High-fidelity epigenetic maintenance in somatic cells: CIZ1 assemblies as molecular shield

Cells in a developing embryo have the capability to become all of the different types of cell in the adult body. As they specialize they gradually turn off genes that are not required, until they only produce products that are essential to their function. One of the ways that their choice is restricted is by chemical modification of specific proteins (histones) that are closely associated with the genes to be shut down. Once established, these modifications must be stable, and also copied each time a cell divides so that daughter cells have the same repertoire as their parents, and therefore access the same genes as their parents. We are studying the processes that ensure histone modifications are faithfully maintained; the 'quality control' that stops a cell from drifting away from its path. Modifications are added by enzymes and removed by other enzymes, so the balance between these, and their relative ability to gain access to histones, governs whether specific histone modifications exist in a particular cell. This is fundamental to what the cell 'is' because it controls which genes it can express.

Recently we showed that a protein called CIZ1 influences the stability of at least three different histone modifications, so that they are absent when CIZ1 is not able to form large assemblies inside the nucleus of cells. We think that these assemblies normally form a shield around selected genes and their histones, to protect them from the enzymes that remove the modifications. This project aims to understand whether interference with shield integrity can destabilize cells, setting them off down a path that could contribute to disease, and possibly aging. We specifically want to explore whether fragments of CIZ1 that are already associated with human diseases, can initiate cellular degeneration by destabilization of CIZ1 assemblies. We aim to use the information to consider how we might intervene (with drugs or diet) to prevent CIZ1 assembly destabilization.

We also want to know whether interference with shield assembly ever happens during normal development, as a way of shifting the balance between 'on' and 'off' enzymes. This would be important to understand in the context of possible interventions. Finally, we plan to test how many different histone modifications are dependent on CIZ1, and under what conditions. So far, our analysis has been in cells taken from mouse embryos that are still developing, so it will be important to widen the picture to dividing and non-dividing cells taken from adult mice, and eventually humans.

CIZ1 forms large RNA-dependent assemblies around the inactive X chromosome, and smaller assemblies throughout the nucleus. Deletion of CIZ1 deregulates approximately 2% of genes in primary fibroblasts, causes loss of H3K27me3, H2AK119ub1 and H4K20me1, and predisposes mice to lymphomas and leukemias. We can model de novo formation of CIZ1 assemblies, and acquisition of histone post-translational modifications (PTMs), by adding ectopic CIZ1 to genetically null cells. This shows that a prion-like polyglutamine domain in the N-terminus of CIZ1 is essential for assemblies to form. Crucially, CIZ1 protein fragments encoding this domain (but lacking other essential elements) can also destabilize endogenous CIZ1 assemblies, leading to loss of histone PTMs. For H2AK119ub1, this is abrogated when deubiquitinylases are chemically inhibited. This suggests a novel function for CIZ1 as a molecular shield that promotes epigenetic stability by protecting chromatin from deubiquitinylases. It is important to understand this because truncated isoforms of CIZ1, similar to those with dominant negative assembly-dispersal capability in our assays, are evident in early stage common solid tumours, and possibly also mimicked by polyglutamine fragments derived from the Huntingtin protein. Here, we ask whether CIZ1 assembly destabilization by dominant negative fragments impacts on gene expression using a transcriptomic approach, and also ask how it affects the location and behaviour of the inactive X chromosome during its replication. We also seek to understand a normal biological context in which alternative splicing appears to modulate assembly formation and dissolution. An independent strand will compare histone PTMs in WT and CIZ1 null fibroblasts, then tissues, to establish the breadth of CIZ1's influence over chromatin state.