Pour assurer la production de milliards de globules rouges, l’érythropoièse doit parfaitement contrôler les processus de prolifération et de différenciation. Ces deux processus sont régulés par l’expression de gènes spécifiques dépendant d’une coordination entre l’activité des facteurs de transcription (FT) et les fonctions épigénétiques portées par exemple par les protéines à bromodomaine. Cette étude se concentre sur les conséquences de l’association ou la dissociation du FT clef de l’érythropoièse GATA-1 avec les FT déterminant pour le cycle cellulaire, pRb et E2F. Dans la première partie de ma thèse, j’ai participé à l’étude du rôle de l’association/dissociation de GATA-1 et FOG-2 avec pRb/E2F dans le contrôle la balance prolifération/différenciation cellulaire. Nos résultats montrent que les souris exprimant une mutation de GATA-1 sur la sérine 310 (GATA-1S310A), qui a la capacité accrue à séquestrer E2F-2, présentent une anémie létale lorsqu’un mécanisme de compensation de production de E2F-2 induit par l’IGF-1 est inhibé. Puis, nous avons trouvé que les propriétés décrites pour GATA-1 sont partagées par le FT FOG-2 et montré que l’abrogation de sa fixation avec pRb induit une perturbation de l’adiposité dans des souris FOG-2pRb-. Dans la deuxième partie, l’expression de c-Myc étant régulé différentiellement par GATA-1 et E2F, j’ai testé si la drogue « JQ1 », premier inhibiteur épigenétique chimique de l’expression de c-Myc, pouvait contrôler l’érythropoièse. Pour cela, j’ai utilisé la ligné érythroleucémique UT7 qui prolifère sans se différencier en présence d’érythropoiétine (stade proérythroblaste). Les résultats montrent que le traitement par JQ1 bloque la prolifération des cellules UT7 et permet de réinitier le programme de différentiation érythroide terminale. J’ai alors recherché les mécanismes moléculaires impliqués dans cette régulation et trouvé que l’inhibition transcriptionnelle de c-Myc par JQ1 est associée à l’inhibition de l’activité transcriptionnelle de STAT5 sans modification de son état de phosphorylation. Enfin, j’ai montré que JQ1 pouvait avoir une activité comparable à celle du TGF-b mais sans implication les voies Smad. Des études in vivo montre que JQ1 augmente la viabilité cellulaire et accélère la maturation des cellules érythroides à la fois chez les souris sauvages et thalassémiques. Cette différence d’action de JQ1 sur l’érythropoièse normale et pathologique implique des modifications épigénétiques différentielles entre ces deux types cellulaires et sont à la base de nouvelles stratégies du traitement du cancer. Le rôle clef de la régulation de l’association/dissociation de GATA-1 ou FOG-2 avec pRb/E2F dans l’érythropoièse et l’adipogénèse, nous a conduit, dans une troisième partie, à déterminer in vivo, les conséquences physiologiques de la séquestration de E2F par pRb. Pour cela nous avons crée une souris transgénique exprimant de façon conditionnelle un peptide contenant la partie N terminale de GATA-1 qui se fixe à pRb (GATA-1Nter). In vitro, ce peptide séquestre E2F dans le complexe GATA-1Nter/pRb et inhibe la prolifération cellulaire de façon irréversible. In vivo, aucune souris transgéniques exprimant le peptide GATA-1Nter n’a pu être sélectionnée et une mortalité au stade embryonnaire est observée. Une expression induite de ce peptide au stade adulte ne produit que des souris chimériques avec une fréquence de recombinaison du transgène GATA-1Nter importante. L’établissement de lignées stables de souris exprimant le peptide GATA-1Nter permettra de déterminer les conséquences physiologiques de la séquestration de E2F dans le complexe GATA-1Nter/pRb. / To ensure the generation of billions of erythrocytes daily, erythropoiesis must be well controlled by proliferation and differentiation processes. These two processes are regulated by expressions of specific genes, coordinated by transcription factors (TFs) and epigenetic factors, such as bromodomain proteins. This study focused on the effects of the binding and dissociation of a key erythroid TF, GATA-1, to the crucial cell cycle TFs, pRb and E2F. In the first part of this thesis, the role of GATA-1 and FOG-2 binding to pRb/E2F in a control balances between cell proliferation and differentiation was studied. Mice bearing a GATA-1 mutation (GATA-1S310A) displayed higher levels of E2F2 sequestration and suffered from fatal anemia when the compensatory pathway of E2F2 production via IGF-1 signaling was also inhibited. The properties described for GATA-1 were found to be common to FOG-2, and the abolition of FOG-2 binding to pRb led to obesity resistance in FOG-2pRb- mice. In the second part of this work, as c-Myc is regulated by GATA-1 and E2F, the first chemical epigenetic inhibitor repressing c-Myc