Plexin a-semaphorin-1a reverse signaling regulates photoreceptor axon guidance in Drosophila

Abstract

While it is well established that Semaphorin family proteins function as axon guidance ligands in invertebrates and vertebrates, several recent studies indicate that the Drosophila Semaphorin-1a (Sema1a), a transmembrane Semaphorin, can also function as a receptor during neural development. The regulator of Sema1a reverse signaling, however, remains unknown. In this study, we show that like Sema1a, the well known Semaphorin receptor Plexin A (PlexA), is required for the proper guidance of photoreceptor (R cell) axons in the Drosophila visual system. Loss of PlexA, like loss of semala, disrupted the association of R-cell growth cones in the optic lobe. Conversely, overexpression of PlexA, like overexpression of sema1a, induced the hyperfasciculation of R-cell axons. Unlike Sema1a, however, the cytoplasmic domain of PlexA is dispensable. Epistasis analysis suggests that PlexA functions upstream of semala. And PlexA and sema1a interact genetically with Rho1. We propose that PlexA regulates Semala reverse signaling in the Drosophila visual system.

Figure 1.

PlexA is required for the proper formation of R1–R6 termination layer in the developing optic lobe. Third-instar larval eye–brain complexes were stained with MAb 24B10 to visualize R-cell axonal projection pattern. A, Wild type. R1–R6 axons stop at the intermediate target region in the lamina (la), where their growth cones expand and associate closely with each other to form a smooth and dense terminal layer. Whereas R7 and R8 axons project through the lamina into the medulla (me). B, sema1a^P1 homozygote. R1–R6 terminal layer was disrupted. R1–R6 growth cones associated loosely with neighboring growth cones and failed to pack into a dense layer. C, D, A similar phenotype was observed in PlexA deficiency mutants (C) and eye-specific PlexA knockdown mutants (D). E–H are enlarged views of the boxed regions in A–D, respectively. Scale bar: A–D, 10 μm; E–H, 5 μm.

Figure 2.

Eye-specific expression of the UAS-PlexA-RNAi transgene effectively decreased the level of PlexA in R cells. Third-instar eye discs were stained with anti-PlexA antibody. A, Wild-type. Strong PlexA staining was detected in developing R-cell clusters in the posterior region of the eye disc. B, PlexA staining was significantly reduced when the level of PlexA was knocked down by expressing a UAS-PlexA-RNAi transgene under control of the eye-specific GMR-GAL4 driver. Scale bar, 5 μm.

Figure 3.

PlexA interacts genetically with sema1a. R-cell axonal projection pattern in third-instar larval eye–brain complexes was visualized with MAb 24B10 staining. A, Wild type. B, Normal R-cell projection pattern was observed in sema1a heterozygotes. C, Larvae in which the level of PlexA was specifically knocked down in the eye displayed a mild phenotype. D, Reducing the dosage of sema1a by 50% in PlexA knockdown larvae increased both penetrance and severity of the phenotype. Scale bar, 20 μm.

Figure 4.

Overexpression of PlexA induced the hyperfasciculation of R-cell axons. A, Wild type. B, Overexpression of Sema1a in R-cell axons induced the formation of thicker R-cell axonal bundles between lamina and medulla. In medulla, R-cell axons formed large clumps. C, Overexpression of PlexA caused a similar hyperfasciculation phenotype. Scale bar, 20 μm.

Figure 5.

The cytoplasmic domain of PlexA is dispensable. A, Wild type. B, Overexpression of PlexA under control of GMR-GAL4 caused the formation of thicker axonal bundles. C, Overexpression of PlexA^Δcyt driven by GMR-GAL4 caused an identical hyperfasciculation phenotype. D, Western blot analysis using anti-HA antibody showed the expression of HA-tagged full-length and cytoplasmic-domain-truncated PlexA in flies under control of GMR-GAL4. The size of transgenic proteins is consistent with the predicted size. E, An eye-specific PlexA knockdown mutant. F, Expression of PlexA^Δcyt largely restored the R1–R6 growth-cone organization pattern in the lamina in PlexA knockdown mutants. Scale bar, 20 μm.

Figure 6.

PlexA functions upstream of sema1a. A, Wild type. B, Overexpression of PlexA under control of GMR-GAL4-induced R-cell axonal hyperfasciculation. C, The PlexA-induced hyperfasciculation phenotype was suppressed when sema1a was disrupted. D, The number of separate axonal bundles that are located between lamina and medulla was counted. The data were normalized with the row number of R-cell clusters in the eye disc. Compared with wild type, overexpression of PlexA induced the formation of thicker bundles and thus significantly decreased the number of separate R-cell axonal bundles (p = 0.0012). Compared with that of PlexA overexpression in wild-type background, the number of separate axonal bundles in PlexA-overexpression mutants in which the sema1a gene was disrupted, was increased significantly (p = 0.0005). Error bars denote SE. Scale bar, 20 μm.

Figure 7.

Sema1a and PlexA interact genetically with Rho1. A, Overexpression of Sema1a induced the formation of thicker axonal bundles. B, Reducing the dosage of ena by half did not modify the Sema1a-induced hyperfasciculation phenotype. C, Reducing the dosage of Rho1 by half significantly enhanced the Sema1a-overexpression phenotype. D, Reducing the dosage of Rho1 by half in wild-type background did not affect R-cell projection pattern. E, Expressing the dominant-negative form Rho1.N19 in R cells also induced an axonal hyperfasciculation phenotype. F, Knocking down PlexA suppressed the Rho1.N19-induced phenotype. Scale bar, 20 μm.