Rearrangement of the actin cytoskeleton is the basis for cell behavior as diverse as cytokinesis, endocytosis, cell shape change, adhesion, and migration. These rearrangements are mediated by members of the Rho family of small GTPases (Schmalzing et al.,1995; Braga et al.,1997; Drechsel et al.,1997; Hall and Nobes,2000). Most GTPases cycle between an active GTP-bound form and an inactive GDP-bound form. Rho GTPases are intracellular signal transducing proteins that interact with and activate downstream effector proteins when bound to GTP.
Three highly conserved members of this family are Rho, Rac, and Cdc42 (reviewed in Jaffe and Hall,2005). In fibroblast cells, Rho regulates the assembly of contractile actin-myosin filaments, while Rac and Cdc42 control actin polymerization during filopodia and lamelipodia formation. Rho is also involved in wound healing in Xenopus oocytes and mouse embryonic fibroblasts (Nobes and Hall,1999; Benink and Bement,2005).
In the Xenopus embryo, extensive morphogenetic cell movements take place during gastrulation and neurulation. Cells of the involuting mesoderm become polarized and perform convergent extension movements, which establish the anteroposterior axis in the embryo. During the involution process, the mesoderm comes into contact with the neuroectoderm and a process called tissue separation ensures that these cell populations do not mix. The neurocetoderm undergoes CE movements similar to the mesoderm, and the formation and closure of the neural tube are other important morphogenetic events.
Rho GTPases in Xenopus embryos are expressed in all tissues that undergo extensive morphogenetic movements during gastrulation and neurulation. There they mediate CE movements, TS and neural tube closure. Both Rho and Rac are activated in the Planar Cell Polarity pathway (PCP) in the mesoderm and neuroectoderm (Wallingford and Harland,2002; Habas et al.,2003; Tahinci and Symes,2003). Other components of PCP signaling such as Paraxial Protocadherin (PAPC) activate Rho but inhibit Rac (Medina et al.,2004; Unterseher et al.,2004). Rho activation is a transient process that can occur in defined compartments of a cell. Here we describe a method that allows us to monitor Rho signaling at a subcellular resolution in early Xenopus embryos. This assay can serve as a valuable tool for the analysis of Rho-mediated signaling in the context of morphogenetic cell movements.
RESULTS AND DISCUSSION
Several methods to detect active Rho have been adapted to the Xenopus system from cell culture. A widely used method consists of precipitating active Rho with the Rho-binding domain of Rhotekin fused to Glutathione-S-transferase (RBD-GST) from protein lysates (Ren et al.,1999). While this is an effective method to determine endogenous global Rho activity levels, it does not provide any information about the subcellular distribution of active Rho. One means with which to approach this problem is to visualize active Rho using a RBD-GFP fusion construct. This method has been used in Xenopus oocyte wounding assays in which mRNA coding for RBD-GFP was microinjected (Benink and Bement,2005). Another method consists of using a single-chain biosensor with intramolecular fluorescence resonance energy transfer (FRET) that measures activation of ectopic Rho, but not the activation status of endogenous Rho (Pertz et al.,2006). Recently this method has been adapted to the Xenopus system to analyze neural crest cell migration by injecting plasmid DNA into 8-cell-stage embryos (Matthews et al.,2008). A simpler means with which to detect localized endogenous Rho activity is to use RBD-GFP protein to stain fixed and permeabilized samples. This approach was first used to analyze cell migration in cell culture scratch-wound assays (Goulimari et al.,2005). We have adapted this method for Xenopus samples to stain cryosections and dorsal marginal zone (DMZ) explants and to detect endogenous activated Rho in situ.
Since pull-down experiments from embryo extracts had shown that Rho-signaling is stronger on the dorsal than on the ventral side of the embryo during gastrulation (Habas et al.,2003; Hukriede et al.,2003), we set out to confirm these results using RBD-GFP staining. Xenopus gastrula embryos were fixed and sectioned sagitally to show the dorsal and ventral lip (Fig. 1A). RBD-GFP protein was expressed in bacteria and purified. After incubating the sections with RBD-GFP protein, we detected active Rho in the ectoderm and the involuting mesoderm but almost none in the endoderm. As expected, the fluorescent signal was stronger on the dorsal than on the ventral side (Fig. 1B,D). This result shows that quantitative differences in endogenous Rho signaling can be visualized using this method. Next, we wanted to detect differences in Rho signaling after experimental manipulation of Rho activity. Injection of constitutively active RhoA (caRho) into the ventral marginal zone led to an increase in the Rho signal as compared to the wild type (WT) ventral lip (Fig. 1B,C). Consistent with this, overexpression of dominant negative RhoA (dnRho) in the dorsal marginal zone led to a reduced level of GTP-Rho (Fig. 1D,E).
