Fig. 3. Manipulation of multiple regulatory mechanisms is required to further break constraint of mouse axial formulae.

a–d In vivo characterisation of miR-196 and Gdf11 individual and compound mouse mutant skeletal phenotypes following treatment with RA-receptor inhibitor AGN193109. a Representative embryonic day (E)18.5 skeletal preparations across genotypes. C = cervical; T = thoracic; L = lumbar; total vertebral number indicated in brackets. b Quantification of total vertebral number across genotype. Raw data is presented in the upper plot (vertical error bar = mean and standard deviation). Mean differences against the respective untreated control are presented in the lower plot as bootstrap sampling distributions. Mean differences are depicted as dots and 95% confidence intervals are indicated by the ends of the vertical error bars. n refers to the number of individual animals used for this analysis. Source data are provided as a Source data file. c Altered skeletal identity surrounding the cervico-thoracic boundary in AGN193109-treated WT and Gdf11 mutant embryos. *=anterior arch of the atlas; TA = tuberculum anterior; yellow arrow indicates a rib anlage on the 8th vertebral element. d Schematic summary of vertebral alterations observed across the AGN193109-treated Gdf11;TKO allelic deletion series, relative to WT. Numbers represent the maximum phenotypic difference observed from the untreated wild-type control. Diff = difference, S-Ca = sacral-caudal vertebrae, question marks indicate dysmorphic and non-quantifiable elements. White drawn line = partially penetrant, often incomplete or unilateral, alterations observed at the first thoracic element (Supp. Table 2).