Douglas A. Melton - US grants

The long term aim of this project is to understand the role of RNA processing in gene expression and animal development. One of the main problems in studying eukaryotic mRNA biogenesis has been isolating the RNA substrate for processing reactions, namely pre-mRNA. This problem has been circumvented by employing an efficient prokaryotic RNA polymerase and promoter to synthesize large amounts of eukaryotic pre-mRNAs in vitro. Most importantly, these in vitro synthesized pre-mRNAs are accurately processed (spliced) following injection into frog oocyte nuclei. The in vitro transcription-oocyte injection system makes it possible to design and synthesize mutant mRNAs and pre-mRNAs and to test their biological function in a living cell. This experimental design will be used to identify and characterize RNA sequences important for RNA splicing, transport, and 3' end formation. In addition, special attention will be given to analyzing how RNAs are localized to particular regions of egg cytoplasm. Several cloned eukaryotic genes, including globin genes, histone genes, and genes encoding maternal mRNAs, will be used to study these processing events. As an extension of the in vivo assays, an in vitro RNA processing system will be established using isolated oocyte nuclei. If successful, these in vitro studies will significantly advance the understanding of the biochemical mechanisms of RNA processing, particularly splicing. The proposed experiments will lead to a better understanding of how animal cells synthesize and regulate the use of messenger RNAs. This basic information may be important for understanding how RNA processing defects cause cellular dysfunction as in the case of some human thalassemias.

The long term objective of this project is to explain how embryonic cells are committed to express certain sets of genes later in development. At present, relatively little is known at the molecular level about now these decisions are made. We have identified and cloned genes that may be involved in the process of embryonic determination in frogs. These include (1) genes that are expressed maternally and whose transcripts show the unusual property of being localized in eggs and (2) genes that contain homeoboxes, called Xhox. The Xhox genes are of interest because the homeobox is a conserved sequence that is found in Drosophila genes known to be important for specifying cell fates in flies. In addition, we propose to analyze the transcriptional regulation of a gene called GS17 because it is first transcribed precisely at the mid-blastula transition, the time at which the embryonic genome is activated. Together these genes provide us with the tools for a detailed study of the molecular biology of early determinative events in frog development. Our specific experimental objectives are: 1. To determine how maternal RNAs are localized to different regions of the frog egg. 2. To determine the function of the proteins translated from these localized mRNAs. 3. To continue developing anti-sense RNA and sense RNA injections using the SP6 system in order to probe gene function during early development. 4. To investigate the developmental expression and function of Xenopus homeobox genes. 5. To identify the cis and trans-acting signals involved in the selective activation of the embryonic genome. The medical significance of this work derives from the possibility that understanding how animals develop and how cells selectively express genes may be helpful for understanding certain diseases.

The long term objective of the work proposed here is to understand the factors that specify cell fates, both differentiation and patterning, in early vertebrate development. Special attention is given to the induction of neural tissue. Recently, it has been shown that peptide growth factors act as morphogens or determinants in that they change the fates of cells from ectodermal to mesodermal pathways during early Xenopus development. Two embryonic peptide growth factors, fibroblast growth factor and the Vgl protein, are involved in mesoderm induction. It is also known that at least one other factor is needed to account for all the mesodermal cell types that are formed. This other factor is likely to be a member of he transforming growth factor beta family and may be the inducing factor secreted from XTC cells, XTC-MIF. We have now cloned several new members of the TGFbeta gene family, in Xenopus, one of which may be the XTC-MIF gene. We propose to study the expression these new TGFbeta genes in early development and determine their roles in mesodermal and/or neural induction. A second major aspect of the proposal stems from our discovery of a neural inducing factor. This factor, secreted from a murine macrophage cell line, induces neural tissue in isolated Xenopus ectodermal cells that would otherwise form only ciliated epidermis. Animal caps treated with PIF organize the induced dorsal CNS and mesodermal tissues into a clear axis with antero-posterior polarity. Taken together, these findings suggest that PIF has many of the properties associated with a Spemann organizer. We propose to purify the PIF protein, clone the gene encoding its Xenopus homologue and study the function and expression of this novel neural inducing factor.

