Made available in DSpace on 2016-08-10T10:38:56Z (GMT). No. of bitstreams: 1 GABRIELA MOURA DE PAULA.pdf: 7198154 bytes, checksum: f6f3fcaaba289c4b737ab5cfc3d5363a (MD5) Previous issue date: 2014-08-29 / Breast cancer is the second most common cancer in the world. It is a complex and heterogeneous disease with diverse clinical features, cellular origin, histological types, mutations, prognosis and therapeutic possibilities. A subpopulation of cells with the ability of self-renewing, increased proliferation and resistance to chemotherapy has been described in breast cancer, named tumor-initiating cells (TICs). The aim of this study was to review and integrate, through meta-analysis, the studies that have investigated the possible associations between tumor-initiating cells, defined by CD44+ CD24- low phenotype and ALDH expression, by using immunohistochemistry, and prognostic aspects of breast cancer. Twenty studies met the inclusion criteria for this meta-analysis, 13 evaluated the CD44+ CD24- low phenotype, 11 evaluated the expression of ALDH and four studies evaluated both markers. The results of the metaanalysis demonstrated statistically significant associations between the triple negative phenotype and the CITs, including both the expression of CD44+ CD24- low phenotype (p < 0.0001), as the expression of ALDH (p = 0.0004). Associations investigated for HER2 overexpression presented conflicting results, whereas CD44+ CD24- low phenotype was not significantly associated with this parameter (p = 0.1989), ALDH expression was significantly associated with HER2 (p < 0.0001). An inverse association, however, statistically significant, was observed between the presence of lymph node metastases and CD44+ CD24- low phenotype (p = 0.0047), while the expression of ALDH was not significantly associated with this parameter (p = 2019). Based on the analyses carried out, it can be concluded that the markers, CD44+ CD24- low and ALDH, represent an important tool in identifying tumor initiating cells in breast carcinomas and that statistical associations observed in this study raise important perspectives for developing molecular therapies based on presence of the CD44+ CD24- low phenotype and ALDH expression in the treatment of triple-negative breast carcinomas. / O câncer de mama é o segundo câncer mais comum do mundo. É uma doença complexa e heterogênea com diversas características clínicas, histológicas, origem celular, mutações, prognóstico e possibilidades terapêuticas. Uma subpopulação de células com a capacidade de auto-renovação, proliferação celular e maior resistência à quimioterápicos tem sido descrita no câncer de mama, sendo estas denominadas célulasiniciadoras tumorais (CITs). O objetivo deste trabalho consistiu em revisar e integrar , por meio de meta-análise, os estudos que investigaram as possíveis associações entre células iniciadoras tumorais, definidas pelo fenótipo CD44+ CD24- low e expressão de ALDH, por método de imuno-histoquímica, e aspectos prognósticos do câncer de mama. Vinte estudos preencheram os critérios de inclusão para esta meta-análise, 13 avaliaram o fenótipo CD44+ CD24- low, 11 avaliaram a expressão de ALDH e quatro estudos avaliaram ambos marcadores. Os resultados da meta-análise demonstram associações estatisticamente significativas entre o fenótipo triplo negativo e as CITs, incluindo tanto a expressão do fenótipo CD44+ CD24- low (p < 0,0001), como a expressão de ALDH (p = 0,0004). Associações investigadas para a hiperexpressão de HER2 apresentaram resultados conflitantes, sendo que fenótipo CD44+ CD24- low não foi associado de forma significativa à este parâmetro (p = 0,1989), enquanto a expressão de ALDH esteve significativamente associada ao HER2 (p < 0,0001). Uma associação inversa, porém, estatisticamente significativa, foi observada entre a presença de metástases linfonodais e o fenótipo CD44+ CD24- low (p = 0,0047), enquanto a expressão de ALDH não esteve significativamente associada a este parâmetro (p = 2019). Com base nas análises realizadas, é possível concluir que os marcadores CD44+ CD24- low e ALDH representam uma importante ferramenta na identificação de células tumorais nos carcinomas mamários e que as associações estatísticas observadas criam perspectivas importantes para o desenvolvimento de terapias moleculares baseadas na presença de CD44+ CD24- low e de ALDH no tratamento dos carcinomas triplo negativos.

