Monday, 11 March 2013

Inflammatory bowel disease in autism: distinctive features?

Where to start with this very long post... where to start?

That autism, some cases of autism, also coincide with various comorbidities sometimes including severe gastrointestinal (GI) issues is a relatively undisputed finding these days. I'm actually getting a little bored of saying this myself on this blog and I'm sure some readers are getting bored of hearing it too.

Health inequality
Lymphocytic infiltration @ Wikipedia  
The reason why I continue to keep hammering away at this line however is because there is a substantial gap between what the peer-reviewed literature is saying about GI factors coexisting with cases of autism and the real-world experiences of many people (children and adults) with autism in terms of getting such symptoms/conditions investigated and properly treated.

A health inequality if ever there was one; indeed why else would the folks over at Treating Autism feel compelled to have to produce a document on the medical comorbidities in autism spectrum disorders highlighting such an issue?

An autism-specific IBD variant?
With this in mind, today's post concerns the paper published by Stephen Walker and colleagues* (open-access) which suggests that when it comes to the more GI disease-related aspect of bowel disturbance linked to autism - as in inflammatory bowel disease (IBD) - it appears that such GI disease might have "distinctive features" onwards to either an autism-specific IBD variant or "a prodromal phase of typical inflammatory bowel disease". We had seen a hint that this study was coming to publication based on an abstract presented as IMFAR 2012 (see here).

OK, I know that this paper takes some people into uncomfortable territory. Bowel disease - lymphoid nodular hyperplasia (LNH) and enterocolitis and autism - means 1998, Lancet and retracted paper and Lord Voldemort style 'he who must not be named'. Indeed search through the Walker paper and you will see that 'he' is definitely not named among the references, or not at least as first author on any paper. This despite the fact that other, well-renowned teams have no issue in citing the retracted paper in question (see here). I'll also point out that the authorship list on the Walker paper includes Dr Arthur Krigsman who has published work in this area** (open-access) previously. Just sayin'.

Anyhow, a few details from the Walker paper bearing in mind it is open to all:

  • The name of the game was "transcriptome profiling of gastrointestinal mucosal biopsy tissue from ASDGI children and three non-ASD control groups (Crohn's disease, ulcerative colitis, and histologically normal)" to ascertain just how similar/different gene expression was. In other words, whether the molecular signature noted in cases of autism and bowel pathology (autism-GI) matches what it seen in other well-known, and relatively well characterised bowel diseases.
  • The focus of analysis was differentially expressed transcripts (DETs) noted in ileal and colonic tissue samples from the various groups under study. The tool of analysis was microarray followed up by quantitative real-time PCR (qPCR) to "validate representative transcripts that showed differential expression by microarray". A sort of scatter gun, see what hits we make approach, followed by a more precise confirmation of said hits. Results were organised according to PCA "to determine similarity among biological replicates". 
  • Participants were 25 children diagnosed with an autism spectrum disorder (mean age around 5 years old) who all presented with a regression in previously acquired skills and some kind of inflammatory-related condition of the bowel (ileitis, colitis, ileocolitis). Indeed all presented with LNH. Most of children with ASD were following a gluten- and casein-free (GFCF) diet. Control groups were rather smaller in number and there was a bit of an age difference between them and the autism-GI group.
  • Results: Lots. Bearing in mind the authors were looking at tissue from two different parts of the GI tract, across quite a few groups, so this is probably not unexpected. For the autism-GI group, not everyone was included in all the analyses based on the two different regions of the gut because of quality/quantity of the RNA derived from them. 
  • For both colonic and ileal mucosa gene expression profiles, the asymptomatic control group (histologically normal) tended to cluster tightly together. Not so when it came to the symptomatic groups, where the autism-GI group in particular showed quite a bit of variability "suggestive of some potential subgroup(s)". Mmm, interesting.
  • Various comparisons were made between the groups based on the DETs identified in the two regions. I would be here all day and night listing the results of all this, so I'm cherry-picking a few bearing in mind it was a mix of up-regulated and down-regulated gene expression. (a) Autism-GI vs. asymptomatic controls (ileal mucosa): "1409 DETs unique to ASD-GI samples". DETs tended to link back to things like inflammatory bowel disease and colitis genes and the inflammatory response. Then we get down to genes related to humoral immune response, antibody production and digestive system development and function. (b) Autism-GI vs. asymptomatic controls (colonic mucosa): "1189 DETs unique to ASD-GI samples". DETs were linked to things like again gastrointestinal disease and also "neurological disease" including schizophrenia (50 genes) (see this post) and "hyperactive disorder" (16 genes) (which I take as meaning something like ADHD).
  • Again focusing on the autism-GI group and the DETs which overlapped both ileal and colonic tissues (178 transcripts), the top associated biological functions were in relation to inflammatory disease, endocrine system development and function, and digestive system development and function. Interesting also that 3 genes also cropped up in the pathway analysis linked to "Valine, Leucine and Isoleucine Degradation".  Did someone say branched-chain amino acids and autism
  • A very large quote: "Taken as a whole, the picture that emerges is one in which GI symptomatic children with ASD in whom cellular infiltrate is present in the ileum and colon have a distinct molecular signature that is consistent with the larger disease categories of gastrointestinal disease, and more specifically, overlaps with Crohn's disease, ulcerative colitis, and autoimmunity".

