Directed differentiation of human induced pluripotent stem cells (iPSCs) into hepatocytes could provide an unlimited source of liver cells, and therefore holds great promise for regenerative medicine, disease modeling, drug screening and toxicology studies. Various methods have been established during the past decade to differentiate human iPSCs into hepatocyte-like cells (HLCs) using growth factors and/or small molecules. However, direct comparison of the differentiation efficiency and the quality of the final HLCs between different methods has rarely been reported. In the current study, two hepatocyte differentiation methods were devised, termed Method 1 and 2, through modifying existing well-known hepatocyte differentiation strategies, and the resultant cells were compared phenotypically and functionally at different stages of hepatocyte differentiation. Compared to Method 1, higher differentiation efficiency and reproducibility were observed in Method 2, which generated highly homogeneous functional HLCs at the end of the differentiation process. The cells exhibited morphology closely resembling primary human hepatocytes and expressed high levels of hepatic protein markers. More importantly, these HLCs demonstrated several essential characteristics of mature hepatocytes, including major serum protein (albumin, fibronectin and α-1 antitrypsin) secretion, urea release, glycogen storage and inducible cytochrome P450 activity. Further transcriptomic comparison of the HLCs derived from the two methods identified 1,481 differentially expressed genes (DEGs); 290 Gene Ontology terms in the biological process category were enriched by these genes, which were further categorized into 34 functional classes. Pathway analysis of the DEGs identified several signaling pathways closely involved in hepatocyte differentiation of pluripotent stem cells, including ‘signaling pathways regulating pluripotency of stem cells’, ‘Wnt signaling pathway’, ‘TGF-beta signaling pathway’ and ‘PI3K-Akt signaling pathway’. These results may provide a molecular basis for the differences observed between the two differentiation methods and suggest ways to further improve hepatocyte differentiation in order to obtain more mature HLCs for biomedical applications.
The advent of induced pluripotent stem cell (iPSC) technology provides a potentially limitless source of cells for biomedical research and industrial and clinical use (
Over the past decade, a variety of protocols have been established for the differentiation of human iPSCs into hepatocytes
In this study, two modified methods of hepatocyte differentiation based on existing well-known strategies were designed, and the resultant HLCs were compared phenotypically and functionally at different stages of the differentiation process. Furthermore, transcriptomic analysis was performed on the final HLCs to explore the molecular basis underlying the differences observed between the two differentiation methods.
Human iPSC cell OARSAi002-A was previously generated in our laboratory from the cord blood of a healthy non-Hispanic white male using self-replicative RNA reprogramming technology (
DE induction was performed using the STEMdiff Definitive Endoderm Kit from STEMCELL Technologies according to the manufacturer's protocol with small modifications. On the day before DE induction (day-1), culture of the OARSAi002-A iPSC line was dissociated into single cells using TrypLE Select Enzyme (1X). Collected cells were resuspended into Cellartis DEF-CS medium supplemented with GF-1, GF-2, and GF-3 and seeded onto a 6-well tissue culture plate pre-coated with LDEV-free, hESC-qualified Matrigel (Corning Inc.) at a cell density of 2.1x105 cells/cm2. The plate was incubated at 37˚C and supplied with 5% CO2, and the cells reached ~90-100% confluence by the following day. On day 0, cells were washed once with DMEM/F12 (Thermo Fisher Scientific, Inc.) and replaced with STEMdiff Endoderm basal medium with 1X supplement MR and 1X supplement CJ. Cells were then incubated with STEMdiff Endoderm basal medium with 1X supplement CJ on days 1-3, with medium replenished daily. On day 4, the protein expression of key DE markers, SOX17 and forkhead box protein A2 (FOXA2), were evaluated by immunocytochemistry staining on 4% paraformaldehyde (PFA)-fixed cells using SOX17 mouse anti-human monoclonal antibody (Abcam) and FOXA2 rabbit anti-human monoclonal antibody (Abcam), as described below.
