- Email: [email protected]
Cryopreserved porcine hepatocyte cultures
Chemico-Biological Interactions 121 (1999) 99 – 115
Cryopreserved porcine hepatocyte cultures H.G. Koebe *, B. Mu¨hling, C.J. Deglmann, F.W. Schildberg Department of Surgery, Klinikum Grosshadern, Ludwig-Maximilians-Uni6ersity of Munich, Marchioninistrasse 15, D-81377 Munich, Germany Received 17 February 1999; accepted 9 March 1999
Abstract Cryopreservation of freshly isolated hepatocytes is regarded the standard technique for long term storage of liver cells. Frankly, we were not successful in reproducing viability rates of about 70% of that which have been reported by most authors as results of various freezing protocols for hepatocyte suspensions. In fact, we saw mostly devastating results. We assume that intracellular ice crystal formation as well as osmotic changes during freezing and thawing of liver cells cause hazardous effects, especially on membranes of cells after enzymatic isolation, and, thus, generally result in a severe loss in number and impaired specific hepatocyte functions in subsequent culture. We tried to improve results by freezing cell cultures instead. We allowed hepatocytes to regain a more stable condition prior to storage and placed them in tissue flasks in a uniform configuration, rather than to pack cell suspensions in vials or bags under rather indefinable conditions. Porcine hepatocytes (viability 929 2%) were isolated from slaughterhouse organs and cultured in a double gel (sandwich) configuration. At day 3, cultures were rate controlled frozen (RCF) and stored in a cell bank for three hours (Group A) or 14 days at − 80°C (Group B), respectively. Non-frozen cells (NF) and cultures subjected to a linear freezing rate of −10°C/min (LFR, Group C) with 3 h of storage served as controls from identical cell batches. Upon thawing, at day 2 of subsequent culture, fluorescence microscopy studies revealed a survival rate of 75910% (mean9S.D.) in the RCF groups. Time of storage (3 h, 14 d) did not influence
* Corresponding author. Tel.: +49-897095-6432; fax: + 49-897095-6433. E-mail address: koebe – [email protected] (H.G. Koebe) 0009-2797/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 0 0 9 - 2 7 9 7 ( 9 9 ) 0 0 0 9 3 - 9
100
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
results. Survival in Group C was 10 9 5%. Cell integrity was measured by LDH-release, which indicated a larger damage of cells in the LFR group, and thereby resembled the morphological findings. Functional parameters, such as albumin synthesis and CYT P 450-activity were comparable to non-frozen liver cells at 48 h after thawing in the RCF groups (A+B), and showed significantly higher levels in these groups as compared to the LFR Group (C). We recommend to freeze hepatocytes in culture in a rate controlled fashion rather than cell suspensions. This way a cell bank of cryopreserved hepatocyte cultures for batch controlled investigations can be easily obtained. This could ameliorate the availability of rare (human) cell material and might also improve the quality of data generated from in vitro experiments in hepatology or pharmacology/toxicology with liver cells from identical sources. It remains to be seen whether this technique might also be of value in hybrid bioartificial liver devices to make these systems become readily available upon clinical demand. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Bioartificial liver; cell culture; cryopreservation; hepatocytes; hepatology; pharmacology/toxicology
1. Introduction The discovery of cryoprotectants (substances with cryoprotective properties) such as dimethyl sulfoxide (Me2SO), 1,2-propanediol, glycerol and others made successful cryopreservation of various cell types a reachable goal [1]. For the cryopreservation of hepatocytes in suspension, many protocols have been proposed and freezing was suggested as a standard procedure for long-term storage of liver cells [2–6]. Cryopreserved hepatocytes would be of considerable value for investigations in the field of hepatology and pharmacology/toxicology [7–9], offering the option to perform repeated investigations on well defined cell batches. Also, the development of bioartificial liver devices highly depends on readily available functional units, i.e. liver cells, for on-demand supply [10–13]. Cryopreserved hepatocytes stored in batch-controlled cell banks could provide the biomaterial for all these applications. However, despite the addition of cryoprotectants, liver cells are very susceptible to freeze-thaw damage and since they do not replicate in culture, special cryopreservation techniques are needed to reduce cell injury and functional impairment of hepatocytes. Most studies so far have investigated on freshly isolated and cryopreserved rat hepatocytes in suspension. The results reported so far vary considerably with survival rates between 30 and 80% in the dye exclusion test [14–17]. It is well known, however, that trypan blue exclusion initially after thawing, does not necessarily correlate to desired cell functions in subsequent culture like attachement, synthesis, or metabolising capacity [18,19]. We have already reported on the successful cryopreservation of cultured hepatocytes [20 – 22] and were supported by the findings of others [23]. Liver cells attached to microcarriers [24] or encapsulated in a semipermeable membrane of polylysine and/or alginate [25] have also been cryopreserved in the
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
101
past with rather little information available on the actual influence of cryopreservation on cell performance as compared to non frozen cells. We used the Double Gel (sandwich) culture configuration instead, which provides a stable three dimensional matrix in which hepatocytes maintain their morphological and functional differentiation for long term periods [26]. In our first study on the feasibility of freezing double gel cultures, survival rates upon thawing were 30% and albumin secretion decreased to 20% of controls [20]. The objective of the present study was to improve these results and to establish a batch controlled cell bank of porcine hepatocytes. For this purpose, we modified the freezing program to prevent temperature rises within the culture due to generated latent heat of fusion. Temperature was measured on-line within the sample and rate controlled freezing was carried through accordingly (Groups A + B). In order to underscore the importance of rate controlled freezing, a different program was used (Group C) that led the specimen through a linear decline in temperature which was likely to result in cell damage through intracellular ice formation. For our study we used pig hepatocytes from slaughterhouse organs [27,28]. They represent an unlimited resource for applied experiments [27– 29] in pharmacology and hepatology. Also, pigs have been suggested as donor species for artificial liver devices that are currently being tested in experimental research and clinical trials [10 – 13,22].
2. Materials and methods
2.1. Reagents and solutions Type IV collagenase, Dulbecco‘s modified Eagle‘s medium (Dulbecco’s Modified Eagles Medium) with 4.5g/l glucose, Me2SO and Leibovitz L-15 medium were from Serva Feinbiochemica GmbH (Heidelberg, Germany). The perfusion buffer contained 154 mmol/l sodium chloride, 5.6 mmol/l potassium chloride, 5 mmol/l glucose, 25 mmol/l sodium bicarbonate and 20 mmol/l HEPES (pH 7.4). The washing buffer contained 120 mmol/l sodium chloride, 6.2 mmol/l potassium chloride, 0.9 mmol/l calcium chloride, 10 mmol/l HEPES and 0.2% (w/v) bovine albumin (pH 7.4). Hydroxycoumarin and ethoxycoumarin, DNA and albumin standards were from Sigma (Deisenhofen, Germany). Antibodies were from Nordic Immunology (Tilburg, NL). All other chemicals were of reagent grade and from various commercially available sources.
2.2. Li6er perfusion and cell isolation We obtained the livers of landrace piglets (both sexes, 4–8 weeks old, weighing 15– 35 kg) from the Munich slaughterhouse. The isolation procedure is described in detail elsewhere [22]; briefly: livers were excised by butchers, put into a plastic bag and transported in ice water to the laboratory within a maximum time of 20 min. In the laboratory, the left liver lobe was dissected from the rest of the organ and
102
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
perfused via the left hepatic vein; perfusion buffer was oxygenated by carbogen for the whole procedure. First the lobe was perfused at 37°C at a flow of 50 ml/min with 1000 ml of buffer containing 1 mmol/l EDTA. Next it was perfused with 300 ml of buffer without EDTA and finally a recirculating perfusion with 200 ml of buffer containing 0.05% collagenase (w/v) and 5 mmol/l calcium chloride was performed; the temperature and flow rate were the same. Next the tissue was dissected with a scalpel blade and the homogenates were shaken for 3 min after adding collagenase solution. The resulting cell suspension was filtered through two nylon meshes (210 and 70 mm grid size). The filtered suspension was filled with 4°C cold washing buffer to a volume of 500 ml and washed three times in a centrifuge at 23 g for 5 min. To separate viable hepatocytes from dead cells, non parenchymal cells and debris, a Percoll™ (Pharmacia Biotechnology, Uppsala, Sweden) gradient centrifugation with 10.8 ml of Percoll™ and 1.2 ml of 10 × DMEM and 17 ml of cell suspension was performed at 76 g. The cell pellets of purified hepatocyte suspension were washed twice with washing buffer at 23 g for 5 min and finally the hepatocytes were resuspended in Leibovitz L-15 medium containing 0.2% (w/v) bovine albumin. Cell viability as judged by the trypan blue exclusion test was 92 9 2% with an average cell yield of 1.0 9 10E7 hepatocytes per g perfused liver tissue.
