Role for the A Domain of Unprocessed Accumulation-Associated Protein (Aap) in the Attachment Phase of the Staphylococcus epidermidis Biofilm Phenotype

Abstract

The polysaccharide intercellular adhesin or the cell wall-anchored accumulation-associated protein (Aap) mediates cellular accumulation during Staphylococcus epidermidis biofilm maturation. Mutation of sortase, which anchors up to 11 proteins (including Aap) to the cell wall, blocked biofilm development by the cerebrospinal fluid isolate CSF41498. Aap was implicated in this phenotype when Western blots and two-dimensional (2D) electrophoresis revealed increased levels of the protein in culture supernatants. Unexpectedly, reduced levels of primary attachment were associated with impaired biofilm formation by CSF41498 srtA and aap mutants. In contrast to previous studies, which implicated Aap proteolytic cleavage and, specifically, the Aap B domains in biofilm accumulation, the CSF41498 Aap protein was unprocessed. Furthermore, aap appeared to play a less important role in the biofilm phenotype of S. epidermidis 1457, in which the Aap protein is processed. Anti-Aap A-domain IgG inhibited primary attachment and biofilm formation in strain CSF41498 but not in strain 1457. The nucleotide sequences of the aap gene A-domain region and cleavage site in strains CSF41498 and 1457 were identical, implicating altered protease activity in the differential Aap processing results in the two strains. These data reveal a new role for the A domain of unprocessed Aap in the attachment phase of biofilm formation and suggest that extracellular protease activity can influence whether Aap contributes to the attachment or accumulation phases of the S. epidermidis biofilm phenotype.

INTRODUCTION

Biofilm formation is the major virulence factor of Staphylococcus epidermidis isolates capable of causing device-related infections. Biofilm formation is a multifactorial process, and a variety of mediators have been identified and characterized. Polysaccharide intercellular adhesin (PIA) is probably the best-studied mediator of biofilm formation in S. epidermidis. PIA or poly-N-acetylglucosamine (PNAG) is an extracellular polysaccharide synthesized and exported by the icaADBC operon-encoded enzymes and has been implicated in biofilm accumulation in S. epidermidis (1, 2). In addition to PIA, the non-covalently bound major cell wall autolysin (AtlE) has been also been implicated in the primary attachment phase of biofilm formation (3, 4).

The transpeptidase sortase A (SrtA) covalently anchors proteins containing an LPXTG motif to the cell wall of a range of Gram-positive bacteria (5–9). S. epidermidis RP62A expresses 11 proteins containing an LPXTG motif (10). Of these, the accumulation-associated protein (Aap) (11, 12), the fibrinogen binding protein (SdrG) (13–15), the collagen binding protein (SdrF) (16), the Bap homologue protein (Bhp) (17), and S. epidermidis surface protein C (SesC) (18, 19) have all been implicated in the biofilm phenotype of S. epidermidis. Aap was previously shown to be essential for the accumulation of S. epidermidis RP62A on polymer surfaces (11). Aap is a multidomain protein (Fig. 1A) which is 54% identical to the SasG protein of Staphylococcus aureus (20). Aap and SasG are organized with a signal peptide at the N terminus, followed by an A domain, predicted to fold to an all-β-structure and implicated in skin colonization and adherence to desquamated epithelial cells (20, 21). Aap contains a variable number of 16-amino-acid repeats at the extreme N terminus of the A region. The B region comprises 3 to 12 repeating G5 and E domains of 128 amino acids followed by a single G5 repeat which are 64% identical in S. aureus and S. epidermidis (22). The protein has a wall-spanning domain and an LPDTG motif at its C terminus (12). The removal of the Aap A domain by an unidentified bacterial protease or a host protease promotes cellular accumulation and biofilm formation (12). The exposed B region promotes cell-cell adhesion via a Zn²⁺-dependent mechanism involving the formation of twisted rope-like structures that link the Aap B domains of adjacent cells (23). Chelation of Zn²⁺ in the growth media inhibited biofilm production by S. epidermidis and S. aureus (24). Antibodies against the B region and addition of exogenous recombinant B-domain protein inhibited biofilm formation in S. epidermidis 5179 (12).

FIG 1.

