Thèse Epigénétique et Organisation 3D du Génome Mécanismes de Contrôle des Gènes Soumis à l'Empreinte en Conditions Physiologiques et Pathologiques H/F - 75

Établissement : Université Paris-Saclay GS Life Sciences and Health
École doctorale : Structure et Dynamique des Systèmes Vivants
Laboratoire de recherche : I2BC - Institut de Biologie Intégrative de la Cellule
Direction de la thèse : Benoit MOINDROT ORCID 0000000339157892
Début de la thèse : 2026-10-01
Date limite de candidature : 2026-03-23T23:59:59

L'empreinte génomique est un mécanisme fondamental de régulation des gènes chez les mammifères, assurant l'expression allélique d'environ 150 gènes chez la souris et l'homme. La perturbation de l'expression des gènes soumis à l'empreinte entraîne des troubles du développement, tels que le syndrome de Silver-Russell (SRS) et le syndrome de Beckwith-Wiedemann (BWS).

La régulation des gènes soumis à l'empreinte repose sur l'intégration de plusieurs mécanismes de régulation épigénétique, notamment la méthylation de l'ADN, les ARN non-codants, les modifications d'histones, la fixation de la protéine architecturale CTCF et/ou une organisation tridimensionnelle spécifique de la chromatine. Comprendre comment ces différents mécanismes interagissent pour contrôler l'expression allélique de ces gènes, et comment ils sont perturbés dans les maladies, reste un défi majeur.

Le directeur de thèse, membre de l'équipe Noordermeer à l'I2BC, étudie ces mécanismes en utilisant des approches génomiques de pointe aux locus Dlk1-Dio3 et Igf2-H19, deux domaines régulés par l'empreinte génomique conservés chez la souris et l'homme. Les travaux précédents et en cours de l'équipe ont montré que :
1- Ces deux domaines présentent une organisation 3D de la chromatine distincte sur chaque allèle, avec des fonctions régulatrices ;
2- Cette organisation 3D est contrôlée par l'empreinte et est altérée chez les patients présentant une perte d'empreinte ;
3- Les deux allèles du locus Dlk1-Dio3 présentent des paysages épigénétique distincts (modification d'histone, association aux lamines nucléaires) et un programme de réplication différent ;
4- Le long ARN non codant Meg3, retenu en cis, contribue à la régulation des gènes imprimés.

S'appuyant sur ces résultats, sur des approches épigénomiques alléliques, et sur des collaborations étroites avec des équipes de recherche clinique et fondamentale, ce projet de thèse s'articule autour de deux grandes questions auxquelles il cherche à répondre :

(a) Quels sont les déterminants régulant en cis le locus Dlk1-Dio3 lors de la différenciation cellulaire ? [60 % du temps]
En utilisant des cellules de souris, le doctorant identifiera les segments génomiques interagissant avec le l'ARN Meg3 et déterminera les facteurs moléculaires contrôlant sa localisation. Des approches ciblées et/ou des cribles permettront d'identifier les partenaires protéiques assurant l'activité régulatrice de l'ARN Meg3 lors de la différenciation. Le paysage épigénétique du locus Dlk1-Dio3 sera aussi cartographié dans des modèles cellulaires différenciés afin d'identifier des éléments régulateurs spécifiques de certains tissus et dont la fonction sera validée par des perturbations (CRISPR-Cas9 ou CRISPRi).

(b) Comment les programmes régulateurs des gènes soumis à l'empreinte sont-ils réorganisés chez les patients présentant une pathologie de l'empreinte ? [40 % du temps]
À partir de cellules dérivées de patients (SRS et BWS), le doctorant évaluera dans quelle mesure la structure allélique de la chromatine du locus IGF2-H19 est reconfigurée chez les patients. Un séquençage Nanopore fournira les profils de méthylation de l'ADN de chaque allèle, offrant une quantification des anomalies dépassant les tests cliniques existants. Les changements dans l'occupation par CTCF et l'organisation 3D de la chromatine seront quantifiés chez certains patients, pour révéler comment les réseaux de régulation y sont perturbés et comment cela explique l'expression aberrantes des gènes.

