La barrière hémato-encéphalique (BHE) régule finement l’apport en oxygène et en nutriments du cerveau et le protège d’éventuels pathogènes, notamment en bloquant le passage des molécules de poids moléculaire supérieur à 400 Da. Malheureusement, cette barrière est un obstacle à la délivrance de nombreux médicaments. Les ultrasons focalisés de basse intensité, combinés avec des microbulles, représentent un outil de choix pour perméabiliser la BHE, de façon sûre et réversible, et ainsi permettre de délivrer efficacement des médicaments ou des agents de contraste dans le cerveau. Dans la première partie de ma thèse, j’ai développé de nouvelles stratégies ultrasonores de perméabilisation de la BHE chez le rongeur. J’ai mesuré l’atténuation du faisceau ultrasonore par le crâne et étudié l’influence des paramètres acoustiques sur l’intensité et la durée de la perméabilisation. Ces développements m’ont ensuite permis de délivrer des nanoparticules dans le cerveau de rongeurs et d’observer cette délivrance par imagerie par résonance magnétique (IRM), tomographie par émission de positrons, imagerie de contraste de phase par rayons-X, spectrométrie de masse ou encore histologie. La seconde partie de mon travail a porté sur l’application de cette technologie ultrasonore à la maladie d’Alzheimer (MA). J’ai tout d’abord optimisé un protocole IRM T2* à très haute résolution permettant l’imagerie ex vivo des plaques amyloïdes de souris modèles de la MA. J’ai développé un traitement semi-automatique des images pour détecter et quantifier la charge amyloïde dans le cortex. Enfin, j’ai évalué la perméabilisation répétée de la BHE en tant que thérapie pour la MA et démontré que des perméabilisations répétées de la BHE pouvaient avoir un effet bénéfique sur la mémoire de rongeurs modèles de la maladie. / The blood-brain barrier (BBB) plays a crucial role in maintaining the hemostasis of the brain and protects it from pathogens. The BBB prevents molecules with a molecular weight higher than 400 Da to enter the brain. Crucial, the BBB becomes a limit to deliver drugs to the brain. Low intensity focused ultrasound and microbubbles are a unique tool to open the BBB, in a safe and reversible way, to deliver drugs to the brain. The first part of the PhD was dedicated to developing new strategies for BBB opening. To do so, I measured the attenuation of the ultrasound beam by the skull and studied the dependency of the intensity and of the duration of the BBB opening on the acoustic parameters. Thanks to these developments, I was able to deliver many kinds of nanoparticles to rodent brains and I could observe their delivery with techniques such as magnetic resonance imaging (MRI), positron emission tomography, phase-contrast X-ray imaging, mass spectroscopy or histology. The second part of my PhD was focused on applying this technology to Alzheimer’s disease (AD). I optimized a T2* MRI protocol at very high resolution for ex vivo imaging of amyloid plaques in the cortex of mice modeling AD. I developed a semi-automatic image treatment to detect and quantify the amyloid load. Finally, I tested a repeated BBB opening as a therapy for AD and showed that repeated BBB openings could have a beneficial impact on the memory on rodents modeling AD.

