Thiol Peroxidases as Major Regulators of Intracellular Levels of Peroxynitrite in Live Saccharomyces cerevisiae Cells

Thiol peroxidases (TP) are ubiquitous and abundant antioxidant proteins of the peroxiredoxin and glutathione peroxidase families that can catalytically and rapidly reduce biologically relevant peroxides, such as hydrogen peroxide and peroxynitrite. However, the TP catalytic cycle is complex, depending on multiple redox reactions and partners, and is subjected to branching and competition points that may limit their peroxide reductase activity in vivo. The goals of the present study were to demonstrate peroxynitrite reductase activity of TP members in live cells in real time and to evaluate its catalytic characteristics. To these ends, we developed a simple fluorescence assay using coumarin boronic acid (CBA), exploiting that fact that TP and CBA compete for peroxynitrite, with the expectation that higher TP peroxynitrite reductase activity will lower the CBA oxidation. TP peroxynitrite reductase activity was evaluated by comparing CBA oxidation in live wild type and genetically modified Δ8 (TP-deficient strain) and Δ8+TSA1 (Δ8 strain that expresses only one TP member, the TSA1 gene) Saccharomyces cerevisiae strains. The results showed that CBA oxidation decreased with cell density and increased with increasing peroxynitrite availability. Additionally, the rate of CBA oxidation decreased in the order Δ8 > Δ8+TSA1 > WT strains both in control and glycerol-adapted (expressing higher TP levels) cells, showing that the CBA competition assay could reliably detect peroxynitrite in real time in live cells, comparing CBA oxidation in strains with reduced and increased TP expression. Finally, there were no signs of compromised TP peroxynitrite reductase activity during experimental runs, even at the highest peroxynitrite levels tested. Altogether, the results show that TP is a major component in the defense of yeast against peroxynitrite insults under basal and increasing stressful conditions.

Only a few cellular species have been identified as possible catalytic peroxynitrite scavengers, including members of the thiol peroxidase family [6]. Thiol peroxidases (TP), including peroxiredoxins (Prx) and glutathione peroxidases (GPx), are ubiquitous, abundant, and widespread proteins that react with a variety of biologically relevant peroxides and are, therefore, regarded as an important part of the cellular antioxidant repertoire [7]. TP may have different biological functions [7,8], but all of them essentially rely on their exceptional reactivity toward hydroperoxides.
The assumption that TP detoxify peroxynitrite is based on the high rate constants by which TP members reduce peroxynitrite to the non-oxidant nitrite (NO 2 − ) anion and in few studies addressing the growth or death of cells under exposure to peroxynitrite [9,10]. However, the catalytic cycle of TP is complex, depending on multiple redox reactions and partners (Scheme 1), a cascade of thiol-disulfide exchange reactions that may limit their peroxide reductase activities. Furthermore, most of evidence for cellular TP activity is from experiments exposing cells to oxidants and monitoring growth or are limited to laboratories that use molecular biology techniques, using genetically encoded probes involving a single TP (GFP-TP fusions) that focus on the monitoring of H 2 O 2 [11]. As a consequence, there is a lack of temporal data describing how TP members globally handle hydroperoxides and peroxynitrite in real-time in live cells.
Only a few cellular species have been identified as possible catalytic peroxynitrite scavengers, including members of the thiol peroxidase family [6]. Thiol peroxidases (TP), including peroxiredoxins (Prx) and glutathione peroxidases (GPx), are ubiquitous, abundant, and widespread proteins that react with a variety of biologically relevant peroxides and are, therefore, regarded as an important part of the cellular antioxidant repertoire [7]. TP may have different biological functions [7,8], but all of them essentially rely on their exceptional reactivity toward hydroperoxides. The assumption that TP detoxify peroxynitrite is based on the high rate constants by which TP members reduce peroxynitrite to the non-oxidant nitrite (NO2 − ) anion and in few studies addressing the growth or death of cells under exposure to peroxynitrite [9,10]. However, the catalytic cycle of TP is complex, depending on multiple redox reactions and partners (Scheme 1), a cascade of thiol-disulfide exchange reactions that may limit their peroxide reductase activities. Furthermore, most of evidence for cellular TP activity is from experiments exposing cells to oxidants and monitoring growth or are limited to laboratories that use molecular biology techniques, using genetically encoded probes involving a single TP (GFP-TP fusions) that focus on the monitoring of H2O2 [11]. As a consequence, there is a lack of temporal data describing how TP members globally handle hydroperoxides and peroxynitrite in real-time in live cells.
The goals of the present study were to demonstrate peroxidase activity of TP members in live cells in real time and to evaluate its catalytic properties. In order to address these needs, we Scheme 1. Generic representation of peroxiredoxins catalytic cycle. (1) Peroxidatic cysteine (S P H) oxidation to sulfenic acid (SOH) by peroxide and the formation of the leaving group (ROH).
The goals of the present study were to demonstrate peroxidase activity of TP members in live cells in real time and to evaluate its catalytic properties. In order to address these needs, we employed a simple and readily accessible fluorescence spectroscopy-based competition assay using a peroxynitrite-reactive boronate compound to monitor peroxynitrite in real-time in Saccharomyces cerevisiae cells. The role of TP was investigated by comparing the boronate oxidation in live wild type and TP-deficient Saccharomyces cerevisiae cells with the expectation that higher TP peroxynitrite reductase activity will lower the coumarin boronic acid (CBA) oxidation.

