- Email: [email protected]
Cryopreservation of isolated rat hepatocytes: A critical evaluation of freezing and thawing conditions
CRYOBIOLOGY
25, 323-330 (1988)
Cryopreservation
of Isolated Rat Hepatocytes: A Critical Evaluation of Freezing and Thawing Conditions
CHRISTOPHE CHESNti AND ANDRIS: GUILLOUZO INSERM U 49, Unite! de Recherches Hkpatologiques, H6pital de Pontchaillou, Rennes, France Various parameters, including the nature and proportion of the constituents of the cryoprotective medium, the cooling rate, and the composition of the thawing medium, were evaluated for the cryopreservation of adult rat hepatocytes. The highest percentage of cells able to survive in culture was obtained by freezing in Ll5 medium containing 16% dimethyl sulfoxide, at a rate of 3”C/min, and by adding 0.8 M glucose to the thawing medium. More than 50% of hepatocytes capable of attachment just after cell isolation kept this property after freezing and survived in primary culture. Dead cells could be eliminated before seeding by centrifugation on a Percoll layer. In culture, frozen cells exhibited a morphology similar to that of unfrozen cells and after 24 hr their protein secretion rate was reduced by only about 40%. 8 1988 Academic PWSS, IX.
The hepatocyte is a valuable experimental model since it performs numerous functions, including synthesis and secretion of plasma proteins, production and storage of energy involving carbohydrate, lipid and amino acid metabolism, and drug metabolism. Marked species differences may exist in various functions between animal species and man. In addition, individual variations may be frequent, particularly in man. Indeed, human hepatocytes exhibit a genetic polymorphism with respect to their drug metabolic capacity. Hepatocytes can easily be isolated in high yields by collagenase perfusion of the liver. Four to six hundred million cells can be produced from a single rat liver and several billion hepatocytes can be produced from a large human liver sample (8). Such numbers of cells, particularly from human origin, are too large to be used just after their isolation. Moreover, the availability of livers from humans and many animal species is limited and cannot be routinely performed. Therefore, it is desirable to store cells in excess. The only conceivable method to achieve long-term storage of paReceived October 19, 1987; accepted February 29, 1988.
renchymal cells appears to be cryopreservation. A successful1 freezing protocol must yield stocks of cells that can be restored to full activity at any time point by thawing. Several freezing protocols have been described for adult hepatocytes. Cryopreservation of both animal and human liver cells has been reported, but in all cases, cell viability and biochemical functions were impaired (4, 5, 13, 15, 20). Freeze-induced damage appeared particularly pronounced when attempts to culture frozen hepatocytes were made (10, 16, 17). This led us to reconsider various aspects of the optimal conditions for hepatocyte cryopreservation using cell attachment to a support as a criterion for the evaluation of cell viability after freezing. MATERIAL
Isolation
AND METHODS
of Hepatocytes
The two-step collagenase perfusion method was used as previously described (7). The livers (180-200 g male Sprague Dawley rats) were first washed with a Ca*+-free Hepes (N-2-hydroxyethylpiperazine-N’2-ethane sulfonic acid)
323 OOll-2240/88 $3.00 Copyright All rights
Q 1988 by Academic Press, Inc. of reproduction in any form reserved.
324
CHESN6
AND
GUILLOUZO
buffer, and then dissociated with the same buffer containing 0.025% collagenase (Boehringer, Mannheim, FRG) and 5 mM CaCl,. After three washings, the cells were counted with a hemacytometer.
tectant, respectively. bation at room centrifugation (50g x purified over Percoll
Deep Freezing of Hepatocytes
Viable hepatocytes in 2-15 ml of medium were separated from dead cells by centrifugation (50g x 5 min) over a 15-ml cushion of a 36% Percoll solution in phosphatebuffered saline (PBS), d = 1.06 (18). Pelleted cells were 90-95% viable and the cells over the meniscus of Percoll were 90-95% dead, as estimated by the trypan blue exclusion test.
Hepatocytes were suspended (4 to 5 x lo6 cells/ml) in Leibovitz L15 medium containing 20% fetal calf serum (FCS), and then an equal volume of L15 medium containing 20% FCS and the cryoprotective agent was slowly introduced (5 min). The cell suspension was distributed in 1.6-ml freezing vials (Nunc, Roskilde, Denmark). After a 30-min incubation at room temperature, the vials were transferred to a cell freezer (Minicool LC40, Air Liquide, Sassenage, France) programmed at 3”C/min until liquid-solid phase transition and at l”C/min thereafter or placed in a - 70°C freezer. In this freezer, different cooling rates were obtained by using several holders: polystyrene packs with holes of the same diameter as the vials (2 or 3”C/min), hemolysis tube holders (6”C/min), and aluminium holders prerefrigerated at - 70°C with holes filled with ethanol (30”C/min) or not (12”C/min). Once the temperature of - 100°C was reached in the cell freezer or after 1 hr in the -70°C freezer, the vials were plunged in liquid nitrogen. Cells were preserved at - 196°C from some hours to several weeks. The slopes of cooling inside the vials were calculated from the curves registered between - 10 and -50°C with the thermic probe of the cell freezer in both the cell freezer and the - 70°C freezer. Thawing of Cells
The vials were quickly placed in a waterbath at 37°C and then the cell suspension from each vial was pipetted slowly and transferred in 15 ml of L15 medium. In all experiments, molarity of glucose was kept proportional to the concentration of cryoprotective agent: 0.5 and 0.8 M glucose corresponded to 10 and 16% (v/v) of cryopro-
Purification
After a lo-min incutemperature and 1 min), the cells were or seeded directly.
of Viable Cells
Culture of the Cells
Hepatocytes were seeded at a density of 7 x IO5cells per petri dish (35 mm in diameter) in 2 ml of the culture medium supplemented with 10% FCS and incubated at 37°C in a humidified atmosphere composed of 95% air-5% CO,. The medium was a mixture of 75% minimal essential medium and 25% 199 medium, buffered with bicarbonate (0.22% w/v), and supplemented with bovine serum albumin (0. I%), bovine insulin (10 t&ml), and antibiotics (penicillin 10 IU/ml, streptomycin 50 pg/ml). Some dishes were coated with rat tail collagen films (14). To obtain these films, collagen was spread on the bottom of the dishes, dried overnight at room temperature, and rinsed with PBS before seeding. Measurement
of Cell Viability
Cell viability was examined both before seeding and after attachment to a support. Viability of freshly isolated or thawed cells was calculated after a 5-min incubation in a 0.05% trypan blue solution in PBS. Viability of attached cells was measured 16-20 hr after seeding. Cell monolayers were rinsed twice with PBS and sonicated (10 set, 10 W, with a Ultrasonic sonicator, New York) in 2 ml of PBS. Intracellular lactate dehydrogenase (LDH) activity was determined using a LDH kit (Roche) on an automatic
CRYOPRESERVATION
325
OF RAT HEPATOCYTES RESULTS
analyzer (Bio-Cobas, Roche, Switzerland). The amount of DNA was measured fluorometrically (2) on the same apparatus. All experiments were performed at least twice. Each value of cell attachment was determined from three or four dishes. The values were statistically treated by the one-way analysis of variance and the Newman Keul’s test.