Ensemble, ce projet de thèse décryptera les mécanismes basés sur la chromatine et les ARN non codants qui contrôlent l'expression allélique des gènes soumis à l'empreinte, et révèlera comment ces mécanismes sont altérés dans les pathologies de l'empreinte. En liant régulation épigénétique fondamentale et approches cliniques, ce projet fournira à une meilleure compréhension mécanistique de la régulation des gènes imprimés, en conditions normales comme pathologiques.

Genomic imprinting is an epigenetic gene regulatory mechanism essential for embryonic development. In mouse and in human, genomic imprinting controls the expression of more than ~150 protein-coding genes and non-coding RNAs (ncRNAs) to ensure their mono-allelic expression, in a parent-of-origin dependent manner. Most imprinted genes are present in clusters, where mono-allelic expression is primarily instructed by regulatory sequences known as imprinted control regions (ICR). ICRs acquire opposing DNA methylation states in the sperm and the oocyte, which are then stably maintained as differentially methylated regions (DMR) through somatic cell divisions (reviewed in Barlow & Bartolomei 2014; Tucci et al. 2019).

The IGF2-H19 and DLK1-DIO3 domains are two evolutionary-conserved imprinted domains. Both contain protein-coding genes expressed from the paternal chromosome, and non-coding RNAs transcribed from the maternal allele (Sanli & Feil 2015). This notably includes Meg3 long non-coding RNA expressed from the maternal Dlk1-Dio3 locus. In both domains, a maternal-specific CTCF binding site overlaps a key regulatory DMR of the locus, and plays a central role in imprinting control (reviewed in Moindrot et al. 2024).

Over the past few years, within the Noordermeer lab, the PhD supervisor has been responsible for the development of allele-resolved genomic assays to interrogate chromatin regulations at imprinted domains with high resolution. Using these approaches, we could show that the mouse Igf2-H19 and Dlk1-Dio3 domains adopt allele-specific 3D chromatin architecture driven by the maternal-specific binding of CTCF protein to a DMR of the locus. We further demonstrated that the allelic 3D structure of the Dlk1-Doi3 locus is functionally important, as its perturbation (either through loss of CTCF binding on the maternal allele, or to ectopic CTCF binding to the paternal allele) leads to aberrant expression of the locus' protein-coding genes (Lleres et al. 2019; Farhadova et al. 2024). Additional work revealed that the paternal and maternal Dlk1-Dio3 alleles exhibit opposing histone modification profiles, distinct associations with the nuclear lamina, and different DNA replication timing programs, all instructed by the underlying DNA methylation imprint (Imaizumi et al., 2026; unpublished data). Lastly, in collaboration, we showed that the maternally-expressed Meg3 long non-coding RNA represses maternal Dlk1 expression (Farhadova et al. 2024), possibly through the cis-recruitment of the Polycomb Repressive Complex 2 (Zhao et al. 2010; Kota et al. 2014; Sanli et al. 2018). Collectively, our work contributed and continues contributing to advance our molecular understanding of the epigenetic layers converging at the Dlk1-Dio3 locus to regulate allelic gene expression programs in cis. However, question remains about the cis-determinants driving the tissue-specific expression of imprinted protein-coding genes at the Dlk1-Dio3 locus, and about how the Meg3 lncRNA, that is retained in cis on the maternal allele, functionally prevent their maternal expression. These pressing questions will be addressed in the proposed PhD project.

The correct expression of imprinted genes is essential for human development and their dysregulation leads to developmental disorders collectively known as imprinting disorders (Eggermann et al. 2023). Silver-Russell Syndrome (SRS) and the Beckwith-Wiedemann Syndrome (BWS) are two imprinted disorders associated with defects at (or near) the imprinted IGF2-H19 domain. Loss of DNA methylation imprint at the IGF2-H19 ICR is the most frequent molecular cause of SRS (Wakeling et al. 2017), whereas gain of methylation at the same region represents one of the etiologies of BWS (Brioude et al. 2018).

