Repeated transplantation of hepatocytes prevents fulminant hepatitis in a rat model of Wilson's disease

Authors


  • Vanessa Sauer and Ramsi Siaj were partly supported by the German Research Foundation (grant SCHM 964/10-1) and Innovative Medical Research (Münster, Germany).

Abstract

The outcome of consecutive hepatocyte transplants was explored in a rat model of Wilson's disease before the onset of fulminant hepatitis without preconditioning regimens. Rats received a high-copper diet in order to induce a rapid induction of liver failure. Sham-operated rats (15/15) developed jaundice and fulminant hepatitis, and they died within 4 weeks of first transplantation. Despite the continuation of a high dietary copper challenge, long-term survival was observed for a notable proportion of the transplanted animals (7/18). All survivors displayed normalized levels of hepatitis-associated serum markers and ceruloplasmin oxidase activity by posttransplant days 50 and 98, respectively. The liver copper concentrations, the liver histology, and the expression of marker genes were significantly restored within 4 months of transplantation in comparison with the control group. The high expression of a copper transporter gene (ATPase Cu++ transporting beta polypeptide) in the livers of the survivors indicated a high rate of repopulation by donor hepatocytes. Our data suggest that repeated cell transplantation can overcome the limitations of a single therapy session in rats with severe hepatic disease by functionally restoring the host liver without preconditioning. Liver Transpl 18:248–259, 2012. © 2011 AASLD.

Wilson's disease (WD) is an autosomal recessive genetic disorder caused by a mutation of the copper transporter gene ATPase Cu++ transporting beta polypeptide (ATP7B), which is located on chromosome 13.1, 2 More than 500 mutations of ATP7B have been described in patients. The diagnosis is established by clinical and biochemical characteristics and lately by gene analysis.3 WD is characterized by a disturbed copper metabolism with an accumulation of copper in the liver, which causes chronic and acute liver disease, and extrahepatic manifestations such as brain damage are common.4 The course of the disease can vary despite similar mutations of ATP7B.5 WD can be effectively treated with anticopper drugs, which reduce the copper load within the liver; however, noncompliance is a major problem.1 Although WD is a rare disease and treatment is available, WD can result in acute liver failure (ALF). It accounts for approximately 5% of ALF cases observed worldwide.6 Almost all patients with WD who develop ALF rapidly die when liver transplantation is not available.6, 7

The Long-Evans Cinnamon (LEC) rat is an inbred strain established from Long-Evans rats; it spontaneously develops fulminant hepatitis at the age of 80 to 120 days.8 Approximately 40% to 60% of the animals die of acute hepatic failure due to dietary copper found in commercial standard chow (approximately 7-15 mg/kg). The remaining animals develop chronic hepatitis and frequently hepatocellular carcinoma.9 LEC rats lack a functional form of Atp7b because of a partial deletion of the gene.10 High concentrations of copper have been observed in the livers of LEC rats, and they are accompanied by elevated levels of iron and zinc.8, 11 The LEC rat is a well-characterized model for gene and cell-based therapy12, 13 and has many characteristics that are also observed in patients (eg, low ceruloplasmin levels, a sensitivity to anticopper therapy, and abnormal p53 expression).8, 10, 14-16 However, abnormal p53 expression is rare in humans, and most patients harbor missense mutations rather than deletions within ATP7B.

In humans, hepatocyte transplantation has recently shown promising results, particularly for metabolic liver disorders caused by a single genetic defect.17-19 Hepatocyte transplantation has been established in various animal models.20-23 Within a few weeks, transplanted cells integrate into the liver parenchyma, exert normal differentiated hepatocyte function, and restore the function of the host liver. In these studies, 2 major issues have been observed that have to be overcome for efficient hepatocyte transplantation: (1) the limited proliferation of donor hepatocytes within the host liver and (2) the competition of donor cells with resident hepatocytes under conditions of stimulated cell growth. Therefore, the proliferation of donor cells has been stimulated in several animal models with genetic, chemical, or physical perturbations within the host liver before transplantation; this is a limitation in the clinical setting. The regimens used to treat recipient livers before transplantation, which are often called preconditioning, can be effectively induced by, for example, agents such as retrorsine that alkylate DNA and proteins, radiation, partial hepatectomy, and temporary liver embolization.24-27 Although these regimens have significantly helped with exploring and explaining important mechanisms involved in cell transplantation, the transfer of these treatments to the clinical setting is difficult because of possibly detrimental effects (eg, the oncogenic potential of drugs). A single hepatocyte transplant in LEC rats was shown to cure WD when a combination of treatments was used for preconditioning (eg, retrorsine and partial hepatectomy, radiation and ischemia reperfusion, or radiation and cholic acid).12, 28-30 In a study that did not involve host preconditioning, a partial amelioration of WD was observed when transplantation was performed in 1- to 2-week-old LEC rats. However, the full impact of cell transplantation on liver failure was difficult to assess in the study because the pups did not present with severe disease, 88% of the animals survived with treatment, and 84% of the animals survived without treatment.31

Previously, we have shown that ATP7B-expressing cells can be selected in tissue cultures.32 We reasoned that the repeated transplantation of functional hepatocytes into LEC rats might favorably modulate in vivo selection after a high-copper diet and overcome previously observed restrictions of a single therapy session and also increase the time for selection. So that we could mimic a clinical situation requiring liver transplantation, transplantation was performed when the animals developed fulminant liver failure. Our study suggests that repeated cell transplantation is effective and advantageous in comparison with a single therapy session and could ameliorate advanced stages of WD within a short time even without preconditioning.

