Identification of extracellular miRNAs as biomarkers of liver fibrosis in a human 3D-culture system

Liver fibrosis is a common consequence of chronic liver injury caused by chemicals, dis-eases or diet. Currently, effective methods for early diagnosis and therapy are lacking. En-hancing our understanding of liver fibrosis mechanisms and identification of early non-invasive biomarkers is crucial to combat the clinical burden the disease causes. Using human 3D cell models, we aimed to investigate fibrosis and identify extracellular miRNAs as poten-tial translational biomarkers of fibrosis. This thesis is divided into three projects, which assess miRNAs as biomarkers using 3D human liver microtissues (MT) (1), the MT response to several compounds i.e. bile salts and environmental pollutants (2) and increasing MT complexity through the addition of a fourth cell type i.e. endothelial cells (3). The research goal of project 1 focussed on assessing the suitability of miRNAs as biomarkers of liver toxicity in vitro, with a specific focus on early detection of hepatic fibrosis and potential extracellular miRNAs specific to fibrosis. Moreover, the functional involvement of selected miRNAs in the HSC activation was also investigated. Project 2 assessed the effects of several compounds, including bile salts and environmental toxins (dioxin and dioxin-like compounds, on the 3D human liver MTs to investigate the po-tential adverse outcomes they may cause. Finally, project 3 aimed to increase the complexity of the multicellular MTs through the addition of an endothelial cell type to further improve the models physiological relevance. We demonstrated that 3D-HepaRG and 3D multicellular MTs are useful tools for investi-gating miRNA markers of hepatotoxicity and fibrosis, respectively. We show that methotrex-ate (MTX, pro-fibrotic) and acetaminophen (APAP, acute injury) elicited compound-specific responses in the MTs. Furthermore, were able to identify four extracellular miRNAs (miR-199a-5p, miR-214-3p, miR-99b-5p and miR-125a-5p) that could contribute towards a non-invasive method to detect liver fibrosis. Three of the miRNAs show direct links to HSC acti-vation. We also identified that the MTs are able to display the expected phenotypic response through exposure to several compounds. Taken together, these results highlight the versatility of the MTs as an in vitro tool for understanding the biological response to different com-pounds. Finally, the incorporation of HUVECs into the MTs was successful and did not affect their ability to recapitulate the final sequence of events leading to fibrosis upon exposure to TGF-β1, MTX and thioacetamide: HSC activation and increased collagen expression. This suggests that the cell-line based model is a suitable and cost-effective alternative for investi-gating liver fibrosis in vitro. In conclusion, the compound-specific responses of the MTs to a range of compounds demonstrates the suitability and versatility of this in vitro model as a tool to assessing biologi-cal and toxicological responses. We were able to identify released miRNAs and determine their phenotypic relevance to liver fibrosis. These miRNAs could contribute towards a panel of potentially translational biomarkers of liver fibrosis. We also enhanced the MTs complexity through the addition of a fourth cell type (endothelial cells) to improve the physiological rele-vance of the in vitro model