The Coronavirus is Mutating

five percent of the cases that we're seeing. But what does that tiny, little single point difference mean one amino acid? Is it something we should be concerned about? That is something that scientists are in the process of figuring out to day. We're joined by one of those scientists, Nevill. Sanjana is a geneticist at the New York Genome Center and New York University. He has been researching and publishing about how the coronavirus is mutating. Nevill, thank you so

much for joining me. Let's start with the call it the medium picture. Let's talk about the changes that we know have happened through mutation in the stars. CoV two virus since the original version surfaced in Wuhan over the course of the spring. Sure, so I think there's several different changes that have been found in different patient populations. Like many RNA viruses, I think one thing that is important to know about stars CoV two is it doesn't

say the same it does mutate. There's elements of how the virus replicates where errors can be made during RNA transcription, and those errors can get packaged into new viruses and then propagated. Most of the time, most of those errors changes or variants or mutations probably have very little effect. But occasionally you have changes that really do have some

functional impact on the virus. So one of the mutations, a single point mutation that you've been working on and that's gotten a fair amount of coverage, not as much, perhaps as it ought, is a mutation at the six hundred and fourteenth place in the genome of the starscoby two virus. Tell us about that one and tell us how it's moved through the population, right, Yeah, So the mutation that's in the spike protein. So Spike is really

amongst the coronavirus proteins, probably the most famous proteins. So Corona just means crown. And the reason that coronavirus has this name with crown in it is because the individual viral particles, the varions are decorated with the spike protein that kind of sticks out and gives it this crown like appearance. So, just out of curiosity, because I've been wondering about this, It's called the spike protein because it literally spikes up and a number of spikes make the crown.

It looks yeah, it looks like a spike. Yeah, it's by far the most distinguishable feature of any pictures we've seen at the virus. So the spike protein, the RNA port encodes a protein and that protein has thousands of amino acids in the six fourteenth position. Is this mutation that we've started to focus on, and the reason that we've started to think about this is totally accidental. My

lab works on crisper and gene editing. We are very interested in using those tools to understand what are the host genes, what are the human genes that are essential for viral entry? And by figuring out which are the essential genes for viral entry, we were hoping that we could find ways to maybe block some of those genes or suppress the activity of some of those genes, and by doing that protect people from the virus entering. And

this was just one of these accidents and science. You often hear about science that when you tell the story of science that it's like super linear, but when you're actually doing science, it actually is not so linear. There's

twists and turns and different paths, sometimes dead ends. And so with the what we were trying to do in the lab to understand what are the host proteins that are required from viral entry, we started using a very safe virus that we use in the lab, and all we did was we attached to it the spike protein on the outside of that virus from stars Kobe two, because we know that the spike protein, because it spikes out, is the first point of contact between human cells and

the coronavirus. And what we found actually was something pretty sad, which was that we were really barely able to get any cells infected with this virus that we had put

the spike protein on. And so at that time in April, we had started to hear about this mutation that was circulating the population and looked like it was increasing it had come about sometime maybe in early February, likely in Europe, based on the best viral population genomics work with sequences viruses, and it looked very rapidly after it's kind of emerged in early February that it started to take over kind

of the world populations. And that's true up to today where it's greater than ninety five percent of the circulating coronavirus today seems to be carrying the spike mutation, so new in February, but now here in July, it really seems to be quite dominant. So we thought, okay, let's modify the spike protein that we have in the lab. Let's just insert this little mutation that seems to be dominating. This is back in April. We did this and let's give it a try in the lab. And what we

found was indeed it solved our problem. Our problem is we couldn't get the virus are kind of safe virus, the pseudovirus that we have decorated with the spike protein, we couldn't get it to enter our human cells. But when we changed it to have this mutation, we were able to see it enter much much better, five to ten times better, and we thought, great, we've solved our

technical problem we had here in the lab. And then a day or two later we started to think about this and we said, wait a minute, this is pretty important. Maybe instead of just running on from this technical problem, we should maybe we should report this. Maybe we should tell people like, look, this is a functional change in how the virus infects humanselves. There are many fascinating things in the story you just told, and I want to

just break them down a bit at a time. So let's start with the kind of astonishing fact that in January, when the first sequencing of the stars COVID two genome began, almost none and maybe none of the viruses that were sequenced had this mutation at place six fourteen. By March the number was noticeable. By April, it was sufficiently noticeable that your lab was thinking, let's try it out. By May it was in seventy percent of reported cases, and

now it's at ninety five percent. So the first question I want to ask is does this count as strong evidence to you, as among other things such geneticist, that there must be some adaptive feature of this change, or is there some way that this could have happened without this particular mutation helping the virus to replicate more successfully. Yeah, so I think scientists are naturally very careful. So when we got this first functional data in late April early May,

we weren't sure whether to believe it. And one of the great things about science during the COVID era is that there's been a lot of sharing and very rapid sharing new results. And in May what we saw were lots of different groups steadying the viral genomics and the evolution of viral sampling through the world, and there was

really quite a dichotomy of views. There were folks who thought this is clear evidence of selection going on for the spike mutation, and then there was completely opposing views, which is less so today, but was more then where people thought, maybe it's something about a founder effect. You know, if hasn't seen coronavirus before, but just the first introduction of coronavirus into that country happens to be the one that carries this variant, and then maybe the next introduction

into that country happens a week later. Well, you know, exponential growth, So if you have a week of lead time, that can really result in a lot more infections, especially because we know asymptomatic people can infect others, you know, I think sitting where we are right now, especially with all the functional data that's come out, there's about five or six different groups that have shown functional data is similar to what we did here in the lab, I

think it's pretty clear that this version of the virus is more transmissible. And it's not just laboratory experiments, but something that we do in our preprint is we look at data from Sheffield University in the UK and also the University of Washington in Seattle. So these are two totally different groups sampling different populations UK population and US population, and they're basically looking at data from the qPCR test, which is the test that involves the nasal swab and

qPCR just stands for quantitative polymerate chain reaction. It's actually quite a standard lab technique and because it's got that q and it quantitative, it can actually detect how many viral copies they are in a particular nasal swab. And what's super consistent over the two sites is the difference between the people who have the D variant or the G variant. These are the two different spike variants. The D is the original one, the G is the new one.

And what both the sites find is that there's about a threefold increase in viral RNA detected in the nasal swabs of people with this new G variant, and that's

consistent between Sheffield and the group in Seattle. And so to me that that suggests, I mean, we can't say something about person to person transmission, but something that we can definitely say with that data is that perhaps within the body, and remember the body is kind of a collection of cells, independent cells, where you know, you can have cells that are infected in cells that are not

that are just fine. So perhaps within the body there's greater infection, greater distribution of the virus amongst cells when they are carrying the SPIKE variant. That sounds like it's a very powerful reason to think that the spike variant is superior with respect from the virus perspective, inferior from our perspective, and that it is adaptive. Let's talk about the question of why it seems to be better at

effectuating transmission. What is it about the crown that works better from the virus's perspective at latching on to your cells in the G variant compared to the D variant with which the virus began. That's a great question. Yet, so if it is more infectious, like we show in a few different cell types, why is it what's you know, how does this one amino acid change create such a big difference in infectivity. That's a great question. We have

the exact same question. And so when you look at the structure of the protein, there's different functional domains across the length of the protein. Something that we noticed is that kind of the closest functional domain to where this mutation is the receptor binding domain. That's the thing that actually makes contact with the human receptor for coronavirus, which we think is in a receptor called ACE two. So we said, okay, it's it's not in the receptor binding domain,

but it's very close to. It's the closest kind of domain of the protein that has this well defined function. So it must be that, right, It must be. You know, something that this mutation is doing. It's increasing its affinity for ACE two. It's just like you know, if you're able to have kind of a tighter handshake, then that's a going to increase infection versus somebody that just kind of waves from a distance. Right, So we set about actually with some collaborators here at NYU to really to

test this. And you know, as is often the case in science, you know, you have some very strong hypothesis about here's what the data should look like, and you know, reality and tells you, hey, you're wrong, And that's exactly what happened here. You know, our hypausis going into this was it was likely stronger binding to the receptor ACE two. We found that there was basically no difference between the purified spike or the spike variant with ACE two binding.

So this was definitely not the right answer. It's fascinating to hear about a hypothesis that doesn't pan out, and it helps the rest of the world to trust science, to realize that it's not that the scientists start with the hypothesis then claim to prove it. Some things work, some things don't. That's part of the process. So what did you do next when you realize that the ACE two wasn't the answer? You know, there are other aspects of what spike you know, which, after all, is just

this little micromachine that helps the virus. It is not there just to have the handshake with the ACE two receptor, it performs a lot of other functions that are very important for the virus to actually enter an inject it's genetic material into the human cell. The handshake is really just the first part. After that, the spike protein actually sheds kind of a piece of it and unveils this very hydrophobic piece, which is a way to say it's fatty.

It's made of lipids, and why is that important. Lipids like to stick to other lipids, and the membranes of our cells are all fats, and so basically by unveiling this hydrophobic piece, it can stick into the plasma membrane, the fatty lipid by layer of our cells, and that way fuse basically make kind of this fusion between the viral membrane, which is also a lipid and the cell membranye.

