Repopulation of the Fibrotic/Cirrhotic RatLiver by Transplanted Hepatic Stem/Progenitor

Cells and Mature HepatocytesMladen I. Yovchev,1,2 Yuhua Xue,3 David A. Shafritz,1,2 Joseph Locker,1,3,4 and Michael Oertel1,2,3,5

Considerable progress has been made in developing antifibrotic agents and other strat-egies to treat liver fibrosis; however, significant long-term restoration of functional livermass has not yet been achieved. Therefore, we investigated whether transplanted hepaticstem/progenitor cells can effectively repopulate the liver with advanced fibrosis/cirrhosis.Stem/progenitor cells derived from fetal livers or mature hepatocytes from DPPIV1 F344rats were transplanted into DPPIV2 rats with thioacetamide (TAA)-induced fibrosis/cir-rhosis; rats were sacrificed 1, 2, or 4 months later. Liver tissues were analyzed by histo-chemistry, hydroxyproline determination, reverse-transcription polymerase chain reaction(RT-PCR), and immunohistochemistry. After chronic TAA administration, DPPIV2 F344rats exhibited progressive fibrosis, cirrhosis, and severe hepatocyte damage. Besides stellatecell activation, increased numbers of stem/progenitor cells (Dlk-11, AFP1, CD1331, Sox-91, FoxJ11) were observed. In conjunction with partial hepatectomy (PH), transplantedstem/progenitor cells engrafted, proliferated competitively compared to host hepatocytes,differentiated into hepatocytic and biliary epithelial cells, and generated new liver masswith extensive long-term liver repopulation (40.8 6 10.3%). Remarkably, more than 20%liver repopulation was achieved in the absence of PH, associated with reduced fibrogenicactivity (e.g., expression of alpha smooth muscle actin, platelet-derived growth factorreceptor b, desmin, vimentin, tissue inhibitor of metalloproteinase-1) and fibrosis(reduced collagen). Furthermore, hepatocytes can also replace liver mass with advancedfibrosis/cirrhosis, but to a lesser extent than fetal liver stem/progenitor cells. Conclusion:This study is a proof of principle demonstration that transplanted epithelial stem/progen-itor cells can restore injured parenchyma in a liver environment with advanced fibrosis/cirrhosis and exhibit antifibrotic effects. (HEPATOLOGY 2014;59:284-295)

Chronic liver disease with cirrhosis is the twelfthleading cause of death in the United States.1

Cirrhosis, the advanced stage of hepatic fibro-sis, is mainly caused by viral infection or alcoholabuse; age over 50 is also recognized as a risk factorfor cirrhosis.2 Liver transplantation is the only effectivetherapeutic option for these patients.3 Because of ashortage of donor organs4 and a dramatic increase inthe mortality rate of patients on the liver transplantwaiting list during the past decade,5 an alternative

strategy to restore liver mass before the endstage wouldrepresent a major clinical advance.

Progressive hepatic fibrosis as a wound-healingresponse to chronic liver injury leads to accumulationof collagen surrounding liver nodules and furtherreplacement of injured parenchyma by scar tissue,resulting in impaired hepatocyte function.2,6 Hepaticstellate cells are the main contributors to the pathoge-nesis of liver fibrosis.7,8 Therefore, these cells have rep-resented the primary target to reduce or reverse fibrosis

Abbreviations: a-SMA, alpha smooth muscle actin; Col1a2, procollagen a2(I); FLSPC, fetal liver stem/progenitor cell; GFAP, glial fibrillary acidic protein;HYP, hydroxyproline; MMP-2, matrix metalloproteinase-2; PH, partial hepatectomy; TAA, thioacetamide; TIMP1, tissue inhibitor of metalloproteinase-1

From the 1Marion Bessin Liver Research Center, Albert Einstein College of Medicine of Yeshiva University, New York, NY; 2Department of Medicine (Divisionof Gastroenterology and Liver Diseases), Albert Einstein College of Medicine of Yeshiva University, New York, NY; 3Department of Pathology (Division of Experi-mental Pathology), University of Pittsburgh, Pittsburgh, PA; 4Department of Pathology, Albert Einstein College of Medicine of Yeshiva University, New York, NY;5McGowan Institute for Regenerative Medicine, University of Pittsburgh, Pittsburgh, PA.

Received November 16, 2012; accepted June 26, 2013.Supported by NIH grant R01 DK090325 (to M.O.) and a Pilot and Feasibility Study (to M.O.) from P30 DK041296 (to D.A.S.), and R01 DK017609 (to

D.A.S.).

284

by developing specific antifibrotic strategies.9,10 Atpresent, however, therapeutic options in humans arequite limited.7,11

Hepatic cell therapy could be an alternative strategyto generate new functional liver parenchyma in the cir-rhotic liver. Stem/progenitor cells—characterized bytheir high proliferative capacity, ability to differentiateinto different lineages, and ability to reconstitute tissuemass12—can be isolated from developing or adult liver,as well as from extrahepatic tissues, and can be trans-planted into normal or preconditioned recipientliver.13-17 To date, rat fetal liver stem/progenitor cells(FLSPCs) exhibit the most favorable characteristics foreffective liver repopulation by cells transplanted intothe (near-)normal liver.13,17-21 Liver repopulation byFLSPCs under nonselective conditions requires onlypartial hepatectomy (PH).13,19 This cell type, there-fore, may represent an excellent resource for restoringhepatocyte mass in a diseased liver environment.

In the present study, we transplanted FLSPCs anddemonstrated that epithelial stem/progenitor cells canengraft, proliferate, and differentiate into hepatocytesin the recipient liver with advanced fibrosis/cirrhosis.Surprisingly, transplantation of FLSPCs leads to con-siderable liver repopulation without the need for PHand reduces active fibrogenesis and net fibrosis. Incomparison, mature hepatocytes also repopulate thethioacetamide (TAA)-induced fibrotic liver, but to alesser extent than FLSPCs. Our model system, there-fore, represents an excellent tool to study novel celltransplantation strategies and to elucidate basic mecha-nisms necessary for successful tissue replacement, criti-cal for development of useful protocols to treatpatients with advanced liver diseases.

