Two odorants, called decanal and undecanal, are reminiscent of orange peel and are perhaps best known for their use in Chanel No°5 perfume. They are also the compounds that help mosquitoes home in on humans, a new study finds. 

The research, published on May 4 in Nature, now shows that when female Aedes aegypti mosquitoes smell humans, a unique pathway—the “human glomerulus”—activates in their tiny brains. The glomerulus responds particularly to decanal and undecanal, which are volatile components of human sebum, an oily substance produced in the sebaceous glands dotted throughout our skin. 

Ae. aegypti spread diseases such as dengue, Zika, and yellow fever. Given the choice, the insects prefer the smell of humans over other animals. “They have some ability to differentiate between a human host and animal hosts, but which odors and neuronal pathways are involved has not been known,” says Marcus Stensmyr, a biologist at Lund University in Sweden who was not involved in the studyThe authors have “managed to pinpoint a specific circuit that is detecting a human volatile, and this circuit is not activated when you present mosquitoes with the odor of animals,” he adds. “The paper is a technical tour-de-force.” 

Designing a smell-o-meter

Coauthor Carolyn McBride, a professor of ecology, evolutionary biology, and neuroscience at Princeton University, explains that the research team started by asking, “what’s different about human odor that makes this mosquito more interested in us than animals?” Using CRISPR gene editing, the team generated mosquitoes that express the calcium sensor GCaMP6f in olfactory sensory neurons. When neurons fire and calcium levels increase, the sensor is activated and fluoresces—something that was previously only possible in model organisms such as mice. “This was the first opportunity where we could actually ask the mosquito, ‘What do you sense now?’” McBride says.

black bag with tubing attached laid out on a cot
A bag used to collect odors from human volunteers
© McBride, Zhao

The researchers then exposed the engineered mosquitoes to human, rat, and sheep odors, using a newly developed method that Christopher Potter, a neuroscientist at Johns Hopkins University School of Medicine who was not involved in the study, describes as an “impressive advancement.” In the set-up, human volunteers who hadn’t showered for a few days lay down in a Teflon bag. The air from the bags was collected and puffed at the mosquitoes through several small pipes. As animal comparisons, live animals were put in a glass chamber, or hair, fur, and wool placed into a bottle, through which air was pumped toward the mosquitoes. “One key innovation was developing an odor-delivery system by which they could precisely and reproducibly capture and then release complex odors from humans and animals to the mosquito,” Potter writes in an email. “Using their new odor-delivery system, they could precisely stimulate the mosquito with [the odor concentrations they’d be expected to encounter in real life].” 

Human and animal odors share many of the same compounds. But some molecules are more abundant in human odor than in animal odor. Previously, it was though that complex odors such as the smell mixture given off by humans would activate many different olfactory processing centers, or glomeruli, in the mosquito brain while animal odor would activate another, partly overlapping, set of glomeruli. “We figured the only way to tell apart [humans and animals] would be some complicated pattern of activity where some neurons were more strongly activated, and these others were less strongly activated,” McBride says, explaining that she anticipated at least five different populations of neurons would be involved.   

To the authors’ surprise, imaging of mosquito olfactory neurons as they smelled either humans, rats, or sheep showed a much simpler picture; in response to either nonhuman animal species, two glomeruli were activated. In response to human odor, one human-specific glomerulus and one of the glomeruli also activated by animals lit up. “It kind of blew us away because we didn’t really think it was going to be that simple,” McBride says. 

“This paper makes the surprising discovery that, even though the odors are complicated mixtures, these mosquitoes can use a simple neural code to differentiate their preferred hosts (us) from other animals,” Laura Duvall, a biologist at Columbia University who was not involved in the study, tells The Scientist in an email.

Attractive aldehydes

Next, the researchers dug deeper to identify which compounds the human-specific glomerulus of mosquitoes responds toOne group of compounds that stood out were aldehydes, which are abundant not only in human and animal odors, but also in smells wafting from plants and soil. The authors determined that animal odors were abundant in shorter-chain aldehydes, while humans had more long-chain aldehydes, McBride says.  

The long-chain carbon aldehydes decanal and undecanal turned out to strongly activate mosquitos’ human-specific glomerulus at physiological concentrations, while the glomerulus tuned to both humans and animals responded to a range of compounds. 

purple mosquito brain with clusters of odor-detecting cells lit up in green
Odor receptors imaged in Aedes aegypti
© Zhilei Zhao

The researchers also found that mosquitoes that smelled a blend of decanal, which activates the human-specific glomerulus, and 1-hexanol, which activates the human-and-animals glomerulus, would fly upwind in search of the source. “Importantly, they also show that these components are behaviorally relevant to the mosquitoes—mosquitoes will track the binary blend of synthetic odorants in the same way that they respond to whole human odor,” notes Duvall. 

The decanal and undecanal are probably derived from sebum, an oily substance that—unlike sweat—is secreted from hair follicles regardless of physical activity. Finding a role for sebum in mosquito attraction is novel, Matthew DeGennaro, a researcher in vector-borne diseases at Florida International University who was not involved in the study, writes in an email. “Previously, most the of the focus has been on human sweat components such as lactic acid or on how the human skin microbiome processes sweat and sebum into our distinct odor.” 

See “Children With Malaria Smell More Attractive to Mosquitoes

While the engineered mosquitoes allowed the researchers to look at responses of olfactory receptors, other neurons that might play a role in detecting humans were not imaged, notes DeGennaro, adding that in Ae. aegypti, a receptor called Ir8a “may play a role given its ability to detect lactic acid and other acidic human odors. . . . Lactic acid is a component of human sweat that cause mosquito attraction in combination with carbon dioxide, not many odors are known to be able to do that.” Currently, it is unknown to what extent these receptors play a role in distinguishing nonhuman animals from humans, writes Potter. 

McBride acknowledges that signals beyond concentrations of decanal and undecanal are likely at work when mosquitoes get very close and decide whether to land on a potential victim. The experiments in a wind tunnel tested how mosquitoes figure out where to fly to, with conditions similar to those a meter or two away from humans. “We don’t think this is the end of the story: this isn’t the only way they recognize humans, but we think it’s one of the biggest parts.”

In future work, Duvall says she’d like to see researchers compare Ae. aegypti with mosquito species that typically bite different animals “to ask whether they use the same type of olfactory coding to distinguish their preferred hosts, or if their preferences are mediated by distinct mechanisms.” 

One potential application of the study’s insights is to the design of new mosquito attractants and repellents, writes Potter. “This could lead to the development of ‘super’ attractants that smell even better to a mosquito than a real human, which can then serve as baits to lure mosquitoes away from humans.” When combined with insecticide, such lures could be used for reducing mosquito populations. However, McBride cautions that more research is needed. “What we can say now is that if you want to design a blend that’s really attractive to these mosquitoes, you should include at least a little bit of decanal or undecanal.” The new neural imaging techniques could be a big step forward to screen for other compounds that inhibit or activate these neurons, McBride adds. “It opens up all sorts of doors to designing better repellents and attractants.”