Three times a week, on a quiet block of Stock Island on the Florida Keys, 25,000 of the world's deadliest creatures are released into the wild.
Not to fear. While these are mosquitoes — the creature responsible for more human deaths each year than any other on the planet — they're all males: It's only the females that bite.
And if they do their jobs, these male mosquitoes will contribute to a significant suppression of the female bugs that do bite—and, more importantly, spread disease.
"This is a very robust mosquito that is causing crazy diseases that are very impactful on people's lives," said Andrea Leal, executive director of the Florida Keys Mosquito Control District. "Aedes aegypti is the most difficult mosquito to control."
The Aedes aegypti mosquito is responsible for spreading diseases like dengue fever, chikungunya, yellow fever, and, most recently, Zika. It's not the deadliest of all the mosquitoes—that title goes to the species that carry malaria—but it's causing the most grief here in the United States.
Through traditional methods, including aerial and on-the-ground spraying of pesticides and very intense vigilance, the Florida Keys Mosquito Control District has been able to reduce the population of Aedes aegypti by 60 percent or more, Leal said.
To do that, the district spends about 10 percent of its $10 million annual budget on that one mosquito. But, Leal said, that level of suppression isn't enough; they're aiming for 90 percent.
"One mosquito out there could be the means of transmitting diseases to multiple people," she said.
That's where those 25,000 mosquitoes come in. The Keys is running a trial to see if it can reduce the population of Aedes aegypti mosquitoes through what may seem like a counterintuitive method: by releasing more Aedes aegypti mosquitoes.
These experimental bugs carry a secret weapon: they're infected with a bacteria called wolbachia that interferes with reproduction, essentially aiming to spread infertility through Aedes aegypti in the area.
"Wolbachia [is] found in many insect species," explained Linda O'Connor, a senior scientist at MosquitoMate, the Kentucky-based developer of wolbachia-mosquitoes that's partnered with the Keys and others to run the trials. "It's found in the reproductive system and it helps them produce eggs. It's a symbiotic bacteria in insects."
But it's not found naturally in Aedes aegypti. So when these males infected with wolbachia are released into the wild, they mate with uninfected females, and the resulting eggs don't hatch.
The test in the Keys is relatively small; 25,000 male mosquitoes released three times a week in a 10-acre area for 16 weeks.
Three thousand miles away, in Fresno, California, a bigger trial just started, with the help of one of the world's largest technology companies.
When Alphabet's Google says it's working on a project called Debug, one could logically assume it's referring to software.
But it's actually working on real bugs: the Aedes aegypti mosquito.
"There are thousands of different mosquitoes, but just one is responsible for spreading almost all of the cases of those four diseases that sicken hundreds of millions of people a year," said Linus Upson, vice president of engineering at Google's health subsidiary Verily, and leader of the Debug project.
Previously the overseer of Google's browser products including Chrome, Upson may not seem the most obvious person to lead a mosquito trial. But he said mosquitoes have been a problem that have stuck with him since he did an undergraduate internship at the National Institutes of Health years ago.
"While working there I learned that mosquitoes are the number one cause of human misery and death for otherwise healthy people of anything else on the planet," Upson said. "I've been fascinated by the problem ever since."
So, he turned from browsers to bugs.
"It's exciting to be able to use a lot of the technology and engineering expertise that we have at Verily to really see if we can solve this problem," Upson said.
Specifically: Rearing millions of mosquitoes at a relatively low cost, and sorting them efficiently into males and females. Because it's only the females that bite, the goal is to release only males into the wild.
This avoids increasing the number of disease-spreading mosquitoes, but also solves a more banal problem, according to Stephen Dobson, the University of Kentucky entomologist who founded MosquitoMate: People are much more likely to let you release thousands of mosquitoes in their yard if the bugs are not going to bite them.
"We've developed automation that can automatically rear mosquitoes, separate the males from the females, and then release them in the field," Upson said. "We've lowered the cost of being able to do this, tremendously."
