Hypothesis: Colony-forming activity of pluripotent stem cell-derived hepatocyte-like cells for stem cell assay

Hepatocyte-like cells (HLCs) generated from human pluripotent stem cells (PSCs) exhibit hepatocytic properties in vitro; however, their engraftment and functionality in vivo remain unsatisfactory. Despite optimization of differentiation protocols, HLCs did not engraft in a mouse model of liver injury. In contrast, organ-derived hepatocytes reproducibly formed colonies in the liver injury mouse model. As an extension of the phenomenon observed in hematopoietic stem cells giving rise to colonies within the spleen, commonly referred to as “colony-forming units in spleen (CFU-s“, we hypothesize that “colony-forming units in liver (CFU-L)“ serves as a reliable indicator of stemness, engraftment, and functionality of hepatocytes. The uniform expression of the randomly inactivated gene in a single colony, as reported by Sugahara et al. 2022, suggests that the colonies generated by isolated hepatocytes likely originate from a single cell. We, therefore, propose that CFU-L can be used to quantify the number of “hepatocytes that engraft and proliferate in vivo“ as a quantitative assay for stem cells that utilize colony-forming ability, similar to that observed in hematopoietic stem cells.

HSCs also display colony-forming ability in vitro (CFU-c: colony-forming units in culture) 9 . HSCs generate many cell lineages from a single cell, and this clonality is an important feature of their function because it allows precise control of the blood system. Mesenchymal stromal cells also exhibit multipotency in vitro and in vivo 10,11 . Likewise, we postulate that hepatocytes can be directly measured in vivo and a comparative analysis of HLCs derived from induced pluripotent stem cells (iPSCs) and organ-derived hepatocytes would enable the validation of the stemness, engraftability, and functionality of hepatocytes.
For instance, iPSC-derived hepatocytes have served as in vitro tools in drug metabolism [16][17][18][19] . In this study, we used iPSCs derived from patients with drug-induced liver injury (DILI) because the immortality of iPSCs allows repeated acquisition of hepatocytes from the same origin.
Chimeric mice with humanized livers have been developed, among which cDNA-uPA/SCID mouse model shows a high replacement rate of human hepatocytes. This chimeric model by using isolated hepatocytes has proven to be useful in human drug metabolism and pharmacokinetics [20][21][22][23][24] . To mimic normal human liver function, chimeric mice with humanized livers are usually generated by the transplantation of hepatocytes. Hepatocytes from patients with ornithine transcarbamylase deficiency (OTCD) form colonies in the liver of cDNA-uPA/SCID mice, and quite surprisingly, these colonies are presumably from a single hepatocyte 25 . We here hypothesize that such colony-forming activity in liver (CFU-L) is directly related to the validation of stemness, viability, and functionality of hepatocytes. This hypothesis is an extension of the idea that bone marrow stem cells form colonies in the spleen 7,8 .

Ethical statement
Human cells in this study were obtained in full compliance with the Ethical Guidelines for Clinical Studies (Ministry of Health, Labor, and Welfare, Japan). The cells were deposited to RIKEN Cell Bank. Animal experiments were performed according to protocols approved by the Institutional Animal Care and Use Committee of the National Research Institute for Child Health and Development.

Cells
Cells were obtained from a patient with DILI (Hep(C)). The cells were maintained in Dulbecco's modified Eagle's medium (DMEM, SIGMA D6429) supplemented with 20% FBS at 37°C in a humidified atmosphere containing 95% air and 5% CO 2 . When the cultures reached subconfluence, the cells were harvested with Trypsin-EDTA solution (cat# 23315, IBL CO., Ltd, Gunma, Japan), and re-plated at a density of 5 x 10 5 cells in a 100-mm dish. Medium changes were carried out twice a week thereafter.

Generation of iPSCs
iPSCs were generated from the patient-derived cells through a reprogramming by Sendai virus infection-mediated expression of OCT4/3, SOX2, KLF4, and c-MYC as previously described 26 .
Edom22iPS#S31 menstrual blood-derived cells were used for comparison purposes 27,28 . Human iPSCs were maintained on irradiated mouse embryonic fibroblasts as previously described 29,30 . The elimination of Sendai virus was confirmed by RT-PCR. Cells just after infection served as a positive control. Sequences of the primer sets for the Sendai virus are shown in Table 1.

Immunocytochemical analysis
Cells were fixed with 4% paraformaldehyde in PBS for 10 min at 4°C. After washing with PBS and treatment with 0.2% Triton X-100 in PBS for 10 min, cells were pre-incubated with blocking buffer (10% goat serum in PBS) for 30 min at room temperature and then exposed to primary antibodies in blocking buffer overnight at 4°C. Following washing with 0.2% PBST, cells were incubated with secondary antibodies; either anti-rabbit or anti-mouse IgG conjugated with Alexa 488 or 546 (1:300) (Invitrogen) in blocking buffer for 30 min at room temperature. Then, the cells were counterstained with DAPI and mounted.

