Development of stem cell-derived hepatocyte models

Stem cell derived models of hepatocytes have potential in toxicity testing and regenerative medicine. Despite global efforts drug-induced liver-injury (DILI) remains a major concern for the pharmaceutical and healthcare industries and is a considerable cause of morbidity and mortality. Therefore, efforts are being made to reduce its incidence through toxicity screening to prevent the release of unsafe compounds to market. Current models of DILI are flawed, either lacking in terms of physiological relevance in the case of hepatic cell lines or in terms of reproducibility and availability in the case of human primary hepatocytes (hPH). Stem cell-derived hepatocyte models offer a theoretically unlimited source of potentially high-quality hepatocytes for toxicity testing. However, currently they are hindered by lack of maturity compared with hPH as in the case of pluripotent stem cell-derived hepatocyte-like cells (PSC-HLCs), or are incompletely phenotyped, in the case of human liver organoids. Hypoxic conditions improve the stemness of PSCs, in addition, during embryogenesis cells are exposed to hypoxic conditions. Furthermore, there is a range of oxygen concentration across the liver lobule which is implicated in their phenotype. We hypothesised that iPSC-HLCs differentiated under hypoxic conditions (5% O2) would adopt a perivenous phenotype with higher CYP expression. Two differentiations protocols, one growth factor-based, the other small molecule-based were also compared. These results showed little difference between oxygen conditions but suggested that cells differentiated under traditional normoxic conditions experience some degree of hypoxia. Additionally, we found little difference between differentiation protocols, suggesting the cheaper, and more reproducible small molecule-based protocol should be used in future studies. Human liver organoids derived from EpCAM-positive cells were established and their phenotype was examined using iTRAQ. These results showed that organoids offer good recapitulation of the phenotype of hPH particularly regarding transporter phenotype. The current demand for donor organs outstrips supply. Consequently, there have been efforts in recent years to find alternate sources of cells and tissue for transplant. PSCs offer a potentially unlimited source of cells for this purpose. However, the risks associated with the use of stem cells for transplant are incompletely understood. There are concerns regarding the tumorigenic potential of stem cells used for transplant, based not only on the inherent ability of pluripotent cells to form teratomas in vivo but also regarding the more insidious genetic changes introduced during isolation and culture. The effects of the 20q11.21 amplicon, a common aberration occurring in approximately 20% of PSC-lines after extended culture and providing a selective advantage in vitro were examined when differentiated into HLCs and injected intrasplenically into SCID mice. We generated luminescent PSCs with known 20q11.21 status, tracked their engraftment over approximately 4 months. Our results demonstrated worsened liver histology and increased luminescence in animals injected with cells containing the amplicon. These studies revealed little no advantage to the use of hypoxic conditions for the production of HLCs but suggest potential for the future use of hyperoxia. The investigation into the proteomic phenotype of organoids revealed a good approximation of many liver specific genes including CYPs, phase II enzymes and transporters but were hindered by the inability to successfully cryopreserve the organoids. Despite issues with engraftment, the presence of 20q11.21 appears to worsen liver histology, future studies should screen against its use. Repeat studies should consider the use of alternate animal models to better facilitate cell engraftmen