ORIGINAL PAPER

Hepatic differentiation of mouse iPS cells and analysis of liverengraftment potential of multistage iPS progeny

Anangi Balasiddaiah & Daniel Moreno &

Laura Guembe & Jesús Prieto & Rafael Aldabe

Received: 22 December 2012 /Accepted: 5 May 2013 /Published online: 30 May 2013# University of Navarra 2013

Abstract Hepatocyte transplantation is considered apromising therapy for patients with liver diseases.Induced pluripotent stem cells (iPSCs) are an unlimitedsource for the generation of functional hepatocytes.While several protocols that direct the differentiation ofiPSCs into hepatocyte-like cells have already beenreported, the liver engraftment potential of iPSC progenyobtained at each step of hepatic differentiation has notyet been thoroughly investigated. In this study, we pres-ent an efficient strategy to differentiate mouse iPSCs intohepatocyte-like cells and evaluate their liver engraftmentpotential at different time points of the protocol (5, 10,15, and 20 days of differentiation). iPSCs were differen-tiated in the presence of cytokines, growth factors, andsmall molecules to finally generate hepatocyte-like cells.

These iPSC-derived hepatocyte-like cells exhibitedhepatocyte-associated functions, such as albumin secre-tion and urea synthesis. When we transplanted iPSCprogeny into the spleen, we found that 15- and 20-dayiPSC progeny engrafted into the livers and further ac-quired hepatocyte morphology. In contrast, 5- and 10-day iPSC progeny were also able to engraft but did notgenerate hepatocyte-like cells in vivo. Our data may aidin improving current protocols geared towards the use ofiPSCs as a new source of liver-targeted cell therapies.

Keywords Hepatic differentiation . Hepatocyte-derived iPSCs . Hepatocyte-like cell transplantation

Introduction

Hepatocytes play key roles in drug metabolism and rep-resent an important focus of research studies investigatingnew medicines and fundamental biological mechanismsin metabolic and viral diseases [8]. While the use of cell-based therapies to treat liver diseases is unquestionablyjustified, progress in the field has been slow because oftwo reasons: the shortage of liver organs from whichhepatocytes can be isolated and the difficulty inmaintaining hepatocytes in culture, as they tend to spon-taneously dedifferentiate in vitro [6]. Interestingly, severalstudies have reported that transplanted hepatocytes canintegrate into the host liver parenchyma and restore nor-mal hepatic functions [5].

In recent years, stem cells have been proposed as anideal source to generate unlimited numbers of hepato-cytes. For this reason, it is crucial to develop robustmethods for the differentiation of mouse and human

J Physiol Biochem (2013) 69:835–845DOI 10.1007/s13105-013-0260-9

Electronic supplementary material The online version of thisarticle (doi:10.1007/s13105-013-0260-9) containssupplementary material, which is available to authorized users.

A. Balasiddaiah :D. Moreno : J. Prieto :R. Aldabe (*)Gene Therapy and Hepatology Area,FIMA University of Navarra,Pamplona, Spaine-mail: rafaldabe@yahoo.com

L. GuembeMorphology Lab, Center for Applied Medical Research(CIMA), FIMA University of Navarra,Pamplona, Spain

J. PrietoCIBER-EHD, University of Navarra,Pamplona, Spain

J. PrietoLiver Unit, University Clinic, University of Navarra,Pamplona, Spain

stem cells into hepatocytes in vitro and investigate theirliver engraftment potential in vivo.

It has been extensively shown that induced pluripo-tent stem cells (iPSCs) can be derived from terminallydifferentiated mouse and human cell types [1, 2, 23, 24].iPSCs were initially generated using pluripotency fac-tors, such as Oct4, Sox2, Klf4, and c-Myc or OCT4,SOX2, NANOG, and LIN28, that activate the transcrip-tional regulatory circuit of pluripotent cells [23, 26].While not identical to embryonic stem cells (ESCs)[3], iPSCs share their defining features: they can formteratomas containing cell types from all three embryonicgerm layers, generate chimeras, and contribute to thegermline [18]. An interesting property of iPSCs is theirepigenetic memory, which has been shown in part tofacilitate their differentiation into their tissue of origin[11]. The differentiation of iPSCs into hepatocyte-likecells has recently been described by many groups [17,19, 21]. However, the potential of differentiated iPSCprogeny to engraft in the liver is still poorly understood.

Molecular cues that drive embryonic liver develop-ment have become the primary source of information fordirecting pluripotent stem cells into hepatocytes [27].Using this information, several research groups havereported that hepatocyte-like cells can be generatedfrom ESCs [20] and iPSCs [19]. In this study, wedescribe the differentiation of mouse iPSCs into func-tional hepatocyte-like cells using a protocol that at-tempts to mimic embryonic liver development.Through the sequential application of defined cyto-kines, we induced the differentiation of iPSCs to formdefinitive endoderm, hepatic endoderm, and finallyhepatocyte-like cells. In addition, we evaluated theliver engraftment potential of iPSC-derived cellpopulations obtained at different stages of hepaticdifferentiation. We show that differentiated iPSC prog-eny at the end of the differentiation program exhibithepatocyte-associated functions, such as albumin (Alb)secretion and urea synthesis. In addition, iPSC prog-eny engrafted into injured mouse livers, and a fractionof the progeny acquired a hepatic phenotype.

