A multi-laboratory evaluation of cryopreserved monkey hepatocyte functions for use in pharmaco–toxicology

Chemico-Biological Interactions 121 (1999) 77–97

A multi-laboratory evaluation of cryopreservedmonkey hepatocyte functions for use in

pharmaco–toxicology�

Georges de Sousa a, Florence Nicolas a,1, Michel Placidi a,Roger Rahmani a,*2,

Marc Benicourt b, Bernard Vannier b, Giocondo Lorenzon b,Karine Mertens c, Sandra Coecke c,Andre Callaerts c, Vera Rogiers c,

Shamas Khan d, Phil Roberts d, Paul Skett d,Alain Fautrel e, Christophe Chesne e, Andre Guillouzo e

a INSERM/Centre de Recherche Agronomique, 41, Boule6ard du Cap, 06606 Antibes, Franceb Centre de Recherches Roussel UCLAF, 102 route de Noisy, 93230 Romain6ille, France

c Department of Toxicology, Vrije Uni6ersiteit Brussel, Laarbeeklaan 103, B-1090 Brussels, Belgiumd West Medical Building, Uni6ersity of Glasgow, Glasgow G12 8QQ, Scotland, UK

e INSERM U49 and Biopredic, 35000 Rennes, France

Abstract

Ethical, economic and technical reasons hinder regular supply of freshly isolated hepato-cytes from higher mammals such as monkey for preclinical evaluation of drugs. Hence, weaimed at developing optimal and reproducible protocols to cryopreserve and thaw parenchy-mal liver cells from this major toxicological species. Before the routine use of these protocols,

Abbre6iations: AE, aldrin epoxydase; ATP, adenosine triphosphate; AZT, 3%-azido-3%-deoxythymidine;CYP, cytochrome P450; DMSO, dimethylsulfoxide; DZP, diazepam; ECOD, 7-ethoxycoumarin O-deethylase; EH, epoxyde hydrolase; FCS, foetal calf serum; GPx, glutathione peroxydase; GR, glu-tathione reductase; GSH, glutathione; GST, glutathione S-transferase; LDH, lactate dehydrogenase;MDZ, midazolam; PVP, polyvinylpyrrolidone.� Supported by CEC DG XI-INSERM contracts (B91/B4-3063/11/1395 and B92/B4-3081/013665)

and the Home Office Animal Procedure Committee.* Corresponding author. Tel.: +33-93-678860; fax: +33-93-673040.E-mail address: [email protected] (R. Rahmani)1 Supported by a grant from Roussel UCLAF-Romainville-France.2 Co-ordinating author.

0009-2797/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved.

PII: S0009 -2797 (99 )00092 -7

G. de Sousa et al. / Chemico-Biological Interactions 121 (1999) 77–9778

we validated them through a multi-laboratory study. Dissociation of the whole animal liverresulted in obtaining 1–5 billion parenchymal cells with a viability of about 86%. Anappropriate fraction (around 20%) of the freshly isolated cells was immediately set inprimary culture and various hepato-specific tests were performed to examine their metabolic,biochemical and toxicological functions as well as their ultrastructural characteristics. Themajor part of the hepatocytes was frozen and their functionality checked using the sameparameters after thawing. The characterization of fresh and thawed monkey hepatocytesdemonstrated the maintenance of various hepato-specific functions. Indeed, cryopreservedhepatocytes were able to survive and to function in culture as well as their fresh counterparts.The ability for synthesis (proteins, ATP, GSH) and conjugation and secretion of biliary acidswas preserved after deep freeze storage. A better stability of drug metabolizing activities thanin rodent hepatocytes was observed in monkey. After thawing, Phase I and Phase II activities(cytochrome P450, ethoxycoumarin-O-deethylase, aldrin epoxidase, epoxide hydrolase, glu-tathione transferase, glutathione reductase and glutathione peroxidase) were well preserved.The metabolic patterns of several drugs were qualitatively and quantitatively similar beforeand after cryopreservation. Lastly, cytotoxicity tests suggested that the freezing/thawing stepsdid not change cell sensitivity to toxic compounds. © 1999 Elsevier Science Ireland Ltd. Allrights reserved.

Keywords: Cryopreservation; Hepatocytes; Monkey; Metabolism; Preclinical drug evaluation

1. Introduction

There is a growing body of evidence that the pharmacological as well as thetoxicological properties of most drugs after their administration in the organism arerelated to their metabolism. Therefore, it has now become essential to determine thebiotransformation pathways of any new potential therapeutic agent, as soon aspossible during its preclinical research and development phases. Such investigationsare usually performed in vivo in animals, studies which suffer from seriousdrawbacks related to the wide interspecies variability in response to xenobioticadministration (metabolism, induction, cytotoxicity). Besides making difficult theextrapolation of animal data to man, these in vivo experimentations also raisedmany ethical, economical and scientific problems.

As the liver is the major active mammalian organ with respect to xenobioticdetoxification or activation processes, hepatocytes from various species have beenincreasingly and routinely used for pharmaco-toxicological and more particularlyfor metabolic studies [1–4]. Indeed, in vitro/in vivo comparative studies on manydrugs clearly indicated that both the qualitative and quantitative interspeciesvariability found in vivo in their hepatic metabolic rates and routes, can bereproduced using this in vitro model. However, although some publications dealtwith dog, monkey or human hepatocytes [5–8], most of the data available wereconcerned with rodent because ethical, technical and economic reasons hinder theregular availability of livers from higher mammals. Nevertheless hepatocytes fromthese species can be successfully isolated in high yields by collagenase perfusion of

G. de Sousa et al. / Chemico-Biological Interactions 121 (1999) 77–97 79

a total liver or a biopsy. The large number of cells (several billion) could not beused in a fresh form after isolation, even for the largest experiment. As the majorityof them are routinely discarded, it appears necessary to develop reproducible andeffective protocols to store the excess hepatocytes. Indeed, besides the importantethical consideration, it could be very convenient to have access, on a regular basis,to viable and functional hepatocytes from higher mammals, monkey for instance.This would allow a more rational design of experiments, in particular for studyinginterindividual or interspecies variability.

In spite of the extensive improvements made in the culture conditions formaintaining hepatocytes in primary culture for several days, no in vitro techniqueallows to date their long-term survival for several months. Cryopreservationappears to be the only conceivable method to achieve such a long term storage andvarious freezing protocols have been described for adult liver parenchymal cells.These methods were usually devised for rat hepatocytes [9,10] and only a fewpublications dealt with monkey [11], dog [12] or human [13–17]. These reportsgenerally demonstrated that the hepatocytes could be cryopreserved and culturedunder good conditions, but showed variable sensitivity to the freeze/thaw processaccording to the species [17,18]. Moreover, only few hepato-specific parameterswere simultaneously checked on fresh and thawed cells in order to establish thereliability of the cryopreservation protocols.

