Abstract
The aim of this study was to establish hepatocyte isolation in pigs, and to evaluate function of isolated hepatocytes after encapsulation, cryopreservation, and transplantation (Tx) in a mouse model of fulminant liver failure (FLF). After isolation, porcine hepatocytes were microencapsulated with alginate-poly-L-Lysine-alginate membranes and cryopreserved. In vitro, albumin production of free and encapsulated hepatocytes were measured by enzyme linked-immunoadsorbent assay. In vivo, encapsulated hepatocytes were transplanted into different groups of mice with FLF and the following experimental groups were performed: group 1, Tx of empty capsules; group 2, Tx of free primary porcine hepatocytes; group 3, Tx of fresh encapsulated porcine hepatocytes; group 4, Tx of cryopreserved encapsulated porcine hepatocytes. In vitro, fresh or cryopreserved encapsulated porcine hepatocytes showed a continuous decreasing metabolic function over 1 week (albumin and urea synthesis, drug catabolism). In vivo, groups 1 and 2 showed similar survival (18% and 25%, respectively, p > 0.05). In groups 3 and 4, Tx of fresh or cryopreserved encapsulated porcine hepatocytes significantly increased survival rate to 75% and 68%, respectively (p < 0.05). Primary porcine hepatocytes maintained metabolic functions after encapsulation and cryopreservation. In mice with FLF, Tx of encapsulated xenogeneic hepatocytes significantly improved survival. These results indicate that porcine hepatocytes can successfully be isolated, encapsulated, stored using cryopreservation, and transplanted into xenogeneic recipients with liver failure and sustain liver metabolic functions.
Keywords
Introduction
Liver transplantation (Tx) is the most effective therapy for acute or chronic hepatic failure, but shortage of organ donors limits its application. Alternative therapies for the treatment of liver failure are under investigation and several approaches are currently tested: (i) extracorporeal hybrid bioartificial liver devices (22,28), (ii) implantable bioartificial liver devices (6,9,10), and (iii) hepatocyte transplantation (1,7,8,16,17,21,36,37,39).
Hepatocyte allotransplantation has been used successfully in rodent (16,21,37) and primates (1,39) as therapy of chronic or acute liver failure. In humans, studies showed that hepatocyte allotransplantation was useful for bridging patients to whole organ Tx, for providing metabolic support during acute liver failure, or for replacing whole organ Tx in metabolic liver diseases (7,8,17,36). However, the shortage of organ donors also limits the availability of human hepatocytes. It is therefore essential to find new sources of tissues for Tx.
Tx of xenogeneic hepatocytes has been considered as unlimited availability of cells to treat liver diseases (4,26). Recently, porcine hepatocytes have been used clinically in bioartificial livers (BAL) (13). As immune responses are still major barriers for successful cell and organ xenotransplantation, semipermeable capsules using polymers have been developed to protect xenogeneic cells (2,11). The small pores of the capsules prevent cells and antibodies to cross, while oxygen, glucose, and nutrients can pass through the membranes and maintain metabolic functions. However, long-term survival of encapsulated cells has been disappointing, preventing its clinical use.
The aim of the present study was to test in vitro and in vivo the function of primary porcine hepatocytes after isolation, microencapsulation, cryopreservation, and xenotransplantation in a mouse model of fulminant liver failure (FLF) established by acetaminophen administration followed by a 30% hepatectomy.
Materials and Methods
Animals
Three-month-old male pigs (n = 5) (Centre Medical Universitaire, Geneva, Switzerland) and weighing 10–15 kg were used as donors of hepatocytes. Eight- to 10-week-old adult male C57BL/6 mice (Charles River Laboratories, France), weighing 25–30 g, were used as recipients. Animals were maintained in conventional housing facilities and experimental protocols were approved by the ethical committee of the Geneva University Medical School and by Geneva veterinary authorities.
