Serine proteases and brain damage – is there a link

Serine proteases and brain damage – is there a link

Please download to get full document.

View again

of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.


Publish on:

Views: 6 | Pages: 9

Extension: PDF | Download: 0

  V IEWPOINT 0166-2236/00/$ – see front matter © 2000 Elsevier Science Ltd. All rights reserved. PII: S0166-2236(00)01617-9 TINS Vol. 23, No. 9, 2000  399 S urprisingly, the brain contains some of the sameproteases and protease-activated receptors (PARs)that serve key roles in blood coagulation and woundhealing (Box1). Whereas these components of theblood clotting and thrombolysis systems probablymediate unique functions in normal and developingbrain, their dual use by blood and brain has interest-ing implications for situations in which theblood–brain barrier (BBB) is compromised. In particu-lar, serine protease or zymogen precursor entry intobrain tissue during cerebrovascular insult might leadto aberrant activation of PARs on neurons and glia, aswell as cleavage of other non-receptor substrates, withpotentially harmful consequences. Recent evidencesuggests that extravasated thrombin, plasmin, tissueplasminogen activator (tPA) and related serine pro-teases have a role in neuropathological situations inwhich the BBB is compromised. Protease-activated receptors Thrombin is a serine protease that is generated byproteolysis of its precursor (prothrombin) at sites of vascular injury. Thrombin stimulation of plateletsand limited proteolysis of fibrinogen enable clot for-mation and wound healing in blood vessels. Many of the actions of thrombin can be traced to PARs. Todate, cDNAs encoding four PAR receptors (PAR1–4)have been identified 1–5 and three of these genes(human  PAR1, PAR2, PAR3 ) map to a single chromo-some (5q13) in a gene cluster 6 ; human  PAR4 maps tochromosome 19p12. Figure1 illustrates how throm-bin interacts with PAR1 through its anion-bindingexosite 7 to subsequently activate these receptors byproteolysis of the extracellular N-terminal domain atArg41, which releases a small peptide and unmasks anew N-terminus 8–9 . The first few amino acids of thenew N-terminus (SFLLRN for PAR1) act as a tetheredligand that binds to another part of the receptor 3,10 toinitiate signaling by heterotrimeric G-protein sub-units G  , G i  and G q  /G o  (Ref. 8). Among thePARs, PAR2 represents a class of trypsin/tryptasereceptors, whereas PAR1, PAR3 and PAR4 are mosteffectively activated by thrombin. PAR1 can also besubmaximally activated by Factor Xa, trypsin andplasmin, whereas Factor VIIa, tPA, cathepsin G, elas-tase and chymotrypsin are inactive at these recep-tors 3 . Interestingly, plasmin can also cleave PAR1 at alinker domain that lies downstream of the SFLLRNactivation sequence, thereby inactivating the recep-tor 11 . PAR1, PAR2 and PAR4, but not PAR3, can beactivated by a short peptide (for example, SFLLRN forPAR1) corresponding to the first few N-terminalresidues revealed by proteolysis (Fig.1). Because PAR-selective peptides can be designed, these peptidesprovide an important tool for evaluating the involve-ment of specific PARs in various processes 12 .Activation of thrombin receptors in both theperiphery and the CNS can influence a host of secondmessengers (Fig.1), including inhibition of adenylylcyclase, stimulation of phosphoinositide turnover,activation of protein kinase C, and activation of mito-gen activated protein kinase (MAP kinase) and phos-phoinositide 3-kinase (PI3 kinase) 8,9 . Thrombin acti-vation of PAR1 in rat primary hippocampal culturesstimulates a rapid increase in intracellular Ca 2  levels 13,14 and a slower reduction of cAMP levels 13 .Thrombin receptors can also activate neuronal andglial tyrosine kinases in hippocampal neurons andstimulate cGMP formation in neuroblastoma cells 15–17 .Although thrombin receptors are constitutively acti-vated by proteolysis, they are rapidly desensitized by avariety of mechanisms, including phosphorylationand internalization through coated pits 8,18–21 . Neitherthrombin-desensitized receptors that are retained atthe cell surface nor receptors that are recycled to theplasma membrane can be activated again by throm-bin, however these receptors are responsive to acti-vation by the peptide SFLLRN (Ref. 22). Localization of protease receptors and inhibitors inthe brain The first thrombin receptor to be cloned, PAR1, iswidely distributed in neurons and glia 23,24 .  