The Catalase KatA Contributes to Microaerophilic H2O2 Priming to Acquire an Improved Oxidative Stress Resistance in Staphylococcus aureus

Staphylococcus aureus has to cope with oxidative stress during infections. In this study, S. aureus was found to be resistant to 100 mM H2O2 during aerobic growth. While KatA was essential for this high aerobic H2O2 resistance, the peroxiredoxin AhpC contributed to detoxification of 0.4 mM H2O2 in the absence of KatA. In addition, the peroxiredoxins AhpC, Tpx and Bcp were found to be required for detoxification of cumene hydroperoxide (CHP). The high H2O2 tolerance of aerobic S. aureus cells was associated with priming by endogenous H2O2 levels, which was supported by an oxidative shift of the bacillithiol redox potential to −291 mV compared to −310 mV in microaerophilic cells. In contrast, S. aureus could be primed by sub-lethal doses of 100 µM H2O2 during microaerophilic growth to acquire an improved resistance towards the otherwise lethal triggering stimulus of 10 mM H2O2. This microaerophilic priming was dependent on increased KatA activity, whereas aerobic cells showed constitutive high KatA activity. Thus, KatA contributes to the high H2O2 resistance of aerobic cells and to microaerophilic H2O2 priming in order to survive the subsequent lethal triggering doses of H2O2, allowing the adaptation of S. aureus under infections to different oxygen environments.


Introduction
Staphylococcus aureus is a major human pathogen that can cause local skin and soft tissue infections, as well as life-threatening diseases, such as septicaemia, endocarditis, necrotizing pneumonia and osteomyelitis [1][2][3]. During infections, S. aureus has to cope with reactive oxygen species (ROS), such as superoxide anion (O 2 •− ) and hydrogen peroxide (H 2 O 2 ) [4], which are produced during the oxidative burst of activated macrophages and neutrophils to kill the invading pathogen [5][6][7][8]. The NADPH oxidase (NOX2) in the phagosomal membrane catalyses the one-electron transfer to molecular oxygen (O 2 ) to generate O 2 •− , which is dismutated to H 2 O 2 either spontaneously or by superoxide dismutases (SODs), including the extracellular SOD3 [6,[8][9][10][11]. Both SOD3 and NOX2 are contained in secretory vesicles in macrophages and neutrophils and the fusion of these vesicles with phagosomes during phagocytosis might provide a mechanism for catalysed H 2 O 2 production [11]. However, in neutrophils, the myeloperoxidase MPO is released from azurophilic granula into the phagosomal lumen, catalysing the dismutation of O 2 •− to H 2 O 2 upon infection [6]. MPO further converts H 2 O 2 with chloride to the highly reactive hypochlorous acid (HOCl), which is the most potent oxidant and microbicidal agent released by activated neutrophils [6,9,10]. In addition, S. aureus encounters endogenous ROS during aerobic respiration due to the stepwise one-electron transfer reactions to O 2 , leading to production of O 2 •− and H 2 O 2 [12]. In the Fenton reaction, H 2 O 2 reacts with free Fe 2+ to generate the highly toxic hydroxyl radical (OH • ), which can damage all cellular macromolecules, resulting in oxidation of proteins, lipids and carbohydrates [12][13][14]. However, H 2 O 2 and HOCl can also function in redox signalling to activate or inactivate specific redox-sensitive regulators, which control defence mechanisms and confer resistance against the oxidants in bacterial pathogens [14,15].
S. aureus uses various enzymatic and non-enzymatic ROS and HOCl detoxification systems, such as a unique catalase (KatA), several peroxiredoxins (AhpC, Tpx, Bcp) and the low-molecular-weight (LMW) thiol bacillithiol (BSH) [14,16]. BSH associates with the bacilliredoxin/BSH/YpdA redox pathway to regenerate oxidized protein thiols and bacillithiol disulfide (BSSB) [16]. We previously constructed a Brx-roGFP2 fused biosensor to monitor the changes in the BSH redox potential (E BSH ) under oxidative stress in S. aureus [17]. This study already revealed that S. aureus is highly resistant to 100 mM H 2 O 2 , since the Brx-roGFP2 biosensor responded only weakly to high H 2 O 2 levels, leading to small E BSH changes [17]. The catalase KatA was identified as the major H 2 O 2 detoxification enzyme, which conferred the constitutive H 2 O 2 -resistant phenotype to aerobically grown S. aureus cells [18,19]. KatA is also important for nasal colonization and mediates protection under macrophage and neutrophil infections [18,[20][21][22]. The peroxiredoxin AhpCF showed compensatory roles in resistance to H 2 O 2 and organic hydroperoxides (OHPs) and contributed to nasal colonization [20]. OHPs (ROOH) are generated during oxidation of polyunsaturated fatty acids of eukaryotic membrane lipids and are reduced by peroxiredoxins to their corresponding organic alcohols [23].
In general, AhpC, Tpx and Bcp can be classified into typical (AhpC) or atypical (Tpx, Bcp) 2-Cys peroxiredoxins based on their thiol-oxidation mechanism between the peroxidatic (C P ) and resolving Cys (C R ), involving inter-or intramolecular disulfides, respectively [24]. The functions and substrates of AhpC, Tpx and Bcp have been previously studied in Escherichia coli. AhpC detoxification of H 2 O 2 leads to formation of an oxidized AhpC dimer, which aggregates to an oligomer with chaperone functions [24,25]. Regeneration of AhpC requires the NADPH-dependent flavin disulfide reductase AhpF as a redox partner [24,25]. The thiol-peroxidase Bcp of E. coli is induced by H 2 O 2 , OHPs and during aerobic growth and confers resistance against H 2 O 2 and OHP stress [26]. The thiol-peroxidase Tpx of E. coli has been shown to catalyse detoxification of H 2 O 2 and OHPs in vitro and is recycled by the Trx/TrxR system [27,28]. In S. aureus, Tpx responds strongly to H 2 O 2 and other thiol-reactive compounds and was oxidized in the redox proteome under HOCl stress [14,29]. However, the detailed functions of the peroxiredoxins AhpC, Tpx and Bcp in peroxide resistance, detoxification and survival have not been studied thus far in S. aureus.
In S. aureus, transcription of katA, ahpCF and bcp is strongly induced only by high levels of 10 mM H 2 O 2 and controlled by the peroxide-responsive PerR repressor [18,20,30]. In Bacillus subtilis, KatA is also a member of the PerR regulon but already inducible by sub-lethal doses of 100 µM H 2 O 2 [31,32]. Pretreatment of B. subtilis cells with sub-lethal H 2 O 2 as "priming stimulus" confers improved resistance towards subsequent lethal H 2 O 2 doses, termed as "triggering stimulus", which are encountered as future stress [33,34]. These terms and abbreviations-priming (P), priming plus triggering (PT) and triggering (T)-were previously introduced within our project SFB973, which was directed to priming and memory of stress responses in different organisms, including bacteria, fungi and plants [35]. In B. subtilis, the H 2 O 2 priming effect was shown to be mediated by KatA, which is induced by a mild stress to prepare the cells for better survival when faced with future lethal oxidative stress [33]. Similarly, H 2 O 2 priming for improved resistance towards the triggering stimulus was dependent on the OxyR-dependent enzymes KatG and AhpCF in E. coli and Salmonella Typhimurium [32,36,37]. Although S. aureus exhibits constitutive H 2 O 2 resistance during aerobic growth, it is unknown whether priming for improved H 2 O 2 resistance is possible under aerobic or microaerophilic conditions.
In this study, we used growth and survival phenotype analyses, Brx-roGFP2 biosensor measurements and transcriptional studies to investigate the functions of KatA and the peroxiredoxins AhpC, Tpx and Bcp in peroxide resistance, detoxification and priming during aerobic and microaerophilic growth. Our results showed that S. aureus is H 2 O 2 primable for improved resistance only under microaerophilic conditions, which are depen-dent on KatA. In contrast, aerobic growth already leads to increased levels of ROS, which causes KatA-dependent aerobic priming for constitutive H 2 O 2 resistance. While KatA confers H 2 O 2 resistance in S. aureus, the peroxiredoxins AhpC, Tpx and Bcp were shown to contribute to survival and resistance under CHP stress and regeneration of reduced E BSH upon recovery from CHP stress.

