Branched actin networks play an important role in cellular processes ranging from cell motility, endocytosis, and adhesion. The Actin-related protein 2/3 (Arp2/3) complex nucleates actin branches from the sides of existing actin filaments. Arp2/3 complex is highly regulated and requires association with ATP, actin monomers, actin filaments and a class of proteins called nucleation promoting factors (NPFs) to undergo an activating conformational change where the actin-related subunits, Arp2 and Arp3, arrange into a filament-like conformation that templates a new actin branch. While some progress has been made, the individual roles of each of these factors on the activating conformational change is poorly understood. In addition, it is still unclear how Arp2/3 complex is held in its inactive state, which is vital for understanding how activation occurs. In this dissertation, we dissect key interfaces in Arp2/3 complex that are responsible for holding it in an inactive state, and specifically evaluate the roles of ATP and WASP, the canonical NPF, in the activating conformational change of Arp2/3 complex. In chapter II, we investigated the contacts made between the Arp2 and Arp3 subunits in their inactive state, and the role of ATP in stimulating the active conformation. We found that two key interfaces, the αE/αF loop in Arp2 and the C-terminus of Arp3, a conserved extension not present in actin, are vital for holding Arp2/3 complex in its autoinhibited state. Evaluation of the role of ATP demonstrated that binding of ATP is required for the activating conformational change and displaces the Arp3 C-terminus, an important step in destabilization of the inactive state. In chapter III, we investigated the mechanism of WASP-induced conformational changes using an engineered crosslinking assay that only forms crosslinks when Arp2/3 is in its active conformation. We discovered that many WASP-related proteins are capable of stimulating this conformational change through a mechanism that involves displacement of the Arp3 C-terminus. Interestingly, purified Arp2/3 complex crosslinked in the active conformation was hyperactive compared to WASP-mediated activation, demonstrating that WASP activation limits nucleation and that actin monomer delivery is not required for nucleation. This dissertation contains unpublished co-authored material.

The Actin Related Protein (ARP) 2/3 complex is a major regulator of the actin cytoskeleton that is implicated in cell morphogenesis in plants. However, a similar role is attributed to the endoplasmic reticulum (ER). My research explored the relationship between the two systems by using transgenic plants simultaneously expressing fluorescent proteins highlighting F-actin and ER organization in living cells. A comparison of F-actin organization in cells of wild type Arabidopsis thaliana and mutants with aberrant actin cytoskeleton suggests bundling in the distorted2 mutant but a relatively fine F-actin arrangement in klunker. These differences correlate with ER organization into cisternae, fenestrated sheets and tubules. A model relating ER-organization to the degree of actin bundling in a cell emerges and is supported by drug-induced interference in actin polymerization, altered ionic conditions and temperature. The study adds to the mechanistic understanding of cell morphogenesis in plants.

Résumé : Le cytosquelette d’actine génère et mécanotransduit des forces. Dans cette étude, nous montrons que l’actine branchée corticale, qui dépend de RAC1, WAVE et des complexes Arp2/3 contenant ARPC1B, est spécifiquement détectée par le senseur Coronin1B, qui signale, via WISp39 et l’inhibiteur de cycline/CDK p21, à la cellule, de progresser dans le cycle cellulaire. En conséquence, la formation d’un lamellipode et la migration persistante des cellules qui en découle, est corrélée à la durée de la phase G1. L’actine branchée corticale détermine l’entrée en phase S des cellules, en intégrant les stimuli solubles des facteurs de croissance et la mécanotransduction des adhérences à la matrice extracellulaire et aux cellules voisines. Le complexe Arp2/3 est globalement sur-exprimé dans le cancer du sein. Parmi ses sous-unités, la sur-expression de l’isoforme ARPC1B est le plus fort facteur prognostique pour les patientes. En outre, l’inhibition du complexe Arp2/3 bloque la prolifération de lignées de carcinomes mammaires et de mélanomes transformées par l’oncogène RAC1, contre laquelle il n’existe pas de thérapie ciblée. La découverte du checkpoint de l’actine branchée corticale apporte ainsi de nouvelles options pronostiques, diagnostiques et thérapeutiques dans les cancers. / The actin cytoskeleton generates and mechanotransducts forces. Here we report that the cortical branched actin that depends on RAC1, WAVE and ARPC1B-containing Arp2/3 complexes is specifically monitored by the Coronin1B sensor, WISp39 and the cyclin-CDK inhibitory protein p21, to control cell cycle progression. Accordingly, the duration of the G1 phase scales with the persistence of single cell migration, ensuing from branched actin and lamellipodium protrusion. Cortical branched actin determines the cell decision to enter into S phase by integrating soluble stimuli from growth factors and mechanotransduced signals, such as substratum rigidity and cell density. The Arp2/3 complex is overall overexpressed in brest cancer. Among its subunits, The ARPC1B isoform overexpression is the strongest prognostic factor for patients. Furthermore, Arp2/3 inhibition prevents the growth of mammary carcinoma and melanoma cell lines transformed by the RAC1 oncogene, for which no targeted therapy is available. The discovery of the cortical branched actin checkpoint thus provides diagnostic and therapeutic opportunities in cancer.

