Enhanced Hepatic Differentiation of Human Induced Pluripotent Stem Cells Using Gas-Permeable Membrane

Abstract

In an effort to construct in vitro liver tissue models for drug screening, human induced pluripotent stem cell (hiPSC)-derived hepatocyte-like cells (iHeps) are a promising cell source alternative to primary human hepatocytes (PHHs). However, thus far, the functions of iHeps are significantly reduced when compared to those of PHHs. In this study, we aimed to determine the role of oxygen supply to the cell layer, which has not been investigated as thoroughly as other conditions, such as soluble factors, the extracellular matrix, or co-culture with nonparenchymal cells, in improving the hepatic function of iHeps. Polydimethylsiloxane (PDMS), which has a high oxygen permeability, was used for cell culture [O₂ (+)], in addition to conventional tissue culture-treated polystyrene [TCPS; O₂ (−)]. During endoderm differentiation, O₂ (+) culture resulted in reduced expression of HIF-1α. As a result, we observed higher expression of the endoderm marker (GATA4). After hepatic differentiation, iHeps generated in O₂ (+) culture showed about a threefold increase per cell in albumin production. Moreover, among the O₂ (+) culture conditions, cells differentiating at a physiological oxygen concentration showed increased expression of several mature hepatic marker genes (ALB, CYP3A4, and CYP1A2) and increased ratio of ALB-positive cells. In O₂ (+) culture, iHeps exhibited a per cell oxygen consumption rate two times higher than that observed in O₂ (−) culture. Furthermore, in sufficiently oxygen-supplied culture systems, cells showed a self-organized, multilayered configuration, and the DNA content in the culture was two to three times higher than that in O₂ (−) culture. Considering the amount of DNA and the ration of iHeps in each culture condition, we could achieve about 7.5 times higher per unit surface area productivity of iHeps in O₂ (+) culture. Therefore, these results demonstrate that direct oxygenation through a PDMS membrane enhances the maturation and efficient production of iHeps. Thus, improving oxygen supply to the cell layer is important for iHeps culture, and direct oxygenation through a PDMS membrane is a promising alternative to conventional TCPS plates for the maturation of iHeps in vitro.

Impact Statement

Although oxygen is a vital nutrient for the hepatocytes in vitro, few reports have focused on its effect during hepatic differentiation of induced pluripotent stem cells (iPSCs). In this report, we performed the hepatic differentiation of human iPSCs (hiPSCs) under different atmospheric oxygen concentrations and oxygen supply fluxes to investigate the effects of oxygen in terms of both the concentration and the supply flux. Results demonstrate that direct oxygenation through a polydimethylsiloxane (PDMS) membrane enhances the maturation and efficient production of hiPSC-derived hepatocyte-like cells (iHeps). Thus, direct oxygenation through a PDMS membrane is a better alternative culture method over conventional tissue culture-treated polystyrene (TCPS) plates for the maturation of hiPSC-derived hepatocytes in vitro.

Introduction

The liver plays crucial roles in drug metabolism and detoxification, and thus, hepatocytes are often used in drug screening, toxicology studies, and in vitro disease models.¹ Although primary human hepatocytes (PHHs) are ideal for in vitro studies,^2,3 they are severely limited in number owing to the shortage of donors.^4,5 As an alternative cell source, animal-derived primary hepatocytes or cell lines have been used for in vitro studies⁶; however, these present challenges are owing to species-specific differences or reduced function, respectively. To overcome these hurdles, human induced pluripotent stem cell (hiPSC)-derived hepatocyte-like cells (iHeps) have attracted considerable attention because hiPSCs can differentiate into almost any cell type and can proliferate indefinitely.^7–9

Many hepatic differentiation protocols of hiPSCs have been reported in the last decade. In these protocols, iHeps are generated by the sequential addition of growth factors that are known to play important roles in liver organogenesis.^10–13 Although the iHeps generated using these protocols show some hepatic functions, such as albumin (ALB) production and cytochrome P450 (CYP) activity, these functions are significantly reduced compared to those in PHHs.¹⁴ To tackle this problem, many researchers have sought to improve culture systems for the further maturation of iHeps, including co-culture^1,15 or three-dimensional (3D) culture.^16–18 Berger et al. used micropatterned co-culture with stromal fibroblasts and reported improvements in the gene expression of several mature hepatic markers, as well as ALB and urea synthesis and several CYP activities.¹ Gieseck et al. reported that iHeps cultured in the Real Architecture for 3D Tissues (RAFT) system showed significantly improved functional maturation compared to iHeps cultured in a conventional two-dimensional (2D) system.¹⁷ Although iHeps cultured using these culture methods showed higher functional maturity compared to those derived from conventional culture systems, these cells still showed a fetal phenotype or reduced function compared to PHHs.

Because oxygen is essential during embryo development and liver organogenesis,^19–21 the oxygen environment of the culture system is also an important factor in the differentiation of hiPSCs and maturation of iHeps in vitro. To enhance oxygen conditions in vitro, some efforts have focused on the atmospheric oxygen conditions. Si-Tayeb et al. reported a hepatic differentiation protocol in which they applied different atmospheric oxygen concentrations to each differentiation step. However, they did not discuss the advantage of applying different oxygen concentrations.¹⁰ Moreover, although they focused on the atmospheric oxygen concentration, they did not control the amount of oxygen supplied to the cell layer, which is what cells utilize for their respiration and energy synthesis.

