Abstract
All centralized nervous systems possess modulatory neurons that arborize broadly across multiple brain regions. Such modulatory systems are critical for proper sensory, motor, and cognitive processing. How single modulatory neurons integrate into circuits within their target destination remains largely unexplored due to difficulties in both labeling individual cells and imaging across distal parts of the CNS. Here, we take advantage of an identified modulatory neuron in Drosophila that arborizes in multiple olfactory neuropils. We demonstrate that this serotonergic neuron has opposing odor responses in its neurites of the antennal lobe and lateral horn, first and second order olfactory neuropils respectively. Specifically, processes of this neuron in the antennal lobe have responses that are inhibitory and odor-independent, while lateral horn responses are excitatory and odor-specific. The results show that widespread modulatory neurons may not function purely as integrate-and-fire cells, but rather their transmitter release is locally regulated based on neuropil. As nearly all vertebrate and invertebrate neurons are subject to synaptic inputs along their dendro-axonic axis, it is likely that our findings generalize across phylogeny and other broadly-projecting modulatory systems.
Significance The centrifugal innervation of neuronal circuits is ubiquitous across centralized nervous systems. Such inputs often arise from modulatory neurons that arborize broadly throughout the brain. How information is integrated in such cells and how release from their distant terminals is regulated remains largely unknown. We show that a serotonergic neuron that innervates multiple stages of odor processing in Drosophila has distinct activity throughout its neurites, including opposite polarity responses in first and second order olfactory neuropils. Disparate activity arises from local interactions within each target region. Our results show that such neurons exhibit dendritic computation rather than somatic integration alone, and that examining local interactions at release sites is critical for understanding centrifugal innervation.