The Competition Assay
The basis for the peroxynitrite competition assay to test the intracellular TP peroxinitrite reductase activity in real-time is presented in Scheme 2. Briefly, cells were continuously exposed to a flux of peroxynitrite (ONOOH/ONOO − ; pKa 6.9) in the presence of CBA, which rapidly and irreversibly (k = 1 x 10 6 M −1 s −1 ) reacts with peroxynitrite (Scheme 2, right arm). This reaction yields nitrite and the fluorescent product 7-hydroxy-coumarin (COH) [12]. Thus, in cells continuously exposed to peroxynitrite, TP and CBA will compete for peroxynitrite. Assuming that the levels of other cellular peroxynitrite reactive constituents remained unchanged, it was expected that the higher the TP peroxynitrite reductase activity would attenuate the rise of fluorescence by reducing the COH accumulation. The total peroxynitrite reductase activity of TP enzymes were accessed by comparing the CBA oxidation rates in live WT and TP-deficient S. cerevisiae strains [13].

The Competition Assay
The basis for the peroxynitrite competition assay to test the intracellular TP peroxinitrite reductase activity in real-time is presented in Scheme 2. Briefly, cells were continuously exposed to a flux of peroxynitrite (ONOOH/ONOO − ; pKa 6.9) in the presence of CBA, which rapidly and irreversibly (k = 1 x 10 6 M -1 s -1 ) reacts with peroxynitrite (Scheme 2, right arm). This reaction yields nitrite and the fluorescent product 7-hydroxy-coumarin (COH) [12]. Thus, in cells continuously exposed to peroxynitrite, TP and CBA will compete for peroxynitrite. Assuming that the levels of other cellular peroxynitrite reactive constituents remained unchanged, it was expected that the higher the TP peroxynitrite reductase activity would attenuate the rise of fluorescence by reducing the COH accumulation. The total peroxynitrite reductase activity of TP enzymes were accessed by comparing the CBA oxidation rates in live WT and TP-deficient S. cerevisiae strains [13]. Scheme 2. Peroxynitrite competition. Peroxynitrite is reduced to nitrite through reactions with thiol peroxidases (TP) and coumarin boronic acid (CBA), reacts directly with other target or undergoes CO2/H + -induced decomposition to radicals.
S. cerevisiae was chosen as the model for the present study because this yeast expresses eight TP members (five Prx and three GPx) and viable strains lacking specific TPs, and all eight isoforms are available [13]. Additionally, S. cerevisiae also expresses a cytochrome c peroxidase (Ccp1), which has been shown to react rapidly with peroxynitrite [14].
Peroxynitrite fluxes were generated by combining an NO • donor and paraquat (PQ), a redox cycler superoxide-generating compound [15]. This combination produces peroxynitrite through reaction between NO • and O2 •− radicals [4], and is herein referred to as the PQ/NO • donor.

CBA Oxidation in Wild Type and Δ8 Yeast Cells during Peroxynitrite Generation
Suspensions of WT and Δ8 S. cerevisiae cells were transferred to wells of a 96-well plate and exposed to the PQ/NO • donor and CBA. The fluorescence of COH was continuously monitored in a plate reader. There was a small CBA oxidation in control experiments in the presence of the NO • donor (in the absence of PQ), probably due to cellular O2 •− production. Notably, CBA oxidation was negligible in control experiments in the absence of the NO • donor (even in the presence of PQ), confirming that the changes in fluorescence were due to peroxynitrite under the experimental conditions of the study. Importantly, CBA oxidation rate increased significantly in cells exposed to PQ/NO • donor and was always higher for the Δ8 strain ( Figure 1A). These results are also shown as the rate of CBA oxidation ( Figure 1B), which is defined in the Figure 1 legend. Scheme 2. Peroxynitrite competition. Peroxynitrite is reduced to nitrite through reactions with thiol peroxidases (TP) and coumarin boronic acid (CBA), reacts directly with other target or undergoes CO 2 /H + -induced decomposition to radicals.
S. cerevisiae was chosen as the model for the present study because this yeast expresses eight TP members (five Prx and three GPx) and viable strains lacking specific TPs, and all eight isoforms are available [13]. Additionally, S. cerevisiae also expresses a cytochrome c peroxidase (Ccp1), which has been shown to react rapidly with peroxynitrite [14].
Peroxynitrite fluxes were generated by combining an NO • donor and paraquat (PQ), a redox cycler superoxide-generating compound [15]. This combination produces peroxynitrite through reaction between NO • and O 2 •− radicals [4], and is herein referred to as the PQ/NO • donor.