Bacterial Strains, Growth and Survival Assays
Bacterial strains, plasmids and primers are described in Tables S1-S3. For genetic manipulation, E. coli was cultivated in Luria Broth (LB) medium. S. aureus COL strains were grown in RPMI medium to an optical density at 500 nm (OD 500 ) of 0.5 and exposed to H 2 O 2 , cumene hydroperoxide (CHP) or hypochlorous acid (HOCl), followed by determination of colony-forming units (CFUs) in survival assays as previously described [38]. Each experiment was performed in at least three independent biological replicates and the results are presented as mean values with standard deviations (SD) from all biological replicates, as indicated in each figure legend. Statistical analysis was performed using Student's unpaired two-tailed t-test with the software Graph Prism. The biochemical compounds were purchased from Sigma Aldrich. The HOCl concentration was determined as previously described [39].

Mutants and Complemented Strains
The S. aureus COL ∆katA mutant and katA complemented strains were previously constructed [40]. The S. aureus ∆ahpC, ∆tpx, ∆bcp and ∆perR deletion mutants were constructed using the temperature-sensitive E. coli-S. aureus shuttle vector pMAD as previously described [41]. In brief, 500 bp of the up-and downstream flanking regions of the specific genes were fused by PCR, digested with BglII and SalI and ligated into pMAD. The constructs were electroporated into the restriction-negative S. aureus RN4220, followed by phage transduction using phage 81 into S. aureus COL [42]. For construction of the ∆ahpC∆katA double mutant, the plasmid pMAD-∆katA of S. aureus RN4220-pMAD-∆katA was transduced by the phage 81 into the S. aureus COL ∆ahpC mutant. Selection of the ∆ahpC, ∆bcp, ∆tpx, ∆perR and ∆ahpC∆katA deletion mutants was performed as previously described [38].

Priming and Triggering Experiments
For priming and triggering, the S. aureus strains were grown aerobically under vigorous agitation in shake flasks in a shaking water bath at 150 rpm or microaerophilically in 50 mL Falcon tubes including 40 mL cultures with closed lids without shaking, as in previous publications [43,44]. At an OD 500 of 0.3, naïve S. aureus cells were primed by adding sub-lethal doses of 0.1 or 1 mM H 2 O 2 , respectively, to the bacterial culture for 30 min. Subsequently, the lethal triggering doses of 10 or 40 mM H 2 O 2 , respectively, were added to the primed bacterial cultures, followed by counting of CFUs after 2 and 4 h of growth. For triggering only, naïve cells were treated with 10 or 40 mM H 2 O 2 at an OD 500 of 0.4, followed by counting of CFUs after 2 and 4 h of growth.

Brx-roGFP2 Biosensor Measurements
To monitor the E BSH changes after H 2 O 2 and CHP stress, we used the Brx-roGFP2 biosensor expressing WT, ∆katA, ∆ahpC, ∆ahpC∆katA, ∆tpx and ∆bcp mutant strains and performed injection assays with the oxidants. For measurements of Brx-roGFP2 oxidation during microaerophilic and aerobic H 2 O 2 priming and triggering experiments, the S. aureus COL strain expressing Brx-roGFP2 was cultivated in LB medium to an OD 540 of 0.3 and challenged with the priming dose of 0.1 mM H 2 O 2 for 30 min, followed by the triggering dose of 10 mM H 2 O 2 , as described above. Samples were harvested from S. aureus cells in the naïve (C), primed (P), primed and triggered (PT) and triggered-only (T) states, alkylated with 10 mM N-ethylmaleimide (NEM), washed and resuspended in PBS with 10 mM NEM. The Brx-roGFP2 oxidation degree (OxD) and E BSH changes were determined in the S. aureus strains during oxidant injection or in samples harvested at C, P, PT and T as previously described [17,45]. For fully reduced and oxidized controls, biosensor strains were treated with 10 mM DTT and 5 mM diamide, respectively. The Brx-roGFP2 fluorescence emission was measured at 510 nm after excitation at 405 and 488 nm using the CLARIOstar microplate reader (BMG Labtech). The OxD of the Brx-roGFP2 biosensor was determined for each sample and normalized to fully reduced and oxidized controls as previously described [17,45].

Northern Blot Analyses
To analyze transcription of katA, dps and ahpCF in the S. aureus COL WT, ∆katA, ∆ahpC and ∆perR mutants using Northern blots, the S. aureus strains were grown in RPMI medium and harvested during the log phase at an OD 500 of 0.4. To investigate katA and ahpC induction in the priming and triggering experiments, S. aureus WT cells were harvested in the naïve (C), primed (P), primed and triggered (PT) and triggered-only (T) states, as explained in the figure legends. RNA isolation was performed using the acid phenol extraction protocol as described previously [46]. Northern blot hybridizations were conducted using digoxigenin-labelled antisense RNA probes for katA, ahpC and dps that were synthesized in vitro using T7 RNA polymerase and the corresponding primers katA-NB-for/rev and ahpC-NB-for/rev, as previously described [46]. The dps antisense RNA probe was constructed previously [44].

Determination of the Catalase Activity Using Native PAGE and Diaminobenzidine Staining
To analyze catalase activities in S. aureus COL strains, protein extracts were prepared under native conditions and 50 µg of each sample was separated by non-denaturing 10% polyacrylamide gel electrophoresis. The gel was stained for catalase activity using 50 µg/mL horseradish peroxidase coupled with 5 mM H 2 O 2 and 0.5 mg/mL diaminobenzidine, as described previously [47,48].

