So, now rather than considering the paternal and the maternal genome in general, I would like to start to consider specific regions of the genome and how they are reprogrammed. So, we know the paternal and the maternal genome are treated differently during epigenetic reprogramming, but in general they are completely cleared between generations. So, now let us think instead about, first of all the repetitive elements. And then about then about the imprint of genes. So if we think first about the repeats, so these repeats, which we've mentioned before, an example of which are the intracisternal A-particles, are spread throughout the genome and they're in very high copy. They in general, want to keep your repeat silenced. Because they can be detrimental when active. They can cause illegitimate combinations between chromosomes. They can activate surroundings genes in terms of the expression levels. So what you generally see here, and shown by the orange line is that they tend to maintain a high level of methylation throughout these two period of epigenetic reprogramming. This obviously makes sense because these genes, you would like to keep them silent. We do know there are some very brief windows in time, particularly in the germ cells and in embryonic stem cells, which are derived from the inner cell mass of a blastocyst, when you see slight demethylation in these repeats. And at the same time you see up regulation of the piRNA's that you'll remember are involved in targeting RNA directed DNA methylation targeting so that we can have remethylation of these repeats again. But in general, they tend to be resistant to these two rounds of epigenetic reprogramming. Instead, the imprinted genes or a second class of genes that we are really going to spend most of this week thinking about. The imprinted genes which are now shown in a black line in this image undergo reprogramming in the primordial germ cells, but resist the reprogramming that's found in early development. In the primordial germ cell development, we know that they are cleared a little bit later than the rest of the genome, but their resetting happens at about the same rate. So, they seem to be restricted a little bit in terms of when they will be cleared, and this might be because they are a special class. So, they need to be treated differently. And so temporarily they are offset from the rest of the genome. So they’re cleared in primordial germ cell development, and they maintain their level of methylation during early embryonic development. Well, why would this be and what is an imprinted gene? Well, genomic imprinting or parental imprinting or gametic imprinting, they all refer to the same thing. And in each case what they're referring to is monoallelic gene expression, and that is expression from just one of the two parental alleles. But this expression occurs in a parent of origin's specific manner. In other words, it's always happening from the one allele that you inherited from a particular parent. If you compare this to X inactivation where you have random X inactivation in the adult, we know you have monoallelic gene expression that is only one of the two alleles of the X chromosome were expressed, but it doesn't happen in the parent of origin specific manner. You can express either the one you inherited from your mom or from your dad. So there's a difference here between genomic imprinting and X inactivation. Imprinted X inactivation which happens pre-implantation in the mouse, is imprinted, that is, that it's always the paternal that is silenced, the paternal X that is silenced. So, if we think again about genomic imprinting, then we think about these two chromosomes pictured here. One that says blue so it's the paternal chromosome. One that's pink that's the maternal chromosome. I've pictured here one imprinted gene. It's imprinted because it's expressed from just the paternal allele, as shown here by this arrow, but not from the maternal allele. So it's expressed from one parental allele and in this case it's in a parent-of-origin specific manner. If you look at all of the cells, they'll only express it from the paternal chromosome. So this seems like a fairly unusual, type of thing to happen in terms of gene expression. So what we know is it's absolutely critical for embryo viability. It's critical for health. So back in the early 80's, they were able to discover, that if you took those pro nuclei that you see in the zygote, in the fertilised egg, and you took two male pro nuclei, so by microdisection you could pull out two male pro nuclei and put them into another enucleated oocyte, you wouldn't get a viable embryo. They'd die before implantation. Similarly, if you took out and you made an embryo that had two female pronuclei, so that both come from the maternal side, again, you would get death of the embryo before implantation. Whereas if you did the same technique, you could control for the technique and pull out a male pronucleus and a female pronucleus and make a new zygote, they'll survive, so it's not that you technically can't do this. It's that you cannot have both sets of chromosomes coming from, from the paternal side or from the maternal side. Now you may begin to suspect that may be that's because you need to have the genetic differences between the parents. This isn't true either. If you do it in inbred strains of mice that are genetically identical every locus, you don't need to have the genetic differences but rather you need to have the epigenetic differences between the two parental alleles. And the epigenetic differences between the parental alleles need to happen at loci that are subjected to genomic imprinting. So, we know that this process of genomic imprinting is critical for viability. And we also know that it's a