Constitutive expression of functional P-glycoprotein in rat hepatoma cells

Eur. J. Biochem. 219, 521 -528 (1994) 0 FEBS 1994

Constitutive expression of functional P-glycoprotein in rat hepatoma cells Olivier FARDEL, Pascal LOYER, ValCrie LECUREUR, Denise GLAISE and AndrC GUILLOUZO Unit6 de Recherches HCpatologiques U 49 de 1’Institut National de la SantC et de la Recherche MCdicale, H6pital Pontchaillou, Rennes, France

(Received August 16/0ctober 25, 1993) - EJB 93 1237/1

P-glycoprotein is a plasma-membrane glycoprotein involved in multidrug resistance. P-glycoprotein overexpression has been demonstrated to occur in tumor cells after cytotoxic drug exposure, but also in some cancers includmg hepatocellular carcinomas before any chemotherapeutic treatment. In order to better analyze this constitutive type of tumoral drug resistance, we have investigated P- glycoprotein expression and function in rat liver tumors induced experimentally by administration of diethylnitrosamine and in two cell clones derived from one of these tumors designated as RHCl and RHC2. High levels of P-glycoprotein mRNAs were found in both liver tumor samples and the two hepatoma cell clones as assessed by Northern blotting; both RHCl and RHC2 cells displayed altered liver functions commonly observed in rat hepatoma cells, particularly the decreased expression of albumin and overexpression of the fetal glutathione S-transferase 7-7. The use of specific multidrug resistance (mdr) probes revealed a major induction of the mdrl gene in liver tumor samples while RHCl and RHC2 cells expressed both mdrl and mdr3 genes without displaying a major alteration in the number of mdr gene copies as assessed by Southern blotting. High amounts of P-glycoprotein were also demonstrated in RHCl and RHC2 cells by Western blotting. These cells were strongly resistant to doxorubicin and vinblastine, two anticancer drugs transported by P- glycoprotein. Doxorubicin intracellular retention was low in RHCl and RHC2 cells, but was strongly enhanced in the presence of verapamil, a known modulator agent of P-glycoprotein; low retention appeared to occur via a drug efflux mechanism, indicating that P-glycoprotein was fully active. These results show that rat hepatoma cells can display elevated levels of functional P- glycoprotein without any prior cytotoxic drug selection and suggest that these cells represent a useful model for analyzing P-glycoprotein regulation in intrinsically clinical drug-resistant cancers.

Resistance to chemotherapy is a major obstacle to the successful treatment of many human cancers. One possible mechanism of this drug resistance is linked to the overexpression of a plasma membrane glycoprotein termed P-glycoprotein [ l , 21. P-glycoprotein is thought to act as an ATP- dependent efflux pump, therefore reducing the intracellular accumulation of antitumor compounds and thus conferring resistance to various structurally and functionnally unrelated cytotoxic drugs, including anthracyclines, Vinca alkaloids, actinomycin D, epipodophyllotoxins and taxol [3,4]. Various compounds such as calcium-channel blockers, cyclosporin and quinidine can inhibit P-glycoprotein-mediated drug efflux and thus reverse the drug resistance phenotype [5]. P- glycoprotein is encoded by mdr genes which constitute a small gene family comprising two members in humans (MDRI and MDR2) and three members in rodents (mdrl, mdr2 and mdr3) [6]. Only MDRI in humans and mdrl and mdr3 (also known as mdrlb and mdrla, respectively) in ro-

Correspondence to 0. Fardel, Institut National de la Sant6 et de la Recherche MCdicale, U 49, Unit6 de Recherches HCpatologiques, HBpital Pontchaillou, F-35033 Rennes, France

Fax: t-33 99 54 01 37. Abbreviations. GST 7-7, glutathione S-transferase 7-7 ; MTT, 3-

[4,S-dimethylthiazol-2-yl]-2,5-diphenyl tetrazoliuni bromide; mdr, multidrug resistance; IC,,, drug concentration producing a 50% reduction in cell numbers.

Enzyme. Glutathione S-transferase (EC 2.5.1 .I 8).

dents have been demonstrated to be involved in drug resistance by transfection experiments [7].

Elevated levels of P-glycoprotein have initially been described in cultured cell lines selected in v i m with increased amounts of a single antitumor compound [8]. Similarly, the overexpression of P-glycoprotein has been demonstrated to occur in human tumors following relapse after chemotherapy. P-glycoprotein expression in this in vivo situation of clini- cally acquired resistance may correspond to that described in in vim-selected cell lines and thus reflects the emergence of resistant cells after treatment [9, 101. An increased expression of P-glycoprotein is also observed in some human cancers derived from different tissues, including colon, kidney, liver or pancreas in the absence of drug treatment [9, 111. Cellular and molecular mechanisms involved in this constitutive drug resistance remain unclear. Moreover, it is not known whether P-glycoprotein intrinsically overexpressed in tumor cells is fully functional since P-glycoprotein activity was usually studied using drug-selected or P-glycoprotein- transfected cells. In order to better analyze and characterize constitutive multidrug resistance, a first step could be to identify cell lines naturally overexpressing P-glycoprotein without prior drug selection, which thus could be relevant to clinical drug resistance in untreated tumors. Such cell lines should be established from tumors exhibiting intrinsic multidrug resistance. Since overexpression of P-glycoprotein is commonly observed in liver tumors of both human and ro-

522

dent origin [12,13], we decided to investigate P-glycoprotein expression and function in cell clones newly obtained from a chemically induced rat liver tumor. Our results demonstrate that these transformed liver cells display intrinsically high levels df functional P-glycoprotein which confer multidrug resistance.

MATERIALS AND METHODS Chemicals

Doxorubicin hydrochloride and vinblastine sulfate were purchased from Roger Bellon laboratories. Verapamil, diethylnitrosamine and cycloheximide were obtained from Sigma.

