Tead family of transcription factors regulate proliferative pathways and determine organ size in many tissues during development ^11,25. However, their role in β-cell development and function are unknown. Examination of islets at different developmental stages revealed that Tead1 was preferentially expressed in adult mouse islets, and highly enriched as compared to other members of the Tead family that are present at 10-fold lower level (Tead2) or not detected (Tead 3 &4) (Figure. 1A). Of the two canonical Tead1 co-activators Yap and Taz, only Taz was robustly expressed (Figure. 1A), as were many upstream signaling components of Hippo pathway (Mst1&2, Lats1&2 and others), suggesting the Hippo-Tead pathway is functional in islets. Expression dynamics of Tead1 by immunostaining during islet development revealed it was most abundant at e15.5 in all pancreatic tissues, but its high expression was restricted to islets around weaning and persisted at a high level in adult β-cells (Figure. 1B). While Taz was robustly expressed throughout, from e15.5 into adult stage, Yap was absent in the islets at all stages examined (Figure. 1B), a finding that is consistent with previous studies ¹⁴. Human islets also expressed robust levels of TEAD1 and Taz but not YAP (Figure 1C). In contrast to mouse islets, human islets express higher levels of Tead4. Based on the robust, yet restricted expression of Tead1 in mature β-cells, we selectively deleted Tead1 in a developmental stage-specific manner by crossing mice with Tead1 floxed (Tead1^F/F) alleles carrying a floxed exon 3 that encodes the DNA binding TEA domain (Figure S1A-C) ²⁶ with three distinct Cre lines, namely Rip-Cre, Ins1-Cre (Thorens), and Mip-CreERT.

Early constitutive deletion at e15.5 (Rip-β-Tead1^-/-), using Rip-Cre deletor mice, led to a loss of β-cell Tead1 expression, as indicated by >80% reduction in transcript (Figure 1D) and near absence Tead1 protein (Figure 1E) in whole islets. Loss of Tead1 immunostaining was specific to β-cells (Figure 1F), but not other hormone-producing islet cells (data not shown). These Rip-β-Tead1^-/- mice with embryonic deletion of Tead1 developed progressive hyperglycemia (fasting and fed) within 2 weeks of weaning at 5 weeks of age. At 2 months of age, they became diabetic with fed plasma glucose >500mg/dl (Figure 1G). Indicative of a severe defect in β-cell secretory function, the Rip-β-Tead1^-/- mice lacked the first phase insulin secretion (Figure 1H), a feature that is lost very early in type 2 diabetes ^27,28, and displayed marked glucose intolerance during an intraperitoneal glucose tolerance testing (GTT) due to a blunted GSIS (Figure 1I-J). These results were not secondary to changes in body weight (Figure S1D) or insulin sensitivity (Figure S1E). Overall energy balance was similar to controls as indicated by food intake and whole body energy expenditure, VC02, V02, and RER (Figure S1F-J). To exclude potential confounding effect of hypothalamic expression or potential secretory dysfunction due to the Rip-Cre transgene ^29,30, we applied another Ins1-Cre deletor line ³¹ for specific genetic targeting of Tead1 in β-cells by generating the Ins1-β-Tead1^-/-mice. These Ins1-β-Tead1^-/- phenocopied the Rip-β-Tead1^-/- indicated by marked glucose intolerance with a significant blunting of GSIS in vivo during GTT (Figure 1K-L).

While the findings in Ins1-β-Tead1^-/- and Rip-β-Tead1^-/-mice demonstrated the requirement of Tead1 function in β-cells in maintaining insulin secretion and glucose homeostasis in vivo, these models that involve embryonic deletion starting at e15.5 could also engender chronic compensatory changes. Hence, we tested whether Tead1 modulates adult β-cell function using Mip-CreERT mice ²⁹ to achieve tamoxifen (TM)-inducible deletion (Mip-β-Tead1^-/- mice). Tamoxifen administration (100mg/kg every other day for 5 doses by gavage) in 5-week-old Mip-β-Tead1^F/F mice largely abolished-Tead1 protein expression in whole islet lysates when examined after 6 weeks, (Figure 1M), along with an absence of Tead1 nuclear immunostaining in ∼85% of β-cells (Figure 1N). Strikingly, this restricted inactivation of Tead1 only in mature β-cells led to a marked glucose intolerance due to a blunting of in vivo GSIS with a loss of early phase insulin secretion (Figure 1O-P), demonstrating severe β-cell secretory deficits within 2 months post-TM treatment. Taken together, the three loss-of-function models of Tead1, during β-cell development (Rip-β-Tead1^-/- and Ins1-β-Tead1^-/-) or after maturation (Mip-β-Tead1^-/-) demonstrate the essential role of Tead1 in maintaining β-cell secretory capacity and glucose homeostasis.