Ongoing unpublished work in the Noordermeer lab aims to dissect allele-specific chromatin alterations in cells derived from patients with imprinting disorders. As a first step in this project, we showed that in three SRS patients, the loss of methylation at the H19-ICR reprograms the 3D chromatin architecture of the paternal IGF2-H19 locus, abolishing the allelic asymmetry normally seen in healthy individuals that normally govern imprinted gene regulation. This initial characterization of 3 SRS patients provides a proof of concept for applying allelic multi-omic approaches to comprehensively characterize epigenetics defects at various scales in patient-derived cells.

Building on an established and productive collaboration with clinicians, who are expert in imprinting disorders (Selenou et al., in-prep) and who manage a large cohort of SRS and BWS patients, the PhD supervisor now aims to extend these analyses to larger cohort of SRS patients, include BWS patients, and assess whether the quantitative characterization of chromatin perturbations can refine molecular diagnosis and improve the stratification of patients with SRS/BWS imprinting disorder. Beyond the translational relevance, these characterization of the various (epi-)genetic alterations seen in the SRS/BWS cohort will also deepen our understanding of both normal and pathological regulation at imprinted domains.

The proposed PhD project integrates mechanistic and translational approaches to further elucidate allele-specific gene-regulatory programs operating at the imprinted Dlk1-Dio3 and Igf2-H19 locus, and their perturbation in diseases. It is structured around two complementary sub-projects:

1- Primary Sub-Project: Molecular determinants of cis-regulation at the Dlk1-Dio3 locus [60% time dedicated]
This project aims to further dissect the mechanisms ensuring allelic gene regulation at the Dlk1-Dio3. It will involve (a) the identification and functional validation of cis-regulatory elements (ie., enhancers, insulators, boundary elements...) controlling Dlk1 expression, and (b) the determination of where Meg3 long non-coding RNA locates on chromatin, the factors governing its localization, and the trans-acting partners mediating its regulatory activity. The lab has the expertise in generating the cellular models (including CRISPR-Cas9) and in implementing genomic assays required for this work (HiC, ChIP, Cut&TAG...), with RNA-chromatin interaction methods (ChIRP-seq or RAP-seq) to be established.

2- Secondary Sub-Project: Chromatin alteration in imprinting disorders [40% time dedicated]
Chromatin alterations will be profiled at various scales-spanning DNA methylation, CTCF binding and higher-order 3D chromatin structure-in cells derived from SRS and BWS patients (peripheral blood cells, iPSC, iPSC-derived organoids). This work will be conducted in the context of an established collaboration with the laboratory of I. Netchine, an expert in imprinting disorders diagnosis, which manages a large cohort of SRS and BWS patients. The PhD project will systematically test if chromatin re-organization is a common hallmark of SRS and BWS and assess how regulatory interactions are perturbed in individual patients. All required state-of-the-art genomic assays (Nanopore sequencing, ChIP-seq or Cut&TAG, Hi-C, all with allelic precision) are already established in the Noordermeer Lab and have been validated in a pilot study restricted to 3 SRS patients.

Collectively, the mechanistic dissection of cis-regulatory programs (sub-project #1) and patient-centered epigenomic analyses (sub-project #2) and will provide an integrated view of imprinting control, both in physiological and pathological contexts.

This PhD project uses imprinted domain as a paradigm to investigate how chromatin (3D) structure and regulatory long non-coding RNAs control gene expression. The project focuses on the Igf2-H19 and Dlk1-Dio3 loci, two evolutionarily-conserved imprinted domains, and aims to elucidate the regulatory circuits governing allelic expression in both physiological and pathological contexts.

The PhD project is structured around two sub-projects.

