Thèse Exploration du Rôle des Éléments Transposables en Tant Qu'Enhancers dans l'Induction d'Un Contexte Pré-Leucémique dans l'Hématopoïèse Clonale de Potentiel Indéterminé H/F - 75

Établissement : Université Paris-Saclay GS Life Sciences and Health
École doctorale : Cancérologie : Biologie - Médecine - Santé
Laboratoire de recherche : Cellules souches hématopoïétiques et développement des hémopathies myéloïdes
Direction de la thèse : Emilie ELVIRA-MATELOT ORCID 0000000240176436
Début de la thèse : 2026-10-01
Date limite de candidature : 2026-04-17T23:59:59

L'hématopoïèse clonale de potentiel indéterminé (CHIP) est associée à un risque accru de leucémie myéloïde, telle que les syndromes myélodysplasiques (SMD), et est donc considérée comme un état préleucémique.

La CHIP est principalement induite par des mutations affectant des régulateurs épigénétiques tels que TET2. Les mutations de TET2 entraînent l'expansion des cellules souches hématopoïétiques (CSH) ainsi que du compartiment myéloïde (biais myéloïde), deux caractéristiques d'un état préleucémique chez les individus présentant une CHIP.

La réorganisation de la marque d'hétérochromatine H3K9me3, qui joue un rôle majeur dans la répression des éléments transposables (TE), est associée à l'induction d'un biais myéloïde. Les TE jouent un rôle crucial dans les réseaux de régulation génique.

En nous basant sur des données préliminaires obtenues dans les CSH murines, nous émettons l'hypothèse que les TE, via la perte d'hétérochromatine et l'ouverture de la chromatine, agissent comme des enhancers afin de promouvoir l'état préleucémique, caractérisé par une augmentation de l'auto-renouvellement et un biais myéloïde, observé lors de la perte de fonction de TET2 dans la CHIP.

Afin d'évaluer le rôle des TE en tant qu'enhancers dans le CHIP, nous utiliserons des approches multi-omiques pour étudier l'ouverture de la chromatine, la méthylation de l'ADN et différentes marques d'histones au niveau des TE : 1- dans des cellules souches et progénitrices hématopoïétiques (HSPC) avec extinction de TET2 par shRNA comparées à des contrôles, en présence ou en absence d'inflammation afin de mimer l'« inflammaging », et 2- dans des HSPC mutées ou non pour TET2 issues d'individus présentant une CHIP. Nous validerons ensuite l'impact de ces TE-enhancers sur le transcriptome et la fonction des HSPC en les délétant ou en les réprimant par CRISPR-(d)Cas9.
Enfin, par la réanalyse de données issues de la littérature obtenues à partir de cellules souches de SMD, nous évaluerons si les TE-enhancers identifiés dans la CHIP pourraient constituer une signature de la leucémogenèse.

Ce projet pourrait, à terme, révéler de nouvelles stratégies permettant de prédire ou de prévenir la leucémie myéloïde chez les individus atteints de CHIP.

Clonal hematopoiesis of indeterminate potential (CHIP) is characterized by the expansion of hematopoietic stem cells (HSCs) with somatic mutations in genes commonly associated with myeloid leukemia, in the absence of hematological disease. CHIP prevalence increases with age, in a context of low-grade inflammation1. CHIP cases are associated with a 10-fold increased risk of leukemia and is thus considered as a preleukemic state. CHIP may evolve to a low-risk myelodysplastic syndrome (LR-MDS). Some of these MDS can then progress towards an high-risk (HR-) MDS and to secondary acute myeloid leukemia (sAML)2, making this disease evolution an ideal model to study the mechanisms of leukemogenesis.
The mutations found in epigenetic factors such as TET2 are among the most frequently occurring mutations in CHIP. TET2 mutation induces the expansion of HSCs and of the myeloid compartment (myeloid bias), both being characteristics of a pre-leukemic state, in individuals with CHIP and mouse models3. Myeloid cells produced by Tet2-mutated HSCs also present enhanced inflammatory properties. This has been linked to the higher risk of developing aged-related diseases4. Inflammation establishes a feedback that selectively further expands mutant HSCs, and also induces the generation of a novel population of myeloid cells in Tet2-mutated mice.
Normal HSCs are intrinsically heterogeneous and, notably, can display skewed differentiation towards specific lineages5. This is correlated with a particular epigenetic state, specifically at enhancers of genes involved in differentiation, that precedes transcription factor (TF) binding and the induction of the corresponding lineage transcriptomic program5.
However, the link between epigenetic changes in HSCs upon TET2 loss of function, the increased HSC self-renewal and the myeloid bias observed is poorly characterized. Deciphering these underlying mechanisms is essential for predicting and preventing aged-related diseases such as myeloid leukemia4.
Loss of the trimethylation of lysine 9 of histone H3 (H3K9me3) repressive mark has been associated with a myeloid bias in human and mouse HSCs6. Heterochromatin, through H3K9me3 and DNA methylation, plays a major role in repressing transposable elements (TEs). TEs, such as long terminal repeats (LTRs) and long interspersed elements (LINE-1/L1), are repetitive sequences that represent almost half of the human and mouse genomes. They have recently been recognized as major contributors of gene regulatory networks. Loss of epigenetic control of pre-existing TE copies can affect gene expression by providing alternative promoters, splicing or polyadenylation signals. TEs are also reservoirs of binding sites for TFs, serving as cis-regulatory elements in a cell- and tissue-specific fashion. TE members of the same subfamily, scattered throughout the genome and enriched for the same motifs for TF binding sites, may regulate different genes involved in the same pathway depending on chromatin accessibility and TF availability7. Even though the most recent subfamilies of TEs are the most epigenetically regulated in the genome, they are generally overlooked in omics data as they present a very low mappability due to quasi-identical copies. We recently showed that H3K9me3 is mainly enriched at L1Md and IAP, the youngest subfamilies of L1 and LTR in mouse. We also showed that H3K9me3 enrichment specifically at intronic L1Md harboring NF-kB binding site motifs is crucial for HSC gene expression and self-renewal capacity8. Interestingly, TET2 was recently shown to regulate chromatin state at TEs in HSPCs and H3K9me3 at L1Md in embryonic stem cells9,10.
Our preliminary data obtained in mouse HSCs show that upon TET2 loss, L1Md and IAP, enriched for binding motifs for TFs involved in HSC self-renewal and myeloid differentiation, such as MEIS1, SPI1, GATA, RUNX, lose H3K9me3, gain H3K4me1 and H3K27ac enhancer marks, and gain chromatin accessibility, suggesting that these TEs might play the role of enhancers in inducing a pre-leukemic state. Even if different in terms of nature, sequence and localization in the genome, the epigenetic mechanisms controlling TEs and the consequences of their deregulation on cell transcriptome are conserved between human and mouse11, suggesting that TEs could also serve as enhancers in inducing a pre-leukemic state in human HSPCs.

