Science Friday - Strain Of Bird Flu Discovered In California | Understanding Bipolar Disorder Through The Genome

Episode Date: January 31, 2025

The outbreak of H5N9 avian influenza occurred at a California duck farm in November 2024. Also, new research pinpoints 298 parts of the genome associated with higher risk of bipolar disorder. This could lead to better treatments.Another Strain Of Bird Flu Discovered In CaliforniaThis week, the World Organization for Animal Health reported that it had been notified by the USDA that a November outbreak of highly pathogenic avian influenza on a California duck farm was caused by a strain not before seen in the United States: H5N9. The dominant bird flu strain circulating currently, H5N1, has led to massive culling of bird flocks, has infected dairy cattle, and has killed almost 500 people around the world since 2003. The US reported its first human death from bird flu earlier this month.Experts stressed that the new strain did not itself appear to be an immediate human threat. But the rise of a new strain is troubling and points to the risk of a viral phenomenon known as “reassortment,” in which different viruses mingled in a host can sometimes exchange bits of viral code, forming new strains.SciFri’s Charles Bergquist joins Flora Lichtman to talk about H5N9 and other stories from the week in science, including a spacewalk that was meant to include a search for microbes on the outside of the International Space Station, a possible positive side effect of scratching an itch, and the discovery of 66 million-year-old fossilized vomit.Understanding Bipolar Disorder Through The GenomeBipolar disorder is one of the most common mental illnesses—it affects an estimated 40 million people worldwide, about 2.8% of the population. Bipolar disorder can cause extreme mood swings, and be debilitating without treatment.In an effort to untangle the mysteries of where bipolar disorder originates, researchers studied the genomes of more than 40,000 people with the condition. When comparing these genomes to those of people without bipolar disorder, the researchers were able to pinpoint 298 different parts of the genome associated with the mental illness. With this better understanding of the genome, better, more targeted treatments for bipolar disorder may be possible.Joining Flora to talk about this research is Dr. Niamh Mullins, assistant professor of psychiatric genomics at the Icahn School of Medicine at Mount Sinai in New York.Transcripts for each segment will be available after the show airs on sciencefriday.com. Subscribe to this podcast. Plus, to stay updated on all things science, sign up for Science Friday's newsletters.

This is Science Friday. I'm Flora Lichten. This week, according to the USDA and the World Organization for Animal Health, a different strain of bird flu has turned up in the U.S. H5N9. It caused a November outbreak of a highly pathogenic avian influenza on a California duck farm. The dominant bird flu strain currently circulating H5N1 has led to massive culling of bird flocks worldwide and is now affecting dairy cattle. The U.S. reported its first human death from H5N1 bird flu earlier this month. Joining me now to talk about this and other stories from the week in science is Cy Fry's Charles Berkw. Hi, Charles. Hey, Flora. So a new kind of bird flu, huh? Yeah, I mean, this gets a little bit weedy.

But when we talk about different kinds of flu, there's like the super big level. First, what type of influenza, A, B, C, D? A and B are the kinds of flu you think about when we talk about people getting affected. with seasonal flu. C is a kind that can affect people, but it's usually a mild. D is mainly a cattle thing. Okay. All of the viruses we're talking about here with the bird flu are flu A.

But then you get into these H and N designations. That talks about what specific form of two proteins called hemaglutin and neuraminidase are on the surface of the virus. Then there can be even like finer distinctions called clades, subclades. But what we're talking about here is influenza A, H5N9, as opposed to influenza A, H5, N1, which is what's been mainly affecting bird flocks. So we're looking at a difference in the neuraminidase protein. Do we know where the strain came from?

So this could have been brought in from somewhere else or another animal host, but viruses have this weird ability called reassortment. If you have a host that gets infected with a couple of different viruses at the same time, sometimes the viruses can actually swap bits of code. I think of it like if you have enough kids at the birthday party, sooner or later somebody's going to be wearing somebody else's socks. I've lived it, Charles. Exactly, right? Whose socks are these?

This new one was found in a duck farm. And ducks are thought to be pretty resistant to the bird flu. It turns out they can be infected without getting very sick and they just sort of carry the flu around. So a duck farm like this where this outbreak occurred can be a good place for that kind of mixing and matching reassortment process to have. happen and create new viruses. Should we be concerned? So officials say that this specific H5N9 virus doesn't look like it poses an immediate threat to humans. But having more kinds of viruses spreading around in animal reservoirs just sort of ups the chance of it making the jump to another species.

Also ups the chance of it mutating into something more threatening. So, you know, this is definitely something to keep an eye on. Let's move on to another microorganism story. There was a spacewalk on Thursday that involved microbes. Tell me about it. Right. Thursday, there was this multi-hour spacewalk from the International Space Station. This was Sunni Williams and Butch Wilmore. They're the two astronauts that have been having unexpectedly extended stay on the ISS. They were the ones who went up on that Boeing Starliner craft that had had thruster problems. So they've been up on the ISS for a while.

