Nature Podcast

This is a transcript of the 19th June edition of the weekly Nature Podcast. Audio files for the current show and archive episodes can be accessed from the Nature Podcast index page (, which also contains details on how to subscribe to the Nature Podcast for FREE, and has troubleshooting top-tips. Send us your feedback to


The Nature Podcast, brought to you by An online community devoted to Mini PROTEAN System Users. Explore, Share, Connect and Win at

End Advertisement

Kerri Smith: Coming up this week, a pair of not-so-identical twin stars leave astronomers scratching their heads.

Keivan G. Stassun: This discovery I think will offend a lot of theorists back to the drawing board to ask what exactly this means for our models of how stars are born.

Michael Hopkin: And how the McDonald's logo makes you hungry for a hamburger.

Geoff Schoenbaum: To think about it, much of human behaviour is really driven by this kind of process, the Starbucks logo, the McDonald's Golden Arches, right, they have acquired this ability to drive behaviour.

Kerri Smith: This is the Nature Podcast. I'm Kerri Smith.

Michael Hopkin: And I'm Mike Hopkin. First this week, yet another creature has joined the genome club. Adam is away from the pod this week, but before he left, he found out about the newest member.

Adam Rutherford: Another week, another genome. Just a month on from the sequence of the frankly freakish platypus, Nature publishes another blockbuster genome sequence. We've had mice, rats, dogs, chickens, and a whole zoo of other critters, but what could it be? The lion, the polar bear, the blue footed booby, nope...this one is a tiny worm-like beast that spends most of its life buried in the sea floor, but the humble Amphioxus holds the key to many important questions of evolutionary biology. I've got authors Daniel Rokhsar and Nick Putnam on the line. Daniel for all of us unfamiliar with bottom - dwelling creatures, you better start by telling us what an Amphioxus actually looks like. Nature 453, 1064–1071 (19 June 2008)

Daniel S. Rokhsar: An Amphioxus is a really small animal, looks kind of like a worm at first glance and it is about few centimetres long, looks a little bit like a, you might think like a small sardine except it is really different and it doesn't really have a head and the structure of its body is much simpler than any fish or land dwelling vertebrates that you might know of.

Adam Rutherford: And it has been studied over the years because it actually shares many of the anatomical features with vertebrates. So how does that work?

Daniel S. Rokhsar: Actually, it was around the time of Darwin when people realized that Amphioxus was not just any old other kind of an animal but was an animal that actually showed affinities with the vertebrate group. It has a stiffening rod running down the centre of its body and that allows it to swim with propulsive movements, just a worm without that rod wouldn't have, that rod is called the notochord and underlies the development of our nervous system, our spinal cord. It also has paired muscle segments in the same way that we have paired ribs and paired muscles associated with them in our bodies. And these and other features were the Great Russian embryologist Kowalevsky to realize that these are actually, even though they are quite different from us they could be evolutionarily related.

Adam Rutherford: So, it has got a good evolutionary biology heritage. What prompted you to actually go on and study its genome?

Daniel S. Rokhsar: I was very interested in looking at the genomes of many different animals with different evolutionary divergences from Humans to try to understand what was the nature of the changes that occurred in genomes throughout evolution of the animals, and so this plays a particularly important role because as we said it, it has this special affinity with vertebrates.

Adam Rutherford: Nick if I can just turn to you now, you have sequenced the Amphioxus genome, and what did it reveal. What is the broad picture like?

Nicholas H. Putnam: So, it was known already that Amphioxus had 19 pairs of chromosomes. And this is a draft genome sequencing project, so we weren't able to reconstruct the complete sequences of those chromosomes, but we were able to reconstruct relatively large pieces that typically have more than a 100 genes on each, and by comparing those pieces to the sequence genomes of vertebrates, we could reconstruct changes that happened after the two lineages diverged and in particular it sheds lot of light on changes that happened in the common ancestor of the vertebrates before the vertebrates diverged.

Adam Rutherford: And what are the major changes that you've seen in sequencing this genome?

Nicholas H. Putnam: Well, the most dramatic one is that there were two rounds of whole genome duplication in that vertebrate stem. This is actually an idea that goes back to the 70s and the Amphioxus genome allows us to have a comparison that is close enough to be informative about our genome that diverged before those doubling events and so it allows to resolve those changes in much more detail.