Materials and MethodsAnimals and TAA Administration. Pregnant

DPPIV1 F344 rats were purchased from CharlesRiver. F344-Tg(EGFP) F455/Rrrc rats were obtainedfrom the Rat Resource and Research Center of theUniversity of Missouri-Columbia and used to providetime pregnant EGFP1 F344 rats. Male DPPIV2

F344 rats were provided by the Liver Research Center,Albert Einstein College of Medicine (AECOM).

Two hundred mg/kg bodyweight (b.w.) TAA wasinjected intraperitoneally into DPPIV2 F344 rats (1.5to 2 months of age) twice weekly for up to 3 monthsprior to cell transplantation, followed by 100 or 200mg/kg b.w. TAA twice weekly after cell infusion. Allanimal studies were conducted under protocolsapproved by the Institutional Animal Care and UseCommittees of AECOM and University of Pittsburghin accordance with National Institutes of Health(NIH) guidelines.

Isolation of Fetal Liver Cells and Hepatocytes.Unfractionated fetal liver cells were isolated fromED14/15 fetal livers of pregnant DPPIV1 orDPPIV1/EGFP1 F344 rats, as described.18,19 Hepato-cytes were isolated from livers of adult DPPIV1 F344rats. Detailed information concerning the cell isolationprocedures can be found in the Supplemental Materi-als and Methods of Ref. 21.

Cell Transplantation and Liver Repopulation.Fetal liver cells (viability >95%) or adult hepatocytes(viability >80%) were transplanted through the portalvein into DPPIV2 F344 rats22 treated with TAA oruntreated recipients with or without 2/3 PH (hepatec-tomized liver lobes were used to assess liver fibrosisand for other studies). After rats were sacrificed at dif-ferent times following cell transplantation, liver repo-pulation was determined by enzyme histochemistry forDPPIV, as described.18,19 For engraftment studies,transplanted fetal liver cells were detected by immuno-histochemistry for EGFP.

Reverse-Transcription Polymerase Chain Reaction(RT-PCR) and Quantitative (q)RT-PCR Analysis.Total RNA was extracted from snap-frozen liver tissuederived from TAA-treated DPPIV2 F344 rats anduntreated age-matched control rats. Qualitative RT-PCR analyses were performed at least twice. Quantita-tive real-time RT-PCR was performed in doublet/trip-licate, as described in the Supplemental Materials andMethods of Ref. 21. A list of the primers is shown inSupporting Table 1.

Histochemistry and Immunohistochemistry. Infor-mation concerning histochemical and immunohisto-chemical analyses can be found in the SupportingMaterials and Methods.

Address reprint requests to: Michael Oertel, Ph.D., University of Pittsburgh, School of Medicine, Dept. of Pathology, 200 Lothrop St., BST S-404, Pittsburgh,PA 15261. E-mail: mio19@pitt.edu; fax: 412-648-1916.

Copyright VC 2013 by the American Association for the Study of Liver Diseases.View this article online at wileyonlinelibrary.com.DOI 10.1002/hep.26615Potential conflict of interest: Nothing to report.Additional Supporting Information may be found in the online version of this article.

HEPATOLOGY, Vol. 59, No. 1, 2014 YOVCHEV ET AL. 285

Hydroxyproline (HYP) Determination. Usingtwo different fragments per liver, the HYP content wasdetermined biochemically, as described.23

Microscopy and Imaging. Tissue slides were exam-ined under an AxioObserver Z1 microscope. Imageswere obtained with an AxioCam ICc3, ICm1, or HRccamera and processed with AxioVision 4.8 or ZENimaging software (Carl Zeiss MicroImaging).

Data Analysis. Data were analyzed using Sigma-Stat 2.01 (SPSS Scientific), GraphPad Prism5 (Graph-Pad), and NIS-Elements D (Nikon) software and arereported as mean 6 SEM.

ResultsCharacterization of Hepatic Fibrosis/Cirrhosis

After TAA Treatment. After chronic TAA administra-tion (200 mg/kg b.w., twice weekly), liver fibrosis wasassessed (Fig. 1) using the Laennec classificationsystem.24 At 6 weeks, the progressive liver injury pro-duces moderate fibrosis and mild cirrhosis, i.e., pre-

dominant nodularity caused by narrow fibrous septabridging portal areas. By 3 months, more advancedfibrosis occurs, leading to moderate to severe cirrhosisin different parts of the liver. Previously narrow septahave thickened; many nodules are surrounded bybroad septa and in some areas extensive tracts offibrous tissue contain a few larger nodules and smallregions of fragmented nodules. In addition, immuno-histochemistry for alpha smooth muscle actin (a-SMA)in liver sections of TAA-treated rats, compared to non-treated control rats, showed increasing numbers of a-SMA-positive cells in intralobular septa in areas offibrosis, indicating activation of hepatic stellate cells.

To determine the expression levels of genes relevantto advanced fibrosis/cirrhosis, we performed RT-PCRanalysis in liver tissues 3 months after TAA administra-tion compared to age-matched nontreated liver (Fig.2A,B). We observed elevated expression of a-SMA, pla-telet-derived growth factor receptor b (PDGFRb), des-min, neural cell adhesion molecule (N-CAM), andvimentin messenger RNA (mRNA) (Fig. 2A). In

Fig. 1. Liver fibrosis in TAA-treated rats. 200 mg/kg TAA was injected intraperitoneally into mutant DPPIV2 F344 rats twice weekly for up to 3months. (A) Normal rat liver and liver at 3 months after administration of TAA are shown. (B) Pathologic changes at 6 weeks and 3 months weredetermined using Masson’s trichrome staining, immunohistochemistry for a-SMA, and hematoxylin and eosin (H&E) staining. Tissue samplesderived from nontreated rats were used as controls. Original magnification, 350 (upper and middle panels), 3200 (lower panels).

286 YOVCHEV ET AL. HEPATOLOGY, January 2014

addition, after induction of advanced fibrosis/cirrhosis,procollagen a2(I) (Col1a2), matrix metalloproteinase-2(MMP-2), MMP-9, tissue inhibitor ofmetalloproteinase-1 (TIMP1), TIMP2 were up-regulated, and glial fibrillary acidic protein (GFAP) wasdown-regulated (Fig. 2A). The expression profiles ofthese genes clearly reflect activation of stellate cells andongoing fibrogenesis. Furthermore, biochemical analysisof the relative HYP content in nontreated versus TAA-treated liver (n 5 6/6 rats) increased from 0.23 6 0.02to 1.38 6 0.18 mg HYP/g liver, indicating advancedfibrosis/cirrhosis at 3 months after TAA administration.

