The oxidation state of the cytoplasmic glutathione redox system does not correlate with replicative lifespan in yeast

Abstract

What is cause and what is consequence of aging and whether reactive oxygen species (ROS) contribute to this phenomenon is debated since more than 50 years. Notwithstanding, little is known about the cellular buffer and redox systems in aging Saccharomyces cerevisiae, which is a model for aging stem cells. Using genetically encoded fluorescent sensors, we measured pH, H₂O₂ levels and the glutathione redox potential compartment-specific in the cytosol of living, replicatively aging yeast cells, growing under fermenting and respiratory conditions until the end of their lifespan. We found that the pH decreases under both conditions at later stages of the replicative lifespan. H₂O₂ levels increase in fermenting cells in the post-replicative stage, but increase continuously with age in respiring cells. The glutathione redox couple becomes also more oxidizing in respiring cells but surprisingly more reducing under fermenting conditions. In strains deleted for the gene encoding glutathione reductase Glr1, such a reduction of the glutathione redox couple with age is not observed. We demonstrate that in vivo Glr1 is activated at lower pH explaining the reduced glutathione potential. The deletion of glr1 dramatically increases the glutathione redox potential especially under respiratory conditions but does not reduce lifespan. Our data demonstrate that pH and the glutathione redox couple is linked through Glr1 and that yeast cells can cope with a high glutathione redox potential without impact on longevity. Our data further suggest that a breakdown of cellular energy metabolism marks the end of replicative lifespan in yeast.

Figure 1

The cytosolic pH decreases in aging yeast cells. (a) Images of logarithmic WΣ cells (left), which are labeled and then initially sorted to result mostly cells without budscars; WΣ cells after the 24 h of growth; and cells at a final stage of aging (see also Supplementary Figures S1 and S2). Budscars colored with solophenylflavin. Scale bars, 2 μm. (b) Budscar distribution of aging WΣ yeast cells at successive FACS (fluorescence-activated cell sorting) enrichment steps at the indicated time points. For every time point, the budscars of 56 to 395 (average 130) single cells were counted. (c) Representative aging curve of WΣ grown under fermenting conditions. Budscars were counted and averaged for every time point. (d) Cellular distribution of pHluorin, roGFP2-Orp1 and roGFP2-Grx1. Wide-field fluorescence image in maximal projection. Scale bar, 2 μm. Z-stacks of confocal images are provided as Supplementary Figures S3, S4, S6–S9. (e) Average cytosolic pH of WΣ cells logarithmically growing on fermentable (2% glucose, SC) and non-fermentable (3% glycerol, 2% ethanol, SG) as indicated. (wt SC, n=27; wt SG, n=31; Error bars represent the standard deviation). (f and g) Development of the population-averaged pH of aging cells grown under fermentative (f, SC-medium with 2% glucose) or respiratory (g, SG-medium with 3% glycerol and 2% ethanol) conditions. The young cells (blue) are the rejuvenated offspring of the old population (red) within the same culture and used as internal control. Standard error of the mean is smaller than the symbols. Numbers below the red symbols indicate the average number of budscars at each time point. Asterisks above the symbols indicate the probability for the statistical significance of the difference between old cells and their rejuvenated offspring within the same culture using Dunn’s multiple comparison test (****P<0.0001). (h) Difference of the population-averaged pH of aging cells and their rejuvenated offspring grown under fermentative (green, SC: 2% glucose) or respiratory conditions (black, SG: 3% glycerol and 2% ethanol) plotted versus the average number of budscars of the old cells. (i) Comparison of budscars of aging fermenting (green) and respiring (black) cells. To clarify the trend, the data were fitted by a Gompertz function (plateau of the fit 23.7±0.08 (SC) and 27.1±0.12 (SG)). The data of several independent aging experiments are shown. The error bars indicate the standard error of the mean (n=56–395). The dotted lines indicate the maximum of the Gompertz function. Inset, average of the maximal budscar average measured in different experiments of WΣ cells grown in SC (green, n=8) or SG (black, n=2) medium. The error bars indicate standard deviation (P=0.0008; Student’s t-test). (j) Replicative competence of aged fermenting (green) and respiring (black) cells at different replicative age (colony-forming units were calculated from the sort of 384 single cells from each time point on YPD-agar plates). The data of 10 (fermentation) and two (respiration) independent aging experiments are shown. (k) Replicative lifespan analysis using the dissecting assay. Two independent assays are shown for cells growing on 2% glucose (Glc)- or 3% glycerol/2% ethanol-containing plates (Gly/EtOH). The difference is highly significant (log-rank Mantel–Cox test P<0.0001; Gehan–Breslow–Wilcoxon test P<0.0001).

