Cell signaling facilitates apical constriction by basolaterally recruiting Arp2/3 via Rac and WAVE

Abstract

Apical constriction is a critical cell shape change that drives cell internalization and tissue bending. How precisely localized actomyosin regulators drive apical constriction remains poorly understood. Caenorhabditis elegans gastrulation provides a valuable model to address this question. The Arp2/3 complex is essential in C. elegans gastrulation. To understand how Arp2/3 is locally regulated, we imaged embryos with endogenously tagged Arp2/3 and its nucleation-promoting factors (NPFs). The three NPFs-WAVE, WASP, and WASH-controlled Arp2/3 localization at distinct subcellular locations. We exploited this finding to study distinct populations of Arp2/3 and found that only WAVE depletion caused penetrant gastrulation defects. WAVE localized basolaterally with Arp2/3 and controlled F-actin levels near cell-cell contacts. WAVE and Arp2/3 localization depended on CED-10/Rac. Establishing ectopic cell contacts recruited WAVE and Arp2/3, identifying the contact as a symmetry-breaking cue for localization of these proteins. These results suggest that cell-cell signaling via Rac activates WAVE and Arp2/3 basolaterally and that basolateral Arp2/3 makes an important contribution to apical constriction.

Figure 1.

Quantitative analysis of Arp2/3 and NPFs localization during apical constriction. (A) Maximum intensity projections of 10 planes spanning a total Z-depth of 5 μm, depicting C. elegans gastrulation from a ventral view with plasma membranes fluorescently labeled (mex-5p::mScarlet-I::PH). Ea and Ep cells are pseudocolored to visualize their internalization over time. The diagram to the right that shows Ea and Ep, along with their neighboring cells, is used throughout the paper to depict cellular and subcellular protein localization. (B) Micrographs from time-lapse movies depicting localization of Arp2/3 (ARX-2::TagRFP), WAVE (GFP-C1^3xFlag::GEX-3), WASP (GFP::WSP-1A), and WASH (WSHC-5::mNG-C1^3xFlag) from a ventral view. White arrowheads point to Ea and Ep cells. White arrows point to vesicle-like structures in the cytoplasm. The diagrams underneath each micrograph highlight the observed localization in Ea, Ep, and neighboring cells. (C) Diagrams representing the four regions of interest for quantification: the Ep-P₄ contact, the other cell–cell contacts, the cytoplasm, and a line scan within the cytoplasm. (D) Violin plots depicting normalized fluorescence intensity of Arp2/3, WAVE, WASP, and WASH at Ep-P₄ contacts, other cell–cell contacts, and the cytoplasm. Insets of Arp2/3 and WASP highlight the difference between the signal at the other cell–cell contacts and the cytoplasm. Measurements were collected 6 min after the division of neighboring mesoderm precursor cells (MSx) (center dot, mean; vertical line, standard deviation (s.d.); outline, the distribution of the data; n ≥ 10 embryos). (E) Representative line scan measurements of Arp2/3, WAVE, WASP, and WASH. (F) Violin plots depicting the differences between the maximum and minimum gray values along the line scan (center dot, mean; vertical line, s.d.; outline, the distribution of the data; n ≥10 embryos). Statistical tests for experiments in D and F were chosen based on the normality and variance of the data. (D) For Arp2/3 and WASP, Welch’s ANOVA was followed by post-hoc Dunnett’s tests; for WAVE, one-way ANOVA was followed by Post-hoc Tukey’s tests; for WASH, Kruskal–Wallis test was followed by post-hoc Dunn’s test. (F) Kruskal–Wallis test was followed by post-hoc Dunn’s test. *P < 0.05, **P < 0.01, ****P < 0.0001. Scale bar: 5 µm.

Figure S1.

