Polyelectrolyte Complexes and their Therapeutic Potential

Polyelectrolyte Complexes and their Therapeutic Potential Dissertation zur Erlangung des akademischen Grades eines Doktors der Naturwissenschaften (Dr. rer. nat.) an der Bayreuther Graduiertenschule für Mathematik und Naturwissenschaften der Universität Bayreuth vorgelegt von Christopher Volker Synatschke geboren in Lemgo Bayreuth, 2013 Die vorliegende Arbeit wurde in der Zeit von September 2009 bis Februar 2013 am Lehrstuhl für Makromolekulare Chemie II der Universität Bayreuth unter Betreuung durch Herrn Prof. Dr. Axel H. E. Müller angefertigt. Vollständiger Abdruck der von der Bayreuther Graduiertenschule für Mathematik und Naturwissenschaften der Universität Bayreuth genehmigten Dissertation zur Erlangung des akademischen Grades eines Doktors der Naturwissenschaften (Dr. rer. nat.). Dissertation eingereicht am: 12. April 2013 Zulassung durch die Prüfungskommision: 16. April 2013 Wissenschaftliches Kolloquium: 10. Juni 2013 Korrigierte Version der Dissertation: 13. Januar 2014 Prüfungsausschuss: Prof. Dr. Axel H. E. Müller (Erstgutachter) Prof. Dr. Ruth Freitag (Zweitgutachterin) Prof. Dr. Patrick Theato (Drittgutachter) Prof. Dr. Stephan Förster Prof. Dr. Andreas Fery (Vorsitz) “If you live each day as if it was your last, someday you'll most certainly be right.” ― Unknown “Why do you go away? So that you can come back. So that you can see the place you came from with new eyes and extra colors. And the people there see you differently, too. Coming back to where you started is not the same as never leaving.” ― Terry Pratchett, A Hat Full of Sky Für meine Mutter Elisabeth Table of Contents i Table of Contents Summary ................................................................................................................................. 1 Zusammenfassung ............................................................................................................... 5 Glossary ................................................................................................................................... 9 Chapter 1 – Introduction ................................................................................................ 13 1. Polymer-Aided Drug Delivery .............................................................................................. 13 1.1. Polymer-Drug Conjugates ................................................................................... 13 1.2 Important Concepts in Drug Delivery: Stealth Effect and EPR ........................... 14 1.3 Polymer Micelles and Vesicles ............................................................................. 16 1.4 Photodynamic Therapy ......................................................................................... 19 1.5 Complex Drug Delivery Systems: Multicompartment Micelles ........................... 20 2. Delivery of Genes ................................................................................................................. 23 2.1 Viral Vectors ......................................................................................................... 24 2.2 Non-Viral Vectors ................................................................................................. 24 2.3 Barriers to a Successful Delivery of Genes .......................................................... 25 2.4 Polymeric Vectors ................................................................................................. 28 3. Aim of this Thesis ................................................................................................................. 32 4. References ............................................................................................................................. 33 Chapter 2 - Overview over the Thesis ........................................................................ 39 1. Influence of Polymer Architecture and Molecular Weight of Poly(2- (Dimethylamino)ethyl Methacrylate) Polycations on Transfection Efficiency and Cell Viability in Gene Delivery ........................................................................................................ 40 2. Nano-Particulate Non-Viral Agent for the Effective Delivery of pDNA and siRNA to Differentiated Cells and Primary Human T Lymphocytes ....................................................... 43 3. Double-Layered Micellar Interpolyelectrolyte Complexes – How Many Shells to a Core? ......................................................................................................................................... 46 4. Micellar Interpolyelectrolyte Complexes With a Compartmentalized Shell ........................ 48 5. Multicompartment Micelles with Adjustable Poly(ethylene glycol) Shell for Efficient in Vivo Photodynamic Therapy ................................................................................................. 50 6. Individual Contributions to Joint Publications ..................................................................... 55 7. References ............................................................................................................................. 59 Zusammenfassung 6 Verzweigungsgrad bei vergleichbaren Molekulargewichten gefunden. Polykationen unterhalb eines kritischen Molekulargewichts von ca. 20 kDa zeigten keine relevante Transfektionseffizienz unabhängig vom N/P Verhältnis. Die PDMAEMA-Sterne mit etwa 20 Armen (Si-PDMAEMA), ausgehend von einem Silsesquioxan-Nanopartikel als Initiatormolekül, zeigten eine überragende Transfektionseffizienz in CHO-K1 Zellen, gekoppelt mit einer geringen Zytotoxizität. Diese verbesserten Transfektionseigenschaften konnten auch in anderen Zelllinien bestätigt werden, obwohl diese als sehr viel schwieriger zu transfizieren gelten. Darunter befanden sich unter anderem konfluente, nicht-teilende C2C12 Zellen sowie ausdifferenzierte humane TLymphozyten. Außerdem konnten die guten Transfektionsergebnisse von Si-PDMAEMA ebenfalls mit Mizellen aus einem amphiphilen Diblockcopolymer erreicht werden. Die Mizellen bestanden aus Polybutadien-block-PDMAEMA (PB-b-PDMAEMA), wobei die Kern-Schale-Struktur der Mizellen in Lösung der von sternförmigen Polymeren ähnelt. Dies wird als Hinweis auf ein generelles Design-Prinzip gedeutet: PDMAEMA Strukturen, die viele Arme ausgehend von einem gemeinsamen zentralen Punkt haben, sind bei der nicht-viralen Gentransfektion besonders effektiv. Beide Polymere (SiPDMAEMA und PB-b-PDMAEMA Mizellen) zeigten außerdem eine hohe Effizienz bei der RNA Interferenz-Therapie, da siRNA ebenfalls effektiv in verschiedene Zelllinien transportiert werden konnte. Der zweite Themenkomplex dieser Dissertation behandelt die Struktur ionischer kompartimentierter Mizellen, nachdem diese mit entgegengesetzt geladenen Polyelektrolyten komplexiert wurden. Abschließend wurde die Fähigkeit solcher Strukturen zum Wirkstofftransport in therapeutischen Anwendungen untersucht. Als Basis für diese Untersuchungen dienten kompartimentierte Mizellen, die aus dem amphiphilen und amphoteren Triblockterpolymer Polybutadien-block-poly(1-methyl-2vinyl pyridinium)-block-polymethacrylsäure (PB-b-P2VPq-b-PMAA; BVqMAA) durch Selbstassemblierung in wässrigen Lösungen eine Kern-Schale-Korona Struktur ausbilden. Der Kern dieser Mizellen besteht aus PB, während sich die diskontinuierliche Schale aus einem IPEK aus P2VPq und PMAA zusammensetzt. Nach außen hin werden die Mizellen durch eine dichte Korona aus überschüssigem, nicht an der Komplexbildung mit P2VPq beteiligtem, PMAA stabilisiert. Da die Korona Ketten bei ausreichend hohem pH Wert negativ geladen sind, konnten sie für die Komplexbildung mit diversen Polykationen Zusammenfassung 7 sowie doppelt hydrophilen Diblockcopolymeren mit einem kationischen Block verwendet werden. Sofern sich das für die Komplexierung verwendete Polykation von dem bereits vorhandenen P2VPq unterschied, wie zum Beispiel im Falle von quaternisiertem PDMAEMA (PDMAEMAq), bildete sich ein neues Kompartiment auf dem ursprünglichen Kern (PB und P2VPq/PMAA IPEK) der Mizellen aus. Dieses Kompartiment bestand aus dem IPEK zwischen PMAA und zugegebenem Polykation und ließ sich von der ersten Schale in elektronenmikroskopischen Aufnahmen deutlich unterscheiden. Es hatte die Form einer durchgängigen Schale, wenn die BVqMAA Mizellen eine kurze bis mittlere Korona-Länge hatten (345 – 550 MAA Einheiten pro BVqMAA Kette). Wurde für die Komplexbildung anstelle eines Homopolymers ein amphiphiles Diblockcopolymer mit einem positiven und einem wasserlöslichen aber ungeladenen Block verwendet, so konnte eine kolloidale Stabilität der erzeugten komplexen Mizellen über den gesamten Mischbereich zwischen Mizellen und Polykationen erreicht werden. Oberhalb einer kritischen Menge an zugegebenem HomoPolykation war dagegen eine Aggregation und makroskopische Phasenseparation der Mizellen zu beobachten. Bei BVqMAA Mizellen mit einer besonders langen PMAA Korona (1350 Einheiten) wurde eine unregelmäßige Verteilung des neu gebildeten IPEK anstelle einer durchgängigen Schale bei der Zugabe von Polykation-Homopolymeren um den Kern gefunden. Als Ursache für diese neuen Strukturen wird eine Grenzflächenminimierung zwischen ursprünglichem Kern und neu gebildetem Kompartiment vermutet. Die Minimierung wird ermöglicht, da die besonders lange Korona das neue Kompartiment effektiv gegen eine Wechselwirkung mit dem wässrigen Medium abschirmen kann und so die Lösungs-Stabilität der gesamten Mizelle nicht negativ beeinflusst wird. Abschließend wurden die Mizellen aus BVqMAA auf ihre Effizienz im Transport von hydrophoben Wirkstoffmolekülen für eine photodynamische Krebstherapie (PDT) sowohl in Zellkultur als auch in Mäusen mit Tumormodellen getestet. Durch die Komplexbildung der Mizellen mit dem doppelt hydrophilen Diblockcopolymer Poly(L-lysin)-blockpoly(ethylenglykol) (PLL-b-PEG) konnte die Zusammensetzung der Mizell-Korona von reinem PMAA kontinuierlich zu einer PEG-Korona verändert werden. Der Einfluss der Korona auf die biologischen Eigenschaften der Mizellen in Abhängigkeit ihrer Zusammensetzung konnte so untersucht werden. Bei einem hohen Anteil an zugegebenem PLL-b-PEG wurde eine stabile, neue Mizellstruktur gefunden, wobei sich Zusammenfassung 8 das neu gebildete IPEK-Kompartiment in Zylinderform senkrecht auf dem Mizellkern stehend ausbildet. Bei Untersuchungen der durch die Wirkstoff-tragenden Mizellen verursachten Zytotoxizität konnte ein deutlicher Einfluss der Korona-Zusammensetzung gefunden werden. BVqMAA Mizellen ohne PLL-b-PEG zeigten die höchste Zytotoxizität gegenüber humanen Lungenkrebszellen (A549). Diese Zytotoxizität wurde mit steigendem Anteil an PLL-b-PEG in der Mizellkorona stetig geringer, wobei der Trend gut mit der in die Zellen aufgenommenen Menge an Wirkstoff korrelierte. Vollständig PEGylierte Mizellen zeigten die geringste Menge an zellulär aufgenommenem Wirkstoff und die geringste Zytotoxizität. In Mäusen wurde eine verlängerte Blutzirkulation im Bereich mehrerer Stunden nach intravenöser Injektion lediglich für vollständig PEGylierte Mizellen beobachtet, während teilweise oder nicht PEGylierte Mizellen innerhalb kurzer Zeit nicht mehr im Blutkreislauf nachweisbar waren. Auch eine signifikante Akkumulation in subkutanen A549-Tumoren 24 h nach der MizellVerabreichung wurde nur für vollständig PEGylierte Mizellen gefunden. Die Menge an Wirkstoff, welche durch die Mizellen in den Tumor transportiert wurde, war ausreichend, um nach einmaliger Injektion der Mizellen und einer einzigen Laser-Bestrahlung eine effiziente Wachstumsunterdrückung des Tumors über einen Zeitraum von 21 Tagen zu erreichen. Somit konnte das Potential der BVqMAA Mizellen für einen Wirkstofftransport in vitro und in vivo erfolgreich nachgewiesen werden. Glossary 9 Glossary AIBN - azobisisobutyro nitrile ANOVA - analysis of variance ATRP - atom transfer radical polymerization BHT - butylated hydroxytoluene b-PEI - branched polyethyleneimine BVqMAA - polybutadiene-block-poly(1-methyl-2-vinyl pyridinium)- block-poly(methacrylic acid) BVT - polybutadiene-block-poly(vinyl pyridine)-block-poly(tertbutyl methacrylate) CD-spectroscopy - circular dichroism spectroscopy cmc - critical micelle concentration CPDB - 2-(2-cyanopropyl)dithio benzoate cryo-TEM - cryogenic transmission electron microscopy CTA - chain transfer agent DAMA - 2-((2-(dimethylamino)ethyl)methylamino)ethyl methacrylate DCM - dichloromethane DCTB - trans-2-[3-(4-tert-Butylphenyl)-2-methyl-2-propenylidene] malononitrile DDS - drug delivery system Dh - hydrodynamic diameter D.I. - dispersity index (light scattering) DLS - dynamic light scattering DMAc - dimethyl acetamide DMAEMA - dimethyl aminoethyl methacrylate DMF - dimethyl formamide DMSO - dimethyl sulfoxide Glossary 10 DNA - desoxyribonucleic acid DP - degree of polymerization Dq - quaternized PDMAEMA EBIB - ethyl 2-bromoisobutyrate EDC - N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride eGFP - enhanced green fluorescent protein ENB - Elite Network of Bavaria EPR - enhanced permeation and retention FCS - fetal calf serum GPC - gel permeation chromatography HMTETA - 1,1,4,7,10,10-hexamethyltriethylenetetramine IC50 - inhibitory concentration for 50 % of cells im-IPEC - intramicellar interpolyelectrolyte complex IPEC - interpolyelectrolyte complex LCC50 - concentration of polyplex for 50 % of viable cells LD50 - lethal dose for 50 % of cells l-PEI - linear polyethylene imine MAA - methacrylic acid MALDI-ToF-MS - matrix assisted laser desorption ionization time of flight mass spectrometry MCM - multicompartment micelle MFI - mean fluorescence intensity Mic-PDMAEMA - micellar PDMAEMA Mn - number average molecular weight MTT - 3-(4,5-dimethylthyazolyl-2)-2,5-diphenyl tetrazolium bromide Mw - weight average molecular weight MWCO - molecular weight cut off Glossary 11 Nagg. - aggregation number NMR - nuclear magnetic resonance N/P (ratio) - nitrogen to phosphate ratio P2VP - poly(2-vinyl pyridine) P2VPq - poly(1-methyl-2-vinyl-pyridinium) PB - polybutadiene PB-b-PDMAEMA - polybutadiene-block-poly(2-(dimethylamino)ethyl methacrylate) PBS - phosphate buffered saline PDAMA - poly(2-((2-(dimethylamino)ethyl)methylamino)ethyl methacrylate) PDAMAq - quaternized PDAMA PDI - polydispersity index PDLL-b-PEG - poly(D,L-lysine)-block-poly(ethylene glycol) PDMAEMA - poly(2-(dimethylamino)ethyl methacrylate) PDMAEMAq - quaternized PDMAEMA pDNA - plasmid DNA PDT - photodynamic therapy PEG - poly(ethylene glycol) PEI - poly(ethylene imine) PEO - poly(ethylene oxide) PI - propidium iodide PLL-b-PEG - poly(L-lysine)-block-poly(ethylene glycol) PMAA - poly(methacrylic acid) PMANa - poly(sodium methacrylate) PS - photosensitizer PtBMA - poly(tert butyl methacrylate) RAFT - reversible addition fragmentation chain transfer (polymerization) Glossary 12 RES - reticuloendothelial system Rh - hydrodynamic diameter RNA - ribonucleic acid ROS - reactive oxygen species SD - standard deviation SEC - size exclusion chromatography Si-PDMAEMA - PDMAEMA from silsesquioxane initiator siRNA - small interfering ribonucleic acid tBMA - tert-butyl methacrylate TE - transfection efficiency TEM - transmission electron microscopy THF - tetrahydrofurane Chapter 1 – Introduction 13 Chapter 1 – Introduction “Nanomedicine” is a catchy name for recent efforts in the medical field to use materials of nanoscopic dimensions for the treatment or diagnosis of various diseases.1, 2 Although the definition is rather vague, the size of such systems is in between that of small molecule drugs (up to several nanometers) and micron-sized objects, resulting in specific beneficial properties. Nanomaterials are comparable in size to proteins, enzymes or viruses and are therefore well suited to specifically interact with biological systems on a cellular or sub-cellular level. Compared to small molecule drugs a higher degree of complexity is inherent to those materials, leading to a multi-functionality of the overall system. For example, in cancer therapy many drug delivery systems (DDS) have been developed from nanostructured materials, which show increased availability of the drug in the organism through a prolonged circulation time in the bloodstream, an enhanced accumulation in the targeted tissue through targeting mechanisms as well as selective release or activation of the drug at the desired site. All these properties of nanoscopic DDS can result in a more effective treatment or diagnosis as compared to conventional methods. Depending on what type of a material is used for the medical application they can be classified into three different categories, namely inorganic, organic or hybrid materials. Quantum dots are well-known examples for inorganic imaging agents in fluorescence microscopy3 and iron oxide nanoparticles are currently in clinical trials for magnetic resonance imaging4. Purely organic platforms can be found in the superstructures of self-assembled small molecules such as liposomes5 or in polymeric systems such as polymer micelles6 and polymer vesicles (polymersomes).7 Consequently, hybrids are a combination of both types of materials into a single DDS. In the following, polymeric systems capable of delivering therapeutically active substances will be described in more detail. 