A central quest in biology is to understand the structure-function relationship of complex biological systems. The brain represents the ultimate complexity of a biological system: (1) the vertebrate brain contains 107-1011 neurons interconnected with glial cells; (2) over tens of diverse cell types are organized in a hierarchical way over an intricate landscape; (3) coordinated electrical and chemical activities of neuronal ensembles generate emergent properties and functions; and (4) each neuron can extend over large volumes with its spatial scales spanning 6 orders of magnitude. As a result, compared to other organ systems, our understanding of the brain remains primitive and obscure in terms of both its structures and its functions. Accordingly, many grand challenges endure in brain sciences, including comprehensively mapping neuronal wiring of the brain, an exhaustive taxonomy of cell types in the brain, and robust diagnostic and therapeutic strategies for brain diseases. These challenges are difficult to tackle with existing microscopy methods, because general trade-offs prevail between number of colors, imaging depth, spatial resolution, imaging throughput, sensitivity, and specificity. Therefore, the quest to understand the brain calls for advances and innovations on novel microscopy methods.The evolution of modern Raman microscopy is fundamentally driven by the development of novel spectroscopy methods. The advancement of molecular spectroscopy in turn pushes forward and benefits from, the progress in vibrational probes, labeling chemistry, and sample processing and transformation. In particular, stimulated Raman scattering (SRS) microscopy offers high sensitivity and fast acquisition for biomedical imaging, by harnessing accelerated vibrational transition from stimulated emission. Bio-orthogonal chemical imaging provides chemical specificity and minimal perturbation for visualizing metabolic dynamics of small molecules, by using tiny vibrational probes such as deuterium and alkyne. Electronic pre-resonance SRS (epr-SRS) microscopy further enhances the sensitivity to the nanomolar level for imaging specific proteins, by exploiting electronic pre-resonance of specially designed Raman dyes. Despite these notable innovations, the imaging depth of these Raman microscopy methods is limited to superficial layers of biological tissues (~100 μm) due to light scattering. This dissertation contributes to the development of advanced Raman microscopy methods for volumetric imaging with extended imaging depth in scattering tissues. For this purpose, we develop a set of tissue clearing strategies tailored to specific Raman imaging modalities. In addition, we develop image analysis methods to extract systems information from volumetric high-dimensional imaging datasets. Equipped with our volumetric imaging and analysis methods, we elucidate intricate structures and functions of the brain at both physiological and pathological conditions, providing implications for brain tumor metabolism and cerebellum development. Chapter 1 introduces an overview of Raman microscopy with particular emphasis on SRS and epr-SRS microscopies. Chapter 2 discusses the principles of tissue clearing with special focus on the basis of light scattering, the working mechanisms of different categories of tissue clearing methods, and the rationale underlying the development and evolution of these tissue clearing methods. Chapter 3 describes the development of volumetric chemical imaging, which brings label-free SRS microscopy, bio-orthogonal chemical imaging, and metabolic imaging to the realm of volumetric imaging with greater than 10-fold depth extension. Chapter 4 depicts the development of volumetric multiplex imaging, which generalizes epr-SRS microscopy to the territory of volumetric imaging. With this method we achieve one-shot imaging of more than 10 colors over millimeter-thick brain tissues, extending the imaging depth of multiplex protein imaging by 10~100 folds. Chapter 5 is a manuscript of an ongoing project on imaging nanocarriers for drug delivery across the blood-brain barrier (BBB). We develop a method of correlative multispectral SRS and fluorescence microscopy to image nanoparticles by SRS with multispectral information and particle counting capability and to image tissue context (especially cerebral vasculature) by fluorescence with high specificity. Using this method, we achieve direct imaging of nanocarriers that cross the BBB with definitive spectral evidence and single particle sensitivity. The preliminary results quantifying the proportion of nanoparticles that cross the BBB provide implications that challenge the current understanding of drug delivery to the brain.

Résumé confidentiel / Confidential abstract

The direct endothelial contact with adjacent astrocytic end-feet is believed to establish blood-brain barrier (BBB) typical characteristics in endothelial cells of the central nervous system (CNS). However, this contact is only present in capillary vessels of the brain parenchyma and absent in larger veins, arteries and vessels within the meninges. To investigate a potential impact of direct endothelial interactions with adjacent astrocytic end-feet on the molecular tight junction (TJ) composition and ultrastructure, we performed a systematic analysis of endothelial cell contacts within the vascular tree of parenchymal and leptomeningeal vessels. Immunofluorescence labeling for claudin-3, claudin-5, zonula occludens-1 and occludin was used to compare the molecular composition, without showing significant differences in their distribution along the vascular tree of parenchymal and leptomeningeal vessels. Furthermore, electron microscopy in combination with quantitative analyses was performed to investigate the endothelial ultrastructure revealing significant differences within the length of endothelial overlaps between the different vessel types. Here, parenchymal arteries exhibit noticeably longer cell contacts compared to capillaries, which could not be observed in leptomeningeal vessels. It was also observed that arterial vessels regularly contain a higher density of endothelial vesicles throughout the parenchyma and meninges as a sign for transendothelial traffic. Hence, endothelial expression of blood-brain barrier typical TJs is not limited to capillary vessels with an intimate contact to surrounding astrocytes, but is also observed in arteries and veins of the brain parenchyma as well as the meninges, the latter of which are lacking a direct astrocyte-endothelial interaction. These vessel-specific characteristics can now be used to address and compare alterations of the BBB in different settings of CNS pathologies.:Table of Content 1. INTRODUCTION 4 1.1 THE BLOOD-BRAIN BARRIER 4 1.2 HISTORY 5 1.3 STRUCTURE AND COMPOSITION 6 1.4 THE ROLE OF THE MICROENVIRONMENT 8 1.4.1 ASTROCYTES 8 1.4.2 PERICYTES 9 1.5 BLOOD BRAIN BARRIER FUNCTION 10 1.5.1 PHYSIOLOGIC CONDITIONS 10 1. 5.2 PATHOLOGIC CONDITIONS 11 2. OPEN QUESTIONS AND SCIENTIFIC APPROACH 12 3. PUBLICATIONS 13 3.1 DIFFERENT SEGMENTS OF THE CEREBRAL VASCULATURE REVEAL SPECIFIC ENDOTHELIAL SPECIFICATIONS, WHILE TIGHT JUNCTION PROTEINS APPEAR EQUALLY DISTRIBUTED 13 3.2 THE BLOOD-BRAIN BARRIER 28 4. SUMMARY 40 5. REFERENCES 43 6. PROOF OF SIGNIFICANT CONTRIBUTION 48 7. DECLARATION OF ACADEMIC HONESTY 49 8. ACKNOWLEDGMENT 50 9. CURRICULUM VITAE 51

