ARTICLE IN PRESS Cytotherapy, 2017; ■■: ■■–■■

Three-dimensional hydrogel culture conditions promote the differentiation of human induced pluripotent stem cells into hepatocytes

YING LUO1, CHENG LOU2, SUI ZHANG3, ZHENGYAN ZHU1, QIANZHE XING2, PENG WANG1, TONG LIU1, HUI LIU1, CHENGLONG LI1, WENXIA SHI1, ZHI DU1 & YINGTANG GAO1 1 Tianjin Key Laboratory of Artificial Cell,Tianjin Institute of Hepatobiliary Disease, Artificial Cell Engineering Technology Research Center of Public Health Ministry,Third Central Hospital of Tianjin,Tianjin, China, 2Department of Hepatobiliary Surgery,Third Central Hospital of Tianjin,Tianjin, China, and 3Department of Cardiology,The University of Texas MD Anderson Cancer Center, Houston,Texas, USA

Abstract Background aims. Human induced pluripotent stem cells (hiPSCs) are becoming increasingly popular in research endeavors due to their potential for clinical application; however, such application is challenging due to limitations such as inferior function and low induction efficiency. In this study, we aimed to establish a three-dimensional (3D) culture condition to mimic the environment in which hepatogenesis occurs in vivo to enhance the differentiation of hiPSCs for large-scale culture and high throughput BAL application. Methods. We used hydrogel to create hepatocyte-like cell (HLC) spheroids in a 3D culture condition and analyzed the cell-behavior and differentiation properties of hiPSCs in a synthetic nanofiber scaffold. Results. We found that treating cells withY-27632 promoted the formation of spheroids, and the cells aggregated more rapidly in a 3D culture condition. The ALB secretion, urea production and glycogen synthesis by HLCs in 3D were significantly higher than those grown in a 2-dimensional culture condition. In addition, the metabolic activities of the CYP450 enzymes were also higher in cells differentiated in the 3D culture condition. Conclusions. 3D hydrogel culture condition can promote differentiation of hiPSCs into hepatocytes. The 3D culture approach could be applied to the differentiation of hiPSCs into hepatocytes for bioartificial liver. Key Words: cell behavior, hepatocyte, human induced pluripotent stem cells, hydrogel

Introduction Liver failure is associated with high morbidity and mortality and is the seventh leading cause of death worldwide [1]. The bioartificial liver (BAL) system is a cell-based external artificial biological device that has synthetic functions and biotransformation activities that are similar to those of the liver [2]. However, the shortage of primary human hepatocytes, the xenotransplantation-related disadvantages of porcine cells and the limited metabolic function of immortalized hepatic cell lines preclude the widespread acceptance of the BAL system. Human induced pluripotent stem cells (hiPSCs) that are reprogrammed from a diverse range of cell types, such as hair follicle mesenchymal stem cells,

peripheral blood mononuclear cells and skin fibroblasts, possess unlimited self-renewal capability and the potential to differentiate into all three germ layers [3–5]. The successful generation of functional hepatocytes from hiPSCs could be a potential cell source for BAL devices. Using a step-wise method and cocktails of growth factors/cytokines to promote hiPSCs to hepatocytes is a current protocol. However, lower induction efficiency and significantly lacking liver function are the greatest obstacle to application.To enhance differentiation efficiency, human serum was used to replace fetal bovine serum (FBS) to mimic the environment of hepatogenesis in our previous study. We found human serum, particularly that acquired relatively soon after hepatectomy, can enhance the differentiation efficiency and functionality [6]. These

Correspondence: Yingtang Gao, PhD,Third Central Hospital of Tianjin, Jintang Road #83, Hedong District,Tianjin 300170, China. E-mail: [email protected] (Received 6 March 2017; accepted 9 August 2017) ISSN 1465-3249 Copyright © 2017 International Society for Cellular Therapy. Published by Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.jcyt.2017.08.008

