The Tremethick Group - Chromatin and transcriptional regulation during development

Chromatin and transcriptional regulation during development

The fate of the eukaryotic cell at all stages of its life cycle is dependent upon the accurate readout of genes encoded by DNA. For example, the development of a single cell into multicellular organism requires precise temporal and spatial regulation of gene transcription. Consequently, certain diseases and developmental disorders are often associated with, and likely to be caused by, aberrant gene expression. It has become increasingly clear over the last decade that eukaryotic gene regulation at the level of transcription is strictly connected to the structural organisation of the genome. The basis for this structural organisation is the nucleosome. Our overall aim is to understand how chromatin structure contributes to the regulation of transcription during development.

Remarkably, a typical eukaryotic cell contains approximately two meters of DNA, which can be squeezed into a nucleus of about 20 µm in diameter. This packaging of DNA is achieved by a hierarchical scheme of folding and compaction into a protein-DNA ensemble called chromatin. At the first level of organisation, approximately two superhelical turns of DNA is wrapped around a protein complex consisting of eight histone molecules. This complete unit, the nucleosome, forms the basic building block of chromatin and is further reorganised into a regular array to form a chromatin fibre. How this fibre subsequently folds into higher order structures is not yet understood. This protein-induced folding of DNA into a complex three-dimensional structure has profound implications with respect to understanding how gene expression is regulated.

As a result of this compaction of the eukaryotic genome, the conformation and accessibility of DNA is dramatically altered. The compaction of a gene into chromatin clearly impedes the transcription process. The cell has therefore devised mechanisms, which reversibly de-compact or remodel chromatin to allow transcription factor binding by altering the stability of protein-DNA interactions in underlying nucleosomes. The multi-subunit structure of the nucleosome (consisting of four histone-dimer subunits) is ideally suited for performing these opposing roles; nucleosomes are both stable enough to compact DNA while at the same time labile enough to allow access of DNA to transcription factors. This lability can be enhanced by targeted modification of the histone proteins or by changing the biochemical composition of the nucleosome. The hypothesis that we are examining is that the stability of protein-DNA interactions in underlying nucleosomes can be altered by the replacement of major histone types with specific histone variants.

There are two important stages with regards to the accurate transcriptional regulation of a gene. The establishment of gene activity, which for most genes occurs during early mammalian development, and the subsequent maintenance of this gene activity throughout many rounds of cell division during the life of the organism. Major global transitions in chromosome and chromatin structure occur early in development when cell lineage and tissue-specific transcriptional patterns of gene expression are established. Very little is known about these structural changes and the mechanisms by which these changes differentially regulate gene transcription. However, it is clear that chromatin plays a fundamental causal role in determining patterns of gene activity.


An important way to control chromatin function is to alter the biochemical make-up of the nucleosome by replacing an individual histone with a histone variant. Interestingly, most histone variants belong to the histone H2A family implying that H2A plays a unique role in the nucleosome. Modulating nucleosomal and higher-order chromatin structure through variation in H2A will have an impact on all nuclear functions.

Our work focuses on a variant of histone H2A referred to as H2A.Z. The presence and high level of conservation from yeast to man (across species, the amino acid sequence of H2A.Z is more conserved than the amino acid sequence of major H2A) shows that H2A.Z plays an important and specific role in chromosome function. This function is essential since in Drosophila and Tetrahymena null mutants die. We also found that the H2A.Z gene is essential for mouse survival with the defect occurring early in development around the time of implantation. However, despite being essential, nothing is known about the specific functional and structural consequences of having H2A.Z incorporated into chromatin.

To begin to understand why H2A.Z is essential for survival, we adopted an in vivo approach. To search for the unique feature(s) of H2A.Z required for its function, we performed amino acid swap experiments in which residues unique to Drosophila H2A.Z were replaced with equivalently positioned histone H2A residues. Mutated H2A.Z genes encoding modified versions of this histone were transformed into Drosophila and tested for their ability to rescue null mutant lethality. Most interestingly, we discovered that the unique and essential feature of H2A.Z lies outside the histone fold in the carboxy-terminal domain. This C-terminal region maps to a short alpha-helix in H2A that is buried deep inside the nucleosome. A region immediately adjacent to this short alpha-helix, located at the surface of the nucleosome, was also found to be important for adult Drosophila survival. Together, this region forms part of a docking domain, a domain involved in stabilising the interaction between the H2A/H2B dimer with the H3/H4 tetramer and in contributing to surface features of the nucleosome. Based on these results, our prediction was that H2A.Z would alter the stability of the nucleosome potentially weakening the interaction between the H2A.Z/H2B dimer with the H3/H4 tetramer and/or modifying the surface of the nucleosome.

Recently, we tested this prediction, in part, by solving the crystal structure of a nucleosome containing H2A.Z (in collaboration with Karolin Luger from Colorado State University). The overall structure is similar to the previously reported structure containing major H2A. In part, consistent with our prediction, distinct localised changes in the docking domain result in a subtle destabilisation between the dimer and the tetramer. However, there is also a stablisation between the two H2A.Z molecules at the back of the nucleosome. Interestingly, salt dissociation experiments indicate that H2A.Z actually increases the overall stability of the nucleosome. Further experiments are in progress to determine whether this is indeed the case. Potentially even more significant, the amino changes in the docking domain of H2A.Z results in an altered nucleosomal surface that includes a metal ion, a more extensive acidic patch, and a larger hole in the centre of the nucleosome. Our current favoured hypothesis is that these surface changes may create a highly specific interaction interface for other nuclear proteins like chromatin remodelling factors and/or modulate nucleosome-nucleosome interactions.

Most recently we tested the prediction that incorporation of H2A.Z into chromatin alters intra- and/or inter- nucleosomal interactions. This was determined by examining whether this histone variant alters the folding pathway from an extended nucleosomal array to highly compacted heterochromatin. In summary, we carried out the first biochemical analysis of a homogenous preparation of H2A.Z-containing nucleosomal arrays. H2A.Z uniquely effects chromatin condensation; intra-nucleosomal interactions are accentuated while fibre-fibre interactions and the subsequent formation of condensed heterochromatin is inhibited. These data suggest that a major function of H2A.Z is to generate a conformational intermediate in the chromatin-folding pathway poised to be assembled into a specialised functional chromosomal domain.

The prediction, from these structural studies, that H2A.Z functions to establish a specialised higher-order chromatin domain was recently tested. Using early mouse embryos at a time when H2AZ null mice die, confocal immunofluorescence experiments demonstrated that H2A.Z was located at constitutive heterochromatin. This is the first description of a function of H2A.Z. Current work is directed to determine, at the molecular level, the role of H2AZ at this specialised domain.


  • 'Hot paper' in Nature Reviews Molecular Cell Biology, February 2013.

Updated:  23 March 2017/Responsible Officer:  Director, JCSMR/Page Contact:  Web Manager