Abstracts from the 22nd International Symposium on Signal Transduction at the Blood–Brain Barriers

s from the 22nd International Symposium on Signal Transduction at the Blood–Brain Barriers Würzburg, Germany. 11–13 September 2019 Published: 11 September 2019 © The Author(s) 2019. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. A1 The brain barriers: gatekeepers of CNS immune privilege Britta Engelhardt University of Bern, Theodor Kocher Institute, Bern, Switzerland Correspondence: Britta Engelhardt ‐ bengel@tki.unibe.ch Fluids Barriers CNS 2019, 16(2): A1 Discoveries leading to an improved understanding of immune surveillance of the central nervous system (CNS) have repeatedly provoked dismissal of the existence of immune privilege of the CNS. Recent rediscoveries of lymphatic vessels within the dura mater surrounding the brain by modern live cell imaging technologies have revived this discussion. Understanding immune privilege of the CNS requires intimate knowledge of its unique anatomy. Employing in vivo and in vitro studies on immune cell migration into the CNS during immune surveillance and neuroinflammation using live cell imaging we have explored the anatomical routes and cellular and molecular mechanisms involved in immune cell trafficking to the CNS. Our findings underscore that endothelial, epithelial and glial brain barriers establish compartments within the CNS that differ strikingly with regard to their accessibility to immune cell subsets. Grant support: This work was supported by the Swiss National Science Foundation, Fidelity International Foundation. A2 Contribution of the blood–brain barrier in Alzheimer’s disease, mouse modelling Amaia Dominguez‐Belloso, Ricardo Figueiredo, Sylvaine Guérit, Sonja Thom, Kavi Devraj, Peter Winter, Stefan Liebner Institute of Neurology (Edinger‐Institute), Johann Wolfgang Goethe‐University Frankfurt, Medical School, Heinrich‐Hoffmann‐Straße 7, 60528, Frankfurt, Germany; GenXPro GmbH, Altenhöferallee 3, 60438, Frankfurt, Germany Correspondence: Amaia.dominguez‐belloso@kgu.de Fluids Barriers CNS 2019, 16(2): A2 The blood–brain barrier (BBB) protects the brain from the detrimental influence of various blood-born, endogenous and exogenous substances, and is known to be directly or indirectly involved in the progression of brain pathologies. Although growing evidence points to a major role of the brain vasculature and the BBB in disease progression, little is known about the specific processes taking place at the neurovascular unit (NVU). In order to shed light on the events taking place at the NVU in brain diseases, we performed next generation sequencing, employing the MACE technology, of mouse brain microvessels (MBMVs) from an Alzheimer’s disease (AD) mouse model, harbouring the Thy1-SwDI triple mutation of amyloid precursor protein (APP), and compared it to aged-matched WT mice at 6 and 18 month-of-age. We show that particularly at 6 month-of-age significant regulation of genes involved in cellular junctions, transporters and angiogenic processes takes place in AD mice. Along with these changes in the expression profile we observed that AD mice have memory and BBB deficits at 6 month-of-age by behavioural and tracer experiments with 3-4KD dextrans, respectively, which became aggravated with age. Interestingly, differences in gene expression between AD and WT mice were less pronounced at 18 month-of-age, possibly suggesting that ageing per se leads to pronounced changes at the NVU, and that the AD condition might only push the system to become pathologic. Together, our results point into a major contribution of a hampered BBB in AD but also in ageing, reemphasizing the vascular aspect of AD and highlighting the BBB as a possible therapeutic target. A3 3Rs in blood–brain barrier research: The power of in vitro models Winfried Neuhaus AIT Austrian Institute of Technology GmbH, Vienna, Austria Correspondence: Winfried Neuhaus ‐ winfried.neuhaus@ait.ac.at Fluids Barriers CNS 2019, 16(2): A3 The principles of the 3Rs concept to alternatives to animal testing were first introduced in 1959 by Bill Russel and Rex Burch in their groundbreaking book “The Principles of Humane Experimental Technique”. Basically, it deals with measures to refine, reduce or replace animal experiments with completely animal-free methods. Although the 3Rs are often stated in research proposals as a goal for method development, success in the sense of the 3Rs is very limited. Hardly any method is developed so far that it can really replace an animal experiment specified by the regulations. In the cosmetics industry, animal experiments have been banned in the EU, in toxicology there is a clear trend towards alternative methods, and in the pharmaceutical industry a rethinking of the sensible use of animal experiments is slowly beginning. Nevertheless, the number of animal experiments is increasing, mainly due to the increased use in basic science. It is important to point out that the 3Rs do not serve a purely ethical purpose. With the help of advanced methods, one can pragmatically obtain much better scientific data, especially if one thinks of species differences and considers the translation of data from animal models to humans. In addition, new non-animal methods are cheaper than animal models and can therefore contribute to saving money. In this lecture, an overview of data and new in vitro models in the blood–brain barrier area will be given, which can contribute to the 3Rs, but also their applicability and limitations will be critically examined. In summary, these non-animal models can be used as great, innovative methods to supplement limitations of animal testing and should not be seen as merely replacing Open Access Fluids and Barriers of the CNS Page 2 of 12 Fluids Barriers CNS 2019, 16(Suppl 2):29 and alternatives to animal experiments. In the future, the common, meaningful use of non-animal models with animal experiments could lead to a significant, scientifically meaningful reduction of animal experiments in order to generate even more relevant data for humans. A4 Epigenetic regulation of BBB integrity in stroke Anuska V. Andjelkovic, Svetlana M. Stamatovic, Chelsea M. Phillips, Gabriela Martinez‐Revollar, Richard F. Keep Department of Pathology and Neurosurgery, University of Michigan, Medical School, Ann Arbor, USA Correspondence: anuskaa@med.umich.edu Fluids Barriers CNS 2019, 16(2): A4 Acute cerebral ischemic injury evokes complex cascade of pathophysiological events at the blood–vascular–parenchymal interface, which evolve over time and space, and result in damage to neuronal cells and edema formation. Stroke survivors usually recover at least some functionality within 3 months, although only 25% return to pre-stroke levels. One problem regarding recovery after stroke is that there are ongoing processes during the chronic phase that increase the risk of recurrent stroke and could also be a solid basis for developing poststroke cognitive impairment (vascular dementia). The new emerging evidence suggest that Blood brain barrier (BBB) recovery is incomplete process which could influent the stroke injury recover, increase risk for the new stroke occurrence and be solid substrata for developing the vascular dementia. Although that several processes are indicated to guide the BBB partial recovery, there are still lack of knowledge whether and how these processes limited the BBB recovery. In the present study we have focused on elucidate the involvement and role of epigenetic mechanism in regulation of post-stroke BBB recovery. The