Thanks to Brilliant for supporting  this episode of SciShow.

Go to Brilliant.org/SciShow to learn how you can take your STEM skills to the next level! [♪ INTRO]. It was a mystery that went  unsolved for more than 25 years..

Across the southeastern United  States, bald eagles were dying. This time, it wasn’t DDT.  But no one knew what it was. Finally, in March of 2021, researchers  announced that they had found the killer: a toxin from a recently  discovered species of bacteria.

And the scientists hope that this  knowledge can better protect eagles and the ecosystems they depend on. But how these scientists uncovered this  mystery is better than any murder TV show that you’ve ever seen, so get  ready for some twists and turns. It all begins in 1994 near a reservoir  in Arkansas called DeGray Lake.

Soon after the eagles had  arrived to spend the winter,   they started turning up dead. Two years later, it happened again. This time, another overwintering  species was affected as well.

Some duck-like birds called coots,  a prey item of eagles, by the way, were having trouble swimming and flying. In necropsies, the non-human  version of an autopsy, the bodies of the affected  birds looked mostly healthy. But in their brains, the  white matter looked spongy.

The axons, the skinny tubes that carry messages between brain cells, were swollen. And their fatty myelin wrappers were riddled  with microscopic, fluid-filled bubbles. Researchers ruled out some possibilities, like prions, or misfolded proteins, and viruses.

And by 1998 they classified  it as a new brain disease:. Avian Vacuolar Myelinopathy, or AVM. No one had ever seen anything  like it in wild birds.

But it did look a lot like the damage  in mammals from certain neurotoxins. Like, from the rat poison bromethalin. Bromethalin works by blocking mitochondria from producing energy in the form of ATP.

Neurons use a lot of ATP to set up their signals. And when this breaks down,  they sort of fall apart.   Trouble was, testing for these and  other toxins turned up nothing.   Meanwhile, the problem was growing. AVM began showing up across the southeastern US.

Over eight winters, it killed at least a  hundred bald eagles and thousands of coots, and it sickened members of  several other bird species. There was growing evidence that  the cause was a natural toxin not a chemical that humans had  dumped into the environment. But despite years of trying,  researchers couldn’t identify it.   Then, in 2005, we have our first breakthrough.

Scientists linked the poisonings  to artificial bodies of water where a certain invasive weed was growing. And where a certain type of  cyanobacteria was growing on the weed. The nuisance weed is called  Hydrilla — or waterthyme.

And these lakes had a lot of it. Hydrilla also grew in similar  lakes without these cyanobacteria. And even though the same types  of birds spent time there, they were not getting sick.

So all signs were pointing to the bacteria. No one had seen this species before,  but others like it are notorious for making potent toxins, including  ones that affect the brain. You may have heard about toxic  algal blooms that poison wildlife and close beaches?

Well, that is actually cyanobacteria, sometimes called blue-green  algae, and they can be nasty. Researchers suspected these new  cyanobacteria were making a toxin that accumulated as it moved up the food chain. This is sort of like what happens  with other toxins like mercury or DDT, the pesticide infamous for  thinning bird eggshells.

Like, waterbirds eat the weeds. Then eagles eat the waterbirds. Especially the sick ones,  because they’re easy to catch.

Since eagles are at the top of the food chain, they likely get the highest dose of toxin. Which may have made AVM  especially deadly for them. But it affected other critters in these lakes too, like snails, tadpoles, fish, and turtles.   Problem was, when researchers  grew the cyanobacteria in the lab, itself, turns out, no easy task  — they didn’t make any toxin.

So even though there was all this  evidence that they were the killer, it was all circumstantial evidence. Now to be clear, it’s not that the  scientists were missing something. The bacteria seem to produce very low  levels of the toxin most of the time, and they need a secret ingredient to do it.

The final clue fell into place when  researchers used a super fancy tool called atmospheric-pressure  matrix-assisted laser desorption/ionization mass spectrometry imaging. ...is what it was called. What is it? Well, it uses a laser to vaporize the  molecules on the surface of an object.

In this case, the objects in  question were bacteria-covered weeds gathered from the wild. Then it weighs and analyzes the molecules. What stood out was a molecule  with five bromine atoms.

Sort of like the ones on  bromethalin, the rat poison. Now, bromine isn’t all that  plentiful in living things. And it is not something you’d necessarily  think to add to culture medium if you were trying to grow cyanobacteria in a lab.

But when the researchers fed  the bacteria some bromine in the form of potassium  bromide, they made the toxin! They named it AETX — short for aetokthonotoxin, which means “poison that  kills the eagle” in Greek. Though, as it turns out, it can  kill plenty of other animals too.

And with that, the mystery was solved. They also made some important connections to what was happening in the reservoirs. Like, when bacteria are stressed, they  crank out a hundred times more toxin.

And they get stressed by things  like water temperatures dropping and the water being stirred up. Researchers think those same  conditions precede bird poisonings. Plus, it turns out Hydrilla accumulates  bromine from the environment.

So the cyanobacteria probably  suck bromine out of the weed. And now we know we should  probably stop using a certain bromine-containing herbicide  to try to control the Hydrilla, because it could be feeding  bromine straight to the bacteria! We can also monitor bromine in the lakes and look for ways to make it  less available to cyanobacteria.

And finally, we can look for better ways  to combat Hydrilla, the invasive weed, and the toxic bacteria. Meanwhile, by studying how the toxin works, we might even be able to find a treatment. Solving this mystery should help  us protect not only bald eagles, but these entire freshwater ecosystems  and the people living near them.

Decades of detective work have finally paid off. Scientists have to be  comfortable with the unknown, they tackle it all the time. And Brilliant’s course Knowledge and Uncertainty can teach you not just how  researchers understand uncertainty, but also how you can deal  with it in your own life.

Brilliant is an online learning platform  that offers tons of interactive courses in science, math, engineering, and computer  science, all designed by top educators. If you’d like to give them a try, you  can check out Brilliant.org/SciShow to sign up and save 20% on an  annual Premium subscription. [♪ OUTRO].