Robust discrimination between self and non-self neurites requires thousands of Dscam1 isoforms

Abstract

Down Syndrome cell adhesion molecule (Dscam) genes encode neuronal cell recognition proteins of the immunoglobulin superfamily. In Drosophila, Dscam1 generates 19,008 different ectodomains by alternative splicing of three exon clusters, each encoding half or a complete variable immunoglobulin domain. Identical isoforms bind to each other, but rarely to isoforms differing at any one of the variable immunoglobulin domains. Binding between isoforms on opposing membranes promotes repulsion. Isoform diversity provides the molecular basis for neurite self-avoidance. Self-avoidance refers to the tendency of branches from the same neuron (self-branches) to selectively avoid one another. To ensure that repulsion is restricted to self-branches, different neurons express different sets of isoforms in a biased stochastic fashion. Genetic studies demonstrated that Dscam1 diversity has a profound role in wiring the fly brain. Here we show how many isoforms are required to provide an identification system that prevents non-self branches from inappropriately recognizing each other. Using homologous recombination, we generated mutant animals encoding 12, 24, 576 and 1,152 potential isoforms. Mutant animals with deletions encoding 4,752 and 14,256 isoforms were also analysed. Branching phenotypes were assessed in three classes of neurons. Branching patterns improved as the potential number of isoforms increased, and this was independent of the identity of the isoforms. Although branching defects in animals with 1,152 potential isoforms remained substantial, animals with 4,752 isoforms were indistinguishable from wild-type controls. Mathematical modelling studies were consistent with the experimental results that thousands of isoforms are necessary to ensure acquisition of unique Dscam1 identities in many neurons. We conclude that thousands of isoforms are essential to provide neurons with a robust discrimination mechanism to distinguish between self and non-self during self-avoidance.

Figure 1. Thousands of isoforms are required for dendrites of dendritic arborization neurons to distinguish between self and non-self

a, Alternative splicing of Dscam1. The inset shows a schematic of Dscam1 isoform-specific homophilic binding. The crystal structure shows that variable domains pair in an antiparallel configuration,. Binding requires matching of variable domains; one mismatch (right, pink arrowheads) disrupts binding. b, Dscam1 alleles with reduced diversity. Alternative exons were replaced with cDNAs encoding exons 7–11 and 5–11 to generate Dscam1^576-isoform and Dscam1^12-isoform alleles, respectively (asterisks). Sequence modified is indicated with blue line. Dscam1^FRT is a control for Dscam1^576-isoform and Dscam1^12-isoform knock-in alleles, and contains an FRT site as a result of homologous recombination. Three different alleles were generated for each mutant. x.y.z refers to combinations of exon 4, 6 and 9, respectively, encoded by alleles. Dscam1^Δ4.1–4.3 and Dscam1^Δ4.4–4.12 were reported previously. c, Self-repulsion of dendritic arborization neurons requires Dscam1. Red arrowheads, self-crossing of class I (vpda) dendrites. d, Representative images of dendritic arborization neuron dendrites in different Dscam1 mutants. See Supplementary Fig. 3 for larger images. All neurons were visualized with anti-HRP antibody (magenta), and class I (vpda) neurons were labelled with GFP (green; appears white because they overlap with magenta). Class III (v′pda) neurons were traced from cell bodies (insets). Yellow arrowheads, crossing between class I and III dendrites. Scale bars, 50 μm. e, Self-repulsion is intact in all mutants with reduced diversity. Numbers in parenthesis denote class I neurons analysed for each genotype. Data in boxplot format (thick line, median; box, 25–75% quartiles; error bar, data within 1.5× quartile range; circles, outliers). ***P < 0.001; Student’s t-test. f, Increasing the number of available isoforms increases overlaps between class I and class III dendrites. Numbers in parenthesis, class I/III pairs analysed. Data in boxplot format. **P < 0.01, ***P < 0.001; Wilcoxon rank test. Thousands of isoforms are required for normal degree of overlap. For genotypes, see Methods.

Figure 2. Thousands of isoforms are required for mushroom body lobe development

a, Schematic of mushroom body (MB) development. The call-out circle represents a model of selective recognition between self-branches mediated by Dscam1 isoform-specific homophilic binding. Dscam1 isoforms are shown as coloured bars. b, Mushroom body lobe morphology in mutant animals, visualized with monoclonal antibody 1D4 (anti-FasII, red). Scale bar, 20 μm. c, Quantification of the mushroom body lobe phenotypes. The numbers in the parenthesis represent the number of mushroom bodies (that is, the number of brain hemispheres) examined for each genotype. See Methods for genotypes.

Figure 3. Branching of posterior scutellar neurons requires thousands of isoforms independent of their identity

a, Representative images of posterior scutellar (pSc) neuron axon branch patterns in homozygous Dscam1 mutants with different degrees of diversity (for Dscam1 deletion mutants over Dscam1^null alleles, see Supplementary Fig. 4). Scale bar, 50 μm. Right panel, branch segment numbers were assigned arbitrarily for scoring (scoring method, Supplementary Fig. 4). b, Average branch patterns for each genotype (for Dscam1 deletion mutants over Dscam1^null alleles, and raw data, see Supplementary Fig. 5). c, Pair-wise comparisons of overall branch patterns. P values were determined using permutation test with Bonferroni correction accounting for 30 different pair-wise comparisons (see Methods). Numbers in parenthesis indicate the number of animals analysed. See Supplementary Fig. 6 for raw statistical data. d–f, Branches that show difference in frequency between each pair are red. P values were calculated using Fisher’s exact test with Bonferroni correction accounting for 16 different branch segments. g, The result of partitioning cluster analysis (homozygous deletion mutants). Each individual sample was clustered based on its branching pattern. Each of the three clusters contains controls, Dscam1^Δ4.1–4.3 mutants, and Dscam1^Δ4.4–4.12 mutants in similar ratios. See Supplementary Fig. 6 for results for knock-in mutants. NS, not significant; *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 4. Mathematical model showing that thousands of isoforms are required to give rise to many unique Dscam1 identities

a, A schematic representation of mathematical modelling used to determine number of neurons that acquire unique Dscam1 identities. Tables of Q_x,i for each condition are shown in Supplementary Fig. 7a. b, The number of neurons that obtain unique identities at more than 95% likelihood with varying degrees of diversity. Both axes are in log scale. See Supplementary Fig. 7b for the condition in which no sharing of isoforms between neurons is allowed.