A relatively recent article in Nature (Kong et al. Rate of de novo mutations and the importance of father's age to disease risk. Nature Volume: 488 Issue: 7412 Pages: 471-475 DOI: 10.1038/nature11396 Published: AUG 23 2012) analyzed what may be the largest amount of parent-child DNA sequence data so far generated to identify novel mutations. Surprisingly, they find that almost all of the variation in the number of mutations is explained by the age of the father.
My apologies if this article was diaried before, but if so, I didn't find it!
According to wikipedia, there are 4000 human diseases caused by single gene defects. This category includes such diseases as sickle cell anemia, huntington's, and haemophelia. It also includes hundreds to thousands that most of us have never heard of.
In addition to these diseases caused by defects in a single gene, there are also complex genetic disorders - diseases where we know there is a heritable effect, but can't pinpoint the exact genetic cause. This includes diseases such as autism, cancer, diabetes, and heart disease.
As early as 1912, before the rapid expansion of modern genetics and our understanding of genes, Dr. Wilhelm Weinberg, who was studying a form of dwarfism named achondroplasia, noticed that the incidence of this disease increased with the age of the parents, and posited a mutational basis for it. The diseases is heritable, but can also arise spontaneously from unaffected parents. Many years later, the gene responsible was identified, and of 154 cases studied, 153 were due to a single amino acid change caused by mutation of a single nucleotide - incidentally this is the nucleotide in the human genome with the highest known mutation rate.
What causes these mutations? In the recent study in Nature (sadly firewalled), a team from Iceland aimed to study spontaneous mutation is in the human population. As part of a very large scale study of human genetics, they sequenced and analyzed the genomes of 78 families; father, mother, and child. By comparing the sequences between the three, they determined whether a specific nucleotide in the child was mutated relative to their parents. In this study they identified 4933 de novo Single Nucleotide Polymorphisms. There are many different types of mutation that can occur, but SNPs (pronounced snips) are where a single letter of the genetic code (G,T,A,C) is switched to another letter, most commonly from a C to a T. Most of the time, such changes have no discernible effects, though as much as 10% of the time these changes could be detrimental in humans.
In this particular study, the researchers had sequenced families where one or more members had either autism, or schizophrenia. Of the 4933 SNPs they identified, 62 were predicted to lead to protein changes. Of these 62, one of them caused a mutation that has previously been linked to schizophrenia, and two of them led to changes in genes that will now be considered relevant for autism.
While identification of these mutations is in itself interesting, what was more interesting about this study is that they were able to (in 5 families) pinpoint whether mutations came from the mother or the father. In these five families, the child had, on average 70 new mutations. Of these, 14 of them (on average) were inherited from the mother, and 55 (on average) were inherited from the father.
While this result was not unexpected, what was surprising is that the rate of mutation increased with the age of the father (and not with the age of the mother), by roughly 2 mutations per year. Further, the father's age accounts for 97% of the variation in the number of new mutations in their study.
This has some interesting ethical and practical implications. As men father children at later ages, it increases the number of mutations in the next generation. Every 6 months we men age, we will (on average) pass on one new mutation to any child we father. The total number of mutations doubles every 16 years. They note that, as the average age of fatherhood in Iceland has increased from 28 in the 1980s, to 33 now, the average number of mutations per child has also increased from 60 to 70.
Also, this quote from the original article is intriguing:
Seeing an association between father’s age and mutation rate is not surprising, but the large linear effect ... is striking. Even more so is the fraction of the variation it explains, which limits the possible contribution by other factors, such as the environment and the genetic and non-genetic differences between individuals, to mutation rate on a population level. Also, even though factors other than father’s age do not seem to contribute substantially to the mutation rate diversity in our data, it does not mean that hazardous environmental conditions could not cause a meaningful increase in mutation rate. Rather, the results indicate that, to estimate such an effect for a specific incident, it is crucial to take the father’s age into account.
What the authors seem to be saying here is that any environmental effect on our genes is dwarfed by the effect that age has on men's sperm. (edit: I may have misunderstood this conclusion, see comments below by Sandino and following discussion)
Why is this the case, and why do the number of mutations women pass along as they age not increase also? The answer is likely due to the differences in the number of times sperm and eggs have divided. While a woman is born with the entire complement of eggs she will have in her life (and these eggs have undergone roughly 24 cell divisions), sperm are regenerated in large numbers throughout a man's lifetime. At the onset of puberty, the cells that will eventually give rise to sperm have undergone approximately 30 cell divisions. Then, these cells will divide again every 16 days (ca. 23 times a year), before going through 6 more divisions to produce a sperm cell.
It turns out that the process of replicating DNA in order to divide is itself mutagenic. While the enzymes that replicate DNA (polymerases) are very good at not making mistakes, they do at times make them. Polymerases make (on average) 1 error every 1 million to 10 million nucleotides. In addition, there is a DNA repair system known as mismatch repair, which scans the DNA following replication looking for errors, and then fixes them. This decreases the error rate 100-1000X, or one mistake every 100 million to 10 billion nucleotides. As the human genome has 6 billion nucleotides, at the high end replication produces less than one mistake every cell division. However, after the perhaps hundreds of divisions an older man's sperm has gone through, even this low probability of making a mistake will lead to the accumulation of mutations.
For more coverage of this article, try the NYT, or scientific american.
I'll leave you with a quote from an accompanying article:
If the paternal-age effect on the de novo mutation rate does lead to substantially impaired health in the children of older fathers, then collecting the sperm of young adult men and cold-storing it for later use could be a wise individual decision. It might also be valuable for public health, as such action could, according to Kong and colleagues' findings, substantially reduce the rate of deterioration of the gene pool in human populations under relaxed selection. By contrast, any attempt to reduce mutation accumulation in our gene pool by restoring selection pressures would probably be much more controversial and painful. Thus, Kong et al. have certainly provided food for thought, on both an individual and a population level.
I'll try and answer any questions in the comments, though you might have to wait a while!