eLife Episode 55: Weaponised insulin

The shellfish that hunt using insulin, and brain pathways linked to addiction...
29 March 2019
Presented by Chris Smith
Production by Chris Smith.


Conus geographus hunting goby fish


The shellfish that release insulin into the water to catch fish, brain activity patterns that predict future addictions, how to do gene drive experiments safely, and is the first author position gender neutral? Chris Smith talks to leading scientists publishing groundbreaking papers in eLife...

In this episode

Conus geographus hunting goby fish

00:33 - Cone snails hunt with fast-acting insulin

A fast-acting insulin from cone snails could help diabetics

Cone snails hunt with fast-acting insulin
Helena Safavi, University of Utah

Some good news for diabetes, because researchers have discovered that a family of venomous shellfish, called cone snails, use a very fast-acting insulin to immobilise their fish prey by causing the target animal’s blood sugar to plummet so it can’t swim off. A fast-acting insulin like this would be extremely useful for human diabetic patients, and now, by studying several different cone snail species, Helena Safavi and her colleagues have figured out how this insulin works, why it’s so rapid in action, and how to copy it, as she explains to Chris Smith...

Helena - Cone snails are all predatory marine snails so they prey on either worms, snails or even fish, and they’re native to the tropics so they live in beautiful areas. They have wonderful shells; there are about 800 different species and every single species has a different shell pattern and they're all very beautiful.

But we are most interested in the compounds these snails make to prey on other animals, and we've long known they produce compounds that can be used in pain relief and pain research. But what we have now found is that some of these species actually make insulin, and they release that insulin into the water that the fish is swimming in and that insulin causes blood sugar to quickly drop, and the fish that is exposed to the insulin is not able to swim away anymore.

Chris - And the cone snail snaps it up and eats it?

Helena - And once the fish is unable to swim away, the cone snail comes and can just swallow it up.

Chris - That's extraordinary to think that this thing is squirting insulin into the water. So to questions spring to mind 1) why doesn't the cone snail end up with very low blood sugar as well or doesn't insulin work in a cone snail, and 2) how does the insulin get out of the water and into the fish?

Helena - It turns out that the insulin the cone snail makes is very different to its own insulin. So the snail makes its own insulin to regulate sugar levels in its own body but the insulin that it sprays into the water is extremely similar to the insulin produced by a fish so it wouldn't be active at its own target receptor. In terms of how it gets into the fish, we think it rapidly enters the body through the gills.

Chris - The thing is though, if you did this with the kinds of insulin that we have in the clinic to give to to humans, they’re quite slow acting aren't they? Whereas a venom has to work really fast in order to immobilise a prey really fast because these things are shellfish, they wouldn't be able to pursue a fast-moving fish, so this stuff must be quick? How does it do it?

Helena - The snail has to make sure that the fish is very rapidly immobilised and the insulin acts very rapidly compared to the insulin that we make and it does that by being a single compound. Our human insulin is very sticky so an individual insulin would stick very rapidly to another insulin and to another insulin and form so-called hexamers. And the snail insulin, because it has to act very rapidly, never forms a hexamer so it can act much faster than our own insulins do.

Chris - So when our insulins go into the body do they have to unstick before they can work then, whereas what the snails are doing, is theirs never stick in the first place so they're immediately available for action?

Helena - Yes, that's exactly right. When we inject, or a diabetic patient injects insulin into the body the hexamer has to first associate into a compound that can then be active, whereas the snail never made the hexamer in the first place so it's immediately active.

Chris - Now if you know this, the obvious question to ask is well why don't we just make with our biotechnology know-how a form of human insulin which can't stick together like that?

Helena - That's a very interesting question. We have actually tried to do this for over 20 years now and we have not solved this problem because at the moment you try to make the human insulin not stick, it's not active anymore so you strip it of its activity so you can still inject it, but it won't do anything in your body anymore. Somehow the snails have solved this long-standing problem by making this insulin in its venom.

