Staphylococcus epidermidis SrrAB Regulates Bacterial Growth and Biofilm Formation Differently under Oxic and Microaerobic Conditions

Abstract

SrrAB expression in Staphylococcus epidermidis strain 1457 (SE1457) was upregulated during a shift from oxic to microaerobic conditions. An srrA deletion (ΔsrrA) mutant was constructed for studying the regulatory function of SrrAB. The deletion resulted in retarded growth and abolished biofilm formation both in vitro and in vivo and under both oxic and microaerobic conditions. Associated with the reduced biofilm formation, the ΔsrrA mutant produced much less polysaccharide intercellular adhesion (PIA) and showed decreased initial adherence capacity. Microarray analysis showed that the srrA mutation affected transcription of 230 genes under microaerobic conditions, and 51 genes under oxic conditions. Quantitative real-time PCR confirmed this observation and showed downregulation of genes involved in maintaining the electron transport chain by supporting cytochrome and quinol-oxidase assembly (e.g., qoxB and ctaA) and in anaerobic metabolism (e.g., pflBA and nrdD). In the ΔsrrA mutant, the expression of the biofilm formation-related gene icaR was upregulated under oxic conditions and downregulated under microaerobic conditions, whereas icaA was downregulated under both conditions. An electrophoretic mobility shift assay further revealed that phosphorylated SrrA bound to the promoter regions of icaR, icaA, qoxB, and pflBA, as well as its own promoter region. These findings demonstrate that in S. epidermidis SrrAB is an autoregulator and regulates biofilm formation in an ica-dependent manner. Under oxic conditions, SrrAB modulates electron transport chain activity by positively regulating qoxBACD transcription. Under microaerobic conditions, it regulates fermentation processes and DNA synthesis by modulating the expression of both the pfl operon and nrdDG.

INTRODUCTION

Staphylococcus epidermidis is an opportunistic pathogen, seldom excreting virulence factors and less aggressive in comparison to Staphylococcus aureus but capable of forming a multilayered biofilm on implanted medical devices, such as vascular catheters, prosthetic joints, artificial heart valves, etc. (1, 2). The bacteria within the biofilm are protected against killing by antibiotics and the host immune system, which contributes to increasing resistance to antimicrobial drugs and persistent infections (3–5). Biofilm-related infections persist until the biomedical implant is removed, resulting in extra trauma and cost to the patients.

Biofilm formation is a complicated process in staphylococci, being regulated by multiple regulatory factors, including Agr P2/P3, SarA, SigB, and two-component signal transduction systems (TCSs) (6–10). TCSs serve as a basic stimulus-response coupling mechanism by which bacteria adapt the environmental changes and consequently play a key role in pathogenesis (11–13). Our previous study revealed that the TCSs LytSR, SaeRS, and ArlRS are involved in S. epidermidis biofilm formation (14–16), whereas the role of the SrrAB (staphylococcal respiratory response) remained unclear.

The SrrAB shares considerable homology with ResDE of Bacillus subtilis (17, 18), and in S. aureus acts as a global regulator of virulence factors (SPA, TSST-1, RNAIII, etc.) in response to oxygen tension (19–22). A study by Yarwood et al. demonstrated that srrAB deletion (in S. aureus MN8) resulted in growth reduction only under anaerobic conditions, and the expression of RNAIII was inversely related to expression of srrAB (20). Throup et al. found that srrA deletion (in S. aureus WCUH29) led to changes in the expression of enzymes involved in fermentative metabolism (e.g., alcohol dehydrogenase, l-lactate dehydrogenase, NADH dehydrogenase, etc.), suggesting a role in the retarded growth of S. aureus under anaerobic conditions (19). In addition, a transposon mutation in srrA resulted in reduction of biofilm formation in S. aureus, although PIA production was increased, suggesting that in S. aureus srrAB affects biofilm formation via an ica-independent pathway (23).

Development of biofilm formation has been described as a two-step process involving an initial attachment, then an aggregation and maturation phase (4, 8). The initial adhesion of bacterial cells to a polymer surfaces is influenced by a number of factors in S. epidermidis, including AtlE, Embp, and other staphylococcal surface-associated proteins (6, 7, 11). In the maturation phase of biofilm development, the most important adhesive biofilm matrix is PIA (polysaccharide intercellular adhesion) (12). The biosynthesis, exportation, and modification of PIA are accomplished by the products of icaADBC operon, and the icaA is negatively regulated by the divergently transcribed icaR gene (3, 4, 24). Besides IcaR, several DNA-binding proteins regulate ica transcription, including SarA, RsbU, ArlR, etc. (10, 13, 16). However, it is possible that other biofilm matrix components are critical for staphylococcal biofilm formation, such as accumulation-associated protein (Aap), and extracellular DNA (eDNA), which mediated cell-cell aggregation and multilayered biofilm formation (25, 26). Environmental factors (such as oxygen limitation, alcohol, NaCl, etc.) may also influence staphylococcal biofilm formation (16, 27).

Much attention has been focused on the relevance of SrrAB as virulence factors, while the mechanisms by which staphylococcal SrrAB regulates biofilm formation have not been investigated in great detail. Here, we come up with new aspects of the role of SrrAB in the regulatory network of biofilm formation in S. epidermidis.

MATERIALS AND METHODS

Ethics statement.

All procedures performed on rabbits were carried out according to relevant national and international guidelines (the Regulations for the Administration of Affairs Concerning Experimental Animals, China, and the National Institutes of Health Guide for the Care and Use of Laboratory Animals) and were approved by the Institutional Animal Care and Use Committee (IACUC) of Shanghai Medical College, Fudan University (IACUC animal project number 20110628-16-qu).

Bacterial strains, plasmids, and growth conditions.

The bacterial strains and plasmids used in the present study are listed in Table 1. S. epidermidis 1457 (SE1457) and S. aureus RN4220 were kindly provided by Yicun Gao from Hong Kong University; S. epidermidis RP62A (accession number NC_002976) (28) was purchased from the American Type Culture Collection (ATCC; Manassas, VA). All staphylococci were routinely cultured in tryptone soy broth (TSB; Oxoid, Basingstoke, United Kingdom) or tryptone soy agar (TSA). For the detection of biofilm formation, S. epidermidis was cultured in TSA medium supplemented with 0.5% glucose. For the transformation of recombinant plasmids, B2 medium (1% casein hydrolysate, 2.5% yeast extract, 0.5% glucose, 2.5% NaCl, 0.1% K₂HPO₄ [pH 7.5]) was used for the recovery of staphylococcal cells after electroporation. Luria-Bertani medium was used for culture of Escherichia coli. Oxic conditions were created by incubation into a flask, in which the culture medium did not exceed 15% of the flask volume, and for microaerobic conditions bacteria were incubated into a syringe fully filled with medium. For static incubation under microaerobic conditions, polystyrene plates inoculated with bacteria were placed in an anaerobic bag with Anaerocult C Mini (Merck KGaA, Darmstadt, Germany). When appropriate, antibiotics were used at the following concentrations: erythromycin (5 μg/ml), spectinomycin (100 μg/ml), chloramphenicol (10 μg/ml), ampicillin (100 μg/ml), and kanamycin (50 μg/ml).

TABLE 1.

Bacterial strains and plasmids used in this study

Plasmid or strain	Descriptiona	Source or reference(s)
Plasmids
pET28a	E. coli expression plasmid; Kmr	Novagen
pET28a-srrA	pET28a harboring the srrA gene, used for SrrA expression	This study
pMAD	Shuttle vector, temperature sensitive; Ampr Emr	30
pMAD-ΔsrrA	Recombinant plasmid	This study
pCN51	Shuttle vector; Ampr Emr	32
pCN51-srrAB	The srrAB gene was cloned into pCN51	This study
pRAB11	Shuttle vector; Ampr Cmr	31
pRAB11-srrA	The srrA gene was cloned into pRAB11	This study
pRAB11-srrB	The srrB gene was cloned into pRAB11	This study
Bacterial strains
S. epidermidis
RP62A	Biofilm positive, genome sequenced, and published	26, 28
1457	Biofilm positive, clinical isolate, wild-type strain	7, 16
ΔsrrA mutant	srrA deletion, Spcr, derivative of S. epidermidis 1457	This study
ΔsrrA(pCN51-srrAB) mutant	ΔsrrA strain complemented with plasmid pCN51-srrAB	This study
ΔsrrA(pRAB11-srrA) mutant	ΔsrrA strain complemented with plasmid pRAB11-srrA	This study
ΔsrrA(pRAB11-srrB) mutant	ΔsrrA strain complemented with plasmid pRAB11-srrB	This study
ΔsrrA(pCN51) mutant	ΔsrrA mutation introduced with plasmid pCN51	This study
S. aureus 4220	Restriction negative, modification positive	14, 32
E. coli
DH5α	supE44 ΔlacU169 (ϕ80dlacZΔM15) hsdR17 recA1 endA1 gyrA96 thi-1 relA1	Invitrogen
BL21(DE3)	F− ompT hsdSB(rB− mB−) gal dcm (DE3)	Invitrogen

^aKm^r, kanamycin resistance; Amp^r, ampicillin resistance; Cm^r, chloramphenicol resistance; Em^r, erythromycin resistance; Spc^r, spectinomycin resistance.