Using isolated Rip-β-Tead1^-/- islets, we next determined the specific functional deficits induced by a loss-of-Tead1 function. Despite normal basal insulin secretion these islets display significantly impaired GSIS with a markedly lower insulin secretion index (Figure 2A), as compared to either Floxed or Rip-Cre controls. Furthermore, in Ins-1 (832/13) cells, Tead1 knockdown yielded a similar blunting of GSIS response (Figure S2A-B), suggesting the cell-autonomous role of Tead1 for GSIS in β-cells. Consistent with the GSIS deficits, there were ∼50-80% reductions of insulin content in Rip-β-Tead1^-/- mice, whether measured in the whole pancreas (Figure 2B), normalized to islet size (Figure 2C) or by cell number (reflected by DNA content) as compared to controls (Figure 2D).

To determine the regulatory pathways mediating Tead1 function in islets, we performed global transcriptome analysis in Rip-β-Tead1^-/- islets that revealed 259 (down) and 296 (up) differentially regulated genes as compared to age-matched floxed controls (Figure S2C and Supplementary table 3). In agreement with the profound defects in GSIS, among the differentially regulated genes by RNA-seq analysis distinct components of the insulin secretory pathway including the glucose transported Glut2 (Slc2a2) and glucokinase (Gck) were down-regulated in Tead1-deficient islets. In addition, key transcription factors known to specify and maintain adult β-cell function were suppressed, such as Pdx1, MafA. (Figure 2E). Further analysis using an unbiased Gene Set Enrichment Analysis (GSEA) revealed a global down-regulation in adult β-Tead1^-/- islets of genes significantly enriched in mature adult as compared to immature neonatal islets (GSE47174) ³² (Figure 2F). Interestingly, many disallowed genes in normal adult β-cells that comprise the neonatal islet transcriptome signature ^33-36 and are highly expressed in immature neonatal islets were not altered in Tead-1 deficient islets (Figure S2D). These findings suggest a loss of mature β-cell characteristics, rather than de-differentiation ⁵ of Tead1-null β-cells. Further support of this notion was found in the transcriptomic signatures of key processes that are critical for acquisition and maintenance of mature secretory function in β-cells, including OXPHOS, substrate metabolism, protein secretion and export, vesicular transport and UPR were consistently suppressed with loss of Tead1 (GSEA, q<0.25) (Figure 2G). We further validated that in Rip-β-Tead1^-/- islets, transcript (Figure 2H) of Ins1, Glut2, Pdx1, NeuroD, MafA and Nkx6.1 were all significantly decreased along with Ucn3, a hallmark of mature β-cells ^37,38. The protein levels of many of these genes were also significantly reduced by immunostaining (Figure 2I) or western blotting (Figure 2J & 2L) with respective quantifications (Figure 2K & 2M). These data collectively demonstrated Tead1 transcriptional control of a regulatory network critical for maintaining functional mature β-cell identity.

We interrogated whether Tead1 exerts direct transcriptional regulation of this gene network driving β-cell function and identity. Analysis of Tead1 occupancy by ChIP-seq in adult mouse islets on regulatory regions of transcription factors that are critical for β-cell function, including Pdx1, MafA, Nkx6.1, revealed its enrichment on these genes that were downregulated by loss of Tead1 (Figure 3A). ATAC-seq further confirmed that the identified promoter regions with Tead1 occupancy in islets occurred in open chromatin regions that are transcriptionally active (Figure 3A). ChIP-qPCR using either endogenous Tead1 (Figure 3B) or tagged Myc-Tead1 (Figure S3A) in mouse insulinoma cell line further validated Tead1 binding to cis-promoter elements (containing Tead consensus binding sequences of GGAATG) of Pdx1, MafA, Nkx6.1, Ucn3.

