Son J , Park MS , Park I , Lee HK , Lee SH , Kang B , Min BH , Ryoo J , Lee S , Bae JS , Kim SH , Park MJ , Lee HS .

	Fig. 1. Interaction between Pick1 and ephrinB1. (A) Full-length Pick1 fused to a C-terminal HA epitope tag (10 ng Pick1-HA) and full-length Xenopus ephrinB1 harboring a Flag epitope tag at the C-terminus (10 ng ephrinB1-Flag) were co-expressed in Xenopus oocytes for Co-IP/western blot analysis. Oocyte lysates were immunoprecipitated with Flag antibody and western blots were performed with the indicated antibodies. (B) Xenopus oocytes were each injected with 10 ng RNAs as indicated and lysates were subjected to Co-IP/western blot analysis. Co-expression with constitutively active FGFR1 K562E (KE) resulted in loss of binding between Pick1 and ephrinB1. However, co-expression with kinase dead FGFR1 C289R/K420A (KD) had no effect on the Pick1 and ephrinB1 interaction. (C) Xenopus oocytes were each injected with 10 ng RNAs as indicated and lysates were subjected to Co-IP/western blot analysis. Even co-expression of mutant ephrinB1 on PBM (Y324F, Y325F, Y324.5F) with constitutively active FGFR1 KE resulted in no effect on the Pick1 and ephrinB1 interaction.
	Fig. 2. Pick1 knock-down causes defects in ventral trunk formation and epidermal cell dissociation. (A) Embryos injected with Pick1MO (10 ng, 20 ng, 40 ng) into both blastomeres at the 2-cell stage. Pick1MO embryos showed bent and shorter body axis, inhibition of eye development, and reduced brain size. The ventral cell dissociation is shown (white circle). (B) Abnormal phenotypes including the ventral lesion caused by the loss of Pick1 significantly rescued by co-injection with 200 pg of 4MT Pick1 mRNA. (C) The histogram denotes the percentage of embryos with defects as described in (B). Note: Pick1 MO-injected embryos showed significant developmental defected embryos (ventral lesion, 94.0 ± 1.41%; eye defect, 78.0 ± 3.54%; bent body, 94.0 ± 1.41%; short body, 14.0 ± 8.49%; n = 60). Co-injection of Pick1MO and 4MT Pick1 into the embryos significantly restored the phenotypic defects as compared to Pick1MO alone (ventral lesion, 6.0 ± 0.71%; eye defect, 4.0 ± 1.41%; bent body, 14.0 ± 4.95%; short body, 12.0 ± 2.83%; n = 55), while over-expression of Pick1 showed minor defects such as a slightly bent axis (bent body, 8.0 ± 2.83%; n = 55). (D) 4MT Pick1-injected embryos were tested for expression. Western blotting was performed using an anti-Flag antibody.
	Fig. 3. Pick1 knock-down causes the disruption of AJs. Embryos were injected with or without 20 ng control MO, 20 ng Pick1 MO, and 200 pg 4MT Pick1 RNA in both the blastomeres of 2-cell stage embryos and those collected at stage 10.5. Immunostaining was performed with phospho-E-cadherin and Flag antibodies on cryosections. (A) Experimental scheme. Schematic showing the region of embryo isolated and examined by immunofluorescence microscopy. Adapted from Lee HS et al. [22]. (B) RT-PCR showed that Pick1 expression was significantly down-regulated, while ephrinB1 was not changed in Pick1MO-injected embryos. The histogram shows the relative intensity of Pick1 mRNA expression. (C) ControlMO-injected embryos showed stable localization of phosphorylated-E-cadherin along cell–cell boundaries in cell junctions. (D) Localization of phosphorylated-E-cadherin was greatly diminished in AJs of Pick1MO-injected embryos. (E) The redistribution of phosphorylated-E-cadherin by Pick1MO was significantly rescued by appropriate amounts of 4MT Pick1 RNA. (F) The localization of Pick1 partially overlapped with that of phosphorylated-E-cadherin, indicating that Pick1 can be present in AJs of epithelial cells and over-expressed Pick1 does not affect cell junction formations. Embryos were injected with 20 ng controlMO, 20 ng Pick1MO, and 200 pg 4MT Pick1 mRNA in both the blastomeres of 2-cell stage embryos. 1st row (Red: phosphorylated-E-cadherin), 2nd row (green: Pick1), 3rd row (merged). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
	Fig. 4. Pick1 repairs ephrinB1-induced disruption of AJs. Embryos were injected with or without 400 pg ephrinB1 and 500 pg Pick1 RNAs in both the blastomeres of 2-cell stage embryos and those collected at stage 10.5. Immunostaining with phospho-E-cadherin, Flag, and HA antibodies was performed on cryosections. (A) Lysates from embryos were tested for the expression of ephrinB1 and Pick1. Western blot was performed using anti-HA and anti-Flag antibodies. (B) Control embryos showed stable localization of phosphorylated-E-cadherin along cell–cell boundaries in cell junctions. (C) The over-expressed ephrinB1 embryo with disrupted cell–cell junctions. (D) The redistribution of phosphorylated-E-cadherin by ephrinB1 over-expression was significantly repaired by appropriate amounts of Pick1 RNA. (E) The over-expressed Pick1 did not affect cell junction formations. Red: Phosphorylated-E-cadherin, Green: ephrinB1, Blue: Pick1, Last row (merged). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
	Supplementary Figure 1. The expression patterns of Pick1 and ephrinB1 in Xenopus embryos by in situ hybridization. (A andB) Xenopus embryos were collected at stages 22 and 32, and then processed for whole-mount in situ hybridization using standard methods with digoxigenin-labeled (Roche; Indianapolis, IN, USA) antisense probes for ephrinB1 and Pick1. At stage 32, ephrinB1 and Pick1 were expressed in the forebrain, eye, and pronephros in the uninjected embryos. Abbreviations: fb, forebrain; e, eye; pn, pronephros.
	Supplementary Figure 2. Association of ephrinB1 with Pick1 in Xenopus epithelial cells. Embryos were injected with or without 300 pg ephrinB1 and 500 pg Pick1 RNAs and were injected with or without 10 ng Pick1MO and 150 pg ephrinB1 RNA into both the blastomeres of 2-cell stage embryos and those collected at stage 9.5. (A) Dissociation degree of cells was scored as: less than 30%, Mild; more than 30% of the animal pole region, Severe. (B) The histogram denotes the percentage of cell dissociation of the injected site for each sample.