We observed increased expression of AFP, Dlk-1,CD133, Sox-9, FoxJ1, and nestin mRNAs (Fig. 2A),indicating increased numbers of progenitor cells14,16,25-28

after induction of advanced fibrosis/cirrhosis. Comparedto normal hepatic tissue, liver samples with advanced

fibrosis also showed down-regulation of glucose-6-phosphatase (G6Pase), asialoglycoprotein receptor(ASGPR), and cytochrome 3A1 (CYP3A1) mRNAs (allof which are related to hepatocyte-specific cell func-tions), indicating hepatocellular damage or loss. In con-trast, biliary epithelial cell-specific genes (cytokeratin-19[CK-19], connexin43, EpCAM) were up-regulated instrongly fibrotic liver (Fig. 2A). These data were con-firmed and fold changes quantified by qRT-PCR analy-sis for selected genes (Fig. 2B).

Repopulation Studies in Rats With TAA-InducedLiver Fibrosis Transplanted With Hepatic Cells. Ina pilot experiment, we tested whether fetal liver cellsare capable of repopulating the fibrotic liver. To inducemoderate hepatic fibrosis, 200 mg/kg TAA wasinjected into DPPIV2 F344 rats twice weekly for 6weeks, followed by a maintenance dose of 100 mg/kgTAA after cell transplantation. Since repopulation ofthe normal liver by FLSPCs occurs only after two-thirds PH,13,19 PH was performed just prior to cellinfusion (�1.5 3 107 ED14 unfractionated fetal livercells, of which �2.5% are AFP1/CK-191 bipotentialstem/progenitor cells17,22). At both 1 and 2 monthsafter cell transplantation, we observed extensive liverrepopulation with more than 50% tissue replacementin many areas of TAA-treated recipient liver (n 5 3rats) (Fig. 3A, upper left and middle panel). In con-trast, only small cell clusters were observed in age-matched rats that did not receive TAA, with very littleliver repopulation at 1 month (n 5 2) (Fig. 3A, upperright panel).

Having demonstrated that transplanted fetal hepaticcells can repopulate a liver with moderate fibrosis, wenext tested whether cell transplantation is feasible inrecipient rats with advanced fibrosis. After inducingadvanced liver fibrosis in DPPIV2 F344 rats (200 mg/kg TAA, twice weekly for 10 weeks; followed by 100mg/kg TAA after cell transplantation), we infused�1.5 3 107 ED14 fetal liver cells into TAA-treatedrats in conjunction with PH. At 2 months after celltransplantation (n 5 3), we observed small and largeDPPIV1 cell clusters in host livers with extensivefibrosis. Many repopulating cell clusters encompassedentire fibrotic lobules (Fig. 3A, lower left panel).Although many areas showed extensive liver repopula-tion with multiple adjacent DPPIV1 regenerating nod-ules, other areas showed only limited repopulation.The majority of transplanted FLSPCs differentiatedinto hepatocytic cells; however, substantial bile ductgeneration, mainly within the fibrotic bands, was alsoobserved (Fig. 4B, below). Furthermore, we trans-planted FLSPCs into TAA-treated rats without PH

Fig. 2. Gene expression in TAA-induced liver fibrosis. (A) Liver RNAextracts from two TAA-treated rats at 3 months after starting TAAadministration compared to two age-matched nontreated rats wereanalyzed for mRNA expression. These experiments were performed atleast twice. (B) Quantitative RT-PCR analysis of selected genes infibrotic liver. Values are mean 6 SEM of liver samples from TAA-treated rats (n 5 4), expressed as fold differences in gene expressioncompared to age-matched nontreated rats (n 5 4). One representativeexperiment for each gene, from at least two replicate experiments, isshown. qRT-PCR analysis was also performed for AFP, Dlk-1, Sox-9,and FoxJ1, expressed in all four fibrotic liver samples (Ct valuesranged from 29 to 33) were undetectable in normal liver.

HEPATOLOGY, Vol. 59, No. 1, 2014 YOVCHEV ET AL. 287

and normal rats without PH (n 5 4/2) and observedscattered repopulation clusters in the fibrotic rat livers.Some of these clusters were of large size (Fig. 3A,lower middle panel), in contrast to normal rats with-out PH in which no liver repopulation was achievedby FLSPCs (Fig. 3A, lower right panel).

Although a limiting factor in liver repopulationmight be the ability of hepatocytes, which are of largesize, to engraft in the fibrotic liver tissue,29 we investi-gated the repopulation potential of differentiatedmature hepatic cells in the TAA fibrosis model. Hepa-tocytes were infused into rats with advanced liverfibrosis/cirrhosis (produced by administration of 200mg/kg TAA, twice weekly for 10-12 weeks; followedby 100 mg/kg TAA after cell transplantation). In twoTAA-treated rats transplanted with �1.5 or 2 3 106

hepatocytes in conjunction with PH, DPPIV1 hepato-cytic clusters were observed in both rats at 2 months,remarkably with up to 10% liver repopulation in therat transplanted with �2 3 106 hepatocytes (Fig. 3B,left panel). In addition, we transplanted �2 or 5 3

106 hepatocytes into TAA-treated rats without PH(n 5 5). Small and larger repopulating hepatocyte

clusters were seen in all rats with advanced fibrosis/cir-rhosis (Fig. 3B, middle panel). In contrast, normaluntreated rats transplanted with similar numbers ofhepatocytes without PH (�5 3 106 cells; n 5 3)showed only single cells in the parenchyma, withoutcluster formation or significant liver repopulation (Fig.3B, right panel).