Le complexe Arp2/3, conservé sur le plan évolutif, joue un rôle central dans la nucléation d’actine branchée, qui entraîne la migration cellulaire, l’endocytose et d’autres processus cellulaire. Récemment, une petite protéine, Arpin, qui inhibe le complexe Arp2/3 au front du lamellipode a été découverte et caractérisée. Sur sa partie C-terminale, Arpin possède un motif acide (A), qui est homologue au motif A des différents NPF (Nucleation Promoting Factor). Il a été prédit qu’Arpin peut se lier à deux sites de liaison au complexe Arp2/3, similaire aux domaines VCA des NPF. Ici, nous utilisons la microscopie électronique de particules uniques pour obtenir une reconstruction 3D du complexe Arp2/3 lié à Arpin, à une résolution de 25 Å. Nous avons montré que la liaison d’Arpin induit la conformation ouverte, standard, du complexe Arp2/3. Nous avons confirmé qu’il y a deux sites de liaison sur le complexe Arp2/3 pour Arpin : un à l’arrière de la sous-unité Arp3, et le second localisé entre les sous-unités Arp2 et ARPC1. La distance entre le complexe Arp2/3 et Arpin (5nm) confirme qu’Arpin interagit avec son partenaire via sa queue acide C-terminale non structurée.Nous avons, ensuite, identifié Tankyrases1/2, comme un nouveau partenaire qui se lie à Arpin, par « pull-down ». De façon intéressante, les sites de liaisons d’Arpin aux Tankyrases et à Arp2/3 se chevauchent. Nous avons, par conséquent, démontré qu’il y a une compétition dose-dépendante entre le domaine ARC4 de Tankyrase1 et le complexe Arp2/3.Pour comprendre les principes de l’interaction entre Arpin et Tankyrases, nous avons créé un mutant d’Arpin (ArpinG218D) qui, in vitro, se lie toujours au complexe Arp2/3, mais plus aux Tankyrases. In vivo, ArpinG218D n’est pas capable d’inhiber le complexe Arp2/3, ce qui suggère que Tankyrase pourrait être nécessaire pour l’interaction entre Arpin et le complexe Arp2/3. Arpin est le facteur responsable du changement de direction des cellules migrantes. Nous avons donc analysé, la migration de cellules MCF10A exprimant soit la forme sauvage d’Arpin (ArpinWT) soit son mutant ArpinG218D en parallèle de la déplétion d’Arpin endogène. Les cellules exprimant ArpinG218D ont une persistance de migration supérieure, similaire à celles déplétées d’Arpin endogène. Nous avons, ainsi, fait l’hypothèse que le mutant ArpinG218D ne peut pas inactiver le complexe Arp2/3 car il n’est pas présent au niveau du lamellipode. Nous avons donc comparé la quantité de protéine d’ArpinWT et d’ArpinG218D dans la fraction membranaire de cellules migrantes. Une différence significative (44%) dans la quantité d’ArpinWT et d’ArpinG218D a confirmé notre hypothèse.Les Tankyrases sont des cibles thérapeutiques dans de nombreux cancers, mais il n’existe pas de modèle structural pour ces protéines grandes et flexibles. Dans ce travail, nous avons, pour la première fois, obtenu deux reconstructions 3D de Tankyrase1 et Tankyrase2 complètes liées à Arpin en utilisant la microscopie électronique de particules uniques. La résolution obtenue (27 Å) a été suffisante pour détecter un changement de conformation dramatique des domaines SAM et PARP de Tankyrase après fixation d’Arpin. Dans notre reconstruction, trois molécules d’Arpin se lient aux domaines ARC1, ARC4 et ARC5 de Tankyrase1. ARC5 a été montré pour être la partie le plus flexible de l’ensemble des domaines ARC.Grâce aux données que nous avons obtenues, nous avons suggéré un modèle de régulation de l’activité d’Arpin par les Tankyrases. Selon notre modèle, les Tankyrases se lient à Arpin dans le cytoplasme, changent sa conformation et amènent Arpin au niveau de la membrane dans le lamellipode. Traduisant les signaux