@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Medicine, Faculty of"@en, "Medical Genetics, Department of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Kadhim, Alexandre Zacharie"@en ; dcterms:issued "2023-09-30T07:00:00Z"@en, "2020"@en ; vivo:relatedDegree "Doctor of Philosophy - PhD"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """The Mediator complex, a coregulator required for RNA pol II activity, interacts with specific transcription factors through its distinct subunits. These interactions promote the expression of defined gene sets both during development and for tissue homeostasis. Transcriptional regulatory networks are critical for the development and function of pancreatic β-cells. To date, few studies examining how transcription factors interact with co-activators, such a Mediator, have been performed in the pancreas. Tail module subunit MED15 is highly expressed in nascent β-cells and required for lipid metabolism, various stress responses, and TGF-β signalling, all of which are important for β-cell function. As such, we hypothesized that MED15 plays a role in β-cells. We found MED15 to be expressed during mouse pancreatogenesis and in mature β-cells. Expression of MED15 was impaired in human T2D islets suggesting it is important for mature β-cell function. After generating a β-cell specific knockout mouse (IM15KO), we observed defects in maturation as assessed by loss of β-cell maturation markers UCN3, MAFA, and GLUT2. In agreement with reduced GLUT2 expression, IM15KO cells had impaired glucose uptake and reduced glucose-stimulated insulin secretion, the hallmark process of β-cell maturation. ChIP-seq analysis determined that MED15 is bound to key GSIS related genes and Co-IP found transcription factors NEUROD1 and NKX6-1 to bind MED15. As the pancreas contains among the highest levels of Zn²⁺ in the body, we also found a role for MED15 in heavy metal stress response. Through this thesis, I provide evidence of the importance of Mediator in β-cell maturation and demonstrate an additional layer of control that modulates transcription factor function. A greater understanding of how Mediator and MED15 regulate β-cell maturation could help refine the generation of cell-based therapies for diabetes."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/75618?expand=metadata"@en ; skos:note """INVESTIGATING THE ROLE OF TRANSCRIPTIONAL COACTIVATOR MED15 IN BETA CELL MATURATION by Alexandre Zacharie Kadhim B.Sc., Simon Fraser University, 2015 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Medical Genetics) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) August 2020 © Alexandre Zacharie Kadhim, 2020 ii The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled: Investigating the Role of Transcriptional Coactivator MED15 in Beta Cell Maturation submitted by Alexandre Zacharie Kadhim in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Medical Genetics Examining Committee: Stefan Taubert, Associate Professor, Department of Medical Genetics, UBC Co-supervisor Francis Lynn, Associate Professor, Department of Surgery, UBC Co-supervisor Bruce Verchere, Professor, Department of Pathology and Laboratory Medicine, UBC Supervisory Committee Member James Johnson, Professor, Department of Cellular & Physiological Sciences, UBC University Examiner Janel Kopp, Assistant Professor, Department of Cellular & Physiological Sciences, UBC University Examiner Additional Supervisory Committee Members: Pamela Hoodless, Professor, Department of Medical Genetics, UBC Supervisory Committee Member William Gibson, Professor, Department of Medical Genetics, UBC Supervisory Committee Member iii Abstract The Mediator complex, a coregulator required for RNA pol II activity, interacts with specific transcription factors through its distinct subunits. These interactions promote the expression of defined gene sets both during development and for tissue homeostasis. Transcriptional regulatory networks are critical for the development and function of pancreatic β-cells. To date, few studies examining how transcription factors interact with co-activators, such a Mediator, have been performed in the pancreas. Tail module subunit MED15 is highly expressed in nascent β-cells and required for lipid metabolism, various stress responses, and TGF-β signalling, all of which are important for β-cell function. As such, we hypothesized that MED15 plays a role in β-cells. We found MED15 to be expressed during mouse pancreatogenesis and in mature β-cells. Expression of MED15 was impaired in human T2D islets suggesting it is important for mature β-cell function. After generating a β-cell specific knockout mouse (IM15KO), we observed defects in maturation as assessed by loss of β-cell maturation markers UCN3, MAFA, and GLUT2. In agreement with reduced GLUT2 expression, IM15KO cells had impaired glucose uptake and reduced glucose-stimulated insulin secretion, the hallmark process of β-cell maturation. ChIP-seq analysis determined that MED15 is bound to key GSIS related genes and Co-IP found transcription factors NEUROD1 and NKX6-1 to bind MED15. As the pancreas contains among the highest levels of Zn2+ in the body, we also found a role for MED15 in heavy metal stress response. Through this thesis, I provide evidence of the importance of Mediator in β-cell maturation and demonstrate an additional layer of control that modulates transcription factor iv function. A greater understanding of how Mediator and MED15 regulate β-cell maturation could help refine the generation of cell-based therapies for diabetes. v Lay Summary Diabetes mellitus impacts over 463 million people worldwide and is a complex disease. In type 2 diabetes, debilitating changes in the pancreas impair the body’s ability to control rising blood sugar levels. Specifically, the cells which produce the sugar-regulating hormone insulin, termed β-cells, become dysfunctional. DNA, the inherited biological blueprint for all organisms, must be read properly by cells to maintain a satisfactory state of health. The process of decoding the information stored in DNA, termed “transcription”, ensures normal development and physiological function. This thesis uncovers a new role for MED15, a protein that physically links DNA to proteins that regulate transcription. My findings suggest that MED15 is critical to produce functionally mature β-cells. This is accomplished through MED15’s interactions with different proteins which influence transcription. Understanding how and why β-cells need MED15 to develop and mature will allow new potential treatments for diabetes by generating mature β-cells for transplantation. vi Preface Animal studies were reviewed and approved by the University of British Columbia Committee on Animal Care under protocols A17-0158 and A18-0213. All experiments were performed and analyzed by A.Z. Kadhim except those outlined below: Chapter 3: Med15fl/fl mice were generated by S. Taubert, R. Cullen, and M. Wang. S. Sasaki performed Nanostring on healthy and T2D islets. E.E. Xu, T. Speckmann, and C. Nian assisted with IPGTT, insulin ELISA, and islet isolation. R. Shi performed mitochondrial function and morphological measurements. E.E. Xu performed wet lab portion of RNA-seq and assisted in bioinformatic analysis. Chapter 4: X. Cheng assisted with ChIP-seq data analysis. B.G. Hoffman assisted in performing chromatin segmentation analyses. C. Nian and T. Speckmann assisted in islet isolation and culturing of MIN6 cells. Chapter 5: E.E. Xu assisted in islet isolation for TEM. M.Y.Y. Lee performed blinded quantification of TEM granules. N. Shomer assisted in performing IF for ZnT8. X. Cheng cultured A549 cells used in ChIP-qPCR. N. Shomer performed culturing of C. elegans and assisted with microscopy. N. Shomer performed quantification of gut granules. Data contained in Chapters 3 and 4 are currently in preparation for submission to a peer-reviewed journal as follows: A.Z. Kadhim, E.E. Xu, S. Sasaki, X. Cheng, S.L.J. Sproul, T. Speckmann, C. Nian, R. Cullen, R. Shi, B.G. Hoffman, D.S. Luciani, S. Taubert, F.C. Lynn. Transcriptional coactivator MED15 is required for β-cell maturation. Data contained in Chapter 3 have been published as part of E.E. Xu’s thesis: E.E. Xu. (2017). The roles of SOX4 and MED15 in the development and maintenance of pancreatic β-cells. Data contained in Chapter 5 have been published as follows: N. Shomer*, A.Z. Kadhim*, J.M. Grants*, X. Cheng, D. Alhusari, F. Bhanshali, A.F. Poon, M.Y.Y Lee, A. Muhuri, J.I. Park, J. Shih, D. Lee, S.V. Lee, F.C. Lynn, S. Taubert. (2019). Mediator subunit MDT-15/MED15 and Nuclear Receptor HIZR-1/HNF4 cooperate to regulate toxic metal stress responses in Caenorhabditis elegans. PLoS Genetics. 15, e1008508. *: co-first authorship Data contained in Chapter 5 have also been published as part of N. Shomer’s thesis: N. Shomer. (2018). A regulatory mechanism of zinc homeostasis involving the mediator subunit MDT-15 and the transcription factor HIZR-1. This thesis was written by A.Z. Kadhim with revisions from F.C. Lynn and S. Taubert. vii Table of Contents Abstract .................................................................................................................................... iii Lay Summary ............................................................................................................................ v Preface ...................................................................................................................................... vi Table of Contents .................................................................................................................... vii List of Tables ........................................................................................................................... xii List of Figures ........................................................................................................................ xiii List of Abbreviations ............................................................................................................. xvi Acknowledgements .............................................................................................................. xviii Dedication .............................................................................................................................. xix Chapter 1: Introduction ......................................................................................................... 20 Diabetes mellitus ....................................................................................................... 20 1.1.1Brief history of diabetes mellitus ............................................................................... 20 Classification and diagnosis of diabetes ..................................................................... 22 1.2.1Type 1 diabetes mellitus ............................................................................................ 22 1.2.2Type 2 diabetes mellitus ............................................................................................ 24 1.2.3Maturity onset diabetes of the young (MODY) .......................................................... 26 Current therapies for diabetes .................................................................................... 27 β-cell function and insulin secretion .......................................................................... 28 1.4.1The role of zinc in the pancreas and insulin secretion ................................................. 30 1.4.2Heavy metal stress response in the pancreas .............................................................. 32 The pancreas and islets of Langerhans: Development and function ............................ 32 viii 1.5.1Endocrine cell types and function .............................................................................. 33 1.5.2Early pancreas, endocrine, and β-cell development .................................................... 34 1.5.3Functional β-cell maturation ...................................................................................... 36 Transcriptional regulation by epigenetic changes ....................................................... 38 1.6.1Transcriptional regulation during β-cell development and maturation ........................ 39 Mediator complex...................................................................................................... 41 1.7.1The structure of Mediator and its modules ................................................................. 43 1.7.2Mediator as a modulator of cell lineage specification ................................................. 45 1.7.3Mediator and human disease ...................................................................................... 46 1.7.4Mediator and metabolism .......................................................................................... 47 1.7.5Tail module subunit MED15...................................................................................... 48 Thesis investigation ................................................................................................... 49 Chapter 2: Materials and Methods ........................................................................................ 51 Animal studies ........................................................................................................... 51 Human islet studies.................................................................................................... 51 Genotyping ................................................................................................................ 52 Pancreatic islet isolation ............................................................................................ 52 Cell culture ................................................................................................................ 53 Transfections and siRNA knockdown ........................................................................ 53 Perifusion GSIS assay ............................................................................................... 54 Static incubation glucose-stimulated insulin secretion (GSIS) assay .......................... 54 Glucose uptake .......................................................................................................... 55 Intraperitoneal glucose tolerance tests (IPGTT) and serum insulin quantification ....... 56 ix Mitochondrial function assay 3D morphometric analysis ........................................... 56 Mitochondrial 3D morphometric analysis .................................................................. 57 RNA extraction and qPCR ......................................................................................... 57 RNA-sequencing ....................................................................................................... 58 ChIP-seq.................................................................................................................... 59 ChIP-qPCR ............................................................................................................... 60 Bioinformatics analysis of RNA-seq and ChIP-seq .................................................... 60 Chromatin segmentation analysis ............................................................................... 61 Tissue processing ...................................................................................................... 61 Immunofluorescence ................................................................................................. 62 Zinc staining with dithizone and FluoZin-3................................................................ 63 Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) ................. 63 Western blotting ........................................................................................................ 64 Co-immunoprecipitation ............................................................................................ 65 Insulin ELISA ........................................................................................................... 65 TEM imaging and analysis ........................................................................................ 66 C. elegans strains and growth conditions ................................................................... 67 C. elegans zinc storage granule analysis .................................................................... 67 Statistics .................................................................................................................... 68 Chapter 3: MED15 is Required for β-Cell Maturation ......................................................... 75 Introduction ............................................................................................................... 75 Results....................................................................................................................... 77 3.2.1MED15 is expressed in mature insulin secreting β-cells ............................................. 77 x 3.2.2Loss of Med15 leads to impaired glucose-stimulated insulin secretion ....................... 81 3.2.3Med15 loss causes defects in glucose uptake in β-cells .............................................. 85 3.2.4Loss of Med15 leads to downstream mitochondrial defects ........................................ 85 3.2.5Loss of Med15 impairs β-cell maturation ................................................................... 87 3.2.6Med15 loss alters mitochondrial networking in β-cells ............................................... 91 Discussion ................................................................................................................. 92 Chapter 4: β-cell Maturation Transcription Factors Bind MED15 at Key Loci ................. 96 Introduction ............................................................................................................... 96 Results....................................................................................................................... 98 4.2.1MED15 and MED1 are found at the Iapp enhancer and promoter .............................. 98 4.2.2MED15 and MED1 have similar yet distinct binding profiles .................................. 100 4.2.3MED15 binds and regulates critical β-cell genes ...................................................... 102 4.2.4MED15 and MED1 are bound to genomic enhancers and promoters ........................ 103 4.2.5MED15 is bound near enhancers of regulated genes ................................................ 105 4.2.6MED15 and MED1 primarily bind intronic and intergenic regions .......................... 107 4.2.7MED15 binds critical β-cell maturation genes in mouse islets .................................. 108 4.2.8ChIP-seq of MED15 bound genes reveals several potential binding partners............ 110 4.2.9MED15 colocalizes with β-cell maturation transcription factors at key loci near β-cell maturation genes ............................................................................................................. 111 4.2.10MED15 interacts with β-cell maturation transcription factors NKX6-1 and NEUROD1 ...................................................................................................................... 114 4.2.11MED15 may also interact with chromatin remodeling complexes .......................... 117 Discussion ............................................................................................................... 120 xi Chapter 5: MED15/MDT-15 is Required for Zinc and Toxic Metal Stress Response ....... 123 Introduction ............................................................................................................. 123 Results..................................................................................................................... 127 5.2.1Loss of Med15 impairs zinc import into insulin secretory granules .......................... 127 5.2.2Loss of Med15 results in impaired intracellular zinc homeostasis............................. 129 5.2.3MED15 binds loci of zinc and cadmium metal detoxification genes......................... 131 5.2.4MED15 binds heavy metal stress response gene MT2A in a Cd2+ inducible fashion . 135 5.2.5C. elegans mdt-15 is required to express genes important for excess zinc storage..... 136 Discussion ............................................................................................................... 140 Chapter 6: Conclusion .......................................................................................................... 144 Research summary .................................................................................................. 144 MED15 is required for functional maturation........................................................... 145 MED15 interacts with several β-cell transcription factors ........................................ 148 MED15 plays a conserved role in heavy metal stress response ................................. 150 Future directions ...................................................................................................... 151 References ............................................................................................................................. 153 Appendices ............................................................................................................................ 185 Appendix A Top 50 MED15 bound genes ........................................................................... 185 Appendix B Top 50 MED1 bound genes ............................................................................. 188 xii List of Tables Table 1: Genotyping primers ..................................................................................................... 68 Table 2: Taqman qPCR primers ................................................................................................ 69 Table 3: Conventional qPCR primers for expression ................................................................. 70 Table 4: Conventional qPCR primers for ChIP .......................................................................... 71 Table 5: Antibody list ............................................................................................................... 72 Table 6: List of worm strains .................................................................................................... 74 Table 7: MED15 and MED1 MIN6 ChIP-seq summary .......................................................... 101 Table 8: MED15 and MED1 islet ChIP-seq summary ............................................................. 108 xiii List of Figures Figure 1: Glucose-stimulated insulin secretion in β-cells ........................................................... 29 Figure 2: Endocrine cell differentiation is driven by transcription factor cascades ..................... 35 Figure 3: Transcriptional machinery and Mediator .................................................................... 42 Figure 4: Mediator is composed of 25-31 subunits which interact with specific transcription factors ....................................................................................................................................... 44 Figure 5: Mediator subunits are broadly expressed at varying levels in mouse pancreatic islets. 78 Figure 6: β-cell maturation transcription factors are coexpressed with MED15 in human β-cells. ................................................................................................................................................. 79 Figure 7: MED15 is expressed in mouse and human adult islets, and insulin positive stem cell derived β-like cells. ................................................................................................................... 80 Figure 8: MED15 is expressed in the nucleus of embryonic and postnatal β-cells in mice. ........ 81 Figure 9: IM15KO mice are glucose intolerant and have less serum insulin. ............................. 82 Figure 10: Loss of Med15 impairs GSIS but does not affect KCl-induced insulin secretion. ...... 83 Figure 11: IM15KO are unable to take up glucose in a timely manner. ...................................... 84 Figure 12: Loss of Med15 does not impair mitochondrial function but may affect mitochondrial networking in β-cells................................................................................................................. 86 Figure 13: GO term analysis of RNA-seq data show MED15 regulates genes important for β-cell function. ................................................................................................................................... 87 Figure 14: MED15 regulates genes important for β-cell maturation. .......................................... 88 Figure 15: Gene expression analysis of IM15KO islets reveal no increases in other endocrine cell types. ........................................................................................................................................ 89 xiv Figure 16: qPCR and immunofluorescence confirm loss of maturation genes in IM15KO islets. ................................................................................................................................................. 90 Figure 17: IM15KO have reduced mitochondrial count but increases in networking ................. 92 Figure 18: Schematic of ChIP-seq workflow ............................................................................. 99 Figure 19: ChIP-qPCR analysis for MED15 and MED1 shows binding at Iapp enhancer and promoter regions ..................................................................................................................... 100 Figure 20: MED15 and MED1 bind similar, yet distinct sets of genes ..................................... 102 Figure 21: MED15 binds and regulates key β-cell maturation genes ........................................ 103 Figure 22: MED15 binds more H3K4me1 regions compared to MED1 in β-cells .................... 104 Figure 23: MED15 and MED1 primarily bind distal enhancers and introns ............................. 106 Figure 24: MED15 and MED1 bind intergenic regions ............................................................ 107 Figure 25: ChIP-seq analysis in IM15KO islets shows loss of binding at Med15 regulated β-cell genes ...................................................................................................................................... 109 Figure 26: MED15 bound genes overlap with genes bound by several other transcriptional factors ..................................................................................................................................... 111 Figure 27: ChIP-seq tracks show MED15 may be associated with β-cell maturation factors near specific genes.......................................................................................................................... 112 Figure 28: MED15/MED1 bound regions at enhancers and motif analysis of all MED15 bound regions .................................................................................................................................... 113 Figure 29: Loss of MED15 is similar to loss of β-cell maturation transcription factors ............ 115 Figure 30: Overlap of MED15 bound genes with NKX6-1 and NEUROD1 bound genes ........ 116 Figure 31: MED15 physically interacts with transcription factors NEUROD1 and NKX6-1 .... 117 Figure 32: Gene ontology term analysis of biological processes using MED15 bound regions. 118 xv Figure 33: Gene ontology term analysis of cellular components using MED15 bound regions. 119 Figure 34: MED15 bound gene overlap with coactivator ChIP-seq and RNA-seq data ............ 120 Figure 35: Zn2+ storage granules in mammalian β-cells and C. elegans intestinal cells ............ 126 Figure 36: Islets of IM15KO mice have reduced expression of zinc transporter Slc30a8/ZNT8 ............................................................................................................................................... 127 Figure 37: IM15KO mouse islets have less and smaller zinc-rich, mature insulin secretory granules .................................................................................................................................. 128 Figure 38: IM15KO islets have reduced zinc content .............................................................. 130 Figure 39: GO-term analysis shows IM15KO bound genes involved in metal ion binding ....... 131 Figure 40: MED15 is associated with the metal detoxification genes Slc30a8, Mt1, and Mt2 in β-cells ........................................................................................................................................ 132 Figure 41: MED15 binds regulatory elements in the Slc30a8 and Mt1 genes in a zinc inducible fashion .................................................................................................................................... 133 Figure 42: Genes required for heavy metal stress response are zinc inducible .......................... 134 Figure 43: Human MED15 binds metallothionein MT2A in a cadmium dependent fashion ...... 135 Figure 44: mdt-15(tm2182) mutants have normal gut granule formation in the absence of Zn2+ supplementation ...................................................................................................................... 137 Figure 45: mdt-15(tm2182) mutants have impaired gut granule formation in conditions of excess Zn2+ ........................................................................................................................................ 138 Figure 46: Impairment of zinc granules is independent of MDT-15’s role in lipid metabolism 140 Figure 47: Summary of MED15’s interactions in β-cells ......................................................... 149 xvi List of Abbreviations ANOVA Analysis of variance ARC Activator-recruited cofactor ARX Aristaless ATAC-seq Assay for transposase-accessible chromatin using sequencing ATP Adenosine triphosphate CDK8 Cyclin dependent kinase 8 ChIP-seq Chromatin immunoprecipitation-sequencing CRSP Cofactor required for Sp1 CTCF CCCTC-binding factor CTD Carboxy terminal domain DRIP Vitamin D receptor-interacting protein DTZ Dithizone E Embryonic day EED Embryonic ectoderm development ELISA Enzyme-linked immunosorbent assay ER Endoplasmic Reticulum G6P Glucose 6 phosphate GCG Glucagon GCK Glucokinase GLP1R Glucagon like peptide 1 receptor GLUT2 Glucose transporter 2 GSIS Glucose-stimulated insulin secretion GTF General transcription factor GWAS Genome wide association study H3K27Ac Histone 3 lysine 27 acetylation H3K27me3 Histone 3 lysine 27 trimethylation H3K4me1 Histone 3 lysine 4 monomethylation HAT Histone acetyltransferase HbA1c Glycated hemoglobin A1c HMT Histone methyltransferase HNF4α Hepatocyte nuclear factor 4α HOMER Hypergeometric optimization of motif enrichment IDR Intrinsically disordered regions IF Immunofluorescence IM15KO Med15fl/fl; Ins1Cre/+ xvii INS Insulin INSR Insulin receptor IP Immunoprecipitation IPGTT Intraperitoneal glucose tolerance test KATP ATP-sensitive potassium channel MAFA V-maf musculoaponeurotic fibrosarcoma oncogene homolog A MED15 Mediator complex subunit 15 MODY Maturity onset diabetes of the young NEUROD1 Neuronal differentiation 1 NEUROG3 Neurogenin 3 NHR-49 Nuclear hormone receptor 49 NKX6-1 NK homeobox 6-1 OGTT Oral glucose tolerance test PDX1 Pancreatic and duodenal homeobox 1 PGC1-α Peroxisome proliferator-activated receptor gamma coactivator 1-α PIC Pre-initiation complex PP Pancreatic polypeptide PRC2 Polycomb repressive complex 2 RNA Pol II RNA polymerase II RNA-seq RNA-sequencing RPKM Reads Per Kilobase of transcript, per Million mapped reads SLC30A8 Solute carrier family 30 member 8 SREBP Sterol response element binding protein SST Somatostatin SYT4 Synaptotagmin 4 T1D Type 1 diabetes T2D Type 2 diabetes TEM Transmission electron microscopy TGF-β Transforming growth factor β TMRE Tetramethylrhodamine, ethyl ester TRAP Thyroid hormone receptor-associated protein t-SNE t-Distributed Stochastic Neighbor Embedding TrxG Trithorax group UCN3 Urocortin 3 VDCC Voltage-dependent calcium channels WT Wild-type ZIP Zrt-/Irt-like protein ZNT Zinc transporter xviii Acknowledgements My graduate school experience has been a greatly rewarding journey. First and foremost, I would like to express the utmost gratitude to my supervisors Dr. Stefan Taubert and Dr. Francis Lynn for all their support and excellent mentorship. I feel fortunate to have been under their supervision; without them, none of this work would have been possible. I would also like to thank the members of my advisory committee: Dr. William Gibson, Dr. Pamela Hoodless, and Dr. Bruce Verchere for providing excellent guidance throughout the progression of my degree. Thank you to all members of the Taubert and Lynn labs, past and present, for your help and intellectually stimulating conversations: Eric Xu, Thilo Speckmann, Nicole Krentz, Paul Sabatini, Naomi Shomer, Kayoung Lee, Grace Goh, Jennifer Grants, Shugo Sasaki, Dahai Zhang, Jean Cheng, Sam Yoon, Shannon Sproul, Deema Alhusari, Marcus Woodley, Kelsie Doering, Forum Bhanshali, Caitlyn Xu, Judith Yan, Meixia Dan, Cuilan Nian, Elizabeth Lin, and any others I have missed. I would like to give a special thanks to Eric Xu and Naomi Shomer, whom I worked closely with and learned so much from. Additionally, I would like to thank members from other labs and people who have provided me with guidance throughout my scientific journey, specifically: CK Wong, Daniel Gamu, Rocky Shi, Daniel Pasula, Peter Zou, Sigrid Alvarez, Søs Skovsø, Dominika Nackiewicz, Shawn Shortill, Stephanie Campbell, Ben Vanderkruk, Yuka Obayashi, Nina Maeshima, Cassie McDonald, Meilin An, Jessica Blaquiere, Vilaiwan Fernandes, Eric Hall, Nathan Wray, Liz Hoesing, Dipa Pradhan-Sundd, Sharan Swarup, and Dr. Esther Verheyen. I would like to give a special thanks to Jessica Blaquiere and Vilaiwan Fernandes for providing me with excellent mentorship during my undergraduate degree. I would like to give a special thanks to Dr. Brad Hoffman and Dr. Dan Luciani for lending their expertise to help me answer important scientific questions. The scientifically stimulating and highly collaborative environment provided at BCCHR and CMMT are among the best anyone can ask for. I would also like to thank the funding agencies who have supported this work: Diabetes Canada, CIHR, and BCCHR. Lastly, I will be forever grateful to my family for their endless support throughout my entire academic career. Thank you, Mom and Dad, for supporting me and always being excited to hear about my research progress. An incredibly special thank you to Tara, for supporting me and sacrificing so much so that I could finish my degree - all complete in the sight of seeds of life with you. xix Dedication To Tara for your endless support. And you and I climb over the sea to the valley20 Chapter 1: Introduction Diabetes mellitus Diabetes mellitus, or diabetes, is a group of serious metabolic disorders which are estimated to be found in over 463 million people worldwide, with approximately 1 in 11 people diagnosed with the disease (Williams et al., 2020). Symptoms of diabetes include polyuria, polydipsia, polyphagia, fatigue, and weight loss. According to Diabetes Canada, approximately 29% of Canadians currently live with diabetes or are prediabetic (Bilandzic and Rosella, 2017). Undiagnosed diabetes can result in prolonged hyperglycemia if left untreated, which can lead to serious complications such as heart disease, stroke, end-stage renal disease, retinopathy, and lower-extremity amputations, among others (Harding et al., 2019). As such, the costs associated with treating diabetes and its complications are expected to rise to $15 billion in Canada by 2022 (Bilandzic and Rosella, 2017). Moreover, diabetes can lead to premature death, with diabetes patients having a lifespan that is reduced by 5-15 years. It is estimated that, in 2010, diabetes accounted for 6.8% of global mortality with 3.96 million deaths worldwide (Roglic and Unwin, 2010). This presents an unmet medical need for treatment and prevention of diabetes. 1.1.1 Brief history of diabetes mellitus The earliest observations of diabetes date back to ancient Egypt, where the mention of a disease causing excessive urination can be found in Ebers Papyrus dating to approximately 1550 BC (von Engelhardt, 1989). The term diabetes, derived from the Greek word diabainein which means “to go through” or “siphon”, was first used during the 2nd century BC (von Engelhardt, 1989). During the 1st century AD, Greek physician Aretaios first described the symptoms of diabetes as intolerable thirst, burning of the intestines, and passage of large amounts of urine 21 (Leopold, 1930). It was not until the 18th century when the distinction between diabetes insipidus and mellitus were made, with the first clue found in 1674 by Thomas Willis, who discovered the sweetness of urine in diabetic individuals (von Engelhardt, 1989). Throughout the 19th century, emerging medical evidence such as the distinction of diabète maigre and diabète gras, emaciated and fatty diabetes respectively, by Étienne Lancereaux further suggested the existence of different forms of the disease (Wright and McIntyre, 2020). In 1869, German medical student Paul Langerhans observed structures in the pancreas measuring 0.1 to 0.24 mm in diameter, which he termed “zellhaufen”, which translates to “cell heaps” (Jolles, 2002). These clusters of cells would later be termed the “islets of Langerhans”. A major breakthrough for diabetes research occurred in 1889 when physicians Oskar Minkowski and Joseph von Mering identified the pancreatic origin of diabetes (Luft, 1989). This was discovered by the observation that pancreatectomized dogs developed symptoms of diabetes and eventually died (v. Mering and Minkowski, 1890). It was later hypothesized by English physiologist Ernest Starling that the pancreas secretes a hormone which allows tissues to utilize sugars (Medvei, 1993). In 1921, Canadian surgeon Frederick Banting, working in the lab of John Macleod at the University of Toronto, began to collaborate with medical student Charles Best to study diabetes (Karamanou, 2016). They found that a crude extract, prepared by ligation, extraction, and freezing of dog pancreata, was able to reduce blood glucose levels in pancreatectomized dogs (Banting et al., 1922). With the help of biochemist James Collip, this extract was further purified for use in humans and named “isletin” (later to be known as insulin). This work led to the discovery of insulin and the subsequent Nobel prize for Banting and Macleod. Thus, this 22 discovery has significantly improved the lives of millions living with diabetes, while simultaneously highlighting the central role of the pancreas in glucose control and diabetes. Classification and diagnosis of diabetes Diabetes is the result of the body’s inability to produce or respond to insulin. Therefore, diagnostic criteria involve measuring plasma glucose and long-term changes associated with hyperglycemia such as glycated hemoglobin A1c (HbA1c). The diagnostic criteria for diabetes includes a fasting plasma glucose level ≥ 7.0 mmol/L, a 2-hour plasma glucose ≥ 11.1 mmol/L in a 75g oral glucose tolerance test (OGTT), and/or a glycated HbA1C ≥ 6.5% (Punthakee et al., 2018). Patients with a fasting plasma glucose level between 6.1 to 6.9 mmol/L, a 2-hour plasma glucose level in a 75g OGTT between 7.8–11.0 mmol/L, or glycated HbA1C between 6.0-6.4% are considered to be pre-diabetic. The majority of cases can be classified as type 1 and type 2 diabetes mellitus (T1D and T2D, respectively). Several other forms of the disease also exist such as gestational diabetes mellitus (GDM), neonatal diabetes mellitus (NDM), mitochondrial diabetes mellitus, and maturity onset diabetes of the young (MODY) also exist (Flannick et al., 2016). With no known cure available, diabetes presents a serious threat to global health. 