Eur. J. Biochem. 205, 847-852 (1992)

0FEBS 1992

Overexpression of the multidrug resistance gene product in adult rat hepatocytes during primary culture Olivier FARDEL ', Damrong RATANASAVANH Pascal LOYER', Brian KETTERER2 and Andre GUILLOUZO' I,

Unit6 de Recherchcs Hepatologiques, Institut National de la SantC et de la Recherche Medicale, Rennes, France Cancer Research Campaign, Molecular Toxicology Research Group, Biochemistry Department, University College London, England (Receivcd October 22/December 19, 1991) - EJB 91 1416

Expression of P-glycoprotein (P-gp), the product of multidrug resistance gene(s), was investigated in primary cultures of normal adult rat hepatocytes. Levels of P-gp mRNAs determined by Northern blotting and of P-gp measured by immunoblotting increased in parallel with time in culture. As in normal liver, P-gp was found to be localized on the membrane of bile canaliculus-like structures. This increased expression of P-gp was associated with decreased intracellular retention of doxorubicin, which could be restored by compounds such as verapamil and cyclosporin; doxorubicin (and also vincristine) was more cytotoxic to early than to late cultures. As in preneoplastic and neoplastic liver, overexpression of P-gp in cultured hepatocytes was associated with differential changes in drugmetabolizing enzymes, including increased glutathione S-transferase 7-7. Functional P-gp overexpression was observed in the absence of xenobiotic exposure or cell division; it could be linked to cellular stress occurring during cell isolation and plating. Increased expression of P-gp was blocked by actinomycin D, indicating its dependence on increased transcription, while cycloheximide led to a superinduction suggesting negative regulation by a protein factor.

Tumor cell lines which acquire resistance to a single anticancer drug often develop cross-resistance to various other structurally unrelated antitumor agents [I]. This multidrug resistance (mdr) is associated with increased expression of mdr gene(s) which encode a 130 - 200-kDa plasma membrane glycoprotein [2 -41. This transmembrane glycoprotein, designed P-gp, has been shown to be involved in energy-dependent drug efflux [5]. Various compounds, such as verapamil and cyclosporin, can inhibit P-gp-mediated drug efflux and thus reverse the mdr phenotype [6]. Similarly, P-gp is overexpressed in many human cancers [7] and is thought to contribute to clinical resistance to a number of cancer chemotherapeutic drugs. P-gp is also present in various normal tissues, including the small intestine, colon, pancreas, kidney, adrenal and liver [8]. In normal liver, the distribution of this glycoprotein is restricted to the biliary surface of hepatocytes [9] and recent studies have demonstrated that P-gp is responsible for ATPdependent drug transport by using canalicular plasma membrane vesicles prepared from rat liver [lo]. P-gp expression is increased in the liver after xenobiotic treatment [l11, during liver regeneration and in preneoplastic and neoplastic liver 1121. This overexpression has been interpreted as reflecting cell injury and/or cellular growth [13] and is associated with ___-

Correspondence to A. Guillouzo, INSERM U 49, HBpital Pontchaillou, F-35033 Rennes CCdex, France Abhreviutimr. GST, glutathione S-transfcrase; LDH, lactate dehydrogenase; mdr, multidrug resistance; NaCl/Pi, phosphatebuffered salinc; P-gp, P-glycoprotein. Enzyme. Glutathione S-transferase (EC 2.5.1.18).

differential changes in drug-metabolizing enzymes [14], characterized by decreased activity of cytochrome P-450 enzymes and increased expression of conjugating enzymes, including glutathione S-transferase P (GST 7-7) (EC 2.5.1.18) [15]. Similar changes in these drug-metabolizing enzymes occur in adult rat hepatocytes during primary culture [16]. This led us further to investigate the expression of P-gp using this in vitro model system. Our results show that P-gp expression is increased during culture in parallel with increased drug resistance.

