Tumor metastasis is a fundamental property of malignant cancer cells and the major cause of death in cancer patients. Recent studies indicate that tumor cell invasion and metastasis may be initiated by the formation of the actin-rich cell protrusions with ECM degradation activity, invadopodia. However, despite extensive research on the biology of invadopodia, very little is known about their specific inducers during tumor progression. Autotaxin (ATX) is a secreted lysophospholipase whose expression levels within tumors correlates strongly with their aggressiveness and invasiveness. ATX produces lyosophosphatidic acid (LPA), a phospholipid with known tumor promoting functions that acts through the G-protein coupled receptors, LPA[subscript 1-6] . Recently, overexpression of ATX and LPA receptors (LPA[subscript 1-3]) has been linked to increased tumor invasion and metastasis in vivo , however, the role of other LPA receptors (LPA[subscript 4-6]) as well as the exact mechanisms by which ATX induces tumor metastasis remain poorly characterized. In order to determine the involvement of ATX and LPA in invadopodia production, we used the fibrosarcoma HT-1080 cells stably transfected with ATX or shRNA targeting ATX in fluorescent matrix degradation assays. Our results demonstrate that ATX is implicated in the production of invadopodia resulting in an increase in both their formation and function. Using LPC or LPA, the substrate and product of ATX, we further show that invadopodia production is dependent on the production of LPA from LPC. Among the LPA receptors, LPA 4 has the highest expression in HT1080 cells. Using LPA[subscript 4] shRNA as well as agonists and inhibitors of the cAMP pathway, we provide evidence that LPA[subscript 4] signaling through the cAMP-EPAC-Rap1 axis, regulates invadopodia formation downstream of ATX. Furthermore, inhibition of Rac1, a known effector of Rap1 and invadopodia formation, abolished EPAC-induced invadopodia production, suggesting downstream participation of Rac1. Finally, results using LPA[subscript 4] shRNA support the requirement of this receptor for in vitro cell invasion and in vivo metastasis formation. Our results suggest that ATX through LPA[subscript 4] is a strong inducer of invadopodia formation that correlates with the ability of the cells to invade and metastasize. This study also revealed an unexpected signaling pathway for cell invasion involving LPA[subscript 4]-driven cAMP production and subsequent activation of the EPAC-Rap1-Rac1 axis.

Cette thèse présente des travaux concernant des phénomènes d'invasion tumorale, aux échelles tissulaire et cellulaire. La première partie est consacrée à deux modèles mathématiques continus. Le premier est un modèle macroscopique de croissance d'un cancer du sein qui se focalise sur la description du passage du stade in situ au stade invasif. Basé sur des équations d'advection d'espèces cellulaires, il tient compte de la géométrie et de l'éventuelle dégradation des tissus, dans le cas où la tumeur produit des enzymes protéolytiques qui permettent l'invasion. Le second modèle concerne le phénomène d'invadopodia, à l'échelle de la cellule. C'est un problème d'interface mobile qui décrit le changement de morphologie des cellules pré-métastatiques qui leur permet de dégrader les tissus pour migrer dans le reste de l'organisme. Chacun de ces deux modèles tient compte des couplages forts inhérents au phénomènes biologiques en jeu.La seconde partie est consacrée aux méthodes numériques développées pour résoudre ces deux problèmes et surmonter les difficultés liées aux couplages et non linéarités. Elles sont construites sur grille cartésienne uniforme, à partir des différences finies et d'une version stabilisée de la méthode Ghost fluid. Leur particularité est de tirer pleinement parti des propriétés de superconvergence de la solution du problème de Poisson, spécifiquement étudiées afin d'aboutir à la résolution des problèmes de cancer du sein et d'invadopodia à l'ordre un ou deux, en fonction de la précision désirée. Cesméthodes peuvent également être utilisées pour résoudre d'autres problèmes d'interface mobile. / In this thesis, we present a study about phenomena of tumor invasion, at the tissues and cell scales.The first part is devoted to two continuous mathematical models. The first one is a macroscopic model for breast cancer growth, which focuses on the transition between the stage in situ and the invasive phase of growth. This model is based on advection equations for cellular species. The geometry and possible tissue damage are taken into account. Invasion occurs when the tumor cells produce proteolytic enzymes. The second model deals with the phenomenon of invadopodia, at the cell scale.This is a free boundary problem, which describes the change in morphology of pre-metastatic cells,enabling them to degrade the tissues and migrate into the rest of the body. Each of these models reflects the strong coupling of biological phenomena.The second part is devoted to numerical methods specifically developed to solve these problems and overcome coupling and nonlinearities. They are built on uniform Cartesian grids, thanks to the finite difference method, and a stabilized version of the Ghost fluid method. Their peculiarity is to take full advantage of superconvergence properties of the Poisson problem solution. These properties are specifically studied, leading to the first or second order numerical computation of the problems ofbreast cancer and invadopodia, depending on the desired accuracy. These methods can also be used to solve other free boundary problems.