CBA Oxidation in Wild Type and ∆8 Yeast Cells during Peroxynitrite Generation
Suspensions of WT and ∆8 S. cerevisiae cells were transferred to wells of a 96-well plate and exposed to the PQ/NO • donor and CBA. The fluorescence of COH was continuously monitored in a plate reader. There was a small CBA oxidation in control experiments in the presence of the NO • donor (in the absence of PQ), probably due to cellular O 2 •− production. Notably, CBA oxidation was negligible in control experiments in the absence of the NO • donor (even in the presence of PQ), confirming that the changes in fluorescence were due to peroxynitrite under the experimental conditions of the study. Importantly, CBA oxidation rate increased significantly in cells exposed to PQ/NO • donor and was always higher for the ∆8 strain ( Figure 1A). These results are also shown as the rate of CBA oxidation ( Figure 1B), which is defined in the Figure 1 legend. PQ redox cycling depends on intracellular reducing agents [12] to generate O2 •− , which is a short-lived species with limited diffusibility through biological membranes. Therefore, the observed peroxynitrite production associated with the PQ/NO • donor must predominantly occur intracellularly, resulting in an environment where TP, other cellular components, and CBA compete for the oxidant.

Effect of Cell Density and NO • Donor Concentration on CBA Oxidation
The PQ/NO • donor-induced CBA oxidation was negligible in cell-free experiments (Figure 2A, labeled as -cells), confirming that paraquat requires intracellular reducing agents for O2 •− production, and that O2 •− and peroxynitrite production occur primarily inside cells. In the presence of cells, the rate of CBA oxidation was higher in the Δ8 strain than the WT strain and decreased with increasing cell density for both WT and Δ8 strains (Figure 2A,B), indicating that peroxynitrite-dependent oxidation of CBA may have partially occurred in the extracellular space.
Interestingly, the cell density-dependent effect on the rate of CBA oxidation was more pronounced for the Δ8 strain than the WT strain ( Figure 2B), which is consistent with the peroxynitrite reductase activity of TP in WT competing better with extracellular CBA oxidation than that of Δ8. The pronounced effect of cell density for Δ8 strain also suggests the existence of a TP-independent peroxynitrite reductase component of the Δ8 strain (and of WT). PQ redox cycling depends on intracellular reducing agents [12] to generate O 2 •− , which is a short-lived species with limited diffusibility through biological membranes. Therefore, the observed peroxynitrite production associated with the PQ/NO • donor must predominantly occur intracellularly, resulting in an environment where TP, other cellular components, and CBA compete for the oxidant.

