Liver disease causes ~2 million deaths annually worldwide. Cirrhosis-related complications account for 50% of deaths, and viral hepatitis and hepatocellular carcinoma (HCC) together account for the other 50% of deaths annually worldwide. Cirrhosis is currently the 11th most common cause of death globally, and liver cancer is the 16th leading cause of death. Cirrhosis and HCC together account for 3.5% of all deaths worldwide (1). Currently, liver transplantation is the most effective treatment option for patients with end-stage liver disease (ESLD). However, <10% of global liver transplantation needs are met (2). The shortage of donor organs and transplant costs are the major limiting factors (3). In addition, the recipient's immune system may reject the transplanted organ (4). Therefore, the development of alternative therapeutic strategies for patients with chronic liver disease is of utmost importance.

Human induced pluripotent stem cells (hiPSCs) are pluripotent stem cells that are induced and reprogrammed from adult cells into undifferentiated cells by exposure to specific mediators. They have the capacity for self-renewal and differentiation into a variety of somatic cells, such as embryonic stem cells (5). HiPSC-derived hepatocyte-like cells (hiPSC-HLCs) exhibit morphological and phenotypic characteristics of primary human hepatocytes (PHHs) (6). hiPSC-HLCs should be incorporated into future studies of liver diseases to solve the problems of organ shortage and prevent recipient immunorejection of PHHs. hiPSC-HLCs can be considered as an ideal source of hepatocytes. In addition, these cells may enable cell-based therapy, exploration of liver disease models, in vitro models for pharmacology and toxicology studies (7) and ESLD treatment.

In this review article, we aimed to provide a perspective on the current organization and provision of transplant services based on specific challenges and environmental settings. We review how these specific requirements are addressed by the existing cell culture systems for the generation of hiPSC-derived hepatocytes, describe their applications for modeling hepatic disorders, and discuss future directions for the use of hiPSCs in the study and treatment of liver diseases.

Protocols for somatic cells transformed into induced pluripotent stem cells are widely used for iPSC reprogramming method. The key to the success of these protocols is the ability to efficiently induce pluripotent cells to adopt a definitive endoderm fate. Earlier studies have shown that mouse fibroblasts can be reprogrammed to develop embryonic stem cell (ESC)-like features, and grow in the presence of four factors: Oct3/4, Sox2, c-Myc, and Klf4 (8). These induced cells were named iPSCs. These insights opened the door to further refinements in the cell differentiation methodology and enabled the derivation of visceral, endodermal-derived tissues. The same protocol was used to successfully generate iPSCs from adult human skin cells (9). In addition, human somatic cells were transformed into PSCs by inducing the expression of a new set of four factors: Oct4, Sox2, Nanog, and Lin28 (10). Many optimization steps have been devised to improve the efficiency of reprogramming factors, such as the use of integration-free methods (8, 11). Episomal vectors, Sendai viruses, and synthetic mRNAs are among the most commonly used methods for generating hiPSCs (12–14) without modifying the host genome, which could interfere with disease modeling or experimental outcomes.

In 2009, hiPSCs were induced to differentiate into hepatic cells for the first time, via a timed administration of various growth factors. The expression of hepatocyte markers and liver-related functions in hiPSC-HLCs was monitored and compared with those in differentiated human ESCs and PHHs (15). This revealed that hepatic cells could be generated from iPSCs but the process took more than 20 d. To overcome this, the differentiation step was revised to a more efficient three-step protocol, allowing rapid generation of HLCs from hiPSCs (Figure 1) (16). Hepatic progenitor-like cells derived from hiPSCs possess the potential for bipotent differentiation into HLCs and cholangiocyte-like cells (17).

For clinical applications, differentiated cells should be assessed by comparing them with primary liver-derived cells, to verify their morphology and expression of liver-specific proteins, such as alpha-fetoprotein (AFP) and albumin. However, neither ESCs nor iPSCs can differentiate into fully mature hepatocytes in vitro. As a consequence of an emerging interest in using iPSCs in regenerative medicine to treat liver diseases, many researchers are focused on developing the most efficient and reproducible approaches for the derivation of high-quality hiPSC-HLCs. Currently, the differentiation strategies for obtaining hiPSC-HLCs generally require a stepwise induction of definitive endoderm, hepatocyte specification, and hepatoblast expansion into mature HLCs. Activin A, Wnt3a (18)and fibroblast growth factor 2 (FGF) signaling (19) play important roles in hiPSC differentiation toward hepatic endoderm, whereas hepatocyte growth factor (HGF) promotes the growth of hepatoblast cells (20, 21). Another factor, the interleukin (IL) 6 family cytokine oncostatin M (OSM), combined with the glucocorticoid dexamethasone (DEX), accelerates the maturation of hepatocytes. Meanwhile, many protocols for iPSC differentiation to hepatocytes have been reported, which involve the use of various growth factors and cytokines, plating techniques, and transduction of key liver-specific transcription factors (22).