Mutações ativadoras no gene KRAS são prevalentes em cancer de pulmão e a as vias de sinalização de RAS estão aumentadas em células iniciadoras de tumor (CITs), que são definidas como células autorrenováveis capazes de iniciar a formação tumoral, sustentar o crescimento tumoral e promover a disseminação tumoral. Entretanto, terapias direcionadas a RAS não foram efetivas até hoje e a identificação de alvos de KRAS que contribuam para o fenótipo oncogênico é necessária. Como a quinase Aurora A (AURKA) já foi implicada, tanto na oncogênese induzida por KRAS, quanto em promover a função das CITs, nós hipotetizamos que a inibição das vias de AURKA seria detrimental para a função de CITs pulmonares portadoras de KRAS oncogênica, desta forma diminuindo o comportamento maligno do câncer de pulmão. Para avaliar a função das CITs, nós usamos ensaios de crescimento de tumoresferas que permitem o crescimento seletivo de CITs in vitro. As linhagens pulmonares positivas para KRAS H358 e A549 formaram tumoresferas em cultura de baixa aderência e, quando comparadas às linhagens parentais, às células oriundas de tumoresferas apresentaram maior capacidade clonogênica in vitro e maior tumorigenicidade in vivo. Além disso, uma análise por qPCR revelou que as células oriundas de tumoresferas possuem expressão aumentada de fatores de células tronco, uma característica de CITs. Em seguida, nós inibimos a AURKA nas linhagens pulmonares positivas para KRAS H358 e A549 por interferência de RNA (RNAi) ou com um inibidor das quinases Aurora (AI II). A inibição de AURKA diminuiu a formação de tumoresferas e o crescimento destas em culturas seriadas, além de reduzir a capacidade clonogênica das células oriundas de tumoresferas. Estes resultados indicam que a AURKA é importante para a autorrenovação e a oncogenicidade de CITs, e que a AURKA induz o fenótipo tronco-tumoral, o que é corroborado pelo achado de que a inibição de AURKA nas tumoresferas reduz a expressão de fatores de célula tronco. Um destes fatores regulados por AURKA é o marcador de superfície de célula tronco CD24. De fato, quando comparadas às células cultivadas de forma aderente, as células oriundas de tumoresferas apresentam maior número de células positivas para CD24 (CD24+) e estes números são reduzidos pelo tratamento com AI II. Finalmente, nós purificamos células H358 CD24+ por citometria de fluxo e mostramos que, quando comparadas às células negativas para CD24, as células CD24+ apresentam maior capacidade de formar tumoresferas em culturas seriadas, e o tratamento com AI II inibe preferencialmente a capacidade de células CD24+ de formarem tumoresferas. Nossos resultados sugerem que uma terapia baseada na inibição de AURKA pode reduzir o número e função de CITs pulmonares portadoras de KRAS oncogênica e, portanto, pode representar uma estratégia terapêutica atraente para reduzir a recidiva e metástase no câncer de pulmão induzido por KRAS. / Activating mutations in KRAS are prevalent in lung cancer and RAS sinaling is enhanced in cancer initiating cells (CICs), which are defined as self-renewing tumor cells able to initiate tumor formation, sustain tumor growth and drive tumor dissemination. However, therapies targeted to oncogenic RAS have been ineffective to date and identification of KRAS targets that impinge on the oncogenic phenotype is warranted. Because Aurora kinase A (AURKA) has been implicated both in RAS oncogenesis and in promoting CIC function, we hypothesized that targeting AURKA pathways would impair KRAS-positive lung CIC function, thereby decreasing lung cancer malignant behavior. To evaluate CIC function, we used tumorsphere assays that allow selective growth of CICs in vitro. KRAS positive lung cancer H358 and A549 cells formed tumorspheres under low attachment conditions, and, when compared to the parental cell lines, sphere-forming cells had increased clonogenic ability in vitro and increased tumorigenicity in vivo. In addition, qPCR analysis revealed that tumorsphere cells displayed increased expression of stem cell factors, a hallmark of CICs. Next, we targeted AURKA in KRAS positive lung cancer H358 and A549 cells by RNA interference (RNAi) or with an Aurora inhibitor (AI II). AURKA targeting decreased tumorsphere formation and growth in serial cultures and reduced clonogenic growth of tumorsphere-forming cells. These results indicate that AURKA is important for CIC selfrenewal and oncogenicity and that AURKA induces a CIC phenotype, which is further underscored by the finding that AURKA targeting in tumorspheres decreases expression of stem cell factors. One such factor shown to be regulated by AURKA is the stem cell surface marker CD24. In fact, when compared to adherent cultures, A549 and H358 tumorspheres display increased numbers of CD24-positive (CD24+) cells and these numbers are reduced by AI II treatment. Finally we purified H358 CD24+cells by flow cytometry and showed that, when compared to CD24-negative cells, CD24+ cells have increased ability to form tumorspheres in serial cultures, and AI II treatment preferentially reduced the ability of CD24+ cells to form tumorspheres. Our results suggest that AURKA inhibition therapy can reduce the number and function of KRAS-positive lung CICs, and, therefore might be an attractive therapeutic strategy to reduce recurrence and metastasis in KRAS-induced lung cancer.

Le neuroblastome (NB) est la tumeur solide extracranienne la plus fréquente et mortelle chez les jeunes enfants. Il se caractérise par une résistance à la chimiothérapie possiblement en partie dû à la présence de cellules initiatrices de tumeurs (TICs). Des études ont mis en évidence le rôle de CD133 comme un marqueur des TICs dans divers types de cancers. Les buts de notre travail étaient d’abord de démontrer les vertus de TICs des cellules exprimant CD133 et ensuite, en utilisant une analyse globale du génome avec des polymorphismes nucléotidiques simples (SNPs), d’effectuer une analyse différentielle entre les TICs et les autres cellules du NB afin d’en identifier les anomalies génétiques spécifiques. Des lignées cellulaires de NB ont été triées par cytométrie de flux afin d’obtenir deux populations: une enrichie en CD133 (CD133high), l’autre faible en CD133 (CD133low). Afin de déterminer si ces populations cellulaires présentent des propriétés de TICs, des essais sur les neurosphères, les colonies en agar mou et les injections orthotopiques de 500 cellules sélectionnées dans 11 souris ont été réalisées. Après une isolation de l’ADN des populations sélectionnées, nous avons effectué une analyse génotypique par SNP utilisant les puces « Affymetrix Genome-Wide Human SNP Array 6.0 ». Pour vérifier l’expression des gènes identifiés, des Western Blots ont été réalisés. Nos résultats ont démontré que la population CD133 avait des propriétés de TICs in vitro et in vivo. L’analyse génotypique différentielle a permis d’identifier deux régions communes (16p13.3 and 19p13.3) dans la population CD133high ayant des gains et deux autres régions (16q12.1 and 21q21.3) dans la population CD133low possédant des pertes d’hétérozygoties (LOH). Aucune perte n’a été