Le phénotype des chondrocytes peut être modulé par des facteurs de croissance comme par des contraintes mécaniques. Nous avons caractérisé le potentiel chondrogénique de la "Bone Morphogenetic Protein" (BMP)-2 sur des chondrocytes murins primaires amplifiés en culture monocouche sur plastique. Nous avons aussi développé un nouveau modèle d’étude de la mécanotransduction par la compression dynamique de ces cellules incluses en hydrogel d’agarose. Nous avons ainsi confirmé l’implication des voies ERK1/2 et p38 dans ces mécanismes, révélé l’activation de Smad2/3 en réponse à la compression et identifié de nouveaux gènes mécano-sensibles. Par ailleurs, nous avons mis en évidence le rôle des intégrines-bêta-1 dans la rigidité du tissu cartilagineux. Nos résultats complètent la connaissance fondamentale des mécanismes de régulation du phénotype chondrocytaire, mais peuvent également contribuer à l'amélioration des techniques de reconstruction du cartilage dans le domaine de l'ingénierie tissulaire. / Chondrocytes phenotype can be modulated by growth factors as well as mechanical stress. We characterised Bone Morphogenetic Protein (BMP)-2 chondrogenic potential on mouse primary chondrocytes expanded in monolayer on culture plastic. Also, we developed a new model to investigate mechanotransduction by applying dynamic compression on these cells embedded in agarose hydrogel. Hence, we confirmed ERK1/2 and p38 pathways implication in such mechanisms, we revealed Smad2/3 activation in response to compression and we identified new mechanosensitive genes. Besides, we highlighted the role of beta-1-integrins in cartilage stiffness. Our results completed the basic knowledge of regulation mechanisms underlying chondrocytes phenotype, but could also contribute to improve techniques for cartilage reconstruction in the field of tissue engineering.

La mécanotransduction correspond à un processus dans lequel la force physique est convertie en signal chimique ou électrique. Ce processus est à la base de nombreuses fonctions physiologiques, y compris le sens du toucher, l’audition, la proprioception et la nociception. Nous ne connaissons pas à ce jour les mécanismes moléculaires à l’origine de la diversité fonctionnelle des mécanorécepteurs. L’objectif de thèse était de fournir 1 caractérisation des canaux mécanosensibles des neurones sensoriels afin d’identifier les mécanismes responsables des propriétés des mécanorécepteurs. 4 types de courants excitateurs ont été identifiés et classés sur la base de leurs cinétiques de relaxation: des courants à relaxation rapide, intermédiaire, lente ou ultra-lente. La relaxation résulte de l’adaptation et de l’inactivation. Nous montrons également que ces courants mécanosensibles possèdent des propriétés spécifiques permettant le codage des différents paramètres du stimulus mécanique. Tous s’activent graduellement en fonction de l’intensité du stimulus mécanique, mais seuls les courants à relaxation lente et ultralente informent sur la persistance du stimulus. A contrario, les courants à relaxation rapide et intermédiaire sont mis en jeu essentiellement par des stimulations rapides, ils traduisent donc la rapidité d’installation du stimulus. Nous avons ensuite identifié un nouveau courant mécanosensible potassique (IKmech) exerçant un effet inhibiteur sur la décharge des mécanorécepteurs. Le profil pharmacologique et les travaux menés sur des souris KO et transgéniques montrent que le courant IKmech est porté par la sous-unité Kv1.1 qui est mécano-susceptible via un mécanisme par lequel la pression altère la sensibilité au potentiel des canaux. En s’opposant aux courants excitateurs, le courant IKmech régule le seuil de décharge des mécano-nocicepteurs et la fréquence de décharge des mécanorécepteurs non nociceptifs. / The somatosensory system mediates fundamental physiological functions, including the senses of touch, pain and proprioception. The aim of my thesis was to understand molecular mechanism of mechanotransduction in mammalian sensory neurons.We identified 4 types of mechanotransducer currents that distribute differentially in cutaneous nociceptors and mechanoreceptors and that differ in desensitization rates. Desensitization of mechanotransducer channels in mechanoreceptors was