Adam Rutherford: So, Daniel if I can just turn back to you, analysis has revealed that there is some pretty unusual and interesting characteristics in the Amphioxus genome, such as a very high level of genetic variation, could you explain that for us?

Daniel S. Rokhsar: If you look at the genomes of individual people there is a human genome but we all differ from each other. We differ at about one out of every 1000 positions, but if you look at Amphioxus and we just sequenced a single animal so it had a mother and a father, who contributed genetic material to it. If you look in that one animal that we sequenced we found 60 times greater variation. It was instead of one out of every 1000, we got one out of every 15 or 20 bases that has variation, so it is a dramatic difference and it, sort of, shows something that we and others have been finding that marine and vertebrate animals that breed by spawning, sperminate out into the water, they produce enormous numbers of offspring and that is presumably related to their large population size and therefore their large variation. Humans on the other hand, underwent this bottleneck "Coming Out Of Africa" and so even though we have multiple billions of people on the planet, the effective size of the human population is much smaller that explains why humans have much less variation in their genomes than even a single individual of Amphioxus.

Adam Rutherford: And how does the high variation of polymorphisms affect selection?

Daniel S. Rokhsar: Well, one of the things it means is that there is a lot more variation for selection to act on, more individuals in a population means that the population has more variation in it and so it's more likely that someone in that population might be a useful adaptation lurking in there.

Adam Rutherford: So, all of this analysis places the Amphioxus somewhere very early in the phylogenetic tree in evolution, where does it fit?

Daniel S. Rokhsar: So if we consider the place of Amphioxus within a group of animals, as we've said there within the chordates and even though we figure that as our closest invertebrate relatives, it's still a pretty ancient group and if you go further out looking at more and more distantly related animals, that lineage unites with the lineages that includes modern sponges and jelly fish, almost exclusively marine, so one of the big questions for our future work is to try to figure out what was going on in that transition from the early animals which presumably have the complexity you associate with jelly fish or sponges to this organization, you have a body axis and so by looking at Amphioxus and comparing with these other genomes we can start to understand what it is that our lineage has in common with all other animals and then also what does our lineage have in special.

Kerri Smith: Daniel Rokhsar and before him Nick Putnam opening that can of worms. They are both at the US Department of Energy Joint Genome Institute and UC Berkeley. The last common ancestor shared by humans and Amphioxus lived over 550 million years ago. Next Geoff Brumfiel reports on some unusual behaviour among two of the night skies more recent additions. A pair of new born twin stars only one million years old.

Geoff Brumfiel: Stars begin their lives as giant clouds of gas, slowly those clouds condense until they become dense enough that nuclear reactions cause them to light up. Given that model you would expect the two stars born out of the same cloud would look identical, but Keivan Stassun at Vanderbilt University in Nashville and his colleagues have found a pair of new born twin stars that appear to be quite a bit different from each other. I phoned him up to find out more about these twins and what they tell us about star birth. Nature 453, 1079–1082 (19 June 2008)

Keivan G. Stassun: There are some basic things about how stars form that we understand in broad strokes. So for example we understand now that stars like the sun basically form from the gravitational collapse of giant clouds of gas and dust and as those clouds of gas and dust contract under gravity, they become denser and harder and eventually become dense enough and hard enough to ignite thermonuclear reactions, which is what is responsible ultimately for the heat and light that they give off.

Geoff Brumfiel: So, occasionally you can have enough gas and dust that two stars can form right next to each other, I guess this must be how binary stars are made, is that the general picture?

Keivan G. Stassun: That's right. So, by definition a binary star system is the system of two stars orbiting about a common centre of gravity. In fact, it turns out that binary star systems are at least as common as single stars. In fact, binary star systems are probably the rule as opposed to the exception. What makes the system that we found interesting and unique and noteworthy are two things: one is that only a very small fraction, roughly one in a thousand of binary star systems can be seen from us at Earth as eclipsing binary star systems. You have to see the binary star system from just the right angle, so that as the stars orbit one another, they periodically pass directly in front of one and another and block out the light. The other is that it also turns out to be, you have to be lucky to find a binary star system, where the two stars happen to have exactly the same mass and so the system that we found is both an eclipsing binary star system and the two components have exactly the same mass.