PAR1 mRNAis widespread in prenatal rat brain tissue, and becomesmore pronounced and confined to particular cell typesin adult animals 23,24 . This developmental profile differs Serine proteases and brain damage – isthere a link? Melissa B. Gingrich and Stephen F. Traynelis The protective blood–brain barrier normally allows diffusion of small molecules such as oxygenand carbon dioxide,and transport of essential nutrients,but excludes large proteins and other blood constituents from the interstitial space of the CNS.However,head trauma,stroke,statusepilepticus and other pathological conditions can all compromise the integrity of this barrier,andallow blood proteins as large as albumin to gain access to the extracellular spaces that surroundneurons and glia.Given their possible entry into brain tissue during cerebrovascular insult,theeffects of blood-derived proteases such as thrombin,tissue plasminogen activator and plasmin inthe CNS have come under increasing scrutiny.Evidence now supports a role for serine proteasesin the sequence of events that can lead to glial scarring,edema,seizure and neuronal death. Trends Neurosci. (2000) 23, 399–407  Melissa B.Gingrich is at theCenter for  Reproduction,University of Virginia,Charlottesville,VA22908-0391,USA and Stephen F. Traynelis is at theDept of  Pharmacology, Emory University School of  Medicine, Atlanta,GA 30322-3090,USA.  400 TINS Vol. 23, No. 9, 2000 from that of prothrombin mRNA, which is transientlyreduced near birth 25 , raising the possibility that serineproteases and their receptors might play different rolesin pre-, neo- and postnatal brain. Hybridization of   PAR1 mRNA is found in distinct cell layers of the cor-tex, subiculum, hypothalamus, thalamus, pretectum,ventral mesencephalon, cerebellum and olfactorybulb. In the hippocampus,  PAR1 transcripts can befound in the granule cell layer of the dentate gyrus,with more diffuse hybridization in the pyramidal celllayer of CA1 (Fig.2). Intense hybridization wasobserved in Purkinje cells of the cerebellum, somebrainstem nuclei and the locus coeruleus (Fig.2). Manyfunctional responses to thrombin in neurons can bemimicked with PAR1 agonist peptides (see below), sug-gesting that thrombin receptors exist in some neurons.The mRNAs for  PAR2 and  PAR4 are not present in thebrain 2,4 , although PAR2 immunoreactivity has beenreported in cultured hippocampal neurons 26 . In addi-tion,  PAR3 can be found at low levels in mouse brain 3 .Recent data suggest that glial and neuronal cellstogether possess their own serine-protease-signalingsystem, containing many of the same componentsthat are present in the blood coagulation/fibrinolysispathways (Box1). Although the connectivity andfunctions of this pathway remain unknown, the pres-ence of protease inhibitors in the brain is particularlyrelevant to the ideas under consideration here. Theseinhibitors are ideally suited to serve protective roles inthe event of blood-derived proteases entering brain tissue. Protease nexin (PN1) and neuroserpin bothbelong to the serpin family of protease inhibitors thatform a complex with the catalytic residues of theircognate protease (Box1). PN1-producing cells caninternalize and degrade PN1-containing complexes 27 .PN1 is most potent as a thrombin inhibitor, althoughat high concentrations it can also complex with plas-min, tPA, urokinase plasminogen activator (uPA) andtrypsin. In addition to neurons, astroglia with perivas-cular endfeet and smooth-muscle cells that form the M.B. Gingrich and S.F. Traynelis  – Serine proteases in the brain V IEWPOINT The clotting cascade is remarkablycomplex, consisting of a multitudeof serine