expression to be described, JQ1, was investigated to see if it could control erythropoiesis. The UT7 erythroleukemia cell line, which proliferates without differentiating was used. This cell line stops differentiation at the proerythroblast stage, in response to erythropoietin. JQ1 treatment inhibited UT7 proliferation and restored terminal erythroid differentiation. The molecular mechanism underlying this regulation by JQ1 was shown that the inhibition of c-Myc expression was associated with the inhibition of STAT5 transcription, with no change in the phosphorylation of this protein. It was found that JQ1 had a putative TGF--like activity, which did not involve the Smad pathway. It was shown in the ex vivo studies that JQ1 increased the viability of erythroid cells and accelerated the maturation of these cells in both WT and thalassemic mice. The observed differences between leukemic and normal erythropoiesis involved differential epigenetic modifications that could be at the basis of new strategies regarding cancer treatment.The key role of the association of GATA-1 or FOG-2 had with pRb/E2F, and the dissociation of these factors, in erythropoiesis and adipogenesis, respectively, led us to investigate, in vivo, the physiological consequences of E2F sequestration by pRb. As a result, transgenic mice displaying conditional expression of a peptide containing the N-terminal part of GATA-1 that binds to pRb (GATA-1Nter) were developed. In vitro, this peptide traps E2F in a GATA-1Nter/pRb complex, resulting in the irreversible inhibition of cell proliferation. The yield of transgenic mice expressing the GATA-1Nter peptide in vivo was unsuccessful, as this expression lead to lethality at the embryonic stage. Using an alternative approach, based on the inducible expression of the peptide in adults, chimeric mice with a high frequency of recombination of the GATA-1Nter transgene were obtained for this study. The establishment of a stable mouse line expressing the GATA-1Nter peptide should make it possible to determine the pathophysiological consequences of E2F sequestration in the GATA-1Nter/pRb complex.

The myc family of oncogenic transcription factors, which includes c-myc, N-myc and L-myc, control major cellular processes such as proliferation and differentiation by integrating upstream signals and orchestrating global gene transcription. They do this largely through dimerising with Max, which together bind to enhancer (E)-box elements in DNA. Myc proteins function similarly but differ in potency and tissue distribution. For instance, N-myc is expressed predominantly during development in undifferentiated cells of the nervous system, whereas c-myc is ubiquitously expressed in all proliferating cells. Myc proteins, when deregulated, are major drivers of tumourigenesis. Myc deregulation occurs in up to 70% of all human cancers and is often associated with the most aggressive forms. For example, MYCN, the gene encoding N-myc, is amplified in 20-30% of neuroblastomas, and amplification strongly correlates with advanced stage and poor prognosis. Myc proteins are therefore considered “most wanted” targets for cancer therapy, but have long been considered undruggable mainly due to challenges in nuclear drug delivery and physically targeting myc directly given that it is a largely disordered protein that lacks discernible clefts and pockets for small molecules to inhabit. Furthermore, c-myc is important in normal tissue maintenance so the effect of its inhibition in humans is difficult to predict. However, recent in vivo studies showed that systemic myc inhibition (using the peptide pan myc inhibitor Omomyc) has mild and reversible side effects and induces tumour regression. This has alleviated concerns about the side effects that myc inhibition might have, and reinforced the promise of myc as a powerful drug target. However, the translation of Omomyc into the clinic has been hindered by poor cellular delivery. In fact, no direct myc inhibitor has yet been approved, indicating that novel approaches are needed. Moreover, inhibitors in development tend to inhibit all myc family proteins. An inhibitor that could specifically target N-myc might improve safety through bypassing c-myc inhibition. This could be used for the treatment of N-myc-driven cancers such as MYCN-amplified neuroblastoma. Nanobodies, camelid-derived single-domain antibodies, are a relatively new drug class. Whilst some are already in clinical trials for a wide range of diseases, these are specific for cell-surface or extracellular targets. However, their properties also make them ideal for use as intracellular antibodies or ‘intrabodies’. For example, they are small (just 12-15 kDa) and highly soluble due to naturally occurring