Suppression of epithelial folding at actomyosin-enriched compartment boundaries downstream of Wingless signalling in Drosophila

Abstract

Epithelial folding shapes embryos and tissues during development. Here, we investigate the coupling between epithelial folding and actomyosin-enriched compartmental boundaries. The mechanistic relationship between the two is unclear, because actomyosin-enriched boundaries are not necessarily associated with folds. Also, some cases of epithelial folding occur independently of actomyosin contractility. We investigated the shallow folds called parasegment grooves that form at boundaries between anterior and posterior compartments in the early Drosophila embryo. We demonstrate that formation of these folds requires the presence of an actomyosin enrichment along the boundary cell-cell contacts. These enrichments, which require Wingless signalling, increase interfacial tension not only at the level of the adherens junctions but also along the lateral surfaces. We find that epithelial folding is normally under inhibitory control because different genetic manipulations, including depletion of the Myosin II phosphatase Flapwing, increase the depth of folds at boundaries. Fold depth correlates with the levels of Bazooka (Baz), the Par-3 homologue, along the boundary cell-cell contacts. Moreover, Wingless and Hedgehog signalling have opposite effects on fold depth at the boundary that correlate with changes in Baz planar polarity.

Keywords: Apico-basal polarity; Contractility; Embryo; Epithelium; Morphogenesis; Planar polarity.

Fig. 1.

Planar polarities and actomyosin contractility at PSBs in wild-type and wingless mutant embryos. (A,B) Representative wild-type (WT) (n=30) (A) and wg^CX4 (n=22) (B) stage 10 embryos imaged by scanning electron microscopy. Scale bars: 100 µm. Parasegmental grooves (asterisks) apparent in WT embryos are absent in wg^CX4 embryos. (C,F) PSBs, corresponding to the interfaces between wingless- and engrailed-expressing cells, enrich actomyosin in WT embryos (C). This enrichment is mostly lost in wg^CX4 embryos (F). (D,E,G,H) Immunostaining against Sqh1P or Baz in WT or wg^CX4 embryos (close-ups shown for Baz). Scale bars: 10 µm. (D′,E′, G′,H′) Corresponding merges with an AJ marker, E-Cadherin (DE-Cad) or phosphotyrosine (pTyr), and PSB marker, Engrailed (En) or en-lacZ (en-lacZ is used to identify PSBs in wg mutants; see Materials and Methods). (D″,E″,G″,H″) Tracings of the PSB (solid line) and control cell-cell contacts (dotted line) to quantify the signal at the level of AJs. (I) Quantification of the fluorescence intensities (f.i.) of proteins along PSBs relative to control interfaces, in WT and wg^CX4 embryos, as log₁₀ (DE-Cad in WT, n=20 PSBs, in wg^CX4, n=20; Sqh1P in WT, n=20, in wg^CX4, n=23; Rok-GFP in WT, n=26, in wg^CX4, n=23; Flw-YFP in WT, n=20, in wg^CX4, n=22; Baz in WT, n=31, in wg^CX4, n=21). Error bars show mean±95% confidence interval (CI). Comparisons between WT and wg^CX4 from Student's t-tests: DE-Cad, P=0.256 (n.s.); Sqh1P, ***P=0.0008; Rok-GFP, ****P<0.0001; Flw-YFP, *P=0.0237; Baz, ***P=0.0002. (J-L) Laser ablations to probe junctional tension at PSBs. (J) Overlay before (green) and after (magenta) ablation of a single cell-cell junction at the PSB (the white rectangle indicates the ablation zone). Scale bar: 5 µm. (J′) Kymograph of the signal along the dashed line in J used to measure distance between cut ends over time (arrows in J; arrowheads in J′). (K,L) Recoil speed upon ablation of cell-cell junctions in WT (K) and wg^CX4 embryos (L) at PSBs or control junctions parallel to AP or DV axes. Control AP junctions in WT, n=32 ablations; control DV junctions in WT, n=18; PSBs in WT, n=20; control DV junctions in wg^CX4, n=18; PSBs in wg^CX4, n=19. Error bars show mean±s.d. Comparisons in K from a Kruskal–Wallis test: AP control versus DV control, P=0.780 (n.s.); PSB versus control, ***P<0.0001. Comparison in L from a Mann–Whitney test: P=0.910 (n.s.). In all figures, anterior is left and dorsal up, unless otherwise stated. Open arrowheads label PSBs.

Fig. 2.

