Acetylation focus over methylation in autism epigenetics?

The paper by Wenjie Sun and colleagues [1] (open-access) provides the blogging fodder for today's post and although based on the science of epigenetics, the usual suspect - DNA methylation - gives way to another concept: histone acetylation with autism in mind. Before heading into the paper myself, I'll draw your attention to some other write-ups of the study including a hat-tip to Jeff Craig and his piece on the topic (see here).

So histone acetylation... I've covered the subject before on this blog (see here) but basically DNA, the stuff that carries the genetic blueprint, complexes with histones to form something called nucleosomes. It's a combination likened to thread wrapped around a spool. Continuing that thread wrapped around a spool analogy, protruding threads called histone tails can be modified in a chemical sense (via processes such as acetylation where an acetyl group is added or deacetylation where one is removed) which can subsequently affect genetic transcription.

Still with me? Good. Sun and colleagues set out to look at histone acetylation in the context of autism; specifically in post-mortem brain samples donated from those deceased who were diagnosed with autism and whether they might show some important changes based on the use of "a histone acetylome-wide association study (HAWAS)."

Specific areas of the brain were assessed using the HAWAS approach - "prefrontal cortex (PFC), temporal cortex (TC), and cerebellum (CB)" - and researchers were looking for a specific type of acetylation mark called H3K27ac linked to gene activation. Based on brain samples from 94 participants ("45 ASD [autism spectrum disorder], 49 control"), a few details emerged:

• Despite the expected heterogeneity across the presentation of autism in terms of whether the diagnosis of autism was syndromic (secondary to an existing condition) or non-syndromic (idiopathic), the authors reported that approaching 70% of the autism cases "shared a common acetylome signature at >5,000 cis-regulatory elements in prefrontal and temporal cortex." In other words, a not uncommon molecular signature in relation to histone acetylation seemed to be present in quite a few of the participant samples included for study.
• Although one needs to be a little cautious about making grand, sweeping claims about how such an 'acetylome signature' comes about, the authors reported "that ASD-specific differential acetylation is driven mostly by.. factors such as environmental influences, SNPs in trans (at a different locus), indels, and larger chromosomal variants." Note the term 'environmental influences' (something I'll come back to shortly).
• When it came to what types of genes were potentially being 'affected' by acetylation, the authors report on quite a diverse spread "involved in synaptic transmission, ion transport, epilepsy, behavioral abnormality, chemokinesis, histone deacetylation, and immunity." Epilepsy and autism is a recurrent theme in the research and clinical literature (see here for example) so there are no great surprise there. 'Immunity' and autism is something else that keeps cropping time and time and time again (see here).
• I appreciate that the authors also acknowledge that whilst autism was the focus on the current work, they do also mention: "By correlating histone acetylation with genotype, we discovered >2,000 histone acetylation quantitative trait loci (haQTLs) in human brain regions, including four candidate causal variants for psychiatric diseases." This opens up the idea that various different psychiatric/behavioural labels might show 'overlap' when it comes to the histone acetylome too.

Interesting stuff by all accounts. I do like the idea that autism research is continuing to look at other areas of gene expression outside of just structural issues to the genome being linked to the condition (or should that be plural). Aside from the fact that people don't walk around with their genes permanently stuck in the 'on or off position' in every tissue all the time, the whole epigenetics field is a welcome complement to more traditional genomics. The focus on gene expression being potentially 'modifiable' might also reunite genetics and environment too (see here).

Criticisms of the Sun study? Well, brain samples from the deceased are a precious resource but not without complications when it comes their use for science (see here). I appreciate that we don't have the technology to look at histone acetylation in real-time or real-life yet with the brain in mind but one has to be cautious about the results from the brains of the deceased who may have passed away for many different reasons. There is also the temptation to move the whole epigenetics 'thing' towards acetylation on the basis of such research, but the methylome still remains potentially important (see here) and probably for more than one reason (see here). Perhaps soon we'll see a study looking at more than one epigenetic factor with autism in mind?

Going back to the concept of 'environmental influences' mentioned in the Sun paper, there are some potentially important repercussions from study results such as these. As with the concept of DNA methylation, one of the important concepts linked to the science of epigenetics is that such chemical alterations affecting the expression of DNA are potentially modifiable. This could mean that particular environmental factors working at critical periods might affect acetylation and methylation patterns and onward the expression of certain genes pertinent to the presentation of something like autism or at least facets of autism. The other scenario is that certain 'conditions' or 'interventions' might 'reverse such changes. On that last point, I might bring in some previous discussions on this blog in relation to something called HDAC (histone deacetylase) inhibitors (see here) that, as their name suggests, have the ability to inhibit the action of histone deacetylases (they remove acetyl groups). Various classes of medicines are classed at HDAC inhibitors including something called valproic acid which has some autism research history (see here for example). It's not therefore beyond the realms of possibility that the actions of certain medicines or other non-genetic factors with an influence on acetylation could be a source for further research in this area.

Independent replication is the next stage in the research process here. Alongside marrying acetylation trends with methylation trends, I do also wonder whether more functional analysis of other tissue(s) outside of just the brain might also be revealing too. I might add that traditional structural genomic issues (all those SNPs and CNVs that are talked about) can still play a role and indeed, might show some association with epigenetic issues too (see here).

And with all this talk of epigenetics and the like, due credit needs to be given to those who've been talking about this for quite a while in the peer-reviewed domain [2]...

To close, we have one final look at Rogue One before touchdown...

----------

[1] Sun W. et al. Histone Acetylome-wide Association Study of Autism Spectrum Disorder. Cell. 2016. Nov 17.

[2] Lasalle JM. Autism genes keep turning up chromatin. OA Autism. 2013 Jun 19;1(2):14.

----------

Sun, W., Poschmann, J., Cruz-Herrera del Rosario, R., Parikshak, N., Hajan, H., Kumar, V., Ramasamy, R., Belgard, T., Elanggovan, B., Wong, C., Mill, J., Geschwind, D., & Prabhakar, S. (2016). Histone Acetylome-wide Association Study of Autism Spectrum Disorder Cell, 167 (5), 1385-2147483647 DOI: 10.1016/j.cell.2016.10.031