AbstractDuring body axis elongation spinal cord neural tissue in the posterior of the embryo is progressively generated from a pool of bipotent progenitors of the stem zone/caudal lateral epiblast, termed neuromesodermal progenitors. Neural differentiation gene transcription begins in a sharply defined position due to loss of fibroblast growth factor (FGF) signalling. This highly coordinated onset of gene expression implies the involvement of a higher order mechanism to co-ordinately regulate hundreds of genes at the same time. One possible mechanism underlying this phenomenon is the regulation of chromatin structure around neural differentiation genes.
The analysis of chromatin structure by fluorescence in situ hybridisation uncovered that chromatin compaction at the neural differentiation gene Pax6 is maintained by FGF signalling in the stem zone/caudal lateral epiblast of the mouse embryo (Patel et al., 2013). Here I show that this process is mediated by ERK signalling, with just one hour of ERK inhibition leading to a decompaction of the Pax6 locus in the mouse embryo. One way in which local chromatin compaction is regulated is by polycomb repressive complexes. They build a machinery that deposit the histone modifications H3K27me3 and H2AK119Ub at target genes and were shown to impose chromatin compaction. To test the possibility of FGF/ERK regulated polycomb mediated chromatin compaction at neural differentiation gene loci in the stem zone/ caudal lateral epiblast a chromatin immunoprecipitation (ChIP) experiment was performed for the histone modification H3K27me3. Preliminary data showed that the Pax6 transcription start site in the stem zone caudal lateral epiblast is marked by H3K27me3, comparable to the known unexpressed polycomb target HoxD11. This indicated the potential for polycomb mediated chromatin compaction at the Pax6 locus.
A human embryonic stem (ES) cell based in vitro model for the generation of spinal cord neural progenitors used to analyse further whether polycomb mediated chromatin compaction determines neural differentiation gene expression and to evaluate its regulation by FGF/ERK signalling. To that end human ES cells were exposed to FGF and Wnt signals for 3 days to generate neuromesodermal progenitors (hNMPs) (Gouti et al., 2014), these were then differentiated into neural progenitors (hNPs) of posterior spinal cord fate. ChIP assays showed that during the differentiation of hNMPs to hNPs the polycomb components Jarid2 and Ring1B dissociate from the Pax6 locus (transcription start site and gene body) and the H3K27me3 mark is removed in order to facilitate Pax6 transcription from a now decompacted locus. When ERK signalling was blocked in the hNMPs for 12 hours the polycomb components Jarid2 and Ring1B dissociated from the locus whereas the H3K27me3 remained correlating with chromatin decompaction; however, no Pax6 gene transcription was observed. These observations suggest that ERK promotes polycomb mediated chromatin compaction. Interestingly, upon only 3 hours of ERK inhibition only Jarid2 was found to have dissociated from the Pax6 locus whereas Ring1B occupancy and H3K27me3 levels remained, thereby opening the possibility for a step wise dissociation/unloading of the polycomb machinery upon FGF/ERK signalling loss. Taken together the data in the hES in vitro system implies that during neural differentiation the loss of FGF/ERK signalling leads to the dissociation of polycomb neural differentiation gene loci and the removal of the H3K27me3 mark in order to facilitate transcription from these loci.
In pluripotent mouse ES cells it was recently revealed that the recruitment of polycomb complexes to differentiation genes relies on activated ERK (Tee et al., 2014). Furthermore, a few polycomb components were shown to be potential phosphorylation targets of FGF/ERK downstream pathways. The data in this thesis suggest a FGF/ERK mediated polycomb regulation that could potentially involve either or both of these mechanisms to regulate gene expression in the stem zone\caudal lateral epiblast of the mouse embryo.
During body axis elongation FGF signalling is attenuated by retinoic acid signalling in the posterior of the mouse embryo (Diez del Corral et al., 2003). The work of this thesis suggests that this retinoic acid mediated attenuation of FGF signalling leads to the dissociation of the polycomb machinery at neural differentiation genes and thereby the de-repression of target gene transcription.
|Date of Award||2016|
|Supervisor||Kate Storey (Supervisor)|