Noses, mouthparts and antennae are just some of the body parts used to detect smells in nature. What all these structures have in common is sensilla– hair-like structures that contain one to a few olfactory neurones,which respond to odours. Olfactory neurones sport a variety of receptor proteins on their cell surfaces, which bind to molecules in the environment. When enough odour molecules have bound to the receptor proteins, the neurone fires, and sends a signal that is interpreted by the animal as an odour. Olfactory sensilla, the neurones within them, and the protein receptors on the neurones' surfaces represent three levels of odour coding. Each level can be grouped into a variety of classes depending on morphology and function. Unfortunately, the relationship between these three levels is complicated: it was previously thought that a single class of receptor proteins does not fall into a single class of olfactory neurones, and a single class of olfactory neurones does not fall into a single class of olfactory sensilla. In other words, the coding relationship was thought to be more complicated than a simple 1:1:1 ratio. That is until Dobritsa and colleagues proved everyone wrong.

In a paper published in Neuron, Dobritsa and co-workers have combined anatomical, electrophysiological and genetic techniques in a study of fruit fly antennae, to challenge all three levels of the troublesome coding cascade. First, the team decided to map the location of two receptor proteins. They designed antibodies to the Or22a and Or22b receptor proteins, which they took as two examples of the full range of possible receptors to label cells that carried the receptor. Electrophysiological techniques were then used on the labelled cells to measure their sensitivities to a range of odours. By looking at the distribution of labelled cells, the team found that both receptors were restricted to a single morphological class of sensilla (LB-1)and a single functional class of sensilla (ab3).

Next, Dobritsa and colleagues found that the two receptor proteins were restricted to a single olfactory neurone type when they combined electrophysiological recording techniques with gene mutation studies. The team showed that when Or22a and Or22b are not expressed, the response of one type of olfactory neurone is affected (ab3A). Usually the ab3A neurone responds to a broad range of different odours and is most sensitive to ethyl butyrate. Interestingly, when Or22a was knocked-out, all ab3A activity disappeared. This means that the Or22a receptor is the only active receptor on the ab3A neurone and is fully responsible for its whole spectrum of odour-response properties. Because ab3A can respond to a range of odours, the presence of many different receptor protein types was expected. Therefore, finding that ab3A, a single neurone class, uses only one receptor protein (Or22a) was a surprising result. By combining this discovery with the fact that the ab3A neurone is restricted to one class of sensilla, Dobrista et al. propose a novel 1:1:1 ratio to describe the relationship between three coding levels; receptor to neurone to sensillum. Further studies of other olfactory receptor genes will be required to determine whether this is a general principle.

Dobritsa, A. A., van der Goes van Naters, W., Warr, C. G.,Steinbrecht, R. A. and Carlson, J. R. (
). Integrating the molecular and cellular basis of odor coding in the Drosophilaantenna.