Most human breast tumors have mutations in the growth factor/phosphatidylinositol 3’ kinase (PI3K) pathway. These can occur in genes coding for receptors, adaptor proteins, catalytic and regulatory subunits of PI3K, downstream kinases, or antagonistic tumor suppressors. While each genetic change results in elevated PI3K signaling, and all major breast cancer subtypes show pathway activation, the specific mutations involved in any one tumor may play an important role in defining tumor subtype, prognosis and sensitivity to therapy. Here, I describe mouse models of PI3K-induced breast cancer. First I generated mice that express Pik3ca cDNA under control of the ROSA26 promoter, in a Cre-dependent and therefore tissue specific way. I have generated four strains of knock-in mice: R26-Pik3cawt, R26-Pik3caE545K, R26-Pik3caH1047R, and R26-Pik3caE545K-H1047R, which can be induced to express wild type, helical domain, kinase domain and double mutant forms of mouse p110α, respectively. Mice expressing mutant Pi3kca develop mammary tumors, but the phenotypic spectrum for each mutation is unique. Indeed, many E545K mammary tumors are ii vascularized, whereas H1047R tumors are not. Using these models, I have compared downstream signaling properties of E545K and H1047R. The potential for improved breast cancer treatment lies in combination therapies that target more than one oncogenic pathway. To develop such treatments, we need good mouse models, and an understanding of the oncogenic network. To this end, my Pik3caH1047R model was mated to p53 and PTEN knockout mice, and to mice with active Notch1 signaling. In each case, genetic cooperation was observed and characterized. Oncogenic PI3K cooperated with p53-loss and active Notch1 to decrease survival and alter tumor phenotype in distinct ways. Loss of PTEN cooperated with oncogenic PI3K to alter tumor type, decrease average age at end point, and increase the number of tumors per mouse. Overall, I have shown that Pik3caE545K and Pik3caH1047R are sufficient to induce mammary tumors, and that tumor characteristics differ with these mutations, and with cooperating genetic changes.

Most human breast tumors have mutations in the growth factor/phosphatidylinositol 3’ kinase (PI3K) pathway. These can occur in genes coding for receptors, adaptor proteins, catalytic and regulatory subunits of PI3K, downstream kinases, or antagonistic tumor suppressors. While each genetic change results in elevated PI3K signaling, and all major breast cancer subtypes show pathway activation, the specific mutations involved in any one tumor may play an important role in defining tumor subtype, prognosis and sensitivity to therapy. Here, I describe mouse models of PI3K-induced breast cancer. First I generated mice that express Pik3ca cDNA under control of the ROSA26 promoter, in a Cre-dependent and therefore tissue specific way. I have generated four strains of knock-in mice: R26-Pik3cawt, R26-Pik3caE545K, R26-Pik3caH1047R, and R26-Pik3caE545K-H1047R, which can be induced to express wild type, helical domain, kinase domain and double mutant forms of mouse p110α, respectively. Mice expressing mutant Pi3kca develop mammary tumors, but the phenotypic spectrum for each mutation is unique. Indeed, many E545K mammary tumors are ii vascularized, whereas H1047R tumors are not. Using these models, I have compared downstream signaling properties of E545K and H1047R. The potential for improved breast cancer treatment lies in combination therapies that target more than one oncogenic pathway. To develop such treatments, we need good mouse models, and an understanding of the oncogenic network. To this end, my Pik3caH1047R model was mated to p53 and PTEN knockout mice, and to mice with active Notch1 signaling. In each case, genetic cooperation was observed and characterized. Oncogenic PI3K cooperated with p53-loss and active Notch1 to decrease survival and alter tumor phenotype in distinct ways. Loss of PTEN cooperated with oncogenic PI3K to alter tumor type, decrease average age at end point, and increase the number of tumors per mouse. Overall, I have shown that Pik3caE545K and Pik3caH1047R are sufficient to induce mammary tumors, and that tumor characteristics differ with these mutations, and with cooperating genetic changes.

