Thesis (Ph. D.)--University of Wisconsin, 1943. / Typescript (carbon copy). Some plates accompanied with leaf of explanatory text. Most ill. are laid in photographs. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references (leaf [29]).

Η παρούσα εργασία εστιάζει στην παρασκευή και μελέτη ανοσολιποσωμάτων είτε για στόχευση των υποδοχέων τρανσφερρίνης του ΑΕΦ είτε με εξειδίκευση για τα Αβ πεπτίδια. Για την πρόσδεση των αντισωμάτων στην επιφάνεια των λιποσωμάτων εφαρμόσαμε μια τεχνική βασισμένη στο σύστημα πρόσδεσης βιοτίνης/στρεπταβιδίνης. Έχει πρόσφατα προταθεί ότι νανοσυστήματα που φέρουν πολλόυς προσδέτες στην επιφάνεια ίσως είναι πιο κατάλληλα για στόχευση ασθενειών ή βιολογικών φραγμών. Σε αυτές τις περιπτώσεις πρέπει να διασφαλιστεί ότι η ικανότητα στόχευσης ενός προσδέτη δεν θα επηρεαστεί από την παρουσία και δευτερου. Σκοπός είναι η παρασκευή ΟΧ-26 (κατά του υποδοχέα τρανσφερρίνης μονοκλωνικό αντίσωμα) νανολιποσωμάτων, η μελέτη διαφόρων παραγόντων κατά την παρασκευή τους και οι μελέτες πρόσληψης από μοντέλο του ΑΕΦ. Ο συγκεκριμένος τύπος ανοσολιποσωμάτων επιλέχθηκε γιατί είναι γνωστό ότι στοχεύουν σε κύτταρα που υπερεκφράζουν τον υποδοχέα της τρανσφερρίνης (TfR). Πρώτος στόχος είναι η παρασκευή ανοσολιποσωμάτων που φέρουν προσδεδεμένο στην επιφάνεια το μονοκλωνικό αντίσωμα ΟΧ-26 που στοχεύει στον υποδοχέα της τρανσφερρίνης και έγιναν μελέτες της πρόσληψης των ΟΧ-26 ανοσολιποσωμάτων από την αθανατοποιημένη κυτταρική σειρά hCMEC/D3. Εκτός από τις μελέτες πρόσληψης έγιναν και μελέτες κυτταρικού εντοπισμού του περιεχομένου των λιποσωμάτων με σκοπό να διασαφηνίσουμε το μηχανισμό πρόσληψης. Τα αποτελέσματα μας έδειξαν ότι η πρόσληψη των ανοσολιποσωμάτων από τα κύτταρα hCMEC/D3 είναι αρκετά υψηλή σε σύγκριση με την πρόσληψη ανοσολιποσωμάτων που φέρουν IgG αντίσωμα στην επιφάνειά τους. Επίσης, η δέσμευση είναι δοσοεξαρτώμενη και εξαρτάται και από την ποσότητα του αντισώματος στην επιφάνεια των λιποσωμάτων. Οσον αφορά στον μηχανισμό πρόσληψης τα αποτελέσματα δείχνουν ότι έχουμε λυσοσωματικό εντοπισμό του περιεχομένου των λιποσωμάτων συνεπώς μπορούμε να ισχυριστούμε ότι η πρόσληψη γίνεται μέσω υποδοχέα και ακολουθεί μεταφορά στα λυσοσώματα. Δεύτερος στόχος είναι η παρασκευή λιποσωμάτων με εξειδίκευση για τα Αβ πεπτίδια. Παρασκευάστηκαν λιποσώματα που φέρουν κουρκουμίνη εγκλωβισμένη στη λιπιδική διπλοστιβάδα και μελετήθηκε η ικανότητα τους να αναστέλουν τη συσσωμάτωση Αβ1-40 μονομερών ή ολιγομερών ή και να αποσσυσωματώνουν ινιδικές μορφές του Αβ1-42 πεπτιδίου. Για τον λόγο αυτό εφαρμόστηκε το πρωτόκολλο θειοφλαβίνης Τ, με τη χρήση του οποίου μπορούμε να ανιχνεύουμε την ύπαρξη β-δομών. Τα αποτελέσματα έδειξαν ότι η λιποσωμική κουρκουμίνη έχει μεγαλύτερη ικανότητα αναστολής της συσσωμάτωσης και αποσσυσωμάτωσης των δομών των Αβ πεπτιδίων σε σύγκριση με την ίδια ποσότητα ελεύθερης κουρκουμίνης. Παρασκευάστηκαν επίσης και αnti-Abeta ανοσολιποσώματα και έγιναν μελέτες προσδιορισμού της συγγένειας πρόσδεσης αυτών σε