But this little dance, you know, is well orchestrated, and so what we eventually did find is that one difference we could see between the variant, the mutant form of spike and the original form of spike is that this kind of processing that enables the protein to go through

this dance that in the end unveils. This piece that sticks into the host cell seems to be different between the wild type spike and the variant spike that it seems to be more resistant to certain kinds of premature unveiling, let's say, of this fatty region, and that actually might help it, because if you think about it, these viral proteins, they don't just come out and start to infect cells.

They're produced inside a cell that's already infected. And if the spike protein has to go through this kind of complicated dance where it changes its confirmation a little bit, if that happens too early, it might be an irreversible change. It might not be able to become functional again. And so if that happens, say in the cell that produces

the spike, then maybe that's too early. It really has to happen after it sees ACE two, it does the handshake with ACE two, and then it can kind of undergo this conformational change to stick itself into the membrane. And if that's happening too early, which is something that our data suggests, that might actually lead to varians that have spike on them. But the spike is not really functional, and perhaps what the mutant spike does is it just

leads to more functional spike on the surface of the viruses. Now, well, let's turn now to the bigger picture consequences of these really remarkable findings that you and your co authors have contributed to making. When the ordinary person hears that possibly the new version of stars covy two is five to ten times better at transmitting itself than the old version,

the natural thought is, oh boy, that's scary. So first question, is there any reason to think that an initial mutation of that sort would be an indicator that there could be future mutations that might similarly improve the transmission rate, that is, make it worse for us better for the virus. Or is it the case that just because this was a random point mutation, there's just no reason to think that there would be some other random mutation that would

make this an even better virus at transmitting itself. Yeah, that's a fantastic question. I mean the real answer is, of course, like many things COVID related, we really don't know.

I think based on this rapid evolution that we've seen just you know, with months of this virus circulating, I think it certainly is not beyond a shadow of a doubt kind of possibility that there might be another mutation, maybe in the spike protein, maybe in some of the other twenty five odd proteins that are in this virus that either leads to increased transmissibility or you could lead to hopefully not but some sort of increased lethality of the virus. And that sounds very scary. I don't think

it has to be scary. I mean, one thing that is very great to see right now, which we certainly didn't have during the nineteen eighteen you know, flu pandemic, is the use of rapidly deployable genomics thing. It's like the work that I'm talking to you about today, where we've been able to very quickly functionally characterize the impact

of some of these mutations. There's a large scientific community of people working on this right now, and I do think we can either react in real time to a lot of these mutations at least understand what their functional impact is, or but the kinds of cool DNA synthesis technologies and RNA synthesis technologies that exist, we can actually make mutations and test them in a massively parallel way to kind of figure out, hey, what does this mutation do,

what does that mutation do, and really fully characterize what's kind of a broad spectrum of what these proteins are capable of. And that way maybe we can start to predict and already think about, well, how do we improve our vaccines, how do we improve our therapeutics. And this has been done before with highly mutating viruses, so this is not a crazy idea to suggest. Perhaps the most mutagenic RNA virus that everyone knows well is HIV, the

virus that causes aids. That virus is tremendously mutagenic, and what was found in the in the early and mid nineties was that drugs that were seemingly effective against HIV didn't really work in the long term, meaning that the virus was able to evolve ways around the drugs. You know.

The real break through was in the late nineties the development of what we now referred to as the cocktail, which was a few different attacks on the virus, three different drugs brought together, and it turns out that even though the virus is very good at mutating that RNA virus, the three drugs together proved to be kind of a knockout punch and still to this date, twenty years twus after the development of the HIV cocktail, it is still effective.

And so I think that provides a really nice roadmap, you know, I think it should inspire us that we've been able to lead with science. Let's talk then about vaccines and reinfection and what the practical consequences will be for those of this observed mutation. Let's start with reinfection. If you're in China, this may matter much more if you're in China than if you're in Europe or the United States. But you got the early version and now here comes the mutated version back around it comes back

via Europe or the United States. I know there was concern initially in China that it might be that whatever immunity people have, and I realize we don't fully know how much community people have when they have been infected, but whatever immunity people did have might no longer be sufficient to hold off this new mutation of the virus. Is there data on that yet or do you have an intuitive sense absent the data, what is likely to be the case with respect your reinfection of people who

got it the first time. Yeah, there's not data from US, But there's data from several other groups that have now started to look either at therapeutic antibodies that are being tested or antibodies isolated from patients who have had a COVID nineteen disease. Course, what they found is that many of these antibodies targets, say, for instance, that ACE two binding domain, which is the part of spike that's kind of on the farthest away from the virus. It's on

the surface really of the virus. And so the good news is again because the mutation doesn't really seem to alter that receptor binding domain so much that most of those antibodies still are very effective against the mutant spike. So that'll suggests that this one mutation probably doesn't wipe out that kind of immunity. And just so I understand, is that because you did a great job of describing the two part process of how the virus gets you.

