Epigenetics and reprogramming - turning back the clock
We've all heard of the genetic code - the letters that spell out the instructions in our DNA. But that's not the whole story. Researchers are increasingly digging into the epigenetic code - the marks that tell cells which genes to use and which to ignore. Plus, we take a look behind the headlines about older fathers and autism, find out what chimps can tell us about our cancer risk, and our gene of the month might be mistaken for a heavy metal band.
In this episode
01:07 - The epigenetic code
The epigenetic code
with Professor Tony Kouzarides, Gurdon Institute, Cambridge
Virtually all the cells in our bodies contain the same DNA. This is the same DNA we were given when we started life as the fusion of an egg and sperm, with half our DNA coming from mum and half from dad. But we have hundreds of different types of cells - from blood to bone, liver to lung and everything in between. So how does this huge diversity arise if all these different cells have the same DNA?
To get some answers I spoke to Professor Tony Kouzarides from the Gurdon Institute at the University of Cambridge.
Tony - Well, the first thing to say that differences come down to different parts of our DNA being turned on and off - in other words, different genes in different types of cells. What we now know is that there's a crucial part of a cell that allows this turning on and off which is a set of proteins called histones around which the DNA is wrapped up. So the DNA is not naked. It's actually occupied and entangled in a set of proteins that control its activity.
Kat - So, it's like a massive load of string actually being organised and wrapped up.
Tony - Yes, DNA is very, very long in each of our cells and has to be put into a very small area. It's about 2 metres long. Each DNA in our cell is 2 metres long.
Kat - And a cell is absolutely tiny.
Tony - It's really, really tiny. And the only way to actually fit the DNA in is to compact it around these proteins. Now because it's very, very dense in these cells, one of the ways in which it has to be controlled is to unwind it and allow a gene to be available to be turned on and off. So, it's like being camouflaged in this compact structure.
Kat - It has to be unwound so that you can read it, so you can actually do something with the gene.
Tony - And use it, that's right. So, genes have to be used and they have to be unwound to be used.
Kat - So, what do we know so far about how DNA is packaged up and some of these signals that are put on it to tell a cell whether to use it or not use it?
Tony - Because it is packaged, there are signals that allow it to be unpackaged and these signals can be coming from the outside that could be environmental, that could be cellular, and they end up by being small, very tiny changes in these protein called histones, and on the DNA ,which allow the unwinding. But also, allow the machinery to come and tell the gene to turn on and off. So there's machines, cellular machines that come and use DNA and make products or turn off products and these are signalled to by these very small changes in histones.
Kat - I've seen images of these and you sort of have the DNA as a string wrapped around these balls of histones, and it's almost like you have little chemical flags on the surface that say, "Here we are, come and use this one. Don't use that one." Do we now really know a lot about the identity of these flags and how they work?
Tony - Well, we know quite a lot about the identity. There's many different types of marks as we call them on chromatin which is the structure - this histone/DNA structure is called chromatin. And what we know a lot about is the enzymes that put on these marks, the different changes on the histones. These are very interesting enzymes because they control many biological processes. In fact, almost all biological processes that involve the regulation of DNA deal with these enzymes, or the other way to look at it, these enzymes control all the biological processes that DNA is involved in.
Kat - So a gene is not going to get switched on without these enzymes, and it's not going to get switched off without these enzymes too.
Tony - Yes, these are crucial enzymes and now, we're discovering that these enzymes are involved not only in the normal way that a cell works, but actually, are misregulated in any disease. Primarily, the one we know of cancer.
Kat - So, what sort of changes do we see in these histone marks in cancer cells?
Tony - Because there's so many of these little changes, we can categorise them and catalogue them, and we can start to understand how they change. But really, that's just cataloguing. What we now are looking for is the genes that change when these enzymes that have put these little tiny marks on are damaged in cancer cells. So, the important thing is the genes that have changed rather than which marks are where.
Kat - You can almost embark on some kind of grand cataloguing thing and say, "This changes, this changes, this changes..." But if it's not actually changing the patterns of which genes are on and off, then it's not making a difference.
Tony - Exactly, and the way we know that the genes are important because now, there are small tiny molecules that can affect these enzymes. So, these are pharmaceutically relevant inhibitors for example of enzymatic activity, you kill the enzyme by adding this small thing to the cell, small chemical, and then you can see that a very small number of changes in genes that come on and off. So then you know that those genes are important in cancer and they are driving the cancer itself.
