Functional study of chromatin factors uncovers strong lineage determining roles and divergent behaviours between normal and malignant haematopoiesis

Abstract

Haematopoiesis relies on the coordinated activities of transcription and chromatin factors (TFs and CFs), which interact to form Genome Regulatory Complexes (GRCs). These dynamic complexes precisely control lineage-specific transcriptional patterns and steer cellular phenotypes. Furthermore, the significance of TFs and CFs is strongly supported by a wealth of evidence from recent studies, which consistently reveal the high recurrence and almost ubiquitous presence of mutations affecting TFs and CFs in haematological malignancies, including acute myeloid leukaemia (AML).

During the last decade, researchers in the field of haematopoiesis have greatly benefited from a comprehensive understanding of normal differentiation roadmaps and the transformation events occurring in malignancies. However, our understanding of CFs contribution to haematopoiesis remains limited. We have yet to determine the main factors (CFs and TFs) within GRCs that govern haematopoietic differentiation, unravel their intricate functional interactions and whether CFs have specific roles or redundantly contribute to lineage determination. Furthermore, we still need to discern which epigenetic mechanisms are disrupted in leukaemia and determine their impact on disease initiation and/or maintenance.

We hypothesised that CFs within GRCs exert distinct roles during haematopoietic lineage determination and that specific CF disruption might contribute to the initiation and/or maintenance of AML.

Firstly, we systematically assessed the lineage-specifying potential of chromatin regulators (CFs and TFs) in murine haematopoietic differentiation trajectories, both ex vivo and in vivo. Ex vivo bulk results unveiled the nuanced and stage-specific roles of GRCs involved in fundamental epigenetic processes during haematopoietic differentiation, including COMPASS methyl-transferases and BAF remodellers. However, a higher level of functional consistency was observed among epigenetic repressor complexes and corepressors. Consistent with the ex vivo effects, in vivo findings at single-cell resolution revealed pronounced lineage-specific trends and functional variability for COMPASS and BAF subcomplexes and suggested that multiple epigenetic repressors, like NuRD, ISWI and N-CoR, help maintain progenitor diversity and balanced lineage distribution by mitigating excessive myelopoiesis.

Our second goal was to unveil the potential interactions amongst TFs and CFs to unravel the composition of key GRCs, shifting the current regulatory paradigm from a single factor to a protein complex-centric view. We examined the dynamic patterns of chromatin accessibility upon TF/CF disruptions and selected representative TF motifs of haematopoietic cell fates to analyse global TF footprints upon every knockout. Disruption of COMPASS and BAF showed downregulation of myeloid TF motif accessibility. In contrast, NuRD and other repressors induced increased accessibility of myeloid TFs associated with inflammatory responses upon knockout.

Lastly, we aimed to dissect the corruption of CFs function in a Npm1c/Flt3-ITD leukaemia model and provide leukaemic-specific vulnerabilities that suggest potential epigenetic therapeutic avenues. scRNA-seq and CITE-seq analyses unveiled leukaemia transcriptomic heterogeneity, with distinct subpopulations exhibiting varying degrees of growth potential. Afterwards, Perturb-seq analysis of the chromatin regulators loss of function across leukaemia clusters confirmed tumour vulnerabilities upon specific perturbations, facilitating the transition of AML-infected cells towards leukaemic differentiated populations with limited cell proliferation and fitness capacities. These observations revealed that leukaemias subvert the function of CFs involved in homeostatic differentiation by aberrantly blocking these processes to sustain their malignant state.