Melanoma is the deadliest form of skin cancer. Prognosis for early stage melanoma patients is excellent, and surgery is often curative for these patients. However, once patients have presented with invasive disease, the average 5-year survival rate drops significantly from over 90% to between 10 and 15%. Several therapies have been developed to target a commonly mutated oncogene BRAF, or its downstream effectors. Unfortunately, while these treatments show robust initial response, most patients relapse within a year. Moreover, therapy-resistant tumors are often more invasive and metastatic. Therefore, it is important to investigate the molecular mechanisms underlying melanoma invasion and metastasis, and to prevent melanoma cell dissemination and metastatic progression. Invadopodia are proteolytic membrane protrusions used by metastatic cancer cells to degrade the extracellular matrix and to facilitate cancer cell invasion and metastasis. In my thesis research I have focused on protein fucosylation and store-operated calcium entry, two separate mechanisms involved in invadopodial regulation. Post translational modifications of proteins are essential for their structure and function. Many cell surface proteins require modifications such as glycosylation for protein-protein interactions, cell adhesion, and signal transduction. Fucosylation is a form of glycosylation that adds L-fucose on glycan structures of proteins. There is evidence indicating that fucosylation plays an important but cancer-type and branching dependent role in cancer progression. Emerging evidence indicates that the fucose salvage pathway and protein fucosylation are altered during melanoma progression and metastasis. Here, we report that the fucose salvage pathway inhibits invadopodia formation and extracellular matrix degradation by promoting α(1,2) fucosylation of cell surface proteins. The activation of the fucose salvage pathway decreases invadopodia numbers and inhibits the proteolytic activity of invadopodia in WM793 melanoma cells. Inhibiting fucokinase, one of the critical enzymes in the fucose salvage pathway, in melanoma cells abrogates L-fucose-mediated inhibition of invadopodia, suggesting dependence on the fucose salvage pathway. The inhibition of invadopodia formation by L-Fucose treatment or fucokinase overexpression could be rescued by treatment with α(1,2), but not α(1,3/4) fucosidase, implicating an α(1,2) fucose linkage-dependent inhibitory effect. The ectopic expression of FUT1, an α(1,2) fucosyltransferase, is sufficient to inhibit invadopodia formation and ECM degradation. Our findings indicate that the fucose salvage pathway can inhibit invadopodia formation, and consequently, invasiveness in melanoma via α(1,2) fucosylation. Re-activation of this pathway in melanoma could be useful for preventing melanoma invasion and metastasis. Calcium is a critical second messenger involved in a multitude of biological processes from cell proliferation to muscle contraction. In melanoma, previous studies have found that activation of the store operated calcium entry (SOCE) channel promotes tumor invasion and metastasis, in vitro and in xenograft models. The expression levels of STIM1, an essential component of the store operated calcium channels, has been found to increase with later stages of melanoma. In melanoma cell lines, the over expression of STIM1 enhances invadopodia number whereas STIM1 knockdown inhibits invadopodia formation. Similarly, gelatin degradation activity is enhanced with STIM1 overexpression and abrogated with STIM1 knockdown, implicating STIM1 as an important factor in the regulation of invadopodia formation and melanoma invasion. Though the studies published have shown a significant role of STIM1 in tumor progression, a robust transgenic animal model has not yet been established. Here, we developed a novel transgenic mouse model which, upon 4-hydroxytamoxifen (4OHT) treatment, induces the BRAFV600E mutation and PTEN, STIM1, and STIM2 deletions in melanocytes via an inducible Cre-lox system. Our investigation found that the loss of STIM1 exacerbates tumor growth and results in tumor formation significantly more quickly than STIM1 wild type mice. Whereas PCR analysis of 4OHT-treated skin showed deletion of STIM1 and PTEN, immunohistochemical staining of these genes in tumors did not convincingly demonstrate complete deletion. Therefore, it remains to be determined whether the effects we observed are due to STIM1 and STIM2 loss. These findings need to be corroborated in the future. Our studies focus on two important mechanisms required for melanoma progression and metastasis. We found that α(1,2) fucosylation is able to inhibit invadopodia formation, and melanoma cell invasion. The reestablishment of α(1,2) fucosylation in melanoma could potentially be exploited to inhibit melanoma metastasis. Additionally, early evidence points to STIM1 having a tumor suppressive role in melanoma oncogenesis and tumor growth based on the transgenic mouse model. Although the phenotype is unexpected, further investigation of this model will likely provide important insight for the complicate roles of SOCE in melanoma initiation and progression.

