Protective role of tuftsin fragment 1-3 in an animal model of intracerebral hemorrhage.код для вставкиСкачать
Protective Role of Tuftsin Fragment 1-3 in an Animal Model of Intracerebral Hemorrhage Jian Wang, MD, PhD, Andrew D. Rogove, MD, PhD, Anna E. Tsirka, MD, and Stella E. Tsirka, PhD Intracerebral hemorrhage (ICH) causes morbidity and mortality and commonly follows the reperfusion after an ischemic event. Tissue plasminogen activator (tPA), a fibrinolytic serine protease, is routinely given for the treatment of stroke. However, tPA also can promote neuronal death, suggesting that caution should be exercised when using it. Furthermore, tPA upon brain injury mediates microglial activation and modulates neuronal survival. To investigate the role of tPA and microglia during brain hemorrhage, we induced experimentally ICH by intracerebral injection of collagenase. Seven days after the introduction of ICH, it persisted in tPA-deficient (tPAⴚ/ⴚ) mice but is drastically reduced in size in wild-type mice. Three weeks after ICH, there are still red blood cells in tPAⴚ/ⴚ but not in wild-type animals. Activated microglia persist around the injury site. When microglial activation is inhibited by tuftsin fragment 1-3 macrophage/microglial inhibitory factor (MIF), the stroke injury volume is significantly reduced, and the neurobehavioral deficits exhibited by the mice are improved. Our results suggest that endogenous tPA assists in the clearance of intracerebral hemorrhage, presumably by affecting microglial activation, and MIF could be a valuable neuroprotective agent for the treatment of ICH. Ann Neurol 2003;54:655– 664 Stroke is the second most common cause of death in the world after heart disease and a leading cause of disability. Intracerebral hemorrhage (ICH) represents at least 10% of all stroke deaths.1 The prognosis of patients after ICH is poor, and the pathogenesis of damage after ICH remains poorly understood. Evidence of cell death due to apoptosis2,3 and inflammation has been described recently.4 – 6 Tissue plasminogen activator (tPA) is a secreted serine protease that converts inactive plasminogen to the active serine protease plasmin. tPA and plasmin, both members of the fibrinolytic system, initiate a potent proteolytic cascade that leads to dissolution of blood clots.7 tPA remains the only Food and Drug Administration–approved treatment for acute stroke.8 However, the increased incidence of symptomatic ICH has put constraints on tPA’s clinical use.9 Furthermore, tPA and plasmin produced in the brain can promote neuronal death after excitotoxin injection10 –13 or tran- sient focal ischemia/reperfusion injury.14 tPA is expressed in the brain by neurons and by microglia, the immunocompetent cells of the brain that have been associated when activated with neurotoxicity. tPA not only is produced by microglia, but also mediates their activation.11 When ICH occurs, the blood–brain barrier becomes disrupted. As a result, macrophages and leukocytes infiltrate the brain parenchyma, and their presence has been proposed to constitute a primary mechanism of cell death. It is also possible that microglia contribute to the observed neuronal death, because the actual disruption of blood–brain barrier can activate microglia. To investigate the role of tPA and microglia in the pathogenesis of the brain injury in ICH, we adapted the mouse collagenase hemorrhage model.15 Extensive activation of microglia was observed around the site of ICH. When we used the tripeptide MIF (macrophage/ microglial inhibitory factor, tuftsin fragment 1-3, From the Department of Pharmacological Sciences, University Medical Center at Stony Brook, Stony Brook, NY. Received Feb 3, 2003, and in revised form Jun 3. Accepted for publication Jul 30, 2003. Current address for Dr Rogove: Department of Neurology, New York Presbyterian Hospital-Weill Cornell Medical Center, New York, NY 10021. Address correspondence to Dr Stella E. Tsirka, Department of Pharmacological Sciences, BST-7, Room 183, University Medical Center at Stony Brook, Stony Brook, NY 11794-8651. E-mail: firstname.lastname@example.org