extracellulaires, la GTPase Rac active Arpin, qui séquentiellement inactive le complexe Arp2/3, tandis que les Tankyrases sont libérées. / The evolutionarily conserved Arp2/3 complex plays a central role in nucleating the branched actin filament arrays that drive cell migration, endocytosis, and other processes. Recently, an inactivator of the Arp2/3 complex at the lamellipodium tip, a small protein, Arpin, was discovered and characterized. On its C-terminus, Arpin possesses an acidic (A) motif, which is homologous to the A-motif of various Nucleation Promoting Factors (NPFs). It was predicted that Arpin can bind at two binding sites to the Arp2/3 complex, similar to VCA domains of NPFs. Here, we used single particle electron microscopy to obtain a 3D reconstruction of the Arp2/3 complex bound to Arpin at a 25Å resolution. We showed that the binding of Arpin causes the standard open conformational of the Arp2/3 complex. We confirmed that there are two binding sites on the Arp2/3 complex for Arpin: one on the back of the Arp3 subunit, and the second is located between Arp2 and ARPC1 subunits. The distance between the Arp2/3 complex and Arpin (5 nm) supports the view that Arpin interacts with its partner via its unstructured C-terminal acidic tail.Next, using the pull-down assay, we identified the new Arpin binding partners, Tankyrases1/2. Interestingly, Tankyrases and the Arp2/3 complex possess overlapping amino acid sequences at Arpin binding sites. Hence, we demonstrated a competition between the ARC4 domain of Tankyrase1 and the Arp2/3 complex in a dose-dependent manner.To understand the principles of Tankyrases-Arpin interaction, we created a mutant Arpin (ArpinG218D) that lacks its ability to interact with Tankyrases, but not with the Arp2/3 complex in vitro. Interestingly, ArpinG218D was not able to inhibit the Arp2/3 complex in vivo, suggesting that Tankyrase may be necessary for Arpin-Arp2/3 complex interaction. Arpin is the turning factor of migrating cells, so we performed a migration analysis of MCF10-A cells expressing either wild type Arpin (ArpinWT) or mutant ArpinG218D in parallel with the depletion of endogenous Arpin. Cells expressing ArpinG218D had higher directional persistence, similar to the cells where the endogenous Arpin was knocked down. Thus, we suggested that mutant ArpinG218D cannot inactivate the Arp2/3 complex since it is not present at the lamellipodial tip. We compared the amount of protein for both ArpinWT and ArpinG218D in the membrane fraction of the migrating cells. A significant difference (44%) in the amount of ArpinWT and Arpin G218D was consistent with our hypothesis.Tankyrases are therapeutic targets in a variety of cancers, but currently there is no structural model available for these large and flexible proteins. In this work, we obtained for the first time two 3D reconstructions of full-length Tankyrase1 and Tankyrase1 bound to Arpin using single particle electron microscopy. The achieved resolution (27Å) was enough to detect a dramatic conformational change in Tankyrase SAM and PARP domains upon binding of Arpin molecules. In our reconstruction, three Arpins were bound to the ARC1, ARC4 and ARC5 domains of Tankyrase1. ARC5 was shown to be the most flexible part of the ARC cluster.Based on the obtained data, we suggested a model of regulation of the activity of Arpin by Tankyrases. According to our model, Tankyrases bind Arpin in the cytoplasm, change their conformational state and bring Arpin closer to the membrane in the lamellipodia. Deciphering the extracellular signals, Rac GTPase activates Arpin, which sequentially inactivates the Arp2/3 complex, while Tankyrases are released.

