Wild type (A, C, F, I), chato (B, D, E, G, J) and Lp mutant (H) embryos at e8.5 were assayed by in situ hybridization with markers expressed at rhombomeres 3 and 5 (Krox20; A-B dorsal and ventral view respectively), neuroepithelia (Sox2; C-E ventral views), somites (Meox1; F-H) and notochord (T; transverse sections in F-H and posterior and ventral views in I-J respectively). In some chato mutants, parts of the neuroepithelium remained open (arrowhead in D), giving the neural tube a wavy appearance. Transverse sections in F-H were hybridized with probes for both T (arrowheads) and the somitic marker Meox1 (arrows). In chato mutants the notochord tissue was embedded in the mesendoderm layer (arrowhead in G) and never formed an individualized notochord rod (arrowhead in F). The notochord of Lp mutants is wider than that of wild type embryos (F-H arrowheads; Greene et al., 1998). Expression of T in chato mutants (J) shows areas where the notochord was wider (w), thinner (t) or absent (a) as compared to wild type embryos (I, arrowhead). (r3) rhombomere 3, (r5) rhombomere 5, (NT) neural tube, (not) notochord, (som) somitic mesoderm.

Whole mount in situ hybridization with Ttr probes to wild type embryos of embryonic stages e7.5 (A, B), e8.0 (C-F) and e8.5 (G-J). Lateral and posterior views respectively illustrate how definitive endoderm (white) grows along the AP axis (length) and narrows laterally (width). Ttr highlights the extraembryonic visceral endoderm (VE). The white tissue covering the embryonic region corresponds to definitive endoderm (DE). Arrows point to white extraembryonic tissue (ext). All pictures were taken at the same magnification. At e7.5, some visceral endoderm cells (arrowhead in A-B) still overlay the exterior of the embryonic region (continuous line in A-B delimits embryonic-extraembryonic parts). Gut closure prevented visualization of all the definitive endoderm in panels I and J. (b’-j’) Representative transverse sections of the embryos in the columns above counterstained with Fast Red. Sections correspond to intermediate levels along the anterior-posterior axis. Only half of the section is shown, with the midline located at the right edge of the panel. (K) Plot of length (blue) and width (green) definitive endoderm measurements in wild type embryos of different stages; data in μm. Dimensions of the endoderm were taken as exemplified by dotted lines in panels G-H. Note that measurements were taken in non-Ttr stained embryos, in which transparency of the tissue allowed for accurate measurements of the whole definitive endoderm. Error bars indicate standard deviation. See Table on Fig. 5H for primary data.

(A) Plot of the average definitive endoderm cell number in wild type (grey) and chato mutants (red). Cells were counted in sections of the definitive endoderm stained with Fast Red at medial levels along the anterior-posterior axis (Fig. 4). Error bars represent standard deviation. (B) Table shows average number of cells ± standard deviation. Number of sections indicates the total number of sections counted for each condition. na, not assayed. *** p<0.0001

Wild type (A, C, E, G) and chato mutant (B, D, F, H) embryos were assayed by in situ hybridization with markers expressed in head mesenchyme-lateral plate mesoderm-somitic mesoderm (Twist; A-B dorsal and ventral view respectively), somites (Meox1; E-F ventrolateral views) and cardiac mesoderm (Nkx2.5; G-H ventral views). Staining for β-galactosidase activity from a nodal-lacZ reporter labeled the lateral plate mesoderm and node of wild type (C) and chato mutant (D) embryos (lateral views). 33% chato mutants (n=184) had condensed somites that appeared narrow and laterally extended (F). In 52% of chato embryos (n=184) somites were not clearly discernible morphologically, but somite markers Twist and Meox1 were expressed at both sides of the midline and marked some imperfectly shaped somites (B, solid arrowheads and not shown). Only 15% chato mutants showed normal somites. Empty arrowheads in A-B point to head mesenchyme. Brackets in C-D highlight the different width of the lateral plate mesoderm in wild type and chato mutant embryos. Brackets in E-F highlight the different width of the somites. Empty arrowheads in H mark the cardiac mesoderm in chato mutants, which fails to fuse at the midline of the embryo. (LPM) Lateral Plate Mesoderm. (som) somites.

(A) Domain structure of Zfp568, which contains two KRAB-A (green)/ KRAB-B (yellow) domains and eleven Zinc fingers (purple). The red star marks the position of the chato point mutation. Red arrow points to the truncation caused by the RRU161 genetrap allele. (B) Sequence comparison of the first KRAB-A domain of Zfp568/Chato with the KRAB-A consensus sequence. Conserved residues are highlighted in green. Grey bars underline residues required for transcriptional repression (Margolin et al., 1994). Red letters indicate the Leu to Pro change caused by the chato point mutation. (C-F) Complementation test between chato and RRU161 genetrap alleles: wild type (F) and mutant embryos of the allele combinations indicated in the panels (C-E) were assayed by double in situ hybridization with T and Meox1 probes. The overall embryonic morphology, as well as defects in somites and midline, are indistinguishable between the different Zfp568 allele combinations. Notochord expression of T was irregular, showing a variable width and interruptions (arrows in D and E). (G-L) In situ hybridization with a Zfp568 probe on wild type embryos at e7.5 (G-H) and e8.5 (J, L). RRU161 mutant embryos, which generate truncated Zfp568 transcripts, were used as negative controls (I, K). Zfp568/chato is expressed in all embryonic and extraembryonic tissues, as confirmed in transverse embryonic sections (H, K-L). Zfp568 was expressed at higher levels in the extraembryonic ectoderm (solid arrowheads). (me) mesoderm, (ect) ectoderm, (end) endoderm, (NT) neural tube, (se) surface epithelia.

Wild type (A-B), chato (C-D) mutant 8 somite stage embryos hybridized with Ttr probes to highlight extraembryonic visceral endoderm (blue) and definitive endoderm (exterior layer of embryonic tissues in white). ext indicates white extraembryonic tissue. (A, C) lateral; (B, D) anterior views. (E) Plot of wild type (blue) and chato (red) definitive endoderm length. (F) Plot of wild type (green) and chato (red) definitive endoderm width. Data in μm. (G) Plot of definitive endoderm LWR in wild type (grey) and chato mutant (red) embryos. Error bars indicate standard deviation. * p<0.05; ** p<0.01 (H) Length and width average measurements ± standard deviation in μm. The number of embryos analyzed for each stage is indicated (# embryos). LWR, length to width ratio; na, not assayed.

Cryosections of wild type (A, C) and chato mutant (B,D) embryos at different embryonic stages were labeled using anti-E-cadherin antibodies (red) and Phospho-Histone H3 antibodies (green). Mitotic cells (green in A-D) in the definitive endoderm (highlighted in red by localization of E-cadherin) are indicated by empty arrowheads. E-cadherin is also present in e7.5 epithelia (ep), embryonic surface ectoderm (se) and extraembryonic visceral endoderm (VE, dashed line). Proliferation of mesoderm and epithelial tissues was not significantly affected between wild type (n=21embryos/380 sections) and chato mutant embryos (n=3 embryos/56 sections) at these stages.