SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. FATE AND ENGRAFTMENT OF TRANSPLANTED HEPATOCYTES
  4. HEPATOCYTE PREPARATION
  5. IMMORTALIZATION
  6. MODE OF INFUSION
  7. METABOLIC FUNCTION OF TRANSPLANTED CELLS
  8. IMMUNODEFICIENT MOUSE MODELS
  9. IMPROVEMENT OF HEPATOCYTE ENGRAFTMENT
  10. CONCLUSIONS
  11. REFERENCES

More than 30 years after the first hepatocyte transplant to treat the Gunn rat, the animal model for Crigler-Najjar syndrome, there are still a number of impediments to hepatocyte transplantation. Numerous animal models are still used in work aimed at improving hepatocyte engraftment and/or long-term function. Although other cell sources, particularly hepatic and extrahepatic stem cells, are being explored, adult hepatocytes remain the cells of choice for the treatment of liver diseases by cell therapy. In recent years, diverse approaches have been developed in various animal models to enhance hepatocyte transduction and amplification in vitro and cell engraftment and functionality in vivo. They have led to significant progress in hepatocyte transplantation for the treatment of patients with metabolic diseases and for bridging patients with acute injury until their own livers regenerate. This review presents and considers the results of this work with a special emphasis on procedures that might be clinically applicable. Liver Transpl 15:7–14, 2009. © 2008 AASLD.

The liver was among the first organs considered for strategies based on the transplantation of isolated cells. The first method for isolating hepatocytes from the liver involved collagenase perfusion in laboratory animals; it was initiated by Berry and Friend in 19691 and developed by Seglen in 1976.2 The first hepatocyte transplant was performed to treat the Gunn rat, the animal model for Crigler-Najjar syndrome, which is congenitally unable to conjugate bilirubin and consequently exhibits lifelong hyperbilirubinemia. The transplant resulted in a decreased plasma bilirubin concentration.3 Later, isolated hepatocytes were transplanted into rats with liver failure induced by dimethylnitrosamine.4, 5 These experiments demonstrated that hepatocyte transplantation could potentially be used for the treatment of liver failure and innate defects of liver-based metabolism. More than 30 years later, these models are still used in work to improve hepatocyte engraftment and/or function.

FATE AND ENGRAFTMENT OF TRANSPLANTED HEPATOCYTES

  1. Top of page
  2. Abstract
  3. FATE AND ENGRAFTMENT OF TRANSPLANTED HEPATOCYTES
  4. HEPATOCYTE PREPARATION
  5. IMMORTALIZATION
  6. MODE OF INFUSION
  7. METABOLIC FUNCTION OF TRANSPLANTED CELLS
  8. IMMUNODEFICIENT MOUSE MODELS
  9. IMPROVEMENT OF HEPATOCYTE ENGRAFTMENT
  10. CONCLUSIONS
  11. REFERENCES

Many studies have shown that hepatocytes (approximately 20–40 μm in diameter) transplanted into rodents via the spleen or the portal vasculature enter through portal vein branches and are entrapped in proximal hepatic sinusoids, which are 6 to 9 μm in diameter; consequently, the hepatocytes are distributed predominantly in periportal regions of the hepatic lobules. Transplanted hepatocytes cause both portal hypertension and transient ischemia-reperfusion injury. The portal hypertension, in experimental animals at least, usually resolves within 2 to 3 hours with no obvious long-term detrimental effects, and microcirculatory abnormalities disappear within 12 hours.6, 7 Numerous hepatocytes (up to 70% of transplanted cells) remain trapped in the portal spaces, and most of them are destroyed by the phagocytic responses of Kupffer cells, which are activated shortly after deposition of hepatocytes in liver sinusoids.8 The remaining cells translocate from sinusoids into the liver plates through a process involving disruption of the sinusoidal endothelium and release of vascular endothelial growth factor by both host and transplanted cells. Subsequently, translocated cells integrate into the liver parenchyma, where gap junctions and bile canaliculi form between transplanted and host hepatocytes without any significant proliferation in adult animals.9–11 In rodents, hepatic remodeling is complete within 3 to 7 days, and the engrafted cells become histologically indistinguishable from host cells. Transplantation of 2 × 107 hepatocytes in rats has led to the engraftment of about 0.5% of the transplanted cells in the recipient livers.12 Only hepatocytes harboring a selective advantage for survival/proliferation can efficiently repopulate a recipient liver, and as a result, many repopulation strategies have been developed using approaches involving the induction of acute or chronic liver injury.13

