-A +A

imprimer la page

Dynamique de la chromatine

Mots-clés : stabilité du génome, dynamique de la chromatine, régulation des fonctions nucléaires

Chef d'équipe : Geneviève Almouzni

EN BREF

Au delà de la séquence d’ADN, son organisation en chromatine dans le noyau cellulaire fournit une source d’information supplémentaire permettant d’étendre les possibilités de régulation du matériel génétique. Comment ces informations sont mises en place, changées et/ou transmises représente un enjeu majeur dans le domaine de l’épigénétique.

Notre équipe cherche à comprendre comment à la fois l’information génétique et son organisation sont établies, propagées/ changées ou maintenues et comment cette organisation peut être modulée au cours du développement, en fonction de l’environnement et en conditions pathologiques.

 

Publications de l'équipe


 



 

THEMATIQUE DE RECHERCHE

Le génome dans le noyau des cellules eucaryotes est organisé en trois dimensions dans l’espace sous la forme d’un complexe nucléoprotéique : la chromatine (Fig. 1).

Figure 1 : Représentation schématique des différents niveaux de compaction de la chromatine. A gauche : représentation artistique, à droite : représentation schématique (Probst et al., 2009)

 

Cette organisation non seulement permet de compacter l’ADN, mais aussi joue un rôle critique dans la régulation des interactions avec l’ADN au cours de son métabolisme, telles que réplication, recombinaison, transcription

Ce « conditionnement » du génome fournit un répertoire élargi d'informations qui s'ajoutent à celles apportées par le code génétique dont l’importance sur le plan épigénétique est un sujet majeur actuellement exploré. L’image suivante en couverture d’un numéro spécial "Epigenetics" de Cell en 2007 illustre ces concepts (Fig. 2). Des pierres de formes différentes représentent l’information génétique. Leur disposition comme l’arrangement de l'ADN en chromatine dans le noyau d’une cellule permet de dessiner un profil spécifique sur un sol pavé. Un profil parental donné peut donner deux cellules filles, l’une retenant le profil parental et l’autre changeant. Leur descendance maintient ensuite le profil dans un lignage spécifique. Ces éléments posent les bases pour un sujet de discussion tel que l’héritabilité épigénétique (Probst et al., 2009). Cette héritabilité au delà de l'ADN, s'adresse à des paramètres non codés génétiquement, cependant stables et héritables lors des divisions cellulaires.

 

Figure 2 : Cette couverture d'un numéro spécial de la revue Cell dédié à l'épigénétique illustre la mise en place et le maintien de profils épigénétiques qui définissent l'identité cellulaire. (Cell volume 128 de février 2007).

 

Une meilleure connaissance de l’organisation en chromatine, sa mise en place et sa stabilité/dynamique au cours des divisions cellulaires dans un lignage cellulaire donné est un enjeu pertinent en biologie du développement, une discipline dans laquelle la notion d’épigenèse a été appréciée dès l’antiquité. Comprendre les principes fondamentaux et mécanismes sous-jacents dans l’organisation de la chromatine sous ses formes variées, leur contrôle et leur modification afin de garantir un maintien de l’intégrité de l’information génétique tout en conservant une certaine plasticité est au cœur de notre programme.

La variabilité de la chromatine par l'utilisation de variants d'histones, les éléments protéiques majeurs de la chromatine et leur modification post-traductionnelle apporte un répertoire large d'information. Selon « l'hypothèse du code histone », les marques post-traductionnelles pourraient être lues par la cellule, et utilisées afin de définir des états de chromatine inertes, réprimés ou actifs. A Chaque type cellulaire un « épigénome » spécifique pourrait être assigné.

Notre équipe cherche à comprendre comment les informations génétiques et épigénétiques sont établies, propagées et maintenues, et comment elles peuvent changer au cours du développement et en réponse à des signaux environnementaux. Des erreurs potentielles peuvent conduire à la dérégulation des fonctions génomiques, ce qui peut avoir des conséquences pour de nombreuses maladies, y compris le cancer. Par conséquent, nous espérons que nos recherches permettront de mieux comprendre l'organisation nucléaire au cours de la vie normale de la cellule, mais également de comprendre des états pathologiques tels que le cancer et, enfin, de traiter le cancer.