observée. Parmi les gènes étudiés, l’expression protéique d’éphrine-A2 était corrélée à celle de CD133 dans 6 tumeurs et 2 lignées cellulaires de NB. De plus, l’augmentation de la concentration d’anticorps anti-éphrine-A2 dans le milieu diminue la taille des neurosphères. Ainsi, la population CD133high, qui a des vertus de TICs, possède des caractéristiques génotypiques différentes par rapport à celle CD133low. La présence d’éphrine-A2 dans les cellules exprimant CD133 souligne son importance dans le développement des TICs. Ces résultats suggèrent la présence de potentielle cible pour de nouvelles thérapeutiques ciblant les TICs mise en évidence par l’étude génomique. / Neuroblastoma (NB) is the most common and deadly extracranial solid tumor of childhood characterized by a resistance to chemotherapy possibly due to the presence of tumor initiating cells (TICs). Studies showed the role of CD133 as a marker of TICs in various types of cancers. Our goals were first to demonstrate the stemness of TICs expressing CD133 and then, using a global genomic analysis with single nucleotide polymorphism (SNPs), to perform a differential analysis between TICs and other cells of NB to identify the specific genetic abnormalities. NB cell lines were sorted by flow cytometry to obtain two populations: one enriched in CD133 (CD133high), the other low in CD133 (CD133low). To determine whether these cell populations have TICs properties, we test the ability of cells to form either neurosphères or, colonies in soft agar and we also test their carcinogenic properties by orthotopic injections of 500 selected cells in 11 mice. After a DNA extraction on selected populations, a differential genotyping analysis has been made with Affymetrix Genome-Wide Human SNP Array 6.0. To verify the expression of the genes identified, Western blots had been made. Our results have demonstrated that CD133high population presented TICs properties in vitro and in vivo. The differential genotyping analysis allowed identifying two gains common regions (16p13.3 and 19p13.3) in CD133high population and two others loss of heterozygosity (LOH) (16q12.1 and 21q21.3) in CD133low population . No losses were observed. Among the genes studied, ephrin-A2 protein expression was correlated to CD133 expression in 6 NB tumors and 2 NB cell lines. Also, ephrin-A2’s increased concentration influenced the neurospheres by decreasing their size. Thereby, CD133high population, which had TICs properties, possess different genotyping characteristics compared to CD133low population. The presence of ephrine-A2 in cells expressing CD133 emphasizes its importance in the development of TICs. These results suggest the presence of potential target for new therapies targeting the TICs demonstrated by the genomic study.

Mutações ativadoras no gene KRAS são prevalentes em cancer de pulmão e a as vias de sinalização de RAS estão aumentadas em células iniciadoras de tumor (CITs), que são definidas como células autorrenováveis capazes de iniciar a formação tumoral, sustentar o crescimento tumoral e promover a disseminação tumoral. Entretanto, terapias direcionadas a RAS não foram efetivas até hoje e a identificação de alvos de KRAS que contribuam para o fenótipo oncogênico é necessária. Como a quinase Aurora A (AURKA) já foi implicada, tanto na oncogênese induzida por KRAS, quanto em promover a função das CITs, nós hipotetizamos que a inibição das vias de AURKA seria detrimental para a função de CITs pulmonares portadoras de KRAS oncogênica, desta forma diminuindo o comportamento maligno do câncer de pulmão. Para avaliar a função das CITs, nós usamos ensaios de crescimento de tumoresferas que permitem o crescimento seletivo de CITs in vitro. As linhagens pulmonares positivas para KRAS H358 e A549 formaram tumoresferas em cultura de baixa aderência e, quando comparadas às linhagens parentais, às células oriundas de tumoresferas apresentaram maior capacidade clonogênica in vitro e maior tumorigenicidade in vivo. Além disso, uma análise por qPCR revelou que as células oriundas de tumoresferas possuem expressão aumentada de fatores de células tronco, uma característica de CITs. Em seguida, nós inibimos a AURKA nas linhagens pulmonares positivas para KRAS H358 e A549 por interferência de RNA (RNAi) ou com um inibidor das quinases Aurora (AI II). A inibição de AURKA diminuiu a formação de tumoresferas e o crescimento destas em culturas seriadas, além de reduzir a capacidade clonogênica das células oriundas de tumoresferas. Estes resultados indicam que a AURKA é importante para a autorrenovação e a oncogenicidade de CITs, e que a AURKA induz o fenótipo tronco-tumoral, o que é corroborado pelo achado de que a inibição de AURKA nas tumoresferas reduz a expressão de fatores de célula tronco. Um destes fatores regulados por AURKA é o marcador de superfície de célula tronco CD24. De fato, quando comparadas às células cultivadas de forma aderente, as células oriundas de tumoresferas apresentam maior número de células positivas para CD24 (CD24+) e estes números são reduzidos pelo tratamento com AI II. Finalmente, nós purificamos células H358 CD24+ por citometria de fluxo e mostramos que, quando comparadas às células negativas para CD24, as células CD24+ apresentam maior capacidade de formar tumoresferas em culturas seriadas, e o tratamento com AI II inibe preferencialmente a capacidade de células CD24+ de formarem tumoresferas. Nossos resultados sugerem que uma terapia baseada na inibição de AURKA pode reduzir o número e função de CITs pulmonares portadoras de KRAS oncogênica e, portanto, pode representar uma estratégia terapêutica atraente para reduzir a recidiva e metástase no câncer de pulmão induzido por KRAS. / Activating mutations in KRAS are prevalent in lung cancer and RAS sinaling is enhanced in cancer initiating cells (CICs), which are defined as self-renewing tumor cells able to initiate tumor formation, sustain tumor growth and drive tumor dissemination. However, therapies targeted to oncogenic RAS have been ineffective to date and identification of KRAS targets that impinge on the oncogenic phenotype is warranted. Because Aurora kinase A (AURKA) has been implicated both in RAS oncogenesis and in promoting CIC function, we hypothesized that targeting AURKA pathways would impair KRAS-positive lung CIC function, thereby decreasing lung cancer malignant behavior. To evaluate CIC function, we used tumorsphere assays that allow selective growth of CICs in vitro. KRAS positive lung cancer H358 and A549 cells formed tumorspheres under low attachment conditions, and, when compared to the parental cell lines, sphere-forming cells had increased clonogenic ability in vitro and increased tumorigenicity in vivo. In addition, qPCR analysis revealed that tumorsphere cells displayed increased expression of stem cell factors, a hallmark of CICs. Next, we targeted AURKA in KRAS positive lung cancer H358 and A549 cells by RNA interference (RNAi) or with an Aurora inhibitor (AI II). AURKA targeting decreased tumorsphere formation and growth in serial cultures and reduced clonogenic growth of tumorsphere-forming cells. These results indicate that AURKA is important for CIC selfrenewal and oncogenicity and that AURKA induces a CIC phenotype, which is further underscored by the finding that AURKA targeting in tumorspheres decreases expression of stem cell factors. One such factor shown to be regulated by AURKA is the stem cell surface marker CD24. In fact, when compared to adherent cultures, A549 and H358 tumorspheres display increased numbers of CD24-positive (CD24+) cells and these numbers are reduced by AI II treatment. Finally we purified H358 CD24+cells by flow cytometry and showed that, when compared to CD24-negative cells, CD24+ cells have increased ability to form tumorspheres in serial cultures, and AI II treatment preferentially reduced the ability of CD24+ cells to form tumorspheres. Our results suggest that AURKA inhibition therapy can reduce the number and function of KRAS-positive lung CICs, and, therefore might be an attractive therapeutic strategy to reduce recurrence and metastasis in KRAS-induced lung cancer.

