Blood glutamate levels in autism meta-analysed

"The meta-analysis provided evidence for higher blood glutamate levels in ASD [autism spectrum disorder]."

That was the research bottom-line reported by Zhen Zheng and colleagues [1] (open-access available here) who surveyed the current peer-reviewed science literature in this area and found something to see based on: "Twelve studies involving 880 participants and 446 incident cases."

Drawing on the idea that glutamate is a rather important amino acid that plays a role in various biological processes including that related to the manufacture of GABA (see here), Zheng et al observed higher circulating blood levels of the stuff; a sort-of proxy for what might also be going on with regards to brain levels of glutamate. That "excess glutamate has been shown to be a potent neurotoxin that leads to neuronal cell death and plays a role in the pathophysiology of some neuropsychiatric disorders" is an important point to make as to the potential implications from the Zheng meta-analysis.

Zheng et al do mention how important glutamate is for the purposes of GABA production and in particular, how issues with glutamate decarboxylase (GAD) - a key enzyme that converts glutamate into GABA - described in some cases of autism [2] might account for the elevated levels of glutamate yet the generally lower levels of GABA seen in autism (see here). I'd be inclined to agree that this is perhaps one of the more important implications for glutamate in autism; particularly when added to the whole 'glutamate linked to epilepsy' bit knowing how close a relationship autism and epilepsy seem to share (see here).

Where next with this research area I hear you ask? Well, I'd like to know a little more not just about glutamate but also another linked amino acid called glutamine. It has already been talked about in the autism research literature a while back (see here) but a lot more follow-up work is required on these two important compounds and what their differing ratio might mean. I'd also like to see more work done on the idea that "the mood stabilizer valproic acid, which exerts neuroprotective effects against glutamate-induced excitotoxicity, is effective in ASD [autism spectrum disorder] with seizures." Yes, I know that valproic acid a.k.a valproate is a bit of a double-edged sword when it comes to autism and other offspring developmental issues under certain circumstances (see here) but much like another research story in autism (see here) timing of exposure seems to be a key issue and one wonders whether other unrelated compounds might also exert a similar neuroprotective effect.

As to the idea that "blood glutamate levels may serve as a potential biomarker in the diagnosis of ASD" made by Zheng and colleagues, we'll wait and see...

----------

[1] Zheng Z. et al. Blood Glutamate Levels in Autism Spectrum Disorder: A Systematic Review and Meta-Analysis. PLoS One. 2016 Jul 8;11(7):e0158688.

[2] Yip J. et al. Decreased GAD65 mRNA levels in select subpopulations of neurons in the cerebellar dentate nuclei in autism: an in situ hybridization study. Autism Res. 2009 Feb;2(1):50-9.

----------

Zheng Z, Zhu T, Qu Y, & Mu D (2016). Blood Glutamate Levels in Autism Spectrum Disorder: A Systematic Review and Meta-Analysis. PloS one, 11 (7) PMID: 27390857

Amino acids and autism in China

Many happy returns @ Paul Whiteley

Questioning Answers is 2 years old today (19th February 2013). Happy Birthday to 'me', or should that be 'it'?

Still a relative newcomer to the blogosphere but still churning out posts on all things autism research and beyond. Just in case you thought that I did actually bake a cake for the occasion, I didn't. But if I had have done (and yes a man can make a cake), it would have looked like the cake shown alongside. So please loyal readers, take an imaginary bite and enjoy.

To task. I've had hold of the short paper by Wen-Jun Tu and colleagues* (open-access) for a few weeks/months but have only now have got round to posting about it. It continues some familiar themes on this blog on (a) the focus on the -omics and application of technologies like mass spectrometry to autism and (b) amino acids revealing some really quite interesting differences in cases of autism vs. not-autism, as they are doing in conditions like schizophrenia and chronic fatigue syndrome also. A bit out of left field but I was also interested to read the paper by Shingyoji et al on plasma amino acid profiles potentially predicting lung cancer too. Wow, these guys get around.

Anyhow. I say it is a short paper but actually the Tu paper is a letter, and although there is relatively little novelty in just looking at amino acid chemistry in autism - what's up and what's down - these days, it does look at autism in quite a different ethnic population (Chinese) compared to quite a lot of the other papers in this area.

Indeed China, as well as emerging as a world superpower albeit with some peculiarities, is also starting to put quite a bit more effort into autism as per papers like this one from McCabe** with the very interesting title: Bamboo shoots after the rain... (hence the cute picture of the baby panda shown below looking so inquiring).

