Le VIH-1, ou Virus de l’ImmunoDéficience Humain de type I, est un pathogène intracellulaire dont le cycle de réplication dépend entièrement des machineries cellulaires. Des interactions entre les protéines virales et cellulaires sont donc essentielles pour l’achèvement de chaque étape du cycle de multiplication du VIH-1. Comprendre comment le virus VIH-1 détourne la machinerie cellulaire à son profit est un élément clé pour le combattre. L’objectif de mon travail de thèse était d’explorer le rôle de la protéine Bécline-1 dans la formation et la libération des particules virales. Cette protéine, composant du complexe phosphatydylinositol-3-Phosphate-kinase III (PI3K-III), est impliquée dans l’autophagie, le trafic intracellulaire et la cytocinèse.Au cours de ma thèse, nous avons développé les outils d’ARN interférence afin de décrypter le rôle de Bécline-1 dans le cycle réplicatif du VIH-1. De manière très intéressante, nous avons mis au jour un nouveau rôle de Bécline-1 lors des étapes précoces de la mitose, jamais décrit, ni publié à ce jour. Un mauvais déroulement de la mitose induit de l’instabilité génomique pouvant conduire à la mort cellulaire ou à une transformation maligne. Au cours de cette thèse, nous avons établi que l’extinction de Bécline-1 induit un défaut d’attachement des chromosomes, via leurs kinétochores, aux microtubules, conduisant à un blocage des cellules en prométaphase. Cette extinction provoque un défaut d’organisation du kinétochore en diminuant le recrutement des protéines CENP-E, CENP-F et ZW10 au niveau de cette structure protéique. Nous avons également montré une interaction directe entre Bécline-1 et Zwint-1, une protéine du kinétochore, jouant un rôle essentiel pour l’accrochage des microtubules aux kinétochores. Enfin, nous avons montré que ce nouveau rôle de Bécline-1 dans l’alignement des chromosomes en mitose est indépendant de son association avec ses partenaires du complexe PI3K-III et de son rôle dans l’autophagie.Le second volet de ma thèse a porté sur l’étude du rôle de Bécline-1 au cours du cycle réplicatif du virus. Nos résultats montrent que cette protéine est nécessaire à l’établissement des phases tardives du cycle viral. Nos expériences de production virale montrent que l’extinction de Bécline-1 provoque une forte accumulation des sous-produits de la protéine Gag du virus dans les cellules, et réduit la libération des particules virales dans le milieu extracellulaire. Par immunofluorescence, nous observons que des composants viraux s’accumulent massivement à la membrane sous l’effet de l’extinction de Bécline-1 dans les cellules. Par ailleurs, ces résultats sont dépendants de l’expression du facteur de restriction BST2 dans la cellule. Ces derniers résultats ouvrent d’importantes perspectives quant au rôle de Bécline-1 et du PI3K-III sur le cycle de réplication du VIH-1.L’ensemble de ces travaux démontre l'importance de la protéine Beclin-1, pour le maintien de l'intégrité du génome pendant la mitose et pour le cycle de réplication du VIH 1. / HIV-1, or Human Immunodeficiency Virus type 1, is an intracellular pathogen whose replication cycle entirely depends on cellular machineries. Interactions between the viral and cellular proteins are therefore essential to the completion of each step of HIV-1's multiplication cycle. Understanding how the HIV-1 virus uses the cellular machinery to its own benefit is a key element to combat it. The objective of my thesis work was to explore the role of the Beclin-1 protein in the formation and the release of viral particles. This protein, a component of the phosphatydylinositol-3-Phosphate-kinase III (PI3K-III) complex, is involved in autophagy, intracellular trafficking and cytokinesis. Throughout my thesis researches, we have developed the RNA interference tools in order to decipher the role of Beclin-1 in HIV-1's replicative cycle. Very interestingly, we have found out a new role of Beclin-1 in the early stages of mitosis which had never been described nor published before. A bad execution of mitosis induces genomic instability which can lead to cell death or malignant transformation. During this thesis work, we have demonstrated that the extinction of Beclin-1 causes a defect in the attachment of chromosomes to microtubules through their kinetochores, leading to the locking of cells in prometaphase. This extinction provokes a defect in the organization of the kinetochore by diminishing the recruitment of the CENP-E, CENP-F and ZW10 proteins at the level of this protein structure. We have also found out that there is a direct interaction between Beclin-1 and Zwint-1, a protein of the kinetochore that plays an essential role in the attachment of the kinetochores to the microtubules. Finally, we have demonstrated that this new role of Beclin-1 in the alignment of the chromosomes during mitosis is independent of both its association with its partners from the PI3K-III complex and its role in autophagy. The second strand of my thesis dealt with