In preliminary experiments (Fig. l), dimethyl sulfoxide (Me,SO), glycerol, and 1,Zpropanediol were tested as cryoprotective agents at a wide range of concentrations and at a low cooling rate (3”Umin); Me,SO was found to be more efficient than glycerol and was selected for further studies. 1,2-Propanediol, though better than glycerol, was markedly less effective than Determination of Protein Synthesis Before and after freezing, each hepato- Me,SO. The appropriate proportion of Me,SO cyte suspension was seeded at different was then determined. The curve of cell vidensities in 24-well dishes. After 24 hr of ability versus the percentage of Me,SO was culture, cell monolayers were washed with rather sharp (Fig. 2). When taking into acPBS and incubated at 37°C in 0.5 ml of culcount both efficiency and toxicity, a conture medium containing 10 &i/ml of centration of Me,SO as high as 16% was [3H]leucine (sp act 0.7 Ci/mmol, Amerfound to give the best results. This optisham, France). After 90 min, media were mum did not change with varying cooling collected and the proteins were precipitated rates (Fig. 2). The optimal incubation time with trichloroacetic acid. The precipitates with Me,SO was around 30 min (Fig. 3). were washed, dissolved in formic acid, and The influence of temperature during incucounted. Cell monolayers were sonicated (5 bation was assayed at 4, 22, and 37°C (Fig. set, 10 W) in 0.5 ml PBS and proteins were measured using the Bradford’s method (1). Each determination was performed in triplicate .
= -2
g 4002
600.
= bte,SO g
400.
200.
A
200
6: i
I / 10
I
I I
$ Propanedid
.I
iGlycerol
14
16
Cryoprotective agent (‘I01 FIG. 1. Effect of the nature of the cryoprotective agents (CA) tested. Hepatocytes were incubated for 20 min with each CA in the culture medium containing 20% FCS and then cooled at 3Wmin. Intracellular LDH was measured in thawed cells after 20 hr of culture. The values are means -t SE (n = 3). **P < 0.01 vs 14% glycerol and 16% 1,2-propanediol. Me,SO was much more efficient than the two other agents and was selected at its optimum concentration, i.e., 16%, in the regime for freezing; under these conditions, the viability after freezing was 30%. As in the following figures, this value was estimated by measurement of intracellular LDH in both seeded and attached cells.
Cryoprotective
agent(%)
FIG. 2. InlIuence of the concentration of Me,SO in the cryoprotective medium at several cooling rates. Hepatocytes were suspended in Ll5 medium containing different proportions of Me,SO and 20% FCS were frozen in a - 70°C freezer at 3”C/min (in a polystyrene pack), 6Wmin (in a hemolysis tube holder), and 30Wmin (in a prerefrigerated aluminium holder with holes containing ethanol). Intracellular LDH was measured in thawed cells after 20 hr of culture. The values are means -C SE (n = 3). **P < 0.01 vs 7, 10, 12, 15, and 20% Me,SO. The optimum concentration of Me,SO was independent on the cooling rate. A 16% concentration was chosen at a cooling rate of 3”C/min; under these conditions the viability after freezing was 33%.
326
CHESNfi
AND
3). There was no difference of attachment when the cell suspensions were incubated at 4 and 22°C but a 30-min incubation at 37°C caused a 50% decrease of viability. The influence of the composition of the aqueous part of the cryoprotective medium was also investigated. Three buffers, namely, Hepes, PBS, and Euro-Collins used for kidney preservation (3), the culture medium (75% minimal essential medium, 25% 199 medium), and L15 medium were tested (Fig. 4). L15 medium gave the best results and was retained. The influence of FCS concentration in the cryoprotective medium was also estimated (Fig. 5). Better results were observed when the cell suspension was diluted in pure FCS but since the difference with lower FCS concentrations was only slight, further experiments were carried out in the presence of 20% FCS. Figure 6 shows that slow cooling rates were better tolerated than faster rates. However, they could lead to sedimentation of the cells during the liquid phase of cooling, since hepatocytes are dense cells and sediment quickly. This phenomenon was found to be toxic and decreased the viability by three- to fivefold. It could be partially
GUILLOUZO
1600
1400
1200
1000 7
5 ;
600..
: a
600-
400. 200-
/IS’ s/25
115
FIG. 4. Effect of the nature of the cryoprotective medium. Hepatocytes were suspended in several media: Hepes buffer used for cell isolation, PBS buffer, L15 medium, culture medium (MEM-199), and EuroCollins buffer. All these media contained 16% Me,SO but no FCS. The incubation lasted 30 min at 22°C except for the Collins buffer which was maintained at 4°C. Cells were frozen at 3Wmin. Intracellular LDH was measured in thawed cells after 20 hr of culture. The values are means f SE (n = 4). **P < 0.01 vs all other media. Ll5 medium gave the best results and was chosen in the regime for freezing; under these conditions, the viability after freezing was 3%.
circumvented by the presence of FCS (because of its viscosity) or quick freezing. However, we preferred to prevent cell sedimentation by stirring the vials during the liquid phase of the cooling procedure. The influence of the cooling rate was not as critical as that of the proportion of cryoprotecL 0 20 40 so tive agent. This led us to select the classical Incubation time (min) cooling equipment using a - 70°C freezer, a FIG. 3. Influence of the time and temperature of simple method well adapted for routine incubation of the cells in the cryoprotective medium work. The cooling rates obtained in a before freezing. Hepatocytes were incubated with 16% Me,SO in Ll5 medium containing 20% FCS during 3 - 70°C freezer were linear until - 50°C and to 60 min and were frozen at 3Wmin. Intracellular the liquid-solid phase transition was only LDH was measured in thawed cells after 20 hr of cul- moderately prolonged compared to that obture. The values are means f SE (n = 3). **P < 0.01 vs 3 min at 22°C and 35 min at 37°C. A 30-min incu- tained with a programmed cell freezer. As expected from the values of cooling bation at room temperature was selected; under these conditions, the viability after freezing was 33%. rates, hepatocyte injury was reduced by
CRYOPRESERVATION
OF RAT
327
HEPATOCYTES
zoo. 6 0
10
20
40
II *’
I a4
Fetal calf serum( %) FIG. 5. Effect of the proportion of FCS in the cryoprotective medium. Hepatocytes were suspended in Ll5 medium containing 16% Me,SO and different proportions of FCS and were frozen at 3Wmin. Intracellular LDH was measured in thawed cells after 20 hr of culture. The values are means f SE (n = 3). **P < 0.01 vs 0%. No marked difference was observed with FCS concentrations equal or superior to 5%. A 20% concentration was chosen in the regime for freezing; under these conditions, the viability after freezing was 36%.