young (3 months) and old mice (18 months) were subject to thromboembolic (TE) stroke based on injection of suspension of platelet-reach microemboli. TE stroke was confirmed and its resolution was followed for 10 days by MRI T2 imaging while BBB permeability was observed and analyzed 1–10 days after stroke using GD-DTPA tracer by MRI T1 imaging. The brain microvessels were collected at day 1 and 10 after TE stroke from area of stroke injury and GD-DTPA hyperpermeability. The isolated brain microvessels from young and old mice were subject of DNA methylation analysis (global DNA methylation assay and Reduced Representation Bisulfite Sequencing (RRBS)) as well histone modification analysis (MODfied Histone peptide Area and ChIP-Seq). Our results indicated profound alteration in DNA methylation and histone monomethylation particularly H3K9, H3K27 and trimethylation (H3K9, H3K79, H3K27) in old mice 10 days after TE stroke. The most affected are proinflammatory and junctional regulator genes. This study provides new information related to the epigenetic regulation of post-stroke BBB and revealed the potential novel therapeutic targets to treat BBB hyperpermeability. A5 Restoring Blood–Brain Barrier Function to Improve Cognition in Alzheimer’s Disease Anika M.S. Hartz, Atcharaporn Ontawong, Yujie Ding, Chutima Srimaroeng, Bjoern Bauer Sanders‐Brown Center on Aging, University of Kentucky, Lexington, USA; Department of Pharmacology and Nutritional Sciences, College of Medicine, University of Kentucky, Lexington, USA; Department of Physiology, Faculty of Medicine, Chiang Mai University, Chiang Mai, Thailand; Division of Physiology, School of Medical Sciences, University of Phayao, Phayao, Thailand; Department of Pharmaceutical Sciences, College of Pharmacy, University of Kentucky, Lexington, USA Correspondence: Anika MS Hartz ‐ anika.hartz@uky.edu Fluids Barriers CNS 2019, 16(2): A5 Introduction: Increasing evidence indicates that blood–brain bar

The blood-brain barrier (BBB) protects the brain from the detrimental influence of various blood-born, endogenous and exogenous substances, and is known to be directly or indirectly involved in the progression of brain pathologies. Although growing evidence points to a major role of the brain vasculature and the BBB in disease progression, little is known about the specific processes taking place at the neurovascular unit (NVU). In order to shed light on the events taking place at the NVU in brain diseases, we performed next generation sequencing, employing the MACE technology, of mouse brain microvessels (MBMVs) from an Alzheimer's disease (AD) mouse model, harbouring the Thy1-SwDI triple mutation of amyloid precursor protein (APP), and compared it to aged-matched WT mice at 6 and 18 month-of-age. We show that particularly at 6 month-of-age significant regulation of genes involved in cellular junctions, transporters and angiogenic processes takes place in AD mice. Along with these changes in the expression profile we observed that AD mice have memory and BBB deficits at 6 month-of-age by behavioural and tracer experiments with 3-4KD dextrans, respectively, which became aggravated with age. Interestingly, differences in gene expression between AD and WT mice were less pronounced at 18 month-of-age, possibly suggesting that ageing per se leads to pronounced changes at the NVU, and that the AD condition might only push the system to become pathologic. Together, our results point into a major contribution of a hampered BBB in AD but also in ageing, reemphasizing the vascular aspect of AD and highlighting the BBB as a possible therapeutic target. The principles of the 3Rs concept to alternatives to animal testing were first introduced in 1959 by Bill Russel and Rex Burch in their groundbreaking book "The Principles of Humane Experimental Technique". Basically, it deals with measures to refine, reduce or replace animal experiments with completely animal-free methods. Although the 3Rs are often stated in research proposals as a goal for method development, success in the sense of the 3Rs is very limited. Hardly any method is developed so far that it can really replace an animal experiment specified by the regulations. In the cosmetics industry, animal experiments have been banned in the EU, in toxicology there is a clear trend towards alternative methods, and in the pharmaceutical industry a rethinking of the sensible use of animal experiments is slowly beginning. Nevertheless, the number of animal experiments is increasing, mainly due to the increased use in basic science. It is important to point out that the 3Rs do not serve a purely ethical purpose. With the help of advanced methods, one can pragmatically obtain much better scientific data, especially if one thinks of species differences and considers the translation of data from animal models to humans. In addition, new non-animal methods are cheaper than animal models and can therefore contribute to saving money. In this lecture, an overview of data and new in vitro models in the blood-brain barrier area will be given, which can contribute to the 3Rs, but also their applicability and limitations will be critically examined. In summary, these non-animal models can be used as great, innovative methods to supplement limitations of animal testing and should not be seen as merely replacing

Open Access
Introduction: Increasing evidence indicates that blood-brain barrier dysfunction contributes to cognitive decline in Alzheimer's disease (AD). Two key elements of barrier dysfunction include (1) reduced levels of the blood-brain barrier transporter P-glycoprotein (P-gp) that clears Aβ from the brain, and (2) barrier leakage. Both reduced P-gp and barrier leakage have been linked to Aβ brain accumulation and cognitive decline. While increasing evidence shows Aβ involvement in barrier dysfunction, the underlying mechanisms remain to be fully defined. Moreover, therapeutic strategies to restore barrier function are currently not available. Thus, there is an unmet critical need to define the mechanism(s) that lead/s to barrier dysfunction and to develop effective intervention strategies to help restore barrier function in AD. In the present study, we test the hypothesis that restoring P-gp through PXR activation and reducing barrier leakage by scavenging reactive oxygen species (ROS) will restore barrier function, increase Aβ clearance, and slow cognitive decline. Methods: To test our hypothesis, transgenic hAPP mice received a purified diet containing the PXR activator pregnenolone 16α-carbonitrile (PCN) or the ROS scavenger N-acetyl-l-cysteine (NAC). We assessed P-gp protein expression and transport activity, Aβ brain and plasma levels, capillary leakage, renal function and cognition to test the PCN or NAC effect in hAPP mice. Results: Feeding hAPP mice for 21 months with the PXR activator PCN restored P-gp protein expression and transport activity, lowered Aβ brain levels, and improved cognition. Further, treating 12-month old mice with NAC for 3 weeks improved renal function and reduced barrier leakage. Conclusion: Our findings suggest that a combination therapy of the PXR activator PCN and the reactive oxygen scavenger NAC will restore P-gp levels and reduce the extent of barrier leakage which has the potential to lower brain Aβ burden and slow cognitive decline. Grant support: This work was supported by the 2R01AG039621 (NIA/ NIH; PI: Hartz); 1R01AG054459 (NIA/NIH; Co-I: Hartz); P30AG028383 (NIA/NIH; PI: Hartz) grants. Natural polyphenols of flavonoid-type (flavonols, flavones, flavanols, flavanones, anthocyanins and isoflavones) and non-flavonoid type (phenolic acids, lignans, stilbenoids) with their anti-oxidative and antiinflammatory properties are considered as promising feed additives in the nutrition of farm animals. These are present in plants either in their glycosylated (with sugar attached), free aglycones (no sugar) or in the less extended esterified form, which dictates their passage, absorption, and bioavailability in the body. Moreover, the biotransformation of polyphenols in enterocytes and liver to conjugated metabolites (glucuronided, methylated, sulphated and combined) seems to reduce their biological activity. It should be emphasized that, the position of conjugation in phytochemical molecules determines the extent of their actual bioactivities. The studies indicated that some forms of polyphenol metabolites, such as 7-glucuronide and 3-glucuronide exhibited high efficiency as antioxidants. We have been using mature female sheep to study isoflavones (daidzein, genistein), flavonols (quercetin derivatives), anthocyanins (cyanidin derivatives) and their metabolites access to the brain. The sheep model has serious advantages: (1) allows for long-lasting serial sampling of peripheral blood parallel with cerebrospinal fluid from the cerebral ventricles, (2) can be done with freely moving animals and, thus, avoids many of the potential complications associated with restraint, (3) the use of ovariectomised, estradiol implanted ewes preserves estradiol receptors in the brain and prevents the variability resulting from the changing level of sex hormones. Grant support: This work was supported by KNOW (Leading National Research Centre) Scientific Consortium "Healthy Animal -Safe Food", decision of Ministry of Science and Higher Education No. 05-1/ KNOW2/2015, and by the National Science.

A6 Polyphenols and their metabolites beyond brain barriers: studies on a sheep model
Iron delivery to the brain is essential for multiple neurological processes such as myelination, neurotransmitter synthesis and, as it is for all organs, for the utilization of oxygen for energy production. Previously, we introduced and established the concept that brain iron uptake is regulated at the level of the endothelial cells of the bloodbrain barrier (BBB), which is contrary to the widely held concept that the BBB endothelial cells serve as a simple conduit for the delivery of transferrin. We also identified H-ferritin, classically considered an intracellular iron storage protein, as a significant iron delivery protein for the brain. These new concepts start to address the question of how the brain acquires iron in a timely manner and adequate amounts. The importance of timely iron delivery during development is clinically manifested in the long term neurological and cognitive impact of developmental iron deficiency. The importance of regulation to adequately manage the amounts of iron delivered in the adult brain is clinically manifested in neurological disorders such as Restless Legs Syndrome (too little iron) and neurodegenerative diseases (too much iron). Despite the prevailing opinion that the BBB was a simple conduit with no apparent mechanism for regulation of brain iron uptake, we identified age, genotype, sex, and systemic iron status as physiological factors that modulate brain iron acquisition. Moreover, we have identified H-ferritin (FtH1) as a significant brain iron delivery protein, particularly during development that augments the traditional transferrin delivery system. Both proteins deliver iron to the brain during development in a mouse model but the pattern for the two proteins is significantly different. We also examined the impact of a mutant form of the human iron homeostatic regulator protein (HFE) protein, which is the most common gene variant in Caucasians. We demonstrate that at postnatal day 22, mutant mice brains take up greater amounts of iron compared to wildtype. In summary, brain iron uptake is mediated by two distinct mechanisms both of which are modulated by age, sex and genotype. Grant support: This work was supported by NIH 5R01NS077678.

Introduction:
In 1897, a woman, Julia Platt, established that neural crest cells (NCCs) in the head participate in the formation of high variety of mesenchymal derived tissues and created the term mesectoderm (now ectomesenchyme) for the mesenchyme of neuroectodermal origin able to develop into a large variety of cells, including forebrain pericytes (FPs). In our hypothesis, FPs can act as multitasking neural crest-derived cells and display a unique angiogenic attitude compared to pericytes of a mesodermal origin homing other different vascular district. Pioneering studies of pericyte-endothelial relationships during human foetal brain vascularization, revealed an intimate interplay of these cells in vessel sprouting events, redefining the view of a late recruitment of pericytes and suggesting that FPs can actually lead the nascent vessel growth and anastomosis. Moreover, recent studies have supported the specific angiogenic activities of FPs, showing tunnelling nanotubes (TNTs) in vessel-tovessel communication during the earliest events of normal as well as brain tumor angiogenic sprouting. Methods: We analysed aspects of FPs embryonic origin and mesenchymal nature, and TNTs and cord-like microtubes (MTs) formation, as possible steps in FP angiogenic activity during both human brain and glioblastoma vascularization, revealing the expression of pericyte, mesenchymal, and neural crest cell markers including NG2, CD146, and Sox10, by high resolution confocal microscopy. Results and conclusions: The results are in agreement with data obtained in murine brain tumor models4 and indicate that during angiogenesis, FPs seem to maintain their original NCC-derived pluripotency, forming TNTs and MTs initially composed solely of pericytes. The leading role of FPs in the earliest events of vascular growth suggests anti-angiogenic therapeutic strategies that combine pericyte TNT and MT targeting. Introduction: Due to the highly specialized nature of the brain barriers tissues, as well as their development and maintenance process, transcriptome profiling is a vital tool to better understand processes in the blood-brain Barrier (BBB). In particular, RNA-seq techniques have provided the opportunity of more thorough studies in many fields of life sciences. Interestingly, the number of studies taking advantage of this technique to study the brain barriers is still very low when compared to the total number of publications concerning the brain barriers. However, we expect that (i) the number of studies using RNA-seq will increase dramatically and (ii) these methodologies will help to bring the knowledge of the BBB to a new level. We present "BBBHub", a web based data hub platform that allows (i) the collection of existing and new data (ii) the integration and organization of omics-data (mainly RNA-seq) (iii) the analysis of the data in a standardized fashion and (iv) the access to the data to researchers in the field. Methods: The BBBHub uses as its backbone the frame work "openBIS", an open source platform for managing scientific information. Furthermore, we will develop an easy-to-use front-end system to allow all researchers to use the without advanced knowledge in computational biology. Results/conclusion: We set up standardized upstream data analysis workflows to allow comparison across datasets and to enable a flexible downstream analysis workflow. Mixed with the search functions this allows for comparison across genes or across whole datasets. These features will be very helpful to understand the utility and limitations of each dataset. In the coming months, we will be adding other features, such as single-cell RNA-seq and isoforms detection workflows.  