In parallel to the development of animal models, the first ex vivo liver-directed gene therapy trial for the treatment of familial hypercholesterolemia (FH) was performed with 5 homozygous patients in 1994.14 However, it has not been repeated since because of the various limitations of this approach, including poor efficacy of hepatocyte transduction, long-term transgene expression, and poor efficacy of cell engraftment. Allogeneic hepatocytes were transplanted into approximately 20 patients with metabolic diseases, resulting in a few cases in some clinical improvement.15 Also, hepatocytes were transplanted into a patient with total argininosuccinate lyase deficiency. The enzyme activity reached 3% of control activity after 8 months, leading to a clinical improvement to a moderate form of the disease.16 Thus, as in animal models, liver cell transplantation can lead to donor cell engraftment in humans with sustained metabolic improvement in rare cases. However, this approach is still far from fully correcting metabolic liver diseases or liver failure.

Despite decades of research, the processes and factors underlying cell engraftment and in situ proliferation are only partially understood, and a good understanding of these mechanisms is essential for the development of new and efficient treatments of human liver diseases. The prevention of early loss of transplanted cells would undoubtedly improve hepatocyte transplantation. First, it has been recently shown that cell-cell interactions between transplanted hepatocytes and hepatic stellate cells modulate hepatocyte engraftment in rat livers. After cell transplantation, soluble signals activating hepatic stellate cells are rapidly induced along with early up-regulated expression of matrix metalloproteinases 2, 3, 9, 13, and 14 and their inhibitors.17 Second, the interaction between integrin receptors and the extracellular matrix plays a role in cell engraftment. The intraportal infusion of a fibronectin-like polymer into recipient rats prior to cell transplantation increases cell engraftment.18 Third, hepatocytes express soluble and membrane-bound forms of tissue factor–dependent activation of coagulation and exert tissue factor–dependent hepatocyte-related procoagulant activity, which can be inhibited by N-acetyl-L-cysteine. This suggests that, as in pancreatic islet transplantation, procoagulant activity might interfere with hepatocyte engraftment.19

HEPATOCYTE PREPARATION

  1. Top of page
  2. Abstract
  3. FATE AND ENGRAFTMENT OF TRANSPLANTED HEPATOCYTES
  4. HEPATOCYTE PREPARATION
  5. IMMORTALIZATION
  6. MODE OF INFUSION
  7. METABOLIC FUNCTION OF TRANSPLANTED CELLS
  8. IMMUNODEFICIENT MOUSE MODELS
  9. IMPROVEMENT OF HEPATOCYTE ENGRAFTMENT
  10. CONCLUSIONS
  11. REFERENCES

Hepatocytes are isolated by collagenase perfusion with modifications of the Seglen technique and variations in the buffer composition.2 However, cell death of freshly isolated hepatocytes occurs, in part, by apoptosis. Apoptosis is initiated when hepatocytes are isolated by the separation of anchorage-dependent hepatocytes from their extracellular matrix components, a process called anoikis or homelessness. This process appears to involve caspase activation following cell-matrix detachment.20–23

Rodent models24 as well as pilot trials using allogeneic hepatocytes to treat a type 1 Crigler-Najjar metabolic disease have revealed the benefits of using freshly isolated and uncultured hepatocytes.25 However, because of the inefficacy of liver cell engraftment, cryopreserved cells are also used in clinical trials using allogeneic hepatocytes, the functionality of which remains lower than that of freshly isolated hepatocytes.26