 

Notre objectif général est de disséquer les mécanismes d'assemblage de la chromatine, en partant de son l'unité de base, le nucléosome, pour aller jusqu'aux niveau d'architecture d'ordre supérieur dans le noyau, comme les régions d'hétérochromatine (Fig. 3).

 

Figure 3 : Au sein de l'unité fondamentale de la chromatine la particule cœur du nucléosome, l'hélice d'ADN s'enroule autour d'un complexe protéique central d'histones. L'organisation régulière répétée de ce motif le long de l'ADN constitue le nucléofilament. L'architecture d'ordre supérieur de cette organisation aboutit finalement à la définition de domaines nucléaires, comme par exemple l'hétérochromatine péricentrique (ici points fluorescents visibles dans un noyau cellulaire de souris ; barre = 5 µm). Les nucléosomes peuvent se former de novo par un processus impliquant la mise en place de nouvelles histones sur de l'ADN libre, par exemple au cours de la réplication lorsque l'ADN nouvellement synthétisé se réorganise en chromatine. Cet assemblage de novo est coordonné au recyclage des anciennes histones provenant des nucléosomes parentaux. D'autres situations comme la réparation ou de la recombinaison de l'ADN, la transcription, la différenciation et le développement conduisent à remplacer ou recycler des histones. 

 

Au cours des dernières années (2005-2010), nous nous sommes intéressés particulièrement aux chaperons d'histone. Comme les histones sont des protéines très basiques, elles ont tendance à interagir de façon non spécifique avec d'autres protéines acides et avec des acides nucléiques ; les protéines chaperons garantissent le bon usage, au bon endroit des histones en les gardant sous surveillance. Ainsi, le transfert des histones d'un site à un autre, permet par exemple, de fournir les provisions d'histones nécessaires pour l'assemblage de la chromatine lors de la réplication de l'ADN. Ainsi, un flux d'histones constamment contrôlé permet à la cellule de s'adapter aux demandes physiologiques au cours du cycle et du développement cellulaires, ainsi qu'en réponse aux lésions de l'ADN.

 

Nous avons caractérisé plusieurs chaperons clés impliqués dans l'assemblage des nucléosomes : CAF-1, HIRA, ASF1a et ASF1b et HJURP. Nous avons identifié CAF-1 comme marqueur de la prolifération cellulaire dans le cancer du sein et Asf1b apparait prometteur comme marqueur pronostique des tumeurs à tendance métastatique. Ces chaperons font partie de complexes multiprotéiques in vivo, avec différentes spécificités pour les variants de chaque histone H3 (Fig. 4) nous amenant à en analyser leur dynamique.

 

Figure 4 : Les dimères d‘histones H3-H4 s'associent aux chaperons d'histones dans des complexes solubles avant incorporation au sein de nucleosomes (figure du haut), suggérant que H3 et H4 sont mis en place sous forme de dimères, comme H2A et H2B, au cours de la formation de novo des nucléosomes (figure du bas). Il reste à évaluer si la réaction inverse implique des intermédiaires et complexes similaires. Les flèches rouges indiquent l'interface impliquée dans la dimérisation de H3 et l'interaction entre H3 et Asf1. Les spécificités des chaperons pour les variants de H3 sont indiquées. (Extrait de Polo SE and Almouzni G (2006) Curr. Opin. Genet. Dev. 16, 104-111)

 

Nous avons observé la dynamique d'une nouvelle incorporation d'histones au cours de la réparation des lésions de la chromatine dues aux UV. Des modifications spécifiques peuvent être décelées sur les histones avant même leur incorporation dans la chromatine. Nos résultats dans l'ensemble apportent une lumière sur les problèmes fondamentaux de la dynamique, du destin et de l'héritage des histones avec leurs marques spécifiques typiques de domaines particuliers de la chromatine. L'un des défis actuels est de comprendre comment le maintien et la duplication des informations génétiques et épigénétiques sont assurés et coordonnés. Notre hypothèse de travail est que les chaperons d'histones fonctionnent selon une "chaîne de montage ", et que leur spécificité pour des variants d'histones particuliers contribue au marquage spécifique de régions définies du génome.

