The main subject of my thesis is the investigation of mechanisms of glioma tumorigenesis associated with the recently identified mutations in isocitrate dehydrogenase. Gliomas account for 80% of primary brain cancers. They represent a diverse group of tumours, and are graded from I-IV based on histopathological features. Whilst grade I tumours may be curable with surgery alone, grade II and III gliomas inevitably progress to glioblastoma multiforme (GBM), which is highly resistant to current therapies and carries a very poor prognosis. Despite an improved understanding of the pathways and mechanisms involved in the development of glioma and its progression to grade IV disease, current and novel treatments have so far failed to significantly improve outcome. Isocitrate dehydrogenase (IDH) enzymes catalyse the oxidative decarboxylation of isocitrate to α-ketoglutarate (α-KG). Somatic mutations in genes encoding IDH1 and IDH2 were first identified in glioma and subsequently in acute myeloid leukemia and other solid tumours. These heterozygous point mutations occur at the arginine residue of the enzymes active site and cause both loss of normal enzyme function and gain-of-function, causing the reduction of α-KG to D-2-hydroxyglutarate (D-2HG), which accumulates. D-2HG may act as an oncometabolite through the inhibition of various α-KG dependent enzymes, stimulating angiogenesis, histone modifications and aberrant DNA methylation. Possibly, IDH1/2 mutations may also cause oncogenic effects through dysregulation of the tricarboxylic acid (TCA) cycle, or by increasing susceptibility to oxidative stress. The exact role of mutant IDH1/2 in tumorigenesis however remains unclear. In the work outlined in this thesis, I have demonstrated that the expression of mutant IDH1/2 in glioma cell lines leads to 2-HG accumultation and a reduction in α-KG production and results in HIF1α accumulation and a reduction in 5hmC production. Furthermore, the brain-specific expression of mutant Idh1 in mice also results in 2-HG accumulation and reduced α-KG production, whilst a reduction in 5hmC levels are also seen. This data appears to support the theory that IDH1/2 mutant activity results in the inhibition of α-KG dependent enzymes, either through the accumulation of 2-HG or due to a reduction in α-KG levels. The brain-specific expression of mutant Idh1 in mice also results in increased cellular proliferation and an increase in the expression of the neural stem cell marker, nestin. However gliomas do not develop, perhaps suggesting that additional mutations are required in conjunction with those occuring in IDH1/2 in order to initiate tumourigenesis. Clinically, IDH1/2 mutations may represent a novel therapeutic target in glioma and may also serve as useful diagnostic, prognostic and predictive biomarkers. However, a better understanding of the pathogenesis of mutant IDH is required, to enable effective IDH1/2 directed therapies to be developed in the future.

Les glioblastomes sont des tumeurs fatales du fait de leur agressivité et du manque de traitements efficaces. La prolifération accrue, le caractère invasif et la résistance à la mort cellulaire leur confèrent une croissance rapide et une invasion du parenchyme cérébral environnant, à l’origine de leur systématique récidive. Exprimée par la composante tumorale en hypoxie mais également par la composante vasculaire, l’AM participe de façon autocrine et paracrine au développement des glioblastomes en favorisant la croissance des cellules tumorales et l’angiogenèse tumorale.Il a été montré que des anticorps polyclonaux dirigés contre les récepteurs de l’AM inhibent in vitro la croissance, la migration et la formation de pseudo-capillaires des cellules endothéliales, suggérant une neutralisation par ces anticorps de certaines étapes de l’angiogenèse. De même, il a été montré in vivo que ces anticorps inhibent la croissance tumorale en supprimant l’angiogenèse et la croissance des cellules tumorales suggérant ainsi que les récepteurs de l’AM constitueraient une bonne cible thérapeutique. Des anticorps capables de reconnaître et neutraliser à la fois l’AM, les CLR, RAMP2 et RAMP3 agissant de la même manière sur la croissance tumorale et l’angiogenèse représenteraient un bénéfice thérapeutique majeur. Des anticorps dirigés contre un peptide chimérique constitué