Why Does the Same Musk Smell Weaker to Some People?

The same fragrance material may smell distinct to one person and very faint to another. Such individual differences are well known for musks. Beyond preference and experience, might people differ at the stage where their noses receive odor molecules?

In 2022, Bingjie Li and colleagues asked participants in China and New York to smell actual odors and compared perception with genetic variation in data from 1,357 people. New clues concerned the musk material Galaxolide and a component of underarm odor. Rather than “diagnosing fragrance preferences from genes,” the study aimed to clarify mechanisms connecting receptors with perception.

Searching in 1,000 people in China and testing in 357 in New York

The initial cohort comprised 1,000 Han Chinese participants aged 18–55 in Tangshan, China. They smelled eight single-component stimuli and two fragrance mixtures, ten in total, and rated intensity and pleasantness on 100-point scales. The researchers compared responses with genetic information from blood samples to search broadly for genetic differences associated with odor perception.

Validation used a separate New York cohort aged 18–50. Although 364 people participated, 357 remained for analysis after checks of response reproducibility and genetic-data quality. Together with the 1,000 Chinese participants, the final genetic-perception analysis included 1,357 people. The validation cohort’s larger olfactory experiment included six stimuli shared with the Chinese study, which were used for this paper’s comparison.

Because people use rating scales differently, the main analyses converted scores into within-person rankings of intensity or pleasantness across the stimuli each participant smelled. The results below therefore primarily concern relative intensity rankings among odors. They do not directly represent detection thresholds or everyday quantities of perfume used.

OR4D6 was associated with Galaxolide intensity

Galaxolide is a synthetic fragrance material with a musky odor. Its perceived intensity was strongly associated with two variants in OR4D6, a gene encoding an olfactory receptor. People who inherited the variant form from both parents tended to rank its intensity lower. Validation in New York confirmed the association at the higher tested concentration.

The representative variant S151T explained 13.27% of variation in intensity rankings in the Chinese cohort and 7.54% in the New York cohort. This means that it statistically accounted for part of the differences between participants. It does not mean that the gene detects 13% of the fragrance or that carriers smell only 13% of it. The combined analysis found no significant association with Galaxolide pleasantness.

However, the mechanism by which OR4D6 might directly detect Galaxolide was not established. Nearby genes encode other musk-related receptors and tend to be inherited together. Although statistical analyses favored OR4D6, experiments in cultured cells did not confirm a response to Galaxolide. The authors also identified poor solubility in the experimental system as a problem.

Does a musky impression arise through just one route?

This finding is interesting because the perceptual grouping of musks does not match a single grouping of molecular structures. Molecules with very different structures can produce similar musky impressions. Yet someone who struggles to perceive one musk does not necessarily struggle with others.

Combining their results with existing research, the authors discussed the possibility that different, or partly overlapping, combinations of receptors respond to different musks and produce similar perceptions. OR4D6 is a strong candidate for investigating this mechanism, but the evidence is insufficient to call it “the gene for musk.” Musks should not be treated as one stimulus: the molecule and testing conditions matter.

For a body-odor component, cell responses and human perception did not align simply

The other new association involved trans-3-methyl-2-hexenoic acid, or 3M2H, a component contributing to characteristic underarm odor, and the receptor OR51B2. A form containing the L134F variant in OR51B2 was associated with a tendency to perceive 3M2H more strongly.

In cultured cells, however, the reference OR51B2 responded to 3M2H while the form containing that variant did not. This was opposite to a simple expectation that stronger cellular responses should mean stronger human perception. The authors suggested that cultured cells might not fully reproduce the environment of olfactory sensory neurons in the nose. Replication of the association and responsiveness of the reference receptor were important, but parts of the pathway to perception remained unresolved.

Because 3M2H is only one component of body odor, reduced perception of it is not the same as inability to perceive body odor as a whole. The authors considered OR51B2 a candidate for research into malodor reduction and communication through body odor. The study did not, however, test deodorant efficacy or effects on interpersonal relationships.

What replicated and unconfirmed associations reveal

The researchers also tested previously reported associations, including the violet-like fragrance material β-ionone and OR5A1. The β-ionone finding clearly replicated in both cohorts. Androstenone and OR7D4 also supported receptor involvement, although replication differed across intensity and pleasantness measures. Evidence was weaker for green-leaf-like cis-3-hexen-1-ol, and the previously reported caproic-acid association was not adequately replicated.

No significant genetic associations were found for the aldehyde-related stimuli or the two fragrance mixtures tested in the Chinese cohort. The authors considered possibilities including many receptors each contributing small effects and influences of cultural or social experience. Failure to find an association here does not establish that genetics has no role.

Presentation methods and concentrations also differed: the Chinese study used pen-shaped devices and the New York study glass vials. Confirmation across different conditions and populations is valuable, but discrepant findings cannot immediately be attributed to ethnicity or culture.

Evolutionary interpretations also require caution

Combining their findings with previous reports, the authors examined a tendency for evolutionarily derived variants to be associated with weaker odor perception than ancestral forms. They interpreted this as supporting the hypothesis that primate olfactory receptor function has declined over evolutionary time.

However, the study was not designed to test that hypothesis directly and may be biased toward variants already discovered and reported. Perceived intensity for particular odors cannot establish that modern humans have uniformly poorer overall smell ability or that a particular person or population has a superior nose.

Summary

The study connected differences in perception of the same musk or body-odor component with specific candidate receptor genes. Its larger scale, inclusion of an East Asian population previously less represented in research, and validation in another cohort were important. It showed that subjective differences can involve variation at the entry point of odor perception.

At the same time, genetic associations, cultured-cell function, and human perceived intensity do not always form a straight line. Galaxolide still lacked cellular confirmation, and 3M2H showed opposite directions in cells and perception. Resolving these discrepancies may help explain how different combinations of receptors produce a single odor impression. The authors outlined a route that combines genetics, olfactory measurements, and functional receptor experiments to investigate these mechanisms. They did not establish a way to predict preferences for entire perfumes from perception of particular materials.

References
Li B, Kamarck ML, Peng Q, et al. (2022). From musk to body odor: Decoding olfaction through genetic variation. PLOS Genetics, 18(2), e1009564. Read the original article
Rights: The original article is published under Creative Commons Attribution 4.0 International (CC BY 4.0).