QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online July 21, 2003
doi: 10.1242/10.1242/dev.00615

This Article Summary Full Text Full Text (PDF) Supplemental Data Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Takebayashi-Suzuki, K. Articles by Suzuki, A. Search for Related Content PubMed PubMed Citation Articles by Takebayashi-Suzuki, K. Articles by Suzuki, A.

Interplay between the tumor suppressor p53 and TGFß signaling shapes embryonic body axes in Xenopus

Kimiko Takebayashi-Suzuki^1,*, Jun Funami^1,*, Daisuke Tokumori^1,*, Akira Saito², Tetsuro Watabe², Kohei Miyazono², Akifumi Kanda¹ and Atsushi Suzuki^1,

¹ Institute for Amphibian Biology, Hiroshima University Graduate School of Science, Kagamiyama 1-3-1, Higashi-Hiroshima 739-8526, Japan
² Department of Molecular Pathology, Graduate School of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan

View larger version (20K): [in a new window]   Fig. 1. Isolation of Xenopus p53 gene as a posteriorizing factor. (A) Expression screening strategy. A gastrula expression library was divided into fractions and mRNA was synthesized from each pool. The synthetic library mRNA was injected in combination with noggin mRNA (200 pg), an anterior neural inducer, in the animal pole of two-cell stage embryos and ectodermal explants (animal caps) were isolated at blastula stages. The positive pools were identified by analyzing the expression of posterior neural marker genes by RT-PCR. (B) Animal caps injected with different amounts of xp53 mRNA (0-400 pg) and noggin (200 pg) mRNA were subjected to RT-PCR at neurula stages (stage 21). `-RT' indicates sibling control embryos processed without reverse transcriptase. `Uninj' indicates uninjected caps. Histone was used as a loading control.

View larger version (61K): [in a new window]   Fig. 2. Overexpression of xp53 induces mesodermal and endodermal markers. (A,B) Overexpression of xp53 causes elongation of animal caps. Uninjected (A) and xp53 mRNA (400 pg)-injected (B) animal caps were photographed at early neurula stage. (C,D) Xenopus and human p53 induce mesodermal and endodermal markers. Animal caps injected with different amounts of p53 mRNA were collected at early gastrula (stage 10; C) and neurula (stage 20; D) stages, and expression of marker genes was analyzed by RT-PCR. Injected mRNAs are 0.1-1.0 ng of xp53 (C; lanes 4-8), 0.2-0.8 ng of xp53 (D; lanes 4-6), 0.8 ng of xp53 (R255T) (D; lane 7) and 0.05-0.8 ng of human p53 (D; lanes 8-12). Although human p53 induced marker gene expression more efficiently than xp53, this is likely to be due to the more efficient translation of human p53, as revealed by in vitro translation (data not shown). (E) Deletion of the C-terminal regulatory domain (RD) enhances xp53-mediated transcription. The ability of wild type and RD mutant of xp53 to induce marker genes was compared by animal cap assays. Note that 100 pg of xp53 RD induces Xhox3 and HoxB9 genes to an extent similar to that obtained by 400 pg of wild-type xp53 (lanes 5 and 6).

View larger version (20K): [in a new window]   Fig. 3. xp53 does not require de novo protein synthesis to induce Xhox3 and Mix.1. (A) Schematic representation of xp53 constructs used in this study. xp53 contains an N-terminal transactivation domain (TAD), a central DNA-binding domain (DBD), a tetramerization domain (TD) and a C-terminal regulatory domain (RD). xp53:GR contains the hormone binding domain of the glucocorticoid hormone receptor (GRHBD) at the C terminus. (B) Conditional activation of xp53:GR by dexamethasone (DEX). Animal caps expressing xp53:GR were prepared at blastula stages (stage 9) and cultured in the absence or presence of 20 µM DEX. The expression of molecular markers was detected by RT-PCR at the stages indicated. Animal caps expressing xE2F(1-88):GR are used as a control for expression of GRHBD (Suzuki and Hemmati-Brivanlou, 2000). (C) xp53 does not require de novo protein synthesis to induce target gene expression. Animal caps expressing xp53:GR were treated with cycloheximide (CHX) for 30 minutes, and then transferred into medium containing both CHX and DEX to activate xp53:GR. The expression of marker genes was detected by RT-PCR after 3 hours of DEX treatment (equivalent to stage 11).

View larger version (45K): [in a new window]   Fig. 4. xp53 interacts with activin and BMP pathways downstream of the receptor activation. (A) A dominant-negative activin type II receptor ( ActR) does not interfere with the marker gene induction by xp53:GR. Animal caps injected with xp53:GR (1 ng) alone or together with ActR (2 ng) were treated with DEX to activate xp53:GR. The expression of marker genes was determined at the gastrula stage (stage 10.5) by RT-PCR. ActR is able to inhibit marker gene expression induced by activin protein (lanes 6 and 7). (B) ActR does not interfere with transcription from a Mix.2 reporter gene induced by xp53:GR. Animal caps injected with a Mix.2 reporter gene in combination with indicated RNA [xp53:GR (1 ng), activin (15 pg) and ActR mRNA (1 ng or 2 ng)] were treated with DEX and harvested at mid-gastrula stage for luciferase assay. The columns indicate the averages of duplicate assays and the error bars indicate the ranges. (C) An antisense morpholino oligonucleotide against xp53 inhibits translation of xp53 mRNA in vitro. In vitro translation of xp53 or xp53Nmut:GR was performed in the presence or absence of antisense xp53 morpholino oligonucleotide (xp53-MO) or five mismatched control-MO (5mis-MO). The [35S]-labeled translation products were analyzed on a SDS-PAGE gel. (D) xp53-MO inhibits endogenous p53 transcriptional-activating activity. Whole embryos were injected in the animal pole with xp53-MO (170 ng) or 5mis-MO (170 ng) at the four-cell stage and followed by injection of a p53 reporter plasmid, p53 (X3), alone or together with xp53Nmut:GR at the eight-cell stage. The injected embryos were subjected to the dual luciferase assay at the gastrula stage. `Vector' indicates a reporter vector lacking p53-responsive elements. The columns indicate the averages of duplicate assays and the error bars indicate the ranges. (E) Knockdown of xp53 partially inhibits target gene expression induced by activin and BMP pathways. xp53-MO (170 ng) or 5mis-MO (170 ng) were injected into two animal blastomeres at the two-cell stage and followed by injection of either Smad2 (1 ng) or Smad1 (2 ng) mRNA into four animal blastomeres of four-cell embryos. xp53Nmut:GR mRNA was injected with Smad mRNA to rescue the effect of xp53-MO (lanes 7 and 8). The injected embryos were treated with DEX to activate xp53:GR from stage 7, and animal caps isolated at stage 9 were subjected to RT-PCR at stage 11.

View larger version (29K): [in a new window]   Fig. 5. xp53 interacts with activin and BMP pathways to regulate the expression of Mix.2 and binds to Smads. (A) Schematic representation of Mix.2 reporter constructs. Open circles indicate putative binding sites for FAST-1, Smad and p53. Arrows indicate the putative transcription start site of the Mix.2 gene. (B) The putative p53-binding site is important for Mix.2 transcription induced by activin. A wild-type or mutant Mix.2 reporter gene was injected with or without activinßB mRNA (5 pg or 15 pg) in the animal pole of four-cell embryos. Animal caps were isolated at blastula stages and harvested at mid-gastrula stage (about 3 hours after isolation) for luciferase assay. `Vector' indicates a negative control vector, pGL3-Basic (Promega). The columns indicate the averages of duplicate assays and the error bars indicate the ranges. Four other independent experiments gave similar results. (C) The putative p53-binding site is important for Mix.2 transcription induced by the BMP signal. A wild-type or mutant Mix.2 reporter was injected with or without CA-ALK2 mRNA (150 pg) in the animal pole of four-cell embryos. The luciferase assay was performed and presented as described in B. Six other independent experiments gave similar results. (D) FAST-1 and Smad-binding sites are important for p53-mediated induction of Mix.2 reporter. A wild-type or mutant Mix.2 reporter was injected with or without xp53:GR mRNA (1 ng) in the animal pole of four-cell embryos. Animal caps were isolated at blastula stages and treated with DEX for 3 hours before determination of luciferase activity. The luciferase assay was performed and presented as described in B. Three other independent experiments gave similar results. (E) xp53 binds to R-Smads. Myc-tagged xp53 (Myc-xp53) or xp53 RD (Myc-xp53 RD) was co-precipitated with FLAG-tagged R-Smads (Smad1, 2, 3, 5 and 8), but not with FLAG-tagged Co-Smad (Smad4) when expressed in COS-7 cells. `BG' indicates nonspecific background signals. IP, immunoprecipitation; IB, immunoblotting.

View larger version (37K): [in a new window]   Fig. 6. p53 binds to Mix.2 gene in vitro and in vivo. (A) p53 binds to the putative p53-binding sites from Mix.2 gene in vitro. Cell extracts from embryos injected with FLAG-tagged human p53 mRNA (lanes 2-7) or uninjected embryos (lane 1) were incubated with a labeled double-stranded oligonucleotide. Unlabeled wild-type (WT) or mutant (Mut) oligonucleotides were added in eightfold (lanes 3 and 5) or 40-fold (lanes 4 and 6) molar excess over labeled oligonucleotide. Arrowheads indicate protein-DNA complex or supershifted complex (supershift). (B) xp53 binds to the proximity of Mix.2 gene in vivo in response to activin and BMP signals. Soluble chromatin was prepared from embryos injected with mRNA as indicated and immunoprecipitated (IP) with antibody against Myc tag. The final DNA extractions were amplified using pairs of primers that cover the regions of Mix.2 and goosecoid (Gsc) genes (see Materials and Methods). `input' represents a portion of the sonicated chromatin prior to immunoprecipitation.

View larger version (58K): [in a new window]   Fig. 7. Functional knockdown of xp53 affects mesoderm formation in embryos. (A) xp53-MO or 5mis-MO (170 ng each) was injected into the ventral and dorsal sides of two-cell stage embryos as shown on the right (animal view). Embryos injected with xp53-MO showed truncation of trunk and posterior structures. (B-G) Whole-mount in situ hybridization for Xbra (B,E), muscle actin (C,F) and -globin (D,G) genes in embryos injected with 5mis-MO (B-D) or xp53-MO (E-G). Embryos injected with xp53-MO demonstrate slightly reduced level of expression of these marker genes.