George Square, Edinburgh EH8 9XD, department of Clinical Chemistry,. University of ..... differences in thymidine kinase activity or other enzymes involved.
Carcinogenesis vol.9 no.7 pp. 1283-1287, 1988
Glutathione and glutathione-dependent enzymes in ovarian adenocarcinoma cell lines derived from a patient before and after the onset of drug resistance: intrinsic differences and cell cycle effects Alex D.Lewis1, John D.Hayes2 and C.Roland Wolf1-3 'imperial Cancer Research Fund, Laboratory of Molecular Pharmacology and Drug Metabolism, Department of Biochemistry, Hugh Robson Building, George Square, Edinburgh EH8 9XD, department of Clinical Chemistry, University of Edinburgh, Edinburgh Royal Infirmary, Edinburgh EH3 9YW, UK 3
To whom reprint requests should be sent
Introduction Acquired drug resistance represents a major problem in the treatment of various cancers which are initially responsive to therapy. Such cancers include small cell lung cancer, breast cancer and adenocarcinoma of the ovary. A variety of factors have been discussed in relation to this phenomenon. These include altered tumour vascularity and oxygenation, a selection for intrinsically resistant cells and drug-induced phenotypic changes which result in the over-expression of genes which confer resistance (1). In the latter case various mechanisms have been proposed including increased drug efflux (2), DNA repair (3) and detoxification (4). Glutathione and glutathione-dependent enzymes are known to play a central role in drug detoxification (4-7). In relation to drug resistance, particular attention has been given to reduced glutathione (GSH*) and the glutathione-S-transferases (GST) •Abbreviations: GSH, reduced glutathione; GST, glutathione-S-transferases; 7GCS, 7-glutamylcysteinyl synthetase; 7GT; 7-glutamyltranspeptidase; GPX, glutathione peroxidase; GRD, glutathione reductase; CDNB, l-chloro,2,4-dinitrobenzene; AMC, 7-amino-4-methyl coumarin.
© IRL Press Limited, Oxford, England
Materials and methods All chemicals were purchased from commercial sources and were of the highest grade of purity available. Details of the derivation and general characterization of the cell lines PE01 and PE04 has been described previously (25). The PE04 cell line was more resistant to both cis platinum and chlorambucil and also showed cross resistance to a variety of other chemotherapeutic agents including adriamycin. No difference in the sensitivity of the lines to 5FU was observed. These ovarian cells, and the MCF-7 human breast cancer cell line, were cultured at 37°C, 100% humidity and 5% CO2 in RPMI 1640 supplemented with 10% (v/v) fetal calf serum (FCS),
1283
Downloaded from http://carcin.oxfordjournals.org/ at University of Dundee on November 1, 2012
The regulation of glutathione and various glutathionedependent enzymes has been studied in two ovarian adenocarcinoma cell lines derived from a patient before (PE01) and after (PE04) the onset of drug resistance to m-platinum, chlorambucil and 5-fluorouracil. Reduced glutathione levels were higher in the drug resistant cells (PE04). This could possibly be attributed to a much higher (6.5-fold) 7-glutamyltranspeptidase activity. In addition, glutathione-S-transferase (GST) and glutathione peroxidase were 2.9- and 2.3-fold higher in this cell line. Analysis of the GST subunit composition showed both cell lines contained high levels of the acidic GST and lower concentrations of a basic isozyme. The difference in GST activity between PE01 and PE04 did not appear to be related to the levels of these GST subunits. GSH, glutathione peroxidase and -y-glutamylcysteinyl synthetase were all found to be regulated during the cell cycle, higher levels being detected in logarithmic versus confluent cultures of PE01 and PE04 and MCF7. This did affect some of the differences between PE01 and PE04 and therefore may be a contributing factor to the differential sensitivity of these cells to cytotoxic compounds. The above data provide the first evidence that tumour cells obtained from a patient before and after the onset of drug resistance have significant differences in glutathione-dependent enzyme content.
(4,7-15). Indeed, there are several reports describing increased expression of GST and GSH in drug resistant cell lines (12 — 15). The GST are a multigene family of dimeric proteins which catalyse the conjugation of GSH to a wide variety of reactive electrophiles, including anti-cancer drugs (16). Characterization of the GST enzymes in human tissues has identified three distinct gene families of proteins which can be classified unambiguously by their subunit compositions as acidic, GST YfYf (subunit Mr 24 800), neutral, GST YbYb (subunit Mr 26 700) and basic, GST YaYa (subunit M r 26 000) (17-19). In relation to drug resistance particular emphasis has been placed on the expression of die acidic GST which has been shown to be over-expressed in drug resistant breast cancer cells, as well as in experimental models such as preneoplasia (12,20). However, other GST subunits may also be important; for example, a CHO cell line made resistant to chlorambucil showed a marked elevation in the Ya/Yc subunits (15). Although the role of GST in conferring drug resistance in preneoplasia and cancer chemotherapy is circumstantial, there is strong evidence from other models that they are an important contributing factor (7). In previous reports we have discussed the possibility that the altered expression of GSH and glutathione-dependent enzymes may represent a general protective response to cytotoxic insult (10,21). Other enzymes such as glutathione peroxidase (GPX) and glutathione reductase also protect cells against cytotoxic compounds, particularly those which generate toxic oxygen intermediates such as superoxides and peroxides (e.g. adriamycin) (22—24). The levels of these enzymes in tumour cells may also determine their susceptibility to a variety of cytotoxic drugs. A significant problem in the study of drug resistance is establishing the relevance of in vitro experiments to clinical observations in vivo. In this study we have taken two ovarian adenocarcinoma cell lines, derived from a patient before (PE01) and after the onset of resistance (PE04) to a combination of cisplatinum, chlorambucil and 5-fluorouracil (25) to establish whether differences in GSH-dependent enzymes can be observed and therefore be a potential factor in the clinically observed acquired drug resistance. Glutathione has been shown to be regulated during the cell cycle (26). In view of the altered sensitivity of cells in logaridimic growth to alkylating and chemotherapeutic agents (27,28) we have investigated the regulation of glutadiione-dependent enzymes during the cell division to establish whether this may be a factor in the differential sensitivity to cytotoxic drugs.
