Refine
Year of publication
- 2015 (9) (remove)
Document Type
- Doctoral Thesis (9) (remove)
Language
- English (9) (remove)
Has Fulltext
- yes (9)
Is part of the Bibliography
- no (9)
Keywords
- Hepatitis B (1)
- anti-CD3 (1)
- anti-CXCL10 (1)
- type 1 diabetes (1)
Institute
- Pharmazie (9) (remove)
In addition to infectious viral particles, hepatitis B virus-replicating cells secrete high amounts of SVPs, which are ssembled by HBsAg in the shape of spheres and filaments but lack any capsid and genome. Filaments are characterized by a much higher amount of the surface protein LHBs as compared to spheres. Spheres are
released via the constitutive secretory pathway, while viral particles are ESCRT-dependently released via MVBs. The interaction of virions with the ESCRT machinery is mediated by α-taxilin that connects the PreS1 domain of LHBs with the ESCRT-component tsg101. Since viral particles and filaments contain a significant amount of LHBs, it is unclear whether filaments are secreted as spheres or released like viral particles. To study the release pathways of HBV filaments in the absence of viral particles, A core-deficient
HBV mutant (1.2×HBVΔCore) was generated by site-directed mutagenesis based on wt1.2x HBV. The start codon of core protein was mutated into stop codon, which was confirmed by DNA sequencing. Data from HBsAg ELISA, Western blot, immunofluorescence microscopy and immunoelectron microscopy showed that the lack of core protein did neither affect the production nor the secretion of HBV SVPs. The intracellular distribution of
LHBs and SHBs showed no difference between wtHBV and the core-deficient mutant expressing cells. Therefore, this system is suitable to investigate the release pathway of HBV filaments in the absence of viral particles. Confocal microscopy analysis of cells cotransfected core-deficient mutants with peYFPRab7 as marker for the endosomal/MVB pathway or with pGalT-eGFP as marker for the trans Golgi apparatus showed that YFP-Rab7, but not GalT-GFP, partially colocalized with LHBs. Furthermore, LHBs could be found in dilated MVBs by immune electron microscopy of ultrathin sections. This was confirmed by isolation of MVBs by cell fractionation using discontinuous sucrose gradient ultracentrifugation and percoll-based linear gradient ultracentrifugation, indicating that filaments enter MVBs in the absence of virion formation. Moreover, inhibition of MVB biogenesis by the small molecular inhibitor U18666A significantly abolished the release of filaments in a dose-dependent manner, but no inhibition could be observed in the production. In contrast, no inhibition on the secretion and production of spheres could be
detected. Inhibition of ESCRT-functionality by coexpression of transdominant negative mutants (Vps4A, Vps4B, CHMP3) abolished the release of filaments while secretion of spheres was not affected. These data indicate that in contrast Abstract 73 to spheres while are secreted via the secretory pathway, filaments are released via ESCRT/MVB pathway like infectious viral particles.
Resistance in glucocorticoid-induced apoptosis is associated with poor prognosis for long term survival in childhood acute lymphoblastic leukemia (ALL). As Smac mimetics have been shown to reactivate apoptosis by antagonizing Inhibitor of Apoptosis (IAP) proteins, we investigate the potential of the Smac mimetic BV6 to overcome glucocorticoid-resistance in ALL. This study shows that BV6 synergistically cooperates with glucocorticoids to trigger apoptosis and to suppress clonogenic growth of pediatric ALL cells. Of note, the BV6/glucocorticoid combination treatment also induces cell death in cells having defects in the apoptotic signaling cascade by inducing a switch from apoptotic to necroptotic cell death. The clinical relevance of our novel combination treatment is underscored by parallel experiments in primary pediatric ALL samples, in which glucocorticoids and BV6 act together to induce cell death in a synergistic manner. Importantly, the addition of BV6 enhances the anti-leukemic effects of glucocorticoids in an in vivo mouse model of pediatric ALL without causing substantial side effects, highlighting the potency of a BV6/glucocorticoid combination treatment. In contrast, BV6 does not increase cytotoxicity of glucocorticoids against several non-malignant cell types of the lympho-hematopoietic system. Furthermore, we have identified the novel underlying mechanism of BV6/glucocorticoid-induced apoptosis by showing that BV6 and glucocorticoids synergistically act together to promote assembly of the ripoptosome, a RIP1/FADD/caspase-8-containing cell death complex. Ripoptosome assembly is critically required for BV6/Dexamethasone-induced cell death, since genetic silencing of its members, i.e. RIP1, reduces ROS production, caspase activation and most importantly cell death induction. BV6/glucocorticoid combination treatment promotes ripoptosome assembly by inhibition of both of its negative regulators, IAP proteins and cFLIP. Thus, we identify that BV6 and glucocorticoids cooperate together to reduce cIAP1, cIAP2 and XIAP protein levels and cFLIP expression. Ripoptosome formation occurs independently of autocrine/paracrine loops of death receptor ligands, since blocking antibodies for TNFα, TRAIL or CD95L or genetic silencing of their corresponding receptors fail to rescue BV6/glucocorticoid-induced cell death. In summary, this study shows that the Smac mimetic BV6 sensitizes for glucocorticoid-induced apoptosis by promoting ripoptosome assembly with important implications for the treatment of childhood ALL.