Notre programme vise à analyser les voies de régulation ciblant les chaperons d'histones afin de contrôler cette chaîne de montage et son réseau de connexion. Notre approche spécifique s'appuie sur des outils et systèmes modèles combinant la biochimie, afin d'étudier les complexes au niveau moléculaire, et cellulaire, afin de les suivre in vivo puis d'étudier des domaines nucléaires spécifiques , par exemple, hétérochromatine centromérique ; Fig. 5). En effet, le centromere, région clef du chromosome qui assure sa distribution à chaque division cellulaire représente un excellent modèle de domaine dont l'identité est définie de manière épigénétique.

 

Figure 5 : (A) Les régions d'hétérochromatine dans les centromères sont composées de nombreuses répétitions d'ADN appelées répétitions sateliites majeures dans la région péricentrique (en vert) et répétitions satellites mineures dans la partie centrique (en rouge). Ces répétitions sont fortement méthylées. Des histones sous-acétylées et des histones H3 méthylées sur la lysine 9 (H3-K9 ; la méthylation des histones est indiquée par une étoile bleue) sont présentes, ainsi que l'histone H3-K9 méthyltransférase, Suv39h. La méthylation de H3-K9 offre des sites de liaison pour les protéines péricentriques HP1 dans des cellules de souris et Swi6 dans la levure Schizosaccharomyces pombe. La région centrique du centromère est représentée par une étoile rouge à huit branches. Le composant ARN est illustré par une ligne rouge entre HP1 et H3. (Adapté de Maison C and Almouzni G (2004) Nature Rev. Mol. Cell. Biol. 5, 296-304). (B) Les régions d'ADN satellite majeur (en vert) et mineur (en rouge) visualisées par hybridation in situ en fluorescence définissent des domaines 3D dans des noyaux de souris en interphase. (Extrait de Guenatri M, Bailly D, Maison C, Almouzni G. (2004) J. Cell Biol. 166, 493-505)

 

Comment la fonction du centromère est influencée par l'organisation en hétérochromatine des regions péricentriques constitute un domaine qui nous intéresse particulièrement. Nous avons ainsi cherché à comprendre comment s'effectue le ciblage des protéines HP1 dans les régions péricentriques. Nous montrons que la modification de HP1 par sumoylation est un évènement déterminant pour son association spécifique avec des ARNs codés dans les régions péricentriques. Nous proposons un modèle pour le ciblage de novo de HP1 dans l'hétérochromatine péricentrique.

Nos études développementales exploitent l'avantage de la génétique chez la souris, et les approches expérimentales simples sur l'amphibien,  Xenopus comme organisme modèle. Ainsi, nous cherchons à valider les hypothèses formulées sur la base  de nos résultats obtenus in vitro. Ces approches devraient aider dans des étapes ultérieures à développer des applications médicales. Nos études les plus récentes illustrent le role critique de facteurs d'assemblage et d'ARN non codants dans le control de l'organisation complexe de l'hétéchromatine dans les étapes de développement préimplantatoire chez l'embryon de souris, une étape durant laquelle une reprogrammation majeure du génome a lieu.

 

Notre équipe a été un membre coordonnateur du Réseau d'excellence Epigénome (2005-2010) : responsable de la communauté de recherche européenne sur l'épigénétique, et membre du Réseau Européen de formation-recherche (RTN) travaillant sur les points de contrôle, la réponse aux lésions de l'ADN et le cancer. Nous sommes maintenant coordonnateur du Réseau d'excellence Epigenesys (2010-2015), qui vise à relever de nouveaux défis pour coordonner les questionnement épigénétiques à des approches de la biologie des systèmes. Nous sommes également impliqués dans deux réseaux de formation européens (ITN), l'un sur l'organisation du nucleosome en 4 dimensions, et l'autre sur l'imagerie de la réponse aux dommage dans l'ADN.