Current address for Dr Anna E. Tsirka: Department of Pediatrics, Division of Pediatric Cardiology, University of Minnesota, Minneapolis, MN 55455. © 2003 American Neurological Association Published by Wiley-Liss, Inc., through Wiley Subscription Services 655 Thr-Lys-Pro) to inhibit the activation of microglia,14 functional improvement and decrease in degenerating neurons were observed. Materials and Methods Animal Procedures tPA-deficient (tPA⫺/⫺)16 and C57BL/6 (wild-type) mice were maintained in the Department of Laboratory Animal Research at Stony Brook with access to food and water ad libitum. tPA⫺/⫺ mice have been backcrossed for 12 generations to the C57Bl/6 background. All experiments were performed in accordance to the National Institutes of Health guide for the care and use of laboratory animals as well as the institutional guidelines established by the Institutional Animal Care and Use Committee (IACUC) committee at Stony Brook. All efforts were made to minimize the use of animals and ensure minimal suffering of those animals used. saline (PBS). Injury volumes were digitally quantified, using the NIH1.62 image software package, on 50m coronal sections using Luxol fast blue/cresyl violet staining.15 Hemorrhagic injury areas were summed from six to eight coronal slices at different levels. Volumes in cubic millimeters were calculated by multiplying the 0.5mm slice thickness by the measured areas. Histology Hematoxylin staining was performed according to standard protocols. Luxol fast blue/cresyl violet staining and the Fluoro-Jade B (FJB) staining were performed according to published protocols.15,17 Cells permeable to FJB were marked for cell death. Degenerating neurons were counted in three fields immediately adjacent to the hematoma using a magnification of ⫻400 over a microscopic field of 0.1mm2 and expressed as cells per square millimeter; areas with large blood vessels were avoided. Immunofluorescence Intracerebral Hemorrhage Model The procedure for inducing ICH was adapted to mice from an established rat protocol.15 In brief, mice weighing approximately 25 to 35gm were anesthetized by intraperitoneal injection of avertin (0.5mg/gm of body weight). To induce hemorrhage, we injected mice with collagenase (0.075U in 500nl saline) unilaterally into the caudate putamen, using the following stereotactic coordinates: 1.0mm posterior and 3.0mm lateral of bregma, 4.0mm in depth. Collagenase was delivered over 2 minutes, and we kept the needle in place for an additional 5 minutes to prevent any reflux. Other mice were infused with 100l MIF (500M, Sigma) in saline. MIF was delivered at a rate of 0.5l/hr via a microosmotic pump (Durect, Cupertino, CA) placed subcutaneously in the back of the animals. A brain infusion cannula connected to the pump was positioned at the coordinates above to deliver the compound. Two days after the onset of infusion, collagenase was injected unilaterally into the caudate putamen of the mice; infusion of MIF continued for another day. The mice were allowed to recover from surgery in a warm environment over a 3-hour period. Mice were carefully observed and monitored for several hours after recovery from anesthesia. MIF-treated mice were killed 24 hours after the collagenase injection. Non–MIF-treated mice were killed on days 1, 7, and 21 after injection. Neurological Deficit MIF-treated and control mice were scored blindly for neurological deficits using a 28-point neurological scoring system18 on day 1 after ICH. The tests included body symmetry, gait, climbing, circling behavior, front limb symmetry, compulsory circling, and whisker response. Each point was graded from 0 to 4. Maximum deficit score was 28. Free-floating sections were washed in PBS for 20 minutes, blocked in 5% normal serum, and incubated with 5-D-4 antibody (which recognizes activated microglia; 1:1,000; Seikagaku, East Falmouth, MA), followed by Alexa 488conjugated secondary antibody (1:1,000; Molecular Probes, Eugene, OR). Control sections were processed as described above, except that primary antibodies was omitted. Control sections