Click here to close
Hello! We notice that you are using Internet Explorer, which is not supported by Xenbase and may cause the site to display incorrectly.
We suggest using a current version of Chrome,
FireFox, or Safari.
Dev Biol
2007 Dec 01;3121:331-43. doi: 10.1016/j.ydbio.2007.09.027.
Show Gene links
Show Anatomy links
Recruitment of Cdc42 through the GAP domain of RLIP participates in remodeling of the actin cytoskeleton and is involved in Xenopus gastrulation.
Boissel L
,
Houssin N
,
Chikh A
,
Rynditch A
,
Van Hove L
,
Moreau J
.
???displayArticle.abstract???
The transduction pathways that branch out of fibroblast growth factor signaling are essential for the induction of the mesoderm and the specification of the vertebrate body plan. One of these pathways is thought to control remodeling of the actin cytoskeleton through the Ral binding protein (RLIP also known as RalBP1), an effector of the small G protein Ral. RLIP contains a region of homology with the GTPase-activating protein (GAP) domain involved in the regulation of GTPases of the Rho family. We demonstrate here that the GAP domain of RLIP is responsible for the stability of the actin cytoskeleton in Xenopus laevis embryos. We also demonstrate that the complete N-terminal domain of RLIP containing the mu2 binding domain (mu2BD) and the GAP domain induces disruption of the actin cytoskeleton when targeted to the plasma membrane. Neither domain, however, has any effect on the actin cytoskeleton when individually targeted to the plasma membrane. We also determined that Cdc42-GDP, but neither Rac-GDP nor Rho-GDP, rescues the effect of expression of the membrane-localized Xenopus RLIP on the actin cytoskeleton. We show that the GAP domain of RLIP interacts in vivo with Cdc42-GTP and Cdc42-GDP. Finally, a single mutation (K244A) in the GAP sequence prevented embryos from gastrulating. These results demonstrate that to participate in the control of the actin cytoskeleton, RLIP needs its complete N-terminal region coding for the mu2BD and the GAP domain. We suggest that RLIP, by coordinating two complementary mechanisms, the endocytosis of clathrin-coated pits and the remodeling of cortical actin, participates in the gastrulation process.
Fig. 1. Identification of the RLIP domain that acts on the actin cytoskeleton. Myc-tagged XRLIP deletion mutants were addressed to the plasma membrane by the CAAX sequence in their C terminus. To compare their relative activities on the actin cytoskeleton, embryos at the two-cell stage were injected into the animal pole of one blastomere with 300 to 500 pg mRNA. (A) Myc-XRLIP (1–636)-CAAX (full-length) and XRLIP divided in two parts: Myc-XRLIP (1–495)-CAAX (coding for μ2BD, RhoGAP and RalBD) or the Myc-XRLIP (330–636)-CAAX (coding for RalBD and POB1/Reps) deletion mutants were tested. For XRLIP (1–636)-CAAX and the mutant containing RhoGAP, and the μ2 and Ral binding domains, depigmentation was observed. (B) Each individual domain of the XRLIP N-terminus: RhoGAP domain (XRLIP (177–387)-CAAX), the μ2 BD (XRLIP (1–230)-CAAX) or the RhoGAP domain did not induce depigmentation. Left panel: Pictures of the animal hemisphere of embryos at the large-cell blastula stage. Middle Panel: Localization of XRLIP-CAAX mutants was analyzed with the Myc antibody 9E10 and secondary antibodies conjugated to FITC. Right panel: Confocal microscopy analysis of the effect of the same mutants on the cortical actin cytoskeleton from the animal pole. The immunofluorescence images of actin cortical cytoskeleton were obtained by rhodamine–phalloidin. Disruption of cortical actin is observed in embryos injected with the same mutants that induced bleaching. On the right, schematic presentation of mutants used. The numbers in brackets represent the amino acids corresponding to the full-length XRLIP that delimits each mutant.
Fig. 2. Mutations that affect the activity of the GAP domain reduce the effect of XRLIP-CAAX on the actin cytoskeleton. (A) Effect of the mutation R208L on the GAP activity of XRLIP. Constructs of wild-type XRLIP (1–636) or XRLIP containing the RhoGAP domain alone (XRLIP (177–387)), associated with the μ2 BD and Ral binding domains (XRLIP (1–495)), associated with the μ2BD (XRLIP (1–387)) or without the 208L mutation were incubated with (γ-32P) GTP-Cdc42. The products were separated by filtration and the residual filter-bound radioactivity was determined after incubation. (B) Embryos were injected with mRNAs coding for XRLIP containing the μ2BD linked to the GAP domain carrying the CAAX sequence (XRLIP (1–387)-CAAX). Three mutations were tested, 208L, 244A and 317V. None of the embryos injected with a mutated GAP domain presented flaws in their pigmentation. (C) Analysis by Western blot of protein expression. Each lane was loaded with protein extracts corresponding to one half embryo and loading was controlled by anti-tubulin and anti-Myc antibodies.