Pour le réseau Epigenesys voir: http://www.epigenesys.org

Publications clés

  • Year of publication : 2015

  • Year of publication : 2014

  • Centromeres are essential for ensuring proper chromosome segregation in eukaryotes. Their definition relies on the presence of a centromere-specific H3 histone variant CenH3, known as CENP-A in mammals. Its overexpression in aggressive cancers raises questions concerning its effect on chromatin dynamics and contribution to tumorigenesis. We find that CenH3 overexpression in human cells leads to ectopic enrichment at sites of active histone turnover involving a heterotypic tetramer containing CenH3-H4 with H3.3-H4. Ectopic localization of this particle depends on the H3.3 chaperone DAXX rather than the dedicated CenH3 chaperone HJURP. This aberrant nucleosome occludes CTCF binding and has a minor effect on gene expression. Cells overexpressing CenH3 are more tolerant of DNA damage. Both the survival advantage and CTCF occlusion in these cells are dependent on DAXX. Our findings illustrate how changes in histone variant levels can disrupt chromatin dynamics and suggests a possible mechanism for cell resistance to anticancer treatments.

  • Correct chromosome segregation requires a unique chromatin environment at centromeres and in their vicinity. Here, we address how the deposition of canonical H2A and H2A.Z histone variants is controlled at pericentric heterochromatin (PHC). Whereas in euchromatin newly synthesized H2A and H2A.Z are deposited throughout the cell cycle, we reveal two discrete waves of deposition at PHC - during mid to late S phase in a replication-dependent manner for H2A and during G1 phase for H2A.Z. This G1 cell cycle restriction is lost when heterochromatin features are altered, leading to the accumulation of H2A.Z at the domain. Interestingly, compromising PHC integrity also impacts upon neighboring centric chromatin, increasing the amount of centromeric CENP-A without changing the timing of its deposition. We conclude that the higher-order chromatin structure at the pericentric domain influences dynamics at the nucleosomal level within centromeric chromatin. The two different modes of rearrangement of the PHC during the cell cycle provide distinct opportunities to replenish one or the other H2A variant, highlighting PHC integrity as a potential signal to regulate the deposition timing and stoichiometry of histone variants at the centromere.

  • Year of publication : 2013

  • The equalization of pericentric heterochromatin from distinct parental origins following fertilization is essential for genome function and development. The recent implication of noncoding transcripts in this process raises questions regarding the connection between RNA and the nuclear organization of distinct chromatin environments. Our study addresses the interrelationship between replication and transcription of the two parental pericentric heterochromatin (PHC) domains and their reorganization during early embryonic development. We demonstrate that the replication of PHC is dispensable for its clustering at the late two-cell stage. In contrast, using parthenogenetic embryos, we show that pericentric transcripts are essential for this reorganization independent of the chromatin marks associated with the PHC domains. Finally, our discovery that only reverse pericentric transcripts are required for both the nuclear reorganization of PHC and development beyond the two-cell stage challenges current views on heterochromatin organization.

  • Understanding how to recover fully functional and transcriptionally active chromatin when its integrity has been challenged by genotoxic stress is a critical issue. Here, by investigating how chromatin dynamics regulate transcriptional activity in response to DNA damage in human cells, we identify a pathway involving the histone chaperone histone regulator A (HIRA) to promote transcription restart after UVC damage. Our mechanistic studies reveal that HIRA accumulates at sites of UVC irradiation upon detection of DNA damage prior to repair and deposits newly synthesized H3.3 histones. This local action of HIRA depends on ubiquitylation events associated with damage recognition. Furthermore, we demonstrate that the early and transient function of HIRA in response to DNA damage primes chromatin for later reactivation of transcription. We propose that HIRA-dependent histone deposition serves as a chromatin bookmarking system to facilitate transcription recovery after genotoxic stress.

