What's in the water?

We’re diving into the stuff that makes up two thirds of the Earth’s surface...
14 November 2019
Presented by Phil Sansom
Production by Phil Sansom.


An underwater view of the ocean surface.


This month we’re diving into the stuff that makes up two thirds of the Earth’s surface. Can you use genetics to figure out what’s in the water? We put the science to the test by making a geneticist guess our mystery fish. Plus, a story about whales and dolphins: what do you lose when you leave the land? Jump in, the water's fine.

What's really inside a fish tank? Read the results of our DNA scavenger hunt here.

In this episode

Fish with a question mark overlaid.

00:32 - Will It Sequence: a mystery fish

Supposedly geneticists can use the water from a fishtank to determine the fish it used to contain. Is it true?

Will It Sequence: a mystery fish
Tony Sapsford, Frog End Aquatics; Louise Fraser, Illumina

Supposedly geneticists can use the water from a fishtank to determine the species of fish it used to contain. Is it true? To put it to the test, Phil Sansom went to meet Tony Sapsford, owner of Frog End Aquatics - where our mystery fish resides...

Tony: My name's Tony and I am the owner of Frog End Aquatics.

Phil - And Frog End Aquatics, in case you can’t hear it in the background, is chock full of bubbling fish tanks.

Tony: Basically it's a large tropical aquarium set up, if you like. There's 86 tanks in total.

Phil: 86?

Tony: 86 tropical tanks, 10 cold water tanks, 2 big koi ponds and 4 marine tanks.

Phil: How many fish are in here?

Tony: Christ almighty, Got to be probably five or six thousand.

Phil - I was there for one type of fish in particular. But as Tony led me past tank after colourful tank, I couldn’t help getting distracted.

Phil: Wow.

Tony: Well as you can see there are seahorses, shrimps. This is saltwater.

Phil: God, I forget how beautiful seahorses are. They're tiny and jet black.

Tony: They are, yep.

Phil: Now Tony, have you got a favourite fish?

Tony: Yes. I mean to be honest with you, these days my favourites are puffers. When I started in the hobby thirty odd years ago that was one of the first things I had, was a giant puffer fish.

Phil: Really?

Tony: That is sort of what got me into it. I had a big one - he was nearly fifteen inches long. He would literally fill up with water and when he wants to go down he would go to the surface and spit the water out.

Phil: Really?

Tony: Yeah, yeah, they are cool.

Phil - Enough distractions. I was there with a job to do.

Tony: Right, we are just heading over here.

Phil - We went over to a tank with a very special type of fish. I had a rinsed-out jam jar, and I handed it to Tony to collect a water sample.

Phil: Oh my god you are putting your hand in!

Tony: It’s not going to hurt me, trust me.

Phil: Are you sure?

Tony: Yep.

Phil: You’re just leaving your hand in there!

Tony: Yeah, literally they are so shy, they will just go and hide behind the up-lift.

Phil: I thought they’d absolutely go for you.

Tony: No, not at all they are very, very shy things.

Phil - I had my sample of water from the tank of my mystery fish.

Phil: What do you expect to find in there?

Tony: Probably fish poo. They should find fish poo and probably any remnants of whatever they have been eating today - I mean they are quite messy feeders.

Now that the sample of water is ready, Phil Sansom took it over to the lab at gene sequencing company Illumina, where Louise Fraser works...

Louise: Hi, I'm Louise Fraser. I work at Illumina and I have a team who focus on extracting DNA from human blood samples, saliva samples.

Phil: Louise, I've got a present for you.

Louise: Wow, this is very different to what we would normally receive. Normally we would receive a sample in a sterile container. So what I'm looking at is a jam jar which appears to be full of water.

Phil: Are you going to have to turn down my sample?

Louise: No, we're always very happy to receive all kinds of samples

Phil: Very diplomatic of you, I am going to give this to you.

Louise: Thank you. It does just look like tap water with a few bits in it.

Phil: Now I'm not going to tell you what kind of fish was in that tank, but just off the bat - any guesses?

Louise: I don't know, goldfish, or some of those very large koi/carp type fish? I don't know. You just can't tell from the water.

Phil: Tell me what exactly you're going to do with the sample first.

Louise: What we have here is about half a litre of water and that volume is too big to easily process in the lab. So we would concentrate any of the DNA that's in that sample by flowing the water through a filter. We would then lyse any of the cellular material in the sample which means to break open or pop open the cells to release the DNA into the solution and we would then be able to isolate the DNA by binding the DNA to a magnetic bead so we can easily pull out the DNA.

Phil: Apart from the fish that was in that water, are you expecting to find any DNA from anything else?

Louise: Yes we would. You would expect to find the microbiome of the fish tank so that bacteria, for example, that may like to live in those fish tanks, and knowing the sequences of those bacteria might help us to identify what the fish was and we might find traces of what was originally in your jam jar.