Materials and methods

Reprogramming GFP+ mouse hepatocytes into iPSCs

Mouse hepatocytes were obtained from 7- to 9-week-old transgenic C57BL/6 Tg14 (act-eGFP) Os-bY01

mice by adopting the two-step collagenase perfusiontechnique. C57BL/6 Tg14 (act-eGFP) Os-bY01 miceexpress GFP under the control of the actin promoter.pMX retroviruses coding for Oct4, Sox2, Klf4, and/or c-Myc were used to infect GFP+ mouse hepatocytes. After2 weeks of infection, iPSC colonies appeared. Thirty-nine reprogrammed iPSC colonies (reprogrammed withor without c-Myc) were selected and expanded on mi-totically inactivated fibroblasts in the presence of iPScell medium as previously reported by Yamanaka andcolleagues [24]. Thirteen clones were selected for fur-ther characterisation. Finally, two clones that werereprogrammed without c-Myc (1.38 and 1.39) wereselected for further characterisation. The 1.38 clonewas used for differentiation and transplantation experi-ments. The mouse embryonic stem cell line D3 wasused as a positive control when analysing embryonicstem cell marker expression in iPSC clones.

Alkaline phosphatase staining

iPSCs were fixed using a 3-mM sodium citrate solu-tion containing 60 % acetone and incubated for 30 s atroom temperature. After washing the cells with dis-tilled water, 1 ml of freshly prepared alkaline phos-phatase staining solution (Fast Blue SALT+NaphtolAS-MX; Sigma) was added, and the plates were incu-bated in the dark for 30 min at room temperature. Thereaction was stopped by washing with water. iPSCsstained violet, indicating the presence of alkalinephosphatase.

Teratoma induction and histological analysis

To induce teratomas, iPSCs were harvested bytrypsinisation and resuspended in DMEM containing10 % FBS without antibiotics. Four- to six-week-oldBALB/c Rag2−/− IL2g−/− mice were anesthetisedwith a 4:1 (vol./vol.) mixture of Imalgen (MerialLabs) and Rompun (Bayer Health Care). One mil-lion iPSCs in a volume of 100 μl were subcutane-ously injected into the dorsal flanks. At 3 to 4 weeksafter injection, the tumours were surgically removed,cut into small sections, and fixed in 4 % formal-dehyde. After 24 h, tissue sections were placed incassettes and stored in 70 % alcohol. Tissue em-bedding in paraffin and hematoxylin eosin stainingwere performed following standard histologicalmethods.

836 A. Balasiddaiah et al.

Albumin detection by ELISA

Tomeasure the Alb secretion of iPSC-derived hepatocyte-like cells, medium was collected from the culture at 48 hafter the medium change and centrifuged to remove thecell debris, and 100μl of supernatant was used to quantifyAlb using a mouse Alb quantitative kit (BethylLaboratories). Alb secretion was normalised to the num-ber of the cells (the values shown were normalised to 0.5million cells).

Liver damage induction

All animal studies were performed in accordance withthe guidelines of the institutional ethics commission.The experimental design employing mice was ap-proved by the Ethics Committee for Animal Testingof the University of Navarra, Pamplona, Spain. MaleBalb/C Rag2−/− γC

−/− mice (5 weeks old) were intra-venously inoculated with 5×109 PFU of recombinantadenovirus encoding thymidine kinase (AdTk) via theretro-orbital plexus in 100 μl of saline solution. Threedays after infection, the mice were administered twodoses of 25 mg/kg of Ganciclovir (GCV; Roche) in-traperitoneally on alternating days.

Transplantation of iPSC progeny

Differentiated iPSCs were trypsinised, washed twicewith PBS, resuspended in PBS (1×106/100 μl), andstored on ice until they were transplanted. At 48 h afterthe administration of a second dose of GCV, the maleRag2−/− γC

−/− mice were anesthetised with isoflurane(Abbott Labs) and kept on a heating pad during theentire procedure. Then, the mice were injected withketoprofen (ketofen 10 %, Merial Labs) to reduce theinflammation and pain caused by the operation. Next,a mid-abdominal incision was made, and 1×106 iPSCprogeny were intrasplenically transplanted, followedby closure of the abdominal incision using silk su-tures. The transplanted mice were then treated withenrofloxacin (Bayer Health Care) diluted to 25 mg/mlin drinking water for 1 week to protect them frompathogenic microorganisms.

Liver engraftment analysis by immunohistochemistry

Mice were sacrificed by cervical dislocation at 4 weeksafter cell transplantation, and their livers were surgically

removed, cut into small sections, and fixed in 4 %formaldehyde. After 24 h of fixation, tissue sectionswere placed in cassettes and stored in 70 % alcohol.Rabbit anti-GFP antibody (1:30,000 dilution, Abcam)and the EnVison anti-rabbit detection system (Dako)were used to detect iPSC-derived cells that engraftedinto mouse livers. Tissue embedding in paraffin and allother procedures were performed following standardimmunohistochemical methods. IHC analysis (ratherthan direct fluorescence detection of GFP) was preferredas livers exhibit high autofluorescence.

