Die Rolle von Ephrin-B2 in der Invasion maligner Gliome

Das Glioblastoma multiforme zählt zu den häufigsten glialen Neoplasien des Menschen und weist zudem unter den Gliomen die höchste Malignität auf. Glioblastompatienten haben trotz aggressiver therapeutischer Ansätze eine mittlere Überlebenszeit von weniger als einem Jahr. Die diffuse Invasion in das umliegende Hirngewebe ist einer der Hauptgründe für die Rezidivbildung und die infauste Prognose von Glioblastompatienten. Neuere Untersuchungen lassen vermuten, dass die starke Invasion auch einer der Gründe für die beobachtete anti-angiogene Resistenz bei der Behandlung von Glioblastomen ist. Das bidirektionale EphB/Ephrin-B-System wurde bei der axonalen Wegfindung als Vermittler repulsiver Signale identifiziert und auch im Zusammenhang der Migration und Invasion von Zellen überprüft. In der vorliegenden Arbeit sollte daher die Funktion der bidirektionalen Eph- und Ephrin-Signaltransduktion in Bezug auf die Glioblastominvasion und Progression untersucht werden. rn Genetische und epigenetische Untersuchungen der EphB/Ephrin-B-Familie in einer Kohorte von Gliompatienten unterschiedlicher Malignitätsgrade identifizierten Ephrin-B2 als mögliches Tumorsuppressorgen. In Übereinstimmung damit führte die Inaktivierung von Ephrin-B2 in einem murinen Gliommodell zu einer verstärkten Invasion und einem erhöhtem Tumorwachstum in vivo. Dies konnte in verschiedenen Invasion-Assays in vitro bestätigt werden. Weiterhin zeigten unsere Untersuchungen, dass Ephrin-B2 transkriptionell durch das hypoxische Mikromilieu HIF-1α-vermittelt reprimiert wird. Da HIF-1α als transkriptioneller Aktivator Ephrin-B2 nicht direkt reprimieren kann, wurden potentielle HIF-1α-regulierte Repressoren untersucht, die für die Ephrin-B2 Herunterregulation verantwortlich sein könnten. Dabei wurde anhand von Ephrin-B2-Promotoranalysen und ChIP-Assays ZEB2 als HIF-1α-induzierbarer Repressor von Ephrin-B2 identifiziert. Zur Bestätigung der Hypothese, dass ZEB2 ein wichtiger Regulator der Tumorinvasion ist, wurden humane ZEB2-Knockdown-Glioblastomzellen generiert und in vitro sowie in vivo untersucht. Im Hinblick auf mögliche therapeutische Anwendungen wurden die ZEB2-Knockdown-Glioblastomzellen zusätzlich im Zusammenhang anti-Angiogenese-induzierter Invasion analysiert. Der Verlust von ZEB2 führte dabei zu einer verringerten Glioblastominvasion und Progression in einem Maus-Xenograft Modell. Die Behandlung der Tumoren mit dem anti-VEGF-Antikörper Avastin resultierte in einer stark erhöhten Invasion, die durch die Inaktivierung von ZEB2 und der dadurch reaktivierten repulsiven Signale von Ephrin-B2 wieder aufgehoben werden konnte. Zusammenfassend konnte in der vorliegenden Arbeit erstmals gezeigt werden, dass Ephrin-B2 als Tumorsuppressor in Gliomen agiert und durch verschiedene Mechanismen wie der genetischen und epigenetischen Kontrolle, aber auch der HIF-1α-vermittelten, ZEB2-abhängigen Repression inaktiviert wird. Dies resultiert in einer Blockade repulsiver Signale, so dass Tumorzellen diffus in das Parenchym und zu den Blutgefäßen migrieren können. Der in dieser Arbeit neu identifizierte Signalweg stellt ein attraktives therapeutisches Ziel zur Inhibition der Tumorzellinvasion dar und ermöglicht darüber hinaus der Ausbildung von Resistenzen gegenüber anti-angiogener Behandlung entgegenzuwirken. rn

