[Jingle]
Interviewer: Adam Levy
Welcome to the Nature Podcast. Coming up in the show, we take a look at a new development in the gene editing of human embryos.
Interviewer: Shamini Bundell
We’ll also meet the physicists who are studying matter’s evil twin: antimatter. Plus, we’re having a look at the first flower.
Interviewer: Adam Levy
This is the Nature Podcastfor August the 3rd2017. I’m Adam Levy.
Interviewer: Shamini Bundell
And I’m Shamini Bundell.
[Jingle]
Interviewer: Adam Levy
There’s been some reporting over the last week about the first US based experiments to edit genomes of viable human embryos. Very few details were known other than the fact that it had been done.
Interviewer: Shamini Bundell
Yes, so this is actually a Naturepaper that has now just been published and that I’ve been looking into this week. But it’s interesting to look at what the early coverage has focused on. An MIT technology review reporter wrote, ‘although none of the embryos were allowed to develop for more than a few days, and there was never any intention of implanting them into a womb, the experiments are a milestone on what may prove to be an inevitable journey towards the birth of the first genetically modified humans.’ And that’s why there has been all this excitement and controversy over this paper. One possible outcome of genome-editing human embryos is this idea of designer babies; that parents would be able to select particular features in their offspring or artificially alter their intelligence or attractiveness. Now, this paper is actually a long way off from those kinds of complex alterations. In fact, what it shows is a very simple change. The aim was to fix a potentially fatal mutation that causes a disease called hypertrophic cardiomyopathy. When I spoke to the lead author, Shoukhrat Mitalipov, he was keen to emphasise the importance of this kind of research for treating inherited diseases like this one.
Interviewee: Shoukhrat Mitalipov
So the idea was that there are inheritable gene mutations in humans that cause a variety of disease and we’ve been trying to understand if today’s technology can allow us to actually correct these mutations in early human embryos.
Interviewer: Shamini Bundell
Hypertrophic cardiomyopathy is a heart condition that often goes undetected but can lead to sudden death in otherwise healthy adults. Most people with the condition will have one disease-causing version of the gene, and one healthy version, meaning that they have a 50% chance of passing it on to their children. One way to avoid this is to use IVF and test the embryos before implanting them in the womb: only implanting those embryos without the mutation. Shoukhrat’s team, however, were interested in whether they could fix the mutant gene in the embryos themselves. To do this, they made use of a popular gene-editing tool.
Interviewee: Shoukhrat Mitalipov
So the powerful technology that we have so far is this so-called genetic scissors, CRISPR, which allows you to go into very specific sequences in the genome and make a cut in a really specific space. So with CRISPR, it doesn’t really correct but it makes cuts, so it makes damage.
Interviewer: Shamini Bundell
Although gene-editing tools can be used to knock out certain genes, or even insert new ones, the aim here was to allow the cell to repair the cut gene itself. Almost all cells will try to automatically repair broken DNA, but the way embryos do this is slightly different.
Interviewee: Shoukhrat Mitalipov
Cells, most cells that we know, when they repair, they will make a mistake. But human embryos, they actually look for blueprints that will tell them how to re-programme this broken piece.
Interviewer: Shamini Bundell
The blueprint that the embryos found was already in the cell. The embryos all had one mutant copy on the gene and one normal, healthy copy. So when the mutated version got cut by CRISPR, the healthy copy was used as a template. This was an unexpected finding and could mean that CRISPR could be used by other diseases caused by a mutation in a single version of a gene. The team also addressed a well-known problem in gene-editing called Mosaicism. This is when the edited embryo ends up with a mosaic of different cells, some of which have been fixed and some of which haven’t. Making the edit at the single cell stage, just after the sperm fertilises the egg should prevent this because every adult cell has descended from this first edited cell. But in many experiments, gene-editing at this stage has still led to a mosaic effect. Shoukhrat thinks he knows why.
Interviewee: Shoukhrat Mitalipov
What they found is that even though we are targeting a single cell level, the DNA is already replicated inside. So the cell hasn’t divided yet but the DNA has already divided.