rapid thawing (in a 37°C water bath) compared to a slower thawing (in ambient air at 22°C): the viability was about 15-20% higher. Addition of glucose to the thawing medium limited the osmotic shock caused by the outflow of cryoprotective agent from the cells. The most effective concentration of glucose was obtained between 0.6 and 1 M when a 16% Me,SO concentration was used during the freezing step (Fig. 7). When the optimized freezing and thawing conditions were employed, the average viability values measured by the trypan blue exclusion test were about 65% for frozen cells versus 83% for nonfrozen cells (Table 1). This test of viability is useful for shortterm experiments on hepatocyte suspensions. However, the values of recovery were overestimated since a fraction of dead cells were eliminated during washing. In addition, the trypan blue test does not give any indication on the capability of thawed cells to be maintained in culture. To evaluate this property, cell attachment was assayed 16 to 20 hr after hepatocyte seeding. This time was selected because frozen cells attached and spread more slowly than non-
10
30
” Cooling
60 rate Vqmin)
FIG. 6. Effect of the cooling rate. Hepatocytes were suspended in Ll5 medium containing 16% Me,SO and 20% FCS and were frozen in a -70°C freezer in a polystyrene pack with a cover (TUrnin) or without (3”C/min), on a hemolysis tube holder (6YYmin) or in a prerefrigerated (at -70°C) aluminium holder with holes containing ethanol (30Wmin) or not (lTC/min), or were plunged directly in liquid nitrogen (6OWmin approximately). After 1 hr in the - 70°C freezer, vials were transferred in liquid nitrogen. Intracellular LDH was measured in thawed cells after 20 hr of culture. The values are means + SE (n = 3). **P < 0.01 vs 6, 12, 30, and 60Wmin. A 3Wmin cooling rate was found to be the most efficient and was selected in the regime for freezing; under these conditions, the viability after freezing was 34%.
frozen cells. The percentage of attached cells was determined by measuring intracellular LDH and DNA content. Both values were directly compared with those calculated from the freshly isolated cell suspension. Lower values were obtained by LDH measurement: they were 62 and 33% for nonfrozen and frozen cells, respectively, while they reached 81 and 47% when DNA content was used as the endpoint for cell attachment evaluation (Table 1). In order to improve cell attachment, a collagen-coated substrate was also tested. A 15% increase in cell attachment was observed when collagen-coated plastic was used as a support (Table 1). To limit attachment of nonviable cells and to obtain confluent cell monolayers from frozen cell suspension, viable cells were separated from nonviable ones over a Percoll layer. Cell monolayers obtained from frozen cells purified over a Percoll layer looked similar to those obtained from nonfrozen cells (Fig. 8).
328
CHESNl?
I
0
a2
0.6
1 Glucose
AND
1.4 (M)
FIG. 7. Effect of the glucose in the thawing medium. Hepatocytes were frozen at 3”Umin in L15 medium containing 20% FCS and 16% Me,SO. The cells were then thawed in L15 medium containing glucose at different concentrations. This step of thawing lasted 10 mitt before centrifugation. Intracellular LDH was measured in thawed cells after 20 hr of culture. The values are means f SE (n = 3). **P < 0.01 vs 0 M glucose. Glucose had a significant effect at 0.8 A4 and was used at this concentration in the regime for freezing; under these conditions, the viability after freezing was 28%.
After 24 hr of culture, protein secretion rate was measured by incorporation of labeled leucine. Whatever the cell density, this cellular activity still represented 5060% in frozen cell cultures compared to the value calculated in nonfrozen cultured hepatocytes (Fig. 9). DISCUSSION
Several studies on various cell types have shown that the composition of the freezing medium and the cooling rate are fundamental parameters in the preservation of cell viability after freezing (19). The results reported here clearly indicate that these parameters are also critical for hepatocyte freezing. As already shown (16) Me,SO was found to be more efficient than glycerol. The best results were obtained with a concentration of 16% which is higher than that used by others (5, 12, 13, 16). However, it may be mentioned that Novicki et al. (16) had noticed that Me,SO gave identical results at 10 and 20%. As observed with pancreatic islets (1 l), a 30-min incubation with this cryoprotective agent
GUILLOUZO
before freezing at 4 or 22°C did not alter cell viability, while a similar incubation at 37°C caused a significant decrease in cell attachment after the freezing-thawing steps. Most authors have used a cooling rate of 1 or 2”C/min. Thus, Le Cam et al. (13) found that cooling rates below 7YYmin were more efficient than faster rates. To our knowledge, only Gomez et al. (6) have recommended a higher cooling rate (39YYmin) obtained by plunging the vials in liquid nitrogen. By contrast, Fuller et al. (4) had found this protocol particularly inefficient. We show that slow cooling rates (around 3”C/min) were better tolerated provided that cell sedimentation was prevented by stirring. The values of cell recovery after freezing greatly varied according to the end point used and the time of measurement. When determinations were made after cell attachment, it may be assumed that only functional cells were counted. Lower values were obtained by intracellular LDH measurement than by DNA determination. This could be explained at least partly by plasma membrane alterations occurring during freezing and thawing processes and resulting in the loss of intracellular LDH. Values obtained with both parameters were higher than most of those previously reported by others (16, 17): the highest values for cell attachment after freezing estimated by various end points did not exceed 20-25% of the total cell population. In this study, depending on the parameter used, 3347% of the initial cell population were still capable of attachment just after freezing. Only Jackson et al. (10) found a similar percentage of cell attachment after freezing: however, this is somewhat surprising since it is only slightly lower than the value determined by the trypan blue exclusion test (42%). After freezing, protein synthesis was decreased probably because of the occurrence of some alterations during freezing
CRYOPRESERVATION
OF RAT
329
HEPATOCYTES
TABLE 1 Viability of Frozen and Nonfrozen Hepatocytes
Cell viability (%) Criterion
Before freezing
After freezing
Cell recovery (%)
Trypan blue test Attachment On plastic LDH DNA On collagen LDH
83 (80-86, n = 5)
65 (60-68, n =5)
78”
60(5&73,
81 (75-85, n = 3)
33 (29-39, n = 5) 47(43-53,n = 3)
55 58
63 (58-68, n = 2)
40(37-43,n
63
n = 5)
= 2)
Note. Attachment of cells was measured 16 to 20 hr after seeding. One hundred percent refered to the total number of cells estimated before freezing by cell counting, LDH, or DNA determination. The ranges of values and the numbers of determinations are given in parentheses. a This value was overestimated due to the loss of dead cells by centrifugation after thawing.