  • Year of publication : 2012

  • Discovering how histone variants that mark distinct chromatin regions affect a developmental program is a major challenge in the epigenetics field. To assess the importance of the H3.3 histone variant and its dedicated histone chaperone HIRA, we used an established developmental model, Xenopus laevis. After the early rapid divisions exploiting a large maternal pool of both replicative H3.2 and replacement H3.3, H3.3 transcripts show a distinct peak of expression at gastrulation. Depletion of both H3.2 and H3.3 leads to an early gastrulation arrest. However, with only H3.3 depletion, defects occur at late gastrulation, impairing further development. Providing exogenous H3.3 mRNAs, but not replicative H3.2 mRNAs, rescues these defects. Notably, downregulation of the H3.3 histone chaperone HIRA similarly impairs late gastrulation, and we find a global defect in H3.3 incorporation into chromatin comparable to H3.3 depletion. We discuss how specific HIRA-dependent H3.3 deposition is required for chromatin dynamics during gastrulation.

  • During immune responses, naive CD4+ T cells differentiate into several T helper (TH) cell subsets under the control of lineage-specifying genes. These subsets (TH1, TH2 and TH17 cells and regulatory T cells) secrete distinct cytokines and are involved in protection against different types of infection. Epigenetic mechanisms are involved in the regulation of these developmental programs, and correlations have been drawn between the levels of particular epigenetic marks and the activity or silencing of specifying genes during differentiation. Nevertheless, the functional relevance of the epigenetic pathways involved in TH cell subset differentiation and commitment is still unclear. Here we explore the role of the SUV39H1-H3K9me3-HP1alpha silencing pathway in the control of TH2 lineage stability. This pathway involves the histone methylase SUV39H1, which participates in the trimethylation of histone H3 on lysine 9 (H3K9me3), a modification that provides binding sites for heterochromatin protein 1alpha (HP1alpha) and promotes transcriptional silencing. This pathway was initially associated with heterochromatin formation and maintenance but can also contribute to the regulation of euchromatic genes. We now propose that the SUV39H1-H3K9me3-HP1alpha pathway participates in maintaining the silencing of TH1 loci, ensuring TH2 lineage stability. In TH2 cells that are deficient in SUV39H1, the ratio between trimethylated and acetylated H3K9 is impaired, and the binding of HP1alpha at the promoters of silenced TH1 genes is reduced. Despite showing normal differentiation, both SUV39H1-deficient TH2 cells and HP1alpha-deficient TH2 cells, in contrast to wild-type cells, expressed TH1 genes when recultured under conditions that drive differentiation into TH1 cells. In a mouse model of TH2-driven allergic asthma, the chemical inhibition or loss of SUV39H1 skewed T-cell responses towards TH1 responses and decreased the lung pathology. These results establish a link between the SUV39H1-H3K9me3-HP1alpha pathway and the stability of TH2 cells, and they identify potential targets for therapeutic intervention in TH2-cell-mediated inflammatory diseases.

  • SUMOylation promotes targeting of HP1alpha to pericentric heterochromatin. Here we identify the SUMO-specific protease SENP7 in mouse as a maintenance factor for HP1alpha accumulation at this location. SENP7 interacts directly with HP1alpha, localizes at HP1-enriched pericentric domains and can deconjugate SUMOylated HP1alpha in vivo. Depletion of SENP7 delocalizes HP1alpha from pericentric heterochromatin without affecting H3K9me3 levels. We propose that following targeting of HP1alpha, a subsequent deSUMOylation event enables HP1alpha retention at these domains.

  • Year of publication : 2011

  • HP1 enrichment at pericentric heterochromatin is considered important for centromere function. Although HP1 binding to H3K9me3 can explain its accumulation at pericentric heterochromatin, how it is initially targeted there remains unclear. Here, in mouse cells, we reveal the presence of long nuclear noncoding transcripts corresponding to major satellite repeats at the periphery of pericentric heterochromatin. Furthermore, we find that major transcripts in the forward orientation specifically associate with SUMO-modified HP1 proteins. We identified this modification as SUMO-1 and mapped it in the hinge domain of HP1alpha. Notably, the hinge domain and its SUMOylation proved critical to promote the initial targeting of HP1alpha to pericentric domains using de novo localization assays, whereas they are dispensable for maintenance of HP1 domains. We propose that SUMO-HP1, through a specific association with major forward transcript, is guided at the pericentric heterochromatin domain to seed further HP1 localization.

