Investigators are reporting on the production of mice with ectopically implanted human liver tissue that remains functional for weeks, synthesizes human proteins, and recapitulates human drug metabolism, drug-drug interactions, and drug-induced liver injury. The researchers claim that in contrast with previous models in which human liver cells are transplanted into immunodeficient mice with liver injury, human ectopic artificial livers (HEALS) can be established rapidly and efficiently in immunocompetent mice with normal liver function.
Reporting in PNAS, Sangeeta N. Bhatia, Ph.D., of MIT, and colleagues describe their experiments in a paper is titled “Humanized mice with ectopic artificial liver tissues.” HEALS scaffolds were generated by coculturing primary human hepatocytes (HEP) with stabilizing mouse fibroblasts (FIB). The cells were then encapsulated with liver endothelial cells in a polymeric scaffold.
The resulting HEALS were about 20 mm in diameter and 250 μm thick and comprised about 0.5 × 106 human hepatocytes. Importantly, the FIBs provided stabilization of the HEPs after isolation, while the LECs improved encapsulated cell function, probably through the secretion of short-range or rapidly turned over soluble factors.
Cultured 3-D HEALS were shown to express the genes for major human drug metabolizing enzymes including phase I detoxification enzymes, phase II conjugating enzymes, phase III transporters, several key transcription factors, and albumin. Importantly, CYP3A4, 1A2, 2D6, 2E1, and the 2C isoforms, which collectively metabolize over 90% of clinical drugs, were expressed by the HEALS. Treating the cultured HEALS with drugs that are known to activate specific metabolizing enzymes resulted in the production of expected enzymes including relevant drug-drug interaction responses.
Reporter-carrying 3-D HEP/FIB HEALS were then transplanted subcutaneously or into an intraperitoneal site in athymic nude mice. Subsequent imaging and biochemical analyses showed that the implants were supported for weeks in vivo (and for up to three months in one cohort of mice), and produced secreted human proteins. Engraftment of HEALS in vivo was successful in over 90% of transplanted mice, with blood vessels growing into and supplying the implants.
Encouragingly, the results suggested that the techniques used to stabilize human hepatocytes prior to implantation protected the cells from death due to anoikis, loss of cellular signaling, and/or compromised oxygen transport during engraftment. It also decreased dependence on hepatotrophic factors from the portal vein.
The authors point out that this enabled them to generate HEAL-humanized mice using cryopreserved hepatocytes from different primary donors in under a week. In contrast, generating humanized mice using cell transplantation methods takes 2–6 months and results in unpredictable and highly variable cell repopulation yields, they claim.
To assess whether mice with HEALS could be used to identify major drug metabolites in vivo, the team then challenged wild-type and humanized athymic nude or immunocompetent C57/BL6 mice with drugs known to be metabolized differently by mice and humans. Whereas the normal mice failed to identify human metabolites of the drugs, the HEAL-humanized mice correctly revealed multiple breakdown products. Similarly, HEAL-humanized mice could be used to identify the effects of drug-drug interactions, including those that lead to liver toxicity.
“We envision that integration of this model into drug development pipelines has the potential to reduce clinical trial attrition rates and accelerate the process by which new therapeutics reach patients,” the authors conclude. They also claim the HEAL mouse model will have applications beyond drug testing, in fields including human liver disease and infection research or humanized immunity.