Development of a Microfluidic Vascularized Liver Model for High Throughput Drug Metabolism and Toxicity Screening

Farhang Ghahremani, Morvarid 1, 2; Chebotarev, Oleg 1, 2; Simmons, Craig 1, 2, 3

1. Department of Mechanical & Industrial Engineering - University of Toronto; 2. Translational Biology & Engineering Program - Ted Rogers Centre for Heart Research; 3. Institute of Biomaterials & Biomedical Engineering - University of Toronto

Drug induced liver toxicity is the major problem in all phases of clinical trials during the drug development process resulting in an unacceptably high failure rates. The methods currently used for pre-clinical drug screening are limited to two dimensional cell cultures and animal testing. In addition to ethical considerations and high costs of these studies, current in vitro based approaches used for screening are inaccurate, in part because they do not represent multi-cellular interactions of native human liver. In particular, liver sinusoidal endothelial cells (LSECs) which modulate hepatocyte CYP expression, urea synthesis, and albumin secretion are absent from these models. Therefore there is an urgent need to develop a predictive, human-based in vitro drug screening and toxicity platform that can be applied early in the development process and overcome limitations of current systems in terms of cost-effectiveness, relevance, and predictability.

To address this need we have developed a high throughput microfluidic platform to model three-dimensional (3D) liver tissues. The device is a 96-multiwell plate where each well is a dual-layered microchannel composed of hepatocytes embedded in a 3D hydrogel below an LSEC layer that is perfused at physiological fluid flow mimicking in vivo structure.

We have currently optimized culture conditions to co-culture human hepatocytes with LSECs and have shown microscopically that hepatocytes cultured in the device for one week perform better than 2D monolayer and equal to sandwich culture models in terms of viability and albumin expression. In addition we assessed viability and function of primary human LSECs grown on the porous polyester membrane, in co-culture with hepatocytes, to primary LSECs grown on TCPS for five days. We were able to demonstrate that LSECs grown on the membrane expressed both standard and liver specific endothelial markers and uptake of acetylated low density lipoprotein uptake comparable to LSECs grown on TCPS.

The next phase of our studies is to establish baseline CYP enzyme expression levels before and after induction with specific drugs. Furthermore, we will optimize our model for metabolic clearance, drug-drug interaction, and hepatotoxicity assays.