Drosophila p120-catenin is crucial for endocytosis of the dynamic E-cadherin-Bazooka complex

Abstract

The intracellular functions of classical cadherins are mediated through the direct binding of two catenins: β-catenin and p120-catenin (also known as CTNND1 in vertebrates, and p120ctn in Drosophila). Whereas β-catenin is crucial for cadherin function, the role of p120-catenin is less clear and appears to vary between organisms. We show here that p120-catenin has a conserved role in regulating the endocytosis of cadherins, but that its ancestral role might have been to promote endocytosis, followed by the acquisition of a new inhibitory role in vertebrates. In Drosophila, p120-catenin facilitates endocytosis of the dynamic E-cadherin-Bazooka subcomplex, which is followed by its recycling. The absence of p120-catenin stabilises this subcomplex at the membrane, reducing the ability of cells to exchange neighbours in embryos and expanding cell-cell contacts in imaginal discs.

Keywords: Cell adhesion; E-cadherin trafficking; Epithelial morphogenesis.

Fig. 1.

p120-catenin is required for the slow component of E-cad and Baz recovery in FRAP. (A,B) ubi::E-cad–GFP distribution in the stage 15 embryo (A) and wing disc (B). The orientation within the embryo and wing (A,B), and direction of 0° and 90° border angles (A) are indicated. Scale bars: 10 µm. (C–H) ubi::E-cad–GFP FRAP at 40–90° (C,D) and 0–10° borders (E,F) in the embryonic epidermis, and in wing discs (G,H). (I–L) UAS::Baz–GFP driven by engrailed::Gal4 FRAP at 40–90° (I,J) and 0-10° borders (K,L) in the embryonic epidermis. Examples of recovery are shown in C,E,G,I and K. Red circles on the prebleached frame (P) outline the bleached spots. Average recovery curves (mean±s.e.m., n=13–24; for specific n values see Table S1) and the best fit curves (solid lines) are shown in D,F,H,J and L.

Fig. 2.

p120-catenin does not affect the amount of ubi::E-cad–GFP in stage 15 embryos and wing discs. (A–D) ubi::E-cad–GFP amounts in stage 15 embryos (A,B) and wing discs (C,D). Examples of direct fluorescence of ubi::E-cad–GFP (A,C) are shown. Scale bars: 5 µm. The quantifications are shown in B and D as box plots (n=21 in B and 19 in D). The box represents the 25–75th percentiles, and the median is indicated. The whiskers show the minimum and maximum values.

Fig. 3.

p120-catenin is not required for internalisation of E-cad for degradation. (A–F) Pulse-chase labelling of E-cad (A,B), E-cad in presence of chloroquine (CQ) (D–F) and Notch extracellular domains (C) in control and p120-catenin mutant wing discs (see Materials and Methods). Examples of immunofluorescence at different time points after pulse-chase labelling are shown in A,C and D, and quantifications of vesicle numbers are in B, E and F (mean±s.e.m., n=4 in control without CQ, and n=5 for all other datapoints). z-stacks were collected every 0.38 µm. Two sets of five images were projected to collect the signal in the 1.9 µm around the adherens junctions (red) and the next 1.9 µm basal to that (black). Arrows indicate examples of intracellular vesicles. Scale bars: 5 µm. (G) The model of two pathways of E-cad endocytosis in Drosophila cells. Endocytosis of the E-cad–Baz subcomplex is p120-catenin-dependent and results in E-cad recycling. The second pathway is p120-catenin-independent, results in E-cad degradation and affects either only immobile E-cad or both subcomplexes (uncertainty indicated with ‘?’). It is unclear whether p120-catenin associates with both E-cad subcomplexes or only with E-cad–Baz.

Fig. 4.

p120-catenin is required for multicellular rosettes and crossing of segment boundaries by cells in stage 15 embryos and for cell shape in wing discs. (A) Stripes of engrailed::Gal4 driving UAS::myr–GFP (red) with cell outlines (anti-E-cad, green) shown. In the example on the left, no cells are outside of the segment boundary (white line). In the example on the right, pairs of cells (arrowheads) are outside of the segment boundary. (B). Examples of three-cell (3), four-cell (4) and five-cell contacts (rosette, R). (C,D) Percentage of rosettes (C) and four-cell contacts (D) between the labelled cells at the boundary and their anterior neighbours. (E) Percentage of stripes with cells expressing the UAS::myr–GFP on the wrong side of the segment boundary. (F) Relative rates of cell pair crossing. In each case the mean±95% confidence interval is shown (n=135 segments from 45 embryos). *P<0.01, °P<0.05 (Chi-square test in C–E and F-test in F). (G–L) Imaginal discs (G) and stage 15 epidermis (J) from controls (p120^−/+) and p120-catenin mutants (p120⁻/Δp120) with the cell outlines (direct fluorescence of uni::E-cad–GFP) shown. Distributions of apical cell surface areas in wing discs (H) and stage 15 cells (K), and cell elongation (ratio of the length of the long to short axes) in wing discs (I) and stage 15 cells (L) in control (blue) and p120-catenin mutants (red). The binned data points (dots) and the best fit lognormal distributions (lines) are shown. In H–I, n=10,363 cells from 19 discs in control, and n=7985 cells from 19 discs in p120-catenin mutants. In K–L, n=1189 cells from 21 embryos in control, and n=1116 cells from 21 embryos in p120-catenin mutants. P values show the probabilities of the distributions being the same.