Les facteurs de transcription (FT) de la famille R2R3-MYB chez les plantes jouent un rôle important dans la formation de la paroi secondaire des cellules de bois, que ce soit en activant ou en réprimant leurs gènes cibles au sein d’un réseau régulationnel complexe. Dans ce travail, nous avons utilisé la transgénèse et l’immunoprécipitation de chromatine associée à un séquençage haut-débit (ChIP-SEQ) pour déterminer la fonction de 3 FT R2R3-MYB chez le peuplier. Les peupliers surexprimant MYB090 ont des rayons moins lignifiés ; les tiges présentent une réduction de croissance et de teneurs en lignines. MYB090 régule ses cibles à l’aide d’un motif très conservé, similaire au motif Gamyb. Ses cibles sont impliquées notamment dans la biosynthèse des lignines, cellulose et xylanes, constituants principaux des parois. Les plantes surexprimant MYB221-SRDX et MYB156 présentent une nette réduction de la lignification des parois de leurs fibres, associée à une réduction de croissance. MYB221 semble avoir pour cibles des gènes codant pour des enzymes du métabolisme, ainsi qu’un autre FT de type R2R3-MYB, dont la régulation passe par un motif conservé de type SMRE (Secondary wall MYB-Responsive Element). En conclusion, la combinaison des approches ChIP-SEQ et de transgénèse montre que MYB090 semble être un répresseur transcriptionnel de la lignification, notamment dans les rayons, et de la formation de la paroi secondaire. De même, MYB156 et MYB221 seraient également des répresseurs de la lignification, dans les fibres et les rayons. Cette thèse ouvre des perspectives sur l’établissement de réseaux de régulation transcriptionnelle de la formation de la paroi secondaire. / Plant R2R3-MYB transcription factors (TF) play an important role in secondary cell wall formation in wood cells, by activating or repressing their target genes within a complex regulatory network. Here, we used genetic engineering and chromatin immunoprecipitation technique, associated to next-generation sequencing (ChIP-SEQ) to determine the function of 3 R2R3-MYB TF in poplar. Plants overexpressing MYB090 had less lignified parenchyma rays. The stem growth and total lignin content were reduced. MYB090 regulates target genes through a highly conserved motif, similar to Gamyb. Its target genes are involved in lignin, cellulose and xylan biosynthesis, which are the major components of secondary cell wall. Poplars overexpressing MYB221-SRDX and MYB156 showed a decrease in fiber cell wall lignification, and a reduced growth. MYB221 have targets encoding for metabolic enzymes but also for another R2R3-MYB TF. MYB221 recognizes its target genes, most probably through SMRE (Secondary wall MYB-Responsive Element) conserved motif. In conclusion, the combination of ChIP-SEQ and genetic engineering approaches shows that MYB090 seems to be a transcriptional repressor of lignification, especially in parenchyma rays. MYB156 and MYB221 are also negative regulators of secondary cell wall lignification, in fibers and parenchyma rays. This work opens new avenues on the building of transcriptional regulatory networks involved in secondary cell wall formation.

Les ascidies présentent des propriétés remarquables au sein des métazoaires qui en font un modèle particulièrement intéressant pour étudier le fonctionnement et l’évolution des éléments cis-régulateurs dans un contexte développemental. Ciona intestinalis et Phallusia mammillata, deux espèces d’ascidies qui ont divergé il y a environ 300 millions d’années, combinent une grande conservation de leurs processus développementaux avec une grande divergence de leur séquence génomique. Pour comprendre comment « fabriquer » des embryons similaires avec des génomes divergents, nous avons identifié les éléments cis-régulateurs actifs au cours du développement de Ciona intestinalis et Phallusia mammillata en développant et en appliquant la méthode de ChIP-Seq sur des modifications d’histones sur des jeunes gastrulae. La définition puis la validation fonctionnelle de différentes catégories d'éléments cis-régulateurs nous a permis de révéler quelques propriétés de la cis-régulation au sein de génomes compacts et intensément remaniés. En sus, les données que nous avons produites constituent une resource fonctionnelle unique pour la caractérisation des éléments cis-régulateurs chez les ascidies et l'étude de leur évolution au sein des Chordés. / Ascidians display remarkable features within metazoans making them particularly suited for the study of function and evolution of cis-regulatory elements in the context of embryonic development. Ciona intestinalis and Phallusia mammillata, two ascidian species that diverged about 300M years ago, combine high conservation of their developmental processes with high divergence of their genome sequence. To understand how to “make” similar embryos with divergent genomes, we identified active cis-regulatory elements during Ciona intestinalis and Phallusia mammillata development by developing and applying the ChIP-Seq method on histone modifications in early-gastrula embryos. Definition then functional validation of different categories of cis-regulatory elements led us to reveal some features of cis-regulation within compact and highly dynamic genomes. Together, our data constitute a unique functional resource for characterizing cis-regulatory elements in ascidians and questioning their evolution within the Chordates.

