β-cell Tead1 deletion leads to diabetes.

(A) Relative gene expression assessed by RT-qPCR of total RNA from isolated islets of 4 month old wild type mice, normalized to housekeeping gene, Tbp. n=4 (B) Representative immunofluorescence staining of Tead1, Taz and Yap in mouse islets at different ages. Scale bar - 50µm. (C) Relative gene expression assessed by RT-qPCR of total RNA from isolated human islets, normalized to housekeeping gene, TBP. n=4 (D) RT-qPCR (normalized to control Tead1F/F) n=4-6; and (E) western blotting for Tead1 in isolated islets from Tead1F/F and Rip-β-Tead1-/-mice. (F) Representative images of immunofluorescence staining of Tead1F/F and Rip-β-Tead1-/- pancreas. Scale bar - 50µm. (G) Plasma glucose level with overnight fasting and after 1 hour refeeding in Tead1F/F and Rip-β-Tead1-/-mice. n=4-6. w-weeks; M-months; F-fasting; RF- refeeding. (H) Acute insulin secretion after glucose stimulation in 10-week-old mice. n=5. (I) Plasma glucose and (J) insulin during glucose tolerance testing (GTT) in overnight fasted 8 weeks old male mice. n=5-6. (K) Plasma glucose and (L) insulin during GTT in overnight fasted 9 weeks old male mice. n=3-4. (M) Western blotting of isolated islets from Tead1F/F and Tead1F/FMip-CreERT+ mice treated with oil or Tamoxifen (TM). (N) Nuclear Tead1 staining is absent from most insulin expressing β-cell in TM-treated Mip-β-Tead1-/- islets. Scale bar – 50µm. (O) Plasma glucose level and (P) insulin levels during the GTT in overnight fasted 5 month old Tead1F/F and Tead1F/FMip-CreERT+ (Mip-β-Tead1-/-) mice treated with oil or TM. n=4. TM, Tamoxifen. All values are mean ± SEM, * - p≤ 0.05.

Tead1 is required for mature β-cell function.

(A) GSIS from isolated islets (ten islets each from 5-6 individual mice of each genotype) on exposure to increasing glucose concentrations. Insulin secretion index is expressed as a fold change over basal secretion in 2.8mM glucose. Insulin content in (B) total pancreas, n=4; in isolated islets expressed as insulin content (C) per equally sized islet or (D) per μg DNA, n=5. (E) Volcano plot from RNA-seq data from differentially expressed genes from isolated islets of 1 year old Rip-β-Tead1-/- compared to Tead1F/F. Some of the genes critical in mature β-cell function are highlighted. (F) GSEA analysis of the differentially expressed downregulated genes in isolated islets of Rip-β-Tead1-/- compared to Tead1F/F as compared to the published gene set (GSE47174) that was upregulated in adult β-cells when compared to neonatal β-cells. Details in the accompanying text. NES – normalized enrichment score. FDR Q value is also shown. (G) Enriched pathways in transcriptome analysis of Rip-β-Tead1-/- compared to Tead1F/F isolated islets. x-axis shows the Normalized enrichment score. Note that in this representation the red bars indicate the enrichment of Rip-β-Tead1-/- over control Tead1F/F indicating these pathways are suppressed in islets from Rip-β-Tead1-/- mice, while the blue bars indicate that the same pathways are upregulated in adult over neonatal β-cells. (H) Expression, by RT-qPCR, of genes critical to mature β-cell function in Tead1F/F and Rip-β-Tead1-/- islets, normalized to housekeeping gene, TopI. n=4. (I) Representative immunofluorescent images of Pdx1, MafA, Ucn3 and Glut2 staining in Tead1F/F and Rip-β-Tead1-/- islets. Scale bar - 50 μm. (J-M) Western blot and quantification of (J and K) Pdx1 protein and (L and M) MafA from isolated islets from Rip-β-Tead1-/- and Tead1F/F mice. All values are mean ± SEM, * - p≤ 0.05.

Tead1 is direct transcriptional regulator of genes for mature β-cell function.

