly detectable expression; qq s prominent expression; y s no expression. U One of the three mice showed qMMP-9 expression. † One of the three mice showed yMMP-9 expression.
tion of type IV collagen revealed in the assay was due to MMP-9. To assess the anatomical distribution of MMP-9
Fig. 4. Western blot analysis of MMP-9 protein after transient ischemia. pro-MMP-9 was evident as 96-kDa characteristic bands in both the non-ischemic and ischemic brain 23 h after reperfusion. Activated MMP-9, as shown by the 88-kDa band, was absent in the non-ischemic brain, but was detected in the ischemic brain. A similar result was obtained from another independent study. c sControl.
after ischemia, immunohistochemistry of MMP-9 was undertaken at 1, 3 and 23 h after reperfusion ŽTable 2.. As shown in Fig. 3, the MMP-9 protein was barely recognized in the normal brain without ischemia ŽFig. 3A,B.. One hour after reperfusion, a slight increase of MMP-9 immunoreactivity was observed in the MCA territory cortex ŽFig. 3C, arrowheads. and a lesser amount in the caudate putamen ŽFig. 3D.. MMP-9 expression was
Fig. 3. Immunohistochemistry of MMP-9 after transient ischemia. MMP-9 was not seen in the non-ischemic brain, including the MCA territory cortex ŽA. and caudate putamen ŽB., while it did appear in the MCA territory cortex ŽC. and to a lesser degree in the caudate putamen ŽD. 1 h after reperfusion. At 3 h, immunoreactivity increased in the ischemic cortex ŽE, arrowheads., while it decreased in the caudate putamen ŽF. compared with 1 h after reperfusion. Twenty-three hours after reperfusion, MMP-9 was observed in the cortex ŽG, arrowheads., however it was barely discernable in the caudate putamen ŽH.. Immunoreactive cells showed heterogenous morphology ŽC, E, G, arrowheads.. An alternative slice incubated without a primary antibody did not show any immunoreactivity ŽI.. The results shown are representative of two independent studies. Scale bar s 20 mm.
M. Fujimura et al.r Brain Research 842 (1999) 92–100
Fig. 5. Reverse zymography after transient ischemia. A similar amount of TIMP-1 was evident as 28-kDa characteristic bands in the ischemic and non-ischemic brain at both 3 and 23 h after reperfusion. The results shown were derived from two samples from each time point. c s Contralateral samples; TIMP-1s control TIMP-1 ŽOncogene Research..
further increased in the MCA territory cortex 3 h after reperfusion ŽFig. 3E, arrowheads., while it was barely detected in the caudate putamen ŽFig. 3F. compared with 1 h after reperfusion. Twenty-three hours after reperfusion, MMP-9 was detected in the MCA territory cortex ŽFig. 3G, arrowheads. but not in the caudate putamen ŽFig. 3H.. Immunoreactive cells showed a relatively heterogenous morphology ŽFig. 3, arrowheads.. There was no immunoreactivity in the control specimens, which were treated without a primary antibody ŽFig. 3I.. To further confirm
97
that the bands were from both the proform and activated form of MMP-9, Western blot analysis was performed using a polyclonal antibody against MMP-9 ŽAnawa.. As shown in Fig. 4, MMP-9 immunoreactivity was evident as bands of molecular mass 96 kDa of the pro-MMP-9 in the non-ischemic sample, and was increased in the sample from the ischemic brain 23 h after reperfusion. The activated MMP-9, as characterized by the 88 kDa band, was absent in the control brain, although it appeared 23 h after reperfusion. These data not only confirm the specificity of the antibody for MMP-9 used in this study, but also suggest that MMP-9 increased after transient FCI. 3.3. TIMP-1 expression after ischemiar reperfusion To evaluate the balance between MMP-9 expression and its endogenous inhibitor, the TIMP-1 level was measured using the reverse zymography technique. A characteristic band of 28 kDa molecular weight was evident in the control sample from the contralateral hemisphere, and was not modified in the ischemic sample ŽFig. 5.. This suggests that the increased expression of MMP-9 was not due to the reduction of TIMP-1.
