"findings suggest a protective effect of CRP" on schizophrenia risk?

Science is often a puzzling endeavour. Sometimes, just when you think that you've got something nailed down, scientific results appear that 'trash' long held, cherished beliefs. So it was with the publication of the results by Fernando Pires Hartwig and colleagues [1] who presented findings looking at "the effect of inflammatory markers on schizophrenia risk" based on the use of "a mendelian randomization (MR) design."

MR, by the way, is a interesting technique based on the principle that "genetic variants that either alter the level of, or mirror the biological effects of, a modifiable environmental exposure that itself alters disease risk should be related to disease risk to the extent predicted by their influence on exposure to the environmental risk factor." It's a technique that has already been applied to inflammatory markers in the context of schizophrenia on more than one occasion (see here and see here). Those inflammatory markers studied have included ones which were covered by the Hartwig paper. Specifically: "Genetically elevated circulating levels of C-reactive protein (CRP), interleukin-1 receptor antagonist (IL-1Ra), and soluble interleukin-6 receptor (sIL-6R)." I should also point out that Hartwig and colleagues have some research form in the area of applying MR to various aspects of medical science (see here).

As per an accompanying editorial on the Hartwig paper [2], the long-and-short of it was that researchers "used 2-sample MR to test for a potentially causal relationship between inflammation and schizophrenia and to improve inference for the association between genes, inflammatory biomarkers, and risk of developing schizophrenia." I can't claim any specific expertise in the use of MR (see here for a good overview [3]) but it appears that data on single-nucleotide polymorphisms (SNPs) in relation to those inflammatory markers was used to test whether said markers might be linked 'causally' to risk of schizophrenia. Their results were interesting: "we did not find strong evidence that lifelong exposure to increased action of these proinflammatory cytokines increases schizophrenia risk, as previously hypothesized" and indeed that: "blockade of IL-6 effects and low CRP levels might instead increase schizophrenia risk." This is contrary to quite a lot of other research in this area (see here for example).

There are a few words of caution to attach to the Hartwig results that need mentioning not least the primary tenet on which analyses are based: genetic variants (SNPs) affecting something like CRP are of primary importance to schizophrenia. I don't for example, see anything in the data looking at gene function/expression being affected as a result of non-structural changes to the genome via something like epigenetic 'alterations' for example (and there is such a thing as epigenetic Mendelian randomization y'know). I say this on the basis that other genes involved potentially involved in processes linked to DNA methylation have also been *associated* with cases of schizophrenia (see here). The authors also caution that their analyses are based on "lifelong exposure to elevated cytokine and CRP levels" and that exposure during 'critical windows' might be the important issue when it comes to any change in schizophrenia risk. Similarly they note that "it is possible that IL-6 and CRP effects on schizophrenia risk are related to a maternal effect (eg, maternal susceptibility to infections during pregnancy), so that our findings are explained by the correlation between maternal and offspring genotypes." This final point is based on the idea that maternal infection during pregnancy (or the biological consequences of) seems to be quite a big risk factor for at least some presentations of schizophrenia (see here) as well as [cautiously] other labels (see here). Here, the importance of a reprogrammed immune system during pregnancy might also come into play alongside any maternal 'susceptibility'.

Personally I'm not yet ready to totally trash the idea that the immune system, and specifically elevations in inflammatory markers such as CRP and other pentraxins, might not be important to some schizophrenia risk in a more detrimental way. I appreciate that one has to be careful when talking about immune system markers and their inflammatory direction (see here for some chatter on IL-6 and its pro- and anti-inflammatory natures) but the existing data is too evident to just discard on the basis of one new study, despite it's scientific prowess. I don't however doubt that there may be several confounding variables linked to increases in CRP in schizophrenia; not least the impact of something like increased body mass index (BMI) that seems to follow some cases of schizophrenia [4]. These variables need to be further explored, particularly in the context of what side-effects pharmacological management of schizophrenia might have (see here). And I also hat-tip the paper by Manu and colleagues [5] applying the Bradford Hill's guidelines on 'causation' to this area and concluding that (upto 2014) "there is insufficient evidence that the replicated, strong association between schizophrenia and elevated inflammatory markers has etiopathological relevance"...

For now however, the Hartwig findings reiterate that science is an ever-changing, ever-evolving process...

----------

[1] Hartwig FP. et al. Inflammatory Biomarkers and Risk of Schizophrenia: A 2-Sample Mendelian Randomization Study. JAMA Psychiatry. 2017. Nov 1.

[2] Byrne E. et al. Inference in Psychiatry via 2-Sample Mendelian Randomization—From Association to Causal Pathway? JAMA Psychiatry. 2017. Nov 1.

[3] Sheehan N. et al. Mendelian Randomisation and Causal Inference in Observational Epidemiology. PLoS Med. 2008; 5(8): e177.

[4] Fernandes BS. et al. C-reactive protein is increased in schizophrenia but is not altered by antipsychotics: meta-analysis and implications. Mol Psychiatry. 2016 Apr;21(4):554-64.

[5] Manu P. et al. Markers of inflammation in schizophrenia: association vs. causation. World Psychiatry. 2014 Jun; 13(2): 189–192.

----------

Meta-meta-analysing MTHFR and autism

"In conclusion, [the] present meta-analysis strongly suggested a significant association of the MTHFR C677T polymorphism with autism."

So said the findings reported by Vandana Rai [1] as yet more discussion emerges on the possible role of issues with the methylenetetrahydrofolate reductase (MTHFR) gene in relation to at least some autism. The reason I've titled this post as a 'meta-meta-analysis' is because we've previously seen meta-analysis done on this polymorphism (SNP) in relation to autism as per other entries on this blog (see here). That and other entries will also provides readers with a little more background on what MTHFR does and why it might be so important to some autism.

This time around Rai looked at 13 studies hitting the criteria for study entry covering nearly 2000 people diagnosed on the autism spectrum compared with over 7000 asymptomatic - not autism - controls. Looking at the various genetic combinations based on zygosity, the author concluded that in both Caucasian and Asian populations, there was an association between the C677T polymorphism and autism.

