The pharmaceutical industry uses a number of different in vivo and in vitro approaches to assess the potential for new drugs to cause liver injury (hepatotoxicity). However, with hepatotoxicity being the second most common cause of drug failure, there are clear limitations in the ability of these models to predict human toxicity. The development of more predictive human safety models incorporating the influence of drug metabolism is urgently required to facilitate development of safer and more cost effective medicines.
The company FibroMed has developed a novel pluripotent stem cell-derived hepatocyte system to predict human drug toxicity and drug induced liver injury. Working with the pharmaceutical industry they have validated their model using a panel of compounds that displayed human hepatotoxicity and operate through known cellular mechanisms. Based on current data, their system is predictive of human hepatocyte toxicity; notably identifying toxicities with IC50 values comparable to freshly isolated human hepatocytes. This approach shows the potential to reduce the number of animals used in safety testing and the current levels of drug attrition due to hepatotoxicity.
The team is keen to evolve the platform to culture the cells in an environment that better mimics liver physiology in vivo. This includes the provision of a stable environment with the constant delivery of nutrients, removal of waste products and presentation of key liver factors using fluid flow across their cells. Evidence exists supporting the importance of fluid shear stress in other cell types, but it has been lacking in the case of hepatocytes.
Through CRACK IT Solutions, FibroMed identified a partner, Kirkstall Ltd, able to provide culture chambers that permit exposure to fluid shear stress, in two and three dimensions. Supported by CRACK IT Solutions funding, both teams embarked on a project to:
- Optimise the seeding of stem cell-derived hepatocytes in fluid shear stress chambers,
- Optimise the level of fluid shear stress that stem cell-derived hepatocytes were exposed to;
- Assess the response of the stem cell-derived hepatocytes to a known hepatotoxin following exposure to fluid shear stress.
To assess the effect of flow on metabolic activity, 2D- and 3D-cultured stem cell-derived hepatocytes were transferred into serially-connected chambers of the Kirkstall Ltd. Quasi-Vivo® system and exposed to a range of fluid shear stress of 2.9 to 6.5×10-5 dynes/cm2 using a peristaltic pump. Metabolic cytochrome P450 (CYP) activity of two isoforms of Cyp1A2 and Cyp2D6 were measured to evaluate the effect of flow on stem cell-derived hepatocyte metabolic activity, and the response to a known hepatotoxin supplied by the pharmaceutical company Bristol-Myers Squibb.
These studies highlighted that stem cell-derived hepatocytes are able to tolerate fluid shear stress and this resulted in improved Cyp2D6 metabolic activity. Cyp1A2 metabolic activity was differentially regulated under these conditions and this will be the subject of future investigation. Importantly, prototype model sensitivity is improved in response to flow, resulting in improved pharmaceutical grade compound metabolism.
The teams have successfully developed a scalable and predictive 2D and 3D human hepatocyte prototype model to study the influence of fluid shear stress on stem cell-derived hepatocyte function. The knowledge obtained from this project provides proof-of-concept that fluid shear stress improves hepatocyte phenotype and sensitivity to a known drug. Therefore the cell-based models generated in this programme of work will be more sensitive tools for use in human drug safety trials and may obviate the need for animal use in safety testing experiments.
They remain open to exploring a variety of options for partnerships from companies interested in screening their internal compounds in their model.
The full output of this study has recently been published Archives of Toxicology.
Rashisi H et al. (2016). Fluid shear stress modulation of hepatocyte-like cell function, Arch Toxicol 90: 1757. doi:10.1007/s00204-016-1689-8.