Acute myeloid leukemia is a hematopoietic stem cell disorder and a type of acute leukemia which is characterized by clonal proliferation of myeloid precursors with a reduced capacity to differentiate into more mature cellular elements. Clinically AML is characterized by a high degree of heterogeneity with respect to chromosome abnormalities, gene mutations, and changes in expression of multiple genes and microRNAs. Cytogenetic abnormalities can be detected in approximately 50% to 60% of newly diagnosed AML patients. Majority of AML cases are associated with chromosomal aberrations, more specifically translocations that often result in gene arrangements and expression of aberrant fusion proteins. This study was carried out with two fusion proteins: PML/RARα and DEK/CAN which results from the translocations t(15;17) and t (6,9) respectively. PML/RARα is the most common translocation (97%) and the main driver in Acute Promyelocytic Leukemia (APL), a wellcharacterized and well treatable subtype of AML. In contrast, DEK/CAN occurs in 1-5% of AML, associated with poor prognosis and defines a high risk group in AML. The expression of PML/RARα results in a fusion protein that acts as a transcriptional repressor by interfering with gene expression programs involved in differentiation, apoptosis, and selfrenewal. Current therapy focused on the targeting of PML/RARα fusion protien. Success has been achieved by using either ATRA, anthracyclines and Arsenic trioxide or their combinations. These agents induce differentiation in PML/RARα positive AML and hence called differentiation therapy. In comparison with ATRA, ATO and anthracyclines are poor cellular differentiation agents. Despite early promise, several studies have reported that differentiation therapy is unable to target/eradicate leukemic stem cells or eradicate the disease. Therefore current therapeutic focus is to eliminate leukemic stem cells and achieve complete molecular remission not only in APL but also in acute lymphoblastic leukemia and chronic myeloid leukemia as well. Key enzymes of the eicosanoid pathways in the arachidonic acid metabolism, such as COX1/2 as well as the 5-LO have been shown to be good targets for leukemic stem cell therapy approach in AML by interfering with the Wntsignaling which is known to be indispensable for the pathogenesis of AML. Recently it was reported that the third eicosanoid pathway based on the cytochrome P450 (CYP) enzymes interferes with Wnt-signaling as well as with the proliferation and mobilization of hematopoietic stem cells...
Das natürlich vorkommende Polyphenol Resveratrol (3,4‘,5-(E)-Trihydroxystilben) ist eine potente chemopräventive Substanz, die in vielen verschiedenen Krebszelllinien wirksam ist. Außerdem verfügt sie über anti-inflammatorische, anti-oxidative und pro-apoptotische Wirkungen. Da Resveratrol auch in Tiermodellen des Typ-2-Diabetes und der nicht-alkoholischen Fettlebererkrankung gute Effekte gezeigt hat, wird in Erwägung gezogen es zur Prävention und Behandlung von metabolischen Erkrankungen einzusetzen. Allerdings liegen, aufgrund von schneller Metabolisierung und geringer Bioverfügbarkeit, die wirksamen Konzentrationen im mikromolaren Bereich. Eine geeignete Strategie, um die anti-tumorale Wirkung und die Bioverfügbarkeit von Resveratrol zu verbessern, scheint die Methylierung der freien Hydroxylgruppen zu sein. Allerdings liefern einige Studien Hinweise darauf, dass diese strukturelle Modifikation der Stilbengrundstruktur zu einer Veränderung des antiproliferativen Wirkmechanismus der methylierten Substanzen führt. Daher führten wir im ersten Teil dieser Arbeit genauere Untersuchungen durch, um die Veränderungen der biologischen Wirkung, die durch die Methylierung der freien Hydroxylgruppen von (E)- und (Z)-Resveratrol verursacht werden, zu charakterisieren. Einen Schwerpunkt bildete die Bestimmung der metabolischen Effekte der methylierten Substanzen. Dabei sollte aufgeklärt werden, ob die Analoga noch immer in der Lage sind bekannte Resveratrol-Targets, wie AMPK, SIRT1 und Phosphodiesterasen, zu modulieren. Zunächst bestätigten wir, dass die methylierten Resveratrolanaloga ST911 (3,4‘,5-Z)-Trimethoxystilben) und ST912 (3,4‘,5-(E)-Trimethoxystilben) einen starken antiproliferativen Effekt auf verschiedene Krebszelllinien ausüben. Wie bereits zuvor beschrieben, konnten wir beobachten, dass ST911 und ST912 das Wachstum von Tumorzellen stärker beeinflussen, als die hydroxylierten Substanzen (E)- und (Z)-Resveratrol. Dies, in Verbindung mit einer vernachlässigbaren zytotoxischen Wirkung und einer deutlich geringeren antiproliferativen Wirkung auf Primärzellen, legt nahe, dass ST911 als potentielles neues Chemotherapeutikum weiter untersucht werden sollte. Zudem zeigten ST911 und ST912 signifikante pro-apoptotische Wirkungen in CaCo-2-Zellen. Auch Resveratrol konnte in diesen Zellen Apoptose auslösen, allerdings erst nach Behandlung mit deutlich höheren Konzentrationen, verglichen mit ST911 und ST912. Eine genauere Charakterisierung der antitumoralen Wirkung von ST911 in HT-29-Zellen zeigte, dass ST911 die Polymerisation von Tubulin zu Mikrotubuli beeinflusst und einen Arrest des Zellzyklus in der Mitose-Phase auslöst. Im Gegensatz dazu führt Resveratrol zu einem Zellzyklus-Arrest in der S-Phase und beeinflusst die Tubulinpolymerisation nicht. Diese Beobachtungen verstärkten die Annahme, dass ST911 ein Mitosehemmer ist und betonten noch einmal die