Oxygen supply to the cell layer in a conventional tissue culture-treated polystyrene (TCPS) plate depends only on diffusion through culture media. Because this diffusion-based oxygen supply is severely limited by the low solubility and diffusivity of oxygen, it is unable to meet the cellular oxygen demand.²² Employing an oxygen-permeable membrane as the bottom of the culture plate is one of the simplest methods for enhancing oxygen supply to the cell layer. Xiao et al. reported that direct oxygenation through a polydimethylsiloxane (PDMS) membrane, which has a high oxygen permeability, could completely meet the cellular oxygen demands of rat hepatocytes in vitro.²⁴ Although many advantages of employing PDMS membranes at the culture plate bottom have been shown in the culture of hepatocytes (rat primary hepatocytes and human hepatoma HepG2 cells) according to previous reports,^23–26 there have been few reports assessing the beneficial effects of direct oxygenation through a PDMS membrane on the hepatic differentiation of hiPSCs.

In this study, we performed the hepatic differentiation of hiPSCs using several different atmospheric oxygen concentrations in both TCPS and PDMS plates to investigate the effects of oxygen supply in terms of both the concentration at the cell layer and the oxygen supply flux. Comparisons of different conditions for differentiation revealed that the expression of hepatic genes, such as ALB, CYP3A4, and CYP1A2, was upregulated in cells differentiated on PDMS plates with physiological oxygen concentrations (5% or 10%). Interestingly, cells differentiated under low oxygen supply conditions exhibited significantly reduced hepatic functions, implying that hepatic differentiation failed under a low oxygen supply. Furthermore, in the PDMS culture system, cells exhibited a self-organized, multilayered configuration, and the DNA content of the culture was two to three times higher than that in TCPS culture. These results clearly demonstrate that improving oxygen supply to the cell layer using PDMS membranes is key to improving the hepatic differentiation of hiPSCs.

Materials and Methods

High- and low-oxygen cell culture

To enable direct oxygenation through PDMS membranes, hiPSCs (TkDN4-M), provided by the Center for Stem Cell Biology and Regenerative Medicine at the University of Tokyo, were cultured on commercially available polycarbonate 24-well plates assembled with a 0.35-mm thick PDMS membrane (Vessel) [O₂ (+) condition]. In the low-oxygen control condition [O₂ (−)], hiPSCs were cultured on PDMS discs inserted into the wells of a conventional TCPS plate (Iwaki) to keep the same surface characteristics both in O₂ (+) and O₂ (−) culture conditions (Fig. 1). Before cell inoculation, the PDMS surfaces were treated with oxygen plasma for 60 s using a YHS-GZA 200 (SAKIGAKE-Semiconductor) and coupled with 3-mercaptopropyltrimethoxysilane (Shinetsu Silicone). Amino groups introduced were subsequently reacted with a cross-linker, N-(4-maleimidobutyryloxy) succinimide (GMBS; Dojindo), activated by 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide, hydrochloride (WSC; Dojindo), and N-hydroxysulfosuccinimide sodium salt (sulfo-NHS; TCI). Finally, the PDMS surfaces were coated with Matrigel (Corning) to allow cells to attach. hiPSCs were inoculated and cultured on Matrigel-coated O₂ (−) and O₂ (+) plates with mTeSR1 medium (Stem Cell Research).

FIG. 1.

Schematic illustration of cell culture on PDMS discs inserted into the wells of TCPS plates [O₂ (−), left] or on PDMS plates [O₂ (+), right]. PDMS, polydimethylsiloxane; TPCS, tissue culture-treated polystyrene.

Hepatic differentiation of hiPSCs

Using 90% confluent hiPSCs, hepatic differentiation was induced using a four-step protocol as previously reported¹⁰ with minor modifications²⁷: endodermal differentiation (step 1), specified hepatic cell differentiation (step 2), hepatoblast differentiation (step 3), and hepatic differentiation (step 4). Briefly, hiPSCs reached 90% confluence were cultured with RPMI medium supplemented with B-27 supplement (Gibco) and 100 ng/mL Activin A (R&D) for 5 days to differentiate into endodermal cells. After that, endodermal cells were cultured with RPMI medium supplemented with B-27 supplement, 10 ng/mL b-FGF (Peprotech), and 20 ng/mL BMP-4 (Humanzyme) for 5 days to induce hepatic progenitor differentiation. Then, cells were cultured with RPMI medium supplemented with 20 ng/mL HGF (Peprotech) for 5 days to induce hepatic differentiation. Finally, cells were cultured with hepatocyte culture medium (HCM; Lonza, without EGF) supplemented with oncostatin M (Peprotech) for 5 days to induce hepatic maturation. To evaluate the effect of oxygen supply during differentiation, we performed hepatic differentiation at various atmospheric oxygen concentrations (20%, conventional culture condition; 5% and 10%, physiological oxygen condition) under both O₂ (−) and O₂ (+) culture conditions (Table 1). We used cells differentiating in O₂ (−) plates at 20% atmospheric oxygen as the control group, as these conditions were used in the original protocol.¹⁰

Table 1.

Atmospheric Oxygen Concentration of Each Culture Condition During Hepatic Differentiation

	Step 1 (%)	Step 2 (%)	Step 3 (%)	Step 4 (%)
O₂ (−)-5%	5	5	5	5
O₂ (−)-10%	10	5	5	10
O₂ (−)-20%	20	5	5	20
O₂ (+)-5%	5	5	5	5
O₂ (+)-10%	10	5	5	10
O₂ (+)-20%	20	5	5	20

Immunostaining

Cells were fixed with 4% paraformaldehyde at room temperature for 30 min and then incubated with 1% Triton-X100 (WAKO) for permeabilization. The cells were further incubated with 1% bovine serum albumin for blocking, followed by incubation with primary antibody overnight at 4°C. Cells were subsequently incubated with secondary antibody for 3 h and then counter-stained with 4′,6-diamidino-2-phenylindole (DAPI; Dojindo) for 3 min. The cells were rinsed with phosphate-buffered saline thrice for 5 min each after each step was finished. Fluorescent images were taken using a confocal microscope (PowerIX70; Olympus). Antibodies used for immunostaining are shown in Table 2. The ratio of marker protein-positive cells to DAPI-positive cells was manually counted using Image J (NIH) from three different areas of each sample.