In a primary sub-project [60% time dedicated], mouse cells will be used as experimental model to extend investigation ongoing in the lab at the Dlk1-Dio3 locus, building on a productive and long-lasting collaboration with the Feil Lab at the IGMM in Montpellier (Lleres et al. 2019; Farhadova et al. 2024; Imaizumi et al., 2026). This work will investigate the chromatin localization and regulatory function of Meg3 long non-coding RNA, as well as identify and characterize cis-regulatory elements controlling Dlk1 expression, in a tissue-specific manner. Allele-specific genomic approaches will be applied in both wild-type and perturbed cellular systems. In collaboration with the Kind lab (Hubrecht Institute, The Netherlands), emerging single-cell methods profiling multiple chromatin features (Lochs et al. 2024) may be considered as a high-risk/high-gain extension, to develop an integrated view of the chromatin logic underlying imprinted gene regulation. Specifically, this sub-project aims to answer the following questions:
1- Where does Meg3 long non-coding RNA locate on chromatin, and how does it functionally mediate its regulatory activity?
2- Which cis-regulatory genetic elements (e.g., enhancers, insulators, boundary elements...) control tissue-specific Dlk1 expression, and how do they mediate these cis-regulatory functions?
3- Can single-cell chromatin profiling approaches, where feasible, complement bulk allele-specific analyses to generate an integrated view of chromatin regulation at imprinted loci, in differentiated cellular models?

In a secondary sub-project [40% time dedicated], patient-derived samples will be used to determine how the allele-specific chromatin structure at the imprinted IGF2-H19 locus is altered in patients, and how these changes contribute to aberrant gene expression. Allele-specific genomic analyses will be performed in patient-derived cells obtained through an established collaboration with clinicians at Saint-Antoine and Armand-Trousseau Hospitals (Paris), who manage a large cohort of patients with Silver-Russell (SRS) and Beckwith-Wiedemann (BWS) imprinting disorders. Specifically, this sub-project aims to answer the following questions:
1- To what extent is chromatin structure altered in patients with imprinted disorders, and how do these alterations underly the aberrant gene expression programs observed in patients?
2- Can the characterization of altered chromatin structure in patients improve the diagnosis of imprinting disorders and refine patient stratification, compared to current molecular diagnostic strategies? Can it help resolve frequently encountered diagnostic impasses?

Collectively, this PhD project will further characterize the chromatin- and non-coding RNA-based regulatory mechanisms governing allelic expressions at imprinted domains. It will elucidate how these mechanisms are perturbed in human imprinting disorders using patients-derived samples. This project bridges fundamental epigenetic regulation and translational studies. It is anchored in the framework of both fundamental epigenetic biology and precision medicine. Ultimately, it will advance our mechanistic understanding of imprinted gene regulation, in normal and pathological contexts.

This PhD project will combine different genomic approaches, in various cellular models.

--------------------------------------------------------------
Methods specific to mouse studies (sub-project #1)
--------------------------------------------------------------
* F1-hybrid mouse embryonic stem cells, already established in the lab, will be used for mechanistic studies in mouse. The functional validation of cis-regulatory elements and trans-acting Meg3 partners will be performed using CRISPR-Cas9 mediated genome editing or CRISPR interference (CRISPRi).

* To identify the molecular trans-acting partners mediating Meg3 regulatory activity, a candidate-based approach will be implemented. Candidate factors will be selected by intersecting a curated list of direct Meg3 binders recently generated in the Feil lab (IGMM, Montpellier, a long-term collaborator [Lleres et al. 2019; Farhadova et al. 2024; Imaizumi et al. 2026]) with published datasets of RNA-binding proteins implicated in chromatin regulation (e.g., from Guo et al. 2024; Trotman et al. 2025). If necessary, a genetic screen will be considered, using a reporter cell line enabling allele-specific monitoring of Dlk1 (i.e., possibly analogous to Aronson et al. 2021). Selected candidate will be functionally validated using perturbation experiments (CRISPR-Cas9).