In that context, the present study aimed at the development of the optimalexperimental conditions for cryopreserving, thawing and culturing hepatocytesisolated from whole monkey liver. Before the routine use of these cells, wetentatively characterized the effects of deep freezing storage on major hepatocytefunctions. This was performed first by checking the cell viability and the attachmentefficacy after cryopreservation, parameters which mainly reflect the plasma mem-brane damage but are not informative enough about the maintenance of cellfunctions. Therefore, a large set of morphological, biochemical, and metabolicparameters were concomitantly tested on hepatocyte suspensions immediately afterthe isolation and thawing procedures, but also on fresh and thawed culturedhepatocytes. Indeed, although primary culture represents a model closer to the invivo situation than cells in suspension, only few studies have evaluated the effectsof cryopreservation on specific functions of human [13–17], dog [12,15] andmonkey [11] hepatocytes maintained in primary culture for several days.

Only data obtained on monkey, one of the major species used in in vivotoxicological studies, are presented in this study. Our ultimate goal is however, toset-up and to characterize by a similar approach, a bank of frozen dog and humanhepatocytes available, when necessary, for pharmaco–toxicological investigations.Once validated, this cryopreserved hepatocyte model would constitute a highlyuseful tool at a preclinical stage of drug development. Indeed, it may allow, undersimilar experimental conditions (in terms of cell density, drug concentration, culturemedia) to study the overall qualitative and quantitative interspecies variability ofdrug biotransformation and transport processes at the hepatocyte level. Moreover,such cells would be of major interest in identifying and/or predicting the origins ofdrug interactions likely to occur by induction or inhibition of hepatic metabolizing

G. de Sousa et al. / Chemico-Biological Interactions 121 (1999) 77–9780

enzymes. Finally using this in vitro tool, biotransformation polymorphism of anynew drug could be rapidly detected in a given species, human in particular, duringits preclinical evaluation. Not only a significant reduction of in vivo experimenta-tions in animals is expected from this strategy, but also a more rational selection ofthe optimal drug candidate for further development.

2. Materials and methods

2.1. Hepatocyte isolation

Four male Cynomolgus monkeys (3.7–4.7 kg) were supplied from Les ElevagesLebeau (France). They were handled in compliance with French regulations onanimal experimentation. Twenty-four hours before the operation, the animalsreceived a light diet with water ad libitum.

After anaesthesia, a large median laparotomy was made. The liver was firstextensively washed in situ via the portal and aortic veins by 1–2 l of Eurocollinssolution, at room temperature, containing 5000 UI/l heparin and saturated with O2.Just before cutting out the liver was flushed with Eurocollins solution at 4°C inorder to cool the organ before its transport in an isotherm package to thelaboratory.

Hepatocytes were isolated from the whole liver by a two-step collagenaseperfusion technique [19]. Briefly, the liver was first washed with Mg2+ and Ca2+

free Hanks balanced salt solution (HBSS) at 37°C. The organ was then perfusedwith a collagenase (Worthington) solution (0.05%, w/v) containing 10 mM CaCl2,until the appearance of the liver indicated tissue dissociation. The Glisson’s capsulewas then disrupted. After purification, freshly isolated cells were resuspended inMedium I consisting of: Williams E medium containing 10% fetal calf serum (FCS)and supplemented with penicillin (50 U/ml), streptomycin (50 mg/ml), netilmicine(50 mg/ml) and insulin (0.1 U/ml). Hepatocyte viability was determined using theErythrosin B exclusion test.

2.2. Cryopreser6ation and thawing

Monkey hepatocytes were cryopreserved and thawed according to the protocolspreviously described [14,16]. Briefly, the hepatocytes were resuspended in LeibovitzL15 medium containing 2% polyvinylpyrrolydone (PVP), 2.5% bovine serum albu-min (BSA), 20% FCS and 10% dimethylsulfoxide (DMSO). The cell suspension wasprepared at a density of 30–40×106 cells per freezing vial. The freezing cycle wasmonitored by a NICOOL ST 20 (Air Liquide, France) and the gradient tempera-ture was −1.9°C/min from 4 to −30°C, and −30°C/min from −30 to −150°C.Vials were stored and dispatched to each laboratory in liquid nitrogen. Cryopre-served cells were thawed by immersing the vials in a 37°C water-bath immediatelyafter removing from the liquid nitrogen storage container. Thawed cell suspensionwas then transferred to L15 medium at 37°C, containing 10% FCS. After washing,

G. de Sousa et al. / Chemico-Biological Interactions 121 (1999) 77–97 81

viable hepatocytes were purified on a discontinuous density gradient with 37% and63% Percoll solutions and were resuspended in Medium I. Viability of thawedhepatocytes was determined as for freshly isolated cells.

2.3. Primary culture

Freshly isolated or thawed hepatocytes in suspension in Medium I were seeded ata density of 0.4×106/ml on collagen I coated Petri dishes (Corning). The plateswere incubated 4 h in a 37°C incubator under a humidified 5% CO2 atmosphere.The medium was then renewed with the same initial medium without FCS butsupplemented with 10−6 M hydrocortisone hemisuccinate (Medium II).

2.4. Cell attachment

Cell attachment efficacy was estimated 4 h after seeding by evaluating the ratioof intracellular lactate dehydrogenase (LDH) content over the total LDH containedin the initial cell suspension [20].

2.5. Light and electron microscopy

Phase contrast microphotographs of hepatocyte cultures were taken at differenttimes after cell seeding. For electron microscopy, the cells were fixed and processedaccording to standard procedures. Briefly, after 4 and 24 h of culture, hepatocyteswere fixed in 2.5% glutaraldehyde buffered with chilled sodium cacodylate 0.1 Mfor 5 min, then postfixed in 1% osmium tetroxide in sodium cacodylate 0.1 M for30 min, dehydrated in graded ethanol and embedded in Epon. Ultrathin sectionswere stained with uranyl acetate and lead citrate before examination.

2.6. Biochemical studies

LDH leakage in the extracellular medium was measured 20 h after seeding [19].Intracellular adenosine triphosphate (ATP) and glutathione (GSH) concentrationswere measured in hepatocytes just after isolation and thawing and 4 and 20 h afterseeding. ATP content was determined according to Lamprecht and Trautschold [21]by means of an ATP Bioluminescence HS kit (Boehringer) on a microplateluminometer (Labsystems). GSH content was measured by an enzymatic method[22,23]. Cholic acid conjugation and secretion were measured according to Boel-sterli et al. [24] using [14C]-cholic acid (specific activity 40 mCi/mmol, NENResearch Products). For protein synthesis studies, 22 h after seeding, culturedhepatocytes were incubated with 14C-leucine (specific activity 342 mCi/mmol,Amersham) for 2 h and the radioactivity incorporated in both secreted andintracellular proteins was estimated.

Immediately after isolation and thawing, 4 and 24 h after seeding, cytosolic andmicrosomal fractions were prepared by differential centrifugations and Phase I andPhase II enzymatic activities were assessed. Total cytochrome P450 (CYP) content

G. de Sousa et al. / Chemico-Biological Interactions 121 (1999) 77–9782

was determined in the microsomal fraction [25]. Ethoxycoumarin-O-deethylase(ECOD) and aldrin epoxidase (AE) activities were determined as described previ-ously [26,27]. Microsomal epoxide hydrolase (EH) activity was measured accordingto the method described by Moody et al. [28] with 3H-labeled cis-stilbene–oxide assubstrate. Cytosolic glutathione S-transferase (GST) activities were measured spec-trophotometrically with 1-chloro-2,4-dinitrobenzene (CDNB) and 1,2-dichloro-4-ni-trobenzene (DCNB) as substrates [29]. Cytosolic glutathione reductase (GR) andglutathione peroxidase (GPx) activities were determined by spectrophotometricanalysis [29,30]. GPx was assessed with hydrogen peroxide and tertiary butylhy-droperoxide as substrates [31]. Total cellular proteins were determined by theBio–Rad protein assay [32].

