An anti-CRISPR system that helps save viruses from destruction

Shamini Bundell

Welcome back to the Nature Podcast. This week: the anti-CRISPR that helps viruses avoid destruction...

Nick Petrić Howe

...and how much of Greenland's melting can be prevented? I'm Nick Petric Howe.

Shamini Bundell

And I'm Shamini Bundell.

<Music>

Shamini Bundell

You've heard of CRISPR, but what about anti-CRISPR? Now, the CRISPR-Cas system is famous thanks to its use as a revolutionary gene editing tool. But its roots lie somewhere rather different. You see, CRISPR-Cas is actually a kind of bacterial immune system, one that recognises invading viruses called phages, and chops up their genetic material protecting against infection. And like all immune systems, it's part of an arms race between host and invader. And this is where anti-CRISPR's come in. You see phages, despite their small genomes aren't helpless. In some cases, they deploy anti-CRISPR systems in an attempt to neutralize a bacterium’s CRISPR-Cas defences. And this week in Nature, a brand-new kind of anti-CRISPR is being described. And it's one that works in a very different way from previously known systems. Reporter Benjamin Thompson spoke with Rafael Pinilla-Redondo, one of the authors of the new paper about the new find, and its potential biotechnology applications. Rafael started with a quick overview of how CRISPR-Cas immunity works in bacteria.

Rafael Pinilla-Redondo

CRISPR is very special, because it's an immune system that has memory against previously infecting viruses and chop them up with these, what we call nucleases, with our gene scissors, and essentially when viruses infect the cell, some CRISPR-Cas components are able to take some short sequences of the viral genome and integrate it into what we call the CRISPR array or the CRISPR memory. And this memory bank is then expressed into these short CRISPR RNAs. And these are what guide the Cas scissors, and when they find a match, then it will chop up DNA, or RNA, depending on the system.

Benjamin Thompson

But viruses aren't defenceless, though, right? They have evolved a number of countermeasures to protect themselves against this fate.

Rafael Pinilla-Redondo

Yeah, I mean, the simplest one would be viruses mutate, and so they mutate in the sequences that would be recognised by CRISPR. So that is not recognised anymore. But then some viruses have also acquired or evolved what we call anti-CRISPR's, or ACRS, that are proteins. And these bind, and specifically blocked the Cas scissors from performing their natural function. And so they essentially inhibit CRISPR immunity.

Benjamin Thompson

And that list of defences against CRISPR has been added to it seems, so you've identified another anti-CRISPR system, one that works in a different way. What did you find?

Rafael Pinilla-Redondo

In this paper, one of the main findings is that we provide the first demonstration that viruses also use RNAs to inhibit the CRISPR-Cas immune response.

Benjamin Thompson

There have been hints then that this system existed through genome studies of viruses that infect bacteria. So phages, what was known beforehand?

Rafael Pinilla-Redondo

Yeah, so previous computational analysis had found that they have these CRISPR repeat like sequences, and that's the minimal unit of CRISPR. But instead of being in the host chromosome, they would be in the genomes of phages. And it's unclear what they were doing. And in our work, we demonstrate that these CRISPR live repeat sequences are used by phages to inhibit the CRISPR-Cas immune response.

Benjamin Thompson

So there was this sense then that some phage genomes contain very small elements of the CRISPR system encoded within their own genomes. And you've been looking at one of these then and shown that when activated, it makes RNA that acts as a molecular mimic to try and defend against the host immune system. When a virus infects a bacterial cell. What happens?

Rafael Pinilla-Redondo

Yeah, so these are molecular mimics of host CRISPR RNA guides, and they're not functional, they're incomplete or faulty. And so they interact with Cas proteins, but not to form functional effector complexes. So, they're making sub complexes that cannot work, and thereby they inhibit the immune response.

Benjamin Thompson

Normally, then the host makes an RNA guide molecule that attaches to the Cas part of the system, and that guides it to the invading genetic material, and the Cas kind of acts as a pair of scissors to chop it up. But in this case, it seems that the mimic binds instead and stops the scissors targeting the viral genome.

Rafael Pinilla-Redondo

Yeah, so it's like a non-functional version that would compete with the standard CRISPR RNA but these complexes are blind, they wouldn't find any viral target.