Yes / Background: Many urothelial carcinomas (UC) contain activating PIK3CA mutations. In telomerase-immortalized normal urothelial cells (TERT-NHUC), ectopic expression of mutant PIK3CA induces PI3K pathway activation, cell proliferation and cell migration. However, it is not clear whether advanced UC tumors are PIK3CA-dependent and whether PI3K pathway inhibition is a good therapeutic option in such cases. Methods: We used retrovirus-mediated delivery of shRNA to knock down mutant PIK3CA in UC cell lines and assessed effects on pathway activation, cell proliferation, migration and tumorigenicity. The effect of the class I PI3K inhibitor GDC-0941 was assessed in a panel of UC cell lines with a range of known molecular alterations in the PI3K pathway. Results: Specific knockdown of PIK3CA inhibited proliferation, migration, anchorage-independent growth and in vivo tumor growth of cells with PIK3CA mutations. Sensitivity to GDC-0941 was dependent on hotspot PIK3CA mutation status. Cells with rare PIK3CA mutations and co-occurring TSC1 or PTEN mutations were less sensitive. Furthermore, downstream PI3K pathway alterations in TSC1 or PTEN or co-occurring AKT1 and RAS gene mutations were associated with GDC-0941 resistance. Conclusions: Mutant PIK3CA is a potent oncogenic driver in many UC cell lines and may represent a valuable therapeutic target in advanced bladder cancer.

La voie PI3K/AKT est impliquée dans la progression de divers cancers humains, et semble jouer un rôle majeur dans le développement de tumeurs lymphoïdes. Elle pourrait être impliquée dans la pathogénie du lymphome folliculaire (LF) par certains mécanismes non identifiés. Les travaux de thèse portent sur l'étude des anomalies de la voie PI3K/AKT dans le LF, dans le but d'apporter une nouvelle cible thérapeutique. 38 biopsies tissulaires de LF humain ont été étudiées pour une analyse mutationnelle du gène PIK3CA dans les exons 9 et 20 par séquençage. Les mêmes échantillons ont été analysés par western blot et immunohistochimie pour détecter l'expression des protéines AKT, AKT phosphorylée (pAKT), et PTEN. Deux cas de lymphadénite ont été utilisés comme témoins.Les résultats obtenus montrent que l'expression d'AKT était présente dans tous les cas de LF et lymphadénite, et 14/38 (37%) échantillons de LF et 2/2 cas de lymphadénite exprimaient pAKT. 9/38 (24%) échantillons de LF ont montré un niveau élevé d'expression de pAKT, alors que 5/38 (13%) cas de LF, et 2/2 échantillons de lymphadénite montraient un faible niveau d'expression de pAKT. L'expression de PTEN a été observée dans 30/38 (79%) cas de LF et 2/2 cas de lymphadénite, tandis que 8/38 (21%) cas ont montré une perte d'expression de PTEN. En outre, 3 cas qui expriment pAKT montrent une perte d'expression de PTEN. Aucune mutation du gène PIK3CA n'a été détectée dans les échantillons étudiés. Ces données suggèrent que la voie PI3K/AKT peut être activée dans certains cas de LF, soit en raison de la phosphorylation d'AKT, soit en raison d'une perte d'expression de PTEN, en absence de mutations de PIK3CA. / The phosphoinositide 3- kinase (PI3K) pathway is involved in the growth of various human cancers, including lymphoid malignancies. However its role in the pathogenesis of follicular lymphoma (FL) has not been yet described.The PhD work focuses on the study of alterations in the PI3K/AKT pathway in follicular lymphoma, in order to provide a new therapeutic target.To clarify this point, biopsy tissue samples from 38 human FL cases were investigated for PIK3CA somatic mutations in exons 9 and 20 using Sanger sequencing. The same samples were analyzed using western blotting and immunohistochemistry to detect expression of AKT, phosphorylated AKT (pAKT), and PTEN proteins. Two cases of benign lymphadenitis were used as controls. AKT expression was present in all FL and lymphadenitis cases. 14/38 (37%) FL and 2/2 lymphadenitis cases expressed pAKT. 9/38 (24%) FL samples showed high level of pAKT, whereas 5/38 (13%) FL cases and 2/2 benign lymphadenitis samples expressed pAKT at low level. PTEN expression was observed in 30/38 (79%) FL and 2/2 benign lymphadenitis cases, whereas 8/38 (21%) of FL cases showed loss of PTEN expression. In addition, 3 cases with positive pAKT did not express PTEN. PIK3CA mutations were not detected in any sample. These data suggest that the PI3K/AKT signaling pathway could be activated in a subset of FL cases, due to either AKT phosphorylation or PTEN downregulation, in the absence of PIK3CA mutations.