Since Pdx1 is a master regulator of mature β-cell function, we undertook a detailed analysis of Tead1 regulation of Pdx1 transcription. Interestingly, we found multiple Tead1 responsive elements in the Pdx1 promoter region as represented in Figure 3C, including the Tead1 consensus sequence present in Area IV region known to be critical for β-cell specific Pdx1 expression and postnatal mature β-cell function ³⁹. In ChIP pull down with Myc-Tead1 in Ins2 cells, this site exhibited the highest enrichment (Figure 3D). In a promoter luciferase assay, expression of Tead1 together with its co-activator, Taz, activated transcription of the native Pdx1-promoter containing Tead1 response elements, demonstrating Tead1 functional regulation of Pdx1 transcription (Figure 3E). Furthermore, Tead1 knockdown decreased Pdx1 protein level in Ins2 cells (Figure 3F-G); whereas its overexpression by adenoviral vector in Ins2 cells increased Pdx1 expression (Figure 3H-I). Overexpression of TEAD1 in additional human β-cell models including human β-cell line (EndoC-βh2) (Figure 3J-K) or human cadaveric islets (Figure 3L) augmented PDX1 expression. It is noteworthy that TEAD1 and PDX1 protein levels detected by western blotting in individual human islet samples exhibited significant linear correlation (Figure 3M). Thus, TEAD1 transcriptional control of PDX1 is conserved in human β-cells. Taken together, these data provide direct support of Tead1 as a direct transcriptional regulator that orchestrates a transcription factor network regulatory network critical to acquire and maintain mature β-cell function.

Tead1 drives cell proliferation in various tissue development processes ^11-13. We postulated that Tead1 deficiency during islet development may limit β-cell proliferation to reduce β-cell mass. Surprisingly, Rip-β-Tead1^-/- islets from 12 week old mice contained increased number of β-cells (Figure 4A), which was also reflected by increased DNA content in similar-sized islets (Figure S4A). However, these β-cells were significantly smaller (Figure 4B-C), which offset the increase in numbers, so β-cell mass in Rip-β-Tead1^-/- mice remained normal compared to that of the floxed controls (Figure 4D). Similar findings were observed when these were repeated in 5-6 month old Ins1-β-Tead1^-/- islets (Figure S4B-D). Tead1 deletion only in adult β-cells in the inducible line of 9 month old adult Mip-β-Tead1^-/- mice yielded similar findings with a significant decrease in single β-cell area (Figure S4E), but resulted in no change in total β-cell area (Figure S4F) or islet size (Figure S4G). Further analysis indicated that increased β-cell number in Rip-β-Tead1^-/- islets was not due to apoptosis (Figure S4H-I), but rather because of higher proliferation rate, as reflected by a significantly elevated Ki67 positive β-cells (2% vs. 0.62% in 12 week old Rip-β-Tead1^-/- and Tead1^F/F controls, respectively; p<0.01) (Figure 4E-F).

To test if Tead1 regulates β-cell proliferation at distinct developmental stages, we analyzed in vivo BrdU incorporation into β-cells in neonatal (postnatal day 12), post-weaning (4 weeks) and adult (12 weeks) mice. Notably, while there was no significant difference in proliferation at postnatal day 12 between Rip-β-Tead1^-/- and Tead1^F/F littermate controls, higher rate of BrdU incorporation was observed at 4 (5.31% vs. 4.08%; p<0.03) and 12 weeks of age (0.5% vs. 0.2%; p<0.05) respectively (Figure 4G-H), with similar increase (0.03% vs. 0.1%; p<0.05)seen in 5-6 month old Ins1-β-Tead1^-/- islets compared to floxed controls (Figure S4D). This data indicate that Tead1 did not contribute physiologic neonatal β-cell proliferative expansion when the proliferative drive is high, but rather was required for attaining proliferative quiescence as β-cells mature in adult mice.

β-cells have a short quiescence phase of ∼7 days after mitosis before they can re-enter cell cycle ⁴⁰. Hence, we assessed if Tead1 could regulate total pool of actively proliferating β-cells by labelling all cycling β-cells over a 6 day period with BrdU via drinking water and found that active proliferation was ∼4 fold higher in Tead1-deficient β-cells than that of controls (17.1% vs. 4.46%) (Figure 4I-J). However, the fraction of proliferating β-cells in S-phase (BrdU-positive) to actively cycling Ki67-positive cells did not differ between the Rip-β-Tead1^-/- and controls (Figure 4K) indicating that loss of Tead1 did not affect cell cycle kinetics, once the cells entered cell cycle. Together, these data supports the notion that higher proliferation in Tead1-deficient β-cells likely occurred as a result of more cells entering cell cycle (G1) from the G0 quiescent state. We then asked if Tead1 affects age-associated proliferative quiescence when most β-cells become resistant to proliferative signals ⁴¹. Tead1 deletion in β-cells in 9 months old TM-treated Mip-β-Tead1^-/- mice showed a 2-fold increase in proliferation (Figure 4L), similar as young adult mice. Thus, Tead1 was required to maintain quiescence in adult β-cells, while its inactivation releases cells from age-associated quiescence into active cell cycle.