MATERIALS AND METHODS Chemicals

Doxorubicin hydrochloride and vincristine sulfate were purchased from Roger Bellon (Neuilly, France). Verapamil, nifedipine, quinidine and amitriptyline were obtained from Sigma Chemical Company (St. Louis, MO), cyclosporin from Sandoz Ltd (Basel, Switzerland). Cell isolation and culture

Hepatocytes from adult male Sprague Dawley rats weighing 180-200 g were isolated by perfusion of the liver with a collagenase solution as previously described [17]. They were seeded at a density of lo5 cells/cm2 on plastic dishes in a medium consisting of 75% minimal essential medium and 25% medium 199, supplemented with 0.2 mg/ml bovine serum albumin, 10 &ml bovine insulin and 1O0h fetal calf serum;

848 4 h after cell seeding and daily thereafter a serum-free medium containing 0.1 pM dexamethasone was used. Isolation of RNA and blot analysis Total RNA was extracted from cultured hepatocytes by the guanidiniuin thiocyanate/cesium chloride method of Chirgwin et al. [I81 as modified by Raymondjean et al. [19]. For dot blotting, aliquots of total RNAs (2.5, 5 and 10 pg) were spotted onto Hybond N sheets (Amersham, Bucks, UK). For Northern blotting, 20 pg total RNAs were subjected to electrophoresis in a denaturing formaldehyde/agarose gel and transferred onto Hybond N sheets. RNA amounts in each lane and transfer efficiency were verified by staining the gel with ethidiuin bromide. The sheets were prehybridized and hybridized with 3'P-labeled probes. P-gp mRNAs were detccted with pCHP1. a hamster 660-base-pair cDNA probe [20], obtained from the American Type Culture Collection (Rockville, MD). The pCHPl sequence has been shown to be within the most highly conserved region of P-gp cDNAs [21]. GST 7-7 and albumin mRNAs were analysed with the rat GST 7-7 cDNA probe (pGSTr7) [22], and a rat albumin cDNA [23] respectively. After hybridization, the sheets were washed, dried and autoradiographed at - 80' C.

Evaluation of intracellular doxorubicin concentration The intracellular concentration of doxorubicin was estimated as described by Schott et al. [29] with slight modifications. Cultured hepatocytes were exposed to doxorubicin (10 pg/ml) for 2 h, with or without a modulating agent, namely verapamil, quinidine, nifedipine, amitriptyline or cyclosporin which are known to reverse the mdr phenotype [6]. The cells were then washed with NaCI/P,, harvested and ultrasonicated. These steps were performed quickly in order to avoid any drug efflux. Proteins were then precipitated with 20% trichloroacetic acid. The acid-soluble part was used to evaluate thc intracellular concentration of doxorubicin by fluorimetry using an excitation wavelength of 485 nm and emission wavelength of 590 nm. Preliminary controls showed no interference of doxorubicin fluorescence by reverting agents. No toxicity of any modulating agent or doxorubicin was apparent at the concentrations used over the incubation period. Evaluation of drug-induced cytotoxicity Hepatocytes were exposed for periods of 24 h to doxorubicin (1 pg/ml) or vincristine (150 pg/ml). Quantitative determination of cytotoxicity was assessed by determining the ratio of lactate dehydrogenase (LDH) activity in the culture medium to total (extracellular intracellular) LDH activity. LDH activity was measured using a LDH UVP kit system from Boehringer (Mannheim, FRG) using a Cobas Bioanalyser (Roche Analytical Instruments, Inc, Nutley, NJ).

+

Preparation of membranes and immunoblotting Crude membranes were prepared from cultured hepatocytes by differential centrifugation as described by Gerinann et al. [24]. Membrane proteins were estimated by the Bio-Rad protein assay [25], using bovine serum albumin as a standard. Proteins were beparated on a sodium dodecyl sulfate/polyacrylamide gel [26] and electrophoretically transferred to nitrocellulose sheets [27]. The concentration of polyacrylamide was 6%. Nitrocellulose sheets were blocked for 1 h with phosphate-buffered saline (NACI/Pi) (140 mM NaCI, 3 mM KCI, 9 mM Na2HP04, 1 mM KH,PO,) containing 3% bovine serum albumin, 10% fetal calf serum and 0.5% Tween 20. C219 monoclonal antibody (Centocor, Inc, Malvern, PA) raised against P-gp [28] was used as the primary antibody at a concentration of 1 pg/ml in the buffer used for blocking. Biotinylated antimouse IgG was used as the secondary antibody. The blot was developed using the Vectastain ABC kit (Vector Laboratory, Burlingame, CA) and 4-chloro-I -naphthol (Sigma) as the substrate. A control blot was performed using the same protocol with non-immune mouse myeloma cell ascites as the primary antibody. Cell extracts from sensitive and multidrug-resistant cell clones isolated froin the rat HTC hepatoma cell line [29] (a gift from Dr. J . Robert, Fondation Bergonii., Bordeaux, France), were used as negative and positive controls of P-gp expression.

