Dem bones dem bones dem dry bones

There is some very interesting research into autism coming out of King Saud University in Saudi Arabia this year. The various papers published in 2011 include this one on fatty acids, this one again on fatty acids and this one on anti-ganglioside auto-antibodies. There is even a trial on-going at the same centre on the use of camel milk in autism, I assume as an alternative to the bovine (cow) version. One study in particular caught my attention looking at serum osteopontin levels in autism.

Osteopontin is a glycoprotein primarily charged with the maintenance of tissue integrity during inflammatory processes. It has some structural duties in relation to bone health but also acts as a cytokine (inducer). The main study findings from Saudi Arabia: children with autism (n=42) had higher levels of serum osteopontin than matched controls; indeed over 80% of the kids with autism presented with increased levels (I assume this means over and above the top reference range). Importantly also that levels seemed to correlate quite well according to severity of symptoms as measured by the CARS: the more affected the person, the higher the levels detected.

This is a relatively small study and as such requires further replication. When however authors start talking about significant differences alongside elevated levels of something present in approaching 80% of a particular group, it grabs my attention the same way that the recent Buie paper did when they suggested lactase deficiency in over 50% of their group with autism. The two functions detailed in relation to osteopontin, bone and immune system, have cropped up a few times in relation to autism spectrum conditions.


'Dem bones, dem bones, dem dry bones' have been studied in autism quite a bit, but of particular interest to me was this paper from Hediger and colleagues. They suggested problems with bone cortical thickness in autism potentially exacerbated by use of a casein-free diet linked to calcium intake (although acknowledging to be also potentially due to accompanying GI disorders, lack of sunlight exposure and /or limited physical activity). I have a post scheduled for posting in the next few weeks specifically on calcium so will discuss this more then. Suffice to say that recent research has not indicated any increased risk of fracture in autism compared with controls.

OK I hear you cry - he is going to use the osteopontin findings to try and explain away the Hediger results and protect his beautiful casein-free diet. Well, no I am not; even aside from the suggestion, outside of autism, that high levels of osteopontin might correlate with things like onset of osteoporosis, at least in post-menopausal women. Of more interest to me in this post are the potential immune effects of osteopontin.

I will warn you that it gets a little complicated from here on in (for you and me both), but please stay with me. Osteopontin has been linked to the production of various cytokines including IL-6 and IL-17 in various states. As an amplifier of the Th-1 immune response (pro-inflammatory), osteopontin has also been suggested to show involvement in inflammatory bowel conditions such as Crohns disease, at least in mice. This paper for example suggested  that osteopontin was a good cop / bad cop depending on whether inflammation was acute or chronic. During an acute event (short-term) it reduced tissue damage and actually helped in mucosal repair. During chronic inflammation, it turns bad guy and promotes further inflammation. The question in relation to the levels detected in the autism group is whether or not the findings represent a reaction to an acute event or a more chronic one and if so against what? (Note: IL-17 has been reported in relation to autism here and IL-6 more widely so, including this paper here).

The autism study authors end their abstract with a question as to whether anti-osteopontin antibodies might be a useful intervention for cases of autism associated with (consistently) high levels of the compound. The literature on the use of such antibodies is interesting. This study for example, suggested that one type of monoclonal antibody might be useful to treat a particular type of arthritis. This study suggested a more general effect on organ atrophy.

I would very much like to see a little more data on osteopontin as a marker and data on the suggested intervention options in the test-tube first before adding my tuppence worth on possible efficacy on either autism symptoms or associated co-morbidities. I do wonder also whether the results obtained are 'transferable' across the different peoples and ethnicites where autism is present given that no other group has yet published on this compound as far as I can see.

I finish with a song.