fast and mediated by channel inactivation and adaptation, which reduces the mechanical force sensed by the transduction channel. Both processes were promoted by negative voltage. These properties of mechanotransducer channels suited them to encode the dynamic parameters of the stimulus. In contrast, inactivation and adaptation of mechanotransducer channels in nociceptors had slow time courses and were suited to encode duration of the stimulus. Thus, desensitization properties of mechanotransducer currents relate to their functions as sensors of phasic and tonic stimuli and enable sensory neurons to achieve efficient stimulus representation.In the second work, we explored the molecular determinants of threshold differences and temporal adaptation among mammalian mechanoreceptors. We identified a novel mechanosensitive K+ current (IKmech) in different classes of mechanosensory neurons from mouse and rat DRGs. IKmech activates slowly in response to mechanical stimulation and is carried by Kv1.1 subunit-containing K+ channels. By antagonizing depolarizing drive induced by excitatory MS currents, IKMech regulates threshold for noxious mechano-perception and temporal adaptation in non-painful mechanosensation. Our work has identified Kv1.1 as an essential molecular element in defining the threshold range of mechanical sensitivity and temporal responses of fibers associated with mechanical perception.

Les événements biochimiques de l'activation T ont été décrits depuis longtemps et sont bien connus. A l'échelle moléculaire la liaison du Récepteur des Cellules T (TCR) présent à la surface du lymphocyte T avec un peptide (du soi ou non soi), ce dernier étant chargé sur le Complexe Majeur d'Histocompatibilité (MHC), présent à la surface des Cellules Présentatrices d'Antigène (CPA) conduit à l'initiation de la réponse immunitaire. La réponse des lymphocytes T est extrêmement spécifique, sensible, robuste et semble présenter des caractéristiques de mécano-transduction. En effet, il a été démontré récemment que le TCR agirait comme un mécanosenseur alors que le lymphocyte T peut sentir la mécanique de son environnement à une échelle cellulaire. Cependant, la majorité de ces études ont été réalisées en opposant un lymphocyte T à un substrat inerte limitant la compréhension sur la contribution éventuelle de chaque partenaire cellulaire car la présence de l'APC peut induire des changements dans l'organisation du lymphocyte T.Le but de cette thèse a été de mettre en place un suivi de l’activation des lymphocytes T pat une APC modèle, sous force grâce à l’utilisation de la microscopie à force atomique. Il a mené à plusieurs développements méthodologiques originaux validés de manière expérimentale avec un système cellulaire modèle (hybridomes murins vs. Cellules COS modifiées pour être des APC pouvant être modifiées à souhait grâce à l’expression d’une grande variété de molécules impliquées dans la réponse immunitaire). / The biochemical events of T activation have been described for a long time and are well known. At the molecular level, the binding of the T-cell receptor (TCR) present on the surface of the T-cell with a peptide (self or non-self) loaded on the Major Histocompatibility Complex (MHC), present on the surface of the Antigen Presenting Cells (APC), leads to the initiation of the immune response. In addition, the T cell response is extremely specific, sensitive, robust and appears to have mechano-transduction characteristics. Indeed, it has recently been demonstrated that the TCR would act as a mechanosensor while the T lymphocyte can feel the mechanics of its environment on a cellular scale. However, the majority of these studies were performed by opposing/facing a T cell to an inert substrate (e.g., bead, functionnlized glass slides molecularly decorated or not), limiting the understanding of the possible contribution of each cell partner because the presence of APC can induce changes in the organization of the T cell. The aim of this thesis was to set up a follow-up of the activation, under force, between a model APC and a T lymphocyte. It has led to several original methodological developments experimentally validated with a model cell system (mouse hybridomas vs. COS cells modified to be complexifiable APCs).