The general area of research considered in this application is that part of developmental biology concerned with how cell fates are specified and how the body plan is established during vertebrate embryogenesis. The specific focus of this project is the development of frog endoderm (the gut and its derivatives) and the organogenesis of the pancreas. In contrast to the understanding of mesodermal and ectodermal development, relatively little is known about how the endoderm is initially formed in any vertebrate. There is evidence that cytoplasmic determinants are responsible for the self-differentiation of endodermal cells in Xenopus and recent work suggests that inductive interactions by peptide growth factors may also play a role. One aim is to study the molecules and mechanisms involved in forming embryonic endoderm. We also propose to study organogenesis by focusing on the pancreas. Our initial aims are to understand how the rudimentary gut tube (endodermal epithelium) is patterned along the anteroposterior axis to form the pancreas at a specific site. Special attention will be given to the hypothesis that signals from the notochord play a role. Specific experimental objectives include: To determine the role of mesoderm (the notochord early on and mesenchymal cells in later development) in gut differentiation with emphasis on pancreatic development. To determine the role of inducing molecules, including Vg1,in endodermal specification. To characterize the molecular signals (both epithelial and mesenchymal) responsible for the specification and differentiation of the pancreas. The medical significance of this work relates to the formation of the pancreas and, in the longer term, a possible treatment for Type I juvenile) diabetes.

DESCRIPTION (provided by applicant): Human embryonic stem (hES) cells are a unique scientific and medical resource. These pluripotent cells are self-renewing and have the capacity for differentiation into many types of cells. The potential for using hES cells in transplantation therapies for numerous diseases, including diabetes and Parkinson's disease, has been widely discussed. Despite their considerable potential, there is rather little known about the genetic regulation of human embryonic stem cells. For example, the genes and signaling pathways that control self-renewal of hES cells have not been identified. Similarly, the genes that regulate hES cell differentiation into various embryonic germ layers and specialized cells are unknown. In the absence of understanding the genetic control of hES cell self-renewal and differentiation, the experimental manipulation of hES cells, either for research or therapeutic purposes, is bound to remain in a rudimentary state. This project will characterize the transcriptional program of human embryonic stem cells and discover genes involved in their self-renewal and differentiation. The genetic program that characterizes undifferentiated hES cells will be analyzed using DNA microarrays. In addition, a gene "trap" strategy will be used to identify genes that potentially control hES differentiation into the three embryonic germ layers. The function of selected genes will be explored using conditional expression during hES cell differentiation. Four lines of hES cells will be used for these studies. They are: H1 (WA01), H9 (WA09), H13 (WA13) and HES-1 (ES01); the NIH registry number is in parentheses.

[unreadable] DESCRIPTION (provided by applicant): [unreadable] Type I diabetes is characterized by the permanent loss of pancreatic insulin-producing beta cells. Our goal is to understand the mechanisms determining pancreatic beta cell mass, as a first step toward the development of regenerative therapies for diabetes. In collaboration with the BCBC Coordinating Center and members of the consortium, we will use transgenic mouse technology to examine the hypothesis that beta cells have a significant regenerative capacity, which is modulated by specific signals. The specific aims are to: [unreadable] 1. Develop a system for regulated ablation of pancreatic beta cells. Our preliminary studies indicate that mice can recover from a pulse of beta cell ablation. We will characterize in detail the physiological and histological aspects of ablation and recovery, in order to understand the mechanisims of beta cell regeneration. [unreadable] 2. Determine the cellular origins of regenerating beta cells. We will employ genetic lineage tracing to definitely determine the contributions of stem cells and pre-existing beta cells to beta cell regeneration. [unreadable] 3. Characterize the signals that regulate beta cell mass. The relative importance of blood-borne signals will be determined, with emphasis on the role of insulin and glucose metabolism. In addition, we will assess the significance of beta cell dedifferentiation in vivo and in vitro. [unreadable] Insights from these studies will be applied to primary beta cell cultures, with the aim of improving yield, preventing dedifferentiation and minimizing cell death [unreadable] [unreadable]

The endocrine pancreas is both the substrate for fundamental questions in developmental biology as well as the target of the disease diabetes mellitus, which affects millions of individuals worldwide. A detailed genetic and molecular understanding of pancreatic endocrine development will be essential if we are to manipulate islet cell fate and numbers in vivo. Our emerging understanding of pancreatic development is one in which, through interactions with surrounding mesenchyme, the initially unpatterned pancreatic epithelium is successively sub-divided into exocrine and endocrine compartments which subsequently differentiate, the latter containing the |3eta cells that produce insulin. Among the factors that dictate islet cell specification are many genes whose identities are now known. In fact, it is now possible to order these genes into a first order genetic regulatory network in terms of order of gene action, expression and functional interdependencies, and hierarchical relationships. This grant therefore poses the question: Do we already know enough, and can we sufficiently augment that knowledge, to begin to use this information to systematically engineer islets and islet cells in vitro? To accomplish this ambitious goal, we will undertake three Specific Aims that are highly integrated with multiple components of SysCODE. In Aim 1, we will generate complete gene lists for key early states in the developing endocrine pancreas. An initial effort in this area has already been accomplished (Gu et al., Development 131, 165-79, 2004). We will now augment this information with data from additional developmental stages, selected mutant states and first-generation proteomic analyses. In Aim 2, in conjunction with the SysCODE Computational Team, we will develop methodology to order these genes and selected proteins into a definitive gene regulatory network (CRN) in a format that is useful to both biologists and tissue engineers. Lastly, in Aim 3 we will collaborate with the SysCODE Tissue Engineering Team to implement a stem cell based, engineered model of pancreatic islet development and we will use the GRNs generated in Aim 2 to optimize pancreatic endocrine fate specification. In the out years of the grant, we will transplant the engineered islets into diabetic mice and assess their physiological function. Collectively, these efforts, in conjunction with the rest of SysCODE, will establish a transforming paradigm for regenerative medicine.