Sortase A contributes to the CSF41498 biofilm phenotype. (A) Schematic representation of the domain organization of Aap. The positions of the signal sequence (S)-, wall (W)-, and membrane (M)-spanning domains are indicated. The numbers of repeated G5 and E domains that comprise the B region differ between strains. (B) CSF41498 biofilms grown in 96-well plates in the presence and absence of proteinase K. EtOH, ethyl alcohol. (C) Biofilm phenotypes of strain CSF41498 and the CSF41498 ΔsrtA mutant harboring the plasmids pLI50 (control) and psrtA4 (srtA) grown in 96-well plates. (D) Biofilm phenotypes of strains CSF41498 and ΔsrtA on glass slides. (E) Immunoblot analysis of PIA production in whole-cell extracts from strains CSF41498 and ΔsrtA harboring the plasmids pLI50 (control) and psrtA4 (srtA) grown overnight at 37°C in BHI medium. Cell extracts were diluted as indicated on the right. Standard deviations and significant differences (*, P < 0.05) are indicated.

The contributions of various combinations of PIA, extracellular DNA, and surface proteins to staphylococcal biofilm production in a species- and strain-dependent manner highlight both the complexity and the importance of the biofilm lifestyle and have important implications, particularly for antibiofilm vaccine development (25). In this study, we investigated the role of protein adhesins in the biofilm phenotype of S. epidermidis cerebrospinal fluid isolate CSF41498 by examining the impact of a srtA mutation. Our findings revealed that the Aap protein of this strain does not undergo significant proteolysis and implicated the A domain of unprocessed Aap in promoting primary attachment and, ultimately, biofilm formation.

MATERIALS AND METHODS

Bacterial strains, plasmids, oligonucleotide primers, media, and growth conditions.

The bacterial strains and plasmids used in this study are listed in Table 1. Escherichia coli strains were grown at 37°C in Luria-Bertani (LB) medium (Sigma) supplemented when required with kanamycin (Kan) (50 μg/ml), carbenicillin (Car) (50 μg/ml), tetracycline (Tet) (5 μg/ml), or chloramphenicol (Cam) (10 μg/ml).

TABLE 1.

Strains and plasmids used in this study

Strain or plasmid	Characteristicsa	Source or reference
Strains
S. epidermidis CSF41498	S. epidermidis cerebrospinal fluid infection isolate; ica+ aap+	26
S. epidermidis CSF-2	CSF41498 derivative; biofilm negative; icaC::IS256Δtnp	25
S. epidermidis CSF41498 ΔsrtA	CSF41498 derivative; ΔsrtA::Tcr	This study
S. epidermidis CSF41498 Δaap	CSF41498 derivative; Δaap::Tcr	This study
S. epidermidis 1457	S. epidermidis central venous catheter infection isolate; ica+ aap+	27
S. epidermidis 1457 Δaap	1457 derivative; Δaap::Tcr	Schaeffer et al., submitted
S. epidermidis 1457 Δica	1457 derivative; ΔicaADBC::Tmpr	28
S. epidermidis RP62A	Biofilm-positive blood culture isolate; ica+ aap+	29
S. aureus RN4220	Restriction-deficient derivative of S. aureus 8325-4	30
E. coli TOPO	recA1 endA1 lac [F′ proAB lac1q Tn10 (Tetr)]	Invitrogen
Plasmids
pCR-Blunt II-TOPO	PCR cloning vector; Kamr Apr	Invitrogen
pLI50	E. coli-Staphylococcus cloning vector; Apr (E. coli) Camr (Staphylococcus)	31
pT181	4,450-bp S. aureus plasmid containing tetA gene	32
pBluescript	Cloning vector; Apr	Stratagene
pBlue::tet	pBluescript containing the tetA gene from pT181 on a 2,236-bp HindIII fragment	This study
psrtA1	2,129-bp PCR product containing the srtA gene and its upstream promoter region from CSF41498 using primers SEsrtA_1 and SEsrtA_2 in pCR-Blunt II-TOPO	This study
psrtA2	2,323-bp SwaI-SmaI fragment containing tetA gene from pBlue::tet cloned into the BstZ17I site of psrtA1	This study
psrtA3	4,283-bp SwaI fragment containing the ΔsrtA::Tcr fragment cloned into the SmaI site of pBT2	This study
psrtA4	2,047-bp SwaI fragment containing the srtA locus cloned from psrtA1 and cloned into pLI50	This study
pALC2073	S. aureus-E. coli shuttle vector containing the tetR gene and xyl-tetO promoter; Apr Camr	33
pALC2073sasG+	pALC2073 containing full-length sasG; Apr Camr	20

^aAp, ampicillin; Tmp, trimethoprim.