No sistema cardiovascular, modificações de pressão e shear stress devido ao fluxo sanguíneo influenciam a morfologia e a patofisiologia dos vasos sanguíneos e do coração. Neste trabalho, estudamos o papel de duas moléculas transmembrânicas do Sistema Renina-Angiotensina: a Enzima Conversora de Angiotensina (ECA) e o Receptor de Angiotensina do tipo I (AT1) como mecanosensoras e mecanotransdutoras dessas forças físicas. A ECA foi por muito tempo conhecida somente por sua ação em converter Angiotensina I em Angiotensina II e por inativar a Bradicinina. Recentemente foi demonstrado que a ECA, além dos efeitos enzimáticos já conhecidos, pode ter sua cauda citoplasmática fosforilada e desencadear vias de sinalização intracelular. Observamos que o shear stress, mas não o estiramento, induziu a diminuição da fosforilação da porção citoplasmática da ECA após 5 minutos de estímulo e se mantém até 18 horas. Demonstramos também que a porção extracelular da ECA tem papel fundamental como mecanosensora e que a via intracelular da JNK participa da mecanotransdução em resposta ao shear stress. Além disto, demonstramos que a diminuição da fosforilação da ECA está associada na diminuição da sua expressão pelo shear stress. O receptor AT1 é a principal molécula efetora das ações da angiotensina II. Recentemente foi demonstrado que esse receptor pode também ser ativado por forças físicas, estiramento celular, independentemente da presença da angiotensina II. No presente estudo, observamos que o receptor AT1 é ativado pelo shear stress e que o Candesartan, mas não o Losartan, é capaz de impedir esta resposta. A via intracelular ativada é dependente de proteína-G e da entrada de cálcio do meio extracelular. Interessantemente, a pré-exposicao dos receptores ao shear stress diminuem a responsividade dos receptores ao peptídeo Angiotensina II porém a Angiotensina II não é capaz de inibir a ativação pelo shear stress.. Em conjunto, demonstramos novos mecanismos de ação da ECA e do AT1 que são duas importantes moléculas do sistema renina angiotensina. A modulação destes componentes por estímulos mecânicos traz novas possibilidades de intervenções farmacologicas sobre esse sistema bem como o melhor entendimento da participação dessas moléculas na fisiopatologia cardiovascular. / Hemodynamic forces such as pressure and shear stress modulate the patophysiolgy of the cardiovascular system. In this study, we investigated two transmembranic key molecules of the renin-angiotensin system (RAS) as mechanosensors and mechanotransducers of physical forces: Angiotensin Converting Enzyme (ACE) and Angiotensin II type 1 Receptor (AT1). ACE is an enzyme that converts angiotensin I in angiotensin II. Recently, it was demonstrated that ACE cytoplasmic tail can be phosphorylated by ACE inhibitors and elicited intracellular cell signaling. Here, we observed that shear stress, but not stretch, decreased ACE cytoplasmic phosphorylation after 5 minutes and maintained up to 18 hours. ACE extracellular portion act as mechanosensor and JNK pathway participate in the mechanotransduction activation. In addition, we also demonstrate that decrease in ACE phosphorylation is involved in ACE expression downregulation by shear stress. AT1 receptor is the main effector molecule of angiotensin II cellular responses. It has recently been shown that AT1 receptor can directly be activated by mechanical stretch stress through an angiotensin-II-independent mechanism. In the present study, we observed that shear stress also activates AT1 receptor which is blocked by Candesartan, but not by Losartan. The intracellular pathway activated by shear stress involves both G-protein and extracellular calcium. Interestingly, preconditioning of AT1 receptor by shear stress impairs its responsiveness to angiotensin II while the pretreatment with angiotensin II still allow AT1 responsiveness to shear stress. Altogether, we demonstrated that ACE and AT1 receptor activates intracellular signal in response to mechanical force. This new concept for the RAS, the modulation of these molecules by mechanical forces gives new insigh into the discovery for pharmacological interventions to the RAS

