Unravelling Epigenetics

Stefanie - Yeah, so epigenetics is a very important factor in ageing and that's something that's become more and more clear in recent years. Specific traits or likes or dislikes, I don't know how much that is down to epigenetics, but I wouldn't be surprised if that is the case.

Apart from enabling us to understand how cells function and develop normally, epigenetics can also give us some new treatments for a range of diseases including cancers and Haematologist Dr. Mark Dawson from Cambridge University works on this very question... 

Chris - I presume your case Mark, it's chiefly going to be diseases or cancers of the blood, leukaemia, is that you're interested in?

Mark: -   That's right.  I spent part of my time at Addenbrooke's Hospital caring for patients with various types of blood cancers including leukaemia as you just said.

Chris: -   Which type?  You've done some pretty instrumental studies looking at the role of epigenetics in these leukaemias.  Which ones were you looking at in particular because they're quite a big family of diseases?

Mark: -   Broadly speaking, my research interest is about the acute leukaemias.  So, these leukaemias are cancers of white blood cells, and somewhere between 5 to 8-10 people per 100,000 of the population develop an acute leukaemia. 

What's sad about this disease is that despite progress in modern medicine and supportive care, we've made very little progress in trying to cure the vast majority of these patients.  In our current therapy, only less than 30% of these patients actually get cured.

Chris: -   When we look back over the history of treating leukaemias though, once upon a time, we just knew people had some kind of disease that made them have enormous numbers of the wrong sort of white blood cells in their blood, and we knew that they got sick and we knew that giving them drugs that were horrendous but killed off those cells since to make them live longer.  Now, we understand quite a bit about what's going on genetically about these disorders.

Mark: -   That's right.  So now, if you take acute leukaemia for instance, we know in over half of these patients what the initiating event is.  So, this is a genetic abnormality that is called a chromosomal translocation.  And what happens here is that part of a chromosome breaks off and fuses abnormally to a completely different chromosome.  What this results in is the fusion of genes from two different chromosomes together.

Chris: -   So if I had say, chromosome  number 1 and a chunk of fell off there, it could take itself to chromosome  number 2 and stick itself on instead of the equivalent chunk of chromosome  number 2.  Where the two joined, I've now got two sections of genetic material linked together, and I've effectively made a new gene by linking that chunk of chromosome to the other chromosome.

Mark: -   That's exactly right.  These are called fusion genes and they're only really present in the cancerous cells.  So, none of our normal cells whilst they have two original copies of what these genes are.  They don't have the fusion together.

Chris: -   And what do those fusion genes do?

Mark: -   The vast majority of them serve to initiate or drive the process of leukaemia and the fusions I've been studying more recently are called the MLL fusions.  So here, one part of this fusion is a gene is called the MLL gene and this produces a protein related to the proteins that you and Stefanie have just been talking about, an epigenetic regulator.  And what it does is it binds very specifically to a small set of genes and it modifies them by laying down a chemical group on the histone protein surrounding these genes.  And what this modification does is really prepare this gene to be turned on.  So the MLL gene is involved in the initiation of gene expression.

Chris: -   So, there's a normal gene there whose job it is to go to a certain part of the DNA of a cell and say, "Right, you are going to turn on and your gene is going to be expressed."  So when we then take that gene and fuse it to another gene, making this funny fusion, it retains the ability to address itself to different bits of the genome, does it?

Mark: -   Exactly.  Its targeting potential is still preserved.

Chris: -   What else happens beside that?

Mark: -   So, what happens is what we've realised more recently is many of the fusions that the MLL genes stuck towards - the other genes - are also genes that code for proteins that are involved in the process of gene expression.  But these proteins really continue the process of gene expression and are involved in the completion of gene expression.  So, what these fusions eventually do is they form this sort of turbo charged driver of gene expression.  It increases the expression of several of the MLL target genes, but also increases the expression of genes that should normally be turned off at this stage.

