Comparative study of CYP1A1 induction by 3-methylcholanthrene in various human hepatic and epidermal cell types

Pergamon

Toxicology

in Vitro 11 (1997) 443-450

Comparative Study of CYPlAl Induction by 3-Methylcholanthrene in Various Human Hepatic and Epidermal Cell Types C. DELESCLUSE*, N. LEDIRAC, G. de SOUSA, M. PRALAVORIO, D. BOTTA-FRIDLUNDT, Y. LETREUTt and R. RAHMANI* *Equipe INSERM. Centre de Recherches INRA, 41 Bd du Cap, 06606, Antibes and tservice

d’Hipatologie,

Hdpital

de la Conception,

147 Bd Baille, 13005 Marseilles,

France

Abstract-Hepatocytes and keratinocytes are among the most widely used cells in pharmaco-toxicology, but a limitation of these models is the provision of human tissues on a regular basis. The suitability of HepG2, HaCaT and HEW cell lines as an acceptable substitute for primary cultures was examined. In these cell types, the effects of 3-methylcholanthrene (3.MC) were analysed on CYPl Al gene expression, a crucial CYP subfamily in the activation of chemical carcinogens. Ethoxyresorufin O-deethylase (EROD) activity was never detected in HESV cells, but in other cell types it was stimulated in a concentration-dependent manner (maximal induction, l-2.5 PM). Above this peak induction the effect fell rapidly. Northern blot analysis of CYPlAl mRNA agreed with the trends obtained for EROD values. However, the decrease of the EROD activity observed at the highest 3-MC concentrations was not correlated with CYPIAI mRNA reduction. This study also demonstrated that 3-MC is capable of significantly inducing CYPlAl in HaCaT cells (17-fold over conrrol), as in human hepatocytes (six- to 1S-fold) and HepG2 (fourfold). Therefore, in contrast to SV40-immortalized keratinocytes (HEW), spontaneously immortalized keratinocytes (HaCaT) may constitute a valuable tool for studying epidermal CYPlAl gene regulation by xenobiotics. 0 1997 Published by Elsevier Science Lrd Abbreviations:

CYP = cytochromes P-450; DMEM = Dulbecco’s DMSO = dimethyl sulfoxide; EROD = ethoxyresorufin U-dcethylase; 3-MC = 3-methylchoianthrene; MFO = mixed function oxidase.

INTRODUCTION

The liver plays a key role in the metabolism of most endogenous and exogenous substances. The reactions are mainly catalysed by cytochromes P-450 (CUP), a superfamily of constitutive and inducible haemoproteins (Okey, 1990). The induction of xenobiotic metabolism by drugs and other chemicals in humans is of clinical and toxicological importance. It has, however, proved difficult to study the effects of inducing agents on xenobiotic-metabolizing enzymes in humans directly and most of the information available in humans has been indirect, using biopsies from surgical patients. As the expression levels of the various enzyme species are higher in hepatocytes than in other cell types (Porter and Coon, 1991), many investigators have used cultured hepatocytes as an in vitro model for studying P-450 activity (Kocarek et al., 1993; Marre et al., 1996; Valles et al., 1995). Liver cell cultures have become especially valuable as a tool for investigating the hepatotoxic action (de Sousa et al., 1991 and 1995) and metabolic profile of drugs (Lacarelle et al., 1992; Nicolas et al., 1995; Valles *Authors

for correspondence.

