XB-ART-35207Dev Biol March 1, 2007; 303 (1): 281-94.
Formation of the left-right axis involves a symmetry-breaking signal originating in the node or its equivalents, which increases TGF-beta signaling on the left side of the embryo and ultimately leads to asymmetric patterning of the viscera. DAN domain proteins are extracellular inhibitors of TGF-beta ligands, and are involved in regulating the left-right axis in chick, mouse and zebrafish. We find that Coco, a Xenopus DAN family member, and two TGF-beta ligands, Xnr1 and derrière, are coexpressed in the posterior paraxial mesoderm at neurula stage. Side-specific protein depletion demonstrated that left-right patterning requires Coco exclusively on the right side, and Xnr1 and derrière exclusively on the left, despite their bilateral expression pattern. In the absence of Coco, the TGF-beta signal is bilateral. Interactions among the three proteins show that derrière is required for normal levels of Xnr1 expression, while Coco directly inhibits both ligands. We conclude that derrière, Xnr1, and Coco define a posttranscriptionally regulated signaling center, which is a necessary link in the signaling chain leading to an increased TGF-beta signal on the left side of the embryo.
PubMed ID: 17239842
Article link: Dev Biol
Genes referenced: dand5 nodal1 gdf3 tnnt2 pitx2 lefty mmut tgfb1 tnni3
Antibodies: Ctnnb1 Ab10 Dand5 Ab1 Tnnt2 Ab1
Morpholinos: dand5 MO1 dand5 MO2 gdf3 MO1 nodal1 MO5
Article Images: [+] show captions
|Supplementary Fig. S3. Cell fate of tier C blastomeres from 32 cell stage embryos, at stage 18. 32 cell stage embryos (upper panel, nomenclature of Nakamura, descendents of the dorsal blastomere from 4 cell stage in light blue, and of the ventral blastomere in pink) were injected in blastomeres C1, C2 (dorsal descendents) and C3 (ventral descendent) with 2 ng LacZ RNA, collected at stage 18, and stained with X-Gal (blue dots). All embryos are cleared. A, B. C1 injection. A: dorsal view. B: sectioned posterior half, internal view. Most of the tracer migrates to the anterior pole, the rest is mostly axial, with some contribution to posterior paraxial mesoderm. C, D. C2 injection. C: anterior half, external view. D: posterior half, internal view. Arrow in D indicates tracer in ventral paraxial mesoderm. E, F. C3 injection. E: anterior half, internal view. Arrow indicates lateral plate mesoderm. F: posterior half, external view. Tracer is lateral to the posterior paraxial mesoderm. Bars in B and D indicate the midline.|
|Supplementary Fig. S4. Coco and Xnr1 MOs do not affect expression of posterior paraxial genes. Double in situ hybridization for the indicated markers (purple) and LacZ (red). Coco MO (3 ng, right dorsal side) and Xnr1 MO (2 ng, left dorsal side) were coinjected with LacZ RNA (1 ng) in the C2 blastomere (32 cell stage), and embryos were collected at stage 18. Inner side of dissected posterior poles is shown, dorsal side up. A-C Xnr1 in situ. D, E derrie in situ. A, D LacZ RNA injections. B, E Coco MO+LacZ RNA. C Xnr1 MO+LacZ RNA..|
|Fig. 1. Coco, Xnr1 and derrie are coexpressed in previous termXenopusnext term neurula. In situ hybridization for Coco (A, F, G, H1), Xnr1 (D, E, H2), and derrie (I, J). Panels A, D, I are inner views of posterior fragments, dorsal side up, dissected as shown in the drawing on upper left (dorsal view, anterior is up); B, C, E are cleared whole embryos, dorsal views, posterior side down; J is a cleared embryo, lateral view, anterior to the left; F is a paraxial longitudinal section of a whole-mount in situ stained embryo, G2 are in situ hybridizations of horizontal paraffin sections. Panels H1 and H2 are consecutive sections (15 μm thick), and H3 is an overlap of H1 and 2, with false color for Coco (green) and Xnr1 (red), and the overlap (yellow). Horizontal lines in H1 and H2 indicate the lateral borders of the notochord. The lower arrowhead in panel J shows the most posterior segment of derrie expression. Scale bars panels in A and B represent 0.2 mm.|