During gastrulation, the involuting mesoderm must stay separated from the outer ectoderm in order for these tissues to move past each other. This morphogenetic behavior is called tissue separation and results the formation of Brachet's cleft. Brachet's cleft in the embryo is marked by intense RBD-GFP protein staining (Fig. 1D, white arrowhead). This staining was lost together with the morphological structure in dnRho-injected embryos (Fig. 1E, open arrowhead), confirming the importance of Rho for tissue separation (Medina et al.,2004).
Rho has also been reported to play an important role in other morphogenetic processes during gastrulation and neurulation such as convergent extension and neural tube closure (Wallingford and Harland,2002; Tahinci and Symes,2003). The dorsal marginal zone (DMZ) tissue, which undergoes CE movements, can easily be explanted and cultured in vitro. Therefore, we set out to analyze Rho activity in DMZ explants. Embryos were injected only into the right side of the dorsal marginal zone so that the uninjected left side could serve as an internal control. Histone 2B fused to mRFP (H2B-mRFP) was always coinjected to mark the injected cells. At stage 10.5, the DMZs were explanted and cultured until stage 12. The fixed explants were stained with RBD-GFP protein and analyzed by confocal microscopy (Fig. 2A). Incubation with GFP protein led to very faint nonspecific staining, while RBD-GFP protein marked cell membranes and nuclei (Fig. 2B, C–F). In the right side of the DMZ injected with dnRho, the level of active Rho dropped dramatically. We observed that the cells in which Rho signaling was blocked were larger in size, often having two nuclei within the same cell (Fig. 2C', and data not shown). This phenomenon could be due to the role of Rho in cytokinesis (Drechsel et al.,1997). It was shown previously by precipitation that Paraxial Protocadherin (PAPC) is necessary for the activation of Rho in the dorsal mesoderm (Medina et al.,2004; Unterseher et al.,2004), and we wanted to confirm this result in situ. Knock-down of PAPC in the DMZ explants was achieved by injecting Morpholino oligonucleotides targeting the two PAPC alleles (MoPAPC). Loss of function of PAPC resulted in a decrease in RhoA activation as judged by RBD-GFP staining. The injected cells were often bigger than the control cells but smaller than dnRhoA-injected cells (Fig. 2D'). A PAPC construct lacking the Morpholino target sequence could rescue Rho activation when coexpressed with MoPAPC in the DMZ (Fig. 2E'). The intracellular domain of PAPC is essential for the activation of Rho (Wang et al.,2008). In agreement with this data, we could show that a PAPC construct lacking the intracellular domain (M-PAPC) could not suppress the PAPC loss-of-function phenotype (Fig. 2F').
RBD-GFP protein staining also allows the analysis of the subcellular distribution of activated Rho. This is best visualized in bipolar cells of the dorsal mesoderm where convergent extension takes place. We compared cells from embryos that were undergoing convergent extension movements with cells from embryos that had already completed this process. For these experiments, DMZs were explanted at the gastrula stage and cultured until the control embryos reached neurula stages. Whole explants or cryosections of the DMZs were stained with RBD-GFP and analyzed by confocal or standard fluorescence microscopy. At stage 15, cells of the paraxial mesoderm showed marked accumulation of active Rho at the tips, reflecting their movement towards the dorsal midline (Fig. 3A, A'). In contrast, cells of the notochord at stage 22, which have completed convergent extension, had a uniform activated Rho staining along the cell membrane (Fig. 3B,B'). These staining patterns corroborate the results from cell culture scratch-wound assays. After scratching of a mouse embryonic fibroblast (MEF) monolayer, the cells at the wound accumulate active Rho at the rear and the leading edge (Fig. 3C). Cells that are not in direct contact with the wound have a faint staining in the cytoplasm and along the membrane with no variation (Fig. 3D). The nuclear staining detected in MEFs is nonspecific (Goulimari et al.,2008).
In this work, we show that RBD-GFP staining can detect quantitative changes in Rho signaling in fixed tissues and allows the identification of zones of endogenous Rho activity in Xenopus embryos. Using fluorescent or confocal microscopy, the subcellular distribution of active Rho can be visualized at different timepoints of development.
