Is Fragrance Preference Written in Your DNA?

The Complete DNA Sequence of Human Olfactory Receptor OR10Z1 

How is it that the same odor molecule smells strong to one person and not at all to another? Or that some find an odor disagreeable while others judge it mildly pleasant?

Extreme person-to-person variation is the hallmark of odor perception. It intrigues and frustrates sensory scientists—not to mention fragrance designers and chefs. But a scientific account of these differences has been elusive. Some psychologists believe personal experience determines odor preference but they offer little more than hand waving to support the idea. In any case, personal experience is unlikely to explain a person’s inability to smell a particular molecule, or to explain why entire segments of the population disagree about an odor’s basic olfactory character (e.g., is it like musk, urine, or sandalwood?).

The answer to the riddle of smell variation may lie in biology. Humans have about 450 different olfactory receptors (ORs), each one coded for by a specific gene. A sensory cell in the nose randomly expresses one of these 450 receptors. Every cell with a given receptor sends a signal to a specific location in the olfactory bulb. An odor molecule activates a particular combination of receptors and olfactory bulb locations. This is the neural pattern that the brain eventually interprets as a smell.

As molecular biologists explore the human genome with ever greater resolution they are finding a rich tapestry of OR gene variation—variation they hope will map onto differences in sensory perception. It’s possible that much of what we regard as personal preference and aesthetics will boil down to biology.

Doron Lancet and his colleagues at the Weizmann Institute of Science summarize the latest findings from several labs in a new paper. They explain that the human OR genes are “organized in several dozen genomic clusters distributed on most chromosomes.” OR genes are “highly conserved,” which means that the human versions are very similar to those of other mammals. However, as species we have a relatively high levels of “pseudogenes”; these are OR genes that have been rendered functionless by a disruptive mutation. Pseudogenes play no role in odor perception—they are just so much junk in the genetic attic.

Lancet et al. describe two main types of genomic variation. The first consists of “segregating pseudogenes.” These are OR genes that have been disabled by mutation in some individuals but that remain functional in others. Theoretically, this could account for why Dick (who has the pseudogene) can’t smell molecule X, but why Jane (who has the working version) can.

A second source of variability is copy number variation. This means that one or sometimes a set of neighboring OR genes has been duplicated in evolution. Some people have only the original copy while others have multiple copies. The duplication process is not perfect and small differences in the copied gene may alter the odor specificity of its receptor. Depending on the genetic lottery, a person’s nose might express zero, one, two, or even three versions of an OR, each of which might respond differently to a given molecule.

To date, variations in the perception of androstenone and isovaleric acid have been traced to genetic variations. But these are merely the first results. The new genomics of olfaction promises many more. Faster and cheaper DNA sequencing technologies are coming on line and soon we will know the complete extent of human OR variation. The fun begins when we cross-reference the genetic data to variations in odor perception.

Imagine a personal genetics of perfume preference. A quick spit sample, a few minutes on the sequencing machine, and bingo!—a printout lists the top twenty perfumes preferred by people with a genotype like yours. It might not be romantic but wouldn’t it be an easier way to start shopping for scent?