Interviewer: Shamini Bundell
The team showed that by using CRISPR at an early stage, targeting the gene in the sperm before fertilisation, they could prevent this problematic mosaic effect and ensure every cell in the growing embryo has the corrected gene. These findings could help make this kind of treatment safer and more effective for eventual use on patients but some people question whether this kind of research should be done at all.
Interviewee: Shoukhrat Mitalipov
Controversy has unfortunately always surrounded us but I tried to explain that this is for the sake of saving children from horrible diseases, inheritable conditions, which need to be treated. And unfortunately these are genetic diseases that need to be treated by genetic approaches.
Interviewer: Shamini Bundell
That was Shoukhrat Mitalipov of the Oregon Health and Science University in the US. There’s been a lot of buzz about this paper at the Natureoffices and beyond. Our reporter Heidi Ledford has written a news story on the piece and she joins us in the studio now. Heidi, now that journalists have actually been able to read the paper, do you think the coverage will have a different tone?
Interviewee: Heidi Ledford
Well it might. Some of the coverage initially was a bit scatter shot, I guess you could say. There were some interesting claims made about how this was the first time genes had been edited in viable human embryos which is not the case. There’s a group in China that had reported that earlier this year.
Interviewer: Shamini Bundell
So if this has been done before – we have done gene-editing on human embryos before – why is this a particular splash, this story?
Interviewee: Heidi Ledford
Well, what people are telling me is that there have been a number of improvements in the technique made in this paper. So there are a lot of safety concerns about trying to edit human embryos in the clinic. Just about anybody you talk to says it’s nowhere near ready for that yet. And some of those safety concerns though are addressed in this paper, not necessarily fully, but they’ve taken steps to address some of them.
Interviewer: Shamini Bundell
And so one of the key issues that people are focusing on is this is germ-line editing. So the germ-line cells in the embryo are going to turn into the sperm and eggs that make the children. So any edits you do to the germ line are going to get passed on and on down through the generations.
Interviewee: Heidi Ledford
That’s right.
Interviewer: Shamini Bundell
And why is that so much more controversial than…?
Interviewee: Heidi Ledford
Oh gosh, it’s much more frightening isn’t it? If you make a mistake, ah, there go the kids and their kid’s kids and everything. It’s a much more difficult situation ethically to wrestle with. When you talk about editing an embryo and making a change that could be passed down to future generations, you have to be very, very serious that you have to edit that embryo. I mean, with other kinds of gene therapies that are performed in adults or even children, you’re not going to have that changed passed down to the next generation and so different benefit-risk kind of analysis going on. That needs to be reserved for situations where there’s no other possibility – this is a last resort treatment.
Interviewer: Shamini Bundell
And people think it’s a slippery slope. Now that we’re doing this for this terrible heart condition, therefore, designer babies around the corner?
Interviewee: Heidi Ledford
I think there are some technical hurdles to that as well. It’s one thing to try and use gene-editing to treat a condition that’s linked to a single mutation in a single gene. A designer baby: that’s a very complex – I mean, and I’m using designer baby as kind of code for, oh, I want this colour hair, and this colour eyes, and I want them to be this smart and this athletic. We have no idea how to engineer that yet. It’s something to discuss. I’m not trying to put it off as something that we shouldn’t be talking about or worried about. It’s something to discuss, to look ahead to, but it’s really thinking a long way down the line I think.
Interviewer: Shamini Bundell
So, what about the scientific community’s response to this paper? Is it as dramatic as it sounds even to a knowledgeable scientist?
Interviewee: Heidi Ledford
It’s a bit more nuanced I’d say but the people that I’ve spoken to so far are pleased to see it, I guess, and to say okay, so they’ve made these improvements and the off target genetic changes that people were very worried about making – this paper doesn’t see any evidence for those. Doesn’t mean they aren’t there, but at least if they’re there, they’re likely at a lower rate.
Interviewer: Shamini Bundell
So that’s when you make accidental mutations?
Interviewee: Heidi Ledford
Exactly. Yeah. So some of these safety concerns that people have had… this paper takes a step towards addressing some of those. But again the agreement across the board is that this is nowhere near ready for primetime – to engineer even a genetic disease, much less a designer baby.