Isolation of Porcine Hepatocytes
Under general anesthesia, a midline laparotomy was performed and the liver was exposed. After ligation of blood vessels at the root of the lateral lobe of the liver, the left lateral lobe was perfused with cold Viaspan® (Bristol Meyers Squibb, Baar, Switzerland). The perfused liver lobe was subsequently explanted and put on ice for the following steps of oxygenated perfusion. The oxygenated perfusion was started at a rate of 60 ml/min for 15 min, using a warm (37°C) oxygenated perfusion Ca2+-free buffer [Hepatocyte Liver Perfusion Medium (1×), Invitrogen, Basel, Switzerland]. The liver tissue was then perfused with collagenase solution (37°C) [Liver Digest Medium (1×), Invitrogen] at a rate of 60 ml/min for 10 min. Hepatocytes were released by mincing and shaking the liver in a petri dish containing cold L-15 Medium (Invitrogen) 10% FBS (Invitrogen). The cell suspension was filtered through a sterile 100-μm nylon mesh into a beaker. Porcine hepatocytes were washed in cold Hepatocyte Wash Medium (Invitrogen) and purified by 60% Percoll (Amersham Biosciences AB, Uppsala, Sweden) density gradient separation and resuspended for culture and encapsulation. Finally, viability was determined by trypan blue exclusion test (32,33).
Primary Culture of Porcine Hepatocytes in Monolayers
Cells were cultured on 12-well Primaria plates (Falcon) at 2.5 × 105 cells/well, in Williams E medium (Invitrogen), with 10% FBS (Invitrogen), penicillin (100 U/ml)/streptomycin (100 μg/ml) (Invitrogen), insulin (10–7 M)(Huminsulin, Lilly France S.A.S, Strasbourg, France), dexamethasone (10–6 M) (Sigma-Aldrich GmbH, Basel, Switzerland), EGF (25 ng/ml) (Sigma-Aldrich GmbH). Next day, we switched to serum-free Williams E medium (Invitrogen) for assessment of albumin secretion.
Evaluation of Metabolic Function
Porcine hepatocytes (primary freshly isolated or immortalized, free or encapsulated) were cultured in serum-free Williams Medium E (Invitrogen) and incubated at 37°C for 1 week. Culture medium was collected every day and stored at −20°C for assessment of albumin secretion by enzyme-linked immunosorbent assay (ELISA; Bethyl, Lucerne, Switzerland).
To analyze urea synthesis and drug catabolism, ammonium sulfate (0.56 mM), lidocaine (1 mg/ml), or diazepam (1 μg/ml) was added to the culture medium at 1, 3, and 7 days after seeding 5 × 104 hepatocytes in each well of a 24-well plate. After 6 h, concentrations of each reagent and urea, a metabolite of ammonium catabolism, were measured and compared between the groups. The measurements were performed by the clinical chemistry unit of Geneva University Hospital using Cobas INTEGRA 400 (Roche Diagnostic, Basel, Switzerland) and DxC 800 system (Beckman Coulter Inc., Brea, CA, USA).
Encapsulation of Porcine Hepatocytes
Isolated porcine hepatocytes were centrifuged at 50 × g for 5 min and the pellet was suspended in sterile MOPS washing-buffer solution (10 mM MOPS + 0.85% NaCl, Inotech, Dottikon, Switzerland) and mixed with 1.5% sodium-alginate solution (Inotech). Encapsulation was performed using an encapsulation device (Encapsulator, Inotech), where the porcine hepatocyte-alginate suspension flowed through a 200-μm nozzle and was mixed with a polymerization solution (10 mM MOPS + 100 mM CaCl2, Inotech) to hepatocyte beads. Poly-L-lysine (PLL) hydrobromide solution 0.05% (Sigma-Aldrich GmbH) was added to the hepatocyte beads to form a PLL-alginate membrane. Then, the beads were immersed in a 0.03% alginate solution (Inotech) to form the outer alginate membrane. After washing, the beads were put into a depolymerization solution (50 mM Na3 citrate, 10 mM MOPS + 0.45% NaCl, Inotech) to dissolve the core alginate, transforming beads into alginate-PLL-alginate capsules of a mean diameter of 400 μm (5) (Fig. 1). Capsules were cultured in serum-free Williams Medium E (Invitrogen) or incubated in Hank's solution (University Hospital of Geneva, Geneva, Switzerland) for cryopreservation or Tx.