While a large body of work has focused on the transcriptional regulation of cellular identity, the role of the mechanical properties of cells and the importance of their physical interactions with the local environment remains less well understood. In this project, we explored the impact of cytoskeleton-generated forces exerted on the nucleus in the context of early embryonic stem (ES) cell fate decisions. We chose to perturb force generating components in the cytoskeleton – notably the molecular motor non-muscle myosin II - and key structural and chromatin binding proteins in the nuclear envelope, notably, the lamins (LMNA), Lamin B receptor (LBR) and components of the LINC complex (nesprins/KASH). The structural proteins in the nuclear envelope regulate both the mechanical response of the nucleus to force and the stabilization of peripheral heterochromatin (repressed genes). Our hypothesis is that reducing forces transmitted directly to chromatin or increasing tethering of peripheral heterochromatin to the nuclear envelope would restrict access to lineage specific genes sequestered at the nuclear lamina and thereby either impair, or delay, differentiation. We found phenotypes in the capacity of mouse ES cells to specify to the neural lineage following our perturbations: overexpression of LMNA, LBR and KASH proteins resulted in a significant fraction of cells that did not express the neuroectoderm marker Sox1 after four days of differentiation, while inhibiting non-muscle myosin II delayed Sox1 expression in the entire population. Overexpression of LMNA and LBR did not affect the ability of the cells to exit the naive pluripotent state, which raises the possibility that the perturbations are halting the cells in a formative phase prior to lineage specification. Future work will focus on looking at genome-wide transcriptional changes accompanying differentiation combined with an analysis of spatial information of differentially regulated genes.

As células endoteliais são capazes de converter o estímulo mecânico em sinais intracelulares e produzir fatores vasoativos como o óxido nítrico (oNO). Evidências recentes sugerem que as beta-arrestinas desempenham um papel importante não somente na dessensibilização e internalização de receptores acoplados à proteína G (GPCR) como também na mecanotransdução. Nós testamos a hipótese de que células endoteliais submetidas ao shear stress (SS) produzem oNO por meio da ativação da via de sinalização dependente de beta-arrestina. Para tal, células endoteliais de veia safena (hSVEC) foram transfectadas com siRNA contra as isoformas 1 e 2 da beta-arrestina e, posteriormente, submetidas ao SS (15 dinas/cm2) durante 10 min. Nós encontramos que as SVEC silenciadas para a beta-arrestina 1/2 (70%) exibiram uma menor produção de nitrito no meio de cultura em resposta ao SS (166±17 vs. 326±44% comparado com hSVEC transfectadas com siRNA controle). Além disso, o silenciamento da beta-arrestina 1 e 2 preveniu os níveis de fosforilação da Akt no resíduo de serina 473 e a fosforilação da eNOS no resíduo de serina 1177, enquanto que a fosforilação da ERK 1/2 manteve-se inalterada. Curiosamente, análises de imunoprecipitação mostraram que a beta-arrestina interage com caveolina-1, um mecanossensor do shear stress, mas não é influenciado pelo SS. Além disso, na situação estática, a beta-arrestina encontra-se em uma localização perinuclear e, após o SS, adquiriu um padrão mais difuso no citosol. Coletivamente, esses dados sugerem que a beta-arrestina e a sinalização downstream Akt/ eNOS são necessárias para a produção de oNO induzido por shear stress em células endoteliais vasculares humana / Endothelial cells are capable of converting mechanical stimuli into intracellular signals generating vasoactive factors such as nitric oxide (oNO). Recent evidence suggests that beta-arrestins play a role not only on G protein-coupled receptors (GPCR) desensibilization but also in mechanotransduction. We tested the hypothesis that beta-arrestin and its downstream signaling influence laminar shear stress (SS)-induced oNO production by endothelial cells. Towards this end, human saphenous vein endothelial cells (hSVEC) transfected with siRNA against beta-arrestins isoforms 1 and 2 were subjected to SS (15 dynes/cm2, 10 minutes). We found that the SS-induced production of nitrite in the cell culture medium from down-expressed beta-arrestin 1/ 2 (70%) SVEC decreased (166±17 vs. 326±44% compared to wild-type hSVEC; P < 0.001). The beta-arrestin 1 and 2 down-regulation in SVEC also inhibited the phosphorylation levels of Akt at the serine residue 473 and the phosphorylation levels of eNOS at the serine residue 1177, whereas ERK phosphorylation remained unchanged. Interestingly, immunoprecipitation analysis showed that beta-arrestin interacts with caveolin-1, a shear stress mechanosensor, which is not influenced by SS despite the fact that the static perinuclear localization of beta-arrestins changed to the cytosol upon SS. Collective these data suggest that beta-arrestin and Akt/eNOS downstream signaling are required for shear stress-induced nitric oxide production in human vascular endothelial cells