What can we surmise from these results? Well potentially quite a bit but with the very strong requirement for replication, replication, replication with a greater number of participants over a longer period of inspection. That all children with ASD included in the Walker study also presented with defined bowel disorder (including LNH) is an important findings. I've talked about inflammatory bowel disease and autism in a previous post based on the paper by Chen and colleagues*** which concluded that LNH was not an inconsistent finding - "apart from intestinal lymphonodular hyperplasia, the majority of these findings were not consistent" - among the various papers looking at bowel disorder in cases of autism. The Walker paper adds to that feeling.

Carbohydrate metabolism
I was interested to read about the potential overlap between the Walker findings and those reported in the Brent Williams paper on carbohydrate metabolism**** in autism (see here and here reloaded). Williams and colleagues reported decreased mRNA expression in cases of autism related to important carbohydrate metabolising enzymes alongside signs and symptoms indicative of things like dysbiosis. Walker says 'yes'; they found similar evidence in their cohort, particularly when it came to the down-regulation of CDX2.

Politics and kingdoms aside, and bearing in mind the heterogeneity present within the autism spectrum (and its fuzzy boundaries), this paper is an important one. Re-opening scientific investigation into how autism is not just the sum of its dyad, and how a diagnosis of autism is seemingly protective of nothing when it comes to comorbidity, are important elements to this work. Realising the very real and very significant pain and discomfort which can occur when such bowel conditions are present alongside autism is a good first step to bringing this particular branch of GI research out of the shadows, and finally asking the question: what can we do about it?

If indeed people want to see real equality when it comes to healthcare access and use for people with autism, continued investigation on the topic of GI disease and autism inclusive of both functional presentation and underlying pathology is absolutely implied.

To close, Blondie and One Way or Another....


* Walker S. et al. Identification of unique gene expression profile in children with regressive autism spectrum disorder (ASD) and ileocolitis. PLoS ONE. 2013; 8: e58058.

** Krigsman A. et al. Clinical presentation and histologic findings at ileocolonoscopy in children with autistic spectrum disorder and chronic gastrointestinal symptoms. Autism Insights. 2010; 2: 11.

*** Chen B. et al. Abnormal gastrointestinal histopathology in children With autism spectrum disorders. J Pediatr Gastroenterol Nutr. February 2011.

**** Williams B. et al. Impaired carbohydrate digestion and transport and mucosal dysbiosis in the intestines of children with autism and gastrointestinal disturbances. PLoS ONE. 2011; 6: e24585.

---------- Walker, S., Fortunato, J., Gonzalez, L., & Krigsman, A. (2013). Identification of Unique Gene Expression Profile in Children with Regressive Autism Spectrum Disorder (ASD) and Ileocolitis PLoS ONE, 8 (3) DOI: 10.1371/journal.pone.0058058


  1. Don't forget the common comorbidity of autism and celiac.Usually there is another GI disease besides celiac.

    It isn't just GI disease.So many diseases are different in autism.There is quite a difference in the diseases between children who have just mitochondrial disease,and those who have bothautism and mito.

  2. Thanks Roger.

    I'm not altogether sure that I would call coeliac (celiac) disease a 'common' comorbidity in cases of autism but it certainly can and does happen.

    and sometimes showing a quite surprising connection:

    You raise an interesting point about how the presentation of autism might affect the presentation of other conditions which I'd like to see explored a little further.

  3. Paul, one of the things that has been bugging me about autism/gut research is anxiety. Anxiety is co-morbid with autism and can cause gut issues in neurotypicals. I wonder if some of these kids just need a chill pill.

  4. Many thanks for the comment Sarah Jane.

    I'm sure many people (with autism or not) would love the option of a chill pill at certain times.

    Joking aside, the study by Micah Mazurek and colleagues on the relationship between GI issues, anxiety and sensory issues perhaps hints at the important relationship between these elements:

    That being said, I'd like to think that management of such severe bowel issues as noted by Walker and colleagues deserves more than just CBT...