How The Nose Knows

By Philip Elmer-DeWitt

The sense of smell is the most primitive of the five senses, a throwback to the primordial mists when the brain was scarcely developed. It is also the least understood sense. The human nose can distinguish an extraordinary bouquet of odors, some 10,000 in all, and other animals can better that. It has long been recognized that moths, for example, are exquisitely sensitive to certain pheromone molecules and can sniff out a potential mate half a mile away. But scientists could not begin to explain precisely how they did it.

Until last week. In a discovery that promises to open up a whole new field of olfactory science, two researchers at Columbia University announced they have isolated what they believe are the first known odor receptors -- individual genes that are active in the nose and nowhere else in the body. What is more, the molecules they found seem to be part of an extended family of smell genes -- perhaps the largest single family in the long strand of mammalian DNA. "We have identified a few hundred genes," says Richard Axel, a professor at Columbia's Howard Hughes Medical Institute. "And there is reason to suspect there may be as many as a thousand."

That is a lot of genes by modern standards. The eye, by contrast, uses only three different types of receptors -- one sensitive to red light, another to green light and another to blue -- to recognize a few thousand different colors. Most of the information processing required to distinguish, say, mauve from chartreuse is actually done by the brain.

The new findings, published in the current issue of the journal Cell, suggest that the sense of smell may work very differently. When odor molecules drift among the millions of tiny cilia located high in the nasal cavity, they seem to slip into certain odor receptors like keys into locks. The fact that there are such a large number of different kinds of odor receptors suggests that much of the work of discriminating among smells is being carried out at a chemical level within the nose itself. Signals from these receptors are then transmitted to the olfactory bulb, the small region of the brain that specializes in identifying fragrances. But since that information has been filtered through the odor receptors before it is passed along, the brain does not have to do very much of its own processing before concluding that what it is confronting is a garlic clove and not a rose.

This makes a certain amount of sense from an evolutionary point of view. Although humans tend to treasure sight above all other senses, primitive animals probably relied more heavily on smell than on vision for their survival. And since their small brain size may have limited their capacity to process large quantities of information, they needed lots of specialized cells to do the work of identifying, say, the smell of food that had spoiled or the odors associated with fertility and reproduction.

The nose, therefore, may be a key to understanding how the brain works. "These molecules will serve as useful tools" for solving a variety of scientific problems, says Linda Buck, who co-authored the Cell article with Axel. This knowledge may even yield some practical benefits. Pesticide makers may be able to design improved insect repellents based on a better understanding of why certain pests are attracted to some people and not to others. And who knows, perfume manufacturers could someday offer custom-made scents that are designed to snare not just any man, but a particular, special someone.