the role of Beclin-1 in the virus replicative cycle. Our results show that this protein is essential to the setting up of the late phases of the viral cycle. Our experiments of viral production have shown that the extinction of Beclin-1 causes a high accumulation of the by-products of the virus' Gag protein in the cells, and reduces the release of viral particles in the extracellular medium. By immunofluorescence, we detected a massive accumulation of viral components on the membrane as a result of the extinction of Beclin-1 in the cells. Besides, these results depend on the expression of the BST2 restriction factor in the cell. These latter results open up significant prospects regarding the role of Beclin-1 and that of the PI3K-III complex in HIV-1's replication cycle. All these researches demonstrate the importance of the Beclin-1 protein in two mechanisms that allow the maintenance of genomic integrity during mitosis and the HIV-1 replication cycle.

Le virus d’Epstein-Barr (EBV) est un herpesvirus dont la séroprévalence est d’environ 90 % de la population adulte mondiale. EBV est associé à de nombreuses pathologies tumorales. La primo infection conduit à l’établissement du virus sous forme latente dans les lymphocytes B mémoires. Au sein de ces cellules B, le génome viral est sous la forme d’un épisome, un ADN circulaire double brin, et une fraction restreinte de gènes viraux est exprimée. Afin de se maintenir aux cours des divisions cellulaires, le génome viral est répliqué en phase S par la machinerie cellulaire et ségrégé lors de la mitose dans chaque cellules filles. La réplication et la ségrégation du génome viral nécessitent 2 facteurs viraux que sont la protéine virale EBNA1 (Epstein-Barr Nuclear Antigen 1) et la région oriP sur le génome viral. En phase S, EBNA1 interagit directement avec l’oriP et y recrute le complexe de pré-réplication de l’ADN. En mitose, EBNA1 ancre l’épisome à la chromatine ce qui permet une ségrégation efficace. Les mécanismes d’interaction entre EBNA1 et la chromatine reste encore flou. Au cours de notre travail, nous avons identifié la protéine RCC1 comme un partenaire potentiel pour la protéine EBNA1 pouvant être impliqué dans l’ancrage d’EBNA1 à la chromatine. Nous avons validé cette interaction et caractérisé les régions d’interactions pour ces deux protéines. Par ailleurs nous avons démontré que RCC1 est recrutée sur l’oriP en présence d’EBNA1 et que ces deux protéines interagissent en mitose. À la lumière de nos résultats et des données de la littérature, nous proposons que l’interaction d’EBNA1 avec la chromatine est dynamique et implique à la fois des interactions directes (AT-Hook, interaction avec les nucléosomes) mais aussi des facteurs cellulaires (RCC1, EBP2 et HMGB2). / Epstein-Barr virus (EBV) is a ubiquitous herpesvirus associated with several human cancers. In proliferating latently-infected cells, the EBV genome persists as a circular plasmid that is replicated once per cell cycle and partitioned at mitosis. Both of these processes require a single viral protein, Epstein Barr nuclear antigen 1 (EBNA1), which binds to two clusters of cognate binding sites within the origin of plasmid replication (oriP). EBNA1 plays an essential role both in viral episome replication, by recruiting the cellular complex of DNA replication onto the oriP, and in the efficient segregation of the viral episomes, by tethering the viral DNA onto the mitotic chromosomes. Whereas the mechanisms of viral DNA replication have been well documented, the mechanisms involved in tethering EBNA1 to the cellular chromatin are far from being understood. Here we have identified Regulator of Chromosome Condensation 1 (RCC1) as a novel EBNA1 cellular partner. RCC1 is the only known nuclear guanine nucleotide exchange factor (RanGEF) for the small GTPase Ran enzyme. RCC1, associated with chromatin, is involved in the formation of RanGTP gradients critical for nucleo-cytoplasmic transport, mitotic spindle formation, and nuclear envelope reassembly after mitosis. We have used several approaches to demonstrate a direct interaction between these two proteins and to identify the regions. involved Moreover, by using Chromatin ImmunoPrecipitation assay (ChIP) we have shown that RCC1 is enriched in the oriP region of mini viral replicons in a manner dependent on EBNA1. Finally, by using a combination of confocal microscopy and FRET analysis to follow the dynamics of interaction between the two proteins throughout the cell cycle, we have demonstrated that EBNA1 and RCC1 closely associate on the chromosomes during metaphase. Taken together, our data strongly suggest an essential role for RCC1 in tethering EBNA1 - linked to the viral episome - to the metaphasic chromosomes. Our results and those of others lead us to the idea that the interaction between EBNA1 with the cellular chromosomes requires several factors such as direct interactions or cellular proteins and these interactions are complementary and / or redundant.