  • Proper genome packaging requires coordination of both DNA and histone metabolism. While histone gene transcription and RNA processing adequately provide for scheduled needs, how histone supply adjusts to unexpected changes in demand remains unknown. Here, we reveal that the histone chaperone nuclear autoantigenic sperm protein (NASP) protects a reservoir of soluble histones H3-H4. The importance of NASP is revealed upon histone overload, engagement of the reservoir during acute replication stress, and perturbation of Asf1 activity. The reservoir can be fine-tuned, increasing or decreasing depending on the level of NASP. Our data suggest that NASP does so by balancing the activity of the heat shock proteins Hsc70 and Hsp90 to direct H3-H4 for degradation by chaperone-mediated autophagy. These insights into NASP function and the existence of a tunable reservoir in mammalian cells demonstrate that contingency is integrated into the histone supply chain to respond to unexpected changes in demand.

  • Establishment of a proper chromatin landscape is central to genome function. Here, we explain H3 variant distribution by specific targeting and dynamics of deposition involving the CAF-1 and HIRA histone chaperones. Impairing replicative H3.1 incorporation via CAF-1 enables an alternative H3.3 deposition at replication sites via HIRA. Conversely, the H3.3 incorporation throughout the cell cycle via HIRA cannot be replaced by H3.1. ChIP-seq analyses reveal correlation between HIRA-dependent H3.3 accumulation and RNA pol II at transcription sites and specific regulatory elements, further supported by their biochemical association. The HIRA complex shows unique DNA binding properties, and depletion of HIRA increases DNA sensitivity to nucleases. We propose that protective nucleosome gap filling of naked DNA by HIRA leads to a broad distribution of H3.3, and HIRA association with Pol II ensures local H3.3 enrichment at specific sites. We discuss the importance of this H3.3 deposition as a salvage pathway to maintain chromatin integrity.

  • Year of publication : 2010

  • At the time of fertilization, the paternal genome lacks the typical configuration and marks characteristic of pericentric heterochromatin. It is thus essential to understand the dynamics of this region during early development, its importance during that time period and how a somatic configuration is attained. Here, we show that pericentric satellites undergo a transient peak in expression precisely at the time of chromocenter formation. This transcription is regulated in a strand-specific manner in time and space and is strongly biased by the parental asymmetry. The transcriptional upregulation follows a developmental clock, yet when replication is blocked chromocenter formation is impeded. Furthermore, interference with major satellite transcripts using locked nucleic acid (LNA)-DNA gapmers results in developmental arrest before completion of chromocenter formation. We conclude that the exquisite strand-specific expression dynamics at major satellites during the 2-cell stage, with both up and downregulation, are necessary events for proper chromocenter organization and developmental progression.

  • Year of publication : 2009

  • The histone H3 variant CenH3, called CENP-A in humans, is central in centromeric chromatin to ensure proper chromosome segregation. In the absence of an underlying DNA sequence, it is still unclear how CENP-A deposition at centromeres is determined. Here, we purified non-nucleosomal CENP-A complexes to identify direct CENP-A partners involved in such a mechanism and identified HJURP. HJURP was not detected in H3.1- or H3.3-containing complexes, indicating its specificity for CENP-A. HJURP centromeric localization is cell cycle regulated, and its transient appearance at the centromere coincides precisely with the proposed time window for new CENP-A deposition. Furthermore, HJURP downregulation leads to a major reduction in CENP-A at centromeres and impairs deposition of newly synthesized CENP-A, causing mitotic defects. We conclude that HJURP is a key factor for CENP-A deposition and maintenance at centromeres.

  • Studies that concern the mechanism of DNA replication have provided a major framework for understanding genetic transmission through multiple cell cycles. Recent work has begun to gain insight into possible means to ensure the stable transmission of information beyond just DNA, and has led to the concept of epigenetic inheritance. Considering chromatin-based information, key candidates have arisen as epigenetic marks, including DNA and histone modifications, histone variants, non-histone chromatin proteins, nuclear RNA as well as higher-order chromatin organization. Understanding the dynamics and stability of these marks through the cell cycle is crucial in maintaining a given chromatin state.