Table 2.

List of Antibodies Used for Immunostaining

Antigen	Host	Distributor	Product No.
GATA4	Goat	Santa Cruz	sc1237
AFP	Rabbit	EnoGene	E90002
ALB	Goat	Bethyl	A80-129A
Alexa Fluor anti-goat 594	Donkey	Invitrogen	A11058
Alexa Fluor anti-rabbit 488	Donkey	Abcam	ab150073

Real-time quantitative polymerase chain reaction

Total RNAs were isolated and purified from samples using TRIzol Reagent (Life Technologies) according to the manufacturer's instructions. The concentration and quality of each RNA sample were assessed with the NanoDrop system (BioSpec-nano; SHIMADZU). Subsequently, complementary DNAs (cDNAs) were reverse-transcribed from 0.5 μg total RNA using Rever Tra Ace qPCR Master Mix with gDNA Remover (TOYOBO) according to the manufacturer's protocol. Real-time quantitative polymerase chain reaction (qPCR) assays were performed with THUNDERBIRD SYBR qPCR Mix (TOYOBO) and assessed using the StepOnePlus qPCR system (Applied Biosystems). Primer sequences used for real-time qPCR are shown in Table 3. β-Actin (bACT) expression was used as the internal control.

Table 3.

List of Real-Time Quantitative Polymerase Chain Reaction Primers

Primer name	Sequence 5′-3′	Temperature (°C)
bACT_forward	CCTCATGAAGATCCTCACCGA	58.97
bACT_reverse	TTGCCAATGGTGATGACCTGG	60.89
OCT4_forward	AGTGGGTGGAGGAAGCTGACAAC	63.7
OCT4_reverse	TCGTTGTGCATAGTCGCTGCTTGA	64
GATA4_forward	AGCACACTGCATCTCTCCTGTG	61.77
GATA4_reverse	CTCCGCTTGTTCTCAGATCCTC	60.53
AFP_forward	TGGGACCCGAACTTTCCA	58.01
AFP_reverse	GGCCACATCCAGGACTAGTTTC	60.68
Albumin_forward	CCTGCTGACTTGCCTTCATTAG	61.3
Albumin_reverse	TGGCATAGCATTCATGAGGA	60.2
CYP1A2_forward	CAGCTTCCTCATCCTCCTGCTA	61.28
CYP1A2_reverse	AGGCTGAGCATCTCATCGCTAC	62.23
CYP3A4_forward	ACATAGCCCAGCAAAGAGCAAC	61.46
CYP3A4_reverse	GTCTGGGATGAGAGCCATCACT	61.55

Measurement of albumin production

The culture medium was changed to fresh HCM (Lonza) on day 20, the last day of differentiation, and differentiated cells were cultured for 24 h. After 24 h of incubation, the culture supernatant was collected for sandwich-type enzyme-linked immunosorbent assay to determine the amount of secreted albumin. Goat anti-human albumin antibody (Bethyl) was used as the capture antibody, horseradish peroxidase-conjugated goat anti-human albumin (Bethyl) was used as the detection antibody, and human reference serum (Bethyl) was used as the standard human albumin solution. The absorbance was measured by the microplate reader iMark (Bio-Rad) at 450 nm with an optical correction of 630 nm.

Measurement of DNA amount

Cells were detached by trypsin-EDTA treatment on day 20 and were homogenized with an Ultra Sonic Homogenizer UH-300 (SMT) before DNA measurement. DNA content was measured according to previously reported protocols.²⁸ DAPI was used as a fluorescent dye, and the intensity was assessed with the RF-5300PC spectrofluorophotometer (SHIMADZU). The excitation wavelength was set at 350 nm, and the emission wavelength at 450 nm.

Measurement of glucose production and lactate consumption

Culture supernatant was collected and stored at −20°C after every medium change. The concentrations of glucose and lactate in the samples were measured with the YSI 2950D Biochemistry Analyzer (YSI).

Measurement of oxygen tension

Oxygen concentrations at the cell layer were measured with a fluorescent oxygen probe (Firesting; Pyro Science) on day 20 as previously described.²⁴ Briefly, the probe was put about 30 μm above the cell layer. The concentrations were monitored continuously and read after they reached their steady state.

Statistical analysis

All data are represented as the mean ± standard deviation. Student's t-tests were used to determine the statistical significance of two experimental groups. Differences with p < 0.05 were considered to be significant.

Results

Direct oxygenation through PDMS enhanced endoderm differentiation in hiPSCs

To evaluate the effect of oxygen supply on endoderm differentiation in hiPSCs, endodermal cells generated under each condition were stained by labeling GATA binding protein 4 (GATA4), a known endodermal marker (Fig. 2A). Almost all cells differentiating at O₂ (−)-20% and O₂ (+)-5%, O₂ (+)-10%, and O₂ (+)-20% expressed GATA4 (89.4% ± 9.6%, 81.4% ± 16.0%, 87.6% ± 3.7%, and 77.2% ± 20.4%, respectively) (Table 4), indicating that these cells properly differentiated toward the endodermal lineage. In contrast, the number of cells expressing GATA4 decreased under O₂ (−)-5% and O₂ (−)-10% conditions (13.2% ± 9.1% and 54.3% ± 18.3%) (Table 4). Notably, almost all cells differentiating at O₂ (−)-5% did not express GATA4. This result indicates that reduced oxygen supply to the cell layer inhibited the endodermal differentiation of hiPSCs.