were devoid of specific staining (n ⱖ 3 for each condition at each time point). Microglial Cultures Primary wild-type microglia were plated as described.19 Collagenase was added at different concentrations to these cells, and their morphology was evaluated 24 hours later. Spectrophotometric Assay for Intracerebral Hemorrhage The hemoglobin content of brains subjected to ICH was quantified with Drabkin’s reagent. Brain hemispheres were obtained from wild-type or tPA⫺/⫺ animals. The tissue (injected vs uninjected) was trimmed to contain only the caudate putamen region and was treated individually as follows.20 Distilled water (500l) was added to each sample, followed by homogenization for 5 minutes, and centrifugation at 5,000g for 5 minutes. The supernatant, which contains the hemoglobin, was collected. Eighty microliters of Drabkin’s reagent (Sigma, St. Louis, MO) was added to a 20l aliquot and allowed to stand for 15 minutes at room temperature. The concentration of cyanmethemoglobin produced was measured at 550nm. A standard curve, reflecting the amount of hemoglobin present, was generated by adding to 100l lysate from untreated tissue 0.1, 0.5, 1.0, and 2.0l of blood obtained from control mice by cardiac puncture after anesthesia. In Situ Matrix Metalloproteinases Activity Hemorrhagic Injury Analysis On day 1 after neurological scoring, mice were killed and their brains were removed, fixed, and dehydrated in 4% paraformaldehyde and 20% sucrose in phosphate-buffered 656 Annals of Neurology Vol 54 No 5 November 2003 To localize the gelatinolytic activity of matrix metalloproteinases (MMPs) in situ, we used fluorescein isothiocyanate– labeled DQ gelatin as a substrate (EnzChek; Molecular Probes). MMP activity yields fluorescent fluorescein isothio- cyanate–conjugated gelatin. Fresh 10m cryostat sections were incubated with reaction buffer (0.05M Tris-HCl, 0.15M NaCl, 5mM CaCl2, and 0.l2mM NaN3, pH 7.6) containing 20g/ml DQ gelatin for 1 hour. MMP activity was visualized by confocal microscopy and photographed. MMP-expressing cells were counted in three fields immediately adjacent to the hematoma site by using a magnification of ⫻400 over a microscopic field of 0.1mm2 and expressed as cells/mm2; areas with large blood vessels were avoided. In vitro collagenase activity was determined using the EnzChek kit, as per manufacturer’s instructions. Statistics Statistical analysis was performed using SAS 8.0 software for PC. The data were analyzed by one-way analysis of variance, followed by multiple linear regression. Multiple regression models included the interaction between tPA and MIF. Statistical significance was set at p value less than 0.05. Results The injection of collagenase is an established model of ICH. Within 30 minutes, bleeding is observed at the injection site and continues for 4 hours. No tissue necrosis is then evident. At 24 hours, two zones of injury can be observed: a central zone of erythrocytes and necrotic tissue, surrounded by a peripheral zone that contains mixed viable and necrotic tissue. Forty-eight hours after ICH, there is increased decay in the central zone, and polymorphonuclear cells invade the periphery of the lesion.21 By 7 days, the central zone contains necrotic ghost cells and pale erythrocytes, whereas the peripheral zone contains a broad band of phagocytic macrophages, and by 21 days the hemorrhagic area resolves to a cystic structure/glial scar bordered by reactive astrocytes. Effects of Tissue Plasminogen Activator on Intracerebral Hemorrhage Clearance In tPA⫺/⫺ mice, an altered time course of hematoma clearance was observed. Seven days after ICH, its presence persisted in these mice. However, the hematoma was drastically reduced in size in wild-type mice. By 21 days, red blood cells were still visible in tPA⫺/⫺ mice, but not in wild-type mice (Fig 1). These results indicate that wild-type and tPA⫺/⫺ mice show a different time course of clearance of the hematoma, suggesting that endogenous tPA assists in its clearance. Our results are in agreement with those published recently in a study using a pig intracerebral