1.2.1 Type 1 diabetes mellitus T1D is a chronic progressive autoimmune disorder, which leads to the destruction of β-cells, thus resulting in insulin deficiency and hyperglycemia (Katsarou et al., 2017). During clinical onset of T1D, immune cells such as CD4+ and CD8+ T cells, dendritic cells, macrophages, and B cells infiltrate islets (Krogvold et al., 2016). This immune dysfunction directly causes loss of β-cells and an absolute reduction in the amount of secreted insulin. In 23 most T1D cases, approximately 70-80% of β-cells are lost, leaving severely reduced islet function (Chen et al., 2017). T1D accounts for 10-15% of all diabetes cases and is most common in children aged 15 or younger, with peaks in presentation at 5-7 years of age (Harjutsalo et al., 2008; Katsarou et al., 2017). T1D is a complex polygenic disorder, with genetic factors demonstrating low penetrance (Katsarou et al., 2017). Among monozygotic twins, the concordance rate is approximately 30%, suggesting that environmental factors such as viral infection or chemical exposure may contribute to disease onset or progression (Knip et al., 2005; Redondo et al., 2008). The prevalence of T1D is highest in Scandinavian countries, the highest being Finland, followed by European countries, North America, and Australia (Atkinson et al., 2014; Katsarou et al., 2017). Interestingly, T1D appears heterogeneous, with certain cases presenting without autoantibodies, yet a complete loss of β-cells (Gianani et al., 2010). Because approximately 70-90% of T1D cases present with an autoimmune component such as autoantibodies, classifications of T1D into subtypes type 1a (autoimmune) and type 1b (idiopathic) have been proposed, with type 1b presentation tending to occur at older ages (between 20-30 years old) (Atkinson et al., 2014). Known autoantibody targets include insulin, zinc transporter 8 (ZNT8), 65-kDa isoform of glutamic acid decarboxylase (GAD65), and insulinoma antigen 2 protein (IA-2) (Calderon and Sacks, 2014). Most individuals presenting with two or more autoantibodies against islet antigens will develop symptoms of T1D, with peak detection of autoantibodies occurring between 1-2 years of age (Insel et al., 2015; Krischer et al., 2015). Although complex, genetic factors such as MHC class II haplotype clusters on chromosome 6p21 (HLA-DR3-DQ2 and HLA-DR4-DQ8) are major risks for developing T1D, with over 90% of T1D patients carrying one of these haplotypes (Noble and Valdes, 2011; Singal 24 and Blajchman, 1973). Conversely, protection from T1D is conferred by the DRB*602 allele (Ettinger and Kwok, 1998). Genome-wide association studies (GWAS) identified over 50 genes contributing to increased risk of T1D including: PTPN22, CTLA4, SH2B3, TYK2, and CLEC16A loci (Katsarou et al., 2017). Interestingly, these genes also confer increased risk for other autoimmune disorders such as autoimmune thyroid and coeliac diseases (Cerolsaletti et al., 2019). 1.2.2 Type 2 diabetes mellitus T2D is a serious metabolic disorder resulting in hyperglycemia and accounts for 90% of diabetes cases. Characterized by the progressive impairment of β-cell function, namely glucose stimulated insulin secretion (GSIS), T2D can lead to long-term complications such as retinopathy, heart disease, and nephropathy (Kahn et al., 2014). During β-cell failure, increased insulin demand caused by peripheral tissue insulin resistance places excess metabolic stress on β-cells, leading to dedifferentiation and death (Szabat et al., 2012). The concomitant increase in insulin biogenesis and secretion can lead to endoplasmic reticulum, hypoxic, and oxidative stresses in β-cells (Back et al., 2012; Gerber and Rutter, 2017; Nyengaard et al., 2004). Moreover, obesity can lead to insulin resistance and subsequently T2D (Wilcox, 2005). Furthermore, obesity induced inflammation, by adipocyte related increases in pro-inflammatory cytokines, may contribute to β-cell dysfunction (de Luca and Olefsky, 2008). As of 2015, countries with the highest numbers of adults living with T2D include China, India, and the United States (Zheng et al., 2018b). Certain regions of the world such as Pacific nations (American Samoa, Polynesia, and Micronesia) and the Middle East (primarily Saudi Arabia) contain hotspots for T2D with prevalence greater than 25% (Al-Rubeaan et al., 2015; 25 Zhou et al., 2016). In Canada, rates of T2D are higher among First Nations populations, with diagnosis occurring at younger ages (Weisman et al., 2018). Rates of T2D among on-reserve First Nations individuals are estimated at 17.2%, and 10.3% for off-reserve populations, as compared to 5% of the general population (Crowshoe et al., 2018). Genetically, common risk alleles are well known to contribute to T2D susceptibility (Fuchsberger et al., 2016). Since the first GWAS reports in 2007 (Scott et al., 2007; Sladek et al., 2007), over 150 genes have been reported to increase susceptibility to T2D (Lawlor et al., 2017a). Risk genes include: transcription factor 7-like 2 (TCF7L2), which regulates insulin secretion (Wang et al., 2013); paired box 4 (PAX4), which promotes β-cell survival (Lorenzo et al., 2017); and zinc transporter SLC30A8, which is important for zinc uptake and insulin granule maturation (Bosco et al., 2010). These genetic susceptibility factors are often compounded by environmental stresses (Kahn et al., 2014). As such, combined genetic and environmental factors can lead to excess β-cell stress and loss of important β-cell transcription factors MAFA, NKX6-1, and PDX1, among others (Guo et al., 2013a). Because these genes are important for β-cell identity and maturity, impairment due to stress can affect expression of important downstream components such as the glucose transporter GLUT2/SLC2A2 (Lawlor et al., 2017a). In an individual harbouring multiple T2DM risk alleles, stresses such as poor diet may lead to accelerated β-cell failure. In humans and rodents, high-fat diets are known to contribute to β-cell dysfunction through lipotoxicity and excess proliferative stress (Chang-Chen et al., 2008; Tuomilehto et al., 2001). Increased secretory demands caused by insulin resistance can also lead to unfolded or misfolded proinsulin and endoplasmic reticulum (ER) stress (Back et al., 2012). Moreover, lipotoxicity can induce oxidative stress and reduce proliferative capacity of β-cells (Ye et al., 26 2019). At present, it is recommended for all individuals who are overweight or at risk for T2D to consider weight loss (DeFronzo et al., 2015). Because obesity and peripheral insulin resistance are associated, improving insulin tolerance would relieve stress on β-cells. Additional changes to diet are recommended as such: increasing fiber consumption, reducing protein intake, reducing simple sugar intake, and limiting fat intake to 7% of total calories (Khazrai et al., 2014). Further behavioural changes, such as high intensity interval training, can improve glucose tolerance and cardiovascular health (Francois and Little, 2015). 1.2.3 Maturity onset diabetes of the young (MODY) Monogenic forms of diabetes, often mislabeled as “type 1 or 2 like”, are caused by defects in single genes. With a prevalence of 2-5% among all diabetes cases within young people, maturity onset diabetes of the young (MODY) is due to autosomal dominant mutations and occurs before the age of 25 (Firdous et al., 2018). Some of the first cases of MODY described were caused by nonsense mutations in glucokinase (GCK – MODY2) (Vionnet et al., 1992). Four years later, mutations in hepatocyte nuclear factor 4 and 1α (HNF4α and HNF1α) were observed and labelled as MODY1 and MODY3, respectively (Yamagata et al., 1996a, 1996b). Subsequently, MODY4 was molecularly described in 1997, when Stoffers et al. described a single nucleotide deletion within codon 63 of the pancreatic and duodenal homeobox 1 (PDX-1) gene (Stoffers et al., 1997). Whereas homozygote or compound heterozygote PDX-1 mutations manifest phenotypically as pancreatic agenesis, heterozygous MODY mutations in PDX-1 cause late-onset diabetes, which may be misdiagnosed as T2D. Thus, mutations that cause MODY affect important developmental regulatory networks, whose dysregulation leads to improper pancreas or β-cell formation. 27 Current therapies for diabetes Depending on the type of diabetes, several therapies exist, including exogenous insulin, pharmaceutical intervention, and islet transplantation. In cases of T1D, administration of insulin by injection or continuous subcutaneous insulin infusion (CSII; via insulin pumps) are required because β-cells are lost (Atkinson et al., 2014; Sora et al., 2019). Studies have demonstrated greater decreases in HbA1c among CSII users compared to insulin injections, suggesting greater glycemic control (Yeh et al., 2012). Other methods for insulin administration, such as inhalation, have been explored but face limitations and safety concerns (Al-Tabakha and Arida, 2008; Oleck et al., 2016). Current pharmaceutical approaches for diabetes also include several classes of drugs with various mechanisms of action (Chaudhury et al., 2017). To increase insulin sensitivity, compounds such as biguanides and thiazolidinediones (TZD) can be used, with the biguanide metformin often being the drug of first choice in T2D (Shurrab and Arafa, 2020). Secretagogues such as sulfonylureas and meglitinides work by increasing insulin secretion from β-cells and are used in T2D and certain types of MODY (Kim, 2015). Peptide analogues such as exendin-4 and liraglutide interact with the glucagon-like peptide 1 receptor (GLP1R) to stimulate insulin secretion through incretin signalling (Aroda, 2018). Glycosurics such as sodium-glucose cotransporter (SGLT2) inhibitors reduce blood glucose by preventing reuptake of glucose in the kidneys and are preferred as a second agent in combination with metformin for T2D (Tentolouris et al., 2019). Finally, α-glucosidase inhibitors do not exert their action via insulin secretion or sensitivity, but slow the digestion of starch to allow endogenous insulin to manage blood glucose levels in a controllable fashion (Van De Laar et al., 2005). 28 Islet transplantation from cadaveric donors has emerged as an effective β-cell replacement therapy for T1D patients (Rickels and Robertson, 2019). Pioneering work in 2000 by James Shapiro at the University of Alberta described the first consistent use of islet transplantation to establish proper glycemic control in T1D patients (Shapiro et al., 2000). Although effective in most cases, issues such as transplant rejection and the limited availability of usable islets present significant hurdles (Rickels et al., 2007). The use of in vitro derived β-cells has shown promise as a treatment for T1D and potentially T2D, presenting a potentially unlimited source of transplantable material (Tremmel et al., 2019). However, current directed differentiation protocols from stem cells do not produce mature β-cells, which are capable of physiologic glucose responsiveness (Rezania et al., 2014). Recent advances have shown significant improvements by attempting to recapitulate the β-cell niche by reaggregating differentiated β-like cells (Hogrebe et al., 2020; Nair et al., 2019; Sneddon et al., 2018; Velazco-Cruz et al., 2019; Veres et al., 2019). While not currently ready for implementation, the generation of stem cell derived β-cells, which are glucose responsive to meet physiological requirements, remains one of the most promising strategies towards a treatment and potential cure for diabetes. To achieve this goal, it is critical that scientists fully understand how mature β-cells form and develop. β-cell function and insulin secretion Insulin is an endocrine peptide hormone responsible for facilitating glucose uptake and maintaining glucose homeostasis (Petersen and Shulman, 2018). In the human body, the β-cells of the pancreatic islets secrete insulin in an oscillatory manner to maintain narrow euglycemic levels between 3.9 and 7.1 mmol/L (Schmitz et al., 2008). Because β-cell dysfunction leads to 29 diabetes, β-cells are central to diabetes etiology (Park and Woo, 2019). As such, the key function of the β-cells is to secrete insulin as needed. GSIS, wherein the β-cells respond to increased glucose levels following meals, is the process by which β-cells secrete insulin as this occurs (Figure 1) (Grodsky et al., 1963). In human β-cells, glucose uptake is controlled by GLUT1 (SLC2A1), with GLUT2 (Slc2a2) being the predominant form in rodents (McCulloch et al., 2011). Upon glucose uptake, glucose is metabolized via glycolysis and oxidative phosphorylation, leading to increased levels of ATP, Figure 1: Glucose-stimulated insulin secretion in β-cells Glucose-stimulated insulin secretion (GSIS) begins with glucose transport into the cell. Following phosphorylation into glucose-6-phosphate (G6P), glycolysis produces pyruvate, which is used to produce ATP. ATP sensitive potassium channels (KATP channel) then close, leading to increased membrane potential. This allows intracellular influx of calcium ions (Ca2+) via voltage-dependent calcium channels (VDCC) and exocytosis of Zn2+ rich insulin secretory granules (illustration created with BioRender.com). 30 increasing the intracellular ATP/ADP ratio. This leads to the closure of ATP-sensitive potassium (KATP) channels, depolarization, and the opening of voltage-dependent Ca2+ channels (VDCC) (Dean and Matthews, 1968). Influx of Ca2+ then causes the docking of insulin secretory granules to the cell membrane and thus insulin secretion into the bloodstream (Grodsky and Bennett, 1966). Therefore, insulin secretion is tightly regulated through cellular processes. Following its secretion by the pancreatic β-cells, insulin regulates glucose metabolism and utilization throughout the body. Specifically, it acts on peripheral tissues such as muscle, the liver, or adipose tissue, wherein it activates the insulin receptor (INSR), a tyrosine kinase that consists of two α and two β glycoprotein subunits (Kido et al., 2001). Binding to the extracellular α subunit induces a conformational change, INSR dimerization, and phosphorylation of insulin responsive substrate (IRS) proteins. Through a complex signalling pathway, this eventually leads to both metabolic and mitogenic effects (Wilcox, 2005). Mitogenic effects of insulin include phosphorylation of MAPK, promoting the expression of genes required for cell growth and differentiation. Conversely, metabolic effects of insulin involve the phosphatidylinositol 3 kinase (PI3K) pathway, and ultimately lead to translocation of the glucose transporter GLUT4 to the cell membrane (De Meyts, 2000). Collectively, these actions decrease blood glucose levels, as glucose is metabolized intracellularly in peripheral tissues. Thus, tightly controlled and sensed insulin action maintains appropriate glucose fuel utilization and storage in the face of rapidly changing supply and demand. 1.4.1 The role of zinc in the pancreas and insulin secretion Zinc is an essential metal required for many biological processes (Hou et al., 2009; Kelleher et al., 2011). In the pancreas, Zn2+ is required for exocrine and endocrine processes 31 including zymogen granule crystallization, insulin packaging, and as a signalling molecule (Chimienti et al., 2004; Guo et al., 2010; Huber and Gershoff, 1973). Within insulin secretory granules, high concentrations of Zn2+ are required to form crystallized insulin (Baker et al., 1988). Crystallized insulin in insulin secretory granules contains two Zn2+ ions with hexameric insulin. Transport of Zn2+ is performed by specialized zinc transporters, namely Zrt/IRT-like protein (ZIP- Slc39 family) and zinc transporter (ZNT- Slc30 family), which shuttle Zn2+ intracellularly and into vesicles, respectively (Bosco et al., 2010). Studies have demonstrated roles for ZIPs Slc39a6, Slc39a7, and Slc39a8, along with ZNTs Slc30a7 and Slc30a8, in insulin secretion (Bellomo et al., 2011; Syring et al., 2016b). Moreover, increasing concentrations of glucose lead to increases in intracellular Zn2+, Slc39a7, Mt-2, Slc39a6, and Slc39a8 (Bellomo et al., 2011). Zn2+ deficiency can lead to impaired growth and glycemic control, and Zn2+ concentrations are often abnormal in individuals with diabetes (Canfield et al., 1984; Hall et al., 2005; Jou et al., 2010; Kinlaw et al., 1983). In β-cells, Zn2+depletion causes a reduction in the expression of transcription factors required to maintain mature β-cell activity and identity, such as Mafa, Nkx6-1, and Pdx1 (see section 1.5.3), suggesting Zn2+ homeostasis plays a key role in this process (Lawson et al., 2018; Nygaard et al., 2014). Further, emerging evidence suggests that increasing concentrations of Zn2+ in media from stem cell derived β-cells may be a marker of β-cell maturation (Ohta et al., 2019). As a signalling molecule, Zn2+ co-secreted with insulin may suppress glucagon secretion from α-cells and serve other paracrine or autocrine functions (Emdin et al., 1980; Ishihara et al., 2003). Thus, maintaining adequate Zn2+ levels in the pancreas is critical for β-cell function. 32 1.4.2 Heavy metal stress response in the pancreas In the human body, concentrations of Zn2+ are among the highest in the pancreas, reaching millimolar concentrations in insulin secretory granules (Li, 2014). Physiologically, excess amounts of Zn2+ can lead to toxicity (Nriagu, 2011). As such, β-cells must protect themselves from heavy metal toxicity by expressing key genes required to sequester excess Zn2+. This is primarily accomplished by metal sequestering metallothioneins (MTs), which are upregulated as Zn2+ concentrations increase (Thirumoorthy et al., 2011). When stimulated with excess Zn2+, β-cells will induce expression of ZnTs and MTs, which allow for proper Zn2+ storage and sequestering respectively (Nygaard et al., 2014). Since glucose stimulation in β-cells results in increased Zn2+ concentrations, this is also accompanied by increases in Mt1 and Mt2, to counteract potential deleterious effects of free Zn2+ (Bellomo et al., 2011). Further, exposure to other heavy metals such as Cd2+ and Ni2+ impairs GSIS (Dover et al., 2018). MTs have also been implicated in the process of β-cell maturation and protection during T2D induced β-cell stress (Cai, 2004; Chen et al., 2001; Qiu et al., 2017). Therefore, heavy metal stress response is important for β-cell function and protection from damage. The pancreas and islets of Langerhans: Development and function The pancreas has both exocrine and endocrine functions. Comprising approximately 95-98% of total pancreatic mass, exocrine tissues such as acinar cells perform their function via ductal cells. During digestion, these cells will secrete a cascade of digestive enzymes and bicarbonate into the intestine (Husain and Thrower, 2009). Other cell types found in islets include blood vessels, neurons, macrophages, and dendritic cells, all of which serve critical functions (Carrero et al., 2016; Mattsson, 2005). As these cell types are not the main focus of this 33 thesis, non-endocrine functions of the pancreas are not discussed further; for a review, see Pandol, 2010. 1.5.1 Endocrine cell types and function Endocrine cells represent approximately 1% of total pancreatic cells and include α, β, δ, ε, and pancreatic polypeptide (PP) cells contained within the islets of Langerhans. In humans, endocrine cells are found in the following proportions: 50-70% β-cells, 20-30% α-cells, less than 10% δ-cells and PP cells, and less than 1% ε-cells. Interestingly, in mice, the proportion of β to α-cells is slightly higher with 60-80% β-cells, 10-20% α-cells, and 5% δ-cells (Dolenšek et al., 2015). Further, humans and mice display different cellular organization within islets, with no clear organization in humans, whereas mice having α-cells near the mantle (Sharon et al., 2018). Endocrine cells vary in their functions. α-cells produce and secrete glucagon, which serves to increase blood glucose levels (Gaisano et al., 2012). Within the islet, paracrine interactions between β- and α-cells maintain glycemic setpoints, which can be further influenced by proportional numbers of each cell type (Rodriguez-Diaz et al., 2018). Somatostatin-secreting δ-cells have inhibitory paracrine crosstalk functions between β and α-cells (Rorsman and Huising, 2018). Structurally, δ-cells display a ‘neuron-like’ morphology to increase contact with other islet cell types. Moreover, emerging evidence suggests δ and β-cells are electrically coupled via gap-junctions to regulate α-cell activity (Briant et al., 2018). PP cells (also known as F-cells) express and secrete PP, which slows gastric emptying, inhibits exocrine secretions, and acts as a satiety hormone (Asakawa et al., 2003). Interestingly, PP does not affect insulin or glucagon secretion, suggesting PP cells act primarily on digestion. Finally, ε-cells produce and secrete ghrelin, which is found in acetylated and unacetylated forms (Sakata et al., 2019). Studies 34 have demonstrated that ghrelin increases blood glucose by decreasing insulin secretion from β-cells (Dezaki et al., 2004). Furthermore, ghrelin’s paracrine action on β-cells suggests that it may play a protective role and increase β-cell mass (Granata et al., 2010). 1.5.2 Early pancreas, endocrine, and β-cell development Pancreatic and endocrine development is a highly orchestrated event. During endocrine cell differentiation in the mouse pancreas (Figure 2), the expression of the transcription factor PDX1 at mouse embryonic day E8.5 marks the pre-pancreatic endoderm; PDX1 expression is maintained throughout early pancreas development but eventually becomes restricted to the β and δ-cell lineages (DiGruccio et al., 2016; Oliver-Krasinski et al., 2009). Pre-pancreatic endoderm cells give rise to the pancreas with distinct dorsal and ventral regions visible at approximately embryonic day (E) 9-9.5 in mice (Wessells and Cohen, 1967). Traditionally, pancreas development has been viewed as a sequence of two waves of morphogenic events (Pictet et al., 1972). During the primary transition, occurring between E9.5-12.5, tissue budding and evagination produces branch-like epithelial protrusions surrounded by mesenchyme (Zhou et al., 2007). At this stage, early endocrine cells which coexpress glucagon, insulin, and/or PP, are present (Herrera et al., 1991). This tissue is comprised of multipotent pancreatic progenitor cells and can give rise to any cell of the pancreas (Pan and Wright, 2011). Pancreatic epithelial morphogenesis effectively segregates the epithelium into “tip” and “trunk” regions containing exocrine, endocrine, and ductal progenitor cells (Villasenor et al., 2010). Furthermore, mature pancreas size is determined at this stage (Pan and Wright, 2011). At E13.5, endocrine cells found in the trunk region experience a large wave of differentiation, a process known as the secondary transition (Pictet et al., 1972). At the molecular 35 level, this stage is marked by massive increases in differentiation and proliferation, with the endocrine lineage being specified by transient expression of the transcription factor Neurogenin3 Figure 2: Endocrine cell differentiation is driven by transcription factor cascades Temporal expression of specific transcription factors is essential for and drives the formation of all pancreatic cell lineages. Pancreatic progenitors give rise to multipotent tip cells, which can differentiate into bipotent trunk cells. These cells can form ductal cells or endocrine progenitors, which produce β, α, ɣ, PP (F), and ε-cells (illustration created with BioRender.com). 36 (NEUROG3). NEUROG3 is required and sufficient to specify the endocrine lineage (Schwitzgebel et al., 2000). Thus, whole body Neurog3 ablation leads to loss of endocrine cells, whereas ectopic expression of NEUROG3 leads to differentiation towards the endocrine lineage (Gradwohl et al., 2000; Schwitzgebel et al., 2000; Villasenor et al., 2008). These endocrine progenitors express transcription factors such as NEUROD1, PAX4, PAX6, and ARX (Zhou et al., 2007). Interestingly, PAX4 and ARX play antagonistic roles in determining the numbers of differentiated β- and α-cells, respectively (Collombat et al., 2003). The differentiation of the insulin expressing β-cell lineage requires expression of NEUROD1, NKX2-2, NKX6-1, PAX4, and PAX6 (Oliver-Krasinski et al., 2009; Sander et al., 2000; Sosa-Pineda et al., 1997; Sussel et al., 1998; Wang et al., 2004). At birth, nascent β-cells appear to have a high proliferative rate that rapidly decreases over time (Kushner, 2006). Although considered differentiated β-cells, these cells are not yet mature, i.e. not capable of responding to glucose at that time (Grasso et al., 1968). 1.5.3 Functional β-cell maturation The process of β-cell maturation is complex and not fully understood. Functionally, the hallmark and definition of β-cell maturation is acquisition of an appropriate GSIS response, i.e., increased insulin secretion following exposure to high levels of glucose (>10mM) (Rozzo et al., 2009). However, one day post-birth (P1), mouse β-cells display a high rate of insulin secretion with low glucose (2.8-5mM), yet fail to increase insulin secretion when stimulated with high glucose (Blum et al., 2012). This suggests a change in glucose sensing threshold as β-cells undergo maturation. In mice, postnatal events such as weaning have been implicated as triggers for the maturation process (Stolovich-Rain et al., 2015). 37 Molecularly, establishment of appropriate GSIS correlates with expression of the peptide hormone urocortin 3 (Ucn3), with a similar increase observed when stem cell derived β-cells are matured in vivo (Blum et al., 2012). In mature islets, UCN3 is cosecreted with insulin and serves paracrine function on δ-cells, inducing secretion of somatostatin (van der Meulen et al., 2015). Interestingly, islets of individuals with T2D demonstrate a loss of mature β-cells, as evidenced by decreased expression of key factors such as UCN3, MAFA, NKX6-1, PDX1, and others (Lawlor et al., 2017a). Emerging evidence also suggests increases in synaptotagmin 4 (SYT4) during β-cell maturation allow for changes in Ca2+ sensitivity (Huang et al., 2018). Reminiscent of the critical, sequential action of transcription factors in endocrine differentiation, several transcription factors such as MAFA, NKX2-2, NKX6-1, NEUROD1, FOXA2, PDX1, and ERRγ induce the expression of key β-cell maturation genes (Bastidas-Ponce et al., 2017; Gu et al., 2010; Liu and Hebrok, 2017; Nishimura et al., 2015; Taylor et al., 2013). In Mafa, Nkx2-2, and Nkx6-1 β-cell specific knockout mice, functional maturation is lost along with decreases in Slc2a2, Slc30a8, and Ucn3, among other genes (Gutiérrez et al., 2016; Nishimura et al., 2015; Taylor et al., 2013). Similarly, the expression of GLUT2/Slc2a2 and ZNT8/Slc30a8 are impaired in β-cell specific Pparγ, and Errγ knockout mouse models, suggesting that the expression of these transcription factors are critical for mature β-cell function (Rosen et al., 2003; Yoshihara et al., 2016a). Moreover, in humans, mutations in certain transcription factors such as PDX1 and NEUROD1 lead to impaired mature β-cell function and MODY (Steck and Winter, 2011). These studies highlight the importance of transcription factors for the expression of genes required for mature β-cell function. The observation that β-cell transcription factors are required to express maturation genes, which can code for other transcription factors, highlights the importance of gene regulatory 38 networks in this process (Servitja and Ferrer, 2004). As such, genetic susceptibility to T2D can arise from variation in islet regulatory regions such as enhancers (Cebola, 2019). As expected, transcription factors important for β-cell maturation may directly bind such enhancer regions (Sun et al., 2018; Tennant et al., 2013). For example, cooperation between PDX1 and FOXA2 leads to proper maturation through expression of Slc2a2, Mafa, and Ucn3 (Bastidas-Ponce et al., 2017). Other regulatory connections in islets include cooperation between HNF4α, HNF1α, and NKX6-1 (Donelan et al., 2010; Eeckhoute et al., 2004; Servitja and Ferrer, 2004). Thus, transcription factors and regulatory networks play essential roles in β-cell maturation. Transcriptional regulation by epigenetic changes Transcription is the process by which cells use DNA as a template to synthesize mRNA. This involves interactions between transcription factors, RNA polymerase II (RNA Pol II), and coactivators at genomic loci (Thompson et al., 1995). These regions contain regulatory elements such as enhancers, promoters, and silencers, which serve important roles in determining the expression of specific genes (Maston et al., 2006). Short stretches of DNA, approximately 140bp, are wrapped around histones and form structural units known as nucleosomes, which are important for the regulation of gene expression (Weber and Henikoff, 2014). Epigenetic posttranslational modifications on histones become instructive scaffolds to further modulate transcriptional behaviour. Chromatin modifications such as histone 3 lysine 27 acetylation (H3K27Ac), deposited and removed by coactivators, which have histone acetyltransferase (HAT) or histone deacetylase (HDAC) function, serve to change chromatin accessibility. Other chromatin marks such as H3K4 mono- and tri-methylation (H3K4me1 and H3K4me3) serve as recognition sites for different factors and are associated with enhancer and 39 promoter regions, respectively (Bannister and Kouzarides, 2011; Ma et al., 2019). Conversely, H3K27me3 rich regions are associated with closed heterochromatin, which represses transcription. These marks are deposited or removed by coactivators containing histone methyltransferase (HMT) or demethylase (HDM) activity. Additional epigenetic modifications include DNA methylation, where DNA methyltransferases (DNMTs) deposit methyl groups at C5 position of cytosines to create 5-methylcytosine to reduce DNA accessibility. Thus, epigenetic changes play an important role in transcriptional regulation. 1.6.1 Transcriptional regulation during β-cell development and maturation Prior to the formation of pancreatic progenitors, differential repressive H3K27me3 and activating H3K4me3 regions specify pancreas and liver tissue (Xu et al., 2011a). As endocrine cells differentiate, increasing histone acetylation by inhibiting histone deacetylases (HDACs) results in more β- and δ-cells (Haumaitre et al., 2008). Similarly, loss of acetylation by overexpression of HDACs results in fewer endocrine cells (Lenoir et al., 2011). Assay for Transposase-Accessible Chromatin using sequencing (ATAC-seq) experiments have shown similar chromatin accessibility patterns in both α- and β-cells, suggesting important changes in chromatin accessibility during endocrine cell specification (Ackermann et al., 2016). During β-cell differentiation, α-cell genes such as Arx must be silenced to prevent their expression. This is accomplished by recruitment of NKX2-2 to the Arx promoter, leading to methylation by its co-regulator Dnmt3 (Dhawan et al., 2011; Papizan et al., 2011). Emerging evidence suggests changes in chromatin accessibility, allowing expression of genes such as NeuroD1, Nkx6-1, and Foxa2, occur during β-cell maturation (Sobel et al., 2019). Moreover, islet-specific changes in enhancer-promoter looping, which is influenced by chromatin accessibility, occur in T2D 40 (Greenwald et al., 2019). This suggests endocrine and β-cell development require proper transcriptional regulation by coactivators. To date, transcriptional coactivator function is poorly understood in the context of β-cell maturation. Evidence suggests islet specific enhancers, marked by H3K27Ac, H3K4me1, and specific transcription factors exist, but it is unclear which co-activators may be involved in establishing these regions (Tennant et al., 2013). p300/CBP has HAT activity, and studies have demonstrated it has a synergistic role alongside NEUROD1, PDX1, and MAFA in regulating insulin expression (Andrali et al., 2008). Moreover, p300/CBP was recently shown to be essential for β-cell development and maturation (Wong et al., 2018). Specifically, p300/CBP β-cell knockout mice show reductions in genes such as Slc2a2, Ucn3, and Iapp, all of which are important markers of maturation. This suggests H3K27Ac is important during β-cell maturation. Gene silencing via H3K27 trimethylation (H3K27me3) is mediated by the Polycomb repressive complex 2 (PRC2) and is also important during pancreas development. Thus, loss of the PRC2 component EZH2 in pancreatic progenitors leads to more endocrine cells and an increase in mature β-cells (Xu et al., 2014). In contrast, loss of the PRC2 component EED leads to impaired β-cell function and terminal differentiation (Lu et al., 2018), suggesting a nuanced function for PRC2. Interestingly, in T2D islets, chromatin dysfunction mimicking PRC2 deletion was observed, suggesting that altered H3K27me3 may exacerbate pathogenesis (Lu et al., 2018). Similarly, loss of Trithorax group proteins (TrxG), which have HMT activities, in endocrine progenitor cells leads to impaired expression of Slc2a2, Ucn3, and Iapp (Campbell, 2019). Collectively, these studies demonstrate an important role for transcriptional coregulators during β-cell development. 41 Mediator complex Originally described in yeast, Mediator is a transcriptional coactivator required for RNA Pol II dependent transcription (Kelleher et al., 1990). Early literature on Mediator described it as thyroid hormone receptor-associated protein (TRAP), activator-recruited cofactor (ARC), cofactor required for Sp1 (CRSP), and vitamin D receptor-interacting protein (DRIP) (Fondell et al., 1996; Näär et al., 1999; Rachez et al., 1998; Ryu and Tjian, 1999; Ryu et al., 1999). Mediator is composed of 25-31 subunits, several of which represent direct docking sites for transcription factors (Bourbon et al., 2004). Mediator also plays a critical role in the formation of the pre-initiation complex (PIC), which is assembled at promoter regions and include the general transcription factors (GTFs) TFIIA, TFIIB, TFIID, TFIIE, TFIIF, TFIIH, Mediator, and RNA Pol II (Figure 3) (Lee and Young, 2000). Distal transcription factors bound to enhancers interact through Mediator to promote PIC assembly and hence modulate transcription. Thus, Mediator serves as a molecular scaffold for RNA Pol II and GTFs, acting as a molecular bridge. Mediator has multiple molecular functions that contribute to its role as a regulator of gene transcription. Following PIC assembly, the carboxy-terminal domain (CTD) of RNA Pol II must be phosphorylated by TFIIH and Mediator kinase Cdk8 to for allow elongation, effectively breaking contact from Mediator (Poss et al., 2013). RNA Pol II pausing at proximal-promoter regions (approximately 20-60 nucleotides away) serves as an additional level of transcription 42 regulation, which is controlled by Mediator and GTFs such as TFIIH (Li and Price, 2012; Poss et al., 2013; Wong et al., 2014). Besides linking distal transcription factors to RNA Pol II, an additional function of Mediator is to promote enhancer-promoter looping via interactions with cohesin and CCCTC-binding factor (CTCF) (Kagey et al., 2010). Components of Mediator also interact with epigenetic regulators containing HMT activity, such as Protein Arginine Figure 3: Transcriptional machinery and Mediator During transcription, RNA Pol II interacts with general transcription factors (GTFs) and Mediator to form the approximately 4.0MDa preinitiation complex (PIC). Mediator has four distinct modules, namely head, middle, tail and kinase, acting as an interface between RNA Pol II and transcription factors. These are found at regulatory elements such as enhancers (illustration created with BioRender.com). 43 Methyltransferase 5 (PRMT5) (Soutourina, 2018; Tsutsui et al., 2013). Thus, Mediator is involved in several aspects of regulating transcription. 1.7.1 The structure of Mediator and its modules Crystallization and cryo-electron microscopy studies have demonstrated Mediator to be large (approximately 1.4MDa), to have a high degree of flexibility, and are the main binding interface between transcription factors and RNA Pol II (Harper and Taatjes, 2018; Poss et al., 2013). Structurally, each subunit of Mediator can be further classified into four distinct modules: head, middle, tail, and kinase (Figure 4). The head, middle, and tail modules are held together by MED14 – the functional “backbone” of Mediator (Cevher et al., 2014). Among modules, the head and middle (with MED14) constitute the core of Mediator and are essential for transcription as they directly contact RNA Pol II to form the PIC (Malik and Roeder, 2010; Soutourina, 2018). The tail module, composed of MED2, MED3, MED5, MED14, MED15, and MED16, serves regulatory purposes through its interactions with specific transcription factors (Dotson et al., 2000; Jeronimo et al., 2016). Together, the head, middle, and tail modules have activating roles in transcription whereas evidence suggests a negative regulatory role for the kinase module (Soutourina, 2018). This is exemplified by the fact that loss of the kinase module leads to increased transcriptional activity (Knuesel et al., 2009; Van De Peppel et al., 2005; Taatjes et al., 2002a). With each module having different roles in transcription, this highlights the extensive interactions Mediator may be involved in. 44 Figure 4: Mediator is composed of 25-31 subunits which interact with specific transcription factors Mediator is a large multi-protein complex composed of varying numbers of subunits depending on tissue and species. Each distinct Mediator module contains a specified number of subunits, which may interact with different transcription factors (adapted from Yin and Wang, 2014). Mediator’s flexibility is enabled by large amounts of intrinsically disordered regions (IDRs), which lack defined 3D conformational structures, allowing Mediator to change conformation throughout transcriptional processes (Taatjes et al., 2002b). Studies indicate that loss of the flexible hinge domain (MED7/MED21) prevents its association with RNA Pol II (Baumli et al., 2005; Nozawa et al., 2017). As such, context dependent structural changes allow Mediator to contact RNA Pol II and several different transcription factors at once. Importantly, Mediator can change conformation based on its binding partners, and certain subunits can be added or removed without perturbing its core integrity (Allen and Taatjes, 2015; Taatjes et al., 2002a). Thus, specific transcription factor-Mediator subunit interactions can selectively activate distinct signalling pathways in individual tissues. Mediator is present in cells with and without 45 the kinase module, which is linked by MED13 and serves to fine tune Mediator function (Knuesel et al., 2009). Similarly, the tail module performs regulatory functions whereas core modules of Mediator consist of the head and middle, which directly contact RNA Pol II (Jeronimo et al., 2016). Additionally, many Mediator subunits such as MED14 and MED17 have demonstrated extensive interactions with other subunits in different modules (Harper and Taatjes, 2018). This exemplifies interactions within Mediator and with transcription factors can influence its activity on transcription. 