Effect of Cell Density and NO • Donor Concentration on CBA Oxidation
The PQ/NO • donor-induced CBA oxidation was negligible in cell-free experiments ( Figure 2A, labeled as -cells), confirming that paraquat requires intracellular reducing agents for O 2 •− production, and that O 2 •− and peroxynitrite production occur primarily inside cells. In the presence of cells, the rate of CBA oxidation was higher in the ∆8 strain than the WT strain and decreased with increasing cell density for both WT and ∆8 strains (Figure 2A,B), indicating that peroxynitrite-dependent oxidation of CBA may have partially occurred in the extracellular space. Interestingly, the cell density-dependent effect on the rate of CBA oxidation was more pronounced for the ∆8 strain than the WT strain ( Figure 2B), which is consistent with the peroxynitrite reductase activity of TP in WT competing better with extracellular CBA oxidation than that of ∆8. The pronounced effect of cell density for ∆8 strain also suggests the existence of a TP-independent peroxynitrite reductase component of the ∆8 strain (and of WT).
The CBA oxidation rate increased linearly with NO • donor concentration in both strains ( Figure 3A), as expected from the competition between NO • and SODs for O 2 •− , and was higher for the ∆8 strain at every NO • donor concentration ( Figure 3B). The CBA oxidation rate increased linearly with NO • donor concentration in both strains ( Figure  3A), as expected from the competition between NO • and SODs for O2 •− , and was higher for the Δ8 strain at every NO • donor concentration ( Figure 3B).
It is worth mentioning that the fluorescence increase was linear throughout the experiments using varying cellular densities and increasing NO • donor concentrations, suggesting that the overall cellular peroxynitrite reductase activity of the WT and Δ8 strains remained unchanged throughout the experimental time scale.   The CBA oxidation rate increased linearly with NO • donor concentration in both strains ( Figure  3A), as expected from the competition between NO • and SODs for O2 •− , and was higher for the Δ8 strain at every NO • donor concentration ( Figure 3B).
It is worth mentioning that the fluorescence increase was linear throughout the experiments using varying cellular densities and increasing NO • donor concentrations, suggesting that the overall cellular peroxynitrite reductase activity of the WT and Δ8 strains remained unchanged throughout the experimental time scale.  It is worth mentioning that the fluorescence increase was linear throughout the experiments using varying cellular densities and increasing NO • donor concentrations, suggesting that the overall cellular peroxynitrite reductase activity of the WT and ∆8 strains remained unchanged throughout the experimental time scale.
Using the data of Figure 3A, the CBA oxidation rate was also plotted as a function of deta/NO concentration during three different time periods for WT ( Figure 4A) and ∆8 ( Figure 4B) strains. There was only a slight acceleration in the CBA oxidation rate over time for the WT strain at all NO • donor concentrations tested, again suggesting that the peroxynitrite reductase activity of WT S. cerevisiae cells remained virtually constant throughout the experimental run and that TP catalytically Antioxidants 2020, 9, 434 6 of 14 remove peroxynitrite. TP were also probably the limiting reactant. Overall, under the experimental conditions, the bulky (considering the intra and extracellular space) TP concentration was roughly between 0.04 and 0.5 µM (this calculation was performed assuming 100 µM TP, an intracellular concentration purposely exaggerated, and a S. cerevisiae diameter between 2 and 5 µm). Using a standard fluorescence analytical curve, the accumulated COH concentration after 1 h for the highest NO • donor concentration employed was estimated to be around 0.8 and 2.4 µM for the WT ( Figure 3A) and ∆8 ( Figure 3B) strains, respectively. It should be pointed out that this estimation did not take into account the peroxynitrite consumed by cellular targets other than CBA (including TP) since the CBA concentration used was not saturating. Thus, it was likely that TP catalytically reduced peroxynitrite under the experimental conditions, as total peroxynitrite production during experimental runs was greater than TP concentrations.
There was only a slight acceleration in the CBA oxidation rate over time for the WT strain at all NO • donor concentrations tested, again suggesting that the peroxynitrite reductase activity of WT S. cerevisiae cells remained virtually constant throughout the experimental run and that TP catalytically remove peroxynitrite. TP were also probably the limiting reactant. Overall, under the experimental conditions, the bulky (considering the intra and extracellular space) TP concentration was roughly between 0.04 and 0.5 μM (this calculation was performed assuming 100 μM TP, an intracellular concentration purposely exaggerated, and a S. cerevisiae diameter between 2 and 5 μm). Using a standard fluorescence analytical curve, the accumulated COH concentration after 1 hour for the highest NO • donor concentration employed was estimated to be around 0.8 and 2.4 μM for the WT ( Figure 3A) and Δ8 ( Figure 3B) strains, respectively. It should be pointed out that this estimation did not take into account the peroxynitrite consumed by cellular targets other than CBA (including TP) since the CBA concentration used was not saturating. Thus, it was likely that TP catalytically reduced peroxynitrite under the experimental conditions, as total peroxynitrite production during experimental runs was greater than TP concentrations.
A similar behavior was observed at the two lowest deta/NO concentrations with the Δ8 strain. For the two highest donor concentrations, for reasons that are not clear, negative slopes were observed. It could have been due to CBA exhaustion or fluorescent detector saturation. It was also observed that the CBA oxidation rate nonlinearly increased with PQ concentration (data not shown). This result was probably due to O2 •− production by PQ depending on other intracellular factors, such as cellular PQ take up, and O2 and/or PQ-reducing cellular component availabilities.

Oxidation of CBA in S. cerevisiae Adapted to Glycerol Medium
S. cerevisiae grown in glucose-rich medium [16][17][18] rely mostly upon fermentation to produce ATP. When glycerol is substituted for glucose as a carbon source, the yeast depend more on mitochondrial respiratory metabolism. One response yeast employ for adapting to glycerol-rich medium involves mitochondrial biogenesis [19]. Accordingly, we detected increased levels of the outer mitochondrial membrane protein porin ( Figure 5A, lanes 4-12; Figure 5B, bottom panel), A similar behavior was observed at the two lowest deta/NO concentrations with the ∆8 strain. For the two highest donor concentrations, for reasons that are not clear, negative slopes were observed. It could have been due to CBA exhaustion or fluorescent detector saturation.
It was also observed that the CBA oxidation rate nonlinearly increased with PQ concentration (data not shown). This result was probably due to O 2 •− production by PQ depending on other intracellular factors, such as cellular PQ take up, and O 2 and/or PQ-reducing cellular component availabilities.