hydrophobic to hydrophilic amino acid substitutions. Their small size and convex shape makes them advantageous in capturing structures in intrinsically disordered proteins and allows them to reach hidden epitopes not accessible to conventional antibodies, which could improve biological activity. Importantly, nanobodies retain the high specificities and affinities of conventional antibodies. Their small, single-domain nature also means they can be engineered with ease to modify aspects of their localisation and/or function. For example, they can be coupled to carrier molecules to facilitate cellular entry, and a nuclear localisation signal (NLS) can be added to drive them into the nucleus. Also, it was recently shown that an F-box domain could also be incorporated into nanobodies to recruit degradation machinery to its antigen, which depletes the antigen from cells via the proteasomal degradation pathway. Due to their highly advantageous properties, nanobodies raised against N-myc might overcome the barriers to targeting N-myc, providing potent and specific means of directly inhibiting N-myc therapeutically, which has not yet been achieved. In this thesis, nine unique nanobodies were raised against N-myc. These included three against the basic helix-loop-helix leucine zipper (bHLH-LZ) domain where Max dimerises, and six against the transactivation domain where numerous regulatory and cofactor proteins bind, such as the E3 ubiquitin ligase Skp2. Nanobodies against the transactivation domain were more specific for N-myc and were shown to inhibit its Skp-2-mediated ubiquitylation. This could provide novel means of eradicating tumours based on a study showing that inhibition of ubiquitylation at this domain triggers a transcriptional ‘switch’ that induces a non-canonical target gene Egr1, leading to p53-independent apoptosis. A nanobody against the bHLH-LZ (Nb C2) was shown to bind both N- and c-myc to similar magnitudes. Its affinity for N-myc bHLH-LZ was superior to that of the small molecule myc inhibitor 10058-F4, which prolongs survival in a MYCN-dependent mouse model of high-risk neuroblastoma. Nb C2 spontaneously transduced cell membranes and its coupling to a novel small molecule carrier (SMoC) enhanced its cellular uptake. Furthermore, the addition of a NLS increased its nuclear localisation. Preliminary experiments showed that Nb C2 might slow proliferation and induce apoptosis in cancer cell lines expressing c-myc, suggesting that Nb C2 might also be effective against cancers characterised by deregulated c-myc. Taken together, data generated in this thesis have revealed intriguing findings that provide a basis for the development of these nanobodies for the treatment of N-myc- and c-myc-driven cancers.

Thesis (Ph. D.)--University of Washington, 2006. / Vita. Includes bibliographical references (leaves 92-109).

Made available in DSpace on 2015-07-22T20:50:35Z (GMT). No. of bitstreams: 0 Previous issue date: 2010-04-28 / Meduloblastoma (MB) é o tumor maligno de sistema nervoso central (SNC) mais comum em criança, compreendendo 20% dos tumores primários de SNC e 40% dos tumores cerebelares da infância. Devido sua forte tendência metastática, o tratamento padrão pós-operatório inclui radio e quimioterapia, cujo impacto causa distúrbios endócrinos e de crescimento, e disfunção neurocognitiva a longo prazo. Frente a esses efeitos negativos, muitas pesquisas em meduloblastoma têm sido realizadas com intuito de obter conhecimento biológico desses tumores para tentar identificar fatores prognósticos moleculares que possam orientar os tratamentos, tornando-os mais específicos e menos agressivos. Alguns estudos em MB têm sugerido que a expressão do oncogene MYC está associada com diminuição da sobrevida e sua superexpressão com maior agressividade do tumor. Por isso, MYC pode ser um indicador importante de prognóstico, além de modulador do comportamento desta doença. Enquanto o gene MYC é expresso em uma variedade de tecidos, a expressão de MYCN, outro membro da família MYC, é restrita a estágios precoces do desenvolvimento embrionário de alguns tecidos apenas, entre eles, o sistema nervoso central e periférico, sendo um mediador importante dos efeitos de ativação na proliferação de células precursoras cerebelares. Dessa forma, quando a expressão de MYCN está desregulada, ela aumenta a tumorigenicidade dessas células podendo dar origem ao MB. Além disso, o gene MYC também é considerado importante regulador da transcrição TERT, gene que codifica uma subunidade catálica de da telomerase, enzima importante para carcinogênese e imortalização de células neoplásicas. A atividade anormal da telomerase está presente em 90% dos cânceres e o aumento de sua atividade está associado a eventos clínicos