Advancing paternal age and risk of autism

2 Dec

This isn’t the first study to look at paternal age as a possible risk factor for autism but it is, I believe, the first meta-analysis of the subject. The conclusions of the study were:

Based on data from a birth cohort, a family-based study and a meta-analysis, we provide the strongest and most consistent evidence available that advancing paternal age at the time of birth of offspring increases the risk of autism. De novo germline mutations, epigenetic alterations and life course toxic exposure may partly explain the observed association. The evidence is substantial enough to justify a search for the underlying mechanisms in both human and animal models

An interesting conclusion for a few reasons. First and foremost the idea of paternal age being a definite risk factor for autism. Secondly the authors don’t shy away from the idea that ‘life course toxic exposure’ may explain the association. Its not exactly a new observation amongst science (despite what some observers think) but its good to see it placed so clearly amongst the other clear risk factors.

There will be those, I predict, who will have a go at this study for somehow ‘blaming’ fathers/parents. It has happened in the past and will no doubt have the same effect on those who’ll attack this study for their own reasons.

12 Responses to “Advancing paternal age and risk of autism”

  1. daedalus2u December 2, 2010 at 20:26 #

    It would be pretty straightforward to look at the cohort of ASD children born to older fathers and see if they do have genetic abnormalities characteristic of gamete damage. My guess is that they will not.

    They might have epigenetic differences for which there are not tests yet. The genetic testing to date has shown no genetic abnormalities in the vast bulk of ASD individuals, those with familial-type ASDs from multiplex families. The CNVs, deletions, duplications are a minority of ASD cases.

    I suspect that having an older father is a “stress” that manifests itself during pregnancy on the fetus while in utero.

    With an older father, the fetus needs to “hurry up” and grow to maturity while the older father can still protect it. In the “wild”, it was the fetuses of older fathers who had their fathers killed while they were still very young by the new alpha male. Stress in utero tends to accelerate development and stress in utero also causes ASDs.

    It would be pretty straightforward to look at the large genetic studies that have found CNVs and look at age of the father at conception and see if there is a correlation. I haven’t seen that anyone has done this and published it. It is so obvious that it is hard to imagine it not being done. If it has been done and not published, maybe it is because there was no correlation (or maybe an inverse correlation, more autism but less CNVs).

  2. Prometheus December 2, 2010 at 20:48 #

    From a genetic damage point of view, the greatest “life course toxic exposure” is to oxygen. Oxygen (in concert with that other environmental toxin, water) causes over 80% of all DNA damage.

    I suppose that if we limited paternal exposure to oxygen, we could dramatically decrease the risk of autism due to advanced paternal age.

    Something to think about.


  3. RAJ December 2, 2010 at 22:58 #

    Copy number variations are associated with ASD, Schizophrenia, ADHD, language disorders, intellectual disability and even AIDS.

    The CNV’s are primarily de novo which means they are not present in the either parent just as Downs Syndrome is not a heritable condition. No doubt , increasing parental age raises the risk of de novo CNV’s just as increased maternal age raises the risk of Down Syndrome.

    De novo mutations, either in the single gene disorders or in CNV’s, are not present in either parent therefore they cannot be associated with the genetic influences underlying the Broad Autism Phenotype.

    That is entirely consistent with the Rutter Hypothesis which states that the genes underlying the Broad autism phenotype may not be the same as the genetic and environmental factors involved in the transition to the handicapping condition.

    What do CNV’s exactly do to ’cause’ any developmental disorder? The answer is no one knows at this point. The only verified gene x evironment interaction that has been identified comes from AIDS researchers. Lower copy number variations in a single gene, CCL3L1, located on chromosome 17Q.11.2 has been identified as representing substantial risk for infection after exposure to HIV.

    To deny that the environment does not disrupt functioning of a de novo (or inherited) CNV or upon common genetic variances like the serotonin transporter gene which is present in 40% of Caucasions is to deny nature.

