The accepted wisdom is that, during development of an embryo, the nerve cells located in the back of the nose extend long tentacles that are drawn to relay stations in the brain just as drones land at a specific airport because signals told the drones to do so. The destination is in charge.

But Mombaerts, head of the Laboratory of Developmental Biology and Neurogenetics and part of the F. M. Kirby Center for Sensory Neuroscience, and Feinstein, a research associate in Mombaerts’ lab, are turning that notion upside down. They now liken the nerve cells involved in smell to commercial airliners whose pilots determine where the planes fly and land within the confines of air and earth. The neurons are piloted by odorant receptors that control the cells’ course and destination.

The difference is fundamental and it suggests a new explanation for how the olfactory system wires itself to perceive smell. The model may also have implications for understanding how other parts of the brain are constructed.

In two papers published back-to-back in the June 11 issue of Cell, Mombaerts, who has a commercial pilot’s license, and Feinstein propose that odorant receptors tell olfactory nerve axons, the hair-like extensions of nerve cells, what to do, thereby dictating how the olfactory bulb in the brain is structured. The studies, which take up 29 pages in Cell and cover more than 40 different gene targeting studies, are the culmination of eight years of research.

 “This is a new way of thinking about brain wiring,” says Mombaerts. “We propose that the olfactory bulb is almost a blank slate; there are no targets for the axons to navigate to, but the axons themselves are targets for each other and thereby sort themselves out in reproducible patterns.”Even with its mind numbing complexity of millions of neurons, olfaction appears to be one of the simpler brain systems, the Rockefeller researchers say. Odor molecules entering the nasal cavity travel to the olfactory epithelium, a patch of tissue almost level with the eyes. There, thousands of receptors stud the fine hair-like cilia of nerve cells embedded in the tissue. Neurons that serve these receptors each send an axon back to one of two olfactory bulbs in the brain, and “like-minded” axons — those originating from neurons that link to the same odorant receptor — terminate together into brain structures known as glomeruli.

When an odor molecule interacts with a receptor, impulses travel up the neurons to the glomeruli, some 2,000 of which reside within each olfactory bulb. The glomeruli act as relay stations, processing passing nerve signals from the olfactory epithelium to other regions of the brain. About 5,000 axons form a glomerulus and branch into 50,000 axon terminals that synapse with the dendrites of 25 or so second-order neurons, all in the space of a sphere with a diameter of one-tenth of a millimeter.

The question for Feinstein and Mombaerts: How do all these axons find their way to such a tiny and precise spot in the brain?

The prevailing view is that axons navigate to glomeruli via positional cues or signaling information secreted by cells in the olfactory bulb, says Feinstein: “People view glomeruli as existing structures, targets which axons were aiming for.” But glomeruli are not seen when the olfactory system begins to develop. How can a developing axon be drawn to a structure which does not yet exist?

In 1996, Mombaerts published findings that odorant receptors were capable of organizing “like” axons into glomeruli. It was a major hint.

“But we had no model for how they could do that,” Mombaerts says. “It was a bizarre notion that receptors could somehow be involved in telling axons where to navigate. As far as anyone knew, the only role odorant receptors had was to smell. We had no proof of another function.”

The Cell studies now detail that proof. The first paper tests whether odorant receptors themselves carry enough information to drive the identity of the axons. Feinstein carried out a series of “gene swaps” between odorant receptors. By implanting these receptors into the genetic material of mice, he was able to trace axons from neurons expressing associated receptors. He found that when the “identity” of the axon, given to it by its receptor, changed, the wiring also was altered.

“The first Cell paper sets forth the concept that axonal identity is primarily determined by the odorant receptor sequence,” says Feinstein. “However, the axonal identity is revealed by what other axons exist nearby during the process of axonal extension to and within the olfactory bulb. We do not believe that the olfactory bulb contains the information for axonal identity; rather than following positional cues, we suggest that growth cones of axons sort themselves.”

The conclusion: “Our experiments reveal that the odorant receptors exist not just at the ends of nerve cells in the olfactory epithelium, but also along their entire length,” Mombaerts says. “We hypothesize that axons with odorant receptors appear to interact stronger with axons carrying similar odorant receptor structures and repel those that are different. As a result, the system self-organizes such that axons eventually coalesce into like-minded bundles.”

The second Cell paper, by Feinstein, Mombaerts and research associates Thomas Bozza, Ivan Rodriguez and Anne Vassalli, picks up where the first paper leaves off to ask how odorant receptor proteins modulate axonal identity. The team genetically substituted the odorant receptor with an unrelated receptor, a G-protein coupled receptor (GPCR) called the b2 adrenergic receptor. “The expressed b2 adrenergic receptor also supported the formation of glomeruli, which indicates that this receptor, which has nothing to do with the smell system, can also work to guide an axon,” Mombaerts says. “That further suggests that the olfactory bulb is not drawing axons in.”

“This paper confirms several predictions of axonal identity,” says Feinstein. “We show that odorant receptors, including the b2 adrenergic receptor we utilized for the gene swap, are expressed in extending axons. We further show that these GPCRs may also promote axon outgrowth, and speculate that these processes of axon identity and axon outgrowth may be a more general phenomena throughout the brain governed by other, similar proteins.”

From an evolutionary perspective, the new model makes perfect sense. “The olfactory system has to be very plastic to accommodate new genes encoding odorant receptors and new populations of neurons,” Mombaerts says. “If you have a rigid view that axons are attracted by existing cues, it would be much more difficult to explain how these new populations of neurons wire themselves.”

July 16, 2004