Refine
Has Fulltext
- yes (31)
Is part of the Bibliography
- yes (31)
Year of publication
Document Type
- Doctoral Thesis (31) (remove)
Keywords
- Antikörper (31) (remove)
Institute
- Theodor-Boveri-Institut für Biowissenschaften (9)
- Institut für Virologie und Immunbiologie (4)
- Medizinische Klinik und Poliklinik II (4)
- Abteilung für Molekulare Innere Medizin (in der Medizinischen Klinik und Poliklinik II) (3)
- Graduate School of Life Sciences (3)
- Julius-von-Sachs-Institut für Biowissenschaften (3)
- Missionsärztliche Klinik (2)
- Neurologische Klinik und Poliklinik (2)
- Pathologisches Institut (2)
- Institut für Molekulare Infektionsbiologie (1)
CD40-restringierte Aktivierung der TRAIL-Todesrezeptoren durch bifunktionelle rekombinante Proteine
(2014)
Der Ligand TRAIL wurde 1997 aufgrund seiner hohen Sequenzhomolgie ge-genüber dem TNFL CD95L entdeckt (28 %). Allerdings besitzt TRAIL, anders als die Liganden CD95L und TNF, die bemerkenswerte Eigenschaft vor allem in veränderten Zellen Apoptose zu induzieren, während gesunde Zellen davor bewahrt werden. Die TRAIL-induzierte Apoptose wird durch die apoptoseinduzierenden Todesrezeptoren TRAILR1 und TRAILR2 vermittelt. Allerdings bindet und aktiviert lösliches TRAIL hauptsächlich den Todesrezeptor TRAILR1, während membrangebundes TRAIL sowohl TRAILR1 als auch TRAILR2 gut aktiviert. In den letzten Jahren wurden verschiedene Methoden entwickelt, um die Bioaktivität löslicher TNFL zu steigern. Hierzu zählen z.B.: Stabilisierung der trimeren Molekülanordnung über die TNC-Domäne, Oligomerisierung des Flag-getaggten Liganden mithilfe des monoklonalen Antikörpers M2, sowie Generierung einer artifiziellen, antigenabhängigen Membranständigkeit. In dieser Arbeit wurde der Oberflächenrezeptor CD40 zur Immobilisierung des generierten Fusionsproteins scFv:CD40-Flag-TNC-TRAIL genutzt. In verschieden Experimenten konnten mit scFv:CD40-Flag-TNC-TRAIL in CD40-exprimierenden Zellen starke Apoptoseinduktion ermittelt werden. Charakteris-tische Kennzeichen und Spaltprodukte der Apoptose konnten ausschließlich in CD40-positiven Tumorzellen detektiert werden. Dabei wurde in allen Versuchen die für die Apoptoseinduktion benötigte Konzentration des Konstrukts mithilfe des Proteinsyntheseinhibitors CHX um das 10- bis 100-fache verringert. Es konnte auch gezeigt werden, dass in CD40-positiven Zellen, nach Stimulation mit scFv:CD40-Flag-TNC-TRAIL, nicht-apoptotische Signalwege verstärkt aktiviert werden. Dies war auf die agonistische Aktivität des monoklonalen Antikörperfragments scFv:CD40 zurückzuführen. Die Antikörperdomäne war folglich nicht nur zur effizienten Aktivierung der TRAIL-Todesrezeptoren mittels Immobilisierung fähig, sondern konnte zusätzlich zur Stimulation des Immunsystems genutzt werden. Zusammenfassend konnte gezeigt werden, dass der lösliche, schwach aktive Ligand TRAIL mittels Oberflächenimmobilisierung über Antigen-Antikörper-Wechselwirkungen in einen hochaktiven Liganden mit lokal begrenzter Toxizität überführt werden kann.
Mithilfe dieses Fusionsproteins ist es somit möglich die selektive Toxizität von TRAIL durch Steigerung seiner Aktivität effizient zu nutzen. Zusätzlich kann durch die Antigenbindung der Wirkungsbereich weiter eingegrenzt werden (CD40-positive Tumoren), wodurch unerwünschte Nebenwirkungen reduziert oder sogar ausgeschaltet werden können. Das in Tumoren oft heruntergefahrene Immunsystem kann CD40-abhängig stimuliert werden, um somit auch Tumorzellen in apoptoseresistenten Stadien zu eliminieren. Basierend auf diesen Ergebnissen können in der Zukunft weitere Studien zur Therapie von TRAIL-resistenten, CD40-exprimierenden Tumoren fortgeführt werden.