  4. Prometheus December 7, 2010 at 21:37 #

    RAJ asserts:

    “To deny that the environment does not disrupt functioning of a de novo (or inherited) CNV or upon common genetic variances like the serotonin transporter gene which is present in 40% of Caucasions is to deny nature.”

    RAJ appears to assuming that autism is the result of a single-gene mutation. Otherwise, why would the association between autism and a common gene variant be such a compelling hallmark of environmental influence?

    In fact, having a common gene variant in combination with a common environmental exposure (e.g. vaccines, mercury, plasticisers, etc.) should be very common. For example, using RAJ’s common gene variant (40% have it) and childhood vaccines (90+% get them), we would expect that at least 36% of the population would have both the gene variant and the environmental exposure.

    Ah, there’s the rub. If autism is the combination of a single gene and environmental exposure(s), the chance of having both cannot greatly exceed the chance of having autism. Most of the environmental exposures I’ve heard proposed as “triggers” for autism are pretty ubiquitous. As a result, the gene variants would have to be relatively rare (about 1 – 5%).

    On the other hand, if we assume that autism is the result of two or more genes acting in concert, then we can have associations between autism and gene variations that are more common than autism. If – for example – the common gene variation was found in 40% of people and the rarer gene variation was found in only 2.5% of people, the chance of having both (assuming no linkage) would be 1% and the common gene variation would still be strongly associated with the combination.

    Now, I’m not saying that there can’t be an environmental trigger for autism, just that an environmental trigger is not required by the available data and the association of common gene variations with autism does not invalidate the genetic-only hypothesis (hypotheses) of autism.

    An interesting corrolary to this is that autism could be the result of an interaction between a common gene variation and a rare environmental exposure. However, none of the popularly espoused environmental exposures fit that description.


  5. daedalus2u December 8, 2010 at 00:29 #

    Prometheus, there is an environmental exposure that does fit that description, low nitric oxide, but I concede that it is not popular.

    The low nitric oxide hypothesis is also consistent with the CNV data, the teratogen data, the familial genetic data, the twin data and all the other data surrounding autism causation and symptomology.

    And, as you should know, the compound that is most protective against the damaging effects of the greatest “life course toxic exposure” is the compound that controls the effects of O2, and also superoxide, nitric oxide.

  6. RAJ December 8, 2010 at 16:12 #

    “RAJ appears to assuming that autism is the result of a single-gene mutation. Otherwise, why would the association between autism and a common gene variant be such a compelling hallmark of environmental influence”?


    Thats not what I am assuming at all. A great deal of excitement is about the de novo copy number variations associated with autism risk. What is not generally understood is that the single gene disorders associated with autism, with the exception of the Fragile X mental retardation syndrome are not heritable at all.

    I’ll give you a short list of single-gene disorders associated with autism risk and the prevelance of de novo mutatations:

    Rhett Syndrome 99% of cases are de novo mutations.

    Downs Syndrome 99% of cases are de novo mutations.

    Velo-Cardio-Facial Syndrome ( 22Q11.2), over 90% of cases are de novo mutations.

    CHARGE Syndrome (Coloboma, Heart defect, Atresia choanae, Retarded growth and development, Genital hypoplasia, Ear anomolies/deafness)most are de novo mutations with a 1-2% recurrance risk.

    Williams Syndrome, 99% are de novo mutations.

    Angelmans Syndrome and Prader-Willi Syndromes are both associated with de novo mutations in 70% of the cases.

    There seems to be a misunderstanding that considers the concept of ‘genetic’ and the concept of ‘heritability’ as being interchangeable.

    The de novo mutations in the genetic syndromes mostly are not present in either parent and therefore are genetic, but not heritable.

    The single gene disorders are associated with neurodevelopmental alterations and intellectual impairment with or without co-ocurring autism.

    Additive factors can also be demonstrated in Downs Syndrome with or without co-occurring autism.
    Ghazuddian in a series of studies reported that first degree relatives (parents and siblings) in Downs Syndrome with co-ocurring autism had an excessive rate of broad autism phenotype features that were not present in first degree relatives in Downs Syndrome without co-occurring autism.