Chris: -   So, you end up with this fusion protein, taking a very potent turn on signal for genes to lots of different places in the genome that wouldn't normally get turned on like that.

Mark: -   Correct.

Chris: -   And that starts the cells dividing and growing in this abnormal cancerous way.

Mark: -   That's right.  Many of these genes that are turned on in this way actually confer upon these cancerous cells a survival and growth advantage, and that's eventually what leads to the process of leukaemia.

Chris: -   Begging the question, now you know that, could you intervene and interrupt the ability of that funny protein to address itself to all these different places in the genome and turn it off?

Mark: -   Yes, so that's exactly what we've been studying.  We've been trying to understand.  What's clear is that these fusion proteins don't work in isolation.  They work with a number of collaborators.  Other proteins, they perform as what we call a protein complex and each component of this protein complex plays an important role in the regulation of this gene expression. 

We looked very deeply to understand what are the components of the MLL fusion protein complex and we found that there are part of this complex that contain a group of proteins whose job it is to anchor the MLL fusions at these specific genes.  They do so by using a special binding module that recognises a chemical group on the histone proteins.  So, binds this and locks on there and keeps the whole protein complex there.

Chris: -   So, you've understood what the molecular Velcro is that sticks this strong expresser of genes to the wrong bits of the genome. So have you got some way of interrupting that process?

Mark -   That's right.  So in collaboration with GSK, a pharmaceutical company, we've developed a molecular decoy so to speak.  This small molecule largely mimics the chemical modification that we see on the histones.  And so, this works as a decoy to draw away the protein complex from being locked on to the genes to away from the genes, and then now, the genes can actually be turned off.

Chris -   Was this in mice or people?

Mark -   So, we use this in a number of different studies, a number of laboratory models of leukaemia, including studying cells and how they grow in a dish, what genes they turn on and turn off, and how this small molecule turns these genes off.  We also were able to show very successfully in models of leukaemia, mouse models of leukaemia, that this small molecule really confers an impressive therapeutic advantage for these mice.

Chris -   Are you doing this in humans now?

Mark -   So, based on our study and other studies from around the world, phase I clinical trials with this compound in patients with cancer have already started in the US, and we're hoping that this will follow on and be rolled out to the UK, where some of our patients can be involved.

Chris -   I suppose it's relevant with David Cameron saying he wants everyone to be in a human DNA database across the country and marry that data to our medical records.  But this, I suppose we should emphasise is one particular kind of leukaemia, and you've had to do this sort of DNA detective story to work out how this disease occurs.  We couldn't just assume that the same compound is going to work in a range of other diseases because they're going to have a different process, aren't they?

Mark -   That's exactly right and we know that is true for this compound.  It works in a very specific subset of cancers, leukaemia being one of those cancers, but it's not a panacea for cancer.  So, it's not likely that this will be what we've been looking for to cure all cancers because different cancers are driven by very different genetic events.

Chris -   Mark, thank you.  Mark Dawson from Cambridge University.

This is a difficult area to research, because you can't conduct a randomised trial like you normally would to test a hypothesis. But a new study gets to the bottom of this tricky question. The verdict? Having children seems to lower your chance of having a long life!.

What scientists would like to do is take a group of people, and randomly assign some of them to have children, and others not to, but this isn't possible, or ethical.

Prof Agerbo and colleagues from Aarhus University Denmark attempted to get around this problem by using a natural experiment. They followed couples undergoing IVF treatment, and examined the differences between those for whom the treatment was successful, and those for whom it wasn't.

Why is this an improvement on previous studies? Previously, couples who are childless out of choice and those who are unable to have children have been put in the same category- this study avoided this, as all childless couples had been through the IVF process, so were not childless by choice.

They also looked at those parents who adopted children, and compared them to childless couples, and those who had a biological child.

Criticisms of previous studies have suggested that the differences found may be due to the fact that not being able to have children is an indicator of poor health.