1. Polymer-Aided Drug Delivery 1.1. Polymer-Drug Conjugates A pioneering idea for the use of polymers in therapeutic applications was introduced by Helmut Ringsdorf in 1975, when he proposed the concept of “polymer therapeutics”.8 In his approach a polymer chain is used to covalently connect several functional molecules, Chapter 1 – Introduction 14 such as a therapeutic drug and a targeting moiety, into one multi-functional macromolecule (Scheme 1-1). The main advantage of such a combination as proposed by Ringsdorf was in the larger availability of the drug in the organism through an increase in blood circulation time due to the higher molecular weight and a better solubilization of poorly water-soluble drugs by linking them to hydrophilic macromolecules. The idea of attaching specific targeting molecules to increase the localization of the drug at the desired site was already included in this approach. Due to the ease of chemical modification of synthetic polymers, the original concept has been explored in much more detail and was considerably expanded to include a wide variety of macromolecular delivery systems for therapeutic applications.9-12 The design of biocompatible13 and degradable polymers14, 15 together with the use of chemical linkers, which release the coupled drugs at appropriate conditions have pushed the field forward and several polymer-drug conjugate formulations are tested in clinical trials or have already been approved for therapeutic use in humans.2, 16 Scheme 1-1. Ringsdorf´s model for a pharmaceutically active polymer-drug conjugate. Reprinted with permission.8 1.2 Important Concepts in Drug Delivery: Stealth Effect and EPR A highly successful synthetic material aiding in the transportation of drugs inside living organisms is poly(ethylene glycol) (PEG). Many DDS are ultimately formulations containing PEG in some form. It is an uncharged, water-soluble polymer that is Chapter 1 – Introduction 15 extensively used in many kinds of consumer products, such as shampoos, crèmes, gels, etc. and is synthesized by anionic polymerization of ethylene oxide. Therefore, PEG is also referred to as poly(ethylene oxide) (PEO) and both names are used regularly. For therapeutic applications PEG is an ideal polymer, because it is non-toxic, biocompatible and has protein repellent properties. Compared to other polymers with similar properties, i.e. polyoxazolines,17 it has the advantage of FDA (U.S. Food and Drug Administration) approval for many applications, consequently triggering an extensive use of PEG polymers in therapeutic problems and resulting in its status as “gold standard”.18 Originally, the term of PEGylation referred to the conjugation of therapeutically active molecules (small molecule drugs, proteins, peptides, DNA, etc.) to PEG chains in order to increase the solubility in water and to protect the respective molecule from degradation, e.g. by enzymes, or immunogenic recognition.19 In the meantime, however, the term is being broadly applied to many kinds of PEG containing structures used in a medical context, for example organic and inorganic nanoparticles, surface coatings and polymeric micelles.20, 21 Many of the desired properties found for PEG-drug conjugates are also occurring for PEGylated particles, where the PEG coating leads to increased circulation time in the bloodstream and a reduction of non-specific interactions especially with proteins, which is summarized in a so-called “shielding” or “stealth effect” of PEG. Depending on the designated application, the molecular weight and grafting density (mushroom or brush conformation22) of PEG chains have to be adjusted. Connected to the beneficial properties conferred by PEG on DDS, a passive targeting mechanism occurring for solid tumors in cancer therapy, known as the enhanced permeation and retention (EPR) effect, has been largely responsible for the success of nanoscopic DDS in that field.23 It was first described by Maeda et al.24 in 1984 and is based on the specific vascular structure in solid tumors, characterized by a high vascular permeability together with an impaired lymphatic drainage as compared to normal tissue, which allows DDS of a specific size range to preferentially accumulate inside the tumor tissue. In order to make use of the EPR effect, the nanoscopic carrier system needs to have a prolonged blood circulation time in the order of several hours, usually associated with a size in between approximately 10 – 500 nm and a dense shielding layer preventing protein adsorption.25 Molecules with a size below 10 nm are rapidly cleared through the kidney, while larger particles up to 15 µm primarily accumulate in liver and spleen and are cleared through the reticuloendothelial system (RES).26 However, if both the surface Chapter 1 – Introduction 22 drugs at once are needed. For the synthetic effort necessary in creating such systems to be worthwhile, the full potential of a selective drug release of several drugs independent of each other and in specific cellular locations must be realized in the future. Chapter 1 – Introduction 23 2. Delivery of Genes “Gene therapy” summarizes a therapeutic approach for the treatment of inheritable or acquired diseases, where defective or missing genes are replaced by an appropriate exogenous gene introduced into the cell.66, 67 Also the down-regulation (silencing) of overactive genes by means of RNA interference can offer promising and previously unavailable therapeutic tools.68 A human gene therapy approach is especially promising for certain types of diseases that are otherwise difficult or even impossible to treat so far. These include genetic diseases like severe combined immunodeficiency, hemophilia or cystic fibrosis, but also acquired illnesses like cancer or AIDS could potentially be addressed with gene therapy.66 One of the biggest challenges for gene therapy is the effective introduction of foreign genetic material into the target cell, because several hurdles exist that protect against just such an event. Consequently, significant efforts have been made in creating delivery vehicles (“vectors”) for nucleic acids that can effectively overcome the different cellular barriers and successfully deliver their cargo to the target (nucleus or cytosol). Inevitably, the delivery of genes for fighting disease in patients will need delivery vehicles capable of delivering their cargo in a complex living organism and not just in cell culture. This environment offers special conditions, which additionally complicate the process of gene delivery before the vector has even reached the target cell and encounters its inherent defense mechanisms. It follows that the same challenges discussed above for in vivo drug delivery apply here as well. However, the intentional modification of cells, summarized under the title of genetic engineering, by introducing recombinant DNA in vitro is an important field in its own right. Many drug molecules are difficult to produce by fully synthetic means because of multi-step procedures requiring protective groups, stereo-selective catalysis and multiple purification steps. Cells can produce these drugs by enzymatic means with absolute stereo-selectivity and in high quantity when genetically engineered to do so. Since the cells are cultivated in cell culture in a bioreactor, many of the restrictions that apply to an in vivo delivery system are absent for this application. Therefore, even vectors which might not be suitable for a gene therapy approach can still be very effective for the delivery of genes in vitro where matters of delivery efficiency and production cost might be more important than blood circulation times or organ distribution. Chapter 1 – Introduction 24 All delivery vehicles, regardless of their final use, are usually divided in two categories, namely viral and non-viral vectors. Some significant differences between the two types exist, leading to specific challenges that need to be overcome before a widespread use can take place. 2.1 Viral Vectors Viruses are heavily used as delivery vehicles for genetic material, because they evolved specifically to introduce their own genome into the host cell. As a result, the delivery efficiency of viral vectors is very high and in many cases specific towards a certain cell line. In addition, viruses can stably incorporate the transported DNA sequence into the host cell genome ensuring continued gene expression, which can be beneficial in the treatment of certain diseases. However, some drawbacks exist for viral vectors. The most significant drawback is the possibility of an immune response of the organism, posing a significant risk to the patient. Also, a repeated application of viral carriers in the same patient could result in a loss of transfection efficiency, due to recognition of the virus by the immune system. Furthermore, large-scale production and chemical modification of the virus capsid is challenging, the latter potentially leading to a change in the selfassembly of the three dimensional virus structure. Some restrictions on the size of the transported DNA can also apply, due to the highly defined structure of the virus. Nevertheless, most clinical trials performed on gene therapy so far have used viral vectors.69, 70 2.2 Non-Viral Vectors Alternative transfection methods, often based on synthetic molecules or artificial particles, have been developed in recent years to overcome the above-mentioned drawbacks of viral vectors. Amongst those are electroporation, the “gene gun” - metal nanoparticles coated with DNA that are shot into the cells - as well as cationic molecules such as lipids and polymers.71 All of these methods are generally considered to be less effective and selective in delivering their genetic cargo as compared to their viral competitors and in most cases only a transient expression of genes can be achieved, since they are lacking mechanisms to permanently introduce the transported DNA sequence Chapter 1 – Introduction 25 into the host genome. For some non-viral systems, but especially in polymeric vectors, a significant cellular toxicity is observed in many cases. Still, significant effort has been put into improving the performance of non-viral vectors, aided by the ease with which chemical modifications can be made on the systems. 2.3 Barriers to a Successful Delivery of Genes The process of a successful delivery of nucleic acids to the target site in the cell with special consideration of the barriers encountered en route is exemplarily described in the following by example of a polycationic non-viral gene delivery vector. On its own, the uptake of free DNA in cells is inefficient, because of the size and negative charge of the DNA. Consequently, methods that make use of delivery mechanisms inherent to the cell (liposomes and polymers) rather than brute force (gene gun and electroporation) first need to compact the DNA to particles of smaller size and neutralize the negative charge of the DNA phosphate groups. Both processes occur when polycations form polyion complexes with nucleic acids (Figure 1-2, I). Such complexes are also termed as “polyplexes” or “inter-polyelectrolyte complexes” (IPECs) and additionally protect the DNA against enzymatic degradation. The properties of polyplexes like size and charge are mostly influenced by the choice of the polycation and do not strongly depend on the DNA used, but even if these physicochemical parameters of polyplexes are known, it is not possible to predict their transfection behavior.70 When mixing DNA and polycations to form polyplexes, generally, an excess of polycation (calculated as the ratio of positive to negative charges or nitrogen over phosphorus, N/P) is used and this leads to the formation of polyplexes with an overall positive charge. These can now efficiently bind to the negatively charged cell membrane, significantly enhancing their cellular uptake in vitro (Figure 1-2, II). Cellular uptake of the polyplexes depends on a multitude of factors such as size, charge, surface chemistry, the presence of targeting functions and possibly also mechanical properties of the polyplexes. It constitutes one of the most critical steps in gene delivery, although a high polyplex uptake is by no means a guarantee for a strong transgene expression. The uptake behavior is also known to vary with cell type, where different pathways may be used, which complicates matters further.72 An uptake through an endosomal pathway is the most common mechanism of polyplex entry into cells. Chapter 1 – Introduction 26 Figure 1-2. Barriers to gene delivery – Design requirements for gene delivery systems include the ability to (I) package therapeutic genes; (II) gain entry into cells; (III) escape the endo-lysosomal pathway; (IV) effect DNA/vector release; (V) traffic through the cytoplasm and into the nucleus; (VI) enable gene expression; and (VII) remain biocompatible. Reprinted with permission.72 Once the polyplex is inside of an endosomal compartment in the cell, it needs to be released into the cytosol and further travel to the nucleus in case of DNA delivery (Figure 1-2, III), while for siRNA delivery reaching the cytosol is often sufficient. During the transport of endosomes to lysosomes ATP-mediated proton pumps acidify the interior of the endosome, finally leading to the activation of nucleases capable of degrading the DNA. Several polymeric vectors make use of this acidification for an endosomal escape. This can be either through a controversially discussed73 process called “proton sponge effect” (PSE)74, 75 where the high buffering capacity of polycations like poly(ethylene imine) (PEI) leads to a water influx and finally bursting of the endosome. Other polymeric systems have used cell penetrating peptide sequences76 or synthetic functions77 mimicking these peptides for interrupting the endosomal membrane and facilitating a release of the polyplex from the endosome. Also the co-delivery of small molecules assisting in endosomal escape has been shown. Even after a polyplex has successfully entered the cytosol of the target cell, there are still several barriers to overcome, before the DNA has reached its designated destination. The mobility of macromolecules and particles, such as polyplexes, is strongly hindered in the cytosol, i.e., due to the cytoskeleton and the high viscosity of the cytosol originating from Chapter 1 – Introduction 27 the high protein content. Viral vectors can rely on active transport mechanisms such as microtubules, which are generally not available for polymeric vectors unless still enclosed in the endosome. Instead polymeric vectors have to form complexes of small size to increase mobility. Nucleic acid degradation is a reoccurring problem in the cytosol and a premature release of the DNA from the complex is detrimental for transfection efficiency mainly for reasons of rapid DNA degradation. However, for the genetic information of the DNA to be accessible it must eventually be released from the polyplex, otherwise transgene expression is hindered. Finding a system with a balanced binding strength towards the DNA, where on the one hand stable complexes that protect the DNA from degradation are formed while still allowing for a release at the correct timepoint, is critical. The final barrier for DNA delivery to eukaryotic cells is the nucleic membrane and the DNA has to cross it in order to induce a successful expression of the desired gene (Figure 1-2, V). Pore complexes can actively transport even large molecules through the membrane and into the nucleus, but appropriate targeting ligands have to be presented in order to activate these pore complexes. Another possibility is during cellular division, where the nucleic membrane breaks down for a short period of time. The lack of pore complex activating peptides is the reason why many polymeric vectors are only efficient in dividing cell lines, while transfection performance drops severely in non-dividing cell lines. Many of the processes involved in the successful delivery of exogenous genetic material to a target cell are not fully understood yet. It is therefore difficult to make any prediction on the transfection performance of a new polymer vector and in many cases careful mapping of the parameter space is necessary to find optimal transfection conditions. Since performance can vary strongly with each different cell line and also depends on factors like the presence or absence of serum proteins in vitro, this process has to be repeated for each vector when applied to a new target. As mentioned above, the delivery of genes under in vivo conditions constitutes a formidable challenge for most transfection systems, especially for non-viral ones. Polyplex stability can be a problem under physiological conditions (increased salt concentration, competing polyions, serum proteins, etc.) and dissociation of the complex Chapter 1 – Introduction 28 can lead to DNA degradation, while an aggregation of polyplexes or the association with proteins in the bloodstream can lead to fast clearance. For intravenous injection of polyplexes the problem of sufficient accumulation at the target site and avoidance of recognition, already discussed for anti-cancer-drug carrying polymer micelles, arises as well. Some solutions are found in the charge neutralization and shielding of the polyplexes, for example through PEGylation, although this generally results in lower uptake efficiency in the target cells. A direct injection of the polyplexes to the target tissue can overcome the problems associated with insufficient accumulation; however, it does not automatically guarantee a successful gene transfection. 2.4 Polymeric Vectors Many different polymers with positive charges have been tested for their potential in transfection. Most of these polymers have amino groups as the charge bearing species and some typical representatives of polymeric non-viral carriers are depicted in Figure 1-3. Polylysine was one of the first polymers to be used for the transfection of cells and has reached pre-clinical trials as a block copolymer with PEG.78, 79 PEI in both its branched and linear architectures has shown remarkably high transfection efficiency and is considered as the “gold standard” for polymeric gene delivery.75 Many chemical modifications have been proposed to further increase the transfection efficiency and especially at addressing the problem of its rather high cytotoxicity.80 Several commercially available transfection reagents such as ExGen500 and jetPEI use linear PEI in their formulations. Szoka et al. first used poly(amido amine) dendrimers for nucleic acid transport and several different dendritic molecules have been studied in detail because of their branched architecture and highly defined structure.81, 82 Recently, some reports used phosphonium containing polymers as an alternative material, proving their principal capability for the delivery of genes in vitro, while exhibiting rather low cytotoxicity as compared to ammonium containing polymers.83-85 Chapter 1 – Introduction 29 Figure 1-3. Polymer structures of polycations regularly used as non-viral transfection agents. To better understand the individual mechanisms that govern a successful delivery of genes to the target cell, it is a necessity to have well-defined starting materials. Almost all synthetic polymers have a distribution of their molecular weight and therefore represent a mixture of individual polymer chains of varying length. Furthermore, reproducibility is only given up to a certain point, since both molecular weight and molecular weight distribution can vary significantly between individual batches. If the distribution of a polymer sample is broad, it becomes difficult to discern between the individual contributions of the different chains making up the overall mixture. Therefore, polymers with a narrow molecular weight distribution, a good control over the molecular weight and high reproducibility between batches is to be preferred. Additionally, access to different polymer architectures can be interesting, since material properties can be significantly influenced through the three-dimensional connection of the monomers in a polymer. By using monomers that can be polymerized with controlled or living polymerization methods, such different polymer architectures in combination with narrow molecular weight distributions have become available.86, 87 With the resulting polymers of defined molecular weight and controllable architecture, the aim is to establish structureproperty relationships for polymeric vectors and possibly even define some guidelines for the design of successful polymeric gene carriers.88 Poly(2-(dimethylamino)ethyl methacrylate) (PDMAEMA) is a polycation that can be easily prepared by living polymerization methods like reversible addition fragmentation chain transfer (RAFT) polymerization,89 atom transfer radical polymerization (ATRP),90 Chapter 1 – Introduction 30 as well as anionic polymerization91 and has shown promising performance as a non-viral gene carrier.92 Similar to many other polymeric gene transfection agents an increase in cytotoxicity that is coupled with an increase in transfection efficiency could be observed. Several studies on the influence of the molecular weight on transfection efficiency and toxicity have also been performed.93, 94 Much effort has been put into decoupling an effective transfection from a significant cytotoxicity. At the very least a decrease of the toxicity to bearable levels, while maintaining a high transfection efficiency has been the aim of several studies. It is believed that part of the toxicity stems from the nondegradable nature of the PDMAEMA. Several groups have therefore prepared DMAEMA containing polymers that can be cleaved into smaller building blocks under physiological conditions, albeit with ambiguous results on both transfection efficiency and toxicity.95-98 A copolymerization of PDMAEMA with other monomers or conjugating PEG to PDMAEMA could successfully reduce the toxic effects of the polymers in many cases. However, not in all cases did a reduction of the toxicity also lead to an increase in transfection efficiency.99 The influence of architectural changes to the PDAMEMA structure was also tested for improving gene delivery efficacy. Non-linear polymer architectures can be synthesized for PDMAEMA for example by using multi-functional initiator molecules. Both star-shaped100-102, branched103, 104 and cylindrical brush-like105 polymers were subsequently tested for their transfection performance and showed superior results as compared to their linear counterparts. Georgiou et al. were the first to use star-shaped PDMAEMAs for gene transfection. They prepared the polymer through an “arm-first” method with a crosslinking monomer introduced into the group transfer polymerization of DMAEMA.106-108 Several other groups have made use of such branched structures in the meantime. In summary, PDMAEMA offers many possibilities to elucidate the mechanisms of gene transfection, due to its chemical versatility. Despite these tremendous improvements in controlling the polymer architecture and molecular weight distribution, many open questions remain unanswered for the moment. Controversial and even conflicting results are repeatedly reported in the literature. In many cases a comparison between different studies is difficult, because transfection protocols, cell lines, reporter genes and material characterization methods are chosen based on preference or availability by the respective groups rather than by standardized rules. This is a major drawback that needs to be addressed in the future in order to better coordinate the individual efforts of each group. Chapter 1 – Introduction 31 As an outlook for nanomedicine, polymeric materials have a good chance of significantly contributing to the field in the future. Polymers are chemically versatile and can easily be tailored to their respective use by skilled synthetic chemists. However, to be able to correctly design the materials, polymer chemists and material scientists in general need to closely collaborate with scientists from other disciplines like biology, biochemistry, pharmaceutical and medical sciences, who have a deep understanding of the processes and related challenges specific to living organisms. Only through a clever design of materials that takes into account all the available knowledge of the different disciplines will a significant improvement become possible, rather than hoping for a “lucky shot” from a single discipline. Chapter 1 – Introduction 38 Chapter 2 – Overview over the Thesis 39 Chapter 2 - Overview over the Thesis This thesis consists of seven chapters including five publications, which are presented in Chapters 3 to 7. Structural characterization of polyelectrolyte nanostructures in aqueous solution and the therapeutic use arising from these materials is the common topic unifying the different chapters. Two different types of polymeric materials were used in the work, which can be divided into star-shaped polycations on one side and multicompartment micelles (MCMs) from ionic triblock terpolymers on the other. In close collaboration with the group of Process Biotechnology at the University of Bayreuth, we explored the biological properties of star-shaped polycations for the delivery of genetic material (transfection) into eukaryotic cells. To better understand the relationship between chemical modifications on the molecular level and biological properties relevant to the transfection process, i.e., cytotoxicity and transfection efficiency (TE) were the main criteria of interest, I synthesized a variety of different polymer structures that were subsequently tested against several types of cell lines. We found general design criteria for star-shaped vectors (Chapter 3 and 4) and could use them to create materials with significantly enhanced transfection properties (Chapter 4) as compared to previously used polymers. In the second part of the thesis, from a starting material consisting of MCMs from triblock terpolymers which have a negatively charged corona in aqueous solution, I investigated the possibilities to alter the micellar structure towards higher complexity through the interaction with oppositely charged (block co-)polymers. Creating new compartments or changing the surface chemistry of the micelles by means of interpolyelectrolyte complex (IPEC) formation was of particular interest (Chapter 5 and 6). Some of these new MCMs were then used to investigate the influence of the corona chemistry on biological properties, while simultaneously demonstrating good drug carrying capacity sufficient for anti-cancer therapy in vitro and in vivo (Chapter 7). The most important results from each of the different parts are discussed in the following. Chapter 2 – Overview over the Thesis 40 1. Influence of Polymer Architecture and Molecular Weight of Poly(2- (Dimethylamino)ethyl Methacrylate) Polycations on Transfection Efficiency and Cell Viability in Gene Delivery Poly(2-(dimethylamino)ethyl methacrylate) (PDMAEMA) is an ideal candidate for investigating structure-property relationships in non-viral gene delivery, because welldefined polymers with varying morphologies can be synthesized by means of controlled polymerization methods such as atom transfer radical polymerization (ATRP).1 Starshaped polycations have only recently been used in gene transfection and there was some evidence for a superior transfection performance of this specific polymer architecture reported in the literature.2, 3 We were especially interested how structural parameters, such as the arm number and arm length of star-shaped polycatione, influence their transfection properties. By using ATRP, I prepared a material library consisting of linear, 3-arm-, and 5-arm-star polymers with different molecular weights for each type of polymer. The two star-shaped DMAEMA polymers were polymerized in a “core-first” approach using multifunctional initiator molecules on the basis of sugars (glucose and saccharose). The final polymer library contained twelve polymers in total and covered a molecular weight range from 16 to 158 kDa (see Table 2-1). Chapter 2 – Overview over the Thesis 41 Table 2-1: Nomenclature and molecular characterization of the linear, 3-arm and 5-arm DMAEMA polymers synthesized with ATRP. Architecture Name N / Moleculea Mn [kDa]b PDIc Linear L110 108 17.1 1.12 L520 518 81.7 - L880 881 138.7 1.73 L1000 1003 157.9 1.80 3-Arm S-395 95 15.9 1.26 S-3210 210 34.0 1.16 S-3300 296 47.5 1.12 S-3600 596 94.7 1.12 S-3710 710 112.6 1.13 5-Arm S-5580 575 91.9 1.09 S-5700 699 111.4 1.12 S-5920 919 146.0 1.10 acalculated from the NMR-molecular weight; bdetermined from NMR conversion data; cmeasured via SEC with DMAc as eluent and poly(methyl methacrylate) as standard. The same polymerization conditions were used for all three types of polymer architectures and well-defined star-shaped PDMAEMAs with narrow molecular weight distributions were obtained. Then, polyplexes were formed with plasmid DNA (pDNA) in increasing ratios of PDMAEMA-nitrogen/DNA-phosphate (N/P ratio) to test them for their cytotoxicity as well as TE in Chinese Hamster Ovary (CHO-K1) cells. For ease of data analysis, the concentration dependent cytotoxicity data for each polymer sample was converted into a single value, defined as the lethal complex concentration for 50 % of the cells (LCC50) which could then be plotted against the molecular weight (Figure 2-1a). Interestingly, a trend was found that points towards a reduced cytotoxicity (high LCC50 value) with increasing degree of branching for comparable molecular weights, i.e., 5-arm stars are less toxic than 3-arm stars which in turn are less toxic than linear polymers. Chapter 2 – Overview over the Thesis 42 Figure 2-1: (A) Plot of the LCC50-values of various PDMAEMAs against the molecular weight of the polymer. LCC50-values were calculated from MTT experiments using CHO-K1 cells. The symbols represent linear (squares), 3-arm star (triangles) and 5-arm star (stars) PDMAEMA. (B) Fraction of transfected cells plotted against the relative viability for the polyplexes from DMAEMA polymers at N/P ratios 2 (black), 5 (dark grey), 10 (light grey) and 20 (open symbols). The average number of monomers per polycation is given next to each entry. Data represent mean value of three independent experiments. The arrow represents the general course of a polycation through the graph with increasing N/P ratio (from grey = low to black = high N/P ratio). Cellular toxicity can have a direct influence on the TE, as cell death decreases the production of reporter genes. It is therefore reasonable to directly connect the toxicity of a Chapter 2 – Overview over the Thesis 43 given N/P ratio with the corresponding TE. To have both of these important parameters visualized in one graph, we introduced a plot of the TE against relative viability for all N/P ratios in the polymer library as depicted in Figure 2-1b. This new type of graph allows the reader to quickly identify the polymeric material and N/P ratio with ideal transfection properties, i.e. the data points that appear in the upper right quadrant number 2. Additionally, this type of graph nicely illustrates the effect of increasing the N/P ratio for a certain DMAEMA polymer, starting with low toxicity and TE, then going through a tradeoff region between increasing TE and toxicity, while finally the toxicity dominates and results in a decreased TE. This behavior is illustrated by the arrow in Figure 2-1b. From the dataset we could further find that polymers below a critical molecular weight of approximately 20 kDa (corresponding to 130 monomer units in case of PDMAEMA), exhibit no significant transfection, suggesting that polymers with an intermediate molecular weight and a branched architecture would be good candidates for gene delivery. 2. Nano-Particulate Non-Viral Agent for the Effective Delivery of pDNA and siRNA to Differentiated Cells and Primary Human T Lymphocytes With the knowledge from the polymer library containing linear, 3and 5-arm star DMAEMA polymers described in Chapter 2.1, we then tested the transfection properties of a star-shaped PDMAEMA with 20 arms (Si-PDMAEMA). This polymer had been obtained again by ATRP “core-first” method using a multifunctional silsesquioxane nanoparticle as the initiator (Scheme 2-1a). Chapter 2 – Overview over the Thesis 44 Scheme 2-1. (A) Chemical structure of star-shaped Si-PDMAEMA synthesized via “core-first” method by ATRP; (B) Star-like PDMAEMA micelle self-assembled from amphiphilic PB-b-PDMAEMA diblock copolymer. Despite its large molecular weight (Mn = 730 kg/mol), we could achieve very good cellular viability coupled with extremely high transfection efficiencies in CHO-K1 cells (74 % TE) with 93 % rel. viability on average that surpassed the best results from the “gold standard” poly(ethylene imine) (PEI) (50 % TE with 94 % rel. viability on average). In various cell lines that are generally considered to be more difficult to transfect than CHO-K1 cells, the Si-PDMAEMA performed better relative to PEI as is exemplarily shown for C2C12 cells (Figure 2-2). These cells stop dividing when the culture plate is densely populated (confluent), which significantly hinders successful transfection for most polymeric vectors. Furthermore, the confluent C2C12 cells could easily be differentiated into myotubes through a change of the culture medium composition. In all of the cases Si-PDMAEMA (Figure 2-2, open symbols) gave higher TE for every single transfection experiment as compared to PEI (Figure 2-2, closed symbols). Chapter 2 – Overview over the Thesis 45 Figure 2-2. Analysis of the percentage of transfected cells against the relative viability after transfection in C2C12 cells. Transfection efficiencies in dividing (, ), non-dividing myoblasts (, ) and myotubes (, ) are plotted against the viability. Black symbols: PEI, white symbols: Si-PDMAEMA. Data shown are from individual transfections. The superior transfection performance of Si-PDMAEMA is based on its general architecture, where a multitude of polymeric arms emanate from a common center, rather than the specific polymer sample: When polymeric micelles from an amphiphilic diblock copolymer (polybutadiene-block-PDMAEMA; PB-b-PDMAEMA) comprising a PB core and PDMAEMA corona, which resemble a star-shaped architecture (Scheme 2-1b), were used for gene delivery, a comparable performance to that of Si-PDMAEMA was found. Both types of nanostructures could also transfect human T lymphocytes with pDNA more efficiently as compared to the standard method of electroporation for these cell types. Additionally, RNA interference could also successfully be performed with both polymeric vectors of star-like architecture, reaching up to 40 % silencing efficiency of the targeted gene in human T lymphocytes. The remarkable transfection performance of this class of polymers, combined with the large number of different cell lines it can be applied to, especially for non-dividing or differentiated cells, makes these star-like polymers highly interesting materials for in vitro gene delivery. Chapter 2 – Overview over the Thesis 46 3. Double-Layered Micellar Interpolyelectrolyte Complexes – How Many Shells to a Core? MCMs are highly complex nanostructures with possible applications in templating, as nanoreactors, for carrying catalysts and in biomedical drug delivery or imaging. By using triblock terpolymers, it is possible to obtain MCMs with two chemically different compartments in the core of the resulting micelles. However, increasing the number of distinguishable compartments to more than two is generally very difficult. Here, we established a new route for the formation of distinguishable and chemically different compartments in MCMs from ionic triblock terpolymers by using the negatively charged corona of those micelles for IPEC formation with oppositely charged polyions. The original MCMs used in this work formed through self-assembly of poly(butadiene)- block-poly(1-methyl-2-vinyl pyridinium)-block-poly(methacrylic acid) (BVqMAA) in aqueous solution. The MAA corona stabilizes the micelles and is easily accessible for further reactions. First, we used quaternized PDMAEMA (PDMAEMAq), which retains its positive charge even at high pH values (pH = 10) where PMAA is fully deprotonated. Upon mixing the polycation solution with BVqMAA micelles, an IPEC was immediately formed, which collapsed onto the core of the micelles. This new IPEC compartment (2nd IPEC) comprising PDMAEMAq and PMAA could be distinguished in cryogenic transmission electron microscopy (cryo-TEM) measurements from the original intramicellar IPEC shell (Vq/MAA). Furthermore, a complete second layer was formed (Figure 2-3, left side) leading to an onion-type morphology. Non-complexed MAA from the original micellar corona served to stabilize the micelles in solution below a critical complexation ratio. The 2nd IPEC layer was found for 3 different block lengths of the MAA block in BVqMAA micelles, ranging from 345 to 550 units. The length of the two DMAEMAq homopolymers (157 and 820 units) used for complex formation did not significantly affect structure formation. If complexation ratios close to charge neutrality were reached, a macroscopic phase separation with precipitation of the MCMs was found. This macroscopic precipitation at high complexation ratio was avoided when instead of a homopolymer the double hydrophilic diblock copolymer of PEG-b-PDMAEMAq was used for IPEC formation (Figure 2-3, right side). The new layered compartment from PDMAEMAq/MAA IPEC developed as before, but the colloidal stability of the particles Chapter 2 – Overview over the Thesis 47 was retained even at charge neutralization, because the PEG segment served as stabilizing corona chains. Figure 2-3: Schematic pathway for the formation of double-layered IPECs from BVqMAA triblock terpolymer micelles and either PDMAEMAq homopolymers (left) or a PEG-b-PDMAEMAq diblock copolymer (right). Scale bars in the insets represent 40 nm. Chapter 2 – Overview over the Thesis 54 drug carrying capacity and therapeutic efficacy of BVqMAA MCMs, while the interactions of the micelles in biological surroundings could be tuned by controlling the corona composition. Chapter 2 – Overview over the Thesis 55 6. Individual Contributions to Joint Publications The results presented within this thesis were obtained in collaboration with other persons and were previously published. In the following the individual contributions of each coauthor are specified. The asterisk indicates the corresponding author(s) of the respective publication. Chapter 3 This work has been published in Biomacromolecules 12, pp. 4247-4255 (2011) under the title: “Influence of Polymer Architecture and Molecular Weight of Poly(2- (dimethylamino)ethyl methacrylate) Polycations on Transfection Efficiency and Cell Viability in Gene Delivery” by Christopher V. Synatschke, Anja Schallon, Valérie Jérôme, Ruth Freitag*, and Axel H. E. Müller* This work was conducted in collaboration with the chair of “Process Biotechnology” at the University of Bayreuth. I synthesized all materials, conducted their physicochemical characterization and wrote the manuscript, except that: A. Schallon was involved in the planning of the experiments, performed all of the cellculture experiments and corrected the manuscript. V. Jérôme, R. Freitag and A.H.E. Müller were involved in scientific discussions and correcting the manuscript. Chapter 2 – Overview over the Thesis 56 Chapter 4 This work has been published in Biomacromolecules 13, pp. 3463−3474 (2012) under the title: “Nanoparticulate Non-Viral Agent for the Effective Delivery of pDNA and siRNA to Differentiated Cells and Primary Human T Lymphocytes” by Anja Schallon, Christopher V. Synatschke, Valérie Jérôme, Axel H. E. Müller , and Ruth Freitag* This work was conducted in collaboration with the chair of “Process Biotechnology” at the University of Bayreuth. I synthesized all materials, conducted their physico-chemical characterization and wrote parts of the manuscript, except that: A. Schallon was involved in the planning of the experiments, performed of the cell-culture experiments and wrote parts of the manuscript. V. Jérôme and R. Freitag were involved in scientific discussions, planning of the cellculture experiments and wrote parts of the manuscript. A.H.E. Müller was involved in scientific discussions and corrected the manuscript. Chapter 5 This work has been published in Soft Matter 7, pp. 1714-1725 (2011) under the title: “Double-Layered Micellar Interpolyelectrolyte Complexes-How Many Shells to a Core?” by Christopher V. Synatschke, Felix H. Schacher*, Melanie Förtsch, Markus Drechsler and Axel H. E. Müller* I conducted all of the experiments and wrote parts of the manuscript, except that: F.H. Schacher synthesized the BVT triblock terpolymers, wrote parts of the manuscript and was involved in the planning of the experiments. M. Förtsch and M. Drechsler conducted all of the cryo-TEM measurements. A.H.E. Müller was involved in scientific discussions and corrected the manuscript. Chapter 2 – Overview over the Thesis 57 Chapter 6 This work has been published in Macromolecules 46, pp. 6466-6474 (2013) under the title: “Micellar Interpolyelectrolyte Complexes With a Compartmentalized Shell” by Christopher V. Synatschke, Tina I. Löbling, Melanie Förtsch, Andreas Hanisch, Felix H. Schacher*, and Axel H. E. Müller*, I synthesized the BVT and BVqMAA polymers, was involved in the planning of the experiments and wrote the manuscript, except that: T.I. Löbling performed all other synthesis, physico-chemical characterization and corrected the manuscript. M. Förtsch conducted all of the cryo-TEM measurements. A. Hanisch assisted with BVT synthesis and corrected the manuscript. F.H. Schacher and A.H.E. Müller were involved in scientific discussions and corrected the manuscript. Chapter 2 – Overview over the Thesis 58 Chapter 7 This work has been published in ACS Nano (DOI: 10.1021/nn4028294) under the title: “Multicompartment Micelles with Adjustable Poly(ethylene glycol) Shell for Efficient in vivo Photodynamic Therapy” by Christopher V. Synatschke, Takahiro Nomoto, Horacio Cabral, Melanie Förtsch, Kazuko Toh, Yu Matsumoto, Kozo Miyazaki, Andreas Hanisch, Felix H. Schacher, Akihiro Kishimura, Nobuhiro Nishiyama, Axel H. E. Müller*, and Kazunori Kataoka* This work was done in collaboration with the group of Prof. Kazunori Kataoka at the University of Tokyo, Japan. I conducted all of the experiments and wrote the manuscript, except that: T. Nomoto was involved in planning of the experiments, conducted some of fluorescence microscopy measurements and corrected the manuscript. H. Cabral conducted part of the animal experiments, was involved in the planning of experiments and corrected the manuscript. M. Förtsch conducted all of the cryo-TEM measurements. K. Toh and Y. Matsumoto conducted fluorescence microscopy measurements on animals. K. Miyazaki assisted with in vivo PDT efficacy measurements and corrected the manuscript. A. Hanisch assisted with BVT synthesis and corrected the manuscript. F.H. Schacher, A. Kishimura, N. Nishiyama, A.H.E. Müller and K. Kataoka were involved in scientific discussions and corrected the manuscript. Chapter 2 – Overview over the Thesis 59 7. References 1. Matyjaszewski, K., Macromolecules 2012, 45, (10), 4015-4039. 2. Nakayama, Y.; Masuda, T.; Nagaishi, M.; Hayashi, M.; Ohira, M.; Harada-Shiba, M., Current Drug Delivery 2005, 2, (1), 53-57. 3. Nemoto, Y.; Borovkov, A.; Zhou, Y. M.; Takewa, Y.; Tatsumi, E.; Nakayama, Y., Bioconjugate Chemistry 2009, 20, (12), 2293-2299. Chapter 2 – Overview over the Thesis 60 Chapter 3 – Influence of Polymer Architecture and Molecular Weight on TE and Cell Viability 61 Chapter 3 Influence of Polymer Architecture and Molecular Weight of Poly(2-(Dimethylamino)ethyl Methacrylate) Polycations on Transfection Efficiency and Cell Viability in Gene Delivery The results of this chapter have been published in Biomacromolecules as: “Influence of Polymer Architecture and Molecular Weight of Poly(2- (Dimethylamino)ethyl Methacrylate) Polycations on Transfection Efficiency and Cell Viability in Gene Delivery” by Christopher V. Synatschke, Anja Schallon, Valérie Jérôme, Ruth Freitag, and Axel H. E. Müller Reprinted with permission from Biomacromolecules 2011, 12, 4247-4255. Copyright 2011 American Chemical Society. Chapter 3 – Influence of Polymer Architecture and Molecular Weight on TE and Cell Viability 62 Chapter 3 – Influence of Polymer Architecture and Molecular Weight on TE and Cell Viability 63 Abstract Nonviral gene delivery with the help of polycations has raised considerable interest in the scientific community over the last decades. Herein, we present a systematic study on the influence of the molecular weight and architecture of poly(2-(dimethylamino)ethyl methacrylate) (PDMAEMA) on the transfection efficiency and the cytotoxicity in CHOK1 cells. A library of well defined homopolymers with a linear and star-shaped topology (3and 5-arm stars) was synthesized via atom transfer radical polymerization (ATRP). The molecular weights of the polycations ranged from 16 to 158 kDa. We found that the cytotoxicity at a given molecular weight decreased with increasing number of arms. For a successful transfection a minimum molecular weight was necessary, since the polymers with a number-average molecular weight, Mn, below 20 kDa showed negligible transfection efficiency at any of the tested polyelectrolyte complex compositions. From the combined analysis of cytotoxicity and transfection data, we propose that polymers with a branched architecture and an intermediate molecular weight are the most promising candidates for efficient gene delivery, since they combine low cytotoxicity with acceptable transfection results. Chapter 3 – Influence of Polymer Architecture and Molecular Weight on TE and Cell Viability 70 Complex Formation pDNA/polymer polyplexes were prepared at room temperature using 3 µg pDNA and varied amounts of the respective polycation stock solution to achieve the indicated PDMAEMA-nitrogen/DNA-phosphate (N/P) ratios. For this purpose, the pDNA was diluted in a final volume of 200 µL of 150 mM NaCl-solution. The required polymer solution was added in a single drop to the pDNA solution and the mixture was immediately vortexed for 10 sec at full speed, followed by incubation at room temperature for 30 min. Transfection of Mammalian Cells Cells were seeded in 2 mL growth medium at a density of 2 x 105 cells/well in 6-well plates 24 h prior to transfection. One hour prior to transfection, cells were rinsed with PBS and supplemented with 2 mL OptiMEM. The polyplex suspension (200 µL) was added to the cells and the plates were centrifuged for 5 min at 200 g and placed for 4 h in the incubator. Afterwards, the medium was removed, 2 mL of fresh growth medium were added, and the cells were further incubated for 20 h. Cells were harvested by trypsinization and resuspended in PBS. The relative expression of eGFP fluorescence of 1 x 104 cells was quantified via flow cytometry using a Cytomics FC 500 equipped with the CXP Analysis research software (Beckman Coulter, Krefeld, Germany). The parameters of the device were set, so that the fluorescence intensity value of the control cells (non-treated) was below 100. All cells showing a fluorescence intensity above this value were recorded as transfected. MTT Assay (Cytotoxicity Studies) The cytotoxicity of the polyplexes at various N/P-ratios was evaluated in 96-well microtitre plates by the MTT assay following essentially the transfection protocol. The CHO-K1 cells were seeded in growth medium at a density of 2 x 104 cells/well 24 h prior to the experiment. One hour prior to the experiment, the medium was discarded and serum-free growth medium was added. Cells were incubated with the indicated polyplex preparation for 4 h, then the medium was replaced by serum-containing medium (analogously to the transfection protocol above). After 20h incubation, cells were rinsed with PBS and further incubated in 200 µL MTT solution (0.5 mg/mL in PBS) for 2 h. The Chapter 3 – Influence of Polymer Architecture and Molecular Weight on TE and Cell Viability 71 solution was aspirated, replaced with 200 µL DMSO and mixed at 150 rpm for 5 min to dissolve the formazan crystals produced in the reaction. Absorbance was then measured at 580 nm in a microplate reader (Genios Pro, Tecan GmbH, Crailsheim, Germany) with untreated cells serving as controls. The lethal complex concentration (LCC50) was defined as polymer concentration of the complex at which 50 % of metabolic activity could be measured. Statistical Analysis Group data are reported as mean ± SD. For transfection results, the Student’s t-test was used to determine whether data groups differed significantly from each other. Statistical significance was defined as having P-values < 0.05 for significance and P-values < 0.01 for great significance. To determine the significance of more than two groups of data, ANOVA was used with a defined P-value < 0.05 for significant differences. Chapter 3 – Influence of Polymer Architecture and Molecular Weight on TE and Cell Viability 72 Results and Discussion The aim of this work was to study the influence of polymer architecture and molecular weight on both transfection efficiency (TE) and cytotoxicity of DMAEMA-based polymers in gene delivery experiments. For this purpose, a library of three different polymer architectures, namely linear, 3-armand 5-arm stars, was synthesized by means of ATRP. We chose this controlled radical polymerization method, as it allows a precise control of the molecular weight of the resulting polymers, while ensuring narrow molecular weight distributions. This is in contrast to many other studies published in the pertinent literature, where the PDMAEMA used for the transfection experiments is frequently prepared via free radical polymerization. Such polymers typically have a broad molecular weight distribution, especially in case of larger molecules.18, 38 With such polydisperse samples it is almost impossible to discern the influence of the individual species. Instead meaningless average values are determined. Any systematic investigation of the influence of molecular parameters (architecture, molecular weight, etc.) on the gene delivery ability requires instead polymers with a minimal heterogeneity in these parameters. This becomes even more important for architectures with a higher complexity such as stars. Furthermore, when ATRP is used to prepare the polycations, functional end-groups remain on the polymer chains, which can be used for subsequent modifications such as fluorophore labelling, as was demonstrated in an earlier publication.28 Polymer Synthesis and Characterization The polymerization process of the linear and star-shaped samples is depicted in Scheme 3-1. We used a core-first approach with functionalized sugars (glucose43 and saccharose44) as initiators for the star polymers (Scheme 3-1B and C, respectively). Due to steric hindrance, the actual arm number per molecule is lower than the number of initiation sites. In our experience, the average arm-number for the glucose-based polymers is around three arms per molecule and we confirmed this for the S-3300 sample through alkaline cleavage of the arms, where an average number of arms of 3.1 was determined. PDMAEMA from the saccharose-based initiator usually has slightly more than five arms (5.4 – 5.6).22 A detailed description of the synthesis of the initiators and synthetic procedure for the cleavage of the arms has been published previously.42, 45 Chapter 3 – Influence of Polymer Architecture and Molecular Weight on TE and Cell Viability 73 Scheme 3-1: Synthesis of the PDMAEMA samples with linear (A), 3-arm (B) and 5-arm (C) star-shaped architecture, from the respective initiators via ATRP. Due to steric hindrance not all initiating sites of the star initiators can form polymer chains. The bottom line shows a schematic representation of the respective polymer architecture. Chapter 3 – Influence of Polymer Architecture and Molecular Weight on TE and Cell Viability 74 Table 3-1: Nomenclature and molecular characterization of the linear, 3-arm and 5-arm DMAEMA polymers synthesized with ATRP. Molecular weight averages were calculated from NMR conversion data and PDI was determined with SEC. Architecture Name N / Moleculea Mn [kDa]b PDIc Linear L110 108 17.1 1.12 L520 518 81.7 - L880 881 138.7 1.73 L1000 1003 157.9 1.80 3-Arm S-395 95 15.9 1.26 S-3210 210 34.0 1.16 S-3300 296 47.5 1.12 S-3600 596 94.7 1.12 S-3710 710 112.6 1.13 5-Arm S-5580 575 91.9 1.09 S-5700 699 111.4 1.12 S-5920 919 146.0 1.10 acalculated from the NMR-molecular weight; bdetermined from NMR conversion data; cmeasured via SEC with DMAc as eluent and poly(methyl methacrylate) as standard. As shown in Table 3-1, for each architecture we synthesized a series of polymers with increasing average molecular weights. The name of each sample is given as the polymer architecture in capital letters, while the subscripts denote the average number of monomers per molecule as determined from the NMR conversion data. SEC measurements confirmed controlled polymerization conditions, as polymers show a rather narrow molecular weight distribution, at least for the star-shaped polymers. One of the linear polymers (L520) could not be detected in the SEC measurements, even at high polymer concentrations (> 5 mg / mL). The reason for this remains obscure. Additionally, the polydispersity indices (PDI) of the linear polymers are considerably higher than those of the star-shaped samples. This could be an indication of an uncontrolled polymerization of the linear samples, caused for example through a complex formation between the growing polymer chains and the copper catalyst. Also, PDMAEMA is known to interact with the column material during size exclusion chromatography, which has to be suppressed through the addition of salt to the eluent. An interaction of the sample with the column material would also result in a broadening of the molecular weight distribution Chapter 3 – Influence of Polymer Architecture and Molecular Weight on TE and Cell Viability 75 and should be more pronounced for the linear samples, because of their less dense structure and increased hydrodynamic radius compared to the branched structures. For all further discussions, in particular the nomenclature of the polymers, the molecular weight determined via NMR was used. Cytotoxicity Studies by MTT Assay The polycation library was then used to systematically study the effect of both molecular weight and polymer architecture on cytotoxicity and gene delivery efficiency in mammalian cells. All experiments were performed using Chinese Hamster Ovary (CHO-K1) cells, as this cell line is well established in our group and also commonly used for recombinant protein production in the biopharmaceutical industry. The cytotoxicity of the polyplexes from the indicated polymers at different N/P ratios was determined by MTT assay. The results are shown in Figure 3-1, where the average of the three experiments is given as the residual cell viability relative to an untreated cell population. Chapter 3 – Influence of Polymer Architecture and Molecular Weight on TE and Cell Viability 76 Figure 3-1: Relative viability of CHO-K1 cells incubated with polyplexes made of linear, 3-arm and 5-arm PDMAEMA of varying molecular weights at increasing N/P ratios (1, 2, 5, 10, 20 and 50). Incubation period was 4 h with polyplexes in serum-free media and 20 h in growth media at a cell seeding density of 2 x 104 cells/well. The results are expressed as a percentage of the control cell culture. Data represent mean ± SD, n ≥ 3. The Students t-test was used to determine the N/P ratios that significantly differ from 100 % viability (*, P < 0.05; #, P < 0.01). As shown in Figure 3-1, all polyplexes became toxic at a sufficiently high concentration of the polycation (N/P ratio), as can be seen from the decreasing average relative viability of the cells. A decrease of viability with increasing N/P ratio can be seen for all polymers used for complex formation except the ones with the lowest molecular weights, namely L110 and S-395, which show the lowest cytotoxicity (highest residual cell viability) at intermediate N/P ratios. Both low molecular weight polymers showed no significant cytotoxicity (P < 0.01) in the Students t-test. Only at N/P = 50 for L110 and at N/P = 20 for S-395 a significant difference (P < 0.05) was observed in the Students t-test compared to 100 % viability, indicating a slightly toxic behavior. Polycations in general are toxic to cells as they interact with various important anionic species found in biological systems including the membrane lipids, (poly)nucleotides and many proteins.39, 46, 47 As the N/P-ratio increases, the fraction of non-complexed polycations in the Chapter 3 – Influence of Polymer Architecture and Molecular Weight on TE and Cell Viability 77 preparation and hence the cytotoxicity increases. The smallest polycations included in our investigation on the other hand show low general toxicities and can be considered quasi non-toxic in the investigated range. In order to simplify the amount of data and allow a better comparison between the individual polymers as well as the different architectures, the LCC50 (50 % lethal concentration of complexes at a fixed DNA concentration) value was determined for each polyplex solution. The LCC50 value then indicates the polycation concentration where the viability compared to the control cells reached 50 %. The value was extrapolated from a plot of the viability against the polymer concentration (see Schallon et al.28 for details). In Figure 3-2 the LCC50 values are plotted against the molecular weight of each polymer in a double logarithmic scale. A smaller LCC50-value represents a more toxic polymer, since a lower molar polycation concentration is necessary to reduce viable cell numbers to 50 %. A steady decrease of the LCC50 value with increasing molecular weight of the polycations can be seen from Figure 3-2. Figure 3-2: Plot of the LCC50-values (50 % lethal complex concentration) of various PDMAEMAs against the molecular weight of the polymer in a double logarithmic scale. The symbols represent linear (square), 3arm star (triangle) and 5-arm star (star) PDMAEMA. LCC50-values were calculated from MTT experiments with CHO-K1 cells. Chapter 3 – Influence of Polymer Architecture and Molecular Weight on TE and Cell Viability 78 More importantly, when