FIG. 2.

(A) Immunohistochemical analysis of GATA4 in endodermal cells on day 5. Blue, DAPI; red, GATA4. Scale bar represents 100 μm. (B) OCT4 (top) and GATA4 (bottom) expression in endoderm cells. Data represent the mean ± SD (n = 3). Asterisks indicate significant difference versus O₂ (−)-20%; *p < 0.05, **p < 0.01, ***p < 0.001. DAPI, 4′,6-diamidino-2-phenylindole; SD, standard deviation.

Table 4.

The Ratio of GATA4-Positive Cells to DAPI-Positive Cells on Day 5

	O₂ (−)			O₂ (+)
	5%	10%	20%	5%	10%	20%
GATA4⁺/DAPI (%)	13.2 ± 9.1	54.3 ± 18.3	89.4 ± 9.6	81.4 ± 16.0	87.6 ± 3.7	77.2 ± 20.4

DAPI, 4′,6-diamidino-2-phenylindole.

Next, we performed real-time qPCR analysis to confirm the gene expression levels of both octamer binding transcription factor 4 (OCT4) and GATA4 in cells differentiating in each culture condition (Fig. 2B). When compared to cells differentiating at O₂ (−)-20%, cells differentiating under lower oxygen supply conditions [O₂ (−)-5% or O₂ (−)-10%] showed higher expression of OCT4 and significantly lower expression of GATA4. In contrast, cells differentiating under O₂ (+) conditions showed significantly higher expression of GATA4. However, there was no significant difference in the expression of OCT4 among cells differentiating at O₂ (−)-20% and O₂ (+)-5%, O₂ (+)-10%, and O₂ (+)-20%. The gene expression levels of OCT4 during differentiation at O₂ (−)-20% decreased during endoderm differentiation, and this reduced expression was maintained to the last stage of differentiation (data not shown). The gene expression levels of OCT4 under these conditions may have reached the lower limit of expression, and thus, we were unable to observe significant differences. Based on these results, direct oxygenation through the PDMS membrane is effective for endoderm differentiation of hiPSCs.

Respiratory state and HIF-1α expression in endoderm cells

To determine the effect of direct oxygenation through PDMS on endoderm differentiation, we first determined the respiratory states of differentiating cells. The glucose and lactate concentrations in the culture supernatants on day 5 were measured, and the ratio of the lactate production (mol) to the glucose consumption (mol) was calculated (Fig. 3A). Cells differentiating under all conditions showed similar values close to 2 (mol/mol), showing that these cells were mostly performing anaerobic respiration.²³ For further evaluation, the gene expression of hypoxia inducible factor 1α (HIF-1α), which is highly expressed under hypoxic conditions,²⁹ was measured (Fig. 3B). As oxygen supply to the cell layer increased, the HIF-1α expression was reduced. These results confirm the relationship between HIF-1α expression and the endodermal differentiation of hiPSCs.

FIG. 3.

(A) Ratio of lactate production (mol)/glucose consumption (mol) under different culture conditions on day 5. (B) HIF-1α expression in endoderm cells. Data represent the mean ± SD (n = 3).

Cellular oxygen consumption was enhanced by direct oxygenation on day 20

To investigate the effect of direct oxygenation in O₂ (+) culture on iHep respiration, we measured the oxygen concentration above the cell layer on day 20 (Fig. 4A) and calculated the oxygen supply flux to the cell layer, which is the same value as the oxygen consumption rate (OCR), determined using the following equations [Eq. (1) for culture media, Eq. (2) for PDMS membrane]. The parameters used for the calculations are shown in Table 5. \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*} F = D \frac { { \left( { p { O_ { 2 , \;air } } - p { O_ { 2 , \;cell } } } \right) K } } { z } \tag { 1 } \end{align*} \end{document} \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*} F = P \frac { { \left( { pO_2 , air - pO_2 , \;cell } \right) } } { z } \tag { 2 } \end{align*} \end{document}

FIG. 4.

(A) Oxygen concentration at the cell layer and (B) calculated value of consumed oxygen on day 20.

Table 5.

Parameters That Were Used for Oxygen Consumption Rate Calculation

Parameter	Definition	Value	Unit
F	Supplied oxygen flux		pmol/(cm²·s)
F_medium	Supplied oxygen flux from the medium		pmol/(cm²·s)
F_PDMS	Supplied oxygen flux from PDMS membrane		pmol/(cm²·s)
D_medium	Diffusion coefficient of oxygen in the medium²⁴	2.0 × 10⁻⁵	cm²/s
K_medium	Oxygen solubility in the medium²⁴	1.19 × 10³	pmol/(cm³·mmHg)
P_PDMS	Permeability coefficient of oxygen in PDMS	1.72 × 10⁻¹	(pmol·cm)/(s·cm²·mmHg)
Z_medium	Depth of the medium	0.26	cm
Z_PDMS	Thickness of the PDMS membrane	0.035	cm
pO_2air	Oxygen partial pressure in incubator		mmHg
pO_2cell	Oxygen partial pressure at cell layer		mmHg

PDMS, polydimethylsiloxane.

The calculated OCRs are shown in Figure 4B. Cells differentiating under O₂ (−) conditions consumed <0.5 pmol-O₂/s/μg DNA. In contrast, cells differentiating at O₂ (+)-5%, O₂ (+)-10%, or O₂ (+)-20% consumed oxygen at rates 2.2, 2.9, and 6.0 times higher, respectively, than the rate in cells at O₂ (−)-20%. Thus, cells in O₂ (+) culture utilize more oxygen for their respiration, suggesting that these cells synthesize energy more efficiently.

