Logo of mbcLink to Publisher's site
Mol Biol Cell. 2009 Mar 1; 20(5): 1493–1508.
PMCID: PMC2649269

Cu, Zn Superoxide Dismutase and NADP(H) Homeostasis Are Required for Tolerance of Endoplasmic Reticulum Stress in Saccharomyces cerevisiae

Sean Munro, Monitoring Editor


Genome-wide screening for sensitivity to chronic endoplasmic reticulum (ER) stress induced by dithiothreitol and tunicamycin (TM) identified mutants deleted for Cu, Zn superoxide dismutase (SOD) function (SOD1, CCS1) or affected in NADPH generation via the pentose phosphate pathway (TKL1, RPE1). TM-induced ER stress led to an increase in cellular superoxide accumulation and an increase in SOD1 expression and Sod1p activity. Prior adaptation of the hac1 mutant deficient in the unfolded protein response (UPR) to the superoxide-generating agent paraquat reduced cell death under ER stress. Overexpression of the ER oxidoreductase Ero1p known to generate hydrogen peroxide in vitro, did not lead to increased superoxide levels in cells subjected to ER stress. The mutants lacking SOD1, TKL1, or RPE1 exhibited decreased UPR induction under ER stress. Sensitivity of the sod1 mutant to ER stress and decreased UPR induction was partially rescued by overexpression of TKL1 encoding transketolase. These data indicate an important role for SOD and cellular NADP(H) in cell survival during ER stress, and it is proposed that accumulation of superoxide affects NADP(H) homeostasis, leading to reduced UPR induction during ER stress.


Aerobic organisms have a range of mechanisms to cope with reactive oxygen species (ROS) derived from endogenous and exogenous sources. ROS can affect cellular components and elicit damage including lipid peroxidation, nucleic acid damage, and oxidation of proteins (Temple et al., 2005 blue right-pointing triangle). Overwhelming production of ROS, termed oxidative stress, ultimately leads to cell death. Cells have many enzymatic and nonenzymatic mechanisms to remove or detoxify ROS to prevent oxidative stress. Nonenzymatic defenses include small molecules such as the tripeptide glutathione and the dithiol-containing thioredoxins. Enzymatic mechanisms include superoxide dismutases (SODs) that convert superoxide anion to hydrogen peroxide and oxygen, whereas the glutathione dependent peroxidases and catalase convert hydrogen peroxide to water and oxygen (Temple et al., 2005 blue right-pointing triangle; Perrone et al., 2008 blue right-pointing triangle).

SODs are highly conserved, although the cellular localization and requirements for metal ion cofactor(s) differ between organisms. Mutations in the human SOD1, encoding the Cu, Zn SOD, have been implicated in ∼20% of the familial motor neuron disease amyotrophic lateral sclerosis (ALS; Rosen et al., 1993 blue right-pointing triangle). In Saccharomyces cerevisiae, the Cu, Zn SOD, Sod1p, localizes mainly in the cytosol but is also found in the mitochondrial intermembrane space (Sturtz et al., 2001 blue right-pointing triangle). Sod1p activity is dependent on the copper chaperone of Sod1p (Ccs1p) for copper loading and activation (Culotta et al., 1997 blue right-pointing triangle). Deletion of SOD1 or CCS1 leads to a number of oxygen-dependent phenotypes, including oxygen sensitivity, slow growth, hypersensitivity to superoxide-generating agents such as paraquat, accelerated aging, and auxotrophy for methionine and lysine (Culotta, 2000 blue right-pointing triangle). Lysine auxotrophy was attributed to oxidation of the iron sulfur center of homoaconitase, whereas methionine auxotrophy was proposed to result from depletion of cellular NADPH by superoxide anion (Slekar et al., 1996 blue right-pointing triangle).

Cellular NADPH is important for tolerance to ROS. Strains deleted for genes involved in the nonoxidative branch of the pentose phosphate pathway that are important for NADPH generation, such as RPE1 (encoding ribose-5 phosphate epimerase) or TKL1 (encoding an isoform of transketolase) are sensitive to H2O2 (Juhnke et al., 1996 blue right-pointing triangle; Thorpe et al., 2004 blue right-pointing triangle) and also defective in adaptation to H2O2 (Ng et al., 2008 blue right-pointing triangle). NADPH is the cofactor for glutathione reductase and thioredoxin reductase, which are required to reduce oxidized glutathione and thioredoxin, respectively. This highlights the importance of NADPH for defense against ROS and maintenance of cellular redox homeostasis (Holmgren, 1989 blue right-pointing triangle; Grant et al., 1996a blue right-pointing triangle; Muller, 1996 blue right-pointing triangle; Temple et al., 2005 blue right-pointing triangle; Ng et al., 2008 blue right-pointing triangle).

The endoplasmic reticulum (ER) contributes to cellular redox balance since formation of disulphide bonds in proteins that traverse the secretory pathway occurs in this organelle. The ER contains chaperones, ER oxidoreductases, protein disulphide isomerases, and glycosylation enzymes to support optimal protein folding. However, a large influx of protein into the ER, resulting in accumulation of aberrantly folded protein, leads to ER stress (Travers et al., 2000 blue right-pointing triangle; Liu and Kaufman, 2003 blue right-pointing triangle; Ron and Walter, 2007 blue right-pointing triangle). The unfolded protein response (UPR) is a highly conserved cellular pathway that alleviates ER stress (Liu and Kaufman, 2003 blue right-pointing triangle; Ron and Walter, 2007 blue right-pointing triangle). In S. cerevisiae, the UPR requires the ER transmembrane kinase Ire1p and its downstream transcription factor Hac1p. On accumulation of misfolded/unfolded proteins, Ire1p dimerizes, autophosphorylates, and splices the intron of HAC1 mRNA. The translated Hac1p then binds to genes whose promoter contains the unfolded protein response element (Cox et al., 1993 blue right-pointing triangle; Cox and Walter, 1996 blue right-pointing triangle; Mori et al., 1996 blue right-pointing triangle; Kawahara et al., 1997 blue right-pointing triangle). Activation of the UPR alleviates ER stress by up-regulating genes including those required for ER expansion, general secretory pathway function, and protein folding and degradation (Travers et al., 2000 blue right-pointing triangle). The Ire1p-Hac1p pathway functions in a linear manner because strains deleted for either HAC1 or IRE1 have an indistinguishable phenotype and a similar gene expression profile (Travers et al., 2000 blue right-pointing triangle), and no alternative Ire1p substrates have been identified (Niwa et al., 2005 blue right-pointing triangle).

ER stress in yeast has been shown to be associated with ROS generation and cell death in a programmed cell death-like manner (Haynes et al., 2004 blue right-pointing triangle; Hauptmann et al., 2006 blue right-pointing triangle), but the species of ROS and cellular mechanisms that detoxify them under ER stress have not been fully elucidated. The role of ROS in yeast programmed cell death has been reviewed (Perrone et al., 2008 blue right-pointing triangle). The ER is a potential source of ROS formation. It has been proposed that 25% of cellular ROS originates from the ER (Tu and Weissman, 2004 blue right-pointing triangle). The use of oxygen as the terminal electron acceptor by the essential ER oxidoreductase Ero1p during oxidative protein folding can result in ROS production (Tu and Weissman, 2004 blue right-pointing triangle). Ero1p generates H2O2 in vitro (Gross et al., 2006 blue right-pointing triangle), but the ROS generated by Ero1p in vivo remains unknown.

Here, we screened the S. cerevisiae deletion collection (Winzeler et al., 1999 blue right-pointing triangle) for mutants sensitive to the glycosylation inhibitor tunicamycin (TM) and reductant dithiothreitol (DTT) to identify cellular processes important for ER stress tolerance or for countering ROS generated in this stress. Exposure of cells to DTT leads to accumulation of unfolded proteins in the ER (Jamsa et al., 1994 blue right-pointing triangle), whereas TM disrupts N-linked glycosylation through inhibition of the dolichyl phosphate–dependent N-acetylglucosamine-1-phosphate transferase, Alg7p (Barnes et al., 1984 blue right-pointing triangle; Kukuruzinska and Lennon, 1995 blue right-pointing triangle). These compounds cause ER stress and activate the UPR (Travers et al., 2000 blue right-pointing triangle).

Among the mutants that were sensitive, we identified those affected in superoxide detoxification (sod1, ccs1) and NADPH production (tkl1, rpe1) to be sensitive to both ER-stressing agents. This prompted us to further investigate their role in ER stress tolerance.


Yeast, Strains, and Plasmids

The strains used (Table 1) include those derived from BY4743 (Open Biosystems, Huntsville, AL) or from CY4 as described previously (Grant et al., 1996b blue right-pointing triangle). Single-gene deletion strains in the CY4 background were generated using PCR to amplify the disrupted allele and flanking region (∼200–400 base pairs) from the strains available from the S. cerevisiae Genome Deletion Project (Winzeler et al., 1999 blue right-pointing triangle). The amplified DNA product was transformed into CY4 with selection on YEPD medium containing 220 μg/ml geneticin (Invitrogen, Invitrogen, Carlsbad, CA). Transformation was performed using the lithium acetate–based method (Ito et al., 1983 blue right-pointing triangle). Each deletant was confirmed by PCR analysis on genomic DNA isolated from the respective strains. Double mutants were generated by mating haploid single mutants and dissecting spores from the resulting heterozygous diploid. The hac1::NatMX allele was generated by transforming EcoRI-digested p4339 (Tong et al., 2001 blue right-pointing triangle) containing the NatMX cassette into ST003 (hac1::KanMX) with selection on YEPD medium containing 100 μg/ml nourseothricin (Sigma, St. Louis, MO) to generate ST004.

Table 1.
S. cerevisaie strains used in this study

The centromeric ERO1::lacZ fusion construct containing 951 base pairs of upstream untranslated region and 42 base pairs of the coding region of ERO1 was generated by ligating the BglII-digested PCR fragment (amplified using the specific oligonucleotides GCAAAGATCTGTGAGGCACAGTGTGGCAATGGC and GAAGAGATCTTACCTTAATTCTTGCTGAAAAATAC) to the BamHI-digested PYLZ-6 (Hermann et al., 1992 blue right-pointing triangle) to generate pST3 (CEN, URA3).

