ARCHIVES
Vol.
OF BIOCHEMISTRY
286, No. 2, May
AND
BIOPHYSICS
1, pp. 311-315,199l
Paraquat-Resistant HeLa Cells: Increased Cellular Content of Glutathione Peroxidase J. Krall,’
M. J. Speranza,
and R. E. Lynch2-4
Research Service, VA Medical Center, and Department School of Medicine, Salt Lake City, Utah 84148
Received
June
4, 1990, and in revised
form
December
of Pathology,
Academic
Press,
Inc.
A variety of agents, both cellular and chemical, reduce O2as they act. The products of the reduction of dioxygen, O,, HzOz, and OH’, are potentially cytocidal by virtue i Present address: Cardiology Division, Department of Medicine, University of Utah School of Medicine, Salt Lake City, Utah 84132. * To whom correspondence should be addressed: VA Medical Center (113), 500 Foothill Drive, Salt Lake City, UT 84148. a Supported by the Research Service of the Veteran’s Administration and by grants from the National Institute of Environmental Health Sciences and the National Heart, Lung, and Blood Institute. ’ Clinical Investigator, Veteran’s Administration. COO3-9861/91
Copyright All
rights
$3.00 1991 by of reproduction
0
Academic in any
Press, Inc. form
reserved.
of Utah
17, 1990
Paraquat-resistant HeLa cells were selected and characterized to determine the mechanism(s) of the toxic action of paraquat for cultured mammalian cells. From HeLa cells already selected for resistance to the toxicity of 40 &M paraquat more resistant cells were selected in 90 PM paraquat. In these more resistant cells (PQRHMSO) one adaptation which occurred at the lesser concentration of paraquat, magnification of the cellular content of both the CuZn- and the Mn-containing superoxide dismutases, did not increase further. Instead, the cellular content of catalase and of glutathione peroxidase increased. The increased cellular content of glutathione peroxidase appeared more likely to have produced increased resistance to paraquat than did the augmented content of catalase because, after cultivation in the absence of paraquat, the cells retained both resistance to 90 PM paraquat and the increased content of glutathione peroxidase, while the content of catalase declined. The cellular content of reduced glutathione in the PQRHMSO cells grown in paraquat was diminished in comparison with that in the cells from which they derived. The data suggest that the PQRHMSO cells resisted the toxicity of paraquat by increasing the cellular activity of glutathione peroxidase as a means of detoxifying the HzOz produced by paraquat. Q 1991
University
of their chemical reactivities. In only a few cases,however, is it known whether the ability to reduce O2 is necessary for the cytocidal effects of this class of agents, which includes some cancerocidal drugs ( 1,2), phagocytes (3), and herbicides (4). One prototype of these agents is paraquat, an agent which promotes a flux of 0; within cells by transferring single electrons catalytically from biological reductants to Oz. There is evidence both in prokaryotes and in eukaryotes that the formation of 0; is a necessary part of the cytocidal effect of paraquat. In Escherichiu coli (4), in Salmonella typhimurium (5), and in cultured mammalian cells (6,7) enriched in various ways for superoxide dismutases, resistance to paraquat was observed. Conversely, in E. coli in which the cellular content of both superoxide dismutases was greatly decreased after both genes for superoxide dismutase were mutationally inactivated, sensitivity to paraquat increased (8). The association between the activity of superoxide dismutases and resistance to paraquat was further strengthened in HeLa cells selected for resistance to 40 PM paraquat in which the cellular content of both the CuZn-containing and the Mn-containing superoxide dismutases increased, in a stable and apparently genetically determined manner (7). Evidence which appears to support the contradictory conclusion has also been presented. E. coli in which the cellular content of the Fe-superoxide dismutase was augmented by transcription of a multicopy plasmid were rendered hypersensitive to the toxicity of paraquat (9). Similarly, in Chinese hamster ovary cells in which the cellular content of the human CuZn-superoxide dismutase was greatly magnified, increased sensitivity to the toxicity of paraquat was observed, although in cells enriched to a lesser extent resistance to paraquat occurred (10). Whether 0; itself is the cytotoxin or whether it serves as the precursor of the cytotoxin is unclear. Evidence has been presented that 0, may engender the actual toxin in an iron-dependent manner (11). One way in which this may occur is by the interaction between 0, and Hz02 to 311
312
KRALL.
SPERANZA.
form OH’ in a reaction catalyzed by ionic iron (12-14). If the actual toxin is the product of a bimolecular reaction between 0, and H202, enzymes which scavenge HzOz should also mitigate the toxicity of paraquat. We have isolated paraquat-resistant HeLa cells to determine the adaptations which produced resistance, anticipating that the adaptations identified would indicate whether 0; or H202 or both are necessary components of paraquat’s actions. The data from cells studied after the first increment in concentration of paraquat (from 0 to 40 PM) suggested that 0; was necessary for paraquat’s toxicity (7). We now describe cells isolated in the next increment in concentration of paraquat, from 40 to 90 PM. The data, documenting an increase in the cellular content of glutathione peroxidase, suggest that paraquat can injure cells also as a consequence of its ability to engender H202. MATERIALS
AND
METHODS
Cell culture. HeLa S3 cells were propagated in Dulbecco’s minimal essential medium (DMEM)’ with 7.5% calf serum at 37°C in 8% COP. Cells resistant to 90 pM paraquat (PQRHMSO) were selected in dishes in which cells already made resistant to 40 pM paraquat (PQRHMIO) were plated. A colony which could grow continuously in the presence of 90 pM paraquat was obtained and expanded in DMEM containing 90 gM paraquat in stationary monolayers in loo-mm dishes. Assuys. To obtain lysates for assays of activity of superoxide dismutases, catalase, or glutathione peroxidase cells were removed from monolayers by trypsinization, collected in PBS, sedimented by centrifugation, washed in PBS, and disrupted by sonication. Unlysed cells and organelles were sedimented by centrifugation at 40,OOOg for 20 min at 4°C. Catalase was assayed spectrophotometrically by the catalase-induced loss of absorbance of 19.5 mM H202 at 240 nm at 25°C in 16.5 mM potassium phosphate buffer, pH 7 (15). Glutathione peroxidase was assayed spectrophotometrically at 340 nm with 73 pM H202 as substrate by a method which couples the oxidation of 5 mM reduced glutathione to the reduction of oxidized glutathione at the expense of 285 pM NADPH by glutathione reductase 0.33 U/ml in 50 mM potassium phosphate buffer, pH 7, at 37°C (16). Superoxide dismutase was assayed by its ability to inhibit the O,-mediated reduction of 10 PM cytochrome c when 0, is generated by the enzymatic turnover of xanthine oxidase acting upon 50 pM xanthine in 50 mM potassium phosphate buffer, pH 7.8, lo-’ M EDTA at 25°C (17). Cellular lysates were subjected to electrophoresis in 7.5% nondenaturing acrylamide gels (18) and stained for the activity of superoxide dismutase (19). Quantitation of the proportion of activity in each of the two bands of activity was performed by densitometry of disc gels after staining for activity, with quantitation of the proportion of total activity in each band by weighing the paper upon which each peak of activity was traced after excising the peak with scissors (20). Except where indicated protein was assayed by the method of Lowry et al. (21). Glucose-g-phosphate dehydrogenase was assayed spectrophotometrically by measurement of the increase in absorbance at 340 nm as 67 pM NADP is reduced enzymatically at 25°C in 33 mM Tris-HCl, pH 7.5, with electrons derived from 0.83 mM glucose 6-phosphate (22). Glutathione reductase was assayed by recording the loss of absorbance at 25°C at 340 nm as 100 pM NADPH is oxidized in the presence of 3.3
6 Abbreviations used: DMEM, PQRHM40, HeLa cells resistant cells resistant to 90 pM paraquat;
Dulbecco’s minimal essential medium; to 40 pM paraquat; PQRHMSO, HeLa PBS, phosphate-buffered saline.