Des liens de caténation sont progressivement crées lors de la réplication de l’ADN et sont responsables de la cohésion des chromatides sœurs. La topoisomérase IV est une topoisomérase de type II impliquée dans la résolution de ces liens de caténation accumulés derrière la fourche de réplication, et lors de la dernière étape de séparation des chromatides sœurs à la fin de la réplication. Nous avons étudié la liaison de la topoIV à l’ADN ainsi que son activité catalytique à l’aide de méthodes de biologie moléculaire et de génomique. Une expérience de ChIPseq a révélé que l’interaction de la topoIV de chez E.coli avec l’ADN est contrôlée par la réplication. Durant la réplication, la topoIV a accès à des centaines de sites sur l’ADN mais ne se lie qu’à quelques sites où elle exerce son activité catalytique. La conformation locale de la chromatine et l’expression des gènes influencent la sélection de certains sites. De plus, une forte liaison et une activité catalytique renforcée a été trouvée au site de résolution des dimers, dif. Le site dif est situé à l’opposé de l’origine de réplication dans le macrodomaine ter. Nous avons montré qu’il existe une interaction physique et fonctionnelle entre la topoIV et la recombinase XerCD, qui agit au site dif. Cette interaction est médiée par MatP, une protéine essentielle dans l’organisation du macrodomaine ter. L’ensemble de ces résultats montre que la topoIV, XerCD/dif et MatP œuvrent ensemble pour permettre l’étape finale de ségrégation des chromosomes lors du cycle cellulaire. / Catenation links between sister chromatids are formed progressively during DNA replication and are involved in the establishment of sister chromatid cohesion. Topo IV is a bacterial type II topoisomerase involved in the removal of catenation links both behind replication forks and after replication during the final separation of sister chromosomes. We have investigated the global DNA-binding and catalytic activity of Topo IV in E. coli using genomic and molecular biology approaches. ChIP-seq revealed that Topo IV interaction with the E. coli chromosome is controlled by DNA replication. During replication, Topo IV has access to most of the genome but only selects a few hundred specific sites for its activity. Local chromatin and gene expression context influence site selection. Moreover strong DNA-binding and catalytic activities are found at the chromosome dimer resolution site, dif, located opposite the origin of replication. We reveal a physical and functional interaction between Topo IV and the XerCD recombinases acting at the dif site. This interaction is modulated by MatP, a protein involved in the organization of the Ter macrodomain. These results show that Topo IV, XerCD/dif and MatP are part of a network dedicated to the final step of chromosome management during the cell cycle.

