Antisense yycG modulates the susceptibility of Staphylococcus aureus to hydrogen peroxide via the sarA

The infectious pathogen Staphylococcus aureus (S. aureus) is primarily associated with osteomyelitis. Hydrogen peroxide drainage is an effective antimicrobial treatment that has been adopted to combat S. aureus infections. Previous investigations have indicated that the antisense RNA (asRNA) strategy negatively modulates S. aureus YycFG TCS, and it significantly disrupts biofilm formation. However, the effects of the antisense yycG RNA (ASyycG) strategy on the susceptibility of biofilm-producing S. aureus to hydrogen peroxide and the mechanisms underlying this effect have not been elucidated to date. Overexpression of ASyycG inhibited the transcription of biofilm formation-related genes, including sarA and icaA. Additionally, the CFU counts and the live bacterial ratios of ASyycG biofilm-producing S. aureus treated with H2O2 were notably reduced across the groups. Notably, the predicted promoter regions of the sarA and icaA genes were directly regulated by YycF. ASyycG was observed to sensitize biofilm-producing S. aureus to H2O2 intervention synergistically via the sarA and thus may represent a supplementary strategy for managing osteomyelitis. However, future in-depth studies should attempt to replicate our findings in animal models, such as the rat osteomyelitis model.


Background
In humans, Staphylococcus aureus (S. aureus) is associated with a variety of diseases, ranging from relatively mild skin and soft tissue infections to life-threatening bacteremia and endocarditis. Hydrogen peroxide drainage is an effective antimicrobial treatment to combat S. aureus colonization and invasion [1]. Notably, it is well-known that 1000-fold greater resistance to antimicrobial agents was observed in the biofilm state than in the planktonic state [2][3][4][5]. During the establishment of biofilms, polysaccharide intercellular adhesion (PIA) is considered to be the key mechanism that provides an essential scaffold for biofilms and a special kind of extracellular matrix that is encoded by the ica operon and sarA [6,7]. The sarA gene has been reported to drive biofilm formation by altering ica transcription and producing PIA [2,4]. In Staphylococcus epidermidis (S. epidermidis), SarA was reported to be an essential positive regulator of the ica operon in biofilm development [8]. Additionally, numerous studies have evaluated the regulatory pathway of the sarA gene and demonstrated that sarA was associated with bacterial oxidation sensing, virulence factors, and biofilm formation in S. aureus [9][10][11].
Two-component signal transduction systems (TCSs) are essential regulators of staphylococcal metabolism and adaptation to environmental changes [12,13]. Among the 16 known TCSs in S. aureus, YycFG is essential for bacterial viability [14]. The YycG histidine kinase is generally anchored by a cytoplasmic membrane and is involved in monitoring environmental changes. Notably, when the responses of this kinase to extracellular stimuli transfer the phosphoryl group to activate the response regulator (RR) YycF, the cellular physiology status, including biofilm organization, adapts to changes [7,14,15]. In a previous study, we demonstrated that the essential YycFG TCS was closely related to biofilm formation and extracellular matrix organization [16]. However, since the YycFG TCS is essential to S. aureus viability, the construction of deletion mutants of YycFG was unsuccessful [17].
Antisense RNA (asRNA) is a single-stranded RNA that is complementary to the target messenger RNA (mRNA). The interaction between the RNA molecules inhibits downstream signal transduction activation. Using this asRNA strategy, we constructed mutant antisense S. aureus strains that express YycFG TCS at low levels. The number of icaA gene transcripts and the level of biofilm formation decreased in the ASyycG-mutant S. aureus, and YycFG was expressed at low levels [18]. However, the potential mechanisms by which the transcriptional regulator YycF modulates sarA and/or icaA expression and the biological effects of this modulation on the susceptibility of biofilm-producing S. aureus to hydrogen peroxide warrant further research.
In the present study, we investigated the role played by antisense yycG in regulating the susceptibility of biofilm-producing S. aureus to H 2 O 2 treatment. In addition, the potential association of YycF with sarA and icaA was evaluated using an electrophoretic mobility shift assay (EMSA). This work primarily investigated whether an antisense yycG interference strategy could sensitize biofilm-producing S. aureus to H 2 O 2 intervention, which may be associated with the direct regulation of YycF to the adjacent genes sarA and icaA.

