molecular oral microbiology molecular oral microbiology

Streptococcus gordonii DL1 adhesin SspB V-region mediates coaggregation via receptor polysaccharide of Actinomyces oris T14V C.R. Back, S.K. Douglas, J.E. Emerson, A.H. Nobbs and H.F. Jenkinson School of Oral and Dental Sciences, University of Bristol, Bristol, UK

Correspondence: Howard F. Jenkinson, School of Oral and Dental Sciences, University of Bristol, Lower Maudlin Street, Bristol BS1 2LY, UK Tel.: +44 1173 424424; fax: + 44 1173 424 313; E-mail: [email protected] Keywords: antigen I/II; cell wall protein; crystal structure; salivary pellicle Accepted 4 May 2015 DOI: 10.1111/omi.12106

SUMMARY Streptococcus gordonii SspA and SspB proteins, members of the antigen I/II (AgI/II) family of Streptococcus adhesins, mediate adherence to cysteine-rich scavenger glycoprotein gp340 and cells of other oral microbial species. In this article we investigated further the mechanism of coaggregation between S. gordonii DL1 and Actinomyces oris T14V. Previous mutational analysis of S. gordonii suggested that SspB was necessary for coaggregation with A. oris T14V. We have confirmed this by showing that Lactococcus lactis surrogate host cells expressing SspB coaggregated with A. oris T14V and PK606 cells, while L. lactis cells expressing SspA did not. Coaggregation occurred independently of expression of A. oris type 1 (FimP) or type 2 (FimA) fimbriae. Polysaccharide was prepared from cells of A. oris T14V and found to contain 1,4-, 4,6- and 3,4-linked glucose, 1,4-linked mannose, and 2,4-linked galactose residues. When immobilized onto plastic wells this polysaccharide supported binding of L. lactis expressing SspB, but not binding of L. lactis expressing other AgI/II family proteins. Purified recombinant NAVP region of SspB, comprising amino acid (aa) residues 41–847, bound A. oris polysaccharide but the C-domain (932– 1470 aa residues) did not. A site-directed deletion of 29 aa residues (D691–718) close to the predicted binding cleft within the SspB V-region © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Molecular Oral Microbiology 30 (2015) 411–424

ablated binding of the NAVP region to polysaccharide. These results infer that the V-region head of SspB recognizes an actinomyces polysaccharide ligand, so further characterizing a lectin-like coaggregation mechanism occurring between two important primary colonizers.

INTRODUCTION The human oral cavity is home to hundreds of different species of microorganism (Dewhirst et al., 2010), present in saliva and on hard or soft tissue surfaces as organized communities (Jenkinson, 2011). Residing within these communities are the bacteria and fungi that may become associated with initiation and progression of disease, e.g. caries or periodontitis (Wright et al., 2013). Control of community composition is therefore a vital factor in limiting disease, and also serves to protect against colonization of the oral cavity by incoming pathogens, and to prime the mucosa to respond to pathogenic microbial challenge (Wu et al., 2014). Streptococcus spp. and Actinomyces spp. make up over 40% of bacteria in early dental plaque (Nyvad & Kilian, 1990). These genera, as well as Gemella, Granulicatella, Rothia and Veillonella, are primary colonizers of the tooth and mucosal surfaces (Aas et al., 2005), and initiate the development of oral cavity 411

S. gordonii SspB binding Actinomyces oris

biofilm communities. In addition to providing a surface for secondary colonizing organisms such as Fusobacterium spp. or Porphyromonas gingivalis (Periasamy & Kolenbrander, 2009), these primary colonizers interact with each other physically (coaggregation/ coadhesion) and metabolically (Kolenbrander et al., 2010; Jenkinson, 2011; Nobbs et al., 2011). In these respects, the ubiquitous oral bacterium Streptococcus gordonii is able to interact with a wide range of coaggregating partners, including Actinomyces oris (Jakubovics et al., 2005), Streptococcus oralis (McNab et al., 1996), Fusobacterium nucleatum (Foster & Kolenbrander, 2004), P. gingivalis (Park et al., 2005), Aggregatibacter actinomycetemcomitans (Stacy et al., 2014) or Candida albicans (Bamford et al., 2009; Dutton et al., 2014). Although the molecular mechanisms of interactions between some of these pairs are well characterized, the interaction between S. gordonii and A. oris at the molecular level is poorly understood. Actinomyces oris expresses two fimbrial variants on its surface designated type 1 and type 2 (Cisar et al., 1988). Type 1 fimbriae have been shown to bind to components of salivary pellicle (Clark et al., 1989). Type 2 fimbriae on the other hand mediate lectin-like coaggregations between A. oris and other oral bacteria (Cisar et al., 1988). In particular, the FimA protein subunits of type 2 fimbriae are responsible for coaggregation of A. oris with streptococci that express cell wall phosphopolysaccharides containing the linkages GalNAcb1-3Gal or Galb1-3GalNAc (Yoshida et al., 2006; Mishra et al., 2011). Species that produce these polysaccharides include S. oralis, Streptococcus mitis, and Streptococcus sanguinis

C.R. Back et al.

(Yang et al., 2014). Conversely, A. oris T14V produces a polysaccharide that aggregates S. sanguinis H1 (Mizuno et al., 1983). Lectin-like interactions between actinomyces and streptococci can therefore be mediated in either direction. The oral (viridans) streptococci express antigen I/II (AgI/II) family polypeptide adhesins (Brady et al., 2010). These proteins are involved in binding a wide range of host receptors (Nobbs et al., 2009), mediate biofilm formation (Pecharki et al., 2005), and may promote coaggregation with oral microbial partners (Brooks et al., 1997; Egland et al., 2001). Streptococcus gordonii expresses two AgI/II family proteins on its surface, designated SspA and SspB, encoded by tandemly arranged genes on the S. gordonii DL1 chromosome (Demuth et al., 1996). It has been previously demonstrated that SspB of S. gordonii DL1, but not AgI/II proteins from Streptococcus mutans or Streptococcus intermedius, interacts with A. oris T14V (Jakubovics et al., 2005). SspB is a 164-kDa protein, and comprises seven distinct regions common to all AgI/II proteins (Brady et al., 2010) (Fig. 1). The protein folds to form a head (V-region) (Forsgren et al., 2009) held out ~60 nm from the cell surface by a stalk, comprising a polyproline type II helix intertwined with an extended a-helix (Larson et al., 2010). The stalk is supported by a globular C-terminal region, containing two intramolecular isopeptide bonds (Forsgren et al., 2010), which is covalently attached to the cell wall via an LPxTG motif (Brady et al., 2010) (see Fig. S1). It is thought that the Nterminal region of SspB determines coaggregation with A. oris T14V (Jakubovics et al., 2005), but the mechanism of this interaction is unknown.

