Novel approach to model Non-Alcoholic Fatty Liver Disease using human Pluripotent Stem Cells.

Non-Alcoholic Fatty Liver Disease (NAFLD) is a growing health burden that affects about 30% of the general population in western countries; being strongly associated with common metabolic disorders as diabetes and obesity, NAFLD incidence is exponentially increasing. The hallmark of NAFLD is steatosis, defined by fat accumulation in the main cell type of the liver, the hepatocytes. Chronic steatosis induces hepatocyte damage and eventually progresses towards liver inflammation, a stage defined as Non-Alcoholic Steatohepatitis (NASH), which culminates with cirrhosis and liver cancer. Given its high prevalence and consequences, NAFLD will have a relevant health and social impact in our society in the near future. Indeed, currently there are no specific therapies approved for the treatment of NAFLD, other than dietary and life style habits modifications, which are solely aimed to slow disease progression. The only available treatment is liver transplantation, an expensive and challenging solution that carries considerable risks and requires lifelong medication. Most importantly, patients often die while on the liver transplant waiting list, due to shortage of liver donors. This scenario highlights the urgency to identify new therapeutic options against NAFLD and NASH progression. Unfortunately, the mechanisms directing NAFLD pathogenesis remain elusive as the discovery of efficient drugs. Both are hampered by the lack of a suitable model that recapitulates all the aspects of the human physiopathology. Indeed, current studies mainly rely on animal models, which do not faithfully mimic human NAFLD. Plus, along with the ethical implications of these studies, in vivo experiments are costly and the clinical translation of the results is often limited.
Therefore, the aim of this project is to generate an in vitro system that recapitulates all the defining features of NAFLD. The different hepatic cell types will be generated taking advantage of human Induced Pluripotent Stem Cells (hIPSCs) and will be grown in a collagen based 3D environment. The hepatic platform will be challenged with free fatty acids commonly included in the western diet. Multicellularity is the crucial component to investigate how hepatocytes steatosis triggers consequent inflammation and fibrosis. Moreover, the system allows for long term experiments, an essential aspect in modelling chronic diseases.
This novel tool will provide an advantageous alternative to perform large scale experiments using only human cells. Validation of the relevance of the system for NAFLD modelling will hold great potential in replacing in vivo models, not limited only to liver disease. Indeed, this study will provide the first proof-of-principle for in vitro modelling of complex diseases sustained by significant inflammatory and fibrotic components. Thus the system could successfully be adjusted to model other pathologies that are associated with inflammation and fibrosis, finding broad applications in the disease modelling field.

Non-Alcoholic Fatty Liver Disease (NAFLD) ranges from simple steatosis to liver inflammation and/or fibrosis, and it is recognized as a risk factor for liver failure and hepatocellular carcinoma. Despite its high prevalence (25-30% in the general population), specific therapy is not available and drug discovery has been hampered by the lack of a suitable system that encompasses all the different aspects of NAFLD. Animal models are widely used to study NAFLD, but they fail to fully reproduce the human pathophysiology while in vitro models are currently not available due to the difficulty to grow primary hepatic cells. In addition, current culture systems do not allow for long term studies, and, most importantly, do not include the diversity of cell types consequent to hepatocyte damage.
This proposal aims to develop a novel platform for modelling the different features of NAFLD using human Induced Pluripotent Stem Cells (hIPSCs). For that, I will take advantage of a culture system that I recently established, consisting of 3D (co)-cultures of hepatocytes, cholangiocytes, macrophages and stellate cells derived from hIPSCs. I will first demonstrate that NAFLD can be modelled by metabolic stimuli in normal hIPSCs or hIPSCs with NAFLD-associated polymorphisms. This will allow to demonstrate the interest of this new platform to study mechanisms of hepatocytes steatosis and lipotoxicity including the processes influencing the inflammatory and the fibrotic response associated with progression of NAFLD. This system will be further validated by direct comparison with the hepatic transcriptome of NAFLD patients. Finally, the last step of this program will establish the interest of this platform for drug screening and target validation. The possibility to model NAFLD using hIPSCs holds great potential in providing a unique alternative to animal models and an attractive approach for the discovery of urgently needed novel therapeutic targets.

The overall incidence of Non-Alcoholic Fatty Liver Disease (NAFLD) in the western world (30%) and the pandemic in metabolic disorders as obesity and diabetes is likely to significantly increase the number of NAFLD patients. Indeed, NAFLD is the hepatic manifestation of the metabolic syndrome, and it is also linked to a higher risk of cardiovascular disorders. Despite the exponentially increasing trend of NAFLD patients over the past decade, there are currently no therapies against NAFLD progression. This significantly contributes to the elevated number of patients with end-stage liver disease requiring liver transplantation for NAFLD complication, namely cirrhosis and hepatocellular carcinoma. NAFLD has a major social and economic impact, representing a growing global health challenge which will overwhelm healthcare systems.
To date, animal models represent the best system to investigate NAFLD pathogenesis and test new therapeutic interventions. Accordingly, more than 20 different NAFLD mice models have been developed, that can be categorised into diet-induced and genetically modified. However, despite the wide variety of models available, they all present some limitations and do not recapitulate the entire clinical manifestation of NAFLD. Therefore, the development of an in vitro humanised platform comprising all the features of NAFLD would have a major impact on several areas, with great 3Rs potential. More specifically, the proposal will impact the 3Rs field in terms of:

- Replacement of animal models for NAFLD: a large number of reports are published yearly on NAFLD in vivo models. In the past 5 years, 486 publications have used a dietary model of NAFLD (metrics calculated considering the methionine-choline deficient (MCD) diet and the high fat high fructose diet models), and 1733 studies were performed on genetic NAFLD models (metrics on the ob/ob and db/db mice). The numbers dramatically increase when including other NAFLD models. Importantly, the decreasing trend of publication with existing models (i.e. publications with MCD model: 59 in 2014, 46 in 2015, 40 in 2016) translates in the development of new in vivo models that aim to overcome current limitations. Therefore, an in vitro system representative of the human physiology will provide an advantageous alternative.

- Replacement of animal models for liver fibrosis: liver fibrosis is a common feature of chronic hepatic damage independently of the aetiology. Thus, the knowledge provided by this model will also significantly contribute to the reduction if not replacement of the models developed to investigate liver fibrosis (publications on liver fibrosis and animal models: 1288 in 2014, 1401 in 2015, 1037 in 2016). Therefore, the NAFLD platform will have an additional impact on a broader number of hepatic disorders characterised by inflammation and fibrosis.

- Replacement of animal model to study fibrosis: organ fibrosis is a common end point of various inflammatory diseases. This complex process results from intense cross-talk among epithelial, immune and fibrotic cells and it is hardly reproducible in vitro. Animal models are intensively used for basic and pharmacologic research in this field (publications on fibrosis and animal models: 1438 in 2014, 1603 in 2015, 1237 in 2016). The proposed system will focus on cellular interactions in NAFLD, but the knowledge derived could be applied to a variety of fibrotic diseases. This proposal could pioneer a novel approach to in vitro fibrosis models, which will reflect on a broader impact on animal research, not limited to liver disease.

- Replacement of animal models for drug screening: mice models are the predominant system used to screen novel compounds against liver damage. The Pharmaceutical industry will greatly benefit from the development of a multicellular in vitro model, which will pave the way to high throughput drug screening that are not yet feasible with the available in vivo or in vitro models.