New bird flu vaccine could tackle multiple variants with one shot

0:23Welcome back to the Nature Podcast. This week, designing a flu vaccine that gives broad immunity. And reviewing 25 years of carbon offsets. I'm Sharmini Bundel. And I'm Benjamin Thompson.

0:46First up on the show this week, we've got a story about new research that's looking to develop a vaccine that offers broad protection against different variants of avian influenza. Now, bird populations around the world are riddled with influenza virus infections. The viruses that cause these infections are broken down into subcategories based on the shape of the molecules found on their surface. Of particular concern are a subtype called haemagglutinin 5 or H5 viruses. Some viruses of this subtype have evolved to become highly pathogenic, devastating flocks of farmed and wild birds around the world.

1:29What's more, these viruses have spilled over into other animals like mammals, often with similar outcomes. There have also been reports of humans contracting H5 influenza from animals, and several hundred deaths have been reported. but to date there has been no evidence of direct human-to-human transmission. If that does come to pass, the results could be incredibly serious. Previous flu pandemics have caused untold numbers of deaths, and so plans are in place to get ahead of H5 influenza. But this is tricky. There are many variants of the virus, with new ones popping up all the time, meaning that multiple candidate vaccines must be prepared and stored, just in case.

2:18But what if there's another way? This week, researchers demonstrate a proof-of-concept approach to produce a single vaccine that could ultimately offer protection to all H5 variants. To find out more about the work, I called up one of the team, Mathilde Richard, from the Erasmus University Medical Centre in the Netherlands, who first gave me an overview of how variant-specific vaccines are developed currently. So two times a year, the WHO is having a meeting where, you know, data are looked at to basically characterise these viruses that are circulating in different places of the world and decides, oh, there's a new variant popping up, very different from everything else we've seen.

3:01We need to make a new candidate vaccine for it that could be distributed quickly to manufacturers in case we would need to scale up production, right? So since the emergence of these viruses, there's been 48 different WHO candidate vaccines that I think represent quite well current and past diversity. But you can imagine that it's a reactive approach. And that's a problem with a very fast evolving field like virology and viruses are always ahead of us, right? So by the time that you select a variant and you make a vaccine seed and all of that, how relevant it is for actually what's currently circulating.

3:37It's hard to really be 100 % sure. And this is where you and your colleagues come in then. What were you looking to do in this work? What we were trying to do is first to understand how these viruses diversified and what is the current diversity. So we wanted to recapitulate that in one big comprehensive way. And by better understanding how these different variants came about, the idea was that you could use this information to try and help develop a vaccine that was more generalised to multiple variants? Yeah, exactly. Against as many as we could, basically. So with one vaccine, we wanted to be able to protect against most variants.

4:16That was the goal. And to do this, then in your paper, you've made this 3D map of the different variants kind of on a grid, and each of them is a sphere, like a planet, I suppose. What is this map showing? So this map is basically showing how similar two viruses are at the antigenic level. So antigenic means really the characteristic of an antigen that it will have in relation to the immune response that it will create in people. So what it shows is that we can basically recapitulate how these viruses changed, you know, over time, antigenically. And basically, the closer they are in space, the more similar they are.

4:56That means that if you were to be vaccinated with a virus, you'll probably be protected against something that's very close to that virus in space, but not against something that's very far away in space, right? So what does this allow you to do then in terms of developing a vaccine? It allows us to understand basically what we're looking at and the space that we're looking at so that we can basically design a vaccine that would elicit an immune response that would be targeted against as much as possible of these different variants. And the second part of your paper, Mathilde, is trying to do that then.

5:29So you've designed some different vaccines that look to produce antibodies that cover off this map. How did you go about doing that? So what we did first is that we looked at variants that were sort of naturally located in the centre of the map, and we tried to understand their characteristics. And we sort of like copied basically nature, We got inspired by nature, right? And so we mutated this hemagglutinin protein, right, which is the surface glycoprotein of the virus. And we mutated it in a way that would sort of resemble whatever was naturally located in the center. We also changed it so that it would actually look a little bit more like a human virus because we know that the next pandemic is going to be caused by a virus that will look more like a virus that's adapted to humans.