Interviewer: Shamini Bundell
Thanks to Heidi Ledford there for coming and telling us about that. Her report is up on nature.com/news. I also spoke to Shoukhrat Mitalipov and you can find his paper and a News & Views at nature.com/nature.
Interviewer: Adam Levy
Stay tuned for slug-inspired sealant and evaporating snowball planets. Those are in the Research Highlights.
Interviewer: Shamini Bundell
But now, Adam is finding out about the physicists researching matter’s mirror image.
[Music]
Interviewer: Adam Levy
At the heart of physics are particles: from electrons to the Higgs Boson, these are the building blocks of the universe we know and love. And every particle of matter has an anti-particle: the physics version of an evil twin. If an electron meets its evil twin – a positron – the pair annihilates in a flash [explosion sound effect]and this raises one very important question…
Interviewee: Jeffrey Hangst
So there’s this huge mystery – I like to refer to it as question zero – it’s why did the universe survive and why is it made of matter instead of antimatter.
Interviewer: Adam Levy
This is Jeffrey Hangst who’s at Aarhus University in Denmark. Jeffrey’s question zero is one of the big mysteries of modern physics. Antimatter particles seem like perfect mirror images of their matter twins. For example, an anti-proton has the same mass and equal and opposite charge to a proton. But if they’re so similar, why didn’t the Big Bang create equal numbers of both? Whatever the explanation, we should count ourselves lucky that it didn’t. If there had been equal amounts of matter and antimatter, it all would have annihilated, leaving nothing to make up galaxies, stars and humans. So the search is on to explain this matter, antimatter difference. Physicists are mostly looking for new particles that might help update their best theories. But Jeffrey is doing something different. He’s measuring antimatter itself, as accurately as possible. This is no easy task.
Interviewee: Jeffrey Hangst
Antimatter doesn’t occur naturally which is kind of inconvenient. We need a high energy accelerator to produce the antimatter and once you have it you have to keep it away from all other matter because it will annihilate if it comes into contact with normal matter. When my colleagues talk about doing measurements with matter or how difficult their experiments are, I just kind of roll my eyes. You want difficult, you try antimatter.
Interviewer: Adam Levy
Despite the difficulties, physicists are getting better and better at handling antimatter. Researchers can now combine an antiproton with a positron which is an anti-electron. Together this forms the most simple possible antiatom, anti-hydrogen. Here’s Michael Doser who leads another team working on antimatter.
Interviewee: Michael Doser
A few years ago no atoms had been trapped at all and they’re still trapping a handful of atoms, so that’s a huge step forward but at the same time you’d like to have thousands or even millions of atoms at your disposal.
Interviewer: Adam Levy
The researchers are now probing all sorts of details of these anti-hydrogen atoms. For example, they’re looking into the wavelengths of light that are absorbed by the atoms. This reveals anti-hydrogen’s internal structure and soon they’re hoping to watch how the atoms are drawn towards earth by gravity. All these delicate measurements are done on the off-chance that there’s something, anything, unexpected about antimatter. Some minute difference between anti-hydrogen and hydrogen. A difference like this could finally explain why our universe began with more matter than antimatter. So far, nothing. When I spoke to Michael I asked him… personally, do you atcually expect to find any surprises?
Interviewee: Michael Doser
No. [Laughter]. No, but the alternative is to look where other people have looked and I don’t find that to be particularly interesting.
Interviewer: Adam Levy
One thing is for sure: even with no indication that they’ll ever find deviations in the antimatter, both Michael and Jeffrey do find working with antimatter deeply rewarding.
Interviewee: Michael Doser
The challenge is on the way. The difficulties and solving those difficulties are satisfaction enough for me.
Interviewee: Jeffrey Hangst
Of course the bad news is that if you find the difference you win the Nobel prize if you find no difference, you don’t, and the work is just as hard either way.
Interviewer: Adam Levy
Only the most exotic theories predict there should be any difference between matter and antimatter. So at this point the Nobel is still looking pretty unlikely. And according to Michael, other physicists aren’t quite sure what to make of this antimatter research.
Interviewee: Michael Doser
Partly they think we’re wasting our time. Partly I think they’re amazed by the complexity of what we’re trying to do and I hope they’re impressed because these are things that take many years, to be able to get to some point of measuring something useful.