S. epidermidis and S. aureus strains were routinely grown at 37°C or 30°C on brain heart infusion (BHI) or tryptone soya broth (TSB) medium (Oxoid, United Kingdom) supplemented with Cam (10 μg/ml) or Tet (5 μg/ml), as indicated.

Oligonucleotide primers SEaap1 (GGCAAACGTAGACAAGGTC) and SEaap2 (CATCGACTGCTTTAGGTGTG) were used to amplify the A-domain region of aap.

Construction of S. epidermidis srtA::Tc^r insertion mutant.

The srtA deletion mutant strain was constructed as follows. A 2,129-bp fragment containing the srtA gene was amplified using primers SEsrtA_1 (AAATTGCTTTCAATATAAAT) and SEsrtA_2 (TTTGCATAAGAATTCGCTCATA). The blunt PCR product was cloned into pCR-Blunt II-TOPO plasmid (Invitrogen) to create psrtA1.

The 2,236-bp tetracycline resistance (Tc^r) gene was cloned from pBlue-tet on a SwaI-SmaI fragment and ligated into the BstZ17I site of psrtA1, yielding psrtA2.

A 4,283-bp SwaI fragment containing the mutant allele from psrtA2 was ligated into pBT2 digested with SmaI to create psrtA3. Temperature-sensitive psrtA3 was electroporated into CSF41498, following electroporation and passage through S. aureus RN4220.

Allele replacement of temperature-sensitive psrtA3 in CSF41498 was achieved by growth at 30°C in the presence of Cam (10 μg/ml) and Tet (5 μg/ml) (3 subcultures) followed by repeated growth (3 subcultures) at 42°C without antibiotic selection and selection of Tc^r colonies on BHI agar plates. Tc^r colonies were then screened for sensitivity to Cam to confirm plasmid loss, and PCR analysis was used to verify the presence of the srtA::Tc^r deletion allele on the chromosome using primers SEsrtA_1 and SEsrtA_2.

To complement the srtA mutation, a 2,047-bp SwaI fragment containing the srtA gene and its upstream promoter region was cloned from psrtA1 and cloned in pLI50 digested with SmaI to create psrtA4, which was ultimately electroporated into the ΔsrtA mutant.

Construction of S. epidermidis CSF41498 aap::Tc^r insertion mutant.

Phage A6C (a kind gift from Holger Rohde) was used to transduce an aap::Tc^r mutation from S. epidermidis 1457 to strain CSF41498 using a previously described protocol (34). A strain 1457 aap::Tc^r mutant was constructed as previously described (C. R. Schaeffer, K. M. Woods, G. M. Longo, M. R. Kiedrowski, A. E. Paharik, H. Büttner, M. Christner, R. J. Boissy, A. R. Horswill, H. Rohde, and P. D. Fey, submitted for publication).

Biofilm assays.

Semiquantitative measurements of biofilm formation under static conditions were determined using Nunclon tissue culture-treated (ΔSurface) 96-well polystyrene plates (Nunc, Denmark) or microscope glass slides, as described previously (35). Each strain was tested at least three times, and averages of the results are presented. A biofilm-positive phenotype was defined as being characterized by an A₄₉₂ of ≥0.17. Proteinase K (Sigma) was added at 10 μg/ml where indicated. α₂-Macroglobulin was added to cultures at 0.25 U/ml where indicated.

Primary attachment assays.

Attachment assays using Nunclon tissue culture-treated 96-well polystyrene plates were carried out using overnight cultures in BHI medium adjusted to an A₆₀₀ of 1.0. A 200-μl volume of each suspension was used to inoculate the microtiter plate wells prior to incubation at 37°C for 1 h. Following incubation, the wells were washed gently 3 times with distilled phosphate-buffered saline (PBS), dried at 65°C for 1 h, and stained for 5 min with 0.4% crystal violet. After staining, the plates were washed gently three times with distilled H₂O and the remaining crystal violet was solubilized using 100 μl of 5% acetic acid. The absorbance of the solubilized crystal violet was measured at A₄₉₂.