La polykystose rénale autosomique dominante (ADPKD) est la maladie monogéniquehumaine la plus fréquente (prévalence 1/800). Les gènes responsables de cette maladie sont PKD1(codant pour PC1) ou PKD2 (codant pour PC2). La maladie évolue vers l’insuffisance rénale terminale.Aujourd’hui, seul un traitement symptomatique est proposé aux malades. Les mécanismes à l’originede l’ADPKD sont mal connus. Les modèles animaux permettent de mieux comprendre laphysiopathologie d’une maladie. Il n’existe pas de bon modèle de polykystose (même causemoléculaire, même mode de transmission, même signes cliniques). En utilisant la transgénose degrands fragments, nous avons créé un modèle de surexpression de PKD2 humain. Le transgène estsous le contrôle de son promoteur naturel humain. Cette souris exprime deux fois plus de PC2 queles sauvages. Elle ne présente que quelques microkystes mais une tubulopathie associant défaut deconcentration des urines et protéinurie tubulaire. La surexpression de PC2 inhibe l’expression degènes codant pour des protéines de la matrice extracellulaire. Le phénotype cellulaire de cesanimaux est remarquable : un tiers des cellules présentent un nombre élevé de centrosomes. Cephénotype cellulaire a été retrouvé chez des souris sous exprimant Pkd2 et chez des souris sousexprimant Pkd1. Ce caractère multicentrosomique est corrigé en incubant les cellules avec uninhibiteur de CaMKII ou en croisant nos souris transgéniques avec une souris KO de Camk2. Nousavons réussi à lier CaMKII, la duplication des centrosomes et les polycystines, in vitro et in vivo. Ceciamène un éclairage nouveau sur la duplication du centrosome et la physiopathologie de l’ADPKD. / Autosomal Dominant Polycystic Kidney Disease (ADPKD) is the most common monogenic human disease (prevalence 1/800). Genes responsible for this disease are PKD1 (encoding PC1) or PKD2 (encoding PC2). The disease progresses to end stage renal disease. Today, only symptomatic treatment is offered to patients. The mechanisms underlying the ADPKD are unknown. Animals models allow better understand the disease’s pathophysiology. There is no good model of ADPKD (same molecular cause, same clinical signs). We created a mice model of human PKD2 overexpression. The transgène is under the control of its human natural promoter. This mouse expresses PC2 twice as much as the wild. It shows only few microcysts but tubulopathy involving lack of urine concentration and tubular proteinuria. PC2 overexpression inhibits the expression of genes encoding proteins of the extracellular matrix. The cellular phenotype of these animals is special : one third of the cells have a high number of centrosomes. This cellular phenotype was found in Pkd2 Knockout mice and in Pkd1 knockout mice. This multicentrosomic character is corrected by incubating the cells with a CaMKII inhibitor or by crossing our transgenic mice with Camk2 knockout mice. We propose a link between CaMKII, Centrosome duplication and polycystin in vitro and in vivo. This brings a new light on centrosome duplication and pathophysiology of ADPKD.

