Holger Willenbring - US grants

[unreadable] DESCRIPTION (provided by applicant): [unreadable] For many liver diseases, including those of inborn genetic origin, transplantation of donor organs is currently the only curative option. The derivation of cellular liver transplants from embryonic stem (ES) cells as an alternative source of tissue would need to be both autologous and free of the genetic disease, requirements that are not easily surmountable and also ethically controversial with ES cells derived from potentially viable fertilized or patient-specific somatic cell nuclear transfer embryos. One approach to produce autologous, disease-free ES cells from patients that may resolve these problems is the derivation of uniparental embryonic stem cells from the patient's gametes. Uniparental embryos such as parthenogenetic embryos have limited developmental capacity due to genomic imprinting but can produce ES cell lines. We have established that hematopoietic stem cells derived from murine uniparental ES cells can reconstitute adult hematopoiesis with no apparent pathology. To investigate the potential of uniparental cells for liver replacement, we now propose to use maternally (oocyte)-derived (parthenogenetic /gynogenetic) and paternally derived (androgenetic; two sperm genomes) fetal ES cell derivatives to functionally repopulate the liver in fumarylacetoacetate hydrolase (Fah) deficient adults. As a therapeutic approach, we will also transplant hepatic progenitors derived in vitro from uniparental ES cells. As each gamete contains a subset of the genome, diploid uniparental embryos are homozygous at a proportion of loci that are heterozygous in the gamete donor. This can be exploited to produce patient-derived ES cells without an errant allele (including large genetic lesions) in diseases associated with heterozygosity. Using the heterozygous phenotypic PiZ mouse model for alpha-1-antitrypsin deficiency, we will perform a proof of principle experiment involving elimination of the PiZ locus in uniparental ES cells. For patients of recessive genetic disorders, mutant allele-free, uniparental ES cell lines derived from parents or siblings could provide immune-compatible transplantable tissue, since MHC homozygous, mutant allele-free, uniparental lines with a matched subset of the recipient's MHC could be identified. Using mouse strains with different MHC loci, we will investigate the engraftment of MHC homozygous cellular liver transplants in MHC heterozygous recipients, i.e. determine the relevance of hybrid resistance.This proposal will investigate the capacity of embryonic stem cells derived from oocytes or sperm only to be used for liver transplantation and correction of genetic liver diseases. These embryonic stem cells would be derived from the respective patient's sperm or oocytes, and thus limit rejection problems associated with the use of existing embryonic stem cell lines. [unreadable] [unreadable] [unreadable]

DESCRIPTION (provided by applicant): Alcoholic liver disease eventually causes liver failure. The only curative therapy for patients with liver failure due to alcoholic liver disease is liver transplantation. Liver cell therapy is not effective in these patients because alcoholic liver disease is invariably associated with liver fibrosis, which impairs the engraftment of transplanted cells. Because of the long-standing shortage of donor livers, many patients with alcoholic liver disease die while waiting for liver transplantation. To improve the outcomes of these patients, we propose to work around liver fibrosis as a barrier to liver cell therapy. We hypothesize that the function of fibrotic livers can be improved by in vivo conversion of the cells that make up the fibrotic tissue into cells with hepatocyte function. Our hypothesis rests on the recent identification of hepatic transcription factors that can convert fibroblasts into therapeutically effective hepatocyte-like cells (iHeps) and success with in vivo conversion of fibrotic heart tissu into functional parenchyma. To establish in vivo iHep generation as a therapy for liver fibrosis, we will pursue 2 aims: (1) To convert myofibroblasts into iHeps in vivo. We will deliver the hepatic transcription factors to myofibroblasts with adenoviral vectors, which have been shown to efficiently transduce myofibroblasts in rat and mouse models of liver fibrosis. Most myofibroblasts are quiescent in advanced liver fibrosis, which will prevent loss of nonintegrating adenoviral vectors before conversion of myofibroblasts into iHeps is completed. Because myofibroblasts are the main source of collagen in the liver, our strategy is expected to not only increase the number of cells with hepatocyte function, but also prevent further collagen deposition in the liver. (2) To establish the use of nonintegrating adenoassociated viral (AAV) vectors for conversion of myofibroblasts into iHeps in vivo. Conventional adenoviral vectors are highly immunogenic and therefore not safe for human application. Because AAV vectors can persist in host cells without toxicity and integration into the genome, they are increasingly used for human gene therapy. To establish targeting of AAV vectors to liver myofibroblasts, we will screen a library of shuffled AAV capsids. In support of the feasibility of this approach, capsids targeting specifically pancreatic ¿-cells have recently been identified in this library. By improvig function and reducing fibrosis of the liver with a viral vector deemed safe for human application, the proposed strategy has potential as an alternative to liver transplantation for therapy of liver failure due to alcoholic liver disease.

