COMMUNICATION TO THE EDITOR An Artificial Liver Sinusoid With a Microfluidic Endothelial-Like Barrier for Primary Hepatocyte Culture Philip J. Lee, 1 Paul J. Hung, 2 Luke P. Lee 1 1 Department of Bioengineering, Biomolecular Nanotechnology Center, Berkeley Sensor and Actuator Center, University of California, Berkeley 485 Evans Hall, Berkeley, California 94720-1762; telephone: 510-642-5855; fax: 510-642-5835; e-mail: lplee@berkeley.edu 2 CellASIC Corporation, San Leandro, California Received 27 November 2006; accepted 22 January 2007 Published online 7 February 2007 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/bit.21360 ABSTRACT: Primary hepatocytes represent a physiologi- cally relevant model for drug toxicity screening. Here, we created a biologically inspired artificial liver sinusoid with a microfluidic endothelial-like barrier having mass transport properties similar to the liver acinus. This unit consisted of a cord of hepatocytes (50 30 500 mm) fed by diffusion of nutrients across the microfluidic endothelial-like barrier from a convective transport vessel (10 nL/min). This con- figuration sustained rat and human hepatocytes for 7 days without an extracellular matrix (ECM) coating. Experiments with the metabolism mediated liver toxicant diclofenac showed no hepatotoxicity after 4 h and an IC 50 of 334 41 mM after 24 h. Biotechnol. Bioeng. 2007;97: 1340–1346. ß 2007 Wiley Periodicals, Inc. KEYWORDS: microfluidic; artificial liver; biomimetic; hepatocyte; bioreactor Introduction Primary cell culture technology has the potential to greatly improve biomedical research by providing a more relevant in vitro model of clinical behavior. With standard culture methods, primary cells rapidly lose tissue specific functions once they are removed from the living organism (Boess et al., 2003; Rodriguez-Antona et al., 2002). In order to overcome this limitation, engineering methods are being developed to better approximate the in vivo culture condition, specifically targeting the cellular microenviron- ment. Much of this research focuses on the use of 3D ECM coatings, due to the available technology and scientific understanding of the biology of cell-ECM interactions (Cukierman et al., 2001; Griffith and Swartz, 2006). Recently, robotic DNA array spotters have been adapted to pattern ECM and biocompatible polymers to screen for their effects on hepatocyte behavior (Flaim et al., 2005; Revzin et al., 2004) and cellular differentiation (Anderson et al., 2004; Flaim et al., 2005). Primary hepatocytes represent a physiologically relevant model for drug toxicity screening, but improved methods are desired to preserve metabolic function in vitro while minimizing cell con- sumption (Battle and Stacey, 2001; LeCluyse et al., 1996). It is well established that ECM coatings such as collagen I are necessary to maintain primary hepatocyte viability in culture, but this has been shown to cause down-regulation of liver specific activity (Ben-Ze’ev et al., 1988). The microscale topology of cell and ECM contacts is known to alter hepatocyte phenotype (Berthiaume et al., 1996; Bhatia et al., 1999), indicating that methods for controlling cellular organization are important. There is a growing body of evidence in cell biology that microscale architecture can be a strong regulator of tissue specific activity (Bissell et al., 2003; Stevens and George, 2005). However, the difficulty of experimentally controlling the micro-architecture of pri- mary cells in vitro prevents more detailed study of this phenomenon. The advance of microfluidic devices for cell culture applications provides a promising route to address this biological problem. Microfluidic environments allow the control of nanoliter fluid volumes and flows that are unavailable with other methods (Walker et al., 2004). The impact of controlling the mass transport microenvironment on cultured hepatocytes was demonstrated in pioneering work using a novel miniature bioreactor (Powers et al., Correspondence to: L.P. Lee 1340 Biotechnology and Bioengineering, Vol. 97, No. 5, August 1, 2007 ß 2007 Wiley Periodicals, Inc.