AND
LYNCH
mM oxidized glutathione in 50 mM Tris-HCl, pH 8 (23). Glutathione was measured spectrophotometrically (24) in lysates of cultured cells after deproteinization of suspensions of cells with an equal volume of 1% picric acid and derivitization of reduced glutathione in an aliquot of the deproteinized lysate with 2-vinylpyridine (25). Mitochondria were prepared by harvesting cells from monolayers with trypsin in iced PBS by centrifugation. Cells were washed in iced 133 mM NaCl, 5 mM KCl, 0.7 mM NazHPOI, 25 mM Tris-HCl pH 7.5, swollen in 10 mM NaCl, 1.5 mM CaCl,, 10 mM Tris-HCl, pH 7.5, for 10 min on ice; and disrupted with eight strokes of a Dounce homogenizer with the tight pestle. An equal volume of 0.68 M sucrose, 2 mM EDTA, 20 mM Tris-HCl, pH 7.5, was added. Nuclei and undisrupted cells were sedimented by two consecutive sedimentations by centrifugation at 1200g for 10 min at 4°C. Mitochondria were collected at 20,OOOg for 20 min at 4°C and washed in 0.34 M sucrose, 1 mM EDTA, 10 mM Tris-HCl, pH 7.5 (26). After disruption by sonication, debris and unlysed mitochondria were sedimented by centrifugation at 40,OOOg for 20 min at 4”C, and the resulting supernatant fluid was assayed for protein (27) and for glutathione peroxidase. Toxicity ofparaquat. HeLa cells were seeded in 60-mm dishes at 200 cells/dish and allowed to adhere overnight in DMEM with 7.5% calf serum. The medium was replaced with paraquat-containing DMEM and the cells were allowed to form colonies at 37°C in 8% COz. After growth for lo-14 days the colonies were fixed and stained with 0.07% Coomassie blue G250 in 50% methanol, 10% acetic acid (28). Colonies of size greater than 50 cells were enumerated with the aid of a dissecting microscope.
RESULTS
Resistance to the toxicity of paruquat. for resistance after each increment in the of paraquat displayed greater resistance to the cells from which they derived (Table to paraquat was retained when paraquat
TABLE
Resistance to Paraquat Paraquat-Resistant
Cells” WT HeLa PQRHM40 PQRHMSO PQRHMSO’
I
of Wild-Type HeLa S3 Cells and of HeLa Cells Derived Therefrom [Paraquat] L&o
Growth in KM paraquat 0 40 90 0
Cells selected concentration paraquat then I). Resistance was removed
(PM)
Mean
SEM
N
15.8 45.6* IO8 3+*,+**
2.1 8.4 11.5 -
6 5 4 2
112
Note. HeLa cells of the types indicated were inoculated into 60-mm dishes at 200 cells/dish. After the cells were allowed to adhere, the medium was changed to DMEM containing various concentrations of paraquat. After growth in 8% CO2 at 37’C for lo-14 days the medium was removed and the resultant colonies were stained with 0.07% COOmassie blue G250 in 50% methanol/lo% acetic acid. Colonies were enumerated in the dry dishes with the use of a dissecting microscope. ’ WT, wild-type; PQRHM40, paraquat-resistant HeLa cells at 40 pM paraquat; PQRHMSO, HeLa cells selected for resistance to 90 PM paraquat; PQRHMSO”, PQRHMSO cells propagated for >5 months in the absence of paraquat. * P < 0.01 vs wild-type. ** P < 0.001 vs wild-type. *** P < 0.001 vs PQRHM40.
GLUTATHIONE
PEROXIDASE
CONTENT
OF
from the medium and the cells were propagated for several months (Table I). The retention of resistance to paraquat many generations after paraquat was removed from the medium indicates that the adaptation is likely to be genetic. Consequently, adaptations in these cells which are found after growth in medium lacking paraquat may have caused resistance; other adaptations which vanish while resistance to paraquat is retained are not likely to have caused resistance but may instead result from effects of growth under stressful conditions. Cellular content of superoxide dismutases, catalase, and glutathione peroxidase. In the cells from which the PQRHMSO cells derived, the PQRHM40, the cellular content of superoxide dismutases had increased (6, 7). However, in the PQRHMSO cells the content of superoxide dismutases did not increase further (Table II). The activity of catalase increased with the degree of resistance to paraquat (Table III). The activity of catalase after further assays is now significantly greater in the PQRHM40 cells than in the wild-type HeLa cells. The interpretation of results of assays of catalase is difficult because of the large variation in activity of catalase even in wild-type HeLa cells. The activity increases in cells grown to a high density and may fluctuate during the cell cycle (R. E. Lynch, unpublished observations). The change in the cellular activity of catalase in the PQRHMSO cells after cultivation in the absence of paraquat suggests that the increased cellular content of catalase was not genetically determined and further, since resistance to the toxicity of paraquat was retained, that enrichment for catalase was not the cause of resistance. In support of the lack of a protective effect of catalase
TABLE
II
Activity of Superoxide Dismutasesin Paraquat-Resistant HeLa Cells Superoxide
dismutase
(U/mg
MnSOD
Resistant
protein)
CuZnSOD
to-PM
Cells WT HeLa PQRHM40 PQRHMSO
paraquat (0) 40 90
Mean
SEM
N
Mean
SEM
N
1.36 3.00’ 3.44*
0.35 1.13 1.20
15 7 4
4.48 6.90* 5.02
0.34 0.47 1.08
15 7 4
Note. HeLa cells, either paraquat-resistant or wild type, were removed from monolayers by treatment with trypsin after growth in DMEM with 7.5% calf serum containing the concentration of paraquat indicated. The cells were sedimented by centrifugation, resuspended in PBS, resedimented by centrifugation, resuspended again, and disrupted by sonication. The unlysed cells and organelles were sedimented at 40,OOOg for 20 min at 4’C and the supernatant fluids were assayed for superoxide dismutase and for protein. * P vs WT HeLa of c 0.01. Statistical comparison of the activity of superoxide dismutases in PQRHMSO cells with the activity in PQRHM40 cells yielded nonsignificant P values.