Transmissible spongiform encephalopathies (TSEs) or prion diseases are a group of infectious neurodegenerative diseases that are associated with misfolding of the cellular form of the cellular prion protein (PrPC) into a disease associated conformer (PrPSc). No therapy for prion diseases is available at present. So far, anti-PrPC vaccination is hampered by immunological tolerance of the mammalian immune system to endogenous PrPC. The aim of this thesis was to set up a new vaccination strategy based on virus-like particles (VLP) to induce anti-PrPC antibody responses in PrPC-competent mice. In a first step it was assessed whether VLP have the capacity to induce antibody responses that are protective against conventional pathogens. For this purpose, VLP displaying the vesicular stomatitis virus-gylcoprotein (VLP-VSV) were generated and tested for their immunogenicity. Similarly to live vesicular stomatitis virus (VSV), replication deficient VLP-VSV induced T help-independent VSV neutralizing IgM responses that switched to the IgG subclass in a T help-dependent manner. Furthermore, type I IFN receptor (IFNAR) triggering only marginally affected VLP-VSV induced neutralizing IgM responses, whereas it was critically required to promote the IgG switch. The analysis of conditional knockout mice with a lymphocyte-specific IFNAR deletion revealed that IFNAR triggering of lymphocytes did not play a crucial role, neither upon VLP-VSV nor VSV immunization. Collectively, these data verified the high immunogenicity of VLP. Therefore, in a next step VLP were generated displaying the C-terminal half of PrP (residues 121-231aa) fused to the platelet derived growth factor receptor (PDGFR) transmembrane region (VLP-PrPD111) for anti-PrPC immunization. On the surface of such retroparticles, PrPC was expressed at high levels as determined by electron microscopy. VLP-PrPD111 immunization of Prnp-deficient (Prnp0/0) mice resulted in antibody response specifically binding the cellular form of PrPC. Upon intravenous injection of wild-type mice, high PrPC-specific IgM responses were induced, whereas the T cell-dependent switch from the IgM to the IgG subclass was less pronounced. As a consequence, anti-PrPC titers were rather short-lived. The impaired subclass switch was probably related with host T cell tolerance to endogenous PrPC. Attempts to increase anti-PrPC IgG responses in wild-type mice via administration of VLP-PrPD111 emulsified in various different adjuvants failed. Nevertheless, in single individuals low IgG antibodies were induced after immunization of VLP-PrPD111 emulsified in CFA. To circumvent T cell tolerance in wild-type mice, a multitude of different immunization strategies was tested, including priming and boosting protocols with different types of VLP or VLP expressing PrPC together with foreign T helper epitopes. Overall, those efforts did not improve anti-PrPC IgG responses in wild-type mice. Interestingly, anti-PrPC antibodies induced in Prnp0/0 mice reduced PrPSc levels in prion infected cell cultures, whereas serum of vaccinated wild-type mice did not. To assess the protective capacity of VLP-PrPD111 induced immune responses, vaccinated wild-type mice were infected with scrapie (RML 5.0). Unfortunately, vaccinated mice did not show a significant delay in the onset of scrapie. In a last part of the thesis it was studied whether in the absence of T cell help activated “memory” B cells were able to produce anti-PrPC specific antibodies. To address this question, PrPC-specific memory B cells were sorted from vaccinated Prnp0/0 mice and adoptively transferred into wild-type recipient mice. Upon VLP-PrPD111 challenge, no PrPC-specific IgG titers were induced in the recipients. Nevertheless, several VLP-PrPD111 challenged recipient mice were protected against scrapie infection. In conclusion, VLP were characterized as highly immunogenic vaccines that were used to elucidate various questions concerning adaptive immune response and basic mechanisms of PrPC-specific tolerance vs. immunity. Remarkably, VLP-PrPD111 was able to induce native PrPC-specific antibodies in wild-type mice but major difficulties associated with PrPC-specific tolerance made efficacious scrapie vaccination impossible. New vaccination approaches are being tested to overcome these limitations.
Cancer cells frequently escape from immune surveillance by down-regulating two important components of the immune defence: antigen-presenting MHC and costimulatory molecules. Therefore several novel anti-tumour compounds that aim to assist the immune system in recognising and fighting cancer are currently under development. Recombinant bispecific antibodies represent one group of such novel therapeutics. They target two different antigens and recruit cytotoxic effector cells to tumour cells. For cancer immunotherapy, bispecific T cell-engaging antibodies are already well characterised. These antibodies target a tumour-associated antigen and CD3ε, the constant molecule of the T cell receptor complex.
On the one hand, this study presents the development of a bispecific antibody targeting CD3ε and the rhabdomyosarcoma-associated fetal acetylcholine receptor. On the other hand, it describes a novel two-part trispecific antibody format for the treatment of leukaemia and other haematological malignancies in the context of haematopoietic stem cell transplantation (HSCT).