Covalent modification of histones is an important factor in the regulation of the chromatin structure implicated in DNA replication, repair, recombination, and transcription, as well as in RNA processing. In recent years, histone methylation has emerged as one of the key modifications regulating chromatin function. However, the mechanisms involved are complex and not well understood. Histone 3 lysine 4 (H3K4) methylation is deposited by a family of histone H3K4 methyltransferases (HMTs) that share a conserved SET domain. In mammalian cells, six family members have been characterized: Setd1a and Setd1b (the mammalian orthologs of yeast Set1) and four Mixed lineage leukemia (Mll) family HMTs, which share limited similarity with yeast Set1 beyond the SET domain. Several studies demonstrated that the H3K4 methyltransferases exist as multiprotein complexes. To functionally dissect H3K4 methyltransferase complexes, GFP tagging of the core subunit Ash2l and the complex-specific subunits Cxxc1 and Wdr82 (Setd1a/b complexes) Men1 (Mll1/2 complexes), and Ptip (Mll3/Mll4 complexes), was used. The fusion proteins were successfully expressed in mouse embryonic stem cells (ES cells), analyzed by confocal microscopy, Mass Spectrometry (MS) and ChIP-seq. Ptip was the only subunit able to bind mitotic chromatin. Additionally, both Ptip and Wdr82 were found to associate with cell cycle regulators, suggesting a possible role of the two proteins or respective complexes in cell cycle regulation. Mass Spectrometry revealed that Wdr82 and Ptip interact with members of he PAF complex, and ChIP-seq showed that Wdr82, Cxxc1 and Ptip positively modulate pluripotency genes. Thus, Setd1a/b and Mll3/4 complexes might act together in the regulation of embryonic stem cells identity. Protein pull downs identified at least one new Setd1a/b interactor, Bod1l that is orthologous to the yeast protein Sgh1, a component of the Set1C complex. Furthermore, our MS and ChIP-seq data suggested that only Mll2 complex binds to bivalent promoters, wheras Mll2 and Setd1a complexes might function together in a set of promoters.

HOX-Gene sind essentielle Transkriptionsfaktoren (TFs), die den Körperplan, die Struktur und die Organbildung während der Entwicklung bestimmen. Diese komplexen Prozesse werden präzise von in verschachtelter Weise exprimierten HOX-Genen reguliert. In vitro Experimente zeigten jedoch, dass die HOX-DNA-Bindungsdomäne stark konserviert ist und oft ähnliche DNA-Sequenzen bindet. Die niedrige biochemische Bindungsspezifität und die hochspezifischen Funktionen stehen oft im Widerspruch und bilden das Schlussthema des so genannten Hox-Paradoxons. Das Paradox besteht aufgrund der folgenden Hindernisse: hohe Proteinhomologie, unspezifischen Antikörper sowie die verschachtelte HOX-Expressionsmuster. Das Ziel dieser Arbeit war, diese Probleme zu überwinden, die HOX-DNA-Bindung in kontrollierten und physiologischen Umstände zu untersuchen und die Bindung von neun Gliedmaßen-spezifischen posterioren HOXA und -D-TFs zu vergleichen. Zu diesem Zweck wurden neun Hühner-HOX-Gene (HOXA- und HOXD9-13) mit dem FLAG markiert und mittels Viren in Gliedmaßen-mesenchymalen Zellen exprimiert. Somit wurde der Vergleich unter identischen und kontrollierten Bedingungen ermöglicht. Im Einklang mit in vivo Funktionsdaten zeigten die HOX-Bindungsprofile, dass zwei direkte Paraloge (z. B. HOXA10 und D10) häufiger dieselben Regionen binden als zwei Nicht-Paraloge (z. B. HOXA9 und A13). Außerdem, die hier beschriebene HOX-DNA-Bindung unterscheidet sich von in vitro Bindung, was darauf hinweist, dass Kofaktoren für deren biologische Funktion wichtig sind. Zusätzlich ergab sich aus dem Bindungsvergleich, dass es zuvor unbekannte Unterschiede zwischen Bindungsweise von HOX-TFs gibt, die zumindest teilweise auf der Häufigkeit von direkter Bindung und Ko-Bindung mit anderen TFs beruhen. Schließlich wurde mit der Kombination von Genetik, Genomik und Biochemie einen neuen HOX-Kofaktor entdeckt, CTCF, der auf ein mögliches Wechselspiel zwischen der HOX-Zielregulation und der Chromatinarchitektur hindeutet. / HOX genes are essential developmental transcription factors (TFs) that pattern the animal body plan, their structures and organs. To precisely control these