Statistical analysis The results of doxorubicin accumulation and drug-induced cytotoxicity studies were analysed by the Student's t test. The criterion of significance of the differences between the means (fstandard deviation) was P < 0.05.

RESULTS Determination of P-gp mRNA levels

Rat hepatocytes were harvested 4 h and 1, 2, 3 and 4 days after seeding. Northern blotting showed that in intact liver and in freshly isolated hepatocytes P-gp mRNAs were barely detectable (Fig. 1). They were increased in cultured hepatocytes, as soon as 4 h after seeding. Higher levels of P-gp mRNAs were reached after 48 - 96 h of culture. To assess phenotypic changes occurring during culture, mRNAs corresponding to GST 7-7, the GST form expressed in fetal liver and hepatoma [15, 301 and to albumin, a specific marker of adult parenchymal cells, were also quantified. Only in 72 - 96-h cultures were the concentrations of mRNAs found to be markedly altered. At that time, GST 7-7 mRNAs became clearly detectable and albumin mRNAs were markedly deFixation and immunocytochemistry creased (Fig. 1). Hepatocyte cultures were briefly washed with NaCI/P, for P-gp mRNAs were also quantified by dot blotting after 1 min, then fixed in 1 % acetic acid/95% ethanol (by vol.) for 4-h exposure of cultures to 3 pg/ml actinomycin D or 60 pg/ml 15 inin at 4 C. After incubation with NaCI/P, containing 10% cycloheximide (Fig. 2). During the first 4 h of culture, addifetal calf serum. C219 (10 pg/ml) was applied for 1 h. After tion of actinomycin D suppressed induction while that of washing, peroxidase-conjugated antimouse IgG (Institut cycloheximide led to a superinduction of P-gp mRNAs. A Pasteur, Paris) was used as the secondary antibody. Peroxidase similar effect of cycloheximide on P-gp mRNAs was also activity was visualized with 3,3'-diaminobenzidine/Hz02. observed after 2 and 4 days of culture. By contrast, both Controls were performed with non-immune mouse myeloma actinomycin D and cycloheximide barely affected GST 7-7 cell ascites. and albumin mRNAs (Fig. 2).

849

Fig. 1. Expression of P-glycoprotein (P-gp), glutathione S-transferase 7-7 ( G S I 7-7) and albumin (Alb) mRNAs in cultured rat hepatocytcs. Each well contains 20 pg total RNAs isolated from rat liver (lane l ) , rreshly isolatcd hepatocytes (lane 2) and cultured hepatocytes harvested 4 h (lane 3). 24 h (lane 4), 48 h (lane 5), 72 h (lane 6) and 96 h (lane 7) after seeding. RNAs wcre transferred to Hybond N sheets afterelcctrophoresis and hybridized with a mdr probe (pCHP1). a GST 7-7 probe (pGSTr7) and a rat albumin probe. Transcript sizes wcre estimated relatively to the migration of 18 S and 28 S rRNA.

Immunoblotting and immunolocalisation of P-gp in cultured hepatocytes Crude membrane extracts were prepared from intact rat liver, freshly isolated hepatocytes and hepatocytes cultivated for 2 and 4 days and were used to investigate P-gp expression by immunoblotting (Fig. 3). P-gp was detected in multidrug-resistant HTC cells as a 140-kDa band but no corresponding band was visualized in thcir sensitive parental counterparts nor in rat h e r or freshly isolated hepatocytes. However, after 2 days of culture, hepatocytcs showed a protein of similar molecular mass that reacted with C219 antibodies and increased in amount after 4 days of culture. In addition, in both resistant HTC cells and cultured normal hepatocytes, two additional bands were observed but comparison with controls showed that they were non-specific. To determine the distribution of P-gp in the plasma membrane domains of cultured hepatocytes the glycoprotein was localized in 4-day cultures by using the indirect immunoperoxidase method. As observed in liver tissue sections, immunostaining was preferentially distributed around bile canaliculus-like structures (Fig. 4). Doxorubicin retention and action of chemical modulators The intracellular accumulation of doxorubicin was measured l, 2, 3 and 4 days after hepatocyte seeding. As shown in Fig. 5, there was a decrease in doxorubicin retention in cultured rat hepatocytes, which became more pronounced as [he culture aged. Intracellular doxorubicin content in 4-day cultures reprcsented only 32% of the value found in I-day cultures ( P < 0.05).

