|
Figure 1. nrp1 and plxna1, and their sema3 ligands, expressed in the Xenopus retina over the initial period of RGC dendritogenesis. (A-F) Lateral views of whole-mount embryos after RNA in situ hybridization with sema3, nrp1 and plxna1 antisense riboprobes at stage 35/36. (G-K) Transverse sections through the retina of stage 31/32 (G-I) and stage 35/36 (J,K) embryos with in situ hybridization performed directly on the tissue sections. (L) Immunolabeling of a stage 37/38 retinal section with an anti-Sema3a antibody. Arrowheads in G and L point to label for sema3a mRNA and protein, respectively. ba, branchial arches; cmz, ciliary marginal zone; e, eye; inl, inner nuclear layer; L, lens; onl, outer nuclear layer; ot, otic vesicle; pe, pigment epithelium; rgcl, retinal ganglion cell layer. Scale bar: in G, 50 μm for G-L.
|
|
Figure 2. Inhibiting Sema3 signaling in RGCs disrupts dendrite polarization. (A) Schematic of the stage 39 retina. The apical and basal surfaces are labeled, and shown are a representative RGC and amacrine cell, basal and apical to the inner plexiform layer (IPL), respectively. (B) Schematic of the electroporation method. (C-J) Eyes were electroporated at stage 28 with CS2-GFP alone (C,I), or with CS2-GFP and CS2-dnnrp1-mt (D,E,F-H,J). (C-E) Cells were visualized in sections by GFP epifluorescence (C,E) and/or by 9E10 immunoreactivity for the myc-tag (D, inset in panel E). A control GFP-expressing RGC (C) sends a thin axon down towards the vitreal surface (basal) and dendrites to the IPL. A stage 39 (D) and a stage 42 (E) dnnrp1-expressing RGC both exhibiting an unbiased morphology, with neurites extending from both apical and basal cell surfaces. Images are oriented with basal down and apical up. Axons are indicated with arrows, and dendrite-like neurites with arrowheads. Schematic drawings of the RGCs in C-E, along with additional examples, to capture the scope of phenotypes, are shown in C′-E′. (F-H) Immunolabeling of a dnnrp1-mt-expressing RGC with a myc antibody (F) and the axonal marker SMI-31 (G). Merge is shown at higher magnification in H. SMI-31 labels the axon (arrow) of the RGC but not a short, basal neurite or an apical dendrite (arrowheads). (I,J) GFP- (I) and dnnrp1- (J) expressing RGC axons visualized at stage 40 in a lateral view of a whole-mount brain by immunostaining with an anti-GFP (I) or an anti-myc (J) antibody, respectively. Pi, pineal gland; Tec, optic tectum. Scale bar: in C, 10 μm for C-E,F,G; in C, 5 μm for H. (K,L) Graphs of the mean number of apical and basal neurites of RGCs expressing dnnrp1 at stage 39 (K) and stage 42 (L). Errors bars represent s.e.m. For K, GFP n=94, dnnrp1 n=106; for L, GFP n=42, dnnrp1 n=57. Data was obtained from at least three independent data replicates. Groups were compared statistically using a two-tailed Mann-Whitney U non-parametric test (***P<0.001).
|
|
Figure 3. Blockade of Plxna1 signaling in RGCs disrupts dendrite polarization. (A-C) Stage 39 RGCs electroporated at stage 28 with mRNA encoding GFP (A), or GFP and dnplxna1 (B,C) were visualized in retinal sections by GFP epifluorescence (A) and/or by immunoreactivity for the vsv-tag (B,C). A control GFP-expressing RGC has a polarized morphology, with dendrites extending towards the inner plexiform layer (IPL), whereas RGCs expressing a truncated Plxna1 receptor have neurites that extend from both apical and basal surfaces of the cell. Images are oriented with basal down and apical up. Axons are indicated with arrows, and dendrite-like neurites with arrowheads. Scale bar: 10 μm. (D) Graph of the mean number of apical and basal neurites of RGCs expressing either GFP and/or dnplxn1 at stage 39. Errors bars represent s.e.m. GFP n=151, dnplxna1 n=189. Data was obtained from at least three independent data replicates. Groups were compared statistically using a two-tailed Mann-Whitney U non-parametric test (***P<0.001).