Interviewer: Adam Levy
The researchers’ ability to create, trap and measure antimatter has come on leaps and bounds in recent years and while no-one can predict what they’ll find, Jeffrey is looking forward to learning more and more about antimatter.
Interviewee: Jeffrey Hangst
So now it’s kind of a new era for us because we can plan ambitious experiments, and so to me it’s like we just reached the crest of a hill and we’re looking down into a valley of opportunity and it’s a fascinating time to be able to work on this.
Interviewer: Adam Levy
That was Jeffrey Hangst and Michael Doser. They lead two antimatter experiments at CERN is Geneva called ALPHA and AEGIS, respectively. To find out more about cutting edge antimatter research, check out the feature in this week’s issue.
Interviewer: Shamini Bundell
Stay tuned for the News. We’re talking exo-moons and P-value pet peeves. But now here’s Charlotte Stoddart with the Research Highlights.
[Jingle]
Interviewer: Charlotte Stoddart
Are you fed up with blister plasters falling off in the swimming pool? Well, look no further than the nearest vegetable patch. Slug slime has inspired an adhesive that stays sticky when wet. The pest that gardeners love to hate secretes a gluey goo that secures them to concrete and cabbage leaves whatever the weather. Scientists investigated how the mucus works and designed a gluey goo of their own. A sticky layer based on cytoson, a common polymer, latches onto the target surface while a backing layer made of hydrogel strengthens the bond by absorbing and dispersing stress. The slime-inspired sealant was put to the test on a pig heart with holes and a leaky rat liver and it successfully stemmed the flow of blood. This ode to ooze was published in Science.
[Jingle]
Interviewer: Charlotte Stoddart
Hundreds of millions of years ago, the earth was entirely locked in ice. Then, thanks to carbon rich gases produced by a volcano, the earth warmed enough to escape its icy bonds and welcome life on board. But other snowball planets might not have such a fruitful future. Planets without earth’s gloriously gaseous conditions are predicted to warm gradually as their stars become brighter. A team modelled this climate evolution on snowball planets with no active carbon cycle and a lack of greenhouse gases. They found that the star power needed to thaw these icy spheres is so tremendous that the planet would overshoot the Goldilocks just right stage and all the water would boil away. Read up on this cool hot topic inNature Geoscience.
Interviewer: Shamini Bundell
I don’t know about you but I love receiving a bunch of reproductive structures on Valentine’s Day. I am of course talking about flowers, the reproductive structures of most plants. The flower evolved between 140 and 250 million years ago, but what did it look like way back then? In a Nature Communicationspaper this week a team of scientists have reconstructed the last common ancestor of all living flowering plants. Reporter Anand Jagatia spoke to lead author Hervé Sauquet from the Paris-Sud University and started by asking him how they managed to reproduce this first flower.
Interviewee: Hervé Sauquet
So unfortunately there’s no direct fossil evidence for these types. There’s a few fossil pollen grains which are probably attributed to the flowering plants but currently there is no fossil flower that is as old as the group itself. So this sets a limit on what we can do with the fossil record for this particular question.
Interviewer: Anand Jagatia
So if there are no fossils from that far back, how can you go about trying to work out what this early ancestral flower might have looked like?
Interviewee: Hervé Sauquet
So, what we did is we used a very general and standard approach which is called ancestral-set-reconstruction and how it works – it’s quite simple. We use the evolutionary tree that connects all of the living species together. We use this tree together with the distribution of particular characteristics of the floral structure as we know it in all of the living species of flowering plants.
Interviewer: Anand Jagatia
Right, okay so you’re kind of chasing the traits of modern plants back through the evolutionary tree. So what kind of traits are we talking about here?
Interviewee: Hervé Sauquet
So we’re actually talking about the number of parts in each of the series of the flowers. We’re talking about the physical arrangement of these parts. We’re talking about the symmetry of the flowers and using the tree and the data, the tips, we can go back in time and make an inference about what happened at different points with different ancestors of living species.
Interviewer: Anand Jagatia
So what did you find when you did this analyses?