Figure 1 Primary structure of SspB [1499 amino acid (aa) residues] from Streptococcus gordonii strain DL1 (NCBI YP_001449531). The main regions are as follows: 1–39 aa, leader (signal) peptide (S); 40–175 aa, N region; 175–463 aa, A repeats; 464–764 aa, V region; 765–909 aa, P repeats; 910–1413 aa, C-domain; 1414–1465 aa, cell wall spanning region (W); 1466–1470 aa, LP9TG motif; 1471–1499 aa, charged tail. Specific repeats are as follows: 175–197 aa, incomplete A repeat; 200–271 aa, A1 repeat; 282–353 aa, A2 repeat; 364–435 aa, A3 repeat; 446–463 aa, incomplete A repeat; 765–803 aa, P1 repeat; 804–842 aa, P2 repeat; 843–881 aa, P3 repeat; 882–909 aa, incomplete P repeat. Note the V-region crystal structure (Forsgren et al., 2009) was derived from S. gordonii strain M5 SspB protein (1500 aa residues) (SwissProt P16952.2).

412

© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Molecular Oral Microbiology 30 (2015) 411–424

C.R. Back et al.

S. gordonii SspB binding Actinomyces oris

This study aimed to identify the receptor present on A. oris that was recognized by SspB, and to determine the functional region(s) within SspB responsible for mediating this binding interaction. Our results show that SspB recognizes a carbohydrate (polysaccharide) receptor produced by A. oris. This interaction is mediated by SspB V-region, and occurs independently of fimbriae production by A. oris. METHODS Microbial growth conditions Bacteria and plasmids used in this study are listed in Table 1. Streptococcus gordonii, S. oralis and A. oris were cultured in BHY medium (3.7% LabM Brain–Heart infusion, 0.5% Bacto-yeast extract) in capped tubes or bottles under static conditions at 37°C. Lactococcus lactis MG1363 was cultured in GM17 broth (3.72% Difco-M17, 0.5% glucose) at 30°C under static conditions. Escherichia coli strains were cultured in Luria–Bertani (LB) medium (Fisher Scientific, Pittsburgh, PA) at 37°C with shaking at 200 r.p.m. Antibiotics were incorporated when necessary at the following concentrations: 100 lg ml 1 ampicillin (E. coli); 5 lg ml 1 erythromycin (S. gordonii or L. lactis).

Coaggregation of S. gordonii and A. oris Bacterial cells were harvested from cultures grown for 16 h at 37°C by centrifugation (5000 g, 10 min) and suspended in coaggregation buffer (1 mM Tris–HCl pH 8.0, 0.1 mM CaCl2, 0.1 mM MgCl2, 0.15 mM NaCl) at an optical density at 600 nm (OD600) of 1.0 (5 9 108 A. oris cells ml 1 or 1 9 109 streptococcal or lactococcal cells ml 1). Various combinations of streptococci, lactococci and actinomyces were tested for coaggregation by mixing equal volumes of two bacterial cell suspensions in glass tubes, with individual cell suspensions as controls. The suspensions were vortex-mixed for 10 s and allowed to stand for 1 h at 22°C. Coaggregation was scored visually from 0 to 4 (Cisar et al., 1979) as follows: 0 = no visible aggregates in the suspension; 1 = small coaggregates but suspension remained opaque; 2 = larger coaggregates but suspension still opaque and no settling of coaggregates; 3 = larger coaggregates that settled leaving a translucent suspension; 4 = large coaggregates that settled immediately leaving a clear fluid. In some experiments the extent of coaggregation was estimated by measuring the OD600 value of the assay fluid after sedimentation (Jenkinson, 1987). For measuring inhibition of coaggregation by glucose,

Table 1 Bacterial strains and plasmids utilized Bacterium

Genotype/Phenotype

References/Source

Streptococcus gordonii DL1 S. gordonii UB622 Streptococcus oralis 34 Actinomyces oris T14V A. oris PK455 A. oris 256-1 A. oris 455-2 A. oris PK606 Lactococcus lactis L. lactis UB1525 L. lactis UB1586 L. lactis UB1559 L. lactis UB1560 L. lactis UB1575 Escherichia coli NovaBlue GigaSingles

Wild type Challis DL1 DsspAsspB::ermAM Wild type Wild type (coaggregation group A) T14V fimA (Fim1+ Fim2 ) T14V fimP (Fim1 Fim2+) T14V fimP fimA (Fim1 Fim2 ) Wild type (coaggregation group D) MG1363 (pUB1000) SspA+ (pUB1000:sspA) EmR SspB+ (pUB1000:sspB) EmR SpaP+ (pUB1000:spaPS. mutans NG8) EmR SpaP+ (pUB1000:spaPS. mutans Guy’s 13) EmR Pas+ (pUB1000:pasS. intermedius ATCC 27335) EmR endA1 hsdR17(rK12 mK12+) supE44 thi-1 recA1 gyrA96 relA1 lac F’[proA+B+ lacIq ZΔM15::Tn10 (TcR)] F ompT hsdS(rB mB ) gal dcm k(DE3) pLysS (CamR) (pQE30:sspB-NAVP) ApR (pQE30:sspB-Cterm) ApR (pQE30:sspB-NAVPD691–718) ApR KmR

Laboratory stock Demuth et al. (1996) Cisar et al. (1979) Cisar et al. (1988) Cisar et al. (1988) Cisar et al. (1988) Cisar et al. (1988) Kolenbrander et al. (1983) Jakubovics et al. (2005) Jakubovics et al. (2005) Jakubovics et al. (2005) Jakubovics et al. (2005) Jakubovics et al. (2005) Jakubovics et al. (2005) Merck Millipore, Darmstadt, Germany

E. E. E. E.

coli coli coli coli

BL21(DE3) BL21(DE3) BL21(DE3) BL21(DE3)

pLys pLys pLys pLys

© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Molecular Oral Microbiology 30 (2015) 411–424

Merck Millipore, Darmstadt, Germany Nobbs et al. (2007) Nobbs et al. (2007) This study

413

S. gordonii SspB binding Actinomyces oris

mannose, galactose or lactose, S. gordonii DL1 or L. lactis SspB+ cells were incubated with A. oris cells in the absence or presence of 0.1 M sugar (final concentration). Coaggregation inhibition was expressed as a percentage (Jenkinson, 1987) based on the OD600 of the assay fluids compared with controls (no sugars). For protease treatment, bacterial cells in coaggregation buffer (OD600 1.0) were incubated with proteinase K (100 lg ml 1) for 30 min at 37°C. The cells were collected by centrifugation, washed twice with coaggregation buffer, and then assayed for coaggregation as described. For periodate treatment, bacterial cells were suspended at OD600 1.0 in 10 mM sodium periodate in 50 mM sodium acetate buffer pH 4.5 and incubated for 1 h at 22°C (Lamont et al., 1992). Controls were incubated in acetate buffer only. The cells were then collected by centrifugation, washed twice with coaggregation buffer, and assayed as described above. Polysaccharide extraction Cultures of A. oris Fim1 Fim2 mutant (Table 1) were grown in TY-Glucose medium (1% Bacto-tryptone, 0.5% yeast extract, 0.3% K2HPO4, 0.2% glucose, pH 7.5) for 16 h at 37°C. Cells were harvested by centrifugation, washed with deionized H2O (dH2O), suspended in dH2O at 100-fold concentration and dripped into liquid N2. The cell droplets were stored at 80°C. A 5–ml Teflon vessel and a tungsten carbide bead (diameter 7 mm) were pre-cooled in liquid N2, the frozen cell droplets were added, and the vessel was shaken for 2.5 min at 2000 r.p.m. using a Micro-Dismembrator (Sartorius Stedim Biotech GmbH, Goettingen, Germany). The powder was suspended in dH2O, centrifuged (7500 g, 10 min, 4°C), and the supernatant was removed and re-centrifuged (12,000 g, 10 min, 4°C). Polysaccharides were then part purified by a method that was adapted from Cisar et al. (1997). The supernatant was added to an equivalent volume of 10% trichloroacetic acid (TCA), vortex mixed, and incubated for 1 h at 4°C. After centrifugation (10,000 g, 15 min) the supernatant was dialyzed against dH2O for 16 h at 4°C. Ice-cold absolute ethanol was added (4:1) to the dialyzed extract, the mixture was vortex mixed and then stored overnight at 20°C. Precipitated polysaccharides were harvested by centrifugation (14,000 g, 30 min, 4°C) and the supernatant was carefully removed. The 414