No description available.

Les résultats des projets actuels apportent une information, sur différents aspects des rôles de la voie PI3K, dans le développement du cancer du sein, et la réponse au traitement. Les projets particuliers couvrent des sujets liés à la voie aux niveaux concernant les récepteurs de la famille HER, activant la voie PI3K, ainsi que PI3K et les effecteurs en découlant. Les effets pronostic et prédictif de la dérégulation de PI3K sont les sujets centraux de la recherche décrite ici. Une baisse d’expression de PI3KR1 est associée à une survie réduite dans notre cohorte de patients. Une attention particulière a été portée aux mutations de PIK3CA communes dans le cancer du sein. Tandis que les mutations de PIK3CA agissent comme des marqueurs de bon pronostic chez les patients anti-HER2-naïfs, ces mutations agissent au contraire comme prédicteurs négatifs de la réponse au traitement par trastuzumab. Les résultats décrits mènent un peu plus vers l’implication de plusieurs voies moléculaires altérées, en particulier la voie de signalisation Wnt, dans la tumorigénèse des cancers du sein PIK3CA mutés. De plus, nous avons testé les taux de lapatinib plasmatique montrant une augmentation pertinente dans les périodes d’état d’équilibre du traitement. Par ailleurs, nous avons démontré des incohérences dans l’évaluation de l’EGFR et proposé des approches pour l’interprétation des comptages d’immunohistochimie et de FISH. Tous ces sujets sont connectés par la 170 voie PI3K, et le besoin d’approfondir les connaissances actuelles, et d’apporter de nouvelles informations utiles applicables dans le futur dans les pratiques cliniques / Results of the presented research projects bring information about several aspects of the PI3K signaling pathway roles in breast cancer development and treatment response. The particular projects covered the subjects connected with the signaling pathway, ranging from the HER family receptors activating the pathway, and PI3K to the downstream levels of signalisation. The prognostic and predictive effect of PI3K deregulation was the central subject of the described research. The decreased expression of PIK3R1 associated with reduced survival of our patients. A special focus was put on the PIK3CA mutations which are common in breast cancer. Whereas the PIK3CA mutations act as a good prognostic marker in patients non-treated with the HER2 inhibitors, these mutations predict a negative response to trastuzumab treatment. The described results, furthermore, draw attention to the role of several altered molecular signaling pathways in breast cancer development, especially to the Wnt signaling pathway. The lapatinib plasma levels showing the relevant increase in comparison with the already described efficient steady-state levels were also described in one of the projects. Moreover, various modifications to EGFR status assessment were compared and showed that EGFR FISH and IHC count interpretation depended significantly on method and thresholds used. All these subjects are connected by the PI3K pathway, the need to deepen current knowledge and bring new useful information applicable in future clinical practice.

Le cancer du sein est le cancer le plus fréquemment diagnostiqué chez la femme et constitue un véritable problème de santé publique. Les progrès de la biologie moléculaire ont permis la caractérisation des principales voies de signalisation et ont mis en évidence l'implication majeure de la signalisation cellulaire dans les processus de cancérogenèse. Des cibles moléculaires ont ainsi été identifiées et ont permis le développement de thérapeutiques dites ciblées, telles que les anticorps monoclonaux ou encore les inhibiteurs de kinase. Malgré ces avancées considérables qui ont permis l'amélioration de la prise en charge des patientes, on constate l'apparition de résistances aux traitements. Ce travail avait pour objectifs d'identifier de nouveaux biomarqueurs et de déterminer leur signification clinique, leur intérêt théranostique ainsi que leur impact sur la réponse aux traitements. Dans un premier temps nous avons étudié les mutations activatrices du gène PIK3CA. Ces mutations sont retrouvées dans 