For HSCT, an HLA-identical donor is preferred, but very rarely available. In an HLA-mismatched setting, the HLA disparity could be exploited for targeted cancer treatment. In the present study, a two-part trispecific HLA-A2 × CD45 × CD3 antibody was developed for potential cases in which the patient is HLA-A2-positive, but the donor is not. This holds true for about half the cases in Germany, since HLA-A2 is the most common HLA molecule found here. Combinatorial targeting of HLA-A2 and the leucocyte-common antigen CD45 allows for highly specific dual-antigen restricted tumour targeting.
More precisely, two single-chain antibody constructs were developed: i) a single-chain variable fragment (scFv) specific for HLA-A2, and ii) a scFv against CD45, both linked to the VL and the VH domain of a CD3ε-specific antibody, respectively. It turned out that, after the concomitant binding of these constructs to the same HLA-A2- and CD45-expressing cell, the unpaired variable domains of a CD3ε-specific antibody assembled to a functional scFv. In a therapeutic situation, this assembly should exclusively occur on the recipient’s blood cancer cells, leading to T cell-mediated cancer cell destruction. In this way, a relapse of disease might be prevented, and standard therapy (radiation and chemotherapy) might be omitted.
For both approaches, the antibody constructs were periplasmically expressed in E. coli, purified via His tag, and biochemically characterised. Their binding to the respective targets was proven by flow cytometry. The stimulatory properties of the antibodies were assayed by measuring IL-2 release after incubation with T cells and antigen-expressing target cells. Both the bispecific antibody against rhabdomyosarcoma and the assembled trispecific antibody against blood cancer mediated T-cell activation in a concentration-dependent manner at nanomolar concentrations. For the trispecific antibody, this effect indeed proved to be dual antigen-restricted, as it could be blocked by prior incubation of either HLA-A2- or CD45-specific scFv and did not occur on single-positive (CD45+) or double-negative (HLA-A2- CD45-) target cells. Furthermore, antibodies from both approaches recruited T cells for tumour cell destruction in vitro.
In Deutschland erkranken jährlich etwa 500.000 Menschen an Krebs, wovon circa
12.000 die Diagnose „Leukämie“ gestellt bekommen [1]. Unter den Leukämien weist
die akute myeloische Leukämie (AML) die ungünstigste Prognose auf, sodass hier
erheblicher Forschungsbedarf besteht. Zusätzlich schnitten viele potentielle Therapeutika,
die sich in bisherigen präklinischen Testsystemen als vielversprechend erwiesen
haben, in klinischen Studien schlecht ab [8]. Ziel dieser Arbeit war daher die
Etablierung eines 3D in vitro Blutgefäß-/Gewebemodells als verbessertes präklinisches
System zur Testung von Therapeutika, die zur erfolgreichen Behandlung von
Leukämien beitragen sollen.
Das 3D Blutgefäßmodell bestand aus humanen primären Endothelzellen, welche
als Monolayer auf der Serosaseite einer dezellularisierten, porzinen, intestinalen Kollagenmatrix
(SIS-Ser) wuchsen. Nach 14-tägiger Zellkultur wurden dem Versuchsansatz
entsprechend nichtadhärente THP-1 Zellen (AML-M5-Zelllinie) und Tipifarnib
oder entsprechende Kontrolllösungen beziehungsweise bimolekulare Antikörperkonstrukte
mit PBMCs als Effektorzellen hinzupipettiert. Nach 5-tägiger Inkubation
mit Tipifarnib beziehungsweise 24-stündiger Behandlung mit Antikörperkonstrukten
wurde der therapiebedingte Anstieg der Apoptoserate in den malignen THP-1 Zellen
mittels durchflusszytometrischer Analyse der Modellüberstände ermittelt. Zum
Ausschluss verbliebener und durchflusszytometrisch zu analysierender Zellen wurde,
stellvertretend für alle Suspensionszellen, eine Anti-CD13/DAB-Färbung durchgeführt,
welche negativ ausfiel. Mögliche Kollateralschäden am Endothel wurden
mittels histologischen Färbemethoden an Gewebeparaffinschnitten untersucht.
In der Durchflusszytometrie zeigte Tipifarnib sowohl im 2D als auch im 3D Modell
äquivalente, dosisabhängige und antileukämische Auswirkungen auf die THP-1 Zellen.