In vitro, la dégradation de la matrice extracellulaire (MEC) par ces cellules invasives est effectuée par les invadopodes, des protrusions cellulaires dotées d’une activité de protéolyse émanant de la membrane ventrale de cellules cultivées sur des substrats bidimensionnels. Le facteur d'échange (GEF) Fgd1, un GEF spécifique de Cdc42, a été montré comme étant un des composants des invadopodes. Nonobstant, la façon dont Fgd1 est régulé n’est pas encore connue. Mon projet visait à élucider la régulation de Fgd1 dans les cellules cancéreuses invasives. J'ai montré que la distribution subcellulaire de la protéine endogène varie au cours des différentes étapes de la formation des invadopodes et que Fgd1 s'accumule aux sites de formation des invadopodes et ce de façon concomittante à l'assemblage du noyau F-actine/cortactine. La partie N –terminale de la molécule est essentielle pour la localisation aux invadopodes. En outre, la Filamine A (FLNa) a été identifiée comme un nouveau partenaire de liaison in vitro et in vivo. J’ai de plus montré que FLNa colocalise avec Fgd1 aux invadopodes et est nécessaire pour leur formation et leur fonction. J'ai émis l'hypothèse que FLNa puisse être impliquée dans la dégradation de Fgd1 par le protéasome, un processus étroitement lié au niveau de l'activation de Cdc42. Dans mon modèle, FLNa agirait comme une protéine d’échafaudage pour connecter Fgd1 et Cdc42 et ainsi réguler localement l'activation de la GTPase, mais également pour stabiliser Fgd1 en empêchant sa dégradation par le protéasome. Dans l'ensemble, mes résultats apportent de nouveaux éléments concernant le rôle de Fgd1 dans la biogénèse des invadopodes / In vitro, degradation of the extracellular matrix (ECM) by invasive tumor cells is carried out by invadopodia, proteolytically active protrusions emanating from the ventral membrane of cells cultured on two dimensional ECM substrates. Identification and characterization of novel molecular components of the invadopodia machinery is now witnessing a relevant interest. The guanine nucleotide exchange factors (GEFs) Fgd1, a specific GEF for Cdc42, has been recently brought into the picture of invadopodia components and regulators. Notwithstanding, how Fgd1 is regulated in invasive cancer cells remains poorly understood. My project aimed to further elucidate Fgd1 regulation in cancer cells. I showed that the cellular distribution of endogenous Fgd1 quantitatively changes along the different stages of invadopodia formation, and that Fgd1 accumulates at invadopodia sites concomitantly with the initial assembling of the actin/cortactin core. The N-terminal proline rich domain (PRD) of Fgd1 is essential for its localization and function at invadopodia. Furthermore, using a yeast two hybrid approach, Filamin A (FLNa) was identified as a novel Fgd1 binding partner and this interaction was validated in vivo. I report that FLNa co-localizes with Fgd1 at invadopodia and is required for their formation and function. I hypothesized that FLNa may be involved in the SCF FWD1/β-TrCP –mediated proteasome degradation of Fgd1, a process that is strictly connected to the Cdc42 activation rate. In my model, FLNa may act as a scaffold to connect Fgd1 and Cdc42 for local activation of the small GTPase, and to increase Fgd1 stability by preventing its proteasomal degradation. Taken together, my findings provide novel insights on the role of Fgd1 in invadopodia biogenesis

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第18502号 / 医博第3922号 / 新制||医||1005(附属図書館) / 31388 / 京都大学大学院医学研究科医学専攻 / (主査)教授 武藤 学, 教授 千葉 勉, 教授 松田 道行 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

No description available.