Biofilm flow cell experiments.

A BioFlux 1000z microfluidic system (Fluxion Biosciences Inc., San Francisco, CA) was used to assess biofilm formation under shear flow conditions. Biofilms were grown in BHI medium supplemented with glucose (1% [wt/vol]). The system was initiated by adding 200 μl of medium to the output wells of a 48-well plate and priming the channels for 5 min at 5.0 dynes/cm². After priming, the medium was aspirated from the output wells and replaced with a 50-μl suspension of bacteria grown to the early-exponential-growth phase and adjusted to an A₆₀₀ of 0.8. A further 50 μl of medium was added to the input wells, and the channels were seeded by pumping from the output wells to the input wells for 3 to 5 s at a speed of 3 dynes/cm². Bacteria were allowed to attach to the surface of the plate for 1 h at 37°C. Excess inoculum solution was aspirated from the output wells, and a further 1 ml of medium was added to the input wells. The flow rate was set at 0.2 dyne/cm² for 24 h (equivalent to 42 μl/hour), and bright-field images were captured every 5 min at ×10 magnification.

Biofilm inhibition assays using antibodies against Aap.

Biofilm inhibition assays were carried out as described by Hussain et al. (11), with some modifications. Various concentrations of polyclonal anti-Aap A-domain IgG and the control anti-rFnBPB IgG were added to the cultures, and these mixtures were incubated at 4°C for 1 h. A 100-μl volume of this suspension was used to inoculate microtiter plate wells prior to incubation at 37°C for 24 h. Microtiter plates were washed three times with distilled H₂O and dried for 1 h at 65°C. Plates were stained with crystal violet, and absorption was measured as previously described. CSF41498 grown in the absence of IgG was used as the positive control, while CSF41498 grown in the presence of anti-rFnbB IgG was the negative control.

Primary attachment inhibition assays using antibodies against Aap.

Primary attachment inhibition assays were carried out as described by Shahrooei et al. with modifications (19). Two concentrations of Aap A- and B-domain antisera and anti-rFnbB IgG (control) were added to cultures, and these mixtures were incubated at 4°C for 1 h. A 200-μl volume of each suspension was used to inoculate microtiter plate wells prior to incubation at 37°C for 1 h. Following incubation, wells were washed gently 3 times with PBS, dried at 45°C for 1 h, and stained with crystal violet, and absorption was measured at A₅₉₅.

PIA quantification.

PIA assays were performed based on the method of Cramton et al. (36) as described previously by Holland et al. (37).

Preparation of extracellular proteins.

Cells from batch cultures grown to the exponential phase in modified Actinomyces defined medium (M-ADM) (48) were harvested by centrifugation, and proteins in the supernatant were precipitated overnight at 4°C by the addition of 10% trichloroacetic acid (TCA). After centrifugation at 16,000 × g for 30 min at 4°C, the protein pellet was resuspended in acetone and sonicated three times for 10 s each time. The samples were then centrifuged at 16,000 × g for 30 min at 4°C, and the resulting pellet was dissolved in 500 μl of two-dimensional (2D) gel electrophoresis rehydration buffer containing 8 M urea, 2% CHAPS {3-[(3-cholamidopropyl)-dimethylammonio]-1-propanesulfonate}, 10 mM dithiothreitol (DTT), and 2% IPG buffer (GE Healthcare Life Sciences) and stored at −20°C.

2D polyacrylamide gel electrophoresis.