Molekulare Charakterisierung der Il-9-Genexpression in T-Helfer-Typ-9-Zellen

CD4+ T-Zellen können in verschiedene T-Helferzellsubpopulationen differenzieren. Dabei hängt es von verschiedensten Milieubedingungen ab, welche Subpopulation sich ausprägt, damit die CD4+ T-Zelle durch die Sekretion verschiedenster Zytokine ihre Funktion im Immunsystem wahrnehmen kann.rnBei der Th9-Subpopulation handelt es sich um einen IL-9-produzierenden Phänotyp, welcher sich in der Anwesenheit von TGF-ß und IL-4 entwickelt39. Als treibender Transkriptionsfaktor für diese Subpopulation wurde das Protein IRF4 beschrieben45. Da dieser Transkriptionsfaktor auch für die Differenzierung weiterer Subpopulationen, wie Th2- und Th17-Zellen von Bedeutung ist30,121, stellte sich die Frage, welcher Interaktionspartner von IRF4 darüber entscheidet, welcher Subtyp sich entwickelt. Deshalb wurde in dieser Arbeit der Transkriptionsfaktor NFATc2 als möglicher Interaktionspartner für IRF4 am murinen Il9 Promotor untersucht. Allerdings zeigten Reportergen¬analysen, dass NFATc2 die IL-9-Produktion in Th9-Zellen inhibiert anstatt sie zu fördern. Th9-Zellen aus NFATc2-defizienten Tieren zeigen folglich im Vergleich zu wildtypischen Th9-Zellen sowohl nach Primär- als auch nach Restimulation eine verstärkte IL-9-Produktion. Der Faktor NFATc2 kann somit als transkriptioneller Aktivator für die IL-9-Expression in Th9-Zellen ausgeschlossen werden. In vivo wurden diese Beobachtungen dadurch untermauert, dass NFATc2-defiziente Tiere im Rahmen des Asthma bronchiale zu einer verstärkten pulmonalen Inflammation neigen und auch einen erhöhten Atemwegswiderstand nach Methacholin-Provokation aufweisen. Diese asthmatischen Symptome konnten durch Applikation eines neutralisierenden Antikörpers für IL-9 wesentlich gemildert werden. In einem B16F10-Melanommodell konnten NFATc2-defiziente Tiere gegenüber dem Wildtyp eine verbesserte anti-Tumorantwort ausprägen. Nach Gabe eines IL-9-neutralisierenden Antikörpers, wurde dieser Effekt wiederum gemildert.rnZusammenfassend lässt sich sagen, dass IRF4 nicht mit NFATc2 am murinen Il9 Promotor interagiert, um die IL-9-Expression in Th9-Zellen zu fördern. Eine NFATc2-Defizienz resultiert sogar in einer gesteigerten IL-9-Produktion, womit ein inhibitorischer Einfluss von NFATc2 in Bezug auf die IL-9-Expression in Th9-Zellen nachgewiesen werden konnte.rn

