Periodic Reporting for period 4 - OlfSwitch (OlfSwitch: Neural circuit switches from molecules to behaviour) Reporting period: 2020-01-01 to 2021-12-31 Summary of the context and overall objectives of the project We are interested in the general problem of how the brain processes information in order to make behavioural decisions. Individual brains cells (neurons) are connected together in a complex network. The set of connections in this network are a key determinant of brain function – an analogy can be drawn to electronic circuit diagrams. Our major goal is to understand what we call elementary circuit motifs, small networks of neurons that serve a very specific function and whose logic may be widely conserved across different kinds of brain. Such basic research into brain function is a crucial complement to applied research in areas such as neurodegeneration and mental health, which are huge issues for our ageing society.We principally uses the fruit fly (Drosophila melanogaster). This has a tiny brain with ~100,000 neurons (one thousand times fewer than a mouse, one million times fewer than a human) but is nevertheless capable of many sophisticated behaviours. We can used genetic tools to identify and manipulate the same neurons from one fly to the next. This allows us to study how they encode information about the outside world, how this controls behaviour and how particular circuit motifs are specified by genes during development. This grant focused on the processing of smells, since they require little brain processing before contributing to decision processes. We studied sexually dimorphic circuits and pathways mediating unlearned responses to different odours. This enabled us to understand how genes can sculpt circuits to specify different behaviours and how signals with opposite or similar behavioural significance are integrated in the brain. We can address these issues with great precision in the fly brain, but they are very likely to reveal conserved principles.In the project we have been able to show 1) how specific neurons with sex-specific connections contribute to sex differences in behaviour 2) identify some of the genes that control these sex differences in brain circuits 3) show that the processing of olfactory information has many more similarities to visual or auditory processing than previously realised 4) identify key principles of brain organisation for innate olfactory behaviour albeit after demonstrating a much greater complexity in these regions than previously appreciated 5) identify specific circuit motifs that allow the interaction between learned and innate behaviours. Taken together these results deepen our understanding of normal brain function in an impactful and influential model system. They also provide molecular and circuit explanations for the origins of different behavioural responses both between the sexes and between learned and innate behaviours. Future work will demonstrate how these organisational principles are conserved or elaborated in vertebrate brains. Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far In Frechter et al eLife 2019 we made a major advance in our field by demonstrating that the lateral horn consists of a large population of stereotyped odour encoders. This confirmed a long-standing hypothesis. Many olfactory systems (including mice) show a bifurcation in the brain into higher order projections that likely support learned vs unlearned behaviour with stereotyped vs non-stereotyped responses. However, our results remain the only unambiguous evidence for this model.The Drosophila "mushroom body" learning and memory centre has been the subject of intense investigation for almost 50 years. Dolan et al Neuron 2018 was a breakthrough since it produced the first specific circuit model for how learning can a) impact circuits in the rest of the fly brain to change behaviour and b) how learned behaviours can compete or suppress innate ones.In Dolan et al Neuron 2018 we identified neurons required for innate attraction. Complementing this in Huoviala et al 2020 we found higher olfactory neurons required for innate aversive behaviour and followed the relevant pathways all the way from the sensory periphery to descending neurons projecting to the nerve cord. Once again this was a first. Once again, we found new convergence motifs downstream of the mushroom body. A second breakthrough was the demonstration that the labelled line encoding of valence in the olfactory periphery breaks down at the level of third order neurons.Finally Galili et al 2022 demonstrated how and likely why flies can smell in stereo – enabling them to localise other flies in the dark. Stereo smell had been demonstrated in artificial scenario over 40 years ago, but no convincing explanation of how flies could do this in naturalistic settings was available until now. These and a range of other results demonstrated that processing of odours has much more similarity to visual or auditory processing than previously realised.Many of these advances depend on new circuit mapping methods. We have had key roles in developing Drosophila connectome resources (Zheng et al Cell 2018, Scheffer et al eLife 2020) and sharing them eg through the virtualflybrain.org website of which I am a PI. We have developed novel computational tools in the R language (see natverse.org) and shared them via github or the public CRAN website (>45,000 downloads over the last 5 years).Finally, we carried out many public dissemination activities. This ranged from local primary schools to the prestigious venue of the Royal Society, the UK's national academy. As the recipient of their Francis Crick medal, I delivered a lecture for ~500 members of the public in Jan 2020. The recorded lecture has been viewed 2700 times on youtube. Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far) We have also made major technological contributions in the course of the project. Our validation and extension work of the NBLAST algorithm (Costa et al Neuron 2016) demonstrated the general applicability of this method which has become a standard in fly neuroscience and also applied to vertebrate models (eg zebrafish, Kunst et al Neuron 2019 and mouse Gouwens et al Cell 2020).Adding to this in Dolan et al. Neuron 2018 we made a crucial new technical step based on the original NBLAST method, showing how neurons labelled and imaged using light microscopy could be cross-identified with a reference Electron Microscopy (EM) dataset revealing comprehensive connectivity information. This method has proven crucial for the rapid success of Drosophila EM connectomics since it allows experimentalists around the world to study the connectivity of their neurons of interest within the two available EM volumes.Schlegel, Bates et al eLife 2021 was one of the first three publications using a major new EM connectomics dataset, the 25000 neuron Drosophila "hemibrain". We developed a range of new analytic methods for EM connectomes such as a stochastic information flow analysis. By comparison with the first EM dataset of the fly brain (Zheng et al Cell 2018) we also demonstrated a surprising and practically important observation: by cell number and connectivity of the same brain were just as similar to each other as they were to a different individual. This suggests that many aspects of development play out stochastically and largely without influencing each other across the midline.These results have significantly advanced the state of the art for circuit analysis in Drosophila. Since the conclusion of OlfSwitch we have entered into new collaborations to extend these methods to mouse connectomics. Related documents Slide1.png Fly_brain.jpg