Other studies have reported strategies for the generation of cholangiocyte-like cells from hiPSCs. The first differentiation of cholangiocyte-like cells from hiPSC hepatoblasts was induced in the presence of growth hormone, epidermal growth factor (EGF), IL-6, and sodium taurocholate (23). Subsequently, efficient differentiation of cholangiocytes from iPSCs was improved by 3D co-culture of hepatoblasts and OP9 stromal cells in the presence of HGF, EGF, and TGF-β (24), and by stimulating cholangiocyte progenitor specification using FGF10, activin A, and retinoic acid (25). Yet another stepwise cholangiocyte differentiation approach involves a definitive endoderm–hepatic specification–hepatic progenitor–cholangiocyte procedure, with Jagged1 and TGF-β supplementation being key for the promotion of iPSC-cholangiocyte formation (26). These cells were subsequently characterized in vitro and in vivo. Induced-cholangiocytes show mature markers, such as SOX9, CK7, CK19, CFTR, AE2, ASBT, AQP1, and SCTR and they are negative for the hepatocyte marker HNF4a.

The liver performs a wide range of fundamental functions, including metabolic, nutrient storage, and detoxification functions. Hence, each protocol for an efficient maturation of iPSCs into HLCs needs to incorporate a thorough and critical evaluation of the hepato-specific transcriptome and enzymatic activities of the obtained cells. Drug metabolizing capacity, urea cycle activity, or bile acid and lipoprotein synthesis and excretion need to be evaluated to ascertain HLC maturation (Figure 2). By contrast, most of the current reports on HLC generation show that HLCs express genes at levels that are characteristic for the fetal liver and display a fetal-like phenotype. According to one study (26), HLC mitochondria display specific morphological changes, such as elongation, swollen cristae, dense matrix, and cytoplasmic migration, with an increased expression of mitochondrial DNA transcription and replication-related genes, and increased oxygen consumption. Following differentiation, HLCs express liver-specific proteins, including albumin and hepatocyte nuclear factor 4 alpha, and show intrinsic hepatocyte functions, including CYP450 activity. However, HLCs also express high levels of AFP, suggesting a persistent immature phenotype or inability to turn off early-stage genes. Furthermore, albumin production, urea production, CYP450 activity, and mitochondrial function of HLCs are significantly lower than those of primary human hepatocytes. Functional indicators, urea synthesis, glycogen synthesis, lipid storage, indocyanine green intake, and low-density lipoprotein intake, are also used to evaluate the function of HLCs (27–31) (Table 1). Liver function after HLC transplantation in disease-specific models in mouse is assessed using glycogen content synthesis, low-density lipoprotein, and indocyanine green intake (32).

Hepatocyte isolation is not easy to perform, and requires expertise and access to primary human liver tissue. An alternative strategy for the acquisition of hepatocytes may grant the researchers access to important control material, thanks to commercially available cryopreserved hepatocytes, or as an alternative to using human liver tissue. Recently, an important and detailed report describing the expression of more than 60 hepatic, pluripotency, and developmental genes has been published to map their changes during liver cell isolation and stem cell maturation into HLC (33). In this report, the specific traits of iPSC-HLC were compared with those of adult liver and fetal liver cells, proving that several hepatic enzymes, whose expression is limited during the fetal period and absent at the postnatal stage, are highly expressed in iPSC-HLCs. In the study, CYP3A4 and CYP3A7 genes were used for a convenient and quick evaluation of hepatic maturation.

At present, some promising stem cell-based transportation treatments for ESLD have been reported. Several studies have demonstrated the restoration of liver function in hematopoietic stem cells (HSCs), mesenchymal stem cells (MSCs), and endothelial progenitor cells (EPCs) in acute liver failure (34–37). Although alleviation of liver dysfunction was observed in patients who have received these treatments in clinical trials, the safety and short-term efficacy of stem cell-based transplantation were not explored in detail and should be further evaluated in large-scale prospective cohort studies (38). In fact, stem cell transplantation as an alternative treatment for ESLD cannot be used for remodeling injured hepatic structure. In theory, iPSC-HLCs can be regarded as the optimal cell replacement treatment for acute liver failure (ALF) and ESLD. Meanwhile, HLCs derived from patient-specific iPSCs are an unlimited source of hepatocytes to treat liver failure.