25% des cancers du sein et sont impliquées dans la résistance au trastuzumab, aux anti-œstrogènes et aux inhibiteurs de mTOR. 149 échantillons de tumeurs de sein infiltrantes ont été analysés par une technique de PCR-HRM (High Resolution Melting) et 118 échantillons par une technique de PCR-ARMS (Amplification Refractory Mutation System). Les résultats des 2 techniques étaient concordants (k=0,845 ; p<0,001) et une relation entre mutations du gène PIK3CA et grade SBR a été mise en évidence, les tumeurs de grade SBR III étant moins fréquemment mutées que les autres (p=0.025 en HRM et p=0.009 en ARMS). Dans un second temps, notre travail a consisté en l'exploration fonctionnelle des voies de signalisation PI3K/AKT/mTOR, RAS/RAF/MAPKinases et P38MAPKinase. Pour cela nous avons analysé le niveau d'expression des phosphoprotéines p-AKT, p-GSK3ß, p-S6 kinase, p-MEK1, p-ERK1/2, p-P90RSK, p-IGF1R ainsi que p-P38MAPK par immuno-analyse multiplexe. Cette partie a comporté 3 études. Une première étude rétrospective sur 45 échantillons de tumeurs mammaires invasives congelées a mis en évidence des niveaux d'expression de P38 et de p-P38 plus élevés dans les tumeurs RE+. La deuxième étude était une étude prospective visant à déterminer des biomarqueurs de réponse à l'association trastuzumab-évérolimus chez des patientes présentant un cancer du sein précoce traitées en préopératoire. Cette étude a révélé une augmentation statistiquement significative du niveau d'expression de p-MEK1 (p=0.012), p-ERK1/2 (p=0.003) et p-P38MAPK (p<0.001) dans le bras de traitement associant l'évérolimus au trastuzumab qui pourrait s'expliquer par la suppression par l'évérolimus d’une boucle de rétrocontrôle négatif contrôlant l'activation de la voie RAS/RAF/MAPKinases. Dans le bras de traitement évaluant le trastuzumab seul, aucune variation du niveau d'expression des phosphoprotéines n'a été mise en évidence, y compris en aval du récepteur HER2, ce qui soulève l'hypothèse d'un mécanisme d'action prédominant immunologique du trastuzumab. La troisième étude qui comparait l'impact du trastuzumab in vitro et en situation clinique confirme la différence des mécanismes d'action mis en jeu en fonction des conditions cellulaires et cliniques. Dans son ensemble, ce travail a mis en évidence que la détermination du statut mutationnel du gène PIK3CA et du niveau d’expression des phosphoprotéines pourrait être utile à une meilleure caractérisation moléculaire des cancers du sein et à l’optimisation de la personnalisation des prescriptions de thérapies ciblées / Breast cancer is the most frequently diagnosed cancer in women and is a real public health problem.Advances in molecular biology have allowed the characterization of the major signaling pathways and revealed their major implication in carcinogenesis processes. Molecular targets have been identified and have enabled the development of targeted therapies, such as monoclonal antibodies, or kinase inhibitors. Despite these considerable advances that have improved the care of patients, emerging of resistance to treatments has been observed. The aim of this work was to identify new tumor biomarkers and determine their clinical significance, their theranostic interest and their impact on the response to targeted therapies. Initially, we studied the activating mutations of the PIK3CA gene. These mutations are found in 25% of breast cancers and are involved in resistance to trastuzumab, antiestrogens and mTOR inhibitors. We analyzed 149 invasive breast tumor samples for PIK3CA gene mutations by PCR-HRM (High Resolution Melting) and 118 by PCR-ARMS (Amplification Refractory Mutation System). The results achieved with the 2 techniques were consistent (k = 0.845; p <0.001) and a relationship between PIK3CA mutations and grade SBR was highlighted with a lower occurrence of mutations in SBR grade III tumors (p=0.025 in HRM and p=0.009 in ARMS). Secondly, we investigated the functional alteration of PI3K/AKT/mTOR, RAS/RAF/MAPKinases and P38MAPKinase signaling pathways. We have analyzed the expression level of phosphoproteins p-AKT, p-GSK3ß, p-S6 kinase, p-MEK1, p-ERK1 / 2, p-P90RSK, p-IGF1R and p-p38MAPK by multiplex immunoanalysis. This