and thawing steps. A number of studies have shown that functional activity levels of freshly isolated and short-term cultured hepatocytes were dependent on the conditions of cell isolation and incubation (9). Cell damage induced by freezing is at least partly reversible since a longer period of culture would reduce functional differences between frozen and nonfrozen cells (5). The relative decrease in protein synthesis not exceeding 50% in our study was markedly less than the 80% reduction calculated by Fuller et al. (5) after 48 hr of culture. Whether protein synthesis was decreased to a similar extent in all attached cells was not determined but this seems quite unlike-
ly, not all having undergone similar alterations during the freezing-thawing process. It may be concluded that reproducible conditions of freezing and thawing hepatocytes defined in this study allow storage of cells which are still able to survive and function in culture. This procedure has been successfully applied to some human hepatocyte suspensions and is presently in use for the preparation of a bank of hepatocytes from various species.
504 . : nonf rozen . : frozen ‘ml
FIG. 8. Morphology of a cell monolayer obtained after a 24-hr culture of frozen and thawed rat hepatocytes (X 120).
FIG. 9. Incorporation of tritiated leucine into secreted proteins. Nonfrozen and frozen hepatocytes were seeded in 24-well dishes at different densities, cultured during 24 hr, and incubated with [3Hlleucine (10 uCi/ml, 500 pi/well) during 90 min. Media were collected; proteins were precipitated and the precipitates were washed and counted. The results were divided by and plotted against cellular proteins.
330
CHESNI.? AND ACKNOWLEDGMENTS
C. Chesnd was supported tinancially by IRIS (Institut de Recherches Intemationales SERVIER). We are grateful to Mrs. D. Glaise for technical assistance and A. Vannier for typing the manuscript. REFERENCES
1. Bradford, M. M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248-254 (1976). 2. Cesarone, C. F., Bolognesi, C., and Santi, L. Improved microfluorometric DNA determination in biological material using 3358 Hoechst. Anal. Biochem. 100, 188-197 (1979). 3. Collins, G. M., and Halasz, N. A. Kidney preservation for transplantation. Lancer 2, 1219 (1969). 4. Fuller, B. J., Grout, B. W., and Woods, R. J. Biochemical and ultrastructural examination of cryopreserved hepatocytes in rat. Cryobiology 19, 493-502 (1982).
5. Fuller, B. J., Morris, G. J., Nutt, L. H., and Attenburrow, V. D. Functional recovery of isolated rat hepatocytes upon thawing from - 196°C. Cryo-Lett. 1, 139-146 (1980). 6. Gomez-L. M. J., Lopez, P., and Castell, J. V. Biochemical functionality and recovery of hepatocytes after deep freezing storage. In Vitro 20, 826832 (1984). 7. Guguen, C., Guillouzo, C., Boisnard, M., Le Cam, A., and Bourel, M. Etude ultrastructurale de monocouches d’hepatocytes de rat adulte cultivts en presence d’hemisuccinate d’hydrocortisone. Biol. Gastroenterol. 8, 223-231 (1975). 8. Guguen-Guillouzo, C., and Guillouzo, A. Methods for preparation of adult and fetal hepatocytes. In “Isolated and Cultured Hepatocytes” (A. Guillouzo and C. Guguen-Guillouzo, Eds.), pp. 1-12. Les Editions INSERM Paris, John Libbey London Eurotext, 1986. 9. Guillouzo, A. Plasma protein production by cultured adult hepatocytes. In “Isolated and Cultured Hepatocytes” (A. Guillouzo and C. Guguen-Guillouzo, Eds.), pp. 155-170. Les Editions INSERM Paris, John Libbey London Eurotext, 1986.
GUILLOUZO
10. Jackson, B. A., Davies, J. E., andchipman, J. K. Cytochrome P-450 activity in hepatocytes following cryopreservation and monolayer culture. Biochem. Pharmacol. 34, 3389-3391 (1985). 11. Kemp, J. A., Hurt, S. N., Brown, J., and Clark, W. R. Recovery and function of human fetal pancreas frozen to - 196°C. Transplantation 32, l&l5 (1981). 12. Kusano, M., Ebata, H., Or&hi, T., Saito, T., and Mito, M. Transplantation of cyropreserved isolated hepatocytes into the rat spleen. Transplant. Proc. 13, 848-854 (1981). 13. Le Cam, A., Guillouzo, A., and Freychet, P. Ultrastructural and biochemical studies of isolated adult rat hepatocytes prepared under hypoxic conditions. Cryopreservation of hepatocytes. Exp. Cell Res. 98, 382-395 (1976). 14. Michalopoulos, G., and Pitot, H. C. Primary culture of parenchymal liver cells on collagen membranes. Morphological and biochemical observations. Exp. Cell Res. 94, 70-78 (1975). 15. Moore, C. J., and Gould, M. N. Metabolism of benzopyrene by cultured human hepatocytes from multiple donors. Carcinogenesis 5, 15771582 (1984). 16. Novicki, D. I., Irons, G. P., Strom, S. C., Jirtle, R., and Michalopoulos, G. Cryopreservation of isolated rat hepatocytes. In Vitro 18, 393-399 (1982).
17. Nutt, L. H., Attenburrow, V. D., and Fuller, B. J. Investigations into repair of freeze/thaw damage in isolated rat hepatocytes. Cryo-Lett. 1, 513-518 (1980). 18. Pertoft, H., Rubin, K., Kjellin,
L., Laurent, T. C., and Klingebom, B. The viability of cells grown or centrifuged in a new density gradient medium: Percoll (TM). Exp. Cell Res. 110,449457 (1977). 19. Renard, J. P. La congelation de I’embryon humain. Med. Sci. 2, 26-34 (1986). 20. Rijntjes, P. J. M., Moshage, H. J., Van Gemert, P. J. L., De Waal, R., and Yap, S. H. Cryopreservation of adult human hepatocytes. The influence of deep freezing storage on the viability, cell seeding, survival, tine structures and albumin synthesis in primary cultures. J. Hepatol. 3, 7-18 (1986).