Assembly of branched actin filaments produces dynamic structures required during membrane associated processes including cell motility and endocytosis. The Actin Related Protein 2/3 (Arp2/3) complex is the only known regulator capable of nucleating actin branches. To specify the sub cellular localization and timing of actin assembly the complex is tightly regulated. Canonical activation of the Arp2/3 complex by Wiskott-Aldrich Syndrome proteins (WASP), requires preformed actin filaments, ensuring the complex nucleates new actin filaments off the sides of preformed filaments. WASP proteins can therefore propagate branch formation but cannot initiate a Y-branch without performed filaments. A key question, then, is what is the source of preformed filaments that seed branched actin network formation in cells? It is unclear how activation of Arp2/3 by multiple regulators is balanced to specify actin filament architectures that are productive in vivo. In this dissertation, we identified WISH/DIP1/SPIN90 (WDS) family proteins as activators of the Arp2/3 complex that do not require preformed filaments, and evaluated whether WDS proteins seed branching nucleation. In chapter II, we dissected the biochemical properties of WDS proteins and found they activate the Arp2/3 complex using a non-WASP like mechanism. Importantly, we discovered WDS-mediated Arp2/3 activation produces linear, unbranched filaments, and this activity is conversed from yeast to mammals. These observations highlight that WDS proteins have the biochemical capacity to seed actin branches. In chapter III, we observed WDS-generated linear filaments can seed WASP-mediated branching directly using single molecule microscopy with fluorescently labeled Dip1. We find that WDS-mediated nucleation co-opts features of branching nucleation. In chapter IV, we investigated how WDS activity is balanced with WASP. We discovered WDS proteins use a single turnover mechanism to activate Arp2/3 and this is conserved during endocytosis. In contrast, WASP-mediated activation is multi-turnover, highlighting a crucial difference between WDS proteins and WASP. Our observations explain how Arp2/3 may limit linear filament production to initiate networks and favor branches during network propagation. Finally, we use fission yeast to show that increasing Dip1 is sufficient to cause defects in actin assembly and the timing of actin patches at sites of endocytosis.

Actin is a highly ubiquitous and evolutionarily conserved protein capable of polymerizing and forming filamentous polymers which play a central role in cell mechanics and motility. Here, we study the in vitro regulation of actin mechanics and dynamics by calponin and caldesmon, two actin binding proteins believed to be involved in regulating cytoskeletal mechanics and structure through mechanisms not currently well understood. Chapters 1 and 2 introduce the reader to actin and its roles in the cell, as well as to the methods and theoretical foundations used in this work. In Chapter 3, we use total internal reflection and confocal fluorescence microscopy to investigate the polymerization dynamics of actin in the presence of a caldesmon C terminal fragment, H32K. We show that H32K stabilizes a nascent structural state of actin without altering the polymerization dynamics of the filament. We also show that H32K stabilized nascent actin has increased affinity for the actin branching protein complex Arp2/3 involved in driving membrane protrusions during cell motility, and propose the nascent state of actin as a possible transient differentiator targeting certain actin binding proteins to actin in vivo. This is to our knowledge the first reported direct functional effect of nascent actin. In Chapter 4, we use fluorescence microscopy to quantify actin bending mechanics in the presence of the binding protein calponin and show that calponin reduces the persistence length of actin. We compare our results to the literature and compare the mechanical change to electron microscopy reconstructions, which suggest that calponin affects actin intermonomer contacts through interactions with actin subdomain 2. In Chapter 5, we expand on the results from Chapter 4 using bulk rheology and show that calponin increases the tensile strength of reconstituted actin networks, similar to the effect seen in whole cells and tissues. We discuss these data within an affine network model and show that the results can be entirely explained in terms of the reduced actin persistence length. We use this to propose a novel physical mechanism for calponin function in vivo. This work elucidates the physical mechanisms of calponin and caldesmon function and their role in regulating the cellular cytoskeleton. / 2031-01-01T00:00:00Z

Le complexe Scar/WAVE génère la formation des lamellipodes par l'intermédiaire du complexe Arp2/3 responsable de la polymérisation de réseaux d'actine branchés. Dans le but d'identifier de nouveaux régulateurs du complexe Scar/WAVE, nous avons conduit un crible en cellules de Drosophiles combinant une approche protéomique à une approche de génomique fonctionnelle. La chaîne lourde de la clathrine a été identifiée au cours de ce crible comme une protéine interagissant avec le complexe Scar/WAVE et dont la déplétion affecte la formation des lamellipodes. Ce rôle de la clathrine dans la formation des lamellipodes peut être découplé de son rôle classique dans le transport vésiculaire en utilisant différentes approches. De plus, la clathrine est localisée au lamellipode en l'absence d'adapteurs et des protéines accessoires de l'endocytose. La surexpression de la clathrine affecte le recrutement membranaire du complexe WAVE réduisant ainsi la vélocité des protrusions membranaire et la migration cellulaire. Par opposition, lorsque la clathrine est envoyée artificiellement à la membrane plasmique par une fusion à une séquence myristoylée, on observe une augmentation du recrutement membranaire du complexe Scar/WAVE, de la vélocité des protrusions membranaires et de la migration cellulaire. L'ensemble de ces résultats montrent que la clathrine envoie le complexe Scar/WAVE à la membrane plasmique et donc contrôle la formation des lamellipodes en plus de son rôle plus classique dans le traffic membranaire. / The Scar/Wave complex (SWC) generates lamellipodia through Arp2/3-dependent polymerization of branched actin networks. In order to identify new SWC regulators, we conducted a screen in Drosophila cells combining proteomics with functional genomics. This screen identified Clathrin Heavy Chain (CHC) as a protein that binds to the SWC and whose depletion affects lamellipodium formation. This role of CHC in lamellipodium formation can be uncoupled from its role in membrane traffic by several experimental approaches. Furthermore, CHC is detected in lamellipodia in the absence of the adaptor and accessory proteins of endocytosis. We found that CHC overexpression decreased membrane recruitment of the SWC, resulting in reduced velocity of protrusions and reduced cell migration. In contrast, when CHC was targeted to the membrane by fusion to a myristoylation sequence, we observed an increase in membrane recruitment of the SWC, in protrusion velocity and in cell migration. Together these data suggest that CHC brings the SWC to the plasma membrane, thereby controlling lamellipodium formation, in addition to its classical role in membrane traffic.

