It is well recognized that the regenerative capacity of pancreatic insulin-producing beta cells declines with age, with the most dramatic reduction occurring shortly after birth. However, the mechanism that accounts for this decline remains poorly understood. We sought to determine the molecular changes that attribute to this decrease in beta cell proliferation. Here, we show that activation of the DNA damage response (DDR) coincides with the decline in beta cell proliferation occurring during early postnatal life. In addition, we discovered a postnatal increase in the expression levels of replication stress-related DNA damage markers such as replication protein A (RPA) and phosphorylated Chk1 (pChk1). Furthermore, inhibition of the downstream effector Wee1 increases beta cell proliferation in vitro and in vivo. The results of this study provide insight into age-dependent changes in beta cells that lead to the increased incidence of cell cycle arrest. We hope that these findings will not only contribute to uncovering mechanisms underlying the age-dependent decline in beta cell proliferation, but also assist in the identification of factors that could potentially be pharmacologically targeted to improve beta cell function and stimulate their expansion for the management and treatment of diabetes mellitus.

The replicative capacity of insulin-producing pancreatic beta cells is dynamically regulated during development, maturation, and aging. Early in life, beta cells proliferate rapidly to expand beta cell mass but quickly become quiescent with age. While this decline is well documented, the mechanisms that underlie this age-dependent beta cell replicative senescence are still poorly understood. Using mouse genetics and in vivo quantitative proteomics approaches, we found that nutrient sensing plays an important role in controlling the proliferation of beta cells. We show that the transcription factor Nkx6.1 is required for expanding beta cell mass during the early wave of rapid postnatal beta cell proliferation by regulating the expression of the nutrient sensing receptors Glut2 and Glp1r. Furthermore, by quantitatively comparing the proteome of islets from young and aged mice, we found dynamic regulation of not only cell cycle proteins, but also proteins critical for beta cell function. Proteins important for insulin secretion and metabolic regulation increased with age, while proteins involved in expanding cell number declined with age. From our proteomic screen, we identified a protein deacetylase upregulated during aging. Pharmacologic inhibition of this protein promoted both rodent and human beta cell proliferation ex vivo, indicating the deacetylation activity of this candidate represses beta cell proliferation. Beta cell-specific deletion of the protein deacetylase increased rodent beta cell proliferation and mass in diabetic mice in vivo. Importantly, inhibition of this protein in human islets ex vivo did not negatively affect beta cell function or survival. Finally, we show that the protein deacetylase specifically regulates beta cell proliferation in conditions of elevated blood glucose through modulating the glucose-dependent MAPK pathway. Overall, our studies have uncovered dynamic regulation of beta cell proliferation from birth to advanced age and identified a viable therapeutic target for enhancing beta cell mass for the treatment of diabetes.

Over the course of development, regulation of gene transcription is the main mechanism by which pluripotent stem cells become restricted to the various distinct cell types found in the mature organism. Among the many different processes that regulate gene transcription, is the control of physical access to DNA and the genes for which it codes. DNA wound around histone proteins forms chromatin and the enzymes that modify the landscape of that chromatin control which regulatory elements, like promoters and enhancers, are active. This process confers different developmental competencies in cells, enabling them to respond uniquely to similar environmental and developmental signals, regulating gene transcription in turn. The study of these processes during in vitro differentiation of stem cells has enabled us and others to draw links between chromatin remodelers, transcription factors and cellular response to inductive cues during human development.

In Chapter 1, I explore the role of the lysine-specific demethylase (LSD1) during human pancreatic development using an in vitro system to differentiate human embryonic stem cells (hESCs) to the pancreatic endocrine lineage. Removal of LSD1 activity during a specific early time window of pancreatic development prevents endocrine formation. Investigation into enhancer regions occupied by LSD1 during this critical time window provided results that support a model in which LSD1-mediated decommissioning renders these enhancers insensitive to activation by external retinoic acid signaling.

In Chapter 2, I report my previous work dissecting the role of the transcription factor neurogenin-3 (NGN3) during human pancreatic development. Using the aforementioned hESC-based in vitro differentiation system, gain and loss-of-function studies showed that NGN3 is both necessary and sufficient to induce endocrine formation in human cells.

A final supplemental chapter provides an example of a hESC-based pancreatic differentiation protocol similar to the one employed for the studies outlined in Chapters 1 and 2 and discusses the importance of such model systems in dissecting the myriad mechanisms of human disease and development.

The insulin producing beta cells of the pancreas are essential for maintaining blood glucose levels. With either the destruction or dysfunction of beta cells leading to the onset of diabetes mellitus, it is necessary to understand the factors that maintain beta cells and their function. Nkx6.1 is a homeodomain transcription factor that is required for beta cell development. It is expressed in pancreatic progenitors and once pancreatic development is complete, Nkx6.1 expression becomes restricted to insulin producing beta cells. Although the functional importance of Nkx6.1 during the earliest steps of pancreatic development is known, it is less clear how Nkx6.1 regulates beta cell development and whether Nkx6.1 is required to maintain adult beta cells. Therefore, we developed mice with a conditional Nkx6.1 loss of function allele to examine the spatial and temporal requirements for Nkx6.1.

By conditionally ablating Nkx6.1 in pancreatic endocrine progenitors of mice, we found that Nkx6.1 promotes beta cell development at the expense of non-beta endocrine cell subtypes. Nkx6.1 achieves this through direct repression of genetic lineage determinants of non-beta endocrine cells. Once specified, Nkx6.1 continues to be required in beta cells to repress delta cell gene programs. Therefore, Nkx6.1 promotes and maintains beta cell identity through repression of alternative endocrine lineages.