ακινητοποιημένα μονομερή Αβ πεπτίδια καθώς και σε ινίδια με την τεχνική SPR. Τα αποτελέσματα έδειξαν ότι η συγγένεια πρόσδεσης των ανοσολιποσωμάτων στα Αβ πεπτίδια και ινίδια είναι υψηλή καθώς και ότι η σταθερά δέσμευσης των ανοσολιποσωμάτων στα Αβ πεπτίδια εξαρτάται από την ποσότητα του αντισώματος στη λιποσωμική επιφάνεια. / This work focuses on the manufacture and study of immunoliposomes either for targeting of transferrin receptors of blood Brain Barrier or with specialization for Aβ peptides. For the attachment of antibodies in the surface of liposomes applied a technique based on the biotin/streptavidin linkage. It has been recently proposed that perhaps multiligant-decorated nanosystems may be more efficient to target specific diseases or biological barriers.insuch cases,it is important to be sure that the targeting potential of one ligand will not be negatively affected by the presence of the second on the same nanosystem. OX-26 (anti-transferrin momoclonal antibody) nanoliposomes werw prepared and after evaluation of several preparative aspects their brain targeting potential was studied on a cellular model of BBB. Manufactured pegylated immunoliposomes coated with the monoclonal antibody OX-26 aimed transferrin receptor and studies of uptake of OX-26 immunoliposomes from cell line hCMEC/D3 were done. Also studies of cell tracking content of immunoliposomes were done in order to clarify the mechanism of intake. Our results showed that the uptake of immunoliposomes from hCMEC/D3 cell line is quite high in comparison with the uptake of immunoliposomes coated with IgG antibody. Moreover the uptake of immunoliposomes is dosedependent and depends on the quantity of antibody in the liposomal surface. With regard to the mechanism of intake, results show that there is lysosomatic localization of the liposomal content so we can say that uptake is achievied through receptor and then followed shipment to lysosomes. In the second part, manufactured liposomes and immunoliposomes with specialization for Ab peptides. In the first case prepared liposomes with curcumin incorporated into the lipid bilayer and the ability to inhibit the linking of Aβ1-40 monomers or oligomers or even to disagreggate fibrillar forms of Aβ1-42 peptide was studied. For this purpose applied the thioflavin–T protocol, using which we can prove the existence of β-structures. The results showed that liposomal curcumin is more capable to inhibit the agreggation or to disaggregate structures of Aβ peptides compared with the same amount of free curcumin. In the second case, prepared anti-Abeta immunoliposomes and studies for the determination of the affinity for immobilized Aβ peptides monomers and fibrils were done with the SPR technique. The results showed that the affinity of immunoliposomes for Aβ monomers and fibrils is high and that the Ka of immunoliposomes in Aβ peptides depends on the amount of antibodies that is tethered on the liposomal surface.