Bei der vorliegenden Arbeit handelt es sich um experimentelle Untersuchungen zu funktionellen Aspekten der neurovaskulären Einheit im Tiermodell. Mittels Licht,- Fluoreszenz- und der Elektronenmikroskopie sowie diverser immunhistochemischer Nachweisverfahren konnten wir verschiedene Populationen der neurovaskulären Einheit näher charakterisieren. So konnten wir nachweisen, dass im Hirnparenchym eine Population CD11c-positiver, dendritischer Zellen existiert, welche im gesunden Gehirn hauptsächlich an Prädilektionsstellen für Entmarkungsherde im Rahmen der Multiplen Sklerose vorkommt. Weiterhin zeigten wir im Tiermodell, dass die über Diphterietoxin vermittelte Oligodendrozytendepletion mit einer Demyelinisierung der Axone im Gehirn einhergeht, wobei die Freisetzung und Drainage der Antigene in zervikale Lymphknoten keine gegen das Gehirn gerichtete Autoimmunität auslöst. Ebenso untersuchten wir den Beitrag endothelialer Tight junctions zur Bluthirnschrankenstörung im Modell der fokalen Ischämie an der Ratte. Hierbei waren wir in der Lage nachzuweisen, dass entgegengesetzt zur herrschenden Lehrmeinung diese nicht verantwortlich für die erhöhte Gefäßpermeabilität im Rahmen des Schlaganfalls im Tiermodell zu sein scheint. Vielmehr konnten wir mit Hilfe der Elektronenmikroskopie einen neuen Mechanismus aufzeigen. Diese Ergebnisse liefern neue Erkenntnisse bezüglich der Interaktion der verschiedenen Populationen der neurovaskulären Einheit und können somit zur Entwicklung neuer Modelle verschiedener Pathologien des Zentralnervensystems beitragen.

Background: Near-IR fluorescence (NIRF) imaging is becoming a promising approach in preclinical tumor detection and clinical image-guided oncological surgery. While heptamethine cyanine dye IR780 has excellent tumor targeting and imaging potential, its hydrophobic property limits its clinical use. In this study, we developed nanoparticle formulations to facilitate the use of IR780 for fluorescent imaging of malignant brain tumor. Methods: Self-assembled IR780-liposomes and IR780-phospholipid micelles were prepared and their NIRF properties were characterized. The intracellular accumulation of IR780-nanoparticles in glioma cells were determined using confocal microscopy. The in vivo brain tumor targeting and NIRF imaging capacity of IR780-nanoparticles were evaluated using U87MG glioma ectopic and orthotopic xenograft models and a spontaneous glioma mouse model driven by RAS/RTK activation. Results: The loading of IR780 into liposomes or phospholipid micelles was efficient. The particle diameter of IR780-liposomes and IR780-phospholipid micelles were 95 and 26nm, respectively. While stock solutions of each preparation were maintained at ready-to-use condition, the IR780-phospholipid micelles were more stable. In tissue culture cells, IR780-nanoparticles prepared by either method accumulated in mitochondria, however, in animals the IR780-phospholipid micelles showed enhanced intra-tumoral accumulation in U87MG ectopic tumors. Moreover, IR780-phospholipid micelles also showed preferred intracranial tumor accumulation and potent NIRF signal intensity in glioma orthotopic models at a real-time, non-invasive manner. Conclusion: The IR780-phospholipid micelles demonstrated tumor-specific NIRF imaging capacity in glioma preclinical mouse models, providing great potential for clinical imaging and image-guided surgery of brain tumors.