|
|
Figure 4. Sema3s can act directly on RGCs to impact dendritogenesis. (A-F) Eye primordia were dissected and dissociated at stage 32, and incubated for 24 hours with conditioned media from COS-7 cells transfected with empty vector as control (A-C), or with transgenes encoding Sema3a, Sema3d (D,E) or Sema3f (F). RGCs were identified by immunolabeling with an antibody that recognizes neurofilament associated antigen (NAA). As seen in the schematic (left), the morphology of the entire cell is revealed with a β-III-tubulin antibody (green), and dendrites (arrowheads) identified as short β-III-tubulin positive processes, showing little or no immunoreactivity with the NAA antibody. By contrast, the axon (arrows) is both β-III-tubulin and NAA positive. Displayed is the β-III-tubulin (A) and NAA (B) immunoreactivity for the control cell shown in phase contrast in C. The β-III-tubulin (D,F) and NAA (E) immunoreactivity for a Sema3d- (D,E) and a Sema3f- (F) exposed cell are presented. Scale bar: 10 μm. (G-J) Graphs showing quantification of the percentage of RGCs with primary dendrites (G), the mean number of primary dendrites/cell (H), the percentage of RGCs with dendrites that branch (I) and the mean number of dendrite branch points (J). Error bars represent s.e.m. N=5-7 independent experiments, with 20-50 RGCs assessed for each condition per experiment. Statistical differences between the groups were assessed by using a one-way ANOVA and a Dunnett’s post-hoc test (*P<0.05).
|
|
Figure 5. Sema3 overexpression disrupts RGC dendrite polarization. (A) Schematic of a stage 39 retina with a representative RGC and amacrine cell to indicate the retinal location and orientation of the photomicrographs shown in B-F and H, respectively. (B-F) Stage 39 RGCs expressing different transgenes. GFP fluorescence is shown in B and F, and myc immunoreactivity in C-E. F is the GFP fluorescence of the myc-positive cell in E. Drawings of the cells in B-E, along with additional RGC examples, to capture the scope of phenotypes, are shown in B′-E′. Arrows show the axons, and arrowheads, dendrite-like neurites. ipl/IPL, inner plexiform layer. Scale bar: 10 μm. (G,H) A Sema3f-expressing photoreceptor (pr), bipolar cell (Bp) and amacrine cell (Am). Scale bar: in B, 15 μm for G; in B, 10 μm for H. (I-K) Graphs showing the mean number of apically or basally oriented neurites for RGCs overexpressing Sema3d/f (I), Sema3a (J) or caplxna1 (K). For I, GFP n=155, Sema3d n=80, Sema3f n=72. For J, GFP n=64, Sema3a n=54. For K, GFP n=98, caplxna1 n=58. (L) Graph showing the mean number of apically or basally oriented neurites for GFP-positive RGCs nearer than (<2; n=48) or farther than (>2; n=8) two cell body diameters from a myc-positive/Sema3f positive cell. Error bars represent s.e.m. Statistical significance in I was determined using a Kruskal-Wallis non-parametric one-way ANOVA, followed by a Dunn’s multiple comparison test (***P<0.001), in J and K by a non-parametric Mann-Whitney U test (***P<0.001), and in L by a two-tailed, unpaired Student’s t-test (*P<0.05). Data were collected from two to four independent experiments.
|
|
Figure 6. Model of Nrp1 and Plxna1 function in dendrite polarization of RGCs. Schematic of the retina, and hypothesized gradients of Sema3a and Sema3f emanating from the lens and inner nuclear layer (INL), respectively. When these gradients are disrupted, by Sema3 overexpression or when RGCs expressing mutant Sema3 receptors are unable to sense the Sema3 gradients, RGC dendrites fail to polarize towards the apical surface of the cell.
|
