MAPK and pro-inflammatory mediators in the walls of brain blood vessels following cerebral ischemia

University dissertation from Lund University, Faculty of Medicine

Abstract: INTRODUCTION Stroke is a serious neurological disease which may lead to death and severe disability [1, 2]. There are two major types of stroke: ischemic and hemorrhagic stroke. Both are associated with disruption of blood flow to a part of the brain with rapid depletion of cellular energy and oxygen, resulting in ionic disturbances and eventually neuronal cell death [3]. The pathologic process that develops after stroke is divided into acute (within hours), sub-acute (hours to days), and chronic (days to months) phases [4, 5]. Obviously, the most effective therapy requires the earliest possible intervention e.g. with removal of a thrombus. However, no specific treatment, apart from thrombolysis, that acts effectively to protect the neurons during the acute phase has yet been developed. Experimental and clinical data show an acute and prolonged inflammatory response in the brain after a stroke. Several investigators have reported that inflammation evolves within a few hours after stroke, and plays an important role in the development of the cerebral lesions [6]. This inflammatory reaction involves activation of resident cells (mainly microglia), infiltration and accumulation of various inflammatory cells (including neutrophils, leukocytes, monocytes, macrophages), and production of pro-inflammatory mediators in the injured brain areas [6, 7]. It has been established that the inflammatory reaction triggered by stroke affects not only the neuronal tissue itself but has impact also on the cerebral arteries [7]. Stroke is a vascular disease and despite extensive research in the area, the physiology and pathophysiology of the neurovascular unit, the complex network of endothelial cells, smooth muscle cells, inflammatory cells and mediators are not fully understood, which is necessary in order to develop effective therapies. The aim of the present thesis was to examine the role of pro-inflammatory mediators in cerebrovascular pathophysiology following stroke. The main focus was directed towards the expression and production of cytokines and inducible nitric oxide synthase (iNOS), the activation of matrix metalloproteinases (MMPs) and mitogen activated protein kinase (MAPK) pathway because microarray work [8] and published data [9] primarily pointed at these. These parameters and the relationships between them were studied in the cerebrovascular walls after ischemic and hemorrhagic strokes. This study lends further support to the view that inflammatory mediators are important contributing factors in brain injury after stroke. It provides evidence that blocking the intracellular signaling pathways involved in the transcription of these mediators may have therapeutic potential, as it may prevent or at least attenuate the inflammatory processes elicited by stroke. Ischemic stroke Ischemic stroke is the most common type of stroke (85% of cases). It is caused by a transient or permanent occlusion of a cerebral artery most often by a thrombus or an embolus [10, 11]. When an ischemic stroke occurs, blood flow to an area of the brain is reduced and the brain cells are starved of oxygen and nutrients, which quickly leads to neuronal cell death and the development of an infarct. The infarct region is divided into two parts: 1) A central part or an ischemic core, where the neurons die and have no chance to survive without rapid reperfusion. 2) A peripheral area or an ischemic penumbra, which surrounds the core [12]. Cells in the penumbra are impaired and cannot function due to compromised metabolism, but do not die immediately and have therefore become a prime target for neuroprotective treatments [13-15]. A number of neurochemical and pathophysiological events are triggered within the ischemic penumbra. As a result of energy depletion, there is disruption of ion homoeostasis, excessive release of excitatory neurotransmitters such as glutamate, calcium channel dysfunction, generation of oxidative stress and free radicals, activation of stress signaling, cell membrane disruption, inflammation, ultimately leading to necrotic and apoptotic cell death [1, 4, 15, 16]. The effect of ischemia on brain cells results not only in loss of structural integrity of brain tissue but affects also blood vessels, partly through the activation of inflammatory events and excess production of vasoconstrictor substances and increased receptor expression [17]. The early inflammatory response, which often is associated with the blood vessels, starts immediately or a few hours after the onset of the ischemia and contributes to the irreversible damage [18-21]. Currently, there are two major ways used for treating ischemic stroke: (i) Dissolution of the clot in the occluded artery by a thrombolytic drug, rt-PA (recombinant tissue-plasminogen activator) [22] and, (ii) administration of neuroprotective agents [23]. Treatment with rt-PA is limited by time and should be administered within 4.5 hours after the onset of stroke to reduce the risk of hemorrhagic transformation [24, 25]. Moreover, rt-PA is associated with the risk of disruption to the blood-brain barrier (BBB) which is due to activation of matrix metalloproteinases [26]. Despite intense research, the results obtained with neuroprotective drugs in clinical trials have not revealed positive results [27, 28]. Hemorrhagic stroke Hemorrhagic stroke (15% of all strokes) is often associated with hypertension, and is due to the rupture of an arterial aneurysm or a vascular malformation [1, 29]. Hemorrhagic stroke is divided into two categories: intracerebral and subarachnoid hemorrhage. Intracerebral hemorrhage (ICH) is due to the rupture of a small artery (arterioles) which bleeds within the brain tissue. It is often associated with chronic high blood pressure and the symptoms often begin with severe headache. Subarachnoid hemorrhage (SAH) occurs when an artery or an arterial aneurysm on the surface of the brain ruptures and bleeds into the space between the pia mater and the arachnoid (subarachnoid space) [1]. The most common cause of the SAH is the spontaneous rupture of an arterial aneurysm. This is associated with acute rise of the intracranial pressure (ICP), reduction of cerebral blood flow (CBF), rapid discharge of blood into the basal cisterns, and delayed cerebral ischemia (DCI), each of which may be fatal. The SAH is most common in women and younger people (below 55 years old). Around 50-70% of patients with SAH die or suffer severe disability, and is the cause of up to 10% of all strokes [30-33]. The disease is biphasic, with an early/short-lived phase that occurs immediately after SAH with a reduction in CBF, followed by a chronic or prolonged phase which is characterized by a varying degree of pathological contraction of cerebral arteries, known as vasospasm [34, 35]. The vasospasm (narrowing of arteries) typically occurs within 5-15 days after SAH and is present in approximately one-third of patients and is accompanied by DCI [36, 37]. It can occur not only at the site of the hemorrhage, but also in brain arteries at a distance from the bleeding. The narrowing of the cerebral vessel lumen leads to reduction in local blood flow and in cerebral metabolism, causing severe cerebral ischemia, with increase in mortality of 1.5-3 folds during the first two weeks after SAH [37-39]. Despite intense research, the pathogenesis of DCI after SAH is not well understood and no specific pharmacological treatment is available. Current treatment recommendations involve management in an intensive care unit. The blood pressure is maintained with consideration to the patient’s neurologic status. In addition, calcium channel blockers, endothelin-1 receptor antagonists, hemodynamic management and endovascular treatment are also used, but these treatments are expensive, time-consuming and only partly effective [40]. Many theories have been advanced to explain the mechanisms responsible for vasospasm and DCI that occur after SAH