Oxygen control in bioreactor drives high yield production of functional hiPSC-like hepatocytes for advanced liver disease modelling

Abstract

Hepatocytes-like cells (HLC) derived from human induced pluripotent stem cells show great promise for cell-based liver therapies and disease modelling. However, their application is currently hindered by the low production yields of existing protocols. We aim to develop a bioprocess able to generate high numbers of HLC. We used stirred-tank bioreactors with a rational control of dissolved oxygen concentration (DO) for the optimization of HLC production as 3D aggregates. We evaluated the impact of controlling DO at physiological levels (4%O2) during hepatic progenitors’ stage on cell proliferation and differentiation efficiency. Whole transcriptome analysis and biochemical assays were performed to provide a detailed characterization of HLC quality attributes. When DO was controlled at 4%O2 during the hepatic progenitors’ stage, cells presented an upregulation of genes associated with hypoxia-inducible factor pathway and a downregulation of oxidative stress genes. This condition promoted higher HLC production (maximum cell concentration: 2 × 106 cell/mL) and improved differentiation efficiencies (80% Albumin-positive cells) when compared to the bioreactor operated under atmospheric oxygen levels (21%O2, 0.6 × 106 cell/mL, 43% Albumin positive cells). These HLC exhibited functional characteristics of hepatocytes: capacity to metabolize drugs, ability to synthesize hepatic metabolites, and inducible cytochrome P450 activity. Bioprocess robustness was confirmed with HLC derived from different donors, including a primary hyperoxaluria type 1 (PH1) patient. The generated PH1.HLC showed metabolic features of PH1 disease with higher secretion of oxalate compared with HLC generated from healthy individuals. This work reports a reproducible bioprocess, that shows the importance of controlling DO at physiological levels to increase HLC production, and the HLC capability to display PH1 disease features.