Psychology Today, December 3, 2012, and The New York Times, August 23, 2012, both cite a study in Nature, August 22, 2012, by Stefansson et al., telling us that as men age, they are more likely to father a child who develops autism and schizophrenia. It says:
"Male sperm-producing cells are constantly dividing and as a result the number of new mutations increases over time. The sperm of a 20 year old man carries about 25 mutations; the number rises at a rate of 2 per year. So the sperm of a 40 year old man may have some 65 new mutations." This raises "the specter of an inevitable decline in the mean fitness of the population.” "After 35, both men and women are more likely to produce children with various kinds of developmental difficulties." It may be "that their union would produce an autistic or schizophrenic child, or a child with a lower IQ".
This may imply that a long lasting preference for fathering children on a late age could lead to permanent changes in the gene pool. If the trend for late parenthood is limited to the western world, China and India perhaps included, this part of mankind will get into a disadvantageous fitness position compared with the third world. That, in its turn, may influence evolution as well.
Apart from a general, cultural trend to late fatherhood, it is also conceivable that men with autistic traits are inclined to postpone getting children and/or to postpone committing themselves to women with a baby wish. If that is true, then the quality of the sperm and the number of mutations are not the sole culprits of the increased chance on autistic and schizophrenic children.
You're referring to the so-called "dad-factor". The article by Reichenberg et al (2006) in "The Archives" wasn't mentioned yet. If you send me an e-mail-address I'll send the PDF, since I believe it isn't legal to desseminate a copyrighted article.
With regard to schizophrenia, conclusion of a recent meta-analysis (Miller et al. Schizophrenia Bulletin 2011) was: "Both advanced paternal age (≥30) and younger paternal age (
There are also interesting results from animal studies. Increased paternal age in mice leads to impairments in social and exploratory behaviors in the offspring. See Rebecca Smith and colleagues in PlosOne 2009.
Increasing the rate of mutations can be seen as the equivalent of a high risk investment or bet. You can lose a lot - meaning end up with neurocircuitry dysfunction that would manifest as (severe) mental illness or hit the jackpot - meaning getting the sort of neurocircuitry that by processing things differently is the foundation of creativity. There is intriguing data about suggesting a continuum of creativity - bipolar - schizophrenia; also a number of prominent scientists and artists had older fathers.
It is pretty clear that there are non-genetic causes to both autism and schizophrenia. Otherwise there would be no differential incidence depending on differential exposure to things in utero.
What does happen in utero is epigenetic programming during differentiation and organogenesis.
Rett Syndrome is caused by loss of the MeCP2 gene on an X chromosome. In males it is (almost) always fatal in utero. Females are mosaic with random silencing of one X chromosome. The MeCP2 protein binds to methylated DNA and so modifies the readout of DNA that has been epigenetically modified. The MeCP2 knock-out mouse recapitulates Rett Syndrome with surprising fidelity, and reactivation of MeCP2 in those ko mice seemingly resolves it completely.
The sample used in Stefansson et al. was highly enriched in individuals with autism or schizophrenia. All the probands fit the mutations vs paternal age correlation, including autism, schizophrenia and other. In other words there were individuals with few mutations and with many mutations in all groups.
The paper by Peterson et al (cited by Stephan Ruhrmann ) is very interesting, it really does show the association is between paternal age at first child, not paternal age at conception. It seems to me that the relationship with paternal age and with de novo mutations is an association and not necessarily causation.
Sperm DNA is epigenetically programmed. Is the effect of paternal age due to differential epigenetic programming of sperm DNA with paternal age? Or with something that correlates with paternal age? It is conceivable that being a father does change the epigenetic programming of sperm that develops later.
It might be possible to see this in children conceived by artificial insemination by donor depending on the age and child status of the donor at the time of donation.
There is controversy over the contruibution of advanced paternal age. Not only the advanced age of the father, but also advanced age of the mother appears to be a risk factor. Reichenberg's paper (Arch Gen Psychiatry. 2006;63:1026-1032) reported an aOR of 5.75 for fathers between 40 and 49, whereas this aOR was only 1.38 in a Californian study published by Shelton et al. (Autism Research, 2010, 3, 30-39). Both studies corrected for maternal age. However, the main difference was that in the Israeli cohort there were 588 mothers over 40, with 4 cases, while the California cohort had 113,080 mothers over 40 with 501 case mothers. The Shelton study appears therefore to be statistically more robust.
Thanks a lot, you all, for your professional comments and useful references on this topic. It is obvious, one should be distrustful of simplistic view on the topic.
- Badges
- Science topic
- Similar topics
- Philosophy
- Esthetics
- Interpretation