such as, endothelial damage [41-43], enhanced smooth muscle cell (SMC) contractility, morphologic changes in vessel walls [44], enhanced levels of free radicals [45-47], increased production and release of potent vasomotor substances such as endothelin-1 (ET-1) and angiotensin II (Ang II) [48, 49], local inflammation and immunological reactions in the vascular wall [50-52]. Yet, the exact mechanisms underlying the vasospasm and the DCI remain unknown [53]. There is evidence that the amount of blood in the subarachnoid space is related to development of vasospasm [54]. Oxyhemoglobin from extravasated blood may be an important trigger of vasospasm and DCI after SAH [55-57] by inducing inflammation [50, 58]. It may in addition correlate with structural damage to the vessel wall [59], release of spasmogenic substances, and inhibition of endothelium dependent relaxation [60, 61]. It is suggested that the extravasated blood could induce generation of free radicals that subsequently may exert a direct local toxic effect on the cerebral arteries [62, 63]. G-protein coupled receptors following stroke Recently, a novel aspect of the pathophysiology of stroke has been suggested, namely that the upregulation of vasoconstrictor receptors in the cerebral arteries after stroke may be an important mechanism in the development of the final damage [64]. Vasoconstrictor receptors such as those of angiotensin II receptor type 1 (AT1) and endothelin-1 receptor type B (ETB) belong to the seven transmembrane G-protein coupled receptor (GPCR) family [65-67]. They are upregulated in the SMCs of cerebral vessels within and associated with the ischemic region after focal ischemic stroke [68] and after SAH [69]. This results in enhanced contractility of the vessels, which further impairs local blood flow and aggravates tissue damage. Importantly, the receptor ligands (angiotensin II and endothelin-1) are formed in the cerebrovascular endothelium. In addition, contractile responses mediated by AT1 and ETB receptors were found to be increased in SMCs of human cerebral arteries after organ culture [70]. Experimental stroke induces upregulation of cerebrovascular contractile receptors in the SMCs which are caused by increased receptor gene transcription induced via activation of specific intracellular signaling pathways (such as MEK-ERK1/2 and PKC pathways) [64]. Importantly, inhibition of these signaling pathways prevents the receptor upregulation, reduces infarct volume after ischemic stroke and improves neurological score and CBF after SAH [71, 72]. This may indicate that the increased cerebrovascular contractility caused by the upregulated receptors contributes to worsening of the brain damage. Inflammation in general and following stroke Inflammation is the body's defense against injurious factors and foreign antigens, e.g., trauma, infection and toxins, and is considered to be both a beneficial and a detrimental element of a pathological process. It is a complicated and multifaceted response, characterized by acute and chronic phases [73, 74]. Among many mechanisms involved in the pathogenesis of stroke, inflammation is increasingly recognized as a key factor. However, all the mediators of the inflammatory response have not been clearly identified [6, 75-77]. There is evidence to suggest that inflammation and immune responses are involved in all three stages of the ischemic cascade, from the acute intravascular process triggered by the interruption of the blood supply to the parenchymal processes that lead to brain damage and subsequently to tissue repair. The early inflammatory response contributes to the ischemic injury, whereas late responses may represent endogenous mechanisms of recovery and repair [78] (Figure 1). When there is a switch from detrimental to beneficial effects might depend on the strength and the duration of the stroke and knowledge about the mechanisms involved is crucial for determining the time-window for effective pharmacotherapy [79]. As mentioned above, reduction in CBF after stroke can result in energy depletion and subsequent neuronal cell death. This triggers an immune response that results in activation of a variety of inflammatory cells and molecules [51, 80, 81]. In the acute phase (minutes to hours), extravasated blood following SAH (or following reperfusion after arterial occlusion in transient ischemia) induces generation of reactive oxygen species (ROS). They may stimulate ischemic cells to produce inflammatory molecules such as cytokines and chemokines which in turn may activate microglial cells and increase leukocyte infiltration. These produce more cytokines, causing an increase in adhesion molecules, which are normally required for the adherence and accumulation of leukocytes and neutrophils to vascular endothelial cells and infiltration of brain parenchyma. In the sub-acute phase (hours to days), increased activation of inflammatory cells results in further production of pro-inflammatory mediators including more cytokines, extracellular MMPs, as well as iNOS which generates nitric oxide (NO) and more ROS [79, 82]. The intravascular accumulation of leukocytes and of platelets results in occlusion of microvessels, leading to hypoxia and further increases in levels of ROS [83, 84]. Activation of mast cells and macrophages can in turn lead to release of histamine (a strong vasoactive substance) and production of more cytokines and proteases [85]. In addition, degradation of extracellular matrix components by MMPs (mostly MMP-9) leads to BBB disruption which contributes to secondary brain damage by releasing serum and blood elements into the brain tissue resulting in vasogenic brain edema and post-ischemic inflammation [83]. Disruption and permeability of the BBB can be either transient or permanent depending on severity of the insult. Permanent disruption is associated with endothelial swelling, astrocyte detachment and blood vessel rupture in the ischemic area, while transient BBB disruption is caused by endothelial hyperpermeability to macromolecules in the penumbra area. This follows a biphasic pattern with an initial opening 2-3 hours after the onset of the insult and a second opening 24-48 hours after reperfusion leading to edema and increased intracranial pressure. All these events involve pro-inflammatory cytokines, adhesion molecules and production of MMPs [86, 87]. Cerebral blood vessels are the first to be exposed to the ischemic insults and their reaction to injury sets the stage for the inflammatory response. Post-ischemic inflammation thus involves activation of microglial and endothelial cells accompanied by migration of peripheral circulating inflammatory cells into the brain such as leukocytes, neutrophils, platelet, mast cells and macrophages. These events amplify signaling along inflammatory cascades increasing the accumulation of toxic molecules that enhance the secondary damage leading to more cell stress, edema, hemorrhage and finally cell death (Figure 1) [76, 79, 84]. On the other hand, many pro-inflammatory mediators play a positive role in late stage of stroke. For example, MMPs have been reported to promote brain regeneration and neurovascular remodeling in the later repair phase [79, 88, 89]. Moreover, macrophages and microglial cells also contribute to tissue recovery by scavenging necrotic debris, by producing anti-inflammatory cytokines and by facilitating plasticity [90] (Figure 1). Yet, despite these beneficial effects there is evidence that administration of anti-inflammatory drugs may reduce infarct volume and improved outcomes in animal models of stroke [91]. On the other hand, to date, clinical trials with anti-inflammatory agents have not been able to demonstrate improved clinical outcome [92, 93]. With better knowledge about which cells and molecules that participate and which mechanisms regulate the inflammatory reactions triggered by cerebral ischemia, it may be possible to identify novel targets for suppression of inflammation following cerebral ischemia and thereby develop more effective stroke therapies. Figure 1. Main inflammatory pathways that