Extraction of bacterial DNA.

Genomic DNA of S. epidermidis was extracted as described by Flamm et al. with minor modifications (29). In brief, staphylococcus cells were treated with lysostaphin (20 μg/ml; Sigma, St. Louis, MO) and proteinase K (100 μg/ml; Merck KGaA, Darmstadt, Germany) and extracted with phenol-chloroform, and the nucleic acids were precipitated with ethanol.

Plasmid DNA from E. coli was extracted with a plasmid purification kit (Qiagen, Hilden, Germany) according to manufacturer's instructions. After harvesting and resuspension, bacterial cells were lysed under alkaline conditions. The lysate was neutralized by the addition of potassium acetate. The cleared lysate was loaded onto a Qiagen-tip by gravity flow, and then the eluted plasmid DNA was concentrated by isopropanol precipitation. Plasmid DNA from S. epidermidis or S. aureus 4220 was extracted using the same method except for an additional step of lysostaphin treatment.

Construction of S. epidermidis srrA deletion mutant and complementary strains.

We first characterized the S. epidermidis srrAB genes in SE1457 by PCR and sequencing and then compared it to that in the genome of the ATCC 35984 strain (GenBank accession number NC_002976). The srrA gene was 726 bp in length, and the srrB gene was 1,770 bp in length. The srrA deletion mutant was constructed by allelic replacement using the temperature-sensitive plasmid pMAD as described previously (30). In brief, the spectinomycin resistance cassette (spc, ∼1 kb) digested with SmaI and BamHI endonucleases (MBI Fermentas, Vilnius, Lithuania) was inserted into the pMAD plasmid, designated pMAD-spc (15). PCR products ∼0.9 kb upstream and downstream flanking the region of srrA were cloned into pMAD-spc. The recombinant plasmid was successively transferred into E. coli DH5α, S. aureus RN4220, and then into SE1457, followed by the process of allelic replacement as performed previously (15, 30). The spectinomycin-resistant and erythromycin-sensitive white colonies were screened as an srrA deletion (ΔsrrA) mutant. The 662-bp fragment (bp +4 to +665 relative to the transcription start site) of srrA, including the REC domain and the major part of Trans_reg-C domain (see Fig. S1 in the supplemental material) was replaced by spectinomycin-resistance cassette (spc, 1,029 bp). The ΔsrrA mutant was verified by PCR and DNA sequencing (see Fig. S2A in the supplemental material). In the ΔsrrA mutant, srrA expression was below the detection level, and srrB expression was downregulated to 14% of that in the wild-type strain, as detected by quantitative reverse transcription-PCR (qRT-PCR) (see Fig. S2B in the supplemental material).

For complementation of the ΔsrrA mutant, we constructed three recombinant expression plasmids containing srrA, srrB, or srrAB genes, respectively. The srrA, srrB, and srrAB genes with the associated Shine-Dalgarno sequence in SE1457 were separately amplified by PCR with the primers pRAB11-srrA-F/pRAB11-srrA-R, pRAB11-srrB-F/pRAB11-srrB-R, and pCN51-srrAB-F/pCN51-srrAB-R, respectively (the sequences are listed in Table 2). Plasmid pRAB11-srrA was constructed from pRAB11 following insertion of a fragment of srrA digested with KpnI and EcoRI, plasmid pRAB11-srrB was constructed from pRAB11 following insertion of a fragment of srrB digested with KpnI and BglII (31), and plasmid pCN51-srrAB was constructed from pCN51 following insertion of a fragment of srrAB digested with BamHI and KpnI (32). The complementary plasmids were transferred into S. aureus RN4220 and then into the ΔsrrA mutant by electroporation, yielding three complementary strains, referred to here as the ΔsrrA(pRAB11-srrA), ΔsrrA(pRAB11-srrB), or ΔsrrA(pCN51-srrAB) strain, respectively. The vector plasmids, pRAB11 or pCN51, were introduced as blank controls into the ΔsrrA mutant and referred to here as the ΔsrrA(pRAB11) or ΔsrrA(pCN51) strain.

TABLE 2.

Primers used in this study^a

Method and primer	Sequence (5′–3′)b	Location (bp)c	Restriction enzyme	Product size (bp)
Construction and identification of srrA deletion mutant
srrA-US-F	GAAGATCTGGAGAGTCAAATGAGTAAAGAAC	1103298–1103320	BglII	884
srrA-US-R	TCCCCCGGGCATACTTTCTACTACCTCCTACA	1102437–1102459	SmaI	884
srrA-DS-F	CGGAATTCAACTGTGTGGGGTGTCGG	1101757–1101774	EcoRI	907
srrA-DS-R	CGGGATCCTCATCAGCCATCTTGTTCG	1100868–1100886	BamHI	907
spc-F	TGGTTCAGCAGTAAATGGTGG			1,029
spc-R	CATCTGTGGTATGGCGGGTA			1,029
srrA, srrB, and srrAB complementation
pCN51-srrAB-F	CGCGGATCCCCTTTGAGTCACTCAATAAC	1102500–1102519	BamHI	2,572
pCN51-srrAB-R	CGGGGTACCTGATACTTTTCAGTTTCTAA	1099948–1099967	KpnI	2,572
pRAB11-srrA-F	GGGGTACCCCTTTGAGTCACTCAATAAC	1102500–1102519	KpnI	806
pRAB11-srrA-R	CGGAATTCCTATTTAGTCGGTTCATCAC	1101714–1101733	EcoRI	806
pRAB11-srrB-F	CGGGGTACCAATGATACAAACTGTGTGG	1101765–1101783	KpnI	1,836
pRAB11-srrB-R	GGAAGATCTTGATACTTTTCAGTTTCTAA	1099948–1099967	BglII	1,836
eDNA quantification using qPCR
gyrB-F	GCTGGACAGATACAAGTT	2611681–2611698		137
gyrB-R	GCTAATGCCTCGTCAATA	2611562–2611579		137
serp0306-F	ATGCCACATCCACGAAAGA	309331–309349		179
serp0306-R	TGTAACTGACAATGCCCAATC	309489–309509		179
lysA-F	TGACAATGGGAGGTACAAGC	988594–988613		76
lysA-R	TGGTCTTCATCGTAAACAATCG	988648–988669		76
leuA-F	GTGAACGGTATTGGTGAAAGAG	1708472–1708493		78
leuA-R	GTGGTCCTTCCTTACATATAAAGC	1708526–1708549		78
SrrA expression
pET-28a-srrA-F	CCGGAATTCATGACTAACGAAATTTTAATCGTTG	1102415–1102439	EcoRI	723
pET-28a-srrA-R	CCGCTCGAGTTTAGTCGGTTCATCACTAGGTT	1101717–1101739	XhoI	723
Amplification of promoter fragments
Psrr-F	ACTTTCTACTACCTCCTA	1102440–1102457		132
Psrr-R	CACCAAAAAGATGTAATT	1102554–1102571		132
PicaA-F	GTATAACAACATTCTATT	2334137–2334154		85
PicaA-R	ATTTTTTCACCTACCTTT	2334204–2334221		85
PicaR-F	ATTCTAAAATCTCCCCCT	2334055–2334072		82
PicaR-R	TGAAACAGTAATATTTGT	2334119–2334136		82
PqoxB-F	TTTTTGACCTCCTAATAC	641914–641931		145
PqoxB-R	AATCTTACAAACCCCGTC	642041–642058		145
PpflB-F	ACTCTCCGCCTCCATTTC	2414402–2414419		151
PpflB-R	TTTATTCACAAACTGTTA	2414535–2414552		151
PrsbU-F	GAAATGCGCCTCCTTACT	1725459–1725476		147
PrsbU-R	GCTTTAGGTTATCCATTC	1725588–1725605		147
PsarA-F	GACACTTTCGTATTTTCATAAGA	279934–279956		160
PsarA-R	ATTAATGAAACCTCCCTATTTA	279797–279818		160
PrpsJ-F	AAGATTCTCGTGAACAATTC	1862226–1862245		119
PrpsJ-R	GATGTCTACACCTGATGG	1862127–1862144		119

^aThe primers were designed using Primer Premier 5 software according to the genomic sequence of S. epidermidis RP62A (GenBank accession number NC_002976).

^bRestriction sites are indicated by underlining.

^cThe locations of primers are indicated according to the S. epidermidis RP62A genome.

Growth curves of SE1457 isogenic srrA mutants.

The growth curves of S. epidermidis strains were determined by measuring the optical density (OD₆₀₀) (15). For oxic growth, overnight bacterial cultures were inoculated into flasks, incubated at 37°C with shaking at 220 rpm, and the OD₆₀₀ values of the cultures were measured at 60-min intervals for 12 h. For microaerobic growth, the cultures were transferred into screw-top 50-ml syringes which were completely filled with medium (with no air bubbles), followed by incubation under vigorous agitation (220 rpm) at 37°C. Then, 200 μl of bacteria in the syringe was removed by syringe every hour until 18 h, and the OD₆₀₀ values were measured.

Semiquantitative detection of biofilm formation of SE1457 isogenic srrA mutants in vitro.