2.7. Drug metabolism studies

Metabolism of five drugs was analysed: two psychotropes, midazolam (MDZ)and diazepam (DZP); an antiviral, 3%-amino-3%-deoxythymidine (AZT), an anal-gesic, phenacetin (PHE) and the steroid, androst 4-ene-3, 17-dione (A4D). Experi-ments were initiated by the addition of one of these compounds to achieve theindicated final concentration: 50 mM MDZ, 10 mM DZP, 10 mM AZT, 200 mMPHE and 100 mM A4D. After appropriate exposure periods, the reaction wasstopped by removing the medium from the cell monolayer and scraping off thehepatocytes. The extracellular and intracellular compartments were analysed byHPLC methods previously described [17,19,33] except for A4D metabolism whichwas analysed according to Hussin and Skett [34].

2.8. Toxicologic studies

Three compounds were used as test molecules: amodiaquine chlorhydrate, ery-thromycin base and furosemide. Following a 20 h incubation in the presence of thetest compounds, the cytotoxic effects were evaluated by measuring the 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl-tetrazolium bromide (MTT) incorporation, ac-cording to the protocol described by Fautrel et al. [35]. Raw data was analyzed bya computer fitting programme in order to determine the inhibitory concentration(IC50) values for each test molecule.

3. Results

Hepatocytes were isolated from four whole monkey livers (mean liver weight:107916 g). The dissociation process lasted around 30 min and yielded 27.6917.9×106 hepatocytes per gram of dissociated tissue. Results presented in Table 1demonstrate that the experimental conditions used for isolating and culturing freshhepatocytes were suitable for monkey cells. The viability of freshly isolated hepato-cytes was estimated at around 86.5%. About 80% of these cells were immediatelycryopreserved and 90.5×109 hepatocytes were seeded in a culture medium

G. de Sousa et al. / Chemico-Biological Interactions 121 (1999) 77–97 83

Table 1Viability and recovery (mean9S.E.M., n=4)

Fresh hepatocytes Cryopreserved hepatocytes

After percollBefore percoll

92.090.6Viability (%) 86.591.3 69.692.45280.5Recovery (%) 10068.694.0Attachment (%) 81.093.0 nda

a nd: not determined. Results are means of four separate experiments9standard deviation.

containing 10% FCS. Within 4 h, 81.0% of the parenchymal cells attached to thecollagen coated support and began to spread (Table 1). Twenty hours after seeding,the cultures were of very high quality as shown by a low LDH leakage (5.691.3%).

After about 2 months of storage in liquid nitrogen, hepatocytes were thawed andshowed a viability 20% lower than freshly isolated cells. To recover a cell suspen-sion with a density of viable cells close to that of non-frozen ones, thawedhepatocytes were purified on a Percoll density gradient. This procedure increasedthe percentage of viable cells to 92% with a yield of more than 60% of the thawedhepatocyte population. However, a lower attachment yield and an increased LDHleakage (�25%) could reflect some membrane injuries resulting from cryopreserva-tion (Table 1). Nevertheless, as shown by the phase contrast and electron micro-scope photographs (Fig. 1), cells monolayers obtained from fresh and thawed cellslooked quite similar. Twenty four hours after seeding, fresh and thawed hepatocytesformed a stable monolayer with a classical polygonal shape and granular cytoplasm(1A, 1B). They exhibited the characteristic features of viable and functional cellswith well-developed endoplasmic reticulum (rough and smooth), numerous mito-chondria and nuclei containing characteristic nucleoli. Like fresh hepatocytes,thawed ones presented glycogen particles. Moreover, the primary cultures exhibitedsome intercellular cavities resembling bile canaliculi (1C, 1D).

A step-wise regular increase of intracellular ATP and GSH was observed both infresh and cryopreserved cultured hepatocytes (Table 2). Moreover, the ATP con-centration measured immediately after thawing was close to that obtained in freshly

Table 2ATP and GSH contents in fresh and cryopreserved monkey hepatocytes in primary culturea

ThawedFresh

ATP GSH ATP GSH

9.0493.09T0 14.2194.79 9.8393.22 7.4992.334 h 14.5294.0913.0593.0115.0894.24 20.9394.84

20.7197.08 46.88912.7224 h 16.7294.48 38.22911.88

a Results are expressed as nanomoles per milligram of protein and are means of four separateexperiments (with three replicates)9standard deviation.

G. de Sousa et al. / Chemico-Biological Interactions 121 (1999) 77–9784

Fig

.1.

Ele

ctro

n(A

,B

)an

dlig

ht(C

,D

)m

icro

scop

icap

pear

ance

ofm

onke

yhe

pato

cyte

scu

ltur

espr

epar

edfr

omfr

esh

(A,

C)

and

thaw

ed(B

,D

)ce

lls24

haf

ter

seed

ing.

G. de Sousa et al. / Chemico-Biological Interactions 121 (1999) 77–97 85

Table 3Conjugation and secretion of 14C-cholic acid in fresh and cryopreserved hepatocytes in primaryculture, 4 and 20 h after seeding, and incorporation of 14C-leucine after 20 h after seedinga

Protrein synthesis/1 hBile salts synthesis

20 h INT EXT4 h

136.25941.51155.81959.01 19.5915.84Fresh 148.82957.65211.039132.73156.969112.68 87.50937.97 7.7593.30Thawed

a Results are expressed as DPM per milligram of protein and are means of four separate experiments.INT: intracellular compartment; EXT: extracellular compartment.

isolated cells, unlike the intracellular GSH which was 45% lower. Twenty hoursafter seeding, ATP and GSH concentrations were 20–25% lower than before deepfreeze storage.

To evaluate the protein and bile salts synthesis and secretion properties of freshand cryopreserved monkey hepatocytes, the incorporation of radiolabelled precur-sors into these molecules was monitored (Table 3). The measurement of bile acidsconjugates (mainly taurocholate), 4 and 20 h after seeding of fresh and thawedhepatocytes, showed a regular synthesis of these compounds in both conditions(Table 3). Their synthesis and excretion in the culture medium was however shownto be increased in thawed cells 20 h after seeding. On the other hand, theincorporation of 14C-leucine in intra- and extra-cellular proteins, demonstrated alsothat both fresh cells and their cryopreserved counterparts were able to synthetizeand export proteins (Table 3). A decrease of the total de novo synthesized proteins(about 39%) was however observed after deep freeze storage.

From the results presented in Table 4, it appeared that the total CYP content wasmaintained over 24 h at the value found in freshly isolated cells. This was also truefor some CYP-related enzymatic activities such as ECOD or AE, as well as forother detoxication activities (GST, GR, GPx and EH). Indeed, when fresh monkeyhepatocytes were seeded, only a slight decrease of these functions (about 18%) wasobserved during the first 4 h. During the following 20 h, CYP and CYP-relatedenzymes activities weakly increased (from 14 to 24%) but whole activities wereroughly maintained at their initial level. When measured immediately after thawing,only the activities of AE and GST seemed to be significantly impaired (30 to 48%decrease). EH was, in contrast, found 21% higher in hepatocytes after cryopreserva-tion than before. When thawed monkey hepatocytes were seeded in primaryculture, the changes in the enzymatic activities were similar to fresh ones, except forCYP and ECOD activities, which dropped, respectively by 42 and 50% of their T0

values. In contrast, after 24 h of culture of thawed hepatocytes, GPx activities werehigher than those found in fresh cells.