1.7.2 Mediator as a modulator of cell lineage specification Many mammalian Mediator subunits play essential roles during various stages of embryonic development, with loss of some Mediator subunits causing embryonic lethality (Soutourina, 2018; Yin and Wang, 2014). In the case of middle module subunit MED1, which interacts with numerous nuclear receptors such as HNF4α and Peroxisome proliferator-activated receptor gamma coactivator 1-α (PGC1α), constitutive null mutant mice die at E11.5 due to hepatic, hematopoietic, and cardiac defects (Chen et al., 2009; Ito et al., 2000; Spitler et al., 2017). Loss of Med31, which is also in the middle module, results in defective cell proliferation, impaired chondrogenesis, and lethality at E16.5 (Risley et al., 2010). In mice lacking kinase module subunit Cdk8, embryonic lethality is observed early, between E2.5-E3.0 (Westerling et al., 2007). This suggests Mediator subunits have requirements at different developmental timepoints. Mediator plays an important role in maintaining embryonic stem cell (ESC) states (Kagey et al., 2010). Through its interactions with OCT4, SOX2, and NANOG, Mediator will bind super-enhancers — regions containing large stretches of enhancers and transcriptional machinery 46 (Kagey et al., 2010; Pott and Lieb, 2015; Whyte et al., 2013). Further, Mediator has been found at cell type specific super-enhancers, suggesting it is required for lineage-specific gene expression (Whyte et al., 2013). For example, kinase module subunit MED12 is required for maintenance of hematopoietic stem cells via a kinase module independent mechanism (Aranda-Orgilles et al., 2016; Borikar and Trowbridge, 2016). During neuronal cell differentiation, MED12, MED13, MED19, and MED26 are required for expression and/or repression of key genes to maintain neuronal states (Ding et al., 2009; Rocha et al., 2010). Moreover, CoIP mass spectrometry experiments using a FLAG-MED15 construct in neuronal stem cells have shown Mediator to bind several factors such as TCF4, SOX2, NFI, JMJD1C, CARM1 and CHD7 at super-enhancer regions (Quevedo et al., 2019a). As such, Mediator subunits influence many development and cell lineage specification pathways, and in some cases the specific transcription factor interactions underlying these roles have been identified. For example, requirements for subunits MED14 and MED1 in adipogenesis likely arise due to loss of PPARγ interactions (Ge et al., 2002, 2008; Grontved et al., 2010). During smooth muscle differentiation, subunits such as MED23 and MED28 have antagonistic roles; MED28 represses smooth muscle cell differentiation whereas MED23 promotes differentiation into smooth muscle (Beyer et al., 2007; Wang et al., 2009). This highlights the importance of Mediator in stem cells and cell lineage specification. 1.7.3 Mediator and human disease Mutations or changes in expression of specific mediator subunit expression are associated with several human diseases, including cancer, neurodevelopmental, cardiovascular, and behavioural disorders (Spaeth et al., 2011). Among the first subunits linked to cancer, MED1 47 interacts with nuclear hormones such as the estrogen receptor subunit α (ERα), with whom it binds to enhancers of estrogen regulated genes (Yang et al., 2018). In hormone-driven cancers such as ERα expressing breast cancers, MED1 expression is increased due to amplification events on chromosome 17q12 (Zhu et al., 1999). Studies in human cancers have implied that many other Mediator subunits have roles in tumourigenesis such as CDK8, MED12, MED19, and MED15 (Brägelmann et al., 2017; Firestein et al., 2008; Lu et al., 2005; Shaikhibrahim et al., 2014; Sun et al., 2011; Syring et al., 2016a; Weber and Garabedian, 2018). Transcriptomic experiments in tumours often reveal improper expression of certain Mediator subunits, which can drive cancer formation. Mediator subunits have also been implicated in developmental disorders such as Charcot-Marie-Tooth disease, postnatal-onset microcephaly, X-linked mental retardation, and Di George syndrome (Soutourina, 2018; Spaeth et al., 2011). These findings highlight the importance of Mediator in human development and disease. 1.7.4 Mediator and metabolism Several Mediator subunits have emerged as important regulators of metabolic processes (Youn et al., 2016). Among all subunits, MED1 is the most prominent as a metabolic regulator due to its interactions with important metabolic regulatory transcription factors such as PPARα, PPARγ, glucocorticoid receptor (GR), CCAAT enhancer binding protein β (C/EBPβ), and PGC1α. Conditional Med1 knockout studies revealed important functions for MED1 in lipid metabolism, hepatic steatosis, skeletal muscle function, and insulin signalling (Jia et al., 2004, 2009, 2014; Li et al., 2008). Tail module subunit MED14 has also been implicated in lipid metabolism, through its interactions with GR and PPARγ (Grontved et al., 2010; Hittelman, 1999). The kinase module subunit MED13 has also emerged as an important factor for 48 cardiovascular and systemic metabolism, with cardiac overexpression leading to improved lipid metabolism (Baskin et al., 2014; Grueter et al., 2012). Moreover, skeletal muscle specific deletion of MED13 improves hepatic steatosis, glucose uptake and utilization, suggesting tissue and therefore context specific function of MED13 (Amoasii et al., 2016). In the context of β-cell biology, few Mediator subunits have been examined. In vitro, the tail module MED25 regulates insulin secretion through HNF4α (Han et al., 2012). However, to date, only two studies have investigated Mediator’s in vivo function in the pancreas or in β-cells, as loss of CDK8 in β-cells leads to improved glucose tolerance and insulin secretion, suggesting that it is a negative regulator of this process (Xue et al., 2019). Previous PhD student Eric Xu characterized MED15 as an important regulator of β-cell development. With respect to early β-cell development, loss of Med15 in all pancreatic cells leads to a reduction in NEUROG3+ endocrine progenitors without affecting acinar cells (Xu, 2016). Additionally, loss of Med15 in endocrine progenitors resulted in fewer β-cells without affecting α-cells (Xu, 2016). When Med15 was deleted specifically in β-cells, no changes in β-cell numbers or insulin secretion was observed, but β-cell function was impaired (Xu, 2016). These observed roles for Mediator in pancreatic cells suggest, in β-cells, specific Mediator subunits play an important role in insulin secretion and warrants further investigation. 1.7.5 Tail module subunit MED15 The tail module of Mediator serves regulatory purposes through its interactions with various transcription factors (Dotson et al., 2000). Within this module, subunit MED15 is glutamine rich and plays an important role in lipid metabolism through its interaction with Sterol regulatory-element binding proteins (SREBPs) (Yang et al., 2006). This function is conserved 49 between species, because in Caenorhabditis elegans, the orthologue of MED15 (MDT-15) interacts with SBP-1 (SREBP) to regulate lipid metabolism and protect against glucotoxicity (Lee et al., 2015; Yang et al., 2006). MDT-15 also interacts with NHR-49, an orthologue of HNF4α/PPARα, to regulate metabolism and protect against oxidative stress, starvation, hypoxia, xenobiotics, heavy metals, and infection with pathogens (Goh et al., 2014, 2018; Hou et al., 2014; Hu et al., 2018; Lee et al., 2019, 2016; Pukkila-Worley et al., 2014; Shomer et al., 2019; Taubert et al., 2006, 2008; Vozdek et al., 2018). As such, this places MDT-15 as an important metabolic regulator in C. elegans. To date, mammalian MED15 has been implicated in TGF-β signalling due to its interactions with SMAD2/3/4 (Kato et al., 2002; Yang et al., 2006; Zhao et al., 2013). Further, these interactions may influence cancer progression, with MED15 being upregulated in certain cancers (Shaikhibrahim et al., 2014). In β-cells, TGF-β plays a role in development and GSIS (Dhawan et al., 2016; Lin et al., 2009; Miralles et al., 1998). Moreover, MED15 and SMAD2/3/4 also interact with a regulatory target of the Hippo pathway TAZ (WWTR1) to maintain stem cell self-renewal (Varelas et al., 2008). Interestingly, deletion of Med15 in mouse T cells resulted in decreased expression of approximately 500 genes and conformational changes in Mediator, but did not affect viability (El Khattabi et al., 2019). This suggests MED15 may play an important role in mammalian development through pathways such as TGF-β signalling. Thesis investigation Few studies of Mediator or MED15 in pancreas development and function have been performed to date (Xu, 2016). Evidence suggests MED15 and its orthologues are important for lipid metabolism, TGF-β/SMAD signalling, oxidative stress resistance, zinc metabolism stress, 50 and hypoxia, all of which are important for pancreas and endocrine cell development and maintenance (Bosco et al., 2010; Dhawan et al., 2016; Gorasia et al., 2015; Nyengaard et al., 2004; Shimano et al., 2007). In mature β-cells, proper expression and function of transcription factors is required (Gu et al., 2010; Sussel et al., 1998; Taylor et al., 2013). As such, we hypothesized that MED15 plays an important role during β-cell development and function through interactions with specific transcription factors at key loci. Because MDT-15 is important for heavy metal stress response (Taubert et al., 2008), which is relevant to the pancreas as it contains among the highest levels of Zn2+, we hypothesized that MED15 is important for heavy metal stress in β-cells. In Chapter 3, we characterize the expression of MED15 in human and mouse β-cells and dissect its role in maturation using a β-cell specific Med15 knockout mouse model. In Chapter 4, we perform genomic analysis for MED15 and MED1 to explore which potential transcription factors or loci MED15 may interact with in β-cells. Finally, in Chapter 5, we use several models such as mouse β-cells, human lung cancer cells, and C. elegans to determine what role MED15/MDT-15 may have in heavy metal stress response. I discuss what role this may play in the context of β-cell maturation and diabetes pathogenesis. 51 Chapter 2: Materials and Methods Animal studies Mouse experiments were approved by the University of British Columbia Animal Care Committee. For generation of Med15 knockout mice, Med15tm1a(KOMP) Wtsi mouse JM8A1.N3 embryonic stem cells were obtained from the trans-NIH Knock-Out Mouse Project (KOMP; www.komp.org). Stem cells were microinjected into B6-Alb 8-cell stage embryos that were implanted into pseudopregnant C57BL/6J females (Mouse Animal Production Service, Centre for Molecular Medicine and Therapeutics). Germline transmission of founders was confirmed by genotyping. Animals were housed under a 12-h light/dark cycle, fed ad libitum with standard chow diet (Lab Diets; 5010). All mouse strains used were on the C57BL/6J background (JAX; 000664). Throughout these studies, males were studied unless otherwise stated. Control mice were littermate Med15fl/fl; Ins1+/+ or Med15+/+; Ins1Cre/+, as indicated, whereas experimental mice were Med15fl/fl; Ins1Cre/+. Human islet studies Human islet studies were approved by the BC Children’s and Women’s Hospital Research Ethics Board. Islets were provided by the Alberta Diabetes Institute Islet Core or University of Alberta Clinical Islet Transplantation Program. Islets were cultured in Connaught Medical Research Laboratories media (CMRL; Gibco #11530037) supplemented with 10% FBS (Gibco LS10082147), 100U/mL penicillin + 100µg/mL streptomycin (Thermo Fisher Scientific SV30010), and 2 mM L-glutamine. 52 Genotyping In brief, mouse ear notch samples were taken using an animal ear punch (VWR; 10806-290) and placed into a labelled 0.2ml flat capped PCR tube (VWR; 93001-120) containing 20% Chelex 100 resin (Biorad; 143-2832), 0.2% Tween-20, and 0.01 mg/ml proteinase K (Roche; 3115836001). Following overnight incubation at 55oC, samples were briefly spun down, and the supernatant was collected for PCR genotyping using appropriate primers (Table 1). Pancreatic islet isolation Mice were euthanized by cervical dislocation and common bile duct injected with 2-5ml of 0.375mg/ml collagenase type XI diluted in Hank’s Balanced Salt Solution (HBSS; 137mM NaCl, 5.4mM KCl, 4.2mM NaH2PO4, 4.1mM KH2PO4, 10mM HEPES, 1mM MgCl2, and 5mM glucose) using a 30G needle. Pancreata were excised and placed in 2ml collagenase in a 50ml tube, incubated for 15 minutes at 37oC, and shaken to release islets from surrounding tissue. HBSS with 1mM CaCl2 was added to stop collagenase activity and centrifuged at 200xg. Supernatant was discarded and the process was repeated twice. Digested pancreas was resuspended in RPMI1640 (Gibco 11875085) with 10% FBS, 100U/mL penicillin +100µg/mL streptomycin, and 2 mM L-glutamine. Resuspended tissue was poured onto a 70µm cell strainer (Corning 352350). Filtered islets were rinsed into a dish using complete RPMI 1640 and picked under a dissecting microscope (Olympus SZX16) with oblique illumination using a P20 micropipette. 53 Cell culture MIN6 mouse insulinoma cells were obtained from Dr. Jun-Ichi Miyazaki (Miyazaki et al., 1990), and A549 lung adenocarcinoma were obtained from American Type Culture Collection (ATCC; CCL-185). MIN6 and A549 cells were maintained in Dulbecco’s Modified Eagle Medium (DMEM; Hyclone # SH30003.03) supplemented with 10% fetal bovine serum (FBS; Gibco #12484028). Cells were passaged for use in experiments between passage 26-35 for MIN6 cells and less than 50 for A549 cells. For cell splitting, adherent cells were washed once with 5 ml per 25cm2 of culture vessel area sterile 1xPBS and 1-2 ml per cm2 0.05% trypsin-EDTA (Gibco; 25200114) was added for 10 minutes. Following inspection for detachment, 2-3x the volume of trypsin of complete DMEM was added to stop digestion and cells were centrifuged for 5 minutes at 200xg. Media was aspirated and cell pellet was resuspended in appropriate volumes of complete DMEM. Cells were plated as needed for expansion of cell lines or subcultured for use in experiments. Transfections and siRNA knockdown For transfection of MED15 expression vector in MIN6 cells, 24 hours prior to transfection cells were split and seeded in a 6-well plate at 0.3x106 cells per well to obtain 60-90% confluency. Lipofectamine 2000 transfection reagent (ThermoFisher; 11668019) was used according to the manufacturer’s protocol. Two micrograms of pCS2+-CMV-MED15 (insert accession # NM_015889) or pEGFP-C1 (GFP control, Addgene #6084) plasmid DNA was diluted in Opti-MEM (ThermoFisher; 31985062) and complexed with Lipofectamine 2000 reagent. DNA-lipid complexes were added to cells, incubated for 5 minutes, and topped up with 3ml of complete DMEM. For siRNA knockdown, MIN6 cells were transfected in a 6-well plate 54 using HiPerFect (Qiagen; 301704) siRNA transfection reagent with 150ng of scrambled (D-001810-0X) or Med15 specific (J-042327-05-0002) siRNAs in 2300µl of serum-free DMEM. Cell lysates were collected 48-hours post transfection. Perifusion GSIS assay Perifusion insulin secretion assay was performed using a Perifusion V2.0.0 system (Biorep). One hundred islets were immobilized in perifusion chambers loaded between Bio-Gel P-4 polyacrylamide beads (Biorep; PERI-BEADS-20). Pumping rate was set at 130 µl/min and the following program was run: 2.8mM glucose for 10 minutes, 25mM glucose for 30 minutes, 2.8mM glucose for 15 minutes, 2.8mM glucose + 10mM ⍺-ketoglutarate for 10 minutes, and 2.8mM glucose for 15 minutes, with all steps in Krebs-Ringer bicarbonate HEPES buffer (KRBH; 114mM NaCl, 20mM HEPES, 4.7mM KCl, 2.5mM CaCl2, 1.2mM KH2PO4, 1.2mM MgSO4, 0.2% BSA, pH 7.4). Ten islets per chamber were removed to assess total insulin and transferred to 500µL of acid ethanol (1M HCl in 70% ethanol in H2O) overnight at 4°C. Remaining islets were lysed in radioimmunoprecipitation assay buffer (RIPA; 150mM NaCl, 1% NP-40, 0.5% Sodium deoxycholate, 0.1% SDS, 50mM pH7.4 Tris), DNA extracted using a DNeasy kit (Qiagen; 69504), and concentration measured using a Qubit 3.0 Fluorometer (ThermoFisher; Q33216). Static incubation glucose-stimulated insulin secretion (GSIS) assay For static incubation, one day prior to experiments, 400 µl of 804G matrix was deposited onto 12-well flat-bottom tissue culture plates. The following day, 804G matrix coated wells were 55 washed three times with sterile water and allowed to briefly air-dry. 1 mL of RPMI 1640 with 10% FBS (Gibco; LS10082147), 100U/mL penicillin, 100µg/mL streptomycin, and 2mM L-glutamine was aliquoted into each well. 30-50 islets were added to each well and incubated overnight at 37oC to allow for adherence to well surface. The following day, islets were washed in 1xPBS and incubated in 500 µL of KRBH supplemented with 2.8mM D-glucose for 1 hour. After preincubation, media was discarded and 500 µL KRBH with 2.8mM D-glucose were added and incubated again for 1 hour for basal secreted insulin levels, and replaced with KRBH supplemented with 16 mM D-glucose or 40 mM KCl for 1 hour for stimulated insulin levels. Finally, 10 islets per well were removed to assess total insulin, removed, and transferred to 500 µL of acid ethanol (1M HCl in 70% ethanol in H2O) and stored at -20°C. The remaining islets were washed with 1xPBS and cell lysate collected in RIPA buffer for DNA concentration as described in previous section. Glucose uptake Glucose uptake assay was performed as described (Yamada et al., 2000). Islets were washed with RPMI1640, dispersed with 0.05% trypsin, and seeded on 804G coated coverslips. The following day, 2.8mM glucose in KRBH without glucose was used to preincubate islet cells followed by either 10 minute incubation or perfusion with 200µM 2-[N-(7-nitrobenz-2-oxa-1, 3-diazol-4-yl)amino]-2deoxy-D-glucose (2-NBDG, Cayman Chemical 186689-07-6) in a Tokai Hit INUBTFP-WSKM stage-top incubator at 37oC. Imaging was performed using confocal microscopy (Leica SP8; Leica Microsystems), background subtracted, and intracellular fluorescence quantified using ImageJ software. 56 Intraperitoneal glucose tolerance tests (IPGTT) and serum insulin quantification For IPGTTs, mice were fasted for 10 hours, weighed, and restrained for blood sample collection using the saphenous vein, 30G needle, and hematocrit capillary tubes (Fisher; 22-362574). Blood glucose levels were measured using a OneTouch UltraMini glucometer (Lifescan; 06720302). D-glucose (40% w/v) was injected at 2 g/kg via peritoneal cavity and blood glucose levels determined at 15, 30, 45, 60, and 90-minute time intervals. Blood was collected at fasting and 10 minutes after intraperitoneal glucose injection for determination of plasma insulin levels using a rodent ELISA kit (Alpco Stellux 80-INSMR-CH10). Blood glucose levels greater than the glucometer detection limit are reported as 33.3 mmol. Mitochondrial function assay 3D morphometric analysis Islets were washed with MEM (Corning; #15-015-CV), dispersed with 0.05% trypsin, and seeded on 804G matrix coated coverslips. The following day, the cells were preincubated for 1 hour with 2.8mM glucose in KRBH. This was followed by 1.5 hours incubation in 2.8 mM glucose, 18 mM glucose or 2.8 mM glucose + 10 mM α-ketoglutarate, with fluorescent 25 nM tetramethylrhodamine ethyl ester (TMRE) (Sigma-Aldrich 87917), 50 nM mitotracker green FM (MTG) (ThermoFisher; M7514) and 0.1µg/mL Hoechst 33342 (Thermo Fisher 62249) added the final 30 min. Cells were imaged in a Tokai Hit INUBTFP-WSKM stage-top incubator at 37oC on a Leica SP8 Confocal Microscope (Concord, Ontario) using a 63xHC Plan Apochromatic water immersion objective (NA=1.2). For quantification of mitochondrial function, TMRE (mitochondrial potential) signal was normalized using MTG (mitochondrial mass) signal. 57 Mitochondrial 3D morphometric analysis For 3D analysis, MTG stained cells were acquired with z-spacing set to optimal Nyquist sampling parameters and deconvolved using Huygens Professional software (Scientific Volume Imaging, Netherlands). Mitochondrial morphological and functional features were extracted and quantified in ImageJ/Fiji using an mitochondria analyzer pipeline and plugin (Chaudhry et al., 2019). Briefly, images were processed using the ImageJ “Subtract Background” and “Sigma Filer Plus” modules, and mitochondria then identified using a local Weighted Mean-based threshold algorithm. 2D morphological descriptors were extracted from the resulting binary image using the “Analyze Particles” command, whereas 3D features were quantified using a combination of the “3D Object Counter” and “Particle Analyzer 3D” commands (part of the MorphoLibJ package) (Legland et al., 2016). RNA extraction and qPCR For MIN6 cells and mouse islets, RNA was extracted using TRIzol reagent (ThermoFisher; 15596026) according to the manufacturer’s protocol. 200 µl of chloroform were added to each sample for every 1ml of TRIzol, briefly vortexed, and centrifuged at 12,000xg for 10 minutes. The aqueous top layer was pipetted into a 1.5 ml tube and 500 µl of 100% isopropanol with 2 µl of glycogen (20mg/ml) was added followed by centrifugation at 12,000xg for 5 minutes. The pellet was washed twice with 70% ethanol and allowed to air dry for 10 minutes at room temperature. Dried pellets containing nucleic acids were resuspended in dimethyl pyrocarbonate (DEPC; Sigma-Aldrich D5758) treated water and treated with TURBO 58 DNA-free DNAse kit (ThermoFisher; AM1907) to remove DNA as per manufacturer’s instructions. RNA was quantified using a NanoDrop 2000 Spectrophotometer (ThermoFisher; ND-2000) and converted into cDNA for gene expression analysis by real-time PCR. For cDNA synthesis, Superscript III Reverse Transcriptase (ThermoFisher; 18080044) was used according to manufacturer’s protocol. All qRT-PCR reactions were performed as triplicates using a ViiA 7 Real-Time PCR System (Thermo Fisher Scientific) using 40 ng of cDNA in a 384-well plate using fast run settings. Taqman primer sequences are listed in Table 2 and SYBR green in 3. RNA-sequencing Total RNA was extracted from islets and rRNA depleted using a RiboGone™ Mammalian-Low Input Ribosomal RNA Removal Kit (Takara Bio USA 634847). Following rRNA depletion, size selection was performed using Agencourt AMPure XP SPRI beads (Beckman Coulter A63881) with a DynaMag-96 side magnetic particle concentrator (Thermo Fisher Scientific 12331D). Quality control analysis using an Agilent RNA Pico 6000 kit with an Agilent 2100 Bioanalyzer (Agilent Technologies 5067-1511) was performed after rRNA depletion. Library preparation was performed using a SMARTer Stranded RNA-seq Kit (Takara Bio USA 634839). Adapter-ligated cDNAs were quantified by Qubit® 3.0 dsDNA high-sensitivity assay kit (Thermo Fisher Scientific Q32854) together with qPCR amplification using standard curves of a known concentration of adapter-ligated libraries. Sequencing was performed on the NextSeq 500 (Illumina, San Diego, CA, USA) with the High Output Reagent Cartridge v2 150 cycles (75 bp x 2). 59 ChIP-seq To perform chromatin immunoprecipitation sequencing (ChIP-seq), MIN6 cells were grown to 2x107 cells per 15cm plate (~80% confluency). One plate of cells (per experiment) was crosslinked with 3% paraformaldehyde for 10 minutes and nuclei were isolated by centrifugation at 200xg at 4oC in Farnham lysis buffer (5 mM PIPES pH 8.0, 85 mM KCl, 0.5% NP-40, and 1x protease inhibitor cocktail; Roche 04 693 159 001). For mouse islets, 1,000 islets were passed through a 30G needle prior to centrifugation for optimal lysis. Pelleted nuclei were resuspended in RIPA buffer and fragmented using an M220 focused ultrasonicator with milliTUBE 1ml with AFA fiber (Covaris; 520130) using the following settings: peak incident power 75W, duty factor 20%, cycles per burst 200, treatment time 410s. Chromatin immunoprecipitation was carried out as described (Johnson et al., 2007) using MED15 antibodies (ProteinTech; 11566-AP) or MED1 antibodies (Bethyl; A300-793A). In brief, 200µ1 of resuspended M-280 sheep anti-rabbit magnetic dynabeads (ThermoFisher; 11201D) were washed with 1xPBS/BSA and incubated with 5ug of antibody at 4oC with rotation for 2-4 hours. Antibody-bead complexes were then washed, resuspended with 100µl of 1xPBS/BSA, and added to chromatin for overnight incubation at 4oC with rotation. Antibody-bead-DNA complexes were then washed with LiCl IP wash buffer (100mM Tris pH 7.5, 500mM LiCl, 1% NP-40, and 1% sodium deoxycholate), eluted with IP elution buffer (1% SDS, 0.1M NaHCO3), and crosslinking was reversed by incubation at 65oC. The chromatin was then digested with proteinase K (Roche; 3115836001), and DNA was extracted using a QIAquick PCR Purification Kit (Qiagen 28104). ChIP-seq libraries were constructed using a ThruPLEX DNA-seq Kit (R400427; Clontech) according to the manufacturer’s protocol. Libraries were sequenced on an Illumina NextSeq500 machine at a 1x76 length for islets and 2x76 for MIN6 cells. 60 ChIP-qPCR Chromatin immunoprecipitation (ChIP) was performed as described above in A549 and MIN6 cells with minor modifications. Briefly, 2x107 cells were grown and incubated for 24 hours with or without 50µM CdCl2 and 100µM ZnSO4 (respectively). The cells were then crosslinked by adding paraformaldehyde to a final concentration of 3% for 10 minutes at room temperature. Immunoprecipitation was performed using MED15 antibody (ProteinTech, 11566-1-AP) as described above. Crosslinking was then reversed by overnight incubation at 65oC in IP elution buffer and DNA purified using QIAquick PCR purification column (Qiagen, 28104). Immunoprecipitated DNA was then quantified via Qubit (ThermoFisher, Q32854) and analyzed by qPCR for appropriate genomic regulatory loci and controls. Primer sequences are listed in Table 4. Bioinformatics analysis of RNA-seq and ChIP-seq For RNA-seq analysis, adapter trimmed FASTQ files were obtained from BaseSpace (Illumina) and each sample’s sequencing lanes were concatenated. ENSEMBL Mus musculus genome GRCm38p.4 FASTQ and GTF annotation file were used for alignments. STAR v2.4.0.1 (Dobin et al., 2013) was used to index genomic FASTQ files, align sample transcriptomes against the genome, and output as sorted by coordinate BAM files. Counts were generated from BAM files with HTSeq (Anders et al., 2015) against annotated genes from GRCm38p.4. DESeq2 (Love et al., 2014) with an FPKM criteria of ≥5 in two or more samples was used to determine statistical significance between differentially expressed genes, applying a Wald test on P values from genes passing filtering. Data was adjusted for multiple testing via Benjamini and Hochberg 61 procedure. For ChIP-seq analysis, fastq files were concatenated and alignment was performed using BWA and filtered to remove PCR duplicates. For MIN6 cells, peaks were called using MACS2 with a q value of 0.01 (Zhang et al., 2008) and BigWig files generated using deeptools2 (Ramírez et al., 2016). For islet data, peaks were called using EaSeq find peak adaptive threshold function on CTRL vs IM15KO (for MED15) and CTRL vs input (for MED1) (Lerdrup et al., 2016). Chromatin segmentation analysis ChIP-seq data from islet chromatin marks were obtained from NCBI SRA (SRP132909 and SRA008281) (Lu et al., 2018; Tennant et al., 2013) and mapped to mm10 mouse genome using Bowtie2 (Langmead and Salzberg, 2012). Duplicate reads were removed and resulting aligned files analyzed using ChromHMM with a k-value of 22 (Ernst and Kellis, 2017). The percent overlap of chromatin state with MED15 and MED1 called peaks was performed using bedtools 2.28.0 intersect (Quinlan and Hall, 2010). Genomic annotation for regions in chromatin states was performed using GREAT basal + extension algorithm (McLean et al., 2010) and compared to RNA-seq regulated genes from Med15 knockout islets. Tissue processing For mouse pancreata, following cardiac perfusion with 4% paraformaldehyde (PFA), tissue was place into Micromesh Loose Biopsy Cassettes (ThermoFisher; B1000735PK) and dehydrated in 50% ethanol for 24 hours followed by 70% ethanol for 24 hours, both steps at 4oC. Tissue was then placed twice in 95% ethanol for 30 minutes, three times in 100% ethanol for 30 62 minutes, twice in xylene for 30 minutes, and embedded in paraffin twice for one hour. For stem cell derived spheroids, approximately 20-30 spheroids were embedded in 1% agarose and dehydrated as described above. Tissue was sectioned using a Shandon Finesse microtome (ThermoFisher; 14047598) with R35 microtome blades (Feather; 207500005) at 5µm intervals. Sections were collected onto prelabelled 75x25mm glass slides (ThermoFisher; ER2951600621T) and allowed to dry overnight at room temperature. Immunofluorescence Tissue was deparaffinized and rehydrated for immunofluorescence as such: twice for 5 minutes in xylene, three times for 2 minutes in 100% ethanol, twice for 2 minutes 95% ethanol, twice for 2 minutes in 70% ethanol, twice for 2 minutes in 50% ethanol, twice for 2 minutes in H2O. Antigen detection was performed using 0.05% citraconic anhydride (Sigma-Aldrich; 125318-25G) for 20 minutes at 95oC, followed by one wash in H2O and three washed in 1xPBS. Tissue was circled with a Super PAP Pen hydrophobic barrier pen (Cedarlane MU22) and blocked using 5% horse serum (ThermoFisher; 16050130) diluted in 1xPBS for 30 minutes. Primary antibodies were diluted in 5% horse serum and incubated at 4oC overnight. Slides were washed three times for 5 minutes in 1xPBS and incubated for 1-2 hours with secondary antibodies at room temperature in the dark. Mounting was performed using SlowFade Antifade Reagent (ThermoFisher; S36972) and slides were scanned on a confocal microscope (Leica SP8; Leica Microsystems). Antibodies used are listed in Table 5. 63 Zinc staining with dithizone and FluoZin-3 50 milligrams of dithizone (Sigma-Aldrich; 43820) was dissolved in 5 ml of DMSO and filtered. Ten microlitres of stock solution was added to 1ml of media containing whole islets and incubated at 37oC for 15 minutes. Islets were washed three times with fresh RPMI 1640 and imaged with a dissecting microscope (Olympus SZX16). FluoZin-3 acetoxymethyl ester (Molecular Probes; F24195) was reconstituted in dimethylsulfoxide (DMSO) to a 1 mM stock solution. For dispersed islets on 804G coated coverslips, FluoZin-3 stock was diluted to 5 µM in RPMI 1640 and incubated for 1 hour at 37oC. Coverslips were washed three times with fresh media and imaged. For staining worms, FluoZin-3 was diluted in M9 buffer (22 mM KH2PO4, 42 mM Na2HPO4, and 86 mM NaCl) to a final concentration of 30 µM and dispensed on NAMM plates. Synchronized L4 stage worms were transferred from NGM-lite plates to NAMM + FluoZin-3 plates and incubated at 20oC for 16 hrs. Worms were then transferred to NGM-lite plates without FluoZin-3 for 30 min to reduce excess FluoZin-3 signal from the intestinal lumen before imaging. Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) Cells or islets were rinsed with 1xPBS, lysed in appropriate volumes of 95 °C NRSB-minus lysis buffer (62.5mM Tris pH 6.8, 1mM Na3VO4, 1mM NaF, 2% SDS, 10% glycerol), spun down (2,000xg for 10 minutes) and supernatant collected. Pierce BCA Protein Assay Kit (ThermoFisher; 23225) was used to quantify protein concentration from cell lysates made by and samples were denatured in 1x SDS loading dye (0.1% 2-Mercaptoethanol, 0.0005% Bromophenol blue, 10% Glycerol, 2% SDS, and 63 mM Tris-HCl pH 6.8) by boiling for 10 minutes. SDS-PAGE (10-15%) 1.5mm 10-well gels were cast using the Mini-PROTEAN gel 64 casting system (Bio-Rad; 1658006FC) as required based on protein size. Gels were placed in a Mini-PROTEAN tetra cell (Bio-Rad; 1658004EDU) and topped up with 1x running buffer (192mM Glycine, 25mM Tris base, 1% SDS, in ddH2O). Protein (30-50µg) was loaded and run alongside BLUelf Prestained Protein Ladder (GeneDireX; PM008-0500) as required based on protein size. Antibodies used are listed in Table 5. Western blotting Polyvinylidene difluoride (PVDF) membranes were soaked in 100% Methanol for 2 minutes and placed in 1x transfer buffer (192mM Glycine, 25mM Tris base, 20% Methanol, in ddH2O). Gels were places in 1x transfer buffer and assembled with PVDF membranes using a Criterion Blotter (Bio-Rad; 1704070) in accordance to manufacturer’s instructions. Transfer was performed at 4oC 70V for 2 hours or 4oC 10V overnight depending on protein size. Following transfer, membrane was rinsed in 1x TBS-T (150 mM NaCl, 50 mM Tris pH 7.5, 0.1 % Tween-20, in ddH2O) and stained with Ponceau S (Sigma-Aldrich; P3504-10G) staining solution (0.1% Ponceau S in 5% acetic acid in ddH2O) to assess protein transfer. Membranes were incubated at room temperature for 30 minutes with 5% milk powder in TBS-T and probed using primary required antibodies overnight at 4oC. Following 3x 15-minute washes in TBS-T, the membrane was probed with secondary antibodies (Jackson ImmunoResearch; 115-035-003, 111-035-144) for 2 hours and developed using Luminata Crescendo Western HRP substrate (EMD Millipore; WBLUR0500) and exposed using Bioflex MSI Film (Mandel Scientific; MED-CLMS810). Antibodies used are listed in Table 5. 65 Co-immunoprecipitation Co-immunoprecipitation (CoIP) was carried out as described (Sabatini et al., 2018) using MED15 (ProteinTech; 1156-1-AP), NKX6-1 (Cell Signaling Technologies; D804R), NEUROD1 (Cell Signaling Technologies; D35G2), Chromogranin A (Cell Signaling Technologies; 60893S) or IgG (Novus Biologics; AB-108-C). In brief, MIN6 cells were harvested with CoIP lysis buffer (150 mM NaCl, 20 mM Tris pH 7.4, 1 mM EDTA, 0.5 mM EGTA, 5% glycerol, 0.5% NP-40, 1 mM Na3VO4, 50 mM NaF, 50 mM β-glycerophosphate, 1x Roche protease inhibitor), insoluble material spun down, and supernatant collected. Protein concentration was determined via BCA assay (Thermo Fisher, 23225), 10% of each lysate was set aside as input, and 2 µg antibody added to 300 µg lysate. Tubes were rotated overnight at 4oC, 2 mg of magnetic Dynabeads added, and bead-antigen-antibody complexes washes three times with CoIP lysis buffer and separated. Laemmli loading buffer (4% SDS, 20% glycerol, 0.004% bromophenol blue, 0.125M Tris-HCl pH 6.8, and 10% 2-mercaptoethanol) was added to beads in CoIP lysis buffer, boiled for 10 minutes and western blot performed for target proteins. Antibodies used are listed in Table 5. Insulin ELISA Chemiluminescent ELISA kits for rodent insulin (ALPCO STELLUX; 80-INSMR-CH10) were utilized according to manufacturer’s instructions. For static incubation GSIS assays, basal samples were used directly; 16 mM glucose incubation media were diluted 5x, 40 mM KCl incubation media were diluted 20x, and acid ethanol total insulin samples were diluted 200x. For perifusion assays, all samples were loaded directly without dilution except total insulin, which 66 was diluted 200x. Five microlitres per well of samples and standards were loaded with 75µL conjugate buffer and incubated for 2 hours on an orbital shaker at 2xg with insulin-antibody conjugate. Plates were each washed six times with wash buffer and 100µL substrate and read on a POLARstar Omega microplate reader (BMG Labtech, Guelph, ON, Canada) or SpectramaxL luminometer (Molecular Devices, San Jose, CA, USA) with 1 second integration time per well. TEM imaging and analysis Freshly isolated mouse islets were placed in 2% glutaraldehyde in PBS for fixation in preparation for transmission electron microscopy. The fixed islets were sent to the Electron Microscopy Facility in the Health Science Centre (McMaster University) and imaged using a JEOL 1200EX TEMSCAN transmission electron microscope. For insulin secretory granule quantification, approximately 15-20 β-cells per genotype were analyzed on 10 random planes at 3000x magnification. CellProfiler 3.0.0 was used to quantify granule counts and size of halo around dense cores (McQuin et al., 2018). A customized pipeline was developed as follows: brightness was normalized across all images, β-cells were identified using the “IdentifyObjectsManually” function based on granule halo morphology (Rorsman and Renström, 2003), dense core of granules were identified using “IdentifyPrimaryObject” function, and halo size identified as automatic expansion of dense core using “IdentifySecondaryObject” function. Granules were quantified as count per β-cell area (normalized count) and halo area per cell area (granule size). 