Where next you might ask? Well, a few possible directions are potentially indicated. First is the idea that screening for the MTHFR C677T SNP does seem to be indicated when a diagnosis of autism is received. By saying that, I'm not suggesting that science has discovered 'a gene for autism' or anything like that (not unless you, for example, count some cases of schizophrenia within that definition of autism as per other work from Rai [2]) but alongside other genetic issues associated with some autism (see here), there is a possible target gene to look at/for. Additional screening of extended family members such as parents and siblings for this SNP might also be similarly indicated, not least because of the various other 'conditions' linked to this SNP [3] and implications for preventative treatment targeting elevated homocysteine for example.

Second, as and when issues with MTHFR are detected, there are some potentially important implications for things like folic acid metabolism (yes, that has been linked to some autism too with some caveats) and potentially onwards the availability of things like methyl groups for important processes such as DNA methylation (again, something looked at with autism in mind). I might also mention some of the literature on homocysteine specifically related to autism might also be something linked here (see here) given the positioning of this compound in relation to the folate and methionine metabolic cycles. Although still in its infancy, talk of MTHFR issues when identified as being part of a 'personalised medicine' approach to autism (see here) provide an important thinking/talking point.

I'd like to think that the MTHFR C677T SNP might offer some quite important clues to at least some types of autism as and when further research is undertaken. With the continuing advances being made using the CRISPR-Cas9 gene editing tool and the ever-increasing CRISPR zoo, the modelling of the MTHFR C677T SNP is set to become quite a bit easier in times to come, and may eventually provide some important advances pertinent to autism and beyond.

Yet again, screening is the first port of call as and when a diagnosis of autism is [eventually] received...

To close, following the very sad news that Ronnie is now with Ronnie, I think it's appropriate to light four candles...

----------

[1] Rai V. Association of methylenetetrahydrofolate reductase (MTHFR) gene C677T polymorphism with autism: evidence of genetic susceptibility. Metab Brain Dis. 2016 Mar 8.

[2] Yadav U. et al. Role of MTHFR C677T gene polymorphism in the susceptibility of schizophrenia: An updated meta-analysis. Asian J Psychiatr. 2016 Apr;20:41-51.

[3] Rajagopalan P. et al. Common folate gene variant, MTHFR C677T, is associated with brain structure in two independent cohorts of people with mild cognitive impairment. NeuroImage : Clinical. 2012;1(1):179-187.

----------

Rai, V. (2016). Association of methylenetetrahydrofolate reductase (MTHFR) gene C677T polymorphism with autism: evidence of genetic susceptibility Metabolic Brain Disease DOI: 10.1007/s11011-016-9815-0

Vitamin D metabolic gene variants and risk for autism

I was really rather happy to see the "preliminary evidence" reported by Rebecca Schmidt and colleagues [1] when it came to examining whether selected vitamin D metabolic gene variants might show linkage to autism spectrum disorder (ASD) based on data derived from the CHARGE initiative.

For quite a while now I've discussed the various peer-reviewed science on the topic of vitamin D deficiency / insufficiency with autism in mind on this blog (see here and see here for example). Specifically, how a diagnosis of ASD seems to offer little protection against issues with vitamin D appearing and what that could mean for important issues such as bone health (see here) for example.

A key component that seemed to be missing from the growing volume of research looking at vitamin D and autism was some discussion about whether the genetics of vitamin production and usage might offer some further clues to how vitamin D might more directly be 'linked' to [some] autism. Schmidt et al have started to put some flesh on to the scientific bones in this area following their previous research discussions on vits and SNPs with autism in mind (see here) .

So: "Maternal, paternal, and child DNA samples for 384 (81%) families of children with ASD and 234 (83%) families of TD [typically developing] children were genotyped for: TaqI, BsmI, FokI, and Cdx2 in the vitamin D receptor (VDR) gene, and CYP27B1 rs4646536, GC rs4588, and CYP2R1 rs10741657." In effect, researchers looked for potential genetic 'issues' with the vitamin D receptor (VDR) gene that have previously been linked to various health issues. They found some potentially interesting information including: "Paternal VDR TaqI homozygous variant genotype was significantly associated with ASD in case-control analysis." Homozygous by the way, refers to the concept of zygosity and the fact we have pairs of chromosomes. Further: "A significant association between decreased ASD risk and child CYP2R1 AA-genotype was found in hybrid log-linear analysis."

This is early days research insofar as the genetics of vitamin D and autism only being mentioned once before in the research literature as per the report from Yan and colleagues [2]. I'm excited at the Schmidt data but am not going to go all out on this very preliminary inspection of vitamin D receptor gene functioning without further large-scale replication and validation studies being carried out including discussions on things like cognitive ability in light of other recent data [3]. Whilst we are however, on the topic of vitamin D and its potential extra-skeletal activities, I'm minded to also bring in the paper by Kaneko and colleagues [4] and their results implying that "vitamin D affects brain serotonin concentrations" with mention of autism among other labels. Reporting on a particularly interesting enzyme - tryptophan hydroxylase (TPH)2 - which has an important role in serotonin metabolism (see here) I'll be watching closely on how vitamin D research with autism in mind also develops in this area.

And then there are the Raftery results [5] to bring to your attention putting further scientific flesh on to the bones about the possibility of a relationship between vitamin D levels and intestinal permeability (see here). Given what has been mentioned about 'leaky gut' and autism in the peer-reviewed literature so far (see here) one might also add this to further investigations in this area...

Music: I am the Monarch of the Sea.

----------

[1] Schmidt RJ. et al. Selected vitamin D metabolic gene variants and risk for autism spectrum disorder in the CHARGE Study. Early Hum Dev. 2015 Jun 11;91(8):483-489.

[2] Yan J. et al. Vitamin D receptor variants in 192 patients with schizophrenia and other psychiatric diseases. Neurosci Lett. 2005 May 20-27;380(1-2):37-41.

[3] Jorde R. et al. Vitamin D and cognitive function: The Tromsø Study. J Neurol Sci. 2015 Jun 7. pii: S0022-510X(15)00350-0.

[4] Kaneko I. et al. 1,25-Dihydroxyvitamin D regulates expression of the tryptophan hydroxylase 2 and leptin genes: implication for behavioral influences of vitamin D. FASEB J. 2015 Jun 12. pii: fj.14-269811.

[5] Raftery T. et al. Effects of vitamin D supplementation on intestinal permeability, cathelicidin and disease markers in Crohn's disease: Results from a randomised double-blind placebo-controlled study. United European Gastroenterol J. 2015 Jun;3(3):294-302.

----------

Schmidt RJ, Hansen RL, Hartiala J, Allayee H, Sconberg JL, Schmidt LC, Volk HE, & Tassone F (2015). Selected vitamin D metabolic gene variants and risk for autism spectrum disorder in the CHARGE Study. Early human development, 91 (8), 483-489 PMID: 26073892

Offspring autism risk and advancing parental age (differences)

Parental age at offspring conception/birth in relation to offspring autism risk has been a recurrent theme in autism research circles for quite a few years now. I've covered it more than once on this blog (see here for example) and the various suggestions that advancing parental age in particular, might elevate the risk of offspring autism.

Set in this context, the paper by Sven Sandin and colleagues [1] (open-access) (a name not unfamiliar to this blog) adds to the research evidence based on their analysis of some 5.7 million children born between 1985 - 2004 resident in one of five countries (Denmark, Israel, Norway, Sweden and Western Australia). Including data on some 30,000 children diagnosed with an autism spectrum disorder (ASD): "Parental ages, sex and birth year were obtained from birth or civil registers."

After quite a bit of statistical modelling and controlling for various potentially confounding variables, several findings were reported pertinent to the authors' data being "the strongest evidence to date supporting the hypothesis that advanced parental ages at the time of birth are independently associated with risk for ASD in the offspring." Outside of "no support for any modification by the sex of the child" researchers also noted a "combined parental age effect" whereby there "was a joint effect of maternal and paternal age with increasing risk of ASD for couples with increasing differences in parental ages."

A few of the finer details of this study have been covered elsewhere (see here). I'll draw your attention to one or two statistics unearthed during the study:

• "relative to fathers aged 20–29 years, fathers 50 years or older had a statistically significantly increased risk for offspring with ASD (RR=1.66 95% CI:1.49–1.85)",
• "Relative to mothers aged 20–29 years, mothers younger than 20 years had a statistically significantly increased risk for offspring with ASD (RR=1.18 95% CI:1.08–1.29)" and 
• the "lowest risk corresponded to couples that generated the majority of births, specifically, 29–39-year-old fathers and 25–35-year-old mothers." 

Those estimates of relative risk (RR) statistics translate into an estimated 66% increased risk for offspring autism if a dad was over 50 years old compared with a dad in their 20s, an 18% increased risk for offspring autism for teen mums compared to 20-something mums and the lowest statistical risk of offspring autism being reported when dads conceive in their 30s coupled with a mid-20 to mid-30 year old mum. The authors also note that "Similar patterns of association, but with slightly higher RRs for the highest parental ages, were evident for AD [autistic disorder]" so completing the message about older parental ages at conception and differing parental ages being relevant across the autism spectrum.

Accepting that this was a huge study in terms of participant numbers and spanning different geographical locations, the authors rightly offer a few words of caution about their methods and data. So: "we lack information about potentially confounding variables such as SES [socio-economic status] and parental psychiatric history" is something to keep in mind [2]. Further: "We cannot rule out the possibility that other factors associated with parental age (for example, length of marriage or partnership, obstetric complications, gestational age and birth weight) have an important role in explaining our results" and "We did not have individual level information on co-morbid ID [intellectual disability] in ASD cases." I'd also suggest that given the growing emphasis on autism or ASD not existing in some sort of diagnostic vacuum (see here) one might reasonably ask whether other comorbidity outside of ID might also play a role in risk estimates.

As to the possible mechanism(s) of effect, well, the authors go through the usual older parents - older sperm and eggs mantra although perhaps bypassing an emerging area outside of just de novo mutations based on the role of epigenetic mechanisms (see here). They do suggest that the 'difference in parental age' factor might suggest "that the increase in risk is not attributable to advancing parental age per se, and that the risk increase cannot be explained solely by an accumulation of point mutations or other genomic alterations in the parents" but say little more on the basis of their collected data.

I might be wrong but I also didn't seem too much in the way of discussion of how parental nutrition might impact on offspring autism risk as per the proposed factor from other work by authors on the Sandin paper in relation to the inter-pregnancy interval (IPI) and autism risk (see here and see here). Although the idea that parental age might affect autism offspring risk, I'd be minded to suggest that this is only the first stage in a journey towards elucidating the particular mechanisms of any effect.

Music: The Pixies - Gigantic.

----------

[1] Sandin S. et al. Autism risk associated with parental age and with increasing difference in age between the parents. Mol Psychiatry. 2015 Jun 9.

[2] Lehti V. et al. Maternal socio-economic status based on occupation and autism spectrum disorders: A national case-control study. Nord J Psychiatry. 2015 Mar 3:1-8.

----------

Sandin S, Schendel D, Magnusson P, Hultman C, Surén P, Susser E, Grønborg T, Gissler M, Gunnes N, Gross R, Henning M, Bresnahan M, Sourander A, Hornig M, Carter K, Francis R, Parner E, Leonard H, Rosanoff M, Stoltenberg C, & Reichenberg A (2015). Autism risk associated with parental age and with increasing difference in age between the parents. Molecular psychiatry PMID: 26055426

Siblings, genetics and the autisms (plural)

The paper by Ryan Yuen and colleagues [1] suggesting that most siblings with autism do not share the same genetic variations thought to contribute to the condition has garnered quite a few media headlines of late (see here and see here).