proteases and co-factorsthat conspire to cleave substrates,specific receptors and each other tocontrol clot formation and dissolu-tion. Not surprisingly, other tissueshave found uses for these interestingsignaling molecules and their recep-tors independent of blood coagula-tion. Genes expressed in the CNSthat were first identified in bloodcoagulation/fibrinolysis pathwaysare highlighted in Fig. I, and includethe zymogen precursors to throm-bin (prothrombin a ), plasmin (plas-minogen b ) and activated protein C(protein C) c . The serine proteases tissue plasminogen activator (tPA) d ,urokinase plasminogen activator (uPA) e , Factor Xa (Ref.f)(whichcleaves prothrombin to thrombin), a protease inhibitor of tPA (PAI-1) e,g , a thrombin receptor [PAR1; (Fig. 2)], a thrombin-binding protein(thrombomodulin) h , and prekininogens i (not shown) are also pro-duced by the CNS. In addition, neurons and glia manufacture manyserine proteases such as neurotrypsin j , neuropsin k and myelen-cephalon-specific protease l , brain type granzyme B (Ref. m),motopsin n , as well as protein inhibitors of thrombin [PN1 (Ref. o); B-43(Ref. p)], and tPA (neuroserpin q,r ). Following the recent identifica-tion of these signaling molecules in brain tissue, much work is neededto connect their actions into understandable pathways and identifyphysiological roles for such pathways. However, once this is known,these pathways might provide exciting new targets for therapeuticintervention in a variety of disease states. References a Weinstein, J.R. et al .(1995) Cellular localization of thrombin receptormRNA in rat brain: expression by mesencephalic dopaminergic neuronsand codistribution with prothrombin mRNA.  J. Neurosci .15, 2906–2919 b Nakajima, K. et al .(1992) Production and secretion of plasminogen incultured rat brain microglia.  FEBS Lett  . 308, 179–182 c Yamamoto, K. and Loskutoff, D.J. (1998) Extrahepatic expression andregulation of protein C in the mouse.  Am. J. Pathol. 153, 547–555 d Sumi, Y. et al. (1992) The expression of tissue and urokinase-typeplasminogen activators in neural development suggests different modesof proteolytic involvement in neuronal growth.  Development 116, 625–637 e Masos, T. and Miskin, R.(1997) mRNAs encoding urokinase-typeplasminogen activator and plasminogen activator inhibitor-1 are elevatedin the mouse brain following kainate-mediated excitation.  Mol. Brain Res .47, 157–169 f  Yamada, T. and Nagai, Y. (1996) Immunohistochemical studies of humantissues with antibody to factor Xa.  Histochem. J. 28, 73–77 g  Wagner, S.L. et al .(1991) Inhibitors of urokinase and thrombin in culturedneural cells.  J. Neurochem. 56, 234–242 h Pindon, A. et al . (1997) Novel expression and localization of activethrombomodulin on the surface of mouse brain astrocytes. Glia 19, 259–268 i Takano, M. et al .(1997) Molecular cloning of cDNAs for mouse low-molecular weight and high-molecular weight prekininogens.  Biochim. Biophys. Acta 1352, 222–230 j Gschwend, T.P. et al .(1997) Neurotrypsin, a novel multidomain serineprotease expressed in the nervous system.  Mol. Cell. Neurosci . 9, 207–219 k Chen, Z.L. et al .(1995) Expression and activity-dependent changes of anovel limbic-serine protease gene in the hippocampus.  J. Neurosci .15,5088–5097 l Scarisbrick, I.A. et al . (1997) Nervous system-specific expression of a novelserine protease: regulation in the adult rat spinal cord by excitotoxicinjury.  J. Neurosci. 17,8156–8168 m Suemoto, T. et al .(1999) cDNA cloning and expression of a novel serineprotease in the mouse brain.  Mol. Brain Res. 70, 273–281 n Iijima, N. et al .(1999) Expression of a serine protease (motopsin PRSS12)mRNA in the mouse brain: in situ hybridization histochemical study.  Mol.Brain Res. 66, 141–149 o Choi, B.H. et al .(1990) Protease nexin-1: localization in the human brainsuggests a protective role against extravasated serine proteases.  Am. J. Pathol. 137  , 741–747 p Nakaya, N. et al. (1996) Cloning of a serine proteinase inhibitor frombovine brain: expression in the brain and characterization of its targetproteinases.  