Two approaches were employed for the hepatic specification and maturation using a combination of growth factors, cytokines and small molecules. To better compare the differentiation efficiency of the two approaches, the hepatic specification was initiated using the same source of DE cells derived above. On day 4 of DE differentiation, cells were washed with DPBS without Ca2+ and Mg2+ (Thermo Fisher Scientific, Inc.) and dissociated into single cells using StemPro Accutase Cell Dissociation Reagent (Thermo Fisher Scientific, Inc.). Cells were split into two 50-ml conical tubes before centrifugation at room temperature and 200 x g for 5 min. Collected cells were then resuspended into two different media for hepatic specification following Methods 1 and 2 (
The first differentiation method was modified based on the work by Jung
Briefly, DE cells in one of the 50-ml conical tubes were resuspended in RPMI Cell Differentiation Medium [including RPMI Basal Medium, 20 ng/ml BMP4 (R&D Systems) and 10 ng/ml FGF2 (Thermo Fisher Scientific, Inc.)] supplemented with 10 µM Y-27632 Rho kinase (ROCK) inhibitor (Tocris Bioscience) on day 4 and seeded onto Matrigel (Corning Inc.) pre-coated 12-well or 24-well plates (Thermo Fisher Scientific, Inc.), at a seeding density of 8x104 cells/cm2. The RPMI Basal Medium was composed of RPMI-1640 HEPES medium, 1X B-27 Supplement with insulin, 1X Minimum Essential Medium non-essential amino acids solution (NEAA) and 1X penicillin/streptomycin (PEST) (all from Thermo Fisher Scientific, Inc.). The plate was shaken gently back and forth, then left and right to ensure even distribution of cells over the surface before being placed in the incubator overnight. The cells were replenished with RPMI Cell Differentiation Medium without ROCK inhibitor the next day and cultured for 5 days with daily medium changes. On days 9-14, the hepatic progenitor cells were further cultured in RPMI Basal medium supplemented with 20 ng/ml HGF with daily medium changes. Starting from day 14, the culture medium was replaced with Hepatocyte Maintenance Medium [Lonza Hepatocyte Culture Medium (HCM) composed of 500 ml 1X HBM Basal Medium and 1X HCM SingleQuots Supplement Pack (both from Lonza Group, Ltd.), supplemented with 20 ng/ml OSM (R&D Systems)]. The SingleQuots Supplement Pack contained 0.5 ml transferrin, 0.5 ml ascorbic acid, 0.5 ml human epidermal growth factor (HEGF), 0.5 ml insulin, 0.5 ml hydrocortisone, 10.5 ml BSA (fatty acid-free) and 0.5 ml gentamicin-amphotericin-1000. All components in the SingleQuots Supplement Pack were added to the medium except for HEGF. The cells were further cultured for 7 days with daily medium changes. After day 21, Hepatocyte Maintenance Medium was replenished every 2-3 days for further maintenance of the differentiated HLCs.
The second differentiation method was modified based on the work by Basma
Briefly, DE cells in the second 50-ml conical tube were resuspended in Hepatic Specification Medium, composed of Hepatocyte Differentiation Basal Medium supplemented with 100 ng/ml HGF, 1% DMSO as well as 10 µM Y-27632 and seeded onto Matrigel pre-coated 12-well or 24-well plates, at a seeding density of 8x104 cells/cm2. The Hepatocyte Differentiation Basal Medium was composed of DMEM/F12, 10% KnockOut Serum Replacement (Thermo Fisher Scientific, Inc.), 1X NEAA, 1X PEST and 1X GlutaMAX Supplement (Thermo Fisher Scientific, Inc.). The plate was gently shaken as above to ensure even distribution of cells over the entire surface and then incubated overnight. The following day, the medium was replaced with fresh Hepatic Specification Medium without ROCK inhibitor, and the cells were cultured for a further 7 days with daily medium changes. On days 12-14, the resultant hepatic progenitors were further cultured in Hepatic Maturation Medium (Hepatocyte Differentiation Basal Medium supplemented with 100 nM DEX and 100 nM dihexa) for 3 days with a daily medium change. Starting from day 15, the cell culture medium was changed to Hepatocyte Maintenance Medium, as described in Method 1. The cells were continually cultured for 6 days with daily medium changes. After day 21, Hepatocyte Maintenance Medium was replaced every 2-3 days for further maintenance of the differentiated HLCs.
