pubmed.ncbi.nlm.nih.gov

MAB21L2, a vertebrate member of the Male-abnormal 21 family, modulates BMP signaling and interacts with SMAD1 - PubMed

  • ️Thu Jan 01 2004

MAB21L2, a vertebrate member of the Male-abnormal 21 family, modulates BMP signaling and interacts with SMAD1

Danila Baldessari et al. BMC Cell Biol. 2004.

Abstract

Background: Through in vivo loss-of-function studies, vertebrate members of the Male abnormal 21 (mab-21) gene family have been implicated in gastrulation, neural tube formation and eye morphogenesis. Despite mounting evidence of their considerable importance in development, the biochemical properties and nature of MAB-21 proteins have remained strikingly elusive. In addition, genetic studies conducted in C. elegans have established that in double mutants mab-21 is epistatic to genes encoding various members of a Transforming Growth Factor beta (TGF-beta) signaling pathway involved in the formation of male-specific sensory organs.

Results: Through a gain-of-function approach, we analyze the interaction of Mab21l2 with a TGF-beta signaling pathway in early vertebrate development. We show that the vertebrate mab-21 homolog Mab21l2 antagonizes the effects of Bone Morphogenetic Protein 4 (BMP4) overexpression in vivo, rescuing the dorsal axis and restoring wild-type distribution of Chordin and Xvent2 transcripts in Xenopus gastrulae. We show that MAB21L2 immunoprecipitates in vivo with the BMP4 effector SMAD1, whilst in vitro it binds SMAD1 and the SMAD1-SMAD4 complex. Finally, when targeted to an heterologous promoter, MAB21L2 acts as a transcriptional repressor.

Conclusions: Our results provide the first biochemical and cellular foundation for future functional studies of mab-21 genes in normal neural development and its pathological disturbances.

PubMed Disclaimer

Figures

Figure 1
Figure 1

Expression of Mab21l2 gene and protein in Xenopus laevis development. A) Double whole mount in situ hybridization for Xmab21l2 (blue) and Xkrox20 (brown). At the end of neurulation (dorsal view of a st. 19 embryo is shown) Xmab21l2 expression (arrows) is restricted to the eye primordium (bottom arrowhead) and midbrain (top arrowhead). Xkrox20 labels the third and fifth rhombomeres (arrows). B) Frontal view of a stage 24 embryo. Xmab21l2 expression (blue, arrows) is restricted to the eye (left, solid arrow) and midbrain (right, solid arrow), clearly posterior to the forebrain marker Emx2 expression domain (brown, empty arrow). C–E: Sections obtained from late tadpole embryos were wholemount-hybridized with Xmab21l2. Three transverse sections, from anterior to posterior, show Xmab21l2 expression in the retina and lens (Ey), in the branchial arches (BA), midbrain (arrow in C), hindbrain (arrow in D), and in the dorsal neural tube, (SC in E). F: Western blot analysis of MAB21l2 protein expression in Xenopus oocytes (stage 0) and Xenopus embryos (stages 3–23). ME: day 12 mouse embryo (positive control). Protein concentrations were quantitated through a Bradford assay and loaded on gel in equal amounts. The experiment shows maternal expression of MAB21l2 (41 kD, arrow), the levels of which decrease around blastula stage to resume zygotically thereafter. In particular, stage 11 gastrulae are positive for the MAB21l2 protein.

Figure 2
Figure 2

Isolated overexpression of Xmab21l2 produces partially dorsalized embryos. A, C, E: control-injected embryos; B, D, F: Xmab21l2-injected embryos. A–D: tailbud-tadpole stages; E, F: gastrula stage. In B: external appearance of Xmab21l2-injected embryos. D: horizontal section of an Xmab21l2-injected section stained with orange-G/aniline blue. Note enlarged notocord (nc). F: expanded and enhanced Chordin expression in Xmab21l2-injected gastrulae (blue signal).