Les cellules vivantes sont capables de réagir aux signaux mécaniques tels que la rigidité de la surface sur laquelle elles adhèrent, les forces de tractions ou compressions auxquelles elles sont soumises, le flux de liquide à la surface de leur membrane ou encore la géométrie de leurs adhésions ou de leur forme globale. Ces signaux influent sur des processus cellulaires tels que la prolifération, la différenciation, la migration et la mort cellulaire. Ces processus sont finement régulés par des réactions biochimiques qui forment un réseau de signalisation. La mécanotransduction est la traduction du signal mécanique en signal biochimique.C’est dans le but d’étudier la mécanotransduction que nous avons étudié l’utilisation d’ultrasons pour stimuler mécaniquement les cellules à des fréquences temporelles et spatiales relativement élevées. De nombreux montages expérimentaux et de nombreuses voies ont été considérées dans cette partie très exploratoire. Nous en retenons finalement des pistes prometteuses pour la continuation future de ce projet.Nous avons développé ce que nous nommons des substrats actifs, qui nous permettent de contrôler à la fois spatialement et temporellement la stimulation mécanique appliquée à des cellules vivantes. Ces substrats actifs consistent en des micropiliers de fer incrustés dans un élastomère peu rigide (PDMS) et manipulés par deux électroaimants. Nous pouvons contrôler dynamiquement le déplacement des piliers qui vont déformer localement et de manière continue la surface. Cette déformation va ensuite déformer en traction ou en compression les cellules vivantes étalées sur la surface à proximité. En employant des marqueurs fluorescents nous pouvons réaliser de la Microscopie de Forces de Traction et surveiller la contrainte appliquée par les piliers aux cellules à travers la surface de PDMS, et nous pouvons étudier la réponse mécanique des cellules. De plus, ces substrats sont compatibles avec la microscopie de fluorescence en cellule vivante, ce qui rend possible l’observation de la réponse cellulaire au niveau morphologique (forme des adhésions focales, activité protrusive, …) et surtout biochimique.En effet, pour étudier la réponse biochimique des cellules après une stimulation mécanique, nous observons par microscopie de fluorescence des biosenseurs portant des paires de fluorophores donneur/accepteur. Ces biosenseurs nous permettent d’observer l’activité de protéines impliquées dans la signalisation cellulaire en calculant l’efficacité de Transfert d’Énergie Résonnant de Förster (FRET) de ces biosenseurs. Pour ce faire, les échantillons sont illuminés alternativement aux longueurs d’ondes d’excitation des fluorophores donneurs puis accepteurs. Le signal de fluorescence est collecté simultanément dans un canal d’émission du donneur et un canal d’émission de l’accepteur. Une grande partie de ma thèse a été consacrée à la mise au point d’une méthode quantitative pour analyser les images de fluorescence afin de mesurer une efficacité de FRET qui ne dépende pas de facteurs expérimentaux ni de la quantité de biosenseurs présents dans les cellules. Nous évaluons alors les différentes méthodes pour déterminer les facteurs de correction répandus corrigeant le débordement de spectre du donneur dans le canal accepteur et l’excitation directe de l’accepteur à la longueur d’onde d’excitation du donneur. Pour obtenir des mesures plus quantitatives, nous avons mis au point une nouvelles méthode pour déterminer 2 facteurs de correction supplémentaires. Nous comparons cette méthode à la seule préexistante et évaluons l’influence des paramètres de traitement des images sur les valeurs d’efficacité de FRET mesurées. / Living cells can react to mechanical signals such as the rigidity of the surface they adhere on, the traction or compression forces applied on them, the liquid flow at their membrane surface or the geometry of their adhesions or of their overall shape. Those signals