desfavoráveis. Outro gene importante é o ASPM (abnormal spindle-like microcephaly associated) que desempenha função fundamental na neurogênese e proliferação celular durante o desenvolvimento cerebral. Esse gene codifica uma proteína de centrossomo e fuso mitótico que permite a divisão celular simétrica em células neuroepiteliais durante o desenvolvimento e aumento do tamanho cerebral. Alterações em ASPM é a causa mais comum de microcefalia primária em humanos e de falha de segregação, induzindo a aneuploidias e instabilidade genética. Além desses genes, outro gene estudado recentemente, como alvo em xv imunoterapia, é o gene PRAME que codifica um antígeno tumoral que está presente em vários tumores, incluindo meduloblastoma. O gene PRAME possui baixa ou ausência de expressão em tecidos normais, por isso é pode ser um forte candidato como alvo em imunoterapia, que é um tratamento menos tóxico. OBJETIVOS: O objetivo desse estudo foi investigar a expressão dos genes MYC, MYCN, TERT, ASPM e PRAME em fragmentos tumorais de meduloblastoma de crianças e tentar correlacionar com os parâmetros clínicos e verificar se há correlação de MYC, MYCN, TERT entre si, uma vez que estão correlacionados. MÉTODOS: Análise de expressão gênica foi realizada através de PCR quantitativa em tempo real, utilizando sistema SYBR Green, em 37 amostras tumorais de crianças, com média de idade de 8 anos. Para comparação de perfil de expressão foi usada duas amostra de cérebro normal. A análise estatística foi realizada nos programas Graph Pad Prism 4 e VassarStats RESULTADOS: Todas nossas amostras superexpressaram o gene MYCN com valor de quantificação relativa (RQ) mediana igual a 31 com p=0.001; assim como, todas nossas amostras também superexpressaram o gene ASPM com mediana igual a 586, p<0.0001. Do total de amostras, 95%, 81% e 84% superexpressaram TERT, MYC e PRAME respectivamente, sendo os valores de RQ (mediana) iguais a 322, p=0.01; 9.2, p<0.0001; 33, p<0.0001. Apesar da elevada expressão dos genes estudados na maioria das amostras estudadas, houve apenas correlação estatística entre a superexpressão de MYCN (p=0.008) e os pacientes que foram a óbito, e de TERT e os pacientes que recidivaram (p=0.0431). Não encontramos outra correlação estatística entre a superexpressão dos genes e as características clínicas dos pacientes. CONCLUSÃO: Os genes MYC, MYCN e TERT estavam superexpressos nas amostras de meduloblastoma analisadas em uma freqüência muito superior ao demonstrado na literatura, o que sugere que esses três genes podem ajudar na identificação de tumores agressivos, uma vez que o pognóstico desses pacientes continua baseado apenas em parâmentros clínicos. A superexpressão de ASPM em todas as amostras estudadas sugere que este gene pode estar envolvido na origem de MB, como parte da neurogênse anormal durante o desenvolvimento embrionário, porém estudoas funcionais devem ser realizados para confirmar essa hipótese. Por fim, o gene PRAME pode ser candidato à marcador de célula tumoral em MB, podendo no futuro ser candidato como alvo em imunoterapias. / To investigate the expression of genes MYC, MYCN and TERT in tumor fragments of pediatric medulloblastoma and correlate gene expression profiles with clinical parameters. Analysis of gene expression was performed by quantitative PCR real time in 37 tumor samples and correlated with clinical and pathological data. All 37 samples overexpressed MYCN gene (p= 0.001), 95% and 84% of the samples overexpressed TERT and MYC, respectively (p<0.0001). Twenty nine (78%) of all samples had concomitant high expression of MYC, MYCN and TERT genes together. Seventeen (59%) were high-risk classification, 10 (34%) were metastatic (M+) stage, two (7%) were anaplastic or largecell/ anaplastic subtype, eight (28%) of patients relapsed, beyond thirteen (45%) suffered partial surgical resection. and fourteen (48%) died. We found correlation between MYC, MYCN and TERT expression (p<0.0001). The identification of a subgroup with concomitant overexpression of the three investigated genes suggests the possibility of using more than one aspect of molecular indicative of unfavorable prognosis that characterizes the group with poor outcome. However, in future this may be enhanced by targeted therapy for the product TERT as proposed in some neoplasms. The identification of molecular events in the medulloblastoma categorization aims to help at-risk groups moving towards individualized medicine. / TEDE / BV UNIFESP: Teses e dissertações