The biotransformation of MDZ was analyzed in intra- and in extracellularcompartments, 4 (data not shown) and 24 h after exposure of hepatocytes to thedrug. MDZ metabolic profiles in fresh and thawed cells were similar. Both Phase I(1-hydroxy, 4-hydroxy and 1,4-dihydroxy-derivatives) and Phase II (glucuronides

G. de Sousa et al. / Chemico-Biological Interactions 121 (1999) 77–9786

and a compound referred to as X, probably sulfo-conjugated) bioproducts wereformed. At 4 h, in fresh as well as in thawed cells, the 1-hydroxylated derivativerepresented the main metabolite whereas the glucuronide became predominant at24 h (Fig. 2). The parent drug accounted only for 12 and 15% of the totalextracellular radioactivity in non-frozen and cryopreserved hepatocytes, respec-tively. No accumulation of midazolam occurred in hepatic cells.

With respect to DZP, four hours after exposure of both fresh and thawed cellswith a 10 mM concentration, the drug was extensively N-demethylated to nor-diazepam. The 3-hydroxylation pathway also contributed to the disappearance ofDZP with a transitional appearance of temazepam, further converted to oxazepamby N-demethylation, both minor metabolites (data not shown). After 24 h incuba-tion, these Phase I bioproducts underwent conjugation with glucuronic acid, butnordiazepam level remained high. Moreover at this time, the substrate was shownto be entirely metabolized in fresh as well as in thawed cultured hepatocytes (Fig.2). No accumulation of diazepam occurred in hepatic cells.

AZT metabolism was studied under the experimental conditions used for thelatter diagnostic substrates. Similar biotransformation derivatives, namely the 3%-amino-3%-deoxythymidine (AMT), the glucuronide of AZT (GAZT) and tritatedwater were obtained in intra- and extracellular media of fresh and thawed hepato-cytes. Twenty-four hours after the beginning of the incubation with 10 mM of AZT,these three metabolites represented 65 and 53% of the total extracellular radioactiv-ity in fresh and cryopreserved cells, respectively, with GAZT as the major metabo-lite (Fig. 3).

Table 4Enzyme activities in fresh and cryopreserved monkey hepatocytes in primary culturea

Fresh Thawed

4 hT=0 24 h4 hT=024 h

6.890.8 4.592.8 3.792.6 4.092.9GST-CDNB 6.691.0 7.291.66.794.36.594.38.595.211.992.8GST-DCNB 14.793.716.193.1

GR 11.290.813.294.1 12.794.3 9.293.1 10.592.210.891.9GPx-HP 0.290.130.1490.080.1390.010.0990.010.0990.020.1190.02

0.2490.030.2990.050.2690.04 0.3490.10.2590.030.3090.04GPx-tBHPCYP 0.5790.48 0.6590.33 0.7590.16 0.5490.17 0.3190.020.7290.37

1.4190.67 1.1790.58 1.3490.61ECOD 1.5890.23 1.2990.25 0.8090.2614719366 5619187AE 7409129103092341648919513349123

17.6494.5116.2592.1816.4092.93 15.4592.8314.1295.3913.3994.11EH

a GST-CDNB: glutathione S-transferase activity (mmol CDNB-GSH/mg prot/min) measured by1-chloro-2,4-dinitrobenzene; GST-DCNB: glutathione S-transferase activity (nmol CDNB-GSH/mgprot/min) measured by dichloronitrobenzene; GR: glutathione reductase and GPx: glutathione peroxi-dase activity (U/mg protein) measured by hydrogen peroxide (HP) and tertiary butyl hydroperoxide(tBHP); CYP: cytochrome P450 (nmol/mg prot); ECOD: ethoxycoumarin O-deethylase (nmol/mgprot/min); AE: aldrin epoxidase (ng dieldrin/mg prot/min); EH: epoxide hydrolase (nmol diol/mgprot/min). Results are means of three separate experiments9standard deviation.

G. de Sousa et al. / Chemico-Biological Interactions 121 (1999) 77–97 87

Fig. 2. Metabolic fate of midazolam (A) and diazepam (B) in fresh (FH) and cryopreserved (CH)monkey hepatocytes in primary culture after 24 h incubation period in extracellular compartment.M-GLU: glucuronides; X: unknown metabolite; Di-OH: 1,4-dihydroxy-midazolam; 4-OH: 4-hydroxy-midazolam; 1-OH: 1-hydroxy-midazolam; MDZ: midazolam. D-GLU: glucuronides; OX- oxazepam;TEM: temazepam; NOR: nordiazepam; DZP: diazepam.

Phenacetin, the fourth selected enzyme marker activity, was converted intoparacetamol and its respective sulfo- and glucurono-conjugates. In this case also,the overall metabolic pattern of this drug was qualitatively similar in fresh and inthawed hepatocytes. However, as for the other compounds tested, large inter-indi-vidual qualitative variations could be observed (Fig. 3).

The metabolism of androst-4-ene-3, 17-dione showed the expected products,namely a predominance of 6b-hydroxylated and 17-reduced metabolites withsmaller amounts of 7a- and 16a-hydroxylated and 5a reduced compounds (Fig. 4).The metabolism of this substrate was followed in culture over 72 h both before andafter cryopreservation. Before cryopreservation there was an increase in all enzymeactivities over the first 24h and then the activities stabilized for the rest of the

G. de Sousa et al. / Chemico-Biological Interactions 121 (1999) 77–9788

period studied (Fig. 4). Directly after cryopreservation (data not shown) theactivities of the enzymes showed a marked increase for the 7a- and 6b-hydroxylasesand 5a-reductase with no significant difference for the 16a-hydroxylase and 17-oxosteroid oxidoreductase. Culturing of the cells after cryopreservation led to a fallin all activities for the first 24 h but subsequently all the activities rose to achievea level above the zero value (Fig. 4). With the exception of the 16a-hydroxylase, allactivities rose to higher values after cryopreservation than before. The enzymeactivities in cultured, cryopreserved cells seemed to have stabilized after 48 h inculture.

Fig. 3. Metabolic fate of AZT (A) in fresh (FH) and cryopreserved (CH) monkey hepatocytes in primaryculture after 24 h incubation period in extracellular compartment, and of phenacetin (B) after 20 hincubation period. THO: tritiated water; AMT: 3%-amino-3%-deoxythymidine; GAZT: 5%-O-glucuronide ofAZT; PAR: paracetamol; SUL-PAR: paracetamol sulfoconjugation; GLU-PAR: paracetamol glu-curonidation.

G. de Sousa et al. / Chemico-Biological Interactions 121 (1999) 77–97 89

Fig. 4. The metabolism of androst-4-ene-3, 17 dione by freshly isolated (A) and crypopreserved monkeyhepatocytes (B) after various times in culture. The enzymes measured were 7a -, 6b-and 16 a-hydroxy-lases, the 17-oxosteroid oxidoreductase and the 5a- reductase. Results are expressed as mean9S.D.(n=4).