PARAQUAT-RESISTANT
HeLa TABLE
Activities
of
Wild-Type
WT HeLa PQRHM40 PQRHMSO PQRHMSO”
-FM paraquat S3
III
Glutathione Peroxidase and of Catalase in and in Paraquat-Resistant HeLa Cells
Growth Cells”
313
CELLS
0 40 90 0
in
Glutathione peroxidase (nmol/min/mg)
Catalase (fimol/min/mg) Mean
SEM
N
Mean
SEM
N
23.1 35.5* 54.2: 32.4
2.1 4.0 13.1 5.9
23 12 12 9
104 100 140** 151**
15 5 15 27
6 6 6 6
Note. Cells of the indicated types were grown in DMEM with 7.5% calf serum with the concentrations of paraquat indicated. The cells were collected by trypsinization, transferred to PBS, sedimented by centrifugation, resuspended in PBS, and resedimented. The washed pellet was suspended in a small volume of PBS and sonicated. The supernatant fluid obtained after centrifugation of the lysate for 20 min at 40,OOOg was assayed for catalase or for glutathione peroxidase. ’ WT, wild-type; PQRHM40, paraquat-resistant HeLa cells at 40 pM paraquat; PQRHMSO, cells grown continuously in the presence of 90 pM paraquat; PQRHMSO”, cells grown continuously for >5 months in the absence of paraquat after selection for resistance to 90 pM paraquat. * P < 0.01 vs WT HeLa S3 cells. ** P < 0.05 vs PQRHM40.
from the toxicity of paraquat, we have observed no resistance to paraquat in L cells enriched for catalase more than 50-fold by transcription of the transfected, integrated cDNA for human catalase (R. E. Lynch, in preparation). In contrast the cellular content of glutathione peroxidase did remain increased in PQRHMSO cells after propagation in the absence of paraquat while resistance to 90 PM paraquat was retained, suggesting that enrichment for glutathione peroxidase is more likely to have caused the increased resistance. Mitochondrial glutathione peroxidase was not increased in the PQRHMSO cells; the activity in lysates of washed mitochondria was 274 f. 49.5 nmol/m/ mg in wild-type HeLa S3 cells, 371+ 39.5 nmol/m/mg in PQRHM40 cells, 297 ? 38.1 nmol/m/mg in PQRHMSO cells. Resistance of PQRHMSO cells to other agents. It was possible that the PQRHMSO cells were resistant to all toxins. That this is not the case is shown by the lack of resistance to two agents which are highly likely to act in distinct ways, G-418 an aminoglycoside cytotoxic for mammalian cells (29) and cadmium. The concentrations at which 50% of cells were killed were similar for G-418 (525, 490, and 562 pg/ml) and for cadmium (0.19, 0.17, and 0.16 PM) in wild-type, PQRHM40, and PQRHMSO cells, respectively. Assays of glutathione, glutathione reductase, andglucose6-phosphate dehydrogenase. To assessthe possibility that other components of the pathway leading to the enzymatic action of glutathione peroxidase could account for the greater resistance to paraquat of the PQRHMSO
314
KRALL.
SPERANZA.
cells, the activities of glucose-g-phosphate dehydrogenase and glutathione reductase were assayed (Table IV). The cellular content of glutathione was assayed as well. Neither the activity of glutathione reductase nor that of glucose-6-phosphate dehydrogenase was increased in the PQRHSO cells. The cellular content of reduced glutathione was moderately diminished, a finding consistent with the greater utilization of reduced glutathione by these cells in the enzymatic turnover of glutathione peroxidase, although a decreased rate of synthesis and/or reduction is not excluded. DISCUSSION The cells described herein (PQRHMSO) were isolated for resistance to 90 PM paraquat from other cells (PQRHM40) already selected for resistance to 40 PM paraquat (7). One adaptation made by cells isolated for resistance to 40 PM paraquat, augmentation of the cellular content of both superoxide dismutases, was not continued in the PQRHMSO cells. Instead, the cellular content of glutathione peroxidase, and possibly that of catalase, increased. The increased cellular content of glutathione peroxidase appeared more likely to have caused increased resistance, because this trait and resistance to paraquat remained associated in PQRHMSO cells after cultivation in the absence of paraquat, whereas the content of catalase regressed to the activity in the cells from which the PQRHMSO cells derived. The retention of increased cellular content of glutathione peroxidase and of increased resistance to paraquat in the PQRHMSO cells after paraquat was removed from the medium for several generations further suggests that enrichment for glutathione peroxidase was a stable genetic adaptation rather than a nonspecific response to growth under stressful conditions. The inference that glutathione peroxidase protects cells containing it from the toxicity of paraquat is strengthened by a further observation. Increased sensitivity to paraquat was observed (R. E. Lynch, in preparation) in DG44 cells depleted of glutathione peroxidase by more than 70% by antisense RNA transcribed from the transfected cDNA for bovine glutathione peroxidase; the concentration of paraquat in which 50% survival occurred was 39 ? 5.4 PM for the untransfected cells and 23.2 + 3.8 PM for the cells in which the content of glutathione peroxidase was markedly decreased. Magnified cellular content of glutathione peroxidase has also been observed in two other kinds of paraquatresistant cells. The cellular content of glutathione peroxidase was increased in a subset of paraquat-resistant HL60 cells in which the activity of superoxide dismutase did not increase (30) and also in paraquat-resistant NIH/ 3T3 cells which had been enriched for the human CuZnsuperoxide dismutase by transcription of the cloned cDNA for the human CuZn-superoxide dismutase (31). The means by which the cellular content of glutathione per-
AND
LYNCH TABLE
IV
Glucose-6-Phosphate Dehydrogenase, Glutathione Reductase, and Reduced Glutathione in Wild-Type and Paraquat-Resistant HeLa Cells Glutathione reductase (pmol/min/ md Cells Wild-type PQRHM40 PQRHMSO
S3
Note. Lysates to Table III and and for glutathione Lysates of HeLa for glutathione glutathione was
GGPD (amol/ min/mg)
Glutathione (nmol/mg protein)
Mean
SD
Mean
SD
Mean
SD
43.6 54.1 50.7
4.0 1.4 0.2
171 198 139
7.8 6.1 1.2
129 126 a2
15.5 25.4 15.6
of HeLa cells were prepared as described in the legend assayed for glucose-g-phosphate dehydrogenase (GGPD) reductase as described under Materials and Methods. cells deproteinized with 1% picric acid were assayed as described under Materials and Methods. Oxidized not detectable.