Results
Antisense yycG modulated oxidative regulation Transcriptome and enrichment analyses showed that overexpression of antisense yycG influenced the pathways associated with biofilm metabolism, virulence, oxidative regulation, and glycolysis/gluconeogenesis utilization by S. aureus ( Fig. 1a and b). Notably by padj values ranking, the 20 KEGG pathways with the greatest differential expression that were involved in S. aureus infection and metabolism were determined to have enrichment score mostly at around 0.5 differences between the antisense yycG overexpression group and ATCC29213 group. The less padj value was, the greater expression diversity in pathway between the antisense yycG overexpression group and ATCC29213 group. Each pathway involved different genes count, such as microbial metabolism in diverse environments involving about 60 genes with a padj value at about 0.05 (Fig. 1b). Moreover, three genes that regulate biofilm formation and virulence were modulated; specifically, sarA was downregulated, and codY and srrA were upregulated (see Fig.  1c). These results indicated that antisense yycG negatively regulated biofilm formation and the expression of the virulence-associated gene sarA.
Antisense yycG sensitized biofilm-producing S. aureus to H 2 O 2 intervention synergistically Through fluorescence microscopy, the live ratios of bacteria were compared. In the ASyycG + H 2 O 2 group (Fig. 2a), the antibacterial effect of H 2 O 2 was significantly enhanced by ASyycG synergistically and exhibited the lowest live bacteria ratio at 22.1 ± 4.6 compared with each group (Fig. 2b). The SEM and immunofluorescence results were similar and exhibited the lowest EPS levels in the ASyycG + H 2 O 2 group. In addition, we could only detect rare cell clusters in ASyycG + H 2 O 2 , even after 24 h of culture (Fig. 3a). Quantitively, we performed a CFU test on each biofilm. Correspondingly, the CFU count of AS S. aureus + H 2 O 2 biofilm was mostly lower in all groups than in the S. aureus ATCC29213 parent group (Fig. 3b). Accordingly, we evaluated the ability of the S. aureus strains to form biofilms. The biomass was quantified via a microtiter dish assay, and ASyycG strains exhibited reduced biofilm formation compared with the S. aureus ATCC29213 group. In particular, the ASyycG strains treated with H 2 O 2 exhibited the strongest decrease in biofilm formation among all groups, which indicated that AS S. aureus + H 2 O 2 biofilms had significantly decreased biofilm growth compared to the other groups (Fig. 4a). These results showed that antisense yycG could significantly enhance the susceptibility of S. aureus to H 2 O 2 and exert negative effects on biofilm-producing S. aureus after H 2 O 2 treatment.
Antisense yycG overexpression inhibited the transcription of the sarA and ica genes Quantitative RT-PCR analyses demonstrated that the expression levels of the genes in the YycFG pathway, including yycG and yycF, were significantly reduced, which may be attributable to the overexpression of ASyycG in the ASyycG group treated with H 2 O 2 . Furthermore, the expression levels of sarA related to the oxide stress reaction and biofilm-associated ica genes were more significantly reduced by ASyycG overexpression in the ASyycG strains treated with H 2 O 2 than in the S. aureus parent ATCC29213 strains (n = 10, P < 0.05; Fig. 4b). In keeping with this result, YycFG pathway protein expression was significantly downregulated in the ASyycG group treated or not treated with H 2 O 2 compared with the ATCC29213 strain. Notably, the protein expression levels of the cognate sensor kinase YycG and the response regulator YycF were elevated in response to H 2 O 2 intervention (Fig. 4c). These results indicated that antisense yycG inhibited the expression of the YycFG pathway, which may have contributed to reductions in downstream icaA and sarA gene expression.
YycF bound to the predicted promoter regions of sarA and icaA genes The purified recombinant YycF protein was visualized via Coomassie staining after SDS-PAGE (Fig. 4d). The promoter region of sarA contained a putative YycF binding consensus motif (Fig. 5a). We employed the bacterial promoter database (http://www.softberry.com/berry. phtml?topic=index&group=programs&subgroup= gfindb), and a putative promoter of the sarA gene was predicted. EMSA demonstrated that the YycF protein was bound to DNA fragments from sarA and icaA promoter regions. As a negative control, to rule out nonspecific binding, we utilized a DNA fragment of the same size as the predicted promoter and with a similar AT: GC mole ratio but missing the YycF consensus binding sequence (Fig. 5b). By EMSAs, we determined that YycF could directly bind to the promoter regions of the sarA and icaA genes and regulate their expression, which contributed to biofilm metabolism and resistance to antibacterial agents, such as H 2 O 2 (Fig. 5c).