S. aureus Exhibits KatA-Dependent H 2 O 2 Resistance during Aerobic Growth
To investigate the roles of the catalases and peroxiredoxins in the constitutive H 2 O 2 resistance of S. aureus COL, phenotype analyses of the ∆katA, ∆ahpC, ∆ahpC∆katA, ∆tpx and ∆bcp mutants and the complemented strains were performed during the aerobic growth under H 2 O 2 stress (Figure 1; Figures S1 and S2). In agreement with previous findings [20], the ∆katA mutant was strongly impaired in growth after exposure to 10 mM H 2 O 2 and did not survive doses of 40 mM H 2 O 2 ( Figure 1A,E). The S. aureus COL wild type (WT) was able to grow with low doses of 1 mM H 2 O 2 and survived to 330 and 725% after 2 and 4 h, respectively ( Figure 1F; Figure S1B). However, the ∆katA mutant was hypersensitive to peroxide stress, since the growth was inhibited by 0.4 and 1 mM H 2 O 2 and only 32% and 4% of cells survived the 1 mM H 2 O 2 treatment after 2 and 4 h, respectively ( Figure 1F; Figure S1B). Complementation of the ∆katA mutant with pRB473-encoded katA could only partially restore the H 2 O 2 resistance to WT level after treatment with 0.4-1 mM H 2 O 2 ( Figure 1F; Figure S1A-C). This incomplete recovery of the WT resistance was due to lower catalase activity in the katA complemented strain, as confirmed using the diaminobenzidine gel staining method ( Figure S1D). only 32% and 4% of cells survived the 1 mM H2O2 treatment after 2 and 4 h, respectively ( Figure 1F; Figure S1B). Complementation of the ΔkatA mutant with pRB473-encoded katA could only partially restore the H2O2 resistance to WT level after treatment with 0.4-1 mM H2O2 ( Figure 1F; Figure S1A-C). This incomplete recovery of the WT resistance was due to lower catalase activity in the katA complemented strain, as confirmed using the diaminobenzidine gel staining method ( Figure S1D). aureus COL WT and ΔkatA mutant cells were exposed to 40 mM H2O2 during the log and stationary phases at OD500 of 0.5 and 2-3, respectively. In (E-G), the survival rates were calculated relative to the untreated control, which was set to 100%. Mean values and standard deviation (SD) from three to four biological replicates are shown. (H) Northern blot analyses of the katA, dps and ahpCF specific transcripts in the S. aureus WT, ΔkatA, ΔahpC and ΔperR mutants. The methylene blue stains below the Northern blot images denote the bands of the 16S and 23S rRNAs used as RNA loading controls. Quantification of the intensities of the Northern blot bands was performed from two biological and three technical replicates using Image J and is shown in the diagrams as fold changes (FCs) of induction of the katA, dps and ahpCF transcripts in the mutants relative to the WT. Error bars represent the SD. The statistics were analysed using Student's unpaired two-tailed t-test in Graph Prism. Symbols: ns p > 0.05, * p ≤ 0.05, ** p ≤ 0.01 and *** p ≤ 0.001.
In addition, the H2O2 resistance of WT cells was 1.5-4-fold further enhanced during the stationary phase, whereas the ΔkatA mutant did not survive 40 mM H2O2 during the log and stationary phases ( Figure 1G). These data indicate that KatA also contributes to the stationary-phase H2O2 resistance in S. aureus. In contrast, the peroxiredoxin-deficient S. aureus Δtpx and Δbcp mutants were not impaired in growth or survival after exposure to 10 and 40 mM H2O2 ( Figure 1C-E). However, the ΔahpC mutant displayed an increased H2O2 resistance ( Figure 1B,E), which was mediated by the PerR-dependent up-regulation (G) S. aureus COL WT and ∆katA mutant cells were exposed to 40 mM H 2 O 2 during the log and stationary phases at OD 500 of 0.5 and 2-3, respectively. In (E-G), the survival rates were calculated relative to the untreated control, which was set to 100%. Mean values and standard deviation (SD) from three to four biological replicates are shown. (H) Northern blot analyses of the katA, dps and ahpCF specific transcripts in the S. aureus WT, ∆katA, ∆ahpC and ∆perR mutants. The methylene blue stains below the Northern blot images denote the bands of the 16S and 23S rRNAs used as RNA loading controls. Quantification of the intensities of the Northern blot bands was performed from two biological and three technical replicates using Image J and is shown in the diagrams as fold changes (FCs) of induction of the katA, dps and ahpCF transcripts in the mutants relative to the WT. Error bars represent the SD. The statistics were analysed using Student's unpaired two-tailed t-test in Graph Prism. Symbols: ns p > 0.05, * p ≤ 0.05, ** p ≤ 0.01 and *** p ≤ 0.001.
In addition, the H 2 O 2 resistance of WT cells was 1.5-4-fold further enhanced during the stationary phase, whereas the ∆katA mutant did not survive 40 mM H 2 O 2 during the log and stationary phases ( Figure 1G). These data indicate that KatA also contributes to the stationary-phase H 2 O 2 resistance in S. aureus. In contrast, the peroxiredoxin-deficient S. aureus ∆tpx and ∆bcp mutants were not impaired in growth or survival after exposure to 10 and 40 mM H 2 O 2 ( Figure 1C-E). However, the ∆ahpC mutant displayed an increased H 2 O 2 resistance ( Figure 1B,E), which was mediated by the PerR-dependent up-regulation of KatA in the ∆ahpC mutant, as shown previously [20]. To validate the derepression of PerR regulon genes in the ∆ahpC mutant, we analysed the transcription of katA and the miniferritin-encoding dps gene, since both were strongly up-regulated under different thiol-stress conditions in the transcriptome of S. aureus WT cells [43,49,50]. The Northern blot results revealed the twofold up-regulation of katA, while dps transcription was induced at an 11-fold higher rate in the ∆ahpC mutant, supporting the derepression of both PerR regulon genes in the ∆ahpC mutant under non-stress conditions ( Figure 1H).
However, the H 2 O 2 -resistant phenotype of the ∆ahpC mutant could not be reversed to WT levels in growth and survival assays upon exposure to 10 and 40 mM H 2 O 2 in the ahpC complemented strain (Figures 1E and S2A). This lack of complementation might be caused by the lower plasmid-borne AhpC expression compared to the highly abundant AhpC in WT cells, as previously observed in the proteome [51]. The catalase activity was higher in the ∆ahpC mutant and ahpC complemented strain than in the WT, explaining the high H 2 O 2 resistance upon complementation ( Figure S2B). To confirm whether KatA and AhpC play additive roles in H 2 O 2 resistance, the growth and survival of the ∆ahpC∆katA double mutant was analysed. In agreement with previous data, the ∆ahpC∆katA mutant showed a slower aerobic growth ( Figure S3A) [20] and displayed 3-fold and 13-fold reduced survival rates after exposure to 1 mM H 2 O 2 for 2 h and 4 h, respectively, as compared to the ∆katA mutant ( Figure 1F).
Taken together, these results indicate that KatA plays the major role of conferring strong H 2 O 2 resistance to aerobic S. aureus cells during the log and the stationary phases, whereas AhpC makes a minor contribution to the H 2 O 2 resistance during the aerobic growth. Thus, KatA was identified as major determinant of the H 2 O 2 resistance in growing and non-growing S. aureus cells.

The ∆katA Mutant Shows a Strong Oxidative Shift in the E BSH after H 2 O 2 Stress and Is Impaired in Its Regeneration of the Reduced State, as Revealed by the Brx-roGFP2 Biosensor
To monitor the changes in the E BSH in the catalase-and peroxiredoxin-deficient mutants, we measured the Brx-roGFP2 biosensor responses after H 2 O 2 stress in the WT and mutant strains. Due to the strong aerobic H 2 O 2 resistance of S. aureus WT cells, the Brx-roGFP2 biosensor responded only weakly to 10 mM H 2 O 2 in our previous studies [17]. Thus, we first used 100 H 2 O 2 for WT cells, leading to fast biosensor oxidation and regeneration of the reduced state of E BSH within two hours, as in our previous studies ( Figure 2A) [17].
No increased biosensor oxidation was measured in WT cells after exposure to 1 mM H 2 O 2 ( Figure 2B). In contrast to WT cells, the biosensor was fully and constitutively oxidized by 1 and 100 mM H 2 O 2 in the ∆katA mutant, indicated by an impaired regeneration of reduced E BSH (Figure 2A,B). The katA mutant was only able to recover the reduced state of E BSH after treatment with 0.4 mM H 2 O 2 ( Figure 2C), suggesting that this low H 2 O 2 level might be detoxified by AhpC. In support of this hypothesis, the Brx-roGFP2 biosensor was quickly oxidized and strongly delayed in the recovery of reduced E BSH after exposure to 0.4 mM H 2 O 2 in the ∆ahpC∆katA double mutant ( Figure S3B). In contrast, the ∆ahpC, ∆tpx and ∆bcp mutants showed similar H 2 O 2 responses and regeneration of reduced E BSH compared to the WT ( Figure 2D-F). The biosensor results confirmed the hypersensitivities of the ∆katA and ∆ahpC∆katA mutants towards H 2 O 2 stress, supporting that KatA was responsible for the rapid detoxification of 100 mM of H 2 O 2 and regeneration of E BSH in S. aureus WT cells, while AhpC could only detoxify low levels of 0.4 mM H 2 O 2 in the absence of KatA. However, the peroxiredoxins Tpx and Bcp were not essential for H 2 O 2 detoxification in S. aureus WT cells. Brx-roGFP2 from plasmid pRB473. H2O2 injection assays were performed in microplates using the CLARIOstar microplate reader, as described previously [45]. The oxidation degrees (OxD) of the Brx-roGFP2 responses were calculated based on the 405/488 nm excitation ratios and normalized to fully reduced (DTT-treated) and fully oxidized (diamide-treated) controls, as described previously [45]. Mean values and SD of the OxD values are presented from three independent biological replicates.