Oxytocin release within the brain modulates several social behaviours in animals and humans. Moreover, low central oxytocin content has been linked to neuropsychiatric disorders, such as anxiety and autism. The exogenous administration of oxytocin has been proposed for therapeutic treatment, but oxytocin does not cross the blood-brain barrier (BBB) in physiologically significant amounts. An alternative approach to oxytocin administration is to stimulate central oxytocin release using melanocortins. Central administration of the naturally occurring melanocortin, α-MSH, has been shown to trigger somatodendritic oxytocin release in vitro. Unfortunately, endogenous melanocortins also do not penetrate the BBB in neuroactive amounts. In this study, I investigated whether systemic administration of synthetic melanocortin receptor 3/4 (MC3/4) agonist, Melanotan-II (MT-II), affects oxytocin neuronal activity and secretion in anaesthetised rats. I hypothesised that systemic administration of MT-II directly (centrally) acts on magnocellular oxytocin neurones to trigger somatodendritic oxytocin release from neurones of the supraoptic nucleus (SON) of the hypothalamus in vivo. Firstly, using double immunohistochemistry against Fos protein, a widely used marker for neural activity, and oxytocin, I showed that intravenous (i.v.; 1 mg/kg), but not intranasal (1 and 30 μg rat), administration of MT-II markedly induced Fos expression in magnocellular oxytocin neurones of the SON and paraventricular nuclei (PVN) of the hypothalamus, and this response was prevented by prior intracerebroventricular (i.c.v.) administration of the melanocortin antagonist, SHU-9119 (1 μg rat). In addition, brain areas receiving peripheral inputs which are involved in the regulation of oxytocin and vasopressin release were also analysed, showing that i.v. MT-II significantly increased Fos expression in the nucleus tractus solitarii (NTS), but not in circumventricular organs of the anteroventral third ventricle (AV3V) region. MT-II-induced Fos in the NTS was not prevented by the i.c.v. melanocortin antagonist. Then, using in vivo electrophysiology, I investigated whether i.v. administration of MT-II affects the electrical activity of SON neurones. Extracellular single-unit recordings from identified magnocellular neurones of the SON showed that MT-II significantly increased the firing rate in oxytocin neurones, however, no significant changes in firing rate were detected in vasopressin neurones. Finally, in vivo oxytocin release experiments showed that i.v. administration of MT-II did not trigger somatodendritic oxytocin release within the SON as measured by microdialysis and subsequent radioimmunoassay. Interestingly, the i.c.v. administration of MT-II (1 μg rat) also failed to trigger oxytocin release within the SON. The analysis of oxytocin content in plasma revealed that the change in oxytocin concentration was significantly greater in i.v. MT-II injected rats compared to vehicle-injected rats. Taken together, these results show that after i.v., but not intranasal, administration of MT-II, the activity of magnocellular neurones of the SON is increased. As previous studies showed that SON oxytocin neurones are inhibited in response to direct application of melanocortin agonists, the actions of i.v. MT-II are likely to be mediated, at least in part, indirectly by activation of inputs from the caudal brainstem.

The global occurrence of metabolic syndrome has reached epidemic proportions and is a contributing factor in the rising incidence of cognitive decline in the aging population. While pre-clinical research has advanced our understanding of many of the mechanisms underlying metabolic syndrome, animal models often do not reflect the complexity of human disease. For example, animal models that investigate the role played by diet in metabolic syndrome have generally focused on a single macronutrient, in particular fat or carbohydrate. As a result, although a balanced diet and increased physical activity are commonly recommended to treat metabolic syndrome symptomatology, their long-term cerebrovascular benefits are uncertain. To address these gaps in knowledge, a “Cafeteria” diet consisting of 16 common ultra-processed grocery store food items was used to model human metabolic syndrome in the rat. I compared rats fed a Cafeteria diet (CAF) to those fed “standard” chow (SD) as well as to a third group that underwent a switch to chow after chronic exposure to the Cafeteria diet (SWT). In a first study, I showed that three months of exposure to the Cafeteria diet produced metabolic syndrome as well as hippocampal neuroinflammation with increased microglial proliferation. These were fully reversed in SWT rats. Nonetheless, the Cafeteria diet did not worsen spatial learning and memory performance as assessed using the Barnes maze. In a second study, brain perfusion was examined using continuous arterial spin labeling magnetic resonance imaging (CASL MRI). Cortical and hippocampal resting perfusion was increased in CAF rats while cerebrovascular reactivity in response to a 10% CO2 vasodilatory challenge was reduced. Furthermore, while resting perfusion improved in SWT rats, cerebrovascular reactivity remained impaired. These cerebral blood flow outcomes were not accompanied by alterations in microvascular architecture or integrity as determined by rat endothelial cell antigen-1 (RECA-1) and immunoglobulin G (IgG) histology. Taken together, these results demonstrate that the Cafeteria diet is an effective model of metabolic syndrome that negatively impacts brain hemodynamic function. Moreover, while a dietary lifestyle intervention can recover peripheral features of metabolic syndrome, neuroinflammation, and resting perfusion, it is insufficient to completely reverse deficits in cerebrovascular reactivity. These findings are compelling as they speak to the detrimental effects of ultra-processed food consumption on cerebrovascular reserve capacity, believed to be an important factor in cognitive decline.