Physiological oxygen supply enhanced the hepatic differentiation of hiPSCs

To investigate the hepatic maturity of iHeps under each condition, cells were differentiated to iHeps subsequently to endodermal differentiation. Based on immunostaining results, we observed many AFP-positive cells in O₂ (−)-20% and O₂ (+)-5%, O₂ (+)-10%, and O₂ (+)-20% condition. In contrast, we observed more cells that were positive for ALB under O₂ (+) conditions than at O₂ (−)-20% (Fig. 5 and Table 6), indicating that hepatic maturation was highly induced in O₂ (+) conditions. Moreover, when we compared cells differentiating under each O₂ (+) culture condition, more cells were positive for ALB when differentiating at physiological oxygen concentrations (5% or 10%). In contrast, under O₂ (−) conditions, cells differentiating at lower oxygen concentrations [O₂ (−)-5% and O₂ (−)-10%] tended to exhibit reduced ALB expression according to immunostaining than those differentiating at O₂ (−)-20%. In particular, we did not detect any ALB-positive cells at O₂ (−)-5%. These cells were also negative for alpha fetoprotein (AFP), indicating that these cells did not differentiate into hepatocytes.

FIG. 5.

Immunohistochemical analysis of AFP and ALB in iHeps on day 20. Blue, DAPI; green, AFP; red, ALB. Scale bar represents 100 μm.

Table 6.

The Ratio of AFP-Positive or ALB-Positive Cells to DAPI-Positive Cells on Day 20

	O₂ (−)			O₂ (+)
	5%	10%	20%	5%	10%	20%
AFP⁺/DAPI (%)	ND	37.6 ± 20.5	55.6 ± 9.4	67.6 ± 0.8	26.0 ± 1.6	45.2 ± 3.8
ALB⁺/DAPI (%)	ND	1.6 ± 1.0	8.1 ± 5.1	25.8 ± 9.2	22.5 ± 1.8	9.3 ± 0.8

Next, we performed real-time qPCR to evaluate the gene expression levels of ALB (Fig. 6A), CYP3A4 (Fig. 6B), and CYP1A2 (Fig. 6C). Consistent with the immunostaining results, cells differentiating under O₂ (+) culture conditions with lower atmospheric oxygen concentrations (5% O₂) showed significantly higher expression of ALB. In contrast, cells differentiating under O₂ (−)-5% conditions showed significantly lower expression of ALB. Similar to the ALB results, cells differentiating at O₂ (+)-5% showed significantly higher expression of CYP3A4 and CYP1A2. For CYP3A4, cells differentiating at O₂ (+)-10% also showed a significant increase in gene expression. However, cells differentiating at O₂ (+)-20% were not significantly different from those differentiating at O₂ (−)-20%. These results suggest that a sufficient oxygen supply with a physiological oxygen concentration induces hepatic maturation. However, these cells still expressed high levels of AFP, which indicates an immature hepatic phenotype (Fig. 6D), and the CYP3A4 activity was quite low in both O₂ (+) and O₂ (−) culture (data not shown).

FIG. 6.

Relative gene expression of ALB (A), CYP3A4 (B), CYP1A2 (C), and AFP (D) and secretion of ALB (E) in iHeps differentiating under each condition on day 20. Data represent the mean ± SD (n = 3). Asterisks indicate significant difference versus O₂ (−)-20%: *p < 0.05, **p < 0.01, ***p < 0.001. iHeps, human induced pluripotent stem cell-derived hepatocyte-like cells.

We also measured the cellular ALB secretion ability to confirm hepatic maturation (Fig. 6E). As a result, cells differentiating in O₂ (+) culture tended to exhibit higher ALB production than those differentiating at O₂ (−)-20%. In contrast, cells differentiating at O₂ (−)-5% showed much lower ALB secretion. Thus, improving oxygen supply to the cell layer was effective in achieving hepatic functional activity.

Multilayered structure obtained with direct oxygenation

To evaluate cell proliferation during hepatic differentiation, we measured the DNA content on day 20 (Fig. 7A). Compared to the DNA content of the O₂ (−)-20% condition, the O₂ (+)-5%, O₂ (+)-10%, and O₂ (+)-20% cultures exhibited 2.7, 2.2, and 2.2 times the amount of DNA, respectively. In other words, we achieved cell numbers two to three times higher than the same culture surface area. Thus, these results indicated that O₂ (+) culture was effective for not only hepatic maturation but also efficient generation of iHeps in per unit surface area. To further investigate this high-density culture, we prepared vertical cross-sections stained with hematoxylin and eosin (H&E) (Fig. 7B). Photographs of the H&E staining of vertical sections revealed that cells cultured under O₂ (+) conditions showed a self-organized, thick, multilayered configuration. In contrast, cells cultured under O₂ (−) conditions formed thin monolayers. Moreover, among the O₂ (+) culture group, the low atmospheric oxygen condition (5% O₂) showed the highest DNA content and the thickest multilayered formation.

FIG. 7.

(A) DNA content of cultures on day 20. Data represent the mean ± SD (n = 3). (B) Photographs of hematoxylin and eosin staining of vertical sections on day 20. Scale bar represents 50 μm.