  • Year of publication : 2007

  • In eukaryotes, DNA is organized into chromatin in a dynamic manner that enables it to be accessed for processes such as transcription and repair. Histones, the chief protein component of chromatin, must be assembled, replaced or exchanged to preserve or change this organization according to cellular needs. Histone chaperones are key actors during histone metabolism. Here we classify known histone chaperones and discuss how they build a network to escort histone proteins. Molecular interactions with histones and their potential specificity or redundancy are also discussed in light of chaperone structural properties. The multiplicity of histone chaperone partners, including histone modifiers, nucleosome remodelers and cell-cycle regulators, is relevant to their coordination with key cellular processes. Given the current interest in chromatin as a source of epigenetic marks, we address the potential contributions of histone chaperones to epigenetic memory and genome stability.

  • Inheritance and maintenance of the DNA sequence and its organization into chromatin are central for eukaryotic life. To orchestrate DNA-replication and -repair processes in the context of chromatin is a challenge, both in terms of accessibility and maintenance of chromatin organization. To meet the challenge of maintenance, cells have evolved efficient nucleosome-assembly pathways and chromatin-maturation mechanisms that reproduce chromatin organization in the wake of DNA replication and repair. The aim of this Review is to describe how these pathways operate and to highlight how the epigenetic landscape may be stably maintained even in the face of dramatic changes in chromatin structure.

  • DNA replication in eukaryotes requires nucleosome disruption ahead of the replication fork and reassembly behind. An unresolved issue concerns how histone dynamics are coordinated with fork progression to maintain chromosomal stability. Here, we characterize a complex in which the human histone chaperone Asf1 and MCM2-7, the putative replicative helicase, are connected through a histone H3-H4 bridge. Depletion of Asf1 by RNA interference impedes DNA unwinding at replication sites, and similar defects arise from overproduction of new histone H3-H4 that compromises Asf1 function. These data link Asf1 chaperone function, histone supply, and replicative unwinding of DNA in chromatin. We propose that Asf1, as a histone acceptor and donor, handles parental and new histones at the replication fork via an Asf1-(H3-H4)-MCM2-7 intermediate and thus provides a means to fine-tune replication fork progression and histone supply and demand.

  • Year of publication : 2006

  • Histone posttranslational modifications (PTMs) and sequence variants regulate genome function. Although accumulating evidence links particular PTM patterns with specific genomic loci, our knowledge concerning where and when these PTMs are imposed remains limited. Here, we find that lysine methylation is absent prior to histone incorporation into chromatin, except at H3K9. Nonnucleosomal H3.1 and H3.3 show distinct enrichments in H3K9me, such that H3.1 contains more K9me1 than H3.3. In addition, H3.3 presents other modifications, including K9/K14 diacetylated and K9me2. Importantly, H3K9me3 was undetectable in both nonnucleosomal variants. Notably, initial modifications on H3 variants can potentiate the action of enzymes as exemplified with Suv39HMTase to produce H3K9me3 as found in pericentric heterochromatin. Although the set of initial modifications present on H3.1 is permissive for further modifications, in H3.3 a subset cannot be K9me3. Thus, initial modifications impact final PTMs within chromatin.

  • Chromatin organization is compromised during the repair of DNA damage. It remains unknown how and to what extent epigenetic information is preserved in vivo. A central question is whether chromatin reorganization involves recycling of parental histones or new histone incorporation. Here, we devise an approach to follow new histone deposition upon UV irradiation in human cells. We show that new H3.1 histones get incorporated in vivo at repair sites. Remarkably we find that H3.1, which is deposited during S phase, is also incorporated outside of S phase. Histone deposition is dependent on nucleotide excision repair (NER), indicating that it occurs at a postrepair stage. The histone chaperone chromatin assembly factor 1 (CAF-1) is directly involved in the histone deposition process in vivo. We conclude that chromatin restoration after damage cannot rely simply on histone recycling. New histone incorporation at repair sites both challenges epigenetic stability and possibly contributes to damage memory.