Characterization of the NPFs during apical constriction. (A) Schematic of the WAVE, WASP, and WASH complexes with endogenously tagged components highlighted with colored outlines. (B) Violin plots depicting normalized fluorescence intensity of Arp2/3, WAVE, WASP, and WASH at Ep-P₄ contacts, other cell–cell contacts, and the cytoplasm. Insets of Arp2/3 and WASP highlight the difference between the signal at the other cell–cell contacts and the cytoplasm. Measurements were collected 0 and 12 min after the division of neighboring mesoderm precursor cells (MSx) (center dot, mean; vertical line, s.d.; outline, the distribution of the data; n ≥ 10 embryos). (C) Schematic of the endogenous wsp-1 locus fused to GFP or mNeonGreen at its N-termini of the isoforms encoding WSP-1A or WSP-1B. (D) DIC (left) and fluorescence (right) micrographs of N2 control and mNeonGreen::WSP-1B embryos from a lateral view. (E) Bar plot depicting relative intensity measurements of whole embryos from N2 and mNeonGreen::WSP-1B worms (n = 15 embryos). (F) Covisualization of WASH and mScarlet-I::CAP-1 from a lateral view. Areas within the white boxes in the upper panel are enlarged in the lower panel to better visualize colocalization between the two proteins. (G) Quantification of colocalization, with a box plot reporting Pearson correlation coefficients at Ep-P₄ contacts, other cell–cell contacts, and the cytoplasm (center line, median; box, IQR; whiskers, min/max range; n = 10 embryos). Statistical tests for experiments in B, E, and G were chosen based on the normality and variance of the data. (B) For Arp2/3 (0 and 12 min) and WASP (12 min), Welch’s ANOVA followed by post-hoc Dunnett’s tests; for WAVE (0 and 12 min), one-way ANOVA followed by Post-hoc Tukey’s tests; for WASH (0 and 12 min) and WASP (0 min), Kruskal–Wallis test followed by post-hoc Dunn’s test. (E) Unpaired t test. (G) One-way ANOVA followed by post-hoc Tukey’s tests. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Scale bar: 5 µm.

Figure 2.

Three NPFs colocalize with Arp2/3 at different cellular and subcellular locations. (A) Covisualization of Arp2/3 with WAVE, WASP, and WASH in gastrulation stage embryos using endogenously tagged alleles. White arrowheads point to Ea and Ep cells. Scale bar: 5 µm. (B) Quantification of colocalization between Arp2/3 and NPFs, with box plots reporting Pearson correlation coefficients at Ep-P₄ contacts, other cell–cell contacts, and the cytoplasm. Statistical tests for experiments in B were chosen based on the normality and variance of the data: For WAVE, one-way ANOVA was followed by Post-hoc Tukey’s tests; for WASP and WASH, Kruskal–Wallis test was followed by post-hoc Dunn’s test (center line, median; box, interquartile range (IQR); whiskers, min/max range; n ≥ 10 embryos; *P < 0.05, **P < 0.01, ****P < 0.0001.

Figure S2.

Quantification of fluorescence in wild-type/knock-in/RNAi-treated embryos permits verification of knockdown effectiveness. (A) DIC (upper) and fluorescence (lower) micrographs of stage-matched wild-type, knock-in, and RNAi-treated embryos mounted side-by-side from a lateral view. Scale bar: 5 µm. (B) Violin plots depicting normalized relative intensity measurements of whole embryos from wild-type, knock-in, and RNAi-treated embryos. The average fluorescence intensity in wild-type embryos is set to 0%, and in knock-in embryos is set to 100%. Statistical tests for experiments in B were chosen based on the normality and variance of the data: for WAVE and WASH, one-way ANOVA followed by post-hoc Tukey’s tests; for WASP, Welch’s ANOVA followed by post-hoc Dunnett’s tests (center dot, mean; vertical line, s.d.; outline, the distribution of the data; n ≥ 10 embryos).

Figure 3.

RNAi against each NPF leads to Arp2/3 reduction at different cellular and subcellular locations. (A) Micrographs from time-lapse movies depicting localization of Arp2/3 in control and WAVE, WASP, or WASH RNAi-treated embryos from a lateral view. White arrowheads point to Ea and Ep cells. The diagrams underneath each micrograph highlight the observed Arp2/3 localization in E and neighboring cells. (B) Violin plots reporting changes in Arp2/3 localization at Ep-P₄ contacts, other cell-cell contacts, and the cytoplasm upon RNAi depletion of WAVE, WASP, and WASH (center dot, mean; vertical line, s.d.; outline, the distribution of the data; n ≥ 10 embryos). (C) Violin plots reporting changes in the differences between the maximum and minimum gray values of the Arp2/3 signal along the line scan upon RNAi depletion of WAVE, WASP, and WASH (center dot, mean; vertical line, s.d.; outline, the distribution of the data; n ≥ 10 embryos). All measurements were collected 6 min after the division of neighboring mesoderm precursor cells (MSx). Statistical tests for experiments in B and C were chosen based on the normality and variance of the data. (B) For WAVE (other cell contacts [control versus RNAi]), unpaired t test with Welch’s correction; for WAVE (others), unpaired t test; for WASP (Ep–P4 contacts [control versus RNAi]), unpaired t test with Welch’s correction; for WASP (others), unpaired t test; for WASH (Ep–P4 contacts [control versus RNAi]), unpaired t test; for WASH (other cell contacts [control versus RNAi]), unpaired t test with Welch’s correction; for WASH (cytoplasm), Mann-Whitney test. (C) For WAVE and WASP, unpaired t test; for WASH, Mann–Whitney test. *P < 0.05, **P < 0.01, ****P < 0.0001. Scale bar: 5 µm.