Animal treatment Hepatomas were chemically induced in rats as previously

described [ 141. Female Sprague-Dawley rats received three doses of diethylnitrosamine (25 mg/kg) 24, 48 and 72 h after a two-third hepatectomy and were maintained with a usual diet until sacrifice 70 weeks after the carcinogenic treatment. At this time, rats treated with the carcinogen have developed hepatocellular carcinomas [ 15, 161.

Cell isolation and culture The two hepatoma cell clones used in this study, desig-

nated RHCl and RHC2, were recently established in our lab- oratory from one diethylnitrosamine-induced rat liver tumor. The tumor was removed 70 weeks after diethylnitrosamine administration and a portion was mechanically dissociated. A highly diluted suspension of isolated cells was cultured in Williams medium supplemented with 10% fetal calf serum, which produced two individual cell clones. These cells were routinely cuItured in William's medium; they were detached by trypsinisation every week. The cultures were used between passages 10 and 50.

The NIH 3T3 drug-sensitive cells used as a negative control for P-glycoprotein expression were maintained in culture conditions similar to those described for the two rat hepatoma cell clones. Freshly isolated normal rat hepatocytes were prepared by the two-step collagenase perfusion method and cultured as previously reported [17].

RNA isolation and analysis Total RNA was extracted from normal or tumoral liver

samples and cultured cells by the guanidinium thiocyanatel cesium chloride method of Chirgwin et al. [18] as mo&fied by Raymondjean et al. [19]. For Northern blotting, 10 pg total RNAs were subjected to electrophoresis in a denaturing formaldehyde/agarose gel and transferred onto Hybond-N sheets. The amount of RNA in each lane and transfer effi- ciency were verified by staining the gel with ethidium bromide. The sheets were prehybridized and hybridized with 32P-labeled probes. P-glycoprotein mRNAs were detected with a hamster pCHPl probe [20] obtained from the Ameri- can Type Culture Collection, and mouse mdrl BBpG4, mdr2 HIpG3 and mdr3 H3pG3 probes (a generous gift from Dr P. Gros, McGill University, Montreal, Canada). The pCHPl sequence has been shown to recognize highly conserved regions of mdr genes [21] while BBpG4, HIpG3 and H3pG3 are mdrl-, mdr2- and mdr3-specific probes, respectively [22]. Glutathione S-transferase 7-7 (GST 7-7), albumin, cyto-

chrome P-450 IIB and glyceraldehyde-3-P-dehydrogenase mRNAs were analyzed with rat GST 7-7 pGSTr7 [23], albumin cDNA [24], cytochrome P-450 2B [25] and glyceraldehyde-3-P-dehydrogenase cDNA [26] probes, respectively. After hybridization, the sheets were washed, dried and autoradiographed at -80°C.

DNA isolation and analysis Cellular DNA was prepared by extraction with phenol/

chlorofodisoamyl alcohol (25 :24: 1, by vol.) followed by precipitation with 2 vol. ethanol as previously described [27]. DNA was digested with the restriction endonuclease PstI and resolved by electrophoresis in a 0.8% agarose gel. The amount of DNA in each lane and enzymic-digestion effi- ciency were verified by staining the gel with ethidium bromide. The DNA was denatured, neutralized and transferred onto Hybond-N sheets. The membranes were prehybridized and hybridized with the 3ZP-labeled P-glycoprotein probe pCHPl. After hybridization, the sheets were washed, dried and autoradiographed at -80°C.

Preparation of membranes and immunoblotting Crude membranes were prepared from cells by dif-

ferential centrifugation as described by Germann et al. [28]. Membrane proteins were estimated by the Bio-Rad protein assay using bovine serum albumin as a standard. Proteins were separated on a SDS/polyacrylamide gel and electropho- retically transferred to nitrocellulose sheets. The concentration of polyacrylamide was 7 %. Nitrocellulose sheets were blocked for 2 h with 20 mM TrisMCl, 0.5 M NaCl, pH 7.5, containing 3 % bovine serum albumin and sequentially incubated with C219 monoclonal antibody (Centocor Inc.) raised against P-glycoprotein [29] and '251-labeled protein A (Amer- sham). After washing, blots were dried and autoradiographed at -80°C. A control blot was performed using the same protocol with non-immune myeloma cell ascites as the primary antibody.

Drug-sensitivity assay Drug effect on cell proliferation was evaluated using the

3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide (MTT) dye assay. Briefly, rat liver tumor-derived cells and NIH 3T3 cells were cultured for 72 h with various concentrations of cytotoxic drug in the absence or presence of verapamil (12.5 pM). Cells were incubated with 0.05% MTT for 2 h at 37°C and blue formazan formed was spectrophoto- metrically assessed as previously described [30]. Growth inhibition was evaluated as ICs0, i.e. the drug concentration providing a 50% reduction of cell numbers as compared to controls cultured in parallel without drug.

Evaluation of intracellular doxorubicin accumulation The intracellular concentration of doxorubicin was esti-

mated as described by Schott et al. [31] with slight modifi- cations. Cells were exposed to doxorubicin (10 pg/ml) for 2 h in the absence or presence of verapamil (12.5 pM) or various diethylnitrosamine concentrations (1 - 250 pM) and washed,with ice-cold 140 mM NaC1, 10 mh4 sodium phosphate, pH 7.4 (NaCVPi) harvested and ultrasonicated. These steps were performed quickly i s order to avoid any drug efflux. Proteins were precipitated with 20% trichloroacetic

523

acid. The acid-soluble fraction was used to evaluate the intracellular concentration of doxorubicin by fluorimetry using excitation and emission wavelengths of 485 nm and 590 nm, respectively. Preliminary controls showed no toxicity of doxorubicin, verapamil or diethylnitrosamine at the concentrations used over the incubation period and no interference of doxorubicin fluorescence by verapamil or diethylnitrosamine.

