Duke researchers work toward an artificial nose to replace K9 dogs

Walk into airport security lines in the United States, and you’ll see K9 dogs sniffing for anything from narcotics to explosives. But what if a robot could do the same thing?

Duke researchers recently published a paper in which they used mouse genes to grow odor receptors that could respond to specific odors. If this could be developed into a device, the electronic nose might be able to detect some of the odors that dogs are searching for—such as cocaine or explosives. 

"If you can recreate that neuronal activation pattern in a [device], then you can use that as a tool to do the same job as K9s,” said Hiroaki Matsunami, senior author of the paper and professor of molecular genetics and microbiology. 

However, he added that there is still a long way to go in successfully building this device to replace K9s, but his lab has shown that such a device is possible in principle. 

Olfactory sensory neurons in the nose detect and discriminate between a vast number of chemicals and chemical mixtures using a combination of receptors. A single chemical can activate multiple receptors and a single receptor can be activated by multiple chemicals, but each receptor is uniquely tuned so that the combination of the receptor activation is unique to each chemical. The brain ultimately interprets what each of these neuronal combination patterns mean. 

“We used 31 receptors—a very small subset of olfactory receptors—and these 31 receptors showed a specific, unique pattern of activation to certain chemicals," he said. "We can use the patterns to discriminate different chemicals." 

By taking some of the mouse receptor DNA and inserting it into human cells, the researchers were able to test how these receptors responded to various odors. When a specific receptor was activated, the cell would start emitting light that the researchers could observe, Matsunami explained.

One of the challenges of working on this research was navigating the large number of olfactory receptors. 

“Humans have about 400 receptors—mice and rats have about 1,100," he said. "So trying to identify the function of each of them is a very daunting task." 

Therefore, as his lab moves forward with its work on this project, he said they hope to focus on expanding the number of olfactory receptors tested. 

In addition, Matsunami explained that the lab is looking to fine-tune the olfactory receptors to be more sensitive to specific odors. 

“Basically, one can evolve an existing protein to be more specific or to be more sensitive," Matsunami said. "For example, we want to detect this specific explosive, but [the receptor] is not very sensitive. Researchers [in other fields] have used evolutionary techniques to introduce changes in various amino acid positions to improve or change the function of proteins, so can we do [the same] with olfactory receptors? We don’t know yet.”

Furthermore, in order for this project to translate into a real-life device, Matsunami highlighted the need for collaboration with other researchers and for more research in general to be done. 

For instance, not only does the size of the device need to be significantly miniaturized to have practical use, but the device would also have to be much faster at detecting activations. Matsunami explained that as of now, it takes about a minute to detect a response. In the context of real-life situations, minutes are too slow—it must be seconds. 

Additionally, the cells used in an actual device should be reusable. Currently, the cells must be replaced after they are stimulated once. 

“Ideally, we [want to] use the same cells to apply different chemicals over and over maybe for a day or maybe even weeks,” Matsunami said. 

To move forward, there must also be more research in the field regarding nasal mucus and its vast array of functions. Similar to the cell’s culture medium, odor chemicals dissolve in the nasal mucus before binding to olfactory receptors in the nose. 

“But [our] medium is not equivalent to nasal mucus because the mucus contains a number of different things—we don’t even know everything in it and we don’t have a good idea as a whole on what the mucus is in fact doing,” Matsunami said. “But we want to mimic the nose as much as possible in the cell and the portable system. So, in that sense, it would be nice to replace the culture medium with a nasal mucus equivalent, as that may increase the efficiency for odor detection.”

Although significant hurdles and gaps await before the actual creation of an artificial, portable “robot nose,” Matsunami’s lab has taken the first steps towards its realization. 

Mona Tong profile
Mona Tong

Mona Tong is a Trinity senior and director of diversity, equity and inclusion analytics for The Chronicle's 117th volume. She was previously news editor for Volume 116.


Share and discuss “Duke researchers work toward an artificial nose to replace K9 dogs” on social media.