Cells generated at different stages of the differentiation process, such as DE cells and HLCs, were fixed with 4% PFA for 20 min at room temperature, rinsed with DPBS with Ca2+ and Mg2+, and permeabilized with 0.3% Triton X-100 in DPBS for 15 min. Cells were then incubated with Image-iT FX signal enhancer (Thermo Fisher Scientific, Inc.) for 40 min at room temperature, followed by incubation with primary antibodies, including FOXA2 rabbit anti-human monoclonal antibody (cat. no. ab108422; Abcam; 1:50), SOX17 mouse anti-human monoclonal antibody (cat. no. ab84990; Abcam; 1:50), hepatocyte nuclear factor 4 α (HNF4A) rabbit anti-human polyclonal antibody (cat. no. HPA004712; Sigma-Aldrich; Merck KGaA; 1:50), cytochrome P450 (CYP)3A4 mouse anti-human monoclonal antibody (cat. no. 67110; ProteinTech Group, Inc.; 1:50), α-fetoprotein (AFP) mouse anti-human monoclonal antibody (cat. no. MAB1368; R&D Systems; 1:20), albumin (ALB) mouse anti-human monoclonal antibody (cat. no. MAB1455; R&D Systems; 1:20), cytokeratin 18 (CK18) mouse anti-human monoclonal antibody (cat. no. MAB7619; R&D Systems; 1:20), and α-1-antitrypsin (A1AT) mouse anti-human monoclonal antibody (cat. no. MAB1268; R&D Systems; 1:20). Cells were incubated with primary antibodies overnight at 4˚C in a humidified environment. The following day, cells were washed three times with DPBS and incubated with the corresponding secondary antibodies conjugated to Alexa Fluor 488 or Alexa Fluor 555 fluorophores (Thermo Fisher Scientific, Inc.) for 2 h at room temperature. The cell nuclei were counterstained with Hoechst 33342 at room temperature for 15 min, and the cell images were captured using the EVOS FL imaging system (Thermo Fisher Scientific, Inc.) with a 20X objective lens and RFP, GFP and DAPI filters. The expression rates of marker proteins and the mean fluorescence intensities of marker protein expression in the cells were quantified using ImageJ version 1.41 (National Institutes of Health) (
HLCs cultured in 12-well plates were incubated in 1 ml/well of Hepatocyte Maintenance Medium for 24 h; the cell culture supernatant was collected and stored at -80˚C until required. Wells without cells were included as blank controls. ELISA kits (all from Abcam) were used to determine the concentrations of ALB (cat. no. ab108788), fibronectin (cat. no. ab229398), and A1AT (cat. no. ab229417) in the cell culture supernatant following the manufacturer's instructions. Results were normalized to protein quantity in each well and presented as the mean ± standard deviation of four independent experiments.
Cells were lysed in cold Pierce RIPA lysis buffer (Thermo Fisher Scientific, Inc.) supplemented with 1X Halt Protease Inhibitor Cocktail (Thermo Fisher Scientific, Inc.) for 5 min. The supernatant was collected after centrifugation at room temperature and 1,600 x g for 10 min. Total protein was quantified using a Pierce BCA Protein Assay kit (Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol.
HLCs cultured in 24-well plates were treated with different concentrations of NH4Cl (0, 2 or 5 mM) in 1 ml Hepatocyte Maintenance Medium for 24 h. The cell culture supernatant was collected and analyzed for urea levels using a Urea Assay kit (cat. no. 83362; Abcam) according to the manufacturer's protocol. Wells without cells were included as blank controls. Results were normalized to protein quantities and presented as the mean ± standard deviation of three independent experiments.
Differentiated HLCs were fixed with 4% PFA as above and stained with a PAS staining kit (Sigma-Aldrich; Merck KGaA) to visualize glycogen storage following the manufacturer's instructions. Phase-contrast images were taken using a BZ-X810 microscope from Keyence Corporation with a 20X objective lens.
CYP metabolic activity of HLCs was analyzed following an adapted version of a previously described protocol (
Cells were harvested on days 17 and 21 of hepatocyte differentiation from Methods 1 and 2 and stored at -80˚C before RNA extraction. Cells were lysed in RLT buffer (Qiagen GmbH) and homogenized using QIAshredder (Qiagen GmbH). Total RNA was extracted from the cell lysates using an EZ1 RNA Cell Mini Kit (Qiagen GmbH) on EZ1 Advanced XL automated RNA purification instrument (Qiagen GmbH) following the manufacturer's instruction. An on-column DNase digestion step was included to remove potential DNA contamination. The total RNA and purity (260/280 nm) were subsequently measured using a NanoDrop 2000 UV-Vis spectrophotometer (NanoDrop Products). The integrity of RNA samples was further assessed using the Agilent 2100 Bioanalyzer with the RNA 6000 Nano Reagent Kit (Agilent Technologies) to obtain the RNA integrity number.