Discussion
Microbial biofilm formation is a principal factor for chronic infectious diseases, such as osteomyelitis; therefore, numerous antibiofilm strategies have been devised for clinical use by surgeons, such as aggressive irrigation and physical removal of infective debris [19,20]. In addition, numerous other promising antibiofilm agents, such as zinc oxide nanoparticles (ZnO NPs), proteinase K, hamamelitannin (HAM) and antimicrobial peptides (AMPs), have also been developed [21,22]. Although 3 % H 2 O 2 is conventionally applied in most bone infection cases to disinfect microorganisms from wounds in clinical practice, the killing ability of H 2 O 2 on pathogenic bacteria is doubtful [23][24][25], especially under biofilm conditions in which the resistance increases 1000fold compared to planktonic bacterial conditions [26][27][28][29]. Biofilms, as a physical barrier, could protect microorganisms from the diffusion of environmental stimuli and confer antibiotic resistance to the microorganisms embedded within the biofilm. Multiple factors, including the existence of metabolically inactive bacterial and immune system evasion, contribute to the resistance of biofilms to disinfectant agents [30]. In this study, transcriptome analysis demonstrated that antisense yycG overexpression downregulated the expression of the oxidative reaction-,  Extracellular polysaccharide substance (EPS), an essential component of biofilms, plays an important role in preventing bacterial colonies from antimicrobial effects. In a previous study, improved killing efficiency of antimicrobial agents for bacteria embedded in EPS could be obtained by combined application with biofilmdispersing enzymes [28,31]. Our study obtained similar results, which showed that the lowest EPS levels and the fewest cell colony clusters were observed by SEM after 24 h of culture in the ASyycG + H 2 O 2 group. The icaADBC operon synthesized polysaccharide intercellular adhesion (PIA) as a significant component of EPS in biofilm production, which is modulated by the only essential TCS, YycFG, to promote cell wall and biofilm metabolism [32]. A plasmid overexpressing a complementary base-pairing antisense YycG (ASyycG) can significantly reduce biofilm formation by inhibiting the YycFG pathway [33]. Notably, when EPSs are disorganized, the resultant exposure and release of residual biofilm cells to these agents enhance their sensitivity. Furthermore, the CFU counting assays and live bacterial ratios indicated that yycG sensitized biofilmproducing S. aureus to H 2 O 2 treatment synergistically. When we combined ASyycG with H 2 O 2 antibacterial agents, not only was the antibacterial effect of H 2 O 2 improved, but a prolonged inhibited biofilm formation effect was also acquired 24 h after treatment with low potential to cause a chronic infection. In vivo research would further be considered to evaluate the synergistic antibacterial effects of ASyycG combined with H 2 O 2 . Although 3 % H 2 O 2 is continually applied to disinfect wounds microorganisms in clinical irrigation, it oxidizes protein, nucleic acid, and lipids of normal healthy cells, which limited the beneficial effect of promoting wound healing becomes [23,24]. However, at comparatively low concentrations, the killing ability of H 2 O 2 on pathogenic bacteria is doubtful [24,25]. In the present study, the application of antisense yycG downregulated the expression of the EPS synthesis-associated gene ica, inhibited biofilm formation and enhanced susceptibility to H 2 O 2 in S. aureus. From this perspective, with the supplementary antisense yycG, an appropriate low therapeutically concentration of H 2 O 2 could be apply to maintain this promoter DNAs were used to out-compete YycF-specific binding. As the negative control, a DNA fragment the same size as the promoter region and similar AT: GC mole ratio, but missing the YycF consensus binding sequence, was used to rule out non-specific binding; c The working model of oxidation regulation by ASyycG via sarA pathway balance by enhancing antibacterial effect and keeping wound healing with ameliorating wound redox extent.
On the other hand, the sarA gene was reported to enhance biofilm formation by altering ica transcription and the production of PIA in S. aureus [2,4]. In S. epidermidis, SarA was demonstrated to be an essential positive regulator of the ica operon in biofilm development [8]. In addition, homologous YycF could directly regulate icaA and sarA gene expression in S. epidermidis [7]. In the present study, the promoter regions of sarA were directly bound by YycF, as demonstrated via EMSA, which showed that YycF could indirectly regulate ica through the sarA pathway.
Regarding environmental stimuli, the essential TCS YycFG (also known as WalRK) was reported to be associated with bacterial oxidation sensing [34,35]. Notably, in the present study, the protein production of the cognate sensor kinase YycG, as well as the response regulator YycF, was enhanced in response to H 2 O 2 intervention, which again confirms that TCS YycFG can be applied to monitor environmental changes. By oxidative stress, biofilm-producing S. aureus and the expression of biofilm-inducing genes, such as sarA, were supposed be elevated [35]. However, the expression levels of the antioxidant stress reaction-related and biofilm-associated sarA and ica genes were significantly reduced by ASyycG RNA in the ASyycG strains treated with H 2 O 2 compared with the parental S. aureus ATCC29213 strains in our study. In addition, the global regulator CodY controls the expression of dozens of metabolism and virulence genes in S. aureus, including genes associated with biofilm formation. The activity of CodY in regulating biofilm formation and cell aggregation was associated with ica expression and PIA production [36]. CodY, a regulatory protein, has been reported to repress virulence gene expression in S. aureus [37]. In present study the expression of CodY was relatively high in ASyycG group, which may contribute to repress the virulence gene expressions in S. aureus. However, whether CodY is involved in sensing bacterial oxidation warrants further investigation, particularly concerning the mechanisms by which it regulates virulence genes. Additionally, the regulatory effect of the srrA gene associated with the YycFG TCS merits further investigation.

