159

Braz J Med Biol Res 37(2) 2004

Glutathione peroxidase induction in superoxide dismutase mutantsBrazilian Journal of Medical and Biological Research (2004) 37: 159-165ISSN 0100-879X

Glutathione peroxidase inductionprotects Saccharomyces cerevisiaesod1∆∆∆∆∆sod2∆∆∆∆∆ double mutants againstoxidative damage

1Laboratório de Estresse Oxidativo, Departamento de Biofísica,2Centro de Biotecnologia, and 3Departamento de Química Inorgânica,Instituto de Química, Universidade Federal do Rio Grande do Sul,Porto Alegre, RS, Brasil

V. Manfredini1, R. Roehrs2,M.C.R. Peralba3,

J.A.P. Henriques1,2,J. Saffi2, A.L.L.P. Ramos1

and M.S. Benfato1

Abstract

Saccharomyces cerevisiae mutants deficient in superoxide dismutasegenes (sod1∆, sod2∆ and the double mutant) were subjected to H2O2

stress in the stationary phase. The highest sensitivity was observed inthe sod2∆ mutant, while the sod1∆sod2∆ double mutant was notsensitive. sod mutants had lower catalase activity (44%) than wild-type cells, independent of H2O2 stress. Untreated cells of sod1∆sod2∆double mutants showed increased glutathione peroxidase activity(126%), while sod1∆ had lower activity (77%) than the wild type.Glutathione levels in sod1∆ were increased (200-260%) after expo-sure to various H2O2 concentrations. In addition, the highest malondi-aldehyde levels could be observed without H2O2 treatment in sod1∆(167%) and sod2∆ (225%) mutants. In contrast, the level of malondi-aldehyde in the sod1∆sod2∆ double mutant was indistinguishablefrom that of the wild type. These results suggest that resistance to H2O2

by sod1∆sod2∆ cells depends on the induction of glutathione peroxi-dase and is independent of catalase, and that glutathione is a primaryantioxidant in the defense against H2O2 in stationary phase sod1∆mutants.

CorrespondenceM.S. Benfato

Departamento de Biofísica

Instituto de Biociências, UFRGS

Av. Bento Gonçalves, 9500

Prédio 43422

91501-970 Porto Alegre, RS

Brasil

Fax: +55-51-3316-7003

E-mail: mbenfato@adufrgs.ufrgs.br

Research supported by FAPERGS,

CNPq and GENOTOX-CBIOT-UFRGS.

Received April 25, 2003

Accepted August 28, 2003

Key words• Catalase• Superoxide dismutase• Glutathione• Hydrogen peroxide• Saccharomyces cerevisiae• Reactive oxygen species

Introduction

Oxygen metabolism may lead to the pro-duction of reactive oxygen species (ROS),i.e., superoxide and hydroxyl radicals andhydrogen peroxide (H2O2), by sequential one-electron reductions. ROS damages all cellu-lar components, including protein, DNA andlipids. A primary source of superoxide is theelectron transport chain in the inner mem-brane of mitochondria, where about 2% of

the oxygen consumed during respiration isincompletely reduced to ROS. To counteractthe oxidative stress resulting from ROS, cellspossess a range of nonenzymatic and enzy-matic defense systems, including glutathi-one (GSH), thioredoxin, superoxide dismu-tase (SOD), catalase (CAT) and glutathioneperoxidase (GPx) (1).

A first line of defense includes the SODenzymes that catalyze the disproportionatecleavage of superoxide to H2O2 and water.

160

Braz J Med Biol Res 37(2) 2004

V. Manfredini et al.

H2O2 is enzymatically catabolized in aerobicorganisms by catalase and several peroxi-dases. Saccharomyces cerevisiae, like mostother eukaryotes, contains CuZnSOD (theproduct of the SOD1 gene) in the cytosol,nucleus, and lysosomes and MnSOD (theproduct of the SOD2 gene) in the mitochon-drial matrix.