Mol. Brain Res. 42, 293–300 q Hastings, G.A. et al .(1997) Neuroserpin, a brain-associated inhibitor of tissue plasminogen activator is localized primarily in neurons.  J. Biol.Chem. 272, 33062–33067 r  Krueger, S.R. et al . (1997) Expression of neuroserpin, an inhibitor of tissueplasminogen activator, in the developing and adult nervous system of themouse.  J. Neurosci . 17, 8984–8996 Box 1.Components of the clotting/fibrinolysis system in the brain trends in Neurosciences  tPA uPAPlasminogenPAI-1 PAI-2Plasmin   Fibrinclot FibrinolysisCoagulation PlateletaggregationFibrinogenThrombinTFFactor VaFactor VIICa 2+ PAR1PAR3Antithrombin III (Transglutaminase) +++ +++++++++++Factor XIIIaProthrombin –  –  –  –  – – Factor XaThrombomodulinprotein C a 2 -antiplasmin Fig. I. Components of the coagulation and fibrinolysis pathways present in the CNS. Components for whichprotein or mRNA are produced in neurons or glia are highlighted in grey.  TINS Vol. 23, No. 9, 2000  401 walls of capillaries and arterioles inthe brain contain PN1, suggestingthat it is well positioned to coun-teract extravasated thrombin 28 .Moreover, PN1 is neuroprotectivein a variety of experimental prepa-rations 29,30 and, interestingly, lev-els of PN1 are increased in hippo-campal CA1 glial cells followingexperimental ischemia 31 and nervelesion 24 . Neuroserpin, which isfound mainly in neurons, effec-tively inhibits tPA. Entry of serine proteases intobrain tissue BBB breakdown associated withcerebrovascular insult reflects alargely nonselective increase in thepermeability of brain capillariesand tight junctions to high-mol-ecular-weight proteins (Box2).Although serine protease extrava-sation into brain tissue duringpathological situations is undocu-mented, the potential entry of ser-ine proteases into the braindeserves careful consideration inthe following three scenarios.(1) Thrombin will enter interstitial fluid during pen-etrating head wound, hemorrhagic stroke, or ruptureof cerebral aneurysms and arteriovenous malforma-tions. Preliminary data show that subdural hematomacan elevate thrombin levels in cerebrospinal fluidfrom 100 p M to 25 n M for a period of more than aweek 32 , suggesting that appreciable amounts of throm-bin can be generated and persist at sites of cerebrovas-cular injury. When bleeding occurs directly withinbrain tissue, active thrombin and other proteases willfreely penetrate the inter-neuronal spaces by diffusionuntil clotting closes off the injured vessels and throm-bin becomes depleted from the clot. Whether serpininhibitors are present at concentrations necessary tofully neutralize the effects of extravasated proteasesremains to be determined. Because prothrombin cir-culates in blood at high concentrations (~1  M ) 33 andvascular injury triggers its rapid conversion to throm-bin, direct entry of thrombin into the interstitialspace, even at 0.02% of prothrombin levels, wouldstill result in the activation of neuronal and glial PAR1(for which thrombin has an EC 50 value of ~50 p M ) 1 .Plasminogen also circulates at high concentrations inblood (2  M ) 34 , and could be cleaved by endogenoustPA upon entry into brain to form plasmin, which canalso activate PAR1 as well as cleave other substrates.(2) Serine proteases might enter brain tissue duringtherapeutic treatment of stroke or neurosurgery. Forexample, the use of thrombin-soaked sponges(Gelfoam) is common for intraoperative hemostasis,and widely considered to be safe 35 . However, thrombincould have harmful consequences independent of sur-gical outcome if, for example, PAR1 activation potenti-ated NMDA-receptor function such that cytotoxic lev-els of Ca 2  entered certain neuronal subpopulations 14 .Results from one animal model used to explore theeffects of thrombin-soaked sponges suggests thatthrombin might precipitate edema and swelling 33 . Theuse of recombinant tPA to establish reperfusion of ischemic tissue after occlusive stroke could also elevatebrain levels of tPA and plasmin, given the increasedpermeability of the BBB to high-molecular-weight pro-teins during ischemia (Box2). The dramatic reductionin neuronal death in rodent models of ischemia causedby the deletion of the gene for tPA (Refs 36,37) high-lights the importance of studying the effects of tPA onneuronal viability further.