Interviewee: Hervé Sauquet
To some extent things we would have predicted, such as the ancestral symmetry of the flower being radial, but also a number of surprises. The flowers were bisexual, containing both female and male parts in the same place rather than in two separate structures. This was completely new. We couldn’t predict it before we ran these experiments and in particular the big surprise was the fact that the ancestral flowers apparently were organised in whirls.
Interviewer: Anand Jagatia
As I understand it the whirls are when the structures of the flower are arranged in concentric circles but the structures can also be arranged as a spiral. Could you give some examples so that people can picture in their heads what they might look like?
Interviewee: Hervé Sauquet
So an example of a flower that has a whorl would be, for instance, a lily flower. An example of a flower that has spirals instead would be the lotus flower. Most of the previously proposed ideas about the ancestral flowers had us assume implicitly that it was spiral and so this was a complete flip in thinking about the ancestral flower. I should specify that we’re not sure about this but all of the data is currently pointing towards this direction.
Interviewer: Anand Jagatia
So, how certain are you that this is what the ancestral flower might have looked like and what would you need in terms of data and experiments to be more sure of this finding?
Interviewee: Hervé Sauquet
So, in this paper, we actually qualified the uncertainty by using a number of methods. What we found is that the uncertainty is, in particular, linked to the choice of the model. So, what I would do to go on and improve the inference in the future would be to conduct new research about the true model of evolution in flowering plants.
Interviewer: Anand Jagatia
If this is what the ancestral flower did look like, what would that tell us about the evolution of modern plants?
Interviewee: Hervé Sauquet
What we found is that if this is correct – it also came with a large number of parts, the parts that we would today refer to as sepals and petals, in the ancestral flower they would be probably more than two whirls, possibly four whorls, of a total of twelve parts. This is much more than most of the living flowers and so what we’ve found is to explain the diversity of living flowers we have nowadays, the very initial phases of floral evolution would have gone through different processes of reduction, of loss of the whorls, or possibly merging of some whirls together to produce simpler flowers initially that later developed into a larger diversity of possibly more complex flowers in some cases and simple flowers in other cases.
Interviewer: Anand Jagatia
What do you think could be some of the possible reasons for that change or that shift?
Interviewee: Hervé Sauquet
So, the way I think of it is we don’t know what happened to produce the first flower but it’s very likely that it was the end point of a long evolutionary history that by chance, and probably with a number of events linked to this, produced this particular structure that we reconstruct today. By all means, with all the time that passed after this ancestor, it had the time to modify itself and it looks as if for flowers it would be better to have fewer parts that are better organised, either in sets of five or sets of three than the original model, which I just see as the imperfect flower.
Interviewer: Shamini Bundell
That was Hervé Sauquet from the Paris-Sud University talking to reporter Anand Jagatia. To read that floral study head to nature.com/Ncomms where there’s also a picture of what the flower might have looked like. And we’ll be tweeting that as well so follow us on @naturepodcast.
Interviewer: Adam Levy
Time now for our weekly News Chat and Davide Castelvecchi has come down to the studio. Hi Davide.
Interviewee: Davide Castelvecchi
Hello Adam.
Interviewer: Adam Levy
Now, scientists have spotted the first moon going around a planet not in our solar system, unless they haven’t. What is going on with this maybe moon?
Interviewee: Davide Castelvecchi
It is a very intriguing hint at this point. There have been hints of so-called exo-moons before, over the years, and it has become one of the biggest, most sought after prizes in the field of exoplanets, but in all the previous cases the hints subsequently disappeared on closer inspection. So, will this one survive? We shall see. In October they will do a follow up observation with the Hubble telescope and maybe some time next year we’ll know.
Interviewer: Adam Levy
So at this point what has actually been observed?
Interviewee: Davide Castelvecchi
It’s one of the most popular techniques for finding exoplanets in general, which is, you stare at a star for a long time and you wait for any dips in its brightness and then if you see any periodic dips then you can deduce that there’s a planet orbiting the star and getting indirectly in the line of sight between the star and us. If you just have a planet and you look at the graph of the luminosity of the star, you see that it’s constant, constant, constant and then it dips. It has a nice symmetric bowl shape, like a cereal bowl, and then it comes back to normal. In this case, what they saw was that it was a cereal bowl that would spill the milk because it was all lopsided and so the idea could be that there’s a moon that peeks out of the planet during this occultation.
