OUP user menu

The adaptive response of anaerobically grown Saccharomyces cerevisiae to hydrogen peroxide is mediated by the Yap1 and Skn7 transcription factors

Anthony G. Beckhouse, Chris M. Grant, Peter J. Rogers, Ian W. Dawes, Vincent J. Higgins
DOI: http://dx.doi.org/10.1111/j.1567-1364.2008.00439.x 1214-1222 First published online: 1 December 2008


The molecular mechanisms involved in the ability of cells to adapt and respond to differing oxygen tensions are of great interest to the pharmaceutical, medical and fermentation industries. The transcriptional profiles reported in previous studies of cells grown under anaerobic, aerobic and dynamic growth conditions have shown significantly altered responses including induction of genes regulated by the oxidative stress transcription factor Yap1p when oxygen was present. The present study investigated the phenotypic changes that occur in cells when shifted from anaerobic to aerobic growth conditions and it was found through mutant analyses that the elevated activity of Yap1p during the shift was mediated by the phospholipid hydroperoxide-sensing protein encoded by GPX3. Cell viability and growth rate were unaffected even though anaerobically grown cells were found to be hypersensitive to low doses of the oxidative stress-inducing compound hydrogen peroxide (H2O2). Adaptation to H2O2 treatment was demonstrated to occur when anaerobically grown wild-type cells were aerated for a short time that was reliant on the Yap1p and Skn7p transcription factors.

  • Saccharomyces cerevisiae
  • oxidative stress
  • anaerobic
  • aerobic


Aerobic organisms primarily respire to generate energy through the breakdown of organic molecules such as glucose, with oxygen as a vital cofactor. Most oxygen molecules are reduced to water during the respiratory process but c. 5% can undergo incomplete reduction to form reactive oxygen species (ROS). ROS are well-known cellular toxicants that induce DNA damage, protein oxidation and lipid peroxidation. To reduce the toxicity of such species, organisms have evolved antioxidant defence systems to counter the damaging effects. If the balance between antioxidant and oxidant is perturbed, the organism is known to have undergone oxidative stress (Jamieson et al., 1998; Dawes et al., 2004). Oxidative stress is of interest in many fields such as the medical and pharmaceutical industries where it has been implicated in the pathology of diseases such as cancer, cardiovascular disease, arthritis and the ageing process (Halliwell et al., 1997; Halliwell & Gutteridge et al., 1999; Finkel & Holbrook et al., 2000).

Under anaerobic conditions, organisms are spared the toxic effects of oxidative stress; however, it has been noted in the yeast Saccharomyces cerevisiae that antioxidant proteins are reduced in level rather than absent and this is thought to be a protective mechanism against possible introduction to an aerobic environment and ROS (Ohmori et al., 1999). In most environmental niches, organisms experience a dynamic flux in oxygen availability and must be able to adapt rapidly for survival. Saccharomyces cerevisiae is an ideal model organism for studying the effect of altering oxygen tension due to its ability to survive in aerobic and anaerobic environments along with its accessible genome sequence (Goffeau et al., 1996) and the availability of single-gene deletion mutants in nonessential genes (Winzeler et al., 1999). Many transcriptional studies have been performed on cells in steady-state anaerobiosis or aerobiosis (ter Linde et al., 1999; Piper et al., 2002), as well as shifts between the two oxygen tensions (Kwast et al., 1998, 2002; Lai et al., 2005, 2006), which identified that differential regulation of genes involved the oxidative stress response, ergosterol and haem biosynthesis, cell-wall maintenance and amino acid biosynthesis. A recent study has also identified up- and downregulated proteins within the proteome under anaerobic and aerobic conditions (de Groot et al., 2007). Such studies led to the observation that transcription factors such as Yap1, Mot3, Upc2 and Sut1 may coordinately regulate transcription of genes during steady state or dynamic oxygen tensions. Altering oxygen tensions in live cells is of interest in the medical field where procedures such as open heart surgery, organ transplant, myocardial infarct and stroke all deprive tissues of oxygen for periods and the resultant reintroduction of oxygenated blood introduces ROS, which can further damage tissues (Fellstrom et al., 1998; Kaminski et al., 2002; Crack & Taylor et al., 2005). Similar damaging effects occur in the brewing industry where oxygen fluctuations can affect fermentation kinetics, flavour profiles, shelf-life and yeast viability due to ROS (O'Connor-Cox & Ingledew et al., 1990; Higgins et al., 2003; Rosenfeld et al., 2003).