Immunofluorescence staining

iPSCs were grown on cover glasses, fixed with coldmethanol, and incubated with primary antibodies di-luted in PBS containing 3 % BSA and 0.05 % Tweenat 37 °C for 30 min. After incubation, the coverglasses were washed three times with PBS. Next, thecover glasses were incubated with secondary anti-bodies diluted in PBS containing 3 % BSA and0.05 % Tween at 37 °C for 30 min. Then, the cellswere washed three times with PBS. Subsequently, thecover glasses were mounted with Vectashield mount-ing medium containing DAPI (Vector laboratories).An inverted microscope (Model DMRB, Leica) wasused for immunocytochemical analysis. The primaryantibodies rabbit anti-Oct4 (1:50, Santa Cruz), mouseanti-SSEA-1 (1:50, Abcam), and rabbit anti-Nanog(1:450, Abcam) were used in this study. The second-ary antibodies anti-rabbit-Cy3 (1:1,500 Sigma) andanti-mouse Alexa 594 (1:500 Sigma) were used in thisstudy.

Differentiation of iPSCs into hepatocyte-like cellsin vitro

Prior to differentiation, iPSCs were trypsinised into asingle-cell suspension and further incubated for30 min at 37 °C and 5 % CO2 on gelatin-coated dishesto deplete feeder fibroblasts. To specify definitiveendoderm (DE), supernatants containing predominant-ly iPSCs were collected, and the cells were countedand resuspended in advanced RPMI containing 10 %FBS, activin A (100 ng/ml), and Wnt3a (50 ng/ml).Then, 9.5×105 cells/well were seeded in 12-wellplates (3.8 cm2/well) coated with 2 % matrigel, andthe plates were incubated at 37 °C and 5 % CO2

overnight. Subsequently, the medium was replaced

Hepatic differentiation of mouse iPS cells 837

with advanced RPMI containing 0.2 % FBS, activin A(100 ng/ml), and Wnt3a (50 ng/ml), and the cells wereincubated for 5 days at 37 °C and 5 % CO2. Themedium was changed every 2 days.

To induce hepatic endoderm from DE, cells weretrypsinised, and 2.5×105 cells were seeded in eachwell of 12-well plates coated with 2 % Matrigel andcultured in advanced RPMI containing 10 % FBS,BMP4 (50 ng/ml; R&D systems), and FGF-2(20 ng/ml; Sigma Aldrich) overnight. After overnightincubation, the medium was replaced with advancedRPMI containing 2 % FBS, BMP4 (50 ng/ml), andFGF-2 (20 ng/ml), and the cells were incubated for5 days at 37 °C and 5 % CO2. The medium waschanged every 2 days.

To induce immature hepatocytes from hepatic en-doderm cells, the medium was replaced with advancedRPMI containing 2 % FBS and HGF (20 ng/ml), andthe cells were incubated for 5 days at 37 °C and 5 %CO2. The medium was changed every 2 days.

To induce hepatocyte maturation, the medium wasreplaced with advanced RPMI containing 2 % FBS,HGF (20 ng/ml), oncostatin M (20 ng/ml), Dex(50 nM), and insulin transferrin selenium (1×), andthe cells were incubated for 5 days at 37 °C and 5 %CO2. The medium was changed every 2 days.

Urea quantification

Urea production in the culture medium of differentiat-ed iPSCs was quantified using the QuantiChrom™Urea assay kit (Bioassay systems). To measure theurea secretion by iPSC-derived hepatocyte-like cells,the medium was collected from the culture at 48 hafter the medium was changed and centrifuged toremove the cell debris, and 50 μl of supernatant wasused to quantify the urea. Urea secretion wasnormalised to the number of the cells (the valuesshown were normalised to 0.5 million cells).

Quantitative real-time PCR

qPCR was performed in a final volume of 20 μlcontaining iQ SYBR green supermix (Biorad), senseand antisense primers (final concentration of 300 nM),and 2 μl of cDNA, which was obtained through thereverse transcription of 1 μg of total RNA. qPCR wasperformed using the iQ 5 Multicolour Real Time PCRDetection system (Biorad). The results were analysed

using the delta CT method 2^ (CT value of internalcontrol-CT value of gene of interest). The CT valuesof genes were normalised to the expression of the house-keeping gene actin. Gene expression values inundifferentiated iPSCs were set to 1, and the gene ex-pression values of other iPSC progeny were expressedas fold change compared with undifferentiated iPSCs.

Primer design

All the primers used for q-PCR were designed usingvector NTI software (Invitrogen) and synthesised bySigma. These primers were designed between intronsto avoid the amplification of genomic DNA. Theprimer sequences of the genes studied are shown inSupplementary Table 1.

Statistical analysis

Statistical analysis was performed using GraphPadPrism5 Software. qPCR data were presented as the meanvalue of one experiment representative of three indepen-dent experiments. Each experiment was performed intriplicate. The data were presented as the mean value±SD, and a one-way ANOVAwas used to compare all thedata with undifferentiated iPSCs.