The augmented β-cell proliferation in 4 week old Rip-β-Tead1^-/- mice (Figure 4G) preceded the onset of hyperglycemia at 5 weeks of age (Figure 1F), suggesting that the observed increase in proliferation is not secondary to hyperglycemia-induced compensation. However, to conclusively exclude non-cell autonomous effects, such as hyperglycemia, as a potential cause of proliferation, we leveraged the mosaic β-cell Tead1 deletion often seen with Mip-CreERT ^29,42. Using this model, we pulse labelled cycling β-cells for 2 hours using BrdU and analyzed islets in which ∼50% of β-cells had deletion of Tead1. We found that within the same pool of islets in Mip-β-Tead1^-/- mice with Tead1-positive and Tead1-negative β-cells, the proliferative rate was ∼ 5 fold higher in Tead1-negative than that of the Tead1-positive β-cells (0.58% vs. 0.11%; p<0.05) (Figure 4M-N), indicative of cell-autonomous effect of Tead1 on proliferation.

Corroborating the unexpected increased proliferation of Tead1-deficient β-cells, global transcriptome analysis of Tead1-deficient islets revealed significant upregulation of pathways involved in cell cycle control, including many G1-S checkpoint regulators (Figure 5A). Expression of Cyclin A2, critical for G1-S phase transition in β-cells ⁴³ was significantly elevated in Rip-β-Tead1^-/- islets (Figure 5B). Though absence of Tead1 reduced Cyclin D1 transcript, its protein level was not altered (data not shown). In contrast, there were marked reductions of p16^INK4a and p19^ARF, key cell cycle inhibitors critical for limiting β-cell proliferation in adult mice ^44-46 (Figure 5B). p16^INK4a protein was nearly absent in Tead1-deficient islets (Figure 5C-D) as well as in Tead1-depleted Ins2 cells (Figure 5E-F, and Figure 4F), while Tead1 overexpression robustly increased p16^INK4a protein levels (Figure 5G). Examination of the Cdkn2a locus that encodes both p16^INK4a and p16^ARF demonstrated Tead1 occupancy in multiple Tead1 consensus motifs (Figure 5H). ChIP-qPCR in Ins2 cells demonstrated Tead1 binding to these motifs within the promoter of p16^INK4a in the Cdkn2a locus (Figure 5I). This was consistent with Chip-seq and ATAC-seq data from adult islets demonstrating Tead1 occupancy within Tead1 consensus motif containing regions of open chromatin in Cdnk2a locus (Figure 5J). Furthermore, Tead1 binding to these sites are functional as it can activate the native Cdkn2a (p16INK4a) promoter when co-expressed with Taz (Figure 5K).

p16 represents a key cell cycle entry barrier to transition from a quiescent state, and Ezh2 has an established role as its critical repressor in β-cells ^41,45,46. To explore whether Tead1 transcriptional regulation of p16^INK4a determines the balance between proliferation and quiescence, and if β-cell proliferation and its mature function were inter-dependent we tested if re-expressing p16 in Tead1-null β-cells in vivo through genetic ablation of its repressor Ezh2 could restore β-cell quiescence and insulin-secretory function. We inactivated Ezh2 in β-cells of Rip-β-Tead1^-/- mice (DKO) by crossing Ezh2 flox mice with Rip-β-Tead1^-/- mice. As expected of Ezh2 regulation of p16, the loss of Ezh2 in Tead1-null β-cells restored the level of p16 to a comparable level in wild type controls (Figure 6A-B). Notably, 6 month old Rip-β-Tead1^-/- β-cells without Ezh2 showed much lower proliferation rates than that seen in Rip-β-Tead1^-/- mice (Figure 4G-H) and at a level comparable to the control flox mice, (Figure 6C-D). Therefore, restoring p16 through the modulation of Ezh2 was sufficient to achieve/rescue proliferative quiescence in Tead1-null β-cells. Interestingly, the restoration of proliferative rates to normal was not sufficient in restoring glucose tolerance in these DKO mice, as these mice exhibited persistence glucose intolerance as compared to the floxed controls though there was some improvement as compared to littermate Rip-β-Tead1^-/- mice (Figure 6E). Thus, Tead1 regulation of proliferative quiescence was mediated by its direct transcriptional regulation of p16^INK4a expression.

