SAX-7/L1CAM and HMR-1/cadherin function redundantly in blastomere compaction and non-muscle myosin accumulation during Caenorhabditis elegans gastrulation

Abstract

Gastrulation is the first major morphogenetic movement in development and requires dynamic regulation of cell adhesion and the cytoskeleton. Caenorhabditis elegans gastrulation begins with the migration of the two endodermal precursors, Ea and Ep, from the surface of the embryo into the interior. Ea/Ep migration provides a relatively simple system to examine the intersection of cell adhesion, cell signaling, and cell movement. Ea/Ep ingression depends on correct cell fate specification and polarization, apical myosin accumulation, and Wnt activated actomyosin contraction that drives apical constriction and ingression (Lee et al., 2006; Nance et al., 2005). Here, we show that Ea/Ep ingression also requires the function of either HMR-1/cadherin or SAX-7/L1CAM. Both cadherin complex components and L1CAM are localized at all sites of cell-cell contact during gastrulation. Either system is sufficient for Ea/Ep ingression, but loss of both together leads to a failure of apical constriction and ingression. Similar results are seen with isolated blastomeres. Ea/Ep are properly specified and appear to display correct apical-basal polarity in sax-7(eq1);hmr-1(RNAi) embryos. Significantly, in sax-7(eq1);hmr-1(RNAi) embryos, Ea and Ep fail to accumulate myosin (NMY-2Colon, two colonsGFP) at their apical surfaces, but in either sax-7(eq1) or hmr-1(RNAi) embryos, apical myosin accumulation is comparable to wild type. Thus, the cadherin and L1CAM adhesion systems are redundantly required for localized myosin accumulation and hence for actomyosin contractility during gastrulation. We also show that sax-7 and hmr-1 function are redundantly required for Wnt-dependent spindle polarization during division of the ABar blastomere, indicating that these cell surface proteins redundantly regulate multiple developmental events in early embryos.

Fig. 1

The CCC and SAX-7 localize to regions of cell-cell contact. All embryos are wildtype unless indicated. (A) Anti-HMP-2 and anti-SAX-7 staining in a 26–28 cell embryo. (B) Anti-HMR-1 staining of (i) 4 cell, (ii) 26–28 cell. (C) Anti-HMR-1 staining of an embryo near the 26–28 cell stage and a Nomarski image of the same focal plane. (D) Anti-SAX-7 staining of (i) 4 cell, (ii) 26–28 cell embryos. (E) Anti-HMR-1 staining of (i) 4 cell, (ii) 26–28 cell sax-7(eq1) embryos. Scale in A smaller than other panels. Scale bars in this and all subsequent figures are 10 μm. Anterior is to the left in all figures. Nuclear signal in Dii and Eii is END-3::GFP.

Fig. 2

Intestinal precursors Ea/Ep fail to ingress in sax-7(eq1); hmr-1(RNAi) embryos. Images are from Nomarski time-lapse movies of the indicated genotypes. E cells are pseudocolored blue in all frames.

Fig. 3

Ea and Ep do not apically constrict in intact sax-7(eq1); hmr-1(RNAi) embryos. (A) Early wild-type embryonic lineage referred to in the text (anterior daughter cells shown to the left). (B) Example of traces used to measure apical membrane length and feret length. (C) Apical membrane length of Ea/Ep at 0, 8 and 16 minutes after the P₄ nuclear envelope was re-established following P₄ birth. Error bars = 95% confidence intervals. *, change from t = 0 to t = 16 significantly different from other genotypes, p < 0.02, n = 5. Other genotypes not significantly different from one another. (D) Data shown represent normalized feret length/apical membrane length at 0 and 16 minutes after the P₄ nuclear envelope was reestablished following P₄ birth. Data were normalized as described Materials and Methods; a completely flat surface has a value of 1, while a half circle has a value of 0. Error bars are 95% confidence intervals. Open diamond, t = 16 value significantly different from wildtype, p < 0.02. Closed diamond, t = 16 value significantly different from wildtype and sax-7, p < 0.01, n = 5.

Fig. 4

SAX-7 and the CCC cooperatively regulate blastomere compaction and displacement of P₄ and MSxx. (A) Cell movements in P₁ isolates over time. Ea/Ep cells are marked with white stars, the initial direction of movement of P₄ was considered to be an indicator of the apical side of the blastomeres (black asterisks) and white arrows show directions of movements of MSxx and P₄. Times shown are post-Ea/Ep birth. Ea and Ep divide prior to the 60 min. time point. Some cells are not visible due to their position along the Z-axis. (B) Length of contact area between Ea/Ep in blastomeres (see Supplemental Fig. S1 for method of measurement). Open diamond, significantly different from sax-7, p < 0.03; not significantly different from sax-7;hmr-1. *, significantly different from wildtype and sax-7, p < 0.0001; n = 5 for all genotypes. (C) Circularity of the Ep blastomere. Open diamond, significantly different from wildtype and sax-7, p < 0.01, n = 5 for each genotype; not significantly different from sax-7;hmr-1. #, not significantly different from wildtype. *, significantly different from wildtype and sax-7, p < 0.001; not significantly different from sax-7;hmr-1. (D) Changes in angular orientation of P₄ and MSxx relative to Ea/Ep as a measure of apical constriction of Ea/Ep. MSxx angle in sax-7;hmr-1 significantly different, p < 0.05, n = 5 for each genotype.

Fig. 5

PAR localization indicates Ea/Ep are correctly polarized in sax-7(eq1); hmr-1(RNAi) embryos. PAR-2::GFP was imaged in Ea/Ep shortly following P₄ birth in wild-type and mutant embryos. Although GFP distribution was monitored throughout all focal planes, single focal planes are shown. The arrowheads point to the apical surface of Ea/Ep (note that vitelline envelope appears white in sax-7(eq1), but is not the apical surface). Arrows point to the basal surface. The promoter used in the PAR-2 expression construct is highly active in P₄ leading to cortical localization. In some embryos PAR-2::GFP is brighter in D than in other cells. This occurs more frequently in sax-7(eq1); hmr-1(RNAi), but has been observed in all genotypes.

Fig. 6

hmr-1(RNAi); sax-7(eq1) embryos fail to accumulate NMY-2 on the apical surfaces of Ea/Ep. (A) NMY-2::GFP levels at apical surface of Ea/Ep over time in wild-type and mutant embryos. The two left-hand panels show NMY-2 accumulation at two successive time points in wild-type and mutant embryos. The right column represents kymographs of the apical surface over time. (B) NMY-2::GFP levels at the apical surface of Ea/Ep over time in wild-type and mutant embryos. *, significantly different from other genotypes, p. < 0.02, n = 5 for each genotype. #, wildtype, hmr-1, and sax-7 not significantly different. (C) NMY-2::GFP levels at the Ea/Ep border over time in wild-type and mutant embryos. *, significantly different from other genotypes, p < = 0.01, n = 5 of each genotype. #, significantly different from hmr-1 and sax-7, p < 0.05. Open diamond, hmr-1 and sax-7 not significantly different.

Fig. 7

Later gastrulation defects are also observed following loss of CCC and SAX-7 function. Nomarski images of the ventral surface of wild-type and mutant embryos. Open gastrulation clefts are outlined with dashed lines. White stars indicate extruded cells.