The upregulation of oxidative stress response genes following a shift from anaerobic to aerobic conditions indicates that cells are exposed to damaging oxidants, which may affect cell viability or growth kinetics. The shift to aeration and subsequent exposure to ROS may also allow cells to adapt to exogenous oxidative stress-causing agents such as hydrogen peroxide (H2O2). Adaptation occurs when cells are exposed to sublethal doses of an oxidant that induces a transient protective response that confers resistance to subsequent doses that would normally be lethal to the cells (Collinson & Dawes et al., 1992; Jamieson et al., 1992; Flattery-O'Brien et al., 1993).

The aims of the present study were to determine the sensitivity of anaerobically grown S. cerevisiae to H2O2, investigate the viability and growth kinetics of cells shifted from anaerobic to aerobic conditions and elucidate any adaptive response and role of transcription factors in the adaptive response during the shift.

Materials and methods

Strains and plasmids

Saccharomyces cerevisiae strains were obtained from the homozygous diploid single deletion collection (Euroscarf http://www.uni-frankfurt.de/fb15/mikro/euroscarf). Strains were derivatives of wild-type BY4743 (MATa/MATαhis31/his3Δ1 leu2Δ0/leu2Δ0 met15/met15Δ0 lys2Δ0/LYS2ura3Δ0/ura3Δ0) with the following mutations: rox1, upc2, mot3, sut1, hap4, yap1, skn7, msn2, msn4, gcn4, skn7yap1 and msn2,4. Plasmid pyDJ73 containing GSH1lacZ was provided by Derek Jamieson, pSC99 and pCEP12 containing TRX2lacZ YRE-CYC1lacZ, respectively, were provided by W. Scott Moye-Rowley and TRX1 lacZ containing TRX1lacZ was supplied by Chris M. Grant.

Growth conditions

Cells were grown in synthetic complete (SC) medium supplemented with components (ergosterol and unsaturated fatty acids) essential for anaerobic growth (Andreasen & Stier et al., 1953, 1954). Anaerobic cultures were grown using methods modified from Skoneczny & Rytka (2000), Panozzo (2002) and Passoth (2003). Essentially, SC medium (Alic et al., 2001) was supplemented with the redox-indicator dye resazurin (2 mg mL−1; Sigma-Aldrich, St. Louis, CA) and the pH was adjusted to 4.5. Media were aliquoted into 100-mL penicillin bottles before degassing with high-purity nitrogen gas (BOC gases, Sydney, NSW, Australia) and sealed with butyl-rubber stoppers and aluminium cap seals (Wheaton Science, Millville, NJ). Ergosterol (Sigma-Aldrich) and Tween 80 (Sigma-Aldrich) were supplemented from a fresh stock (ergosterol 4 mg mL−1; Tween 80, 40% v/v in ethanol) at a volume of 0.5 mL per 100 mL of medium. The reducing agent sodium dithionite (Sigma-Aldrich) was added to the media before inoculation at a concentration of 100 mg L−1 to reduce trace amounts of oxygen. Aerobic cultures were grown in media identical to the anaerobic cultures including the supplements (except nitrogen gas). All incubations of liquid cultures were performed at 30 °C with shaking at 250 r.p.m. Anaerobic manipulations and anaerobic incubation of solid media were performed in an anaerobic hood (Modular Atmosphere Controlled System) (DW Scientific, Shipley, Yorkshire, UK) flushed with a gas mixture of N2 : CO2 : H2 at a ratio of 75 : 20 : 5.

Aeration of anaerobically grown cells was performed by adding the culture to a sterile prewarmed (30 °C) flask where the culture volume was at least 1/5 of the flask volume. Aeration was performed in a 30 °C incubator with shaking at 250 r.p.m.

YEPD medium (2% w/v glucose, 2% w/v Bacto peptone, 1% yeast extract) (Sigma-Aldrich) was used for dilution of cells and YEPD agar (YEPD plus 2% w/v type 3 agar) was used for cultivating cells during viability determination.

Viability measurement of cultures treated with H2O2

Anaerobic and aerobic cultures were grown to an OD600 nm of 1 and subsequently treated with 0, 0.1, 0.25, 0.5, 0.75 and 1.5 mM H2O2 for 10 min. An aliquot of 1 mL was removed and serially diluted appropriately in liquid YEPD. Serial dilutions (200 μL) were then spread onto YEPD agar in triplicate and incubated at 30 °C for 3 days to obtain viable counts.