Finally, cytotoxicity tests performed on freshly isolated and cryopreserved hepa-tocytes in primary culture which were exposed for 20 h with three well-documentedxenobiotics, amodiaquin, erythromycin and furosemide, demonstrated very closeIC50 values for these drugs. These results suggest that the freezing/thawing steps didnot affect the cell sensitivity to toxic compounds (Table 5).

4. Discussion

The adjustment, in terms of perfusion media composition, time and flow-rate ofa method initially developed for isolating human hepatocytes from whole organ

G. de Sousa et al. / Chemico-Biological Interactions 121 (1999) 77–9790

[19], gave 2.891.8 billion hepatocytes from a whole monkey liver (around 28million/g of tissue). Cellular viability and recovery were similar to those alreadypublished [16,19]. For each experiment, about 80% of the hepatocyte populationwas immediately frozen. Two different cryopreservation protocols were firstlycompared: one originally devised by de Sousa et al. for human hepatocytes [14,16],the other developed by Chesne and Guillouzo for rat hepatocytes [9]. Thesemethods differed mainly with respects to the freezing medium composition, thedensity of the frozen cells and the cooling technique, but both allowed the recoveryof viable and functional hepatocytes after thawing. However, the first technique [14]was selected by all the participants as it allowed, i) the cryopreservation of largeramounts of hepatocytes (30–40×106 cells/vial instead of 5×106) required for themulticentre study, ii) the use of an automatic cooling machine which increased theexperimental reproducibility and iii) much more vials to be processed simulta-neously and under the same experimental conditions.

In accordance with the published works in the field of hepatocyte cryopreserva-tion [9,10,36,37], the freezing medium contained 10% of the classical permeantcryoprotectant, DMSO. This solvent not only reduces ice nucleation but is alsosupposed to have a protective effect on cell membranes by interacting electrostati-cally with phospholipids [38]. Other non-penetrating agents holding cryoprotectiveproperties such as PVP, BSA and FCS were also used. These compounds increasethe medium viscosity therefore avoiding cell sedimentation during the cryopreserva-tion process. Finally, to meet the requirement of theoretical cryopreservationprinciples [39], a two-step cooling rate (first −1.9°C/min then −30°C/min) waschosen. This allowed a progressive cell dehydration, therefore limiting the forma-tion of harmful intracellular ice and avoiding high cellular solute concentrations[39].

This protocol resulted in the optimal freezing of most of the isolated hepatocytes.Indeed, about 80% of the viable cells obtained after isolation were recovered afterdeep freeze storage, data which is in agreement with the best results published onhuman and monkey hepatocytes [16,17,40]. However 20% of non-viable cells wasobtained after thawing, which resulted in a rather poor attachment of the hepato-cytes when set in primary culture. The viable cell population was therefore purifiedby Percoll density gradient from the whole thawed hepatocyte suspension. Thisprocedure improved the hepatocyte viability (from 70 to 92%) and was demon-

Table 5Cytotoxicity assays by means of the MTT test

Compound IC50 (mg/ml)

ThawedFresh

Amodiaquin HCl 9.992.1 14.194.2427.09138.5478.0984.1Erythromycin

439.7974.1Furosemide 435.09213.6

G. de Sousa et al. / Chemico-Biological Interactions 121 (1999) 77–97 91

strated by other authors to lead to a better recovery of hepato-specific functions[10,16,18,40]. It, however, resulted in a loss of about 30% of the thawed hepatocytepopulation.

When plated, these purified hepatocytes rapidly attached to their collagen coatedsupport and exhibited similar morphological aspects to fresh cells. However, inspite of having respected the basic rules of cryopreservation (in terms of cryoprotec-tants, slow cooling and fast warming rates), storage at −196°C resulted in a 50%decrease of hepatocyte attachment efficacy and significant increase of membranepermeability. This confirmed the data published so far [11,12,17] which showedthat, although attachment of cryopreserved hepatocytes depended on the protocoland species used, it was usually reduced to about 20–30% as compared to freshcells. These impairments could reflect the membrane injuries resulting from thedrastic physico–chemical changes, such as cell dehydration/rehydration or watercrystallization, provoked by the freeze/thaw processes [41–43].

Despite the crucial role of ATP in cell function and integrity, due in particular toits involvement in overall cellular biosynthesis pathways, only few studies reportedthe effect of cryopreservation on its intracellular concentration. Numerous exoge-nous factors, such as hypothermia [44] or anoxia [45] were shown to modify ATPconcentration and to induce cell injuries [45,46], but data on the effects of deepfreeze storage are conflicting. Indeed, Toschakov et al. showed a 82% decrease ofintracellular ATP after this procedure [47]. Similarly, Lawrence and Benfordshowed a 39% decrease of this parameter 4 h after seeding of cryopreserved rathepatocytes, as compared to fresh ones [10]. In contrast Zaleski et al. measuredsimilar ATP concentrations in fresh and thawed rat hepatocytes [48], and Kasai andMito concluded that cryopreservation did not affect ATP concentrations in doghepatocytes [12]. Our results agree with the latter work, demonstrating the preserva-tion of the cellular ATP pool and therefore of the integrity of the multienzymaticsystem (such as mitochondrial cytochromes) involved in its production.

The maintenance of hepatocyte integrity also depends on intracellular GSHcontent. It could in particular drop when organs are preserved in hypothermia [49]or during the hepatocytes isolation procedure [50]. As it was already observed byLawrence and Benford using rat hepatocytes [10], our data showed that monkeyhepatocytes have lost 45% of their GSH content after cryopreservation. Thisdecrease may either result from a consumption of this tripeptide by detoxicationreactions, or from a significant GSH leakage into the extracellular medium due tomembrane injuries. Nevertheless, thawed hepatocytes, as their fresh counterparts inculture, kept their ability to synthesize GSH. This seems indicative of the return tothe in vivo situation after the GSH depletion due to the isolation process, aspreviously reported in fresh rat hepatocytes [51]. An over-stimulation of celldetoxication mechanisms resulting from an additional oxidative stress provoked bythe freeze/thaw process could be ruled out since monkey hepatocytes behavedsimilarly before and after cryopreservation.

Fresh cultured monkey hepatocytes as human and rat ones [16,17] were able tosynthesize and export proteins. These functions were kept but reduced aftercryopreservation, as previously observed for human and rat cells [9,10,13]. This is

G. de Sousa et al. / Chemico-Biological Interactions 121 (1999) 77–9792

probably a consequence of the membrane injuries occurring during the freeze/thawprocesses, which could result in the impairment of the amino-acids transport fromthe extracellular medium [43]. Dou et al. showed a recovery of these functions incryopreserved human hepatocytes after 38 h of culture [16]. On the other hand,synthesis of albumin or apolipoprotein AI by cultured monkey hepatocytes was notaffected by cryopreservation [11].

Cultured thawed hepatocytes, as fresh ones, were able to conjugate and to secretebile acids. However, the excretion was shown to be increased 20 h after seeding.This is possibly related, as mentioned below, either to membrane damages or toactivation of the cellular detoxication mechanisms in response to the freezing/thawing.

