Back in 2010, Chris Stone was spending a lot of nights under a mosquito net in a greenhouse in Ohio. Loose in the greenhouse’s artificially tropical air were many mosquitoes, and Stone, a graduate student, was there to serve as bait in a series of experiments. And while he was there, he was doing some hard thinking.
In the last ten years or so, more than a billion insecticide-treated bed nets have been distributed across the world in regions where malaria is endemic. Spread by mosquitoes, malaria is a gruesome disease; in 2003, the World Health Organization reported it killed a million children a year, 90% of whom were under the age of five. But recently the bed nets, along with insecticide sprayed inside houses, have had dramatic effects. As a rule, many major malaria mosquitoes like to bite at night, inside homes. If they land on the nets, tantalized by sleepers’ CO2-laden breath, or on the home’s treated walls, they die swiftly. In areas where substantial numbers of people sleep under mosquito nets, malaria rates have plummeted.
Lying there under the mosquito net in Ohio, though, Stone was preoccupied with another idea. If mosquitoes land on treated nets, they’ll die. Sure. But what about the ones that don’t land? What if they start having their meal before people go to bed or can wait until morning? What if they start biting outside instead? “Sitting in that room, I had a lot of time to think about it,” he says.
Indeed, this potential problem, known as behavioral resistance, has been on some malaria biologists’ minds for a while. These days, it’s clear that overusing antibiotics, for instance, has caused the evolution of resistant bacteria, superbugs that stand up to much of what we throw at them. Insecticide resistance, similarly, is a problem when it comes to controlling mosquitoes. In both cases, if we kill all the individuals who are susceptible, the ones who are left over—the ones who by some quirk of biology or genetics can withstand our weapons—will become the parents of future generations.
But can something as simple as a treated net change the way mosquitoes actually behave, undermining our efforts? The answer may be yes.
An Ominous Insight
Behavioral resistance began to crop up in the scientific literature on malaria decades ago. The World Health Organization, back in 1955, launched a campaign called the Global Malaria Eradication Programme. In many places where malaria was endemic, the insecticide DDT was sprayed in homes. Malaria cases dropped, but eventually, many of the mosquito species that spread the disease evolved resistance. The individuals who could survive DDT took over the population. Other complications in the form of wars, lapses in funding, and logistical problems sent malaria cases climbing again. The program was eventually abandoned.
But before it was, in 1975, biologist Brian Taylor wrote that he had observed something peculiar in the Solomon Islands in southeast Asia. Every six months through most of the 1960s, as part of the program, DDT had been applied to the interior walls of homes and porches, even on the undersides of houses built on stilts—anywhere a mosquito might land to rest and digest in a human home. Two of the species that spread malaria there either disappeared or were greatly reduced under the force of the assault. One called Anopheles farauti, however, stuck around. And now, some said, it behaved like a different mosquito. Before it had fed indoors—now it seemed it was much more likely to be found feeding outdoors. Taylor organized an experiment: For nights on end, people indoors and out collected the mosquitoes that landed on them. After an application of DDT, they did the same.
Before the treatment, people caught mosquitoes all throughout the night. They also caught them in substantial numbers inside as the night wore on. Afterward the treatment, however, the mosquitoes were far, far more active in the early evening. Both indoors and out, their numbers went down as the hours passed. Taylor suggested that the DDT was killing off the mosquitoes that went indoors during the night to feed, leaving behind those which bit people before they retreated to the protection of their homes. It was a fascinating, if ominous, insight.
As the years have passed, quashing malaria’s spread has come back on the global agenda. The latest chapter, in the early part of this century, has been the introduction of treated bed nets. The nets don’t just kill mosquitoes on contact, they also provide a physical barrier at night when people are most vulnerable. They last a long time—each net offers protection for at least three years, even with regular washing. However, researchers are now documenting in a steady tide of papers that mosquitoes are evolving resistance to the nets’ insecticides, just as they did to DDT.
By this point, resistance isn’t surprising. This is just how evolution works: As in physics, each action has an equal and opposite reaction. Push a pendulum, watch it swing. Kill all the mosquitoes that can’t take an insecticide, watch their surviving brethren multiply. It is possible to get ahead of resistance, though, if we can apply pressure from enough directions. Understanding how resistance arises, and exactly what form it takes, is required to help plan our counterattacks.
There is plenty of research on how mosquitoes are escaping the effects of insecticides, physiologically speaking. There are several ways they have adapted: they make more enzymes to break down the poison before it can hurt them, they alter the proteins it targets so they can become immune, and so on. But there is less work about the effect Stone was thinking about in Ohio and which Taylor observed more than a quarter century before—the problem of behavior. In the last few years, a handful of papers have begun to approach it with field experiments, literature reviews, and mathematical models.
Stone, after his late-night brainstorming sessions as mosquito bait, put together a model with colleagues that looked at the situations in which behavioral resistance was likely to arise. They focused on the size and habits of the human population, as well as how hard or easy it would be for mosquitoes to sustain themselves between meals of blood by sipping nectar (many mosquitoes do get calories from nectar, though they prefer blood). They found that in situations where there were plenty of people out and about at dusk and dawn, as in a city, it’s not very hard for the mosquitoes to find humans who aren’t under bed nets to bite. In that situation, behavioral resistance by feeding at these times was more likely to develop. “That model that I created suggested there might be something to that,” he says. “But,” he cautions, “there’s no field evidence to support that at this point.”
In a 2013 review article in Malaria Journal, referring to a number of older observations in the South Pacific, researchers primarily based in Australia called for greater awareness of the potential danger of behavioral changes in the age of widespread bed nets. They note that after the application of DDT, changes similar to what Brian Taylor saw back in the 1970s were seen in Papua New Guinea, Vanatu, and other islands. Crucially, in the Solomon Islands, after DDT was no longer used as much, the mosquitoes kept their early evening, outdoor feeding behavior, which continues today. “We were very surprised that it didn’t go back to all-night,” says Nigel Beebe, a mosquito biologist at University of Queensland and the CSIRO who is an author of the paper. That suggested that it really was being passed on from parent to offspring, a genetic reshaping of the species’ habits, rather than being a short-term shift. “It’s difficult to know all the answers, but these are just the observations,” Beebe continues. “And when the bed nets were rolled out into these areas, it seemed to strengthen this early night biting profile,” at least in some places.
In Tanzania, insecticide-treated bed nets there appear to have shifted the behavior of Anopheles funestus, another malaria vector, so that it bites more during the day, according to work by Tanya Russell of James Cook University, one of the authors of that review paper, and colleagues. In Senegal, another group of researchers report collecting enormous numbers of An. funestus biting in broad daylight—though it’s been known as a night-feeding species—and suggest bed nets may be the driver. The authors warn malaria rates’ precipitous declines could be threatened by these changes to behavior.
Still, so far, the evidence for behavioral change is not quite as simple as it might seem. The same mosquito species that show a change in response to nets in one place show no change in another. It isn’t always clear whether the before and after conditions are directly comparable, cautions one meticulous review of the subject in the journal Evolution, and there is so much geographical and seasonal variation in mosquito numbers that it’s challenging to nail down whether the effect you’re seeing is from the bed nets or indoor insecticides and not from something else. As well, if the insecticides on the net act as repellents—if they are unpleasant for mosquitoes to be near, rather than merely lethal to those that touch it—the change in behavior won’t be genetic, but just a reaction. “Those things are all the complexities that need to be unraveled,” says Fred Gould, an evolutionary biologist at North Carolina State University and an author of the Evolution review. “And in general, studying behavior is hard. With behavior, you have to be pretty much of a sleuth to really get in there and figure out what’s going on.”
All the same, if behavioral resistance does occur, it could undermine bed nets’ success, Gould says. Entomologists in agriculture know that heritable behavior changes can be just as unpleasant as resistance to insecticides, he says. The corn rootworm, for example, is a pest that lays eggs in cornfields so its larvae will come up the next year and feast on the roots. Farmers have evaded it by rotating crops so what was a cornfield one year will be soybeans the next. By the 1990s, however, the strategy wasn’t working as well. It turned out that the rootworms had changed—instead of hatching every year, one species was hatching every other year, to be there when the corn returned. Another species was leaping into neighboring soybean fields to wait for them to take their turn as cornfields in the next season.
One Step Ahead
Those examples serve as a reminder that, when it comes to battling pests, whether they are corn rootworms or malaria mosquitoes, we can’t rest on our laurels. It pays to have a multi-pronged approach. If mosquitoes start biting outside more or earlier in the day, we’ll need new defenses to meet them on their new ground. One option, says Tom Burkot, a vector biologist with the Australian Institute of Tropical Health and James Cook University who does field work in the Solomon Islands, is exploiting the fact that mosquitoes will often have nectar meals if they can’t find people.
“If there’s a sugar source nearby, they may top up their energy with a little sip of sugar,” he says. Sugar traps with an oral toxin—one that only affects mosquitoes, not bees, for instance—could help drive their numbers down even if the insecticides on the nets and in the houses don’t work as well as before. Several other options are also on the table, including treatments that get at mosquito larvae, new building codes to make houses more mosquito proof, and insecticides that kill mosquitoes at different phases of their life cycles to avoid the evolution of resistance.
Even if mosquitoes do change their behavior, bed nets will continue to impede the spread of malaria to at least some degree, Burkot says. The reductions in transmission are nothing short of astonishing—25%, 50%, even 80% fewer cases in some places. And that’s with less than ideal use of bed nets. “We’re using these imperfect tools, applied imperfectly,” he says. Some people don’t sleep under them, or if they do, not every night. Distributing them can be a challenge, and the nets can tear. “Yet despite all these significant challenges, we’ve been able to have a hell of an impact on malaria.”
The problem, of course, is that we have to stay at least one step ahead of evolution. Coming at it with many different tools is our best to bet to help keep resistance, whether it is physiological or behavioral, under control. “The way forward is we’ve got to optimize these interventions that are working,” Burkot says. “We have to extend their life.”