The net findings reported by Tu and colleagues reflect a few things:

• Based on quite a small participant group of children diagnosed with DSM-IV autism (n=20) compared with asymptomatic controls (n=20), there were some very distinguishing plasma amino acid results found.
• Tandem mass spectrometry was the main analytical method for determining amino acids complemented by immunoassay for detecting circulating neurotransmitters such as plasma dopamine.
• Levels of some amino acids were significantly elevated (lysine, glutamate - glutamic acid and homocysteine); others were depressed (tryptophan, tyrosine, glutamine) in the autism group compared with controls.



Ailuropoda melanoleuca @ Wikipedia  

• When it came to the level of significance, the biggest group differences were in the elevated levels of leucine (p=0.000 apparently), higher homocysteine (same p-value again) and elevated plasma dopamine (ditto on the p-value).

Of course I don't really need to say too much about these findings that have not already been said. Glutamate and glutamine are already on the autism research radar for quite a few reasons; same goes for homocysteine and it's link into things like methylation and the folate metabolic pathway. Tryptophan is a potentially important one bearing in mind its metabolism into things like melatonin among other things. Leucine? Well think branched chain amino acids and that rather interesting study by Novarino and colleagues*** (see this post) in relation to autism, and one cannot help but wonder if there might be some overlap.

Interestingly, one of my papers on the gluten- and casein-free (GFCF) diet (open-access) gets a mention in the text, with the authors seemingly worried about how their results might be further worsened if and when a GFCF diet is instigated following on from some similar suggestion by Arnold and colleagues****. Indeed this is an issue which has more recently been discussed in the meta-analysis by Sharp and colleagues*****. I'm minded to respond that rather than worry about how things could 'get any worse', a closer inspection of why they have the results they have and indeed, dealing with what they actually found, might be a good starting point, accepting the study by Jim Adams and colleagues (see this post) on what might be achieved by micronutrient supplementation. That and the fact that they reference gastrointestinal (GI) issues as potentially being involved with their results which begs the question: why not try and 'sort out' the GI issues or least one of them?

From the ethnicity point of view, the Tu study is an important one given the overlap between their observations and what has been found in other more Western populations allowing for the genetic, environmental and epigenetic differences that one might envisage and the eternal question of whether autism is presented the same worldwide. Such research actually makes a really good case for doing a little bit more cross-collaborative work among different peoples in different countries with autism, based not just on genetics as seems to have been the case so far, but also more functional biochemistry too.

Because this blog is the big 2 now, and given the association between this age and the word 'terrible', a song which I always thought best encapuslates a toddler tantrum from Nirvana (sorry about the language). Toodle pip.

----------

* Tu WJ. et al. Application of LC-MS/MS analysis of plasma amino acids profiles in children with autism. J Clin Biochem Nutr. 2012; 51: 248-249.

** McCabe H. Bamboo shoots after the rain: Development and challenges of autism intervention in China. Autism. November 2012.

*** Novarino G. et al. Mutations in BCKD-kinase lead to a potentially treatable form of autism with epilepsy. Science. 2012; 338: 394-397.

**** Arnold GL. et al. Plasma amino acids profiles in children with autism: potential risk of nutritional deficiencies. J Autism Dev Disord. 2003; 33: 449-454.

***** Sharp WG. et al. Feeding Problems and Nutrient Intake in Children with Autism Spectrum Disorders: A Meta-analysis and Comprehensive Review of the Literature. J Autism Dev Disord. February 2013.

----------

Tu WJ, Chen H, & He J (2012). Application of LC-MS/MS analysis of plasma amino acids profiles in children with autism. Journal of clinical biochemistry and nutrition, 51 (3), 248-9 PMID: 23170055

Metabolomics and Chronic Fatigue Syndrome

There were lots of things I could have blogged about in this post. The recent EEG autism biomarkers study by Duffy and Als* (full-text) which seems to be generating lots and lots of interest, despite the fact that we've kinda been here before (see this post on the Dark Arts); the interesting case study reported by Sildorf and colleagues** on remission without insulin therapy following use of a gluten-free diet in a pediatric case of type-1 diabetes; or even the high prevalence of vitamin D deficiency in psychiatric in-patients reported by Menkes and colleagues*** (full-text). All noteworthy findings and very much within the remit of this blog and its previous posts.

But instead, I'm turning my attention to another paper by Armstrong and colleagues**** and their application of a field close to my research heart, turning the scientific eye of metabolomics to Chronic Fatigue Syndrome (CFS).

Aside from a metabolomics link and the use of some quite powerful analytical technology, NMR, which has been previously applied to both autism (here) and schizophrenia research (here), this study also caught my eye because of what they reported finding.