Conclusions
The effects of the antisense yycG RNA strategy on the susceptibility of biofilm-producing S. aureus to H 2 O 2 treatment were investigated. Recombinant shuttle plasmids were used to overexpress an antisense yycG RNA (ASyycG) and were transformed into the S. aureus strain to construct ASyycG S. aureus. The results of this study indicated that yycG overexpression inhibited the transcription of biofilm formation-related genes, including sarA and icaA. Additionally, the CFU counts of ASyycG biofilms treated with H 2 O 2 were mostly decreased. Notably, YycF could bind to the predicted promoter regions of sarA and icaA genes, demonstrating that susceptibility of biofilm-producing S. aureus to H 2 O 2 intervention may be regulated by ASyycG via the sarA and icaA pathways. In summary, these results indicate that ASyycG synergistically sensitizes biofilm-producing S. aureus to H 2 O 2 treatment and thus may be considered a supplementary strategy for managing osteomyelitis. Considering the possibility of antisense yycG breakdown by H 2 O 2 , a stabilizing vector, such as nanographene oxide, may be utilized in future research to support the clinical strategy of combining H 2 O 2 with antisense yycG.

Bacterial strains and biofilms growth conditions
S. aureus strain ATCC29213, provided by the Department of Laboratory Medicine (West China Hospital, Sichuan University, Chengdu, China), was cultured in tryptic soy broth (TSB) [16]. In a previous study, we performed phenotypic experiments to ensure that S. aureus strain ATCC29213 was a biofilm-producing pathogen [33]. After overnight incubation at 37°C and 5 % CO 2 , 500 µL of S. aureus suspension was inoculated into 10 mL fresh TSB medium to mid-logarithmic phase (optical density at 600 nm [OD 600 ] = 0.5). For the formation of biofilms, sterilized glass disks (diameter of 10 mm) were dropped into log-phased S. aureus sterile 24-well microtiter plates for 24-h coculture biofilms. This basic culture section was prepared for the subsequent tests.

Construction of S. aureus antisense yycG overexpression strains
To observe the effect of antisense yycG in S. aureus, we first constructed antisense yycG-overexpressing S. aureus strains according to the following protocols. Recombination of the plasmid pDL278 containing the antisense yycG fragment was conducted by inserting the antisense yycG (ASyycG) sequence into restriction sites between BamHI and EcoRI, which was subsequently synthesized by Sangon Biotech (Shanghai, China). According to our previous protocol [38], an ASyycG-overexpressing S. aureus (ASyycG mutants) was constructed by adding 200 ng recombinant pDL278 ASyycG plasmid with 1 µg/ mL competence-stimulating peptide (CSP) into a 250-µL mid-exponential-phase S. aureus suspension for 60 min [33]. The empty pDL278 plasmid did not exert any effects on the viability of S. aureus [39].

