Expression and inducibility of UDP-glucuronosyltransferases 1-naphthol in human cultured hepatocytes and hepatocarcinoma cell lines

Life Sciems, Vol. 60,No. 2.2,pp. 19451951,1997 copyright0 1997J%mier .SckaccInc. Printed in the USA. All ri&htsmerwd am-3205/97 517.00t .oo ELSEVIER

PII SOO24.3205(97)00159-S

EXPRESSION AND INDUCIBILITY OF UDP-GLUCURONOSYLTRANSFERASES l-NAPHTHOL IN HUMAN CULTURED HEPATOCYTES AND HEPATOCARCINOMA CELL LINES

Amr Abid’ , Nicole Sabolovic* , Jacques Magdalou’ Centre du Mtdicament, URA CNRS 597.30 rue Lionnois, 54000 Nancy, France (Received in final form February 20,1997)

Summaq The UDP-glucuronosyltransferase (UGTs) isoforms involved in the conjugation of 1-naphthol were characterized in human cultured hepatocytes and in two human hepatocarcinoma cell lines, KYN-2 and Mz-Hepl in terms of expression, kinetics and induction by drugs. Their properties were compared to those of UGTI*6 stably expressed in the V79 cell line (V79UGT1*6), which glucuronidates I-naphthol preferentially. The determination of kinetic constants for glucuronidation of 1-naphthol revealed a two-site model in human hepatocytes, but a one-site model in the two hepatocarcinoma cell lines. Southern blot analysis of RT-PCR products, showed that the UGT1*6 mRNA was expressed in KYN-2, but not in Mz-Hep-1 cells. However, a mRNA encoding a UGT different from UGTl*6 was expressed in Mz-Hep-1 cells. The two inducers, l3-naphthoflavone and rifampicin exerted a differential effect, depending on the cell lines considered. Altogether, the results suggest that, in hepatocytes, two UGT isoforms, which glucuronidate 1-naphthol are expressed and are differentialy regulated by inducers. Both KYN-2 and Mz-Hep-1 cells express one of the two different UGT isoforms found in hepatocytes. The UGT isoform present in KYN-2 cells corresponds to UGT1*6, whereas in Mz-Hepl cells the UGT isoform present was different from UGTl*6 and UGT1*7. KeyWords: UDP-gIucuronosykransferase, induction, expression, hepatocytes, hepatoma cells

Human cell lines are useful systems to investigate the metabolism of drugs and xenobiotics, and also to study the regulation of the enzymes which are in charge of their biotransformation. Among them, cultured hepatocytes represent an attractive cellular model to gain information on the fate of drugs in human liver. However, the limited availability of human hepatocytes, for ethical reasons, has impaired their use as in vitro models. Despite the fact that they do not reflect the interindividual variability observed in human, the hepatoma cell lines are generally easier to handle and present a stable expression of drug metabolizing enzymes. Two human hepatocarcinoma cell lines KYN-2 (1) and Mz-Hep-1 (2) have been recently established. They exhibited some hepatic functions such as secretion of albumin, transferrin and ferritin, and were recently shown to present hydroxylation and glucuronidation reactions (3). UDP-glucuronosyltransferases (UGTs, EC 2.4.1.17) belong to a multigenic family of enzymes that catalyze the transfer of glucuronic acid from UDP-glucuronic acid to structurally unrelated substances resulting in the formation of water-soluble B-(D)-glucuronides (4). Carcinogens chemically related to benzo(a)pyrene, the drug paracetamol and phenolic molecules, such as 1-naphthol, are substrates of the human UGT1*6 or UGT1*7 isoform (56). These proteins share * PresentAddress.:Laboratoire de Phammcologie, URA CNRS 1288, FacuIt6 de Mtdecine, Avenue de la For& de Haye, 54505 Vandoeuvre les Nancy, France.