67 C. elegans strains and growth conditions C. elegans strains were maintained on nematode growth medium (NGM)-lite (0.2% NaCl, 0.4% tryptone, 0.3% KH2PO4, 0.05% K2HPO4) agar plates, supplemented with 5 µg/mL cholesterol at 20°C. All strains used in this study are listed in Table 6. NGM-lite plates were used unless otherwise indicated for all experiments. Plates were seeded with Escherichia coli OP50 except for RNAi, for which HT115 was used. Zinc (ZnSO4) and cadmium (CdCl2) were supplemented in noble agar minimal media (NAMM) and NGM plates, respectively, at indicated concentrations. For all analyses, worms were synchronized by standard sodium hypochlorite treatment and starvation of isolated eggs on unseeded NGM-lite plates; the next day, synchronized L1 larvae were collected, placed on seeded plates at the desired densities, and grown until being harvested at the desired developmental stage, as indicated. C. elegans zinc storage granule analysis Prior to imaging, FluoZin-3 stained worms were transferred onto slides containing a drop of solidified 2% (w/v) agarose containing 15µM NaN3 (Sigma-Aldrich; S2002) to induce paralysis. A Leica SP8 confocal microscope with Leica LAS X software was used to acquire z-stacks of worm intestinal cells at intervals of 0.7µm across each worm. Quantification of 3D zinc granules in intestinal cells of the gut was performed using ImageJ (Schneider et al., 2012). Background fluorescence from non-intestinal cells and intestinal lumen was manually subtracted from each micrograph to improve automatic quantification of granules. Background was subtracted within intestinal cells and sections were smoothed with the “Sigma Filter Plus” (edge-preserving noise reduction) function. Images were adjusted with Auto threshold of “MaxEntropy” and Auto local threshold of the mean to improve automatic granule recognition 68 regardless of signal. Zinc storage granules where then counted automatically using the “3D Objects Counter” function. Statistics All data are represented as mean ± standard error of the mean (SEM) unless otherwise stated. For statistical analyses, Prism 6 or 7 software was used (GraphPad Software, La Jolla, CA, USA). Two-tailed unpaired Student’s t-tests, Mann-Whitney U-tests, one-way ANOVA with post-hoc Tukey tests, Kruskal-Wallis H tests with post-hoc Dunn’s tests, or multiple t-tests with Holm-Sidak multiple comparison correction were performed as appropriate (unless otherwise specified), with p≤0.05 considered significant. Table 1: Genotyping primers Target & Amplicon Primer 1 Primer 2 Primer 3 Ins1+/+ - 488bp Ins1Cre/+ - 675bp GGAAGCAGAATTCCAGATACTTG GTCAAACAGCATCTTTGTGGTC GCTGGAAGATGGCGATTAGC Med15+/+ - 321bp Med15fl/+ - 262bp TAAGGTGCTGTGTGTGGTTGGC GTAATGATGAGGCTAAAAGGGCTG N/A 69 Table 2: Taqman qPCR primers Gene Probe Primer 1 Primer 2 Dye Gusb TCTAGCTGGAAAT GTTCACTGCCCTG CACCCCTACCAC TTACATCG ACTTTGCCACCC TCATCC FAM/ZEN/IAB KFQ Mafa ACTTCTCGCTCTCCAGAATGTGCC AGTCGTGCCGCTTCAAG CGCCAACTTCTCGTATTTCTCC FAM/ZEN/IAB KFQ Ucn3 AATGAAGCAAGTTGCAAGGGCAGG CATTGCTTCTCGGCTTACCT CTCCTCTAGTCACCTCCTTCTT FAM/ZEN/IAB KFQ Slc2a2 TGGACAGAAGAGCAGTAGCAGACAC CACATCCTACTTGGCCTATCTG CTTTGCCCTGACTTCCTCTT FAM/ZEN/IAB KFQ Nkx6.1 ACTTCGGAGAATGAGGAGGATGACGA CAGGACCAAGTGGAGAAAGAAG GGGTCCAGAGGTTTGTTGTAAT FAM/ZEN/IAB KFQ 70 Table 3: Conventional qPCR primers for expression Gene Primer 1 Primer 2 18S rRNA (human) GCCGCTAGAGGTGAAATTCTTG CTTTCGCTCTGGTCCGTCTT GAPDH (human) GGCCTCCAAGGAGTAAGACC AGGGGAGATTCAGTGTGTG GUSB (human) GAAAATATGTGGTTGGAGAGCTCATT CCGAGTGAAGATCCCCTTTTTA MT1X (human) ACCACGCTTTTCATCTGTCC GAGCAGTTGGGGTCCATTTC MT2A (human) AACCTGTCCCGACTCTAG GAAGTCGCGTTCTTTACA 18s rRNA (mouse) AGTCCCTGCCCTTTGTACACA CGATCCGAGGGCCTCACTA Mt1 (mouse) CCTTCTCCTCACTTACTCCGTAGC GGAGCCGCCGGTGGA Mt2 (mouse) TCCTGTGCCTCCGATGGAT TGCAGGAAGTACATTTGCATTGT Slc30a8 (mouse) TGCACAGTCTACACATCTGGTCACT TGGCTGGCAGCTGTAGCA Med15 (mouse) TCTTCCAACCAAACAGCAGG TGGTTGAAGACAGGTGAACG 71 Table 4: Conventional qPCR primers for ChIP Gene Primer 1 Primer 2 MT2A promoter (human) GGTGGTCAAGAGGTGTTTACTT GACTCTTGGATTGGTGTCTCTG MT2A (-10kb) (human) TGCGGAACTGTGAGTCAATTA AGAGGAGAGAGTGAGCAAGT MT1X promoter (human) CGAGGTGGAGCCAAAGG GCAAGGAGAAGCAGGAGTT MT1X (-10kb) (human) TTCTCTTCTCGCTTGGGAAC AAGCAGCGGAGGAAGTAAAG Mt1 promoter (mouse) CGGACTCGTCCAACGACTATAA CGCCAACTAAAGGTGCCTATTC Mt1 (+10kb) (mouse) CCAGAACAACCCAGTCCTAAA AGGGTCTTGCTCTCCAGATA Slc30a8 enhancer (mouse) GAAACAGGGCAGGTACTCAA TTACAGCGAGAGGCAAGTAAG Slc30a8 (-10kb) (mouse) AAGATGTGGCTCAGTGGTAAA GCCTGTCTCATGGGTCTTATT Iapp enhancer (mouse) TCCTCAGTCCAGCGGTAATA CAATCGTCCTCTCTTCCTGTTC Iapp promoter (mouse) TGGCTATGACTTCTCCATTTCC GTGGGCCATCAACACATTAAC Lep promoter (mouse) GGAACATGCATCCGCTAGAA GCAGAGGTGACTTCCAGATTAC 72 Table 5: Antibody list Target Host Species Catalogue Number Usage and Concentration MED15 Rabbit ProteinTech (11566-1-AP) IF (1:50), WB (1:100), CoIP (2µg), and ChIP (5µg) UCN3 Rabbit Gift from Mark Huising (#7218) (van der Meulen et al., 2015) IF (1:10000) GLUT2 Goat Santa-Cruz (C-19) IF (1:50) ZNT8 Rabbit Gift from Howard Davidson (Wenzlau et al., 2011) IF (1:1000) PDX1 Mouse DSHB (F6A11) IF (1:250) NKX6-1 Rabbit Cell Signaling Technologies (D804R) CoIP (2µg) NKX6-1 Mouse DSHB (F55A10) CoIP (2µg) MED1 Rabbit Bethyl (A300-793A) ChIP (5µg) NEUROD1 Rabbit Cell Signaling Technologies (D35G2) CoIP (2µg) MAFA Rabbit Bethyl (A300-611A) IF (1:100) β-Tubulin Mouse Abcam (ab6046) WB (1:1000) IgG control Rabbit Novus Biologics (AB-108-C) CoIP (2µg) CHGA Rabbit Cell Signaling Technologies (60893S) CoIP (2µg) β-Actin Mouse Abcam (ab8227) WB (1:1000) INS Guinea Pig Dako (A0564) IF (1:500) 73 Target Host Species Catalogue Number Usage and Concentration GCG Mouse Sigma-Aldrich (G2654) IF (1:1000) SST Rat Millipore (MAB354) IF (1:100) 74 Table 6: List of worm strains Strain Strain name Reference Wild type N2 (Brenner, 1974) mdt-15(tm2182) III XA7702 (Taubert et al., 2008) cdk-8(tm1238) I XA7703 (Grants et al., 2016) hizr-1(am286) X WU1563 (Warnhoff et al., 2017) 75 Chapter 3: MED15 is Required for β-Cell Maturation Introduction Correct spatiotemporal expression of specific gene programs is critical for normal β-cell development and function in both animal models and the human population (Gradwohl et al., 2000; Steck and Winter, 2011; Zhou et al., 2007). In particular, mutations in some transcription factors lead to monogenic forms of diabetes in humans, known as maturity onset diabetes of the young (MODY), demonstrating the importance of proper transcriptional regulation for β-cell development and function (Steck and Winter, 2011). Transcription factors such as MAFA, PDX1, NKX2-2, NKX6-1, ERRγ, and NEUROD1 are regulated at the expression level and via protein-protein interactions at parturition to drive expression of the β-cell maturation program (Gu et al., 2010; Nishimura et al., 2015; Oliver-Krasinski et al., 2009; Stolovich-Rain et al., 2015; Sussel et al., 1998; Taylor et al., 2013; Yoshihara et al., 2016a). However, our understanding of the transcriptional mechanisms that drive β-cell maturation remains incomplete. Transcription factors exert their actions through RNA Pol II. Thus, interactions that bridge transcription factors to RNA Pol II play a critical role in transcriptional regulation. In eukaryotes, an essential link between transcription factors and RNA Pol II is the transcriptional coactivator Mediator (Poss et al., 2013). Functionally, conformational changes allow Mediator to drive specific gene programs, and depending on the exact protein-protein interactions, allow engagement at many different promoters (Haberle et al., 2019; Taatjes, 2010). Mediator is composed of four distinct modules, the head, middle, tail, and kinase modules (Allen and Taatjes, 2015; Tsai et al., 2014, 2017). Structure-function analyses in yeast have determined that the middle and head modules are essential for transcription, whereas the kinase and tail modules serve regulatory functions (Plaschka et al., 2015; Soutourina, 2018). The tail module acts as a 76 direct docking site for transcription factors; loss of tail module subunits causes reduced expression of specific genes without affecting viability in yeast (Ansari and Morse, 2013; Kemmeren et al., 2014). However, the function of most Mediator subunits in higher eukaryote development and physiology is not well understood. The tail module subunit MED15 is a regulator of lipid metabolism through its interaction with SREBP (Yang et al., 2006), which is important for GSIS (Diraison et al., 2008; Shimano et al., 2007). The MED15–SREBP functional interaction is conserved in C. elegans, with orthologs MDT-15 and SBP-1 protecting against glucotoxicity (Lee et al., 2015). Furthermore, C. elegans MDT-15 protects from hypoxia and oxidative stress through interactions with transcription factors NHR-49 and SKN-1, orthologs of HNF4⍺ and NRF2 (Goh et al., 2014, 2018; Vozdek et al., 2018). Notably, hypoxia and oxidative stress also occur in β-cells of patients with diabetes (Gerber and Rutter, 2017; Nyengaard et al., 2004; Park and Woo, 2019). Moreover, during vertebrate development and in certain human cancers, MED15 is required for TGF-β signalling through its interactions with the transcription factors SMAD2/3/4 (Kato et al., 2002; Yang et al., 2006; Zhao et al., 2013). In rodent β-cells, TGF-β signalling is important both during development and for GSIS (Brown and Schneyer, 2010; Dhawan et al., 2016; El-Gohary et al., 2013; Velazco-Cruz et al., 2019; Wu et al., 2014). Because MED15 orthologs participate in these processes and interact with molecular partners relevant to β-cell development and maturation, we hypothesized that MED15 may act as a nexus that couples β-cell specific transcription factors to core promoters, thus promoting β-cell maturation and function. In the present study, we demonstrate that MED15 is expressed in nascent and mature β-cells and that its expression is impaired in islets of individuals with T2D. In mice, deletion of Med15 in β-cells resulted in severe glucose intolerance due to impaired β-77 cell maturation. These data place MED15 as a critical β-cell coactivator required for maturation and function. Results 3.2.1 MED15 is expressed in mature insulin secreting β-cells Few studies of Mediator and its subunits in the developing and adult pancreas exist. To determine the expression and biological relevance of Med15 in pancreatic islets, we performed analysis of bulk RNA-seq data in healthy 8-week C57Bl/6 mouse pancreatic islets and characterized the expression of Mediator subunits. Mediator subunits were found to be expressed in islets at varying levels (Figure 5A-D). Excluding the kinase module, Med15 expression was found to be higher relative to most other subunits, along with Med14, Med25, and middle module subunit Med1 (Figure 5A, B). When considering the dissociable kinase module, Med13 and Cdk8 were found to be highly expressed relative to all other subunits (Figure 5D). Previous findings in embryonic mouse pancreata indicate dynamic expression of Med15 throughout development, with the highest expression found in 8-week adult islets (Xu, 2016). Although mRNA and protein expression often do not correlate (Koussounadis et al., 2015), these data suggest expression of many Mediator subunits in adult islets and a potential relevance for MED15 in the pancreas. As bulk RNA-seq analyses on whole islets allow insight into endocrine cell mRNA expression but not by cell type, we next explored a publicly available single cell RNA-seq dataset from human pancreatic cells (Segerstolpe et al., 2016). To determine if MED15 is coexpressed with β-cell maturation transcription factors, we analyzed the five endocrine cell clusters (Figure 6A) with regard to MED15 expression (Figure 6B). Compared to MED15, 78 middle module subunit MED1 had lower expression in endocrine and β-cells (Figure 6C). Within the β-cell cluster, the expression of the β-cell maturation transcription factors NKX6-1, MAFA, AMed15Med14Med25Med23Med24Med16Med290100200300400Expression (RPKM)BMed6Med28Med22Med8Med17Med19Med20Med30Med11Med180100200300400500Expression (RPKM)CMed13Cdk8CcncMed12050010001500Expression (RPKM)DMed1Med10Med7Med21Med4Med31Med9Med27Med260100200300400Expression (RPKM)Tail Module Middle ModuleHead Module Kinase ModuleFigure 5: Mediator subunits are broadly expressed at varying levels in mouse pancreatic islets. Expression of Mediator (A) tail module, (B) middle module, (C) head module, and (D) kinase module subunits in 8-week mouse pancreatic islets, as determined by bulk RNA-seq expressed as Reads Per Kilobase of transcript, per Million mapped reads (RPKM). 79 and PDX1 overlapped with MED15 expression (Figure 6B, D-F). These data suggest that MED15 is expressed in mature insulin secreting β-cells in humans. In addition, MED15 was also expressed at similar levels in a-, d-, and g-cells, suggests it functions in these cell types as well. Figure 6: β-cell maturation transcription factors are coexpressed with MED15 in human β-cells. Data mining of human islet single cell RNA-seq (Segerstolpe et al., 2016) with (A) t-Distributed Stochastic Neighbor Embedding (t-SNE) plot showing clusters separated by endocrine cell type. Expression of (B) MED15 (circle represents β-cell cluster), (C) MED1, (D) NKX6-1, (E) MAFA, and (F) PDX1 in each endocrine cell cluster. To confirm protein expression of MED15 in insulin producing β-cells and pancreatic islets, we performed western blots for MED15 in these tissues. We found expression of MED15 in glucose responsive cells such as mouse insulinoma cells (MIN6), mouse islets, and human islets (Figure 7). Additionally, stage 6, day 11 (S6D11) stem cell-derived β-cell spheroids, which produce insulin, also robustly expressed MED15 (Figure 7A). Lastly, NanoString gene expression analysis for MED15 expression in islets from healthy and T2D human donors was performed by Dr. Shugo Sasaki. Notably, MED15 expression was significantly decreased in MED15NKX6-1 MAFAMED1A BEDCF PDX180 islets from T2D donors compared to islets from healthy donors (Figure 7B). Overall, MED15 expression is conserved across human and mouse insulin-producing cells, with its impaired expression in T2D, suggesting a role for MED15 in T2D pathogenesis. Figure 7: MED15 is expressed in mouse and human adult islets, and insulin positive stem cell derived β-like cells. (A) Western blot of MED15 in MIN6, mouse and human islets, and stage 6 day 11 (S6D11) differentiated β-like cell spheroids (n=2). (B) Normalized gene expression as determined by NanoString of MED15 in human islets from healthy and T2D donors (n=21 (healthy), n=12 (T2D) p≤0.05, Student’s t-test). To determine if MED15 is present in islet cells during specific developmental stages, we performed immunofluorescence staining in mouse pancreata throughout embryonic development. We found that MED15 was highly expressed in the nucleus of nascent insulin-positive β-cells at embryonic day 13.5 (E13.5) (Figure 8A). At E18.5, staining demonstrated MED15TubulinMIN6cellsMouse isletsHuman isletsS6D11spheroidsA130-63-Healthy T2D0100200300400MED15 Relative ExpressionB Human Islets*81 nuclear MED15 expression in some β-cells and in glucagon-positive ⍺-cells (Figure 8B). In 8-week old adult islets, MED15 expression was found primarily in the nucleus of β- and ⍺-cells (Figure 8C). Interestingly, MED15 was persistently present in the nucleus of β-cells in 8-week old adult islets, but less so at E18.5 (Figure 8B, C). As β-cell maturation occurs in mice during the early postnatal period, approximately spanning the timeframe from E18.5 to 8-week old adults (Stolovich-Rain et al., 2015), these data suggest that MED15 may play a role early in β-cell development and during maturation. Figure 8: MED15 is expressed in the nucleus of embryonic and postnatal β-cells in mice. Immunofluorescence of MED15 (green), glucagon (GCG, white), insulin (INS, red), and nuclei (Hoescht-33342, blue) in (A) E13.5, (B) E18.5, and (C) 8-week mouse pancreas sections (n=3, scale bars = 25µm). 3.2.2 Loss of Med15 leads to impaired glucose-stimulated insulin secretion To determine if MED15 is required for β-cell function, we studied 8-week old Med15 knockout Ins1Cre β-cell specific knock-in line (IM15KO; Thorens et al., 2015) mice. To control for gene dosage of insulin, control mice were Ins1Cre/+; Med15+/+ (CTRL) unless otherwise stated. Functionally, during glucose challenge with intraperitoneal glucose tolerance tests 8 WeeksE18.5AE13.5MED15/INS/GCG/DNAB C82 (IPGTTs), IM15KO mice were severely glucose intolerant (Figure 9A) from 30 minutes onward. At 10 minutes, IM15KO mice had significantly reduced serum insulin levels (Figure 9B), suggesting defective β-cell mass or function. Figure 9: IM15KO mice are glucose intolerant and have less serum insulin. (A) Intraperitoneal glucose tolerance test (IPGTT) in 8-week CTRL (black circles; n=10), and IM15KO mice (red squares; n=4; * = p≤0.05, Student’s t-test). (B) Serum insulin measurements of mice in (A) at time 0 and 10 minutes in CTRL (black circles) and IM15KO (red squares) mice (n=3; * = p≤0.05; Student’s t-test). As a previous report demonstrated that IM15KO islets have no changes in insulin expression or β-cell mass (Xu, 2016), we assessed β-cell function in these mice. Glucose stimulated insulin secretion (GSIS) was measured using a perifusion assay with glucose. To bypass glucose uptake and glycolysis, a-ketoglutarate was used to stimulate insulin secretion. In IM15KO islets, we observed a significantly abrogated response to 25mM glucose but no change in insulin secretion when directly stimulated with 10mM a-ketoglutarate (Figure 10A-B). This 83 suggests a defect upstream of mitochondrial function. When analyzing GSIS by first vs. second phase (time 10-20 and 20-40 respectively), we observed a defect in both first and second phase insulin secretion (Figure 10A). To delineate where the observed defects in GSIS lies, we Figure 10: Loss of Med15 impairs GSIS but does not affect KCl-induced insulin secretion. (A) Normalized insulin secretion from perifusion assay of 8-week CTRL (black circles) versus IM15KO (red squares) mouse islets stimulated with 25mM glucose and 10mM α-ketoglutarate (n=3, * = p≤0.05, two-way ANOVA with Sidak multiple-comparison test). (B) Total insulin measured from islets in (F) as measured by ELISA. (C) Static glucose-stimulated insulin secretion from 8-week old CTRL and IM15KO islets (n=3; * = p≤0.05; Student’s t-test). (D) Total insulin quantification from (F) (n=3). 84 performed a static incubation secretion assay with glucose and 40mM KCl to depolarize the cells. We observed significantly decreased GSIS but no changes in KCl-stimulated insulin secretion (Figure 10C-D). Taken together, these data demonstrate that IM15KO islets have defective GSIS but are responsive to other secretagogues such as a-ketoglutarate and KCl, suggesting an immature β-cell phenotype. Figure 11: IM15KO are unable to take up glucose in a timely manner. (A) Representative confocal microscope images of dispersed islets after 10 minutes of incubation with glucose analogue 2-NBDG. (B) Quantification of glucose uptake as measured by 2-NBDG in 8-week old CTRL (black circles) and IM15KO (red squares) dispersed islets expressed as relative fluorescence units (RFU) (n=3; * = p≤0.05, Student’s t-test). (C) Quantification of perifusion with 2-NBDG at 0, 10, 20, and 30 minutes (n=3). 2-NBDGCTRLIM15KOA02004006008002-NBDG uptake (RFU)*2-NBDG uptake (RFU)BCTime (minutes)0 10 20 3085 3.2.3 Med15 loss causes defects in glucose uptake in β-cells As the observed defective GSIS in IM15KO islets was found to be upstream of both ATP-sensitive potassium (K-ATP) channels and mitochondrial activity, as determined by KCl and α-ketoglutarate stimulated insulin secretion, we performed a glucose uptake assay to test the first step of GSIS. In brief, dispersed islets were incubated with a fluorescent glucose analogue 2-NBDG and measured using a confocal microscope. After incubation for 10 minutes, 2-NBDG uptake was found to be significantly reduced in IM15KO islets compared to control (Figure 11A, B). Using a perifusion chamber, gradual addition of 2-NBDG showed reduced accumulation in IM15KO islets at 10 minutes, but at 30 minutes, IM15KO islets had intracellular glucose levels comparable to control islets, suggesting a slower rate of glucose uptake in IM15KO (Figure 11C). Taken together, these data suggest that MED15 is required for proper glucose uptake and β-cell function. 3.2.4 Loss of Med15 leads to downstream mitochondrial defects To further characterize IM15KO islets, we performed staining for mitochondrial mass in IM15KO and control islets under different stimulatory conditions (collaboration with Rocky Shi, Luciani lab). In β-cells, mitochondrial function is essential for insulin secretion, which is enhanced by the generation of metabolic coupling factors (Wollheim and Maechler, 2002). As such, mitochondrial fission and fusion, which maintain functional mitochondrial networks, are essential processes for β-cells. In brief, dispersed islets were stained with mitotracker green (MTG) and tetramethylrhodamine (TMRE) to assess mitochondrial mass and membrane potential respectively. Visual inspection of IM15KO mitochondria under high glucose suggest that mitochondrial fusion, fission, and overall networking may be impaired (Figure 12A). When 86 stimulated with 18mM glucose or 10mM a-ketoglutarate, we observed no changes in mitochondrial mass (Figure 12B). IM15KO islets had significantly decreased normalized B C050100150Average MTG Intensity (AU)2.8mM 18mM 2.8mM + 10mM Į.HWRJOXWDUDWH*OXFRVHFRQFHQWUDWLRQVCTRL IM15.O2.8mM 18mM 2.8mM + 10mM Į.HWRJOXWDUDWH*OXFRVHFRQFHQWUDWLRQV0123TMRE/MTG Intensity (AU)*0,7275$&.(5/TMRE/DNACTRL ,0.2AFigure 12: Loss of Med15 does not impair mitochondrial function but may affect mitochondrial networking in β-cells. (A) Mitotracker green FM (MITOTRACKER- MTG) and Tetramethylrhodamine ethyl ester (TMRE) intensity between CTRL and IM15KO islets stimulated with 18mM glucose. (B) Mitochondrial mass, as measured by MTG intensity in dispersed islets from 8-week old CTRL and IM15KO mice when stimulated with 18 mM glucose or 10mM α-ketoglutarate (n=4 mice). (C) Normalized mitochondrial function, as measured by TMRE/MTG intensity in dispersed islets from 8-week old CTRL and IM15KO mice when stimulated with 18 mM glucose or 10mM α-ketoglutarate (n=4 mice; * = p≤0.05; two-way ANOVA with Sidak multiple-comparison test). 87 mitochondrial membrane potential when stimulated with 18mM glucose but no change when stimulated with 10mM a-ketoglutarate (Figure 12C). This is consistent with the observed defect in GSIS, whereby stimulation with a-ketoglutarate is sufficient to elicit a secretory response (Figure 10A). These data further suggest that IM15KO islets have defective glucose uptake but sufficient mitochondrial function. 3.2.5 Loss of Med15 impairs β-cell maturation GSIS is the hallmark of functional β-cell maturation. Thus, we sought to determine if MED15 is required for the expression of genes involved in this process. We performed alignment and data analysis on RNA-seq data from 8-week IM15KO and Ins1+/+;Med15fl/fl control islets. Gene ontology (GO) term analysis of genes with reduced expression revealed terms required for glucose metabolism and insulin secretion (Figure 13A), which supports defects in mature β-cell Figure 13: GO term analysis of RNA-seq data show MED15 regulates genes important for β-cell function. (A) GO-terms overrepresented in genes differentially expressed in 8-week old CTRL and IM15KO Islets. (B) Reactome Gene Set Enrichment Analysis (GSEA) pathway database of IM15KO differentially expressed genes as determined by RNA-seq (n=5). CTRL IM15KOSlc2a2MafaIappG6pc2Slc30a8Slc4a7Slc28a2Atp2a2Regulation of geneexpression in beta cells Transmembrane transportof small moleculesRegulation of betacell developmentSlc mediated transportReactome GSEAA0 2 4 6 8Cation TransportRegulation of Hormone Levels Regulation of Insulin SecretionIon TransportRegulation of Biological Quality-Log (p-value)Slc30a8,Ucp2, Ucn3, Sytl4, Glp1r, G6pc2, Trpm2, Nr1h4, HadhB88 function in the IM15KO islet. When using the REACTOME GO term dataset, genes involved in β-cell development and gene expression were enriched (Figure 13B). These data suggest MED15 is required for expression of key β-cell maturation and function genes. The most significantly downregulated genes in IM15KO islets included the cytochrome c oxidase subunit VIa polypeptide 2 (Cox6a2) encoding a mitochondrial electron transport chain (ETC) component, the β-cell maturation markers urocortin 3 (Ucn3) and Mafa, the glucose transporter Slc2a2/Glut2, and the enzyme endoplasmic reticulum oxidoreductase 1β (Ero1lb), which is important for disulphide bond formation during insulin maturation (Figure 14) (Zito et al., 2010). Among genes associated with β-cell maturation, IM15KO islets also expressed significantly less Sylt4, Iapp, and Slc30a8. Figure 14: MED15 regulates genes important for β-cell maturation. (A) Expression of genes known to be expressed in immature and mature β-cells, respectively, in 8-week old control (CTRL; black circles) and Med15 knockout (IM15KO; red squares) islets, as determined by RNA-seq analysis (n=5). 89 Aside from Mafa, no significant changes were observed for transcription factors associates with β-cell maturation such as Nkx6-1, NeuroD1, or Pdx1 (Figure 15A). This suggest that MED15 acts in parallel to or downstream of most β-cell maturation transcription factors. Surprisingly, IM15KO islets did not show increased expression of markers of immature β-cells, RPKM relative to controlGcgArxIrx2 Ttr SstHhex0.00.51.01.52.0-cells -cellsSST/INS/DNACTRLB IM15KOAFigure 15: Gene expression analysis of IM15KO islets reveal no increases in other endocrine cell types. (A) Expression of genes known to be expressed in various islet endocrine cell types (as indicated) in 8-week old control (CTRL; black circles) and Med15 knockout (IM15KO; red squares) islets, as determined by RNA-seq analysis (n=5). (B) Immunofluorescent staining for insulin (INS-green), somatostatin (SST-red), and DNA (Hoechst 33342-blue). 90 such as Mafb, Hk1, and Ldha. Often, when β-cells experience stress such as prolonged hyperglycemia (as is the case in T1D or T2D), they begin to transdifferentiate into other endocrine cell types (Szabat et al., 2012). Genes predominantly expressed by ⍺- or δ-cells such as Gcg or Sst were not changed in IM15KO islets (Figure 15A). Fluorescent staining for Figure 16: qPCR and immunofluorescence confirm loss of maturation genes in IM15KO islets. (A) Expression for select β-cell maturation genes in 8-week old control (CTRL; black circles) versus Med15 knockout (IM15KO; red squares) islets as determined by Taqman qPCR (n=4). (B) Immunofluorescence of 8-week CTRL versus IM15KO mouse pancreatic sections for select maturation markers (n=3; scale bars = 25µm). 91 somatostatin showed no changes in numbers of δ-cells (Figure 15B). This suggests that, in IM15KO islets, no transdifferentiation into other endocrine cell types may occur, although lineage tracing experiments are needed to confirm this. These data suggest that MED15 plays a role in the expression of specific genes required for GSIS and β-cell maturation. Using Ins1Cre/+; Med15+/+ mice to control for insulin gene dosage, we confirmed loss at the mRNA and protein level of various key β-cell genes by qPCR and immunofluorescence, respectively. In agreement with the RNA-seq data, control and IM15KO islets showed loss of Mafa, Ucn3, and Slc2a2 (Figure 16A, B), which are all required for establishing GSIS (Blum et al., 2012; Guillam et al., 2000; Nishimura et al., 2015). Collectively, these results demonstrate that IM15KO islets do not express key β-cell genes required for functional maturity. 3.2.6 Med15 loss alters mitochondrial networking in β-cells As we observed drastic changes in the expression of the mitochondrial gene Cox6a2 upon Med15 loss, we explored possible changes in mitochondrial dynamics in collaboration with the Luciani lab at BCCHRI. Using MTG stain and 3D confocal imaging, we measured changes in mitochondrial counts and volume. Interestingly, dispersed IM15KO cells had significantly fewer mitochondria per cell with each mitochondria having increased volume (Figure 17). This may be associated with changes in mitochondrial networking associated with β-cell maturation (Kaufman et al., 2015). Further, β-cells wherein the maturation transcription factor Nkx6-1 has been depleted also display increases in mitochondrial volume (Špaček et al., 2017). Taken together, these data suggest MED15 may be required to promote changes in mitochondrial 92 dynamics during β-cell maturation. Figure 17: IM15KO have reduced mitochondrial count but increases in networking (A) Three-dimensional reconstruction of mitochondrial network in dispersed islets from 8-week old control and IM15KO mice (n=3 mice). (B) Mitochondrial count per cell measured in (A). Each dot represents one cell (n=3 mice; * = p≤0.05, Student’s t-test). (C) Mitochondrial volume measured in (A). Each dot represents one cell (n=3; * = p≤0.05; Student’s t-test). Discussion β-cell function is essential to maintain glucose control in humans and mammals, and β-cell dysfunction underlies most forms of diabetes. Functional β-cell maturation, defined by appropriate GSIS, is essential and achieved through the expression of specific genes such as Mafa and Slc2a2 (Guillam et al., 2000; Kaneto and Matsuoka, 2015). Together, these genes are required for establishing appropriate GSIS, which is acquired during early postnatal life and is the hallmark of β-cell maturation (Blum et al., 2012; Stolovich-Rain et al., 2015). Transcription factors play essential roles in the differentiation, maturation, and function of the pancreatic β-cells (Bastidas-Ponce et al., 2017; Conrad et al., 2014; Donelan et al., 2010; Sabatini and Lynn, 2015). However, how individual transcription factors interface with RNA Pol II, and whether 93 these interactions can be harnessed to promote β-cell differentiation, has not been studied. In this study, we demonstrate that the transcriptional co-activator MED15, a subunit of the evolutionarily conserved Mediator complex, is essential for β-cell maturation. Specifically, genetic ablation of Med15 compromised glucose uptake, mitochondrial networking, and ultimately precipitated glucose intolerance. In mature mouse pancreatic islets, Mediator tail module subunits including Med15 were expressed at low but detectable levels. Previous studies in MIN6 cells demonstrated that the tail module subunit MED25 interacts with transcription factor HNF4⍺ to enhance insulin secretion (Han et al., 2012), suggesting a metabolic role for the tail module in β-cell development and/or function. HNF4⍺ mutations cause one subtype of MODY (MODY 1), and variants of HNF4⍺ are also associated with T2D (Wang et al., 2000); thus, the link of Mediator to transcription factors important for β-cell function is evident from this study. It is interesting to consider that multiple Mediator subunits may contact various β-cell transcription factors, perhaps acting in synergistic or combinatorial fashion to promote the expression of genes involved in driving maturity and GSIS in vivo. Contact of two Mediator subunits by one transcription factor, the glucocorticoid receptor, has been shown before (Chen et al., 2006) and represents a possible model here as well, especially as the C. elegans orthologue of MED15 interacts physically and functionally with multiple HNF4-like Nuclear Receptors (Grants et al., 2015; Shomer et al., 2019). When analyzing single cell RNA-seq data from human pancreatic islets (Segerstolpe et al., 2016), we found MED15 to be enriched in a population of cells also displaying high NKX6-1, PDX1, and MAFA levels, suggesting a role for MED15 in maturation or function in β-cells. This was further supported by impaired expression of MED15 in human T2D islets. In developing 94 mouse pancreata, MED15 was dynamically expressed from E13.5, E18.5, and in mature β-cells, with E13.5 showing high nuclear expression of MED15. Interestingly, MED15 was also found in the cytoplasm, which is consistent with its known interaction with E3-ubiquitin ligase Tripartite Motif Containing 11 (TRIM11) (Ishikawa et al., 2006). Furthermore, observed changes in cytoplasmic to nuclear localization have been observed with other β-cell maturation transcription factors such as PDX1 and MAFA (Ghaye et al., 2015; Kataoka et al., 2002; Semache et al., 2014; Spaeth et al., 2017). Thus, MED15 is expressed in β-cells and may be translocated into the nucleus during maturation, warranting follow up studies to explore this hypothesis. To functionally study MED15 in a mammalian model, we ablated it genetically in mouse pancreatic β-cells (Xu, 2016). In the same IM15KO mouse model, we observed that loss of Med15 in β-cells caused impaired glucose tolerance without affecting β-cell numbers. The observed hyperglycemia is likely caused primarily by defective β-cell glucose uptake. This was supported by my functional analyses, suggesting reduced mitochondrial function under high glucose conditions, likely driven by reduced glycolytic product availability. We also observed subtle differences in mitochondrial morphology that mirror those seen in other knockout models with impaired β-cell maturation, including a reduced mitochondrial count of Nkx6-1 knockout islets (Špaček et al., 2017) and an increased mitochondrial volume in Errγ knockout islets (Špaček et al., 2017; Yoshihara et al., 2016b). Downstream of β-cell glucose uptake, several genes important for insulin biogenesis, namely Ero1lb and Slc30a8 (Wijesekara et al., 2010; Zito et al., 2010), were also decreased upon Med15 loss. Collectively, loss of these MED15 regulated genes may contribute to the impaired glucose tolerance observed in IM15KO mice. 95 Furthermore, it remains unclear if changing the amount of one Mediator subunit would affect the activity of the entire complex. We speculate that providing MED15 earlier in development might improve maturation by increasing MED15 availability, turnover in the complex, recruitment to enhancers by transcription factors, and stability of transcriptional machinery. Further investigation into stabilizing MED15 by reducing its degradation via the E3-ubiquitin ligase proteasome system (Ishikawa et al., 2006) or into modulating specific MED15-transcription factor interactions pharmacologically, as done for the MED15-SREBP interaction (Zhao et al., 2014), may prove useful in improving stem cell differentiation for the treatment of those with insulin-dependent diabetes. 96 Chapter 4: β-cell Maturation Transcription Factors Bind MED15 at Key Loci Introduction During mouse development, transcription factors orchestrate critical gene expression cascades (Wilson et al., 2003). In the pancreas, several gene regulatory networks control cellular differentiation into various cell types, including endocrine cells of the islet (Arda et al., 2013). As postnatal β-cells mature and become glucose responsive, the expression of key genes and markers such as Slc2a2, G6pc2, Ero1lb, and Slc30a8 play important roles to permit glucose metabolism and insulin secretion (Guillam et al., 2000; Komatsu et al., 2013; van der Meulen et al., 2015; Nishimura et al., 2015; Qiu et al., 2017). Transcription factors such as PDX1, NKX6-1, NEUROD1, ISL1, ERRγ, and HNF4⍺ are required for the formation of mature β-cells via their target genes (Bastidas-Ponce et al., 2017; Gu et al., 2010; Wang et al., 2000; Yoshihara et al., 2016b), and are often impaired in different forms of diabetes such as T2D and MODY (Guo et al., 2013a; Steck and Winter, 2011). These proteins are essential nodes in gene regulatory networks important for β-cell maturation (Arda et al., 2013). Although transcription factors have been widely studied in the context of β-cell maturation, how they interact with coactivators, such as Mediator to drive expression, is less well understood. Mediator subunits interact with transcription factors at specific genomic loci to regulate gene expression via RNA pol II (Soutourina, 2018; Yin and Wang, 2014). The subunits of Mediator dynamically change as transcription occurs and its composition can vary across different tissues (Poss et al., 2013). Mediator can also change conformation based on its binding partners, and certain subunits can associate or dissociate without perturbing its core integrity (Harper and Taatjes, 2018; Poss et al., 2013; Soutourina, 2018; Tsai et al., 2017; Verger et al., 2019). Functionally, conformation changes allow Mediator to drive specific gene programs, and NKX6-1 MAFA 97 depending on the exact protein-protein interactions, allow engagement at many different promoters or enhancers (Haberle et al., 2019; Taatjes, 2010). Thus, interactions between Mediator and transcription factors are critical to regulate gene expression and to drive developmental and physiological programs. During transcription, Mediator is recruited to H3K4me1/H3K27Ac rich active enhancer regions and H3K4me3/H3K27Ac rich promoter regions (Sharifi-Zarchi et al., 2017). Regulatory regions such as super-enhancers also contain high levels of Mediator, transcription factors, and other coactivators (Whyte et al., 2013). Recent evidence demonstrates that the loss of Mediator tail module results in changed H3K27Ac deposition (El Khattabi et al., 2019). In hematopoietic stem cells, loss of MED12 results in reduced p300 binding and H3K27Ac (Aranda-Orgilles et al., 2016) . Interestingly, in p300/CBP histone acetyltransferase knockout mice, expression of key β-cell maturation genes such as Ucn3 and Slc2a2 are impaired due to loss of H3K27Ac at promoters (Wong et al., 2018). Thus, MED15 may also interact with histone acetyltransferases to induce expression of β-cell maturation genes. Mediator also interacts with chromatin remodeling complexes and the chromatin looping proteins such as LIM domain-binding protein 1 (LDB1) (Acevedo and Kraus, 2003; Gilmour et al., 2018; Kagey et al., 2010; Krivega and Dean, 2017; Sharma and Fondell, 2002). MED15 interacts with chromatin looping factors Cohesin and CCCTC-binding factor (CTCF) to position enhancers near promoters (Kagey et al., 2010). In β-cells, loss of chromatin looping factor (LDB1) impairs expression of maturation markers such as Slc2a2, Ucn3, and Mafa (Ediger et al., 2017). Emerging evidence also suggests changes in enhancer-promoter looping may contribute to diabetic states due to loss of islet specific 3D chromatin interactions (Cebola, 2019; 98 Greenwald et al., 2019). Thus, MED15 may also interact with chromatin remodeling or looping proteins in addition to transcription factors during β-cell maturation. Because Mediator has not been widely studied in pancreatic β-cells, we performed chromatin immunoprecipitation and co-immunoprecipitation experiments for MED15 and MED1 in pancreatic islets. We found that MED15 associates with actively transcribed genomic enhancer regions and directly regulates a suite of β-cell maturation genes. Mechanistically, MED15 shares regulated genes and chromatin targets with NKX6-1 and NEUROD1, and it binds these transcription factors physically in co-immunoprecipitation assays. Taken together, these data emphasize the importance of MED15 as a regulator of β-cell maturation via its interactions with transcription factors known to be critical for this process. Results 4.2.1 MED15 and MED1 are found at the Iapp enhancer and promoter To determine which genomic loci the tail module subunit MED15 is bound to in β-cells, we performed chromatin immunoprecipitation sequencing (ChIP-seq) analysis in MIN6 cells for MED15 (Figure 18). ChIP-seq for middle module subunit MED1 was performed in parallel to compare with MED15 binding sites since several reports have demonstrated successful ChIP-seq for this subunit (Harms et al., 2015; Whyte et al., 2013). Based on my previous finding in IM15KO islets whereby MED15 is required for β-cell maturation (Chapter 3), we hypothesized that MED15 would be found at regulatory regions, namely enhancers and promoters, of critical β-cell maturation genes. 99 Figure 18: Schematic of ChIP-seq workflow Diagram of Mediator composition (Grants et al., 2015) and ChIP-seq workflow highlighting position of MED15 and MED1 subunits within tail and middle module respectively. Initial attempts to ChIP MED15 in MIN6 cells by crosslinking with 1% PFA were unsuccessful, perhaps because Mediator does not directly contact DNA. Performing chromatin fixation with 3% PFA and sonication yielded fragments of ~300bp (Figure 19A). Following DNA extraction of immunoprecipitated chromatin, ChIP-qPCR was performed at putative enhancer and promoter regions of the Iapp gene to determine if ChIP was successful (Figure 19B-C). These regions were selected based on chromatin accessibility (ATAC-seq), enhancer (H3K4me1 broad domains) and promoter (H3K4me3) chromatin mark data (Figure 23B) (Avrahami et al., 2015; Lawlor et al., 2017b), and because Iapp shows significantly reduced expression in IM15KO islets (Figure 14). For both regions, MED15 and MED1 ChIP showed enrichment over IgG ChIP; in contrast, a Leptin (Lep) promoter control locus was not enriched in MED15 or MED1 ChIP (Figure 19C). These data demonstrate successful ChIP for MED15 and MED1 at the Iapp enhancer and promoter. Hence, the ChIP-ed DNA was deemed suitable for deep sequencing to identify MED15 and MED1 bound regions in unbiased fashion. RNA polymerase IIRegulatory elementsMed15Med1X-linkQC and qPCR ChIP validationSequencing and AnalysisSonicationMed15 and Med1 ChIP100 Figure 19: ChIP-qPCR analysis for MED15 and MED1 shows binding at Iapp enhancer and promoter regions (A) DNA gel electrophoresis of MIN6 chromatin before (left) and after (right) sonication. (B) Integrated Genome Viewer (IGV) tracks of previously published ATAC-seq, H3K4Me1, and H3K4Me3 ChIP-seq data (Avrahami et al., 2015; Lawlor et al., 2017b). Arrows represent qPCR primer target loci. (C) ChIP-qPCR for MED15 and MED1 in MIN6 cells expressed as % input (n=3). 4.2.2 MED15 and MED1 have similar yet distinct binding profiles To determine in an unbiased fashion, which regions are bound by MED15 and MED1, MED15 and MED1 immunoprecipitated DNA was sequenced. This yielded 41,934 peaks for 101 MED15 with 14,207 associated genes and 58,311 peaks for MED1 with 14,863 associated genes using a q-value cutoff of 0.01 (Table 7, Appendix A-B). Table 7: MED15 and MED1 MIN6 ChIP-seq summary # of Peaks # of bound Genes Top 25 bound genes MED15 41,934 14,207 G6pc2, Cpe, Bace2, Calm2, Kcne4, Camk2n1, Smg7, Dhx40, Glp1r, Dynlt3, Cirbp, Ptbp1, Fem1b, Eapp, Alyref2, Spc25, Mir691, Usf2, Srrm1, Zfp428, Mreg, Dusp4, Dleu2, Slc30a8 MED1 58,311 14,863 Kcnh1, 1700012I11Rik, Nufip1, Zfhx3, Zc3h4, Scg2, A430035B10Rik, Smim18, Gcgr, Vwa5b1, Ddx31, Rwdd4a, Uaca, Acsl5, Nav2, Ahcyl1, Ccdc132, 1700085C21Rik, C1qc Cdh26, Xylt1, Pcdh15, Resp18, Fmn1 When analyzing the overlap between genes bound by MED1 and MED15, we observed a 71% overlap (Figure 20A). These included genes involved in β-cell maturation such as Iapp, Slc2a2, Slc30a8, and G6pc2. Interestingly, MED15 alone was bound near critical β-cell maturation transcription factor Mafa and mitochondrial respiratory chain component Cox6a2. Conversely, MED1 but not MED15 bound Ldha and Cat, which are repressed in mature β-cells (Schuit et al., 2012). This agrees with a specific role for MED15 in β-cell maturation. Visually inspecting MED15 and MED1 ChIP-seq tracks, we found a similar yet distinct binding patterns for MED1 and MED15 (Figure 20B). These data suggest that although most bound loci are shared by MED15 and MED1, certain regions may contain a higher amount of MED15 or MED1, which could serve functional purposes during transcription. 