Forskolin induces FXR expression and enhances maturation of iPSC-derived hepatocyte-like cells

Abstract

The generation of iPSC-derived hepatocyte-like cells (HLCs) is a powerful tool for studying liver diseases, their therapy as well as drug development. iPSC-derived disease models benefit from their diverse origin of patients, enabling the study of disease-associated mutations and, when considering more than one iPSC line to reflect a more diverse genetic background compared to immortalized cell lines. Unfortunately, the use of iPSC-derived HLCs is limited due to their lack of maturity and a rather fetal phenotype. Commercial kits and complicated 3D-protocols are cost- and time-intensive and hardly useable for smaller working groups. In this study, we optimized our previously published protocol by fine-tuning the initial cell number, exchanging antibiotics and basal medium composition and introducing the small molecule forskolin during the HLC maturation step. We thereby contribute to the liver research field by providing a simple, cost- and time-effective 2D differentiation protocol. We generate functional HLCs with significantly increased HLC hallmark gene (ALB, HNF4α, and CYP3A4) and protein (ALB) expression, as well as significantly elevated inducible CYP3A4 activity.

Keywords: Forskolin; cytochrome P450 activity; hepatocyte-like cells (HLCs); in vitro Differentiation; induced pluripotent stem cells (iPSCs).

FIGURE 1

Generation of iPSC-derived hepatocyte-like cells (HLCs). (A) Time line for differentiation of iPSCs over 21 days, indicating Definitive Endoderm induction by CHIR99021 on d1, and Activin A treatment on d1-d3. Hepatic Endoderm was induced by adding DMSO on d4-d8. Maturation of HLCs was induced by insulin, hepatocyte growth factor (HGF), dexamethasone (Dex), oncostatin M (OSM) and forskolin (Forsko) treatment on d8-d21. (B) Comparison of protocol 1 and 2, changes in the media composition are marked in red. (C) Morphology of HLCs on day 20/21 generated with protocol 1 or 2.

FIGURE 2

Characterization of HLCs generated with protocol 1 or protocol 2. (A) Representative immunofluorescence images of ECAD (red), ALB (green), HNF4α (purple) in HLCs differentiated with protocol 1 or two respectively. DNA was stained with Hoechst 33258. (B) Representative immunofluorescence images of FXR (green). DNA was stained with Hoechst 33258. (C) Western blots of AFP and ALB for 3 biological replicates of HLCs generated with protocol 1 or two respectively. (D) Fold change of protein expression (mean±SD) of three biological replicates of HLCs generated with protocol 1 or 2, normalized to housekeeping protein β-actin, relative to protocol 1. Two-tailed Student’s t-test was performed to calculate significances (*p < 0.05, **p < 0.01). Relative mRNA expression of (E) OCT4 (F) AFP, ALB, CYP2D6, CYP3A4 and FXR normalized to the housekeeping gene RPL0 of HLCs generated with protocol 1 or two relative to iPSCs, respectively. Bar plots show mean± SE from three biological replicates. Two-tailed Student’s t-test was performed to calculate significances (*p < 0.05, **p < 0.01, ***p < 0.001). (G) Agarose gel electrophoresis showing bands of corresponding sizes indicating amplicons of housekeeping gene RPL0 and FXR after RT-qPCR for iPSCs, and HLCs generated with protocol 1 or 2, respectively. (H) CYP3A4 activity determined by metabolization of luciferin-IPA as relative light units (R.L.U) normalized to total protein in HLCs generated by protocol 1 or 2 with and without rifampicin (Rifa) treatment for 24 h. Bar plots show mean±SD and significances were calculated by using two-tailed Student’s t-test (*p < 0.05, ***p < 0.001). (I) Secreted urea into the supernatant within 24 h. (J) Uptake and release after 6 h of indocyanine green (ICG) of HLCs generated with protocol 1 or 2. (K) Glycogen storage shown via Periodic Acid–Schiff (PAS) reaction on fixated HLCs generated with protocol 1 or 2 respectively.