Elizabeth Clare: bats, cats and naked mole rats

Prof. Clare sampling air for eDNA. Image courtesy of Elizabeth Clare

"Click here to listen to the full podcast episode"

What if you could tell what animals had passed through an area without ever having to see the animal? Or figure out whether a shipwreck at the bottom of a dark and murky ocean contained human remains buried in the mud? These are very real possibilities thanks to advances in the field of environmental DNA, or eDNA for short.

One of the pioneers in this field is Professor Elizabeth Clare at York University, Toronto Ontario. Her lab has been developing techniques for sucking eDNA out of the air for use in conservation. But before she tells us more about that, I asked her to explain what is eDNA…

Elizabeth: eDNA stands for environmental DNA. And it's basically any material we collect that doesn't come directly from an individual or an animal. So if I come and swab your cheek, that's a direct sample that I'm taking from you. But if I just collect something in your environment where you've shed little bits of skin cells and hair cells and DNA into your environment, that is environmental DNA.

Elizabeth: So it's a generalised term for all the DNA material that doesn't come directly from an animal or a person.

Sally: And how much of this eDNA is around? Is there anywhere on Earth that doesn't have eDNA?

Elizabeth: Well, that's the most amazing thing. A few years ago, if you'd asked me that question, I think we would've said it was fairly rare and was hard to find.

Elizabeth: But now increasingly, the more places we look, the more places we find it. It's been found on the surfaces of leaves, you can wash a forest and get DNA. We're finding it in footprints left behind in snow, trapped in honey, trapped on the surfaces of insects. As they walk around, you can swab the insects and get other insects DNA off them.

Elizabeth: It's just everywhere. I think we live in this soup of DNA floating everywhere, and it's in the soil, it's in the air, it's in the water. It's just pervasive everywhere.

Elizabeth: And it's probably the hottest topic in the molecular ecology field right now is where can I get more environmental DNA? And most recently, what we got the idea to do was actually filter the air.

Elizabeth: We know that things like viruses and bacteria and pollens, a lot of pathogens have literally evolved the ability to disperse through the air. So we know we can collect those. We've known that for a while. But the concept that little degrading cells from animals could also be literally floating around in the air to collect was a relatively new idea.

Sally: And how do you go about sampling the air without knowing that you are just sampling like your own breath at the same time? Because we are a 70 kilo lump of DNA excreting material.

Elizabeth: We absolutely are. So we got the idea as the sort of crazy concept. And my lab has a history of trying slightly crazy things and sometimes they work and sometimes they don't.

Elizabeth: But we knew that the DNA was being shed into the environment because we know it's in the soil. So there was not a big leap to think that, well, if it's going into soil, it must get there through the air. And so we had this idea to try and find the most likely place where we might get concentrations of DNA in air.

Elizabeth: And in the facility I worked, we had colonies of naked mole rats. And so we thought, well, if we're gonna try it somewhere, that is the most likely place that we'll find a signal.

Sally: So in the lab, inside a room, you've got what these tanks of dirt and these, what are they, hand sized rodents?

Elizabeth: Yeah. They're sort of bigger than a mouse and smaller than a Guinea pig.

Elizabeth: And they live in these big colonies and so we took the air from inside their burrows and then also from in the room. To our total shock, not only did we get tons of DNA inside the burrows, we got the DNA in the room as well. And that room was full of naked mole rat DNA, people DNA from the people who were looking after the animals.

Elizabeth: We even found dog DNA in the room and that one puzzled us for a really long time until we tracked back and found that one of the people who fed the naked mole rats looked after his mother's dog on the weekend.

Elizabeth: And so now we got this idea that rather being a hard thing to find, that we would only find in like really tiny trace amounts inside the boroughs, we're actually seeing DNA being tracked back and forth between homes by the researchers themselves. And so it's actually turned out to be easier to collect than we thought.

Sally: It's easier to collect, but then you must be swamped with data when you are sequencing it, 'cause you've got the opposite problem, right?