Fig.2. Effects of actinomycin D and cycloheximide treatment on Pglycoprotcin (P-gp), glutathionc S-transferase 7-7 (GST 7-7) and albumin (Alb) mRNAs in cultured rat hepatocytes. Each well contains an aliquot of total RNAs (10, 5 or 2.5 pg) isolated from rat liver (I), freshly isolatcd hepatocytes (2) and cultured hepatocytcs harvcstcd 4 h (3), 48 h (4) and 96 h ( 5 ) al-ter seeding. Cultured cells were either untreated (UT) or treated for 4 h by 3 Fg/ml actinomycin D (AC) or 60 pg/ml cycloheximide (CX) before harvesting. Dot-blot sheets were then hybridized with a rndr probe (pCHP1). a CST 7-7 prohe (pGSTr7) and a rat albumin probe. Amounts of total RNAs (pg) spotted for each line are indicated at the right of the figure.

Fig. 3. Western blot analysis of membrane proteins obtained from cultured hepatocytes. Crude membrane fractions were prepared from rat liver (lane I), freshly isolated hepatocytes (lane 2 ) and cultured hepatocytcs harvested 2 days (lane 3) and 4 days (lane 4) after seeding. 150 pg membrane proteins/lane were separated on SDS/PAGE and transferred to a nitrocellulose sheet. After incubation with C219 antibody (A), blots were developed as described in Materials and Methods. The multidrug-resistant HTC hepatoma cell liric (lane 6) and its parental sensitive line (lane 5) were used as positive and negative controls for P-gp expression. Non-specific C219 reactive bands were detected by using non-immune mouse myeloma cell ascites as primary antibody (B). Arrows indicate the position of P-gp. The position oTmolccular size standards (in kDa) is indicated between the two photographs.

To verify that decreased doxorubicin retention was related to P-gp-mediated drug efflux, the effect of 25 pM verapamil on doxorubicin accumulation into hepatocytes was studied 1, 2, 3 and 4 days after cell seeding (Fig. 5). At day 1, verapamil did not significantly affect doxorubicin accumulation in cells. indicating that P-gp-mediated drug efflux was low. At day 2

8 50

dox dox + vr

7-

-.4

3

2

DAY'S Fig. 5. Intracellular doxorubicin concentration and effect of verapamil in cultured rat hepatocytes. Cultured hepatocytes were incubated with 10 pg/ml doxorubicin alone (dox) or with 25 pM verapamil (dox + vr) during 2 h at 1, 2, 3 and 4 days after cell seeding. Intracellular doxorubicin concentration was then determined by the fluorimetric method as described in Materials and Methods. The values are the mean f SD of three experiments in quintuplicate; *P < 0.05.

Fig. 4. lmmunocytochemical localization of P-glycoprotein in 4-day cultured rat hepatocytes. Cell cultures were fixed in 1 % acetic acid/ 95% ethanol (by vol.) at 4-C and then incubated with mouse monoclonal P-gp antibody C219 (A) at a concentration of 10 pg/ml or with non-immune mousc myeloma cell ascites (B). Immunological reaction was then as described in Materials and Methods. Immunostaining spccifically located in bile canaliculus-like structures is indicated by arrows. Original magnification x 350. Bar = 20 pm.

it slightly, but significantly ( P < 0.05), increased doxorubicin retention into the cells. However at 4 days of culture, a striking effcct of verapamil resulting in 97% increase in doxorubicin accumulation over controls was observed ( P < 0.05). Other well known modulating agents of P-gp activity on doxorubicin efflux, namely cyclosporin, quinidine, nifedipine and amitriptyline, were tested 4 days after hepatocyte seeding (Fig. 6). At the dose tested (25 pM) cyclosporin had an effect quite similar to that of verapamil on the accumulation of doxorubicin in cultured rat liver parenchymal cells; quinidine and amitriptyline also had a significant although weaker effect (P< 0.05) while nifedipine was ineffective, even at higher concentration (50 pM) (data not shown). Cytotoxicity induced by doxorubicin and vincristine

