Characterization of the fibrinogen-binding surface protein Fbl of Staphylococcus lugdunensis

Jennifer Mitchell1, Anne Tristan2 and Timothy J. Foster1

1 Microbiology Department, Moyne Institute of Preventive Medicine, Trinity College, Dublin 2, Ireland
2 INSERM E0230, Faculté Laennec, IFR 62, rue G. Paradin, 69008 Lyon, France

Correspondence
Timothy J. Foster
tfoster{at}tcd.ie


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The fbl gene of Staphylococcus lugdunensis encodes a protein Fbl that is 58 % identical to the clumping factor A (ClfA) of Staphylococcus aureus. The fbl gene was present in eight clinical isolates of S. lugdunensis. When Fbl was expressed on the surface of Lactococcus lactis it promoted adherence to immobilized fibrinogen and cell clumping in a fibrinogen solution. Purified recombinant Fbl region A bound to immobilized fibrinogen in a dose-dependent manner and inhibited the adherence of both Fbl-expressing and ClfA-expressing strains of L. lactis to fibrinogen. Adherence of S. lugdunensis and L. lactis Fbl+ to immobilized fibrinogen was also inhibited by rabbit anti-Fbl region A antibodies and rabbit anti-ClfA region A antibodies, as well as by human immunoglobulin with a high level of anti-ClfA antibodies. Alignment of the A domains of CflA and Fbl revealed that all of the ClfA residues implicated in binding to the {gamma}-chain of fibrinogen are conserved in Fbl. Nevertheless Fbl had a tenfold lower affinity for fibrinogen, suggesting that sequence differences that occur elsewhere in the protein, possibly in {beta}-strand E of domain N2, affect ligand binding.


Abbreviations: ClfA, clumping factor A; CoNS, coagulase-negative Staphylococcus


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Staphylococcus lugdunensis is a species of coagulase-negative Staphylococcus (CoNS) and was first described by Freney et al. (1988). It is a commensal of human skin and can occasionally cause serious invasive infections such as osteomyelitis, peritonitis, soft tissue abscesses and infective endocarditis, diseases that are more usually associated with Staphylococcus aureus (Van der Mee-Marquet et al., 2003; Bellamy & Barkham 2002; Kragsbjerg et al., 2000). S. lugdunensis accounts for 18 % of infective endocarditis and 44 % of native valve endocarditis caused by CoNS (Patel et al., 2000). S. lugdunensis endocarditis is more aggressive than that caused by other CoNS and resembles that caused by S. aureus.

S. lugdunensis expresses several potential virulence factors, including the SLUSH synergistic toxins, haemolysins, extracellular enzymes and a glycocalyx (Donvito et al., 1997; Leung et al., 1998). This paper reports the properties of the fibrinogen-binding protein of S. lugdunensis which is closely related to clumping factor A, a fibrinogen-binding protein of S. aureus.

The clumping factor A protein (ClfA) is an important virulence factor of S. aureus. It contributes to the pathogenesis of experimental septic arthritis and endocarditis in rabbits (Josefsson et al., 2001; Moreillon et al., 1995). It binds to the extreme C-terminus of the {gamma}-chain of fibrinogen at the same site as the platelet integrin GPIIbIIIa (Medved et al., 1997; McDevitt et al., 1997). ClfA is the archetype of a family of surface-associated proteins with similar structural organization. The primary translation product comprises a secretory signal sequence followed by a 520 residue ligand-binding domain, then repeats of the dipeptide DS, a short cell-wall-spanning region, and a C-terminal domain that comprises a sortase signal LPXTG, a hydrophobic membrane-spanning domain and a series of positive-charged residues (McDevitt et al., 1994; Fig. 1). The DS repeats act as a flexible stalk to extend the ligand-binding A domain from the cell surface (Hartford et al., 1997). The signal sequence is removed during protein secretion, the sortase cleaves LPXTG between the T and G, and in a transpeptidation reaction it joins the protein to uncross-linked nascent peptidoglycan precursor, which is then polymerized into new cell wall (Perry et al., 2002; Mazmanian et al., 2001).



View larger version (45K):
[in this window]
[in a new window]
 
Fig. 1. (a) Schematic representation of the domain structure of Fbl and ClfA. The percentage identities of domains are indicated. Fbl does not contain the SLAAVA metalloprotease cleavage motif that bisects the N2 and N3 regions of ClfA. The repeated R region of Fbl differs from that of ClfA because it comprises a SDSDSA hexapeptide motif rather than the SD repeats in ClfA. (b) CLUSTALW alignment of the amino acid sequences of the A domains of Fbl and ClfA. The SLAAVA motif of ClfA is shaded grey, and the individual residues shown by site-directed mutagenesis to be important in ligand binding by ClfA are bold and underlined. Residues 327–335, which line the putative latching trough, and 532–538, comprising the putative latch, are highlighted by black shading.

 
Structural analysis of ClfA and the related fibrinogen-binding proteins SdrG and ClfB revealed that the ligand-binding A domain is composed of three subdomains called N1, N2 and N3 (Deivanayagam et al., 2002; Ponnuraj et al., 2003). The structure of smallest truncate of the ClfA A domain that retains the ability to bind fibrinogen (residues 220–559; sometimes called N2N3) has been solved. Each subdomain comprises nine {beta}-strands that form a novel IgG-type fold (Deivanayagam et al., 2002). The fibrinogen {gamma}-chain peptide-binding site is located in a hydrophobic groove at the junction between N2 and N3. Substitution mutants of several residues predicted to contact the peptide ligand caused defects in fibrinogen binding (Hartford et al., 2001; Deivanayagam et al., 2002).

This paper describes the fibrinogen-binding protein Fbl of S. lugdunensis and compares it with ClfA of S. aureus. It studies both purified recombinant protein and the native protein displayed on the surface of the surrogate Gram-positive host Lactococcus lactis.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains and growth conditions.
Bacterial strains are listed in Table 1. L. lactis MG1363 was grown at 30 °C in M17 medium (Oxoid) supplemented with 0·5 % glucose. S. aureus was grown at 37 °C in tryptic soy broth or agar (Difco). S. lugdunensis was grown in L broth and L agar. Escherichia coli XL-1 Blue was grown at 37 °C in Luria–Bertani medium. When appropriate, antibiotics were added at the following concentrations: erythromycin, 5 µg ml–1 for L. lactis and 10 µg ml–1 for S. aureus; ampicillin, 100 µg ml–1 for E. coli. Bacterial stocks were kept at –70 °C in broth supplemented with 10 % (v/v) glycerol.