mechanistischen Unterschiede zwischen Resveratrol und den methylierten Analoga. Interessanterweise konnte ST911 die hepatische Fettakkumulation in einem in-vitro-Steatosemodell nicht beeinflussen, während eine Behandlung mit Resveratrol zu einer signifikanten Reduktion der intrahepatischen Triglyzeride führte. Dieses Experiment lässt vermuten, dass die stärkere antiproliferative Wirkung von ST911, keine erhöhte Aktivität in metabolischen Krankheitsmodellen nach sich zieht. Die beobachteten Unterschiede im Steatosemodell führten zu der Frage, ob die methylierten Analoga noch immer in der Lage sind die gleichen metabolischen Targetgene zu modulieren, die in der Literatur für Resveratrol beschrieben sind. Vor kurzem wurden Phosphodiesterasen (PDEs) als direkte Targets von Resveratrol identifiziert. Die Inhibition von PDEs durch Resveratrol führt zu einem Anstieg der intrazellulären cAMP-Konzentration. Diese wiederum aktiviert die bekannten Resveratrol-Targetgene AMPK und SIRT1. Unsere Experimente zeigten, dass ST911 und ST912 keinen Einfluss auf die intrazelluläre cAMP-Konzentration haben. Zusätzlich konnten wir keine AMPK- oder SIRT1-abhängigen Veränderungen der Genexpression beobachten. Dies ist ein Hinweis darauf, dass die Substanzen ihre zellulären Effekte vermutlich nicht über eine Modulation von PDEs, AMPK oder SIRT1 vermitteln. Zusammenfassend liefert der erste Teil der Arbeit Beweise dafür, dass ST911 keine positiven Effekte in metabolischen Krankheitsmodellen ausübt. Dies liegt vermutlich in einem Aktivitätsverlust gegenüber den metabolischen Targetgenen von Resveratrol begründet. Des Weiteren unterstützen unsere Ergebnisse frühere Arbeiten, die zeigen konnten, dass ST911 an Tubulin bindet und die Polymerisation zu Mikrotubuli verhindert. Weiterhin bestätigen unsere Daten, dass die Methylierung von Resveratrol zu einer grundlegenden Veränderung des Wirkmechanismus dieser Substanzen führt, die von einem kompletten Verlust der metabolischen Aktivität begleitet wird. Dies sollte bei zukünftigen Leitstrukturoptimierungen mit Resveratrol berücksichtigt werden. Im ersten Teil dieser Arbeit konnte außerdem gezeigt werden, dass Resveratrol die Gentranskription des nukleären Rezeptors SHP (aus dem Englischen: small heterodimer partner) stark induziert. Der Mechanismus dieser Induktion scheint von der Aktivität von AMPK und SIRT1 abhängig zu sein. Diese Ergebnisse konnten unser Verständnis der vielseitigen biologischen Wirkungen von Resveratrol erweitern. Dennoch sollte die Relevanz der SHP-Induktion für die Effekte von Resveratrol auf metabolische Krankheiten und Tumorwachstum noch weiter untersucht werden. Während der Experimente für den ersten Teil der Arbeit stellten wir fest, dass der AMPK-Inhibitor Compound C (CC) in der Lage war, die wachstumshemmende Wirkung von ST911 signifikant zu reduzieren. Die Untersuchung dieses sogenannten „Rescue-Effektes“ wird durch die Tatsache bestärkt, dass eine steigende Anzahl von Tumoren resistent gegenüber Chemotherapeutika ist. Außerdem fehlen spezifische Antidota für akute Intoxikationen mit Mitosehemmern. Daher zielten die folgenden Experimente darauf ab den Rescue-Effekt näher zu charakterisieren und die zugrundeliegenden Wirkmechanismen aufzuklären. Zunächst zeigten Knockdown-Experimente, dass der Rescue-Effekt unabhängig von der AMPK-inhibierenden Wirkung von CC vermittelt wird. Da CC ein ATP-kompetitiver Inhibitor der AMPK ist und zuvor bereits gezeigt wurde, dass es auch eine große Zahl anderer Kinasen inhibieren kann, vermuteten wir, dass der Rescue-Effekt mit diesen Off-Target-Effekten von CC zusammenhängt. Als nächstes testeten wir, ob die wachstumshemmenden Effekte von anderen Mitosehemmern auch durch CC aufgehoben werden können. Wir wählten verschiedene etablierte Substanzen, die dafür bekannt sind mit Mikrotubuli zu interagieren: Colchicin, das Vinca-Alkaloid Vinblastin, Disorazol A und das aus Taxus-Arten isolierte Paclitaxel. Die ersten drei dieser Substanzen haben eine depolymerisierende Wirkung auf die Mikrotubuli, während Paclitaxel zu einer stärkeren Polymerisierung führt. Zudem binden diese Substanzen an drei verschiedenen Bindestellen am Tubulin. Interessanterweise zeigten unsere Versuche, dass CC die antiproliferative Wirkung aller getesteten Mitosehemmer auf HT-29-Zellen, unabhängig von der Bindestelle, abschwächen kann. Des Weiteren konnte CC die Wirkung der pro-apoptotischen Substanz Staurosporin nicht reduzieren. Diese Ergebnisse weisen darauf hin, dass eher die tubulinbindenden, als die pro-apoptotischen Eigenschaften, von ST911 für den Rescue-Effekt verantwortlich sind. Um zu untersuchen, ob der Rescue-Effekt mit einer kompetitiven Bindung von CC und Mitosehemmern an Mikrotubuli erklärt werden kann, führten wir eine Immunfluoreszenzfärbung von ?