hemorrhage model.22 To ascertain that the difference in clearance is not caused by a difference in the original collagenase injury, we measured the initial levels of hemoglobin in the injected tissue in wild-type and tPA⫺/⫺ animals, as Fig 1. Endogenous tissue plasminogen activator (tPA) assists in intracerebral hemorrhage (ICH) clearance. Coronal sections through the site of collagenase delivery were collected at different time points after ICH and stained with hematoxylin. Wild-type and tPA⫺/⫺ mice show a different time course of clearance of the hematoma. Note the presence of red blood cells, even at the later time points in tPA⫺/⫺ mice. Magnification: left, ⫻100; right, ⫻400. Wang et al: tPA, ICH, and Microglial Activation 657 an indicator of the bleeding introduced by the injection. No significant difference between wild-type and tPA⫺/⫺ mice can be observed 18 to 24 hours after collagenase injection, indicating that the clearance delays in tPA⫺/⫺ mice can be attributed only to the absence of tPA (Fig 2). Effect of Macrophage/Microglial Inhibitory Factor on Microglial Activation after Intracerebral Hemorrhage The persistence of the hematoma could be attributed to the absence of the fibrinolytic properties of tPA/ plasmin, to limited phagocytic activity (because microglial activation is attenuated in tPA⫺/⫺ mice11), or to both. We first evaluated the presence of activated microglia surrounding the injury site. Activated microglia (5-D-4 –positive, round cells, Fig 3) were detected in both wild-type and tPA⫺/⫺ mice 1 day after collagenase injection, but fewer activated microglia could be seen in tPA⫺/⫺ mice. Seven days after ICH, activated microglia persisted in tPA⫺/⫺, but their numbers were drastically reduced in wild-type mice. By 21 days, the Fig 2. The initial collagenase-induced intracerebral hemorrhage (ICH) is similar in wild-type and tPA⫺/⫺ mice. Total hemoglobin levels were measured in lysates from the injected caudate putamen of mice. A standard curve was made from lysates from control (uninjected) animals. No significant difference in hemoglobin levels after induction of ICH was observed between the two strains of mice (n ⫽ 4). tPA ⫽ tissue plasminogen activator. numbers of activated microglial had returned to normal in wild-type mice but persisted in tPA⫺/⫺ mice. Collagenase alone had no effect on microglial activation in culture (see Fig 3E). Because there was a difference between wild-type and tPA⫺/⫺ mice in microglial appearance and activation profile, we used an inhibitor of microglial activation before the injury. In an excitotoxic model, MIF had prevented microglial activation.19 In the ICH paradigm, inhibition of microglial activation by MIF (whose delivery started 2 days before the collagenase) was evident 1 day after ICH both in wild-type and tPA⫺/⫺ mice (see Fig 3). MIF had only a minimal inhibitory effect on collagenase, reducing its activity by approximately 10% (Fig 4F). Effect of Macrophage/Microglial Inhibitory Factor on Stroke Volume and Neuronal Cell Death Because MIF inhibited microglial activation, we evaluated its effect on injury volume and the ensuing neuronal death. MIF treatment reduced injury volume by 60 to 70% 1 day after ICH, from 9.5 ⫾ 3.3 to 3.6 ⫾ 0.7mm3 in tPA⫺/⫺ mice ( p ⬍ 0.005), and from 11.6 ⫾ 4.7 to 3.4 ⫾ 0.6mm3 in wild-type mice ( p ⬍ 0.005), respectively (see Fig 4), indicating that activated microglia contribute to injury during ICH. To examine whether neuronal death was evident at the site of hemorrhage, we used the FJB histological staining. MIF reduced the number of degenerating neurons by nearly 50% (Fig 5), from 240 ⫾ 39/mm2 to 126 ⫾ 30/mm2 in tPA⫺/⫺ mice (n ⫽ 5; p ⬍ 0.001) and from 272 ⫾ 61/mm2 to 152 ⫾ 31/mm2 in wild-type mice (n ⫽ 5; p ⬍ 0.001). Because there is no difference in degenerated neurons between tPA⫺/⫺ and wild-type mice, we suggest that factors other than neuronal loss contribute to morbidity/mortality after hemorrhage, that is, edema leading to mass effect and herniation. Therefore, MIF may be either affecting the edema volume or inhibiting the properties of infiltrating