A centromeric vector containing the GAL1-promoter, multiple-cloning site and CYC1 terminator was generated by digesting JMB671 DNA (a kind gift from Joe Mymryk, University of Western Ontario, London, Ontario, Canada) with ScaI and NgoMIV. The resulting 1.4-kbp fragment was ligated to pRS413 digested with the same enzymes to generate p413GAL1 (CEN, HIS3). The galactose-inducible ERO1 plasmid was generated by ligating a BamHI-XhoI–digested PCR fragment amplified from genomic DNA using oligonucleotides GTATTGGATCCATGAGATTAACAACCGCC and CATATAAGATAACTCGAGTTGCTC to vector p413GAL1 digested with BamHI and SalI generating pST1 (GAL1-ERO1, CEN, HIS3). Amplifying ERO1 using ATCGGGATCCATGAGATTAAGAACCGCCATTGC and CTGAGATGAGTTTTTGTTCTTGTATATCTAGCTTATAGGAAAT, followed by a second round of PCR amplification using ATCGGGATCCATGAGATTAGAACCGCCATTGC and CTACTCGAGCTACAGATCCTCTTCTGAGATGAGTTTTTGTTC, generated addition of a myc tag to the carboxyl-terminal of ERO1. The resulting 1.7-kbp PCR fragment was digested with BamHI and XhoI and ligated to p413GAL1 digested with BamHI and SalI to generate pST2 (GAL1-ERO1myc, CEN, HIS3). Plasmids were verified by sequencing.

pAC595 (CUP1-CPY*, CEN, URA3; Haynes et al., 2004 blue right-pointing triangle) and pGEV-TRP (CEN, TRP1), a variant of pGEV-HIS (Gao and Pinkham, 2000 blue right-pointing triangle) was a kind gift from Antony Cooper (Garvan Medical Research Institute, Sydney, Australia). The UPRE::lacZ (CEN, LEU2) fusion construct was provided by Peter Walter (University of California, San Francisco, San Francisco, CA). The pTKL1 (2μ, LEU2; Slekar et al., 1996 blue right-pointing triangle) was a kind gift from Munira Basari (National Institutes of Health, Bethesda, MD). pES67 (GAL1-CPY*HA, CEN, LEU2; Spear and Ng, 2003 blue right-pointing triangle) was kindly provided by Davis Ng (Temasek Life Sciences, Singapore). pAF112 (GAL1-ERO1myc, CEN, URA3), pCS452 (GAL1-ero1-C150A-C295A-myc, CEN, URA3) and pCS504 (GAL1-ero1-C150A-C295A-C100A-C105A-myc, CEN, URA3) described by Sevier et al. (2007) blue right-pointing triangle was kindly provided by Carolyn Sevier and Chris Kaiser (both from the Massachusetts Institute of Technology, Cambridge, MA).

Growth Conditions and Tolerance Analysis

Yeast strains were grown in YEPD medium containing 1% (wt/vol) yeast extract, 2% (wt/vol) peptone, and 2% (wt/vol) glucose or synthetic complete medium (SC) containing 0.17% (wt/vol) yeast nitrogen base without amino acid and ammonium sulfate, 0.5% (wt/vol) ammonium sulfate, and 2% (wt/vol) glucose. Auxotrophic supplements were added as specified by the manufacturer's instructions (Sigma) with appropriate amino acid omitted for auxotrophic selection. For anaerobic growth conditions, media were supplemented with 0.1% (vol/vol) Tween 80 and 20 mg/l ergosterol, and plates were maintained in an anaerobic jar using an anaerobic pack and oxygen indicator to verify anaerobiosis (Oxoid, Adelaide, Australia). For tolerance of strains to various compounds, strains were grown in cultures for 2 d, the OD600 was adjusted to 1.0, and the culture was serially diluted 10-fold before spotting on agar plates containing, paraquat, TM, or DTT. Agar was cooled to 55°C before addition of the indicated compounds. Agar plates containing the compound(s) were prepared fresh before use. Growth curves were generated by sampling cultures in a shake flask using a conventional spectrophotometer or growth of cultures in the automated plate reader Bioscreen C (Oy Growth Curves Ab) with constant temperature (30°C) and continuous shaking (“high” speed) and OD600 read every 30 min.

Genome-Wide Screen for Mutant Sensitive to TM and DTT

Cells were grown for 3 d in 96-well plates containing 200 μl of YEPD before replicating into 200 μl of YEPD containing 6 μg/ml TM or 20 mM DTT in 96-well plates using a 96-pin replicator. Cells were incubated for 3 d at 30°C without agitation. Cultures were resuspended using a 96-well plate shaker before determination of cell density using a 96-well plate reader. Strains that showed complete inhibition of growth in both TM- and DTT-containing media were identified as sensitive. The strains sensitive to these compounds were rearrayed and retested for verification.

SOD Assay

The in-gel SOD activity assay was performed as described (White et al., 1993 blue right-pointing triangle; Huang et al., 2002 blue right-pointing triangle; Sanchez et al., 2005 blue right-pointing triangle). Yeast strains were grown to exponential phase, treated with TM for the indicated time, and harvested, and cells were disrupted using acid-washed glass beads and a mini-bead beater (Biospec Scientific, Bartlesville, OK; 1 min at high speed, 4°C) in 0.1 M Tris-HCl (pH 8) containing 20% (vol/vol) glycerol and 2 mM phenylmethanesulphonyl fluoride (PMSF). The lysate was clarified by centrifugation (5 min 10,000 × g at 4°C), and 10 μg was run in a 10% (wt/vol) Tris-acrylamide gel. Gels were stained for 10 min in 36 mM potassium phosphate buffer (pH 8) containing 34.5 μM riboflavin, 0.19 mM nitroblue tetrazolium, and 175 μl tetramethylethylenediamine. Gels were exposed to light for 10 min or until good contrast was achieved.

Western Blot Analysis

Cells were grown as stated, and ∼40 ml of cells was harvested and lysed in lysis buffer (0.1 M Tris-HCl, pH 8, 20% [vol/vol] glycerol, 2 mM PMSF, 1% SDS). Protein content of cell lysate was determined and adjusted to 2 μg/μl in sample loading buffer (5% [vol/vol] β-mercaptoethanol, 40% [vol/vol] glycerol, Tris-HCl, pH 6.8). Samples were boiled for 10 min, and 30 μg of each sample was loaded onto 14% (wt/vol) polyacralymide Tris-SDS gel for determination of Sod1p or a 4–12% NuPAGE gel (Invitrogen) for determination of Ero1mycp, CPY*HAp or Pgk1p. Antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-Sod1p and anti-Pgk1p were diluted to 1:1000 and 1:5000, respectively. Anti-hemagglutinin (HA) and anti-myc were diluted to 1:1000 and 1:500, respectively. The blots were quantitated using Fuji image gauge software (Tokyo, Japan). Intensity of the Sod1p band was normalized relative to that of Pgk1p and expressed as a fold change relative to untreated cells at zero time.

Dihydroethidium and Propidium Iodide Staining and FACS Analysis

Dihydroethidium (DHE) and propidium iodide (PI) staining of cells were performed essentially as described (Deere et al., 1998 blue right-pointing triangle; Madeo et al., 1999 blue right-pointing triangle). For DHE staining, cells (500 μl) were stained with 5 μg/ml final concentration of DHE for 10 min and viewed under a fluorescence microscope without further manipulation. For PI staining, cells (500 μl) were harvested (1 min 13,000 × g), resuspended in an equal volume of ice-cold phosphate-buffered saline (PBS) and stained with 3 μg/ml final concentration of PI for 20 min. Approximately 500 cells were counted in duplicate in duplicate experiments. For FACS analysis of DHE stained cells, cells were washed in PBS, stained as above and analyzed using a FACSAria (BD Bioscience, NSW, Australia). For each sample, 20,000 cells were analyzed.

β-Galactosidase Assay

Cells were grown to exponential phase (OD600 ∼0.5), harvested and analyzed for β-galactosidase activity as described (Rose and Botstein, 1983 blue right-pointing triangle). β-galactosidase activity was expressed as units (U) of ONPG (O-nitrophenyl-4β4-D- galactopyranoside) hydrolyzed (nmol) per min over total protein (mg) and the extent of induction was the ratio of the β-galactosidase activity of the treated sample to that of the untreated sample. For each strain, three independent transformants were assayed.

Multicopy-suppressor Screen

The sod1 strain from BY4743 background was transformed with a YEp13-based yeast chromosomal gene library (2μ, LEU2) and incubated at 30°C on SC media without leucine (SC-LEU). The resulting ∼2000 transformants were replica plated onto SC-LEU medium containing 1.5 μg/ml TM. Plasmids from colonies resistant to TM were isolated individually, transformed into the sod1 mutant of CY4 background and verified as conferring TM resistance. Only plasmids that conferred resistance upon verification were sequenced. A 5.2-kb insert containing ALG7 was identified in two of the positive plasmids and a 5.1 kb insert containing TKL1 was identified in one positive plasmid.

Determination of Intracellular Glutathione Levels

Reduced (GSH) and oxidized glutathione (GSSG) were estimated as described previously (Vandeputte et al., 1994 blue right-pointing triangle; Grant et al., 1996b blue right-pointing triangle). Cells were grown in SC medium (40 ml) to an OD600 of 1 (∼2 × 107 cells/ml), treated or untreated with TM (1 μg/ml) for 4 h and harvested by centrifugation (10,000 × g; 20 s), washed in 3 ml of ice-cold PBS buffer, resuspended in 350 μl of ice-cold 8 mM HCl containing 3% (wt/vol) 5-sulfosalicilic acid and broken using a mini-bead beater (1 min high speed, 4°C). The lysate was clarified by centrifugation (5 min, 10,000 × g; 4°C) and the supernatant taken to determine total glutathione. GSSG was determined by reacting GSH with 2-vinylpyridine before the assay. Levels of GSH and GSSG were calculated as described in (Vandeputte et al., 1994 blue right-pointing triangle).

Estimation of Intracellular Redox Environment

Intracellular values of the GSSG/2GSH half-cell redox potential (glutathione Eh) were determined as an indicator of the cellular redox environment (Schafer and Buettner, 2001 blue right-pointing triangle). These were calculated from intracellular concentrations of GSH and GSSG using the Nernst equation: Eh′ = E0 − 2.303(RT/nF)log10[(GSH)2/(GSSG)]. Glutathione Eh′ values are expressed in mV; E0 is the standard potential for reduced glutathione (−0.24 mV) at pH 7, R is gas constant (8.31 J mol−1 K−1), T is the absolute temperature (303K), n is the number of electrons transferred (2), and F is the Faraday constant (96,406 J V−1 mol−1). To estimate intracellular glutathione concentrations cell volumes were estimated from cell major (a) and minor (b) axes using the formula (Wheals, 1982 blue right-pointing triangle): V = πa2b/6. Cell dimensions were determined microscopically at 100× magnification using a graticule calibrated against a slide micrometer grating. For each strain at least 100 cells were measured, and the average cell volume was determined.