(3) Serine protease extravasation might occur fol-lowing disruption of the BBB during pregnancy-related hypertension (pre-eclampsia), status epilepti-cus, occlusive stroke, infection, inflammation andclosed head injury. Although historically there are fewdirect measurements of blood-derived serine proteasesin brain tissue in these conditions because of proteaseinstability and previous lack of appreciation for theirpotential role in the CNS, post-ischemic extravascularappearance of larger markers such as fluorescein iso-thiocyanate-conjugated dextan (71.2 kDa) and albu-min (66 kDa) suggests that extravasation of proteinssuch as thrombin, plasminogen and other blood pro-teases are possible (Box2). Moreover, some serine pro-teases themselves might increase BBB permeability 38 . Pathological effects of serine proteases on neuronsand glial cells The molecular and cellular processes that lead toneuronal death following brain trauma and cer-ebrovascular insult are diverse and currently underintense investigation. Serine proteases trigger a varietyof effects in neurons and glia that are often associatedwith brain damage (Fig.3). Neurite outgrowth PAR1 activation causes neurite retraction in mouseneuroblastoma cells and chick motoneurons 39–41 whereas plasminogen and plasmin appear to enhanceneurite outgrowth 42 . PAR1-stimulated neurite retraction M.B. Gingrich and S.F. Traynelis  – Serine proteases in the brain V IEWPOINT GTP G q PLCG i AC    PAR1 bg GTP PO 4 PO 4 GTP trends in Neurosciences  GTP   Thrombin G q  G q bg G i bg (a) PAR1  GDPGDP N 1  N 1 N 1 PLCG i AC (b) PAR1 bg cAMP [Ca] i2+ ,PKC,TyrK[Ca] i2+ ,PKC,TyrK GTP (c) cAMPSFLLRN 37323456 TLDPR/ SFLLRN PNDKYEPFSSKGR/ SLIGKV DGTSHVTGTLPIK/TFRGAPPNSFEEFPLPAPR/ GYPGQV CANDSDTL hPAR1hPAR2hPAR3hPAR4GTP N 2 Fig. 1. Mechanism and consequences of protease-activated receptor 1 activation. (a) The thrombin receptor pro-tease-activated receptor 1(PAR1) is a seven-transmembrane receptor that associates with active thrombin through a fibrinogen/hirudin-binding domain, and is subsequently cleaved by thrombin at the recognition sequence LDPRS after Arg41. The sequence around the biochemically determined cleavage sites of four PARs is shown below. Bold letters denote sequences that correspond to peptides capable of activating the receptor directly; underlined amino acids rep-resent hirudin-like domains. (b) Cleavage of PAR1 creates a new N  42 -terminus (N  2  ) starting at Ser42, the first six residues of which can act as a tethered ligand to activate the thrombin receptor. Thrombin-receptor activation in neur-ons and glial cells can inhibit adenylyl cyclase (AC) through G i   resulting in decreased production of cAMP. Thrombin-receptor activation can also activate phospholipase C (PLC) through G q   . stimulating the formation of inositol (1,4,5)-trisphosphate, resulting in an increase in intracellular Ca  2   , and stimulation of Ca  2  -sensitive isoforms of protein kinase C (PKC). Non-receptor tyrosine kinases (TyrK) can also be activated. Liberation of G   subunits of heterotrimeric G-pro-tein complexes might have additional consequences, but have not been studied for this receptor. Following activation,receptor phosphorylation (shown by PO 4  ) occurs by G-protein-coupled receptor kinase and other kinases. (c) Uncleaved PAR1 can be activated directly by small peptides (e.g. SFLLRN) that match the new N  42 -terminus unmasked by throm-bin cleavage. These peptides can initiate the same signaling cascades as thrombin cleavage.  