The biofilm-forming ability of S. epidermidis strains in vitro was determined by semiquantitative plate assay (14). In brief, overnight cultures of SE1457, ΔsrrA, ΔsrrA(pCN51-srrAB), ΔsrrA(pRAB11-srrA), and ΔsrrA(pCN51) strains were diluted with TSB medium containing 0.5% glucose, inoculated into a polystyrene 96-well microplate (Corning Inc., NY), and incubated statically at 37°C for 6, 12, 24, or 48 h under oxic conditions or for 12, 24, 48, or 72 h under microaerobic conditions (Anaerocult C Mini). After incubation, the plates were washed with phosphate-buffered saline (PBS), fixed with methanol, and stained with 2% crystal violet. The OD₅₇₀ was measured using a spectrophotometer (DTX880; Beckman Coulter, Fullerton, CA). Three independent experiments were carried out.

Detection of biofilm formation of SE1457 isogenic srrA mutants in vivo.

The biofilm-forming ability of S. epidermidis strains in vivo was determined by using a New Zealand rabbit subcutaneous foreign body infection model as described by He et al. with minor modifications (33). Disks were cut from polyethylene 96-well plates (8-mm diameter, 1-mm thickness, with a 2-mm projecting rim or chimb), sterilized with 75% ethanol, washed with sterile distilled water, and then disinfected by using UV light. The rabbit (2.0 to 2.5 kg, female) was anesthetized with pentobarbital sodium (5 mg/kg, administered intravenously), and four incisions (10 mm) were made on the back bilaterally along the spine after removal of the fur. The subcutis was then carefully dissected to form a 2-cm-by-3-cm cavity. After three disks were implanted into each cavity, 1 ml of bacteria (∼10⁸ CFU) suspended in fresh TSB was injected into the cavity. The same volume of TSB was injected as a control. To minimize the effect of between-animal variation, SE1457, ΔsrrA, and ΔsrrA(pCN51-srrAB) strains were separately injected into cavities of the same rabbit.

At 72 h after bacterial inoculation, the rabbits were euthanized, and the implants were taken out, washed with PBS, and observed by using scanning electron microscopy (SEM). The biofilms were scraped from the disks, and the viable bacteria were determined by CFU counting as previously described (16, 33). Five independent experiments were carried out.

Initial adherence capacity of SE1457 isogenic srrA mutants.

Primary attachment of SE1457 isogenic srrA mutant strains to a polystyrene surface was assessed as described previously (7, 15, 34), with a modification. Briefly, overnight cultures of the SE1457, ΔsrrA, ΔsrrA(pCN51-srrAB), and ΔsrrA(pCN51) strains were inoculated into TSB and cultured at 37°C. After growth to an OD₆₀₀ of 0.6 to 0.8, the bacteria were adjusted to an OD₆₀₀ of 0.1 with PBS and inoculated into six-well plates (2 ml/well; Nunc, Roskilde, Denmark). After incubation at 37°C for 2 h, the plates were washed gently with PBS and observed under microcopy using a 40-fold objective lens. The numbers of attached cells in photomicrographs (at least five microscopic fields per sample) were counted by using ImageJ software. In addition, the adhesion capacity of SE1457 isogenic srrA mutant strains was determined by crystal violet staining. Staphylococcal strains grown at 37°C to an OD₆₀₀ of 1.0 were pipetted into a 96-well microplate (200 μl/well) and incubated at 37°C for 2 h, followed by washing with PBS, and the subsequent procedures were the same as those used for the semiquantitative biofilm formation assay measuring the OD₅₇₀ using a spectrophotometer. Three independent experiments were carried out.

Assay of PIA in biofilms of SE1457 isogenic srrA mutants.

Polysaccharide intercellular adhesion (PIA) in the biofilms of SE1457 isogenic srrA mutant strains was semiquantified by dot blot assay with wheat germ agglutinin-horseradish peroxidase (WGA-HRP) conjugate as described by Wu et al. (16) and Gerke et al. (35). In brief, overnight cultures of S. epidermidis strains were inoculated into a six-well plates (Nunc) and incubated at 37°C for 24 h under both oxic and microaerobic conditions (Anaerocult C Mini). Biofilms were scraped off from the bottom of the wells, resuspended in 0.5 M EDTA (3 μl per 1 mg [wet weight]), and centrifuged (13,000 × g, 5 min) after heating at 100°C for 5 min. The supernatant was treated with proteinase K (20 mg/ml) at 37°C for 3 h and inactivated at 100°C for 5 min. Serial dilutions of the PIA extract were transferred to a nitrocellulose membrane (Millipore, Billerica, MA) using a 96-well dot blot device (Biometra GmbH, Gottingen, Germany). The air-dried membrane was blocked with 5% (wt/vol) skim milk and subsequently incubated with WGA (3.2 μg/ml) conjugated for 1 h with HRP (Lectinotest Laboratory, Lviv, Ukraine). The HRP activity was visualized by chromogenic detection using 4-chloride-1-naphthol (Sigma) as the substrate. The quantitation (titer) of the PIA was represented as the highest dilution of the supernatant detectable.

Detection of Aap.

Accumulation-associated protein (Aap) expression of SE1457 isogenic srrA mutant strains was determined by Western blotting with an Aap-specific monoclonal antibody (MAb_18B6) made in our laboratory (26). In brief, 24-h biofilm and 12-h planktonic cells of S. epidermidis strains were collected and adjusted to an identical OD₆₀₀ after being washed with PBS. The bacteria were treated with lysostaphin (Sigma) and centrifuged (20,000 × g) at 4°C for 30 min. The supernatants were separated using SDS-PAGE (7%) and blotted onto polyvinylidene fluoride membrane (0.45 μm; Millipore) by electrotransfer. The membrane was incubated with MAb_18B6 (10 ng/ml) and then with goat anti-mouse IgG conjugated with HRP (Santa Cruz, Santa Cruz, CA) and then visualized by using an ECL Western blotting system (Thermo Fisher Scientific, Waltham, MA).

Quantification of extracellular DNA.

The isolation of extracellular DNA (eDNA) from biofilms was performed as described previously (15, 26). In brief, the 24-h biofilms cultured in a 96-well polystyrene plate were chilled at 4°C for 1 h, and EDTA was added at a final concentration of 2.5 mM. After measurement of the OD₆₀₀ of unwashed biofilm (biofilm biomass), eDNA extraction solution (50 mM Tris-HCl, 10 mM ETDA, 500 mM NaCl [pH 8.0]) was added to the wells. The biofilms were scraped off and centrifuged (13,000 × g) for 5 min at 4°C. The eDNA in the supernatant was extracted with phenol-chloroform-isoamyl alcohol (25:24:1), precipitated with 100% alcohol, and resuspended in Tris-EDTA buffer.

The amount of eDNA was quantified by qPCR with SYBR Premix ExTaq (TaKaRa Bio, Inc., Shiga, Japan) using the gyrB (gyrase B), serp0306 (ferrichrome transport ATP-binding protein A), leuA (2-isopropylmalate synthase), and lysA (diaminopimelate decarboxylase A) primers listed in Table 2. Each gene in the qPCR was assayed in triplicate for three independent experiments. The relative quantitation of eDNA in each sample was calculated as total the eDNA (in ng) divided by the biofilm biomass (OD₆₀₀).

Observation of biofilms with CLSM and SEM.

For observation of bacterial biofilms under confocal laser scanning microscopy (CLSM; Leica TCS SP5, Mannheim, Germany), overnight cultures of SE1457, ΔsrrA, and ΔsrrA(pCN51-srrAB) strains were inoculated into Fluorodishes (2 ml/dish; FD35-100; WPI, Sarasota, FL) and incubated statically at 37°C for 24 h (7, 26). The biofilms on the dishes were then rinsed gently with 0.85% NaCl and observed using CLSM with SYTO9 and propidium iodide (PI) staining (Live/Dead kit; Invitrogen, Carlsbad, CA). The z-stack composite confocal photomicrographs of viable cells (green) and dead cells (red) were generated using Leica LAS AF software. The fluorescence of each stack was quantified using ImageJ software. At least three independent experiments were carried out.

For observation of bacterial biofilms under SEM (JSM-6700F; JEOL, Tokyo, Japan), staphylococcal SE1457, ΔsrrA, and ΔsrrA(pCN51-srrAB) strains were cultured in a six-well plate (35-mm diameter) with three sterile disks (8-mm diameter, 1-mm thickness, with a 2-mm chimb) in each well. After 24 h of incubation at 37°C, the disks were rinsed with PBS, fixed with 2.5% glutaraldehyde in PBS, vacuum dried for 72 h, sputtered with platinum, and then observed under a field emission source instrument.

RNA isolation and microarray analysis.

Total RNA was isolated by using an RNeasy minikit (Qiagen) according to the manufacturer's instructions. In brief, bacterial cultures in a flask for oxic conditions or in 50-ml syringes for microaerobic condition were harvested after 6 h of incubation at 37°C with shaking. The cell pellets were washed with ice-cold 0.85% NaCl and then homogenized using 0.1-mm zirconia-silica beads in a Mini-BeadBeater (Biospec, Bartlesville, OK) at a speed of 4,800 rpm for 40 s as a cycle for five times with 1-min intervals on ice in each cycle. The RNA eluted from the silica-based filter was extracted with phenol-chloroform-isoamyl alcohol and precipitated with ethanol. The quantity and quality of the total RNA were assessed by using a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE) and gel electrophoresis.