25, 323-330 (1988)
Cryopreservation
of Isolated Rat Hepatocytes: A Critical Evaluation of Freezing and Thawing Conditions
CHRISTOPHE CHESNti AND ANDRIS: GUILLOUZO INSERM U 49, Unite! de Recherches Hkpatologiques, H6pital de Pontchaillou, Rennes, France Various parameters, including the nature and proportion of the constituents of the cryoprotective medium, the cooling rate, and the composition of the thawing medium, were evaluated for the cryopreservation of adult rat hepatocytes. The highest percentage of cells able to survive in culture was obtained by freezing in Ll5 medium containing 16% dimethyl sulfoxide, at a rate of 3”C/min, and by adding 0.8 M glucose to the thawing medium. More than 50% of hepatocytes capable of attachment just after cell isolation kept this property after freezing and survived in primary culture. Dead cells could be eliminated before seeding by centrifugation on a Percoll layer. In culture, frozen cells exhibited a morphology similar to that of unfrozen cells and after 24 hr their protein secretion rate was reduced by only about 40%. 8 1988 Academic PWSS, IX.
The hepatocyte is a valuable experimental model since it performs numerous functions, including synthesis and secretion of plasma proteins, production and storage of energy involving carbohydrate, lipid and amino acid metabolism, and drug metabolism. Marked species differences may exist in various functions between animal species and man. In addition, individual variations may be frequent, particularly in man. Indeed, human hepatocytes exhibit a genetic polymorphism with respect to their drug metabolic capacity. Hepatocytes can easily be isolated in high yields by collagenase perfusion of the liver. Four to six hundred million cells can be produced from a single rat liver and several billion hepatocytes can be produced from a large human liver sample (8). Such numbers of cells, particularly from human origin, are too large to be used just after their isolation. Moreover, the availability of livers from humans and many animal species is limited and cannot be routinely performed. Therefore, it is desirable to store cells in excess. The only conceivable method to achieve long-term storage of paReceived October 19, 1987; accepted February 29, 1988.
renchymal cells appears to be cryopreservation. A successful1 freezing protocol must yield stocks of cells that can be restored to full activity at any time point by thawing. Several freezing protocols have been described for adult hepatocytes. Cryopreservation of both animal and human liver cells has been reported, but in all cases, cell viability and biochemical functions were impaired (4, 5, 13, 15, 20). Freeze-induced damage appeared particularly pronounced when attempts to culture frozen hepatocytes were made (10, 16, 17). This led us to reconsider various aspects of the optimal conditions for hepatocyte cryopreservation using cell attachment to a support as a criterion for the evaluation of cell viability after freezing. MATERIAL
Isolation
AND METHODS
of Hepatocytes
The two-step collagenase perfusion method was used as previously described (7). The livers (180-200 g male Sprague Dawley rats) were first washed with a Ca*+-free Hepes (N-2-hydroxyethylpiperazine-N’2-ethane sulfonic acid)
323 OOll-2240/88 $3.00 Copyright All rights
Q 1988 by Academic Press, Inc. of reproduction in any form reserved.
324
CHESN6
AND
GUILLOUZO
buffer, and then dissociated with the same buffer containing 0.025% collagenase (Boehringer, Mannheim, FRG) and 5 mM CaCl,. After three washings, the cells were counted with a hemacytometer.
tectant, respectively. bation at room centrifugation (50g x purified over Percoll
Deep Freezing of Hepatocytes
Viable hepatocytes in 2-15 ml of medium were separated from dead cells by centrifugation (50g x 5 min) over a 15-ml cushion of a 36% Percoll solution in phosphatebuffered saline (PBS), d = 1.06 (18). Pelleted cells were 90-95% viable and the cells over the meniscus of Percoll were 90-95% dead, as estimated by the trypan blue exclusion test.
Hepatocytes were suspended (4 to 5 x lo6 cells/ml) in Leibovitz L15 medium containing 20% fetal calf serum (FCS), and then an equal volume of L15 medium containing 20% FCS and the cryoprotective agent was slowly introduced (5 min). The cell suspension was distributed in 1.6-ml freezing vials (Nunc, Roskilde, Denmark). After a 30-min incubation at room temperature, the vials were transferred to a cell freezer (Minicool LC40, Air Liquide, Sassenage, France) programmed at 3”C/min until liquid-solid phase transition and at l”C/min thereafter or placed in a - 70°C freezer. In this freezer, different cooling rates were obtained by using several holders: polystyrene packs with holes of the same diameter as the vials (2 or 3”C/min), hemolysis tube holders (6”C/min), and aluminium holders prerefrigerated at - 70°C with holes filled with ethanol (30”C/min) or not (12”C/min). Once the temperature of - 100°C was reached in the cell freezer or after 1 hr in the -70°C freezer, the vials were plunged in liquid nitrogen. Cells were preserved at - 196°C from some hours to several weeks. The slopes of cooling inside the vials were calculated from the curves registered between - 10 and -50°C with the thermic probe of the cell freezer in both the cell freezer and the - 70°C freezer. Thawing of Cells
The vials were quickly placed in a waterbath at 37°C and then the cell suspension from each vial was pipetted slowly and transferred in 15 ml of L15 medium. In all experiments, molarity of glucose was kept proportional to the concentration of cryoprotective agent: 0.5 and 0.8 M glucose corresponded to 10 and 16% (v/v) of cryopro-
Purification
After a lo-min incutemperature and 1 min), the cells were or seeded directly.
of Viable Cells
Culture of the Cells
Hepatocytes were seeded at a density of 7 x IO5cells per petri dish (35 mm in diameter) in 2 ml of the culture medium supplemented with 10% FCS and incubated at 37°C in a humidified atmosphere composed of 95% air-5% CO,. The medium was a mixture of 75% minimal essential medium and 25% 199 medium, buffered with bicarbonate (0.22% w/v), and supplemented with bovine serum albumin (0. I%), bovine insulin (10 t&ml), and antibiotics (penicillin 10 IU/ml, streptomycin 50 pg/ml). Some dishes were coated with rat tail collagen films (14). To obtain these films, collagen was spread on the bottom of the dishes, dried overnight at room temperature, and rinsed with PBS before seeding. Measurement
of Cell Viability
Cell viability was examined both before seeding and after attachment to a support. Viability of freshly isolated or thawed cells was calculated after a 5-min incubation in a 0.05% trypan blue solution in PBS. Viability of attached cells was measured 16-20 hr after seeding. Cell monolayers were rinsed twice with PBS and sonicated (10 set, 10 W, with a Ultrasonic sonicator, New York) in 2 ml of PBS. Intracellular lactate dehydrogenase (LDH) activity was determined using a LDH kit (Roche) on an automatic
CRYOPRESERVATION
325
OF RAT HEPATOCYTES RESULTS
analyzer (Bio-Cobas, Roche, Switzerland). The amount of DNA was measured fluorometrically (2) on the same apparatus. All experiments were performed at least twice. Each value of cell attachment was determined from three or four dishes. The values were statistically treated by the one-way analysis of variance and the Newman Keul’s test.