102 Figure 20: MED15 and MED1 bind similar, yet distinct sets of genes (A) Overlap of MED15 and MED1 bound genes as determined by ChIP-seq (note, Venn diagram not to scale). (B) ChIP-seq tracks of MED15, MED1, and previously published data of RNA Pol2, ATAC-seq, H3K4me1, and H3K4me3 at Iapp gene (n=3) (Avrahami et al., 2015; Lawlor et al., 2017b; Lu et al., 2018). 4.2.3 MED15 binds and regulates critical β-cell genes To determine which genes are bound and regulated by MED15, we compared sets of genes bound and regulated by MED15, as determined by ChIP-seq and RNA-seq, respectively. We observed 37% of MED15 activated (i.e. downregulated in IM15KO mice) genes to be bound by MED15, including genes associated with β-cell maturation: Slc2a2, Iapp, Slc30a8, Ucn3, 103 Ero1lb, and G6pc2 (Figure 21A). Vice versa, 31% of MED15 repressed (i.e. upregulated in IM15KO mice) genes were bound by MED15, including Apob and Txnip (Figure 21A), both of which are upregulated in states of impaired insulin secretion (Hou et al., 2019; Shalev, 2014). When also including MED1 bound genes in this overlap analysis, we found 73 genes to be bound and regulated preferentially by MED15 including Mafa, Cox6a2, and Slc2a1. These data demonstrate that MED15 is bound to and likely directly regulates critical β-cell maturation genes in islets. Figure 21: MED15 binds and regulates key β-cell maturation genes (A) Overlap of genes bound by MED15 (ChIP-seq) and differentially expressed in IM15KO islets (RNA-seq) as determined. (B) Overlap of MED15 bound, MED1 bound, and IM15KO differentially expressed genes. 4.2.4 MED15 and MED1 are bound to genomic enhancers and promoters To determine which genomic features MED15 binds to in islets, we collaborated with the Hoffman lab to identify chromatin marks associated with regions bound by MED15 and/or MED1 (Tennant et al., 2013). Because Mediator associates with enhancer and promoter loci, we first compared overlaps between H3K4me1, H3K27Ac, and H3K4me3 enriched regions 104 (enhancer, accessible, and promoter regions respectively) and MED15 and MED1 binding sites. Whereas MED15 and MED1 associate with all three chromatin marks, at H3K4me1 enhancer regions, MED15 bound more regions compared to MED1 (Figure 22). Figure 22: MED15 binds more H3K4me1 regions compared to MED1 in β-cells MED15 and MED1 bound regions overlap at H3K4me1, H3K27Ac, and H3K4me3 marked regions. Score signal represents % overlap between regions ±5kb from region center. To further characterize the genomic regions bound by MED15 and MED1, we expanded the analysis to chromatin segmentation, including 8 chromatin marks in combination (Ernst and Kellis, 2017). This allows delineation of distal vs. proximal active enhancers, H3K27me3+H3K4me1 rich bivalent, and H3K27me3 + H3K9me3 repressed gene states. Interestingly, for both MED15 and MED1, the most ChIP-seq peaks were found at regions with 105 no chromatin marks (Figure 23A – orange and blue bars), labelled as states 13 and 16 (S13 and S16). When excluding S13 and S16 regions in the analysis, distal enhancer/elongation, proximal/cis-regulatory, and active-TSS regions were found to have binding of MED15 and MED1 in that order (Figure 23A-B). The lowest percentage of MED15 and MED1 peaks occurred at bivalent and repressive chromatin marks, suggesting MED15 and MED1 may preferentially activate gene expression (Figure 23A-B). Collectively, these findings suggest that MED15 and MED1 are bound primarily to regions without common chromatin marks and preferentially at distal and proximal enhancer regions. 4.2.5 MED15 is bound near enhancers of regulated genes To determine the chromatin state of MED15 bound and regulated genes, we performed the same analysis with MED15 bound regions corresponding to genes differentially expressed by RNA-seq in IM15KO islets (Figure 23B – red and black bars). Genes decreased in IM15KO islets showed the same pattern as overall MED15 and MED1 bound genes (Figure 23B), i.e. the majority of these genes featured MED15 bound at corresponding distal enhancer/elongation sites, proximal/cis-regulatory region, and active-TSS regions (Figure 23C). If MED15 were to directly repress gene expression, we would expect to observe a majority of genes increased in IM15KO to lie in bivalent or repressive states. As previously observed, however, only a low percentage of MED15 repressed genes (i.e. increased in IM15KO islets) featured corresponding MED15 binding sites with bivalent or repressed states (Figure 23C), suggesting that MED15 preferentially activates gene expression in islets. Overall, this analysis suggests MED15 is primarily bound to proximal and distal enhancer regions to induce gene expression. 106 Figure 23: MED15 and MED1 primarily bind distal enhancers and introns (A) Chromatin state segmentation analysis of MED15 bound (blue), MED1 bound (orange), MED15 bound/IM15KO downregulated (red) and MED15 bound/IM15KO upregulated (black) genes at Active TSS, Proximal Cis-Reg, Bivalent, and Repressed states. (B) Sum of bound and (C) regulated genes at each region. BActive TSSProximal/Cis-RegDistal Enhancer/ElongationBivalentRepressed0102030Sum of % Bound GenesAH3K4me3H3K27acH3K4me1H3K27me1H3K36me3H3K9me3H3K27me2H3K27me30102030400246810ChIP-Seq Peaks (%)Med1Med15Downregulated Upregulated Med15 Regulated/Bound Genes (%)Active TSS Proximal/Cis-RegBivalent Repressed Distal Enhancer/ElongationHistone Modifications1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22S13 S16RNA-seqChIP-seq1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22Active TSSProximal/Cis-RegDistal Enhancer/ElongationBivalentRepressed010203040Sum of % Regulated GenesC107 4.2.6 MED15 and MED1 primarily bind intronic and intergenic regions Because we observed the highest percentage of MED15 and MED1 bound regions to have no chromatin marks (state 13 and 16 - S13/16), we further examined these regions by functional annotation (McLean et al., 2010). Intriguingly, the majority of S13/16 regions are intergenic or intronic, which is in accordance when comparing all MED15 and MED1 bound regions to these regions (Figure 24). This is in line with and corroborates previous findings showing that Mediator binds introns and gene boundaries (Chereji et al., 2017). These regions may serve regulatory purposes for chromatin looping, which Mediator is involved in (Quevedo et al., 2019b; Whyte et al., 2013). MED15 and MED1 may therefore play a role in chromatin looping in β-cells. Figure 24: MED15 and MED1 bind intergenic regions Classification of regions bound by MED15, MED1, chromatin state 13 (S13), and state 16 (S16), expressed as % of total regions size. 13 and S16 represent intronic and intergenic regions. TSS = Transcriptional Start Site, TTS = Transcription Termination Site, UTR = Untranslated Region. AIntronIntergenicPromoter-TSS TTS3' UTRExonNon-CodingOther020406080% of PeaksS13S16Med1Med15108 4.2.7 MED15 binds critical β-cell maturation genes in mouse islets To study MED15 and MED1 binding in mouse islets, we performed ChIP-seq on islets from control and IM15KO 8-week old mice. Using a q-value cutoff of 0.01, we detected 42,206 peaks associated with 15,398 genes for MED15 and 34,297 peaks associated with 8,549 genes for MED1 (Table 8). Table 8: MED15 and MED1 islet ChIP-seq summary # of Peaks # of bound Genes Top 25 bound genes MED15 42,206 15,398 Maf, Serpinf2, 4930440C22Rik, Vmn2r75, Sult1c2, Tnfaip8, Ldlrad2, Homer1, Nptn, Macrod2, Mat2b, Pmpcb, Snrpn, Tpp2, BC048403, Gm8179, Ube2l3, Pik3c3, Plcd4, Ppp1r3c, T, Cntnap5a, Gm5475, Mbd3l1 MED1 34,297 8,549 BC048403, Gm8179, Ube2l3, Mat2b, Pik3c3, Plcd4, Ppp1r3c, T, Cntnap5a, Gm5475, Mbd3l1, Mettl4, Mir3108, Nedd4, Rps6, Setbp1, Tmx4, 1700047E10Rik, 3000002C10Rik, 4930525D18Rik, 4933402J10Rik, Fgfr2, Fstl5, Gpn1 Approximately 70% of binding events observed in control islets were reduced or abolished in IM15KO islets, yet retained MED1 binding (Figure 25A). This indicates that Mediator can still assemble on genomic regulatory regions even without MED15. To determine what type of genes were no longer bound by MED15 in IM15KO islets, we compared the ChIP-seq datasets to a sets of genes annotated by cell type, i.e. specific for each endocrine cell lineage, 109 etc. (DiGruccio et al., 2016). As expected, loss of MED15 caused an especially large reduction inFigure 25: ChIP-seq analysis in IM15KO islets shows loss of binding at Med15 regulated β-cell genes (A) Heatmap of Med15 and Med1 ChIP-seq at Med15 bound regions in CTRL vs. IM15KO islets (n=2). Colour intensity represents ChIP-seq peak signal at Med15 peak regions. (B) Percent difference of bound genes in CTRL vs. IM15KO sorted by endocrine cell type. (C) Overlap of Med15 and Med1 bound genes (as determined by ChIP-seq in islets) with genes differentially expressed in islets of IM15KO mice. (D) GO term analysis of genes bound and regulated by Med15 in islets. 110 binding to genes associated with the β-cell lineage (Figure 25B). Next, to identify potential direct targets of MED15 in islets and β-cells, we compared the MED15 and MED1 ChIP-seq datasets to our list of genes differentially expressed in IM15KO islets. We observed 58% of differentially expressed genes to be bound by MED15, including Ucn3, Iapp, and Ero1lb (Figure 25C, D). Of all regulated genes, 33% were bound by both MED15 and MED1, including Slc2a2, Glp1r, and Slc30a8 (Figure 25C). As expected, GO term analysis of MED15 bound and regulated genes revealed terms such as insulin secretion (Figure 25D). These findings confirm that MED15 binds to key β-cell maturation genes and show Mediator is able to assemble without MED15. 4.2.8 ChIP-seq of MED15 bound genes reveals several potential binding partners Mediator subunits interact with transcription factors to exert their function. To date, several transcription factors have been reported to interact with MED15 and its orthologues including SREBP/SREBF1, SMAD2/3, and TAZ in mammals, and NHR-49/HNF4 in C. elegans (Taatjes et al., 2002b; Taubert et al., 2006; Varelas et al., 2008; Yang et al., 2006; Zhao et al., 2013). Thus, to determine which transcription factors may interact with MED15 in β-cells to drive maturation, we analyzed the set of MED15 bound genes for overlaps with publicly available ChIP-seq datasets in various tissues (Chen et al., 2013; Kuleshov et al., 2016). Among gene sets that showed significant overlaps with MED15 bound genes, we found several known binding partners including SMAD2, SMAD3, and SREBF1 (Figure 26). C. elegans NHR-49 orthologues HNF4γ and PPAR⍺ (Lee et al., 2016) were also bound to sets of genes that overlap with MED15 bound genes (Figure 26), suggesting these interactions may be conserved across species. In the context of the pancreas, several critical β-cell maturation transcription factors including ESRRγ, NEUROD1, PAX6, ISL1, INSM1, and FOXA2 were found to bind genes that 111 significantly overlap with MED15 bound genes (Figure 26). These data suggest several transcription factor partners, which may interact with MED15 to drive β-cell maturation. Figure 26: MED15 bound genes overlap with genes bound by several other transcriptional factors Significant overlap of MED15 bound gene with other transcription factor ChIP-seq datasets from various tissues (Kuleshov et al., 2016). All gene sets displayed had significant overlap (p≤0.05) as determined by Fisher’s exact test. 4.2.9 MED15 colocalizes with β-cell maturation transcription factors at key loci near β-cell maturation genes As several β-cell maturation transcription factors bound similar genes to MED15 (Figure 26), we visually inspected ChIP-seq tracks of certain factors for colocalization at specific genes with MED15. To determine if MED15 colocalizes at β-cell maturation genes it regulates, we first studied regions near Slc2a2, Mafa, and Iapp. MED15 bound active enhancer and promoter regions near these genes, and these sites also showed binding by the transcription factors NKX6-1, NKX2-2, PAX6, PDX1, FOXA2, or NEUROD1 in islets or β-cells (Figure 27A-C) (Gutiérrez et al., 2016; Jia et al., 2015; Perelis et al., 2015; Swisa et al., 2017; Taylor et al., 2013). Near the EsrrSmad5Smad4Smad2NeuroD1Pax6 Isl1Smad1Insm1Smad3 Rxr Rfx6Hnf4PparSmad6Foxa2Hnf1Srebf1Smad9051015-Log10(p-value)Med15 bound genes112 Slc2a2 locus, both MED15 and MED1 were found between the enhancer and promoter (Figure 27A), a feature indicative of enhancer-promoter looping, which Mediator is involved in (Kagey et al., 2010). Near the gene encoding the critical β-cell transcription factor Mafa, we observed a MED15 specific peak at a distal intronic enhancer but no peak near the promoter (Figure 27B). As expected, MED15 and MED1 were also found near Iapp, along with multiple other transcription factors (Figure 27C). This suggests that MED15 may interact with β-cell maturation transcription factors at enhancer and promoter regions. Figure 27: ChIP-seq tracks show MED15 may be associated with β-cell maturation factors near specific genes (A-C) ChIP-Seq tracks of MED15, MED1, RNA-Pol II, FOXA2, NKX2-2, RFX6, PDX1, PAX6, NKX6-1, and NEUROD1. Arrowheads represent peaks colocalized with β-cell transcription factors required for maturation and arrow represents peak indicative of enhancer-promoter looping. To determine overlap between MED15 bound regions and regions bound by various other β-cell transcription factors on a larger scale, we performed clustering analysis using NKX6-1, Med15 Nkx6-1H3K4me1/Enhancer-1.5kb +1.5kbPdx1 Foxa2 NeuroD1Med1DB CAMed15Med1RNA Pol IIFoxA2Nkx2-2Rfx6Pdx1Pax6Nkx6-1NeuroD10 10 20 0 4 0Relative Position (kb)EnhancerChromatin State PromoterSlc2a2 Mafa8 10EnhancerIappDistal enhancer Promoter113 PDX1, FOXA2, and NEUROD1 ChIP-seq datasets at islet enhancer (H3K4me1 high) regions (Figure 28A). Out of 41,934 total sites, we identified a cluster of MED15 and MED1 bound sites that contained 12,286 regions with a strong binding signal of the aforementioned transcription factors (Figure 28A). De novo motif analysis of all MED15 bound regions identified enrichment Figure 28: MED15/MED1 bound regions at enhancers and motif analysis of all MED15 bound regions (A) Heatmap of colocalized binding at 12,286 regions between MED15, MED1, and β-cell transcription factors NKX6-1, PDX1, FOXA2, and Neurod1. (B) HOMER de novo motif analysis of all MED15 bound regions sorted by match p-value (determined by hypergeometric test). NKX6-1 and NEUROD1 binding elements are highlighted within the PITX3 motif. (C) HOMER known motif analysis of MED15, done similarly as described in (B). 114 of motifs with similarities to known NKX6-1 and NEUROD1 binding motifs (Figure 28B) (Jia et al., 2015; Taylor et al., 2013). Similarly, when scanning MED15 bound regions for enriched, previously known motifs, we identified a significant overrepresentation of a NKX6-1 binding motif, among others (Figure 28B). Interestingly, the de novo analysis also detected a significant match with ZNF143 (Figure 28A), a novel chromatin looping factor (Ye et al., 2020). Taken together, these data suggest that MED15 may interact with multiple transcription factor partners and coactivators primarily at enhancer regions to induce the expression of β-cell maturation genes. 4.2.10 MED15 interacts with β-cell maturation transcription factors NKX6-1 and NEUROD1 To determine which β-cell transcription factors regulate similar sets of genes as does MED15, we compared gene expression profiles from various transcription factor mutants in the pancreas (Lachmann et al., 2018). Thus, we found that gene sets controlled by transcription factors related to β-cell maturation, including PDX1, PAX6, NKX2-2, RFX6, GLIS3, ISL1, NEUROD1, and NKX6-1, significantly resembled the list of genes downregulated in IM15KO mouse islets (Figure 29). Intriguingly, some of these knockout models show striking similarities to the β-cell immaturity phenotypes of IM15KO mice including NKX6-1 and NEUROD1 knockout mice (Gu et al., 2010; Taylor et al., 2013). This suggested an interaction between MED15 and either of these two transcription factors. 115 Figure 29: Loss of MED15 is similar to loss of β-cell maturation transcription factors Overlap of IM15KO downregulated genes with genes deregulated after knockout of specific transcription factors (Kuleshov et al., 2016). All gene sets displayed had significant overlap (p≤0.05) as determined by Fisher’s exact test. To test whether MED15 and NEUROD1 or NKX6-1 bind similar sets of genes, we compared MED15, NKX6-1 and NEUROD1 ChIP-seq datasets (Jia et al., 2015; Taylor et al., 2013). We found a significant overlap of genes co-bound by MED15 and NKX6-1 (Figure 30A) as well as NEUROD1 (Figure 30B). These include several genes downregulated in IM15KO islets, such as: Slc2a2, Mafa, Slc30a8, G6pc2, and Ero1lb. Because these genes are associated with mature β-cell function, this supports the model that MED15 interacts with NKX6-1 and/or NEUROD1 to promote or maintain β-cell maturation. IM15KO Downregulated genes0 2 4 6 8 10Nkx6-1NeuroD1Isl1Glis3Rfx6Nkx2-2Pax6Pdx1-Log10(p-value)116 Figure 30: Overlap of MED15 bound genes with NKX6-1 and NEUROD1 bound genes (A) Overlap of MED15 with NKX6-1 bound and (B) NEUROD1 bound genes (Venn diagrams not to scale). Significance determined by Fisher's exact test. As no known MED15 binding partners in β-cells have been reported, we sought to determine which interaction may be involved in the induction of β-cell maturation. To test this, we performed co-immunoprecipitation (CoIP) for MED15 and transcription factors in MIN6 cells. After screening several candidate binding partners, we found that MED15 co-immunoprecipitates with NEUROD1 and NKX6-1 (Figure 31A). Reciprocal CoIP for NKX6-1 revealed that this interaction appears to be specific to MED15, because the middle module subunit MED1 did not co-immunoprecipitate NKX6-1 (Figure 31B). This is not surprising as MED15 and MED1 are found in different modules of Mediator. As expected, MED1 is immunoprecipitated by MED15 (Figure 31C), suggesting NKX6-1 binds MED15 specifically or via another tail module subunit. 117 Figure 31: MED15 physically interacts with transcription factors NEUROD1 and NKX6-1 (A) Co-immunoprecipitation of NEUROD1 and NKX6-1 with MED15. CTRL = non-specific IgG control. (B) Co-immunoprecipitation of MED15 and MED1 with NKX6-1. CTRL = ChgA antibody (non-specific IgG control) (C) Co-immunoprecipitation of MED15 with NKX6-1. CTRL = ChgA antibody (non-specific IgG control). Input: 30 ug of lysate for (A-C) (n=3). 4.2.11 MED15 may also interact with chromatin remodeling complexes Emerging evidence suggests that deletion of Mediator tail module subunits impairs chromatin accessibility via loss of H3K27Ac (El Khattabi et al., 2019). To determine which nuclear complexes MED15 may associate with and what biological processes it may be involved 118 in, we performed gene ontology analysis using MED15 bound regions (McLean et al., 2010). When using the ‘biological processes’ GO terms, MED15 bound regions were found within regions associated with protein acetylation and histone acetylation (Figure 32). Histone acetyltransferases are critical regulators of β-cell maturation and function (Wong, 2018; Wong et al., 2018), suggesting that MED15 may interact with these complexes as well as transcription factors to drive β-cell maturation and function. Figure 32: Gene ontology term analysis of biological processes using MED15 bound regions GO term - Biological Process analysis using MED15 bound regions using GREAT (McLean et al., 2010). Significant determined by hypergeometric test. When analyzing bound regions using the ‘cellular component’ GO terms, MED15 bound regions were found to within regions associated with histone methyltransferase complexes (Figure 33). As histone methyltransferase complexes play important roles for β-cell function and maturation (Campbell, 2019; Lu et al., 2018), MED15 may also interact with such complexes to exert its function. Taken together, MED15 may interact with several components of the transcriptional machinery to exert its function in islets. 119 Figure 33: Gene ontology term analysis of cellular components using MED15 bound regions GO term – Cellular Component analysis using MED15 bound regions using GREAT (McLean et al., 2010). Significant determined by hypergeometric test. To determine which chromatin remodeling complexes MED15 may interact with, we analyzed the MED15 bound genes for overlaps with publicly available ChIP-seq datasets from various tissues (Chen et al., 2013; Kuleshov et al., 2016). Proteins that shared significant sets of bound sites with MED15 included the histone acetyltransferases complex components NCOA1 and E1A-binding protein p400 (EP400), the chromatin looping factors CTCF and ZNF143 (Xu et al., 2011b; Ye et al., 2020), and several SWI/SNF-related, matrix-associated, actin-dependent regulators of chromatin (SMARC) type chromatin remodeling complex components (Figure 34A). Because MED15 bound regions associated with HAT and HMT activity, we also studied overlaps between genes regulated by MED15 and genes regulated by the p300/CBP histone acetyltransferases or the EED/PRC2 histone methyltransferase (Lu et al., 2018; Wong et al., 2018). We found 17 genes that were deregulated in islet or β-cell knockout models of all three proteins, including genes involved in β-cell maturation, such as Ucn3 and Ero1lb (Figure 34B). We observed a larger overlap between MED15 and p300/CBPKO regulated genes, including Slc2a2 and Iapp, than between MED15 and PRC2 regulated genes. This agrees with 120 the model that MED15 and p300/CBP primarily upregulate β-cell genes, whereas PRC2 is involved in the repression of such genes (Lu et al., 2018; Xu et al., 2014). Taken together, these data suggest that MED15 may interact with histone acetyltransferase complexes and chromatin looping or remodeling complexes to allow expression of β-cell maturation genes. Figure 34: MED15 bound gene overlap with coactivator ChIP-seq and RNA-seq data (A) The bar graphs show the significance of the overlap of MED15 bound gene with those bound by specific coactivators, chromatin remodeling, and looping proteins, as determine by ChIP-seq datasets from various tissues. (B) The Venn diagram shows the overlap of genes downregulated in IM15KO, p300/CBP knockout (p300/CBPKO), and Eed knockout (EeD/PRC2KO) mice. Discussion In the previous chapter we showed that loss of Med15 causes defects in β-cell maturation. In this chapter we set out to determine the transcriptional regulatory mechanism underlying these defects, by performing chromatin occupancy analysis of MED15 and MED1 in MIN6 cells and islets followed by computational comparison of the resulting data to publicly available datasets for other transcriptional coregulators and transcription factors. we observed that MED15 is bound to 37% of MED15 induced and 31% of MED15 repressed genes, with bound and induced 121 genes including Slc2a2, Mafa, and Slc30a8. In agreement with this finding, chromatin state segmentation analysis determined that MED15 is enriched at actively transcribed chromatin regions including islet enhancers. Previous studies showed that MED15 and MED1 was primarily found in intergenic regions and gene boundaries, a known characteristic of Mediator (Chereji et al., 2017; Paul et al., 2015), highlighting the fidelity of the ChIP-seq datasets we generated. Interestingly, enhancers showed higher MED15 occupancy compared to MED1 (Figure 22), possibly due to its interaction with β-cell maturation driving transcription factors. These data are consistent with the known role of Mediator in transcribing active genes (Allen and Taatjes, 2015), but for the first time provide insight into the Mediator regulome in β-cells. Mediator subunit interactions with transcription factor partners can selectively activate distinct gene regulatory networks in cell and developmental specific manners (Yin and Wang, 2014). Here, we show that MED15 physically interacts with NKX6-1 and NEUROD1, but it may also bind other transcription factors at actively transcribed regions (Figure 27, 31) to confer β-cell maturation. In contrast, MED1 did not bind NKX6-1 in CoIP assays, and MED1 binding was not detected at certain β-cell maturation gene enhancers, such as Mafa and Slc2a2 (Figure 27, 31A). These results demonstrate that, in β-cells, individual Mediator-transcription factor partnerships may dictate how Mediator governs gene expression and cell-type specific functions. As deletion of Med15 has a similar, yet distinct phenotype to loss of Nkx6-1 (Taylor et al., 2013) and NeuroD1 in β-cells (Gu et al., 2010), it is likely that MED15 may have NKX6-1 or NEUROD1 independent functions. For examples, loss of Nkx6-1 lead to increased expression of Neurog3 and polyhormonal cells whereas loss of Neurod1 increased expression of Npy and disallowed genes such as Ldha (Gu et al., 2010; Taylor et al., 2013). Overall, IM15KO mice 122 have less severe phenotypes compared to Nkx6-1 or NeuroD1 knockout β-cells. This suggests MED15 may fine tune transcriptional activity of NKX6-1 or NEUROD1. Moreover, Mediator interacts with CTCF and the cohesin complex at intergenic boundaries to enable chromatin looping (Chereji et al., 2017; Kagey et al., 2010). Such chromatin looping is important for pancreatic islet function and may be altered in T2D (Greenwald et al., 2019; Mishra and Hawkins, 2017; Pasquali et al., 2014; Thurner et al., 2018). Additionally, Mediator is required for the formation of dense regions of enhancers known as super-enhancers (Whyte et al., 2013). Thus, it will be of interest to determine if MED15 promotes enhancer-promoter looping or interactions at super-enhancers to express genes required for β-cell maturation or function, and whether loss of these interactions predisposes to diabetes. Using gene set enrichment analysis, we found that MED15 may associate with histone acetyltransferases and methyltransferases, which are emerging as factors critical for β-cell maturation (Campbell, 2019; Campbell et al., 2019; Lu et al., 2018; Wong, 2018). It is currently unclear if MED15 interacts with chromatin remodeling proteins in β-cells. Regions such as the kinase-inducible domain interacting (KIX) serve as transcription factor binding sites and are shared between MED15 and p300/CBP (Thakur et al., 2014). Gene set enrichment analysis of β-cell specific p300/CBP knockout mice suggests similar phenotypes to NKX6-1 and NKX2-2 (Wong et al., 2018). Similarities between KIX domains may confer similar binding partners to MED15 and p300/CBP, which could lead to increased chromatin accessibility and subsequent activation of gene expression at enhancer or promoter regions. Future studies such as CoIP followed by mass spectrometry in β-cells would serve to map out MED15 interactions and prove useful to delineate its mechanism of action during this developmental process. 123 Chapter 5: MED15/MDT-15 is Required for Zinc and Toxic Metal Stress Response Introduction In pancreatic β-cells, mature insulin secretory granules contain Zn2+ co-crystallized with insulin hexamers packaged into vesicle-like structures (Hou et al., 2009). Zn2+ is transported into the cytosol of β-cells by Zrt- and Irt-like proteins (ZIP; aka SLC39A family proteins) proteins and into insulin secretory granules by the cation diffusion facilitators (CDFs; also known as zinc transporter (ZnT) or solute carrier 30 (SLC30) family proteins) (Bosco et al., 2010; Lemaire et al., 2009). When visualized with transmission electron microscopy (TEM), insulin granules appear as circles with a dark dense core surrounded by a distinct white halo (Fava et al., 2012) (Figure 35A). In β-cells lacking Slc30a8 and thus featuring reduced zinc availability, these granules appear faint when imaged; this has been determined to represent immature insulin granules (Lemaire et al., 2009; Wijesekara et al., 2010). Although Slc30a8 knockout mice have no changes in glucose tolerance (Pound et al., 2012), combined loss of Slc30a7 and Slc30a8 leads to impaired GSIS (Syring et al., 2016b). Furthermore, changes in glucose concentrations dynamically affect influx of Zn2+, expression of ZIP transporters, and expression of metal sequestering metallothioneins (MTs) (Bellomo et al., 2011). MTs in humans have been implicated to be protective against diabetic insults (Cai, 2004), whereas mutations in SLC30A8 can have either protective or harmful effects (Flannick et al., 2014; Sladek et al., 2007). Thus, maintaining zinc homeostasis plays a critical role for mature β-cell function. The pancreas contains one of the highest concentrations of zinc in the body, with insulin secretory granules reaching millimolar concentrations (Li, 2014). β-cells supplemented with 124 excess Zn2+ had increased insulin secretion and expression of Mt1 and Slc30a8 (Nygaard et al., 2014). Conversely, depletion of Zn2+, results in less insulin content and decreased expression of important β-cell transcription factors Hnf1b, Hnf4a, MafA, Mnx-1, Nkx2.2, Nkx6.1, Pax4, Pax6 and Pdx-1, suggesting zinc homeostasis is critical for β-cell function (Lawson et al., 2018; Nygaard et al., 2014). In the human body, concentrations of Zn2+ above 60ng/l can lead to toxicity (Nriagu, 2011). As such, β-cells must maintain a safe balance of Zn2+ to prevent toxicity while maintaining sufficient levels for insulin crystallization and other processes (Zheng et al., 2018a). Contrary to zinc, cadmium is highly toxic and plays no known biological role. As Cd2+ and Zn2+ share a similar electron configurations, the high toxicity observed by Cd2+ is due to its displacement of Zn2+ in enzymatic reactions (Zhang et al., 2014). To protect against influx of heavy metals and potential cytotoxicity, cells must rapidly express genes such as MTs and zinc transporters to maintain adequate levels of Zn2+ or Cd2+ (Thirumoorthy et al., 2007). Additionally, cadmium supplementation in rats leads to its accumulation in the pancreas and exposure to cadmium has been linked to reduced serum insulin in humans (Lei et al., 2005, 2006), highlighting the susceptibility of the pancreas and requirement to protect against heavy metal toxicity (Balamurugan et al., 2004). Of interest to human disease, cadmium is often found in cigarette smoke, which can exacerbate associated lung disease (Brzóska and Moniuszko-Jakoniuk, 2001; Ganguly et al., 2018); disruptions in zinc homeostasis can further lead to increased lung damage when exposed to cigarette smoke (Knoell et al., 2020) . The nematode worm C. elegans features an adaptive response to toxic heavy metal exposures that resembles that one seen in mammals. In worms fed excess zinc, lysosome-related organelles known as gut granules form and serve as a storage site for excess zinc to protect 125 against deleterious effects (Roh et al., 2012). These granules are rich in CDF-2, the orthologue of mammalian Slc30a8/ZNT8, which serves to transport Zn2+ into granules (Figure 35B) (Roh et al., 2012). Because β-cells contain among one of the highest concentrations of zinc in the body (Hou et al., 2009), this presents a parallel whereby mammalian insulin secretory granules have similar features to gut granules in worms (Figure 35A, B). Interestingly, the MED15 orthologue mdt-15 is required to induce cdf-2 as well as the heavy metal responsive MT genes mtl-1 and mtl-2 in conditions of cadmium or high zinc (Taubert et al., 2008). Thus, MED15/MDT-15 may have a conserved function in worms and mammals to maintain zinc homeostasis by inducing the expression of heavy metal stress response genes. Herein, we demonstrate MED15 is an important regulator of zinc homeostasis, a mechanism which may be conserved between tissues and species. We show that loss of Med15 in β-cells leads to a reduction in zinc transporter Slc30a8/ZNT8 and to impaired insulin crystallization and maturation in secretory granules. Using various dyes to assess zinc content, we find loss of Med15 leads to a reduction in total zinc content. When stressed with high levels of Zn2+ or the presence of Cd2+, we observe MED15 to bind genes associated with heavy metal stress response and Zn2+ transport in mouse β-cells and in human lung cells. We find that MED15 may also be required for downstream expression of heavy metal stress response genes. Further, we find the orthologue of MED15, MDT-15, to be required for proper Zn2+ storage by gut granule formation in C. elegans fed excess Zn2+. Taken together, we show the importance of MED15 in zinc homeostasis and heavy metal stress response. 126 Figure 35: Zn2+ storage granules in mammalian β-cells and C. elegans intestinal cells Schematic of Zn2+ storage granules (created with BioRender.com) found in (A) β-cell insulin secretory granules and (B) C. elegans intestinal gut granules. ZNT8/CDF-2 serve to transport free Zn2+ into these granules. Zn2+Zn2+Zn2+Zn2+ insulinhexamersZNT8Zn2+Zn2+Zn2+Zn2+Zn2+Zn2+Zn2+Sequestered Zn2+Zn2+Zn2+Zn2+Zn2+CDF-23DQFUHDWLFȕFHOO C. elegans intestineZn2+Zn2+Zn2+Zn2+Zn2+Zn2+Zn2+Zn2+ Zn2+Zn2+Zn2+Zn2+Zn2+Zn2+Zn2+Zn2+Zn2+Zn2+ Zn2+Zn2+Zn2+Zn2+Zn2+Zn2+Zn2+Zn2+Zn2+Zn2+Zn2+Zn2+Zn2+Zn2+Zn2+A BInsulinsecretory granulesGut granules127 Results 5.2.1 Loss of Med15 impairs zinc import into insulin secretory granules As we observed a significant reduction in the zinc transporter gene Slc30a8 in IM15KO islets (Figure 14), we performed immunofluorescent staining for Slc30a8/ZNT8. As expected, IM15KO islets had reduced expression of ZNT8 compared to CTRL (Ins1-Cre+/+; Med15fl/fl) mice (Figure 36). This confirms loss of ZNT8 protein expression in IM15KO islets, suggesting defective insulin granule maturation due to impaired Zn2+ uptake caused by reduced expression of ZNT8. Figure 36: Islets of IM15KO mice have reduced expression of zinc transporter Slc30a8/ZNT8 Immunofluorescent staining of 8-week old control CTRL (Ins1-Cre+/+; Med15fl/fl) and knockout (IM15KO) mice pancreata for ZNT8 (green), PDX1 (red) and DNA (Hoechst 33342– blue), (scale bars = 25µm, n=3, 1 sample shown). ZNT8/PDX1/DNA CTRLIM15KOZNT8/PDX1/DNA (63x)128 To determine if IM15KO mice have fewer mature insulin secretory granules, we performed TEM on glutaraldehyde-fixed pancreata from CTRL and IM15KO mice. In CTRL β-Figure 37: IM15KO mouse islets have less and smaller zinc-rich, mature insulin secretory granules (A) TEM of 8-week control (CTRL) and knockout (IM15KO) islets. Arrowheads represent mature insulin secretory granules (scale bars = 4µm). (B) Quantification of mature insulin secretory granules from (A) (n=3-4, * = p≤0.05, unpaired Student’s t-test). (C) Quantification of mature insulin secretory granule size from (A) (n=3-4, * = p≤0.05, unpaired Student’s t-test). CTRL IM15KOGFCAB0.00.10.20.30.4Granule count/cell area (%)CTRL IM15KOGranule size (AU)0.180.200.220.240.260.28**129 cells, we observed a high number of mature dark granules (as identified by dense core granule halo), whereas IM15KO β-cells contained fewer of these granules and more faint, immature granules (Figure 37A). Quantifying the number of mature granules in these β-cells revealed significantly fewer mature dense core granules per β-cell area (Figure 37B) and significantly smaller granules in IM15KO compared to CTRL β-cells (Figure 37C). These data demonstrate a loss of mature dense core granules in IM15KO cells, which may be due to loss of Slc30a8/ZNT8 and Zn2+ uptake into insulin secretory granules. 5.2.2 Loss of Med15 results in impaired intracellular zinc homeostasis Because the endocrine pancreas contains a high level of ZNT8 (Davidson et al., 2014) and we observed decreases in ZNT8 in IM15KO (Figure 36), we hypothesized that IM15KO islets would have less total Zn2+ content. We used the metal ion binding dye dithizone (DTZ), which is used to assess islet purity, to assess islet zinc content in IM15KO islets (Hansen et al., 1989). In IM15KO mice, we observed a decrease in DTZ binding (red colour) compared to CTRL islets, suggesting less zinc content when Med15 is deleted (Figure 38A). This observation is consistent with Slc30a8 knockout mice (Lemaire et al., 2009), suggesting it is a direct effect of less zinc in islets and insulin secretory granules. To confirm that IM15KO islets have a reduction in Zn2+ content, we stained dispersed IM15KO islets with the cell permeable Zn2+ binding dye FluoZin-3, which is used to assess islet viability (Byun et al., 2014; Jayaraman, 2008; El Muayed et al., 2010). In IM15KO cells, we observed a reduction in cytoplasmic FluoZin-3 signal and a more punctate patterning, suggesting reduced zinc levels (Figure 38B). Quantification of fluorescence intensity showed a significantly 130 reduced FluoZin-3 signal in IM15KO cells (Figure 38C). Taken together, these data show that Figure 38: IM15KO islets have reduced zinc content (A) Dithizone (DTZ-red) staining of 8-week CTRL and IM15KO islets (n=3; 1 sample shown). (B) FluoZin-3 (green) and DNA (Hoechst – blue) staining of 8-week CTRL and IM15KO dispersed islets (scale bars = 25µm, n=3; 1 sample shown). (C) Quantification of normalized FluoZin-3 mean fluorescence intensity in (B) (n=3, * = p≤0.05, unpaired Student’s t-test). 131 loss of Med15 perturbs zinc homeostasis and leads to an overall reduction in zinc content. 5.2.3 MED15 binds loci of zinc and cadmium metal detoxification genes The above data show that loss of Med15 causes alters zinc homeostasis in pancreatic islets. This could be a direct effect, i.e. by MED15 associating with and regulating genes involved in zinc uptake, storage, and/or sequestration, or indirectly. To determine if MED15 associates with the loci of zinc homeostasis related proteins, we performed GO term analysis of the MED15 ChIP-seq data using molecular function terms and found that MED15 bound genes are significantly enriched in term relates to metal and ion binding and transport (Figure 39). Specifically, MED15 bound genes included the ZnT family genes Slc30a8, Slc30a7, and Slc30a5; the ZIP family genes Slc39a6 and Slc39a14; and the metallothionein genes Mt1 and Mt2. As all of these gene products interact with Zn2+ to maintain zinc homeostasis (Bellomo et al., 2011; Kelleher et al., 2011; Nygaard et al., 2014), these findings suggest MED15 may directly bind and regulate genes involved in zinc homeostasis other than Slc30a8. Figure 39: GO-term analysis shows IM15KO bound genes involved in metal ion binding GO term – Molecular Function analysis using Med15 bound regions. Significance determined by hypergeometric test. 0 2 4 6 8 10Monovalent inorganic cation transmembrane transportSmall molecule bindingMetal ion transmembrane transporter activityIon bindingInorganic cation transmembrane transporter activityIon transmembrane transporter activityAnion bindingTransporter activityTransmembrane transporter activity-Log10(p-value)132 Visually inspecting MED15 ChIP-seq tracks near the bound and regulated gene Slc30a8, we found MED15 and other β-cell maturation transcription factors such as FOXA2, NKX6-1, and NEUROD1 at distal enhancer and promoter regions (Figure 40A). This suggests that Figure 40: MED15 is associated with the metal detoxification genes Slc30a8, Mt1, and Mt2 in β-cells (A) Chromatin state segmentation and ChIP-Seq tracks of MED1, MED15 and previously published RNA-Pol II, FOXA2, NKX2-2, RFX6, PDX1, PAX6, NKX6-1, and NEUROD1 near Slc30a8. Arrow represents peak at upstream enhancer region. (B) MED15 ChIP-seq track with previously published RNA Pol2, H3K4me1, and H3K4me3 at Mt1, Mt2, Mt3, and Mt4 regions. 