-Tubulin durch. Wir konnten beobachten, dass die Tubulinpolymerisation und die Funktion des Spindelapparates in Zellen, die mit Mitosehemmern behandelt wurden, deutlich eingeschränkt waren. Außerdem stellten wir fest, dass CC nicht in der Lage ist die Zerstörung des Tubulingerüstes durch die Mitosehemmer zu verhindern. Eine Einzelbehandlung mit CC hatte keine Wirkung auf die Polymerisation des Tubulin zu Mikrotubuli. Insgesamt legen diese Daten nahe, dass CC nicht direkt an Mikrotubuli binden kann, um mit den Mitosehemmern um eine Bindung zu kompetitieren. Um diese Hypothese zu stärken, führten wir, in Kooperation mit Dr. Jennifer Herrmann (Helmholtz Institut für Pharmazeutische Forschung, Saarbrücken) SPR-Experimente mit Chips durch, auf denen Tubulin immobilisiert wurde. Die Messungen zeigten, das CC nicht in der Lage war gebundenes Disorazol A von der Bindestelle am Tubulin zu verdrängen. Dies zeigte nun deutlich, dass der Rescue-Effekt nicht auf einer Kompetition von CC und Mitosehemmern um Tubulinbindestellen beruht. Zellzyklusanalysen zeigten, dass die kombinierte Behandlung mit ST911 und CC zu einer Abschwächung des durch ST911 verursachten G2/M-Arrestes führt. Da wir zuvor bereits eine Beeinflussung der direkten Targets von CC und Mitosehemmern, AMPK oder Tubulin, ausgeschlossen hatten, schlussfolgerten wir, dass CC vermutlich mit anderen zellulären Signalwegen interagiert, die zu den beschriebenen Veränderungen des Zellwachstums und der Zellzyklusprogression führen. Eine Literaturrecherche ergab, dass ein erhöhter intrazellulärer Polyaminspiegel, die Aktivierung des PI3K/Akt-Signalweges oder eine erhöhte Aktivität des Transkriptionsfaktors c-Myc zu einer Abschwächung eines G2/M-Arrestes führen können. Daher fokussierten wir die weiteren Experimente auf die Untersuchung einer möglichen Beteiligung dieser Targets an der Vermittlung des Rescue-Effektes. Wir zeigten, dass CC die Expression der Spermidin/Spermin-N1-Acetyltransferase (SSAT) erhöhen kann. Die SSAT ist ein Enzym, das an der Biosynthese der Polyamine beteiligt ist. Zusätzlich beobachteten wir, dass die Behandlung mit CC nach 4 h zu einer Erhöhung von phosphoryliertem und damit aktiviertem Akt (pAkt) führt. Die zusätzliche Behandlung mit Wortmannin, einer Substanz, welche die Phosphorylierung von Akt hemmen kann, führte zu einer Abschwächung des Rescue-Effektes. Insgesamt weisen diese Ergebnisse darauf hin, dass eine Aktivierung von Akt-Signalwegen und ein Einfluss auf die Polyaminbiosynthese, zumindest teilweise, mit dem Rescue-Effekt zusammenhängen können. Die Überexpression von c-Myc, einem Transkriptionsfaktor, der eng mit dem Akt-Signalweg und der Biosynthese von Polyaminen zusammenhängt, ist oft mit einer erhöhten Zellproliferation verbunden. Wir untersuchten die zellulären Proteinmengen von c-Myc mittels Western Blot und entdeckten, dass nach der Behandlung mit Mitosehemmern zusätzliche Banden für c-Myc auf den Blots auftauchten. Diese Ergebnisse geben einen Hinweis auf eine posttranslationale Modifikation von c-Myc nach der Behandlung mit Mitosehemmern. Durch Kombination mit CC wurden die zusätzlichen Banden abgeschwächt und die Gesamtmenge an c-Myc-Protein nahm nach längeren Inkubationszeiten rapide ab. Dies legt nahe, dass die posttranslationale Modifikation von c-Myc zum Abbau des Proteins führt und, dass CC dies abschwächen kann. Verschiedene Arbeiten zeigten bereits, dass c-Myc phosphoryliert wird und nach Konjugation mit Ubiquitin vom Proteasom abgebaut wird. Daher überprüften wir, ob eine Inhibition des Proteasoms mit MG-132 zu einem ähnlichen Rescue-Effekt führt wie mit CC. Tatsächlich führte die Behandlung mit ST911 in Kombination mit MG-132 zu einer Zunahme der Zellproliferation, wie sie vorher bereits für CC beobachtet wurde. Dies bestärkte die Theorie, dass der proteasomale Abbau von c-Myc eine Rolle beim Rescue-Effekt spielen kann. Als nächstes untersuchten wir die Phosphorylierungen von c-Myc am Ser62 und Thr58. Diese Phosphorylierungen spielen eine wichtige Rolle beim Abbau von c-Myc, indem Sie das Protein für die Konjugation mit Ubiquitin markieren. Die densitometrische Auswertung der Western Blots ergab, dass die Behandlung mit ST911 initial zu einem Anstieg von phospho-c-Myc führt, dem eine schnelle Abnahme zu späteren Zeitpunkten folgt. Außerdem konnte gezeigt werden, dass dieser Anstieg von phospho-c-Myc durch Kombination mit CC reduziert wurde. Dies unterstützt die Hypothese, dass ST911 den proteasomalen Abbau von c-Myc begünstigt und CC dies verhindern kann. Dies ist eine mögliche Erklärung für die erhöhte Zellproliferation, die für die durch CC „geretteten“ Zellen beobachtet wurde. Allerdings konnte das direkte Target, das für die Vermittlung des Rescue-Effektes durch CC verantwortlich ist, bisher nicht identifiziert werden. DYRKs (aus dem Englischen: Dual-specificity tyrosine-phosphorylation-regulated kinases) sind wichtige Regulatoren von Proteinstabilität und –abbau während der Zellzyklusprogression. Vor kurzem wurde gezeigt, dass DYRK1A und DYRK2 c-Myc am Ser62 phosphorylieren können und es dadurch für den proteasomalen Abbau markieren. Interessanterweise wurde CC bereits in einer früheren Publikation als potenter Inhibitor verschiedener DYRKs beschrieben. Allerdings wurde die Hemmung der DYRKs durch CC in diesem Artikel nur in einer einzelnen Konzentration getestet. Daher bestimmten wir in einem in-vitro-Kinaseassay in Kooperation mit Dr. Matthias Engel (Universität des Saarlandes, Saarbrücken) die IC50-Werte für CC gegenüber DYRK1A, DYRK1B und DYRK2. Unsere Ergebnisse zeigten deutlich, dass CC ein bevorzugter Inhibitor von DYRK1A und DYRK1B (IC50-Wert von etwa 1 µM) ist, aber auch DYRK2 hemmen kann (IC50-Wert von etwa 5 µM). Da sich die vermutete Bindestelle von CC in der stark konservierten Kinasedomäne befindet, ist eine unspezifische Inhibition verschiedener DYRKs nicht überraschend. Genexpressionsanalysen zeigten, dass HT-29 und HepG2 vergleichbare Mengen an DYRK1A exprimieren, während DYRK1B und DYRK2 deutlich weniger in HepG2 vorhanden sind. Vorige Experimente hatten gezeigt, dass HepG2 weniger sensitiv für ST911 und den