Kat - And presumably then these molecules, these chemicals that you have that can change these patterns, could they potentially be useful for treating cancer?
Tony - Well in fact, that's the whole point of developing these molecules, because pharmaceutical companies now realise that this area is very exciting with respect to pharmacological intervention against cancer. And therefore, they're developing these small molecules to treat patients and we as a lab are collaborating with these companies to work out exactly how these small molecules work to inhibit cancer.
Kat - We first discovered the structure of DNA in 1953 when Watson and Crick figured out that it was a double helix. What happened after that? How long did it take to really figure out this level of regulation of DNA and how it's controlled?
Tony - For some time, it's been known that the DNA is packaged into the cell in this way by wrapping around histones. But really, the modern era of regulation came when the first enzymes were discovered that put down these little marks that change genes, and this was in 1996. It's a very new field and it's progressing very rapidly because a lot of scientists are realising that all the work that they've been doing in the past may relate to how chromatin and histones change, and this is called epigenetics.
Kat - What do you think the potential is for epigenetic-changing drugs in the future for treating cancer and treating other diseases?
Tony - I think it's a major untapped source of targets for many, many diseases, and it's really the tip of the iceberg with what we know at the moment. It's a bit like saying, 'genes control everything'. Well actually, they don't. Our environment controls a lot of what our genes do and this is the missing link - how is it that external events can actually control what our genes do and how our genes can be misregulated in disease. That comes all down to epigenetics.
Kat - That was Professor Tony Kouzarides from the Gurdon Institute at the University of Cambridge.
08:20 - Older fathers and autism
Older fathers and autism
with Nell Barrie, science writer
Kat - It's time to take a look at the top stories from this month with science writer Nell Barrie. Writer. So, there's really been one big story in the news this month and this is about older fathers. What's this one about?
Nell - So, this was looking at the number of mutations that old fathers tend to pass on to their children and specifically focusing on how that might increase the risk of having children with autism.
Kat - Because the headlines were really dramatic. They were saying, "Old dads causing autism and schizophrenia" and it was enough to really worry all my older male friends. What have they actually done in this study?
Nell - So, they were looking at how many new mutations fathers pass on to their children. So, that's mutations that arise in the sperm, not mutations that the father has himself. And what they found out by looking at a whole group of different families and comparing the genes. They looked at 78 groups comprising the mother, the father, and the child. So they were figuring out what mutations in the child weren't present in the mum or the dad, and where those mutations have come from. And they've discovered that the older father, the more new mutations he'll pass on to the child that have popped up in his sperm.
Kat - And in these children, they were more likely to have these conditions.
Nell - Yes, exactly. So, it looked at autism and schizophrenia and found out that children with more of these new mutations were more likely to have those conditions.
Kat - So, what does that this actually mean because on the surface, this is like, "Oh my God! All these mutations are causing autism" and older dads maybe - should they not have kids?
Nell - I think probably the main thing to bear in mind is that autism is a really, really complicated condition. So it's certainly not a case of saying, "You've got this gene. Therefore, you have autism." It just does not work like that. We know that it's quite heritable, meaning that in families where there's lots of autism you can see it passing through the generations, but that doesn't mean that that explains all of it. We also know that it seems to be becoming more common and people aren't quite sure why that is. So, I think a lot of the kind of hype around this was maybe that perhaps this explains why more children have autism now because dads tend to be older. But it's not that simple. It's not that easy to make that connection based on this research. We don't quite know enough yet, but it's certainly very exciting to see that that could be a cause of this perhaps.
Kat - Because I remember a couple of months ago in the Naked Genetics podcast, we covered a story just discussing that they'd discovered a whole bunch more of potential mutations that were involved in autism, and I think it's an incredibly complex disease. But it's also interesting about - there's a nice little vignette about the way that fathers' ages have changed in Iceland because probably round about the early 1900s, the average age from a dad was about 35 in Iceland and that dropped to under 30 during the 20th century, and then it's gone back up to 33. So in fact, fathers were older. I don't know if that means that the rates of autism have changed. It's very hard because we diagnose it differently now.
Nell - Yeah, it is really difficult because I mean, a lot of the research has made people think that perhaps some of the rise is due to just people knowing more about autism. So, children are more likely to be diagnosed because doctors are kind of looking for it. They're more aware and this is sort of perhaps arguing that maybe some of it is real, maybe it's to do with older dads, but we don't know about those patterns further back in time. So, it's not very clear what the sort of the changes could have been and I think it's quite interesting just thinking about it from a kind of social point of view I guess as well because we tend to see families being a little bit older when they have kids now, but that might change in the future. And it's to do with the way people think about how they plan their lives, their jobs, all that kind of thing. So, it's an interesting kind of crossover between genetics and looking at a whole much bigger area as well.