part includes 3 studies. The first study was a retrospective study of 45 frozen samples of invasive breast tumors in which we observed that the level of expression of P38 and p-P38 was higher in the ER + tumors. The second study was a prospective study to identify biomarkers of response to trastuzumab-everolimus association in patients with early breast cancer treated preoperatively. This study showed a statistically significant increase of the expression level of p-MEK1 (p = 0.012), p-ERK1/2 (p = 0.003) and p-p38MAPK (p<0.001) in arm treated by trastuzumab associated with everolimus. It could be explained by the repression of a negative feedback loop involving S6K, PI3K and RAS by everolimus, leading to the activation of RAS/RAF/MAPKinases signaling pathway. In the control arm investigating trastuzumab alone, no significant variations of the level of expression of phosphoproteins was demonstrated, raising the hypothesis of the implementation of a predominant immunological mechanism of action for Trastuzumab. The third study that compared effect of trastuzumab in vitro and in clinical setting confirms that trastuzumab has different modes of action when evaluated in cells and in clinical conditions. As a whole, this work showed that determining the mutation status of PIK3CA and the expression level of phosphoproteins could be useful to refine the molecular characterization of breast cancers and optimize the criteria used to personalize the prescription of targeted therapies

Numerous mutations contribute to tumorigenesis of cancer cells. For most of them it remains unclear whether they are driver or passenger mutations. A classic knock-out to study their function in cancer cells used to take a lot of effort. The CRISPR/Cas-system can be used as a programmable “genome editing” tool. In this work, oncogenes have been inactivated with the CRISPR/Cas-system. Considering off-targets, Streptococcus pyogenes sgRNAs can be designed for 88% of the known cancer mutations. The activity of 15 sgRNAs, targeting 13 mutations in proto-oncogenes (deletions, insertions and point mutations), has been tested with a RFP-GFP-reporter plasmid. For 13 sgRNAs, activity prediction scores correlated with measured activity. Furthermore, sgRNAs have shown preferential binding to mutated versions of targeted proto-oncogene sequence and did not induce double strand breaks in the wild type sequence. For 10 sgRNAs, the activity against their target sequence has been more than 4 times higher than against the wild type sequence. Most of those sgRNAs target insertions or deletions and fewer target point mutations. Permanent knock-out of three mutated proto-oncogenes NPM1, BRAF and PIK3CA has been achieved with a lentiviral expression of CRISPR/Cas. Accordingly, effects on proliferation and phenotype have been studied. Knock-out of NPM1 c.863_864insTCTG mutation has been studied in heterozygous mutated OCI AML3 cell line. Proliferation was strongly inhibited by the corresponding sgRNA. Cells arrested in G0/1-phase of cell cycle (77%) compared to control cells (56%), although no difference was observed for sub-G1 phase, indicating no induction of apoptosis. Cells treated with NPM1 sgRNA had 88% reduced expression of NPM1 c.863_864insTCTG mRNA as well as less cytoplasmic localization of nucleophosmin as assessed by immunostaining. The activity of sgRNA has been confirmed by deep sequencing, showing a shift of wild type to mutated allele ratio from 51:49 to 68:32. This effect was enhanced by the additional treatment with the NHEJ inhibitor SCR7. A BRAFV600E sgRNA was tested in homozygously mutated melanoma cell lines A-375 and SK MEL-28. No differences were detected in comparison to controls. However, in the CRC cell line RKO, heterozygous for BRAFV600E and PIK3CAH1047R, proliferation was inhibited through sgRNAs against either BRAF or PIK3CA. A combination of both had no synergistic effect on proliferation. Activity and specificity of the sgRNA targeting BRAF were confirmed by deep sequencing, while the PIK3CA sgRNA showed a moderate induction of double strand breaks also in the wild type allele. The relation of wild type to mutated allele of BRAF was changed from 32:68 before