respond to injury after a stroke. The generation of ROS and free radicals that occur after stroke triggers inflammatory responses. This involves activation of cytokines and chemokines which leads to activation of inflammatory cells such as microglia and leukocytes causing more production of inflammatory mediators (cytokines, iNOS, MMPs and more ROS) which then lead to brain edema, hemorrhage and cell death. Thus, these early inflammatory responses contribute to ischemic injury, whereas late responses may represent endogenous mechanisms of recovery and repair through activation of anti-inflammatory cytokines, scavenging necrotic debris by microglia and neurovascular remodeling by MMPs. Major inflammatory mediators in cerebral ischemia In this present thesis, I have studied the expression of some of the major cytokines (IL-1ß, IL-6, TNF-α, TNF-R1 and R2), of MMP-9 (BBB associated) and of iNOS (potential toxic molecule) in cerebral vessel walls. Increased levels and activation of these factors may lead to exacerbation of vasoconstriction, resulting in decreased CBF and enhanced neuronal damage following a stroke. Cytokines Cytokines are recognized as small proteins, generally associated with inflammation, immune activation, cell differentiation and hematopoiesis [94]. Most cytokines are pleiotropic and have multiple biologic activities that generally act over a short distance, during short periods of time and at low concentrations. They are produced and expressed by different cell types such as astrocytes, macrophages, monocytes, microglia, platelets, endothelial and smooth muscle cells, neurons, fibroblasts and neutrophils [52, 95, 96]. Normally, they have a beneficial role, but when their expression increases in an imbalanced fashion they become detrimental [97]. Evidence for the involvement of cytokines in the pathology following stroke comes from the detection of their high levels in CSF and plasma of patients [98, 99]. It is thought that increased production and activation of such cytokines in vessel walls after cerebral ischemia/reperfusion may facilitate and expand the ischemic core by inducing secondary brain damage (brain swelling, impaired microcirculation, hemorrhage and inflammation) that typically develops after a delay of hours or days after the original ischemia, trauma or SAH [100]. It is well known that cytokines are involved in the upregulation and activation of adhesion molecules, MMPs, leukocytes, microglial, increased leukocyte-endothelium interaction and increase in vasoconstrictor substances like ET-1 following cerebral ischemia [52, 76, 101]. Tumour necrosis factor-α (TNF-α), interleukin-6 (IL-6) and interleukin-1ß (IL-1ß) are the main cytokines which initiate inflammatory reactions and induce expression of other cytokines and inflammatory mediators after stroke. Ischemic brain has been shown to produce increased levels of TNF-α, IL-6 and IL-1ß, which are considered as a part of the damaging response [102]. Inhibiting the expression of these pro-inflammatory cytokines has been reported to reduce brain infarct size in animal models of stroke [103]. TNF-α TNF-α is a pleiotropic cytokine and exists as either a transmembrane or soluble protein. It is involved in the disruption of the BBB, as well as in inflammatory, thrombogenic and vascular changes associated with brain injury [104]. This cytokine promotes inflammation by stimulation of acute-phase protein secretion, enhances the permeability of endothelial cells to leukocytes, and the expression of adhesion molecules and other cytokines into the ischemic area [105, 106]. In addition, it has been suggested to stimulate angiogenesis after cerebral ischemia through induced expression of angiogenesis-related genes [107, 108]. It is known as a strong immunomediator, which is rapidly upregulated early in neuronal cells in and around the ischemic penumbra, and is associated with neuronal necrosis or apoptosis [105]. TNF-α effects are mediated via two receptors, TNF-R1 and TNF-R2, on the cell surface [109]. TNF-R1 is expressed on all cell types, can be activated by both membrane-bound and soluble forms of TNF-α and is a major signaling receptor for TNF-α. The TNF-R2 is expressed primarily on endothelial cells, responds to the membrane-bound form of TNF-α, and mediates limited biological responses [109]. There is evidence that TNF-α and its receptors may activate nuclear factor-κB (NF-κB), a transcription factor whose activation leads to expression of several genes involved in inflammation and cell proliferation [110-112]. In addition, NF-κB is involved in signaling cell death as well as cell survival, and the balance between these signals determines the toxic degree of TNF-α [112, 113]. TNF-α appears then to be not only neurotoxic but also neuroprotective. Increased TNF-α levels have been observed in brain tissue, plasma and CSF in several CNS diseases such as Alzheimer’s, multiple sclerosis and Parkinson’s [114-116]. Accordingly, a recent study demonstrated that blocking TNF-α significantly reduced infarct size after both permanent and transient MCAO, suggesting the involvement of TNF-α in neuronal cell damage [104]. In contrast, there is evidence to suggest that brain injury after ischemia becomes worse in mice lacking TNF-R1, suggesting that TNF-α mediates neuroprotection through this receptor [117]. The function of TNF-α appears to differ between brain regions. TNF-α released for instance in the striatum is considered as neurodegenerative, while release in the hippocampus has been suggested to promote neuroprotection [112]. Several investigators have suggested that the detrimental effects are activated in the early phase of the inflammatory process whereas the beneficial effects take place in the later phases [79]. IL-6 IL-6 is an endogenous and hematopoietic cytokine that plays multiple roles in the central nervous system during infection and after traumatic injuries. It is involved in induction of B-cell differentiation and helps to attract T-lymphocytes into the brain, contributing to exacerbation of the inflammatory response [79]. To exert its biological effects, IL-6 binds to its receptor, IL-6Rα, which can be either soluble or membrane-bound [118]. IL-6 is often induced together with TNF-α and IL-1ß in different conditions, and circulating IL-6 plays an important role in the induction of acute phase reactions [119]. Several studies have revealed that the expression of IL-6 is detected at an early time point, 4-6 hours after onset of ischemia, and at a later point at 24-48 hours, and that it remains detectable for up to 14 days [120, 121]. The level of IL-6 in CSF is significantly increased from days 3-6 in patients with vasospasm compared to patients with no symptoms of vasospasm, suggesting that IL-6 might be involved in inducing CVS after SAH [122]. In addition, in animal models of stroke, there is an enhanced expression of IL-6 in neuronal cells [123]. However, the exact role of IL-6 in cerebral ischemic is still unclear. For example, the high levels of IL-6 in plasma of patients with acute brain ischemia, are strongly associated with stroke severity and long-term clinical outcome [124]. On the other hand, IL-6 deficient mice show similar infarct size compared to the wild type, suggesting that it does not participate in ischemia pathogenesis [125]. IL-1ß IL-1ß is a member of the IL-1 family and is rapidly produced in the brain after cerebral ischemia [105, 126]. IL-1ß is involved in development of brain damage following cerebral ischemia and blockade of IL-1ß converting enzyme activity, reduced infarct size and improved behavioural deficit [127]. IL-1ß acts by binding to its two transmembrane receptors, type I IL-1 receptor (IL-1R1) and type II IL-1 receptor (IL-1R2) initiating signaling cascades that result in expression of inflammatory genes [128-130]. In addition, IL-1ß plays an important role in the acute stress-induced worsening of behavioural and neurological outcomes and increased infarct size after MCAO in rat [131]. A previous study indicated that IL-1ß expressed in vascular SMCs after SAH, mediates SMC apoptosis and