Chris -  So put me out of my misery; what has the snail been able to do that human industry over the last couple of decades couldn't? It must have discovered some kind of clever trick that we hadn't thought of?

Helena - Yes. So we think that the snail insulin binds to the human insulin receptor or the fish insulin receptor in a different fashion. It uses a slightly different surface on the receptor so the area that it uses to bind to the receptor is a little different to our human insulin, and this is how the snail has solved this.

Chris - But critically, what that means is that if you can copy what the snail does you could potentially make a human insulin that's very fast acting and not sticky in that way so when it   went into the human it would very quickly gain control of their blood sugar?

Helena - Yes, and that's exactly what we are currently trying to do, and we have made very good progress on this that we're planning to hopefully publish soon in the future. So what we’ve done is to try to learn as much as we can from the snail insulins, the different ones that we have found, and then go back to human insulin and make it non-sticky and yet active. And we have the first compound that we are hoping to put into the clinic sometime in the future.


A smoker enjoying a pint of beer

Brain activity patterns predict addiction
Edmund Rolls, Oxford Centre for Computational Neuroscience

Is there such a thing as an addictive personality? And if there is, what is it about the brain that produces that mindset? As he explains to Chris Smith, at Oxford Edmund Rolls has been using an MRI scanner to try to find out...

Edmund - We're very interested in the brain mechanisms of certain sorts of behaviour, including addictive behaviour. In this particular paper smoking and drinking. And what we have discovered is one part of the brain - the medial orbital frontal cortex - has different connectivity in people who drink; and the lateral orbital frontal cortex has different connectivity in those who smoke.

Chris - Who did you look at and how?

Edmund - We studied very large samples - of eight hundred thirty one participants, in one case from a US project called the Human Connectome Project, and we followed that up with eleven hundred and seventy six participants from London. What we did was to measure the connectivity between different brain regions while people were resting in a brain scanner. We used functional magnetic resonance imaging. What we found is that just when we measured the connectivity between different brain regions - when they weren't doing anything: they weren't smoking or drinking at the time - we found that we could predict who, later on in life, would smoke and who would drink.

Chris - Now what fraction of the people you studied were smokers and drinkers and what fraction were not?

Edmund - The youngest participants were 14 in this study. At that stage very few of them had started smoking or drinking. They were brain scanned again at the age of 19, and at 19 some of them had started to drink. And we correlated their brain connectivity with their smoking and drinking. The lateral orbital frontal cortex had low functional connectivity in smokers, and we think smokers were giving themselves nicotine in order to increase the connectivity between brain regions. And we think that that is the basis for that type of semi-addictive behaviour: that it makes one more alert because perhaps one's connectivity of the lateral orbital frontal cortex but also in fact overall of all brain areas is slightly low in smokers.

Chris - This is like functionally self medicating isn't it to pep up the areas which are underperforming?

Edmund - That's one of the ideas that resulted from this particular study. One could consider that that's what happens in smoking. The other interesting thing that happens in this investigation is that, associated with the smoking, and the undergone activity of the lateral orbital frontal cortex, people were more impulsive. So we think that one thing that may facilitate smoking is impulsiveness. And it's important to have discovered this, because that has implications for helping people.

Chris - Now if someone is making up for - for want of a better phrase - a paucity of functional connectivity between two brain regions and the nicotine is doing that, when a person who has been an entrenched smoker gives up, what are the consequences for their cognition?

Edmund - We don't know. My advice to someone would be that if you feel that you need a bit more stimulation then don't smoke but take a small amount of nicotine in the form of a patch and that might restore your alertness that perhaps previously you sought by smoking.

Chris - Because one would infer from what you're saying that if there is this reduction in connectivity and that the the smokers are doing this in order to improve the communication between these disparate brain areas, that if you don't have the nicotine signal there then there would be a consequence for cognition either either an IQ decrement or an attentional decrement. That would suggest then that if that's manifest at 14 because you're saying that when you looked at the youngest participants who hadn't yet become smokers and they showed similar patterns, do you see underperformance or a tendency towards attentional problems in those young people?