Phil: Oh no.

Louise: That depends how well it's been washed out.

Phil: If you can get to the jam and figure out what kind of jam that was. I'll be so impressed.

Louise: No promises, we will aim for the fish and we'll see what else we get.

Phil: Now what about the food that the fish was eating. Are you expecting to find that as well?

Louise: We might find DNA sequences from that. I'm not quite sure what goes into fish food.

Phil: Are you excited?

Louise: I'm very excited. I'll be very interested to see the result, see what we can find. Before, we were able to do high depth sequencing on samples like this, you would only look for things that you knew were there, but with sequencing we can look for everything and then try and figure out what was there in the first place.

Phil: Good luck!

Louise: Alright, thank-you very much!

A whale shark swimming amongst fish.

06:10 - Ocean DNA and marine conservation

Getting DNA out of water is an extremely useful tool for wildlife conversation...

Ocean DNA and marine conservation
Mike Bunce, EPA NZ

Getting DNA out of water is an extremely useful tool for wildlife conversation. You can use genetics to conduct a ‘marine audit’, where you categorise all the living things in an ocean environment to figure out how healthy it is. Mike Bunce, the Chief Scientist at New Zealand’s Environmental Protection Agency, is one of the leading figures in this field. Chris Smith spoke to him last year when he was leading a lab at Australia’s Curtin University, researching trace and environmental DNA. How does it work?

Mike - What environmental DNA is, is our ability to sort of capture DNA that's just exuded into any biological sample. So what we do is, we'll take a big bucket of sea water, we’ll filter that onto a very small membrane to capture all the particles that are floating around in that water column, tease out the DNA molecules or photocopy up specific bits of that DNA, that can tell us things like what fish can we find contained within that.

Chris - So this is like a DNA fishing expedition in the sense that you don't know what's in that bucket of water. You just know there’s some DNA in there and then asking ‘well actually, what is in here" by comparing it to DNA sequences we know.

Mike - That's right. When people generate DNA sequences from known organisms that generates the reference barcode collection, then we can transfer onto an environmental sample like seawater, and we can look at all the fish DNA barcodes within that, or all of the crab DNA barcodes from crustaceans. And we can compare those to the references and then make inferences about what is present there, ‘are we finding new species?’. There’s quite a high profile example of the HMS Sydney, which is a shipwreck that went down in World War II. We got some water from that, extracted the DNA from it, and we found a fish that's only 90 per cent related to anything on the database. So this is probably a fish not yet known, because most of the fish are actually on reference databases now, so I think we've discovered something new.

Chris - Where is this DNA coming from? Is this DNA that's been, for want of a better phrase, pooped out by the fish that you are effectively finding?

Mike - Fecal material that's being pooped out of anything is one major source of it. But organisms in any environment defecate, urinate, slough cells off, things drop off it - it's just that in a marine environment, all of this DNA is sort of homogenized into a nice little soup ready for us to collect up. We've got a natural made blender that's already been sort of blended together for us.

Chris - Someone mentioned to me that, talking of blending things, that you're actually looking at some outputs from one of the largest fish in the sea.

Mike - That's right. That’s part of the work that we're doing with one of our PhD students. They've collected poo from whale sharks. And so, when they defecate, when they're up at the surface, they get out a net and try and scoop some of that material up and then we take it back to the lab. We blend it up, we use our environmental DNA bar coding approaches and look at what we can tease out of it.

Chris - Without being too graphic, what does whale shark poo look like? Is this not just liquid?

Mike - Yeah, it's a big plume of brown goop that comes out the back end and you do have to use a net to collect it because there’s little solids floating around in there, and we literally just scoop it up. So it's not overly pleasant to send a diver swimming through a massive plume of whale shark poo, but you've got to make these sacrifices for science.

Chris - What are you finding in there?

Mike - There's a couple of things we're finding. First of all, we can get whale shark DNA out of the poop, and that's significant because we can then sample that environment for whale sharks non-invasively, without touching the organisms. But we're also getting the window into all those, sort of, zooplankton, copepods, decapods, all these small little krill-like creatures, that big filter feeders pick up. And what we're trying to do, because we've got whale shark samples from around the world that we've collected now, is to try and understand how they're eating different things, at different times of the year, in different geographical locations.

Chris - And you'll be able to track that?

Mike - Well, hopefully. We've only got about 25 different whale shark poo samples that have been sent in by some of our collaborators, because they're quite rare. Most of the time they don’t poo when they're on the surface, apparently. So, we try and pick up what we can, where we can.

Chris - The idea being, then, you'll be able to track not just where they go and what they're eating, but when they're eating it, and therefore what food supplies they depend on where. So we'll have a better idea about conservation, that kind of thing.