  • Year of publication : 2005

  • Maintenance of chromosomal integrity requires tight coordination of histone biosynthesis with DNA replication. Here, we show that extracts from human cells exposed to replication stress display an increased capacity to support replication-coupled chromatin assembly. While in unperturbed S phase, hAsf1 existed in equilibrium between an active form and an inactive histone-free pool, replication stress mobilized the majority of hAsf1 into an active multichaperone complex together with histones. This active multichaperone complex was limiting for chromatin assembly in S phase extracts, and hAsf1 was required for the enhanced assembly activity in cells exposed to replication stress. Consistently, siRNA-mediated knockdown of hAsf1 impaired the kinetics of S phase progression. Together, these data suggest that hAsf1 provides the cells with a buffering system for histone excess generated in response to stalled replication and explains how mammalian cells maintain a critical "active" histone pool available for deposition during recovery from replication stresses.

  • Year of publication : 2004

  • To investigate how the complex organization of heterochromatin is reproduced at each replication cycle, we examined the fate of HP1-rich pericentric domains in mouse cells. We find that replication occurs mainly at the surface of these domains where both PCNA and chromatin assembly factor 1 (CAF-1) are located. Pulse-chase experiments combined with high-resolution analysis and 3D modeling show that within 90 min newly replicated DNA become internalized inside the domain. Remarkably, during this time period, a specific subset of HP1 molecules (alpha and gamma) coinciding with CAF-1 and replicative sites is resistant to RNase treatment. Furthermore, these replication-associated HP1 molecules are detected in Suv39 knockout cells, which otherwise lack stable HP1 staining at pericentric heterochromatin. This replicative pool of HP1 molecules disappears completely following p150CAF-1 siRNA treatment. We conclude that during replication, the interaction of HP1 with p150CAF-1 is essential to promote delivery of HP1 molecules to heterochromatic sites, where they are subsequently retained by further interactions with methylated H3-K9 and RNA.

  • Deposition of the major histone H3 (H3.1) is coupled to DNA synthesis during DNA replication and possibly DNA repair, whereas histone variant H3.3 serves as the replacement variant for the DNA-synthesis-independent deposition pathway. To address how histones H3.1 and H3.3 are deposited into chromatin through distinct pathways, we have purified deposition machineries for these histones. The H3.1 and H3.3 complexes contain distinct histone chaperones, CAF-1 and HIRA, that we show are necessary to mediate DNA-synthesis-dependent and -independent nucleosome assembly, respectively. Notably, these complexes possess one molecule each of H3.1/H3.3 and H4, suggesting that histones H3 and H4 exist as dimeric units that are important intermediates in nucleosome formation. This finding provides new insights into possible mechanisms for maintenance of epigenetic information after chromatin duplication.

  • Year of publication : 2002

  • Post-translational modification of histone tails is thought to modulate higher-order chromatin structure. Combinations of modifications including acetylation, phosphorylation and methylation have been proposed to provide marks recognized by specific proteins. This is exemplified, in both mammalian cells and fission yeast, by transcriptionally silent constitutive pericentric heterochromatin. Such heterochromatin contains histones that are generally hypoacetylated and methylated by Suv39h methyltransferases at lysine 9 of histone H3 (H3-K9). Each of these modification states has been implicated in the maintenance of HP1 protein-binding at pericentric heterochromatin, in transcriptional silencing and in centromere function. In particular, H3-K9 methylation is thought to provide a marking system for the establishment and maintenance of stably repressed regions and heterochromatin subdomains. To address the question of how these two types of modifications, as well as other unidentified parameters, function to maintain pericentric heterochromatin, we used a combination of histone deacetylase inhibitors, RNAse treatments and an antibody raised against methylated branched H3-K9 peptides. Our results show that both H3-K9 acetylation and methylation can occur on independent sets of H3 molecules in pericentric heterochromatin. In addition, we identify an RNA- and histone modification-dependent structure that brings methylated H3-K9 tails together in a specific configuration required for the accumulation of HP1 proteins in these domains.