In addition to regulating beta cell identity, we determined that Nkx6.1 is also a master regulator of mature beta cell function. By directly regulating genes required for central beta cell functions, Nkx6.1 maintains insulin biosynthesis and glucose import and metabolism. Consequently, deletion of Nkx6.1 in beta cells of adult mice causes the rapid onset of diabetes. By maintaining glucose import, Nkx6.1 also indirectly influences glucose stimulated beta cell proliferation. This role for Nkx6.1 is essential, as loss of Nkx6.1 in neonatal beta cells results in decreased beta cell mass expansion. Finally, we show that Nkx6.1 maintains identity of adult beta cells, as Nkx6.1-deficient beta cells adopt delta cell characteristics over time.

Overall, our studies uncover the functional importance of a beta cell specific transcription factor, Nkx6.1, as a central regulator of pancreatic beta cell identity and cellular function.

In response to external signaling factors, human pluripotent stem cells (hPSCs) can be induced to differentiate into diverse cell types. This unique property establishes hPSCs as a valuable model system to study human developmental processes. Furthermore, mature cell types generated through in vitro differentiation protocols can be studied in place of primary cells in circumstances where primary cells are difficult to obtain or exist in small numbers.Here we used established protocols to differentiate hPSCs towards the pancreatic beta cell fate. Using genomic assays, we characterized changes in the epigenome that accompany the commitment of differentiating cells to the pancreatic organ lineage. Focusing on distal regulatory elements, we investigated the function of FOXA pioneer transcription factors in the activation of pancreatic enhancers. We find that these factors are necessary for pancreatic lineage commitment and show that differential DNA sequence specifies distinct temporal patterns of FOXA recruitment to pancreatic enhancers, with profound effects on gene expression. Using models of liver and lung development, we show relevance of our findings across endodermal lineages. We next identified a group of pancreatic enhancers that are activated in response to retinoic acid signaling and whose subsequent deactivation is dependent on the enzyme LSD1. We demonstrate a critical role for LSD1 in limiting the duration of signal-dependent enhancer activation during pancreatic lineage commitment. We then integrated epigenomic analyses with data from genome-wide association studies to identify type 2 diabetes (T2D) risk variants within pancreatic enhancers. Using gene editing in hPSCs, we assigned target genes to variant-containing pancreatic enhancers and performed knockdown experiments in zebrafish to determine the developmental roles of these target genes. These experiments identify a developmental contribution to risk of T2D pathogenesis. Finally, we established a co-culture system between mature beta-like cells derived from induced pluripotent stem cells of donors with type 1 diabetes (T1D) and autoreactive T cell clones. This system is currently being used to identify a potential contribution of direct antigen presentation by beta cells to T1D pathogenesis. Altogether, our work demonstrates the utility of in vitro hPSC differentiation systems in modeling pancreatic development, disease risk, and disease pathogenesis.

Maturity-onset diabetes of the young type 1 (MODY1) is a form of monogenic diabetes caused by single base pair mutation in the hepatocyte nuclear factor-4-alpha (HNF4) gene that results in neonatal hyperinsulinemic hypoglycemia, macrosomia, and early onset diabetes. Although many HNF4 mutations have been identified in patients and their families, the molecular mechanism of MODY1 progression remains unknown. This study aims to understand the mechanism underlying HNF4 R141X mutations in pancreatic beta cells using human embryonic stem cell (hESC)-derived beta cells and genome editing CRISPR/Cas9 technology. By differentiating hESCs into insulin-producing beta cells using a stepwise protocol, we found that in vitro generated beta cells carrying the R141X mutation show phenotypes similar to clinical presentation of MODY1. At the immature stage, which is likely equivalent to the neonatal stage in vivo, mutant beta cells hypersecrete insulin at low glucose, which can be reduced by treatment of the potassium channel activator diaxozide. At the stage when wild-type (WT) beta cells become functionally mature and secrete insulin in response to glucose levels, HNF4 mutants hypersecrete insulin in both low and high glucose conditions. At both stages, mutant beta cells consistently secrete higher insulin in response to KCl-induced depolarization than WT. Furthermore, RNA sequencing of immature beta cells revealed upregulation of genes involved in cell cytotoxicity in R141X mutants. This is validated by an increase in cleaved Caspase-3 in mutants, suggesting that HNF4-mutated cells might be prone to cell death. Overall, our results show that hESC-generated R141X mutant beta cells behave similarly to beta cells from MODY1 patients and this mutation may induce apoptosis in beta cells.

One of future potential treatments for Type 1 diabetes is the use of stem cell-derived beta-cells (SC-beta-cells) to replace cells lost to autoimmune activity, but it is facing various challenges. The problematic composition and different cell populations are their examples. Since human primary pancreatic islets do not include them, it is assumed that stem-cell-derived enterochromaffin-like-cells (SC-ECs) are an aberrant population. However, similarity of serotonin synthesis between both SC-ECs and beta-cells suggests a potential hypothesis of SC-ECs being a transient population of mature SC-beta cells. Based on computational single cell analysis, CDX2 is found to be a major transcription factor determining SC-EC lineage fate. By generating CRISPR/Cas9 CDX2 knockout hESCs, here we aim to validate CDX2’s role in the SC-EC lineage.