Determination of Intracellular NADP(H) Level

Intracellular NADP(H) levels were determined using a 3-(4,5-dimethylthiazolyl-2)-2,5-diphenyltetrazolium bromide (MTT) and phenazine ethosulfate (PES) cycling assay based on that described by Gibson and Larher (1997 blue right-pointing triangle; Gibon and Larher, 1997) blue right-pointing triangle. Cultures were grown to exponential phase and subjected to the desired treatment condition, two aliquots of cells (equivalent to 10 OD units) were isolated. For quantification of NADPH or NADP+ cells were lysed in 0.1 M KOH or 0.1 M HCl, respectively. Acid extraction was used to destroy NADPH but to preserve NADP+, while alkaline extraction was used to destroy NADP+ and preserve NADPH (Lowry et al., 1961 blue right-pointing triangle). Cells were disrupted using glass beads and a mini-bead beater at 4°C. Cell lysates were incubated on ice for 15 min and clarified by centrifugation (13,000 × g, 10 min). Acid and alkaline extracted cell lysates were incubated for 10 or 5 min, respectively, at 70°C and chilled on ice for 5 min before initiating the assay. Ten microliters of cell lysate was added to 150 μl of 0.26 M triethanolamine hydrochloride buffer (pH 7.4) containing 2 mM glucose-6-phosphate, PES and MTT, and 4 mM EDTA. Glucose-6-phosphate dehydrogenase (specifically prepared for cycling assays; Sigma) was added to a final concentration of 200 U/ml to each reaction. Color development was monitored at 570 nm over 30 min. The concentration of cellular NADP(H) was determined using the linear range of the standard curve (0.5–5 μM) and normalized to the cell density (OD600) of the harvested culture.

Statistical Analysis

Statistical analyses were performed using paired, two-tailed Student's t test.


ROS accumulation has been detected under conditions of sustained ER stress (Haynes et al., 2004 blue right-pointing triangle; Hauptmann et al., 2006 blue right-pointing triangle). To study the relationship between ROS generation and ER function, we sought to identify genes that are involved in the cellular response to chronic ER stress. A previous screen for mutants affected in tolerance to transient ER stress using reducing agents (DTT, 2-mercaptoethanol) or TM highlighted the importance of several cellular functions but did not identify mutants involved in cellular responses to ROS (Chen et al., 2005 blue right-pointing triangle).

Genome-Wide Deletant Screen Identifies Cellular Processes Required for Tolerance to Chronic DTT and Tunicamycin Exposure

To identify genes important for tolerance of chronic ER stress, the S. cerevisiae genome-wide deletion collection was screened for mutants exhibiting sensitivity to TM and DTT. The diploid yeast mutant collection was screened for growth after 72 h in YEPD medium containing either 20 mM DTT or 6 μg/ml TM using 96-well plates. These concentrations completely inhibited growth of the hac1 strain but not of the wild type.

Only those strains sensitive to both compounds were identified as positive to increase the stringency of the screen. As expected, cellular processes previously shown by microarray or mutant analyses to be important for ER stress tolerance, such as the unfolded protein response, cell integrity signaling pathway, vacuolar protein sorting, vacuolar acidification and processes involved in secretory pathway functions (Travers et al., 2000 blue right-pointing triangle; Bonilla et al., 2002 blue right-pointing triangle; Bonilla and Cunningham, 2003 blue right-pointing triangle; Chen et al., 2005 blue right-pointing triangle; Kimata et al., 2006 blue right-pointing triangle) were also identified in this screen (Table 2). However, additional processes including those involved in oxidative stress responses (YAP1, HYR1), NADPH generation (GND1, RPE1, TKL1) and dismutation of the superoxide anion (SOD1 and CCS1) were identified (Table 2). The pentose phosphate pathway is involved in adaptation to, and tolerance of, oxidative stress (Juhnke et al., 1996 blue right-pointing triangle; Thorpe et al., 2004 blue right-pointing triangle; Ng et al., 2008 blue right-pointing triangle). Sod1p detoxifies superoxide anion to H2O2 and water and the Cu2+-loading chaperone Ccs1p is required for its activation (Gralla and Valentine, 1991 blue right-pointing triangle; Culotta et al., 1997 blue right-pointing triangle). These data highlighted that antioxidant functions are very relevant to cell survival of chronic ER stress. Because the main aim was to study the link between ER stress and ROS-related processes, we investigated further the role of SOD and the NADPH-generating system during ER stress tolerance.

Table 2.
Genes whose deletion causes sensitivity to both TM and DTT

Sensitivity of sod1 and ccs1 to Tunicamycin or DTT Is Dependent on Oxygen

Because the sensitivity of the above mutants to agents inducing ER stress might have stemmed from reduced tolerance of ROS generated during ER stress, the sensitivity of each mutant to TM, DTT and paraquat (a superoxide generator) was further assessed on solid SC medium (Figure 1) under aerobic and anaerobic conditions. Hac1p has an established role in the response of cells to ER stress (Cox and Walter, 1996 blue right-pointing triangle; Travers et al., 2000 blue right-pointing triangle); therefore, the hac1 mutant was included in this and subsequent experiments.

Figure 1.
Oxygen-dependent sensitivity of sod1 and ccs1 to TM and DTT. The indicated strains (BY4743 background) were grown for 2 d, the cell density was adjusted to OD600 1.0-, 0.1-, and 0.01-, and 5-μl aliquots were spotted on SC solid medium as control ...

The paraquat sensitivity of sod1 and ccs1 cells under aerobic conditions (Figure 1A), and the reversal of this phenotype under anaerobic conditions (Figure 1B) were consistent with the importance of Cu, Zn SOD in protection against superoxide anion-induced damage (Gralla and Valentine, 1991 blue right-pointing triangle; Culotta et al., 1997 blue right-pointing triangle). Importantly, the TM and DTT sensitivity of the sod1 and ccs1 mutants was completely reversed under anaerobic conditions indicating that the sensitivity of these strains to ER-stress generating agents was oxygen-, and most likely ROS-dependent. Anaerobiosis only partially rescued the TM and DTT sensitivity of the tkl1 mutant, and the DTT sensitivity of the rpe1 mutant, indicating that maintenance of cellular NADPH during ER stress conditions is important regardless of the availability of oxygen. In contrast the TM and DTT sensitivity of the hac1 mutant was not influenced by oxygen availability indicating an essential role of the UPR during ER stress under both aerobic and anaerobic conditions. The oxygen-dependent sensitivity of the sod1 and ccs1 mutants to ER stress indicates that the requirement of Cu, Zn SOD (Sod1p) for ER stress tolerance may be due to its role in detoxification of superoxide anions.

Sod1p and Its Activity Increase during ER Stress

To examine whether Sod1p and its activity were affected under ER stress, the levels of Sod1p and SOD activities due to Sod1p and Sod2p were quantified in wild-type and hac1 cells treated with TM over 8 h.

In wild-type cells, TM treatment led to a small but significant increase in SOD activity (4 h, p < 0.05; Figure 2, A and B). In hac1 cells, the level of Sod1p and SOD activity increased steadily between 2 and 8 h and both Sod1p levels and SOD activity increased approximately threefold after 8 h treatment relative to their respective values before treatment (Figure 2, A and B). Untreated wild-type or hac1 cells showed no increase in Sod1p or its activity during different growth phase examined (Supplemental Figure S1). The activity of Sod2p also increased upon exposure of cells to TM, however in contrast to SOD1, deletion of SOD2 did not lead to DTT or TM sensitivity on solid media (Supplemental Figure S1), indicating a minor role of Sod2p in ER stress tolerance.

Figure 2.
Effect of TM treatment on the level and activity of Sod1p. (A) Induction of Sod1p and SOD activity in wild-type (CY4) and hac1 (ST003) strains. The strains were grown to exponential phase (OD 0.5), TM was added to 1 μg/ml, and samples collected ...

To determine if superoxide was generated during ER stress, dihydroethidium (DHE), a dye relatively specific for this radical (Zhao et al., 2003 blue right-pointing triangle), was used to stain wild-type and hac1 cells treated with TM over a period of 8 h, and cell viability was determined using propidium iodide (PI) staining. In untreated wild-type or hac1 cells there was no increase in DHE or PI staining during the experimental period. For treated wild-type and hac1 cells the percentage of DHE-stained cells gradually increased, with hac1 cells showing a higher percentage of staining compared with the wild type after 4 h (Figure 2C). This increase in DHE-stained cells also correlated with an increase in positive PI staining, although the percentage of PI-stained cells was less than those stained with DHE (Figure 2C).

If superoxide anion production contributes to death of cells during ER stress, then prior adaptation of cells to increase their resistance to superoxide anion should lead to increased survival of subsequent ER stress. Exposure of cells to low, sublethal doses of oxidants confers resistance to subsequent treatment with otherwise lethal doses of the reactive oxygen species (Collinson and Dawes, 1992 blue right-pointing triangle; Jamieson, 1992 blue right-pointing triangle; Flattery-O'Brien et al., 1993 blue right-pointing triangle; Alic et al., 2004 blue right-pointing triangle). To ensure that any cross-protective effect was not mediated through induction of the UPR, the hac1 mutant was used to test this hypothesis. Treatment of hac1 cells with a sublethal dose (25 μM) of paraquat did not lead to a change in colony forming units (CFUs) relative to untreated cells (Figure 2D). Treatment of hac1 cells (not pretreated with paraquat) with TM for 1 or 2 h led to markedly decreased cell viability (CFUs; Figure 2D). In contrast, pretreatment of hac1 cells with 25 μM paraquat for 1 h conferred a significant protective effect to a subsequent treatment with the same dose of TM (Figure 2D). Because paraquat pretreatment conferred protection to TM, these data support the hypothesis that superoxide anion accumulation contributes to cell death during ER stress.

Cells simultaneously deleted for genes encoding the cytosolic and peroxisomal catalases (CTT1 and CTA1), and γ-glutamylcysteine synthetase (GSH1) exhibit hypersensitivity to exogenous H2O2 treatment (Grant et al., 1998 blue right-pointing triangle). To assess whether H2O2 detoxification is important for ER stress tolerance, we tested the sensitivity of gsh1, ctt1 cta1, and ctt1 cta1 gsh1 mutants to DTT and TM. These mutants showed wild-type sensitivity to DTT and TM (Supplemental Figure S1), indicating that these processes involved in detoxification of H2O2 play a minor role in ER stress tolerance.