O neuroblastoma é o tumor sólido extra-cranial mais comum e mortal da infância, sendo o tempo de sobrevida nos casos mais agressivos ainda muito curto. Uma das esperanças nesses casos é que os estudos moleculares possam fornecer informações sobre os genes ou as vias moleculares que governam a patogênese dos neuroblastomas. Pois, há poucos genes como o MYCN, que foi descrito por estar diretamente ligado ao neuroblastoma. A amplificação deste oncogene ocorre em pouco mais de 25% dos neuroblastomas e é considerada como o mais importante marcador de prognóstico nestes tumores, sendo fortemente relacionada aos estádios avançados da doença e falha no tratamento. Outros genes do amplicon do MYCN, incluindo o DDX1 \"DEAD box polypeptide 1 gene\" e o NAG \"neuroblastoma-amplified gene\", estão sendo observados por se apresentarem co-amplificados com o MYCN. Entretanto, a importância deste fenômeno no prognóstico ainda é desconhecida. Os objetivos deste trabalho foram determinar qual o melhor método para estudar a amplificação dos genes MYCN, DDX1 e NAG, além de esclarecer a importância da coamplificação dos genes DDX1 e NAG no prognóstico. Procedimento: O número de cópias dos genes MYCN, DDX1 e NAG foi determinado por PCR em Tempo Real e PCR convencional em 100 neuroblastomas primários. Os dados da PCR em Tempo Real foram analisados por quantificação absoluta e relativa. Os resultados da PCR convencional foram analisados por eletroforese em gel de agarose, medindo a intensidade das bandas formadas no gel no sistema Kodak. A relevância da amplificação gênica como marcador de prognóstico foi avaliada em 74 pacientes, dos quais nós obtivemos o acompanhamento clínico. Resultados: Nos 74 casos estudados, ambos os métodos demonstraram que a amplificação do MYCN estava associada com os estádios mais avançados da doença. A análise das curvas de sobrevida livre de progressão confirmou que pacientes com ausência de amplificação do MYCN apresentavam maior tempo de sobrevida. Nós também analisamos a amplificação do DDX1 nas mesmas amostras incluindo aquelas com ausência de amplificação de MYCN. Não foi encontrada nenhuma relação entre a co-amplificação com idade ao diagnóstico ou tempo de sobrevida. Conclusões: Os métodos aplicados para calcular o número de cópias dos genes na PCR em Tempo Real mostraram-se equivalentes. A PCR em Tempo Real apresentou maior acurácia nos resultados quando comparada à PCR convencional. A análise da sobrevida não demonstrou relação entre a amplificação dos genes DDX1 e/ou NAG com piora no prognóstico. / Neuroblastoma is the most common and deadly extra-cranial solid childhood tumor. Survival rates for aggressive neuroblastomas are still disappointingly low. One of the hopes is that molecular studies will provide insights into the genes and molecular pathways that govern neuroblastoma pathogenesis. However, at present only a few genes as MYCN have been directly linked to neuroblastoma. MYCN oncogene amplification, occurring in up to 25% of neuroblastomas, has been considered the most important prognostic factor, strongly correlating to advanced stage disease and treatment failure. Another genes in the MYCN amplicon, including the DEAD box polypeptide 1 (DDX1) gene, and neuroblastoma-amplified gene (NAG gene), have been found to be frequently co-amplified with MYCN in NB. But the prognostic significance of the coamplification remains unclear. The aims of this study were to evaluate which is the best method to study the gene amplification of those three genes MYCN, DDX1 and NAG, as well as clarify the prognostic significance of the co-amplification or DDX1 and NAG with MYCN. Procedure: The gene copy numbers of MYCN, DDX1, and NAG were determined by the real-time quantitative polymerase chain reaction and conventional polymerase chain reaction in 100 primary NBs. Real-Time data were analyzed by absolute and relative quantification. For conventional PCR, samples were electrophoresed on a 2% agarose gel and the intensity of each band evaluated by Kodak image software. To evaluate of the prognostic significance of the gene amplification we had only 74 cases in witch we could analyze the follow-up. Results: In all 74 cases, both methods demonstrated that MYCN amplification was associated mainly with advanced cancer stages, and the analysis of overall survival confirmed that patients without MYCN amplification had a cumulative survival significantly higher than patients with oncogene amplification. We also studied DDX1 and NAG amplification for all NB samples even that without MYCN amplification. No relationship between any gene co-amplification status and disease stage, age at diagnosis, or overall survival was found. Conclusions: The two methods used to calculate gene copy number for Real Time PCR assay shown to be equivalent. Real Time PCR assay shown to be more accurate to study gene amplification than conventional PCR assay. Survival analysis pointed out that DDX1 and/or NAG amplification has no additional adverse effect on prognosis.