2.3. Hepatocyte cultures The culture technique is described in detail elsewhere [21,26]; briefly: culture medium for the first 24 h consisted of Dulbecco’s Modified Eagle Medium supplemented with insulin (125 mU/ml), hydrocortisone (60 ng/ml), glucagon (10 ng/ml), gentamicin(100 mg/ml), penicillin (100 mU/ml) and 5% (v/v) fetal calf serum. For the sandwich culture configuration T-flasks (25 cm2) were precoated with 1 ml of collagen solution by mixing 1 part of 10× DMEM (pH 7.4.) and 9 parts of collagen (0.83 mg/ml) to a final collagen concentration of 0.75 mg/ml. Type I collagen was prepared from rat tail tendons. After gelling of the matrix in a humidified CO2 controlled incubator [37°C, 5% (v/v) CO2] 2 ml of cell suspension (1.5 × 10E6 cells/ml) was distributed on the culture flasks. To allow attachment of hepatocytes, cultures were kept in the incubator for the first 24 h; next, cell medium was replaced by the second layer of collagen. Two ml of culture medium (Serum free) were added and changed every 48 h. Probes for analysis were collected at the given intervals and stored at −20°C before measurement.
2.4. Cryopreser6ation On day three of culture, cells were incubated with DMSO as a cryoprotectant at a concentration of 20% (v/v) prior to freezing to achieve a final concentration of 10% (v/v) within the culture matrix. Cells were stored in the incubator for 15 min at 37°C. Following removal of the supernatant, the T-flasks were put in a metal rack (Fig. 1) which was stored at a defined position in the freezing chamber of a programmable freezing unit (Nicool ST 20, L’Air Liquide, Sassenage, France; Fig.
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
103
2). Temperature levels were recorded at two defined sites within the chamber (Philips recorder PM 8251, Hewlett-Packard standard multimeter) to ensure controlled cooling of the specimen. Measurement of the actual temperature within those hepatocyte cultures that served as samples was performed by using a platinum resistor PT 100 (MRIV, Munich, Germany) which was dipped in the culture matrix (Fig. 3). This technique allows a close follow up of the measured temperature curves within the samples and supports an accurate steering of the Nitrogen inlet. To avoid random differences, three independent experiments were performed with n= 4 cultures for each group. This way, three consecutive cell preparations led to an overall of n= 12 cultures per group. n= 3 non-frozen cultures per isolation formed a group of n= 9. The recorded temperature curves (Groups A+B and Group C, respectively) were almost identical in all three experimental set ups (data not shown). Three experimental groups were included: Group A (n=12): Cells were exposed to the rate controlled freezing program (RCF, Table 1; Fig. 4) and stored for 3 h at − 80°C. Group B (n=12) Same program as under Group A, but stored for 14 days at − 80°C. Group C (n= 12) Cells underwent a linear freezing program (LFP, Table 2) at – 10°C/min and were stored for 3 h at − 80°C. Control (n=9) No freezing.
Fig. 1. Hepatocyte culture T-flasks in a metal rack for up to 12 cultures with 3 ×106 cells each.
104
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
Fig. 2. Programmable freezing unit Nicool ST 20.
2.5. Thawing and remo6al of the cryoprotectant After storage, hepatocyte cultures were thawed by adding 5 ml of a 37°C warm 10% (v/v) solution of DMSO and placing the T-flasks in a waterbath at 40°C for 10 sec. Next, the thawing solution was removed and the cultures were incubated with decreasing Me2SO solutions (7.5, 5, 2.5, and 0%); each solution was added for 5 min at + 4°C to remove the cryoprotectant. Finally, hepatocyte cultures were transferred into the incubator and cultured in cell medium at 37°C.
2.6. Viability assay In order to compare the number of viable cells in cryopreserved cultures with unfrozen controls, the determination of cell viability was set at day five of culture (48 h after thawing) in all groups by staining with fluorescein diacetate and propidium iodine [36]. Briefly, fluorescein diacetate (Sigma Chemie, Deisenhofen, Germany) was dissolved in acetone (1 mg/ml) and frozen at − 20°C before use. Medium was aspirated and replaced by fresh medium (1 ml)+ 5 ml of fluorescein diacetate and propidium iodinesolution. Cultures were held for 5 min at 37°C before being washed in PBS for 20 min at 4°C. Next, fluoroscopic images were captured using a Nikon Diaphot 200 inverted microscope equipped for fluorescence (B2A filter, Duesseldorf, Germany). Two microscopic fields per culture dish were randomly selected and viable cells stained with fluorescein diacetate (green) as well
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
105
Fig. 3. Tissue flask with implanted PT-100 probe for in matrix temperature registration (Demo-Matrix without cells in culture).
as iodine stained nuclei of cells (orange) with destroyed membranes were counted. The number of viable cells of posthypothermic cultures was calculated versus control culture dishes and expressed as percent 9 S.D.
2.7. Biochemical assays Probes for biochemical assessments were collected every 48 h from the first day of culture to the 11th day. Albumin was determined immunologically by using an antibody to porcine albumin (rat-anti-pig) and a peroxidase-conjugated antibody (goat anti-rat IgG peroxidase). Extinction was measured at 450 nm with a Titertek Multiscan Plus Reader (MK II, Flow Laboratories, Meckenheim, Germany). Cytochrome P 450 activity (formation of Hydroxycoumarin) was measured fluorometrically by a modified method according to Edwards [30]. LDH-release was determined in the supernatant photometrically at 339 nm; pyruvate and NADH/ H+ served as substrates. DNA content of hepatocytes was analysed on a TKO 100 Table 1 Rate controlled freezing program (RCF, Groups A+B) Temperature (°C)
Freezing rate
4 to 0 0 to −11 −11 −11 to −80
−1°C/min 0.5°C 15 min −3°C/min
106
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
Fig. 4. Freezing protocol under the rate controlled program. Prior to the final freezing step, cultures were held at –11°C to complete extracellular ice formation. Nitrogen inflow was regulated to compensate for the generation of latent heat of fusion at this point to prevent cultures from osmotic stress due to phase changes.
fluorometer (Hoefer Scientific Instruments, San Francisco, CA; lex = 365 nm; lem =460 nm). Samples were measured three times, each, to avoid random differences. Values are given as mean 9S.D.
3. Results
3.1. Cell morphology Figs. 5 and 6 show the fluorescence microscopic appearance of hepatocytes in groups A and B after 3 h and 14 days storage, respectively. Upon thawing, after 48 h of subsequent culture, the number of hepatocytes was only slightly reduced and viable cells (75 9 10%) showed no remarkable changes in morphology. In contrast, cultures of group C (linear freezing) mainly appeared with stained nuclei of dead cells; viability in this group was 109 5% (Fig. 7). Fig. 8 shows controls, 5th day of culture. Table 2 Linear freezing program (LFP, Group C) Temperature (°C)
Freezing rate
4 to −80
−10°C/min
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
107
Fig. 5. Fluorescence microscopy of Group A cells (rate controlled program, 3h storage), 48 h after thawing, Magn. 300 × .
3.2. Albumin secretion Albumin secretion in groups A, B, C, and control from day one up to day 11 of culture is shown in Fig. 9. Control cultures (non-frozen) showed increasing values from day 5 of culture with a maximum albumin secretion at day 9 (105.3 ng/h/mg DNA). Albumin secretion in groups A and B declined markedly after the freezethaw-cycle, but reached a maximum level on day 9 of culture (A: 127 ng/h/mg DNA, B: 78.4 ng/h/mg DNA). Cultures of Group C almost ceased albumin secretion after thawing and values approached zero.
3.3. Ethoxycoumarin-O-Deethylase-acti6ity Metabolic activity (CYP 450) in unfrozen hepatocytes declined gently from day 1 through 11. Cultures, preserved with the rate controlled program, showed a marked loss of CYP 450 activity upon thawing (Groups A+B), but cells regained their function up to the level of the control groups at day 4 post cryopreservation. In contrast, Group C showed a steady decline towards zero values (Fig. 10).
108
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
3.4. LDH-release LDH-release in the control group displayed a stable course over the culture period of 11 days and an increasing release towards the end, indicating cell death. In groups A and B a maximum LDH activity in the supernatant was seen following cryopreservation, indicating a cell loss due to freeze-thaw damage, which decreased towards the end of culture. Group C showed a lesser increase of LDH release upon thawing and a decline of values towards zero during subsequent culture, suggesting a severe lack in cell based enzyme amounts (Fig. 11). In addition to the morphology pattern observed, this indicates a more severe damage, cells most likely received during freezing, with a subsequent reduction of quantitative LDH-amounts during the washing steps of thawing.