Previous work on freshly isolated rat hepatocytes in primary culture demon-strated that both Phase I and Phase II biotransformation activities continuouslydeclined as a function of culture time [51,52]. The present study demonstrated thatthese activities were much more stable in fresh monkey hepatocytes than in rodentones. Several authors showed that the maintenance of these activities after thawingnot only depended of the cryopreservation protocol used but also on the multi-en-zyme system considered. For instance, Powis et al. [18] reported a decrease ofglobal CYP and related activities in thawed rat and dog hepatocytes. Theirmaintenance [53] or even their increase [36] in rat hepatocytes after deep freezestorage was also described. Although only few data are available on non-CYPdependent parameters, Diener et al. showed a better preservation of membraneenzyme activities (such as CYP, UGT or EH) than cytosolic ones (as GST andsulfotransferases), in cryopreserved rat hepatocytes [40].

Our results show that, independently of their subcellular localization, Phase I andPhase II activities are qualitatively entirely recovered after monkey hepatocytecryopreservation. However a partial quantitative loss of some activities was ob-served (AE, GST), which reflects a decrease in the proportion of functionalhepatocytes, despite their excellent viability. This could also be due to the enzymedegradation provoked by physico–chemical freeze/thaw shocks (changes of saltconcentration and pH resulting in the denaturation of the membrane or cytosolicproteins). The decrease in activities could hardly be explained, as suggested by someauthors [18,40], by the loss of co-factors caused by membrane injuries, since themajor co-factors required (NADPH, GSH) for optimal cell functioning were addedto the reaction mixture.

Our results demonstrated a better stability of non-CYP dependent activities(GST, GR, GPx) than CYP-related ones (CYP, ECOD, AE), in cultured thawedmonkey hepatocytes. To our knowledge, no data have been published on theseparameters in cryopreserved hepatocytes from this species. Some studies carried outon human [10] and rodent [10,17] hepatocytes maintained in culture for 4–72 h,showed a similar CYP content before and after cryopreservation. In contrast,Jackson et al. observed a sharper decrease of the ECOD activity during culture ofthawed rat hepatocytes as compared to that estimated in fresh cells [36]. Moreover,according to Loretz et al. [36], nearly 100% of this activity were recovered aftercryopreservation and culture of rat hepatocytes. Conversely, about a 20% increase

G. de Sousa et al. / Chemico-Biological Interactions 121 (1999) 77–97 93

of EH was obtained after cryopreservation. Such a phenomenon has already beenobserved by Jackson et al. [36] in cryopreserved rat hepatocytes, and may resultfrom a selection of some hepatocyte sub-populations during the cryopreservationand/or thawing. Indeed, it has been established that some enzymes, such as GST orECOD in the rat [54] or EH in man [55], are found in higher proportions inperivenous than in periportal hepatocytes.

In order to evaluate the integrity of thawed hepatocyte biotransformationenzymes, five drugs, MDZ, DZP, AZT, PHE and A4D, were selected as diagnosticsubstrates. The choice was made based on the following considerations: i) a goodknowledge of their pharmacokinetics and more particularly of their metabolism inhuman and various animal species; ii) a qualitative as well as a quantitativeinterspecies variability of these processes; and iii) finally, the involvement of a largepanel of metabolic enzymes in their hepatic activation/deactivation. For all thesecompounds an excellent correlation was obtained between the metabolic profilesdescribed in vivo and those observed in fresh monkey hepatocytes. This holds trueboth in terms of Phase I (mono and di-hydroxylation, N-demethylation, reduc-tion…) and Phase II (mainly sulfo- and glucurono-conjugation) biotransformationpathways, therefore demonstrating a good maintenance of the correspondingenzymes families or subfamilies after hepatocyte isolation. Furthermore, as con-cerns the effects of deep freeze storage, similar metabolic profiles were found infresh and thawed hepatocytes, not only qualitatively but also quantitatively. Thesedata confirmed the excellent preservation of some important phase I and phase IIdetoxication systems, as already found by Dou et al. [16] on thawed humanhepatocytes by using a single diagnostic substrate.

Finally, several publications have pointed out the potential interest of cryopre-served rat, dog and human hepatocytes [9,14,16] in performing in vitro toxicologicalinvestigations. In this respect, our data on monkey hepatocytes indicated thatfreeze/thaw processes had no major effect on the sensitivity of the hepatocytestowards hepatotoxic compounds and confirmed the results obtained by otherauthors.

In conclusion, the functional characterization of fresh and thawed monkeyhepatocytes showed that the freezing and thawing conditions defined in this studywere reproducible and allow the storage of cells able to survive and to function inprimary culture over at least 24 h and maybe up to 72 h. Indeed, in spite of theimpairment of some parameters, the large majority of the hepato–specific biochem-ical functions investigated were qualitatively well preserved. Furthermore, thepresent study demonstrated that the xenobiotic metabolizing function, whichconstitutes one of the most interesting hepatocyte properties regarding drug re-search and development, were globally maintained. Lastly, deep freeze storageseemed not to modify the hepatocyte sensitivity to toxic compounds. Therefore,according to these set of criteria, cryopreserved monkey hepatocytes appeared to beas reliable as fresh ones, for use in various pharmaco–toxicological area includingmetabolic studies on new molecules and the their effect on drug metabolizingenzymes expression [56,57]

G. de Sousa et al. / Chemico-Biological Interactions 121 (1999) 77–9794

To our knowledge, this multi-laboratory study constitutes the first work dealingwith the effects of cryopreservation on monkey hepatocytes by analyzing such alarge number of hepato–specific parameters. Experiments are now being performedin order to develop the optimal purification, cryopreservation, thawing and primaryculture conditions for human and dog hepatocytes, using a similar validationstrategy. This would result in the establishment of a bank of cryopreservedhepatocytes, from higher mammals, which have their specific functions preservedafter long term storage periods (several months or years) in liquid nitrogen. Thiscell bank would constitute a permanent and routinely applicable tool for phar-maco–toxicological purposes. It would, in particular, allow comparative studiesabout the interspecies variability on drug metabolism and toxicity, independent ofthe ethical, scientific and technical limits linked to the in situ perfusion of animalliver and of the difficulty in working on human liver. It could also serve ininvestigating or even predicting the carcinogenic or enzyme inducing potentials ofenvironmental xenobiotics.

Besides these pharmaco–toxicological applications, such in vitro models wouldbe of great interest in various clinical domains, in particular for studying viral(hepatitis) or parasitic (malaria) diseases where hepatocytes represents one of themain targets of pathogen infection. Finally, due to the increasing demand of donorlivers for transplantation and the limited possibility of preserving these organsfunctional over long time-periods, their availability becomes more and more scarce.Hepatocytes constitute, in that context, a useful tool for trying to optimise humanliver preservation solutions and use, and may be a promising alternative to wholeliver transplantation through the development of extracorporal bioreactors [58] orhepatocyte transplantation [59,60] techniques.

The development of these biomedical applications is, however, limited until nowby the lack of a regular supply of fresh human hepatocytes. The creation of a bankof cryopreserved cells would greatly facilitate the use of such an in vitro model inthe study of hepatic functions and the understanding, even the treatment, of somehepatic pathologies of chemical, infectious or genetic origins.