Les modèles animaux actuellement disponibles pour l’étude des leucémies humaines ne sont pas adaptés pour le développement optimal de nouvelles thérapies ciblées. Au cours de ce projet, qui s’inscrit dans une double perspective fondamentale et industrielle, nous avons cherché à générer un modèle versatile de leucémogénèse humaine chez la souris humanisée BRGS (BALB/c Rag2-/- IL-2Rγc-/- SIRPα.NOD). Les animaux sont greffés avec des progéniteurs hématopoïétiques transduits par des vecteurs lentiviraux surexprimant les oncogènes MYC et BCL2 et placés sous le contrôle d’un promoteur ubiquitaire (EF1α ou SFFV). Le suivi longitudinal des animaux sur une période de 5 mois montre que seule la construction SFFV/Myc-T2A-Bcl2 entraîne la transformation des progéniteurs hématopoïétiques. Entre 12 à 14 semaines post-greffe, >90% des animaux développent des lympho-proliférations de type pro-B (CD19+CD10+CD9+CD20-cytIgM-) infiltrant principalement la rate et la moelle osseuse, et circulant en abondance dans le sang. Le caractère transmissible des tumeurs est validé par des greffes secondaires de tumeurs spléniques. Les cultures in vitro de progéniteurs hématopoïétiques suggèrent que l’émergence des blastes est liée à la réactivation d’un programme B latent dans les précurseurs T, dont le développement est bloqué. Nous avons développé en parallèle un modèle de tumeur autologue. L’ensemble de ces résultats valide le modèle de leucémogénèse humaine développé chez la souris humanisée BRGS et ouvre des perspectives pour la caractérisation fonctionnelle des mécanismes de leucémogénèse, et la validation préclinique de nouvelles stratégies anti-tumorales. / Existing animal models for the study of human leukemia are not accurate for the proper development of innovative, targeted therapies. The aim of this project, which contains both a fundamental and an industrial perspective, therefore was to develop a new, versatile model of human leukemogenesis in the BRGS (BALB/c Rag2-/- IL-2Rγc-/- SIRPα.NOD) humanized mouse. Animals are grafted with hematopoietic progenitors transduced with lentivirals vectors to allow overexpression of MYC and BCL2 proteins under the control of an ubiquitous promotor (EF1α or SFFV). Longitudinal monitoring of the animals over five months shows that only the SFFV/Myc-T2A-Bcl2 construction induces the transformation of humans hematopoietics progenitors. Between 12 and 14 weeks post-transplantation, more than 90% of the animals develop pro-B lympho-proliferations (CD19+CD10+CD9+CD20-cytIgM-), with tumor cells being mainly found in the spleen, the bone marrow and in blood. Tumor transferability is achievable through secondary transplantation in immunodeficient mouse recipients. In vitro culture of bone marrow T cell progenitors suggest that the blasts arise from these cells after reactivation of a latent B cell program with blockade of their T cell development. In parallel, we have also developed an autologous tumor model. Altogether, these results validate the human leukemogenesis model constructed here in humanized BRGS mice and provide attractive prospects regarding the functional characterization of leukemogenesis and a preclinical validation of new anti-tumor strategies.