133 MED15 may interact with β-cell maturation transcription factors to control the expression of Slc30a8. Visual inspection of metallothionein genes (Mt1, Mt2, Mt3, and Mt4) revealed that MED15 bound to the Mt1 and Mt2 promoter regions (Figure 40B). Interestingly, H3K4me3/promoter marks were only detected at Mt1, Mt2, and Mt3, suggesting that Mt4 is not expressed in islets (Figure 40B). These data suggest that MED15 directly binds regulatory elements in the Slc30a8, Mt1, and Mt2 genes. To test whether MED15 binding to these genes is regulated by excess zinc, we performed ChIP-qPCR experiments in MIN6 with and without 100µM Zn2+. We found that MED15 binding to the Slc30a8 enhancer and Mt1 promoter was increased by the addition of 100µM Zn2+ (Figure 41), suggesting MED15 is recruited to these regions to increase expression of gene products to counteract increases in Zn2+ concentrations. In this context, MED15 may bind a transcription factor in a zinc inducible fashion and recruited to these regions. Figure 41: MED15 binds regulatory elements in the Slc30a8 and Mt1 genes in a zinc inducible fashion MED15 ChIP-qPCR analysis in MIN6 cells at the Slc30a8 enhancer and Mt1 promoter regions ± 100µM Zn2+ (n=3, * = p≤0.05, unpaired Student’s t-test). Med15RNA Pol2H3K4me1H3K4me3Mt1Mt2Mt4 Mt3Slc30a8 enhancerSlc30a8 CTRLMt1 promoterMt1 CTRL012345% InputMed15Med15 + 100µM ZnAB**134 Because we observed recruitment of MED15 to genes required for zinc homeostasis when supplemented with zinc, we hypothesized that MED15 is required for the increased expression of such genes in these conditions. As IM15KO islets differ developmentally from CTRL islets in terms or maturation, determining if stress response to heavy metals may be confounded by these fundamental differences. Instead, acute knockdown of Med15 in β-cell subsequently stressed with high Zn2+ is preferable. However, attempts to knock down Med15 in MIN6 cells proved unsuccessful (Figure 42A). In line with the increased MED15 binding observed at the Slc30a8 enhancer and Mt1 promoter in high zinc treated MIN6 cells, Slc30a8 and siMed15CTRL Med15 OE CTRL100-42-ABMed15ActinMt1Mt2Slc30a1Slc30a4Slc30a5Slc30a7Slc30a9Slc30a8mRNA normalized to 18srRNA (fold induction)0246810 &75/ȝ0=Q2+&75/ȝ0=Q2+&75/ȝ0=Q2+,0.2ȝ0=Q2+,0.2ȝ0=Q2+,0.2ȝ0=Q2+Slc30a8 Mt1Mt2Med1505101520Normalized fold change —0=n—0=n0.05**n.s.CFigure 42: Genes required for heavy metal stress response are zinc inducible (A) Western blot in MIN6 cells for MED15 and Actin with CTRL (left = CMV-GFP and right = scrambled siRNA) and CMV-MED15 or siMED15 (n=3). (B) qPCR for Slc30a8, Mt1, Mt2, and Med15 in MIN6 cells ± 50µM Zn2+ (n=3, * = p≤0.05, n.s. = not significant, p≥0.05, unpaired Student’s t-test). 135 Mt1, as well as Mt2, were induced upon high zinc stress (Figure 42B). Interestingly, expression of Med15 itself was not induced by zinc (Figure 42B). Nevertheless, my data suggest that MED15 is recruited to Slc30a8 and Mt1 in β-cells in a Zn2+ dependent manner, possibly as a regulatory mechanism to lower increasing Zn2+ concentrations. 5.2.4 MED15 binds heavy metal stress response gene MT2A in a Cd2+ inducible fashion To study putative MED15 dependent heavy metal responsive gene regulation in another model, we performed ChIP-qPCR in A549 lung adenocarcinoma cells exposed to 5µM cadmium. Because Cd2+ and Zn2+ share similar properties, cellular stress response effectors, such Figure 43: Human MED15 binds metallothionein MT2A in a cadmium dependent fashion (A) Previously published ChIP-seq track of RNA Pol2 in A549 cells at metallothionein gene loci (data from Gertz et al., 2013). (B) ChIP-qPCR for MED15 in A549 cells ± 5µM Cd2+ (n=3, * = p≤0.05, unpaired Student’s t-test). RNA Pol2A549 CellsMT4 MT3 MT2A MT1L MT1A MT1F MT1XABMT1X promoterMT1X -10kb CTRLMT2A promoterMT2A -10kb CTRL0.00.51.01.52.02.5% InputMED15 ChIP CTRLMED15 + 5µM Cd*136 as increasing MTs and zinc transporters, are also shared (Balamurugan et al., 2004). Visually inspecting publicly available RNA Pol2 ChIP-seq in A549 cells (Figure 43A) (Gertz et al., 2013), we observed binding at MT2A and MT1X. This suggests that, among other MT genes, MT2A and MT1X are actively transcribed in this cell type. To determine if MED15 is recruited to these loci in a Cd2+ dependent fashion, we performed ChIP-qPCR for MED15 in A549 treated with and without 5µM Cd2+. We found that MED15 binds preferentially to the MT2A promoter, compared to the MT1X promoter, upon heavy metal stress with Cd2+ (Figure 43B). Importantly, knockdown of MED15 results in an abrogated induction of MT2A by exogenous Cd2+ in these cells (Shomer et al., 2019). Taken together with the observed role for MED15 in β-cell heavy metal stress response, these data suggest that MED15’s role in the heavy metal stress response may be conserved in human lung cells. 5.2.5 C. elegans mdt-15 is required to express genes important for excess zinc storage The C. elegans ortholog of MED15, MDT-15, was previously shown to play a role in zinc and cadmium responsive transcription (Taubert et al., 2008). To delineate if MDT-15 is involved in the heavy metal stress response in a similar manner to MED15 in β-cells, we collaborated with Naomi Shomer to perform confocal microscopy on mdt-15 mutant worms grown on high zinc containing media (Shomer, 2018). When grown on plates containing zinc binding dye FluoZin-3 without supplemental zinc, we observed fluorescent gut granules in control (WT) and mdt-15(tm2182) mutant worms at comparable levels (Figure 44A, B). This suggests that, when unstressed, mdt-15(tm2182) mutant worms are able to effectively form gut granules for normal zinc storage. 137 Figure 44: mdt-15(tm2182) mutants have normal gut granule formation in the absence of Zn2+ supplementation (A) Representative micrographs of wild-type (WT = strain N2) and mdt-15(tm2182) mutant worms fed the Zn2+ binding dye FluoZin-3 and 0µM Zn2+ supplementation (scale bars = 20µm). (B) Quantification of gut granules in (A) as determined by FluoZin-3 (n=10-11, error bars represent standard deviation, n.s. = not significant, p≥0.05, unpaired Student’s t-test). When supplemented with 200µM Zn2+ and FluoZin-3, WT worms displayed a large number of punctate gut granules rich in Zn2+ (Figure 45A). Interestingly, in mdt-15(tm2182) mutant worms, we observed a significant reduction in the number and size of gut granules as well as an increase in cytosolic free zinc (Figure 45A, B). This suggests that loss of MDT-15 FLUOZIN-3 DICWT ȝ0=Q2+ )mdt-15(tm2182) ȝ0=Q2+ ) A B138 leads to impaired zinc storage. To determine if this phenotype is specific to MDT-15’s role in Figure 45: mdt-15(tm2182) mutants have impaired gut granule formation in conditions of excess Zn2+ (A) Representative micrographs of WT, mdt-15(tm2182), cdk-8(tm1238), and hizr-1(am286) worms fed FluoZin-3 and 200µM Zn2+ (scale bars = 20µm). (B) Quantification of gut granules in WT, mdt-15(tm2182), cdk-8(tm1238) (A) as determined by FluoZin-3 signal (n=10-13, error bars represent standard deviation, **** p < 0.0001, One-way ANOVA, multiple comparisons, Dunnett correction, compared to WT). (C) Quantification of gut granules in hizr-1(am286) (n=13, error bars represent standard deviation, *** p < 0.001, unpaired Student’s t-test). FLUOZIN-3 DICWT ȝ0=Q2+ )cdk-8(tm1238) ȝ0=Q2+ ) mdt-15(tm2182) ȝ0=Q2+ )+ hizr-1(am286) ȝ0=Q2+ ) A BC139 Mediator, we studied Mediator kinase cdk-8 mutant worms (Grants et al., 2016). In cdk-8(tm1238) null mutant worms, this phenotype was not observed, suggesting it is specific to MDT-15’s function in zinc metabolism (Figure 45A, B). A nuclear hormone receptor known as high-zinc–activated nuclear receptor (HIZR-1) induces transcription of heavy metal stress response genes in C. elegans (Warnhoff et al., 2017). The hizr-1 null mutant hizr-1(am286) displayed a similar phenotype to mdt-15(tm2182) worms (Figure 45A, C), suggesting MDT15 and HIZR-1 may cooperate within the same pathway. Follow up yeast two-hybrid (Y2H) studies determined that MDT-15 and HIZR-1 interact in a zinc inducible fashion (Shomer et al., 2019). Taken together, these data suggest MED15’s function in zinc homeostasis and heavy metal stress response may be conserved across species. mdt-15 mutants have defects in lipid metabolism (Taubert et al., 2006) and this may contribute to the observed reduction in gut granules by disruptions in lipid membrane formation. mdt-15(tm2182) mutants have impaired expression of fatty acid metabolism enzymes such as fat-6 (a stearoyl-CoA desaturase) (Taubert et al., 2006). To determine if the contribution of defective lipid metabolism in mdt-15(tm2182) mutants is sufficient to cause impaired gut granule formation, worms were fed fat-6 RNAi and placed on high zinc containing media. As expected, control (CTRL) RNAi worms on high zinc formed gut granules, with mdt-15 RNAi worms having significantly fewer granules (Figure 46A, B). Interestingly, fat-6 RNAi fed worms did not show a similar phenotype, with similar number of gut granules compared to CTRL RNAi fed worms. These findings suggest the observed defects in zinc metabolism are due to MDT-15’s direct role in zinc metabolism. 140 Figure 46: Impairment of zinc granules is independent of MDT-15’s role in lipid metabolism (A) Representative micrographs of worms fed control (CTRL), mdt-15, or fat-6 RNAi, FluoZin-3, and 200µM Zn2+ (scale bars = 20µm). (B) Quantification of gut granules in (A) as determined by FluoZin-3 signal (n=6 for control, 3 for mdt-15, and 5 for fat-6 RNAi, error bars represent standard deviation, * p < 0.05, n.s. = not significant, One-way ANOVA, multiple comparisons, Dunnett correction, compared to control RNAi). Discussion In this Chapter, we identified a conserved role for MED15 in zinc metabolism and heavy metal stress response. Loss of Med15 in pancreatic β-cells leads to less total zinc content as well as less and smaller mature insulin secretory granules, likely caused at least in part by a reduction in Slc30a8 expression. MED15 is found near genes associated with zinc metabolism and heavy FLUOZIN-3CTRL RNAi ȝ0=Q2+ )Amdt-15 RNAi ȝ0=Q2+ )fat-6 RNAi ȝ0=Q2+ )DIC B141 metal stress response. In two cell types, MED15 binding to zinc metabolism genes increases upon high Zn2+ stress and possibly regulates their expression. In C. elegans, loss of MDT-15 results in impaired formation of gut granules, the primary storage site for excess Zn2+, independently of its role in lipid metabolism. These data suggest a role for MED15 in zinc metabolism and heavy metal stress response which may be conserved between species and tissues. In the context of β-cell maturation, Zn2+ plays an important role in mature insulin formation (Baker et al., 1988). ZnT, ZIP, and MT family proteins have been implicated in the pathology of diabetes (Bellomo et al., 2011; Flannick et al., 2014; Yang et al., 2008). Coactivators such as MED15 may interact with transcription factors to induce expression of genes required for zinc metabolism. In agreement with this model, a decrease in mature insulin secretory granules and in Slc30a8 expression is observed in Nkx6-1 knockout islets (Taylor et al., 2013), which MED15 physically interacts with. Furthermore, single cell RNA-seq analysis of immature and mature β-cells showed enrichment in Mt1 and Mt2 in mature β-cells (Qiu et al., 2017), indicating a role for other metal metabolism genes in this process. Because MED15 binds regulatory elements near the Mt1 and Slc30a8 genes in an inducible fashion in β-cells, this may enhance the expression of such genes during the maturation process. Mechanistically, MED15 may bind an unknown transcription factor in a zinc inducible fashion and be recruited to these regions. Together, this suggests MED15 may directly contribute towards expression of genes required for zinc metabolism during β-cell maturation. When stimulated with glucose, intracellular Zn2+ concentrations increase along with expression of MT and ZIP genes (Bellomo et al., 2011). Similarly, stimulation with Zn2+ induced insulin secretion and the expression of such genes (Nygaard et al., 2014). When MIN6 cells were 142 stimulated with Zn2+, MED15 showed increased recruitment to the Slc30a8 and Mt1 genes (Figure 41). Thus, MED15 may be required to maintain adequate Zn2+ homeostasis during GSIS, which is critical for β-cell function. Experiments to demonstrate if loss of MED15 may prevent insulin secretion when stimulated with Zn2+ could be performed to test this hypothesis. Additionally, as IM15KO islets have markedly reduced expression of Glut2/Slc2a2 (Figure 14, 16B) and less glucose uptake (Figure 11), which may contribute to less intracellular Zn2+. Furthermore, because β-cells contain among the highest levels of Zn2+ in the body (Davidson et al., 2014), protection from associated toxicity is important and the loss of such cytoprotection may contribute to diabetes (Cai, 2004; Chen et al., 2015; Thirumoorthy et al., 2011). Thus, MED15 may protect from heavy metal induced toxicity by inducing protective genes. Taken together, the requirement of MED15 for proper GSIS may relate to changes observed in zinc metabolism and heavy metal stress response. In human lung cancer cells, MED15 bound the promoter of MT2A when stimulated with Cd2+ (Figure 43B). Interestingly, MT2A induction by Cd2+ was abrogated when MED15 was impaired (Shomer et al., 2019). Because Cd2+ and Zn2+ are closely related, this suggests MED15’s role in zinc and cadmium detoxification may be conserved in humans. In the context of tumourigenesis, MED15 is upregulated in certain cancers such as head and neck squamous cell carcinoma, breast cancer, and renal cell carcinoma, with high expression of MED15 relating to poor prognosis (Shaikhibrahim et al., 2014, 2015; Weiten et al., 2018). Interestingly, in these types of tumours, MT2A expression is also increased and correlates with poor prognosis due to increased resistance to drugs and evasion of apoptosis (Kim et al., 2011; Pal et al., 2014; Raudenska et al., 2015). This suggests that MED15 may be a therapeutic target in these cancers due to its requirement for increased expression of heavy metal stress response genes. 143 In C. elegans, the role of MDT-15 in xenobiotic detoxification is well defined (Goh et al., 2014; Grants et al., 2015; Taubert et al., 2008). In the context of heavy metal stress response, loss of mdt-15 impairs the expression of heavy metal stress response genes (Taubert et al., 2008). Furthermore, the observed impairment of Zn2+ storage and homeostasis in mdt-15 mutant worms (Figure 45) presents a parallel to impaired Zn2+ rich insulin secretory granule pattern in β-cells lacking MED15. This suggests a conserved role for MED15/MDT-15 during Zn2+ homeostasis and high zinc stress response. Interestingly, in C. elegans, the transcription factor HIZR-1 binds MDT-15 in a zinc stimulated fashion, and the expression of heavy metal stress response genes require both proteins (Shomer et al., 2019). The closest orthologue of HIZR-1 in mammals is HNF4⍺ (MODY1) by sequence and PPAR⍺ by structure (Lee et al., 2016; Taubert et al., 2011). As β-cell phenotypes relating to loss of HNF4⍺ are similar to IM15KO (Boj et al., 2010), MED15 may interact with an HNF4-like binding partner for expression of β-cell maturation or expression of heavy metal stress response. Supporting this hypothesis, tail module subunit Med25 interacts with HNF⍺ to enhance insulin secretion (Han et al., 2012), suggesting close proximity to MED15. Closely related to PPAR⍺, Pparβ/δ knockout mice have reduced expression of Slc30a8 and several ZIP genes, suggesting MED15 may interact with PPARβ/δ (Iglesias et al., 2012) to maintain zinc homeostasis. Future experiments to determine if MED15 interacts with any transcription factors in a zinc inducible fashion would provide valuable insight. 144 Chapter 6: Conclusion Research summary Transcription factor action is critical for β-cell maturation and function. In several forms of diabetes, transcription factors are impaired (Lawlor et al., 2017a; Steck and Winter, 2011). The concept that transcriptional coactivators such as Mediator play an important role in this process is fueled by insufficient research on coactivators during β-cell development. To date, very few studies on Mediator subunits in the pancreas exist. Mediator has previously been implicated in cell lineage specification, with certain subunits having important functions through unique subunit-transcription factor interactions (Yin and Wang, 2014). As Mediator is the primary interface for transcription factors and RNA Pol II, this placed Mediator as a potential transcriptional nexus during β-cell development, by modulating the expression of gene programs. MED15 has been implicated in TGF-β signalling and lipid metabolism through its interactions with SMADs and SREBPs, respectively, both of which are important for β-cell biology (Dhawan et al., 2016; Kato et al., 2002; Shimano et al., 2007; Zhao et al., 2014).To date, few studies of Mediator in the pancreas have been performed. One study by previous PhD student Eric Xu has demonstrated the importance of MED15 in β-cell development. In the same IM15KO mouse model, loss of Med15 in β-cells resulted in impaired glucose tolerance, loss of maturation genes, and impaired GSIS (Xu, 2016). This thus warranted further questioning as to how MED15 performs its action: 1. Although deletion of Med15 leads to impaired β-cell functional maturation without affecting β-cell mass, where does the primary defect lie? 2. Which transcription factor(s) and genomic regions does MED15 interact with in β-cells to drive maturation? 145 3. What other functions does MED15 have which pertain to insulin secretion and zinc metabolism? The goal of my thesis was to delineate specific roles for MED15, and therefore Mediator, in β-cell maturation. The importance of these findings will allow better understanding as to how MED15 performs its function in β-cells. MED15 is required for functional maturation In Chapter 3, we examined the role of MED15 during β-cell maturation, hypothesizing that MED15 was a critical regulator of this process. To test this hypothesis, we used the Cre-Lox system to knock out Med15 in β-cell specific fashion. To first characterize MED15 expression in β-cells, we studied the expression of Mediator subunits in human and mouse pancreatic tissues and found MED15 to be enriched in nascent and mature β-cells (Figure 7, 8). MED15 staining in developing mouse pancreata showed MED15 expression in the nucleus of E13.5 β-cells, and primarily cytoplasmic (with some nuclear MED15) in E18.5 β-cells (Figure 8). These findings, taken together with previous known roles for MED15 in the pancreas (Xu, 2016), suggest a specific role for MED15 in early β-cell development which could be further investigated. Although it is unclear how higher expression of MED15 during development may affect function in the context of Mediator, we speculate more MED15 turnover within the complex may increase specific transcription factor interactions. In mature 8-week islets, MED15 expression was found in the nucleus of β-cells, suggesting it may translocate to the nucleus during maturation. Nuclear-cytoplasmic shuttling, which may be impaired during prolonged hyperglycemia, has been observed for several important β-cell transcription factors, such as PDX1, NKX6-1, MAFA, and FOXO1 (Guo et al., 146 2013b; Meur et al., 2011; Semache et al., 2014). In this context, binding partners may recruit MED15 to the nucleus during important β-cell maturation events. In the cytoplasm, MED15 interacts with E3-ubiquitin ligase TRIM11, which leads to its degradation via ubiquitination (Ishikawa et al., 2006). This suggests MED15 may be degraded in the cytoplasm prior to maturation. Collectively, these findings demonstrate that MED15 is expressed in developing and mature insulin secreting β-cells. In individuals with T2D or MODY, the defects in first phase insulin secretion are among the earliest detectable β-cell impairments (Gerich, 2002; Vaxillaire et al., 1999). Resembling this phenotype, we observed severe glucose intolerance and reduced plasma insulin after glucose challenge, alongside impaired first and second phase GSIS in 8-week IM15KO mice (Figure 10A). However, when stimulated with the mitochondrial substrate α-ketoglutarate or depolarized with KCl, IM15KO islets were able to secrete insulin (Figure 10A, C). In rodent β-cells, glucose uptake is performed by Glut2/Slc2a2, the loss of which impairs first phase but does not affect second phase insulin secretion (Guillam et al., 2000). As such, if GLUT2/Slc2a2 was the only defect in IM15KO islets, we would expect intact second phase insulin secretion. Interestingly, Pparβ/δ β-cell specific knockout mice have no changes in first phase, but enhanced second phase insulin secretion (Iglesias et al., 2012), suggesting a possible antagonistic role for MED15 and PPARβ/δ. Taken together, this suggests the primary defect in IM15KO islets is glucose uptake due to loss of Glut2/Slc2a2 expression, though other defects may underlie the observed phenotype. We used 3D mitochondrial imaging to show that mitochondrial networking is impaired in IM15KO islets, with fewer mitochondria yet increased mitochondrial volume (Figure 17). These features have been observed in islets from T2D patients and in islets from diabetic rodents 147 (Anello et al., 2005; Bindokas et al., 2003). Interestingly, impaired expression of maturation transcription factors Nkx6-1 and Errγ in β-cells results in similar mitochondrial phenotypes, with increases in mitochondrial volume (Špaček et al., 2017; Yoshihara et al., 2016a). However, despite the loss of Med15 causing impaired mitochondrial dynamics, insulin secretion upon stimulation with α-ketoglutarate was normal (Figure 10A). This suggests changes in mitochondrial dynamics are not sufficient to explain the observed loss of GSIS, supporting the notion that defective glucose uptake is a main metabolic defect in IM15KO mice. Bulk RNA-seq and immunofluorescence experiments in islets of 8-week old IM15KO mice revealed substantially reduced expression of key maturation genes and markers, such as Ucn3, Slc2a2, Mafa, Slc30a8, Iapp, and G6pc2, among others (Figure 14). It is unclear if expression of Mafa is lost due to impaired glucose uptake, as Mafa expression is glucose responsive (Han et al., 2007; Vanderford et al., 2007). As several of these genes are lost upon deletion of many transcription factors important for β-cell maturation (Gu et al., 2010; Gutiérrez et al., 2016; Jia et al., 2015; Taylor et al., 2013), this suggests a similar regulatory mechanism for maturation involving MED15. Interestingly, aside from Mafa, the expression of β-cell maturation transcription factors Pdx1, Nkx6-1, and NeuroD1 remained unchanged in IM15KO (Figure 14), demonstrating that MED15 function is downstream of transcription factor expression and may serve a modulatory role. Taken together, these findings suggest that the main role of MED15 in β-cell maturation is to induce expression of genes such as Slc2a2, Slc30a8, G6pc2, and Mafa to allow functional maturation and GSIS to occur. 148 MED15 interacts with several β-cell transcription factors Mediator’s interactions with transcription factors at specific regulatory regions are critical for its function (Jeronimo et al., 2016; Malik and Roeder, 2010; Soutourina, 2018; Taatjes, 2010). The finding that MED15 is enriched at actively transcribed chromatin and islet enhancers with β-cell maturation transcription factors: NKX6-1, NKX2-2, PAX6, PDX1, FOXA2, or NEUROD1 suggests MED15 may bind multiple transcription factors. In the context of β-cell maturation, this positions MED15 as a possible nexus for transcriptional modulation. Mechanistically, it is unclear if MED15 functions as a scaffold or via interactions with chromatin looping factors. Among other bound genomic regions, MED15 and MED1 were bound to a high number of intergenic regions without chromatin marks that were tested (Figure 23, 24). This known characteristic of Mediator suggests it may interact with chromatin looping proteins at gene boundaries (Chereji et al., 2017), which loop enhancers to promoters during transcription (Kagey et al., 2010). Preliminary CoIPs for CTCF and MED15 in MIN6 cells suggested an interaction (data not shown), which has previously reported in other cell types (Kagey et al., 2010). In β-cells, loss of chromatin looping factor LDB1 leads to reduced expression of Ucn3, Slc2a2, Mafa, and therefore maturation (Ediger et al., 2017). Interestingly, MED15 binding motif analysis showed similarity to novel chromatin looping factor ZNF143 (Ye et al., 2020). Further experiments to determine which chromatin looping factor MED15 may interact with may prove important in the field of β-cell maturation. Animal knockout studies for various transcription factors have pinpointed roles for specific factors in β-cell maturation (Doyle and Sussel, 2007; Gu et al., 2010; Jia et al., 2015; Taylor et al., 2013; Yoshihara et al., 2016b). Performing CoIP experiments for several of these transcription factors, we found that MED15 binds NEUROD1 and NKX6-1 (Figure 31A, B). It is 149 currently unclear if these interactions are direct and future experiments such as Y2H assays or glutathione-S-transferase (GST) pull-down experiments could be performed to assess this. As MED15 interacts with SMAD2/3/4, we also performed CoIPs for SMAD2 in MIN6 cells and saw an interaction (data not shown). Emerging evidence suggests a role for TGF-β signalling in β-cell maturation (Velazco-Cruz et al., 2019) and as such, this warrants further investigation. Taken together, these data suggest that MED15 interacts with NEUROD1, NKX6-1, and potentially other transcription factors (Figure 47). Figure 47: Summary of MED15’s interactions in β-cells In β-cells, MED15 interacts with NKX6-1 and NEUROD1 to induce β-cell maturation and establish glucose stimulated insulin secretion (GSIS). Possible other interactions which contribute to this process may exist (red circle). 150 MED15 plays a conserved role in heavy metal stress response As we observed decreased expression in zinc transporter Slc30a8 in IM15KO, we performed immunofluorescence, TEM, and zinc staining in these islets (Figure 36-38). These findings suggested reduced total zinc content and impaired mature insulin secretory granules. Because Slc30a8 knockout mice have a similar phenotype yet have sufficient insulin secretion and glucose tolerance (Lemaire et al., 2009), this suggests that glucose uptake is the primary defect in IM15KO islets. Similarly, in Nkx6-1 knockout islets, decreased expression of Slc30a8 and impaired mature insulin secretory granules are observed (Taylor et al., 2013). In the context of mature β-cell function, changes in Zn2+ influx are important (Bellomo et al., 2011). In islets of T2D patients and diabetic mice, expression of MT genes have been implicated as protective (Cai, 2004; Chen et al., 2001). Moreover, single-cell RNA-seq datasets have shown upregulation of Mt1 and Mt2 in mature β-cells, suggesting increased influx of Zn2+ as maturation occurs (Qiu et al., 2017). This has been shown in other models such as stem cell derived β-cells (Ohta et al., 2019). As the pancreas contains one of the highest concentrations of Zn2+ in the body (Bosco et al., 2010), β-cells must protect themselves from heavy metal stress. A role for MDT-15 as a regulator of heavy metal stress response has previously been described (Taubert et al., 2008), but whether this function of MED15 was conserved in mammalian organisms was not known. We showed that, in human and mouse cells, MED15 binds to the enhancers of heavy metal stress response genes in a Zn2+ and Cd2+ dependent fashion (Figure 41, 43). In mdt-15 mutant worms, we observed impaired zinc storage in gut granules, which is the primary mechanism by which worms sequester excess Zn2+ (Figure 45). Interestingly, our lab found that MDT-15 interacts with HIZR-1 in a zinc inducible fashion (Shomer et al., 2019). HIZR-1 shares sequence homology 151 with HNF4α, which suggest that MED15 may interact with HNF4α in a similar fashion. Although CoIP experiments with MED15 and HNF4α in MIN6 cells did not show an interaction (data not shown), addition of Zn2+, or another stimulatory signal such a posttranslational modification, may allow increased binding. As such, further studies to determine which transcription factor MED15 may interact with to induce expression of heavy metal stress response genes will be required. Future directions Experiments to delineate a role for MED15 in mature β-cell function could prove useful. For example, to determine if MED15 is required for maintenance of mature β-cells, the use of the inducible Cre-Lox system would allow deletion after maturation events occur (Sauer, 1998). Additionally, placing these mice on high fat diet would reveal whether MED15 has protective function against diabetic insults. This would determine if loss of MED15 prevents differentiation into mature β-cells or if it renders them susceptible to loss of mature function. Using IM15KO islets, rescue experiments to reintroduce expression GLUT2/Slc2a2 would allow us to determine if glucose uptake is in fact the primary defect. If reintroducing GLUT2 allows IM15KO islets to perform GSIS, this would pinpoint glucose uptake as the key defect in IM15KO β-cells. Although we found that MED15 interacts with NKX6-1 and NEUROD1 (Figure 31), several other transcription factors may also bind MED15, as suggested by my ChIP-seq analysis (Figure 27). To determine which factors interact with MED15, CoIP followed by mass spectrometry could be used to pull down MED15 in β-cells, as previously done in neural stem cells (Quevedo et al., 2019a). This would allow for unbiased identification of MED15’s interacting partners. Furthermore, this approach could be used in human pancreatic islets or stem 152 cell derived β-cells, pinpointing which factors interact with human MED15. As MED15 ChIP-seq data implied chromatin looping as a function, future experiments such as chromosome conformation capture (Mishra and Hawkins, 2017) in CTRL and IM15KO islets could be performed. If MED15 interactions with chromatin looping partners were essential for its function, we would expect loss of enhancer-promoter looping. In parallel, experiments such as HiChIP would allow studying the potential association of MED15 with looped regions (Mumbach et al., 2016). As we were unable to effectively knockdown Med15 in MIN6 cells to determine if it is required for expression of heavy metal stress response genes; generation of a Med15-/- MIN6 cell line by CRISPR-Cas9 genome editing would allow testing of this hypothesis. Moreover, expression of MED15 is upregulated in certain cancers and associated with poor-prognosis (Shaikhibrahim et al., 2015; Weiten et al., 2018). Interestingly, increased metallothionein expression shows a similar correlation in similar cancers (Si and Lang, 2018). As such, experiments to test whether knockdown of MED15 would lead to reduced expression of metallothioneins, and therefore reduced tumourigenesis would be useful. Similarly, chemotherapy agents such as cisplatin and carboplatin contain the heavy metal platinum. Investigating whether increased MED15 expression in cancers correlates with increased resistance to these agents, via upregulation of heavy metal stress response genes may prove clinically useful. 153 References Acevedo, M.L., and Kraus, W.L. (2003). Mediator and p300/CBP-Steroid Receptor Coactivator Complexes Have Distinct Roles, but Function Synergistically, during Estrogen Receptor -Dependent Transcription with Chromatin Templates. Mol. Cell. Biol. 23, 335–348. Ackermann, A.M., Wang, Z., Schug, J., Naji, A., and Kaestner, K.H. (2016). Integration of ATAC-seq and RNA-seq identifies human alpha cell and beta cell signature genes. Mol. Metab. 5, 233–244. Al-Rubeaan, K., Al-Manaa, H.A., Khoja, T.A., Ahmad, N.A., Al-Sharqawi, A.H., Siddiqui, K., Alnaqeb, D., Aburisheh, K.H., Youssef, A.M., Al-Batel, A., et al. (2015). Epidemiology of abnormal glucose metabolism in a country facing its epidemic: SAUDI-DM study. J. Diabetes 7, 622–632. Al-Tabakha, M., and Arida, A. (2008). Recent challenges in insulin delivery systems: A review. Indian J. Pharm. Sci. 70, 278–286. Allen, B.L., and Taatjes, D.J. (2015). The Mediator complex: a central integrator of transcription. Nat. Rev. Mol. Cell Biol. 16, 155–166. Amoasii, L., Holland, W., Sanchez-Ortiz, E., Baskin, K.K., Pearson, M., Burgess, S.C., Nelson, B.R., Bassel-Duby, R., and Olson, E.N. (2016). A MED13-dependent skeletal muscle gene program controls systemic glucose homeostasis and hepatic metabolism. Genes Dev. 30, 434–446. Anders, S., Pyl, P.T., and Huber, W. (2015). HTSeq--a Python framework to work with high-throughput sequencing data. Bioinformatics 31, 166–169. Andrali, S.S., Smapley, M.L., Vanderford, N.L., and Özcan, S. (2008). Glucose regulation of insulin gene expression in pancreatic β-cells. Biochem. J. 415, 1–10. Anello, M., Lupi, R., Spampinato, D., Piro, S., Masini, M., Boggi, U., Del Prato, S., Rabuazzo, A.M., Purrello, F., and Marchetti, P. (2005). Functional and morphological alterations of mitochondria in pancreatic beta cells from type 2 diabetic patients. Diabetologia 48, 282–289. Ansari, S.A., and Morse, R.H. (2013). Mechanisms of Mediator complex action in transcriptional activation. Cell. Mol. Life Sci. 70, 2743–2756. Aranda-Orgilles, B., Saldaña-Meyer, R., Wang, E., Trompouki, E., Fassl, A., Lau, S., Mullenders, J., Rocha, P.P., Raviram, R., Guillamot, M., et al. (2016). MED12 Regulates HSC-Specific Enhancers Independently of Mediator Kinase Activity to Control Hematopoiesis. Cell Stem Cell 19, 784–799. 154 Arda, H.E., Benitez, C.M., and Kim, S.K. (2013). Developmental Cell Gene Regulatory Networks Governing Pancreas Development. Aroda, V.R. (2018). A review of GLP-1 receptor agonists: Evolution and advancement, through the lens of randomised controlled trials. Diabetes, Obes. Metab. 20, 22–33. Asakawa, A., Inui, A., Yuzuriha, H., Ueno, N., Katsuura, G., Fujimiya, M., Fujino, M.A., Niijima, A., Meguid, M.M., and Kasuga, M. (2003). Characterization of the effects of pancreatic polypeptide in the regulation of energy balance. Gastroenterology 124, 1325–1336. Atkinson, M.A., Eisenbarth, G.S., and Michels, A.W. (2014). Type 1 diabetes. Lancet 383, 69–82. Avrahami, D., Li, C., Zhang, J., Schug, J., Avrahami, R., Rao, S., Stadler, M.B., Burger, L., Schübeler, D., Glaser, B., et al. (2015). Aging-dependent demethylation of regulatory elements correlates with chromatin state and improved β cell function. Cell Metab. 22, 619–632. Back, S.H., Kang, S.-W., Han, J., and Chung, H.-T. (2012). Endoplasmic reticulum stress in the β-cell pathogenesis of type 2 diabetes. Exp. Diabetes Res. 2012, 618396. Baker, E.N., Blundell, T.L., Cutfield, J.F., Cutfield, S.M., Dodson, E.J., Dodson, G.G., Hodgkin, D.M., Hubbard, R.E., Isaacs, N.W., and Reynolds, C.D. (1988). The structure of 2Zn pig insulin crystals at 1.5 Å resolution. Philos. Trans. R. Soc. London. B, Biol. Sci. 319, 369–456. Balamurugan, K., Egli, D., Selvaraj, A., Zhang, B., Georgiev, O., and Schaffner, W. (2004). Metal-responsive transcription factor (MTF-1) and heavy metal stress response in Drosophila and mammalian cells: A functional comparison. Biol. Chem. 385, 597–603. Bannister, A.J., and Kouzarides, T. (2011). Regulation of chromatin by histone modifications. Cell Res. 21, 381–395. Banting, F., Best, C., Collip, J., Physicians, W.C.-T.A.A., and 1922, U. (1922). The effect produced on diabetes by extracts of pancreas. Trans Assoc Am Physicians. Baskin, K.K., Grueter, C.E., Kusminski, C.M., Holland, W.L., Bookout, A.L., Satapati, S., Kong, Y.M., Burgess, S.C., Malloy, C.R., Scherer, P.E., et al. (2014). MED 13-dependent signaling from the heart confers leanness by enhancing metabolism in adipose tissue and liver . EMBO Mol. Med. 6, 1610–1621. Bastidas-Ponce, A., Roscioni, S.S., Burtscher, I., Bader, E., Sterr, M., Bakhti, M., and Lickert, H. (2017). Foxa2 and Pdx1 cooperatively regulate postnatal maturation of pancreatic β-cells. Mol. Metab. 6, 524–534. Baumli, S., Hoeppner, S., and Cramer, P. (2005). A conserved mediator hinge revealed in the structure of the MED7·MED21 (Med7·Srb7) heterodimer. J. Biol. Chem. 280, 18171–18178. Bellomo, E.A., Meur, G., and Rutter, G.A. (2011). Glucose regulates free cytosolic Zn 2+ 155 concentration, Slc39 (ZiP), and metallothionein gene expression in primary pancreatic islet β-cells. J. Biol. Chem. 286, 25778–25789. Beyer, K.S., Beauchamp, R.L., Lee, M.F., Gusella, J.F., Näär, A.M., and Ramesh, V. (2007). Mediator subunit MED28 (Magicin) is a repressor of smooth muscle cell differentiation. J. Biol. Chem. 282, 32152–32157. Bilandzic, A., and Rosella, L. (2017). The cost of diabetes in Canada over 10 years: Applying attributable health care costs to a diabetes incidence prediction model. Heal. Promot. Chronic Dis. Prev. Canada 37, 49–53. Bindokas, V.P., Kuznetsov, A., Sreenan, S., Polonsky, K.S., Roe, M.W., and Philipson, L.H. (2003). Visualizing superoxide production in normal and diabetic rat islets of Langerhans. J. Biol. Chem. 278, 9796–9801. Blum, B., Hrvatin, S.S.Š., Schuetz, C., Bonal, C., Rezania, A., and Melton, D.A. (2012). Functional beta-cell maturation is marked by an increased glucose threshold and by expression of urocortin 3. Nat. Biotechnol. 30, 261–264. Boj, S.F., Petrov, D., and Ferrer, J. (2010). Epistasis of Transcriptomes Reveals Synergism between Transcriptional Activators Hnf1α and Hnf4α. PLoS Genet. 6, e1000970. Borikar, S., and Trowbridge, J.J. (2016). The Mediator of Hematopoietic Stem Cell Homeostasis. Cell Stem Cell 19, 677–678. Bosco, M.D., Mohanasundaram, D.M., Drogemuller, C.J., Lang, C.J., Zalewski, P.D., and Coates, P.T. (2010). Zinc and Zinc Transporter Regulation in Pancreatic Islets and the Potential Role of Zinc in Islet Transplantation. Rev. Diabet. Stud. 7, 263–274. Bourbon, H.-M., Aguilera, A., Ansari, A.Z., Asturias, F.J., Berk, A.J., Bjorklund, S., Blackwell, T.K., Borggrefe, T., Carey, M., Carlson, M., et al. (2004). A unified nomenclature for protein subunits of mediator complexes linking transcriptional regulators to RNA polymerase II. Mol. Cell 14, 553–557. Brägelmann, J., Klümper, N., Offermann, A., Von Mässenhausen, A., Böhm, D., Deng, M., Queisser, A., Sanders, C., Syring, I., Merseburger, A.S., et al. (2017). Pan-cancer analysis of the mediator complex transcriptome identifies CDK19 and CDK8 as therapeutic targets in advanced prostate cancer. Clin. Cancer Res. 23, 1829–1840. Brenner, S. (1974). The genetics of Caenorhabditis elegans. Genetics 77, 71–94. Briant, L.J.B., Reinbothe, T.M., Spiliotis, I., Miranda, C., Rodriguez, B., and Rorsman, P. (2018). δ-cells and β-cells are electrically coupled and regulate α-cell activity via somatostatin. J. Physiol. 596, 197–215. Brown, M.L., and Schneyer, A.L. (2010). Emerging roles for the TGFbeta family in pancreatic beta-cell homeostasis. Trends Endocrinol. Metab. 21, 441–448. 156 Brzóska, M.M., and Moniuszko-Jakoniuk, J. (2001). Interactions between cadmium and zinc in the organism. Food Chem. Toxicol. 39, 967–980. Byun, H.R., Choi, J.A., and Koh, J.Y. (2014). The role of metallothionein-3 in streptozotocin-induced beta-islet cell death and diabetes in mice. Metallomics 6, 1748–1757. Cai, L. (2004). Metallothionein as an Adaptive Protein Prevents Diabetes and its Toxicity. Nonlinearity Biol. Toxicol. Med. 2, 154014204904643. Calderon, B., and Sacks, D.B. (2014). Islet Autoantibodies and Type 1 Diabetes: Does the Evidence Support Screening? Clin. Chem. 60, 438–440. Campbell, S.A. (2019). Exploring Trithorax Group complex activity in pancreas and endocrine progenitor cell development. Campbell, S.A., McDonald, C.L., Krentz, N.A.J., Lynn, F.C., and Hoffman, B.G. (2019). TrxG Complex Catalytic and Non-catalytic Activity Play Distinct Roles in Pancreas Progenitor Specification and Differentiation. Cell Rep. 28, 1830-1844.e6. Canfield, W.K., Michael Hambidge, K., and Johnson, L.K. (1984). Zinc nutriture in type i diabetes mellitus: relationship to growth measures and metabolic control. J. Pediatr. Gastroenterol. Nutr. 3, 577–584. Carrero, J.A., Ferris, S.T., and Unanue, E.R. (2016). Macrophages and dendritic cells in islets of Langerhans in diabetic autoimmunity: a lesson on cell interactions in a mini-organ. Curr. Opin. Immunol. 