Staphylococcus aureus SrrAB Affects Susceptibility to Hydrogen Peroxide and Co-Existence with Streptococcus sanguinis

Staphylococcus aureus is a pathogen and a commensal bacterial species that is found in humans. Bacterial two-component systems (TCSs) sense and respond to environmental stresses, which include antimicrobial agents produced by other bacteria. In this study, we analyzed the relation between the TCS SrrAB and susceptibility to the hydrogen peroxide (H2O2) that is produced by Streptococcus sanguinis, which is a commensal oral streptococcus. An srrA-inactivated S. aureus mutant demonstrated low susceptibility to the H2O2 produced by S. sanguinis. We investigated the expression of anti-oxidant factors in the mutant. The expression of katA in the mutant was significantly higher than in the wild-type (WT) in the presence or absence of 0.4 mM H2O2. The expression of dps in the mutant was significantly increased compared with the WT in the presence of H2O2 but not in the absence of H2O2. A katA or a dps-inactivated mutant had high susceptibility to H2O2 compared with WT. In addition, we found that the nitric oxide detoxification protein (flavohemoglobin: Hmp), which is regulated by SrrAB, was related to H2O2 susceptibility. The hmp-inactivated mutant had slightly lower susceptibility to the H2O2 produced by S. sanguinis than did WT. When a srrA-inactivated mutant or the WT were co-cultured with S. sanguinis, the population percentage of the mutant was significantly higher than the WT. In conclusion, SrrAB regulates katA, dps and hmp expression and affects H2O2 susceptibility. Our findings suggest that SrrAB is related in vivo to the co-existence of S. aureus with S. sanguinis.


Introduction
Staphylococcus aureus is a human pathogen that causes several diseases such as suppurative diseases, food poisoning and toxic shock syndrome [1,2]. Recently, methicillin-resistant S. aureus epidemics in hospitals have become a worldwide health problem [3][4][5]. S. aureus is a commensal bacterium found in humans that has been isolated from the skin and nasal mucosa of healthy subjects with a frequency of 20 to 60% [6,7]. Additionally, S. aureus is known to inhabit the oral cavity, including the oral mucosa, gingiva and dental plaque [8][9][10].
In a commensal bacterial flora, many bacteria produce anti-bacterial agents such as bacteriocins [11,12] and hydrogen peroxide compete with other bacterium [13][14][15]. It was demonstrated in virginal flora that H 2 O 2 -producing lactobacilli inhibited the growth of pathogens [16,17]. In oral flora, viridans group streptococci produced H 2 O 2 and had an antagonistic effect on pathogens [13][14][15]. Streptococcus sanguinis is an oral bacterium that is found primarily in dental plaques and has been reported to be an H 2 O 2 -producing species. Several reports have demonstrated that the H 2 O 2 produced by S. sanguinis can kill other oral bacterial species [18,19]. Uehara et al. reported that viridans group streptococci containing S. sanguinis inhibit colonization with S. aureus in newborns, which has been attributed to H 2 O 2 [20,21]. On the other hand, S. aureus was reported to possess several factors that confer resistance to H 2 O 2 , such as catalase (KatA), alkyl hydroperoxide reductase (AhpC) and DNA-binding proteins from starved cells (Dps) [22][23][24]. KatA and AhpC are enzymes that decompose H 2 O 2 . Dps is an inhibitor of hydroxyl radical (ÁOH) production from H 2 O 2 in the presence of iron via the Fenton chemistry. Therefore, the biological relevance of interactions between S. sanguinis, a resident of the oral cavity, and S. aureus is uncertain.
Two-component systems (TCSs) are composed of a sensor kinase and a response regulator and are bacterial-specific gene regulation systems. When a sensor kinase senses a stimulant in the extracellular environment, the response regulator is phosphorylated and regulates several genes to facilitate adaptation to the environment [25]. Recently, several TCSs have been reported to be important for adaptation to H 2 O 2 stress. In Escherichia coli, Salmonella enterica Serovar Typhimurium and Haemophilus influenzae, ArcAB has an oxygen sensing function and is essential for resisting reactive oxygen species, including H 2 O 2 [26][27][28]. In S. aureus, Sun et al. demonstrated that two TCSs (AgrCA and AirSR) affected the susceptibility to H 2 O 2 [29,30].
The TCS SrrAB is a known oxygen sensor in S. aureus and regulates several virulence genes under low oxygen conditions [31][32][33], as well as anaerobic metabolism genes and a flavohemoglobin hmp under low oxygen conditions or upon exposure to nitric oxide (NO) [34,35]. However, the relation between susceptibility to H 2 O 2 and SrrAB is unknown. In this study, we investigated the effects of SrrAB on susceptibility to the H 2 O 2 produced by S. sanguinis.