Among the technology Verily's applying to the mosquito-sorting problem: Computer vision algorithms.
"We actually take a picture of every single mosquito that we're going to release," Upson said. Computers then determine whether the mosquito is male or female, and only release the males.
The technology is still in the validation phase, Upson said. He personally had spent the previous weekend reviewing more than 30,000 mosquito images to make sure no females were being released.
"But ultimately," he explained, "the computers can do this much better than humans can."
Descriptions of the Verily/MosquitoMate approach scrupulously point out the wolbachia mosquitoes are not genetically modified organisms. That may be because of some local pushback to that method in the Florida Keys.
Oxitec, a British biotechnology company recently acquired by synthetic biology firm Intrexon, has developed a genetically modified mosquito that works similarly to wolbachia: Modified male mosquitoes are released to mate with wild females, and the resulting offspring are unable to survive.
In field trials in Brazil, the Cayman Islands and elsewhere, Oxitec says it's been able to suppress Aedes aegypti populations by more than 90 percent.
It, too, had planned to conduct a trial in the Florida Keys. But a countywide vote on the general election ballot in November presented a speed bump: of 33 districts, 31 voted in favor of a GMO mosquito trial. One of the two districts that didn't was Key Haven, the area where the trial was planned.
Oxitec's Derric Nimmo said the company is now waiting on Food and Drug Administration clearance to conduct a trial in a different area of the Keys, hoping to start this year. The FDA last year released an environmental assessment declaring the Oxitec trial was unlikely to have a significant impact on the environment.
The GMO and wolbachia approaches are on separate regulatory paths; while Oxitec's GMO mosquitoes are currently being overseen by the FDA, MosquitoMate's wolbachia bugs are under Environmental Protection Agency oversight, being characterized as a microbial pesticide.
Nimmo said the FDA may consider transferring oversight of the Oxitec mosquitoes to the EPA. But the public pushback against the GMO approach, however small, has had an impact.
"We looked at all of the different options," Verily's Upson said. "The wolbachia approach was very attractive from a regulatory and public acceptance standpoint ... It was just the easiest way for us to get started."
Despite the difference in methods, the GMO and wolbachia approaches have the same end goal: Reducing the Aedes aegypti population in a targeted way. Neither technology persists in the environment. If modified male mosquitoes are no longer released, eventually it's likely the wild mosquitoes would repopulate.
Technology theoretically exists to try to wipe out populations more broadly, though, through genetic modifications that would get carried through generations.
"With the gene drive system, we could simply spread infertility," explained Kevin Esvelt, an assistant professor at the MIT Media Lab who runs a group called Sculpting Evolution.
A gene drive is very powerful: Instead of leaving inheritance up to chance, a gene drive system ensures that particular DNA is passed along.
"It spreads indefinitely," Esvelt said.
Considering the toll mosquitoes take on humanity, one might ask: Why don't we kill them all?
Unfortunately, nobody in the scientific community seems to think this is a good idea.
"The history of mankind tells you mosquitoes are bad news," said Dr. Anthony Fauci, director of the National Institutes of Allergy and Infectious Diseases.
Still, he said, "gene drive is a double-edged sword: it could possibly be the most effective way to control mosquitoes on the one hand; on the other hand, if there are unintended consequences, it may be a bit more difficult to turn the process off."
Esvelt points out it's not actually the mosquitoes we want to kill, it's the diseases they carry.
"Our real enemies — the things we really do want to drive extinct — are viruses and parasites," Esvelt said.
"We don't necessarily want the mosquitoes that carry it gone," he added. "So if we can suppress them down to levels where they can no longer transmit the virus and we can remove all of the other places where it's present, then we can eradicate the disease without permanently harming the mosquito."
For now, we may be stuck with the bugs. But if new methods are successful, perhaps one day they'll return to being itchy pests, and relinquish their title of world's most deadly animal.