Benjamin Thompson

So, it's just a numbers game, then in terms of is making enough of the mimic to block the system working?

Rafael Pinilla-Redondo

Yeah, so it's diluting the functional complexes in the cell by forming these non functional ones.

Benjamin Thompson

And how widespread do you think this system could be?

Rafael Pinilla-Redondo

We've only characterized one in detail. But we've done some computational analysis. I mean, we have an idea that they are widespread. But I think a lot more characterisation needs to be done now to confirm that some of these are also inhibitors, because it could be that some of these that we predict could be doing something else.

Benjamin Thompson

And you mentioned earlier on that viruses, despite having these really quite minimal genomes, in many cases have got potentially multiple defence systems. You talked about proteins, for example, to stop Cas, and now you've got this RNA mimic system as well. I mean, is it a surprise that viruses have so many ways of escaping defence systems?

Rafael Pinilla-Redondo

No, I think it's part of the arms race, right? In also bacteria don't only use CRISPR, they have different antiviral mechanisms. On the one side, you have bacteria employing all of these defensive mechanisms. And on the other side, you have viruses coming up with strategies to evade or block all of these immune systems.

Benjamin Thompson

I mean, in terms of the arms race, bacteria still exist. So the viruses haven't won. I mean, in terms of your system, do you think there's an anti, anti-RNA based CRISPR system to be found?

Rafael Pinilla-Redondo

Very likely, I mean, there's indication that there's anti, anti-CRISPR. It's a bit complicated, but there is an indication that anti, anti-CRISPR proteins because they inhibit the activity of anti-CRISPR expression.

Benjamin Thompson

And in terms of CRISPR-Cas, of course, this system, once it was discovered, and maybe laid out, it's been quite a game changer, it has to be said for molecular biology and gene editing, and so forth, do you think there's the chance that the anti-CRISPR system you describe has uses outside of just protecting viruses when they tried to infect bacteria?

Rafael Pinilla-Redondo

Anti-CRISPR's have found applications in biotechnology, many of these applications need an off switch. And so anti-CRISPR proteins have been used to turn off CRISPR-Cas activity. But the problem with anti-CRISPR proteins is that they're very difficult to identify in nature, what I think it's a very interesting feature of these RNA based anti-CRISPR's. You know, by learning how they work and how they have evolved, that we might be able to rationally design anti-CRISPR's on demand for different applications. And I think some application could be in enhancing phage therapy that specifically killed bacteria. So phage therapy is becoming more and more popular because of the rise of antimicrobial resistant bugs. And so it's basically the use of phages that specifically kill bacteria. And of course, if you want to introduce phage therapy, you want to bypass host immune strategies, right. And so if we able to equip phages, with anti-CRISPR strategies that would be more effective at taking over bacterial populations and killing nasty bugs.

Benjamin Thompson

Finally, then what surprised you the most about this system then, because there was evidence that it did exist, and you've shown that it works, what caught your eye when you're doing the work?

Rafael Pinilla-Redondo

I think it's just fascinating how viruses are employing components of the bacterial immune system to fight that bacterial immune system. It taps into like a very beautiful or like very widespread theme across the evolution of viral evasion strategies across the tree of life. So, we have viruses in humans, plants, and so on that are employing similar strategies to inhibit the host immune response. So, it really highlights this double edged nature of host immune components, where they ironically can be co-opted by viruses and used against the immune system to slow it down, like fighting fire with fire.

Shamini Bundell

That was Rafael Pinilla-Redondo from the University of Copenhagen in Denmark, to read his paper look out for a link in the show notes.

Nick Petrić Howe

Coming up, projections for how much humanity has to do to prevent the worst melting of the Greenland ice sheet. Right now though, it's time for the Research Highlights with Noah Baker.

