Enrichment of high-functioning human iPS cell-derived hepatocyte-like cells for pharmaceutical research

Abstract

Human iPS cell-derived hepatocyte-like cells are expected to be utilized in pharmaceutical research. However, the purity of high-functioning hepatocyte-like cells is not high enough. In particular, the purity of cytochrome P450 3A4 (CYP3A4), which is a representative hepatic drug-metabolizing enzyme, positive cells is still quite low (approximately 20%). To address this problem, we established the CYP3A4-NeoR-EGFP transgenic reporter human iPS cell line (CYP3A4-NeoR-EGFP iPS cells) by using genome editing technology. The CYP3A4-NeoR-EGFP iPS cells were differentiated into hepatocyte-like cells, and then the hepatocyte-like cells were treated with neomycin to concentrate the hepatocyte-like cells which strongly express CYP3A4. After the neomycin treatment, the percentage of CYP3A4-positive cells was higher than 80%. The gene expression levels of various drug-metabolizing enzymes, transporters, and hepatic transcription factors were significantly enhanced by neomycin treatment. In addition, the CYP1A2, 2C19, 2D6, and 3A4 activities and biliary excretion capacities were significantly increased by neomycin treatment. We also confirmed that the detection sensitivity of drug-inducing hepatotoxicity was enhanced by neomycin treatment. We succeeded in obtaining human iPS cell-derived hepatocyte-like cells that highly express CYP3A4 at high purity. We believe that our high-purity and high-functioning hepatocyte-like cells could be used to evaluate the risk of drug candidates.

Introduction

Human induced pluripotent stem (iPS) cell-derived hepatocyte-like cells are expected to be utilized in pharmaceutical development, such as drug screening and drug metabolism testing. Many research groups, including ours, have been engaged in developing an efficient hepatocyte differentiation protocol from human iPS cells [[1], [2], [3], [4], [5]]. Hepatocyte-like cells have the capacity to secrete albumin (ALB), urea, and bile acid and also have drug metabolism capacity. However, several issues must be overcome before hepatocyte-like cells can be used in pharmaceutical applications. It has been reported that the hepatocyte-like cells are similar to fetal hepatocytes rather than adult hepatocytes [6,7]. In addition, although almost all of the hepatocyte-like cells are positive for early hepatic markers (such as ALB), some of the hepatocyte-like cells are negative for late hepatic markers (such as cytochrome P450 (CYP)) [8]. These facts suggest that it is still difficult to generate mature hepatocyte-like cells at high purity.

To obtain mature hepatocyte-like cells at high purity, some groups have sorted mature hepatocyte-like cells by using hepatocyte-specific cell surface markers [8,9]. Basma et al. have reported that mature hepatocyte-like cells could be concentrated by sorting asialoglycoprotein receptor 1 (ASGR1)-positive cells [9]. Mallanna et al. have also reported that mature hepatocyte-like cells could be concentrated by sorting sodium taurocholate cotransporting polypeptide (NTCP)-positive cells [8]. However, although cell suspensions of mature hepatocyte-like cells can be obtained by cell sorting, it is difficult to replace the sorted mature hepatocyte-like cells, because hepatocyte-like cells easily lose their hepatic functions during the cell reattachment process. In the drug screening and CYP induction experiments, attached hepatocytes, not suspended hepatocytes, are highly required. Thus, the cell sorting technology might not be suitable for preparing attached mature hepatocyte-like cells at high purity. Therefore, we decided to purify mature hepatocyte-like cells by drug-selection.

In this study, we attempted to establish a CYP3A4-neomycin resistance gene (NeoR)-enhanced green fluorescent protein (EGFP) transgenic reporter human iPS cell line (CYP3A4-NeoR-EGFP iPS cells) by using clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated protein-9 (Cas9) technology. Although it is known that the genome editing efficiency of human ES/iPS cell clones is quite low at the heterochromatin locus [10,11], we have recently found that RAD51 recombinase (RAD51) overexpression and valproic acid treatment enhanced genome targeting efficiency in human ES/iPS cells regardless of the transcriptional activity of the targeted locus [12]. Therefore, we targeted the CYP3A4 locus by using our genome editing technology. After the hepatocyte differentiation, we tried to concentrate the hepatocyte-like cells that strongly expressed CYP3A4 by treating the CYP3A4-NeoR-EGFP iPS cell-derived hepatocyte-like cells with neomycin. In addition, we examined the pharmaceutical value of CYP3A4-positive high-functioning hepatocyte-like cells.