Dans les organes lymphoïdes secondaires, les lymphocytes B acquièrent des antigènes immobilisés à la surface de cellules voisines. L’engagement du BCR (récepteur des cellules B) avec de tels antigènes induit la formation d’une synapse immunologique et la polarisation des lymphocytes B. Cette polarisation inclut le repositionnement du centrosome à la synapse immunologique ainsi que le recrutement et la sécrétion locale des lysosomes qui sont nécessaires à l’extraction, l’apprêtement et la présentation des antigènes sur les molécules du complexe majeur d’histocomptabilité de classe II (CMH-II) aux lymphocytes T CD4+ pré-activés. Des travaux précurseurs menés dans le laboratoire ont permis de mettre en évidence les premiers acteurs moléculaires impliqués dans ce processus. Cependant, le mécanisme précis gouvernant la polarisation du centrosome demeure encore aujourd’hui inconnu. Le travail réalisé pendant cette thèse avait pour objectif d’identifier de nouveaux régulateurs contrôlant la polarisation du centrosome dans les lymphocytes B après engagement du BCR avec des antigènes immobilisés. De plus, au regard du rôle grandissant joué par le microenvironnement tissulaire dans l’activation des lymphocytes B ainsi que dans la modulation de leurs fonctions, nous avons étudié l’effet de la protéine extracellulaire Galectine-8 sur la régulation de la capacité des lymphocytes B à se polariser et à extraire et présenter des antigènes immobilisés. Le travail présenté dans ce manuscrit montre que la présence du complexe Arp2/3 au centrosome des lymphocytes B non activés permet la nucléation locale de filaments d’actine qui permettent, grâce à leur interaction avec le complexe LINC, de lier le centrosome au noyau. L’activation des lymphocytes B induit la déplétion partielle du complexe Arp2/3 du centrosome qui est recruté à la synapse immunologique par la protéine HS1. Ceci induit une diminution de la nucléation d’actine au centrosome entraînant la séparation entre le centrosome et le noyau et permettant la polarisation du centrosome vers la synapse. De plus, nous montrons que la présence de la protéine Galectine-8 dans le milieu extracellulaire favorise le recrutement et la sécrétion des lysosomes à la synapse immunologique, conférant aux lymphocytes B une meilleure capacité à extraire et présenter des antigènes immobilisés. Nos résultats mettent en évidence des mécanismes inattendus régulant la polarisation des lymphocytes B en réponse à une stimulation antigénique et soulèvent des questions intéressantes concernant la régulation coordonnée de ces mécanismes qui confèrent aux lymphocytes B la capacité d’extraire, d’apprêter et de présenter des antigènes immobilisés efficacement. / In secondary lymphoid organs, B cells acquire antigens that are tethered at the surface of neighboring cells. Engagement of the B cell receptor (BCR) with such immobilized antigens leads to the formation of an immune synapse and the subsequent polarization of B cells. This includes the repositioning of the centrosome towards the immune synapse as well as the recruitment and local secretion of lysosomes required for efficient antigen extraction, processing and presentation onto class II major histocompatibility complex (MHC-II) molecules to primed CD4+ T cells. Pioneer work performed in the lab has highlighted the first molecular players involved in this process. However, the precise mechanism governing centrosome polarization remains to be fully elucidated. The work performed during this thesis aimed at identifying new regulators supporting centrosome polarization in B lymphocytes upon BCR engagement with immobilized antigens. In addition, in view of the emerging role played by the tissue microenvironment in shaping B cell activation and functions we investigated whether extracellular Galectin-8 modulates the ability of B cells to polarize, extract and present immobilized antigens. We show here that, in resting lymphocytes, centrosome-associated Arp2/3 (actin related protein-2/3) locally nucleates F-actin, which is needed for centrosome tethering to the nucleus via the LINC (linker of nucleoskeleton and cytoskeleton) complex. Upon lymphocyte activation, Arp2/3 is partially depleted from the centrosome as a result of its HS1-dependent recruitment to the immune synapse. This leads to a reduction in F-actin nucleation at the centrosome and thereby allows its detachment from the nucleus and polarization to the synapse. In addition, we show that extracellular Galectin-8 favors lysosome recruitment and secretion at the immune synapse, hence providing B cells with an enhanced capacity to extract and present immobilized antigens. Our findings highlight unexpected mechanisms that tune B cell polarity in response to antigenic stimulation and raise exciting questions concerning the coordinated regulation of these mechanisms to provide B cells with the capacity to efficiently extract, process and present surface-tethered antigens.