L'invasion cellulaire est un processus basé sur la dynamique des invadosomes, qui sont desstructures acto-adhesives capables de dégrader localement la matrice extracellulaire. Lesinvadosomes sont composés d'un coeur riche en actine-F entouré par un assemblage desprotéines d'adhésion conduisant à la dégradation de la matrice extracellulaire par lerecrutement et la sécrétion des protéases. Les cellules MEF exprimantes la formeconstitutivement active de la kinase Src (SrcY527F) forment des invadosomes quis'organisent sous forme d'anneaux ou rosettes. Paxilline et hic-5 (une protéine homologue àla paxilline) ont des rôles spécifiques dans la morphologie cellulaire et la plasticité pendantl'invasion cellulaire. La structure de ces deux protéines se caractérise par la présence dedomaines LIM dans la partie C-terminale et de motifs LD dans la partie N-terminale. Cettesimilarité suggère que ces protéines peuvent avoir des fonctions redondantes. L'importance dela famille des protéines paxillines dans la formation des invadosomes a été examiné en tentantd'induire des invadosomes via l'expression de la forme constitutivement active de la kinaseSrc (SrcY527F) dans des cellules déplétées en paxilline et hic-5 (cellules pax-/- et déplétées enhic-5 par shRNA). La formation des invadosomes a été totalement bloquéequand l'expression de ces deux protéines a été diminuée de manière significativesimultanément indiquant leur redondance fonctionnelle. La ré-expression de nombreuxmutants de la paxilline dans ce contexte cellulaire a permis de montrer que les domaines LIMde paxilline n'étaient pas nécessaires pour la formation de l'anneau des invadosomes. Aucontraire, ces études de structure-fonctions ont permis de mettre en évidence le rôle essentieldes motifs LD3 et LD5 considérablement dans l'assemblage des rosettes. En outre, l'ordreprécis de chaque LD dans la molécule de paxilline est essentiel pour leur travail coopératifdans la régulation des invadosomes. Après avoir montré la redondance fonctionnelle entrepaxillin et hic-5, nous avons voulu déterminer leurs fonctions spécifiques. Pour cela,différents traitements siRNAs nous ont permis de diminuer spécifiquement l'expression depaxillin et/ou hic-5 dans des cellules formant des invadosomes. La déplétion rapide de lapaxilline a réduit la formation des rosettes alors que celle de hic-5 a affecté de manièreimportante la dégradation de la matrice extracellulaire. En effet, l'identification despartenaires spécifiques de paxilline et hic-5 à révélé que cette dernière interagissaitspécifiquement avec IQGAP1. Cette protéine est un facteur essentiel pour le recrutement del'exocyst, un complexe régulant la sécrétion locale de métalloprotéases dans lesinvadosomes. En conclusion, il apparaît que l'action complémentaire de la paxilline et hic-5 aun rôle essentiel dans la régulation des fonctions cellulaires assurées par les invadosomes. / Cellular invasion is based on invadosome dynamics where acto-adhesive machinery iscoupled with extracellular matrix proteolysis. Each invadosome unit is composed by a denseF-actin core surrounded by a ring of adhesion molecules that localizes intense ECMdegradation activity via the recruitment and secretion of lytic enzymes. Invadosomes in MEFstransformed with an activated form of the Src kinase (SrcY527F) auto-assemble into circularmeta-structures called rings or rosettes. Paxillin and the closely related family member hic-5(hydrogen peroxide inducible clone 5) have been recently identified as critical determinants ofcell morphology and plasticity during cell invasion with distinct signaling functions.However, the two proteins have a similar overall domain structure, including amino-terminalLD motifs and carboxyl-terminal LIM domains suggesting overlapping or redundantfunctions. We examined the importance of this class of proteins in invadosome formation, byexpressing SrcY527F in pax-/- cells depleted or not of hic-5. Only when the expression of bothproteins was significantly decreased, invadosome formation was blocked indicating theirfunctional redundancy. Importantly, LIM domain of paxillin was dispensable for ringformation whereas individual depletion of LD3 and LD5 motifs dramatically reducedinvadosome rosette assembly. Furthermore, the precise order of each LD in the paxillin'smolecule was necessary to allow their cooperativity during invadosome formation. Beyondtheir redundancy, their distinct roles in invadosomes were assessed via siRNA strategy aimingat the acute depletion of each molecule. Rapid depletion of paxillin reduced invadosome ringformation whereas hic-5 silencing dramatically affected extracellular matrix degradation. Hic-5 role in ECM degradation was enhanced by the fact that it was found to specifically interactwith IQGAP1, a CDC42- effector that is essential to recruit the exocyst complex ininvadosomes. In summary, the functional redundancy of paxillin and hic-5 suggests anextensive cross-talk between these two closely related proteins in the regulation ofinvadosome-based cellular functions.