Kat - They were saying, "Maybe men should bank their sperm." You know, bank your sperm in your early 20s and then get it back out of the freezer again if you want to be a dad. But I think one of the really interesting things to think about is what does this mean for us as a species because obviously, mutations are good. They're what drive the natural variation in populations and this is what drives evolution and an awful lot of mutations are just neutral. They're not bad and they're not good. Obviously, some mutations are good. So maybe this increased rate of mutations in dads could be a good thing, do you reckon?
Nell - Absolutely, yeah and I think - I mean, there were some commentary about this saying, "Oh well, yeah, doom and gloom." This could just mean that our genes are getting progressively worse as mothers and fathers get older over time. And I sort of thought, "Well, it's all relative because the mutations that you have are only bad or good, depending on the environment that you're in." So, as our environments changes, the way people live their lives change, you don't know what's a bad or a good mutation, and it's all kind of how you use it I guess.
Kat - Well it's certainly really interesting developments in what's an on-going and quite complicated story. So, thanks very much for chatting about it, Nell. That's Nell Barrie, Science Writer.
13:16 - New genome encyclopedia
New genome encyclopedia
More than 400 scientists from 32 research institutes around the world have published the most comprehensive analysis of the human genome to date in an epic series of 30 papers published in Nature, Science and other journals. ENCODE - the Encyclopedia of DNA elements - aims to provide a comprehensive database of all the functional DNA elements within our genome, including genes, regulatory switches and much more.
The researchers analysed nearly 150 different types of human cells, and concluded that at least 80 per cent of the whole genome is functional in some way - although less than 2 per cent actually codes for proteins - finally laying to rest the idea that most of the genome is "junk DNA". The finding is still controversial however, as some scientists think that much of this is still genetic clutter, accumulated during evolution.
As well as detailing functional elements in the genome, ENCODE also acts as a massive catalogue of gene faults and variations that may be involved in a wide range of diseases, providing fresh leads for researchers around the world.
14:20 - Epigenetic analysis from blood spots
Epigenetic analysis from blood spots
Keeping with the theme of our show, our other news stories are focused on epigenetics. Scientists led by Dr Vardhman Rakyan at Queen Mary, University of London, have shown that purified DNA from Guthrie cards - the filter papers used to collect tiny spots of newborn blood - could reveal important information about the epigenetic state of the child's genome, as well as just their genes. Writing in the journal Genome Research, the scientists analysed the epigenetic state of samples of DNA from Guthrie cards and compared it to fresh DNA, finding that it was a reliable match.
The small blood spots on Guthrie cards are used to test for important diseases such as phenylketonuria and cystic fibrosis. But as scientists find out more about the role of epigenetics in disease, being able to retrospectively analyse these samples could be a vital tool in helping to predict the risk of diseases such as cancer, heart diseases and diabetes, as well as helping researchers to understand what's going on as these conditions develop.
15:17 - Genetic switch controls hereditary heart defect
Genetic switch controls hereditary heart defect
Writing in the journal Developmental Cell this month, researchers led by Dr Anne Voss and Dr Tim Thomas at the Walter and Eliza Hall Institute in Australia reveal a genetic "switch" that might explain variations in how severely children are affected by an inherited heart condition called Di George syndrome, which affects roughly one in every 4,000 babies. The condition is caused by a fault in a specific region of human chromosome 22, and many of the symptoms are linked to loss of a gene in that region called Tbx1, although they can range from mild to severe. Even identical twins with the same mutation may have different conditions, suggesting than non-DNA factors are at work.
Using a mouse model of Di George syndrome, the researchers discovered that levels of a protein called MOZ control the levels of Tbx1, which in turn control how severe a child's symptoms are. MOZ is an enzyme that adds epigenetics marks to histones - the proteins that package DNA - so it acts independently of mutations in the underlying DNA sequence. Intriguingly, the researchers also found that defects were more severe in pups born to mothers whose diet was rich in vitamin A but also had low levels of MOZ. This research sheds light on the interactions between mutations, epigenetics and diet, helping to explain more about the complex factors that underlie birth defects.