DESCRIPTION (provided by applicant): The research in this proposal will apply genetic, cell biological and chemical screening approaches with the goal of identifying novel treatments for type 2 diabetes (T2D), thus addressing thematic areas 1 and 2 of this grants program. Due to a relative deficiency in pancreatic beta cells, T2D patients are unable to produce sufficient insulin to control their blood glucose levels. If the number of beta cells in T2D patients could be increased, their glycemic control could be significantly improved, thus delaying or preventing the devastating consequences of chronic hyperglycemia. It is well established that beta cells possess the capacity to dramatically increase their numbers by replication, suggesting the possibility of harnessing this replicative potential as a therapeutic avenue for T2D. However, genes that specifically control beta cell replication are largely unknown. Likewise, small molecules that can boost beta cell replication rates have not been described. To identify genes that control beta cell replication, we propose to use microarray analysis to compare the transcription profiles of actively replicating populations of beta cells to the profiles of quiescent populations. In this way, we will be able to identify candidate regulatory genes whose expression either positively or negatively correlates with replication status. In parallel, we will undertake a high-throughput small molecule screen on cultured pancreatic islets to identify compounds that can stimulate the rate of beta cell replication in vitro. Genes and compounds identified by these two analyses will then be tested in mouse models to determine their ability to regulate beta cell replication in vivo. Thus, the studies in this proposal are likely to lead to the identification of novel therapeutic targets and drug candidates for the treatment of diabetes. PUBLIC HEALTH RELEVANCE: The incidence of type 2 diabetes is rapidly increasing in the United States and worldwide. The negative impacts of long-term diabetes on a patient's health are significant, and carry with them a heavy financial burden. The therapies that will be developed as a result of the proposed research could be used to treat type 2 diabetes, and delay or prevent the devastating consequences of long-term disease. Thus, these therapies will both alleviate patient suffering while simultaneously reducing the financial costs of this disease.

DESCRIPTION (provided by applicant): Lineage reprogramming, whereby cells of adult organs are converted from one specialized cell type into another, has emerged in recent years as a promising approach to regenerate cells lost due to disease or injury. This regenerative approach could prove valuable for the treatment of Type 1 diabetes where insulin secreting [unreadable]-cells are destroyed by autoimmune attacks. The long-term goal of this proposal is to devise strategies to reprogram adult cells of endodermal organs into pancreatic ?-cells for cell replacement therapies. In a recent study, our research groups showed that pancreatic exocrine cells can be reprogrammed into insulin secreting ?-cells in adult animals by a combination of three transcription factors. We propose to expand on this study and build a set of new tools to investigate the mechanisms of reprogramming and study the reprogramming of several related endodermal cell types, including pancreatic exocrine cells, liver cells and intestine cells, into [unreadable]-cells. In Specific Aim I, we will develop a new generation of viral vectors to dissect the molecular and epigenetic mechanisms of in vivo reprogramming of exocrine cells into ?-cells. In Specific Aim II, we will develop inducible mouse genetic models to investigate the in vivo reprogramming of adult endodermal cells to ?-cells. In Specific Aim III, we will use chemical and genetic screens to reprogram hepatocytes to ?-cells ex vivo. Together, these studies are expected to yield important insights into the mechanism of ?-cell reprogramming and help define additional cell types and reprogramming methodologies to regenerate ?-cells in the adult. Such tools and knowledge will form the foundation for developing novel cell replacement therapies for Type 1 diabetes. PUBLIC HEALTH RELEVANCE: Our proposed studies are broadly aimed at developing novel cell replacement therapies for pancreatic ?-cells in order to treat type 1 diabetes. Specifically, we will develop new research tools to define the cell types and conditions whereby cells of pancreas, liver, and intestine can be converted into functional ?-cells.