To test the viability and growth kinetics of cells when shifted from anaerobic to aerobic growth conditions, exponential-phase cells (OD600 nm=1) were sampled in an anaerobic hood and serially diluted. For viability tests, appropriate dilutions (200 μL) were spread on YEPD agar in duplicate sets of triplicates. One triplicate set was removed to an aerobic incubator at 30 °C; the other set remained in the anaerobic hood for incubation at 30 °C for 3 days. Growth kinetics of anaerobic and aerated samples were monitored by measurement at an OD600 nm.

Adaptation to H2O2 treatment with a period of aeration

Cultures were grown under anaerobic conditions to the exponential phase (OD600 nm=1) and an aliquot (10 mL) was removed for aeration in a prewarmed 50-mL flask. The anaerobic culture was then treated with 1.5 mM H2O2 (Ajax Chemicals, Sydney, NSW, Australia) for a period of 60 min. Samples were removed, serially diluted in YEPD and spread on YEPD agar in triplicate every 15 min. Following incubation for 30 min, the aerated cultures were treated with 1.5 mM H2O2 for 60 min, with samples taken every 15 min. Viability was determined as a percentage of the mean number of cells before H2O2 treatment.

β-Galactosidase assays

Specific activity of the β-galactosidase in extracts of cells containing the Yap1p reporter construct in pCEP12 was assayed essentially as described previously (Rose & Botstein et al., 1983). Cell protein in the extract was measured with bovine serum albumin (Sigma-Aldrich) was used as a standard and protein assay dye reagent (Bio-Rad, Hercules, CA) was used to bind proteins as per the manufacturer's protocol (Bradford et al., 1976). Specific activity was expressed as nanomoles of o-nitrophenyl-β-d-galactopyranoside (ONPG) (Sigma-Aldrich) hydrolysed per minute per microgram of total protein. Each data point was obtained from triplicate cultures and results were expressed as the mean and SD. Statistical analyses were performed using t-tests to determine significance at a 95% confidence level.

Cell harvesting, RNA extraction and Northern blotting

Cultures were grown under anaerobic conditions to the exponential growth phase (OD600 nm=1) and then transferred to a prewarmed 250-mL flask and incubated with shaking at 30 °C for indicated periods of time. Anaerobic samples were harvested immediately in a prechilled tube containing ice by centrifugation at 4 °C. Cells were snap-frozen in liquid nitrogen and stored at −80 °C until RNA extraction was performed. The aerated cells were harvested in the same manner.

Total RNA was obtained by breaking cells with acid-washed glass beads in the Trizol reagent (Invitrogen, Carlsbad, CA) using a mini-bead beater and extracted according to the manufacturer's instructions. The quality of RNA was determined using agarose gel electrophoresis and ethidium bromide staining as well as by spectrophotometric analysis (A260 nm/A280 nm) using a Nanodrop spectrophotometer (Nanodrop Technologies, Wilmington, DE). Northern blotting was performed as described in Sambrook (1989). Probes were generated by random primer labelling of PCR-amplified regions of the TRR1, ARG4 and ACT1 ORFs with α-32P dCTP and α-32P dATP (Perkin Elmer Life Sciences, Melbourne, Vic., Australia). Prehybridization and hybridization were performed at 65 °C in a RapidHyb buffer (Amersham Biosciences, UK). Radioactivity of membrane-bound probes was imaged and quantified using an FLA-5100 phosphorimager and Image Gauge V4 (Fujifilm, Japan).