4. Discussion If we succeeded in making things work, cryopreservation of liver cells had a high potential of advantages that can be summarised as follows: In animal experiments, crypreservation of isolated cell material would lead to a reduction in the number of animals needed, because cells in excess of what is required at the moment could be stored and employed for subsequent investigations.
Fig. 6. Fluorescence microscopy of Group B cells (rate controlled program, 14 days storage), 48 h after thawing, Magn. 300 ×.
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
109
Fig. 7. Fluorescence microscopy of Group C cells (linear freezing program, 3 h storage), 48 h after thawing, Magn. 300 × .
A cell bank of cryopreserved hepatocytes would make it possible to carry out time-independent investigations with parenchymal cells from the same batch. For research groups that do not have access to viable tissues for cell harvesting, thawing of frozen liver cells would mean a reliable supply with necessary biomaterial. Such a cell bank offers the unique opportunity to hold a stock of cryopreserved parenchymal cells from species which are usually difficult to obtain (e.g. human). These batches could be further characterised and subclassified according to individual donor characteristics and employed for in vitro testing of xenobiotics, for instance. A batch-controlled cell bank of cryopreserved hepatocytes could provide cells on demand. This could be of crucial importance to applications, where a time dependency of cell supply exists, such as in the field of bioartificial liver development. Ever since the first publication by Polge et al. [1] on the successful revival of spermatozoa following cryopreservation in 1949, many freezing protocols for different cells and biological tissues have been proposed. Despite the numerous studies, only a few things are being agreed upon: 1. cryopreservation generally results in a loss of cell viability 2. the addition of a cryoprotectant reduces cell damage 3. for every cell-type there is an optimal freezing rate [31–33].
110
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
The aim of our studies was to develop a freezing program and to define conditions under which a batch-controlled cell bank of cryopreserved hepatocytes could be successfully introduced. For our experiments we used porcine hepatocytes from slaughterhouse organs, a method which avoids animal experiments just for the purpose of cell harvesting [27,28]. Most studies have reported on the successful freezing of hepatocyte suspensions [31– 35], techniques which we could never reproduce. We have seen two major disadvantages connected with the cryopreservation of hepatocyte suspensions: 1. Parenchymal cells with membrane alterations from enzymatic digestion during the isolation process are probably more vulnerable to osmotic changes during the freeze-thaw process and present with a substantial loss in cell number and widely reduced hepatocyte functions in the early culture periods. 2. Only a small percentage of cells attach to the matrix after cryopreservation [8,14 – 19,32]. We have refrained from freezing of cell suspensions and recommend to use protective solutions for hypothermic (+ 4°C) short term storage of hepatocyte suspensions instead [22]. For long term storage of liver cells, we have cultured hepatocytes in a sandwich configuration between two layers of collagen prior to freeze storage. The presence of this stable extracellular matrix has led to an
Fig. 8. Fluorescence microscopy of control group cells (no freezing), 5th day of culture, Magn. 300 × .
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
111
Fig. 9. Albumin secretion in Groups A– C and control group before and up to 8 days post freezing. Groups reflect different storage periods (Group A: 3 h, Group B: 14 days) and freezing programs (Group A+ B: Rate Controlled Freezing; Group C: Linear Freezing Program). Values are given as Means 9S.D. (n =12/9). Non-frozen control cultures showed a maximum secretion at day 9 of culture. Groups A and B showed a marked depression immediately post thaw and a recovery of secretional activity to the level of control cultures. Group C showed a depression of Albumin secretion activity towards zero values.
improvement in cell survival in vitro, presumably due to repair of membranes and the cytoskeleton prior to freezing [23,26]. Since parenchymal cells like hepatocytes do not proliferate in vitro, an optimal freeze-thaw protocol is needed to achieve
Fig. 10. CYP 450-dependent formation of 7-Hydroxycoumarin in control and cryopreserved cultures over 11 days of culture, values are given as means 9 S.D. (n = 12/9). Control cultures show a mild decline of activity over the entire period. Cultures of groups A +B react with low values directly after freezing but recover quickly to reach control level at day 7. Group C shows no recovery of CYP 450 activity after linear freezing.
112
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
Fig. 11. LDH-release as marker for cell integrity. Values are given as means 9S.D. (n = 12/9). Control cultures show a stable course and increasing levels towards the end of the investigational period. Groups A – C show a marked release post cryopreservation. Levels in groups A +B indicate a lesser drop of overall enzyme content due to freezing and thawing as compared to group C.
maximum cell survival rates. The cryoprotectant itself, its concentration and the rates of cooling and thawing are crucial for cell survival. Novicki et al. [19] found that Dimethylsulfoxide (DMSO, Me2SO) was superior to glycerol; Chesne´ and Guillouzo [14] showed that DMSO had better cryoprotective capacities, as compared to glycerol or 1,2-propanediol and Loretz et al. [17] also proved DMSO to be better than glycerol, PVP and dextrans. Therefore, we also used Me2SO as cryoprotectant. Powis et al. [8] and Diener et al. [2] recommended 10% as the optimal concentration. Taking these findings into account, we added Me2SO at a concentration of 20% (v/v) in the supernatant medium which led to a final concentrarion of 10% within the matrix. As for the freezing program; most authors recommend slow cooling rates and rapid thawing in an waterbath. An exception here is Gomez-Lechon et al. [35] who found a rapid freezing to be best. We focused on slow cooling rates and used platinum resistors placed within the culture matrix to record the temperature course and to regulate the nitrogen inflow via this route [21]. To preserve a maximum of culture flasks in the freezing chamber a metal rack was used with 12 T-flasks in it (Fig. 2). To assess the need for rate-controlled freezing, we compared our program (Groups A+ B) to a linear freezing protocol with a freezing rate of -10°C/min under otherwise identical conditions (Group C). To assess the viabiltiy of hepatocyte cultures properly we introduced a vital stain with Fluoresceine-Diacetate and Propidium Iodide which revealed a survival rate of 75910% in the rate-adjusted cryopreserved cell cultures (Groups A and B); the linear (= undifferentiated) freezing program led to a recovery rate of about 10%. This result is a significant improvement compared to our first attempt to freeze sandwich cultures [20] and is in accordance with earlier reports of this technique from our group [21,22].
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
113
A high recovery rate is of crucial importance to freezing protocols for cell cultures. In the sandwich culture configuration, dead cells and cell debris cannot be removed from the system. Also, it is advisable to improve the freezing technique to the utmost extent, since functional recovery closely follows the morphological appearance. Albumin secretion in groups that were cryopreserved under rate controlled conditions fully recovered 4 days after thawing and was comparable to non-frozen cultures. It is well known that CYP 450 activity of hepatocytes in vitro decreases with time, even in the more advanced culture configurations. However, the double gel system prolongs the time of detectable phase I cell activity in vitro [21,22,26]. Cytochrome P 450 activity as judged by the formation of Hydroxycoumarin also regained its activity 48 h after thawing to a level comparable to the control group. To judge cell integrity we measured the LDH-release. As expected, cryopreserved hepatocytes exhibited a high release after thawing in early subsequent culture. On day 7 of culture, LDH-release was comparable to non-frozen cultures. Surprisingly the increase in LDH-activity post thaw in the supernatant of Group C was lower than that in Goups A and B. This was due to the fact that in this group LDH was removed by the incubation with decreasing DMSO solutions immediately after thawing (data not shown). From these data it might be concluded that cryopreservation still imposes a hazardous effect on cells, even if hepatocytes are frozen after a recreation period in early cell culture has been granted. In summary, we want to advocate the use of a rate-controlled freezing program that allows satisfactory functional recovery of hepatocytes sandwiched in a double gel configuration before stored. Porcine hepatocytes in this culture technique can be cryopreserved with a survival rate of at least 75% and stored in a batch-controlled cell bank for up to 14 days. This result could potentially be even further improved and extended to other species or special applications. Thus the successful cryopreservation of hepatocyte cultures will mean a benefit for on demand supply of liver cells, either for batch controlled, time-independent investigations in the field of hepatology/ pharmacology or in the development of hybrid bioartificial liver devices.
Acknowledgements We thank M. Kussmaul and U. Cramer for outstanding technical assistance and the butchers of the Munich slaughterhouse for their kind support. We also pay our gratitude to M.M. Heiss, M.D. for the opportunity to measure ethoxycoumarin deethylase on the multiplate scanner fluorimeter. This work was financed in part by the Deutsche Forschungsgemeinschaft (Ko 978/4-1) and the Bundesministerium fu¨r Bildung, Wissenschaft und Forschung (BEO/31: 0311247).