Fig. 3. Only the dominant-negative form of Cdc42 rescues the depigmentation phenotype induced by XRLIP-CAAX. The animal hemisphere of embryos was injected with mRNAs coding for Cdc42, Rho or Rac in an inactive form, with Cdc42 WT alone, or with RLIP-CAAX containing the μ2BD and the GAP domain (amino acids 1–387). Individually, the small G proteins of the Rho family do not induce pigmentation (D, G, J and M), whereas XRLIP-CAAX coinjected with either Rho, Rac inactivated forms or Cdc42 wt induces bleaching of the animal hemisphere (H, I, K and L). Only Cdc42 17N rescues the bleaching of embryos (E and F). Right panel corresponds to an enlargement of the area delimited by the yellow rectangle.
Fig. 4. Synergistic effect of XRLIP-CAAX and Cdc42 12V C188S. (A) The animal hemisphere of embryos were injected with 200 pg XRLIP-CAAX mRNA or 50 pg Cdc42 V12 C188S mRNA alone, or coinjected with the same amounts of mRNAs. Only coinjection induced bleaching of the embryo. At least 40 embryos of two independent experiments were scored for each mRNA microinjected. (E) Each lane was loaded with protein extracts corresponding to one half embryo and tested with the appropriate antibodies (anti Myc for the XRLIP-CAAX constructs and anti-Cdc42).
Fig. 5. Cdc42 interacts with the GAP domain of XRLIP. (A) In vitro, GST-RalA, Rac and Cdc42 loaded with γ-32P GTP were mixed with or without protein extracts from 10 embryos, precipitated with glutathione-Sepharose beads, washed, eluted in SDS and separated by SDS–PAGE followed by autoradiography. Cdc42 and RalA interacted with XRLIP, whereas GST or Rac did not present a signal with anti-Myc antibody. (B) Co-immunoprecipitation (Co-IP) of endogenous Cdc42 with Myc-XRLIP or XRLIP-CAAX. (C) Co-IP of endogenous Cdc42 with different deletion mutants of XRLIP (see Fig. 1 for details of domains tested). (D) Co-IP of endogenous RLIP with Myc-Cdc42 17N or 12V. Data are representatives of at least three independent experiments.
Fig. 6. Effect of RLIP deletion mutants and GAP domain mutant induces severe on gastrulation. (A) Embryos at four-cell stage were injected into the marginal zone. (d, e and f) Vegetal view of embryos at 12.5 stage. (a, b and c) cross-section of embryos corresponding to the same stage of development. White arrows indicate the archenteron and the arrowheads the site of blastopore formation. (B) Schematic representation of the XRLIP deletion mutants without the CAAX sequence, used in embryos shown in the upper pictures. Letters in bracket corresponds to the mutants used in the pictures. (C) Embryos injected with mRNAs coding for WT XRLIP or mutants into the four blastomeres in the marginal zone. (D) Quantification of embryos with an abnormal phenotype at the 35–36 stage. Embryos were injected with 0,8 ng mRNAs in the marginal zone at the four-cell stage in the two dorsal blastomeres (grey) and in all four blastomeres (white). (E) Phenotypes of Xenopus embryos injected in the dorsal marginal zone with RNA encoding XRLIP 244A. (F) Embryos coinjected with XRLIP or XRLIP 244A and GFP mRNAs into the two dorsal blastomeres at the four-cell stage. The cells that express XRLIP are identified by GFP fluorescence during the different stages of early development. The embryos injected with RLIP 244A show fluorescence in a more posterior location than embryos injected with XRLIP.
Fig. 7. Model for recruiting of protein complex by RLIP during gastrulation. Activation of FGF sequentially recruits proteins that activate Ras. Activated Ras interacts with RalGDS that in turn activates Ral. RLIP then interacts with Ral and is used as a platform allowing interaction with many proteins such as Cdc42, through its GAP domain, and μ2 a subunit of the adaptor protein complex AP-2. Together Cdc42 and μ2 are involved in endocytosis and locally control the dynamics of cortical cytoskeleton actin. (?) Corresponds to the interaction with unidentified proteins.