results in enhanced aneurysm formation [132]. Several investigations have shown that IL-1ß, IL-6 and TNF-α follow approximately the same time course of expression after both global and focal cerebral ischemia. Thus, there is an early increase in the levels of TNF-α (1 h), IL-6 and IL-1ß (3-6 h) and a later increase at 2 days post-MCAO, which have been observed in cerebral cortex, striatum and hippocampus [121, 133, 134]. Matrix Metalloproteinases The MMPs represent a family of zinc-dependent proteolytic enzymes with the ability to break down extracellular matrix (ECM) proteins and to cleave other non-ECM molecules ranging from growth factors, cytokines and binding proteins to cell surface receptors [135]. MMPs are involved in extracellular matrix remodeling, wound healing and angiogenesis. They are normally found in the cytosol in an inactive form, but when cleaved by proteases, such as plasmin or other MMPs, they convert to their active form [136]. The proteolytic activity of MMPs is tightly controlled by tissue inhibitors of MMPs (TIMPs). TIMPs are specific endogenous molecules, which by binding to pro-MMPs inhibit the activation of MMPs. Imbalance between production of MMPs and TIMPs plays an important role following stroke [137, 138]. Pfefferkorn and Rosenberg showed in an experimental stroke model that inhibition of MMP reduces infarct volume, brain edema and hemorrhage [139], supporting the notion that MMPs mediate the degradation of the neurovascular matrix and thereby promoting injury of the BBB. Increased expression of MMP levels may contribute to inflammation, particularly those of MMP-2 and MMP-9 which have been shown to be upregulated in cerebral ischemia, however at different time points [140, 141]. Permanent MCAO in rats resulted in production of MMP-2 peaking at 5 days post-ischemia while MMP-9 peaked at 24-48 hours when the BBB was maximally opened. It was concluded that MMP-9 had a more significant role as compared to MMP-2 [141-144]. At the same time, MMPs seem to play yet another role in the later phase of cerebral ischemia, being involved in vascular plasticity and recovery through increase of vascular endothelial growth factor (VEGF) signals [145]. MMP-9 MMP-9 or gelatinase B is a pro-inflammatory protease that is produced during inflammatory responses by astrocytes, microglia, endothelial cells, neutrophils and macrophages. MMP-9 activates both AP-1 and NF-κB transcription factors which in turn activate pro-inflammatory cascades [146]. MMP-9 is able to degrade major components of the endothelial basal lamina (type IV collagen, laminin and fibronectin), playing an important role in the disruption of the BBB thereby contributing to the development of edema following ischemia/reperfusion [142, 143, 147]. In support, it has been reported that administration of a MMP-9 inhibitor prevented the degradation and abolished the BBB disruption after focal ischemia in rat [142]. MMP-9 is also expressed in human brain tissue after ischemic and hemorrhagic stroke [148]. Furthermore, degradation of basal lamina through activation of MMP-9 after cerebral ischemia leads to loss of astrocytes and endothelial cell contacts, resulting in hemorrhagic transformation [149]. Besides this, an early increase in MMP-9 expression has been observed in microvascular walls, which has been suggested to be the primary cause of microvasular hemorrhage after cerebral ischemia [138]. Accordingly, MMP-9 deficient animals showed a smaller infarct area in an experimental stroke model, as compared to wild type controls [150]. Additionally, it has been reported that the levels of MMP-9 in plasma appear to correlate with infarct volume and severity of stroke [151]. iNOS Another potent biological molecule that is expressed during inflammatory reactions in the CNS is nitric oxide (NO) which is induced by nitric oxide synthase (NOS) [152]. NO is an important signaling molecule, involved in numerous physiological processes such as neuronal communication, host defense and regulation of vascular tone. Three different isoforms of NOS exist: neuronal NOS (nNOS), endothelial NOS (eNOS) and the inducible form (iNOS). nNOS and eNOS are constitutively expressed and calcium-dependent, whereas iNOS is expressed after immunologic challenge and neuronal injury and is calcium-dependent under most circumstances [153]. Activation of iNOS produces toxic levels of nitric oxide and is considered one of the key inflammatory mediators produced by different cells [76, 154]. There is evidence that iNOS can enhance glutamate release, resulting in ATP depletion in the ischemic infarct area after transient focal cerebral ischemia in rats [155]. In addition, it has been reported that iNOS is not present in the CNS under physiologic conditions, but its expression can be stimulated by cytokines [51]. The activity of iNOS is strongly linked to that of COX enzymes because of the interaction between the two systems. There is evidence that the production of iNOS increases with the activity of COX-2, which results in generation of more free radicals [52]. Moreover, iNOS may in addition cause DNA damage in cerebral ischemia through the formation of peroxynitrite [156]. A previous study has revealed that iNOS expression is significantly increased in vascular endothelial and smooth muscle cells at 7 days post-SAH, and that its intensity is greatest in animals with angiographic vasospasm [157]. Furthermore, the mRNA level of iNOS is increased mostly in vascular tissue at 1-7 days in different experimental SAH models [158]. Expression of iNOS has also been reported in both permanent and transient MCAO at 12-48 hours post-ischemia in inflammatory cells of the brain parenchyma and in cerebral blood vessels [159, 160]. The involvement of iNOS in cerebral ischemia is confirmed by the observation that mice lacking the iNOS gene have significantly reduced infarct volumes compared with wild-type controls [161]. Accordingly, administrations of iNOS blockers after cerebral ischemia attenuate the damage, decrease the infarct volume and improve neurological outcome [162, 163]. Intracellular Signaling Cerebral ischemia activates several cell signaling pathways that are crucial for cell survival or damage, initiating complex cascades of events at genomic, molecular and cellular levels in all types of cells in the CNS. Inflammation, which occurs after both ischemic and hemorrhagic stroke, is very important in this context and may participate as a further enhancer in the ischemic cells [164]. Studies of signal transduction pathways that regulate the inflammatory genes have mostly focused on mitogen activated protein kinases (MAPKs), one of the cascades activated in response to cerebral ischemia [165]. In this thesis, the attention has been directed towards evaluation of the role of Raf-MEK-ERK1/2 signaling transducers in regulating the expression of pro-inflammatory mediators in cerebral vessels following both ischemia and hemorrhagic stroke. The work provides some clues to explaining beneficial effects observed with inhibitors of these cascades in experimental models. Mitogen activated protein kinases MAPKs are involved in the transduction of cellular responses, mediating signaling from the extracellular environment to the nucleus and other intracellular targets [166, 167]. In response to extracellular stimuli, MAPKs regulate a broad range of intracellular activities from metabolism, motility, inflammation, differentiation and proliferation to cell death and survival [168]. The transduction of signals is made through activation of protein kinases and protein phosphatases [168, 169]. The MAPK family consists of three major groups, including extracellular signal-regulated kinase (ERK), p38 and stress-activated protein kinase c-Jun N-terminal kinase (SAPK/JNK). Each MAPK signaling pathway contains a three-tiered kinase cascade comprising a MAPKKK, that can activate MAPKK, which in turn activates and phosphorylates MAPKs [170]. Activation of MAPK pathways regulates the activity of a number of transcription factors that are present in the cytoplasm or the cell nucleus, such as