View this table:
[in this window]
[in a new window]
 
Table 1. Bacterial strains

 
Manipulation of DNA.
DNA manipulations were performed using standard methods. Enzymes were obtained from Roche Molecular Biochemicals or Promega and were used as directed by the manufacturer. Genomic DNA was isolated from a 2 ml stationary-phase culture of S. aureus using the AGTC bacterial genomic DNA purification kit (Edge BioSystems), adapted for use for staphylococci by the incorporation of lysostaphin (0·01 mg ml–1 final concentration; Ambicin L recombinant lysostaphin, Applied Microbiology). E. coli plasmid DNA was purified using a Wizard Mini-prep kit (Promega) or the Qiagen Plasmid Midi-prep kit. Standard methods for DNA manipulation were used in constructing plasmids, and for Southern blot hybridization, PCR products were labelled with digoxigenin (Boehringer Mannheim) during amplification by Taq DNA polymerase (Gibco-BRL).

Cloning of fbl in L. lactis.
The fbl gene of S. lugdunensis strain N920143 was amplified from the putative translational initiation codon to the TAA stop codon by using Pfu polymerase. It was then cloned into the L. lactis expression vector pKS80 (Hartford et al., 2001), which provides a lactococcal promoter, initiation codon, ribosome-binding site and a translational coupling mechanism to ensure efficient initiation of translation of the cloned gene. The primers used are listed in Table 2.


View this table:
[in this window]
[in a new window]
 
Table 2. Primers

Restriction endonuclease (RE) sites are underlined.

 
Constructing a variant that secreted the Fbl A region from L. lactis.
In order to construct a L. lactis vector that secreted recombinant Fbl region A, a reverse primer recognizing DNA encoding the C-terminal amino acids of Fbl region A, and a forward primer recognizing the stop codon of Fbl and adjoining flanking sequence, including the transcription terminator of plasmid pKS80 (Table 2), were used to amplify a 7 kb fragment from plasmid pKS80 : fbl. This was then purified using the High Pure PCR Product Purification Kit from Roche Diagnostics. The product was cut using SacI, then repurified, ligated and transformed into L. lactis MG1363. Clones secreting rFbl region A carrying a truncated plasmid pKS80 : fblA were identified by testing supernatants by dot immunoblotting with anti-ClfA region A antibodies.

Purification of rFbl40–534.
L. lactis pKS80 : fblA was inoculated into GM17 supplemented with 5 µg erythromycin ml–1 and grown for 16 h at 30 °C. Cells were pelleted by centrifugation at 7000 r.p.m. (Sorvall GS3 rotor, 5320 g) and 445 g ammonium sulphate l–1 was added to the supernatant and mixed well. The precipitated protein was removed, dissolved in 50 ml H2O and dialysed against 10 mM Tris/HCl, 0·9 % NaCl, pH 7·4 at 4 °C. This was concentrated tenfold in a stirred-cell ultrafiltration chamber (Amicon) with a 50 kDa cut-off membrane. Proteins were separated by size fractionation through Sephadex G-50. Peak fractions were applied to a 5 ml HiTrap Q Sepharose column (Amersham Pharmacia Biotech). Bound protein was eluted with a continuous linear gradient of NaCl (50–500 mM; total volume 100 ml) in 20 mM Tris, 2 mM EDTA (pH 7·9). Peak fractions were analysed by SDS-PAGE, then pooled and concentrated.

Antibodies.
Rabbit antibodies to the A region of ClfA were kindly provided by Judy Higgins (Microbiology Department, Trinity College, Dublin). Antibodies to rFbl40–534 were raised in a young New Zealand White rabbit whose pre-immune sera showed no reaction with S. lugdunensis wall-associated antigens in Western blots. Protein (25 µg in 10mM Tris/HCl, 0·9 % NaCl, pH 7·4) was emulsified with an equal volume of Freund's complete adjuvant (500 µl) and injected subcutaneously. Three subsequent injections given at 2 week intervals contained Freund's incomplete adjuvant. The rabbits were bled, the serum was recovered and IgG was purified.

SDS-PAGE and Western immunoblotting.
Bacterial cells were suspended to an OD600 of 60 in 30 % raffinose and 20 mM MgCl2 in a final volume of 500 µl. A 10 µl volume of lysostaphin (10 mg ml–1) and 8 µl of protease inhibitors (Complete mixture, Roche Molecular Biochemicals) were added and the suspension was incubated at 37 °C for 20 min. Protoplasts were removed by centrifugation at 6000 g for 15 min. Supernatants containing wall-associated proteins and protoplasts in 30 % raffinose were prepared for electrophoresis by boiling for 5 min in final sample buffer (0·125 M Tris/HCl, 4 %, w/v, SDS, 20 % glycerol, 10 %, v/v, 2-mercaptoethanol, 0·002 %, w/v, bromphenol blue) and analysed in 10 % (w/v) acrylamide gels. Gels were stained with Coomassie blue or electrophoretically transferred to PVDF Western-blotting membranes (Boehringer) by the semi-dry system (Bio-Rad). Membranes were blocked for 15 h at 4 °C with 10 % (w/v) blocking reagent (Marvel milk powder).

Anti-ClfA antibodies were used because they allowed detection of ClfA and Fbl in the same blot, whereas anti-Fbl antibodies cross-reacted too weakly to allow detection of ClfA in the same sample. They were used at a 1 : 1000 dilution followed by use of protein-A-conjugated horseradish peroxidase (Sigma; a 1 mg ml–1 stock diluted 1 : 500) or goat anti-rabbit IgG at a 1 : 1000 dilution (Dako) to detect bound antibody. Membranes were developed using LumiGLO chemiluminescent substrate (New England Biolabs) according to the manufacturer's instructions and exposed to X-ray film.

