Attenuation of temporary focal cerebral ischemic injury in the mouse following transfection with interleukin-1 receptor antagonist

Molecular Brain Research 72 Ž1999. 129–137 www.elsevier.comrlocaterbres

Research report

Attenuation of temporary focal cerebral ischemic injury in the mouse following transfection with interleukin-1 receptor antagonist Guo-Yuan Yang a

a,)

, Ying Mao a,c , Liang-Fu Zhou c , Wen Ye a , Xiao-Hong Liu a , Chao Gong a , A. Lorris Betz, a,b,d

Department of Surgery (Neurosurgery), School of Medicine, UniÕersity of Michigan, 5550 Kresge I r 0532, 1500 East Medical Center Dr., Ann Arbor, MI 48109-0532, USA b Department of Pediatrics, UniÕersity of Michigan, Ann Arbor, MI 48109, USA c Department of Neurosurgery, Hua Shan Hospital, Shanghai Medical UniÕersity, Shanghai 200040, China d Department of Neurology, UniÕersity of Michigan, Ann Arbor, MI 48109, USA Accepted 15 June 1999

Abstract The proinflammatory cytokine interleukin-1 beta ŽIL-1b . is thought to play an important role in the stimulation of the inflammatory response following ischemia and reperfusion. This study investigated the inflammatory effect of IL-1b during transient focal cerebral ischemia and reperfusion in the mouse transduced with the interleukin-1 receptor antagonist ŽIL-1ra. gene. An adenoviral vector encoding, either the human IL-1ra gene ŽAdRSVIL-1ra. or the LacZ gene ŽAdRSVlacZ. or normal saline, were injected into the right lateral ventricles of adult CD-1 mice Ž n s 96.. Five days later, the mice received 1 h temporary middle cerebral artery occlusion ŽtMACAO. followed by 23 h reperfusion. Cerebral blood flow ŽCBF., infarct volume, blood–brain barrier ŽBBB. permeability, and the number of intracellular adhesion molecule-1 positive vessels were measured to determine the effect of IL-1b during postischemic reperfusion. Infarct volume in the AdRSVIL-1ra-transduced mice was markedly reduced compared to the AdRSVlacZ-transduced and saline-injected mice Ž36.0 " 5.3 mm3 vs. 60.0 " 6.2 mm3, 69.5 " 6.3 mm3, after 23 h of reperfusion, n s 6–8 per group, p - 0.05.. BBB disruption and intracellular adhesion molecule-1 expression Ž135 " 23 vs. 311 " 40 and 357 " 51, n s 6–8 per group, p - 0.05. in the AdRSVIL-1ra-transduced mice were also less than that of the AdRSVlacZ-transduced and saline-injected mice. Our studies demonstrated that overexpression of IL-1ra in the mouse brain can downregulate intracellular adhesion molecule-1 expression both in the cortex and basal ganglia, which suggests that IL-1b may play an important role in the activation of the inflammatory response during focal cerebral ischemia by promoting leukocyte adhesion to endothelial cells. The decrease of BBB disruption in AdRSVIL-1ra-transduced mice suggests that the endothelial cells may be a target for IL-1b during postischemic reperfusion. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Adenovirus; Blood–brain barrier; Ischemia; Mouse; Intracellular adhesion molecule-1; Interleukin-1; Temporary

1. Introduction Postischemic reperfusion counteracts the ischemic injury process and serves to return at least some of the reversibly injured tissue to a functional state w17x. However, there is a developing consensus that return of blood flow in the postischemic period has potentially deleterious effects w19x. The mechanism of reperfusion injury is diffe-

) Corresponding [email protected]

author.

Fax:

q 1-734-763-7322;

E-mail:

rent from the mechanism of ischemic injury w34x. Interactions between blood components and the damaged tissue can lead to further injury. Reperfusion produces delayed postischemic hypoperfusion w20x, cell swelling w6x, free radical damage w31x and apoptosis w5x. The release of cytokines could aggravate brain injury that follows postischemic reperfusion. Cytokines, for example, interleukin-1b ŽIL-1b . and tumor necrosis factor-alpha ŽTNFa ., may stimulate vascular endothelium, which upregulate intercellular adhesion molecule-1 ŽICAM-1. and endothelial leukocyte adhesion molecule-1 ŽELAM-1. expression, and lead to neutrophil adhesion w40x. The neutrophil accumula-

0169-328Xr99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 9 - 3 2 8 X Ž 9 9 . 0 0 2 0 5 - 3

