Menu

Close

The majority of histone acetylation is a consequence of transcription

Benjamin J.E. Martin, Julie Brind’Amour, Kristoffer N. Jensen, Anastasia Kuzmin, Zhen Cheng Liu, Matthew Lorincz, LeAnn J. Howe

Preprint posted on September 30, 2019 https://www.biorxiv.org/content/10.1101/785998v1.full#F3

Acetylation, / without RNA Pol II? / No chance in yeast cells!

Selected by Gabriel Aughey

Background

Gene expression in eukaryotic cells is associated with post-translational modifications of histone proteins. One of the best-known histone modifications is lysine acetylation, which is known to be enriched at actively transcribed loci. Whilst these modifications have been well characterised for many years, it remains unclear what the functional role of histone acetylation is, or even whether it promotes transcription or is a consequence of it. In this preprint, Martin et al address some of these long-standing questions – demonstrating that transcription is required for histone acetylation.

Key findings:

RNA polymerase association with chromatin is required for histone acetylation.

The authors of this study aimed to inhibit transcription in yeast to determine whether histone acetylation was disrupted. This was achieved by the use of the chemical inhibitor 1,10pt which had previously been shown to eliminate transcription in treated cells. Upon treatment with 1,10pt a loss of RNA polymerase II (RNAPII) was observed at transcribed loci. Remarkably, a concomitant loss of six different histone acetylation marks was seen after 30mins of drug treatment (Figure 1). The authors go on to show that the loss of acetylation during ablation of transcription requires active histone deacetylases (HDAC), but inhibition of HDAC was not sufficient to restore acetylation. These data raise the possibility that rather the loss of histone acetyltransferases (HATs) is the mechanism by which histone acetylation is depleted when transcription is disrupted.

 

Fig 1. (Figure 1A in preprint). Blots indicate that all histone acetylation marks assayed are depleted following transcriptional inhibition.

 

Transcription is required for HAT association with genes.

Next the authors turn their attention towards the HATs that may be responsible for RNAPII dependent histone acetylation. They demonstrate that inhibition of RNAPII causes the HAT subunit Epl1 to be mislocalised. In 1,10pt treated cells Epl1 was depleted in gene bodies, becoming newly enriched in promoter regions. In untreated cells Epl1 was found to bind to upstream regions of transcribed genes corresponding to transcription factor binding sites, indicating that Epl1 is targeted to genes by transcriptional activators. A truncated version of Epl1, thought to be unable to interact with transcriptional activators, lost this upstream localisation.

 

HAT activity is regulated after recruitment to chromatin.

The authors noticed that the genomic occupancy of Epl1 did not reflect the distribution of acetylated histones. Furthermore, when Epl1 was mislocalised to upstream promoter regions during inhibition of transcription, no concomitant increase in acetylation was seen at these sites. These data indicate that HAT activity is regulated downstream of recruitment.

When RNAPII accumulation at the 5’ ends of genes (as indicated by NET-seq CRAC-seq and chromatin bound RNA-seq) was compared to histone acetylation, a strong correlation was observed. These regions are thought to be loci at which RNAPII progress is slow. Sequences in transcribed genes that were thought to form more stable nucleosomes correlated with impaired transcriptional elongation (i.e. increased 5’ RNAPII enrichment) and histone acetylation indicating that these strongly positioned nucleosomes may be involved in the regulation of HAT activity.

Summary

Overall, Martin et al’s data provide compelling evidence to suggest that the majority of histone acetylation occurs as a consequence of RNAPII association with chromatin, due to loss of HAT recruitment. Whilst it remains unclear how HAT activity is regulated downstream of RNAPII, this preprint provides compelling clues as to the regulation of acetylation and transcriptional.

 

Posted on: 18th October 2019

Read preprint (No Ratings Yet)




  • Author's response

    LeAnn Howe shared

    1. In your experiments RNAPII recruitment to transcribed loci is disrupted. Do you think that polymerase occupancy alone without productive transcription would be sufficient for histone acetylation, or is elongation required?

    My favourite hypothesis is that the interaction of the histone tails with DNA makes them poor substrates for HATs.  During the process of transcription these interactions are disrupted, allowing acetylation by available HATs. If this hypothesis is correct, then it is productive transcription that dictates acetylation.  The one caveat to this is that we really don’t have a sense for how quickly RNAPII passes through a nucleosomes and thus the temporal window during which acetylation could occur.  If it happens quickly, then transcription alone might not be sufficient.  Indeed, there is little acetylation at the 3’ ends of genes, despite the fact that there are HATs present.  Instead, acetylation might primarily happen when RNAPII gets “stuck” at nucleosomes, in which case it is occupancy that is important.

     

    1. You showed that truncation of Epl1 was sufficient to reduce association with promoter regions but not gene bodies. If targeting of Epl1 to promoter regions is not required for its recruitment to genes, how can you explain this result? I.e. what is the purpose of its association with those regions? Do you know if these regions remain acetylated when Epl1 is truncated?

    One hypothesis is that the Esa1/Yng2/Epl1 is targeted to transcribed regions through affinity for RNAPII-disrupted nucleosomes, where it acetylates histones, facilitating RNAP passage.  Surprisingly, truncation of Epl1 does not result in loss of bulk acetylation.

     

    1. A recent study demonstrated that H3K27ac precedes transcription in zebrafish (https://dev.biologists.org/content/146/19/dev179127). Can you reconcile these findings with your data?

    Unfortunately I have not had time to read this one yet, but it seems interesting.

     

    1. Do you think your work has implications for the role of histone acetylation as an epigenetic marker in the inheritance of cell states?

    It depends on your definition of epigenetic.  If you are referring to acetylation as a marker of inheritance through DNA replication, then I would argue that it is too short-lived of a PTM for this function.

    Have your say

    Your email address will not be published. Required fields are marked *

    This site uses Akismet to reduce spam. Learn how your comment data is processed.

    Sign up to customise the site to your preferences and to receive alerts

    Register here
    Close