Neue aussichtsreiche Pharmazeutika scheitern regelmäßig in späten Entwicklungsphasen1 und stehen somit nicht als wertvolle Wirkstoffe zur Verfügung. Ein Grund hierfür ist die komplexe Pharmakinetik und der Mangel an geeigneten in-vitro Modellen. Daher befasst sich diese Dissertation mit der Entwicklung neuartiger in-vitro Membranmodelle am Beispiel der Blut-Hirn-Schranke (BHS). Zu diesem Zweck wird der aktuelle Stand der Technik vorgestellt und anschließend das Konzept eines Mikrofluidikchips, in welchem mittels Dielektrophorese an eine zuvor erstellte Polyamidmembran CaCo-2-Zellen assembliert wurden. Die Auslegung und Optimierung des Chip-Designs, die Entwicklung der in-situ Membran sowie die Ermittlung der Randbedingungen sind wesentliche Bestandteil dieser Arbeit. Es konnte mittels FEM-Simulationen und Assemblierungsversuchen ein Modell erzeugt werden, mit dem ein Chipdesign entwickelt werden konnte, dass zum einen ein günstigeres Verhältnis von Zellflächen- und Abluminalen Volumen aufweist und zum anderen möglichst wenig Zellen für den Aufbau benötigt. Dieses System bietet somit ein hohes Potenzial für die Herstellung verbesserter in-vitro Modelle. Jedoch konnte auch durch die Charakterisierung mit Rhodamin, Fluorescein und FITC-Dextran aufgezeigt werden, dass dieser Vorteil durch spezifische und unspezifische Bindungen an der größeren Oberfläche z.T. reduziert wird, abhängig vom verwendeten Chipmaterial und untersuchten Wirkstoff. Als neuartig kann die in-situ Herstellung einer vertikalen Polyamidmembran in einem Polymerchip bezeichnet werden, die im Rahmen dieser Arbeit entwickelt wurde. Für diese wurden die Parameter zur optimale Collagenbeschichtung für die Zelladhäsion ermittelt, sowie der Einfluss auf die Zellvitalität untersucht. Des Weiteren wurde das Medium zur Dielektrophorese und zur Kultivierung der Zellen ohne CO2-Begasung optimiert.:1 Abkürzungsverzeichnis 2 Formelzeichen 3 Einleitung 4 Grundlagen 5 Chipdesign 6 Herstellung und Charakterisierung der Stützmembran 7 Entwicklung des Zellkulturmodells 8 Zusammenfassung 9 Ausblick 10 Anhang 11 Literaturverzeichnis 12 Abbildungsverzeichnis / New promising pharmaceuticals regularly fail at late stages of development1 and thus do not become available as valuable active substances. Two of the main reasons for such failures are the complex pharmacokinetics and the lack of adequate in-vitro models. Therefore, this dissertation focuses on the development of novel in-vitro membrane models at the example of the blood-brain-barrier (BBB). It starts by presenting current investigations and the state-of-the-art technology and continues with the concept for a microfluidic chip in which CaCo-2 cells were assembled with dielectrophoresis on an in-situ membrane. The essential part of this work was to design and optimize this microfluidic chip, to develop an in-situ membrane to catch the assembled cells and to investigate the required boundary conditions. FEM simulations and assembling experiments conducted to the creation of a model to develop a chip design with a better ratio between cell area and abluminal volume, as well a low number of cells needed for the creation of this model. Such system might have a high potential to establish more sensitive in-vitro models than the current Transwell model. However, it was also demonstrated that this advantage is reduced by specific and nonspecific binding on the larger surface, depending on the chip material and the investigated test substance, shown during the chip characterization by using Rhodamine, Fluorescein and FITC-Dextran. Furthermore, the creation of a vertical polyamide in-situ membrane in a polymer chip like in this work, can be described as novel. To assemble cells on this supporting membrane, a protocol for a collagen coating as well for the dielectrophoresis medium were developed. Also, a modified culture medium was investigated, to allow the cultivation on standard atmospheric conditions.:1 Abkürzungsverzeichnis 2 Formelzeichen 3 Einleitung 4 Grundlagen 5 Chipdesign 6 Herstellung und Charakterisierung der Stützmembran 7 Entwicklung des Zellkulturmodells 8 Zusammenfassung 9 Ausblick 10 Anhang 11 Literaturverzeichnis 12 Abbildungsverzeichnis