Global gene expression profiling of HLCs derived from differentiation Methods 1 and 2 was conducted using GeneChip PrimeView Human Gene Expression Arrays (Affymetrix; Thermo Fisher Scientific, Inc.) as described previously (
The biotin-labeled cRNA was then purified, fragmented and hybridized to the arrays in the GeneChip Hybridization Oven 645 (Affymetrix; Thermo Fisher Scientific, Inc.). After hybridization, the array chips were stained and washed using the GeneChip Fluidics Station 450. The chips were then scanned using GeneChip Scanner 3000 7G and the scanned image (.DAT) files were further processed using GeneChip Command Console software version 4.0 (Affymetrix; Thermo Fisher Scientific, Inc.) to produce cell intensity (.CEL) files. All arrays were assessed for data quality using Expression Console software version 1.3 (Affymetrix; Thermo Fisher Scientific, Inc.) before further data analysis. The dataset has been deposited in Gene Expression Omnibus (GEO;
The robust multi-array average algorithm (
To identify differentially expressed genes (DEGs), statistical analysis between two experimental groups was conducted using Affymetrix Transcriptome Analysis Console software 4.0, based on ebayes ANOVA. For each comparison, the fold change of every annotated gene, together with their ANOVA P-value or false discovery rate, was used to identify DEGs with cutoff values indicated in the text.
The online tool Database for Annotation, Visualization, and Integrated Discovery (DAVID) (
Data are presented as the mean ± standard deviation. Statistical analysis of the phenotypical and functional data was performed using GraphPad Prism version 9.2.0 (GraphPad Software Inc.). For the serum protein secretion and CYP activity data, a Student's t-test was used to analyze the statistical significance of differences between two groups. For the urea synthesis and release data, one-way ANOVA was performed followed by a Bonferroni post hoc test.
To obtain a homogenous cell population of human iPSCs for HLC differentiation, the cells were cultured in the Cellartis DEF-CS Culture System. Unlike the traditional colony-forming culture system (e.g., culturing iPSCs using E8 medium or mTeSR medium on Matrigel-coated tissue culture plates) (
DE induction from iPSCs represents a crucial first step for subsequent efficient hepatic differentiation. To achieve highly efficient and consistent DE induction, DE induction was performed using the STEMdiff Definitive Endoderm Kit. After single-cell dissociation of iPSCs with TrypLE Select Enzyme, cells were plated on Matrigel-coated 6-well tissue culture plates with ROCK inhibitor. The cells formed a subconfluent monolayer the next day, thus making it possible for the growth factors to act homogeneously on the stem cell population. During the 4 days of DE induction, the cell morphology kept evolving, and a confluent monolayer of homogeneous DE cells with a spiky shape was obtained at the end of the induction without any aggregates or multicellular structures (
To determine a more efficient method for hepatic specification and hepatocyte maturation, and to obtain highly homogeneous functional HLCs, two modified strategies were designed and evaluated (
Immunocytochemistry staining of hepatic marker proteins was performed to assess the cell maturity of differentiated HLCs. Cells on day 21 from both differentiation methods demonstrated characteristic nuclear staining of HNF4A, and cytoplasm expression of AFP, ALB, CK18, A1AT and CYP3A4 (
Next, whether the HLCs differentiated from both Methods 1 and 2 displayed key hepatic functional characteristics was assessed. Serum protein secretion is one of the common indicators of hepatic maturation; therefore, the ability of HLCs to secrete ALB, fibronectin and A1AT was assessed by ELISA. As shown in
The urea cycle is a crucial detoxification pathway that converts ammonia into urea. Since urea is almost exclusively produced by the liver, ureogenic capacity is commonly used as an indicator of a differentiated hepatic phenotype. As demonstrated in
Another essential function of hepatocytes
Hepatocytes are capable of clearing xenobiotics via metabolism through CYP isoenzymes (
To obtain an unbiased assessment and comparison of the HLCs differentiated from Method 1 and 2, whole transcriptome gene expression analysis on cells at early maturation (day 17) and late maturation (day 21) stages of hepatocyte differentiation was performed and this was further compared with PHHs. All RNA samples passed the RNA quality analysis with an RNA integrity number ≥8.0. Before data analysis, all arrays in this study were assessed for data quality with all quality assessment metrics (including spike-in controls during target preparation and hybridization) within default boundaries.
PCA of the microarray data (
To determine whether extended culture (beyond day 17) could improve the maturation of the HLCs, gene expression in HLCs between day 21 and 17 for both methods were compared. The results demonstrated that only 173 genes exhibited altered expression between days 17 and 21 in Method 1, with 140 upregulated and 33 downregulated genes. In contrast, 942 genes exhibited altered expression between days 17 and 21 in Method 2, including 381 upregulated and 561 downregulated genes (
Further comparison of gene expression in HLCs on day 21 from both methods to PHHs revealed substantial differences between HLCs and PHHs. In total, 14,595 DEGs were identified between HLCs generated from Method 1 and PHHs. Similarly, 12,537 DEGs were found between HLCs generated from Method 2 and PHHs (
To further compare HLCs generated from the two differentiation methods, PCA was performed among the cell samples on day 21. As shown in
To elucidate cellular functions impacted by the DEGs between the two differentiation methods, the DEGs were annotated using DAVID to find GO terms overrepresented in the BP category. The 1,481 DEGs between the cell samples generated from the two methods on day 21 resulted in a total of 290 GO terms. Using the CateGOrizer online tool, these GO terms were further categorized into 34 functional classes within the pre-defined set of parent/ancestor GO terms.
KEGG pathway analysis was performed to explore the molecular basis of the differences between the two hepatocyte differentiation methods. Several KEGG pathways that are closely involved in hepatocyte differentiation of pluripotent stem cells were enriched by the 1,481 DEGs (
Through phenotypical, functional and transcriptomic comparison of the two modified hepatocyte differentiation methods, which were based on existing well-known strategies, a stepwise improved method was established (shown in
Single-cell culture of iPSCs in the Cellartis DEF-CS culture system served as a key prerequisite for subsequent homogeneous hepatocyte differentiation. Heterogeneity of iPSCs, in terms of their morphology, self-renewal ability, pluripotency, differentiation bias and other traits, represents one of the critical challenges facing the stem cell field (
DE induction from iPSCs represents a critical step for subsequent efficient hepatic differentiation (
Upon successful DE formation, cell differentiation was next directed towards hepatic specification through two modified methods. Although starting from the same source of DE cells, considerable differences were observed during the differentiation process and in the final differentiated HLCs between these two methods, including cell morphology, differentiation efficiency, homogeneity and functional characteristics. Unlike cells from Method 1, which only displayed relatively mild cell morphological changes and showed an immature hepatic phenotype at the end of the differentiation, cells undergoing the differentiation process of Method 2 demonstrated significant morphological changes toward a final cell morphology closely resembling that of PHHs. Early studies have shown that there is an inverse relationship between cell proliferation and differentiation. Precursor cells continue dividing before acquiring a fully differentiated state, while terminal differentiation usually coincides with proliferation arrest and exit from the division cycle (
The present study showed that DMSO, combined with a high concentration of HGF, was capable of guiding highly efficient and homogeneous hepatic specification. DMSO has been used in previous studies, either separately or with cytokines/growth factors, for multiple lineage differentiation, including hepatic differentiation (
Following hepatic specification, further treatment of the cells with DEX, dihexa and OSM boosted cell maturation into functional HLCs. The role of DEX in hepatic maturation has been well established (
It is interesting to note that HLCs generated by Method 2 displayed the conspicuous formation of intracytoplasmic lipid droplets compared to those from Method 1. It is well established that the liver plays an important role in lipid metabolism, and hepatocytes store large amounts of lipid in the form of droplets for energy storage (
Transcriptomic analysis allows for comprehensive assessment and unbiased comparison of biological samples. In addition to the phenotypical and functional studies described above, the HLCs generated from the two hepatocyte differentiation methods were further compared by whole genome gene expression profiling. HLCs generated from Method 2 were found to be slightly better than those generated from Method 1 in terms of the overall gene expression profile similarity to PHHs. However, regardless of the differentiation methods, vast differences existed between HLCs and PHHs, with >12,000 DEGs identified between HLCs and PHHs; very similar findings were reported by Godoy
With this in mind, the current study explored the molecular basis underlying the differences observed between the two differentiation methods. Transcriptomic comparison of the HLCs derived by the two methods on day 21 identified 1,481 DEGs. Functional analysis of the DEGs revealed that, in addition to ‘metabolism’, other functions such as ‘development’, ‘cell communication’, ‘signal transduction’ and ‘cell differentiation’ were associated with these DEGs, highlighting their involvement and influence on hepatocyte differentiation. Furthermore, pathway analysis of the DEGs identified several signaling pathways closely involved in hepatocyte differentiation of pluripotent stem cells. Among them, the ‘ECM-receptor interaction’ and ‘focal adhesion pathways’ play essential roles in cell proliferation, cell differentiation and cell survival, and are pivotal in tissue and organ morphogenesis (
In conclusion, based on the phenotypical, functional and transcriptomic comparison of the two modified hepatocyte differentiation methods, an improved method for efficient generation of homogeneous functional hepatocytes from iPSCs was developed. Compared with the HLCs derived from existing protocols evaluated in our group, the final differentiated HLCs derived from this enhanced method were highly homogeneous, both phenotypically and functionally. Combining the easy scale-up of the DEF-CS single-cell culture system for iPSCs, the highly efficient DE induction by the STEMdiff Definitive Endoderm kit, the robust hepatic specification guided by DMSO and HGF, and the hepatocyte maturation promoted by DEX, dihexa and OSM, this improved method may allow for robust generation of homogeneous functional hepatocytes from human stem cells and opens up new possibilities for drug discovery, hepatotoxicity studies and liver regenerative therapy. Moreover, the results of the transcriptomics analysis provide a molecular basis for the differences observed between the two differentiation methods, offer new insight into gene regulation in hepatogenesis
We would like to thank Dr Marianne Miliotis-Solomotis and Dr Mary E. Torrence (Office of Applied Research and Safety Assessment, Center for Food Safety and Applied Nutrition, U.S. Food and Drug Administration, Laurel, MD, USA) for their critical review of the manuscript and their support for this work.
The microarray dataset has been deposited in Gene Expression Omnibus (GEO;
XG and RL conceived and designed the study, performed the experiments, analyzed the data and wrote the manuscript. RL performed the hepatocyte differentiation and the functional assays. XG performed the transcriptomic analysis. YZ analyzed the CYP assay samples using LC/MS. JJY and RLS contributed to the conception of the study, critically reviewed the manuscript. All authors read and approved the final manuscript. XG and RL confirmed the authenticity of all the raw data.
Not applicable.
Not applicable.
The authors declare that they have no competing interests.
Schematic illustration of HLC differentiation from human iPSCs. (A) Cell states and sequential steps of the differentiation process. Differentiation timeline, media and growth factors used at each stage of (B) Differentiation Method 1 and (C) Differentiation Method 2. HCM, hepatocyte culture medium; iPSC, induced pluripotent stem cell; BMP4, bone morphogenetic protein 4; FGF2, fibroblast growth factor 2; HGF, hepatocyte growth factor; HEGF, human epidermal growth factor; OSM, oncostatin M; KOS, knockout serum replacement; DEX, dexamethasone.
Representative images showing sequential morphological changes during the process of HLC differentiation. (A) DE induction. Hepatic specification and HLC maturation of (B) Method 1 and (C) Method 2. Scale bar, 400 µm; inset scale bar, 100 µm. D, day; DE, definitive endoderm.
Immunocytochemistry staining showing the expression of characteristic stage-specific protein markers at major stages of HLC differentiation. (A) OCT4, SSEA4, TRA-1-60, and SOX2 for iPSCs. (B) SOX17 and FOXA2 for DE cells. HNF4A, AFP, ALB, CK18, A1AT, and CYP3A4 for HLCs derived from (C) Method 1 and (D) Method 2. Scale bar, 400 µm. (E) Quantification of fluorescence intensity of hepatic protein marker immunostaining. n=4, *P<0.05, **P<0.01, ***P<0.001 using a two tailed Student's t-test. HLC, hepatocyte-like cell; DE, definitive endoderm; iPSC, induced pluripotent stem cell.
Functional characterization of HLCs. (A) Secretion of serum proteins ALB, fibronectin and A1AT. (B) Urea synthesis and release upon incubation with different concentrations of NH4Cl. (C) Periodic acid-Schiff staining showing glycogen storage. Scale bar, 100 µm. (D) Basal activities of CYP1A, CYP2D6, CYP2B6, CYP3A, and CYP2C9. (E) Induction of CYP1A, CYP2D6, CYP2B6, CYP3A and CYP2C9, respectively, after treatment with OME or PB. For CYP activity, 1 unit (U)=1 pmol metabolite/mg protein/h, and the metabolites are acetaminophen, 1'-hydroxy bufuralol, hydroxy bupropion, 1'-hydroxy midazolam and 4'-hydroxy diclofenac for CYP1A, CYP2D6, CYP2B6, CYP3A, and CYP2C9, respectively. All values are presented as the mean ± standard deviation. n=4 for serum protein secretion and CYP activity, and n=3 for urea synthesis. *P<0.05, **P<0.01, ***P<0.001. HLC, hepatocyte-like cell; ALB, albumin; A1AT, α-1-antitrypsin; CYP, cytochrome P450; OME, omeprazole; PB, phenobarbital; HLC_M1, HLCs derived from method 1; HLC_M2, HLCs derived from method 2.
Global gene expression of HLCs derived from hepatocyte differentiation Method 1 (M1) and Method 2 (M2). (A) Principal component analysis using all probe sets on the array to cluster samples based on their similarities in global gene expression level. The two axes PC1 and PC2, represent the first two principal components identified by the analysis. The percentage contribution of each component to the overall source of variation is included in the parenthesis following the component name. The cell samples collected at different stages of hepatocyte differentiation from different differentiation methods are color-coded, as shown in the graph. HLCs were collected on day 17 (D17, round) and day 21 (D21, diamond) after differentiation by Method 1 (blue) and Method 2 (red). PHHs (green square) were also included for comparison. (B) Hierarchical cluster analysis based on all genes. Expression data are in log2 scale and color-coded as shown in the scheme in the bottom right corner. The dendrogram on the bottom shows clusters of genes, while the dendrogram on the right shows clusters of samples. (C) Volcano plots showing the DEGs between HLCs on D17 and 21 by differentiation Method 1 and differentiation Method 2. (D) Volcano plots showing the DEGs between PHHs and HLCs (D21) derived from differentiation Method 1 and differentiation Method 2. All DEGs are shown as red dots. Downregulated genes are on the top-left corner with numbers denoted by green arrow and upregulated genes are on the top-right corner denoted by red arrow. The total number of DEGs is included in the parenthesis under the title of the panel. HLC, hepatocyte-like cell; PC, principal component; PHHs, primary human hepatocytes; DEG, differentially expressed gene.
Comparison of HLCs on day 21 (D21) between differentiation Method 1 (M1) and Method 2 (M2). (A) Principal component analysis based on all probe sets. (B) Volcano plots showing the DEGs between HLCs (D21) from differentiation M1 and M2. All DEGs are shown as red dots. Downregulated genes are on the top-left corner with numbers denoted by green arrow, and upregulated genes are on the top-right corner denoted by red arrow. (C) Top 10 functional classes associated with the DEGs between two differentiation methods. HLC, hepatocyte-like cell; PC, principal component; DEG, differentially expressed gene.
Wnt signaling pathway. Differentially expressed genes identified between the two differentiation methods are highlighted with red stars.
List of KEGG pathways enriched by the DEGs between the two differentiation methods that are closely involved in hepatocyte differentiation of pluripotent stem cells.
KEGG pathway | Number of DEGs | Percentage | Fold enrichment | P-value |
---|---|---|---|---|
ECM-receptor interaction | 15 | 1.6 | 3.0 | <0.0001 |
TGF-beta signaling pathway | 13 | 1.4 | 2.7 | 0.003 |
Focal adhesion | 23 | 2.5 | 2.0 | 0.003 |
Cytokine-cytokine receptor interaction | 24 | 2.6 | 1.7 | 0.010 |
Hematopoietic cell lineage | 11 | 1.2 | 2.2 | 0.024 |
Signaling pathways regulating pluripotency of stem cells | 14 | 1.5 | 1.8 | 0.053 |
PI3K-Akt signaling pathway | 27 | 2.9 | 1.4 | 0.080 |
Wnt signaling pathway | 13 | 1.4 | 1.7 | 0.090 |
KEGG, Kyoto Encyclopedia of Genes and Genomes; DEG, differentially expressed genes; ECM extracellular matrix.