Fig 2: TCF4 and SOX11 act synergistically in corpus callosum formation. (A) Representative overview images of Luxol Fast Blue staining at the position of the AC and the caudal body of the CC. Images below the overview images are magnifications of the boxed areas. Yellow dashed lines indicate the CC crossing the midline. Red dashed lines indicate the AC. In Tcf4 and Sox11 double haploinsufficient mice, agenesis of the AC and agenesis of the splenium and caudal part of the body of the CC can be observed. Quantification of slices showing a CC is presented on the right (n=5, P-values were determined with Mann–Whitney-U tests; mean±s.d., WT 2664±137.64 µm; Tcf4Het 2320±56.57; Sox11Het 2648±77.56; Tcf4Het×Sox11Het 1816±19.60; WT versus Tcf4Het: P=0.008; WT versus Tcf4Het×Sox11Het: P=0.011; Tcf4Het versus Tcf4Het×Sox11Het: P=0.011; Sox11Het versus Tcf4Het×Sox11Het: P=0.011). Scale bars: 1000 µm. For more overview and magnification images, see Fig. S7. (B) Venn diagram of the overlap of DEGs and the targets of the Sox11 regulon in the Satb2 cluster. Possible common targets are depicted below. (C) Selection of disease associations enriched in the list of common targets of TCF4 and SOX11 in the Satb2 cluster. (D) Selection of GO terms associated with the common targets of TCF4 and SOX11 in the Satb2 cluster. GO terms for neurogenesis, neuronal differentiation and axonogenesis were enriched. (E) Schematic of Plxna2 and Dcx genes and the location of the ECRs with TCF4- and SOX11-binding capacity (green box) relative to the transcriptional start site (TSS). The magenta bar indicates the regions with possible binding sites for TCF4 (blue) and SOX11 (red). For DNA sequences, see Fig. S9. (F) Relative luciferase reporter gene activity under the control of the regulatory regions of Plxna2 and Dcx in transiently transfected HEK cells co-expressing TCF4A, TCF4B, SOX11 and a combination of either TCF4A or TCF4B with SOX11 (n=3, P-values were determined with two-tailed Student's t-tests; presented as mean±s.d., transfection with empty CAG-GFP vector was set to 1 for each regulatory region; Dcx: SOX11 versus TCF4A+SOX11: P=0.387; SOX11 versus TCF4B+SOX11: P=0.006; Plxna2: SOX11 versus TCF4A+SOX11: P=0.462; SOX11 versus TCF4B+SOX11: P=0.025). (G) For electrophoretic mobility shift assay, oligonucleotides containing potential SOX11-binding sites and E-Boxes from the Plxna2 and Dcx ECRs (for sequences, see Fig. 5E or Fig. S9) were incubated without cell extract (-), or in the presence of extracts from HEK293 cells transfected with empty expression vector (‘C’) or expression vector for the SOX11 high mobility group domain (amino acids 38-126 of SOX11) or the bHLH domain of TCF4B (amino acids 491-617 of TCF4B).

Fig 3: PKA interacts with SOX11. (a) Scatter plot showing the outcome of the quantitative MS-based interactome study for Strep/FLAG SOX11 after recombinant expression in Neuro2a cells. The x axis represents the ratios (bait vs control) of the reverse (label switch) SILAC experiments while the ratios of the forward experiments are presented by the y axis. Significant hits with a ratio >2.0 for the forward and <0.5 for the reverse experiments are shown in green. The SOX11 interactors Prkar1a (RIa), AKAP8 and AKAP12, involved in PKA regulation, are highlighted. (b) Immunofluorescent analysis of the co-localization of SOX11 and pCREB in the cortex of embryos on E15.5. SOX11 (in red), pCREB (in green), DCX (in grey). Scale bars 50 µm. (c) Immunofluorescent analysis of the co-localization of SOX11 and pCREB in the subgranular zone of the dentate gyrus (DG) of adult mice. SOX11 (in red), pCREB (in green), DCX (in grey). Scale bars 20 µm. (d) Co-immunoprecipitation assay conducted with brain extracts from embryos on E15.5. Blotting with an anti-PKAc Mouse antibody showed a more intense band in the IP: PKAc sample compared to the Input while the PKAc band in the IP: SOX11 was more intense than the one in the IP: GFP. Blotting with an anti-SOX11 Rat antibody showed that the SOX11 band on the IP: SOX11 was more intense than the one on the Input. Likewise, in the IP: PKAc sample the SOX11 band was more intense when compared with the negative control, IP: GFP. The blots presented are cropped.

Fig 4: SOX11’s non-phosphorylatable form leads to nuclear localization. (A–G) Representative line intensity plots of HEK293T cells transfected with SOX11 wildtype (WT) and mutants. The blue line represents DAPI intensity, the red line represents SOX11 intensity. (A'–G') Immunostaining for SOX11 (red) and DAPI (blue) of cells transfected with SOX11 WT and mutants were to determine the subcellular localization of SOX11 phospho-mutants. Note that the C3-Sox11pNON (B') and C3-Sox11pN3W7 (D') are almost exclusively localized in the nucleus. Arrow: cell with nuclear and cytoplasmic localization of SOX11. Arrow head: cell with only nuclear localization of SOX11. Scale bars: 20 µm. (A?–G?) Percentage of cells with nuclear localization (N) or nuclear and cytoplasmic (N + C) localization of the different SOX11 mutants.

Fig 5: Sox11 regulates stem cell activity and lineage status of Brca1.1516 mouse mammary tumour cells. (A) Expression of Sox11 in Brca1.1516 cells after CRISPR/Cas9 gene editing. qRT-PCR showing knockdown of Sox11 with guide #1 and guide #2 compared to sg-control. N=3 independent experiments. Three replicates per group per experiment. (B) Colony formation assay results from Brca1.1516 cells when Sox11 levels were reduced. N=3 independent experiments. Two replicates per group per experiment. (C) Sphere-forming capacity of sg-control and sg-Sox11-1 cells. Single cells were grown in the presence of 4% Matrigel for 10 days before image analysis. N=3 independent experiments. Two replicates per group per experiment. Scale bar: 2000 µm. (D) Western blot showing fibronectin, N-cadherin, E-cadherin, Mex3A and Rcor2 after knockdown of Sox11 (guide #1) compared to sg-control. N=3 independent experiments. Data are mean±s.d. Paired two-tailed Student's t-test.