402 TINS Vol. 23, No. 9, 2000 in NG108-15 neuroblastoma cells is blocked by ADP-ribosylation of the Rho GTP-binding protein 8,20 . Thus,in pathological situations in which neuronal connec-tions are disrupted, thrombin might antagonize theability of neurites to make appropriate connections bycausing retraction of neuronal processes. Studies of granule-cell mossy-fiber elongation, development of the neuromuscular junction and development of dopaminergic neurons in vitro provide parallel supportfor the involvement of tPA (Ref. 43), thrombin 44 andPAR1 (Ref. 45) in synapse and neurite remodeling. Glial proliferation Glial scars are often associated with brain injury, andproliferating glia within these scars form a barriertoregenerating axons, confounding neural repair.Picomolar concentrations of thrombin, as well as thePAR1-agonist peptide, stimulate astrocyte proliferationand reverse astrocyte stellation in culture, indicatingthat thrombin extravasated from vasculature might bemitotic in vivo 16,46 . Several proteins appear to be rapidlyand transiently phosphorylated on tyrosine residuesfollowing PAR1 activation in cultured astrocytes.Furthermore, the tyrosine kinase inhibitor, herbimycinA, and the kinase inhibitors, staurosporine and H7, canblock thrombin-mediated cell proliferation 16,47 .Thrombin has also been proposed to upregulate glialexpression of thrombomodulin in in vitro models of glial injury 48 . Whole-animal studies also suggest thatthrombin can induce astrogliosis. Infusion of throm-bin into the rat caudate nucleus causes reactive gliosis,infiltration of inflammatory cells and induction of angiogenesis 49,50 . These features resemble the inflam-mation and glial-scar formation typical of cerebrovas-cular damage in traumatic head injury. Edema, neuronal survival and apoptosis Thrombin and its endogenous inhibitor, PN1, canalso influence neuronal survival in in vitro and in vivo models of brain insult and ischemia. Thrombin appearsto trigger edema, which can damage both white andgray matter 33,51 . Thrombin-induced edema is potenti-ated by thrombolytics such as tPA, which can competefor endogenous thrombin inhibitors such as PN1(Ref.52). Although application of low concentrationsof thrombin (  1 n M ) can protect cultured glial cells andneurons from a variety of metabolic insults, highthrombin concentrations can exacerbate cell death 30,53 .Consistent with these in vitro studies, administration of thrombin at low concentrations (50 p M ) is neuroprotec-tive in animal models of ischemia, but exacerbatesischemic death at higher concentrations (50 n M ) 54 .Recent evidence suggests that PAR1 activation caninduce cell death in hippocampal cultures andmotoneurons in the developing avian spinal cord; celldeath was accompanied by signs of apoptosis and couldbe blocked by caspase inhibitors 17,41,55 . Thrombin-induced apoptosis was blocked by tyrosine kinaseinhibitors and inhibitors of RhoA, a GTP-binding pro-tein known to be involved in apoptotic pathways. Hyperexcitability and seizure Post-traumatic seizures that follow stroke and braininjury are often associated with hemorrhage into braintissue. Epilepsy complicates 7% of civilian head-traumapatients and up to 34% of combat casualties. About afifth of individuals with severe head injuries sufferunprovoked seizures at some time in the 30-year periodfollowing the srcinal injury. Robust indicators of post-traumatic epilepsy include severity of head injury,occurrence of an early seizure, subdural hematoma andintracerebral hemorrhage 56–58 . In addition, 2–10% of patients with acute cerebrovascular disease and17–25% of patients with intracerebral bleeding of unknown cause suffer seizures 59–62 . Three recent reportssupport the possibility that serine-protease extravasa-tion during intracerebral hemorrhage, through acti-vation of PARs or cleavage of other substrates, con-tributes to the cascade of events that underliepost-traumatic seizures. First, Lee et al. 63 infused micro-spheres the size of red blood cells, preincubated with orwithout quantities of thrombin found in a typicalhematoma, into the basal ganglia of rats. Focal motorseizures were observed immediately after recovery fromanesthesia only in the rats that received thrombin,with resolution within one to two hours. Althoughiron from heme 64 and synaptic