This observation suggests that VEGF164 promotes axon outgrowth only from RGCs that express NRP1. To confirm that VEGF164 promotes RGC axon growth in a NRP1-dependent fashion, we used a function-blocking antibody specific for NRP1 (Fantin et al., 2010). Control experiments demonstrated that axon outgrowth in the absence of VEGF164 was not altered by isotype control IgG or NRP1 antibody and that outgrowth from ventrotemporal retina, where RGCs lack NRP1 expression, remained at baseline levels when VEGF164 was added together with control

IgG or NRP1 antibody (Figures 6E and 6F). In contrast, axon outgrowth from NRP1-positive dorsotemporal explants was increased significantly when VEGF164 was added together with IgG and this VEGF164-induced enhancement of growth was blocked completely TSA HDAC by the NRP1 antibody (Figures 6E and 6F). We conclude that VEGF164 promotes the growth of presumptive contralaterally projecting RGC axons through its receptor, NRP1. Previous studies demonstrated selleck a role for the NRP1 coreceptor FLK1 in axon regeneration after VEGF treatment of postnatal RGC explants (Böcker-Meffert et al., 2002). However, Flk1 was not expressed obviously in RGCs at E12.5 or E14.5, when they extend axons through the chiasm ( Figure S3B). Consistent with this finding, a previously validated function-blocking antibody that is specific for FLK1 and

blocks VEGF-A signaling in endothelial cells ( Gerhardt et al., 2003) did not inhibit the response of

RGC axons to VEGF164 ( Figures S4C and S4D). We conclude that VEGF164 signals through NRP1 in embryonic RGC axons independently of FLK1. To address if VEGF acts directly on RGC axons as aminophylline a guidance signal, we used the growth cone turning assay (Lohof et al., 1992). In this assay, a pipette is placed at an angle of 45° to the initial direction of axon extension, and test substances are puffed into the medium to establish a gradient. As expected, we found that growth cones from both ventrotemporal retina, which gives rise to NRP1-negative, ipsilaterally projecting RGCs, and dorsotemporal retina, which gives rise to NRP1-positive, contralaterally projecting RGCs, grew randomly in a gradient of PBS (Figures 7A–7F; mean turning angle of ventrotemporal axons: −0.1° ± 3.4°, n = 12; mean turning angle of dorsotemporal axons: 0.5° ± 5.1°, n = 10). Random growth of both ventrotemporal and dorsotemporal growth cones occurred also in a VEGF120 gradient (Figures 7C–7F and S5; mean turning angle of ventrotemporal growth cones: 3.5° ± 4.0°, n = 10; mean turning angle of dorsotemporal growth cones: −2.0° ± 2.3°, n = 9). We also found that VEGF164 did not induce significant turning of ventrotemporal growth cones (Figures 7C and 7D; mean turning angle: 5.9° ± 3.7°, n = 11). In contrast, dorsotemporal RGC growth cones were attracted strongly by a gradient of VEGF164 (Figures 7A, 7B, 7E, and 7F; mean turning angle: 21.5° ± 5.8°, n = 9, p < 0.01 compared to PBS).