Geoff Brumfiel: But aside from having the same mass, I understand that they are not quite as similar as you might expect, given they are the same size and came from the same cloud of gas.

Keivan G. Stassun: That's exactly right and that ultimately is what was surprising to us. Naively, we and I think most of our colleagues in this business would expect that two stars of the same mass in a binary star system would be more or less identical in every respect. The basic reason for that is that mass is destiny for a star. The amount of material that a star is born with basically determines everything about its birth, its life and even ultimately the manner in which it will die and so if you have two stars that are born at the same time, out of the same stuff, and weigh exactly the same, they are for all intents and purposes, we think, the same star.

Geoff Brumfiel: So, what's the difference here? What are you seeing that's different in these two stars?

Keivan G. Stassun: Well, we are seeing some pretty dramatic differences in the some of the physical properties specifically the surface temperatures of the two stars are quite different by about 300 degrees that corresponds to about a 10% difference in their surface temperatures, even more dramatically their intrinsic brightness's or their luminosities are very different. One is roughly twice as bright as the other. We even have a hint that their physical diameters are slightly different by utmost 10% that is something we have to follow up and confirm. But in any case, the surface temperatures and the intrinsic brightness's of these two stars are very different.

Geoff Brumfiel: The fact that they are bright; one is brighter than the other; does that mean that it is burning its fuel faster or...?

Keivan G. Stassun: We know that stars like these will start out, cool and bright but as they contract under gravity because they are becoming smaller will dim down as they become mature full fledged stars and so the differences in temperatures and brightness's that we see are actually exactly what you would expect if one of the stars is older than the other. So it is almost like the twins in this binary star system, one of the siblings represent a slightly older state of the other.

Geoff Brumfiel: How do you think that happened?

Keivan G. Stassun: We believe what we found is that the two siblings in the system which I should add are extremely young, one million years old, which is about as young as stars can be first to observe them in humans terms that corresponds about a one-day-old infant. What we are basically saying is that we have observed a birth order in these twins. Now, of course with human identical twins there is a very good reason why one has to be born a little bit before the other one. It is not at all clear to us why two stars would have a birth order and unfortunately our current theories or current theoretical models of how binary star systems form are largely silent on this question of whether there should be a birth order or not and if there is a birth order how much of a time difference there should be in that birth order. And so this discovery I think will offend a lot of theorists who work on the birth of stars back to the drawing board to ask what exactly this means for our models of how stars are born.

Kerri Smith: Vanderbilt University's Keivan Stassun.


Kerri Smith: Next I'm doing a bit of colouring in. When you're looking at a clump of cells or a piece of tissue, it helps to be able to pick out the bits you're interested in from the picture and many biomedical imaging techniques use colour to help track cells or molecules of interest, but one scanning method magnetic resonance imaging exists only in black and white. Current ways of upping the resolution of MRI images rely on adding in chemicals that affect the brightness of the image. The trouble with injecting colour into this picture is that the kinds of magnetic particles that might be used to mark out different cells or areas only have two variants, magnetic or not magnetic and what researchers would like is a range of states or colours and that is where a new report by a group led by Gary Zabow who works at the National Institute of Standard and Technology at the NIH might help. Nature 453, 1058–1063 (19 June 2008)

Gary Zabow: What we're trying to do basically is create something that is, sort of, like a colour tag for MRI. MRI is typically a black and white technology, so if you can somehow add colour to it then you could hope to get a lot more information out of the images.

Kerri Smith: Nanoparticles are already in use in certain types of biomedical imaging, aren't they?

Gary Zabow: Yeah, it is a big idea in a lot of places.

Kerri Smith: And MRI is a pretty special case than in that it doesn't really have anything like this.

Gary Zabow: Yeah, this idea of having a coloured tag is quite pervasive in biology, I mean, there's a variety of different things as fluorescent proteins or quantum dots where you can engineer whatever colour you want. Having it is very useful, because once you put the silver colour tag or beacon on to something, you can track that thing and if you have a variety of colours you can track a variety of things. It is just easy to see them if you are just looking at the colour rather than the object itself and the issue with those is that in order to see the colour, I mean it is obvious you have to see it, so you need to have optical access to get at it. So it is fine if you're doing some experiments in a Petri dish in a lab bench or something like that but it doesn't really work that well if you try to go inside the body and you would like to see things inside the body too, because you know the conditions there are different to what you've on your lab bench and that's really where the MRI side comes in, because the MRI let's you see inside the body, but on the other hand MRI does not actually have the colour information that you get with regular optical access, so what we are trying to do here is trying to add that colour, try and do really an analogy to the colour that you have in the optical spectrum, we are trying to do it in the radiofrequency MRI spectrum.