Microarray analysis was performed by in situ synthesis of 14,527, 60-mer-long oligonucleotide probes (Agilent, Palo Alto, CA), as previously described by Lou et al. (15) and Charbonnier et al. (36), which cover more than 95% of the open reading frames (ORFs) annotated in S. epidermidis strains ATCC 12228 (accession number NC_004461) and ATCC 35984 (accession number NC_002976). Total RNAs (10 μg) from SE1457 were labeled by cyanine-3 dCTP (green Cy3; Perkin-Elmer Life Sciences, Boston, MA) using the SuperScript II (Invitrogen, Basel, Switzerland). Purified genomic DNA from the reference strains was labeled with cyanine-5 dCTP (red Cy5) and used for microarray normalization. Mixtures of Cy5-labeled DNA and Cy3-labeled cDNA were hybridized and scanned in a dedicated oven as previously described (37). The fluorescence intensities were quantified using Agilent Feature Extraction software (version 8). The data were normalized to baseline using red channel data as a control. A false discovery rate value of 5% (P value cutoff, 0.05) was used for variance analysis of three biological replicates, and an arbitrary threshold of 3.0-fold was used for defining significant differences in expression ratios.

qRT-PCR.

The RNA extracted from bacterial strains was treated with DNase I and reverse transcribed into cDNA using iScript reverse transcriptase (Bio-Rad, Hercules, CA) by incubation for 5 min at 25°C, followed by 30 min at 42°C and 5 min at 85°C. Next, qPCRs were performed using SYBR green PCR reagents (Premix EX Taq; TaKaRa Biotechnology, Dalian, China) in the MasterCycler Realplex system (Eppendorf AG, Hamburg, Germany), with the amplification conditions as 95°C for 30 s, followed by 40 cycles of 95°C for 5 s and 60°C for 34 s, followed in turn by melting-curve analysis. A housekeeping gene gyrB (DNA gyrase subunit B) was used to normalize the transcript levels of genes in the qPCRs. All qRT-PCRs were carried out in triplicate with at least three independent RNA samples. The sequences of the primers were designed using Beacon designer software (Premier Biosoft International, Ltd., Palo Alto, CA) and are listed in Table 3.

TABLE 3.

Primers used for transcriptional analysis by qRT-PCR^a

Primer	Sequence (5′–3′)	Location (bp)b	Product size (bp)
sarA-F	GTAATGAACACGATGAAAGAACT	279526–279548	103
sarA-R	GCTTCTGTGATACGGTTGT	279446–279464	103
rsbU-F	GCTTATGGACATTCACAA	1724925–1724942	121
rsbU-R	GATTCATCTCTTCATACAGT	1724822–1724841	121
icaA-F	ATCAAGCGAAGTCAATCTC	2334781–2334799	127
icaA-R	CAGCAATATCCTCAGTAATCA	2334887–2334907	127
icaR-F	GCACATCGCTTTGGATAA	2333646–2333663	146
icaR-R	TTAACAGTGAATATACTTGGTCTT	2333518–2333541	146
atlE-F	CAATTACAGGAGACACAAT	631207–631225	149
atlE-R	TCATTATCATTAGAAGCAGTT	631077–631097	149
aap-F	CGAGGAATTACAATCATCACA	2460758–2460778	166
aap-R	CGTAGTTGGCGGTATATCTA	2460613–2460632	166
srrA-F	TCACCTAGAGAAGTAGTATT	1102108–1102127	130
srrA-R	GAGCGTCATTATCAATCA	1101998–1102015	130
srrB-F	TCCATAGTAGACGGTATAGT	1100559–1100578	136
srrB-R	ATAATCCTTCAGCATCCATA	1100443–1100462	136
ctaA-F	CTACGATTATTATGACCTT	703606–703624	146
ctaA-R	ACTCAGTTCTATAATTGTT	703479–703497	146
qoxB-F	TCTATGGATACAATGACAAGTT	641416–641437	126
qoxB-R	TGAGTTACGACCTCTGAA	641312–641329	126
serp0257-F	AACCTGGAGAAGCATTAG	265234–265251	101
serp0257-R	TAGCGTTACACCTGTTAC	265317–265334	101
serp2257-F	AGGTAATGCTGGCTTATCT	2286255–2286273	110
serp2257-R	CGAATGCGTTGACTGTAA	2286164–2286181	110
pflA-F	ACACTTACACTCCGTTGA	2412105–2412122	140
pflA-R	CTTCTCTTGATGGTTCGTTA	2411983–2412002	140
serp2381-F	AGAAGGTAATCAAGTTGT	2428977–2428994	136
serp2381-R	CGTATTATATTGTTGTAGCA	2428859–2428878	136
lacA-F	GGAAGACAACGATTATGAT	1841284–1841302	135
lacA-R	GCACCATAGGCATCTATA	1841168–1841185	135
ureF-F	TTAGGTGTAGATGTGGAAT	1898186–1898204	148
ureF-R	CGTGTCTTCTCAATATGG	1898317–1898334	148
rbsK-F	GCAGGTATTCATACACAAT	2125633–2125651	150
rbsK-R	CACACTCATCTCAACATC	2125502–2125519	150
betB-F	TATCCATTACTTCAAGCATCT	2209851–2209871	128
betB-R	CCAACTTCCTCCATCAAT	2209744–2209761	128
cysH-F	TTGGTGCTGAGAGTATGG	2227130–2227147	148
cysH-R	TTAATGCGTAATTGCGGATAT	2227000–2227020	148
rplB-F	AAGATGGAATCATTGCTAA	1860237–1860255	130
rplB-R	TGACCTACTTGTAATCCT	1860126–1860143	130
rpsJ-F	AAGATTCTCGTGAACAATTC	1862226–1862245	119
rpsJ-R	GATGTCTACACCTGATGG	1862127–1862144	119
opp1B-F	TGATTCCATTATTGATTGTAGTGA	2419641–2419664	111
opp1B-R	GCGTTATATTAGGCGTTCC	2419554–2419572	111
rpoA-F	TGAAGTTAGTGAAGATGCTA	1849776–1849795	113
rpoA-R	CTGGTAATGAAGATAGTAGGA	1849683–1849703	113
nrdD-F	GATAGTAATACATTCTCAACAA	2217415–2217436	145
nrdD-R	ATGGATGGTAATCTAAGTC	2217292–2217310	145

^aPrimers used in qRT-PCR were designed with Beacon Designer 7 software according to the genomic sequence of S. epidermidis RP62A (GenBank accession number NC_002976).

^bThe locations of primers are indicated according to the S. epidermidis RP62A genome.

Expression and purification of recombinant SrrA.

To determine the DNA-binding properties of SrrA, a recombinant expression plasmid (pET28a-srrA) was constructed by inserting the srrA fragment amplified from SE1457 with the primers pET-28a-srrA-F/pET-28a-srrA-R (listed in Table 2) into vector pET28a(+) and transferred into E. coli BL21(DE3). When bacteria were grown to an OD₆₀₀ of 0.6 at 37°C, 0.8 mM IPTG (isopropyl-β-d-thiogalactopyranoside) was added for overnight incubation at 22°C. The cells resuspended in lysis buffer (50 mM NaH₂PO₄ [pH 8.0], 300 mM NaCl, 0.1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride) were sonicated and centrifuged at 15,000 × g for 30 min, and the supernatants were loaded onto a nickel-nitrilotriacetic acid column (Qiagen). His-tagged SrrA was eluted using a linear gradient of 20 to 300 mM imidazole and enriched by ultrafiltration, and the protein concentration was determined by using a Bradford protein quantification kit (Tiangen, Beijing, China).

EMSA.

To determine the interaction between SrrA and the promoter regions of putative target genes, electrophoresis mobility shift assay (EMSA) was carried out using a digoxigenin gel shift kit (Roche Diagnostics GmbH, Mannheim, Germany) according to the manufacturer's instructions. In brief, the predicted promoter regions of icaA, icaR (∼80-bp fragments), srrAB, qoxB, pflB, sarA, and rsbU (∼140-bp fragments) were amplified by PCR with the primers listed in Table 2. The DNA fragments were purified by using a gel extraction kit (Qiagen) and labeled with digoxigenin using terminal transferase. Purified His-tagged SrrA was phosphorylated (SrrA-P) by incubation with 50 mM acetyl phosphate (Sigma) for 1 h at room temperature. Each gel shift assay included the probe labeled with digoxigenin plus increasing concentrations of SrrA-P (ranging from 1.2 to 0.3 μM in 2-fold dilutions); a 125-fold molar excess of unlabeled probe as a competitor was also added to the labeled probe plus 1.2 μM SrrA-P with a labeled probe as a control. The 119-bp coding sequence of S. epidermidis rpsJ was designated a negative control for SrrA-DNA binding. All samples were incubated at 25°C for 20 min, separated by electrophoresis on 6% nondenaturing polyacrylamide gel, and blotted onto a positively charged nylon membrane (Millipore). The blots were incubated with alkaline phosphatase conjugated anti-digoxigenin antibody, followed by chloro-5-substituted adamantyl-1,2-dioxetane phosphate (CSPD) solution for chemiluminescent detection, and then exposed to X-ray film.