Differential cell type-specific transcriptional regulation of the CYP1A1 gene

Cytochrome P450 1A1 (CYP1A1) monooxygenase plays an important role in the metabolism of environmental pollutants such as polycyclic aromatic hydrocarbons (PAHs) and halogenated polycyclic aromatic hydrocarbons (HAHs). Oxidation of these compounds converts them to the metabolites that subsequently can be conjugated to hydrophilic endogenous entities e.g. glutathione. Derivates generated in this way are water soluble and can be excreted in bile or urine, which is a defense mechanism. Besides detoxification, metabolism by CYP1A1 may lead to deleterious effects since the highly reactive intermediate metabolites are able to react with DNA and thus cause mutagenic effects, as it is in the case of benzo(a) pyrene (B[a]P). CYP1A1 is normally not expressed or expressed at a very low level in the cells but it is inducible by many PAHs and HAHs e.g. by B[a]P or 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). Transcriptional activation of the CYP1A1 gene is mediated by aryl hydrocarbon receptor (AHR), a basic-helix-loop-helix (bHLH) transcription factor. In the absence of a ligand AHR stays predominantly in the cytoplasm. Ligand binding causes translocation of AHR to the nuclear compartment, its heterodimerization with another bHLH protein, the aryl hydrocarbon nuclear translocator (ARNT) and binding of the AHR/ARNT heterodimer to a DNA motif designated dioxin responsive element (DRE). This process leads to the transcriptional activation of the responsive genes containing DREs in their regulatory regions, e.g. that coding for CYP1A1. TCDD is the most potent known agonist of AHR. Since it is not metabolized by the activated enzymes, exposure to this compound leads to a persisting activation of AHR resulting in diverse toxic effects in the organism. To enlighten the molecular mechanisms that mediate the toxicity of xenobiotics like TCDD and related compounds, the AHR-dependent regulation of the CYP1A1 gene was investigated in two cell lines: human cervix carcinoma (HeLa) and mouse hepatoma (Hepa). Study of AHR activation and its consequence concerning expression of the CYP1A1 enzyme confirmed the TCDD-dependent formation of the AHR/ARNT complex on DRE leading to an increase of the CYP1A1 transcription in Hepa cells. In contrast, in HeLa cells formation of the AHR/ARNT heterodimer and binding of a protein complex containing AHR and ARNT to DRE occurred naturally in the absence of TCDD. Moreover, treatment with TCDD did not affect the AHR/ARNT dimer formation and binding of these proteins to DRE in these cells. Even though the constitutive complex on DRE exists in HeLa, transcription of the CYP1A1 gene was not increased. Furthermore, the CYP1A1 level in HeLa cells remained unchanged in the presence of TCDD suggesting repressional mechanism of the AHR complex function which may hinder the TCDD-dependent mechanisms in these cells. Similar to the native, the mouse CYP1A1-driven reporter constructs containing different regulatory elements were not inducible by TCDD in HeLa cells, which supported a presence of cell type specific trans-acting factor in HeLa cells able to repress both the native CYP1A1 and CYP1A1-driven reporter genes rather than species specific differences between CYP1A1 genes of human and rodent origin. The different regulation of the AHR-mediated transcription of CYP1A1 gene in Hepa and HeLa cells was further explored in order to elucidate two aspects of the AHR function: (I) mechanism involved in the activation of AHR in the absence of exogenous ligand and (II) factor that repress function of the exogenous ligand-independent AHR/ARNT complex. Since preliminary studies revealed that the activation of PKA causes an activation of AHR in Hepa cells in the absence of TCDD, the PKA-dependent signalling pathway was the proposed endogenous mechanism leading to the TCDD-independent activation of AHR in HeLa cells. Activation of PKA by forskolin or db-cAMP as well as inhibition of the kinase by H89 