Comparative Repopulation Studies in Rats WithAdvanced Fibrosis/Cirrhosis Transplanted WithFLSPCs Versus Mature Hepatocytes. For definitivelong-term repopulation studies under the most strin-gent fibrosis conditions, we infused cells into rats at 3months after starting TAA administration (200 mg/kg)and continued with the same TAA dose after cell infu-sion. In these experiments we attempted to inoculateequivalent numbers of hepatic epithelial lineage cells,based on our previous analysis showing that“bipotential” stem/progenitor cells (AFP1/CK-191)represent �2.5% of total unfractionated ED14 fetalliver cells.17 We therefore estimated that 8 3 107

unfractionated fetal liver cells contained �2 3 106

“bipotential” FLSPCs, comparable to 2 3 106 maturehepatocytes. To obtain sufficient numbers of cells for

Fig. 3. Enhanced repopulation of fibrotic liver by transplanted ED14 FLSPCs and adult hepatocytes. DPPIV enzyme histochemistry showingrepopulation by unfractionated wild-type ED14 FLSPCs (�1.5 3 107, A) and adult hepatocytes (�2 3 106, B, left panel; �5 3 106, B, mid-dle and right panel) transplanted into DPPIV2 TAA-treated rats at 6, 10, or 12 weeks after initiating TAA administration (200 mg/kg TAA, fol-lowed by a maintenance dose of 100 mg/kg TAA twice weekly after cell transplantation) versus nontreated normal rats in conjunction with (1)or without PH (2). Rats were sacrificed at 1 or 2 months (mo) after cell infusion. Single cells/small cell clusters are highlighted by arrowheads(A,B, right panels). Original magnification, 350.

288 YOVCHEV ET AL. HEPATOLOGY, January 2014

these studies, we isolated unfractionated hepatic stem/progenitor cells from ED15 fetal livers.

While maintaining the TAA dose after cell trans-plantation into advanced fibrotic rat liver (Figs. 4, 5),levels of 35.7 6 6.4% and 40.8 6 10.3% repopula-tion were achieved with FLSPCs at 2 and 4 months,respectively (n 5 4/4). FLSPCs differentiated intohepatocytes (Fig. 4A) and bile duct cells. The largeDPPIV1 clusters of hepatocytes typically had DPPIV1

bile ducts along the edges of fibrous septae (Fig. 4). Insome cases, DPPIV1 bile ducts extended into

surrounding DPPIV-negative regions (Fig. 4B), pre-sumably resulting from a stimulus for bile duct prolif-eration in the injured liver. The cells formed largeDPPIV1 clusters with extensive tissue replacement(Fig. 4C,E). In comparison, substantial numbers oftransplanted mature hepatocytes engrafted in the cir-rhotic liver, proliferated long-term, and replaced dis-eased liver mass (Fig. 4A, right panels, 4D). However,liver repopulation levels with mature hepatocytes werelower at 2 and 4 months after cell transplantation (8.36 2.0% and 10.5 6 3.2%, respectively; n 5 3/4)

Fig. 4. Repopulation of transplanted cells in host liver with progressing fibrosis/cirrhosis. Donor cells (�8 3 107 fetal liver cells versus �23 106 hepatocytes) were infused in conjunction with PH into rats at 3 months after starting TAA administration, which was continued at thesame dose thereafter (200 mg/kg, twice weekly). Liver repopulation was analyzed at 2 (A-C) and 4 months (D,E) after cell transplantation.ED15 FLSPCs differentiated into hepatocytes (A, lower left panel) and bile ducts (B). Original magnification, 350 (A, upper panels), 3200 (A,lower panels; B). (C-E) Whole liver sections with the highest repopulation level observed at 2 and 4 months after cell transplantation are shown.Images contain 30, 41, or 64 merged adjacent microscopic fields (original magnification, 350).

HEPATOLOGY, Vol. 59, No. 1, 2014 YOVCHEV ET AL. 289

compared to that obtained with FLSPCs (35.7 6

6.4% and 40.8 6 10.3%, respectively). Although therewas higher repopulation with transplanted stem/pro-genitor cells, which indicates a higher engraftment orproliferation rate, our findings with mature hepato-cytes also represent a significant new observation inthe fibrotic liver.

Simultaneous immunohistochemical analysis forDPPIV (CD26) and a-SMA (Fig. 5A) showed thatDPPIV1 cell clusters derived from transplantedFLSPCs completely replaced host hepatocytes withinliver nodules surrounded by fibrous host tissue con-taining a-SMA1 cells (Fig. 5A, left panels), a phenom-enon also observed after hepatocyte transplantation

Fig. 5. Expansion of transplanted ED15 FLSPCs and hepatocytes transplanted in conjunction with PH into recipient rats with progressing fibro-sis/cirrhosis. (A) Localization of transplanted cell clusters in recipient liver with advanced fibrosis/cirrhosis using simultaneous immunohistochem-istry for DPPIV (CD26) and a-SMA. (B) Detection of proliferating cells in normal, fibrotic, and recipient liver after cell transplantation, usingcostaining for CD26 and Ki-67. (C) Functional integration of transplanted cells in cirrhotic liver. Normal lobular distribution and gradients of glu-cose-6-phosphatase expression (C, upper right panel) in untreated wt DPPIV1 F344 rat liver (C, upper left panel) compared to cirrhotic liver after3 months TAA administration (C, right panel, 2nd row) in mutant F344 rats negative for DPPIV (C, left panel, 2nd row). The panels in C, 3rd to6th row, compare DPPIV and G6Pase staining of nearby liver sections. Where transplanted cells have proliferated and generated new hepato-cytes, large DPPIV1 cell clusters showed normal expression of glucose-6-phosphatase. Panels in the 4th row demonstrate that transplantedstem/progenitor cells differentiate into both DPPIV1 bile ducts (asterisks) and hepatocytes, but only hepatocytic clusters express glucose-6-phosphatase. Original magnification, 350 (A, upper panels; C), 3100 (A, lower panels; B).

290 YOVCHEV ET AL. HEPATOLOGY, January 2014

(Fig. 5A, upper right panel). Double-label immunohis-tochemistry for DPPIV (CD26) and Ki-67 (Fig. 5B)showed that FLSPC and hepatocyte-derived cell clus-ters contained actively proliferating cells for up to 4months (Fig. 5B, middle and lower panels) and“competed” with proliferating host hepatocytes (Fig.5B, upper right panel). Furthermore, DPPIV andG6Pase expressing hepatocytic cells were detected at 2and 4 months after transplantation of FLSPCs orhepatocytes (Fig. 5C), demonstrating hepatocyte-specific metabolic activity of transplanted cells.