RNA extraction analyses
To investigate the alterations in the whole transcriptome between S. aureus ATCC29213 and ASyycG mutants, whole transcriptome analysis was applied. First, samples of S. aureus strain ATCC29213 and ASyycG mutants were cultured in 10 mL of fresh TSB media into the mid-logarithmic phase. Total RNA was isolated and quantified as previously described [33]. Agarose gel electrophoresis (1 %) was run to check for degradation or contamination of the RNA. We used a NanoPhotometer® spectrophotometer (IMPLEN, CA, USA) for RNA quality and quantity checks. The integrity of the extracted RNA sample was assessed using the RNA Nano 6000 Assay Kit of the Bioanalyzer 2100 system (Agilent Technologies, CA, USA).

Library preparation for whole transcriptome sequencing
After the abovementioned RNA extraction, exactly 5 µg RNA per sample for the S. aureus ATCC29213 and ASyycG mutant groups was used as input for library preparation. Sequencing libraries were generated using the NEBNext® Ultra™ RNA Library Prep Kit for Illumina® (NEB, USA) following the manufacturer's recommendations. Next, index codes were added to attribute sequences to each sample. The rRNA was removed using a specialized kit. Fragmentation was performed using divalent cations under elevated temperature in NEBNext First Strand Synthesis Reaction Buffer (5X). Using a fragmented mRNA template, first-strand cDNA was synthesized using random hexamer primers in the M-MuLV Reverse transcription system. The RNA was digested by RNase H, and then second strand cDNA synthesis was subsequently performed using DNA Polymerase I and RNase H. The remaining overhangs were converted into blunt ends via exonuclease/polymerase activities. After adenylation of the 3' ends of DNA fragments, NEBNext adapter with a hairpin loop structure was ligated. Next, USER Enzyme (NEB, USA) was employed to degrade the second strand of cDNA. The library fragments were purified using the AMPure XP system (Beckman Coulter, Beverly, USA) to select cDNA fragments with a length of 250~300 bp. Then, PCR was performed using Phusion High-Fidelity DNA polymerase, universal PCR primers, and Index (X) Primer. Similarly, the above PCR products were purified (AMPure XP system), and library quality was assessed on the Agilent Bioanalyzer 2100 system. The cluster generation of the index-coded samples was performed on a cBot Cluster Generation System using TruSeq PE Cluster Kit v3-cBot-HS (Illumina) according to the manufacturer's instructions. After clustering, sequencing was performed on an Illumina Nova-Seq platform, whereby 150-bp paired-end reads were generated.

Differential expression analysis
Following the above library preparation, differential expression between the two groups was analyzed. In this analysis, high-quality clean data were obtained by eliminating reads with adapters and low-quality reads from raw data. Bowtie2 was applied to map the abovefiltered reads to the reference genome. To quantify the gene expression level, HTSeq v0.6.1 was used to count the read numbers mapped to each gene. Next, the number of fragments per kilobase of transcript sequence per million base pairs sequenced (FPKM) of each gene was calculated. Differential expression analysis of two groups was performed using the DESeq R package (1.18.0) [40]. DESeq provided statistical routines for determining differential expression in digital gene expression data using the negative binomial distribution-based model. The resulting P-values were adjusted using Benjamini and Hochberg's approach to control the false discovery rate. Genes with adjusted P-values < 0.05 found via DESeq were classified as differentially expressed.

KEGG enrichment analysis of differentially expressed genes
After the procedure described above was performed, raw data regarding differentially expressed genes were acquired. To identify the functions of differentially expressed genes between the S. aureus and ASyycG groups, Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis was applied. KEGG is a database resource for elucidating high-level functions and utilities of biological systems, such as cells, organisms, and ecosystems, from molecular-level information, especially large-scale molecular datasets generated by genome sequencing and other high-throughput experimental technologies (http://www.genome.jp/kegg/) [41]. In this study, we used KOBAS software to test the statistical enrichment of differentially expressed genes in KEGG pathways. The results showed 20 KEGG pathways with adjusted P-values of approximately 0.5, involving the S. aureus infection pathway, and enrichment score differences were detected between the antisense yycG overexpression group and ATCC29213 group.