MnSOD is believed to be the major meansof protection against mitochondrial super-oxide (2). The sources of ROS relevant toCuZnSOD are less clear because of thelocation of this enzyme within the cytosol.Neither MnSOD nor CuZnSOD is strictlyessential; however, the loss of CuZnSODhas dramatic phenotypic consequences inyeast. Yeast strains lacking CuZnSOD(sod1∆) show several defects during aerobicgrowth. These include reduced growth ratesin glycerol and ethanol, auxotrophy for lysineand methionine or cysteine, a higher rate ofspontaneous mutation, more rapid loss ofviability in the stationary phase, increasediron uptake, elevated levels of “free iron”,and sensitivity to redox-cycling drugs suchas paraquat or menadione (3-8). sod2∆mutants are oxygen-sensitive and, whenrequired to utilize oxygen, grow poorly andare particularly sensitive to paraquat (6).sod1∆sod2∆ double mutants are more se-verely affected, exhibiting essentially all thecharacteristics of the single mutant pheno-types.

The full growth cycle of a yeast culturebegins with the exponential phase of growth(log phase) and progresses through thediauxic shift to the true stationary phase. Inthe log phase, the cells use glucose to pro-duce energy via glycolysis. The diauxic shiftoccurs when fermentable nutrients becomelimited and energy metabolism shifts to res-piration, the growth rate slows, and cellsutilize ethanol and other two- and three-carbon compounds for energy production. Inthe total absence of any nutrients, the cellsenter the true stationary phase; no cell divi-sion occurs, the metabolic rate slows and

cells can survive for weeks to months.Stationary phase yeast resemble the ma-

jority of cells of multicellular organisms intwo important aspects: 1) their main sourceof energy is mitochondrial respiration, and2) the cells have exited the cell cycle andentered the G0 phase. With increasing timein the stationary phase, damage accumu-lates. This damage cannot be excluded bycell synthesis and division, which is notoccurring, and thus must be prevented orrepaired. For this reason the yeast S. cerevi-siae has been extensively exploited as a mo-del for advancing our understanding of cel-lular defenses against ROS.

In the present study we investigated therole of nonenzymatic and enzymatic defensesystems after H2O2 treatment in sod mutants.We found that sod1∆ and sod1∆sod2∆ doublemutants show very little or no sensitivity toH2O2 in the stationary phase. The doublemutant displays increased GPx activity andreduced malondialdehyde (MDA) levelscompared to wild-type cells, whereas CATis decreased. The sod1∆ mutant has the high-est total GSH content after H2O2. Thus, theresistance to H2O2 in sod1∆sod2∆ cells isdependent on the induction of GPx, whereasCAT does not appear to be required. Wesuggest here that GSH is a primary antioxi-dant in the defense against H2O2 in sod1∆mutants.

Material and Methods

Strains of Saccharomyces cerevisiae

The wild-type S. cerevisiae strain EG103and the isogenic mutant strains sod1∆, sod2∆and sod1∆sod2∆ were kindly provided byDr. E. Gralla (University of California, LosAngeles, CA, USA). Disruption of the SOD1and SOD2 genes was performed as describedpreviously (4,9). MG5312 was a gift fromDr. M. Brendel (J.W. Goethe-University,Frankfurt, Germany). The relevant genotypesof the strains are listed in Table 1.

161

Braz J Med Biol Res 37(2) 2004

Glutathione peroxidase induction in superoxide dismutase mutants

Media and growth conditions

Yeast strains were grown at 30ºC in YPDliquid medium containing 2% glucose, 1%yeast extract, 2% bacto-peptone, or selectivemedium supplemented with the appropriatenutrients (SD medium with 2% glucose,0.67% yeast nitrogen base without aminoacids, plus nutrients). The flask volume/me-dium ratio was 2:1 (microaerophilic condi-tions) and flasks were shaken at 200 rpm.For solid medium 2% agar was added (10).The cell lines lacking CuZnSOD were con-tinually monitored for suppressor activity (4,6).

Hydrogen peroxide treatment

Yeast cells were grown to the stationaryphase (1-2 x 108 cells/ml) in YPD medium at30ºC. Cells were harvested and washed insterile saline (0.9% NaCl) and the cell pel-lets were resuspended in saline and treatedwith increasing concentrations of H2O2 (0.5-10 mM) at 30ºC for 1 h.