Figure 4.

NPF RNAi by themselves and in combination lead to different degrees of gastrulation defects. (A) Micrographs from time-lapse DIC movies in eight different backgrounds, with time indicated on the left from the MSx cell division. E lineage cells are outlined and pseudocolored in green. Gastrulation defects (Ea or Ep cell dividing before being fully covered by neighboring cells) are indicated with white arrowheads. An enclosed outline and the absence of arrowheads indicate that endodermal precursors became internalized at the 2E stage, as seen in wild-type embryos. Scale bar: 5 µm. (B) The bar graph (left) and Venn diagram (right) summarize the effects of NPF RNAi, both individually and in combination, on gastrulation. The heat map represents the different penetrance levels, with darker colors indicating a higher percentage of gastrulation defects. Fisher’s exact test was used for categorical data in B. *P < 0.05, ****P < 0.0001.

Figure 5.

WAVE regulates F-actin levels at multiple cell–cell contact regions. (A) Micrographs from time-lapse movies depicting localization of F-actin in control and WAVE RNAi-treated embryos from a ventral view. Scale bar: 5 µm. Green arrows, MS front (Ea-MSxx contact); yellow arrows, MS back (MSxx-ABxxxx); pink arrows, Ea–Ep contact; purple arrows, Ea/p-ABxxxx contacts; grey arrows, Ep–P₄ contact. White arrowheads point to Ea and Ep cells. (B) Violin plots reporting changes in F-actin levels at MS front, MS back, Ea–Ep contact, Ea/p–ABxxxx contacts, and Ep–P₄ contact upon RNAi depletion of WAVE (center dot, mean; vertical line, s.d.; outline, the distribution of the data; n = 10 embryos). (C) Violin plots reporting changes in the ratio of F-actin at the front versus back of the MSxx cell (center dot, mean; vertical line, s.d.; outline, the distribution of the data; n = 10 embryos). Statistical tests for experiments in B and C were chosen based on the normality and variance of the data. (B) Unpaired t test. (C) Mann–Whitney test. **P < 0.01, ****P < 0.0001.

Figure S3.

WAVE RNAi did not affect E-cadherin levels at cell–cell contacts. (A) Micrographs from time-lapse movies depicting localization of E-cadherin in control and WAVE RNAi-treated embryos from a lateral view. Scale bar: 5 µm. White arrowheads point to Ea and Ep cells. (B) Violin plots reporting changes in E-cadherin levels at Ep-P₄ contacts, non-Ep-P₄ cell–cell contacts, and Ea–MSxx contacts upon RNAi depletion of WAVE (center dot, mean; vertical line, s.d.; outline, the distribution of the data; n = 10 embryos). Unpaired t test was used in B based on the normality and variance of the data.

Figure 6.

The Rac1 GTPase CED-10 recruits WAVE and Arp2/3 at cell–cell contact and contributes to apical constriction. (A) Micrographs from time-lapse movies depicting localization of WAVE and Arp2/3 in control and ced-10 RNAi-treated embryos from a lateral view. Scale bar: 5 µm. The diagrams underneath each micrograph highlight the observed Arp2/3 localization in E and neighboring cells. (B) Violin plots reporting changes in WAVE and Arp2/3 localization at Ep-P₄ contacts, other cell–cell contacts, and the cytoplasm upon RNAi depletion of CED-10 (center dot, mean; vertical line, s.d.; outline, the distribution of the data; n = 11 embryos). (C) Micrographs from time-lapse DIC movies of wild-type and ced-10 RNAi-treated embryos with time on the left from MSa/p cell division. E-lineage cells are outlined and pseudocolored in green. Gastrulation defects (Ea or Ep cell dividing before being fully covered by neighboring cells) are indicated with white arrowheads. Scale bar: 5 µm. (D) Bar graph summarizing gastrulation defects caused by ced-10 RNAi. Statistical tests for experiments in B were chosen based on the normality and variance of the data: For WAVE (cytoplasm [control versus RNAi]), Mann–Whitney test; for WAVE (others), unpaired t test; for Arp2/3 (Ep-P4 contacts [control versus RNAi]), unpaired t test with Welch’s correction; For Arp2/3 (others), unpaired t test. Fisher’s exact test was used for categorical data in D. *P < 0.05, ***P < 0.001, ****P < 0.0001.