Discussion

To use iHeps for engineering in vitro liver models, further maturation of iHeps is essential. The improvement of the culture environment is one of the keys to the hepatic differentiation of hiPSCs. While many approaches have focused on the composition of the culture medium^10–13 or co-culture with nonparenchymal cells,^1,15 few reports have focused on oxygen conditions of the culture system during hepatic differentiation, which is an important factor for human development¹⁹ and hepatocyte culture.³⁰ Although Si-Tayeb et al. applied different oxygen pressures during differentiation, the advantages of the different oxygen concentrations were not discussed.^10,20 In this study, we used PDMS membranes as the bottom of the culture plates to supply oxygen directly to the cell layer, and we evaluated the effect of oxygen supply on the hepatic differentiation of hiPSCs. When compared to cells differentiating in O₂ (−) culture, we observed higher expression of GATA4 in endoderm cells differentiating in O₂ (+) culture, as well as significant improvements in the expression of several mature hepatic marker genes (ALB, CYP1A2, and CYP3A4) and ALB production in iHeps generated in O₂ (+) culture with physiological oxygen concentration (5% or 10% O₂). Moreover, oxygen flux measurements revealed that iHeps generated in O₂ (+) culture exhibited a higher consumption of oxygen. Under such sufficient oxygen supply conditions, cells spontaneously formed multilayered structures, and thus, the efficient generation of iHeps was achieved.

Oxygen plays an essential role in embryo development and liver organogenesis.²⁰ In vivo, oxygen is supplied by blood flow. During liver organogenesis, early liver development occurs under low oxygen supply conditions because the blood vessels are not yet developed.¹⁹ In contrast, as vascular formation occurs along with liver bud formation,³⁰ in the latter stage of liver organogenesis,³¹ the amount of oxygen supplied to the cell markedly increases and cells utilize more oxygen. In our study, oxygen supply to the cell layer was improved by using an oxygen-permeable membrane as the bottom of culture plate. Oxygen supply to the cell layer is therefore determined by the concentration gradient between the atmospheric and pericellular oxygen concentrations and can be expressed using Fick's Law.²⁴ While in the conventional TCPS plate, oxygen supply to the cell layer depends only on the diffusion of oxygen through the layer of culture medium, in the PDMS plate, it is supplied through both the culture medium and the PDMS membrane. Based on the results of the oxygen concentration measurements, we observed similar oxygen concentrations at the cell layers in the O₂ (−)-20% (10.77%) and O₂ (+)-10% (9.08%) cultures and in the O₂ (−)-10% (4.81%) and O₂ (+)-5% (4.24%) cultures. However, the supplied oxygen flux to the cell in O₂ (+) was about two to four times higher than that in O₂ (−) culture. A comparison of the permeability coefficient of oxygen through both the culture medium and PDMS membrane revealed that PDMS has a permeability coefficient that is about seven times higher (P_PDMS: 1.72 × 10⁻¹ pmol·cm/s·cm²·mmHg; P_medium (D_medium × K_medium): 2.38 × 10⁻² pmol·cm/s·cm²·mmHg). Thus, PDMS contributes considerably to improving the oxygen supply and cellular oxygen consumption. Moreover, considering that the oxygen concentration in the liver roble is between 4% and 11%,³² this implies that the improved maturation observed in O₂ (+) culture with a 5% or 10% oxygen concentration occurred through the replication of both an in vivo-like oxygen concentration and an in vivo-like oxygen supply.

In most hepatic differentiation protocol, many expensive cytokines were added to induce differentiation. Thus, the cost-effective differentiation methods would be attractive.³³ In sufficient oxygen-supplied condition using PDMS membrane, we revealed cells spontaneously formed multilayered configuration. Especially, in O₂ (+)-5% or O₂ (+)-10% condition, we achieved about 2.5 times higher DNA content than that in O₂ (−)-20% condition. Moreover, we observed about three times higher ratio of ALB-positive cells in these conditions than that in O₂ (−)-20% condition. Based on these results, we achieved about 7.5 times higher per unit surface area productivity of iHeps. Thus, O₂ (+) culture under physiological oxygen concentration was effective for efficient generation of iHeps. Furthermore, it is suggested the formation of multilayered configuration is also effective on hepatic maturation. Many researchers show the importance of 3D culture compared to 2D culture^16,17 for hepatic differentiation of hiPSCs, especially, the regeneration of cell–cell junction is critical. Although cells formed a thin monolayer in the O₂ (−) condition, cells spontaneously formed 3D configuration in O₂ (+) culture. Thus, in addition to the replication of in vivo-like oxygen concentration and oxygen supply, replication of in vivo-like cell–cell interaction by direct oxygenation is also effective for hepatic differentiation of hiPSCs.

Our findings also revealed that direct oxygenation through PDMS is effective for endodermal differentiation, even though vascularization has not yet occurred during early liver organogenesis. Although one advantage of direct oxygenation through the PDMS membrane is enabling cells to utilize sufficient oxygen for respiration, in this study, we found that endodermal cells differentiating under both O₂ (−) and O₂ (+) culture conditions performed anaerobic respiration (Fig. 3). However, we also found that HIF-1α, which is highly expressed under hypoxic conditions, was downregulated by PDMS culture. Several reports have revealed that Wnt signaling plays an important role during early liver organogenesis,^34–37 and thus, some hepatic differentiation protocols add Wnt3A cytokines to the culture medium to promote endoderm differentiation.^12,38 The relationship between Wnt signaling and HIF-1α expression is widely known, and Wnt signaling is activated by the inhibition of HIF-1α gene expression.³⁹ This suggests that the inhibition of HIF-1α using the PDMS membrane contributes to the effective differentiation of hiPSCs toward endodermal cells.