First it shakes hands and then it's in the door, and then it takes off its hat or something like that or its mask and reveals the fatty lipid bind to you and then you're stuck. Are you saying that the reason that the antibodies are likely to work. Nevertheless, is that they primarily target the initial handshake, and as you showed in your initial lab efforts, there's not actually a major change in the nature of the handshake derived from this mutation. It's the later part. It's the taking

off the mask. Yeah, I think that's that's exactly the case. Let's talk about vaccines in that case. Sure. So. Obviously, some of the vaccines seek to replicate precisely the antibodies that occur naturally, but there are also different kinds of vaccines that are being experimented with. Now. There are these so called trojan horse vaccines like the Oxford approach, there's

the RNA vaccine like the Maderna approach. What does the evolution in the virus suggest with respect to those vaccines that would those also be just as effective on the earlier version as the late version or is it trickier than that? Yeah, I think the implication for vaccines is something that definitely merits some research. So there's about one hundred and thirty or so vaccines under in various stages

of clinical development right now. In terms of how we think about virus manufacturing, one thing that I think was really impressive was how quickly some of these, especially the new clique acid based vaccines, the DNA and the RNA vaccines. How quickly we can go from sequencing the virus, you know, the first coronavirus sequence that was released in January, to having a vaccine ready to go, which was also in January,

as you mentioned with maderna. And so what might be more important than being too worried about is this vaccine out of date? Has it kept up with all the newest spike mutations? Is to think, how can we develop a process or a pipeline where we can quickly capture population genomic data and I mean circulating virus data, sequence the genomes quickly, and then quickly update the vaccines to take into account new restraints and this kind of whatever you want to call it, like tightly closed loop sort

of system or something like that. You know, that's really that could be a powerful process that doesn't just protect against this spike variant, but perhaps any future spike variants we might be worried about. Are there any general lessons from other viruses and the course of their mutations that

are relevant to us here? Yeah? I think you know, HIV is a great example because enormous resources were dedicated starting the late eighties onwards to fighting HIV, and there it still took more than ten years to have a truly effective therapy. So that's I mean, that's one thing to say, but we did end up with an effective therapy. Another reason why HIV, I think is a particularly good example is because, as some folks might know, there has been a long, decades long quest for a vaccine for HIV.

But today, even though there's any promising candidates kind of better time now than ever before, we still don't have an approved vaccine. This is a virus where we've known about it since the eighties, So that's you know, to me that that's kind of a scary thing, right that we can expend tremendous effort and many years and it still can be difficult to have vaccines. Now, that's one case.

There are other cases where vaccines have been developed in much shorter periods of time, just a few years here. I mean, we really do have the whole world focused on this, so I'm you know, I'm all for these optimistic estimates that we hear from you respected infectious disease doctors like doctor Fauci and others of you know, six months to a year, and I certainly hope that that's

the case. I mean, we all want our lives to go back to normal, but I think it's important with the historical perspective, we have to say that, you know, developing safe, effective therapies, safe effective vaccines, it's not easy, and I'm only hopeful that there's so much effort going

into it right now that it will greatly accelerate those efforts. Well, thank you for the work that you're doing and for explaining why we should be a concern about the irritation and why we should also recognize that it's not necessarily the end of the world. Thanks for the opportunity to

be here. To me, it's a rather remarkable fact that we can see in real time how the stars Cove two virus has been evolving the fact that the variation at place six fourteen was not visible almost at all in January, in February, was noticeable in March, was really noticeable in April, was at seventy percent in May, and it is now at ninety five percent provides significant reason to think that it's actually doing something to help the

virus transmit itself. Nevland his group have suggested that by improving the spike protein, the new version maybe five to ten times better at transmitting the virus than the version that existed before, and they've made significant progress in trying to figure out where and why that is happening. The consequences of this development are significant, and they're also subtle. On the one hand, Nevil says we shouldn't assume that just because there's been one mutation that made the disease

easier to spread, that there will be others. There might be, there might not be. On the other hand, he says, sometimes the reality is that we do get rapid evolution in a virus in a way that makes it difficult to contain the virus with a vaccine. The upshot is that we need to watch the development of this virus quickly. The good news is we can now do that. The speed and cheapness of sequencing genomes now makes it possible in almost real time, to track what's happening in a virus.

