Back

Modular presynaptic assemblages scale to postsynaptic partner number

Punal, V. M.; Thornton-Kolbe, E. M.; Dhillon, J.; Rogow, J. A.; Clowney, E. J.

2026-01-08 neuroscience
10.64898/2026.01.07.698224 bioRxiv
Show abstract

Behavioral diversification can arise through, and is constrained by, evolutionary and inter-individual differences in neural circuit development. Moreover, alteration of focal neural parameters changes the environment in which cells connect into circuits. In the mushroom body, an associative learning center of arthropods, the number of principal Kenyon cells varies widely across species and among individuals. How such variation is developmentally accommodated by projection neurons, which provide sensory input to Kenyon cells, is not understood. In Drosophila melanogaster, we previously demonstrated that projection neurons scale their presynaptic bouton number to Kenyon cell population size. Here, we identify the developmental mechanisms underlying this input flexibility. Boutons arise from projection neuron axonal collaterals; we find that a PNs collateral number is subtype-specific and serves as the substrate through which bouton number scales to Kenyon cell population size. Independent of projection neuron identity or Kenyon cell number, individual collaterals most often produce just one bouton, suggesting collaterals are modular cell biological bouton units. Developing projection neurons initially overproduce nascent collaterals in the early pupa. The set of nascent collaterals that mature and eventually bear boutons is conditional on Kenyon cell number, thereby executing scaling. Finally, early boutons bear filopodia that frequently contact neighboring PN processes, suggesting that bouton-bouton interactions contribute to shaping these structures.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.