A la fin de la vie d’une cellule, différentes forces mécaniques permettent la séparation des chromosomes. Nos données préliminaires suggèrent l’existence d’un autre mécanisme provenant du centre du fuseau mitotique, non décrit dans l’embryon une cellule de C. elegans qui permettrait la séparation des chromosomes. Dans cette cellule, les microtubules kinétochoriens n’appliquent aucune force mécaniques sur les chromosomes durant l’anaphase. Il a été décrit que les chromosomes sont séparés grâce au déplacement des centrosomes via les forces de traction corticales. A l’aide de la microchirurgie laser dans les embryons une cellule de C. elegans, j’ai montré qu’en détruisant physiquement un ou deux centrosomes, les chromosomes continuent de se séparer, révélant l’existence d’une force de propulsion interne au fuseau mitotique (Nahaboo et al., 2015). En combinant la destruction de centrosomes et l’inactivation génétique, nous avons caractérisé les rôles de gènes favorisant ou freinant cette force de propulsion. J’ai observé que la kinésine-5, BMK-1, et le crosslinker MAP-65/SPD-1 freinent cette force de propulsion. Alors que dans d’autres espèces ces protéines favorisent la séparation des chromosomes. Nous avons remarqué que les protéines RanGTP et CLASP, favorisant de la nucléation et la polymérisation des microtubules, aident cette force de propulsion. Ces propriétés suggèrent que la polymérisation des microtubules au centre du fuseau est requise pour permettre la séparation des chromosomes durant la mitose.Par manque d’outils adéquats afin d’altérer la dynamique des microtubules, nous avons collaboré avec l’équipe de biochimistes du Dr. D. Trauner à Munich en Allemagne. Ils ont synthétisé la molécule photoactivable, Photostatin (PST), permettant la dépolymérisation des microtubules en quelques secondes (Borowiak et al., 2015). Entre 390 - 430 nm, PST est activé, dépolymérisant les microtubules, alors qu’entre 500 – 530 nm, PST est inactivé, permettant la polymérisation normale des microtubules. J’ai mesuré que la croissance des microtubules avec PST actif est absente dans des cellules Hela. J’ai montré que le cycle cellulaire dans l’embryon de C. elegans est arrêté localement en présence de PST actif. Nous avons alors montré que PST contrôle optiquement la dynamique des microtubules, in vitro, in cellulo et in vivo, de manière non invasive, rapide, locale et réversible. En résume, j’ai identifié une nouvelle force permettant la séparation des chromosomes à l’aide des approches moléculaires et biophysiques, et j’ai aidé à la caractérisation PST, un antimicrotubule photoactivable de manière locale et réversible. / In mitosis, different mechanical forces are involved in chromosome segregation. In C. elegans one-cell embryos, preliminary data suggest that an unknown mechanism, coming from inside the mitotic spindle, could influence chromosome separation. In those cells, it has been showed that kinetochore microtubule activity is absent. Thanks to external pulling forces, centrosome separation drives chromosome segregation. By using microsurgery inside the one-cell C. elegans embryos, we have shown that destroying one or two centrosomes did not prevent chromosome separation, revealing the existence of an outward pushing force (Nahaboo et al., 2015). By combining gene inactivation and centrosome destruction, we showed that the kinesin-5 and the crosslinker SPD-1 act as a brake on this pushing force, whereas they enhance chromosome segregation in other species. Moreover, we identified a novel role for the two microtubule-growth and nucleation agents, RanGTP and CLASP, in the establishment of the centrosome-independent force during anaphase. Their involvement raises the interesting possibility that microtubule polymerization of midzone microtubules is required to sustain chromosome segregation during mitosis. Then, we aim to reversibility affect microtubule dynamics in the central spindle. Because of the lack of adequate tools, we have collaborated with biochemists from Dr. D. Trauner lab, in Munich, Germany, who are specialized in photoactivable drugs. They have synthetized a photoswitable drug, Photostatin (PST), which can depolymerize microtubules in few seconds in an on/off manner (Borowiak et al., 2015). Under blue light (390 - 430 nm), PST is activated leading to microtubule depolymerization, whereas under green light (500 - 530 nm), PST is activated which does not affect microtubule dynamics. I measured that microtubule growing is absent in presence of activated PST in Hela cells. I also showed that cell cycle can be stopped thank to activated PST in multiple cell C. elegans embryos. We have shown that PST can control microtubule dynamics thanks to visible light in vitro, in cellulo and in vivo, as an on/off switch, in a non-invasive, local and reversible manner.