1944

UGT Isofonas in Hepatocytes

Vol. 60, No. 22,1!397

the same C-terminal end encoded by 4 common exons (exons 2-5), which correspond to the UDP-glucuronic acid binding site domain, whereas the N-terminal end, encoded by a different exon 1 is supposed to interact with the aglycon and therefore confers the substrate specificity to the enzyme (7). On the other hand, UGT isoforms are differentially induced by drugs in rat, but little is known about inducibility and regulation of UGT isoforms in man, when compared to cytochrome P450 (8). Using antibodies raised against the N-terminal part of UGTI*6, we have previously shown that this isoform contributed up to 50% for the glucuronidation of I-naphthol in human liver (9). From a toxicological point of view, it is important to determine which other isoforms are also implicated in the glucuronidation of this compound and other structurally related molecules. The aim of this work was to characterize the UGTs which glucuronidate 1-naphthol using cultured human hepatocytes and hepatoma cell lines, KYN-2 and Mz-Hep-1. The characterization was performed in terms of expression, inducibility by drugs and kinetic constants determination. Methods Preparation of human hepatocytes. The human liver samples used in this study were obtained in

accordance with the French regulations concerning the tissue transplantations. Liver samples (lobeetomy) were obtained from patients with a secondary hepatic tumor. Human liver samples (I, II, III and IV) were a generous gift of Dr P. Maurel (INSERM U 128, Montpellier, France), whereas liver sample V, was from Centre Alexis Vautrin (Vandoeuvre 12s Nancy, France). The hepatocytes were prepared from peritumoral tissues by a two-steps collagenase perfusion method, as described (10). The yields averaged 10’ cells, with a viability higher than 92 %, as determined by the Trypan blue exclusion test. Gel? culture and treatments. Hepatoma cells, KYN-2 and Mz-Hep- 1 were kindly donated by Dr M. Kojiro (Kurume University, School of Medicine, Kurume-shi, Japan) and Dr W.G. Dippold (Johannes Gutenberg-Unversitit, Mainz, Germany), respectively. Hepatoma and hepatocytes were cultured in Dulbecco’s modified Eagle medium (Gibco-BRL, France) supplemented as described previously (11). The cultures were maintained at 37’C in a humidified atmosphere containing 5% CO,. B-Naphthoflavone (Sigma, France) and rifampicin (Sigma) were dissolved in dimethylsulfoxide (Sigma) and added to the medium 24 h after plating the cells. In control experiments, the cultures received the same amount of solvent. The medium and inducers were renewed every 24 h. After 72 h of treatment, the cells were harvested by scraping in ice-cold phosphate-buffered saline (pH 7.4) and stored at -80°C until required. V79 cells stably expressing UGTl*6 were cultured as previously described (12). They were kindly donated by Dr. S. FournelGigleux (Nancy, France). Microsomes from hepatocytes were prepared according to Dragacci et al. (13). Enzymatic assay. I-Naphthol glucuronidation activity was carried out with 0.1 mg of cellular

homogenates of hepatoma cells or with native hepatocyte microsomes, and was measured using l-[14C]naphthol (100,000 dpm; 9,4 mCi/mmol; Sigma) (14). The incubation was carried out in 100 mM Tris-HCl (pH 7.4) containing 5 mM MgCl,, and 0.5 mM I-naphthol (Sigma) and 4 mM UDP-glucuronic acid (Boehringer, Germany). The assay was incubated for 15 min at 37°C. The protein content was determined according to Bradford (15), with bovine serum albumin as a standard. Kinetic constunts determination. Apparent kinetic constants (Km and V_> were determined in

cellular homogenates using linear least-squares regression analysis of double-reciprocal plots corresponding to, at least, 11 and 8 different concentrations of 1-naphthol and UDP-glucuronic acid, respectively. The concentration of 1-naphthol varied from 0.125 mM to 0.5 mM for a constant UDP-glucuronic acid concentration of 4 mM. The UDP-glucuronic acid concentration then varied from 0.01 mM to 4 mM for a constant concentration of 0.5 mM 1-naphthol. K,,, and V_ welE: expressed as mean + S.D. of 3 independent measurements. RT-PCR assay. Total RNA was extracted from cells by the guanidium thiocyanate-phenol-

chloroform method (16). cDNA was prepared from 10 pg of total RNA by reverse transcription at 42°C for 2 h using Superscript RNase H reverse transcriptase (8 U/l.& Gibco-BRL) in mixture

