p120-catenin-dependent junctional recruitment of Shroom3 is required for apical constriction during lens pit morphogenesis

Abstract

Apical constriction (AC) is a widely utilized mechanism of cell shape change whereby epithelial cells transform from a cylindrical to conical shape, which can facilitate morphogenetic movements during embryonic development. Invertebrate epithelial cells undergoing AC depend on the contraction of apical cortex-spanning actomyosin filaments that generate force on the apical junctions and pull them toward the middle of the cell, effectively reducing the apical circumference. A current challenge is to determine whether these mechanisms are conserved in vertebrates and to identify the molecules responsible for linking apical junctions with the AC machinery. Utilizing the developing mouse eye as a model, we have uncovered evidence that lens placode AC may be partially dependent on apically positioned myosin-containing filaments associated with the zonula adherens. In addition we found that, among several junctional components, p120-catenin genetically interacts with Shroom3, a protein required for AC during embryonic morphogenesis. Further analysis revealed that, similar to Shroom3, p120-catenin is required for AC of lens cells. Finally, we determined that p120-catenin functions by recruiting Shroom3 to adherens junctions. Together, these data identify a novel role for p120-catenin during AC and further define the mechanisms required for vertebrate AC.

Keywords: Apical constriction; Invagination; Lens pit morphogenesis; Shroom3; delta1 catenin (Ctnnd1); p120-catenin.

Fig. 1.

Myosin IIb filaments span the apical domain of apically constricting cells of the lens placode. (A) En face view of whole-mount E10.0 mouse lens placode colabeled for β-catenin and myosin IIb or F-actin. The white dashed line indicates the region of the lens placode. (B) Prominent actomyosin filaments in A that do not colocalize with junctions and that span the apical surface of cells were traced with black lines. The radius of the central lens placode region is half that of the outer peripheral boundary. (C) Higher magnification of representative regions from lens placodes fluorescently labeled for myosin IIb or F-actin. (D,E) The average percentage of cells with actomyosin filaments in the central or peripheral region (n=5) (D) and the average apical area of all cells from three placodes (E). Error bars indicate s.e.m.; *P<0.05. (F-H) En face view of lens placodes at the level of apical junctions, with magnification of the bracketed regions shown to the right. The far right panel is an image taken 2 µm below the apical junctions. White arrowheads demarcate fibers associated with apical junctions that appear to be under tension (F,H) or fibers that appear to traverse the junctions (G). White arrows indicate the apical position of the fibers. (I) Approximate positions of the E10.0 en face images shown in the panels indicated. (J,K) En face view of head ectoderm. (L-O) En face view of nasal placode. Scale bars: 50 µm in A; 5 µm in C,F-K,M-O; 20 µm in L.

Fig. 2.

Apical myosin IIb filaments are contractile. (A) Schematic detailing how bicellular junction displacement is measured. Red brackets indicate the measured dimension. (B-D) En face view of whole-mount E10.0 embryos cultured in control media (B) or media containing Blebbistatin (C) or Calyculin A (D). Immunostaining is for myosin IIb (red) and β-catenin (green). The bracketed regions are magnified to the right. Arrowheads indicate myosin filaments and the dashed lines outline the junctional and/or the myosin filament signal. (E) Representative en face and z-projection (top row) images of Calyculin A-treated placodes immunolabeled as indicated. Note the separation from myosin signal from the junctions. (F) Average width of β-catenin-labeled junctions and the distance of junctional displacement. (G) Average fold change in apical areas measured using boundaries determined from junctional labeling, and the average number of cells with filaments. Error bars indicate s.e.m.; *P<0.05 versus control and ^#P<0.05 versus filament-nonassociated junctional width. Scale bars: 5 µm.

Fig. 3.

Shroom3 genetically interacts with p120-catenin. (A-H) Representative images of whole E10.75 mouse embryos (A-D) and anterior views of E10.75 mouse heads (E-H) of the indicated genotypes. (I-L) Representative images of the eye from E12.5 mouse embryos. Arrowheads indicate the location of ventral optic cup abnormalities. Scale bars: 500 µm in A-H; 125 µm in I-L.

Fig. 4.