Edmund - That's not clear yet. The overall point would be that it's possible that those who choose to smoke may have slightly less sort of activation of their brains, and so they may be likely to benefit from stimulation of their brains. But we haven't taken it further than that. What I could say is that we made a second discovery: the same individuals were reporting how much drinking they performed. Those individuals had slight over-connectivity of the whole of their brain, and in particular for the medial orbital frontal cortex. Now that's extremely interesting because the medial orbital frontal cortex is a part of the brain involved in rewards. The account we came to, therefore, is that it's possible that those who drink have a particularly sensitive reward system and it's that that attracts them to alcohol. Now the interesting point here, in relation to addiction in general, is that we now have two types of addictive behaviour which have quite different underlying brain mechanisms. 

Chris - And what do you think, putting all this together, the take home message is? How should cognitive neuroscientists think differently about people who use these agents, and how might doctors and clinicians dealing with people who do, approach their subjects differently because of what you've found?

Edmund - I think this has very important implications for understanding addiction in general. There's been an emphasis in the field of neuroscience research on addiction to think primarily of a chemical called dopamine which may underlie a lot of addictive behaviour. But if so, that would be a unitary phenomenon. But here we have two other types of addictive behaviour - smoking and drinking - which can't be accounted for by a simple unitary explanation. There are different things that are happening in different parts of the brain. That then leads us towards potential treatments for different types of addictive behaviour. And the sort of treatment that then could be helpful for those who have smoked is that it may be helpful to suggest a nicotine patch until their brain perhaps normalises its sensitivity a little bit after they haven't smoked for some time. And similarly, for those who may have an oversensitive medial orbital frontal cortex to rewards, which attracts them to alcohol, then at least some cognitive understanding of what's happening in their brain would help individuals to understand their own behaviour better.

Shown here in drosophila, a “gene drive” is a gene editing technique that biases the inheritance of a genetic element or trait so that it rapidly increases in frequency in a population.

13:03 - Gene drives tested safely

Are laboratory-based gene drives a reasonable representation of what would happen in the wild?

Gene drives tested safely
Jackson Champer, Cornell University

A “gene drive” is a gene editing technique that biases the inheritance of a genetic element or trait so that it rapidly increases in frequency in a population. In a fast-breeding species, this means that, within just a small number of generations, almost the entire population will carry the gene drive. This could be used to protect an endangered species by conferring resistance to a pest, or even to eradicate or neutralise a problem or invasive organism, and scientists are testing these concepts in insects - and even in mammals - at the moment. But with this power comes enormous risk: what might happen, for instance, if an experimental gene drive escaped the confines of the laboratory and ran amok in the wild? For this reason, gene drive technologies have been engineered with in-built Achilles Heels. But are these constructs really “field relevant”? Speaking with Chris Smith, Jackson Champer has built and tested some to find out…

Jackson - In this study we developed ways to study gene drive in the lab without risk of them spreading in the wild if they were accidentally released.

Chris - We need to back up a little bit. First up, what is a gene drive?

Jackson - A gene drive is basically a piece of DNA that's part of an organism's genome that contains a nuclease, which is something that can cleave a very specific portion of DNA. It cleaves a bit of DNA on a chromosome that does not have the gene drive, and then copies itself onto that chromosome. So instead of the organism passing down a gene drive to only half of its offspring, it passes it down to all of its offspring thus allowing the gene drive to spread through an entire population. You can almost describe it as a cut and then a copy paste to fix the cut.

Chris - And what sorts of genetic elements are people talking about pasting in and driving through a population like this?

Jackson - Well there's two basic types of gene drive: a "population modification" gene drive, where we basically try to spread a genetic payload through the entire population. This could be something like a special gene that prevents mosquitoes from transmitting malaria. The other type of gene drive is a "population suppression" gene drive. This type of gene drive is designed to completely eliminate a particular species, either from a small region or globally. And this of course has been best studied in malaria mosquitoes. So there could be two ways to use gene drive to get rid of malaria: either make the mosquitoes unable to transmit the malaria, or get rid of the mosquitoes in the first place.