For dose-response curves, aliquots of cellswere diluted in saline and plated in triplicateonto YPD to obtain viable counts after 3-5days of growth at 30ºC.

Enzyme activities

Crude extracts were prepared by glassbead lysis as follows: cells were suspendedin lysis buffer (50 mM Tris, 150 mM NaCl,50 mM EDTA, pH 7.2) with an equal vol-ume of acid-washed 425-600 µm glass beadsand phenylmethylsulfonyl fluoride, vortexedfor 10-15 cycles (30 s each), followed by 30s of cooling. The mixture was then micro-centrifuged for 2 min to remove the cellulardebris and glass beads (6).

CAT activity was determined spectro-photometrically by monitoring the disappear-ance of H2O2 at 240 nm (11). GPx activitywas determined by monitoring the NADPHconsumption rate at 340 nm (12). Proteinconcentration was determined by the

Bradford assay (13).

Assay of total glutathione

Total GSH was monitored by a recentlydeveloped microbiological method that usesa GSH auxotrophic yeast strain as a sensorfor the presence of GSH, GSSG and γ-Glu-Cys (14). The indicator strain, S. cerevisiaeMG5312 (Table 1), homozygous for themutant allele gsh1∆URA3, was grown to acell density of 2 x 108 cells/ml, washed twicewith and then suspended in potassium phos-phate buffer (25 mM KH2PO4, 50 mMNa2HPO4, pH 7.0) to a concentration of 2 x107 cells/ml. One hundred microliters of thissuspension was added to 3.5 ml top-agar(0.4% agar in potassium phosphate buffer)at a temperature of 48ºC and poured imme-diately onto synthetic medium lacking GSH.After the top-agar had solidified, 20 µl of thesamples to be tested was applied to a sterilepaper disc 10 mm in diameter on the agarsurface. Plates were incubated at 30ºC for 3days. Growth of MG5312 depends on thepresence either GSH or GSSG. Growth zoneson agar plates seeded with MG5312 permitthe quantitative determination as little as 0.1µg GSH (for details see Ref. 14).

Preparation of yeast samples. After H2O2

treatment yeast cells were harvested, washedtwice and suspended in potassium phosphatebuffer to a final concentration of 4%. Cellsuspensions were heat-treated for 15 min at85ºC and cell debris was removed by cen-trifugation at 11,700 g.

Table 1. Yeast strains used in this study.

Yeast strain Genotype Ref.

EG103 (DBY746) MATa leu2-3, 112 his3∆1 trp1-289 ura3-52 GAL+ 4EG118 EG103 with sod1∆::URA3 4EG110 EG103 with sod2∆::TRP1 9EG133 EG103 with sod1∆::URA3 sod2∆::TRP1 9MG5312 MATa/MATα ade2-101/ade2-101 LEU2/leu2∆1 LYS2/lys2-801 14

TRP1/trp1-∆1 TRP5/trp5-a ura3-52/ura3-52gsh1-∆1::URA3/gsh1-∆1 URA3 (rho-)

162

Braz J Med Biol Res 37(2) 2004

V. Manfredini et al.

HPLC using an amino-phase column withacetonitrile, 30 mM Tris buffer, pH 7.4 (1:9v/v). The effluent was monitored at 267 nm(15).

Statistical analysis

Results are reported as means ± SD andwere analyzed by the Student t-test. Valuesof P < 0.05 were considered to be statisti-cally significant.

Results

Wild-type and sod null mutant cultures inthe stationary phase of growth under mi-croaerophilic conditions were treated withincreasing levels of H2O2 for 1 h at 30ºCunder non-growth conditions. The results(Figure 1) showed that the sod2∆ mutantwas more sensitive to H2O2 than its isogenicparent strain, while the sensitivity of sod1∆was similar to that of the wild type. Surpris-ingly, the sensitivity of the sod1∆sod2∆double mutant was indistinguishable fromthat of the wild type.