c-MYC est un facteur de transcription oncogénique dont l’expression est dérégulée dans 78% des gliomes. On observe d’ailleurs une corrélation positive entre sa surexpression et le grade des gliomes. De plus, cette surexpression serait essentielle à la survie des cellules souches tumorales, cellules qui seraient davantage résistantes à la chimiothérapie et à la radiothérapie en plus d’avoir un caractère plus invasif. Il a aussi été démontré que l’inhibition de c-MYC par ARN interférents peut sensibiliser les cellules cancéreuses à l’apoptose et réduire leur prolifération. Sa surexpression relative dans les glioblastomes (GBM) est signe de la malignité et l’espérance de vie des patients atteints par ces tumeurs est réduite chez les patients plus âgés. c-MYC doit s’hétérodimériser avec MAX, son partenaire obligatoire afin de se lier aux promoteurs de ses gènes cibles contenant des EBox (CANNTG) et ainsi activer leur transcription. Cependant, il a été proposé que MAX pourrait homodimériser et agir comme antagoniste en compétitionnant pour les mêmes sites de reconnaissance que l’hétérodimère c-MYC/MAX sur l’ADN. Notre étude vise donc à évaluer l’effet dose-dépendant d’un traitement exogène de MAX*WT, correspondant à une version tronquée du facteur de transcription MAX, sur différentes lignées cellulaires de GBM. Nous avons d’abord étudié les capacités de la protéine à transloquer dans les cellules par microscopie. Ceci a permis de déterminer que le peptide s’internalise rapidement (15 minutes) pour ensuite s’accumuler au niveau nucléaire (24 h, 48 h). Par la suite, des analyses de FACScan ont démonté qu’un traitement de 72 heures provoque une inhibition de la prolifération cellulaire. À l’aide de chambres de Boyden et d’essais de croissance en sphéroïdes dans une matrice de Matrigel(indice supérieur TM]), nous avons observé une diminution importante du caractère invasif des lignées de gliomes malins suite au traitement avec MAX*WT. Ces résultats démontrent que la protéine MAX*WT semble avoir un effet antinéoplasique sur plusieurs lignées de gliomes malins et que la voie de signalisation de c-MYC pourrait constituer une cible thérapeutique intéressante.

Dans le monde actuel, il demeure encore que le cancer colorectal soit une des premières causes de mortalité par cancer. Le développement et la progression de ce cancer fait intervenir des événements complexes comme les mutations oncogéniques mais aussi la surexpression de différents récepteurs membranaires dont les intégrines. Dans l'intestin grêle, l'intégrine α1β1 est exprimée uniquement par les cellules prolifératives de la crypte. Ce profil supporte le rôle rapporté pour cette intégrine dans le recrutement de la cavéoline-1 et l’activation en aval, via Shc, de la voie proliférative Ras/ERK dans les fibroblastes. L’expression de la sous-unité α1 n’a pas été décrite dans le tissu colique normal et tumoral. Chez la souris, l'intégrine α1β1 soutient la motilité des cellules tumorales mammaires et, de concert avec le facteur oncogénique Kras, elle potentialise la croissance et l’invasion tumorales. Ces rôles ne sont pas connus dans le cancer colorectal. L’objectif dans cette étude est d’analyser l’expression de la sous-unité α1 dans le cancer colorectal, de définir les facteurs de régulation de son expression et d’identifier les rôles de l’intégrine α1β1 dans la progression du cancer colorectal. Mes expériences de recherche ont permis de montrer que la sous-unité α1 est présente dans 65% des adénocarcinomes colorectaux et que son expression est observée non seulement dans les cellules épithéliales mais aussi dans les cellules du microenvironnement de la tumeur. Mes résultats ont contribué à l’identification de la sous-unité α1 comme une nouvelle cible directe de l’oncogène MYC dans le cancer colorectal. J’ai démontré que MYC se lie directement au promoteur du gène ITGA1 au niveau de deux éléments de réponse spécifiques. Ce contrôle s’est reflété par une corrélation significative entre l’expression des deux protéines MYC et α1 dans 72% des adénocarcinomes étudiés. J’ai pu également montrer pour la première fois que l’intégrine α1β1 est importante pour la prolifération des cellules tumorales colorectales ainsi que pour leur migration et leur survie. De plus, en utilisant des modèles de xénogreffes, j’ai observé que cette intégrine est importante pour la croissance des tumeurs colorectales, et son absence induit également une importante nécrose tumorale. Mes découvertes apportent de nouvelles connaissances dans la compréhension de l’implication de l’intégrine α1β1 dans la carcinogenèse colorectale. Cette intégrine pourrait représenter une cible thérapeutique intéressante dans le traitement du cancer colorectal.

In this dissertation, we explore the transcriptional regulatory roles of Gquadruplex- forming motifs and the involvement of specific transcriptional factors, which interact with the same elements, in the control of human c-myc and bcl-2 gene expression. The G-quadruplex structures within the NHE III1 region of the c-myc promoter and their ability to repress transcription has been well established. However, a longstanding unanswered question is how these stable DNA secondary structures are transformed to activate c-myc transcription. NDPK-B has been recognized as an activator of c-myc transcription via interactions with NHE III1 region of the c-myc gene promoter. Through the use of RNAi, we confirmed the transcriptional regulatory role of NDPK-B. We demonstrate that NDPK-B has DNA binding activity and the nuclease activity results from a contaminating protein. NDPK-B preferentially binds to the singlestranded guanine-rich strand of the c-myc NHE III₁. Potassium ions and G-quadruplexinteractive agents, which stabilize G-quadruplex structures, had an inhibitory effect on NDPK-B DNA binding activity. Based on our studies, we have proposed a stepwise trapping-out of the NHE III1 region in a single-stranded form, thus allowing singlestranded transcription factors to bind and activate c-myc transcription. This model provides a rationale for how the stabilization of G-quadruplexes within the c-myc gene promoter region can inhibit NDPK-B from activating c-myc transcription. Similarly, the human bcl-2 gene contains a GC-rich region within its promoter region, which is critical in the regulation of bcl-2 expression. We demonstrate that the guanine rich strand within this region can form three intramolecular G-quadruplex structures. Based on NMR studies, the central G-quadruplex forms a mixed parallel/antiparallel structure with three tetrads connected by loops of one, seven, and three bases. The Gquadruplex structures in the bcl-2 promoter extends beyond the ability to form any one of three separate G-quadruplexes to each having the capacity to form either three or six different loop isomers. This suggests that targeting these individual structures could lead to different biological outcomes. We also found that Telomestatin upregulates bcl-2 gene expression, which we propose is a result of inhibiting the binding of the WT1 repressor protein by the formation of a drug-stabilized G-quadruplex structure.

The activated product of the c-MYC proto-oncogene is one of the strongest known activators of carcinogenesis. It has been estimated that as many as one-seventh of all cancer deaths are associated with alterations in the c-MYC gene or its expression [1]. Therefore, understanding the regulation of c-MYC expression is a key factor in understanding carcinogenesis in many histologic classes of malignancy. The nuclease hypersensitive element (NHE) III₁ region of the c-MYC promoter has been shown to be particularly important in regulating c-MYC expression. Specifically, the formation of a G-quadruplex structure appears to promote repression of c-MYC transcription. In this dissertation, we investigate the role that nucleolin, a critical player in ribosome biogenesis and cell stress sensing, plays on the transcriptional regulation of the c-MYC promoter through its interaction with the c-MYC G-quadruplex structure. Our studies initiated with the design of a c-MYC G-quadruplex affinity column intended to trap potential c-MYC G-quadruplex-binding proteins that were then identified by LC-MS/MS. After careful examination of the literature of the list of potential c-MYC G-quadruplexbinding proteins, we realized that several of the proteins identified had been previously reported to interact directly with nucleolin. Consequently, we chose to focus our studies on nucleolin, as it could be a central regulator of the (NHE) III region. By performing chromatin immunoprecipitation in HeLa cells, we found that nucleolin indeed interacts with the c-MYC promoter region containing the NHE III₁ element. This binding activity was confirmed by both electromobility shift assay and polymerase stop assay. We provide evidence that nucleolin can induce the formation of the c-MYC G-quadruplex structure from single-stranded DNA, both in linear and circular DNA forms. We show that upon binding, nucleolin increases the stability of the c-MYC G-quadruplex structure leading to repression of c-MYC promoter activity. We also show that nucleolin binds with much higher affinity to G-quadruplex structures with topology similar to that of the parallel c-MYC G-quadruplex, such as those found in the VEGF and PDGF-A promoters; in comparison to G-quadruplexes found in telomeres or the c-MYB promoter, whose have significantly different topology. Interestingly, we also demonstrate that nucleolin binds with higher affinity to the c-MYC G-quadruplex than to its consensus RNA substrate, the nucleolin recognition element (NRE). Furthermore, we show that the C-terminal domain of nucleolin is critical for its interaction and stabilization of the c-MYC G-quadruplex structure. Lastly, we show that the binding of nucleolin to the (NHE) III region causes repression of c-MYC transcription. On the basis of these results, we propose that nucleolin may play an important role in the transcriptional regulation of c-MYC in vivo by inducing the formation of the c-MYC G-quadruplex structure.