Adherence assays.
Adherence of bacterial cells to immobilized fibrinogen was performed as described by Hartford et al. (1997). Briefly, microtitre plates were coated with 0·5 µg ml–1 of the protein in PBS and incubated overnight at 4 °C. Bovine serum albumin (2 %, w/v, in PBS) was added and incubated for 1 h at 37 °C. The plates were washed three times with PBS, and 100 µl bacterial cell suspension (1x108 c.f.u. ml–1) was added. The plates were incubated at 37 °C for 2 h and then washed three times with PBS. Bound cells were fixed with formaldehyde (25 %, v/v) for 30 min and then stained with crystal violet (0·5 % v/v) for 1 min. Absorbance was measured at 570 nm in an ELISA plate reader (Labsystems Multiskan Plus).

Inhibition of adherence.
The ability of recombinant Fbl region A and recombinant ClfA region A to inhibit the adherence of L. lactis strains to immobilized fibrinogen was determined as described by Hartford et al. (1997), with modifications. Microtitre plates were coated with 10 µg fibrinogen ml–1 in PBS and incubated overnight at 4 °C. Bovine serum albumin (2 %, w/v, in PBS) was added and the plates were incubated for 1 h at 37 °C. A 100 µl volume of rFbl was added in doubling dilutions and incubated with gentle agitation for 1 h at room temperature. After this, 100 µl bacterial cell suspension (1x108 c.f.u. ml–1) was added, the plates were incubated with gentle agitation at room temperature for 1·5 h, and then washed three times with PBS. Bound cells were fixed with formaldehyde (25 %, v/v) for 30 min and stained with crystal violet (0·5 %, v/v) for 1 min. Absorbance was measured at 570 nm in an ELISA plate reader (Labsystems Multiskan Plus).

The same procedure was used to study anti-Fbl antibody inhibition of adherence, except that bacteria were incubated with 35 mg ml–1 anti-Fbl antibodies or pre-immune serum for 30 min prior to measuring adherence.

Fibrinogen binding by Fbl region A.
The ability of recombinant proteins to bind to immobilized fibrinogen was analysed using an ELISA-type assay. Microtitre plates were coated with fibrinogen (10 µg ml–1 in PBS) at 4 °C. The plates were washed three times with PBS containing 0·5 % Tween 20, and blocked with 2 % bovine serum albumin for 2 h at 37 °C. After an additional three washes with PBS containing 0·5 % Tween 20, rFbl40–534 was added and the plates were incubated at 37 °C for 2 h. The wells were washed again and incubated with anti-ClfA region A antibodies at 37 °C for 1 h. After further washing, horseradish-peroxidase-labelled goat anti-rabbit IgG (Sigma) was added at a 1 : 1000 dilution. Following incubation at 37 °C for 1 h and washing with PBS, 100 µl chromogenic substrate (580 µg tetramethylbenzidine ml–1 and 0·0001 % H2O2 in 0·1 M sodium acetate buffer, pH 5·2) was added per well and developed for 10 min in the dark. The reaction was stopped by the addition of 2 M H2SO4 (50 µl per well). Plates were read at 450 nm. The apparent dissociation constant (KD) of each protein for fibrinogen was the protein concentration that resulted in 50 % binding (Davis et al., 2001).

Measurement of cell clumping.
S. aureus and S. lugdunensis were grown to stationary phase with aeration, harvested by centrifugation at 7000 r.p.m. (Sorvall GS3 rotor, 5320 g) for 10 min, and washed in PBS. L. lactis strains were grown statically at 30 °C for 16 h and harvested in the same way. A suspension of 4x108 c.f.u. in a 20 µl volume was added to fibrinogen (Calbiochem, 1 mg ml–1 and doubling dilutions thereof) in a 96-well microtitre dish. The highest dilution of fibrinogen that produced clumping after 5 min agitation was defined as the clumping titre (Hartford et al., 1997).

Whole-cell blotting.
Cells from stationary-phase cultures were harvested by centrifugation, washed in PBS, and resuspended at an OD600 of 100. Volumes of 10 µl were spotted onto a nitrocellulose membrane (Protran) and dried. The membrane was blocked with 10 % (w/v) skim milk in PBS for 1 h, and probed with anti-ClfA region A antibodies at a 1 : 1000 dilution. Thereafter the membranes were treated as for Western immunoblotting.

Whole cells were tested for fibrinogen-binding activity by probing duplicate membranes with 10 ml of a 1 mg ml–1 solution of fibrinogen in 10 % (w/v) skim milk in PBS for 1 h. Membranes were subsequently washed and probed with anti-fibrinogen antibodies conjugated to horseradish peroxidase (Dako) at a 1 : 1000 dilution. Thereafter the membranes were treated as for Western immunoblotting.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Analysis of Fbl in S. lugdunensis strains
The sequence of the gene encoding a putative fibrinogen-binding protein called Fbl has been lodged in the NCBI database (GenBank accession no. AF404823). Fbl has close sequence and organizational similarity to ClfA of S. aureus (Fig. 1). It is notable that the region of Fbl that corresponds to the minimum ligand-binding A region of ClfA (ClfA residues 220–559; N2 and N3) is 60 % identical, whereas the N1 domains are only 19 % identical. Each of the residues that are known to be required for fibrinogen binding by ClfA (P336, Y338, K389, I387, Y256, A254, E526, V527) are conserved in Fbl (Deivanayagam et al., 2002; Hartford et al., 2001; Fig. 1). However, the metalloprotease cleavage motif SLAAVA in ClfA is not present in Fbl. The ClfA A domain is linked to the cell wall anchoring region by DS repeats encoded by the 18 bp DNA repeat GAY TCN GAY TCN GAY AGY, whereas Fbl region R comprises DSDSDA repeats encoded by GAY TCN GAY AGY GAY GCR (where R is a purine, Y is a pyrimidine and N is any base).

Southern hybridization analysis using a probe corresponding to bases 1400–3001 of the clfA gene was performed with eight S. lugdunensis strains. The probe encoded the C-terminal end of the A-domain, the SD repeat region and the wall-spanning region of ClfA, including the LPDTG motif. A single hybridizing band of 9 kb was revealed in each strain when the genomic DNA was cut with HindIII (Fig. 2 shows one representative strain), suggesting that the fbl gene is ubiquitous in S. lugdunensis. The clfA probe hybridized to fbl because of identical bases in codons 1, 2, 3 and 5 of the 18 bp repeats.