very traditional epigenetic mechanism to study. So, just like X inactivation that we discussed last week, just as we said that it's really important to be able to study this model system, to be able to work out how epigenetic control works spread throughout the genome. Similarly, parental imprinting or genomic imprinting is this model system that we can study. So how do you get this parent-of-origin specific gene expression. Well the first thing is that you have a controlling element. So there imprint control regions so ICR's or there otherwise known as differentially methylated regions so DMR's or differentially methylated domains DMD's or sometimes imprint control elements ICE's. They all, essentially, mean the same thing. So if you're reading an article you can think this is basically, synonymous. I'll refer to them as ICR's, but each imprinted gene can be controlled by the, an ICR. So the imprint itself is associated universally with DNA methylation at that imprint control region. So, if a gene is paternally imprinted it will have methylation on the paternal imprint control region, but not the maternal imprint control region of that same gene. So, the consequence of that DNA methylation, though, the consequence of that imprint at the imprint control region is different depending on the cluster. So in this case, its not the instance where DNA methylation is universally associated with gene silencing like it would be at a CpG island. Rather, it's more complex than that. And so the way that each imprint control region actually brings about gene silencing differs by cluster, or differs by each individual locus. The instances that we will talk about long noncoding RNA so, the action of a long noncoding RNA bringing about imprinted gene expression or enhancer and insulator blocking. Although these aren't the two only mechanisms by which an ICR can bring about imprinted expression. So, as we said in the previous couple of slides, the imprint control region methylation is established during primordial germ cell development. And it's in this way that you can get the parent-of-origin specific marks. So, I'm going to go through and explain this in two different diagrams. So here, as I showed you before, you have reprogramming and resetting of the epigenetic marks at the imprinted loci in the sperm and the eggs. And so, therefore, because they're developing in a particular parent. So, the sperm are developing and they'll get the paternal set of marks. Whereas, by contrast, the eggs are developing in the mother and they will get a maternal set of marks. These marks are then maintained through these periods of reprogramming and early development. So since they are maintained, they remain as being maternal or paternal or whatever was setup in the egg or the sperm. So, for a long time, we didn't know how they would escape this reprogramming that happens in early development but we now understand this involves maternally derived or oocyte derived maternal effect proteins. These are proteins that are expressed in the egg and they then seem to bind to these imprint control regions and protect those imprint control regions from the very widespread demethylation that's happening during this time period. So, if we think about this in a slightly different way or in a slightly different diagram, we know that once you have these epigenetic states established in the zygote shown here. And what I'm showing is just one chromosome that you've inherited from the mom in pink and one from the dad in blue. And if we think about two imprint control regions that are shown here we've got a methylated one that's maternally methylated, so maternally imprinted. Which is then unmethylated on the paternal allele. But we have a second one that is methylated in the paternal allele, so it's paternally imprinted, and it's unmethylated on the maternal chromosome. So this level, this difference between the paternal chromosomes and these imprinted expression patterns and imprinted DNA methylation in this case is maintained in the somatic cells of the embryo and, as shown here, maintained in the somatic cells of the adult. But what happens in the primordial germ cells is very important. So when the primordial germ cells are developing, they erase the DNA methylation marks as shown here. So in the primordial germ cells you now have no methylation at these imprinted control regions. And then resetting happens differently, depending on whether that embryo is female or male. If it's a female, you'll have resetting of the female specific or the maternal specific epigenetic marks. In other words, you'll have all oocytes develop with their haploid genome, and they'll have said one copy of the, the chromosomes. And they'll have methylation at that maternally imprinted locus, but no methylation at the paternally imprinted locus here. And the reciprocal will be true if the embryo is a male. If the embryo is a male, then you'll have methylation of the paternally imprinted locus, but no methylation of the maternally imprinted locus. So it's because this resetting only happens during primordial germ cell development when it's still clear the parent of origin, that is the developing into either a female or a male because this clearing happens just once but escapes re-programming in early development. This is how you can have imprinted control regions having DNA methylation only on one parental allele. And therefore how you can bring about this imprinted gene expression. So in the next lecture we're going to talk more about imprinted genes in particularly think about the location of imprinted genes spread throughout the genome and how we'll have a little bit aside to think how you actually measure DNA methylation at imprint control regions or other places in the genome.