The above results demonstrate that induction of ER stress in wild-type cells by a low sub-lethal dose (LD) of TM was associated with a transient increase in Sod1p activity. More severe stress imposed by the same dose of TM but in a hac1 deletant led to an extensive increase in superoxide anion production as determined by DHE staining, induction of Sod1p, and increase in Sod1p activity (Figure 2). The present data demonstrate that TM-induced ER stress is associated with accumulation of ROS including superoxide anion.

SOD1 Is Required for Wild-type UPR Induction

Given the central role of the unfolded protein response in ER stress tolerance (Travers et al., 2000 blue right-pointing triangle; Liu and Kaufman, 2003 blue right-pointing triangle; Ron and Walter, 2007 blue right-pointing triangle), we sought to determine whether the extent of UPR induction is dependent on the level of Sod1p activity. Expression levels of two lacZ reporter constructs were determined in the wild-type and sod1 and hac1 cells. The extent of Hac1p activation was determined using a reporter containing the UPR element (UPRE) derived from KAR2 fused to the Escherichia coli lacZ gene (UPRE::lacZ; Cox et al., 1993 blue right-pointing triangle), whereas the other reporter consisted of the 951-base pair upstream sequence of ERO1 fused to the lacZ gene (ERO1::lacZ); this construct indicates the expression of ERO1, which is regulated by Hac1p and the heat-shock transcription factor Hsf1p (Takemori et al., 2006 blue right-pointing triangle).

Consistent with previous findings (Cox et al., 1993 blue right-pointing triangle; Travers et al., 2000 blue right-pointing triangle; Takemori et al., 2006 blue right-pointing triangle), there was almost no induction of either ERO1::lacZ or UPRE::lacZ, respectively, in hac1 cells upon TM treatment (Figure 3). In the sod1 strain, the extent of induction of both ERO1::lacZ and UPRE::lacZ was ∼50% of that observed in wild-type cells, indicating that deletion of SOD1 affects activation of the UPR, and this may contribute to sod1 sensitivity to ER stress observed in a hac1 mutant. Deletion of SOD1 did not affect β-galactosidase activity per se, because the β-galactosidase activity and protein level correlates in the sod1 mutant (Supplemental Figure S2). However, the requirement for Sod1p to remove superoxide generated during ER stress (Figure 2), and the oxygen-dependent sensitivity of sod1 to ER stress (Figure 1) must also account for the sensitivity of the sod1 mutant.

Figure 3.
Cells deleted of SOD1 have reduced UPR induction. Wild-type (CY4), sod1 (ST002), and hac1 (ST003) cells were transformed with (A) ERO1::lacZ or (B) UPRE::lacZ. Transformants were grown to exponential phase (OD600 ∼0.5), and TM was added to 1 μg/ml ...

ERO1 Does Not Contribute to Superoxide Generated during ER Stress

The essential oxidoreductase Ero1p is required for disulphide bond formation in the ER and was previously shown to generate ROS both in vitro and in vivo (Haynes et al., 2004 blue right-pointing triangle; Gross et al., 2006 blue right-pointing triangle). We therefore sought to determine whether Ero1p generates superoxide anions in cells subjected to ER stress.

Carboxypeptidase Y (CPY, encoded by PRC1) is a nonspecific peptidase of the vacuolar lumen that undergoes folding in the ER and maturation in the trans-Golgi network (Stevens et al., 1982 blue right-pointing triangle; Valls et al., 1987 blue right-pointing triangle). Expression of a terminally misfolded form of CPY encoded by PRC1–1, denoted CPY*, causes ER stress and activation of the UPR due to accumulation of this protein in the ER (Hiller et al., 1996 blue right-pointing triangle; Spear and Ng, 2003 blue right-pointing triangle; Haynes et al., 2004 blue right-pointing triangle). Therefore, ERO1 was chronically overexpressed in conjunction with CPY* to induce ER stress.

Because Sod1p levels in wild-type cells vary with carbon source (data not shown), a β-estradiol–inducible system that is compatible with the GAL1 promoter and does not require a change in carbon source (Gao and Pinkham, 2000 blue right-pointing triangle) was used to induce CPY*HA and ERO1myc from plasmids encoding C-terminally HA-tagged PRC1–1 (CPY*HA) and a C-terminally myc-tagged ERO1 (ERO1myc) fused to the GAL1 promoter. The Ero1-myc protein was functional because it rescued the temperature-sensitive ero1–1 mutant at the nonpermissive temperature and increased DTT resistance when overexpressed in wild-type cells (data not shown). Efficacy of the GAL1-CPY*HA construct was established previously (Spear and Ng, 2003 blue right-pointing triangle). Western blot analysis indicated that both proteins could be induced in wild-type and hac1 cells using this system (Figure 4A). Chronic overexpression of CPY*HA and/or ERO1myc had no effect on growth of wild-type cells. However, overexpression of CPY*HA with or without ERO1myc coexpression inhibited growth of UPR-deficient cells (hac1), whereas overexpression of ERO1myc alone had no obvious effect on growth of either wild-type or hac1 cells (Figure 4B).

Figure 4.
Wild-type Ero1p does not contribute to superoxide production during CPY*-induced ER stress. Wild-type (CY4), sod1 (ST002), and hac1 (ST003) cells harboring the pGEV-TRP1 plasmid were transformed with pRS315 or pRS413 and/or pES67 (GAL1-CPY*HA) or pST2 ...

If Ero1p generates appreciable quantities of superoxide in vivo, we surmised that the sod1 mutant may exhibit sensitivity to Ero1p overexpression. Overexpression of CPY*HA and/or ERO1myc had no obvious effect on growth of sod1 cells (Figure 4C). Overexpression of CPY*HA in sod1 cells led to growth inhibition under a non-LD of DTT (2 mM) compared with sod1 cells expressing the control vector alone, supporting the hypothesis that Sod1p is required for wild-type ER stress tolerance. Overexpression of ERO1myc alone or in conjunction with CPY*HA did not further increase the sensitivity of sod1 cells to DTT stress when compared with sod1 cells overexpressing CPY*HA alone. Furthermore, ERO1myc overexpression increased tolerance of sod1 cells to a concentration of DTT (4 mM) that inhibited the growth of the same cells transformed with the control vectors (Figure 4C). These data indicate that the absence of Sod1p does not render cells more sensitive to increased levels of Ero1p in CPY*HA and/or DTT-induced ER stress or to the effect of increased Ero1p expression in the presence of DTT. It is worth noting that overexpression of ERO1myc does not confer increased tolerance of the hac1 mutant to TM (data not shown), indicating that Ero1p protects cells against ER stress in a manner more related to protein thiol redox status.

To examine further whether Ero1p may generate superoxide under ER stress, CPY*HA and/or ERO1myc were acutely overexpressed in hac1 cells that were stained with DHE over a period of 6 h (Figure 4F). Cell density (Figure 4D) and cell viability using PI staining (Figure 4E) were also monitored. Induction of CPY*HA alone or in conjunction with ERO1myc led to a reduction in growth, whereas ERO1myc had a lesser effect on hac1 cells (Figure 4D). Overexpression of CPY*HA led to an increase in DHE-stained cells (Figure 4F), indicating that CPY*HA-induced ER stress also leads to superoxide production as was observed in TM-treated cells (Figure 2, C and D). ERO1myc overexpression alone, or in conjunction with CPY*HA, did not lead to an increase in DHE-stained cells compared with cells overexpressing CPY*HA alone. Furthermore, FACS analysis of these cells after 6 h indicated that hac1 cells overexpressing CPY*HA alone or in conjunction with ERO1myc displayed an identical distribution of the DHE fluorescence (Figure 4G). Overexpression of CPY*HA and/or ERO1myc results in an increase in PI-stained hac1 cells, indicating increased cell death when the level of these proteins was increased.

Ero1p can self-regulate its oxidase activity, because of the presence of two cysteine residues (C150 and C295; Sevier et al., 2007 blue right-pointing triangle). Mutation of these residues to alanine disrupts the ability of Ero1p to regulate its oxidizing activity (Sevier et al., 2007 blue right-pointing triangle). Overexpression of this hyperactive ERO1 allele (ero1-C150A-C295A) inhibits growth of cells in a manner dependent on the oxidase activity of the protein (Sevier et al., 2007 blue right-pointing triangle). To determine if cells expressing the hyperactive allele of Ero1p generate superoxide, ero1-C150A-295A-myc was overexpressed in hac1 cells using the β-estradiol–inducible system. When ero1-C150A-295A-myc was overexpressed in hac1 cells, no obvious difference in growth compared with wild-type ERO1myc overexpression in hac1 cells was observed on solid media using spot test analysis (data not shown). However, a more sensitive approach was used to detect growth differences by monitoring cell density using an automated continuous culture system (Figure 5, A–D). Overexpression of ero1-C150A-C295A-myc resulted in retarded growth. This retardation of growth was dependent on Ero1p hyperactivity because expression of the control inactive form (ero1-C150A-C295A-C100A-C105A-myc; Sevier et al., 2007 blue right-pointing triangle) did not affect growth relative to wild-type ERO1myc or the control vector (Figure 5A). Overexpression of ero1-C150A-C295A-myc increased resistance of hac1 cells to DTT, similar to that observed for overexpression of wild-type ERO1myc (Figure 5B). These data are therefore consistent with the previous observations of Sevier et al. (2007) blue right-pointing triangle.

Figure 5.
An hyperactive allele of ERO1 does not increase superoxide anion production. The hac1 mutant (ST003) harboring the pGEV-TRP1 was transformed with pRS315 or pRS316 (control vectors) and/or pAF112 (GAL1-ERO1myc), pCS452 (GAL1-ero1-C150A-C295A-myc), pCS504 ...

Overexpression of ero1-C150A-295A-myc or wild-type ERO1myc in conjunction with CPY*HA led to a slight increase in growth of hac1 relative to cells overexpressing CPY*HA alone (Figure 5C). A similar trend was observed for the above constructs for hac1 cells treated with DTT (Figure 5D). To determine whether the reduction in growth was associated with increased accumulation of superoxide, the extent of DHE fluorescence in hac1 cells overexpressing ERO1myc or ero1-C150A-295A-myc in the presence or absence of CPY*HA overexpression was determined. Cells expressing ero1-C150A-295A-myc yielded a DHE fluorescence profile similar to those expressing ERO1myc or the control vectors (Figure 5E). Importantly, expressing CPY*HA alone led to increased DHE-associated fluorescence (Figure 5E), and this was not further increased by coexpression of ero1-C150A-295A-myc or ERO1myc. (Figure 5F). These data indicate that Ero1p does not generate appreciable levels of superoxide anions in cells.