oxidase was magnified in these two instances is not yet clear although enrichment for glutathione peroxidase in cells transcribing the transfected cDNA for human CuZnsuperoxide dismutase had been observed (32). The modest magnitude of the enrichment for glutathione peroxidase may seem insufficient to account for resistance to paraquat. HeLa cells appear to be relatively rich in glutathione peroxidase in comparison with other lines of cultured cells. This characteristic could permit the acquisition of resistance to paraquat by magnification of the cellular content of superoxide dismutases and catalase without a change in the cellular content of glutathione peroxidase when wild-type HeLa cells became resistant to 40 /IM paraquat. Constitutive enrichment for glutathione peroxidase may also enable PQRHM40 cells to resist the next increment in the concentration of paraquat (from 40 to 90 PM) by a modest increase in cellular content of glutathione peroxidase. The unusual structural feature of glutathione peroxidase, a single selenocysteine per subunit, may also impose a limit on the magnitude to which the cellular content can be augmented. The codon for phosphoserine, the amino acid which serves as a precursor of selenocysteine, is a termination codon (33, 34) recognized by a unique tRNA (35). The cellular abundance of this unique tRNA may be sufficient only for synthesis of the usual cellular content of glutathione peroxidase but not for substantially greater rates of synthesis. It may, as a consequence, not be possible to enrich cells markedly for glutathione peroxidase without first increasing the cellular content of the phosphoserine-inserting, opal-suppressing tRNA. The association between increased resistance to paraquat and enrichment for glutathione peroxidase suggests that HzOz, in addition to 0;) mediates the toxicity of paraquat. If 0; formed within HeLa cells escapes into
GLUTATHIONE
PEROXIDASE
CONTENT
the medium, enrichment for superoxide dismutases in the PQRHM40 cells will augment the concentration of HzOz within cells by capturing 0, before it can escape and converting it to HzOz. Indeed, the ready transit of 0; through one cellular membrane of eukaryotic cells, that of the erythrocyte, has been demonstrated (36). As a result of the increased cellular content of superoxide dismutase in the PQRHM40 cells the dependency of the toxic action of paraquat on 0; would diminish at the cost of an enhanced flux of HzOz. Since HzOz is itself cytotoxic, survival of the PQRHM40 cells in the augmented flux of H202 generated by the increase in concentration of paraquat from 40 to 90 PM necessitates blunting the cytotoxic effects of Hz02. Although other adaptations could also confer resistance, the PQRHMSO cells have magnified the cellular content of glutathione peroxidase. The mechanism by which these HeLa cells have increased the cellular content of superoxide dismutases and catalase in the first step of resistance and of glutathione peroxidase in the second step is not clear. No amplification of genes for superoxide dismutases or catalase was detected in PQRHM40 or PQRHMSO cells in Southern blots probed with fragments of cDNAs for the human CuZnsuperoxide dismutase (37) for the human Mn-superoxide dismutase, and for human catalase (38,39) after digestion of genomic DNA with P&I or with BgZII. Although it seems more likely that distinct adaptations occurred in the PQRHM40 and and the PQRHMSO cells because the activities of distinct sets of enzymes increased in each step, it is conceivable that a single adaptation which occurred in the PQRHM40 cells was magnified further in the PQRHMSO cells. How HeLa cells have increased cellular content of these enzymes remains to be determined. REFERENCES 1. Berlin,
V., and Haseltine,
W. A. (1981)
J. Biol.
Chem.
266,4747-
OF
Chem.
3. Babior, Invest.
L., Zweier, J. L., Levy, 260,6820-6826. B. M., Kipnes, 52, 741-744.
4. Hassan,
H. M.,
A., and Myers,
C. E. (1985)
R. S., and Curnutte,
J. Biol.
J. T. (1973)
12. Beauchamp, 13. Haber,
I. (1978)
J. Biol.
Chem.
253,8143-
5. Moody, C. S., and Hassan, 79,2855-2859.
H. M. (1982)
6. Bagley, A. C., Krall, J., and Lynch, Sci. USA 83,3189-3193. 7. Krall, J., Bagley, A. C., Mullenbach, Lynch, R. E. (1988) J. Biol. Chem.
8. Carlioz,
A., and Touati,
9. Scott, Chem.
M. D., Meshnick, 262,3640-3645.
10. Elroy-Stein, 615-622.
O., Bernstein,
D. (1986)
Proc. Natl.
R. E. (1986)
Acad. Proc.
G. T., Hallewell, 263,1910-1914. EMBO
N&l.
Acad.
R. A., and
J. W. (1987) Y. (1986)
F., and Weiss,
14. Halliwell,
EMBO
J. Biol. J. 5.
I. (1970)
J. Biol.
Chem.
M. (1986) 245,4641-
J. (1934)
Proc.
R. Sot. London
Ser. A
147,
B., and Gutteridge,
J. M. C. (1986)
Arch. Biochem.
Biophys.
246,501-514. 15. Beers, R. F., and Sizer, I. W. (1952) J. Biol. Chem. 195, 133-140. 16. Paglia, D. E., and Valentine, W. N. (1967) J. Lab. Clin. Med. 70, 158-169.
17. McCord, 6055. 18. Malinowski,
J. M., and Fridovich, D. P., and Fridovich,
I. (1969)
J. Biol.
I. (1979)
Chem.
Biochemistry
244,
6049-
18, 237-
244. 19. Beauchamp,
C., and Fridovich,
I. (1971)
Anal.
Biochem.
44,
276-
267. 20. Hassan, H. M., and Fridovich, I. (1977) J. Bacterial. 129, 15741583. 21. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193,265-275. 22. Langdon, R. G. (1966) in Methods in Enzymology (Wood, W. A., 23. 24. 25. 26.
Ed.), Vol. 9, pp. 126-131, Academic Press, New York. Beutler, E. (1969) J. Clin. Invest. 48, 1957-1966. Tietze, F. (1969) Anal. Biochem. 27, 502-522. Griffith, 0. W. (1980) Anal. Biochem. 106, 207-212. Hauswirth, W. W., Lim, L. O., Dujon, B., and Turner, G. (1987) in Mitochondria: A Practical Approach (Darley-Usmar, V. M., Rickwood, D., and Wilson, M. T., Eds.), pp. 176 and 177, IRL Press, Oxford/Washington, DC. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254.
27. 28. Raetz,
C. R. H., Wermuth, M. M., McIntyre, T. M., Esko, J. D., and Wing, D. C. (1982) Proc. Natl. Acad. Sci. USA 79,3223-3227. 29. Colbere-Garapin, F., Horodniceanu, F., Kourilsky, P., and Garapin, A-C. (1981) J. Mol. Biol. 150, 1-14. 30. Kelner, M. J., and Bagnell, R. (1990) Free Radicals Biol. Med. 9, 149-153. 31. Kelner, M. J., and Bagnell, R. (1990) J. Biol. Chem. 269, 10872-
10875. 32. Ceballos,
I., Delabar, J. M., Nicole, A., Lynch, R. E., Hallewell, R. A., Kamoun, P., and Sinet, P. M. (1988) Biochim. Biophys. Actu
949,58-64. 33. Chambers, I., Frampton,
35. 36.
J., Goldfarb, P., Affara, N., McBain, W., and Harrison, P. R. (1986) EMBO J. 5, 1221-1227. Mullenbach, G. T., Tabrizi, A., Irvine, B. D., Bell, G. I., and Hallewell, R. A. (1987) Nucl. Acids Res. 16,5484. Lee, B. J., de la Pena, P., Tobian, J. A., Zasloff, M., and Hatfield, D. (1987) Proc. Natl. Acad. Sci. USA 84,6384-6388. Lynch, R. E., and Fridovich, I. (1978) J. Biol. Chem. 253, 2838-
2845. 37. Hallewell,
R. A., Masiarz, F. R., Najarian, R. C., Puma, J. P., Quiroga, M. R., Randolph, A., Sanchez-Pescador, R., Scandella, C. J., Smith, B., Steimer, K. S., and Mullenbach, G. T. (1985) Nucl. Acids Res.
13,2017-2034. 36. Korneluk, R. G., Quan,
J. 6,623-630.
S. R., and Eaton, Y., and Groner,
Sci. USA
J., and Chevion,
332-351.
34.
and Fridovicb,
C., and Fridovich,
315
CELLS
4646.
J. Clin.
8148.
HeLa
11. Korbashi, P., Kohen, R., Katzhendler, J. Biol. Chem. 261,12471-12476.
4756. 2. Gianni,
PARAQUAT-RESISTANT
F., Lewis, W. H., Guise, K. S., Willard, H. F., Holmes, M. T., and Gravel, R. A. (1984) J. Biol. Chem. 259, 13819-13823. 39. Quan, F., Korneluk, R. G., Tropak, M. B., and Gravel, R. A. (1986) Nucl. Acids Res. 14, 5321-5335.