Symmetric cell division in gram-negative bacteria is essential for generating two equal-sized daughter cells, each containing cellular material crucial for growth and future replication. The Min system, comprised of proteins MinC, MinD and MinE, is particularly important for this process since its deletion leads to minicells incapable of further replication. This thesis focuses on the interactions involving MinE that are important for allowing cell division at the mid-cell and for directing the dynamic localization of MinD that is observed in vivo. Previous experiments have shown that the MinE protein contains an N-terminal region that is required to stimulate MinD-catalyzed ATP hydrolysis in the Min protein interaction cycle. However, MinD-binding residues in MinE identified by in vitro MinD ATPase assays were subsequently found to be buried in the hydrophobic dimeric interface in the MinE structure, raising the possibility that these residues are not directly involved in the interaction. To address this issue, the ability of N-terminal MinE peptides to stimulate MinD activity was studied to determine the role of these residues in MinD activation. Our results implied that MinE likely undergoes a change in conformation or oligomerization state before binding MinD. In addition we performed circular dichroism spectroscopy of MinE. The data suggest that direct interactions between MinE and the lipid membrane can lead to conformational changes in MinE. Using NMR spectroscopy in an attempt to observe this structure change, different membrane-mimetic environments were tested. However the results strongly suggest that structural studies on the membrane-bound state of MinE will pose significant challenges. Taken together, the results in this thesis open the door for further exploration of the interactions involving MinE in order to gain a better understanding of the dynamic localization patterns formed by these proteins in vivo.

TIF1γ est une protéine nucléaire de 1127 acides aminés possédant deux activités : une activité d'E3-ubiquitine ligase et des fonctions de régulateur transcriptionnel. TIF1γ exerce majoritairement ses fonctions dans les processus de développement embryonnaire et de différenciation cellulaire, notamment via son implication dans la voie de signalisation du TGFβ. Le rôle anti-tumoral de TIF1γ a été mis en évidence dans plusieurs modèles murins et son expression est diminuée dans de nombreuses tumeurs humaines de diverses origines tissulaires. Néanmoins, les mécanismes moléculaires et cellulaires par lesquels TIF1γ exerce ses fonctions oncosuppressives sont méconnus. Dans ces travaux, nous avons pu mettre en évidence le rôle inhibiteur de TIF1γ sur la transition épithélio-mésenchymateuse (EMT, Epithelial-to- Mesenchymal Transition) médiée par le TGFβ in vivo, permettant ainsi de limiter les propriétés agressives des cellules tumorales. De plus, nous avons décrit l'implication de TIF1γ dans la progression de la mitose et le point de contrôle du fuseau mitotique : les cellules n'exprimant plus TIF1γ présentent de nombreuses anomalies nucléaires ainsi qu'une forte aneuploïdie associée à une résistance aux agents ciblant les microtubules, molécules classiquement utilisées en chimiothérapie. De plus, nous avons pu corréler la faible expression de TIF1γ à une augmentation de l'instabilité chromosomique dans différentes tumeurs humaines. Ainsi, nos travaux ont permis de mettre en évidence le phénotype cellulaire induit par la perte de TIF1γ dans les cellules tumorales : instabilité chromosomique, résistance aux traitements chimiothérapeutiques et acquisition de propriétés invasives / TIF1γ / TRIM33 (Transcriptional Intermediary Factor 1γ / TRIpartite Motif-containing 33) is a 1,127 amino acids nuclear protein with two biochemical activities: an E3-ubiquitin ligase activity and transcriptional regulatory functions. TIF1γ is ubiquitously expressed in many organisms and exerts its functions mainly in the processes of embryonic development and cell differentiation, particularly through its involvement in the TGFβ signaling pathway. The oncosuppressive functions of TIF1γ have been demonstrated in several mouse models and its expression is reduced in many human tumors of various tissue origins. Nevertheless, the molecular and cellular mechanisms driving TIF1γ anti-tumoral activities are unknown. In this work, we highlight its inhibitory role on TGFβ-mediated EMT (Epithelial-to-Mesenchymal Transition) in vivo, thus limiting the aggressive properties of tumor cells. In addition, we describe TIF1γ involvement in mitotic progression and the Spindle Assembly Checkpoint (SAC): TIF1γ deleted cells display many nuclear abnormalities, aneuploidy and resistance to spindle microtubule-disrupting agents, which are drugs classically used in chemotherapeutic treatments. Finally, we correlated the low expression level of TIF1γ to an increased rate of chromosomal instability in different human tumors. Thus, our work has highlighted the tumor suppressor role of TIF1γ: its deletion in tumor cells induce chromosomal instability, resistance to chemotherapeutic treatments and acquisition of invasive properties