Kerri Smith: So people have previously then been using so called contrast agents in MRI, these kind of chemicals that make tissues brighter and easier to examine, which isn't exactly the same thing, but I suppose that is the history of this, isn't it?

Gary Zabow: Yeah, I mean contrast agents are used a lot. It's basically something that, you know, the doctor or physician will use to try and target, say a particular organ or something in your body and the way it works is it basically makes that particular part of the image either brighter or darker than the surrounding, so that the doctor can focus in on that particular piece. That is really where the name contrast agent comes from; it changes the contrast of the image. So, if you think of a sort of a grey scale image, you make something bright or darker and what we are trying to do is sort of a contrast agent, it is a little bit different, because we are trying to come up with something which is, sort of, an analogy to colour instead of changing brightness and darkness.

Kerri Smith: And because in that way it is a qualitatively different approach, you have come at it from a different perspective, haven't you from those developing chemical agents.

Gary Zabow: So, lot of the contrast agents that are out there are chemically synthesized, and so these things are either, you know, some sort of liquid with some magnetic molecules or may be magnetic nanoparticles in it, but they are all chemically synthesized and we are coming at it differently, we are taking it top down micro fabrication approach. The reason that we are doing that is it gives us more control over what we create. It is obviously more complicated than the chemicals synthesis route but we hope that in the end we get a bit more bang of the buck because we can really engineer precise shapes and geometries into the magnetic particles.

Kerri Smith: And how exactly can you get these different colours into the magnetic particles?

Gary Zabow: Traditional magnetic nanoparticles typically all look very similar in an MRI. You can certainly track them and see what they are doing and people have done that but these magnets don't exactly come in different colours, it's either magnetic or it is not, it's sort of like a binary signal and to try and get the different colours we need to have more control over the exact shape and geometry of the micro particle. When we take this micro fabrication approach, we are basically using the same techniques, the same technology that you would more commonly associate with the microchip industry. So the same sort of patterning tool they could use to make the microchip in your computer for example, that sort of what we are doing here to make these micro nanoparticles and it just gives us more control, so by doing that we can change the shapes and set up particular geometries and then each different shape or geometry effectively corresponds to a different colour.

Kerri Smith: What kind of different things will these particles eventually then be able to label, is it different tissues, different tissue density?

Gary Zabow: It's, I mean, I should stress this is really early days right now, we have just done a proof of principle to experiment on this but our hope is that we could take these things and hopefully get them inside of cells, so you might imagine if you had two different types of particles if one is effectively red and the other one blue say, then may be you would be able to attach the red one that goes inside of a normal cell and the blue one into something like a cancerous cell and then be able to watch in the body and really see what happens and that's really just the idea of having different colours that you could attach it to different things.

Kerri Smith: That was Gary Zabow.

Michael Hopkin: Finally this week, do you hanker for hamburger when you see the Golden Arches or feel like a cup of coffee if you spy a Starbucks across the street? Well, as much as you might hate to admit it that's because these and other corporate logos have become so entrenched in our minds that they actually help to dictate our behaviour. Charlotte Stoddart has more.

Charlotte Stoddart: If you're a Big Mac lover, walking past McDonald's while you're on a diet puts you in a quandary. You think of how tasty and satisfying that big burger would be, but you also know that eating one won't help you lose weight. The McDonald's logo is an example of what psychologists call conditioned reinforcers - things we learn to associate with reward. Geoff Schoenbaum and colleagues at the University Of Maryland School Of Medicine have been looking into the thought processes that underlie conditioned reinforcement. They found that the two processes that drive it, the evocation of emotions and thoughts of the reward, use different brain circuits suggesting that drug addiction and eating disorders could be the result of an imbalance of these two. I spoke to Geoff and he began by explaining how conditioned reinforcement can be studied in the lab. Nature advance online publication (18 June 2008)