Evaluation of doxorubicin emux Cells were incubated with doxorubicin (10 pg/ml) for 2 h

as described above, washed three times with ice-cold and NaCIRi and incubated again in drug-free medium for 1 h or 2 h in the absence or presence of verapamil (12.5 pM). The intracellular doxorubicin concentration was further determined by the fluorimetric method as described above. Intra- cellular drug retention was expressed as a percentage of the initial drug intracellular accumulation.

Assay of protein synthesis Protein synthesis was determined by measuring the incor-

poration of ['4C]leucine (Amersham) in trichloroacetic-pre- cipitable macromolecules as previously described [32].

Statistical analysis The results of cellular doxorubicin accumulation and ef-

flux studies were analyzed by the Student's t-test. The crite- rion of significance of the differences between the means (%standard deviation) was P < 0.05.

RESULTS P-glycoprotein &A levels in liver tumor samples re-

moved from rats 70 weeks after diethylnitrosamine administration were estimated by Northern blotting (Fig. 1A). High amounts of a 4.5-kb P-glycoprotein mRNA were evidenced in neoplasic livers compared to normal livers with the use of the mdrl-specific probe BBpG4. No signal was detected by using mdr2 HIpG3 and mdr3 H3pG3 probes in both normal and tumoral livers (data not shown). RHCl and RHC2 cell clones, established from one of the diethylnitrosamine-induced liver tumors, also demonstrated increased levels of mdrl mRNAs compared to normal rat hepatocytes or NIH 3T3 cells (Fig. 1B). Moreover, hybridization with the mdr3 probe revealed elevated levels of 5.1-kb mdr3 mRNAs in RHCl and RHC2 cells. This high expression of both mdrl and mdr3 genes in the liver tumor-derived cell clones was also demonstrated by using the pCHPl probe which recognizes highly conserved regions of mdr genes and thus indicated both mdrl and mdr3 mRNAs (Fig. 1B) P-glycoprotein mRNA levels in RHCl and RHC2 cells shown in Fig. lB, were observed in RHCl and RHC2 cells maintained in culture for 30 passages. Similar values were obtained from cells analyzed after 10 or 50 passages.

Northern-blot analyses were also performed in order to determine the expression of various liver functions in RHCl and RHC2 cells. These tumor liver-derived cells were found to express low levels of albumin and cytochrome P-450 2B mRNAs in contrast to normal hepatocytes (Fig. 2). GST 7-7, a fetal liver marker [33], was strongly expressed in both RHCl and RHC2 cell clones, but was not detectable in normal liver parenchymal cells (Fig. 2).

A

B

I

I

II

m

kb

- 4.5

- 4.5

- 5.1

1 5 . 1

4.5

Fig. 1. Expression of P-glycoprotein mRNAs in diethylnitrosamine-induced hepatoma cells. Each well contains 10 pg total RNAs isolated from (A) normal rat livers (NL) and diethylnitrosamine-induced rat liver tumors (LT) and (B) normal freshly isolated rat hepatocytes (NH), NIH 3T3 cells, RHCl and RHC2 cells isolated from one of the diethylnitrosamine-induced rat liver tumors. RNAs were transferred to Hybond-N sheets after electrophoresis and hybridized with mdrl (I) and mdr3 (11) gene-specific probes and with pCHPl probe (111) which recognizes highly conserved regions of mdr genes. Transcript sizes (mkb) were estimated relative to the migration of 18 S and 28 S rRNA.

Since P-glycoprotein mRNA overexpression in cultured drug-resistant tumor cells has been demonstrated to be fre- quently associated with mdr gene amplification [4], Southern blotting analyses of genomic DNA isolated from RHCl and RHC2 cells and normal hepatocytes were performed in order to examine the number of mdr gene copies (Fig. 3). Hybrid- ization with pCHPl probe revealed a similar PsA restriction pattern and amount of mdr genes in both tumoral liver cells and normal hepatocytes.

Crude membrane fractions were further prepared from RHCl and RHC2 cells as well as from normal hepatocytes and were used to investigate P-glycoprotein expression by Western-blot analysis (Fig. 4). A C219-reactive band of 140 kDa corresponding to P-glycoprotein was markedly overexpressed in the two liver tumor-derived cell clones compared to normal rat hepatocytes. No reactive band of 140 kDa was observed in a control blot performed with non-immune myeloma cell ascite as the primary antibody (data not shown).

Drug sensitivity experiments were performed using the MTT dye assay in order to determine the effect on cell proliferation of doxorubicin and vinblastine, two cytotoxic com-

5 24

A

kb

+ 2.4 -205

+ -116.5

- 80

6 c2.1 -49.5

C -0.94

Fig. 2. Expression of albumin, cytochrome P-450 2B and glutathione-S-transferase 7-7 (GST 7-7) mRNAs in RHCl and RHC2 cells. Each well contains 10 pg total RNAs isolated from normal freshly isolated rat hepatocytes (NH), rat RHCl and RHC2 cells. RNAs were transferred to Hybond-N sheets after electrophoresis and hybridized with albumin (A), cytochrome P-450 2B (B) and GST 7-7 (C) probes, respectively. Transcript sizes were estimated relative to the migration of 18 S and 28 S rRNA. The exposure times were 72 h (albumin and cytochrome P-450 2B) and 18 h (GST 7-7).