0887-2333/97/%17.00 + 0.00 0 SSDI 0887-2333(97)00077-5

1997 Published

modified Eagle’s medium; FCS = foctal calf serum;

et al., 1993), as well as their interaction with other therapeutic agents or xenobiotics (Rahmani et al., 1993). However, primary hepatocyte cultures are phenotypically unstable, losing many of their differentiated functions, including mixed function oxidase (MFO) activity in 48-72 hr (Grant et al., 1987); furthermore, being non-proliferating they have a limited lifespan in culture. This represents a serious drawback to using this experimental model for long-term in vitro pharmacotoxicological studies. The HepG2 human hepatoma cell line is an immortal cell culture system that was found to be a good model for studying hepatic drug metabolism and its induction in humans (Doostdar et al., 1993). Indeed, these cells carry out both MFO and conjugation reactions (Doostdar et al., 1990; Duthie et al., 1988) and activate xenobiotics to cytotoxic and/or mutagenic metabolites (Duthie and Grant, 1989; Huh et al., 1982). Keratinocytes are also (but to a lesser extent) widely used cells in pharmacotoxicology. As hepatocytes, they present the advantage of being both the effector of biotransformation and the target system for toxicity. They contain numerous cytochrome P-450 enzymes and, in particular, CYPlAl isoforms (Merk et al., 1993). Moreover, it has been shown

by Elsevier Science Ltd. All rights

reserved.

Printed

in Great

Britain

C.

444

Delescluse ef al

recently that polycyclic aromatic hydrocarbons, such as 3-methylcholanthrene (3-MC), are capable of inducing CYPIAI in these cells (Raffali et al., 1994; Reiners et al., 1990) but the induction factor was much lower than in hepatocytes (Reiners et al., 1990). Cytochrome P4501AI can also be rapidly induced in these cells during the process of spontaneously induced or calcium-induced terminal differentiation in the absence of any xenobiotics (Sadek and Allen-Hoffmann, 1994). A major limitation of keratinocytes is the provision of tissues of human origin on a regular basis and, even when this is possible, there are problems of sexual differences and genetic polymorphism. To avoid these disadvantages, we therefore examined the suitability of epidermal cell lines as acceptable substitutes for primary cultures. In this paper, we demonstrate that 3-MC is capable of significantly inducing CYPlAl in HaCaT cells, as in hepatocytes and HepG2, but not in HESV. Therefore, in contrast to SV40-immortalized keratinocytes, spontaneously immortalized keratinocytes may constitute a valuable tool for studying epidermal CYPIAI gene regulation by xenobiotics. MATERIALS AND METHODS

Chemicals

Dulbecco’s modified Eagle’s medium (DMEM), penicillin-streptomycin, L-glutamine, sodium pyruvate, Eagle’s non-essential amino acids and foetal calf serum (FCS) were from Eurobio. Keratinocyte growth medium (MCDB 153) was from Clonetics. Williams’ medium, dimethyl sulfoxide (DMSO), glucosed-phosphate, glucose-6-phosphate dehydrogenase, P-NADP, and 3-MC were from Sigma. 7-Ethoxyresorufin and resorufin were from Boehringer. Hydrocortisone hemisuccinate was from Roussel-Uclaf. Cell culture and treatment Human hepatocytes were obtained from liver biopsies resected from secondary tumours. Hepatocytes were obtained, as previously described, by a two-step collagenase perfusion (Fabre et al., 1988). Freshly isolated cells were resuspended in William’s medium containing 10% FCS and supplemented with penicillin (50 U/ml) streptomycin (50 fig/ml), and insulin (0.1 U/ml). Freshly isolated cells were seeded on collagen type l-coated dishes. Plates containing hepatocytes were incubated for 4 hr at 37°C under a humidified 5% CO2 atmosphere. The medium was then renewed with the same initial medium without FCS but supplemented with hydrocortisone hemisuccinate and containing increasing concentrations of 3-MC over 72 hr, with a change of medium every 24 hr. Keratinocytes were purchased from Clonetics. They were cultured in keratinocyte growth medium:

MCDB 153, and incubated at 37°C in a humidified atmosphere containing 5% CO?. At 70% confluency, this medium was replaced by DMEM supplemented with 10% FCS, penicillin (100 U/ml), streptomycin (100 pg/ml), sodium pyruvate (1 mM), non-essential amino acids (0.1 mM) and L-glutamine (2 mM) for 24 hr. HepG2 and HaCaT were cultured in DMEM supplemented with 10% FCS, penicillin (100 U/ml), streptomycin (100 pg/ml), sodium pyruvate (1 mM), non-essential amino acids (0.1 mM) and L-glutamine (2 rnM). Cells were seeded on IOO-mm diameter plates for Northern blot analysis, or 96-well microtitre plates for ethoxyresorufin O-deethylase (EROD), and grown to 80% confluency. Then, 3-MC, dissolved in DMSO (final concentration < 0.5%) was added to the cultures for 72 hr, with a change of medium every 24 hr. EROD activity assay

After treatment, the medium was discarded and 25 ~1 TMB buffer (Tris-HCI 100 mM pH 7.8, MgClz 2.5 mM; bovine serum albumin 0.06%) containing 1% glycerol was added per well and the cultures were frozen at - 80°C. After thawing, TMB buffer (200 pi/well) containing glucose-6-phosphate (3 mM), NADP (0.5 mM), dicoumarol (10 PM), ethoxyresorufin (2 PM) and glucose-dphosphate dehydrogenase (0.1 U/ml) were added. The EROD activity was 37°C by spectrofluorimetry measured at (A,, = 600 nm, i.,, = 535 nm) by following the kinetics of appearance of resorufin from ethoxyresorufin. mRNA analysis

The cultures were washed twice with cold phosphate buffered saline and extracted with guanidium buffer (guanidium 4 M; sodium citrate 25 mM pH 7.4, sarcosylO.5% and /?-mercaptoethanol 0.1 M). Total RNA was isolated by the acidic phenol extraction procedure (Chomczynski and Sacchi, 1987). 20 pg RNA were size-fractionated on a 0.9% agarose gel containing 10% formaldehyde and transferred to a nitrocellulose membrane. Hybridization was performed with 480 bp cDNA insert of human CYPIAI mRNA corresponding to nucleotides + 310 to + 790 and labelled using ‘*P Biolabs kit. RESULTS AND DISCUSSION

In this study, human hepatocytes and HepG2 hepatoma cells were used as positive controls for the induction in liver cells. For keratinocytes, we had the choice between different cell types. First, we chose HaCaT cells (Boukamp et al., 1988) which are spontaneously immortalized human keratinocytes; they have kept their full capability to differentiate terminally into corneocytes when calcium is shifted from low (0.15 mr+i)to high (1.5 mM) concentrations.

CYPl Al induction in human cells

34-K CONCENTRATIONS

445

Q.IM)

Fig. 1. Dose-dependent increase of EROD activity in human hepatocytes. Cells were treated with DMSO (0.5%), or with various concentrations of 3-MC (0.5-20 FM) for 72 hr. EROD activity was monitored by kinetic appearance of resorufin at 37’C.

Secondly, we used the HEW cell line (Banks-Schlegel and Howley, 1983); these cells were immortalized by transfection with the cDNA coding for the large T antigen of the SV40 virus; they have lost the capability to differentiate terminally. Nothing is known about the constitutive activity and the induction potential of CYPlAi by 3-MC in these two cell lines. Figure 1 shows that 3-MC induces EROD activity in human hepatocytes. This increase is dose-dependent with the maximal value at a concentration of 2.5 pM. At higher doses, 3-MC decreases EROD activity although, as revealed by the neutral red and the MTT tests, these concentrations were shown not

3-K

to be cytotoxic. 3-MC also induces EROD activity in a dose-dependent manner in HepG2 cells (Fig. 2). The concentration for the maximal induction value (10 pmol/min/mg protein) is 0.75 PM, which is threefold lower than that observed in hepatocytes. In HepG2 cells, non toxic high doses also decrease EROD activity as for hepatocytes. In this case, our results confirm those already reported in the literature. 3-MC is also capable of dose-dependently inducing EROD activity in human keratinocytes (Fig. 3). The maximal enzyme activity of about 1 pmot/min/mg protein is sixfold lower than the value obtained for hepatocytes (6 pmol/min/mg protein). Maximal

CONCENTRATIONS

(pm)

Fig. 2. Dose-dependent increase of EROD activity in HepG2 cell line. Cells were treated with DMSO (0.5%) or with various concentrations of 3-MC (0.05-10 PM) for 72 hr. EROD activity was monitored by kinetic appearance of resorufin at 37’C.