The blood-brain barrier BBB refers to the set of specialized endothelial cells that line the vasculature in the brain and effectively control movement of molecules into and out of the brain. While necessary for proper brain function, the BBB blocks 98% of drugs from entering the brain and is the most significant barrier to developing therapies for neurodegenerative diseases. Active transport allows some specific molecules to cross the BBB, but therapeutic development using this route has had limited success. A number of techniques have been used to bypass the BBB, but are often highly invasive and ineffective. Over the last two decades, a minimally invasive technique to transiently open the BBB has been under development that utilizes transcranial focused ultrasound (FUS) in combination with intravascular microbubble contrast agents. This method is often carried out in conjunction with magnetic resonance imaging (MRI) to guide and assess BBB opening and has been referred to as MRI guided FUS (MRgFUS). Because of the utility of mouse models of neurological disease and the exploratory nature of MRgFUS, systems that allow BBB opening in mice are a useful and necessary tool to develop and evaluate this method for clinical application. In this dissertation project, a custom built, cost-effective FUS system for opening the BBB in mice was developed, with the objective of using this device to deliver therapeutics to the brain. Being a custom device, it was necessary to evaluate the ultrasound output, verify in vivo safety, and anticipate the therapeutic effect. The scope of the work herein consists of the design, construction, and evaluation of system that fulfills these requirements. The final constructed system cost was an order of magnitude less than any commercially available MRgFUS system. At this low price point, the hardware could allow the implementation of the methodology in many more research areas than previously possible. Additionally, to anticipate the therapeutic effect, molecules of pharmacologically-relevant sizes were delivered to brain with a novel, multispectral approach. Results demonstrated that the device was able to safely open the BBB, and macromolecule delivery showed that both molecule size and FUS pressure both influence the amount and distribution of molecules in the brain. Using different ultrasound pressures, the threshold for BBB opening was found to be ≥ 180 kPa (0.13 MI). The threshold for damage was found to be ≥ 420 kPa (0.30 MI), and was minor at this pressure, but extensive for higher pressure (870 kPa, 0.62 MI), in which minor damage was caused by this pressure. Performing a novel implementation of a diffusion model on the fluorescence images of 500, 70, and 3 kDa dextran resulted in calculated diffusion coefficients of 0.032 ± 0.015, 12 ± 6.0, and 0.13 ± 0.094 square microns per second, respectively.

>Magister Scientiae - MSc / The blood-brain barrier (BBB) is an interface between the brain parenchyma and the circulating system. This barrier plays a vital role in protecting the CNS by restricting free paracellular diffusion of molecules from the systemic circulation. Methamphetamine (MA) is a highly addictive psychostimulant and has demonstrated neurotoxic properties as well as the ability to compromise the BBB. MA exposure is strongly linked with increased oxidative stress which can result in a decrease in the integrity of the BBB.The aim of this study was to investigate in vitro effects of pure and street MA "tik" on DNA proliferation and cell cycles in mouse brain endothelial (bEnd5) cells. Trypan blue was used to determine effects of MA (0.0001M-1mM) on cell viability and % cell growth. The Cell Titer Glo® luminescent assay and nonradioactive analogue, 5-bromo-2'-deoxyuridine (BrdU) was used to detect ATP and DNA levels, respectively. Cell cycles (propidium iodide incorporation) were analysed using flow cytometry. Statistical analysis was performed using Wilcoxin Rank Sum Test in which P<0.05 was denoted as significant. Results of this study showed that: 1. Viability of bEnd5 cells exposed to all selected concentrations of MA were unaffected when compared to controls (P>0.05). 2. % Cell growth was suppressed by MA exposure at 96hrs in comparison to that of controls (P≤0.03). 3. Cells exposed to MA had significant higher ATP concentrations than control cells at 96hrs (P ≤.0.03) 4. DNA synthesis was markedly suppressed in cells exposed to pure MA and street MA sample 4 (P≤0.03), while was similar and higher in cells exposed to street MA sample 1 (P=0.39), and street MA sample 2 and 3 (P≤0.04), respectively at 96hrs. 5. bEnd5 cell were arrested between 72 and 96hrs at the G1-S phase. In conclusion, this study demonstrated pure and illicit samples of MA obtained from forensic police did not affect the viability of bEnd5 cells, however resulted in the significant suppression of their cell numbers. This growth inhibition may be due to MA-induced cell cycle arrest at the G1-S phase. The study also showed that compounds found in the samples of street MA produced results significantly different to that of pure MA.

Magister Scientiae (Medical Bioscience) - MSc(MBS) / The blood brain barrier (BBB) is a signaling interface between the blood and the central nervous system (CNS), which prohibits the entry of harmful blood-borne substances into the brain micro-environment, thus maintaining brain homeostasis. The crucial role of the BBB is protecting the CNS, which may adversely be affected by alcohol. The central component of the BBB, endothelial cells (ECs), regulates BBB transport by regulating the permeability both transcellularly and through their paracellular junctions, by structures called tight junctions (TJs) that are composed of proteins. The aim of this study was to investigate the in vitro effects of ethanol (EtOH) and fermented rooibos (Rf) on a monolayer of bEnd5 mouse brain ECs, by determining the effects of EtOH and Rf on bEnd5 (i) cell viability (ii) cell proliferation (iii) rate of cell division (iv) cell toxicity (v) claudin-5 transcription (vi) permeability across a monolayer of bEnd5 ECs and (vii) morphology, for a selected experimental timeline of 24, 48, 72, and 96hrs. We then investigated if the simultaneous exposure of Rf and EtOH could reverse or alleviate the EtOHinduced effects on the bEnd5 ECs. EtOH metabolism induces oxidative stress and results in a range of adverse physiological effects. Aspathalus linearis (rooibos) contains many phenolic compounds, of which the main antioxidant activity is attributed to aspalathin. Our underlining hypothesis is that the antioxidants in an aqueous rooibos extract may therefore protect against the potential oxidant damaging effects of alcohol on the BBB. Cells were exposed for 24hrs to selected concentrations of EtOH (25mM and 100mM), a concentration of Rf containing equivalent of 1.9nM aspalathin, and the combinations of EtOH and Rf. Cell viability and cell toxicity was determined, while cell proliferation and rate of cell division was estimated using the trypan blue exclusion assay. Real time quantitative PCR was implemented to quantify claudin-5 transcription, normalized against housekeeping genes, GAPDH and HPRT. Transepithelial electrical resistance (TEER) was measured using the Ohm Millicell-electrical resistance system, while bEnd5 monolayer morphology was analysed using the Zeiss scanning electron microscope. Both concentrations of EtOH led to an overall decrease in cell viability, and a decreased number of live cells across 72hrs. Consistent with this, EtOH resulted in increased cell toxicity across the 96hr experimental timeframe and a diminished rate of cell division. The transcription of claudin-5 in bEnd5 ECs exposed to 25mM and 100mM EtOH varied dramatically across the 96hr timeframe. While 25mM EtOH resulted in an overall decrease in TEER, cells exposed to 100mM EtOH only decreased TEER between 48 and 96hrs. Morphologically, both concentrations of EtOH led to compromised paracellular spaces as endorsed by high definition SEM analysis. The administration of Rf on its own resulted in an initial decrease in viability, followed by recovery between 72 and 96hrs. Exposure to Rf diminished live cell numbers at 72 and 96hrs, accompanied by a compromised rate of cell division and an overall increase in cell toxicity. In addition, Rf down-regulated claudin-5 transcription across the course of the experiment, particularly between 24 and 48hrs. In alignment with this, Rf also led to an increase in BBB permeability from 24 to 96hrs. However, SEM studies were not able to discriminate any differences between control and Rf treated cells. Our study showed that the BBB could be protected against the adverse effects of EtOH, and this at the plasma concentration induced by 500ml’s of Rooibos tea. The simultaneous exposure of Rf and EtOH was able to negate the effects of EtOH on cell viability, cell proliferation, and cell toxicity but exacerbated the effects of EtOH on claudin-5 transcription and paracellular permeability. Morphologically, co-exposure with Rf only reversed the effects of 25mM EtOH while exacerbating the effects of 100mM EtOH at 96hrs. In conclusion, EtOH was shown to be detrimental to the integrity of bEnd5 ECs, and the addition of a minuscule quantity of the Rf extract was able to partially alleviate excess ROS-induced effects.