very diverse processes HOX genes are expressed in a nested fashion and regulate their targets in a context specific way. However, in vitro experiments indicated that HOX DNA binding domain (Homeodomain) is remarkably rigid and often binds very similar DNA sequences. This discrepancy between high functional specificity and low in vitro biochemical specificity is at the core of a problem termed Hox paradox. This paradox persists due to several biological and technical obstacles; namely high HOX protein homology and lack of sufficiently specific antibodies as well as nested HOX expression pattern. The aim of this study was to address these problems, study HOX-DNA binding in a controlled, Hox-native environment and to compare HOX-DNA binding of nine posterior vertebrate HOXA and HOXD TFs. To do this, nine chicken HOX genes (HOXA9-13 and HOXD9-13) were FLAG-tagged and virally expressed in chicken mesenchymal limb-derived cells enabling comparison of their binding in an identical setup and controlled conditions. HOX binding profiles uncovered two direct paralogues (i.e. HOXA10 and D10) bind more often same regions than two non-paralogues (i.e. HOXA9 and A13) reminiscent of in vivo functional data. Moreover, the here described in vivo HOX-DNA binding differs from in vitro binding, indicating the importance of cofactors and biological context for HOX binding and functional outcome. Additionally, binding comparison uncovered previously unknown differences between binding modes of HOX-TFs that at least partially rely on the abundance of direct binding and co-binding with other TFs. Finally, with combination of genetics, genomics and biochemistry a novel HOX cofactor, CCCTC binding factor (CTCF), was discovered suggesting potential interplay between HOX target regulation and chromatin architecture.

Mutationen von Transkriptionsfaktoren (TF) betreffen nicht nur die Funktion des TFs, sondern auch die Expression seiner Zielgene und liegen häufig angeborenen Entwicklungsdefekten zugrunde. Über 20 Mutationen in HOXD13, einem TF der die Entwicklung der Extremitäten kontrolliert, sind bisher als Ursache verschiedenartiger Extremitätenfehlbildungen entdeckt worden. Eine molekularbiologische Grundlage für die Vielgestaltigkeit der HOXD13-Mutationen ist jedoch unbekannt. Die bisherigen Methoden zur funktionellen Charakterisierung von TF-Mutationen ermöglichten eine lediglich eingeschränkte Interpretation der molekularen Pathomechanismen. Die kürzlich entwickelte ChIP-seq Methode ermöglicht eine umfassende, funktionelle Charakterisierung eines TFs. In dieser Arbeit wurde eine Methode etabliert, um eine Vielzahl von Transkriptionsfaktoren und TF-Mutationen systematisch zu untersuchen. Zur Validierung wurden zwei neue Punktmutationen in HOXD13, p.Q317K und p.R298Q, charakterisiert. Beide Mutationen betreffen die DNA-bindende Domäne von HOXD13, rufen aber stark unterschiedliche Fehlbildungen hervor. Die Ergebnisse zeigen, dass die HOXD13Q317K Mutante eine veränderte Sequenzspezifität aufweist, welche nun jener eines anderen TFs, PITX1, ähnelt. Auch genomweit zeigt HOXD13Q317K ein Bindungsprofil, welches eher PITX1 als HOXD13wt entspricht. Durch weitere, unabhängige Analysen und Experimente wurde bestätigt, dass die p.Q317K Mutation HOXD13 in einen TF mit PITX1-ähnlichen Eigenschaften verändert. Die HOXD13R298Q-Mutante zeigt eine weitgehend unveränderte Bindungssequenz gegenüber HOXD13wt, jedoch eine veränderte Zusammensetzung der genomischen Bindestellen. Dies weist, in Kombination mit dem humanen Phänotyp auf einen dominant-negativen Pathomechanismus dieser Mutanten hin. Zusammengenommen zeigt diese Arbeit durch die Erhebung von experimentellen Daten, dass klar unterscheidbare molekularbiologische Mechanismen den HOXD13Q317K- und HOXD13R298Q-Mutationen zugrunde liegen. / Mutations in transcription factors (TF) do not only affect the function of the TF, but also the expression of its target genes and are frequently underlying congenital malformations. More than 20 distinct pathogenic mutations in HOXD13, a TF controlling limb development, have been associated with a broad range of limb malformations. However, a molecular basis underlying the variability of HOXD13-associated phenotypes remains elusive. To date, the