View larger version (43K):
[in this window]
[in a new window]
 
Fig. 2. (a) Southern hybridization of genomic DNA from S. lugdunensis N920143 probed using DIG-labelled DNA corresponding to the S. aureus clfA gene encoding the ClfA protein SD-repeat region to the C-terminus. Lane 1, S. lugdunensis N920143 genomic DNA cut with HindIII. Lane 2, S. lugdunensis N920143 genomic DNA cut with BamHI. (b) PCR amplification of the entire fbl gene (2721 bp) from S. lugdunensis N920143 genomic DNA. (c) PCR analysis of DNA flanking the repeated R region of fbl. Lane M, size markers. Lanes 1–8, S. lugdunensis strains: 1, N920143; 2, N930432; 3, N940025; 4, N940084; 5, N940113; 6, N940135; 7, N910319; 8, N910320.

 
PCR analysis of the fbl gene of S. lugdunensis using primers that amplified across the repeat region R revealed two distinct size classes (Fig. 2). This and the Southern data are consistent with other studies indicating that S. lugdunensis is a highly clonal species (Dufour et al., 2002).

The eight S. lugdunensis strains were tested for their ability to adhere to immobilized fibrinogen. Each strain adhered to fibrinogen although the level of binding varied (Fig. 3). Adherence of each was completely inhibited by anti-Fbl antibodies, whereas pre-immune serum had no effect. This demonstrates that adherence of S. lugdunensis to immobilized fibrinogen is promoted exclusively by Fbl.



View larger version (51K):
[in this window]
[in a new window]
 
Fig. 3. Adherence of S. lugdunensis strains to immobilized fibrinogen. ELISA plates (96-well) were coated with 10 µg fibrinogen ml–1. S. lugdunensis grown to stationary phase was collected by centrifugation, washed and resuspended to an OD600 of 1·0. Cells were preincubated with 35 mg ml–1 anti-Fbl region A antibodies, 35 mg ml–1 rabbit antibodies or PBS. Cells were stained with crystal violet and washed with PBS. The absorbance of cells bound to fibrinogen-coated microtitre plate wells was measured in an ELISA plate reader at 570 nm. Values represent the means of triplicate wells. This experiment was carried out three times with similar results.

 
Expression of Fbl by L. lactis
In order to determine unambiguously whether Fbl is a fibrinogen-binding adhesin similar to ClfA of S. aureus, the fbl gene from S. lugdunensis strain N920143 was cloned into the L. lactis expression vector pKS80 and transformed into L. lactis strain MG1363 to form pKS80 : fbl. Transformant colonies were identified by whole-cell dot immunoblotting with anti-ClfA antibodies, and they were shown to react with fibrinogen when probed with fibrinogen and anti-fibrinogen antibodies. L. lactis Fbl+ was tested by comparative whole-cell dot immunoblotting in order to determine the relative levels of expression of Fbl protein on the surface of S. lugdunensis and L. lactis Fbl+. As the blots in Fig. 4 show, L. lactis Fbl+ expressed a level of Fbl protein that was ~64-fold higher than that of S. lugdunensis N920143. Anti-Fbl antibodies did not bind to plasmid-free L. lactis, and pre-immune serum did not react with S. lugdunensis or L. lactis (not shown).



View larger version (23K):
[in this window]
[in a new window]
 
Fig. 4. Expression of Fbl on the surface of S. lugdunensis N920143 and L. lactis Fbl+. S. lugdunensis N920143 and L. lactis Fbl+ cells at an OD600 of 2·0 were serially diluted twofold and spotted onto nitrocellulose membranes. The membranes were probed with polyclonal anti-Fbl region A antibodies and visualized using horseradish-peroxidase-conjugated goat anti-rabbit IgG. This experiment was carried out twice with similar results.

 
L. lactis Fbl+ was tested for its ability to adhere to immobilized fibrinogen (Fig. 5). L. lactis Fbl+ adhered in a dose-dependent and saturable manner that was similar to L. lactis expressing ClfA from S. aureus Newman. L. lactis Fbl+ was also tested for its ability to clump in soluble fibrinogen in comparison to L. lactis ClfA+, S. aureus strain Newman and S. lugdunensis N920143. Clumping of the S. lugdunensis strain was not detected in soluble fibrinogen, whereas overexpression of Fbl by L. lactis produced a clumping titre of 306, which was only twofold lower than the titre of 612 for L. lactis ClfA+. S. aureus Newman also produced a clumping titre of 612. These data show that Fbl is strongly expressed in L. lactis, and that it is a fibrinogen-binding adhesin and clumping factor.



View larger version (14K):
[in this window]
[in a new window]
 
Fig. 5. Adherence of L. lactis Fbl+ and L. lactis ClfA+ to immobilized fibrinogen. ELISA plates (96-well) were coated with varying concentrations of fibrinogen. L. lactis Fbl+ ({bullet}) and L. lactis ClfA+ ({blacksquare}) cells grown to stationary phase were collected by centrifugation, washed and resuspended to an OD600 of 1·0. L. lactis pKS80 ({lozenge}) was used as a control. The absorbance of cells stained with crystal violet was measured in an ELISA plate reader at 570 nm. Values represent the means of triplicate wells. This experiment was carried out three times with similar results.

 
Recombinant Fbl region A
Many attempts were made to clone the DNA encoding the A domain of Fbl into the expression vectors pQE30 and pGEX, and into cloning vectors pUC18 and pBluescript in E. coli. Despite changing cloning sites and using host strains that supported the establishment of unstable chimeras, low frequencies of transformation were always obtained, and any putative recombinants had undergone rearrangements. An alternative strategy to express recombinant Fbl A domain protein involved an inverse PCR reaction using plasmid pKS80 : fbl as a template in order to delete the DNA encoding the SD repeat region and the cell-wall- and membrane-spanning region of Fbl. These sequences were replaced with a stop codon. The resulting plasmid (pKS80 : fblA) expressed a protein that was 495 amino acids in length comprising only the A domain of Fbl which was secreted into the culture medium. The recombinant protein rFbl40–534 was purified by ammonium sulphate precipitation and size exclusion chromatography, followed by anion-exchange chromatography. It had an apparent molecular mass of 70 kDa determined by SDS-PAGE, compared to its predicted molecular mass of 54·33 kDa (data not shown).