durch CC vermittelten Rescue-Effekt waren. Wir schlussfolgerten, dass die unterschiedliche Expression der DYRK-Formen eine mögliche Erklärung für diese Unterschiede sein könnte. Daher entschieden wir uns für eine nähere Untersuchung von DRK1B und DYRK2. Experimente mit verschiedenen Inhibitoren der DYRKs zeigten, dass diese Substanzen, ähnlich wie CC, in der Lage waren die antiproliferative Wirkung von ST911 abzuschwächen. Diese Ergebnisse wurden in nachfolgenden Knockdown-Experimenten bestätigt. Dies legt nahe, dass die DYRKs zumindest teilweise für die Vermittlung des Rescue-Effektes verantwortlich sind. Zusammenfassend man kann sagen, dass der Rescue-Effekt vermutlich mit der Biosynthese von Polyaminen, dem Akt-Signalweg und dem proteasomalen Abbau von c-Myc zusammenhängt. Des Weiteren scheint die direkte Inhibition von DYRKs durch CC ein vielversprechender Ansatz für die Erklärung des Effektes zu sein. Allerdings konnte in keinem der Experimente eine kompletten Aufhebung des Rescue-Effektes durch CC gezeigt werden. Daher gehen wir davon aus, dass verschiedene Targets in die Vermittlung des Rescue-Effektes involviert sind. Dies ist höchstwahrscheinlich auf eine unspezifische, ATP-kompetitive Hemmung verschiedener Kinasen durch CC zurückzuführen. Nichtsdestotrotz, sind eine nähere Untersuchung von DYRKs im Rahmen der Therapieresistenz von Tumoren und eine genauere Aufklärung der am Rescue-Effekt beteiligten Signalwege eine interessantes Feld für weitere Untersuchungen.
The endocannabinoids (EC), their synthetizing and metabolizing enzymes, and the cannabinoid (CB) receptors comprise the endocannabinoid system (ECS) that has been detected by Yasuo et al. (2010) in rodent and human brain areas essential for circadian rhythmic control and hormone secretion. The EC are secreted in the pars tuberalis formation (PT) of the pituitary gland and unfold their effect as ligands on cannabinoid receptors type 1 (CB1) in the pars distalis (PD). The CB1 is mostly expressed on folliculo-stellate (fs) cells of the PD. The fs cells execute regulative and supportive functions to adjacent hormone-producing cells (Allaerts and Venkelecom, 2005; Mitsuishi et al., 2013). The lipid and calcium binding protein Annexin A1 (Anx A1) and the cell membrane permeable compound nitric oxide (NO) have been detected in the fs cells (Woods et al., 1990; Devnath and Inoue, 2008). There are published findings indicating strong influence of Anx A1 and NO on hormone production (Taylor et al. 1993; Venkelecom et al, 1997). The hypothesis of this study is that the EC influence hormonal secretion by acting upon CB1 receptors on fs cells and thus activating or inhibiting Anx A1 and NO that directly affect adjacent glandular cells.
Prevalently cell models were used to carry out the experimental work. The TtT/GF and Tpit/F1 cell lines represent the fs cells, the AtT20/D16v stand for the ACTH-producing corticotroph (C) cells, and GH4C1 for the PRL-producing lactotroph (L) cells. Whenever comparison with an integrity model was possible tissue from C3H mice was used. Chemoluminescent and photometrical detection, enzyme-linked immunosorbent assay (ELISA), fluorescence-activated cell sorting (FACS), immunoblot (IB), immunocyto- and immunohisto-chemical analysis (ICC, IHC), in situ hybridization (ISH), and (q) PCR methods were used as assaying tools to investigate CB1, Anx A1, the Anx A1 receptor - Fpr-rs1, NO, ACTH, and PRL.
CB1 was detected on the fs, C, and L cell models. The presence of fatty acid amide hydrolase (FAAH, an EC degrading enzyme) was confirmed in the fs cells. Incubations of the fs cells with CB1 agonists (2-AG, AEA, WIN) and antagonist (otenabant) were performed and resulting increase of Anx A1, and inhibition of NO were detected. Anx A1 binding sites, known as formyl peptide like receptor – related sequence 1 (Fpr-rs1) were identified on the C and L cells. The hormone-producing cells were treated with a 2-AG, Anx A1, and NO and the resulting changes in the levels of ACTH and PRL were detected. Anx A1 acted stimulatory on ACTH in the C AtT20/D16v cell and inhibitory on PRL in the L GH4C1 cell. NO inhibited both ACTH and PRL release. Additional analysis of the levels of expression of mRNA for Anx A1 and Fpr-rs1 in murine PD tissue demonstrated that while the expression of the first was not influenced by time, the expression of the latter was activated during the subjective day.
The here presented study shows that EC influence the ACTH release stimulatory through activating Anx A1 and inhibiting NO. As for PRL, the EC unfold an inhibition through activating Anx A1, and stimulation through inhibiting NO. A clear regulatory linkeage between the EC and ACTH and PRL control is revealed, involving the fs cells with possible time-dependence.
Hematopoietic stem cells (HSCs) have the unique abilities of life-long self-renewal and multi-lineage differentiation. They are routinely used in BM or stem cell transplantations to reconstitute the blood system of patients suffering from malignant or monogenic blood disorders. For an adequate production of each blood cell lineage in homeostasis and under stress conditions, the fate choice of HSCs to either self-renew or to differentiate must be strictly controlled. The incomplete understanding of the molecular mechanisms that control this balance makes it still impossible to maintain or expand undifferentiated HSCs in culture for advanced regenerative medical purposes.