Figure 2

Change of H₂O₂ levels and glutathione-reducing potential in aging yeast cells. (a and b) Development of the population-averaged H₂O₂ production (measured as % probe oxidation of the roGFP2-Orp1 reporter) in aging fermenting (a) and respiring (b) cells. Red, old cells; blue, rejuvenated offspring in the same culture. Standard error of the mean is smaller than the symbols. Asterisks below or above the symbols indicate the probability for the statistical significance of the difference between the old cells and their rejuvenated offspring within the same culture using Dunn’s multiple comparison test (****P<0.0001). The inset shows the difference Δ %OxD between the old and young cells from two independent experiments plotted against the age in average number of budscars within the population. (c and d) Redox potential of the GSSG/2GSH redox couple in terms of roGFP2-Grx1 probe oxidation in aging fermenting (c) and respiring (d) cells. Red, old cells; blue, rejuvenated offspring in the same culture. Standard error of the mean is smaller than the symbols. Asterisks below the symbols indicate the probability for the statistical significance of the difference between the old cells and their rejuvenated offspring within the same culture using Dunn’s multiple comparison test (****P<0.0001). The inset shows the difference Δ %OxD between the old and young cells from two independent experiments plotted against the age in average number of budscars within the population.

Figure 3

Δglr1 cells have a highly oxidized cytosol but no reduced lifespan under fermentative growth conditions. (a) Glutathione potential in aging WΣ-Δglr1 cells growing under fermentative conditions; red, old cells; blue, rejuvenated young cells in the same culture. Standard error of the mean is smaller than the symbols. Asterisks above the symbols indicate the probability for the statistical significance of the difference between old cells and their rejuvenated offspring within the same culture using Dunn’s multiple comparison test (****P<0.0001). The inset shows the steady-state probe oxidation for wild-type (n=44) and Δglr1 (n=7) cells. Error bars are the standard error of the mean of n independent cultures. (b) Glutathione potential in aging WΣ-Δglr1 cells growing under respiratory conditions. (Error bars and asterisks as in a). (c) Difference Δ %OxD of old and young WΣ-Δglr1 cells growing under fermenting (green) or respiratory (turquois) conditions plotted versus the average number of budscars of the populations. (d) Comparison of the population-averaged number of budscars of wild-type (SC: gray triangles, SG: black inverted triangles) and Δglr (SC: magenta circles; SG: purple squares) cells, enriched from the flow-cytometric aging assay. A Gompertz function was fit to the data to visualize the trend. The data for wild-type cells are the same as in Figure 1i shown for comparison. (e) Replicative competence of aged wild-type and Δglr1 cells grown under fermenting (SC) and respiring (SG) conditions at different replicative age. Colony-forming units were calculated from the sort of 384 single viable cells (as judged by GFP fluorescence) from each time point on YPD-agar plates. The data of 10 (wild type, SC), two (wild type, SG), two (Δglr1, SC) and two (Δglr1, SG) independent aging experiments are shown. The data for wild-type cells are the same as in Figure 1j shown for comparison. (f) Classical lifespan determination of WΣ wild-type (gray and black) and Δglr1 (magenta and purple) by micromanipulation on 2% glucose (Glc)- or 3% glycerol- and 2% ethanol (GlyEtOH)-containing plates. For wt (Glc: median=23; n=71; GlyEtOH: median=30.5; n=92; same data as in Figure 1k shown for comparison). For Δglr1 (Glc: median=22; n=72; GlyEtOH: median=24; n=94). Data from two independent experiments each shown in one survival curve for clarity.