Introduction: Due to the blood-brain barrier (BBB) drug delivery to the brain still represents a challenge. Nanoparticles (NP) have the potential to cross this barrier and to function as carrier systems. However, many in vitro models used for studying these transport mechanisms are not of human origin, lack essential characteristics and may therefore limit the translation of the generated results from the laboratories to the clinic. Methods: NP of 100 nm size in diameter were formulated by a combination of miniemulsion and solvent evaporation and, composed of Poly(lactic acid) (PLLA) and Poly(lactic-co-glycolic acid) (PLGA), stabilized by Tween 80. To study the NP-cell-interaction via flow cytometry and microscopy, the NP had a fluorescent dye encapsulated. The physicochemical characteristics of the NP were determined by transmission electron microscopy. Mass spectroscopy was performed to analyze the protein corona. Barrier-building endothelial cells were differentiated from human induced pluripotent stem cells. By measuring the transendothelial electrical resistance (TEER) the barrier formation was investigated. After NP exposure of 20 h, the toxic potential of all NPs was determined via crystal violet, FVS660 and CCK-8 assays. Results: Having low batch-to-batch deviations, only slight differences of the protein corona occurred among the different polymeric compositions. A model of the BBB consisting of human endothelial cells with a TEER of 2000 Ohm cm 2 for at least 80 h was established. Being highly internalized by the cells, the NP exposure did not have any toxic effects on the cells. The barrier function was mainly unaffected with only one PLGA-NP leading to a significant decrease of the TEER. Conclusion: An in vitro BBB model based on human cells generating a highly tight barrier was established. The NP exposure study points to a promising potential for BBB permeation of these NP and demonstrates the suitability of the BBB model for in vitro transport studies. Grant support: This work was supported by the Evonik Stiftung. Introduction: A variety of blood-brain barrier (BBB) in vitro models have been developed over the years (Avdeef, Deli, Neuhaus, 2015). In human BBB models, based on primary or immortalized human cells, the desired barrier properties can´t be achieved. The differentiation of human induced pluripotent stem cells (hiPSC) into brain capillary endothelial cells (BCEC) is a promising approach to achieve a human BBB model with in vivo like barrier properties (Lippmann et al., 2012, 2014). Although transwell models are in use because of the simplicity in generation, static models do not reflect the shear stress generated by blood flow in physiological conditions. A major drawback is also the lack of direct cellular interactions, only indirect effect of astrocytes and pericytes on BCECs can be explored in transwell co culture models. It is known that in 2D models, endothelial cells undergo a phenotypic drift, Results: The three cell types (astrocytes, pericytes and endothelial cells) can self-assemble and form spheroids where astrocytes form a central core, followed by a layer of pericytes and an outer layer of endothelial cells, reflecting the BBB physiology Conclusions: Our 3D hiPSC-generated BBB spheroids are novel and offer the potential for disease modelling and pharmaceutical studies. Grant support: This work was supported by the Graduate School of Life Sciences, Würzburg, IM2PACT Consortium. Introduction: In recent years research on the development of flow-based in vitro blood-brain barrier models has increased progressively. Dynamic hollow-fiber in vitro (DIV-) models offer the possibility to study the blood-brain barrier (BBB) under shear stress. In comparison to microfluidic models, one major advantage is that significant amounts of cell mass and medium can be obtained for subsequent analysis. Cell viability and optimal culture conditions were assessed by glucose and lactate measurements with the BioProfile ® 100Plus, whereby BBB integrity was monitored by TEER measurement and permeability studies at different flow rates with the paracellular marker FITC-Dextran 4000 (FD4). Changes at the mRNA level of tight junction proteins including CLDN1-25, ZO1-3, occludin, tricellulin, JAM 1-3 as well as a series of ABC-and SLC-transporters were analyzed with a "BBB-chip", an adapted high-throughput qPCR Fluidigm chip, and compared to static transwell models. Results: Protocols for the long-term cultivation of hCMEC/D3 or hiPS-BCEC were successfully established. Adjustments of the medium composition showed a serum-dependency not only for cellular growth rates but also for BBB properties of the hCMEC/D3. Paracellular permeability rates of FD4 were dependent on the applied flow rate. Differences in the gene expression between DIV-models and transwell models indicate a complex regulation of BBB properties induced by flow. Conclusion: The established DIV-models are suitable for long-term studies of the human BBB for six weeks or up to several months offering the possibility of chronic disease-modeling such as Alzheimer's disease. Grant support: This work was supported by the Stiftung SET zur Förderung der Erforschung von Ersatz-und Ergänzungsmethoden zur Einschränkung von Tierversuchen, project 060 to W.N. and M.M. Introduction: Tumor Treating Fields (TTFields) have been established as adjuvant therapy for glioblastoma patients. The blood brain barrier (BBB) tightly controls the influx of the majority of compounds from blood to brain and may hinder delivery of drugs for treatment of brain tumors. We assessed the influence of TTFields on BBB permeability in vivo. Methods: For 72 h rats were treated with 100 kHz TTFields and thereupon i.v. injected with Evan's Blue (EB) which directly binds to Albumin. EB was extracted after brain homogenization and Fluids Barriers CNS 2019, 16(Suppl 2):29 quantified to evaluate effects on the BBB. To assess vessel structure, cryosections of rat brains were prepared following TTFields application and stained for tight junction proteins Claudin-5 and Occludin, respectively, and for immunoglobulin G (IgG). In addition, serial dynamic contrast-enhanced DCE-MRI with Gadolinium contrast agent was performed before and after TTFields application. Results: EB accumulation in the rat brain was significantly increased by TTFields application. The vessel structure became diffuse compared to control cryosections of rat brains in treated animals; Claudin 5 and Occludin were delocalized and IgG was found throughout the brain tissue. Serial DCE-MRI demonstrated significantly increased accumulation of Gadolinium in the brain, observed directly after 72 h of TTFields application. The effect of TTFields on the BBB disappeared 96 h after end of treatment and no difference in contrast enhancement between controls and TTFields treated animals was detectable. Conclusion: Application of TTFields at 100 kHz may have the potential to deliver drugs usually unable to cross the BBB to the brain by altering BBB integrity and permeability. Utilizing TTFields to open the BBB and its subsequent recovery could be a clinical approach of drug delivery for treatment of brain tumors and other diseases of the central nervous system. Grant support: This work was supported by Novocure Ltd.

