Fig 2: Transplantations confirm the growth promoting-effect in kif5aa mutant tecta.(A) Representative pictures of single in vivo imaged RGC axons after blastula stage transplantions from wild-type donors into a wild-type tectum (left panel), from kif5aa mutants into a wild-type tectum (middle panel) or from a wild-type donor into a kif5aa mutant tectum (right panel). The same cell was analyzed at 5 dpf (upper panel) and 7 dpf (middle panel). Scale bars = 20 μm. Schematics of RGC arbor complexity and size in the lower panel. In orange: Ntf3 overexpressing kif5aa mutant tectum. D = dorsal, V = ventral, R = rostral, C = caudal. (B) Quantification of total branch length of transplanted RGC axons at 5 and 7 dpf. Kif5aa mutant cell arbors are significantly smaller than wild-type cell arbors when growing into a wild-type tectum (p < 0.01). Wild-type cells built larger arbors when growing into a kif5aa mutant tectum (p < 0.05) (5 dpf: n = 14, 35, 6; 7 dpf: n = 14, 19, 5). (C) Schematic illustrating growth behavior of RGC axons in wildtype (upper panel) and kif5aa mutant tecta (lower panel) and upon loss of TrkC signaling (middle panel). Wild-type RGCs start to grow into the wild-type neuropil at 3 dpf. They grow highly active filopodial protrusions and start to form complex axonal arbors. At 5 dpf they reach their final size and maintain their branch shape at 7 dpf. When TrkC signaling is blocked by overexpression of a dominant negative receptor (ntrk3adN-GFP), wild-type cells show a substantially reduced arbor complexity (middle panel). Kif5aa mutant RGC arbors show a delay of ingrowth into the tectal neuropil. This is followed by a period of highly active growth with abundant filopodia formation. This results in highly complex arbors at 7 dpf. The delay of RGC growth is cell autonomous (Figure 8—figure supplement 1). The lack of retinal input leads to an upregulation of ntf3 expression by tectal cells and constitutes a growth-promoting environment.DOI: http://dx.doi.org/10.7554/eLife.05061.023

Fig 3: Silencing of all Retinal Ganglion Cells by BoTx expression leads to Ntf3 upregulation.(A) Expression of botulinum toxin light chain B (Brunger et al., 2008; Nevin et al., 2008) in all RGCs leads to expansion of melanosomes similar to blind mutant fish as shown in Figure 1—figure supplement 1. Scale bar = 200 μm. (B) Relative expression levels of ntf3 in 4-dpf old wildtype, kif5aa mutant, and Tg(Isl2b:Gal4, UAS:BoTxLCB-GFP) transgenic embryos. ntf3 is upregulated to a similar extent in Tg(Isl2b:Gal4, UAS:BoTxLCB-GFP) transgenic embryos as in kif5aa mutants. (C) Western blotting confirms the overexpression of Ntf3 protein in embryos without presynaptic activity.DOI: http://dx.doi.org/10.7554/eLife.05061.020

Fig 4: Blumenkohl mutant RGC arbors and RGC arbors growing into a Ntf3 overexpressing tectum do not show increased filopodia dynamics.(A) Upper panel: Axonal arbor of a single blumenkohl mutant RGC at 5 and 7 dpf. (indicated by an arrow). D = dorsal, V = ventral, R = rostral, C = caudal. Lower panel: Tracings of an axonal arbor at time point zero. In red: Overlay of filopodia formed and retracted within 10 min (1 frame/2 min). Scale bars = 20 μm. (B) Upper panel: Axonal arbor of a single RGC axonal arbor growing into a Ntf3 overexpressing tectum (compare Figure 7) at 5 and 7 dpf. Lower panel: Tracing of an axonal arbor at time point zero. In red: Overlay of filopodia formed and retracted within 10 min (1 frame/2 min). Scale bars = 20 μm. (C) Quantification of filopodia numbers formed and retracted within 10 min per cell at 5 and 7 dpf. No increased rate of filopodia formation can be observed in blumenkohl mutant RGC arbors and arbors of RGCs growing into a Ntf3 overexpressing tectum while kif5aa mutant axonal arbors form and retract significantly more filopodias. Graphs for wild-type and kif5aa mutant RGC arbors are identical to Figure 7.DOI: http://dx.doi.org/10.7554/eLife.05061.022

Fig 5: Transplantation of kif5aa mutant RGCs into a blumenkohl mutant acceptor leads to an increased growth compared to transplantation into a wild-type acceptor.(A) Representative pictures of single in vivo imaged RGC axons after blastula stage transplantions from kif5aa mutant donors into a wild-type tectum (left panel) and from kif5aa mutants into a blumenkohl mutant tectum (right panel). The same cell was analyzed at 5 dpf (upper panel) and 7 dpf (middle panel). Scale bars = 20 μm. Schematics of RGC arbor complexity and size in the lower panel. In orange: Ntf3 overexpressing blumenkohl mutant tectum. D = dorsal, V = ventral, R = rostral, C = caudal. Pictures in the left panel are identical to Figure 8. (B) Quantification of total branch length of transplanted RGC axons at 5 and 7 dpf. The reduced size of kif5aa mutant axonal arbors when growing into a wild-type tectum is partially rescued when transplanted into a blumenkohl mutant environment (p < 0.05) (5 dpf: n = 35, 8; 7 dpf: n = 19, 8). The graph for kif5aa mutant cells into a wild-type host is identical to Figure 8.DOI: http://dx.doi.org/10.7554/eLife.05061.025