Activation of Yap1p during the shift from anaerobic to aerobic conditions

Lai (2006) reported that a rapid transcriptional response occurred when anoxic yeast cultures were shifted into an aerobic environment. Of the transcripts identified to alter within 5 min of the shift, many are known to belong to the oxidative stress regulon, which is regulated by the Yap1p transcription factor; therefore, we measured the activation of the Yap1p transcriptional activator during aeration of anaerobically grown cells. A wild-type strain harbouring the YRElacZ plasmid (Wu & Moye-Rowley et al., 1994) containing multiple Yap1p response elements (YRE's) located upstream of a minimal CYC1lacZ fusion gene reporter was used. As a control, the plasmid was also transformed into a strain (Δgpx3) that is unable to sense the presence of H2O2 and therefore unable to mount a Yap1p-mediated response to H2O2 (Delaunay et al., 2002). The strains were grown anaerobically and then shifted to aerobic conditions. Samples were taken at the intervals indicated and assayed for β-galactosidase activity. It can be seen in Fig. 1a that there was a significant induction of the reporter construct within 30 min of the shift, which remained at a high level for 120 min. The activity of the Yap1 reporter construct in the Δgpx3 strain remained at relatively low levels during the shift compared with the wild type; however, there was a statistically significant increase (P<0.05) in the level of the reporter construct during the shift compared with the levels seen in anaerobically grown cultures (t=0 min). Control cultures of each strain harbouring the reporter construct were grown under aerobic conditions and both strains showed a significant increase in activation during aerobic growth compared with anaerobic growth but not at the levels observed during the shift. We also tested whether known marker genes in the oxidative stress regulon were also upregulated in response to the aeration. The lacZ reporter constructs for TRX2 and GSH1 were transformed into wild-type yeast and assayed for β-galactosidase activity over an aeration time course of 2 h. A control reporter construct was also included (TRX1lacZ) in the experiment because it is known to be unresponsive during oxidative challenge (Lee et al., 1999; Garrido & Grant et al., 2002). The results in Fig. 1b clearly show that both TRX2 and GSH1 reporter levels were significantly increased (P<0.05) after 30 min of aeration compared with anaerobic cells, indicating that the response to aeration includes an oxidative stress response.

Figure 1

β-Galactosidase assays of Yap1 activity and Yap1-regulated genes during a shift from anaerobic to aerobic conditions. Anaerobic cultures of yeast containing reporter constructs were grown in triplicate to exponential phase and shifted to aerobic conditions. Samples were taken at the indicated times. Aerobic cultures were grown to the exponential phase and harvested as a control. (a) WT yeast strain BY4743 and mutant gpx3 containing the Yap1p activity reporter construct YRE-CYC1-lacZ. (b) WT yeast strain BY4743 containing TRX1lacZ, TRX2lacZ or GSH1lacZ reporter constructs. Average measurements are indicated with error bars (±SD) included. *Indicates the first time-point where a significant (P<0.05) difference from time zero occurred or whether the aerobic level of β-galactosidase activity was significantly (P<0.05) greater than anaerobic levels.

Viability and growth kinetics of shifted yeast cells

The results presented thus far indicate that cells transcribe oxidative stress response genes rapidly (but not exclusively) and antioxidant genes are upregulated when cells are shifted from anaerobic growth to aerobic conditions. Therefore, it was of importance to determine the viability and growth kinetics of the cells in which oxidative stress was signalled following the shift in oxygen tension. To test whether aeration of anaerobically grown samples would affect the viability or kinetics of cell growth, CFU plate tests were performed. Anaerobic cultures were grown to an OD600 nm=1.0 and placed a Modular Atmosphere Controlled System anaerobic chamber at 30 °C. Aliquots were removed from the flasks, serially diluted and plated onto solid YEPD media and incubated in either an anaerobic chamber or removed and incubated aerobically at 30 °C. There was no significant difference in viability as indicated in Fig. 2a, but due to the length in time required for CFUs to arise, kinetics of the cultures were also measured after the aliquots were removed to determine whether any perturbation in growth was observed. A further aliquot was removed from anaerobic cultures into prewarmed flasks and compared with the anaerobic culture growth kinetics. ODs were measured for 9 h after shifting and the results are illustrated in Fig. 2b. No differences in kinetics were observed between shifted and anaerobic cultures until 6 h postshift where aerobic cells changed the carbon source and began to generate energy through respiration. This is a slightly unexpected result as the transcriptional profiles reported by Lai (2006) indicated that cells underwent cellular reprogramming, which may have resulted in some growth retardation but this was clearly not the case. Because no effect was observed on the viability or the growth kinetics of shifted yeast cultures but the transcriptional response and level of antioxidant reporter protein indicated that cells may experience oxidative challenge, it was then of interest to determine whether addition of an oxidant at low concentrations to anaerobic cells would affect cell viability.