Engraftment and Repopulation of the Liver WithAdvanced Fibrosis/Cirrhosis After Transplantationof Fetal Liver Cells in the Absence of PH. Since weshowed that FLSPCs can form cell clusters in thefibrotic liver without PH (Fig. 3A), we performedadditional cell transplantations to determine whethersubstantial liver repopulation could be obtained underthese conditions. These studies required infusion of�1.5 3 108 donor cells into DPPIV2 rats withoutPH at 3 months after TAA administration. At4 months after cell transplantation, 23.8 6 4.4% liverrepopulation was achieved and G6Pase-expressing, dif-ferentiated cells were integrated in the cirrhotic liverenvironment. Compared to nontransplanted TAA-treated livers, analysis of mRNA showed that FLSPCtransplantation up-regulated genes related to specifichepatocytic functions (G6P, 3.7-fold; CYP3A1, 3.5-fold; TAT, 1.8-fold; n 5 3/3) (Supporting Figure 1).

The above findings suggested that a significant num-ber of FLSPCs can engraft in the cirrhotic liver andsubstantial repopulation can be achieved in the absenceof PH. FLSPC transplantation under these conditionsis well tolerated and we observed a mortality of 11%.

We next studied the dynamics of donor cell engraft-ment and expansion immediately after FLSPC infu-sion. Since DPPIV is not expressed before ED18,ED15 FLSPCs were isolated from EGFP-marked

transgenic F344 rats to identify engrafted cells. Weinfused �1.5 3 108 EGFP-expressing fetal liver cellsinto three DPPIV2 rats at 3 months after TAA admin-istration without PH. At days 1 and 3 after cell infu-sion, single EGFP1 cells (Fig. 6A) and small groups ofEGFP1 cells (Fig. 6B) were detected in the host liverparenchyma, respectively, demonstrating successfulengraftment of transplanted stem/progenitor cells intothe cirrhotic liver. By day 7, expanding fetal liver cellsformed small cell clusters primarily along the borderof fibrotic bands (Fig. 6C), demonstrating ongoingrepopulation in the cirrhotic liver tissue environment.

Effect of Stem/Progenitor Cells on Fibrogene-sis. Having demonstrated that transplanted fetalhepatic cells can engraft and significantly repopulatethe recipient liver with advanced fibrosis/cirrhosis, wenext determined whether stem/progenitor cells caneffect fibrogenesis and the extent of liver fibrosis. Afterinducing advanced fibrosis in DPPIV2 rats (200 mg/kg TAA, twice weekly for 3 months), we infused�1.8 3 108 unfractionated ED15 fetal liver cells intoTAA-treated rats that had not undergone PH (n 5 6).Two months later, TAA administration was discontin-ued and rats were sacrificed 5 weeks later. Other ratsreceived identical TAA-treatment without cell trans-plantation (n 5 6). Repopulation analysis of the celltransplant recipients showed that 26.9 6 6.3% of theliver mass was repopulated by FLSPC-derived hepato-cytes that expressed albumin at the same level asobserved in adjacent host liver tissue (Fig. 7A,B).Selective expression of glutamine synthetase in the cen-trilobular regions of engrafted liver tissue suggestedcomplete zonal differentiation by repopulating FLSPC-derived hepatocytes (Fig. 7B, lower panels). Doublelabel immunohistochemistry for CD26 and CK-19/EpCAM demonstrated that transplanted stem/progeni-tor cells also differentiated into bile duct cells(Fig. 7C). Furthermore, a marked decrease of a-SMA,

Fig. 6. Engraftment and early expansion of ED15 FLSPCs transplanted into recipients with advanced fibrosis/cirrhosis without PH. Donor fetalliver cells (�1.5 3 108) were infused into rats at 3 months after starting TAA administration (200 mg/kg, twice weekly). Single EGFP1 cellswere detected at day 1 (A; arrowheads), which divided at day 3 (B), and formed small cell clusters by day 7 (C). Original magnification, 3100(A-C), 3630 (insets).

HEPATOLOGY, Vol. 59, No. 1, 2014 YOVCHEV ET AL. 291

PDGFRb, desmin, vimentin, TIMP1, and TIMP2mRNA was observed in the FLSPC-transplanted group(Fig. 8A), indicating a dramatic reduction of thefibrinogenic process. This also correlated with amarked reduction in activated a-SMA-positive stellatecells (Fig. 8B). Histological examination of Sirius Red-stained sections showed more macronodules in FLSPCrecipients, i.e., intact regions without internal fibrosis

(Fig. 8C). Quantification of Sirius Red-stained colla-gen showed less collagen in the liver of FLSPC recipi-ents compared to nontransplanted rats (15.4 6 2.8%versus 19.6 6 0.5%; Fig. 8C, right panel; see alsoCol1a1 mRNA levels in Fig. 8A), although thesechanges were not statistically significant.

Discussion

Several rodent models of cirrhosis have been estab-lished to study the mechanism of fibrosis progressionor antifibrotic therapies (reviewed30). To develop a celltransplantation model for epithelial stem/progenitorcells in a cirrhotic recipient background, we inducedfibrosis/cirrhosis in the mutant DPPIV2 F344 rat,31

an inbred strain originally used to follow the fate oftransplanted wild-type DPPIV1 hepatocytes inDPPIV2 recipients.32 TAA-induced liver fibrosis wasselected in preference to CCl4 and other known fibro-sis models because it produces more extensive and sta-ble fibrosis and is most similar to human fibrosis inclinical progression.30,33,34 We demonstrated thatadvanced fibrosis/cirrhosis was established at 3 monthsafter chronic TAA administration, indicated by charac-teristic hepatic lesions and collagen deposition.33 Thecirrhotic liver showed increased HYP, a-SMA,PDGFRb, procollagen, TIMP1, and MMP-2, indicat-ing increased numbers of activated stellate cells andongoing fibrogenesis,8,35,36 and decreased GFAP, whichis down-regulated in activated stellate cells in advancedfibrosis.37 Advanced fibrosis/cirrhosis in the recipientliver was further supported by decreased levels ofunique hepatocyte-specific mRNA transcripts (e.g.,ASGPR, CYP3A1, and G6Pase mRNA) (see also Fig.5C). Finally, an increased number/activation of cholan-giocytes, which secrete fibrogenic growth factors andactivate stellate cells in fibrotic/cirrhotic liver,30 wasreflected by augmented CK-19, connexin43, andEpCAM levels in TAA-treated liver.