Chris - The worry is, of course, that, if one does this, once you set these hares running you might not be able to rein them back in?

Jackson - Right. There are some strategies that could potentially let you rein-in a gene drive, though it's still an open question on how effective these might be. What we were most concerned with in our manuscript is if someone were to develop a gene drive and then, by accident, it was released into the wild, it could then either spread it to all the organisms, or suppress them, which would be an undesired result if this happened by accident. So we came up with a way for scientists to study these gene drives in insects without worrying about any negative consequences, if some of these insects were accidentally released from the laboratory

Chris - How does that work? 

Jackson - Well we used two different strategies for this. One was using a special target site. These gene drives need to cleave DNA at a very precise spot in order for the gene drive to work. And we simply made it so that that spot was an artificially-inserted piece of DNA. That means that our gene drive can only work in these lab insects that have this synthetic target site. If it met a wild-type insect, it simply wouldn't be able to cleave and thus would not be able to spread in the population. The other method we used was simply to not use a complete gene drive. Instead, we used a split-drive strategy in which one of the important components of the gene drive was provided separately. This separate component would not copy itself, thus preventing the gene drive from having an exponential spread in wild type organisms. This split drive could only spread if we provided it with this supporting element in the lab.

Chris - Of course one criticism of this is that this is still, nevertheless, a laboratory and these sorts of artificial constructs do not exist in nature. So while that's great because it enables you to study this - and the behavior of gene drive technologies - without the risk of an environmental escape, one must ask how relevant is this?

Jackson - In this study we took many performance measurements of both of these different types of gene drives and we found that they worked the same basic way that a full standard homing type gene drive would work according to some of our previous studies. Because of this, we believe that the things we learn about gene drive in the lab using these safe drives - the synthetics, target sites and split drives - will be fully applicable to full gene drives that might later be released in the wild.

Chris - This obviously applies to insects. What about when we try and take this further afield, if we want to, say, go to a remote island that's now being plagued by rats and other rodents and eradicate those; is what we're seeing in drosophila relevant to other species?

Jackson - Yes it is. It seems as though the same basic mechanisms of gene drives and the ways to improve them will likely be the same across species. That said, just because you have a gene drive in the fruit fly doesn't mean you're going to have a gene drive in a mouse right away. There's still a lot of important experiments you would need to do to make a gene drive in a new species. But, at the same time, you can still apply a lot of the lessons that we can learn from the fruit fly traditionally in biomedical literature.

A man and a woman in discussion.

19:48 - What's in a name?

Who goes first on shared authorship papers, men or women...

What's in a name?
Nichole Broderick, University of Connecticut

What’s in a name? Well if it’s the person whose name is first on a paper, it’s pretty important, because that tells the scientific community who, ostensibly, did most of the work. So what happens when two people share the first authorship? Who goes first? And does it make a difference if they’re male, or female? Speaking with Chris Smith, Nichole Broderick…

Nichole - Traditionally in biomedical literature, the first author is the person who really led the study who did the work and then the last author's name is usually the person who got the money for the study or runs the lab or institute. But we end up calling papers by that first name and that kind of sticks is something that part becomes part of our conversation and how we talk about the work.

Chris - I suppose that matters, because when a person goes for a job, if they have got a string of papers with their name at the front of them and they've been at a conference presenting that work and their name is the one that everyone's aware of, their prospects can only be improved by that can't they?

Nichole - It gives them that ownership and that place of being the person at the helm of that work. And so it really does indicate they've been the one driving it which can be misguided if there's another person who also was working with them, but is the second person in the name of that co-authored list.

Chris - So how did you actually approach this and what did you do? Where did you get the data from to do the study?