Figure 2

Viability and growth kinetics of yeast shifted from anaerobic to aerobic conditions. Anaerobic, exponential phase yeast cultures were shifted to aerobic conditions. (a) The number of viable cells measured using a CFU plate test of triplicate samples plated onto solid YEPD and incubated in aerobic or anaerobic conditions. (b) The growth of triplicate cultures measured using OD600 nm after the exponential phase had been reached in anaerobic conditions and samples were split into aerobic and anaerobic environments. Error bars are included (±SD).

Anaerobically grown cells are hypersensitive to H2O2

Previous studies have shown that anaerobic yeast cells do express basal levels of antioxidant enzymes that are hypothesized to prepare cells against sudden exposure to oxygen and its radicals (Hortner et al., 1982; Ohmori et al., 1999). The sensitivity of anaerobically grown cells to oxidative challenge is not well understood. Susceptibility to H2O2 was used to determine whether cells grown anaerobically were more sensitive to the oxidant than cells grown under aerobic conditions. The exponential-phase cells cultured under anaerobic or aerobic conditions were treated with different concentrations of H2O2 for 10 min and their viabilities were determined. Cells that had been grown under anaerobic conditions showed a marked reduction in viability (85%) at a relatively low dose of H2O2 (0.1 mM) compared with untreated cells. As the concentration of H2O2 increased to a maximum of 1.5 mM, the viability of the anaerobically grown cells decreased dramatically (Fig. 3). Aerobically grown cultures did not show any decrease in viability across the H2O2 concentrations used. Such a large difference in viabilities at very low concentrations indicates that aeration of anaerobic cells introduces a very small amount of oxidative stress that elucidates a transcriptional and physiological response but is sufficiently low that it does not affect the growth and viability of cells. Therefore, it may be possible that aeration of anaerobically cultured cells may be sufficient to induce an adaptive response to an exogenous oxidative challenge.

Figure 3

Sensitivity of anaerobic and aerobic exponential phase yeast cultures to H2O2. Wild-type yeast strain (BY4743) was inoculated and grown to exponential phase in aerobic or anaerobic conditions then treated with indicated concentrations of H2O2 for 10 min. CFU plate tests were used to quantify viable cells. Survival is expressed as the colony-forming ability in each incubation condition relative to that before treatment. Data are the means of three independent experiments with error bars (±SD) included.

Anaerobic cultures of yeast adapt to H2O2 toxicity when aerated

Yeast elicits an adaptive response that confers resistance to normally toxic levels of H2O2 when pretreated with low doses of the oxidant (Collinson & Dawes et al., 1992). Cultures of aerobic and anaerobic yeast cells were treated with 1.5 mM H2O2 and samples were removed and tested for viability at 15-min intervals for a total of 60 min. Additional cultures of anaerobically grown cells were aerated for a period of 30 min before addition of 1.5 mM H2O2 and samples were then removed and measured for viability in the same manner as the anaerobic and aerobic cultures. As shown in Fig. 3, 1.5 mM H2O2 treatment for 10 min was lethal to anaerobically grown cells but did not affect aerobically cultured cells. Similarly, Fig. 4 shows that anaerobic cells were hypersensitive to treatment with an oxidant compared with aerobic cells when treated for 60 min. Cells that were aerated before oxidant addition behaved differently. The period of aeration generated a resistance to the oxidative challenge in a manner similar to an adaptive response. Transcription factors play a major role in eliciting an adaptive response to stress; therefore, we tested many transcription factor mutants identified through transcriptional and proteomic studies.

Figure 4

Survival of H2O2 treated anaerobic, aerobic and anaerobic shifted yeast cultures. Exponential phase cultures of yeast strain BY4743 were grown anaerobically, aerobically and anaerobically with a 30-min aeration were treated with 1.5 mM H2O2. Survival was measured using CFU plate tests and is expressed as the colony-forming ability relative to that before treatment. Data are the means of three independent experiments with error bars (±SD) included but not visible due to low error.