Mike - Yeah, well, at its very base, conservation of species is about conservation of habitats. And when you know the food web of whatever species you're trying to conserve, you've got a better indication of how they might respond.

Chris - But how do you know that the signals you're getting correspond to something that the animal has actually eaten and not, given how sensitive your techniques are, not just stuff that’s floating around in the water they've just pooed into?

Mike - Again, it's a good question and the simple answer is we don't. It's probably a combination of both. We end up with these sort of Russian doll effects where even if a large krill has eaten other types of organisms and a whale shark eats that, we end up with, you know, dolls within dolls within dolls. And so to answer that question well frankly, we don't really know,  but it's better than the information that we've got at the moment, which is just looking at a big plume of brown stuff.

Chris - I suppose one of the major benefits of doing this is that you're effectively auditing what's out there, without actually having to go out there apart from armed with a bucket, where previously it would be fishing expeditions, it would be diving expeditions, and relentless counting expeditions. This is much easier

Mike - It is easier. We can literally wade into the ocean and scoop up a bucket. But I will say there's lots of different methods for auditing marine ecosystems from everything from basic underwater cameras, through to visual sensors, and DNA is really just added another quite powerful part of that toolkit. And there's many scientific methods. The more proxies or more ways you've got at looking at a question, the better your answers are going to be. Truly, environmental DNA is a powerful part of this new toolkit because it can look at multiple levels. It doesn't just look at fish, which is historically perhaps how people assayed marine environments, assess them. We get to look at all the crustaceans and even the bacteria that are contained within that and phytoplankton and corals, because collectively that is what makes up the base of the food web and marine ecosystems.

The tail of a piranha.

14:09 - Mystery fish revealed!

The results are in. Did you guess right?

Mystery fish revealed!
Louise Fraser, Illumina; Tony Sapsford, Frog End Aquatics

After a week and a half, the results are in. Phil Sansom went back to Illumina’s Louise Fraser to see if she could identify our mystery fish. Did you guess right?

Louise - The results just came in yesterday.

Phil - Ready to guess?

Louise - Oh, absolutely! What we got was around 500 million DNA sequencing reads, and we compared them to a database that contains all the species that have ever been sequenced completely. And about 98% of the reads actually didn’t align to anything in the database. Most likely those would be bacterial samples…

Phil - But those aren’t new-to-science creatures, they just haven’t had their genome sequenced yet?

Louise - That’s right, they’re just things that aren’t currently in the database.

Phil - OK, 98% unknowns…

Louise - Of the remaining 2% they broadly fall into four different groups. 95% of the classified reads are bacterial; 5% are from animals, and 1% from plants; and then there’s a tiny fraction that come from fungi and archaea. The most common animal species in the water sample was actually human. But the second-most-common animal was a family of fish called cichlids. And so that’s what we think was in the tank.

Phil - Is that your final answer? Louise, I hate to tell you… it’s not a cichlid! Any other ideas?

Louise - Yeah, so there’s low level of DNA from a number of other animals, such as rice fishes, carp, and even piranhas.

Phil - OK, this is interesting, because it’s actually one of those.

Louise - OK…

Phil - Do you want to guess which one?

Louise - Gosh, I’d like to think it was the piranhas, maybe.

Phil - It is piranhas!

Louise - It is piranhas?

Phil - It was piranhas!

Louise - OK, wow. That was a brave person that took that sample.

Phil - Did you guess right? Our mystery fish is indeed a piranha! Now Louise got piranhas as one of the results, but it was far from top of the list. What’s going on - well, the water in the aquarium tanks isn’t separated from each other, it’s actually filtered through a number of them. So i guess we didn’t make Louise identify a piranha from a piranha tank - so much as identify a piranha from a whole aquarium filled with different fish.

Louise - It would be interesting to know if there are cichlids within the same aquarium, and whether we’re just picking that up from a different tank.

Phil - There are, in quite a few of them.

Louise - Right. That might make sense then. There was also some other interesting DNA sequences that we picked up. Some mouse, or rodent DNA in there...

Phil - You should have all the pieces to the puzzle by now.

Louise - Oh, is it part of the fish food?

Phil - Oh yeah.

Louise - OK, right, that makes sense.

Here’s the audio that was hidden earlier - Phil Sansom learning about piranhas from Tony Sapsford at Frog End Aquatics...

Tony - Do you want me to say what they are?

Phil - Yeah, please.

Tony - Okay, this is the piranha tank. They've got red underbellies, red tails, large eyes. I mean they are only babies have only just got their teeth, so they’re only about 4 inches long at the moment, maybe 5 inches.

Phil - You can barely see them.

Tony - You can barely see the teeth.

Phil - Oh my God, you're putting your hand in.

Tony - It's not gonna hurt me trust me.

Phil - Are you sure?

Tony - Yeah.

Phil - Are they not aggressive?