Hepatocytes were exposed to doxorubicin and vincristine for 24 h after 24 h, 48 h or 72 h of culture. Cytotoxicity was assayed by the ratio released LDH/total LDH. Both compounds provoked an extensive release of the enzyme in the culture medium when added to 24-h hepatocyte cultures (Fig. 7). Cytotoxic effects were significantly decreased ( P < 0.05) when doxorubicin and vincristine were added to

, I

I

C

NF

AT

QD

VR

CY

Modulating agent Fig. 6. Effect of various modulating agents on intracellular concentration of doxorubicin in 4-day cultured hepatocytes. After 4 days of culturc, hcpatocytes werc incubated for 2 h with 10 pg/ml of doxorubicin alone ( C ) or with various modulating agents (25pM), namely nifedipine (NF), amitriptyline (AT), quinidine (QD), verapamil (VR) and cyclosporin (CY). The values are thc mean f SD of three cxperiments in quintuplicate; * P < 0.05.

48-h cultures and cytotoxicity was limited, if present at all, when the drugs were added at 72 h.

851 1.0 i

I

1

untreated dox

vcr

1

2

3

DAYS Fig. 7. Cytotoxic effect of doxorubicin and vincristine on cultured hepatocytes as a function of culturc time. Cultured hepatocytes werc exposcd to 1 pg/ml doxorubicin (dox) or 150 &ml vincristine (vcr) for 24 h. Drugs were added 24, 48 and 72 h after cell seeding. The ratio of relcascd LDH/total LDH indicated the drug cytotoxic effect. The values are thc mean f SD of three expcriments in quintuplicate; * P < 0.05.

DISCUSSION Previous studies have shown that the expression of drugmetabolizing enzymes is rapidly altered with time in rat hepatocytes in primary culture [I 61. The results reported here, using Northern and Western blotting methods, show that Pgp, the product of the mdr gene(s), is overexpressed during the same period. A marked increase in P-gp mRNAs was associated with the induction of a 140-kDa P-gp. This value of 140 kDa is in the range of those previously reported for this protein, namely 130- 200 kDa [4]. Such variation in molecular mass is probably the result of differences in the glycosylation of this transmembrane glycoprotein [31]. As in normal liver, in vitvo P-gp localisation was restricted to the biliary surface of hepatocytes. Another unrelated antigen, specific to the bile canalicular membrane 132, 331, was also reported to be similarly preserved in isolated rat hepatocytes

zymes, namely decreased cytochrome P-450enzymes and increased GST 7-7 [30]. Similar changes in P-gp and GST 7-7 expression have been reported during the development of xenobiotic resistance in the Soh-Farber model of hepatocarcinogenesis [4, 351, in certain tumor cell lines [36] and in rat liver cells after transfection with v-H-ras or v-raf oncogenes [37], suggesting the existence of a common mechanism of regulation of these detoxifying systems [38]. However, while overexpression of fetal marker GST 7-7 can be related to the frequent reappearance of fetal proteins in vitro [30], this is probably not the case for P-gp which is present in low and similar amounts in both fetal and adult liver [12]. Moreover, increased expression of Pgp occurred earlier than that of GST 7-7 in cultured liver parenchymal cells. These observations suggest that P-gp and GST 7-7 are differently regulated in normal hepatocytes. As previously reported by others, [3H]thymidine incorporation and mitotic indexes are very low in standard culture conditions [39, 401. The mitotic index calcukdted over a 24-h period did not exceed 0.2%, 1.2% and 2.1% after 2, 3 and 4 days of culture respectively, and no mitotic figure was observed during the first 24 h (Loyer et al., unpublished results). This suggests that increased expression of P-gp in isolated hepatocytes cultured in standard conditions occurs in the absence of cell division. This increase was also not related to xenobiotic treatment. The P-gp overexpression, starting as soon as 4 h after cell plating, could be a cellular stress response resulting from the isolation of hepatocytes and their exposure to an unfamiliar environment. Indeed P-gp has been shown to be regulated by environmental stress [41] including heat shock and exposure to heavy metal ions. It is presumably significant that the human MDR 1 gene promoter contains two heat-shock consensus elements [41]. The precise mechanism responsible for P-gp overexpression in cultured hepatocytes remains to be elucichted. Active gene transcription is required as indicated by the inhibitory effect of actinomycin D. In addition, inhibition of protein synthesis by cycloheximide leading to a superinduction of Pgp mRNAs suggests that a protein factor negatively controls expression of P-gp in cultured liver parenchymal cells. In conclusion, cultured rat hepatocytes represent an interesting cell model of spontaneous drug resistance. This in vitro system may be suitable for further investigations to understand better mechanism(s) that lead to P-gp overexpression. We thank M. Boisnard-Rissel, A. Fautrel, and F. Morel for their excellcnt collaboration. This work was supported in part by the Associution pour la Recherche sur 161 Cancer (grant 6324).