Elk-1, ATF-2, C-Myc, NF-κB, and AP-1 components, as well as c-Fos and c-Jun [170, 171]. Activation of these transcription factors leads in turn to the expression of target genes, resulting in biological responses. It has been demonstrated that various MAPKs have overlapping functions, with the same transcription factor sometimes being activated by two or more MAPKs [172]. ERK1/2 is thus involved in differentiation, proliferation, meiosis, learning and memory in nerve cells and is activated by oxidative stress and mitogenic stimuli such as growth factors, cytokines and GPCRs [173-175]. In addition to the ERK pathway, the p38 and JNK pathways have been demonstrated to be involved in inflammation, cell survival and apoptosis, and can be activated by inflammatory cytokines and changes in shear stress. These kinases are activated by phosphorylation on both threonine and tyrosine residues, which may phosphorylate intracellular enzymes and transcription factors [172, 176]. The balance between ERK and p38-JNK has been suggested to regulate cell fate, mediating survival or death [177]. MAPK and inflammation in cerebral ischemia As mentioned before, cerebral ischemia is a pathophysiological condition caused by decreases in blood supply to the brain that results in deprivation of oxygen and glucose, leading eventually to cell death, inflammation and tissue repair [178]. In response to inflammatory stimuli that activate macrophages and initiate leukocyte infiltration, intracellular signaling pathways are activated and carry the signals needed to further activate the production of inflammatory mediators [179]. Cytokines such as TNF-α, IL-1ß and IL-6 can act as intracellular messengers because they have low molecular weights [180]. They act through binding to their respective receptors and/or Toll like receptors. Activation of the receptors triggers major intracellular signaling pathways, leading to activation of transcription factors such as NF-κB and AP-1, which in turn produce more cytokines involved in secondary damage [179]. Regulation of the expression of these genes by MAPK signaling, especially via ERK1/2, plays important roles in cerebra ischemia [167]. Several investigators have suggested a role for the ERK pathway in the regulation of cytokine expression following cerebral ischemia. For example, studies have shown that TNF-α can increase the permeability of the BBB via activation of the ERK1/2 pathway and increase the expression of TNF-R1 and TNF-R2. Treatment with a MEK1/2 inhibitor inactivates this signaling pathway and decreases the expression of the TNF receptors [181, 182]. However, it has been suggested that the main biological response to p38 activation involves the production and activation of inflammatory genes such as cytokines, COX2 and collagenase-1, while, inhibition of p38 may reduce pro-inflammatory cytokines in several inflammatory cells [179, 183]. Activation of all three major MAPK pathways has been reported in experimental cerebral ischemia [9, 175, 184, 185] and activation of ERK1/2 is reported in humans after ischemic stroke [186]. Activation of JNK and p38 appears to be detrimental after a stroke and their inhibition decreases infarct size and prevents neuronal apoptosis [177, 187-189], while ERK1/2 activation can be both beneficial and detrimental [190]. There is evidence pointing at the activation of ERK1/2 in cerebral arteries after MCAO [72], after SAH [49] and in cultures [185]. In addition, several studies have reported on the involvement of the MEK/ERK/MAPK pathway in the regulation of CVS after experimental SAH [191, 192]. It has been suggested that activation of the ERK1/2 pathway increases neurological damage by increasing ROS and oxidative stress-related cell death, promoting inflammation after stroke [190]. ERK1/2 activity may also stimulate inflammation by upregulation of IL-1ß, which results in necrosis [180]. Wang and co-workers revealed that the activation of ERK1/2 in the brain following ischemia/reperfusion is associated with cell death and brain injury while inhibition of ERK1/2 by a specific MEK1/2 inhibitor provided neuroprotection in cerebral ischemia through suppression of IL-1ß expression [193]. Administration of inhibitors of the MEK/ERK1/2 pathway has been found to attenuate ischemic injury and improve neurological outcome [68, 194, 195]. On the other hand, ERK1/2 activity may also block apoptosis by increasing the level of the anti-apoptotic protein BCL-2 or by inhibiting the pro-apoptotic protein Bad [168]. Moreover, it has been reported that ERK1/2 mediates protection after cerebral hypoxic-ischemic injury through activation of neurotrophins such as brain derived neurotrophic factor (BDNF), resulting in survival of neurons in the neonatal brain [196]. According to earlier studies, the protein kinase ERK1/2 is activated in the early phase after stroke [71]. Here, in this work, we suggest that this activation is probably an early “switch-on” mechanism involved in the increased release or formation of vasoconstrictor receptors, cytokines and of other mediators. Cytokine stimulation, in turn, activates p38 and JNK pathways which results in induction of more inflammatory genes, causing more damage. Therefore, early inhibition of this pathway may provide novel interesting targets for anti-inflammatory therapy following stroke (Figure 2). Figure 2. Regulation and production of pro-inflammatory mediators through early activation of the ERK1/2 MAPK pathway, which is stimulated by stress, cytokines and CGRP. Hypothesis We hypothesize that the increase in expression of cerebrovascular pro-inflammatory mediators that is seen after cerebral ischemia occurs via increased inflammatory gene transcription induced via activation of MAPK-MEK-ERK1/2 signaling pathway. This results in reduced CBF, larger brain damage and worsened neurological function. By blocking this signaling pathway in time, it will prevent the enhanced transcription of inflammatory genes and is associated with improved outcome after stroke. AIMS The general aim of this thesis is to investigate the role of pro-inflammatory mediators and their regulation in the wall of cerebral vessels following cerebral ischemia. More specifically: • To investigate the expression of MMP-9 and TIMP-1 in cerebrovascular SMC following focal cerebral ischemia and to determine if their expression is regulated via the MEK/ERK pathway. • To investigate the expression of pro-inflammatory cytokines in the walls of cerebral vessels after MCAO and to compare the inhibition of the inflammatory reaction with 1) a specific MEK1/2 inhibitor (U0126) to block transcription, and 2) a combined blockade of the AT1 and ETA receptors. • To determine if the time-course and upregulation of pro-inflammatory mediators in the walls of cerebral arteries and microvessels after SAH is associated with the MEK-ERK1/2 pathway. • To investigate if treatment with specific Raf or/and MEK1/2 inhibitors given as late as 6 hours after induction of SAH would prevent the upregulation of pro-inflammatory mediators, prevent SAH-induced decrease in CBF and improve functional neurological outcome. • To examine if the expression of TNF-α and TNF receptors in the wall of cerebral arteries in two in vivo models (MCAO, SAH) and in an in vitro model of isolated cerebral arteries segments (organ culture), is regulated via the MEK/ERK pathway. GENERAL METHODS Animal surgery procedure MCAO model (papers I, II and V) Transient middle cerebral artery occlusion (MCAO) was induced in male Wistar rats by an intraluminal filament technique described by Memezawa et al [197]. The rats were housed under controlled temperature and humidity with free access to water and food. Anesthesia was induced using 4.5 % halothane or isoflurane in N2O:O2 (70:30); thereafter the rats were kept anesthetized by inhalation of 1.5 % halothane or isoflurane on a mask. To confirm a proper occlusion of the right MCA, a laser-Doppler probe was fixed on the skull measuring regional cortical blood flow. A polyethylene catheter was inserted into a tail artery for measurements of mean arterial blood pressure (MAP), pH, pO2, pCO2, and plasma glucose. A rectal temperature probe connected to a homeothermal blanket was inserted for maintenance of a body temperature of 37 ° C during the operational procedure. An incision was made in the midline of the neck and the right common, external and internal carotid arteries were exposed. The common and external carotid arteries were permanently ligated by sutures. A filament was inserted into the internal carotid artery via an incision in the common carotid artery, and further advanced until the rounded tip reached the entrance of the right MCA. The resulting occlusion was made visible by laser-Doppler as an abrupt reduction of cerebral blood flow of about 80-90 %. Finally, the filament was fixed by a suture and the rats were allowed to wake up. Two hours after occlusion, the rats were re-anesthetized to allow for withdrawal of the filament, and subsequently achieve reperfusion as verified by laser-Doppler recording [198, 199]. SAH model (papers III, IV and V) Subarachnoid hemorrhage (SAH) was induced in male Sprague-Dawley rats by a model originally described by Svendgaard et al [200] and in detail by Prunell et al [201]. In this model, fresh and non-heparinized blood is administered into the subarachnoid space at an intracranial pressure (ICP) equal to the mean arterial pressure. The rats were anesthetized, intubated and artificially ventilated with inhalation of 0.5-1.5% halothane (paper III) or 1-2% isoflurane (paper IV) in N2O/O2 (70:30) during the surgical procedure. Respiration was monitored by regularly withdrawing blood samples to a blood gas analyzer. A temperature probe was inserted into the rectum of each rat to record the body temperature, which was maintained at 37ºC by a heating pad. An arterial catheter was placed in the tail artery to measure blood pressure and a catheter to measure the ICP was placed in the subarachnoid space. A laser-Doppler probe was placed to measure cortical cerebral blood flow (CBF). A 27G blunt cannula with a side hole facing right was placed 6.5 mm anterior to the bregma in the midline at an angle of 30º to the vertical plane placing the tip of the needle just in front of the chiasma opticum. After 30 minutes of equilibration, 250 µl of blood was withdrawn from the tail catheter and injected manually into the prechiasmatic cistern at a pressure equal to the mean arterial blood pressure. Subsequently, rats were maintained under anesthesia for another 60 minutes in order to allow the animal to recover. The ICP catheter was cut and sealed with a removable plug 2 cm from the tip. The tail catheter, the needle and the laser-Doppler probe were removed and incisions closed. The rats were then revitalized and extubated. Drug administration U0126 (papers I, II and IV) U0126 is a specific MEK1/2 inhibitor that inhibits the ERK1/2 pathway by binding to and inhibiting the enzyme activity of MEK1/2, inhibiting thereby the activation and phosphorylation of ERK1/2 [180]. U0126 was used in both MCAO and SAH experimental models. In the MCAO model, 30 mg/kg body weight of U0126 (obtained from Sigma, St Louis, MI, U.S.A.) were administered intraperitoneally either immediately after starting reperfusion (0 hours) or at 6 or 12 hours after the start of reperfusion; in both cases, the first injection was followed by a second one at 24 hours. The animals were then sacrificed 48 hours after occlusion (papers I and II). In the SAH experimental model, 0.22 µg/kg body weight of U0126 were administered intracisternally in two strategies: (i) treatment started at 6 hours after SAH induction, was repeated at 12, 24 and 36 hours and animals were then sacrificed at 48 hours; or (ii) treatment started at 6 hours after SAH induction, was repeated at 12 and 24 hours and animals were sacrificed at 72 hours post-SAH (paper IV). The U0126 doses were chosen on the basis of previous studies [68, 202]. SB386023+b (paper III) SB386023+b is a specific B-Raf inhibitor, which inhibits MAPKKK upstream of MEK/ERK1/2 pathway [203]. The substance (20 µl of a 10-6M solution; kind gift from Dr. AA Parsons, GSK; UK), were administered intracisternally. Treatment started at different time points after SAH induction (0, 6 or12 hours) with repeated injections over a period of 36 hours. Animals were subsequently sacrificed at 48 hours post-SAH. The dose of SB386023+b was chosen based on a previous study on isolated arteries [185]. Neurological evaluation (papers I, II and IV) Neurological evaluations were performed for all survival MCAO and SAH animals. The MCAO animals were examined neurologically before recirculation and immediately before they were sacrificed, at 48 hours after MCAO, using an established scoring system described in Table 1 [198, 199]. A rotating pole test was used to evaluate gross sensorimotor function (integration and coordination of movements as well as balance) of SAH animals. This examined the ability of the animals to traverse a rotating pole, which was either steady or rotating at different speeds (3 or 10 rpm) [204]. The performance of the rat was scored according to the described in Table 2. Table 1. Score Interpretation 0 No visible deficits. 1 Contralateral forelimb flexion, when held by tail. 2 Decreased grip of contralateral forelimb. 3 Spontaneous movement in all directions, but contralateral circling if pulled by tail. 4 Spontaneous contralateral circling. 5 Death. Table 2. Score Interpretation 1 Animal is unable to balance on the pole and falls off immediately. 2 Animal balances on the pole but has severe difficulties crossing the pole and moves less than 30 cm. 3 Animal embraces the pole with the paws and does not reach the end of the pole but manages to move more than 30 cm. 4 Animal traverses the pole but embraces the pole with the paws and/or jumps with the hind legs. 5 6 Animal traverses the pole with normal posture but with more than 3-4 foot slips. Animal traverses the pole perfectly with less than 3-4 foot slips. Brain damage evaluation (papers I and II) Coronal slices (2 mm thick) were obtained from brains of MCAO operated rats and stained by 1 % 2, 3, 5-triphenyltetrazolium chloride (TTC) dissolved in buffer solution at 37˚C for 20 minutes. Following TTC staining, normal brain tissue appeared bright red, while regions of damage were pale-white. The size of the ischemic damage was calculated as a percentage of the total brain volume calculated from the slices, using the software program Brain Damage Calculator 1.1 (MB Teknikkonsult) [72]. Cerebral blood flow measurement (paper III) The SAH rat was intubated and artificially ventilated with inhalation of 0.5% to 1.5% halothane in N2O/O2 (70:30) during the surgical procedure. The anesthesia and the respiration were monitored by regularly withdrawing arterial blood samples for blood gas analysis. A catheter to measure MABP was placed in the right femoral artery and a catheter for blood sampling was placed in the left femoral artery. This catheter was connected to a constant velocity withdrawal pump for mechanical integration of tracer concentration. Another catheter was inserted in one femoral vein for injection of heparin and for infusion of the radioactive tracer. After 30 minutes of equilibration, a bolus injection of 50 µCi of 14C-iodoantipyrine 4[N-methyl-14C] was administered intravenously and subsequently, 122 µl of arterial blood was withdrawn over 20 seconds. Thereafter, the rat was decapitated, the brain removed and chilled to -50°C. The ß-radioactivity scintillation counting was performed on the blood samples with a program that included quench correction. The 14C activity in the tissue was determined on cryo-sections of the brain. The sections were exposed to X-ray films together with 14C methyl methacrylate standards and exposed for 20-30 days. The 14C content was determined in several brain regions and CBF was calculated from the brain tissue 14C activity determined by autoradiography using the equation of Sakurada [205] and Gjedde [206]. Organ culture (paper V) The organ culture has been described previously by Adner and co-workers [207]. Male Wistar rats were anesthetized with