Fig. 6. Evaluation of BBB permeability by the Evans blue extravasation method after transient ischemia. No interstitial expression of Evans blue was detected in the non-ischemic brain ŽA, C. or in the ischemic brain 1 h after reperfusion ŽB., whereas marked extravasation of Evans blue Žred in color. was seen in the ischemic brain 3 h after reperfusion ŽD.. The results shown were derived from two samples from each time point. Scale bar s 100 mm.
98
M. Fujimura et al.r Brain Research 842 (1999) 92–100
3.4. Vasogenic edema after transient MCA occlusion BBB permeability was evaluated in the ischemic hemisphere as well as in the contralateral side 1 and 3 h after 60 min of MCA occlusion. No Evans blue was detected in the non-ischemic hemispheres ŽFig. 6A,C. Žmean O.D.s 0.315 " 0.081; mean " S.D.. and was not increased in the ischemic hemisphere 1 h after ischemia ŽFig. 6B. Žmean O.D.s 0.470 " 0.268.. Four hours after ischemia, the Evans blue signal was significantly increased in the ischemic hemisphere ŽFig. 6D. compared to the non-ischemic hemisphere Žmean O.D.s 4.251 " 0.899, P - 0.0001..
4. Discussion The current study provides evidence that activated MMP-9 appears as early as 3 h after 60 min of transient FCI, and that a significant increase of pro-MMP-9 occurred in a time dependent manner during reperfusion. These conclusions are derived from the following observations. First, a constitutive expression of pro-MMP-9 in the control specimens was evident as a 96-kDa characteristic band by both zymography ŽFig. 1. and Western blot analysis ŽFig. 4., and markedly increased during reperfusion after FCI ŽFig. 1.. Second, activated MMP-9, characterized as an 88 kDa band, was absent in the control specimens, whereas it appeared as early as 3 h and further increased at 23 h after reperfusion ŽFig. 1.. Moreover, the activated nature of this 88 kDa MMP-9 was supported by the presence of a significant increase of type IV collagenase activity in our extracted samples obtained from ischemic brains 3 h after reperfusion ŽFig. 2.. Up-regulation of pro-MMP-9 was first reported by Rosenberg and his colleagues using a permanent FCI model of spontaneous hypertensive rats w24x, in which a marked increase of pro-MMP-9 was shown 12 h after permanent ischemia, but no activated MMP-9 was found after ischemia. In their study, pro-MMP-9 was not detected in the control specimens, while it appeared 12 h after the initiation of ischemia w24x. However, the activated forms of MMP-9 and MMP-2 were not detected w24x. As for transient FCI, a more recent study by Rosenberg et al. w27x successfully showed the significant increase of pro-MMP-9 48 h after 2 h of MCA occlusion; however, the activated form of MMP-9 was not detected. The discrepancy between our results in the current study and these previous studies could be due to the following reasons: Ž1. we employed the protein extraction method with detergent extraction and affinity-support purification, which was recently reported to be capable of recovering gelatinase activity over a 4- to 10-fold range using zymography w36x. The use of this method could contribute to the early detection of activated MMP-9 Ž88
kDa. as well as to the detection of pro-MMP-9 Ž96 kDa. in the control specimens; andror Ž2. since we used the mouse model of 1 h focal ischemia and subsequent reperfusion w33x, a difference in the species used in this study could have affected the results of MMP-9 induction andror activation. Reperfusion injury is thought to play a critical role in cerebral infarction and BBB disruption w34x. Although reperfusion of ischemic tissue with thrombolytic agents after a short period of ischemia is considered to reduce infarct volume w1x, reperfusion at a later period exacerbates ischemic damage. In the present study, we used the transient FCI model of 60 min of MCA occlusion and subsequent reperfusion, which consistently produced infarction and brain edema in the MCA territory area on the ischemic side w18,33x. The present study demonstrated the evidence of BBB disruption by the Evans blue extravasation