CD133, a pentaspan glycoprotein, has long been known to represent aggressive, stem-like populations across various human malignancies. While its expression correlates with numerous clinical outcomes including disease progression, metastasis, recurrence, and poor overall survival in numerous cancers, little is currently known about its function. In the brain cancer glioblastoma (GBM), CD133-expressing cells have previously been shown to initiate tumours, evade therapy and interestingly, self-renew, a key property of cancer stem cells. With an implied signalling role in driving self-renewal, we aim to elucidate the role of CD133 in glioblastoma. To understand the role of CD133, we aim to study its protein-protein interactions using the proximity-dependent labelling technique known as miniTurboID. By tagging proteins of interest with a promiscuous biotin ligase at both protein termini, potential interactors can be biotinylated and identified by subsequent mass spectrometry. While miniTurboID has traditionally been performed by synthetic transgenes expressing the tagged proteins of interest in commercial cell lines, overexpression may not recapitulate its native function. Thus, using CRISPR technology, we aim to insert the miniTurboID ligase at both the N- and C-terminus of CD133 in patient-derived human GBM lines. Although little is currently known about CD133 function, development of targeted therapies has presented a promising strategy in pre-clinical studies. In the Singh Lab, we previously developed a chimeric antigen receptor T-cell, or CAR-T, comprised of a T-cell expressing a synthetic receptor capable of recognizing a tumor-associated antigen and activating cytolytic-killing directed towards the target cell. Currently, CAR-T therapies are autologous, or patient-derived, in nature which may host a myriad of concerns including patient-specific qualitative and quantitative T-cell dysfunction, inconsistent generation of CAR products, and availability to rapidly progressing patients. To circumvent this concern, “off-the-shelf”, donor-derived or allogeneic CAR-T products may be generated for use in GBM patients. However, in addition to CAR integration, allogeneic products must be additionally modified to eradicate expression of the endogenous TCR, as this would induce a phenomenon known as graft versus host disease, in which healthy tissues are targeted. Thus, in this thesis, we show gene editing potential in human GBMs to perform an endogenous genomic knock-in of miniTurboID. With the identification of interacting proteins, defining the subsequent functionality of CD133 may elucidate oncogenic cellular programs, and highlight common nodes of interaction within divergent cell signaling pathways. To develop an allogeneic CAR-T product, we designed a two-step approach in which the CAR sequence was integrated into the TCR gene for simultaneous knock-out. We later show early pre-clinical efficacy in comparison to traditional autologous CAR-T in our patient-derived models of human GBM. Thus, by using CD133 as a centralizing concept in this thesis, we ultimately hope to develop our biological understanding of CD133, while testing the therapeutic development of a donor-derived CAR-T therapy. / Thesis / Master of Science (MSc) / Glioblastoma (GBM) is one of the most common malignant brain tumors in adults. Despite an aggressive therapy regimen, almost all patients relapse 7-9 months post-diagnosis. Therapy failure and poor patient outcome may be attributed to a small population of cells known as glioblastoma stem cells, or GSCs, that are able to escape therapy and seed disease recurrence. GSCs are most notably identified by the cell surface protein CD133, which has previously been shown to associate with pro-tumor properties including treatment resistance, tumor growth, maintenance, progression and metastasis. While expression of CD133 in cancer has been heavily characterized, little is currently known about its function. One such avenue to understand its mechanism of action in cancer, and more particularly GBM, is to define its interactions with other proteins. Protein-protein interactions play a pivotal part as the backbone of signalling pathways that drive tumor development and growth. Therefore, defining and mapping the CD133 interaction network may help us understand how this protein governs regulation of GSCs, and ultimately, GBM progression. While the biology of CD133 has yet to be elucidated, targeting CD133 on GSCs has presented a promising therapeutic strategy for patients with GBM. Previously in the Singh Lab, we developed an engineered T-cell therapy, known as a CAR-T, that can recognize CD133 to induce tumor cell death. While this showed success in our animal models of human GBM, other considerations must be addressed on its path to clinical development. As of current, CAR-T therapies are generated from T-cells taken from cancer patients. This hosts a myriad of concerns including the quality of patient T-cells, the time and cost to manufacture, and its availability for patients with rapidly progressing disease. To circumvent this issue, donor-derived CAR-T cells can be genetically engineered for safe usage in GBM patients as a readily available, “off-the-shelf” therapy. To define the function of CD133, we have attempted to use a technique known as BioID, which tags the protein of interest with a smaller biotin ligase. This biotin ligase can subsequently tag proteins that come within the vicinity of CD133, that may later be identified by sequencing as potential interactors. As current use of BioID may not reliably mimic the interaction of CD133, we sought to genetically engineer human GBM lines with the BioID protein to more closely resemble tumor-relevant behaviours of CD133. To develop a donor-derived CAR-T therapy, we similarly used genetic engineering of T-cells to ensure specific targeting of tumor cells with CD133, while sparing healthy tissues. By using CD133 as a centralizing concept in this thesis, we ultimately hope to develop our biological understanding of CD133, while testing the therapeutic development of a donor-derived CAR-T therapy.

