Retinal explants from E8 chick embryos were cultured on laminin (A, E), E-cadherin (B, F), N-cadherin (C, G) or R-cadherin (D, H) substrates in the presence of antibody control (A–D) or adhesion blocking antibodies to chick N-cadherin (E–H). Antibodies against chick N-cadherin that recognize N-cadherin expressed on the surface of the chick retinal explant inhibited neurite outgrowth on human N-cadherin (G) and human R-cadherin (H) substrates, whereas they had no effect on neurite outgrowth on laminin (E) or human E-cadherin (F) substrate. Scale bar, 200 μm. (I) E-cadherin-Fc, N-cadherin-Fc, R-cadherin-Fc or E8 chick retina lysate was separated by SDS-PAGE, transferred to nitrocellulose membrane and probed with antibodies to chick N-cadherin (NCD-2), intracellular human E-cadherin, intracellular human N-cadherin, chick R-cadherin (RCD-2), human Fc, extracellular human E-cadherin, extracellular human/chick N-cadherin or extracellular human R-cadherin. E-cadherin, N-cadherin and R-cadherin protein migrates at ~120 kDa, ~130 kDa or ~125 kDa respectively.

DiI labeling of neurites on a laminin substrate (white bars) demonstrates the morphology of a small growth cone with few filopodia (A). Growth cones on an E-cadherin substrate (black bars) were very large and broad with few filopodial processes (B). On an N-cadherin substrate (hatched bars) growth cones were moderate with more filopodial processes (C). Growth cones on an R-cadherin substrate (gray bars) have hybrid morphology of broad growth cones with several filopodia (D). Representative images are shown. Arrowheads indicate the location of filopodia that were counted. The average number of filopodia per growth cone (E) and growth cone area (F) was quantitated for each substrate. Scale bar, 10 μm.

E8 chick retinal explants were cultured on laminin (A, D, G, J), E-cadherin (B, E, H, K), or N-cadherin (C, F, I, L) substrates and infected with HSV encoding GFP control (A–C), Cdc42 DN (D–F), Rac1 DN (G–I) or RhoA DN (J–L). Infection Cdc42 DN, Rac1 DN or RhoA DN had no effect on laminin (D, G, J) when compared to control (A). Infection with Cdc42 DN and RhoA DN also had no effect on E-cadherin-mediated neurite outgrowth (E, K). However perturbation of E-cadherin-mediated neurite outgrowth was observed when infected with Rac1 DN (H) as compared to control (B). Infection with Cdc42 DN or Rac1 DN decreased neurite outgrowth on an N-cadherin substrate (F, I) when compared to control (C). Infection with RhoA DN had no effect on N-cadherin-mediated neurite outgrowth (L). Neurite length (M) and neurite density (N) were quantitated in the presence of GFP (white bars), Cdc42 DN (black bars), Rac1 DN (hatched bars) or RhoA DN (gray bars). The asterisk denotes statistically significant changes in neurite length (p ≤ 0.001) or density (p ≤ 0.001, p ≤ 0.01 for Rac1 DN on N-cadherin) compared to control. Scale bar, 200 μm.

E8 chick retinal explants were cultured on laminin (A, E), E-cadherin (B, F), N-cadherin (C, G), or R-cadherin (D, H) in the presence of control (A–D) or Rac1 inhibitor (E–H). Cultures incubated with Rac1 inhibitor had no effect on laminin-dependent neurite outgrowth (E) when compared to control (A). Perturbation of neurite outgrowth was observed on E-cadherin (F), N-cadherin (G) and R-cadherin (H) in the presence of the Rac1 inhibitor when compared to vehicle control (B–D). Neurite length (I) and neurite density (J) were quantitated in the presence of the Rac1 inhibitor (black bars) or control (white bars). The asterisk denotes statistically significant changes in neurite length or density (p ≤ 0.001) compared to control. Scale bar, 200 μm.

E8 chick retinal explants were cultured on laminin (A, E, I, M), E-cadherin (B, F, J, N), N-cadherin (C, G, K, O) or R-cadherin (D, H, L, P) in the presence of either DMSO control (A–D), Rottlerin (E–H), Gö6976 (I–L) or Gö6983 (M–P). No effect on laminin-dependent neurite outgrowth was observed in the presence of any inhibitor (E, I, M) when compared to DMSO control (A). Perturbation of neurite outgrowth was observed on an E-cadherin substrate when cultures were incubated in the presence of Rottlerin (F) or Gö6983 (N), compared to DMSO control (B). Incubation with Gö6976 (J) had no effect on E-cadherin-mediated neurite outgrowth. On an N-cadherin substrate Rottlerin (G), Gö6976 (K), or Gö6983 (O) had no effect on neurite outgrowth when compared to DMSO control (C). A reduction in neurite outgrowth was observed on an R-cadherin substrate in the presence of Rottlerin (H), Gö6976 (L) or Gö6983 (P) when compared to DMSO control (D). Neurite length (Q) and neurite density (R) were quantitated in the presence of Rottlerin (black bars), Gö6976 (hatched bars), Gö6983 (gray bars) or DMSO control (white bars). The asterisk denotes statistically significant changes in neurite length (p ≤ 0.005) or density (p ≤ 0.001, p ≤ 0.05 for R-cadherin) compared to control. Scale bar, 200 μm.