The aim of this thesis was the identification and molecular characterisation of mechanisms that control the decision of HSCs to self-renew or to differentiate, and how they are connected to extrinsic cytokine signaling control. Prior to this thesis, a screening for genes upregulated under self-renewal promoting thrombopoietin (TPO) signaling via the transcription factors STAT5A/B in HSCs was conducted, and Growth arrest and DNA damage inducible 45 gamma (Gadd45g) was one of the regulated genes. GADD45G was described as stress sensor, DNA-damage response and tumor suppressor gene, that is epigenetically silenced in many solid tumors and leukemia. Furthermore, Gadd45g is upregulated in aged HSCs with impaired multi-lineage reconstitution abilities, and it is induced by differentiation promoting cytokines in GM-committed cells. However, the function of GADD45G in LT-HSCs was unknown. All these points warrant further investigation to unravel the function of GADD45G on early cell fate decisions of HSCs in hematopoiesis.
The expression of Gadd45g was stimulated by hematopoietic cytokines TPO, IL3 and IL6 both in HSCs and MPPs, making GADD45G an interesting target to focus on. To simulate the cytokine-induced expression GADD45G was lentivirally transduced in HSCs. Surprisingly, GADD45G did not induce cell cycle arrest or cell death in hematopoietic cells neither in vitro nor in vivo, as reported in many cell lines. Instead GADD45G revealed an enhanced and markedly accelerated differentiation of HSCs into mainly myelomonocytic cells, similar as observed for IL3 and IL6 containing cultures. Also in vivo, GADD45G rapidly initiates the differentiation program in HSCs at the expense of self-renewal and long-term engraftment, as shown by serial HSC transplantation experiments. Along the same line, HSCs from Gadd45g-knock out mice exhibited an increased self-renewal. In vitro, Gadd45g-/- progenitors showed higher and prolonged colony formation potential and slower expansion after cytokine stimulation. The loss of Gadd45g increased HSC self-renewal and improved repopulation in secondary recipients, determined by serial competitive transplantations. Taken together, GADD45G could be identified as molecular link between differentiation-promoting cytokine signaling and rapid differentiation induction in murine LT-HSCs.
As presented in this thesis the differentiation induction of GADD45G was mediated by the activation of the cascade of MAP3K4 – MKK6 –p38 MAPK. Small molecule inhibition of p38, but not JNK, blocked the GADD45G-induced differentiation. GADD45G binds to MAP3K4 and releases its auto-inhibitory loop by a change in confirmation, initiating this cascade. Phosphoflow cytometry demonstrated the activation of p38 and a downstream kinase MK2 by GADD45G expression in MPPs. Furthermore, the expression of constitutive active MAP3K4 and MKK6 were able to phenocopy GADD45G-induced differentiation, which could be blocked by p38 inhibition.
The other two family members GADD45A and B also induced accelerated differentiation in LT-HSCs. Interestingly, only GADD45G suppressed the differentiation into megakaryocyte and erythrocyte (Mek/E) lineage cells suggesting a role of GADD45G in lineage choice. Long-term time-lapse microscopy-based cell tracking of single LT-HSCs and their progeny revealed that, once GADD45G is expressed, the development of LT-HSCs into granulocyte-macrophage-committed progeny occurred within 36 hours, and uncovered a selective lineage choice with a severe reduction in Mek/E cells. Furthermore, no megakaryocytic-erythroid progenitors (MEPs) could develop from HSPCs in BM 2 weeks after transplantation suggesting a very early selection against Mek/E cell fates. In line with these findings, GADD45G-transduced MEPs could not expand or form colonies in vitro, demonstrating that the differentiation program induced by GADD45G is not compatible with Mek/E lineage fate. Gene expression profiling of HSCs indicated that GADD45G promotes myelomonocytic differentiation programs over programs for self-renewal or megakaryo-/ erythropoiesis. The here identified differentiation induction potential of GADD45G is so strong that the expression of GADD45G in primary acute myeloid leukemia (AML) cells inhibited their expansion accompanied by enhanced differentiation and increased apoptosis.
The here presented work shows that IL3 and IL6 induce a differentiation program in HSCs via GADD45G and p38 closing the link of extrinsic cytokine signaling and differentiation induction. Since the loss of Gadd45g increased the self-renewal and slowed HSC differentiation, this may be utilized, i.e. by p38 inhibition, to ex vivo maintain and expand HSCs by preventing cytokine-induced differentiation. Furthermore, Re-expression of GADD45G may overcome the differentiation block in leukemia to eliminate these cells by driving them into terminal differentiation and apoptosis.
Life-saving pig-to-human xenotransplantation is a promising technology with the potential to balance the shortage of human organs in allotransplantation. Before this approach is applied on solid vascularized organs, several barriers must be overcome. Patient safety is menaced by infectious porcine endogenous retroviruses (PERV) which are able to infect human cell lines in vitro. Successful infection with PERV is associated with diverse life-threatening consequences including gene disruption, tumorigenicity, immune suppression as well as PERV proliferation throughout the whole human body. This could cause a catastrophic xenozonoosis leading to the emergence of new forms of pathogens and pandemic diseases similar to AIDS. However, in vivo, there is hitherto no incidence of any infection with PERV in preclinical xenotransplantations performed in the past.
PERV infection of human peripheral blood mononuclear cells (huPBMC) is a critical issue discussed controversially in several studies. It is essential to address the sensitivity of huPBMC to infection by PERV since it is generally one of the first retroviral targets upon viral invasion and infection of the human body. To assess definitely if huPBMC are infected productively by PERV, target cells were challenged with the highest infectious PERV class, recombinant PERV-A/C, in different assays. Modern and standard methods to detect PERV at different stages of viral cycles were used to monitor PERV development upon contact with host cells. Indeed, PERV-A/C in supernatants of producer cell lines failed to infect mitogen-activated huPBMC. Neither retroviral reverse transcriptase (RT) nor viral RNA packaged in virus particles were observed in supernatants of cells exposed to viral supernatants. In addition, provirus was not detected in huPBMC until 56 days p. i. with PERV-A/C. Independently of the virus load applied, culture conditions of huPBMC or administration of polybrene as enhancer, PERV was unable to infect huPBMC. Results suggest that PERV in supernatants lack sufficient infectious potential to be productively generated in huPBMC.
