DEV021899 Supplementary Material

Files in this Data Supplement:

• Supplemental Figure S1 -

  Fig. S1. Expression of Wnt targets in response to stabilized β-catenin is dependent on the timing, rather than the length, of activation. (A-D) Extending the length of exposure to stabilized β-catenin does not activate Pax7 in the ventral domains (4 and 5). In these experiments, TM (150 mg/kg body weight) was given at E8.5 and the embryos were harvested at E11.5. Although ventral progenitors were exposed to stabilized β-catenin for the same length of time as embryos injected at E7.5 and harvested at E10.5, ectopic Pax7 cells were not induced in domains 4 and 5. (E,F) Quantification of ectopic cells in five equally divided domains in the ventral spinal cord. Each dot represents the percentage of ectopic cells in one embryo. Each line represents the average percentage from three embryos. The total number of ectopic cells is indicated by n. (G-J) Gsh1/2 showed a time-dependent activation in response to activated Wnt signaling at different stages. Embryos were induced with TM at different stages and analyzed at E10.5 (G,H,J), or early E11.5 (I). White line, dorsoventral boundary. Ectopic Gsh1/2 cells are indicated with white arrows. Scale bar: 50 µm in A-D,I; 41 µm in G,H,J.

• Supplemental Figure S2 -

  Fig. S2. Activation of Wnt signaling in the ventral spinal cord is partially dependent on Gli3 and does not involve activation of BMP signaling. (A-C) Antibody staining of β-galactosidase showing that Cre activity is detected in a few cells within the floor plate (bound by white arrowheads). (D-I) Activation of Wnt signaling using Olig1-Cre results in partial respecification of cell fates. Although Msx1/2 proteins were found throughout the ventral spinal cord (E,F), Pax7 was not detected in pMN/pV3 domains (H,I, brackets). Instead, Pax7 was found only in more-dorsal progenitors where Olig1-Cre activity was occasionally detected. (J,K) Stabilized β-catenin did not activate BMP signaling as there was no change in the expression of phosphorylated Smad1/5/8. (L-N) Gli3 is partially responsible for the inhibition of Nkx6.1 when stabilized β-catenin is expressed. Embryos shown are at E10.5 unless otherwise indicated. Scale bar: 50 µm in D-N.

• Supplemental Figure S3 -

  Fig. S3. Removal of β-catenin using Olig1-Cre did not affect the production of Shh or the response to Shh in E9.5 embryos. (A-D) RNA in situ hybridization of Gli1 and Ptch1, two downstream targets of Shh, in E9.5 spinal cord sections. The range of expression is indicated by brackets. (E,F) RNA in situ hybridization of Shh in E9.5 spinal cord sections. At this stage, although the mutant floor plate is less compact, the number of Shh-expressing floor plate cells in mutants is comparable to that in WT. (G) The average number of Shh-expressing cells ± s.e.m. is indicated above the bar. At least three embryos, with three forelimb sections from each embryo, were used in the analysis. NS, not significant, based on Student’s t-test.

• Supplemental Figure S4 -

  Fig. S4. Expression of stabilized β-catenin or removal of the β-catenin gene does not appear to significantly alter the response to Shh signaling or the expression of genes required for Shh signaling. (A-I) RNA in situ hybridization was performed on E10.5 embryos. At this stage, most of the ventral cell types have been specified. The range of expression is indicated by brackets. Scale bar: 100 µm.

• Supplemental Figure S5 -

  Fig. S5. Deletion of the β-catenin gene in early development results in the expansion of ventral cell types. (A,C) When Olig1-Cre was used to delete the β-catenin gene, although there was expansion in V3 cell types, no changes in Olig2 or Nkx6.1 domains were detected. (B,D) Antibody staining of β-galactosidase showing the extent of Cre activity in the pMN domain. (E-J) Expansion of ventral neurons in β-catenin mutant mosaic embryos. The deletion was mediated by an inducible Cre (expressed from Gli1-CreER allele). TM (125 mg/kg body weight) was delivered at 9:00 h on E8.5. In these mosaic embryos, some cells were mutant for β-catenin whereas the others were wild-type. (E,H) Antibody staining of β-galactosidase from Rosa26-lacZ indicates the mosaic pattern of recombination in the spinal cord. White bracket indicates the expansion of Nxk2.2⁺ cells. (F,I) Expansion of Olig2⁺ pMN in the ventral spinal cord. (G,J) Expansion of the Isl1/2⁺ MN domain. Embryos were analyzed at early E10.5. The ventral expansion was observed in at least three mosaic embryos. Control embryos were from the same litter. Scale bar: 50 µm in E,F.

• Supplemental Figure S6 -

  Fig. S6. TCF/LEF-lacZ transgenic reporter was not expressed in all cells expressing stabilized β-catenin at E10.5. (A-A′′) In Olig1-Cre;Ctnnb1^gof;R26R/+ embryos, ∼96% of cells expressing ectopic Msx1/2 coexpress β-galactosidase (from R26R). The total number of cells counted was 505, from five sections. (B-B′′) TCF/LEF-lacZ reporter was expressed in a mosaic pattern in progenitors ectopically expressing Msx1/2 (arrowhead in B′), although most of the post-mitotic neurons coexpressed β-galactosidase (arrow in B′). The mosaic pattern of TCF/LEF-lacZ expression, as compared with that of the R26R (A′), suggests that cell counts using TCF/LEF-lacZ as a marker for Wnt signaling activation will be underestimated. (C) Percentage of ectopic Msx1/2 or Pax7 cells that also express TCF/LEF-lacZ in Gli1-CreER;Ctnnb1^gof;Tcf-lacZ embryos at E10.5 when TM was given at different stages. The number in parenthesis indicates the number of cells. The total number of sections used in the counting is also indicated. Between 70 and 90% of the progenitors that ectopically express Msx1/2 or Pax7 also express TCF/LEF-lacZ reporter. Because TCF/LEF-lacZ is expressed in a mosaic pattern, the percentage is likely to be underestimated. (D,E) Reduction of Irx3 in cells outside of the pMN/pV3 domain suggests possible non-cell-autonomous effects in Olig1-Cre;Ctnnb1^lof embryos.