Bacterial strains and growth conditions
The bacterial strains used in this study are listed in Table 1. S. aureus was grown in 5 ml of tryptic soy broth (TSB) (Becton Dickinson Microbiology Systems, Cockeysville, MD, USA) in test tubes (18 mm diameter × 150 mm tall) at 37°C under aerobic conditions with shaking (120 rpm). S. sanguinis was aerobically grown in 5 ml of TSB in test tubes (18 mm diameter × 150 mm tall) at 37°C under 5% CO 2 without shaking. Tetracycline (Tc; 5 μg / ml) and chloramphenicol (Cp; 3 μg / ml) were added for the maintenance of S. aureus mutant strains. Ampicillin (100 μg / ml) and spectinomycin (50 μg / ml) were added for the maintenance of E. coli mutant strains.

Construction of S. aureus mutants
The srrA-inactivated mutants were previously constructed [36,37]. The genes dps, katA, hmp and perR were inactivated in S. aureus strain MW2 using the thermosensitive plasmid pCL52.1 by a previously described method [38]. Gene complementation was performed in the srrAinactivated mutants using pCL8, which is an E. coli-S. aureus shuttle vector [39]. Entire sequences of srrAB with their own promoters were amplified by PCR. The amplified DNA was cloned into the pCL8 vector using E. coli XLII-Blue cells. The constructs were purified and electroporated into S. aureus RN4220, which was the recipient for the foreign plasmid [40]. Then, the plasmid was transduced into the mutant strains using the phage 80 alpha [41]. As a control strain for co-culture assays, strain MW2 harbouring the empty pCL8 was constructed. The primers used are listed in Table 2.
Direct assay for evaluating susceptibility to H 2 O 2 produced by S. sanguinis The direct assay method was modified from a previously described method [42]. A total of 5 μl of S. sanguinis (10 8 cells / ml) was dropped onto a tryptic soy agar (TSA) plate. After 16 h of aerobic incubation at 37°C under 5% CO 2 , the mid-log phase (cell density 660 nm = 0.8) of S. aureus strains (10 7 cells) mixed with 6 ml of pre-warmed tryptic soy soft agar (0.5% agar) was poured over the plates. The plates were incubated overnight at 37°C under aerobic conditions. To analyze the effects of anaerobic conditions on the production of an antibacterial agent, S. sanguinis was grown on TSA plates anaerobically using a GasPak system (Mitsubishi Gas Chemical Company Inc., Tokyo, Japan). Then, after pouring tryptic soy soft agar containing S. aureus, the plate was incubated overnight at 37°C under anaerobic conditions. To neutralize the H 2 O 2 produced by S. sanguinis, 20 μl of bovine liver catalase (100 μg / ml) (Sigma-Aldrich, St. Louis, MO, USA) was dropped onto the area surrounding the S. sanguinis colony, and the direct assay was performed under aerobic conditions. The diameter of the S. aureus inhibition zone was measured in three directions to evaluate the inhibitory size. Three independent experiments were performed and are expressed as the mean ± SD.