In this study, none of the participants, even those who ended up with children, had been able to conceive naturally, reducing the risk of this being an important factor in the findings.

So what did they find? They followed the parents for between 3 and 14 years after starting IVF, and found that childless women had a rate of death 4 times higher than those with a child, whether it was biological or adopted. Men with children also benefited, being half as likely to die as their childless counterparts.

The team also looked at instances of Psychiatric illness, and found that the rate amongst those with biological children did not differ to that of the childless couples (apart from in the case of substance abuse, which was lower).

This goes against previous, less well controlled studies, which found higher rates amongst childless couples, but this can be explained if having an illness makes people less likely to have children.

Interestingly, parents who adopted children had a lower rate of psychiatric illness that either of the other groups.  This may be because the rigorous selection criteria rules out any prospective adoptive parents who have an underlying, undiagnosed, psychiatric illness.

This study does have its limitations- the authors admit that income, education and age may have confounding roles, however it provides an intriguing glimpse into the benefits of child-rearing.. you may think they are driving you into an early grave, but your kids might actually be helping you to live a long and healthy life!

A new scientific journal is being launched this week in Cambridge. It aims to be distinct and different from existing journals, so we invited the executive Managing Editor, Dr Mark Patterson, to tell us how...

Robert Tjian, Howard Hughes Medical Institute - We do need an alternative to the pre-existing process of publication that every scientists has to undergo and allowing the process to occur in a way that's much more efficient and serves a purpose of getting the most up-to-date high quality scientific information in the hands of the scientific community is really the goal we should go for.

Herbert Jackle, Max Planck Institute - We did a new journal because the journals which are on the market don't have the mechanisms to select the best possible science.

Mark Wolpert, Wellcome Trust - This will be a journal for scientists edited by scientists.  So scientists will be at the heart of the decision making process.  They'll be the best peer reviewers and then scientific editors will make the decision.  So, it will be a journal for peers by peers.

Robert Tjian -   I certainly hope that a defining feature of this journal would be rapid, transparent, scientifically based editorial decisions, one in which we don't at all sacrifice the quality of science, but we make the process much, much more efficient.

Ginny -   The journal is called eLife.  The Managing Editor is Mark Patterson.  So, rapid, efficient, transparent publication uncompromising on quality.  It sounds great what we've just heard on that soundtrack, but how are you going to do this better than a normal journal?

Mark P. -   One of the initial things that we're going to do, you heard a lot of emphasis on quality.  So, one of the things that eLife is going to do is to be a great journal for publishing very influential and important science across all of biology and medicine, and make that work openly available. 

Anyone with an interest in that work can read it and they can do whatever they want with it, which is really important because most of the science that you hear about in programmes like this or read about in the media, maybe 90% of that, you have to pay to read. 

The first step really is to make research openly available.  So that's the first thing that eLife is going to do differently, but you also heard other themes in those talks there, and they were from the people behind the project, the funders behind the project. 

One of the other themes was that the journal should be run by scientists for scientists and so, another thing that's different about eLife is that we have a group of 200 scientists who are committed to a different way of reviewing the work, taking work through peer review, so it's more efficient and more rapid, and more constructive than you see in a conventional journal. 

With just one example of something that we're doing differently is that when the reviews, the review comments, are sent to the authors, what happens is the authors don't actually see the full reports.  Instead, what happens is the editor who's handling that manuscript assimilates and consolidates the comments of the reviewers, after a discussion amongst the reviewers, so that the authors just receive a single set of instructions, and they know exactly what they need to do in order to get the work revised and then published.  And that makes the whole thing happen much more quickly.  So, those were a couple of ways for which eLife will be different.

Ginny -   That sounds great, but if it's so good, why hasn't it been done already?

Mark P. -   Yeah, that's a great question.  There are some pretty powerful forces which keep the system of journals and the way the articles are published in journals, operating in the way that it has done for decades or even centuries, and I think there are probably two main forces at the moment. 