de l’enchainement de séquences peptidiques des protéines CLR, RAMP2, RAMP3 et du peptide AM sont en cours. Le traitement par ces anticorps diminue la croissance des cellules tumorales ainsi que leurs migration et invasion. Ces résultats très encourageants nous permettent pour le moment de valider la faisabilité du concept d’anticorps développés à partir d’un peptide chimérique pour neutraliser le système AM/AMR dans le but d’envisager dans le futur une application thérapeutique. / Glioblastoma are fatal tumors because of their aggressiveness and lack of effective treatments. The increased proliferation, the invasiveness and resistance to cell death gives them a rapid growth and invasion of brain parenchyma surrounding the origin of their systematic recurrence. Expressed by the tumoral component in hypoxia but also by the vascular component, the AM participate by an autocrine and paracrine way, the development of glioblastoma by promoting tumor ell growth and tumor angiogenesis.It was shown that polyclonal antibodies directed against the AM receptor inhibit in vitro growth, migration and the formation of pseudo-capillary of endothelial cells, suggesting neutralization by theses antibodies in certain stages of angiogenesis. Similarly, it has been shown in vivo that these antibodies inhibit tumor growth by suppressing angiogenesis and tumor cell growth, suggesting that the AM receptor would be a good therapeutic target. Antibodies that recognized and neutralized both the AM, the CLR, RAMP2 and RAMP3 acting the same way on tumor growth and angiogenesis represent a major therapeutic benefit. Antibodies against a chimeric peptide consisting of peptide sequence of CLR, RAMP2, RAMP3 and AM peptide are in progress. Treatment with these antibodies decreases the growth of tumor cells, their migration and invasion. These encouraging results allow us the time to validate the feasibility of the concept of antibodies developed from a chimeric peptide to neutralize the AM/AMR system in order to consider in the future therapeutic application.

Thesis advisor: Thomas Seyfried / Lipids, in all their forms from structural components of the membranes (phosphoglycerides, glycolglycerolipids) to signaling molecules (IP3, DAG, prostaglandins, etc.,) post-translational modification of proteins (palmitoylated, farnesylated, prenylated, and GPI anchoring) play an essential role in cancer cell survival, proliferation, and metastasis. Alteration in structural lipids can impair transport, and signaling cascades. Abnormalities in lipids, such as cardiolipin (Ptd2Gro), impair mitochondrial function, bioenergetics, and could play a role in precipitatting the high incidence of spontaneous tumors in VM/Dk mice. This thesis explores the role of glucose and glutamine in their incorporation into lipids in the VM-M3 murine glioblastoma cell line as well as the inheritance of brain cardiolipin fatty acids abnormalities in VM/Dk mice. I used labeled [14C]-U-D-glucose and [14C]-U-L-glutamine to examine the profile of de novo lipid biosynthesis in the VM-M3 cell line. The major lipids synthesized included phosphatidylcholine (PtdCho), phosphatidylethanolamine (EtnGpl), phosphatidylinositol (PtdIns), phosphatidylserine (PtdSer), sphingomyelin (CerPCho), bis(monoacylglycero)phosphate (BMP) / phosphatidic acid (PtdOH), cholesterol (C), Ptd2Gro, and the gangliosides. The data show that the incorporation of labeled glucose and glutamine into synthesized lipids was dependent on the type of growth environment, and that the VM-M3 glioblastoma cells could acquire lipids, especially cholesterol, from the external environment for growth and proliferation. In addition, this thesis also explores and evaluates the abnormality of Ptd2Gro fatty acid composition in VM mice in comparison to B6 mice. Although previously reported, I confirmed the finding in the abnormal cardiolipin fatty acid composition in the VM mice. The abnormal brain cardiolipin fatty acid composition was found to be inherited as an autosomal dominant trait in reciprocal B6 x VM F1 hybrids for both male and female. Impaired cognitive awareness under hypoxia observed for the VM mice and reciprocal F1 hybrids is associated with abnormalities in neural lipid composition. / Thesis (PhD) — Boston College, 2014. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Biology.