In preliminary experiments (Fig. l), dimethyl sulfoxide (Me,SO), glycerol, and 1,Zpropanediol were tested as cryoprotective agents at a wide range of concentrations and at a low cooling rate (3”Umin); Me,SO was found to be more efficient than glycerol and was selected for further studies. 1,2-Propanediol, though better than glycerol, was markedly less effective than Determination of Protein Synthesis Before and after freezing, each hepato- Me,SO. The appropriate proportion of Me,SO cyte suspension was seeded at different was then determined. The curve of cell vidensities in 24-well dishes. After 24 hr of ability versus the percentage of Me,SO was culture, cell monolayers were washed with rather sharp (Fig. 2). When taking into acPBS and incubated at 37°C in 0.5 ml of culcount both efficiency and toxicity, a conture medium containing 10 &i/ml of centration of Me,SO as high as 16% was [3H]leucine (sp act 0.7 Ci/mmol, Amerfound to give the best results. This optisham, France). After 90 min, media were mum did not change with varying cooling collected and the proteins were precipitated rates (Fig. 2). The optimal incubation time with trichloroacetic acid. The precipitates with Me,SO was around 30 min (Fig. 3). were washed, dissolved in formic acid, and The influence of temperature during incucounted. Cell monolayers were sonicated (5 bation was assayed at 4, 22, and 37°C (Fig. set, 10 W) in 0.5 ml PBS and proteins were measured using the Bradford’s method (1). Each determination was performed in triplicate .
= -2
g 4002
600.
= bte,SO g
400.
200.
A
200
6: i
I / 10
I
I I
$ Propanedid
.I
iGlycerol
14
16
Cryoprotective agent (‘I01 FIG. 1. Effect of the nature of the cryoprotective agents (CA) tested. Hepatocytes were incubated for 20 min with each CA in the culture medium containing 20% FCS and then cooled at 3Wmin. Intracellular LDH was measured in thawed cells after 20 hr of culture. The values are means -t SE (n = 3). **P < 0.01 vs 14% glycerol and 16% 1,2-propanediol. Me,SO was much more efficient than the two other agents and was selected at its optimum concentration, i.e., 16%, in the regime for freezing; under these conditions, the viability after freezing was 30%. As in the following figures, this value was estimated by measurement of intracellular LDH in both seeded and attached cells.
Cryoprotective
agent(%)
FIG. 2. InlIuence of the concentration of Me,SO in the cryoprotective medium at several cooling rates. Hepatocytes were suspended in Ll5 medium containing different proportions of Me,SO and 20% FCS were frozen in a - 70°C freezer at 3”C/min (in a polystyrene pack), 6Wmin (in a hemolysis tube holder), and 30Wmin (in a prerefrigerated aluminium holder with holes containing ethanol). Intracellular LDH was measured in thawed cells after 20 hr of culture. The values are means -C SE (n = 3). **P < 0.01 vs 7, 10, 12, 15, and 20% Me,SO. The optimum concentration of Me,SO was independent on the cooling rate. A 16% concentration was chosen at a cooling rate of 3”C/min; under these conditions the viability after freezing was 33%.
326
CHESNfi
AND
3). There was no difference of attachment when the cell suspensions were incubated at 4 and 22°C but a 30-min incubation at 37°C caused a 50% decrease of viability. The influence of the composition of the aqueous part of the cryoprotective medium was also investigated. Three buffers, namely, Hepes, PBS, and Euro-Collins used for kidney preservation (3), the culture medium (75% minimal essential medium, 25% 199 medium), and L15 medium were tested (Fig. 4). L15 medium gave the best results and was retained. The influence of FCS concentration in the cryoprotective medium was also estimated (Fig. 5). Better results were observed when the cell suspension was diluted in pure FCS but since the difference with lower FCS concentrations was only slight, further experiments were carried out in the presence of 20% FCS. Figure 6 shows that slow cooling rates were better tolerated than faster rates. However, they could lead to sedimentation of the cells during the liquid phase of cooling, since hepatocytes are dense cells and sediment quickly. This phenomenon was found to be toxic and decreased the viability by three- to fivefold. It could be partially
GUILLOUZO
1600
1400
1200
1000 7
5 ;
600..
: a
600-
400. 200-
/IS’ s/25
115
FIG. 4. Effect of the nature of the cryoprotective medium. Hepatocytes were suspended in several media: Hepes buffer used for cell isolation, PBS buffer, L15 medium, culture medium (MEM-199), and EuroCollins buffer. All these media contained 16% Me,SO but no FCS. The incubation lasted 30 min at 22°C except for the Collins buffer which was maintained at 4°C. Cells were frozen at 3Wmin. Intracellular LDH was measured in thawed cells after 20 hr of culture. The values are means f SE (n = 4). **P < 0.01 vs all other media. Ll5 medium gave the best results and was chosen in the regime for freezing; under these conditions, the viability after freezing was 3%.
circumvented by the presence of FCS (because of its viscosity) or quick freezing. However, we preferred to prevent cell sedimentation by stirring the vials during the liquid phase of the cooling procedure. The influence of the cooling rate was not as critical as that of the proportion of cryoprotecL 0 20 40 so tive agent. This led us to select the classical Incubation time (min) cooling equipment using a - 70°C freezer, a FIG. 3. Influence of the time and temperature of simple method well adapted for routine incubation of the cells in the cryoprotective medium work. The cooling rates obtained in a before freezing. Hepatocytes were incubated with 16% Me,SO in Ll5 medium containing 20% FCS during 3 - 70°C freezer were linear until - 50°C and to 60 min and were frozen at 3Wmin. Intracellular the liquid-solid phase transition was only LDH was measured in thawed cells after 20 hr of cul- moderately prolonged compared to that obture. The values are means f SE (n = 3). **P < 0.01 vs 3 min at 22°C and 35 min at 37°C. A 30-min incu- tained with a programmed cell freezer. As expected from the values of cooling bation at room temperature was selected; under these conditions, the viability after freezing was 33%. rates, hepatocyte injury was reduced by
CRYOPRESERVATION
OF RAT
327
HEPATOCYTES
zoo. 6 0
10
20
40
II *’
I a4
Fetal calf serum( %) FIG. 5. Effect of the proportion of FCS in the cryoprotective medium. Hepatocytes were suspended in Ll5 medium containing 16% Me,SO and different proportions of FCS and were frozen at 3Wmin. Intracellular LDH was measured in thawed cells after 20 hr of culture. The values are means f SE (n = 3). **P < 0.01 vs 0%. No marked difference was observed with FCS concentrations equal or superior to 5%. A 20% concentration was chosen in the regime for freezing; under these conditions, the viability after freezing was 36%.