WI.

P-gp overexpression in cultured rat hepatocytes paralleled a decrease in doxorubicin retention. Various modulating compounds known to reverse drug resistance, such as verapamil, cyclosporin, amitriptyline and quinidine, affected the intracellular accumulation of doxorubicin while, as in normal liver, nifedipine was ineffective [lo]. These results show that P-gp is active in cultured normal rat hepatocytes and has properties similar to those occurring in mdr tumor cells [6] and in liver canalicular membrane vesicles [lo]. Such an overexpression of functional P-gp could compromise the use of cultured hepatocytes as models for drug metabolism by intact liver [16], especially anticancerous compounds. The enhanced expression of P-gp may contribute to the related increased resistance of cultured hepatocytes to cytotoxic drugs. However, this could also be due to changes in drug-metabolizing en-

REFERENCES 1. Ling, V. & Thompson, L. H. (1974) J . Cell. Plzysiol. 83, 103116. 2. Gottesman,M.M.&Pastan,I.(1988)J.Biol. Chem.262,1216312166. 3. Van Dcr Bliek, A . M. & Borst, P. (1989) Adv. Cuncer.. Rcs. 52, 165 - 263. 4. Endicott, J. A.& Ling, V.(1989) Annu. Rev. Biochem. 58. 137171. 5 . Cornwell, M. M., Tsuruo, T., Gottesman, M . M. & Pastan, 1. (1987) FASEBJ. I , 51 -54. 6. Ford, J. M.&Hart, W. N. (1990) Phnrmacol. Rev. 42,155-199. 7. Goldstein, L. J., Galski, H., Fojo, A., Willingham, M., Lai. S. H., Gazdar, A,, Pirker, R., Green, A., Crist, W., Brodeur. G.

852 M., Lieber, M.. Cossma, J., Gottesman, M . M. & Pastan, I. (1989) J . Nutl Cancer Inst. 81, 216-124. 8. Fojo, A. T., Vedak, K., Slamon, D. J., Poplack, D. G., Gottesman, M. M. & Pastan, 1. (1987) Proc. Null Acad. Sci. USA 84,265 - 269. 9. Thiebaut, F., Tsuruo, T., Hamada, H., Gottesman, M . M., Pastan. I . & Willingham, M. (1987) Proc. Nut1 Acad. Sci. USA 84.7735 -7138. 10. Karnimoto. Y., Gatmartan, Z., Hsu, J. & Arias, I. M. (1989) J . Biol. C'hem. 264, 1 1693 - 11698. 1 I . Burt. R. K. & Thorgeirsson, S. S. (1988) J . Nut1 Cunccr Insf. 80, 1383 - 1386. 12. Thorgeirsson, S. S., Huber, B. E., Sonel, S., Fojo, A,, Pastan, I. & Gottesrnan, M. M. (1987) Science 236, 1120-1122. 13. Teeter, L. D.. Becker, F. F., Chisari, F. V., Li, D. & Kuo, M . T. (1990) Mol. Cell. Biol. 10. 5728-5735. 14. Farbcr. E. (1984) Cancer Rm. 44, 5463-5474. 15. Satoh. K.. Kitahara. A., Soma, Y., Inaba, Y . , Hataydma, I. & Sato, K. (1 985) Proc. Nutl Acud. Sci. USA 82, 3964- 3968. 16. Guillouzo, A . . Morel, F., Ratanasavanh, D., Chesne, C. & Guguen-Guillouzo, C. (1 990) Toxic. in Vitro 4, 41 5 -427. 17. Gugucn, C., Guillouzo, A., Boisnard, M., Le Cam, A. & Bourel, M.(1 975) Biol. Gastromterol. 8,223 - 231. 18. Chirgwin, J. M., Przybyla, E. A,, MacDonald, R. J. & Rutter, W. .I. (1979) Biochemistry IS, 5295 - 5299. 19. Raymondjean. M., Kneip. B. & Schapira, G. (1983) Biochimie 65.65 - 70. 20. Riordan, J. R., Dcuchars. K.. Kartner, N., Alar, N., Trent, J . & Ling. V. (1985) Nature 316, 817-819. 21. Ng, W. F., Sarangi, F., Zastawny. R. L., Vcinot Drebot, L. & Ling. V . (1990) Mol. Cell. Biol. 9, 1224-1232. 22. Pernblc. S.. Taylor. J. B. & Ketterer, B. (1986) Biochem. J . 240, 885 - 889. 23. Sargent. T. D.. Wu, J. R., Sala-Trkpat, J. M., Wallace, R. B.. Reyes, A, A. & Bonner, J. (1979) Proc. Nutl Acud. Sci. IISA 76. 3256- 3260. 24. Gerrnann, U. A.. Gottesrnan, M. M. & Pastan, I. (1989) J . Biol. Chertl. 264, 7418 - 7424. 25. Bradford, M. M. A. (1976) Anal. Biochetn. 72,248-254.