C.

446

0

Delescluse er ol

0.1

0.5

3-W

1.0

5.0

10.0

CONCENTRATIONS (CIM)

Fig. 3. Dose-dependent increase of EROD activity in human keratinocytes. Cells were treated with DMSO (0.5%). or with various concentrations of 3-MC (0.1-10 p(M) for 72 hr. EROD activity was monitored by kinetic appearance of resorufin at 37,C.

concentration values are between 0.1 and 0.5 PM which, again, are fivefold lower than those observed

in hepatocytes. Our data are in agreement with the results of Reiners et al. (1990). However, these authors showed a difference of 2000-fold between hepatocyte and keratinocyte EROD activities, instead of sixfold as in our experiments. This difference could be explained by the interspecies variability, since mice, but not human, keratinocytes were used. As for the two cell types of hepatic origin, non-toxic high doses of 3-MC also inhibit EROD activity.

Figure 4 shows that 3-MC induces EROD activity in the spontaneously immortalized HaCaT cell line in a dose-dependent manner. The maximal activity and inducing concentration are similar to those observed in HepG2 (around 10 pmol/min/mg protein and 0.75 pM, respectively). Again, high doses that are not toxic per se inhibit EROD activity. In contrast, the SV40-immortalized HEW cell line does not show any basal EROD activity and 3-MC was unable to induce it. Analysis of CYPlA1/2 mRNA Northern blots agrees with the trends obtained for EROD values and

Fig. 4. Dose-dependent increase of EROD activity in HaCaT cell line. Cells were treated with DMSO (OS%), or with various concentrations of 3-MC (0.05IO PM) for 72 hr. EROD activity was monitored by kinetic appearance of resorufin at 37°C.

HEPATOCYTE

12345

HEPGZ

‘-CYPIAI

KERATINOCYTE

123Q56 .

..

-CY?lAl

HACAT

”

-ACTIN

123456

-CYPlAl

HESV

123

Plate 1. Induction of CYPlA1/2 mRNA in cultures of human hepatocytes, HepG2, keratinocytes, HaCaT and HEW. Cells were treated with either DMSO or various concentrations of chemicals. 2CL30 yg total RNA was analysed and CYPlA1/2 mRNA was revealed with a radiolabelled CYPlAlj2 probe as described in Materials and Methods. Chemical concentrations: (hepatocytes) lanes 1, 2, 3, 4, 5: 0.05, 0.1, 0.5, 1, 2 PM 3-MC; (HepG2) lanes 1,2, 3,4, 5, 6, 7: control, 0.5,0.75, 1, 2, 5, 10 PM 3-MC; (keratinocytes) lanes 1, 2, 3, 4, 5, 6: control, 0.1, 0.5, 1, 5, 10 PM 3-MC; (HaCaT) lanes 1, 2, 3, 4, 5, 6: control, 0.5, 1, 2, 5, 10 PM 3-MC; (HESV) lanes 1, 2: control, 1 PM 3-MC; lane 3: 5 ug cycloheximide/ml.