The UPR and Sod1p Protect Cells against ER Stress in an Overlapping Pathway

The requirement of Sod1p for wild-type UPR induction and the increase in Sod1p and its activity in cells treated with TM (Figures 2 and and3)3) indicated a link between the unfolded protein response and detoxification of superoxide. To determine whether these two processes function in the same or different pathways for ER-stress tolerance, a sod1 hac1 strain was generated and examined for growth and tolerance to ER stress and the superoxide-generating agent paraquat.

The resulting sod1 hac1 strain was viable and its growth was compared with the respective single deletants and the wild type (Figure 6A). The four strains showed a similar growth rate in early exponential phase. The hac1 strain reached a final cell yield similar to wild-type cells, whereas sod1 and sod1 hac1 strains had a lower yield than the wild type. The sod1 and sod1 hac1 strains showed a reduced growth rate relative to the wild type and hac1 strains as the cells approached diauxic shift/respiratory phase, with the sod1 hac1 strain displaying a more pronounced growth retardation. As cells approach the diauxic shift, ROS generation increases, leading to increased SOD1 transcription (Maris et al., 2001 blue right-pointing triangle). These data indicate that in the absence of Sod1p, the UPR may also play a role in transition of cells in the diauxic shift.

Figure 6.
The unfolded protein response and Sod1p protect cells against ER stress in an overlapping manner. (A) Growth of the sod1 hac1 (ST007) mutant. Indicated strains were grown overnight and reinocculated into fresh medium at a final OD600 of 0.1 and incubated ...

To further elucidate the relationship between Sod1p and the unfolded protein response to ER stress, chemically (i.e., TM and DTT) and genetically (i.e., CPY*) induced ER stresses were used. Although induction of CPY* under the control of a CUP1 promoter at 30°C led to inhibition of hac1 cells, growth of sod1 cells was unaffected (Figure 6B). This result is consistent with the effect of CPY*HA induction under the control of the GAL1 promoter using the inducible β-estradiol system (Figure 4C). In contrast, at 37°C (a temperature known to increase unfolded/misfolded proteins), induction of CPY* using 400 or 800 μM CuSO4 resulted in pronounced inhibition of sod1 cells. This inhibition was not due to the presence of elevated Cu2+ because growth of sod1 cells was similar to that of wild-type cells expressing the empty vector in Cu2+-containing media (Figure 6B). The sod1 hac1 mutant was also more sensitive to CPY* induction (Figure 6B) at 30 or 37°C relative to the sod1 and hac1 single mutants. The ER stress sensitivity of sod1 cells was therefore correlated with the level of ER stress experienced by cells, because the sod1 mutant with a defective UPR was more susceptible to ER stress as observed in the double mutant.

Deletion of SOD1 in hac1 cells also led to increased sensitivity to either TM or DTT (Figure 6C). Moreover, although growth of the sod1 strain under ER stress was rescued under anaerobic conditions, that of the sod1 hac1 double mutant was not (Figure 6C). This indicates that the effects of SOD1 deletion in the hac1 mutant were due to accumulation of ROS. The sensitivity of sod1 hac1 to the superoxide-generating agent paraquat reflected that of the sod1 deletant, indicating that the UPR plays little role in tolerance of superoxide.

These results further support the hypothesis that ER stress leads to ROS production (Harding et al., 2003 blue right-pointing triangle; Haynes et al., 2004 blue right-pointing triangle; Hauptmann et al., 2006 blue right-pointing triangle). The ROS are likely to be superoxide because SOD activity was up-regulated in wild-type and hac1 cells under ER stress (Figure 2, A and B). Furthermore, deletion of SOD1 increased sensitivity of wild-type or UPR-deficient (hac1) cells to all conditions that generate ER stress (TM, DTT, and CPY*). We therefore rationalized that Sod1p protects cells against ER stress both in the presence and absence of the UPR. Collectively, these results indicate that SOD activity and the unfolded protein response are both required for maximal tolerance of chronic ER stress, and these two processes function in an overlapping manner.

Multiple Copies of TKL1 Suppress the Sensitivity of the sod1 Mutant to TM

To identify additional factors linking Sod1p to resistance to ER stress, a screen for high-copy suppressors of the TM sensitivity of sod1 cells was performed. The sod1 mutant was transformed with a multicopy gene library and the resulting ∼2000 transformants were replica-plated on to media containing a lethal concentration (LC) of TM that would otherwise inhibited growth of sod1 cells.

Two plasmids conferred TM resistance on sod1 cells: the first harbored ALG7 and the second TKL1. ALG7 encodes dolichyl-P–dependent N-acetylglucosamine-1-P transferase, a known target of TM (Barnes et al., 1984 blue right-pointing triangle; Kukuruzinska and Lennon, 1995 blue right-pointing triangle), whereas TKL1 encodes a transketolase isoform that acts in the pentose phosphate pathway that is important for NADPH generation. Overexpression of TKL1 also rescued the sensitivity of sod1 cells to DTT (data not shown).

If the rescue of sod1 to ER stress by TKL1 overexpression was due to an increase in the generation of NADPH, then this should rely on Zwf1p, the first committed step of the pentose phosphate pathway (Nogae and Johnston, 1990 blue right-pointing triangle). Overexpressing TKL1 in the wild-type and zwf1 and zwf1 sod1 strains had minimal effect on their tolerance to TM, whereas the sod1 strain showed increased resistance to this stress (Figure 7A). Similar results were obtained with DTT (data not shown). The effect of TKL1 overexpression on sod1 was not a general effect on UPR-deficient cells because TKL1 overexpression did not suppress the sensitivity of hac1 to TM (Figure 7B). The data verify that the rescue of sod1 sensitivity to ER stress by TKL1 overexpression depends on the NADPH-generating components of the pentose phosphate pathway. From Figure 7A, the zwf1 sod1 strain was more sensitive to ER stress than the sod1 strain, indicating that there is some contribution of ZWF1 to combating ER stress.

Figure 7.
Overexpression of TKL1 rescues sod1 from ER stress, affects NADP(H) levels, and restores UPR induction of the sod1 mutant. (A) Wild-type (1783), sod1 (KS105), zwf1 (KS113), and zwf1 sod1 (KS117) and (B) wild-type (CY4), sod1 (ST002), and hac1 (ST003) ...

To examine the effect of TKL1 overexpression on NADP(H) homeostasis of sod1 cells, NADP+ and NADPH levels were determined in the presence and absence of TM treatment (Figure 7C). Overexpression of TKL1 led to a significant increase in the total pool of NADP(H). Surprisingly, this was due to increased accumulation of NADP+ (Figure 7C). TM treatment of sod1 cells overexpressing TKL1 did not affect NADP(H) homeostasis further relative to TKL1 overexpression alone. This is interesting because exposure of sod1 cells (without TKL1 overexpression) to TM led to a ∼40% increase in NADP+. These data indicate there is an increased burden on NADP(H) homeostasis and on the NADPH-generating systems during ER stress. Because the level of NADPH did not change upon TM treatment and/or after TKL1 overexpression, one possible explanation of this data is that the protective affect of TKL1 overexpression on sod1 ER-stress tolerance (Figure 7A and B) may be due to altered flux through the NADPH-generating pathway.

Mutants Affected in NADPH Synthesis Are Affected in Induction of the UPR

Because TKL1 overexpression affected the NADP(H) homeostasis of sod1 cells and conferred increased tolerance of these cells to ER stress (Figure 7, A and B) and because sod1 cells exhibit reduced UPR induction (Figure 3), we wanted to see whether TKL1 overexpression rescued the unfolded protein response in the sod1 mutant.

Overexpression of TKL1 restored ERO1::lacZ induction in sod1 and had no effect on wild-type cells (Figure 7D). These data indicate the failure of the sod1 strain to fully induce the UPR can be attributed to disruption of NADP(H) homeostasis. If perturbation of NADP(H) homeostasis leads to reduced UPR induction, we rationalized that mutants affected in NADPH production would also be affected in UPR induction. The extent of TM induction of both ERO1::lacZ and UPRE::lacZ reporter constructs was significantly lower in tkl1 and rpe1 strains (defective in components of the pentose phosphate pathway) when compared with the wild type (Figure 7, E and F). These data indicate that disruption of NADP(H) homeostasis affects the induction of the UPR and provide insight into the effects of cellular NADP(H) on the unfolded protein response.

TM-induced ER Stress Alters Cellular Glutathione Levels

NADPH is important for numerous processes including maintenance of the thioredoxin and the glutathione redox couples. Given the important role of glutathione in regulation of ER oxidation state (Cuozzo and Kaiser, 1999 blue right-pointing triangle), the impact of secretory pathway dysfunction on cellular glutathione homeostasis (Haynes et al., 2004 blue right-pointing triangle; Perrone et al., 2005 blue right-pointing triangle), and the proposal that increased GSH levels induce the UPR (Trotter and Grant, 2002 blue right-pointing triangle), we sought to examine whether Sod1p is required for maintaining intracellular glutathione homeostasis of cells under ER stress.

TM treatment (4 h) led to increased GSSG and GSH in wild-type cells, as well as hac1, sod1, and sod1 hac1 mutants (Figure 8, A and B). This increase was more marked in the sod1 hac1 double mutant. In untreated cells of sod1 and sod1 hac1, GSH and GSSG levels were also elevated, and this increased further on TM treatment. Overexpression of TKL1 had little effect on GSSG or GSH levels in most strains with or without TM, only in the sod1 hac1 double mutant was there a significant (p < 0.05) reduction in GSSG and GSH (Figure 8, A and B). Hence TKL1 overexpression only affected glutathione levels under extreme ER stress, and loss of Sod1p caused by TM treatment of the sod1 hac1 mutant.

Figure 8.
SOD1 is required to maintain cellular redox balance. (A) GSH and (B) GSSG were estimated in cells of the indicated strains untreated, or treated with TM, for 4 h. Error bars, the SE of at least four replicates from three independent experiments. (C) Cellular ...