Applying the concept of whole-genome sequencing whereby the complete genetic blueprint of a person is mapped to provide "the most comprehensive collection of an individual's genetic variation" [2], 340 genomes from 85 families with two children with a diagnosis of autism or autism spectrum disorder (ASD) were analysed. Yuen et al reported that examination of de novo and rare inherited SNPs previously linked to cases of autism were not present in some 70% of their "affected siblings" group. Further, that less than a third of siblings shared the same autism-related gene changes, potentially over-turning the idea that most affected siblings (with the same parentage) share the same genetic issues as being related to their autistic symptoms and label. That all being said, the authors did report that: "Brothers and sisters who shared autism-related mutations displayed more similar symptoms than those who did not".

This is a significant paper in quite a few ways although there are caveats. The results only covered 85 sibling-pairs and the number of autism-related genetic variations inspected was relatively limited. One therefore needs to be a little cautious about sweeping generalisations to the very wide autism spectrum, heterogeneity, comorbidity and all. “The findings suggest that there is significant genetic diversity among people which autism” was one of the commentaries on the Yuen study which I would definitely second (see here). The 'people with autism are like snowflakes' analogy has also been banded around to illustrate that idea of diversity; something that I would also agree with, although perhaps preferring the slightly more scientific idea that 'autism' should perhaps be replaced by the more plural idea: the autisms. Exactly how many 'autisms' there are, remains to be seen as it does in other areas of psychiatry.

Ways forward following the Yuen study? Well, I might suggest that alongside replicating the work in a larger cohort, one might also entertain the idea that structural issues associated with the genome might also be complemented by a little more focus on gene functions and that rising star discipline called epigenetics (see here). The idea for example, that even identical twins might vary in their 'methylomic' profile (see here) is gaining traction in autism research circles to potentially account for some of the missing heritability which has been reported in recent years (see here). I'm not saying that the significant resources ploughed into the genetic roots of autism is all bunk; merely that the idea that genes and environment might synergistically [and variably] act on autism risk should be given a lot more credence, alongside the role that common variants might play in [some] autism (see here). Air pollution is one environmental example perhaps requiring a little more study (see here), although I hasten to stress not the only variable which might need further investigation.

Finally, as part of the Google - Autism Speaks MSSNG initiative (see here), the 'de-identified' data from the Yuen study has been uploaded to the 'cloud' for other researchers to utilise in further investigations. Again, a very good idea for those interested in this branch of autism research but again with the proviso that autism is a very, very heterogeneous condition often including quite a bit of enhanced risk for various comorbidity...

----------

[1] Yuen RKC. et al. Whole-genome sequencing of quartet families with autism spectrum disorder. Nature Medicine. 2015. Jan 26.

[2] Ng PC. & Kirkness EF. Whole genome sequencing. Methods Mol Biol. 2010;628:215-26.

----------

Yuen, R., Thiruvahindrapuram, B., Merico, D., Walker, S., Tammimies, K., Hoang, N., Chrysler, C., Nalpathamkalam, T., Pellecchia, G., Liu, Y., Gazzellone, M., D'Abate, L., Deneault, E., Howe, J., Liu, R., Thompson, A., Zarrei, M., Uddin, M., Marshall, C., Ring, R., Zwaigenbaum, L., Ray, P., Weksberg, R., Carter, M., Fernandez, B., Roberts, W., Szatmari, P., & Scherer, S. (2015). Whole-genome sequencing of quartet families with autism spectrum disorder Nature Medicine DOI: 10.1038/nm.3792

Increasing parental age and autism severity?

An interesting paper by David Geier and colleagues [1] (open-access here) caught my eye recently, concluding that there was a lack of support for the suggestion that: "increasing parental age was associated with increasing autism spectrum disorder phenotypic severity".

"the snozzberries taste like snozzberries".

Before progressing through the paper and its possible implications, the eagle-eyed out there might have already spotted the name Dr Brian Hooker on the authorship list of the Geier paper. Outside of his other peer-reviewed work [2], I probably only need to mention the letters 'CDC' and everything that has [so far] followed including (at the time of writing) a removal statement retraction statement (updated: 04/10/14) for another paper [3]...

Anyhow, the idea behind the Geier paper stems from the quite widely disseminated notion that there may be a connection between increasing parental age at conceiving and an increased risk of offspring autism. I've covered it a few times on this blog (see here and see here). The authors elaborate about a recent hypothesis suggesting that "there must be a linkage between increasing genetic load and increasing parental age in autism spectrum disorder pathogenesis" based on studies like the one from Kong and colleagues [4] (covered in a previous post) and Lampi and colleagues [5]. Further, that as a consequence of an increasing genetic load (all those SNPs et al), "there should be a significant relationship between increasing parental age and increasing autism spectrum disorder phenotypic severity of subjects diagnosed with an autism spectrum disorder".

The paper is open-access but maybe a few details are in order:

• Participants (N=351), diagnosed with DSM-IV autism, were drawn from "patients presenting for outpatient genetic consultations at the ASD Centers, LLC". Mean age was approximately 9 years of age, most male and most reporting developmental regression following birth. Details of age of parents at time of offspring birth were analysed alongside use of the ATEC (Autism Treatment Evaluation Checklist) at initial clinical presentation. These variables formed the crux of the study.
• Results: "Overall, it was observed that no significant relationships were observed between increasing autism spectrum disorder phenotypic severity and increasing maternal or paternal age". Except, that is, for something that seemed to suggest that older maternal age at birth of child seemed to correlates with "improved sociability" in offspring. The authors report that their observations: "provide important insights into the apparent lack of a relationship between increasing parental age and increasing autism spectrum disorder phenotypic severity".

Of course one has to be careful with any study of correlation/association, particularly when it comes to something as simple as just looking at ATEC scores of severity of behaviours in the autism domains and parents age at time of birth of their children. I personally would also have liked to see some further discussion on whether the broader autism phenotype (BAP) for example, might have been an influencing variable too in light of studies like the one from Hasegawa and colleagues [6]. Also, the participant group is quite large - as the authors note - but even there I think back to the sort of sample numbers that those [big data] studies in Taiwan are including (see here) as to where we should be heading.

That all being said, I don't want to downplay the Geier results. Another quote might be useful here: "most observed de novo genetic events are unconnected to an autism spectrum disorder diagnosis, and those that do confer risk are distributed across many genes and are not necessarily sufficient for disease". This ties in rather nicely with the recent discussions on common variations and autism risk (see here) and how Gaugler and colleagues [7] questioned how much weight to give to de novo mutations in the grand scheme of autism 'causation'. This also might imply that non-genetic events, or at least non-structural genetic events headed under the general banner of environment might also play some contributory role to at least some cases of autism. Again, something which has cropped up on this blog before (see here).

Music to close. Given the recent vote near these parts, one of Scotland's most famous exports... Franz Ferdinand and Do You Want To.

----------

[1] Geier DA. et al. An Evaluation of the Effect of Increasing Parental Age on the Phenotypic Severity of Autism Spectrum Disorder. J Child Neurol. 2014 Aug 27. pii: 0883073814541478.

[2] Hooker B. et al. Methodological issues and evidence of malfeasance in research purporting to show thimerosal in vaccines is safe. Biomed Res Int. 2014;2014:247218.

[3] Hooker BS. Measles-mumps-rubella vaccination timing and autism among young african american boys: a reanalysis of CDC data. Transl Neurodegener. 2014; 3: 16.

[4] Kong A. et al. Rate of de novo mutations and the importance of father's age to disease risk. Nature. 2012 Aug 23;488(7412):471-5.

[5] Lampi KM. et al. Parental age and risk of autism spectrum disorders in a Finnish national birth cohort. J Autism Dev Disord. 2013 Nov;43(11):2526-35.

[6] Hasegawa C. et al. Broader autism phenotype in mothers predicts social responsiveness in young children with autism spectrum disorders. Psychiatry Clin Neurosci. 2014 Jun 6. doi: 10.1111/pcn.12210.

[7] Gaugler T. et al. Most genetic risk for autism resides with common variation. Nature Genetics. 2014. July 20.

----------

Geier DA, Hooker BS, Kern JK, Sykes LK, & Geier MR (2014). An Evaluation of the Effect of Increasing Parental Age on the Phenotypic Severity of Autism Spectrum Disorder. Journal of child neurology PMID: 25163730

The schizophrenias (plural)

A micropost if you will, to draw your attention to the paper by Javier Arnedo and colleagues [1] mentioning the concept of 'the schizophrenias' (plural). Some media coverage of this paper can be found here and here. The crux of the paper is that although currently unified by a diagnostic label, schizophrenia seems to be comprised of various conditions: "caused by a moderate number of separate genotypic networks associated with several distinct clinical syndromes".

"... dogs and cats living together... mass hysteria!"

I'm going to say little else about this findings aside from stressing how (a) this re-conceptualisation of schizophrenia into a more plural condition is not a million miles away from moves in other areas of psychiatry (see here) and (b) the reliance on genetic mutation (SNPs) in the paper whilst interesting, perhaps overlooks other non-structural genomic factors potentially implicated in cases of the schizophrenias (see here). That also there may be some 'common ground' between the schizophrenias and other conditions (see here) might also be important for this growing tide of psychiatric plurality.

Music to close. Mr Blue Sky (live).

----------

[1] Arnedo J. et al. Uncovering the Hidden Risk Architecture of the Schizophrenias: Confirmation in Three Independent Genome-Wide Association Studies. Am J Psychiatry. 2014. September 15.

----------

Javier Arnedo, Dragan M. Svrakic, Coral del Val, Rocío Romero-Zaliz, Helena Hernández-Cuervo, Molecular Genetics of Schizophrenia Consortium, Ayman H. Fanous, Michele T. Pato, Carlos N. Pato, Gabriel A. de Erausquin, C. Robert Cloninger, & Igor Zwir (2014). Uncovering the Hidden Risk Architecture of the Schizophrenias: Confirmation in Three Independent Genome-Wide Association Studies The American Journal of Psychiatry : doi:10.1176/appi.ajp.2014.14040435

Homocysteine, MTHFR and schizophrenia studied AND meta-analysed

"Our study suggests that increased plasma total homocysteine levels may be associated with an increased risk of schizophrenia". Further: "The meta-analysis of the Japanese genetic association studies demonstrated a significant association between the MTHFR C677T polymorphism and schizophrenia".

MTHFR (again!) @ Paul Whiteley

So said the results of the study and meta-analysis carried out by Akira Nishi and colleagues [1] (open-access) looking at the 'big H' alongside everyone's genetic Scrabble favourite MTHFR (methylenetetrahydrofolate reductase (NAD(P)H)).

The Nishi paper represents pretty good scientific value for money given that authors not only looked at plasma levels of total homocysteine in nearly 400 participants diagnosed with schizophrenia compared with nearly 1000 controls, they also genotyped for the MTHFR C677T polymorphism [2] (describing an amino acid substitution which reduces the activity of the enzyme methylenetetrahydrofolate reductase and results in elevated homocysteine levels) in a further 1700 participants with schizophrenia compared against over 3000 asymptomatic controls. For good measure, the authors then carried out a meta-analysis of the scientific literature looking at homocysteine and schizophrenia as a function of gender. Phew.

As per the opening paragraph, authors reported "significantly elevated plasma total homocysteine levels in patients with schizophrenia compared with controls, in both male and female subjects". The results of their meta-analysis confirmed such elevations in homocysteine "although antipsychotic medication might influence this outcome". Combined with the association made between a diagnosis of schizophrenia and the MTHFR SNP studied, all adds up to "disrupted 1-carbon metabolism [having] an important role in the pathophysiology of schizophrenia".