Hepatocytes are highly susceptible to the freeze-thaw process. There are detrimental effects of cryopreservation on hepatocyte structure and metabolic function, including cell attachment, which is important to the engraftment of transplanted cells in the liver. Various different factors affect the viability and function of thawed cells. They include the preincubation of hepatocytes with cytoprotective compounds to promote recovery from the isolation process prior to cryopreservation, the freezing solution, and the freezing protocol. There are differences in the resistance of hepatocytes to cryopreservation according to the species and the developmental age. For example, preincubation with fructose improves the viability and attachment efficiency of rat hepatocytes and improves the attachment efficiency of human hepatocytes.27 In our hands, simian hepatocytes are easier to cryopreserve than human hepatocytes, and fetal cells are easier to cryopreserve than adult cells (unpublished data, 2008). Most published studies have been performed with human cells. It is nonetheless also important to cryopreserve hepatocytes from animal models so as to make the best use of these models, which are very expensive; this also helps avoid time-consuming hepatocyte isolation. The addition of trehalose or other related oligosaccharides to the cryopreservation medium improves the postthaw cell viability and plating efficiency of both rat and human hepatocytes. Trehalose is a naturally occurring (though not found in the human body) disaccharide containing 2 glucose molecules. This saccharide was tested because lower organisms such as fungi, yeast, bacteria, and insects, which have the ability to survive complete freezing and/or drying, accumulate a large amount of trehalose.28, 29

Pig hepatocytes are considered to be a ready source of metabolic functions for use in bioartificial livers. One study reported that the use of University of Wisconsin solution rather than standard media maintained the functions of cryopreserved porcine hepatocytes not only in vitro but also in vivo; also, intrasplenic transplantation of University of Wisconsin solution–cryopreserved hepatocytes improved the survival of rats treated with D-galactosamine.30

It has been suggested that cryopreservation is a further stimulus for apoptotic death of freshly isolated hepatocytes. The involvement of apoptosis through caspase activation in pig hepatocyte death occurring during cryopreservation was evaluated through testing of the cytoprotective effect of a global caspase inhibitor, benzyloxycarbonyl-valyl-alanyl-DL-aspartyl-fluoromethylketone. In the experiment, 16% of the benzyloxycarbonyl-valyl-alanyl-DL-aspartyl-fluoromethylketone–treated hepatocytes, but 36% of the control cells, were apoptotic 24 hours after thawing.31 Thawed simian hepatocytes cryopreserved in the same medium (in the presence of 50-micromol vitamin E) were functional in vivo after transplantation into newborn mice.32

Hepatocytes have also been encapsulated prior to cryopreservation. Encapsulated cryopreserved rat hepatocytes injected intraperitoneally into a mouse model of fulminant liver failure remained viable up to 2 weeks post-transplantation.33

Various methods of bioencapsulation have been developed to maintain the specific functions and phenotype of the cells. They include supplementation of factors, use of appropriate substrates, and cocultivation of hepatocytes with other types of cells, which can be of liver or nonliver origin. For example, hepatocytes have been encapsulated with bone marrow cells including stem cells.34 Syngeneic bioencapsulated bone marrow cells transdifferentiated into hepatocyte-like cells in a rat model of acute injury.35

Techniques to microencapsulate hepatocytes within synthetic semipermeable membranes have been developed to enable the transplantation of cell cultures without the need for immunosuppression. Hepatocytes can be microencapsulated within a collagen matrix enveloped in an ultrathin sodium alginate copolymer membrane, which allows molecules such as glucose, albumin, and clotting factors to diffuse freely but prevent the microencapsulated cells from getting out. The major field of research in which encapsulated hepatocytes have been applied is that concerning the treatment of liver failure either after intraperitoneal or intrasplenic transplantation into rats.36–38

An alginate matrix (1% concentration) has been used to transplant porcine hepatocytes into a primate spleen; the objective was to enhance hepatocyte engraftment in the spleen by inhibiting immediate translocation of cells into the portal circulation.39