43, 54–59. Cebola, I. (2019). Pancreatic Islet Transcriptional Enhancers and Diabetes. Curr. Diab. Rep. 19, 1–12. Cerolsaletti, K., Hao, W., and Greenbaum, C.J. (2019). Genetics coming of age in type 1 diabetes. Diabetes Care 42, 189–191. Cevher, M.A., Shi, Y., Li, D., Chait, B.T., Malik, S., and Roeder, R.G. (2014). Reconstitution of active human core Mediator complex reveals a critical role of the MED14 subunit. Nat. Struct. Mol. Biol. 21, 1028–1034. Chang-Chen, K.J., Mullur, R., and Bernal-Mizrachi, E. (2008). β-cell failure as a complication of diabetes. Rev. Endocr. Metab. Disord. 9, 329–343. Chaudhry, A., Shi, R., and Luciani, D.S. (2019). A pipeline for multidimensional confocal analysis of mitochondrial morphology, function and dynamics in pancreatic β-cells. BioRxiv 687749. Chaudhury, A., Duvoor, C., Reddy Dendi, V.S., Kraleti, S., Chada, A., Ravilla, R., Marco, A., Shekhawat, N.S., Montales, M.T., Kuriakose, K., et al. (2017). Clinical Review of Antidiabetic Drugs: Implications for Type 2 Diabetes Mellitus Management. Front. Endocrinol. (Lausanne). 157 8. Chen, C., Cohrs, C.M., Stertmann, J., Bozsak, R., and Speier, S. (2017). Human beta cell mass and function in diabetes: Recent advances in knowledge and technologies to understand disease pathogenesis. Mol. Metab. 6, 943–957. Chen, E.Y., Tan, C.M., Kou, Y., Duan, Q., Wang, Z., Meirelles, G. V., Clark, N.R., and Ma’ayan, A. (2013). Enrichr: Interactive and collaborative HTML5 gene list enrichment analysis tool. BMC Bioinformatics 14. Chen, H., Carlson, E.C., Pellet, L., Moritz, J.T., and Epstein, P.N. (2001). Overexpression of Metallothionein in Pancreatic β-Cells Reduces Streptozotocin-Induced DNA Damage and Diabetes. Diabetes 50, 2040–2046. Chen, S., Han, J., and Liu, Y. (2015). Dual Opposing Roles of Metallothionein Overexpression in C57BL/6J Mouse Pancreatic β-Cells. PLoS One 10, e0137583. Chen, W., Rogatsky, I., and Garabedian, M.J. (2006). MED14 and MED1 differentially regulate target-specific gene activation by the glucocorticoid receptor. Mol. Endocrinol. 20, 560–572. Chen, W., Yang, Q., and Roeder, R.G. (2009). Dynamic Interactions and Cooperative Functions of PGC-1α and MED1 in TRα-Mediated Activation of the Brown-Fat-Specific UCP-1 Gene. Mol. Cell 35, 755–768. Chereji, R. V, Bharatula, V., Elfving, N., Blomberg, J., Larsson, M., Morozov, A. V, Broach, J.R., and Björklund, S. (2017). Mediator binds to boundaries of chromosomal interaction domains and to proteins involved in DNA looping, RNA metabolism, chromatin remodeling, and actin assembly. Nucleic Acids Res. 45, 8806–8821. Chimienti, F., Devergnas, S., Favier, A., and Seve, M. (2004). Identification and cloning of a β-cell-specific zinc transporter, ZnT-8, localized into insulin secretory granules. Diabetes 53, 2330–2337. Collombat, P., Mansouri, A., Hecksher-Sørensen, J., Serup, P., Krull, J., Gradwohl, G., and Gruss, P. (2003). Opposing actions of Arx and Pax4 in endocrine pancreas development. Genes Dev. 17, 2591–2603. Conrad, E., Stein, R., and Hunter, C.S. (2014). Revealing transcription factors during human pancreatic β cell development. Trends Endocrinol. Metab. 25, 407–414. Courtade, J.A., Wang, E.Y., Yen, P., Dai, D.L., Soukhatcheva, G., Orban, P.C., and Verchere, C.B. (2017). Loss of prohormone convertase 2 promotes beta cell dysfunction in a rodent transplant model expressing human pro-islet amyloid polypeptide. Diabetologia 60, 453–463. Crowshoe, L., Dannenbaum, D., Green, M., Ma, R.H., Naqshbandi, M., Msc, H., and Toth, E. (2018). 2018 Clinical Practice Guidelines Type 2 Diabetes and Indigenous Peoples Diabetes Canada Clinical Practice Guidelines Expert Committee. 158 Davidson, H.W., Wenzlau, J.M., and O’Brien, R.M. (2014). Zinc transporter 8 (ZnT8) and β cell function. Trends Endocrinol. Metab. 25, 415–424. Dean, P.M., and Matthews, E.K. (1968). Electrical activity in pancreatic islet cells. Nature 219, 389–390. DeFronzo, R.A., Ferrannini, E., Groop, L., Henry, R.R., Herman, W.H., Holst, J.J., Hu, F.B., Kahn, C.R., Raz, I., Shulman, G.I., et al. (2015). Type 2 diabetes mellitus. Nat. Rev. Dis. Prim. 1, 1–22. Dezaki, K., Hosoda, H., Kakei, M., Hashiguchi, S., Watanabe, M., Kangawa, K., and Yada, T. (2004). Endogenous ghrelin in pancreatic islets restricts insulin release by attenuating Ca2+ signaling in β-cells: Implication in the glycemic control in rodents. Diabetes 53, 3142–3151. Dhawan, S., Georgia, S., Tschen, S. ing, Fan, G., and Bhushan, A. (2011). Pancreatic β Cell Identity Is Maintained by DNA Methylation-Mediated Repression of Arx. Dev. Cell 20, 419–429. Dhawan, S., Dirice, E., Kulkarni, R.N., and Bhushan, A. (2016). Inhibition of TGF-beta signaling promotes human pancreatic beta cell replication. Diabetes. DiGruccio, M.R., Mawla, A.M., Donaldson, C.J., Noguchi, G.M., Vaughan, J., Cowing-Zitron, C., van der Meulen, T., and Huising, M.O. (2016). Comprehensive alpha, beta and delta cell transcriptomes reveal that ghrelin selectively activates delta cells and promotes somatostatin release from pancreatic islets. Mol. Metab. 5, 449–458. Ding, N., Tomomori-Sato, C., Sato, S., Conaway, R.C., Conaway, J.W., and Boyer, T.G. (2009). MED19 and MED26 are synergistic functional targets of the RE1 silencing transcription factor in epigenetic silencing of neuronal gene expression. J. Biol. Chem. 284, 2648–2656. Diraison, F., Ravier, M.A., Richards, S.K., Smith, R.M., Shimano, H., and Rutter, G.A. (2008). SREBP1 is required for the induction by glucose of pancreatic beta-cell genes involved in glucose sensing. J. Lipid Res. 49, 814–822. Dobin, A., Davis, C.A., Schlesinger, F., Drenkow, J., Zaleski, C., Jha, S., Batut, P., Chaisson, M., and Gingeras, T.R. (2013). STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21. Dolenšek, J., Slak Rupnik, M., Stožer, A., Dolen sek, J., and Sto zer, A. (2015). Structural similarities and differences between the human and the mouse pancreas. Donelan, W., Koya, V., Li, S.-W.W., and Yang, L.-J.J. (2010). Distinct regulation of hepatic nuclear factor 1alpha by NKX6.1 in pancreatic beta cells. J. Biol. Chem. 285, 12181–12189. Dotson, M.R., Yuan, C.X., Roeder, R.G., Myers, L.C., Gustafsson, C.M., Jiang, Y.W., Li, Y., Kornberg, R.D., and Asturias, F.J. (2000). Structural organization of yeast and mammalian mediator complexes. Proc. Natl. Acad. Sci. U. S. A. 97, 14307–14310. 159 Dover, E.N., Patel, N.Y., and Stýblo, M. (2018). Impact of in vitro heavy metal exposure on pancreatic β-cell function. Toxicol. Lett. 299, 137–144. Doyle, M.J., and Sussel, L. (2007). Nkx2.2 regulates β-cell function in the mature islet. Diabetes 56, 1999–2007. Ediger, B.N., Lim, H.W., Juliana, C., Groff, D.N., Williams, L.Q.T., Dominguez, G., Liu, J.H., Taylor, B.L., Walp, E.R., Kameswaran, V., et al. (2017). LIM domain-binding 1 maintains the terminally differentiated state of pancreatic β cells. J. Clin. Invest. 127, 215–229. Eeckhoute, J., Formstecher, P., and Laine, B. (2004). Hepatocyte Nuclear Factor 4a enhances the Hepatocyte Nuclear Factor 1a-mediated activation of transcription. El-Gohary, Y., Tulachan, S., Guo, P., Welsh, C., Wiersch, J., Prasadan, K., Paredes, J., Shiota, C., Xiao, X., Wada, Y., et al. (2013). Smad signaling pathways regulate pancreatic endocrine development. Dev. Biol. 378, 83–93. Emdin, S.O., Dodson, G.G., Cutfield, J.M., and Cutfield, S.M. (1980). Role of zinc in insulin biosynthesis - Some possible zinc-insulin interactions in the pancreatic B-cell. Diabetologia 19, 174–182. von Engelhardt, D. (1989). Outlines of Historical Development. In Diabetes Its Medical and Cultural History, (Springer Berlin Heidelberg), pp. 3–12. Ernst, J., and Kellis, M. (2017). Chromatin-state discovery and genome annotation with ChromHMM. Nat. Protoc. 12, 2478–2492. Ettinger, R.A., and Kwok, W.W. (1998). A Peptide Binding Motif for HLA-DQA1*0102/DQB1*0602, the Class II MHC Molecule Associated with Dominant Protection in Insulin-Dependent Diabetes Mellitus. J. Immunol. 151, 3627–3637. Fava, E., Dehghany, J., Ouwendijk, J., Müller, A., Niederlein, A., Verkade, P., Meyer-Hermann, M., and Solimena, M. (2012). Novel standards in the measurement of rat insulin granules combining electron microscopy, high-content image analysis and in silico modelling. Diabetologia 55, 1013–1023. Firdous, P., Nissar, K., Ali, S., Ganai, B.A., Shabir, U., Hassan, T., and Masoodi, S.R. (2018). Genetic testing of maturity-onset diabetes of the young current status and future perspectives. Front. Endocrinol. (Lausanne). 9, 253. Firestein, R., Bass, A.J., Kim, S.Y., Dunn, I.F., Silver, S.J., Guney, I., Freed, E., Ligon, A.H., Vena, N., Ogino, S., et al. (2008). CDK8 is a colorectal cancer oncogene that regulates β-catenin activity. Nature 455, 547–551. Flannick, J., Thorleifsson, G., Beer, N.L., Jacobs, S.B.R., Grarup, N., Burtt, N.P., Mahajan, A., Fuchsberger, C., Atzmon, G., Benediktsson, R., et al. (2014). Loss-of-function mutations in SLC30A8 protect against type 2 diabetes. Nat. Genet. 46, 357–363. 160 Flannick, J., Johansson, S., and Njølstad, P.R. (2016). Common and rare forms of diabetes mellitus: Towards a continuum of diabetes subtypes. Nat. Rev. Endocrinol. 12, 394–406. Fondell, J.D., Ge, H., and Roeder, R.G. (1996). Ligand induction of a transcriptionally active thyroid hormone receptor coactivator complex. Proc. Natl. Acad. Sci. U. S. A. 93, 8329–8333. Francois, M.E., and Little, J.P. (2015). Effectiveness and safety of high-intensity interval training in patients with type 2 diabetes. Diabetes Spectr. 28, 39–44. Fuchsberger, C., Flannick, J., Teslovich, T.M., Mahajan, A., Agarwala, V., Gaulton, K.J., Ma, C., Fontanillas, P., Moutsianas, L., McCarthy, D.J., et al. (2016). The genetic architecture of type 2 diabetes. Nature 536, 41–47. Gaisano, H.Y., MacDonald, P.E., and Vranic, M. (2012). Glucagon secretion and signaling in the development of diabetes. Front. Physiol. 3 SEP, 349. Ganguly, K., Levänen, B., Palmberg, L., Åkesson, A., and Lindén, A. (2018). Cadmium in tobacco smokers: A neglected link to lung disease? Eur. Respir. Rev. 27. Ge, K., Guermah, M., Yuan, C.-X., Ito, M., Wallberg, A.E., Spiegelman, B.M., and Roeder, R.G. (2002). Transcription coactivator TRAP220 is required for PPARγ2-stimulated adipogenesis. Nature 417, 563–567. Ge, K., Cho, Y.-W., Guo, H., Hong, T.B., Guermah, M., Ito, M., Yu, H., Kalkum, M., and Roeder, R.G. (2008). Alternative Mechanisms by Which Mediator Subunit MED1/TRAP220 Regulates Peroxisome Proliferator-Activated Receptor γ-Stimulated Adipogenesis and Target Gene Expression. Mol. Cell. Biol. 28, 1081–1091. Gerber, P.A., and Rutter, G.A. (2017). The Role of Oxidative Stress and Hypoxia in Pancreatic Beta-Cell Dysfunction in Diabetes Mellitus. Antioxid. Redox Signal. 26, 501–518. Gerich, J.E. (2002). Is reduced first-phase insulin release the earliest detectable abnormality in individuals destined to develop type 2 diabetes? In Diabetes, (American Diabetes Association), pp. S117–S121. Gertz, J., Savic, D., Varley, K.E., Partridge, E.C., Safi, A., Jain, P., Cooper, G.M., Reddy, T.E., Crawford, G.E., and Myers, R.M. (2013). Distinct properties of cell-type-specific and shared transcription factor binding sites. Mol. Cell 52, 25–36. Ghaye, A.P., Bergemann, D., Tarifeño-Saldivia, E., Flasse, L.C., Von Berg, V., Peers, B., Voz, M.L., and Manfroid, I. (2015). Progenitor potential of nkx6.1-expressing cells throughout zebrafish life and during beta cell regeneration. BMC Biol. 13, 70. Gianani, R., Campbell-Thompson, M., Sarkar, S.A., Wasserfall, C., Pugliese, A., Solis, J.M., Kent, S.C., Hering, B.J., West, E., Steck, A., et al. (2010). Dimorphic histopathology of long-standing childhood-onset diabetes. Diabetologia 53, 690–698. 161 Gilmour, J., Assi, S.A., Noailles, L., Lichtinger, M., Obier, N., and Bonifer, C. (2018). The Co-operation of RUNX1 with LDB1, CDK9 and BRD4 Drives Transcription Factor Complex Relocation during Haematopoietic Specification. Sci. Rep. 8, 1–14. Goh, G.Y.S., Martelli, K.L., Parhar, K.S., Kwong, A.W.L., Wong, M.A., Mah, A., Hou, N.S., and Taubert, S. (2014). The conserved Mediator subunit MDT-15 is required for oxidative stress responses in Caenorhabditis elegans. Aging Cell 13, 70–79. Goh, G.Y.S., Winter, J.J., Bhanshali, F., Doering, K.R.S., Lai, R., Lee, K., Veal, E.A., and Taubert, S. (2018). NHR-49/HNF4 integrates regulation of fatty acid metabolism with a protective transcriptional response to oxidative stress and fasting. Aging Cell 17, e12743. Gorasia, D.G., Dudek, N.L., Veith, P.D., Shankar, R., Safavi-Hemami, H., Williamson, N.A., Reynolds, E.C., Hubbard, M.J., and Purcell, A.W. (2015). Pancreatic Beta Cells Are Highly Susceptible to Oxidative and ER Stresses during the Development of Diabetes. J. Proteome Res. 14, 688–699. Gradwohl, G., Dierich, A., LeMeur, M., and Guillemot, F. (2000). neurogenin3 is required for the development of the four endocrine cell lineages of the pancreas. Proc. Natl. Acad. Sci. U. S. A. 97, 1607–1611. Granata, R., Volante, M., Settanni, F., Gauna, C., Ghé, C., Annunziata, M., Deidda, B., Gesmundo, I., Abribat, T., Van Der Lely, A.J., et al. (2010). Unacylated ghrelin and obestatin increase islet cell mass and prevent diabetes in streptozotocin-treated newborn rats. J. Mol. Endocrinol. 45, 9–17. Grants, J.M., Goh, G.Y.S., and Taubert, S. (2015). The Mediator complex of Caenorhabditis elegans: insights into the developmental and physiological roles of a conserved transcriptional coregulator. Nucleic Acids Res. 43, 2442–2453. Grants, J.M., Ying, L.T.L., Yoda, A., You, C.C., Okano, H., Sawa, H., and Taubert, S. (2016). The mediator kinase module restrains epidermal growth factor receptor signaling and represses vulval cell fate specification in Caenorhabditis elegans. Genetics 202, 583–599. Grasso, S., Messina, A., Saporito, N., and Reitano, G. (1968). Serum-insulin response to glucose and aminoacids in the premature infant. Lancet 2, 755–756. Greenwald, W.W., Chiou, J., Yan, J., Qiu, Y., Dai, N., Wang, A., Nariai, N., Aylward, A., Han, J.Y., Kadakia, N., et al. (2019). Pancreatic islet chromatin accessibility and conformation reveals distal enhancer networks of type 2 diabetes risk. Nat. Commun. 10, 1–12. Grodsky, G.M., and Bennett, L.L. (1966). Cation requirements for insulin secretion in the isolated perfused pancreas. Diabetes 15, 910–913. Grodsky, G.M., Batts, A.A., Bennett, L.L., Vcella, C., McWilliams, N.B., and Smith, D.F. (1963). Effects of carbohydrates on secretion of insulin from isolated rat pancreas. Am. J. Physiol. 205, 638–644. 162 Grontved, L., Madsen, M.S., Boergesen, M., Roeder, R.G., and Mandrup, S. (2010). MED14 Tethers Mediator to the N-Terminal Domain of Peroxisome Proliferator-Activated Receptor and Is Required for Full Transcriptional Activity and Adipogenesis. Mol. Cell. Biol. 30, 2155–2169. Grueter, C.E., Van Rooij, E., Johnson, B.A., Deleon, S.M., Sutherland, L.B., Qi, X., Gautron, L., Elmquist, J.K., Bassel-Duby, R., and Olson, E.N. (2012). A cardiac MicroRNA governs systemic energy homeostasis by regulation of MED13. Cell 149, 671–683. Gu, C., Stein, G.H., Pan, N., Goebbels, S., Hörnberg, H., Nave, K.-A., Herrera, P., White, P., Kaestner, K.H., Sussel, L., et al. (2010). Pancreatic beta cells require NeuroD to achieve and maintain functional maturity. Cell Metab. 11, 298–310. Guillam, M.T., Dupraz, P., and Thorens, B. (2000). Glucose uptake, utilization, and signaling in GLUT2-null islets. Diabetes 49, 1485–1491. Guo, L., Lichten, L.A., Ryu, M.S., Liuzzi, J.P., Wang, F., and Cousins, R.J. (2010). STAT5-glucocorticoid receptor interaction and MTF-1 regulate the expression of ZnT2 (Slc30a2) in pancreatic acinar cells. Proc. Natl. Acad. Sci. U. S. A. 107, 2818–2823. Guo, S., Dai, C., Guo, M., Taylor, B., Harmon, J.S., Sander, M., Robertson, R.P., Powers, A.C., and Stein, R. (2013a). Inactivation of specific β cell transcription factors in type 2 diabetes. J. Clin. Invest. 123, 3305–3316. Guo, S., Dai, C., Guo, M., Taylor, B., Harmon, J.S., Sander, M., Robertson, R.P., Powers, A.C., and Stein, R. (2013b). Inactivation of specific β cell transcription factors in type 2 diabetes. J. Clin. Invest. 123, 3305–3316. Gutiérrez, G.D., Bender, A.S., Cirulli, V., Mastracci, T.L., Kelly, S.M., Tsirigos, A., Kaestner, K.H., and Sussel, L. (2016). Pancreatic β cell identity requires continual repression of non–β cell programs. J. Clin. Invest. 127, 244–259. Haberle, V., Arnold, C.D., Pagani, M., Rath, M., Schernhuber, K., and Stark, A. (2019). Transcriptional cofactors display specificity for distinct types of core promoters. Nature 570, 122–126. Hall, A.G., Kelleher, S.L., Lönnerdal, B., and Philipps, A.F. (2005). A graded model of dietary zinc deficiency: Effects on growth, insulin-like growth factor-I, and the glucose/insulin axis in weanling rats. J. Pediatr. Gastroenterol. Nutr. 41, 72–80. Han, E.H., Rha, G.B., and Chi, Y.-I. (2012). MED25 Is a Mediator Component of HNF4α-Driven Transcription Leading to Insulin Secretion in Pancreatic Beta-Cells. PLoS One 7, e44007. Han, S., Aramata, S., Yasuda, K., and Kataoka, K. (2007). MafA Stability in Pancreatic β Cells Is Regulated by Glucose and Is Dependent on Its Constitutive Phosphorylation at Multiple Sites by Glycogen Synthase Kinase 3. Mol. Cell. Biol. 27, 6593–6605. 163 Hansen, W.A., Christie, M.R., Kahn, R., Norgaard, A., Abel, I., Petersen, A.M., Jorgensen, D.W., Baekkeskov, S., Nielsen, J.H., Lernmark, A., et al. (1989). Supravital dithizone staining in the isolation of human and rat pancreatic islets. Diabetes Res. 10, 53–57. Harding, J.L., Pavkov, M.E., Magliano, D.J., Shaw, J.E., and Gregg, E.W. (2019). Global trends in diabetes complications: a review of current evidence. Diabetologia 62, 3–16. Harjutsalo, V., Sjöberg, L., and Tuomilehto, J. (2008). Time trends in the incidence of type 1 diabetes in Finnish children: a cohort study. Lancet 371, 1777–1782. Harms, M.J., Lim, H.W., Ho, Y., Shapira, S.N., Ishibashi, J., Rajakumari, S., Steger, D.J., Lazar, M.A., Won, K.J., and Seale, P. (2015). PRDM16 binds MED1 and controls chromatin architecture to determine a brown fat transcriptional program. Genes Dev. 29, 298–307. Harper, T.M., and Taatjes, D.J. (2018). The complex structure and function of Mediator. J. Biol. Chem. 293, 13778–13785. Haumaitre, C., Lenoir, O., and Scharfmann, R. (2008). Histone Deacetylase Inhibitors Modify Pancreatic Cell Fate Determination and Amplify Endocrine Progenitors. Mol. Cell. Biol. 28, 6373–6383. Herrera, P.L., Huarte, J., Sanvito, F., Meda, P., Orci, L., and Vassalli, J.D. (1991). Embryogenesis of the murine endocrine pancreas; early expression of pancreatic polypeptide gene. Development 113. Hittelman, A.B. (1999). Differential regulation of glucocorticoid receptor transcriptional activation via AF-1-associated proteins. EMBO J. 18, 5380–5388. Hogrebe, N.J., Augsornworawat, P., Maxwell, K.G., Velazco-Cruz, L., and Millman, J.R. (2020). Targeting the cytoskeleton to direct pancreatic differentiation of human pluripotent stem cells. Nat. Biotechnol. 38, 460–470. Hou, J.C., Min, L., and Pessin, J.E. (2009). Insulin granule biogenesis, trafficking and exocytosis. Vitam. Horm. 80, 473–506. Hou, L., Tang, S., Wu, B.J., Ong, K.L., Westerterp, M., Barter, P.J., Cochran, B.J., Tabet, F., and Rye, K.A. (2019). Apolipoprotein A-l improves pancreatic β-cell function independent of the ATP-binding cassette transporters ABCA1 and ABCG1. FASEB J. 33, 8479–8489. Hou, N.S., Gutschmidt, A., Choi, D.Y., Pather, K., Shi, X., Watts, J.L., Hoppe, T., and Taubert, S. (2014). Activation of the endoplasmic reticulum unfolded protein response by lipid disequilibrium without disturbed proteostasis in vivo. Proc. Natl. Acad. Sci. 111, E2271–E2280. Hu, Q., D’Amora, D.R., MacNeil, L.T., Walhout, A.J.M., and Kubiseski, T.J. (2018). The caenorhabditis elegans oxidative stress response requires the NHR-49 transcription factor. G3 Genes, Genomes, Genet. 8, 3857–3863. 164 Huang, C., Walker, E.M., Dadi, P.K., Hu, R., Xu, Y., Zhang, W., Sanavia, T., Mun, J., Liu, J., Nair, G.G., et al. (2018). Synaptotagmin 4 Regulates Pancreatic β Cell Maturation by Modulating the Ca2+ Sensitivity of Insulin Secretion Vesicles. Dev. Cell 45, 347-361.e5. Huber, A.M., and Gershoff, S.N. (1973). Effect of zinc deficiency in rats on insulin release from the pancreas. J. Nutr. 103, 1739–1744. Husain, S., and Thrower, E. (2009). Molecular and cellular regulation of pancreatic acinar cell function. Curr. Opin. Gastroenterol. 25, 466–471. Iglesias, J., Barg, S., Vallois, D., Lahiri, S., Roger, C., Yessoufou, A., Pradevand, S., McDonald, A., Bonal, C., Reimann, F., et al. (2012). PPARβ/δ affects pancreatic β cell mass and insulin secretion in mice. J. Clin. Invest. 122, 4105–4117. Insel, R.A., Dunne, J.L., Atkinson, M.A., Chiang, J.L., Dabelea, D., Gottlieb, P.A., Greenbaum, C.J., Herold, K.C., Krischer, J.P., Lernmark, A., et al. (2015). Staging presymptomatic type 1 diabetes: A scientific statement of jdrf, the endocrine society, and the American diabetes association. Diabetes Care 38, 1964–1974. Ishihara, H., Maechler, P., Gjinovci, A., Herrera, P.L., and Wollheim, C.B. (2003). Islet beta-cell secretion determines glucagon release from neighbouring alpha-cells. Nat Cell Biol 5, 330–335. Ishikawa, H., Tachikawa, H., Miura, Y., and Takahashi, N. (2006). TRIM11 binds to and destabilizes a key component of the activator-mediated cofactor complex (ARC105) through the ubiquitin-proteasome system. FEBS Lett. 580, 4784–4792. Ito, M., Yuan, C.X., Okano, H.J., Darnell, R.B., and Roeder, R.G. (2000). Involvement of the TRAP220 component of the TRAP/SMCC coactivator complex in embryonic development and thyroid hormone action. Mol. Cell 5, 683–693. Jayaraman, S. (2008). A novel method for the detection of viable human pancreatic beta cells by flow cytometry using fluorophores that selectively detect labile zinc, mitochondrial membrane potential and protein thiols. Cytom. Part A 73A, 615–625. Jeronimo, C., Langelier, M.F., Bataille, A.R., Pascal, J.M., Pugh, B.F., and Robert, F. (2016). Tail and Kinase Modules Differently Regulate Core Mediator Recruitment and Function In Vivo. Mol. Cell 64, 455–466. Jia, S., Ivanov, A., Blasevic, D., Muller, T., Purfurst, B., Sun, W., Chen, W., Poy, M.N., Rajewsky, N., and Birchmeier, C. (2015). Insm1 cooperates with Neurod1 and Foxa2 to maintain mature pancreatic -cell function. EMBO J. 34, 1417–1433. Jia, Y., Qi, C., Kashireddi, P., Surapureddi, S., Zhu, Y.J., Rao, M.S., Le Roith, D., Chambon, P., Gonzalez, F.J., and Reddy, J.K. (2004). Transcription coactivator PBP, the peroxisome proliferator-activated receptor (PPAR)-binding protein, is required for PPARα-regulated gene expression in liver. J. Biol. Chem. 279, 24427–24434. 165 Jia, Y., Viswakarma, N., Fu, T., Yu, S., Rao, M.S., Borensztajn, J., and Reddy, J.K. (2009). Conditional ablation of mediator subunit MED1 (MED1/PPARBP) gene in mouse liver attenuates glucocorticoid receptor agonist dexamethasone-induced hepatic steatosis. Gene Expr. 14, 291–306. Jia, Y., Viswakarma, N., and Reddy, J.K. (2014). Med1 subunit of the mediator complex in nuclear receptor-regulated energy metabolism, liver regeneration, and hepatocarcinogenesis. Gene Expr. 16, 63–75. Johnson, D.S., Mortazavi, A., Myers, R.M., and Wold, B. (2007). Genome-Wide Mapping of in Vivo Protein-DNA Interactions. Science (80-. ). 316, 1497–1502. Jolles, S. (2002). Paul Langerhans. J. Clin. Pathol. 55, 243. Jou, M.-Y., Philipps, A.F., and Lönnerdal, B. (2010). Maternal Zinc Deficiency in Rats Affects Growth and Glucose Metabolism in the Offspring by Inducing Insulin Resistance Postnatally. J. Nutr. 140, 1621–1627. Kagey, M.H., Newman, J.J., Bilodeau, S., Zhan, Y., Orlando, D.A., van Berkum, N.L., Ebmeier, C.C., Goossens, J., Rahl, P.B., Levine, S.S., et al. (2010). Mediator and cohesin connect gene expression and chromatin architecture. Nature 467, 430–435. Kahn, S.E., Cooper, M.E., and Del Prato, S. (2014). Pathophysiology and treatment of type 2 diabetes: perspectives on the past, present, and future. Lancet (London, England) 383, 1068–1083. Kaneto, H., and Matsuoka, T. (2015). Role of Pancreatic Transcription Factors in Maintenance of Mature β-Cell Function. Int. J. Mol. Sci. 16, 6281–6297. Karamanou, M. (2016). Milestones in the history of diabetes mellitus: The main contributors. World J. Diabetes 7, 1. Kataoka, K., Han, S. iee, Shioda, S., Hirai, M., Nishizawa, M., and Handa, H. (2002). MafA is a glucose-regulated and pancreatic β-cell-specific transcriptional activator for the insulin gene. J. Biol. Chem. 277, 49903–49910. Kato, Y., Habas, R., Katsuyama, Y., Näär, A.M., and He, X. (2002). A component of the ARC/Mediator complex required for TGFβ/Nodal signalling. Nature 418, 641–646. Katsarou, A., Gudbjörnsdottir, S., Rawshani, A., Dabelea, D., Bonifacio, E., Anderson, B.J., Jacobsen, L.M., Schatz, D.A., and Lernmark, A. (2017). Type 1 diabetes mellitus. Nat. Rev. Dis. Prim. 3, 1–17. Kaufman, B.A., Li, C., and Soleimanpour, S.A. (2015). Mitochondrial regulation of β-cell function: Maintaining the momentum for insulin release. Mol. Aspects Med. 42, 91–104. Kelleher, R.J., Flanagan, P.M., and Kornberg, R.D. (1990). A novel mediator between activator 166 proteins and the RNA polymerase II transcription apparatus. Cell 61, 1209–1215. Kelleher, S.L., McCormick, N.H., Velasquez, V., and Lopez, V. (2011). Zinc in Specialized Secretory Tissues: Roles in the Pancreas, Prostate, and Mammary Gland. Adv. Nutr. 2, 101–111. Kemmeren, P., Sameith, K., van de Pasch, L.A.L., Benschop, J.J., Lenstra, T.L., Margaritis, T., O’Duibhir, E., Apweiler, E., van Wageningen, S., Ko, C.W., et al. (2014). Large-Scale Genetic Perturbations Reveal Regulatory Networks and an Abundance of Gene-Specific Repressors. Cell 157, 740–752. El Khattabi, L., Zhao, H., Kalchschmidt, J., Young, N., Jung, S., Van Blerkom, P., Kieffer-Kwon, P., Kieffer-Kwon, K.R., Park, S., Wang, X., et al. (2019). A Pliable Mediator Acts as a Functional Rather Than an Architectural Bridge between Promoters and Enhancers. Cell 178, 1145-1158.e20. Khazrai, Y.M., Defeudis, G., and Pozzilli, P. (2014). Effect of diet on type 2 diabetes mellitus: a review. Diabetes. Metab. Res. Rev. 30, 24–33. Kido, Y., Nakae, J., and Accili, D. (2001). The Insulin Receptor and Its Cellular Targets 1 . J. Clin. Endocrinol. Metab. 86, 972–979. Kim, S.H. (2015). Maturity-onset diabetes of the young: What do clinicians need to know? Diabetes Metab. J. 39, 468–477. Kim, H.G., Kim, J.Y., Han, E.H., Hwang, Y.P., Choi, J.H., Park, B.H., and Jeong, H.G. (2011). Metallothionein-2A overexpression increases the expression of matrix metalloproteinase-9 and invasion of breast cancer cells. FEBS Lett. 585, 421–428. Kinlaw, W.B., Levine, A.S., Morley, J.E., Silvis, S.E., and McClain, C.J. (1983). Abnormal zinc metabolism in type II diabetes mellitus. Am. J. Med. 75, 273–277. Knip, M., Veijola, R., Virtanen, S.M., Hyöty, H., Vaarala, O., and Åkerblom, H.K. (2005). Environmental triggers and determinants of type 1 diabetes. Diabetes 54, S125–S136. Knoell, D.L., Smith, D., Bao, S., Sapkota, M., Wyatt, T.A., Zweier, J.L., Flury, J., Borchers, M.T., and Knutson, M. (2020). Imbalance in zinc homeostasis enhances lung Tissue Loss following cigarette smoke exposure. J. Trace Elem. Med. Biol. 60. Knuesel, M.T., Meyer, K.D., Bernecky, C., and Taatjes, D.J. (2009). The human CDK8 subcomplex is a molecular switch that controls Mediator coactivator function. Genes Dev. 23, 439–451. Komatsu, M., Takei, M., Ishii, H., and Sato, Y. (2013). Glucose-stimulated insulin secretion: A newer perspective. J. Diabetes Investig. 4, 511–516. Koussounadis, A., Langdon, S.P., Um, I.H., Harrison, D.J., and Smith, V.A. (2015). Relationship between differentially expressed mRNA and mRNA-protein correlations in a xenograft model 167 system. Sci. Rep. 5, 1–9. Krischer, J.P., Lynch, K.F., Schatz, D.A., Ilonen, J., Lernmark, Å., Hagopian, W.A., Rewers, M.J., She, J.X., Simell, O.G., Toppari, J., et al. (2015). The 6 year incidence of diabetes-associated autoantibodies in genetically at-risk children: the TEDDY study. Diabetologia 58, 980–987. Krivega, I., and Dean, A. (2017). LDB1-mediated enhancer looping can be established independent of mediator and cohesin. Nucleic Acids Res. 45, 8255–8268. Krogvold, L., Wiberg, A., Edwin, B., Buanes, T., Jahnsen, F.L., Hanssen, K.F., Larsson, E., Korsgren, O., Skog, O., and Dahl-Jørgensen, K. (2016). Insulitis and characterisation of infiltrating T cells in surgical pancreatic tail resections from patients at onset of type 1 diabetes. Diabetologia 59, 492–501. Kuleshov, M. V., Jones, M.R., Rouillard, A.D., Fernandez, N.F., Duan, Q., Wang, Z., Koplev, S., Jenkins, S.L., Jagodnik, K.M., Lachmann, A., et al. (2016). Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res. 44, W90–W97. Kushner, J.A. (2006). β-cell growth: An unusual paradigm of organogenesis that is cyclin D2/Cdk4 dependent. Cell Cycle 5, 234–237. Van De Laar, F.A., Lucassen, P.L., Akkermans, R.P., Van De Lisdonk, E.H., Rutten, G.E., and Van Weel, C. (2005). α-Glucosidase inhibitors for patients with type 2 diabetes: Results from a Cochrane systematic review and meta-analysis. Diabetes Care 28, 154–163. Lachmann, A., Torre, D., Keenan, A.B., Jagodnik, K.M., Lee, H.J., Wang, L., Silverstein, M.C., and Ma’ayan, A. (2018). Massive mining of publicly available RNA-seq data from human and mouse. Nat. Commun. 9, 1366. Langmead, B., and Salzberg, S.L. (2012). Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359. Lawlor, N., Khetan, S., Ucar, D., and Stitzel, M.L. (2017a). Genomics of Islet (Dys)function and Type 2 Diabetes. Trends Genet. 33, 244–255. Lawlor, N., Youn, A., Kursawe, R., Ucar, D., and Stitzel, M.L. (2017b). Alpha TC1 and Beta-TC-6 genomic profiling uncovers both shared and distinct transcriptional regulatory features with their primary islet counterparts. Sci. Rep. 7, 11959. Lawson, R., Maret, W., and Hogstrand, C. (2018). Prolonged stimulation of insulin release from MIN6 cells causes zinc depletion and loss of β-cell markers. J. Trace Elem. Med. Biol. 49, 51–59. Lee, T.I., and Young, R.A. (2000). Transcription of Eukaryotic Protein-Coding Genes. Annu. Rev. Genet. 34, 77–137. 168 Lee, D., Jeong, D.-E., Son, H.G., Yamaoka, Y., Kim, H., Seo, K., Khan, A.A., Roh, T.-Y., Moon, D.W., Lee, Y., et al. (2015). SREBP and MDT-15 protect C. elegans from glucose-induced accelerated aging by preventing accumulation of saturated fat. Genes Dev. 29, 2490–2503. Lee, D., An, S.W.A., Jung, Y., Yamaoka, Y., Ryu, Y., Goh, G.Y.S., Beigi, A., Yang, J.S., Jung, G.Y., Ma, D.K., et al. (2019). MDT-15/MED15 permits longevity at low temperature via enhancing lipidostasis and proteostasis. PLoS Biol. 17. Lee, K., Goh, G.Y.S., Wong, M.A., Klassen, T.L., and Taubert, S. (2016). Gain-of-function alleles in caenorhabditis elegans nuclear hormone receptor nhr-49 are functionally distinct. PLoS One 11. Legland, D., Arganda-Carreras, I., and Andrey, P. (2016). MorphoLibJ: integrated library and plugins for mathematical morphology with ImageJ. Bioinformatics 32, btw413. Lei, L., Jin, T., and Zhou, Y. (2005). [The toxic effects of cadmium on pancreas.]. Zhonghua Lao Dong Wei Sheng Zhi Ye Bing Za Zhi 23, 45–49. Lei, L., Jin, T., and Zhou, Y. (2006). [The effects of cadmium on the levels of insulin in smelters]. Zhonghua Lao Dong Wei Sheng Zhi Ye Bing Za Zhi 24, 3–6. Lemaire, K., Ravier, M.A., Schraenen, A., Creemers, J.W.M., Van de Plas, R., Granvik, M., Van Lommel, L., Waelkens, E., Chimienti, F., Rutter, G.A., et al. (2009). Insulin crystallization depends on zinc transporter ZnT8 expression, but is not required for normal glucose homeostasis in mice. Proc. Natl. Acad. Sci. 106, 14872–14877. Lenoir, O., Flosseau, K., Ma, F.X., Blondeau, B., Mai, A., Bassel-Duby, R., Ravassard, P., Olson, E.N., Haumaitre, C., and Scharfmann, R. (2011). Specific control of pancreatic endocrine β- and δ-cell mass by class IIa histone deacetylases HDAC4, HDAC5, and HDAC9. Diabetes 60, 2861–2871. Leopold, E. (1930). Aretaeus the Cappadocian: His Contribution to Diabetes Mellitus. Ann. Med. Hist. 2, 424–435. Lerdrup, M., Johansen, J.V., Agrawal-Singh, S., and Hansen, K. (2016). An interactive environment for agile analysis and visualization of ChIP-sequencing data. Nat. Struct. Mol. Biol. 23, 349–357. Li, Y. V. (2014). Zinc and insulin in pancreatic beta-cells. Endocrine 45, 178–189. Li, T., and Price, D.H. (2012). Gdown1: Making a link between Mediator and RNA polymerase II elongation control. Transcription 3, 177–180. Li, H., Gade, P., Nallar, S.C., Raha, A., Roy, S.K., Karra, S., Reddy, J.K., Reddy, S.P., and Kalvakolanu, D. V. (2008). The Med1 subunit of transcriptional mediator plays a central role in regulating CCAAT/enhancer-binding protein-β-driven transcription in response to interferon-γ. J. 169 Biol. Chem. 283, 13077–13086. Lin, H.M., Lee, J.H., Yadav, H., Kamaraju, A.K., Liu, E., Zhigang, D., Vieira, A., Kim, S.J., Collins, H., Matschinsky, F., et al. (2009). Transforming growth factor-β/Smad3 signaling regulates insulin gene transcription and pancreatic islet β-cell function. J. Biol. Chem. 284, 12246–12257. Liu, J.S.E., and Hebrok, M. (2017). All mixed up: defining roles for β-cell subtypes in mature islets. Genes Dev. 31, 228–240. Lorenzo, P., Juárez-Vicente, F., Cobo-Vuilleumier, N., García-Domínguez, M., and Gauthier, B. (2017). The Diabetes-Linked Transcription Factor PAX4: From Gene to Functional Consequences. Genes (Basel). 8, 101. Love, M.I., Huber, W., and Anders, S. (2014). Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550. Lu, M., Zhang, L., Maul, R.S., Sartippour, M.R., Norris, A., Whitelegge, J., Rao, J.Y., and Brooks, M.N. (2005). The novel gene EG-1 stimulates cellular proliferation. Cancer Res. 65, 6159–6166. Lu, T.T.-H., Heyne, S., Dror, E., Casas, E., Leonhardt, L., Boenke, T., Yang, C.-H., Sagar, Arrigoni, L., Dalgaard, K., et al. (2018). The Polycomb-Dependent Epigenome Controls β Cell Dysfunction, Dedifferentiation, and Diabetes. Cell Metab. 27, 1294-1308.e7. de Luca, C., and Olefsky, J.M. (2008). Inflammation and insulin resistance. FEBS Lett. 582, 97–105. Luft, R. (1989). Oskar Minkowski: Discovery of the pancreatic origin of diabetes, 1889. Diabetologia 32, 399–401. Ma, F., Jiang, S., and Zhang, C. yang (2019). Recent advances in histone modification and histone modifying enzyme assays. Expert Rev. Mol. Diagn. 19, 27–36. Malik, S., and Roeder, R.G. (2010). The metazoan Mediator co-activator complex as an integrative hub for transcriptional regulation. Nat. Rev. Genet. 11, 761–772. Maston, G.A., Evans, S.K., Green, M.R., Maston GA, Evans SK, G.M., Maston, G.A., Evans, S.K., and Green, M.R. (2006). Transcriptional regulatory elements in the human genome. Annu Rev Genomics Hum Genet. 7, 29–59. Mattsson, G. (2005). The Endothelial Cells in Islets of Langerhans. Ups. J. Med. Sci. 110, 1–15. McCulloch, L.J., van de Bunt, M., Braun, M., Frayn, K.N., Clark, A., and Gloyn, A.L. (2011). GLUT2 (SLC2A2) is not the principal glucose transporter in human pancreatic beta cells: Implications for understanding genetic association signals at this locus. Mol. Genet. Metab. 104, 648–653. 170 McLean, C.Y., Bristor, D., Hiller, M., Clarke, S.L., Schaar, B.T., Lowe, C.B., Wenger, A.M., and Bejerano, G. (2010). GREAT improves functional interpretation of cis-regulatory regions. Nat. Biotechnol. 28, 495–501. McQuin, C., Goodman, A., Chernyshev, V., Kamentsky, L., Cimini, B.A., Karhohs, K.W., Doan, M., Ding, L., Rafelski, S.M., Thirstrup, D., et al. (2018). CellProfiler 3.0: Next-generation image processing for biology. PLOS Biol. 16, e2005970. Medvei, V. (1993). The history of clinical endocrinology: a comprehensive account of endocrinology from earliest times to the present day. v. Mering, J., and Minkowski, O. (1890). Diabetes mellitus nach Pankreasexstirpation. Arch. Für Exp. Pathol. Und Pharmakologie 26, 371–387. van der Meulen, T., Donaldson, C.J., Cáceres, E., Hunter, A.E., Cowing-Zitron, C., Pound, L.D., Adams, M.W., Zembrzycki, A., Grove, K.L., and Huising, M.O. (2015). Urocortin3 mediates somatostatin-dependent negative feedback control of insulin secretion. Nat. Med. 21, 769–776. Meur, G., Qian, Q., Da Silva Xavier, G., Pullen, T.J., Tsuboi, T., McKinnon, C., Fletcher, L., Tavaré, J.M., Hughes, S., Johnson, P., et al. (2011). Nucleo-cytosolic shuttling of FoxO1 directly regulates mouse Ins2 but not Ins1 gene expression in pancreatic beta cells (MIN6. J. Biol. Chem. 286, 13647–13656. De Meyts, P. (2000). The Insulin Receptor and Its Signal Transduction Network (MDText.com, Inc.). Miralles, F., Battelino, T., Czernichow, P., and Scharfmann, R. (1998). TGF-β plays a key role in morphogenesis of the pancreatic islets of langerhans by controlling the activity of the matrix metalloproteinase MMP-2. J. Cell Biol. 143, 827–836. Mishra, A., and Hawkins, R.D. (2017). Three-dimensional genome architecture and emerging technologies: looping in disease. Genome Med. 9, 87. Miyazaki, J.I., Araki, K., Yamato, E., Ikegami, H., Asano, T., Shibasaki, Y., Oka, Y., Yamamura, K.I., and Miyazaki, J.I. (1990). Establishment of a pancreatic β cell line that retains glucose-inducible insulin secretion: Special reference to expression of glucose transporter isoforms. Endocrinology 127, 126–132. El Muayed, M., Billings, L.K., Raja, M.R., Zhang, X., Park, P.J., Newman, M. V., Kaufman, D.B., O’Halloran, T. V., and Lowe, W.L. (2010). Acute cytokine-mediated downregulation of the zinc transporter ZnT8 alters pancreatic β-cell function. J. Endocrinol. 206, 159–169. Mumbach, M.R., Rubin, A.J., Flynn, R.A., Dai, C., Khavari, P.A., Greenleaf, W.J., and Chang, H.Y. (2016). HiChIP: Efficient and sensitive analysis of protein-directed genome architecture. Nat. Methods 13, 919–922. Näär, A.M., Beaurang, P.A., Zhou, S., Abraham, S., Solomon, W., and Tjian, R. (1999). 171 Composite co-activator ARC mediates chromatin-directed transcriptional activation. Nature 398, 828–832. Nair, G.G., Liu, J.S., Russ, H.A., Tran, S., Saxton, M.S., Chen, R., Juang, C., Li, M., Nguyen, V.Q., Giacometti, S., et al. (2019). Recapitulating endocrine cell clustering in culture promotes maturation of human stem-cell-derived β cells. Nat. Cell Biol. 21, 263–274. Nishimura, W., Takahashi, S., and Yasuda, K. (2015). MafA is critical for maintenance of the mature beta cell phenotype in mice. Diabetologia 58, 566–574. Noble, J.A., and Valdes, A.M. (2011). Genetics of the HLA region in the prediction of type 1 diabetes. Curr. Diab. Rep. 11, 533–542. Nozawa, K., Schneider, T.R., and Cramer, P. (2017). Core Mediator structure at 3.4 Å extends model of transcription initiation complex. Nature 545, 248–251. Nriagu, J. (2011). Zinc Toxicity in Humans. In Encyclopedia of Environmental Health, (Elsevier Inc.), pp. 801–807. Nyengaard, J.R., Ido, Y., Kilo, C., and Williamson, J.R. (2004). Interactions Between Hyperglycemia and Hypoxia. Diabetes 53, 2931–2938. Nygaard, S.B., Larsen, A., Knuhtsen, A., Rungby, J., and Smidt, K. (2014). Effects of zinc supplementation and zinc chelation on in vitro β-cell function in INS-1E cells. BMC Res. Notes 7, 84. Ohta, S., Ikemoto, T., Wada, Y., Saito, Y., Yamada, S., Imura, S., Morine, Y., and Shimada, M. (2019). A change in the zinc ion concentration reflects the maturation of insulin-producing cells generated from adipose-derived mesenchymal stem cells. Sci. Rep. 9, 1–10. Oleck, J., Kassam, S., and Goldman, J.D. (2016). Commentary: Why was inhaled insulin a failure in the market? Diabetes Spectr. 29, 180–184. Oliver-Krasinski, J.M., Kasner, M.T., Yang, J., Crutchlow, M.F., Rustgi, A.K., Kaestner, K.H., and Stoffers, D.A. (2009). The diabetes gene Pdx1 regulates the transcriptional network of pancreatic endocrine progenitor cells in mice. J. Clin. Invest. 