Bei Applikation der Antikörperkonstrukte ließ lediglich die Kombination beider Hemibodies signifikante Effekte auf die THP-1 Zellen erkennen. Dabei zeigten sich
bei konstanten Konzentrationen der Antikörperkonstrukte im 3D Modell deutlich
höhere Apoptoseraten (58%) als im 2D Modell (38%). Stellt man Vergleiche von
Tipifarnib mit den T-Zell-rekrutierenden Antikörperkonstrukten an, so ließen sich
im 2D Modell ähnliche Apoptoseraten in den THP-1 Zellen erzielen (jeweils 38% bei
Anwendung von 500 nM Tipifarnib). In den 3D Modellen erzielten jedoch die niedriger
konzentrierten Antikörperkonstrukte bei kürzerer Inkubationsdauer eine noch
höhere spezifische Apoptoserate in den THP-1 Zellen (im Mittel 58%) als 500 nM
Tipifarnib (mittlere Apoptoserate 40%). Bezüglich der Nebenwirkungen ließ sich
im 3D Modell nach Applikation von Antikörperkonstrukten kein wesentlicher Einfluss
auf das Endothel erkennen, während Tipifarnib/DMSO als auch die mit DMSO
versetzten Kontrolllösungen zu einer dosisabhänigen Destruktion des ursprünglichen
Endothelzellmonolayers führten. Damit stellt die hier beschriebene, hoch spezifische,
Hemibody-vermittelte Immuntherapie einen vielversprechenden Ansatz für zukünftige
onkologische Therapien dar.
Mithilfe des etablierten humanen 3D in vitro Modells konnte im Vergleich zur
konventionellen Zellkultur eine natürlichere Mikroumgebung für Zellen geschaffen
und die Auswirkungen der Testsubstanzen sowohl auf maligne Zellen, als auch auf
die Gefäßstrukturen untersucht werden.
Staphylococcus aureus ist einer der häufigsten Erreger von nosokomialen Infektionen. Diese grampositiven Bakterien verursachen neben harmlosen oberflächlichen Hautinfektionen auch lebensbedrohliche Systeminfektionen. Ein großes Problem in der Therapie von S. aureus-Infektionen stellen die zunehmenden Multiresistenzen dar. Die Entwicklung neuer Antibiotika wird zukünftig wahrscheinlich nicht ausreichen, da immer wieder neue Resistenzen der Bakterien zu erwarten sind. Es besteht daher dringender Bedarf an der Entwicklung alternativer Therapieformen im Kampf gegen multiresistente Problemkeime wie S. aureus. Eine Möglichkeit besteht in der Immuntherapie, zum Beispiel durch Gewinnung von monoklonalen Antikörpern gegen geeignete Targetstrukturen von S. aureus. Ziel dieser Arbeit war es, zunächst zwei Proteine IsaA und IsaB herzustellen, um diese Proteine für Immunisierungsstudien zu nutzen. Zunächst wurde das gereinigte IsaA-Protein verwendet, um ein Kaninchen zu immunisieren. Mit den daraus gewonnenen Antikörpern wurden dann erste Tierversuche begonnen, um die Bedingungen für den therapeutischen Einatz von gegen IsaA-gerichteten Antikörpern zu ermitteln und die Wirksamkeit einer Antikörper-Behandlung zu evaluieren. Für die Herstellung der gewünschten Proteine wurden die Gensequenzen zunächst aus verschiedenen S. aureus-Stämmen mittels PCR amplifiziert und in den kommerziellen Expressionsvektor pQE30 kloniert. Die amplifizierte Gensequenz stammt aus den klinischen Stämmen 418 (IsaA) bzw. 134 (IsaB). Nach der Klonierung wurden geeignete Expressions- und Reinigungsstrategien entwickelt. Dabei wurden folgende Bedingungen als optimal für Wachstum und Überexpression herausgearbeitet: IsaA: Induktion der Überexpression mit 100 µM IPTG, 3 h Wachstum bei 37°C. IsaB: Induktion der Überexpression mit 100 µM IPTG, 4 h Wachstum bei 37°C. Es stellte sich auch heraus, dass IsaA zunächst in nur unzureichender Quantität vorhanden bzw. exprimiert worden war. Die Vermutung, dass IsaA überwiegend im Pellet in sogenannten Einschlusskörpern (inclusion bodies) eingeschlossen war, erklärte dieses Phänomen. Das Protein konnte erfolgreich aus dem Pellet isoliert werden. Die Produktion und Aufreinigung beider Proteine IsaA und IsaB unter optimierten Bedingungen ergab, dass beide Proteine nun in ausreichender Menge und Konzentration für die folgende Immunisierung und die weiteren Arbeiten vorlagen. Aus Kaninchen, die mit IsaA immunisiert wurden, konnten polyklonale Antikörper gewonnen werden, die die Grundlage für einen ersten Tierversuch mit 24 Ratten bildeten. Hierbei zeigte sich, dass die Tiere, die mit 1.000.000.000 Bakterien infiziert worden waren deutlich stärkere Infektionszeichen aufwiesen als diejenigen, die mit 100.000.000 Bakterien infiziert worden waren. Weiterhin wurde deutlich, dass die Tiere, die Serum (mit Antikörper gegen IsaA) erhalten hatten, gegenüber den Vergleichstieren mit Placebo einen deutlichen Vorteil hinsichtlich Infektionszeichen und Immunantwort hatten. Somit belegen die tierexperimentiellen Ergebnisse in dieser Arbeit erstmalig den therapeutischen Nutzen von Antikörpern gegen IsaA. IsaA ist demnach ein geeignetes Target für eine Immuntherapie gegen S. aureus.
