Skip to main content
. 2014 Apr 22;3:e01699. doi: 10.7554/eLife.01699

Figure 7. EGFR mediates a probabilistic branch refinement process.

(A–B) EGFR localization examined by expressing UAS-EGFRGFP (green) in the DCNs (red, UAS-cherryRFP) during pupal development at (A) 56 hr APF and (B) 72 hr APF. EGFRGFP expression was observed in a punctate pattern in the cell bodies (insets in A and B) and along the axonal branches (A and B). Images A/B and A′/B′ were subjected to thresholding and merged (A‴/B‴). Differential localization results in branches with (A‴, arrowheads) and without (A‴, arrows) EGFRGFP at 56 hr APF, whereas most if not all branches contain EGFRGFP at 72 hr APF (B‴, arrowheads). High magnification shows EGFR localization at branches at 56 hr APF (A2) and 72 hr APF (B2). (C) Z-stack projections from live imaging time-lapse videos of control axons at around 40 hr APF between t0 = 0 min (C1) and t2 = 10 min (C3) with 5-min intervals. (D) Z-stack projections from live imaging time-lapse videos of EGFRDN axons at around 40 hr APF between t0 = 0 min (D1) and t2 = 10 min (D3) with 5 min intervals. Arrows indicate branches being pruned while arrowheads point to growing branches. (E) Visualization of growth (green) and retraction (purple) events between t0 = 0 min (C1) and t1 = 5 min (C2) in control. (F) Visualization of growth (green) and retraction (purple) events between t1 = 5 min (D2) and t2 = 10 min (D3) in EGFRDN. (G) Quantification of growth and retraction dynamics at branches using the tracer tool shows significant decrease in branch lengths in EGFRDN compared to control. Control (growth) 7.75 ± 2.65 (n = 8), EGFRDN (growth) 2.97 ± 0.56 (n = 9, p<0.001). Control (retraction) 7.4 ± 2.28 (n = 8), EGFRDN (retraction) 3 ± 1.08 (n = 8, p<0.001). Horizontal lines represent the mean for each data set. t test. ***p<0.001. The scale bars represent 20 µm.

DOI: http://dx.doi.org/10.7554/eLife.01699.021

Figure 7.

Figure 7—figure supplement 1. Localization of EGFR.

Figure 7—figure supplement 1.

(A) Expression of genomic EGFRGFP (green) during brain development at 48 hr APF and the neuropil marked with DN-cadherin (blue). (B) Axonal arbor of lateral neurons ventral (LNv) (green) in wild type and EGFRDN. (C–E) EGFR localization by expressing UAS-EGFRGFP (green) in the adult DCNs (red, UAS-cherryRFP) in one brain hemisphere (C), in the cell bodies (D) and in the axonal branches (E). The scale bars represent 20 µm except in C with 100 µm. (F) EGFR localization by expressing UAS-EGFRGFP (green) in the DCNs (red, UAS-cherryRFP) of two different individual flies during pupal development at 56 hr APF. Differential localization results in branches with (arrowheads) and without (arrows) EGFR localization at 56 hr APF. The scale bars represent 20 µm. (G) EGFR localization examined by expressing UAS-EGFRGFP (green) in the LNvs (red, UAS-lacZ) at (G) larval stage (L3) and in (G′) adult. The scale bar represents 60 µm.
DOI: http://dx.doi.org/10.7554/eLife.01699.022
Figure 7—figure supplement 2. DCN branch pattern in cultured pupal brains.

Figure 7—figure supplement 2.

(A) DCN branch morphology of a brain dissected from pupae at 48 hr APF and then cultured under standard conditions (‘Materials and methods’) for 48 hr. Morphology of the neuropils has been visualized by nc82 staining (magenta). (B–C) Axon branch morphology in the optic lobe of pupal brain dissected at 48 hr APF and cultured for (B) 24 hr and (C) 48 hr. (D) EGFR shows differential and dynamic localization in developing dorsal cluster neurons in vivo. UAS-CD8-RFP and UAS-EGFR-GFP were expressed with ato-Gal4 in wild-type Drosophila brains. Intact eye–brain complexes were imaged live at 45% APF. Maximum projection images demonstrating a single DCN axon terminal from live imaging time-lapse videos at t = 0 min (left) t = 18 min (middle) and t = 46 min (right), for both channels (upper) and only EGFR channel (lower). Two directly opposing branches of the same DCN axon were followed over time. At t = 0 both branches have significant levels of EGFR signal (arrow and arrowhead). 18 min later upper (arrow) branch retains its EGFR signal while lower (arrowhead) branch demonstrates a significant decrease. 28 min later the lower branch demonstrates a slight increase in the signal while the upper branch almost completely loses it. Scale bars correspond to 5 µm in all images.
DOI: http://dx.doi.org/10.7554/eLife.01699.023
Figure 7—figure supplement 3. UAS-EGFRGFP localizes and functions similar to endogenous EGFR.

Figure 7—figure supplement 3.

(A) Expression of UAS-EGFR-GFP in photoreceptor neurons at developing L3 eye disc using GMR-Gal4. (A′) Expression of endogenous EGFR revealed by immunohistochemistry on developing Canton-S L3 eye disc. The scale bars represent 20 µm. (B–D′) EGFR-GFP overexpression in the wing produces increase in vein tissue. (B) Control bearing Dpp-Gal4, showing the wild-type vein pattern, (B′) Zoom in of a ROI in (B). (C) Flies over-expressing wild-type-untagged EGFR in the wing using Dpp-Gal4 (along vein L3, black arrow) show a vein-specific increase in vein thickness (yellow arrow) and formation of ectopic veins (yellow arrowheads), (C′) zoom in of a ROI in (C). (D) Flies expressing EGFR-GFP using Dpp-Gal4 produces similar phenotypes to wild type EGFR over-expression (yellow arrow and arrowhead), (D′) zoom in of a ROI in (D).
DOI: http://dx.doi.org/10.7554/eLife.01699.024
Figure 7—figure supplement 4. Overexpression of wild-type EGFR does not cause a significant increase in axonal branching.

Figure 7—figure supplement 4.

Quantification of the number of axonal branches in DCNs overexpressing untagged or GFP-tagged EGFR. Neither causes a significant increase in the average number of axonal branches.
DOI: http://dx.doi.org/10.7554/eLife.01699.025