Experience-dependent plasticity of the peripheral olfactory code

The sense of smell - olfaction - is key to many activities, including finding food, locating a mate and for social interactions. It is strongly connected with learning and memory, which in turn depend on changes in how the brain processes information - referred to as 'neural plasticity'. For example, some diseases involving impaired memory are associated with a reduced ability to smell; learning about smells is easier when an individual has experienced them before; and newborns show preferences for odours that their mothers experienced during pregnancy. Although many of these effects take place in the brain, this kind of 'plasticity' can also take place at the first stage of processing - in the neurons in the nose. However, we know very little about this peripheral plasticity. We hypothesise that changes in the peripheral sensory code form an essential component of neural plasticity underlying altered olfactory behavioural responses. This hypothesis is supported by our recent work showing that peripheral olfactory sensory neurons (OSNs) can generate variable responses to exactly the same odour. This implies that the sensory code reaching the brain is not predictable for a given odour and that plasticity in the sensory code occurs in the periphery. We will study the fruitfly larva. The neural 'wiring diagram' of a larval nose is the same as in humans, but much simpler - it has only 21 OSNs, compared to millions in humans. Each of these OSNs is unique because they possess different types of odour receptors. Because of this, we can record the electrical firing activity of individual OSNs in a whole larva and find out how each one responds to a range of odours. Furthermore, we can genetically alter larvae so that only 1 or 2 identified OSNs are capable of responding, or we can selectively kill individual identified OSNs so that they cannot respond. Behavioural tests tell us which odours larvae can detect and which ones they prefer and, just like more complex animals, larvae show altered responses to odours dependent on previous experience of specific odours. They can also learn to respond to an odour that is associated with a sugar reward. We will first describe the changes that occur in the electrical responses of individual OSNs in larvae that have been trained using specific odours. We expect to find OSN-specific changes in the sensory code following prior exposure of the larva, or its mother, to an odour. We also predict that the ability to learn associations between specific odours and food reward will be enhanced. Once we have acquired the electrophysiological data, we will construct a computer model of how the peripheral olfactory code works in this simple organism. This model will be based on real firing data from the responses of individual OSNs to a range of odours. It will then be used to make predictions about how well the system can discriminate particular odours. We hypothesise that prior olfactory experience and accompanying peripheral neural plasticity will improve discrimination ability and we will test this both on real larvae and using the computer model. The model will also predict which OSNs are the most important for discriminating specific odours and we will test this using larvae in which these OSNs have been selectively killed. The numerical simplicity of our larvae will allow us to gain new insights into sensory coding and how this links to behaviour in a whole animal. The computer model can be developed for application to more complex sensory systems. In addition our work will be relevant to the development of improved odour-based insect pest control strategies.

We will address the hypothesis that plastic changes in the peripheral sensory code form an essential component of neural plasticity underlying altered olfactory responses following behavioural conditioning. This hypothesis arises from our recent finding of unexplained variability in the responses of primary olfactory sensory neurons (OSNs) in Drosophila larvae. It challenges the view that the OSN population functions as a hard-wired set of 'labelled lines' whose activity forms an unvarying 'combinatorial code' for a given odour. Specific hypotheses are: 1. Non-associative changes in olfactory behavioural responses are reflected in specific changes in the peripheral neural code. 2. Associative conditioning does not directly involve peripheral plasticity. However, the capacity to show associative conditioning is influenced by previous non-associative olfactory experience. 3. OSN response plasticity conveys a functional advantage by increasing the ability of the OSN population code to discriminate specific odours. Using the simple olfactory system of the Drosophila larva, we will combine electrophysiological recordings from OSNs in situ, genetic manipulation of receptor expression in specific OSNs, and computational modelling approaches. Neural correlates of non-associative conditioning will be defined using single-functional-OSN larvae in which specific OSNs can be identified. The effect of prior odour experience on the ability of wildtype larvae to show associative conditioning will be measured. Computational modelling based on real OSN firing data will be used to explore the information-processing implications of OSN response plasticity and to make predictions about changes in the ability of the system to discriminate odours following conditioning. These predictions will then be tested experimentally. We expect our work to advance the understanding of olfactory coding, by generating a predictive model that could be applied to other, more complex systems.

Who will benefit from the research? Beneficiaries will be: - The academic scientific community, especially those with an interest in sensory neuroscience, neural plasticity, computational neuroscience, Drosophila biology, and insect pest control. - Commercial companies with an interest in olfaction and/or pest control. - Students taking specialist courses in sensory neuroscience and insect biology. - School children and the general public. How will they benefit? Our findings will be of wide interest to the field of sensory neuroscience, including other modalities. The predictive model of olfactory coding will be applicable to other organisms, thereby broadening the importance of our study. Our findings will also be of interest to those working on links between olfactory processing and memory in mammals. Our research will be of economic relevance to the development of improved chemical control strategies for pest species of insects including disease vectors. The computational model will also be of interest to the commercial sector developing products linked to olfaction - e.g. perfumes, deodorants - and will contribute to research aimed at developing a realistic artificial nose. Students and the public have a real fascination with how animals use their senses to interact with the outside world. They also relate to science which has relevance to human biology and disease, such as sensory and memory disorders. Our public engagement work will reinforce and feed these interests, thus promoting an interest in science generally. What will be done to ensure that they benefit? We will continue to publish in high impact journals. We will also present our work at international and national conferences, both specialist and general. Once launched, the UK CARMEN data-sharing project will provide a platform for sharing our electrophysiological data with the sensory neuroscience community via an internet portal. RP is the Manchester representative of CARMEN. We will engage with the commercial R&D sector with the aid of our Faculty Research Business Managers and the University of Manchester Intellectual Property Company (UMIP). All three applicants are active in teaching specialist courses relevant to the proposed work, and new findings will be disseminated by this route. The Faculty has a full time Schools Liaison Officer and the applicants are active in presenting their science to school children. Engagement with the general public will include our web presence (facilitated by the Faculty Media Officer), public science lectures, and contribution to the 2011 Pestival Insect Arts Festival in London.