Prostate cancer (CaP) is the second leading cause of cancer related deaths in men in the United States. Currently, androgen depletion is an essential strategy for CaP combined with surgery, chemotherapy and radiation. Hormone independent cancer stem cells escaping conventional therapy present a major therapeutic challenge. The available treatment regimens for hormone resistant CaP are only palliative and marginally increase survival. Therefore, novel strategies to eradicate CaP including stem cells are imperative. Oncolytic virus (OV) therapy is a novel approach that overcomes the limitations posed by radiation and chemotherapy. Oncolytic virotherapy of cancer is based on the use of replication competent, tumor selective viruses with limited toxicity. Newcastle Disease Virus (NDV), an avian paramyxovirus, is a safe and promising OV successfully used in many clinical trials. NDV is inherently tumor selective and cytotoxic but replication restricted in normal cells. But, systemically delivered NDV fails to reach solid tumors in therapeutic concentrations and also spreads poorly within the tumors due to barriers including complement, innate immunity and extracellular matrix. Overcoming these hurdles is paramount to realize the exceptional oncolytic efficacy of NDV. Therefore, we engineered the fusion (F) glycoprotein of NDV and generated a recombinant NDV (rNDV) cleavable exclusively by prostate specific antigen (PSA). The rNDV replicated efficiently and specifically only in prostate cancer (CaP) cells but failed to replicate in the absence of PSA. Further, PSA-cleavable rNDV caused specific lysis of androgen independent and dependent/responsive CaP cells with a mean effective concentration (EC50) ranging from 0.01 to 0.1 multiplicity of infection (MOI). PSA retargeted rNDV efficiently lysed three-dimensional prostaspheres, suggesting efficacy in vivo. Also, PSA-cleavable NDV failed to replicate in chicken embryos, indicating absence of pathogenicity to its natural host, chickens. Prostaspheres generated from DU-145 CaP cell line derived xenografts showed self-renewal, proliferative and clonogenic potential in vitro, and exhibited increased tumorigenicity in vivo. Embryonic stem and progenitor cell markers like Nanog, Nestin and CD44 were overexpressed in spheres as compared to the cell line suggesting prostaspheres comprise tumor-initiating cells from CaP. Xenograft and cell line derived prostaspheres were permissive for rNDV replication, when the fusion protein was activated by exogenous PSA. The EC50 against tumor initiating cells was 0.11-0.14 MOI, suggesting an excellent therapeutic margin for in vivo studies. PSA retargeting is likely to enhance the therapeutic index of rNDV owing to tumor restricted replication and enhanced fusogenicity. Our results suggest PSA retargeted rNDV selectively replicates and lyse PSA producing CaP cells including tumor-initiating cells and is a promising candidate for immediate Phase I/II clinical trials. / Ph. D.

As neoplasias mamárias apresentam um grande desafio tanto para a medicina humana, quanto para a medicina veterinária. Esses tumores apresentam ampla heterogeneidade intertumoral e intratumoral, dificultando assim a busca por tratamentos eficazes. Recentemente, pesquisadores tem voltado sua atenção para uma população de células que apresentam características muito semelhantes as células-tronco. São as chamadas células iniciadoras de tumores (CITs). Estas são descritas como as principais responsáveis por falhas nas quimioterapias e no surgimento de recidivas tumorais, devido ao seu potencial tumorigênico, de auto-renovação e de resistência a drogas antineoplásicas. Entretanto, o estudo dessas células é limitado pelas dificuldades no isolamento e na caracterização pós-enriquecimento dessas células, devido à perda do fenótipo em modelos in vitro. Sendo assim, vários estudos estão buscando maneiras alternativas de enriquecer essa população. Uma das maneiras mais utilizadas, baseia-se na indução do processo de transição epitélio-mesenquimal, através da superexpressão de fatores de transcrição como SNAI1, SLUG, ZEB1 e ZEB2. Sendo assim, nós objetivamos expressar de maneira exógena os fatores de transcrição SLUG e ZEB1 em células oriundas de carcinomas mamários caninos, caracterizar seus efeitos nessas células e observar se esses fatores de transcrição seriam capazes de induzir o fenótipo de CIT. Primeiramente, quatro amostras de carcinomas mamários caninos foram analisados quanto sua morfologia e os níveis de expressão gênica de quatro fatores de transcrição associados a transição epitélio-mesenquimal: SLUG, STAT3, ZEB1 e ZEB2. Após, nós selecionamos duas dessas amostras (CC-20E e CL-28E), que apresentavam morfologia típica de células epiteliais e baixa expressão dos fatores de transcrição citados acima e expressamos de maneira exógena e de forma estável os fatores de transcrição SLUG e ZEB1, através do processo de transdução lentiviral. Entretanto, apenas a transdução com os plasmídeos contendo a região