Millionen Menschen weltweit leiden an den verschiedensten Autoimmunerkrankungen. Diese Krankheiten entstehen, wenn das Immunsystem gesundes körpereigenes Gewebe angreift und zerstört. An der Pathogenese sind sowohl Komponenten des angeborenen Immunsystems als auch Bestandteile des adaptiven Immunsystems, wie Lymphozyten und Antikörper, beteiligt. Da die Ursachen und molekularen Mechanismen der Pathogenese dieser Erkrankungen bis heute weitgehend unbekannt sind, wurden in dieser Arbeit autoaggressive Lymphozyten bei den humanen Autoimmunerkrankungen Polymyositis und Multiple Sklerose näher untersucht. Die Polymyositis ist eine chronisch entzündliche Erkrankung der Skelettmuskulatur. Die Muskelfasern werden dabei von zytotoxischen CD8+ gd-T-Lymphozyten infiltriert, attackiert und schließlich zerstört. In einem seltenen Fall der Polymyositis wurden die Muskelzellen hingegen in ähnlicher Weise von CD8- gd-T-Lymphozyten angegriffen. Die gd-T-Lymphozyten waren monoklonal expandiert und ihr Rezeptor, im Folgenden als M88 bezeichnet, wurde als Vg1.3+Vd2+ identifiziert. Frühere Untersuchungen der Antigenspezifität dieser Zellen zeigten, dass M88 mehrere funktionell und strukturell verschiedene Proteine aus unterschiedlichen Spezies erkennt. Die Bindung erfolgt spezifisch durch die Antigenerkennungsregionen beider Rezeptorketten von M88. In dieser Arbeit wurden verschiedene bakterielle und humane Proteine des Translationsapparates als Antigene von M88 identifiziert. Weitere ausführliche Untersuchungen eines paradigmatischen bakteriellen Antigens, dem Translationsinitiationsfaktor EcIF1, zeigten, dass M88 an Oberflächen-exponierte Konformationsepitope von Proteinen bindet. Interessanterweise erkennt M88 mehrere humane Aminoacyl-tRNA-Synthetasen, Antigene, die in anderen Formen der Myositis von Autoantikörpern angegriffen werden. Diese Beobachtung ergibt eine bemerkenswerte Verbindung zwischen T-Zell- und Antikörper-vermittelten B-Zell-Antworten bei der autoimmunen Myositis. Bei der Multiplen Sklerose ist das zentrale Nervensystem betroffen. Autoaggressive Lymphozyten greifen die Myelinschicht der Nervenzellen im Gehirn und Rückenmark an und zerstören sie. Im Liquor cerebrospinalis von Patienten lassen sich klonal expandierte und affinitätsgereifte B-Zellen sowie „oligoklonale Banden“ (OKB) Antikörper nachweisen. Obwohl diese Merkmale auf eine Antigen-induzierte Immunantwort hindeuten, sind die zugrundeliegenden Antigene und die Rolle der OKB bei der Pathogenese bis heute unbekannt. In dieser Arbeit wurde die Antigenspezifität von fünf IgG OKB-Antikörpern aus drei Patienten untersucht. Durch verschiedene proteinbiochemische Methoden konnten intrazelluläre Kandidatenantigene identifiziert werden. Interessanterweise sind darunter mehrere nukleäre Proteine, die an der Transkriptionsregulation oder der RNA-Prozessierung beteiligt sind. Reaktivitäten gegen intrazelluläre Antigene treten auch bei anderen Autoimmunerkrankungen, wie beispielsweise dem systemischen Lupus erythematodes, auf. Diese Ergebnisse könnten auf einen allgemeinen Mechanismus der Entstehung und Funktion von Autoantikörpern bei diesen humanen Autoimmunerkrankungen hindeuten.
