A microfabricated 3D tissue engineered "Liver on a Chip" : information content assays for in vitro drug metabolism studies

Abstract

(cont.) approaches to improving hepatocyte function in culture have been described, not all of the important functions--specifically the biotransformation functions of the liver--can as yet be replicated at desired in ivo levels, especially in culture formats amenable to routine use in drug development. The in vivo microenvironment of hepatocytes in the liver capillary bed includes signaling mechanisms mediated by cell-cell and cell-matrix interactions, soluble factors, and mechanical forces. This thesis focuses on the design, fabrication, modeling and characterization of a microfabricated bioreactor system that attempts to mimic the in vivo microenvironment by allowing for the three dimensional morphogenesis of liver tissue under continuous perfusion conditions. A key feature of the bioreactor that was designed is the distribution of cells into many tiny ([approximately]0.001 cm³) tissue units that are uniformly perfused with culture medium. The total mass of tissue in the system is readily adjusted for applications requiring only a few thousand cells to those requiring over a million cells by keeping the microenvironment the same and scaling the total number of tissue units in the reactor. Using a computational fluid dynamic model in ADINA® and a species conservation mass transfer model in FEMLAB®, the design of the bioreactor and the fluidic circuit was optimized to mimic physiological shear stress rates ...

Recent reports indicate that it takes nearly $800 million dollars and 10-15 years of development time to bring a drug to market. The pre-clinical stage of the drug development process includes a panel of screens with in vitro models followed by comprehensive studies in animals to make quantitative and qualitative predictions of the main pharmacodynamic, pharmacokinetic, and toxicological properties of the candidate drug. Nearly 90% of the lead candidates identified by current in vitro screens fail to become drugs. Among lead compounds that progress to Phase I clinical trials, more than 50% fail due to unforeseen human liver toxicity and bioavailability issues. Clearly, better methods are needed to predict human responses to drugs. The liver is the most important site of drug metabolism and a variety of ex vivo and in vitro model systems have therefore been developed to mimic key aspects of the in vivo biotransformation pathways of human liver-- a pre-requisite for a good, predictive pharmacologically relevant screen. Drug metabolism or biotransformation in the liver involves a set of Phase I (or p450 mediated) and Phase II enzyme reactions that affect the overall therapeutic and toxic profile of a drug. The liver is also a key site of drug toxicity following biotransformation, a response that is desirable but difficult to mimic in vitro. A major barrier to predictive liver metabolism and toxicology is the rapid (hours) loss of liver-specific functions in isolated hepatocytes when maintained under standard in itrom cell culture condition. This loss of function may be especially important in predicting toxicology, where the time scale for toxic response may greatly exceed the time scale for loss of hepatocyte function in culture. Although a wide variety of