Sally: You thought you wouldn't get much DNA and so you sequence it and then you see what you've got. But now it turns out there's so much DNA. How do you go about working out what species, each bit of DNA is from ?

Elizabeth: So you're quite right. We're basically collecting a soup, a mixed soup of all different kinds of DNA and all kinds of stuff in that sample. We target a tiny little piece of DNA, a really small fragment of the most abundant form of DNA that we can get from animals, the mitochondrial DNA. And these are bits of DNA markers that are known to be able to differentiate different species. We can't tell apart individuals, we can tell apart species.

Elizabeth: And we amplify that signal for a target group like mammals or birds, and then we sequence it on mass and then we can play some pretty simple bioinformatic tricks on the computer to separate all those signals out. We have these wonderful big libraries of known DNA we can compare to. And it's kind of like the children's card game Go Fish, you know: I've got one of these, have you got a match to it in your database? And we look for these perfect matches. We're doing really well at identifying the sources of those signals, in most cases, down to the species level.

Barking mad?

Dingos at the zoo make sure the researchers are following the correct eDNA sampling protocol. Image coutesy of Elizabeth Clare.

Sally: Not to underplay how incredible it is that you can find DNA from the air, but okay, you've managed to say that there are more rats in a mole rat lab. Well done! How else can this be used?

Elizabeth: Well, because we got excited by this, this is basically proof of concept. Is it even possible to get DNA out? So then we set out to do some real exploration experiments. The first thing we did, was went to a local zoo, and the zoo is great because it's a concentration of non-native species.

Elizabeth: And if I go to the British countryside near a zoo and I find tiger, there's only one source for tiger DNA in that environment. So we know...

Sally: You hope!

Elizabeth: We really hope! But we know then with complete certainty where that signal came from. So we can figure out is this a real signal? And how far did it come? So that was our first sort of controlled experiment that we did, and we were really successful with that.

Sally: So how far away from, say the tiger enclosure were you able to find tiger eDNA?

Elizabeth: In our first experiments, we estimated up to about 300 meters. The signal was traveling quite far. Then, yeah, it wasn't bad in terms of getting it across. And so then we decided to go and test this under different circumstances, much more normal, wild circumstances.

Elizabeth: So the first thing that happened was we were contacted by a team of researchers who had been studying plant DNA in dust. And they had been collecting dust out in the environment and analysing plant DNA from it. And they said to us, "Well, we found plants, but we never looked for animals. Can you check our samples for animal DNA?"

Elizabeth: This is Southern US in Texas in a really well known range land. And so we looked at their every two week samples and we found really good signals of local wildlife. So for instance, in the season where their area dries out and the vegetation disappears, you really don't find many animals out there.

Elizabeth: But once it cools off and the vegetation comes back, suddenly all the animals get really active and it's actually breeding season. And we suddenly start seeing spikes in other interesting animals, pocket gophers and frogs.

Sally: And while you're almost not just getting a presence-absence data, but I'm imagining a curve where it's like oh suddenly it's rocketed up and then slowly it'll decay over time.

Elizabeth: Pretty much. So each of our samples is really just a presence-absence test. But when you get lots of those samples over a period of time, you start to see patterns. One of the most specific signals we saw was this little ephemeral toad, which normally lives effectively in the mud. And when there's a sudden environmental change and it gets wet, it comes out and breeds.

Elizabeth: And in this range land, there was a dried out pond, and then there was a sudden signal of toads. And we went back and tracked the environmental conditions, and in the two weeks before the toad signal appeared, there was a massive rainstorm. And so we think the rainstorm basically hydrated this little ephemeral pond.

Elizabeth: The toads get active, come out breed, we catch their DNA, and then they go away again.

Sally: And of course no one needs to have actually seen the toad for you to know what had happened and piece all bits of the puzzle together.

Elizabeth: Exactly. And so with the sort of success of this passive dust collection, we decided to go and really test the system and we actually invented a whole new form of filter that works on batteries.

Cave of curiosities

Elizabeth Clare sampling eDNA from the air inside one of the caves where bats roost. Image coutesy of Elizabeth Clare.