It's nice they finally were able to get outside for a bit of EVA and look around. Fresh air. Fresh air in the spacesuits. They spent over five hours outside on Thursday, and mostly they were working to remove a failed piece of communications equipment that needed some repair. Apparently, this was a bit stuck in place. Wilmore said he had to jiggle, jiggle, jiggle to get it to come out. But while they were out there, they also took swabs of the outside of the U.S. portions of the ISS looking for bacteria. Why? So, you know, stuff is clean when it's set up into orbit.

Steralization's a big deal in the spaceflight world, but humans aren't clean. And the ISS has been in orbit for a long time. They were mainly swabbing around exterior vents from the life support system. And there are other chances for microbes to move around, say like an astronaut touches a spacesuit or tool inside the ISS and then goes outside and uses that spacesuit or tool. there's a chance for bacteria to make the transfer. And why did they care, whether there's a microbe on the outside of the ISS or not? Yeah, so the Russian crew has done similar swab studies before. This is the first one on the U.S. side.

And the question here is whether or not any bacteria that made it outside could survive on the outside. Because, you know, if there are bacteria that could survive on the outside of the station, you're in vacuum, you're exposed to solar radiation. first, that has implications for finding life elsewhere. You know, what can a bacteria really take and still be viable? It also would raise questions about what we would have to do to protect other worlds from being contaminated with our own life. Right. If microbes can survive the vacuum of space, we've got to be even cleaner than we already are. Exactly. Just wiping it with the bleach wipe isn't good enough.

While we're on the subject of human bodies, there's new reasons. research into itching and scratching? Yeah. You know, when you have an itch, it can feel really, really satisfying to scratch it, right? Sure. But at the same time, that scratching is not good for your skin. If it gets out hand, you scratch too much. It can lead to more damage and inflammation, which makes you uncomfortable, which makes you want to scratch more. And dermatologists call this the itch scratch cycle. But that thing of scratching and it's feeling good has kind of been a bit of a mystery. If it's just a damaging thing, you'd think that we would have evolved not to enjoy scratching. Right. Right. It's a paradox. Yeah. But this week in the journal science, researchers report that

apparently, at least in mice, scratching can also boost immune defenses against bacterial infections right at that injury site. So there's actually some benefit to scratching. Okay. Okay. It sounds like we scratched the itch on that mystery. Yeah. What's it good for? Why are people researching this? So they say that this new immune response pathway might be something that they could target with a drug. So that may be a benefit there. There are also people with a chronic itch condition. They always are itchy.

And learning more about why we scratch could potentially lead to hopefully some kind of relief for them. Let's move on to something that makes me feel itchy sometimes. Artificial intelligence. It has been a busy week. Yes. It seems like every other minute some complies. is announcing a new revision to their AI model. It's more accurate. It's faster. It's more realistic. Late last week, a Chinese company made big news with this model called DeepSeek R1.

And then later this week, the Chinese mega company, Alibaba, announced its own updated AI model that it said is even better than that. What was up with Deepseek? Yeah. So the big thing here is that all AI systems take a lot of computing power, like a lot a lot of computing power. And the Deepseek and Alibaba products both apparently have come up with ways to be much more efficient, training and giving results with a lot less computing power, and so less cost. And this was to the point that when Deepseek demonstrated this R1 model, the stocks of the computer company, Nvidia, and a few other companies had significant drops

because if these models hold up, maybe the high-powered computer processors these companies make are going to be less in demand. Okay, besides the stock hit, why else do we care? You know, why do we care that there's another AI company pushing a new technique or model? Yeah, so there's all different layers of drama here. So some people are worried that this represents a shift in AI leadership away from the U.S., which, you know, true. It might be. But then there are other complaints similar to the ones that you see centered around the media app TikTok.

You know, if you use this model, you're sharing all your data with China, potentially its government, who knows what happens then? And to add on to all of that, OpenAI, those are the chat GPT folks are complaining that Deepseek may have copied some of its work or trained its model on chat GPT output, which is somewhat ironic given the concerns over these models training on data around the web and books and art without necessarily getting permission from the creators in the first place. It's a pot kettle situation. Exactly.