ARTICLE IN PRESS 2

Y. Luo et al.

results suggested that conditions that mimic early organogenesis could enhance hepatocyte differentiation and the functionality of hiPSCs. Recently, several studies have demonstrated that three-dimensional (3D) culture conditions create a pragmatic microenvironment and mimic in vivo development, enhancing hepatocyte differentiation and the functionality of human embryonic stem cells (hESCs) and hiPSCs compared with 2-dimensional (2D) culture conditions [7–9].Various 3D culture conditions have been used, including the formation of selfaggregated spheroids on low-attachment surfaces [8,10], the encapsulation of hepatocytes in alginate and the embedding of cells in synthetic biomaterials [11]. Compared with the other two conditions, synthetic biomaterials provide 3D structures, extracellular matrix (ECM)-mimicking stiffness and an environment that facilitates the diffusion of nutrients and cellular growth factors [12–14]. Nanofiber hydrogel comprises a animal-free synthetic biomaterial that can mimic native ECM functions and thus support the adhesion and differentiation of hiPSCs. This gel is made by interweaving a self-assembling polypeptide, and the pore size is 50–200 nm.These peptides are completely synthetic, thus avoiding the potential pathogenicity of animal-derived materials. This 3D nanofiber hydrogel is critical for meeting future demands because of the advantages of biocompatibility and retrieval and the lower risk of immunogenic reaction. Thus, embedding cells in a 3D biomaterial construct including a nanofiber scaffold is expected to provide a microenvironment to improve the induction efficiency of hepatocyte differentiation for BAL applications [12,15,16]. Recent advances in 3D culture techniques have enabled the development of promising scaffolds for the differentiation of hiPSCs. However, to the best of our knowledge, few studies have applied this nanofiber hydrogel to hiPSC differentiation, and few have tried to determine how this completely synthetic hydrogel influences cell activities and differentiation. In this study, we focused on the cell-behavior and differentiation properties of hiPSCs in a synthetic nanofiber scaffold and designed the fabrication method in 3D culture condition. Methods Ethics approval and consent to participate This study was approved by the institutional review board of Tianjin Third Central Hospital,Tianjin, China (file no. 13010). All patients provided written informed consent. The methods were carried out in accordance with the approved guidelines. All experimental protocols were approved by the institutional review board.

Maintenance of hiPSCs cultures Three human iPSC lines were used for hepatic differentiation (hiPSCs-HF1, hiPSCs-HF2 and hiPSCsEC1). hiPSCs-HF1 and hiPSCs-HF2 were induced from IMR-90 human fibroblasts using Oct4, Sox2, Klf4 and c-myc. hiPSCs-EC1 were induced from urine of renal epithelial cells using Oct4, Sox2, Klf4 and c-myc. In our laboratory, theses hiPSC cell lines were routinely passaged on a feeder layer of mitomycin-C (Roche)–inactivated mouse embryonic fibroblasts. Briefly, cells were maintained on an inactivated feeder layer and were cultured in hiPSC medium (Dulbecco’s Modified Eagle Medium [DMEM]/F12 supplemented with 20% knockout serum replacement, 8 ng/ mL basic fibroblast growth factor [bFGF], 1 mmol/L nonessential amino acids, 1 mmol/L L-glutamine and 0.1 mmol/L 2-mercaptoethanol; all from Thermo Fisher Scientific) at 37°C in 5% CO2. Before differentiation, cells were harvested and separated indigestion solution containing Dispase (Stemcell Technologies), and they were then transferred to plates coated with Matrigel (BD Biosciences) grown in mTesR1 (Stemcell Technologies). Hepatic differentiation of hiPSCs Before differentiation, hiPSCs were used to form embryoid bodies (EBs). Feeder-free hiPSCs were digested to a single-cell suspension by Accutase (Stemcell Technologies), and they were cultured in AggreWell 400 plates (Stemcell Technologies; ~ 1.2 × 106 cells/well) for 24 h, and EBs (consisting of ~1000 cells) were later transferred to Ultra Low Cluster Plates (Corning) and cultured for another 24 h in DMEM/F12 supplemented with 20% knockout serum replacement, 1 mmol/L nonessential amino acids, 1 mmol/L L-glutamine and 0.1 mmol/L 2-mercaptoethanol. In the first stage, EBs were cultured on Matrigel-coated well plates (~100 EBs/well) and maintained in DMEM/ F12 supplemented with 100 ng/mL recombinant Activin-A and 100 ng/mL bFGF (all from R&D Systems). The concentration of fetal bovine serum (FBS) was 0% for the first 24 h, 0.2% for the second 24 h and 2.0% for the last 24 h. In the second stage, the differentiated cells were further induced to hepatoblasts in DMEM/F12 media containing 10% FBS, 1 mmol/L nonessential amino acids, 1 mmol/L L-glutamine, 1% dimethyl sulfoxide (Sigma-Aldrich) and 100 ng/mL hepatocyte growth factor (HGF; R&D Systems) for the following 8 days. In the third stage, hepatoblasts were switched to maturation media containing 10% FBS, 1 mmol/L nonessential amino acids, 1 mmol/L L-glutamine and 0.1 µmol/L dexamethasone (Sigma-Aldrich) for 3 days. For 3D differentiation, we initiated the differentiation of the cells in the same condition as the 2D culture until the first day of the