Tony - Only in like a pack, when they're feeding and stuff like that. Like if he had a sick one in there, they’d annihilate it.

Phil - Am I right to be scared of them, or should  I not fear them?

Tony - No. No, it's nothing to be scared of. Unless you're gonna jump into a lake in the Amazon with a whole pack of them, I don't think there’s anything to worry about. Most people who buy fish like that think ‘oh it's great. I’ve got to have a piranha’, but then they get bored of it very quickly because they don't do anything.

Phil - Do people buy them because they think they're being hard?

Tony - Yeah, basically that's it. ‘Oh I got one of them, it’s a piranha’. And all they do is get big and boring.

You can read the full results from the fish tank water here.

Bottlenose dolphin

19:08 - Whales and dolphins: the lost genes

Tens of millions of years ago whales and dolphins evolved into the oceans. They lost genes in the process...

Whales and dolphins: the lost genes
Michael Hiller, Max Planck Institute of Molecular Cell Biology and Genetics

Tens of millions of years ago cetaceans - whales and dolphins - evolved back into the oceans. To survive, they needed complex new features like fins and blowholes, as well as the genes to make them. But it seems they also lost certain genes as well. 85, in fact, according to a new study by German and American scientists. Producer Mariana Marasoiu heard more from Micheal Hiller...

Mariana - When you hear of whales and dolphins, what comes to mind? Perhaps smooth shiny animals in the depths of the sea, surrounded by dark blue-green water with faint shimmering from the light far above. Sometimes jumping out of the water for a breath of air, or just for a little play.

Wait. But whales and dolphins are mammals! Millions of years ago, they used to live on land. Their ancestors probably looked like wolves on hooves. How did they end up living underwater? Michael Hiller and his colleagues from the Max Planck Institute of Molecular Cell Biology and Genetics investigated a certain type of genetic change that may have helped whales and dolphins in this transition from land  to water.

Michael - We pretty much used the genomes of living whales and dolphins, and we search for genes that exhibit the same mutation in these species. If you have exactly the same mutation in the genomes of different mammals that are related, this tells you that this mutation occurred already in the ancestors of these species. And with this we could single out 85 genes that were lost during this transition from land to water.

Mariana - Why would we want to lose a gene?

Michael - I think there are 2 main principles that could explain when genes actually get lost. One is - and this probably explains the majority of the gene losses that we found in this study - that the function of the gene is no longer required. And then there is no selection pressure to maintain the gene, and it will eventually accumulate mutations that destroy its function and then we would call the gene ‘lost’. In the second, potentially more interesting, although probably also much rarer, mechanism is: that the loss of the genes is actually a benefit. We think we found in this study a few cases where the loss of genes could have been beneficial for the ancestors of whales and dolphins when they actually transitioned from land to water.

Mariana - One example that they found that seems to have been very useful in this transition is the elimination of a protein that repairs DNA damage. Because DNA is basically just a chemical entity, it’s subject to assault from the environment and interaction with other chemical molecules which can damage it. DNA damage happens all the time, but the body has repair proteins that walk over the DNA molecule and try to fix any damage that occurs. Isn’t it a bad thing that whales lost one of these proteins? Well, it turns out that these proteins are sometimes a bit faulty, and some of them are more accurate at repair than others.

Michael - We found that whales and dolphins have lost the protein that’s the most sloppy. What is interesting is that the ability to repair DNA damage is actually increased, so the fidelity is higher. And the mechanism behind why losing this protein then enhances the fidelity of DNA repair is most likely that the other proteins that are more accurate actually compensate for the loss.

Mariana - Having an increased ability to repair DNA damage may explain how whales and dolphins are able to survive for so long under water, as DNA suffers an increased risk of damage during the deep diving and surfacing cycle from oxidative stress,when there is an imbalance between reactive oxygen species and antioxidants. And this is not the only gene loss-based adaptation that seems to help cetaceans thrive. Some of the genes that help with blood clotting were also inactivated, which suggests a reduced risk of blood clots forming inside blood vessels as they get more and more compressed during those deep diving periods. The team may have also found a gene loss that explains how whales and dolphins are able to sleep underwater without drowning.

Michael - The loss of the melatonin synthesising gene, so melatonin being the sleep regulating hormone, is potentially a link or association with a particular form of sleep that cetaceans - whales and dolphins - have: and this is they only sleep with one brain hemisphere at a time while the other brain hemisphere is awake, and likely coordinates movements and coming back to the surface for breathing.

Mariana - While the researchers looked in this study at only one type of change in the genome, the loss of genes, there are many other types of genetic mutations and changes that happened.

Michael - There is definitely much more to learn and our understanding of the changes in the DNA that are required to turn a land animal into an aquatic animal is at the moment quite rudimentary, so hopefully we can make progress on that in the future.


Using nanopore sequencing would have given you these results within hours!

They were generated within hours!

Add a comment