Karyotypic analysis
Karyotypic analysis was contracted out to Nihon Gene Research Laboratories Inc. (Sendai, Japan).

Cytochrome P450 induction
The expression levels of three major CYP enzymes, CYP1A2, CYP2B6, and CYP3A4, were examined to evaluate CYP induction in HepaKI. HepaKI was treated with omeprazole, phenobarbital, and rifampicin, and controls were treated with dimethyl sulfoxide (DMSO), as described in the previous study 17 .

Statistical analysis
Statistical analysis was performed using the unpaired two-tailed Student's t-test.

Generation of drug-induced hepatic injury iPSCs
We generated iPSCs from a patient with hepatic failure by Sendai virus infection-mediated expression of OCT4/3, SOX2, KLF4, and c-MYC ( Figure 1A). When these reprogramming factors were SOX2, NANOG, and OCT4/3, which was consistent with the profile observed in iPSCs ( Figure 1F).
To investigate multipotency in vitro, iPSCs were differentiated into ectodermal, mesodermal, and endodermal lineages. The differentiation of iPSC-K was confirmed by immunostaining using antibodies against β -tubulin III (TUJ1), α -smooth muscle actin (SMA), and α -fetoprotein (AFP) as ectodermal, mesodermal, and endodermal markers, respectively ( Figure 1G, H). To address whether the iPSC-K has the competence to differentiate into specific tissues in vivo, teratomas were formed by implantation of iPSC-K in the subcutaneous tissue of immunodeficient NOD/SCID mice. iPSC-K produced teratomas within 6-10 weeks after implantation. Histological analysis of paraffin-embedded sections demonstrated that the three primary germ layers were generated as shown by the presence of ectodermal, mesodermal, and endodermal tissues in the teratoma ( Figure 1I), implying iPSC-K has the potential for multilineage differentiation in vitro and in vivo. Among iPSC-K clones, #25, #66, and #100 generated a larger area of liver-like tissues in the teratomas, while the other clones did not.

Hepatic differentiation
We investigated the efficiency of hepatic differentiation of iPSC-K by two different methods Protocol H and Protocol S (described in detail in Figure 2A, B, C). The differentiated cells exhibited hepatocyte-like morphology, i.e. a polygonal and/or cuboidal shape that had tight cell-cell contact when generated by either method (Figure 2D, E) and the expression of the hepatocyte-associated genes were comparable. Quantitative analysis revealed that iPSC-K expressed the genes for AFP, ALB, and α -antitrypsin (AAT) 21 days after the start of induction ( Figure 2F, G, H). We employed Protocol S and used iPSC-K#25 hereafter for the iPSC-K experiments because iPSC-K#25 exhibited morphology that most resembled primary hepatocytes, and high expression of hepatocyte-associated genes. Time-course analysis revealed that similar expression levels of liver-associated genes were observed at 21, 28, and 35 days after the start of hepatic induction ( Figure 2I, J).

Verification of the manufacturing process
To ensure the consistent production of HLCs from iPSC-K, we established a master cell bank and used a working cell bank (WCB) for starting material. We then developed a standard operating procedure for hepatic differentiation ( Figure 3A). HepaKI developed into a cell type that was positive for both AFP and ALB ( Figure 3B). Karyotypic analysis showed that the same chromosomes, without any aberration, were present in the WCB stock as parental fibroblastic cells from the patient ( Figure 3C, D). Exome analysis revealed that iPSC-K had no significant single nucleotide alterations in a homozygous manner. To verify the procedure, we repeatedly manufactured HLCs from the WCB.

Induction of the genes for cyp1A2, 2B6, and 3A4
To investigate whether HepaKI exhibits CYP induction, we exposed HepaKI to omeprazole, phenobarbital, and rifampicin for 24, 48, and 48 h, respectively ( Figure 4). Expression of the genes for AFP and ALB was unchanged with exposure to these drugs. Cyp1A2 was increased after exposure to omeprazole and phenobarbital, while Cyp2B6 remained unchanged ( Figure 4C, D). Interestingly, Cyp3A4 was up-regulated 57.2-fold, on average, upon exposure to rifampicin ( Figure 4E).