rapid thawing (in a 37°C water bath) compared to a slower thawing (in ambient air at 22°C): the viability was about 15-20% higher. Addition of glucose to the thawing medium limited the osmotic shock caused by the outflow of cryoprotective agent from the cells. The most effective concentration of glucose was obtained between 0.6 and 1 M when a 16% Me,SO concentration was used during the freezing step (Fig. 7). When the optimized freezing and thawing conditions were employed, the average viability values measured by the trypan blue exclusion test were about 65% for frozen cells versus 83% for nonfrozen cells (Table 1). This test of viability is useful for shortterm experiments on hepatocyte suspensions. However, the values of recovery were overestimated since a fraction of dead cells were eliminated during washing. In addition, the trypan blue test does not give any indication on the capability of thawed cells to be maintained in culture. To evaluate this property, cell attachment was assayed 16 to 20 hr after hepatocyte seeding. This time was selected because frozen cells attached and spread more slowly than non-
10
30
” Cooling
60 rate Vqmin)
FIG. 6. Effect of the cooling rate. Hepatocytes were suspended in Ll5 medium containing 16% Me,SO and 20% FCS and were frozen in a -70°C freezer in a polystyrene pack with a cover (TUrnin) or without (3”C/min), on a hemolysis tube holder (6YYmin) or in a prerefrigerated (at -70°C) aluminium holder with holes containing ethanol (30Wmin) or not (lTC/min), or were plunged directly in liquid nitrogen (6OWmin approximately). After 1 hr in the - 70°C freezer, vials were transferred in liquid nitrogen. Intracellular LDH was measured in thawed cells after 20 hr of culture. The values are means + SE (n = 3). **P < 0.01 vs 6, 12, 30, and 60Wmin. A 3Wmin cooling rate was found to be the most efficient and was selected in the regime for freezing; under these conditions, the viability after freezing was 34%.
frozen cells. The percentage of attached cells was determined by measuring intracellular LDH and DNA content. Both values were directly compared with those calculated from the freshly isolated cell suspension. Lower values were obtained by LDH measurement: they were 62 and 33% for nonfrozen and frozen cells, respectively, while they reached 81 and 47% when DNA content was used as the endpoint for cell attachment evaluation (Table 1). In order to improve cell attachment, a collagen-coated substrate was also tested. A 15% increase in cell attachment was observed when collagen-coated plastic was used as a support (Table 1). To limit attachment of nonviable cells and to obtain confluent cell monolayers from frozen cell suspension, viable cells were separated from nonviable ones over a Percoll layer. Cell monolayers obtained from frozen cells purified over a Percoll layer looked similar to those obtained from nonfrozen cells (Fig. 8).
328
CHESNl?
I
0
a2
0.6
1 Glucose
AND
1.4 (M)
FIG. 7. Effect of the glucose in the thawing medium. Hepatocytes were frozen at 3”Umin in L15 medium containing 20% FCS and 16% Me,SO. The cells were then thawed in L15 medium containing glucose at different concentrations. This step of thawing lasted 10 mitt before centrifugation. Intracellular LDH was measured in thawed cells after 20 hr of culture. The values are means f SE (n = 3). **P < 0.01 vs 0 M glucose. Glucose had a significant effect at 0.8 A4 and was used at this concentration in the regime for freezing; under these conditions, the viability after freezing was 28%.
After 24 hr of culture, protein secretion rate was measured by incorporation of labeled leucine. Whatever the cell density, this cellular activity still represented 5060% in frozen cell cultures compared to the value calculated in nonfrozen cultured hepatocytes (Fig. 9). DISCUSSION
Several studies on various cell types have shown that the composition of the freezing medium and the cooling rate are fundamental parameters in the preservation of cell viability after freezing (19). The results reported here clearly indicate that these parameters are also critical for hepatocyte freezing. As already shown (16) Me,SO was found to be more efficient than glycerol. The best results were obtained with a concentration of 16% which is higher than that used by others (5, 12, 13, 16). However, it may be mentioned that Novicki et al. (16) had noticed that Me,SO gave identical results at 10 and 20%. As observed with pancreatic islets (1 l), a 30-min incubation with this cryoprotective agent
GUILLOUZO
before freezing at 4 or 22°C did not alter cell viability, while a similar incubation at 37°C caused a significant decrease in cell attachment after the freezing-thawing steps. Most authors have used a cooling rate of 1 or 2”C/min. Thus, Le Cam et al. (13) found that cooling rates below 7YYmin were more efficient than faster rates. To our knowledge, only Gomez et al. (6) have recommended a higher cooling rate (39YYmin) obtained by plunging the vials in liquid nitrogen. By contrast, Fuller et al. (4) had found this protocol particularly inefficient. We show that slow cooling rates (around 3”C/min) were better tolerated provided that cell sedimentation was prevented by stirring. The values of cell recovery after freezing greatly varied according to the end point used and the time of measurement. When determinations were made after cell attachment, it may be assumed that only functional cells were counted. Lower values were obtained by intracellular LDH measurement than by DNA determination. This could be explained at least partly by plasma membrane alterations occurring during freezing and thawing processes and resulting in the loss of intracellular LDH. Values obtained with both parameters were higher than most of those previously reported by others (16, 17): the highest values for cell attachment after freezing estimated by various end points did not exceed 20-25% of the total cell population. In this study, depending on the parameter used, 3347% of the initial cell population were still capable of attachment just after freezing. Only Jackson et al. (10) found a similar percentage of cell attachment after freezing: however, this is somewhat surprising since it is only slightly lower than the value determined by the trypan blue exclusion test (42%). After freezing, protein synthesis was decreased probably because of the occurrence of some alterations during freezing
CRYOPRESERVATION
OF RAT
329
HEPATOCYTES
TABLE 1 Viability of Frozen and Nonfrozen Hepatocytes
Cell viability (%) Criterion
Before freezing
After freezing
Cell recovery (%)
Trypan blue test Attachment On plastic LDH DNA On collagen LDH
83 (80-86, n = 5)
65 (60-68, n =5)
78”
60(5&73,
81 (75-85, n = 3)
33 (29-39, n = 5) 47(43-53,n = 3)
55 58
63 (58-68, n = 2)
40(37-43,n
63
n = 5)
= 2)
Note. Attachment of cells was measured 16 to 20 hr after seeding. One hundred percent refered to the total number of cells estimated before freezing by cell counting, LDH, or DNA determination. The ranges of values and the numbers of determinations are given in parentheses. a This value was overestimated due to the loss of dead cells by centrifugation after thawing.