Vol. 60, No. 22, 1997

1945

UGT Isofortnsin Hepatocytes

containing 50 mM Tris-HCl buffer (pH 8.3), 75 mM KCl, 3 mM MgCl,, 2 mM dithiothreitol, 0.4 PM primers B or D and 0.5 mM dNTP. TABLE I gives the sequences and the localization on the UGTI *6 cDNA of the primers used for the RT-PCR. The reverse transcription products were used for PCR amplification in a solution containing 67 mM Tris-HCI (pH 8.8) 16.6 mM (NH&SO,, 1.5 mM MgCl,, 0.4 pM of each primers, 0.2 mM each dNTP and 0.01 % (v/v> Tween-20. Reaction mixtures were overlaid with mineral oil and heated at 94°C for 5 min, and then 0.05 U/ml of BioTaq polymerase (Eurobio, France) was added at 80°C. The samples were subjected to one cycle at 55’C for 2 min and 72°C for 40 min to synthesize the second cDNA strand. Amplification was performed for 40 cycles at 94’C for 1 min, 55°C for 1 min and 72’C for 2 min using a thermocycler (Perkin-Elmer Cetus Corp., USA). The final extension step at 72°C lasted for 10 min. The couple of primers used for PCR were A and B, A and E and C and D. Southern blots. RT-PCR products were separated in 1% (w/v) agarose gel and transferred onto Hybon-N’membrane (Amersham, Les Ulis, France). The membrane was hybridized to PstI fragment of human UGTZ*6 cDNA (17), labeled by random priming with [a-32P]dATP (3000 Ci/mmol; Amersham), for 18 h at 65’Cin 5X SSC (1X SSC = 0.15 M sodium chloride, 0.015 M sodium citrate), 5X Denhardt’s solution, 0.5 % (w/v) sodium dodecyl sulfate and 0.1 mg/ml of heat-denatured salmon sperm DNA. The membrane was then washed twice with 2X SSC, 0.1% (w/v) sodium dodecyl sulfate at room temperature and twice at 65’C in 0.1X SSC containing 0.1% (w/v) sodium dodecyl sulfate. Autoradiography was carried out by exposure to Kodak X-ray ftirn at -80°C with intensifying screens. Zmmunoblor analysis. Microsomal proteins were separated by 10% (w/v) polyacrylamide gel

electrophoresis containing sodium dodecyl sulfate according to Laemmli (18). The separated proteins were transferred onto Immobilon-P membrane (Millipore, France). Immunoblotting was carried out using sheep antibodies against the N-terminal part of human UGT1*6 (9). TABLE I Primers Used for the RT-PCR Analysis Localization on

Primers

Sequences

UGTl*6 cDNA

S-CGCGAAGGATITCTGCAGGGGTITTCTTCTTAGC

S-AGGATCCTCTCAATGGGTCITGG A

Exon 1

; C D

5’GCCATATGGCTTGCCT

Exon 5 Exon 1

E

CC’ITCGCTCA

5’-GACTCGAGT CGACAAGP 5’-GCGAAlTCTTGAGGACAGCTGATGCG

Polyadenylation site Exon 1

Catalytic properties of I-naphthol glucuronidation. The apparent kinetic constants K, and V,, for the glucuronidation of I-naphthol in hepatocytes, hepatoma cell lines and in V79 cells stably expressing the UGT1*6 (V79UGT1*6), as reference, have been determined and compared (TABLE II). In hepatocytes,

the kinetics were biphasic (Fig l), and allowed the determination

of

two K, and V,,,,,values corresponding to two classes of UGT with high and low affinity toward 1-naphthol. A difference of 5 and 11-fold was observed between the K, values measured for

I-naphthol and UDP-glucuronic acid, respectively. On the other hand, the glucuronidation of I-napthol in KYN-2 and Mz-Hep-1 cells followed simple Michaelis Menten kinetics, like in V79UGT1*6 cells. The apparent K,,, values for 1-naphthol and UDP-glucuronic acid determined in KYN-2 cells were similar to those determined in V79UGT1*6 cells but differed from those determined in Mz-Hep-1 cells. Finally, the K, values for 1-naphthol and UDP-glucuronic acid determined in KYN-2 and Mz-Hep-1 cells were similar to those determined in human hepatocytes. Altogether, the results obtained with V79UGT1*6, KYN-2 and Mz-Hep-1 cells suggest that in hepatocytes, the UGT presenting the highest affinity toward 1-naphthol exhibited the lowest affinity toward UDP-glucuronic acid and vice versa.