C.R. Back et al.

pellets were left to air dry, and then dissolved in dH2O and stored at 20°C. Overall it was calculated that approximately 35 mg dry weight of extract was obtained from 500-ml culture of A. oris. The extract was further analyzed by sodium dodecyl sulfate– polyacrylamide gel electrophoresis (SDS–PAGE) to check for proteins, and by performing a Bradford protein assay, but there was < 0.1% protein present. Polysaccharide concentration was measured as glucose equivalents by the method of Dubois et al. (1956). The experiments generally yielded approximately 35 mg glucose equivalents of polysaccharides from 500 ml culture of A. oris. Polysaccharide analyses Polysaccharide extracts were first subjected to intact mass analysis. Extracts (4.4 mg ml 1) were injected onto an Applied Biosciences 4700 mass spectrometer for analysis by matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF/TOF) mass spectrometry. Glycosyl residue composition was analyzed by gas chromatography-mass spectrometry (GS-MS) of the per-O-trimethylsilyl derivatives of the monosaccharide methyl glycosides produced from the sample by acid methanolysis. The analysis tested for arabinose, fucose, glucuronic acid, mannose, glucose, Nacetyl-galactosamine, rhamnose, xylose, galacturonic acid, galactose, N-acetyl-mannosamine and Nacetyl-glucosamine. For linkage analysis, the sample was permethylated, depolymerized, reduced and acetylated, with the partially methylated alditol acetates analyzed by gas chromatography (GC)-MS. The analysis detected linkage positions for arabinose, fucose, glucuronic acid, mannose, glucose, rhamnose, xylose, galacturonic acid and galactose. Cell adhesion assay Polysaccharide extract (in water) was air dried at 37°C onto microtiter plate wells (Immulon 2HB 96 wells). In some experiments, purified gp340 (kindly €mberg, Umea Uniprovided by Professor Nicklas Stro versity, Sweden) in 50 mM Na2CO3-NaHCO3 pH 9 coating buffer was adsorbed onto plate wells for 16 h at 4°C (Jakubovics et al., 2005). Non-specific binding sites on the plastic were blocked with 1% bovine serum albumin (1 h at 22°C) and the wells were then washed with TBSC (10 mM Tris–HCl pH 7.6, 150 mM NaCl, © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Molecular Oral Microbiology 30 (2015) 411–424

C.R. Back et al.

5 mM CaCl2). Lactococcal cells in TBSC (OD600 1.0, approx. 109 cells ml 1) were added to microtitre plate wells (50 ll per well) and incubated for 2 h at 37°C. Wells were washed three times with TBS (10 mM Tris– HCl pH 7.2 containing 0.15 M NaCl) and bound cells were fixed with 25% formaldehyde for 30 min at 22°C. The wells were washed three times with TBS, and bound cells were stained with 0.5% crystal violet for 10 min. The wells were washed three times with TBS, residual dye was dissolved in 50 ll 7% acetic acid and cell biomass bound was quantified by measuring A595 (Jakubovics et al., 2005). Enzyme treatments of polysaccharide extract Actinomyces oris polysaccharide extract (0.4 mg ml 1 in TBS pH 7.0) was incubated with 10 U ml 1 salivary a-amylase (Sigma-Aldrich, St Louis, MO) or Penicillium dextranase (Sigma-Aldrich), or buffer only, for 1 h at 37°C. The mixtures were diluted 1 : 5 with absolute ethanol, vortex mixed and incubated for 16 h at 20°C. The extracts were centrifuged (14,000 g, 30 min, 4°C), the supernatants were removed, and the pellets were left to air dry. The pellets were then suspended in dH2O, concentrations were determined (Dubois et al., 1956), and the extracts were adjusted to 0.4 mg ml 1. Cloning of SspB sequences and generation of deletion mutant The NAVP coding region of SspB (41–847 aa residues) (Fig. 1) and C-domain coding region (932–1470 aa) (Fig. 1) were cloned into pQE30 as previously described (Nobbs et al., 2007). A short deletion of 29 aa residues (see Fig. S2 for details) was introduced into the SspB-NAVP coding region by inverse polymerase chain reaction (PCR) of pQE30:SspB-NAVP (Nobbs et al., 2007) with Phusion polymerase (NEB). The following primer pair was used: D691-720-F GCAGAAGCCGGCATTGTCGGTGAAATAACTCAAT CG and D691-720-R GCAGAAGCCGGCAGCACCGT ACCAATTATTTGGACTAG. The resulting PCR amplicon was digested with appropriate restriction enzymes (sites underlined in the above primers), ligated, and transformed into E. coli Gigasinglesâ. Colonies were screened by PCR and restriction enzyme digestion for desired plasmids, and the inserts were then fully sequenced. © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Molecular Oral Microbiology 30 (2015) 411–424

S. gordonii SspB binding Actinomyces oris

Purification of recombinant SspB fragments All recombinant proteins carried N-terminal x6His tags. The E. coli BL21 cells containing expression plasmids were grown in LB medium (100 ll ampicillin ml 1) at 37°C with shaking at 220 r.p.m. to OD600 0.4–0.6. Isopropyl-b-D-thiogalactopyranoside (1 mM) was added and the cultures were incubated for a further 16 h at 25°C, with shaking at 220 r.p.m. Cells were harvested by centrifugation, and the pellet was suspended in 5 ml BugBusterâ (Novagen, Darmstadt, Germany) per gram of cell pellet (wet weight). Benzonase (1 ll ml 1 of Bugbuster) was added, and the mixture was gently rocked for 30 min at 22°C. The slurry was centrifuged and the supernatant was collected. His-tagged proteins were then recovered by metal ion-affinity chromatography as previously described (Nobbs et al., 2007). Western immunoblot of SspB-NAVP fractions Portions of eluted fractions from affinity columns were subjected to SDS–PAGE, as described previously (Bamford et al., 2010), and proteins were stained with Coomassie Blue or analyzed by Western immunoblot to check for purity. Blots were probed with rabbit antibodies to S. mutans AgI/II (SpaP; kindly provided by Professor Charles Kelly, KCL), or with monoclonal mouse tetra-His antibody (Qiagen, Hilden, Germany) diluted 1 : 1000. Antibody binding was detected using horseradish peroxidase-conjugated swine anti-rabbit antibody or goat anti-mouse antibody (diluted 1 : 1000), respectively. The purity of the N-terminal NAVP recombinant protein fragments is shown in Fig. 2. The C-terminal fragment purity was as shown in Nobbs et al. (2007). Adherence to A. oris polysaccharide extract Polysaccharide extract was air dried onto microtiter plate wells (Immulon 2HB 96 wells) as before. Nonspecific binding sites on the plastic were blocked with 1% bovine serum albumin (1 h at 22°C) and the wells were then washed with TBSC. Recombinant proteins in TBSC were added to the wells (5 lg per well) and the plate was incubated for 1 h at 37°C. Wells were washed once with TBS, and incubated with anti-tetraHis antibody (diluted 1 : 1000 in TBS containing 1% Tween 20 and 3% BSA) (TBSTB) for 1 h at 37°C. The 415