Statistical analysis.

Data from the biofilm assay, the initial attachment assay, and the CFU enumeration assay were analyzed by the GraphPad Prism program (San Diego, CA) using the Student t test. Differences with a P value of <0.05 were considered statistically significant.

Microarray data accession number.

The complete microarray data set is posted in the Gene Expression Omnibus database (http://www.ncbi.nlm.nih.gov/geo/) under accession numbers GPL13532 for the platform design and GSE47101 for the original data set.

RESULTS

The ΔsrrA mutant displayed growth defects under oxic and microaerobic conditions.

To assess whether srrAB expression responds to oxygen stress, the transcription of srrAB in SE1457 was analyzed by qRT-PCR during the shift from oxic to microaerobic conditions. Indeed, both srrA and srrB expressions were upregulated (a 2- or 3-fold increase) under microaerobic conditions, whereas the expression of arlRS monitored as a control showed no obvious change (Fig. 1A), indicating that SrrAB displayed function differently under oxic and microaerobic conditions.

FIG 1.

Effect of srrA deletion on the S. epidermidis growth. (A) Transcriptional levels of srrA and srrB in the S. epidermidis strain 1457 under microaerobic stress. After culture for 4 h under oxic conditions, SE1457 was transferred into a 50-ml syringe (sealed entirely with no bubbles inside) for 0.5 or 1 h of incubation under microaerobic conditions (−O₂). Bacterial cells were collected, and the total RNA was extracted. The relative expression levels of srrA and srrB were analyzed by qRT-PCR in comparison to the transcription level of gyrB (housekeeping gene). The two-component regulatory system (TCS) arlRS was used as a control. Data are represented as means ± the standard deviations (SD) from three independent experiments. (B) The srrA deletion mutant displayed a defect in growth of S. epidermidis. SE1457 and its isogenic srrA mutant strains were streaked on TSA plates and incubated at 37°C for 24 h under oxic conditions (+O₂) or under microaerobic conditions (−O₂). The results represent one of three independent experiments. (C) Growth curves of SE1457 isogenic srrA mutants. Under oxic conditions (+O₂), S. epidermidis strains were cultured in a flask (1:8 culture-to-flask volume ratio) at 37°C with shaking. For microaerobic conditions (−O₂), cultures were inoculated into 50-ml syringes, all air bubbles were removed, and the syringes were incubated at 37°C with shaking. The cultures were measured hourly at OD₆₀₀. The experiments were repeated at least three times, and a representative set of growth curves is shown.

The S. aureus (WCUH29) srrA mutant displayed a marked reduction in growth under anaerobic conditions, whereas it showed no differences in colony size or growth rate under oxic conditions compared to the parent strain (19). In contrast, we found here that under oxic conditions the SE1457 ΔsrrA strain formed smaller colonies than parent strain and that, under microaerobic conditions, no ΔsrrA colonies were evident on TSA plates even after 24 h of incubation (Fig. 1B).

In liquid medium under oxic conditions, SE1457 entered early log phase (OD₆₀₀ of 0.65) by 4 h after inoculation, but the ΔsrrA mutant took about 6 h to reach a similar growth level (OD₆₀₀ of 0.74). The growth of the srrAB complementation strain ΔsrrA(pCN51-srrAB) was restored to the a wild-type level, whereas the growth of the srrA complementation strain ΔsrrA(pRAB11-srrA) was partially recovered. Under microaerobic conditions, to reach an OD₆₀₀ of 1.0 the ΔsrrA mutant needed 15 h of incubation, whereas the parent strain needed only 7 h. The ΔsrrA(pCN51-srrAB) and ΔsrrA(pRAB11-srrA) complementation strains required 6 or 9 h of incubation, respectively, to reach an OD₆₀₀ of 1.0 (Fig. 1C). Under either oxic or microaerobic conditions, the transformation of pCN51 or pRAB11-srrB had no effect on the growth of the ΔsrrA mutant.

Deletion of srrA impaired biofilm development in vitro.

The impact of the srrA deletion on the biofilm formation of S. epidermidis in vitro was investigated by a semiquantitative microplate assay. Bacterial biofilm formation was monitored at 6, 12, 24, and 48 h under oxic conditions or at 12, 24, 48, and 72 h under microaerobic conditions.

Under oxic conditions, the ΔsrrA mutant also produced less biofilm than did the parent strain at these time points, whereas the biofilm of the ΔsrrA mutant (OD₅₇₀ = 0.61 ± 0.03) was dramatically decreased compared to the wild-type strain (OD₅₇₀ = 2.65 ± 0.08) (Fig. 2A and B). Under microaerobic conditions, no biofilm formation was observed in the ΔsrrA mutant at 12 h or 24 h. After 48 h of incubation, biofilm produced by the ΔsrrA mutant had an OD₅₇₀ of 0.25 ± 0.02, which is significantly less than that of its wild-type counterpart (OD₅₇₀ = 0.93 ± 0.04) (Fig. 2A and C).

FIG 2.

Effect of srrA deletion on the in vitro biofilm formation of S. epidermidis. Overnight cultures of the S. epidermidis strains were 1:200 diluted with TSB medium and inoculated into 96-well polystyrene plates in triplicate. (A) After static incubation for 6, 12, 24, and 48 h under oxic conditions (+O₂) or for 12, 24, 48, and 72 h under microaerobic conditions (−O₂), biofilms were stained with crystal violet and observed. (B and C) Biofilm formation under oxic and microaerobic conditions was detected at OD₅₇₀. The experiments were repeated three times, and the data represent means ± the SD. **, P < 0.01 (ΔsrrA mutant versus SE1457).

Under both oxic and microaerobic conditions, the biofilm-forming ability was restored in the ΔsrrA(pCN51-srrAB) complementation strain (OD₅₇₀ = 1.93 ± 0.14 and 0.95 ± 0.11, respectively) and partially restored in the ΔsrrA(pRAB11-srrA) strain (OD₅₇₀ = 1.13 ± 0.03 and 0.67 ± 0.02, respectively). Transformation of the vector alone had no effect on ΔsrrA strain biofilm formation.

When biofilms of the SE1457, ΔsrrA, and ΔsrrA(pCN51-srrAB) strains that had been cultured under oxic conditions were observed under CLSM with Live/Dead staining, the thickness of the ΔsrrA mutant biofilm was much less (5.97 ± 0.34 μm) than that of the parent strain (22.2 ± 2.06 μm) (P < 0.01), and the thickness was restored by complementation with pCN51-srrAB (10.16 ± 1.51 μm). There were more dead cells in the ΔsrrA biofilm than in the wild-type counterpart (PI/total = 0.23 versus 0.072, P < 0.01), whereas fewer dead cells were observed in ΔsrrA(pCN51-srrAB) biofilm (PI/total = 0.017) (Fig. 3).

FIG 3.

Observation of S. epidermidis biofilm under CLSM. The 24-h biofilms cultured in vitro were visualized using Live/Dead viability staining under CLSM. The three-dimension structural images (zoom 1, ×63 magnification) were reconstructed, and the thickness of biofilm was measured using Imaris software. The top layer, middle layer, and bottom layer within a biofilm are shown. The viable and dead cells were stained in green (SYTO9) and red (PI), respectively. The amount of fluorescence from the bottom to the top layer of the biofilm was determined using ImageJ software (zoom 3, ×63 magnification). The value of the PI/total florescence indicates the proportion of dead cells within the biofilm. The figures represent one of three independent experiments.

Deletion of srrA abolished biofilm formation in vivo.

To determine whether srrA deletion had an impact on in vivo biofilm formation, a rabbit subcutaneous foreign body infection model was used. Staphylococcal strains (10⁸ CFU) were injected into the cavities with implanted polystyrene disks on the animal's back. After 72 h, biofilm on the disks was observed under SEM. SE1457 formed a compact biofilm covered with secreted substance on the implanted disks, while the ΔsrrA mutant formed only a few bacterial clusters. The amount of biofilm produced by the ΔsrrA(pCN51-srrAB) complementation strain was similar to that of the wild-type strain (Fig. 4A). The viable bacterial cells in the ΔsrrA biofilm formed in vivo (CFU = 8.82 × 10³) were significantly fewer than that in those of the SE1457 (CFU = 5.49 × 10⁴) (P < 0.01) and ΔsrrA(pCN51-srrAB) (CFU = 6.14 × 10⁴) strains (Fig. 4B).

FIG 4.

Effect of srrA deletion on biofilm formation by S. epidermidis in vivo. The New Zealand rabbit model of local S. epidermidis biofilm infection was used. Subcutaneous incisions were made on the backs of the animal. Sterile polyethylene disks were implanted, and then overnight bacterial cultures (10⁸ CFU) were resuspended in 1 ml of TSB and inoculated into the cavities. (A) After 72 h, the disks were removed, fixed with 2.5% glutaraldehyde, and observed using SEM. As a control, 24-h biofilms cultured in vitro were observed under SEM. (B) The biofilms were scraped from the disks, and the viable bacteria were determined by CFU counting. The data are from one of three independent experiments. **, P < 0.01 (ΔsrrA mutant versus SE1457).