Oxidation of CBA in S. cerevisiae Adapted to Glycerol Medium
S. cerevisiae grown in glucose-rich medium [16][17][18] rely mostly upon fermentation to produce ATP. When glycerol is substituted for glucose as a carbon source, the yeast depend more on mitochondrial respiratory metabolism. One response yeast employ for adapting to glycerol-rich medium involves mitochondrial biogenesis [19]. Accordingly, we detected increased levels of the outer mitochondrial membrane protein porin ( Figure 5A, lanes 4-12; Figure 5B, bottom panel), which has been used as a mitochondrial biogenesis marker [20]. A similar result was observed with cells grown in galactose or lactate-rich media. Increased porin expression was also observed in the stationary growth phase of cells grown in glucose ( Figure 5A, lane 3; Figure 5B, bottom). Previous studies have demonstrated that growing S. cerevisiae cells in these other carbon sources generates more mitochondrial oxidants and that the yeast, in turn, adapt by modulating the expression levels of antioxidant enzymes such as superoxide dismutase, catalase, and TP [21][22][23]. Indeed, mitochondrial peroxiredoxin 1 (Prx1) expression levels increased in yeast cells grown in media containing galactose, ethanol/glycerol, or lactate when compared to cells grown in glucose ( Figure 5A,B, top panel). Additionally, the WT strain displayed increased TSA1 protein expression levels when grown in glycerol ( Figure 6A,B, compare lanes 1 and 5).
which has been used as a mitochondrial biogenesis marker [20]. A similar result was observed with cells grown in galactose or lactate-rich media. Increased porin expression was also observed in the stationary growth phase of cells grown in glucose ( Figure 5A, lane 3; Figure 5B, bottom). Previous studies have demonstrated that growing S. cerevisiae cells in these other carbon sources generates more mitochondrial oxidants and that the yeast, in turn, adapt by modulating the expression levels of antioxidant enzymes such as superoxide dismutase, catalase, and TP [21][22][23]. Indeed, mitochondrial peroxiredoxin 1 (Prx1) expression levels increased in yeast cells grown in media containing galactose, ethanol/glycerol, or lactate when compared to cells grown in glucose ( Figure  5A,B, top panel). Additionally, the WT strain displayed increased TSA1 protein expression levels when grown in glycerol ( Figure 6A,B, compare lanes 1 and 5). . Yeast WT cells were grown in different carbon sources and collected in three different growth phases, as determined by the OD600. Cell extracts were prepared and proteins were separated by SDS-PAGE gel electrophoresis followed by immunoblotting with specific antibodies at the (a) early exponential (OD600 =v1.0), (b) mid-late exponential (OD600 = 2.0), and (c) stationary (OD600 > 2.0) growth phases. Pgk1 protein was used as the gel loading control. All experiments were performed in triplicate and repeated at least three times, yielding consistent results.
It was hypothesized that once the WT strain was glycerol-adapted, the peroxynitrite availability would be lower than that in WT cells grown in glucose, both because of increased prevention of peroxynitrite formation (by induction of SODs expression) and increased TP expression. To this end, experiments such as those described in Figures 2 and 3 were conducted with WT and Δ8 glycerol-adapted cells.
Control experiments in the absence of PQ ( Figure 6B, labeled -PQ) detected a slight decrease in the rate of CBA oxidation in glycerol-adapted cells when compared to the respective control group. As expected, in the presence of PQ, the CBA oxidation rate increased with NO • donor concentration in both control and glycerol-adapted cells. Additionally, glycerol-adapted WT cells had reduced CBA oxidation rates relative to control cells over the entire range of NO • donor concentrations tested ( Figure 6B), which is indicative of reduced peroxynitrite availability. Interestingly, the CBA oxidation rate of the glycerol-adapted Δ8 strain was also lower than the respective control cells ( Figure 6B). This finding is possibly related to decreased peroxynitrite formation and may also indicate that a TP-independent peroxynitrite reductase activity was present in these cells and that this system is up-regulated under respiratory conditions.