H 2 O 2 susceptibility test
Mid-log phase (cell density at 660 nm = 0.8) S. aureus strains were washed with PBS and re-suspended in TSB. Then, 0.5 × 10 8 cells were inoculated into 10 ml of TSB or TSB containing 0.4 mM H 2 O 2 in a test tube (18 mm diameter × 150 mm tall) and grown aerobically at 37°C with shaking at 120 rpm. Bacterial growth was monitored to measure the bacterial density (OD 660 nm) for 2 to 10 h using the spectraphotometer miniphoto 518R (Taitec Corporation, Saitama, Japan).

Quantitative PCR
A small amount of the S. aureus strains (10 8 cells) was inoculated in 10 ml of TSB and grown aerobically to mid-log phase (cell density at 660 nm = 0.8) at 37°C with shaking at 120 rpm. The cultures were transferred to a centrifuge tube and treated with or without 0.4 mM H 2 O 2 for 10 min at 37°C with shaking (120 rpm). RNA extraction was performed using a FastRNA Pro Blue Kit (MP Biomedicals, Santa Ana, CA, USA) according to the manufacturer's protocol. One microgram of total RNA was reverse-transcribed into cDNA using a Transcriptor First Strand cDNA Synthesis Kit (Roche Diagnostics, Basel, Switzerland). Using cDNA as the template, quantitative PCR was performed using a LightCycler Nano (Roche Diagnostics). The primers used are listed in Table 2. Transcriptional levels were determined using 2 -ΔΔCt methods [43]. The Ct value of 16S rRNA in 1000-fold diluted cDNA was used as a reference. The means of Ct values in WT untreated with H 2 O 2 (N = 5) were used as the calibrator. All test and calibrator samples were normalized to the ΔCt value (ΔCt (test) = Ct (target test) -(reference test) , ΔCt (calibrator) = Ct (target calibrator) -(reference calibrator) . Then, the ΔΔCt value was determined (ΔΔCt = ΔCt (test) -ΔCt (calibrator) ). The relative expression level was calculated using the formula F = 2 -ΔΔCt . Individual experiments were performed three or five times, and the results expressed as the mean ± SD.
Restriction sites are underlined. doi:10.1371/journal.pone.0159768.t002 Co-culture assay Co-culture assays were performed using a previously described method [42]. Mid-log phase cells (cell density at 660 nm = 0.8) of the S. sanguinis and S. aureus strains were adjusted to 2 × 10 8 cells / ml using PBS. The same volume of S. aureus and S. sanguinis was mixed and 20 μl of the mixture was dropped onto a TSA plate. The plate was incubated for 2 h at 37°C under 5% CO 2 . The agar in the spotted area was excised and incorporated into 500 μl of PBS. Then, the agar was vigorously mixed to detach the bacterial cells from the agar. Appropriate dilutions were plated on TSA plates containing Cp (3 μg / ml), Tc (5 μg / ml), or ofloxacin (Oflx) (1 μg / ml) because of different susceptibilities to antibiotics. WT S. aureus (MW2:: pCL8) were selected with Cp. S. aureus mutants, and complemented strains were selected with Tc. S. sanguinis was selected with Oflx. After an overnight incubation at 37°C under 5% CO 2 , CFUs were determined, and the population percentage for each S. aureus strain was calculated.
To analyze the effects of pre-culturing S. sanguinis, ten microliters of S. sanguinis (10 8 cells / ml) was dropped onto a TSA plate and the plate was incubated for 1 h at 37°C under 5% CO 2 .
Then, ten microliters of S. aureus (10 8 cells / ml) was dropped onto a S. sanguinis colony and the plate was incubated for 2 h at 37°C under 5% CO 2 . The population percentage of S. aureus was determined by the method described above. Three independent experiments were performed and the results are expressed as the mean ± SD.

