So we're going to move on now from the very particular epigenetic marks that we've covered in the last few lectures. So DNA methylation and various histone modifications, to think about how these modifications might bring about changes in the chromatin itself. How the chromatin is packaged more densely or more sparsely. And to think about this, we really need to consider the Chromatin remodelling complexes. So these complexes use energy in the form of ATP within the cell. This is the usual energy currency and when it's hydrolyzed and it releases the energy. This energy's able to be used by the chromatin remodelling complexes to actually shift where the nucleosomes are. So I think the best way to think about the chromatin remodelling complexes is that they are the ones doing the real work in the cell. They're the ones that can densely pack down the nucleosomes or make them more sparse more sparsely located. And as you remember from the earlier lectures, it's really this accessibility of the DNA to the transcriptional machinery or the DNA replication machinery or the DNA repair machinery which is relevant. So, what one of the things that when the chromatin remodelling complexes first started being characterised, one of things is was that people found it hard to really remember that they can either be activators or repressors. They're neither one nor the other necessarily. It really depends on context. And this is because, as I just mentioned, they can make them more densely packed, the nucleosomes, more densely packed or more sparsely packed. They really have both of these functionalities. Their primary role is just to move the nucleosomes. In addition to moving the nucleosomes around, the chromatin remodelling complexes can also be involved in removing the nucleosomes entirely, called nucleosome eviction or disassembly. And this happens sometimes around transcriptional start sites around the promoters of genes. And it's known that when you completely remove the nucleosome, the DNA clearly has much as movement and accessibility to these transcription factors as it possibly could. In the next slide, after this, I'll be able to show you a little movie of this. We also know that the histone the chromatin remodelling complexes are involved in histone variance deposition. What I'll come to in one of the next lectures is talking about these histone variants. And they need to be incorporated into the nucleosomes by these chromatin remodelling complexes. And finally, the chromatin remodelling complexes can have a role in nucleosome turnover. Now what we've been speaking about in previous lectures is how you can have mitotic heritability of epigenetic marks. And clearly, if the nucleosomes can be turned over, they can be refreshed or renewed without relying on changes in the cell cycle and this could alter mitotic heritability. So I guess the question is, how is it that these complexes are actually able to move the nucleosomes around? And the answer is that they disrupt the interactions between the negatively charged DNA and the positively charged histones. And by disrupting these they can actually shift where the nucleosomes bind, where the histones are actually binding in the DNA itself. So now I'll show you a short movie of this. It will show both nucleosome eviction or disassembly, as well as sliding of these nucleosomes along the DNA. So you can see here nucleosome disassembly, these histone proteins, that are shown in grey, are completely removed from the DNA. And now this loop of DNA with the purple backbone, as you can see, is really free to move as much as it likes. Next you'll see the chromatin remodelling complex in red, come and bind to the nucleosome. And you can see it shifting it along, just inch by inch, disrupting these interactions and it can slide the nucleosomes so that they're more closely spaced to one another. So this is really how the Chromatin remodelling complexes work. So these Chromatin remodelling complexes come in three flavours, three different types. Each of them have a different histone tail modification that they recognise. But they all possess ATPase activity. So they have the ability to hydrolize the ATP, and provide the energy for the chromatin remodelling. The first of these complexes is called the SWI-SNF complex. And it was named based on the function that was described in yeast. So SWItch/Sucrose Non Fermentable is what it stands for. It contains a bromodomain which you'll remember from previous lectures recognises an acetylated histone. So it's going to bind in regions that are acetylated and therefore are actively being transcribed. It has this bromodomain recognising the acetylated histone and in addition to this has of course the ATPase domain. The second class that we'll talk about, is the ISWI, or Imitation SWI. So of course, by its name, you can imagine there when it was described they thought this looks a bit like a SWI-SNF, but not quite. In this case, it contains a SANT domain. And we don't yet know exactly what a SANT domain recognises, except that it appears to recognise a modified histone tail. And, of course, also has the ATPase subunit. And the final class, I called the CHD chromatin remodellers. And this stands for Chromo domain and Helicase-like Domain. So, the chromo domain, we've also dealt with before. And it recognises the methylated histone tail. And remember methylated histones can be found in different contexts. And it's context dependent, whether or not it may be an active chromatin or inactive chromatin. So, heterochromatin or euchromatin. And, depending on the chromo domain, which protein it's found in, that will alter the function. In addition to this chromo domain, they also contain a helicase domain, and helicases are normally involved in unwinding DNA at the site of transcription or the site of DNA replication. And in this case, again, they contain the ATPase that's required for chromatin remodelling. So, we know that chromatin remodellers are really essential for viability. And we know this from studies that have been performed in mouse. So in mice, what they've done is they've produced knockout strains of mice. So this means they've deleted particular genes. And if they delete factors that are involved in chromatin remodelling, then this is incompatible with life. In addition, if you delete these chromatin remodellers one at a time in a particular tissue, say for example in the skin, we know that this results in aberrant differentiation. So, and this is believed to be because although you can rescue the viability of the animal, the skin or any other particular tissue needs to rapidly differentiate at particular times. And because of the role of chromatin remodellers in moving around the