Fig. 3. Southern-blot analysis of RHCl and RHC2 cell genomic DNA. Each well contains 10 pg PstI-digested DNA isolated from normal freshly isolated rat hepatocytes (NH), RHCl and RHC2 cells. DNA was transferred to Hybond-N sheets after electrophoresis in 0.8% agarose gel and hybridized with the P-glycoprotein probe pCHPl. DNA size markers (kb) are indicated on the right. The exposure time was 30 h.

pounds which are actively transported by P-glycoprotein. NIH 3T3 cells appeared to be highly sensitive to both doxorubicin and vinblastine, while RHCl and RHC2 cells were markedly more resistant to these drugs. Comparison of IC,, values thus indicated that RHCl and RHC2 cells were 169- fold and 103-fold more resistant to doxorubicin than NIH 3T3 cells (Table 1). Vinblastine IC,, values were also 37.4- fold Bnd 25.7-fold enhanced for RHCl and RHC2 cells,

Fig. 4. Western-blot analysis of membrane proteins obtained from RHCl and RHC2 cells. Crude membrane fractions were prepared from normal freshly isolated rat hepatocytes (NH), NIH 3T3 cells, rat RHCl and RHC2 cells. 100 pg membrane proteins were separated on SDSPAGE and transferred onto a nitrocellulose sheet. After incubation with C219 antibody raised against P-glycoprotein, the blot was developed as described in the Materials and Methods section. Arrows indicate the position of P-glycoprotein. The position of molecular mass standards ( m a ) is indicated on the right.

respectively, compared to sensitive NIH 3T3 cells. Incuba- tion of the cytotoxic compounds with 12.5 pM verapamil strongly decreased the resistance of both RHCl and RHC2 cell clones to doxorubicin and vinblastine, but only mar- ginally affect the effect of the drugs on NIH 3T3 cells (Table 1).

Functional P-glycoprotein activity in RHCl and RHC2 cells was further analyzed by investigation of cellular doxorubicin accumulation and efflux. As shown in Fig. 5 , treatment of cells with doxorubicin led to lower accumulation of the drug in the tumoral liver cells than in NIH 3T3 cells. Addition of verapamil (12.5 pM) significantly increased doxorubicin retention in RHCl and RHC2 cell clones, but not in NIH 3T3 cells (Fig. 6). In contrast, the addition of diethylnitrosamine at various concentrations did not affect intracellular doxorubicin accumulation in RHCl and RHC2 cells (Fig. 6). Efflux studies also demonstrated that the cellular doxorubicin concentration in liver tumor-derived cells strongly decreased during post-incubation in drug-free medium, but was unaffected in NIH 3T3 cells (Fig. 7). This doxorubicin efflux from RHCl and RHC2 cells was significantly inhibited by coincubation with 12.5 pM verapamil (Fig. 7).

Previous studies had shown that cycloheximide treatment strongly induced P-glycoprotein in primary cultures of rat hepatocytes [34, 351. In order to evaluate the effect of protein-synthesis inhibition on P-glycoprotein levels in RHCl and RHC2 cells, cells were exposed to cycloheximide (60 pg/ml) for 6 h. Protein synthesis estimated by the deter- mination of the incorporation of ['4C]leucine was strongly inhibited after cycloheximide treatment in normal hepatocytes, RHCl and RHC2 cells, not exceeding 3.2, 1.9 and 2.1% of the values found in their untreated counterparts, respectively. Concomitantly, an increase in mdrl mRNA levels was demonstrated in liver tumor-derived cells after cycloheximide exposure, but to a lesser extent than in normal hepatocytes, particularly in RHC2 cells (Fig. 8). The use of mdr3-specific probe revealed no major alteration in mdr3 mRNA levels in cells after cycloheximide treatment. Glycer- aldehyde-3-P-dehydrogenase mRN4 levels were not affected

525

Table 1. Sensitivity of RHCl and RHC2 cells to doxorubicin and vinblastine. RHC1, RHC2 and NIH 3T3 cells were cultured for 72 h with various concentrations of cytotoxic drug in the absence or presence of verapamil (12.5 pM). The drug effects on cell proliferation were determined by using the MTT dye assay as described in the Materials and Methods section. Growth inhibition is expressed as IC5,, i.e. the drug concentration providing a 50% reduction of cell numbers as compared to controls cultured in parallel without drug.

Cell type IC,, for doxorubicin IC,, for vinblastine

- verapamil + verapamil - verapamil + verapamil

~~~ ~

NIH 3T3 13.20 t 1.02 9.08 2 1.53 0.98 5 0.023 0.42 ? 0.09 RHCl 2236 +- 183 105 ? 15.81 36.66 5 9.56 1.80 t 0.50 RHC2 1357 5 2 8 1 49.65 ? 20.70 25.16 ? 6.48 1.08 5 0.31

Fig. 5. Intracellular doxorubicin accumulation in RHCl and RHC2 cells. NIH 3T3, RHCl and RHC2 cells were incubated with 10 p g / d doxorubicin for 2 h. The intracellular doxorubicin concentration was determined using the fluorimetric method as described in the Materials and Methods section. The values are the mean 2 SD of three independent experiments in quintuplicate; *, P < 0.05.

VR 1 T

500

400

3 0

200

100

n " Hll l ST3 R H S ,

I VR

RHCZ

Fig. 6. Effect of verapamil (VR) and diethylnitrosamine (DENA) on doxorubicin retention in RHCl and RHC2 cells. Cells were incubated with 10 pg/ml doxorubicin for 2 h with verapamil (12.5 pM) or various concentrations of DENA (1-250 pM)., The intracellular doxorubicin concentration was determined using the fluorimetric method as described in the Materials and Methods section. The values are expressed relative to the drug accumulation values observed in the presence of doxorubicin alone and are the mean -tSD of three experiments in quintuplicate; *, P < 0.05.

by protein synthesis inhibition in both cultured normal and tumoral liver cells (Fig. 8).

DISCUSSION A way to better understand the cellular and molecular

mechanisms involved in constitutive multidrug resistance of tumors is to isolate cell lines naturally overexpressing P-glycoprotein. We report in the present study high expression of functional P-glycoprotein in two hepatoma cell clones, RHCI and RHC2, derived from an experimentally induced rat liver tumor. These cells expressed albumin, although at a low level, as assessed by Northern blotting, indicating that they originate from parenchymal liver cells. RHCl and RHC2 cells were also found to have low cytochrome P-450 2B and elevated GST 7-7 levels, which represent phenotypic alterations in phase-I and phase-I1 metabolizing enzymes usually observed during hepatocarcinogenesis [36, 371. In addition, RHCl and RHC2 cells were highly tumorigenic in nude mice, indicating that they are transformed (data not shown). All the data clearly suggest that RHCl and RHC2 cells can be identified as rat hepatoma cells.

RHCl and RHC2 cells were derived from a rat liver tumor induced by administration of diethylnitrosamine following partial hepatectomy without promotion with 2-ace- tyl aminofluorene in contrast to the Solt-Farber method [38]. The neoplasic lesions induced in the liver by this carcinogenic treatment displayed elevated levels of P-glycoprotein as demonstrated by Northern blotting. Previous studies have indicated that P-glycoprotein is overexpressed in liver tumors induced in rats by the Solt-Farber protocol or by N-nitrosomorpholine administration [39-411 and in mice by various chemical carcinogen treatments or by transgenesis using a hepatitis-B-virus gene [13]. Taken together, these data strongly suggest that P-glycoprotein overexpression could represent a general feature of neoplasic liver. Moreover, in contrast to previous studies on P-glycoprotein expression during rat hepatocarcinogenesis [39-411, we were able to demonstrate using specific mdr gene probes that this increased P- glycoprotein expression during rat hepatocarcinogenesis may occur via a marked induction of the mdrl gene. This con- trasts with a report on mouse liver tumors showing that the overexpressed mdr gene was mdr3 [13]. Such discrepancies in the nature of mdr gene(s) induced during rat and mouse liver carcinogenesis could in fact reflect species differences in P-glycoprotein regulation. Variations in P-glycoprotein regulation recently observed between rat and mouse hepatocytes in primary cultures support this conclusion [42].

High levels of P-glycoprotein mRNAs were demonstrated in RHCl and RHC2 cells by Northern blotting and

526

z u f B I)

?!

2 n W

A

NIH 3T3

0- 1 i 0

120 1

60

40

I \ T *

R H C l " ! :

0 1 2

'*O 1

60

2o 1 R H C P 0-

0 1 2

Time (h)

Fig.7. Doxorubicin efflux from RHCl and RHC2 cell. Cultured NIH 3T3, RHCl and RHC2 cells were incubated with doxorubicin (10 Fg/ml) for 2 h, washed three times with ice-cold NaCL/P, and incubated again in drug-free medium for 1 h or 2 h. These studies were performed in the absence (0) or the presence (0) of verapamil (12.5 pM). The intracellular doxorubicin concentration was further determined by the fluorimetric method as described above. Intracel- lular retained drug values were expressed relative to the initial drug accumulation values. The values are the mean ? SD of three independent experiments in quintuplicate; *, P < 0.05.

were associated with an elevated amount of a 140-kDa P- glycoprotein as assessed by immunoblotting. This value of 140 kDa is in the range of those values previously reported for this protein, namely 130-200 kDa [3]. Such a variation in molecular mass probably results from differences in the glycosylation of the P-glycoprotein [43]. P-glycoprotein was also present in normal hepatocytes as indicated by Western blotting, thus probably reflecting the physiological presence in these cells of mdr gene transcripts recently detected by using the sensitive RNase-protection assay [44].

The increased expression of P-glycoprotein in liver tumor-derived cells paralleled the poor accumulation of doxorubicin, which is markedly enhanced in the presence of the known P-glycoprotein modulator agent verapamil [5]. This low intracellular retention of doxorubicin appeared to occur at least partly via an efflux mechanism sensitive to verapamil. These results suggest that P-glycoprotein is fully

B

C

Fig. 8. Effect of cyclobeximide treatment on P-glycoprotein mFWA levels in normal rat hepatocytes and RHCl and RHC2 cells. Each well contains 10 pg total RNAs isolated from untreated (U) or cycloheximide-treated (Cx) cells. Cultured normal hepatocytes (NH) and RHCl and RHC2 cells were exposed to cycloheximide (60 pg/ml) for 6 h. RNAs were transferred to Hybond-N sheets after electrophoresis and hybridized with mdrl (A)- and mdr3 (B)- gene-specific probes and a glyceraldehyde-3-P-dehydrogenase (C) probe. Transcript sizes (mkb) were estimated relative to the migration of 18 S and 28 S rRNA.

active in RHCl and RHC2 cells. This conclusion is also supported by the results of drug-sensitivity assays. RHCl and RHC2 cells were highly resistant to doxorubicin and vinblastine compared to sensitive NIH 3T3 cells and coincubation with verapamil strongly decreased the degree of resistance of these hepatoma cells.

The high expression of P-glycoprotein in RHCl and RHC2 cells appears to be stable in culture since similar elevated P-glycoprotein mRNA levels were observed in cells maintained in culture for 10, 30 or 50 passages. It did not occur after in vitro cytotoxic drug selection as classically described for multidrug-resistant cell lines. Moreover, it did not result from an in vivo selection with a P-glycoprotein- related drug since diethylnitrosamine, the carcinogen agent used to induce the liver tumor, is not a substrate for P-glycoprotein. Indeed, no multidrug-resistant cell lines have been shown to be cross-resistant to diethylnitrosamine [39] and we have demonstrated that diethylnitrosamine did not in- terfere with P-glycoprotein-related doxorubicin transport in RHCl and RHC2 cells. Therefore, this clearly suggests that the drug resistance observed in RHCl and RHC2 cells may be an intrinsic phenomenon which thus could represent at least partly an in vitro counterpart of in vivo liver tumor drug resistance.

It is not known whether P-glycoprotein overexpression is a general phenomenon in hepatoma cell lines as in in vivo liver tumors. Some rat chemically induced hepatoma cell lines, including H35 (Reuber) and Yoshida cells generated after exposure to 2-acetylaminofluorene and dimethylamino- benzene, respectively [45, 461, have been recently reported to display elevated levels of P-glycoprotein without prior exposure to anticancer drugs in vitro [47, 481. However, these cell lines have been established for more than 30 years and their drug resistance could therefore reflect phenotypic alteration or evolution during culture and no data are available on P-glycoprotein expression in the liver tumors from which

527

they were isolated. En contrast to rat hepatoma cells, mouse hepatoma Hepa cells have been demonstrated to have similar sensitivity to doxorubicin and vinblastine as NIH 3T3 cells 1491, and five human hepatocarcinoma cell lines studied by Shen et al. 1501 displayed very low levels of P-glycoprotein. These results clearly suggest that overexpression of functional P-glycoprotein is not a general feature of liver tumor- derived cell lines. These variations in P-glycoprotein expression in cultured hepatoma cells could in fact reflect phenotypic changes during culture or tumor heterogeneity from which the cell lines were derived.

Northern-blot analyses and hybridization with mdr-specific probes have shown that m d r l and mdr3 genes, the two mdr genes involved in drug resistance in rodents [7], are both expressed in RHCl and RHC2 hepatoma cells, while only the m d r l gene seems to be strongly induced in the liver tumor from which the cell clones were isolated. These data favor the idea that RHCl and RHC2 cell clones derive from tumoral cells exhibiting both mdrl and mdr3 expression, but which represent only a small fraction of the whole tumoral liver cell population, thus accounting for the failure to ob- serve mdr3 transcripts in tumor samples. Alternatively, it could be speculated that mdr3 was induced in RHCl and RHC2 cells in response to cell isolation and culture. P-glycoprotein overexpression has been reported to occur in primary cultures of adult rat hepatocytes and has been hypothesized to reflect a cellular-stress response resulting from the exposure of liver cells to an unfamiliar environment [51]. How- ever, this induction seems to be principally related to an increase in expression of the mdrl gene 1521.

The precise mechanisms responsible for P-glycoprotein induction in hepatoma cells remain to be elucidated. As demonstrated by Southern blotting, P-glycoprotein expression in RHCl and RHC2 cells was not associated with major alterations in the number of mdr gene copies in contrast to many other selected drug-resistant cell lines [4]. This indi- cates that P-glycoprotein induction in these cells may require activation of mechanisms acting at a transcriptional or post- transcriptional level.

P-glycoprotein expression in normal hepatocytes has previously been shown to be strongly induced by cycloheximide treatment [34]. These data suggest that P-glycoprotein expression is at least partly negatively regulated by a labile protein factor, through a transcriptional mechanism as assessed by nuclear run-on experiments [35]. In order to precisely determine the relevance of this putative labile repressor factor in our drug-resistant hepatoma cells, P-glycoprotein expression in RHCl and RHC2 cells was analyzed after protein synthesis inhibition by Northern blotting. Re- sults indicated that the m d r l gene, the mdr gene strongly induced in normal hepatocytes, was less increased in hepatoma cells, particularly in RHC2 cells. The reasons for this weaker inducibility of the mdr l gene in hepatoma cells are unclear, but it cannot be linked to cycloheximide resistance since this compound inhibited strongly protein synthesis in both hepatocytes and RHCl and RHC2 cells. It could reflect alteration in mdrl negative regulation via the labile repressor in liver tumor-derived cells. Further studies are needed to support this hypothesis.

In summary, the results reported in this study demonstrated that two cell clones isolated from a chemically induced rat liver tumor displayed elevated levels of P-glycoprotein. This P-glycoprotein was fully functional and con- ferred multidrug resistance without any prior cytotoxic drug selection. This intrinsic drug resistance model should conse-

quently constitute a useful system to better understand cellular and molecular mechanisms of P-glycoprotein regulation in naturally clinical drug-resistant cancers, particularly liver tumors.

We are grateful to Dr P. Gros for the gift of the mdr-specific probes and to Dr C. Guguen-Guillouzo for critical reading of the manuscript. This work was supported by the Institut National de la Sante‘ et de la Recherche Mkdicale, the Association pour la Re- cherche sur le Cancer and the Ligue Nationale contre te Cancer (Comite‘ d’llle et Vilaine) 0. Fardel and P. Loyer are recipients of fellowships from the Association pour la Recherche sur le Cancer.

REFERENCES 1. Gottesman, M. M. & Pastan, I. (1988) The multidrug transpor-

ter, a double-edged sword, J. Biol. Chem. 262,12163-12166. 2. Van Der Bliek, A. M. & Borst, P. (1989) Multidrug resistance,

Adv. Cancer Res. 52, 165-263. 3. Endicott, J. A. & Ling, V. (1989) The biochemistry of P-glyco-

protein-mediated multidrug resistance, Annu. Rev. Biochem. 58, 137-171.

4. Roninson, I. B. (1992) The role of the MDRl (P-glycoprotein) gene in multidrug resistance in vitro and in vivo, Biochem. Pharmacol, 43, 95-102.

5. Ford, J. M. & Hart, W. N. (1990) Pharmacology of drugs that alter multidrug resistance in cancer, Pharmacol. Rev. 8, 137- 171.

6. Thorgeirsson, S . S., Silvermann, J. A., Gant, T. W. & Marino, P. (1991) Multidrug resistance gene family and chemical car- cinogens, Pharmacol. & Ther. 49, 283-292.

7. Devault, A. & Gros, P. (1990) Two members of the mouse mdr gene family confer multidrug resistance with overlapping but distinct drug specificities, Mol. Cell. Biol. 10, 1652-1663.

8. Juliano, R. L. & Ling, V. (1976) A surface glycoprotein modu- lating drug permeability in Chinese hamster ovary cell mu- tants, Biochim. Biophys. Acta 455, 152-162.

9. Goldstein, L. J., Galski, H., Fojo, A., Willingham, M., Lai, S. H., Gazdar, A., Pirker, R., Green, A., Crist, W., Brodeur, G. M., Lieber, M., Cossma, J., Gottesman, M. M. & Pastan I. (1989) Expression of a multidrug resistance gene in human cancer, J. Natl Cancer Inst. (Bethesda) 181, 116-124.

10. Nooter, K. & Henveijer, H. (1991) Multidrug resistance (mdr) in human cancer, Br. J. Cancer 63, 663-669.

11. Fojo, A. T., Ueda, K., Slamon, D., Poplack, D. G., Gottesman, M. M. & Pastan, I. (1987) Expression of a multidrug-resistance gene in human tumors and tissues, Proc. Natl Acad. Sci. USA 84, 265-269.

12. Huang, C., Wu, M., Xu, G., Li, D., Cheng, H., Tu, Z., Jiang, H. & Gu, J. (1992) Overexpression of the MDRl gene and P- glycoprotein in human hepatocellular carcinoma, J. Natl Can- cer Inst. (Bethesda) 84, 260-262.

13. Teeter, L. D., Becker, E F,, Chisari, F. V., Li, D. & Kuo, M. T. (1990) Overexpression of the multidrug resistance gene mdr3 in spontaneous and chemically induced mouse hepatocellular carcinomas, Mol. Cell. Biol. 10, 5728-5735.

14. Beaumont, C., Lerumeur, E., Guillouzo, C., Latinier, M. F., Bourel, M. & Guillouzo, A. (1981) Ferritin is increased in diethylnitrosamine altered rat hepatocytes, Biochem. Biophys. Res. Commun. 99, 879-885.

15. Squire, R. A. & Levitt, M. H. (1975) Report of a workshop on classification of specific hepatocellular lesions in rats, Cancer Res. 35, 3214-3223.

16. Schemer, E. & Emmelot, P. (1975) Foci of altered liver cells induced by a single dose of diethylnitrosamine and partial hepatectomy : Their contribution to hepatocarcinogenesis in rat, Eul: J. Cancer 11, 145-154.

17. Guguen, C., Guillouzo, A., Boisnard, M., Le Cam, A. & Bourel, M. (1975) Etude ultrastructurale de monocouches d’hkpato- cytes de rat adulte cultivCs en prksence d’hkmisuccinate d’hy- drocortisone, Biol. Gastro-enterol. 8, 223 -231.

528

18. Chirgwin, J. M., Przybyla, E. A., MacDonald, R. J. & Rutter, W. J. (1979) Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease, Biochemistry 18, 5294 - 5299.

19. Raymondjean, M., Kneip, B. & Schapira G. (1983) Preparation and characterization of mRNAs from rat heart muscle, Biochi- mie (Paris) 65, 65-70.

20. Riordan, J. R., Deuchars, K., Kartner, N., Alar, N., Trent, J. & Ling, V. (1985) Amplification of P-glycoprotein genes in multidrug resistant mammalian cell lines, Nature 316, 817-819.

21. Ng, W. F., Sarangi, F., Zastawny, R. L., Veinot Drebot, L. & Ling, V. (1989) Identification of members of the P-glycoprotein multigene family, Mol. Cell. Biol. 9, 1224-1232.

22. Raymond, M., Rose, E., Housman, D. E. & Gros, P. (1990) Physical mapping, amplification and overexpression of the mouse mdr gene family in multidrug-resistant cells, Mol. Cell. Biol. 10, 1642-1651.

23. Pemble, S., Taylor, J. B. & Ketterer, B. (1986) Tissue distribution of rat glutathione transferase subunit 7, a hepatoma marker, Biochem. J. 240, 885-889.

24. Sargent, T. D., Wu, J. R., Sala-Trkpat, J. M., Wallace, R. B., Reyes, A. A. & Bonner, J. (1979) The rat serum albumin gene: analysis of cloned sequences, Proc. Natl Acad. Sci. USA 76,3256-3260.

25. Adesnik, M., Bar-Nun, S., Maschio, F., Zunich, M., Lippman, A. & Bard, E. (1981) Mechanism of induction of cytochrome P-450 by phenobarbital, J. Biol. Chem. 256, 10340-10345.

26. Fort, P., Marty, L., Piechaczyk, M., Sabrouty, S. E., Dai, C., Jeanteur, P. & Blanchard, J. M. (1985) Various rat adult tissues express only one major mRNA species from the glyceraldehyde-3-phosphate-dehydrogenase multigenic family, Nu- cleic Acids Res. 13, 1431-1443.