experimental methods used to functionally characters TF mutations have allowed only limited insights into the underlying molecular pathomechanisms. The recently developed ChIP-seq technology has proven to be a powerful method to profile the binding characteristics of TFs; however a number of technical hurdles hinder its application for functional characterization of mutant TFs. This work describes the establishment of a ChIP-seq approach to investigate a wide spectrum of TFs and TF mutations. The approach was applied to characterize two previously unknown missense mutations in HOXD13, p.Q317K and p.R298Q, which both alter the DNA-binding domain of HOXD13 but cause very different disease phenotypes. The results show that the HOXD13Q317K mutant has an altered sequence specificity that resembles the recognition sequence of another TF, PITX1. Further, the genome-wide binding pattern of HOXD13Q317K shifts towards a more PITX1-like binding pattern. Even further analysis and viral overexpression in chicken limb buds confirm that the mutation partially converts HOXD13Q317K into a TF with PITX1-like properties. The HOXD13R298Q has a largely unchanged sequence specificity, but an altered composition of genomic binding sites. This, in combination with the human phenotype, indicates that the mutant might act in a dominant-negative manner. Collectively, this work shows through generation of direct experimental evidence, that clearly distinct molecular mechanisms underlie the pathogenicity of HOXD13Q317K and HOXD13R298Q mutations.

Hypoxia regulates many hundreds of genes with important roles in ischemic and neoplastic disorders. Central to this response are the hypoxia inducible transcription factors (HIF). This work aimed to better understand the direct transcriptional response to HIF by mapping HIF-binding sites across the genome using chromatin immunoprecipitation coupled to high-throughput sequencing (ChIP-seq). ChIP-seq for HIF in MCF-7 breast cancer cells under hypoxic conditions revealed more than 400 high-stringency HIF-binding sites genome-wide. Each member of the HIF heterodimer was present with near complete concordance. Binding of the two principle isoforms revealed a high degree of overlap with no differences in the DNA-binding motif. HIF-binding was associated with upregulation, but not downregulation of genes indicating that it functions as a transcriptional activator but not as a repressor. HIF-binding occurred preferentially at gene promoters, but was also present at promoter-distant sites, which were also associated with gene regulation, implicating long-range interactions in hypoxic gene activation. HIF-binding was associated with markers of open chromatin and active enhancers that were present in normoxia, indicating that HIF-binding sites are already “prepared” to bind HIF before the hypoxic stimulus. Analysis of normoxic and hypoxic RNA pol2 and H3K4me3 signals revealed distinctive hypoxia-inducible changes unique to HIF-binding genes. Comparable numbers of HIF-binding sites were observed in a second cell line (von Hippel-Lindau defective 786-O renal cancer cells) as in MCF-7 breast cancer cells, although approximately 65% were unique to 786-O cells. These unique sites were more frequently promoter-distant. Correlation with expression analyses from renal tumours indicated that many HIF-binding genes were upregulated in renal cancer. One such RCC unique promoter-distant HIF-binding site was identified at an intergenic locus on chromosome 11q13.3 that has been associated with renal cancer in Genome-Wide Association Studies. The HIF-binding site was in high linkage disequilibrium with the disease associated SNP and had the epigenetic hallmarks of an enhancer. Analysis of pan-genomic expression analyses identified the cell-cycle regulator cyclin D1 as highly HIF-regulated, and a physical association between the HIF-binding site and the CCND1 promoter could be determined. Furthermore, in a renal cancer cell line heterozygous at this locus, the RCC-protective allele disrupted HIF-binding leading to an allelic imbalance in cyclin D1 expression.

Thèse décrivant l'écriture d'outils spécialisés facilitant l'analyse de grandes quantités de données provenant de technologie de séquencage haut débit.