macrophages. Furthermore, persistent microglia could lead to additional cell death, thereby leading to additional deficits. Fig 3. Microglial activation after intracerebral hemorrhage. Microglial cells (5-D-4 –positive) are present near the hematoma in wild-type and tPA⫺/⫺ mice. Microglial activation peaked in wild-type animals around day 7, but not in tPA⫺/⫺ mice, where significant microglial activation is evident at day 21. After infusion of macrophage/microglial inhibitory factor (MIF), microglial activation was efficiently inhibited at day 1 both in wild-type and tPA⫺/⫺ mice. Magnification: columns A, C, ⫻100; column B, D, ⫻400. Scale bars ⫽ 20m. (E) Addition of collagenase had no effect on microglial activation. LPS ⫽ lipopolysaccharide. tPA ⫽ tissue plasminogen activator. 658 Annals of Neurology Vol 54 No 5 November 2003 ‹ Figure 3 Wang et al: tPA, ICH, and Microglial Activation 659 Fig 4. Macrophage/microglial inhibitory factor (MIF) reduced injury volume in wild-type and tPA⫺/⫺ mice. Coronal sections were collected at day 1 after intracerebral hemorrhage (ICH) and stained for Luxol fast blue/cresyl violet. (A) Control (ctr) tPA⫺/⫺ mice; (B) MIF-treated tPA⫺/⫺ mice; (C ) control wild-type mice; (D) MIF-treated wild-type mice. (E) MIF-treated wild-type (n ⫽ 5), and tPA⫺/⫺ (n ⫽ 5) mice had smaller injury volumes than control-treated wild-type (n ⫽ 6) or tPA⫺/⫺ (n ⫽ 6) mice. Values are means ⫾ SD. (F) Collagenase was incubated for different time periods with 0.5mM MIF or 1,10-phenanthroline (specific inhibitor), and its residual enzymatic activity was determined. tPA ⫽ tissue plasminogen activator; wt ⫽ wild type. 660 Annals of Neurology Vol 54 No 5 November 2003 Fig 5. Macrophage/microglial inhibitory factor (MIF) reduced the number of neurons degenerating after intracerebral hemorrhage (ICH). (A) Fluoro-Jade B (FJB) histological staining of neurons in coronal sections collected at different time-points after the collagenase injection. Evenly distributed, intensely labeled neurons and processes are observed in the peri-ICH region on day 1 in wild-type and tPA⫺/⫺ mice. There is persistence of degenerating neurons in tPA⫺/⫺ mice 7 days after ICH, but not in the wild-type mice. Magnification: columns a, c: ⫻100; columns b, d: ⫻400. Scale bars ⫽ 20m. (B) Quantification of the numbers of degenerating neurons on day 1 in wild-type and tPA⫺/⫺ control and MIF-treated mice. MIF-treated mice had fewer degenerating neurons than control wild-type or tPA⫺/⫺ mice. n ⫽ 5 for each genotype. Values are means ⫾ SD. Effect of Macrophage/Microglial Inhibitory Factor on Intracerebral Hemorrhage–Induced Neurobehavioral Deficits Intracerebral hemorrhage usually is accompanied by characteristic behavioral deficits. To determine whether the ICH-induced pathological and molecular events that showed improvement after MIF pretreatment were paralleled by neurobehavioral recovery, we performed repeated assessments of the animals on day 1 after ICH. MIF significantly improved the neurobehavioral score of the animals compared with non– MIF-treated animals; the score changed from 8.2 ⫾ 0.8 to 5.4 ⫾ 1.1 in tPA⫺/⫺ mice ( p ⬍ 0.005) and from 7.5 ⫾ 1.5 to 5.0 ⫾ 0.7 in wild-type mice ( p ⬍ 0.005; Fig 6). Matrix Metalloproteinase Activity after Intracerebral Hemorrhage The potential interaction between collagenase injury and other proteolytic systems, especially other MMPs, which could be activated by collagenase or plasmin, Fig 6. Macrophage/microglial inhibitory factor (MIF) improved neurological functional outcome on day 1 in both wild-type and tPA⫺/⫺ mice, compared with control mice: from 8.2 to 5.4 in tPA⫺/⫺, and from 7.5 to 5 in wild-type mice, respectively. Values are means ⫾ SD. tPA ⫽ tissue plasminogen activator. Wang et al: tPA, ICH, and Microglial Activation 661 Fig 7. Matrix metalloproteinase (MMP) activation after intracerebral hemorrhage (ICH) on day 1 in control, wild-type, and tPA⫺/⫺ mice. (A) MMP-expressing cells are present in the injury area in all animals. Magnification: columns a, c, ⫻100; columns b, d: ⫻400. Scale bars ⫽ 20m. (B) Quantification of MMP-expressing