One is the fact that the publishers that publish subscription base journals make an awful lot of money out of it.  So, there's the strong commercial incentive to keep things as they are. 

On the other hand, the scientists who publish their work, they have to publish in journals with an established reputation.  And if most of those journals are subscription based journals, then you've got a kind of a cycle that reinforces itself. 

Now having said that, there are similarly powerful forces of change as well, beginning to operate.  eLife will provide some real momentum towards open access by providing highly prestigious journal home for really great science, and make that science openly available to everybody. 

But there are many other publishers now, new publishers now like the (Public Library of Science) that is showing how this kind of way of publishing can work successfully.  So, we've come so far, we've got to about 10 to 15% of the literature is now available to everyone.  There's a very powerful momentum that will take us further, but there's still quite a lot of work to do.

Ginny -   So, most journals, they make their money through subscriptions.  People pay to read them.  If you're not making money via that way, how are you doing this?  Who's funding it?

Mark P. -   The three voices that you heard at the beginning in that segment there were from three funders.  And so, eLife is supported by three of the most prestigious funders in the world of research, the Wellcome-Trust in the UK, the Howard Hughes Medical Institute in the United States, and the Max Planck Society in Germany. 

In our case, we're funded and our job is to respond to the kind of vision of the funders behind the project, and launch the best possible journal we can. 

Other publishers are showing through other business models, and other approaches, how this kind of approach to open access publishing can operate in a sustainable way through other business models.

Ginny -   And just one final question, how are you going to make this appealing to the scientists, who at the moment want to get published in Nature, or one of the really prestigious journals?  How are you going to make sure that your journal is just as appealing as those?

Mark P. -   Well, it's another very important question.  I think the things that we have going for us are, that we have - as I mentioned, these three very prestigious funders who are behind the project, so I think that lends a huge amount of credibility to the scientific community who are considering submitting to the journal.  We also have a terrific group of 200 scientists who are responsible for running the journal. 

I think those two things will really help us to attract great work and we have already started publishing work now, and we are in fact, receiving some really terrific science. 

I think the other thing that we have to offer that's very important is just the speed of the process.  Especially for people we are early in their careers; they cannot afford to wait around for months and months to get their work published.  And it's not uncommon actually that people can wait - you know, spend more than a year to get a great piece of science published.  And so, you know, that's another reason why I think eLife can offer something very special to scientists.

Mark D. - Yes, so what we now know is that many of the infantile leukaemias actually had their origins in utero. So, when we go and have a look at whenever any baby is born - and you would've seen this with your own kids - you have a heel prick test and this is taken away to be tested for a number of common genetic diseases. When you go back and have a look at these samples [from a leukaemia patient], for instance these chromosomal translocations, these abnormal fusion genes that drive leukaemia, are they present when the baby is actually born? In a number of cases, the answer is yes. So, this abnormal genetic event is not inherited, but for some reason, happens while the baby is in utero.

Mark D. - An interesting question and a philosophical question. I'm not sure we've been made to evolve that quickly. Radiation plays a very important role in treatment and especially in the treatment of some cancers, but I'm not sure that would be a very effective way for us to try and evolve.

Chris - It's reassuring. I'll avoid Chernobyl in the future then!

Stefanie: - There are some human behaviours that are definitely hard-wired in the genetic code. Like the fear of heights, from living in trees and having to hold on, or fear of darkness coming from the fear of being eaten by a predator. I think there are some genes that we know are involved in this, but I think it's still a question scientists are working on today.

Stefanie - Wow, interesting question. One could imagine that this could eventually lead to something like this, but I think we're still in the process of understanding the basic mechanisms behind everything. Mark D. - I'll have to agree with Stefanie. I think it's possible. Our field of epigenetics is still very much in its infancy and understanding how genes are controlled, we're just starting to get a grip on this. So, it may be, in the future, we might be able to bespoke this process a bit better.