A cyanobacterial neurotoxin produced by blue-green algae is responsible for die-offs of bald eagles and other birds dating back to the 1990s, scientists reported this week.

The toxin causes lesions to form in the brain and is only produced when the algae come into contact with a form of bromine called bromide, which is introduced to watersheds via herbicides and other human-related activities, the researchers reported March 25 in Science.

"The fact that we are seeing recovery of bald eagles even in states where this does occur means that probably the population as a whole isn't being affected," said Susan Wilde, an associate professor of aquatic science at the University of Georgia and last author of the paper. "My concern is we don't want it to spread to more locations; then, I think it would be more significant."

The first sign of trouble came in the winter of 1994-95, when bald eagles began dying around DeGray Lake in Arkansas. Over the next two years, more than 70 dead eagles were found. 

When researchers examined the brains of the dead birds, they found that the fatty myelin sheaths of their neurons had swelled and split apart. Scientists suspected that the disease, called vacuolar myelinopathy, was caused by exposure to some kind of toxin. 

Vacuolar myelinopathy has since spread across bird populations in the southeastern United States, and causes neurological symptoms such as weakness and impaired coordination.  

During an epidemic in the late 1990s and early 2000s, scientists released mallards and American coots into North Carolina's Lake Surf and saw the birds sicken within a week. 

"They weren't able to fly and when they came up from underwater they couldn't right themselves," Wilde said. "[The scientists] realized that whatever was occurring ... was something that could happen pretty rapidly." 

To investigate how the birds were becoming sick, Wilde and her colleagues visited lakes where birds had fallen ill. Their water and sediment samples turned up clean, but the team did realize that an invasive plant called Hydrilla verticillata, or waterthyme, was growing at all the sites. 

They discovered a new species of cyanobacteria, which they named Aetokthonos hydrillicola, growing on up to 95% of the plant leaves.

Hydrilla grows most densely in late autumn, at the same time that migrating waterfowl arrive. The famished birds gobble up the Hydrilla leaves. 

"That is when we see the sickest birds," Wilde said. The coots are targeted in turn by the eagles.

After identifying the cyanobacteria growing on the undersides of the leaves, the researchers found that eating algae-infested Hydrilla leaves — or another animal that had grazed on the leaves — caused vacuolar myelinopathy to develop in waterfowl, birds of prey, reptiles, amphibians and fish. 

At this point, Wilde and her team suspected that a toxin produced by the cyanobacteria was the culprit. Puzzlingly, when they grew the algae in the lab and fed it to chickens, the birds remained healthy. 

However, when the team analyzed wild cyanobacteria colonies, they identified a molecule that contained five bromine atoms. When the researchers added potassium bromide to their cyanobacteria cultures, the algae produced the molecule, which caused vacuolar myelinopathy when fed to chickens.

The researchers named the molecule aetokthonotoxin, and speculate that its function may be to deter herbivores or algal competitors. It's fat-soluble, which means that it can build up in the tissues of the animals that eat it. 

The researchers were able to detect aetokthonotoxin in the liver, breast and thighs of two deceased wild coots. Furthermore, they only observed aetokthonotoxin at lakes where cases of vacuolar myelinopathy had been documented. 

Bromide is found in fumigants, herbicides (including those used to combat Hydrilla infestations), flame retardants and water treatment plants, and is also used to reduce mercury emissions from coal-fired power plants. It occurs naturally in the Earth's crust and seawater. The researchers discovered that Hydrilla can accumulate much higher concentrations of bromide than the surrounding water or sediments.

Wilde and her team found that aetokthonotoxin concentrations peaked in November. At this point in the fall, Hydrilla plants are beginning to break down, perhaps providing the cyanobacteria with a ready supply of bromide. The researchers suspect that certain environmental characteristics such as temperature and seasonal mixing of water layers may also be important.

"We're beginning to understand the water quality and sediment parameters that might encourage not only the algae to grow, but also allow it to make that toxin molecule," Wilde said.

There haven't been any known cases of humans developing the disease. Still, it's imperative to determine how vulnerable mammals are to aetokthonotoxin and the potential health risks to people from consuming diseased fish and waterbirds, Wilde and her colleagues wrote in the paper.

"A toxin produced by cyanobacteria that colonize a highly invasive plant, which has the capacity to affect diverse animal phyla, should not be underestimated in its potential impact on our environment," they concluded. 

Meanwhile, there are ways to limit the spread of vacuolar myelinopathy, Wilde says. To prevent Hydrilla from being transported to new reservoirs, people should thoroughly wash their boats when moving between sites. Another strategy is to control dense infestations of aquatic plants using fish rather than bromide-containing chemicals. 

In recent years, sterile grass carp have been deployed to eliminate Hydrilla from sites where bald eagles stricken with vacuolar myelinopathy have been recorded since the 1990s, including Arkansas's DeGray Lake and the J. Strom Thurmond Reservoir in Georgia and South Carolina. The latter introduced grass carp in 2017. No eagle deaths from vacuolar myelinopathy have been reported there since 2018.

The study, "Hunting the eagle killer: A cyanobacterial neurotoxin causes vacuolar myelinopathy," published March 25 in Science, was authored by Steffen Breinlinger, Christoph Wagner and Timo H. J. Niedermeyer, Martin Luther University Halle-Wittenberg; Tabitha J. Phillips, Brigette N. Haram, Susan M. Williams, James D. Lauderdale, H. Dayton Wilde, Wesley Gerrin and Susan B. Wilde, University of Georgia; Jan Mareš, Biology Centre of the Czech Academy of Sciences, Institute of Microbiology of the Czech Academy of Sciences and University of South Bohemia; José A. Martínez Yerena, Biology Centre of the Czech Academy of Sciences and University of South Bohemia; Pavel Hrouzek and Roman Sobotka, Institute of Microbiology of the Czech Academy of Sciences and University of South Bohemia; Andreja Kust, Biology Centre of the Czech Academy of Sciences; W. Matthew Henderson and John W. Washington, U.S. Environmental Protection Agency; Peter Schmieder, Leibniz-Forschungsinstitut für Molekulare Pharmakologie; Benedikt Geier and Manuel Liebeke, Max Planck Institute for Marine Microbiology; and Heike Enke, Cyano Biotech GmbH.