Xenopus Embryo Manipulations and Microinjections
Embryos were obtained by in vitro fertilization, dejellied, and microinjected as previously described (Medina et al.,2004). Staging was performed according to Nieuwkoop and Faber (1967). Capped sense mRNAs for microinjections were synthesized from linearized plasmids with the SP6 mMessage mMachine kit (Ambion) according to the manufacturer's instructions. For cryosections, embryos were injected at the 4-cell stage into the ventral or dorsal marginal zone. For dorsal mesoderm explants, embryos were injected at the 8-cell stage into the right half of the dorsal marginal zone only. The injected amounts of DNA or mRNA varied depending on the construct: constitutively active (ca) V14RhoA 100 pg DNA, dominant negative (dn) N19RhoA 200 pg DNA, PAPC 200 pg mRNA, M-PAPC 200 pg mRNA, H2B-mRFP 200 pg mRNA. Morpholino oligonucleotides against both PAPC alleles (MoPAPC1 and MoPAPC2) were injected as a mixture totaling 40 ng.
Cryosections were generated largely as has been described (Fagotto and Gumbiner,1994). Embryos were fixed in MEMFA (0.1M MOPS, pH 7.4, 2 mM EGTA, 1 mM MgSO4, 3.7% formaldehyde) for 1 hr at room temperature, washed in PBS, rinsed in 100 mM Tris, pH 7.4, 100 mM NaCl for 1 hr, and washed again in PBS. The embryos were then embedded in 15% fish gelatin, 15% sucrose overnight followed by 25% fish gelatin, 15% sucrose overnight. Embryos were frozen in 15% gelatin at −80°C. Twelve-micrometer sections were cut at −19°C, collected on precoated glass slides (Fisher Scientific), and dried at 37°C overnight. The dried cryosections were fixed with acetone for 5 min and stained with RBD-GFP.
Dorsal Marginal Zone (DMZ) Explants
Dorsal marginal zone explants were prepared at stage 10.5 and cultured in 1× MBSH (88 mM NaCl, 1 mM KCl, 2.4 mM NaHCO3, 0.82 mM MgSO4, 0.41 mM CaCl2, 0.33 mM Ca(NO3)2, 10 mM HEPES, pH 7.4, 10 μg/ml streptomycine sulphate, and 10 μg/ml penicillin) until stage 12–13. Explants were fixed in MEMFA for 30 min at RT, washed 6 times in PBS, and stained.
RBD-GFP Probe and RhoA Activity Staining
His10-tagged RBD-GFP protein was purified under native conditions according to The QIAexpressionist (Qiagen). RBD in H10GFP-spacer-MCS1-vector was expressed in Rosetta DE3 Escherichia coli. Protein expression was induced with 1 mM isopropyl 1-thio-β-D-galactopyranoside at 30°C for 4 hr. The bacterial pellet was resuspended in lysis buffer (50 mM NaH2PO4, 300 mM NaCl, 10 mM imidazole, pH 8) and sonicated. The supernatant of the bacterial suspension was incubated with nickel-nitrilotriacetic acid-agarose (Qiagen) for 1 hr at 4°C. The beads were washed with washing buffer (50 mM NaH2PO4, 300 mM NaCl, 20 mM imidazole, pH 8) and eluated 3 times with elution buffer (50 mM NaH2PO4, 300 mM NaCl, 250 mM imidazole, pH 8). The eluate was dialysed overnight at 4°C in dialysis buffer (25 mM Tris, 10 mM MgCl2, 50 mM NaCl, 5% glycerol, 1 mM dithiothreitol) and concentrated with Amicon Ultra tubes. RBD-GFP protein is stable at −80°C for up to 2 months.
MEF cells were stained as described (Goulimari et al.,2005). Sections (or DMZs) were permeabilized with 0.3% TritonX-100/PBS for 10 min, washed with PBS for 15 min and with H2O for 20 sec. Specimens were blocked in 3% BSA, 5% (20%) normal goat serum, 0.1M glycine for 1 hr at room temperature. Incubation with 10 μg RBD-GFP protein in blocking solution was carried out overnight at 4°C in the dark. The sections (DMZs) were washed 3 (6) times with PBS for 10 min and mounted using Mowiol (Calbiochem). For simultaneous RBD-GFP and antibody staining, RBD-GFP protein can be added to the fluorescent secondary antibody.
Microscopy, Image Acquisition, and Processing:
Fluorescent images were acquired using a Zeiss Axiophot equipped with a Leica DC350FX camera. Confocal images were acquired using a Nikon C1Si confocal laser scanning system on a Nikon Ti fully automated inverted microscope. Maximum z-stack projection was performed using Nikon EZ-C1 Free Viewer Gold Version 3.30 build 647. Further image processing was carried out with Adobe Photoshop CS3 Extended Version 10.0.1.
We thank J.B. Wallingford for providing the H2B-mRFP plasmid and U. Engel for advice and assistance with the microscopy. The confocal images were acquired at the Nikon Imaging Center at the University Heidelberg. This work was supported by a research grant from the Deutsche Forschungsgemeinschaft to H.S. (STE-613/4-1).