How many mutations do parents pass on?

Every one of us got half of our genetic information from each of our parents. But how many new mutations did Mum, or Dad, hand on to us? And if Mum and Dad had waited a few more years before they had us, how much difference would that have made? Now, thanks to a remarkable study looking at whole families conceived over significant periods of time, we know. Thomas Sasani is at the University of Utah…

Thomas - Every generation, we receive half of our DNA from each of our parents. But what we're really interested in figuring out is how many sorts of new mutations are present in the DNA that we inherit. Essentially, how many “spelling errors” are there in the genetic code that we receive from our parents?

Chris - Did we not know that already?

Thomas - So the actual estimated number - this sort of number of spelling errors - was actually pretty well appreciated, and it's been estimated at a number of times before. But what's less well appreciated is how all sorts of different other factors, like the age of your parents when you're born, actually affects that number.

Chris - Because we often say “have your children when you're young, because A you're more fertile and B the risk of having things like chromosomal abnormalities increases with parental age”. So, is that true then?

Thomas - Yes that's absolutely true. In this study actually we were a little bit less interested in sort of the large chromosomal changes - things like the aneuploidy that you mentioned - and more interested in sort of individual spelling errors. So single letter changes in the DNA that you inherit from your parents a “T” instead of a “C”...

Chris - Did you have to therefore go and laboriously read the genetic code of loads and loads of individuals to work out how different they are from each parent?

Thomas - Essentially yeah, that's exactly what we did. In practice, what this amounts to is you sequence the entire genome of both parents, and you sequence the genomes of all of their children, and then you sort of go child by child and compare their genome sequence to the sequence of both of their parents and you just look for the places in the genome where they're different; where even though you would expect that child to share half of their DNA with each parent, at a small number of sites they'll actually have unique mutations that neither of their parents had.

Chris - And have you done the experiment where you could, for instance, take a family and the parents who are obviously correspondingly younger when they have one child, and older when they have subsequent children, and then you could compare how many changes there are all in each of the subsequent children and therefore relate that to parental age?

Thomas - Yes that's exactly right. And really one of the fantastic things about the dataset that we were working with, normally when studies like this are done you use two parents and maybe one or two children. But in this study we had some families that had up to 16 children. And so those children represent - as you were saying - parents when they were, say, 20 years old all the way up to when they were maybe 40 years old. And so we are able to compare the number of new mutations we see in children born to young parents and children born to that same set of parents, but when they were much older.

Chris - What trends emerged?

Thomas - So, overall, on average the number of these new mutations that you see in kids definitely increases as parents get older. So it amounts to about one and a half new mutations every year dads get older, and about point five additional mutations every year mum gets older.

Chris - That's quite a lot isn't it? And interesting that there's that disparity between the sexes. Does that reflect the fact that sperm are made as a continuous process from ongoing divisions in stem cells, whereas eggs get made when actually the individual is themselves developing from an egg and therefore the chances of the DNA in the egg becoming mutated is lower?

Thomas - Yeah. That's a really great question and I think for a long time that's really been the sense in the field is that every year after puberty, sperm cells - or the stem cells that will eventually develop into sperm cells - they're constantly dividing and the idea is that every time you have to copy your genome there's a chance for an error to occur. Now it turns out that there's a hypothesis that the constant sperm cell divisions are not the only reason that there's this kind of bias of de novo mutations in fathers, and that there may be things like accumulating DNA damage and other causes. But, certainly, the increased number of divisions in the sperm cells are probably contributing a little bit to that bias that we see in dads.

Chris - When you say a little bit, do you have a feeling for when most of these mutations are acquired? Because one other possibility, surely, is that not just the sperm and the eggs contributing, but if the embryo itself is not faithfully copying DNA, maybe quite a big amount of that mutation burden could occur after fertilisation has occurred?

Thomas - Yeah that's a great point. We actually were able to get a sense of this in our paper, largely owing to the fact that we had a lot of these large and three generation families. So we were able to apply a strategy whereby we were able to figure out essentially if these new mutations had occurred after the sperm and the egg came together to produce an embryo, or if they happened before. And we actually found that about 10 percent of all of these new mutations that we saw were likely occurring in that embryo.

Chris - And was that itself influenced by the age of the sperm and the egg? So were older parents more likely to have highly mutating embryos, or was the mutation rate in the embryo at about 10 percent regardless of parental age?

Thomas - Yes. So in the embryo at least, we didn't see that the number of new mutations was really affected by the age of the parents. However, overall, the mothers and fathers in our study were generally under the age of 40 or 50. And it's possible that there may be an affected age in much much older parents, but we really weren't able to detect that, at least in our dataset.

Chris - So although you can say that, with increasing parental age, there does appear to be a higher mutation burden being handed on to kids, but, presumably, one constraint to this study is you can't say at the moment whether or not that's going to have a clinical impact?

Thomas - Yeah that's right. So if you think about the number of mutations that we're seeing - 70 new mutations on average in a child - and the number of new mutations might increase by one point five or point five per year depending on the parent you're looking at. But again this is out of - you know - 3 billion or 6 billion total letters in the human genome. And so, in practice, not very many of these mutations are actually landing in genes - regions of the genome - that actually make functional protein. And so at least at this point it's it's tough to say how frequently we'd expect these to be damaging, or cause disease, but certainly there are a number of rare genetic diseases caused by these kinds of mutations. And so having a good handle on how frequently they occur is I think important in its own right.