A14
Cargo-conjugation to Tat affects its potential to facilitate brain endothelial cell adherence, uptake, and barrier Introduction and purpose: The 11-residue cationic peptide Tat1 is a promising tool to facilitate delivery of peptide drug entities not only into cells but likely also across biological barriers, such as the blood-brain barrier (BBB). Conjugation of a cargo to Tat may, however, influence on the permeation properties of the resulting Tat-cargo construct. In order to evaluate whether Tat conjugation is a viable strategy for brain delivery of drug entities, we evaluate the chemical stability and the BBB interacting and permeating properties of Tat as well as of Tat conjugated to the clinically relevant peptide drug entities NR2B9c (1 kDa)2 and N-dimer (2.1 kDa)3. All constructs were fluorophorelabelled for quantification and visualization purposes. Methods and results: The chemical stability of NR2B9c, Tat, Tat-NR2B9c, and Tat-N-dimer was assessed by UPLC in PBS and cell media supplemented with 10% FBS. All constructs were stable in PBS. In cell media supplemented with 10% FBS ~ 90% of initial NR2B9c and Tat-Ndimer remained intact following 4 h incubation, whereas only ~ 50% and ~ 30% of initial Tat and Tat-NR2B9c, respectively, remained intact. The propensity of Tat to facilitate brain endothelial cell adherence -and uptake as well as BBB permeation of NR2B9c and the N-dimer was assessed using primary bovine brain endothelial cells cultured in contact with rat astrocytes4. Tat facilitated endothelial cell membrane adherence and uptake of NR2B9c to a greater extent than the N-dimer. In addition, uptake of Tat and Tat-NR2B9c, but not NR2B9c and Tat-N-dimer, was observed in astrocytes cultured on the basolateral ("brain") side. Following a 3 h BBB permeation study, Tat-NR2B9c appeared to permeate the barrier to a greater extent that NR2B9c, Tat, and Tat-N-dimer. Conclusions: Conjugation of Tat to NR2B9c did not affect cell membrane adherence and cell uptake, whereas conjugation to the N-dimer limited cell membrane adherence and cell uptake. On the other hand, the Tat-N-dimer remained intact in cell media; whereas extensive degradation was observed of Tat and Tat-NR2B9c. Near future studies will elucidate whether the superior BBB permeation of Tat-NR2B9c, when quantified via the conjugated fluorophore, is due to enhanced permeation of a degradation product or reflects permeation of the intact construct. Grant support: This work was supported by the Lundbeck Foundation Research Initiative on Brain Barriers and Drug Delivery (RIBBDD). Introduction: Cerebral Small Vessel Disease (SVD) causes 25% of stroke and 45% of dementia worldwide. Despite its prevalence, little is known about the mechanisms that cause or propagate the disease. However, neurovascular unit (NVU) dysfunction and perturbation of the matrisome are thought to play a key role in Blood-Brain Barrier (BBB) breakdown and SVD progression. Monogenic SVD caused by mutations in Collagen IV α1/α2 (COL4A1/2), a highly abundant matrisome protein, offer a tractable platform to study SVD pathophysiology in vitro. Methods: Wild-type control (WT) and COL4A2G702D patient-derived human induced pluripotent stem cells (hiPSCs) were differentiated into brain microvascular endothelial cells (BMECs), astrocytes and mural cells, using previously published protocols. The cell types were characterised using qPCR, immunocytochemistry, flow cytometry and functional tests, including transendothelial electrical resistance (TEER), permeability to FITC-Dextran, LDL-uptake and tube formation. The three cell types were then combined into a Transwell-based co-culture model of the BBB. Results: hiPSC-derived BMECs exhibit high TEER of 1500-2000 Ohm cm 2 , which is increased and maintained over two weeks in coculture with astrocytes and in tri-culture with astrocytes and mural cells. WT and COL4A2G702D BMECs are impermeable to 10 kDa FITC-Dextran, uptake AF488-LDL and are capable of tube formation. Preliminary results from the COL4A2G702D SVD model suggest that COL4A2G702D BMECs exhibit increased expression of MMPs 2 and 9, relative to WT, as well as reduced Collagen IV deposition and lower tight junction protein expression.