rearrangements 65 arethought to be involved in the expression of post-trau-matic seizures and post-traumatic epilepsy, this reportproposes that a component of cerebral hematomas(thrombin) could also contribute to post-traumaticseizure in individuals with intracerebral hemorrhage. M.B. Gingrich and S.F. Traynelis  – Serine proteases in the brain V IEWPOINT trends in Neurosciences  Mo5RF Sp5VCA4vLCCA2SGSOSPSR (b)(a) Fig. 2. PAR1 expression in the CNS. (a) Hybridization of a cRNA probe for PAR1 to a rostral section of rat hippocampus shows prominent expression in the stratum granulosum (SG) of the dentate gyrus and dif-fuse expression in stratum pyramidale (SP) of the CA1 and CA2 in a dark-field, low-magnification photomicrograph. Small clusters of hybridization were observed throughout the CA1 region in strata radia-tum (SR) and oriens (SO) and subiculum, but not in CA3. Scale bar,400  m. (b) mRNA for the PAR1 thrombin receptor is widely distributed in the cerebellum and pontine tegmentum, as shown by dark-field, low-magnification photomicrographs of  in situ hybridization of PAR1 cRNAprobes in coronal tissue sections from rat. Striking hybridization was observed in several cell layers, including the cerebellar Purkinje cell layer,tegmental and vestibular nuclei along the floor of the fourth ventricle (4v), motor nucleus of the trigeminal nerve (Mo5), ventral cochlear nucleus (VCA) and locus coeruleus (LC). Labeled cells are also visible inthe reticular formation (RF) and the spinal trigeminal nucleus (Sp5).Scale bar, 650  m Reproduced, with permission, from Ref. 23.  TINS Vol. 23, No. 9, 2000  403 Second, NMDA-receptors are excitatory synaptic lig-and-gated ion channels that participate in seizure ini-tiation and maintenance. The finding that PAR1 activation can potentiate NMDA-receptor-elicitedresponses of CA1 pyramidal cells in hippocampal slicesand in recombinant systems (Fig.4) might underlie theability of thrombin to increase neuronal excitability tothe point of seizure 14 . In addition, reduction of GABA-mediated inhibition by plasmin generated bytPA-mediated cleavage of extravasated plasminogen 66 could increase neuronal excitability synergistically.Third, consistent with the ability of thrombin to pre-cipitate seizure 63 , mice engineered to lack the serine-protease inhibitor PN1 have an increased susceptibility M.B. Gingrich and S.F. Traynelis  – Serine proteases in the brain V IEWPOINT The blood–brain barrier (BBB) comprises endothelial cells, withtheir connecting tight junctions, and astrocytic perivascular endfeetthat surround brain capillaries. This barrier can be disrupted bymechanical and hemostatic forces, as well as situations in whichendothelial-cell health is compromised. The processes and conse-quences surrounding BBB breakdown during brain injury, stroke,status epilepticus or pregnancy-related hypertension are varied andcomplicated. Not only is extravasation of small molecules and bloodproteins likely to occur rapidly during ischemia (Table I), but alsoinflammatory responses and endothelial ultrastructural changesoccur within minutes of the ischemic episode a–c . Thus, experimentsand arguments that attempt to single out one process as importantfor cell survival from the many associated with BBB breakdownrequire convincing proof and thus should be viewed with cautiousconsideration. Similarly, the possible entry of any blood constituentinto the brain should not readily be dismissed until its absence inbrain, following increases in vascular permeability, is proven. TableI lists some of the experimental models of cerebral vascular insult forwhich changes in blood–brain permeability have been detected.These results suggest that blood–brain permeability will also beincreased in humans during the initial hours following stroke. References a Selmaj, K.(1996) Pathophysiology of the blood–brain barrier. Springer Semin. Immunopathol. 18, 57–73 b Staddon, J.M. and Rubin, L.L. (1996) Cell adhesion, cell junctions andblood–brain barrier. Curr. Opin. Neurobiol . 