Center News

Study deepens understanding of how brain perceives scents

Through large-scale analysis of odor detection, Linda Buck and colleagues find extraordinary diversity, bias in odor recognition

July 18, 2011
Dr. Linda Buck

“More in-depth understanding is important because the neurons in the nose and the receptors they use to detect odorants are the ones that provide information to the brain that the brain translates into odor perceptions,” said Dr. Linda Buck, Basic Sciences Division.

Photo by Jim Linna

Hutchinson Center researchers have gained a broader and more detailed understanding of how mammals perceive odors and send such sensory input to the brain. The findings, by principal investigator Dr. Linda Buck of the Basic Sciences Division, were published June 22 in The Journal of Neuroscience. The work builds on previous Nobel Prize-winning research by Buck.

Mammals can perceive and discriminate a myriad of chemicals as having a distinct odor. Smells are initially detected by odorant receptors on olfactory sensory neurons in the nose. In a mouse, each neuron expresses one of 1,000 different receptor genes. The neurons transmit signals to the olfactory bulb of the brain, which then sends signals to the olfactory cortex. From there, the sensory signals are transmitted to other brain areas, ultimately resulting in odor perceptions. Although neurons and their expressed receptors constitute the fundamental units of sensory input to the brain, a comprehensive understanding of how they encode odor identities was lacking.

Buck, along with lab colleagues Drs. Xiaolan Ye, Luis Saraiva and Kiyomitsu Nara (now at Fukushima Medical University), tested the responses of 3,000 mouse olfactory neurons to 125  odorants with diverse structures and perceived odors. Mice have about 1,000 different odorant receptors and humans have about 350, but the receptors are likely to function similarly in odorant detection.

“More in-depth understanding is important because the neurons in the nose and the receptors they use to detect odorants are the ones that provide information to the brain that the brain translates into odor perceptions,” Buck said. “In order to understand the mechanisms that underlie perception, it is essential to first know what these initial inputs to the brain consist of.”

‘Combinatorial receptor codes’

Buck’s lab previously showed that each neuron uses only one type of receptor, which means the information that each neuron sends to the brain reflects the odorant specificity of a single type of receptor. They also showed that one odorant can be recognized by multiple receptors and that one receptor can recognize more than one odorant, but that different odorants are recognized—and thereby encoded—by different combinations of receptors. Thus, there are “combinatorial receptor codes” for odors. In fact, even odorants with very similar structures have different combinatorial receptor codes, explaining how they can be perceived as having very different odors, like orange versus sweat.

With this study, the researchers sought to understand how many receptors are needed to recognize a single odorant and if that number differs for different odorants. Additionally, they attempted to determine if individual receptors are narrowly or broadly tuned to recognize a few odorants with similar chemical structures or many odorants with differing structures. And given that receptors are used in a combination, what determines our perception of an odor as, say, minty instead of fishy?
 
The researchers extracted single olfactory sensory neurons from the nose and loaded them with a fluorescent dye that senses changes in intracellular calcium. While imaging the neurons with a microscope, they flowed fluid containing odorants over them and used imaging software to monitor changes in fluorescence emitted by the dye inside the neuron as it was exposed to the odorants. A change in fluorescence indicated an increase in calcium, indicative of the neuron's response.  
 
Diversity in neuron response

The research revealed extraordinary diversity, but also bias, in odorant recognition by the neurons and receptors. A total of 217 neurons responded to odorant mixtures in these experiments. The study showed most odorant neurons are narrowly tuned to detect a subset of odorants with related structures and often related odors, but that the repertoire also includes broadly tuned components. Strikingly, the vast majority of odorants activated a unique set of neurons, usually two or more in combination. The resulting combinatorial codes varied in size among odorants and sometimes contained both narrowly and broadly tuned components.

While many of the neurons recognized multiple odorants, some appeared specific for a given pheromone or other animal-associated compound. The neurons were highly specific for a pheromone in male mouse urine that accelerates female puberty onset, two odorants with a fecal odor, an odorant that smells like decaying flesh, or individual musk odorants. This raises the possibility that signals derived from some of the neurons and receptors might be able to elicit an innate behavior.

The vast majority of odorants tested were recognized by a unique set of neurons. Of 125 test odorants, 102 stimulated one or more neurons and 96 out of 102 stimulated a unique set of neurons. 77 percent of the odorants were recognized by more than one neuron and only 23 percent by a single neuron, showing that the combinatorial coding scheme extends to a wide variety of odorants with different structures and perceived odors. 

Determinants of odor quality

The researchers found that some neurons recognized only a single odorant or odorants that shared a particular odor quality, such as fishy, minty, woody, or fruity. However, most odorants were also recognized by neurons that responded to odorants with different odor qualities. They concluded that odor quality is likely to be determined by the combination of neurons and receptors, but that there may be some neurons and receptors that do convey a particular quality, such as fishy.    
 
“It is satisfying to finally have a broad view of how the repertoire functions because there were so many unknowns before,” Buck said. “The enormous diversity seen in the recognition properties of individual neurons and receptors is truly remarkable.”

Buck and colleagues hope to use the receptors isolated from mouse neurons that recognize certain odorants as a means of identifying human receptors with similar functions.

Howard Hughes Medical Institute, Japan’s Mitsubishi Kagaku Institute of Life Sciences and a grant from the National Institutes of Health’s National Institute on Deafness and Other Communication Disorders supported the research.

Fred Hutchinson Cancer Research Center is a world leader in research to prevent, detect and treat cancer and other life-threatening diseases.