S. aureus Shows KatA-Dependent Microaerophilic H2O2 Priming to Acquire an Improved Resistance towards Lethal H2O2 Doses
The previous data revealed that KatA confers strong H2O2 resistance during aerobic growth. However, the role of KatA in the priming of S. aureus for improved H2O2 resistance during microaerophilic conditions was not investigated. Thus, the S. aureus WT and the ΔkatA mutant were grown under microaerophilic conditions to the log phase and primed with 0.1 mM H2O2 for 30 min, followed by triggering with 10 mM H2O2 ( Figure  3A). The growth and survival were analysed in naïve (C), primed (P), primed and triggered (PT) and triggered bacteria (T) ( Figure 3A).
The primed S. aureus WT and ΔkatA mutant (P) were not impaired in growth and survival under 0.1 mM H2O2 ( Figure 3B-E). However, the primed and triggered WT (PT) could acquire an improved resistance towards the triggering stimulus of 10 mM H2O2 compared to the triggering-only state (T) ( Figure 3B,D). Specifically, PT bacteria showed survival rates of 52-72%, whereas T bacteria were almost killed and survived only to <0.07% after 10 mM H2O2 treatment. This indicates that S. aureus is primable for improved oxidative stress resistance during microaerophilic growth ( Figure 3B,D). However, in contrast to the WT, the primed ΔkatA mutant strain was unable to acquire the improved resistance towards otherwise lethal doses of 10 mM H2O2 under microaerophilic conditions ( Figure 3C,E). Both PT and T bacteria of the ΔkatA mutant were strongly impaired in injection assays were performed in microplates using the CLARIOstar microplate reader, as described previously [45]. The oxidation degrees (OxD) of the Brx-roGFP2 responses were calculated based on the 405/488 nm excitation ratios and normalized to fully reduced (DTT-treated) and fully oxidized (diamide-treated) controls, as described previously [45]. Mean values and SD of the OxD values are presented from three independent biological replicates.

S. aureus Shows KatA-Dependent Microaerophilic H 2 O 2 Priming to Acquire an Improved Resistance towards Lethal H 2 O 2 Doses
The previous data revealed that KatA confers strong H 2 O 2 resistance during aerobic growth. However, the role of KatA in the priming of S. aureus for improved H 2 O 2 resistance during microaerophilic conditions was not investigated. Thus, the S. aureus WT and the ∆katA mutant were grown under microaerophilic conditions to the log phase and primed with 0.1 mM H 2 O 2 for 30 min, followed by triggering with 10 mM H 2 O 2 ( Figure 3A). The growth and survival were analysed in naïve (C), primed (P), primed and triggered (PT) and triggered bacteria (T) ( Figure 3A).
The primed S. aureus WT and ∆katA mutant (P) were not impaired in growth and survival under 0.1 mM H 2 O 2 ( Figure 3B-E). However, the primed and triggered WT (PT) could acquire an improved resistance towards the triggering stimulus of 10 mM H 2 O 2 compared to the triggering-only state (T) ( Figure 3B,D). Specifically, PT bacteria showed survival rates of 52-72%, whereas T bacteria were almost killed and survived only to <0.07% after 10 mM H 2 O 2 treatment. This indicates that S. aureus is primable for improved oxidative stress resistance during microaerophilic growth ( Figure 3B,D). However, in contrast to the WT, the primed ∆katA mutant strain was unable to acquire the improved resistance towards otherwise lethal doses of 10 mM H 2 O 2 under microaerophilic conditions ( Figure 3C,E). Both PT and T bacteria of the ∆katA mutant were strongly impaired in growth and completely killed after treatment with 10 mM H 2 O 2 as a triggering stimulus ( Figure 3C,E).
However, due to its lower catalase activity from plasmid-based KatA expression ( Figure S1D), the katA complemented strain did not recover the improved H 2 O 2 resistance upon microaerophilic priming, resulting in growth inhibition and killing of PT and T bacteria ( Figure S4). Overall, these results indicate that KatA is responsible for S. aureus priming for improved resistance against upcoming lethal H 2 O 2 stress under microaerophilic conditions. Thus, KatA confers the constitutive resistance during aerobic growth and prepares S. aureus for future oxidative stress under microaerophilic conditions. growth and completely killed after treatment with 10 mM H2O2 as a triggering stimulus ( Figure 3C,E). However, due to its lower catalase activity from plasmid-based KatA expression (Figure S1D), the katA complemented strain did not recover the improved H2O2 resistance upon microaerophilic priming, resulting in growth inhibition and killing of PT and T bacteria ( Figure S4). Overall, these results indicate that KatA is responsible for S. aureus priming for improved resistance against upcoming lethal H2O2 stress under microaerophilic conditions. Thus, KatA confers the constitutive resistance during aerobic growth and prepares S. aureus for future oxidative stress under microaerophilic conditions. aureus WT and ΔkatA mutant strains were grown microaerophilically and primed during the log phase with 0.1 mM H2O2 for ~30 min (P), followed by treatment with 10 mM H2O2 as a triggering stimulus (PT). The growth curves (B,C) and survival rates (D,E) were measured in naïve (C), primed (P), primed and triggered (PT) and triggered-only bacteria (T). The survival rates were calculated after 2 and 4 h of H2O2 stress relative to untreated control cells. The results are from three to four biological replicates. Error bars represent the SD. The statistics were calculated using Student's unpaired two-tailed t-test in Graph Prism. Symbols: ns p > 0.05, * p < 0.05,** p ≤ 0.01 and *** p ≤ 0.001.

S. aureus Is Not Primable for Improved H2O2 Resistance during Aerobic Growth
Next, priming and triggering experiments were performed in aerobically grown S. aureus cells to analyse whether the constitutive H2O2 resistance could be further enhanced in primed cells (Figure 4). First, we used the same H2O2 doses for the priming (0.1 mM) and triggering (10 mM) experiments as applied in the microaerophilic experiments (Figure 4A). As expected, there were no differences in growth and survival between PT and T bacteria after exposure to 10 mM H2O2 during the aerobic growth ( Figure 4B,C). Both PT and T bacteria were similarly resistant and fully survived the 10 mM H2O2 triggering dose. Thus, the constitutive resistance of aerobic S. aureus cells towards 10 mM H2O2 could not be further enhanced by pre-exposure to the priming stimulus of 0.1 mM H2O2 ( Figure  4B,C). As shown before, this constitutive H2O2 resistance of aerobically grown S. aureus cells was dependent on KatA ( Figure 1A,E,F).
We further increased the H2O2 doses for priming (1 mM H2O2) and triggering (40 mM) of S. aureus during the aerobic growth ( Figure 4D). However, higher priming doses also could not improve the growth and survival of PT bacteria in response to the subse-

