Two arms secure gene transcription at the right sites

The Cell Reports paper, a collaborative effort between labs in Oxford, Japan and the USA, reveals how a key histone-modifying enzyme is targeted to promoters and helps ensure appropriate gene expression (1). The findings provide an important step towards understanding how an epigenetic change can influence transcription.

From bacteria through to yeast and mammals, huge efforts have been made to understand how DNA-encoded information is used. In higher organisms, this includes elucidating how epigenetic factors control the way in which genes are switched on and off. 'We understand a lot about transcription factors that recognise specific DNA sequences,' says Professor Klose, 'but how do we bring in the epigenetic component? Does epigenetic information influence how our DNA sequences are used, thereby supporting complex human processes and ultimately differentiating us from more simple species?'

One important epigenetic feature is chromatin modification, which includes H3K4me - methylation of histone H3 on lysine 4 - a mark that is generally associated with regulatory regions of DNA. 'We know that most actively transcribed genes sit in an epigenetic state that is permissive for gene transcription and that this is associated with H3K4me,' explains Professor Klose. 'Active genes have histones with this methylation, whereas inactive ones don't. But we don't know why or how this happens.'

Work in the Klose lab focuses on understanding epigenetics in the context of CpG islands (CGIs). These are regulatory elements made up of non-methylated CpG dinucleotides. CGIs are associated with promoters and are epigenetically modified, including acquiring H3K4me. However, not all CGIs have high levels of H3K4me, raising the question of how the methylation complex is recruited to the appropriate subset of CGIs so that H3K4me can be placed on them.

In higher organisms, there are six large multi-protein complexes that can place this methylation on histone H3, one of which, SET1, is believed to methylate histones at gene promoters. Little is known about how these chromatin-modifying complexes function inside cells, including how they recognise their target sites in the genome. So the Klose group embarked on a comprehensive study of SET1, combining live-cell imaging with functional genomics to understand how the complex is targeted to the appropriate CGI gene promoters.

Former Wellcome-Trust funded DPhil student David Brown and postdoc Vincenzo Di Cerbo, involved in the work from its early days, were joined by Sir Henry Wellcome Postdoctoral Fellow Angelika Feldmann to provide computational expertise. Other Klose lab members came on board in what turned out to be a substantial team effort that also involved collaborators in the department and in Japan and Colorado.

The group chose mouse ES and epithelial cells to explore the role of the CFP1 protein, a component of the SET1 complex that carries a DNA-binding domain that can recognise non-methylated CpGs found in CGIs and a domain that is thought to bind to H3K4me. CFP1 is found at CGIs but it is not known how it binds to chromatin in vivo and whether its binding helps to guide SET1 complex to specific target sites.

The group looked at how fluorescently tagged wild-type and mutated CFP1 interacted with chromatin and combined this with ChIP-seq to map where the proteins were bound in the genome. These experiments showed that CFP1 bound non-methylated DNA and that it preferentially recognised H3K4me. Further studies with ES cells in which CFP1 could be knocked out demonstrated that it was largely CFP1 that determined where SET1 was recruited, thereby targeting the chromatin-modifying complex to active CGI promoters.

Having uncovered a mechanism for how the SET1 complex is recruited to the appropriate targets, the group then explored what affect this had on transcription. They analysed transcription genome-wide so that they could understand the extent of impact of SET1 binding. When they analysed CFP1 target genes in ES cells lacking CFP1, they found that histone modification was reduced and that gene expression was affected. Although the affect was not significant for the most highly transcribed genes, it was for more moderately transcribed ones - perhaps indicating that CFP1 plays a greater role in shaping the expression of the latter.

The findings reveal how the SET1 complex uses the multivalent interactions of CFP1 with CGI chromatin - first reading the methylation state of the DNA and then the histone methylation mark - to help it deposit H3K4me on its target genes. The group suggests that this two-pronged process may allow SET1 to initially sample through CGIs and if it encounters the methylated histones, its interaction with chromatin is stabilised, thereby amplifying the H3K4me state and sustaining transcription. Whilst this provides an explanation for how the SET1 complex regulates H3K4me at actively transcribed CGI-associated genes, it remains unclear how the H3K4me state would be initiated in the first place.

The next step for the Klose group will be to elucidate how the specific chromatin state that CFP1 helps to set up directly affects transcription - the missing link in the field more widely, says Professor Klose. 'There are many descriptions about epigenetic states and the outcome on transcription but very little is known about the molecular mechanisms underlying how the epigenome influences transcription,' he comments. Future studies in the lab will explore how the chromatin state affects recruitment and elongation of RNA polymerase.

As more is discovered about the epigenetic components controlling the genome, the divide between the simplest and most complex eukaryotes appears to become even more pronounced. Higher eukaryotes are multicellular and their genomes show a lot of redundancy, so additional layers of control are crucial. The steps described in this paper are just one part of a sophisticated and complex network that regulates the genome of these organisms. 'This type of epigenetic control helps constrain the genome and maintains sensible transcription,' says Professor Klose. 'In cancer, it is these sorts of complexes that are targeted, 'loosening up' the system up and allowing transcriptional events to occur that shouldn't.

Reference

1. The SET1 Complex Selects Actively Transcribed Target Genes via Multivalent Interaction with CpG Island Chromatin. Brown DA et al. Cell Reports 20:10, 5 September 2017, p2313-2327

Work in the Klose lab was funded by the Wellcome Trust, the European Research Council, the Lister Institute of Preventive Medicine, and European Molecular Biology Organisation.

Jane Itzhaki
11th September 2017