|
|
||||||||
Address correspondence to Mr. Tom D. R. Lloyd, Department of Surgery, University Hospitals of Leicester, Glenfield General Hospital,, Groby Road, Leicester LE3 9QP, UK; tel 0116 287 1471; fax 0116 287 9852; e-mail: tdlloyd{at}ntlworld.co.uk.
| Abstract |
|---|
|
|
|---|
(received 1 November 2003; accepted 21 November 2003)
Keywords: Bioartificial liver, hepatocytes, cryopreservation, hepatocyte culture
| Introduction |
|---|
|
|
|---|
An ideal protocol for hepatocyte cryopreservation has been sought for many years without any one protocol gaining universal acceptance [11,12]. Many studies have been performed on various species, with different methods and endpoints, making comparisons difficult [1320]. Despite this uncertainty, it appears that a slow rate of temperature reduction, use of dimethylsulphoxide (DMSO) as the cryoprotective agent, and a rapid rate of thawing are prerequisites to a successful protocol.
As part of our project to develop a porcine bioartificial liver, we assessed the suitability of cryopreserved porcine hepatocytes to provide hepatic support in vitro. A simple, easily applicable protocol was developed to facilitate mass production and enhance consistency. Three facets of the porcine hepatocyte cryopreservation process were selected for this study. The first was to establish the optimal freeze rate protocol, the second was to determine the optimal concentration of porcine hepatocytes in the cryopreservation solution, and the third was to test the influence of a pre-incubation step prior to cryopreservation and subsequent culture.
| Materials and Methods |
|---|
|
|
|---|
Hepatocyte isolation. The whole liver was removed and the inferior vena cava was cannulated with a tube (1 cm diameter) sutured into position. The liver was perfused with pre-cooled (4°C) Soltran kidney perfusion solution (Baxter Healthcare, Ltd., UK), containing 1 µl/ml of gentamicin (Gibco, Scotland). A clip was placed on the portal vein to prevent uncontrolled leakage and to maintain adequate tissue perfusion. Each liver was immediately placed on ice and transported in this solution to the laboratory.
Liver digestion and hepatocyte isolation were performed as described by Koebe et al [21]. Three oxygenated porcine perfusion buffers were sequentially perfused through the liver circuit prior to liver digestion and were then discarded. The stock buffer (5 L) contained NaCl, 154 mM; KCL, 5.6 mM; glucose, 5 mM; NaHCO3, 25 mM; and HEPES, 20 mM. Buffer 1 contained 500 ml of the stock buffer with dexamethasone, 40 mg/L. Buffer 2 contained 500 ml of the stock buffer with EGTA, 1 mM. Buffer 3 was 100 ml of the stock buffer. Liver digestion was achieved by addition of pre-warmed oxygenated porcine digestion medium (Buffer 4, 800 ml of stock buffer containing 0.5 mg of collagenase (Sigma-Aldrich, UK)) and perfused through the liver. This was circulated at a rate of 180 ml/min for 1525 min. When the liver became soft to the touch, it was emulsified manually and the cell suspension was filtered (250 µm, 100 µm, and 75 µm pore meshes), washed, and viable cells were counted using trypan blue exclusion.
Hepatocyte culture. Immediately following isolation, the porcine hepatocytes were cultured on collagen-coated 12-well plates (Becton-Dickinson, France) at a seeding density of 7x105 hepatocytes/well. Porcine hepatocytes were cultured in an adapted tissue medium described by Koebe et al [21], which consisted of Dulbeccos Modified Eagles Medium (DMEM) supplemented with glucose, 4.5 g/L; insulin, 125 mU/ml; hydrocortisone, 60 ng/ml; glucagon, 10 ng/ml; foetal calf serum, 10% v/v; and fungizone (Gibco, Scotland), 1.6 ml/500 ml.
Cryopreservation and thawing. Porcine hepatocytes were suspended in ice-cold DMEM at a concentration of 5x106 hepatocytes/ml (except during the concentration experiment, when other hepatocyte densities were investigated) and 20% (by final volume) foetal calf serum (Sigma) was added. A 10% (by final volume) solution of DMSO (Sigma) was added over 34 min, just prior to pipetting the hepatocytes into 2 ml cryovials (Nalgene). The cryovials were subjected to the appropriate freezing protocols and stored in liquid nitrogen for 710 days before thawing. Thawing was performed rapidly by submerging the vials in a 37°C water bath prior to evaluation.