influence cellular processes such as proliferation, differentiation, migration or cell death. Those processes are tightly regulated by biochemical reactions that constitute a signaling network. Mechanotransduction is the translation of the mechanical signal into the biochemical one.In order to study mechanotransduction, we have considered the use of ultrasounds to mechanically stimulate cells at relatively high temporal and spatial frequencies. Numerous setups and options have been considered in this very exploratory project. Finally, we will retain some promising leads for the continuation of this project.We have developed what we call active substrates that allows us to control both spatially and temporally the mechanical stimulation on living cells. Those active substrates consist of iron micropillars embedded in a soft elastomer and actuated by 2 electromagnets. We can control dynamically the displacement of the pillar that will deform locally and continuously the surface. This deformation will then deform in traction or in compression the living cells spread on the surface nearby. Thanks to fluorescent trackers we can perform Traction Force Microscopy and monitor the stress applied by the pillars to the cells through the PDMS surface, and we can look at the mechanical response of the cells. Moreover, those substrates are compatible with live cell fluorescence microscopy, which makes possible the observation of the cellular response at the morphological level (focal adhesions, protrusive activity, …) and most importantly at the biochemical level.Indeed, in order to study the cellular biochemical response after a mechanical stimulation, we use fluorescence microscopy to observe biosensors containing pairs of donor/acceptor fluorophores. Those biosensors allow us to monitor the activity of proteins implied in cellular signaling by computing the Förster Resonance Energy Transfer (FRET) efficiency of those biosensors. To do so, samples are alternatively excited at donor and acceptor excitation wavelengths. The fluorescence signal is then simultaneously measured in donor and acceptor emission channels. A substantial part of my thesis has been dedicated to the development of a quantitative method to analyze fluorescence images in order to measure FRET efficiencies that do not depend on experimental factors or biosensors concentration in cells. We assess different methods to compute standard correction factors that account for spectral bleed-through and direct excitation of acceptors at donor excitation wavelength. To obtain more quantitative measurements, we have developed a new method to compute 2 additional correction factors. We compare this method with the only one preexisting, and we assess the influence of image processing parameters on FRET efficiency values.

Fibronectin is an important protein that is able to bind to other fibronectin molecules and to cell surface receptors. In doing so, the interactions fibronectin can perform is important for the processes of cell migration and tissue formation. Understanding the properties of fibronectin and fibril assembly is useful for areas such as wound healing, where fibronectin molecules are assembled to protect the tissue and to perform other tasks. Because of these reasons, it is important to understand how fibronectin is assembled and how its properties affect the fibril assembly, which in return affects the way the cell matrix operates. Previously published papers illustrate that the properties of fibronectin affect the mechanotransduction process, the cell conversion of mechanical stimulus to chemical, and this leads to various changes of the fibril assembly. However, the question that now comes to focus is what variables affect the fibril assembly? The two main variables that come into question is the substrate stiffness (ksub) (pN/nm) and the actin velocity (Vu) (nm/s). In order to test this hypothesis, several fibril assembly simulations were performed via MATLAB based upon the Weinberg-Mair-Lemmon Fibronectin Model. These simulations were performed by varying the parameters of substrate stiffness and actin velocity as well as fibril size, which affect the various measurements of the fibronectin, such as stretched length, relaxed length, etc. Through these various experiments, it was determined that the actin velocity and fibril size had the greatest impacts in affecting the fibronectin’s properties and its assembly.