ARTICLE IN PRESS 3D culture promotes hiPSC-differentiated hepatocytes

3

Figure 1. Generation of HLCs from hiPSCs by sequential application of monolayer and 3D culture systems. (A) Stepwise 3D differentiation strategy. In the first stage, embryoid bodies (EBs) were cultured on Matrigel-coated well plates and maintained in DMEM/F12 supplemented with 100 ng/mL recombinant Activin-A (Act) and 100 ng/mL bFGF. In the second stage, the differentiated cells were embedded into 0.25% (w/v) hydrogel and induced to hepatoblasts in DMEM/F12 media containing 1% dimethyl sulfoxide and 100 ng/mL hepatocyte growth factor (HGF). In the third stage, hepatoblasts were switched to maturation media containing 0.1 µmol/L dexamethasone (Dex). (B) Bright field microscopic image shows the morphological changes of hiPSCs with EBs formation in monolayer cultures on first (day 1) and third (day 3) days of differentiation. (C) Immunofluorescence images showing differentiated cells with scattering distribution could express FOXA2 on day 3. For nuclear staining (blue) 4,6-diamidino-2-phenylindole was used. (D) Flow cytometry analysis of definitive endoderm maker CXCR4 on day 3. Bar = 50 µm.

second stage when they were inoculated into the hydrogel (Beaver). Before inoculation, cells were treated with 10 µmol/L ROCK inhibitor Y-27632 (ROCKi; Merck Millipore) for 2 h before they were detached from the culture dish with Accutase. The undifferentiated cells were seeded at a density of 106/mL in 0.25% (w/v) hydrogel supplemented with 10 µmol/L ROCKi, which was removed after 48 h. Cells were differentiated with same media as the 2D cultures in the 3D

culture condition (Figure 1A). Independent culture experiments were performed three times. Nanofiber hydrogel preparation and cell seeding Prepared the appropriate volume of hydrogel (peptide hydrogel, 1% w/v, Beaver) in a microtube by diluting the stock with sterile 20% sucrose to generate a 2 × concentration. The cells were disassociated to a