6:13And avian and human viruses are very different. So we modified it such a way that it would look a little bit more like what could be the next pandemic virus. Through these changes, we sort of engineered a variant that was already more central in space. But most importantly, when administered it to animals, was also eliciting an immune response that was also centered in space and reactive. And you made this vaccine and tested it against some very different flu variants. What did you see there? So we use ferrets. Those are the animal that we use for flu research. and we've seen basically protection.

6:48So once we vaccinated the ferrets with our vaccine antigen and we infected afterwards the ferrets with two different H5 viruses, very different and basically opposite sides of the map, we saw that the ferrets were protected as well as if you were comparing vaccines that were matched to those two viruses, right? So this is what we call the standard of care. If there is a new virus emerging, the best vaccine will always be the vaccine that is matched to the virus that's emerging. So our central vaccine was non-inferior to the two homologous matched vaccines against two viruses that are very different in the map.

7:27So what you're saying is that your vaccine, which is non-specific to any particular strain, was doing, in this small study it has to be said, as well as vaccines for a specific strain. How did you feel when you saw the results? So some of this, especially one of this challenge, was very severe. So steroids that were what we call mock vaccinated. Just to see the extent of the disease and the severity of the disease, we actually, by day four, two ferrets were found dead because it's a very, very severe infection. This virus not only cause pneumonia in the lungs, but in this case of that specific virus also disseminated to other sites.

8:03So for instance, the liver and the spleen. And so we were very pleased to see that with a vaccine, as you said, that's not matched. We actually were able to protect against such severe disease as well as a major vaccine. So we were very happy to see that it was working very well. As you say, this is very encouraging for your research, but of course, it's important not to run away with this. I think you say in your paper, this is very much a proof of concept. This is testing a small number of variants in one group of flu viruses. This isn't the pan flu vaccine that people are hoping for. No, certainly not.

8:34This is not a universal flu vaccine, right? A universal flu vaccine would be something that would protect against all type of influenza. So going to all these different age subtypes, this is not the case, right? This is why we call it subtype wide, basically, and broad immunity, but within the same subtype, right? And so it is a vaccine that will be, I think, if it works as well in people, instrumental for pandemic preparedness. So what you could do is early stage of a pandemic, vaccinate people because you sort of vaccinate agnostic to whatever is coming because you are central in space either pre-pandemic so vaccinating in the idea of perhaps having a pandemic or at a very early phase before a vaccine that's matched to the actual virus that's emerging is made and made available because the problem is that there's always some you know month even can go up to six months before you can actually have a vaccine that's available by that time there's many people have been infected and your paper's out now what questions does this work leave to answer, do you think?

9:37So the modifications that we did, you know, I told you that we engineered that specific vaccine in a certain way, we still need to really understand the fundamental of why it's working so well. So this is for sure a question that is left for us. The other question that we continue having is whether we could take this approach for other flu subtypes. And that's something that we certainly want to explore. And it leaves, I think, question of whether this vaccine is really going to be useful for people. And on that then, finally, what do you think about the chances of it being applicable in humans, do you think?

10:13This is something that needs to be tested. And this is something that is currently evaluated in a phase one trial. I think that the responses that we see in ferrets can be widely, at least from past studies, extrapolated to people. And so we see a very good correlation between the type of responses we see in animals and hearing ferrets compared to what we can see in people. But we don't know until we test. That was Mathilde Richard from the Erasmus University Medical Centre in the Netherlands. To read her paper, look out for a link in the show notes. Coming up in the show, we've got the latest from the Nature Briefing, but right now it is time for the research highlights with Dan Fox.

10:57Physicists have worked out how to generate beams of muons without using a huge particle accelerator. Muons are subatomic particles, similar to electrons but much heavier. They form in the upper atmosphere when cosmic rays collide with particles and can be used to sneak a peek into hard-to-reach places. Muography, that's imaging with muons, has been used to find a hidden chamber in the Great Pyramid of Giza and even to peer inside volcanic magma chambers. However, making these measurements using cosmic ray-derived muons can take many months. Muons can also be made in a particle accelerator, but that is impractical for fieldwork.