Prominin-1 is a lipid raft–associated, cholesterol-binding membrane glycoprotein selectively associated with plasma membrane protrusions and extracellular vesicles derived therefrom. Despite its worldwide use for stem cell isolation and its clinical importance in cancer-initiating cells and photoreceptor morphogenesis the function of prominin-1 remains elusive. This prompted me to investigate its role in the architecture and dynamics of microvilli and primary cilia at the apical plasma membrane of Madin-Darby canine kidney (MDCK) cells. Therefore, stably transfected cell lines were established expressing human prominin-1 splice variant 1 or 2. Upon the overexpression of prominin-1 the number of individual microvilli and clusters of them increased significantly. I also noticed alterations in their architecture, i.e. branching microvilli. Fascinatingly, two point mutations (Pro37→Ala and Tyr41→Ser) in the ganglioside GM1-binding motif of prominin-1 increased the number of branched microvilli and generated irregular ones with knob-like structures at their tip. Additionally, the release of prominin-1+ vesicles was impaired. Interestingly, both phenotypes were suppressed by the inhibition of the phosphoinositide 3-kinase (PI3K) or the Arp2/3 complex. Impaired interaction of prominin-1 with the PI3K through the introduction of an additional mutation (Tyr828→Phe) in its PI3K-binding site also reduced the amount of structurally altered microvilli. Thus, the interaction of prominin-1 with the PI3K may drive the conversion of the docking phospholipid phosphatidylinositol(4,5)-bisphosphate into phosphatidylinositol(3,4,5)-trisphosphate resulting in the uncoupling of the microvillar membrane from the underlying actin filaments thereby creating irregular/knob-like microvilli. Simultaneously, the phospholipid conversion might modulate the activity of regulators and/or activators of the Arp2/3 complex leading to the branching of microvilli. The overexpression of human prominin-1 also increased the length of primary cilia. Remarkably, a mutation in the histone deacetylase 6-binding site that mimics acetylation produces shorter cilia in cells expressing human prominin-1.s2. Additionally, it stimulates membrane vesicle release and dome formation. Above these striking observations, I observed branching cilia and cilia with a pearling shape. Collectively, the data suggest that a complex interplay of prominin-1 with its lipid and protein interaction partners regulates the architecture and dynamics of cellular protrusions. Finally, a growing number of studies use canine prominin-1 as an antigenic marker despite the absence of specific antibodies. Studies investigating its expression in dog tissues or cells derived therefrom rely on antibodies directed against its human and murine orthologs. To determine its cross-species immunoreactivity I cloned canine prominin-1 and overexpressed it as a green fluorescent protein fusion protein in MDCK cells. Here, I show that the genomic structure of the canine prom1 gene is similar to that of human and mouse. Canine prominin-1 shows the common characteristics of the prominin-1 family but the primary structure is poorly conserved. Like human and mouse protein, it is targeted to the apical membrane of MDCK cells and specifically enriched in microvilli and primary cilia. Immunocytochemistry, flow cytometry and immunoblotting techniques revealed that none of the applied antibodies against human or mouse prominin-1 recognizes the canine protein.

La migration cellulaire possède deux volets antagonistes ; nécessaire à plusieurs processus physiologiques tels que la réponse immunitaire, elle peut également induire la mort d’un organisme en permettant les cellules cancéreuses d’envahir des organes sains. In vivo, la migration s’effectue dans des milieux complexes et confinés qui imposent une forte déformabilité aux cellules migratoires. Récemment, divers études ont montré que le noyau impose la limite de la déformabilité cellulaire lors de la migration en 3D (Wolf et al. JCB, 2013; Harada et al. JCB, 2013). Il a, en effet, été montré que la migration cellulaire peut être augmentée en diminuant la rigidité nucléaire (Wolf et al. JCB, 2013). Cependant, il existe une limite de rigidité nucléaire en dessous de laquelle la migration cellulaire peut être inhibée via l’inhibition de la survie cellulaire (Harada et al. JCB, 2013). Les cellules cancéreuses qui migrent à des vitesses relativement faibles (µm/heure) et ont des noyaux rigides surmontent la limite imposée par la déformation nucléaire en dégradant et élargissant le milieu extracellulaire. Les cellules immunitaires telles que les neutrophiles qui migrent rapidement (10 µm/mn) et ont des noyaux mous sont connus pour mourir aux sites