Binding of anti-Fbl antibodies to ClfA and of anti-ClfA antibodies to Fbl
The interaction of rabbit anti-Fbl and anti-ClfA A domain antibodies with recombinant bacterial proteins was analysed by ELISA (data not shown). The anti-Fbl antibodies had a higher affinity for the cognate antigen rFbl40–534 than for the ClfA A domain rClfA40–534. Similarly, rabbit anti-rClfA A domain antibodies bound more strongly to the cognate antigen rClfA40–559 than to rFbl40–534.

Anti-Fbl antibodies potently inhibited adherence of S. lugdunensis N920143 to immobilized fibrinogen, confirming data in Fig. 3. Higher amounts were required to inhibit adherence of L. lactis Fbl+, reflecting a higher level of Fbl expression by this strain (Fig. 6a). Complete inhibition of S. aureus Newman was only achieved at the highest concentration of antibody tested, whereas L. lactis ClfA+ was only inhibited by 50 %. Once again this reflects a different level of protein expression, and also shows that the anti-Fbl antibodies contain molecules that neutralize the fibrinogen-binding activity of ClfA.



View larger version (14K):
[in this window]
[in a new window]
 
Fig. 6. The inhibitory effects of anti-rFbl40–534 antibodies and anti-rClfA40–559 antibodies on Fbl- and ClfA-mediated adherence to fibrinogen. S. aureus Newman ({square}), S. lugdunensis NN920143 ({circ}), L. lactis ClfA+ ({blacksquare}) and L. lactis Fbl+ ({bullet}) cells were pre-incubated with varying concentrations of anti-rFbl40–534 antibodies (a), anti-ClfA40–559 antibodies (b) or Veronate (c). Cells that adhered to fibrinogen-coated microtitre plate wells were stained with crystal violet and the absorbance at 570 nm was measured with an ELISA plate reader. Values represent the means of triplicate wells and are expressed as the percentage inhibition of binding of cells to fibrinogen compared to control (no antibody). These experiments were repeated three times with similar results.

 
Similar results were obtained when the rabbit anti-ClfA antibodies were tested for their ability to inhibit bacterial adherence to immobilized fibrinogen (Fig. 6b). However, the anti-ClfA antibodies were more potent inhibitors of Fbl-expressing bacteria than anti-Fbl antibodies were of inhibiting of ClfA-expressing bacteria.

Pooled human IgG enriched for anti-ClfA antibodies (Veronate; Inhibitex) was tested for its ability to block ClfA- and Fbl-mediated fibrinogen binding. Veronate inhibited both ClfA- and Fbl-expressing bacteria from adhering to immobilized fibrinogen in a concentration-dependent manner (Fig. 6). Veronate blocked S. lugdunensis more efficiently than L. lactis Fbl+. This is presumably because there is less Fbl protein expressed on the surface of S. lugdunensis than L. lactis Fbl+. Similarly Veronate blocked the adherence of S. aureus Newman to immobilized fibrinogen more efficiently than it blocked L. lactis ClfA+.

Western immunoblotting analysis
S. lugdunensis N920143 was digested with lysostaphin in the presence of raffinose in order to stabilize protoplasts and solubilize cell-wall-associated proteins. Analysis by Western immunoblotting with anti-ClfA antibodies detected no immunoreactive proteins in the cell wall fraction (Fig. 7). Proteins of ~175 and ~60 kDa were detected in the protoplast fraction. No immunoreactive proteins were detected with pre-immune serum. The larger protein probably represents the native form of Fbl, and the smaller is probably a breakdown product. The predicted molecular mass of Fbl is 90 kDa, but like ClfA it appears to migrate aberrantly in SDS-PAGE at close to twice the predicted size (Hartford et al., 1997).



View larger version (65K):
[in this window]
[in a new window]
 
Fig. 7. Expression of Fbl by S. lugdunensis and L. lactis (pKS80 : fbl). Lane 1, stabilized protoplasts of S. lugdunensis 1003. Lane 2, lysostaphin-solubilized cell wall fraction of S. lugdunensis 1003. Lane 3, stabilized protoplast fraction of L. lactis Fbl+. Lane 4, mutanolysin/lysozyme-solubilized cell wall fraction of L. lactis Fbl+. Lane 5, mutanolysin/lysozyme-solubilized cell wall fraction of L. lactis ClfA+. Lane 6, stabilized protoplast fraction of L. lactis pKS80. Lane 7, mutanolysin/lysozyme-solubilized cell wall fraction of L. lactis pKS80. Membranes were probed with anti-ClfA region A antibodies followed by goat anti-rabbit IgG conjugated to horseradish peroxidase.

 
Cell wall and protoplast fractions of L. lactis Fbl+ were analysed following mutanolysin digestion. The cell wall fraction of L. lactis Fbl+ did not contain any immunoreactive proteins, whereas immunoreactive proteins were detected in the control L. lactis expressing ClfA, a protein known to be sorted to the cell wall fraction. The protoplast fraction of L. lactis Fbl+ contained several immunoreactive proteins, the largest of which (175 kDa) is probably the intact protein. The smaller bands are likely to be degradation products of Fbl, because the antibodies did not react with any proteins in the L. lactis control. It seems that Fbl is not sorted to the cell wall, either in its native host or in L. lactis, but remains associated with the protoplast fraction as precursor II (Perry et al., 2002)

Recombinant Fbl40–534 binding to fibrinogen
Recombinant Fbl40–534 bound to immobilized fibrinogen in ELISA-type ligand-binding assays in a dose-dependent and saturable fashion, indicating a specific interaction. This formally demonstrates that the ligand-binding region of Fbl is in the A domain between residues 40 and 534. rFbl had an apparent KD of 1·5 µM compared with 150 nM for rClfA when bound proteins were detected by anti-Fbl antibodies (Fig. 8a) and anti-ClfA antibodies (Fig. 8b). The higher affinity of the antibody for the cognate antigen did not affect the apparent KD.