IMMORTALIZATION

  1. Top of page
  2. Abstract
  3. FATE AND ENGRAFTMENT OF TRANSPLANTED HEPATOCYTES
  4. HEPATOCYTE PREPARATION
  5. IMMORTALIZATION
  6. MODE OF INFUSION
  7. METABOLIC FUNCTION OF TRANSPLANTED CELLS
  8. IMMUNODEFICIENT MOUSE MODELS
  9. IMPROVEMENT OF HEPATOCYTE ENGRAFTMENT
  10. CONCLUSIONS
  11. REFERENCES

Transplantation of animal models with immortalized hepatocytes is based on the hypothesis that an ideal alternative to the transplantation of primary human hepatocytes would be the use of a clonal cell line that could be grown in culture and exhibit the characteristics of differentiated nontransformed hepatocytes following transplantation. Human and rat hepatocytes have been transduced with a retroviral vector expressing the immortalizing simian virus 40 large T antigen (SV40Tag) gene.40, 41 This approach allows hepatocyte expansion in vitro, but transplantation of immortalized cells cannot currently be envisaged in clinical settings. A more appropriate approach was first described by Westerman and Leboulch42 and called reversible immortalization. The oncogene, in the retroviral vector, was flanked by lox P sequences, which allowed its excision upon CRE recombinase transduction into the cells. Reversible immortalization was achieved by immortalization of rodent cells with a retroviral vector expressing the SV40Tag gene and a suicide gene—herpes simplex virus thymidine kinase (HSV-tk)—flanked by a pair of loxP recombination targets.43 When transplanted into the spleens of rats with liver failure, the immortalized hepatocyte clones prevented the development of hyperammonemia-induced hepatic encephalopathy. The protection was reversed by treatment with ganciclovir, which kills HSV-tk–expressing cells. The same approach was applied to human hepatocytes.44 One of the resulting immortalized clones, NKNT-3, a reversible simian 40 T-antigen-immortalized human hepatocyte line, expressed the genes of differentiated liver functions and proliferated. After an adenoviral delivery of CRE recombinase and subsequent differential selection, SV40TAg was efficiently eliminated from the NKNT-3 cells by CRE/Lox-mediated site-specific recombination. When transplanted into the spleens of rats under transient immunosuppression, the reverted hepatocytes were able to provide life-saving metabolic support during acute liver failure induced by 90% hepatectomy. A similar approach has been used to rescue mice from lethal doses of acetaminophen.45 In a model of immortalized simian fetal liver progenitor cells, the CRE recombinase was fused to a mutant of the estrogen-receptor ligand binding domains, and its nuclear translocation and excision activity was consequently dependent on the addition of tamoxifen to the transduced cells.46 In this cell model, hepatocytes from which SV40Tag had been excised stopped dividing and became apoptotic and therefore could not be transplanted, in contrast to NKNT-3 cells.47 Moreover, the transgene was not excised in all cells, and SV40Tag induced karyotypic instability.

Recently, a more sophisticated approach was evaluated in a xenotransplantation model. Human hepatocytes were immortalized with a retroviral vector carrying complementary DNAs for the catalytic subunit of human telomerase reverse transcriptase and enhanced green fluorescent protein, flanked by lox P sites. Green fluorescent protein–positive hepatocytes were isolated by flow cytometric cell sorting, and 1 clone was then transfected with a plasmid expressing the CRE recombinase protein fused to 2 mutant estrogen-receptor ligand-binding domains. The reverted hepatocyte population was selected by enhanced green fluorescent protein–negative cell sorting and transplanted into a pig model of acute failure induced by D-galactosamine injection. Recipient pigs received a single intraportal injection of 1 × 109 cells, which was equivalent to 5% of the mass of the whole pig liver. The pigs received a daily intramuscular injection of FK506 for 8 days. When they were sacrificed 3 months post-transplantation, there was no evidence of tumor development in the organs of the pigs.48