<Music>

Noah Baker

A queen named Thyra might have been the most powerful person in Viking Age Denmark, according to a new study of ancient ruins. The name Thyra is mentioned in inscriptions on four carved memorials called runestones in southwestern Denmark, all four date to the mid 10th century, but two are in a site roughly 30 kilometres from the others. Researchers analysed the shape of carved characters the carving technique and the language used to determine whether all refer to the same woman. And the similarities they found suggest that two stones were in fact carved by the same person. And that detail along with other names carved into stone indicated that yes, indeed, all four stones did refer to the same person. The team say this suggests Thyra was a particularly powerful and celebrated figure key to the creation of the Danish state. Not even Thyra's famous son, Danish King Harold Bluetooth, is mentioned on that many stones. The authors say this suggests women had even more influence in Viking Age Denmark than was previously thought. You can find more on that research in Antiquity.

<Music>

Noah Baker

Astronomers have found one of the fluffiest planets ever discovered. Now you didn't mishear me this exoplanet is known as a super puff planet. And it's mostly gas, giving it a density that's only 1.5% that of Earth. The planet given the catchy name TOI-1420B is among the least dense of all the 5500 known extrasolar planets. It was discovered when it passed in front of a star causing a weekly dip in the light recorded by a NASA satellite. This transit as it's known was then investigated further using observations from ground-based telescopes, which allowed a team of astronomers to estimate its size and mass, whilst TOI-1420B is about 25 times heavier than Earth, it's also more than 1700 times larger, which makes for a very puffy planet. The team concluded that the atmosphere, mostly helium and hydrogen, makes up most of its mass, and speculate that it was formed farther away from its star and migrated closer in by interacting with a large, unseen companion planet. You can read more on that study in the Astronomical Journal.

<Music>

Nick Petrić Howe

When you think about climate change, probably one of the first things you think of is melting ice. Take the Greenland ice sheet, this vast sheet of ice covering almost 80% of Greenland holds around 3 million cubic kilometres of ice. If all of that was to melt, it would lead to around seven meters of sea level rise. In fact, with current climate warming, the ice sheet is already melting. And we're already committed to some amount of sea level rise, a potential disaster for low lying and flood prone areas. But exactly at what temperatures the ice sheet would melt, how quickly and how much humanity can actually prevent this from happening are complex questions that climate scientists are trying to figure out. But thanks to a new paper in Nature, we may have an idea. It looks at the specific temperatures at which this huge stretch of ice would transition from its current state to so called rapid melting, known as its tipping point. I caught up with one of the authors of new paper, Nils Bochow and he told me more about what we already knew about the Greenland ice sheet's tipping points.

Nils Bochow

So, if you cross this critical threshold that we call tipping point, then we expect that there's an abrupt change in the system. For the Greenland ice sheet, we expect that we transition to another stable state that can be a completely ice-free Greenland, for example. But what hasn't been done so far is that if we crossed these critical thresholds, we might have some time to reduce the temperatures again, before we actually have this change, because the ice needs a lot of time to melt like on the scale of hundreds, thousands to maybe even 10,000s of years.

Nick Petrić Howe

And so when you say tipping point, I imagined that it's quite a sudden thing, but from your description, it is a sudden change to a different state. But the actual melting would take a while, is that correct?

Nils Bochow

Exactly. So I have to define what is sudden actually, sudden means for Greenland for example, that it is, as I said, several thousand to 10,000 years.

Nick Petrić Howe

But nonetheless, there is a specific or at least there should be a specific temperature, global mean temperature, at which point it will transition to this other state.

Nils Bochow

We have this critical threshold and you can express it in terms of global mean temperature. And then our study we find it's between 1.7 and 2.3 degrees above pre-industrial temperatures, where we have this abrupt change. So that means if we stay at temperatures above these critical thresholds, we would lose the Greenland ice sheet completely.

Nick Petrić Howe

And on what sort of timescale would we lose the Greenland ice sheet in this case, if we go to this 1.7, 2.3?

Nils Bochow

If you stay very close to these thresholds, it can take several thousands of years, even up to 100,000 years if we basically exactly on the threshold temperature. But the more we go above the threshold, the faster it is. So, if we go to six, seven degrees above pre-industrial, it can take maybe only 1000 years.

Nick Petrić Howe

And so to get to these particular numbers, that 1.7 and the 2.3, what did you do? How did you sort of go about determining these thresholds?