Oxidation of CBA in Δ8+TSA1 S. cerevisiae Strain
To further evaluate the role of TP as peroxynitrite reductase, we performed experiments with the Δ8+TSA1 strain. This yeast strain expresses the TSA1 gene under the control of a constitutive promoter (the endogenous translation elongation factor, TEF-1 alpha) [24]. The expression of TSA1 by the Δ8+TSA1 strain was confirmed at the protein level, and was comparable to the TSA1 expression levels in the WT strain ( Figure 6A, compare lanes 1 and 4). Notably, Δ8+TSA1 was significantly more resistant to peroxynitrite-dependent CBA oxidation when compared to the Δ8 strain ( Figure 7A), and the CBA oxidation rate of the glycerol-adapted Δ8+TSA1 cells was indistinguishable from the glycerol-adapted WT strain ( Figure 7B). These observations further show that TP is an important peroxynitrite reductase system. It was hypothesized that once the WT strain was glycerol-adapted, the peroxynitrite availability would be lower than that in WT cells grown in glucose, both because of increased prevention of peroxynitrite formation (by induction of SODs expression) and increased TP expression. To this end, experiments such as those described in Figures 2 and 3 were conducted with WT and ∆8 glycerol-adapted cells.
Control experiments in the absence of PQ ( Figure 6B, labeled -PQ) detected a slight decrease in the rate of CBA oxidation in glycerol-adapted cells when compared to the respective control group. As expected, in the presence of PQ, the CBA oxidation rate increased with NO • donor concentration in both control and glycerol-adapted cells. Additionally, glycerol-adapted WT cells had reduced CBA oxidation rates relative to control cells over the entire range of NO • donor concentrations tested ( Figure 6B), which is indicative of reduced peroxynitrite availability. Interestingly, the CBA oxidation rate of the glycerol-adapted ∆8 strain was also lower than the respective control cells ( Figure 6B). This finding is possibly related to decreased peroxynitrite formation and may also indicate that a TP-independent peroxynitrite reductase activity was present in these cells and that this system is up-regulated under respiratory conditions.

Oxidation of CBA in ∆8+TSA1 S. cerevisiae Strain
To further evaluate the role of TP as peroxynitrite reductase, we performed experiments with the ∆8+TSA1 strain. This yeast strain expresses the TSA1 gene under the control of a constitutive promoter (the endogenous translation elongation factor, TEF-1 alpha) [24]. The expression of TSA1 by the ∆8+TSA1 strain was confirmed at the protein level, and was comparable to the TSA1 expression levels in the WT strain ( Figure 6A, compare lanes 1 and 4). Notably, ∆8+TSA1 was significantly more resistant to peroxynitrite-dependent CBA oxidation when compared to the ∆8 strain ( Figure 7A), and the CBA oxidation rate of the glycerol-adapted ∆8+TSA1 cells was indistinguishable from the glycerol-adapted WT strain ( Figure 7B). These observations further show that TP is an important peroxynitrite reductase system.

Oxidation of CBA in Δ8+TSA1 S. cerevisiae Strain
To further evaluate the role of TP as peroxynitrite reductase, we performed experiments with the Δ8+TSA1 strain. This yeast strain expresses the TSA1 gene under the control of a constitutive promoter (the endogenous translation elongation factor, TEF-1 alpha) [24]. The expression of TSA1 by the Δ8+TSA1 strain was confirmed at the protein level, and was comparable to the TSA1 expression levels in the WT strain ( Figure 6A, compare lanes 1 and 4). Notably, Δ8+TSA1 was significantly more resistant to peroxynitrite-dependent CBA oxidation when compared to the Δ8 strain ( Figure 7A), and the CBA oxidation rate of the glycerol-adapted Δ8+TSA1 cells was indistinguishable from the glycerol-adapted WT strain ( Figure 7B). These observations further show that TP is an important peroxynitrite reductase system.

The Growth of Different S. cerevisiae Strains Challenged with Peroxynitrite Fluxes
The growth of S. cerevisiae strains exposed to peroxynitrite fluxes generated by the PQ/NO • donor were monitored for 24 h (Figure 8A,B). Under simulated normal conditions, the growth of the

The Growth of Different S. cerevisiae Strains Challenged with Peroxynitrite Fluxes
The growth of S. cerevisiae strains exposed to peroxynitrite fluxes generated by the PQ/NO • donor were monitored for 24 h (Figure 8A,B). Under simulated normal conditions, the growth of the WT, ∆TSA1, and ∆8 strains was similar. However, when exposed to peroxynitrite fluxes, the ∆8 strain was drastically affected. These results are in agreement with earlier data [9,25] showing that TP members are important peroxynitrite scavengers.
Antioxidants 2020, 9, x FOR PEER REVIEW 9 of 15 WT, ΔTSA1, and Δ8 strains was similar. However, when exposed to peroxynitrite fluxes, the Δ8 strain was drastically affected. These results are in agreement with earlier data [9] [25] showing that TP members are important peroxynitrite scavengers.

