When smelling the same smoky odor, some people find it strong and unpleasant, while others are less bothered. Receptor function in the nose was also implicated in these differences. For the odor molecule guaiacol, people with less responsive variants of the receptor OR10G4 tended to rate the odor as weaker and more pleasant.
However, the interest of this study goes beyond finding a single gene for smoky-scent preferences. It linked the screening of hundreds of olfactory receptors with questions about which molecules they respond to, how much their function differs between genetic variants, and whether those differences also appear in human perception.
The machinery for receiving odors differs from person to person
Olfactory receptors are proteins that receive odor molecules. Even receptors with the same name can differ in some of their constituent amino acids because of genetic variation. The researchers investigated whether these small differences altered receptor responses.
They began with 511 gene clones covering 394 human olfactory receptors. Because these included different variants of the same receptor, the number of clones was not the same as the number of receptor types. They expressed them in cultured cells, screened their responses to 73 odor stimuli in stages, and tested promising combinations at different concentrations.
This confirmed responsive receptor–molecule pairings for 27 receptor types. Eighteen of these had not previously had a known corresponding molecule. The work thus expanded the map linking odor molecules to their reception points in the nose.
Functional differences in 17 of the 27 receptors studied
Comparisons between receptor gene variants found differences in concentration–response behavior for 17 of the 27 receptor types, or about 63%. The differences were not uniform: some variants responded more readily to weak stimuli, while others had smaller maximum responses.
In a separate comparison, variants of 15 receptors were examined using 55 chemically diverse stimuli. Although response magnitudes differed between variants, the rank order of molecules eliciting stronger responses was generally preserved. The main pattern was therefore variation in response strength within the same receptor, rather than a wholesale change in its molecular preferences.
The amount of receptor present at the cell surface did not adequately explain the response differences. Neither the location of the genetic changes nor existing computational predictions of function sufficiently predicted the measured functional differences, underscoring the importance of measuring responses in cells.
The researchers also applied their experimental results to public genomic data from 1,092 people, estimating numerous receptor-function differences between individuals. The main text reports an average of 16–22 differences among 54 gene copies, representing two copies of each of the 27 receptors examined. However, the Methods section also gives a maximum of 48, so the exact interpretation of these counts requires caution. At a minimum, the findings suggest that individual receptor variation needs to be examined as diverse response differences, rather than simply whether a receptor works or does not work.
For guaiacol, cellular differences were also linked to human ratings
Do different cellular responses mean that people experience an odor differently? The researchers examined OR10G4 in more detail to address this. In cell experiments, this receptor responded to vanillin and ethyl vanillin as well as guaiacol.
For the comparison with perception, they used data from 308 people whose OR10G4 sequences could be determined, from an earlier study of 391 participants who had rated odor intensity and pleasantness. Participants aged 19–66 had evaluated 66 odors and solvents. In the analysis, ratings were converted to ranks within each person to reduce differences such as individual use of rating scales.
The analysis considered ratings at the higher of two presentation concentrations. This asks a different question from measuring the concentration at which a faint odor first becomes detectable.
When four common genetic variants were compared in cells, some had weaker responses to guaiacol, while others did not produce responses clearly distinguishable from control cells. Carriers of the two less responsive variants tended to rank guaiacol lower in intensity and higher in pleasantness.
A model using OR10G4 variants explained an adjusted 15.4% of the variation in intensity ratings and 10.3% of the variation in pleasantness ratings. These values describe explanatory power for variation within this participant group. They do not indicate what fraction of an individual’s likes or dislikes is determined by this gene.
No association for vanillin, despite the same receptor responding
Importantly, OR10G4 variants were not significantly associated with intensity or pleasantness ratings for vanillin or ethyl vanillin. Among the stimuli compared, including other odors, guaiacol was the one with a confirmed association with this receptor’s variants.
In cells, guaiacol activated OR10G4 at lower concentrations than the vanillin compounds did. The authors suggest that, for molecules that activate a receptor only weakly, differences between its variants may be less apparent in overall perception. Odor perception also involves input from other receptors and subsequent brain processing, so cellular response differences do not always translate directly into perceptual differences.
The study examined the gene sequences encoding receptors in detail. It did not comprehensively cover regions affecting how much receptor is produced, or differences in gene copy number. The perceptual data were also a reanalysis of an existing participant group, not a new replication of the association in a separate group. The authors note that other receptors, as well as cultural and genetic backgrounds, may contribute to individual differences.
Summary
The study expanded known pairings between odor molecules and receptors, and found functional differences between genetic variants in 17 of 27 receptor types examined. This provides evidence for individual diversity in how the nose receives odors.
Specifically, people with less responsive OR10G4 variants tended to rate guaiacol as weaker and more pleasant. The same association was not found for vanillin compounds that activated the receptor in cells. Understanding individual differences in scent perception therefore requires attention not only to which receptor is involved, but also to which molecule is smelled and under what conditions.
The research was supported by US NIH grants, DARPA’s RealNose project, and other sources. The authors declared no competing financial interests.
Reference
Mainland JD, Keller A, Li YR, et al. The missense of smell: functional variability in the human odorant receptor repertoire. Nature Neuroscience. 2014;17:114–120 (published online in 2013). doi:10.1038/nn.3598. Full text hosted by an author’s institution.