DESCRIPTION (provided by applicant): The goal of this project is to reconstruct human Type 1 diabetes in a mouse model system. This goal will be accomplished through the achievement of two specific aims. First, we will develop optimized immunodeficient mice that will provide an ideal vessel for the three human tissue types most relevant to T1D, namely the hematopoietic system, the thymus, and pancreatic beta-cells. These mice will be tested and validated using human fetal tissues. Second, we will generate each of the three tissue types listed above using directed differentiation of induced pluripotent stem cells derived from a panel of T1D patients. These T1D tissues will be implanted in the immunodeficient mice, and the mice will be monitored to observe the onset of autoantibody production and autoimmune destruction of beta-cells. This mouse model of human T1D will allow, for the first time, a detailed analysis of the development of autoimmunity in real time, thus opening the possibility of identifying novel therapeutic avenues for the treatment of the disease. This system will also be used to test the relative contributions of various risk alleles and environmental factors to the emergence of the T1D phenotype. The novel mouse strains and directed differentiation protocols developed as part of this project will be of great use not only for the study of T1D, but will be of broad use for the study of immunology and transplantation medicine across many diseases. PUBLIC HEALTH RELEVANCE: This research will identify new ways to treat Type 1 diabetes. In addition to this, new strains of mice will be developed over the course of the project that will be of great use to a large community of researchers studying human immunology. Also as a part of this project, new methods will be developed to convert induced pluripotent stem cells into clinically useful cell types, such as blood cells and insulin-producing beta- cells, which could ultimately be transplanted into patients lacking those cells due to injury or disease.

? DESCRIPTION: Mellitus results from failure pancreatic islets leading to an increase in morbidity and mortality. Mechanistic studies of this disease are hindered by low availability, high variabiliy and the cost of human islets. Our recent advances have led to the first successful method to generate mature, glucose sensing- insulin secreting b cells from human embryonic stem (ES) cells in vitro. This method, and its application using human iPS cells, provides a virtually unlimited supply of standardized human ? cells. Moreover, as the ? cells can be prepared from patient iPS cells, normal and diseased states can be analyzed. This advance provides a renewable source of b cells for cell replacement therapy for insulin dependent diabetics and the opportunity to perform rigorous disease modeling to identify therapeutic targets for all diabetics. Despite these advances, challenges remain. Robust, sensitive and routine technologies to assess ? cell function are lacking. Further, it is unlikely that b cells by themselves will recapitulate the complex biology involved in islet function. As such, the proposed research aims to combine approaches in stem cell and islet biology with tissue engineering to design, build and test new technologies for generating human islets in vitro and assessing their function in microfluidic devices. Using reverse engineering principles we will design and build a bio-inspired microfluidic chip that supports the survival and function of cell clusters containing b cells. This islet chip design will enable rigorous and sensitive evaluation of ? cell function that goes beyond current technologies. This chip will also provide a platform to evaluate human cadaveric islets by quantifying their functional variability. In parallel, we seek to generate whole islets i vitro using a combination of top-down and bottom-up tissue engineering approaches. Endocrine progenitors from human stem cells (ES and iPS) will be introduced to a chip designed to screen a combination of substrates, matrixes and mechanical forces to identify a niche that supports differentiation to islet-like structures with all endocrine cell types. The resulting stem cell-derved islets will be evaluated in our islet chip to describe the functional differences between these ES-islets, bcells alone and cadaveric islets. Finally, we will use these technologies for disease modeling and drug screening by generating healthy and diseased islets from iPS cells representing different disease states (healthy, type 1, type 2 diabetes, MODY) and evaluate the function and response of these islets to diabetes drugs. These studies will provide validated technologies that will increase our understanding of diabetes and speed development of new therapies.