Never before in the history of pandemics has it been possible to keep as close an eye on the genetic variation and evolutionary pressures that are taking place within a disease as it spreads. The worrisome bit is that no matter how much science we have, and no matter how sophisticated we are at understanding what's happening, we don't necessarily have all the tools to solve the problem. We're going

to continue to watch this story. If the virus evolves more, you can be sure that we will talk about it right here on deep background. We'll be right back. Welcome to this week's playback. Hi, this is Anne with the Warranty Department. Our records show that your vehicle warranty has expired or it's about to expire. That is a sound that no one likes to hear, the sound of a

robocall reaching you on your mobile phone. The Supreme Court weighed in on the robocall issued this past week in a case where Justice Brett Kavanaugh wrote, Americans passionately disagree about many things, but they are largely united in their disdain for robocalls. The Supreme Court struck down a twenty fifteen law that made an exception from the general ban on robocalls to your cell phone for collection of debts that are backed by the government, which would include, for example,

your student loans. On the surface, nothing could sound more straightforward. How great that the Supreme Court, in the exercise of its infinite wisdom, has protected us further from robocalls. But that's not really what was going on. What was actually happening at the Supreme Court was an intense fight between the courts conservatives and the courts liberals about what standard they should use to analyze questions about the freedom of speech.

The conservatives want to use the highest standard called strict scrutiny, where the Court almost always strikes down a law that is seen to implicate free speech, and the liberals are concerned that the conservatives are going to use free speech doctrine to overturn progressive regulation on things like food and drug regulation, workplace safety, or the regulation of the al of securities on the stock market. To understand what was really going on beneath the service in the robotcall case,

you need thirty seconds on the constitutional law of free speech. Historically, the Supreme Court applied its toughest level of scrutiny to laws that seemed to treat different statements differently from each other based on the ideas expressed in them. Sometimes the Court called that viewpoint discrimination. That is a law that treats two people differently based on their viewpoint. Perhaps one is a Republican and one is a Democrat, and the

law treats them differently under those circumstances. The Court always said, we're going to look at this law very carefully, and we're almost certainly going to strike it down. That all changed a few years ago in a case called Read against the Town of Gilbert, when Justice Clarence Thomas wrote an opinion saying that strict scrutinies should apply whenever a law differentiated between different kinds of expression based on their content, not based on the ideas they expressed, but just based

on their content at all. In the robocall case, the Supreme Court relied on exactly that idea. The Court held that what was wrong with the government giving an exception to the robocall ban for government backed debt collection is that to do so, it had to ask what is the robocall about? The mine? You're asking what the robocall

is about, said the Court. You're looking at the content of expression, and any law, the Court said, that looks at the content of expression automatically gets stick scrutiny and gets struck down. Writing in dissent, Justice Stephen Bryer forcefully expressed his serious worry that applying that kind of content based analysis to government regulations about speech could end up invalidating the laws that tell companies you must disclose what's in your product, you must tell the truth about your

security's offerings. You must provide workplace warnings to workers so that they know what the dangers are that are facing them in all of those instances. Brier pointed out, the rule in question regulates content. It says that certain things must be said and other things must not be said. Bryer is worried that the Conservatives are going to use this very idea that all content based rules deserve strict scrutiny, to chip away and maybe go all in and strike

down huge swaths of government regulation. All of this may sound to you pretty far from robo calls, but you know what, that's often how things happen at the Supreme Court. On the surface, a decision that seems to touch on something relatively minor, the irritation of robocalls, and it seems to give you a good result. Meanwhile, beneath the surface, a long run battle between conservatives and liberals for the future of our country and how government is allowed to

operate under the constraints of our constitution. They're going to be several more Supreme Court decisions in the next week or two, which are likely to be high profile and significant. Will come back to those in a future playback. Until next time, Be careful, be safe, Be well. Deep background is brought to you by Pushkin Industries. Our producer is Lydia Jane Cott, with mastering by Jason Gambrell and Martin Gonzalez.

Our showrunner is Sophie mckibbon. Our theme music is composed by Luis GERA special thanks to the Pushkin Brass, Malcolm Gladwell, Jacob Weisberg, and Mia Lobel. I'm Noah Feldman. I also write a regular column for Bloomberg Opinion, which you can find at Bloomberg dot com slash Feldman. To discover Bloomberg's original slate of podcasts, wrote Bloomberg dot Com slash Podcasts and one last thing. I just wrote a book called The Arab Winter, a Tragedy. I would be delighted if

you checked it out. If you liked what you heard today, please write a review, or tell a friend. You can always let me know what you think. On Twitter, my handle is Noah R. Feldman. This is deep background