Nichole - So we honestly used Google Scholar primarily, and went and just searched for co-author or equal contribution and searched through different journals and we were able to find those papers that way. 

Chris - Are all of the papers where that has happened., do they state clearly that there has been a co-authorship and that authorship was shared equally - all the work was shared equally between the first second and maybe even sometimes first second and third names on that list?

Nichole - We found as much as eleven!

Chris - Really. Eleven first authors!

Nichole - Yes! It's usually two, sometimes three, but there were some extremes. You will find somewhere, but what's interesting is that in some journals as they've moved from more of their paper versions now onto an online format, you won't necessarily see it unless you actually download the paper. So that was somewhat tricky but we were able to find them.

Chris - And how many papers did you extract data for?

Nichole - So we looked at around 3000 papers over a dozen different journals all in the biomedical area.

Chris - And over what time span? 

Nichole - So we did, starting in about 1995, 97 going through 2017.

Chris - Okay. And so you crunched all these numbers; you know which authors have contributed. Do you know who's a male who's a female author?

Nichole - Some of the Anglo names were a bit easier for us to sort of decipher: a Bridget versus a Jonathan. But obviously science is very international. So we ended up resorting to searching for images of people, finding them on LinkedIn pages, or research gate so that we could definitively indicate whether or not they were male or female. So we had to exclude some papers where we couldn't determine that.

Chris - Indeed, because there were some "Ashleys" and some "Leslies" that may well be indeterminate!

Nichole - Well those we checked. We always checked.

Chris - And what you're asking is: "So we've got multiple author paper and we've got either a man and a man, a man and a woman, or a woman and a woman, as these authors upfront". What sort of numbers did you see in terms of how did those break down. And when you actually then compared who was first and who was second, did you see numbers that were equivalent to chance, or was there bias?

Nichole - What we found first of all was the male-male combination was by far the most common. And then if we looked at mixed genders, there did appear to be a bias against women in the first position. So it was more common, if it was a mixed gender paper, for there to be a male in the first position and a female in the second position.

Chris - Any idea why you might be seeing that? Because when people decide the author list and they've contributed equally, sometimes they'll reach some kind of decision about alphabetical order, or they'll toss a coin. Was there any explanation offered as to why this might be the case?

Nichole - The one thing that we found was interesting was we only found about one or two papers that actually said "this order was decided by alphabetical order". In most cases that's never indicated. If you talk to scientists, anecdotally, I think they would tell you that sometimes that flipping of a coin and stuff happens less frequently than you might expect. But we do note that over time that has gone down. We did look for what was occurring by time and it is becoming more equal.

Chris - Did you phone up any of these people, where you saw this going on, to ask them or how did you reach that decision?

Nichole - We hesitated to actually do a study so that we could survey for those sorts of questions; it might be an interesting follow-up that we've talked about. At the time we really wanted to get the story out so that people could talk about it and think about it.

Chris - And if nothing else maybe lobby journals to say look this needs to be made clearer in the papers. If you found such a paucity of examples where this is actually explained, it kind of argues that we really ought to be putting that on the paper shouldn't we?

Nichole - Exactly and that's one of the points that we make and actually we've been getting feedback from that aspect where people have told us that their journals are you know their editor in chiefs of journals that it's something that they're journals moving towards is having a very clear statement about how these decisions are made. 

Chris - You must be quite pleased though by the fact that as you look at the scientific timeline of publication that this effect appears to weaken?

Nichole - Absolutely. It was very reassuring to see that there was a correction. Who knows exactly why but it's wonderful to see and it's great to think that we're approaching equilibrium with.

Chris - With that sort of ordering it's reassuring to me, as a as a biomedical scientist, that we're in that situation. But what about these physicists? Are you going to go and probe them next to make sure that they're not being naughty?

Nichole - One of the dilemmas there is just the number of women physicists that are in training it's still an area and still one of the sciences where we really need to increase the numbers. So I think that'll be one of the tricky aspects of it but obviously would be great to look at.


Add a comment