All animal experiments were approved by the Institutional Animal Care and Use Committee of Baylor College of Medicine and the University of Pittsburgh. Tead1 floxed ES cells were obtained from KOMP (Clone Name: EPDO207_2_H11), wherein the DNA binding domain coding Exon 3 is floxed. Targeted ES cells were injected into blastocysts to generate chimeric mice and those that demonstrated germline transmission were crossed with mice expressing FLP recombinase to remove the neomycin resistance cassette to generate Tead1 loxP alleles [floxed exon: ENSMUSE00000203030]. The homologous recombination was verified by long range PCR using primer set #1 (forward: GCAGTGCTCTCAGGAGTGCTGAGTGCGAC; reverse: AGAAGCCACAGTGCCCTGGAAGTGT) and primer set #2 (forward: GCCTTCTGAGTGCTGGCATTAAAGG; reverse: CACTGCAGCGCGCATAGCCTATGCTCCTTC) (Figure S1). Mice were genotyped using Tead1-flox-forward (GCCTTCTGAGTGCTGGCATTAAAGG) and reverse (AAGGCAGACTCCTTCATTGGAATGG) primers. Floxed Tead1 (Tead1^F/F) mice were crossed with Rip-Cre or Ins1-Cre (Thorens) transgenic mice to generate β-cell-specific Tead1 deletion during embryonic development. For adult β-cell inducible deletion, Tead1 ^F/F mice were crossed with Mip-CreERT transgenic mouse (Ins1 promoter driven Cre-ER) to generate Tead1 ^F/F Mip-CreERT⁺ mice that when treated with Tamoxifen (TM) 100 mg/ kg every other day by gavage for 5 doses induced β-cell-specific Tead1 deletion (Mip-β-Tead1^-/-). Ezh2 flox mice were obtained from Jackson labs (Stock#02616) and crossed into the Tead1^F/F background to obtain double floxed mice and crossed with Rip-Cre mice to obtain double knockout mice. Mice were maintained under standard 12-h light-dark cycles with ad lib access to food and water, unless specified otherwise.

Glucose tolerance tests (GTTs) and acute insulin secretion tests were performed in 16 hour overnight fasted mice administrated 1.5 g/kg or 3 g/kg D-glucose i.p. respectively. Insulin tolerance test (ITT) was performed in 16 hour overnight fasted mice administrated 0.75 U/kg insulin i.p.

Metabolic cage measurements were performed as described previously ⁷². Oxygen consumption (VO₂), energy expenditure, carbon dioxide production (VCO2), and the respiratory exchange ratio (RER) were measured with the comprehensive laboratory animal monitoring system (Columbus Instruments) in individual cages without bedding. Mice were acclimated to the metabolic cages for 4 days prior to the start of data collection. Data was collected for 72 h. All data was analyzed and averaged for the dark and light cycles separately.

Islets were isolated as previously described ^72,73. Briefly, 1mg/ml collagenase P (Roche) dissolved in HBSS was injected into the common bile duct of anesthetized mice and the pancreas was removed, minced and incubated in collagenase solution at 37°C for 17 min. After three washes with 10% FBS, the islets were pelleted at 200 g and purified on Histopaque (Sigma) gradient. Islets from each mouse were individually purified by handpick for all experiments. The islets were incubated in 11.1 mM RPMI-1640 supplemented with 10% FBS and 1% penicillin/streptomycin, L-Glutamine overnight. GSIS studies were performed as described previously ⁷³. Ten similar sized islets from individual mice were used, with at least three mice from each genotype, under each experimental condition. Incubation times for ex vivo insulin secretion studies were for 30 min for each condition and the results were normalized to the DNA content or to the total insulin content of the islets measured after acid ethanol extraction at the end of the experiment. All insulin assays were performed using mouse insulin ELISA kit (Mercodia). Insulin content was assayed after acid-ethanol extraction from whole pancreas and isolated islets and normalized to DNA and protein measured from the same extract. Similar protocol was used for the GSIS experiments in 832/13 cells.

