SdrC induces staphylococcal biofilm formation through a homophilic interaction

Abstract

The molecular pathogenesis of many Staphylococcus aureus infections involves growth of bacteria as biofilm. In addition to polysaccharide intercellular adhesin (PIA) and extracellular DNA, surface proteins appear to mediate the transition of bacteria from planktonic growth to sessile lifestyle as well as biofilm growth, and can enable these processes even in the absence of PIA expression. However, the molecular mechanisms by which surface proteins contribute to biofilm formation are incompletely understood. Here we demonstrate that self-association of the serine-aspartate repeat protein SdrC promotes both bacterial adherence to surfaces and biofilm formation. However, this homophilic interaction is not required for the attachment of bacteria to abiotic surfaces. We identified the subdomain that mediates SdrC dimerization and subsequent cell-cell interactions. In addition, we determined that two adjacently located amino acid sequences within this subdomain are required for the SdrC homophilic interaction. Comparative amino acid sequence analysis indicated that these binding sites are conserved. In summary, our study identifies SdrC as a novel molecular determinant in staphylococcal biofilm formation and describes the mechanism responsible for intercellular interactions. Furthermore, these findings contribute to a growing body of evidence suggesting that homophilic interactions between surface proteins present on neighbouring bacteria induce biofilm growth.

Fig. 1. Specificity of enriched peptide-displaying phage

A. Schematic overall representation of SdrC domains. S, signal sequence; A region, composed of N1, N2 and N3; B repeats, B1 and B2; R, serine-aspartic acid repeat region; W, wall-spanning fragment; M, transmembrane domain; C, cytoplasmic tail; LPETG, cell wall anchoring motif; VAAPQ, cleavage site; the arrow indicates the enzymatic cleavage position. Also shown, are the enriched phage-displayed peptides aligned with the relevant SdrC sequence. Consensus sequences are highlighted in yellow. B. Purified phage clones binding to immobilized SdrC_N2N3, SdrG_N2N3 and BSA. Binding was detected with an anti-M13-HRP antibody. The data shown is representative of three individual experiments performed in triplicate. Bars represent mean ± standard error of the mean (SEM); *p<0.001.

Fig. 2. The N2 subdomain mediates SdrC dimerization

A. Biotin-labeled recombinant SdrC subdomains (1µM) binding to unlabeled segments. The interaction was detected with avidin-HRP. B and C. Dose-dependent and saturable binding of N2 (red) and N2N3 (blue) subdomains to one another. Increasing concentrations of biotin-labeled subdomains were incubated with either immobilized (B) unlabeled N2 or (C) N2N3. The apparent dissociation constant (K_D) calculated using the equation ΔP = ΔP_max [protein]/ (K_D + [protein]) ranged between 0.2 – 0.3 µM. D and E. Inhibition of N2N3 subdomain dimerization by phage displaying consensus peptide sequences (D) or anti-SdrC_N2N3 antibodies (E). Immobilized recombinant SdrC_N2N3 protein was first allowed to interact with phage or antibodies and then incubated with biotin-labeled recombinant SdrC_N2 protein. Insertless phage and anti-ClfB antibodies were used as negative controls. Binding was detected with an anti-M13-HRP antibody or avidin-HRP. The data shown is representative of three individual experiments performed in triplicate. Bars represent mean ± SEM; *p<0.001, **p<0.01. F. N2 subdomain-mediated dimer formation demonstrated by gel permeation chromatography. Pure recombinant SdrC subdomains were separated on a Sephadex 200 5/150 GL column at a flow rate of 0.3 ml min⁻¹. G. Interpolated relative molecular masses of SdrC species from the linear regression based on column calibration with thyroglobulin (Thy, 659 KDa), ferritin (Fer, 440 KDa), aldolase (Ald, 158 KDa), conalbumin (Con, 75 KDa), ovalbumin (Ova, 44 KDa), chymotrypsin (Chy, 29 KDa), cytochrome c (Cytc, 12.4 KDa), aprotinin (Fer, 6.5 KDa). Gel phase distribution coefficients (K_AV) were calculated from the respective elution volumes (Ve) and represented as a function of molecular mass