treatment to 51:49 afterwards. This effect can be explained by a “re mutation” to the wild type after DSB via HDR with wild type sister chromatid as template. This effect was observed for PIK3CA sgRNA to a lesser extent. In conclusion, these results show the applicability of the CRISPR/Cas-system for the inactivation of mutated proto-oncogenes.:List of tables III List of figures IV List of abbreviations V 1 Introduction 1 1.1 Cancer 1 1.2 Oncogenes 2 1.2.1 Role in cancer 2 1.2.2 Targeted therapies 3 1.2.3 NPM1 5 1.2.4 BRAF 6 1.2.5 PIK3CA 7 1.3 CRISPR/Cas-system 7 1.4 Aim and motivation 10 2 Material and Methods 11 2.1 Design of sgRNAs 11 2.2 Plasmids 11 2.3 Cell culture 12 2.4 FACS analysis 14 2.5 T7 assay 14 2.6 Cell cycle analysis 15 2.7 Immunostaining 15 2.8 Apoptosis assay 16 2.9 Quantification of mutant NPM1 transcripts 16 2.10 Deep sequencing 16 2.11 Statistical Analysis 19 3 Results 20 3.1 Design of sgRNAs targeting oncogenes 20 3.2 Evaluation of sgRNA efficacy and selectivity 23 3.3 Effects of oncogene knock-out in cancer cell lines 27 3.3.1 Targeting NPM1 in AML cells 27 3.3.2 Targeting BRAF in melanoma cells 30 3.3.3 Targeting BRAF and PIK3CA in colorectal carcinoma cells 31 4 Discussion 37 4.1 The design of sgRNAs is possible for most cancer mutations 37 4.2 sgRNAs targeting oncogenes have to be tested 37 4.3 Oncogenes can be knocked out with the CRISPR/Cas-system 37 4.3.1 NPM1 in AML cells 37 4.3.2 BRAF in melanoma cells 38 4.3.3 BRAF and PIK3CA in CRC cells 38 4.4 Advantages and disadvantages to target oncogenes with the CRISPR/Cas-system 40 4.5 Concluding remarks 41 5 Original Article 43 6 Summary 47 7 Zusammenfassung 49 List of references 51 Appendix VIII / In Krebszellen tritt eine Vielzahl von Mutationen auf. Für den Großteil der Mutationen ist ungeklärt, ob es sich um krebsverursachende oder passagere Mutationen handelt. Ein gezieltes Ausschalten (Knock-out) dieser Gene zur Untersuchung ihrer Funktion in Krebszellen war bisher mit großem Aufwand verbunden. Das CRISPR/Cas-System lässt sich als programmierbares „Genome-editing“ Werkzeug einsetzen und wurde in der vorliegenden Arbeit verwendet, um gezielt mutierte Protoonkogene zu inaktivieren. Für 88% der bekannten, in Krebszellen auftretenden Mutationen lassen sich, unter Berücksichtigung von off-targets, Streptococcus pyogenes sgRNAs entwerfen. Mit Hilfe eines RFP-GFP-Reporter-Plasmides wurde die Aktivität von 15 sgRNAs gegen 13 Mutationen (Deletionen, Insertionen und Punktmutationen) in Protoonkogenen überprüft. Für 13 der sgRNAs zeigte sich eine Aktivität, die mit der Vorhersage durch den Algorithmus korrelierte. Außerdem wurde gezeigt, dass die sgRNAs spezifisch genug binden, um zwar bei der mutierten Sequenz eines Protoonkogens, jedoch nicht bei der Wildtyp-Sequenz Doppelstrangbrüche zu erzeugen. Unter den sgRNAs waren 10 mit mehr als 4-fach höherer Aktivität bei komplett übereinstimmender Zielsequenz gegenüber der Wildtyp-Sequenz. Diese spezifischen sgRNAs waren vor allem gegen Insertions- oder Deletionsmutationen gerichtet, einige auch gegen Punktmutationen. Durch permanente, lentivirale Expression von CRISPR/Cas wurden die Effekte eines Knock-out von drei mutierten Protoonkogenen, NPM1, BRAF und PIK3CA, auf das Wachstum und phänotypische Aspekte humaner Krebszelllinien untersucht. Ein Knock-out der NPM1 c.863_864insTCTG Mutation wurde in heterozygot mutierten OCI AML3 Zellen untersucht, es zeigte sich eine starke Proliferationshemmung. In der Zellzyklusanalyse trat ein G0/1-Arrest dieser Zellen (77%) im Vergleich mit Kontroll-Zellen (56%) auf, jedoch keine Unterschiede in der sub-G1-Analyse, sodass nicht von einer vermehrten Apoptose auszugehen ist. Die mit sgRNA behandelten OCI-AML3 Zellen zeigten sowohl eine um 88% verminderte NPM1 c.863_864insTCTG mRNA-Expression als auch verminderte zytoplasmatische Sublokalisation des Nucleophosmins in der Immunfärbung. Die hohe Aktivität der gRNA gegen mutiertes NPM1 wurde durch Deep Sequencing bestätigt, außerdem hat sich das Verhältnis vom Wildtyp- zu mutiertem Allel von 51:49 zu 68:32 verschoben. Dieser Effekt wurde durch Zugabe des NHEJ-Hemmstoffes SCR7 noch verstärkt. Die sgRNA gegen BRAFV600E wurde in den homozygot