Interactions between the Ig-Superfamily Proteins DIP-α and Dpr6/10 Regulate Assembly of Neural Circuits

Abstract

Drosophila Dpr (21 paralogs) and DIP proteins (11 paralogs) are cell recognition molecules of the immunoglobulin superfamily (IgSF) that form a complex protein interaction network. DIP and Dpr proteins are expressed in a synaptic layer-specific fashion in the visual system. How interactions between these proteins regulate layer-specific synaptic circuitry is not known. Here we establish that DIP-α and its interacting partners Dpr6 and Dpr10 regulate multiple processes, including arborization within layers, synapse number, layer specificity, and cell survival. We demonstrate that heterophilic binding between Dpr6/10 and DIP-α and homophilic binding between DIP-α proteins promote interactions between processes in vivo. Knockin mutants disrupting the DIP/Dpr binding interface reveal a role for these proteins during normal development, while ectopic expression studies support an instructive role for interactions between DIPs and Dprs in circuit development. These studies support an important role for the DIP/Dpr protein interaction network in regulating cell-type-specific connectivity patterns.

Keywords: DIP/Dpr proteins; cell survival; connectome; development; layer-specific circuit; specificity; synapse.

Figure 1.. DIP-α, Dpr6, and Dpr10 Proteins Are Localized to Neuronal Processes

(A) Schematic of the adult Drosophila visual system. Cell types studied in this paper are shown. Adapted from Fischbach and Dittrich (1989). (B and C) Wild-type Dm4 and Dm12 clones labeled by Multicolor Flip-out (MCFO) using Dm4- and Dm12-specific GAL4 lines. Dm4 neuron terminals tile; Dm12 terminals overlap. (D) K_D for interactions (see Cosmanescu et al., 2018). (E–H) Developing medulla at 24 hr APF. (E–E”) Antibody staining of wild-type pupa as indicated. Yellow dotted line, cell body region in the medulla; green dotted lines, medulla neuropil. (F) Dm4 axons in the medulla neuropil. Image, maximum intensity projection from a 10 mm stack. Arrow, Dm4 target layer labeled by anti-chaoptin (red) and myrGFP-labeled Dm4 neurons (green). (G) Dm4 dendrites contacting Dpr10-protein-rich layer. Green, myrGFP-labeled Dm4 neurons; red, Dpr10 antibody staining. (H) L3 contacts Dm4 at 24 hr APF. Green, myrGFP labeled Dm4 neuron; red, a myr-tdTomato-labeled L3 growth cone. See also Figures S1–S3.

Figure 2.. DIP-α Is Required to Promote the Layer-Specific Patterning of Dm4 and Dm12 Axon Terminals

(A–D) Single Dm4 neurons labeled using MARCM. (E–H) Single Dm12 neurons labeled using MARCM. DIP-α ^het-homo: this mutation disrupts both heterophilic and homophilic binding. Isoleucine 83 was changed to aspartic acid (I83D). DIP-α ^homo: this mutant protein disrupts homophilic binding only. It is a double mutant with alanine 78 changed to lysine and asparagine 94 converted to aspartic acid (A78K+N94D) (see text and Cosmanescu et al., 2018). (I and J) Single Dm12 neurons labeled using MARCM at 35 hr APF. Blue and orange dotted lines indicate two different medulla layers. (K) Arbor coverage of Dm4 neurons within M3 (n: 11–16 neurons, except for DIP-α ^het-homo, which is 6. DIP-α mutant, p < 0.0001; DIP-α ^het-homo, p = 0.0005, unpaired t test). (L) Dm12 targeting defects to M8 (brains analyzed, 7–13; number of labeled neurons analyzed, 40–166). Note that all mutant neurons are generated in an otherwise wild-type background (i.e., heterozygous background). (K) and (L) are represented as mean ± SD. See also Figures S4–S6.