nucleosomes, it's thought that the chromatin remodellers really allow for the rapid activation or silencing of gene expression. And so that's why they might be particularly important in differentiation. So one of the really interesting things to think about is how is it that these chromatin remodellers work with the histone modifiers? That is those proteins that lay down the histone methylation or acetylation marks. Well I showed you with those three types of complexes, they each recognise modified histone tails. Either the SANT domain, which recognises some form of modified histone tail. The bromodomain recognising acetylated histone or the methyl the chromo domain recognising the methylated histone tail. But moreover, there are other complexes, for example the NURD complex, which have within that one complex, within that one chromatin remodelling complex, the ability not only to remodel the chromatin, in this case through the protein called CHD3 or CHD4, it's a chromo domain protein, but also through having histone deacetylases monitoring. So it's also got a histone deacetylate which can remove the histone acetylation marks in the surrounding chromatin. One of the opening questions in this field, as well as in terms of both chromatin remodelling and histone marks, but also more broadly in the epigenetics field is what's the order in the hierarchy. So how do we understand what happens first and what happens next? Or do things happen simultaneously? So in this case, does the chromatin remodelling happen first, and then a change in histone mark? Or because the chromatin remodelling factors recognise modified histones. Do the modifications to the histone tails happen first or do they happen simultaneously? Well here it seems that probably they happen either simultaneously or the histone marks come first. This isn't something that we can broadly say will be true and it will be more likely to vary by circumstance. So this leads me on to talk about some of the open questions that there are in this field of epigenetics. So it's still a relatively young field in terms of science as a whole. Really it's taken off, since perhaps the year 2000, that we've really had a big acceleration in the amount of research performed in this field of epigenetics. So, some of the questions that are still really outstanding and that I'm going to highlight throughout our lectures how is it that these epigenetic marks can be mitotically heritable? We can explain it in some specific instances. For example, in DNA methylation, we know the mechanism of mitotic heritability, but that certainly isn't always the case. Another question which I just mentioned was, what's the interplay between different epigenetic marks? How do they work together and how do they bring about transcriptional silencing? The fourth of these questions is, what are the factors? What's the identity of the epigenetic factors that lay down these epigenetic marks, remove them, read them, move the nucleosomes around? While I've been able to describe to you quite a number of these, it still is true that in mammals we don't know the identity of all of these players. And finally, even when we do know the chromatin remodelling complexes or the particular the particular histone modification complexes. How are they actually being directed to particular sites in the DNA? How do they know where to go to change the chromatin packaging? And this is something I'll come to in the later lectures on long noncoding RNAs. So, I wanted to just highlight at this time that this fourth question down about the identity of the factors that are involved in epigenetic control is one of the things that my laboratory works on. And I'm going to try and highlight to you in just one slide, here or there, the sort of things that my lab works on, to demonstrate some of the open questions that and how we address them in epigenetics. So, the way that I like to think of it, is that in the genome, there are perhaps 25,000 or 30,000 genes. So these are all the grey puzzle pieces. They're the ones that do the work in the cell of different varieties. However, there might be about 1000 genes that are involved in epigenetic control. And these if you like, can be seen as the regulators. They're the stop/go men, the people that are describing to all the other genes and bossing the other genes around, and telling them what to do. But as I said in mammals, we don't know all of these thousand genes. We’re guessing about the potential epigenetic modifiers when we make these lists of a thousand genes. So what we're interested to do is to say, which genes actually participate in epigenetic modification of the genome and how does this differ between cellular scenarios, different cellular scenarios? And the idea about why we want to try and find all of these genes is because they are like puzzle pieces, and without having all of the pieces of the puzzle, it's really impossible to see what picture they are making. Or, in other words, it's impossible to see what the molecular makeup of epigenetic control is. And so we'd like to understand what's happening at the molecular level. And so we first want to assemble a full list of the players in this epigenetic control. So to summarise what I've told you about the epigenetic marks and also chromatin remodelling, if we take first of all at the bottom here a naive locus. Again I'm showing you the nucleosomes just as an octagon, a grey octagon. The DNA is the black string here coming through. The open circles in this case are the CpG dinucleotides. And you can see the histone tails coming off here, away from the core of the nucleosome. When a region is found to be active, chromatin remodelling complexes may have been involved in opening up the chromatin in this area, so that the nucleosomes are more sparsely packed. And then, at an active locus, you'll also find a lack of DNA methylations, so DNA hypomethylation. And along with that, you'll find histone acetylation and often methylation of H3K4 so lysine 4 histone H3. By contrast at a silent locus that is inactive, so transcriptionally inactive is shown by this cross here. The chromatin will have been more densely packaged down using the chromatin remodelling complexes. And at these inactive regions, you're likely to find more likely to find that the DNA is methylated and you'll find a variety of repressive histone marks. The examples that I've drawn here are H3K9 methylation and H3K27 methylation. In fact, this is a figment of my imagination. You would very rarely find all of these three marks together, as I've drawn in this instance, just here. But, this is just to remind you the sort of marks that we find in an inactive locus, and that the nucleosomes are more densely packed.
