Epidermal growth factor and autism continued

I've talked before about epidermal growth factor (EGF) in relation to autism spectrum disorders (ASDs) on this blog. In that entry (see here) the suggestion was that plasma levels of EGF were 'generally' (and I use the word very loosely) reported to be lower in grouped cases of ASD.

Backing up a little, EGF is as its name suggests, a fairly abundant growth factor involved in various cell growth. Obviously that description of its function covers quite a lot of ground, so let me then focus your attention on areas like its role in relation to the gastrointestinal (GI) tract for example, and its connections (or not) to things like intestinal barrier function in certain circumstances. You can perhaps already see where I might be going with all this GI chatter in light of some corresponding interest in the GI tract with cases of autism in mind.

Enter then even more study on EGF levels in autism by way of the paper by Anthony Russo* (open-access). Dr Russo had already received a mention on my previous post on EGF with regards to some work done on another growth factor, HGF - hepatocyte growth factor - being decreased in cases of autism and comorbid GI issues. In the latest paper, the suggestion was that group plasma EGF levels were once again lower in cases of autism compared to asymptomatic controls. That and one or two correlations reported with other biological values and presented symptoms.

As per my usual cop-out, I'm not going to heavily into the paper because it is open-access and you can check out the details yourself. I would direct you to Figure 1 of the paper showing the comparisons between the groups (which included both an autism group and a wider autism spectrum - pervasive developmental disorder - group) illustrating just how different control values were from both the autism and PDD (that's PDD, not PPD!) values accepting the relatively small participant numbers included for study.

Russo also points to a positive correlation between EGF levels in the autism group with another compound included for investigation, HMGB1 - high-mobility group protein B1 - and a negative correlation with GABA (see here). Accepting that correlation doesn't necessarily mean anything more than correlation, I was quite interested in HMGB1 and the inflammation link; even more so the suggestion of a toll-like receptor 4 (TLR-4) and matrix metalloproteinase-9 (MMP-9) link** (at least in a mouse model of ischemic stroke).

I'll admit to being a little confused about the paper's link between EGF and HMGB1 though.

There was a suggestion that despite HMGB1 levels not being reported in the paper: "We have also found significantly increased plasma HMGB1 in our autistic group (P = 0.02; unpublished data)." So unpublished but higher levels. But just before the paper reported: "EGF levels correlated with HMGB1 levels (r = 0.5; P = 0.009)". The direction of the correlation is positive; albeit with an r-value not exactly as much as one would like, so one would assume that lower EGF would correlate with lower HMGB1, and higher EGF would correlate with higher HMGB1. Or am I not seeing something here?

Indeed I was slightly annoyed to see quite a few other references to 'unpublished' data in this paper specifically with regards to the statement of no difference in EGF values between GI associated and non-GI associated autism groups. I'd prefer to see the data included in the paper please (even as supplementary material), particularly when followed by a statement like "decreased EGF may be associated with increased inflammation but not necessarily increased inflammation often associated with GI disease in children with autism".

I have a few other issues with this paper too: the use of 'we' where there is only one author listed who seems to have done all the work which means sample blinding was not possible, and a summary sentence: "We suggest that decreased EGF is associated with increased plasma GABA levels".. Er, yes and no, given that the p-value for the correlation escaped significance coming in at p=0.06. Trend, perhaps. Association, hmmm.... assuming, that is, you adhere to the magical p=0.05.

Sorry if I've come across as being a bit pedantic about all this. Take no notice of me, it's been a long week: school half-term, catching up with work in between and oh, did I mention that I'm also involved in writing a book (TBC)? The important point to take from the Russo paper is that of the lower plasma EGF values and how they add to the existing literature on EGF in cases of autism.

Now we just have to work out what it all means.... answers on a postcard please.

To finish, Bastille and their very catchy tune 'Pompeii'.  And please, no questions about what the Romans ever did for us?

----------

* Russo A. Decreased Epidermal Growth Factor (EGF) Associated with HMGB1 and Increased Hyperactivity in Children with Autism. Biomark Insights. 2013; 8: 35-41.

** Qui J. et al. High-Mobility Group Box 1 Promotes Metalloproteinase-9 Upregulation Through Toll-Like Receptor 4 After Cerebral Ischemia. Stroke. 2010; 41: 2077-2082.

---------

Russo AJ (2013). Decreased Epidermal Growth Factor (EGF) Associated with HMGB1 and Increased Hyperactivity in Children with Autism. Biomarker insights, 8, 35-41 PMID: 23645980