PSB grooves are deepened by depletion of the Myosin-II phosphatase Flapwing. (A) Immunostaining against GFP reveals degradation of Flw-YFP in an embryo expressing UAS-deGradFP in the prd-Gal4 domain (yellow dotted lines). Scale bar: 50 µm. (B-B″) The same genotype at higher magnification, immunostained against GFP (B), Sqh1P (B′) and merged (B″). Scale bar: 10 µm. Filled arrowheads indicate PSBs in prd-Gal4 domains; open arrowheads indicate control PSBs. (C-D′) SEM images of late stage 10 control and deGradFP embryos. Asterisks indicate shallow parasegmental grooves in control embryos; filled arrowheads indicate deepened parasegmental grooves in Flw-depleted domains (close-up in D′). Scale bars: 50 µm. (E) Blind quantification of embryos with shallow only versus deep grooves in sibling embryos shown in C and D. Comparison from Fisher's exact test, **P=0.0016. (F) Quantification of the fluorescence intensities (f.i.) of proteins at PSBs in deGradFP-expressing and -nonexpressing domains (prd-Gal4 positive or negative), relative to control interfaces, as log₁₀ (for both domains, Sqh1P, n=22 PSBs; Baz and DE-Cad, n=18). Error bars show mean±95% CI. Comparisons between prd-Gal4-positive and -negative PSBs from Mann–Whitney tests: Sqh1P, ***P=0.0002; Baz, **P=0.0016; DE-Cad, P=0.9626 (n.s.). (G) PSB position and fold depth relative to Wg, En and prd-Gal4 expression domains. (H-H″) Immunostaining against GFP (H) and En (H′) (H″, merge) showing the deep groove at the PSB in the Flw-depleted domain (filled arrowheads) and the shallow groove (open arrowheads) in the control domain. Scale bar: 20 µm.

Fig. 3.

Increased epithelial folding at ectopic PSBs in wingless-overexpressing embryos. (A,B) Location of wg-, hh- and en-expressing cells relative to endogenous and ectopic PSBs at stage 10. Ectopic PSBs form at posterior edges of the enlarged Engrailed domain in arm>wg embryos. (C) Position of the actomyosin enrichment at endogenous (magenta) and ectopic (cyan) PSBs in arm>wg embryos. (D,D′) Immunostaining of arm>wg early stage 10 embryos against Sqh1P (D), En and pTyr (merged in D′). Scale bar: 10 µm. (D″) Traces of endogenous and ectopic PSBs (solid lines) and control junctions (dotted line). (E) Quantification of the fluorescence intensities (f.i.) of proteins in arm>wg embryos along the endogenous (open circles) and ectopic (solid circles) PSB junctions, relative to control interfaces, as log₁₀ (for both boundaries, Sqh1P, n=20; Rok-GFP, n=17; Flw-YFP, n=20; Baz, n=22). Error bars show mean±95% CI. Comparisons between PSBs and ectopic boundaries from Student's t-tests: Sqh1P, *P=0.025; Rok-GFP, ***P=0.0002; Flw-YFP, P=0.76 (n.s.); Baz, ****P<0.0001. (F) Recoil speeds following laser ablation of endogenous and ectopic PSB cell junctions, and control DV-oriented junctions. Control DV junctions, n=20 ablations; PSB, n=25; ectopic, n=26. Error bars show mean±s.d. Comparisons from one-way ANOVA: DV controls versus PSBs or ectopics, ****P<0.0001; PSB versus ectopics, P=0.998 (n.s.). (G) Position of the shallow and deep folds at endogenous and ectopic PSBs, respectively, in arm>wg embryos. (H-H″) Sagittal view showing difference in folding at the endogenous and ectopic PSBs of an arm>wg embryo stained for alpha-Catenin (α-Cat) (H), En (H′) and merged with Discs Large (Dlg) (H″). Scale bar: 10 µm. (I-J′) SEM of stage 10 (I) wild-type and (J,J′) arm>wg embryos. Endogenous PSBs barely indent the surface of the embryo (asterisks in I), whereas ectopic PSBs form deep grooves (J, close-up in J′). WT, n=30 embryos; arm>wg, n=62 embryos, of which 56 had deep folds. (K,K′) SEM of stage 10 arm>wg embryos injected with Rho kinase inhibitor Y-27632, showing two examples; n=10 embryos, of which 9 had no grooves. Scale bars: 100 µm. Open arrowheads indicate endogenous PSBs; filled arrowheads indicate ectopic PSBs.

Fig. 4.

Baz overexpression increases epithelial folding at actomyosin-enriched boundaries. (A) Positions of deep folds at endogenous PSBs in Baz-overexpressing embryos (MTD>bazGFP). (B,B′) SEM of MTD>bazGFP embryos (B) at stage 10 and (B′) stage 11, showing deep folding at PSBs. n=28 embryos (24 show deepened folds, 85.7%). Scale bars: 100 µm. (C) Recoil speeds following laser ablation of DV-oriented control and PSB junctions in MTD>bazGFP embryos. Control DV junctions, n=21 ablations; PSB, n=25. Error bars show mean±s.d. Comparison from a Student's t-test: **P=0.0032. (D-D″) Grazing section of an early stage 10 MTD>bazGFP embryo, immunostained against GFP (D) and Engrailed (D′) (merged in D″), showing deep PSB folds. Scale bar: 20 µm. (E,F) Quantification of the absolute fluorescence intensities (f.i.) of Baz (E) and Sqh1P (F) at PSBs and control DV-oriented interfaces in wild-type and MTD>bazGFP embryos. For PSBs (in both WT and MTD>bazGFP embryos, for both Sqh1P and Baz), n=21 boundaries; controls, n=42. Error bars show mean±95% CI. Comparisons from Student's t-tests: Baz in WT, ***P=0.0001; Baz in MTD>bazGFP, ****P<0.0001; Sqh1P in WT, *P=0.0368; Sqh1P in MTD>bazGFP, P=0.139 (n.s.).