I don't mind saying that the Nishi results are really rather interesting to me. The links between schizophrenia and homocysteine have been talked about previously on this blog (see here) including the potential usefulness of folic acid and vitamin B12 for some cases of schizophrenia (see here). Nishi et al also talk about another potentially important part of their results with some mention of DNA methylation, something which also crosses over into other areas of research interest too (see here).

And so the evidence continues to stack up for the big H and MTHFR in some cases of schizophrenia...

Music then. The Pixies and Debaser.

----------

[1] Nishi A. et al. Meta-analyses of Blood Homocysteine Levels for Gender and Genetic Association
Studies of the MTHFR C677T Polymorphism in Schizophrenia. Schizophrenia Bulletin. 2014; 40: 1154-1163.

[2] Gilbody S. et al. Methylenetetrahydrofolate reductase (MTHFR) genetic polymorphisms and psychiatric disorders: a HuGE review. Am J Epidemiol. 2007 Jan 1;165(1):1-13.

----------

Nishi A, Numata S, Tajima A, Kinoshita M, Kikuchi K, Shimodera S, Tomotake M, Ohi K, Hashimoto R, Imoto I, Takeda M, & Ohmori T (2014). Meta-analyses of Blood Homocysteine Levels for Gender and Genetic Association Studies of the MTHFR C677T Polymorphism in Schizophrenia. Schizophrenia bulletin PMID: 24535549

Common variation and the genetics of autism

The paper by Trent Gaugler and colleagues [1] reporting that the genetic architecture of the autism spectrum disorders (ASDs) seems in the most part to be due to "common variation" over and above "rare variants or spontaneous glitches" adds to the quite voluminous literature in this area.

Everything in proportion? @ Wikipedia 

Based on an analysis of "a unique epidemiological sample from Sweden" researchers looked at DNA variations in some 3000 individuals with autism and asymptomatic controls. They were able to model their findings "based mostly on combined effects of multiple genes and non-shared environmental factors" including some "synthesis of results from other studies".

Their results: "Most genetic risk for autism comes from common inherited gene variations that can be found in many individuals without the disorder" as per one write-up of the study results. Spontaneous mutations - those so-called de novo mutations which seem to be of growing interest to autism research - were reported to only 'modestly' increase risk of the condition (2.6% of the total risk). About 40% of the risk was unaccounted for, but combined with those common inherited gene variations, made up about 90% of the total risk or liability for ASD.

Quite a lot of the discussion about these results has focused on the issue of tiny genetic effects which many people not on the autism spectrum have present in their genome adding up into something with "substantial impact" when present together. Other research has hinted at similar things as for example, in the paper by St Pourcain and colleagues [2] looking at the genetics of social communication issues.

Whilst I do think that the Gaugler paper is an important one, I am minded to suggest a few words of caution. First and foremost is the reliance on observed genetic variation in the current paper. Although no expert in genetics, my very basic knowledge is that such variations are structural in nature as per issues like single-nucleotide polymorphisms (SNPs). The presence of such mutations (which we all have by the way, dotted around our genomic landscape) whilst of interest, don't actually though tell you an awful lot about the function of particular genes as a consequence of those point mutations unless further studies are conducted. Genes for example expressing protein can be affected by such mutations but, as we've come to realise in the past few decades, gene expression is also to some degree affected by other variables, as per the rise and rise of the science of epigenetics and the focus on non-structural effects on the genome. It's beyond the scope of this post to go too heavily into epigenetics and autism, but the research forays so far have provided some interesting data on issues like DNA methylation and autism (see here) and potential knock-on effects (see here). Importantly, structural variations might not necessarily be the same, or have the same effects, as epigenetic variations although the two may work synergistically.

Second, and I hate to bang on about this, but autism or ASD does not normally appear in some sort of diagnostic vacuum. As per the Gillberg work on the ESSENCE of autism (see here) or the 'big data' studies from the likes of Kohane and colleagues (see here), not only is autism an extremely heterogeneous condition in terms of presentation, but also a condition more than likely to co-exist alongside some heightened risk of certain comorbidity. It's all well and good saying that cumulative common genetic variants raise the risk of autism but, as per other biomarker discussions, we might very well replace the word autism with something like attention-deficit hyperactivity disorder (ADHD) or epilepsy or even something more somatic along the lines of the various work looking at autoimmune conditions appearing alongside autism. In short, genetic risk might be related to other things outside of just autism or its individual traits, and as I was reminded recently: "correlation is not the same as causation" (thanks Natasa). Oh, and then there is the RDoC initiative to consider...

Finally, it is a glaring omission in quite a bit of the coverage of this paper that the 41% of risk "unaccounted for" does not receive more interest than it has. I don't want to speculate on what might be included in the array of factors involved in this category (outside of my previous chatter on possible epigenetic factors) but will again draw your attention to other work on the old genetics-environment relationship with autism in mind and the question of heritability (see here and see here). That also one media piece talking about the Gaugler study is quoted as saying: "On their own, none of these common variants will have sufficient impact to cause autism" is an important detail which implies both cumulative effects and possibly the input of some external force(s). And those effects may very well cross the nature-nuture debate in some instances as per the results from Mitchell and colleagues talked about in a previous post.

Deciphering the genetic architecture of autism is still very much a work in progress. This latest contribution to the issue is important not least for the conclusions arrived at with talk of an additive model and it's intersection with common genetic mutations present in the general population. That being said, I still want to see more from the discipline. I'd like to see a more comprehensive analysis taking into account both genetic and epigenetic factors crossing environmental contributions too. I'd also like to see more focus on smaller groups on the autism spectrum as a function of things like developmental trajectory (see here) or response to certain interventions (see here). And for those who seem to be using this work as a hammer against environment being related to cases of autism, just remember, there may be many, many routes towards a clinical diagnosis...

----------

[1] Gaugler T. et al. Most genetic risk for autism resides with common variation. Nature Genetics. 2014. July 20.

[2] St Pourcain B. et al. Common variation contributes to the genetic architecture of social communication traits. Mol Autism. 2013 Sep 18;4(1):34.

----------

Gaugler T, Klei L, Sanders SJ, Bodea CA, Goldberg AP, Lee AB, Mahajan M, Manaa D, Pawitan Y, Reichert J, Ripke S, Sandin S, Sklar P, Svantesson O, Reichenberg A, Hultman CM, Devlin B, Roeder K, & Buxbaum JD (2014). Most genetic risk for autism resides with common variation. Nature genetics PMID: 25038753

HLA alleles and specific language impairment

I've talked about the [almighty] major histocompatibility complex (MHC) before on this blog (see here). The important duties that this biological system performs in relation to identification and communication insofar as antigen presentation and importantly, the process of differentiation between 'self' and 'non-self' from an immunological perspective, are not to be sniffed at.

Harold? @ Wikipedia

I've found myself quite interested in all things MHC (also referring to the human leukocyte antigen, HLA, genes) from quite a few different perspectives. Be it with reference to the genetics of coeliac (celiac) disease (see here) or the very preliminary work being discussed on HERVs [human endogenous retroviruses], or rather HERV proteins, as being possible superantigens in relation to a condition such as myalgic encephalomyelitis (ME) (see here), there's lots to think about when it comes to the MHC.

So when I stumbled across the paper by Ron Nudel and colleagues [1] (open-access here) talking about "a possible role for HLA loci in language disorders" my attention was piqued. Before progressing into some of the details behind the Nudel paper, it's important to note the authorship team involved in this research paper. Alongside some quite well-known names attached to autism research (see here and see here for example), I was also interested to see Gillie Baird as one of the research team. Prof. Baird has impressed me in recent times; not only with her research involvement looking into the whole functional bowel issues in autism research (see here) but also with that recent 'leaky gut' paper with autism in mind (see here) which despite some methodological 'issues' gave me cause to smile that this area was at last starting to be taken seriously in more orthodox circles. Who knows, one day even the NICE guidance might recognise it as a potentially important comorbidity?

No mind, a few details from the Nudel paper are noteworthy, accepting that the paper is open-access:

• This was a multi-centre study looking at "the possible involvement of HLA loci in SLI [specific language impairment]". Loci by the way, is the plural of locus, and with genes in mind, refers to specific locations on a gene, DNA sequence or chromosome; a sort of area or postcode if you will. Participants were mostly derived from the SLIC (SLI consortium) groups with additional families included. All were assessed as having a SLI; all did not have autism, but some had comorbid features of ADHD and/or dyslexia. A control group was also included for some of the analyses.
• OK here's where is starts to get a little bit complicated. I found quite a nice website talking about the ways and means that HLA genes are indexed in terms of nomenclature. The complexity of the HLA genes and their molecules derives from the fact that HLA genes are highly polymorphic (see here). Regular readers might have heard me talk about SNPs (Single Nucleotide Polymorphisms) before reflecting single changes to the genetic alphabet. In the case of HLA genes it gets a little more complex (see here) and, as such, specific terminology has been introduced. I'll also draw your attention the description of an allele too (see here). 
• So, first researchers went looking for SNPs "across the HLA [genetic] region". Further, they then imputed HLA type from the SNP data to see if there was any connection between HLA type(s) and SLI or the specific parts of SLI from a psychometric point of view.
• Results: The HLA-A locus seemed to show a possible connection for "susceptibility to SLI". The A1 allele (HLA-A A1) in particular came out as the "most highly positively correlated allele". When it came to looking at SLI cases vs. controls, the "DR10 allele of HLA-DRB1 was more frequent in individuals with SLI than population controls". There are some other findings reported in terms of parental inheritance patterns but I'm going to leave those for now.
• The authors conclude that their preliminary data requires further study but: "provide an intriguing link to those described by previous studies of other neurodevelopmental disorders suggesting a possible role for HLA loci in language disorders".

I'm sure you can see why I was so interested in this paper. The bigger picture of the HLA (MHC if you prefer) being potentially implicated in something other than just immune function and all those 'self' and 'non-self' duties, opens up a whole new world of research. Given also my increasing interest in all things autoimmunity with an autism research slant, I'm minded to suggest that this may go some much further than anyone ever suspected [2]. That being said I would direct readers to the graphs showing the frequencies of alleles across the different HLA genes to see how aside from the HLA-DRB1 DR10 allele, this is more of a 'shades of grey' over and above 'there or not' pattern of findings when comparing SLI and non-SLI groups.

Looking back at some of the quite limited autism research in this area, I did find some clues that HLA involvement in SLI should perhaps not be totally unexpected as a function of the potential overlap between autism and SLI [3]. I say this acknowledging the debates are still on-going about the links or not between the two conditions and also that Nudel and colleagues ruled out autism as a diagnosis in their particular cohort. As an aside, readers might also find a recent post by Dorothy Bishop of interest, and in particular her model C for language development.

HLA alleles and autism? Well, you might say this is an emerging area in light of results like those from Mostafa and colleagues [4]. I dare say there may be other HLA overlap with other findings from autism research (see the open-access review from Torres and colleagues [5] for further reading). More recently I also stumbled across the paper by Al-Hakbany and colleagues [6] (open-access) which continued discussions about overlapping alleles (I'll probably formulate a separate post on that research soon).

The late Reed Warren also talked about the null allele of C4B [7] (see here for some more explanation on this) quite a few years back now, and as intimated in a past blog entry on his later work, reading disorders and ADHD actually turned up as being more relevant to the null allele than autism did (67%, 56% and 48% respectively). OK SLI is not necessarily reading disorder or ADHD, but as Nudel and colleagues noted there is a "high degree of co-occurrence for SLI and ADHD or dyslexia".

I think it's also important to recognise that those HLA types talked about in the Nudel study aren't just limited to behavioural or psychiatric conditions. I'm not going to go through the myriad of conditions and clinical findings related to the HLA alleles identified, but suffice to say that we perhaps need to look at a bigger picture when interpreting this data. This includes whether there may be other comorbidity present in those SLI cases - including somatic comorbidity - which might link into HLA genotype as per previous work looking at coeliac disease and ADHD for example (see here). SLI like autism like every other behavioural, developmental or psychiatric condition does not just exist in a diagnostic vacuum.