the different architectures are compared for a similar molecular weight, the 5-arm star polymers show the lowest toxicities, followed by the 3arm stars and finally the linear polymers. Our data therefore hints towards a decrease of cytotoxicity of PDMAEMA with increasing arm-number. In this context, a study by Newland et al. is of interest.48 The authors describe a highly branched PDMAEMA prepared through the copolymerization of DMAEMA with ethylene glycol dimethacrylate. In their experiments, they observed a decrease in the cytotoxicity of irregularly branched PDMAEMA compared to linear PDMAEMA, which suggests that polycation cytotoxicity may decrease not only with increasing arm-number as indicated by our results, but in general as a consequence of increased branching. Similar trends were also described by Xu et al. for different types of branched DMAEMA polymers.25, 26, 37 The reasons for the cytotoxicity of polycations in general and PDMAEMA in particular, have been investigated in several studies. It was found that polycations have a tendency to interact with the cellular membrane and membrane proteins.18, 46 A decrease in the membrane potential was observed, which points towards the formation of holes in the cellular membrane.49 Furthermore, an interaction of the polycation with important proteins and RNA in the cytosol has been speculated upon.11 The general conception is that polymers carrying more charges per molecule, e.g. larger polycations, are more toxic, because they have a stronger tendency to bind negatively charged peptides and eventually precipitate within the cytosol.17, 18 In a recent study with well-defined linear PDMAEMA from ATRP it was found that these polymers induce cytotoxicity through a cooperative effect from both membrane disruption and apoptosis.39 However, the specific mechanism may also depend on the cell type.47 As a consequence of this observation, several groups have synthesized degradable polycations made from small building blocks, which indeed showed a reduced cytotoxicity compared to their non-degradable analogues.50-53 However, such an explanation cannot be applied here, since the polymers used in this study are not biodegradable. The reduced cytotoxicity is therefore very likely a result of the unique star-shaped architecture of those molecules. A star-shaped molecule has the highest density in the core, which decreases with increasing distance from the core.54 The interaction of the various positively charged nitrogen atoms in such a star with rather flexible polyanions such as pDNA should not be hampered significantly. In fact little differences can be observed in the stability of polyplexes from the same type of Chapter 3 – Influence of Polymer Architecture and Molecular Weight on TE and Cell Viability 79 polycation, regardless of its architecture.55 In case of a putative interaction with a globular protein molecule or even more importantly the phospholipids in the cellular membrane, however, we propose that not all nitrogens in a star-shaped polymer can participate in the interaction, since those from the inner part of the star would be sterically excluded. Since interaction with the phospholipids and the concomitant formation of holes in the membrane is an important mediator of cellular cytotoxicity, branched and in particular star-shaped and dendritic polymers should be less toxic than their linear counterparts, while transfection efficiency would not necessarily be affected. Transfection Studies In order to investigate the structure-function-relationship in regard to gene transfer capability, the polycations were then evaluated as transfection agents. We chose the eGFP transgene and flow cytometry for analysis, since the choice of a reporter gene assay proved not to be crucial for the experimental outcome, as recently reviewed by van Gaal et al.56 In order to successfully transport genetic material into the nucleus of eukaryotic cells, several barriers have to be overcome.7 First, the DNA has to be condensed by the polycation into small and positively charged particles, so called polyplexes. These polyplexes then have to be transported into the cell through the cellular membrane. The commonly accepted mechanism for the uptake of polyplexes by cells is via endocytosis.19, 57 Alternatively, it was proposed that some polycations (e.g. dendrimers) have the ability to directly penetrate the cellular membrane, as could be shown in experiments on model membranes.58 In any case, after crossing the cellular membrane the polyplex needs to protect the genetic material from degradation. Furthermore, the polyplex has to be transported to the nucleus before the genetic material crosses the nuclear membrane. When all of these barriers have successfully been overcome, the genetic information can be processed leading to the expression of transgenes. In our case, the transgene was the enhanced green fluorescent protein (eGFP). Successfully transfected cells appear green in this case and thus are easily detected by flow cytometry. From these data, the transfection efficiency (TE) was calculated as percentage of green fluorescent cells within the total cell population analyzed, Figure 3-3. N/P ratios above values of 20 were not tested, as the cytotoxicity of the polycations at N/P = 20 increases (see Figure 3-1) generally resulting in a low TE. Chapter 3 – Influence of Polymer Architecture and Molecular Weight on TE and Cell Viability 86 significant transfection. Additionally, at least for the star-shaped polymers, an increase in molecular weight does not necessarily result in an improved gene delivery. Our data suggests, that an ideal combination between low cytotoxicity and high transfection should be achievable with a branched structure displaying an intermediate molecular weight. In the future, it needs to be clarified, whether star-shaped polymers with more than 5 arms can further reduce the cytotoxicity, which could lead to an increase in transfection efficiency. Also, other structures of e.g. irregularly branched polymers should be tested. Acknowledgements C. V. Synatschke gratefully acknowledges funding by the state of Bavaria through a BayEFG scholarship and ongoing support by the Elite Network of Bavaria (ENB). The authors would like to thank D. V. Pergushov (Moscow), A. Kishimura (Tokyo) and C. B. Tsvetanov (Sofia) for helpful discussions. M. Böhm is acknowledged for performing the SEC measurements. Chapter 3 – Influence of Polymer Architecture and Molecular Weight on TE and Cell Viability 87 References 1. de Smedt, S. C.; Demeester, J.; Hennink, W. E Pharmaceutical Research 2000, 17, 113-126. 2. He, C.-X.; Tabata, Y.; Gao, J.-Q. International Journal of Pharmaceuticals 2010, 386, 232-242. 3. Jeong, J. H.; Kim, S. W.; Park, T. G. Progress in Polymer Science 2007, 32, 12391274. 4. Treco, D. A.; Selden, R. F. Molecular Medicine Today 1995, 1, 314-321. 5. Wong, S. Y.; Pelet, J. M.; Putnam, D. Progress in Polymer Science 2007, 32, 799837. 6. Godbey, W. T.; Mikos, A. G. Journal of Controlled Release 2001, 72, 115-125. 7. Mintzer, M. A.; Simanek, E. E. Chemical Reviews 2009, 109, 259-302. 8. Verma, I. M.; Somia, N. Nature 1997, 389, 239-242. 9. Boussif, O.; Lezoualch, F.; Zanta, M. A.; Mergny, M. D.; Scherman, D.; Demeneix, B.; Behr, J. P. Proceedings of the National Acadamy of Science of the United States of America 1995, 92, 7297-7301. 10. Godbey, W. T.; Wu, K. K.; Mikos, A. G. Journal of Controlled Release 1999, 60, 149-160. 11. Parhamifar, L.; Larsen, A. K.; Hunter, A. C.; Andresen, T. L.; Moghimi, S. M Soft Matter 2010, 6, 4001-4009. 12. von Gersdorff, K.; Sanders, N. N.; Vandenbroucke, R.; de Smedt, S. C.; Wagner, E.; Ogris, M. Molecular Therapy 2006, 14, 745-753. 13. Deshpande, M. C.; Davies, M. C.; Garnett, M. C.; Williams, P. M.; Armitage, D.; Bailey, L.; Vamvakaki, M.; Armes, S. P.; Stolnik, S. Journal of Controlled Release 2004, 97, 143-156. 14. Dubruel, P.; Schacht, E. Macromolecular Bioscience 2006, 6, 789-810. 15. Funhoff, A. M.; van Nostrum, C. F.; Lok, M. C.; Kruijtzer, J. A. W.; Crommelin, D. J. A.; Hennink, W. E. Journal of Controlled Release 2005, 101, 233-246. 16. Lam, J. K. W.; Armes, S. P.; Stolnik, S. Journal of Drug Targeting 2011, 19, 5666. 17. van de Wetering, P.; Cherng, J. Y.; Talsma, H.; Crommelin, D. J. A.; Hennink, W. E. Journal of Controlled Release 1998, 53, 145-153. 18. van de Wetering, P.; Cherng, J.-Y.; Talsma, H.; Hennink, W. E. Journal of Controlled Release 1997, 49, 59-69. 19. van der Aa, M.; Huth, U. S.; Hafele, S. Y.; Schubert, R.; Oosting, R. S.; Mastrobattista, E.; Hennink, W. E.; Peschka-Suss, R.; Koning, G. A.; Crommelin, D. J. A. Pharmaceutical Research 2007, 24, 1590-1598. 20. Creutz, S.; Teyssie, P.; Jerome, R. Macromolecules 1997, 30, 6-9. 21. Rungsardthong, U.; Deshpande, M.; Bailey, L.; Vamvakaki, M.; Armes, S. P.; Garnett, M. C.; Stolnik, S. Journal of Controlled Release 2001, 73, 359-380. 22. Plamper, F. A.; Schmalz, A.; Penott-Chang, E.; Drechsler, M.; Jusufi, A.; Ballauff, M.; Müller, A. H. E. Macromolecules 2007, 40, 5689-5697. 23. Sahnoun, M.; Charreyre, M.-T.; Veron, L.; Delair, T.; D'Agosto, F. Journal of Polymer Science, Part A: Polymer Chemistry 2005, 43, 3551-3565. 24. Heath, W. H.; Senyurt, A. F.; Layman, J.; Long, T. E. Macromolecular Chemistry and Physics 2007, 208, 1243-1249. 25. Wang, Z. H.; Li, W. B.; Ma, J.; Tang, G. P.; Yang, W. T.; Xu, F. J. Macromolecules 2011, 44, 230-239. Chapter 3 – Influence of Polymer Architecture and Molecular Weight on TE and Cell Viability 88 26. Xu, F. J.; Ping, Y.; Ma, J.; Tang, G. P.; Yang, W. T.; Li, J.; Kang, E. T.; Neoh, K. G. Bioconjugate Chemistry 2009, 20, 1449-1458. 27. Xu, F. J.; Yang, W. T. Progress in Polymer Science 2011, 36, 1099-1131. 28. Schallon, A.; Jerome, V.; Walther, A.; Synatschke, C. V.; Müller, A. H. E.; Freitag, R. Reactive and Functional Polymers 2010, 70, 1-10. 29. Georgiou, T. K.; Phylactou, L. A.; Patrickios, C. S. Biomacromolecules 2006, 7, 3505-3512. 30. Georgiou, T. K.; Vamvakaki, M.; Patrickios, C. S.; Yamasaki, E. N.; Phylactou, L. A. Biomacromolecules 2004, 5, 2221-2229. 31. Georgiou, T. K.; Vamvakaki, M.; Phylactou, L. A.; Patrickios, C. S. Biomacromolecules 2005, 6, 2990-2997. 32. Pafiti, K. S.; Mastroyiannopoulos, N. P.; Phylactou, L. A.; Patrickios, C. S. Biomacromolecules 2011, 12, 1468-1479. 33. Dai, F.; Sun, P.; Liu, Y.; Liu, W. Biomaterials 2010, 31, 559-569. 34. Li, J. S.; Guo, Z. Z.; Xin, J. Y.; Zhao, G. L.; Xiao, H. N. Carbohydrate Polymers 2010, 79, 277-283. 35. Loh, X. J.; Zhang, Z. X.; Mya, K. Y.; Wu, Y. L.; He, C. B.; Li, J. Journal of Materials Chemistry 2010, 20, 10634-10642. 36. Nemoto, Y.; Borovkov, A.; Zhou, Y. M.; Takewa, Y.; Tatsumi, E.; Nakayama, Y. Bioconjugate Chemistry 2009, 20, 2293-2299. 37. Xu, F. J.; Zhang, Z. X.; Ping, Y.; Li, J.; Kang, E. T.; Neoh, K. G. Biomacromolecules 2009, 10, 285-293. 38. Layman, J. M.; Ramirez, S. M.; Green, M. D.; Long, T. E. Biomacromolecules 2009, 10, 1244-1252. 39. Cai, J. G.; Yue, Y. A.; Rui, D.; Zhang, Y. F.; Liu, S. Y.; Wu, C. Macromolecules 2011, 44, 2050-2057. 40. zg n, S. Akdemir, . asenpusch, G. Maucksch, C. Golas, M. M. Sander, B.; Stark, H.; Imker, R.; Lutz, J.-F.; Rudolph, C. Biomacromolecules 2009, 11, 39-50. 41. Venkataraman, S.; Ong, W. L.; Ong, Z. Y.; Loo, S. C. J.; Ee, P. L. R.; Yang, Y. Y. Biomaterials 2011, 32, 2369-2378. 42. Plamper, F. A.; Becker, H.; Lanzendorfer, M.; Patel, M.; Wittemann, A.; Ballauff, M.; Müller, A. H. E. Macromolecular Chemistry and Physics 2005, 206, 18131825. 43. Haddleton, D. M.; Edmonds, R.; Heming, A. M.; Kelly, E. J.; Kukulj, D. New Journal of Chemistry 1999, 23, 477-479. 44. Stenzel-Rosenbaum, M. H.; Davis, T. P.; Chen, V. K.; Fane, A. G. Macromolecules 2001, 34, 5433-5438. 45. Plamper, F. A.; Ruppel, M.; Schmalz, A.; Borisov, O.; Ballauff, M.; Müller, A. H. E. Macromolecules 2007, 40, 8361-8366. 46. Fischer, D.; Li, Y.; Ahlemeyer, B.; Krieglstein, J.; Kissel, T. Biomaterials 2003, 24, 1121-1131. 47. Rawlinson, L.-A. B.; O'Brien, P. J.; Brayden, D. J. Journal of Controlled Release 2010, 146, 84-92. 48. Newland, B.; Tai, H.; Zheng, Y.; Velasco, D.; Di Luca, A.; Howdle, S. M.; Alexander, C.; Wang, W.; Pandit, A. Chemical Communications 2010, 46, 46984700. 49. Prevette, L. E.; Mullen, D. G.; Banaszak Holl, M. M. Molecluar Pharmaceutics 2010, 7, 2370-2370. Chapter 3 – Influence of Polymer Architecture and Molecular Weight on TE and Cell Viability 89 50. Deng, R.; Yue, Y.; Jin, F.; Chen, Y. C.; Kung, H. F.; Lin, M. C. M.; Wu, C. Journal of Controlled Release 2009, 140, 40-46. 51. Luten, J.; van Nostrum, C. F.; de Smedt, S. C.; Hennink, W. E. Journal of Controlled Release 2008, 126, 97-110. 52. Wen, Y. T.; Pan, S. R.; Luo, X.; Zhang, X.; Zhang, W.; Feng, M. Bioconjugate Chemistry 2009, 20, 322-332. 53. You, Y.-Z.; Manickam, D. S.; Zhou, Q.-H.; Oupický, D. Journal of Controlled Release 2007, 122, 217-225. 54. Borisov, O.; Zhulina, E.; Leermakers, F.; Ballauff, M.; Müller, A. Self Organized Nanostructures of Amphiphilic Block Copolymers I; Springer: Berlin/Heidelberg, 2011; Vol. 241, pp 1-55. 55. Schallon, A.; Synatschke, C. V.; Pergushov, D. V.; Jerome, V.; Müller, A. H. E.; Freitag, R. Langmuir 2011, 27, 12042-12051. 56. van Gaal, E. V. B.; van Eijk, R.; Oosting, R. S.; Kok, R. J.; Hennink, W. E.; Crommelin, D. J. A.; Mastrobattista, E. Journal of Controlled Release 2011, 154, 218-232. 57. Elouahabi, A.; Ruysschaert, J.-M. Molecualr Therapeutics 2005, 11, 336-347. 58. Ainalem, M. L.; Campbell, R. A.; Khalid, S.; Gillams, R. J.; Rennie, A. R.; Nylander, T. Journal of Physical Chemistry B 2010, 114, 7229-7244. 59. Akinc, A.; Thomas, M.; Klibanov, A. M.; Langer, R. Journal of Genetic Medicine 2005, 7, 657-663. 60. Yang, S.; May, S. Journal of Chemical Physics 2008, 129, 185105. 61. Godbey, W. T.; Wu, K. K.; Mikos, A. G. Journal of Biomedical Materials Research, Part A 1999, 45, 268-275. Chapter 3 – Influence of Polymer Architecture and Molecular Weight on TE and Cell Viability 90 Supplementary Material Figure 3-S1: Mean fluorescence intensity (MFI) for the transfection efficiency data shown in Figure 3-3. Statistical significance was tested by ANOVA. Data represent mean values of five independent experiments  SD. Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 91 Chapter 4 Nano-Particulate Non-Viral Agent for the Effective Delivery of pDNA and siRNA to Differentiated Cells and Primary Human T Lymphocytes The results from this chapter have been published in Biomacromolecules as: “Nano-Particulate Non-Viral Agent for the Effective Delivery of pDNA and siRNA to Differentiated Cells and Primary Human T Lymphocytes” by Anja Schallon, Christopher V. Synatschke, Valérie Jérôme, Axel H. E. Müller and Ruth Freitag*. Reprinted with permission from Biomacromolecules 2012, 13, 3463–3474. Copyright 2012 American Chemical Society. Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 92 Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 93 Abstract Delivery of polynucleotides such as plasmid DNA (pDNA) and siRNA to non-dividing and primary cells by non-viral vectors presents a considerable challenge. In this contribution, we introduce a novel type of PDMAEMA-based star-shaped nanoparticles that (i) are efficient transfection agents in clinically relevant and difficult-to-transfect human cells (Jurkat T cells, primary T lymphocytes) and (ii) can efficiently deliver siRNA to human primary T lymphocytes resulting to more than 40 % silencing of the targeted gene. Transfection efficiencies achieved by the new vectors in serum-free medium are generally high and only slightly reduced in the presence of serum, while cytotoxicity and cell membrane disruptive potential at physiological pH are low. Therefore, these novel agents are expected to be promising carriers for non-viral gene transfer. Moreover, we propose a general design principle for the construction of polycationic nanoparticles capable of delivering nucleic acids to the above-mentioned cells. Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 94 Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 95 Introduction The interest in designing fully synthetic non-viral vectors for gene delivery has never waned, 1 although in the past the application range of such agents was limited. Compared to viral vectors, 2, 3 transfection efficiencies of non-viral vectors in general are low and none of the known agents can efficiently transfect non-dividing and/or differentiated cells. 4 For some innovative medical therapies, e.g. RNA interference (siRNA), delivery is still considered a substantial bottleneck. 5, 6 Currently used non-viral delivery agents are typically based on cationic polymers, polypeptides, or lipids, 7-10 with poly(ethylene imine), PEI, being a major player in the field of commercial products. The performance of non-viral transfection agents is typically discussed based on the various stages of the delivery process. The poor performance of non-viral vectors in transfecting suspension or non-dividing cells has been linked to problems in transfection complex attachment to the cellular membrane leading to inefficient endocytosis 11 and to a presumed inability to transgress the intact nuclear membrane, 12 respectively. In recent years, we have nevertheless seen a number of studies, which link size and structure of non-viral polycationic transfection agents to their performance. 12, 13 Increasing size of the polycation often correlates with improved transfection efficiency, but also with an increase in cytotoxicity; the latter being presumably due to a more pronounced disruptive interaction with the cellular membrane. 14, 15 Concomitantly evidence is building up that non-linear polymer structures are more efficient transfection agents than linear polymers of the same size. 2, 16-19 In this context, poly(2- (dimethylamino)ethyl methacrylate) (PDMAEMA), first described in the mid-90’s by Cherng and co-workers, 20 has become an important probe molecule in transfection studies, since PDMAEMA can be synthesized by a number of controlled polymerization methods (e.g., anionic polymerization and atom transfer radical polymerization (ATRP)). Thus, rather homogeneous polycations of different topologies (e.g. linear, branched) become available. In this context, suitable methods for the core-first synthesis of nanoparticular multi-armed DMAEMA stars using ATRP have recently become available, 21, 22 extending the basis for detailed investigation of structure-function relationships. In addition, such star shaped architectures can also be produced by using block copolymers which can self-assemble to micelles structures given suitable solvents. 23 The resulting Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 102 the cells, 2 mL of pre-warmed fresh growth medium were added, and the cells were further incubated for 20 h. For primary human T lymphocytes, transgene delivery was carried out 3 to 5 days after PBMC preparation and cultivation in QPBL medium. 5 x 105 cells were placed in 1.5 mL Opti-MEM medium per well in 6-well plates. 