Monday, 2 April 2012

Gastrointestinal inflammation and immune activation in schizophrenia

Monday 2nd April 2012. It is World Autism Awareness Day today and, in light of the 1 in 88 figure recently published by the CDC, perhaps more poignant than ever. Today's post is not specifically autism research orientated but might provide an idea or two for any budding autism researcher out there.

Every now and again a paper comes along which seems to fill in a gap in the research literature. I'm not saying that said paper is anything definitive, as if such a thing exists in science, and in particular science about behaviour and psychology. But rather another piece of the puzzle seems to fit into a spare slot.

The paper in question on this occasion is this one by Emily Severance and colleagues* reporting on the presence of gastrointestinal inflammation and associated immune activation linked to cases of schizophrenia. I'm not the first to blog about this paper as viewers will see from the excellent shakedown from Dr Deans [see here]. Regular readers of this blog will probably understand why I had also to post about this paper with keywords like gastrointestinal (GI), inflammation, immune system and schizophrenia; all of which have littered various posts on a possible gut-brain connection. I should perhaps point out that as is usually the case, Dr Severance has 'some research form' in this area as per papers like this one and this one. I probably don't even need to mention the quite widely known papers of some of the other authors with particular focus on Faith Dickerson and Robert Yolken.

Where to start with this research. Well, I very recently talked about schizophrenia and some indication of overlap with cases on the autism spectrum. Schizophrenia has also figured in this post on comorbidity with coeliac (celiac) disease and the potential for a gluten-free diet to solve more than just underlying gastrointestinal pathology à la the work of the late Curt Dohan. The appearance of our parasitic friend, Toxoplasma gondii might also be of some relevance to this recent piece of research although more peripherally than anything else.

Details of the current paper are shown below with apologies that I can't link to the full-text of the article as things stand so you'll have to take my word for it:

  • Similar to some research undertaken in autism, the starting point for this paper is that there is some evidence of 'issues' with the gastrointestinal barrier in some cases of schizophrenia - the so-called leaky gut. The source of this permeability has not been definitively uncovered yet, but the speculation is that gastrointestinal inflammation as per what is observed in inflammatory bowel conditions such as Crohn's disease, might be a candidate. Sounds familiar doesn't it?
  • This study examined a marker of intestinal inflammation in the measurement of anti-Saccharomyces cerevisiae IgG antibodies (ASCA) more traditionally used as part of the diagnostic work-up for diagnosing Crohn's disease. I should note that Saccharomyces cerevisiae is not to be confused with a relation Saccharomyces boulardii covered in an earlier post. Alongside, various measures of exposure to food antigens and infectious agents in serum and plasma were also examined based on IgG antibodies in various patient groups at different stages of their condition. My recent post on the finding of Sutterella species in some cases of autism provides some logic for looking at IgG antibodies as a sort of peripheral indicator of GI contents meeting the immune system where it really shouldn't meet.
  • Two geographically separate participant groups were included for study: cohort 1 based in Baltimore, USA and cohort 2 based in Cologne, Germany. Within these two cohorts, cohort 1 groups were split into non-recent onset schizophrenia (SZ) (n=193), recent onset schizophrenia (ROSZ) (n=67) and controls (CO) (n=207); cohort 2 were split into first episode schizophrenia medicated (n=63) and first episode schizophrenia unmedicated (n=40). Mean ages across the cohorts and groups ranged from 22 years - 42 years with a real mix of genders and ethnic groups.
  • The findings: levels of ASCAs from the Baltimore cohort were significantly higher in both SZ and ROSZ groups compared with controls (p<0.004 and p<0.00001). Levels of ASCAs also significantly correlated with anti-casein and anti-gluten IgG antibodies in the SZ group and anti-gluten IgG antibodies with the ROSZ group. No significant correlation was found for the food antibodies and ASCAs in the non-psychiatric control group.
  • Looking at any effect from pharmacotherapuetic treatment, attention turned to cohort 2 based in Germany. Levels of ASCAs were significantly higher in unmedicated participants than medicated participants (taking anti-psychotics) and males seemed to be a key factor in this relationship. Indeed the unmedicated group also showed the only significant correlation with both IgG anti-casein and anti-gluten antibodies driven this time by females.
  • IgG antibodies to that most cunning of protozoan, T.gondii, suggested some possible connection with anti-casein antibodies in the ROSZ group with a particularly male slant (males in this group also showing some connection between T.gondii antibodies and anti-gluten antibodies with a slight typo in the paper when it came to the R-squared value).