The cancer stem cell hypothesis is a cancer development model which elicited great interest in the last decades stating that cancer heterogeneity arises from a stem cell through asymmetrical division. The Cancer Stem Cell subset is described as the only population to be tumorigenic and having the potential to renew. Conventional therapy often fails to eradicate CSC resulting in tumor relapse. Consequently, it is of great inter-est to eliminate this subset of cells to provide the best patient outcome. In the last years several approaches to target CSC were developed, one of them being immunotherapeu-tic targeting with antibodies. Since markers associated with CSC are also expressed on normal stem cells or healthy adjacent tissue in colorectal cancer, dual targeting strate-gies are preferred over targeting only a single antigen. Subsequently, the idea of dual targeting two CSC markers in parallel by a newly developed split T cell-engaging anti-body format termed as Hemibodies emerged. In a preliminary single cell RNA sequenc-ing analysis of colorectal cancer cells CD133, CD24, CD166 and CEA were identified as suitable targets for the combinatorial targeting strategy. Therefore, this study focused on trispecific and trivalent Hemibodies comprising a split binding moiety against CD3 and a binding moiety against either CD133, CD24, CD166 or CEA to overcome the occurrence of resistance and to efficiently eradicate all tumor cells including the CSC compartment. The study showed that the Hemibody combinations CD133xCD24, CD133xCD166 and CD133xCEA are able to eliminate double positive CHO cells with high efficacy while having a high specificity indicated by no killing of single antigen positive cells. A thera-peutic window ranging between one to two log levels could be achieved for all combina-tions mentioned above. The combinations CD133xCD24 and CD133xCD166 further-more proved its efficacy and specificity on established colorectal cancer cell lines. Be-sides the evaluation of specificity and efficacy the already introduced 1st generation of Hemibodies could be improved into a 2nd generation Hemibody format with increased half-life, stability and production yield. In future experiments the applicability of above-mentioned Hemibodies will be proven on patient-derived micro tumors to also include variables like tumor microenvironment and infiltration.
Functionalization of cells, extracellular matrix components and proteins for therapeutic application
(2019)
Glycosylation is a biochemical process leading to the formation of glycoconjugates by linking glycans (carbohydrates) to proteins, lipids and various small molecules. The glycans are formed by one or more monosaccharides that are covalently attached, thus offering a broad variety depending on their composition, site of glycan linkage, length and ramification. This special nature provides an exceptional and fine tunable possibility in fields of information transfer, recognition, stability and pharmacokinetic. Due to their intra- and extracellular omnipresence, glycans fulfill an essential role in the regulation of different endogenous processes (e.g. hormone action, immune surveillance, inflammatory response) and act as a key element for maintenance of homeostasis. The strategy of metabolic glycoengineering enables the integration of structural similar but chemically modified monosaccharide building blocks into the natural given glycosylation pathways, thereby anchoring them in the carbohydrate architecture of de novo synthesized glycoconjugates. The available unnatural sugar molecules which are similar to endogenous sugar molecules show minimal perturbation in cell function and - based on their multitude functional groups - offer the potential of side directed coupling with a target substance/structure as well as the development of new biological properties. The chemical-enzymatic strategy of glycoengineering provides a valuable complement to genetic approaches.
This thesis primarily focuses on potential fields of application for glycoengineering and its further use in clinic and research. The last section of this work outlines a genetic approach, using special Escherichia coli systems, to integrate chemically tunable amino acids into the biosynthetic pathway of proteins, enabling specific and site-directed coupling with target substances. With the genetic information of the methanogen archaea, Methanosarcina barkeri, the E. coli. system is able to insert a further amino acid, the pyrrolysine, at the ribosomal site during translation of the protein. The natural stop-codon UAG (amber codon) is used for this newly obtained proteinogenic amino acid.
Chapter I describes two systems for the integration of chemically tunable monosaccharides and presents methods for characterizing these systems. Moreover, it gives a general overview of the structure as well as intended use of glycans and illustrates different glycosylation pathways. Furthermore, the strategy of metabolic glycoengineering is demonstrated. In this context, the structure of basic building blocks and the epimerization of monosaccharides during their metabolic fate are discussed.
Chapter II translates the concept of metabolic glycoengineering to the extracellular network produced by fibroblasts. The incorporation of chemically modified sugar components in the matrix provides an innovative, elegant and biocompatible method for site-directed coupling of target substances. Resident cells, which are involved in the de novo synthesis of matrices, as well as isolated matrices were characterized and compared to unmodified resident cells and matrices. The natural capacity of the matrix can be extended by metabolic glycoengineering and enables the selective immobilization of a variety of therapeutic substances by combining enzymatic and bioorthogonal reaction strategies. This approach expands the natural ability of extracellular matrix (ECM), like the storage of specific growth factors and the recruitment of surface receptors along with synergistic effects of bound substances. By the selection of the cell type, the production of a wide range of different matrices is possible.