Susceptibility of the srrA-inactivated mutants to H 2 O 2 produced by S. sanguinis
A direct assay demonstrated that the srrA-inactivated MW2 mutant showed a small inhibition zone surrounding S. sanguinis compared with the WT and that the small zone of the mutant was restored by complementation with srrAB (Fig 1A and 1C). Additionally, we investigated the susceptibility of an srrA-inactivated TY34 mutant to S. sanguinis and found that the mutant had a small inhibition zone compared with the WT (Fig 1C). Under anaerobic conditions, S. aureus WT showed no inhibition zone, and no inhibition zone was observed after catalase treatment (Fig 1B). In the growth curve experiment, the growth of the srrA-inactivated mutant was higher than the WT in the presence of 0.4 mM H 2 O 2 . Statistical significance was observed between WT and the mutant in the presence of 0.4 mM H 2 O 2 at 10 h incubation. This phenotype in the mutant was restored by complementation with srrAB (Fig 2).

Expression of anti-oxidant factors and hmp in the srrA-inactivated mutant
We used quantitative PCR to investigate the expression of three anti-oxidant factors (katA, dps and ahpC) in the srrA-inactivated mutant exposed to 0.4 mM H 2 O 2 for 10 min. The expression of these three factors in the WT, the srrA-inactivated mutant and the complemented strain was increased by H 2 O 2 treatment. Compared with WT, the expression of katA was significantly higher in the mutant in the presence or absence of H 2 O 2 treatment. The high level was restored in the srrAB-complemented strain. The expression of dps in the mutant did not increase in the absence of H 2 O 2 , but the expression was significantly higher in the mutant treated with H 2 O 2 . The increased expression in the mutant was restored by complementation. The expression of ahpC in the mutant was slightly increased, but the expression was decreased in the mutant compared to the WT when treated with H 2 O 2 (Fig 3). Next, we focused on the expression of hmp because hmp expression is regulated by SrrAB in S. aureus [34,35] and is related to oxidative stress in S. enterica Serovar Typhimurium [44,45]. The expression of hmp in the srrA-inactivated mutant was significantly less than in WT treated or untreated with H 2 O 2 . The expression pattern in the complemented strain was similar to that of WT. The expression of hmp in the WT was increased 2.4-fold by H 2 O 2 treatment (Fig 3).

Susceptibility of H 2 O 2 and expression of anti-oxidant factors in the perRinactivated mutant
PerR is related to the regulation of anti-oxidant factors in S. aureus [46]. We analyzed the susceptibility of the perR-inactivated mutant to the H 2 O 2 produced by S. sanguinis. As shown in Fig 4A, a perR-inactivated mutant strain had significantly lower susceptibility to the H 2 O 2 produced by S. sanguinis than the srrA-inactivated mutant. The expression of katA, dps and ahpC was significantly increased in the perR-inactivated mutant in the absence of H 2 O 2 ( Fig 4B).  Susceptibility of the katA, dps or hmp-inactivated mutant to H 2 O 2 produced by S. sanguinis Because the expression of the two factors (katA and dps) was increased in the srrA-inactivated mutant treated with H 2 O 2 , we constructed a mutant at each locus and performed a direct assay to identify the factor(s) that affected susceptibility to H 2 O 2 . The katA and dps-inactivated mutants had a large inhibition zone compared with the WT (Fig 5). Additionally, we analyzed  the susceptibility of the hmp-inactivated mutant to H 2 O 2 , and found that the mutant had a small inhibition zone compared with the WT (Fig 5).