Obesity is an independent risk factor for ischaemic stroke and may worsen stroke outcome. Therefore, the objective of this thesis was to identify potential interactions between obesity and stroke, with particular focus on inflammatory mechanisms. Ischaemic stroke was surgically induced in obese (ob/ob) and control (ob/-) mice by middle cerebral artery occlusion (MCAO). Outcome was worse in ob/ob mice, which showed rapid and severe ischaemic and vascular damage, characterised by blood vessel haemorrhage as early as 30 minutes post-stroke. The early vascular damage is likely an important driver of subsequent ischaemic damage by exacerbating inflammation in the brain. Assessment of systemic inflammation in ob/ob mice found that stroke had triggered a deregulated inflammatory response in the plasma, spleen, liver and bone marrow, and increased adipose tissue inflammation. Increased vascular damage was hypothesised to mediate the worse outcome found in obese rodents. The structural and inflammatory state of the vasculature was therefore assessed in ob/ob mice prior to and after stroke. Prior to stroke, increased vascular inflammation was found in ob/ob mice, though no differences in tight junction expression or structural alterations were noted. Stroke resulted in vascular damage that was similar on structural and molecular levels between genotypes, though ob/ob mice showed an increase in transcytosis in endothelial cells which may mediate their enhanced blood-brain barrier breakdown. Some studies have found obese patients have better outcome after stroke, perhaps because they are protected from detrimental post-stroke weight loss. The metabolic response to MCAO was therefore assessed, with ob/ob mice showing altered lipid metabolism post-stroke, such as increased fatty acid release from adipose tissue into the plasma. The potential role of fatty acid release in triggering adipose inflammation was also assessed in adipose tissue explants, though limited evidence for an inflammatory response to lipolysis was found. Overall, these findings suggest rapid and enhanced vascular damage may drive worse stroke outcome in ob/ob mice, as may their altered immunological and metabolic states.

The permeation of drugs across the blood-brain barrier (BBB) is a prerequisite for central nervous system (CNS) drug penetration. The BBB, possessing efflux transporters and tight junctions, limits drug penetration to the brain. Consequently, the discovery of novel drugs to treat CNS diseases remains problematic and is lagging behind other therapeutic areas. In vitro assays have progressed understanding of the factors that govern brain penetration. Central nervous system drug penetration is now thought to be modulated by three main processes, namely BBB permeability, active transport at the BBB and drug binding in blood and brain tissue. A more integrated approach to CNS drug discovery programmes is emerging which encompasses these processes in order to examine the rate and extent of drug brain penetration across species and improve predictions in human.A primary porcine in vitro BBB model was developed and characterised for the prediction of CNS drug permeability in vivo. Characterisation confirmed that the model exhibited physiologically realistic cell architecture, the formation of tight junction protein complexes, transcellular electrical resistance consistently >2000 Ω.cm2, functional expression the P-gp efflux transporter and ?-glutamyl transpeptidase and alkaline phosphatase activities.Transport of 12 centrally acting test drugs was investigated across four in vitro BBB models in order make comparisons between models and to generate in vitro permeability and efflux measurements. Blood-brain barrier permeability and active efflux processes are two major influences on the rate of drug penetration across the BBB. Species differences in fublood and fubrain, two prime influences on the extent of drug penetration, were investigated using equilibrium dialysis. Fraction unbound in brain was shown to be comparable across species suggesting that species differences in brain penetration could be due to variation in fublood for drugs that cross the BBB by passive diffusion, and/or species differences in transporter characteristics for drugs that are subject to active transport processes at the BBB. An in-house hybrid-PBPK rat CNS model was used to predict calculated rat Kp,uu using in vitro permeability, efflux, fublood and fubrain parameters generated during this work. The predicted Kp,uu generated using the rat CNS hybrid-PBPK model were within 3-fold of calculated Kp,uu. The rat CNS hybrid-PBPK model has potential use, as a tool for drug discovery scientists to aid the prediction of the extent of drug penetration in the early stages of drug discovery.This work has demonstrated that in vitro permeability and unbound drug fraction can be used to predict CNS drug penetration.

Ischemic stroke is characterized by a disruption of blood supply to a part of the brain tissue, which leads to a focal ischemic infarct. The expression and activity of MMP-9 is increased in ischemic stroke and is considered to be one of the main factors responsible for damages to the cerebral vasculature, resulting in compromised blood-brain barrier (BBB) integrity. However, the regulatory mechanisms of MMP-9 expression and activity are not well established in ischemic stroke. Since hypoxia/ischemia and reperfusion generates reactive oxygen species (ROS), I hypothesize that ROS is one of factors involved in up-regulation of MMP-9 expression in brain cells and ROS-mediated effect may occur via MAPK signaling pathway. My study has provided the evidence that ROS is responsible for an increase in MMP-9 expression in astrocytes mediated via MAPK-AP1 signaling pathway. Preliminary studies with an in vitro model of the BBB suggest that inhibition of MMP-9 is a critical component of reducing ROS-induced BBB permeability.