Doctor Scientiae / The blood-brain barrier (BBB) is a restrictive interface located between the blood circulation and the central nervous system (CNS), regulating the homeostatic environment of the neuronal milieu, by controlling the permeability of the cerebrovasculature. Currently, we cannot fully comprehend the regulatory features and the complexity of BBB morphology to allow for intervention clinically. The thesis consists of four publications. The methodology paper proposes a novel experimental design to visualize the morphological architecture of immortalized mouse brain endothelial cell lines (bEnd3/bEnd5). The brain endothelial cells (BECs) were grown on cellulose matrices and fixed in 2.5 % glutaraldehyde in preparation for visualization of the paracellular (PC) spaces between adjacent BECs, employing high-resolution electron microscopy (HREM), with vested interest in the morphological profile of the developing BEC. The second publication addresses and reports on the nanosized detail of BEC monolayer morphology utilizing high-resolution scanning electron microscopy (HR-SEM) and published the first descriptions of the extrusion of a basement membrane from developing in vitro BECs. Moreover, we categorized and discussed two types of nanotubule (NT) development specific for the establishment of the BEC monolayers. NTs can occur via nanovesicle extrusion onto the BEC membrane surfaces, which fuse, forming tunneling NTs (TUNTs) between adjacent BECs. Furthermore, cytoplasmic extensions of BEC membrane leading edges give rise to tethering NT (TENTs), which result in overlapping regions across the PC spaces, resulting in PC occlusion. BEC NT communication is illuminated in a third publication utilizing immunofluorescence microscopy, which reports on the molecular, cytoskeletal elements governing NT formation. This study shows, for the first time, f-actin and α-tubulin cytoskeletal proteins extending between the soma of the cells and NT cytoskeletal structures within an in vitro BBB model. Thereafter, the effects depolymerizing agents, Cytochalasin D and Nocodazole, were investigated on f-actin and α-tubulin cytoskeletal protein generation,functionality of NT morphology, cell division and permeability. For the first time, we show that f-actin possesses an additional function, key to tight junction, plaque protein organization. Moreover, it facilitates TENT formation, essential for cytoplasmic projection across PC spaces. Conversely, α-tubulin facilitates known functions: (i) transportation, (ii) cytokinesis, (iii) cellular division, and (iv) possesses a novel function as the molecular cytoskeletal backbone of TENTs, which facilitates BBB impermeability. A critical review evaluates past literature, in light of the current findings emanating from this study. The review critiques the concept of BEC cilia, which have been reported in the literature, comprised of tubulin and actin, but at low-resolution. In the light of our novel observations, nowhere in transmission electron microscopy do we observe cilia on the BECs, we postulate that NTs have been misnamed and mischaracterized as cilia. The thesis endeavors to elucidate the complexity of BEC nanostructures by examining the emerging role of the nanoscopic landscape of BBB development and the changing nature of BEC morphology, NT formation and associated cytoarchitectural underpinnings governing NT morphology. The research study attempts to, with a view to create new avenues for treating brain pathology, revolutionize our interpretation of barrier-genesis on a nanoscale.

The brain's resistance to ischemic injury can be transiently augmented by prior exposure to a sub-lethal stress stimulus, i.e. preconditioning. It has been reported that Toll-like receptors (TLRs) are involved in the preconditioning-induced protective effect against ischemic brain injury. In this study, we investigated the effect of preconditioning with a TLR2 specific ligand, Pam3CSK4, on focal cerebral ischemia/reperfusion (I/R) injury in mice. Pam3CSK4 was administered systemically 24 h before the mice were subjected to focal cerebral ischemia (1 h) followed by reperfusion. Cerebral infarct size was determined, blood brain barrier (BBB) permeability was evaluated, and expression of tight-junction proteins were examined after focal cerebral I/R. Results showed that pre-treatment with Pam3CSK significantly reduced brain infarct size (1.9 ± 0.5% vs 9.4 ± 2.2%) compared with the untreated I/R group. Pam3CSK4 pre-treatment also significantly reduced acute mortality (4.3% vs 24.2%), preserved neurological function (8.22 ± 0.64 vs 3.91 ± 0.57), and attenuated brain edema (84.61 ± 0.08% vs 85.29 ± 0.09%) after cerebral I/R. In addition, Pam3CSK4 pre-treatment preserved BBB function as evidenced by decreased leakage of serum albumin (0.528 ± 0.026 vs 0.771 ± 0.059) and Evans Blue (9.23 ± 0.72 μg/mg vs 12.56 ± 0.65 μg/mg) into brain tissue. Pam3CSK4 pre-treatment also attenuated the loss of the tight junction protein occludin in response to brain I/R injury. These results suggest that TLR2 is a new target of ischemic preconditioning in the brain and preconditioning with a TLR2 specific ligand will protect the brain from I/R injury.

Although it is well-recognized that catecholamines are generally unable to penetrate the developed blood-brain barrier (BBB) to gain entry into brain, except at circumventricular sites where the BBB is absent or deficient, onto-genetic development of this barrier seems to have escaped systematic study. To explore BBB development, several approaches were used. In the first study rats were treated once on a specific day of postnatal ontogeny, as early as the day of birth, with the neurotoxin 6-hydroxydopa-mine (6-OHDA; 60 mg/kg), and then terminated in adulthood for regional analysis of endogenous norepinephrine (NE) content of brain. In another study, rats were treated once, on a specific day of postnatal ontogeny, with the BBB-perme-able neurotoxin 6-hydroxydopa (6-OHDOPA; 60 mg/kg) following pretreatment with the BBB-impermeable amino acid decarboxylase inhibitor carbidopa (100 mg/kg IP), then terminated in adulthood for regional analysis of endogenous NE content of brain. In the third study rats were treated once, on a specific day of postnatal ontogeny, with the analog [3H]metaraminol, and terminated 1 hour later for determination of regional distribution of tritium in brain. On the basis of [3H]metaraminol distribution and NE depletions after neurotoxin treatments, it is evident that the BBB in neocortex, striatum, cerebellum and other brain regions forms in stages over a period of at least 2 weeks from birth. Moreover, because the BBB consists of several element (physical-, ion-restrictive-, and enzymatic-barrier), the method employed will derive data mainly applicable to the targeted aspect of the barrier, which may or may not necessarily coincide with elements of the barrier that have a different rate of ontogenetic development. Accordingly, it is evident that some aspects of physical- and ion-restrictive elements of the BBB form within approximately the first week after birth in rat neocortex and striatum, while enzymatic elements of the BBB form more than than 2 week later. Regardless, the BBB forms at earlier times in forebrain vs hindbrain regions.

Aim: The general aim of this research is development and evaluation of novel methods for predicting active transport over the human blood-brain-barrier (BBB), while this project specifically aims to i) review the literature and select suitable methods and substrates, ii) develop models for determining in vitro kinetic properties of selected compounds, analyze the in vitro data using the developed models and to use Maximum Transport Activity (MTA) approach (Karlgren et al., 2012), iii) perform Physiology Based Pharmacokinetic (PBPK) modelling and compare to in vivo literature data. Background: Drug permeation to the brain through blood circulation is primarily limited by blood-brain barrier (BBB), due to existence of tight junctions in endothelial cells of blood vessels as well as active efflux and influx transporters in the barrier. Toxicity and CNS related side effects can be caused by peripheral targeted drugs crossing BBB. Hence, prediction of BBB permeability and estimation of drug concentration in the brain tissue are challenging in drug discovery. To resolve this, estimating the human BBB permeability using improved in vitro and in silico predictive models can be a facilitator. Methods: In vitro data provided by the Drug Delivery research group was used to develop in vitro predictive models for BBB penetration of Verapamil, Risperidone, and Prazosin using R-studio 1.2.5. The MTA approach was adjusted for extrapolation of BBB in vitro transporter activity to in vivo condition. For PBPK modelling, we took advantage of PK-Sim® to simulate drug disposition and time profile of Risperidone in human and animal species. Results: It was shown that MDR1 is the major transporter for efflux transport of Prazosin and Risperidone in brain while both BCRP and MDR1 have similar impact on transport of Verapamil. Furthermore, it was presented in PBPK models that the predicted brain concentration of Risperidone increases in rat and nonhuman primate (NHP) when MDR1 And BCRP are knocked out while the brain concentration of Risperidone in dog is not affected by expression level of the efflux transporters. Conclusion: Both MDR1 and BCRP are contributing in efflux transport of Verapamil, Risperidone, and Prazosin across the BBB. Additionally, expression of the efflux transporters shown to have an impact on brain exposure of Risperidone in animal PBPK models.