and thawing steps. A number of studies have shown that functional activity levels of freshly isolated and short-term cultured hepatocytes were dependent on the conditions of cell isolation and incubation (9). Cell damage induced by freezing is at least partly reversible since a longer period of culture would reduce functional differences between frozen and nonfrozen cells (5). The relative decrease in protein synthesis not exceeding 50% in our study was markedly less than the 80% reduction calculated by Fuller et al. (5) after 48 hr of culture. Whether protein synthesis was decreased to a similar extent in all attached cells was not determined but this seems quite unlike-
ly, not all having undergone similar alterations during the freezing-thawing process. It may be concluded that reproducible conditions of freezing and thawing hepatocytes defined in this study allow storage of cells which are still able to survive and function in culture. This procedure has been successfully applied to some human hepatocyte suspensions and is presently in use for the preparation of a bank of hepatocytes from various species.
504 . : nonf rozen . : frozen ‘ml
FIG. 8. Morphology of a cell monolayer obtained after a 24-hr culture of frozen and thawed rat hepatocytes (X 120).
FIG. 9. Incorporation of tritiated leucine into secreted proteins. Nonfrozen and frozen hepatocytes were seeded in 24-well dishes at different densities, cultured during 24 hr, and incubated with [3Hlleucine (10 uCi/ml, 500 pi/well) during 90 min. Media were collected; proteins were precipitated and the precipitates were washed and counted. The results were divided by and plotted against cellular proteins.
330
CHESNI.? AND ACKNOWLEDGMENTS
C. Chesnd was supported tinancially by IRIS (Institut de Recherches Intemationales SERVIER). We are grateful to Mrs. D. Glaise for technical assistance and A. Vannier for typing the manuscript. REFERENCES
1. Bradford, M. M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248-254 (1976). 2. Cesarone, C. F., Bolognesi, C., and Santi, L. Improved microfluorometric DNA determination in biological material using 3358 Hoechst. Anal. Biochem. 100, 188-197 (1979). 3. Collins, G. M., and Halasz, N. A. Kidney preservation for transplantation. Lancer 2, 1219 (1969). 4. Fuller, B. J., Grout, B. W., and Woods, R. J. Biochemical and ultrastructural examination of cryopreserved hepatocytes in rat. Cryobiology 19, 493-502 (1982).
5. Fuller, B. J., Morris, G. J., Nutt, L. H., and Attenburrow, V. D. Functional recovery of isolated rat hepatocytes upon thawing from - 196°C. Cryo-Lett. 1, 139-146 (1980). 6. Gomez-L. M. J., Lopez, P., and Castell, J. V. Biochemical functionality and recovery of hepatocytes after deep freezing storage. In Vitro 20, 826832 (1984). 7. Guguen, C., Guillouzo, C., Boisnard, M., Le Cam, A., and Bourel, M. Etude ultrastructurale de monocouches d’hepatocytes de rat adulte cultivts en presence d’hemisuccinate d’hydrocortisone. Biol. Gastroenterol. 8, 223-231 (1975). 8. Guguen-Guillouzo, C., and Guillouzo, A. Methods for preparation of adult and fetal hepatocytes. In “Isolated and Cultured Hepatocytes” (A. Guillouzo and C. Guguen-Guillouzo, Eds.), pp. 1-12. Les Editions INSERM Paris, John Libbey London Eurotext, 1986. 9. Guillouzo, A. Plasma protein production by cultured adult hepatocytes. In “Isolated and Cultured Hepatocytes” (A. Guillouzo and C. Guguen-Guillouzo, Eds.), pp. 155-170. Les Editions INSERM Paris, John Libbey London Eurotext, 1986.
GUILLOUZO
10. Jackson, B. A., Davies, J. E., andchipman, J. K. Cytochrome P-450 activity in hepatocytes following cryopreservation and monolayer culture. Biochem. Pharmacol. 34, 3389-3391 (1985). 11. Kemp, J. A., Hurt, S. N., Brown, J., and Clark, W. R. Recovery and function of human fetal pancreas frozen to - 196°C. Transplantation 32, l&l5 (1981). 12. Kusano, M., Ebata, H., Or&hi, T., Saito, T., and Mito, M. Transplantation of cyropreserved isolated hepatocytes into the rat spleen. Transplant. Proc. 13, 848-854 (1981). 13. Le Cam, A., Guillouzo, A., and Freychet, P. Ultrastructural and biochemical studies of isolated adult rat hepatocytes prepared under hypoxic conditions. Cryopreservation of hepatocytes. Exp. Cell Res. 98, 382-395 (1976). 14. Michalopoulos, G., and Pitot, H. C. Primary culture of parenchymal liver cells on collagen membranes. Morphological and biochemical observations. Exp. Cell Res. 94, 70-78 (1975). 15. Moore, C. J., and Gould, M. N. Metabolism of benzopyrene by cultured human hepatocytes from multiple donors. Carcinogenesis 5, 15771582 (1984). 16. Novicki, D. I., Irons, G. P., Strom, S. C., Jirtle, R., and Michalopoulos, G. Cryopreservation of isolated rat hepatocytes. In Vitro 18, 393-399 (1982).
17. Nutt, L. H., Attenburrow, V. D., and Fuller, B. J. Investigations into repair of freeze/thaw damage in isolated rat hepatocytes. Cryo-Lett. 1, 513-518 (1980). 18. Pertoft, H., Rubin, K., Kjellin,
L., Laurent, T. C., and Klingebom, B. The viability of cells grown or centrifuged in a new density gradient medium: Percoll (TM). Exp. Cell Res. 110,449457 (1977). 19. Renard, J. P. La congelation de I’embryon humain. Med. Sci. 2, 26-34 (1986). 20. Rijntjes, P. J. M., Moshage, H. J., Van Gemert, P. J. L., De Waal, R., and Yap, S. H. Cryopreservation of adult human hepatocytes. The influence of deep freezing storage on the viability, cell seeding, survival, tine structures and albumin synthesis in primary cultures. J. Hepatol. 3, 7-18 (1986).