447

CYPIAI induction in human cells

demonstrates that 3-MC provokes a dose-dependent accumulation of CYPlAl mRNA (Plate 1) in hepatocytes, HepG2, keratinocytes and HaCaT, but not in HESV. We have shown, as others (Wang et al., 1995) that the protein synthesis inhibitor cycloheximide was able to induce CYPlAl mRNA. In order to check whether the CYPIAI gene was repressed or absent in this cell line, we treated these cells with this compound. Plate I shows that cycloheximide did not increase CYPIAI mRNA, suggesting that the gene is not present or not functional. However, another possibility may exist, which is the lack or the non-functionality of the Ah receptor. Whatever the explanation, it seems that the process of immortalization by the SV40 large T antigen has disturbed the cascade of events leading to CYPIAI induction and also the capability of these cells to differentiate terminally. This correlates well with the fact that in keratinocytes, the process of differentiation is accompanied by an induction of CYPI Al. Plate I also reveals that 3-MC induces CYPIAI as well CYPlA2 genes in hepatocytes, but only CYPl Al in keratinocytes and the two cell lines HepG2 and HaCaT, owing to the lack of expression of the corresponding gene in these cells. Finally the highest 3-MC concentrations do not decrease CYPI Al mRNA (Plate I). These data are in contrast to those obtained for EROD activity suggesting an effect of 3-MC either at the translational level, or an enzyme inhibition by an excess of substrate. In conclusion, our results confirm that HepG2 may be a substitute for hepatocytes. Moreover, they demonstrate that the spontaneously immortalized cell line HaCaT may constitute a valuable tool for studying epidermal CYPI Al gene regulation by xenobiotics.

449

enzyme activities during the growth of human HepG2 hepatoma cells. Xenobiotica 20, 43Wl.

Doostdar H., Grant M. H., Melvin W. T.. Wolf C. R. and Burke M. D. (1993) The effects of inducing agents on cytochrome P450 and UDP-glucuronyltransferase activities in human HepG2 hepatoma cells. Biochemical Pharmacology 46, 6i9-635.

_

Duthie S. J.. Coleman C. S. and Grant M. H. (1988) Status of reduced glutathion in the human hepatdma &II line,

HepG2. Biochemical Pharmacology 37, 3365-3368. Duthie S. J. and Grant M. H. (1989) The role of reductive and oxidative metabolism in the toxicity of mitoxantrone, adriamycin and menadione in human liver derived HepG2 hepatoma cells. British Journal of Cancer 60, 56571.

Fabre G., Rahmani R., Placidi M.. Combalbert J.. Covo J.. Cano J. P., Coulange C., Ducros M. and Rampal M. (1988) Characterisation of midazolam metabolism using human hepatic microsomal fractions and hepatocytes in suspension obtained by perfusing whole human liver. Biochemical Pharmacology 31, 4389-4397.

Grant M. H., Burke M. D.. Hawsworth G. M., Duthie S. J., Engeset J. and Petrie J. C. (1987) Human adult hepatocytes in primary monolayer culture: maintenance of mixed function oxidase and conjugation pathways of drug metabolism. Biochemical Pharmacology 36, 231 l2316. Huh N.. Nemoto N. and Utakoji T. (1982) Metabolic activation of benzo(a)pyrene, aflatoxin B2 and dimethylnitroamine by a human hepatoma cell line. Muration Research 94, 339-348.

Kocarek T. A., Schuetz E. G. and Guzelian P. S. (1993) Transient induction of cytochrome P450 IAI mRNA by culture medium component in primary cultures of adult rat hepatocytes. In Vitro Cell Deoelopment and Biology 29A, 62-66. Lacarelle B., Marre F., Durand A., Davi H. and Rahmani R. (1991) Metabolism of Minaprine in human and animal hepatocytes and liver microsomes. Prediction of metabolism in vivo. Xenobiotica 21, 317-329. Marre F.. Sanderink G., de Sousa G., Gaillard C., Martinet M. and Rahmani R. (1996) Hepatic biotransformation of Docetaxel (Taxotere’g) in ciwo: involvement of the CYP3A subfamily in man. Cancer Research 56, 129&1302.

Merk H., Jugert F. and Frankenberg S. (1993) KeratiREFERENCES

Banks-Schlegel S. P. and Howley P. M. (1983) Differentiation of human epidermal cells transformed by SV40. Journal of Cell Biology 96, 330-331.