* To map the cis-determinants controlling allelic expression at the Dlk1-Dio3 locus, epi-genenomic profiling (histone modification, chromatin accessibility...) will be performed (see below how) in in-vitro differentiated cellular models in which paternal Dlk1 is actively expressed. These datasets will be used to identify potential enhancers, insulators, boundary elements... If required, complementary analyses will be extended to mouses tissues (preferably in F1-hybrid animals, or alternatively from inbred lines).

* To determine Meg3 long non-coding RNA interaction with chromatin, CHIRP-seq (Chu et al. 2011) or RAP-seq (Engreitz et al. 2013) will be set-up and used.

* Correct DNA methylation levels at the key DMRs of the locus will be monitored by Bisulfite-cloning-sequencing, or by qPCR quantification after digestion by methylation sensitive enzymes (e.g., as in Farhadova et al. 2024; Imaizumi et al. 2026). If required for the project, high-throughput assays will be considered (i.e., RBBS, EM-seq, Nanopore Sequencing...)

--------------------------------------------------------------
Methods specific to human studies:
--------------------------------------------------------------
* Patient-derived cellular samples: A collaboration is already in place with the laboratory of I. Netchine and F. Brioude (Centre de Recherche Saint-Antoine, Trousseau Hospital; both hospital practitioners), who manage a very large cohort of SRS and BWS patients (> 100 patients). In this PhD project, they ensure our access the patient-derived cells and cellular models. Specifically, Peripheral Blood Mononuclear Cells (PBMCs) are routinely isolated from SRS/BWS blood in their lab; they have previously generated induced Pluripotent Stem Cells (iPSC) from SRS patients and master their differentiation into clinically-relevant organoids (Selenou et al., in-prep); they comply with the ethical and legal requirements for handling and sharing patient samples. Transmission of cellular samples between their lab and ours has already been done. For the PhD project, PBMC, iPSC and in-vitro differentiated iPSCs from patients and appropriate controls will be used.

* DNA methylation defects will be quantified by Nanopore-Sequencing using Dorado basecaller and the latest basecalling models. Adaptive sampling will be used to ensure sufficient sequencing depth at imprinted domains. Genetic variants will be identified using Clair3 (Zheng et al. 2022) and phased using LongPhase (Lin et al. 2022). DNA methylation information will be extracted using MethylArtist and modkit tools.

--------------------------------------------------------------
Methods for both human and mouse studies:
--------------------------------------------------------------
* To determine the 3D structure of the chromatin in human and mouse samples, Region Capture Hi-C assay will be used (Arima kit; Target Enrichment using bespoke probes synthetized by Agilent; Hi-C pro tools for the bioinformatic analyses). The strain-specific variants (mouse studies), and the one determined by Nanopore Sequencing (human studies), will be used to generate allele-specific 3D structure maps (whenever possible in human individuals, depending on SNP density). These methodologies are already in place in the Noordermeer Laboratory (Farhadova et al. 2024; Imaizumi et al. 2026; unpublished data).

* To determine other aspect of chromatin organization (CTCF binding, histone modifications...), ChIP-seq and/or Cut&TAG will be used. Both methodologies are established in the Noordermeer lab (Chang et al. 2023; unpublished). A commercial kit will be used for the study of chromatin accessibility (i.e., ATAC-seq). Targeted-sequencing strategy (already established, Imaizumi et al. 2026) will be implemented to ensure sufficient coverage at discriminative SNP and generate reliable allele-specific epigenomic tracks at sufficient resolution (i.e., up to 1kb).

* Changes in gene expression will be determined by RT-qPCR, and allelism determined by Sanger sequencing of the amplified qPCR product. Whenever pertinent, RNA-seq will be used. The team has experience RNA-seq.

* All bioinformatic analyses will be done using the I2BC computational cluster, in close interaction with the bioinformatician engineer of the team. Most of the bioinformatic tools and analyses are already established in the team. We are organizing an HDS-compliant solutions to host genomic data from SRS/BWS patients and healthy donors.