Project Summary/Abstract This application targets the unmet medical needs of patients with bile duct paucity who often develop severe cholestatic liver injury that currently is only curable by liver transplantation. Our application builds on our finding that hepatocytes can form intrahepatic bile ducts (IHBDs) that function to reverse cholestasis in a mouse model of severe IHBD paucity. Conversion of hepatocytes-to-cholangiocytes has been reported, but our mouse model establishes that hepatocytes can build a therapeutically effective biliary system from scratch. We hypothesize that our mouse model has revealed the full potential of hepatocyte-to-cholangiocyte conversion because of the severity of its IHBD paucity and its unique genetic makeup, affecting both NOTCH and TGFbeta signaling, the main regulators of bile duct development. We propose to identify the mechanisms responsible for spontaneous IHBD restoration from hepatocytes in our mouse model with the long-term goal of enlightening tissue engineering approaches and to develop a therapy for diseases associated with bile duct paucity. In Aim 1 we will define the interplay between NOTCH and TGFbeta driving conversion of hepatocytes- to-cholangiocytes and assembly into IHBDs. For this we will use in vivo mouse models to perform pharmacologic and genetic approaches modulating TGFbeta signaling for examining the ?steps? (conversion and morphogenesis) of hepatocyte-derived de novo IHBD formation. In addition, we will investigate the mechanism by which loss of a TGFbeta effector influences the association of transcriptional cis-regulatory complexes to induce hepatocyte-to-cholangiocyte transdifferentiation. Aim 2, in which we will define the transcriptional network driving conversion of hepatocytes-to-cholangiocytes and assembly into IHBDs will also inform these efforts. For this we will investigate in converting hepatocytes in vivo which transcription factors are active and validate their efficacy in primary hepatocytes stabilized in a micropatterned co-culture system and a cholangiosphere morphogenesis assay. Using the insight gained from these experiments, we will express effectors of biliary differentiation and tube formation in vivo using non-integrating, nontoxic adeno-associated viral vectors for gene delivery to hepatocytes. Our research will generate new insight into the molecular regulation of hepatic cell identity, biliary development and regeneration.

PROJECT SUMMARY/ABSTRACT ? UCSF LIVER CENTER For the project summary/abstract please refer to the OVERALL/OVERVIEW section.

PROJECT SUMMARY/ABSTRACT ? UCSF LIVER CENTER For the project summary/abstract please refer to the OVERALL/OVERVIEW section.

Project Summary/Abstract Alcoholic liver disease is a major clinical challenge because it is associated with severe complications like liver cirrhosis and alcoholic hepatitis, which carry a high mortality. Our understanding of the mechanisms underlying alcoholic liver disease is still evolving; however, previous research has identified cells involved in the liver's response to alcohol-induced injury, including myofibroblasts, macrophages and oval cells (called ductular reaction in humans). Because of their unique and important functions, each of these cell types is a promising therapeutic target. Macrophages initiate and promote liver inflammation and activate hepatic stellate cells, which leads to the formation of collagen-secreting myofibroblasts and liver fibrosis and cirrhosis. Other types of macrophages resolve fibrosis and oval cells may produce new hepatocytes, although oval cell accumulation does not appear to prevent liver failure in patients with alcoholic hepatitis. Efficient gene delivery to these cell types would facilitate inhibiting or activating these functions, which would have many applications in research and therapy of alcoholic liver disease. To achieve this overall goal, we aim to target nonintegrating nontoxic adenoassociated viral (AAV) vectors to (1) myofibroblasts, (2) macrophage subsets, i.e., Kupffer cells and pro- inflammatory and anti-inflammatory infiltrating macrophages, and (3) oval cells in vivo. For this, we formed a collaboration that combines our expertise in AAV vector engineering and the cellular and molecular biology of liver injury and regeneration. To facilitate specific experimentation and support clinical translation, we will not only target AAV vectors to each of these cell types but also detarget them from every other organ and cell type in the body using a workflow that combines state-of-the-art AAV capsid evolution technology, faithful mouse models of alcoholic liver disease, next-generation sequencing-based analysis of vector biodistribution and on- target and off-target regulation of vector gene expression. To demonstrate the efficacy of the new synthetic AAV capsids, we will use them to determine which of the targeted cell types is most susceptible to in vivo reprogramming into hepatocytes. We hypothesize that oval cells can be most efficiently induced to become hepatocytes because they derive from closely related cholangiocytes or hepatocytes themselves. Therefore, in addition to providing efficient and precise tools for studies of alcohol-induced liver injury, the proposed project may establish opportunities for new therapies for liver failure and other life-threatening complications of alcoholic liver disease.