|Fig. 2. Depletion of Coco protein randomizes the left right axis. (A) Alignment of the Coco and Coco2 sequences. (B) Design of Coco MO1 and 2. A mix of both was used. (C) Coco MO specifically inhibits translation of injected and endogenous wild-type Coco RNA but not of a 5′ mutated RNA (Mut Coco). Western blots. Upper panel. 100 pg Coco or Mut Coco RNA were injected in the animal pole of each blastomere at the two-cell stage, alone (lanes 2, 5) or with 20 ng of either Coco MO (lanes 3, 6) or Coco Mut MO (lane 4). Animal caps were collected at stage 10. Coco protein was detected with anti-Coco antibody (1:10,000). Lower panel. Expression of endogenous Coco protein is inhibited by Coco MOs. The diagram represents a neurula-stage embryos, dorsal view, with the dissected dorsal posterior fragment boxed. Each lane contains 100 dissected dorsal posterior fragments (stage 18) from embryos injected dorsally with 3 ng control MO or 3 ng Coco MO. Positive control was overexpressed Coco FLAG RNA. Polyclonal antibodies for Coco (1:500) and β-catenin (1:10,000) were used. (D) Stage 46 embryos, dorsal (whole, D, H) and ventral (details) views. (D) Control embryos. (H) Embryos injected at four-cell stage in the dorsal right blastomere with 3 ng Coco MO. (E, I) Immunohistochemistry (peroxidase), (F, J) immunofluorescence for cardiac troponin. V: ventricle; A: atrium; Ao: aorta. Curved arrows indicate the direction of the outflow tract and of the gut looping, straight arrows in panels F and J indicate the antero-posterior axis. Panel I is a complete inversion (Situs inversus), panel K is an isolated gut inversion and panel L is an isolated heart inversion (Situs ambiguus). Hearts are contoured in panels G1, K1 and L1, and G2, K2, and L2 show the position of the gall bladder by UV autofluorescence in the same embryos.|
|Fig. 3. The effect of Coco MO on asymmetrically expressed genes. Stage 256 embryos were stained by in situ hybridization for Pitx2c (A), Xnr1 (D), and Xlefty-a (G). Left and right sides of the same embryos are shown. Panels G are dorsal views of cleared embryos, showing midline expression of Xlefty-a. Embryos injected in the right dorsal blastomere with 3 ng Coco MO have mostly bilateral expression of Pitx2c (B), Xnr1 (E), and Xlefty-a (H). Coinjection of Mut Coco RNA (10 pg) restores left-side expression in Coco MO-injected embryos (Pitx2c C, Xnr1 F, Xlefty-a I). Panels A, D, G are controls. Embryos in panels A were coinjected with 1 ng LacZ RNA and stained with Red Gal, to show that the right side was correctly targeted.|
|Fig. 4. Coco RNA prevents expression of left side markers. Embryos were injected on the left dorsal side of four-cell stage embryos with Coco RNA (10 pg) and collected at stage 24. In situ hybridization for Pitx2c (A, B) and Xnr1 (C, D). Left and right sides of the same embryos are shown. Arrowheads in panels C and D indicate the posterior, bilateral expression domain of Xnr1, not affected by Coco RNA.|
|Fig. 5. Depletion of Der protein on the left side randomizes the leftight axis. Embryos were injected at the four-cell stage with 2 ng Der MO in the dorsal left blastomere, and collected at stage 46 for phenotype, or 25 for marker expression. (A, B) Stage 46 embryos, ventral view. (A) Control, (B) Xnr1 MO injected (Situs inversus). Panels A1 and B1 show the normal head and trunk of the same embryos. The contour of the hearts is indicated, and arrows show the direction of the outflow tract of the heart and of gut looping. (C) Effect of Der MO on expression of Pitx2c (C, D) and Xnr1 (E, F). Left and right sides of the same embryos are shown. Embryos injected with Der MO are devoid of Xnr1 and Pitx2c expression in the anterior left LPM (D, F).|