Figure 3
Figure 3

Xmab21l2 rescues dorsal structures in BMP4 injected embryos. (A) As previously described, injection of 0.6 ng of BMP4 mRNA gave rise to ventralized embryos. The most severely ventralized phenotype of BMP4 injection, called bauchstück, was scored with a dorso-anterior index (DAI) of 0 (top), whereas normally dorsalized embryos were assigned a DAI of 5 (bottom). (B) Embryos were injected with 0.6 ng BMP4 (n = 111), with 0.6 ng BMP4 plus 0.8 ng Mab21l2 (n = 83), or with 0.6 ng BMP4 plus 1.6 ng Mab21l2 (n = 101). In each group, we scored the percentage of complete ventralization (DAI = 0, grey bars) and normal dorsal axis formation (DAI = 5, black bars) in embryos that completed development. Intermediate classes are omitted in the plot. Coinjection of embryos with Mab21l2 significantly increased the percentage of correctly dosalized embryos (see text for details). Statistical analysis was conducted using the Chi square algorithm (1 df). *: p = 0.0005; **: p = 0.0087.

Figure 4
Figure 4

Xmab21l2 restores the normal expression of Xvent2 and Chordin in BMP4-injected embryos. Chordin (A–C) and Xvent2 (D–E) expression were analyzed by whole mount in situ hybridization. Embryos are shown in a vegetal view, dorsal to the top, ventral to the bottom. (A, D) Wild type expression of Chordin (A) and Xvent2 (D) in stage 10.5 gastrulae; (B, E) embryos injected with 1.2 ng of BMP4 mRNA alone showed reduced Chordin expression (B) (22/25) and expanded Xvent2 expression (E) (16/25); (C, F) embryos co-injected with 1.2 ng of BMP4 and 1.6 ng Xmab21l2 mRNA showed a rescue of wild type Chordin (C) (17/36) and Xvent2 expression (F) (24/29). (G) Percentage of embryos exhibiting wild type (black bars) or reduced (grey bars) Chordin gene expression in embryos injected with 1.2 ng BMP4 alone, and in embryos coinjected with 1.6 ng Mab21l2. Coinjection of Mab21l2 significantly increased the number of embryos exhibiting a wild type Chordin expression pattern; *: p = 0.004. (H) Percentage of embryos exhibiting wild type (black bars) or expanded (grey bars) Xvent2 gene expression in embryos injected with 1.2 ng BMP4 and in embryos coinjected with 1.6 ng Mab21l2. Again, Mab21l2 significantly increased the number of embryos exhibiting a wild type Xvent2 expression pattern; **: p = 0.00044. Statistical analysis was conducted using the Chi square algorithm (1 df).

Figure 5
Figure 5

Mab21l2 interacts with Smad1 in vitro and in vivo. (A) Lanes 1, 2: immunoblotting of total lysates from P19 cells, mock- and Smad1-transfected, respectively. P19 cells do not express detectable levels of endogenous SMAD1. Lanes 3–8: Affinity chromatography (pull-down) experiment. High-stringency eluates from untransfected P19 were separated by SDS-PAGE and transfered onto nitrocellulose filters. The blots were immunolabeled using an anti-SMAD1 antibody. Lanes 3–6: the in vitro interaction is not strictly dependent on activation of BMP signaling: the pull-down experiment was performed with 20 μl (3, 4) and 40 μl of P19 cell lysates (5, 6). Lysates came from P19 cells treated (4, 6) or untreated (3, 5) with BMP4. Lanes 7, 8: an His-ZZ-MAB21L2 protein (HisMab) synthesized in E. coli (lane 8) was coupled to a sepharose-Ig resin and incubated with cell lysates of P19 cells treated with BMP4. As a negative control, an in-vitro synthesized His-ZZ incubated with the same P19 lysates fails to pull down a 53 kD band. Arrows: SMAD1 (53 kD). (B) direct interaction between SMAD1 and MAB21L2; no direct interaction between SMAD4 and MAB21L2. An His-ZZ-MAB21L2 protein (+Mab) synthesized in E. coli (lane 2) was coupled to a sepharose-Ig resin and incubated with in vitro-translated SMAD1 (lanes 2), SMAD4 (lanes 4), and SMAD1 + SMAD4 (lanes 6). Negative controls were represented by His-ZZ-coupled resins (-Mab) incubated with the same in vitro-synthesized proteins (lanes 3, 5, 7). Lanes 1 and 8 contain in vitro-synthesized SMAD1 and SMAD4, respectively. (C) BMP4-dependent in vivo co-immunoprecipitation of flag-SMAD1 and myc-MAB21L2 in P19 cells. Cells were mock-transfected, transfected with flag-SMAD1 and/or with myc-MAB21L2, as indicated, and either treated with BMP4 or left untreated. In lanes 3–7, cell lysates were immunoprecipitated with a monoclonal anti-flag antibody. Cell lysates in lanes 1, 2 and immunoprecipitates in lanes 3–7 were gel fractionated and transfered to nitrocellulose filters. Blots were immunolabeled with an anti myc antibody. Arrow (42 kD) points to a band in lanes 2, 3) corresponding to myc-MAB21L2. (D) in vivo co-immunoprecipitation of flag-SMAD1 and myc-MAB21L2 in stage 11 Xenopus embryos, facilitated by BMP4 overexpression. Embryos were water-injected, injected with flag-Smad1 and/or with myc-Mab21l2 RNA, as indicated, and coinjected with BMP4 where indicated. In lanes 3–7, cell lysates were immunoprecipitated with a monoclonal anti-flag antibody. Embryo lysates in lanes 1, 2 and immunoprecipitates in lanes 3–6 were gel fractionated and transfered to nitrocellulase filters. Blots were immunolabeled with an anti myc antibody. Arrow (42 kD) points to a band in lanes 2–4) corresponding to myc-MAB21L2.