View larger version (16K):
[in this window]
[in a new window]
 
Fig. 8. Binding of recombinant Fbl40–534 and ClfA40–559 to immobilized fibrinogen. Microtitre wells were coated with a solution of 10 µg human fibrinogen ml–1. Increasing concentrations of rFbl40–534 ({circ}) or rClf40–559 ({square}) were added and bound protein was detected by addition of anti-Fbl region A antibody (a) or anti-ClfA region A antibody (b), followed by goat anti-rabbit antibodies conjugated to alkaline phosphatase and development with p-nitrophenyl phosphate. Absorbance was measured at 450 nm. Values represent the means of triplicate wells. This experiment was repeated three times with similar results.

 
Adherence of L. lactis Fbl+ and L. lactis ClfA+ to immobilized fibrinogen was inhibited by rFbl40–534 in a dose-dependent manner (Fig. 9b). Recombinant rClfA40–559 (Fig. 9a) also blocked L. lactis Fbl+ adherence, whereas recombinant ClfB197–542 had no effect (data not shown). This indicates that Fbl and ClfA bind to a similar region of the C-terminal end of the {gamma}-chain of fibrinogen. L. lactis Fbl+ also adhered to purified fibrinogen {gamma}-chain in a dose-dependent and saturable manner, whereas L. lactis ClfB+, which adheres to the {alpha}-chain, and L. lactis pKS80 did not bind to the {gamma}-chain (data not shown).



View larger version (13K):
[in this window]
[in a new window]
 
Fig. 9. Inhibition of adherence of L. lactis Fbl+ ({bullet}) and L. lactis ClfA+ ({blacksquare}) to microtitre wells coated with fibrinogen by (a) rClfA40–559 and (b) rFbl40–534. Adherent cells were stained with crystal violet and absorbance at 570 nm was measured in an ELISA plate reader. Values represent the means of triplicate wells and are expressed as a percentage of controls without recombinant protein. These experiments were repeated three times with similar results.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
This paper shows that Fbl is a fibrinogen-binding adhesin and a clumping factor. Because a genetic system has not been established in S. lugdunensis, it was not possible to generate a null mutant of fbl in S. lugdunensis. Thus, the identification of Fbl as a surface-located fibrinogen-binding protein was achieved by demonstrating that anti-Fbl region A antibodies blocked adhesion to immobilized fibrinogen and reacted with a surface-associated protein in whole-cell immunoblots.

The clonality of S. lugdunensis was suggested by PCR amplification of the DNA encoding the R region of Fbl, which yielded two distinct sizes that differed by ~100 bp. When similar analysis was performed with clfA, many different sizes of R region were observed, a feature that reflects the more diverse population structure of S. aureus shown by PFGE (McDevitt & Foster, 1995; van der Mee-Marquet et al., 2003).

When expressed on the surface of L. lactis, Fbl behaved in a similar manner to ClfA. It promoted adherence to immobilized fibrinogen and cell clumping in a fibrinogen solution. Recombinant ClfA region A inhibited the adherence of L. lactis Fbl+ to immobilized fibrinogen, indicating that Fbl interacts with the {gamma}-chain of fibrinogen. Thus it can be concluded that Fbl binds to the same region of the {gamma}-chain of fibrinogen as ClfA, but whether precisely the same residues are recognized is not known.

S. lugdunensis cells did not clump in soluble fibrinogen. However when Fbl was expressed in L. lactis, Fbl-promoted clumping did occur. This suggests that the level of expression of the fibrinogen-binding protein determines the ability of cells to form clumps. Western immunoblotting and whole-cell dot immunoblotting blotting showed that there is ~64-fold more Fbl on the surface of L. lactis Fbl+ than on S. lugdunensis N920143.

The predicted molecular mass of full-length Fbl is 95 kDa. However, when protoplasts of S. lugdunensis and L. lactis Fbl+ were analysed by SDS-PAGE and Western immunoblotting, the highest immunoreactive band had an apparent molecular mass of 175 kDa. Aberrant migration at approximately twice the predicted molecular mass is a property of other Clf/Sdr proteins and is attributed to the presence of the multiple S residues in region R (Hartford et al., 1997). When Fbl was isolated from L. lactis Fbl+, the protein appeared to have undergone proteolytic degradation. This is likely to happen during cell wall digestion and probably does not affect the functionality of surface-exposed Fbl on intact cells. Studies reported by other authors concluded that the majority of molecules of heterologously expressed protein were intact and fully functional on the surface of L. lactis cells, and that degradation only occurred during solubilization of the proteins when they became exposed to a membrane-associated HtrA-like protease (O'Brien et al., 2002; Poquet et al., 2000; Miyoshi et al., 2002).

It is interesting to observe that while ClfA is covalently sorted to the cell wall peptidoglycan in S. aureus and L. lactis, Fbl does not appear to be sorted to the cell wall in either host. A sortase signal LPKTG is present in Fbl. However, other factors must interfere with its ability to become anchored to the cell wall. It has been shown that changes to a consensus sequence SIRK-G/S in the signal sequence of some staphylococcal surface proteins can impede efficient sorting (Bae & Schneewind, 2003). Differences in the signal sequence of Fbl at residues F9, H14 and K15 compared to ClfA might provide an explanation as to why Fbl sorting is inefficient.

A major focus of this study was to determine whether antibodies recognizing S. aureus ClfA would also bind to S. lugdunensis Fbl. S. lugdunensis is often misdiagnosed as S. aureus, and is a frequent cause of invasive endocarditis. Polyclonal anti-ClfA antibodies inhibited S. aureus infections in animals, but it is not clear whether antibodies that block ClfA binding to fibrinogen would have the same effect on S. lugdunensis. Anti-ClfA region A antibodies inhibited Fbl-mediated bacterial adherence to immobilized fibrinogen in a concentration-dependent manner, and anti-Fbl antibodies inhibited ClfA-mediated adherence. This indicates that Fbl and ClfA are sufficiently similar in amino acid sequence and in three-dimensional structure that they share epitopes that enable antibodies raised against one protein to inhibit the other.