130

G.-Y. Yang et al.r Molecular Brain Research 72 (1999) 129–137

tion could plug capillaries and aggravate reperfusion injury w30x. Although the injuries which result from ischemia and reperfusion seem similar, the mechanisms of action of these mediators Žincluding cytokines and adhesion molecules. may be different. IL-1b is a potent mediator of inflammation in most tissues. Numerous studies have demonstrated that IL-1b mRNA and protein are increased during cerebral ischemia. IL-1b may initiate a cascade of inflammatory responses after cerebral ischemia and cause brain cell death w2x. Ischemic brain injury is greatly attenuated in IL-1b converting enzyme ŽICE. knock out mice w18,32x as well as IL-1 receptor antagonist ŽIL-1ra.-treated animals w15,28x. Expression of foreign genes in the brain is now feasible, giving us a useful tool to study the pathogenesis of cerebral ischemia in vivo. We have demonstrated that an adenovirus carrying the IL-1ra gene ŽAdRSVIL-1ra.injected intraventricularly leads to the overexpression of IL-1ra protein in the rodent brain for several days w3,44x. Our previous studies established that overexpression of IL-1ra can reduce infarction volume following permanent middle cerebral artery occlusion ŽpMCAO. in rats and mice. Later, we demonstrated that overexpression of IL-1ra in the brain can attenuate edema formation and inflammatory responses in the pMCAO mouse model w44x. However, there is little known of the effect of IL-1 in reperfusion brain injury although IL-1 mRNA expression was increased following transient brain ischemia in the rat w12x. In this study using AdRSVIL-1 gene-transduced mice, we examined CBF, infarct volume, blood–brain barrier ŽBBB. permeability, and ICAM-1 following 1 h temporary middle cerebral artery occlusion ŽtMCAO. and 23 h reperfusion. The specific aims are to determine: Ž1. if overexpression of IL-1ra in mice will decrease reperfusion brain injury; and Ž2. the possible roles that IL-1 may play during tMCAO.

2. Materials and methods 2.1. Experimental group Procedures using laboratory animals were approved by the institutional animal care and use committee. Adult male CD-1 mice ŽCharles River, n s 96. weighing 30–35 g were divided into three groups: IL-1ra group Ž n s 32. in which adenoviral vector encoding the human IL-1ra cDNA was injected into the right lateral ventricle; lacZ group Ž n s 32. in which adenoviral vector encoding the LacZ gene was injected in the same way and this group was used as a control group Žspecific gene. to exclude the effect of IL-1ra during the study; the saline control group Ž n s 32. was injected with the same amount of 0.9% normal saline and this group was used as a control group Žvector. to exclude the influence of the adenoviral vector.

2.2. AdenoÕiral gene transfer in mouse brain The construction of replication-deficient human adenovirus serotype 5-derived adenoviral vectors is described elsewhere w29x. Two different recombinant virus vectors, one containing the human IL-1ra cDNA and the other containing the Escherichia coli lacZ gene, were used. These vectors are designated as AdRSVIL-1ra and AdRSVlacZ, respectively. Rous sarcoma virus ŽRSV. promoter was used to drive gene transcription in these two groups. The virus vectors are identical except for the particular cDNA. The mice were anesthetized with 4% chloral hydrate Ž400 mgrkg, i.p.. and placed in a stereotactic frame ŽKopf Model 921, David Kopf Instruments, Tujunga, CA.. A burr hole was drilled in the pericranium 1 mm lateral to the sagittal suture and 1 mm posterior to the coronal suture. A 33-gauge needle attached to a 10 ml Hamilton syringe was stereotactically inserted into the right lateral ventricle 2.5 mm under the surface of cortex. One ml of adenoviral suspension containing 1 = 10 12 particlesrml was injected at a rate of 0.2 mlrmin. The needle was then withdrawn over 5 min. The hole was sealed with bone wax, the wound was closed with suture, and the animals were allowed to recover. 2.3. MCAO and reperfusion Five days later, AdRSVIL-1ra, AdRSVlacZ-transduced, and saline-injected mice were reanesthetized with 4% isoflurane in 70% N2 Or30%O2 and maintained at 1.5% isoflurane using a face mask. The body temperature was maintained at 37.0 " 0.58C with a heated blanket using a feedback control system Ž73A, Yellow Springs, OH.. The left femoral artery was cannulated with PE-10 tubing to monitor the changes of blood pressure during the MCAO. Blood gases were evaluated before and after tMCAO. The tMCAO method was described in our previous studies w42x. Briefly, the left common carotid artery was exposed via a ventral midline incision. The internal carotid artery was isolated and its branch, the pterygopalatine artery, was ligated close to its origin. A 2 cm length of 5–0 rounded tip nylon suture was advanced from the external carotid artery through the common carotid artery and then up to the internal carotid artery for a distance of 11.0 " 0.5 mm. Restoration of MCA blood flow in mice subjected to 1 h of temporary occlusion was achieved by withdrawing the suture to the stump of external carotid artery. After closing the skin incision, mice were allowed to recover from anesthesia, returned to their cages, and allowed to move and eat freely. 2.4. Laser-doppler flowmeter We used a Laser-Doppler Flowmetry ŽLDF. monitor ŽModel BPM 2 System, Vasamedics, St. Paul, MN.

G.-Y. Yang et al.r Molecular Brain Research 72 (1999) 129–137 Table 1 Physiological parameters before and after tMCAO in mice Measurements were performed 15 min before and 15 min after tMCAO. Values are mean"S.D. for 8–10 mice in each group. IL-1ra: IL-1ra gene-transduced mice, LacZ: LacZ gene-transduced mice, Saline: salineinjected mice intraventricularly. MABP ŽmmHg.

Blood pH

PaO 2 ŽmmHg.

PaCO 2 ŽmmHg.