ARTICLE IN PRESS 4

Y. Luo et al.

single cell and suspended in 10% (w/v) sucrose solution at a final concentration of 106 cells/mL. A volume of this suspension was gently mixed with an equal volume of the peptide solution and then carefully added hydrogel/cell/sucrose mixture to the center of the well, without introducing bubbles. The media was changed two times over the next hour to further equilibrate the pH of the hydrogel. Analysis of the cell behavior in nanofiber hydrogel 3D scaffold Cell cultures were monitored by phase contrast microscopy (DM IL LED, Leica), and images were taken at regular time intervals. The average diameter and number of spheroids was recorded using images taken under a phase-contrast microscope and using the medium axis of the spheroid. To determine the percentage of viable cells, spheroids were stained by propidium iodide (PI) and Hoechst33342. Differentiated cells were pulsed with 10 µg/mL Hochest33342 for 10 min at room temperature, and 50 µg/mL PI for 15 min at 4°C. After the incubation, the stained cells were analyzed by fluorescence microscopy using UV and 488 nm. Human hepatocyte isolation and culture Normal human livers were obtained from patients with liver-occupying lesions but no cirrhosis who required hepatectomy at Tianjin Third Central Hospitals after obtaining informed permission from their parents. All specimens were prepared in accordance with the guidance of the Ethic Committee of Tianjin Third Central Hospitals. The human livers were collected in cold phosphate-buffered saline (PBS) containing 100 U/mL penicillin and 100 U/mL streptomycin. Tissue dissociation and hepatocyte isolation were performed using a two-step liver digest medium perfusion procedure.The isolated fresh human hepatocytes were seeded at pre-coated Rat Tail Collagen I (12.5 µg/ cm2) culture plate supplemented with HepatoZYMESFM (all from Thermo Fisher Scientific). Four donors of fresh primary human hepatocytes (PHs) were used as control in the experiments, and the data were averaged across these four donors of PHs. Flow cytometry On day 3 of differentiation, cells were dissociated in 0.05% trypsin–ethylenediaminetetraacetic acid and washed twice in PBS. Cells were incubated with 20 µL of the solution mouse anti-human CXCR4 antibody or the isotype controls for 30 min (BD Biosciences). Cells were washed twice in 1 mL PBS and resuspended in 500 µL PBS. Flow cytometric analysis was

performed with BD-FACS Canto II Flow Cytometer (BD Biosciences). Acquired data was analyzed with FlowJo7.6 software.

Immunofluorescence cell staining Spheroids were collected, fixed with 4% formaldehyde for 24 h, cryoprotected in 30% sucrose and embedded in Tissue-Tek OCT compound (Sakura). Spheroid cryosections permeabilized with 0.1% Triton X-100 for 10 min and blocked in 1% bovine serum album in PBS for 30 min at room temperature. Next, the cells were incubated at 4°C overnight with the following primary antibodies: alpha-fetoprotein (AFP, Santa Cruz Biotechnology), forkhead box protein A2 (FOXA2, Santa Cruz Biotechnology), Zonula occludens-1 (ZO-1, Abcam), ALB (Bethyl), F-actin (Abcam) and asialoglycoprotein receptor (ASGPR, Santa Cruz Biotechnology). Alexa Fluor 488-conjugated and Alexa Fluor 546-conjugated secondary antibodies (Thermo Fisher Scientific) were applied to the cells for 1 h at room temperature. Cell nuclei were stained with 4,6-diamidino-2phenylindole. Isotype contrast immunoglobulin G was added to the control group. The cells were analyzed with confocal Imaging System (UltraVIEW VoX; PerkinElmer).

Measurement of albumin and urea production by enzyme-linked immunosorbent assay On day 14 of differentiation, the medium was changed to serum-free hepatocyte culture media (HepatoZYME-SFM) supplemented with 5 mmol/L NH4Cl, and cells were cultured for 24 h. The supernatant was collected and stored at −20°C until assay. Albumin (ALB) was determined by a commercially available enzyme-linked immunosorbent assay (ELISA) quantization kit (Bethyl), and urea production was determined by QuantiChrom™ Urea Assay Kit (BioAssay Systems). ALB secretion and urea production were normalized to total cell protein content per well.

Periodic acid Schiff staining for glycogen To detect glycogen, 14-day differentiated cells were stained using a Periodic acid Schiff (PAS) staining kit (Bestbio). Briefly, cells were fixed in 4% paraformaldehyde for 10 min, and they were subsequently oxidized by incubation with a 1% periodic acid solution for 10 min. The cells were later incubated with freshly prepared Schiff’s reagent at 37°C for 30 min. After the cells were rinsed with sulfurous acid, nuclei were counterstained by incubation with Mayer’s hematoxylin for 5 min.