Abbreviations:

α-SMA, alpha-smooth muscle actin; ACTA2, actin alpha 2 smooth muscle aorta; ALF, acute liver failure; ALT, alanine aminotransferase; ANGPT1, angiopoietin 1; AST, aspartate aminotransferase; ATP7B, ATPase Cu++ transporting beta polypeptide; CCND3, cyclin D3; cDNA, complementary DNA; CDKN1A, cyclin-dependent kinase inhibitor 1A; COL1A1, collagen type I alpha 1; CP, ceruloplasmin oxidase; Ct, cycle threshold; CYP2E1, cytochrome P450 2E1; Fah, fumarylacetoacetate hydrolase; FLT1, fms-related tyrosine kinase 1; HGF, hepatocyte growth factor; HMOX1, heme oxygenase (decycling) 1; HO-1, heme oxygenase 1; HPRT1, hypoxanthine phosphoribosyltransferase 1; ICAM1, intercellular adhesion molecule 1; iNOS, inducible nitric oxide synthase; KDR, kinase insert domain receptor; KITLG, KIT ligand; LEA, Long-Evans Agouti; LEC, Long-Evans Cinnamon; mRNA, messenger RNA; NOS2, nitric oxide synthase 2 inducible; PDGFR, platelet-derived growth factor receptor; PCR, polymerase chain reaction; SCF, stem cell factor; SREBF1, sterol regulatory element binding transcription factor 1; SREBP1, sterol regulatory element binding protein 1; TEK, TEK tyrosine kinase endothelial; TIE2, tyrosine kinase with immunoglobulin-like and epidermal growth factor–like domains 2; TIMP1, TIMP metallopeptidase inhibitor 1; VCAM1, vascular cell adhesion molecule 1; VEGF, vascular endothelial growth factor; VEGFR, vascular endothelial growth factor receptor; WD, Wilson's disease.

MATERIALS AND METHODS

Animals

Long-Evans Agouti (LEA) rats and LEC rats expressing wild-type Atp7b and a deletion mutant of Atp7b10 were kindly provided by S. Gupta (Albert Einstein University College of Medicine, Bronx, NY). All rats were maintained in a special animal care facility of the university on a 12-hour light/dark cycle. LEC rats were bred from heterozygote females and homozygote males and were genotyped by polymerase chain reaction (PCR) as previously described.33 The occurrence of severe jaundice with severe behavioral disturbances and weight loss (>20%) was accompanied by coma and was followed by the sacrifice of the animals. All procedures and treatments for the animals were approved by local authorities.

Hepatocyte Isolation

Hepatocytes were isolated from livers of male and female LEA rats with a standard collagenase in situ perfusion technique (type 4, Worthington Biochemical) with minor modifications as described previously.34 All solutions were prewarmed to 37°C before use and were flushed through the portal vein at a flow velocity of 10 mL/minute. In situ perfusion was performed as follows. A first rinsing of the liver was performed for 4 to 5 minutes with buffer A [10 mM 4-(2-hydroxyethyl)-1-piperazine ethanesulfonic acid, 3 mM potassium chloride, 130 mM sodium chloride, 1 mM monosodium phosphate monohydrate, and 10 mM D-glucose], which contained 0.5 mM ethylene glycol tetraacetic acid (Sigma); this was followed by a second rinsing with buffer A only. In the next step, the liver was perfused with collagenase (0.4 mg/mL), which was resuspended in buffer A and 0.25 mM calcium dichloride. The perfusion was stopped when softening of the liver tissue was observed. The collagenase-digested liver was removed from the animal, transferred to a tissue culture plate, and manually disrupted with a scalpel and a cell dispenser under a flow chamber. The cells were resuspended in phosphate-buffered saline and filtered through a mesh with a pore size of 50 μm. The cells were centrifuged at a low spin (50g) for 5 minutes and were resuspended in 0.7 mL of isotonic sodium chloride for transplantation. The cell viability (>85%) was analyzed with trypan blue dye exclusion.

Serum Samples

Blood samples were obtained by retrobulbar puncture under anesthesia with isoflurane and were collected in ethylene diamine tetraacetic acid–coated tubes (Sarstaedt). Analyses of the aspartate aminotransferase (AST) and alanine aminotransferase (ALT) activity and measurements of serum total bilirubin were performed photometrically with a Cobas modular system (Roche Diagnostics, Mannheim, Germany). A modified protocol from Schosinsky et al.35 was used to determine the ceruloplasmin oxidase (CP) activity; it was adapted to a 96-well plate. The absorbance was determined at 540 nm, and international units were calculated after the subtraction of values derived from 15- and 5-minute incubation periods.