Chapter III focuses on the target-oriented modification of cell surface membranes of living fibroblast and human embryonic kidney cells. Chemically modified monosaccharides are inserted by means of metabolic glycoengineering and are then presented on the cell surface. These monosaccharides can later be covalently coupled, by “strain promoted azide-alkyne cycloaddition“ (SPAAC) and/or “copper(I)-catalyzed azide-alkyne cycloaddition“ (CuAAC), to the target substance. Due to the toxicity of the copper catalysator in the CuAAC, cytotoxicity analyses were conducted to determine the in vivo tolerable range for the use of CuAAC on living cell systems. Finally, the efficacy of both bioorthogonal reactions was compared.
Chapter IV outlines two versatile carrier – spacer – payload delivery systems based on an enzymatic cleavable linker, triggered by disease associated protease. In the selection of carrier systems (i) polyethylene glycol (PEG), a well-studied, Food and Drug Administration approved substance and very common tool to increase the pharmacokinetic properties of therapeutic agents, was chosen as a carrier for non-targeting systems and (ii) Revacept, a human glycoprotein VI antibody, was chosen as a carrier for targeting systems. The protease sensitive cleavable linker was genetically inserted into the N-terminal region of fibroblast growth factor 2 (FGF-2) without jeopardizing protein activity. By exchanging the protease sensitive sequence or the therapeutic payload, both systems represent a promising and adaptable approach for establishing therapeutic systems with bioresponsive release, tailored to pre-existing conditions.
In summary, by site-specific functionalization of various delivery platforms, this thesis establishes an essential cornerstone for promising strategies advancing clinical application. The outlined platforms ensure high flexibility due to exchanging single or multiple elements of the system, individually tailoring them to the respective disease or target site.
Functionally active (conformational) autoantibodies directed against the β1-adrenergic receptor (β1-AR) are supposed to have a pathogenic relevance in human heart failure, particularly in idiopathic dilated cardiomyopathy (DCM). Prevalence of anti-β1-autoantibodies (anti-β1-aabs) in the healthy population is almost negligible, whereas it amounts to up to 30% in heart failure patients with idiopathic DCM. As β1-ARs are not restricted to the heart and are also highly expressed in particular segments of the nephron, it is conceivable that such autoantibodies might also affect kidney function to some extent through the activation of renal β1-ARs.
In the kidney, β1-ARs are highly abundant in the juxtaglomerular apparatus, the distal convoluted tubules, the collecting duct, and the renal arteries. However, the functional significance of β1-ARs at these particular sites along the nephron is poorly understood, as are the effects of conformational stimulating anti-β1-aabs on renal β1-ARs. From the available literature, it is well known that the β1-adrenergic system is involved in, e.g., the regulation of renin-secretion from juxtaglomerular cells. In addition, the β1-adrenergic system is thought to be involved in the regulation of the urine pH via type B-intercalated cells in the collecting duct. In contrast, the regulation of salt- and fluid-secretion in the medullary collecting duct appears to occur independently from the SNS.
As a consequence, the present work aimed to unravel the potential pathophysiological links between renal function, alterations in the cardiovascular system, and circulating agonist-like anti- β1-abs. We analyzed possible renal effects of anti-β1-abs in a human-analogous rat model. After immunization with a GST-fusion protein containing the second extracellular loop (β1-ECII) of the human β1-AR, Lewis-rats develop functionally active, stimulating, conformational anti-β1-ECII-abs. Within the first 6 months, anti-β1-ECII-ab-positive animals develop a hypertensive phenotype, which after 9 months evolves into a DCM phenotype.
In n=40 GST/ β1-ECII-immunized Lewis rats and n=40 age-matched, 0.9% NaCl-injected control animals, we sequentially (i.e. at months 1, 2, 3, 6, 9, 12, 15, and 18 after start of immunization) analyzed the changes in renal function on a molecular, functional, and structural level. We could show that the presence of stimulating anti-β1-ECII-abs – even though having detrimental effects on the heart – has only a minor impact on kidney function and structure. Within the first 3 months after induction of anti-β1-ECII-abs, the levels and activity of renin were significantly increased in immunized compared to corresponding control animals, which was confirmed by experiments on isolated perfused kidneys, in which anti-β1-ECII-abs were able to directly induce the liberation of renin. However, within several weeks the initial anti-β1-ECII-ab-mediated RAAS activation was counter-regulated by auto-regulatory mechanisms activated in the kidney. Similarly, glomerular filtration rate (GFR) and renal blood flow (RBF) were initially decreased in the presence of the stimulating anti-β1-ECII-abs, but returned to control values within 3 months after immunization of the animals. Although expression of several pro-fibrotic markers was significantly up-regulated in anti-β1-ECII-ab-positive rats, no significant differences were noted on a histomorphological level with regard to the occurrence of renal fibrosis, glomerular damage, tubular damage, and perivascular fibrosis. Only a mild decrease in glomerular filtration function was observed in the kidneys of anti-β1-ECII-ab-positive animals from immunization-month 12 on, apparent by increased levels of urinary protein.