Figure S4.

Characterization of the Rac proteins during apical constriction. (A) Micrographs from time-lapse movies depicting localization of WAVE and Arp2/3 in control and mig-2 RNAi-treated embryos from a lateral view. The diagrams underneath each micrograph highlight the observed Arp2/3 localization in E and neighboring cells. (B) Violin plots reporting changes in WAVE and Arp2/3 localization at Ep-P₄ contacts, other cell–cell contacts, and the cytoplasm upon RNAi depletion of MIG-2 (center dot, mean; vertical line, s.d.; outline, the distribution of the data; n = 10 embryos). (C) Micrographs from time-lapse movies depicting localization of mNeonGreen::CED-10 from a lateral view. White arrowheads point to Ea and Ep cells. (D) DIC (up) and fluorescence (down) micrographs of stage-matched wild-type, knock-in, and ced-10 RNAi-treated embryos mounted side-by-side from a lateral view. (E) Violin plot depicting normalized relative intensity measurements of whole embryos from wild-type, knock-in, and ced-10 RNAi-treated embryos, with average fluorescence intensity in wild-type embryos set to 0% and knock-in embryos set to 100% (center dot, mean; vertical line, s.d.; outline, the distribution of the data; n = 11 embryos). Statistical tests for experiments in B and E were chosen based on the normality and variance of the data. (B) For Ep-P4 contacts (control versus RNAi) and other cell contacts (control versus RNAi), Mann–Whitney test; for cytoplasm (control versus RNAi), unpaired t test. (E) Welch’s ANOVA followed by post-hoc Dunnett’s tests. Scale bar: 5 µm.

Figure 7.

Newly established ectopic cell–cell contact acts as a symmetry-breaking cue to recruit WAVE and Arp2/3. (A) Micrographs from time-lapse movies of chimeric embryos created by combining two embryos expressing WAVE and Arp2/3. The DIC channel of the whole chimera is shown on the left, and two fluorescence channels of a blow-up of the chimeric contact (outlined region) are shown on the right. Yellow arrows point to ectopic cell–cell contacts. (B) Violin plots reporting normalized fluorescence intensity of WAVE and Arp2/3 at ectopic/endogenous cell–cell contacts and the contact-free cell apex (center dot, mean; vertical line, s.d.; outline, the distribution of the data; n = 15 chimeras). Statistical tests for experiments in B were chosen based on the normality and variance of the data: For WAVE, Kruskal–Wallis test followed by post-hoc Dunn’s test; for Arp2/3, one-way ANOVA followed by post-hoc Tukey’s tests. **P < 0.01, ***P < 0.001, ****P < 0.0001. Scale bar: 5 µm.

Figure 8.

Summary. (A) The three NPFs in C. elegans, WAVE, WASP, and WASH complexes, colocalize with Arp2/3 and control Arp2/3 localization at distinct subcellular locations. (B) Cell–cell signaling makes an essential contribution to apical constriction by basolaterally recruiting Arp2/3 via Rac and WAVE. (C) Rac signaling and myosin-activating kinase are active at opposing locations in migrating and apically constricting cells.

Figure S5.

Myosin II partial RNAi rescued gastrulation defects caused by WAVE depletion. (A) Micrographs from time-lapse DIC movies in three backgrounds with time on the left from MSa/p cell division. E lineage cells are outlined and pseudocolored in green. Gastrulation defects (Ea or Ep cell dividing before being fully covered by neighboring cells) are indicated with white arrowheads. An enclosed outline and absence of the arrowhead indicate that endodermal precursors became internalized at the 2E stage, as in wild-type embryos. Scale bar: 5 µm. (B) The bar graph summarizes the percentage of embryos that have gastrulation defects. Fisher’s exact test was used for categorical data in B. ****P < 0.0001.