In order to approximate xenotransplantation scenarios, different PERV producing cells including PHA-activated porcine PBMC (poPBMC) were adopted as virus source in co-cultivation studies with huPBMC. In this case, expression of viral RNA was successfully measured. However, RT activity did not increase until 28 days p. e. with PERV producer cells which indicates that viral particles devoid of infectious capacity were released from non-productively infected cells.
On the other hand, co-cultivation of both virus producer and virus recipients increases the contact pressure between PERV and target cells. Consequently, PERV was able to be detected at least as provirus in huPBMC. Although virions produced were not functional, presence of provirus in infected cells will sooner or later provoke expression of provirus. This could lead to chromosomal rearrangements as well as virus reinfection and insertional mutagenesis.
Ecotropic PERV-C displays a restricted host range to porcine cells. Given its ability to serve as template to form recombinant xenotropic PERV-A/C, PERV-C represents a potent hazard in the course of xenotransplantation. Thus, isolation and functional characterization of PERV-C in the genome of pigs in use and intended for xenotransplantation is necessary to analyze the genetics of these virions as well as to select animals lacking proviral PERV-C or to generate transgenic PERV-C negative donors.
PERV-C was isolated from the genome of a female SLAd/d haplotype pig via screening of a bacteriophage library which was constructed from the genomic DNA of poPBMC extracted from this PERV non-transmitting sow. Upon genetic complementation of provirus using a PCR fragment infectious ability of full-length PERV-C clones was investigated in cell culture. PERV-C clones were successfully reproduced in susceptible porcine cells as RT activity as well as viral RNA were detected in supernatants of infected cells 56 days p. i. Furthermore, presence of proviruses in challenged cells was confirmed by nested PCR.
PERV-C clones were also isolated from a bacteriophage library generated on genomic DNA of an Auckland island pig of the DPF colony, whose individuals display a PERV-null phenotype and are already in use for xenotransplantation, and of a Göttingen minipig, whose relatives serve as animal models to study human diseases. In contrast to PERV clones isolated from the female SLAd/d haplotype sow PERV-C clones of the Auckland island pig as well as of the Göttingen minipig were not functional and therefore unable to infect target cells. This confirms the PERV-null phenotype which renders these animals putative candidates as donors in xenotransplantation. On the other hand, presence of functional PERV-C in SLAd/d haplotype pigs exerts a negative impact on patient safety in xenotransplantation. The suitability of these animals as potent organ donors should be intensively investigated.
In conclusion, PERV of all classes pose a virological risk in xenotransplantation which should not be ignored. Since exclusion of all PERV from donor herds is impossible, generation of transgenic humanized animals lacking genomic infectious PERV represents the best strategy to guarantee patient safety in future life-saving pig-to-human xenotransplantation.
Type 1 Diabetes (T1D) is an autoimmune disorder in which the own immune system attacks the insulin producing _-cells in the pancreas. Therapy of T1D with anti-CD3 antibodies (aCD3) leads to a blockade of the autoimmune process in animal models and patients resulting in reduced insulin need. Unfortunately, this effect is only temporal and the insulin need increases after a few years. In the first approach, I aimed at a blockade of the cellular re-entry into the islets of Langerhans after aCD3 treatment by neutralising the key chemokine CXCL10, which is important for the T cell migration. In the second approach I tried to block the transmigration of leukocytes trough the endothelial layer into inflamed tissue with an anti-JAM-C antibody (aJAM-C) after aCD3 treatment.
I used the well-established RIP-LCMV-GP mouse model of T1D. As target autoantigen in the _-cells, such mice express the glycoprotein (GP) of the lymphocytic choriomeningitis virus (LCMV) under control of the rat insulin promoter (RIP). These mice develop T1D within 10 to 14 days only after LCMV-infection. In the combination therapy (CT) I treated diabetic RIP-LCMV-GP mice with 3 5g aCD3 per mouse (3 injections in 3 days) followed by administration of a neutralising anti-CXCL10 (CT) or aJAM-C (CT-J) monoclonal antibody (8 injections of 100 5g per mouse over 2.5 weeks).
CT reverted T1D in RIP-LCMV-GP mice significantly (CT: 67 % reversion; control: 16 % reversion) and with superior efficacy to monotherapies with aCD3 (38 % reversion) and aCXCL10 (36 % reversion).
The CD8 T cells in the spleen have fully regenerated at day 31 after infection. However, the frequency of islet antigen (GP)-specific CD8 T-cells was significantly reduced by 73 % in the spleen after CT compared to isotype control treated mice. In contrast, in aCD3 treated mice the T cells were only reduced by 56 % of the frequency of isotype control treated mice. Flow cytometry and immunohistological examinations demonstrated a marked reduction of CD8 T cells in the pancreas of CT treated mice. Importantly, the number of GP-specific CD8 T cells was reduced dramatically by 78 % in the pancreas of CT treated mice, whereas aCD3 treatment led to a less pronounced reduction of the GP-specific CD8 T cell number (23 %). This reduction of infiltration was long lasting since in the pancreas of CT treated mice the _-cells produce insulin and there were almost no infiltrating T cells present at day 182 post-infection. aCD3 treated mice also showed many insulin producing cells after 182 days post-infection. Nevertheless, their pancreas displayed also some infiltrates around the islets.
In order to confirm my data I treated non-obese diabetic (NOD) mice with CT. In contrast to RIP-LCMV-GP mice, NOD mice develop spontaneous T1D within 15 to 30 weeks after birth, due to a mutation in the CTLA-4 gene. Strikingly CT cured 55 % of diabetic NOD mice, whereas only 30 % showed T1D reversion with aCD3 alone and none reverted after isotype control administration.