Using the TAA-induced experimental model of liverfibrosis/cirrhosis, we made five major observations.First, we showed that rat fetal liver-derived epithelialstem/progenitor cells can engraft into the recipientliver with advanced fibrosis/cirrhosis and differentiateinto hepatocytes, i.e., cells with hepatocyte-specificmorphology and metabolic function. Second, theengrafted cells expand and replace failing liver masswithin a short time after cell infusion. Third, efficientliver repopulation by transplanted epithelial stem/pro-genitor cells can be achieved in a densely fibrotic liverwithout an additional stimulus provided by liverregeneration. Fourth, the engrafted liver exhibited

Fig. 7. Repopulation and functional incorporation of differentiated,donor-derived cells after transplantation without PH. ED15 fetal livercells (�1.8 3 108 cells) were transplanted into DPPIV2 rats 3months after starting TAA administration (A). Images contain 65 or 25tiled microscopic fields (original magnification, 350 and 3100). (B)Patterns of DPPIV (CD26) and albumin expression, and DPPIV and glu-tamine synthetase (GS) expression in consecutive liver sections dem-onstrate that DPPIV1 cells differentiated into hepatocytes and becomeincorporated into morphologically normal hepatocyte plates. (C) Trans-planted FLSPCs showed a canalicular pattern of CD26 staining charac-teristic of mature hepatocytes and also differentiated into bile ductscoexpressing CD26 with CK-19 or EpCAM (see arrows). Original magni-fication, 310 and 35 (B), 340 (C).

292 YOVCHEV ET AL. HEPATOLOGY, January 2014

reduced fibrinogenic activity. Fifth, we also showedthat transplanted mature hepatocytes proliferate andform large cell clusters with substantial tissuereplacement.

Rodent fibrosis models are crucial to investigate theefficiency of antifibrotic agents.30 Since it is impossible

to distinguish between the antiinflammatory and anti-fibrotic effects of agents tested in hepatotoxin-inducedfibrosis models, carbon tetrachloride (CCl4) or TAA isgenerally withdrawn during drug administration andthe rate of fibrosis recovery is determined to assess theeffectiveness of the tested treatment.30 Because the

Fig. 8. Effect of FLSPC transplantation on hepatic fibrogenesis. ED15 fetal liver cells (�1.8 3 108 cells) were transplanted into DPPIV2 rats3 months after starting TAA administration, which was continued for 2 months thereafter. Five weeks later rats were sacrificed. (A) QuantitativeRT-PCR analysis for mRNA of genes for stellate cell activation and fibrogenesis. Values are mean 6 SEM of liver samples from FLSPC-transplanted fibrotic rats (n 5 6) or nontransplanted fibrotic rats (n 5 6) compared to age-matched normal control rats, set at a value of 1 (n5 3). One representative example from at least three replicate experiments is shown. (B) Immunohistochemical detection of a-SMA-positive cellsin rat livers with (left) or without (right) cell transplantation. Images contain 25 adjacent microscopic fields (original magnification, 3100). (C)Selected areas of Sirius Red-stained tissue sections of rat livers with (left) or without (middle) cell transplantation. Quantification of Sirius Red-stained collagen (right). Values are mean 6 SEM of whole liver sections (n 5 6/6).

HEPATOLOGY, Vol. 59, No. 1, 2014 YOVCHEV ET AL. 293

main focus of the present study was to assess whethertransplanted epithelial stem/progenitor cells can restorehepatic parenchyma in a chronically injured liver envi-ronment during evolution of fibrosis/cirrhosis, we con-tinued TAA administration after cell infusion. Then,to evaluate whether transplanted FLSPCs have an anti-fibrotic effect, in some studies we discontinued theTAA administration after successful cell engraftmentand repopulation.

Potential obstacles to effective repopulation of fibrotictissue include infarction of the liver by infused cells orpoor engraftment of transplanted cells. Indeed, fibroticrats infused through the portal vein with 5 3 106 hepa-tocytes in conjunction with PH died within 48 hours(n 5 3). Infusion of 2 3 106 cells was better tolerated,although a noticeable mortality was still observed (datanot shown). Rat FLSPCs are much smaller than adulthepatocytes (10-12 lm versus 20-35 lm diameter,respectively13; human fetal cells15), which allowed us toinfuse high numbers of unfractionated fetal liver cells (83 107 or 1.5 3 108 cells, contains �2 3 106 or 4 3

106 “bipotential” FLSPCs, respectively), with or with-out PH. Importantly, a preliminary study of FLSPCsenriched by immunomagnetic bead cell sorting showedthat we can significantly increase the number ofFLSPCs transplanted without increasing the total cellnumber infused (see Supporting Figure 2).

Previously, we have demonstrated that FLSPCs caneffectively repopulate the (near-)normal liver, but onlyin conjunction with PH,13,19 suggesting that PH isrequired for their engraftment and/or expansion.19

However, the present study showed substantial earlyengraftment and efficient repopulation after FLSPCinfusion into the TAA-treated recipient liver withoutPH. These results suggest that chronic injury duringevolution of cirrhosis, or the altered cirrhotic livermicroenvironment, favors engraftment and prolifera-tion of transplanted epithelial stem/progenitor cells.However, to achieve long-term correction of cirrhosisafter hepatic stem cell transplantation, additional mod-ifications of the microenvironment may be necessary.38

During the past 2 decades, several model systemshave been developed to study liver repopulation bytransplanted hepatic cells (reviewed17). In these mod-els, substantial liver replacement by transplanted wild-type hepatocytes was achieved by genetic modificationsof the host liver causing massive hepatic injury39,40 orby inhibition of the proliferative capacity of host hepa-tocytes through administration of DNA-damagingagents or liver irradiation.41-43 Liver replacement wasobserved in the a1-antitrypsin deficient transgenicmouse, in which the proliferation of endogenous

hepatocytes is impaired.44 These repopulation modelsare characterized by a strong growth advantage oftransplanted cells compared to host hepatocytes.Although previous studies demonstrated increased sur-vival of rats with decompensated liver cirrhosis afterintrasplenic hepatocyte transplantation,45 to ourknowledge there is no previous report showing signifi-cant hepatic tissue replacement by transplanted epithe-lial stem/progenitor cells in an experimental model ofadvanced liver fibrosis/cirrhosis.