This all sounds very messy. Yeah, messy is definitely a good word for it. We should also note that a lot of these models from private commercial companies, you know, they're kind of black boxes. We don't see the code. We don't see what it's doing in there. So we really only have the company's claims on some of these efficiency measures. So, yeah, we're going to really have to see how this all shakes.

out. And on top of that, there's probably going to be something new next week and the week after that. It's sloppy out there. I think that leads right to our last story. Ancient vomit is in the news this week. I would like to hear every single detail. Yes. So, you know, this is a story about a fossil hunter discovering what is apparently fossilized 66 million year old fish vomits. This is from a sample in Denmark, and the fossil hunter found what looked like a bunch of mixed up bits of fossilized sea lilies embedded in a piece of chalk. These sea lilies apparently have a bunch of calcium rich parts that are not particularly digestible. So something, probably a fish, chowed down on a bunch of lilies, and then yacked up the bits it couldn't digest. And now, fast forward 66 million

years and it's going to be in a museum. Sea lilies were found in there. What is that? Yeah, so don't think of like lilies like frog sitting on a lily pad. Sea lilies are actually cryinoids. They're these delicate, almost feathery-looking things that live at the bottom of the sea. And the thought is that in the Cretaceous, these were a tasty meal for some fish, but just couldn't stomach some of the harder parts of the lilies. This sounds goofy. Like, why do we can't? But fossils like this can give a better idea of sort of the food chain in some of these early situations where tracing out who ate what is kind of an important thing to know. And probably hard to figure out without ancient vomit. Right. We have so many gaps that any clues like this are helpful.

What does it look like? Well, I mean, first it's embedded in chalk. So this is all sort of a beigeish white stuff. and there are cylindrical chunks kind of jumbled up next to each other. Chunks. That makes sense. Exactly. That adds up.

Yeah. And I know that you like fun words. So just as fossilized dung has a name, coprolite, fossilized vomit apparently has the delightful name of regurgitalite. Regurgitalite? Regurgitalite. And now you know. I am so glad to know that. Charles Burkwest Science Friday's senior producer.

We have to take a quick break, but when we come back, what a new genetic study is telling us about the biology of bipolar disorder. It's pointing us towards an involvement of the gut brain axis, and we do know some of these cell types are capable of producing serotonin. Don't go away. A new study on bipolar disorder, a condition that's characterized by extreme mood swings, including sometimes manic episodes followed by bouts. of depression. Despite the fact that bipolar disorder affects 40 million people worldwide, it's often misdiagnosed and its underlying biology isn't fully understood. A new study is filling in some gaps. Researchers looked at the DNA of almost 3 million people and found new regions of the genome associated with the condition. Here to tell us more about what these

findings might mean for treatment and our understanding of bipolar disorder is my guest. Dr. Neve Mullins, assistant professor of psychiatric genomics at the Icon School of Medicine at Mount Sinai in New York. Welcome to Science Friday, Neve. Thank you. Glad to be here. Okay, so tell me a little bit about what you found in this study. In this study, we found 298 genomic regions linked to bipolar disorder. So each of these regions is a window into the biology of the disorder. Each one can include many genes. And we did a variety of analyses to map the genetic variance in these regions to particular genes. And this allowed us to identify 36 genes that had robust evidence for their involvement in the disorder. 36 genes. So is that a lot? Was that what you were expecting? We know that bipolar disorder is very polygenic. So there's no one risk

gene, but there's many risk variants throughout the genome that each increase risk by a tiny amount. So we expect that there are many genes involved in risk for the disorder. Okay, so you found 298 regions associated with 36 genes. Were many of these new? Had we known about them before? Of these 298 regions, 267 of them have been linked to bipolar disorder for the first time in this study. So it's a fourfold increase in the number of regions that were known previously, and that's thanks to being able to conduct this larger study, including many more participants. A fourfold increase. Okay, yeah, I mean, I want to understand this. Is that because we've never sampled this many people before or the sampling methods have gotten better? How do you explain that jump? It's really driven by the number of

participants in the study. So every genetic study is a mixture of signal and noise. And we need enough participants in the study. So that signal outweighs the noise. And we're able to detect the regions of the genome involved. And since each of these genetic variants confers a very small effect on risk, we needed hundreds of thousands of samples to be able to detect these regions. Were any of the regions surprising or were they associated with other things that add up for what we know about bipolar disorder? Many of the genes that we identified have roles in synaptic signaling. So communication between different cells in the brain. And we did also see in this study some evidence of the involvement of certain cell types

outside of the brain. So we saw two cell types highlighted from the intestine and the pancreas. And so this was a novel finding for the first time, pointing us towards some cell types outside of the brain. And we do know some of these cell types are capable of producing serotonin, which affects circulating serotonin and brain serotonin levels. So it's pointing us towards an involvement of the gut brain axis, but we do need further research to better understand the potential mechanism underlying this. Wow, so the pancreas and the intestines might be involved in bipolar disorder? Yes, this is what the current data suggests, a link there between the gut brain axis, which was a surprising finding from this study. It's not something that has.

had come up previously in our previous genetic studies of the disorder. So there's definitely something novel there to explore in future studies. Are all this genetic regions sort of equally predictive for the illness? Some of them increased risk for the disorder more than others. So one of the ways these genetic studies could have an impact is in building genetic risk predictors for bipolar disorder or polygenic risk scores. So these are scores that summarize an individual's genetic risk for bipolar disorder based on the results of this study. And so the genetic regions can increase risk by different amounts, but really we want to look at the risk across the entire genome, and we sum that together to create a polygenic risk score, which is a single number, which summarizes,

a person's genetic risk for the disorder. Hmm. And are there multiple presentations of bipolar disorder? And do they correlate somehow with these different regions? If you have changes in different regions, do you have a different presentation of the condition? We know that bipolar disorder is heterogeneous clinically. So there are different presentations. For example, there are subtypes.