Discussion
The TP catalytic cycle is complex and subjected to numerous side reactions. TP members reduce peroxynitrite (and other different hydroperoxides) with high rate constants via a conserved peroxidatic cysteine (CP), which is oxidized to sulfenic acid in the process (Scheme 1, step 1). Depending on the TP member and redox conditions, this sulfenic acid species can have different fates. For example, the sulfenic acid can be overoxidized by the hydroperoxide. However, a condensation reaction between the resulting Cp sulfenic acid and a second conserved residue called the resolving cysteine (CR), yielding a disulfide bridged form (Scheme 1, step 2), typically occurs following oxidation [26]. The disulfide TP forms can also have different fates, activating signaling events through covalent interactions with other proteins, including transcription factors [27], or be fully reduced back to the resting peroxide reactive form via thiol-based oxidoreductase partners (Scheme 1, step 3) [28][29][30]. In Saccharomyces cerevisiae, the oxidized form of both Prx and GPx are reduced, through thiol-disulfide exchange reactions, by each one of the three oxidoreductases called thioredoxin (Trx) [31]. The resulting disulfide formed in Trx is subsequently reduced via NADPH-dependent thiol-disulfide exchange reactions with the flavoenzyme thioredoxin reductase (TR) (Scheme 1, steps 4 and 5; reviewed in [32]). TP-mediated reduction of peroxynitrite is rapid, with rate constants varying between 10 6 and 10 8 M −1 s −1 [10,33,34],depending on TP member and substrate. Few previous studies measuring the rate constant of step 2 (Scheme 1) have shown that this reaction significantly varies among TP members [31,32,35,36], but it is typically a rapid unimolecular first-order process, which is complete within a few seconds. The kinetics of the thioldisulfide exchange reaction downstream from the resolution step has only been assessed for a limited number of redox partners, with rate constants ranging from 10 5 to 10 6 M −1 s −1 [34,[37][38][39].
Although the several steps of the TP catalytic cycle exhibit high reaction rate constants, TP peroxide reductase turnover is directly dependent on the kinetics of each successive step, the concentrations of downstream redox partners, and on the aforementioned branching points of the whole catalytic cycle species. Perturbations in any of these parameters could potentially limit TP peroxide reductase catalytic activity. For example, steady-state NADPH oxidation measurements in the presence of the entire TP system (Scheme 1) underestimated the rate constant of the reaction between the Cp of several Prx enzymes and different peroxides (step 1) by 3 to 4 orders of

