We propose to apply four complementary technologies in a Quantitative Systems Pharmacology approach to create a human experimental model of non-alcoholic fatty liver disease (NAFLD), the most rapidly growing disease, and to use the model to test novel therapeutic strategies:1) Implement a vascularized, liver acinus microphysiological system (vLAMPS) constructed with human patient-derived, liver cells, as an experimental model to recapitulate early NAFLD phenotypes and as a platform to experimentally test novel therapeutics; 2) Building on our experience in computational and systems biology, we will use RNAseq data from normal and NAFLD patients to infer pathways of disease progression, to identify the potential molecular protein targets that are in the inferred pathways, and to use our latent factor modeling approach and 3D similarity models to identify drugs that statistically interact with the targets in these pathways; 3) We will employ our highly efficient processes for generating mature iPSC-derived hepatocytes combined with gene editing to incorporate disease engineered iPSC hepatocytes (conditional gain/loss of function) into the vLAMPS to begin testing patient specific therapies; and 4) Apply phenotypic drug screening technologies. NAFLD encompasses a spectrum of liver damage ranging from simple steatosis (NAFL) to more serious non- alcoholic steatohepatitis (NASH), cirrhosis and hepatocellular carcinoma (HCC). Cirrhosis and HCC resulting from progressive damage to the liver have become the third most common causes of liver transplants. The disease pathogenesis of NAFLD is complex and confounded by the considerable inter-individual differences in disease susceptibility, progression and complications, suggesting the need for a patient specific approach. Studies have identified NAFLD associated gene signatures and single nucleotide polymorphisms (SNPs). In particular, the SIRT1 gene that is downregulated in NAFLD, has been identified as a key regulator of lipogenesis, gluconeogenesis, ER stress, fatty acid oxidation, urea cycle and the antioxidant response in hepatocytes. A SNP in the patatin-like phospho-lipase domain-containing 3 (PNPLA3) gene is strongly associated with hepatic steatosis, fibrosis, cirrhosis, and HCC. However, there continues to be major gaps in our understanding of the pathogenesis of NAFLD. For example, despite its strong association with NAFLD, the functional significance of the PNPLA3 variant is unknown. A major limitation in the elucidation of a mechanistic role of PNPLA3 in NAFLD has been the interspecies differences in its expression and tissue-specific distribution, suggesting the need for human cell models. This combination of the technologies and approaches is expected to lead to new strategies for development of repurposed and new therapeutics with the potential to slow or halt the progression of early NAFLD to the more advanced, life threatening stages.
We propose to further develop and to apply a vascularized, human, 3D microfluidic liver model (vLAMPS), comprised of patient-derived, primary hepatocytes (or engineered, iPSC-derived hepatocytes containing key SNPs and mutations) together with liver sinusoidal endothelial, stellate and Kupffer cells to recapitulate non-alcoholic fatty liver disease (NAFLD) phenotypes and genotypes and to test novel therapies. Computational biology is being used to define the pathways of disease progression, to identify targets within the pathways, to predict drugs that statistically could interact with the targets to potentially halt or reverse disease phenotypes, and then profile these drugs and combinations to test the predictions. The absence of approved drugs and excellent animal models suggests that novel experimental models containing human cells are needed to study the pathogenesis of NAFLD and to test novel therapies.