method 3 h but not 1 h after reperfusion ŽFig. 6D., as confirmed by the semi-quantitative analysis of Evans blue extravasation using the densitometer. These data are consistent with a previous report showing the increase of brain sucrose uptake 4 h after 2 h of MCA occlusion w27x. At the same time point as Evans blue leakage, we detected the appearance of activated MMP-9 ŽFig. 1., which may suggest the possible involvement of MMP-9 activation in early BBB dysfunction. In fact, it was reported that the MMP inhibitor BB-1101 reduced the BBB opening 3 h after transient FCI w27x, which may also support the idea that activation of MMP-9 contributes to BBB dysfunction at the early stage of transient FCI. Reactive oxygen species ŽROS. are produced during reperfusion, and have been implicated in BBB disruption and the expansion of infarction after FCI. Based on this, we have reported that antioxidant enzymes, such as superoxide dismutase ŽSOD.-1, play a protective role in transient focal ischemia w6,7,16,18,33x, suggesting that free radical production during reperfusion exacerbates cerebral infarction. SOD-1 transgenic mice show a significant reduction in infarct volume after transient focal ischemia w16,33x, whereas infarction is increased in SOD-1 knockouts after transient ischemia. Although the molecular cascade between free radical production and the exacerbation of ischemiarreperfusion injury is still controversial, we may find some clue in recent in vitro studies. It was reported that inflammatory molecules such as intercellular adhesion molecule-1 ŽICAM-1. w2x, and MMPs w22,29x could be induced in cultured endothelial cells after hypoxiarreoxygenation. Also, in vivo studies have shown that ICAM-1 is markedly induced after ischemiarreperfusion, and that ICAM-1 knockouts show a smaller infarction after transient ischemia w10,30x, which suggests the deleterious role of this molecule in ischemiarreperfusion injury. In fact, antibodies against ICAM-1 reduced infarction after transient FCI w8x, but not after permanent FCI w35x. In the present study, no activated MMP-9 was detected in the control specimens, while it did appear at 3 h
M. Fujimura et al.r Brain Research 842 (1999) 92–100
and was further increased 23 h after 60 min of transient FCI ŽFig. 1.. Furthermore, a time dependent increase of pro-MMP-9 was seen during reperfusion after ischemia. Although further examination is necessary to elucidate the direct relationship between free radical production during reperfusion and MMP-9 induction, our data suggest that reperfusion after FCI may contribute to the activation of MMP-9 as well as the increase of pro-MMP-9. The exact mechanisms involved in the induction and activation of MMPs after ischemiarreperfusion are not well defined. Since MMP-9 has an activator protein-1 ŽAP-1. binding site in its promoter region, it is conceivable that the induction of pro-MMP-9 is associated with the alteration of AP-1 binding activity during reperfusion. In fact, a significant increase in AP-1 binding activity was detected after 1 h of reperfusion following 90 min of MCA occlusion in rats w28x, which is compatible with our results showing pro-MMP-9 induction 3 h after reperfusion. It is also conceivable that free radical production during reperfusion affects the activation of MMP-9, since it has been suggested that ROS regulate the activity of MMPs in vitro w22,29x. Based on this, we are currently undertaking experiments to elucidate whether the induction and activation of MMP-9 is affected in transgenic mice that overexpresss the anti-oxidant enzyme, SOD-1. In the present study, different patterns of MMP-9 expression were seen between each time point by immunohistochemistry ŽFig. 3.. One hour after reperfusion, an increase in MMP-9 expression was seen in the entire area of the ischemic brain ŽFig. 3C,D., while at later time points such as 3 and 23 h, more MMP-9 expression was detected in the MCA territory cortex ŽFig. 3E,G. compared with the ischemic core of the caudate putamen ŽFig. 3F,H.. It is unclear why there was a regional