Read our COVID-19 research and news.

Can You Smell Yourself?

You might not be able to pick your fingerprint out of an inky lineup, but your brain knows what you smell like. For the first time, scientists have shown that people recognize their own scent based on their particular combination of major histocompatibility complex (MHC) proteins, molecules similar to those used by animals to choose their mates. The discovery suggests that humans can also exploit the molecules to differentiate between people.

"This is definitely new and exciting," says Frank Zufall, a neurobiologist at Saarland University's School of Medicine in Homburg, Germany, who was not involved in the work. "This type of experiment had never been done on humans before."

MHC peptides are found on the surface of almost all cells in the human body, helping inform the immune system that the cells are ours. Because a given combination of MHC peptides—called an MHC type—is unique to a person, they can help the body recognize invading pathogens and foreign cells. Over the past 2 decades, scientists have discovered that the molecules also foster communication between animals, including mice and fish. Stickleback fish, for example, choose mates with different MHC types than their own. Then, in 1995, researchers conducted the now famous "sweaty T-shirt study," which concluded that women prefer the smell of men who have different MHC genes than themselves. But no studies had shown a clear-cut physiological response to MHC proteins.

In the new work, Thomas Boehm, a biologist at the Max Planck Institute of Immunobiology and Epigenetics in Freiburg, Germany, and colleagues first tested whether women can recognize lab-made MHC proteins resembling their own. After showering, 22 women applied two different solutions to their armpits and decided which odor they liked better. The experiment was repeated two to six times for each participant. Women preferred to wear a synthetic scent containing their own MHC proteins, but only if they were nonsmokers and didn't have a cold. The study did not determine which scents women preferred on other people, but past studies on perfume have shown that individuals prefer different smells on themselves than on others.

The researchers wanted to know whether the preferences were truly rooted in the brain's response to the proteins. So next, they used functional magnetic resonance imaging to measure changes in the brains of 19 different women when they smelled the various solutions, in aerosol form puffed toward their noses. "Sure enough, there again was a clear difference between the response to self and non-self peptides," Boehm says. "There was a particular region of the brain that was only activated by peptides resembling a person's own MHC molecules." The brain had a similar response to all non-self MHC combinations, suggesting that any preference for how other people smell is a preference for non-self, not for particular MHC types.

Claus Wedekind, a biologist at the University of Lausanne in Switzerland who spearheaded the original smelly T-shirt study, says the results fit well with his research over the past decade. "After our original T-shirt study, I had the impression that people had preferences between different MHC types," he says. "But based on later studies, it seemed that people are actually just distinguishing between self and non-self. This new paper certainly confirms this view."

Past studies on perfumes have shown that different scents amplify the natural aromas of different MHC types -- peach might mesh best with your own smell, whereas vanilla might jibe with your best friend's odor. Boehm says his group's new findings on distinguishing self from non-self smells, which appear online today in the Proceedings of the Royal Society B, could help researchers understand why people prefer different perfumes on themselves than on others. They might, he says, choose to wear a perfume that amplifies their own MHC peptides, but they favor perfumes on another person that amplify a non-self MHC type. But questions on the physiology of sensing MHC peptides still abound. Researchers don't know which receptors in the nose actually sense MHC proteins, because humans don't have the vomeronasal organ that animals use to sniff out the molecules. "We would really like to continue this research to identify the receptors that recognize these peptides in humans," Boehm says.

Other molecules the human body produces could also influence individual smells and scent preferences, Zufall says. The individuality of people's microbiomes—the collection of microbes living in and on us—could also be linked to the body's odor or preferences, Wedekind says. "We just don't know the full physiology yet," he said, "But this is a good start."