(A) Tead1 occupancy in the active promoters of displayed genes that have open chromatin in mouse islets by ATAC-seq and ChIP-seq. Red boxes indicate overlapping regions of Tead1 occupancy (Tead1 Chip-seq) and open chromatin (ATAC-seq). Input control is also shown. (B) ChIP in Ins-2 cells with Tead1 or control IgG antibody and qPCR with primers flanking putative Tead1 response elements. The y-axis represents the ratio of pulldown DNA to input DNA. n=3. (C) Tead1 response elements (Tead1 RE) in promoter region of mouse Pdx1 is shown schematically and (D) ChIP in Myc-Tead1 over-expressed Ins-2 cells with Myc or control IgG antibody and qPCR with putative Tead1 response element on mouse Pdx1 promoter. (E) Promoter luciferase assay with native Pdx1 promoter-Tead1 response element (RE:Luc) reporter in Ins-2 cells. y-axis represents the relative luminescence units (RLU). n=3. (F-G) Western blotting for Pdx1 protein in Tead1 knock down Ins2 cell line with quantification relative to α-tubulin as housekeeping control (G). (H-I) Western blotting and quantification for PDX1 protein in TEAD1 overexpressing (Tead1 O/E) in mouse Ins2 cell line, in (J and K) human EndoC-β-h2 cell line and in (L) human islets. Control cells are transfected with empty vector. (M) Correlation analysis of PDX1 protein and TEAD1 in human islets (n=8). All values are mean ± the standard errors of the means, * - p≤ 0.05.

Tead1 deletion enhances β-cell proliferation.

(A) β-cell number in 1000 µm2 in Rip-β-Tead1-/- mice and Tead1F/F mouse islet. N=3. (B) Single β-cell area in Rip-β-Tead1-/- mice and Tead1F/F mouse pancreas at 3 months age. n=3. (C) β-cell size distribution in Tead1F/F mice (blue) and Rip-β-Tead1-/- mice (red) at 3 month age. All values are means ± SEM. n=3. (D) Total pancreatic β-cell area from Rip-β-Tead1-/- mice and Tead1F/F mice at 3 months age. n=3. (E-H) Quantification of β-cell proliferation in Tead1F/F and Rip-β-Tead1-/- mice at 3 months. n=3. BrdU was injected (G and H) 2 hrs prior to sacrifice, or (I and J) administered continuously by water for 6 days prior to sacrifice. Representative microscopy images of Rip-β-Tead1-/- mice and Tead1F/F mouse pancreas at 3 months age and quantification of (E and F) Ki67 expressing and (G and I) BrdU positive β-cells. n=4. Scale bar – 50µm. (K) The fraction of BrdU-positive β-cells to Ki67-positive β-cells in the Rip-β-Tead1-/- and controls. n=3-4 (L) Quantification of Brdu-positive β-cells in in 11 month old TM-treated Mip-β-Tead1-/- mice and controls. n=4. (M) Brdu-positive cells in Tead-negative and Tead1-positive β-cell subpopulations in 11 month old TM-treated Mip-β-Tead1-/- mice. n=4. All values are mean ±the standard errors of the means, * - p ≤ 0.05. N.S.-not significant. (N) Mosaic deletion of Tead1 in TM-treated Mip-β-Tead1-/- β-cells leads to intra-islet mosaicism of Tead1 expression in β-cells. Representative confocal microscopy images of immunofluorescent staining of pancreas sections from Mip-β-Tead1-/- mice indicating Tead1-neg (red arrow) and Tead1-pos (white arrow) β-cells in 3 different mice.

Tead1 is direct transcriptional regulator of genes for β-cell proliferation.

(A) Transcriptome analysis of isolated Rip-β-Tead1-/- and control Tead1F/F islets from 12 weeks old male mice with significantly enriched pathways by Gene Ontology pathway analysis. All proliferation related pathways are highlighted in red. Details in the accompanying text. NES – normalized enrichment score. FDR Q value is also shown. (B) Relative expression of cell cycle related genes in Rip-β-Tead1-/- and Tead1F/F islets, normalized to housekeeping gene, TopI. n=4. (C) Western blot of p16INK4a protein and (D) quantification in isolated islets from Rip-β-Tead1-/- and Tead1F/F mice. n=4-5. (E and F) Western blotting for p16INK4a in (E) shRNA Tead1 knockdown and scrambled Ins-2 cells and quantification compared to α-tubulin as housekeeping control (F). This is from the same lysate as shown in Figure 3F. (G) Tead1 overexpression and control in Ins-2 cells. (H) Tead1 response elements (Tead1 RE) in promoter region of mouse Cdkn2a. The Cdkn2a locus encoding p16INK4a and p19ARF is shown schematically. (I) ChIP of Myc-Tead1 over-expressing Ins-2 cells with Myc or control IgG antibody and qPCR with primers flanking putative Tead1 response elements. The y-axis represents the ratio of pulldown DNA to input DNA. n=3. (J) Tead1 occupancy in the active promoter/intronic region of Cdkn2a that have open chromatin in mouse islets by ATAC-seq and ChIP-seq. Red box indicate overlapping regions of Tead1 occupancy (Tead1 Chip-seq) and open chromatin (ATAC-seq). Input control is also shown. (K) Promoter luciferase assay with native Cdkn2a promoter-Tead1 response element (RE:Luc) reporter in Ins-2 cells. y-axis represents the relative luminescence units (RLU). n=3. All values are mean ±the standard errors of the means, * - p≤ 0.05.

