Figure 7: Endocytosis is essential for activation of the PCP pathway and stability of vessels during retinal vascularization.

(a) Blockade of endocytosis by dynasore intraocular injection at P3 induces retinal vasculature regression quantified at P5. The number of branch points after isolectin fluorescein isothiocyanate (FITC) staining, in the superficial plexus was decreased and capillary regression was increased in retinas injected with dynasore when compared with retinas injected with DMSO (control), as quantified by the ratio of collagen IV-positive/isolectin FITC-negative branches to total branches. n=6 DMSO injected; n=6 dynasore injected. Unpaired t-test. Scale bars, 100 μm. (b) Aortic rings were implanted in Matrigel and the area of sprouting was quantified 72 h after plating. Aortic rings from P7 Pdzrn3^iECKO (iEC KO) mice produce significantly smaller and less-organized sprouts, as compared with aortic rings from Pdzrn3^iECWT (iEC WT) mice. Treatment of wild-type rings with dynasore reduced the number and length of sprouts while treatment of iEC KO rings with Pdzrn3 lentivirus (lenti Pdzrn3) restores their ability to form longer sprouts; n=7 iEC KO; n=7 iEC WT, sprouting surface area was measured per aorta. Unpaired t-test. Scale bars, 50 μm. (c) At P5, the Wnt canonical pathway is activated constitutively and the c-Jun-dependent PCP pathway repressed in iEC KO retina compared with iEC WT retina, as indicated by anti-active β-catenin (ABC) and anti p–c Jun immunoblot analysis in retina lysates. (d) The level of the active form of c-Jun is downregulated at P5 by blockade of endocytosis by intraocular injection of dynasore at P3, indicating that endocytosis is an important step for activation of the planar cell polarity signalling during retinal vascularization. The level of p–c-Jun was normalized using tubulin. Expression of ABC and PDZRN3 was unchanged under dynasore treatment in contrast to that in Pdzrn3 deletion (iEC KO versus iEC WT) conditions. NS, not significant.