There are currently only a few pioneering humanstudies of mature or fetal hepatic cell transplantationsin patients with chronic liver diseases.46-48 Neverthe-less, animal studies must provide critical understandingof the basic requirements and mechanisms for effectiveliver repopulation. In the present study, using experi-mental conditions that reflect circumstances similar tohuman fibrosis/cirrhosis, we demonstrated that trans-planted progenitor cells can efficiently proliferate aftertheir engraftment and are capable of differentiatinginto hepatic cell lineages. In conjunction with replace-ment of 20%-30% of hepatocytic mass by FLSPCs,hepatic fibrogenesis was reduced, as evidenced byreduced stellate cell activation, decreased expression offibrogenesis genes, and reduced collagen in the tissue.Thus, transplantation of epithelial stem/progenitor orFLSPC-like cells engineered by way of iPS cell tech-nology, perhaps combined with targeted antifibrotictherapy, holds great promise for treatment of patientswith endstage liver diseases.

Acknowledgment: The authors thank Dr. Scott L.Friedman (Division of Liver Diseases, Mount SinaiSchool of Medicine, New York, NY) for advice inusing the TAA-induced fibrosis model and Ms.Amanda Franklin for excellent technical assistance.

Author contributions: M.I.Y. and Y.X. carried outexperiments and analyzed data. D.A.S. contributed tothe experimental design and data analyses. J.L. per-formed histological subclassification of fibrosis/cirrho-sis. M.O. designed the studies and performedexperiments, analyzed data, and wrote the article. Allauthors read and commented on the article.

References

1. Xu J, Kochanek KD, Murphy SL, Tejada-Vera B. Deaths: final data for2007. Natl Vital Stat Rep 2010;58:1-135.

2. Schuppan D, Afdhal NH. Liver cirrhosis. Lancet 2008;371:838-851.3. Cardenas A, Gines P. Management of patients with cirrhosis awaiting

liver transplantation. Gut 2011;60:412-421.4. Perera MT, Mirza DF, Elias E. Liver transplantation: issues for the next

20 years. J Gastroenterol Hepatol 2009;24:S124-S131.

294 YOVCHEV ET AL. HEPATOLOGY, January 2014

5. http://optn.transplant.hrsa.gov/latestData/rptData.asp6. Hernandez-Gea V, Friedman SL. Pathogenesis of liver fibrosis. Annu

Rev Pathol 2011;6:425-456.7. Henderson NC, Iredale JP. Liver fibrosis: cellular mechanisms of pro-

gression and resolution. Clin Sci (Lond) 2007;112:265-280.8. Friedman SL. Mechanisms of hepatic fibrogenesis. Gastroenterology

2008;134:1655-1669.9. Schuppan D, Popov Y. Rationale and targets for antifibrotic therapies.

Gastroenterol Clin Biol 2009;33:949-957.10. Kisseleva T, Brenner DA. Anti-fibrogenic strategies and the regression

of fibrosis. Best Pract Res Clin Gastroenterol 2011;25:305-317.11. Friedman SL. Evolving challenges in hepatic fibrosis. Nat Rev Gastro-

enterol Hepatol 2010;7:425-436.12. Loeffler M, Potten CS. Stem cells and cellular pedigrees — a concep-

tual introduction. In: Potten CS, editor. Stem cells. London: AcademicPress; 1997. p 1-29.

13. Dabeva MD, Petkov PM, Sandhu J, Oren R, Laconi E, Hurston E,et al. Proliferation and differentiation of fetal liver epithelial progenitorcells after transplantation into adult rat liver. Am J Pathol 2000;156:2017-2031.

14. Suzuki A, Sekiya S, Onishi M, Oshima N, Kiyonari H, Nakauchi H,et al. Flow cytometric isolation and clonal identification of self-renewing bipotent hepatic progenitor cells in adult mouse liver. HEPA-

TOLOGY 2008;48:1964-1978.15. Schmelzer E, Zhang L, Bruce A, Wauthier E, Ludlow J, Yao HL, et al.

Human hepatic stem cells from fetal and postnatal donors. J Exp Med2007;204:1973-1987.

16. Cardinale V, Wang Y, Carpino G, Cui CB, Gatto M, Rossi M, et al.Multipotent stem/progenitor cells in human biliary tree give rise tohepatocytes, cholangiocytes, and pancreatic islets. HEPATOLOGY 2011;54:2159-2172.

17. Oertel M. Fetal liver cell transplantation as a potential alternative towhole liver transplantation? J Gastroenterol 2011;46:953-965.

18. Sandhu JS, Petkov PM, Dabeva MD, Shafritz DA. Stem cell propertiesand repopulation of the rat liver by fetal liver epithelial progenitor cells.Am J Pathol 2001;159:1323-1334.

19. Oertel M, Menthena A, Dabeva MD, Shafritz DA. Cell competitionleads to high level of normal liver reconstitution by transplanted fetalliver stem/progenitor cells. Gastroenterology 2006;130:507-520.

20. Oertel M, Menthena A, Chen YQ, Teisner B, Jensen CH, Shafritz DA.Purification of fetal stem/progenitor cells containing all the repopulationpotential for normal adult rat liver. Gastroenterology 2008;134:823-832.

21. Menthena A, Koehler C, Sandhu JS, Yovchev MI, Hurston E, ShafritzDA, et al. Activin A, p15INK4b signaling, and cell competition promotestem/progenitor cell repopulation of livers in aging rats. Gastroenterol-ogy 2011;140:1009-1020.

22. Oertel M, Shafritz DA. Liver repopulation by fetal liver stem/progeni-tor cells. In: Soto-Gutierrez A, Navarro-Alvarez N, Fox IJ, editors.Methods in bioengineering: cell transplantation. Boston: Artech House;2011. p 201-217.

23. Gomes AT, Bastos CG, Afonso CL, Medrado BF, Andrade ZA. Howvariable are hydroxyproline determinations made in different samplesof the same liver? Clin Biochem 2006;39:1160-1163.