ARTICLE IN PRESS 3D culture promotes hiPSC-differentiated hepatocytes

5

Table I. Primer sequences and product sizes. Gene name G6Pase CYP2C8 CYP2C9 CYP2C18 CYP1A2 CYP2D6 CYP3A7 CYP7A1 CYP3A4 GAPDH

Primer sequences

Product size (bp)

Annealing temperature (°C)

Sense 5′-TTTGGGTAGCTGTGATTGGAGACT-3′ Antisense 5′-TCTCACAGGTTACAGGGAACTGCT-3′ Sense 5′-CCCATGCAGTGACCACTGATAC-3′ Antisense 5′-TGGCAGAGAAACAATCCCTTTG-3′ Sense 5′-AATGAAAACATCAAGATTTTGAGCAGC-3′ Antisense 5′-CCAAACAAGTCAACTGCAGTGTTTTC-3′ Sense 5′-GGTGCTCTGTCTCTCCTGTTTGT-3′ Antisense 5′-CAACACCACAATGGGCTTCAG-3′ Sense 5′-ATCCAGATATGCAATAATTTTCCCAC-3′ Antisense 5′-TGTCTCTGTCCCAGCTCCAAGT-3 Sense 5′-CATAGTGGTGGCTGACCTGTTCT-3′ Antisense 5′-GTGATGAGTGTCGTTCCCTTAGG-3 Sense 5′-GGGCAGGAAAGCTCCACACACAC-3′ Antisense 5′-CGGGTTCCATATAGATAGAGGAGTAT-3′ Sense 5′-GAACCCAGAAGCAATGAAAGC-3′ Antisense 5′- GAAATCCTCCTTAGCTGTCCG-3′ Sense 5′-TGTCCTACCATAAGGGCTTTTGTATG-3′ Antisense 5′- TTTCACTAGCACTGTTTTGATCATGTCA-3′ Sense 5′-GGGCATCCTGGGCTACACTGA-3′ Antisense 5′-CAAATTCGTTGTCATACCAGGAAATG-3′

147

65

341

60

284

60

211

60

267

60

303

60

175

60

202

60

137

60

143

60

CYP, Cytochrome P450; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; G6Pase, glucose-6-phosphatase.

RNA isolation and quantitative reverse transcription polymerase chain reaction Total cellular RNA was extracted using the Trizol Reagent (Thermo Fisher Scientific), and it served as a template for cDNA synthesis, which was performed using the superscript II first-strand synthesis system (Promega). SYBR Green quantitative reverse transcription polymerase chain reaction (qRT-PCR) was performed using the ViiA 7 Real-Time PCR System (Applied Biosystems).The change in the Target gene expression normalized to the housekeeping gene GAPDH and relative to the expression at PHs was calculated for each group where ΔΔCT = (CT target- CT GAPDH) each group – (CT target- CT GAPDH) PHs group. The amplification efficiencies of the target and reference genes were approximately equal. The oligonucleotide sequences of the primers designed for real-time RT-PCR are listed in Table I. CYP450 induction and CYP450 activity assays The differentiated 14 days cells cultured in 2D and 3D culture conditions were treated with 10 mmol/L rifampicin (Aladdin) for 48 h, which is known to induce cytochrome 2C9 (CYP2C9) and 3A4 (CYP3A4), and the cells were also treated with 1 µmol/L ketoconazole (Aladdin) or 2 mmol/L sulfaphenazole (Santa Cruz) for 48 h, which are inhibitors for CYP3A4 and2C9, respectively. Subsequently, the CYP3A4 and 2C9 activities of the cells were measured using a P450GloTM CYP2C9 and 3A4 Assay Kit (Promega). We performed nonlytic assays according to the manufacturer’s instructions. Activity was measured using a

luminometer (GloMax-96 Microplate luminometer; Promega).Values were corrected to account for background.The CYP activity was normalized to total cell protein content per well. Statistical methods Statistical analysis was performed with SPSS 19.0 statistical software. The results are presented as means ± SD from at least three independent experiments. Differences among multiple groups were analyzed using one-way analysis of variance. Comparisons between two groups were conducted using Student’s t-test. A P value