cells on day 1 in wild-type and tPA⫺/⫺ mice. No significant difference was observed between the two genotypes of mice (n ⫽ 5). Values shown are means ⫾ SD. was evaluated. MMP activity was determined in situ in PBS-injected mice or in wild-type and tPA⫺/⫺ mice on day 1 after ICH (Fig 7A). MMP activity was clearly identifiable in the injury area of wild-type, tPA⫺/⫺, and PBS-injected mice, indicating that the hemorrhagic injury can activate MMPs. The gelatinolytic activity appeared to be associated with neuronal cells. No significant difference (271 ⫾ 41/mm2 vs 247 ⫾ 52/ mm2; p ⫽ 0.105) could be observed in MMP- 662 Annals of Neurology Vol 54 No 5 November 2003 expressing cells between wild-type (n ⫽ 5) and tPA⫺/⫺ mice (n ⫽ 5) on day 1 after ICH (see Fig 7B). Discussion Experimental and clinical studies have demonstrated that tPA administration is beneficial when treatment is initiated within 3 hours (intravenously) or up to 6 hours (intraarterial delivery) of the onset of ischemic stroke.9,23 However, the presence of significant amounts of tPA has resulted in increased incidence of local ICH, which in turn has worsened the stroke outcome. The role of endogenous tPA in ICH is not clear. Results obtained from a focal ischemia/reperfusion injury paradigm in the absence of induced thrombi in tPA⫺/⫺ mice indicate that tPA contributes to neuronal damage.10 The data here indicate that endogenous tPA assists in ICH clearance but also affects microglial activation. Inhibition of microglial activation by MIF reduces the injury volume and degenerating neurons and improves neurological functional outcome. Therefore, when intraarterial tPA is injected to dissolve the clot, it might be beneficial to coinject MIF, because it would cross the permeable blood–brain barrier in stroke patients. In addition, it may be possible to wash the evacuated hemorrhage bed with MIF when ICH is treated surgically. Inflammation contributes to brain injury after ICH.4 – 6 Our results indicate that, besides macrophage infiltration, microglial activation does occur in and around hematomas. The presence of activated microglia also has been documented in another model of ICH, in rats injected intracerebrally with autologous blood.5 We show that there is a functional consequence to the presence of activated microglia, because MIF was able to reduce the cell death observed and improved the neurological functional outcome. The inhibition of timely microglial activation results in a corresponding decrease in TNF-␣ and tPA secretion.13,14,19 Consistent with the notion that MIF inhibits microglial activation and macrophage migration, we show that MIF not only reduces the degenerating neurons and injury volume, but also improves neurological functional outcome. Using multiple regression analysis, we correlated neuron density, lesion volume, and neurological scores in wild-type and tPA⫺/⫺ animals that had received control or MIF treatment. As described in Results, MIF treatment produced significant differences in all parameters examined ( p ⬍ 0.001, p ⬍ 0.005, and p ⬍ 0.005, respectively). In conclusion, our study provides support for the idea that activated microglia are important contributors to brain injury during ICH. tPA, because of its thrombolytic role, is beneficial for hematoma clearance; however, its indirect role, microglial activation, should be kept in check. In this respect, the recent report in experimental models of excitotoxicity using desmoteplase, a vampire bat–derived tPA, provides an interesting possibility of a thrombolytic enzyme devoid of microglial activation capabilities.24 Our results also suggest that MIF could be a valuable neuroprotective agent for the treatment of ICH. Further studies in other hemorrhagic stroke models are necessary to determine the optimal dose and administration of MIF in ICH. This work was supported by grants from the National Institutes of Health (ROINS042168, S.E.T.); a fellowship from the American Heart Association (0225701T, J.W.), and by a National Institute of Diabetes and Digestive and Kidney Diseases NRSA grant (T32DK007521, A.D.R.). We thank members of the Tsirka lab and Dr R. Thiex for critical reading of the manuscript. References 1. Qureshi AI, Tuhrim S, Broderick JP, et al. Spontaneous intracerebral hemorrhage. N Engl J Med 2001;344:1450 –1460. 