Sunday 22 April 2012

S100B protein and autism

I am a little late getting to this paper given that other bloggers have discussed some of the ins and outs of its content and potential implications for autism spectrum conditions. Hopefully I can add some additional value to this area of endeavour with this post.

I freely admit that until the paper by Al-Ayadhi and Mostafa* was published I had never come across the S100B protein ever. Indeed had someone asked me 'what is the S100B protein' I might have been tempted to say either some kind of experimental protein shake or perhaps some elaborate lab-grown meat alternative as per this story.

As it happens S100B protein is a member of the S100 proteins, so named because the protein is 100% soluble in ammonium sulphate at neutral pH. S100B protein is synthesised in quite considerable amounts by astrocytes and has been tied into various functions including long-term neuronal synaptic plasticity. At elevated levels however, S100B protein has been associated with neuronal death among other things as per this study looking at levels in Olympic boxers.

With this in mind together with some speculation on a link with autoimmunity, or more specifically autoantibodies tied into neurological damage, the authors set about looking at S100B as a marker of potential neurological damage combined with antiribosomal P protein antibodies. For those avid viewers, antiribosomal P protein antibodies have already been looked at by this Saudi autism research group before, specifically with plasma neurokinin A levels in mind.

A summary of the current paper:

  • Sixty-four children diagnosed with DSM-IV autism, mean age 8.4 years were included for study and compared against 46 age- and sex-matched asymptomatic controls.
  • Serum levels of S100B protein were measured by ELISA. Actually, samples were all analysed twice just to make sure there was some degree of consistency in the findings. At the same time, serum IgG and IgM antiribosomal P protein antibodies were also measured. ELISA relies on the principle of antibody-antigen interaction which can sometimes be affected by cross-reactivity. In this study, such confounders were not observed in any significant measure.
  • The results: serum levels of S100B protein were elevated in the autism group compared to controls (p<0.001). There was quite a bit of overlap between the groups so no AUC=1 for this study group this time around. Based on CARS scores, severity of autism also seemed to be a factor: more severe cases seemed to show elevated levels compared with milder or moderate cases (according to CARS).
  • Increased serum levels of antiribosomal P protein antibodies were also reported in the autism group compared with controls and again severity of CARS autism was also a factor.
  • When it came to plotting any correlation between the two variables, S100B and antiribosomal P protein antibodies, it was a case of close but no statistical cigar.

These results are interesting insofar as the main findings suggestive of increased serum levels of S100B in autism and a kind of sliding scale involvement when taking severity of presented symptoms into account. I would hasten to add that S100B protein has also cropped up in other conditions including schizophrenia and bipolar disorder although questions about whether this is a central finding or perhaps related to issues with glucose metabolism for example, remain unanswered. I note for example, markers of body weight and body mass were not recorded in the recent study thus remaining a potential confounding variable.

Alongside the other research being undertaken by this group with autism as a specific focus, a fuzzy picture is starting to be built up of quite a complex relationship between the immune system, various neurological functions and autism, at least in the small numbers of cases included in their studies.

To finish, I was torn between 2 songs for this post. Instead of deciding I offer a choice: The Pixies and Monkey gone to heaven or The Descendents and I'm the one. It's your choice.

* Al-Ayadhi L. & Mostafa GA. A lack of association between elevated serum levels of S100B protein and autoimmunity in autistic children. Journal of Neuroinflammation. March 2012.
DOI: 10.1186/1742-2094-9-54


  1. Hi Paul Whitelely -

    I haven't read that paper either, but man there go those Egyptians firing off on something interesting.

    I did see the 's100' family of proteins implicated previously in autism.

    One of a series that Enzo released near all at once, only to go seemingly silent in autism area.

    Also, you display significant coolness (p > .01) in suggesting either The Pixies or Decemberists in one post. My recommendation for you involves MC Frontalot.

    - pD

  2. Many thanks for the comment pD.

    I am going to look further at the AGE-RAGE axis paper. I'm drawn to the whole 'low-grade inflammation' and oxidative stress angles as potentially being relevant. Again another potential area where promising findings seem to be subject to the wax and wane of trends and funding.

    Thanks for the coolness comment. Cool is something that I have never been particularly good at but at least the musical influences of college and university have not been wasted!


Note: only a member of this blog may post a comment.