114
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
References [1] C. Polge, A.U. Smith, A.S. Parkes, Revival of spermatozoa after vitrification and dehydration at low temperatures, Nature 164 (1949) 666. [2] B. Diener, D. Utesch, N. Beer, H. Durk, F. Oesch, A method for the cryopreservation of liver parenchymal cells for studies of xenobiotics, Cryobiology 30 (1993) 116 – 127. [3] M. Dou, G. De Sousa, B. Lacarelle, M. Placidi, P. Lechene de la Porte, M. Domingo, H. Lafont, R. Rahmani, Thawed human hepatocytes in primary culture, Cryobiology 29 (1992) 454 – 469. [4] B.J. Fuller, B.W. Grout, R.J. Woods, Biochemical and ultrastructural examination of cryopreserved hepatocytes in rat, Cryobiology 19 (1982) 493 – 502. [5] B.A. Jackson, J.E. Davies, J.K. Chipman, Cytochrome P-450 activity in hepatocytes following cryopreservation and monolayer culture, Biochem. Pharmacol. 34 (1985) 3389 – 3391. [6] A.Y. Petrenko, S.P. Mazur, Inhibition of biotransformation of xenobiotic p-nitroanisole after cryopreservation of isolated rat hepatocytes, Cryobiology 30 (1993) 158 – 163. [7] J. Zaleski, J. Richburg, F.C. Kauffman, Preservation of the rate and profile of xenobiotic metabolism in rat hepatocytes stored in liquid nitrogen, Biochem. Pharmacol. 46 (1993) 111 – 116. [8] G. Powis, K.S. Santone, D.C. Melder, L. Thomas, D.J. Moore, T.J. Wilke, Cryopreservation of rat and dog hepatocytes for studies of xenobiotic metabolism and activation, Drug Metab. Dispos. Biol. Fate Chem. 15 (1987) 826–832. [9] K.S. Santone, D.C. Melder, G. Powis, Studies of chemical toxicity to fresh and cryopreserved rat hepatocytes, Toxicol. Appl. Pharmacol. 97 (1989) 370 – 376. [10] J. Gerlach, K. Kloppel, H.H. Schauwecker, R. Tauber, C. Muller, E.S. Bucherl, Use of hepatocytes in adhesion and suspension cultures for liver support bioreactors, Int. J. Artif. Organs 12 (1989) 788–792. [11] J. Gerlach, T. Trost, C.J. Ryan, M. Meissler, O. Hole, C. Muller, P. Neuhaus, Hybrid liver support system in a short term application on hepatectomized pigs, Int. J. Artif. Organs 17 (1994) 549 – 553. [12] M. Takahashi, H. Ishikura, C. Takahashi, Y. Nakajima, M. Matsushita, H. Matsue, K. Sato, H. Noto, K. Taguchi, et al., Immunologic considerations in the use of cultured porcine hepatocytes as a hybrid artificial liver. Anti-porcine hepatocyte human serum, J. Am. Soc. Art. Inter. Organs 39 (1993) M242–M246. [13] J. Uchino, H. Matsue, M. Takahashi, Y. Nakajima, M. Matsushita, T. Hamada, E. Hashimura, A hybrid artificial liver system. Function of cultured monolayer pig hepatocytes in plasma from hepatic failure patients, Trans. Am. Soc. Art. Inter. Organs 37 (1991) M337 – M338. [14] C. Chesne, A. Guillouzo, Cryopreservation of isolated rat hepatocytes: a critical evaluation of freezing and thawing conditions, Cryobiology 25 (1988) 323 – 330. [15] G.K. Innes, B.J. Fuller, K.E. Hobbs, Functional testing of hepatocytes following their recovery from cryopreservation, Cryobiology 25 (1988) 23 – 30. [16] A. Le Cam, A. Guillouzo, P. Freychet, Ultrastructual and biochemical studies of isolated adult rat hepatocytes prepared under hypoxic conditions. Cryopreservation of hepatocytes, Exper. Cell Res. 98 (1976) 382–395. [17] L.J. Loretz, A.P. Li, M.W. Flye, A.G. Wilson, Optimization of cryopreservation procedures for rat and human hepatocytes, Xenobiotica 19 (1989) 489 – 498. [18] C. Chesne, C. Guyomard, A. Fautrel, M.G. Poullain, B. Fremond, H. De Jong, A. Guillouzo, Viability and function in primary culture of adult hepatocytes from various animal species and human beings after cryopreservation, Hepatology 18 (1993) 406 – 414. [19] D.L. Novicki, G.P. Irons, S.C. Strom, R. Jirtle, G. Michalopoulos, Cryopreservation of isolated rat hepatocytes, In Vitro 18 (1982) 393– 399. [20] H.G. Koebe, J.C. Dunn, M. Toner, L.M. Sterling, A. Hubel, E.G. Cravalho, M.L. Yarmush, R.G. Tompkins, A new approach to the cryopreservation of hepatocytes in a sandwich culture configuration, Cryobiology 27 (1990) 576–584. [21] H.G. Koebe, C. Da¨hnhardt, J. Mu¨ller-Ho¨cker, H. Wagner, F.W. Schildberg, Cryopreservation of porcine hepatocyte cultures, Cryobiology 33 (1995) 127 – 141.
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
115
[22] H.G. Koebe, S.A. Pahernik, W.E. Thasler, F.W. Schildberg, Porcine hepatocytes for biohybrid artificial liver devices—a comparison of hypothermic storage techniques, Artif Organs 20 (1996) 1181–1190. [23] I.H. Borel Rinkes, M. Toner, S.J. Sheeha, R.G. Tompkins, M.L. Yarmush, Long-term functional recovery of hepatocytes after cryopreservation in a three-dimensional culture configuration, Cell Trans. 1 (1992) 281–292. [24] V. Dixit, R. Darvasi, M. Arthur, K. Lewin, G. Gitnick, Cryopreserved microencapsulated hepatocytes – transplantation studies in Gunn rats, Transplantation 55 (1993) 616 – 622. [25] J. Rozga, M.D. Holzman, M.S. Ro, D.W. Griffin, D.F. Neuzil, T. Giorgio, A.D. Moscioni, A.A. Demetriou, Development of a hybrid bioartificial liver, Ann. Surg. 217 (1993) 502 – 509. [26] J.C. Dunn, M.L. Yarmush, H.G. Koebe, R.G. Tompkins, Hepatocyte function and extracellular matrix geometry: Long-term culture in a sandwich configuration, FASEB J. 3 (1989) 174 – 177. [27] H.G. Koebe, S. Pahernik, M. Sproede, W. Thasler, F.W. Schildberg, Porcine hepatocytes from slaughterhouse organs: An unlimited resource for bioartificial liver devices, J. Am. Soc. Artif. Inter. Organs 41 (2) (1995) 189–193. [28] H.G. Koebe, S. Pahernik, W. Thasler, F.W. Schildberg, High yield isolation of porcine hepatocytes from slaughter house organs, Biomater. Artif. Cells Artif. Organs 22 (1994) A24 (abstract). [29] B. Nordlinger, M.E. Bouma, S.R. Wang, F. Ballet, N. Verthier, C. Huguet, R. Infante, High-yield preparation of porcine hepatocytes for long survival after transplantation in the spleen, Eur. Surg. Res. 17 (1985) 377–382. [30] A.M. Edwards, M.L. Glistak, C.M. Lucas, P.A. Wilson, 7-Ethoxycoumarin deethylase activity as a convenient measure of liver drug metabolizing enzymes: regulation in cultured rat hepatocytes, Biochem. Pharmacol. 33 (1984) 1537 – 1546. [31] P. Mazur, Cryobiology: the freezing of biological systems, Science 168 (1970) 939 – 949. [32] H.T. Meryman, Freezing injury and its prevention in living cells, Annu. Rev. Biophys. Bioeng. 3 (1974) 341–363. [33] H.T. Meryman, R.J. Williams, M.S. Douglas, Freezing injury from ‘solution effects’ and its prevention by natural or artificial cryoprotection, Cryobiology 14 (1977) 287 – 302. [34] J.A. Coundouris, M.H. Grant, J. Engeset, J.C. Petrie, G.M. Hawksworth, Cryopreservation of human adult hepatocytes for use in drug metabolism and toxicity studies, Xenobiotica 23 (1993) 1399–1409. [35] M.J. Gomez-Lechon, P. Lopez, J.V. Castell, Biochemical functionality and recovery of hepatocytes after deep freezing storage, In Vitro 20 (1984) 826 – 832. [36] S.L. Nyberg, R.A. Shatford, W.D. Payne, W.S. Hu, F.B. Cerra, Staining with fluorescein diacetate correlates with hepatocyte function, Biotech. Histochem. 68 (1993) 56 – 63.
