Fig 2: Identification of specific mutations of SMARCA4 in CRC patients.a Oncoprint plot showing the genomic profiles of SMARCA4, SMARCA2, SMARCB1, SMARCC1, SMARCC2, SMARCE1, ARID1A, ARID1B, and other CRC relevant genes, data compiled from the cBioPortal based on original data from TCGA CRC project. b Detection of the SMARCA4 gene showing the c. 3721C>T (p. Arg1157Trp) mutation in tumor samples of CRC patients using DNA sequencing.

Fig 3: An ARID1B patient had enlarged ganglionic eminence.a, Graph representation of AR, PBX1, DLX6, OTX2, and OLIG1 TF modules, which are most strongly enriched in ARID1B DEG (Two-sided Fisher exact test p-value < 0.01, top 4 odds ratio). The targets highlighted in yellow are SFARI genes, and the targets highlighted in black are TF modules enriched in SFARI genes. b, Top, sanger sequencing showing a duplication mutation (red arrow) identified in one of the alleles of patient 1. Bottom, sanger sequencing showing the mutation was repaired and this cell line is used as an isogenic control. Sequencing was done on fragments amplified from genomic DNA extracted from the two iPS cell lines. c, SNP array genotyping of PBMCs and iPS cells from patient 2 shows a microdeletion (dark green arrow) identified on chromosome 6. d-f, Bar plot showing the absolute GE (d), cortex (e) and total brain (f) volumes of control (mean ± S.D.; n = 5, age macheted) and patient 2 at two gestational stages. g-h, Bar plots showing normalized GE to cortex volume (g) and total brain volume (h) of control (mean ± S.D.; n = 5, age macheted) and patient 2 at two gestational stages.

Fig 4: Non-canonical Activin signaling pathways are elevated after loss of Arid1b.a Western blot of ARID1B, p-SMAD2, SMAD2, p-AKT, AKT, p-P38, P38, p-ERK, ERK, p-JNK, JNK, p-COFILIN, COFILIN, p-NF-κB, NF-κB, and β-ACTIN in proximal incisor mesenchyme from control and Gli1-CreER;Arid1bfl/fl mice. b–g Immunostaining of p-COFILIN (red) with β-gal (green) at Gli1-LacZ mouse incisor (b, e) and comparison of p-COFILIN expression in control (c, f) and Gli1-CreER;Arid1bfl/fl (d, g) mouse incisors. e–g represent high-magnification images of boxes in (b–d), respectively. White dotted lines outline the cervical loop. Yellow arrows indicate positive cells. n = 3. h–m Immunostaining of p-JNK (red) with β-gal (green) in Gli1-LacZ mouse incisor (h, k) and comparison of p-JNK expression in control (i, l) and Gli1-CreER;Arid1bfl/fl (j, m) mouse incisors. k–m represent high-magnification images of boxes in (h–j), respectively. White dotted lines outline the cervical loop. Yellow arrows indicate positive cells. n = 3. n–o Immunostaining of p-ERK (red) with β-gal (green) in Gli1-LacZ. o represents the high-magnification image of (n). White dotted line outlines the cervical loop. The asterisk in (o) represents the absence of co-stained cells. p–u Immunostaining of p-ERK (red) in control (p–r) and Gli1-CreER;Arid1bfl/fl (s–u) mouse incisors. q, r represent the high-magnification image of boxes in (p), t, u represent the high-magnification image of boxes in (s). White dotted lines outline the cervical loop. Yellow arrows indicate positive cells. n = 3. Scale bars, 100 μm.

Fig 5: Loss of Arid1b in GLI1+ lineage impairs mouse incisor growth and tissue homeotasis.a–d Immunostaining of ARID1B (red) in control (a, b) mouse incisor and co-immunostaining of ARID1B (red)/β-galactosidase (β-gal) (green) in Gli1-LacZ mouse incisor (c, d). b, d represent the high-magnification image of the box in (a, c). White dotted lines outline the cervical loop. White arrows indicate ARID1B+ cells. Yellow arrows indicate the ARID1B+/GLI1+ cells. e–j Notch movement assay in control (e–g) and Gli1-CreER;Arid1bfl/fl (h–j) mice. Yellow dotted lines show the notch position. Blue lines show the gingival margin. TMX, tamoxifen; wpt, week post-tamoxifen injection. Illustration below (e–k) created with BioRender.com released under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 Internatioanl license. k Quantification of the notch movement every other day from D2 to D10. Data are mean ± SEM, control n = 5, Gli1-CreER;Arid1bfl/fl n = 7, unpaired two-tailed Student’s t-test. D2, p = 0.0083; D4, p = 0.0103; D6, p = 0.0001; D8, p = 0.0003; D10, p = 0.0001. l–o MicroCT and HE staining of control (l, m) and Gli1-CreER;Arid1bfl/fl (n, o) mouse incisors at 3 months after tamoxifen induction. l, n MicroCT of control (l) and Gli1-CreER;Arid1bfl/fl (n) mouse incisors. White arrow indicates the narrowed dental pulp. m, o HE staining of control (m) and Gli1-CreER;Arid1bfl/fl (o) mouse incisors. Yellow arrows indicate the initiation of odontoblast polarization. Green arrow indicates the disordered alignment of odontoblasts. White two-way arrows indicate the dentin thickness. Black asterisk indicates stacked and distorted dentin. Boxes in m and o are shown at higher magnification on the right. p, q Dspp (red) in situ hybridization in control (p) and Gli1-CreER;Arid1bfl/fl (q) mouse incisors. White dotted lines outline the cervical loop. Yellow dotted lines show the cervical loop bending point to the odontoblast initiation distance. Unfilled arrows indicate the distance between the yellow dotted lines. n = 3. r Quantification of the dental pulp cavity. Data are mean ± SEM, n = 3, unpaired two-tailed Student’s t-test. p = 0.0113. s Quantification of the distance of Dspp+ cells to the cervical loop. Data are mean ± SEM, n = 3, unpaired two-tailed Student’s t-test. p < 0.0001. Source data are provided as a Source Data file. Scale bars: 2 mm (e–j, l, n); 100 μm (other images).