Intron-targeted mutagenesis reveals roles for Dscam1 RNA pairing architecture-driven splicing bias in neuronal wiring

Abstract

Drosophila melanogaster Down syndrome cell adhesion molecule (Dscam1) can generate 38,016 different isoforms through largely stochastic, yet highly biased, alternative splicing. These isoforms are required for nervous functions. However, the functional significance of splicing bias remains unknown. Here, we provide evidence that Dscam1 splicing bias is required for mushroom body (MB) axonal wiring. We generate mutant flies with normal overall protein levels and an identical number but global changes in exon 4 and 9 isoform bias (DscamD4D À/À and DscamD9D À/À ), respectively. In contrast to DscamD4D À/À , DscamD9D À/À exhibits remarkable MB defects, suggesting a variable domain-specific requirement for isoform bias. Importantly, changes in isoform bias cause axonal defects but do not influence the self-avoidance of axonal branches. We conclude that, in contrast to the isoform number that provides the molecular basis for neurite self-avoidance, isoform bias may play a role in MB axonal wiring by influencing non-repulsive signaling.

Cis mutagenesis in vivo reveals extensive noncanonical functions of Dscam1 isoforms in neuronal wiring

“…Notably, except for a few exon 9s, this pattern of change in the frequency of most exon 9 variants in Dscam1 Δ9uD4 mutants was largely in contrast with the changing pattern in Dscam1 Δ9D mutants, which lack the downstream docking site (Figure S2E). (Hong et al, 2021) Thus, this splicing outcome of these mutant flies seems consistent with the regulatory model by dual docking sites and bidirectional base pairings. Interestingly, this led us to construct mutant flies encoding a comparable number of Dscam1 isoforms as WT albeit with global changes in exon 9 variant composition (Figure 1E).…”

“…Numerous studies demonstrate that Dscam1 functions to mediate dendrite patterning via homophilic repulsion in mechanosensory dendritic arborization (da) neurons. (Hattori et al, 2009;Hong et al, 2021;Hughes et al, 2007;Matthews et al, 2007;Soba et al, 2007) However, our present study shows that Dscam1 Δ9uD exhibit significant increases in dendritic branch number and total dendritic length without disrupting dendritic spacing or inter-neuronal repulsion. No obvious dendritic self-crossings were observed, and the overlapping between class I (vpda) and class III (v′pda) dendrites was not affected.…”

“…Purified products were pooled on Illumina MiSeq platform (Illumina, San Diego) according to the standard protocol by G-BIO.We calculated the relative expression levels of Dscam1 exon 9s according to the method as previously described. (Hong et al, 2021)…”

“…We have previously identified the docking site in the intron upstream of exon 10 of Drosophila Dscam1, which may pair with the selector sequence downstream of only a few exon 9 variants. (Hong et al, 2021;Yang et al, 2011) Considering the presence of dual docking sites and bidirectional base pairings in the exon 9 cluster of distantly related lepidopteran, coleopteran and hymenopteran Dscam1 (Figure S1A), (Dong et al, 2021;Yue et al, 2016) we speculated that an analogous docking site may be located in the intron upstream exon 9.1 of Drosophila Dscam1. However, sequence alignment revealed only relatively limited conservation across Drosophila (Figure S1B).…”

Hidden RNA pairings counteract the “first-come, first-served” splicing principle to drive stochastic choice in Dscam1 splice variants

“…A series of knockout mutant flies were constructed by targeted CRISPR-Cas9 with varying degrees of deletions of the docking site sequence and its downstream sequence (designated Dscam ΔDock1–ΔDock4 ). Cryptic exons upstream of exon 6.1 are indicated ( 43 ). ( B ) RT-PCR diagram from the head of WT and mutant flies.…”