d’infections. Les cellules dendritiques, de la famille des cellules immunitaires, ont une fonction de présentation d’antigènes qui requiert à la fois une grande capacité migratoire et de survie. Elles représentent donc un modèle cellulaire intéressant pour l’étude de la déformation nucléaire chez les cellules qui migrent rapidement et survivent longtemps. Durant mon doctorat, j’ai étudié le mécanisme grâce auquel les cellules dendritiques déforment leurs noyaux afin de migrer de manière efficace en milieux confinés tout en préservant un haut taux de survie. J’ai utilisé un système expérimental nouveau et original consistant en des microcannaux avec des constrictions (Heuzé et al. MMB, 2011). Ces canaux, combinés à des manipulations génétiques et de la video microscopie nous ont permis de montré que les cellules dendritiques possèdent un mécanisme spécifique, indépendant de celui utilisé pour leur migration, leur permettant de déformer leurs noyaux tout en migrant dans des milieux hautement confinés. Ce mécanisme est basé sur la génération d’un réseau d’actin, autour du noyau, nucléé par Arp2/3 et indépendant du moteur Myosin II. Ce réseau d’actine co-localise avec des sites de rupture de la Lamin A/C. De plus, réduire la quantité de Lamin A/C dans les cellules dendritiques inhibe la formation de ce réseau d’actin perinucléaire. Basés sur ces résultats, nous avons proposé un nouveau mécanisme de déformation du noyau lors de la migration en milieux confinés basé sur Arp2/3 qui, en nucléant un réseau d’actine autour du noyau permet de casser la lamin A/C diminuant ainsi la tension de surface nucléaire et permettant le passage noyau. / Cell migration has two opposite faces; necessary for many physiological processes such as immune response, it can also lead to the organism death by allowing metastatic cells to invade new organs. In vivo migration often occurs in complex 3D environments which impose high cellular deformability. Recently, cellular deformability during 3D migration has been shown to be limited by the nucleus (Wolf et al. JCB, 2013). For instance, cell migration can be increased by decreasing nuclear stiffness. However, below a given nuclear stiffness 3D cell migration can be reduced as a result of impaired cell survival (Harada et al. JCB, 2014). Cancer cells which display slow migration and have rather stiff nuclei have been shown to overcome the physical limits of 3D migration through adhesion combined to matrix degradation or high actomyosin contraction (Wolf et al. JCB, 2013). Immune cells such as neutrophils which are fast moving cells with soft nuclei have been reported to die at sites of infection. Interestingly, dendritic cells function as antigen presenting cells requires high migratory ability as well as high survival. They thus constitute an interesting model for studying nuclear deformation in fast moving and long lived cells. During my PhD, I studied the mechanism by which dendritic cells deform their nuclei to achieve proper migration in highly confining space while preserving a high survival rate. I used an original micro fabricated experimental set up (Heuzé et al. MMB, 2011) consisting of microchannels with constrictions to mimic cellular transmigration. Those channels combined with genetic manipulation and live cell imaging followed by image processing were used to assess the mechanism dendritic cells use to deform their nucleus, which we found to be specific and not required for cell motility per se. I showed that dendritic cells overcome the physical limitation imposed by nuclear deformation through small gaps by nucleating an Arp2/3 based actin network around the nucleus. Surprisingly, the formation of this actin network is independent of myosin II based contraction. This actin accumulation around the nucleus co-localized with sites of nuclear Lamin A/C breakage. Moreover, Lamin A/C depletion in dendritic cells leads to the disappearance of this actin ring and the release of the need for Arp2/3 for nuclear deformation. We thus propose a new mechanism of nuclear squeezing through narrow gaps based on an Arp2/3 nucleated actin meshwork which, by transiently breaking the Lamin A/C network, releases the nuclear surface tension and allows nuclear thus cell passage through micrometric constrictions. Lamin A/C repolymerization around the nucleus at the exit of constrictions would then restore nuclear stiffness, allowing cell survival. Interestingly, this actin accumulation around the nucleus was also observed in vivo in migrating macrophages but not in HL-60 derived neutrophils. Taken together, our data suggest that the Arp2/3 based nuclear squeezing mechanism would be a general feature of highly migratory cells which need to survive long enough to accomplish their functions.