Histological Analysis

Formalin-fixed liver tissue was embedded in paraffin and was used for standard hematoxylin-eosin staining. The morphological parameters were recorded after staining and were evaluated by a blinded observer for polyploidy, steatosis, apoptosis, and proliferation. Cell proliferation was assessed with an antibody against the Ki-67 antigen (Vector Laboratories), and the detection of copper in liver samples was performed with 5-(4-dimethylaminobenzylidene)-rhodanine staining according to standard protocols. Approximately 40 to 60 microscopic fields were analyzed per group with standard evaluation protocols.

Copper Determination

Liver samples were taken at the time of sacrifice or from survivors after resection (weeks 14-16) and dehydrated for 72 hours at 70°C; the weights were determined with an ME235S analytical balance (Sartorius, Göttingen, Germany). The samples were dissolved in 300 μL of nitric acid (65%; Merck), and this was followed by overnight incubation at 60°C. Urine was collected for 18 hours and was centrifuged at 20,000g for 5 minutes. The copper concentration was determined with flame atomic absorption spectroscopy (AA6300, Shimadzu, Kyoto, Japan).

Real-Time PCR

Fresh liver samples were taken via resection from survivors or at the time of sacrifice, frozen immediately in liquid nitrogen, and stored at −80°C before the total RNA was extracted (RNeasy kit). Reverse transcription was performed with SuperScript II (Invitrogen). Atp7b primers were designed that were specific for LEA rats and corresponded to a region lacking in LEC rats10: 5′-ACAGAGGTCCAACCGCTCAC-3′ (forward) and 5′-CATTCCCGGTTTCCAATCAG-3′ (reverse). For the detection of the housekeeping gene β-actin, the following primers were used: 5′-ATCGCTGACAGGATGCAGAAG-3′ and 5′-TCAGGAGGAGCAATGATCTTGA-3′. All primers were synthesized by TIB MOLBIOL (Berlin, Germany). A SYBR Green PCR kit (Eurogentec, Seraing, Belgium) was used for quantitative real-time PCR, which was performed with an ABI-Prism 7900HT sequence detector (PE Applied Biosystems). Cycle threshold (Ct) values were first normalized to the expression of the housekeeping gene with the ΔCt method. Each PCR experiment included serial dilutions (0.1%, 1.0%, 10%, and 100%) of reference control complementary DNA (cDNA) obtained from an LEA rat that was spiked with cDNA from an LEC rat. The values were used for the calculation of a standard curve with arbitrary PCR units with respect to the reference control LEA rat. Marker genes specific for liver regeneration were chosen to represent angiogenesis [angiopoietin 1 (ANGPT1), vascular endothelial growth factor A (Vegfa), fms-related tyrosine kinase 1 (FLT1), kinase insert domain receptor (KDR), and TEK tyrosine kinase endothelial (Tek)], antioxidative properties [heme oxygenase (decycling) 1 (HMOX1), nitric oxide synthase 2 inducible (NOS2), and cytochrome P450 2E1 (CYP2E1)], cell adhesion [intercellular adhesion molecule 1 (ICAM1) and vascular cell adhesion molecule 1 (VCAM1)], cell cycling [cyclin D3 (CCND3) and cyclin-dependent kinase inhibitor 1A (CDKN1A)], extra-cellular matrix [actin alpha 2 smooth muscle aorta (ACTA2), collagen type I alpha 1 (COL1A1), and TIMP metallopeptidase inhibitor 1 (TIMP1)], fat metabolism [sterol regulatory element binding transcription factor 1 (SREBF1)], and growth factors [hepatocyte growth factor (HGF), KIT ligand (KITLG), and platelet-derived growth factor receptor B (PDGFR); Table 1]. Three independent experiments were performed for each sample, and the Ct values were normalized to the expression of the housekeeping gene hypoxanthine phosphoribosyltransferase 1 (HPRT1) with the ΔΔCt method.

Table 1. Marker Genes for Liver Disease
Gene SymbolAccession NumberPrimer (Sense/Antisense)
ANGPT1NM053546GCACAAAGGACGCTGATAACG/TAGATTGGAAGGGCCACAGG
COL1A1NM053304CTCTGGAGCTTCTGGTCCTG/GGGAGACCGTTGAGTCCATC
CCND3NM012766CCAGCACTCCCACAGATGTC/CCCTCGGGCTTCAGATATGG
CDKN1A (p21)U24174CTTGCACTCTGGTGTCTCACG/ATCGGCGCTTGGAGTGATAG
CYP2E1NM031543CAGCCATGAGTTTCCAGATCC/CCAACACACACACGCTTTCC
FLT1 (VEGFR1)NM019306CAAAGCTCTGATGACCGAACTC/TCCACGATCACCATCAGAGG
HMOX1 (HO-1)NM012580AGAGGCTAAGACCGCCTTCC/AGGCCTCTGGCGAAGAAAC
HGFNM017017GTCAGCACCATCAAGGCAAG/TGAGTGGGCCACCATAATCTC
HPRT1NM012583AAAGGACCTCTCGAAGTGTTGG/AAGGGCATATCCAACAACAAAC
ICAM1BC081837TTAGCTCCCGTGGGAGTATCAC/CCGCAATGATCAGTACCAACAC
KDR (VEGFR2)NM013062GAGCCCTCATGTCCGAACTC/AGTCTTTCCCAGAGCGGAAG
KITLG (SCF)NM001012459AGGTCCGTTCTGAGCAGGAG/GCCTGTGTCTCGAAGTCTCG
NOS2 (iNOS)NM012611GGCTACCACTTTGAAGAAGCTG/CGTTTCTGGCTCTTGAGCTG
PDGFRB (PDGFR)NM031525CAAGCCTGATGCTGCTGATG/GTCGCAGGAGATGGTGGAAG
ACTA2 (α-SMA)X06801CCGTGACATCAAGGAGAAGC/GCCCATCAGGCAGTTCGTAG
SREBF1 (SREBP1)AF286470AGAAGGCCAGTGGGTACCTG/TGCGGGCCACAAGAAGTAG
TEK (TIE2)XM342863AGGTCGAGTTCGAGGACAGG/CCCTGTCCACGGTCATAGTTAG
TIMP1BC099821CCTGGTTCCCTGGCATAATC/TTGCAAGGGATGGCTGAAC
VEGFA (VEGF)NM031836CAGTTCGAGGAAAGGGAAAGG/CAAATGCTTTCTCCGCTCTG
VCAM1NM012889CCGGCATTTATGTATGTGAAGG/TGACGCTCTTAGATGGGAAGAC

