Thursday 13 September 2012

MTHFR and autism reloaded

MTHFR @ Paul Whiteley
I've talked before about the enzyme methylenetetrahydrofolate reductase, or MTHFR to its friends, and the suggested link with autism.

A few points are worth remembering:

  • MTHFR serves a primary function in reducing the compound 5,10-methylenetetrahydrofolate to 5-methyltetrahydrofolate. 
  • On a grander scale, 5-methyltetrahydrofolate, the reduced and methylated form of folic acid, is an important methyl group donor for the recycling of homocysteine back to methionine utilising vitamin B12 along the way. 

'What's the big deal?' you might be thinking to yourself, as hopefully you are still reading this post. Well, there may be a few consequences of disruption along this pathway: elevated levels of homocysteine (discussed in this post), reduced levels of S-adenosylmethionine (SAMe), etc.

If I had to boil it down to a single important effect, it would probably be that reductions in the levels of SAMe might have some profound effects on methylation patterns in DNA (see here). Indeed with autism in mind, methlylation capability has been questioned in both people with autism (here) and their parents (here) bearing in mind the need for a lot more investigation in this area and my continued fixation on methylation (see a previous post). So then the MTHFR enzyme serves an important function. 

In my first post about MTHFR and autism, I talked about point mutations, SNPs, of the MTHFR gene (which control production of the enzyme) and how one particular SNP, C677T (whether homozygous or heterozygous) might show some link to cases of autism and onwards to the production of MTHFR (see this example study by Goin-Kochel and colleagues*). That and the recent suggestion that increasing doses of folic acid might affect the 'risk' of offspring autism as a function of maternal C677 status (see here), makes for some interesting observations on this very interesting SNP.

A recent paper by Guo and colleagues** extends the interest in the MTHFR C677T SNP in relation to autism, with their suggestion, quote: "MTHFR C677T is a risk factor of autism in Chinese Han children". 

Guo et al reported the presence of the C677T SNP in 16% of their cohort of children with autism, compared with 8.6% of their asymptomatic controls. This contrast between the groups is pretty similar to that reported by Mohammad and colleagues*** (16.3% autism vs. 6.5% controls) in a different ethnic population. Perhaps also just as interesting was the suggestion by Guo that 'overactivity' based on selected ADI-R scores seemed to be related to the presence of the SNP. A pretty similar finding to that presented by Goin-Kochel* bearing in mind the ins and outs of the ADI-R.

Taken cumulatively, it does appear as though there might be some connection between the C677T SNP, MTHFR and autism, albeit in a relatively small proportion of cases. Before however we get carried away, bear in mind that the same SNP has been linked to lots and lots of other different conditions including type-2 diabetes (here) and hypertension (here) in specific populations. There is even the suggestion that this SNP might be a risk allele for decreased kidney function too (see here full-text). The specificity of the finding is therefore something that requires quite a bit more research and whether such findings might actually indicate a subset of people with autism at potentially greater risk of such conditions either currently or in later life.

Indeed with that 'overactivity' finding in mind - I assume to be related to something like hyperactivity - the results looking at the C677T SNP in relation to conditions exemplified by hyperactivity such as ADHD, are less than impressive at the current time as per this paper by Ergul and colleagues****. That being said, other psychiatric diagnoses looked at by Ergul et al have shown some possible relationship (here) but I'm going to leave that for another day.

I admit to being slightly less than impressed with the whole area of genetic markers related to autism given the genetic complexity that surrounds the condition and all the additional issues relating to heterogeneity and comorbidity. This despite the recent gene pathway analysis paper from Skafidas and colleagues which, as several people have pointed out, didn't seem to talk too much about false-positives and the like. That's not to say however that genetics may not be an important part of autism but rather as we are starting to realise from the whole epigenetics revolution for example, that genomic changes might not the 'be all and end all'. There's so much more to DNA.

Bearing this in mind however, I'm intrigued by the suggestion that a SNP in MTHFR, most probably with other SNPs in other genes too, might have the capacity to trigger a cascade of biochemical actions which potentially ends up with reduced DNA methylation, and likely more issues with the function of DNA in cases of autism as in other conditions. A sort of 'genes causing an epigenetic effect' model if you like, which itself raises quite a lot of important questions about how we are the way we are.

To finish, it's been a while since I offered some musical accompaniment to my posts. To correct this I link to a song which I've been hearing a lot on the radio recently Razorlight and In the Morning.


* Goin-Kochel RP. et al. The MTHFR 677C-->T polymorphism and behaviors in children with autism: exploratory genotype-phenotype correlations. Autism Research. 2009; 2: 98-108.

** Guo T. et al. Methylenetetrahydrofolate reductase polymorphisms C677T and risk of autism in the Chinese Han population. Genetic Testing & Molecular Markers. July 2012.

*** Mohammad NS. et al. Aberrations in folate metabolic pathway and altered susceptibility to autism. Psychiatric Genetics. 2009; 19: 171-176.

**** Ergul E. et al. Methylenetetrahydrofolate reductase gene polymorphisms in Turkish children with attention-deficit/hyperactivity disorder. Genetic Testing & Molecular Biomarkers. 2012; 16: 67-69.

Guo T, Chen H, Liu B, Ji W, & Yang C (2012). Methylenetetrahydrofolate Reductase Polymorphisms C677T and Risk of Autism in the Chinese Han Population. Genetic Testing and Molecular Biomarkers PMID: 22775456


  1. This is interesting to say the least considering I am husband to a homozygous C677T wife and father to a daughter born with Spina Bifida along with hydrocephalus. The methionine cycle is very interesting to me. What do you feel is most important nutritionally for C677T SNP patients? I'm not looking for a rec, but wondering if research has yielded anything regarding supplementation with methylated folate, b-12, or the use of SAM-e supplements themselves.

  2. Thanks for the comment Gabe A.

    I'm being quite careful in what I say here so as not to be giving medical advice (I'm only a lowly Ph.D) as per this blogs guidelines.

    One could argue that everything you mentioned might come in handy where the C677T SNP (homozygous?) has been shown to affect levels/metabolism of 5-MTHF.

    Certainly in autism circles folate has been indicated - see this post:

    bearing in mind potential issues with receptors and autoantibodies:

    B12 and SAMe have has less of a mention at the moment. Sorry I can't say much more. Best advice would be to speak to your healthcare provider.

  3. Father of a kid on the spectrum for whom M-B12 seems to be a great help in controlling anxiety. I'm also heterozygous for C677T.

  4. Many thanks for dropping by Stephen.

    An interesting observation on the B-12 side of things in relation to anxiety.

  5. Hi Paul. Another anecdote: for anxiety, I've noticed that a couple of grams or so of glutamine works wonders. The funny thing is just how fast I've seen it work -- just minutes. I wonder if it even has time to cross the BBB, or whether it just provide gastro-intestinal relief. Obviously, this is not a controlled trial. ;)

  6. Thanks again Stephen.

    Glutamine is a bit of favourite amino acid to this blog covered on a couple of posts here and on a sister blog:

    I found this article (full-text) for a little more information on glutamine transport by the BBB:

    Also, controlled trials are important but never underestimate the power of the N=1....


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