in both HeLa and Hepa cells did not lead to alterations in the AHR interaction with ARNT in the absence of TCDD and had no effect on binding of these proteins to DRE. Moreover, the modulators of PKA did not influence the CYP1A1 activity in these cells in the presence and in the absence of TCDD. Thus, an involvement of PKA in the regulation of the CYP1A1 Gen in HeLa cells was not evaluated in the course of this study. Repression of genes by transcription factors bound to their responsive elements in the absence of ligands has been described for nuclear receptors. These receptors interact with protein complex containing histone deacetylase (HDAC), enzyme responsible for the repressional effect. Thus, a participation of histone deacetylase in the transcriptional modulation of CYP1A1 gene by the constitutively DNA-bound AHR/ARNT complex was supposed. Inhibition of the HDAC activity by trichostatin A (TSA) or sodium butyrate (NaBu) led to an increase of the CYP1A1 transcription in the presence but not in the absence of TCDD in Hepa and HeLa cells. Since amount of the AHR and ARNT proteins remained unchanged upon treatment of the cells with TSA or NaBu, the transcriptional upregulation of CYP1A1 gene was not due to an increased expression of the regulatory proteins. These findings strongly suggest an involvement of HDAC in the repression of the CYP1A1 gene. Similar to the native human CYP1A1 also the mouse CYP1A1-driven reporter gene transfected into HeLa cells was repressed by histone deacetylase since the presence of TSA or NaBu led to an increase in the reporter activity. Induction of reporter gene did not require a presence of the promoter or negative regulatory regions of the CYP1A1 gene. A promoter-distal fragment containing three DREs together with surrounding sequences was sufficient to mediate the effects of the HDAC inhibitors suggesting that the AHR/ARNT binding to its specific DNA recognition site may be important for the CYP1A1 repression. Histone deacetylase is recruited to the specific genes by corepressors, proteins that bind to the transcription factors and interact with other members of the HDAC complex. Western blot analyses revealed a presence of HDAC1 and the corepressors mSin3A (mammalian homolog of yeast Sin3) and SMRT (silencing mediator for retinoid and thyroid hormone receptor) in both cell types, while the corepressor NCoR (nuclear receptor corepressor) was expressed exclusively in HeLa cells. Thus the high inducibility of CYP1A1 in Hepa cells may be due to the absence of NCoR in these cells in contrast to the non-responsive HeLa cells, where the presence of NCoR would support repression of the gene by histone deacetylase. This hypothesis was verified in reporter gene experiments where expression constructs coding for the particular members of the HDAC complex were cotransfected in Hepa cells together with the TCDD-inducible reporter constructs containing the CYP1A1 regulatory sequences. An overexpression of NCoR however did not decrease but instead led to a slight increase of the reporter gene activity in the cells. The expected inhibition was observed solely in the case of SMRT that slightly reduced constitutive and TCDD-induced reporter gene activity. A simultaneous expression of NCoR and SMRT shown no further effects and coexpression of HDAC1 with the two corepressors did not alter this situation. Thus, additional factors that are likely involved in the repression of CYP1A1 gene by HDAC complex remained to be identified. Taking together, characterisation of an exogenous ligand independent AHR/ARNT complex on DRE in HeLa cells that repress transcription of the CYP1A1 gene creates a model system enabling investigation of endogenous processes involved in the regulation of AHR function. This study implicates HDAC-mediated repression of CYP1A1 gene that contributes to the xenobiotic-induced expression in a tissue specific manner. Elucidation of these processes gains an insight into mechanisms leading to deleterious effects of TCDD and related compounds.