Quantitative PCR (qPCR) from cDNA made from DNase-digested RNA was performed with gene-specific primers with SYBR green mix and normalized to the expression of housekeeping genes (Tbp, TopI, and Actb). The sequences of the primers used are available in Supplementary table 2.

Islets were isolated from each mouse individually as described above and 150–200 islets were snap frozen immediately. Western blotting was performed, as described before ⁷². Briefly, total protein was extracted from 150-200 islets mice using a protein lysis buffer with a cocktail of protease inhibitors and separated by a standard SDS-PAGE and transferred to a nitrocellulose membrane. Tead1 antibody (rabbit polyclonal, 1:10,000; Abcam), p16 antibody (mouse monoclonal, 1:1,000; Abcam), Pdx1 (rabbit polyclonal, 1:10,000; BCBC), Mafa antibody (rabbit polyclonal, 1:1,000; Bethyl lab Inc.), Ezh2 antibody (rabbit polyclonal, 1:1,000; Cell Signaling), α-tubulin (mouse monoclonal, 1:10,000; Cell Signaling), Hsp90 (mouse monoclonal, 1:1,000; Santa Cruz) and β-actin antibody (rabbit monoclonal-horseradish peroxidase conjugate, 1:10,000; Cell Signaling), as a housekeeping control, were visualized by enhanced chemiluminescence (Pierce).

Mouse pancreas were harvested, embedded in paraffin and sectioned to 5 µm thickness. Immunostaining was performed as previously described ⁷². Primary antibodies used were Glut2 (rabbit polyclonal, 1:200; Millipore), Insulin (guinea pig ployclonal, 1:500; Abcam), Glucagon (rabbit polyclonal, 1:400; Dako), Tead1 (rabbit polyclonal, 1:200; Abcam), Taz (rabbit polyclonal, 1:200; Santa Curz), Yap (rabbit poly clonal 1:200; Cell signaling), Pdx1 (guinea pig polyclonal, 1:500; Abcam), Mafa (rabbit polyclonal, 1:100; Bethyl lab), Ucn3 (rabbit polyclonal, 1:2,000; MGI), Ki67 (rabbit poly clonalm 1:50; Abcam), Brdu (rat polyclonal, 1:1,000; Abcam), Caspase-3 (rabbit polyclonal, 1:200; Cell signaling), and Cleaved Caspase-3 (rabbit monoclonal, 1:200; Cell signaling). To measure total pancreatic islet area, pancreas paraffin blocks from mice were cut 5 µm thickness sections spaced 150 µm apart between each slide. 5 pancreatic sections per mouse were stained and measured islets area using ImageJ software. For assessing number of Brdu positive and Ki67 expression cells in β-cell, 3,000 to 5,000 cells were counted from mouse pancreatic sections in 3 to 4 mice from each group.

832/13 (Ins-1, courtesy of Dr. Christopher Newgard) rat β-cell line was cultured as described previously ⁷². Ins-2 murine β-cell line (Courtesy of Akio Koizumi) ⁷⁴ was cultured in high glucose DMEM (4.5 g/L) containing 15% fetal bovine serum, 150 μM beta-2-mercaptoethanol, and Penicillin (100 units/ml)/Streptomycin (100 μg/ml). EndoC-β H2 (purchased from EndoCells and courtesy of Drs. Ravassard and Scharfmann) ⁷⁵ were cultured according to the manufacturer’s instruction. EndoC-βH2 cells were cultured in DMEM (1 g/L Glucose) containing 2% BSA fraction V, 50 μoM 2-mercaptoethanol, 10 mM nicotinamide, 5.5 μg/ml transferrin, 6.7 ng/ml sodium selenite, Penicillin (100 units/ml)/Streptomycin (100 μg/ml). Tissue culture flasks and dishes were coated with DMEM (4.5 g/L Glucose) containing fibronectin (2 μg/ml), 1 % ECM, and Penicillin (100 units/ml)/Streptomycin (100 μg/ml). De-identified human islets were obtained from the Integrated Islet Distribution Program (IIDP). Donor characteristics are shown in Supplementary table 3. Received human islets were selected by hand picking and cultured in CMRL1066 (Mediatech, Corning) supplemented with 10% FBS, 1% P/S and 2 mM L-glutamine. Intact islets were used for Verteporfin (VP) treatment experiments. For adenovirus (AdV) infection, human islets were disassociated with 0.05% Trypsin as previously described ⁷⁶ and subjected to infection at 20 MOI, the islets re-aggregated spontaneously and formed pseudoislets in 2 days. Protein samples were collected post 72hrs of AdV infection. All cell lines were initially authenticated by the donating investigator or vendor and were tested to be mycoplasma free.