CO2 and decapitated. The brains were quickly removed and chilled in ice-cold bicarbonate buffer solution. The cerebral arteries were removed and dissected free from the brain and surrounding tissue. The artery segments were placed individually in wells with 2 ml serum-free Dulbecco’s modified Eagle’s medium (DMEM) supplemented with streptomycin (100 µg/ml) and penicillin (100 U/ml). Incubation was performed at 37°C in humidified 5 % CO2 in air for 24 or 48 hours in the presence or absence of the intracellular signal inhibitors (a NF-κB inhibitor, IMD-0354; 30 nM; the specific MEK1/2 inhibitor, U0126; 10µM, the specific B-Raf inhibitor, SB386023-b; 10µM). These substances were added to the medium in the beginning and at 24 hours of the culture. IMD-0354 was shown to specifically inhibiting the phosphorylation of IκB by IκB kinases, thus preventing NF-κB release and activation [208]. Molecular techniques Real-time PCR (paper III) Real time polymerase chain reaction (RT-PCR) is a sensitive method used for the detection of mRNA expression of a specific gene in a tissue homogenate. Cerebral arteries were immediately dissected out free from the brain and cleaned from connective tissue and blood. Total cellular RNA was extracted from the cerebral arteries using the Trizol RNA isolation kit according to the supplier’s instructions (Invitrogen, Taastrup, Denmark). Total RNA was determined using a Gene Quant Pro spectrophotometer measuring absorbance at 260/280 nm. Reverse transcription of total RNA to cDNA was performed using the GeneAmp RNA kit (Perkin-Elmer Applied Biosystems, USA) in a Perkin-Elmer DNA thermal cycler, using random hexamers as primers. Real-time quantitative PCR was performed in a GeneAmp 5700 sequence detection system using the GeneAmp SYBER® Green kit (Perkin-Elmer Applied Biosystems, USA) with the cDNA synthesized above as template. Specific primers were designed by using the Primer Express software program and no-template controls for each primer pair were included in all experiments. The real–time PCR consists of a system that is able to evaluate the amount of DNA in each PCR cycle via the detection of a fluorescent dye binding double-strand DNA. The amount of mRNA for each gene was calculated relative to the amount of the housekeeping genes elongation factor-1(EF-1) and ß-actin, which were used as endogenous standards as they are continuously expressed in cells. For detailed description of the real-time PCR method, please see the paper III. Immunohistochemistry (papers I-V) Indirect immunofluorescence staining was used for the detection and localization of specific proteins in cerebral arteries, microvessels and surrounding brain tissue. Briefly, the cerebral arteries and surrounding brain tissue were dissected out, placed into Tissue TEK, frozen on dry ice, and sectioned into 10-µm-thick slices in a cryostat. Cryostat sections were fixed for 10 minutes in ice cold acetone and thereafter rehydrated in phosphate buffer solution (PBS) containing 0.25% Triton X-100 for 15 minutes. The tissue sections were then permeabilized and blocked for 1 hour in blocking solution and thereafter were incubated over night at 4 ˚C with the primary antibody of interest. The sections were subsequently washed with PBS and incubated with the appropriate secondary antibody conjugated with a fluorophore for 1 hour at room temperature. After washing with PBS, the slides were mounted with anti-fading mounting medium and sections were photographed with a confocal microscope at the appropriate wavelengths. The same procedure was used for the negative controls except that either the primary antibody or the secondary antibody was omitted to verify that there was no autofluorescence or unspecific labeling. Fluorescence intensity was used as a semi-quantitative measure of the level of expression of the proteins in the samples and was determined using the ImageJ software ( Western blot (papers I, III and V) Western blotting is a method commonly used in order to identify and quantify a certain protein in a sample of tissue homogenate or extract. The cerebral arteries were harvested, frozen in liquid nitrogen and homogenized in cell extract denaturing buffer. Whole cell lysates were sonicated on ice and the supernatants were collected as protein samples. Total protein concentration was determined using a Bio-Rad DC kit (Bio-Rad, Hercules, CA, USA). Equal amounts of protein were loaded onto a gel and separated by sodium dodecyl sulfate-PAGE. Molecular weight markers were loaded on each gel for protein band identification. After separation, proteins were transferred to a nitrocellulose membrane by electroblotting. The membrane was then blocked for 1 hour at room temperature and incubated with the primary antibody of interest overnight at 4ºC, followed by incubation with a horseradish peroxidase (HRP)-conjugated secondary antibody for 1 hour at room temperature. The labeled proteins were developed using the LumiSensor Chemiluminescent HRP Substrate kit (GenScript, Piscataaway, NJ, USA). The membrane was visualized using a Fujifilm LAS-1000 Luminescent Image Analyzer and band intensity was quantified using Image Gauge Version 4.0. The levels of ß-actin were used as a control for the normalization of the target proteins and data were expressed as a percentage of control. Statistics Statistical analyses were performed using the nonparametric Kruskal-Wallis with Dunn´s post hoc test for comparison between more than two groups and Mann-Whitney test for comparison between two groups, using Graph Pad Prism v.5 (Graph Pad software, Inc., La Jolla, CA). Data were expressed as the mean ± standard error of the mean (S.E.M) and P-values less than 0.05 were considered significant. Ethics For MCAO the experimental procedures were approved by the University Animal Ethics Committee in Sweden (M43-07). For SAH all procedures were carried out strictly in accordance with national laws and guidelines and were approved by the Danish Animal Experimentation Inspectorate (license no. 2066/561-1139) and the Ethical Committee for Laboratory Animal Experiments at the University of Lund. RESULTS AND DISCUSSION Cerebrovascular expression of pro-inflammatory mediators following MCAO (papers I and II) MMP-9 and TIMP-1 expression The BBB plays an important role in protecting the neuronal environment. Endothelial cells of brain arteries and capillaries have tight junctions, which can restrict molecules from moving between the blood and the brain. MMP-9 has been reported to be involved in the disruption of the BBB by degrading the tight junction proteins claudin-5 and occludin between the endothelial cells and extracellular matrix molecules constituting the basal lamina surrounding the endothelial cells. When the integrity of the BBB is lost, inflammatory cells and fluid can pass to the brain, causing hemorrhage, vasogenic edema and neuronal cell death [142, 209, 210]. There exist evidence pointing at angiotensin II and endothelin-1 increase following cerebral edema. They may induce elevated MMP-9 expression in rat vascular SMCs and astrocytes through activation of AT1 and ETB receptors after focal cerebral ischemia and in culture [211, 212]. Interestingly, administration of inhibitors towards these receptors reduced MMP-9 expression and ischemic injury [211, 212]. Therefore, the aim of paper I was to examine the early changes in the expression of MMP-9 and TIMP-1 in the walls of brain vessels at 48 hours post-MCAO. Results from immunohistochemistry showed markedly enhanced expression of these proteins in the SMCs of the middle cerebral arteries (MCAs) and in associated microvessels within the ischemic region (Figure 3) and not in vessels on the contralateral side. This was confirmed with western blot analysis that showed the protein levels of MMP-9 and TIMP-1 were significantly increased in MCAs after MCAO as compared to control groups (Figure 4). The results are in agreement with a previous study that reported increased in MMP-9 mRNA levels in the MCA at 24 hours after focal ischemia [9]. Another studies confirmed the presence and increase in MMP-9 mRNA