Deletion of srrA diminished initial attachment of S. epidermidis.

Biofilm formation by S. epidermidis is generally a two-step process involving an initial attachment and a subsequent maturation phase. The initial attachment of the ΔsrrA mutant to polystyrene plates was determined by using ImageJ software. Attached cells of the ΔsrrA strain (about 1.35 × 10³per field) were 3.4-fold less than those of the parent strain (about 4.54 × 10³per field) and of the ΔsrrA(pCN51-srrAB) complementation strain (about 5.61 × 10³) (Fig. 5A), and this was confirmed by crystal violet staining with measurement at OD₅₇₀ (Fig. 5B).

FIG 5.

Effect of srrA deletion on initial adherence capacity in vitro. (A and B) Mid-log-phase bacterial cells were adjusted to an OD₆₀₀ of 0.1 in PBS and then added to a 6-well plate (2 ml/well) (A) or a 96-well microplate (200 μl/well) (B). The plates were then incubated at 37°C for 2 h. The attached bacterial cells on the 6-well plate were counted by microcopy (A), and attached cells on the 96-well microplate were determined at OD₅₇₀ after crystal violet staining (B). The data represent one of three independent experiments. **, P < 0.01 (ΔsrrA mutant versus SE1457).

Deletion of srrA affected the biofilm matrix production in S. epidermidis.

To investigate the effect of srrA deletion on the biofilm matrix production, the release of PIA, eDNA, and Aap were determined in the SE1457, ΔsrrA, ΔsrrA(pCN51-srrAB), and ΔsrrA(pCN51) strains. PIA, a major factor affecting biofilm accumulation, was detected semiquantitatively with WGA-HRP. Under both oxic and microaerobic conditions, the ΔsrrA mutant strain and the ΔsrrA(pCN51) vector control strain produced less PIA than either SE1457 or the ΔsrrA(pCN51-srrAB) complementation strain (Fig. 6A). No differences of Aap expression in biofilms or planktonic cells of the SE1457, ΔsrrA, and ΔsrrA(pCN51-srrAB) strains were detected by Western blotting with monoclonal antibody (18B6) against the Aap protein B repeat region (Fig. 6B). The relative concentrations of eDNA in 24-h biofilms of the ΔsrrA mutant and the vector control ΔsrrA(pCN51) strain were ∼5-fold higher than that of either the SE1457 parent strain (P < 0.01) or the ΔsrrA(pCN51-srrAB) complementation strain, as shown in Fig. 6C.

FIG 6.

Effect of srrA deletion on the extracellular matrix biosynthesis of S. epidermidis. (A) PIA biosynthesis semiquantified by dot blot assay with WGA. Twenty-four-hour biofilms, cultured aerobically (+O₂) or microaerobically (−O₂) were scraped off and suspended in EDTA (3 μl/1 mg [wet weight]). Serial dilutions of the PIA extractions were spotted onto nitrocellulose membranes, subsequently incubated with WGA conjugated with HRP, and visualized using chromogenic detection. (B) Aap expression of SE1457 and its isogenic srrA deletion mutants. Twenty-four-hour biofilms and 12-h planktonic bacteria were collected after being washed with PBS. Lysostaphin-treated samples with identical OD₆₀₀s were centrifuged at 20,000 × g for 30 min at 4°C. The supernatant were separated using 7% SDS-PAGE, and the gel pieces that carried >130-kDa proteins were used for Western blotting (upper panel).The remaining gel pieces were stained using Coomassie blue as an endogenous reference (lower panel). MAb_18B6 (10 ng/ml) was used for the primary antibody. Immunoreactivity was detected with an ECL Western blotting system after incubation with HRP-conjugated secondary antibody. (C) eDNA quantified by qPCRs of four chromosomal loci (gyrB, serp0306, leuA, and lysA). Unwashed 24-h biofilms were measured at OD₆₀₀ in order to normalize to biofilm biomass and then used for eDNA isolation by phenol-chloroform-isoamyl alcohol extraction and ethanol precipitation. The results are shown as nanograms per biomass of eDNA (means ± the SD) and were derived from three independent experiments. **, P < 0.01 (ΔsrrA mutant versus SE1457).

In the ΔsrrA mutant the transcriptional profile was severely altered compared to the wild-type strain.

DNA microarray was used to compare the transcriptional profiles of the SE1457 and ΔsrrA strains under oxic or microaerobic conditions. Under microaerobic conditions, 230 differentially expressed genes were found, which were involved in respiratory and energy metabolism, biofilm formation, and cell wall biosynthesis, etc. Among them, 118 genes were upregulated and 112 were downregulated in the ΔsrrA mutant. The latter included genes involved in cytochrome and quinol-oxidase biosynthesis and assembly (e.g., qoxACD, ctaAB, and atpC), as well as anaerobic metabolism-related genes such as pflBA (formate acetyltransferase), nrdD (anaerobic ribonucleoside triphosphate reductase), serp0257 (alcohol dehydrogenase), serp2257 (acetoin reductase), serp2133 (d-lactate dehydrogenase), and serp2381 (NADH:flavin oxidoreductase/fumarate reductase flavoprotein subunit) (Table 4). Under oxic conditions the srrA mutation affected the expression of 51 genes; 33 genes were upregulated, and 18 genes were downregulated. Similar to findings under microaerobic conditions, downregulation was observed in respiratory and energy metabolism-related genes, including srrA, serp2324 (branched-chain alpha-keto acid dehydrogenase subunit E2), serp2327 (acetoin dehydrogenase, E3 component, dihydrolipoamide dehydrogenase), serp2379 (acetoindiacetyl reductase), ppdK (pyruvate phosphate dikinase), and nrdDG. Protein synthesis-related genes, including rpsORTD and rplNS, were upregulated (Table 5). These results were confirmed by qRT-PCR and suggested that retarded growth may result from low expression levels of the genes involved in the respiratory electron transport chain and anaerobic metabolism.

TABLE 4.

Transcription levels of genes involved in growth and biofilm formation of the S. epidermidis ΔsrrA mutant under microaerobic conditions

Gene function and ORF	Gene	Description or predicted function	Expression ratio (mutant/WT)a
			Microarray	qRT-PCR
Respiratory chain and energy metabolism
SERP1055	srrA	DNA-binding response regulator	0.10	0.00001
SERP1054	srrB	Histidine kinase sensor	UD	0.09 ± 0.02
SERP0705	ctaA	Cytochrome oxidase assembly protein	0.22	0.04 ± 0.01
SERP0706	ctaB	Protoheme IX farnesyltransferase	0.28	ND
SERP0646	qoxB	Quinol oxidase subunit II	UD	0.31 ± 0.08
SERP0645	qoxA	Quinol oxidase subunit I	0.30	ND
SERP0644	qoxC	Quinol oxidase polypeptide III	0.20	ND
SERP0643	qoxD	Quinol oxidase polypeptide IV	0.25	ND
SERP2381		NADH:flavinoxidoreductase/fumarate reductase flavoprotein subunit	0.21	0.20 ± 0.17
SERP0257		Alcohol dehydrogenase	0.32	0.08 ± 0.04
SERP2112		Alcohol dehydrogenase, zinc containing	0.28	ND
SERP2257		Acetoin reductase, oxidoreductase, ligand is NAD	0.18	0.02 ± 0.01
SERP2365	pflA	Pyruvate formate-lyase-activating enzyme	0.15	0.15 ± 0.11
SERP2366	pflB	Formateacetyltransferase	0.31	ND
SERP2183	nrdD	Anaerobic ribonucleoside triphosphate reductase	0.33	ND
SERP1795	lacA	Galactose-6-phosphate isomerase LacA subunit	0.32	1.72 ± 0.37
SERP1793	lacC	Tagatose-6-phosphate kinase	0.30	ND
SERP1791	lacF	PTS system, lactose-specific IIA component	0.31	ND
SERP1873	ureF	Urease accessory protein UreF	0.34	1.38 ± 0.6
SERP1874	ureG	Urease accessory protein UreG	0.33	ND
SERP1875	ureD	Urease accessory protein UreD	0.33	ND
SERP2100	rbsK	Ribokinase, catalyzes the phosphorylation of ribose to ribose-5-phosphate using ATP, this reaction is the first step in the ribose metabolism	0.16	0.14 ± 0.06
SERP2101		d-Ribose pyranase	0.27	ND
SERP2102	rbsU	Ribose transporter RbsU	0.30	ND
SERP2347	bioB	Biotin synthase	0.32	0.44 ± 0.20
SERP2396	bioD	Dethiobiotinsynthetase	0.34	ND
SERP2190	cysI	Sulfite reductase (NADPH) hemoprotein beta-component	3.68	ND
SERP2191	cysJ	Sulfite reductase (NADPH) flavoprotein	4.33	ND
SERP2192	cysH	Phosophoadenylyl-sulfate reductase	5.22	ND
SERP2176	betA	Choline dehydrogenase	3.84	ND
SERP2177	betB	Betaine aldehyde dehydrogenase, ligand is NAD	4.25	7.29 ± 2.56
Biofilm formation
SERP2292	icaR	Intercellular adhesion regulator	0.33	0.32 ± 0.07
SERP2293	icaA	N-Glycosyltransferase	UD	0.10 ± 0.02
SERP2398	aap	Accumulation associated protein	UD	0.89 ± 0.21
SERP0274	sarA	Accessory regulator A	UD	0.87 ± 0.33
SERP1680	rsbU	Sigma factor B regulator protein	UD	1.35 ± 0.26
Protein synthesis
SERP0044	rpsF	30S ribosomal protein S6	3.8	ND
SERP1832	rpsJ	30S ribosomal protein S10	5.09	6.85 ± 2.06
SERP0186	rpsL	30S ribosomal protein S12	3.04	ND
SERP1807	rpsM	30S ribosomal protein S13	3.34	ND
SERP1818	rpsN	30S ribosomal protein S14	5.58	ND
SERP1822	rpsQ	30S ribosomal protein S17	3.11	ND
SERP1828	rplB	50S ribosomal protein L2	4.17	17.1 ± 3.22
SERP1831	rplC	50S ribosomal protein L3	4.77	ND
SERP1821	rplN	50S ribosomal protein L14	3.09	ND
SERP1824	rplP	50S ribosomal protein L16	3.41	ND
SERP1815	rplR	50S ribosomal protein L18	4.37	ND
SERP1826	rplV	50S ribosomal protein L22	3.4	ND
SERP1820	rplX	50S ribosomal protein L24	3.11	ND
SERP2371	opp-1B	Peptide ABC transporter, permease protein 1B	0.19	0.25 ± 0.15
SERP2370	opp-1C	Peptide ABC transporter, permease protein 1C	0.26	ND
Transcription
SERP1805	rpoA	DNA-directed RNA polymerase subunit alpha	3.24	3.54 ± 2.28
SERP0183	rpoB	DNA-directed RNA polymerase subunit beta	2.96	ND
SERP0184	rpoC	DNA-directed RNA polymerase subunit beta′	3.53	ND
SERP1127	rpoD	RNA polymerase sigma factor	4.07	ND
SERP1677	rpoF	RNA polymerase sigma factor SigB	3.2	ND

