Saturday, 5 November 2011

Grandparental age and future generation risk

Most people with an interest in autism research will, at some point, have heard about the proposed equation: increasing paternal age = greater risk of autism diagnosis in offspring. I say 'equation', but unlike most mathematical and statistical formulae, the 'certainty' of the advancing parental age hypothesis of autism is actually not as certain as one might have been led to believe from papers like that included in the recent Nature series. Risk is risk not certainty.

On one of the sister/brother/sibling(?) blogs to this one, I linked to a study published last summer (2010) by Golding and colleagues*. The data derived from the ALSPAC study, suggested that parental (paternal and maternal) age effects whilst potentially important to autism might actually be secondary to age effects going further back a generation. In particular grandmother's age came under the spotlight and the revelation that Grandmum's who were over 35 years old at conception/birth of mums of children with autism might be over-represented in the cases of autism identified during the ALSPAC study.

Golding and colleagues were very cautious about their findings and provided some limited speculation of how the 'meiotic mismatch methylation (3M) hypothesis' might come into play. I can't claim to fully understand all of the 3M hypothesis but risk alleles, DNA methylation and the subsequent skipping of a generation are the watchwords. Speculative but interesting.

A more recent study has also emerged discussing similar concepts as potentially being relevant to schizophrenia. The paper by Frans and colleagues** based on another very substantial cohort number suggested again that older dads (particularly those above 55 years of age) were associated with an increased risk of offspring developing schizophrenia. They also found that older maternal grandfathers age (at conception of mums) might be associated with an increased risk of schizophrenia. The authors reason some involvement of the X chromosome 'differentially' being involved.

Whilst I am in no way suggesting that autism and schizophrenia are one and the same, and bearing in mind the multitude of factors, variables and biases inherent to this kind of cross-linking work, I do find it interesting that another set of commonalities might be coming to the research surface. Quite a lot is being made about SNPs, CNVs and such like in research examining developmental, behavioural and psychiatrically-defined conditions but little has been so far offered about how and why such point mutations, deletions and insertions might come about. Yes, they may be spontaneous, they may be age-related and potentially even evolutionary but what is/are the underlying process(es)?

What these collected research suggest is that generationally, there may be good reason to start looking back at extended family as and when possible for some of the genetic clues to conditions like autism. I know that in some cases this might not be possible, but given that autism is a condition manifesting in early childhood, there are good odds that grandparents and beyond might still be around. I hasten to add that I am not talking about looking for any 'eccentric grandfathers' or anything behaviourally-defined at this point following the slew of speculation asking questions like did this person or that person have autism or Asperger syndrome. Rather asking some sensible questions about how things like age at conception and the possibility that other environmental factors which parents, grandparents (and even great grandparents) were brought up in and exposed to might have the propensity to affect the genetic make-up of subsequent generations. Bear in mind also that just because we have only fairly recently been able to catalogue such genetic features in lots of different conditions and asymptomy, does not mean that they haven't been around mutating, adding and subtracting themselves, for whatever reason, for an awful lot longer.

Just in case you need to visualise how generational research could be done, have a look at this paper from a few years back..

* Golding J. et al. Parental and grandparental ages in the autistic spectrum disorders: a birth cohort study. PLoS ONE. April 2010.
** Frans EM. et al. Advanced parental and grandparental age and schizophrenia: a three-generation perspective. Schizophrenia Research. October 2011.