IL-1ra Before ischemia After ischemia

103"2 104"1

7.42"0.04 7.33"0.03

159"4 154"11

35"3 40"3

LacZ Before ischemia After ischemia

110"2 113"3

7.38"0.03 7.34"0.02

163"6 151"7

39"2 40"2

Saline Before ischemia After ischemia

102"2 104"2

7.39"0.02 7.36"0.04

164"10 159"8

34"3 39"2

equipped with a small caliber probe of 0.7 mm diameter ŽP-433, Vasamedics. to measure surface cerebral blood flow ŽCBF.. Three points were measured: ŽA. 3.5 mm lateral to the sagittal suture and 1 mm posterior to the coronal suture on the left side corresponding to the ischemic core region; ŽB. the same point on the opposite side served as a control region; and ŽC. 1 mm lateral to the sagittal suture on the left side and 1 mm posterior to the coronal suture corresponding to the perifocal region. The laser-Doppler probe was advanced steadily in a perpendicular plane to rest on the skull surface. Our preliminary study demonstrated that the changes of blood flow measured through the cranial bone in mice were not significantly different from flow measured with a dural contact

131

probe Ž100 " 5% vs. 102 " 6%, n s 6, p ) 0.05.. Baseline LDF recordings of these three regions were made prior to MCAO, 5 min after MCAO, and 5 min after reperfusion. CBF values were calculated and expressed as a percentage of baseline values w42x. 2.5. Measurement of infarct Õolume Following 1 h of tMCAO and 23 h or 7 days reperfusion, the animals were killed and the brains were removed and frozen immediately on dry ice. Six coronal sections Ž20 mm in thickness. from 1 to 7 mm from the frontal pole were cut and mounted on slides. Hematoxylin and eosin ŽH and E. staining was used to identify the infarct area. Using an NIH 1.61 image analysis system, infarction volume was calculated by summing the infarction areas multiplied by the distance between the sections. The influence of edema on the infarction volume was corrected according to the method described by Swanson et al. w36x. 2.6. Immunostaining of extraÕasated endogenous albumin AdRSVIL-1ra, AdRSVlacZ-transduced, and saline control mice were studied for BBB integrity. The brains were removed and frozen following 1 h of tMCAO and 23 h or 7 days reperfusion. Ten-mm-thick sections were cut using a cryostat ŽCM1800, Leica, Germany. and mounted on precleaned Probe-On-Slides ŽFisher Scientific, Pittsburgh, PA.. After drying, the sections were fixed with Carnoy fixative solution Ž60% methanol, 30% chloroform, and 10% acetate. for 20 min, and rinsed three times with 100 mmol phosphate buffered saline ŽPBS, Sigma, 2.6 mM

Table 2 The changes of CBF during 1 h occlusion and 23 h reperfusion Measurements were performed 15 min before, 15 min after MCAO, 15 min and 23 h following reperfusion. Values are mean " S.E.; N s 8–10 in each group. AdRSVIL-1ra: IL-1ra gene-transduced mice, AdRSVlacZ: LacZ gene-transduced mice, saline: saline-injected mice intraventricularly. AdRSVIL-1ra Žpercentage of baseline.

AdRSVLacZ Žpercentage of baseline.

Saline Žpercentage of baseline.

Ischemic core Baseline Occlusion Ž15 min. Reperfusion Ž15 min. Reperfusion Ž24 h.

100 " 0 12 " 1 76 " 5 65 " 4

100 " 0 15 " 1 70 " 3 53 " 4

100 " 0 13 " 2 66 " 3 55 " 3

Perifocal Baseline Occlusion Ž15 min. Reperfusion Ž15 min. Reperfusion Ž24 h.

100 " 0 53 " 6 76 " 2 81 " 12

100 " 0 54 " 4 61 " 5 57 " 6

100 " 0 48 " 3 78 " 7 59 " 3

Contralateral Baseline Occlusion Ž15 min. Reperfusion Ž15 min. Reperfusion Ž24 h.

100 " 0 100 " 3 101 " 6 84 " 13

100 " 0 101 " 1 90 " 7 67 " 6

100 " 0 99 " 2 109 " 7 73 " 5

132

G.-Y. Yang et al.r Molecular Brain Research 72 (1999) 129–137

coronal sections measured at q1.34, q0.86, q0.38, y0.10, y0.58, y1.06 related to Bregma in each mouse were selected to calculate the albumin immunostaining area w13x. 2.7. ICAM-1 immunohistochemistry

Fig. 1. Bar graph shows infarct volume in the AdRSVIL-1ra, AdRSVlacZ-transduced, and saline-injected mice following 1 h tMCAO and 23 h or 7 days of reperfusion. Five days prior to MCAO, the mice received an intraventricular injection containing 1=10 9 particles of AdRSVIL-1ra or AdRSVlacZ, or same volume of saline. Injured tissue was identified by H&E staining. Data are mean"S.E.; ns6–8 in each group. U p - 0.05, the AdRSVIL-1ra-transduced mice vs. the AdRSVlacZ-transduced and the saline-injected mice.

KCl, 1.4 mM KH 2 PO4 , 136 mM NaCl, 8 mM Na 2 HPO4 , pH 7.4.. Nonspecific binding was blocked using 10% normal goat serum for 30 min at room temperature. The sections were incubated in a 1:400 dilution of rabbit anti-mouse albumin primary antibody ŽCappel, Downington, PA. overnight at 48C. After treatment with 1% H 2 O 2 in 30%r70% methanolrPBS solution, the sections were incubated with biotinylated goat anti-rabbit secondary antibody ŽVector Lab., San Francisco, CA. for 90 min at room temperature followed by an ABC process ŽABC-Elite Kit, Vector Lab... Finally, the sections were treated with stable DAB ŽResearch Genetics, Huntsville, AL. as a peroxidase substrate. Negative controls used the same concentration of normal rabbit IgG instead of primary antibody. BBB disruption was detected by dark brown staining indicating albumin leakage into tissue. Immunostained sections were analyzed with an NIH 1.61 image analysis system. Six

After decapitation, the brains were removed and frozen immediately in 2-methyl-butane solution ŽMallinckroft. at y408C. Coronal cryostat sections Ž20 mm in thickness. located at Bregma q0.38 mm were cut and mounted on precleaned Probe-on-Slides. After drying, the sections were fixed with Cornoy fixative solution for 20 min. Nonspecific binding was blocked using 10% normal rabbit serum for 30 min at room temperature. The sections were incubated in a 1:300 dilution of rat monoclonal anti-mouse CD54 primary antibody ŽCaltag, San Francisco, CA. overnight at 48C. After treatment with 0.3% H 2 O 2 in 30%r70% methanolrPBS solution, the sections were incubated with biotinylated rabbit anti-rat secondary antibody ŽVector Lab.. at a 1:200 dilution for 90 min at room temperature followed by an ABC process. Finally, the sections were treated with stable DAB and counter-stained with Hematoxylin. The sections were mounted in buffered glycerin. A negative control was carried out using the same concentration of normal rat IgG instead of primary antibody. The total number of deep brown-stained positive vessels were counted in one cross-section Žq0.38., including the cortex and basal ganglia region. ICAM-1 vessel counting was done manually to exclude the infiltrating cells occasionally expressing ICAM-1. 2.8. Statistical analysis All data are expressed as mean " S.E. Parametric data among the AdRSVIL-1ra, AdRSVlacZ-transduced, and saline control groups were compared using ANOVA and

Fig. 2. Graph showing albumin positive immunostaining area expressed as a % of the ipsilateral hemisphere in the AdRSVIL-1ra, AdRSVlacZ-transduced, and saline-injected mice. The mice received 1 h tMCAO and 23 h ŽA. or 7 days ŽB. of reperfusion. The brain sections were chosen at q1.34, q0.86, q0.38, y0.10, y0.58, y1.06 mm related to the Bregma. The albumin positive stained areas were smaller in the AdRSVIL-1ra-transduced mice. Values are mean " S.E.; n s 6–8 in each group; U p - 0.05, the AdRSVIL-1ra-transduced mice vs. the AdRSVlacZ-transduced and the saline-injected mice.

G.-Y. Yang et al.r Molecular Brain Research 72 (1999) 129–137

two-tailed Student’s t-test with Bonferroni adjustment. A probability value of less than 5% was considered statistically significant.

133

mm3 and 69.5 " 6.3 mm3, respectively; p - 0.05; Fig. 1.. The infarct volume measurements 7 days following tMCAO in these three groups were parallel to the 1 day post-tMCAO group; there were no further reductions observed after longer reperfusion times.

3. Results 3.4. Immunostaining of extraÕasated endogenous albumin 3.1. Physiological parameters Introduction of adenoviral vector or normal saline into the lateral ventricle did not influence the survival rate of mice in any of the three groups. No signs of general toxicity or neurological deficit were found, and there was no surgical mortality. Five days later, three groups of mice underwent 1 h tMCAO and 23 h of reperfusion. The results showed that the blood pressure was stable during the surgical procedure and there was no significant difference in blood pressure among the three groups of mice. There were also no significant differences in pO 2 , pCO 2 , and the saturation of oxygen among the three groups before and after ischemia and reperfusion ŽTable 1..

BBB permeability was measured as the area of albumin positive staining. Albumin staining was negative in the contralateral hemisphere in these three tMCAO groups of

3.2. CBF changes Surface CBF was measured by LDF: CBF in the ischemic core region dropped sharply to 12 " 1%, 15 " 1%, and 13 " 2% of the baseline in AdRSVIL-1ra, AdRSVlacZ-transduced, and saline-injected mice, respectively ŽTable 2.. There were no significant differences among the three groups. The CBF in the contralateral hemisphere in all three groups remained normal during the MCAO which demonstrates the occurrence of focal cerebral ischemia, not global cerebral ischemia. CBF in the ischemic core region following reperfusion returned to 76 " 5%, 70 " 4%, and 66 " 8% of the baseline in the AdRSVIL-1ra, AdRSVlacZ-transduced, and saline-injected mice, respectively. After 23 h of reperfusion, the CBF remained higher than 50% of baseline flow both in the core and perifocal regions in all three groups. Surface CBF changes were not significantly different during ischemia and following reperfusion, which suggests adenoviral gene transfer did not affect the blood flow, and the smaller brain lesion in the AdRSVIL-1ra-transduced mice were not due to incomplete occlusion of the middle cerebral artery or greater collateral flow capacity. 3.3. Infarct Õolume Brain infarct volume was determined in AdRSVIL-1ra, AdRSVlacZ-transduced, and saline-injected mice after 1 h tMCAO and 23 h reperfusion: the total volume of ischemic brain injury in the AdRSVIL-1ra-transduced mice was 40–48% smaller than that of the AdRSVlacZ-transduced and saline-injected mice Ž36.0 " 5.3 mm3 vs. 60.0 " 6.2

Fig. 3. Photographs represent the distribution of extravasated endogenous albumin Ždark area. in the AdRSVIL-1ra, the AdRSVlacZ-transduced, and the saline-injected mice following 1 h tMCAO and 23 h reperfusion. Dark brown staining indicates damaged areas immunostained with monoclonal anti-albumin antibody. Photographs show the distribution of extravasated endogenous albumin in the ischemic region from ŽA. AdRSVIL-1ra-transduced; ŽB. AdRSVlacZ-transduced; and ŽC. saline-injected mice. These photographs demonstrate that BBB disruption is significantly smaller in the AdRSVIL-1ra-transduced mice than in the AdRSVlacZ-transduced and saline-injected mice.