A very brief summary:

• Blood samples from a small patient group (n=11) diagnosed with CFS were analysed and compared with a small asymptomatic control group (n=10) via NMR. These participant groups are quite small but bear in mind the task ahead of the researchers given the number of potential compounds present in a blood sample (see this paper by Psychogios and colleagues***** on how many they found in human serum). I hasten to add that even in our lab we have found considerably more compounds to be present in other mediums like urine.
• Significant reductions in the amino acids glutamine and ornithine were detected in the CFS group compared with controls. These findings also correlated with other metabolites linked to glucogenic amino acids and metabolites of the urea cycle.
• I believe this to be only one of a handful of papers (that I can find) that discusses the application of techniques like NMR to CFS. On other occasions some interesting bacterial findings have been reported as per this paper by Sheedy and colleagues****** who I think might have been part of the same group (including Dr Henry Butt and the whole CFSUM1 and CFSUM2 episode).

My attention was immediately drawn to the reductions in glutamine which were present in this small participant group. Drawn because of the previously discussed 'possibility' of gut hyperpermeability in cases of CFS and the various suggestions that glutamine might play a role in intestinal permeability, or at least as an aid to improving such permeability in both animals (here) and humans (here). I might also add that lower levels of plasma glutamine have also cropped up in autism research too, bearing in mind that no direct connection between the conditions is intended. 

The literature on glutamine and CFS is still a little sparse. Aside from this paper looking at glutamine levels being related to something called 'overtraining syndrome' there's really not that much more to compare with. I note that exercise can induce changes to intestinal permeability (here) but wouldn't like to speculate on the mechanisms of this effect and any role for glutamine. I am on purpose also excluding the findings on glutamine related to ancillary conditions like fibromyalgia such as a raised glutamate/glutamine ratio (now where have I seen that before?)

I'm going to stop there with this interesting area of research which requires replication with much greater numbers. I've not really discussed the ornithine findings reported, simply because (a) there is even less research on this with CFS in mind, and (b) ornithine in relation to the urea cycle gets really, really complicated and my head is starting to hurt. Suffice to say that amino acid chemistry once again reveals itself as a possible correlate to another one of our heterogeneous medical syndromes without clues to causation. The question is: is the link with glutamine causation, association or just epiphenomenal?

----------

* Duffy FH. & Als H. A stable pattern of EEG spectral coherence distinguishes children with autism from neuro-typical controls - a large case control study. BMC Medicine. June 2012.

** Sildorf SM. et al. Remission without insulin therapy on gluten-free diet in a 6-year old boy with type 1 diabetes mellitus. BMJ Case Reports. June 2012

*** Menkes DB. et al. Vitamin D status of psychiatric inpatients in New Zealand's Waikato region. BMC Psychiatry. June 2012.

**** Armstrong CW. et al. NMR metabolic profiling of serum identifies amino acid disturbances in Chronic Fatigue Syndrome. Clinica Chimica Acta. June 2012.

***** Psychgios N. et al. The human serum metabolome. PLoS ONE. 2011; 6: e16957.

****** Sheedy JR. et al. Increased d-lactic Acid intestinal bacteria in patients with chronic fatigue syndrome. In Vivo. 2009; 23: 621-628.

The glutamate:glutamine ratio and autism

There they go again.

An Athenian butterfly @ Paul Whiteley

That Saudi Arabian autism research group who figured so heavily on this blog in 2011 have yet another publication out in this one by Abu Shmais and colleagues* (full-text) examining nitrogen-related parameters and autism. I'll admit that I'm a fan of this group; not only for the quantity of research that they are producing but the fact that there is clear evidence of 'blue-sky' thinking in their research material. These chaps are quietly contributing to some really interesting avenues in some cases of autism spectrum conditions, albeit at the moment based on quite small participant numbers.

Their latest offering asks a few interesting questions but before summarising some of them, I just want to run a sentence by you from the paper's introduction: "A recent epidemiological study in Saudi Arabia established the autism prevalence at 6:1000 (Talat; unpublished data, personal communication)". Bearing in mind that there are some important differences between personal communications about unpublished data and the public communication of the peer-reviewed variety, lots and lots of questions immediately start to appear in my mind about how this figure was arrived at, what criteria for autism used and if accurate, why the slight discrepancy with the [less than] 1 in 100 figure usually cited for the UK, US, Canada, etc. (or 1 in 38 if you believe the SK study last year). Answers on a postcard please addressed to...