Determination of optimal freezing rate. Three methods were evaluated. The first was the method described by Diener et al [22], utilising a step-wise program with a computer-controlled Planer Cryo 1016 series machine (Planer, Sunbury on Thames, UK). The second method used the Nalgene propan-2-ol device, which allows a reduction in temperature of approximately 1°C/ min when placed in a 80°C freezer. The device was left in the freezer for 46 hr before the cryovials were removed for storage in a liquid nitrogen container. The third method was to put the cryovials in a container and place the container in a 20°C freezer for 1 hr before transfer to a 80°C freezer for 3 hr. The vials were then stored in liquid nitrogen.
Determination of optimal hepatocyte concentration. Four cell concentrations were evaluated (2.5x106, 5x106, 1x107, and 2x107 cells/ml of solution). The constituents of the cryopreservation solution were used as outlined above. The cryopreservation was performed using the Nalgene propan-2-ol device.
Investigation of pre-incubation step. Following hepatocyte isolation, 1.5x108 viable hepatocytes were suspended in 80 ml of pre-warmed culture medium within two spinner flasks in a 37°C incubator (5% CO2). A further 20 ml of culture media was used to wash the tube to ensure suspension of all cells, resulting in a cell density of 1.5x106 hepatocytes/ml. The spinner flasks (Cellspin, Integra Biosciences) were set to rotate at 30 rpm and left for 1 hr and 16 hr, respectively. After incubation, the cells were transferred to a 150 ml tube and centrifuged (100 x g, 4°C, 5 min) to pellet the cells. Following trypan blue exclusion cell count, 2.5x107 viable hepatocytes were cultured as above. The remaining viable cells were cryopreserved in solution at a concentration of 5.0x106 cells /ml in the Nalgene device.
Of the remaining isolated hepatocytes, 1.0x108 were cryopreserved with the Nalgene device and 3x107 hepatocytes were cultured directly on collagen-coated 12-well plates to serve as a control.
Determination of cell viability and attachment. Total cell number and percentage cell viability were determined using trypan blue exclusion. Cell return refers to the percentage of viable hepatocytes returned following cryopreservation, compared to the number of viable hepatocytes originally frozen. The number of attached hepatocytes was estimated by measuring the protein content of the cells using the Bio-Rad protein assay kit (Bio-Rad, Hemel Hempstead, UK), based on the intensity of the coomasie blue colour formed with protein. The absorbance was measured at 405 nm and compared to a standard curve.
Biochemical parameters. Release of lactate dehydrogenase (LDH) was measured using the Sigma kit. Briefly, lactate dehydrogenase catalyses the reversible conversion of lactic acid to pyruvic acid. Pyruvic acid reacts with 2,4-dinitrophenylhydrazine to form hydrazine, which is measured by spectrophotometry.
Phase II metabolic activity was evaluated by measuring the ability of the hepatocytes to conjugate bilirubin. Bilirubin mixed isomers (Sigma-Aldrich, Gillingham, UK) were added (29 µl of a solution containing 2.9 mg of bilirubin mixed isomers plus 1.25 ml of 50 mM NaOH and 3.75 ml of 100 mM tris-HCl) to hepatocytes in culture and the supernatant was collected after 4 hr. Bilirubin was assayed by a reagent kit (Sigma-Aldrich, Gillingham, UK ), based on reaction of conjugated and unconjugated bilirubin with diazotized sulfanilic acid in the presence of dimethylsulphoxide (DMSO) to generate azobilirubin, which was measured by spectrophotometry.
Cytochrome P450 3A4 activity was measured by the deethylation of lignocaine to produce monoethylglycinexylidide (MEGX), which was isolated by HPLC and monitored at 214 nm. The assay was developed in our department, based on published methods [2325].
Statistics. Statistical analyses were performed by the SPSS computer program, using the ANOVA general linear model. Results were expressed as mean ± SD; p values <0.05 were considered significant.
| Results |
|---|
|
|
|---|
|
|
|
Hepatocyte concentration. When hepatocyte attachment at day 2 was compared, there was little variation among the 4 hepatocyte concentrations. Fresh hepatocytes maintained their attachment up to day 5 of culture. Cryopreserved hepatocytes showed significantly reduced attachment at day 5, compared to fresh hepatocytes (p <0.05), but no significant differences were noted among the cryopreserved hepatocytes prepared by the three methods (Nalgene device vs Planer apparatus vs freezer method).