Chick retinal explants from E6, E7, E8, E9 and E10 were isolated and cultured on an R-cadherin substrate (A). E8 chick retina explants were also sectioned parallel to the optic fissure and explants from each region of the retina were cultured on R-cadherin (B) or N-cadherin (C) substrates. Images were acquired from a location corresponding to the outer third of each explant. Each number indicates the explant number (e.g. 1 and 6 are most peripheral). Dorsal (D), ventral (V), nasal (N), temporal (T). Scale bar, 200 μm.

E8 retinal explants from the dorsal-nasal or ventral-nasal region of the retina were cultured on laminin (A), N-cadherin (B) or R-cadherin (C) substrates in the presence of a control antibody or NCD-2. NCD-2 had no effect on laminin (A). Neurite outgrowth from both the dorsal-nasal region and ventral nasal region of the retina was blocked in the presence of the chick specific NCD-2 antibody on an R-cadherin substrate. On an N-cadherin substrate, NCD-2 blocked neurite outgrowth from the dorsal-nasal and ventral-nasal region of the retina. Scale bar, 200 μm.

E8 chick retinal explants were cultured on laminin (A, E), E-cadherin (B, F), N-cadherin (C, G) or R-cadherin (D, H) substrates in the presence of 6.5 μM scrambled control (SPTPμ-Tat) (A–D) or PTPμ wedge peptide (WPTPμ-Tat) (E–H). Cultures incubated with WPTPμ-Tat had little to no effect on laminin-dependent neurite outgrowth (E) when compared to SPTPμ-Tat control (A), however neurite outgrowth decreased in the presence of WPTPμ-Tat on E-cadherin (F), N-cadherin (G) and R-cadherin (H) when compared to SPTPμ-Tat control (B–D). Neurite length (I) and neurite density (J) were quantitated in the presence of WPTPμ-Tat (black bars) or SPTPμ-Tat control (white bars). The asterisk denotes statistically significant changes in neurite length or density (p ≤ 0.001) compared to control. Scale bar, 200 μm.

E8 chick retinal explants were cultured on laminin (A, C, E, G) or R-cadherin (B, D, F, H) in the presence of control (A, B), dominant negative Cdc42 (C, D), dominant negative Rac1 (E, F) or the RhoA inhibitor (C3 transferase) (G, H). Transduction with all three Rho GTPases had no effect on laminin (C, E, G) compared to control (A). Dominant negative Cdc42 (D) and Rac1 (F) decreased neurite outgrowth on an R-cadherin substrate compared to control (B). The RhoA inhibitor had no effect on R-cadherin-mediated neurite outgrowth (H). Neurite length (I) and neurite density (J) were quantitated in the presence of control (white bars), Cdc42 DN (black bars), Rac1 DN (hatched bars) or RhoA inhibitor (gray bars). The asterisk denotes statistically significant changes in neurite length (p ≤ 0.001) or density (p ≤ 0.001) compared to control. Scale bar 200 μm.

E8 chick retinal explants were cultured on laminin (A, E), E-cadherin (B, F), N-cadherin (C, G) or R-cadherin (D, H) in the presence of scrambled control (A–D) or IQGAP1 inhibitor (E–H). Cultures incubated with IQGAP1 inhibitor had no effect on laminin (E) or E-cadherin-dependent neurite outgrowth (F) when compared to scrambled control (A, B). Decreased neurite outgrowth was observed on an N-cadherin (G) and R-cadherin (H) substrate when compared to scrambled control (C, D). Neurite length (I) and neurite density (J) were quantitated in the presence of IQGAP1 Inhibitor (black bars) or scrambled control (white bars). The asterisk denotes statistically significant changes in neurite length (p ≤ 0.001) or density (p ≤ 0.001, p ≤ 0.01 for R-cadherin) compared to control. Scale bar, 200 μm.

Model summarizing the potential signaling pathways involved in E-cadherin, N-cadherin and R-cadherin-mediated neurite outgrowth. See discussion for details.