Vol.
OF BIOCHEMISTRY
286, No. 2, May
AND
BIOPHYSICS
1, pp. 311-315,199l
Paraquat-Resistant HeLa Cells: Increased Cellular Content of Glutathione Peroxidase J. Krall,’
M. J. Speranza,
and R. E. Lynch2-4
Research Service, VA Medical Center, and Department School of Medicine, Salt Lake City, Utah 84148
Received
June
4, 1990, and in revised
form
December
of Pathology,
Academic
Press,
Inc.
A variety of agents, both cellular and chemical, reduce O2as they act. The products of the reduction of dioxygen, O,, HzOz, and OH’, are potentially cytocidal by virtue i Present address: Cardiology Division, Department of Medicine, University of Utah School of Medicine, Salt Lake City, Utah 84132. * To whom correspondence should be addressed: VA Medical Center (113), 500 Foothill Drive, Salt Lake City, UT 84148. a Supported by the Research Service of the Veteran’s Administration and by grants from the National Institute of Environmental Health Sciences and the National Heart, Lung, and Blood Institute. ’ Clinical Investigator, Veteran’s Administration. COO3-9861/91
Copyright All
rights
$3.00 1991 by of reproduction
0
Academic in any
Press, Inc. form
reserved.
of Utah
17, 1990
Paraquat-resistant HeLa cells were selected and characterized to determine the mechanism(s) of the toxic action of paraquat for cultured mammalian cells. From HeLa cells already selected for resistance to the toxicity of 40 &M paraquat more resistant cells were selected in 90 PM paraquat. In these more resistant cells (PQRHMSO) one adaptation which occurred at the lesser concentration of paraquat, magnification of the cellular content of both the CuZn- and the Mn-containing superoxide dismutases, did not increase further. Instead, the cellular content of catalase and of glutathione peroxidase increased. The increased cellular content of glutathione peroxidase appeared more likely to have produced increased resistance to paraquat than did the augmented content of catalase because, after cultivation in the absence of paraquat, the cells retained both resistance to 90 PM paraquat and the increased content of glutathione peroxidase, while the content of catalase declined. The cellular content of reduced glutathione in the PQRHMSO cells grown in paraquat was diminished in comparison with that in the cells from which they derived. The data suggest that the PQRHMSO cells resisted the toxicity of paraquat by increasing the cellular activity of glutathione peroxidase as a means of detoxifying the HzOz produced by paraquat. Q 1991
University
of their chemical reactivities. In only a few cases,however, is it known whether the ability to reduce O2 is necessary for the cytocidal effects of this class of agents, which includes some cancerocidal drugs ( 1,2), phagocytes (3), and herbicides (4). One prototype of these agents is paraquat, an agent which promotes a flux of 0; within cells by transferring single electrons catalytically from biological reductants to Oz. There is evidence both in prokaryotes and in eukaryotes that the formation of 0; is a necessary part of the cytocidal effect of paraquat. In Escherichiu coli (4), in Salmonella typhimurium (5), and in cultured mammalian cells (6,7) enriched in various ways for superoxide dismutases, resistance to paraquat was observed. Conversely, in E. coli in which the cellular content of both superoxide dismutases was greatly decreased after both genes for superoxide dismutase were mutationally inactivated, sensitivity to paraquat increased (8). The association between the activity of superoxide dismutases and resistance to paraquat was further strengthened in HeLa cells selected for resistance to 40 PM paraquat in which the cellular content of both the CuZn-containing and the Mn-containing superoxide dismutases increased, in a stable and apparently genetically determined manner (7). Evidence which appears to support the contradictory conclusion has also been presented. E. coli in which the cellular content of the Fe-superoxide dismutase was augmented by transcription of a multicopy plasmid were rendered hypersensitive to the toxicity of paraquat (9). Similarly, in Chinese hamster ovary cells in which the cellular content of the human CuZn-superoxide dismutase was greatly magnified, increased sensitivity to the toxicity of paraquat was observed, although in cells enriched to a lesser extent resistance to paraquat occurred (10). Whether 0; itself is the cytotoxin or whether it serves as the precursor of the cytotoxin is unclear. Evidence has been presented that 0, may engender the actual toxin in an iron-dependent manner (11). One way in which this may occur is by the interaction between 0, and Hz02 to 311
312
KRALL.
SPERANZA.
form OH’ in a reaction catalyzed by ionic iron (12-14). If the actual toxin is the product of a bimolecular reaction between 0, and H202, enzymes which scavenge HzOz should also mitigate the toxicity of paraquat. We have isolated paraquat-resistant HeLa cells to determine the adaptations which produced resistance, anticipating that the adaptations identified would indicate whether 0; or H202 or both are necessary components of paraquat’s actions. The data from cells studied after the first increment in concentration of paraquat (from 0 to 40 PM) suggested that 0; was necessary for paraquat’s toxicity (7). We now describe cells isolated in the next increment in concentration of paraquat, from 40 to 90 PM. The data, documenting an increase in the cellular content of glutathione peroxidase, suggest that paraquat can injure cells also as a consequence of its ability to engender H202. MATERIALS
AND
METHODS
Cell culture. HeLa S3 cells were propagated in Dulbecco’s minimal essential medium (DMEM)’ with 7.5% calf serum at 37°C in 8% COP. Cells resistant to 90 pM paraquat (PQRHMSO) were selected in dishes in which cells already made resistant to 40 pM paraquat (PQRHMIO) were plated. A colony which could grow continuously in the presence of 90 pM paraquat was obtained and expanded in DMEM containing 90 gM paraquat in stationary monolayers in loo-mm dishes. Assuys. To obtain lysates for assays of activity of superoxide dismutases, catalase, or glutathione peroxidase cells were removed from monolayers by trypsinization, collected in PBS, sedimented by centrifugation, washed in PBS, and disrupted by sonication. Unlysed cells and organelles were sedimented by centrifugation at 40,OOOg for 20 min at 4°C. Catalase was assayed spectrophotometrically by the catalase-induced loss of absorbance of 19.5 mM H202 at 240 nm at 25°C in 16.5 mM potassium phosphate buffer, pH 7 (15). Glutathione peroxidase was assayed spectrophotometrically at 340 nm with 73 pM H202 as substrate by a method which couples the oxidation of 5 mM reduced glutathione to the reduction of oxidized glutathione at the expense of 285 pM NADPH by glutathione reductase 0.33 U/ml in 50 mM potassium phosphate buffer, pH 7, at 37°C (16). Superoxide dismutase was assayed by its ability to inhibit the O,-mediated reduction of 10 PM cytochrome c when 0, is generated by the enzymatic turnover of xanthine oxidase acting upon 50 pM xanthine in 50 mM potassium phosphate buffer, pH 7.8, lo-’ M EDTA at 25°C (17). Cellular lysates were subjected to electrophoresis in 7.5% nondenaturing acrylamide gels (18) and stained for the activity of superoxide dismutase (19). Quantitation of the proportion of activity in each of the two bands of activity was performed by densitometry of disc gels after staining for activity, with quantitation of the proportion of total activity in each band by weighing the paper upon which each peak of activity was traced after excising the peak with scissors (20). Except where indicated protein was assayed by the method of Lowry et al. (21). Glucose-g-phosphate dehydrogenase was assayed spectrophotometrically by measurement of the increase in absorbance at 340 nm as 67 pM NADP is reduced enzymatically at 25°C in 33 mM Tris-HCl, pH 7.5, with electrons derived from 0.83 mM glucose 6-phosphate (22). Glutathione reductase was assayed by recording the loss of absorbance at 25°C at 340 nm as 100 pM NADPH is oxidized in the presence of 3.3
6 Abbreviations used: DMEM, PQRHM40, HeLa cells resistant cells resistant to 90 pM paraquat;
Dulbecco’s minimal essential medium; to 40 pM paraquat; PQRHMSO, HeLa PBS, phosphate-buffered saline.
AND
LYNCH
mM oxidized glutathione in 50 mM Tris-HCl, pH 8 (23). Glutathione was measured spectrophotometrically (24) in lysates of cultured cells after deproteinization of suspensions of cells with an equal volume of 1% picric acid and derivitization of reduced glutathione in an aliquot of the deproteinized lysate with 2-vinylpyridine (25). Mitochondria were prepared by harvesting cells from monolayers with trypsin in iced PBS by centrifugation. Cells were washed in iced 133 mM NaCl, 5 mM KCl, 0.7 mM NazHPOI, 25 mM Tris-HCl pH 7.5, swollen in 10 mM NaCl, 1.5 mM CaCl,, 10 mM Tris-HCl, pH 7.5, for 10 min on ice; and disrupted with eight strokes of a Dounce homogenizer with the tight pestle. An equal volume of 0.68 M sucrose, 2 mM EDTA, 20 mM Tris-HCl, pH 7.5, was added. Nuclei and undisrupted cells were sedimented by two consecutive sedimentations by centrifugation at 1200g for 10 min at 4°C. Mitochondria were collected at 20,OOOg for 20 min at 4°C and washed in 0.34 M sucrose, 1 mM EDTA, 10 mM Tris-HCl, pH 7.5 (26). After disruption by sonication, debris and unlysed mitochondria were sedimented by centrifugation at 40,OOOg for 20 min at 4”C, and the resulting supernatant fluid was assayed for protein (27) and for glutathione peroxidase. Toxicity ofparaquat. HeLa cells were seeded in 60-mm dishes at 200 cells/dish and allowed to adhere overnight in DMEM with 7.5% calf serum. The medium was replaced with paraquat-containing DMEM and the cells were allowed to form colonies at 37°C in 8% COz. After growth for lo-14 days the colonies were fixed and stained with 0.07% Coomassie blue G250 in 50% methanol, 10% acetic acid (28). Colonies of size greater than 50 cells were enumerated with the aid of a dissecting microscope.
RESULTS
Resistance to the toxicity of paruquat. for resistance after each increment in the of paraquat displayed greater resistance to the cells from which they derived (Table to paraquat was retained when paraquat
TABLE
Resistance to Paraquat Paraquat-Resistant
Cells” WT HeLa PQRHM40 PQRHMSO PQRHMSO’
I
of Wild-Type HeLa S3 Cells and of HeLa Cells Derived Therefrom [Paraquat] L&o
Growth in KM paraquat 0 40 90 0
Cells selected concentration paraquat then I). Resistance was removed
(PM)
Mean
SEM
N
15.8 45.6* IO8 3+*,+**
2.1 8.4 11.5 -
6 5 4 2
112
Note. HeLa cells of the types indicated were inoculated into 60-mm dishes at 200 cells/dish. After the cells were allowed to adhere, the medium was changed to DMEM containing various concentrations of paraquat. After growth in 8% CO2 at 37’C for lo-14 days the medium was removed and the resultant colonies were stained with 0.07% COOmassie blue G250 in 50% methanol/lo% acetic acid. Colonies were enumerated in the dry dishes with the use of a dissecting microscope. ’ WT, wild-type; PQRHM40, paraquat-resistant HeLa cells at 40 pM paraquat; PQRHMSO, HeLa cells selected for resistance to 90 PM paraquat; PQRHMSO”, PQRHMSO cells propagated for >5 months in the absence of paraquat. * P < 0.01 vs wild-type. ** P < 0.001 vs wild-type. *** P < 0.001 vs PQRHM40.
GLUTATHIONE
PEROXIDASE
CONTENT
OF
from the medium and the cells were propagated for several months (Table I). The retention of resistance to paraquat many generations after paraquat was removed from the medium indicates that the adaptation is likely to be genetic. Consequently, adaptations in these cells which are found after growth in medium lacking paraquat may have caused resistance; other adaptations which vanish while resistance to paraquat is retained are not likely to have caused resistance but may instead result from effects of growth under stressful conditions. Cellular content of superoxide dismutases, catalase, and glutathione peroxidase. In the cells from which the PQRHMSO cells derived, the PQRHM40, the cellular content of superoxide dismutases had increased (6, 7). However, in the PQRHMSO cells the content of superoxide dismutases did not increase further (Table II). The activity of catalase increased with the degree of resistance to paraquat (Table III). The activity of catalase after further assays is now significantly greater in the PQRHM40 cells than in the wild-type HeLa cells. The interpretation of results of assays of catalase is difficult because of the large variation in activity of catalase even in wild-type HeLa cells. The activity increases in cells grown to a high density and may fluctuate during the cell cycle (R. E. Lynch, unpublished observations). The change in the cellular activity of catalase in the PQRHMSO cells after cultivation in the absence of paraquat suggests that the increased cellular content of catalase was not genetically determined and further, since resistance to the toxicity of paraquat was retained, that enrichment for catalase was not the cause of resistance. In support of the lack of a protective effect of catalase
TABLE
II
Activity of Superoxide Dismutasesin Paraquat-Resistant HeLa Cells Superoxide
dismutase
(U/mg
MnSOD
Resistant
protein)
CuZnSOD
to-PM
Cells WT HeLa PQRHM40 PQRHMSO
paraquat (0) 40 90
Mean
SEM
N
Mean
SEM
N
1.36 3.00’ 3.44*
0.35 1.13 1.20
15 7 4
4.48 6.90* 5.02
0.34 0.47 1.08
15 7 4
Note. HeLa cells, either paraquat-resistant or wild type, were removed from monolayers by treatment with trypsin after growth in DMEM with 7.5% calf serum containing the concentration of paraquat indicated. The cells were sedimented by centrifugation, resuspended in PBS, resedimented by centrifugation, resuspended again, and disrupted by sonication. The unlysed cells and organelles were sedimented at 40,OOOg for 20 min at 4’C and the supernatant fluids were assayed for superoxide dismutase and for protein. * P vs WT HeLa of c 0.01. Statistical comparison of the activity of superoxide dismutases in PQRHMSO cells with the activity in PQRHM40 cells yielded nonsignificant P values.