11:40Now though, a team of researchers have developed a new method to make muons. First, they used a laser plasma accelerator to generate beams of high-energy electrons. These electrons are then collided with a metal target to produce muon pairs, which can be detected even through 90cm of concrete. The team say their system will outperform cosmic ray muons for imaging by reducing exposure time by orders of magnitude. You don't need a muon source to find that research, though. it's in physical review accelerators and beams.

12:18The discovery of bird DNA in bat guano suggests that a handful of bat species are hunting migrating birds. To understand how bats could pull off such attacks, researchers attached altitude and audio sensors to 14 noctual bats in southern Spain during bird migration season. The team recorded a total of 611 insect hunts involving short, fast attacks. However, two other hunts lasted longer, started with a vertical sprint and involved more than 25 times more echolocation than did the average insect hunt. One successful attack ended with a recording of a bird distress call, later identified as belonging to a European and robin, and chewing sounds on the recording confirmed that the successful hunter ate its prey on the wing.

13:13The authors think that bird wings found on the ground in areas where the bats hunt suggest that the bats bite them off to immobilise their prey. Echolocate that research in science.

13:34and finally here we are it's briefing chat time so we're going to discuss a couple of articles that have been highlighted in the nature briefing which is nature's daily roundup of science news so ben you first this week what have you got for me yeah i've got a story that i read about in nature and it's based on a nature paper in fact and it's about metagraph metagraph metagraph the shorthand way to describe Metagraph is that it's essentially Google search, but for DNA. I love that. I love the sound of that, Googling DNA. What kind of things are we talking about here? Is anyone listening who has spent a lot of time working in the biological or biomedical sciences, and I put my hand up here as well, dear listener, you do spend a lot of time searching and analyzing things like DNA, RNA, protein sequences, a lot of searching of the internet.

14:21Now, Now, these data are housed in these online databases, right? Public repositories. And these can contain billions of letters of amino acid sequence or DNA sequence or what have you. And it can be really, really hard to search these things, right? And although Metagraph has been described as Google for DNA, it's actually a bit more like a YouTube search, right? So YouTube, if you search red balloon, it'll search inside the video for a red balloon rather than stuff that's just been tagged as it. And that's what this does. It can uncover patterns hidden inside big sequencing data, even if they're not tagged.

14:56So when you're talking about you want to search for DNA, are you talking about I would like to search for the DNA of this particular species? Or are you talking about searching sequences? Yeah. So let's say you've got a particular sequence and you want to compare it to other stuff. That can actually be quite hard. And the motivation behind Metagraph was to address an accessibility problem, right? Because data repositories have grown in size at an enormous rate. And this really presents challenges. If you want to compare sequences in disparate databases that are just growing and growing and growing, that can be really hard.

15:29Because a lot of these are what are called raw sequencing reads. They've just been uploaded from the sequencer. And maybe they haven't been assembled into genomes fully. Maybe they haven't been tagged properly. So they are, in some cases, just long sequences of DNA or RNA or proteins or whatever. And these, as I say, can be quite fragmented. They can be very, very numerous to search directly. So this is where Metagraph comes in. And what the folk behind this new system have done is they've integrated data from seven publicly funded databases. And this contains millions of DNA sequence sets, billions of amino acid sequence sets.

16:05And this is from across the tree of life. We've got microbes, plants, humans, and they've developed this search engine. And you can put in the sequence that you're interested in. And it will search these integrated archives of raw data and throw out where it can be found. So normally, I might have to go through different databases, like know where my thing that I'm looking for might be, know what kind of thing it might be to go and look. Is it in this genome? Is it in that genome? Whereas with Metagraph, theoretically, I could just type out a sort of string of a sequence and it will tell me places where that is.

16:39And it has been used. So to demonstrate the utility of Metagraph, the authors behind the work used it to scan almost 250 ,000 human gut microbiome samples for genetic indicators of antibiotic resistance around the world. And they say that this analysis took about an hour on a high-powered computer. So it could well be very, very useful for people who spend a lot of time searching in databases. And as I say, there are a lot of people doing that right now, I'm sure while they're listening to the nature podcast yeah but it's not the only one of these it has to be said the article describes some other nascent search tools which can look at stretches of DNA and what have you and it seems like this could open up a lot of new discoveries they describe it as pentascale genomics here right so who knows what those findings will be but folk are quite excited about this and say the most impactful applications are yet to come Well, thanks, Ben.