134

G.-Y. Yang et al.r Molecular Brain Research 72 (1999) 129–137

mice. The area of albumin positive staining in the AdRSVIL-1ra-transduced mice was much smaller compared to the AdRSVlacZ-transduced and saline-treated mice following 23 h reperfusion ŽFig. 2A. and even after 7 days reperfusion ŽFig. 2B.. Fig. 3 shows the distribution of extravasated endogenous albumin in the AdRSVIL-1ra, AdRSVlacZ-transduced, and saline-treated mice following 1 h occlusion and 23 h reperfusion. 3.5. Immunostaining of ICAM-1 There were few ICAM-1 positive vessels in the contralateral hemisphere among the AdRSVIL-1ra, AdRSVlacZ-transduced, and saline-injected groups Žless than 10 positive vessels per section. and there were no statistical differences among the three groups Ž p ) 0.05.. ICAM-1 positive vessels in the ipsilateral cortex of AdRSVIL-1ra-transduced mice were fewer than those in the other two groups following tMCAO Ž135 " 23 vs. 311 " 40 and 357 " 51, respectively; p - 0.05; Fig. 4.. Similarly, there was a significant difference in ICAM-1 positive vessels in ischemic basal ganglia Ž90 " 21, vs. 207 " 43 and 259 " 43, p - 0.05.. Compared to the AdRSVlacZ-transduced and saline-injected mice, ICAM-1 expression in the AdRSVIL-1ra-transduced mice was significantly decreased, especially in the border zone between the infarct core and perifocal regions ŽFig. 5.. It has been found ICAM-1 positive vessels were expressed in the ischemic core at the early time point of tMCAO, but moved to the perifocal region at later time points Ž23 h following reperfusion. since a large amount of brain cells

Fig. 5. Photographs represent ICAM-1 immunohistochemical staining in the AdRSVIL-1ra ŽA., AdRSVlacZ-transduced ŽB., and saline-treated ŽC. mice following 1 h tMCAO and 23 h reperfusion. The sections show that ICAM-1 positive vessels are expressed in the ischemic hemisphere, especially in the infarct bound zone. These photographs demonstrate that decreased expression of ICAM-1 positive vessels in the AdRSVIL-1ratransduced mice, compared to the AdRSVlacZ-transduced and salinetreated mice. Scale bar, 30 mm.

were dead at the late times. The negative control section showed no ICAM-1 positive staining. Fig. 4. Bar graph showing ICAM-1 positive vessels in the ischemic cortex and basal ganglia in the AdRSVIL-1ra, AdRSVlacZ-transduced, and saline-injected mice following 1 h tMCAO and 23 h reperfusion. There are no significant differences in the number of ICAM-1 positive vessels in the contralateral hemisphere among the three groups Ždata not show here.. Compared to the AdRSVlacZ-transduced and saline-injected mice, the number of ICAM-1 positive vessels in the AdRSVIL-1ra-transduced mice were significantly reduced both in the ischemic cortex and basal ganglia. Values are mean"S.E.; ns6–8 in each group; U p- 0.05, the AdRSVIL-1ra-transduced mice vs. the AdRSVlacZ-transduced and the saline-injected mice.

4. Discussion Our study demonstrates that not only infarct volume but also BBB disruption and ICAM-1 expression are significantly reduced in the AdRSVIL-1ra-transduced mice following postischemic reperfusion. These findings are consistent with previous studies in which IL-1ra reduced

G.-Y. Yang et al.r Molecular Brain Research 72 (1999) 129–137

ischemic brain injury w2,15,28,44x. Since the differences in anatomical vessels and collateral blood flow may influence the degree of severity of the stroke insult, it is important to evaluate the level of CBF to make sure the occlusion is successful w41x. Our study shows that the levels of cortical blood flow in the ischemic core in the AdRSVIL-1ra, AdRSVlacZ-transduced, and saline-injected mice all decreased to 12–15% of the baseline CBF measurement after tMCAO and returned to 66–76% of the baseline immediately after reperfusion. Following 23 h reperfusion, the blood flow remained higher than 50% of the baseline flow both in ischemic core and perifocal regions in these three groups of mice. This result suggests that adenoviral transfection does not affect CBF in the mouse brain and that the protective effect of IL-1ra during postischemic reperfusion is not likely due to reduced CBF since no significant differences were detected in blood flow among these three groups. It has been proposed that during reperfusion activated microglial cells produce various inflammatory cytokines including IL-1b. Generally, the possible mechanisms of reperfusion brain injury produced by IL-1b are upregulating proinflammatory mediators w4,7,8,10,11,16x. Release of IL-1 during postischemic reperfusion may stimulate the synthesis of other cytokines such as IL-2, IL-6, and TNFa w9x. These cytokines share many central actions with IL-1, and the increase of these cytokines could in turn exacerbate brain injury w24x. IL-1 could stimulate endothelial cells producing neutrophil chemotactic factor w35x. IL-1 could also induce IL-8 production in the brain and blood during early reperfusion w26x. At present, the most recognized functions of IL-1 are the induction of endothelial cell adhesion molecule expression and the promotion of neutrophil tissue infiltration. Our data demonstrate that local overexpression of IL-1ra in the brain can reduce infarct volume after 24 h of tMCAO. Further studies demonstrate that there was no further reduction in infarct volume and BBB disruption in the AdRSVIL-1ra-transduced mice after longer periods of reperfusion. This suggests that the effect of IL-1ra on cerebral ischemia is most beneficial shortly after the onset of brain ischemia. Our results demonstrated that BBB disruption was significantly attenuated in the AdRSVIL-1ra-transduced mice, suggesting that proinflammatory cytokines may cause brain damage partly by disrupting the BBB. Plasma protein leakage across the BBB into the infarct area was seen 2–3 h after pMCAO, earlier than the appearance of histological injury w21x. In the reperfusion insult, increased vascular permeability appears much earlier than in permanent ischemic insult w1x. BBB injury could play an important role in the pathogenesis of ischemic and reperfusion brain injury, especially during reperfusion. Tomita and Fukuuchi w38x suggested that many neurotoxic factors play roles in ischemia which damages the BBB, followed by transudationrexudition, edema and necrosis. Several investigators