200 µL of the polyplex suspension prepared as described above were added and the plate placed for 4 h in the incubator. Afterwards, 0.5 mL of the medium was removed taking care not to disturb the cells, 2 mL of pre-warmed fresh QPBL medium were added, and the cells were further incubated for 48 h. Electroporation of T lymphocytes was performed as previously described. 42 Briefly, 20 µg DNA were pulsed with 5.0 x 106 cells in Opti-MEM at 250 V, 950 µF (BioRad Gene Pulse X Cell). Immediately after electroporation, cells were incubated for 10 min at 37°C followed by transfer into pre-warmed fresh QPBL medium (2 mL) in a 6-well plate. For analysis, adherent cells were harvested by trypsinization and suspension cells by centrifugation and resuspended in DPBS. For determination of the viability, dead cells were identified via counterstaining with propidium iodide (PI) or trypan blue. The relative expression of EGFP fluorescence of 1 x 104 cells was quantified via flow cytometry. Cells were initially evaluated by scatter properties (FSC/SSC) in order to select a region representing single non-apoptotic cells (elimination of dead cells, debris and cellular aggregations). This gated region (R0) was further analyzed for fluorescence (PI/EGFP). Dot plots with log of the red fluorescence intensity (PI) on the x-axis and log of the green fluorescence intensity (EGFP) on the y-axis were used to estimate the percentage of EGFP-expressing cells in the main non-apoptotic cell population (gate R0). Negative controls (N/P 0, non-transfected cells or cells transfected with an irrelevant pDNA) were used to set the position of quadrants separating GFP-positive living cells (upper left), GFP-positive dead cells (upper right), GFP-negative living cells (lower left) and GFPnegative dead cells (lower right). These quadrants were applied for the analysis of transfected cells and percentage cell number / total cell number in the gated region were calculated for each quadrant. Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 103 Delivery of siRNA siRNAs were: hCD4-siRNA (sense) 5’- GCAAUUGCUAGUGUUCGGAUUGACUGC - 3', (anti-sense), 5’ – GCAGUCAAUCCGAACACUAGCAAUUGC - 3’ EGFP-siRNA 50 (sense) 5'’ – AAGCUGACCCUGAAGUUCAUCUGCACC - 3', (antisense), 5’ – GGUGCAGAUGAACUUCAGGGUCAGCUU - 3’ (all Eurofins MWG peron). The siRNAs were obtained as duplex and solubilized into 1x siMAX universal buffer (6 mM HEPES, 20 mM KCl, 200 µM MgCl2, pH 7.3; Eurofins MWG Operon). EGFP-siRNA and hCD4-siRNA were used as negative control in the experiments involving CD4 knockdown in T lymphocytes and EGFP knockdown in CHO-EGFP-VEGFA cells, respectively. For siRNA delivery into EGFP expressing CHO cells (CHO-EGFP-VEGFA cells 48), cells were seeded at a density of 0.25 x 105 cells per well in 24-well plates 20 h prior to delivery. One hour prior to delivery, cells were rinsed with DPBS and supplemented with 0.2 mL Opti-MEM medium. siRNA-polyplexes were prepared by diluting siRNAs stock solution (10 µM) in a total volume of 50 µL Opti-MEM to the indicated final concentration and adding sufficient amounts of the polycation stock solution to reach the desired N/P ratio. The mixture was immediately mixed by vortexing at full speed followed by incubation at room temperature for 15 min. The polyplex suspension (50 µL) was added to the cells, the plates were centrifuged for 5 min at 200 g, and placed for 4 h in the incubator. Afterwards, the supernatant was removed by aspiration, 1 mL of fresh growth medium was added, and the cells were further incubated for 20 h (Si-PDMAEMA) and 30 h (Si-PDMAEMA and Mic-PDMAEMA). The relative expression of EGFP fluorescence of 1 x 104 cells was quantified. For determination of the viability, dead cells were identified via counterstaining with propidium iodide. Cells were initially evaluated by scatter properties (FSC/SSC) in order to select a region representing single non-apoptotic cells (elimination of dead cells, debris and cellular aggregations). This gated region (R0) was further analyzed for fluorescence (PI/EGFP). To assess the efficiency of the siRNA to knockdown the EGFP expression, the median fluorescence intensity (FI) values were compared. For siRNA delivery to T lymphocytes, cells were washed twice with DPBS and plated in 250 µL Opti-MEM in 24-well plates at 5 x 105 cells per well for 1 h prior to transfection. Polyplexes were prepared and added to the wells as described above. The plates were centrifuged for 5 min at 200 g and placed for 4 h in the incubator. Afterwards, 700 µL of pre-warmed fresh QPBL medium were added per well and the cells were further Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 104 incubated for at least 24 h without removing the complexes or replacing the medium until subsequent analysis by flow cytometry. Analysis of CD4 Expression For analysis of CD4 expression, cells were harvested by centrifugation (5 min, 200 g, 4 °C) and resuspended in DPBS. T lymphocytes expressing CD4 receptors were identified by immunofluorescence via staining the cells with FITC-conjugated anti-CD4 or FITC-conjugated isotype control antibodies (according to manufacturer’s instructions) on the day of siRNA delivery and at least 24 h after delivery. Dead cells were identified via counterstaining with propidium iodide. The relative expression of CD4 was quantified via flow cytometry. Cells were initially evaluated by scatter properties (FSC/SSC) in order to select a region representing single non-apoptotic cells and to eliminate debris which always compose a significant fraction in an activated primary lymphocyte cell culture (gate ”lympho”) and by (SSC/PI) in order to select the living cells (PI-negative population) (gate ”living”). The expression of the CD4 protein was assessed in histogram plots (green fluorescence intensity on the x-axis and cell number on the y-axis) representing the intensity of the CD4-FITC fluorescence (CD4low: fluorescence intensity between 70 and 170; CD4high: fluorescence intensity > 170) in the living T lymphocytes (defined as a sub-population of gate “lympho” and gate “living”). Cytotoxicity / Vitality Assay (MTT) The toxicity of the polycations was tested (concentration range 0.001 mg/mL to 5.0 mg/mL, 8 replicate experiments each) according to the ISO 10993-5 protocol by MTT assay using L929 murine fibroblasts, cultured in MEM supplemented with 10 % FCS, as test cells. The cells were seeded at a density of 1 x 105 cells per well 24 h prior to the experiment in 96-well plates. The concentration of the MTT stock solution was 1 mg/mL. As 100 % viability control, untreated cells were used. The absorbance was measured using a plate reader (Genios Pro, Tecan, Germany); wavelength 580 nm. Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 105 For T lymphocytes, the MTT assay was performed as follows. 5 x 105 human T lymphocytes in 500 µL Opti-MEM medium were transferred into 1.5 mL Eppendorf tubes. The polymer was added and the cells were incubated for 4 h at 37°C in the incubator. Then, the Opti-MEM medium was replaced by QPBL medium. All medium exchange and washing steps were done by centrifugation (5 min, 200 g, 4°C). The cells were further incubated for 20 h. Then cells were rinsed with DPBS and further incubated in 200 µL MTT solution (1.0 mg/mL in RPMI 1640 without phenol red) for 2 h. The tubes were centrifuged for 5 min at 600 g, the MTT solution was discarded and 200 µL isopropanol were added to the cell pellet. The tubes were mixed at 150 rpm for 5 min to dissolve the formazan crystals produced in the reaction. Absorbance was measured at 580 nm in the microplate reader with untreated cells serving as controls. For data evaluation, Origin 6.1 (OriginLab Corporation, Northampton, USA) software was used, the x-scale was plotted logarithmically and a nonlinear fit was used to obtain the LD50 values. Hemolysis Test The membrane damaging properties of the polymers was quantified by analyzing the release of hemoglobin from human erythrocytes, according to Parnham and Wetzig. 51 The erythrocytes-containing blood fraction obtained after Ficoll gradient separation was centrifuged at 700 g for 10 min. The obtained pellet was washed three times with cold DPBS pH 7.4 by centrifugation at 700 g for 10 min and re-suspension in the same buffer. Polymer solutions were prepared in DPBS buffer and 100 µL were added to the erythrocytes (100 µL) to give final concentrations in the range of 0.001 to 5.0 mg/mL and incubated for 60 min under constant shaking at 37°C. After centrifugation (700 g, 10 min), the supernatant was analyzed for released hemoglobin at 580 nm. The absorbance was measured using a plate reader (Genios Pro, Tecan, Germany). For comparison, collected erythrocytes were washed with DPBS and either lysed with 0.2 % Triton X-100 yielding the 100 % lysis control value (A100) or resuspended in DPBS as reference (A0). The analysis was repeated with blood from at least six independent donors. The hemolytic activity of the polycations was calculated as follow: Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 106 % hemolysis = 100*(A – A0)/(A100 – A0) with A: absorbance of the sample, A100: absorbance at 100 % hemolysis, A0: absorbance at 0 % hemolysis. Statistical Analysis Group data are reported as mean  s.e.m. For transfection results, the Student’s t-test was used to determine whether data groups differed significantly from each other. Statistical significance was defined as having P < 0.05. Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 107 Results and Discussion Polymer Synthesis and Characterization As basis for the investigation, a well-defined 20-armed star (Si-PDMAEMA) was synthesized via ATRP of DMAEMA, from a silsesquioxane initiator core (Scheme 4-1a) based on a procedure previously published by one of our groups. 21 Progress of the polymerization was monitored by following the monomer consumption via 1H-NMR as evidenced by the decrease in the integral of the two vinyl protons at 6.21 ppm and 5.61 ppm. Trioxane was added as an internal standard to the mixture, because it gives a characteristic signal at a chemical shift of 5.14 ppm and does not participate in the reaction. The polymerization was quenched at 42 % monomers conversion, which corresponds to an average degree of polymerization (DP, or number of DMAEMA units per molecule) of 4,570, or a number average molecular weight Mn of 730 kDa assuming that the monomers were homogenously distributed among all growing particles. Due to steric hindrance, not all putative initiation sites of the silsesquioxane initiator can be expected to start a polymer chain21 and the number of arms per star is therefore below the theoretically possible 58. In order to determine the average number and length of the arms, these were cleaved off and their molecular weight distribution was determined by gel permeation chromatography (GPC). The number average Mn of the arms was 23,500 Da, and the weight average Mw was 33,500 Da, which corresponds to a polydispersity index (PDI) of 1.42 and a DP of 235 monomers per arm. From this, an average number of 19.5 arms was calculated for the produced Si-PDMAEMA. The number of arms is in good agreement with the values of 19 to 24 arms that have been determined for other starshaped polymers prepared using the same initiator. 21 A z-average hydrodynamic radius, <Rh>z, app., of 37.2 ± 3.5 nm was determined for the Si-PDMAEMA by dynamic light scattering (DLS). It is known that branched poly(ethylene imine) (b-PEI) shows a high buffering capacity, because of the large number of nitrogen atoms in its chemical structure. The so-called “proton sponge effect” is a result of this buffering capacity and it is said to result in an effective escape of PEI-based-polyplexes from the endosome, explaining the high transfection efficiency of this polymer. 52, 53 We conducted potentiometric titration experiments for the Si-PDMAEMA star-shaped polymer as well as for b-PEI. From the titration experiments apparent pKa values of pKa, Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 108 app.(b-PEI) = 7.05 and pKa, app.(Si-PDMAEMA) = 6.73 were determined. The titration was performed at a mass concentration of 0.5 mg/mL in MilliQ water (corresponding to molar nitrogen concentration of 11.6 mM and 3.13 mM for b-PEI and Si-PDMAEMA, respectively) with 0.1M HCl solution. b-PEI had a larger buffering region ranging from pH 9.6 – 4.2 than Si-PDMAEMA, which showed a buffering effect from pH 8.9 – 4.6. The overall amount of buffered HCl was greater for b-PEI (2.61 mL) than for SiPDMAEMA (1.25 mL), which is explained by the higher molar nitrogen concentration of b-PEI. In the buffering region relevant for the endosomal environment (pH 7.4 – 5) b-PEI shows a slightly better buffering capacity by mass concentration than Si-PDMAEMA, with 1.27 mL of buffered 0.1M HCl for b-PEI and 0.94 mL for Si-PDMAEMA. Elemental analysis of the polycations indicated inter alia the presence of 15.55 ppm Cu, i.e. not all of the copper catalyst had been removed from the final product during dialysis, probably due to strong binding by the PDMAEMA arms. However, given the low cytotoxicity of Si-PDMAEMA, see below, significant release of copper ions during application is unlikely. To demonstrate the general potential of multi-arm star-shaped polycationic nanoparticles as polynucleotide delivery vehicles, a second structure of similar design was produced via self-assembly (formation of star-like micelles) of an amphiphilic diblock copolymer. For this purpose, polybutadiene-block-poly(2-(dimethylamino)ethyl methacrylate) (PB290-bPDMAEMA240) was synthesized via sequential living anionic polymerization as previously published. 44 The diblock copolymer had a number-average molecular weight, Mn, of 53,500 Da and a very narrow PDI of 1.07. A micellar, star-shaped gene delivery agent, Mic-PDMAEMA, was obtained via the self-assembly of PB290-b-PDMAEMA240 into micelles upon a change in solvent from THF, which solubilizes the entire diblock copolymer, to PBS, which only solubilizes the PDMAEMA block (Scheme 4-1b). An <Rh>z, app. of 27 ± 3 nm was determined for the Mic-PDMAEMA by DLS. The hydrodynamic radii of the two delivery agents are thus in the same range. Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 109 Scheme 4-1. Synthetic procedure for Si-PDMAEMA preparation from multi-functional initiator via ATRP (a) and self-assembly of amphiphilic diblock copolymer PB290-b-PDMAEMA240 to Mic-PDMAEMA starshaped micelles (b). The first step in polynucleotides delivery is the formation of polyplexes between the polycationic delivery agent and the negatively charged polynucleotide. Surface charge of polyplexes is an important factor especially for unspecific uptake into cells by adsorptive endocytosis mediated by proteoglycans. 54 We determined the zeta potential of the SiPDMAEMA and Mic-PDMAEMA polyplexes at various polymer N / DNA P-ratios (Table 4-S1). The zeta potential of the polyplexes increased with increasing N/P ratio and at N/P ratios equal to or larger than 5 positive values up to +10.5 mV were obtained. pDNA Delivery by Si-PDMAEMA The Si-PDMAEMA was tested by standardized transfection procedures, using EGFP as reporter gene, in a panel of model cell lines including adherent (CHO-K1, HEK-293, Wi38, A549) and suspension (Jurkat) cells. In addition, C2C12 cells were used as model for non-dividing and differentiated cells and human T lymphocytes were used as example for primary cells. b-PEI (25 kDa), i.e. the standard non-viral transfection agent used in our Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 110 group, served as control. While depending on N/P-ratio and cell type, transfection efficiencies were consistently higher for Si-PDMAEMA than for b-PEI at slightly reduced cytotoxicities (Table 4-1). This included the Jurkat cells, where the best transfection efficiency was 46.1 % ± 3.7 for Si-PDMAEMA compared to 6.2 % ± 2.6 for b-PEI (n  5). In order to exclude a false positive measurement, we also performed transfection with a blank pDNA (“control plasmid”). The data presented in Table 4-1 and Figure 4-S2 clearly demonstrate that the measured increase in fluorescence in the SiPDMAEMA transfected cells is due to EGFP expression and not to an increased autofluorescence of the cells due to polymer accumulation. Jurkat cells are suspension cells, which are difficult to transfect with non-viral vectors because they sparsely internalize cationic complexes. 11 The low values obtained for b-PEI are in accordance with the data from the literature 55, 56 and were thus expected, while the much better value obtained for Si-PDMAEMA were a first indication for a fundamentally different performance of this transfection agent. Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 111 Table 4-1. Transfection efficiency and cell viability after transfection with polyplexes based on b-PEI and Si-PDMAEMA. Cell line Transfection efficiency (%) Viability (%) PEI Si-PDMAEMA PEI Si-PDMAEMA N/P 5 36.6 ± 12.0 59.8 ± 18.9 95.3 ± 3.7 93.7 ± 4.4 CHO-K1 N/P 10 49.1 ± 17.6 73.5 ± 8.0 94.4± 3.9 92.9 ± 3.0 N/P 20 40.9 ± 20.6 70.5 ± 6.3 87.3 ± 8.7 91.6 ± 3.2 N/P 5 n. d. 0.2 ± 0.2 n. d. 98.5 ± 0.7 Control plasmid a N/P 10 n. d. 1.6 ± 1.1 n. d. 97.4 ± 2.8 N/P 20 n. d. 0.8 ± 0.1 n. d. 95.0 ± 1.9 N/P 5 23.5 ± 9.7 33.9 ± 15.0 83.9 ± 10.1 88.0 ± 6.8 HEK-293 N/P 10 36.1 ± 8.2 50.3 ± 15.1 82.7 ± 14.3 90.8 ± 5.9 N/P 20 32.8 ± 7.7 55.2 ± 15.2 79.1 ± 17.7 83.4 ± 9.6 N/P 5 20.3 ± 3.1 9.8 ± 3.5 71.0 ± 6.5 76.8 ± 7.3 Wi-38 N/P 10 19.2 ± 3.9 26.9 ± 5.5 66.6 ± 5.7 73.3 ± 6.4 N/P 20 2.1 ± 1.7 20.9 ± 6.4 39.4 ± 16.6 53.8 ± 7.9 N/P 5 21.7 ± 21.2 45.8 ± 28.0 93.2 ± 4.7 91.4 ± 4.0 A549 N/P 10 36.1 ± 14.2 48.6 ± 22.0 74.8 ± 7.9 86.0 ± 9.4 N/P 20 20.0 ± 7.7 43.0 ± 9.8 48.7 ± 16.1 70.1 ± 21.3 N/P 3 0.6 ± 0.4 24.1 ± 0.0 81. 4 ± 2.9 76.6 ± 4.8 Jurkat b, c N/P 5 2.0 ± 1.4 33.3 ± 2.7 84.3 ± 4.2 54.4 ± 0.6 N/P 10 4.4 ± 3.1 46.1 ± 3.7 69.0 ± 19.3 44.2 ± 15.3 N/P 20 6.2 ± 2.6 28.6 ± 16.3 39.9 ± 22.0 53.6 ± 17.7 N/P 5 n. d. 0.4 ± 0.1 n. d. 71.8 ± 3.5 Control plasmid a N/P 10 n. d. 0.3 ± 0.0 n. d. 60.6 ± 7.7 N/P 20 n. d. 0.4 ± 0.2 n. d. 53.1 ± 2.4 The cells were transfected with pEGFP-N1 (EGFP expression plasmid) and in the case of CHO-K1 and Jurkat cells additional transfections were performed with a control plasmid (pIVEX2.3-UK, non-EGFP expression plasmid) to exclude that accumulation of polymer merely induces an increase of the cell autofluorescence. DNA concentration: 15 µg/mL. The EGFP expression was measured 24 h after transfection by flow cytometry and analyzed as described in the materials and methods section. The transfection efficiency data represent the percentage of living cells expressing EGFP in the non-apoptotic cell population defined by scatter properties as determined by flow cytometry analysis. For determination of the viability, dead cells were identified via counterstaining with propidium iodide. a, b: Representative flow cytometry dot plots are provided in the supplementary information (Figure 4-S1 and Figure 4-S2). Data represent mean ± s.e.m., n ≥ 5, safe for c: n ≥ 3. n. d.: not determined. Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 118 Table 4-3. Relative release of hemoglobin in percent (mean  s.e.m.) by human red blood cells after 60 min incubation with different concentrations of the cationic polymers at 37°C (n = 6). Polymer 0.001 mg/mL 0.005 mg/mL 0.01 mg/mL 0.05 mg/mL 0.1 mg/mL 0.5 mg/mL 1.0 mg/mL 5.0 mg/mL b-PEI 0.1 ± 0.8 5.0 ± 1.9 5.4 ± 3.1 16.2 ± 5.0 26.2 ± 9.4 63.3 ± 12.9 75.2 ± 12.0 101.5 ± 8.4 SiPDMAEMA 1.4 ± 2.1 0.9 ± 1.9 0.9 ± 0.2 7.1 ± 1.9 13.2 ± 5.8 22.0 ± 1.5 23.4 ± 5.0 34.1 ± 7.2 siRNA-Mediated Knockdown of Gene Expression in Recombinant CHO Cells and T Lymphocytes RNA interference (RNAi) represents a promising technology for gene-specific knockdown, e.g. in the context of developing new therapeutic approaches. 72 However, a critical factor still limiting the use of siRNA as therapeutic is delivering siRNA to its intracellular target site as recently reviewed. 32, 73 Recombinant CHO cells constitutively expressing EGFP 48 and human primary T lymphocytes were used to evaluate the potential of Si-PDMAEMA to deliver siRNA into the cells and mediate gene silencing. Prior to all silencing experiments, preliminary tests were performed in order to estimate the most suitable siRNA concentration and N/P ratio for gene silencing. Therefore, referring to published contributions, 36, 74 25 and 50 nM siRNA were tested in parallel to various charge ratios (N/P 3 to 20). Optimized conditions were found to be 25 or 50 nM siRNA and a N/P ratio of 10 gave the best results. A significant silencing effect was not detected for incubation time shorter than 30 h, probably due to the high stability of the targeted proteins (EGFP t1/2 ≥ 24 h 75; CD4 t1/2 = 20 h 76) (data not shown). The results presented below were obtained under optimized conditions and reflect the maximal knockdown achieved so far. In the recombinant CHO cells, knockdown after incubation with the complexes containing 50 nM EGFP-siRNA was evaluated by flow cytometry analysis of the EGFP fluorescence in comparison to cells where delivery of siRNA had been attempted using b-PEI (Table 4-4). b-PEI/siRNA polyplexes achieved at most a 16 % reduction of the EGFP expression. This low knockdown efficiency of PEI is in agreement with data published elsewhere. 77, 78 Si-PDMAEMA triggered a significantly higher knockdown (54.6 %). For both polycations, only a minimal effect on cell viability, which remained within 85 % of the non-transfected cells, was observed. Similar levels were observed in isolated cases for PDMAEMA/siRNA polyplexes in lung cancer cells. 79, 80 Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 119 Table 4-4. Gene silencing in recombinant CHO cells constitutively expressing EGFP. N/P ratio 0 5 10 15 naked siRNA 5.3 - - - b-PEI - 0.0 16.0 n.d. Si-PDMAEMA - 1.7 54.6 51.0 EGFP expression was determined by flow cytometry 30h after siRNA delivery using either b-PEI or SiPDMAEMA. siRNA concentration: 50 nM at N/P 10. Data represent percentage of knockdown of EGFP expression compared to control cells. Viabilities were estimated by propidium iodide staining prior to flow cytometry analysis. n.d.: not determined T lymphocytes are known to be particularly resistant to siRNA uptake enforced by conventional non-viral delivery methods excepting Nucleofection® 42 and antibody fragment-peptide fusion protein-based delivery. 81 Based on the promising results obtained in the recombinant CHO cell line, we subsequently investigated the potential of Si-PDMAEMA for knockdown of CD4-expression in human T lymphocytes. In preliminary experiments, screening for optimized delivery conditions, we were able to show that Si-PDMAEMA-based delivery of hCD4-siRNA led to significantly higher silencing effect than the one obtained after b-PEI-based delivery (data not shown). In order to confirm this observation, siRNA delivery/knockdown was repeated with T lymphocytes isolated from another donor. In addition, linear PEI (l-PEI, 25 kDa) was used instead of b-PEI. Cells mock-delivered with EGFP-siRNA served as control. Knockdown after 30 h incubation with the complexes was evaluated by flow cytometry analysis. Delivery of the siRNA with l-PEI had no effect on the level of CD4 expression. Si-PDMAEMA, on the other hand, achieved a 2.3-fold decrease of CD4high and a 2.3-fold increase of the CD4low populations, respectively (Table 4-5). Viability was again within 85 % of the non-transfected cells in all cases. This is, to our knowledge, the first time that a PDMAEMA-based polycation were used successfully to deliver siRNA into human primary T lymphocytes and leading to the specific knock-down of the targeted gene. Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 120 Table 4-5. Analysis of CD4 expression 30 h after human T lymphocytes were either mock delivered (EGFP-siRNA) or delivered with hCD4-siRNA (25 nM each). hCD4-siRNA EGFP-siRNA CD4high CD4low CD4high CD4low l-PEI 77.5 22.5 79.9 20.1 Si-PDMAEMA 29.6 70.3 69 31 The siRNA was delivered complexed with l-PEI or Si-PDMAEMA (N/P ratio 10). Data represent the percentage of CD4high and CD4low cells within the viable CD4+ population. The viability was estimated by propidium iodide staining prior to flow cytometry analysis. For comparison: cells submitted to the same medium changes as the transfected ones, but not receiving any siRNA displayed 66.7 % CD4high and 33.5 % CD4low. Representative flow cytometry dot plots and histograms are provided in the supplementary information (Figure 4-S4). Verification of the General Design Principle for Improved Non-Viral Transfection Agents In order to verify our initial hypothesis that many arms emanating from a common center is a general design principle for the construction of efficient non-viral polynucleotide delivery vehicles, star-like polymer micelles (Mic-PDMAEMA) were produced and their efficiency as potential transfection reagent was explored under standard conditions in Jurkat cells. The micelle core-based structure was as efficient as Si-PDMAEMA as shown by the achieved transfection efficiency ranging from 11 to 35 % transfected cells depending on the N/P ratio, although transfection at N/P ratio of 20 led to high cytotoxicity (Table 4-6). Table 4-6. Summary of the Jurkat cells transfection with star-like PDMAEMA-based micelles (MicPDMAEMA) in serum-free medium. N/P ratio Transfection efficiency (%) Viability (%) N/P 3 11.5 ± 1.4 88.1 ± 3.2 N/P 5 31.0 ± 3.0 76.2 ± 13.3 N/P 10 21.2 ± 0.0 79.4 ± 0.4 N/P 20 35.0 ± 17.4 32.4 ± 20.9 The cells were transfected with pEGFP-N1 (EGFP expression plasmid). DNA concentration: 15 µg/mL. Polymer concentrations were adjusted to the indicated N/P ratios. The EGFP expression was measured 24 h after transfection by flow cytometry. The transfection efficiency data represent the percentage of living cells expressing EGFP in the non-apoptotic cell population defined by scatter properties as determined by flow cytometry analysis. The viability was estimated by propidium iodide staining prior to flow cytometry analysis. Data represent mean ± s.e.m. (n ≥ 3) representative flow cytometry dot plots are provided in the supplementary information (Figure 4-S5). Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 121 Recently, we have in addition published data showing that magnetic core-shell nanoparticles displaying similar architecture, can efficiently deliver pDNA to CHO-K1 cells yielding more than 50 % transfected cells. 82 Taken together, these results sustain our initial hypothesis that star-shaped architectures with a large number of arms irradiating from the core are efficient non-viral vectors. Additional fine tuning of the composition of the co-polymers to enhance endosomal release could further increase the transfection efficiency of the star-like micelles as recently demonstrated in one monocyte cell line by Manganiello and co-workers. 83 The ability of Mic-PDMAEMA to deliver siRNA was tested using the EGFP knockdown in the recombinant CHO cells stably expressing this protein as test system. Here also, preliminary screening of the optimum for siRNA concentration and N/P ratio was performed (data not shown) and the results presented below reflect data obtained under these optimized conditions. Knockdown after 30 h incubation with the polyplexes (N/P ratio of 20) containing either EGFP-siRNA or hCD4-siRNA (used as control) was evaluated by flow cytometry (Figure 4-4). Figure 4-4. siRNA mediated silencing of the EGFP-expression in recombinant CHO cells constitutively expressing this protein determined after delivery with b-PEI or Mic-PDMAEMA; siRNA concentration: 25 nM, N/P ratio: 20. CHO-EGFP-VEGF cells were either mock delivered (hCD4siRNA, white bars) or delivered with EGFP-siRNA (black bars). 30 h post-transfection, the cells were counterstained with propidium iodide to identify dead cells and analyzed for EGFP expression by flow cytometry. Data represent percentage of knockdown of EGFP expression in viable cells compared to non-treated control cells (mean ± s.e.m., n ≥ 3). Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 122 Whereas b-PEI/siRNA polyplexes achieved at most a 20 % reduction in the fluorescence geometric mean, Mic-PDMAEMA delivery resulted in a 43.6 % knockdown and thus performed almost in the same range as Si-PDMAEMA (Table 4-4). As before, cell viability was only minimally affected (> 85 % in all cases). Conclusions The large polycationic nanoparticles introduced here display, to our knowledge, a never before observed capability to deliver nuclei acids to human primary T lymphocytes and to non-dividing cells and thus, have considerable advantages over conventional polycations for gene delivery. In particular, the proposed new transfection reagent synthesized from an inorganic core (Si-PDMAEMA) displays high potentiality for transfection of primary, non-dividing and differentiated cells as well as a broad compatibility with established cell lines. An additional construct, produced along the same design principle and containing a polybutadiene core (Mic-PDMAEMA), also showed more efficient pDNA–deliveries than PEI to CHO and Jurkat cells. Furthermore, we established a first proof of principle that Si-PDMAEMA and Mic-PDMAEMA can be used for gene silencing using small interfering RNA (siRNA) in CHO cells and human primary T lymphocytes. In this context, polymers based on diblock copolymers are of particular interest because their production is easy and further modification, e.g., including a targeting sequence, would be possible. As far as we know, star-like architectures reported before were generally less efficient than conventional transfection reagents for gene delivery in particular when “hard-to-transfect” cells were concerned. This work establishes that the design principle of many arms emanating from a common center results in efficient polynucleotide delivery vehicles independent of the core material and therefore offers advanced possibilities for the development improved gene vectors in particular for primary cells. Moreover, due to its low disruptive potential for cell membranes (hemolytic activity) at physiological pH and its ability to transfect cells in the presence of serum, SiPDMAEMA might become an attractive system for further in vivo evaluations. Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 123 Acknowledgments The authors would like to thank Prof. Förster (Physical Chemistry I and Bayreuth Center for Colloids and Interfaces, University of Bayreuth) for access to the Zetasizer. C. V. Synatschke acknowledges funding through a BayEFG scholarship and support from the Elite Network of Bavaria. PB290-b-PDMAEMA240 block polymers were kindly supplied by F. Schacher (Macromolecular Chemistry II, University of Bayreuth, Germany). Alexandra Rinkenauer is acknowledged for performing some of the siRNA delivery experiments. Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 124 References 1. Grigsby, C. L.; Leong, K. W. Journal of the Royal Society Interface 2010, 7 Suppl 1, 67-82. 2. Mintzer, M. A.; Simanek, E. E. Chemical Reviews 2009, 109, (2), 259-302. 3. Verma, I. M. Molecular Medicine 1994, 1, (1), 2-3. 4. van Gaal, E. V.; van Eijk, R.; Oosting, R. S.; Kok, R. J.; Hennink, W. E.; Crommelin, D. J.; Mastrobattista, E. Journal of Controlled Release 2011, 154, (3), 218-232. 5. Shegokar, R.; Al Shaal, L.; Mishra, P. R. Pharmazie 2011, 66, (5), 313-318. 6. Sliva, K.; Schnierle, B. S. Virology Journal 2010, 7, 248. 7. Donkuru, M.; Badea, I.; Wettig, S.; Verrall, R.; Elsabahy, M.; Foldvari, M. Nanomedicine 2010, 5, (7), 1103-1127. 8. Elsabahy, M.; Nazarali, A.; Foldvari, M. Current Drug Delivery 2011, 8, (3), 235244. 9. Laufer, S. D.; Restle, T. Current Pharmaceutical Design 2008, 14, (34), 36373655. 10. Midoux, P.; Pichon, C.; Yaouanc, J. J.; Jaffres, P. A. British Journal of Pharmacology 2009, 157, (2), 166-178. 11. Labat-Moleur, F.; Steffan, A. M.; Brisson, C.; Perron, H.; Feugeas, O.; Furstenberger, P.; Oberling, F.; Brambilla, E.; Behr, J. P. Gene Therapy 1996, 3, (11), 1010-1017. 12. Pathak, A.; Patnaik, S.; Gupta, K. C. Biotechnology Journal 2009, 4, (11), 15591572. 13. Gao, X.; Kim, K. S.; Liu, D. AAPS Journal 2007, 9, (1), E92-104. 14. Fischer, D.; Li, Y.; Ahlemeyer, B.; Krieglstein, J.; Kissel, T. Biomaterials 2003, 24, (7), 1121-1131. 15. Hunter, A. C. Adv. Drug Delivery Reviews 2006, 58, (14), 1523-1531. 16. Al-Dosari, M. S.; Gao, X. AAPS Journal 2009, 11, (4), 671-681. 17. Nakayama, Y. Accounts of Chemical Research 2012, 45, (7), 994-1004. 18. Schallon, A.; Jérôme, V.; Walther, A.; Synatschke, C. V.; Müller, A. H. E.; Freitag, R. Reactive and Functional Polymers 2010, 70, 1-10. 19. Synatschke, C. V.; Schallon, A.; Jérôme, V.; Freitag, R.; Müller, A. H. E. Biomacromolecules 2011, 12, 4247 - 4255. 20. Cherng, J.-Y.; van de Wetering, P.; Talsma, H.; Crommelin, D. J. A.; Hennink, W. E. Pharmaceutical Research 1996, 13, (7), 1038-1042. 21. Plamper, F. A.; Schmalz, A.; Penott-Chang, E.; Drechsler, M.; Jusufi, A.; Ballauff, M.; Müller, A. H. E. Macromolecules 2007, 40, (16), 5689-5697. 22. Wang, J. S.; Greszta, D.; Matyjaszewski, K. Abstract Papers, Joint Conference – Chemical Institute of Canada and American Chemical Society 1995, 210, 227PMSE. 23. Harada, A.; Kataoka, K. Progress in Polymer Science 2006, 31, (11), 949-982. 24. Miyata, K.; Christie, R. J.; Kataoka, K. Reactive and Functional Polymers 2011, 71, (3), 227-234. 25. Newland, B.; Tai, H.; Zheng, Y.; Velasco, D.; Di Luca, A.; Howdle, S. M.; Alexander, C.; Wang, W.; Pandit, A. Chemical Communications 2010, 46, (26), 4698-4700. 26. Zhong, Z.; Song, Y.; Engbersen, J. F.; Lok, M. C.; Hennink, W. E.; Feijen, J. Journal of Controlled Release 2005, 109, (1-3), 317-329. Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 125 27. Layman, J. M.; Ramirez, S. M.; Green, M. D.; Long, T. E. Biomacromolecules 2009, 10, (5), 1244-1252. 28. van de Wetering, P.; Cherng, J. Y.; Talsma, H.; Crommelin, D. J. A.; Hennink, W. E. Journal of Controlled Release 1998, 53, (1-3), 145-153. 29. Georgiou, T. K.; Vamvakaki, M.; Patrickios, C. S.; Yamasaki, E. N.; Phylactou, L. A. Biomacromolecules 2004, 5, (6), 2221-2229. 30 Xu, F. J.; Zhang, Z. X.; Ping, Y.; Li, J.; Kang, E. T.; Neoh, K. G. Biomacromolecules 2009, 10, (2), 285-293. 31. Aliabadi, . M. Landry, B. Sun, C. Tang, T. Uludağ, . Biomaterials 2012, 33, (8), 2546-2569. 32. Scholz, C.; Wagner, E. Journal of Controlled Release 2012, 161, (2), 554-565. 33. Zhang, S.; Zhao, Y.; Zhi, D.; Zhang, S. Bioorganic Chemistry 2012, 40, (0), 1018. 34. Benoit, D. S.; Henry, S. M.; Shubin, A. D.; Hoffman, A. S.; Stayton, P. S. Molecular Pharmaceutics 2010, 7, (2), 442-455. 35. Convertine, A. J.; Benoit, D. S. W.; Duvall, C. L.; Hoffman, A. S.; Stayton, P. S. Journal of Controlled Release 2009, 133, (3), 221-229. 36. Convertine, A. J.; Diab, C.; Prieve, M.; Paschal, A.; Hoffman, A. S.; Johnson, P. H.; Stayton, P. S. Biomacromolecules 2010, 11, (11), 2904-2911. 37. Kong, W.-H.; Sung, D.-K.; Shim, Y.-H.; Bae, K. H.; Dubois, P.; Park, T. G.; Kim, J.-H.; Seo, S.-W. Journal of Controlled Release 2009, 138, (2), 141-147. 38. Pafiti, K. S.; Patrickios, C. S.; Georgiou, T. K.; Yamasaki, E. N.; Mastroyiannopoulos, N. P.; Phylactou, L. A. European Polymer Journal 2012, 48, (8), 1422-1430. 39. Zhu, C.; Jung, S.; Luo, S.; Meng, F.; Zhu, X.; Park, T. G.; Zhong, Z. Biomaterials 2010, 31, (8), 2408-2416. 40. Gary, D. J.; Lee, H.; Sharma, R.; Lee, J. S.; Kim, Y.; Cui, Z. Y.; Jia, D.; Bowman, V. D.; Chipman, P. R.; Wan, L.; Zou, Y.; Mao, G.; Park, K.; Herbert, B. S.; Konieczny, S. F.; Won, Y. Y. ACS Nano 2011, 5, (5), 3493-4505. 41. Lin, D.; Huang, Y.; Jiang, Q.; Zhang, W.; Yue, X.; Guo, S.; Xiao, P.; Du, Q.; Xing, J.; Deng, L.; Liang, Z.; Dong, A. Biomaterials 2011, 32, (33), 8730-8742. 42. Goffinet, C.; Keppler, O. T. The FASEB Journal 2006, 20, (3), 500-502. 43. Weber, N. D.; Merkel, O. M.; Kissel, T.; Munoz-Fernandez, M. A. Journal of Controlled Release 2012, 157, (1), 55-63. 44. Schacher, F.; Müllner, M.; Schmalz, H.; Müller, A. H. E. Macromolecular Chemistry and Physics 2009, 210, (3-4), 256-262. 45. Mori, H.; Müller, A. H. E.; Klee, J. E. Journal of the American Chemical Society 2003, 125, (13), 3712-3713. 46. Mori, H.; Walther, A.; Andre, X.; Lanzendorfer, M. G.; Müller, A. H. E. Macromolecules 2004, 37, (6), 2054-2066. 47. Muthukrishnan, S.; Plamper, F.; Mori, H.; Müller, A. H. E. Macromolecules 2005, 38, (26), 10631-10642. 48. Freimark, D.; Jérôme, V.; Freitag, R. Biotechnology Journal 2010, 5, (1), 24-31. 49. Chou, L.-F.; Chou, W.-G. Cell Biology International 1999, 23, (10), 663-670. 50. Kim, D.-H.; Behlke, M. A.; Rose, S. D.; Chang, M.-S.; Choi, S.; Rossi, J. J. Nature Biotechnology 2005, 23, (2), 222-226. 51. Parnham, M. J.; Wetzig, H. Chemistry and Physics of Lipids 1993, 64, (1-3), 263274. Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 126 52. Boussif, O.; Lezoualc'h, F.; Zanta, M. A.; Mergny, M. D.; Scherman, D.; Demeneix, B.; Behr, J. P. Proceedings of the National Academy of Science of the United States of America 1995, 92, (16), 7297-7301. 53. Jeon, O.; Lim, H. W.; Lee, M.; Song, S. J.; Kim, B. S. Journal of Drug Targeting 2007, 15, (3), 190-198. 54. Mislick, K. A.; Baldeschwieler, J. D. Proceedings of the National Academy of Science of the United States of America 1996, 93, (22), 12349-12354. 55. Guillem, V. M.; Tormo, M.; Moret, I.; Benet, I.; Garcia-Conde, J.; Crespo, A.; Alino, S. F. Journal of Controlled Release 2002, 83, (1), 133-146. 56. Wang, D.-A.; Narang, A. S.; Kotb, M.; Gaber, A. O.; Miller, D. D.; Kim, S. W.; Mahato, R. I. Biomacromolecules 2002, 3, (6), 1197-1207. 57. Atherton, G. T.; Travers, H.; Deed, R.; Norton, J. D. Cell Growth and Differentiation1996, 7, (8), 1059-1066. 58. Ehrhardt, C.; Schmolke, M.; Matzke, A.; Knoblauch, A.; Will, C.; Wixler, V.; Ludwig, S. Signal Transduction 2006, 6, (3), 179-184. 59. Balci, B.; Dincer, P. Biotechnology Journal 2009, 4, (7), 1042-1045. 60. Billiet, L.; Gomez, J.-P.; Berchel, M.; Jaffrès, P.