26. Laemmli, U. K. (1970) Nature 227, 680-685. 27. Towbin, H., Staehelin, T. & Gordon, J. (1979) Proc. Nut1 Acud. Sci. U S A 76,4350-4354. 28. Kartner, N., Evcrnden-Porelle, D., Bradley, G. & Ling. V. (1985) Nature 316.820 - 823. 29. Schott, B., Huet, S., Benchekroun, M. N., Londos-Gagliardi, D., Vrignaud, P., Montaudan, 0.&Robert, J. (1989) in Anficuncer drugs (Tapiero, H., Robert, J . & Lampidis, T. J., cds) vol. 191,

30.

31.

32. 33. 34.

35.

36.

pp. 245-252, Les Editions INSERM, Paris & John Libbey Eurotext, London. Vandenberghe, Y., Morel, F., Pcmble, S., Taylor, J . B.. Rogiers, V., Ratanasavanh, D., Vercruysse, A., Kettercr, B. & Guillouzo, A. (1990) Mol. Pharmacol. 37, 372-316. Grcenbergcr, L. M., Williams, S. S. & Horwitz, S. B. (1987) J . B i d . Chem. 262,13685-13689. Maurice, M., Durand-Schneidcr, A. M., Garbaz, M. & Feldmann. G. (1 985) Eur. J . Cell. B i d . 99, 122 - 1 29. Scoazec, J. Y., Maurice, M., Moreau, M. & Feldmann, G. (1988) Cancer Res. 48,6882 - 6890. Maurice, M., Rogier, E., Cassio, D. & Feldmann, G. (1988) ./. Cell Sci. YO, 79 -92. Fairchild, C. R., Ivy, S. P., Rushmore, T., Lcc. G.. Koo. P.. Goldsmith, M. E., Myers, C. E., Farber, E. & Cowan. K . H. (1987) Proc. Nut1 Acad. Sci. USA 84,7701 -7705. Moscow, J., Fairchild, C. R., Madden, M. J . , Ransom. D. T.. Wieand, H. S., O'Brien, E. E., Poplack, D. G., Cossman. J.. Myers, C. E. & Cowan, K. H. (1989) Cancer Res. 4Y, 1422-

1428. 37. Burt, R. K . , Garfield, S., Johnson, K. &Thorgcirsson. S. S. (1988) Carcinogenesis 9 , 2329 - 2332. 38. Gottesman, M. M. (1988) J . Nutl Cancer Insz. 80,1352- 1353. 39. McCowan, J. A. & Bucher, N. L. R. (1983) In Vitro 19. 159166. 40. Mitaka, T., Sattler, C. A., Sattler, G. L.. Sargcnt. L. M. & Pilot. H. C. (1991) Hepatology Y3,21-30. 41, Chin, K.-H., Tanaka, S., Darlington, G., Pastan, I . & Gottcsrnan, M. M. (1990) J . Biol. Chem. 265, 221 -226.