Nils Bochow

So, what we did is that we use two independent ice sheet models and we basically start with present-day conditions of the ice sheet of the climate. And then we warm the surface temperature around the Greenland ice sheet, above the Greenland ice sheet. And then we also looked at the latest generation of climate models. And we look what is the relationship between the temperature in Greenland and the global mean temperature, then we get this factor that we can use to translate the temperature in Greenland to the global mean temperature.

Nick Petrić Howe

I mean, when you put it like that sounds very straightforward. So how has this paper sort of transcended our previous understanding of the Greenland ice sheet?

Nils Bochow

To our knowledge, it's very much in agreement with previous research or studies. So, I think the consensus is between 1.5 and three degrees in previous studies. So, I would say we are in the middle of the consensus of critical thresholds for the Greenland ice sheet.

Nick Petrić Howe

But then one interesting thing about your paper is that you also looked at what would happen if we're able to mitigate these temperatures in the future, if we're able to bring the sort of temperatures down, if we overshoot these thresholds. What did you find in that case? What sort of timescales were better or worse for the Greenland ice sheet, and then maybe it sounds obvious, like quicker, the better. But could you tell me a little bit more about that?

Rafael Pinilla-Redondo

Yeah very simple, the quicker the better, but it depends. So, we have the overshoot temperature, so the temperature we reach at the end of the century, then we have what we call convergence time that this time, we need to go back to a temperature and simply put, the longer the convergence time, the more sea level rise you have and the higher the overshoot temperature, the more sea level rise we have, so... But for very short timescales, that is, within 100 years, then the overshoot temperature is not that important, but the temperature we reach within these 100 years is the most important thing. Because for the Greenland ice sheet, 100 or 200 years of warming is a relatively short period. So the ice needs some time to react. And 100, 200 years is not that long. But the important thing that we have to keep in mind is, even if we are able to mitigate a large loss of the ice sheet, by the warming that we have today, we already lock in some sea level rise, because we warm and the ice is reacting so slow, that it also reacts, of course, slow to our cooling again.

Nick Petrić Howe

And so what would you say then, are the implications of your findings?

Nils Bochow

The implications are, as I said, first that we already today determined to some extent what sea level rise we will have, from the Greenland ice sheet, in the maybe next 200 years, 100 years. But the other thing is also that we have a chance to mitigate large sea level rise, if there is a political will to either limit the global warming today, which is the easy option. Or if we reduce the temperatures after we reach the maximum temperature again.

Nick Petrić Howe

That was Nils Bochow from the Arctic University of Norway. For more info, check out the show notes for a link to the paper.

Shamini Bundell

Finally on the show, it's time for the Briefing Chat, the part of the show where we discuss a couple of stories highlighted in the Nature Briefing. So Nick, why don't you go first this week?

Nick Petrić Howe

Well Shamini I've got a joke for you to start my section of the Briefing Chat this week—

Shamini Bundell

—yeah, great—

Nick Petrić Howe

—so how many ecologists does it take to answer a question?

Shamini Bundell

I don't know, Nick, how many ecologists does it take to answer a question?

Nick Petrić Howe

246. But they all get different answers.

Shamini Bundell

Hurray! I mean, I don't get it. But hurray! Please explain your joke. Explain the punch line. That's, that's always a good sign.

Nick Petrić Howe

So it's not the best joke, but it does illustrate the story that I'm talking about this week. So, this week, I was reading an article in Nature about a sort of reproducibility trial where a number of ecologists were sat down, and they were trying to answer the same questions using the same data set. And then some other researchers looked at what they came up with to see if they were consistent or not.

Shamini Bundell

Oh, so this was a sort of artificially set up reproducibility test for a bunch of individual ecologists then?

Nick Petrić Howe

Yeah, exactly. So, the way it was set up is that the ecologists were given one of two data sets and a question to answer. So the questions were, to what extent is the growth of nesting blue tits influenced by competition with siblings? Or how does grass cover influence eucalyptus species seedling recruitment? So standard sort of ecology-based questions, and they were given the same data set depending on which question they were trying to answer, and then sent away to analyse the data. And what this study found is that people came up with very different answers. Thus, my sort of joke at the start.