Conclusions:
The triple co-culture model offers a platform to assess the comparative functional capacity of the control and diseased cell types. Abnormal MMP expression and Collagen IV protein deposition are consistent with what is seen in patients and strengthens the argument that matrisome alterations could be important in COL4A1/2 SVD. Grant support: This work was supported by the ACT-VAD (Stroke Association, British Heart Foundation, Alzheimer's Society) and The Rosetrees Trust. In neurodegenerative disorders, concentrations of extracellular neurotransmitter in the central nervous system (CNS) are out of balance. Glutamatergic dysregulation and extracellular glycine accumulation cause excitotoxicity and subsequent neuronal death. This process is largely mediated by N-methyl-d-aspartate receptor (NMDAR) activation. NMDARs are ionotropic receptors permeable for positive ions. In the CNS, they are composed by two obligatory NR1 subunits containing glycine binding sites and two NR2 subunits that are activated by glutamate. NR3 subunits which are less abundant in the CNS are activated by glycine. While neuronal NMDARs are well studied, their functional role on endothelial cells (ECs) is still to be investigated. ECs of the Blood Brain Barrier (BBB) are exposed to the extracellular milieu of the CNS and therefore to high concentrations of neurotransmitters. In this study, we investigated the effect of NMDAR modulators on murine brain microvascular endothelial cells (MBMECs). Western blot experiments show expression of NR1, NR2a and NR3a subunits on MBMECs. Application of typical NR2 binding site activators (glutamate and quinolinic acid) did neither evoke any NMDAR mediated currents in electrophysiological experiments, nor affect Ca 2+ levels. No changes in transendothelial electrical resistance (TEER) values of MBMECs were observed. However, glycine application to whole cell recorded MBMECs evoked a bigger NMDAR mediated current compared to control conditions and induced an increase in intracellular Ca 2+ levels. Functionally, these changes were reflected in a reduced TEER and impaired MBMEC migration. Application of glycine but not glutamate binding site inhibitors rescued these effects. Taken together, these results suggest that glycine activated NMDARs composed of NR1 and NR3a subunits are expressed on MBMECs and that their activation triggers a loss in barrier function which is most likely mediated by intracellular Ca 2+ changes. Grant Support: The Groups of Sven G. Meuth and Stefan Bittner are funded by the "Transregional Collaborative Research Center (CRC) SFB TR-128 Initiating/effector versus regulatory mechanisms in Multiple Sclerosis -progress towards tackling the disease"; the Group of Sven G. Meuth and Michael Platten were funded by the DFG "FOR 2289: Kalzium-Homöostase bei Neuroinflammation und -degeneration: Neue Ansatzpunkte für die Therapie der multiplen Sklerose?"

A17 Glycine activation of NMDA Receptors on brain endothelial cells reduces barrier integrity mediated by increased intracellular calcium level
In the development of neuroinflammatory diseases, alterations of the blood brain barrier (BBB) represent key events. Factors that cause endothelial cells (ECs) of the BBB to develop an immune cell-permissive phenotype are poorly understood. It is known that EC-involving vascular dysfunction can be caused by kynurenine pathway (KP) metabolites. The enzyme indolamin-2,3-Dioxygenase (IDO-1) catalyzes the initial step of the KP in which tryptophan (TRP) is converted into various biologically active metabolites of which kynurneine and kynurenic acid are known to activate the aryl hydrocarbon receptor (AhR), a ligand binding transcription factor that mediates immune responses. In this study, we investigated the effect of IDO-1 induction and subsequent KP metabolite production on T-cell migration across murine brain microvascular endothelial cells (MBMECs) under inflammatory conditions. We could show that inflammation of MBMECs induced with IFNγ and TNFα results in IDO-1 expression and KP metabolite production which causes AhR activation and its translocation into the nucleus. Migration of T-cells across MBMEC monolayers from IDO-1−/− mice is reduced under inflammatory conditions in comparison to WT mice. The effect is mediated by a decrease in ICAM-1 expression. Interestingly, pharmacological inhibition of the AhR also results in a reduced T-cell migration across MBMEC monolayers from WT mice and it is accompanied by a decrease in ICAM-1 expression under inflammatory conditions. Additionally, these findings were confirmed in MBMECs obtained from AhR−/− mice, where a decreased T-cell migration under inflammation was also detected. Taken together, these results indicate a new endogenous mechanism in which IDO-1 activation through pro-inflammatory stimuli activates the AhR, which The majority of supratentorial (ST) ependymomas (EPN) in children is driven by distinct gene fusions between the partners C11orf95 and RELA. Tumors belonging to the resulting molecular group ST-EPN-RELA are characterized by activation of NF-κB signaling and deregulation of the p53 pathway. In contrast to surgery and radiotherapy, chemotherapeutic concepts have so far failed in the clinical management of these tumors. Using RNAi and drug screening methods, we identified p53 reactivation and inhibition of nuclear protein shuttling as promising therapeutic approaches. While low dose dactinomycin can successfully reestablish p53 function in ST-EPN-RELA cells, penetration of the drug through the blood-brain barrier (BBB) is poor. We are currently exploring alternative strategies such as combination with efflux pump inhibitors or liposomal packaging to overcome these constraints. In addition, idasanutlin, a brain penetrant MDM2 inhibitor, and selinexor, a covalent, slowly reversible, and selective inhibitor of XPO1 being the sole nuclear exporter of p53, have entered preclinical in vivo studies in ST-EPN-RELA models. Inhibition of XPO1 also enhances nuclear localization of IκBα, which is expected to result in downregulation of NF-κB transcriptional activity in ST-EPN-RELA cells. In order to identify optimal dosing strategies and assess effects of combinatorial treatment approaches on BBB penetration, we here apply cerebral microdialysis combined with ultraperformance liquid chromatography coupled to tandem mass spectrometry (UPLC-MS/MS) analyses, which allows exact, continuous, and time-dependent drug quantification in tumors or healthy tissue in freely-moving experimental animals. Introduction: Choroid plexuses (CPs) are localized in the brain ventricles (the third, two lateral and the fourth) and form interface between the blood and the cerebrospinal fluid (CSF). They are composed of tightly connected epithelial cells, mainly responsible for active CSF production, surrounding a core in which blood vessels are fenestrated. Proper CSF production is dependent on: (1) maintaining of fenestration of CP capillaries (vascular endothelial growth factor/receptors-VEGF/VEGFR) and water transport through CP epithelium (aquaporin 1-AQP1, transcellular and claudin 2-CLDN2, paracellular passage). This study evaluates mRNA expression of mentioned above factors in the ovine CPs collected from different ventricles. Methods: Studies were performed on adult ewes (4-5 years old, n = 12) during the period of increasing days (April). CPs were collected separately from the fourth ventricle and from the third ventricle together with the lateral ventricles immediately after sacrifice at 3 different time points: 11:00 (n = 4), 16:00 (n = 4), and 22:00 (n = 4). The real-time PCR analysis of mRNA expression of VEGF isoforms (VEGF164 and VEGF120) and its receptors (VEGFR-1 and VEGFR-2), AQP1 and CLDN-2 were performed. Results: The significantly (p < 0.05) higher expression of VEGF164 in all time points, and VEGF120 at 16:00, was observed for CPs from fourth compared to lateral ventricles. The expression of VEGFRs did not differ between CP types. The expression of AQP1 was significantly (p < 0.05) lower in CPs from fourth compare to lateral ventricles at 2 time points: 11:00 and 22:00, and showed common tendency (p = 0,.54) at 16:00. The expression of CLDN2 was slightly lower (p = 0.067) in CPs from fourth compare to lateral ventricles, but only at 16:00. Conclusion: It seems to be important to consider CPs from the fourth and lateral ventricles, separately when studying VEGF, AQP1 and CLDN2.