.
Cryopreserved porcine hepatocyte cultures H.G. Koebe *, B. Mu¨hling, C.J. Deglmann, F.W. Schildberg Department of Surgery, Klinikum Grosshadern, Ludwig-Maximilians-Uni6ersity of Munich, Marchioninistrasse 15, D-81377 Munich, Germany Received 17 February 1999; accepted 9 March 1999
Abstract Cryopreservation of freshly isolated hepatocytes is regarded the standard technique for long term storage of liver cells. Frankly, we were not successful in reproducing viability rates of about 70% of that which have been reported by most authors as results of various freezing protocols for hepatocyte suspensions. In fact, we saw mostly devastating results. We assume that intracellular ice crystal formation as well as osmotic changes during freezing and thawing of liver cells cause hazardous effects, especially on membranes of cells after enzymatic isolation, and, thus, generally result in a severe loss in number and impaired specific hepatocyte functions in subsequent culture. We tried to improve results by freezing cell cultures instead. We allowed hepatocytes to regain a more stable condition prior to storage and placed them in tissue flasks in a uniform configuration, rather than to pack cell suspensions in vials or bags under rather indefinable conditions. Porcine hepatocytes (viability 929 2%) were isolated from slaughterhouse organs and cultured in a double gel (sandwich) configuration. At day 3, cultures were rate controlled frozen (RCF) and stored in a cell bank for three hours (Group A) or 14 days at − 80°C (Group B), respectively. Non-frozen cells (NF) and cultures subjected to a linear freezing rate of −10°C/min (LFR, Group C) with 3 h of storage served as controls from identical cell batches. Upon thawing, at day 2 of subsequent culture, fluorescence microscopy studies revealed a survival rate of 75910% (mean9S.D.) in the RCF groups. Time of storage (3 h, 14 d) did not influence
* Corresponding author. Tel.: +49-897095-6432; fax: + 49-897095-6433. E-mail address: koebe – [email protected] (H.G. Koebe) 0009-2797/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 0 0 9 - 2 7 9 7 ( 9 9 ) 0 0 0 9 3 - 9
100
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
results. Survival in Group C was 10 9 5%. Cell integrity was measured by LDH-release, which indicated a larger damage of cells in the LFR group, and thereby resembled the morphological findings. Functional parameters, such as albumin synthesis and CYT P 450-activity were comparable to non-frozen liver cells at 48 h after thawing in the RCF groups (A+B), and showed significantly higher levels in these groups as compared to the LFR Group (C). We recommend to freeze hepatocytes in culture in a rate controlled fashion rather than cell suspensions. This way a cell bank of cryopreserved hepatocyte cultures for batch controlled investigations can be easily obtained. This could ameliorate the availability of rare (human) cell material and might also improve the quality of data generated from in vitro experiments in hepatology or pharmacology/toxicology with liver cells from identical sources. It remains to be seen whether this technique might also be of value in hybrid bioartificial liver devices to make these systems become readily available upon clinical demand. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Bioartificial liver; cell culture; cryopreservation; hepatocytes; hepatology; pharmacology/toxicology
1. Introduction The discovery of cryoprotectants (substances with cryoprotective properties) such as dimethyl sulfoxide (Me2SO), 1,2-propanediol, glycerol and others made successful cryopreservation of various cell types a reachable goal [1]. For the cryopreservation of hepatocytes in suspension, many protocols have been proposed and freezing was suggested as a standard procedure for long-term storage of liver cells [2–6]. Cryopreserved hepatocytes would be of considerable value for investigations in the field of hepatology and pharmacology/toxicology [7–9], offering the option to perform repeated investigations on well defined cell batches. Also, the development of bioartificial liver devices highly depends on readily available functional units, i.e. liver cells, for on-demand supply [10–13]. Cryopreserved hepatocytes stored in batch-controlled cell banks could provide the biomaterial for all these applications. However, despite the addition of cryoprotectants, liver cells are very susceptible to freeze-thaw damage and since they do not replicate in culture, special cryopreservation techniques are needed to reduce cell injury and functional impairment of hepatocytes. Most studies so far have investigated on freshly isolated and cryopreserved rat hepatocytes in suspension. The results reported so far vary considerably with survival rates between 30 and 80% in the dye exclusion test [14–17]. It is well known, however, that trypan blue exclusion initially after thawing, does not necessarily correlate to desired cell functions in subsequent culture like attachement, synthesis, or metabolising capacity [18,19]. We have already reported on the successful cryopreservation of cultured hepatocytes [20 – 22] and were supported by the findings of others [23]. Liver cells attached to microcarriers [24] or encapsulated in a semipermeable membrane of polylysine and/or alginate [25] have also been cryopreserved in the
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
101
past with rather little information available on the actual influence of cryopreservation on cell performance as compared to non frozen cells. We used the Double Gel (sandwich) culture configuration instead, which provides a stable three dimensional matrix in which hepatocytes maintain their morphological and functional differentiation for long term periods [26]. In our first study on the feasibility of freezing double gel cultures, survival rates upon thawing were 30% and albumin secretion decreased to 20% of controls [20]. The objective of the present study was to improve these results and to establish a batch controlled cell bank of porcine hepatocytes. For this purpose, we modified the freezing program to prevent temperature rises within the culture due to generated latent heat of fusion. Temperature was measured on-line within the sample and rate controlled freezing was carried through accordingly (Groups A + B). In order to underscore the importance of rate controlled freezing, a different program was used (Group C) that led the specimen through a linear decline in temperature which was likely to result in cell damage through intracellular ice formation. For our study we used pig hepatocytes from slaughterhouse organs [27,28]. They represent an unlimited resource for applied experiments [27– 29] in pharmacology and hepatology. Also, pigs have been suggested as donor species for artificial liver devices that are currently being tested in experimental research and clinical trials [10 – 13,22].
2. Materials and methods
2.1. Reagents and solutions Type IV collagenase, Dulbecco‘s modified Eagle‘s medium (Dulbecco’s Modified Eagles Medium) with 4.5g/l glucose, Me2SO and Leibovitz L-15 medium were from Serva Feinbiochemica GmbH (Heidelberg, Germany). The perfusion buffer contained 154 mmol/l sodium chloride, 5.6 mmol/l potassium chloride, 5 mmol/l glucose, 25 mmol/l sodium bicarbonate and 20 mmol/l HEPES (pH 7.4). The washing buffer contained 120 mmol/l sodium chloride, 6.2 mmol/l potassium chloride, 0.9 mmol/l calcium chloride, 10 mmol/l HEPES and 0.2% (w/v) bovine albumin (pH 7.4). Hydroxycoumarin and ethoxycoumarin, DNA and albumin standards were from Sigma (Deisenhofen, Germany). Antibodies were from Nordic Immunology (Tilburg, NL). All other chemicals were of reagent grade and from various commercially available sources.
2.2. Li6er perfusion and cell isolation We obtained the livers of landrace piglets (both sexes, 4–8 weeks old, weighing 15– 35 kg) from the Munich slaughterhouse. The isolation procedure is described in detail elsewhere [22]; briefly: livers were excised by butchers, put into a plastic bag and transported in ice water to the laboratory within a maximum time of 20 min. In the laboratory, the left liver lobe was dissected from the rest of the organ and
102
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
perfused via the left hepatic vein; perfusion buffer was oxygenated by carbogen for the whole procedure. First the lobe was perfused at 37°C at a flow of 50 ml/min with 1000 ml of buffer containing 1 mmol/l EDTA. Next it was perfused with 300 ml of buffer without EDTA and finally a recirculating perfusion with 200 ml of buffer containing 0.05% collagenase (w/v) and 5 mmol/l calcium chloride was performed; the temperature and flow rate were the same. Next the tissue was dissected with a scalpel blade and the homogenates were shaken for 3 min after adding collagenase solution. The resulting cell suspension was filtered through two nylon meshes (210 and 70 mm grid size). The filtered suspension was filled with 4°C cold washing buffer to a volume of 500 ml and washed three times in a centrifuge at 23 g for 5 min. To separate viable hepatocytes from dead cells, non parenchymal cells and debris, a Percoll™ (Pharmacia Biotechnology, Uppsala, Sweden) gradient centrifugation with 10.8 ml of Percoll™ and 1.2 ml of 10 × DMEM and 17 ml of cell suspension was performed at 76 g. The cell pellets of purified hepatocyte suspension were washed twice with washing buffer at 23 g for 5 min and finally the hepatocytes were resuspended in Leibovitz L-15 medium containing 0.2% (w/v) bovine albumin. Cell viability as judged by the trypan blue exclusion test was 92 9 2% with an average cell yield of 1.0 9 10E7 hepatocytes per g perfused liver tissue.