Lentiviral shRNA against Tead1 and scrambled shRNA controls were obtained from Thermo Scientific and used to generate stable cell lines with mouse insulinoma (832/13 and Ins-2) cells. The following shRNAs were used: shRNA-Tead1#1: 5’-CTCGCCAATGTGTGAATATA-3’, shRNA-Tead1#3: 5’-GACTGAACCTGGTTAATTTA-3’. The knockdown efficiency was confirmed by RT-qPCR and western blotting.

Native Pdx1 and Cdkn2a promoter containing Tead1 binding element were amplified by PCR, cloned to pGL3 basic, mouse insulinoma cell line Ins-2 was used for the promoter activity assay.

Chromatin immunoprecipitation (ChIP) was performed using standard protocols in Ins-2 cells with Tead1 and Myc antibody or control IgG for pulldown. Immunoprecipitated DNA was then used as the template for qPCR with primers flanking conserved Tead1 binding element on promoters of target genes.

Biotin-labeled aRNA was generated from total RNA from isolated islets from 3 mice of each genotype at 12 weeks of age and was tested for integrity on the Agilent Bioanalyzer and quantified using the NanoDrop® ND-1000 spectrophotometer. 12 μg of labeled aRNA samples were fragmented and re-checked for concentration and size. Hybridization cocktails containing Affymetrix spike-in controls, 12 μg of each fragmented labeled aRNA and the Arcturus Turbo Blocking Reagent heated to 99 °C for 5 minutes, then incubated at 45 °C for 5 minutes. The hybridization cocktails were loaded onto Affymetrix GeneChip Mouse 430 2.0 arrays. The arrays were hybridized for 16 hours at 45 °C with rotation at 60 rpm in the Affymetrix GeneChip® Hybridization Oven 640.The arrays were washed and stained with a streptavidin, R-phycoerythrin conjugate stain using the Affymetrix GeneChip® Fluidics Station 450. Signal amplification was done using biotinylated antistreptavidin.The stained arrays were scanned on the Affymetrix GeneChip® Scanner 3000. The images were analyzed and quality control metrics recorded using Affymetrix Command Console software version 3.0.0. The data was normalized by quantiles and then significant genes analyzed using multiple sample corrected t-testing and 1.5 fold change and by gene ontology terms and MSigDb searches.

Total RNA samples with RNA integrity number (RIN) ≥8 were used for transcriptome sequencing. Total RNA (10ng) from a pooled sample from 1 year old mice (n=3) each for two independent samples for each group was used for amplified double-stranded cDNA that was sheared to 200-300bp and ligated to Illumina paired-end adaptors using the Illumina TruSeq DNA library preparation kit according to the manufacturer’s instructions (Illumina, San Diego, CA). PCR amplification was performed to obtain the final cDNA library using Illumina kits. Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA) analysis was used to verify fragment size after amplification, library size and concentration before clustering. A total of 10pM of the library was then used for paired-end sequencing on the HiSeq 2500 at the Sequencing core at Brigham Young University. All further analysis were performed using the CLC genomics workbench ver 12. Raw RNA sequencing reads were mapped to the mouse reference genome build 10 (UCSCmm10/GRCm38). Mapped reads were counted using the feature counts and differential expression between the samples analyzed using multiple hypothesis testing. Pathway analysis: Gene Set Enrichment (GSEA, version 2.2.3, http://software.broadinstitute.org/gsea/) was performed on normalized count per million (CPM) or on ranked gene lists. Ranked lists were created based on the expression levels of DEGs in Rip-Tead1 KO vs. Flox controls (q<0.05). Significance was assessed by analyzing signal-to-noise ratio and gene permutations based on 1,000 permutations. Molecular signature database (MSigDB) 3.0 curated gene sets for hallmark and canonical pathways were used for the analysis. Significant gene sets with enrichment score and a q value cutoff of 0.05 are presented.