Acknowledgements

We are deeply grateful to Dr Bretau and Dr Sandowsky of the Institute forTropical Medicine, for their help in performing experimentations on animals. Thesupport the Home Office Animal Procedures Committee to one of the laboratory(PS) is also gratefully acknowledged.

References

[1] R. Rahmani, B. Richard, G. Fabre, J.P. Cano, Extrapolation of preclinical pharmacokinetic datato therapeutic drug use, Xenobiotica 18 (1988b) 71–88.

G. de Sousa et al. / Chemico-Biological Interactions 121 (1999) 77–97 95

[2] S.A. Wrighton, M. Vandenbranden, J.C. Stevens, L.A. Shipley, B.J. Ring, A.E. Rettie, J.R.Cashman, In vitro methods for assessing human hepatic drug metabolism: their use in drugdevelopment, Drug Metab. Rev. 25 (1993) 453–484.

[3] A. Guillouzo, F. Morel, O. Fardel, B. Meunier, Use of human hepatocyte cultures for drugmetabolism studies, Toxicology 82 (1993) 209–219.

[4] A.P. Li, Primary hepatocyte culture as an in vitro toxicological system of the liver, in: S.C. Gad(Ed.), In Vitro Toxicology, Raven Press, UK, 1994, pp. 195–220.

[5] T. Seddon, I. Michelle, R.J. Chenery, Comparative drug metabolism of diazepam in hepatocytesisolated from man, rat, monkey and dog, Biochem. Pharmacol. 38 (1989) 1657–1665.

[6] B. Richard, G. Fabre, G. de Sousa, I. Fabre, R. Rahmani, J.-P. Cano, Interspecies variability inmitoxantrone metabolism using primary cultures of hepatocytes isolated from rat, rabbit andhumans, Biochem. Pharmacol. 41 (1991) 255–262.

[7] R.J. Chenery, A. Ayrton, H.G. Oldham, P. Standring, S.J. Norman, T. Seddon, R. Kirby,Diazepam metabolism in cultured hepatocytes from rat, rabbit, dog, guinea pig, and man, DrugMetab. Dispo. 15 (1987) 312–317.

[8] R. Rahmani, G. de Sousa, F. Marre, F. Nicolas, M. Placidi, in: V. Rogiers, W. Sonck, E. Shephard,A. Vercruysse (Eds.), Potential of freshly isolated and cryopreserved human hepatocytes in drugresearch and development, In Human cells in in vitro pharmaco–toxicology, 1993 pp. 117–138.

[9] C. Chesne, A. Guillouzo, Cryopreservation of isolated rat hepatocytes: a critical evaluation offreezing and thawing conditions, Cryobiology 25 (1988) 323–330.

[10] J.N. Lawrence, D.J. Benford, Development of an optimal method for the cryopreservation ofhepatocytes and their subsequent monolayer culture, Tox. in Vitro 5 (1991) 39–50.

[11] E.L. Sun, D.G. Aspar, R.G. Ulrich, G.W. Melchior, Cryopreservation of cynomolgus monkey(Macaca Fascicularis) hepatocytes for subsequent culture and protein synthesis studies, In VitroCell. Dev. Biol. 25 (1990) 147–150.

[12] S. Kasai, M. Mito, Large-scale cryopreservation of isolated dog hepatocytes, Cryobiology 30 (1993)1–11.

[13] P.J.M. Rijntjes, H.J. Moshage, P.J.L. Van Gemert, R. De Waal, S.H. Yap, Cryopreservation ofadult human hepatocytes: the influence of deep freezing storage on the viability, cell seeding,survival, fine structures and albumin synthesis in primary culture, J. Hepatol. 3 (1986) 7–18.

[14] G. de Sousa, M. Dou, D. Barbe, B. Lacarelle, M. Placidi, R. Rahmani, Freshly isolated orcryopreserved human hepatocytes in primary culture: Influence of drug metabolism on hepatotoxi-city, Tox. in Vitro 5/6 (1991) 483–486.

[15] C. Chesne, C. Guyomard, L. Grislain, C. Clerc, A. Fautrel, A. Guillouzo, Use of cryopreservedanimal and human hepatocytes for cytotoxicity studies, Tox. in Vitro 5/6 (1991) 479–482.

[16] 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.

[17] 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 andhuman beings after cryopreservation, Hepatology 18 (1993) 406–414.

[18] G. Powis, K.S. Santone, D.C. Melder, L. Thomas, D.J. Moore, T.J. Wilke, Cryopreservation of ratand dog hepatocytes for studies of xenobiotic metabolism and activation, Drug Metab. Disp. 15(1987) 826–832.

[19] G. Fabre, R. Rahmani, M. Placidi, J. Combalbert, J. Cov, J.-P. Cano, C. Coulange, M. Ducros, M.Rampal, Characterization of midazolam metabolism using human hepatic microsomal fractions andhepatocytes in suspension obtained by perfusing whole human livers, Biochem. Pharmacol. 37(1988) 4389–4397.

[20] F. Wroblevski, J.C. Ladue, Lactic dehydrogenase in blood, Proc. Soc. Exp. Biol. 90 (1955)210–214.

[21] Lamprecht, H.U. Bergmeyer (Ed.), Trautschold, Methods of Enzymatic Analysis (1974) 2101.[22] M.E. Anderson, Determination of glutathione and glutathione disulfide in biological samples,

Methods in Enzymol. 113 (1985) 548–555.[23] M.A. Baker, G.J. Cerniglia, A. Zaman, Microtiter plate assay for the measurement of glutathione

and glutathione disulfide in large numbers of biological samples, Anal. Biochem. 190 (1990)360–365.

G. de Sousa et al. / Chemico-Biological Interactions 121 (1999) 77–9796

[24] U.A. Boelsterli, P. Bouis, J.F. Brouillard, P. Donatsch, In vitro toxicity assessment of cyclosporinA and its analogs in a primary rat hepatocyte culture model, Toxicol. Appl. Pharmacol. 96 (1988)212–221.

[25] T. Omura, R. Sato, The carbon monoxide-binding pigment of liver microsomes, J. Biol. Chem. 239(1964) 2370–2378.

[26] V. Rogiers, L. Adriaenssens, Y. Vandenberghe, E. Gepts, A. Callaerts, A. Vercruysse, Criticalevaluation of 7-ethoxycoumarin O-deethylase activity measurements in intact isolated rat hepato-cytes, Xenobiotica 16 (1986) 817–828.

[27] V. Rogiers, Y. Vandenberghe, A. Callaerts, W. Sonck, V. Maes, A. Vercruysse, The inducing andinhibiting effects of sodium valproate in vivo on the biotransformation systems of xenobiotics inisolated rat hepatocytes, Xenobiotica 18 (1988) 665–673.

[28] D.E. Moody, M.H. Silva, B.D. Hammock, Epoxide hydrolysis in the cytosol of rat liver, kidney,and testis. Measurement in the presence of glutathione and the effect of dietary Clofibrate,Biochem. Pharmacol. 35 (1985) 2073–2080.

[29] W.H. Habig, M.J. Pabst, W.B. Jakoby, Glutathione S-transferases: the first enzymatic step inmercapturic acid formation, J. Biol. Chem. 249 (1974) 7130–7139.