^aWT, wild type; UD, under the detection level in the microarray analysis; ND, not done. qRT-PCR data are given as means ± standard deviations of results from three independent experiments.

TABLE 5.

Transcription levels of genes involved in growth and biofilm formation of the S. epidermidis ΔsrrA strain under oxic conditions

Gene function and ORF	Gene	Description or predicted function	Expression ratio (mutant/WT)a
			Microarray	qRT-PCR
Respiratory chain and energy metabolism
SERP1055	srrA	DNA-binding response regulator	0.30	0.00001
SERP1054	srrB	Histidine kinase sensor	UD	0.10 ± 0.02
SERP0705	ctaA	Cytochrome oxidase assembly protein	UD	0.18 ± 0.08
SERP0646	qoxB	Quinol oxidase subunit II	UD	0.17 ± 0.05
SERP2381		NADH:flavinoxidoreductase/fumarate reductase flavoprotein subunit	UD	0.35 ± 0.21
SERP0257		Alcohol dehydrogenase	UD	0.16 ± 0.07
SERP2257		Acetoin reductase, oxidoreductase, ligand is NAD	UD	0.13 ± 0.07
SERP2327		Acetoin dehydrogenase, E3 component, dihydrolipoamide dehydrogenase	0.31	ND
SERP2379		Acetoin (diacetyl) reductase	0.28	ND
SERP2324		Branched-chain alpha-keto acid dehydrogenase subunit E2	0.32	ND
SERP2365	pflA	Pyruvate formate-lyase-activating enzyme	UD	0.12 ± 0.05
SERP2170	ppdK	Pyruvate phosphate dikinase	0.23	ND
SERP2182	nrdG	Anaerobic ribonucleoside-triphosphate reductase activating protein	0.33	ND
SERP2183	nrdD	Anaerobic ribonucleoside triphosphate reductase	0.30	ND
SERP1795	lacA	Galactose-6-phosphate isomerase LacA subunit	UD	0.39 ± 0.12
SERP1873	ureF	Urease accessory protein UreF	UD	0.27 ± 0.15
SERP2192	cysH	Phosophoadenylyl-sulfate reductase	UD	5.26 ± 1.36
Biofilm formation
SERP2292	icaR	Intercellular adhesion regulator	UD	5.14 ± 0.63
SERP2293	icaA	N-Glycosyltransferase	UD	0.13 ± 0.06
SERP2295	icaB	Intercellular adhesion protein B	UD	0.17 ± 0.1
SERP2294	icaD	Intercellular adhesion protein D	UD	ND
SERP0636	atlE	Bifunctional autolysin	UD	0.25 ± 0.12
SERP2398	aap	Accumulation-associated protein	UD	0.68 ± 0.14
SERP0274	sarA	Accessory regulator A	UD	0.74 ± 0.21
SERP1680	rsbU	Sigma factor B regulator protein	UD	0.93 ± 0.29
Protein synthesis
SERP0840	rpsO	30S ribosomal protein S15	3.16	ND
SERP0046	rpsR	30S ribosomal protein S18	3.25	ND
SERP1153	rpsT	30S ribosomal protein S20	4.56	ND
SERP1284	rpsD	30S ribosomal protein S4	4.39	ND
SERP1821	rplN	50S ribosomal protein L14	3.88	ND
SERP0807	rplS	50S ribosomal protein L19	3.69	ND
SERP0001	rpmH	50S ribosomal protein L34	3.54	ND
SERP2371	opp-1B	Peptide ABC transporter, permease protein 1B	0.34	1.01 ± 0.55
SERP2105	pgsA	Poly-gamma-glutamate synthesis protein PgsA	3.39	ND
SERP2107	pgsB	Poly-gamma-glutamate synthesis protein PgsB	3.51	ND
Transcription
SERP0926	parC	DNA topoisomerase IV subunit A	2.96	ND
SERP0838	truB	tRNA pseudouridine synthase B	3.10	ND
SERP1307	trmB	tRNA [guanine-N(7)-]-methyltransferase	2.95	ND

^aWT, wild type; UD, under the detection level in the microarray analysis; ND, not done. qRT-PCR data are given as means ± standard deviations of results from three independent experiments.

Transcription levels of biofilm-related genes were also confirmed by qRT-PCR. Compared to SE1457, icaR in the ΔsrrA mutant was downregulated 3.1-fold under microaerobic conditions but upregulated 5.1-fold under oxic conditions. The transcription levels of icaA and atlE in the ΔsrrA mutant were downregulated about 10- and 5-fold, respectively, under microaerobic conditions, and 8- and 4-fold under oxic conditions. In addition, no significant difference in the expression of rsbU, sarA, and aap was detected by qRT-PCR between SE1457 and the ΔsrrA mutant under either condition (Tables 4 and 5).

Binding of recombinant SrrA protein to the putative promoter regions.

To further study the regulation role of SrrAB in the biofilm formation, EMSA was carried out with digoxigenin-labeled putative promoter regions and recombinant SrrA (His-tagged SrrA). The 132-bp DNA fragment upstream of srr (p-srr) formed a shifted complex with phosphorylated SrrA (SrrA-P) in a dose-dependent manner but not with unphosphorylated SrrA (Fig. 7A and B, lane 2 to lane 4). The addition of a 125-fold excess of unlabeled p-srr as a specific competitor blocked SrrA-DIG-DNA complex formation (Fig. 7B, lane 5).

FIG 7.

EMSA analysis of S. epidermidis SrrA with the putative promoter regions. His-tagged SrrA was purified and phosphorylated (SrrA-P) by incubation with 50 mM acetyl phosphate. The putative promoter regions of srrAB, icaR, icaA, qoxB, pflB, sarA, and rsbU genes were PCR amplified. DNA probes were purified and labeled with digoxigenin (Dig). Gel shift reactions were performed by incubating labeled probe with increasing concentrations of SrrA-P (range, 0.3 to 1.2 μM). Lane 1 and lane 5 of each blot contained a no-protein control and a 125-fold excess of unlabeled probe competitor control, respectively. All samples were electrophoresed on a nondenaturing polyacrylamide gel and blotted onto nylon membrane. After incubation with antidigoxigenin antibody, CSPD chemiluminescent reagent was added. The arrows indicate the positions of phosphorylated SrrA-bound probes; triangles indicate the positions of free probes. The DNA fragment within the rpsJ coding region was used as a negative control.

SrrA-P resulted in a mobility shift of the 82-bp, 85-bp, 145-bp, or 151-bp fragments upstream of icaR, icaA, qoxB, or pflB, respectively (Fig. 7C, D, E, and F). SrrA-P did not bind to the fragment upstream of sarA and rsbU (Fig. 7G and H). As a negative control, a 119-bp DNA fragment of rpsJ gene did not form a complex with Srr-P under the same conditions (Fig. 7I).The results indicated that SrrA-P was able to bind specifically to the promoter regions of certain biofilm-related genes, as well as some genes involved in respiratory metabolism.

