Construction of a culture protocol for functional bile canaliculi formation to apply human iPS cell-derived hepatocytes for cholestasis evaluation

Cholestatic toxicity causes the failure of pharmaceutical agents during drug development and, thus, should be identified at an early stage of drug discovery and development. The formation of functional bile canaliculi in human hepatocytes is required for in vitro cholestasis toxicity tests conducted during the early stage of drug development. In this study, we investigated the culture conditions required for the formation of bile canaliculi using human-induced pluripotent stem cell-derived hepatocytes (hiPSC-Heps). When hiPSC-Heps were sandwich-cultured under the condition we established, extended bile canaliculi were formed on the whole well surfaces. Biliary efflux transporters were localized in the formed bile canaliculi structures which had junctional complexes. After the model substrates of the biliary efflux transporters were taken up into cells, their subsequent excretion into the bile canaliculi was observed and was found to be impeded by each inhibitor of the biliary efflux transporter. These findings suggest that bile canaliculi have transporter-specific bile excretion abilities. We will continue to study the application of this culture protocol to cell-based cholestasis assay system. As a result, the culture protocol could lead to a highly predictable, robust cell-based cholestasis assay system because it forms functional bile canaliculi reproducibly and efficiently.

Culture schedule for the extended bile canaliculi and cell morphology before or after sandwich culture in maintenance medium (one on the left) or LTM medium (three on the right). Cell morphology was observed using a phase contrast microscope. The red circles show areas where bile canaliculi were formed before the sandwich culture. Observation of the bile canaliculi using FDA. hiPSC-Heps were cultured in a LTM medium for 28 days and then, they were sandwich-cultured in an LTM medium for 7 days. The biliary efflux assay was performed using FDA. Phase contrast microscope image (left) and fluorescence image (

Relative expression
Expression of MRP2 when hiPSC-Heps were short-term cultured in a TLM medium or maintenance medium. hiPSC-Heps were cultured in a TLM medium or maintenance medium for 4 days after a 5-day culture according to the manufacturer's protocol. The bars show the relative expression levels of MRP2. Pooled RNA from human liver was used for the standard curve, and the expression level was set as one. The relative expression level was calculated using the equation of the line for the standard curve. Data are presented as means ± S.D. (n = 3).
Observation of the bile canaliculi using FDA or tauro-nor-THCA-24-DBD. hiPSC-Heps were cultured in an LTM medium for 28 days and sandwich-cultured in a maintenance medium for another 7 days. The biliary efflux assay was performed using FDA or tauro-nor-THCA-24-DBD. Fluorescence images show that fluorescein or tauro-nor-THCA-24-DBD accumulated in bile canaliculi. Culture schedule for the extended bile canaliculi formation using LTM medium and maintenance medium from CDI.

Tauro-nor-THCA-24-DBD
Expression of the genes related to bile acid efflux during culture in a long-term maintenance medium. The bars show the relative expression levels of FXR, CYP7A1, MRP2, and BSEP. Pooled human liver RNA was used for the standard curve, and the expression level was set as one. The relative expression level was calculated using the equation of the line for the standard curve. Data are presented as means ± S.D. (n = 3).

CYP7A1
Gene expressions of OATP1B1, and NTCP when the bile canaliculi were formed. Human-induced pluripotent stem cellderived hepatocytes were cultured in a long-term maintenance medium for 28 days and then sandwich-cultured in a maintenance medium for 9 days. The expression levels of OATP1B1, and NTCP were measured using quantitative polymerase chain reaction before and after sandwich culture. The bars show the relative expression levels OATP1B1, or NTCP. Pooled RNA from human liver was used for the standard curve, and the expression level was set as one. The relative expression level was calculated using the equation of the line for the standard curve. Data are presented as means ± S.D. (n = 3). Effect of a biliary efflux transporter inhibitor on the excretion of the model substrate into bile canaliculi. Human-induced pluripotent stem cell-derived hepatocytes were cultured in a long-term maintenance medium for 28 days and then sandwich-cultured in a maintenance medium for 9 days. After that, the effect of a biliary efflux transporter inhibitor on the excretion of the model substrate into bile canaliculi was examined. Fluorescence images were taken at the same exposure time with and without inhibitor. The intensity of ten intracellular regions in the fluorescence images was measured using ImageJ (Abramoff, M., Magalhaes, P. & Ram, S. Biophotonics Int. 11, 36-42 (2004)), and the average value was calculated. A) Fluorescein intensity in the intracellular area, B) tauro-nor-THCA-24-DBD intensity in the intracellular area. Comparison of bile canaliculi formation between batches. hiPSC-Heps were cultured in a long-term maintenance medium for 28 days and then sandwich-cultured in a maintenance medium for 9 days. Then, the biliary efflux assay was performed in each batch culture using FDA or CDFDA. Phase contrast microscope image (top) and fluorescence image (bottom). Fluorescence images show fluorescein or CDF accumulated in bile canaliculi. Cytochrome P450 activity when the bile canaliculi were formed. Human-induced pluripotent stem cell-derived hepatocytes were cultured in a long-term maintenance medium for 28 days and then sandwich-cultured in a maintenance medium for 9 days. Then, the metabolism test was performed. The red line shows the maximum value of activity in 8 lots of human cryopreserved hepatocytes. The blue line shows the minimum value of activity in 8 lots of human cryopreserved hepatocytes. Data are presented as means ± S.D. (n = 3).  Scale-down culture for bile canaliculi formation. hiPSC-Heps were cultured in a long-term maintenance medium for 28 days and then sandwich-cultured in a maintenance medium for 10 days on a 96well plate. Then, the biliary efflux assay was performed in three wells. Fluorescence images show CDF accumulated in bile canaliculi. Comparison of biliary efflux transporter expression in sandwich cultures using different mediums. hiPSC-Heps were cultured in a long-term maintenance medium for 28 days and then sandwichcultured in a long-term maintenance medium or maintenance medium for 9 days. The bars show the relative expression levels of MRP2 or BSEP after sandwich culture (n=1). Pooled RNA from human liver was used for the standard curve, and the expression level was set as one. The relative expression level was calculated using the equation of the line for the standard curve. 1.E-03 1.E-02 1.E-01 1.E+00 0 9 CYP2E1 10 0 10 -1 10 -2 10 -3

Culture days in LTM medium (days) relative expression
Gene expression of CYP2E1 when the bile canaliculi were formed. Human-induced pluripotent stem cell-derived hepatocytes were cultured in a long-term maintenance medium for 28 days and then, they were sandwich-cultured in a maintenance medium for 9 days. The expression levels of major CYPs were measured using quantitative polymerase chain reaction before and after sandwich culture. The bar shows the relative expression levels of CYPs. Pooled RNA from human liver was used for the standard curve, and the expression level was set as one. The relative expression level was calculated using the equation of the line for the standard curve. The red line shows the maximum value of expression in 4 lots of human cryopreserved hepatocytes. The blue line shows the minimum value of expression in 4 lots of human cryopreserved hepatocytes. Data are presented as means ± S.D. (n = 3).