Geoffrey Schoenbaum: If you teach an animal that a neutral cue predicts a primary reward, something that the animal eats biologically like food or sex or something like that. What you find is the animal then works, press the lever, pull the chain, in order to get that cue, in order to turn that light on or play that tone and they will do this pretty robustly even if the actual reward is never again delivered. So, the cue itself becomes able to reinforce behaviour and this might sound kind of esoteric but if you think about it much of human behaviour is really driven by this kind of process, the Starbucks logo, the McDonald's Golden Arches, right, these things are not biologically reinforcing of their own right but they've acquired this ability to drive behaviour so this is essentially the same process that we are trying to study on a very simple level.

Charlotte Stoddart: And before your new study what did we know about how this conditioned reinforcement controls behaviour?

Geoffrey Schoenbaum: It was thought that the conditioned reinforcers were able to control behaviour primarily only because they acquired some value in themselves, so essentially the idea is that when you do something as simple as pairing a light with food, what we know empirically is that the animals actually come to associate that light not only with the food itself and the sensory properties of the food but also with the general emotion that the food evokes, its happiness or may be it makes the rat remember it is hungry, so the cue becomes able to directly activate that emotion and the idea was that the conditioned reinforcement seemed to be mostly, if not solely, mediated by this kind of general affective property.

Charlotte Stoddart: And do the experiments that you did on rats then support this idea that motivated behaviour is driven by emotions?

Geoffrey Schoenbaum: The experiments that we did in rats directly contradict that idea and so we use, sort of, novel training procedures to teach rats some ways so that the cues would not acquire this general emotional representation. They became associated with the outcome and essentially would evoke parts of the outcome, but didn't themselves acquire any motivational value and when we did this what we found was that parts of the outcome when they were isolated that that information did motivate the animals to do work just as well as emotional information.

Charlotte Stoddart: So, your work is showing then that there two types of thought process that drive this motivated behaviour. One where the driving force is a desire for the outcome, say for the hamburger, and the other where it is a more general emotional feeling towards the thing. Are these two processes acting through the same part of the brain?

Geoffrey Schoenbaum: So we go on to show further is that in fact the ability of the animals to use thoughts of the outcome, that the cue predicts to guard their behaviour does in fact appear to tend critically on one area of the brain which is the orbital frontal cortex and we haven't done sort of a full dissociation but it suggests that there may be in fact different neural circuits that would allow any animal to mobilize this kind of information in response to a conditioned reinforcer.

Charlotte Stoddart: What do your findings then that there are these two different brain circuits tell us about human behaviour and perhaps about human behaviour disorders?

Geoffrey Schoenbaum: Well, so the fact that there are potentially two different brain circuits that mediate these two different kinds of information, would allow you then to use these different kinds of information differentially to control your behaviour and obviously in most situations or in many situations you would assume that these two kinds of information would point at the same behavioural response, right. But the fact that they are segregated in these two different neural systems would allow you to mobilize independently and perhaps in some situations you might get a different answer from these two systems, right. One system might tell you to pursue the response may be the value system which does not know anything about the outcome that is being predicted whereas the more cognitive system, the system that is actually projecting forward and thinking about what might be obtained by this response might be able to be more selective and so in a situation, in human behaviour that might allow you to, you know, not go in to the McDonald's if you are on a diet and you're not interested in the French fries and hamburgers, even though you know you have this other part of the brain that is responding to the Golden Arches and signalling their value and you can think of your own life in some cases that works pretty well and in some cases it doesn't work so well and that might point to a situation in which you have, you know, some imbalance between the operation of the systems and for us a great example of where that might happen is a situation such as addiction. So in addiction, cues drive behaviour quite strongly in addicts often out of our proportion to their expressed desire for the actual drug. So, our study would suggest that when you are looking at why those cues are driving the behaviour you need to think carefully about what brain systems they are tapping into and what kind of information they are evoking.

Kerri Smith: Geoff Schoenbaum talking to Charlotte. That's all from us this week. Do join us next time when we've got massive craters and underwater volcanoes for you. I'm Kerri Smith.

Michael Hopkin: And I'm Mike Hopkin. Thanks for listening.


The Nature Podcast is sponsored by Bio-Rad, at the centre of scientific discovery for over 50 years, on the web at

End Advertisement