UGT Isoforms in Hepatocytes

Vol. 60, No. 22, 1997

6-

-0.04

-0.02

0.00

0.02

0.04

0.06

0.08

0.10

l/[l-naphtholl Q.l.M-1) 60

50

40

l/v

30 20 10 0 -20

40

60

80

100

120

l/[UDP-GkAl @hI -1)

Fig. 1 Determination of kinetic constants for glucuronidation of 1-naphthol in human hepatocytes. The results presented are means of three independent determinations and show plots of reciprocal of the initial reaction velocity against the reciprocal of the I-naphthol A (or [UDP-GlcA], B) concentration. The line of best fit, as calculated by linear least squares regression, is also indicated. v is expressed as nmol/min x mg protein.

Vol. 60, No. 22, 1997

1947

UGT Isoforms in Hepatocytes

TABLE II Kinetics of 1-Naphthol Glucuronidation Kml Cells Human hepatocytes 1-naphthol UDP-GlcA*

K m2

Vmax2

nmoYmin per mg of proteins

NM 53fl 2880 f 130

V maxI

244*09 260 f 12

1.6kO.l 3.1 + 0.1

KYN-2 1-naphthol UDP-GlcA

293 f 23 291fO9

27.4 & 1.6 9.7 + 0.2

Mz-Hep-1 I-naphthol UDP-GlcA

52f2 2990 f 460

2.9 * 0.1 4.8 + 0.8

V79UGT1*6 1-naphthol UDP-GlcA

473 + 144 591+ 168

75.5 f 18.9 33.8 +_05.5

0.50 f 0.02 2.4 + 0.2

* UDP-GlcA: UDP-glucuronic acid

Analysis of the VGT conjugating I-naphthol by RT-PCR method. The RT-PCR method was used

to characterize the expression of UGT 1-naphthol isoforms in KYN-2 and Mz-Hepl hepatoma cells. This strategy was based on known sequences of VGTI *6 cDNA (16). Primers A and B were used to amplify a 1586 bp-fragment from the coding sequence of VGTl*6 cDNA. The RT-PCR products were analyzed by Southern blotting with the labeled PsrI fragment of VGTl*6 cDNA which corresponds to exon 1. A single band of the predicted size was observed in V79UGT1*6 and in KYN-2 cells (Fig 2A). This result indicates that the I-naphthol UGT isoform expressed in KYN-2 cells was UGT1*6. By contrast, no band could be detected in Mz-Hepl, indicating that the UGT1*6 mRNA was not expressed in this cell line. Primers A and E were used to amplify a 701 bp-fragment corresponding to the exon 1 coding sequence of VGTl*6. Fig 2B shows the Southern blot developed with the PstI fragment of VGTI *6 cDNA as probe. A single band of the predicted size was observed in all samples. This result suggested that the mRNA encoding UGT 1-naphthol in Mz-Hepl had an exon 1 similar to that of UGT1*6 mRNA. Primers D and C were used to amplify a 2200 bp fragment of UGTI *6 cDNA. A southern blot developed with the PstI probe showed that in V79UGT1*6 samples, a single band with the correct size was obtained, whereas in Mz-Hep-1 cells one band with a size of 1600 bp was detected. This result shows that the 1-naphthol UGT in the Mz-Hep-1 cell line was different from UGT1*6 (Fig 3). To confirm these results, a Western blot was performed using antibodies developed against the N-terminal part of human UGT1*6 encoded by exon 1. Immunoblot analyzes revealed the presence of a protein in all samples, with a molecular weight (53-55 kDa) similar to that of UGT1*6 (Fig 4). Effects of&naphthoflavone

and n$ampicin on I-naphthol glucaronidation. Inducers were used in an attempt to further characterize the UGT isoforms present in the hepatocarcinoma cell lines. KYN-2 and Mz-Hepl cells were treated with B-naphthoflavone and rifampicin at 3 concentrations, which varied from 50 ~JMto 100 pM. In KYN-2 cells, I-naphthol glucuronidation was significantly induced by 75 l.tM 8-naphthoflavone (Fig 5A). However, in Mz-Hep-1 cells, I%naphthoflavone treatment failed to stimulate the glucuronidation of 1-naphthol (Fig 5B). On the other hand, in Mz-Hepl cells this activity was increased by rifampicin treatment at 100 pM (Fig 5B). The effects of R-naphthoflavone and rifampicin treatments were studied on cultured hepatocytes from different donors. A marked interindividual variability was observed. D-Naphthoflavone and rifampicin increased 1-naphthol glucuronidation activity in cultured hepatocytes from donors I and II, but this activity in hepatocytes from donor III was increased by rifampicin but not by &naphthoflavone treatment. In contrast, 1-naphthol glucuronidation activity in hepatocytes from donor IV was increased by l3-naphthoflavone treatment but not by rifampicin treatment. 1-Naphthol glucuronidation activity in cultured bepatocytes from donor V was not

1948

UGT Isoforms in Hepatocytes

Vol. 60, No. 22, 1!297

sensitive to l3-naphthoflavone treatment, whereas rifampicin treatment lowered this activity (Pig 5C).