MODE OF INFUSION

  1. Top of page
  2. Abstract
  3. FATE AND ENGRAFTMENT OF TRANSPLANTED HEPATOCYTES
  4. HEPATOCYTE PREPARATION
  5. IMMORTALIZATION
  6. MODE OF INFUSION
  7. METABOLIC FUNCTION OF TRANSPLANTED CELLS
  8. IMMUNODEFICIENT MOUSE MODELS
  9. IMPROVEMENT OF HEPATOCYTE ENGRAFTMENT
  10. CONCLUSIONS
  11. REFERENCES

In cases of metabolic liver disease and acute liver failure in which the liver architecture is intact, the presence of a physiological matrix and the availability of a portal blood supply make the liver the optimal site for hepatocyte transplantation. Generally, the intraportal route has been used in animal models of metabolic liver disorders. When the hepatic architecture is modified, as in cirrhosis, transplantation of hepatocytes into the liver may result in prolonged portal hypertension and embolization of the cells to the lungs.49 Thus, ectopic sites for hepatocyte engraftment, particularly the peritoneal cavity and subcutaneous tissues, have been investigated. Unfortunately, these sites do not efficiently support cell viability for long periods because the transplanted cells have no direct and instant access to oxygen and nutrients. However, the survival of hepatocytes transplanted into the peritoneal cavity can be prolonged by encapsulation in alginate or attachment to collagen-coated beads.50

The renal capsular space has also been used, but it can accommodate only a small number of liver cells. Of all nonhepatic organs, the spleen has proven to be the best site for hepatocyte engraftment51; hepatocytes infused into the portal vein or infused into the spleen undergo blood flow–mediated translocation to the hepatic sinusoids. However, a large proportion remains in the spleen, which can serve as a site for long-term survival and function of engrafted hepatocytes.52

Intrasplenic hepatocyte transplantation has been performed in animal models of chronic liver failure; such clinical conditions are likely to require engraftment at this site. In rats, 2 × 107 hepatocytes (representing approximately 3% of the host liver mass) to 7.5 × 107 hepatocytes (approximately 12.5% of the host liver mass) were transplanted into the splenic parenchyma, and only a transient portal hypertension was observed. Radioactive labeling was used to analyze the hepatocyte distribution: approximately 26% of the cells remained in the spleen, 72% were located in the liver, and 2% were in the lungs, where they were entrapped by the small diameter of pulmonary capillaries. This resulted in some cases in pulmonary embolism.53 Almost all hepatocytes deposited on the pulmonary capillaries in rats were cleared within 24 hours.54 Once engrafted, the hepatocytes survived indefinitely.

METABOLIC FUNCTION OF TRANSPLANTED CELLS

  1. Top of page
  2. Abstract
  3. FATE AND ENGRAFTMENT OF TRANSPLANTED HEPATOCYTES
  4. HEPATOCYTE PREPARATION
  5. IMMORTALIZATION
  6. MODE OF INFUSION
  7. METABOLIC FUNCTION OF TRANSPLANTED CELLS
  8. IMMUNODEFICIENT MOUSE MODELS
  9. IMPROVEMENT OF HEPATOCYTE ENGRAFTMENT
  10. CONCLUSIONS
  11. REFERENCES

Hepatocytes are metabolically specialized cells, and liver genes are differentially expressed along a portocentral axis, allowing for metabolic zonation. Mechanisms directing position-specific liver gene regulation are incompletely understood, although the Wnt/beta-catenin pathway and antigen-presenting cells seem to be involved.55 To analyze transplanted cell expression, syngeneic hepatocytes were transplanted into dipeptidyl peptidase IV–deficient F344 rats. Transplanted hepatocytes in perivenous areas exhibited inducible cytochrome P450 activity, which was not expressed by periportal hepatocytes. Moreover, when these cells were transplanted into the livers of suckling rat pups, cytochrome P450 activity was rapidly down-regulated.56 Another recent study confirmed the high degree of plasticity of gene expression in hepatocytes subjected to a change in microenvironment and the predominant role of the liver microenvironment in directing position-specific gene expression.57 Engrafted hepatocytes interact with neighboring hepatocytes, allowing interactions with regulatory signals and persistence of specialized hepatocellular function.