The protein concentration in the extracts was determined using a 2D Quant kit (GE Healthcare Life Sciences). A volume corresponding to 20 μg protein was diluted with rehydration buffer and placed in a reswelling cassette with linear IPG strips (GE Healthcare Life Sciences) (18 cm in length, pH 4 to 7) on top. Rehydration was carried out at room temperature for 30 h using silicone oil. Isoelectric focusing was carried out using a Multiphor II electrophoresis system (GE Healthcare Life Sciences), with cooling water at 15°C supplied by a Pharmacia Multitemp II refrigerated bath. The focusing was initiated at 150 V for 1 h and continued at 300 V for 3 h, 600 V for 3 h, 1,200 V for 12 h, and, finally, 3,500 V for 20 h. After focusing, the IPG strips were stored at −80°C. Before the second dimension was run, the IPG strips were equilibrated first in 50 mM Tris buffer (pH 6.8) containing 2% SDS, 26% glycerol, and 16 mM DTT for 15 min and then in 50 mM Tris buffer (pH 6.8) containing 2% SDS, 26% glycerol, 250 mM iodoacetamide, and 0.005% bromophenol blue for another 15 min. The equilibrated IPG strips were embedded on top of 7% or 14% polyacrylamide gels (20 by 20 by 0.1 cm) using 0.5% (wt/vol) molten agarose. SDS-PAGE was performed at a constant current of 15 mA per gel and at 10°C overnight in a Protean II xi cell (Bio-Rad) with high-range Rainbow molecular mass standards (GE Healthcare Life Sciences) run on the acidic side of the IPG strips. Gels were stained with silver according to the protocol from GE Healthcare Life Sciences. All gels were run in triplicate, and only those protein spots occurring in all gels were considered further. Proteins were considered to show increased or decreased production only when the trend was reproduced in all three gels.

Surface protein Western blots.

Cell wall-associated proteins of S. epidermidis were prepared as previously described (38). Stationary-phase cultures were harvested, washed in PBS, and adjusted to an A₆₀₀ of 40 in lysis buffer (50 mM Tris-HCl, 20 mM MgCl₂; pH 7.5) supplemented with 30% (wt/vol) raffinose and Complete protease inhibitors (Roche) (40 μl/ml). Cell wall proteins were solubilized by incubation with lysostaphin (Ambi Products, Lawrence, NY) (200 μg/ml) for 10 min at 37°C. Protoplasts were removed by centrifugation at 12,000 × g for 10 min, and the supernatant containing solubilized cell wall proteins was aspirated and boiled for 5 min in Laemmli sample buffer (Sigma). For supernatant fractions, bacteria were removed from an overnight culture by centrifugation, and the supernatant was passed through a 0.2-μm-pore-size filter. Protein was concentrated to 20% of the original volume on a 9,000-Da-molecular-mass-cutoff spin column.

Proteins were separated on 7.5% (wt/vol) polyacrylamide gels, transferred onto polyvinylidene difluoride membranes (Roche), and blocked in 10% (wt/vol) skimmed-milk proteins. Blots were probed with anti-Aap A-domain (1:20,000) antibodies and detected using horseradish peroxidase (HRP)-conjugated goat anti-rabbit antibodies (Dako) (1:3,000). Reactive bands were visualized using a LumiGLO reagent and peroxide detection system (Cell Signaling Technology).

Antibodies.

Polyclonal rabbit antiserum was used throughout. In order to purify antigen-specific IgG, serum was passed through a protein A column. IgG was affinity purified against the cognate antigen (recombinant His-tagged Aap A domain corresponding to amino acids 53 to 608 or recombinant His-tagged FnBPB A domain). Aap B domain antibodies raised against SasG B domains were described previously (38) and cross-react with Aap B domains. PIA antiserum was a gift from G. B. Pier.

Statistical analysis.

Two-tailed, two-sample equal-variance Student's t tests (Microsoft Excel 2007) were used to determine statistically significant differences in assays performed during this study. A significant difference was indicated as a P value of <0.05.

RESULTS

Impaired biofilm production by a CSF41498 srtA mutant.