[30] N. Tanigushi, Y. Tsukada, K. Mukuo, H. Hirai, Effect of hepatocarcinogenic azo dyes onglutathione and related enzymes in rat liver, GANN. 65 (1974) 381–387.

[31] I. Calberg, B. Mannervik, Purification and characterization of the flavoenzyme glutathione reduc-tase from rat liver, J. Biol. Chem. 250 (1975) 5475–5480.

[32] M.M. Bradford, A rapid and sensitive method for the quantification of mg quantities of proteinutilizing the principal of protein dye binding. Anal Biochem. 72 (1976) 248–254.

[33] F. Nicolas, G. de Sousa, P. Thomas, M. Placidi, G. Lorenzon, R. Rahmani, Comparativemetabolism of 3%-azido-3%-deoxythymidine in cultured hepatocytes from rat, dog, monkey and man,Drug. Metab. Dispo. 23 (1995) 308–313.

[34] A.H. Hussin, P. Skett, Maintenance of steroid metabolism in primary-culture of adult rat hepato-cytes in serum-free medium, Biochem. Soc. Trans. 14 (1986) 914–915.

[35] A. Fautrel, C. Chesne, A. Guillouzo, G. de Sousa, M. Placidi, R. Rahmani et al., A multicentrestudy of acute in vitro cytotoxicity in rat liver cells, Tox. in vitro 5/6 (1991) 543–547.

[36] B.A. Jackson, J.E. Davies, J.K. Chipman, Cytochrome P-450 activity in hepatocytes followingcryopreservation and monolayer culture, Biochem. Pharmacol. 34 (1985) 3389–3391.

[37] L.J. Loretz, A.P. Li, M.W. Flye, A.G.E. Wilson, Optimization of cryopreservation procedures forrat human hepatocytes, Xenobiotica 19 (1989) 489–498.

[38] T.J. Anchordoguy, C.A. Cecchini, J.H. Crowe, L.M. Crowe, Insights into the cryoprotectivemechanism of dimethyl sulfoxide for phospholipid bilayers, Cryobiology 28 (1991) 467–473.

[39] J. Farrant, General observations on cell preservation, in: M.J. Ashwood-Smith, J. Farrant (Eds.),In Low temperature preservation in medicine and biology, Pitman Medical, USA, 1980, pp. 1–18.

[40] B. Diener, D. Utesch, N. Beer, H. Durk, F. Oesch, A method for the cryopreservation of liverparenchymal cells for studies of xenobiotics, Cryobiology 30 (1993) 116–127.

[41] B.W.W. Grout, B.J. Fuller, Hepatocyte plasma membrane response to hypertonic stress and relatedcell viability, Cryo-Lett. 3 (1982) 51–56.

[42] B.J. Fuller, B.W. Grout, R.J. Woods, Biochemical and ultrastructural examination of cryopreservedhepatocytes in rat, Cryobiology 5 (1982) 493–502.

[43] R. De Loecker, B.J. Fuller, J. Gruwez, W. De Loecker, The effects of cryopreservation onmembrane integrity, membrane transport, and protein synthesis in rat hepatocytes, Cryobiology 27(1990) 143–152.

[44] F.O. Belzer, J.H. Southard, Principles of solid-organ preservation by cold storage, Transplantation45 (1988) 673–676.

[45] J.D. Blankensteijn, O.T. Terpstra, Liver preservation: the past and the future, Hepatology 13 (1991)1235–1250.

[46] F.A.M. Redegeld, R.M.W. Moison, A.S. Koster, J. Noordhoek, Depletion of ATP but not of GSHaffects viability of rat hepatocytes, Eur. J. Pharmacol. 228 (1992) 229–236.

[47] V.Y. Toshchakov, G.I. Morozova, N.A. Onishchenko, Energy metabolism of cryopreserved rathepatocytes, Cryo-Lett. 12 (1991) 259–272.

G. de Sousa et al. / Chemico-Biological Interactions 121 (1999) 77–97 97

[48] J. Zaleski, J. Richburg, F.C. Kauffman, Preservation of the rate and profile of xenobioticmetabolism in rat hepatocytes stored in liquid nitrogen, Biochem. Pharmacol. 46 (1993) 111–116.

[49] P.K. Vreugdenhil, F.O. Belzer, J.H. Southard, Effect of cold storage on tissue and cellularglutathione, Cryobiology 28 (1991) 143–149.

[50] J. Vina, R. Hems, H.A. Krebs, Maintenance of glutathione content in isolated hepatocytes,Biochem. J. (1978) 627–630.

[51] K. Mertens, V. Rogiers, A. Vercruysse, Glutathione dependent detoxication in adult rat hepatocytesunder various culture conditions, Arch. Toxicol. 67 (1993) 680–685.

[52] V. Rogiers, Y. Vandenberghe, A. Callaerts, G. Verleye, M. Cornet, K. Mertens, W. Sonck, A.Vercruysse, Phase I and phase II xenobiotic biotransformation in cultures and co-cultures of adultrat hepatocytes, Biochem. Pharmacol. 40 (1990) 1701–1706.

[53] J.G. Shaddock, J.E. Snawder, D.A. Casciano, Cryopreservation and long-term storage of primaryrat hepatocytes: effects on substrate-specific cytochrome P450-dependent activities and unscheduledDNA synthesis, Cell. Biol. Toxicol. 9 (1993) 345–357.

[54] E.M. Suolinna, K.E. Penttila, B.M. Winell, A.C. Sjoholm, K.O. Lindros, Drug metabolism byperiportal and perivenous rat hepatocytes: comparison of phase I and phase II reactions and theirinducibility during culture, Biochem. Pharmacol. 38 (1989) 1329–1334.

[55] D. Ratanasavanh, P. Beaune, F. Morel, J.-P. Flinois, F.P. Guengerich, A. Guillouzo, Intralobulardistribution and quantitation of cytochrome P-450 enzymes in human liver as a function of age,Hepatology 13 (1991) 1142–1151.

[56] G. de Sousa, S. Langouet, F. Nicolas, G. Lorenzon, M. Placidi, R. Rahamni, R. Rahmani, A.Guillouzo, Increase of cytochrome P-450 1A and glutathione transferase transcript in culturedhepatocytes from dogs, monkeys and humans after cryopreservation, Cell Biol. Tox. 12 (1996)351–358.

[57] A.P. Li, P. Maurel, M.J. Gomez-Lechon, L.C. Cheng, M. Jurima-Romet, Preclinical evaluation ofdrug-drug interaction potential: present status of the application of primary human hepatocytes inthe evaluation of cytochrome P450 induction, Chem. Biol. Interact. 107 (1997) 5–16.

[58] B. Fremond, C. Malandin, C. Guyomard, C. Chesne, A. Guillouzo, J.-P. Campion, Correction ofbilirubin conjugation in the Gunn rat using hepatocytes immobilized in alginate gel beads as anextracorporeal bioartificial liver, Cell Transplant. 2 (1993) 453–460.

[59] F.D. Ledley, Hepatic gene therapy: present and future, Hepatology 18 (1993) 1263–1273.[60] A.A. Demetriou, A. Felcher, A.D. Moscioni, Hepatocyte transplantation: a potential treatment for

liver disease, Digest Dis. Sci. 36 (1991) 1320–1326.

.