Shroom3 or p120 deficiency similarly disrupts lens pit invagination. (A) Cryosections from mouse embryo lenses during lens pit invagination (E10.0-10.75) and at E15.5. Sections were fluorescently labeled for β-catenin (red) alone (stages I-IV) or with F-actin (green) and Hoechst (blue) (E15.5). Scale bars: 25 µm, except 50 µm for E15.5. The white lines mark a straight epithelial shape; arrowheads indicate lens pit hinge points, whereas white arrows indicate their absence. Yellow arrows point to the lack of cornea (Cor.) and lens separation. (B) Average (x,y) coordinates (black line) for the apical and basal boundary of lens pits. The flanking gray area represents the average (x,y) coordinate plus or minus the s.e.m. (C) Average apical curvature values for the central or lateral regions of the lens pit calculated from the (x,y) coordinates determined for B. (D) Comparison of stage III lens pit cell width. *P<0.05 for p120^flox/flox versus Ap2α-cre; p120^flox/flox; ^#P<0.05 for p120^flox/flox, AP2a-cre; p120^flox/flox or Shroom3^Gt/Gt versus Shroom3^+/Gt; p120^+/flox. (E) Comparison of stage III lens pit cell height. The width of the bars is also representative of the average cell width along the apicobasal axis based on the data in D and is proportional to the cell height. (F) Non-central cryosections of stage IV lens pits in the central and hinge point regions where the apical surface is exposed for multiple cells fluorescently labeled for F-actin. Scale bars: 5 µm. (G,H) Average apical area based on immunolabeling of stage IV lens pit cryosections or en face images of stage I placodes. (C,E,G,H) *P<0.05; ^#P<0.05 compared with the appropriate control genotype. Error bars indicate s.e.m.

Fig. 5.

Junctional displacement inhibition and mislocalization of myosin IIb/Shroom3 in p120-catenin-deficient lens pits. (A) Magnified en face view of lens placode bicellular junctions associated with myosin filaments. Dashed lines indicate the localization of the junctional and/or apically positioned filaments. (B) Comparison of the average junctional displacement. *P<0.05 versus junctional width of a filament-nonassociated bicellular junction; ^#P<0.05 versus control genotype. (C-Q) Cryosections of mouse embryo lenses immunolabeled with antibodies specific to p120-catenin (C-E), Shroom3 (G-I), myosin IIb (K-M) or phalloidin to label F-actin (O-Q). The central (c) or upper hinge point (h) regions are magnified from the bracketed areas. (F) The average intensity of p120-catenin signal along a straight line drawn across apical junctions. NS, not significant. (J,N,R) The average apical signal of Shroom3 (J), myosin IIb (N) and F-actin (R) of the central or hinge point regions. *P<0.05 versus wild type. Error bars indicate s.e.m. Scale bars: 25 µm.

Fig. 6.

Junctional p120-catenin is required for Shroom3-induced AC. (A-F) Control (MDCK^wt) or p120-catenin-deficient (MDCK^shp120) MDCK cells transiently transfected with Shroom3 and immunolabeled for p120-catenin (A,B), ZO-1 (C,D) or Shroom3 and ZO-1 (E,F). (G) Apical area comparison of control and Shroom3⁺ MDCK^wt or MDCK^shp120 cells. *P<0.05 versus control MDCK^wt cells; ^#P<0.05 versus Shroom3⁺ MDCK^wt cells. (H-K) Shroom3⁺ and N-cadherin^wt+ or N-cadherin^AAA+ MDCK cells immunolabeled for Shroom3 (red) and β-catenin (green). (L) Average apical:basal area ratios of transfected MDCK cells. *P<0.05 versus non-transgenic MDCK cells; ^#P<0.05 versus Shroom3/N-cadherin^wt-expressing cells. Error bars indicate s.e.m. The number of cells measured is indicated. Scale bars: 15 µm.

Fig. 7.

Shroom3 junctional localization is facilitated by p120 catenin. (A-E) Stable transgenic lines of endogenous cadherin-deficient A431D cells expressing N-cadherin^wt (A-C) or p120-catenin binding-deficient N-cadherin^AAA (D,E) transiently transfected with Shroom3 (B,E) or the Rock binding-deficient Shroom3 mutant (Shroom3^R1838C) (C) immunolabeled for p120-catenin (red) (A,D) or Shroom3 (red) (B,C,E). The bracketed junctions are magnified in the panels beneath. (F-H) Comparison of the average fluorescence intensity values for p120-catenin (F) or Shroom3 (G,H) along line intervals perpendicular to the cell junctions. Error bars indicate s.e.m.; *P<0.05 versus wild-type construct for the bracketed data points. N.S., not significant. Scale bars: 20 µm.