So, the final class of non-coding RNAs we want to think about are the long non-coding RNAs. So, as you'd suggests what, from the name, these are much longer than those short RNAs we're talking about before the piRNA's or the microRNAs. And in this case, they're now more than 200 nucleotides in length. In addition, according to their definition they are usually spliced. So, they undergo splicing like a regular messenger RNA would. They are capped so they have a five prime cap like a messenger RNA would. And then have polyA tail so they're polyadenylated. They also have the same sort of promoter as a regular gene would have. And so they're subject to expression in a developmentally controlled manner. But what's interesting about them is they tend to be or they should be constrained to the nucleus, at least predominantly. And this is looked at by using a method known as RNA fluorescence in situ hybridisation. So you look at where the RNA is within a cell. Using the homology of a probe that can bind to RNA in particular. And you want to see for long noncoding RNA that the vast majority of the RNA is staying with the DNA in the nucleus. So the numbers of RNAs that long noncoding RNAs that are found in the genome the estimates vary dramatically. And so perhaps there is 10,000 to 200,000 that are found in mammalian genome. The vast majority of these found at very, very low copy numbers. So there may be only a few copies of these, these long noncoding RNAs in each nucleus. So the question now becomes what are these long noncoding RNAs doing? And are the ones that are found at such low copy really doing anything, or are they perhaps just the result of transcriptional read-through, or transcriptional noise in the genome. What's interesting about the ones that have been characterised is that these long noncoding RNAs appear to regulate epigenetic processes. And this is what's interesting for them in relation to epigenetic control in this course. So these long noncoding RNAs can regulate many different processes. And as I said, there are a very large number of them. But there's a smaller set of them that have been characterised in any detail. We're going to talk about Xist as well as HOTAIR, and these two long noncoding RNAs have roles in X inactivation for Xist and Hox gene silencing for HOTAIR. You'll recall X inactivation, the dosage compensation mechanism in female mammals. It actually involves a very large number of long noncoding RNAs. But, the prime example is Xist. And this was indeed, the first long noncoding RNA to be discovered back in 1991. There are many more that have been discovered since that have a role in X inactivation, including Tsix. Genomic imprinting, which I introduced in the last lecture just with a couple of sentences, this mono allelic expression of genes between that differs, or is based on, the parent of origin of the gene. Also can involve long noncoding RNAs that have been very well studied. And Hox gene silencing involved HOTAIR which we'll talk about. This long noncoding RNA called HOTAIR. So, Hox gene silencing we have a large number of Hox genes in our genome and these are involved in setting up the segmentation through the body axis. So, for example if you think of your vertebrae. So each Hox gene needs to be on at a particular region, and then is switched off in other regions. And this switching off involves this long noncoding RNA called HOTAIR. Finally there are other, many other processes that are regulated in, with long noncoding RNAs including DNA damage response. So, what do these long noncoding RNAs actually do? Well, you'll remember one of these longstanding questions in the field of epigenetics that I brought up in the previous lecture was, how are the epigenetic complexes so, the complexes, the laying down of DNA methylation or histone modifications, or that are moving around the nucleosomes, the chromatin remodellers, how are these actually targeted to specific sites in the DNA? We know that most sites actually don't have much DNA binding specificity. so your most complexes don't have the DNA binding specificity all of their own. So we also know there are transcription factors, these factors that that would bind with RNA polymerase. And they can provide some sequence specificity. However, the sequence specificity that's provided by a transcription factor is, tends to not be, particularly high. So they may recognise maybe a 6 base recognition sequence or even a 12 base recognition sequence. But if you calculate the number of times that these 6 or 12 base recognition sequences can be found in the mammalian genome. It doesn't allow for unique targeting of any complex but rather there'll be thousands of these sites spread throughout the genome. So finally we also said that we can, the chromatin remodelling complexes, for example, will be targeted based on previous histone marks or histone, they will bind to histone, other histone marks. And we know also there are some epigenetic complexes that lay down a histone modification that also bind a previously made histone