Because the levels of GSH and GSSG reflect cellular redox state (Schafer and Buettner, 2001 blue right-pointing triangle; Drakulic et al., 2005 blue right-pointing triangle) and glutathione is the major redox buffer in most cells (Schafer and Buettner, 2001 blue right-pointing triangle), we estimated the cellular Eh′ of sod1, hac1, and sod1 hac1 strains under TM stress using the GSH/GSSG redox couple (Figure 8C). We also determined whether TKL1 overexpression had any effect on cellular Eh′ in these strains. The sod1 and sod1 hac1 strains had a significantly more oxidizing Eh′ compared with wild-type or hac1 cells (p < 0.005). Although deletion of SOD1, TM-induced ER stress, and TKL1 overexpression all affected GSH or GSSG levels in some way, only loss of Sod1p caused any substantial change (p < 0.005) in cellular glutathione redox potential. Treatment of cells with TM with or without TKL1 overexpression barely affected cellular redox in all strains tested. This is consistent with the observation that cells can maintain their cellular redox state even when treated with H2O2 (Ng et al., 2008 blue right-pointing triangle).

Because Sod1p is important to maintain ER stress tolerance and cellular glutathione redox potential, we sought to determine whether a change to more oxidizing conditions by deletion of the glutathione reductase gene (GLR1) would affect ER stress tolerance further. Intriguingly, we were unable to generate the double sod1 glr1 mutant under aerobic conditions, but did recover it anaerobically (Supplemental Figure S3). The mutant displayed a severe oxygen-dependent growth defect, which was partially rescued by TKL1 overexpression because dissection of a heterozygous diploid carrying the overexpression plasmid enabled aerobic isolation of the double mutant. Even this condition led to poor aerobic growth, which precluded use of the strain for analysis of ER stress tolerance.

Interestingly, glutathione may play a less direct role than NADPH during ER stress because deletion of GLR1 (Muller, 1996 blue right-pointing triangle; Grant, 2001 blue right-pointing triangle) or GSH1 did not result in TM or DTT sensitivity (data not shown).


The genome-wide screen for mutants sensitive to chronic TM and DTT stress identified genes encoding functions including the Cu, Zn SOD (Sod1p), the copper chaperone of Sod1p (Ccs1p), and those involved in NADPH generation to be important for chronic ER-stress tolerance. Several observations support the hypothesis that superoxide accumulation affects ER stress tolerance and that ER stress is associated with increased generation of superoxide: 1) the oxygen-dependent sensitivity of sod1 and ccs1 to TM and DTT; 2) the increase in Sod1p levels and Sod1p activity after treatment of cells with TM; 3) increased detection of superoxide anions in cells overexpressing CPY* or treated with TM; and 4) preadaptation of cells to superoxide affords protection to ER stress. These observations also demonstrate the importance of Sod1p activity in cell survival during ER stress. Although ROS production upstream of cell death during ER stress has previously been demonstrated (Hauptmann et al., 2006 blue right-pointing triangle; Haynes et al., 2004 blue right-pointing triangle), here we show that superoxide is at least one of the species of ROS produced during ER stress. The preadaptation data, together with the specificity of the Sod1p effect (mutants affecting other antioxidant functions do not show sensitivity to ER stress) indicate that there is a link between ER stress and superoxide production, and it is very unlikely that all of the above observations are due to a separate and independent additive effect of loss of Sod1p and of ER stress. However it is possible that some of the ROS detected during ER stress were a consequence of the onset of cell death.

ROS have been reported to accumulate during ER stress not only in S. cerevisiae (Haynes et al., 2004 blue right-pointing triangle; Hauptmann et al., 2006 blue right-pointing triangle) but also in Caenorhabditis elegans (Harding et al., 2003 blue right-pointing triangle). ROS and ER stress play an important role in pathogenesis of many human diseases including diabetes, atherosclerosis, and neurodegenerative diseases (Kincaid and Cooper, 2007 blue right-pointing triangle; Malhotra and Kaufman, 2007 blue right-pointing triangle). In each of these cases ROS accumulation was detected, but the species present was not fully elucidated. Despite this, the extent of ROS production was shown to be affected by the level of ERO1 (Harding et al., 2003 blue right-pointing triangle; Haynes et al., 2004 blue right-pointing triangle).

In this study, overexpression of CPY* in hac1 cells led to increased production of superoxide. However, overexpression of ERO1 alone, or in conjunction with CPY*, did not lead to any further increase in the level of superoxide. Furthermore, overexpression of Ero1p did not increase sensitivity of sod1 cells to CPY* and/or DTT exposure. These results indicate that appreciable levels of superoxide anions are unlikely to be generated during Ero1p-dependent oxidative protein folding under the conditions used.

Ero1p generates stoichiometric quantities of H2O2 during disulphide bond formation in vitro (Gross et al., 2006 blue right-pointing triangle). H2O2 accumulation during ER stress may lead to particularly serious consequences for the cell if superoxide accumulation occurred simultaneously since reduction of metals ions including Fe3+ and Cu2+ by superoxide can promote generation of the highly reactive hydroxyl radical from H2O2 via Haber-Weiss and Fenton reactions (Perrone et al., 2008 blue right-pointing triangle). Although cells possess effective means for selective removal of superoxide and H2O2, this is not the case for hydroxyl radicals (Perrone et al., 2008 blue right-pointing triangle). It is therefore important for cells to possess a mechanism to modulate Ero1p to minimize excess H2O2 production. Sevier et al. (2007) blue right-pointing triangle have shown that self-regulation of Ero1p oxidase activity is regulated by oxidation of two cysteine residues at positions 150 and 295 of the protein. The authors proposed that these cysteine residues play a role in minimizing excess H2O2 production by inhibiting continuous reduction of Ero1p by oxygen, thereby avoiding genesis of an hyper-oxidizing condition in the ER. Because ER stress induces expression of ERO1 (Travers et al., 2000 blue right-pointing triangle; Takemori et al., 2006 blue right-pointing triangle), regulation of Ero1p activity may be important to minimize H2O2 production under conditions of elevated superoxide anion accumulation, e.g., during ER stress as demonstrated in this study. Here, however, overexpression of the mutant ero1-C150A-295A did not lead to overaccumulation of superoxide, further supporting the proposal that Ero1p is not a source of superoxide in vivo.

Interestingly, although simultaneous deletion of cytosolic and peroxisomal catalases (CTT1, CTA1) and γ-glutamyl-cysteine synthetase (GSH1) leads to extreme sensitivity to exogenous H2O2, (Grant et al., 1998 blue right-pointing triangle), it did not affect tolerance to ER-stressing agents. Overexpression of ERO1 in the ctt1 cta1 gsh1 triple mutant did not inhibit growth in the presence or absence of the ER-stressing agent DTT (data not shown). These data indicate that although it is possible that Ero1p generates H2O2 in vivo, the levels produced are not sufficient to overwhelm any residual H2O2 detoxification activity or damage coping capacity present in the ctt1 cta1 gsh1 triple mutant. This is in contrast to the reduced tolerance of this mutant to exogenous H2O2 when this oxidant was added extracellularly at millimolar concentrations (Grant et al., 1998 blue right-pointing triangle).

The data above appear to conflict with previous findings by Haynes et al. (2004) blue right-pointing triangle who demonstrated a role for Ero1p in ROS generation during ER stress in vivo. This difference may be due to the nature of ER stress experienced by cells and to different strain backgrounds. Here, overexpression of CPY* in hac1 cells led to increased superoxide production, and increasing Ero1p levels did not affect the level of superoxide produced. In the study by Haynes et al. (2004) blue right-pointing triangle, whereas CPY* was used to generate ER stress, ROS production associated with ERO1 overexpression was demonstrated to occur in ire1 erv29 cells. The possible contribution of strain background in influencing the outcome is further supported by the demonstration that deletion of IRE1 in erv29, harboring a constitutively active allele of HSF1, led to reduced tolerance to TM and CPY* (Liu and Chang, 2008 blue right-pointing triangle). In contrast, Haynes et al. demonstrated that deletion of IRE1 in the erv29 mutant reduced sensitivity to CPY* overexpression. Additional genetic differences between the strain backgrounds used to conduct these studies may also have affected observations made. This may include difference in genetic backgrounds leading to altered tolerance of strains to ROS as observed previously (Veal et al., 2003 blue right-pointing triangle).

Superoxide accumulation in the sod1 mutant also reduced tolerance to ER stress by dampening activation of the UPR. These changes, at least in part, were mediated through effects of superoxide on NADP(H) homeostasis because overexpression of TKL1, reversed sensitivity of sod1 cells to TM and DTT, and restored UPR induction to wild-type levels. Moreover, mutants disrupted in the nonoxidative branch of NADPH synthesis exhibited a defect in UPR induction that correlated with sensitivity to ER stress. A model summarizing the role of Sod1p and NADP(H) homeostasis in ER stress tolerance is given in Figure 9.

Figure 9.
Proposed model for the role of Sod1p under ER stress and its effect on cellular NADP(H), glutathione, and the unfolded protein response.

The UPR is essential to counter ER stress (Travers et al., 2000 blue right-pointing triangle). The decreased tolerance of sod1 cells to ER stress is at least in part due to the effects of superoxide accumulation on NADP(H) homeostasis, which in turn affects the magnitude of UPR induction (Figure 9). This proposal is supported by the observation that TKL1 overexpression rescued the ER stress sensitivity and the UPR induction defect of sod1 cells. However, TKL1 overexpression did not fully restore the sensitivity of sod1 cells to DTT or TM in contrast to anaerobiosis, indicating that superoxide accumulation in cells also reduces ER stress tolerance through mechanisms other than UPR induction. Therefore, the sensitivity of sod1 cells to ER stress is primarily attributed to superoxide accumulation, and this accumulation leads to disruption of both NADP(H) homeostasis and UPR induction (Figure 9). Increased superoxide accumulation in hac1 cells does not appear to affect ER stress tolerance through altered NADP(H) homeostasis because TKL1 overexpression did not affect their tolerance to ER stress. Moreover, it is unlikely that superoxide accumulation per se in hac1 cells during ER stress was a primary cause of death because anaerobiosis did not increase tolerance of these cells to ER stress.