    This would suggest the presence of a gene by gene interplay etiology but it is still not heritable as the Downs Syndrome mutation is not present in either parent.

    Where do these de novo mutations come from? Evolutionary theory would implicate environmental factors as being central to mutagenesis.

  7. daedalus2u December 9, 2010 at 00:11 #

    RAJ, the de novo mutations occur most likely during gamete formation or before the fertilized egg becomes multi-cellular. It is not possible for them to be due to any environmental exposure any time later.

    It is widely recognized that the familial type of autism (the one that is highly heritible) has genetics that are virtually completely impenetrable, with no single gene amounting to more than a few percent of causation.

  8. Prometheus December 9, 2010 at 02:17 #

    RAJ asserts:

    “Where do these de novo mutations come from? Evolutionary theory would implicate environmental factors as being central to mutagenesis.”

    Evolutionary theory is silent on the cause of mutations. In evolution, the environment selects for or against certain mutations (and is neutral on the majority of them) regardless of what might have caused them.

    RAJ also seems quite concerned that people might be confusing genetic disorders with heritable disorders. Well, I can’t speak for everyone, but I’m pretty clear on the difference. Curiously, RAJ points out Downs Syndrome as a non-heritable genetic disorder when there is a small percentage of Downs Syndrome that is heritable because of a translocation of part of chromosome 21. But that’s just quibbling.

    It seems that RAJ is missing the reason for the interest in the genetics of autism. It’s not because it will allow us to blame the parents for their bad genes (a sort of “Refrigerator Mother” redux), but because knowing the genetic basis of autism will help point us toward the neurological basis of autism and – perhaps – the reason for its protean nature.

    As for the cause of the de novo mutations associated with autism, I’ve noted above that the most common cause of mutation is oxygen. While this is an “environmental factor”, it’s not the sort of environmental factor people will want to ban.

    Gene array studies have shown that de novo mutations – even large ones – are surprisingly common. It appears that autistic people don’t necessarily have more mutations, their mutations just happen to be in a sensitive area.

    Also, as Daedalus2U mentions above, any “environmental” exposure responsible for de novo mutations in children would have to occur in their parents, so that pretty much eliminates the majority of “exposures” that are popularly held to cause autism. A vaccine a child receives at age 15 months (or even at birth) cannot cause the same mutation in all cells, not even if the vaccine contains mercury.

    As for things that can cause de novo mutations in the germ cells (sperm and egg cells) of the parents of autistic children, the list is quite extensive, encompassing the already-mentioned oxygen (hard to reduce exposure to that one), thousands of plant compounds (see: psoralens and DNA intercalation), ionizing radiation (but not cell phones or microwave ovens) and a vast array of chemical compounds both synthetic and naturally occuring.

    None of these mutagens is uniquely capable of causing autism, so there cannot be a “smoking gun” that one can point to and say “This causes autism!” except in the general sense that any of them could – as well as they can cause any other genetic disorder caused by a mutation.

    So, there may be an “environmental” contribution to de novo mutations that lead to autism, but most of them will be due to oxygen. Good luck getting a group of parents behind banning that “chemical”. On second thought, given the low level of rational thought in groups like AoA, they might just try to get oxygen banned.


  9. RAJ December 9, 2010 at 13:14 #


    You seem to always infer what others believe based on your own beliefs. As far as the environments possible role in mutagenesis I’ll refer you to Daniel Geschwind, a behavioral geneticist, a true believer in genetic determinism who in a review of the genetics of autism wrote:

    “These data suggest one of many potential mechanisms through which environmental factors could play a role in creating de novo genetic events causing autism, that is, accumulation of mutations in the male germline. Such factors could occur in isolation or in conjunction with genetic susceptibility loci. In the latter case, certain inherited haplotypes, for example, could render specific regions more vulnerable to mutagens, thus increasing the frequency of mutational events. Alternatively, certain regions may be more vulnerable to other environmental factors that could affect chromatin structure or gene expression, leading to epigenetic causes of autism”