To date, the application of iPSCs in ALF and liver failure has been tested in animal models. Indeed, iPSC-HLC transplantation improves the general condition in a mouse model of CCl₄-induced liver injury (39). Further, the survival rate of immunodeficient mice with ALF significantly increased and liver fibrosis levels in mice with chronic liver injury significantly decreased after hiPSC-HLC transplantation (40). According to another study, hiPSC-HLCs rescue drug-induced ALF in rodents, and hiPSC-HLCs have functional and proliferative potential for liver regeneration after transplantation in an ALF model (41, 42). These studies indicated the potential of these cells in regenerative medicine for future clinical applications. Of note, the aforementioned studies only focused on rodent models rather than primates. To date, no studies have illustrated the application of iPSC-HLCs in human liver failure.

The first transplantation of hiPSCs in regenerative medicine took place in 2014, and was performed to treat a 77-year-old female patient with polypoidal choroidal vasculopathy in both eyes. Autologous iPSCs were generated from the patient's skin fibroblasts, differentiated into retinal pigment epithelium, and then transplanted into the eye. Immune response to autologous transplantation was observed 18 months postoperatively, and the results indicated no complications. The retinal pigment epithelial sheet that had been transplanted survived well, and the corrected visual acuity of the treated eye did not improve or worsen.

Unfortunately, not all cases are as successful as the autologous transplant discussed above. Another clinical trial involving three patients with age-related macular degeneration involved intravitreal administration of stem cells derived from autologous adipose tissue. Post-treatment observation revealed vision loss associated with ocular hypertension, hemorrhagic retinopathy, vitreous hemorrhage, combined traction and rhegmatogenous retinal detachment, or lens dislocation (43).

The above trials demonstrate the need for further investigation of stem cell therapy and, more specifically, the use of hiPSCs in regenerative medicine to treat ESLD.

Another important application of iPSCs is assessing the therapeutic effect and toxicity of drugs, as has been reported in many studies (93, 94). The effect of drugs is influenced by the genetic background and other complex factors. It is easier to screen drugs using iPSCs than via pharmacological testing of animals. Consequently, organoids, as an innovative technique for drug screening and toxicity assessment, have been extensively researched. However, iPSC-derived organoids in 2D culture cannot be used for drug screening because iPSCs in such culture receive similar stimuli as those in monolayer and cannot be used to model the physiological microenvironment. By contrast, 3D culture models the in vivo microenvironment and approximates the physiological conditions. Many studies have reported high-throughput screening of small molecule libraries for drug development and toxicity assessment for liver diseases using iPSC-hepatocytes (59, 95, 96). Generation of organoids in several liver cell co-cultures yields a more sensitive cellular model than that constructed using single cell type. Recently, Broutier et al. (97) developed organoids representing the tumor structure and reflecting the expression profile of hepatocellular carcinoma, cholangiocarcinoma, and hepatocellular cholangiocarcinoma. These organoids open up new opportunities for drug testing and personalized medicine, and can be used for the generation of tumor bio-banks to be used as screening platforms (98). Additionally, “liver-on-a-chip” as a platform for drug development and toxicology testing can be used in pharmacokinetic and pharmacodynamic studies. In one study, liver organoids were generated using hepatocytes and cholangiocytes on a perfusable microcapillary chip (99). These organoids were then used to test the dose- and time-dependent hepatotoxic effects of acetaminophen. Thus, “organ-on-a-chip” represents a novel and valid platform for drug testing.

While iPSCs have considerable applications in regenerative medicine, the genomic stability of these cells, such as the occurrence of genomic and epigenetic aberrations, copy number variation (CNV), and single-nucleotide polymorphisms (SNPs), is a matter of concern. Genomic aberrations in human PSCs (hPSCs) include abnormal karyotypes, such as recurrent trisomy of chromosomes 12, 17, or X, and aneuploidies of sub-chromosomal regions, such as duplications of the 12p, 17q, or 20q11.21 loci (100). Some studies have demonstrated that iPSCs, unlike somatic cells, contain regions of uniparental disomy (UPD), sharing the equivalent chromosomal and sub-chromosomal characteristics (101). These abnormalities may confer a selective advantage to certain genes during prolonged culturing.

In addition to genomic aberrations, other studies have focused on epigenetic aberrations, but the possibility of hiPSC application for complex diseases remains unknown (102). Epigenetic variations are often observed in different iPSC lines, and prolonged periods of culture (103) may affect disease modeling and clinical applications. For example, reactivating X-chromosome inactivation (XCI) and erosion of inactive X chromosome (Xi) silencing in female hiPSCs during iPSC reprogramming may trigger phenotypic and epigenetic changes in iPSCs, reducing their differentiation potential and increasing their tumorigenicity (104). Furthermore, local epigenetic variations in iPSCs, such as cellular memory and aberrant methylation loci, have been noted. Cellular memory leads to incomplete reprogramming, as DNA hypomethylation and histone modification at specific loci render iPSCs remain similar to the source cell. Further, iPSCs with cellular memory are susceptible to preferential differentiation into the cell type that they had been derived from (105–108). The iPSC-associated methylated loci contain certain imprinted loci and other genomic regions. Some alteration of genomic imprinted loci occurs during cell reprogramming or long-term culture (109–112). For instance, aberrant silencing of the D1K1-Dio3 imprinted locus is functionally associated with the failure to generate iPSCs in mouse during reprogramming (113, 114). Methylome profiling has been used to detect differentially methylated regions (DMRs) in hESCs and hiPSCs (115). In the study, all hypermethylated CG DMRs in hiPSCs were recognized as reprogramming-induced aberrancies.