Type 1 bipolar disorder is characterized by manic episodes and depressive effects. episodes, whereas type 2 bipolar disorder is characterized by hypomanic episodes and depressive episodes. So we also see some differences on the genetic level. We've been able to study type 1 and type 2 bipolar disorder separately. And what we see is that they're highly genetically similar, but they're not genetically identical. So some differences on the genetic level could partially explain the differences in clinical presentations for people who experience bipolar. disorder. So let's talk about the applications for this research, and you touched on them a little bit, but do you see this research underpinning a new genetic test for diagnosing bipolar disorder?

We won't be able to use a genetic test to diagnose bipolar disorder because it's not completely genetic. So when we think about genetic tests, we need to think about them as tests for a genetic risk or genetic susceptibility. Currently, the genetic risk predictors that we're able to construct based on genetic studies, they're not yet powerful enough to be used in clinical practice, but we do know that the performance of these genetic risk predictors will continue to improve as we conduct larger and more powerful genetic studies. And so there are a number of possibilities in the long term for being able to use tests for genetic risk in clinical practice. And there are some studies ongoing incorporating genetic risk predictors into clinical care, particularly for physical disorders,

such as breast cancer or heart disease, to investigate how useful these genetic risk predictors could be in terms of stratifying patients who would benefit from earlier screening or earlier intervention to reduce their risk. So I expect that we'll see more studies in psychiatry in the next few years to investigate how useful polygenic risk scores could be in informing clinical care in terms of assisting with the diagnosis, early intervention, perhaps even treatment selection, alongside the current clinical tools. Do these findings point to new treatments? Yes, we've seen based on the 36 genes that we identified in this study that many of the proteins encoded by these genes can be targeted by drugs. Some are actually the targets of existing medications that are

currently in use for treating other disorders. So there could be opportunities for repurposing some medications for treating bipolar disorder and there could be some novel drug targets amongst these genes. Of course, we need further research into each of these drug targets, of course, clinical trials before any new medications could be used in patient care. But the results so far are promising. And we know that drugs that have genetic support for their target or their mechanism of action have a higher chance of success in the drug development pipeline. So that's encouraging. If we zoom out for a minute, how well understood is bipolar disorder?

Until recent years, our knowledge about the genetics of bipolar disorder has lagged behind other psychiatric disorders like schizophrenia or autism, which received more attention and more research funding. But in the last few years, that's really started to change. Our studies on bipolar disorder have rapidly grown to now including hundreds of thousands of participants, which has enabled us to identify these 298 genomic regions involved. So it's slightly surpassed similar genetic studies on schizophrenia, and our knowledge base on the genetics and biology has really started to improve in the last couple of years.

Why has it lagged behind? It's received less attention, less research funding than some other psychiatric disorders. The studies on bipolar disorder were smaller. than those of other psychiatric disorders. And coming back to that heterogeneity that you mentioned in clinical presentation, again, we need really large sample sizes in order for the signal to outweigh the noise in the genetic studies and be able to detect these regions that are involved in the disorder. So it's really a combination of increasing sample size through international collaboration.

In this study, we meta-analyzed the results from 79 different studies from all around the world. So it's bringing together all of that data through international collaboration, having more funding to be able to do this work, has really driven the knowledge forward. Does stigma around mental illness and around this disorder prevent us from doing these studies or has it contributed to the lag in our understanding? understanding of them. Yes, there's certainly stigma still surrounding psychiatric disorders, but I think that, you know, understanding the genetics and biology of these disorders could actually contribute to reducing the stigma surrounding them. So having a better understanding of how these conditions arise and participants in these studies often find it very meaningful to be involved in the research, even if it doesn't impact them directly right now, it may have.

have impact in the future for other patients. And so contributing to these genetic studies is a way that I think we can reduce the stigma surrounding bipolar disorder and other psychiatric disorders. That's about all the time we have for today. Thank you so much for joining us. Thank you for having me. Dr. Neve Mullins, Assistant Professor of Psychiatric Genomics at the Icon School of Medicine at Mount Sinai in New York. And that is about all we have. have time for. Lots of folks helped make the show happen, including Rasha Aurety, Dee Pieder, Sandsy, Sandy Roberts, Shoshana Bugsbaum. I'm Flora Lickman. Thanks for listening.