S. gordonii SspB binding Actinomyces oris

A

C.R. Back et al.

B

Figure 2 SspB-NAVP region recombinant fragment purification. Western immunoblots of (A) NAVP fractions C1-C5 or (B) NAVP(D691–718) fractions C1–C4 eluted from Ni2+ column, blotted onto nitrocellulose, and probed with polyclonal antibody to AgI/II (anti-SpaP), or monoclonal anti-tetra-His antibody.

wells were then washed twice with TBST (TBS containing 0.1% Tween-20), incubated for 1 h at 37°C with anti-mouse-horseradish peroxidase antibody (diluted 1 : 2000 in TBSTB), washed once with TBST and twice with TBS. Developing reagent (0.102 M Na2HPO4, 0.049 M citric acid, 0.0013% H2O2, 4 mg ophenylenediamine) was added to the wells, the plate was incubated in the dark for 10 min at 22°C, and the reactions were stopped with 0.65 M H2SO4. Adherence was quantified by measuring A490 (Nobbs et al., 2007). RESULTS Actinomyces oris T14V-S. gordonii DL1 coaggregation is fimbriae independent Coaggregation of S. oralis 34 with A. oris T14V is known to be type 2 fimbriae-dependent (Cisar et al., 1988; Mishra et al., 2011). We confirmed this by assaying coaggregation of S. oralis 34 with a range

of A. oris T14V fimbrial mutants (Table 2). Mutants in fimA, defective in production of type 2 fimbriae, were unable to coaggregate with S. oralis 34. Conversely, coaggregation of S. gordonii DL1 with A. oris occurred in the presence or absence of fimbriae expression by the actinomyces (Table 2). The A. oris T14V coaggregated with S. gordonii DL1 to form a dense aggregate of cells that settled quickly to the bottom of the tube (see Fig. S3). Streptococcus gordonii DL1 also coaggregated with A. oris T14V lacking type 1 fimbriae, lacking type 2 fimbriae, or lacking both fimbrial types (see Fig. S3). This implied that fimbriae were not involved in the coaggregation (Table 2). Streptococcus gordonii SspB mediates coaggregation with A. oris By contrast, S. gordonii DsspAsspB cells did not coaggregate with A. oris T14V, Fim mutants, or

Table 2 Coaggregation scores of Actinomyces oris strains with Streptococcus gordonii, Streptococcus oralis, or Lactococcus lactis expressing S. gordonii SspA or SspB proteins Coaggregation score1 with:

Actinomyces strain

S. oralis 34

S. gordonii DL1

S. gordonii DL1 DsspAsspB

L. lactis MG1363 (pUB1000)

L. lactis MG1363 (pUB1000-SspA+)

L. lactis MG1363 (pUB1000-SspB+)

Wild-type T14V T14V Fim1+ Fim2 T14V Fim1 Fim2+ T14V Fim1 Fim2 PK6062

2 0 2 0 0

4 4 4 4 4

0 0 0 0 0

0 0 0 0 0

0 0 0 0 0

4 4 4 4 4

1

Visual scoring system from 4 (clumped sediment) to 0 (no coaggregation) as described in Methods. Coaggregation Group D (Kolenbrander et al., 2006).

2

416

© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Molecular Oral Microbiology 30 (2015) 411–424

C.R. Back et al.

S. gordonii SspB binding Actinomyces oris

Table 3 Effects of proteinase K treatment on Streptococcus gordonii or Streptococcus oralis coaggregation with Actinomyces oris Coaggregation score1 with A. oris strain: Streptococcus strain

T14V

T14V Fim1+ Fim2

T14V Fim1 Fim2+

T14V Fim1 Fim2

PK606

S. S. S. S.

4 0 0 3

3 0 0 0

4 0 0 4

4 0 0 0

3 0 0 0

gordonii DL1 gordonii DL1 + protease2 gordonii DL1 DsspAsspB + protease oralis 34 + protease

1

Visual scoring system from 4 (clumped sediment) to 0 (no coaggregation) as described in Methods. 2 9 109 cells ml 1 incubated with 100 lg ml 1 proteinase K for 30 min at 37°C.

2

coaggregation group D A. oris PK606 (Table 2). This confirmed a direct role for one or both of the AgI/II family proteins in mediating S. gordonii coaggregation with A. oris (Egland et al., 2001). Lactococcus lactis has been identified as an effective surrogate host for determining adherence functions of AgI/II proteins (Jakubovics et al., 2005). Accordingly L. lactis strains expressing SspA or SspB, or wild-type (WT) control (MG1363) were tested for coaggregation with A. oris T14V. The L. lactis WT cells did not coaggregate with A. oris and neither did L. lactis expressing SspA (Table 2; see Fig. S3). However, L. lactis cells expressing SspB showed a coaggregation phenotype similar to S. gordonii DL1 (Table 2; see Fig. S3). These results suggested that SspB, but not SspA, was a coaggregation adhesin for A. oris T14V, as well as for A. oris PK606 (Table 2). Coaggregations of S. gordonii DL1 or L. lactis expressing SspB with A. oris T14V were tested for inhibition by several different carbohydrates. Coaggregation of S. gordonii DL1 with A. oris T14V was partially inhibited by 0.1 M glucose (33% inhibition) or mannose (36%), and inhibited more by 0.1 M galactose (49% inhibition) or 0.1 M lactose (58%) (data not shown). Lactococcus lactis expressing SspB was not inhibited in coaggregation with A. oris T14V by any of the sugars tested (data not shown). Streptococcus gordonii SspB recognizes receptor carbohydrate on A. oris To determine further the biochemical nature of this interaction, S. gordonii or L. lactis cells were incubated with proteinase K (100 lg ml 1, 37°C, 30 min), washed and tested for coaggregation. Incubation of S. gordonii with protease ablated coaggregation with A. oris T14V and PK606 (Table 3). Proteasetreatment of L. lactis expressing SspB also abrogated © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Molecular Oral Microbiology 30 (2015) 411–424

coaggregation with A. oris T14V and PK606. This concurred with the evidence that surface protein SspB mediated coaggregation. Treatment of S. oralis 34 cells with protease did not affect coaggregation with A. oris T14V or with the Fim1 mutant (Table 3). Protease treatment of A. oris T14V had little or no effect on the coaggregation reactions with S. gordonii DL1 or L. lactis expressing SspB (Table 4). However, treatment of A. oris T14V cells with periodate led to reduced coaggregation with S. gordonii DL1 and ablated coaggregation with L. lactis cells expressing SspB (Table 4). Similar effects were seen by incubating A. oris Fim1 Fim2 with protease or periodate, respectively (Table 4). Collectively these results showed that a non-fimbrial and non-protein component of A. oris functioned as a receptor for SspB. This was most likely to be a periodate-sensitive receptor carbohydrate component of the cell wall or surface layers. Table 4 Effects of proteinase K or periodate treatments on Actinomyces oris coaggregation with Streptococcus gordonii or Lactococcus lactis expressing S. gordonii SspB protein Coaggregation score1 with:

A. oris strain

S. gordonii DL1

L. lactis MG1363 (pUB1000-SspB+)

T14V T14V T14V T14V T14V T14V

4 4 1 4 4 0

4 4 0 4 4 0

+ protease2 + periodate3 Fim1 Fim2 Fim1 Fim2 + protease Fim1 Fim2 + periodate

1

Visual scoring system from 4 (clumped sediment) to 0 (no coaggregation) as described in Methods. 2 1 9 109 cells ml 1 incubated with 100 lg ml 1 proteinase K for 30 min at 37°C. 3 1 9 109 cells ml 1 incubated with 10 mM sodium periodate for 1 h at 22°C.

417

S. gordonii SspB binding Actinomyces oris

C.R. Back et al.

Extraction and characterization of A. oris receptor polysaccharide

Actinomyces oris polysaccharide supports SspBmediated bacterial adhesion

To obtain polysaccharides from A. oris T14V, cells were physically disrupted (see Methods), fractionated, and mixed with 10% TCA. Polysaccharides were then collected from the TCA-soluble fraction following ethanol precipitation. The components were first subjected to MALDI-TOF mass spectrometry to investigate overall structural features. In the spectrum (Fig. 3) there was a series of peaks, decreasing in height and separated by 160–180 m/z (molecular mass). This was characteristic of a repeating structure of regular units, possibly pentose or hexose in view of the molecular mass differences. However, there was evidence of a secondary lessregular repeat structure being present in the samples, appearing on the readout as smaller peaks close to the main peaks (Fig. 3). Composition and linkage analyses were performed by GC-MS. The primary composition analysis indicated that the sample consisted of 90% glucose (Glc). The linkage analyses provided a more comprehensive overview of the structures present comprising: 87.6% 1,4-Glc; 6.9% terminalGlc; 4.2% 4,6-Glc; 0.6% 3,4-Glc; 0.4% 1,4-mannose; and 0.3% 2,4-galactose. Collectively the results suggested that the extract contained primarily a1,4-linked glucan with a minor polysaccharide component comprising mannose, galactose and glucose.

To analyze the binding of SspB and other AgI/II family proteins to A. oris polysaccharide, portions of the polysaccharide preparation were immobilized onto microtiter-plate wells. Binding of lactococcal cells was measured by crystal violet assay (see Methods). Lactococcus lactis MG1363 WT cells showed no adherence to the immobilized polysaccharide (Fig. 4). Furthermore, none of the AgI/II family proteins SspA, SpaPNG8, SpaPGuys or Pas mediated binding of L. lactis cells to polysaccharide. However, L. lactis cells expressing SspB bound avidly to increasing amounts of the immobilized polysaccharide (Fig. 4), indicating that SspB had specificity for interaction with the polysaccharide. Treatment of the polysaccharide extract with aamylase reduced the overall carbohydrate concentration by ~40%, but did not reduce the levels of binding by L. lactis expressing SspB (data not shown). Likewise treatment with dextranase had no effects on binding levels. We also tested binding of L. lactis cells expressing SspB to a range of immobilized substrata including amylose (a1,4-glucose chains), glycogen (a1,4-glucan with a1,6-branches), dextran (a1,6-glucan with a1,3-branches), laminarin (b1,3-glucan with b1,6-branches) and yeast mannan (a1,6mannose with a1,2- and a1,3-branches). None of these substrata supported adherence of L. lactis expressing SspB (data not shown).

A

B

C D

418

Figure 3 Characterization of Actinomyces oris T14V polysaccharide by MALDI-TOF mass spectrometry. Extract (4.4 mg ml 1) was injected onto an Applied Biosciences 4700 mass spectrometer for analysis by MALDI-TOF spectroscopy. Insert shows examples of carbohydrate linkages present within the polysaccharide from the results of linkage analysis. (A) 4-linked glucopyranosyl (4-Glc); (B) 2,4-linked galactopyranosyl (2,4Gal); (C) 4,6-linked glucopyranosyl (4,6-Glc); and (D) 4-linked mannopyranosyl (4-Man).

© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Molecular Oral Microbiology 30 (2015) 411–424

C.R. Back et al.

S. gordonii SspB binding Actinomyces oris

Figure 4 Adherence of Lactococcus lactis cells expressing various AgI/II family proteins to wells coated with polysaccharide from Actinomyces oris T14V. The L. lactis expression constructs were described previously (Jakubovics et al., 2005): SspA (Streptococcus gordonii), SspB (S. gordonii), SpaP (Streptococcus mutans NG8), SpaPGuys (S. mutans Guys13), and Pas (Streptococcus intermedius). Lactococcus lactis harboring empty vector pUB1000 was negative control (wild type; WT). Increasing concentrations of polysaccharide were air dried onto Immulon microtiter plates and blocked using 1% bovine serum albumin. The L. lactis cells (5 9 107) were added to each well, and cells were detected by staining with crystal violet (see Methods). Blocked wells that were not coated with the polysaccharides were subtracted as background values. Error bars represent  standard deviation of two independent experiments performed in triplicate.

SspB V-region binds polysaccharide The SspB protein V-region folds to form a head supported by a stalk generated by interaction of the A (alanine-rich) and P (proline-rich) regions (Brady et al., 2010; Larson et al., 2010). The structure overall is held by a globular C-domain (Forsgren et al., 2010) that is covalently linked to the cell wall via a conventional LPxTG sortase-recognition motif (see Fig. S1). To test binding of these two main SspB protein regions to polysaccharide, we generated x6His-tagged recombinant N-terminal (NAVP) region and C-domain (C) fragments of the SspB polypeptide (Fig. 1). Binding of these fragments to immobilized polysaccharide was assayed with anti-tetra-His antibody and with horseradish peroxidase-linked secondary antibody. The SspB NAVP region showed binding to the immobilized polysaccharide, whereas the SspB C-domain demonstrated no binding (Fig. 5). To determine if the V-region alone was responsible, a site-directed deletion of 29 aa residues was generated in a sequence implicated to form a cap over the potential binding pocket (Forsgren et al., 2009) (see Fig. S2). This deletion was predicted to remove regions b16, b17 and b18 from the structure © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Molecular Oral Microbiology 30 (2015) 411–424