119, 1888–1898. Pal, D., Sharma, U., Singh, S.K., Mandal, A.K., and Prasad, R. (2014). Metallothionein gene expression in renal cell carcinoma. Indian J. Urol. 30, 241–244. Pan, F.C., and Wright, C. (2011). Pancreas organogenesis: From bud to plexus to gland. Dev. Dyn. 240, 530–565. Pandol, S.J. (2010). Regulation of Whole-Organ Pancreatic Secretion. In The Exocrine Pancreas, (Morgan & Claypool Life Sciences), p. Papizan, J.B., Singer, R.A., Tschen, S.I., Dhawan, S., Friel, J.M., Hipkens, S.B., Magnuson, 172 M.A., Bhushan, A., and Sussel, L. (2011). Nkx2.2 repressor complex regulates islet β-cell specification and prevents β-to-α-cell reprogramming. Genes Dev. 25, 2291–2305. Park, Y.J., and Woo, M. (2019). Pancreatic β cells: Gatekeepers of type 2 diabetes. J. Cell Biol. 218, 1094–1095. Pasquali, L., Gaulton, K.J., Rodríguez-Seguí, S.A., Mularoni, L., Miguel-Escalada, I., Akerman, İ., Tena, J.J., Morán, I., Gómez-Marín, C., van de Bunt, M., et al. (2014). Pancreatic islet enhancer clusters enriched in type 2 diabetes risk-associated variants. Nat. Genet. 46, 136–143. Paul, E., Zhu, Z.I., Landsman, D., and Morse, R.H. (2015). Genome-Wide Association of Mediator and RNA Polymerase II in Wild-Type and Mediator Mutant Yeast. Mol. Cell. Biol. 35, 331–342. Van De Peppel, J., Kettelarij, N., Van Bakel, H., Kockelkorn, T.T.J.P., Van Leenen, D., and Holstege, F.C.P. (2005). Mediator expression profiling epistasis reveals a signal transduction pathway with antagonistic submodules and highly specific downstream targets. Mol. Cell 19, 511–522. Perelis, M., Marcheva, B., Moynihan Ramsey, K., Schipma, M.J., Hutchison, A.L., Taguchi, A., Peek, C.B., Hong, H., Huang, W., Omura, C., et al. (2015). Pancreatic cell enhancers regulate rhythmic transcription of genes controlling insulin secretion. Science (80-. ). 350, aac4250–aac4250. Petersen, M.C., and Shulman, G.I. (2018). Mechanisms of insulin action and insulin resistance. Physiol. Rev. 98, 2133–2223. Pictet, R.L., Clark, W.R., Williams, R.H., and Rutter, W.J. (1972). An ultrastructural analysis of the developing embryonic pancreas. Dev. Biol. 29, 436–467. Plaschka, C., Larivière, L., Wenzeck, L., Seizl, M., Hemann, M., Tegunov, D., Petrotchenko, E. V., Borchers, C.H., Baumeister, W., Herzog, F., et al. (2015). Architecture of the RNA polymerase II–Mediator core initiation complex. Nature 518, 376–380. Poss, Z.C., Ebmeier, C.C., and Taatjes, D.J. (2013). The Mediator complex and transcription regulation. Crit. Rev. Biochem. Mol. Biol. 48, 575–608. Pott, S., and Lieb, J.D. (2015). What are super-enhancers? Nat. Genet. 47, 8–12. Pound, L.D., Sarkar, S.A., Ustione, A., Dadi, P.K., Shadoan, M.K., Lee, C.E., Walters, J.A., Shiota, M., McGuinness, O.P., Jacobson, D.A., et al. (2012). The physiological effects of deleting the mouse Slc30a8 gene encoding zinc transporter-8 are influenced by gender and genetic background. PLoS One 7. Pukkila-Worley, R., Feinbaum, R.L., McEwan, D.L., Conery, A.L., and Ausubel, F.M. (2014). The Evolutionarily Conserved Mediator Subunit MDT-15/MED15 Links Protective Innate Immune Responses and Xenobiotic Detoxification. PLoS Pathog. 10. 173 Punthakee, Z., Goldenberg, R., and Katz, P. (2018). Definition, Classification and Diagnosis of Diabetes, Prediabetes and Metabolic Syndrome. Can. J. Diabetes 42, S10–S15. Qiu, W.L., Zhang, Y.W., Feng, Y., Li, L.C., Yang, L., and Xu, C.R. (2017). Deciphering Pancreatic Islet β Cell and α Cell Maturation Pathways and Characteristic Features at the Single-Cell Level. Cell Metab. 25, 1194-1205.e4. Quevedo, M., Meert, L., Dekker, M.R., Dekkers, D.H.W., Brandsma, J.H., van den Berg, D.L.C., Ozgür, Z., IJcken, W.F.J. va., Demmers, J., Fornerod, M., et al. (2019a). Mediator complex interaction partners organize the transcriptional network that defines neural stem cells. Nat. Commun. 10, 1–15. Quevedo, M., Meert, L., Dekker, M.R., Dekkers, D.H.W., Brandsma, J.H., van den Berg, D.L.C., Ozgür, Z., IJcken, W.F.J. va., Demmers, J., Fornerod, M., et al. (2019b). Mediator complex interaction partners organize the transcriptional network that defines neural stem cells. Nat. Commun. 10, 1–15. Quinlan, A.R., and Hall, I.M. (2010). BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics 26, 841–842. Rachez, C., Suldan, Z., Ward, J., Chang, C.P.B., Burakov, D., Erdjument-Bromage, H., Tempst, P., and Freedman, L.P. (1998). A novel protein complex that interacts with the vitamin D3 receptor in a ligand-dependent manner and enhances VDR transactivation in a cell-free system. Genes Dev. 12, 1787–1800. Ramírez, F., Ryan, D.P., Grüning, B., Bhardwaj, V., Kilpert, F., Richter, A.S., Heyne, S., Dündar, F., and Manke, T. (2016). deepTools2: a next generation web server for deep-sequencing data analysis. Nucleic Acids Res. 44, W160–W165. Raudenska, M., Sztalmachova, M., Gumulec, J., Fojtu, M., Polanska, H., Balvan, J., Feith, M., Binkova, H., Horakova, Z., Kostrica, R., et al. (2015). Prognostic significance of the tumour-adjacent tissue in head and neck cancers. Tumor Biol. 36, 9929–9939. Redondo, M.J., Jeffrey, J., Fain, P.R., Eisenbarth, G.S., and Orban, T. (2008). Concordance for islet autoimmunity among monozygotic twins. N. Engl. J. Med. 359, 2849–2850. Rezania, A., Bruin, J.E., Arora, P., Rubin, A., Batushansky, I., Asadi, A., O’Dwyer, S., Quiskamp, N., Mojibian, M., Albrecht, T., et al. (2014). Reversal of diabetes with insulin-producing cells derived in vitro from human pluripotent stem cells. Nat. Biotechnol. 32, 1121–1133. Rickels, M.R., and Robertson, R.P. (2019). Pancreatic Islet Transplantation in Humans: Recent Progress and Future Directions. Endocr. Rev. 40, 631–668. Rickels, M.R., Kamoun, M., Kearns, J., Markmann, J.F., and Naji, A. (2007). Evidence for Allograft Rejection in an Islet Transplant Recipient and Effect on β-Cell Secretory Capacity. J. Clin. Endocrinol. Metab. 92, 2410–2414. 174 Risley, M.D., Clowes, C., Yu, M., Mitchell, K., and Hentges, K.E. (2010). The Mediator complex protein Med31 is required for embryonic growth and cell proliferation during mammalian development. Dev. Biol. 342, 146–156. Rocha, P.P., Scholze, M., Bleiß, W., and Schrewe, H. (2010). Med12 is essential for early mouse development and for canonical Wnt and Wnt/PCP signaling. Development 137, 2723–2731. Rodriguez-Diaz, R., Molano, R.D., Weitz, J.R., Abdulreda, M.H., Berman, D.M., Leibiger, B., Leibiger, I.B., Kenyon, N.S., Ricordi, C., Pileggi, A., et al. (2018). Paracrine Interactions within the Pancreatic Islet Determine the Glycemic Set Point. Cell Metab. 27, 549-558.e4. Roglic, G., and Unwin, N. (2010). Mortality attributable to diabetes: Estimates for the year 2010. Diabetes Res. Clin. Pract. 87, 15–19. Roh, H.C., Collier, S., Guthrie, J., Robertson, J.D., and Kornfeld, K. (2012). Lysosome-related organelles in intestinal cells are a zinc storage site in C. elegans. Cell Metab. 15, 88–99. Rorsman, P., and Huising, M.O. (2018). The somatostatin-secreting pancreatic δ-cell in health and disease. Nat. Rev. Endocrinol. 14, 404–414. Rorsman, P., and Renström, E. (2003). Insulin granule dynamics in pancreatic beta cells. Diabetologia 46, 1029–1045. Rosen, E.D., Kulkarni, R.N., Sarraf, P., Ozcan, U., Okada, T., Hsu, C.-H., Eisenman, D., Magnuson, M.A., Gonzalez, F.J., Kahn, C.R., et al. (2003). Targeted Elimination of Peroxisome Proliferator-Activated Receptor γ in β Cells Leads to Abnormalities in Islet Mass without Compromising Glucose Homeostasis. Mol. Cell. Biol. 23, 7222–7229. Rozzo, A., Meneghel-Rozzo, T., Delakorda, S.L., Yang, S.B., and Rupnik, M. (2009). Exocytosis of insulin: In vivo maturation of mouse endocrine pancreas. In Annals of the New York Academy of Sciences, (Blackwell Publishing Inc.), pp. 53–62. Ryu, S., and Tjian, R. (1999). Purification of transcription cofactor complex CRSP. Proc. Natl. Acad. Sci. U. S. A. 96, 7137–7142. Ryu, S., Zhou, S., Ladurner, A.G., and Tjian, R. (1999). The transcriptional cofactor complex CRSP is required for activity of the enhancer-binding protein Sp1. Nature 397, 446–450. Sabatini, P. V, and Lynn, F.C. (2015). All-encomPASsing regulation of β-cells: PAS domain proteins in β-cell dysfunction and diabetes. Trends Endocrinol. Metab. 26, 49–57. Sabatini, P. V., Speckmann, T., Nian, C., Glavas, M.M., Wong, C.K., Yoon, J.S., Kin, T., Shapiro, A.M.J., Gibson, W.T., Verchere, C.B., et al. (2018). Neuronal PAS Domain Protein 4 Suppression of Oxygen Sensing Optimizes Metabolism during Excitation of Neuroendocrine Cells. Cell Rep. 22, 163–174. Sakata, N., Yoshimatsu, G., and Kodama, S. (2019). Development and characteristics of 175 pancreatic epsilon cells. Int. J. Mol. Sci. 20. Sander, M., Sussel, L., Conners, J., Scheel, D., Kalamaras, J., Dela Cruz, F., Schwitzgebel, V., Hayes-Jordan, A., and German, M. (2000). Homeobox gene Nkx6.1 lies downstream of Nkx2.2 in the major pathway of β-cell formation in the pancreas. Development 127, 5533–5540. Sauer, B. (1998). Inducible gene targeting in mice using the Cre/lox system. Methods 14, 381–392. Schmitz, O., Rungby, J., Edge, L., and Juhl, C.B. (2008). On high-frequency insulin oscillations. Ageing Res. Rev. 7, 301–305. Schneider, C.A., Rasband, W.S., and Eliceiri, K.W. (2012). NIH Image to ImageJ: 25 years of image analysis. Nat. Methods 9, 671–675. Schuit, F., Van Lommel, L., Granvik, M., Goyvaerts, L., De Faudeur, G., Schraenen, A., and Lemaire, K. (2012). β-cell-specific gene repression: A mechanism to protect against inappropriate or maladjusted insulin secretion? Diabetes 61, 969–975. Schwitzgebel, V.M., Scheel, D.W., Conners, J.R., Kalamaras, J., Lee, J.E., Anderson, D.J., Sussel, L., Johnson, J.D., and German, M.S. (2000). Expression of neurogenin3 reveals an islet cell precursor population in the pancreas. Development 127, 3533–3542. Scott, L.J., Mohlke, K.L., Bonnycastle, L.L., Willer, C.J., Li, Y., Duren, W.L., Erdos, M.R., Stringham, H.M., Chines, P.S., Jackson, A.U., et al. (2007). A Genome-Wide Association Study of Type 2 Diabetes in Finns Detects Multiple Susceptibility Variants. Science (80-. ). 316, 1341–1345. Segerstolpe, Å., Palasantza, A., Eliasson, P., Andersson, E.-M., Andréasson, A.-C., Sun, X., Picelli, S., Sabirsh, A., Clausen, M., Bjursell, M.K., et al. (2016). Single-Cell Transcriptome Profiling of Human Pancreatic Islets in Health and Type 2 Diabetes. Cell Metab. 24, 593–607. Semache, M., Ghislain, J., Zarrouki, B., Tremblay, C., and Poitout, V. (2014). Pancreatic and duodenal homeobox-1 nuclear localization is regulated by glucose in dispersed rat islets but not in insulin-secreting cell lines. Islets 6. Servitja, J.M., and Ferrer, J. (2004). Transcriptional networks controlling pancreatic development and beta cell function. Diabetologia 47, 597–613. Shaikhibrahim, Z., Menon, R., Braun, M., Offermann, A., Queisser, A., Boehm, D., Vogel, W., Rüenauver, K., Ruiz, C., Zellweger, T., et al. (2014). MED15, encoding a subunit of the mediator complex, is overexpressed at high frequency in castration-resistant prostate cancer. Int. J. Cancer 135, 19–26. Shaikhibrahim, Z., Offermann, A., Halbach, R., Vogel, W., Braun, M., Kristiansen, G., Bootz, F., Wenzel, J., Mikut, R., Lengerke, C., et al. (2015). Clinical and molecular implications of MED15 in head and neck squamous cell carcinoma. Am. J. Pathol. 185, 1114–1122. 176 Shalev, A. (2014). Minireview: Thioredoxin-interacting protein: Regulation and function in the pancreatic β-cell. Mol. Endocrinol. 28, 1211–1220. Shapiro, A.M.J., Lakey, J.R.T., Ryan, E.A., Korbutt, G.S., Toth, E., Warnock, G.L., Kneteman, N.M., and Rajotte, R. V. (2000). Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. N. Engl. J. Med. 343, 230–238. Sharifi-Zarchi, A., Gerovska, D., Adachi, K., Totonchi, M., Pezeshk, H., Taft, R.J., Schöler, H.R., Chitsaz, H., Sadeghi, M., Baharvand, H., et al. (2017). DNA methylation regulates discrimination of enhancers from promoters through a H3K4me1-H3K4me3 seesaw mechanism. BMC Genomics 18, 964. Sharma, D., and Fondell, J.D. (2002). Ordered recruitment of histone acetyltransferases and the TRAP/Mediator complex to thyroid hormone-responsive promoters in vivo. Proc. Natl. Acad. Sci. U. S. A. 99, 7934–7939. Sharon, N., Chawla, R., Mueller, J., Gifford, D.K., Trapnell, C., and Melton, D. (2018). A Peninsular Structure Coordinates Asynchronous Differentiation with Morphogenesis to Generate Pancreatic Islets. Cell 176, 790–804. Shimano, H., Amemiya-Kudo, M., Takahashi, A., Kato, T., Ishikawa, M., and Yamada, N. (2007). Sterol regulatory element?binding protein-1c and pancreatic ?-cell dysfunction. Diabetes, Obes. Metab. 9, 133–139. Shomer, N. (2018). A regulatory mechanism of zinc homeostasis involving the mediator subunit MDT-15 and the transcription factor HIZR-1. Shomer, N., Zacharie Kadhim, A., Margaret Grants, J., Cheng, X., Alhusari, D., Bhanshali, F., Poon, A.F.Y., Ya Lee, M.Y., Muhuri, A., In Park, J., et al. (2019). Mediator subunit MDT-15/MED15 and Nuclear Receptor HIZR-1/HNF4 cooperate to regulate toxic metal stress responses in Caenorhabditis elegans. PLoS Genet. 15, e1008508. Shurrab, N.T., and Arafa, E.S.A. (2020). Metformin: A review of its therapeutic efficacy and adverse effects. Obes. Med. 17, 100186. Si, M., and Lang, J. (2018). The roles of metallothioneins in carcinogenesis. J. Hematol. Oncol. 11, 1–20. Singal, D.P., and Blajchman, M.A. (1973). Histocompatibility (HL A) antigens, lymphocytotoxic antibodies and tissue antibodies in patients with diabetes mellitus. Diabetes 22, 429–432. Sladek, R., Rocheleau, G., Rung, J., Dina, C., Shen, L., Serre, D., Boutin, P., Vincent, D., Belisle, A., Hadjadj, S., et al. (2007). A genome-wide association study identifies novel risk loci for type 2 diabetes. Nature 445, 881–885. Sneddon, J.B., Tang, Q., Stock, P., Bluestone, J.A., Roy, S., Desai, T., and Hebrok, M. (2018). 177 Cell Stem Cell Perspective Stem Cell Therapies for Treating Diabetes: Progress and Remaining Challenges. Sobel, J., Guay, C., Rodriguez-Trejo, A., Stoll, L., Menoud, V., and Regazzi, R. (2019). Chromatin accessibility variations across pancreatic islet maturation. BioRxiv 782318. Sora, N.D., Shashpal, F., Bond, E.A., and Jenkins, A.J. (2019). Insulin Pumps: Review of Technological Advancement in Diabetes Management. Am. J. Med. Sci. 358, 326–331. Sosa-Pineda, B., Chowdhury, K., Torres, M., Oliver, G., and Gruss, P. (1997). The Pax4 gene is essential for differentiation of insulin-producing β cells in the mammalian pancreas. Nature 386, 399–402. Soutourina, J. (2018). Transcription regulation by the Mediator complex. Nat. Rev. Mol. Cell Biol. 19, 262–274. Špaček, T., Pavluch, V., Alán, L., Capková, N., Engstová, H., Dlasková, A., Berková, Z., Saudek, F., and Ježek, P. (2017). Nkx6.1 decline accompanies mitochondrial DNA reduction but subtle nucleoid size decrease in pancreatic islet β-cells of diabetic Goto Kakizaki rats. Sci. Rep. 7, 15674. Spaeth, J.M., Kim, N.H., and Boyer, T.G. (2011). Mediator and Human Disease. Semin. Cell Dev. Biol. 22, 776. Spaeth, J.M., Gupte, M., Perelis, M., Yang, Y.-P., Cyphert, H., Guo, S., Liu, J.-H., Guo, M., Bass, J., Magnuson, M.A., et al. (2017). Defining a Novel Role for the Pdx1 Transcription Factor in Islet β-Cell Maturation and Proliferation During Weaning. Diabetes 66, 2830–2839. Spitler, K.M., Ponce, J.M., Oudit, G.Y., Hall, D.D., and Grueter, C.E. (2017). Cardiac med1 deletion promotes early lethality, cardiac remodeling, and transcriptional reprogramming. Am. J. Physiol. - Hear. Circ. Physiol. 312, H768–H780. Steck, A.K., and Winter, W.E. (2011). Review on monogenic diabetes. Curr. Opin. Endocrinol. Diabetes Obes. 18, 252–258. Stoffers, D.A., Ferrer, J., Clarke, W.L., and Habener, J.F. (1997). Early-onset type-ll diabetes mellitus (Mody4) linked to ipf1. Nat. Genet. 17, 139. Stolovich-Rain, M., Enk, J., Vikesa, J., Nielsen, F.C., Saada, A., Glaser, B., and Dor, Y. (2015). Weaning Triggers a Maturation Step of Pancreatic β Cells. Dev. Cell 32, 535–545. Sun, M., Jiang, R., Li, J.D., Luo, S.L., Gao, H.W., Jin, C.Y., Shi, D.L., Wang, C.G., Wang, B., and Zhang, X.Y. (2011). MED19 promotes proliferation and tumorigenesis of lung cancer. Mol. Cell. Biochem. 355, 27–33. Sun, W., Yao, S., Tang, J., Liu, S., Chen, J., Deng, D., and Zeng, C. (2018). Integrative analysis of super enhancer SNPs for type 2 diabetes. PLoS One 13. 178 Sussel, L., Kalamaras, J., Hartigan-O’Connor, D.J., Meneses, J.J., Pedersen, R.A., Rubenstein, J.L.R., and German, M.S. (1998). Mice lacking the homeodomain transcription factor Nkx2.2 have diabetes due to arrested differentiation of pancreatic β cells. Development 125, 2213–2221. Swisa, A., Avrahami, D., Eden, N., Zhang, J., Feleke, E., Dahan, T., Cohen-Tayar, Y., Stolovich-Rain, M., Kaestner, K.H., Glaser, B., et al. (2017). PAX6 maintains β cell identity by repressing genes of alternative islet cell types. J. Clin. Invest. 127, 230–243. Syring, I., Klümper, N., Offermann, A., Braun, M., Deng, M., Boehm, D., Queisser, A., von Mässenhausen, A., Brägelmann, J., Vogel, W., et al. (2016a). Comprehensive analysis of the transcriptional profile of the Mediator complex across human cancer types. Oncotarget 7, 23043–23055. Syring, K.E., Boortz, K.A., Oeser, J.K., Ustione, A., Platt, K.A., Shadoan, M.K., McGuinness, O.P., Piston, D.W., Powell, D.R., and O’Brien, R.M. (2016b). Combined deletion of Slc30a7 and Slc30a8 unmasks a critical role for ZnT8 in glucose-stimulated insulin secretion. Endocrinology 157, 4534–4541. Szabat, M., Lynn, F.C., Hoffman, B.G., Kieffer, T.J., Allan, D.W., and Johnson, J.D. (2012). Maintenance of β-cell maturity and plasticity in the adult pancreas: developmental biology concepts in adult physiology. Diabetes 61, 1365–1371. Taatjes, D.J. (2010). The human Mediator complex: a versatile, genome-wide regulator of transcription. Trends Biochem. Sci. 35, 315–322. Taatjes, D.J., Näär, A.M., Andel, F., Nogales, E., and Tjian, R. (2002a). Structure, function, and activator-induced conformations of the CRSP coactivator. Science (80-. ). 295, 1058–1062. Taatjes, D.J., Näär, A.M., Andel, F., Nogales, E., and Tjian, R. (2002b). Structure, function, and activator-induced conformations of the CRSP coactivator. Science (80-. ). 295, 1058–1062. Taubert, S., Van Gilst, M.R., Hansen, M., and Yamamoto, K.R. (2006). A Mediator subunit, MDT-15, integrates regulation of fatty acid metabolism by NHR-49-dependent and -independent pathways in C. elegans. Genes Dev. 20, 1137–1149. Taubert, S., Hansen, M., Van Gilst, M.R., Cooper, S.B., and Yamamoto, K.R. (2008). The Mediator Subunit MDT-15 Confers Metabolic Adaptation to Ingested Material. PLoS Genet. 4, e1000021. Taubert, S., Ward, J.D., and Yamamoto, K.R. (2011). Nuclear hormone receptors in nematodes: Evolution and function. Mol. Cell. Endocrinol. 334, 49–55. Taylor, B.L., Liu, F.-F., and Sander, M. (2013). Nkx6.1 is essential for maintaining the functional state of pancreatic beta cells. Cell Rep. 4, 1262–1275. Tennant, B.R., Robertson, A.G., Kramer, M., Li, L., Zhang, X., Beach, M., Thiessen, N., Chiu, R., Mungall, K., Whiting, C.J., et al. (2013). Identification and analysis of murine pancreatic islet 179 enhancers. Diabetologia 56, 542–552. Tentolouris, A., Vlachakis, P., Tzeravini, E., Eleftheriadou, I., and Tentolouris, N. (2019). SGLT2 inhibitors: A review of their antidiabetic and cardioprotective effects. Int. J. Environ. Res. Public Health 16. Thakur, J.K., Yadav, A., and Yadav, G. (2014). Molecular recognition by the KIX domain and its role in gene regulation. Nucleic Acids Res. 42, 2112–2125. Thirumoorthy, N., Manisenthil Kumar, K.T., Sundar, A.S., Panayappan, L., and Chatterjee, M. (2007). Metallothionein: An overview. World J. Gastroenterol. 13, 993–996. Thirumoorthy, N., Shyam Sunder, A., Manisenthil Kumar, K.T., Senthil kumar, M., Ganesh, G.N.K., and Chatterjee, M. (2011). A review of metallothionein isoforms and their role in pathophysiology. World J. Surg. Oncol. 9, 54. Thompson, C.M., Young, R.A., Hayashibara, K., Brown, P., and Kornberg, R. (1995). General requirement for RNA polymerase II holoenzymes in vivo. Proc. Natl. Acad. Sci. 92, 4587–4590. Thurner, M., van de Bunt, M., Torres, J.M., Mahajan, A., Nylander, V., Bennett, A.J., Gaulton, K.J., Barrett, A., Burrows, C., Bell, C.G., et al. (2018). Integration of human pancreatic islet genomic data refines regulatory mechanisms at Type 2 Diabetes susceptibility loci. Elife 7. Tremmel, D.M., Mitchell, S.A., Sackett, S.D., and Odorico, J.S. (2019). Mimicking nature-made beta cells: Recent advances towards stem cell-derived islets. Curr. Opin. Organ Transplant. 24, 574–581. Tsai, K.-L., Tomomori-Sato, C., Sato, S., Conaway, R.C., Conaway, J.W., and Asturias, F.J. (2014). Subunit architecture and functional modular rearrangements of the transcriptional mediator complex. Cell 157, 1430–1444. Tsai, K.-L., Yu, X., Gopalan, S., Chao, T.-C., Zhang, Y., Florens, L., Washburn, M.P., Murakami, K., Conaway, R.C., Conaway, J.W., et al. (2017). Mediator structure and rearrangements required for holoenzyme formation. Nature 544, 196–201. Tsutsui, T., Fukasawa, R., Shinmyouzu, K., Nakagawa, R., Tobe, K., Tanaka, A., and Ohkuma, Y. (2013). Mediator complex recruits epigenetic regulators via its two cyclin-dependent kinase subunits to repress transcription of immune response genes. J. Biol. Chem. 288, 20955–20965. Tuomilehto, J., Lindström, J., Eriksson, J.G., Valle, T.T., Hämäläinen, H., Ilanne-Parikka, P., Keinänen-Kiukaanniemi, S., Laakso, M., Louheranta, A., Rastas, M., et al. (2001). Prevention of Type 2 Diabetes Mellitus by Changes in Lifestyle among Subjects with Impaired Glucose Tolerance. N. Engl. J. Med. 344, 1343–1350. Vanderford, N.L., Andrali, S.S., and Özcan, S. (2007). Glucose induces MafA expression in pancreatic beta cell lines via the hexosamine biosynthetic pathway. J. Biol. Chem. 282, 1577–1584. 180 Varelas, X., Sakuma, R., Samavarchi-Tehrani, P., Peerani, R., Rao, B.M., Dembowy, J., Yaffe, M.B., Zandstra, P.W., and Wrana, J.L. (2008). TAZ controls Smad nucleocytoplasmic shuttling and regulates human embryonic stem-cell self-renewal. Nat. Cell Biol. 10, 837–848. Vaxillaire, M., Pueyo, M.E., Clément, K., Fiet, J., Timsit, J., Philippe, J., Robert, J.J., Tappy, L., Froguel, P., and Velho, G. (1999). Insulin secretion and insulin sensitivity in diabetic and non-diabetic subjects with hepatic nuclear factor-1α (maturity-onset diabetes of the young-3) mutations. Eur. J. Endocrinol. 141, 609–618. Velazco-Cruz, L., Song, J., Maxwell, K.G., Goedegebuure, M.M., Augsornworawat, P., Hogrebe, N.J., and Millman, J.R. (2019). Acquisition of Dynamic Function in Human Stem Cell-Derived β Cells. Stem Cell Reports 12, 351–365. Veres, A., Faust, A.L., Bushnell, H.L., Engquist, E.N., Kenty, J.H.-R., Harb, G., Poh, Y.-C., Sintov, E., Gürtler, M., Pagliuca, F.W., et al. (2019). Charting cellular identity during human in vitro β-cell differentiation. Nature 569, 368–373. Verger, A., Monté, D., and Villeret, V. (2019). Twenty years of Mediator complex structural studies. Biochem. Soc. Trans. 47, 399–410. Villasenor, A., Chong, D.C., and Cleaver, O. (2008). Biphasic Ngn3 expression in the developing pancreas. Dev. Dyn. 237, 3270–3279. Villasenor, A., Chong, D.C., Henkemeyer, M., and Cleaver, O. (2010). Epithelial dynamics of pancreatic branching morphogenesis. Development 137, 4295–4305. Vionnet, N., Stoffel, M., Takeda, J., Yasuda, K., Bell, G.I., Zouali, H., Lesage, S., Velho, G., Iris, F., Passa, P., et al. (1992). Nonsense mutation in the glucokinase gene causes early-onset non-insulin-dependent diabetes mellitus. Nature 356, 721–722. Vozdek, R., Long, Y., and Ma, D.K. (2018). The receptor tyrosine kinase HIR-1 coordinates HIF-independent responses to hypoxia and extracellular matrix injury. Sci. Signal. 11, eaat0138. Wang, H., Maechler, P., Antinozzi, P.A., Hagenfeldt, K.A., and Wollheim, C.B. (2000). Hepatocyte nuclear factor 4alpha regulates the expression of pancreatic beta -cell genes implicated in glucose metabolism and nutrient-induced insulin secretion. J. Biol. Chem. 275, 35953–35959. Wang, J., Elghazi, L., Parker, S.E., Kizilocak, H., Asano, M., Sussel, L., and Sosa-Pineda, B. (2004). The concerted activities of Pax4 and Nkx2.2 are essential to initiate pancreatic beta-cell differentiation. Dev. Biol. 266, 178–189. Wang, J., Zhang, J., Li, L., Wang, Y., Wang, Q., Zhai, Y., You, H., and Hu, D. (2013). Association of rs12255372 in the TCF7L2 gene with type 2 diabetes mellitus: a meta-analysis. Brazilian J. Med. Biol. Res. = Rev. Bras. Pesqui. Medicas e Biol. 46, 382–393. Wang, W., Huang, L., Huang, Y., Yin, J. wen, Berk, A.J., Friedman, J.M., and Wang, G. (2009). 181 Mediator MED23 Links Insulin Signaling to the Adipogenesis Transcription Cascade. Dev. Cell 16, 764–771. Warnhoff, K., Roh, H.C., Kocsisova, Z., Tan, C.H., Morrison, A., Croswell, D., Schneider, D.L., and Kornfeld, K. (2017). The Nuclear Receptor HIZR-1 Uses Zinc as a Ligand to Mediate Homeostasis in Response to High Zinc. PLoS Biol. 15, e2000094. Weber, C.M., and Henikoff, S. (2014). Histone variants: dynamic punctuation in transcription. Genes Dev. 28, 672–682. Weber, H., and Garabedian, M.J. (2018). The mediator complex in genomic and non-genomic signaling in cancer. Steroids 133, 8–14. Weisman, A., Fazli, G.S., Johns, A., and Booth, G.L. (2018). Evolving Trends in the Epidemiology, Risk Factors, and Prevention of Type 2 Diabetes: A Review. Can. J. Cardiol. 34, 552–564. Weiten, R., Müller, T., Schmidt, D., Steiner, S., Kristiansen, G., Müller, S.C., Ellinger, J., and Syring, I. (2018). The Mediator complex subunit MED15, a promoter of tumour progression and metastatic spread in renal cell carcinoma. Cancer Biomarkers 21, 839–847. Wenzlau, J.M., Frisch, L.M., Hutton, J.C., and Davidson, H.W. (2011). Mapping of conformational autoantibody epitopes in ZNT8. Diabetes. Metab. Res. Rev. 27, 883–886. Wessells, N.K., and Cohen, J.H. (1967). Early pancreas organogenesis: Morphogenesis, tissue interactions, and mass effects. Dev. Biol. 15, 237–270. Westerling, T., Kuuluvainen, E., and Mäkelä, T.P. (2007). Cdk8 Is Essential for Preimplantation Mouse Development. Mol. Cell. Biol. 27, 6177–6182. Whyte, W.A., Orlando, D.A., Hnisz, D., Abraham, B.J., Lin, C.Y., Kagey, M.H., Rahl, P.B., Lee, T.I., and Young, R.A. (2013). Master transcription factors and mediator establish super-enhancers at key cell identity genes. Cell 153, 307–319. Wijesekara, N., Dai, F.F., Hardy, A.B., Giglou, P.R., Bhattacharjee, A., Koshkin, V., Chimienti, F., Gaisano, H.Y., Rutter, G.A., and Wheeler, M.B. (2010). Beta cell-specific Znt8 deletion in mice causes marked defects in insulin processing, crystallisation and secretion. Diabetologia 53, 1656–1668. Wilcox, G. (2005). Insulin and insulin resistance. Clin. Biochem. Rev. 26, 19–39. Williams, R., Karuranga, S., Malanda, B., Saeedi, P., Basit, A., Besançon, S., Bommer, C., Esteghamati, A., Ogurtsova, K., Zhang, P., et al. (2020). Global and regional estimates and projections of diabetes-related health expenditure: Results from the International Diabetes Federation Diabetes Atlas, 9th edition. Diabetes Res. Clin. Pract. 0. Wilson, M.E., Scheel, D., and German, M.S. (2003). Gene expression cascades in pancreatic 182 development. Mech. Dev. 120, 65–80. Wollheim, C.B., and Maechler, P. (2002). β-cell mitochondria and insulin secretion: Messenger role of nucleotides and metabolites. In Diabetes, (American Diabetes Association), pp. S37–S42. Wong, C.K. (2018). The role of p300 transcriptional coactivators in pancreatic beta cells. Wong, C.K., Wade-Vallance, A.K., Luciani, D.S., Brindle, P.K., Lynn, F.C., and Gibson, W.T. (2018). The p300 and CBP transcriptional coactivators are required for β-cell and α-cell proliferation. Diabetes 67, 412–422. Wong, K.H., Jin, Y., and Struhl, K. (2014). TFIIH Phosphorylation of the Pol II CTD Stimulates Mediator Dissociation from the Preinitiation Complex and Promoter Escape. Mol. Cell 54, 601–612. Wright, J.R., and McIntyre, L. (2020). Misread and mistaken: Étienne Lancereaux’s enduring legacy in the classification of diabetes mellitus. J. Med. Biogr. Wu, H., Mezghenna, K., Marmol, P., Guo, T., Moliner, A., Yang, S.-N., Berggren, P.-O., and Ibáñez, C.F. (2014). Differential regulation of mouse pancreatic islet insulin secretion and Smad proteins by activin ligands. Diabetologia 57, 148–156. Xu, E.E. (2016). The roles of Sox4 and Med15 in the development and maintenance of pancreatic beta-cells. 45–46. Xu, C., Li, L., Donahue, G., Ying, L., Zhang, Y., Gadue, P., and Zaret, K.S. (2014). Dynamics of genomic H 3 K 27me3 domains and role of EZH 2 during pancreatic endocrine specification. EMBO J. 33, 2157–2170. Xu, C.R., Cole, P.A., Meyers, D.J., Kormish, J., Dent, S., and Zaret, K.S. (2011a). Chromatin “prepattern” and histone modifiers in a fate choice for liver and pancreas. Science (80-. ). 332, 963–966. Xu, Z., Wei, G., Chepelev, I., Zhao, K., and Felsenfeld, G. (2011b). Mapping of INS promoter interactions reveals its role in long-range regulation of SYT8 transcription. Nat. Struct. Mol. Biol. 18, 372–378. Xue, J., Scotti, E., and Correspondence, M.S. (2019). CDK8 Regulates Insulin Secretion and Mediates Postnatal and Stress-Induced Expression of Neuropeptides in Pancreatic β Cells. Yamada, K., Nakata, M., Horimoto, N., Saito, M., Matsuoka, H., and Inagaki, N. (2000). Measurement of Glucose Uptake and Intracellular Calcium Concentration in Single, Living Pancreatic β-Cells. J. Biol. Chem. 275, 22278–22283. Yamagata, K., Oda, N., Kaisaki, P.J., Menzel, S., Furuta, H., Vaxillaire, M., Southam, L., Cox, R.D., Lathrop, G.M., Boriraj, V.V., et al. (1996a). Mutations in the hepatocyte nuclear factor-1α gene in maturity-onset diabetes of the young (MODY3). Nature 384, 455–458. 183 Yamagata, K., Furuta, H., Oda, N., Kaisaki, P.J., Menzel, S., Cox, N.J., Fajans, S.S., Signorini, S., Stoffel, M., and Bell, G.I. (1996b). Mutations in the hepatocyte nuclear factor-4α gene in maturity-onset diabetes of the young (MODY1). Nature 384, 458–460. Yang, F., Vought, B.W., Satterlee, J.S., Walker, A.K., Jim Sun, Z.-Y., Watts, J.L., DeBeaumont, R., Saito, R.M., Hyberts, S.G., Yang, S., et al. (2006). An ARC/Mediator subunit required for SREBP control of cholesterol and lipid homeostasis. Nature 442, 700–704. Yang, L., Li, H., Yu, T., Zhao, H., Cherian, M.G., Cai, L., and Liu, Y. (2008). Polymorphisms in metallothionein-1 and -2 genes associated with the risk of type 2 diabetes mellitus and its complications. Am. J. Physiol. - Endocrinol. Metab. 294. Yang, Y., Leonard, M., Zhang, Y., Zhao, D., Mahmoud, C., Khan, S., Wang, J., Lower, E.E., and Zhang, X. (2018). Tumor Biology and Immunology HER2-Driven Breast Tumorigenesis Relies upon Interactions of the Estrogen Receptor with Coactivator MED1. Ye, B., Yang, G., Li, Y., Zhang, C., Wang, Q., and Yu, G. (2020). ZNF143 in Chromatin Looping and Gene Regulation. Front. Genet. 11, 1–10. Ye, R., Onodera, T., and Scherer, P.E. (2019). Lipotoxicity and B cell maintenance in obesity and type 2 diabetes. J. Endocr. Soc. 3, 617–631. Yeh, H.C., Brown, T.T., Maruthur, N., Ranasinghe, P., Berger, Z., Suh, Y.D., Wilson, L.M., Haberl, E.B., Brick, J., Bass, E.B., et al. (2012). Comparative effectiveness and safety of methods of insulin delivery and glucose monitoring for diabetes mellitus: A systematic review and meta-analysis. Ann. Intern. Med. 157, 336–347. Yin, J. -w., and Wang, G. (2014). The Mediator complex: a master coordinator of transcription and cell lineage development. Development 141, 977–987. Yoshihara, E., Wei, Z., Lin, C.S., Fang, S., Ahmadian, M., Kida, Y., Tseng, T., Dai, Y., Yu, R.T., Liddle, C., et al. (2016a). ERRγ Is Required for the Metabolic Maturation of Therapeutically Functional Glucose-Responsive β Cells. Cell Metab. 23, 622–634. Yoshihara, E., Wei, Z., Lin, C.S., Fang, S., Ahmadian, M., Kida, Y., Tseng, T., Dai, Y., Yu, R.T., Liddle, C., et al. (2016b). ERRγ Is Required for the Metabolic Maturation of Therapeutically Functional Glucose-Responsive β Cells. Cell Metab. 23, 622–634. Youn, D.Y., Xiaoli, A.M., Pessin, J.E., and Yang, F. (2016). Regulation of metabolism by the Mediator complex. Biophys. Reports 2, 69–77. Zhang, D., Liu, J., Gao, J., Shahzad, M., Han, Z., Wang, Z., Li, J., and Sjölinder, H. (2014). Zinc Supplementation Protects against Cadmium Accumulation and Cytotoxicity in Madin-Darby Bovine Kidney Cells. PLoS One 9, e103427. Zhang, Y., Liu, T., Meyer, C.A., Eeckhoute, J., Johnson, D.S., Bernstein, B.E., Nussbaum, C., Myers, R.M., Brown, M., Li, W., et al. (2008). Model-based Analysis of ChIP-Seq (MACS). 184 Genome Biol. 9, R137. Zhao, M., Yang, X., Fu, Y., Wang, H., Ning, Y., Yan, J., Chen, Y.-G., and Wang, G. (2013). Mediator MED15 modulates transforming growth factor beta (TGFβ)/Smad signaling and breast cancer cell metastasis. J. Mol. Cell Biol. 5, 57–60. Zhao, X., Xiaoli, Zong, H., Abdulla, A., Yang, E.S.T., Wang, Q., Ji, J.-Y., Pessin, J.E., Das, B.C., and Yang, F. (2014). Inhibition of SREBP transcriptional activity by a boron-containing compound improves lipid homeostasis in diet-induced obesity. Diabetes 63, 2464–2473. Zheng, T., Liu, S., Bai, Y., Cheng, N., Buka, S., Yang, A., Shi, K., Zhang, X., Li, Y., Xu, S., et al. (2018a). Curre Res Diabetes & Obes J Current Understanding of the Relationship between Metal Exposures and Risk of Type 2 Diabetes. Rev. Artic. 7. Zheng, Y., Ley, S.H., and Hu, F.B. (2018b). Global aetiology and epidemiology of type 2 diabetes mellitus and its complications. Nat. Rev. Endocrinol. 14, 88–98. Zhou, B., Lu, Y., Hajifathalian, K., Bentham, J., Cesare, M. Di, Danaei, G., Bixby, H., Cowan, M., Ali, M., Taddei, C., et al. (2016). Worldwide trends in diabetes since 1980: a pooled analysis of 751 population-based studies with 4.4 million participants. Lancet (London, England) 387, 1513–1530. Zhou, Q., Law, A.C., Rajagopal, J., Anderson, W.J., Gray, P.A., and Melton, D.A. (2007). A Multipotent Progenitor Domain Guides Pancreatic Organogenesis. Dev. Cell 13, 103–114. Zhu, Y., Qi, C., Jain, S., Le Beau, M.M., Espinosa, R., Atkins, G.B., Lazar, M.A., Yeldandi, A. V., Rao, M.S., and Reddy, J.K. (1999). Amplification and overexpression of peroxisome proliferator-activated receptor binding protein (PBP/PPARBP) gene in breast cancer. Proc. Natl. Acad. Sci. U. S. A. 96, 10848–10853. Zito, E., Chin, K.-T., Blais, J., Harding, H.P., and Ron, D. (2010). ERO1-beta, a pancreas-specific disulfide oxidase, promotes insulin biogenesis and glucose homeostasis. J. Cell Biol. 188, 821–832. . 185 Appendices Appendix A Top 50 MED15 bound genes Gene name Peak Score Annotation G6pc2 388.48328 intron (NM_001289857, intron 2 of 3) Cpe 377.97482 intron (NM_013494, intron 1 of 8) Bace2 371.05331 intron (NM_019517, intron 1 of 8) Calm2 324.10184 intron (NM_007589, intron 1 of 5) (CpG) Kcne4 314.11444 Intergenic (RMER1A|Other|Other) Camk2n1 307.0079 intron (NM_025451, intron 1 of 1) Smg7 289.89618 promoter-TSS (NM_001160256) Dhx40 274.41495 intron (NR_106195, intron 4 of 8) Glp1r 270.50897 Intergenic Dynlt3 265.85788 Intergenic (MamRep605|LTR?|LTR?) Cirbp 262.63889 intron (NM_007705, intron 1 of 6) (CpG) Ptbp1 261.93716 promoter-TSS (NM_008956) Fem1b 258.61404 Intergenic (L2|LINE|L2) Eapp 255.23448 Intergenic Alyref2 252.22842 Intergenic Spc25 250.17554 intron (NM_001199123, intron 4 of 6) (B1_Mus2|SINE|Alu) Mir691 245.22934 intron (NM_175549, intron 2 of 27) Usf2 244.53548 promoter-TSS (NM_011680) Srrm1 243.46674 promoter-TSS (NM_001130477) 186 Zfp428 241.00117 intron (NR_110958, intron 1 of 2) (CpG) Mreg 241.00117 Intergenic Dusp4 240.57451 intron (NM_176933, intron 1 of 3) (CpG-13950) Dleu2 233.55324 intron (NR_028264, intron 1 of 6) (CpG) Slc30a8 229.6792 Intergenic ((GCTTG)n|Simple_repeat|Simple_repeat) Gse1 224.82874 promoter-TSS (NM_198671) Hivep2 223.57687 intron (NM_010437, intron 2 of 8) 1700034G24Rik 221.87996 Intergenic Rap1gap2 220.85352 intron (NM_001015046, intron 2 of 24) Isl1 218.65726 intron (NM_021459, intron 1 of 5) Hsp90ab1 217.89175 promoter-TSS (NM_008302) Dst 214.72679 intron (NM_001276764, intron 4 of 102) Hcar2 212.1226 Intergenic (B2_Mm1a|SINE|B2) Pxk 209.19244 intron (NM_145458, intron 4 of 17) Lrp8 208.86624 Intergenic Itgb1 206.28522 promoter-TSS (NM_010578) Xpo1 204.0802 5' UTR (NM_134014, exon 2 of 26) Ppp1r1a 203.28049 intron (NM_021391, intron 1 of 6) Psmg4 200.77592 intron (NM_001033167, intron 3 of 10) Marcks 200.53613 intron (NM_008538, intron 1 of 1) Acsbg1 199.16379 Intergenic (ORR1D1|LTR|ERVL-MaLR) Ilf3 197.91699 promoter-TSS (NM_001277322) Mir1894 195.94714 intron (NM_001163818, intron 1 of 18) Gm14920 195.8313 Intergenic (RLTR9A3B|LTR|ERVK) 187 Shank2 195.61423 intron (NM_001113373, intron 4 of 15) Prim2 194.45122 Intergenic (RMER19B2|LTR|ERVK) Rad9b 192.74396 promoter-TSS (NM_019780) Xylt1 192.62978 Intergenic Elavl4 192.2318 promoter-TSS (NM_001038698) Pgrmc2 191.95171 Intergenic Ing3 190.91864 promoter-TSS (NM_023626) 188 Appendix B Top 50 MED1 bound genes Gene Name Peak Score Annotation Kcnh1 99.24611 intron (NM_010600, intron 1 of 10) 1700012I11Rik 94.81264 intron (NR_045140, intron 7 of 8) Nufip1 91.64334 Intergenic ((TTTC)n|Simple_repeat|Simple_repeat) Zfhx3 91.11443 intron (NM_007496, intron 3 of 9) Zc3h4 89.1293 intron (NM_198631, intron 1 of 13) Scg2 88.60381 Intergenic (AT_rich|Low_complexity|Low_complexity) A430035B10Rik 87.32587 intron (NR_033518, intron 3 of 3) (L1Md_F2|LINE|L1) Smim18 85.62307 intron (NM_001286168, intron 5 of 7) Gcgr 84.97294 Intergenic (T-rich|Low_complexity|Low_complexity) Vwa5b1 84.09539 intron (NM_029401, intron 8 of 21) Ddx31 84.08543 intron (NM_001033294, intron 14 of 19) Rwdd4a 81.93472 Intergenic (MER45R|DNA|hAT-Tip100) Uaca 81.35915 intron (NM_028283, intron 1 of 18) Acsl5 79.49333 intron (NM_027976, intron 1 of 21) Nav2 78.13068 intron (NM_001111016, intron 1 of 38) Ahcyl1 78.05226 Intergenic (L1MCa|LINE|L1) Ccdc132 77.29261 intron (NM_001167751, intron 2 of 26) 1700085C21Rik 77.08764 Intergenic C1qc 76.91346 TTS (NM_007572) Cdh26 76.88364 Intergenic (MTD|LTR|ERVL-MaLR) Xylt1 76.88364 Intergenic Pcdh15 76.49445 Intergenic (L1M2|LINE|L1) 189 Resp18 76.17947 Intergenic Fmn1 75.9371 intron (NM_001285459, intron 13 of 16) B3galt5 75.9371 Intergenic Syt13 75.41393 intron (NM_030725, intron 4 of 5) Slc30a8 75.06465 Intergenic ((TTTTC)n|Simple_repeat|Simple_repeat) G6pc2 74.98505 intron (NM_001289857, intron 2 of 3) Rptoros 74.5135 intron (NR_045313, intron 1 of 2) (B4A|SINE|B4) Wbscr16 74.42128 intron (NM_033572, intron 9 of 10) Gm16894 74.2457 promoter-TSS (NR_037980) C130026I21Rik 74.01318 Intergenic ((TTTC)n|Simple_repeat|Simple_repeat) Rfx2 73.94263 intron (NM_027787, intron 1 of 17) Siah1b 73.87369 promoter-TSS (NM_009173) Ankrd60 73.29493 promoter-TSS (NM_001199955) Atp13a1 73.24262 intron (NM_133224, intron 1 of 25) (PB1D9|SINE|Alu) Ufm1 73.24262 Intergenic Ccser1 73.04315 Intergenic (T-rich|Low_complexity|Low_complexity) Prkca 72.89782 intron (NM_011101, intron 3 of 16) Ppard 72.76284 intron (NM_027185, intron 9 of 10) (B1_Mm|SINE|Alu) Sept9 72.3872 intron (NM_001113486, intron 2 of 11) (MTE2a|LTR|ERVL-MaLR) 4930578M01Rik 72.24453 Intergenic Foxa2 71.93475 Intergenic (L1MA4|LINE|L1) Pdcd4 71.93475 intron (NM_001170847, intron 8 of 13) Slc16a12 71.89878 intron (NM_172838, intron 1 of 6) (CT-rich|Low_complexity|Low_complexity) 190 Pan2 71.46757 Intergenic Cacnb2 71.1736 intron (NM_023116, intron 1 of 13) ((T)n|Simple_repeat|Simple_repeat) 9530052E02Rik 71.12265 Intergenic ((T)n|Simple_repeat|Simple_repeat) 1700028D13Rik 71.12265 Intergenic ((CAAA)n|Simple_repeat|Simple_repeat) Gm15545 71.10378 intron (NR_045266, intron 1 of 1) """@en ; edm:hasType "Thesis/Dissertation"@en ; vivo:dateIssued "2020-11"@en ; edm:isShownAt "10.14288/1.0392886"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Medical Genetics"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "Attribution-NonCommercial-NoDerivatives 4.0 International"@* ; ns0:rightsURI "http://creativecommons.org/licenses/by-nc-nd/4.0/"@* ; ns0:scholarLevel "Graduate"@en ; dcterms:title "Investigating the role of transcriptional coactivator MED15 in beta cell maturation"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/75618"@en .