Mum’s genes, dad’s genes
As Rebecca mentioned, one of the first talks we heard was from Marisa Bartolomei, Professor of Cell and Developmental Biology at the University of Pennsylvania, who gave the 2017 Genetics Society Medal Lecture about her work on genomic imprinting. Kat Arney started by asking her to explain exactly what imprinting actually is.
Marisa - Genomic imprinting is a process that affects only a small number of genes. It’s mammalian. The bottom line of genomic imprinting is that these small number of genes are only expressed either from their maternal allele or their paternal allele – the copy they inherited from mum or the copy they inherit from dad. The other copy is repressed.
Kat - For most of the genes, we get one copy from mum, one copy from dad and effectively, it’s a backup system. If one breaks the other one works. But these genes, it’s only mum’s copy or only dad’s copy. Is that in all cells that it’s only mum or dad’s copy that is switched on and the other one is switched off?
Marisa: So for certain genes, it’s in all cells, but for other genes, there is tissue specific imprinting. So for example, there may only be expression from one of the parental copies only in the brain or only in the placenta, and other tissues normally have both copies. So there's a lot of different variations on this. But for other genes, it’s everywhere. Everywhere the gene is expressed, it’s only expressed from one of the copies. If it’s the mother’s copy, it’s always the mother’s copy.
Kat - What sort of genes are we talking about that are only active if they come from mum or only active if they come from dad? Are there common traits between these sort of genes?
Marisa - Yes. So I think when we first started studying it and thinking about it, we thought that most of these genes were going to be growth regulatory genes and genes that were important in the embryo and the foetus. But as time has gone on, we found genes that have other properties that are beyond foetal growth that are properties in behaviour, properties in postnatal energy homeostasis.
So, their properties are pretty diverse and you can no longer just say, well, these are growth regulatory types of genes. It may also be that there are only a few whose imprinting is important and that others have gone along as bystanders. They were just in the region of an imprinted gene and they went along.
But at this point, it’s not very easy for us to tell. That said, we can say that imprinting is pretty highly conserved between mice and humans, so there must be something there that maintains that selection for imprinted genes.
Kat - I do remember the kind of idea that the genes that are active from dad are the sort of genes that make foetuses grow really big because they're trying to suck all the resources out of mum and then the genes that are expressed from the mum’s copy are the ones that are trying to supress that. But you're saying that it’s kind of a bit more complicated than that.
Marisa - I mean, I think it’s a bit more complicated although that is still the prevailing theory and people who study imprinting from an evolutionary perspective really seem to focus on those types of traits. But you could say that behavioural traits are also something that’s important for maternal or paternal care.
We’re not really that clear on the evolutionary purpose of imprinting but we do know it’s there and we’re interested in studying what these genes are, how they're regulated and what the consequences there are to human health.
Kat - So let’s unpack that a little bit more. What do we know about how this is controlled, how in these cells, it’s only the copy that comes from dad is active and mum’s copy is off, or vice versa when only mum’s copy is on and dad’s copy is switched off? What are the molecular nuts and bolts that are making that activity or silencing happen?
Marisa - One of the things we know for certain is that most imprinted genes are found grouped in clusters in the genome. We also know that they're regulated by these small bits of DNA, maybe only one kilobase in length. These bits of DNA seem to have epigenetic modifications that are put on in a parental specific way. So, there may be DNA methylation or chromatin modifications.
They come on either in the father’s germline or the mother’s germline. So at the time when eggs and sperm then meet for the fertilisation event, they already have their maternal specific marks or their paternal specific marks. Once they go into a fertilised egg, they're saying, “I came from mum. I came from dad. I’ll be active. I’ll be repressed.”
And so, it’s the germline where a lot of these events of imprint establishment take place and then after fertilisation, they're read by the normal machinery that reads everything else and the genome to turn things on, turn things off during development.
Kat - I always like to think of these epigenetic modifications as almost like sticky notes or Post-It notes. So there, you're saying that they're being put on in the eggs. When mums are making eggs, it’s like, “Okay, these genes all came from me” and in dad’s sperm is like, “These genes all came from me.” But then if the embryo then when it grows up and make eggs or sperm depending on whether it’s a male or female, presumably, all those marks have to be taken off again and reset.
Marisa: Right. Then the sticky notes are sort of torn up, disposed of and then as they go through becoming eggs or becoming sperm, they're saying, “Okay, they know I'm an egg. I'm a sperm” partly by their genetic composition part, by their environment. And they say, “Okay, I'm an egg now. I will put the marks on during growth of the oocyte and right before ovulation.”
Whereas if they’re in the sperm, they’ll say, “Okay, I know I'm in the male germline. I will put on my marks.” And so, they know that but they take what's present in the somatic tissue, erase it and restart again. And that’s really critical for the lifecycle of these imprints.
Kat - So we have this cycle of imprinting of marks coming from mum, marks coming from dad, making an embryo with mum’s genes on or dad’s genes on, wiping them out, starting again, round and round, and round. So we know that that’s kind of how it works normally. But what about some situations where it’s gone wrong? Are there diseases that are associated with problems in this imprinting?
Marisa - So, there are quite a few imprinting disorders such as the overgrowth and undergrowth disorders Beckwith-Wiedemann syndrome and Silver-Russell syndrome as well as some neural behavioural disorders such as Angelman syndrome and Prader-Willi syndrome as well as other imprinting disorders. There are cases where there's just deletions or mutations in the expressed copy of a gene that leads to this disorder.
But there are also cases of which depending upon the disorder how prevalent they are where they're just mistakes in that marking of the imprint that occurs in the germline or maybe something happens during gametogenesis where it’s erased a little bit or after fertilisation, something happens. It’s not clear but what we can see in some of these children that have Beckwith-Wiedemann or Silver-Russell, they’ll have a mosaic pattern in their cells so that some other cells have their DNA methylation that looks right, normal, and some cells have lost it.
And so, we don’t know if there's something that happens very early in development that leads to this if there's an environmental perturbation. One of the things that my lab as well as few others have studied is the effect of assisted reproductive technologies on imprints. Using a mouse model, we have shown that some of the techniques used in assisted reproduction can lead to some mistakes in maintaining appropriate imprints that could lead to these disorders.
Kat - Basically IVF, in vitro fertilisation.
Marisa - Correct and there's been an association with increased incidence of Beckwith-Wiedemann syndrome. So the environment that an egg or a sperm, or an early embryo in can be very important for appropriate imprints.
Kat - Now that we know that there are these imprinted genes and starting to understand how they work, what are still the big questions? What do we still really need to know?
Marisa - I think there still are a lot of big questions and that is, how are these genes recognised in the germline to be modified or marked with their parental origins to their imprint. That’s another thing that we and others are still working on. So there still are a lot of really important questions to answer and again, since it affects human health, it is something that we really want to drill down.
Kat - Marisa Bartolomei from the University of Pennsylvania.