The purpose of this study was the investigation of several transport systems expressed at the BBB. The identification and functional characterization of such transport systems is essential to provide a basis for strategies to regulate drug disposition into the brain. Immortalized rat brain endothelial cells (RBE4 cells) have been used in this study as an in vitro model of the BBB. The present study has shown that the RBE4 cells are a suitable model of the BBB for transporter studies. These cells do express the amino acid transport systems L and y+, which are known to be present at the BBB. The uptake of L-tryptophan, a neutral amino acid transported by system L, exhibited a half saturation constant (Kt) of 31 µM and a maximal velocity rate (Vmax) of about 1 nmol/mg/min in RBE4 cells. The kinetic constants of the L-arginine uptake, representing system y+ transport activity, into RBE4 cells were determined with a Kt value of about 55 µM and a Vmax of 0.56 nmol/mg/min. Furthermore the expression of two sodium dependent transporters, the 5-HT transporter (SERT) and the organic cation/carnitine transporter OCTN2, was shown at the RBE4 cells. Uptake studies with radiolabeled 5-HT exhibited a saturable, sodium dependent transport at RBE4 cells with a Kt value of about 0.40 µM and a Vmax of about 52 fmol/mg/min. L-carnitine and TEA (tetraethylammonium) are known to be transported by the OCTN2 transporter. The uptake of L-carnitine into RBE4 cells was shown to be sodium dependent and saturable with a Kt value of 54 µM and a maximal velocity of about 3.6 pmol/mg/min. In contrast, the organic cation TEA follows a sodium independent uptake mechanism at RBE4 cells. Also a sodium independent choline uptake into the cells was discovered but the molecular identity remained unknown. This saturable choline transport exhibited a Kt value of about 22 µM and a maximal velocity of about 52 pmol/mg/min.

The purpose of this study was the investigation of several transport systems expressed at the BBB. The identification and functional characterization of such transport systems is essential to provide a basis for strategies to regulate drug disposition into the brain. Immortalized rat brain endothelial cells (RBE4 cells) have been used in this study as an in vitro model of the BBB. The present study has shown that the RBE4 cells are a suitable model of the BBB for transporter studies. These cells do express the amino acid transport systems L and y+, which are known to be present at the BBB. The uptake of L-tryptophan, a neutral amino acid transported by system L, exhibited a half saturation constant (Kt) of 31 µM and a maximal velocity rate (Vmax) of about 1 nmol/mg/min in RBE4 cells. The kinetic constants of the L-arginine uptake, representing system y+ transport activity, into RBE4 cells were determined with a Kt value of about 55 µM and a Vmax of 0.56 nmol/mg/min. Furthermore the expression of two sodium dependent transporters, the 5-HT transporter (SERT) and the organic cation/carnitine transporter OCTN2, was shown at the RBE4 cells. Uptake studies with radiolabeled 5-HT exhibited a saturable, sodium dependent transport at RBE4 cells with a Kt value of about 0.40 µM and a Vmax of about 52 fmol/mg/min. L-carnitine and TEA (tetraethylammonium) are known to be transported by the OCTN2 transporter. The uptake of L-carnitine into RBE4 cells was shown to be sodium dependent and saturable with a Kt value of 54 µM and a maximal velocity of about 3.6 pmol/mg/min. In contrast, the organic cation TEA follows a sodium independent uptake mechanism at RBE4 cells. Also a sodium independent choline uptake into the cells was discovered but the molecular identity remained unknown. This saturable choline transport exhibited a Kt value of about 22 µM and a maximal velocity of about 52 pmol/mg/min.