A

B

Fig. 2 Southern blot analysis of RT-PCR obtained with primers A and B (A) and A and E (B). RT-PCR products were electrophoresed on 1% (w/v) gel agarose and transferred onto the membrane which was blotted with PstI fragment of human UGTZ*6 cDNA. Lane 1, Mz-Hep- 1; lane 2, KYN-2; lane 3, V79UGT1*6. The arrow indicates the UGTZ *6 cDNA.

23.1 9.41 6.55 4.34

2.32 2.02

kDa

I

2

3

4

Western blot analysis of expressed 1-naphthol UDP-glucuronosyltransferases. Immunoblot was developed with anti-human UGT1*6 antibodies. Lane 1, V79UGT1*6; lane 2, KYN-2; lane 3, Mz-Hep-1; lane 4, hepatocytes. Fig. 3 Southern blot analysis of RT-PCR products obtained with primers C and D. Lane 1, Mz-Hepl; lane 2, V79UGT1*6.

Vol. 60,No. 22,1!#7

UGT

Isofortnsin Hepatocytes

1949

The present work reports the characterization of UGTs involved in the glucuronidation of 1-naphthol and their induction by drugs. This study was performed with human hepatocytes and two hepatoma cell lines, KYN-2 and Mz-Hepl. V79 cells stably expressing human UGT1*6 were used as a reference. We have shown that the glucuronidation of 1-naphthol in human hepatocytes was carried out by two kinetically different UGT isoforms. This is based on the fact that the kinetic of I-naphthol glucuronidation is biphasic, and therefore two sets of K,,, were determined for 1-naphthol and UDP-glucuronic acid. A large difference (ratio of 11) was observed for K,,, values of UDP-glucuronic acid and not for those of 1-naphthol (ratio of 5). On the other hand, in human hepatoma cell lines KYN-2 and Mz-Hep-1, I-naphthol was glucuronidated by one UGT isoform. UGT 1-naphthol present in KYN-2 and Mz-Hepl were kinetically different. However, these UGTs were kinetically similar to those determined in human hepatocytes. One of these two UGTs (present in KYN-2 and in human hepatocytes) had a similar kinetic constants as those determined in our reference, V79UGT1*6 cells. In fact, these results suggest that the UGT isoform, which supports 1-naphthol glucuronidation in KYN-2 cells could be UGT1*6, whereas in Mz-Hep-1 cells I-naphthol glucuronidation was supported by another isoform. Biphasic kinetics concerning the glucuronidation of 1-naphthol have been reported with human liver microsomes (19-20), but never in cultured cells. The two corresponding sets of apparent K,,, values toward 1-naphthol and UDP-glucuronic acid were in the range of those measured in this work. In order to identify the isoforms a RT-PCR strategy was used to characterize the expressed mRNA of UGT 1-naphthol in KYN-2 and Mz-Hep- 1. By this method, we showed that the mRNA of UGT 1-naphthol expressed in KYN-2 cells is UGT1*6. However, in Mz-Hep-1 cells the mRNA expressed differed from that encoding UGTl*6. On the other hand, UGT 1-naphthol isoform expressed in Mz-Hep-I cells possessed an exon 1 similar to that of UGT1*6. These results were corroborated by Western blot analysis using antibodies raised against the N-terminal end of UGT1*6. It is known that UGT1*6 and UGT1*7 have a similar C-terminal part (UDP-glucuronic acid binding site), which is encoded by exons 2 to exon 5, but have different exon 1 (aglycone binding site). Therefore, the UGT 1-naphthol expressed in Mz-Hep-1 cells was not UGT1*7. This result was confirmed by the glucuronidation toward the marker substrate thymol, which is exclusively supported by UGTl*7 (6). The activity was very low (7 pmoYmin per mg proteins), whereas 1-naphthol glucuronidation activity was about 3 nmol/min per mg of proteins (results not shown). In addition, the apparent Km determined toward UDP-glucuronic acid in KYN-2 cells was radically different to that determined in Mz-Hep-I cells. This result further indicated that the glucuronidation of 1-naphthol in Mz-Hep-1 cells was not supported by UGT1*6 or by UGT1*7, the difference would be localized on exon-5. Cane et al. (10) reported that in human cultured hepatocytes, the glucuronidation of 1-naphthol was not affected by S-naphthoflavone treatment. On the other hand, Fabre et al. (21) observed an