in vivo analysis of Hepa-KI cells in the kidney
To investigate engraftability and functionality, we implanted human PSC-derived HLCs into the spleen of cDNA-uPA/SCID mice, however, PSC-derived HLCs were not engrafted and only transiently human ALB was elevated. We then implanted organ-derived hepatocytes, i.e. cryopreserved hepatocytes obtained from the surplus liver during liver transplantation in cDNA-uPA/SCID mice. The hepatocytes formed multiple colonies as detected by a human-specific antibody to CYP2C ( Figure 5A, B), and blood human ALB was elevated.
We also implanted HepaKI HLCs under the renal capsules of immunodeficient mice. HepaKI successfully engrafted and displayed trabeculae of monomorphic polygonal cells with uniform round/oval nuclei and abundant granular eosinophilic cytoplasm, with capillary vessels (Figure 5C-F).
We also observed extramedullary erythropoiesis at the implanted sites in the subrenal capsules ( Figure   5F).

DISCUSSION
In this study, HLCs were prepared from PSCs and characterized as hepatocytes, and engraftment was examined. In vitro, HepaKI showed characteristics as hepatocytes. However, despite the optimization of differentiation protocols, HLCs did not engraft in a mouse model of hepatic failure. There has been a report of engraftment of PSC-derived human HLCs in mice 32 , and the lack of engraftment in this study may be due to the insufficient protocol for differentiation into hepatocytes that engraft and proliferate in vivo. PSCs themselves have the potential for hepatocyte differentiation, in which hepatocytes clearly appeared as colonies or a cell population in teratomas ( Figure 1I, panel for hepatocytes). Distinct colonies of hepatocytes in the teratomas suggest that these colonies are formed from a single hepatocyte/hepatic progenitor/hepatic stem cell during teratoma formation. In addition, HLCs formed trabeculae and exhibited extramedullary erythropoiesis along with colonies of hepatocytes when implanted under the renal capsule ( Figure 5D), suggesting a simulation of liver development with the presence of fetal liver cells or hepatoblasts. Hepatocytes in the renal capsule generated a cell population with a trabeculae structure; these trabeculae could have been formed by a large number of differentiated cells or a single proliferative hepatocyte. It is likely that a single hepatocyte proliferates and induces erythropoiesis, similar to that of the embryonic liver.
The CFU-s assay can quantify HSCs in bone marrow, and this idea has contributed to fundamental studies in bone marrow transplantation 7,8 . Likewise, as for hepatocyte transplantation, we have confirmed that human hepatocytes survived in the liver of immunocompromised mice 33 . The use of genetically engineered cDNA-uPA/SCID mice, in which liver parenchymal cells are progressively impaired, resulted in more pronounced engraftment of transplanted hepatocytes 7,8 . The colonies generated by isolated human hepatocytes in the liver of cDNA-uPA/SCID mice are thought to be derived from a single cell. The expression pattern of OTC genes in a single colony is uniform ( Figure   5G, H) 25 , despite the random inactivation of the OTC genes on the X chromosome 34 . We, therefore, speculate that "colony-forming activity in liver" can be used to quantify "hepatocytes that engraft and proliferate in vivo" as a quantitative assay for stem cells that utilize colony-forming ability like bone marrow stem cells ( Figure 5I). "Hepatocytes that engraft and proliferate in vivo" encompass the conceptual notion of liver stem/progenitor cells.

This research was supported by AMED; by KAKENHI; by the Grant of National Center for Child
Health and Development. Computation time was provided by the computer cluster HA8000/RS210 at the Center for Regenerative Medicine, National Research Institute for Child Health and Development.

Acknowledgments
We would like to express our sincere thanks to N. Ito and K. Miyado for the fruitful discussion, to M.
Ichinose for providing expert technical assistance, to C. Ketcham for English editing and proofreading, and to E. Suzuki and K. Saito for secretarial work.

Competing financial interests
AU is a co-researcher with MTI Ltd., Terumo Corp., BONAC Corp., Kaneka Corp., CellSeed Inc., ROHTO Pharmaceutical Co., Ltd., SEKISUI MEDICAL Co., Ltd., Metcela Inc., PhoenixBio Co., Ltd., Dai Nippon Printing Co., Ltd. AU is a stockholder of TMU Science Ltd., Morikuni Ltd., and Japan Tissue Engineering Co., Ltd. The other authors declare that there is no conflict of interest regarding the work described herein. All authors have read and approved the manuscript.

Author Contribution Statement
AU designed experiments. KI, MT, SY, and TY performed experiments. KI and AU analyzed data. KI, MYI, KT, AN, and MK contributed reagents, materials, and analysis tools. MT, HA, HN, TK, and NO discussed the data and manuscript. AU and KI wrote this manuscript.

Additional Information
The read data have been submitted to the Sequence Read Archive (SRA) under accession number