Boukamp P.. Petrussevska R. T., Breitkreutz D.. Hornung J., Markham A. and Fusenig N. (1988) Normal keratinization in a spontaneously immortalized aneuploid human keratinocyte cell line. Journal of Ce// Biology 106, 761-771. Chomczynski P. and Sacchi N. (1987) Single-step method of RNA isolation by acid guanidium thiocyanate-phenolchloroform extraction. Analytical Biochemisrry 162, 156159. de Sousa G., Dou M., Barbe D.. Lacarelle B.. Placidi M. and Rahmani R. (1991) Freshly isolated or cryopreserved human hepatocytes in primary-culture: influence of drug metabolism on hepatotoxicity. Toxicology in Vitro 5, 483-486.

de Sousa G.. Nicolas F., Valles B.. Coassolo P. and Rahmani R. (1995) Relationships between in cifro and in oioo hepatic biotransformation of drugs in humans and animals: pharmaco-toxicological consequences. Cell Biology and Toxicology II, 147-I 53.

Doostdar H., Demoz A., Burke M. D., Melvin W. T. and Grant M. H. (1990) Variation in drug-metabolizing

nocytes and reconstructed epidermis for in cirro pharmacological and toxicological testing. In Human cells in In Vitro Pharmaco-toxicology: Present Status within Europe. Edited by V. Rogiers, W. Sonck. E. Shephard

and A. Vercruysse. pp. 4%59. Vubpress. Brussels. Nicolas F., de Sousa G.. Valles B.. Coassolo P. and Rahmani R. (1995) Comparative metabolism of AZT (3’-Azido-3’Deoxythymidine) in cultured hepatocytes from rat, dog, monkey and man. Drug Metabolism and Disposition 23, 308-3 13. Okey A. B. (1990) Enzyme induction in the cytochrome P450 system. Pharmacology and Therapeutics 45.241-298. Porter T. D. and Coon M. J. (1991) Cytochrome P450, multiplicity of isoforms. substrates and catalytic and regulatory mechanisms. Journal of Biological Chemistry 45, 13469-13472.

Raffali F.. Rougier A. and Roguet R. (1994) Measurement and modulation of cytochrome-P450-dependent enzyme activity in cultured human keratinocytes. Skin Pharmacology 7, 345-354.

Rahmani R., de Sousa G.. Marre F.. Nicolas F. and Placidi M. (1993) Potential of freshly isolated and cryopreserved human hepatocytes in drug research and development. In Human cells in In Vitro Pharmaco-toxicology: Present Starus within Europe. Edited by V. Rogiers. W. Sonck, E.

Shephard and A. Vercruysse. pp. 117-138. Vubpress. Brussels.

4.50

C. Delescluse et ol

Reiners J. J., Cantu A. R., Pavone A., Smith S. C., Gardner C. R. and Laskin D. L. (1990) Fluorescence assay for per-cell estimation of cytochrome P-450-dependent monooxygenase activities in keratinocytes suspensions and cultures. Analytical Biochemisrry 188, 31 l-324. Sadek C. M. and Allen-Hoffmann B. L. (1994) Cytochrome P450lAl is rapidly induced in normal human keratinocytes in the absence of xenobiotics. Journal q/ Biological Chemistry 269, 16067-l 6074. Valles B., Coassolo P., de Sousa G., Aubert C. and Rahmani R. (1993) In citro hepatic biotransformation of moclobemide Ro I l-l 163 in human and rat. Xenohiotica 23, 1101-1111.

Valles B., Schiller C. D., Coassolo P., de Sousa G., Wyss R., Jaek D., Viger-Chougnet A. and Rahmani R. (1995) Metabolism of Mofarotene in hepatocytes and liver microsomes from different species. Comparison with in viuo data and evaluation of the cytochrome P450 isoenzymes involved in human biotransformation. Drug Metabolism and Disposition 23, 1051-1057.

Wang X., Thomsen J. S., Santostefano M., Rosengren R., Safe S. and Perdew G. H. (1995) Comparative properties of the nuclear aryl hydrocarbon (Ah) receptor complex from several human cell lines. European Journal of Pharmacology

293, 191-205.