Interviewer: Adam Levy
You’ve mentioned that we’re not sure about this; we’re not sure that this is definitely a moon and that’s something that the researchers hope to be sure of in the next few months. Why is there such a big fuss about this right now if it’s still so up in the air?
Interviewee: Davide Castelvecchi
The researchers were not planning to make this announcement because they were afraid of being billed as sensationalists and so on but then to find confirmation of this hint, they booked observation time on the Hubble Space Telescope. The Hubble records are public so another astronomer saw that they had a slot in October and tweeted about it excitedly and said ‘oh, if David Kipping is booking time on the Hubble then there’s only one possible explanation: he’s seen a hint of an exo-moon’. So, to get ahead of the news, the researchers at the last minute added a section to the paper that was originally going to be just about negative results.
Interviewer: Adam Levy
Let’s move onto our second story which is from last week and it’s all about the statistical measure, the P-value. Before we get onto what’s new about the P value, what is a P value?
Interviewee: Davide Castelvecchi
So the P value is one of the most popular measures of the statistical relevance of a result and it says, well what if you wanted to test that there is no effect, for example, that chocolate does not raise your intelligence. You want to make sure that you rule out that so-called null hypothesis with enough certainty and so in some fields of science the standards for ruling out the null hypothesis are extremely strict and in other fields there’s been extremely long controversy over whether they’re strong enough.
Interviewer: Adam Levy
So what would be the problem if a P value isn’t strong enough? What would that actually risk in practical terms?
Interviewee: Davide Castelvecchi
Well, it raises issues of reproducibility because a lot of times results that have a quote-unquote low enough P value oftentimes turn out to be false positives and that could be because of multiple reasons… Either because the statistical sample was not large enough, the experiment was not designed well enough or maybe the experiment was just one of many questions that the researchers addressed – by a statistical fluke they decided to publish the one that had a low P value.
Interviewer: Adam Levy
So what’s the proposed solution to these potential errors, I guess?
Interviewee: Davide Castelvecchi
So the authors of this study, they’re not claiming to have a one size fits all solution or to have a magic wand that will fix the reproducibility crisis but they’re saying in all fields where the standards are low right now, it would help to lower the P value threshold by an order of magnitudes, so instead of 0.05, let’s just start from 0.005.
Interviewer: Adam Levy
Wouldn’t this mean that some results which are true, some findings that are true, could get missed?
Interviewee: Davide Castelvecchi
Yes it could and this is also something they discuss and they say, well this is not the only thing. You have to have a holistic approach to the problem and one other important issue is to raise, to do better experiments and to raise the size of your statistical sample.
Interviewer: Adam Levy
Now, they’re certainly not the only people to complain about the P value. There have been other attempts to try and correct it in one way or another hasn’t there, right?
Interviewee: Davide Castelvecchi
Yes, and it’s an on-going problem. The gold standard is to reproduce findings. There is also the other important issue that has been raised with negative results that are not published because if you do ten experiments and you just publish the one that finds the result you want, say for example, about a drug being effective for one particular symptom, then you are presenting a skewed picture.
Interviewer: Adam Levy
Do you think measures like this will actually get taken up? I feel like every few months there’s a new proposal to try and correct these practices. I don’t hear much about them actually getting implemented.
Interviewee: Davide Castelvecchi
I think the very fact that these issues are being discussed a lot and they’re getting coverage, including in Nature, suggests that people are paying more attention and that maybe things will change.
Interviewer: Adam Levy
Well we’ll see if P values do indeed start to get a little lower soon. Davide, thanks for joining us. As ever, for all the latest science news make sure to head to nature.com/news.
Interviewer: Shamini Bundell
That’s all for this week. But before you go there’s just time to mention that Nature, in partnership with Eppendorf, has announced its winner of the Eppendorf Award for Young European Investigators. The award recognises outstanding work in biomedical science and you can listen to an interview with this year’s winner, Tom Baden over at go.nature.com/eyia2017. I’m Shamini Bundell.
Interviewer: Adam Levy
And I’m Adam Levy
[Jingle]