S. aureus Is Not Primable for Improved H 2 O 2 Resistance during Aerobic Growth
Next, priming and triggering experiments were performed in aerobically grown S. aureus cells to analyse whether the constitutive H 2 O 2 resistance could be further enhanced in primed cells (Figure 4). First, we used the same H 2 O 2 doses for the priming (0.1 mM) and triggering (10 mM) experiments as applied in the microaerophilic experiments ( Figure 4A). As expected, there were no differences in growth and survival between PT and T bacteria after exposure to 10 mM H 2 O 2 during the aerobic growth ( Figure 4B,C). Both PT and T bacteria were similarly resistant and fully survived the 10 mM H 2 O 2 triggering dose. Thus, the constitutive resistance of aerobic S. aureus cells towards 10 mM H 2 O 2 could not be further enhanced by pre-exposure to the priming stimulus of 0.1 mM H 2 O 2 ( Figure 4B,C). As shown before, this constitutive H 2 O 2 resistance of aerobically grown S. aureus cells was dependent on KatA ( Figure 1A,E,F).
We further increased the H 2 O 2 doses for priming (1 mM H 2 O 2 ) and triggering (40 mM) of S. aureus during the aerobic growth ( Figure 4D). However, higher priming doses also could not improve the growth and survival of PT bacteria in response to the subsequent 40 mM H 2 O 2 stress compared to the T bacteria treated with 40 mM H 2 O 2 only ( Figure 4E,F).
Both PT and T bacteria were strongly impaired in growth under 40 mM H 2 O 2 stress and showed survival rates of <10% after 2 h. Small survival differences of 4% were observed after 4 h in T versus PT bacteria but not at the 2 h time point ( Figure 4F). In conclusion, these different priming setups support that S. aureus is not primable for enhanced H 2 O 2 resistance under aerobic conditions. We hypothesize that ROS production during aerobic respiration acts as a priming stimulus to induce KatA, which confers the high H 2 O 2 resistance.
Antioxidants 2022, 11, x FOR PEER REVIEW 9 of 20 quent 40 mM H2O2 stress compared to the T bacteria treated with 40 mM H2O2 only (Figure 4E,F). Both PT and T bacteria were strongly impaired in growth under 40 mM H2O2 stress and showed survival rates of <10% after 2 h. Small survival differences of 4% were observed after 4 h in T versus PT bacteria but not at the 2 h time point ( Figure 4F). In conclusion, these different priming setups support that S. aureus is not primable for enhanced H2O2 resistance under aerobic conditions. We hypothesize that ROS production during aerobic respiration acts as a priming stimulus to induce KatA, which confers the high H2O2 resistance. , respectively, as triggering stimuli (PT). The growth curves (B,E) and survival rates (C,F) were measured for naïve (C), primed (P), primed and triggered (PT) and triggered-only bacteria (T). Survival rates of the H2O2-treated cells were calculated relative to the untreated control. The results are from three to four biological replicates. Error bars represent the SD. The statistics were calculated using Student's unpaired two-tailed t-test in Graph Prism. Symbols: ns p > 0.05 and * p < 0.05.

Microaerophilic H2O2 Priming Causes Increased Transcription of KatA and Elevated KatA Activity, which Confers Improved Resistance towards Lethal H2O2 Doses in S. aureus
Northern blot analyses were used to study whether microaerophilic H2O2 priming induces transcription of the PerR-dependent katA gene and the ahpCF operon in S. aureus ( Figure 5). The results revealed that transcription of katA was significantly up-regulated by 1.8-fold upon microaerophilic priming with 0.1 mM H2O2 ( Figure 5B,D), whereas the basal level of katA transcription was already higher under aerobic conditions and could not be further induced during aerobic H2O2 priming ( Figure 5C,E). Transcription of the ahpCF operon was not significantly induced during microaerophilic priming ( Figure  5B,G). However, transcription of katA and ahpCF was strongly reduced after triggering by 10 mM H2O2 under microaerophilic conditions, since the triggering dose was lethal (Figure 5B,D,G). Transcription of katA decreased even in the PT bacteria compared to P cells, which highlights the high efficiency of the KatA protein for fast removal of H2O2 in PT bacteria ( Figure 5B,D). Under aerobic conditions, PT and T bacteria did not show significantly enhanced transcription of katA and ahpCF, supporting that the constitutive H2O2 resistance could not be further enhanced by pre-exposure to the priming dose ( Figure  5C,E,H). . The growth curves (B,E) and survival rates (C,F) were measured for naïve (C), primed (P), primed and triggered (PT) and triggered-only bacteria (T). Survival rates of the H 2 O 2 -treated cells were calculated relative to the untreated control. The results are from three to four biological replicates. Error bars represent the SD. The statistics were calculated using Student's unpaired two-tailed t-test in Graph Prism. Symbols: ns p > 0.05 and * p < 0.05.