and protein levels in the ischemic region at 24 hours after MCAO with reperfusion. This was associated with reduction in tight junction proteins in cerebrovascular endothelial cells and administration of a MMP blocker and in MMP-9 knock-out animals prevented the degradation of tight junction proteins, reduced the BBB opening and vasogenic edema [142, 143]. Moreover, It has been reported that in rats with 2 hours transient MCAO, a maximally increase in MMP-9 was associated with maximal brain sucrose uptake at 48 hours after reperfusion [146]. Figure 3. Confocal microscopy images of the MCA, cerebral microvessels (Mic.V), and surrounding brain tissue (Brain). Immunofluorescence labeling corresponding to MMP-9 (A-E) or TIMP-1 (F-J). Images represent the vehicle control group (contralateral side) (A, F), MCAO plus vehicle group (epsilateral side) (B, G), MCAO plus U0126 with start at 0 hour (C, H), with start at 6 hours (D, I) or at 12 hours (E, J) groups. There was a significant increase in MMP-9 protein levels in the smooth muscle cell layer of ischemic vessels as compared to vessels from the control group. TIMP-1 expression was upregulated in SMCs and in the proximity of the adventitia layer of ischemic vessels as compared to control vessels. Treatment with U0126 starting at zero and 6 hours, but not 12 hours after occlusion, prevented the increase in MMP-9 and TIMP-1 protein expression. Figure 4. Western blot showing MMP-9 and TIMP-1 protein expression levels in the MCA 48 hours after MCAO using ß-actin as a loading control. Treatment with U0126 at 0 hour post occlusion decreased the MCAO-induced enhanced expression of MMP-9 and TIMP-1 proteins. Data are expressed as mean ± S.E.M., n = 4. 'P < 0.05, ''P < 0.01. In addition, we showed that TIMP-1 was increased 48 hours after MCAO, probably in effort to balance the elevated expression of MMP-9 after induction of MCAO. Thus, we suggest that an imbalance between MMPs and TIMPs expression following cerebral ischemia may result in opening of the BBB and increase vessels permeability in reperfusion injury, contributing to cerebral edema and more brain damage. To determine the cellular source of MMP-9 and TIMP-1, we performed co-localization studies using a SMC-actin specific antibody. MMP-9 immunoreactivity was localized to the cytoplasm in SMCs of the cerebral vessels. TIMP-1 was also localized in the SMCs of the medial layer but it was mainly located closer to the adventitia layer of the cerebral vessel walls (Figure 5). To confirm this, we performed co-localization studies using CD31 (as a marker of endothelium cells) neither MMP-9 nor TIMP-1 revealed any major co-localization with CD31; hence, the upregulation occurs in the medial layer. In addition, some vessels were studied after mechanical removal of endothelium. After this procedure the localization of the above proteins in the SMCs was still the same and the SMC localization confirmed. Interestingly, following double staining with GFAP (a selective marker of astrocytes), we noted that the expressions of MMP-9 and TIMP-1 were not associated with glial or astrocyte end-feet in the vessel walls. This confirmed and supports that the transcriptional upregulation takes place in the vascular SMCs themselves. There was a rich network of GFAP-positive astrocytes in the cerebral cortex tissue and around the microvessels. This is in agreement with a previous study that demonstrated the presence of astrocytic end-feet surrounding the microvasculature [213]. Figure 5. Double immunofluorescence staining for MMP-9 or TIMP-1 and SMCs actin in the MCA after MCAO. Photographs show the localization of MMP-9 and TIMP-1 (green) and of actin (red) in smooth muscle cells, and their co-localization (yellow fluorescence in the merged picture). Cytokines and iNOS expression associated with infarct volume and neurological scores It is well established that the neuroinflammatory process is complex and involves numerous pathways and molecules in the brain. However, relatively little information is available on the role of the cerebrovascular SMCs in this process following cerebral ischemia. Therefore, the next step was to investigate early changes in cytokine expression in the wall of cerebral vessels after focal ischemia. Previously, Vikman and co-workers revealed, using microarray and qPCR analysis, that upregulation of cytokine genes occurs in the walls of the cerebral arteries at 24 hours after cerebral ischemia and after organ culture [8, 9]. To investigate if the upregulated genes are translated to proteins, immunohistochemistry was used to detect TNF-α, IL-6 and IL-1ß protein expression in brain vessel walls at 48 hours after MCAO. We observed significantly enhanced expression of TNF-α, IL-6 and IL-1ß proteins in the walls of MCA and brain microvessels (paper II). Notably, this enhanced expression was primarily located in the cytoplasm of the SMCs (co-localization with actin), while a weak expression was in addition seen for IL-6 and IL-1ß in the endothelial cells. Taken together with the results obtained by Vikman at 24 hours, our results support the notion that a transcriptional event is involved. It is thought that cytokines such as TNF-α, IL-6 and IL-1ß are involved in the development of secondary brain damage and are associated with increased infarct size through upregulation and activation of adhesion molecules, leukocyte infiltration and MMP-9 activation [76, 96, 214]. At 48 hours after MCAO, we also revealed that acute cerebral ischemia followed by reperfusion in the rat is accompanied by an infarct volume of 25 ± 2% of total cerebrum and a poor neurological score (Figure 6). The results are in agreement with those of previous studies. In parallel there is elevated expression of TNF-α and IL-1ß in the cortex after both transient and permanent MCAO in rat [104, 215]. It has been reported that intracerebroventricular injection of antibodies against TNF-α and IL-1ß starting at 30 minutes before permanent MCAO or immediately after reperfusion following transient MCAO reduced infarct volume [104, 215]. Vila and co-workers have reported high levels of IL-6 and TNF-α in plasma and CSF of patients within the first 48 hours after ischemic stroke onset, which correlated with early neurological deterioration, raised body temperature, and a larger infarct volume [216]. In addition, stroke is associated with enhanced expression of some GPCRs that mediate vasoconstriction. We have recently revealed that cytokines can, at least in vitro, enhance this expression [217]. Thus, there may appear a link between the expression of cytokines and the brain damage. In conjunction with an inflammatory reaction there is formation of iNOS [218]. This is another inflammatory mediator which is expressed in the brain following cerebral ischemia and its expression can be stimulated by cytokines [51]. In addition, iNOS is involved in secondary brain injury through production of COX and free radicals, which are putative mediators of BBB disruption and brain edema, leading to increased infarct volume [219, 220]. Therefore, we decided to investigate if there is upregulation of iNOS in the vessel walls at 48 hours post-ischemia. In concert with the findings for cytokines and MMP-9, we noted a marked expression of iNOS in the SMCs of both MCA and microvessels on the ischemic side compared to the contralateral side. This finding is supported by a previous study by Iadecola of iNOS mRNA and protein expression in the ischemic brain at 48 hours after MCAO [221]. One important observation we made was that the upregulation of cytokines, iNOS and MMPs occurred not only in large cerebral arteries but also in associated microvessels in the affected brain region. This indicates that both large cerebral arteries and associated microvessels in the ischemic region are actively participating in inflammatory responses and production of inflammatory mediators following focal ischemic stroke. This provides the first direct evidence of an associated vascular mechanism, involving both large cerebral arteries and brain microvessels. MEK-ERK1/2 pathway regulating upregulat