Cultured encapsulated porcine hepatocytes at day 3. Scale bar: 400 μm.
Cryopreservation of Encapsulated Porcine Hepatocytes
After addition of 10% DMSO (v/v) (Merck, Darmstadt, Germany) and 10% FBS (Invitrogen) to Williams E culture medium, cell suspensions of freshly isolated and encapsulated porcine hepatocytes were transferred into 2-ml cryogenic tubes at 1 × 107 cells/ml and placed in the freezing chamber of a programmable freezing unit (Kryo 10 series III, Planer, UK) for cryopreservation and frozen using a computerized rate-controlled freezing program. Briefly, the sample temperature was reduced at 1°C/min from 4 to 0°C, then reduced at 0.5°C/min from 0 to −11°C, kept at −11°C for 15 min, and then reduced at 3°C/min from −11 to −80°C. Finally, the cryogenic tubes were stored in liquid nitrogen for 10 days. The encapsulated porcine hepatocytes were rapidly thawed by immersing the cryogenic tubes in a 37°C waterbath (<1 min), and the encapsulated hepatocytes were gently washed by sedimentation three times with Hank's solution. One portion of encapsulated hepatocytes were resuspended in supplemented serum-free Williams E culture medium, as specified above, and incubated at 37°C in 5% CO2 for 7 days.
Induction of Fulminant Liver Failure
A model of fulminant liver failure was established in C57/BL6 mice by acetaminophen administration (700 mg/kg, IP) followed 15 h later by a 30% hepatectomy, as previously described (19,27).
Experimental Groups and Transplantation of Encapsulated Hepatocytes
We performed the intraperitoneal Tx by injecting >50 × 106 free or encapsulated porcine hepatocyte suspended in 3 ml Hank's through an 18-gauge needle into the abdomen cavity at the end of 30% hepatectomy.
The following experimental groups were defined: group 1 (n = 11) Tx of empty capsules; group 2 (n = 12) Tx of free primary porcine hepatocytes; group 3 (n = 36) Tx of encapsulated porcine hepatocytes; group 4 (n = 38) Tx of cryopreserved encapsulated porcine hepatocytes.
Survival Follow-up and Histology
In each group, mouse survival was followed up at different time points, two animals were euthanized at regular intervals (i.e., 2 weeks and at 1 month post-Tx), and histopathology of microcapsules and liver tissues were performed.
Statistical Analysis
Statistical analysis was performed using STATISTICA (STATISTICA 5.5 Software for Windows, Statsoft Inc, Tulsa, OK). Survival curves were calculated by Kaplan-Meier method and analyzed with chi-square test. A value of p < 0.05 was considered statistically significant.
Results
Viability
We isolated 8.05 × 106 ± 1.86 × 106 porcine hepatocytes per gram of liver tissue. Cell viability was measured after isolation at 76 ± 8%, after encapsulation at 67 ± 7% and after cryopreservation at 54 ± 5%. Viability after encapsulation and cryopreservation was evaluated on the surface of cell aggregates, but we could not evaluate their centers.
Evaluation of Metabolic Function
Porcine hepatocytes under various conditions (free or encapsulated, fresh or cryopreserved) showed a continuous decrease of albumin and urea synthesis and drug catabolism (lidocaine and diazepam) over the first week of culture (Fig. 2). The values reported in Figure 2A and B represent the appearance of albumin and urea in the culture medium synthesized by porcine hepatocytes. The values reported in Figure 2C–E represent ammonium sulfate, diazepam, and lidocaine clearance of the culture media. Thus, an increase of drug concentration over time represents a decrease of hepatocyte catabolism.