Although iHeps differentiating under O₂ (+) culture conditions with physiological oxygen concentrations showed significantly higher expression of several mature hepatic markers and increased the production of ALB compared to cells differentiating under O₂ (−) conditions, they still exhibited some fetal hepatocyte phenotypes, such as the high expression of AFP, whose expression is downregulated in PHHs, or the quite low metabolic activity of CYP3A4. Several promising approaches have been reported for the further maturation of iHeps, including co-culture with nonparenchymal cells¹⁵ and 3D culture,^16–18 in addition to the optimization of soluble factors added to the culture medium during differentiation protocols. However, these approaches have not yet resulted in fully mature iHeps. In these studies, oxygen supply to the cell layer has not been considered. As a synergistic approach of co-cultures and oxygen supply improvements, Xiao et al. reported that co-culture of rat hepatocytes with nonparenchymal cells under PDMS membrane-based direct oxygenation enabled in vivo-like hierarchical double-layered microstructure on the PDMS membranes and maintained higher levels of hepatic function for prolonged time.²⁵ In the near future, through integration of various advanced culture methods such as 3D culture or co-culture with the improved oxygenation as evidenced in this study, better physiologically relevant in vitro liver tissue models with prolonged duration can be developed for efficacy/toxicity screenings of drugs or liver disease models. Therefore, co-cultures with various nonparenchymal cells with sufficient oxygen supply through PDMS membranes are expected to enhance further maturation of iHeps in vitro.

In summary, we have shown sufficient oxygen supply to the cell layer through the use of a PDMS membrane to be effective in the hepatic maturation of iHeps. We demonstrated the effect of direct oxygenation through the PDMS membrane by performing hepatic differentiation of hiPSCs under several different oxygen concentrations with varying methods of supplying oxygen. The combination of the PDMS membrane and a physiological oxygen concentration contributes to an improvement in cellular oxygen consumption. Hepatic differentiation under a sufficient oxygen supply can promote the further maturation of iHeps. Moreover, direct oxygenation through the PDMS membrane also promotes the self-organization of a 3D configuration during hepatic differentiation, leading to an increase in per area-based productivity. Thus, direct oxygenation through a PDMS membrane, rather than a conventional TCPS plate, is a promising fundamental culture method for the maturation of iHeps.

Footnotes

Acknowledgments

This work was supported by the Japan Agency for Medical Research and Development (AMED, “Development of Cell Production and Processing Systems for Commercialization of Regenerative Medicine”). We thank Vecell, Inc. for supplying the permeability coefficient of the PDMS membrane. K.K. is grateful for the financial support from SEUT (Graduate School of Engineering, The University of Tokyo, Doctoral Student Special Incentives Program).

Disclosure Statement

No competing financial interests exist.

References

Berger

D.R.

, Ware

B.R.

, Davidson

M.D.

, Allsup

S.R.

, and Khetani

S.R.

Enhancing the functional maturity of induced pluripotent stem cell-derived human hepatocytes by controlled presentation of cell–cell interactions in vitro. Hepatology, 61, 1370, 2015.

Lin

J.H.

CYP induction-mediated drug interactions: in vitro assessment and clinical implications. Pharm Res, 23, 1089, 2006.

Hewitt

N.J.

, Lecluyse

E.L.

, and Ferguson

S.S.

Induction of hepatic cytochrome P450 enzymes: methods, mechanisms, recommendations, and in vitro-in vivo correlations. Xenobiotica, 37, 1196, 2007.

Godoy

, Hewitt

N.J.

, Albrecht

, et al. Recent advances in 2D and 3D in vitro systems using primary hepatocytes, alternative hepatocyte sources and non-parenchymal liver cells and their use in investigating mechanisms of hepatotoxicity, cell signaling and ADME. Arch Toxicol, 87, 1315, 2013.

LeCluyse

E.L.

, Witek

R.P.

, Andersen

M.E.

, and Powers

M.J.

Organotypic liver culture models: meeting current challenges in toxicity testing. Crit Rev Toxicol, 42, 501, 2012.

Bale

S.S.

, Vernetti

, Senutovitch

, et al. In vitro platforms for evaluating liver toxicity. Exp Biol Med (Maywood), 239, 1180, 2014.

Sauer

, Roy-Chowdhury

, Guha

, and Roy-Chowdhury

Induced pluripotent stem cells as a source of hepatocytes. Curr Pathobiol Rep, 2, 11, 2014.

Ebert

A.D.

, Liang

, and Wu

J.C.

Induced pluripotent stem cells as a disease modeling and drug screening platform. J Cardiovasc Pharmacol, 60, 408, 2012.

Schwartz

R.E.

, Fleming

H.E.

, Khetani

S.R.

, and Bhatia

S.N.

Pluripotent stem cell-derived hepatocyte-like cells. Biotechnol Adv, 32, 504, 2014.

10.

Si-Tayeb

, Noto

F.K.

, Nagaoka

, et al. Highly efficient generation of human hepatocyte-like cells from induced pluripotent stem cells. Hepatology, 51, 297, 2010.

11.

Song

, Cai

, Liu

, et al. Efficient generation of hepatocyte-like cells from human induced pluripotent stem cells. Cell Res, 19, 1233, 2009.

12.

Chen

Y.F.

, Tseng

C.Y.

, Wang

H.W.

, Kuo

H.C.

, Yang

V.W.

, and Lee

O.K.

Rapid generation of mature hepatocyte-like cells from human induced pluripotent stem cells by an efficient three-step protocol. Hepatology, 55, 1193, 2012.

13.

Takayama

, Inamura

, Kawabata

, et al. Efficient generation of functional hepatocytes from human embryonic stem cells and induced pluripotent stem cells by HNF4α transduction. Mol Ther, 20, 127, 2012.

14.

Nakamura

, Enosawa

, and Umezawa

Parameters of stem cell-derived functional hepatocytes. Organ Biol, 21, 33, 2014.