IMMUNODEFICIENT MOUSE MODELS

  1. Top of page
  2. Abstract
  3. FATE AND ENGRAFTMENT OF TRANSPLANTED HEPATOCYTES
  4. HEPATOCYTE PREPARATION
  5. IMMORTALIZATION
  6. MODE OF INFUSION
  7. METABOLIC FUNCTION OF TRANSPLANTED CELLS
  8. IMMUNODEFICIENT MOUSE MODELS
  9. IMPROVEMENT OF HEPATOCYTE ENGRAFTMENT
  10. CONCLUSIONS
  11. REFERENCES

Successful transplantation of human hepatocytes into mice or rats requires the recipient animals to not reject the graft and eventually provide a supportive niche that promotes engraftment of the cells. Small animal models able to harbor functional human liver cell xenografts have been developed through the crossing of various strains of immunodeficient mice with animals with metabolic diseases or severe chronic liver diseases.58 The most powerful model so far is a mouse homozygous for targeted mutations in the interleukin-2-receptor gamma-chain locus; these mutations cause severe impairments in T and B cell development and function and also completely prevent natural killer cell development.59 Urokinase plasminogen activator transgenic or fumarylacetoacetate hydrolase−/− mice, backcrossed with these recombination activation gene 2−/−gamma(c)−/− knockout mice, allow engraftment and repopulation by xenogeneic hepatocytes60–62 (see Table 1).

Table 1. Animal Models
Animal ModelReferencesGene InvolvedCorresponding Human Disease
  1. Abbreviations: DPPIV, dipeptidyl peptidase IV; Fah, fumarylacetoacetate hydrolase; LDL, low-density lipoprotein; Mdr2, multidrug resistance protein 2; Rag2, recombination activation gene 2; uPA, urokinase plasminogen activator.

Fah−/− mouse60Fumarylacetoacetate hydrolaseHereditary tyrosinemia type 1
Mdr2−/− mouse71Multidrug resistance protein 2Progressive familial intrahepatic cholestasis
uPA+/+ mouse13, 61, 62Urokinase-type plasminogen activatorChronic liver injury
Rag2−/−gamma(c)−/− mouse59Interleukin 2 receptor gamma chainX-linked severe combined immunodeficiency
DPPIV rat6–9, 10, 11, 17, 18, 49, 56, 57Dipeptidyl peptidase IV
Gunn rat3Uridine diphosphoglucuronate glucuronosyltransferase-1A1Crigler-Najjar type 1
Long-Evans Cinnamon rat69ATPB7Wilson disease
Watanabe rabbit75LDL receptorFamilial hypercholesterolemia type 1

Syngeneic and immunodeficient models have been very useful for defining the notion of selective advantage of transplanted hepatocytes and evaluating different methods of liver repopulation. However, they do not reflect the situation in humans because nearly all hepatocyte transplants have used (and those in the future will use) allogeneic donors.

Indeed, hepatocyte transplantation is limited by immunosuppression regimens applied to transplanted patients. For example, rapamycin has a deleterious effect on the engraftment and proliferation of engrafted hepatocytes.63 Apoptosis seems to play a role in grafted cell rejection, which seems to be involved in the time-limited clinical improvement of patients transplanted for metabolic diseases.64 Humoral rejection has been described as a major cause of allograft injury, and macrophages have recently been implicated in mediating CD4+ T cell–dependent injury of transplanted hepatocytes.65

Autotransplantation of ex vivo genetically modified hepatocytes is an alternative strategy to hepatocyte transplantation that can now be considered. Hepatocytes can be used fresh rather than cryopreserved, and third-generation lentiviral vectors allow the stable integration of transgenes into the genome of nondividing differentiated cells, including hepatocytes, and may provide durable expression of therapeutic genes.32, 66, 67 The use of autologous cells also overcomes the problems associated with donor scarcity and immunosuppression.