modification. But again this doesn't allow for that specificity. We don't get to come down to just a few loci, or one allele in by using just these mechanisms. So, the features of long noncoding RNAs somehow allow them to direct these epigenetic complexes, at least in some instances that we know about at the moment. And they appear able to do this in cis, so in other words. On the chromosome from which they are transcribed. Or in trans, so acting somewhere else. They do this amongst other functions. There are many functions that we'll come to, and we're just going to go through these ones that are concentrating on them directing the epigenetic machinery. So, what are the features that allow them to be involved in this targeting, and especially to allow this allele-specificity? Well, because they are transcribed. While they are being transcribed from a particular locus from a particular chromosome they are still tethered there. They're still tethered to the place from which they're being transcribed. So this allows for that allele specificity. And it also shows how they might act in cis. Now the reason they can have provide specificity that's greater than that of a transcription factor is simply because of their length. Because of the length of the molecule, the length being over 200 nucleotides in length and in fact most long noncoding RNA is about which we know something at the moment are several kilo-bases or tens of kilo-bases in length. So this means thousands of bases, of course. This really gives unparalleled site-specificity to these long noncoding RNAs. But it's important to note that this specificity and to the allele or specificity based on the sequence is not necessarily always required for their function. So what are some of those functions? What are some of the mechanisms by which they work? The two that we'll focus on are long noncoding RNAs acting as guides, and the example that we like to talk about here is Xist. We'll also talk about how HOTAIR acts as a guide. The difference being, that Xist acts in cis, whereas HOTAIR acts in trans. And I'll go through this in more detail. We know that these long noncoding RNAs can act as scaffolds. And again HOTAIR is the example here. But in addition to these two mechanisms, there are many others that long noncoding RNAs many other mechanisms by which long noncoding RNAs can function. They can act as decoys. They can act as signals, and what this means is that it's actually the act of transcription itself running through a locus, which has the function rather than the product, rather than the long noncoding RNA that's produced. They can act as enhancers and this is has always been shown to work in cis. So, you can have transcription at one point which alters transcription nearby. And they can even act as reservoirs. And in this context, H19, which is one of the best characterised long non coding RNAs, acts as a reservoir of a micro RNA. So it can sit around being a stable long noncoding RNA, that can then be processed to release the microRNA that it contains. So, while there are so many different ways that these long noncoding RNA's can actually work, there is a common feature. And that is that there's the formation of an RNA protein complex. And an, this RNA protein complex is able to influence gene expression. So I'll just briefly describe by what I mean when things act as a guide versus a scaffold and then in the next lecture we'll go through an example of each of these. So as a guide what I mean is that this epigenetic complex which I've drawn here will bind to the noncoding RNA. The noncoding RNA is drawn in red. And so we have this long non-coding RNA coming along. And it is tethering an epigenetic complex to, a particular region. The reason that it's site specific is that there is, through here, this sequence identity, which is allowing this RNA to bind to the DNA in a sequence specific manner. In other words, the long noncoding RNA is guiding the epigenetic machinery. This epigenetic machinery can then participate in whatever it normally does. In this case, histone methylation. So in this way one complex is being guided to a specific site in the DNA. However when these long noncoding RNAs act as a scaffold now, rather than just recruiting one complex, they tend to recruit more than one, in this case two. So different regions of the long noncoding RNA. So, this region here, or the region down here, combine to two different epigenetic complexes. And so in this way they act as a scaffold. They unite two complexes so that they are tethered to one another, and act as a molecular scaffold to bring epigenetic complexes somewhere else. This scaffolding may not necessarily require that their long noncoding RNAs provide sequence specificity. But, you have then got tethered together these two different complexes which are then able to act on chromatin and at the same location tethered by these scaffolding long noncoding RNAs. So in the next lecture we'll talk about examples of each of these cases and go through their particular functions.