24. Kim MY, Cho MY, Baik SK, Park HJ, Jeon HK, Im CK, et al. Histo-logical subclassification of cirrhosis using the Laennec fibrosis scoringsystem correlates with clinical stage and grade of portal hypertension. JHepatol 2011;55:1004-1009.

25. Sell S. Distribution of alpha-fetoprotein- and albumin-containing cellsin the livers of Fischer rats fed four cycles of N-2-fluorenylacetamide.Cancer Res 1978;38:3107-3113.

26. Jensen CH, Jauho EI, Santoni-Rugiu E, Holmskov U, Teisner B,Tygstrup N, et al. Transit-amplifying ductular (oval) cells and theirhepatocytic progeny are characterized by a novel and distinctive expres-sion of delta-like protein/preadipocyte factor 1/fetal antigen 1. Am JPathol 2004;164:1347-1359.

27. Yovchev MI, Grozdanov PN, Joseph B, Gupta S, Dabeva MD. Novelhepatic progenitor cell surface markers in the adult rat liver. HEPATO-

LOGY 2007;45:139-149.

28. Dorrell C, Erker L, Schug J, Kopp JL, Canaday PS, Fox AJ, et al. Pro-spective isolation of a bipotential clonogenic liver progenitor cell inadult mice. Genes Dev 2011;25:1193-1203.

29. Fox IJ, Roy-Chowdhury J. Hepatocyte transplantation. J Hepatol2004;40:878-886.

30. Popov Y, Schuppan D. Targeting liver fibrosis: strategies for develop-ment and validation of antifibrotic therapies. HEPATOLOGY 2009;50:1294-1306.

31. Thompson NL, Hixson DC, Callanan H, Panzica M, Flanagan D, Faris

RA, et al. A Fischer rat substrain deficient in dipeptidyl peptidase IV

activity makes normal steady-state RNA levels and an altered protein.

Use as a liver-cell transplantation model. Biochem J 1991;273:497-502.32. Rajvanshi PA, Kerr A, Bhargava KK, Burk RD, Gupta S. Studies of

liver repopulation using the dipeptidyl peptidase IV deficient rat and

other rodent recipients: cell size and structure relationships regulate

capacity for increased transplanted hepatocytes mass in the liver lobule.

HEPATOLOGY 1996;23:482-496.33. M€uller A, Machnik F, Zimmermann T, Schubert H. Thioacetamide-

induced cirrhosis-like liver lesions in rats—usefulness and reliability of

this animal model. Exp Pathol 1988;34:229-236.34. Jeong DH, Jang JJ, Lee SJ, Lee JH, Lim IK, Lee MJ, et al. Expression

patterns of cell cycle-related proteins in a rat cirrhotic model induced

by CCl4 or thioacetamide. J Gastroenterol 2001;36:24-32.35. Popov Y, Patsenker E, Bauer M, Niedobitek E, Schulze-Krebs A,

Schuppan D. Halofuginone induces matrix metalloproteinases in rat

hepatic stellate cells via activation of p38 and NFkappaB. J Biol Chem

2006;281:15090-15098.36. Safadi R, Ohta M, Alvarez CE, Fiel MI, Bansal M, Mehal WZ, et al.

Immune stimulation of hepatic fibrogenesis by CD8 cells and attenua-

tion by transgenic interleukin-10 from hepatocytes. Gastroenterology

2004;127:870-882.37. Niki T, De Bleser PJ, Xu G, Van Den Berg K, Wisse E, Geerts A. Com-

parison of glial fibrillary acidic protein and desmin staining in normal

and CCl4-induced fibrotic rat livers. HEPATOLOGY 1996;23:1538-1545.38. Liu L, Yannam GR, Nishikawa T, Yamamoto T, Basma H, Ito R, et al.

The microenvironment in hepatocyte regeneration and function in rats

with advanced cirrhosis. HEPATOLOGY 2012;55:1529-1539.39. Rhim J, Sandgren EP, Degen JL, Palmiter RD, Brinster RL. Replace-

ment of diseased mouse liver by hepatic cell transplantation. Science1994;263:1149-1152.

40. Overturf K, Al-Dhalimy M, Tanguay R, Brantly M, Ou CN, Finegold M,et al. Hepatocytes corrected by gene therapy are selected in vivo in a murinemodel of hereditary tyrosinaemia type I. Nat Genet 1996;12:266-273.

41. Laconi E, Oren R, Mukhopadhay DK, Hurston E, Laconi S, Pani P,et al. Long-term, near total liver replacement by transplantation of iso-lated hepatocytes. Am J Pathol 1998;153:319-329.

42. Witek RP, Fisher SH, Petersen BE. Monocrotaline, an alternative toretrorsine-based hepatocyte transplantation in rodents. Cell Transplant2005;14:41-47.

43. Guha C, Sharma A, Gupta S, Alfieri A, Gorla GR, Gagandeep S, et al.Amelioration of radiation-induced liver damage in partially hepatectom-ized rats by hepatocytes transplantation. Cancer Res 1999;59:5871-5874.

44. Ding J, Yannam GR, Roy-Chowdhury N, Hidvegi T, Basma H,Rennard SI, et al. Spontaneous hepatic repopulation in transgenic miceexpressing mutant human a1-antitrypsin by wild-type donor hepato-cytes. J Clin Invest 2011;121:1930-1934.

45. Kobayashi N, Ito M, Nakamura J, Cai J, Gao C, Hammel JM, et al.Hepatocyte transplantation in rats with decompensated cirrhosis. HEPA-

TOLOGY 2000;31:851-857.46. Fisher RA, Strom SC. Human hepatocyte transplantation: worldwide

results. Transplantation 2006;82:441-449.47. Fitzpatrick E, Mitry RR, Dhawan A. Human hepatocyte transplanta-

tion: state of the art. J Intern Med 2009;266:339-357.48. Khan AA, Shaik MV, Parveen N, Rajendraprasad A, Aleem MA,

Habeeb MA, et al. Human fetal liver-derived stem cell transplantationas supportive modality in the management of end-stage decompensatedliver cirrhosis. Cell Transplant 2010;19:409-418.

HEPATOLOGY, Vol. 59, No. 1, 2014 YOVCHEV ET AL. 295