135

reported that oxygen free radical scavengers attenuated the permeability of BBB after temporary cerebral ischemia, w33,37,48x which suggests that oxygen free radicals may cause ischemic brain injury partly by disrupting the BBB. Increased BBB permeability following transient cerebral ischemia in rats could be attenuated by administration of an NOS inhibitor w45x. Rao et al. w27x reported that Sadenosyl-L-methionine tosylate, an anti-inflammatory drug, affects the outcome of ischemic injury by reducing BBB breakdown. Although we cannot exclude the possibility that decreased BBB disruption is due to the smaller infarction, BBB disruption may be the common pathway of many injury mechanisms after transient cerebral ischemia. Increased expression of adhesion molecules has been demonstrated following cerebral ischemia and reperfusion. Several research groups reported that ICAM-1 and ELAM1 are upregulated from 1–3 h up to 1 week following tMCAO in rats w23,39,47x. In the baboon MCAO model, Okada et al. w25x found that ICAM-1 and ELAM-1 upregulated and similarly localized to the endothelium of postcapillary microvasculature in the ischemic penumbra. Another supportive result is the inhibition of endothelial interactions with the leukocyte counterpart of ICAM-1 binding and the reduction in infarct size by 45–50% in the rat transient MCAO model w46x. PMNL begin to infiltrate into the ischemic tissue between 6 and 12 h after reperfusion w14,22x. Using the RP-3 monoclonal antibody, Matsuo et al. w22x reported a dramatic reduction in both neutrophil accumulation and infarct volume in the rat MCAO model. Furthermore, in a baboon transient focal ischemia model, anti-CD18 mAb administrated 25 min prior to reperfusion led to an increase in reflow in microvessels of various sizes w24x. Our study demonstrates that overexpression of IL-1ra in the mouse brain markedly attenuates ICAM-1 expression in the ischemic hemisphere. During reperfusion, IL-1 may stimulate endothelial cells to overexpress ICAM-1, which may facilitate the adhesion of leukocytes on the wall of the microvessels and the migration of leukocytes into the parenchyma. Our studies shows that ICAM-1 stained microvessels are markedly increased in the ischemic hemisphere in all three groups after 1 h occlusion and 23 h reperfusion. The perifocal region, which has less ischemic injury and more blood supply than the ischemic core region, had densely distributed ICAM-1 positive microvessels in all three groups. Our results suggest that ICAM-1 may facilitate migration of the leukocytes from less injured areas to the ischemic core region. The destructive effects of leukocytes, such as obstructing capillaries, damaging BBB, constricting microvasculature, and releasing cytotoxic mediators on the border zone, could lead injured brain cells in the perifocal area to irreversible damage. We believe that decreased ICAM-1 expression may be the reason for reduced infarct volume since our previous study w43x demonstrated that increased ICAM-1 expression occurred as early as 2 h after MCAO and infarction usually happened 8 h

136

G.-Y. Yang et al.r Molecular Brain Research 72 (1999) 129–137

after MCAO; however, determining which is the cause and which is the effect will require additional research. In summary, overexpression of IL-1ra in the mouse brain using an adenoviral vector provides a useful tool to explore the mechanism of action of IL-1 during postischemic reperfusion. Brain infarct volume and ICAM-1 expression were significantly attenuated in the AdRSVIL1ra-transduced mice following tMCAO, suggesting that IL-1b cause upregulation of ICAM-1. Reduced BBB disruption in the IL-1ra gene-transduced mice suggests proinflammatory cytokines may cause brain damage partly through disrupting BBB.

w16x

w17x w18x

w19x

w20x

Acknowledgements

w21x

These studies were supported by grants from the National Institutes of Health R01 NS-35089 ŽGYY. and R01 NS-23870 ŽALB.. The authors thank Ms. Kathleen Donahoe for editorial assistance.