Co-culture of the S. aureus srrA-inactivated mutant with S. sanguinis
In a preliminary experiment, we demonstrated that the strain MW2 harbouring an empty pCL8 vector (MW2::pCL8) showed an inhibition zone similar to that of strain MW2 with no vector (S1 Fig). Therefore, we used this strain as a WT control for the co-culture assays. Additionally, we analyzed the growth of each S. aureus strain and S. sanguinis on TSA plates for 2 h and found that the growth was approximately the same among the S. aureus strains but that S. sanguinis grew approximately 2-fold more rapidly compared to the S. aureus strains (S1 Table). Fig 6A shows the population percentages for the S. aureus strains co-cultured with S. sanguinis for 2 h. The mutant population was approximately 2-fold larger than the WT. Fig 6B   Fig 6. Co-culture of the srrA-inactivated mutant with S. sanguinis. (A) The population percentages of S. aureus MW2 WT harbouring an empty pCL8 vector (MW2::pCL8), the srrA-inactivated mutant and the complemented strain when co-cultured with S. sanguinis were measured by co-culture assay as described in the Materials and Methods section. (B) The population percentage of MW2 strains co-cultured with precultured (37°C under 5% CO 2 for 1 h) S. sanguinis. The data are the mean ± SD of three biological independent experiments. Significant differences compared with WT were determined by Dunnett's test (*, P < 0.05; **, P < 0.001). shows the population percentages of the S. aureus strains when S. sanguinis was pre-cultured on a TSA plate for a 1 h. Before the co-culture assay, we demonstrated in a preliminary experiment that the number of S. sanguinis cells increased 4-fold after 1 h incubation when S. sanguinis cells alone were spotted on a TSA plate (S1 Table). The mutant population was 18-fold larger than that of the WT.