A.D.Lewis, J.D.Hayes and C.R.Wolf
Table I. Glutathione levels and glutathione-dependent enzyme activities in logarithmic and confluent cultures of PEOl and PE04 Thymidine incorporation PEOl C L PE04 C L
3805 12236 29577 72682
387 ± ± 1232* ± 6174 ± 12260*
Activity GSH 27.8 42.9 37.4 77.7
± 6.7 ± 3.3** ± 5.2 + + ± 19.4*
7GCS 16.8 23.1 12.2 23.6
± ± ± ±
0.64 5.6** 0.8 4.7*
0.06 0.08 0.39 0.33
± ± ± ±
0.01 0.03 0.04 + 0.07
62.3 62.9 87.4 89.5
GPX
GST
GRD
7GT
± 5.8 ± 5.0 ± 12.3 ± 16.2
74.0 95 215 213
± 9.3 ± 30 ± 15 + ± 39
11.6 38.9 27.0 52.1
± ± ± ±
0.7 1.9* 0.8+ 2.1*
Thymidine incorporation is expressed as cpm per mg protein. GSH concentration is expressed as nmol/mg soluble protein. Enzyme activities are expressed as nmol/min/mg protein with the exception of 7GT which is expressed as nmol/min/106 cells. Other details are given in Materials and methods section. Abbreviations used are 7GCS, 7-glutamylcysteinyl synthetase; 7GT, 7-glutamyltranspeptidase; GRD, glutathione reductase; GST, glutathione-S-transferase measured using CDNB; and GPX, glutathione peroxidase measured using H 2 O 2 . C and L are confluent and logarithmic cultures, respectively. The values are shown as means ± standard deviations from three separate experiments carried out in triplicate. *Significantly different from confluent culture. P < 0.001. "Significantly different from confluent culture. P < 0.01. +Significantly different from PEOl. P < 0.001. ++Significantly different from PEOl. P < 0.01. streptomycin (100 /ig/ml) and penicillin (100 IU/ml). Insulin (2.5 mg/ml) was also added to cultures of PEOl and PE04. Periodic assays for mycoplasma were carried out routinely and were always found to be negative. Flasks were seeded at 2 x 105 or 5 x Iff1 cells/ml to obtain either confluent or logarithmic cultures respectively. Cell synchronisation was achieved by arresting the cells in G0IGl of the cell cycle by reducing the concentration of fetal calf serum from 10% to 0.5% (29). Logarithmic growth was achieved by then increasing the FCS concentration to 10%. Cells were harvested 18 h later.
Glutathione and enzymic assays GSH levels were measured by high performance liquid chromatography following derivitisation with monobromobimane (Calbiochem, Cambridge Science, Cambridge, UK) (31). The stability of the GSH was established by the addition of known amounts of GSH to a cell sample following harvesting. GSH in the samples was found to be stable for up to 1 month if stored at -70°C. GST activities towards the substrates l-chloro-2,4-dinitrobenzene and ethacrynic acid were determined by the method of Habig et al. (32). The peroxidase activity of the basic GST towards cumene hydroperoxide was assessed by measuring the oxidation of NADPH at 340 nm (33). Glutathione reductase and glutathione peroxidase activities were assayed fluorimetrically, as described by Weiss et al. (34). 7-Glutamylcysteinyl synthetase ( T G C S ) , the rate determining enzyme of GSH biosynthesis (35), was measured spectrophotometrically using a coupled reaction involving the measurement of NADH oxidation at 340 nm (36). The paniculate fraction of the cell preparation was taken for the assay of 7GT activity (37) using 7-glutamyl-7-amino-4-methyl coumarin as a substrate (Universal Biological Ltd, Cambridge, UK). All assays were carried out at 37°C. Statistical significance was determined using the Students /-test. Protein estimations were by the method of Lowry et al. (38) using bovine serum albumin as the standard. Western blot analysis Western blots were carried out using a modified version of that described by Towbin et al. (39) as previously described by Adams et al. (40). Following SDS—polyacrylamide gel electrophoresis and transfer to nitrocellulose, filters were washed for two 10 min periods in 50 mM Tris/HCl buffer, pH 7.9 containing 0.15 M NaCl, and 0.05% (v:v) Tween 20 (TBST). Filters were blocked with 3% BSA in TBST for 1 h followed by two 10 min washes with TBST. Specific GST antisera (33,41) (diluted 1:500) was then added for 1 h, followed by four
1284
600 -
500 -
>
400 -
>
IO