predominance in the cortex rather than in the ischemic core of the caudate putamen at the later time points. Since the core of the caudate putamen is assumed to be undergoing more severe ischemia than the MCA territory cortex that has collateral circulation such as leptomeningeal anastomosis, it is conceivable that severe ischemia in the ischemic core during MCA occlusion may contribute to less MMP-9 expression in the caudate putamen at the later time points. As for MMP-9 cell populations, they could be induced in a variety of cells in the central nervous system, including endothelial cells w19x, astrocytes w12x, oligodendrocytes w31x, microglia w13x, and neurons w3x. Also in the model of permanent FCI in rats, early MMP-9 expression on endothelial cells as well as on neutrophils was reported, while there was no alteration of TIMP-1 after ischemia w23x. In the present study, immunohistochemistry revealed the heterogeneous morphology of the immunopositive cells in the area of the MCA territory on the ischemic side ŽFig. 3.. They are likely to involve neuronal ŽFig. 3E, arrowheads. and endothelial cell ŽFig. 3G. populations. A definition of the cell populations of these immunopositive cells remains to be elucidated in a future study. Double staining
99
of MMP-9 and various cell markers would be useful to clarify this issue. In conclusion, we have shown that MMP-9 was activated as early as 3 h after transient FCI, and that proMMP-9 also increased in a time dependent manner after transient ischemia. Although further examination is necessary to elucidate the relationship between the activation of MMP-9 and ischemiarreperfusion injury, our results suggest the possibility that early activation of MMP-9 may contribute to BBB dysfunction and subsequent formation of vasogenic edema after transient FCI. It is suggested that pharmacological or therapeutic intervention should be considered in targeting the activation andror induction of MMP-9 in FCI.
Acknowledgements The authors are grateful to Ms. Cheryl Christensen for her editorial assistance and Ms. Liza Reola, Mr. Bernard Calagui and Ms. Jane O. Kim for their technical assistance. This study was supported by National Institutes of Health grants NS25372, NS14543, NS36147, NS38653 and NO1 NS82386. P.H.C. is a recipient of the Jacob Javits Neuroscience Investigator Award.
References w1x The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group, Tissue plasminogen activator for acute ischemic stroke, N. Engl. J. Med. 333 Ž1995. 1581–1587. w2x T. Aoki, Y. Suzuki, K. Nishio, K. Suzuki, A. Miyata, Y. Oyamada, M. Mori, H. Fujita, K. Yamaguchi, Effect of antioxidants on hyperoxia-induced ICAM-1 expression in human endothelial cells, Adv. Exp. Med. Biol. 411 Ž1997. 503–511. w3x J.R. Backstrom, G.P. Lim, M.J. Cullen, Z.A. Tokes, Matrix metalloproteinase-9 ŽMMP-9. is synthesized in neurons of the human hippocampus and is capable of degrading the amyloid-beta peptide Ž1–40., J. Neurosci. 16 Ž1996. 7910–7919. w4x L. Belayev, R. Busto, W. Zhao, M.D. Ginsberg, Quantitative evaluation of blood–brain barrier permeability following middle cerebral artery occlusion in rats, Brain Res. 739 Ž1996. 88–96. w5x G. Bruno, R. Todor, I. Lewis, D. Chyatte, Vascular extracellular matrix remodeling in cerebral aneurysms, J. Neurosurg. 89 Ž1998. 431–440. w6x P.H. Chan, C.J. Epstein, Y. Li, T.T. Huang, E. Carlson, H. Kinouchi, G. Yang, H. Kamii, S. Mikawa, T. Kondo et al., Transgenic mice and knockout mutants in the study of oxidative stress in brain injury, J. Neurotrauma 12 Ž1995. 815–824. w7x P.H. Chan, Role of oxidants in ischemic brain damage, Stroke 27 Ž1996. 1124–1129. w8x M. Chopp, Y. Li, N. Jiang, R.L. Zhang, J. Prostak, Antibodies against adhesion molecules reduce apoptosis after transient middle cerebral artery occlusion in rat brain, J. Cereb. Blood Flow Metab. 16 Ž1996. 578–584. w9x A.W. Clark, C.A. Krekoski, S.S. Bou, K.R. Chapman, D.R. Edwards, Increased gelatinase A ŽMMP-2. and gelatinase B ŽMMP-9. activities in human brain after focal ischemia, Neurosci. Lett. 238 Ž1997. 53–56.