Transplantation

At the age of 4 to 5 weeks, LEC rats (males, 218 ± 37 g; females, 148 ± 23 g) began to receive a high-copper diet consisting of standard rat chow (copper at 13 mg/kg; diet 1324, Altromin, Lage, Germany) and tap water with copper at 20 mg/L [copper(II) chloride; Sigma-Aldrich]. The high-copper diet was maintained throughout the observation period. All operations were performed under inhalatory nitrous oxide/oxygen (2:1)–isoflurane (1.5 vol %) anesthesia (1-chloro-2,2,2,-trifluoroethyl difluoromethyl ether; Abbott). Lateral minilaparotomy was performed. The caudal pole of the spleen was exposed, and a ligature was applied. Viable hepatocytes (1 × 107) were injected (25 gauge) intrasplenically for 15 to 20 seconds within 2 hours after isolation. Afterwards, the ligature was tightened, and the abdomen was closed in 2 layers. Liver samples were taken at the time of sacrifice or after resection (survivors), which was performed from the posterior part of the caudate lobe (approximately 4% of the liver mass).36 LEC rats received hepatocytes or underwent sham transplantation and received isotonic sodium chloride. For a pilot experiment, 1 hepatocyte transplant was performed in 7- to 8-week-old animals. In the other experiments, LEC rats received 3 consecutive hepatocyte transplants at the ages of 7 to 8, 8 to 9, and 9 to 10 weeks. LEA rats that were fed a copper diet (n = 4) or standard chow (n = 4) served as controls so that we could monitor for any effects of the high-copper diet.

Statistical Analysis

SPSS 18.0 was used for the statistical analysis. All values are presented as means and standard errors or standard deviations. Analyses were performed with the Student 2-tailed t test, a 1-way analysis of variance, the Mann-Whitney U test, or the Kruskal-Wallis test. Bonferroni corrections were used. Survival was analyzed with the Kaplan-Meier estimation method. A P value < 0.05 was considered to be statistically significant.

RESULTS

Delayed Onset of Hepatitis After Repeated Hepatocyte Transplantation

As observed by other investigators, a single session of hepatocyte transplantation resulted in poor therapeutic outcomes when no liver preconditioning was used.28 In a pilot experiment, we investigated the survival of LEC rats that were challenged with a high-copper diet for 3 weeks. The animals received a single transplant of 1 × 107 hepatocytes at the age of approximately 7- to 8-weeks. None of the animals (n = 10) receiving a single transplant survived past day 22. All animals developed fulminant hepatitis (data not shown). Next, the impact of 3 consecutive transplants was explored with the same experimental setup (Fig. 1). The time points and the levels of hepatitis-associated serum markers were characterized in LEC rats before and after hepatocyte transplantation (Fig. 2). In comparison with the LEA rat control group, all the LEC rats developed hepatitis and showed significantly increased levels of serum markers at individual time points (0.3 ± 0 mg/dL for bilirubin, 123 ± 6 U/L for AST, and 54 ± 0 U/L for ALT). All sham-operated animals (15/15) and 11 transplanted animals (11/18) showed severe jaundice (bilirubin level > 2.0 mg/dL; Table 2). The onset of hepatitis was delayed in 7 of the 18 transplanted LEC rats; these rats showed no jaundice within approximately 10 days in comparison with the other LEC rats. The period of elevated serum markers was short (mean = 15 ± 3 days) in these animals; thereafter, the levels of all 3 markers normalized between days 30 and 50, and they remained normal to the end of the observation period (Fig. 2C). LEC rats that received a high-copper diet but did not undergo surgery served as controls and showed a course of disease similar to that of sham-operated animals (data not shown).

Figure 1.

Experimental design. LEC rats at the age of 4 to 5 weeks received a high-copper diet, which was started 3 weeks (week −3) before first transplantation (week 0) and was maintained up to the end of the observation period (week 30). The animals received 3 consecutive transplants of hepatocytes, which were obtained from a healthy LEA rat. Control LEC rats underwent sham operations. Blood was routinely taken throughout the observation period. The livers were inspected after sacrifice or resection (weeks 14-16).