Transcription factor involvement in anaerobic to aerobic adaptive response

Transcriptional analyses of the anaerobic to aerobic shift by Lai (2006), identified clustered sets of genes with similar expression profiles. These sets allowed in silico identification of consensus promoter-binding motifs that may play a role in regulation of the genes within the cluster. A set of transcription factors identified in that study and a proteomic study of de Groot (2007) included Rox1, Upc2, Mot3, Sut1, Hap4, Yap1, Skn7, Msn2, Msn4 and Gcn4, and strains with a single deletion of each of these transcription factors were obtained from the yeast single nonessential deletion collection (Winzeler et al., 1999). Strains harbouring double mutations (skn7 yap1 and msn2 msn4) were also obtained as the two transcription factors often function together in stress responses. Mutants were then used to test the ability of aerated cells to adapt to an oxidant challenge after a period of aeration. When compared with the wild-type response in Fig. 4, all mutant strains behaved in a manner similar to the wild type, except for strains that harboured knock-outs of yap1, skn7 and the double skn7 yap1 (Fig. 5a–c, respectively). The data clearly show that the Δyap1 strain was defective in its ability to mount an adaptive response to H2O2 after a period of aeration. A similar response was seen in the Δskn7 strain shown in Fig. 5b; however, the Δskn7 mutant did demonstrate a level of adaptive response that was absent in the Δyap1 strain, but was still impaired relative to the wild type. The Δskn7Δyap1 double mutant had a similar phenotype compared with the single Δyap1 strain when grown anaerobically; however, it was more sensitive to the oxidant when grown under aerobic conditions (Fig. 5c). Anaerobic cultures of the three mutant strains (Fig. 5) were only slightly more sensitive to H2O2 treatment than anaerobic wild-type cells (Fig. 4), indicating that Yap1p and Skn7p do not have a major role when growing anaerobically. Conversely, the double mutant was more sensitive to oxidant challenge than the wild type and each single mutant under aerobic conditions, demonstrating the importance of Yap1p and Skn7p for aerobic growth. The strains harbouring mutations were also tested for their ability to survive exposure to air relative to continued anaerobic growth, and Fig. 5d indicates that the shift to an aerobic environment did not affect the ability of the cells to survive. Interestingly, the Δgcn4 mutant was unable to grow under the anaerobic conditions used. To determine whether Gcn4p was activated under anaerobic conditions, we measured the abundance of Gcn4p target gene ARG4 using Northern analysis. Figure 6 clearly shows that ARG4 is rapidly downregulated upon aeration of anaerobically grown cells whereas an oxidative stress gene representative TRR1 was rapidly upregulated and house-keeping gene ACT1 expression remained constant in response to aeration.

Figure 5

Survival of H2O2 treated anaerobic, aerobic and shifted yeast strains. Exponential phase cultures of yeast were grown in the above conditions and treated with 1.5 mM H2O2. (a) Δyap1, (b) Δskn7, (c) Δskn7Δyap1. Survival is expressed as the colony-forming ability relative to that before treatment. (d) Exponential phase cultures of the WT BY4743 strain and three mutants were grown to exponential phase and plated onto solid YEPD under anaerobic conditions in duplicate. Cultures were incubated either aerobically or anaerobically. Data are the means of three independent experiments with error bars (±SD) included.

Figure 6

Northern blot analysis of ACT1, ARG4 and TRR1 during a shift from an aerobic to anaerobic environment. Wild-type yeast strain BY4743 was grown in anaerobic conditions to exponential phase then aerated for a period of 60 min with samples taken at the intervals indicated. Total RNA was extracted and run on a denaturing agarose gel then transferred to nylon filters. Radioactively labelled probes for ACT1, ARG4 and TRR1 were hybridized to the filters. Scanning and quantification were performed using a phosphorimager. Top panel is the relative level of expression of each gene compared with that in an anaerobically growing sample and is a representative example of a triplicate experiment. Lower panel is the scanned image produced using the phosphorimager of the labelled filters.


The ability of organisms to adapt rapidly to environmental change is extremely important for survival. The genetic and molecular mechanisms involved in this phenomenon are described here for the response of S. cerevisiae to aeration after a period of anaerobic growth. The above data show that this yeast has the capacity to adapt very rapidly to the shift from anaerobiosis to aerobiosis without adjusting its growth rate or losing viability. This indicates the presence of active homeostatic systems, including one based on the Yap1p transcription factor associated with ensuring that ROS do not present a problem to the cell.