Fig. 5.

Fold suppression and Baz depletion is rescued at ectopic PSBs in absence of Hedgehog. (A) Lack of deep grooves at endogenous and ectopic PSBs, in arm>wg embryos in a null hh mutant background. (B,C) SEM of stage 10 embryos with arm>wg (B) showing the deep grooves at ectopic PSBs (n=62 embryos), which are much shallower in arm>wg, hh^AC/hh^AC (C) (n=19 embryos, of which 17 had shallow grooves). Scale bars: 100 µm. (D,E) Quantification of the fluorescence intensities (f.i.) of Sqh1P (D) and Baz (E) in arm>wg and arm>wg, hh^AC/hh^AC embryos along the PSB (open circles) and ectopic (filled circles) junctions, relative to control cell interfaces, as log₁₀. Error bars show mean±95% CI. Sqh1P in arm>wg, n=20; Sqh1P in arm>wg, hh^AC/hh^AC, n=23; Baz in arm>wg, n=22; Baz in arm>wg, hh^AC/hh^AC, n=22. Comparisons in E from one-sample Student's t-tests: Baz at ectopics in arm>wg, difference from 0, *P=0.0399; Baz at ectopics in arm>wg, hh^AC/hh^AC, difference from 0, ****P<0.0001.

Fig. 6.

Cell behaviours during folding at endogenous and ectopic PSBs in live embryos. (A-B‴) Frames at 10-min intervals from time-lapse imaging of a live stage 10 embryo expressing Gap43-mCherry, with cells abutting the PSB highlighted, in a wild-type (A-A‴) and an arm>wg (B-B‴) embryo. Dashed lines highlight the boundaries; the asterisks mark cells that delaminate from the epithelium; arrows indicate cells undergoing intercalation; arrowheads indicate cells undergoing intercalation events associated with cell divisions. Anterior is left and ventral is up. Scale bars: 10 µm.

Fig. 7.

Measuring AJ lowering and apical constriction at endogenous and ectopic PSBs. (A,B) Confocal stack projections of stage 10 wild-type and arm>wg embryos immunostained for E-Cadherin (green), phalloidin (cyan) and Engrailed (magenta). The positions of endogenous (open arrowheads) and ectopic (filled arrowheads) PSBs are indicated. Scale bars: 20 µm. (A′,B′) Close-up of x-z optical sections through the stacks shown in A and B. Scale bars: 5 µm. (C,D) Cell segmentation of image stacks shown in A and B. Cells depicted in green, magenta and cyan are control cells and endogenous and ectopic PSB-abutting cells, respectively. Scale bars: 20 µm. (E,F) 3D cell reconstructions of representative example cells from wild-type (E) and arm>wg (F) embryos. Control cells, green; endogenous PSB-abutting cells, magenta; ectopic PSB-abutting cell, cyan. Scale bars: 2 µm. (G,H) Histograms showing the distance separating the AJs from the top of the cell for control, endogenous PSB and ectopic PSB cell-cell junctions, in wild-type (G) and arm>wg (H) embryos. In wild-type embryos, the mean AJ positions are 0.67 µm (below the top of the cell) for controls and 0.86 µm for PSBs (n=2120 pixels in PSB junctions; n=2481 in controls). In arm>wg embryos, the mean AJ positions are 0.71 µm for controls, 0.84 µm for endogenous PSBs and 1.06 µm for ectopic PSBs (n=1139 pixels in endogenous PSB junctions; n=1309 in controls; n=1398 in ectopic PSBs). Outliers (>3 µm between AJ and top of cell) are omitted; they account for 3.1% and 2.5% of the data in wild-type and arm>wg embryos, respectively. (I,J) Quantification of apical areas of control cells and cells abutting the endogenous and ectopic PSBs, in wild-type (I) and arm>wg (J) embryos (cell numbers: in wild-type controls, n=144; PSBs, n=48; in arm>wg controls, n=94; PSBs, n=44 PSBs; ectopic PSBs, n=34). Comparison in I from a Student's t-test: P=0.338 (n.s.). Comparisons in J from a Kruskal–Wallis test: all pairs, n.s. Error bars show mean±95% CI. Three embryos of each genotype (shown in Fig. S5) were analysed in G-J.