w22x

w23x

References w1x S. Albayrak, Q. Zhao, B.K. Siesjo, M.L. Smith, Effect of transient focal ischemia on blood–brain barrier permeability in the rat: correlation to cell injury, Acta Neuropathologica 94 Ž1997. 158–163. w2x A.L. Betz, G.P. Schielke, G.Y. Yang, Interleukin-1 in cerebral ischemia, Keio J. Med. 45 Ž1996. 230–237. w3x A.L. Betz, G.Y. Yang, B.L. Davidson, Attenuation of stroke size in rats using an adenoviral vector to induce overexpression of interleukin-1 receptor antagonist in brain, J. Cereb. Blood Flow Metab. 15 Ž1995. 547–551. w4x C.M. Blatteis, Neuronal mechanisms in the pyrogenic and acute phase responses to interleukin-1, Int. J. Neurosci. 38 Ž1988. 223–232. w5x D.W. Choi, Ischemia-induced neuronal apoptosis, Curr. Opin. Neurobiol. 6 Ž1996. 667–672. w6x D.R. DiBona, W.J.J. Powell, Quantitative correlation between cell swelling and necrosis in myocardial ischemia in dogs, Circ. Res. 47 Ž1980. 653–665. w7x C.A. Dinarello, The biology of interleukin-1, in: T. Kishimoto ŽEd.., Interleukins: Molecular Biology and Immunology, Karger, Basel, Switzerland, 1992, pp. 1–32. w8x C.A. Dinarello, The role of interleukin-1 in host responses to infectious diseases, Infectious Agents and Disease 1 Ž1992. 227–236. w9x C.A. Dinarello, Biologic basis for interleukin-1 in disease, Blood 87 Ž1996. 2095–2147. w10x C.A. Dinarello, R.C. Thompson, Blocking IL-1: interleukin 1 receptor antagonist in vivo and in vitro, Immunology Today 12 Ž1991. 404–410. w11x A.A. Farooqui, L.A. Horrocks, Excitatory amino acid receptors, neuronal membrane phospholipid metabolism and neurological disorders, Brain. Res. Rev. 16 Ž1992. 171–191. w12x G.Z. Feuerstein, T. Liu, F.C. Barone, Cytokines, inflammation, and brain injury: role of tumor necrosis factor-alpha, Cerebrovasc. Brain Metab. Rev. 6 Ž1994. 341–360. w13x K.B.J. Franklin, Paxinos G., The mouse brain in stereotaxic coordinates, 1997, pp. 20–40. w14x J.H. Garcia, K.-F. Liu, Y. Yoshida, J. Lian, S. Chen, G. del Zoppo, Influx of leukocytes and platelets in an evolving brain infarct, Am. J. Pathol. 144 Ž1994. 188–199. w15x J.H. Garcia, K.F. Lui, J.K. Relton, Interleukin-1 receptor antagonist

w24x

w25x

w26x

w27x

w28x

w29x

w30x

w31x

w32x

w33x

w34x

decreases the number of necrotic neurons in rats with middle cerebral artery occlusion, Am. J. Pathol. 147 Ž1995. 1477–1486. D. Giulian, J. Woodward, D.G. Young, J.F. Krebs, L.B. Lachman, Interleukin-1 injected into mammalian brain stimulates astrogliosis and neovascularization, J. Neurosci. 8 Ž1988. 2485–2490. J.M. Hallenbeck, A.J. Dutka, Background review and current concepts of reperfusion injury, Arch. Neurol. 47 Ž1990. 1245–1254. H. Hara, K. Fink, M. Endres, R.M. Friedlander, V. Gagliardini, J. Yuan, M.A. Moskowitz, Attenuation of transient focal cerebral ischemic injury in transgenic mice expressing a mutant ICE inhibitory protein, J. Cereb. Blood Flow Metab. 17 Ž1997. 370–375. J.E. Krawisz, P. Sharon, W.F. Stenson, Quantitative assay for acute intestinal inflammation based on myeloperoxidase activity, Gastroenterology 87 Ž1984. 1344–1350. D.E. Levy, R.L. van Uitert, C.L. Pike, Delayed postischemic hypoperfusion: a potentially damaging consequence of stroke, Neurology 29 Ž1979. 1245–1252. E.M. Loberg, B. Hassel, F. Fonnum, A. Torvik, Early entry of plasma proteins into damaged neurons in brain infarcts. An immunohistochemical study on experimental animals, Apmis 102 Ž1994. 771–776. Y. Matsuo, H. Onodera, Y. Shiga, M. Nakamura, M. Ninomiya, T. Kihara, K. Kogura, Correlation between myeloperoxidase-quantified neutrophil accumulation and ischemic brain injury in the rat: effects of neutrophil depletion, Stroke 25 Ž1994. 1469–1475. Y. Matsuo, H. Onodera, Y. Shiga, H. Shozuhara, M. Ninomiya, T. Kihara, T. Tamatani, M. Miyasaka, K. Kogura, Role of cell adhesion molecules in brain injury after transient middle cerebral artery occlusion in the rat, Brain Res. 656 Ž1994. 344–352. E. Mori, G.J. del Zoppo, D. Chambers, B.R. Copeland, K.E. Arfors, Inhibition of polymorphonuclear leukocyte adherence suppresses no-reflow after focal cerebral ischemia in baboons, Stroke 23 Ž5. Ž1992. 712–718. Y. Okada, B.R. Copeland, E. Mori, M.M. Tung, W.S. Thomas, G.J. del Zoppo, P-selectin and intercellular adhesion molecule-1 expression after focal brain ischemia and reperfusion, Stroke 25 Ž1994. 202–211. H. Onodera, Y. Yamasaki, Y. Matsuo, Ischemia and neuroimmunology — ischemiarreperfusion injury, Nippon Rinsho 52 Ž1994. 2995–2999. A.M. Rao, M.K. Baskaya, M.E. Maley, M.S. Kindy, R.J. Dempsey, Beneficial effects of S-adenosyl-L-methionine on blood–brain barrier breakdown and neuronal survival after transient cerebral ischemia in gerbils, Brain Res. Mol. Brain Res. 44 Ž1997. 134–138. J.K. Relton, D. Martin, D. Russell, Effects of peripheral IL-1ra treatment after focal cerebral ischemia in the rat, Soc. Neurosci. 19 Ž1993. 6732. B.J. Roessler, J.W. Hartman, D.K. Vallance, J.M. Latta, S.L. Janich, B.L. Davidson, Inhibition of IL-1 induced effects in synoviocytes transduced with the human IL-1 receptor antagonist cDNA using an adenoviral vector, Hum. Gene Ther. 6 Ž1995. 307–316. J.L. Romson, B.G. Hook, S.L. Kunkel, G.D. Abrams, M.A. Schork, B.R. Lucchesi, Reduction of the extent of ischemic myocardial injury by neutrophil depletion in the dog, Circulation 67 Ž1983. 1016–1023. G.T. Rowe, N.H. Manson, M. Caplan, M.L. Hess, Hydrogen peroxide and hydroxyl radical mediation of activated leukocyte depression of cardiac sarcoplasmic reticulum, Circ. Res. 53 Ž1983. 584–591. G.P. Schielke, G.Y. Yang, B.D. Shivers, A.L. Betz, Reduced ischemic brain injury in interleukin-1 beta converting enzyme-deficient mice, J. Cereb. Blood Flow. Metab. 18 Ž1998. 180–185. C.L. Schleien, B. Eberle, D.H. Shaffner, R.C. Koehler, R.J. Traystman, Reduced blood–brain barrier permeability after cardiac arrest by conjugated superoxide dismutase and catalase in piglets, Stroke 25 Ž1994. 1830–1834. B.K. Siesjo, P. Siesjo, Mechanisms of secondary brain injury, Eur. J. Anaesthesiol. 13 Ž1996. 247–268.

G.-Y. Yang et al.r Molecular Brain Research 72 (1999) 129–137 w35x R.M. Strieter, S.L. Kunkel, H.J. Showell, Endothelial cell gene expression of a neutrophil chemotactic factor by TNFa , LPS, and IL-1b, Science 243 Ž1989. 1467–1469. w36x R.A. Swanson, M.T. Morton, W.G. Tsao, R.A. Savalos, C. Davidson, F.R. Sharp, A semiautomated method for measuring brain infarct volume, J. Cereb. Blood Flow Metab. 10 Ž2. Ž1990. 290–293. w37x E. Tasdemiroglu, P.D. Christenberry, J.L. Ardell, R.B. Chronister, A.E. Taylor, Effects of antioxidants on the blood–brain barrier and postischemic hyperemia, Acta Neurochir. 131 Ž1994. 302–309. w38x M. Tomita, Y. Fukuuchi, Leukocytes, macrophages and secondary brain damage following cerebral ischemia, Acta Neurochir. 66 Ž1996. 32–39. w39x X. Wang, G.Z. Feuerstein, Induced expression of adhesion molecules following focal brain ischemia, J. Neurotrauma 12 Ž1995. 825–832. w40x D. Wong, K. Dorovini Zis, Upregulation of intercellular adhesion molecule-1 ŽICAM-1. expression in primary cultures of human brain microvessel endothelial cells by cytokines and lipopolysaccharide, J. Neuroimmunol. 39 Ž1992. 11–22. w41x G.Y. Yang, A.L. Betz, J.T. Hoff, The effects of blood or plasma clot on brain edema in the rat with intracerebral hemorrhage, Acta Neurochir. 60 ŽSuppl. Ž1994. 555–557. w42x G.Y. 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.

137

w43x G.Y. Yang, C. Gong, Z. Qin, Y. Mao, X. Liu, A.L. Betz, Tumor necrosis factor alpha expression is associated with increased blood– brain barrier permeability following temporary focal cerebral ischemia in mice, Brain Res. Mol. Brain Res. Ž1998. in press. w44x G.Y. Yang, Y. Zhao, B.L. Davidson, A.L. Betz, Overexpression of interleukin-1 receptor antagonist in the mouse brain reduces ischemic brain injury, Brain Res. 751 Ž1997. 181–188. w45x J. Zhang, H. Benveniste, B. Klitzman, C.A. Piantadosi, Nitric oxide synthase inhibition and extracellular glutamate concentration after cerebral ischemiarreperfusion, Stroke 26 Ž1995. 298–304. w46x R.L. Zhang, M. Chopp, Y. Li, C. Zaloga, N. Jiang, M. Jone, M. Miyasaka, P. Ward, Anti-ICAM-1 antibody reduces ischemic cell damage after transient middle cerebral artery occlusion in the rat, Neurology 44 Ž1994. 1747–1751. w47x R.L. Zhang, M. Chopp, C. Zaloga, Z.G. Zhang, N. Jiang, S.C. Gautam, W.X. Tang, W. Tsang, D.C. Anderson, A.M. Manning, The temporal profiles of ICAM-1 protein and mRNA expression after transient MCA occlusion in the rat, Brain Res. 682 Ž1995. 182–188. w48x S.L. Zuckerman, R. Mirro, W.M. Armstead, M. Shibata, C.W. Leffler, Indomethacin reduces ischemia-induced alteration of blood–brain barrier transport in piglets, Am. J. Physiol. 266 Ž1994. H2198–2203.