2.3. Hepatocyte cultures The culture technique is described in detail elsewhere [21,26]; briefly: culture medium for the first 24 h consisted of Dulbecco’s Modified Eagle Medium supplemented with insulin (125 mU/ml), hydrocortisone (60 ng/ml), glucagon (10 ng/ml), gentamicin(100 mg/ml), penicillin (100 mU/ml) and 5% (v/v) fetal calf serum. For the sandwich culture configuration T-flasks (25 cm2) were precoated with 1 ml of collagen solution by mixing 1 part of 10× DMEM (pH 7.4.) and 9 parts of collagen (0.83 mg/ml) to a final collagen concentration of 0.75 mg/ml. Type I collagen was prepared from rat tail tendons. After gelling of the matrix in a humidified CO2 controlled incubator [37°C, 5% (v/v) CO2] 2 ml of cell suspension (1.5 × 10E6 cells/ml) was distributed on the culture flasks. To allow attachment of hepatocytes, cultures were kept in the incubator for the first 24 h; next, cell medium was replaced by the second layer of collagen. Two ml of culture medium (Serum free) were added and changed every 48 h. Probes for analysis were collected at the given intervals and stored at −20°C before measurement.
2.4. Cryopreser6ation On day three of culture, cells were incubated with DMSO as a cryoprotectant at a concentration of 20% (v/v) prior to freezing to achieve a final concentration of 10% (v/v) within the culture matrix. Cells were stored in the incubator for 15 min at 37°C. Following removal of the supernatant, the T-flasks were put in a metal rack (Fig. 1) which was stored at a defined position in the freezing chamber of a programmable freezing unit (Nicool ST 20, L’Air Liquide, Sassenage, France; Fig.
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
103
2). Temperature levels were recorded at two defined sites within the chamber (Philips recorder PM 8251, Hewlett-Packard standard multimeter) to ensure controlled cooling of the specimen. Measurement of the actual temperature within those hepatocyte cultures that served as samples was performed by using a platinum resistor PT 100 (MRIV, Munich, Germany) which was dipped in the culture matrix (Fig. 3). This technique allows a close follow up of the measured temperature curves within the samples and supports an accurate steering of the Nitrogen inlet. To avoid random differences, three independent experiments were performed with n= 4 cultures for each group. This way, three consecutive cell preparations led to an overall of n= 12 cultures per group. n= 3 non-frozen cultures per isolation formed a group of n= 9. The recorded temperature curves (Groups A+B and Group C, respectively) were almost identical in all three experimental set ups (data not shown). Three experimental groups were included: Group A (n=12): Cells were exposed to the rate controlled freezing program (RCF, Table 1; Fig. 4) and stored for 3 h at − 80°C. Group B (n=12) Same program as under Group A, but stored for 14 days at − 80°C. Group C (n= 12) Cells underwent a linear freezing program (LFP, Table 2) at – 10°C/min and were stored for 3 h at − 80°C. Control (n=9) No freezing.
Fig. 1. Hepatocyte culture T-flasks in a metal rack for up to 12 cultures with 3 ×106 cells each.
104
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
Fig. 2. Programmable freezing unit Nicool ST 20.
2.5. Thawing and remo6al of the cryoprotectant After storage, hepatocyte cultures were thawed by adding 5 ml of a 37°C warm 10% (v/v) solution of DMSO and placing the T-flasks in a waterbath at 40°C for 10 sec. Next, the thawing solution was removed and the cultures were incubated with decreasing Me2SO solutions (7.5, 5, 2.5, and 0%); each solution was added for 5 min at + 4°C to remove the cryoprotectant. Finally, hepatocyte cultures were transferred into the incubator and cultured in cell medium at 37°C.
2.6. Viability assay In order to compare the number of viable cells in cryopreserved cultures with unfrozen controls, the determination of cell viability was set at day five of culture (48 h after thawing) in all groups by staining with fluorescein diacetate and propidium iodine [36]. Briefly, fluorescein diacetate (Sigma Chemie, Deisenhofen, Germany) was dissolved in acetone (1 mg/ml) and frozen at − 20°C before use. Medium was aspirated and replaced by fresh medium (1 ml)+ 5 ml of fluorescein diacetate and propidium iodinesolution. Cultures were held for 5 min at 37°C before being washed in PBS for 20 min at 4°C. Next, fluoroscopic images were captured using a Nikon Diaphot 200 inverted microscope equipped for fluorescence (B2A filter, Duesseldorf, Germany). Two microscopic fields per culture dish were randomly selected and viable cells stained with fluorescein diacetate (green) as well
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
105
Fig. 3. Tissue flask with implanted PT-100 probe for in matrix temperature registration (Demo-Matrix without cells in culture).
as iodine stained nuclei of cells (orange) with destroyed membranes were counted. The number of viable cells of posthypothermic cultures was calculated versus control culture dishes and expressed as percent 9 S.D.
2.7. Biochemical assays Probes for biochemical assessments were collected every 48 h from the first day of culture to the 11th day. Albumin was determined immunologically by using an antibody to porcine albumin (rat-anti-pig) and a peroxidase-conjugated antibody (goat anti-rat IgG peroxidase). Extinction was measured at 450 nm with a Titertek Multiscan Plus Reader (MK II, Flow Laboratories, Meckenheim, Germany). Cytochrome P 450 activity (formation of Hydroxycoumarin) was measured fluorometrically by a modified method according to Edwards [30]. LDH-release was determined in the supernatant photometrically at 339 nm; pyruvate and NADH/ H+ served as substrates. DNA content of hepatocytes was analysed on a TKO 100 Table 1 Rate controlled freezing program (RCF, Groups A+B) Temperature (°C)
Freezing rate
4 to 0 0 to −11 −11 −11 to −80
−1°C/min 0.5°C 15 min −3°C/min
106
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
Fig. 4. Freezing protocol under the rate controlled program. Prior to the final freezing step, cultures were held at –11°C to complete extracellular ice formation. Nitrogen inflow was regulated to compensate for the generation of latent heat of fusion at this point to prevent cultures from osmotic stress due to phase changes.
fluorometer (Hoefer Scientific Instruments, San Francisco, CA; lex = 365 nm; lem =460 nm). Samples were measured three times, each, to avoid random differences. Values are given as mean 9S.D.
3. Results
3.1. Cell morphology Figs. 5 and 6 show the fluorescence microscopic appearance of hepatocytes in groups A and B after 3 h and 14 days storage, respectively. Upon thawing, after 48 h of subsequent culture, the number of hepatocytes was only slightly reduced and viable cells (75 9 10%) showed no remarkable changes in morphology. In contrast, cultures of group C (linear freezing) mainly appeared with stained nuclei of dead cells; viability in this group was 109 5% (Fig. 7). Fig. 8 shows controls, 5th day of culture. Table 2 Linear freezing program (LFP, Group C) Temperature (°C)
Freezing rate
4 to −80
−10°C/min
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
107
Fig. 5. Fluorescence microscopy of Group A cells (rate controlled program, 3h storage), 48 h after thawing, Magn. 300 × .
3.2. Albumin secretion Albumin secretion in groups A, B, C, and control from day one up to day 11 of culture is shown in Fig. 9. Control cultures (non-frozen) showed increasing values from day 5 of culture with a maximum albumin secretion at day 9 (105.3 ng/h/mg DNA). Albumin secretion in groups A and B declined markedly after the freezethaw-cycle, but reached a maximum level on day 9 of culture (A: 127 ng/h/mg DNA, B: 78.4 ng/h/mg DNA). Cultures of Group C almost ceased albumin secretion after thawing and values approached zero.
3.3. Ethoxycoumarin-O-Deethylase-acti6ity Metabolic activity (CYP 450) in unfrozen hepatocytes declined gently from day 1 through 11. Cultures, preserved with the rate controlled program, showed a marked loss of CYP 450 activity upon thawing (Groups A+B), but cells regained their function up to the level of the control groups at day 4 post cryopreservation. In contrast, Group C showed a steady decline towards zero values (Fig. 10).
108
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
3.4. LDH-release LDH-release in the control group displayed a stable course over the culture period of 11 days and an increasing release towards the end, indicating cell death. In groups A and B a maximum LDH activity in the supernatant was seen following cryopreservation, indicating a cell loss due to freeze-thaw damage, which decreased towards the end of culture. Group C showed a lesser increase of LDH release upon thawing and a decline of values towards zero during subsequent culture, suggesting a severe lack in cell based enzyme amounts (Fig. 11). In addition to the morphology pattern observed, this indicates a more severe damage, cells most likely received during freezing, with a subsequent reduction of quantitative LDH-amounts during the washing steps of thawing.