Microaerophilic H 2 O 2 Priming Causes Increased Transcription of KatA and Elevated KatA Activity, which Confers Improved Resistance towards Lethal H 2 O 2 Doses in S. aureus
Northern blot analyses were used to study whether microaerophilic H 2 O 2 priming induces transcription of the PerR-dependent katA gene and the ahpCF operon in S. aureus ( Figure 5). The results revealed that transcription of katA was significantly up-regulated by 1.8-fold upon microaerophilic priming with 0.1 mM H 2 O 2 ( Figure 5B,D), whereas the basal level of katA transcription was already higher under aerobic conditions and could not be further induced during aerobic H 2 O 2 priming ( Figure 5C,E). Transcription of the ahpCF operon was not significantly induced during microaerophilic priming ( Figure 5B,G). However, transcription of katA and ahpCF was strongly reduced after triggering by 10 mM H 2 O 2 under microaerophilic conditions, since the triggering dose was lethal (Figure 5B,D,G). Transcription of katA decreased even in the PT bacteria compared to P cells, which highlights the high efficiency of the KatA protein for fast removal of H 2 O 2 in PT bacteria ( Figure 5B,D). Under aerobic conditions, PT and T bacteria did not show significantly enhanced transcription of katA and ahpCF, supporting that the constitutive H 2 O 2 resistance could not be further enhanced by pre-exposure to the priming dose ( Figure 5C,E,H).
The katA transcript levels could be confirmed by catalase activities during the microaerophilic and aerobic priming experiments ( Figure 5F). Specifically, the basal activity of KatA was very low during the microaerophilic growth, whereas aerobic cells showed a constitutive high catalase activity. The KatA activity could be only enhanced upon microaerophilic priming but not during aerobic priming due to the high constitutive resistance ( Figure 5F). Further consistent with the Northern blots, the KatA activity decreased strongly in PT and T bacteria during microaerophilic priming. Together, these results reveal that microaerophilic priming induces katA transcription and KatA activity, which confers improved resistance against otherwise lethal H 2 O 2 stress.
Antioxidants 2022, 11, x FOR PEER REVIEW 10 of 20 The katA transcript levels could be confirmed by catalase activities during the microaerophilic and aerobic priming experiments ( Figure 5F). Specifically, the basal activity of KatA was very low during the microaerophilic growth, whereas aerobic cells showed a constitutive high catalase activity. The KatA activity could be only enhanced upon microaerophilic priming but not during aerobic priming due to the high constitutive resistance ( Figure 5F). Further consistent with the Northern blots, the KatA activity decreased strongly in PT and T bacteria during microaerophilic priming. Together, these results reveal that microaerophilic priming induces katA transcription and KatA activity, which confers improved resistance against otherwise lethal H2O2 stress. (B,C) To analyze transcription of katA and ahpCF using Northern blots, RNA was isolated from S. aureus WT cells in the naïve (C1, C2), primed (P), primed and triggered (PT) and triggered (T) states. The band intensities of the katA (D,E) and ahpCF specific transcripts (G,H) were quantified from two biological replicates using ImageJ. The transcriptional induction of katA and ahpCF was calculated as fold change relative to the control C1, which was set to 1. Error bars represent the SD. The statistics were calculated using ordinary one-way ANOVA and Dunnet's multiple comparisons test in Graph Prism. Symbols: ns p > 0.05; * p ≤ 0.05; ** p ≤ 0.01 and *** p≤ 0.001. (F) The catalase activity was analysed in cell extracts of the S. aureus WT during microaerophilic and aerobic H2O2 priming for C2, P, PT and T states using native PAGE and diaminobenzidine staining. The catalase activity assays were performed in two biological and two technical replicates. (I,J) The response of the Brx-roGFP2 biosensor was measured in S. aureus COL grown in LB medium under microaerophilic and aerobic conditions in naïve (C1, C2), primed (P), primed and triggered (PT) and triggered (T) cells. The C1 and C2 samples were harvested at OD500 of 0.3 and 0.4, respectively, and the P, PT and T bacteria were harvested after 10 min of H2O2 exposure (maximum biosensor oxidation). Samples were blocked with 10 mM NEM and the fluorescence excitation maxima were measured at 405 and 488 nm using the microplate reader. OxD values and the EBSH of Brx-roGFP2 were calculated using the 405/408 nm excitation ratio, as described in the Materials and Methods section. Mean values of three biological replicates are shown, error bars represent the SD and p-values were calculated using Student's unpaired two-tailed t-test in Graph Prism software. Symbols: ns p > 0.05; * p ≤0.05; ** p ≤ 0.01; ***p ≤ 0.001 and ***p ≤ 0.0001. (B,C) To analyze transcription of katA and ahpCF using Northern blots, RNA was isolated from S. aureus WT cells in the naïve (C1, C2), primed (P), primed and triggered (PT) and triggered (T) states. The band intensities of the katA (D,E) and ahpCF specific transcripts (G,H) were quantified from two biological replicates using ImageJ. The transcriptional induction of katA and ahpCF was calculated as fold change relative to the control C1, which was set to 1. Error bars represent the SD. The statistics were calculated using ordinary one-way ANOVA and Dunnet's multiple comparisons test in Graph Prism. Symbols: ns p > 0.05; * p ≤ 0.05; ** p ≤ 0.01 and *** p≤ 0.001. (F) The catalase activity was analysed in cell extracts of the S. aureus WT during microaerophilic and aerobic H 2 O 2 priming for C2, P, PT and T states using native PAGE and diaminobenzidine staining. The catalase activity assays were performed in two biological and two technical replicates. (I,J) The response of the Brx-roGFP2 biosensor was measured in S. aureus COL grown in LB medium under microaerophilic and aerobic conditions in naïve (C1, C2), primed (P), primed and triggered (PT) and triggered (T) cells. The C1 and C2 samples were harvested at OD 500 of 0.3 and 0.4, respectively, and the P, PT and T bacteria were harvested after 10 min of H 2 O 2 exposure (maximum biosensor oxidation). Samples were blocked with 10 mM NEM and the fluorescence excitation maxima were measured at 405 and 488 nm using the microplate reader. OxD values and the E BSH of Brx-roGFP2 were calculated using the 405/408 nm excitation ratio, as described in the Section 2. Mean values of three biological replicates are shown, error bars represent the SD and p-values were calculated using Student's unpaired two-tailed t-test in Graph Prism software. Symbols: ns p > 0.05; * p ≤0.05; ** p ≤ 0.01; *** p ≤ 0.001 and **** p ≤ 0.0001.

Microaerophilic H 2 O 2 Priming Leads to a Strong Oxidative Shift in the E BSH
We were interested in the E BSH differences between aerobic and microaerophilic growth conditions, supporting the enhanced ROS levels in aerobic cells as endogenous priming stimuli. Moreover, we aimed to analyse the Brx-roGFP2 biosensor response upon microaerophilic priming and triggering to investigate if the KatA induction upon microaerophilic priming is accompanied by a change in the Brx-roGFP2 biosensor oxidation. The comparison of the basal biosensor oxidation revealed a more reducing basal OxD of 0.1 and an E BSH of −310 mV during microaerophilic growth (C1) compared to the OxD of 0.3 and the E BSH of −291 mV during aerobic growth (C1) (Figure 5I,J). Thus, the higher basal oxidation in aerobic cells accounted for the increased ROS level due to aerobic respiration. Upon microaerophilic priming, the Brx-roGFP2 biosensor showed a fivefold increased OxD and an oxidized E BSH of −281 mV, which was further oxidized in PT and T bacteria. In contrast, the biosensor did not respond to aerobic priming and showed increased oxidation only in aerobic PT and T bacteria ( Figure 5I,J). These results support that microaerophilic priming leads to an oxidative shift in the E BSH from −310 mV to −281 mV, resulting in increased catalase expression that primes the cells for improved H 2 O 2 resistance. Due to the higher basal oxidation in aerobic cells, priming did not change the high E BSH of −291 mV, which is consistent with the constitutive KatA expression ( Figure 5I,J). These results on the E BSH differences of~20 mV between microaerophilic and aerobic cells strongly support that respiratory H 2 O 2 primes aerobic cells for constitutive H 2 O 2 resistance.

The Peroxiredoxins AhpC, Tpx and Bcp Mediate CHP Resistance in S. aureus
While the role of KatA in aerobic H 2 O 2 resistance and microaerophilic H 2 O 2 priming was clearly revealed, the peroxiredoxins AhpC, Tpx and Bcp could also function in the resistance to organic hydroperoxides in S. aureus. Using growth and survival assays, the phenotypes of the catalase and peroxiredoxin-deficient mutants were analysed after CHP treatment ( Figure 6). While the growth of the ∆katA and ∆bcp mutants was not affected by 0.15 mM CHP stress, the CHP-treated ∆ahpC and ∆tpx mutants showed slightly reduced growth rates ( Figure 6A-D), which could be restored to WT levels upon complementation ( Figure 6E,F). In addition, the ∆ahpC, ∆tpx and ∆bcp mutants showed significantly decreased survival rates after 4 h of CHP stress, supporting that the peroxiredoxins confer protection against CHP stress in S. aureus ( Figure 6G). These CHP-sensitive survival phenotypes of the peroxiredoxin-deficient mutants could be restored to WT levels in the ahpC, tpx and bcp complemented strains ( Figure 6H). However, the slightly increased CHP resistance of the ∆katA mutant could not be reverted to WT levels in the katA complemented strain ( Figure 6G), probably due to the partial complementation by plasmid-based KatA expression ( Figure S1D).

Peroxiredoxin-Deficient ΔahpC, Δtpx and Δbcp Mutants Are Delayed in CHP Detoxification as Revealed by Brx-roGFP2 Measurements
To investigate the impact of the peroxiredoxins in the maintenance of the reduced state of EBSH in S. aureus, Brx-roGFP2 biosensor measurements were performed after 0.5 mM CHP stress (Figure 7). The Brx-roGFP2 biosensor was similarly quickly oxidized by 0.5 mM CHP in the WT, ΔkatA, ΔahpC, Δtpx and Δbcp mutants. However, while the WT and ΔkatA mutant could regenerate the reduced state of EBSH within 2 h, the ΔahpC mutant was unable to regenerate the basal level of EBSH during the recovery phase from CHP stress ( Figure 7A,B). In addition, both the Δbcp and Δtpx mutants showed significant delays in recovery of the reduced state of EBSH upon CHP stress as compared to the WT ( Figure  7C,D). These biosensor results support that the peroxiredoxins AhpC, Tpx and Bcp are important for CHP detoxification and contribute to regeneration of EBSH during the recovery phase from CHP stress in S. aureus. In contrast, KatA does not contribute to CHP detoxification and resistance in S. aureus. aureus COL WT, ∆katA, ∆ahpC, ∆tpx and ∆bcp mutants (G) and the katA (G), ahpC, tpx and bcp (H) complemented strains were determined at 2 and 4 h after exposure to 0.25 mM CHP relative to the untreated control, which was set to 100%. Mean values and SD of four to six biological replicates are presented. The statistics were calculated using Student's unpaired two-tailed t-test in Graph Prism. Symbols: ns p > 0.05, * p ≤ 0.05, ** p ≤ 0.01 and *** p ≤ 0.001.