Even though anti-β1-ECII-abs were able to induce mild changes in renal function, their effects were not strong enough to critically damage the kidneys in our rat-model. Differences between immunized anti-β1-ECII-ab-positive and corresponding control rats at later time-points (that is, from immunization-month 12 on) are most likely secondary to the progressive heart failure phenotype that immunized animals develop in the course of the experiment.
The present study is the first to focus on the effects of stimulating anti-β1-ECII-abs on the kidney, and on the prevalence of these effects for the heart (referred to as cardio-renal crosstalk). Although our results were obtained in a rat model, they might contribute to better understand the situation in anti-β1-AR-aab-positive human patients. Following the results of our experiments, treatment of such patients should focus on direct and specific neutralization/elimination of stimulating anti-β1-ECII-aab or at least comprise therapeutic strategies that counteract the anti-β1-ECII-aab-effects on the heart by standard treatment for heart failure (i.e. ACE inhibitors, AT1-receptor blockers, and β-blockers) according to current guidelines.
Bispecific T cell engager (BiTE) display a novel design among the class of bispecific antibodies and hold great promise to fight diverse cancers. BiTE molecules consist of two different binding entities derived from two human IgG antibodies connected by a short peptide linker. Their binding arms are directed against the CD3e chain of the T cell receptor on T cells and against an antigen that is specific for (e.g., CD19 for lymphoma in MT103) or over-expressed on (e.g., EpCAM for epithelial cancer in MT110) tumor cells. Without requirement for pre- or co-stimulation, BiTE molecules efficiently redirect CD3+ T cells towards tumor cells expressing the relevant target antigen. Only a BiTE molecule simultaneously bound to both tumor cell and T cell activates the T cell to exert its cytolytic function resulting in tumor cell death. In T cells stimulated with both BiTE and target cells, elevated levels of caspase activation and increased expression of cytotoxic and signaling proteins are observed. These include cytolytic proteins granzyme B and perforin, activation markers CD69 and CD25 and adhesion molecules CD2 and LFA-1. Activated T cells secrete the usual mix of cytokines, among them pro-inflammatory cytokines IFN-g and TNF-a. The membrane of tumor cells expressing the relevant target antigen is perforated during the attack of BiTE-stimulated effector cells as can be concluded from adenylate kinase release from the cytosol of tumor cells. Ca2+-chelator EGTA completely blocked BiTE-mediated activation of caspases and tumor cell lysis. As perforin is strictly Ca2+-dependent, a major role for this pore-forming protein is assumed for the elimination of tumor cells via BiTE-stimulated T cells. Granzyme B and caspases are main players in BiTE-mediated elimination of tumor cells. Inhibitors of granzyme B or caspases reduce or block, respectively the activation of caspases. However, other signals of apoptosis (cleavage of PARP and fragmentation of DNA) were only reduced by granzyme B inhibitor or caspase inhibitor. Most interestingly, the lytic capacity of BiTE molecules was not impaired by granzyme B inhibitor or caspase inhibitor. It seems that there is no requirement for granzyme B and caspases to be present simultaneously. Instead the data presented provide evidence that they can be replaced one at a time by related proteins. Pre-incubation of effector cells with the glucocorticoids dexamethasone or methylprednisolone resulted in markedly decreased secretion of cytokines by T cells yet only a small reduction in the expression of activation markers and adhesion molecules on T cells and specific lysis of tumor cells upon BiTE stimulation. Soluble factors secreted in an undirected manner by BiTE-stimulated T cells do not mediate tumor cell death by themselves. Bystander cells negative for the antigen that is recognized by the BiTE molecule will not be compromised by BiTE activity. The cytokine TGF-b reduced proliferation as well as granzyme B and perforin expression of BiTE-stimulated T cells. Redirected lysis by BiTE-activated T cells was also decreased under the influence of TGF-b, however lysis was still performed at a reasonable rate (72 % of target cells). TGF-b does not exert a deleterious effect on lytic potential of BiTE-stimulated T cells. The minimal anticipated biological effect level for the BiTE MT110 was determined for the entry of MT110 into phase I clinical studies. Experiments analyzing redirected lysis of tumor cells, expression of activation marker CD25 and cytokine release by T cells revealed a MABEL value of 50 pg/ml for MT110.