IMPROVEMENT OF HEPATOCYTE ENGRAFTMENT

  1. Top of page
  2. Abstract
  3. FATE AND ENGRAFTMENT OF TRANSPLANTED HEPATOCYTES
  4. HEPATOCYTE PREPARATION
  5. IMMORTALIZATION
  6. MODE OF INFUSION
  7. METABOLIC FUNCTION OF TRANSPLANTED CELLS
  8. IMMUNODEFICIENT MOUSE MODELS
  9. IMPROVEMENT OF HEPATOCYTE ENGRAFTMENT
  10. CONCLUSIONS
  11. REFERENCES

Not only are new rodent models of cell transplantation required, but large animal models (rabbits, pigs, dogs, or nonhuman primates) are needed. Small animal models are needed (1) for complete elucidation of the molecular mechanisms of cell engraftment and evaluation of the effects of the various different factors and (2) for the assessment of the functionality of transplanted cells in the long term. Among rodents, the rat is the best suited for hepatocyte isolation as it is more reproducible than in mice and the surgery is easier.

Research with large animal models is also essential to define procedures that can be applied clinically. Various important issues also need to be addressed in such animal models, including how many cells can be transplanted safely either once or repeatedly and the numbers of cells needed to achieve therapeutic goals. The rabbit is a lagomorph and considered to be a large animal model: rabbits are very sensitive to stress and anesthesia, and they also are prone to developing portal thrombosis after cell transplantation even if given high doses of heparin. Pigs are suitable for the assessment of hepatocyte transplantation and bioartificial livers and have been recently reported to be a reliable model for liver failure.68 Nonhuman primates are the most closely related to humans. The macaca liver, including its vasculature, is more similar than the livers of other nonprimate mammals to the human liver. As in man, the left lobe makes up about one-fifth of the liver and can be easily perfused. The disadvantages, however, include (1) the confinement of the animals and the facilities required (only 1 in France) and (2) the high costs of each experiment, a result of the use of such animals and their long-term housing.

It has been recently shown that liver conditioning by irradiation causes prolonged cell cycle block and promotes preferential proliferation of transplanted hepatocytes.69, 70

However, this procedure and others used in rodents, including hepatectomy and injection of toxins, cannot be used in patients: such liver conditioning carries unacceptable clinical risks. Attempts to increase the proportion of engrafted hepatocytes by stimulating liver regeneration have been limited by the lack of growth advantage for transplanted cells over resident cells. However, there is 1 exception: the mouse model (multidrug resistance protein 2 knockout) of progressive familial intrahepatic cholestasis type 3 due to mutations in the multidrug resistance protein 3 gene encoding the hepatocanalicular phospholipid translocator. In this model, the absence of phospholipids in bile causes chronic bile salt–induced damage to hepatocytes, resulting in partial repopulation of transplanted livers.71

In rats, the occlusion of portal branches of the 2 anterior liver lobes (70% of the total liver mass) results in a regeneration response in the remaining nonoccluded lobes leading to their hypertrophy.72, 73 This procedure, portal branch ligation, also favors efficient retroviral transduction of hepatocytes in vivo.74 A similar approach was performed in Watanabe hyperlipidemic rabbits. Five months after the transplantation of hepatocytes into portal branch ligation–stimulated Watanabe rabbits, the decrease in cholesterol was more pronounced and sustained than that in nonligated animals.75 This experiment demonstrated that this liver regeneration stimulus enhanced the population of transplanted hepatocytes and their functional effect in a large animal model of liver metabolic disease.

Portal Vein Occlusion in Nonhuman Primates

Autologous genetically labeled hepatocytes isolated from a 20% hepatectomy were transplanted into nonhuman primates. The infusion of 400 million β-galactosidase–expressing hepatocytes, equivalent to 4% of the liver mass, resulted in engraftment of less than 2% of the liver mass. As in rodent models, at least 50% of the transplanted hepatocytes were lost during the process.76, 77 Consequently, the 20% hepatectomy was not sufficient to induce liver regeneration and increase the engraftment rate of transplanted cells.