Results

Generation and characterisation of iPSCs

In this study, we chose endoderm-derived cells, primarymouse hepatocytes, as the source of somatic cells forreprogramming because it has been shown thathepatocyte-derived iPSCs offer many advantages, suchas being less tumorigenic compared with iPSCs derivedfrom other somatic cells [2] and retaining epigeneticmemory of their origin cells. The latter advantage mayfacilitate efficient hepatocyte differentiation, as it hasrecently been shown that iPSCs harbour residual DNAmethylation signatures characteristic of their somatictissue of origin [11].

Hepatocytes derived from mice that constitutivelyexpress GFP under the control of the actin promoter wereused to generate iPSCs for the subsequent detection oftransplanted iPSC-derived cells. We reprogrammedGFP+mouse hepatocytes into iPSCs by overexpressingthe reprogramming factors Oct4, Sox2, Klf4, and/or c-

838 A. Balasiddaiah et al.

Myc as described by Yamanaka and colleagues [2].Similar to the results of that group, the hepatocyteslost their morphology upon overexpression of thesefactors, and iPSC colonies were formed. When thesecolonies were selected and expanded on mitoticallyinactivated fibroblasts in the presence of leukaemiainhibitory factor, they acquired ESC-like morphology(Fig. 1a).

When we omitted c-Myc from the cocktail ofreprogrammed factors, we observed a similar numberof reprogrammed hepatocyte colonies compared withcolonies reprogrammed with the four factors (data notshown), which is contrary to a previous report thatshowed that hepatocyte reprogramming without c-Myc decreases colony numbers by 20–40 %. Thus,in our study, c-Myc addition did not seem to affect theefficiency of hepatocyte reprogramming. In agreementwith this observation, when we evaluated c-Myc

levels in primary mouse hepatocytes and fibroblasts,we found that mouse hepatocytes expressed ten timesmore c-Myc than fibroblasts (Supplementary Fig. 1c),suggesting the possible role of endogenous c-Myc inhepatocyte reprogramming. Because it has beenshown that mice transplanted with iPSCs generatedwithout c-Myc do not develop tumours [16], in thisstudy, we used iPSCs generated without c-Myc fordifferentiation and transplantation experiments.

To determine whether hepatocyte-derived iPSCs(named Hep-iPSCs hereafter) expressed ESC-specificgenes, we performed qPCR and observed that similarto the D3 ESC line, Hep-iPSCs expressed pluripotencymarkers, such as Oct4, Nanog, Sox2, Klf4, Ecat, Eras,Fbx15, and Rex1 (Fig. 1c, d). In addition, immunofluo-rescence data confirmed the expression of the keypluripotency markers Oct4 and Nanog in these iPSCclones (Supplementary Fig. 1a). Moreover, Hep-iPSCs-

Fig. 1 Characterisation of iPSC clones. a From left to right,morphology of primary hepatocytes, a hepatocyte-derived iPSCcolony, a stable iPSC line, and GFP retrovirus-infected mousehepatocytes, which served as the negative control (magnification,×10). b Teratoma formation by iPSCs. Hematoxylin- and eosin-stained sections of teratomas derived upon the subcutaneous injec-tion of iPSCs into immunodeficient mice are shown (×10). c Anal-ysis of endogenous pluripotency factors Oct4, Nanog, Sox2, andKlf4 in iPSC clones 1.38 and 1.39; hepatocytes; and the D3 ESC

line with qPCR. d Analysis of other ESC-specific genes Ecat, Eras,Fbx15, and Rex1 in iPSC clones, hepatocytes, and the D3 cell linewith qPCR. The D3 cell line and primary mouse hepatocytes wereused as positive and negative controls, respectively. The expressionof ESC-specific genes is represented as fold change compared withD3 expression. The expression levels of ESC genes in the D3 cellline is set to 1. Gene expression is normalised to β-actin. The datarepresent the average values of triplicates from the same experiment,and error bars represent the standard deviation

Hepatic differentiation of mouse iPS cells 839

induced teratomas containing derivatives of all threegerm layers, mesoderm, endoderm, and ectoderm(Fig. 1b), and stained positive for alkaline phosphatase(Supplementary Fig. 1b). Altogether, these results dem-onstrate that Hep-iPSCs are pluripotent.

Hepatic differentiation of Hep-iPSCs and characterisationof in vitro differentiated Hep-iPSCs

Existing protocols on differentiating ESCs or iPSCsinto hepatocytes have largely been based on themolecular cues that drive liver development duringembryogenesis [20]. These can be summarised intothree stages: endoderm induction, hepatic specifica-tion, and hepatic maturation. Building on the pro-tocols reported by others [10, 19, 21], we describedthe differentiation of Hep-iPSCs into hepatocyte-likecells (Fig. 2a). Using this 20-day differentiation pro-tocol, we differentiated Hep-iPSCs into DE cells andfurther into hepatocyte-like cells (Fig. 2b). Cellswere obtained at multiple stages of differentiation,and qPCR was performed to evaluate the kinetics of

endoderm and hepatic marker induction duringdifferentiation.