6, 622–627 c Tomita, M. and Fukuuchi, Y. (1996) Leukocytes, macrophages and secondarybrain damage following cerebral ischemia.  Acta Neurochir. (Suppl.) 66, 32–39 d Laursen, H. et al . (1993) Cerebrovascular permeability and brain edema.  Acta Neuropathol. 86, 378–385 e Belayev, L. et al .(1995) Post-ischemic administration of HU-211, a novelnon-competitive NMDA antagonist, protects against blood–brain barrierdisruption in photochemical cortical infarction in rats: a quantitativestudy.  Brain Res. 702, 266–270 f  Yang, G-Y. et al .(1994) Experimental intracerebral hemorrhage:relationship between brain edema, blood flow, and blood–brain barrierpermeability in rats.  J. Neurosurg. 81, 93–102 g  Preston, E. and Foster, D.O. (1997) Evidence for pore-like opening of theblood–brain barrier following forebrain ischemia in rats.  Brain Res .761,4–10 h Baskaya, M.K. et al . (1997) Protective effects of ifenprodil on ischemicinjury size, blood–brain barrier breakdown, and edema formation in focalcerebral ischemia. Neurosurg. 40, 364–371 i Dogan, A. et al . (1997) Effects of ifenprodil, a polyamine site NMDAreceptor antagonist, on reperfusion injury after transient focal cerebralischemia.  J. Neurosurg. 87, 921–926 j Wang, Y.F. et al .(1998) Tissue plasminogen activator (tPA) increasesneuronal damage after focal cerebral ischemia in wild-type andtPA-deficient mice. Nat. Med. 4, 228–231 k Du, C. et al .(1996) Dextrorphan reduces infarct volume, vascularinjury, and brain edema after ischemic brain injury.  J. Neurotrauma 13,215–221 l Uyama, O. et al . (1988) Quantitative evaluation of vascular permeabilityin the gerbil brain after transient ischemia using evans blue fluorescence.  J. Cereb. Blood Flow Metab . 8, 282–284 m Preston, E. et al. (1993) Three openings of the blood–brain barrierproduced by forebrain ischemia in the rat. Neurosci. Lett  . 149, 75–78 n Harlt, R. et al .(1997) Blood–brain barrier breakdown occurs early aftertraumatic brain injury and is not related to white blood cell adherence.  Acta Neurochir. (Suppl.) 70, 240–242 o Barzo, P. et al . (1997) Acute blood–brain barrier changes in experimentalclosed head injury as measured by MRI and Gd-DTPA.  Acta Neurochir.(Suppl.) 70, 243–246 p Sato, S. et al . (1997) Blood-brain barrier disruption, HSP70 expression andapoptosis due to 3-nitropropionic acid, a mitochondrial toxin.  ActaNeurochir. (Suppl.) 70, 237–239 q Dietrich, W.D. et al . (1992) Intraventricular infusion of N -methyl- D -aspartate.  Acta Neuropathol. 8 4, 621–629 r  Correale, J. et al . (1998) Status epilepticus increases CSF levels of neuron-specific enolase and alters blood–brain barrier. Neurology  50, 1388–1391 s Skultetyova, I. et al . (1998) Stress-induced increase in blood–brain barrierpermeability in control and monosodium glutamate-treated rats.  Brain Res. Bull. 45, 175–178 Box 2.Changes in blood–brain barrier integrity after cerebral insult TABLE 1. Pathological situations associated with blood–brain barrier breakdownCerebral insultMarkerMolecularDuration (h)AnimalRefs weight(kDa) Cortical photochemical thrombosis d Albumin66.02–24RatdCortical photochemical thrombosis e Evans blue1.024RateIntracerebral hemorrhage f   -aminoisobutyric acid0.112–48Ratf Focal ischemia/hypotension g Inulin5.06RatgFocal ischemia (permanent) h Evans blue1.0ImmediateCathTransient (180 min) ischemia i Evans blue1.0ImmediateRatiTransient (120–180 min) ischemia [67]tPA administered IV70.024 tPA  /  micejTransient (90 min) ischemia  j Dextran71.04Ratk Transient (60 min) ischemia k  Evans blue1.01–3GerbillTransient (10 min) ischemia l Sucrose0.30–24RatmFluid percussion brain injury m Flourescein0.31–6RabbitnClosed head injury n Gd-DPTA (MRI)0.60.5RatoClosed head injury/hypoxia n Gd-DPTA (MRI)0.60.5RatoMitochondrial dysfunction o Evans blue1.03–72RatpVentricular injection of NMDA p Horse radish peroxidase44.00.25RatqStatus epilepticus q Albumin66.024HumanrImmobilization stress r Albumin66.0ImmediateRats Abbreviations: Gd-DPTA, gadolinium-DPTA; tPA, tissue plasminogen activation.
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks