Zebrafish gut motility, SHANK3 and autism?

I have to admit that I've always been a little bit bewildered by the use of the zebrafish (Danio rerio) in autism research. It's not that I don't understand the [careful] use of 'animal models' for autism research, or that zebrafish make for good research model material [1] from a genetic and biological point of view. It's just that zebrafish seem so far removed from real life autism; it's heterogeneity, complexity and also its 'humanness'.

The findings published by David James and colleagues [2] kinda changed my mind a little bit. Their findings supporting "mutations in SHANK3 as causal for GI [gastrointestinal] transit and motility abnormalities" caught my attention. Even more so when I realised that this was not the first time that I've blogged about SHANK3 disruptions having more than just 'brain effects' (see here), their connection to the gut (intestinal barrier function no less) and what this could mean for some autism.

James et al undertook their study on the basis that: (a) "gastrointestinal (GI) distress is a commonly reported but a poorly understood co-occurring symptom" alongside many instances of autism (yes it is), and (b) autism research still only has a preliminary idea of why gut issues are over-represented when it comes to autism. Given that zebrafish have been previously used to examine "GI dysfunction in Hirschsprung’s and chronic intestinal pseudo-obstruction diseases" as well as that previous research on gut barrier issues associated with SHANK3 disruptions [3], the research began.

I can't really claim any major expertise in the hows-and-whys of genetically manipulating zebrafish to "generate a zebrafish model of PMS [Phelan-McDermid syndrome]... a condition caused by mutations in the SHANK3 gene" but it was nonetheless achieved. I understand that the technique known as CRISPR/Cas9 was utilised as part of the research strategy and led to the production of mice with "shank3 loss-of-function mutations." Researchers subsequently set about studying their SHANK3 mutant zebrafish specifically focused on "the digestive tract (DT) structure and function." To do this we are told that "videos of gut peristalsis in intact transparent 7-day-old zebrafish larvae after feeding with a chicken egg yolk emulsion" were captured. Peristalsis by the way, refers to the waves generated by muscle movement that, in the gut, keep things moving from top to bottom. Researchers also fed microscopic "fluorescent beads" to mutant and non-mutant (wild type) zebrafish larvae in order to measure digestive tract (DT) transit as a function of that SHANK3 dysregulation. A few other experiments were also carried out and reported on pertinent to their research focus.

"Our studies are the first to establish DT dysmotility as a robust phenotype in any SHANK3 mutant animal model of ASD [autism spectrum disorder]." Hypomobility denoting a reduction in gut motility was found, something that perhaps ties in with other more general autism-related research literature (see here). The magnitude of the reduced gut motility reported by James and colleagues was quite notable: "Comparatively, shank3abΔC +/− larvae took longer than 12 h to begin passing the microspheres and some individuals had not passed the remainder even after 24 h post consumption." Authors talk about 'sloshing' as being potentially important to this increased transit time "where the microspheres would repeatedly move anteriorly and posteriorly between the intestinal bulb and upper-intestine."

Another detail mentioned by James et al relates to their attempt to "rescue the DT dysmotility phenotype" via an injection of "mRNA encoding either the longest human SHANK3 isoform that includes all SHANK3 protein domains (5t, n = 19) or a shorter human SHANK3 isoform that includes only the C-terminal proline-rich and SAM domains (32t, n = 6) into fertilized eggs from shank3abΔC +/− mutants." This 'we can rebuild him' molecular rescue attempt wasn't a complete success, although did partially improve the DT (digestive tract) transit time.

OK, so one needs to remember that this is still work based on the examination of zebrafish. It's work that looked at one particular genetic 'issue' noted in a genetic condition that manifests behaviour(s) that look like autism [3]. There are, as you can see, various issues that perhaps stand in the way of making any sweeping generalisations back to 'all autism'. But in light of the other research in this area similarly linking gut-related parameters back to SHANK3 disruptions, a trend is beginning to appear. A trend that taps into other important concepts in autism research that "brain, gut, and microbiome" represent emerging research areas with respect to autism, that 'autism genes are probably not just genes for autism' in a brain-behaviour sense (see here) and onward that gastrointestinal (GI) issues appearing alongside autism may be much more than just 'comorbidity' for some...

----------

[1] Sakai C. et al. Zebrafish Models of Neurodevelopmental Disorders: Past, Present, and Future. Front Mol Neurosci. 2018 Aug 29;11:294.

[2] James DM. et al. Intestinal dysmotility in a zebrafish (Danio rerio) shank3a;shank3b mutant model of autism. Molecular Autism. 2019; 10:3.

[3] Wei SC. et al. SHANK3 Regulates Intestinal Barrier Function Through Modulating ZO-1 Expression Through the PKCε-dependent Pathway. Inflamm Bowel Dis. 2017 Oct;23(10):1730-1740.

----------

Headline fail: "Autism traits could be 'edited' out genetic trial suggests"

The Telegraph June 25 2018

"Autism traits could be 'edited' out genetic trial suggests" was one of the headlines that accompanied the publication of the findings by Bumwhee Lee and colleagues [1].

The Lee article details some interesting science following the use of something called CRISPR-Cas9 gene editing (see here) or more precisely the use of a new-ish development to this technique - "CRISPR–Gold, a nonviral delivery vehicle for the CRISPR–Cas9 ribonucleoprotein." CRISPR is one of the hottest things in science at the moment, as the words 'find, cut and paste' move to a genetic level (see here) and promises so much. In the Lee study, the target was the metabotropic glutamate receptor 5 (mGluR5) gene as a move to "efficiently reduce local mGluR5 levels in the striatum."

Oh, did I also mention that this research was done using mice? Indeed, this was a study of mice engineered to display some of the molecular and behavioural characteristics of a condition called Fragile X syndrome (FXS). As such, authors reported that the use of CRISPR-Gold injections into the striatum of said FXS mice correlated with a reduction in certain behaviours such as obsessive digging and leaping into the air. The authors opine that such behaviours 'overlap' with those noted in autism (FXS has a 'connection' to autism) and voilà, a link to autism is made.

Aside from that brief overview of the findings from Lee et al just mentioned, I'm not going to go too much into the nitty-gritty of the actual results. A cobbler should stick to his last and all that, and others have done a far better job than I ever could in discussing the science (see here). I do however want to make a case that the 'autism traits could be edited out' headline represents a fail when covering the Lee findings.

I say 'headline fail' in the title of this post because well, it is. Not only does it assume that obsessive digging and sporadic leaping into the air made by mice are singularly autistic traits, it takes a few sentences before the word 'mice' is even mentioned in the coverage. I've talked before about the caution(s) needed when translating animal findings to real people (see here) and how autism in particular, seems to be a label ripe for mass sweeping generalisations from 'autistic animals' to autistic people. I'm not saying that some of the features of autism are uniquely human (see here) but rather that is it premature to even imply that the traits of autism can be 'edited out' on the basis of a single mouse or other animal genetic study.

I've already mentioned about a 'connection' between FXS and autism but it is perhaps also important to realise that there seem to be many routes that bring someone to a diagnosis of autism. FXS is one condition that manifests autistic traits but it is not the only one and certainly science does not yet know everything there is to know about the genetics of autism and FXS. And just before anyone starts talking about autism being universally 'in-born' and 'genetic' as it is [assumed] in FXS, well, the peer-reviewed research evidence might just disagree with you (see here for one example)...

Finally there's another aspect to this work that requires sensitive media handling: the ethics of 'editing out' autistic traits. I know this is a 'hot potato' area, as an increasingly vocal - certainly on social media - group of people on the autism spectrum talk about their strengths as well as their disabilities. Much of this discussion is framed around the notion that autism is not something separate from who they are but rather a fundamental part of who they are. If one takes this viewpoint, it is logical to assume that 'editing out' autistic traits might mean something rather ominous to some people...

The point I'm trying to get across is that the Lee paper is seemingly good science. It faithfully reported the results of an exciting new technology that holds promise for many different labels, conditions and diseases (see here). The issue however, is that the reporting of such research needs to be accurate and responsible. Headlines in particular, need to mention the word 'mouse' if it was a mouse study. They need to avoid sweeping generalisations that infer that digging and leaping behaviour in animals are generalisable as autistic traits (certainly the latest ICD-11 schedule says nothing about such behaviours), and they need to be sensitive to the fact that 'editing out' may very well provoke significant anxiety among some people on the autism spectrum. All for the sake of an attention-grabbing headline...

----------

[1] Lee B. et al. Nanoparticle delivery of CRISPR into the brain rescues a mouse model of fragile X syndrome from exaggerated repetitive behaviours. Nature Biomedical Engineering. 2018. June 25.

----------

CRISPR-Cas9 and autism research

If you feel brave enough, today I will direct your reading attention to the paper by Michael Williams and colleagues [1] detailing the application of a particularly important genome editing technique called CRISPR-Cas9 [2] to autism-related science.

Titled: "A Retroviral CRISPR-Cas9 System for Cellular Autism-Associated Phenotype Discovery in Developing Neurons" the Williams paper probably won't win any awards for plain English but don't be fooled about just how important this paper might be in the grand era of 'we can edit genomes' and how this might translate into modelling particular types of autism or genetic issues linked to autism in mice or other animals for example.

I really wish that I could say I was an expert on CRISPR-Cas9 and understood every detail included in the Williams paper but alas, I'm not and I didn't. Bearing in mind my non-expertise ('a cobbler should stick to his last') I did want to include it on this blog given the excitement in this area. Take my observations however, with a large pinch of salt...

So a definition of CRISPR [clustered regularly interspaced short palindromic repeats] -Cas9 - well, I don't want to reinvent the wheel so I'll use that offered in reference [2] with full credit given to the writer (Steph Yin): "Here’s how CRISPR/Cas works in bacteria: When bacteria encounter an invading source of DNA, such as from a virus, they can copy and incorporate segments of the foreign DNA into their genome as “spacers” between the short DNA repeats in CRISPR. These spacers enhance the bacteria’s immune response by providing a template for RNA molecules to quickly identify and target the same DNA sequence in the event of future viral infections. If the RNA molecules recognize an incoming sequence of foreign DNA, they guide the CRISPR complex to that sequence. There, the bacteria’s Cas proteins, which are specialized for cutting DNA, splice and disable the invading gene." The application of this process outside of just bacteria was subsequently recognised and a 'gene editing tool' was eventually born whereby a CRISPR-Cas9 system could replace any gene sequence.

Clear as mud right?

Well, Williams et al add to a small but emerging peer-reviewed research base at the time of writing suggesting that CRISPR-Cas9 might provide some important insights into at least 'some' autism. Their particular idea was to "mimic nonsense PTEN mutations from autism patients in developing mouse neurons" on the back of some previous research suggesting that various genetic issues with PTEN might be present in some autism [2]. Nonsense mutation by the way, normally ends in 'nonfunctional proteins' based on the knowledge that [some] genes provide the template to make proteins.

To achieve such mutations in PTEN authors used "retroviral implementation of the CRISPR-Cas9 system" where engineered retroviruses were purposed to deliver something mimicking a genetic mutation previously noted in cases of autism that were then injected into "the hippocampus of postnatal day 7 (P7) mice." Researchers then monitored the retrovirus infected cells to see what they looked like in terms of carrying the mutation and hence showing loss of PTEN function or not. They noted that there was a degree of 'hit-and-miss' based on their approach but were "able to clearly discern the established hypertrophic phenotype due to loss of Pten function across the cell population on average."

Not content with such molecular engineering, authors also turned their attention to designing viruses "to target a gene that has recently been associated with autism, KATNAL2." KATNAL2 has been described by other authors as a 'genuine' autism risk factor [3] (er, right...) and on that basis researchers designed a retrovirus carrying a mutation designed to disrupt expression of the gene. After some preliminary work to test out how successful their retrovirus delivered mutation was in the test tube, they injected it and/or a control retrovirus into the brain of another set of mice. They found some interesting changes in the experimental retrovirus-infected brains pertinent to "decreased dendritic arborization of developing neurons." In layman's terms this equates as evidence of "disruption of normal neuronal development" that "may lead to synaptic circuit dysfunction underlying the autism phenotype."

As per my earlier 'pinch of salt' sentiments I am not offering any authoritative opinion about the Williams paper and the techniques included. My interpretation is just that; interested readers are advised to do a little more reading around this subject before quoting my text as 'truth'. What I do hope that I've got across is the message that CRISPR-Cas9 and the delivery of engineered genetic mutations via something like a retrovirus is already here and will no doubt be impacting on autism research in times to come. In an era where 'the autism gene' has been replaced by a more general model of many different genes potentially producing many different autisms (see here), one can perhaps see how focusing in on specific genes linked to 'some' autism might be ripe for this kind of analysis (see here). I say all that recognising that whilst many would love to be able to say that autism is solely a genetic condition, the role of non-genetic factors variably affecting risk is not to be forgotten (see here).

We will see what else emerges in the peer-reviewed domain in this brave new world...

----------

[1] Williams MR. et al. A Retroviral CRISPR-Cas9 System for Cellular Autism-Associated Phenotype Discovery in Developing Neurons. Sci Rep. 2016 May 10;6:25611.

[2] Yin S. What Is CRISPR/Cas9 and Why Is It Suddenly Everywhere? Motherboard. 2015. April 30.

[3] Neale BM. et al. Patterns and rates of exonic de novo mutations in autism spectrum disorders. Nature. 2012 Apr 4;485(7397):242-5.

----------

Williams MR, Fricano-Kugler CJ, Getz SA, Skelton PD, Lee J, Rizzuto CP, Geller JS, Li M, & Luikart BW (2016). A Retroviral CRISPR-Cas9 System for Cellular Autism-Associated Phenotype Discovery in Developing Neurons. Scientific reports, 6 PMID: 27161796