-A.; Le Gall, T.; Montier, T.; Bertrand, E.; Cheradame, H.; Guégan, P.; Mével, M.; Pitard, B.; Benvegnu, T.; Lehn, P.; Pichon, C.; Midoux, P. Biomaterials 2012, 33, (10), 2980-2990. 61. Dodds, E.; Dunckley, M. G.; Naujoks, K.; Michaelis, U.; Dickson, G. Gene Therapy 1998, 5, (4), 542-551. 62. Grosse, S.; Thévenot, G.; Monsigny, M.; Fajac, I. Journal of Gene Medicine 2006, 8, (7), 845-851. 63. Hamm, A.; Krott, N.; Breibach, I.; Blindt, R.; Bosserhoff, A. K. Tissue Engineering 2002, 8, (2), 235-245. 64. Guo, W.; Lee, R. J. Journal of Controlled Release 2001, 77, (1-2), 131-138. 65. Kircheis, R.; Wightman, L.; Wagner, E. Advanced Drug Delivery Reviews 2001, 53, (3), 341-358. 66. Laga, R.; Carlisle, R.; Tangney, M.; Ulbrich, K.; Seymour, L. W. Journal of Controlled Release 2012, 161, (2), 537-553. 67. Merdan, T.; Kopecek, J.; Kissel, T. Advanced Drug Delivery Reviews 2002, 54, (5), 715-758. 68. Ogris, M.; Brunner, S.; Schuller, S.; Kircheis, R.; Wagner, E. Gene Therapy 1999, 6, (4), 595-605. 69. Üzgün, S.; Akdemir, O.; Hasenpusch, G.; Maucksch, C.; Golas, M. M.; Sander, B.; Stark, H.; Imker, R.; Lutz, J.-F.; Rudolph, C. Biomacromolecules 2010, 11, (1), 39-50. 70. Verbaan, F. J.; Oussoren, C.; van Dam, I. M.; Takakura, Y.; Hashida, M.; Crommelin, D. J.; Hennink, W. E.; Storm, G. International Journal of Pharmaceutics 2001, 214, (1-2), 99-101. 71. Dai, F.; Liu, W. Biomaterials 2011, 32, (2), 628-638. 72. Castanotto, D.; Rossi, J. J. Nature 2009, 457, (7228), 426-433. 73. Takahashi, Y.; Nishikawa, M.; Takakura, Y. Advanced Drug Delivery Reviews 2009, 61, (9), 760-766. 74. Cuccato, G.; Polynikis, A.; Siciliano, V.; Graziano, M.; di Bernardo, M.; di Bernardo, D. BMC Systems Biology 2011, 5, (19), 1-12. 75. Maliyekkel, A.; Davis, B. M.; Roninson, I. B. Cell Cycle 2006, 5, (20), 23902395. Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 127 76. Novina, C. D.; Murray, M. F.; Dykxhoorn, D. M.; Beresford, P. J.; Riess, J.; Lee, S.-K.; Collman, R. G.; Lieberman, J.; Shankar, P.; Sharp, P. A. Nature Medicine 2002, 8, (7), 681-686. 77. Hassani, Z.; Lemkine, G. F.; Erbacher, P.; Palmier, K.; Alfama, G.; Giovannangeli, C.; Behr, J. P.; Demeneix, B. A. Journal of Gene Medicine 2005, 7, (2), 198-207. 78. Mao, S.; Germershaus, O.; Fischer, D.; Linn, T.; Schnepf, R.; Kissel, T. Pharmaceutical Research 2005, 22, (12), 2058-2068. 79. Varkouhi, A. K.; Lammers, T.; Schiffelers, R. M.; van Steenbergen, M. J.; Hennink, W. E.; Storm, G. European Journal of Pharmaceutics Biopharmaceutics 2011, 77, (3), 450-457. 80. Varkouhi, A. K.; Mountrichas, G.; Schiffelers, R. M.; Lammers, T.; Storm, G.; Pispas, S.; Hennink, W. E. European Journal of Pharmaceutical Sciences 2012, 45, (4), 459-466. 81. Kumar, P.; Ban, H. S.; Kim, S. S.; Wu, H.; Pearson, T.; Greiner, D. L.; Laouar, A.; Yao, J.; Haridas, V.; Habiro, K.; Yang, Y. G.; Jeong, J. H.; Lee, K. Y.; Kim, Y. H.; Kim, S. W.; Peipp, M.; Fey, G. H.; Manjunath, N.; Shultz, L. D.; Lee, S. K.; Shankar, P. Cell 2008, 134, (4), 577-586. 82. Majewski, A. P.; Schallon, A.; Jérôme, V.; Freitag, R.; Müller, A. H. E.; Schmalz, H. Biomacromolecules 2012, 13, (3), 857-866. 83. Manganiello, M. J.; Cheng, C.; Convertine, A. J.; Bryers, J. D.; Stayton, P. S. Biomaterials 2012, 33, (7), 2301-2309. Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 134 Figure 4-S4 Analysis of CD4 expression after human T lymphocytes were either mock delivered (EGFPsiRNA) or delivered with hCD4-siRNA. Prior to transfection, Peripheral blood mononuclear cells (PBMCs) were cultivated for 2 – 3 days in QPBL medium to stimulate proliferation of the T lymphocytes. On the day of transfection the cells are ≥ 95 % CD3+ with blast morphology. For transfection, polyplexes were formed with either b-PEI or Si-PDMAEMA and 25 nM siRNA at a N/P ratio of 10. 30 h post-transfection, the cells were stained with CD4-FITC antibody, counterstained with propidium iodide for estimation of the dead cells and then analyzed for CD4 expression by flow cytometry. Cells were initially evaluated by scatter properties (FSC/SSC) in order to select a region representing single non-apoptotic cells (gate “lympho”) and for scatter and fluorescence (SSC/PI) in order to select the living cells (gate “living”). Non-treated cells (“control”), otherwise similarly treated were used to set the regions defined as “CD4high” and “CD4low”. The expression of the CD4 protein was assessed in histogram plots (green fluorescence intensity on the x-axis and cell number on the y-axis) representing the intensity of the CD4-FITC fluorescence (CD4low: fluorescence intensity between 70 and 170; CD4high: fluorescence intensity > 170) in the living T lymphocytes (defined as a sub-population of gate “lympho” and gate “living”). The data are presented as histograms overlays (“control cells”: line and fill color gray “l-PEI”: black line “Si-PDMAEMA”: red line). Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 135 Figure 4-S5 Green fluorescencent protein flow cytometry analysis after transfection of Jurkat cells with Mic-PDMAEMA.For transfection polyplexes were formed with Mic-PDMAEMA and 15 µg/mL pEGFPN1 at N/P 5, 10, and 20. 24 h post-transfection, the cells were analyzed for EGFP expression. Cells were initially evaluated by scatter properties (FSC/SSC) in order to select a region representing single nonapoptotic cells. This gated region (R0) was further analyzed for fluorescence (PI/EGFP). Dot plots with log of the red fluorescence intensity (PI) on the x-axis and log of the green fluorescence intensity (EGFP) on the y-axis were used to estimate the percentage of EGFP-expressing cells in the main non-apoptotic cell population (gate R0). Negative controls (control; non-transfected cells) were used to set the position of quadrants separating GFP-positive living cells (upper left), GFP-positive dead cells (upper right), GFPnegative living cells (lower left) and GFP-negative dead cells (lower right). These quadrants were applied for the analysis of transfected cells and percentage cell number / total cell number in the gated region were calculated for each quadrant. Chapter 4 – Effective Delivery of Nucleic Acids to Differentiated Cells and Human T Lymphocytes 136 Chapter 5 – Double-Layered Micellar Interpolyelectrolyte Complexes – How Many Shells to a Core? 137 Chapter 5 Double-Layered Micellar Interpolyelectrolyte Complexes – How Many Shells to a Core? The results from this chapter have been published in Soft Matter as: “Double-Layered Micellar Interpolyelectrolyte Complexes – How Many Shells to a Core?” by Christopher V. Synatschke, Felix H. Schacher,* Melanie Förtsch, Markus Drechsler, and Axel H. E. Müller* Chapter 5 – Double-Layered Micellar Interpolyelectrolyte Complexes – How Many Shells to a Core? 138 Chapter 5 – Double-Layered Micellar Interpolyelectrolyte Complexes – How Many Shells to a Core? 139 Abstract We report on the formation of double-layered micellar interpolyelectrolyte complexes (IPECs) from ABC triblock terpolymer precursor micelles and hydrophilic homoor block copolymers. Polybutadiene-block-poly(1-methyl-2-vinyl pyridinium)-blockpoly(sodium methacrylate) (PB-b-P2VPq-b-PMANa) block terpolymers form micelles in aqueous solution at high pH exhibiting a PB core, a P2VPq/PMANa intramicellar IPEC (im-IPEC) shell, and a PMANa corona, which is negatively charged. Upon mixing with either positively charged, quaternized poly(N,N-dimethylaminoethyl methacrylate) (PDMAEMAq) homopolymers or its double-hydrophilic block copolymer with poly(ethylene oxide) (PEO-b-PDMAEMAq), a further IPEC shell is formed, rendering core-shell-shell-corona aggregates. The effects of the ratio of positive to negative charges, Z+/-, the composition of the block terpolymer micelles, and the length of the added Dq block were investigated. We show that within a certain Z+/- regime stable complex micellar IPECs featuring two distinguishable IPEC shells are formed. The so-formed complex particles were analyzed by dynamic light scattering and cryogenic transmission electron microscopy. Chapter 5 – Double-Layered Micellar Interpolyelectrolyte Complexes – How Many Shells to a Core? 140 Chapter 5 – Double-Layered Micellar Interpolyelectrolyte Complexes – How Many Shells to a Core? 141 Introduction Self-assembly of block coand terpolymers in solution has received considerable interest during the past decade.1, 2 Typically, such processes lead to the formation of spherical micelles,3, 4 cylindrical or rod-like aggregates,5-7 or vesicles.8 Particular interest is devoted to the control of size, size distribution, shape, or the number or type of functional groups present on the surface of such particles. This can be achieved by controlling the kinetics of the self-assembly process,9 through changing the polymer composition or architecture,10 or via the variation of external parameters like the employed solvent,11, 12 pH,13 salinity,14 or temperature.15 Another possibility to influence self-assembly processes in solution is to employ block copolymers with charged compartments, or polyelectrolyte segments.16, 17 Such polyelectrolytes can be natural (e.g., polynucleic acids) or synthetic polyanions or - cations and can be further subdivided into weak (e.g., poly(methacrylic acid), PMAA) or strong (e.g., poly(styrene sulfonate)) species.18 Recent research interest in such materials has been primarily based on intrinsic properties such as water solubility, very strong interand intra-chain interactions, ionic conductivity, and surface activity.19 Mixing of two different block copolymers with polyelectrolyte segments bearing opposite charges in aqueous solution leads to electrostatic co-assembly and the formation of interpolyelectrolyte complexes (IPECs).20-22 The driving force is the entropy gain from the release of the counterions. Such IPECs are hydrophobic yet are still able to participate in dynamic polyion exchange reactions in aqueous media.23, 24 If weak polyelectrolytes like PMAA are used, the complex formation is pH-dependent.25 In addition, the IPEC formation is reversible: the addition of large amounts of salt leads to a screening of the charges and to a breakup of the complexes.20, 26 If preformed micelles with a charged corona are mixed with oppositely charged polyelectrolytes, an elegant route towards complex micellar architectures is opened. This has been demonstrated for micelles exhibiting a soft polyisobutylene core and a PMAA corona25, 26 or for more complex corecompartmentalized block terpolymer micelles.27 Within this contribution we demonstrate for the first time the formation of two distinctly different adjacent IPEC shells within the same complex micellar aggregate. As starting material we employ multicompartment micelles exhibiting a soft polybutadiene (PB) core, a discontinuous shell consisting of an intramicellar IPEC (im-IPEC) between Chapter 5 – Double-Layered Micellar Interpolyelectrolyte Complexes – How Many Shells to a Core? 142 quaternized poly(2-vinylpyridine) (P2VPq) and poly(sodium methacrylate) (PMANa), and a corona of excess PMANa, thus rendering colloidal objects with a negative surface charge. These micelles are formed if ampholytic block terpolymers, polybutadiene-blockpoly(1-methyl-2-vinyl pyridinium)-block-poly(sodium methacrylate) (PB-b-P2VPq-bPMANa), self-assemble in aqueous solution at pH 10, conditions where methacrylic acid is negatively charged.28 To this is added another solution containing a positively charged polyelectrolyte, being either quaternized poly(N,N-dimethylaminoethyl methacrylate) (PDMAEMAq) homopolymers of different chain length or the corresponding doublehydrophilic block copolymer with poly(ethylene oxide) (PEO-b-PDMAEMAq). This then results in further IPEC formation between the PMANa corona and PDMAEMAq and the generation of a second IPEC shell. The whole process is shown in Scheme 5-1. Scheme 5-1: Schematic pathway for the formation of double-layered IPECs from PB-b-P2VPq-b-PMANa block terpolymer micelles and either PDMAEMAq homopolymers (left) or a PEO-b-PDMAEMAq diblock copolymer (right). Chapter 5 – Double-Layered Micellar Interpolyelectrolyte Complexes – How Many Shells to a Core? 143 In case of PEO-b-PDMAEMAq, the PEO chains then serve as the corona of the generated core-shell-shell-corona particle and maintain a good solubility in aqueous media. We investigated two PDMAEMAq homopolymers of different chain length, in short: Dq162 and Dq820, and one double-hydrophilic block copolymer, EO325Dq157, the subscripts denoting the degree of polymerization of the corresponding block. As precursor micelles three different block terpolymers, PB800-b-P2VPq190-b-PMANa345, PB800-b-P2VPq190-bPMANa465, and PB800-b-P2VPq190-b-PMANa550, were used. In the following sections a shorter nomenclature, BVqMANax, will be used for the precursor micelles, as the first two blocks (B and Vq) have the same degree of polymerization in all cases shown in this manuscript. The complexations were performed at different Z+/- values. We define Z+/- by dividing the number of cationic Dq monomer units added to the micellar solution divided by the number of free (non-complexed) anionic MANa units present, as shown in equation (5-1). / Dq MANa Vq n Znn  (5-1) The structure and the stability of the formed micellar IPECs were analyzed using dynamic light scattering (DLS) and cyrogenic transmission electron microscopy (cryo-TEM). Experimental Synthesis Materials. The solvents for the preparation of the micellar solutions were purchased in p.a. grade and used as delivered. Dimethyl sulfate (Me2SO4, >99 %, Aldrich) and hydrochloric acid (32 %) were used as received. Buffer solutions with pH 10 (H3BO3/KCl/NaOH) were obtained from Fluka and contained about 0.3 wt. % salt. The initiator for ATRP, ethylbromo isobutyrate (EBIB) was distilled and stored under nitrogen. HMTETA was distilled prior to use. Anisole (p.a. grade, Fluka) and CuBr (>99 %, Aldrich) were used as received. Acknowledgements 246 The MC2-Team deserves a very BIG thanks. Working in the group has been a wonderful experience, and I have never met a group that shared a similar spirit as we did in MC2. Thanks for the everyday help in the laboratories, for many a coffee, cake, beer and Schnapps shared in the kitchen, the countless barbeques and all the fun we had on our trips to conferences. The technicians kept the lab in good working condition. I want to thank Melanie Förtsch and Annika Pfaffenberger for taking care of all my TEM samples and together with Dane Blasser for forming my “MALDI-Team”. Together we kept the old lady in good shape. Marietta Böhm, I thank for always being there to measure the tricky samples I prepared for SEC. Further, Annette Krökel took care of much of the administrative work occurring in the lab, and could always tell me where to find the odd missing piece of equipment, glassware or consumable. Thank you for that. I also thank the “good soul” of the MC2, Gaby Oliver, who somehow always managed to get things done with the university administration even on very short notice. She worked her miracle ways to re-organize the group money and always found a follow-up contract from various sources for me. Many people came and went in MC2 during my time there. Thanks for the remarkable atmosphere you created (in no particular order): Markus Müllner, Andreas Hanisch, Thomas Ruhland, Alexander Majewski, André Gröschel, Tina Löbling, Eva Betthausen, Andrea Wolf, Stephan Weiß, Sandrine Tea, André Pfaff, Alexander Schmalz, Joachim Schmelz, Stefan Reinicke, Zhicheng Zheng, Francesca Bennet, Holger Schmalz, Markus Drechsler, Anja Goldmann, Jiayin Yuan, Andreas Walther, Jun Ling, Jie Kong, Weian Zhang, Marina Krekhova, Stefan Döhler, Shohei Ida, Tony Granville, Hülya Arslan, Ivan Babin, Meirav Ben-Lulu and Tomohiro irano. The “Russian Mafia” consisting of Dmitry (Dima) Pergushov, Larissa Siegolaeva, Alexander (Sasha) Yakimansky and Oleg Borisov, as well as the corresponding “Spanish Mafia”: Ramón Novoa-Carballal, Lourdes Pastor-Pérez and Ainhoa Tolentino Chivite. My students and “ iwi´s” also deserve a big thank you for all the effort they put into their work: Annika Eckardt, Fabian Pooch, Irina Weber and Tobias Rudolph. I would also like to thank Prof. Ruth Freitag for the good collaboration I had with her group and for the support, especially in the final part of my thesis. Acknowledgements 247 From conferences, meetings and collaborations I thank those people who did influence my work, or discussed ideas. Steffen Weidner, Jana Falkenhagen, Volker Sauerland, Christo Tsvetanov, Darrin Pochan, Christopher Barner-Kowollik, Helmut Schlaad, Felix Plamper, Vladimir Tsukruk, Jürgen Senker, Andreas Fery, Takuzo Aida, Matthias Karg, Markus Retsch, Helmut Ringsdorf, Amir Fahmi, Harald Pasch, Ian Manners and Eugenia Kumacheva. During my 6-months research stay at the University of Tokyo, I met many people who made my time there an exceptional experience. First among those people is Prof. Kazunori Kataoka, whom I deeply respect and who has impressed me as being a researcher of rare quality. Within his group or connected with the University of Tokyo I want to thank Horacio Cabral, Takahiro Nomoto, Stefanie Deshayes, Philippe Saint-Cricq Riviere, Tomeu (The Bastard) Soberats, Adrien Kaeser, Xiao Ling, Kazuko Toh, Yu Matsumoto, Sayan Chuanoi, Yuuki Mochida, Mitsuru Naito, Akihiro Kishimura, Nobuhiro Nishiyama, R. James (Jim) Christie and Yutaka Miura. I am further grateful for financial support I received from various sources during my work, most importantly from the State of Bavaria, the German Academic Exchange Service, the Center for Medical Systems Innovation and the SFB 840. Many friends have supported me throughout the PhD process and I cannot name you all. Let it suffice to say you are an important part of my life. Thank you Anja, for all the love and support you gave me during those wonderful 6 years that we shared. I wish you the best for your future and hope to remain a small part of that. Most importantly, I want to thank my family. My mother Elisabeth I thank for supporting me unconditionally, always listening to my adventures, successes and in some rare cases my frustrations. Both my sisters Dagmar and Franziska have been an inspiration to me and have shown me choices I might not have known without them. Acknowledgements 248 249 “There is a theory which states that if ever anyone discovers exactly what the Universe is for and why it is here, it will instantly disappear and be replaced by something even more bizarre and inexplicable. There is another theory which states that this has already happened.” ― Douglas Adams, The Restaurant at the End of the Universe 250 251 Erklärung Die vorliegende Arbeit wurde von mir selbständig verfasst und ich habe dabei keine anderen als die von mir angegebenen Hilfsmittel oder Quellen verwendet. Ferner habe ich nicht versucht, anderweitig mit oder ohne Erfolg eine Dissertation einzureichen oder mich einer Doktorprüfung zu unterziehen. Bayreuth, den 15. Januar 2014 Christopher Volker Synatschke