Tead1 regulates proliferative quiescence by its direct transcriptional regulation of p16INK4a expression.

(A) western blotting for Tead1, Ezh2 and p16 and (B) quantification of p16 protein in isolated islets from Tead1F/FEzh2F/F and Rip-β-Tead1-/-Ezh2-/- mice. n=4-6. (C) Representative images of immunofluorescence staining for BrdU and insulin in Tead1F/FEzh2F/F and Rip-β-Tead1-/-Ezh2-/- pancreas. BrdU was injected 2 hrs prior to sacrifice. (D) Quantification of β-cell proliferation in Tead1F/FEzh2F/F and Rip-β-Tead1-/-Ezh2-/- mice at 10-11 weeks. n=3. (E) Blood glucose during GTT in overnight fasted 9-10 weeks old male mice. n=4-6.

Schematic model of reciprocal regulation of β-cell proliferation and mature function by Tead1.

Generation of Tead1 floxed mouse strain and β-cell specific Tead1 knock out mice.

(A) Generation of Tead1 conditional knockout allele, and β-cell specific knockout mouse models. (B) Genomic PCR showing the recombined fragment of the Tead1 allele in islet and heart from Tead1F/F and Rip-β-Tead1-/-mice. (C) Tead1 flox allele and its deletion in Rip-β-Tead1-/-, verified by long-range PCR (right) in Tead1F/F and Rip-β-Tead1-/-islet. (D) Body weight of Tead1F/F and Rip-β-Tead1-/-mice. n=4-6. (E) Plasma glucose represented as a percent drop from baseline during insulin tolerance test in 12 weeks old male mice. n=5. All values are mean ± SEM. (F) Food intake measured over a 3-day period. n=4. Energy expenditure and VO2 (G and H), VCO2 (I), and RER (J) were assessed as described in Methods in supplementary material. n=4. All values are mean ±SEM.

Cell-autonomous regulation of mature β-cell function by Tead1.

(A,B) Tead1-depleted β-cells have impaired GSIS in vitro. (A) Western blotting for Tead1 in shRNA Tead1 knockdown and scrambled Ins-1 (832/13) cells. (B) GSIS in response to increasing glucose concentration in Tead1-knockdown and scrambled control 832/13 cells. n= 4. All values are mean ± SEM, * - p≤ 0.05. (C) Transcriptome analysis heat map showing up and downregulated genes in isolated islets from 12 week old Rip-β-Tead1-/- compared to Tead1F/F mice. (D) GSEA analysis of the differentially expressed upregulated genes in isolated islets of Rip-β-Tead1-/- compared to Tead1F/F as compared to the published gene set (GSE47174) that was upregulated in adult β-cells when compared to neonatal β-cells. Details in the accompanying text. FDR Q value did not show any significance.

Tead1 regulates mature β-cell function-related genes.

(A) ChIP of Myc-Tead1 over-expressing Ins-2 cells with Myc or control IgG antibody and qPCR with primers flanking putative Tead1 response elements in promoters of displayed genes. The y-axis represents the ratio of pulldown DNA to input DNA. n=3.

Tead1 regulates β-cell proliferative quiescence.

(A) Genomic DNA content in similar sized islets from Rip-β-Tead1-/- and floxed control mice. n=5. (B,C, and D) Quantification of (B) β-cell size, (C) β-cell area and (D) BrdU positive β-cells in 5-6 month old Ins-β-Tead1-/- and floxed control mice n=3-4. (E, F, and G) Quantification of (E) β-cell size, (F) β-cell area and (G) average islet size in Mip-β-Tead1-/- and control mice n=3-4. (H and I) Apoptosis is not altered in Tead1-deficient β-cells. Representative image of (H) TUNEL staining and (I) immunohistochemistry for Caspase-3 and cleaved Caspase-3 in Rip-β-Tead1-/- and control islet. DNase I treatment was used as a positive control for TUNEL staining. Scale bar – 50μm.