OBJECTIVE: Glioblastoma Multiforme (GBM), one of the most malignant forms of primary brain tumors, is characterized by its highly heterogenous genetic composition, aggressive infiltration of surrounding tissue, and resistance to current treatments. Gene expression analysis has characterized GBM into four main types, with a significant portion belonging to the Classical subtype, typified by overexpression and/or mutation of the epidermal growth factor receptor (EGFR). Also common to this subtype of GBM is the loss of crucial tumor suppressor genes Ink4A/ARF and PTEN, which contribute to the invasive nature and unregulated proliferation that underlie the GBM pathology. The high rate of tumor recurrence post treatment with surgical resection, chemotherapy, and radiation has driven the pursuit of more effective molecularly targeted therapies. This study was undertaken to determine the effects of two types of small molecule tyrosine kinase inhibitors on cells overexpressing wild-type EGFR in the context of their respective complements of tumor suppressor genes. METHODS: Several cell lines were established from mouse models of EGFR wild-type (EGFRWT) driven gliomagenesis and treated with 10 μM of type I tyrosine kinase inhibitors Gefitinib (Iressa®, Astra Zeneca), CI-1033 (Canertinib, Pfizer), or Dimethyl Sulfoxide vehicle. Cells were exposed to each drug treatment as part of a time course ranging from 0 to 24 hours and then evaluated by trypan blue exclusion and Western blot analysis for cell viability and molecular and biochemical effects respectively. RESULTS: Evaluation of cell viability indicated that CI-1033 caused a greater increase in cell death than gefitinib when compared to control treated cells regardless of the tumor suppressors lost. Gefitinib was found to cause cell death only in cells expressing the PTEN tumor suppressor whereas CI-1033 showed similar levels of cell death for cells deficient in Ink4A/ARF or both Ink4A/ARF and PTEN tumor suppressors. Western blot analysis revealed that CI-1033 more effectively inhibited EGFR compared to gefitinib. Treatment with both gefitinib and CI-1033 effectively blocked phosphorylation of EGFR, but this effect was less pronounced with gefitinib treatment. Further analysis of downstream signaling molecules showed a greater presence of cleaved caspase 3, a hallmark of apoptosis, in gefitinib treated cells expressing PTEN than in those cells treated with CI-1033. Cells deficient in both Ink4A/ARF and PTEN did not demonstrate any induction of cleaved caspase 3 following either treatment. CONCLUSIONS: Based on the significant differences in cell viability between treatments, CI-1033 is an overall more effective inhibitor of EGFRWT expressing cells lacking PTEN, while gefitinib and CI-1033 were found to be similarly effective in cells expressing PTEN. The results of western blot analysis indicate that total and irreversible EGFR inhibition may be necessary to induce cell death in a manner that effectively terminates downstream cell signaling. It is likely that CI-1033, unlike gefitinib, induces apoptosis in a caspase-independent manner, which may be one of the many differences in downstream effects produced by these two drugs. Further research is necessary to determine the extent to which each inhibitor shuts down proliferative cell signaling pathways such as PI3K-AKT and MEK-ERK signaling pathways downstream of EGFR. Overall, these data indicate that genotype plays an important role in the determination of therapeutic response and may aid in the evaluation of clinical prognoses.

O glioblastoma (GBM) é o tumor cerebral primário mais comum e o mais grave tumor de células da glia. O GBM é um tumor astrocítico de grau IV caracterizado pela proliferação descontrolada, infiltrado difuso, tendência à necrose, angiogênese, resistência a apoptose e grande heterogeneidade genética. Apesar da terapia abranger a remoção cirúrgica máxima, a radioterapia e a quimioterapia, o tumor torna-se resistente à drogas utilizadas no tratamento levando o paciente a recorrência e morte em menos de 15 meses após o diagnóstico. Uma alternativa para o tratamento do GBM é a imunoterapia, a qual é capaz de estimular o sistema imunológico do próprio paciente a gerar uma resposta específica e duradoura que pode proteger contra a recorrência da doença. Uma dessas alternativas envolve o uso de vacinas de DNA codificando antígenos