27. Gripon, P., Diot, C., ThCzC, N., Fourel, I., Loreal, O., BrCchot, C. & Guguen-Guillouzo, C. (1988) Hepatitis B virus infection of adult human hepatocytes cultured in the presence of di- methyl-sulfoxide, J. Virol. 62, 4136-4143.

28. Germann, U. A., Gottesman, M. M. & Pastan, I. (1989) Expres- sion of a multidrug resistance-adenosine deaminase fusion gene J . Biol. Chem. 264,7418-7424.

29. Kartner, N., Evernden-Porelle, D., Bradley, G. & Ling, V. (1985) Detection of P-glycoprotein in multidrug-resistant cell lines by monoclonal antibodies, Nature 316, 820- 823.

30. Carmichael, J., DeGraff, W. G., Gazdar, A. F., Minna, J.D. & Mitchell, J. B. (1987) Evaluation of a tetrazolium-based semi- automated colorimetric assay : assessment of chemosensitivity testing, Cancer Res. 47, 936-942.

31. Schott, B., Huet, S., Benchekroun, M. N., Londos-Gagliardi, D., Vrignaud, P., Montaudon, D. & Robert, J. (1989) Pharmaco- logical and molecular characterization of three rodent cell lines in culture selected for resistance to doxorubicin, in An- ticancer drugs (Tapiero, H., Robert J. & Lampidis, T. J., eds) vol. 191, pp. 245-252, Les Editions Institut National de la SantC et de la Recherche MCdicale, Paris and John Libbey Eurotext, London.

32. Guyomdrd, C., ChesnC, C., Meunier, B., Fautrel, A., Clerc, C., Morel, F., Rissel, M., Campion, J.-P. & Guillouzo, A. (1990) Primary culture of adult rat hepatocytes after 48-hour preser- vation of the liver with cold UW solution, Hepatology 12,

33. Vandenberghe, Y., Morel, F., Pemble, S., Taylor, J. B., Rogiers, V., Ratanasavanh, D., Vercruysse, A., Ketterer, B. & Guil- louzo, A. (1990) Changes in expression of mRNA coding for glutathione S-transferase subunits 1-2 and 7 in cultured rat hepatocytes, Mol. Pharmacol. 37, 372-376.

34. Fardel, O., Ratanasavanh, D., Loyer, P., Ketterer, B. & Guil- louzo, A. (1992) Overexpression of the multidrug resistance gene product in adult rat hepatocytes during primary culture, Eul: J. Biochem. 205, 847-852.

35. Gant, T. W., Silverman, J. A. & Thorgeirsson, S. S. (1992) Reg- ulation of P-glycoprotein gene expression in hepatocyte cul-

1329-1 336.

tures and liver cell lines by a trans-acting transcriptional repressor, Nucleic Acids Res. 20, 2841 -2846.

36. Farber, E. (1984) Cellular biochemistry of the stepwise develop- ment of cancer with chemicals: G.H.A. Clowes memorial lec- ture, Cancer Res. 44, 5463 -5474.

37. Satoh, K., Kitahara, A,, Soma, Y., Inaba, Y., Hatayama, I. & Sato, K. (1985) Purification, induction and distribution of pla- cental glutathione transferase: A new marker enzyme for preneoplastic cells in the rat chemical hepatocarcinogenesis, Proc. Natl Acad. Sci. USA 82, 3964-3968.

38. Solt, D. B. & Farber, E. (1976) New principle for the analysis of chemical carcinogenesis, Nature 263, 702-703.

39. Thorgeirsson, S. S., Huber B. E., Sonel, S., Fojo, A., Pastan, I. & Gottesman, M. M. (1987) Expression of the multidrug resistance gene in hepatocarcinogenesis and regenerating rat liver, Science 236, 1120-1122.

40. Fairchild, C. R., Ivy, S. P., Rushmore, T., Lee, G., Koo, P., Goldsmith, M. E., Myers, C. E., Farber, E. & Cowan, K. H. (1 987) Carcinogen-induced mdr overexpression is associated with xenobiotic resistance in rat preneoplastic liver nodules and hepatocellular carcinomas, Proc. Natl Acad. Sci. USA 84,

41, Volm, M., Zerban, H., Mattern, J. & Efferth, T. (1990) Overex- pression of P-glycoprotein in rat hepatocellular carcinomas induced with N-nitrosomorpholine, Carcinogenesis 11, 169- 172.

42. Fardel, O., Morel, F. & Guillouzo, A. (1993) P-glycoprotein expression in human, mouse, hamster and rat hepatocytes in primary culture, Carcinogenesis 14, 781 -783.

43. Greenberger, L. M., Williams, S. S. & Honvitz, S. B. (1987) Biosynthesis of heterogeneous forms of multidrug resistance- associated glycoproteins, J. Biol. Chem. 262, 13 685 - 13 689.

44. Brown, P. C., Thorgeirsson, S. S. & Silverman, J. A. (1993) Cloning and regulation of the rat mdr2 gene, Nucleic Acids Res. 21, 3885-3891.

45. Reuber, M. D. (1961) A transplantable bile-secreting hepatocel- Mar carcinoma in the rat, J. Natl Cancer Inst. (Bethesda) 26, 891 -897.

46. Yoshida, T. (1956) Contributions of the ascites hepatoma to the concept of malignancy of cancer, Ann. NY Acad. Sci. 63,

47. Ellis, A. L., Munger, C. E., Bunch, R. T., Woods, K. E., Ran- dolph, J. K., Boise, L., Swerdlow, P. S., Zwelling, L. A,, Hinds, M., Yanovich, S. & Gewirtz, D. A. (1992) Components of intrinsic drug resistance in the rat hepatoma, Biachem. Pharmacol. 43,331 -342.

48. Miyamoto, K.-I., Wakusawa, S. & Nakamura, S. (1992) Drug resistance dependent on different molecular size P-glycoproteins in Yoshida rat ascites hepatoma cells, Biochern. Pharma- col. 43, 1143-1145.

49. Ikeguchi, M., Teeter, L. D., Eckersberg, T., Ganapathi, R. & Tien Kuo, M. (1991) Structural and functional analyses of the promoter of the murine multidrug resistance gene mdr3/mdrl a reveal a negative element containing the AP-1 binding site, DNA Cell Biol. 10, 639-649.

50. Shen, D.-W., Lu, Y.-G., Chin, K.-V., Pastan, I. & Gottesman, M. M. (1991) Human hepatocellular carcinoma cell lines exhibit multidrug resistance unrelated to MDRl gene expression, J. Cell Sci. 98, 317-322.

51. Fardel, O., Loyer, P, Morel, F., Ratanasavanh, D. & Guillouzo, A. (1992) Modulation of multidrug resistance gene expression in rat hepatocytes maintained under various culture conditions, Biochem. Pharmacol. 44, 2259-2262.

52. Fardel, O., Lecureur, V. & Guillouzo, A. (1993) Regulation by dexamethasone of P-gl ycoprotein expression in cultured rat hepatocytes, FEBS Lett. 327, 189-193.

7701 -7705.

852- 881.

Date post:	14-May-2023
Category:	Documents
Upload:	independent
View:	0 times
Download:	0 times

Constitutive expression of functional P-glycoprotein in rat hepatoma cells

Documents