100
M. Fujimura et al.r Brain Research 842 (1999) 92–100
w10x E.S. Connolly Jr., C.J. Winfree, T.A. Springer, Y. Naka, H. Liao, S.D. Yan, D.M. Stern, R.A. Solomon, J.C. Gutierrez-Ramos, D.J. Pinsky, Cerebral protection in homozygous null ICAM-1 mice after middle cerebral artery occlusion. Role of neutrophil adhesion in the pathogenesis of stroke, J. Clin. Invest. 97 Ž1996. 209–216. w11x M. Fujimura, Y. Morita-Fujimura, K. Murakami, M. Kawase, P.H. Chan, Cytosolic redistribution of cytochrome c after transient focal cerebral ischemia in rats, J. Cereb. Blood Flow Metab. 18 Ž1998. 1239–1247. w12x P.E. Gottschall, X. Yu, Cytokines regulate gelatinase A and B Žmatrix metalloproteinase 2 and 9. activity in cultured rat astrocytes, J. Neurochem. 64 Ž1995. 1513–1520. w13x P.E. Gottschall, X. Yu, B. Bing, Increased production of gelatinase B Žmatrix metalloproteinase-9. and interleukin-6 by activated rat microglia in culture, J. Neurosci. Res. 42 Ž1995. 335–342. w14x G.F. Hamann, Y. Okada, R. Fitridge, G.J. del Zoppo, Microvascular basal lamina antigens disappear during cerebral ischemia and reperfusion, Stroke 26 Ž1995. 2120–2126. w15x S.C. Kim, M. Singh, J. Huang, C.J. Prestigiacomo, C.J. Winfree, R.A. Solomon, E.S. Connolly Jr., Matrix metalloproteinase-9 in cerebral aneurysms, Neurosurgery 41 Ž1997. 642–666. w16x H. Kinouchi, C.J. Epstein, T. Mizui, E. Carlson, S.F. Chen, P.H. Chan, Attenuation of focal cerebral ischemic injury in transgenic mice overexpressing CuZn superoxide dismutase, Proc. Natl. Acad. Sci. U.S.A. 88 Ž1991. 11158–11162. w17x D.E. Kleiner, W.G. Stetler-Stevenson, Quantitative zymography: detection of picogram quantities of gelatinases, Anal. Biochem. 218 Ž1994. 325–329. w18x T. Kondo, A.G. Reaume, T.T. Huang, E. Carlson, K. Murakami, S.F. Chen, E.K. Hoffman, R.W. Scott, C.J. Epstein, P.H. Chan, Reduction of CuZn-superoxide dismutase activity exacerbates neuronal cell injury and edema formation after transient focal cerebral ischemia, J. Neurosci. 17 Ž1997. 4180–4189. w19x M. Nguyen, J. Arkell, C.J. Jackson, Active and tissue inhibitor of matrix metalloproteinase-free gelatinase B accumulates within human microvascular endothelial vesicles, J. Biol. Chem. 273 Ž1998. 5400–5404. w20x G.W. Oliver, J.D. Leferson, W.G. Stetler-Stevenson, D.E. Kleiner, Quantitative reverse zymography: analysis of picogram amounts of metalloproteinase inhibitors using gelatinase A and B reverse zymograms, Anal. Biochem. 244 Ž1997. 161–166. w21x A.R. Quesada, M.M. Barbacid, E. Mira, P. Fernandez-Resa, G. Marquez, M. Aracil, Evaluation of fluorometric and zymographic methods as activity assays for stromelysins and gelatinases, Clin. Exp. Metastasis 15 Ž1997. 26–32. w22x S. Rajagopalan, X.P. Meng, S. Ramasamy, D.G. Harrison, Z.S. Galis, Reactive oxygen species produced by macrophage-derived foam cells regulate the activity of vascular matrix metalloproteinases in vitro. Implications for atherosclerotic plaque stability, J. Clin. Invest. 98 Ž1996. 2572–2579.