RNA helicases are found in all kingdoms of life, functioning in all aspects of RNA biology mainly through modulating structures of RNA and ribonucleoprotein (RNP) complex. RNA structures have fundamental impacts on steps in gene expression, including transcription, pre-mRNA processing, and translation. However, the precise roles and regulatory mechanisms of RNA structures in co- and post-transcriptional processes remain elusive. By probing genome-wide RNA structures in vivo, a recent study suggested that ATP-dependent factors, such as RNA helicases, maintain the actively unfolded state of RNAs. Among all RNA helicases, DEAD-box proteins form the largest family in eukaryotes, and have been shown to remodel RNA/RNP structures both in vitro and in vivo. Nevertheless, for the majority of these enzymes, it is largely unclear what RNAs are targeted and where they modulate RNA/RNP structures to regulate co-transcriptional processes. To fill the gap, my research focused on identification of the RNAs and structures targeted by the DEAD-box protein Dbp2 in S. cerevisiae to uncover the cellular processes that Dbp2 is involved in.<br><div><div>My studies revealed a role of Dbp2 in transcriptional termination. Dbp2 binds to ~34% of yeast mRNAs and all snoRNAs, and loss of DBP2 leads to a termination defect as evidenced by RNA polymerase II (RNAPII) accumulation at 3’ ends of these genes. In addition, the binding pattern of Dbp2 in mRNAs is highly similar to Nrd1 and Nab3 in the Nrd1-Nab3-Sen1 (NNS) termination complex, and deletion of DBP2 leads to reduced recruitment of Nrd1 to its target genomic loci. In Dbp2 and NNS targeted 3’ UTRs, RNA structural changes resulted from DBP2 deletion also overlap polyadenylation elements and correlate with inefficient termination, and loss of stable structure in the 3’ UTR bypasses the requirement for Dbp2. These findings lead to a model that Dbp2 promotes efficient termination of transcription through RNA structure remodeling.</div><div>Interestingly, my research also revealed the requirement of DBP2 for efficient splicing, as loss of DBP2 leads to accumulation of unspliced pre-mRNAs. Moreover, this function is dependent on the helicase activity of Dbp2. Further studies are needed to characterize the molecular mechanism of how Dbp2 facilitates splicing in cells. Overall, my research demonstrated that DEAD-box RNA helicases remodel mRNA structure in vivo and that structural alteration can be essential for proper gene expression.</div></div>

Regulation of gene expression is an essential process for all living organisms. Transcriptional regulation, associated with chromatin, is governed by: (1) DNA sequence, which creates regulatory sites (promoters, enhancers and silencers), where sequence motifs and features (e. g. CpG) can attract transcription factors (TFs) and influence chromatin structure or RNA polymerase II (Pol II) binding, initiation and elongation; (2) non-sequence, epigenetic factors - histone modifications, TF binding, chromatin remodelling (histone placement, eviction and reconstitution), and non-coding RNA regulation. These factors interact with each other, creating a complex network of interactions. In this thesis I describe computational studies of heterochromatin factors in regulation of gene and repeat expression, an analysis of active regulatory elements, and global analyses of big datasets in C. elegans. I first show that a team of heterochromatin factors - HPL-2/HP1, LIN-13, LIN-61, LET-418/Mi-2, and H3K9me2 histone methyltransferase MET-2/SETDB1 - collaborates with piRNA and nuclear RNAi pathways to silence repetitive elements and protect the germline. I also found that the TACBGTA motif is particularly enriched on repeats and heterochromatin factors binding sites, and that repeat elements are derepressed in the soma during normal C. elegans ageing. I then describe the work on active regulatory regions. I show that CFP-1/CXXC1 binds CpG dense, nucleosome depleted promoters and, along SET-2, is required for H3K4me3 deposition at these loci. Using expression profiling I determined that the majority of CFP-1 binding targets are not significantly mis-regulated in cfp-1 mutants, but are weakly upregulated in bulk analyses. I also show that CFP-1 functionally interacts with the Sin3S/HDAC complex. In cfp-1 mutant I observed both loss and gain of SIN-3 binding, depending on chromatin context. Finally, I performed a data driven study on a large collection of ChIP-seq profiles using non-parametric sparse factor analyses (NSFA) and compared it to other, unsupervised machine learning algorithms. This study uncovered interactions and structure in genomic datasets. In addition, I present a collection of computational tools and methods I developed to facilitate processing, storage, retrieval, annotation, and analyses of large datasets in genomics.