Hepatitis B virus (HBV) infection, which increases the risk of cirrhosis and hepatocellular carcinoma and requires lifelong treatment, has become a major global health problem. However, host factors essential to the HBV life cycle are still unclear, and the development of new drugs is needed. Cells derived from the human hepatoma cell line HepG2 and engineered to overexpress sodium taurocholate cotransporting polypeptide (NTCP: a receptor for HBV), termed HepG2/NTCP cells, are widely used as the cell-based HBV infection and replication systems for HBV research. We recently found that human hepatoma cell line Li23-derived cells overexpressing NTCP (A8 cells subcloned from Li23 cells), whose gene expression profile was distinct from that of HepG2/NTCP cells, were also sensitive to HBV infection. However, the HBV susceptibility of A8 cells was around 1/100 that of HepG2/NTCP cells. Since we considered that plural cell assay systems will be needed for the objective evaluation of anti-HBV reagents, as we previously demonstrated in hepatitis C virus research, we here attempted to develop a new Li23 cell-derived assay system equivalent to that using HepG2/NTCP cells. By repeated subcloning of A8 cells, we successfully established a new cell line (A8.15.78.10) exhibiting high HBV susceptibility equal to that of HepG2/NTCP cells. Characterization of A8.15.78.10 cells revealed that the increase of HBV susceptibility was correlated with increases in the protein and glycosylation levels of NTCP, and with decreased expression of STING, a factor contributing to innate immunity. Finally, we performed a comparative evaluation of HBV entry inhibitors (cyclosporin A and rosiglitazone) by an HBV/secNL reporter assay using A8.15.78.10 cells or HepG2/NTCP cells. The results confirmed that cyclosporin A exhibited anti-HBV activity in both cell lines, as previously reported. However, we found that rosiglitazone did not show the anti-HBV activity in A8.15.78.10 cells, although it worked in HepG2/NTCP cells as previously reported. This suggested that the difference in anti-HBV activity between cyclosporin A and rosiglitazone was due to the different types of cells used for the assay. In conclusion, plural assay systems using different types of cells are required for the objective and impartial evaluation of anti-HBV reagents. (C) 2019 Elsevier Inc. All rights reserved.