Figure 6
Figure 6

When targeted to a heterologous promoter, Mab21l2 behaves as a strong transcriptional repressor. (A) An in vitro-translated chimeric protein containing the GAL4 DNA binding domain fused to the N-terminus of MAB21L2 was used in an electrophoretic mobility shift assay. An end-labeled ds-DNA containing the GAL4 nucleotide binding site (5XUAS) was incubated in native conditions with in vitro translated GAL4-MAB21L2 (G-M) and separated electrophoretically by nondenaturing PAGE. As negative controls, we used free oligonucleotides (F) and rabbit reticulocyte lysates (RRL). As a positive control we used a previously tested GAL4 fusion protein, GAL4-D9NT (G-D9) [40]. The arrow indicates a bandshift specific for GAL4-MAB21L2 and absent in negative controls. u: unbound. (B) 5XUAS-luc transcription is clearly downregulated in cells transfected with GAL4-MAB21L2 (G-M) vs. untransfected COS7 cells or same cells transfected with GAL4 DBD, G-D) alone (*: p = 0.0019). No downregulation of 5XUAS-luc activity was observed in COS7 cells transfected with a fusion of GAL4 and the N-terminal domain of the HOXB3 protein, which possesses no transcriptional activation or repression activity [24] (ns: not significant). Statistical analysis was conducted using the T test method (two tails).

Similar articles

Cited by

References

    1. Chow KL, Emmons SW. HOM-C/Hox genes and four interacting loci determine the morphogenetic properties of single cells in the nematode male tail. Development. 1994;120:2579–92. - PubMed
    1. Chow KL, Hall DH, Emmons SW. The mab-21 gene of Caenorhabditis elegans encodes a novel protein required for choice of alternate cell fates. Development. 1995;121:3615–26. - PubMed
    1. Margolis RL, Stine OC, McInnis MG, Ranen NG, Rubinsztein DC, Leggo J, Brando LV, Kidwai AS, Loev SJ, Breschel TS, Callahan C, Simpson SG, DePaulo JR, McMahon FJ, Jain S, Paykel ES, Walsh C, DeLisi LE, Crow TJ, Torrey EF, Ashworth RG, Macke JP, Nathans J, Ross CA. cDNA cloning of a human homologue of the Caenorhabditis elegans cell fate-determining gene mab-21: expression, chromosomal localization and analysis of a highly polymorphic (CAG)n trinucleotide repeat. Human Molecular Genetics. 1996;5:607–16. doi: 10.1093/hmg/5.5.607. - DOI - PubMed
    1. Smith M, Woodroffe A, Smith R, Holguin S, Martinez J, Filipek PA, Modahl C, Moore B, Bocian ME, Mays L, Laulhere T, Flodman P, Spence MA. Molecular genetic delineation of a deletion of chromosome 13q12-->q13 in a patient with autism and auditory processing deficits. Cytogenet Genome Res. 2002;98:233–9. doi: 10.1159/000071040. - DOI - PubMed
    1. Mariani M, Corradi A, Baldessari D, Malgaretti N, Pozzoli O, Fesce R, Martinez S, Boncinelli E, Consalez GG. Mab21, the mouse homolog of a C. elegans homeotic regulator, participates in cerebellar, midbrain and eye development. Mechanisms of Development. 1998;79:131–135. doi: 10.1016/S0925-4773(98)00180-4. - DOI - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources