Pax6 limits the competence of developing cerebral cortical cells to respond to inductive intercellular signals
Fig 9
Morphogen regulation of ectopic gene expression following Pax6 loss.
(A) The experimental procedure for (B-E): TAM was administered at E9.5 to delete either one (control) or both (cKO) Pax6 allele(s), with Cre-deleted cells expressing GFP; coronal slices were cultured on E13.5 with Bmp4 or vehicle alone for 2 DIV; slices were analysed using qRT-PCR or sectioned. (B) Concentration-response measured using qRT-PCR: Gsx2 levels (averages ± SEM; values were relative to the average level in control cortex treated with 0 Bmp4) in control and Pax6 cKO slices with increasing concentrations of Bmp4 (n = 3 independent cultures at each concentration). Two-way ANOVA showed significant effects of genotype (p < 0.001), of Bmp4 concentration (p < 0.005), and an interaction effect (p < 0.01). Differences between genotypes at each Bmp4 concentration were tested with Bonferroni’s method for comparison of means (****, p < 0.001) (S9 Data). (C) Immunoreactivity for Gsx2 and GFP in telencephalic slices from Pax6 cKOs cultured with vehicle alone or Bmp4. Scale bar: 0.1 mm. (D) Concentration-response measured using qRT-PCR: Prdm13 levels (averages ± SEM; values are relative to the average level in Pax6 cKO cortex treated with 0 Bmp4) in control and Pax6 cKO slices with increasing concentrations of Bmp4 (n = 3 independent cultures at each concentration). Two-way ANOVA showed significant effects of genotype (p < 0.005), of Bmp4 concentration (p < 0.05), and an interaction effect (p < 0.05). Differences between genotypes at each Bmp4 concentration were tested with Bonferroni’s method for comparison of means (***, p < 0.005) (S9 Data). (E) In situ hybridizations for Prdm13 and immunoreactivity for GFP in telencephalic slices from Pax6 cKOs cultured with vehicle alone or Bmp4. Green arrows indicate Prdm13 expression in lateral cortex. Scale bar: 0.1 mm. (F) A hypothesis of how Shh and Bmp4 might combine to generate the observed spatial patterns of Gsx2 and Prdm13 expression in the embryonic cortex after Pax6 deletion. Deletion might increase the probability of Gsx2 being activated in cells exposed to physiological levels of Shh above a threshold (broken line). In the medial cortex, exposure to the levels of Bmp above a threshold (central broken line) might reduce the probability of Gsx2 activation. Cells exposed to intermediate levels of Bmp (between upper and lower broken lines) might have an increased probability of expressing Prdm13. (G) Waddington’s epigenetic landscape analogy, used to illustrate our main conclusions. A saddle-node bifurcation illustrates Pax6’s normal action, closing a valley on the left (RGP). Pax6 deletion opens this valley, creating a subcritical pitchfork bifurcation where cells emerging from the transition state can enter either of 2 valleys (eGC). Increasing exposure to Shh tilts the landscape to the left making it more likely that the cell will enter the open valley on the left; increasing exposure to Bmp has the opposite effect. DIV, day in vitro; eGC, ectopic GABAergic cell; GFP, green fluorescent protein; Pax6 cKO, Pax6 conditional knockout; qRT-PCR, quantitative real-time PCR; RGP, radial glial progenitor; TAM, tamoxifen.