Still, I'll say again that this is a potentially very important paper which potentially opens up a whole new world of research...

Music. The life and times of Nelson Mandela will always remain in the public consciousness. I always fondly remember the Special AKA tribute to the great man.

----------

[1] Nudel R. et al. Associations of HLA alleles with specific language impairment. J Neurodev Disord. 2014 Jan 17;6(1):1.

[2] Choudhury N. & Benasich AA. A family aggregation study: the influence of family history and other risk factors on language development. J Speech Lang Hear Res. 2003 Apr;46(2):261-72.

[3] Bishop DV. Autism and specific language impairment: categorical distinction or continuum? Novartis Found Symp. 2003;251:213-26

[4] Mostafa GA. et al. The link between some alleles on human leukocyte antigen system and autism in children. J Neuroimmunol. 2013 Feb 15;255(1-2):70-4.

[5] Torres AR. et al. HLA Immune Function Genes in Autism. Autism Res Treat. 2012;2012:959073.

[6] Al-Hakbany M. et al. The Relationship of HLA Class I and II Alleles and Haplotypes with Autism: A Case Control Study. Autism Res Treat. 2014; 2014: 242048

[7] Warren RP. et al. Increased frequency of the null allele at the complement C4b locus in autism. Clin Exp Immunol. 1991 Mar;83(3):438-40.

----------

Nudel R, Simpson NH, Baird G, O Hare A, Conti-Ramsden G, Bolton PF, Hennessy ER, Monaco AP, Knight JC, Winney B, Fisher SE, & Newbury DF (2014). Associations of HLA alleles with specific language impairment. Journal of neurodevelopmental disorders, 6 (1) PMID: 24433325

Put it in the (autism science) replicator

Replication is one of the most important processes in those things we call science and the scientific method. Outside of conjuring up images of a certain 'Make it so' Captain with his "Earl Grey, Hot", scientific replication provides the community at large with some degree of reassurance that a research finding was not just a fluke or an artefact of a particular sample of people or a method used or an interpretation of results. As per the recent BBC article on vitamin use: "Looking at any one individual study won't be very revealing to answer the question of whether vitamin supplementation is good for you."

Uncanny likeness @ Wikipedia 

With the issue of replication in mind, I was interested to read the Letter to the Editor from Robinson and colleagues* (open-access) who set about trying to replicate the findings from Skafidas and colleagues** (open-access) and their notion that science might be making some in-roads into the detection of "genetic biomarkers [that] can correctly classify ASD from non-ASD individuals". I posted about the Skafidas study at the time also (see here) and their analysis of single-nucleotide polymorphisms (SNPs) in relation to autism spectrum disorder (ASD).

The Robinson letter reports an attempt to replicate the Skafidas findings based on an independent analysis of data from the Psychiatric Genomics Consortium (PGC) "which includes ~5400 cases, more than three times the number used in the original [Skafidas] report". I'm not on this occasions going to get the fine-toothed comb out on both papers because they're open-access so free for anyone to read.

The conclusions from Robinson et al are pretty clear: "We find no evidence that the implicated SNPs, the classifier or the pathways named in Skafidas et al.1 are associated with ASDs. We therefore conclude that the classifier, as presented, cannot be used in a general way to predict ASDs, and consequently is unlikely to have any translational value."

Obviously such findings are both a blow to autism research and also the original authors who first proposed the classifier model, who I don't doubt probably invested quite a lot of time, effort and funds into getting their experiments done and results published (and published in a Nature journal). The ego also takes a bit of a knock under such circumstances, believe me (see here and here and here).

The Robinson data however re-emphasize the importance of replication in autism research. Perhaps just as important, they also reaffirm that autism is a tremendously difficult set of conditions to study. As is often the case when it comes to a heterogeneous condition like autism (or should that be the autisms) often carrying more than its fair share of comorbidity (see here) including risk of certain somatic conditions (see here), consistent findings are often few and far between. Indeed, that the use of the label autism, whilst providing a way of classifying certain types of behaviour and their impact on a person's life, is not necessarily the best thing for research purposes, as was vocalised through the grudge match that was DSM V vs. RDoC (see here).

The added realisation that outside of no one single SNP being linked to all autism (see here) there may be a significant degree of overlap when it comes to the genetics of the autisms with other developmental and psychiatrically defined conditions (see here) implies that it's going to be some time yet before any genetic biomarkers or test is going to be able to accurately classify autism, sorry the autisms with any great accuracy. Then there is the question of what such a test would accomplish. I've not even mentioned the fact that autism, whilst having a genetic component, is probably not without it's [variable] partner in crime, environment (however you want to define this) when it comes to aetiology. And don't even mention that other area of increasing interest, epigenomics (see here)... which in recent days has seen some interesting papers published (see here and see here).

I suppose in the spirit of all this talk on replication, the last question should be: who next is going to try and replicate the Robinson results? Indeed, does science any longer need the 'Letter to the Editor' in light of the rolling out of PubMed Commons?

----------

* Robinson EB. et al. Response to ‘Predicting the diagnosis of autism spectrum disorder using gene pathway analysis’. Molecular Pyschiatry. 2013: Oct 22. doi: 10.1038/mp.2013.125

** Skafidas E. et al. Predicting the diagnosis of autism spectrum disorder using gene pathway analysis. Molecular Psychiatry. 2013; Sep 11. doi: 10.1038/mp.2012.126

----------

E B Robinson, D Howrigan, J Yang, S Ripke, V Anttila, L E Duncan, L Jostins, J C Barrett, S E Medland, D G MacArthur, G Breen, M C O'Donovan, N R Wray, B Devlin, M J Daly, P M Visscher, P F Sullivan, B M Neale (2013). Response to ‘Predicting the diagnosis of autism spectrum disorder using gene pathway analysis’ Molecular Psychiatry DOI: 10.1038/mp.2013.125