mutierten Melanom-Zelllinien A-375 und SK-MEL-28 getestet. Bei Proliferationsversuchen zeigten sich keine Unterschiede im Vergleich zu Kontrollzellen. In der kolorektalen Krebszelllinie RKO, die heterozygot BRAFV600E und PIK3CAH1047R ist, zeigte sich bei der Testung von sgRNAs gegen BRAF, PIK3CA und Kombination beider sgRNAs eine Wachstumshemmung. Jedoch lag kein synergistischer Effekt bei sgRNA-Kombination vor. Zudem bestätigten sich Aktivität und Spezifität der sgRNA gegen BRAF im Deep Sequencing, während die sgRNA gegen PIK3CA in mäßigem Umfang Doppelstrangbrüche im Wildtyp-Allel verursachte. Das Verhältnis vom Wildtyp- zu mutiertem BRAF Allel verschob sich von 32:68 ohne sgRNA zu 51:49 nach sgRNA-Behandlung. Eine mögliche Erklärung dieser Beobachtung ist die Rückmutation zum Wildtyp-Allel nach Doppelstrangbruch mit Hilfe homologer Rekombination durch das Wildtyp-Schwesterchromatid. Für PIK3CA konnte dieser Effekt in schwächerem Ausmaß ebenfalls beobachtet werden. Zusammengefasst zeigen diese Ergebnisse, dass das CRISPR/Cas-System zur Inaktivierung mutierter Protoonkogene genutzt werden kann.:List of tables III List of figures IV List of abbreviations V 1 Introduction 1 1.1 Cancer 1 1.2 Oncogenes 2 1.2.1 Role in cancer 2 1.2.2 Targeted therapies 3 1.2.3 NPM1 5 1.2.4 BRAF 6 1.2.5 PIK3CA 7 1.3 CRISPR/Cas-system 7 1.4 Aim and motivation 10 2 Material and Methods 11 2.1 Design of sgRNAs 11 2.2 Plasmids 11 2.3 Cell culture 12 2.4 FACS analysis 14 2.5 T7 assay 14 2.6 Cell cycle analysis 15 2.7 Immunostaining 15 2.8 Apoptosis assay 16 2.9 Quantification of mutant NPM1 transcripts 16 2.10 Deep sequencing 16 2.11 Statistical Analysis 19 3 Results 20 3.1 Design of sgRNAs targeting oncogenes 20 3.2 Evaluation of sgRNA efficacy and selectivity 23 3.3 Effects of oncogene knock-out in cancer cell lines 27 3.3.1 Targeting NPM1 in AML cells 27 3.3.2 Targeting BRAF in melanoma cells 30 3.3.3 Targeting BRAF and PIK3CA in colorectal carcinoma cells 31 4 Discussion 37 4.1 The design of sgRNAs is possible for most cancer mutations 37 4.2 sgRNAs targeting oncogenes have to be tested 37 4.3 Oncogenes can be knocked out with the CRISPR/Cas-system 37 4.3.1 NPM1 in AML cells 37 4.3.2 BRAF in melanoma cells 38 4.3.3 BRAF and PIK3CA in CRC cells 38 4.4 Advantages and disadvantages to target oncogenes with the CRISPR/Cas-system 40 4.5 Concluding remarks 41 5 Original Article 43 6 Summary 47 7 Zusammenfassung 49 List of references 51 Appendix VIII

Mosaic, activating mutations in PIK3CA, the gene encoding the catalytic p110α subunit of class IA phosphatidylinositol 3-kinase (PI3K), are the cause of rare, developmental growth disorders collectively known as PIK3CA-Related Overgrowth Spectrum (PROS). Given the pressing need for targeted therapy and evidence for tissue- and cell lineage-specific distribution of PIK3CA mutations in PROS, developmental models of this disease will be a key asset for preclinical drug testing and for a better understanding of PIK3CA activation in development. This PhD project addressed the lack of human, developmental PROS models by establishing isogenic series of human induced pluripotent stem cells (iPSCs) with endogenously expressed, activating PIK3CA mutations. This involved the optimisation of a CRISPR/Cas9 protocol for efficient knockin of different PIK3CA variants into human iPSCs. An isogenic iPSC series was established with cells expressing either wild-type PIK3CA or PIK3CA-H1047R, knocked into either one or both endogenous alleles. In parallel, mosaic patient- derived fibroblast cultures were reprogrammed to obtain isogenic wild-type and heterozygous iPSCs expressing PIK3CA-E418K. The models were used in comprehensive signalling studies, providing new insights into PI3K signalling in human iPSCs and how it is perturbed by genetic p110α activation. PIK3CA-E418K, a rare variant in both PROS and cancer, caused minimal pathway activation, in contrast to the highly recurrent variant PIK3CA-H1047R which induced strong PI3K signalling in both heterozygous and homozygous iPSCs according to a graded pattern. Studies of clinically relevant PI3K pathway inhibitors provided proof-of-concept that stem cell-based PROS models can be used