Bacterial exposure to oxidants
After genetic alternation analysis and associated functional enrichment screening, we examined biofilm characteristics associated with resistance strength to chemical situlas, particularly H 2 O 2 , by overexpressing ASyycG. To determine the susceptibility of S. aureus ATCC29213 and ASyycG mutants to oxidants, 24-h-old biofilms derived from these groups were treated with 3 % H 2 O 2 (for clinical irrigation, H 2 O 2 is usually 3 %) for 30 min at 37°C. Residual H 2 O 2 was removed by diluting biofilm samples in PBS three times [23,42]. The biofilm sample disks were cultured in TSB medium for 24 h after treatment.

Microtiter dish assay and colony-forming unit (CFU) counting
The microtiter dish assay was applied to examine the biomass of 24-h-old biofilms after interventions with crystal violet (CV). For the formation of biofilms, samples of S. aureus strain ATCC29213 and ASyycG mutants were cultured into the midlogarithmic phase in 1 mL TSB media, which were dropped into sterile 24-well microtiter plates for 24 h. After 24 h of culture, the culture medium was removed. We washed the established biofilms three times with PBS solution. The dye bound to the biofilms was assayed after 0.1 % (w/v) crystal violet staining for 15 min followed by 1 mL destaining solution (ethanol/acetone = 8:2). Subsequently, 100 µL of the solution was transferred into a new 96-well plate, and the absorbance was measured with a microplate reader (ELX800, Gene) at OD 600 nm [33].
For the CFU counting test, the treated biofilms were subsequently cultured in TSB medium. After 24 h of culturing, the biofilms were collected via ultrasound in 1 mL of PBS for 15 min. The acquired microbiological samples were serially diluted 6-fold (from 10 − 1 to 10 − 6 ) in PBS and spread onto TSA agar plates for incubation at 37°C in 5 % CO 2 for 24 h. To determine cell viability, CFU per millimeter of microbiological suspension was counted [16].
Furthermore, the biofilms treated for 24 h were labeled with SYTO9 (LIVE/DEAD Bacterial Viability Kit reagent; BacLight, Invitrogen, Grand Island, NY, USA); live cells appeared green, while dead cells were stained red with propidium iodide. The cells were visualized using epifluorescence microscopy (Nikon Eclipse TE-2000 S, Melville, NY) at 40×magnification. Notably, five random fields in each specimen were selected [16].

Characterizing biofilm morphology
Scanning electron microscopy (SEM) was applied to assess the structure of biofilms. The samples were diluted twice using PBS and fixed with 2.5 % glutaraldehyde for 4 h. Fixed samples were serially dehydrated with increasing concentrated ethanol solutions from 30 to 100 % and dried using a critical point dryer. Subsequently, samples were coated with gold powder. We obtained micrographs using a scanning electron microscope (Inspect, Hillsboro, OR, USA) [16].

Analysis of gene expression using quantitative real-time PCR
To further validate the biofilm-associated genes and YycFG pathway modulated by ASyycG according to the enrichment results, quantitative real-time PCR was conducted according to the following protocols. Total RNA was extracted and purified with the MasterPure™ RNA Purification Kit (Epicentre Technologies, Epicentre, Madison, WI, USA) from 24-h-treated biofilm samples obtained from the S. aureus group, S. aureus + H 2 O 2 group, AS yycG group, and AS yycG + H 2 O 2 group following the manufacturer's instructions. Contaminating genomic DNA was digested and removed with Turbo RNase-free DNase I (Ambion, New York, NY USA) according to the recommended instructions. Next, the purity (A260/A280) and concentration of RNA were determined using a NanoDrop 2000 spectrophotometer (Thermo Scientific, Waltham, MA, USA). The purified RNA was reverse-transcribed to cDNA using random hexamers or gene-specific primers (Table 1) with the RevertAid First Strand cDNA Synthesis Kit (Thermo Scientific). We conducted quantitative real-time polymerase chain reaction (qRT-PCR) assays using the primers listed in Table 1, with the 16Sr RNA gene serving as an internal control [16,33]. Each sample was analyzed in triplicate, and the threshold cycle values (CT) were quantified.