DISCUSSION

SrrAB in S. aureus modulates biofilm formation and expression of virulence factors (such as tst, spa, agr, ica, etc.) under oxic and anaerobic conditions (20–22, 38–40). However, the role of SrrAB in regulation of S. epidermidis biofilm formation and growth is not clear. In the present study, we first compared the protein sequence of SrrA in S. epidermidis strain 1457 to that in S. aureus strain COL. They shared 90.5% identity (see Fig. S1A in the supplemental material). SE1457 srrA and srrB were oriented in tandem and overlapped by 20 nucleotides. srrAB forms an operon with a putative promoter upstream of srrA and a transcription terminator structure (ATATATGAAAAACGCCTGCGACTCAGAGTGATGTCTCAGGCGTTTTTTTGTATATA, where boldface nucleotides represent reverse complement sequences to form a hairpin structure that may function in transcription termination) located 81 bp downstream of the srrB translational stop codon. By RT-PCR, a single mRNA covering srrAB was verified (see Fig. S1B in the supplemental material). We then found that oxygen limitation induced the expression of srrAB in S. epidermidis, whereas stressors such as vancomycin, ethanol, or high NaCl had no effect (see Fig. S3 in the supplemental material). This indicated that S. epidermidis SrrAB selectively responds to microaerobic stress.

To study the role of SrrAB in regulating biofilm formation and growth of S. epidermidis, an srrA deletion (ΔsrrA) mutant from SE1457 was constructed. The ΔsrrA mutant exhibited impaired biofilm formation and delayed growth under both oxic and microaerobic conditions, which was restored by complementation with srrAB. A double gene deletion mutant (ΔsrrAB) was constructed and showed a phenotype similar to that of the ΔsrrA mutant. This demonstrates that SrrAB regulates S. epidermidis growth under both oxic and microaerobic conditions. However, in S. aureus, srrAB regulated bacterial growth only under anaerobic conditions (20–22, 38), suggesting that the role of SrrAB in the growth of S. epidermidis is different from that in S. aureus. When we further analyzed the transcriptional profile by microarray and qRT-PCR, we found that the mRNA levels of qoxBACD, ctaA, pflBA, nrdDG, etc., were significantly reduced in the srrA mutant compared to the parent strain (Tables 4 and 5).

The qoxBACD operon encodes the cytochrome aa₃-type quinol oxidase, one of the two terminal oxidases in S. aureus. Terminal oxidases deficiency usually results in an inability to respire and a severe growth defect in bacteria. Hammer et al. (41) found that a double mutant lacking both qoxB and cydB showed significantly reduced aerobic growth and a small-colony variant phenotype, indicating that the mutant was unable to respire aerobically. Similar results were found by Kinkel et al. (42): the qoxBACD mutant exhibited a modest growth defect under aerobic conditions, and either srrAB or qoxBACD mutant strains were unable to reach a maximal final cell density. In the present study, the expression of the qoxBACD operon in the ΔsrrA mutant was downregulated, and SrrA was able to bind the promoter region of qoxBACD (Fig. 7). Therefore, we speculated that the aerobic growth retardation of the S. epidermidis srrA mutant may have resulted from the downregulation of the qoxBACD operon impairing respiratory chain reaction.

Under anaerobic conditions, expression levels of pflBA operon and nrdDG were reported highly induced in S. aureus (43, 44). PflA is an activating enzyme of PflB, a pyruvate formatelyase that catalyzes the reversible conversion of pyruvate to formate, thereby producing acetyl coenzyme A. Thus, the pflBA operon is important for energy supply when pyruvate is available and favors the growth of cells under fermentation conditions (45). The protein encoded by nrdDG is a class III ribonucleotide reductase that catalyzes the synthesis of deoxynucleoside triphosphates (dNTPs) via the reduction of NTPs under anaerobic conditions (17, 42, 46). In the present study, under microaerobic conditions, the S. epidermidis ΔsrrA mutant displayed severe growth retardation and inability to reach maximal final cell density, which may be related to the downregulation of pflBA and nrdDG expression and a subsequent decrease in fermentation and DNA replication. In addition, SrrA was able to bind the promoter region of pflBA (Fig. 7). This indicates that under microaerobic conditions SrrAB regulates S. epidermidis growth via pflBA (Fig. 8).

FIG 8.

Proposed model of srrAB regulation in S. epidermidis. SrrB represents the membrane-associated sensor kinase that becomes activated and autophosphorylated (indicated by circled “P”s) in the absence of O₂ (indicated by bright red spheres). The SrrB-P phosphorylates SrrA to SrrA-P, which acts as a response regulator that directly positively controls its own srrAB operon, as well as the ica, qox, and pfl operons (solid lines). Genes that are indirectly positively regulated are indicated by dotted lines. At the same time, SrrA-P acts also as a repressor for icaR, which encodes the repressor of the ica operon. +O₂, under oxic conditions; −O₂, under microaerobic conditions.

Besides its effect on bacterial growth, SrrAB regulates S. aureus biofilm formation (22, 39, 42). In the present study, srrA deletion resulted in a decrease in S. epidermidis biofilm formation under both oxic and microaerobic conditions. Although under oxic conditions the ΔsrrA mutant needed two more hours to enter stationary phase than did the wild-type strain (Fig. 1), the optical density of the two strains was similar after entering stationary phase (see Fig. S4 in the supplemental material). This indicates that under oxic conditions decreased biofilm formation by the ΔsrrA mutant may not be directly related to growth defects. Nevertheless, under microaerobic conditions, where the biofilm formed by the ΔsrrA mutant was less than that by SE1457, and the ΔsrrA mutant was unable to reach the optical density as high as SE1457 even if the culture time was extended to 48 h (Fig. 2A and C). This suggests that under microaerobic conditions an effect of growth retardation on decreased biofilm formation in the srrA mutant cannot be excluded.

Then, we further analyzed the possible mechanism that led to biofilm reduction in the S. epidermidis ΔsrrA mutant. The initial adherence, as a crucial step in biofilm formation, was decreased in the ΔsrrA mutant (Fig. 5), which indicated that SrrAB participated in the early stage of biofilm development. PIA production, which is regarded as the most important intercellular adherence factor and glue in the accumulation stage of biofilm formation in staphylococci (4, 9), was decreased in the ΔsrrA mutant compared to the parent strain, especially under microaerobic conditions (Fig. 6). In the S. aureus srrA transposon mutant, PIA synthesis was increased, whereas biofilm formation was decreased (23). This suggests that the mechanisms of biofilm formation regulated by SrrAB in S. epidermidis differ from those of S. aureus.

In S. aureus, SrrAB regulates biofilm formation via IcaR, a repressor of the ica operon. Under aerobic conditions, S. aureus SrrAB decreased biofilm formation by upregulating icaR expression but enhanced biofilm formation through downregulating icaR expression under microaerobic conditions (39). In the present study, we found that the transcription of icaR was upregulated in the S. epidermidis ΔsrrA mutant under oxic conditions but downregulated under microaerobic conditions; icaA expression was downregulated under both conditions. Under oxic conditions, S. epidermidis SrrAB positively regulated icaADBC expression and downregulated icaR expression, which is correlated with PIA production and biofilm formation. Under microaerobic conditions, the transcription of both icaR and icaADBC in the ΔsrrA mutant were downregulated, and PIA synthesis was decreased. We demonstrated that phosphorylated SrrA of S. epidermidis bound to the promoter regions of icaR and icaA. This suggests that the S. epidermidis SrrAB response to oxygen variation is to modulate biofilm formation in an ica-dependent pathway (Fig. 8).

Other intercellular matrix components may play important role in the biofilm formation. Extracellular DNA (eDNA) is released following bacterial autolysis (7, 15). The amount of eDNA within ΔsrrA mutant biofilm was more than that in the parent strain. The percentage of dead cells in ΔsrrA biofilm was much higher than that in the parent strain biofilm, although there was no difference between the ΔsrrA and SE1457 strains in cell viability in planktonic conditions or in Triton X-100-induced autolysis (see Fig. S5 in the supplemental material). Kinkel et al. (42) demonstrated that the srrAB mutant of S. aureus UAMS-1 had reduced capacity to form biofilm under static aeration conditions and that its biofilm contained significantly more dead cells than did the wild-type biofilm, and this was correlated with the loss of its structural integrity. It has been reported that pflBA was upregulated in the deeper layer of the biofilm, which may be related with the survival of biofilm cells in that place (45). In the present study, transcription of pfl operon was downregulated in the srrA mutant of S. epidermidis, which may be associated with the increased proportion of dead cells found in ΔsrrA biofilm.

In summary, S. epidermidis SrrAB responds to microaerobic stress and modulates biofilm formation in an ica-dependent manner. The mechanism that SrrAB regulates bacterial growth varies with environmental oxygen concentration: under oxic conditions, SrrAB modulates respiratory chain reaction by positively regulating qoxBACD transcription, while under microaerobic conditions it regulates fermentation processes and DNA replication via the pfl operon and nrdDG (Fig. 8). These results provide new insight into the SrrAB-mediated regulation of biofilm formation and the growth of S. epidermidis.