Nick Petrić Howe

I can understand that there might be some variation depending on the sort of statistical methods used, but surely in the same dataset with that many ecologists, there's only there's only so many different answers that you can get what what were they coming up with?

Nick Petrić Howe

Well, for the blue tit one, most people agreed that sibling competition negative athlete effected nestling growth, but it's the size of the effect where the differences were. So, people were quite widely different on that, for the eucalyptus species question, it was really quite different. So on average, there was no effect found. But some people found really strong positive effects, some people found really strong negative effects. So, there's a real wide variety in there. And as to why this is that people are coming up with different answers to the same question with the same dataset, according to the authors of the study that was testing this, which is a preprint — so, it's not been peer reviewed — but according to them, this actually reflects the participants training and how they sort of set sample sizes. So more sort of their background and how they approach data rather than anything inherent to the data itself.

Shamini Bundell

So we've reported on sort of reproducibility issues before, which usually involves people taking particular paper and trying to reproduce it. So this is obviously a sort of really big study, which I mean, to me the results sound kind of worrying, what have people been saying?

Nick Petrić Howe

Well, the authors of this study say that what we normally do is we treat individual papers findings as pretty definitive, but they say that these results show that we can't rely on any individual study to tell us like the whole story. And it also sort of speaks to a larger problem. So you've probably heard of the sort of reproducibility crisis, this is something that people think of, particularly in the social and psychological sciences — but I don't think any particular branch of science is immune to it. And that's actually where this idea of getting a bunch of people to study the same data came from, from social sciences, it's something called the many analysts method. And yeah, it just sort of may give energy to certain movements that are trying to improve reproducibility in science, generally, because it shows that there are many different ways to answer a scientific question, and so therefore there needs to be more effort to put into trying to reproduce it better or coming up with better systems to understand data.

Nick Petrić Howe

And do they have a suggestion for what these better systems might be? You sort of suggested, you know, you can't just rely on one paper. But here, they've got quite a few people working on the same thing, and it's still not very reliable.

Nick Petrić Howe

Yeah. So the authors say that people could use practices common in other fields to show like the range of potential results you can get. So in economics, for example, there's something called a robustness test, which requires researchers to analyse their data in several ways, and then assess the variation they get from analysing the data in these different ways. And so that can give you an indication of how much variability there is in answering the question. But you know, the other issue that this study raises as well, is that ecology itself is quite a tricky field, because like, there's a lot of variability in the natural world, like, it's not a controlled conditions a lot of time is very observational. So it could just be inherent to the field as well, that people are very used to sort of observational data. And, you know, dealing with a lot of variation in how to interpret it.

Shamini Bundell

I feel like there's a lot of potential for more of these particular kinds of data analysis studies, as you say, across different fields and more studying of the reproducibility crisis. So hopefully, we'll have more of that on the podcast as this story progresses.

Nick Petrić Howe

Well, what I'm looking forward to hearing from you is to reproduce my excellent Briefing Chat, and for you to tell me about your story this week Shamini.

Shamini Bundell

A bit different, bit different from yours, not sure whether this counts as anywhere near a reproduction. My story this week is about building roads on the moon — paving over the moon.

Nick Petrić Howe

I mean, this sounds fascinating, but I must say, I don't think there's a great amount of traffic on the moon. So where's the push coming for for this?

Shamini Bundell

Well, it's, I mean, compared to the olden days, I'd say the moon has quite a lot of traffic these days, all these rovers and landers and all sorts of things. So I've been reading about this in Nature and The Guardian and this is based on a Scientific Reports paper. And yeah, there is a real need for some sort of... when I say paving or roads, what I really mean is a flat solid surface on the moon. That is not, and this is the key, dusty, the moon is annoyingly dusty. So if you see the footage of astronauts bouncing around and driving their little vehicles over it, they kick off all this dust. The surface of the Moon is called like the lunar regolith and it flies up into the air and the moon has really low gravity, so apparently this dust gets everywhere, and just causes all sorts of problems. It can interfere with scientific instruments, it can sort of start wearing things away. There have been times where you've got vehicles where like the radiator has got covered in the dust, and then it's overheated. So this whole like pavements on the moon thing is all about trying to overcome this dust issue.