4. Discussion If we succeeded in making things work, cryopreservation of liver cells had a high potential of advantages that can be summarised as follows: In animal experiments, crypreservation of isolated cell material would lead to a reduction in the number of animals needed, because cells in excess of what is required at the moment could be stored and employed for subsequent investigations.
Fig. 6. Fluorescence microscopy of Group B cells (rate controlled program, 14 days storage), 48 h after thawing, Magn. 300 ×.
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
109
Fig. 7. Fluorescence microscopy of Group C cells (linear freezing program, 3 h storage), 48 h after thawing, Magn. 300 × .
A cell bank of cryopreserved hepatocytes would make it possible to carry out time-independent investigations with parenchymal cells from the same batch. For research groups that do not have access to viable tissues for cell harvesting, thawing of frozen liver cells would mean a reliable supply with necessary biomaterial. Such a cell bank offers the unique opportunity to hold a stock of cryopreserved parenchymal cells from species which are usually difficult to obtain (e.g. human). These batches could be further characterised and subclassified according to individual donor characteristics and employed for in vitro testing of xenobiotics, for instance. A batch-controlled cell bank of cryopreserved hepatocytes could provide cells on demand. This could be of crucial importance to applications, where a time dependency of cell supply exists, such as in the field of bioartificial liver development. Ever since the first publication by Polge et al. [1] on the successful revival of spermatozoa following cryopreservation in 1949, many freezing protocols for different cells and biological tissues have been proposed. Despite the numerous studies, only a few things are being agreed upon: 1. cryopreservation generally results in a loss of cell viability 2. the addition of a cryoprotectant reduces cell damage 3. for every cell-type there is an optimal freezing rate [31–33].
110
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
The aim of our studies was to develop a freezing program and to define conditions under which a batch-controlled cell bank of cryopreserved hepatocytes could be successfully introduced. For our experiments we used porcine hepatocytes from slaughterhouse organs, a method which avoids animal experiments just for the purpose of cell harvesting [27,28]. Most studies have reported on the successful freezing of hepatocyte suspensions [31– 35], techniques which we could never reproduce. We have seen two major disadvantages connected with the cryopreservation of hepatocyte suspensions: 1. Parenchymal cells with membrane alterations from enzymatic digestion during the isolation process are probably more vulnerable to osmotic changes during the freeze-thaw process and present with a substantial loss in cell number and widely reduced hepatocyte functions in the early culture periods. 2. Only a small percentage of cells attach to the matrix after cryopreservation [8,14 – 19,32]. We have refrained from freezing of cell suspensions and recommend to use protective solutions for hypothermic (+ 4°C) short term storage of hepatocyte suspensions instead [22]. For long term storage of liver cells, we have cultured hepatocytes in a sandwich configuration between two layers of collagen prior to freeze storage. The presence of this stable extracellular matrix has led to an
Fig. 8. Fluorescence microscopy of control group cells (no freezing), 5th day of culture, Magn. 300 × .
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
111
Fig. 9. Albumin secretion in Groups A– C and control group before and up to 8 days post freezing. Groups reflect different storage periods (Group A: 3 h, Group B: 14 days) and freezing programs (Group A+ B: Rate Controlled Freezing; Group C: Linear Freezing Program). Values are given as Means 9S.D. (n =12/9). Non-frozen control cultures showed a maximum secretion at day 9 of culture. Groups A and B showed a marked depression immediately post thaw and a recovery of secretional activity to the level of control cultures. Group C showed a depression of Albumin secretion activity towards zero values.
improvement in cell survival in vitro, presumably due to repair of membranes and the cytoskeleton prior to freezing [23,26]. Since parenchymal cells like hepatocytes do not proliferate in vitro, an optimal freeze-thaw protocol is needed to achieve
Fig. 10. CYP 450-dependent formation of 7-Hydroxycoumarin in control and cryopreserved cultures over 11 days of culture, values are given as means 9 S.D. (n = 12/9). Control cultures show a mild decline of activity over the entire period. Cultures of groups A +B react with low values directly after freezing but recover quickly to reach control level at day 7. Group C shows no recovery of CYP 450 activity after linear freezing.
112
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
Fig. 11. LDH-release as marker for cell integrity. Values are given as means 9S.D. (n = 12/9). Control cultures show a stable course and increasing levels towards the end of the investigational period. Groups A – C show a marked release post cryopreservation. Levels in groups A +B indicate a lesser drop of overall enzyme content due to freezing and thawing as compared to group C.
maximum cell survival rates. The cryoprotectant itself, its concentration and the rates of cooling and thawing are crucial for cell survival. Novicki et al. [19] found that Dimethylsulfoxide (DMSO, Me2SO) was superior to glycerol; Chesne´ and Guillouzo [14] showed that DMSO had better cryoprotective capacities, as compared to glycerol or 1,2-propanediol and Loretz et al. [17] also proved DMSO to be better than glycerol, PVP and dextrans. Therefore, we also used Me2SO as cryoprotectant. Powis et al. [8] and Diener et al. [2] recommended 10% as the optimal concentration. Taking these findings into account, we added Me2SO at a concentration of 20% (v/v) in the supernatant medium which led to a final concentrarion of 10% within the matrix. As for the freezing program; most authors recommend slow cooling rates and rapid thawing in an waterbath. An exception here is Gomez-Lechon et al. [35] who found a rapid freezing to be best. We focused on slow cooling rates and used platinum resistors placed within the culture matrix to record the temperature course and to regulate the nitrogen inflow via this route [21]. To preserve a maximum of culture flasks in the freezing chamber a metal rack was used with 12 T-flasks in it (Fig. 2). To assess the need for rate-controlled freezing, we compared our program (Groups A+ B) to a linear freezing protocol with a freezing rate of -10°C/min under otherwise identical conditions (Group C). To assess the viabiltiy of hepatocyte cultures properly we introduced a vital stain with Fluoresceine-Diacetate and Propidium Iodide which revealed a survival rate of 75910% in the rate-adjusted cryopreserved cell cultures (Groups A and B); the linear (= undifferentiated) freezing program led to a recovery rate of about 10%. This result is a significant improvement compared to our first attempt to freeze sandwich cultures [20] and is in accordance with earlier reports of this technique from our group [21,22].
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
113
A high recovery rate is of crucial importance to freezing protocols for cell cultures. In the sandwich culture configuration, dead cells and cell debris cannot be removed from the system. Also, it is advisable to improve the freezing technique to the utmost extent, since functional recovery closely follows the morphological appearance. Albumin secretion in groups that were cryopreserved under rate controlled conditions fully recovered 4 days after thawing and was comparable to non-frozen cultures. It is well known that CYP 450 activity of hepatocytes in vitro decreases with time, even in the more advanced culture configurations. However, the double gel system prolongs the time of detectable phase I cell activity in vitro [21,22,26]. Cytochrome P 450 activity as judged by the formation of Hydroxycoumarin also regained its activity 48 h after thawing to a level comparable to the control group. To judge cell integrity we measured the LDH-release. As expected, cryopreserved hepatocytes exhibited a high release after thawing in early subsequent culture. On day 7 of culture, LDH-release was comparable to non-frozen cultures. Surprisingly the increase in LDH-activity post thaw in the supernatant of Group C was lower than that in Goups A and B. This was due to the fact that in this group LDH was removed by the incubation with decreasing DMSO solutions immediately after thawing (data not shown). From these data it might be concluded that cryopreservation still imposes a hazardous effect on cells, even if hepatocytes are frozen after a recreation period in early cell culture has been granted. In summary, we want to advocate the use of a rate-controlled freezing program that allows satisfactory functional recovery of hepatocytes sandwiched in a double gel configuration before stored. Porcine hepatocytes in this culture technique can be cryopreserved with a survival rate of at least 75% and stored in a batch-controlled cell bank for up to 14 days. This result could potentially be even further improved and extended to other species or special applications. Thus the successful cryopreservation of hepatocyte cultures will mean a benefit for on demand supply of liver cells, either for batch controlled, time-independent investigations in the field of hepatology/ pharmacology or in the development of hybrid bioartificial liver devices.
Acknowledgements We thank M. Kussmaul and U. Cramer for outstanding technical assistance and the butchers of the Munich slaughterhouse for their kind support. We also pay our gratitude to M.M. Heiss, M.D. for the opportunity to measure ethoxycoumarin deethylase on the multiplate scanner fluorimeter. This work was financed in part by the Deutsche Forschungsgemeinschaft (Ko 978/4-1) and the Bundesministerium fu¨r Bildung, Wissenschaft und Forschung (BEO/31: 0311247).
