New preprint on connectomic analysis of taste circuits

We have a new preprint on bioRxiv describing the organization of Drosophila taste circuits! This study was led by co-first authors Sydney Walker, an Emory undergraduate majoring in Quantitative Sciences, and Marco Peña Garcia, a neuroscience PhD student.

Our lab’s research focuses on how taste circuits in Drosophila transform sensory inputs into behavioral responses, such as promoting or suppressing food consumption. Work over many decades has given us a pretty good understanding of how sensory cells in the taste system work, including which tastes they respond to and what their response dynamics look like. But the next steps of taste processing have been super mysterious. Taste sensory cells send information into a region of the brain called the subesophageal zone (SEZ), but until recently we had no idea which cells they talk to and what downstream taste circuits look like. Which brain regions are involved in taste processing? Do circuits for processing different tastes overlap?

We can now answer these questions thanks to the recent release of a whole-brain connectome, a map of all the neurons and connections in the entire fly brain. We traced connections from four types of taste sensory neurons (called gustatory receptor neurons or GRNs) that each respond to a different taste: sugar, water, bitter, or salt + amino acids (termed “IR94e” neurons). We first identified the neurons that each type of GRN connects with, called second-order neurons or “2Ns”. As you can see from the picture below, almost all the 2Ns are located within the SEZ, which is the bottom part of the brain where the GRN projections are. The 2Ns for different taste modalities are largely non-overlapping, with the exception of overlap between sugar and water 2Ns. That’s probably because both sugar and water tastes promote ingestion, as they are nutrients that the fly needs, so they would need to activate some of the same pathways.

We found that the anatomical and functional properties of 2Ns vary by taste modality, such as whether they are excitatory or inhibitory neurons. We then traced connections from 2Ns to identify third-order neurons (“3Ns”), shown below. Compared to 2Ns, 3Ns showed way more overlap between taste modalities and way more projections outside the SEZ, with different 3Ns preferentially innervating different brain regions. We also found that 3Ns can receive either excitatory or inhibitory inputs, and many 3Ns are predicted to be inhibited rather than excited by taste.

The paper has a lot more analyses of the anatomical and functional properties of taste neurons, convergence of taste inputs within and across modalities, and comparing connectivity to predicted neuronal activity using computational simulations.

Our main takeaways are:

1) Pathways for processing different tastes remain largely separate at the second layer but begin to converge at the third layer, which would create neurons that have diverse taste response profiles rather than responding to just one taste.

2) Neurons that process different taste modalities have different anatomical and functional properties, including projections to different brain regions, which likely relate to the behavioral role of each taste.

3) Early stages of taste processing primarily occur within the SEZ, with an expansion to other brain regions at the third layer, and these include regions linked to feeding, olfactory processing, and learning.

We think that this study provides an overview of early taste processing in flies that will guide functional studies in the future. Let us know if you have any feedback on the paper!