w23x A.M. Romanic, R.F. White, A.J. Arleth, E.H. Ohlstein, F.C. Barone, Matrix metalloproteinase expression increases after cerebral focal ischemia in rats: inhibition of matrix metalloproteinase-9 reduces infarct size, Stroke 29 Ž1998. 1020–1030. w24x G.A. Rosenberg, M. Navratil, F. Barone, G. Feuerstein, Proteolytic cascade enzymes increase in focal cerebral ischemia in rat, J. Cereb. Blood Flow Metab. 16 Ž1996. 360–366. w25x G.A. Rosenberg, J.E. Dencoff, N. Correa Jr., M. Reiners, C.C. Ford, Effect of steroids on CSF matrix metalloproteinases in multiple sclerosis: relation to blood–brain barrier injury, Neurology 46 Ž1996. 1626–1632. w26x G.A. Rosenberg, M. Navratil, Metalloproteinase inhibition blocks edema in intracerebral hemorrhage in the rat, Neurology 48 Ž1997. 921–926. w27x G.A. Rosenberg, E.Y. Estrada, J.E. Dencoff, Matrix metalloproteinases and TIMPs are associated with blood–brain-barrier opening after reperfusion in rat brain, Stroke 29 Ž1998. 2189–2195. w28x A. Salminen, P.K. Liu, C.Y. Hsu, Alteration of transcription factor binding activities in the ischemic rat brain, Biochem. Biophys. Res. Commun. 212 Ž1995. 939–944. w29x M. Shingu, T. Isayama, C. Yasutake, T. Naono, M. Nobunaga, K. Tomari, K. Horie, Y. Goto, Role of oxygen radicals and IL-6 in IL-1-dependent cartilage matrix degradation, Inflammation 18 Ž1994. 613–623. w30x S.G. Soriano, S.A. Lipton, Y.F. Wang, M. Xiao, T.A. Springer, J.C. Gutierrez-Ramos, P.R. Hickey, Intercellular adhesion molecule-1deficient mice are less susceptible to cerebral ischemia-reperfusion injury, Ann. Neurol. 39 Ž1996. 618–624. w31x J.H. Uhm, N.P. Dooley, L.Y. Oh, V.W. Yong, Oligodendrocytes utilize a matrix metalloproteinase, MMP-9, to extend processes along an astrocyte extracellular matrix, Glia 22 Ž1998. 53–63. w32x H.E. Van Wart, H. Birkedal-Hansen, The cysteine switch: a principle of regulation of metalloproteinase activity with potential applicability to the entire matrix metalloproteinase gene family, Proc. Natl. Acad. Sci. U.S.A. 87 Ž1990. 5578–5582. w33x G. Yang, P.H. Chan, J. Chen, E. Carlson, S.F. Chen, P. Weinstein, C.J. Epstein, H. Kamii, Human copper–zinc superoxide dismutase transgenic mice are highly resistant to reperfusion injury after focal cerebral ischemia, Stroke 25 Ž1994. 165–170. w34x G.Y. Yang, A.L. Betz, Reperfusion-induced injury to the blood–brain barrier after middle cerebral artery occlusion in rats, Stroke 25 Ž1994. 1658–1664, Discussion 1664-1655. w35x R.L. Zhang, M. Chopp, N. Jiang, W.X. Tang, J. Prostak, A.M. Manning, D.C. Anderson, Anti-intercellular adhesion molecule-1 antibody reduces ischemic cell damage after transient but not permanent middle cerebral artery occlusion in the Wistar rat, Stroke 26 Ž1995. 1438–1442. w36x J.W. Zhang, P.E. Gottschall, Zymographic measurement of gelatinase activity in brain tissue after detergent extraction and affinitysupport purification, J. Neurosci. Methods 76 Ž1997. 15–20.