Figure 3.. Ectopic Expression of Dpr10 Promotes Mistargeting of Dm4 and Dm12 Terminals

(A–D”) Wild-type (A–A”); dpr6, dpr10 double mutant animals (B–B”); T4-GAL4-driven expression of UAS-Dpr10 in the medulla M10 layer, in a dpr6, dpr10 double mutant animal (C–C”); and T4-GAL4-driven expression of UAS-Dpr10 in the M10 layer, in a DIP-α, dpr6, dpr10 triple mutant animal (D–D”). (A–D) Schematic of targeting events at 24 hr APF. Red neuron, Dm4 expressing DIP-α; white neuron, Dm4 in DIP-α^null animal; blue neuron, expressing Dpr10; gray neuron, no expression of Dpr10. (E and E’) T4-GAL4 driving expression of a Dpr10 mutant (Dpr10^het) protein that does not bind to DIP-α does not promote Dm4 mistargeting. (F and F’) T4-GAL4 drives expression of UAS-Dpr10 in the medulla M10 layer, in an otherwise wild-type background. Note phenotype is not as strong as in a dpr6,dpr10 double mutant background (see C”). (A’–D’, A”–D”, E, E’, F, and F’) Images at 48 hr APF. Note layers at 48 hr are further apart than at 24 hr (schematically shown in A–D) due to intercalary growth of other neuronal processes. Red, myr-tdTomato-labeled Dm4; blue, anti-Dpr10 antibody; white arrowheads, Dm4 terminals in M3; white arrows, Dm4s mistargeted to M10. (G–J) Wild-type Dm12 neurons target to the M3 layer in adults (G); T4-GAL4 driving expression of UAS-Dpr10 in the medulla M10 layer, in an otherwise wild-type animal, causes Dm12 to mistarget to the M10 layer. Note phenotype is not as strong as in a dpr6,dpr10 double mutant background (H); expressing a Dpr10 isoform (Dpr10^het) that disrupts heterophilic binding does not cause Dm12 mistargeting (I); T4-GAL4 drives expression of UAS-Dpr10 in the M10 layer, in a DIP-α mutant animal; removing DIP-α suppresses mistargeting of Dm12 to the M10 layer induced by T4-Gal4 driven UAS-Dpr10 expression (J); green, myr-tdTomato- labeled Dm12; blue, anti-Dpr10 antibody; yellow arrowheads, Dm12 terminals at M3; yellow arrows, Dm12s mistargeted to M10. See also Figures S7 and S8.

Figure 4.. DIP-α Regulates the Number and Distribution of Presynaptic Sites

(A) MARCM-STaR to assess the distribution of Brp puncta (active zone marker) in identified neuron types. Schematic diagram for MARCM-STaR (for genotype, see below). Note that in (C), (D), (F), and (G), single mutant neurons in an otherwise wild-type background are visualized. (B–D) Dm4: wild-type (B), DIP-α mutant (C), and DIP-α^homo (D) Dm4 neurons labeled with MARCM-STaR. (E–G) Dm12: wild-type (E), DIP-α mutant (F), and DIP-α^homo (G) Dm12 neurons with presynaptic sites labeled with MARCM-STaR. (B–G) Red, myr-tdTomato; green, Brp puncta. Note in (F) that Dm12 neurons with processes in a deeper layer (M8) also accumulate Brp puncta consistent with elaborating synapses in an inappropriate layer. (H) Quantification of the total number of presynaptic sites per neuron at the M3 layer (Dm4: number of labeled neurons, 5–11; Dm12: number of labeled neurons, 8–12; ***p < 0.001; **p < 0.01). (I) Quantification of the density of presynaptic sites in the M3 layer (Dm4 group: number of terminals [N.B. one terminal/column], 16–36 [exception, 108 terminals were analyzed in DIP-α^homo], p < 0.0001; Dm12 group: number of labeled neurons, 6–10, p = 0.001; unpaired t test). (H) and (I) are represented as mean ± SD. The genotype for MARCM-STaR is DIP-α, FRT9–2/Act-GAL80, FRT9–2; hs-Flp:PEST/UAS-R, LexAOP-myrTdTom; Dm4, or Dm12-GAL4/Brp-RSR-smGFP-V5–2A-LexA. See also Figure S6.

