Back

Stochastic splicing and deterministic inclusion of exon variables promote diversification of Down Syndrome Cell Adhesion Molecule expression

Lassota, A.; Dix, T. C.; Soller, M.

2025-05-26 molecular biology
10.1101/2025.05.21.655277 bioRxiv
Show abstract

Mutually alternative splicing in Down Syndrome Cell Adhesion Molecule (Dscam) gene of arthropods generates extraordinary molecular diversity producing tens of thousands of isoforms. From three clusters of variable exons directing homophilic interactions, one single exon is selected. Homophilic repulsion of identical isoforms directs branching of axons running in neuronal tracts, and of dendrites for generation of overlapping dendritic fields through selection of different variables in neighbouring cells. Here, we investigate the spatial inclusion of Dscam variable alternative exons in Drosophila and honey bees using reporter genes and in situ hybridizations, respectively. In Drosophila, we find that Dscam variable clusters 4 and 9 splicing is not always productive in reporters, resulting in suppressed expression in optic lobes and variable expression across identical cells in salivary glands and photoreceptor fields. However, in photoreceptor neurons, we find repetitive inclusion of specific variables suggesting that stochastic expression is generated at the level of splicing of the variable cluster, but inclusion of variables follows a deterministic path. Likewise, we find in larval brains, inclusion of exon 4 and 9 variables in compartmentalised and repetitive patterns. In foraging honey bees, inclusion of exon 4 and 10 variables occurs in compartmentalized patterns differing between mushroom body lobes and individuals. This indicates that initial equal inclusion of variables is directed to compartmentalized inclusion through experience. These findings detail a new model of experience directed alternative splicing in Dscam incorporating stochasticity through splicing productivity and deterministic selection of individual isoforms.

Matching journals

The top 5 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.