Figure 2.

The transplantation of hepatocytes resulted in transient episodes of hepatitis in LEC rats. The levels of hepatitis-associated serum markers are shown for all animals. LEC rats (A) underwent sham transplantation (n = 15), (B) received hepatocytes and had severe jaundice (n = 11), or (C) received hepatocytes and did not have severe jaundice (n = 7). Each line represents the values for 1 animal. Data are depicted with respect to the time of first transplantation (day 0). (A,B) Some animals rapidly developed fulminant hepatitis, and the final values could not be recorded. The values on the x and y axes differ for parts A and B and part C.

Table 2. Onset of Hepatitis in LEC Rats After the Transplantation of Hepatocytes
 JaundiceNo Jaundice
Sham-Transplanted Rats (n = 15)Transplanted Rats (n = 11)Transplanted Rats (n = 7)
  • NOTE: The data are presented as means and standard errors.

  • *

    Significant (P < 0.05) versus other two groups.

Maximum bilirubin value   
 mg/dL25 ± 423 ± 50.8 ± 0.2*
 Days13 ± 217 ± 728 ± 2*
Maximum AST value   
 U/L1287 ± 1671072 ± 198876 ± 79*
 Days13 ± 217 ± 727 ± 2*
Maximum ALT value   
 U/L1203 ± 1391004 ± 1091295 ± 137
 Days10 ± 214 ± 626 ± 2*

Results of Repeated Hepatocyte Transplantation for Long-Term Survival

The delayed onset of hepatitis and the normalization of hepatitis-associated serum markers after repeated hepatocyte transplantation was accompanied by survival for 7 LEC rats (38.9%; Fig. 3). Sham-transplanted LEC rats (15/15) developed fulminant hepatitis and died within 4 weeks (mean 12 ± 6 days) of first transplantation. A moderate improvement in survival was observed in 11 LEC rats that underwent hepatocyte transplantation (mean = 27 ± 16 days). Notably, LEA rats that received a high-copper diet remained healthy, did not develop hepatitis, and survived (data not shown).

Figure 3.

Improved survival of LEC rats after hepatocyte transplantation. The survival of sham-transplanted rats and hepatocyte-transplanted rats (Tx) was determined from the time of first transplantation (day 0) to the end of the observation period (week 30). Each data point indicates 1 animal.

Significantly Reduced Copper Content in the Livers and Urine of Survivors

One hallmark of WD is an elevated liver copper content, which is also observed in LEC rats.8 The copper content in the livers of survivors significantly decreased over time (Fig. 4A). During weeks 14 to 16, the liver copper level was lower in survivors versus sham-transplanted animals. Within a time period of approximately 4 months, the liver copper content decreased further to a value that was, however, significantly higher in comparison with control LEA rats. The levels of liver copper in nonsurvivors and sham-transplanted LEC rats at the time of sacrifice were almost identical. In order to determine whether copper was excreted from survivors via the kidneys, the copper content in urine was determined in LEC rats (Fig. 4B). The urine copper concentration in survivors was almost identical to the values of control LEA rats and was significantly lower than the values of sham-operated LEC rats.

Figure 4.

The transplantation of hepatocytes resulted in a reduction of copper in the livers and urine of survivors. (A) The livers of survivors (Txs) were analyzed after resection (weeks 14-16) and on the day of sacrifice (week 30). The livers of nonsurvivors (Txns) and sham-transplanted LEC rats were taken at the time of sacrifice. LEA rats that received a high-copper diet served as controls. Data for 7 animals in each group were recorded. (B) Urine was collected for 18 hours from survivors, sham-transplanted LEC rats, and LEA rats that received a high-copper diet. Urine was collected during week 30 from survivors and on the day of sacrifice from the sham-operated animals. Three animals were analyzed in each group. Means and standard errors are shown. Asterisks indicate significance (P < 0.05).

Restored CP Levels After Hepatocyte Transplantation

The level of CP is significantly reduced in patients with WD and in LEC rats because of a malfunctioning copper transporter gene (Atp7b).37 Elevated CP activity (>50 U/L) was detected in all survivors within 43 days (median = 29 ± 12 days) of transplantation (Fig. 5A). Before day 98 (median = 49 ± 20 days), all survivors displayed high levels of CP activity (mean > 210 U/L) that were almost identical to the levels of LEA control rats. CP activity was maintained in survivors at high levels to the end of the observation period (Fig. 5B). Notably, most nonsurvivors that received hepatocytes but developed fulminant hepatitis displayed various degrees of low CP activity (median = 41 ± 75 U/L).

Figure 5.

The CP activity was restored in survivors after hepatocyte transplantation. (A) The serum CP activity is depicted for all survivors (n = 7) on an individual basis until the restoration of activity (day 100). Each line represents 1 animal. (B) Means and standard deviations are shown for the CP activity of surviving LEC rats (Txs) on the day of transplantation (day 0) and during weeks 14 and week 30. LEC rats that received hepatocytes but developed fulminant hepatitis (nonsurvivors, Txns), sham-transplanted rats, and LEA rats are shown as controls. Seven animals were analyzed in each group. The asterisk indicates significance (P < 0.05).