Strategies to identify and further characterize novel and known genes involved in regulation of skeletogenesis

372 osteochondrodysplasias and genetically determined dysostoses were reported in 2007 [Superti-Furga and Unger, 2007]. For 215 of these conditions, an association with one or more genes can be stated, while the molecular changes for the remaining syndromes remain illusive to date. Thus, the present dissertation aims at the identification of novel genes involved in processes regarding cartilage/ bone formation, growth, differentiation and homeostasis, which may serve as candidate genes for the above mentioned conditions. Two different approaches were undertaken. Firstly, a high throughput EST sequencing project from a human fetal cartilage library was performed to identify novel genes in early skeletal development (20th week of gestation until 2nd year of life) that could be investigated as potential candidate genes. 5000 EST sequences were generated and analyzed representing 1573 individual transcripts, corresponding to known (1400) and to novel, yet uncharacterized genes (173). About 7% of the proteins were already described in cartilage/ bone development or homeostasis, showing that the generated library is tissue specific. The remaining profile of this library was compared to previously published libraries from different time points (8th–12th, 18th–20th week and adult human cartilage) that also showed a similar distribution, reflecting the quality of the presented library analyzed. Furthermore, three potential candidate genes (LRRC59, CRELD2, ZNF577) were further investigated and their potential involvement in skeletogenesis was discussed. Secondly, a disease-orientated approach was undertaken to identify downstream targets of LMX1B, the gene causing Nail-Patella syndrome (NPS), and to investigate similar conditions. Like NPS, Genitopatellar syndrome (GPS) is characterized by aplasia or hypoplasia of the patella and renal anomalies. Therefore, six GPS patients were enrolled in a study to investigate the molecular changes responsible for this relatively rare disease. A 3.07 Mb deletion including LMX1B and NR5A1 (SF1) was found in one female patient that showed features of both NPS and GPS and investigations revealed a 46,XY karyotype and ovotestes indicating true hermaphroditism. The microdeletion was not seen in any of the five other patients with GPS features only, but a potential regulatory element between the two genes cannot be ruled out yet. Since Lmx1b is expressed in the dorsal limb bud and in podocytes, proteomic approaches and expression profiling were performed with murine material of the limbs and the kidneys to identify its downstream targets. After 2D-gel electrophoresis with protein extracts from E13.5 fore limb buds and newborn kidneys of Lmx1b wild type and knock-out mice and mass spectrometry analysis, only two proteins, agrin and carbonic anhydrase 2, remained of interest, but further analysis of the two genes did not show a transcriptional down regulation by Lmx1b. The focus was switched to expression profiles and RNA from newborn Lmx1b wild type and knock-out kidneys was compared by microarray analysis. Potential Lmx1b targets were almost impossible to study, because of the early death of Lmx1b deficient mice, when the glomeruli, containing podocytes, are still immature. Because Lmx1b is also expressed during limb development, RNA from wild type and knock-out Lmx1b E11.5 fore limb buds was investigated by microarray, revealing four potential Lmx1b downstream targets: neuropilin 2, single-stranded DNA binding protein 2, peroxisome proliferative activated receptor, gamma, co-activator 1 alpha, and short stature homeobox 2. Whole mount in situ hybridization strengthened a potential down regulation of neuropilin 2 by Lmx1b, but further investigations including in situ hybridization and protein-protein interaction studies will be needed.

LRP1 modulates APP trafficking and APP metabolism within compartments of the secretory pathway. Increased AICD generation is ineffective in nuclear translocation and transcriptional activation

LRP1 modulates APP trafficking and metabolism within compartments of the secretory pathway The amyloid precursor protein (APP) is the parent protein to the amyloid beta peptide (Abeta) and is a central player in Alzheimer’s disease (AD) pathology. Abeta liberation depends on APP cleavage by beta- and gamma-secretases. To date, only a unilateral view of APP processing exists, excluding other proteins, which might be transported together and/or processed dependent on each other by the secretases described above. The low density lipoprotein receptor related protein 1 (LRP1) was shown to function as such a mediator of APP processing at multiple steps. Newly synthesized LRP1 can interact with APP, implying an interaction between these two proteins early in the secretory pathway. Therefore, we wanted to investigate whether LRP1 can mediate APP trafficking along the secretory pathway, and, if so, whether it affects APP processing. Indeed, we demonstrate that APP trafficking is strongly influenced by LRP1 transport through the endoplasmic reticulum (ER) and Golgi compartments. LRP1-constructs with ER- and Golgi-retention motifs (LRP-CT KKAA, LRP-CT KKFF) had the capacity to retard APP trafficking at the respective steps in the secretory pathway. Here, we provide evidence that APP metabolism occurs in close conjunction with LRP1 trafficking, highlighting a new role of lipoprotein receptors in neurodegenerative diseases. Increased AICD generation is ineffective in nuclear translocation and transcriptional activity A sequence of amyloid precursor protein (APP) cleavages gives rise to the APP intracellular domain (AICD) together with amyloid beta peptide (Abeta) and/or p3 fragment. One of the environmental factors identified favouring the accumulation of AICD appears to be a rise in intracellular pH. This accumulation is a result of an abrogated cleavage event and does not extend to other secretase substrates. AICD can activate the transcription of artificially expressed constructs and many downstream gene targets have been discussed. Here we further identified the metabolism and subcellular localization of the constructs used in this well documented gene reporter assay. We also co-examined the mechanistic lead up to the AICD accumulation and explored possible significances for its increased expression. We found that most of the AICD generated under pH neutralized conditions is likely that cleaved from C83. Furthermore, the AICD surplus is not transcriptionally active but rather remains membrane tethered and free in the cytosol where it interacts with Fe65. However, Fe65 is still essential in AICD mediated transcriptional transactivation although its exact role in this set of events is unclear.