So, one marker of intestinal inflammation was elevated (on the whole) in participants with schizophrenia compared with controls and depending on whether you had recent onset or not, and whether you were under medication or not, seemed to determine your antibody response to foods containing gluten and casein as a marker of permeability and its effects.

If I were to be at all critical of this work, I could have done with seeing a few more direct measures of things like inflammation such as levels of C-reactive protein and perhaps a more direct measure of gut permeability would not have gone amiss. I'm not going to get too heavy into things like measuring those tight junction proteins and the type of permeability present but it would have been great to see more detail.

There are lots of 'possibilities' arising from these findings.  For some reasons I am taken back to a post last year on a the study by Col. David Niebuhr and colleagues regarding the presence of casein antibodies as potentially being predictive of future symptom onset of schizophrenia. I know that the two sets of results are not wholly compatible but it does make me wonder about whether the ASCA test might have similarly been predictive.

The differences between medicated and unmedicated participants in terms of ASCAs also ask an important question on whether anti-psychotics might have more than just an anti-psychotic mode of action in terms of pharmacology. The answer, with quite high probability, is yes as per studies like this one and this one. Notice the focus on an anti-inflammatory effect.

The late Curt Dohan (and indeed Dr Dohan Jr. continuing the family research tradition) would perhaps be casting a smile at reading this text based on his ideas stretching back 40 years now about a possible connection between food - gluten and casein containing foods - and schizophrenia, at least some cases of schizophrenia. I have to wonder how these results might possibly tie into similar observations seen comorbid to some cases of autism and whether applying the tests and methods used in the current study to autism, anti-Saccharomyces cerevisiae IgG antibodies, anti-casein and anti-gluten IgG (and T.gondii IgG antibodies) might also be a study worth doing.

To finish, the JCB song.

* Severance EG. et al. Gastrointestinal inflammation and associated immune activation in schizophrenia. Schizophrenia Research. March 2012.
DOI: 10.1016/j.schres.2012.02.025


  1. Hi Paul Whiteley -

    Are there particular molecular properties of gluten/caesin that make them good candidates for measurement of innate immune response in the blood? I guess the lack of testing things actually in the gut bothers me a little bit about this study too; while asca is used for Crohn's, that doesn't necessarily mean it couldn't be increased in other conditions.

    Were food antigens other than gluten/caesin tested and found to have normal innate immune responses?

    In other words, what evidence do we have that this is a problem with specific antigens, as opposed to a generalized upregulation of the innate immune response to any stimulation?

    Don't get me wrong; I'm a believer in dietary effects that go far beyond our understandings, but these questions have been floating around my head for a little while now. I saw the Dean post on this study, and wasn't able to generate the same over the top enthusiasm for it.

    - pD

  2. Thanks for the comment pD.

    Now you are asking some questions. I perhaps need to do a little more research on gluten and casein proteins/peptides in terms of innate immune response. Having said that I remember Dr Jyonouchi's paper quite a while back which might hold some answers:

    Regarding the ASCA side of things, a colleague of mine did happen to talk to a paediatric GI man about this and he found it to be something which might be interesting to undertake. As far as I am aware, ASCA is really a test for permeability associated/following inflammation (I might be wrong) as per IgG antibodies being produced when permeability allows gut contents to meet the immune system where it shouldn't.

    Unfortunately in this paper, the authors only report on looking for IgG antibodies to gluten and casein and T.gondii so we are limited by what they report on.

    I think we need to bring some 'big guns' into this area of work to complement Buie and colleagues; thinking about people like Alessio Fasano and a favourite research group of mine headed by David Sanders and Marios Hadjivassilliou. (Pretty please)