The impact of CT on GP-specific T cells (Teff) was stronger in the RIP LCMV-GP than in the NOD model. In contrast, regulatory T cells (Tregs) were induced predominantly in NOD mice rather than in RIP-LCMV-GP mice. However, looking at the Treg/Teff ratio and compared to isotype control antibody treated mice, I found a significant 4-fold increase in the pancreas of CT treated RIP LCMV-GP mice and a 17-fold increase in the PDLN of CT treated NOD mice. In addition, a tendency for an increase in Treg/Teff ratio was obtained in the spleen of CT-treated RIP LCMV-GP as well as NOD mice compared to aCD3 and isotype control antibody treated mice.
In the second combination therapy with neutralising aJAM-C, CT-J (51 % reversion) slightly improved the aCD3 therapy (41 % reversion). However, there was no significant difference between CT-J and aCD3 administration in terms of total CD8 and GP-specific CD8 T cells.
JAM-C also interacts with the integrin receptor macrophage-1 antigen (MAC-1), which is among others expressed by neutrophils. Accordingly, JAM-C could be involved in neutrophil transmigration to the pancreas. Indeed, I found a significant reduction for the infiltrating neutrophils into the pancreas of mice after CT-J compared to aCD3 monotherapy.
In summary the addition of aJAM-C to aCD3 monotherapy showed a small improvement, which was associated with a reduced neutrophil migration into the pancreas. However, JAM C seemed to play only a minor role in T1D development and some other adhesion molecules might be more important. Nevertheless, the combination of aCD3 and aCXCL10 resulted in a significant and long lasting reduction of aggressive T cells in the pancreas in two independent mouse models. Furthermore a protective immune balance was obtained. Since both antibodies are available for as well as tested in humans and the therapy is only for a short period of time after disease onset, this combination therapy might kick-start a novel therapy for T1D.
HDAC inhibitors (HDACI), a new class of anticancer agents, induce apoptosis in many cancer entities. JNJ-26481585 is a second generation class І HDACI that displays improved efficacy in preclinical studies compared to the established HDACI SAHA (Vorinostat). Therefore, this study aims at evaluating the effects of JNJ-26481585 on human rhabdomyosarcoma (RMS) and at identifying novel synergistic interactions of JNJ-26481585 or the more common HDACI SAHA with different anticancer drugs in RMS cells. Indeed, we show that JNJ-26481585 and SAHA significantly increase chemotherapeutic drug-induced apoptosis in embryonal and alveolar RMS cell lines, when used in combination with chemotherapeutic agents (i.e. doxorubicin, etoposide, vincristine, and cyclophosphamide) which are currently used in the clinic for the treatment of RMS.
We demonstrate that JNJ-26481585 as single agent and in combination with doxorubicin induces apoptosis, which is characterized by activation of the caspase cascade, PARP cleavage, and DNA fragmentation. Induction of caspase-dependent apoptotic cell death is confirmed by the use of the broad-range caspase inhibitor zVAD.fmk, which significantly decreases both JNJ-26481585-triggered and combination treatment-mediated DNA fragmentation, and in addition completely abrogates loss of cell viability. Importantly, JNJ-26481585 significantly inhibits tumor growth in vivo in two preclinical RMS models, i.e. the chicken chorioallantoic membrane (CAM) model and a xenograft mouse model, supporting the notion that JNJ-26481585 hampers tumor maintenance. Also, in combination with doxorubicin JNJ-26481585 significantly reduces tumor growth in in vivo experiments using the CAM model.
Mechanistically, we identify that JNJ-26481585-induced apoptosis is mediated via the intrinsic apoptotic pathway, since we observe increased loss of mitochondrial membrane potential and activation of the proapoptotic Bcl-2 family members Bax and Bak. Interestingly, we find that JNJ-26481585 triggers induction of Bim, Bmf, Puma, and Noxa on mRNA level as well as on protein level, pointing to an altered transcription of BH3-only proteins as important event for the Bax/Bak-mediated loss of mitochondrial membrane potential as well as mitochondrial apoptosis induction upon JNJ-26481585 treatment. JNJ-26481585-initiated activation of Bax and Bak is not prevented with the addition of zVAD.fmk, suggesting that JNJ-26481585 first disrupts the mitochondria and subsequently activates the caspase cascade. When JNJ-26481585 is used in combination with doxorubicin, we observe not only an increase of proapoptotic Bcl-2 proteins, but also a decrease in the level of the antiapoptotic mitochondrial proteins Bcl-2, Mcl-1, and Bcl-xL. This indicates that Bax, Bak, Bim, and Noxa are crucial for JNJ-26481585-induced as well as JNJ/Dox treatment-induced apoptosis, since RNAi mediated silencing of Bax, Bak, Bim, and Noxa significantly impedes DNA fragmentation upon those treatments.
Furthermore, ectopic overexpression of Bcl-2 profoundly impairs both JNJ-26481585 and combination treatment-mediated apoptosis, abrogates caspase cleavage, and reduces activation of Bax and Bak, underlining the hypothesis that JNJ-26481585 initially targets the mitochondria and then activates caspases.
With the more commonly used HDACI SAHA we confirm the results obtained with the HDACI JNJ-26481585, since combination treatment with SAHA and doxorubicin also induces intrinsic apoptosis, which can be significantly diminished by zVAD.fmk or ectopic overexpression of Bcl-2. Treatment with SAHA and doxorubicin also affects expression levels of pro- and antiapoptotic mitochondrial proteins, thus shifting the balance towards the proapoptotic mitochondrial machinery, resulting in Bax/Bak activation, caspase activation, and subsequently apoptosis.
Taken together, we provide evidence that the HDACIs JNJ-26481585 and SAHA are promising therapeutic agents for the treatment of RMS and that combination regimens with HDACIs represent an efficient strategy to prime RMS cells for chemotherapy-induced apoptosis. These findings have important implications for mitochondrial apoptosis-targeted therapies of RMS.