codificante de SLUG foi eficiente. Sendo assim, nós avaliamos os efeitos da expressão exógena de SLUG nas células CC-20E e CL-28E, quanto a alteração de morfologia e expressão de filamentos intermediários como citoqueratina, vimentina e actina. Além disso, nós avaliamos se a expressão exógena de SLUG poderia regular a expressão de outros genes associados a EMT, além de genes supressores de tumores, alvos de SLUG. Por fim, nós avaliamos se a expressão exógena de SLUG poderia induzir ao fenótipo de CITs, verificando se havia alteração na sensibilidade das células aos quimioterápicos doxorrubicina e paclitaxel, além de avaliar o potencial tumorigênico e de auto-renovação dessas células em cultivos de baixa aderência. A expressão exógena de SLUG nas células CC-20E e CL-28E, não induziu a alterações na morfologia epitelial das células. Entretanto, as células alteraram sua disposição em monocamada no cultivo, formando tipos de túbulos semidiferenciados, característicos do processo de EMT híbrido ou parcial. Além, disso, houve um equilíbrio entre a expressão dos filamentos intermediários de citoqueratina e vimentina nas células, além do aumento na expressão dos genes CDH1 (E-caderina) e CDH2 (N-caderina), resultado que sustentou a indução de EMT parcial. O processo de EMT parcial induziu maior resistência ao quimioterápico paclitaxel, além de potencializar a tumorigenecidade e a capacidade de auto-renovação das células em cultivos de baixa aderência. Sendo assim, no presente estudo, nós obtivemos um cultivo com características que mimetizam as CITs, demonstrando que os processos que induzem esse fenótipo são semelhantes tanto na espécie canina, quanto em humanos, sustentando a hipótese de que neoplasias mamárias caninas podem servir como modelo para o estudo das CITs e, consequentemente, do desenvolvimento neoplásico de tumores sólidos. / Mammary neoplasms present a major challenge for both human and veterinary medicine, due to intertumoral and intratumoral heterogeneity, hindering the search for effective treatments. Recently, researchers has highlighted a population of cells with features very similar to stem cells. Known as Tumor-Initiating Cells (TICs), they are described as the main responsible for chemotherapy failures and tumor recurrence, due to their tumorigenic potential, self-renewal ability and drug resistance. The study of TICs is limited mainly by their difficult isolation owing to specific markers absence, and furthermore, cells lose their phenotype when placed in vitro. Therefore, several studies are seeking for alternatives to enrich this population in regular cultures. One way is based on the epithelial-mesenchymal transition induction through of transcription factors overexpression, such as SNAI1, SLUG, ZEB1 e ZEB2. So, the aim of this study was to overexpresse the SLUG and ZEB1 transcription factors in a cell culture derived from canine mammary carcinomas, evaluate its effects and observe whether these transcription factors would be capable of inducing the TIC phenotype. First, four canine mammary carcinomas cell cultures were analyzed for their morphology and gene expression levels of four transcription factors associated with epithelial-mesenchymal transition: SLUG, STAT3, and ZEB1 ZEB2. After, we selected two samples (CC-20E and CL-28E) with typical morphology of epithelial cells and low expression of the transcription factors mentioned above. We then overexpress, stably, the transcription factors SLUG and ZEB1 by lentiviral transduction, However, only SLUG transduction was efficient. Then, we evaluated the effects of SLUG overexpression in CC-20E and CL-28E cells as the change of morphology, expression of intermediate filaments as cytokeratin, vimentin and actin. In addition, we evaluated whether SLUG overexpression could regulate the expression of other EMT-associated genes as well as tumor suppressor genes, and assessed evaluated the tumorigenic potential and self-renewal of these cells in low adherence cultures. Finally, we assessed whether SLUG overexpression could induce drug resistance through doxorubicin and paclitaxel sensivity assay. The SLUG overexpression did not induce modification in epithelial cell morphology, however, cells changed their arrangement in monolayer culture, inducing the semidifferentiated tubules, typical of hybrid or partial EMT process. In, addition, there was a balanced expression between cytokeratin and vimentin, possibly explained by an increase in CDH1 expression (E-cadherin) and CDH2 (N-cadherin) typical of partial EMT. Furthermore, the partial EMT generated cells presenting paclitaxel resistance, and enhanced the tumorigenic potential and self-renewal capacity of the cells on low adherent plates. Thus, in this study, we obtained a cell culture exhibiting features that mimics the TICs, demonstrating the mechanisms which regulate this phenotype are similar in dogs and humans, supporting the hypothesis that canine mammary carcinomas are a great model for the study of TICs and solid tumors development.