Restoration of Liver Injury in Survivors

To monitor the efficacy of repeated hepatocyte transplantation for the copper metabolism of LEC rats,8 we analyzed the liver histology with various stains (Fig. 6A). Hematoxylin-eosin stains of livers obtained from survivors revealed no detrimental alterations of the regular liver architecture in comparison with LEA control rats, although areas of minor inflammation and regeneration were sporadically observed. Livers of sham-operated LEC rats and nonsurvivors developing fulminant hepatitis showed a highly irregular hepatocyte architecture with enlarged nuclei, inflammation, and occasionally necrosis. The livers of survivors contained a few confined areas of high-copper deposition; however, the absolute number of such cell areas was significantly lower in the livers of survivors versus the livers of controls (Fig. 6B). Liver sections obtained from survivors were significantly stained with an antibody directed against the epitope Ki-67, but liver sections from the other 2 groups were not; this suggested the induction of focal regeneration by transplanted hepatocytes in the survivors.

Figure 6.

The liver pathology was ameliorated in the survivors after hepatocyte transplantation. (A) Liver staining with (a-c) hematoxylin-eosin, (d-e) rhodanine, and (g-i) Ki-67 antibodies. The livers were obtained from (a,d,g) an LEC rat that survived after hepatocyte transplantation, (b,e,h) a sham control, and (c,f,i) an LEA rat. One representative stain is shown for each group. The arrows indicate (b) enlarged nuclei within an irregular polyploid parenchyma, (e) focal areas of copper deposition, and (g) singular clusters of proliferative cells. The liver of the survivor 14 weeks after transplantation was normal except for (a) minor hepatocellular aneuploidy, (d) singular areas of copper deposition, and (g) sporadic regenerative foci. The histological findings for the sham control and the LEC rat that developed fulminant hepatitis did not differ (data not shown). The scale bar equals 50 μm. (B) Quantitative analysis of rhodanine and Ki-67 stains. The means and standard errors for the number of positive areas for survivors (Txs), sham-operated rats, and LEA rats are shown (40-60 microscopic fields were used). Asterisks indicate significance (P < 0.05). (C) Expression of marker genes in survivors and LEA rats. Four animals were analyzed in each group. The Ct values were normalized to the housekeeping gene and were compared between the groups with the ΔΔCt method. The means of the fold changes are presented. A value of 1 indicates identical expression in the groups.

We next characterized the expression of marker genes in the livers of LEC rats that survived after repeated hepatocyte transplantation. Altogether, 19 marker genes (Table 1) were analyzed in survivors, and the results were compared to those for LEA rats (Fig. 6C). The values obtained for most marker genes (17/19) were within a factor range of 2; this indicated that the expression levels of the respective genes in the livers of survivors and LEA rats were almost identical. The levels of Tek and Cdkn1a expression leading to angiogenesis and cell cycle arrest (factor = 2.6 ± 1.3 and factor = 3.3 ± 2.7, respectively) were slightly increased in survivors. However, the numbers did not reach significant levels. Notably, the expression of marker genes in LEC rats during fulminant hepatitis significantly differed from the values obtained from survivors (data not shown).

Significantly Increased Level of Liver Repopulation After Repeated Hepatocyte Transplantation

The extent of repopulation by repeated hepatocyte transplantation was assessed in the livers of LEC rats via the determination of Atp7b messenger RNA (mRNA) with real-time PCR (Fig. 7). Atp7b could not be detected in sham-operated LEC rats.10 However, in most nonsurvivors that received LEA hepatocytes but died because of fulminant hepatitis, detectable levels of Atp7b were observed. The level of Atp7b in this group, however, was lower than the level in LEA rats by at least 1 order of magnitude. In contrast, the level of Atp7b in all survivors (7/7) was within the range observed for LEA rats, and this suggested that repopulation by LEA hepatocytes occurred with high efficiency by month 3 after transplantation in the long-term survivors.

Figure 7.

Hepatocyte transplantation resulted in high rates of repopulation in survivors. The relative expression of Atp7b mRNA in the liver was determined. Data were obtained from LEC rats that received hepatocytes but developed fulminant hepatitis (nonsurvivors, Txns), survivors (Txs), sham-operated LEC rats, and LEA rats on a standard diet. Samples were taken on the day of sacrifice (nonsurvivors, sham-operated LEC rats, and LEA rats) or during weeks 14 to 16 (survivors). Atp7b expression was not detectable in sham-operated LEC rats. Gene expression was normalized to β-actin with the ΔCt method. Arbitrary PCR expression units were calculated with a standard curve obtained with the cDNA of an LEA rat. Seven animals were analyzed in each group. Data were obtained from 2 independent experiments and are represented as box plots. n.d., not determined.

Impact of Disease Progression for Repeated Cell-Based Therapy

To explore parameters that may predict the outcomes of therapy, we analyzed LEC rats that underwent repeated hepatocyte transplantation on an individual basis at the time of first transplantation (Table 3). Four of the nonsurvivors (4/11) showed highly elevated levels of hepatitis-associated serum markers at the time of first transplantation (bilirubin ≥ 0.3 mg/dL, AST > 300 U/L, and ALT > 500 U/L). None of the survivors (0/7) showed such markedly elevated levels at this time point, although the ALT values of all animals were significantly higher than the values obtained from LEA rats; this indicated that these animals had ongoing disease activity on the day of first transplantation. None of the values, however, reached significant levels when the 2 groups were compared.