NADPH is required for maintenance of the thioredoxin and glutathione redox systems. The effect of deleting SOD1 or affecting the NADPH biosynthesis pathway on induction of the UPR may therefore be due to changes in the glutathione and/or thioredoxin redox couples. It is unlikely that increased accumulation of oxidized thioredoxin led to reduced UPR induction in the above strains, because deletion of thioredoxin reductase (Trr1p) causes oxidized thioredoxin to accumulate (Holmgren, 1989 blue right-pointing triangle) but leads to hyper-induction of the UPR and increased resistance to TM (Trotter and Grant, 2002 blue right-pointing triangle). Deletion of TRR1 also led to a approximately threefold increase in intracellular GSH accumulation, leading to the proposal that the elevated UPR in trr1 mutants is a consequence of elevated GSH levels, which in turn leads to accumulation of misfolded proteins in the ER (Trotter and Grant, 2002 blue right-pointing triangle). Overexpression of GSH1 also leads to a 30% increase in UPR induction (Cuozzo and Kaiser, 1999 blue right-pointing triangle). However, sod1 (this study) and tkl1 and rpel (Ng et al., 2008 blue right-pointing triangle) mutants all accumulate higher levels of GSH relative to wild-type cells, yet exhibit a reduced UPR induction. These data indicate that elevated GSH in sod1, tkl1, and rpe1 mutants is unlikely to account for altered UPR induction in these cells. Disruption of SOD1, TKL1, or RPE1 also leads to increased accumulation of oxidized glutathione (GSSG) in cells. Deletion of GLR1 also leads to increased accumulation of GSSG, but this mutant exhibited a similar induction of the UPR as wild-type cells in response to DTT (Trotter and Grant, 2002 blue right-pointing triangle). Analysis of the intracellular glutathione redox potential showed that sod1 cells had a more oxidizing environment (this study), whereas the tkl1 and rpe1 mutants display a wild-type redox potential (Ng et al., 2008 blue right-pointing triangle). Therefore the altered UPR observed in sod1, tkl1, and rpe1 cells was more closely correlated with altered NADP(H) homeostasis and was unlikely to have resulted from altered thioredoxin, GSH, GSSG, or the intracellular redox environment.

ER stress caused a change in NADP(H) homeostasis, whereby the level of total NADP(H) and NADP+ increased significantly during this stress, and the absolute level of NADPH in sod1 cells during ER stress was comparable to that in untreated cells. This indicates that ER stress is unlikely to disrupt NADPH synthesis directly because this would be expected to lead to increased NADP+ at the expense of NADPH. Rather ER stress appears to increase the burden on the NADPH-generating system leading to accumulation of NADP+. Overexpression of TKL1 therefore may afford protection of sod1 cells to ER stress, not only through reversal of the UPR defect, but also by increasing flux through the NADPH-generating system, thereby facilitating processes that require NADPH.

This study highlights the vital role of Sod1p and NADP(H) homeostasis in induction of the UPR and survival of cells during ER stress. Accumulation of superoxide in cells may lead to reduced UPR induction and elevated ER stress. Because the role of the UPR is to alleviate ER stress and ER stress is associated with elevated superoxide production, accumulation of this ROS in cells before and/or during ER stress may pose serious consequences for cell survival by promoting a downward spiral toward cell death. H2O2 formation by the ER or other processes in the cell may further increase the deleterious consequences of ER stress and superoxide accumulation, by promoting production of the highly reactive hydroxyl radical. These findings illustrate the interconnected nature of ER stress, superoxide anions, and the UPR and NADPH homeostatic systems, which may have important implications for disease processes particularly those associated with ER stress and ROS, including neurodegenerative disease and diabetes.

Supplementary Material

[Supplemental Materials]


The authors thank Antony Cooper, Chris Kaiser, Carolyn Sevier, Davis Ng, Joe Mymryk, Munira Basrai, Peter Walter, and Valeria Culotta (Johns Hopkins University, Baltimore, MD) for strains and plasmids provided in this study and Anna Kitanovic for advice on the NADP(H) assay. S.-X.T. is a recipient of an Endeavor International Postgraduate Research Scholarship from the Australian government. This research was supported by grants from the Australian Research Council and Cancer Council New South Wales.

Abbreviations used:

ERendoplasmic reticulum
ROSreactive oxygen species
UPRunfolded protein response.


This article was published online ahead of print in MBC in Press (http://www.molbiolcell.org/cgi/doi/10.1091/mbc.E08-07-0697) on January 7, 2009.