Effect of pre-incubation. The data suggested that pre-incubation for 1 hr prior to cryopreservation may improve the attachment of cryopreserved hepatocytes in culture, but this finding did not reach statistical significance. As described above, there was a general reduction of hepatocyte attachment at day 2 and day 5 of culture following cryopreservation. Pre-incubation did not have a significant positive impact on this decline.
Biochemical assays
Freeze rate.
Cryopreserved hepatocyte cultures (2 and 5 days) showed higher LDH leakage than cultured fresh hepatocytes, reflecting the expected damage caused by cryopreservation (Table 1
). No significant differences were noted between the mean LDH leakage values following hepatocyte culture at day 2 or day 5.
|
Cytochrome P450 3A4 activity. The mean MEGX value for hepatocytes cultured immediately following isolation was lower than those cryopreserved with the Nalgene device and measured at day 2. CYP3A4 activity increased in cultured fresh hepatocytes from day 2 to 5 (7.5±9.7 µg/ml vs 13.9±11.9 µg/ml) although this difference did not reach significance. Such an increase was not observed in the post-cryopreserved hepatocytes. With the Nalgene device, reduction of hepatocyte 3A4 function, evaluated following cryopreservation, approached significance (12.4±8.9 µg/ml at day 2 vs 5.2±4.7 µg/ml at day 5, p = 0.08). The Planer apparatus also demonstrated a reduction in 3A4 activity, but this was not significant (8.7±4.8 µg/ml at day 2 vs. 4.0±5.6 µg/ ml at day 5, p = 0.17). At day 5 of culture, MEGX production was greatest in the fresh hepatocytes, while the freezer method showed the highest 3A4 activity of the cryopreserved hepatocytes. However, the differences among the freezing methods were not statistically significant.
Hepatocyte concentration.
As shown in Table 2
, on day 2 of culture less LDH leakage was observed for the fresh cells than for cryopreserved hepatocytes. A concentration of 5x106 hepatocytes/ml was associated with the highest LDH leakage. By day 5, LDH leakage from hepatocytes cultured fresh was not greatly changed, but cryopreserved cells showed a marked reduction. Significant reduction in leakage from day 2 to day 5 occurred at the 5x106 hepatocyte concentration (p <0.05). Comparing the other values on day 2 and 5 revealed no significant differences in leakage. At day 2 of culture, LDH leakage for 5x106 hepatocytes/ml vs fresh hepatocytes approached significance (p = 0.10). At day 5 of culture, no significant differences were observed.
|
|
There were no significant diferences among the MEGX results obtained in the pre-incubation experiments, indicating that cryopreserved cells are as good as non-cryopreserved cells and that pre-incubation has a negligible effect on hepatocyte function, based on cytochrome P450 3A4 activity.
| Discussion |
|---|
|
|
|---|
Effect of freeze rate on post-cryopreservation hepatocytes. The freezing rate is considered a major factor that affects post-thaw viability of hepatocytes. Many studies have compared simple rates of freezing (for example placing hepatocytes directly into a freezer [27]) and complex controlled rates (where the temperature is reduced in a controlled fashion using sophisticated freezing apparatuses). The controlled rate methods have been advocated following experiments using rat hepatocytes [28]. The whole freezing process hinges on the avoidance of intracellular ice formation, especially for cells in suspension. Although DMSO provides some protection by allowing more controlled cellular dehydration, the actual rate of freezing is also important. If the rate is too fast, water cannot exit and the formation of intracellular ice damages the cells. If the rate is too slow, hepatocytes are exposed to excessive cellular dehydration and mechanical effects of external ice are evident [29,30].
By controlling and altering the freeze process, it is hoped to optimize the time and temperature at critical stages of this process. Some processes involve a shock-cooling phase [31]. Here the temperature is suddenly reduced to compensate for the latent heat of fusion, which increases the temperature as ice forms and may be harmful to cells. By quickly reducing the temperature, the extra energy is absorbed and smoother cooling progression occurs. The ideal freezing protocol is far from clear and discrepancies exist, which may reflect differences in hepatocyte preparation and experimental variations [32,33]. Guillouzo et al [12] reviewed the situation in 1999 and concluded that there was no convincing benefit from complex controlled cooling, compared to simply placing the hepatocytes first in a 20°C freezer and then in a 80°C freezer.