PROJECT SUMMARY/ABSTRACT Obesity-related disorders, particularly type-2 diabetes mellitus (T2DM), continuously increase in the US and worldwide with an estimated 1.9 billion overweight adults and over 650 million obese individuals globally. While the mechanistic underpinnings of obesity-induced T2DM remain a topic of investigation, central features include a pro-inflammatory environment and dysregulated lipolysis in adipose tissue leading to elevated levels of circulating free fatty acids with subsequent ectopic accumulation of lipids in multiple tissues. The combination of nutrient excess and pro-inflammatory signaling in turn results in insulin resistance in multiple tissues impairing glucose uptake by muscle and adipose tissue and release by the liver as well as ß-cell function, ultimately resulting in overt diabetes. Interrogation of the complex interplay between these key tissues has, thus far, only been possible using animal models, which do not lend themselves to high-throughput approaches and frequently deviate from humans in key metabolic features, thus greatly impeding efforts to discover treatments for insulin resistance and T2DM. Here we propose to develop an essential set of human induced pluripotent stem cell (iPSC)-derived key metabolic tissues for glucose and fatty acid uptake/release, i.e., liver (L) and adipose (A) tissue, and insulin secretion, i.e., islets (I), in conjunction with an immune component, i.e., macrophages, using interconnected microphysiological systems (MPS). This LAI-MPS will allow for the pharmacological interrogation of glucose and insulin sensitivity in the context of normal tissue interactions, lipid overload and chronic inflammation to address the following major current shortfalls. In 6 milestones we will progress from the 1) generation and metabolic characterization of human iPSC-derived hepatocytes, adipocytes, ß-cells and macrophages ? to 2) Development of optimized microfluidic devices for iPSC-derived hepatocytes, adipocytes and ß-cells ? to 3) Establish on-chip insulin and glucose sensitivity assays for, WAT and islet MPS. As part of the UH3 phase we will then begin integration of MPS platforms by 4) integration of liver and fat MPS with common medium and determination of insulin sensitivity using in-line sensors ? and 5) Use liver and WAT MPS for the generation and quantitation of insulin resistance following scaling of WAT MPS and inclusion of pro-inflammatory macrophages ? and finally 6) integrate islet, liver, and WAT MPS and determine impact of pharmacological and pro-inflammatory modulation on glucose tolerance and ß-cell function. Ultimately, this disruptive technology will enable the rapid screening of pharmacological and environmental compounds for beneficial or detrimental effects on insulin sensitivity and for the detection of pharmacogenetic interactions.

Project Summary/Abstract Pioneering clinical trials of hepatocyte-targeted liver gene therapy in hemophilias have established the principal therapeutic efficacy of adenoassociated virus (AAV) vectors. Clinical trials have also revealed limitations of current AAV vectors, mainly lower than expected therapeutic efficacy and dose-dependent toxicity undermining both safety and efficacy. These clinical findings show that animal models, even monkeys or mice engrafted with human hepatocytes, fail to accurately predict the performance of AAV vectors in humans. The limitations of current AAV vectors need to be overcome for broad application of AAV liver gene therapy because most liver diseases require more hepatocytes to be transduced to achieve a therapeutic effect than hemophilias. To maximize the therapeutic effect that can be achieved within a safe AAV vector dose limit, we will identify AAV capsids that transduce hepatocytes in the human liver with maximum efficiency but no or limited off-targeting. For this, we will harness the near-clinical conditions provided by normothermic machine perfusion (NMP) of human livers. We will compare capsids currently used in clinical trials of liver gene therapy to engineered capsids reported to efficiently target mouse liver or human hepatocytes engrafted in mice. To compare capsids side by side, we have established analysis of AAV vector-expressed barcodes by single-cell RNA sequencing of comprehensive cell populations isolated from human livers after NMP. We will also target capsids de novo in human livers maintained by NMP by screening a library of 1 million chimeric capsids generated by shuffling of the DNA sequences of naturally occurring AAV capsids. In addition to achieving unprecedented levels of efficiency and specificity of transduction of hepatocytes, we aim to establish transduction of cholangiocytes, thereby facilitating the development of gene therapies for biliary diseases. We will also target activated hepatic stellate cells (myofibroblasts), the source of excessive collagen in liver fibrosis, and reactive cholangiocytes, which form the ductular reaction characteristic for cholestatic liver diseases. Efficient and specific in vivo gene delivery to these pathogenic and abundant cells will facilitate therapeutic strategies based on inactivation or repurposing, for example, reprogramming into hepatocytes. By identifying or generating capsids that transduce hepatocytes and other therapeutically relevant liver cell types with the highest level of efficiency and specificity our results will directly inform clinical trials that are at the planning stage and provide the basis for extending the reach of AAV liver gene therapy to common liver diseases like fatty liver disease and biliary diseases, including liver fibrosis as their common end stage.