Transcriptional regulation and functional characterization of the tumor suppressor genes Hugl-1 and Hugl-2

Cancer is a multi-step process in which both the activation of oncogenes and the inactivation of tumor suppressor genes alter the normal cellular programs to a state of proliferation and growth. The regulation of a number of tumor suppressor genes and the mechanism underlying the tumor suppression have been intensively studied. Hugl-1 and Hugl-2, the human homologues of Drosophila lgl are shown to be down-regulated in a variety of cancers including breast, colon, lung and melanoma, but the mechanism responsible for loss of expression is not yet known. The regulation of gene expression is influenced by factors inducing or repressing transcription. The present study was focused on the identification and characterization of the active promoters of Hugl-1 and Hugl-2. Further, the regulation of the promoter and functional consequences of this regulation by specific transcription factors was analyzed. Experiments to delineate the function of the mouse homologue of Hugl-2, mgl2 using transgenic mice model were performed. This study shows that the active promoter for both Hugl-1 and Hugl-2 is located 1000bp upstream of transcription start sites. The study also provides first insight into the regulation of Hugl-2 by an important EMT transcriptional regulator, Snail. Direct binding of Snail to four E-boxes present in Hugl-2 promoter region results in repression of Hugl-2 expression. Hugl-1 and Hugl-2 plays pivotal role in establishment and maintenance of cell polarity in a diversity of cell types and organisms. Loss of epithelial cell polarity is a prerequisite for cancer progression and metastasis and is an important step in inducing EMT in cells. Regulation of Hugl-2 by Snail suggests one of the initial events towards loss of epithelial cell polarity during Snail-mediated EMT. Another important finding of this study is the induction of Hugl-2 expression can reverse the Snail-driven EMT. Inducing Hugl-2 in Snail expressing cells results in the re-expression of epithelial markers E-cadherin and Cytokeratin-18. Further, Hugl-2 also reduces the rate of tumor growth, cell migration and induces the epithelial phenotype in 3D culture model in cells expressing Snail. Studies to gain insight into the signaling pathways involved in reversing Snail-mediated EMT revealed that induction of Hugl-2 expression interferes with the activation of extracellular receptor kinase, Erk. Functional aspects of mammalian lgl in vivo was investigated by establishing mgl2 conditional knockout mice. Though disruption of mgl2 gene in hepatic tissues did not alter the growth and development, ubiquitous disruption of mgl2 gene causes embryonic lethality which is evident by the fact that no mgl2-/- mice were born.

Rolle der FOS-Proteine und des MAPK-Signallings in der Regulation der zellulären Antwort auf genotoxischen Stress