Elizabeth: We modified some other designs that were out there to something that we could carry with us.

Sally: Yeah, what does it look like, your eDNA sampling machine?

Elizabeth: Well, my sampling machine is a combination of a fan and some 3D printed housing.

Sally: And how big is it?

Elizabeth: We've made multiple different versions. One is the size of a coin. It runs off a USB, like you can just plug...

Sally: A coin?!

Elizabeth: Yeah, sort of a big, like if you're in the UK...

Sally: A 2p piece. Or a 2 pound coin.

Elizabeth: Yeah. It's about maybe less than an inch across. If you're, you know, in the continental US or Canada, it's more like a silver dollar or a loonie.

Sally: That's tiny!

Elizabeth: It's tiny and it runs on a little battery that costs us $9 at the local business supply. Just the kind of thing you charge your cell phone off.

Sally: God, you could pop that anywhere.

Elizabeth: Yeah! Our large ones are only about four inches across at most. They're not very big at all, and we are using sort of an intermediate size, so we found the easiest to use. It's more powerful in terms of its sucking ability.

Elizabeth: So we packed as many of these as we could into literally our carry on luggage. And my brilliant student Nina and I headed down to Central America to a field site we work on. And so we're bat biologists, most of us on this big team that goes every year and we study this population.

Elizabeth: And two of us are the taxonomists. So we are in charge of identifying every single individual bat that comes in. And over the course of two weeks, more than a thousand individual bats are brought into this room for identification, and then they go out again that same night.

Sally: How do you catch a bat?

Elizabeth: Well, you use a kind of net, it's very, very fine filaments, like a fish net, but much, much finer.

Elizabeth: And they fly into it and they get caught and then you pull them out. And you put them in a little cloth bag because they just curl up and go to sleep in there and you carry them back.

Sally: Oh, so it's like what they do for birds.

Elizabeth: It's exactly the same as birds. Although sometimes even finer nets.

Elizabeth: But if you think about this room that we're in, it's like a cave. So think of it as an artificial cave. And all the bats come and go every night. We catch them in our nets, we bring them back, they go into the room, we identify them, and then they go outta the room again as if they were coming and going. And so in this highly controlled environment, we know precisely how many of every single species came and went from that room, on what days.

Elizabeth: We had, I think we had 35 species of bat that came and went from that room and the people and other stuff that came and went. And the question was, how much of this could we have recovered with DNA only in the room? And the answer was astonishingly high amounts. We got about 90% accuracy at diagnosing or describing the complexity of that community. We picked out bats where we only had one individual in that room, and there was a hundred other bats that night.

Elizabeth: But it also, our own activities had a really interesting effect on our data. So one of the biggest questions people ask is, could you estimate the number of individuals from the sample? So we know that Species X was there. Can I tell you how many? And the answer was astonishingly, yes, there's a quite good relationship between number of individuals and DNA signal. But then there were these four weird exceptions, and it took forever for me to work out why four species were way off this relationship. And it turns out I'm the problem.

Elizabeth: So I'm the chief identifier. I'm the chief taxonomist. My job is to look at every single individual and confirm its identification. For some bats that's really easy. It's that you don't look at it, it's obvious. And for others it takes a lot more effort. I can do most species in a few seconds and some take a little longer and you have to look and you've got your microscope or you've got your hand lens or your, in my case, my reading glasses on, trying to see little features and identify them.

Elizabeth: And when I finally looked at this graph, we'd made of number of animals and signal of DNA, and I suddenly realised there were three groups of bats that were hugely overrepresented. We had too much DNA for the numbers there. They were the ones that are the hardest to identify. So I spent more time looking at those individuals to confirm who they were.

Suck it up!

Researchers sucking air from a bat roost inside a tree hollow to sample it for eDNA. Image coutesy of Elizabeth Clare.

Elizabeth: And so my actions as the local bat identification expert actually influenced the amount of DNA in the air that we sampled. And then there was one bat where its DNA was really low compared to what we would've expected given its abundance. And those were vampire bats. You don't even have to look at those. If you touch the bag they're in, they emit a noise that's really characteristic and you go, "Oh, it's a vampire bat!"