The thorough microarray experiment performed by Lai (2006) identified a rapid transcriptional response for yeast cells shifted from anaerobic to aerobic conditions with in silico clustering of gene families that may show coregulation through transcription factors. A similar study by Higgins (2003) performed on an industrial-scale shift of yeast cells from anaerobic to aerobic conditions also identified genes involved in the oxidative stress response and ergosterol biosynthesis, but they included knock-out studies to further suggest that ergosterol content and oxidative stress can affect the viability of cells. The results presented here show that aeration of anaerobically growing cells produced no effect on cell viability or growth rate when a single stress was introduced, indicating that the transcriptional response to oxygenation may play a role in adaptation of cells to further oxidative challenge. Reporter gene assays were performed to measure the activity of the oxidative stress response transcription factor Yap1 and also expression of genes under its control during the shift. There was rapid upregulation of Yap1 activity and Yap1 target genes during the shift to an aerated environment and the values measured were significantly higher than those found in steady-state aerobic cultures, indicating that cells react to overwhelm oxidative challenges transiently before steady-state homeostasis. It has been shown previously that genes involved in the response to oxidative stress (thioredoxin, thioredoxin reductase and glutathione reductase) play an important role in maintaining the redox potential within the cell (Drakulic et al., 2005). The upregulation of genes involved in the oxidative stress response following a shift from anaerobiosis to aerobic conditions indicated that an oxidative stress was encountered and that the cells required an alteration in redox homeostasis upon introduction into an aerobic environment. The rapidity of this response may play a role in sustaining growth rate and survival that was observed when shifted to aerobic environments.

A previous report studying the effect of fatty acids in the membrane of anaerobic and aerobic cells indicated that anaerobically grown cells were more sensitive to H2O2 compared with aerobically grown cells but this required very high levels of the oxidant to produce the phenotype (Steels et al., 1994). This was an interesting observation as a later report detected oxidative stress response enzymes at much lower levels than aerobically grown cells, which was hypothesized to be more of a protective mechanism in the event of exposure to very low levels of oxygen and its radicals (Ohmori et al., 1999). Therefore, we tested the tolerance of anaerobically cultured cells to low and high doses of H2O2 and found that the cells were hypersensitive to very low levels of oxidant. Because, in these experiments, anaerobically grown cells were able to induce an oxidative stress response without affecting the viability or the kinetics when aerated, we sought to determine whether aeration of anaerobic cells would induce an adaptive response to an additional oxidative challenge. Anaerobically grown cells have shown adaptability to other types of redox stresses previously (Krantz et al., 2004); therefore, it seemed likely that adaptation to an oxidant through aeration may occur. The adaptive response to oxidative stress in yeast is well studied in aerobic environments and is known to involve specific transcription factors Yap1 and Skn7 (Lee et al., 1999; Ng et al., 2008). Yap1 activation is dependent on the sensing protein Gpx3 (Delaunay et al., 2002) but is also known to be activated by Gpx3-independent mechanisms (Kuge et al., 2001; Azevedo et al., 2003). In silico identification of transcription factors through microarray analyses identified many transcription factors that may play a role in adapting to the aerated environment (Higgins et al., 2003; Lai et al., 2006); therefore, transcription factor mutants were tested for their ability to adapt to an oxidant challenge via aeration. This study identified that Yap1 has a major role in the adaptive process through aeration in a Gpx3-dependent manner. Skn7 was shown to have a minor role in the process but when deleted together with Yap1 did not produce additive loss in adaptability. This clearly implicates Yap1 as the major transcription factor involved in the ability to adapt to an oxidant after the onset of aeration. Other transcription factor mutations that were tested did not affect the adaptive response; however, it was noted that the gcn4 strain was unable to grow under the conditions tested. Gcn4 is a transcriptional activator of amino acid metabolism genes (Hinnebusch et al., 1984) and many microarray experiments have identified its target genes to be upregulated under anaerobic conditions (ter Linde et al., 1999; Piper et al., 2002; Lai et al., 2006). Interestingly, in an attempt to identify essential genes for anaerobic growth, Gcn4 was omitted (Snoek & Steensma et al., 2006). In contrast, we report that the gcn4 strain did not grow under the anaerobic conditions used, which may indicate an important role for amino acid biosynthesis for growth under anaerobic conditions.


We wish to acknowledge Nazif Alic, Cristy Gelling, Ruby Lin, Bronwyn Robertson and Nick Matigian for their assistance in RNA and microarray analysis, W. Scott Moye-Rowley and Derek Jamieson for providing plasmids and Frank Estruch for the msn2,4 strain. This work was funded by an Australian Research Council Linkage grant and Carlton and United Beverages. A.G.B. was funded by an Australian Postgraduate Award and C.M.G. was funded by the Wellcome Trust.


  • Editor: Isak Pretorius


View Abstract