Western blotting
To determine the protein production of YycF and YycG, Western blotting was performed as described below. S. aureus ATCC 29,213 and ASyycG mutants were cultured to the mid-logarithmic phase (optical density at 600 nm [OD600] = 0.5). Then, we treated the S. aureus planktonic cultures with 100 mM H 2 O 2 for 60 min, washed them and resuspended them in PBS (pH 7.2). The bacterial cells, including the S. aureus group, S. aureus + H 2 O 2 group, AS yycG group, and AS yycG + H 2 O 2 group, were mechanically disrupted as previously described [31]. For Western blot analysis, equal amounts of protein (30 µg) were mixed with 2X SDS-PAGE Sample Loading Buffer (Sangon Biotech, Shanghai, China) in boiling water for 10 min and loaded on 10 % SDS-PAGE gels (Bio-Rad). Proteins were fractionated and electrotransferred to polyvinylidene fluoride (PVDF) membranes (Biosharp Biotech, Shanghai, China). Membranes were blocked in TSBT buffer (100 mM Tris-HCl, 2.5 mM NaCl) containing 5 % w/v nonfat dry milk at room temperature for 2 h. To measure YycG and YycF production, membranes were incubated with purified YycGand YycF-specific antibodies (1:500, HuaBio Biotechnology, Hangzhou, China) for 2 h at room temperature, washed in Tris-buffered saline containing 0.1 % Tween 20, and incubated with horseradish peroxidase (HRP)conjugated goat anti-rabbit secondary antibody (1:5,000) for 2 h at room temperature as previously described [28]. Protein immunoreactive bands were visualized using an Immobilon Western Chemiluminescent kit (Millipore, Billerica, MA, USA). We utilized a Coomassie-stained gel to estimate equal loading of the samples for total bacterial lysis.

Electrophoretic mobility shift assay (EMSA)
To determine whether the YycF protein could directly bind to the promoter regions of the sarA and icaA genes, electrophoretic mobility shift assays (EMSAs) were conducted. To generate YycFHis-Tag fusion proteins, the ORF was amplified and restricted. Then, purified products were cloned into digested pET-22b (Novagen) to yield pET-yycG by Huabio Biotech (Hangzhou, China). Plasmids were transformed into E. coli BL21. Recombinant proteins were isolated from 500 µL of culture after a 3-h induction with 1 mM IPTG. After cell lysis, recombinant proteins were purified through affinity chromatography on Ni 2+ NTA agarose (Qiagen) as previously described [43]. The purified protein was visualized via Coomassie staining after SDS-PAGE.
EMSA was used to determine whether the sarA or icaA gene was directly regulated by YycF as previously described [43] with modifications. A PCR amplicon was generated from S. aureus ATCC29213 genomic DNA using primers labeled with the 5' FAM (Roche) (see Table 1). Before the binding reaction, DNA fragments were purified according to the manufacturer's instructions (Tiangen Biotech, Beijing, China). Labeled DNA fragments (0.02 pmol) were incubated with increasing amounts of recombinant YycF (0, 20, 40, and 60 pmol) and a 100-fold excess of unlabeled DNA fragments (cold DNA) as a competitor. The reaction mixture (20 µL) contained 50 % glycerinum, labeled DNA fragments, and YycF proteins.

Data analysis
All statistical data were analyzed in SPSS 16.0 (SPSS Inc., Chicago, IL, USA). The Shapiro-Wilk test was used to analyze the distribution of data, and the Bartlett test was used to determine the homogeneity of variances. For parametric testing, we adopted one-way ANOVA to assess the statistical significance of variables followed by the Tukey test. In the data, P-values < 0.05 were considered to indicate significant differences.  5'-ACTACTGCTGCGTTAATAAT -3' yycF 5' -TGGCGAAAGAAGACATCA -3'/ Ref. [30] 5' -AACCCGTTACAAATCCTG-3' yycG 5' -CGGGGCGTTCAAAAGACTTT -3'/ Ref. [30] 5' -TCTGAACCTTTGAACACACGT -3' The program Primer3 (https://bioinfo.ut.ee/primer3-0.4.0/) was used to design the sarA and codY primers. For the sarA gene, the forward primer targeted starting at nucleotide 194, and the reverse primer targeted starting at nucleotide 429. For the codY gene, the forward primer targeted starting at nucleotide 352, and the reverse primer targeted starting at nucleotide 501