A disseminação metastática de células tumorais malignas envolve múltiplas etapas e é uma das principais causas de mortalidade. O microambiente tumoral apresenta um importante papel no processo de tumorigênese. A laminina é um componente do microambiente que pode ser clivado em peptídeos bioativos, como o peptídeo AG73 (domínio globular da cadeia 1) que aumenta a formação/atividade de invadopódios. Invadopódios são protrusões de membrana ricas em actina com atividade proteolítica da matriz pericelular. Além da actina, os invadopódios exibem outras proteínas importantes, como a cortactina e o MT1-MMP. Recentemente, descobriu-se que a podoplanina pode desempenhar um papel importante na atividade dos invadopódios. A podoplanina é uma glicoproteína transmembrana do tipo I, intimamente associada à progressão maligna do câncer. A podoplanina e o peptídeo AG73 influenciam a biologia do câncer. Tal particularidade levou-nos a investigar o papel do peptídeo AG73 na regulação da da podoplanina e invadopódios em linhagem celular derivada de carcinoma de células escamosas oral (Cal 27). As células foram cultivadas em DMEM com SFB e tratadas com o peptídeo AG73. As células tratadas por peptídeo de sequência embaralhada serviram como controles. Imunofluorescência foi realizada para analisar os níveis de proteína de podoplanina, cortactina e MT1-MMP. Os resultados mostraram que o peptídeo AG73 diminuiu os níveis de podoplanina nas células Cal27 em comparação aos controles. Nenhuma alteração foi observada em relação à cortactina e MT1-MMP. Estes resultados foram confirmados por imunofluorescência. As células tratadas com AG73 diminuíram a distribuição da podoplanina em todo o citoplasma em comparação com os controles. Nossos resultados sugerem que o peptídeo derivado da laminina AG73 inibe a podoplanina, uma molécula reguladora do câncer, em células malignas humanas. / The metastatic spread of malignant tumor cells involves multiple steps, and is one of the major causes of mortality. The tumor microenvironment has an important role in tumorigenesis process. Laminin is a component of the microenvironment that can be cleaved into bioactive peptides, such as peptide AG73 (globular domain of 1 chain) that increase invadopodia formation/activity. Invadopodia are actin-rich membrane protrusions with proteolytic activity of peri-cellular matrix. In addition to actin invadopodia exhibit other key proteins such as cortactin and MT1-MMP. Recently it has been discovered that podoplanin may play an important role in the invadopodia activity. Podoplanin is a type I transmembrane glycoprotein, closely associated with the malignant progression of cancer. Podoplanin and the peptide AG73 influence cancer biology. This prompted us to investigate the role of peptide AG73 in the regulation of podoplanin and invadopodia formation in a cell line derived from oral squamous cell carcinoma (Cal 27). Cells were cultured in DMEM with FBS and treated with peptide AG73. Cells treated by scrambled peptide served as controls. Immunoblot was carried out to analyze protein levels of podoplanin, cortactin and MT1-MMP. Results showed that the peptide AG73 decreased podoplanin levels in Cal27 cells compared to controls. No alterations were observed with regard to cortactin and MT1-MMP. These results were confirmed by immunofluorescence. Cells treated by AG73 decreased podoplanin distribution throughout the cytoplasm compared to controls. Our results suggest that the laminin-derived peptide AG73 inhibits podoplanin, a cancer regulator molecule, in human malignant cells.