(Fig. 6) and to displace b13. The binding pocket would consequently be more exposed (see Fig. S4). The NAVP(D691–718) deletion fragment showed no binding to A. oris polysaccharide (Fig. 5). In separate assays we measured binding of the NAVP and NAVP(D691–718) fragments to gp340, the recognized main receptor for AgI/II polypeptides (Brady et al., 2010). We found that the NAVP(D691–718) deletion did not affect binding to immobilized gp340 (data not shown). These results suggest therefore that the V region of SspB specifically interacts with A. oris T14V polysaccharide. DISCUSSION The physical interactions between microorganisms are believed to be crucial processes in the development of oral cavity biofilms. These interactions, which may be detected by coaggregation assays in the laboratory, enable microorganisms to co-adhere to oral or dental surfaces, co-metabolize, and form microbial communities. Coaggregation reactions occur between early colonizers, between early and late colonizers, and between late colonizers (Kolenbrander et al.,

419

S. gordonii SspB binding Actinomyces oris

C.R. Back et al.

Figure 5 Adherence of SspB recombinant protein fragments to immobilized Actinomyces oris T14V polysaccharide. Equal amounts of each recombinant protein (5 lg) were added to wells containing increasing amounts of immobilized polysaccharide. Recombinant protein bound was detected by enzyme-linked immunosorbent assay with anti-tetra-His primary antibody and anti-mouse IgG-horseradish peroxidase-linked secondary antibody, as described in Methods. Values were corrected for adherence of recombinant proteins to blocked wells (no polysaccharide). Error bars represent  standard deviation from three independent experiments performed in duplicate.

A

B

Figure 6 Crystal structure of the Streptococcus gordonii SspB V-region. A the two subdomains A and B (SDA and SDB) flank a central bsandwich, consisting of two anti-parallel b-sheets packed against each other, with a putative binding pocket stabilized by a Ca2+ ion (silver sphere) (Forsgren et al., 2009). The a-helices are labeled from 1 to 3, and b-sheets are labeled from 1 to 20. N terminus = blue, C terminus = red. PDB accession code: 2wd6. (B) A model of the V-region with a 29 amino acid residue deletion (A691–G718), equivalent to the sequence colored dark gray in (A). The model (B) was generated using PHYRE2 (Kelley & Sternberg, 2009).

2006; Periasamy & Kolenbrander, 2009). In this study we focused on the interaction between S. gordonii DL1 and A. oris T14V, with the knowledge that mutants in S. gordonii deleted in sspB were unable to coaggregate with A. oris T14V (Egland et al., 2001). We have now shown that SspB directly mediates this reaction. By expressing the protein on coaggregationinert surrogate host cells of L. lactis, these were endowed with the ability to coaggregate with A. oris 420

strains T14V and PK606. The interaction mediated by SspB was not inhibited by the inclusion of 0.1 M glucose, mannose, galactose or lactose in the coaggregation reactions. Actinomyces oris T14V is a member of coaggregation group A (Kolenbrander et al., 1983, 2006), and provides both adhesins and receptors for coaggregation with oral streptococci. The major coaggregation adhesin on strain T14V is the type 2 fimbria, © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Molecular Oral Microbiology 30 (2015) 411–424

C.R. Back et al.

which mediates lactose-sensitive binding of actinomyces cells to strains of S. oralis, S. mitis and S. sanguinis (see Introduction). A coaggregation receptor on A. oris T14V, recognized by S. sanguinis H1, was identified as a complex cell wall-associated polysaccharide of unknown structure. The polysaccharide units comprised 6-deoxy-L-talose (6DOT), rhamnose (Rha), and N-acetylgalactosamine (GalNAc) (Mizuno et al., 1983). The isolated polysaccharide strongly aggregated S. sanguinis H1 but was much less effective at inhibiting coaggregation. The polysaccharide also weakly aggregated S. sanguinis (gordonii) DL1. Actinomyces oris PK606 is a member of coaggregation group D (Kolenbrander et al., 2006), which is also reported to express a receptor polysaccharide of unknown structure and composition. Our data suggest that SspB recognizes a ligand within a polysaccharide produced by A. oris T14V that is also present on the surface of PK606. Clearly to further this study it would be relevant to extract polysaccharide from strain PK606 and, together with the polysaccharide that we have already identified, try to improve the structural analyses to determine the specific ligand for SspB. However, based on our composition and linkage analyses, it would seem that the polysaccharide is not identical to that identified by Mizuno et al. (1983). Our analyses would have detected all relevant carbohydrates (see Methods). Our polysaccharide might be more related to the extracellular polysaccharides from A. oris T14V and T14Av, described by Imai & Kuramitsu (1983), which contained Gal, Glc and Man, in addition to 6-DOT, Rha and GalNAc. The polysaccharide that we prepared was not active in coaggregating S. gordonii cells or L. lactis cells expressing SspB, and neither did it measurably inhibit coaggregation of these bacteria with A. oris T14V (data not shown). This would be in keeping with the notion that the SspB receptor polysaccharide was not a major component of the preparation which, nevertheless, supported adherence of SspB-expressing bacteria. We tried to get more structural information on the polysaccharide by H-NMR but the mixture was too complex. We also tried to fractionate the polysaccharide further by chromatography but lost the receptor activity. The polysaccharide was extracted from disrupted cells, and so would have included intracellular polysaccharide in addition to cell wall material © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Molecular Oral Microbiology 30 (2015) 411–424

S. gordonii SspB binding Actinomyces oris

and surface-associated polysaccharide. To preferentially obtain surface-associated polysaccharides we extracted A. oris T14V cells with ice-cold 0.1 M NaOH, to retrieve alkali-soluble polysaccharides, but no SspB-binding activity was detected. We also precipitated polysaccharides from spent culture medium, but no activity was obtained. Further structural information will not be possible therefore until the receptor activity is purified to much higher specific activity. Previous work suggested that the N-terminal region of SspB (including the A and V domains) determined coaggregation specificity with A. oris T14V (Jakubovics et al., 2005). As none of the other AgI/II family proteins tested were active in this respect it was hypothesized that the variable domain of the SspB protein might be the region responsible for adhesion specificity. The V-region of SspB is only 27% identical to the V-region of SspA (Demuth et al., 1996), whereas the two C-domains are 98% identical. The crystal structure of the SspB V-domain has a central b-supersandwich, which consists of two anti-parallel b sheets folded against one another (Fig. 6). The central region is flanked by two sub-domains (SDA and SDB). There is a central cavity that is generally hypothesized to be a binding pocket, and is stabilized by a Ca2+ ion. This ion is coordinated by three water molecules and four separate protein atoms: Ser648 O and Oc, Asn650 Od1 from b sheet 13, and Glu664 Oe1 from b sheet 14. The binding cleft of SspB appears to be smaller than that of the S. mutans SpaP protein V-domain (Troffer-Charlier et al., 2002), and so this is thought to indicate differential ligand specificities (Forsgren et al., 2009). By deleting aa residues 691–718 and exposing the binding pocket (see Fig. S4), we abrogated binding of NAVP(D691–718) to polysaccharide, but not to gp340. Taking into consideration the overall structure of SspB, and the regions along the stalk and within the C-domain designated to bind gp340 (Brady et al., 1992; Moisset et al., 1994; Senpuku et al., 1995) (see Fig. S1), it is strongly suggested that the A. oris T14V polysaccharide interacts only with the V-region binding pocket. In addition, the deletion mutagenesis did not affect binding levels to gp340, supporting evidence that the V-region is not necessary for SspB binding to immobilized gp340 (Jakubovics et al., 2005; Maddocks et al., 2011). More precise mutagenesis would be needed to determine the critical residues involved in polysaccharide binding, and to 421