Au cours de ma thèse, je me suis intéressée à une population cellulaire rare et peu étudiée de l’épithélium intestinal ; les cellules tuft. La fonction de ces cellules fut longtemps débattue dans la littérature, jusqu’à ce que nous découvrions leur fonction dans l’initiation de la réponse immune de type II en réponse à une infection parasitaire. De manière intéressante, ces cellules sont présentes de manière massive et transitoire au sein des lésions adénomateuses précoces et certains groupes ont suggéré l’implication de ces cellules en tant que cellules souches tumorales. Les principaux objectifs de ma thèse ont été de déterminer le rôle des cellules tuft au cours de la tumorigenèse intestinale et colorectale.Nous avons montré que l’absence de cellules tuft impacte le processus de tumorigenèse à la fois dans l’intestin grêle, dans des souris de la lignée Apc14/+, et au niveau du colon, après traitement avec un agent carcinogène. Nos données indiquent que si les cellules tuft n’agissent pas en tant que cellules souches tumorales, leur absence impacte certaines populations de cellules immunitaires. Afin de déterminer les mécanismes permettant aux cellules tuft de moduler le microenvironnement immunitaire, nous avons identifié par analyse transcriptomique de cellules tuft isolées par cytométrie en flux, des gènes codant pour des médiateurs connus pour être impliqués dans la communication avec le système immunitaire. Des analyses in-vivo, permettront de valider d’un point de vue fonctionnel l’implication de ces médiateurs immunitaires dans la fonction immuno-régulatrice des cellules tuft ainsi que dans le développement tumoral.L’ensemble de ces travaux a permis d’identifier une fonction immuno-régulatrice des cellules tuft au cours d’une infection parasitaire, mais aussi très probablement lors du développement tumoral. La compréhension des mécanismes permettant aux cellules tuft de moduler certaines populations de cellules immunitaires permettra d’identifier des cibles d’intérêt thérapeutique potentiel pour le traitement de patients atteints d’un cancer colorectal. / I focused my PhD project on a scare epithelial cell population referred as tuft cells. Their function has been debated for decades in the literature, until we discovered their crucial role in the initiation of the so-called type-2 immune response following parasitic infection. Interestingly, tuft cells are present in early adenomatous intestinal lesions and literature suggested that these cells could act as cancer stem cells. The main objective of my PhD was to determine the tuft cell function during intestinal and colorectal cancer.We showed that tuft cells deficiency impacts both intestinal and colorectal tumorigenesis process, using Apc14/+ mouse strain and chemically induced carcinogenesis model, respectively. Our data indicate that tuft cells are not cancer stem cells, but that these cells are able to regulate immune cell populations. To get more insights into mechanisms allowing tuft cells to modulate the immune microenvironment, we identified, by transcriptomic analysis of FACS-isolated tuft cells, specific genes encoding mediators involved in the crosstalk with the immune system. Functional in-vivo validation of the most relevant candidates will identify tuft cells derived factors crucial for the immune-regulatory tuft cell function and for tumor development.This work allowed to highlight the immune-regulatory function of tuft cells during parasitic infection and likely during tumor development. A better knowledge of the mechanisms allowing tuft cells to shape either a pro- or an anti-tumoral microenvironment, will potentially paves the way for new therapeutic strategies regarding intestinal and colorectal tumorigenesis.

La présence de métastases est un facteur de mauvais pronostic dans les cancers solides et une meilleure compréhension de la dissémination tumorale est nécessaire afin d'améliorer la prise en charge de ces formes avancées. Les cellules tumorales circulantes (CTC) représentent un intérêt majeur dans la pathologie tumorale, d'une part sur le plan clinique en tant que marqueur prédictif et pronostique et d'autre part sur le plan de la compréhension des mécanismes impliqués dans la formation des métastases. Les CTC sont rares et hétérogènes et restent mal caractérisées, et ce, particulièrement dans le cancer colorectal. Une partie de ces cellules aurait un phénotype de cellules initiatrices de tumeur (CIT) leur permettant de former des métastases, de résister aux traitements et par conséquent d'être responsables des rechutes. Une meilleure connaissance des CTC possédant un phénotype de CIT représente donc un enjeu majeur. L'objectif de ce travail a été d'identifier et de caractériser les CTC avec un potentiel de cellules initiatrices de tumeur dans le cancer colorectal en se basant sur les propriétés fonctionnelles des CIT. Nous avons ainsi, pour la première fois, pu établir deux modèles permettant de répondre à cet objectif. D'une part des lignées de CTC avec un fort potentiel de CIT obtenues à partir d'échantillons sanguins de patients atteints de cancer colorectal, et d'autre part, nous avons mis en place un modèle murin de dissémination tumorale par xénogreffe orthotopique permettant d'isoler les CTC. / Liver or lung metastases represent a poor prognosis in colorectal cancer patients and better understanding tumor spreading became essential to improve patient care. Circulating tumor cells (CTC) is considered as a promising tool, both as prognostic marker and as tool to study mechanisms involved in metastasis development. CTCs are rare and heterogeneous and remain poorly characterized especially in colorectal cancer. It is accepted that at least some of the CTC have a tumor initiating cell (TIC) phenotype that could be responsible for metastasis, chemoresistance and consequently lead to relapse. A deep characterization of CTC became thus an urgent unmet need. The aim of this work was to identify and characterize CTC with TIC properties in colorectal cancer, on the basis of their functional properties. To reach this aim, we established for the first time and characterized CTC lines from blood sample of colorectal cancer patient, and we also developed an orthotopic xenograft mouse model in which tumoral cells are circulating in the blood.

