Our laboratory is interested in epigenetic reprogramming in mammalian development and ageing. We previously discovered some of the mechanisms that lead to genome-wide erasure of DNA methylation in mammalian embryos and primordial germ cells. We continue work on this topic but have also recently started to think about potential links between reprogramming and zygotic genome activation (ZGA). ZGA is the major transcriptional activation of the embryonic genome which occurs at the 2-cell stage in mouse and at the 4-8 cell stage in human embryos.
We carried out a Crispr activation screen with single cell transcriptional read-out in mouse embryonic stem cells (ESCs) in order to identify upstream regulators of ZGA, and identified amongst others Dppa2, Smarca5, and Patz1 as candidate regulators. We are currently investigating the putative roles in ZGA of these factors in vivo in mouse embryos, including by a protein knockdown method called TRIM-away.
We have developed a number of single cell multi-omics sequencing technologies including scNMT-seq which reads out the transcriptome, methylome, and chromatin accessibility from the same single cell. Most recently, we have combined high throughput spatial with single cell transcriptomics to elucidate mouse development. Using single cell multi-omics and advanced computational techniques, we examined the earliest cell fate decisions underlying gastrulation in the mouse, by which the three primary germ layers (endoderm, mesoderm, ectoderm) are established.
We found that the three lineages are primarily characterised by enhancer epigenetic dynamics, in that hundreds of ectoderm enhancers are hypomethylated and their chromatin accessible in the ectoderm but not in the other two lineages. We were however surprised to find that how the three lineages get to this pattern in time is very different between them.
Ectoderm enhancers were already hypomethylated and accessible in the early epiblast (where the linked genes are not expressed) whilst endoderm and mesoderm enhancers were methylated and inaccessible at this stage. They then become acutely demethylated and accessible as cells migrate through the primitive streak and are exposed to a new signalling environment which results in their fate acquisition.
Hence ectoderm (but not mesoderm or endoderm) enhancers are ‘primed’ in early development for their later action. This may explain at least in part the so-called default hypothesis of neuroectoderm differentiation whereby embryonic stem cells released from pluripotency spontaneously differentiated into neurectoderm (but not mesoderm or endoderm).
Priming of enhancers may be an important step for their eventual activation which is needed for expression of linked genes. We are currently aiming to identify proteins which bind to enhancer DNA and instigate priming. We have already found two proteins, Dppa2 and Dppa4, that bind to bivalent promoters of developmental genes and are needed to prime these genes in pluripotent stem cells for their future transcriptional activation during embryogenesis.
Epigenetic information degrades during ageing of an organism, potentially in a programmed fashion. The DNA methylation ageing clock for example measures changes of methylation with age and from this predicts age with high precision. Excitingly, the methylation clock is reset to 0 when somatic cells are reprogrammed to induced pluripotent stem cells (iPSCs).
We have asked if partially reprogramming human fibroblasts might rejuvenate cells from older individuals. Reprogramming 60 or so year old fibroblasts to the so-called maturation phase followed by switching off of reprogramming factors allowed the cells to return to their original fibroblast identity but these were now measuring 25-30 years old by their transcriptome, methylome, and expression of functionally important collagen proteins.
This suggests the possibility of rejuvenating various human cell types for cell based therapies, and moreover the potential of dissociating pluripotency induction from putative rejuvenation networks which in the future may be safer to apply in therapeutic settings.