Nick Petrić Howe

And I guess, you know, when you're going to the moon cargo space is important, so i'm imagining people aren't taking, like tarmac and things to lay down paving for roads and that sort of thing. So how would they actually go about making these flat, less dusty areas?

Shamini Bundell

Well, exactly. They're not about to transport any building materials from Earth to the Moon. So what have we got to work with on the moon? Well, what we've got is dust—

Nick Petrić Howe

—dust is the answer and the problem—

Shamini Bundell

—dust is the problem and the answer. What they've done... So this is, I will note that this is an experiment that has been done on Earth, not on the Moon. So it still needs to be reproduced accurately on the moon. But they've got on the Earth, a particular material thats basically used as a stand in for lunar regolith, like fake Moon dust basically. And their methods for turning that dusty substance into something hard and solid, that doesn't get everywhere, is aiming a giant laser at it, and basically melting it into a sort of hard kind of glassy substance.

Nick Petrić Howe

Wow. I mean, that sounds like a super Sci-Fi way to solve this problem. But again, we don't have like Death Stars, lasers that can go a long way from Earth to the Moon. So how would they do this, like on the lunar surface itself?

Shamini Bundell

You could theoretically take a big laser to the Moon. Their plan is actually to use sort of concentrated sunlight to basically get a big lens that's going to focus the sunlight. So rather than having a solar powered laser, just use the sunlight directly, focus it, get enough heat to melt this dust, which will hopefully work with the lunar dust as well as it does in the lab. This obviously, you know, some things might be different, they don't exactly know — tests obviously need doing on that. But they reckon that if they had a lens around two meters wide, that will be suitable for their current sort of technique that they've developed. And potentially, it could even be like a foldable lens, so you don't have to take a giant glass magnifying glass up in your spaceship with you. And they've been doing a lot of testing on what the sort of beam needs to look like. So, they're not just sort of like taking a huge giant beam and zapping a straight line and being like boom road. It's relatively small, so the beam is I think, just under five centimeters in diameter, and they've been working on, right, let's make a little shape. So you've actually sort of drawn a triangle with this beam. So you've got this sort of triangular, hard bit, sort of hard, glassy bit, it goes about two centimeters down into the dust, I think the sides are about 22 centimeters wide. And they figured out that they basically need to be careful not to overlap. If the beam tries to sort of go over an area that's already been melted, it tends to crack this stuff is kind of brittle, but also has pretty decent compressive strength, which is sort of pretty important if you want to land things on it or drive things on it. And then we're working with shapes that can all fit together, basically. So you can kind of slot all these shapes together and make this big flat surface.

Nick Petrić Howe

I mean, it sounds, as I say, sounds super cool. And sci-fi, so I'm getting all sorts of ideas in my head. But this is a lab test, how far does it need to go to become a sort of lunar reality?

Shamini Bundell

Yeah, this is very much promising initial results with a lot more research needed, like I said, like, does this work under low gravity? What happens if you build your little road, and then you have sort of rocket thrusters hitting them, lots of lots of unknowns, kind of a lot of potential as well, because you know, building on the Moon, some of the people behind it have even been talking about whether this can be useful for building structures on the Moon, because, again, you need to work with the material that you've got there, you can't really take a lot of it with you. And so there have been a lot of questions about okay, what kind of material properties does this sort of resulting glassy substance have? And how useful is this going to be for different kinds of structure? So tons of research to go but yeah, very fun to sort of picture and imagine they have some sort of exciting illustrations of their hypothetical future moon with these nice little interlocking triangles acting as a sort of landing pad and pathway coming away from it. So yeah, pretty cool.

Nick Petrić Howe

No it sounds super cool. And listeners we will actually put some links in the show notes, where you can see pictures of some Sci-Fi recreations of what this might look like, or the actual melted materials themselves. And we'll put that in the show notes, along with links to the stories that we discussed, and a link of where you can sign up to Nature Briefing to get more stories like them direct to your inbox.

Shamini Bundell

And that is all for this week. As always, you can keep in touch with us. We're on X our handle is @naturepodcast, or you can send an email to podcast@nature.com. I'm Shamini Bundell...

Nick Petrić Howe

...and I’m Nick Petric Howe. See you next time.