Freshly isolated and cryopreserved encapsulated porcine hepatocytes showed a continuous decrease of metabolic function during the first week of culture, becoming almost undetectable at 7 days. In (A), albumin synthesis concentration was measured on culture supernatants by ELISA and results are given in ng/ml; (B) urea synthesis, (C) ammonium sulfate catabolism, (D) diazepam catabolism, and (E) lidocaine catabolism. The results were given in μmol/L and mmol/L and normalized to total protein. The values reported in (A) and (B) represent the appearance of albumin and urea in the culture medium synthesized by porcine hepatocytes. The values reported in (C–E) represent ammonium sulfate, diazepam, and lidocaine clearance of the culture media. Thus, an increase of drug concentration over time represents a decrease of hepatocyte catabolism. There was no significant difference in albumin and urea synthesis or drug catabolism over time for the three conditions of porcine hepatocytes (free primary hepatocytes, fresh encapsulated, and cryopreserved encapsulated hepatocytes) (p > 0.05).
There was no significant difference in albumin and urea synthesis or drug catabolism over time for the three conditions of porcine hepatocytes (free primary hepatocytes, fresh encapsulated, and cryopreserved encapsulated hepatocytes (Fig. 2A–E).
In Vivo Studies
FLF was achieved by combination of acetaminophen and 30% hepatectomy, resulting in a survival of approximately 20%. In groups 1 and 2, survival remained unmodified (18% and 25%, respectively), compared to untreated mice (p > 0.05). In groups 3 and 4, Tx of freshly isolated or cryopreserved encapsulated porcine hepatocytes significantly increased survival rate to 75% and 68%, respectively (p < 0.01) (Fig. 3).

Kaplan-Meier graph of cumulative survival of animals with fulminant liver failure. In group 1 (n = 11, transplantation of empty capsules) and group 2 (n = 12, transplantation of free primary porcine hepatocytes) survival was 18% and 25%, respectively (p > 0.05). In group 3 (n = 36, transplantation of encapsulated fresh porcine hepatocytes) and group 4 (n = 38, transplantation of encapsulated cryopreserved porcine hepatocytes) survival rate was significantly increased to 75% and 68%, respectively (p < 0.01).
Histopathology
Histopathology of native liver tissue demonstrated massive hemorrhage and extensive hepatocyte necrosis at 15 h after acetaminophen administration (Fig. 4A). In contrast, surviving mice showed complete recovery of native liver tissue 2 weeks post-Tx (Fig. 4B).

(A) Histopathology of native liver tissue demonstrating massive hemorrhage and extensive hepatocyte necrosis at 15 h after acetaminophen administration (hematoxylin-eosin). Scale bar: 20 μm. (B) Surviving mice showed complete recovery of native liver tissue 2 weeks posttransplantation (hematoxylin-eosin). Scale bar: 20 μm.
At 1-month post-Tx, histopathological examination showed fibrosis on the surface of transplanted capsules surrounded by inflammatory cells (Fig. 5).

Histopathological examination of transplanted porcine hepatocyte capsules showed fibrotic formation at 1-month posttransplantation (hematoxylin-eosin). Scale bar: 400 μm.
Discussion
Liver allotransplantation is considered the only effective therapy for severe hepatic disorders, such as fulminant liver failure (FLF), where need for adequate therapy is required urgently and liver or cell Tx must be performed within hours. Therefore, a method of hepatocyte storage must be developed, as preparation of encapsulated primary is time consuming.
Kunieda et al. reported that cryopreservation using UW solution allows storage of rat hepatocytes long term to make best use of available hepatocytes (15). Hengstler et al. showed that cryopreserved human hepatocytes demonstrated a high cell viability and over 60% cytochrome P450 activity when compared to freshly isolated hepatocytes (12). Thus, cryopreservation provides an easy, low-cost method for preserving hepatocytes with maintenance of good viability and could play an important role in cell Tx or therapy for fulminant hepatic failure.
In our study, we tested the in vitro function of porcine hepatocytes after isolation, encapsulation, and cryopreservation. Over 1 week in culture, decreasing albumin and urea synthesis, as well as drug catabolism (diazepam and lidocaine), were observed for porcine hepatocytes, regardless if the cells were freshly isolated, free, encapsulated, or cryopreserved (Fig. 2A–E). These results are in accordance with several studies showing that free primary hepatocytes are difficult to maintain in culture (14,18,25). Interestingly, encapsulated porcine hepatocytes, fresh or cryopreserved, demonstrated a higher albumin secretion at 1 week of culture, compared to free primary hepatocytes (Fig. 2A). This result suggests that capsules represent an appropriate environment for maintenance of biological functions.