Epithelial-mesenchymal transition (EMT) regulates the cellular processes of migration, growth, and proliferation - as well as the collective cellular process of tissue remodeling - in response to mechanical and chemical stimuli in the cellular microenvironment. Cells of the epithelium form cell-cell junctions with adjacent cells to function as a barrier between the body and its environment. By distributing localized stress throughout the tissue, this mechanical coupling between cells maintains tensional homeostasis in epithelial tissue structures and provides positional information for regulating cellular processes. Whereas in vitro and in vivo models fail to capture the complex interconnectedness of EMT-associated signaling networks, previous computational models have succinctly reproduced components of the EMT program. In this work, we have developed a computational framework to evaluate the mechanochemical signaling dynamics of EMT at the molecular, cellular, and tissue scale. First, we established a model of cell-matrix and cell-cell feedback for predicting mechanical force distributions within an epithelial monolayer. These findings suggest that tensional homeostasis is the result of cytoskeletal stress distribution across cell-cell junctions, which organizes otherwise migratory cells into a stable epithelial monolayer. However, differences in phenotype-specific cell characteristics led to discrepancies in the experimental and computational observations. To better understand the role of mechanical cell-cell feedback in regulating EMT-dependent cellular processes, we introduce an EMT gene regulatory network of key epithelial and mesenchymal markers, E-cadherin and N-cadherin, coupled to a mechanically-sensitive intracellular signaling cascade. Together these signaling networks integrate mechanical cell-cell feedback with EMT-associated gene regulation. Using this approach, we demonstrate that the phenotype-specific properties collectively account for discrepancies in the computational and experimental observations. Additionally, mechanical cell-cell feedback suppresses the EMT program, which is reflected in the gene expression of the heterogeneous cell population. Together, these findings advance our understanding of the complex interplay in cell-cell and cell-matrix feedback during EMT of both normal physiological processes as well as disease progression.

O câncer de mama é o tipo mais freqüente e o segundo mais letal no mundo. Embora ass taxas de sobrevida dos pacientes tenham aumentado consideravelmente nas últimas décadas, indicadores prognósticos desfavoráveis são associados a pacientes com diagnóstico em fase avançada e presença de metástases, frequentemente associadas à existência de células-tronco tumorais (CTT). As CTT são indiferenciadas e capazes de autorrenovação e diferenciação, o que as torna fundamentais para a manutenção da heterogeneidade celular intratumoral. As CTTs são altamente invasivas, tumorigênicas e resistentes a tratamentos convencionais, sendo frequentemente associadas ao surgimento de metástase e recidiva após tratamento. O microambiente tumoral modula as CTT por meio de células e da matriz extracelular (MEC), uma estrutura biologicamente dinâmica, complexa e que regula processos celulares como migração, invasão e diferenciação. A MEC é composta por uma grande variedade de moléculas, peptídeos e macromoléculas, sendo o colágeno seu componente mais abundante. A alta densidade mamográfica é frequentemente associada a elevada rigidez da MEC e deposição aumentada de colágeno fibrilar, principalmente colágeno tipo I (Col I), e é um dos maiores fatores de risco independentes para o desenvolvimento do câncer de mama. A alta densidade de Col I e rigidez da MEC também está associada à maior agressividade tumoral e metástase. Col I também induz o fenótipo tronco tumoral em diversos tipos celulares tumorais, embora o papel da densidade sobre este efeito seja pouco esclarecido. Nosso estudo avaliou a hipótese de a alta densidade de Col I induzir o fenótipo tronco tumoral. Cultivamos linhagens normais (MCF-10A) e tumorais (MDA-MB-231 e MCF-7) de mama em géis de baixa, média e alta densidade de Col I. Também cultivamos células em superfície bidimensional (2D) e em suspensão para geração de mamoesferas (ME), representando o cultivo tradicional e de enriquecimento de CTTs, respectivamente. Avaliamos os níveis do imunofenótipo tronco (CD44+CD24-), expressão gênica e proteica de marcadores de CTTs e de resposta mecânica ao substrato (mecanotransdução), bem como potencial clonogênico, autorrenovação celular e alinhamento fibrilar de géis de Col I. Alta densidade de Col I elevou os níveis da subpopulação CD44+CD24- e inibiu o alongamento celular da linhagem MDA-MB-231, porém não modulou a expressão de marcadores de CTT, bem como potencial clonogênico, autorrenovação celular e alinhamento fibrilar de géis de Col I. A alta densidade de Col I induziu aumento dos níveis totais da isoforma variante da glicoproteína CD44 (CD44v), receptor de estrógeno (RE &#945) e do fator de pluripotência Sox2 em linhagem MCF-7 derivada de ME. Entretanto, os níveis nucleares dos fatores de transcrição (RE &#945 e Sox2) permaneceram inalterados. Em comum, a alta densidade de Col I não elevou os níveis nucleares do mecanotransdutor YAP em linhagens MDAMB-231 e MCF-7 derivada de ME. Concluímos que a alta densidade de Col I induz parcialmente o fenótipo molecular, mas não o funcional, de células tumorais mamárias. / Breast cancer is the most frequent and second deadliest cancer type worldwide. Although patient survival rates have increased considerably in recent decades, unfavorable prognostic indicators are associated with patients with advanced disease stage at diagnosis and presence of metastases, frequently associated with the existence of cancer stem cells (CSC). CSC are undifferentiated and capable of self renewal and differentiation, making them fundamental for the maintenance of intratumoral cellular heterogeneity. CTTs are highly invasive, tumorigenic and resistant to conventional treatments, and are frequently associated with the onset of metastasis and relapse after treatment. The tumor microenvironment modulates CTT by means of cells and the extracellular matrix (ECM), a biologically complex and dynamic structure that regulates cell processes such as migration, invasion and differentiation. ECM is composed of a large variety of molecules, peptides and macromolecules, with collagen being its most abundant component. High mammographic density is often associated with high MEC stiffness and increased deposition of fibrillar collagen, mainly type I collagen (Coll I), and is one of the main independent risk factors for breast câncer development. High Coll I density and ECM stifness are also associated with increased tumor aggressiveness and metastasis. Coll I also induces tumor stemness in several tumor cell types, although the role of its density on this effect is unclear. Our study evaluated the hypothesis that high Coll I density induces the tumor stemness. We cultured normal-like- (MCF-10A) and tumoral (MDA-MB-231 and MCF-7) breast cell lines in low-, medium- and high-density Coll I gels. We also cultured cells in twodimensional (2D) surface and in suspension for the generation of mammospheres (MS), representing the traditional cell culture and CSC enrichment, respectively. We evaluated the levels of the CSC immunophenotype (CD44+CD24), gene/protein expression of CSC markers and mechanical response to the substrate (mechanotransduction), as well as the clonogenic potential, cell self-renewal and fibrillar alignment of Col I gels. High Coll I density increased the levels of the CD44+CD24- subpopulation and inhibited cell elongation of the MDA-MB-231 cell line, but did not modulate the expression of CSC markers as well as clonogenic potential, cell self-renewal and fibrillar alignment of Col I gels. High Coll I density increased total levels of the variant CD44 glycoprotein (CD44v), estrogen receptor (ER) and the pluripotency factor Sox2 in MS-derived MCF-7. However, the nuclear levels of the transcription factors (ER &#945 and Sox2) remained unchanged. In common, high Coll I density did not increase nuclear levels of the mechanotransducer YAP in MDA-MB- 231 and MS-derived cell lines. We conclude that high Coll I density partially induces the molecular stemness, but not the functional, phenotype of mammary tumor cells.