inter-individual variation in the effects of g-naphthoflavone and rifampicin treatments on 1-naphthol glucuronidation in human hepatocytes. These results are in agreement with our data. Recently, we have shown that the UGT1*6 expression in human cultured hepatocytes was repressed by rifampicin treatment, whereas I-naphthol glucuronidation activity was not affected by this treatment (11). In fact, the induction of 1-naphthol glucuronidation activity by rifampicin in human hepatocytes was due to the induction of UGT1*6, but could be due to the induction of UGT 1-naphthol isoform characterized in Mz-Hep-1 cells. In addition, we have recently shown that the UGT I-naphthol isofotm present in KYN-2 cells (i.e. UGT1*6) was sensitive to the effect of DMSO, which is not the case of UGT 1-naphthol isoform present in Mz-Hep-1. In addition, the apparent absence of the concentration-dependent induction of I-naphthol glucuronidation activity by &naphthoflavone is due to the fact that this activity is lowered by DMSO, and therefore B-naphthoflavone have to overcome the negatif effect of DMSO (3). These results further support that human hepatocytes would contain two different isoforms which are differentially sensitive to the induction potency of B-naphtoflavone and rifampicin. The apparent V_ for 1-naphthol glucuronidation in cultured human hepatocytes supported by UGTl*6 was 12 times lower than that in the hepatocarcinoma cell line KYN-2. These results could be related with immunoblot analysis, and indicated that the high V_ determined in KYN-2

1950

UGT Isoformsin Hepatocytes

Vol. 60, No. 22, 1997

cells was associated with an elevated expression level of UGT1*6 protein. Bock et al. (5) also reported a persistent increase in activity and expression of UGT 1-naphthol in rat liver preneoplastic and neoplastic nodules. The increase in UGT 1-naphthol activity was concomitant with other alterations in the pattern of drug metabolizing enzymes, such as a decrease in the cytochrome P450 activities and an inctease in the activities of glutathione-S-transferases, epoxide hydrolase and y-glutamyltransferase (22). These results suggest that the increase in the glucuronidation level of 1-naphthol was associated with the increase in UGTI*6 expression in KYN-2.

it

Inducer concentrations @M)

Inducer concentrations (pM)

Samples

Fig. 5 Effects of B-naphthoflavone and rifampicin treatment on glucuronidation of I-naphthol. Cells were treated with 50,751 and 100 PM B-naphthoflavone ([7 ) and rifampicin (I)_ Controls were treated with DMSO (I). (A) KYN-2 cells; (B) Mz-Hep-1 cells. (C) Hepatocytes were treated with 50 pM B-naphthoflavone and rifampicin, and results are means of two determinations. For hepatoma cells, the results are expressed as mean f S.D. of triplicate measurements. * Significantly different from control (r-Student test, P < 0.05); ** significantly different from control (z-Student test, P < 0.01). In conclusion, using three human hepatic cellular models and V79 cells stably expressing UGTl*6 as a reference, we have demonstrated that 1-naphthol glucuronidation was supported by two UGT, UGTl*6 and another UGT different from UGTl*6 and UGT1*7. These proteins were

Vol. 60, No. 22, 1997

UGT Isoforms in Hepatocytes

1951

differentially regulated by B-naphthoflavone and rifampicin. We have shown, for the first time, that human UGT1*6 was inducible by O-naphthoflavone and was overexpressed in the human hepatocarcinoma cell line KYN-2.

We would like to thank Dr M. Kojiro (Kurume University, School of Medecine, Kurume-Shi, Japan) for KYN-2 cells, Dr W.G. Dippold (Department of Internal Medecine, Pathology, Genetic and Microbiology, Johannes Gutenberg-Universitat, Mainz, Germany) for Mz-Hep-1 cells and Dr P. Maurel (INSERM U 128, Montpellier, France). This work was supported in part by the Pale Europeen de Sante, Nancy, France.

76:

!: 10.

11. 12.

13. 14.

::* 17: 18. 19. 20. 21. 22.

H. YANO, M. MARUIWA, T. MURAKAMI, K. FUKUNORI, Y. ITO, S. SUGIHARA and M. KOJIRO, Acta Pathol. Jpn. 3 953-956 (1988). W.G. DIPPOLD, H.P. DIENES, A. KNUTI-I, W. SACHSSE, W. PRELLWITZ, D. BITIERSUERMANN and K.H. MEYER ZUM BUSCHENFELD, Hepatology 6,1112- 1119 (1985). A. ABID, J MAGDALOU and N. SABOLOVIC, Cell. Biol. Toxicol.,u 115-125 (1996). J. MAGDALOU, Y. HOCHMAN and D. ZAKIM, J. Biol. Chem. a 13624-13629 (1982). K.W. BOCK, A. FORSTER, H. GSCHAIDMEIER, M. BRUCK, P. MUNZEL, W. SCHARECK, S. FOURNEL-GIGLEUX and B. BURCHELL, Biochem. Pharmacol. ti 18091814 (1993). T. EBNER and B. BURCHELL, Drug Metab. Dispos. 2L 50-55 (1993). B. BURCHELL, D.W. NEBERT, D.R. NELSON, BOCK K.W., T. IYANNAGI, P.L.M. JANSEN, D. LANCET, G.J. MULDER, J.R. CHOWDHURY, G. SIEST, T.R. TEPHLY and PI. MACKENZIE, DNA Cell Biol. UT 487-494 (1991). C.H. BRIERLEY and B. BURCHELL, BioAssays. fi 749-754 (1993). M. OUZZINE, T. PILLOT, S. FOURNEL-GIGLEUX, 3. MAGDALOU, B. BURCHELL and G. SIEST, Arch. B&hem. Biophys. m 196-204 (1994 ). J.P. CANO, G. FABRE, P. MAUREL, N. BICHET, Y. BERGER and P. VIC, w . . of Ge, G. Siest, J. Magdalou, and B. Burchell (Eds) 249-260, INSERM/John Libbey Eurotext Ltd, London (1988). A. ABID, N. SABOLOVIC, A.M. BAIT, G. SIEST, P. MAUREL, F. GUILLEMIN and J. MAGDALOU, Cell. Pharmacol. _l.257-262 (1994). E. BATTAGLIA, M. PRITCHARD, M. OUZZINE, S. FOURNEL-GIGLEUX, A. RADOMINSKA, G. SIEST and J. MAGDALOU, Arch. Biochem. Biophys. 3_a9 266-272 (1994). S. DRAGACCI, J. THOMASSIN, J. MAGDALOU, H. SOUHAILI-EL-AMRI, P. BOISSEL and G. SIEST, Eur. J. Clin. Pharmacol. z 485-491 (1987). G. OTANI, M.M. ABOU-EL-MAKAREM and K.W. BOCK, Biochem. Pharmacol. z 12931297 (1976). M. BRADFORD, Anal. B&hem. 22 248-254 (1976). P. CHOMCZYNSKI and N. SACCHI, Anal. Biochem. 162 156-159 (1987). D.S. HARDING, S. FOURNEL-GIGLEUX, M.R. JACKSON and B. BURCHELL, Proc. Natl. Acad. Sci. USA 85 838 l-8385 (1988). U.K. LAEMMLI, Nature 222 680-685 (1970). 3.0. MINERS, K.J. LILLYWHITE, A.P. MATTHEWS, M.E. JONES and D.J. BIRKETI, Biochem. Pharmacol. z 656-67 1 (1988). K.W. BOCK, G. BURNNER, H. HOENSCH, E. HUBER and D. JOSTING, Eur. J. Clin. Pharmac. &I 367-373 (1978). G. FABRE, M. BOURRIE, E. MARTI, I. FABRE, C. ROQUE, B. SAINT AUBERT, H. JOYEUX, P. MAUREL, Y. BERGER and J.P. CANO, B&hem. Pharmacol. (Life Sci. Adv.) e79-91(1990). E. FARBER, Z.Y. CHEN, L. HARRIS, G. LEE, S. RINAUDO, W.M. ROOMI, J. ROTSTEIN and E. SEMPLE, Liver Cell Carcinoma, P. Bannasch, D. Keppler, and G. Weber (Eds), 273-291, Kluwer Academic Publishers, London (1989).