Table 3. Parameters at the Time of Transplantation
AnimalSexDeath (Days)Bilirubin (mg/dL)*AST (U/L)*ALT (U/L)*CP > 20 U/L (Days)Maximum CP (U/L)Maximum CP (Days)Atp7b (U)
  • NOTE: Survivors were recorded up to day 42; by that time, all nonsurvivors and sham-transplanted animals had died.

  • ND, not determined.

  • *

    At the time of first transplantation (day 0).

  • The expression of Atp7b was analyzed in the surviving animals during weeks 14 to 16.

Nonsurvivor 1Female113.811961370Not detectedNDND≤0.1 ± 0
Nonsurvivor 2Female50.5487971Not detected10 ± 205 ± 1
Nonsurvivor 3Female20.9366723Not detected14 ± 10ND
Nonsurvivor 4Female80.3305587Not detected14 ± 505 ± 1
Nonsurvivor 5Male170.0132186Not detected0.1 ± 70ND
Nonsurvivor 6Male270.01491477221 ± 726ND
Nonsurvivor 7Male270.112598772 ± 31728 ± 4
Nonsurvivor 8Female400.1136923521 ± 835Not detected
Nonsurvivor 9Male370.114792Not detected13 ± 337≤0.1 ± 0
Nonsurvivor 10Female330.114483Not detected17 ± 1289 ± 1
Nonsurvivor 11Female420.115978Not detected5 ± 214ND
Survivor 1Female0.12523827117 ± 929112 ± 2
Survivor 2Female0.016416740107 ± 140117 ± 14
Survivor 3Male0.11731012861 ± 33597 ± 2
Survivor 4Male0.0104981579 ± 203479 ± 8
Survivor 5Male0.1123902982 ± 1429126 ± 4
Survivor 6Male0.1130823534 ± 335102 ± 6
Survivor 7Female0.11226822161 ± 2129131 ± 5

DISCUSSION

Hepatocyte transplantation represents a promising alternative therapy for the management of patients with fulminant hepatic failure, end-stage liver disease, or metabolic liver disease, who have previously been shown to have an improved prognosis after orthotopic liver transplantation. This study has demonstrated that repeated transplantation can prevent life-threatening liver disease and suggests that multiple hepatocyte transplants within a short period of time during the onset of fulminant hepatitis may overcome the limitations of a single cell therapy session. In a notable proportion of the LEC rats, long-term survival was observed after repeated hepatocyte transplantation. Survivors displayed normalized levels of hepatitis-associated markers, significantly reduced levels of liver copper, restored CP activity, improved liver histology, and normalized levels of a set of marker genes important in liver disease. Male and female rats benefited almost equally from therapy. The data suggest that repeated hepatocyte transplantation is followed by the cure of severe liver disease even when potentially harmful preconditioning regimens are not used, at least in the experimental setting used here.

The number of engrafted hepatocytes, which was estimated with the CP activity and the PCR analysis of Atp7b, suggested that an almost complete repopulation could be achieved by month 3 in LEC rats. Thereafter, the animals remained healthy, and this suggested that the high level of liver repopulation remained stable at least for the observation period of almost 8 months. As many as 2 × 108 hepatocytes (equivalent to approximately 13% of the total weight of a rat liver) have been transplanted into rats without significant detrimental effects on the liver architecture.38 In all, we used 3 × 107 viable hepatocytes in 3 consecutive transplants; this number was only moderately higher than the number of cells (1 × 107) in other transplantation studies of LEC rats.28-30 However, as shown in the fumarylacetoacetate hydrolase (Fah) mouse model, 100 to 1000 hepatocytes are sufficient for liver repopulation,22 and this suggests that the repeated transplantation of fewer cells might also be efficient for the treatment of LEC rats. No significant difference was observed between male and female rats after transplantation, although female rats had a lower body weight but received the same number of hepatocytes. In contrast, a previous report tentatively indicated differing levels of serum markers and CP activity between the sexes.39 Therefore, the relationship between the number of transplanted cells and the body weight and/or hormone status with respect to the efficacy of cell-based therapy for WD needs further investigation.

In vitro, the selective proliferation of cells expressing functional ATP7B (including hepatoma cell lines) was recently shown in tissue cultures with high concentrations of copper.32 A similar selection of functional hepatocytes may also be present in LEC rats.12, 28 An exceptionally high rate of liver repopulation without any preconditioning has been reported for genetically modified animal models in which transplanted hepatocytes have a marked proliferative advantage (eg, the urokinase-type plasminogen activator–transgenic and Fah-null mouse).21, 22 The selection of Fah-positive hepatocytes seems to be extraordinarily high and can occur in the livers of patients after a spontaneous mutational reversion of the Fah gene followed by clonal cell selection, which gives rise to Fah-positive nodules.40 However, such spontaneous selection does not reverse hereditary tyrosinemia type 1 in most patients because the number of reverted hepatocytes is too low.41 Although a similarly high repopulation efficiency was observed in our rat model of WD (in which spontaneous revertants have not been detected), the mechanisms of repopulation and/or selection are likely different from those of the Fah model, in which selection is due to intrinsic factors and not to dietary copper.42 The individual metabolic consequences of the cellular defect, the particular changes in the liver architecture after injury, and the time constraints of cells for escaping from cell death are specific for individual diseases and represent some of the main driving forces behind the complex selective repopulation process.