CSF41498 is a PIA-producing S. epidermidis strain isolated from cerebrospinal fluid (37, 39). Although PIA has been shown to be necessary for biofilm formation in this strain (37, 39), growth in the presence of proteinase K inhibited biofilm production in BHI medium or BHI medium supplemented with 1% glucose but not in BHI medium supplemented with 4% NaCl or 4% ethanol, which activates icaADBC transcription and PIA production (26) (Fig. 1B). These data implicated a protein adhesin(s) in the CSF41498 biofilm phenotype under specific growth conditions. To investigate the involvement of LPXTG-anchored protein adhesins in CSF41498 biofilm production, allele replacement was used to construct a srtA::Tc^r mutation. When grown in BHI and BHI glucose media, the srtA::Tc^r mutant displayed a reduced capacity to form biofilm in tissue-culture-treated 96-well plates (Fig. 1C) and on glass slides (Fig. 1D) despite exhibiting a growth rate similar to that seen with strain CSF41498 (data not shown). Consistent with our biofilm dispersal assays performed with proteinase K (Fig. 1B), the srtA mutation had no impact on CSF41498 biofilm in BHI NaCl medium (Fig. 1C). The biofilm defect of CSF41498 ΔsrtA was complemented by a multicopy srtA plasmid (psrtA4) (Fig. 1C). The colony morphologies of strains CSF41498 and CSF41498 ΔsrtA were examined on Congo red agar, which can be used as an indicator medium for PIA production (40). Both strains produced dark red, crusty colonies indicative of PIA production (data not shown). Immunoassay slot blot analyses revealed that PIA expression levels in BHI medium were similar in strains CSF41498, CSF41498 ΔsrtA, and CSF41498 ΔsrtA psrtA4 (Fig. 1E). Thus, the strain CSF41498 ΔsrtA biofilm defect cannot be attributed to reduced PIA expression.

Reduced biofilm-forming capacity of a srtA mutant is associated with defective cell wall anchoring of Aap.

Supernatant protein fractions were prepared, and 2D gel electrophoresis revealed that a large (∼220-kDa) protein was more abundant in the supernatant fraction of strain CSF41498 ΔsrtA than in that of strain CSF41498 on a silver-stained 2D gel (Fig. 2A). As silver-stained gels are more sensitive but less quantitative, the protein fractions were also separated on a 2D gel and stained with Coomassie brilliant blue G (Fig. 2B), further revealing that the levels of this ∼220-kDa protein were higher in the ΔsrtA culture supernatant. Mass spectrometry performed commercially on spots taken from the ΔsrtA supernatant fraction identified the ∼220-kDa protein as the LPXTG motif-containing accumulation-associated protein (Aap).

FIG 2.

Mutation of srtA is associated with secretion of Aap into the supernatant. (A and B) Released proteins from S. epidermidis CSF41498 and ΔsrtA resolved on 2D PAGE gels and stained with silver (A) and Coomassie blue (B). (C) Western blot analysis of Aap expression using A-domain IgG in supernatant and cell wall fractions of strains CSF41498, RP62A (control), and CSF41498 ΔsrtA.

Western blot analysis using anti-A-domain Aap antibodies on cell wall and supernatant protein fractions from strains CSF41498 and CSF41498 ΔsrtA confirmed that Aap was more abundant in the secreted fraction of the mutant and that this corresponded to reduced levels of Aap in the cell wall fraction (Fig. 2C). S. epidermidis RP62A was included as a control. Based on this finding, we hypothesized that impaired biofilm formation in CSF41498 ΔsrtA may be associated with a defect in the anchoring of Aap to the cell wall.

Aap is required for primary attachment and biofilm formation in S. epidermidis CSF41498 but not 1457.

To evaluate the role of aap in the CSF41498 biofilm phenotype, we used phage A6C to transduce a Δaap::Tc^r mutation from S. epidermidis 1457 to S. epidermidis CSF41498. The CSF41498 Δaap mutant, in which the A and B domains of aap are replaced with tetM, was confirmed using PCR (data not shown) and pulsed-field gel electrophoresis (data not shown).

Similar to our findings with the CSF41498 ΔsrtA mutant, the Δaap mutation significantly reduced biofilm production in BHI and BHI glucose medium but not in BHI NaCl medium (Fig. 3A). Interestingly, biofilm assays performed with strain 1457 and its isogenic aap mutant revealed no significant difference, suggesting that Aap may play a less important role in the 1457 biofilm phenotype (Fig. 3A). Under shear flow conditions, CSF41498 formed robust biofilms when grown in BHI glucose medium, and, similarly to our findings under static growth conditions, the aap mutation impaired this phenotype (Fig. 3B).

FIG 3.

