Periodic Reporting for period 2 - EpiCellLineage (Epigenetic regulation of cell fate during early mammalian development)
Reporting period: 2022-07-01 to 2023-12-31
Objective 1: A targeting principle for bivalent domains and multi-lineage commitment
This objective addresses the targeting of bivalency to promoters of developmental genes by Dppa2,4 and the consequences for lineage commitment when bivalency is lost. It will address the sequence of events that leads to loss of H3K4me3 and H3K27me3, loss of chromatin accessibility, gain of DNA methylation, and ineffective gene activation and multi-lineage differentiation in Dppa2,4 knockouts. We will also investigate the in vivo consequences for gastrulation and organogenesis of lack of bivalency.
Objective 2: An epigenetic priming system for neuroectoderm enhancers
This objective explores the factors that cause priming of neuroectoderm enhancers in the early epiblast. DNA binding proteins that prime these enhancers will be identified, and their roles in neural development will be determined. We will also address the role of TET proteins in enhancer hypomethylation, and potentially in their function.
Objective 3: In vivo manipulation of lineage- or organ-defining enhancer and promoter systems
In this objective we will be combinatorially editing 100s of enhancers in an in vitro differentiation system, combined with single cell transcriptional read-out to link enhancer combinations to transcription and differentiation outcomes. Following this screening phase, we will edit informative combinations of enhancers and promoters in vivo to determine their impact on cell fate commitment. This will for the first time connect regulatory element activity at scale to developmental phenotypes and outcomes.
This objective addresses the targeting of bivalency to promoters of developmental genes by Dppa2,4 and the consequences for lineage commitment when bivalency is lost. It will address the sequence of events that leads to loss of H3K4me3 and H3K27me3, loss of chromatin accessibility, gain of DNA methylation, and ineffective gene activation and multi-lineage differentiation in Dppa2,4 knockouts. We will also investigate the in vivo consequences for gastrulation and organogenesis of lack of bivalency.
Objective 2: An epigenetic priming system for neuroectoderm enhancers
This objective explores the factors that cause priming of neuroectoderm enhancers in the early epiblast. DNA binding proteins that prime these enhancers will be identified, and their roles in neural development will be determined. We will also address the role of TET proteins in enhancer hypomethylation, and potentially in their function.
Objective 3: In vivo manipulation of lineage- or organ-defining enhancer and promoter systems
In this objective we will be combinatorially editing 100s of enhancers in an in vitro differentiation system, combined with single cell transcriptional read-out to link enhancer combinations to transcription and differentiation outcomes. Following this screening phase, we will edit informative combinations of enhancers and promoters in vivo to determine their impact on cell fate commitment. This will for the first time connect regulatory element activity at scale to developmental phenotypes and outcomes.
To investigate the role of DPPA2/4 and the consequences of bivalency loss on lineage commitment, we used previously characterised DPPA2/4 KO mouse embryonic stem cells to generate gastruloids. Morphological and protein-level observations show that DPPA2/4 KO are able to specify multiple cell types in this system, however KO gastruloids show a premature elongation, possibly due to a faster proliferation rate. To uncover the mechanism behind this phenotype, we performed 10X single cell RNA sequencing which is currently being analysed by our collaborators in John Marioni’s Lab.
Work package 2
To study priming, primed neuroectoderm enhancers were identified (in E10.5 brain) using ChIP-seq for H3K27ac and H3K4me1. These enhancers are also primed in induced pluripotent stem cells and therefore HiPSci cell lines were used to understand how altering the primary genome sequence changes the priming potential of these enhancers. We identified cases where there is loss of accessibility and loss of H3K27ac at sites where mutations overlap transcription factor binding motifs. We also identified sites where mutations may have altered the transcription factor binding motif causing a de novo site to be generated. The functional outcome of altering these enhancers needs to be validated experimentally to determine the true relevance of these mutations but this analysis provides a subset of mutations and enhancers that may be important for priming potential.
To study the function of enhancer hypomethylation, enhancers active at germ layer specification were identified using H3K27ac ChIP-seq. From this lineage tracing enhancers were identified which have hypomethylated exclusively in the tissues that were derived from that germ layer. These lineage tracing enhancers are active at gastrulation but do not appear to be active at later developmental stages or in adulthood. While most gastrulation enhancers have no discernible methylation memory, using 81 enhancers (2462 cytosines) has the potential to differentiate the germ layer origin of cells from a variety of developmental and postnatal stages. This memory is maintained throughout life and provides the potential to determine germ layer origin of cells. It is lost upon transdifferentiation suggesting that, while it is stable in normal development and ageing, it will be perturbed under specific conditions. This provides a set of regions which can be used to identify shared developmental origins of cells. Future work could identify the role of TET proteins on enhancer hypomethylation at these sites using TET KO cells and how any altered hypomethylation changes cell identity both in development and adulthood.
Work package 3
As a model to perform screens, we established and characterised gastruloids as an in vitro complex system of lineage commitment. We defined intra- and inter- gastruloid heterogeneity and with our collaborators in John Marioni’s lab we developed a deconvolution tool which allows us to discriminate between heterogeneity and perturbation signals (Rosen et al 2022, BiorivxID: 509783). We generated mouse embryonic stem cell clonal lines containing a CRISPRi system and thoroughly characterised their derivative gastruloids using 10X single cell RNA sequencing. We found that cell composition of gastruloids derived from the CRISPRi clonal lines differs from that of WT-derived gastruloids, which was valuable information to design our screen.
We focused on targeting enhancers associated with the gene regulatory network Tbx6-Sox2-Mesp2, which is essential for lineage commitment of neuromesodermal progenitors towards either neural or mesodermal faith. To do so, we used a method developed in the lab (in slico ChIP, Argelaguet et al 2022, BiorivxID:496239) where chromatin accessibility and RNA expression are used to determine new regulatory regions. We built a library of 150 target regions for 50 putative enhancers, created gastruloids and profiled them using 10X single cell RNA sequencing combined with TAP-seq and MULTI-seq. This will allow us to get a single-gastruloid resolution landscape of the impact of enhancer inhibition in transcription factor expression, cell fate, and single-gastruloid cell composition. To our knowledge, this is the first time that TAP-seq and MULTI-seq have been combined which will provide useful insights into how these systems can be used in parallel for in vitro screens.
Work package 2
To study priming, primed neuroectoderm enhancers were identified (in E10.5 brain) using ChIP-seq for H3K27ac and H3K4me1. These enhancers are also primed in induced pluripotent stem cells and therefore HiPSci cell lines were used to understand how altering the primary genome sequence changes the priming potential of these enhancers. We identified cases where there is loss of accessibility and loss of H3K27ac at sites where mutations overlap transcription factor binding motifs. We also identified sites where mutations may have altered the transcription factor binding motif causing a de novo site to be generated. The functional outcome of altering these enhancers needs to be validated experimentally to determine the true relevance of these mutations but this analysis provides a subset of mutations and enhancers that may be important for priming potential.
To study the function of enhancer hypomethylation, enhancers active at germ layer specification were identified using H3K27ac ChIP-seq. From this lineage tracing enhancers were identified which have hypomethylated exclusively in the tissues that were derived from that germ layer. These lineage tracing enhancers are active at gastrulation but do not appear to be active at later developmental stages or in adulthood. While most gastrulation enhancers have no discernible methylation memory, using 81 enhancers (2462 cytosines) has the potential to differentiate the germ layer origin of cells from a variety of developmental and postnatal stages. This memory is maintained throughout life and provides the potential to determine germ layer origin of cells. It is lost upon transdifferentiation suggesting that, while it is stable in normal development and ageing, it will be perturbed under specific conditions. This provides a set of regions which can be used to identify shared developmental origins of cells. Future work could identify the role of TET proteins on enhancer hypomethylation at these sites using TET KO cells and how any altered hypomethylation changes cell identity both in development and adulthood.
Work package 3
As a model to perform screens, we established and characterised gastruloids as an in vitro complex system of lineage commitment. We defined intra- and inter- gastruloid heterogeneity and with our collaborators in John Marioni’s lab we developed a deconvolution tool which allows us to discriminate between heterogeneity and perturbation signals (Rosen et al 2022, BiorivxID: 509783). We generated mouse embryonic stem cell clonal lines containing a CRISPRi system and thoroughly characterised their derivative gastruloids using 10X single cell RNA sequencing. We found that cell composition of gastruloids derived from the CRISPRi clonal lines differs from that of WT-derived gastruloids, which was valuable information to design our screen.
We focused on targeting enhancers associated with the gene regulatory network Tbx6-Sox2-Mesp2, which is essential for lineage commitment of neuromesodermal progenitors towards either neural or mesodermal faith. To do so, we used a method developed in the lab (in slico ChIP, Argelaguet et al 2022, BiorivxID:496239) where chromatin accessibility and RNA expression are used to determine new regulatory regions. We built a library of 150 target regions for 50 putative enhancers, created gastruloids and profiled them using 10X single cell RNA sequencing combined with TAP-seq and MULTI-seq. This will allow us to get a single-gastruloid resolution landscape of the impact of enhancer inhibition in transcription factor expression, cell fate, and single-gastruloid cell composition. To our knowledge, this is the first time that TAP-seq and MULTI-seq have been combined which will provide useful insights into how these systems can be used in parallel for in vitro screens.
We performed 10X single cell RNA sequencing which is currently being analysed to address bivalncey in DPPA2/4
The results of such analysis might provide clues to uncover mechanisms related to the function of DPPA2/4 that have not been reported before.
The results of such analysis might provide clues to uncover mechanisms related to the function of DPPA2/4 that have not been reported before.