In vivo, intraperitoneal Tx of empty capsules (group 1) or free primary porcine hepatocytes (group 2) did not improve survival of mice with FLF, indicating a rapid loss of transplanted cells, possibly due to innate or humoral rejection mechanisms.
In contrast, Tx of encapsulated fresh porcine hepatocytes (group 3) significantly increased survival of mice with FLF from 20% to 75% (p < 0.01), indicating that encapsulation protects hepatocytes from recipient immune responses.
Tx of cryopreserved encapsulated cells (group 4) similarly improved survival from 20% to 68% (p < 0.01), demonstrating that cryopreservation did not significantly alter hepatocyte function (Fig. 3).
Other authors have used Tx of xenogeneic hepatocytes in animal models of acute liver failure. Roger et al. introduced porcine hepatocytes in hollow fibers protected by a semipermeable membrane and transplanted these intraperitoneally into rats with acute liver failure induced by 95% hepatectomy. Survival rate was increased from 27% in controls to 64% in treated animals (29). More recently, Sarkis et al. and Benoist et al. showed that macroencapsulated cryopreserved porcine hepatocytes were as effective as freshly isolated hepatocytes when transplanted in a rat model of acute liver failure induced by a two-stage 95% hepatectomy (3,30,31). Other groups have used porcine hepatocytes to improve liver function and prolong the survival of rats with cirrhosis (24,35). We are not aware of other studies using cryopreservation and microencapsulation of porcine hepatocytes for therapy of fulminant liver failure in a mouse model, which is close to clinical situations.
For our mouse model, we started to administrate an overdose of acetaminophen alone to induce fulminant hepatic failure. We observed a high variability in the mortality of animals with an on–off response. Combination of acetaminophen followed by 30% hepatectomy induced a stable mortality around 80% (19,27). Therefore, this model results in a similar mortality as observed clinically in patients with fulminant hepatic failure. In the literature, few studies have used a model of acute liver failure induced by acetaminophen alone in a transplant setting (38). This model was mainly used to study toxicology and pharmacology of acetaminophen (34). An alternative animal model of acute liver failure is the 90% hepatectomy, but is usually performed in rats (23). One research group has used this 90% hepatectomy model in mice and reported that all animals died within 24 h after hepatectomy (20). In conclusion, our model combining acetaminophen followed by a 30% hepatectomy induced a stable mortality of 80%, which is similar to the clinical situation.
Histopathology of liver tissue showed massive hemorrhage and extensive hepatocyte necrosis at 15 h after acetaminophen administration (Fig. 4A). After 2 weeks, histopathological examination showed that surviving mice had a complete recovery of native liver tissue (Fig. 4B). At 1 month after Tx, histology showed fibrosis formation on the surface of microcapsules that were surrounded by inflammatory cells (Fig. 5). This fibrosis clearly indicates that the capsule material used in these experiments would not be appropriate for treatment of chronic liver diseases. However, the function of encapsulated cells was sufficient to allow regeneration of native liver tissue.
In conclusion, the data presented in this study indicate that xenogeneic porcine hepatocytes can be isolated, encapsulated, and cryopreserved and still maintain biological functions in vitro. When transplanted intraperitoneally, fresh or cryopreserved encapsulated xenogeneic porcine hepatocytes provided life-supporting liver-specific metabolic functions for a short period of time and allowed regeneration of native liver tissue without immunosuppressive therapy. These findings have great potential for clinical use, as encapsulated xenogeneic could be easily stored as described here and injected intraperitoneally in patients with FLF.
Footnotes
Acknowledgments
This work was supported by a Swiss National Research Fund grant (to L.H.B. and P.M.) #3200-067157.01/1. We thank David Matthey-Doret, Corinne Sinigaglia, Nadine Pernin, and Solange Charvier for their technical support.