Portal vein occlusion is a widely practiced technique in humans. To explore its effect on hepatocyte engraftment, 50% of the liver portal territories were either ligated or embolized with biological glue prior to cell transplantation in nonhuman primates (Macaca mulatta).78 The left lateral lobe, which accounts for about 20% of the liver mass, was resected for hepatocyte isolation. Isolated hepatocytes were immediately labeled with the Hoechst fluorescent dye and transplanted via the portal vein. Liver regeneration was induced by both procedures, but it was significantly higher after partial portal embolization. Transplanted hepatocytes engrafted more efficiently after embolization than after ligation, and this led to the replacement of 10% of the liver mass.

In humans, portal embolization is performed before major hepatectomy, and the embolized liver is generally removed. For cell transplantation approaches, the embolized liver is not removed, and the long-term effects of permanent portal occlusion with a nonabsorbable material are not known. Therefore, it would be safer to use an absorbable embolizing material prior to hepatocyte transplantation.

We tested the effect of reversible portal vein embolization using an absorbable material (Curaspon powder) rather than gelfoam strips (known to obstruct larger portal branches and therefore probably resulting in early reperfusion of embolized liver). Obstruction of the portal branches with Curaspon powder resulted in hypertrophy of the nonembolized segments and an increase in the nonembolized liver volume of approximately 40%. Full portal recanalization was observed approximately 15 days after embolization and was sufficient to induce significant hepatocyte proliferation, which peaked 3 days after surgery.79

The use of an absorbable material would be safer before hepatocyte transplantation because the embolized liver is not resected. Reversible portal vein embolization induced significant liver regeneration of the nonembolized segments, and this suggests that such an approach may be suitable for cell transplantation. Our preliminary data obtained with nonhuman primates suggest that reversible portal vein embolization could be safely used in combination with hepatocyte transplantation for patients with severe metabolic diseases.

CONCLUSIONS

  1. Top of page
  2. Abstract
  3. FATE AND ENGRAFTMENT OF TRANSPLANTED HEPATOCYTES
  4. HEPATOCYTE PREPARATION
  5. IMMORTALIZATION
  6. MODE OF INFUSION
  7. METABOLIC FUNCTION OF TRANSPLANTED CELLS
  8. IMMUNODEFICIENT MOUSE MODELS
  9. IMPROVEMENT OF HEPATOCYTE ENGRAFTMENT
  10. CONCLUSIONS
  11. REFERENCES

In recent years, the development of different animal models has allowed significant progress in hepatocyte transplantation.

Furthermore, hepatic tissue engineering using primary hepatocytes is an emerging therapeutic approach to liver diseases.80 Two recent studies reported (1) engraftment of functional hepatocytes in a neovascularized subcutaneous cavity in mice81 and (2) a method to manipulate uniform sheets of hepatic tissue allowing the formation, in vivo, of a 3-dimensional miniature liver system that maintained its biological function for several months.82

Numerous approaches to isolating stem cells of hepatic or extrahepatic origin, including embryonic stem cells, are being developed. However, extensive work is still required to assess the number of cells that need to be expanded and differentiated, and the functionality of the different cell types needs to be carefully addressed in animal models. Thus, despite the numerous limitations, adult hepatocytes remain the cells of choice for the treatment of liver diseases by cell therapy.

REFERENCES

  1. Top of page
  2. Abstract
  3. FATE AND ENGRAFTMENT OF TRANSPLANTED HEPATOCYTES
  4. HEPATOCYTE PREPARATION
  5. IMMORTALIZATION
  6. MODE OF INFUSION
  7. METABOLIC FUNCTION OF TRANSPLANTED CELLS
  8. IMMUNODEFICIENT MOUSE MODELS
  9. IMPROVEMENT OF HEPATOCYTE ENGRAFTMENT
  10. CONCLUSIONS
  11. REFERENCES