Fbl has a tenfold lower affinity for fibrinogen than ClfA. It is possible that this is due in part to the absence of residues equivalent to residues 550–559 in ClfA. ClfA220–559 had a higher affinity for fibrinogen than ClfA220–550, suggesting functional importance for residues 550–559 (McDevitt et al., 1997). Nevertheless ClfA220–550 was still able to bind to fibrinogen in a dose-dependent and saturable manner. It is possible that residues 550–559 play a role in stabilizing the interaction between ClfA and fibrinogen, while not actually contributing directly to the structure of the binding domain (Deivanayagam et al., 2002).

The crystal structure of rClfA220–559 was solved without the bound ligand in place (apo-ClfA). Solving the structure of the fibrinogen {gamma}-chain peptide co-crystallized with the minimum binding domain of ClfA region A would promote understanding of how exactly the terminal residues of the {gamma}-chain of fibrinogen interact with the residues of ClfA. Ponnuraj et al. (2003) successfully co-crystallized the minimum binding domain of SdrG, the fibrinogen-binding protein of Staphylococcus epidermidis, with a modified fibrinogen {beta}-chain peptide. They solved the structure of SdrG bound to its ligand and compared it with the structure of the unbound apo-SdrG. This led to the proposal of a ‘dock, lock and latch’ model for binding, whereby an open form of SdrG first binds to the fibrinogen {beta}-chain peptide, a loop at the end of SdrG N3 then encloses the peptide within the binding groove and the peptide is locked into position. This could be a universal mechanism for Clf-Sdr proteins binding to peptide ligands.

The residues in ClfA that have been shown to interact directly with the extreme C-terminal residues of the {gamma}-chain peptide of fibrinogen (AGDV) are conserved in Fbl. They are located in the hydrophobic trench between N2 and N3. The majority of amino acid residue differences between Fbl and ClfA occur away from the fibrinogen-binding site and are proposed to contribute to antigenic differences. However Fbl has a tenfold lower affinity for fibrinogen than ClfA. It is not clear if the ‘dock, lock and latch’ mechanism applies to Fbl and ClfA because the three-dimensional structure of ClfA has the putative latching peptide wrapped around N3 in the apo form. Using SdrG as a model, the putative latching peptide of ClfA would comprise residues at the extreme C-terminus of domain N3 (532–537). Only one of these residues varies in Fbl (Fig. 1). However, it is possible that the amino acids in {beta}-strand E, in domain N2 lining the putative latching trough, which vary between Fbl and ClfA, affect the affinity of Fbl for fibrinogen. This region has diverged considerably between ClfA and Fbl and there is a greater proportion of charged residues in the former (Fig. 1). In order to test this hypothesis, residues in ClfA could be changed to those of Fbl, and vice versa, particularly in the critical region of ClfA277DDVK280 bearing charged residues which are absent in the corresponding Fbl sequence.

Fbl and ClfA bind to a similar region of the {gamma}-chain of fibrinogen. However, differences may occur in the precise interaction between residues in the binding trough and the {gamma}-chain. Comparison of the interaction of the variant fibrinogen {gamma}-chain peptides with rFbl40–534 and rClfA40–559 may shed some light on this.

Considering that Fbl exhibits many similar characteristics to ClfA, it is reasonable to assume that it is one of the main virulence factors of S. lugdunensis. It is encouraging to note that antibodies generated against bothFbl A domain and ClfA A domain recognize the heterologous proteins, thus indicating that common antigenic epitopes must be present on each protein allow antibodies (including those in pooled human IgG) to inhibit fibrinogen binding by Fbl. This suggests that polyclonal human IVIG Veronate (Vernachio et al., 2003) is likely to protect against S. lugdunensis as well as S. aureus infections.


   ACKNOWLEDGEMENTS
 
This project was supported by The Wellcome Trust (061617), Enterprise Ireland, the Higher Education Authority and Inhibitex Inc.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bae, T. & Schneewind, O. (2003). The YSIRK-G/S motif of staphylococcal protein A and its role in efficiency of signal peptide processing. J Bacteriol 185, 2910–2919.[Abstract/Free Full Text]

Bellamy, R. & Barkham, T. (2002). Staphylococcus lugdunensis infection sites: predominance of abscesses in the pelvic girdle region. Clin Infect Dis 35, E32–34.[CrossRef][Medline]

Davis, S. L., Gurusiddappa, S., McCrea, K. W., Perkins, S. & Höök, M. (2001). SdrG, a fibrinogen-binding bacterial adhesin of the microbial surface components recognizing adhesive matrix molecules subfamily from Staphylococcus epidermidis, targets the thrombin cleavage site in the B{beta} chain. J Biol Chem 276, 27799–27805.[Abstract/Free Full Text]

Deivanayagam, C. C., Wann, E. R., Chen, W., Carson, M., Rajashankar, K. R., Höök, M. & Narayana, S. V. (2002). A novel variant of the immunoglobulin fold in surface adhesins of Staphylococcus aureus: crystal structure of the fibrinogen-binding MSCRAMM, clumping factor A. EMBO J 21, 6660–6672.[Abstract/Free Full Text]

Donvito, B., Etienne, J., Denoroy, L., Greenland, T., Benito, Y. & Vandenesch, F. (1997). Synergistic hemolytic activity of Staphylococcus lugdunensis is mediated by three peptides encoded by a non-agr genetic locus. Infect Immun 65, 95–100.[Abstract]

Dufour, P., Jarraud, S., Vandenesch, F., Greenland, T., Novick, R. P., Bes, M., Etienne, J. & Lina, G. (2002). High genetic variability of the agr locus in Staphylococcus species. J Bacteriol 184, 1180–1186.[Abstract/Free Full Text]

Duthie, E. S. & Lorenz, L. L. (1952). Staphylococcal coagulase; mode of action and antigenicity. J Gen Microbiol 6, 95–107.

Freney, J., Brun, Y., Bes, M., Meugnier, H., Grimont, F., Grimont, P., Nervi, C. & Fleurette, J. (1988). Staphylococcus lugdunensis sp. nov and Staphylococcus schleiferi sp. nov., two species from human clinical specimens. Int J Syst Bacteriol 38, 168–172.