16:38 - Sea lampreys shed genes
Sea lampreys shed genes
A new study in the journal Current Biology shows how sea lampreys get round the problem of controlling when and where to switch genes on and off - they simply get rid of them. Sea lampreys are simple, parasitic eel-like vertebrates, which are considered to be a pest in some parts of the world. The researchers, led by Chris Amemiya at the Benaroya Research Institute in the US, discovered that lampreys shed more than a thousand genes from their cells during development - around a fifth of their whole genome. The only cells to keep all their genes intact are primordial germ cells, which go on to make eggs and sperm.
Looking closer, the scientists found that many of the lost genes are involved in pluripotency - the key characteristic of stem cells. In higher species, these genes are only needed during development or in stem cells, and have to be switched off. But they can be reactivated in diseases such as cancer. By simply getting rid of these risky genes, the sea lampreys sidestep this problem altogether. The researchers believe their findings will help to shed light on how gene regulation has changed through evolution, both in health and in disease.
17:45 - Epigenetic differences between chimps and humans
Epigenetic differences between chimps and humans
Soojin Yi and her colleagues at Georgia Tech may have found an explanation for why humans are so different from chimps, even though we share 96 per cent of our DNA with our furry friends. Publishing in the American Journal of Human Genetics, the researchers studied samples of brain tissue from both species, looking at levels of DNA methylation - an epigenetic mark that can switch genes off. They found hundreds of places where methylation levels were significantly lower in humans than in chimps, with an unusually high proportion in genes related to diseases such as autism, alcoholism and other addictions and even cancer.
Intriguingly, chimps have a lower risk of cancer than humans, something the researchers think may be linked to the difference in methylation. Although it's still early days for this research and a lot more work needs to be done to show how significant these differences are, the findings hint that epigenetic differences arising during evolution could have implications for a range of human diseases.
19:01 - Reprogramming cells
with Professor Mandy Fisher, MRC Clinical Sciences Centre
We've already heard how epigenetic marks tell cells which genes to use, creating different types of tissue. But is this always a one-way process? And can we ever turn the clock back on cells that have decided their fate? The answer from the lab seems to be yes.
To find out more, I spoke to Professor Mandy Fisher, Director of the MRC Clinical Sciences Centre in London.
Mandy - I think in the vast majority of cases of different cells and different tissues of the body, if they do divide, they give rise to more of the same. And in general, they don't go backwards in terms of being less specialised. However, experimentally, you can take cells from the skin and you can reset their developmental clocks, turning them right back to completely unspecialised cells, similar to those that you find shortly after the egg meets the sperm.
Kat - So these are the embryonic stem cells.
Mandy - Exactly so. So, this is a huge opportunity because it at least in principle suggests that you may be able to take skin derived from a patient and make bespoke stem cells to regenerate certain tissues. Now, we're a long way from that, but in principle, if that were possible, it would be extremely advantageous. You'd have none of the problems of rejection that normally plague many stem cell replacement therapies.
Kat - When you do these experiments, when you take say, a skin cell and you reprogram it, what are you doing to it to set the clock back?
Mandy - So, different investigators think about this in sort of different ways, I would say, but for me, a nice way of thinking about it is really that you're erasing or removing all the marks on the DNA so that it is the blueprint which you can then program to become something different.
Kat - How do you do that? How do you take these marks off? How do you put them back on in a different way?
Mandy - So experimentally, there are three sort of tried and tested approaches that's something called nuclear transfer and this was pioneered by people like John Gurdon.
Kat - Basically, cloning.
Mandy - Exactly so. So, a second route that's become very famous in the last sort of 5 years was pioneered by Japanese scientist called Shinya Yamanaka, and that is something called IPS. And this is a way of introducing transcription factors - proteins that bind DNA and turn genes on - introducing this cocktail of factors into skin cells, and reprogramming them by that route.
Kat - So, you basically chuck in a bunch of stuff and it erases the marks and it resets these to stem cells. That's incredibly powerful.
Mandy - It is amazingly powerful and I think all good money would be on Shinya for winning a Nobel Prize for this. It's an amazingly brave experiment that he did.
Kat - And what's the final way that you can reprogram cells?
Mandy - The third way is an experimental method. It's arguable whether it occurs to any extent in living tissue normally and this is by fusing a skin cell to an embryonic stem cell. So, this is a cell fusion method, which was developed many, many years ago by Henry Harris in Oxford, and I think it's beginning to gain popularity and being used widely to try to understand the underlying mechanisms by which you can turn back the clock. It's not going to be useful probably in any kind of therapy. It's really to try to get at what is provided by the embryonic stem cell that allows the fibroblast or the skin cell to reset its clock.