Sally: I have to ask, what does the vampire bat sound like?

Elizabeth: They're a bit creepy cats. They kind of emit this little screaming noise and it's really characteristic. So you know, you don't really have to do much to confirm them. You can just open the bag, look in and go, "Yeah, that's what it is!" and put it back. So those ones I don't handle very much. And those are our four exceptions to the nice little prediction rule: the hard to identify, the really easy to identify.

Sally: And it really is the exceptions that prove the rule.

Elizabeth: Yes, exactly! But we also had some really interesting things. So when you do this kind of work, there's lots of DNA you would just normally discard because it doesn't meet lots of your quality requirements for a signal.

Elizabeth: We decided because this was so new and we had no idea it was really there, that we would try to track down every single one of those bits of noise that we nearly always throw out. And some of them were really interesting. So for instance, there were four sources of bat DNA that were not possible in Central America. These are African species. So why was their DNA in that room?

Sally: Oh, this is a puzzle! Ooh, is it a bat lab? Has someone got some dried taxidermied African bats as reference samples in there? Has someone been eating bats?

Elizabeth: No. But several of the individuals who came had been working with those animals a month previously in the wild in Africa.

Sally: It's like the dog on your researchers coat in the lab, the mole rate lab!

Elizabeth: It's the dog in the mole rat lab!

Elizabeth: And so what we think was happening was we're actually seeing these minute traces of DNA coming on people's luggage and being introduced into the environment.

Sally: So your definition of clean must be so much stricter than everybody else's definition of clean.

Elizabeth: My definition of clean just got a lot higher! So then we had one more experiment that we did, and this was our first try to truly apply this to do something in biology.

Elizabeth: So we went out and deployed the same samplers in a whole variety of different roosts where the bats are actually living in the wild. And some of these are roosts that we know really well that we've been studying for a long time. Others are roosts that we can't enter. They're too dangerous, they're unstable. We can't get physically to them.

Elizabeth: And so there were some roosts that we've never been able to survey fully. And so we put our samplers in all of these different places and we started measuring wild populations. And again, worked brilliantly well. We got signals from all the bats we would expect to be there. The coolest thing for us was that in one of these tree roosts, a giant big hollow tree, we caught the signal of a bat we've never seen. And every sampler we put in that tree recorded the same thing.

Elizabeth: So we got multiple signals that it's real for a bat, which the key that I helped write for that location: we think white-winged vampire bats are present, but we've never caught one - keep an out. So we had written a prediction that they were there. We felt they were there. We thought they were there.

Sally: Why did you think they were there if you hadn't seen them up to this point?

Elizabeth: It's the right environment. It's the right habitat. There's other vampire bats there. These ones are a bit hard to catch in nets. They tend to avoid them pretty well, and we were just waiting for someone to bring one back. And the DNA samplers found them in the tree!

Sally: And that just goes to show how useful these kind of technologies can be for conservation purposes as well. How is eDNA being used already in a kind of very practical sense? Not just, "Oh, can we see if it works?" But actually we are using this as a tool to find out information we couldn't otherwise get.

Elizabeth: Well, mostly it's being used in aquatic systems. So there it's become very, very routine to include aquatic sampling in your environmental DNA sample.

Elizabeth: Things like searching for invasives or rare species, just like our vampire bat that we predicted was there but hadn't confirmed yet. So we know that these kind of technologies work really well.

Elizabeth: I work in collaboration with one of the world's leading environmental DNA companies called Nature Metrics. They are helping us develop the technology further and one of the things that Nature Metrics does is works with a lot of governments and conservation organisations providing high quality facilities and data for conservation applications. And so one of the things that is coming down the line, I think, is the idea that it's being built into law, built into regulatory monitoring programs.

Elizabeth: So we're gonna start seeing a lot more requirement of environmental filtering of different kinds of sources to learn about the environment as part of our regulation of our own impact on the planet.

That was Elizabeth Clare from York University, Toronto, Ontario.