Discussion
The TP catalytic cycle is complex and subjected to numerous side reactions. TP members reduce peroxynitrite (and other different hydroperoxides) with high rate constants via a conserved peroxidatic cysteine (C P ), which is oxidized to sulfenic acid in the process (Scheme 1, step 1). Depending on the TP member and redox conditions, this sulfenic acid species can have different fates. For example, the sulfenic acid can be overoxidized by the hydroperoxide. However, a condensation reaction between the resulting Cp sulfenic acid and a second conserved residue called the resolving cysteine (C R ), yielding a disulfide bridged form (Scheme 1, step 2), typically occurs following oxidation [26]. The disulfide TP forms can also have different fates, activating signaling events through covalent interactions with other proteins, including transcription factors [27], or be fully reduced back to the resting peroxide reactive form via thiol-based oxidoreductase partners (Scheme 1, step 3) [28][29][30]. In Saccharomyces cerevisiae, the oxidized form of both Prx and GPx are reduced, through thiol-disulfide exchange reactions, by each one of the three oxidoreductases called thioredoxin (Trx) [31]. The resulting disulfide formed in Trx is subsequently reduced via NADPH-dependent thiol-disulfide exchange reactions with the flavoenzyme thioredoxin reductase (TR) (Scheme 1, steps 4 and 5; reviewed in [32]). TP-mediated reduction of peroxynitrite is rapid, with rate constants varying between 10 6 and 10 8 M −1 s −1 [10,33,34], depending on TP member and substrate. Few previous studies measuring the rate constant of step 2 (Scheme 1) have shown that this reaction significantly varies among TP members [31,32,35,36], but it is typically a rapid unimolecular first-order process, which is complete within a few seconds. The kinetics of the thiol-disulfide exchange reaction downstream from the resolution step has only been assessed for a limited number of redox partners, with rate constants ranging from 10 5 to 10 6 M −1 s −1 [34,[37][38][39].
Although the several steps of the TP catalytic cycle exhibit high reaction rate constants, TP peroxide reductase turnover is directly dependent on the kinetics of each successive step, the concentrations of downstream redox partners, and on the aforementioned branching points of the whole catalytic cycle species. Perturbations in any of these parameters could potentially limit TP peroxide reductase catalytic activity. For example, steady-state NADPH oxidation measurements in the presence of the entire TP system (Scheme 1) underestimated the rate constant of the reaction between the Cp of several Prx enzymes and different peroxides (step 1) by 3 to 4 orders of magnitude, possibly because experiments were performed with limited concentrations of the downstream redox partners. Only more direct kinetic approaches revealed the exceptionally high rate constants for this step. In cells, the TP catalytic system may be limited still by other parameters such as the substrate and/or TP compartmentalization, expression of TP redox system members, redox partner competition between TPs and other cellular oxidases, cellular redox status, and Cp hyperoxidation [26,27]. Furthermore, it has been demonstrated that TP enzymes undergo structural switches, from a fully-folded decamer to a locally unfolded dimer, during catalysis [26,40]. Individually or collectively, these variables represent numerous potential bottlenecks for the TP peroxide reductase activity in living cells. Accordingly, Bayer et al. [41] have suggested that Prx 2 remove hydrogen peroxide from red blood cells in a stoichiometric manner.
In the present study, we developed a fluorescence-based competition assay that is capable of monitoring, in real-time, the TP-mediated peroxidase activity in live cells. The assay can be used to measure TP peroxynitrite reductase activity in presumably any cell type, provided that the appropriate control experiments are performed. The CBA approach also allowed us to make some other interesting observations. The ability of our CBA assay to monitor peroxynitrite in real time gave us evidence that TP members constitute a relevant and catalytic peroxynitrite reductase system in living S. cerevisiae cells. Accordingly, the CBA oxidation rate (i.e., fluorescence increase) was linear throughout most of the 60-90 min time span of the experiments, with no signs of partway through acceleration of CBA oxidation, suggesting that TP levels and activities were not compromised during experimental runs, even though the accumulated peroxynitrite formed exceeded total TP concentration. These results are in line with TP kinetic properties and add the fact that the potential cellular bottlenecks factors do not limit the peroxynitrite activity to the point where removal of the peroxides becomes stoichiometric in S. cerevisiae cells under different respiratory metabolic and stressful conditions. Similar results would likely be observed with other biologically relevant TP substrates, such as hydrogen peroxide. In addition, the conclusion that TP catalytically scavenge peroxynitrite and other peroxides may be cautiously extrapolated to mammalian cells, whose TP reactivity, kinetic properties, and catalytic cycle are not fundamentally different from yeast TP. The CBA competition assay could in theory be used to test this expectation using simple cellular models. Such peroxynitrite reductase activity of mammalian cells is particularly important for activated macrophages, potentially helping sustain their functions during episodes of infection and inflammation, where peroxynitrite formation is likely and a long-lasting, macrophage-mediated immune response is critical.
The ∆8 strain results suggest the existence of a TP-independent peroxynitrite reductase species. It was previously shown, using a fundamentally different experimental approach, that ∆8 strain survival and growth under oxidative conditions is strictly dependent on mitochondrial Ccp1 heme-protein [24]. The ferric species of this protein rapidly reduces hydrogen peroxide and peroxynitrite (to nitrite) and is concomitantly oxidized to a compound 1-like heme-peroxidase, which is fully reduced back to the ferric, peroxynitrite reactive species by two successive one-electron reductions by mitochondrial ferrous cytochrome c [30], and is thus highly dependent on the reducing power of the mitochondria. Interestingly, the ∆8 strain that naturally acquired an extra copy of chromosome XI, where the CCP1 gene is located (consequently increasing Ccp1 protein expression), was positively selected. As well as this, genetic deletion of chromosome XI or the CCP1 gene, specifically, is lethal to the ∆8 strain [24]. Thus, Ccp1 apparently could serve as a backup system when the TP catalytic cycle is compromised [24]. In the present study, we showed that CBA oxidation in ∆8 strain exposed to fluxes of peroxynitrite decreases with increasing cell densities ( Figure 2) and attenuates in glycerol adapted ∆8 cells when compared to ∆8 control cells ( Figure 6B). Taken together, these results suggest that a TP-independent peroxynitrite reductase system exists in ∆8 strain (in WT strain), and that it is upregulated under respiratory conditions, which is consistent with a role for Ccp1. However, our data indicate that Ccp1 cannot fully compensate for TP deficiency, as the levels of CBA oxidation were always noticeably higher for the ∆8 when compared to the WT strain.
In summary, our work provides a global picture of the peroxynitrite metabolism in living S. cerevisiae in real time, showing that TP sustain a catalytic peroxynitrite reductase activity under simulated normal and increasingly stressful conditions.

SDS-PAGE and Western Blotting
Proteins were separated by SDS-PAGE. Routinely, proteins were transferred to nitrocellulose membranes (Amersham Biosciences Protran Premium, GE Healthcare) and incubated with primary antibodies (anti-porin (Thermo Fisher Scientific-16G9E6BC4), anti-Pgk1 (Nordic Immunology-NE130/7S), and anti-Tsa1 antibody, from Fernando Gomes [36]). After washing the membranes, the specific secondary antibodies (anti-rabbit IgG, HRP-linked (Cell Signaling Technology) and anti-mouse IgG HRP-linked (Cell Signaling Technology)) were added. The immunoblots were developed using the ECL prime Western blotting detection reagent (GE Healthcare). The normalized band densitometry measurements were performed with the aid of the ImageJ software.