15.

Takebe

, Sekine

, Enomura

, et al. Vascularized and functional human liver from an iPSC-derived organ bud transplant. Nature, 499, 481, 2013.

16.

Takayama

, Kawabata

, Nagamoto

, et al. 3D spheroid culture of hESC/hiPSC-derived hepatocyte-like cells for drug toxicity testing. Biomaterials, 34, 1781, 2013.

17.

Gieseck

R.L.

III , Hannan

N.R.F.

, Bort

et al., Maturation of induced pluripotent stem cell derived hepatocytes by 3D-culture. PloS One, 9, e86372, 2014.

18.

Nagamoto

, Tashiro

, Takayama

, et al. The promotion of hepatic maturation of human pluripotent stem cells in 3D co-culture using type I collagen and Swiss 3T3 cell sheets. Biomaterials, 33, 4526, 2012.

19.

Simon

M.C.

, and Keith

The role of oxygen availability in embryonic development and stem cell function. Nat Rev Mol Cell Biol, 9, 285, 2008.

20.

Behbahan

I.S.

, Duan

, Lam

, et al. New approaches in the differentiation of human embryonic stem cells and induced pluripotent stem cells toward hepatocytes. Stem Cell Rev, 7, 748, 2011.

21.

Hakim

, Kaitsuka

, Raeed

J.M.

, et al. High oxygen condition facilitates the differentiation of mouse and human pluripotent stem cells into pancreatic progenitors and insulin-producing cells. J Biol Chem, 289, 9623, 2014.

22.

Stevens

K.M.

Oxygen requirements for liver cells in vitro. Nature, 206, 199, 1965.

23.

Evenou

, Fujii

, and Sakai

Spontaneous formation of highly functional three-dimensional multilayer from human hepatoma Hep G2 cells cultured on an oxygen-permeable polydimethylsiloxane membrane. Tissue Eng, 16, 311, 2010.

24.

Xiao

, Shinohara

, Komori

, Sakai

, Matsui

, and Osada

The importance of physiological oxygen concentrations in the sandwich cultures of rat hepatocytes on gas-permeable membranes. Biotechnol Prog, 30, 1401, 2014.

25.

Xiao

, Perry

, Komori

, and Sakai

New physiologically-relevant liver tissue model based on hierarchically cocultured primary rat hepatocytes with liver endothelial cells. Integr Biol, 7, 1412, 2015.

26.

Hamon

, Hanada

, Fujii

, and Sakai

Direct oxygen supply with polydimethylsiloxane (PDMS) membranes induces a spontaneous organization of thick heterogeneous liver tissues from rat fetal liver cells in vitro. Cell Transplant, 21, 401, 2012.

27.

Kido

, Koui

, Suzuki

, et al. CPM is a useful cell surface marker to isolate expandable bi-potential liver progenitor cells derived from human iPS cells. Stem Cell Rep, 5, 508, 2015.

28.

Horiguchi

, Chowdhury

M.M.

, Sakai

, and Tabata

Proliferation, morphology, and pluripotency of mouse induced pluripotent stem cells in three different types of alginate beads for mass production. Biotechnol Prog, 30, 896, 2014.

29.

Bruick

R.K.

Oxygen sensing in the hypoxic response pathway: regulation of the hypoxia-inducible transcription factor. Genes Dev, 17, 2614, 2003.

30.

Nahmias

, Berthiaume

, and Yarmush

M.L.

Integration of technologies for hepatic tissue engineering. Adv Biochem Eng Biotechnol, 103, 309, 2007.

31.

Collardeau-Frachon

, and Scoazec

J.Y.

Vascular development and differentiation during human liver organogenesis. Anat Rec, 291, 614, 2008.

32.

Jungermann

, and Kietzmann

Role of oxygen in the zonation of carbohydrate metabolism and gene expression in liver. Kidney Int, 51, 402, 1997.

33.

Tasnim

, Phan

, Toh

Y.C.

, and Yu

Cost-effective differentiation of hepatocyte-like cells from human pluripotent stem cells using small molecules. Biomaterials, 70, 115, 2015.

34.

Hay

D.C.

, Fletcher

, Payne

, et al. Highly efficient differentiation of hESCs to functional hepatic endoderm requires ActivinA and Wnt3a signaling. Proc Natl Acad Sci U S A, 105, 12301, 2008.

35.

Nostro

M.C.

, Sarangi

, Ogawa

, et al. Stage-specific signaling through TGFβ family members and WNT regulates patterning and pancreatic specification of human pluripotent stem cells. Development, 138, 861, 2011.

36.

Davidson

K.C.

, Adams

A.M.

, Goodson

J.M.

, et al. Wnt/β-catenin signaling promotes differentiation, not self-renewal, of human embryonic stem cells and is repressed by Oct4. Proc Natl Acad Sci U S A, 109, 4485, 2012.

37.

Nostro

M.C.

, Cheng

, Keller

G.M.

, and Gadue

Wnt, activin, and BMP signaling regulate distinct stages in the developmental pathway from embryonic stem cells to blood. Cell Stem Cell, 2, 60, 2008.

38.

Nakamura

, Saeki

, Mitsumoto

, et al. Feeder-free and serum-free production of hepatocytes, cholangiocytes, and their proliferating progenitors from human pluripotent stem cells: application to liver-specific functional and cytotoxic assays. Cell Reprogram, 14, 171, 2012.

39.

Kaidi

, Williams

A.C.

, and Paraskeva

Interaction between β-catenin and HIF-1 promotes cellular adaptation to hypoxia. Nat Cell Biol, 9, 210, 2007.