Some Skin in the Game

If from a neuroscientist’s perspective smell is the least understood of the senses, then from a perfumer’s point of view skin must be the least understood of the organs. We wear scent on the skin and it is from this biological surface that we judge a perfume to be an aesthetic success or failure. As a substrate for the perfumer’s creativity, skin is what wood panels, canvas or watercolor paper are to the painter. The biology of skin may determine why a perfume evolves differently on different people. One would think that the fragrance industry knows a lot about the interaction of skin and scent, but one would be wrong. What little is known is gleaned from studies aimed primarily at other topics.

Skin is also the broadcast medium for a range of endogenous bodily odors. These intrinsically human scents reflect emotional state, diet, sex, and genetics. One might think that by now scientists have thoroughly catalogued these odors and analyzed their function. But again, one would be wrong. As several recent studies show, we are still discovering unexpected aspects of human skin scent.

Take the glands of Montgomery for example. (Don’t tell me you haven’t heard of them.) Montgomery’s glands are found in the areola of the female nipple; they look like goose bumps and are anatomically a combination of sebaceous and lactiferous units. They become conspicuous during late pregnancy and lactation. Beginning after parturition they exude a fluid that is distinct from breast milk. The function of this fluid has been something of a mystery since W.F. Montgomery first described the areolar glands in 1837.

French psychologist Benoist Schaal is an expert in the olfactory cues that draw a newborn baby to its mother’s breast. He and his colleagues have focused increasingly on the role of Montgomery gland (MG) fluid. In a new paper, they demonstrate that 3-day old infants respond selectively to these secretions. Newborns turn their heads and make mouthing movements when presented with MG fluid collected from nursing mothers. They also show changes in heart rate and respiration, presumably in response to volatiles in the MG fluid (curiously, the fluid has almost no detectable odor to adults). 

Compared to human breast milk or cow’s milk, MG fluid provokes quicker and more intense orienting behavior. Finally, bottle-fed infants show the same behavior as breast fed babies, indicating that the response is in-born and automatic, not learned. Schaal’s team thinks they have the key to the mystery: MG secretions are a specialized scent signal that guides the infant to the mother’s nipple and promotes successful nursing. (Hint to new moms: don’t go overboard with the wet-wipes before feedings.)

The skin scents that make us human also make us a target for insects, notably Anopheles gambiae, the African mosquito that transmits malaria. It is well known that mosquitoes home in on carbon dioxide, ammonia and L-lactic acid (a breakdown product in human sweat). Another set of molecules—aliphatic carboxylic acids such as propionic, butanoic and pentanoic acids—were thought to be part of the BO bouquet that attracts mosquitoes. But which ones and in what proportions? Finding the answer could greatly improve the effectiveness of scent traps and reduce the incidence of potentially infectious bug bites.

In a new paper, a large international research team describes how they methodically varied the combinations and concentrations of CO2, ammonia, L-lactic acid and seven aliphatic carboxylic acids to come up with a synthetic human scent that, under some conditions, is even more attractive to mosquitoes than the real thing. The team tested the synthetic lures in the field in a village in southeastern Tanzania. The optimized odor blend was more attractive than human scent when the two samples were set up in huts located 10 to 100 meters apart. When mosquitoes had to choose between sample in close proximity in the same hut, the synthetic lure no longer had an advantage. Still, bait stations strategically located outside a village could draw significant numbers of insects away from homes in the village.

From a chemist’s point of view, human BO is extremely complex. What’s remarkable here is that a human scent realistic enough to draw mosquitoes can be built with nine simple molecules. (That simplified “aroma models” can provide high fidelity olfactory impressions of complex odors is a theme I develop in What the Nose Knows.)

Finally, a provocative new study in the Journal of Forensic Sciences takes the power of the minimalist odor model even further. A group at Florida International University did extensive chemical analysis of volatile compounds in the scent collected from the hands of volunteers. From the 37 different molecules identified, the researchers selected 24 that were common to set of ten volunteers. Using the relative proportions of these 24 molecules the FIU scientists could identify individual hand-scent samples with high accuracy. This means that in principle a standard set of volatile compounds could provide personal BO “bar codes” searchable in a computer database. Welcome to the world of olfactory biometrics.