  • Alic N., Felder T., Temple M. D., Gloeckner C., Higgins V. J., Briza P., Dawes I. W. Genome-wide transcriptional responses to a lipid hydroperoxide: adaptation occurs without induction of oxidant defenses. Free Radic. Biol. Med. 2004;37:23–35. [PubMed]
  • Barnes G., Hansen W. J., Holcomb C. L., Rine J. Asparagine-linked glycosylation in Saccharomyces cerevisiae: genetic analysis of an early step. Mol. Cell. Biol. 1984;4:2381–2388. [PMC free article] [PubMed]
  • Bonilla M., Cunningham K. W. Mitogen-activated protein kinase stimulation of Ca(2+) signaling is required for survival of endoplasmic reticulum stress in yeast. Mol. Biol. Cell. 2003;14:4296–4305. [PMC free article] [PubMed]
  • Bonilla M., Nastase K. K., Cunningham K. W. Essential role of calcineurin in response to endoplasmic reticulum stress. EMBO J. 2002;21:2343–2353. [PMC free article] [PubMed]
  • Chen Y., Feldman D. E., Deng C., Brown J. A., De Giacomo A. F., Gaw A. F., Shi G., Le Q. T., Brown J. M., Koong A. C. Identification of mitogen-activated protein kinase signaling pathways that confer resistance to endoplasmic reticulum stress in Saccharomyces cerevisiae. Mol. Cancer Res. 2005;3:669–677. [PubMed]
  • Collinson L. P., Dawes I. W. Inducibility of the response of yeast cells to peroxide stress. J. Gen. Microbiol. 1992;138:329–335. [PubMed]
  • Cox J. S., Shamu C. E., Walter P. Transcriptional induction of genes encoding endoplasmic reticulum resident proteins requires a transmembrane protein kinase. Cell. 1993;73:1197–1206. [PubMed]
  • Cox J. S., Walter P. A novel mechanism for regulating activity of a transcription factor that controls the unfolded protein response. Cell. 1996;87:391–404. [PubMed]
  • Culotta V. C. Superoxide dismutase, oxidative stress, and cell metabolism. Curr. Top. Cell Regul. 2000;36:117–132. [PubMed]
  • Culotta V. C., Klomp L. W., Strain J., Casareno R. L., Krems B., Gitlin J. D. The copper chaperone for superoxide dismutase. J. Biol. Chem. 1997;272:23469–23472. [PubMed]
  • Cuozzo J. W., Kaiser C. A. Competition between glutathione and protein thiols for disulphide-bond formation. Nat. Cell Biol. 1999;1:130–135. [PubMed]
  • Deere D., Shen J., Vesey G., Bell P., Bissinger P., Veal D. Flow cytometry and cell sorting for yeast viability assessment and cell selection. Yeast. 1998;14:147–160. [PubMed]
  • Drakulic T., Temple M. D., Guido R., Jarolim S., Breitenbach M., Attfield P. V., Dawes I. W. Involvement of oxidative stress response genes in redox homeostasis, the level of reactive oxygen species, and ageing in Saccharomyces cerevisiae. FEMS Yeast Res. 2005;12:1215–1228. [PubMed]
  • Flattery-O'Brien J., Collinson L. P., Dawes I. W. Saccharomyces cerevisiae has an inducible response to menadione which differs from that to hydrogen peroxide. J. Gen. Microbiol. 1993;139:501–507. [PubMed]
  • Gao C. Y., Pinkham J. L. Tightly regulated, beta-estradiol dose-dependent expression system for yeast. Biotechniques. 2000;29:1226–1231. [PubMed]
  • Gibon Y., Larher F. Cycling assay for nicotinamide adenine dinucleotides: NaCl precipitation and ethanol solubilization of the reduced tetrazolium. Anal. Biochem. 1997;251:153–157. [PubMed]
  • Gralla E. B., Valentine J. S. Null mutants of Saccharomyces cerevisiae Cu, Zn superoxide dismutase: characterization and spontaneous mutation rates. J. Bacteriol. 1991;173:5918–5920. [PMC free article] [PubMed]
  • Grant C. M. Role of the glutathione/glutaredoxin and thioredoxin systems in yeast growth and response to stress conditions. Mol. Microbiol. 2001;39:533–541. [PubMed]
  • Grant C. M., Collinson L. P., Roe J. H., Dawes I. W. Yeast glutathione reductase is required for protection against oxidative stress and is a target gene for yAP-1 transcriptional regulation. Mol. Microbiol. 1996a;21:171–179. [PubMed]
  • Grant C. M., MacIver F. H., Dawes I. W. Glutathione is an essential metabolite required for resistance to oxidative stress in the yeast Saccharomyces cerevisiae. Curr. Genet. 1996b;29:511–515. [PubMed]
  • Grant C. M., Perrone G., Dawes I. W. Glutathione and catalase provide overlapping defenses for protection against hydrogen peroxide in the yeast Saccharomyces cerevisiae. Biochem. Biophys. Res. Commun. 1998;253:893–898. [PubMed]
  • Gross E., Sevier C. S., Heldman N., Vitu E., Bentzur M., Kaiser C. A., Thorpe C., Fass D. Generating disulfides enzymatically: reaction products and electron acceptors of the endoplasmic reticulum thiol oxidase Ero1p. Proc. Natl. Acad. Sci. USA. 2006;103:299–304. [PMC free article] [PubMed]
  • Harding H. P., et al. An integrated stress response regulates amino acid metabolism and resistance to oxidative stress. Mol. Cell. 2003;11:619–633. [PubMed]
  • Hauptmann P., Riel C., Kunz-Schughart L. A., Frohlich K. U., Madeo F., Lehle L. Defects in N-glycosylation induce apoptosis in yeast. Mol. Microbiol. 2006;59:765–778. [PubMed]
  • Haynes C. M., Titus E. A., Cooper A. A. Degradation of misfolded proteins prevents ER-derived oxidative stress and cell death. Mol. Cell. 2004;15:767–776. [PubMed]
  • Hermann H., Hacker U., Bandlow W., Magdolen V. pYLZ vectors: Saccharomyces cerevisiae/Escherichia coli shuttle plasmids to analyze yeast promoters. Gene. 1992;119:137–141. [PubMed]
  • Hiller M. M., Finger A., Schweiger M., Wolf D. H. ER degradation of a misfolded luminal protein by the cytosolic ubiquitin-proteasome pathway. Science. 1996;273:1725–1728. [PubMed]
  • Holmgren A. Thioredoxin and glutaredoxin systems. J. Biol. Chem. 1989;264:13963–13966. [PubMed]
  • Huang T. T., Raineri I., Eggerding F., Epstein C. J. Transgenic and mutant mice for oxygen free radical studies. Methods Enzymol. 2002;349:191–213. [PubMed]
  • Ito H., Fukuda Y., Murata K., Kimura A. Transformation of intact yeast cells treated with alkali cations. J. Bacteriol. 1983;153:163–168. [PMC free article] [PubMed]
  • Jamieson D. J. Saccharomyces cerevisiae has distinct adaptive responses to both hydrogen peroxide and menadione. J. Bacteriol. 1992;174:6678–6681. [PMC free article] [PubMed]
  • Jamsa E., Simonen M., Makarow M. Selective retention of secretory proteins in the yeast endoplasmic reticulum by treatment of cells with a reducing agent. Yeast. 1994;10:355–370. [PubMed]
  • Juhnke H., Krems B., Kotter P., Entian K. D. Mutants that show increased sensitivity to hydrogen peroxide reveal an important role for the pentose phosphate pathway in protection of yeast against oxidative stress. Mol. Gen. Genet. 1996;252:456–464. [PubMed]
  • Kawahara T., Yanagi H., Yura T., Mori K. Endoplasmic reticulum stress-induced mRNA splicing permits synthesis of transcription factor Hac1p/Ern4p that activates the unfolded protein response. Mol. Biol. Cell. 1997;8:1845–1862. [PMC free article] [PubMed]
  • Kimata Y., Ishiwata-Kimata Y., Yamada S., Kohno K. Yeast unfolded protein response pathway regulates expression of genes for anti-oxidative stress and for cell surface proteins. Genes Cells. 2006;11:59–69. [PubMed]
  • Kincaid M. M., Cooper A. A. ERADicate ER stress or die trying. Antioxid. Redox Signal. 2007;9:2373–2387. [PubMed]
  • Kukuruzinska M. A., Lennon K. Diminished activity of the first N-glycosylation enzyme, dolichol-P-dependent N-acetylglucosamine-1-P transferase (GPT), gives rise to mutant phenotypes in yeast. Biochim. Biophys. Acta. 1995;1247:51–59. [PubMed]
  • Liu C. Y., Kaufman R. J. The unfolded protein response. J. Cell Sci. 2003;116:1861–1862. [PubMed]
  • Liu Y., Chang A. Heat shock response relieves ER stress. EMBO J. 2008;27:1049–1059. [PMC free article] [PubMed]
  • Lowry O. H., Passonneau J. V., Rock M. K. The stability of pyridine nucleotides. J. Biol. Chem. 1961;236:2756–2759. [PubMed]
  • Madeo F., Frohlich E., Ligr M., Grey M., Sigrist S. J., Wolf D. H., Frohlich K. U. Oxygen stress: a regulator of apoptosis in yeast. J. Cell Biol. 1999;145:757–767. [PMC free article] [PubMed]
  • Malhotra J. D., Kaufman R. J. Endoplasmic reticulum stress and oxidative stress: a vicious cycle or a double-edged sword? Antioxid. Redox Signal. 2007;9:2277–2293. [PubMed]
  • Maris A. F., Assumpcao A. L., Bonatto D., Brendel M., Henriques J. A. Diauxic shift-induced stress resistance against hydroperoxides in Saccharomyces cerevisiae is not an adaptive stress response and does not depend on functional mitochondria. Curr. Genet. 2001;39:137–149. [PubMed]
  • Mori K., Kawahara T., Yoshida H., Yanagi H., Yura T. Signalling from endoplasmic reticulum to nucleus: transcription factor with a basic-leucine zipper motif is required for the unfolded protein-response pathway. Genes Cells. 1996;1:803–817. [PubMed]
  • Muller E. G. A glutathione reductase mutant of yeast accumulates high levels of oxidized glutathione and requires thioredoxin for growth. Mol. Biol. Cell. 1996;7:1805–1813. [PMC free article] [PubMed]
  • Ng C. H., Tan S. X., Perrone G. G., Thorpe G. W., Higgins V. J., Dawes I. W. Adaptation to hydrogen peroxide in Saccharomyces cerevisiae: the role of NADPH-generating systems and the SKN7 transcription factor. Free Radic. Biol. Med. 2008;44:1131–1145. [PubMed]
  • Niwa M., Patil C. K., DeRisi J., Walter P. Genome-scale approaches for discovering novel nonconventional splicing substrates of the Ire1 nuclease. Genome Biol. 2005;6:R3. [PMC free article] [PubMed]
  • Nogae I., Johnston M. Isolation and characterization of the ZWF1 gene of Saccharomyces cerevisiae, encoding glucose-6-phosphate dehydrogenase. Gene. 1990;96:161–169. [PubMed]
  • Perrone G. G., Grant C. M., Dawes I. W. Genetic and environmental factors influencing glutathione homeostasis in Saccharomyces cerevisiae. Mol. Biol. Cell. 2005;16:218–230. [PMC free article] [PubMed]
  • Perrone G. G., Tan S. X., Dawes I. W. Reactive oxygen species and yeast apoptosis. Biochim. Biophys. Acta. 2008;1783:1354–1368. [PubMed]
  • Ron D., Walter P. Signal integration in the endoplasmic reticulum unfolded protein response. Nat. Rev. Mol. Cell Biol. 2007;8:519–529. [PubMed]
  • Rose M., Botstein D. Construction and use of gene fusions to lacZ (beta-galactosidase) that are expressed in yeast. Methods Enzymol. 1983;101:167–180. [PubMed]
  • Rosen D. R., et al. Mutations in Cu/Zn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis. Nature. 1993;362:59–62. [PubMed]
  • Sanchez R. J., Srinivasan C., Munroe W. H., Wallace M. A., Martins J., Kao T. Y., Le K., Gralla E. B., Valentine J. S. Exogenous manganous ion at millimolar levels rescues all known dioxygen-sensitive phenotypes of yeast lacking CuZnSOD. J. Biol. Inorg. Chem. 2005;10:913–923. [PubMed]
  • Schafer F. Q., Buettner G. R. Redox environment of the cell as viewed through the redox state of the glutathione disulfide/glutathione couple. Free Radic. Biol. Med. 2001;30:1191–1212. [PubMed]
  • Sevier C. S., Qu H., Heldman N., Gross E., Fass D., Kaiser C. A. Modulation of cellular disulfide-bond formation and the ER redox environment by feedback regulation of Ero1. Cell. 2007;129:333–344. [PubMed]
  • Slekar K. H., Kosman D. J., Culotta V. C. The yeast copper/zinc superoxide dismutase and the pentose phosphate pathway play overlapping roles in oxidative stress protection. J. Biol. Chem. 1996;271:28831–28836. [PubMed]
  • Spear E. D., Ng D. T. Stress tolerance of misfolded carboxypeptidase Y requires maintenance of protein trafficking and degradative pathways. Mol. Biol. Cell. 2003;14:2756–2767. [PMC free article] [PubMed]
  • Stevens T., Esmon B., Schekman R. Early stages in the yeast secretory pathway are required for transport of carboxypeptidase Y to the vacuole. Cell. 1982;30:439–448. [PubMed]
  • Sturtz L. A., Diekert K., Jensen L. T., Lill R., Culotta V. C. A fraction of yeast Cu, Zn-superoxide dismutase and its metallochaperone, CCS, localize to the intermembrane space of mitochondria. A physiological role for SOD1 in guarding against mitochondrial oxidative damage. J. Biol. Chem. 2001;276:38084–38089. [PubMed]
  • Takemori Y., Sakaguchi A., Matsuda S., Mizukami Y., Sakurai H. Stress-induced transcription of the endoplasmic reticulum oxidoreductin gene ERO1 in the yeast Saccharomyces cerevisiae. Mol. Genet. Genomics. 2006;275:89–96. [PubMed]
  • Temple M. D., Perrone G. G., Dawes I. W. Complex cellular responses to reactive oxygen species. Trends Cell Biol. 2005;15:319–326. [PubMed]
  • Thorpe G. W., Fong C. S., Alic N., Higgins V. J., Dawes I. W. Cells have distinct mechanisms to maintain protection against different reactive oxygen species: oxidative-stress-response genes. Proc. Natl. Acad. Sci. USA. 2004;101:6564–6569. [PMC free article] [PubMed]
  • Tong A. H., et al. Systematic genetic analysis with ordered arrays of yeast deletion mutants. Science. 2001;294:2364–2368. [PubMed]
  • Travers K. J., Patil C. K., Wodicka L., Lockhart D. J., Weissman J. S., Walter P. Functional and genomic analyses reveal an essential coordination between the unfolded protein response and ER-associated degradation. Cell. 2000;101:249–258. [PubMed]
  • Trotter E. W., Grant C. M. Thioredoxins are required for protection against a reductive stress in the yeast Saccharomyces cerevisiae. Mol. Microbiol. 2002;46:869–878. [PubMed]
  • Tu B. P., Weissman J. S. Oxidative protein folding in eukaryotes: mechanisms and consequences. J. Cell Biol. 2004;164:341–346. [PMC free article] [PubMed]
  • Valls L. A., Hunter C. P., Rothman J. H., Stevens T. H. Protein sorting in yeast: the localization determinant of yeast vacuolar carboxypeptidase Y resides in the propeptide. Cell. 1987;48:887–897. [PubMed]
  • Vandeputte C., Guizon I., Genestie-Denis I., Vannier B., Lorenzon G. A microtiter plate assay for total glutathione and glutathione disulfide contents in cultured/isolated cells: performance study of a new miniaturized protocol. Cell Biol. Toxicol. 1994;10:415–421. [PubMed]
  • Veal E. A., Ross S. J., Malakasi P., Peacock E., Morgan B. A. Ybp1 is required for the hydrogen peroxide-induced oxidation of the Yap1 transcription factor. J. Biol. Chem. 2003;278:30896–30904. [PubMed]
  • Wheals A. E. Size control models of Saccharomyces cerevisiae cell proliferation. Mol. Cell. Biol. 1982;2:361–368. [PMC free article] [PubMed]
  • White C. W., Nguyen D. H., Suzuki K., Taniguchi N., Rusakow L. S., Avraham K. B., Groner Y. Expression of manganese superoxide dismutase is not altered in transgenic mice with elevated level of copper-zinc superoxide dismutase. Free Radic. Biol. Med. 1993;15:629–636. [PubMed]
  • Winzeler E. A., et al. Functional characterization of the S. cerevisiae genome by gene deletion and parallel analysis. Science. 1999;285:901–906. [PubMed]
  • Zhao H., Kalivendi S., Zhang H., Joseph J., Nithipatikom K., Vasquez-Vivar J., Kalyanaraman B. Superoxide reacts with hydroethidine but forms a fluorescent product that is distinctly different from ethidium: potential implications in intracellular fluorescence detection of superoxide. Free Radic. Biol. Med. 2003;34:1359–1368. [PubMed]

Articles from Molecular Biology of the Cell are provided here courtesy of American Society for Cell Biology
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...