Figure 5.. DIP-α Antagonizes Dm4 Apoptosis in Early Pupal Development

(A) Dm4 cell number in adult whole-animal wild-type or null mutants as indicated (n = 8–18 optic lobes; ***p < 0.0001, **p = 0.0003, unpaired t test). (B) Dm4 cell number at different developmental stages (n: 4–20 optic lobes). (C) Schematic of apoliner transgene. TM, trans-membrane domain; nls, nuclear localization signal; dotted line, caspase-sensitive cleavage site. (D) Dm4 neurons (%) in all animals analyzed that are positive for active caspase (nuclear eGFP) (n = 14, 13 optic lobes for DIP-α and wild-type). (E) Apoliner expression pattern in Dm4 neurons at 24 hr APF. Upon caspase activation, the apoliner sensor is cleaved and eGFP translocates to the nucleus, leaving mRFP at membranes. Arrow indicates a Dm4 cell body with nuclear GFP and cytoplasmic RFP, while arrowheads point to Dm4 cell bodies with membrane-tethered eGFP and mRFP. (F) Dm4 neurons with (arrow) or without (arrowhead) hid-GFP expression in wild-type or DIP-α null animals as indicated. (G) Percentage of Dm4 neurons that are positive for the expression of hid-GFP or grim^MiMIC in wild-type or DIP-α null animals (n: for the hid-GFP group, 26 and 28 optic lobes; for grim^MiMIC group, 8 and 6 optic lobes; ***p = 0.0007, unpaired t test). (H) Dm4 cell number in adult wild-type and DIP-α mutants is increased in animals heterozygous for hid or H99. The H99 deficiency removes four pro- apoptotic genes (hid, rpr, grim, and skl) (n: 5–16 optic lobes, **p = 0.0023, 0.0002, 0.0001 from left to right; ***p < 0.0001; unpaired t test). (A), (B), (D), (G), and (H) are represented as mean ± SD. See also Figures S9–S11.

Figure 6.. Ligand Binding to DIP-α Regulates Cell Survival

(A) Dm4 cell number in DIP-α mutant alleles defective for heterophilic and homophilic binding (DIP-α ^het+homo) or homophilic binding only (DIP- α ^homo) (n: 6–10 optic lobes; ***p < 0.0001, unpaired t test). (B) Total Dm4 cell number in dpr10 mutant allele defective for heterophilic binding. Note with the exception of wild-type, all other animals were null for dpr6 (n: 6–13 optic lobes; ***p < 0.0001, unpaired t test). (C) Total Dm4 cell number in wild-type animals and dpr6,dpr10 double mutant animals with and without expression of Dpr6 or Dpr10 transgenes in different neuronal populations, as indicated (n: 6–18 optic lobes; ***p < 0.0001, unpaired t test). (D) Domain structure and binding affinity of DIP- α ^Nectin1 and Dpr10^Nectin3 chimeras. Homodimerization K_Ds were determined by analytical ultracentrifugation and the K_D for heterophilic binding was determined by SPR. (E) Total Dm4 cell number in DIP-α^Nectin1 knockin flies, and in knockin flies expressing Dpr10^Nectin3 driven by the combined activity of dpr6-GAL4 and dpr10-GAL4 (n: 10–24 optic lobes; ***p < 0.0001, unpaired t test). (F) Total Dm4 cell number in rescue experiments using UAS-DIP-α driven by various GAL4 drivers in dpr6,dpr10 double mutants. Note that although loss of DIP-α homophilic binding does not lead to a reduction in cell number (see A), ectopic expression of DIP-α can rescue cell survival in animals lacking Dpr6 and Dpr10 through homophilic binding (n = 8–15 optic lobes; ***p < 0.0001, unpaired t test). Cell numbers for (A) and (B) were quantified as adults and (C), (E), and (F) in pupae at 48 hr APF. (A)–(C), (E), and (F) are represented as mean ± SD. See also Figures S6, S12, and S13.