Does X Mark the Spot for ASD?

X-chromosome epigenetics could explain the incidence of autism

Posted Apr 08, 2011

Wikimedia commons
Source: Wikimedia commons

In February last year I published a post suggesting that Asperger syndrome (AS) might be the result of what I called lingering Lyonization. By that I meant the accidentally-retained non-expression of key X chromosome genes derived from the mother. The fact that males get their one and only X chromosome from the mother but that females get one from each parent seemed to explain the highly-male-biased sex ratio in AS. Most daughters would be protected, but occasional skewed expression in favor of an affected maternal X chromosome in critical parts of a woman's brain might produce a result analogous to that in male AS cases.

Today I can update that post and report that my suggestion was brilliantly pre-empted in 2008 by Julie R. Jones and others (Jones JR, Skinner C, Friez MJ, Schwartz CE, Stevenson RE. 2008. Hypothesis: Dysregulation of methylation of brain-expressed genes on the X chromosome and autism spectrum disorders. Am J Med Genet Part A 146A:2213-2220).

The paper compellingly argues the case for an epigenetic basis for autism. In its simplest form, the authors' hypothesis is that dysregulation of X chromosome genes expressed in the brain predisposes to autism spectrum disorders (ASDs) such as AS. They then go on to make the same point that I did in my original post, arguing that males are particularly vulnerable thanks to having only one X chromosome.

However, they also speculate on the epigenetic mechanisms that might be responsible, and raise a possibility that had not occurred to me. This is that the dysregulation might result in the up-regulation of certain genes, "resulting in enhanced production of brain-expressed proteins," as well as in the down-regulation and reduced production which I had envisaged. Such a possibility did not figure in my thinking because, according to the imprinted brain theory, ASD results from the over-expression of paternal and/or the under-expression of maternal and X chromosome genes. The possibility that Jones et al. raise certainly cannot be ruled out as they eloquently argue, but according to the imprinted brain theory enhanced expression of X genes would be much more likely to be associated with psychotic spectrum disorders (PSDs) such as schizophrenia which, according to the theory, have the exact opposite pattern of expression: reduced paternal and/or enhanced maternal and X chromosome gene expression.

Whether this is the case or not remains to be discovered, but recent research on the expression of imprinted and X chromosome genes in the mouse brain sets a number of intriguing precedents. Imprinting (that is, parent-specific gene expression) was found at 1300 places in the mouse genome, with 347 non-sex chromosome genes having sex-specific imprints. This means that their expression was determined by the sex of the individual in which they found themselves, and clearly this could be another factor affecting the sex ratio of incidence of disorders such as ASD if something similar were true of the human brain.

Even more intriguingly, a bias in the expression of the maternal X chromosome was found in the medial pre-frontal cortex (mPFC) of the mouse brain. As I pointed out in the previous and earlier posts, the mPFC is now known to be critical in mentalizing functions in the human brain, and up- or down-regulation of genes expressed in this region is likely to be implicated in both ASD and PSD thanks to the symptomatic hypo-mentalism of ASD and hyper-mentalism of PSD.

Indeed, such an effect might go some way towards explaining the sex-ratio seen in PSDs such as schizophrenia in much the same way that Jones et al. suggest it might in relation to ASD--at least as far as the imprinted brain theory is concerned. The problem here is that the basic symmetry of the model and its links with sex differences suggest that, just as there is a preponderance of males on the autistic side of the mentalistic spectrum, so there should be a female preponderance on the psychotic side. Although there is some evidence that this is the case (especially at the high-functioning end of the psychotic spectrum where, for example, women outnumber men 3:1 in the incidence of borderline personality disorder), severe psychosis seems, if anything, to affect more males than females. However, if the male's single X were once again the factor explaining his greater vulnerability to variations in X gene expression both up as well as down, the paradox would vanish--at least as far as the imprinted brain theory is concerned.

The recent study from which I am quoting (Gregg C, et al. (2010) Science 329: 643-648) concludes that "parental expression bias emerges as a major mode of epigenetic regulation in the brain," just as Jones et al. and--the imprinted brain theory--argue. Today this is a much more widely accepted possibility than it was even three years ago when Jones et al. published their pioneering hypothesis. Whether X chromosome gene expression really can explain the sex-bias in the incidence of ASD remains to be seen, and still more uncertain is the role that X chromosome genes play in PSD. But whatever the outcome, Jones et al are to be congratulated for being the first to publish what may well come to be seen as a classic hypothesis. And of course, if the suggestion regarding PSD made in this post were to prove true, these authors might have explained even more about the epigenetics of mental illness than they originally imagined!

(With thanks to Julie Jones for bringing this matter to my attention and apologies to her and the other authors of this paper for my initial failure to cite their prior publication.)