114
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
References [1] C. Polge, A.U. Smith, A.S. Parkes, Revival of spermatozoa after vitrification and dehydration at low temperatures, Nature 164 (1949) 666. [2] B. Diener, D. Utesch, N. Beer, H. Durk, F. Oesch, A method for the cryopreservation of liver parenchymal cells for studies of xenobiotics, Cryobiology 30 (1993) 116 – 127. [3] M. Dou, G. De Sousa, B. Lacarelle, M. Placidi, P. Lechene de la Porte, M. Domingo, H. Lafont, R. Rahmani, Thawed human hepatocytes in primary culture, Cryobiology 29 (1992) 454 – 469. [4] B.J. Fuller, B.W. Grout, R.J. Woods, Biochemical and ultrastructural examination of cryopreserved hepatocytes in rat, Cryobiology 19 (1982) 493 – 502. [5] B.A. Jackson, J.E. Davies, J.K. Chipman, Cytochrome P-450 activity in hepatocytes following cryopreservation and monolayer culture, Biochem. Pharmacol. 34 (1985) 3389 – 3391. [6] A.Y. Petrenko, S.P. Mazur, Inhibition of biotransformation of xenobiotic p-nitroanisole after cryopreservation of isolated rat hepatocytes, Cryobiology 30 (1993) 158 – 163. [7] J. Zaleski, J. Richburg, F.C. Kauffman, Preservation of the rate and profile of xenobiotic metabolism in rat hepatocytes stored in liquid nitrogen, Biochem. Pharmacol. 46 (1993) 111 – 116. [8] G. Powis, K.S. Santone, D.C. Melder, L. Thomas, D.J. Moore, T.J. Wilke, Cryopreservation of rat and dog hepatocytes for studies of xenobiotic metabolism and activation, Drug Metab. Dispos. Biol. Fate Chem. 15 (1987) 826–832. [9] K.S. Santone, D.C. Melder, G. Powis, Studies of chemical toxicity to fresh and cryopreserved rat hepatocytes, Toxicol. Appl. Pharmacol. 97 (1989) 370 – 376. [10] J. Gerlach, K. Kloppel, H.H. Schauwecker, R. Tauber, C. Muller, E.S. Bucherl, Use of hepatocytes in adhesion and suspension cultures for liver support bioreactors, Int. J. Artif. Organs 12 (1989) 788–792. [11] J. Gerlach, T. Trost, C.J. Ryan, M. Meissler, O. Hole, C. Muller, P. Neuhaus, Hybrid liver support system in a short term application on hepatectomized pigs, Int. J. Artif. Organs 17 (1994) 549 – 553. [12] M. Takahashi, H. Ishikura, C. Takahashi, Y. Nakajima, M. Matsushita, H. Matsue, K. Sato, H. Noto, K. Taguchi, et al., Immunologic considerations in the use of cultured porcine hepatocytes as a hybrid artificial liver. Anti-porcine hepatocyte human serum, J. Am. Soc. Art. Inter. Organs 39 (1993) M242–M246. [13] J. Uchino, H. Matsue, M. Takahashi, Y. Nakajima, M. Matsushita, T. Hamada, E. Hashimura, A hybrid artificial liver system. Function of cultured monolayer pig hepatocytes in plasma from hepatic failure patients, Trans. Am. Soc. Art. Inter. Organs 37 (1991) M337 – M338. [14] C. Chesne, A. Guillouzo, Cryopreservation of isolated rat hepatocytes: a critical evaluation of freezing and thawing conditions, Cryobiology 25 (1988) 323 – 330. [15] G.K. Innes, B.J. Fuller, K.E. Hobbs, Functional testing of hepatocytes following their recovery from cryopreservation, Cryobiology 25 (1988) 23 – 30. [16] A. Le Cam, A. Guillouzo, P. Freychet, Ultrastructual and biochemical studies of isolated adult rat hepatocytes prepared under hypoxic conditions. Cryopreservation of hepatocytes, Exper. Cell Res. 98 (1976) 382–395. [17] L.J. Loretz, A.P. Li, M.W. Flye, A.G. Wilson, Optimization of cryopreservation procedures for rat and human hepatocytes, Xenobiotica 19 (1989) 489 – 498. [18] C. Chesne, C. Guyomard, A. Fautrel, M.G. Poullain, B. Fremond, H. De Jong, A. Guillouzo, Viability and function in primary culture of adult hepatocytes from various animal species and human beings after cryopreservation, Hepatology 18 (1993) 406 – 414. [19] D.L. Novicki, G.P. Irons, S.C. Strom, R. Jirtle, G. Michalopoulos, Cryopreservation of isolated rat hepatocytes, In Vitro 18 (1982) 393– 399. [20] H.G. Koebe, J.C. Dunn, M. Toner, L.M. Sterling, A. Hubel, E.G. Cravalho, M.L. Yarmush, R.G. Tompkins, A new approach to the cryopreservation of hepatocytes in a sandwich culture configuration, Cryobiology 27 (1990) 576–584. [21] H.G. Koebe, C. Da¨hnhardt, J. Mu¨ller-Ho¨cker, H. Wagner, F.W. Schildberg, Cryopreservation of porcine hepatocyte cultures, Cryobiology 33 (1995) 127 – 141.
H.G. Koebe et al. / Chemico-Biological Interactions 121 (1999) 99–115
115
[22] H.G. Koebe, S.A. Pahernik, W.E. Thasler, F.W. Schildberg, Porcine hepatocytes for biohybrid artificial liver devices—a comparison of hypothermic storage techniques, Artif Organs 20 (1996) 1181–1190. [23] I.H. Borel Rinkes, M. Toner, S.J. Sheeha, R.G. Tompkins, M.L. Yarmush, Long-term functional recovery of hepatocytes after cryopreservation in a three-dimensional culture configuration, Cell Trans. 1 (1992) 281–292. [24] V. Dixit, R. Darvasi, M. Arthur, K. Lewin, G. Gitnick, Cryopreserved microencapsulated hepatocytes – transplantation studies in Gunn rats, Transplantation 55 (1993) 616 – 622. [25] J. Rozga, M.D. Holzman, M.S. Ro, D.W. Griffin, D.F. Neuzil, T. Giorgio, A.D. Moscioni, A.A. Demetriou, Development of a hybrid bioartificial liver, Ann. Surg. 217 (1993) 502 – 509. [26] J.C. Dunn, M.L. Yarmush, H.G. Koebe, R.G. Tompkins, Hepatocyte function and extracellular matrix geometry: Long-term culture in a sandwich configuration, FASEB J. 3 (1989) 174 – 177. [27] H.G. Koebe, S. Pahernik, M. Sproede, W. Thasler, F.W. Schildberg, Porcine hepatocytes from slaughterhouse organs: An unlimited resource for bioartificial liver devices, J. Am. Soc. Artif. Inter. Organs 41 (2) (1995) 189–193. [28] H.G. Koebe, S. Pahernik, W. Thasler, F.W. Schildberg, High yield isolation of porcine hepatocytes from slaughter house organs, Biomater. Artif. Cells Artif. Organs 22 (1994) A24 (abstract). [29] B. Nordlinger, M.E. Bouma, S.R. Wang, F. Ballet, N. Verthier, C. Huguet, R. Infante, High-yield preparation of porcine hepatocytes for long survival after transplantation in the spleen, Eur. Surg. Res. 17 (1985) 377–382. [30] A.M. Edwards, M.L. Glistak, C.M. Lucas, P.A. Wilson, 7-Ethoxycoumarin deethylase activity as a convenient measure of liver drug metabolizing enzymes: regulation in cultured rat hepatocytes, Biochem. Pharmacol. 33 (1984) 1537 – 1546. [31] P. Mazur, Cryobiology: the freezing of biological systems, Science 168 (1970) 939 – 949. [32] H.T. Meryman, Freezing injury and its prevention in living cells, Annu. Rev. Biophys. Bioeng. 3 (1974) 341–363. [33] H.T. Meryman, R.J. Williams, M.S. Douglas, Freezing injury from ‘solution effects’ and its prevention by natural or artificial cryoprotection, Cryobiology 14 (1977) 287 – 302. [34] J.A. Coundouris, M.H. Grant, J. Engeset, J.C. Petrie, G.M. Hawksworth, Cryopreservation of human adult hepatocytes for use in drug metabolism and toxicity studies, Xenobiotica 23 (1993) 1399–1409. [35] M.J. Gomez-Lechon, P. Lopez, J.V. Castell, Biochemical functionality and recovery of hepatocytes after deep freezing storage, In Vitro 20 (1984) 826 – 832. [36] S.L. Nyberg, R.A. Shatford, W.D. Payne, W.S. Hu, F.B. Cerra, Staining with fluorescein diacetate correlates with hepatocyte function, Biotech. Histochem. 68 (1993) 56 – 63.
.