PARAQUAT-RESISTANT
HeLa TABLE
Activities
of
Wild-Type
WT HeLa PQRHM40 PQRHMSO PQRHMSO”
-FM paraquat S3
III
Glutathione Peroxidase and of Catalase in and in Paraquat-Resistant HeLa Cells
Growth Cells”
313
CELLS
0 40 90 0
in
Glutathione peroxidase (nmol/min/mg)
Catalase (fimol/min/mg) Mean
SEM
N
Mean
SEM
N
23.1 35.5* 54.2: 32.4
2.1 4.0 13.1 5.9
23 12 12 9
104 100 140** 151**
15 5 15 27
6 6 6 6
Note. Cells of the indicated types were grown in DMEM with 7.5% calf serum with the concentrations of paraquat indicated. The cells were collected by trypsinization, transferred to PBS, sedimented by centrifugation, resuspended in PBS, and resedimented. The washed pellet was suspended in a small volume of PBS and sonicated. The supernatant fluid obtained after centrifugation of the lysate for 20 min at 40,OOOg was assayed for catalase or for glutathione peroxidase. ’ WT, wild-type; PQRHM40, paraquat-resistant HeLa cells at 40 pM paraquat; PQRHMSO, cells grown continuously in the presence of 90 pM paraquat; PQRHMSO”, cells grown continuously for >5 months in the absence of paraquat after selection for resistance to 90 pM paraquat. * P < 0.01 vs WT HeLa S3 cells. ** P < 0.05 vs PQRHM40.
from the toxicity of paraquat, we have observed no resistance to paraquat in L cells enriched for catalase more than 50-fold by transcription of the transfected, integrated cDNA for human catalase (R. E. Lynch, in preparation). In contrast the cellular content of glutathione peroxidase did remain increased in PQRHMSO cells after propagation in the absence of paraquat while resistance to 90 PM paraquat was retained, suggesting that enrichment for glutathione peroxidase is more likely to have caused the increased resistance. Mitochondrial glutathione peroxidase was not increased in the PQRHMSO cells; the activity in lysates of washed mitochondria was 274 f. 49.5 nmol/m/ mg in wild-type HeLa S3 cells, 371+ 39.5 nmol/m/mg in PQRHM40 cells, 297 ? 38.1 nmol/m/mg in PQRHMSO cells. Resistance of PQRHMSO cells to other agents. It was possible that the PQRHMSO cells were resistant to all toxins. That this is not the case is shown by the lack of resistance to two agents which are highly likely to act in distinct ways, G-418 an aminoglycoside cytotoxic for mammalian cells (29) and cadmium. The concentrations at which 50% of cells were killed were similar for G-418 (525, 490, and 562 pg/ml) and for cadmium (0.19, 0.17, and 0.16 PM) in wild-type, PQRHM40, and PQRHMSO cells, respectively. Assays of glutathione, glutathione reductase, andglucose6-phosphate dehydrogenase. To assessthe possibility that other components of the pathway leading to the enzymatic action of glutathione peroxidase could account for the greater resistance to paraquat of the PQRHMSO
314
KRALL.
SPERANZA.
cells, the activities of glucose-g-phosphate dehydrogenase and glutathione reductase were assayed (Table IV). The cellular content of glutathione was assayed as well. Neither the activity of glutathione reductase nor that of glucose-6-phosphate dehydrogenase was increased in the PQRHSO cells. The cellular content of reduced glutathione was moderately diminished, a finding consistent with the greater utilization of reduced glutathione by these cells in the enzymatic turnover of glutathione peroxidase, although a decreased rate of synthesis and/or reduction is not excluded. DISCUSSION The cells described herein (PQRHMSO) were isolated for resistance to 90 PM paraquat from other cells (PQRHM40) already selected for resistance to 40 PM paraquat (7). One adaptation made by cells isolated for resistance to 40 PM paraquat, augmentation of the cellular content of both superoxide dismutases, was not continued in the PQRHMSO cells. Instead, the cellular content of glutathione peroxidase, and possibly that of catalase, increased. The increased cellular content of glutathione peroxidase appeared more likely to have caused increased resistance, because this trait and resistance to paraquat remained associated in PQRHMSO cells after cultivation in the absence of paraquat, whereas the content of catalase regressed to the activity in the cells from which the PQRHMSO cells derived. The retention of increased cellular content of glutathione peroxidase and of increased resistance to paraquat in the PQRHMSO cells after paraquat was removed from the medium for several generations further suggests that enrichment for glutathione peroxidase was a stable genetic adaptation rather than a nonspecific response to growth under stressful conditions. The inference that glutathione peroxidase protects cells containing it from the toxicity of paraquat is strengthened by a further observation. Increased sensitivity to paraquat was observed (R. E. Lynch, in preparation) in DG44 cells depleted of glutathione peroxidase by more than 70% by antisense RNA transcribed from the transfected cDNA for bovine glutathione peroxidase; the concentration of paraquat in which 50% survival occurred was 39 ? 5.4 PM for the untransfected cells and 23.2 + 3.8 PM for the cells in which the content of glutathione peroxidase was markedly decreased. Magnified cellular content of glutathione peroxidase has also been observed in two other kinds of paraquatresistant cells. The cellular content of glutathione peroxidase was increased in a subset of paraquat-resistant HL60 cells in which the activity of superoxide dismutase did not increase (30) and also in paraquat-resistant NIH/ 3T3 cells which had been enriched for the human CuZnsuperoxide dismutase by transcription of the cloned cDNA for the human CuZn-superoxide dismutase (31). The means by which the cellular content of glutathione per-
AND
LYNCH TABLE
IV
Glucose-6-Phosphate Dehydrogenase, Glutathione Reductase, and Reduced Glutathione in Wild-Type and Paraquat-Resistant HeLa Cells Glutathione reductase (pmol/min/ md Cells Wild-type PQRHM40 PQRHMSO
S3
Note. Lysates to Table III and and for glutathione Lysates of HeLa for glutathione glutathione was
GGPD (amol/ min/mg)
Glutathione (nmol/mg protein)
Mean
SD
Mean
SD
Mean
SD
43.6 54.1 50.7
4.0 1.4 0.2
171 198 139
7.8 6.1 1.2
129 126 a2
15.5 25.4 15.6
of HeLa cells were prepared as described in the legend assayed for glucose-g-phosphate dehydrogenase (GGPD) reductase as described under Materials and Methods. cells deproteinized with 1% picric acid were assayed as described under Materials and Methods. Oxidized not detectable.
oxidase was magnified in these two instances is not yet clear although enrichment for glutathione peroxidase in cells transcribing the transfected cDNA for human CuZnsuperoxide dismutase had been observed (32). The modest magnitude of the enrichment for glutathione peroxidase may seem insufficient to account for resistance to paraquat. HeLa cells appear to be relatively rich in glutathione peroxidase in comparison with other lines of cultured cells. This characteristic could permit the acquisition of resistance to paraquat by magnification of the cellular content of superoxide dismutases and catalase without a change in the cellular content of glutathione peroxidase when wild-type HeLa cells became resistant to 40 /IM paraquat. Constitutive enrichment for glutathione peroxidase may also enable PQRHM40 cells to resist the next increment in the concentration of paraquat (from 40 to 90 PM) by a modest increase in cellular content of glutathione peroxidase. The unusual structural feature of glutathione peroxidase, a single selenocysteine per subunit, may also impose a limit on the magnitude to which the cellular content can be augmented. The codon for phosphoserine, the amino acid which serves as a precursor of selenocysteine, is a termination codon (33, 34) recognized by a unique tRNA (35). The cellular abundance of this unique tRNA may be sufficient only for synthesis of the usual cellular content of glutathione peroxidase but not for substantially greater rates of synthesis. It may, as a consequence, not be possible to enrich cells markedly for glutathione peroxidase without first increasing the cellular content of the phosphoserine-inserting, opal-suppressing tRNA. The association between increased resistance to paraquat and enrichment for glutathione peroxidase suggests that HzOz, in addition to 0;) mediates the toxicity of paraquat. If 0; formed within HeLa cells escapes into
GLUTATHIONE
PEROXIDASE
CONTENT
the medium, enrichment for superoxide dismutases in the PQRHM40 cells will augment the concentration of HzOz within cells by capturing 0, before it can escape and converting it to HzOz. Indeed, the ready transit of 0; through one cellular membrane of eukaryotic cells, that of the erythrocyte, has been demonstrated (36). As a result of the increased cellular content of superoxide dismutase in the PQRHM40 cells the dependency of the toxic action of paraquat on 0; would diminish at the cost of an enhanced flux of HzOz. Since HzOz is itself cytotoxic, survival of the PQRHM40 cells in the augmented flux of H202 generated by the increase in concentration of paraquat from 40 to 90 PM necessitates blunting the cytotoxic effects of Hz02. Although other adaptations could also confer resistance, the PQRHMSO cells have magnified the cellular content of glutathione peroxidase. The mechanism by which these HeLa cells have increased the cellular content of superoxide dismutases and catalase in the first step of resistance and of glutathione peroxidase in the second step is not clear. No amplification of genes for superoxide dismutases or catalase was detected in PQRHM40 or PQRHMSO cells in Southern blots probed with fragments of cDNAs for the human CuZnsuperoxide dismutase (37) for the human Mn-superoxide dismutase, and for human catalase (38,39) after digestion of genomic DNA with P&I or with BgZII. Although it seems more likely that distinct adaptations occurred in the PQRHM40 and and the PQRHMSO cells because the activities of distinct sets of enzymes increased in each step, it is conceivable that a single adaptation which occurred in the PQRHM40 cells was magnified further in the PQRHMSO cells. How HeLa cells have increased cellular content of these enzymes remains to be determined. REFERENCES 1. Berlin,
V., and Haseltine,
W. A. (1981)
J. Biol.
Chem.
266,4747-
OF
Chem.
3. Babior, Invest.
L., Zweier, J. L., Levy, 260,6820-6826. B. M., Kipnes, 52, 741-744.
4. Hassan,
H. M.,
A., and Myers,
C. E. (1985)
R. S., and Curnutte,
J. Biol.
J. T. (1973)
12. Beauchamp, 13. Haber,
I. (1978)
J. Biol.
Chem.
253,8143-
5. Moody, C. S., and Hassan, 79,2855-2859.
H. M. (1982)
6. Bagley, A. C., Krall, J., and Lynch, Sci. USA 83,3189-3193. 7. Krall, J., Bagley, A. C., Mullenbach, Lynch, R. E. (1988) J. Biol. Chem.
8. Carlioz,
A., and Touati,
9. Scott, Chem.
M. D., Meshnick, 262,3640-3645.
10. Elroy-Stein, 615-622.
O., Bernstein,
D. (1986)
Proc. Natl.
R. E. (1986)
Acad. Proc.
G. T., Hallewell, 263,1910-1914. EMBO
N&l.
Acad.
R. A., and
J. W. (1987) Y. (1986)
F., and Weiss,
14. Halliwell,
EMBO
J. Biol. J. 5.
I. (1970)
J. Biol.
Chem.
M. (1986) 245,4641-
J. (1934)
Proc.
R. Sot. London
Ser. A
147,
B., and Gutteridge,
J. M. C. (1986)
Arch. Biochem.
Biophys.
246,501-514. 15. Beers, R. F., and Sizer, I. W. (1952) J. Biol. Chem. 195, 133-140. 16. Paglia, D. E., and Valentine, W. N. (1967) J. Lab. Clin. Med. 70, 158-169.
17. McCord, 6055. 18. Malinowski,
J. M., and Fridovich, D. P., and Fridovich,
I. (1969)
J. Biol.
I. (1979)
Chem.
Biochemistry
244,
6049-
18, 237-
244. 19. Beauchamp,
C., and Fridovich,
I. (1971)
Anal.
Biochem.
44,
276-
267. 20. Hassan, H. M., and Fridovich, I. (1977) J. Bacterial. 129, 15741583. 21. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193,265-275. 22. Langdon, R. G. (1966) in Methods in Enzymology (Wood, W. A., 23. 24. 25. 26.
Ed.), Vol. 9, pp. 126-131, Academic Press, New York. Beutler, E. (1969) J. Clin. Invest. 48, 1957-1966. Tietze, F. (1969) Anal. Biochem. 27, 502-522. Griffith, 0. W. (1980) Anal. Biochem. 106, 207-212. Hauswirth, W. W., Lim, L. O., Dujon, B., and Turner, G. (1987) in Mitochondria: A Practical Approach (Darley-Usmar, V. M., Rickwood, D., and Wilson, M. T., Eds.), pp. 176 and 177, IRL Press, Oxford/Washington, DC. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254.
27. 28. Raetz,
C. R. H., Wermuth, M. M., McIntyre, T. M., Esko, J. D., and Wing, D. C. (1982) Proc. Natl. Acad. Sci. USA 79,3223-3227. 29. Colbere-Garapin, F., Horodniceanu, F., Kourilsky, P., and Garapin, A-C. (1981) J. Mol. Biol. 150, 1-14. 30. Kelner, M. J., and Bagnell, R. (1990) Free Radicals Biol. Med. 9, 149-153. 31. Kelner, M. J., and Bagnell, R. (1990) J. Biol. Chem. 269, 10872-
10875. 32. Ceballos,
I., Delabar, J. M., Nicole, A., Lynch, R. E., Hallewell, R. A., Kamoun, P., and Sinet, P. M. (1988) Biochim. Biophys. Actu
949,58-64. 33. Chambers, I., Frampton,
35. 36.
J., Goldfarb, P., Affara, N., McBain, W., and Harrison, P. R. (1986) EMBO J. 5, 1221-1227. Mullenbach, G. T., Tabrizi, A., Irvine, B. D., Bell, G. I., and Hallewell, R. A. (1987) Nucl. Acids Res. 16,5484. Lee, B. J., de la Pena, P., Tobian, J. A., Zasloff, M., and Hatfield, D. (1987) Proc. Natl. Acad. Sci. USA 84,6384-6388. Lynch, R. E., and Fridovich, I. (1978) J. Biol. Chem. 253, 2838-
2845. 37. Hallewell,
R. A., Masiarz, F. R., Najarian, R. C., Puma, J. P., Quiroga, M. R., Randolph, A., Sanchez-Pescador, R., Scandella, C. J., Smith, B., Steimer, K. S., and Mullenbach, G. T. (1985) Nucl. Acids Res.
13,2017-2034. 36. Korneluk, R. G., Quan,
J. 6,623-630.
S. R., and Eaton, Y., and Groner,
Sci. USA
J., and Chevion,
332-351.
34.
and Fridovicb,
C., and Fridovich,
315
CELLS
4646.
J. Clin.
8148.
HeLa
11. Korbashi, P., Kohen, R., Katzhendler, J. Biol. Chem. 261,12471-12476.
4756. 2. Gianni,
PARAQUAT-RESISTANT
F., Lewis, W. H., Guise, K. S., Willard, H. F., Holmes, M. T., and Gravel, R. A. (1984) J. Biol. Chem. 259, 13819-13823. 39. Quan, F., Korneluk, R. G., Tropak, M. B., and Gravel, R. A. (1986) Nucl. Acids Res. 14, 5321-5335.