Figure 4

Aged yeast cells buffer an H₂O₂ shock more quickly than young cells through a Glr1-dependent pathway. (a) Response kinetics of glutathione potential after addition of 2 mmol/l H₂O₂ to aging wild-type cells growing under fermentative conditions at the indicated time points. Moving average over 25 cells (raw data are shown in Supplementary Figure S4A). (b) Same kinetics as in a for WΣ-Δglr1 but with 0.5 mmol/l H₂O₂. Moving average over 25 cells (raw data are shown in Supplementary Figure S4B). (c) Steady-state oxidation of the GSSG/2GSH redox couple in wild-type and Δglr1 strains and upon glr1 overexpression. GSSG/2GSH redox couple measured as % oxidation of the roGFP2-Grx1 probe in fermenting cells. Ectopic expression of glr1 can complement the knockout, but does not lead to a more reduced glutathione potential in wild-type cells (wt, n=1; wt+Glr1, n=4; Δglr1, n=2; Δglr1+Glr1, n=4. Error bars are the s.d.). (d and e) Glutathione potential of fermenting young wild-type (d) and glr1-overexpressing (e) cells in response to 2 mmol/l H₂O₂. The dashed green line indicates steady-state oxidation. (f and g) roGFP2-Orp1 probe oxidation in fermenting young wild-type (f) and glr1-overexpressing (g) cells in response to 2 mmol/l H₂O₂.

Figure 5

The activity of Glr1 is pH dependent in vivo. (a and b) Change of cytosolic pH (a) and GSSG/2GSH redox couple oxidation (b) in response to addition of 25 μmol/l DNP (2,4-dinitrophenol) in wild-type yeast growing in synthetic medium with glucose. Both pHluorin and roGFP2-Grx1 were measured within the same culture but in different cells, one marked with TagBFP. (c and d) Gradually lowering the cytosolic pH of a fermenting (c) and respiring (d) yeast culture by different concentrations of DNP (0–100 μmol/l). The relative oxidation of the cytosolic GSSG/2GSH redox couple of differentially tagged co-cultured wild-type (green), Δglr1 (magenta) and glr1-overexpressing (↑glr1, blue) cells was determined simultaneously 30 min after DNP addition. The gray area is outside the unambiguous range of the pHluorin reporter and pH measurements in this area are imprecise. The open circles in d indicate a pH upshift by the addition of Tris base. Standard error of the mean is smaller then the symbols. (e) Cytosolic pH of wild-type WΣ and Δglr1 strains under fermenting (2% glucose) and respiring (3% glycerol+2% ethanol) conditions (data for wild-type WΣ cells are the same as shown in Figure 1g and shown for comparison); wt glucose, n=27; wt glycerol, n=31, Δglr1 glucose, n=13; Δglr1 glycerol, n=13. Error bars are the s.d. (f) Gradual uncoupling of cells by DNP titration. The cytosolic pH, measured 30 min after DNP addition, for fermenting cells (blue colors) and respiring cells (red colors) is shown. The pH upshift was achieved by increasing amounts of Tris base as indicated by the arrow. Shown are three independent experiments.

Figure 6

Model for age-dependent changes in pH, H₂O₂ levels and glutathione potential. (a) In young fermenting cells, glycolysis runs at an optimal rate and ATP levels are sufficiently high to power the plasma membrane proton pump Pma1, which is able to maintain a high cytosolic pH. Glr1 operates at suboptimal rates keeping the GSSG/2GSH redox couple in a slightly oxidized steady state. In old cells, ATP levels are reduced possibly due to the impairment of glycolysis by elevated H₂O₂ levels and/or other age-associated factors. The proton pumps (like Pma1) work at limiting rates and cannot maintain the high cytosolic pH anymore. Glr1 is activated by the lower pH and works at its maximal rate, leading to a more-reducing glutathione potential. Top panels indicate idealized curves for the changes in pH, H₂O₂ (roGFP2-Orp1) and glutathione potential (roGFP2-Grx1). (b) In young respiring cells, ATP is produced by oxidative phosphorylation (OxPhos). The respiring cells already have a lower cytosolic pH in which Glr1 shows its highest activity, leading to a more-reducing steady-state glutathione potential as compared with the fermenting cells. In old respiring cells, mitochondrial energy production may be less efficient, which possibly leads to generation of more ROS including H₂O₂. This constitutive burden may also affect the steady state of the GSSG/2GSH redox system and leads to a more -oxidizing glutathione potential. Top panels indicate idealized curves for the changes in pH, H₂O₂ (roGFP2-Orp1) and oxidation state of the GSSG/2GSH redox couple (roGFP2-Grx1).