In order to determine the possible mechan-isms that may contribute to H2O2 resistancein the sod1∆sod2∆ and sod1∆ mutants, anti-oxidant enzyme activities were assayed (Fig-ures 2-4). Decreased CAT activity was ob-served either with or without H2O2 treatmentin simple and double sod mutants (Figures 2and 3). On the other hand, GPx activitieswere significantly increased in sod2∆ andsod1∆sod2∆ mutants without treatment,while sod1∆ showed lower GPx activity thanthe wild type (Figure 2). After H2O2 treat-ment, GPx levels were significantly increasedwhen exposed to 0.25-0.5 mM H2O2 andstrongly reduced at 2.5 mM H2O2 in thewild-type and mutant strains (Figure 4). Atthe highest doses of H2O2 (5-10 mM), GPxlevels were the same as in untreated cells forsod2∆, while wild-type and sod1∆sod2∆ lev-els increased over control cell levels andsod1∆ maintained lower GPx activities.

Figure 1. Sensitivity of sod nullmutants exposed to H2O2. Thestrains used were: SOD+,sod1∆, sod2∆ and sod1∆sod2∆.Data are reported as the mean ±SD of three independent experi-ments.

Figure 2. Activity of catalase andglutathione peroxidase in un-treated wild-type and sod mu-tant cells. Data are reported asthe mean ± SD of three inde-pendent experiments. *P < 0.05compared to SOD+ (Student t-test).

Act

ivity

(U

/mg

prot

ein) 45

30

15

00 5 10

SOD+ sod1∆ sod1∆sod2∆sod2∆

H2O2 (mM)

Figure 3. Activity of catalase instationary wild-type and sod mu-tant cells exposed to H2O2. Thestrains used were: SOD+,sod1∆, sod2∆ and sod1∆sod2∆.Data are reported as the mean ±SD of three independent experi-ments. aP < 0.05 compared toSOD+-untreated cells; bP < 0.05compared to sod1∆-untreatedcells; cP < 0.05 compared tosod2∆-untreated cells; dP < 0.05compared to sod1∆sod2∆-un-treated cells (Student t-test).

Act

ivity

(U

/mg

prot

ein) 40

SOD+ sod1∆ sod2∆ sod1∆sod2∆

Glutathione peroxidase Catalase

30

20

10

0

Sen

sitiv

ity (

%)

100

SOD+ sod1∆

sod1∆sod2∆ sod2∆

100 2 4 6 8 10

H2O2 (mM)

Malondialdehyde determination

MDA was measured by HPLC by themethod described by Esterbauer and Cheese-man (15). Briefly, an aqueous sample con-taining MDA, pH 6.5-8.0, was separated by

dd

b b

c c

a

**

*

*

*

*

163

Braz J Med Biol Res 37(2) 2004

Glutathione peroxidase induction in superoxide dismutase mutants

The total GSH levels in these mutantswere increased at 0.5 mM H2O2 and werereduced 6- to 7-fold at 1 mM H2O2 in thewild type, while the strain lacking both SODsshowed a reduction in GSH levels both withand without treatment (Figure 5). The singlesod1∆ mutant showed increased GSH at 0.5to 5 mM H2O2. Little increase in the GSHlevel was caused by H2O2 treatment in thesingle sod2∆ mutant.

The lipid peroxidation index was deter-mined on the basis of MDA levels (Figure 6).The highest MDA levels were observed with-out H2O2 treatment in simple sod mutants.On the other hand, the MDA level was indis-tinguishable from that of the wild type in thesod1∆sod2∆ double mutant.

Discussion

Metabolically active cells consume oxy-gen intensively; under limited aeration, thisleads to low oxygen concentration within thecells. In stationary cultures, metabolism isconsiderably reduced and the partial pres-sure of oxygen can increase as a result ofdecreased metabolism. Under these condi-tions of high oxidative stress, one-electronreactions of autoxidation-prone cellular com-ponents may produce superoxide and otherROS. Mitochondrial respiration has beensuggested to be the major source of ROSunder these conditions, even or especiallyunder conditions of low aeration.

Oxidative stress may be a factor limitingthe survival of microorganisms in long-termstationary culture. The increases in antioxi-dant content and in the levels of antioxidantenzymes, including SOD, in yeast enteringthe stationary phase may, therefore, consti-tute an adaptive response to the enhancedoxidative damage (16). The resistance toH2O2 in sod1∆sod2∆ cells is dependent onthe induction of GPx rather than CAT, thelatter apparently being unnecessary. In theabsence only of MnSOD (sod2∆ mutant),GPx was induced, but this did not prevent

Act

ivity

(U

/mg

prot

ein) 45

30

15

00 5 10

SOD+ sod1∆ sod1∆sod2∆sod2∆

H2O2 (mM)

bcdabd

b

c

d

acd

Figure 4. Activity of glutathioneperoxidase in wild-type and sodmutant cells exposed to H2O2.The strains used were: SOD+,sod1∆, sod2∆ and sod1∆sod2∆.Data are reported as the mean ±SD of three independent experi-ments. aP < 0.05 compared toSOD+-untreated cells; bP < 0.05compared to sod1∆-untreatedcells; cP < 0.05 compared tosod2∆-untreated cells; dP < 0.05compared to sod1∆sod2∆-untreated cells (Student t-test).

Figure 6. Malondialdehyde (MDA) determination in untreated wild-type and sod mutantcells. Data are reported as the mean ± SD of four independent experiments. *P < 0.05compared to SOD+ (Student t-test).

SOD+

sod1∆

sod1∆sod2∆sod2∆12

µM M

DA

/mg

prot

ein

7

6

5

4

3

2

1

0

123456781234567812345678123456781234567812345678123456781234567812345678123456781234567812345678123456781234567812345678123456781234567812345678

*

*

µg G

SH

/mg

prot

ein

5

123123123123123123123123123123123123123

123123123123123123123123123123123123123123123123

12341234123412341234123412341234123412341234

123123123123123123123123123123123123

a

bc

d

b

b

a dd

00 0.5 1 5

4

3

2

1

H2O2 (mM)

SOD+ sod1∆ sod1∆sod2∆sod2∆1212

Figure 5. Total glutathione (GSH) content of wild-type and sod mutant cells exposed toH2O2. The total GSH was monitored by a microbiological method that uses a GSHauxotrophic yeast strain as a sensor for the presence of GSH, GSSG and γ-Glu-Cys. Dataare reported as the mean ± SD of two independent experiments. aP < 0.05 compared toSOD+-untreated cells; bP < 0.05 compared to sod1∆-untreated cells; cP < 0.05 comparedto sod2∆-untreated cells; dP < 0.05 compared to sod1∆sod2∆-untreated cells (Student t-test).

164

Braz J Med Biol Res 37(2) 2004

V. Manfredini et al.

cell mortality (70% at 10 mM H2O2; Figure 1).However, the cells lacking CuZnSOD (sod1∆mutant) were resistant to H2O2 (Figure 1).These results cannot be explained by GPx orCAT induction, because both GPx and CATactivities were lower in the sod1∆ mutant(Figures 2-4).

We also investigated the levels of totalGSH in these mutants and in the wild type(Figure 5). The sod1∆ mutant showed in-creased GSH after H2O2 treatment. Theseresults indicate that GSH is a primary anti-oxidant in the defense against H2O2 in sod1∆cells. Mixed disulfides of GSSG with pro-teins accumulate in tissues subjected to oxi-dative stress, in both the mitochondria andthe cytosol (17). GSSG can inhibit proteinsynthesis in animal and plant cells (18,19).These actions of GSSG may explain whycells keep intracellular GSSG levels verylow under normal conditions, and why cellsexport GSSG when they are under oxidativestress (20-23). The sod1∆sod2∆ double mu-tant showed reduced total GSH levels withor without H2O2 treatment (Figure 5), whilea slight increase in GSH levels was observedin the sod2∆ mutant after H2O2 treatment.We therefore suggest that GSSG and/or mixeddisulfides are exported by these cells. It isknown that GSSG levels are similar in expo-nential phase cells and their media duringgrowth conditions and that the extracellularGSH level is elevated following treatmentwith H2O2 concentrations greater than 0.5mM (24). We obtained similar results in thestationary phase (Figure 5).

The highest levels of antioxidant defenses,including GPx or total GSH, were observedin sod mutants. CAT levels, however, werelower and the enzyme was not induced afterH2O2 treatment. SOD enzymes catalyze thedisproportionate cleavage of superoxide, pro-ducing oxygen and H2O2; so sod mutantsshould contain less H2O2. H2O2 is enzymati-

cally detoxified within the cells by CAT andGPx; however, GPx also detoxifies otherperoxides. The acatalasemic mutant cells inthe stationary phase were much more sensi-tive to H2O2 stress than wild-type cells. Inaddition, the ability of acatalasemic cells toshow adaptation to H2O2 treatment has beenshown to be distinctly inferior to that of thewild type (25). These results suggest thatCAT is not essential for yeast cells undernormal conditions, but plays an importantrole in the acquisition of tolerance to oxida-tive stress in the adaptive response. The sodmutants did not induce CAT in spite of stressin their intracellular environment (Figure 3).If superoxide is not adequately removed itcould inhibit CAT (26). However, sod mu-tants induced GPx (Figure 4).

These results suggest that sod mutantsproduce other type(s) of intracellular perox-ides. It has been shown recently that the GPxgenes of S. cerevisiae encode phospholipidhydroperoxide GPx and that these enzymesprotect yeast against phospholipid hydroper-oxides as well as nonphospholipid peroxidesduring oxidative stress (27,28). Therefore,we can suggest that single sod mutants haveincreased lipid peroxidation as well as pro-tein damage (7,8). Our results showed thatlack of CuZnSOD or MnSOD causes a sig-nificant increase, 71 and 124%, respectively,in the levels of MDA, an indicator of lipidperoxidation (Figure 6). However, cells lack-ing both CuZnSOD and MnSOD enzymesdid not show elevated MDA levels. Thesefindings are in agreement with the GPx in-duction and resistance to H2O2 observed insod1∆sod2∆.

Acknowledgments

The authors wish to thank Dr. C. Gaylardfor a critical English revision, and Dr. C.R.Carlini for generously providing equipment.

165

Braz J Med Biol Res 37(2) 2004

Glutathione peroxidase induction in superoxide dismutase mutants

References

1. Halliwell B & Gutteridge JMC (1999). Free Radicals in Biology andMedicine. 3rd edn. Oxford University Press, New York.

2. Gralla EB & Kosman DJ (1992). Molecular genetics of superoxidedismutases in yeasts and related fungi. Advances in Genetics, 30:251-319.

3. Farr SB, D’Ari R & Touati D (1986). Oxygen-dependent mutagenesisin Escherichia coli lacking superoxide-dismutase. Proceedings ofthe National Academy of Sciences, USA, 83: 8268-8272.

4. Gralla EB & Valentine JS (1991). Null mutants of Saccharomycescerevisiae Cu,Zn superoxide dismutase: characterization and spon-taneous mutation rates. Journal of Bacteriology, 173: 5918-5920.

5. Bilinski T, Krawiec Z, Liczmanski A & Litwinska J (1985). Is hydroxylradical generated by the Fenton reaction in vivo? Biochemical andBiophysical Research Communications, 130: 533-539.

6. Longo VD, Gralla EB & Valentine JS (1996). Superoxide dismutaseactivity is essential for stationary phase survival in Saccharomycescerevisiae. Journal of Biological Chemistry, 271: 12275-12280.

7. De Freitas JM, Liba A, Meneghini R, Valentinie JS & Gralla EB(2000). Yeast lacking superoxide dismutase show altered iron ho-meostasis. Role of oxidative stress in iron metabolism. Journal ofBiological Chemistry, 275: 11645-11649.

8. Srinivasan C, Liba A, Imlay JA, Valentine JS & Gralla EB (2000). Yeastlacking superoxide dismutase(s) show elevated levels of “free iron”by whole cell electron paramagnetic resonance. Journal of Biologi-cal Chemistry, 275: 29187-29192.

9. Liu XF, Elashvili I, Gralla EB, Valentine JS, Lapinskas P & Culotta VC(1992). Yeast lacking superoxide-dismutase - isolation of geneticsuppressors. Journal of Biological Chemistry, 267: 18298-18302.

10. Shermann F, Fink GR & Hicks JB (1986). Methods in Yeast Genet-ics. Cold Spring Harbor, New York.

11. Taniguchi N & Gutteridge JMC (2000). Experimental Protocols forReactive Oxygen and Nitrogen Species. 1st edn. Oxford UniversityPress, New York.

12. Pinto RE & Bartley W (1969). Effect of age and sex on GSHreductaseand GSHperoxidase activities and on aerobic glutathione oxidationin rat liver homogenates. Biochemical Journal, 112: 109-115.

13. Bradford M (1976). A rapid and sensitive method for the quantita-tion of microgram quantities of protein utilizing the principle ofprotein-dye binding. Analytical Biochemistry, 72: 248-254.

14. Schmidt M, Grey M & Brendel M (1996). A microbiological assay forthe quantitative determination of glutathione. BioTechniques, 21:881-886.

15. Esterbauer H & Cheeseman KH (1990). Determination of aldehydiclipid peroxidation products: malonaldehyde and 4-hydroxynonenal.

Methods in Enzymology, 186: 407-421.16. Jakubowski W, Bilinski T & Bartosz G (2000). Oxidative stress during

aging of stationary cultures of the yeast Saccharomyces cerevisiae.Free Radical Biology and Medicine, 28: 659-664.

17. Ravichandran V, Seres T, Moriguchi T, Thomas JA & Johnston RB(1994). S-Thiolation of glyceraldehyde-3-phosphate dehydrogenaseinduced by the phagocytosis-associated respiratory burst in bloodmonocytes. Journal of Biological Chemistry, 269: 25010-25015.

18. Schuppekoistinen I, Gerdes R, Moldeus P & Cotgreave IA (1994).Studies on the reversibility of protein S-thiolation in human endothe-lial cells. Archives of Biochemistry and Biophysics, 315: 226-234.

19. Dhindsa RS (1987). Glutathione status and protein synthesis duringdrought and subsequent rehydration in Tortula ruralis. Plant Physiol-ogy, 83: 816-919.

20. Hirrlinger J, Schulz JB & Dringen R (2002). Effects of dopamine onthe glutathione metabolism of cultured astroglial cells. Implicationsfor Parkinson’s disease. Journal of Neurochemistry, 82: 458-467.

21. Wernerman J, Luo JL & Hammarqvist F (1999). Glutathione statusin critically-ill patients; possibility of modulation by antioxidants.Proceedings of the Nutrition Society, 58: 677-680.

22. Spooren AAMG & Evelo CTA (1997). Hydroxylamine treatment in-creases glutathione-protein and protein-protein binding in humanerythrocytes. Blood Cells, Molecules, and Diseases, 17: 323-336.

23. Lapshina EA & Bartosz G (1995). What determines the antioxidantpotential of yeast cells. Biochemistry and Molecular Biology Inter-national, 37: 949-957.

24. Grant CM, Perrone G & Dawes IW (1998). Glutathione and catalaseprovide overlapping defenses for protection against hydrogen per-oxide in the yeast Saccharomyces cerevisiae. Biochemical andBiophysical Research Comunications, 253: 893-898.

25. Izawa S, Inoue Y & Kimura A (1996). Importance of catalase in theadaptive response to hydrogen peroxide: analysis of acatalasaemicSaccharomyces cerevisiae. Biochemical Journal, 320: 61-67.

26. Lardinois OM (1995). Reactions of bovine liver catalase with super-oxide radicals and hydrogen-peroxide. Free Radical Research, 22:251-274.

27. Inoue Y, Matsuda T, Sugiyama K, Izawa S & Kimura A (1999).Genetic analysis of glutathione peroxidase in oxidative stress re-sponse of Saccharomyces cerevisiae. Journal of Biological Chemis-try, 274: 27002-27009.

28. Avery AM & Avery SM (2001). Saccharomyces cerevisiae expressesthree phospholipid hydroperoxide glutathione peroxidases. Journalof Biological Chemistry, 276: 33730-33735.