Based on qPCR data, Nanog and Oct4 showeddecreased expression towards the end of differentia-tion (Fig. 3), suggesting the concurrent differentiationof Hep-iPSCs and loss of pluripotency, as previouslydescribed by others [10].

It has been previously determined that ESCs andiPSCs can be directed towards endoderm cellscharacterised by the expression of Gsc, Sox17, FoxA2,and other DE markers upon in vitro differentiation in thepresence of cytokines such as activin A and Wnt3a [4,10, 15, 25]. To analyse the expression kinetics of theseendodermal markers, we performed qPCR. Goosecoid(Gsc), a gene expressed in the meso-endoderm and DE,was upregulated only in day 5 cultures and declinedrapidly as differentiation progressed (Fig. 3). Moreover,Sox17, DE marker, and Hnf1α/β, pivotal factors in theestablishment of the early liver development peaked inday 5 cultures and declined progressively towards theend (Fig. 3; Supplementary Fig. 2). Finally, FoxA2,which is abundantly expressed in DE and mature

Fig. 2 Hepatic differentiation of iPSCs. a Schematic diagram showing the stepwise hepatic differentiation protocol. b Morphologicalchanges in iPSCs at different stages of differentiation (×10)

840 A. Balasiddaiah et al.

hepatocytes, was induced in day 5 cultures andmaintained throughout differentiation (Fig. 3). Thesedata suggest that DE cells were generated at day 5 ofdifferentiation.

It has been shown that cells differentiated from ESCsand iPSCs at later stages of differentiation express hepaticmarkers, such as Ttr, Afp, Tdo, and Alb [7, 9, 19, 22].During embryonic liver development, Ttr and Afp areexpressed in hepatoblasts [12, 13]. To verify the ex-pression of Ttr and Afp in differentiated Hep-iPSCs,we performed qPCR and found that Ttr and Afp wereupregulated at days 5 and 10 after differentiation asreported by others [19] and were decreased at laterstages of differentiation (Fig. 3). Tdo, a marker ofterminally differentiated hepatocytes, was induced atday 15, followed by a sharp increase at day 20, where-as Alb, another marker of matured hepatocytes, wasobserved only at day 20 (Fig. 3). In accordance withthese results, we observed upregulation of C/EBP-β at20 days after differentiation but curiously not of theC/EBP-α (Supplementary Fig 2). Accordingly, whenwe compared the expression of several hepatocyte

markers between iPSCs progeny differentiated for20 days and primary hepatocytes, we found lowerexpression in iPSC progeny suggesting that maturationis incomplete (Supplementary Table 2).

Hepatocytes perform numerous synthetic and detoxi-fication functions. Synthetic functions can be assessed bymeasuring the ability of a cell population to secrete Alb,while detoxification functions can be evaluated by testingthe ability of cells to synthesise urea [19], as hepatocytesconvert ammonia, a toxic substance, to urea. Therefore, totest the ability of iPSC-derived hepatocyte-like cells toproduce Alb and urea, we performed ELISA and ureaquantification assay, respectively. In contrast to otherstudies that reported Alb secretion at early time points[19, 22], we detectedAlb secretion only at day 20 (Fig. 4),which coincided with the induction of Alb transcripts onday 20. Furthermore, when we quantified urea levels insupernatants collected at 5, 10, 15, and 20 days of differ-entiation, we found that all iPSC progeny, except thosefrom day 5, had the ability to synthesise and secrete ureainto supernatants (Fig. 4), which is not in accordance withprevious reports that reported the detection of urea only at

Fig. 3 Characterisation of differentiated iPSCs. qPCR expres-sion analysis of the pluripotency markers Oct4 and Nanog; theendoderm markers Gsc, Sox17, and FoxA2; and the hepatocytemarkers Ttr, Afp, Tdo, and Alb in iPSC progeny obtained at 5,10, 15, and 20 days. The CT values of all genes were normalisedto the CT values of β-actin. The y-axis represents the fold

change of gene expression compared with undifferentiatediPSCs. “UN” in the x-axis represents iPSCs without differenti-ation (undifferentiated iPSCs). One-way ANOVA test was usedfor statistical analysis (*p<0.05; **p<0.01; ***p<0.001). Er-ror bars represent the standard deviation. The data represent theaverage values of triplicates of one experiment

Hepatic differentiation of mouse iPS cells 841

the final stage of differentiation [19]. However, when weanalysed the expression of liver specific arginase-1, anenzyme involved in the production of urea in the liver, wefound that it is induced only at day 20 (SupplementaryFig. 2).

Taken together, these results confirm the presenceof functional hepatocyte-like cells at the end of differ-entiation, despite the persistent expression of the im-mature hepatocyte marker AFP.

Analysis of liver engraftment potentialof differentiated Hep-iPSCs

It has been previously demonstrated that iPSC-derivedhepatocyte-like cells can engraft into the liver upon trans-plantation [7, 19]. However, most research groups did not

study the liver engraftment potential of iPSC-derivedcells obtained at different stages of differentiation.

A persistent liver damage for an efficient engraftmentof hepatic cells in the liver is necessary. Thus, we haveused amouse model that develops liver damage upon thehepatic expression of thymidine kinase and GCVadmin-istration for transplanting differentiated iPSC progenyobtained at different stages of differentiation and evalu-ated their engraftment potential in the liver. AdTk–GCV-induced liver damage is characterised by an inflamma-tory process accompanied by increased hepatocyte sizeand enlarged nuclei (Fig. 5a, b). This effect is maintainedfor 10 weeks and then is progressively resolved.

Using AdTk–GCV mouse model, we transplantedmultistage iPSC progeny in the spleen and we analysedcellular engraftment 4 weeks later. While screening theliver parenchyma, we observed small colonies resem-bling hepatocyte morphology that were positive forGFP in the liver in two out of five mice transplanted withday 15 iPSC progeny and two out of ten micetransplanted with day 20 iPSC progeny, thus indicatingthat they were derived from transplanted cells (Fig. 5c, d).In contrast, no hepatocyte-like cells were observed inanimals transplanted with iPSCs differentiated for 5 and10 days. Instead, we observed GFP-positive cells that didnot exhibit hepatocyte morphology. Day 5 iPSC progenywere mainly distributed as clusters, which could be at-tributed to the higher proliferation capacity of DE cells,while day 10 iPSC progeny were scattered among thehost parenchyma, indicating that they were more differ-entiated compared with DE cells (Fig. 6). Cells from allfour stages were engrafted into the damaged livers with-out any preferential localisation, and importantly, thesecells did not form teratomas in the livers. These resultssuggest that days 5 and 10 iPSC progeny engraft into theliver but may not have the potential to differentiate intohepatocytes in damaged livers, while days 15 and 20iPSC progenies have the potential to engraft, and theyexhibit hepatocyte morphology upon transplantation.These in vivo data corroborate with our in vitro data,which showed that hepatocyte markers were mainly ob-served during the last two stages of differentiation.

Discussion

iPSCs are one of the most promising stem cell typesthat can be used as an alternative source to producemature hepatocytes for liver-targeted cell therapies. In

Fig. 4 Functional characterisation of differentiated iPSCs. aAlb protein concentration in supernatants obtained at differentstages of differentiation. b Urea concentration in supernatantsobtained at different stages of differentiation. After 48 h ofculture, supernatants were collected for Alb and urea analysis,and values were normalised by the number of cells (per 0.5million cells). The data represent the average values of tripli-cates from the same experiment, and error bars represent thestandard deviation. One-way ANOVA test was used for statisti-cal analysis (*p<0.05; **p<0.01; ***p<0.001)

842 A. Balasiddaiah et al.

Fig. 5 Liver engraftment of iPSC progeny following AdTk–GCV-mediated liver injury. Images show hematoxylin–eosin-stained liver sections of a control mouse, b mouse 4 weeks afterAdTk–GCV treatment, c mouse transplanted with iPSCs

differentiated for 15 days, and d mouse transplanted with iPSCsdifferentiated for 20 days. Lower panels of (c) and (d) showimmunohistochemistry of GFP-positive cells present in mouseliver parenchyma (×10)

Fig. 6 Liver engraftment ofiPSC progeny differentiatedfor 5 and 10 days. Imagesshow the engraftment ofdays 5 (a) and 10 (b) iPSCprogeny. Images in the upperand lower panels show he-matoxylin and eosin stainingand GFP immunohistochem-istry of seriated liver sections,respectively (×10)

Hepatic differentiation of mouse iPS cells 843

this study, we describe an in vitro hepatocyte differenti-ation protocol that, similar to other studies, sequentiallydifferentiates mouse iPSCs into cells expressing DEmarkers, followed by the acquisition of some hepatoblastmarkers and, finally, mature hepatocyte characteristics.During embryonic liver development, Ttr and Afp areexpressed in hepatoblasts and the extraembryonic endo-derm [12, 13]. The early expression of Ttr and Afp onday 5 of differentiation suggests that some cells mightdifferentiate into extraembryonic endoderm. However,the increased expression of Ttr and Afp at day 10 sug-gests that hepatoblast specification indeed occurred atthis stage. In addition, upregulation of HNF-1α andHNF-1β at earlier stages of differentiation indicates thatthis protocol mimics the events of early embryonic liverdevelopment in vivo. The expression of Tdo, Alb, andC/EBPβ, during the final stage indicates the commitmentof iPSC progeny to mature hepatocytes. However, at thefinal stage of differentiation, iPSC progeny still expressedAfp but not C/EBPα, which indicates that the cellsobtained at this stage were not completely mature, asshown by other reports [7, 19]. In addition, we observedAlb expression only at the last stage of differentiation,which is not in accordance with previous reports thatshowed that Alb expression begins early in differentiation[19, 22]. However, at the end of the differentiation pro-gram, iPSCs exhibited some hepatocyte-associated func-tions, such as Alb secretion and urea synthesis, whichindicates the presence of functional hepatocyte-like cellsin culture. In this study, it appears that differentiated cellsacquire the capacity to synthesise urea earlier than Albsynthesis, as high levels of secreted urea were detected atearly time points. These differences compared with otherreports could be because we used hepatocyte-derivediPSCs for differentiation or differentiation protocols, asthere is no universal protocol to differentiate stem cellsinto hepatocytes.

In this study, we showed that upon transplantation,differentiated iPSCs obtained at all stages engrafted intodamaged livers, but only 15- and 20-day iPSC progenyexhibited hepatocyte morphology. However, in contrastto our findings, a previous report showed that early-stagehepatic cells, such as DE cells, hepatic progenitor cells,can efficiently differentiate into hepatocyte-like cells invivo [14]. This apparent discrepancy between our resultsand theirs may be explained by the differences in theanimal models used. They used a chronic liver cirrhosismodel (i.e., chronic DMN-treated NSGmice), whichmayhave caused a different form of liver damage compared

with that induced by AdTk and GCV treatment. In thisanimal model, AdTk and GCV induce chronic liver dam-age that lasts for 10 weeks, and mice subsequently recov-er from the damage because of oval cell-mediated liverregeneration. Therefore, in different animal models, oncethe transplanted cells reach the liver, they are exposed todifferent cytokines and growth factors, which can affectthe engraftment, fate, and function of transplanted cells.In addition, their study involved differentiated human andnot mouse iPSCs for transplantation. Furthermore, wehave observed that the generated hepatocyte-like cellsare not fully mature, and this characteristic could be alimiting factor for efficient liver repopulation as thisphenomenon was not observed when mouse hepatocytesare transplanted.

In summary, we show here the differentiation ofHep-iPSCs into hepatocyte-like cells and the engraft-ment potential of multistage iPSC progeny using amouse model of liver damage based on Tk and GCV.These findings may aid in improving current differen-tiation strategies in the effort of generating transplant-able liver cells for liver therapy.

Acknowledgments We thank Dr. Miguel Barajas for providingus retroviral plasmids for reprogramming. We thank Dr. AngeloPorciuncula for not only correcting the English but also for somevaluable comments. This work is supported by Foundation forApplied Medical Research (FIMA), University of Navarra, Spain.

References

1. Aasen T, Raya A, Barrero MJ, Garreta E, Consiglio A,Gonzalez F, Vassena R, Bilic J, Pekarik V, Tiscornia G,Edel M, Boue S, Izpisua Belmonte JC (2008) Efficientand rapid generation of induced pluripotent stem cells fromhuman keratinocytes. Nat Biotechnol 26:1276h–1284h

2. Aoi T, Yae K, Nakagawa M, Ichisaka T, Okita K, TakahashiK, Chiba T, Yamanaka S (2008) Generation of pluripotentstem cells from adult mouse liver and stomach cells.Science 321:699–702

3. Chin MH, Mason MJ, Xie W, Volinia S, Singer M, PetersonC, Ambartsumyan G, Aimiuwu O, Richter L, Zhang J,Khvorostov I, Ott V, Grunstein M, Lavon N, BenvenistyN, Croce CM, Clark AT, Baxter T, Pyle AD, Teitell MA,Pelegrini M, Plath K, Lowry WE (2009) Induced pluripo-tent stem cells and embryonic stem cells are distinguishedby gene expression signatures. Cell Stem Cell 5:111–123

4. D'Amour KA, Agulnick AD, Eliazer S, Kelly OG, Kroon E,Baetge EE (2005) Efficient differentiation of human embry-onic stem cells to definitive endoderm. Nat Biotechnol23:1534–1541

5. Dhawan A, Strom SC, Sokal E, Fox IJ (2010) Humanhepatocyte transplantation. Methods Mol Biol 640:525–534

844 A. Balasiddaiah et al.

6. Elaut G, Henkens T, Papeleu P, Snykers S, Vinken M,Vanhaecke T, Rogiers V (2006) Molecular mechanismsunderlying the dedifferentiation process of isolated hepato-cytes and their cultures. Curr Drug Metab 7:629–660

7. Gai H, Nguyen DM, Moon YJ, Aguila JR, Fink LM, WardDC & Ma Y (2010) Generation of murine hepatic lineagecells from induced pluripotent stem cells. Differentiation;research in biological diversity 79:171–181

8. Guguen-Guillouzo C, Guillouzo A (2010) General reviewon in vitro hepatocyte models and their applications.Methods Mol Biol 640:1–40

9. Hamazaki T, Iiboshi Y, Oka M, Papst PJ, Meacham AM,Zon LI, Terada N (2001) Hepatic maturation in differenti-ating embryonic stem cells in vitro. FEBS Lett 497:15–19

10. Hay DC, Zhao DB, Fletcher J, Hewitt ZA, McLean D,Urruticoechea-Uriguen A, Black JR, Elcombe C, Ross JA,Wolf R, Cui W (2008) Efficient differentiation of hepatocytesfrom human embryonic stem cells exhibiting markers recapit-ulating liver development in vivo. Stem Cells 26:894–902

11. Kim K, Doi A, Wen B, Ng K, Zhao R, Cahan P, Kim J, AryeeMJ, Ji H, Ehrlich LI, Yabuuchi A, Takeuchi A, Cunniff KC,Hongguang H, McKinney-Freeman S, Naveiras O, Yoon TJ,Irizarry RA, Jung N, Seita J, Hanna J, Murakami P, JaenischR,Weissleder R, Orkin SH,Weissman IL, Feinberg AP, DaleyGQ (2010) Epigenetic memory in induced pluripotent stemcells. Nature 467:285–290

12. Kwon GS, Fraser ST, Eakin GS, Mangano M, Isern J, SahrKE, Hadjantonakis AK, Baron MH (2006) Tg(Afp-GFP)expression marks primitive and definitive endoderm line-ages during mouse development. Dev Dyn 235:2549–2558

13. Kwon GS, Hadjantonakis AK (2009) Transthyretin mousetransgenes direct RFP expression or Cre-mediated recombi-nation throughout the visceral endoderm. Genesis 47:447–455

14. Liu H, Kim Y, Sharkis S, Marchionni L, Jang YY (2011) Invivo liver regeneration potential of human induced pluripotentstem cells from diverse origins. Sci Transl Med 3:82ra39

15. Morrison GM, Oikonomopoulou I, Migueles RP, Soneji S,Livigni A, Enver T, Brickman JM (2008) Anterior definitiveendoderm from ESCs reveals a role for FGF signaling. CellStem Cell 3:402–415

16. Nakagawa M, Koyanagi M, Tanabe K, Takahashi K,Ichisaka T, Aoi T, Okita K, Mochiduki Y, Takizawa N,Yamanaka S (2008) Generation of induced pluripotent stemcells without Myc from mouse and human fibroblasts. NatBiotechnol 26:101–106

17. Nakamura N, Saeki K, Mitsumoto M, Matsuyama S, NishioM, Saeki K, HasegawaM, Miyagawa Y, Ohkita H, Kiyokawa

N, Toyoda M, Akutsu H, Umezawa A, Yuo A (2012) Feeder-Free and Serum-Free Production of Hepatocytes,Cholangiocytes, and Their Proliferating Progenitors fromHuman Pluripotent Stem Cells: Application to Liver-SpecificFunctional and Cytotoxic Assays. Cell Reprogram 14:171–185

18. Okita K, Ichisaka T, Yamanaka S (2007) Generation ofgermline-competent induced pluripotent stem cells. Nature448:313–317

19. Sancho-Bru P, Roelandt P, Narain N, Pauwelyn K, NotelaersT, Shimizu T, Ott M, Verfaillie C (2011) Directed differentia-tion of murine-induced pluripotent stem cells to functionalhepatocyte-like cells. J Hepatol 54:98–107

20. Snykers S, De Kock J, Rogiers V, Vanhaecke T (2009) Invitro differentiation of embryonic and adult stem cells intohepatocytes: state of the art. Stem Cells 27:577–605

21. Song Z, Cai J, Liu Y, Zhao D, Yong J, Duo S, Song X, Guo Y,Zhao Y, Qin H, Yin X, Wu C, Che J, Lu S, Ding M, Deng H(2009) Efficient generation of hepatocyte-like cells from hu-man induced pluripotent stem cells. Cell Res 19:1233–1242

22. Soto-Gutierrez A, Kobayashi N, Rivas-Carrillo JD, Navarro-Alvarez N, Zhao D, Okitsu T, Noguchi H, BasmaH, Tabata Y,Chen Y, Tanaka K, Narushima M, Miki A, Ueda T, Jun HS,Yoon JW, Lebkowski J, Tanaka N, Fox IJ (2006) Reversal ofmouse hepatic failure using an implanted liver-assist devicecontaining ES cell-derived hepatocytes. Nat Biotechnol24:1412–1419

23. Takahashi K, Tanabe K, Ohnuki M, Narita M, Ichisaka T,Tomoda K, Yamanaka S (2007) Induction of pluripotentstem cells from adult human fibroblasts by defined factors.Cell 131:861–872

24. Takahashi K, Yamanaka S (2006) Induction of pluripotentstem cells from mouse embryonic and adult fibroblast cul-tures by defined factors. Cell 126:663–676

25. Yasunaga M, Tada S, Torikai-Nishikawa S, Nakano Y,Okada M, Jakt LM, Nishikawa S, Chiba T, Era T,Nishikawa S (2005) Induction and monitoring of definitiveand visceral endoderm differentiation of mouse ES cells.Nat Biotechnol 23:1542–1550

26. Yu J, Vodyanik MA, Smuga-Otto K, Antosiewicz-BourgetJ, Frane JL, Tian S, Nie J, Jonsdottir GA, Ruotti V, StewartR, Slukvin II, Thomson JA (2007) Induced pluripotent stemcell lines derived from human somatic cells. Science318:1917–1920

27. Zorn AM, Wells JM (2009) Vertebrate endoderm develop-ment and organ formation. Annu Rev Cell Dev Biol25:221–251

Hepatic differentiation of mouse iPS cells 845