S. gordonii SspB binding Actinomyces oris

prove that the polysaccharide bound within the predicted binding pocket (trench). The SspB protein has multiple binding functions and a new activity directly involved in coaggregation of early colonizers has been characterized here. The SspB protein is now known to have at least three different microbial partners, A. oris, P. gingivalis (Brooks et al., 1997) and C. albicans (Silverman et al., 2010), in addition to recognizing gp340, fibronectin (Nobbs et al., 2007) and collagen (Love et al., 1997). The interaction of SspB with A. oris seems to be located to the V-region. On the other hand, interaction of SspB with P. gingivalis is located in the C-domain, downstream from the P-rich repeats (Brooks et al., 1997). The discrete locations of various binding sites on SspB for interaction with different microorganisms, and host molecules, therefore ensures maximum protein functionality. This could be envisaged as being a key mechanism for generating diversity in the development of oral microbial communities, and underlies the potential importance of SspB (and other AgI/II proteins) in construction of the oral microbiome. ACKNOWLEDGEMENTS We would like to thank Professors Charles Kelly and €mberg for the provision of antiserum and Nicklas Stro gp340, respectively; the University of Bristol School of Chemistry MS Service (Dr Paul Gates) for performing MALDI-TOF MS; the Complex Carbohydrate Research Centre, University of Georgia, Athens, GA, USA for composition and linkage analyses; and Dr Paul Race, School of Biochemistry, for helpful discussions. We thank Sara Rego for assistance with preparing sequence data and running protein analysis software. A Society for General Microbiology UK Vacation Studentship awarded to SKD is gratefully acknowledged. REFERENCES Aas, J.A., Paster, B.J., Stokes, L.N., Olsen, I. and Dewhirst, F.E. (2005) Defining the normal bacterial flora of the oral cavity. J Clin Microbiol 43: 5721– 5732. Bamford, C.V., d’Mello, A., Nobbs, A.H. et al. (2009) Streptococcus gordonii modulates Candida albicans biofilm formation through intergeneric communication. Infect Immun 77: 3696–3704.

422

C.R. Back et al.

Bamford, C.V., Francescutti, T., Cameron, C.E., Jenkinson, H.F. and Dymock, D. (2010) Characterization of a novel family of fibronectin-binding proteins with M23 peptidase domains from Treponema denticola. Mol Oral Microbiol 25: 369–383. Brady, L.J., Piacentini, D.A., Crowley, P.J., Oyston, P.C.F. and Bleiweis, A.S. (1992) Differentiation of salivary agglutinin-mediated adherence and aggregation of mutans streptococci by use of monoclonal antibodies against the major surface adhesin P1. Infect Immun 60: 1008–1017. Brady, L.J., Maddocks, S.E., Larson, M.R. et al. (2010) The changing faces of Streptococcus antigen I/II polypeptide family adhesins. Mol Microbiol 77: 276–286. Brooks, W., Demuth, D.R., Gil, S. and Lamont, R.J. (1997) Identification of a Streptococcus gordonii SspB domain that mediates adhesion to Porphyromonas gingivalis. Infect Immun 65: 3753–3758. Cisar, J.O., Kolenbrander, P.E. and McIntire, F.C. (1979) The specificity of coaggregation reactions between human oral streptococci and strains of Actinomyces viscosus or Actinomyces naeslundii. Infect Immun 24: 742–752. Cisar, J.O., Vatter, A.E., Clark, W.B., Curl, S.H., HurstCalderone, S. and Sandberg, A.L. (1988) Mutants of Actinomyces viscosus T14V lacking type 1, type 2, or both types of fimbriae. Infect Immun 56: 2984–2989. Cisar, J.O., Sandberg, A.L., Reddy, G.P., Abeygunawardana, C. and Bush, C.A. (1997) Structural and antigenic types of cell wall polysaccharides from viridans group streptococci with receptors for oral actinomyces and streptococcal lectins. Infect Immun 65: 5035–5041. Clark, W.B., Beem, J.E., Nesbitt, W.E., Cisar, J.O., Tseng, C.C. and Levine, M.J. (1989) Pellicle receptors for Actinomyces viscosus type 1 fimbriae in vitro. Infect Immun 57: 3003–3008. Demuth, D.R., Duan, Y., Brooks, W., Holmes, A.R., McNab, R. and Jenkinson, H.F. (1996) Tandem genes encode cell-surface polypeptides SspA and SspB which mediate adhesion of the oral bacterium Streptococcus gordonii to human and bacterial receptors. Mol Microbiol 20: 403–413. Dewhirst, F.E., Chen, T., Izard, J. et al. (2010) The human oral microbiome. J Bacteriol 192: 5002–5017. Dubois, M., Gilles, K.A., Hamilton, J.K., Rebers, P.A. and Smith, F. (1956) Colorimetric method for determination of sugars and related substances. Anal Chem 28: 350–356. Dutton, L.C., Nobbs, A.H., Jepson, K. et al. (2014) O-mannosylation in Candida albicans enables development of interkingdom biofilm communities. mBio 5: e00911.

© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Molecular Oral Microbiology 30 (2015) 411–424

C.R. Back et al.