? DESCRIPTION (provided by applicant): To date, studies of human type 1 diabetes (T1D) have failed to provide a mechanistic understanding of the causes of the disease, largely because patients must be analyzed long after the autoimmune attack was initiated. Our ignorance of the key molecules and cells mediating the initiation and progression of human T1D may well underlie the paucity of significant new therapeutic interventions. Here we propose to reconstruct human T1D, using iPS-derived ? cells, thymic epithelial cells (TEC), and immune systems derived from T1D patients implanted in a novel immunodeficient mouse model based on the NOD-scid-IL2rgnull (NSG) strain. To accomplish our goal, we propose 2 aims. Aim 1 will validate human immune and pancreatic beta cell functions in optimized immunodeficient mice (OPTI-MICE). Currently available OPTI-MICE will be improved using genetic techniques to: 1) enhance engraftment of human cells; 2) allow for spontaneous and inducible hyperglycemia; 3) support expression of human HLA alleles and cytokines; and 4) knockout mouse genes that impair human cell engraftment and function. This suite of improvements will be validated using implantation of fetal human stem cells and tissues. Aim 2 will reconstruct human T1D in mice using cells derived from Type 1 diabetic iPS cells. These iPS cells will be used to produce the three key cell types: hematopoietic stem cells (HSC) that will generate immune systems, TEC, and ?-cells, all integral to the pathology of T1D. These cells will be derived through the use of directed differentiation and reprogramming strategies. We have been successful in generating functional human ? cells from human control and T1D patient iPS cells by a recently developed multi-step protocol for directed differentiation, providing a standardized and reproducible source of ? cells our studies. Functional human HSC will be generated using two technologies: 1) directed differentiation of iPS cells and 2) reprogramming of differentiated hematopoietic cells using defined factors to derive induced-HSCs. Functional human TEC will be generated using directed differentiation protocols similar to those used to achieve fully differentiated human ? cells. Each cell type will be subjected to rigorous analysis in vitro and in vivo to ensur full functionality. Differentiated cells derived from a single donor will be co-transplanted into OPTI-MICE, thus reconstituting an individual patient's disease in an animal model. Transplanted mice will be carefully monitored for the emergence of autoantibodies and autoreactive T-cells, and for destruction of ?-cells. This new model of human T1D will permit detailed observation, manipulation, and analysis of T1D as it progresses, enabling us to determine which cells and antigens initiate T1D. Such mechanistic insights will properly inform new approaches to curing, or even preventing, this disease. To accomplish our goal, we have assembled an interactive team of researchers formed using seed monies from the Helmsley Charitable Trust. We have now been working together and meeting regularly for over 5 years and have expertise in the relevant areas required to accomplish this project.

DESCRIPTION (provided by applicant): The diagnosis of Type 1 diabetes (T1D) is currently limited by methods that detect the disease at relatively late stages, when blood glucose levels rise and autoimmune attack is well underway. By this stage, given the treatments presently available (insulin injection), the b cell loss is de facto, irreversible. Biomarkers for early detetion of ? cell stress, before this late stage, would improve diagnosis and open up new windows to monitor and possibly treat the disease. We will produce functionally mature, never-before-stressed, ? cells from human pluripotent stem cells and use these cells to identify biomarkers for the earliest stages of ? cell stress by providing various stress conditions ex vivo. In additin, we can use stem cells from T1D and control patients to investigate differences in their ? ell responses to stress. In other words, we can ask: are ? cells in T1D patients especially sensitive to stress or in other ways defective before an immune attack? Successful identification of early ? ell-specific biomarkers for stress will enable earlier detection and diagnosis, better tracking f progress from asymptomatic to clinical presentation, and a more mechanistic understanding of the earliest stages and possibly initiating events of human T1D.

Recent advances in research conducted at Harvard Stem Cell and Regenerative Biology, Harvard Stem Cell Institute and later further developed at Semma Therapeutics, demonstrated that large quantities of insulin-producing cells can be generated from human iPSC. Since human iPSC can be derived from a simple blood sample from any patient, this technology now allows for generation of unlimited number of insulin-producing cells in a personalized manner. Other clinical research has shown that normal glycemic control can be restored in type 1 diabetes patients when they are transplanted with enough insulin-producing cells. The Boston Autologous Beta Cell Therapy [BAIRT] Program was formed between Brigham and Women Hospital (BWH), Dana Farber Cancer Institute (DFCI), Harvard Stem Cell Institute (HSCI), Joslin Diabetes Center (Joslin), and Semma Therapeutics; this group aims to use these approaches to derive patient-specific iPSC lines, manufacture clinical grade insulin-producing cell products, and perform autologous cell transplants to treat diabetes patients. As part of the BAIRT program, the goal of the proposed project is to establish a cGMP standard operating protocol to derive iPSC lines from freshly collected human blood. The protocol has been developed by HSCI in collaboration with DFCI as well as input from other partners in the Boston Autologous Beta Cell Therapy Program. It is expected during the project period DFCI will establish the capacity and know-how to begin deriving and banking cGMP-grade patient-specific iPSC lines from blood samples collected from diabetes patients. Joslin and BWH have begun the recruitment of suitable candidates with diabetes resulted from pancreatectomies, so that research subjects will be available to provide their cells for iPSC derivation. The project will facilitate the testing and development of assays to characterize and qualify iPSC products intended for autologous cell therapies. This valuable information will prove crucial toward developing personalized cell therapies, a long-sought goal in regenerative medicine.