Introduction:
The blood-brain barrier (BBB) is a physical barrier formed by endothelial cells of the central nervous system and cells of the neurovascular unit. Transport across the barrier is tightly regulated by the presence of tight junctions (TJs) composed of a number of transmembrane proteins including claudins, occludin and junctional adhesion molecules. Claudin-5 is of particular importance in maintaining integrity of the BBB and changes in its expression has been implicated in a number of disorders including stroke, traumatic brain injury and psychiatric disorders such as schizophrenia. The aim of this study was to assess the integrity of the blood-brain barrier in a cohort of 60 postmortem brain samples from the Stanley Medical Research Institute, including 15 bipolar disorder, 15 major depression, 15 schizophrenia and 15 unaffected controls. Methods: The integrity of the blood-brain barrier was analysed by immunohistochemical (IHC) analysis of fresh frozen brain sections from the hippocampus and orbital frontal cortex. Changes in transcript levels of TJ proteins were assessed by qPCR of RNA isolated from Fluids Barriers CNS 2019, 16(Suppl 2):29 the caudal cingulate, cerebellum, parietal, occipital and premotor cortex. Results: IHC analysis of claudin-5 protein levels revealed decreased claudin-5 expression in the hippocampus of major depression and schizophrenia. RNA transcript levels of claudin-5 were upregulated in the cerebellum and occipital cortex of bipolar disorder. Correlation analysis looking at age of onset or duration of each disease versus TJ expression were performed and showed negative correlations between levels of TJ proteins and duration of bipolar disorder across multiple brain regions.

Conclusion:
Here we showed that the integrity of the BBB has been compromised in a cohort of patients with neuropsychiatric disorders. This suggests an important role of BBB disruption in the development and prognosis of each disorder that could be targeted therapeutically in the future.  The blood brain barrier (BBB) positioned along blood vessels of the central nervous system (CNS) is a highly selective and tightly regulated barrier. The principal components of the BBB are endothelial cells (EC), which have developed numerous specialisations which limit movement of material from blood to brain and vice versa. Paracellular transport is restricted due to the presence of tight junctions (TJ), dynamic structures linking adjacent ECs. Claudins, a family of transmembrane proteins, act as the main determinants of TJ properties, with claudin-5 (CLDN5) being the most enriched. Previous work in the lab has identified a link between CLDN5 suppression and psychosis through an inducible knockdown mouse model. Schizophrenia is a neurodevelopmental disorder that affects almost 1% of the population. However, in patients with 22q11 deletion syndrome (22q11DS), this prevalence rises to 30%. Due to a microdeletion within the 22q11.21 chromosomal region, this population is haploinsufficent for the CLDN5 gene. Importantly, CLDN5 is a dosage sensitive gene that also has polymorphisms which result in reduced gene expression. The lab has previously demonstrated that the reduced level of CLDN5 expression that results from carrying both a low expression allele and being haploinsufficent, may predispose individuals to schizophrenia. Therefore, our main objective is to determine if restoring CLDN5 expression at the BBB prevents a schizophrenia like phenotype. To do this, we will use an adeno-associated viral (AAV) vector to increase CLDN5 expression. As the overall aim is to translate this molecular modulation strategy into a clinical setting, the efficacy of the AAV must be optimised. This has been done by screening various plasmids in vitro, examining changes in gene expression and its temporal nature, the effect of promoter specificity and the use of inducible expression systems. Their effects on the expression of CLDN5 in vitro was investigated by IHC, qPCR and WB analysis.  Introduction: The central nervous system (CNS) is an immune privileged site. The brain barriers separate the CNS into compartments that significantly vary with respect to their accessibility to immune cell subsets and their drainage of antigens from the CNS to the peripheral immune system. It has been shown that the blood-brain barrier (BBB) and blood cerebrospinal fluid barrier (BCSFB) can be breached by activated immune cells to reach the cerebrospinal fluid (CSF) filled ventricular and subarachnoid spaces, allowing for CNS immune surveillance without the risk of disturbing homeostasis of the CNS parenchyma, which is lined by the glia limitans. Recent studies have proposed the existence of a glymphatic system in the CNS where CSF mixes with the interstitial fluid (ISF) of the CNS parenchyma and is drained via dural lymphatic vessels into peripheral lymph nodes. These concepts ignore the existence of the brain barriers that may prohibit such a fluid exchange. Methods: We here propose that disturbance of the brain barriers may cause CNS pathology and thus disruption of the CNS immune privilege. To investigate this hypothesis, we are using and establishing novel reporter mice allowing for visualization of the brain barriers by two photon intravital microscopy in the brain and spinal cord of mice during health and in animal models of multiple sclerosis and Alzheimer's Disease. This approach will allow us to visualize the distribution of cellular and humoral factors of the immune system in the different CNS compartments bordered by the brain barriers. Grant support: This work was supported by the Swiss Government Excellence Scholarship; FBRI. Introduction: Receptor-mediated transcytosis (RMT) is the potential method for transferring biopharmaceuticals across the blood-brain barrier (BBB), and the cell-surface proteins, internalized into brain capillary endothelial cells, serve as promising candidates for this process. The present study aimed to identify cell-surface proteins internalized into human brain microvascular endothelial cells using surface biotinylation method and SWATH-based quantitative proteomics. Methods: After biotinylation of cell-surface proteins, an internalization study was performed with 20% human serum. The biotinylated proteins internalized into the cells were recovered by streptavidin beads, followed by identification and quantification by SWATH-based quantitative proteomics. Results: Using SWATH-MS-based quantitative proteomics, 378 cellsurface biotinylated proteins were identified in hCMEC/D3 cells, a model of human brain microvascular endothelial cell, and 125 proteins were identified to be internalized into hCMEC/D3. Compared to identification and expression levels of proteins in HUVEC, a model of peripheral microvascular endothelial cell, 35 selectively internalized proteins were identified in hCMEC/D3 cells. RNAseq database analysis showed 10 proteins with high mRNA expression in human