Mapping aroma: smells light up distinct brain parts.

Researchers decades ago mapped out the brain's sense of touch, with patches of neurons corresponding to body parts, such as a hand, a lip, or the torso. A new study suggests that the sense of smell may have its own brain atlas. The finding adds to a growing body of research on smell, which scientists haven't studied as much as touch, hearing, or sight.

Last year, Linda Buck of the Fred Hutchinson Cancer Research Center in Seattle and Richard Axel of Columbia University shared the Nobel Prize in Physiology or Medicine for their work in deciphering the mechanisms behind smell (SN: 10/9/04,p. 229). Over the past 15 years, the two scientists have independently worked out how scents are perceived by olfactory neurons in the nose. Their work has also detailed how these neurons transmit signals to the olfactory bulb, a structure at the front of the brain.

However, researchers still knew little about how the brain processes odor information in the olfactory cortex, a region in the temporal lobe, which extends along the sides of the brain.

"We know quite a bit about the mechanisms used to encode touch and sight, but we knew nothing about how different odorants are encoded in the olfactory cortex," says Zhihua Zou of the Fred Hutchinson center.

To fill in that gap, Zou, Buck, and fellow Fred Hutchinson researcher Fusheng Li exposed each of a series of mice for several minutes to a single, pure chemical with a distinctive odor, such as that of apples, vanilla, fish, or urine. Then, the researchers examined fine slices of each mouse's brain. Using antibodies that bind to a protein that brain cells produce only when they're active, the researchers looked to see which neurons had been turned on by each odor.

Although each mouse's olfactory cortex responded slightly differently to a given odor, the patchwork of activated neurons was strikingly similar from mouse to mouse. Furthermore, higher concentrations of a scent activated larger areas of the brain than lower concentrations did.

The researchers found different patterns of activated neurons in the brains of mice exposed to distinctive scents.

However, there were large overlaps in patterns between mice exposed to different scents that are chemically similar, such as two molecules that have thiols, which are sulfur-containing structures.

"It suggests that there may be some logic in the way that odorant molecules are represented in the cortex," says Zou. The researchers report these findings in the May 24 Proceedings of the National Academy of Sciences.

Donald Wilson, who studies smell at the University of Oklahoma in Norman, calls the new study "very exciting." However, he notes that other recent findings suggest that the brain handles familiar scents differently than it handles novel ones. Because the new study examined lab mice that had experienced only a limited environment, the brain map that emerged might not be representative of that of wild mice, let alone other animals.

"The problem is that, generally, these animals have a relatively impoverished olfactory world. When they come out of the cage, they've smelled each other, food, [excrement], and that's about it," he says.

Wilson suggests that a valuable olfactory brain map will require study of animals that have been exposed to a variety of odors in the natural world.