Egland, P.G., Du, L.D. and Kolenbrander, P.E. (2001) Identification of independent Streptococcus gordonii SspA and SspB functions in coaggregation with Actinomyces naeslundii. Infect Immun 69: 7512–7516. Forsgren, N., Lamont, R.J. and Persson, K. (2009) Crystal structure of the variable domain of the Streptococcus gordonii surface protein SspB. Protein Sci 18: 1896– 1905. Forsgren, N., Lamont, R.J. and Persson, K. (2010) Two intramolecular isopeptide bonds are identified in the crystal structure of the Streptococcus gordonii SspB Cterminal domain. J Mol Biol 397: 740–751. Foster, J.S. and Kolenbrander, P.E. (2004) Development of a multispecies oral bacterial community in a salivaconditioned flow cell. Appl Environ Microbiol 70: 4340– 4348. Imai, S. and Kuramitsu, H. (1983) Chemical characterization of extracellular polysaccharides produced by Actinomyces viscosus T14V and T14Av. Infect Immun 39: 1059–1066. Jakubovics, N.S., Stromberg, N., van Dolleweerd, C.J., Kelly, C.G. and Jenkinson, H.F. (2005) Differential binding specificities of oral streptococcal antigen I/II family adhesins for human or bacterial ligands. Mol Microbiol 55: 1591–1605. Jenkinson, H.F. (1987) Novobiocin-resistant mutants of Streptococcus sanguis with reduced cell hydrophobicity and defective in coaggregation. J Gen Microbiol 133: 1909–1918. Jenkinson, H.F. (2011) Beyond the oral microbiome. Environ Microbiol 13: 3077–3087. Kelley, L.A. and Sternberg, M.J.E. (2009) Protein structure predication on the Web: a case study using the Phyre server. Nat Protoc 4: 363–371. Kolenbrander, P.E., Inouye, Y. and Holdeman, L.V. (1983) New Actinomyces and Streptococcus coaggregation groups among human oral isolates from the same site. Infect Immun 41: 501–506. Kolenbrander, P.E., Palmer, R.J. Jr, Rickard, A.H., Jakubovics, N.S., Chalmers, N.I. and Diaz, P.I. (2006) Bacterial interactions and successions during plaque development. Periodontol 2000 42: 47–79. Kolenbrander, P.E., Palmer, R.J. Jr, Periasamy, S. and Jakubovics, N.S. (2010) Oral multispecies biofilm development and the key role of cell–cell distance. Nat Rev Microbiol 8: 471–480. Lamont, R.J., Hersey, S.G. and Rosan, B. (1992) Characterization of the adherence of Porphyromonas gingivalis to oral streptococci. Oral Microbiol Immunol 7: 193–197.

© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Molecular Oral Microbiology 30 (2015) 411–424

S. gordonii SspB binding Actinomyces oris

Larson, M.R., Rajashankar, K.R., Patel, M.H. et al. (2010) Elongated fibrillar structure of a streptococcal adhesin assembled by the high-affinity association of a- and PPII-helices. Proc Natl Acad Sci USA 107: 5983–5988. Love, R.M., McMillan, M.D. and Jenkinson, H.F. (1997) Invasion of dentinal tubules by oral streptococci is associated with collagen recognition mediated by the antigen I/II family of polypeptides. Infect Immun 65: 5157–5164. Maddocks, S.E., Wright, C.J., Nobbs, A.H. et al. (2011) Streptococcus pyogenes antigen I/II-family polypeptide AspA shows differential ligand-binding properties and mediates biofilm formation. Mol Microbiol 81: 1034– 1049. McNab, R., Holmes, A.R., Clarke, J.M., Tannock, G.W. and Jenkinson, H.F. (1996) Cell surface polypeptide CshA mediates binding of Streptococcus gordonii to other oral bacteria and to immobilized fibronectin. Infect Immun 64: 4204–4210. Mishra, A., Devarajan, B., Reardon, M.E. et al. (2011) Two autonomous structural modules in the fimbrial shaft adhesin FimA mediate Actinomyces interactions with streptococci and host cells during oral biofilm development. Mol Microbiol 81: 1205–1220. Mizuno, J., Cisar, J.O., Vatter, A.E., Fennessey, P.V. and McIntire, F.C. (1983) A factor from Actinomyces viscosus T14V that specifically aggregates Streptococcus sanguis H1. Infect Immun 40: 1204–1213. Moisset, A., Schatz, N., Lepoivre, Y. et al. (1994) Conservation of salivary glycoprotein-interacting and human immunoglobulin G-cross-reactive domains of antigen I/II in oral streptococci. Infect Immun 62: 184–193. Nobbs, A.H., Shearer, B.H., Drobni, M., Jepson, M.A. and Jenkinson, H.F. (2007) Adherence and internalization of Streptococcus gordonii by epithelial cells involves b1 integrin recognition by SspA and SspB (antigen I/II family) polypeptides. Cell Microbiol 9: 65–83. Nobbs, A.H., Lamont, R.J. and Jenkinson, H.F. (2009) Streptococcus adherence and colonization. Microbiol Mol Biol Rev 73: 407–450. Nobbs, A.H., Jenkinson, H.F. and Jakubovics, N.S. (2011) Stick to your gums: mechanisms of oral microbial adherence. J Dent Res 90: 1271–1278. Nyvad, B. and Kilian, M. (1990) Comparison of the initial streptococcal microflora on dental enamel in cariesactive and in caries-inactive individuals. Caries Res 24: 267–272. Park, Y., Simionato, M.R., Sekiya, K. et al. (2005) Short fimbriae of Porphyromonas gingivalis and their role in coadhesion with Streptococcus gordonii. Infect Immun 73: 3983–3989.

423

S. gordonii SspB binding Actinomyces oris

Pecharki, D., Petersen, F.C., Assev, S. and Scheie, A.A. (2005) Involvement of antigen I/II surface proteins in Streptococcus mutans and Streptococcus intermedius biofilm formation. Oral Microbiol Immunol 20: 366–371. Periasamy, S. and Kolenbrander, P.E. (2009) Mutualistic biofilm communities develop with Porphyromonas gingivalis and initial, early, and late colonizers of enamel. J Bacteriol 191: 6804–6811. Senpuku, H., Miyauchi, T., Hanada, N. and Nisizawa, T. (1995) An antigenic peptide inducing cross-reacting antibodies inhibiting the interaction of Streptococcus mutans PAc with human salivary components. Infect Immun 63: 4695–4703. Silverman, R.J., Nobbs, A.H., Vickerman, M.M., Barbour, M.E. and Jenkinson, H.F. (2010) Interaction of Candida albicans cell wall Als3 protein with Streptococcus gordonii SspB adhesin promotes development of mixed-species communities. Infect Immun 78: 4644–4652. Stacy, A., Everett, J., Jorth, P., Trivedi, U., Rumbaugh, K.P. and Whiteley, M. (2014) Bacterial fight-andflight responses enhance virulence in a polymicrobial infection. Proc Natl Acad Sci USA 111: 7819–7824. Troffer-Charlier, N., Ogier, J., Moras, D. and Cavarelli, J. (2002) Crystal structure of the V-region of Streptococcus

424

C.R. Back et al.

mutans antigen I/II at 2.4  A resolution suggests a sugar preformed binding site. J Mol Biol 318: 179–188. Wright, C.J., Burns, L.H., Jack, A.A. et al. (2013) Microbial interactions in building of communities. Mol Oral Microbiol 28: 83–101. Wu, R.Q., Zhang, D.F., Tu, E., Chen, Q.M. and Chen, W. (2014) The mucosal immune system in the oral cavity – an orchestra of T cell diversity. Int J Oral Sci 6: 125– 132. Yang, J., Yoshida, Y. and Cisor, J.O. (2014) Genetic basis of coaggregation receptor polysaccharide biosynthesis in Streptococcus sanguinis and related species. Mol Oral Microbiol 29: 24–31. Yoshida, Y., Palmer, R.J. Jr, Yang, J., Kolenbrander, P.E. and Cisar, J.O. (2006) Streptococcal receptor polysaccharides: recognition molecules for oral biofilm formation. BMC Oral Health 6(Suppl 1): S12.

SUPPORTING INFORMATION Additional supporting information may be found in the online version of this article at the publisher’s web-site.

© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Molecular Oral Microbiology 30 (2015) 411–424