17:32So my story today I want to tell you about. So this was a Guardian article that was flagged in the briefing about a systematic review of the literature from the annual review of environment and resources all about carbon offsets. This whole system of carbon offsets, carbon credits, which has been around for a while and, you know, criticised and trying to give an overall impression of whether it's working. Well, before we get into the answer, then maybe just give us a brief 101 on the carbon credit system. Yes, this has been going on for a while now. And the basic idea is that if I am doing a lot of polluting, I'm a big corporation.

18:11I'm pumping out lots of carbon dioxide into the atmosphere. That's not great. I want to do something about that. Instead of actually trying to cut my emissions, what I could do is I could say, Ben, it would be much cheaper for you to, let's say, cut your emissions or maybe go and plant some trees, maybe something abroad where it's going to be cheaper to do that. If you do something that takes out the same amount of carbon dioxide from the atmosphere that I would have otherwise cut, it kind of evens out. So there's this sort of buying system where you sell me a credit and then I can say, aha, you know, I've cancelled out my emissions or even my activities are now carbon neutral because I've completely offset it with these carbon credits.

18:55OK, and that sounds sort of sensible, I suppose. But you're saying that there has been criticism of this system. Yes. So there have been criticisms of lots of sort of problems that come up in practice with this for a long time. and what this review is doing is basically saying these are deep-seated systemic problems that incremental change is not going to solve and it's not just due to a few bad apples. So I'll give you some examples. This is what the researchers said were the worst problems. So one problem might be, hey Ben, build a wind farm and I'm going to get the credits for that when actually you were going to do it anyway.

19:30Yeah, I was thinking about it. You know, that was the plan anyway. I'm not really increasing any benefit to the environment there. It might be the fact that your project that I'm paying for isn't permanent. An example given here is maybe I'll pay you to plant a bunch of trees, but they later burn down in a wildfire. So I'm not having the benefit anymore. It could be things that aren't very thorough or comprehensive. So let's say you could protect a little part of the forest and be like, great, this is now protected. But the logging is still happening just in another part of the forest. or even projects that are double counted, where there's a buyer and a seller and they both claim for the same project.

20:11So it's like double the carbon credits, essentially. So far, roundly negative. Are there any upsides to this endeavour? There are things that do work. So I will say that, again, in this review, they found that there are a few areas where there might be some success or there might be sort of easily fixable problems. So certainly the potential to improve it. So one of the authors here said, you know, we don't want to throw the baby out with the bathwater. But overall, I must say the sort of stats and the overview is pretty negative. So there was a previous meta-analysis from Nature Communications last year that investigated some carbon credits, and only 16 % of them showed real reductions in greenhouse gas emissions out of everything that's tried.

20:56And there was also a quote from this co-author of this study, which said, we have assessed 25 years of evidence and almost everything up until this point has failed. Crikey. OK, which leads to the question then, if this isn't fit for purpose as is, what's to be done? So there was a few recommendations in here. And as I said, there are things that they found that worked or could be fixed. and there are things that don't. They recommend urgently phasing out any offsets that are not actively sucking out carbon dioxide from the atmosphere. I think they think that high-quality carbon dioxide removal and storage is a really important way to do it.

21:35And, I mean, you might have seen ads of companies saying, we've gone carbon neutral, we're green, it's all great, we've essentially got no emissions, basically not allowing people to claim that when, if you look into it, it is not in fact true. well a sobering story there Sharmini let's leave it there for this week's briefing chat listeners if you'd like more on either of those stories or you want to sign up for the nature briefing to get more like them delivered directly to your inbox head over to the show notes for some links and that is all for this week as always you can reach out to us on X or Blue Sky our handle is at nature podcast or just send us an email podcast at nature dot com I'm Sharmini Bandel and I'm Benjamin Thompson see you next time

22:20Thank you.