So in the previous lectures, we were thinking about gene by gene, whether or not it might be hypermethylated or have aberrant histone of modifications. But now I want to think about not gene by gene, but rather groups of genes, and so the long range epigenetic alterations. So long-range means that they're actually covering megabase regions that may encapsulate tens of genes, at least many genes in these intervals that are all epigenetically abnormal. So, long-range epigenetic silencing, or l-res, and long-range epigenetic activation, which has been described more recently, were both first discovered in prostate cancer compared with prostate epithelium. What they were able to find was that the histone modifications and the DNA methylation were aberrant at large intervals spread throughout the cancer genome compared with the normal tissue. So, it seems at least for LRES, these epigenetic silencing that which has been known about for a longer period of time. That it also seems to occur in other cancer types, it's not at all restricted to prostate cancer. But where other investigators have looked they can find it in many other types of cancer. It remains to be determined if long-range epigenetic activation is the same, because mainly the groups haven't looked yet. But it seems likely that you'll both have these large regions of activation and large regions of silencing compared with the normal nucleus. There's a second terminology for this, which is a slight variant, which is called LOCKs rather than LRES or LREA. And LOCKs are regions that are large organised chromatin lysine altered regions. It's the same sort of thing. So mostly when we've been talking about post-translation histone modifications we have been thinking about modifications to lysines. So K4 methylation is lysine 4 methylation being activating, K9 and K27, lysine 9 and lysine 27 methylation. So it's a very similar idea, that, although it's not the same groups that have found these particular types of long epigenetic silencing. It's the same general consequence. So, this long range epigenetic silencing, occurs in a few different varieties, and I'll just go through these briefly. So for example, in some regions, these large regions, megabase regions, That have different epigenetic marks to normal, you'll have in the context of normal cells, active genes, shown here by these arrows. And so, in the case of the active genes, you'll have hypomethylated DNA, so not too much DNA methylation. And you can see the methyl groups are being shown here as these little circles on the DNA. Which are either open or closed being unmethylated when they're opened or methylated when they're closed. And they'll also be histone acetylation on histone H3 and H4 shown here as these triangles. But if you look in this same region now in cancer cells then each of these genes slightly variably dependent on the gene. But within a whole region you'll have DNA hypermethylation, so you'll be able to see here that there are more of those circles. Those CPG dinucleotides are coloured in and methylated. The chromatin is compacted so these nucleosomes are much more densely packaged. And indeed you have hypo-acetylated residues because acetylation is associated with activation. And then you have methylation of H3K9 and methylation of H3K27. So these are really the same features that I told you about for those CPG islands that are hypermethylated in general. However, what I'm now telling you is that they have clearer enlarged blocks. These megabase intervals so that is thousands and millions in fact of bases altogether. These were at large regions. So, in this case you're going from a context of an active gene, regions of active genes to regions of silent genes in cancer. But it's not exclusively happening where you always go from active genes to silent genes. But you can also have this long range epigenetic silencing to really consolidate that silencing signal at regions that are already inactive in a normal cell. So, say for example, in a normal cell, you have a region that has in this case H3K9 methylation shown here. It's, but it doesn't have DNA methylation and it doesn't have H3K27 methylation but it is still hypoacetylated. In the context of cancer then you might sometimes have some DNA methylation coming on. So hypermethylation in some of the genes with this interval but now you may gain an additional repressive epigenetic mark. And in this case you're gaining H3K27 methylation. So by now having 2 different epigeneticly silenced marks, so H3K9 and H3K27 methylation. This is really a consolidation of that epigenetic silencing. Meaning it would be more difficult to activate these genes that are spread, again, through a large genomic interval. Finally, within these large intervals you can have some sort of swap or an exchange of repressive chromatin marks in a whole region. So you can start out with these regions, which are generally hypomethylated, but have both H3 and H4, H3 and K9 and K27 methylation. And now what can happen is that you can gain DNA methylation, but lose H3K27 methylation. In each case, we still have inactive genes, but you've got different reasons for why those genes have become inactive. So this is perhaps the reason why if we look at DNA methylation marks as one instance genome wide. They may be better at detecting a particular tumour type than just gene expression would. Because the DNA methylation changes occur at many places in the genome, where actually as you'll see here the expression status does not change. And so this may give more sensitive signal as to what's happening in defining genotype. So to think about these long range epigenetic silencing a long range epigenetic activation. Very simply in terms of the chromosome level if we think about this normal cell which has three chromosomes. In comparison to a cancer cell in which now still has 3 chromosomes. You'll have these large regions showing in the red blocks that are all showing long range. All the genes in this integral are showing long range epigenetic silencing. In other words they are silenced epigenetically. You might in here have three distinct little suburbs if you like, so three distinct little regions. Which have in one case they were genes that were active normally. And they became silent and so now they've a particular set of epigenetic marks. But, you might have another region which has a different set of epigenetic marks and a final region with a different set again. Just like I describe to you now, there could be three options, you could have genes that were active becoming silent. Genes that were silent being consolidated in their silence with additional epigenetic marks. Or finally genes that were silent exchanging particular repressive epigenetic marks in the context of cancer. But neighbouring these regions you can also have regions of activation. So I haven't been through the examples of the particular histone marks that are found in these active regions. But the point remains that you have whole megabase spanning intervals that have become now active. So genes have become active and they have the appropriate chromatin signature to go with that activation. So why is it that you would have these large regions that are all epigeneticallly activated or inactivated together. We can't really start to claim that its to do with just the function of those genes. Because if it were to do with just the function of each individual gene. It's unlikely that the all within a large megabase or several megabase interval, would all be tumour suppressors, for example. Or all the oncogenes. So there must be something else going on there, something else that's driving the cancer. And when we think about these large regions, I think you need to start to think back again to the nuclear architecture. So, if we think about the nuclear architecture, we know that chromosomes in general are located in one region. But active regions of that chromosome loop into the centre of the nucleus and inactive regions loop out to the nuclear periphery. So one potential consequence is that there's actually an alteration in the underlying nuclear architecure. And so we are just reading that out as these long range epigenetic silencing. So, we don't really know very much about the specifics of the nuclear architecture yet, in the context of cancer. But while we don't know much about the specifics, because these studies are definitely still ongoing, we know that aberrant nuclear architecture and unusual nuclear organisation or nuclear disorganisation actually, has been used to diagnose cancer for over a 140 years by pathologists. So, if you take a cancer cell and a normal cell, and you put them side by side, the nucleus of each is quite distinctly different. So, we know there can be different nuclear size. There can be different nuclear shape. The ploidy can be different, in other words how many copies of the chromosomes they have. So in a normal cell you'll have two copies of each autosome, but this isn't usually true in a cancer cell. They'll often have the wrong number of chromosomes. So maybe they're tetraploids that have four copies of every autosome. And you can see nuclear organisation with normal histiological stains. So as you stain for the nucleus, you can see slightly denser staining regions and slightly lighter staining regions. So you can see effectively hetrochromatin versus euchromatin. And this will also commonly be abnormal in the context of cancer. So we know that nuclear architecture is basically one if the hallmarks, if you like. Or aberrant nuclear architecture is one of the hallmarks of cancer that's been used for more than a century to be able to diagnose a neoplastic tissue. There are fundamental differences but we still don't know what the specifics are. So hopefully in the coming years we'll be able to know more about how the nuclear architecture is organised, is different in cancer as well. At specific, levels, perhaps there are different regions that are associated with the nuclear periphery. Perhaps there are different regions associated, in the internal part of the nucleus. And this will hopefully lead to further understanding of how cancer progresses. In the next lecture we're going to, be thinking about disrupted non-coding RNA's and cancer.