tumorais e proteínas imunomoduladoras capazes de ativar eficientemente linfócitos B e T específicos aos antígenos presentes no tumor. Nesse contexto, o objetivo do presente trabalho foi construir vacinas de DNA utilizando-se os genes dos antígenos EGFRvIII, cERBB2, MAGE e GLEA de GBM e os genes das proteínas imunomoduladoras hsp65, hsp70, gp96 e gD e avaliar suas respectivas imunogenicidades. Os genes foram avaliados in silico, sintetizados in vitro e utilizados na construção das vacinas de DNA. Ferramentas de biologia molecular e o vetor pVAX foram utilizados para obtenção das vacinas. Elas foram caracterizadas por sequenciamento e western blot e utilizadas na imunização de camundongos C57BL/6. As imunizações foram realizadas com três doses em intervalos de 12 dias combinando um antígeno tumoral e uma proteína imunomoduladora na forma de vacina de DNA. A imunogenicidade foi avaliada 20 dias após a última dose. Os ensaios ex vivo foram realizados com o soro dos animais imunizados para dosagem de anticorpos específicos contra os antígenos tumorais e com as células do baço que foram re-estimuladas com as proteínas EGFRvIII, cERBB2, MAGE e GLEA para identificar a presença de células específicas aos antígenos tumorais. Como resultado, a vacina pVAXgDGLEA foi a única capaz de induzir anticorpos do subtipo IgG2a anti-GLEA. As vacinas pVAXgDGLEA, pVAXgDEGFRvIII e pVAXgDMAGE foram capazes de ativar células específicas aos antígenos que após o re-estímulo responderam rapidamente com produção de IFN-g e IL-10. A proteína imunomoduladora gD foi, portanto, capaz de ajudar na indução de um padrão de resposta Th1, específica aos antígenos de GBM, importante no combate ao tumor e a IL-10 pode favorecer e/ou balancear a resposta no cérebro que deve ser eficaz, mas não exacerbada. / Glioblastoma multiforme (GBM) is the most common form of primary brain cancer and the most severe tumour affecting glia cells. GBM is a grade IV astrocytoma known by uncontrolled proliferation, diffused infiltrate, necrosis tendency, angiogenesis, apoptosis resistance and a wide genetic heterogeneity. The standard of care consists of maximal surgical resection, followed by a combination of radiation and chemotherapy. Despite that, tumour becomes resistant to drugs used to treatment, and the patient experiences recurrence followed by death in less than 15 months after diagnosis. An alternative in GBM treatment could be immunotherapy which aims to stimulate patients immunological system in order to obtain a specific and long-term response that can protect against recurrence. One of these alternatives involves the use of DNA vaccines codifying tumoral antigens and immunomodulatory proteins that can effectively activate tumour antigen specific B and T lymphocytes. In this context, the objective of this work was the construction of DNA vaccines using GBM antigen genes (EGFRvIII, cERBB2, MAGE e GLEA) and immunomodulatory proteins (hsp65, hsp70, gp96 e gD), followed by their immunogenicity evaluation. Genes were evaluated in silico, synthesized in vitro and used in DNA vaccines construction. Molecular biology tools and the pVAX vector were used to obtain the vaccine. They were characterized by sequencing, western blot and were used in the immunization of C57BL/6 mice. Immunizations were performed in 3 doses of a DNA vaccine combining a tumoral antigen and an immunomodulatory protein at each 12 days. Immunogenicity was evaluated 20 days after the last dose. The ex vivo assays were performed with the serum of immunized animals for antibody evaluation and spleen cells were stimulated with EGFRvIII, cERBB2, MAGE e GLEA proteins to assess tumoral antigen specific cells. The pVAXgDGLEA vaccine was the only able to induce IgG2a subtype anti-GLEA antibodies. Vaccines pVAXgDGLEA, pVAXgDEGFRvIII e pVAXgDMAGE were able to activate antigen-specific cells that produced IFN-g e IL-10 quickly after reestimulation. The gD immunomodulatory protein was able to induce a Th1 immune response, specific to GBM antigens, which is important in tumor combat while IL-10 could favor and/or balance the response in brain, which should be effective but not exacerbated.

by Xin-xia Tian. / Thesis (Ph.D.)--Chinese University of Hong Kong, 1999. / Includes bibliographical references (p. 195-212). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Mode of access: World Wide Web. / Abstracts in English and Chinese.

Glioblastoma multiforme (GBM) is an aggressive Grade IV astrocytoma with a poor survival rate. This is largely due to the GBM tumor cells migrating away from the primary tumor site along white matter tracts and blood vessels leading to secondary tumor sites. It is unknown whether the microenvironment nanotopography influences the biomechanical properties of the tumor cells. Although these tumor cells have an innate propensity to migrate, we believe that the nanotopography changes the biomechanical properties to enhance the migratory phenotype. To study this, we used an in vitro polycaprolactone aligned nanofiber film that mimics the nanotopography of the white matter tracts and blood vessels to investigate the mechanical properties of the GBM tumor cells. Our data demonstrate that the cytoskeletal stiffness, traction force, and focal adhesion area are inherently lower in invasive GBM tumor cells compared to healthy astrocytes. Moreover, the tumor cytoskeletal stiffness was significantly reduced when cultured on the aligned nanofiber films compared to smooth and randomly aligned nanofibers films. Analysis of gene expression also showed that tumor cells cultured on the aligned nanotopography upregulated key migratory genes and downregulated key proliferative genes. In addition, cell cycle analysis exhibited a reduced proliferative state on aligned nanofibers, highlighting the dichotomy between proliferation and migration observed in GBM. Finally, focal adhesions of tumor cells were larger and more elliptical when grown on the aligned fibers, suggesting a more migratory state. Therefore, our data demonstrate that the invasive potential is elevated when the tumor cells are cultured on an aligned nanotopography. This in vitro model can further be used to identify the GBM tumor cells’ response in a mimetic in vivo tumor microenvironment and elucidate how the aligned nanotopography transduces into altered gene and protein expression, thus providing a mechanism to target to inhibit the enhanced migratory behavior observed in these cells.

Glioblastoma is a highly malignant tumor which mostly recurs locally around the resected contrast enhancement. However, it is difficult to identify tumor invasiveness pre-surgically, especially in non-enhancing areas. Thus, the aim of this thesis was to utilize multimodal MR technique to identify and characterize the peritumoral progression zone that eventually leads to tumor progression. Patients with newly diagnosed cerebral glioblastoma were included consecutively from our cohort between 2010 and2014. The presurgical MRI sequences included volumetric T1-weighted with contrast, FLAIR, T2-weighted, diffusion-weighted imaging, diffusion tensor and perfusion MR imaging. Postsurgical and follow-up MRI included structural and ADC images. Image deformation, caused by disease nature and surgical procedure, renders routine coregistration methods inadequate for MRIs comparison between different time points. Therefore, a two-staged non-linear semi-automatic coregistration method was developed from the modification of the linear FLIRT and non-linear FNIRT functions in FMRIB’s Software Library (FSL). Utilising the above mentioned coregistration method, a volumetric study was conducted to analyse the extent of resection based on different MR techniques, including T1 weighted with contrast, FLAIR and DTI measures of isotropy (DTI-p) and anisotropy (DTI-q). The results showed that patients can have a better clinical outcome with a larger resection of the abnormal DTI q areas. Further study of the imaging characteristics of abnormal peritumoural DTI-q areas, using MRS and DCS-MRI, showed a higher Choline/NAA ratio (p = 0.035), especially higher Choline (p = 0.022), in these areas when compared to normal DTI-q areas. This was indicative of tumour activity in the peritumoural abnormal DTI-q areas. The peritumoural progression areas were found to have distinct imaging characteristics. In these progression areas, compared to non-progression areas within a 10 mm border around the contrast enhancing lesion, there was higher signal intensity in FLAIR (p = 0.02), and T1C (p < 0.001), and there were lower intensity in ADC (p = 0.029) and DTI-p (p < 0.001). Further applying radiomics features showed that 35 first order features and 77 second order features were significantly different between progression and non-progression areas. By using supervised convolutional neural network, there was an overall accuracy of 92.4% in the training set (n = 37) and 78.5% in the validation set (n=14). In summary, multimodal MR imaging, particularly diffusion tensor imaging, can demonstrate distinct characteristics in areas of potential progression on preoperative MRI, which can be considered potential targets for treatment. Further application of radiomics and machine learning can be potentially useful when identifying the tumor invasive margin before the surgery.

Glioblastoma (GB) é o tumor cerebral humano mais frequente e maligno. O prognóstico dos pacientes com GB permanece alarmante, principalmente devido à baixa eficácia das estratégias terapêuticas atuais além da natureza invasiva desse tipo de câncer. Uma característica de tumores sólidos é apresentar áreas hipóxicas. Microambientes hipóxicos contribuem para a progressão do câncer, resistência ao tratamento e ao prognóstico ruim da doença. Dessa forma, neste trabalho, desenvolvemos um modelo in vitro que se aproxima ao microambiente hipóxico tumoral in vivo. Nossos resultados sugerem que a privação de oxigênio (PO) em combinação com ausência de soro forneceu um ambiente favorável à desdiferenciação das células C6 em células tronco tumorais (CTT). A doxazosina (DOX), um antihipertensivo utilizado na clínica, apresenta efeitos antitumorais em diversos tipos de câncer. Assim, avaliamos o efeito antitumoral da doxazosina em linhagens de glioma de rato (C6) e humano (U138-MG) e a toxicidade do fármaco em culturas primárias de astrócitos e culturas organotípicas de hipocampo. A doxazosina induziu morte celular nas linhagens de glioma e apresentou baixa neurotoxicidade. Além disso, o fármaco inibiu a via da PI3K/Akt e ativou GSK-3β e p53, resultando em indução de apoptose e parada no ciclo celular na fase G0/G1. Considerando a importância da mitocôndria na plasticidade de células tumorais e a resistência ao tratamento apresentada pelos GB, nós analisamos os efeitos da DOX nas interações entre núcleo e mitocôndrias. A DOX induziu biogênese mitocondrial e apoptose nas células C6 e diminuiu secreção de TNF-α. Ao analisar a estrutura química da DOX, encontramos diversas características que indicam que ela possui autofluorescência. Dessa forma, caracterizamos a autofluorescência da DOX em diversos meios. Observamos que há um padrão de distribuição do fármaco nas células C6: ele se encontra ao redor do núcleo e parece estar vesiculado. A superexpressão do receptor do fator de crescimento epidermal (EGFR) está relacionada às formas mais malignas e resistentes de GBs. Portanto, analisamos se a ação antiglioma da DOX está envolvida com EGFR. O tratamento com DOX foi capaz de diminuir os níveis de p-EGFR. O co-tratamento de DOX e AG1478 (inibidor de receptores de tirosina cinase) diminuiu a fosforilação de EGFR e causou necrose. Em vista disso, nossos resultados sugerem que o mecanismo de ação da DOX envolve sinalização de EGFR. Por fim, nós avaliamos os efeitos da DOX livre e nanoencapsulada (DOX-NC) em modelos in vitro e in vivo de glioma. Demonstramos que a DOX-NC induziu morte celular em linhagem de glioma em concentrações 100 vezes menores do que as que utilizamos para a DOX na sua forma livre. Além disso, analisamos o efeito da DOX-NC em culturas organotípicas de hipocampo e, da mesma maneira que o observado com a DOX, a DOX-NC demonstrou baixa toxicidade e diminuiu a área tumoral in vivo, sendo seletiva para células cancerosas. Nesta tese, alteramos o microambiente in vitro de células de glioma, avaliamos os efeitos antitumorais da doxazosina em sua forma livre e nanoencapsulada. Além disso, utilizamos modelos de glioma in vitro e in vivo e em diversos parâmetros celulares, descrevemos a autofluorescência do fármaco e também utilizamos essa característica para avaliar a captação e a distribuição da doxazosina em células de glioma. Nossos resultados contribuíram para aproximar os modelos de estudo com o que ocorre em gliomas in vivo e para evidenciar o potencial terapêutico da doxazosina como um agente antiglioma com baixa toxicidade neural e sistêmica. / Glioblastoma (GB) is the most frequent and malignant human brain tumor. The prognosis of patients with GB remains dismal, mainly due to the low effectiveness of current therapeutic strategies and the invasive nature of this type of cancer. A characteristic of solid tumors is the presence of hypoxic areas. Hypoxic microenvironments contribute to cancer progression, resistance to treatment, and poor prognosis of the disease. Thus, we developed an in vitro model that approximates the hypoxic tumor microenvironment found in vivo. Our results suggest that OD in combination with absence of serum provided an environment favorable to the dedifferentiation of C6 cells in cancer stem cells. Doxazosin (DOX), an antihypertensive used in the clinic, has antitumor effects in several types of cancer. Thus, we evaluated the antitumor effect of DOX on rat (C6) and human (U138-MG) glioma cell lines and drug toxicity in primary astrocyte cultures and organotypic hippocampal cultures. DOX induced cell death in glioma lines and showed low neurotoxicity. In addition, the drug inhibited the PI3K/Akt pathway and activated GSK-3β and p53, resulting in induction of apoptosis and cell cycle arrest in the G0/G1 phase. Considering the importance of mitochondria in the plasticity of tumor cells and the resistance to treatment presented by glioblastomas, we analyzed the effects of DOX the interactions between nucleus and mitochondria. DOX induced mitochondrial biogenesis and apoptosis in C6 cells, and decreased TNF-α secretion. When analyzing the chemical structure doxazosin, we find several characteristics that indicate that it has autofluorescence. Thus, we characterized the autofluorescence of DOX in several media. We observed that there is a pattern of distribution of the drug in the cell: it is around the nucleus and appears to be vesiculated. Overexpression of the Epidermal Growth Factor Receptor (EGFR) is related to the more malignant and resistant forms of GBs. Therefore, we analyzed whether the antiglioma action of DOX is involved with EGFR. DOX treatment was able to decrease p-EGFR levels. Co-treatment of DOX and AG1478 (a receptor tyrosine kinase inhibitor) decreased EGFR phosphorylation and caused necrosis. Thus, our results suggest that the mechanism of action of doxazosin involves EGFR signaling. We evaluated the effects of free and nanoencapsulated doxazosin (DOX-NC) in in vitro and in vivo models of glioma. We demonstrated that nanoencapsulated doxazosin (DOX-NC) induced cell death in a glioma line at concentrations 100 times lower than those used for doxazosin in its free form. In addition, we analyzed the effect of DOX-NC on organotypic hippocampal cultures and, in the same way as observed with DOX, DOX-NC demonstrated low toxicity and decreased tumor area in vivo, being selective for cancer cells. In this dissertation, we altered the in vitro microenvironment of glioma cells, we evaluated the antitumor effects of doxazosin in its free and nanoencapsulated form. In addition, we used glioma models in vitro and in vivo and analyzed several cellular parameters, we described the autofluorescence of the drug and also used this characteristic to evaluate the uptake and distribution of doxazosin in glioma cells. Our results have contributed to approximate the study models with what occurs in gliomas in vivo and to evidence the therapeutic potential of doxazosin as an antiglioma agent with low neural and systemic toxicity.

Les glioblastomes sont des tumeurs fatales du fait de leur agressivité et du manque de traitements efficaces. La prolifération accrue, le caractère invasif et la résistance à la mort cellulaire du compartiment tumoral des glioblastomes leur confèrent une croissance rapide et une invasion du parenchyme cérébral environnant, à l’origine de leur systématique récidive. De plus, le processus d’angiogenèse au sein de ces tumeurs participe activement au mauvais pronostic en développant une forte vascularisation qui favorise leur croissance.Un peptide vasoactif l’Adrénomédulline (AM), est exprimé de façon ubiquitaire chez l’homme et a de ce fait un large champ d’action à travers l’organisme dont la progression tumorale en agissant de façon autocrine et paracrine via ses récepteurs CLR/RAMP2 et CLR/RAMP3 (« AMR »). Or l’expression de l’AM est corrélée avec le grade des gliomes. De plus, il a été démontré au sein de notre équipe que des anticorps polyclonaux anti-AM développés au laboratoire inhibent la prolifération des cellules gliales tumorales de glioblastomes in vitro ainsi que la croissance tumorale in vivo. Il a également été montré que des anticorps polyclonaux (anti-AMR) dirigés contre les récepteurs de l’AM (CLR, RAMP2 et RAMP3) inhibent in vitro la croissance, la migration et la formation en pseudo-capillaires des cellules endothéliales (HUVECs), suggérant une neutralisation par ces anticorps, de certaines étapes de l’angiogenèse. De même, il a été démontré dans des modèles in vivo que les anticorps anti-AMR inhibent la croissance tumorale en supprimant l’angiogenèse et la croissance des cellules tumorales, suggérant ainsi que les récepteurs de l’AM constitueraient une bonne cible thérapeutique. Ces études ayant été effectuées à partir de lignées cellulaires, nous avons caractérisé le système AM/AMR au sein des composantes gliales et microvasculaires provenant des cultures