2. Matsushita K, Meng W, Wang X, et al. Evidence for apoptosis after intracerebral hemorrhage in rat striatum. J Cereb Blood Flow Metab 2000;20:396 – 404. 3. Gong C, Boulis N, Qian J, et al. Intracerebral hemorrhageinduced neuronal death. Neurosurgery 2001;48:875– 882; discussion, 882– 883. 4. Hickenbottom SL, Grotta JC, Strong R, et al. Nuclear factorkappa B and cell death after experimental intracerebral hemorrhage in rats. Stroke 1999;30:2472–2477; discussion, 2477– 2478. 5. Gong C, Hoff JT, Keep RF. Acute inflammatory reaction following experimental intracerebral hemorrhage in rat. Brain Res 2000;871:57– 65. 6. Castillo J, Davalos A, Alvarez-Sabin J, et al. Molecular signatures of brain injury after intracerebral hemorrhage. Neurology 2002;58:624 – 629. 7. Vassalli JD, Sappino AP, Belin D. The plasminogen activator/ plasmin system. J Clin Invest 1991;88:1067–1072. 8. Meschia JF, Miller DA, Brott TG. Thrombolytic treatment of acute ischemic stroke. Mayo Clin Proc 2002;77:542–551. 9. Zivin JA. Thrombolytic stroke therapy: past, present, and future. Neurology 1999;53:14 –19. 10. Wang YF, Tsirka SE, Strickland S, et al. Tissue plasminogen activator (tPA) increases neuronal damage after focal cerebral ischemia in wild-type and tPA-deficient mice. Nat Med 1998; 4:228 –231. 11. Rogove AD, Siao C, Keyt B, et al. Activation of microglia reveals a non-proteolytic cytokine function for tissue plasminogen activator in the central nervous system. J Cell Sci 1999;112: 4007– 4016. 12. Tsirka SE, Rogove AD, Bugge TH, et al. An extracellular proteolytic cascade promotes neuronal degeneration in the mouse hippocampus. J Neurosci 1997;17:543–552. 13. Thanos S, Mey J, Wild M. Treatment of the adult retina with microglia-suppressing factors retards axotomy-induced neuronal degradation and enhances axonal regeneration in vivo and in vitro. J Neurosci 1993;13:455– 466. 14. Auriault C, Joseph M, Tartar A, Capron A. Characterization and synthesis of a macrophage inhibitory peptide from the second constant domain of human immunoglobulin G. FEBS Lett 1983;153:11–15. 15. Clark W, Gunion-Rinker L, Lessov N, Hazel K. Citicole treatment for experimental intracerebral hemorrhage in mice. Stroke 1998;29:2136 –2140. 16. Carmeliet P, Schoonjans L, Kieckens L, et al. Physiological consequences of loss of plasminogen activator gene function in mice. Nature 1994;368:419 – 424. 17. Schmued LC, Hopkins KJ. Fluoro-Jade B: a high affinity fluorescent marker for the localization of neuronal degeneration. Brain Res 2000;874:123–130. 18. Clark WM, Lessov NS, Dixon MP, Eckenstein F. Monofilament intraluminal middle cerebral artery occlusion in the mouse. Neurol Res 1997;19:641– 648. Wang et al: tPA, ICH, and Microglial Activation 663 19. Rogove AD, Tsirka SE. Neurotoxic responses by microglia elicited by excitotoxic injury in the mouse hippocampus. Curr Biol 1998;8:19 –25. 20. Choudhri T, Hoh B, Solomon R, et al. Use of spectrophotometric hemoglobin assay to objectively quantify intracerebral hemorrhage in mice. Stroke 1997;28:2296 –2302. 21. Peeling J, Yan H-J, Corbett D, et al. Effect of FK-506 on inflammation and behavioral outcome following intracerebral hemorrhage in rat. Exp Neurol 2001;167:341–347. 22. Rohde V, Rohde I, Thiex R, et al. Fibrinolysis therapy achieved with tissue plasminogen activator and aspiration of 664 Annals of Neurology Vol 54 No 5 November 2003 the liquefied clot after experimental intracerebral hemorrhage:rapid reduction in hematoma volume but intensification of delayed edema formation. J Neurosurg 2002;97: 954 –962. 23. Overgaard K, Sperling B, Boysen G, et al. Thrombolytic therapy in acute ischemic stroke. A Danish pilot study. Stroke 1993;24:1439 –1446. 24. Liberatore G, Samson A, Bladin C, et al. Vampire bat salivary plasminogen activator (desmoteplase): a unique fibrinolytic enzyme that does not promote neurodegeneration. Stroke 2003; 34:537–543.