Skip to main content

Neural Basis of Olfactory Perception in Drosophila

Final Report Summary - OLFPERCEPT (Neural Basis of Olfactory Perception in Drosophila)

The OlfPercept project set out to study higher olfactory circuits relevant to unlearned (or innate) behaviours with a strong genetic component. The overarching theme was to understand, at the level of circuits of identified neurons, how raw sensory information is processed to prepare for decision making and the initiation of motor responses. We used the olfactory system because in both vertebrates and invertebrates, just one or two synapses separates sensory neurons from neurons that project to brain regions required for memory formation or innate behaviours.

We proposed that a specific higher olfactory centre in the fruit fly Drosophila, called the lateral horn would show an organisational and functional logic relevant to innate olfactory behaviour. We began by focussing on neurons that might be involved in processing pheromones that contribute to sexually dimorphic behaviour. We carried out a brain-wide search for neuroanatomical sex differences and found numerous examples across the brain, including a significant number in the lateral horn. Detailed analysis of these differences indicated that they had the potential to change connectivity in several olfactory pathways. This work together with related studies from the group of Barry Dickson, substantially revised the prevailing view that physiological rather than neuroanatomical differences were likely to be the main driver of sexually dimorphic behaviour.

We studied some of these differences in detail by recording the activity of identified neurons in flies as they smelled pheromones or other odours. We were able to describe, to our knowledge for the first time in any animal, a changeover switch in the brain which rerouted sensory information to different groups of target neurons in male and female brains. This is an example of a simple circuit motif that may recur in many situations.

These pheromone responsive neurons were genetically, morphologically and functionally stereotyped from one animal to the next. Such stereotypy suggests that the distinct circuit logic might contribute to innate differences in the behavioural responses of male and female flies to pheromone. We have now found that stereotyped anatomy and physiology is the norm in the lateral horn. We have catalogued almost 100 cell types in this area (based on morphological criteria) and recorded from about 50 of these classes. We find that although these neurons are extremely heterogeneous anatomically and functionally (including tuning breadth and the odour categories to which they respond), these responses are highly stereotyped. This is in stark contrast to the other main olfactory centre, the mushroom body, and suggests that the lateral horn contains genetically specified integrator neurons for both pheromonal and non-pheromonal odours, that are likely to be the driver for innate olfactory behaviour. Evidence for such a divergence between stereotyped and non-stereotyped olfactory projections has also been found in mice, suggesting that this may be a general principle.

Our main scientific results have been augmented by a number of related collaborations. With the group of Richard Benton, we have mapped the projections of the fly's "second nose" into the brain and shown that these neurons, while separated from the main family of chemosensors in the periphery and in the antennal lobe (the first olfactory processing centre) are anatomically integrated in the lateral horn. We have also found (again with the Benton group) that information about a food-derived cue (PAA) that can synergistically promote male courtship converges at a specific location in the lateral horn with female pheromone information – this suggests a site for neural integration of these two signals.

This work has been supported by the development of numerous experimental and analytical techniques for circuit mapping, published and in review, that we have shared openly with colleagues. Experimental techniques include transgenic flies (and mouse viral vectors) enabling ultra-rapid multi-colour labelling of fixed tissue, genetically encoded EM markers and pilot work on a new transsynaptic labelling system. Analytical techniques include an open source suite of tools for visualisation and analysis of neuroanatomical data.