As neoplasias mamárias apresentam um grande desafio tanto para a medicina humana, quanto para a medicina veterinária. Esses tumores apresentam ampla heterogeneidade intertumoral e intratumoral, dificultando assim a busca por tratamentos eficazes. Recentemente, pesquisadores tem voltado sua atenção para uma população de células que apresentam características muito semelhantes as células-tronco. São as chamadas células iniciadoras de tumores (CITs). Estas são descritas como as principais responsáveis por falhas nas quimioterapias e no surgimento de recidivas tumorais, devido ao seu potencial tumorigênico, de auto-renovação e de resistência a drogas antineoplásicas. Entretanto, o estudo dessas células é limitado pelas dificuldades no isolamento e na caracterização pós-enriquecimento dessas células, devido à perda do fenótipo em modelos in vitro. Sendo assim, vários estudos estão buscando maneiras alternativas de enriquecer essa população. Uma das maneiras mais utilizadas, baseia-se na indução do processo de transição epitélio-mesenquimal, através da superexpressão de fatores de transcrição como SNAI1, SLUG, ZEB1 e ZEB2. Sendo assim, nós objetivamos expressar de maneira exógena os fatores de transcrição SLUG e ZEB1 em células oriundas de carcinomas mamários caninos, caracterizar seus efeitos nessas células e observar se esses fatores de transcrição seriam capazes de induzir o fenótipo de CIT. Primeiramente, quatro amostras de carcinomas mamários caninos foram analisados quanto sua morfologia e os níveis de expressão gênica de quatro fatores de transcrição associados a transição epitélio-mesenquimal: SLUG, STAT3, ZEB1 e ZEB2. Após, nós selecionamos duas dessas amostras (CC-20E e CL-28E), que apresentavam morfologia típica de células epiteliais e baixa expressão dos fatores de transcrição citados acima e expressamos de maneira exógena e de forma estável os fatores de transcrição SLUG e ZEB1, através do processo de transdução lentiviral. Entretanto, apenas a transdução com os plasmídeos contendo a região codificante de SLUG foi eficiente. Sendo assim, nós avaliamos os efeitos da expressão exógena de SLUG nas células CC-20E e CL-28E, quanto a alteração de morfologia e expressão de filamentos intermediários como citoqueratina, vimentina e actina. Além disso, nós avaliamos se a expressão exógena de SLUG poderia regular a expressão de outros genes associados a EMT, além de genes supressores de tumores, alvos de SLUG. Por fim, nós avaliamos se a expressão exógena de SLUG poderia induzir ao fenótipo de CITs, verificando se havia alteração na sensibilidade das células aos quimioterápicos doxorrubicina e paclitaxel, além de avaliar o potencial tumorigênico e de auto-renovação dessas células em cultivos de baixa aderência. A expressão exógena de SLUG nas células CC-20E e CL-28E, não induziu a alterações na morfologia epitelial das células. Entretanto, as células alteraram sua disposição em monocamada no cultivo, formando tipos de túbulos semidiferenciados, característicos do processo de EMT híbrido ou parcial. Além, disso, houve um equilíbrio entre a expressão dos filamentos intermediários de citoqueratina e vimentina nas células, além do aumento na expressão dos genes CDH1 (E-caderina) e CDH2 (N-caderina), resultado que sustentou a indução de EMT parcial. O processo de EMT parcial induziu maior resistência ao quimioterápico paclitaxel, além de potencializar a tumorigenecidade e a capacidade de auto-renovação das células em cultivos de baixa aderência. Sendo assim, no presente estudo, nós obtivemos um cultivo com características que mimetizam as CITs, demonstrando que os processos que induzem esse fenótipo são semelhantes tanto na espécie canina, quanto em humanos, sustentando a hipótese de que neoplasias mamárias caninas podem servir como modelo para o estudo das CITs e, consequentemente, do desenvolvimento neoplásico de tumores sólidos. / Mammary neoplasms present a major challenge for both human and veterinary medicine, due to intertumoral and intratumoral heterogeneity, hindering the search for effective treatments. Recently, researchers has highlighted a population of cells with features very similar to stem cells. Known as Tumor-Initiating Cells (TICs), they are described as the main responsible for chemotherapy failures and tumor recurrence, due to their tumorigenic potential, self-renewal ability and drug resistance. The study of TICs is limited mainly by their difficult isolation owing to specific markers absence, and furthermore, cells lose their phenotype when placed in vitro. Therefore, several studies are seeking for alternatives to enrich this population in regular cultures. One way is based on the epithelial-mesenchymal transition induction through of transcription factors overexpression, such as SNAI1, SLUG, ZEB1 e ZEB2. So, the aim of this study was to overexpresse the SLUG and ZEB1 transcription factors in a cell culture derived from canine mammary carcinomas, evaluate its effects and observe whether these transcription factors would be capable of inducing the TIC phenotype. First, four canine mammary carcinomas cell cultures were analyzed for their morphology and gene expression levels of four transcription factors associated with epithelial-mesenchymal transition: SLUG, STAT3, and ZEB1 ZEB2. After, we selected two samples (CC-20E and CL-28E) with typical morphology of epithelial cells and low expression of the transcription factors mentioned above. We then overexpress, stably, the transcription factors SLUG and ZEB1 by lentiviral transduction, However, only SLUG transduction was efficient. Then, we evaluated the effects of SLUG overexpression in CC-20E and CL-28E cells as the change of morphology, expression of intermediate filaments as cytokeratin, vimentin and actin. In addition, we evaluated whether SLUG overexpression could regulate the expression of other EMT-associated genes as well as tumor suppressor genes, and assessed evaluated the tumorigenic potential and self-renewal of these cells in low adherence cultures. Finally, we assessed whether SLUG overexpression could induce drug resistance through doxorubicin and paclitaxel sensivity assay. The SLUG overexpression did not induce modification in epithelial cell morphology, however, cells changed their arrangement in monolayer culture, inducing the semidifferentiated tubules, typical of hybrid or partial EMT process. In, addition, there was a balanced expression between cytokeratin and vimentin, possibly explained by an increase in CDH1 expression (E-cadherin) and CDH2 (N-cadherin) typical of partial EMT. Furthermore, the partial EMT generated cells presenting paclitaxel resistance, and enhanced the tumorigenic potential and self-renewal capacity of the cells on low adherent plates. Thus, in this study, we obtained a cell culture exhibiting features that mimics the TICs, demonstrating the mechanisms which regulate this phenotype are similar in dogs and humans, supporting the hypothesis that canine mammary carcinomas are a great model for the study of TICs and solid tumors development.