Biofilm phenotypes of CSF41498 and 1457 aap mutants. (A) Biofilm phenotypes of strains CSF41498 and 1457 and their respective aap mutants grown for 24 h at 37°C in BHI, BHI glucose, and BHI NaCl media in hydrophilic 96-well polystyrene plates. (B) Biofilm accumulation of strains CSF41498 and Δaap under flow conditions grown in BHI glucose medium in a BioFlux 1000z instrument under a shear flow rate of 0.6 dyne cm⁻². Bright-field images were captured at ×10 magnification after 18 h.

Western blot analysis with A-domain IgG confirmed the absence of Aap expression in the Δaap mutant and revealed lower levels of Aap associated with the cell wall of the ΔsrtA mutant (Fig. 4). Importantly, these Western blots also revealed that full-length 220-kDa Aap was expressed in strain CSF41498 (and in strain RP62A, which was included as a control) whereas the smaller 150-kDa processed Aap protein was produced by strain 1457 and very little 220-kDa protein was detected (Fig. 4). A proteolytic cleavage event removes the N-terminal A domain of Aap (41) and exposes the repeated G5 and E domains which can mediate intracellular accumulation (23, 24). To investigate why Aap is processed differently in strain CSF41498 and strain 1457, the A-domain regions and proteolytic cleavage sites of the aap genes in both strains were amplified using primers SEaap1 and SEaap2 and sequenced. The A-domain aap sequences were identical to each other (data not shown), indicating that reduced or absent protease activity may be responsible for impaired Aap proteolysis in strain CSF41498.

FIG 4.

Western blot analysis of Aap expression using A-domain IgG in cell wall fractions of RP62A, 1457, CSF41498, CSF41498 ΔsrtA, and CSF41498 Δaap cultures grown to stationary phase in BHI medium.

Inhibition of proteolytic processing by α₂-macroglobulin was previously shown to interfere with Aap B-domain-mediated biofilm accumulation in strain 5179 (12). Here we demonstrated that growth of strain CSF41498 in BHI medium containing α₂-macroglobulin had no impact on biofilm formation, suggesting that Aap processing is not involved in mediating biofilm formation in this strain (data not shown). Anti-Aap A-domain IgG inhibited biofilm formation by CSF41498 in a dose-dependent manner (Fig. 5A). Antibody raised against FnbB (anti-FnbB IgG), an unrelated protein from S. aureus, was used as a negative control and had no effect on biofilm formation (Fig. 5A).

FIG 5.

A role for Aap in the attachment and biofilm phenotypes of strain CSF41498. (A) Inhibition of CSF41498 biofilms grown for 24 h at 37°C in BHI medium in hydrophilic 96-well polystyrene plates by IgG specific for the A-domain of Aap or IgG specific for FnbB (negative control). (B) Primary attachment of strains CSF41498, Δaap, Δaap pALC2073, and Δaap pALC2073sasG to hydrophilic 96-well plates after incubation for 1 h in BHI medium at 37°C. (C) Primary attachment of 1457 and its isogenic aap mutant to hydrophilic 96-well plates after incubation for 1 h in BHI media at 37°C. All experiments were carried out at least 3 times, and standard deviations are indicated. *, significant difference (P < 0.05). (D) Primary attachment of strains CSF41498 and 1457 to hydrophilic 96-well plates after incubation at 37°C for 1 h in TSB medium supplemented with various concentrations of antibodies to the A or B domains of Aap.

The CSF41498 Δaap mutant displayed a significant decrease in its capacity to attach to hydrophilic polystyrene (Fig. 5B). Because efforts to introduce a plasmid carrying the S. epidermidis aap gene from strain 5179 into the CSF41498 aap mutant were unsuccessful, we instead complemented the mutant using SasG from S. aureus, which is 54% identical to Aap and which we have previously shown to be implicated in the S. aureus biofilm phenotype (40). These data demonstrated that heterologous expression of sasG restored the primary attachment rates of the aap mutant to wild-type levels (Fig. 5B) and further highlight the functional conservation of SasG and Aap. Primary attachment rates of strain 1457 and its isogenic Δaap mutant were similar (Fig. 5C). The CSF41498 srtA mutant exhibited significantly reduced attachment rates compared to the wild type, whereas primary attachment was not significantly affected in a CSF41498 icaC mutant, which was defective in intercellular accumulation (see Fig. S1 in the supplemental material).