Back to the paper:

• The analysis of nitrogen containing compounds and their metabolism makes up the main reason for this paper so inevitably compounds such as ammonia (NH3) get a mention. More specifically how the the body harvests nitrogen from lots of different sources (including those funny amino acids which I seem to been talking a lot about recently) and consequently disposes of it. 
• In terms of nitrogen metabolism, and concepts like the nitrogen balance, some familiar names crop up noted previously in autism research including glutamate and glutamine and how these compounds are involved in the removal of ammonia from the brain in the absence of important urea cycle enzymes.
• Fasting blood samples were collected from 20 participants with autism and compared with 20 controls. There is a bit of typo in the methods section which states that controls were all male and in the next breath, 16 males and 4 females(!) 
• Based on analyses looking at enzyme activity (5'-nucleotidase, adenosine deaminase and glutamate dehydrogenase), ammonia, urea, creatine, nitric oxide (NO), GABA, glutamate and glutamine - catch my breath - there were a few interesting results.
• Mean plasma levels of creatine, glutamate and GABA were elevated in autism compared to control means. By contrast plasma glutamine levels were lower. Looking at the values and standard deviations, these differences seemed pretty pronounced particularly across the glutamate:glutamine ratio. The results coincided with significant increases in 5'-nucleotidase and a marginal mean decrease in adenosine deaminase activity (p=0.048).
• A little bit of statistical wizardry (ROC) applied to the results suggested that the glutamate:glutamine ratio was the best measure of determining autism samples from controls based on a ratio cut-off value of 0.906. 

So what does all this tell us? Well, lots of things; some perhaps more important than others bearing in mind the small participant numbers. I'm going to focus specifically on the glutamate - glutamine findings as per the title of the post because the results are cumulatively most significant, as once again amino acid chemistry red flags are raised. I should also point out that plasma levels of glutamate and glutamine might not necessarily be the same as 'brain levels' as per the blood-brain barrier (BBB) and its amino acids transporters [note: I have a post scheduled soon on P-glycoprotein and BBB transport].

Not for the first time have elevated levels of glutamate and lower levels of glutamine been reported in autism research. The paper by Shimmura and colleagues** (open-access) found similar results in their analysis and once again pointed to the potential discriminatory power of this finding in cases of autism independent of learning disability from asymptomatic controls. Even Jim Adams and colleagues have reported similar trends (at least with regards to glutamate) on top of other group findings. As for searching the brain, well there is some evidence of issues there too at least in certain areas of the brain.

Fine, the result is pretty consistent but what does it mean?

Most commentary on glutamate reverts back to it being an excitatory neurotransmitter with reference to the hyperglutaminergic hypothesis of autism. I don't really want to go to heavily into the details of the hypothesis in this post but alongside elevated levels of glutamate, there is a corresponding decrease in activity of enzymes linked to glutamate metabolism (GAD65 & 67) and increased gliosis. Only a few days ago, a new paper*** was discussing the presence of autoantibodies to GAD65 in a subgroup of children with autism (and ADHD) which is probably not good news either.

The glutamine findings are also quite interesting. Abu Shmais and colleagues talk about how glutamine is an essential part of the process removing ammonia from the brain via processes such as glutamine synthetase. Low levels of glutamine may potentially reflect issues with this process. Outside of this however, other potential roles for glutamine need to be highlighted seemingly independent of the brain, in the liver and gut for example. The evidence for a connection between glutamine and glutathione (GSH) is also at the back of my mind bearing in mind the evidence so far on GSH in relation to autism.

A post on a sister blog discussed the issue of glutamine and its relationship to some gastrointestinal findings with its simplified role as the 'food of the gut' in terms of things like its effect on gut barrier function. It is therefore conceivable that low plasma levels of glutamine are probably not going to be particularly good for maintaining a healthy gut barrier; even outside of other confounding issues such as sulphation, etc. It might seem like a peripheral finding but more and more there is a realisation that physiology and psychology are linked as per the recent constipation - language impairment findings in autism.

So, another finding which requires large scale replication and another possible 'marker' for autism research is discussed. As time goes on, I am really starting to appreciate just how important amino acid chemistry is to lots of different functions and states outside of the classsic phenylalanine and PKU direction.

To finish something a little bit different as I ask the question: are you a man or a muppet (or perhaps neither)?

----------

* Abu Shmais GA. et al. Mechanism of nitrogen metabolism-related parameters and enzyme activities in the pathophysiology of autism. Journal of Neurodevelopmental disorders. February 2012.
DOI: 10.1186/1866-1955-4-4

** Shimmura C. et al. Alteration of plasma glutamate and glutamine levels in children with high-functioning autism. PLoS ONE. October 2011.
DOI: 10.1371/journal.pone.0025340

*** Rout UK. et al. Presence of GAD65 autoantibodies in the serum of children with autism or ADHD. European Child & Adolescent Psychiatry. February 2012

----------

Abu Shmais GA, Al-Ayadhi LY, Al-Dbass AM, & El-Ansary AK (2012). Mechanism of nitrogen metabolism-related parameters and enzyme activities in the pathophysiology of autism. Journal of neurodevelopmental disorders, 4 (1) PMID: 22958401