By studying histone marks, DNA methylation, chromatin accessibility and TF motifs at TEs, and performing functional analysis in human HSPC knocked-down for TET2, or collected from CHIP individuals, we aim to test the hypothesis that poorly studied TE sequences may be a significant source of enhancers in inducing an altered stemness and myeloid program in CHIP and leukemogenesis.

We aim to investigate:
1-the role of TEs as enhancers in inducing a preleukemic state in TET2 deficient HSPC
2-the role of TEs as enhancers in inducing a preleukemic state in TET2 mutated CHIP
3-The possibility that TE-enhancer could be used as a signature for leukemogenesis

Aim1. TEs as enhancers in inducing a preleukemic state in TET2 deficient HSPCs
To assess the potential role of TEs as enhancers in inducing a preleukemic state in TET2 deficient HSPCs, we will knock-down TET2 (TET2-KD) by transducing an shRNA directed against TET212 in CD34+ cells collected from cord blood , treated or not with LPS to mimic inflammaging. We will then perform H3K9me3, H3K4me1 (poised enhancer), H3K27ac (activated enhancer) CUT&Tag, ATAC-seq (chromatin accessibility) and RNA-seq experiments. We will also localize DNA methylation changes at the most recent subfamilies of TEs by long-read DNA sequencing using nanopore technology (precise mapping of repeated reads originating from TEs).
Copies of TEs losing heterochromatin, gaining chromatin accessibility, marks of (active) enhancers, and enriched for myeloid TF motifs (TE-enhancers) in TET2 KD vs control HSPCs will be selected for further analysis.
Next, we will decipher if TE-enhancers regulate the transcription of genes involved in HSC self-renewal or myeloid differentiation by integrating these results with differential gene expression (RNA-seq in TET2KD HSPCs ± LPS).
Finally, we will validate the impact of the selected TE-enhancers on the transcriptome of HSPCs by 1- deleting candidate TE-enhancers using Cas9/gRNA RNP complexes into TET2-KD HSPCs, as previously performed8; 2- restoring heterochromatin at selected TF motifs in TE-enhancers by transducing a CRISPR-dCas9 lentiviral vector fused to either KRAB (H3K9me3 recruitment) or DNMT3A (DNA methylation) and guide RNAs targeting selected TF motifs13. We will assess HSPC transcriptomes and differentiation through RNA-seq, in vitro liquid culture and methylcelllulose assays.
Aim2. TEs as enhancers in inducing a preleukemic state in TET2 mutated CHIP
In order to test the role of TEs as enhancers in inducing a pre-leukemic state in HSCs in CHIP, we will work on BM from TET2 mutated CHIP individuals vs controls (BoHeme study- Coll Borhane Guezguez, Mainz Germany) and perform single cell (sc) long-read multiome (RNA-seq + ATAC-seq)) to compare chromatin accessibility and gene expression in mutated vs non-mutated cells in the same individuals (technics currently under optimization at the GR genomic platform- N. Droin). We will pay a particular attention to the TE-enhancer identified above, and check if their derepression (gain in chromatin accessibility) is associated with the deregulation of neighboring genes involved in HSC self-renewal and/or myeloid differentiation in HSCs and/or their progeny (RNA-seq).

Aim3. TE-enhancer as a signature for leukemogenesis?
MDS stem cell acquire a precocious myeloid-like chromatin accessibility signature, which can be used to stratify disease status and predict disease prognosis15.
In order to assess if TE-enhancers found in CHIP could be a signature of leukemogenesis, we will reanalyze ATAC-seq data obtained in CD34+CD38- and CD34+CD38+ stem and progenitor cells from MDS patients vs controls15, keeping multiple mapping reads originating from TEs, and will compare TE-enhancers in CHIP (aim1-2) to TE-enhancers in MDS. By correlating the TE-enhancer-MDS signature to disease status (HR vs LR-MDS) and prognosis, we will then check if chromatin accessibility at TEs could improve the stratitification and prognosis of the disease.