Aap has not previously been implicated in primary attachment, suggesting that this role of Aap in the CSF41498 biofilm phenotype may be a feature of this strain. To investigate this and to further confirm the role of the A domain in primary attachment, antibody inhibition assays were carried out on both strain CSF41498 and strain 1457. Interestingly, IgG raised against the A domain of Aap completely inhibited primary attachment of CSF41498 but had no effect on the attachment of 1457 (Fig. 5D). Antibodies against the B domain of Aap had no significant impact on primary attachment of either isolate (Fig. 5D). These data suggest a role for the A domain of unprocessed Aap in the primary attachment stage of the CSF41498 but not the 1457 biofilm phenotype.

DISCUSSION

Protein adhesins and PIA are important in S. epidermidis biofilm formation, but their relative contributions are strain dependent (42). Growth of strain CSF41498 in the presence of proteinase K significantly inhibited biofilm formation as did mutation of srtA. Analysis of the extracellular protein fractions of the wild type and the srtA mutant identified the secretion of Aap as the only discernible impact of srtA mutation on the secreted protein profile under our conditions. Mutation of aap confirmed its role in biofilm in strain CSF41498. Aap was previously implicated in biofilm accumulation (12, 42, 43) via a mechanism believed to involve proteolytic cleavage of the protein and exposure of the B domains, which promote homophilic interactions between adjacent cells (24). However, Aap is expressed in strain CSF41498 as an unprocessed 220-kDa protein. Furthermore, primary attachment assays and antibody inhibition assays demonstrated that Aap promotes primary cell adherence to surfaces in CSF41498 via its A domain. Thus, these data reveal a new role for the A domain of unprocessed Aap in the attachment stage of the CSF41498 biofilm phenotype. Nevertheless, PIA remains essential for CSF41498 biofilm accumulation and, interestingly, osmotic stress, which activates the icaADBC operon and PIA production, overcomes the biofilm defect in the CSF41498 aap mutant. The experiments described in this report did not reveal a role for Aap in the biofilm phenotype of strain 1457, which we attribute to a more prominent role for PIA in this strain. However, experiments described in a separate report did reveal a role for Aap in optimal biofilm production by strain 1457, particularly under shear flow conditions and in an ica-negative background (Schaeffer et al., submitted).

As its name implies, the accumulation-associated protein was originally reported to promote biofilm accumulation and not primary attachment (11, 43). Anti-Aap A-domain IgG interfered with primary attachment of strain CSF41498 but not strain 1457, and Western blot analyses revealed that Aap is proteolytically processed in strain 1457 but not in strain CSF41498. Because the aap gene sequences corresponding to the A domain and cleavage site in strains 1457 and CSF41498 are identical, it would appear that the activity of the protease(s) responsible for Aap processing in CSF41498 is different.

We have previously reported that the S. aureus Aap homologue SasG promotes intercellular accumulation via the B domain following proteolytic processing (38). However, extensive analysis performed with protease inhibitors and protease mutants failed to identify an enzyme involved in proteolysis of SasG and we concluded that SasG may undergo spontaneous proteolysis at labile peptide bonds within the G5 and E domains (38). Given that similar patterns of Aap processing are not observed in strains CSF41498 and 1457, these data suggest that the Aap B domains are unlikely to undergo spontaneous proteolysis. In this context, it is interesting that expression of the S. epidermidis serine protease Esp inhibits S. aureus biofilm formation and nasal colonization (44) and that levels of extracellular serine protease activity differ significantly among clinical isolates of S. epidermidis (45). Esp has been shown to process the major autolysin (46) and a range of proteins involved in S. aureus biofilm formation (47). It is therefore tempting to speculate that reduced activity of one or more proteases in strain CSF41498 may explain why Aap is not processed in this strain.

Many cell wall-anchored proteins of staphylococci are multifunctional, and distinct domains mediate adherence to different ligands and contribute to distinct phenotypes. The A domain of Aap has already been shown to mediate binding to human corneocytes in S. epidermidis NCTC 11047 (21). Our findings reveal a new role for Aap in mediating primary attachment, adding to our understanding of the S. epidermidis biofilm phenotype. Intriguingly, Aap can participate in either the primary attachment phase or the accumulation phase of biofilm formation, apparently depending on the extracellular protease activity of the host strain.