primaires de glioblastomes de patients puis nous avons testé l’effet de l’AM dans certains phénomènes caractéristiques des glioblastomes comme l’invasion et certaines étapes de l’angiogenèse tumorale. Il résulte de cette étude que l’AM est trois fois plus secrétée par la composante microvasculaire tumorale au sein de laquelle l’AM favorise la migration, l’invasion et l’organisation en pseudo-capillaires. De plus, l’AM favorise l’invasion des cellules de la composante gliale tumorale.Des anticorps capables de reconnaître et neutraliser à la fois l’AM, le CLR, RAMP2 et RAMP3 agissant de la même manière sur la croissance tumorale et l’angiogenèse représenteraient un bénéfice thérapeutique majeure. Le laboratoire a alors développé des anticorps dirigés contre un peptide chimérique constitué de l’enchainement de séquences peptidiques des protéines CLR, RAMP2, RAMP3 et du peptide AM (« AMRc »). Les tests effectués avec les anticorps anti-AMRc nous permettent d’affirmer leur efficacité sur le système AM/AMR. Le traitement par ces anticorps diminue la croissance in vitro et in vivo des cellules tumorales de glioblastomes U87, ainsi que leur migration et leur invasion et la densité vasculaire au sein de xénogreffes, ce qui suggère leur effet sur l’angiogenèse tumorale. De plus, le traitement par ces anticorps augmente la perméabilité du modèle de cellules endothéliale microvasculaires HMECs. Ces résultats très encourageants, nous permettent donc pour le moment de valider la faisabilité du concept d’anticorps développés à partir d’un peptide chimère pour neutraliser le sytème AM/AMR, dans le but d’envisager dans le futur une application thérapeutique. / Glioblastomas are fatal tumors because of their aggressiveness and the lack of effective treatments. The increased proliferation, invasiveness and resistance in cell death of glioblastomas tumoral compartment confer them a fast growth and an invasion of the surrounding cerebral parenchyma, at the origin of glioblastomas systematic recurrence. Furthermore, angiogenesis within these tumors participates actively in the poor prognosis developing a strong vascularization, which favors their growth.Adrenomedullin (AM), is a vasoactive peptide ubiquitously expressed in humans and thus induces multiple biological actions through the body as tumor growth, via autocrine and paracrine activation of its receptors CLR/RAMP2 and CLR/RAMP3 (« AMR »). But AM expression is correlated with gliomas grading and our team demonstrated that polyclonal antibodies « anti-AM » developped in the laboratory inhibited in vitro glioblastoma tumoral cells proliferation and in vivo tumor growth. Polyclonal antibodies (anti-AMR) directed against AM receptors (CLR, RAMP2 and RAMP3) also inhibited in vitro growth, migration and endothelial cells (HUVECs) pseudo-capillar formation, suggesting neutralization of some steps of angiogenesis. Moreover, anti-AMR antibodies inhibited in vivo tumor growth by suppression of angiogenesis, suggesting AM receptors as a therapeutical target.These studies have been done from lineage cells. We thus characterized AM/AMR system within glial and microvascular components from patients glioblastoma primary cultures and we investigated AM impact in some stages of angiogenesis. We showed that AM was three fold higher expressed by microvascular cells in whom AM induces migration, invasion and organization into a meshwork of capillary-like tubular structures. AM increased too, glial tumoral cells migration and invasion.Antibodies able to recognize and neutralize AM, CLR, RAMP2, RAMP3 and acting in tumor growth and angiogenesis would represent a major therapeutic benefit. Then, the laboratory developped polyclonal antibodies directed against one chimeric peptide synthesized with AM, CLR, RAMP2 and RAMP3 peptide sequences (named « AMRc »). The tests made with anti-AMRc antibodies allow us to assert their efficiency on the AM / AMR system. Glioblastoma cell line U87’s growth decreases in vitro and in vivo after treatment with anti-AMRc antibodies, as well as its migration, invasion and vascular density inside tumor xenografts, suggesting an impact in tumor angiogenesis. Furthermore, the treatment with these antibodies increases the microvascular endothelial HMECs permeability. These promising results allow us to validate the feasibility of a concept of antibodies developed against one peptide to neutralize AM/AMR system, in order to envisage a future therapeutic application.