for preclinical drug testing, and demonstrated that p110α is likely to be the main catalytic isoform mediating canonical PI3K signalling in human iPSCs. Differentiation assays revealed allele dose-dependent effects of PIK3CA-H1047R on stemness, with homozygous iPSCs exhibiting widespread transcriptome remodelling affect- ing genes implicated in cancer and development. Accordingly, these cells showed increased expression of pluripotency genes such as NANOG and NODAL, resulting in self-sustained "stemness" in embryoid body and teratoma assays. In comparison, heterozygous mutants behaved similar to wild-type controls under all differentiation paradigms. Furthermore, evidence was obtained that strong activation of PI3K signalling is fully compatible with definitive endoderm formation, arguing against cell-autonomous differentiation defects as the cause of endoderm sparing in PROS. In summary, these studies demonstrate the utility of human stem cell-based models of PROS for preclinical drug testing and for improved understanding of class IA PI3K signalling in human development. They are also likely to be useful in efforts to obtain a better understanding of PIK3CA-H1047R in human cancer.

This thesis is comprised of two parts. In the first part we study the influence of four frequently disputed genes on the susceptibility for developing prostate cancer, and in the second part we attempt to establish a basic understanding of the molecular genetic events in penile cancer. In a prostate cancer cohort we have investigated the relation of prostate cancer risk and single nucleotide polymorphisms (SNPs) in four different genes coding for the androgen receptor (AR), the vitamin D receptor (VDR), insulin (INS) and insulin receptor substrate 1 (IRS1). Despite strong biological indications of an involvement of these genes in prostate carcinogenesis, the results from different studies are contradictory and inconclusive. The action of the AR varies between individuals in part owing to a repetitive CAG sequence (polyglutamine) in the first exon of the AR gene. The results presented in this thesis show that in our cohort of prostate cancer patients the average number of repeats is 20.1, which is significantly (p&lt;0.001) fewer repeats compared to healthy control individuals, where the average is 22.5 repeats. We find a 4.94 fold (p=0.00003) increased risk of developing prostate cancer associated with having short repeat lengths (≤19 repeats), compared with long repeats (≥23 repeats). In paper I we also study the TaqI polymorphism in the VDR gene, and find that it does not modify the risk of prostate cancer. In the INS gene we study the +1127 PstI polymorphism and find no overall effect on the risk of prostate cancer. However, we do find that the CC genotype is associated with low grade disease defined as having a Gleason score ≤6 (OR=1.46; p=0.018). In the IRS1 gene we study the G972R polymorphism and observe that the R allele is significantly associated with a 2.44 fold increased prostate cancer risk (p=0.010). The knowledge of molecular genetic events in penile cancer is very scarce and to date very few genes have been identified to be involved in penile carcinogenesis. We chose therefore to analyse the penile cancer samples using genome-wide high-density SNP arrays. We find major regions of frequent copy number gain in chromosome arms 3q, 5p and 8q, and slightly less frequent in 1p, 16q and 20q. The chromosomal regions of most frequent copy number losses are 3p, 4q, 11p and 13q. We suggest four candidate genes residing in these areas, the PIK3CA gene (3q26.32), the hTERT gene (5p15.33), the MYC gene (8q24.21) and the FHIT gene (3p14.2). The mutational status of the PIK3CA and PTEN genes in the PI3K/AKT pathway and the HRAS, KRAS, NRAS and BRAF genes in the RAS/MAPK pathway was assessed in the penile cancer samples. We find the PIK3CA, HRAS and KRAS genes to be mutated in 29%, 7% and 3% of the cases, respectively. All mutations are mutually exclusive. In total the PI3K/AKT and RAS/MAPK pathways were found to be activated through mutation or amplification in 64% of the cases, indicating the significance of these pathways in the aetiology of penile cancer.