    Transmission of germ-cell mutations to the offspring may also encounter gender-specific influences. Gender differences in susceptibility to chemically derived alterations in imprinting patterns may pose a threat for the health of the offspring and may also be transmitted to future generations

    As far as mutagens specifically ‘causing’ autism well that is certainly true, but de novo mutations have been reported in all the major neurodevelopmental disorders (Autism, Schizophrenia, ADHD, Intellectual disability) which would be expected since neurodevelopmental disorders are all multifactorial. De novo mutations are associated with risk for neuroanatomical alterations regardless of diagnosis. In fact, no diagnosis specific susceptabilty gene in the mental disorders has ever been identified so you can make the same argument that neither de novo mutations or candidate susceptability genes are specific to autism just as no mutagen is specifically linked to autism which leads to the same conclusion, the etiology of autism is unknown.

  10. Prometheus December 9, 2010 at 18:43 #


    I’m trying to understand your position, but I can only work with what you have written. If I misinterpret your position, perhaps it is because I have my own biases (as do we all), but perhaps you are not being clear.

    With all due deferrence to Dr. Geschwind, he is speculating – there is no data showing that he is correct (or, to be fair, incorrect). In fact, the article you cite was written before the major genome-screening studies that showed how common de novo mutations were.

    The Jiang et al study Dr. Geschwind cites in support of his statement, “…certain regions may be more vulnerable to other environmental factors that could affect chromatin structure or gene expression, leading to epigenetic causes of autism.” is actually about how the “Angelman gene” (UBE3A) is involved in imprinting of the genome and how this (at least partially) explains the Angelman/Prader-Willi dichotomy in paternal vs maternal 15q anomalies. This is an “environmental factor” only in the sense that the functioning of the UBE3A gene (and its product, the E6-AP ubiquitin ligase) alters the intracellular environment.

    The Eichenlaub-Ritter et al paper discusses how gender differences in both gene expression and gametogenesis can impact the mutagenic potential of certain compounds and how genetic imprinting may also be affected, leading to differences in how mutations – de novo mutations – are passed to the offspring. This – again – does not support the claim (made by many, if not yourself) that specific mutagens preferrentially cause specific mutations.

    Clearly, there are mutational “hot spots” (this has been known since the 1940’s) and there are mutations that impair the cells’ ability to deal with certain types of mutation (see: xeroderma pigmentosa) as well as certain types of mutagens. However, absent any data showing that a specific gene or class of genes involved in DNA repair or “toxin” mediation is consistently (or even frequently) associated with autism, we are left with the conclusion that no single environmental factor or group of external environmental factors can be associated with a significant fraction of autism cases at this time.

    Note that I’m not saying that environmental factors cannot cause autism or that no cases of autism have ever (or will ever) be caused by environmental factors. The current definition of “autism” is far too broad to make those sort of claims. However, the currently available data – including the data on de novo mutations – do not support a specific environmental cause or environmental “trigger” involved in a significant fraction of autism cases.

    If you want to insist that environmental factors NOS (not otherwise specified) are a potential cause of de novo mutations that lead to autism, I can support that. However, keep in mind that the types of de novo copy number variations found in the genome-screening studies are not the typical point mutations seen as the result of mutagens. The sort of large-scale deletions and duplications those studies found are much more likely to be the result of faulty crossing-over during meiosis, and that – in the absence of parental mutations of the proteins involved in crossing-over (which hasn’t been found) – is most likely due to random chance.


  11. testsieger kaffeevollautomaten September 8, 2013 at 03:45 #

    What a stuff of un-ambiguity and preserveness of valuable familiarity on the topic of unpredicted


  1. Tweets that mention Autism Blog - Advancing paternal age and risk of autism « Left Brain/Right Brain -- - December 2, 2010

    […] This post was mentioned on Twitter by Kev and Brandon Blietz, Alltop Autism. Alltop Autism said: Advancing paternal age and risk of autism […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: