Mosquitoes prowling for a blood meal are drawn to the plumes of carbon dioxide exhaled with each human breath. But they also buzz around dirty socks and worn clothes—gravitating toward skin odor even in the absence of a panting human—leaving scientists to puzzle over what internal guide drives mosquitoes to dinner.

A closer examination of mosquitoes' neurons and behavior has now revealed that mosquitoes suss out eau de human in the same way they sense carbon dioxide. The insects rely on the same receptors and the same neural pathway to detect bare ankles and CO2 alike.

The olfactory breakthrough cracks open new possibilities for future mosquito control worldwide. By tapping existing compounds known to block one class of olfactory neurons, scientists might be able to better camouflage humans from mosquito detection or draw mosquitoes away from humans with chemical bait. Reducing even some number of bites could help prevent cases of malaria, yellow fever and dengue—all contracted and transmitted via the bite of the female mosquito.

The research all started with smelly feet.

A team of researchers from the University of California, Riverside, placed odor-collecting beads in their own socks to pick up a scent and placed them in an experimental wind tunnel with hungry mosquitoes, where the beads acted as a stand-in for a human. Then the researchers doused dengue-spreading mosquitoes with a chemical that shuts down the mosquitoes' ability to detect CO2 by blocking a receptor called neuron–cpA, a type of sensory cell found on a small pair of antennalike structures near the mouth called the maxillary palps. The cpA-blocked mosquitoes were about 75 percent less likely to seek out the odoriferous beads than were mosquitoes doused with a material that did not interfere with cpA. About 85 percent of the cpA-blocked mosquitoes did not make it to the odor beads even though their flight ability was unaffected by the chemical. In contrast, all but 35 percent of uninhibited mosquitoes made it to the beads. The study, which is the first to identify a causal link between a specific type of neuron and mosquitoes' human odor-sensing capability, appears in the December 5 issue of Cell.

This new research indicates that the maxillary palps, rather than the antennae themselves, serve to seek out humans. "Many scientists, including us, had focused mostly on the mosquito antennae" for sensing human odors outside of exhaled carbon dioxide, says senior study author Anandasankar Ray, an entomologist at the University of California, Riverside. Mosquitoes' knack for sensing humans through cpA was "completely unanticipated," he says.

Human musk is a complex medley of skin, sweat and bacteria, and scientists still do not know why one person's skin makes for a more attractive dish than her or his neighbor's. The team did, however, discover that mosquitoes respond more strongly to a mixture of carbon dioxide and skin odorants than to either one of those tantalizers alone.

Based in part on the new cpA findings, the scientists compiled a list of thousands of chemical candidates for future bug repellents. They winnowed that list down to 138 substances found in nature and then made further cuts to arrive at a few top contenders that were cheap, smelled pleasant (to humans) and had been demonstrated to be safe for some human uses.

One such chemical, ethyl pyruvate, is a fruity-scented cpA inhibitor that is already approved as a flavor agent in food and can be found in some wines. The team showed that mosquitoes were less interested in a gloved human hand with ethyl pyruvate on it than in an untreated glove. Such a compound, the authors write, could provide a safe and effective alternative to DEET, a repellent which is not widely used in disease-prone areas in Africa and Asia, largely due to cost and the need for consistent application on all exposed body parts. Ideally, ethyl pyruvate would be cheaper than DEET and sprayed on bed nets or other surfaces to act as a short-term human invisibility cloak to keep mosquitoes away, Ray says.

Mosquitoes do have human-sensing tools other than cpA. But even preventing a small number of bites could have a huge impact because a female mosquito must engorge herself with a blood meal at least twice to transmit a disease such as malaria or dengue—once to contract the disease, and again to pass it along.

Beyond discovering how to mask the scent of humans, the team's experiments also alighted on a chemical that attracts mosquitoes. The researchers found that just as mosquitoes are attracted to carbon dioxide, they are also drawn to a peppermint-scented compound called cyclopentanone, a chemical that triggers the same activation pathway through the cpA neuron. That discovery could set the stage for cheaper, simpler mosquito traps than those using carbon dioxide, which often involve dry ice or burning propane and are difficult to deploy in developing countries. In the lab, a cyclopentanone trap was shown to be just as effective as existing CO2 traps. "To our knowledge, no other chemical has been able to successfully lure mosquitoes to traps in large numbers in the absence of CO2, let alone at rates comparable with CO2," the researchers report in their study.

In 2011 Ray's group published a study in Nature identifying several chemicals that could activate or inhibit the cpA receptor—which at the time was known for sensing carbon dioxide but not human odors—yet those chemicals can smell offensive and are not safe for human exposures at the necessary concentrations, he says. (Scientific American is part of Nature Publishing Group.) The safety of some of the materials identified in these new experiments, too, would have to be evaluated. Some work on ethyl pyruvate, for example, suggests repeated exposure to skin may cause cracking, flaking or drying.

Although the new research looks "promising," the newly identified compounds have not been tested outside of the lab, says James Logan, scientific director of the Arthropod Control Product Test Center at the London School of Hygiene and Tropical Medicine, who did not contribute to the study. "It is a long and expensive process to get compounds to market and into practice," Logan says. The most useful repellent compound, he adds, "would be one which repels at a distance, providing spatial repellency without the need to apply directly to the skin."