Kat - Yamanaka's method of making these stem cells by chucking in cocktail of factors, how are these factors found that can reprogram cells? Are they normally found in the body doing reprogramming?
Mandy - He did a huge screening protocol. As I said, an extraordinarily brave experiment to try to find what was required to reset the lineage clock, if you like. The factors that he pulled out from that screen turned out to be factors that are present very early in the developing embryo. So, I guess he could have taken a good guess and not have gone through the various labours of screening and asking questions about all of these factors. But he came up with the goods and he showed that four, just four factors were sufficient to reset the fibroblast clock, and turn those cells back into cells that very much resemble embryonic stem cells. And I guess what people now are trying to do is to use that same thinking, that same screening protocol to ask whether you can turn back the clock to slightly different stages in development to make somatic stem cells that might be important for blood formation or cardiac stem cells for repairing the heart rather than going all the way back to the fertilised egg, the embryo.
Kat - And a very speculative question, how far away do you think we are from seeing this kind of technology actually making it to clinical use. Give me a guess.
Mandy - Well, I would say that your guess might be as good as mine. I think the experimental evidence is accumulating, but it's a great opportunity. There are massive concerns about safety and I think we're at least, in my view, 5 to 10 years away from being able to apply that to a clinical situation.
Kat - Are you excited about it?
Mandy - Yes, I am. I think it's probably one of the most interesting fields to be working in right now. I think we got to be cautious and make sure all of our thinking is good, but I think it's a hugely exciting time.
Kat - That was Professor Mandy Fisher from the MRC Clinical Sciences Centre.
How do mutations happen?
Louise - This month's question is being answered by Dr. Philip Zegerman from the Gurdon Institute in Cambridge. Kashefa Farooqi emailed us to ask, "At the molecular level, how does mutation occur in the genome and how does exposure to radiation for example cause a point in mutation in the
Philip - And so, our genome is really just a long organic molecule of repeating units which are called bases and it's the order and composition of these bases that really determines your genome, what your genes do. And if you do get changes to the base composition or the organisation of these As, Cs, Gs, and Ts, that's a process that's called mutation. So this polymer, this organic polymer that is your genome is actually sensitive to damage by chemicals or high energy radiation. What UV does, for example, is it causes these bases that your genome is made out of to stick together really in a process that's crosslinking. And x-rays have sufficient energy that actually make your DNA helix, your double helix to actually cause it to break. So it can actually cause the backbone of the DNA to be nicked or broken, and these are called strand breaks. I've told you that the genome is essentially a long string of bases, 3.2 billion of them in the case of humans, but the important point is that every cell in the body has a perfect copy of your genome which means that every time your cell divides, it has to make a perfect copy of this 3.2 million bases. This essentially involves a process where your DNA is duplicated by a special group of enzymes which are called DNA polymerases. And these polymerases travel along your DNA, copying it as faithfully as they can, but if they hit a problem, for example, a strand break, a lesion or perhaps two bases that have been stuck together by UV light, then your DNA polymerases can't replicate very well, and they have a high propensity to cause errors.
27:48 - Gene of the month - Mothers Against Decapentaplegic
Gene of the month - Mothers Against Decapentaplegic
with Kat Arney
And finally, our gene of the month is Mothers Against Decapentaplegic - which sounds more like the name of a death metal band than something in the genome. As with so many of these unusually named genes, it was first discovered in fruit flies - ah, those wacky fly geneticists.
It all starts back in 1982 with the discovery of a gene named decapentaplegic, which makes a protein that is vital for fruit fly development. Decapentaplegic is important for the creation of 15 imaginal discs inside a fly larva - these are the precursors of adult organs such as the antennae, wings, limbs and more. Flies with faulty decapentaplegic don't develop any of these structures, hence the name - decapenta means fifteen, while plegic means paralysis.
Mothers against decapentaplegic was found in 1995, named as a humorous nod towards campaigning organisations such as Mothers Against Drunk Driving. If a female fly carries a fault in the gene, it acts to switch off decapentaplegic in her embryos. Further research revealed a whole family of related proteins across many species, including humans, collectively known as SMADs. These proteins are involved in sending signals inside cells, telling them to stop dividing. Unsurprisingly, faulty SMADs are implicated in cancer, and they're an active topic of research for many scientists around the world.