Peroxiredoxin-Deficient ∆ahpC, ∆tpx and ∆bcp Mutants Are Delayed in CHP Detoxification as Revealed by Brx-roGFP2 Measurements
To investigate the impact of the peroxiredoxins in the maintenance of the reduced state of E BSH in S. aureus, Brx-roGFP2 biosensor measurements were performed after 0.5 mM CHP stress (Figure 7). The Brx-roGFP2 biosensor was similarly quickly oxidized by 0.5 mM CHP in the WT, ∆katA, ∆ahpC, ∆tpx and ∆bcp mutants. However, while the WT and ∆katA mutant could regenerate the reduced state of E BSH within 2 h, the ∆ahpC mutant was unable to regenerate the basal level of E BSH during the recovery phase from CHP stress ( Figure 7A,B). In addition, both the ∆bcp and ∆tpx mutants showed significant delays in recovery of the reduced state of E BSH upon CHP stress as compared to the WT ( Figure 7C,D). These biosensor results support that the peroxiredoxins AhpC, Tpx and Bcp are important for CHP detoxification and contribute to regeneration of E BSH during the recovery phase from CHP stress in S. aureus. In contrast, KatA does not contribute to CHP detoxification and resistance in S. aureus. , Δtpx (C) and Δbcp mutants expressing Brx-roGFP2 from plasmid pRB473. H2O2 injection assays were performed in microplates and the biosensor responses measured using the CLARIOstar microplate reader, as described previously [45]. (D) The oxidation degree (OxD) of the Brx-roGFP2 response was calculated based on the 405/488 nm excitation ratio and normalized to fully reduced (DTT-treated) and oxidized (diamidetreated) controls. Mean values and SD of three biological replicates are shown. The statistics were calculated using Student's unpaired two-tailed t-test in Graph Prism. Symbols: ns p > 0.05, * p ≤ 0.05, ** p ≤ 0.01 and *** p ≤ 0.001.

KatA and Peroxiredoxins Do Not Contribute to Protection against HOCl Stress
Previous transcriptome analyses revealed an increased transcription of the PerR regulon under HOCl stress in S. aureus [38]. Thus, growth curves and survival assays were used to analyse the phenotypes of the ΔkatA, ΔahpC, Δtpx and Δbcp mutants under HOCl stress. However, none of these mutants showed significant defects in growth or survival after HOCl stress compared to the WT, indicating that the catalase and peroxiredoxins do not contribute to HOCl detoxification and resistance in S. aureus ( Figure S5A-E).

Discussion
Here, we investigated the roles of catalase and peroxiredoxins in the peroxide stress resistance and priming of S. aureus during aerobic and microaerophilic growth. Our results revealed that S. aureus is not primable towards improved H2O2 resistance during aerobic growth due to its constitutive H2O2 resistance, which was shown to be dependent on the catalase KatA. Moreover, we found that microaerophilic S. aureus cells can be primed to acquire an enhanced resistance towards lethal H2O2 doses, which was also mediated by KatA ( Figure 3B,D). In addition, the peroxiredoxins AhpC, Tpx and Bcp were shown to contribute to the CHP detoxification and resistance in S. aureus to ensure the maintenance of the redox balance upon recovery from stress. , ∆tpx (C) and ∆bcp mutants expressing Brx-roGFP2 from plasmid pRB473. H 2 O 2 injection assays were performed in microplates and the biosensor responses measured using the CLARIOstar microplate reader, as described previously [45]. (D) The oxidation degree (OxD) of the Brx-roGFP2 response was calculated based on the 405/488 nm excitation ratio and normalized to fully reduced (DTT-treated) and oxidized (diamide-treated) controls. Mean values and SD of three biological replicates are shown. The statistics were calculated using Student's unpaired two-tailed t-test in Graph Prism. Symbols: ns p > 0.05, * p ≤ 0.05, ** p ≤ 0.01 and *** p ≤ 0.001.

KatA and Peroxiredoxins Do Not Contribute to Protection against HOCl Stress
Previous transcriptome analyses revealed an increased transcription of the PerR regulon under HOCl stress in S. aureus [38]. Thus, growth curves and survival assays were used to analyse the phenotypes of the ∆katA, ∆ahpC, ∆tpx and ∆bcp mutants under HOCl stress. However, none of these mutants showed significant defects in growth or survival after HOCl stress compared to the WT, indicating that the catalase and peroxiredoxins do not contribute to HOCl detoxification and resistance in S. aureus ( Figure S5A-E).