La migration et l'invasion tumorale sont des étapes clés de la cascade métastatique. Les microtubules (MT) contribuent à la division cellulaire et constitue la cible des agents de chimiothérapie anti-MT (ACM). Ce sont des structures dynamiques qui s'ancrent aux structures cellulaires périphériques. Durant ma thèse, j’ai étudié comment les protéines régulant le bout « + » des MT (+TIP) contribuent à la migration cellulaire et à la dégradation de la matrice extracellulaire. D’abord j’ai étudié l'impact de l'eribuline, un ACM dépolymérisant, sur la migration de cellules mammaires. L'éribuline s'est avérée empêcher l'ancrage des MT, modifier leur dynamique et inhiber la migration dirigée cellulaire, phénomène que nous avons expliqué par son action sur la +TIP EB1 mais surtout par la délocalisation de la tubuline polymérase ch-TOG de l'extrémité + des MT. Puis, nous avons examiné le rôle des +TIP dans la dégradation de la matrice, par les invadopodes, de petites protrusions riches en actine dégradant la matrice. La déplétion de EB1 et ses partenaires, APC et ACF7, régulaient négativement l’action des invadopodes, laissant supposer la présence d'un complexe fonctionnel : EB1, APC et ACF7 régulant négativement les invadopodes. En parallèle, par analyse protéomique systématique des composant des invadopodes, nous avons identifié de nouveaux proches voisins de TKS5, protéines indispensable à la formation des invadopodes, dont une protéine associée aux MT, MAP4. Au total, la régulation de la dynamique des +TIP est importante pour la migration et l'invasion et développer des stratégies ciblées contre ces acteurs pourrait améliorer la prise en charge du cancer du sein métastatique. / Migration and invasion are key steps in the metastatic cascade. Microtubules (MT) are involved in cell division and are dammaged by MT tagetting agents(MTA), a widely used chemotherapy drugs. MT are dynamic structures anchored to peripheral cell structures. During this work, I studied how proteins that regulates the "+" end of MT (+ TIP) cell migration and extracellular matrix degradation. First I adressed the impact of eribulin, a new depolymerizing MTA, on mammary cell migration. Eribulin was found to prevent the anchoring of MT to cell cortex, to modify their dynamics and to inhibit cell migration, a phenomenon that we explained by its action on +TIP EB1 but more precisely by the delocalization of tubulin polymerase ch-TOG. Next we investigated the role of TIPs in invadopodia matrix degradation , which are actin-rich protrusion specialized in matrix digestion. The depletion of EB1 and its partners, APC and ACF7, negatively regulated the action of invadopodia, assumed the presence of a complex complex: EB1, APC and ACF7 negatively regulating invadopodia activity. In parallel, by systematic proteomic analysis of the component of the invadopodia, we identified new close neighbors of TKS5, an essential proteins in invadopodia formation, including a MT associated protein MAP4. In conclusion, the regulation of + TIP dynamics is important for migration and invasion and developping targeted strategies against them could improve the management of metastatic breast cancer.

Les cellules BHK transformées par l'oncogène v-Src forment des invadopodia qui s'organisent successivement sous forme de paquets, anneaux et enfin ceintures d'invadopodia. L'expansion des anneaux d'invadopodia est due à la néoformation d'invadopodia à la périphérie de l'anneau et au désassemblage simultané des invadopodia situés au centre de l'anneau. L'orthovanadate, inhibiteur de tyrosine phosphatases, génère des expansions très rapides indiquant l'implication de phosphorylations sur tyrosine dans la formation des invadopodia à la périphérie et leur désassemblage au centre. La paxilline, une protéine hautement phosphorylée, responsable du désassemblage des adhérences focales, est également présente dans les invadopodia et induit le désassemblage des invadopodia au centre de l'anneau (processus indispensable à la formation et à l'expansion des anneaux), grace a un processus de phosphorylation de la paxilline sur les tyrosines 31 et 118, qui en retour active la MAP kinase Erk et la calpaine, responsable du clivage protéique des composants des invadopodia.