Die mittlere Überlebenszeit nach Erkennung eines Glioblastoms ohne Behandlung liegt bei 3 Monaten und kann durch die Behandlung mit Temozolomid (TMZ) auf etwa 15 Monate gesteigert werden. Neben TMZ sind die chlorethylierenden Nitrosoharnstoffe die meistversprechendsten und am häufigsten eingesetzten Chemotherapeutika in der Gliomtherapie. Hier liegt die mittlere Überlebenszeit bei 17,3 Monaten. Um die Therapie des Glioblastoms noch effektiver zu gestalten und Resistenzen zu begegnen, werden unterschiedlichste Ansätze untersucht. Eine zentrale Rolle spielen hierbei das activator protein 1 (AP-1) und die mitogen aktivierten Proteinkinasen (MAPK), deren Funktion in bisherigen Arbeiten noch unzureichend beleuchtet wurde.rnBesonders mit der Rolle des AP-1-bildenden Proteins FRA-1 in der Therapie des Glioblastoms haben sich bisher nur wenige Arbeiten beschäftigt, weshalb im ersten Teil der vorliegenden Arbeit dessen Funktion in der Regulation der Chemosensitivität gegenüber dem chlorethylierenden Agenz ACNU genauer untersucht wurde. Es konnte gezeigt werden, dass die FRA 1-Expression durch Behandlung mit ACNU induziert wird. Die Induktion erfolgte über die beiden MAPKs ERK1/2 und p38K. JNK hatte keinen Einfluss auf die Induktion. Durch die Herunterregulation der FRA-1-Expression mit Hilfe von siRNA und eines shRNA exprimierenden Plasmids kam es zu einer signifikanten Sensitivierung gegenüber ACNU. Dabei konnte gezeigt werden, dass die Herunterregulation der FRA-1-Expression in einer verminderten AP 1-Bildung, bedingt durch eine reduzierte Menge an FRA-1 im AP-1-Komplex resultiert. Die Sensitivierung gegenüber ACNU ist weder durch eine Veränderung in der DNA-Reparatur, noch in der Modulation der FAS-Ligand- bzw. FAS-Rezeptor-Expression bedingt. Auch die hier untersuchten BCL 2-Familienmitglieder wiesen keine Unterschiede in der Expression durch Modulation der FRA 1-Expression auf. Allerdings kam es durch die verminderte FRA-1-Expression zu einer Reduktion der Zellzahl in der G2/M-Phase nach Behandlung mit ACNU. Diese ging einher mit einer reduzierten Menge an phosphoryliertem und unphosphoryliertem CHK1, weshalb davon auszugehen ist, dass FRA 1 nach ACNU-Behandlung in Gliomzellen vor der Apoptose schützt, indem es modulierend auf die Zellzykluskontrolle einwirkt.rnIm zweiten Teil dieser Arbeit wurde die Regulation der apoptotischen Antwort nach Behandlung mit ACNU und TMZ genauer beleuchtet, wobei ein spezielles Augen¬merk auf AP 1 und die MAPKs gelegt wurde. Hier konnte gezeigt werden, dass die Apoptose nach Behandlung mit ACNU bzw. TMZ sowohl durch Spaltung von Pro-Caspase 8, als auch Pro-Caspase 9 eingeleitet wird. Dabei akkumulierte in beiden Fällen p53 vermehrt im Zellkern. Eine Inhibierung der transkriptionellen Aktivität von p53 führte nach ACNU-Behandlung zu einer Sensitivierung der Zellen, nach TMZ-Behandlung kam es zu einem leichten Anstieg in der Vitälität. Der FAS-Rezeptor wurde nach ACNU- und nach TMZ-Behandlung aktiviert und auch die DNA-Reparaturproteine DDB2 und XPC wurden in beiden Fällen vermehrt exprimiert. Für die MAPKs JNK und ERK1/2 konnte gezeigt werden, dass diese pro-apoptotisch wirken. Die AP-1-Bildung nach ACNU-Behandlung erfolgte bereits nach 24 h und war von langer Dauer, wohingegen nach TMZ-Behandlung nur eine transiente AP 1-Bildung zu relativ späten Zeitpunkten detektiert werden konnte. Ebenso konnte für das AP-1-Zielgen FAS-Ligand nach ACNU-Behandlung eine relativ schnelle, lang anhaltende Aktivierung detektiert werden, wohingegen nach TMZ-Behandlung zu einem späten Zeitpunkt ein kurzer Anstieg im Signal zu verzeichnen war. In späteren Experimenten konnte gezeigt werden, dass das BCL-2-Familienmitglied BIM eine zentrale Rolle in der Regulation des intrinsischen Apoptosesignalweges nach Behandlung mit ACNU und TMZ spielt. Die hier entstanden Ergebnisse tragen entscheidend zum Verständnis der durch diese beiden Agenzien gesteuerten, apoptotischen Signalwege bei und bieten eine fundierte Grundlage für weitere Untersuchungen.rn