Discussion
Here, we investigated the roles of catalase and peroxiredoxins in the peroxide stress resistance and priming of S. aureus during aerobic and microaerophilic growth. Our results revealed that S. aureus is not primable towards improved H 2 O 2 resistance during aerobic growth due to its constitutive H 2 O 2 resistance, which was shown to be dependent on the catalase KatA. Moreover, we found that microaerophilic S. aureus cells can be primed to acquire an enhanced resistance towards lethal H 2 O 2 doses, which was also mediated by KatA ( Figure 3B,D). In addition, the peroxiredoxins AhpC, Tpx and Bcp were shown to contribute to the CHP detoxification and resistance in S. aureus to ensure the maintenance of the redox balance upon recovery from stress.
The roles of KatA and AhpC in the peroxide resistance of S. aureus have been previously demonstrated [20]. In this work, we additionally used Brx-roGFP2 biosensor measurements, showing the impact of KatA and AhpC on the level of H 2 O 2 detoxification and regeneration of reduced E BSH under oxidative stress. Without KatA, S. aureus cells are highly sensitive to oxidants and only able to remove low doses of 0.4 mM H 2 O 2 , which were shown to be detoxified by AhpC. In addition, the ∆ahpC∆katA double mutant showed an increased sensitivity towards H 2 O 2 stress in survival assays as compared to the ∆katA mutant. These data confirm that KatA and AhpCF have compensatory roles in H 2 O 2 resistance to ensure the survival of S. aureus under oxidative stress [20]. Expression of katA, bcp, dps and the ahpCF operon is controlled by the peroxide-sensing PerR repressor, which is inactivated by H 2 O 2 due to Fe 2+ -catalysed histidine oxidation in S. aureus [18,52,53]. Due to PerR derepression in the ∆ahpC mutant, katA and dps expression was elevated, as confirmed here using Northern blots and shown previously [20]. Thus, the higher KatA expression level in the ∆ahpC mutant mediates the enhanced H 2 O 2 resistance, confirming previous results in S. aureus and B. subtilis [20,54,55].
In addition, we showed that aerobically grown S. aureus acquire an improved resistance to H 2 O 2 during the stationary phase, which also depends on KatA. The oxidative stress resistance was also enhanced during the stationary phase in other bacteria, such as B. subtilis and E. coli [56,57]. In addition, KatA activity was elevated during the stationary phase in S. aureus and B. subtilis [20,48]. Altogether, our results on aerobic S. aureus cells reveal that KatA is the major player that confers the strong constitutive H 2 O 2 resistance during the log and stationary phases, while the peroxiredoxin AhpC plays an additional role of scavenging 0.4 mM H 2 O 2 in the absence of KatA. Apart from its major role as an H 2 O 2 scavenger, the catalase also provides heme and iron as cofactors for cellular metabolism, which could contribute to S. aureus survival under oxidative stress.
We further showed that S. aureus is primable with sub-lethal doses of 0.1 mM H 2 O 2 to acquire an improved resistance towards otherwise lethal doses of 10 mM H 2 O 2 during the microaerophilic growth. This microaerophilic H 2 O 2 priming was found to be dependent on KatA, which was transcriptionally induced and showed a higher catalase activity upon challenge with the priming dose. These results are consistent with previous data showing increased KatA activity by exposure to 100 µM H 2 O 2 during oxygen limitation [52]. We showed that microaerophilic priming by KatA prepares the cells to better survive the lethal triggering stress, resulting in improved growth and survival of S. aureus. In contrast, due to their constitutive H 2 O 2 resistance, aerobic S. aureus cells were not primable to acquire higher resistance towards lethal H 2 O 2 doses. Aerobic priming was not possible with 0.1 mM or 1 mM H 2 O 2 since katA transcription and catalase activity were already elevated in naïve cells and could not be further increased in primed cells. The higher basal level of E BSH of −291 mV in aerobic cells compared to the more reducing basal E BSH of −310 mV in microaerophilic cells strongly indicates an increased ROS level in aerobic cells due to aerobic respiration. These biosensor data support that S. aureus is primed by endogenous ROS generated by aerobic respiration to achieve their constitutive H 2 O 2 resistance phenotype ( Figure 8). Thus, the PerR regulon is already up-regulated during aerobic growth due to ROS generated during aerobic respiration [52], resulting in the H 2 O 2 -resistant phenotype.
This up-regulation of the PerR regulon in S. aureus during the aerobic growth was caused by the hypersensitive PerR repressor, which is poised by very low endogenous levels of H 2 O 2 generated during aerobic respiration [52]. The endogenous H 2 O 2 concentration in aerobic E. coli cells was determined as~50 nM [58]. While PerR of S. aureus is hypersensitive to endogenous H 2 O 2 levels during aerobic growth, the PerR protein of B. subtilis is less sensitive and cannot sense such low H 2 O 2 levels originating from aerobic respiration [52]. Thus, B. subtilis can be primed during aerobic growth with 0.1 mM H 2 O 2 , leading to PerR inactivation and derepression of the PerR-controlled KatA, which confers an adaptive resistance against the otherwise lethal triggering dose of 10 mM H 2 O 2 [33,34,52,55,59].   [13,61,62]. In mammalian cells, the physiological intracellular H 2 O 2 concentration was estimated in the range of 1-100 nM, while the extracellular H 2 O 2 was 100-fold higher [63]. Thus, S. aureus might not experience such high doses of 10-100 mM H 2 O 2 during interaction with immune cells. However, as a commensal bacterium, S. aureus colonizes the anterior nares and the nasopharynx together with competing microbes, such as Streptococcus pneumoniae, which generates millimolar levels of H 2 O 2 by the lactate and pyruvate oxidases to kill competitive bacteria [64][65][66][67][68][69]. In a mouse infection model, KatA was required for nasal colonization of S. aureus [18]. Furthermore, the katA mutant was impaired in survival towards H 2 O 2 produced by S. pneumoniae during nasal colonization [20,70]. Thus, microaerophilic H 2 O 2 priming of S. aureus might provide an advantage in its ecological niche to resist the high H 2 O 2 levels generated by competing microbes.
In contrast, KatA and the peroxiredoxins are not directly involved in the defence against HOCl stress in S. aureus. However, KatA might contribute to lowering external H 2 O 2 and, subsequently, HOCl levels in the neutrophil phagosome. In support of this notion, KatA was found to be induced upon macrophage infection and to be essential for survival of S. aureus inside macrophages [21]. In addition, S. aureus strains with high catalase activity were more resistant to killing by neutrophils compared to strains with lower catalase activity [22]. These data support that KatA is an important defence mechanism in S. aureus against the respiratory burst of macrophages and neutrophils.
In addition, the peroxiredoxins AhpC, Tpx and Bcp were found to confer protection against CHP stress in S. aureus cells. Using Brx-roGFP2 biosensor measurements, the peroxiredoxin-deficient ∆ahpC, ∆tpx and ∆bcp mutants showed delayed regeneration of the reduced state of E BSH after recovery from CHP stress. Thus, AhpC, Tpx and Bcp function in CHP detoxification and contribute to the maintenance of the cellular redox balance. The role of AhpC in CHP resistance has been previously demonstrated in S. aureus [20]. Similarly, the AhpC homologs of B. subtilis, E. coli and S. Typhimurium conferred resistance towards CHP stress [55,71]. In E. coli, the ∆tpx and ∆bcp mutants were more sensitive towards various OHPs, indicating that these peroxiredoxins are more specific to reduction of organic peroxide substrates [26,28]. Kinetic assays of the E. coli Tpx protein demonstrated the substrate specificity towards alkyl hydroperoxides over H 2 O 2 [27]. Similarly, Bcp of E. coli has a fivefold higher V max /K m value for linoleic acid hydroperoxide as a substrate compared to H 2 O 2 [26].

Conclusions
Taken together, we have shown that the catalase KatA is the major player in aerobic H 2 O 2 resistance in S. aureus and mediates priming to endogenous ROS levels generated during aerobic respiration to confer the constitutive H 2 O 2 resistance towards the triggering stimulus in aerobic cells. In addition, KatA mediates microaerophilic priming by low H 2 O 2 levels to prepare S. aureus cells for improved and adaptive resistance against otherwise lethal H 2 O 2 doses. Furthermore, the peroxiredoxins AhpC, Tpx and Bcp were shown to contribute to CHP detoxification and resistance to ensure the survival and regeneration of the reduced E BSH in S. aureus (Figure 8). In future studies, we aim to elucidate the functions of KatA and the peroxiredoxins in signal transduction, as redox-active chaperones and in cellular metabolism during aerobic growth and under oxidative stress in S. aureus.

Supplementary Materials:
The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/antiox11091793/s1: Figure S1: Complementation of KatA increases H 2 O 2 resistance in S. aureus, but the catalase activity in the katA complemented strain is lower compared to the WT. Figure S2: Growth and catalase activity of the ahpC complemented strain. Figure S3: The ∆katA∆ahpC double mutant is hypersensitive towards H 2 O 2 and delayed in the recovery of reduced E BSH after 0.4 mM H 2 O 2 . Figure S4: The katA complemented strain is not primable towards increased H 2 O 2 resistance during microaerophilic growth. Figure S5: KatA and the peroxiredoxins AhpC, Tpx and Bcp are not involved in HOCl detoxification in S. aureus. Figure S6: The Northern blot images using the katA specific RNA probe show the katA transcripts (images of Figure 1H; two bioreplicates, three technical replicates). Figure S7: The Northern blot images using the ahpC specific RNA probe show the ahpCF operon transcripts (images of Figure 1H; two bioreplicates, three technical replicates). Figure S8: The Northern blot images using the dps specific RNA probe show the dps operon transcripts (images of Figure 1H; two bioreplicates, three technical replicates). Figure S9: The Northern blot images of priming experiments using the katA specific RNA probe show the katA transcripts (images of Figure 5B,C; three bioreplicates). Figure S10: The Northern blot images of priming experiments using the ahpC specific RNA probe show the ahpCF transcripts (images of Figure 5B,C; three bioreplicates). Figure S11: Native gels of catalase stains used for KatA activity assays (images of Figure 5F; two bioreplicates, two technical replicates). Figure S12: Native gels of catalase stains used for KatA activity assays (images of Figure S1D; three bioreplicates). Figure S13: Native gels of catalase stains used for KatA activity assays (images of Figure S2B; three bioreplicates).