brain endothelial cells, and 8 of these 10 proteins showed high levels of internalization into hCEMC/D3 cells. These 8 proteins were expressed in the plasma membrane fraction of HBMEC/Cibeta another model of human brain microvascular endothelial cell, as well as in hCMEC/D3. Based on immunohistochemistry-based atlas of protein expression profiles in normal tissues, 6 of the 8 proteins were at least localized in human brain microvessels. Conclusion: We identified 6 cell-surface proteins selectively internalized into human brain microvascular endothelial cells, and our findings provide novel strategies for the development of a BBB-permeable drug delivery system. Grant support: This work was supported by the AMED, JST-CREST, Grant-in-Aid for Scientific Research (B). In animal models of epilepsy induced by kainic acid, the primary ligand of ionotropic kainate receptors, BBB damage occurs, but the effect of kainate on brain endothelial cells has not been studied. Our aim was to examine the presence of kainate receptors and the effects of kainate on cultured cells of the BBB and to test anti-inflammatory and antioxidant drugs to protect against kainate-induced changes. Rat primary brain endothelial, pericyte and astroglial cell cultures were used to study cell viability by impedance measurement and cell morphology. BBB permeability was measured on a model made from the three cell types. The production of nitrogen monoxide and reactive oxygen species was followed by fluorescent probes. Four drugs were tested to protect against kainate-induced changes. The mRNA expression of kainate receptors and nitric oxide synthases (NOS) were studied by RT-PCR. The presence of kainate receptors was demonstrated on brain endothelial cells by PCR. Kainate damaged brain endothelial cells in a concentration-and time-dependent way and made the immunostaining of claudin-5 discontinuous at the cell border indicating the opening of the barrier. The viability or the morphology of astroglial cells were unchanged. Kainic acid increased the permeability of the BBB model for marker molecules and the production of nitric oxide (NO) in brain endothelial cells. Dexamethasone, simvastatin and edaravone protected against the reduced cell viability and increased permeability caused by kainate. Dexamethasone attenuated the elevated NO production and iNOS/NOS2 mRNS expression caused by kainate treatment. In conclusion, we demonstrated the presence of kainate receptors on brain endothelial cells and that kainate directly damages brain endothelial cells. Our results suggest that kainate receptors of brain endothelial cells could be therapeutic targets in epilepsy to attenuate BBB damage. Grant support: This work was supported by the bilateral grant from the Hungarian Academy of Sciences. Introduction: Creatine transporter (CRT) is involved in supplying creatine to neurons across the blood-brain barrier (BBB). Deletion of CRT causes cerebral creatine deficiency syndromes (CCDSs). Cyclocreatine, a creatine analog, may be a promising drug for treating CCDS patients. However, whether cyclocreatine is a substrate for CRT or not, and the transporters involved in its transport across the BBB remain unclear. This study aimed to elucidate the involvement of CRT in cyclocreatine transport using 2 cell lines. One line is HEK293 cells, which possess endogenous CRT-mediated creatine uptake, and the other is hCMEC/ D3 cells, which are immortalized human brain microvascular endothelial cells. Methods: Cells were incubated with [14C]cyclocreatine (9 µM) and [14C]creatine (9 µM) for the indicated times at 37 °C in the presence or absence of several compounds. siRNAs were transfected by lipofection. Protein and mRNA expressions were quantified by LC-MS/MS and qPCR, respectively.

Results:
[14C]cyclocreatine was taken up in a time-dependent manner by HEK293 cells, and exhibited saturable kinetics, with a Km of 188 µM and Vmax of 273 pmol/mg protein/min. The uptake of [14C] cyclocreatine by HEK293 cells was inhibited 97% by 10 mM unlabeled creatine, and 96% by 10 mM guanidinopropionic acid (GPA), a specific inhibitor of CRT. Treatment of HEK293 cells with CRT-targeting siRNA reduced CRT protein expression by 42%, and inhibited uptake of [14C]cyclocreatine by 33% and [14C]creatine by 37%. In hCMEC/D3 cells, uptake of [14C]cyclocreatine was suppressed by 91%, both by unlabeled creatine and GPA. Treatment of hCMEC/D3 cells with CRTtargeting siRNA reduced CRT mRNA expression by 97%, and inhibited uptake of [14C]cyclocreatine by 70% and [14C]creatine by 63%. Conclusion: The present results suggest that cyclocreatine is a CRT substrate; CRT at the BBB is the predominant contributor to influx transport of cyclocreatine into the brain. Our findings provide important insights for treating CCDS patients with cyclocreatine. Grant support: This work was supported by AMED, Grant-in-Aid for Scientific Research (B). Introduction: One pathological characteristic of late-onset Alzheimer's disease (AD) is the accumulation of amyloid-β (Aβ) peptides. Dysregulation of the blood-brain barrier (BBB) confers the risk for suffering from AD and contributes to disease progression. The most associated genetic risk factor is the E4 allele of apolipoprotein E (APOE). APOE is involved in a number of metabolic pathways including lipid transport, Aβ aggregation, and Aβ clearance. However, the molecular and cellular signaling pathways regulated by APOE4 are currently poorly understood. The differentiation of BBB cells using patientderived induced pluripotent stem cells (iPSCs) is a promising approach to analyze AD disease mechanisms in the context of APOE4 and its role in BBB breakdown. Methods: First, we identified the APOE status of our AD patients and healthy matched controls. Then, episomal vectors were used for the generation of iPSCs from B-lymphoblastoid cell lines of patients carrying the APOE4 allele and healthy donors with homozygous APOE3 alleles. Following the verification of pluripotency the established iPSCs were used for differentiation into cells of the BBB system. Expression levels of BBB specific markers and barrier functionality were analyzed. Results: After successful generation of AD-specific iPSCs and proven pluripotency, we differentiated them into cells of the BBB system, in particular endothelial cells and astrocytes. Efficient differentiation was shown by the expression of cell-and BBB-specific markers including OCLN and TJP1. Barrier functionality was demonstrated by hindered paracellular transport of sodium fluorescein and transendothelial electrical resistance values > 1000 Ohm*cm2 for both AD patients and matching controls. Conclusion: Overall, we established a patient-and disease-specific BBB model suitable to investigate AD-associated genetic risk variants, resulting pathogenic phenotypes, and underlying disease mechanisms providing potential targets for AD treatment. Grant support: This work was supported by the HiPSTAR, BMBF (01EK1608C).