Fig. 3. SdrC contributes to staphylococcal biofilm formation

A and B. Biofilm formation on plastic plates in TSB 1% glucose medium (TBSG) by (A) S. aureus Newman, Newman ΔsdrC, Newman ΔsdrD, Newman ΔsdrCDE, and Newman ΔclfB or (B) S. aureus Newman DU6023 pCU1, DU6023 pCU1 sdrC, DU6023 pCU1 sdrD, and DU6023 pCU1 sdrE. Bacteria were added to microtiter wells at OD_{600 nm} = 0.01. C and D. Biofilm formation in M17 medium containing 0.5% glucose (GM17) by (C) L. lactis pKS80 or L. lactis pKS80 constitutively expressing SdrC, SdrE, ClfB, FnBPA or SasG or (D) L. lactis pNZ8037, L. lactis pNZ8037 sdrC, and L. lactis pNZ8037 clfB. Bacteria were added to microtiter wells at OD_{600 nm} = 0.01. Heterologous protein expression was induced with increasing concentrations of nisin. Static biofilm formation was measured by staining with 0.5% crystal violet (CV). The data shown is representative of three individual experiments performed in triplicate. Bars represent mean ± SEM; *p<0.001, ***p<0.05. Representative images of wells containing the CV extracted from previously stained biofilm are shown above the graphs. E. L. lactis pKS80, pKS80 sdrC, pKS80 sdrE colony spreading motility overnight incubation at 30°C. Bacteria from overnight cultures (2µl) were spotted on soft agar GM17 plates. The data shown is representative of six individual experiments. Bars represent mean ± SEM; *p<0.001. Representative images of colony swarming motility are shown above the graphs.

Fig. 4. Inhibition of SdrC dimerization disrupts biofilm formation

A. L. lactis pKS80 sdrC biofilm inhibition by anti-SdrC_N2N3 serum (range, 0 – 1 µM). L. lactis pKS80 sasG was used as negative control. B. L. lactis pKS80 sdrC biofilm inhibition by either recombinant SdrC_N2, SdrC_N3 SdrC_N2N3 proteins (range, 0 – 2.4 µM). Increasing concentrations of proteins were added to the plates at the same time as bacteria and incubated for 24 hours a 30°C. C. Inhibition of L. lactis pKS80 sdrC initial adherence to 96-well microtiter plates by anti-SdrC antibodies. Bacteria were grown overnight, washed in PBS, resuspended at OD_{600 nm} = 1 and added to plastic plates in the presence of either recombinant SdrC_N2N3 (3 µM) or antibodies (1 µM). After one hour, unbound bacteria were removed by washing with PBS. L. lactis pKS80 and L. lactis pKS80 sasG was used as negative control. Bound bacteria were stained with 0.5% CV. The data shown is representative of three individual experiments performed in triplicate. Bars represent mean ± SEM; *p<0.001. Representative images of wells containing the CV extracted from previously stained biofilm are shown above the graphs.

Fig. 5. Manganese inhibits SdrC-mediated biofilm formation

A. L. lactis pKS80 sdrC biofilm formation in the presence of metal ions. Bacteria (OD_{600 nm} = 0.01) and metal ions (1 mM) were added to microtiter wells at the same time and incubated for 24 hours a 30°C. L. lactis pKS80 clfB was used a positive control. B. EDTA metal chelation restores the ability of both SdrC and ClfB to promote biofilm growth of heterologous host L. lactis. The data shown is representative of two individual experiments performed in triplicate. C. Biotin-labeled rSdrC_N2 binding to immobilized rSdrC_N2N3 in the presence of Mn²⁺ (1µM) or Mn²⁺ (1µM) chelated with EDTA (10 µM). Bars represent mean ± SEM; **p<0.01. Representative images of wells containing the CV extracted from previously stained biofilm are shown above the graphs.

Fig. 6. SdrC contribution to biofilm formation is strain-dependent

A and B. Biofilm formation by staphylococcal laboratory and clinical strains in the presence of recombinant SdrC_N2 protein (3 µM), (A) plastic and (B) immobilized plasma (20 ng protein ml⁻¹). The data shown is representative of four individual experiments performed in triplicate. Bars represent mean ± SEM. C. SdrC nucleic acid sequences from 134 staphylococcal genomes deposited in the PATRIC database were analyzed for mutations in the dimerization site identified by phage display. The consensus sequence present in S. aureus Newman is highlighted in yellow; examples of polymorphisms are highlighted in green.