(ENGLISH) Ubiquitin signaling is a key regulatory mechanism for many important cellular processes such as transcription, differentiation and cell division. Cell division requires duplication of all genetic material during S-phase followed by its precise partitioning between two daughter cells during mitosis. Misregulation of the complex mitotic machinery may lead to aneuploidy and genomic instability, known drivers of tumorigenesis. Indeed, systematic genetic analysis of many cancer tissues over the last decades, indicates the presence of severe chromosome abnormalities in thousands of cancer tissue samples. In this work, I investigated the function of two components of ubiquitin signaling, the deubiquitinating enzyme UCHL3 and the E3 ubiquitin ligase TRIM15. The hypothesized role of E3 ligase TRIM15 in the cell cycle regulation could not be confirmed by our experiments, but I observed an effect on cell adhesion and motility instead. UCHL3 was identified using high-content visual siRNA screen, as a critical factor controlling genome segregation and integrity. Interestingly, it has been previously reported that UCHL3 levels are altered in various cancer types, especially colon cancer. My data demonstrate that UCHL3 drives proper alignment of chromosomes at the metaphase plate by facilitating...

L’établissement et le maintien du patron d’expression génique sont d’une importance critique pour l’identité cellulaire. Les protéines du groupe Polycomb (PcG) agissent sur la chromatine afin de maintenir la répression génique de ses gènes cibles à travers les cycles cellulaires de façon épigénétique. Toutefois, durant la mitose, la structure de la chromatine est grandement altérée par la répression de la transcription, la condensation de la chromatine et le relâchement de nombreux facteurs de transcription. Une question se pose alors : comment les protéines PcG peuvent-elles maintenir leur fonction à travers la mitose ? En interphase, les protéines PcG sont liées à leurs cibles sur la chromatine. Durant la mitose, la majorité des protéines PcG se libèrent de la chromatine mais une petite fraction persiste. Selon l’hypothèse du mitotic bookmarking, cette fraction agirait comme un ensemble de marqueurs guidant le recrutement des protéines PcG en fin de mitose pour maintenir le profil d’expression génique de la cellule. Cependant, nous ne savons pas comment ce recrutement à lieu, ni comment une fraction de protéines PcG est retenue à la chromatine. Afin de répondre à ces questions, un crible à ARN interférent a été établi pour identifier des facteurs contrôlant la liaison des protéines PcG à la chromatine à travers le cycle cellulaire. Quoiqu’une confirmation soit nécessaire, les facteurs spécifiques à l’interphase sont enrichis en protéines co-purifiant avec la protéine PcG testée et en hélicases alors que ceux spécifiques à la mitose sont enrichis en candidats liés aux protéines du groupe Trithorax (TrxG). / A critical part of cell identity is the establishment and maintenance of gene expression patterns. Polycomb group proteins (PcG) act on chromatin to maintain gene repression through cell cycles (epigenetically). However, during mitosis, chromatin structure is greatly altered by transcription repression, chromatin condensation, and the release of many transcription factors. A question then arises: how can PcG proteins maintain their function through mitosis? During interphase, PcG proteins are bound to their chromatin targets. During mitosis, most PcG proteins are released from chromatin, but a small fraction remains bound to chromatin. According to the mitotic bookmarking hypothesis, this fraction acts as a set of markers to guide the recruitment of PcG proteins at the end of mitosis to maintain the gene expression profile. However, we do not know how this recruitment takes place, nor do we know how a fraction of PcG proteins is retained on chromatin. To address these questions, an RNAi screen was established to identify factors that control the binding of PcG proteins to chromatin across the cell cycle. Although a confirmation is necessary, factors identified from interphase cells were enriched in proteins co-purifying with the tested PcG protein and in helicases while mitosis specific factors were enriched in Trithorax group (TrxG) protein related candidates.

Myosin II and actin are two major components of the ingressing cortex during cytokinesis. However, their structural dynamics and functions during cytokinesis are still poorly understood. To study the role of myosin II in cortical actin turnover, dividing normal rat kidney (NRK) cells were treated with blebbistatin, a potent inhibitor of the non-muscle myosin II ATPase. Blebbistatin caused a strong inhibition of actin filament turnover and cytokinesis. Local release of blebbistatin at the equator caused inhibition of cytokinesis, while treatment in the polar region also caused a high frequency of abnormal cytokinesis, suggesting that myosin II may play a global role. These observations indicate that myosin II ATPase is essential for actin turnover and remodeling during cytokinesis. To further study the mechanism of myosin II and actin recruitment to the cytokinetic furrow, equatorial cortex were observed with total internal reflection fluorescence microscope (TIRF-M) coupled with spatial temporal image correlation spectroscopy (STICS) and a new approach termed temporal differential microscopy (TDM). The results indicated at least partially independent mechanisms for the early equatorial recruitment of myosin II and actin filaments. Cortical myosin II showed no detectable directional flow toward the equator. In addition to de novo equatorial assembly, localized inhibition of disassembly appeared to contribute to the formation of the equatorial myosin II band. In contrast, actin filaments underwent a striking, myosin II dependent flux toward the equator. However, myosin II was not required for equatorial actin concentration, suggesting that there was a flux-independent, de novo mechanism. The study was then extended to retraction fibers found typically on mitotic adherent cells, to address the hypothesis that they may facilitate post-mitotic spreading. Cells with retraction fibers showed increased spreading speed in post-mitotic spreading compared to cells without retraction fibers. In addition, micromanipulation study suggested that retraction fibers may guide the direction of post-mitotic spreading. Focal adhesion proteins were present at the tips of retraction fibers, and may act as small nucleators for focal adhesions reassembly that help cell quickly respread and regrow focal adhesions. These findings may suggest a general mechanism utilized by adherent cells to facilitate post-mitotic spreading and reoccupy their previous territory.

The three-dimensional organization of the human genome is non-random in interphase cells. Heterochromatin is highly clustered at the nuclear periphery, adjacent to nucleoli, and near centromeres. These localizations are reshuffled during mitosis when the chromosomes are condensed, nucleoli disassembled, and the nuclear envelope broken down. After cytokinesis, heterochromatin is re-localized to the domains described above. However, the mechanisms by which this localization is coordinated are not well understood. This dissertation will present evidence showing that both CAF-1 p150 and Ki-67 regulate nuclear structure throughout the human cell cycle. Chromatin Assembly Factor 1 (CAF-1) is a highly conserved three-subunit protein complex which deposits histones (H3/H4)2 heterotetramers onto replicating DNA during S-phase of the cell cycle. The N-terminal domain of the largest subunit of CAF-1 (p150N) is dispensable for histone deposition, and instead regulates the localization of specific loci (Nucleolar-Associated Domains, or “NADs”) and several proteins to the nucleolus during interphase. One of the proteins regulated by p150N is Ki-67, a protein widely used as a clinical marker of cellular proliferation. Depletion of Ki-67 decreases the association of NADs to the nucleolus in a manner similar to that of p150. Ki-67 is also a fundamental component of the perichromosomal layer (PCL), a sheath of proteins that surrounds all condensed chromosomes during mitosis. A subset of p150 localizes to the PCL during mitosis, and depletion of p150 disrupts Ki-67 localization to the PCL. This activity was mapped to the Sumoylation Interacting Motif (SIM) within p150N, which is also required for the localization of NADs and Ki-67 to the nucleolus during interphase. Together, these studies indicate that p150N coordinates the three-dimensional arrangement of both interphase and mitotic chromosomes via Ki-67.