Flash-frozen tissue was sent to Active Motif to perform the ATAC-seq assay. The tissue was manually disassociated, isolated nuclei were quantified using a hemocytometer, and 100,000 nuclei were tagmented as previously described (Buenrostro et al. 2013), with some modifications based on (Corces et al. 2017) using the enzyme and buffer provided in the Nextera Library Prep Kit (Illumina). Tagmented DNA was then purified using the MinElute PCR purification kit (Qiagen), amplified with 10 cycles of PCR, and purified using Agencourt AMPure SPRI beads (Beckman Coulter). Resulting material was quantified using the KAPA Library Quantification Kit for Illumina platforms (KAPA Biosystems), and sequenced with PE42 sequencing on the NextSeq 500 sequencer (Illumina). Analysis of ATAC-seq data was very similar to the analysis of ChIP-Seq data. Reads were aligned using the BWA algorithm (mem mode; default settings). Duplicate reads were removed, only reads mapping as matched pairs and only uniquely mapped reads (mapping quality >= 1) were used for further analysis. Alignments were extended in silico at their 3’-ends to a length of 200 bp and assigned to 32-nt bins along the genome. The resulting histograms (genomic “signal maps”) were stored in bigWig files. Peaks were identified using the MACS 2.1.0 algorithm at a cutoff of p-value 1e-7, without control file, and with the –nomodel option. Peaks that were on the ENCODE blacklist of known false ChIP-Seq peaks were removed.

Frozen tissue was sent to Active Motif Services (Carlsbad, CA) to be processed for ChIP-Seq. In brief, tissue was submersed in PBS + 1% formaldehyde, cut into small pieces and incubated at room temperature for 15 minutes. Fixation was stopped by the addition of 0.125 M glycine (final). The tissue pieces were then treated with a TissueTearer and finally spun down and washed 2x in PBS. Chromatin was isolated by the addition of lysis buffer, followed by disruption with a Dounce homogenizer. Lysates were sonicated and the DNA sheared to an average length of 300-500 bp. Genomic DNA (Input) was prepared by treating aliquots of chromatin with RNase, proteinase K and heat for de-crosslinking, followed by ethanol precipitation. Pellets were resuspended and the resulting DNA was quantified on a NanoDrop spectrophotometer. Extrapolation to the original chromatin volume allowed quantitation of the total chromatin yield. An aliquot of chromatin (10 ug for islets and 30 ug for hearts) was precleared with protein G agarose beads (Invitrogen). Genomic DNA regions of interest were isolated using 20 ul of antibody against Tead1. Complexes were washed, eluted from the beads with SDS buffer, and subjected to RNase and proteinase K treatment. Crosslinks were reversed by incubation overnight at 65 C, and ChIP DNA was purified by phenol-chloroform extraction and ethanol precipitation. Quantitative PCR (QPCR) reactions were carried out in triplicate on specific genomic regions using SYBR Green Supermix (Bio-Rad). The resulting signals were normalized for primer efficiency by carrying out QPCR for each primer pair using Input DNA. Illumina sequencing libraries were prepared from the ChIP and Input DNAs by the standard consecutive enzymatic steps of end-polishing, dA-addition, and adaptor ligation. Steps were performed on an automated system (Apollo 342, Wafergen Biosystems/Takara). After a final PCR amplification step, the resulting DNA libraries were quantified and sequenced on Illumina’s NextSeq 500 (75 nt reads, single end). Reads were aligned to the mouse genome (mm10) using the BWA algorithm (default settings). Duplicate reads were removed and only uniquely mapped reads (mapping quality >= 25) were used for further analysis. Alignments were extended in silico at their 3’-ends to a length of 200 bp, which is the average genomic fragment length in the size-selected library, and assigned to 32-nt bins along the genome. The resulting histograms (genomic “signal maps”) were stored in bigWig files. Peak locations were determined using the MACS algorithm (v2.1.0) with a cutoff of p-value = 1e-7. Peaks that were on the ENCODE blacklist of known false ChIP-Seq peaks were removed.

All statistical testing was performed either by two-tailed Student’s t-test, assuming unequal variance for two groups or ANOVA for multiple samples with p ≤ 0.05 considered significant between the groups.

Expression data have been deposited in GEO: 12 weeks (GSE139228) and 1 year (GSE139152).

Chip-seq and ATAC-seq data have been deposited in GEO under GEO accession GSE157513