As recently shown for LEC rats, radiation, partial hepatectomy, and the addition of bile salts could significantly enhance the efficiency of hepatocyte repopulation, and this suggests the induction of favorable selection conditions by such preconditioning.12, 28-30 Although preconditioning was not used here, an almost complete repopulation was observed in some of the animals. It cannot be ruled out that the high-copper diet used in our study may have contributed to the observed high repopulation efficiency. Our model differs from previous reports because of the stringent onset of fulminant hepatitis (100%); only 40% to 60% of animals have been affected in other studies.12, 43 However, the copper diet that we used was only moderately higher in comparison with other studies.8, 39, 44, 45 Because all the animals were prone to developing fulminant hepatitis, this is the first study in which the full therapeutic impact of hepatocyte transplantation for the amelioration of WD could be analyzed.

Although repeated hepatocyte transplantation has been studied in several animal models, its impact on severe disease has not been predominantly investigated. In comparison with a single intraportal infusion of cells, 6 transplants of normal hepatocytes into analbuminemic Nagase rats significantly improved plasma albumin expression.46 In the dipeptidyl peptidase IV mouse model, liver repopulation could be enhanced to approximately 5% by 3 transplants via the spleen in comparison with a single transplant.38 However, the estimated repopulation efficiencies of these models and other models have been low.47 Most transplanted hepatocytes are subjected to rapid degradation, especially in healthy animals. Liver injury, therefore, seems to be a prerequisite for repopulation by transplanted hepatocytes. On the other hand, advanced liver disease limits the efficiency of cell-based therapy (eg, patients with chronic disease). Interestingly, a minor injury may suffice for efficient repopulation, as recently observed in a transgenic mouse model of alpha-1-antitrypsin.48 Repeated transplants of hepatocytes at different stages of a progressive disease may thus enhance the overall probability of favorable time points for cell engraftment. After engraftment, a strong proliferative stimulus for cell transplants may be present via growth factors released during the course of fulminant hepatitis,49, 50 which can also have different magnitudes at individual time points during the course of ongoing disease.

In a significant proportion of the animals, fulminant hepatitis was prevented by repeated hepatocyte transplantation; however, 11 animals in this experimental group died. Most of these animals underwent transplantation 3 times but showed only modest CP activity and had low levels of Atp7b in the liver. This suggested that low amounts of the transplanted cells were engrafted into the liver or that these cells could not proliferate extensively before the onset of fulminant hepatitis (or both). It is conceivable that the severity of disease at the time of transplantation affects the therapeutic outcome. Our analysis of individual disease parameters (as determined by hepatitis-associated serum markers at the time of transplantation) indicates that hepatocyte transplantation may be less effective at an advanced stage of disease. Transplantation at earlier time points or shortened intervals between transplant procedures might further improve the repopulation efficiency in animals with advanced disease and might thus improve the overall therapy outcome. Despite the high rate of liver repopulation achieved here without preconditioning, other regimens can result in improved therapeutic outcomes after a single transplant, as recently demonstrated in many excellent studies by Gupta et al.28-30 The pretreatment of the liver or the modification of isolated hepatocytes may thus further enhance the proliferation of transplanted cells and improve their attachment to sinusoids.51-53 However, pretreatment regimens must be safe. Notably, the treatment of patients by repeated hepatocyte transplantation has not been followed by detrimental effects.17, 54, 55

In humans, the abrogation of anticopper therapy, which results in a higher copper intake, is also followed by a rapid and strict onset of hepatic failure.56, 57 This resembles the rapid onset of fulminant hepatitis in LEC rats after a high-copper diet. Repeated hepatocyte transplantation might, therefore, be effective in the clinical situation of fulminant hepatitis in patients with WD. However, the models used here and elsewhere12, 28-30 can not fully address the clinical situation of chronic disease observed in patients who mostly develop hepatic failure in a late decade of life after several years of chronic disease. Animals in our study were 7 to 8 weeks old at the time of first transplantation. Also, advanced chronic disease was treated by cell therapy in none of the other LEC studies.12, 28-30 Late developments of WD such as hepatocellular carcinoma, which is frequently observed in LEC rats after 1 year,43 were not investigated. Whether repeated hepatocyte transplantation or other regimens used to enhance the therapeutic efficiency of cell therapy might be effective for patients with advanced chronic liver disease remains to be determined. Taken together, the experiments suggest that repeated cell transplantation can result in a significantly improved therapeutic outcome for the LEC rat model of WD, and it may represent a supplemental therapeutic option for WD and other liver diseases.

Acknowledgements

The authors thank J. Gerss for assisting with the statistical analysis and Y. Avsar and S. Chandhok for critically reading the manuscript.

Ancillary