CNVs and SNVs can be introduced into iPSCs as they may already exist in source cell lines or be acquired during the reprogramming process. hiPSCs contain more CNVs than hESCs, source cells, and somatic cells (116). Therefore, it is possible that iPSCs are more susceptible to CNV generation during cell reprogramming. However, sequencing-based findings suggest the occurrence of few or no detectable de novo CNVs in iPSCs (117–121). By contrast, low-grade genetic mosaicism of CNVs in source cells was tracked to iPSC derivation. Hence, it is possible that low-grade genetic mosaicism of somatic cells is the major source of CNVs in iPSCs. In addition, several studies have discussed the potential relationship between SNVs and iPSC generation. Most studies revealed no specific functional enrichment of genes with SNVs in iPSCs (117, 120, 122, 123). Most iPSC-manifested SNVs appear to be randomly distributed in the genome and are functionally irrelevant to iPSC generation. Therefore, the safety of iPSCs for use in regenerative medicine still faces many challenges.

Another safety concern has been raised regarding immunogenicity and tumorigenicity of the iPSC technology. Both genetic and epigenetic instability arising from iPSC reprogramming increases the risk of immunogenicity and tumorigenicity in vivo during iPSC-derived hepatic cell-associated therapy. Theoretically, the recipients of autologous iPSC-HLC transplants should not reject the transplants (124). However, in a teratoma mouse model, immune rejection in recipients was observed after iPSC transplantation (125). Undifferentiated PSCs, which possess the privilege property of immune tolerance because of low MHC-I antigen expression and the absence of MCH-II antigen expression, are expected to be less immunogenic than iPSCs (126–129). However, abnormal epigenetic differences between iPSCs and PSCs could contribute to the expression of immunogenic antigens during iPSC differentiation (128). Further, iPSC tumorigenicity poses a challenge for the development of individualized iPSC-HLC therapy. Pluripotency acquired by somatic cells upon reprogramming methods can increase genomic instability on the chromosomal and sub-chromosomal levels, contributing to the risk of tumorigenic transformation (130). Therefore, before the clinical development of iPSCs, current reprogramming technologies need to be optimized to minimize the occurrence of immunogenicity and tumorigenicity (Figure 3).

Together with the burgeoning application of stem cell-based techniques, iPSC technology has been incorporated into new approaches such as -omics–related research, nuclear reprogramming, gene-editing technology, RNAi, tissue engineering, medical devices, high-throughput screens (HTS), and humanized chimeric animal models. iPSCs provide promising opportunities to study novel therapies for liver diseases using cell properties of self-renewal and differentiation for the generation of HLCs. Recent studies have demonstrated that iPSC-derived hepatocytes are applicable for in vitro studies of complex liver disorders, such as viral hepatitis, inherited metabolic disorders, non-alcoholic liver diseases, cirrhosis, and HCC (Figure 4). HLCs can also be used as cell therapy to repair and regenerate liver mass to treat liver disease and prolong patient survival. The genetic and molecular mechanisms underlying liver disorders are an emerging area of research. By studying iPSC-HLCs as a possible treatment option, researchers are able to gain valuable insight into in vitro disease modeling and personalized medicine.

Nevertheless, several limitations still exist that prevent the clinical application of iPSC technology. The protocols for reprogramming and differentiation of hiPSCs into HLCs should be optimized and standardized to increase the efficiency of inducibility and promote HLC maturation. The risk of potential tumorigenicity associated with genomic and epigenetic variations should also be assessed. In addition, to commercialize iPSCs for clinical applications, further investigation into the limitations and challenges of using iPSCs is needed. Very few scientific reports have tested HLC injection/transplantation in life-threatening models of congenital diseases. To the best of our knowledge, to date, no single study has been able to demonstrate correction of amino acid or neurotransmitter abnormalities by injecting iPSC-HLCs. Nonetheless, with rapid technological advances in stem cell therapy, hiPSCs are likely to become effective and safe treatment of liver diseases in the future.

LQ provided the concept of this manuscript and prepared the figures. LZ, KP, and XL drafted this manuscript and prepared the figures. SDWB, RN, SB, and YL revised the manuscript. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.