Hartford, O., Francois, P., Vaudaux, P. & Foster, T. J. (1997). The dipeptide repeat region of the fibrinogen-binding protein (clumping factor) is required for functional expression of the fibrinogen-binding domain on the Staphylococcus aureus cell surface. Mol Microbiol 25, 1065–1076.[Medline]

Hartford, O. M., Wann, E. R., Höök, M. & Foster, T. J. (2001). Identification of residues in the Staphylococcus aureus fibrinogen-binding MSCRAMM clumping factor A (ClfA) that are important for ligand binding. J Biol Chem 276, 2466–2473.[Abstract/Free Full Text]

Josefsson, E., Hartford, O., O'Brien, L., Patti, J. M. & Foster, T. (2001). Protection against experimental Staphylococcus aureus arthritis by vaccination with clumping factor A, a novel virulence determinant. J Infect Dis 184, 1572–1580.[CrossRef][Medline]

Kragsbjerg, P., Bomfim-Loogna, J., Tornqvist, E. & Soderquist, B. (2000). Development of antimicrobial resistance in Staphylococcus lugdunensis during treatment – report of a case of bacterial arthritis, vertebral osteomyelitis and infective endocarditis. Clin Microbiol Infect 6, 496–499.[CrossRef][Medline]

Leung, M. J., Nuttall, N., Pryce, T. M., Coombs, G. W. & Pearman, J. W. (1998). Colony variation in Staphylococcus lugdunensis. J Clin Microbiol 36, 3096–3098.[Abstract/Free Full Text]

Mazmanian, S. K., Ton-That, H. & Schneewind, O. (2001). Sortase-catalysed anchoring of surface proteins to the cell wall of Staphylococcus aureus. Mol Microbiol 40, 1049–1057.[CrossRef][Medline]

McDevitt, D., Francois, P., Vaudaux, P. & Foster, T. J. (1994). Molecular characterization of the clumping factor (fibrinogen receptor) of Staphylococcus aureus. Mol Microbiol 11, 237–248.[Medline]

McDevitt, D., Francois, P., Vaudaux, P. & Foster, T. J. (1995). Identification of the ligand-binding domain of the surface-located fibrinogen receptor (clumping factor) of Staphylococcus aureus. Mol Microbiol 16, 895–907.[Medline]

McDevitt, D., Nanavaty, T., House-Pompeo, K., Bell, E., Turner, N., McIntire, L., Foster, T. & Höök, M. (1997). Characterization of the interaction between the Staphylococcus aureus clumping factor (ClfA) and fibrinogen. Eur J Biochem 247, 416–424.[Abstract]

Medved, L., Litvinovich, S., Ugarova, T., Matsuka, Y. & Ingham, K. (1997). Domain structure and functional activity of the recombinant human fibrinogen {gamma}-module ({gamma}148–411). Biochemistry 36, 4685–4693.[CrossRef][Medline]

Miyoshi, A., Poquet, I., Azevedo, V., Commissaire, J., Bermudez-Humaran, L., Domakova, E., Le Loir, Y., Oliveira, S. C., Gruss, A. & Langella, P. (2002). Controlled production of stable heterologous proteins in Lactococcus lactis. Appl Environ Microbiol 68, 141–146.

Moreillon, P., Entenza, J. M., Francioli, P., McDevitt, D., Foster, T. J., Francois, P. & Vaudaux, P. (1995). Role of Staphylococcus aureus coagulase and clumping factor in pathogenesis of experimental endocarditis. Infect Immun 63, 4738–4743.[Abstract]

O'Brien, L. M., Kerrigen, S. W., Kaw, G., Hogan, M., Penades, J., Litt, D., Fitzgerald, D. J., Foster, T. J. & Cox, D. (2002). Multiple mechanisms for the activation of human platelet aggregation by Staphylococcus aureus. Roles for the clumping factors ClfA and ClfB, the serine-aspartate repeat protein SdrE and protein A. Mol Microbiol 44, 1033–1044.[CrossRef][Medline]

Patel, R., Piper, K. E., Rouse, M. S., Uhl, J. R., Cockerill, F. R. & Steckelberg, J. M. (2000). Frequency of isolation of Staphylococcus lugdunensis among staphylococcal isolates causing endocarditis: a 20-year experience. J Clin Microbiol 38, 4262–4263.[Abstract/Free Full Text]

Perry, A. M., Ton-That, H., Mazmanian, S. K. & Schneewind, O. (2002). Anchoring of surface proteins to the cell wall of Staphylococcus aureus. III. Lipid II is an in vivo peptidoglycan substrate for sortase-catalyzed surface protein anchoring. J Biol Chem 277, 16241–16248.[Abstract/Free Full Text]

Ponnuraj, K., Bowden, M. G., Davis, S., Gurusiddappa, S., Moore, D., Choe, D., Xu, Y., Höök, M. & Narayana, S. V. (2003). A ‘dock, lock, and latch’ structural model for a staphylococcal adhesin binding to fibrinogen. Cell 115, 217–228.[Medline]

Poquet, I., Saint, V., Seznec, E., Simoes, N., Bolotin, A. & Gruss, A. (2000). HtrA is the unique surface housekeeping protease in Lactococcus lactis and is required for natural protein processing. Mol Microbiol 35, 1042–1051.[CrossRef][Medline]

van der Mee-Marquet, N., Achard, A., Mereghetti, L., Danton, A., Minier, M. & Quentin, R. (2003). Staphylococcus lugdunensis infections: high frequency of inguinal area carriage. J Clin Microbiol 41, 1404–1409.[Abstract/Free Full Text]

Vernachio, J., Bayer, A. S., Le, T. & 7 other authors (2003). Anti-clumping factor A immunoglobulin reduces the duration of methicillin-resistant Staphylococcus aureus bacteremia in an experimental model of infective endocarditis. Antimicrob Agents Chemother 47, 3400–3406.[Abstract/Free Full Text]

Wells, J. M., Wilson, P. W. & Le Page, R. W. F. (1993). Improved cloning vectors and transformation procedures for Lactococcus lactis. J Appl Bacteriol 74, 629–636.[Medline]

Received 18 May 2004; revised 26 July 2004; accepted 6 August 2004.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Mitchell, J.
Articles by Foster, T. J.
Articles citing this Article
PubMed
PubMed Citation
Articles by Mitchell, J.
Articles by Foster, T. J.
Agricola
Articles by Mitchell, J.
Articles by Foster, T. J.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2004 Society for General Microbiology.