Discussion
We demonstrated in this study that an srrA-inactivated mutant has a smaller inhibition zone surrounding S. sanguinis than does the WT by a direct assay and that this inhibition was completely relieved by anaerobic incubation or catalase treatment (Fig 1). In addition, the mutant had a low susceptibility to H 2 O 2 (Fig 2). Therefore, the small inhibition zone of the srrA mutant was caused by the low susceptibility to the H 2 O 2 produced by S. sanguinis. Additionally, we demonstrated that the expression of both katA and dps was increased in the mutant exposed to H 2 O 2 (Fig 3). Based on these findings, we concluded that the low susceptibility of the srrA mutant to H 2 O 2 was primarily due to the increased expression of katA and dps.
SrrAB acts as a sensor for low oxygen tension and NO and regulates several factors that facilitate adaptation to these conditions. SrrAB regulates several virulence genes (tst, spa and icaA) under anaerobic or low oxygen conditions [31][32][33]. The expression of genes involved in anaerobic respiratory pathways (pflAB, adhE and nrdDG), cytochrome assembly and biosynthesis (qoxABCD, cydAB and hemABCX), iron-sulfur cluster repair (scdA) and NO detoxification protein (hmp) were altered in the srrAB mutant under low oxygen or NO stress conditions [34,35]. Furthermore, phosphatidylinositol-specific phospholipase C (plc) was regulated via SrrAB by hypochlorous acid or polymorphonuclear leukocytes [47]. However, the regulation of katA and dps by SrrAB has not been demonstrated. Recently, Windham et al. reported that SrrAB modulates S. aureus (strain UAMS-1) cell death in high glucose conditions and that an srrAB mutant had increased susceptibility to H 2 O 2 . They attributed the increased susceptibility to H 2 O 2 in the srrAB mutant to the production of endogenous reactive oxygen species by the expression of cidABC via SrrAB [48]. This report contains results conflicting with our results using S. aureus strain MW2 and TY34 (Figs 1 and 2). We investigated the expression of cidA in the srrA mutant of MW2 and found that the expression of cidA was significantly repressed by SrrAB ( S2 Fig). Therefore, we think that the effect of cidA in the srrA mutant is not much below the background of MW2 and TY34.
Previously, Horsburgh et al. reported that katA and dps expression in S. aureus was repressed by PerR, which is a Fur family protein [46]. As shown in Fig 4A, a perR-inactivated mutant showed lower susceptibility to H 2 O 2 than the srrA-inactivated mutant. Therefore, we analyzed the relation between SrrAB and PerR. First, we investigated perR gene expression in the srrA-inactivated mutant and found that perR gene expression was unaltered ( S3 Fig). Then, we investigated the expression of anti-oxidant factors, and found a higher expression of katA, dps and ahpC in the perR-inactivated mutant in the absence of H 2 O 2 treatment ( Fig  4B). Conversely, the increased expression of dps was not observed in the srrA-inactivated mutant untreated with H 2 O 2 (Fig 3). These results suggest that the increased expression of dps in the srrA mutant is not directly related to PerR. PerR is a repressor for several anti-oxidant factors, and this repression was alleviated by H 2 O 2 [49]. The increased expression of katA, dps and ahpC in the WT and the mutant treated with H 2 O 2 (Fig 3) indicates that PerR is also involved in the expression of these factors. Compared with the WT, a higher level of katA and dps transcripts was observed in the srrA-inactivated mutant treated with H 2 O 2 ( Fig  3). These results indicate that SrrAB together with PerR is independently involved in katA and dps regulation. The increased expression of these genes might be an indirect effect of a change in the redox-potential in the srrA-inactivated mutant because the mutant showed a decreased expression of the genes responsible for cytochrome assembly and heme biosynthesis in the electron transport chain [35]. However, because the expression pattern of katA and dps in the mutant was different (Fig 3), further studies will be required to clarify the link between SrrAB and katA or dps.
In addition, we demonstrated for the first time that Hmp was associated with H 2 O 2 susceptibility in S. aureus. A relationship between Hmp and susceptibility to oxidative stress has been reported in E. coli and S. typhimurium [50,44]. In the presence of NO, Hmp converts NO to nitrate (NO 3 -) by the reaction NO + O 2 + e -! NO 3 utilizing an electron from the reduction of flavin adenine dinucleotide (FAD) [51]. In the absence of NO, Hmp has the potential to generate superoxide anion radicals (O 2 -) by the reaction O 2 + e -! O 2 utilizing an electron from the reduction of FAD [52]. In addition, Hmp is associated with the production of ÁOH from H 2 O 2 via the Fenton chemistry in the absence of NO [44]. Based on these reports, it is thought that hmp inactivation in S. aureus suppresses the generation of intracellular oxidative stress, and the mutant showed lower susceptibility to H 2 O 2 than the WT. NsrR, which is a Rrf2 family transcription repressor, was demonstrated to repress the generation of oxidative stress in the absence of NO by repressing the expression of hmp in several bacterial species, including E. coli, S. typhimurium and B. subtilis [53]. The inactivation of nsrR results in high susceptibility to H 2 O 2 in S. typhimurium [45]. However, we could not find the gene nsrR or an nsrR homologue in the S. aureus genome database. TCS, SrrAB and/or ResDE have been reported to regulate Hmp in the presence of NO in S. aureus and Bacillus subtilis [34,35,54]. In B. subtilis, ResDE regulates Hmp expression in an NsrR-dependent manner [55], whereas in S. aureus, Hmp was dependent on SrrAB regulation. We suggest that the expression of hmp is regulated by SrrAB and affects the susceptibility to H 2 O 2 .
In a co-culture assay, the percentage of the srrA-inactivated mutant was high in a mixed culture with S. sanguinis. Because several oral streptococci, such as S. sanguinis, S. parasanguinis, S. gordonii and S. oralis, can produce H 2 O 2 [56][57][58], S. aureus requires H 2 O 2 resistance to survive in the oral cavity. In the oral cavity, S. aureus can colonize under anaerobic (dental plaque and gingival sulcus) and aerobic conditions (oral mucosa) [8][9][10]. Therefore, S. aureus can modulate its susceptibility to H 2 O 2 by SrrAB activity and coexist with H 2 O 2-producing oral streptococci, including S. sanguinis. Further studies will be required to analyze the functions of SrrAB involved in the co-existence with H 2 O 2 -producing bacteria in vivo, particularly in the oral cavity. The expression of perR in mid-log phase (cell density at 660 nm = 0.8) cells of S. aureus MW2 WT, srrA-inactivated mutant and the complemented strain grown in TSB was determined by quantitative PCR as described in the Materials and Methods section. The data are the mean ± SD of five biological independent experiments. Significant differences compared with WT were determined by Dunnett's test (N.S., not significant). (TIF) S1 Table. Bacterial growth of S. sanguinis and S. aureus strains on TSA plates. (DOCX)