In experiments using porcine and rat hepatocytes, various step-wise cooling programs have been advocated [1,2,7,22]. In this study, we found that a complex computer-controlled cooling system (Planer apparatus) was not significantly superior to other techniques. The simple freezer method, in agreement with Guillouzo et al [12], proved as effective. Alexandre et al [13] found with human hepatocytes that the Nalgene device was as effective as computer-controlled rate reduction. Although we did not find any method to be superior in terms of biochemical parameters, the Nalgene device produced the highest return of viable hepatocytes and was therefore chosen for future experiments.
Deterioration of hepatocyte attachment was consistently seen as the post-cryopreservation hepatocyte cultures progressed from day 2 to 5, in comparison to fresh cultures. This appears to be an inevitable consequence of cryopreservation, although the hepatocytes that remain show good cellular repair, with comparable LDH leakage and functional ability [1].
Effect of hepatocyte concentration on post-cryopreserved hepatocytes. Storage at high cell density would save space, which is an important factor for large-scale tissue banking. Published studies have investigated variations in cell densities during cryopreservation, but few studies have established an optimum cell density. Madan et al [34] noted that in published studies the concentration of stored hepatocytes has ranged from 106107 cells/ml of solution. De Loecker et al [35] found with rat hepatocytes that as the density of cells decreased, cell viability as measured by trypan blue exclusion increased. As cell density increased, the attachment measured by radioactive uptake diminished, following 1 hr of culture. In canine hepatocytes, cell viability measured by trypan blue exclusion was better at concentrations <107cells/ml than at higher density [22,36]. These findings support the hypothesis that, although there are many reasons for cellular injury during freezing, membrane to membrane contact between hepatocytes leads to more serious damage; hence lower cell densities should yield improved viability [35].
The hepatocyte concentrations evaluated in this investigation had no significant influence on the measured parameters. The 5x106 hepatocyte/ml concentration appeared to return more hepatocytes and to sustain bilirubin conjugation better than the others concentrations tested. The least dense and the most dense concentrations showed the widest variations. One problem was cell clumping following isolation, which may have impaired the accuracy of trypan blue exclusion and associated haemocytometer cell counts.
Effect of pre-incubation on hepatocytes post-cryopreservation. Culturing hepatocytes in suspension following isolation, but prior to cryopreservation, is thought to allow the cells time to recover from the isolation process before the rigors of the freezing process. Darr and Hubel [37] cultured porcine hepatocytes in Williams E solution (with additives) in a spinner flask for 4 to 48 hr at 90 rpm following isolation, before the cells were seeded for culture. On fresh culture, the albumin secretion increased after a long pre-incubation period (0.25 µg/ml/hr at time 0, increasing to 1.4 µg/ml/hr at 48 hr). Further experiments examined pre-incubation and subsequent cryopreservation and thawing. Pre-incubation up to 24 hr conveyed significant improvement of albumin production upon thawing. Unlike fresh cells, preincubation for more than 24 hr showed marked reduction in albumin secretion. They concluded that aggregates of cells form during the culture in suspension, which do not survive the freezing process as efficiently and thus do not last as long in culture [37].
Koebe et al [38,39] performed experiments on cryopreservation of hepatocytes in monolayer culture, using rat and porcine cells. Prior to cryopreservation, they cultured the hepatocytes on a monolayer in DMEM with additives. Culturing for 3 days before cryopreservation gave the best results. Moreover, culturing for 7 days prior to freezing in a sandwich configuration conferred survival benefits after thawing [40].
The effect of pre-incubation on porcine hepatocytes was disappointing in our studies. Pre-incubation for 16 hr did increase the percentage of viable hepatocytes after cryopreservation, but the process itself reduced the viable cell population by 80%. Possible explanations include too high a spin rate and hence detrimental shear forces. In respect to functional ability, pre-incubated hepatocytes did not perform any better than fresh cells before or after cryopreservation. Further investigations on this facet of cryopreservation are underway in our laboratory.
| Conclusions |
|---|
|
|
|---|
| Acknowledgements |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |