Porphyromonas gingivalis Minor Fimbriae Are Required for Cell-Cell Interactions

Abstract

Two distinctive types of fimbriae have been identified in Porphyromonas gingivalis. In this report, we demonstrate that minor fimbriae are involved in P. gingivalis autoaggregation and colonization. A mutant with a deficiency in minor fimbriae can bind to a saliva-coated surface but does not form microcolonies as the wild-type strain does.

Porphyromonas gingivalis is a gram-negative bacterium closely associated with severe adult periodontitis (16). It is well known that the bacterium can adhere to different surfaces in the oral cavity such as teeth, oral mucosa, and other oral bacteria (8). This adherence ability is important for bacterial colonization and is driven predominantly by peritrichous fimbriae. Two distinct fimbriae are present on the surface of P. gingivalis cells (1). Major fimbriae are long, peritrichous, filamentous components. The subunit of the major fimbriae is a 41-kDa protein (FimA, fimbrillin) and is encoded by the fimA gene (2). A thin, short secondary fimbrial structure, termed minor fimbriae or short fimbriae (15), is composed of a 67-kDa protein encoded by the mfa1 gene (4). Both major and minor fimbriae appear to contribute to the pathogenicity of P. gingivalis (1). The primary role of FimA is to promote bacterial attachment to oral surfaces. A deficiency in the fimA gene leads to a diminished capacity to adhere to human gingival fibroblasts and epithelial cells (5) and attenuation of periodontal bone loss in a gnotobiotic-rat model (10). Major fimbriae are also required for P. gingivalis intracellular invasion. P. gingivalis fimbria-deficient mutants have a reduced ability to invade human gingival epithelial and KB cells (11, 18, 19). Furthermore, major fimbriae interrupt the host immune system by inducing human peripheral macrophages and neutrophils to overproduce several proinflammatory cytokines such as interleukin-1 (IL-l), IL-6, and tumor necrosis factor alpha (12, 13). Although the role of minor fimbriae in P. gingivalis virulence is less well understood, Hiramine et al. (6) have shown that the 67-kDa fimbrial protein induces IL-1α, IL-1β, and tumor necrosis factor alpha expression in mouse peritoneal macrophages, suggesting their possible involvement in the inflammatory response during the development of periodontal disease. Recent research has shown that minor fimbriae are necessary for the development of P. gingivalis biofilms on streptococcal substrates (7). Coadhesion of P. gingivalis-Streptococcus gordonii requires specific recognition between the 67-kDa and SspB proteins.

In this study, we demonstrate that minor fimbriae are required for P. gingivalis autoaggregation by examining a group of P. gingivalis fimbrial mutants. Our results show that only strains possessing the two distinct fimbriae are able to develop into mature monospecies biofilms. We provide evidence that while major fimbriae are responsible for P. gingivalis attachment and initiation of colonization, minor fimbriae are involved in the formation of microcolonies and maturation of P. gingivalis biofilms.

Minor fimbriae (Mfa1) are essential for P. gingivalis cell-cell aggregation.

To test the hypothesis that minor fimbriae function distinctively from major fimbriae, a group of fimbrial mutants, including a ΔfimA mutant, a Δmfa1 mutant, and a ΔfimA Δmfa1 double mutant, were constructed and examined for the ability to autoaggregate in Trypticase soy broth (TSB) supplemented with yeast extract (1 mg/ml), hemin (5 μg/ml), and menadione (1 μg/ml). P. gingivalis ATCC 33277 was used as the parental strain for mutant construction. To construct the ΔfimA mutant, a 2.5-kb NheI DNA fragment containing the tetracycline gene tetA(Q) (9) was inserted into the fimA gene cloned into plasmid pFIM. For construction of the Δmfa1 mutant, a 2.1-kb ermF-ermAM cassette (3) was inserted into the mfa1 gene cloned into plasmid pMFA. The resulting plasmids were linearized with XhoI and introduced into P. gingivalis 33277 by electroporation. Electroporation was carried out by a modification of the procedure of Fletcher et al. (3). P. gingivalis 33277 competent cells were obtained by suspending early-log-phase cells in electroporation buffer (10% glycerol, 1.0 mM MgCl₂). The cells were incubated with linearized plasmid pFIM or pMFA and pulsed with a Bio-Rad (Hercules, CA) Gene Pulser at 2.5 kV. The cells were then immediately added to the TSB and incubated anaerobically for 16 h. The ΔfimA mutant (FAT) was selected on TSB agar plates containing tetracycline (0.5 μg/ml), and the Δmfa1 mutant (MFAE) was selected on TSB agar plates containing erythromycin (5 μg/ml). To construct a ΔfimA Δmfa1 double mutant, competent cells of the Δmfa1 mutant, together with linearized plasmid pFIM, were pulsed with a Bio-Rad (Hercules, CA) Gene Pulser. The ΔfimA Δmfa1 double mutant (DFAET) was selected on TSB agar plates containing erythromycin (5 μg/ml) and tetracycline (0.5 μg/ml). All mutants were confirmed by PCR. The growth rates of these mutants were compared with that of wild-type strain 33277. Dilutions (1:100) of overnight bacterial cultures were inoculated into TSB supplemented with 100 μg/ml gentamicin, and the bacteria were grown anaerobically at 37°C. At various times, 1-ml aliquots were taken and the optical density at 600 nm was determined. Wild-type strain 33277 and the fimbrial mutants did not show any significant growth curve differences over a period of 72 h (Fig. 1), suggesting that the growth rate was not affected by the fimbrial mutations.

FIG. 1.

Comparison of the growth curves of P. gingivalis strains. Cells were grown in TSB medium. Shown in the curves are means of four samples, with error bars representing the standard error of the mean. One-milliliter aliquots were taken, and the OD₆₀₀ was measured over a period of 72 h.

To determine the production of major and minor fimbriae in these mutants, surface proteins were isolated by sonication and centrifugation from bacteria grown to mid-exponential phase. The soluble proteins were separated by 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and fimbrial production was detected by Western blotting with a polyclonal anti-FimA or anti-Mfa1 antibody. This analysis revealed that insertional mutations in the fimA and/or mfa1 genes completely abolished fimbrial production (Fig. 2A).

FIG. 2.

Aggregation of P. gingivalis strains in TSB. (A) Immunodetection of fimbrial production with rabbit polyclonal anti-FimA and anti-Mfa1 antibodies. Lane 1, wild-type strain 33277 (fimA⁺ mfa1⁺); lane 2, strain DFAET (ΔfimA Δmfa1); lane 3, strain SMFC (fimA⁺ mfa1⁺); lane 4, strain FAT (ΔfimA mfa1⁺); lane 5, strain MFAE (fimA⁺ Δmfa1). (B) P. gingivalis cells were grown in TSB for 24 h. Aggregated cells collected at the bottoms of the test tubes. (C) The number of P. gingivalis cells was estimated in a spectrophotometer as described by Soukos et al. (17). OD₆₀₀ was determined. Readings of supernatant were taken from 9 ml of top culture, and readings of aggregation were from 1 ml of bottom culture. An OD₆₀₀ of 0.1 equals approximately 10⁸ P. gingivalis cells per ml.

We observed that P. gingivalis cells display aggregative behavior, manifested by some cell gathering and sinking to the bottom of the test tubes, when the bacteria were grown in TSB and that the cell aggregation phenotypes of the wild-type strain and the derivative strains were correlated with expression and production of the Mfa1 protein (Fig. 2B). After growth for 24 h in TSB, P. gingivalis 33277 showed some degree of aggregation, and approximately 50% of the cells sank to the bottom of the test tube (Fig. 2C). P. gingivalis ΔfimA Δmfa1 double mutant DFAET, expressing neither fimA nor mfa1, did not show visible aggregation in TSB medium. The mfa1 mutant MFAE was unable to express mfa1 and, similar to P. gingivalis DFAET, lost its aggregation ability. The highest degree of aggregation was observed in the fimA mutant FAT, with 90% aggregation (P < 0.01). The evident autoaggregation of FAT may result from low expression of the major fimbriae, which exposes the shorter minor fimbriae. Autoaggregation was restored in a complemented Δmfa1 strain (SMFC) carrying plasmid pT-MFA containing a 2.5-kb fragment of the upstream and coding region of the mfa1 gene (15), suggesting that this phenotype did not result from polar effects. These data demonstrate that the degree of P. gingivalis aggregation correlated with the expression levels of mfa1 and fimA. Up-regulation of mfa1 and down-regulation of fimA may promote autoaggregation of P. gingivalis.

Role of major and minor fimbriae in P. gingivalis colonization.

The observation that minor fimbriae are associated with cell aggregation led us to hypothesize that minor fimbriae may also play an important role in biofilm formation by P. gingivalis. Current models of initial biofilm development in gram-negative organisms include two major steps. In the first step, bacteria attach to the surface in a monolayer. The cells then aggregate into microcolonies. In Pseudomonas aeruginosa, flagella are required for formation of the organism monolayer and type IV pili promote the cell-cell interaction that assembles a monolayer into microcolonies (14). We speculate that multiple extracellular proteins may also be involved in P. gingivalis biofilm formation. To test this possibility, biofilm formation experiments were performed with six-well polystyrene microtiter dishes (Corning, Inc., Corning, NY) containing TSB medium. The wells were precoated with human whole saliva. The plates were inoculated with individual P. gingivalis strains (10⁸ cells) and incubated at 37°C in an anaerobic chamber for 4 h. After the nonattached cells were removed, the wells were washed three times with phosphate-buffered saline (PBS). The attached bacteria were subsequently grown in TSB anaerobically for 72 h with a culture medium (TSB) change every 24 h. At various time intervals, bound bacteria were suspended in 1 ml of PBS and quantified by determination of the optical density at 600 nm (OD₆₀₀). As shown in Fig. 3, P. gingivalis ΔfimA mutant FAT and ΔfimA Δmfa1 double mutant DFAET displayed poor adherence to the surface at all time points, which is likely due to its lack of FimA production. Wild-type strain 33277 bound to the saliva-coated wells and formed microcolonies visible to the naked eye after 48 h. The number of attached cells peaked at 72 h. P. gingivalis Δmfa1 mutant MFAE was able to bind to the surfaces of the wells. There was no significant difference in the degree of bacterial attachment between wild-type strain 33277 and mfa1-deficient mutant MFAE during the first 4 h of incubation, suggesting that these two strains have similar adherence abilities. However, the number of attached P. gingivalis Δmfa1 mutant cells was only slightly increased after initial binding. The mfa1-deficient mutant also failed to form microcolonies visible to the naked eye after 48 h of growth. These findings support the aggregation results, suggesting that the minor fimbriae are involved in cell-cell aggregation, an essential step in microcolony formation.

FIG. 3.

Quantitation of biofilm formation by P. gingivalis on saliva-coated polystyrene wells. P. gingivalis biofilms were allowed to form in saliva-coated six-well polystyrene microtiter dishes. The cells bound to the well surfaces were suspended in 1 ml PBS for 4, 16, 24, 40, 48, 64, and 72 h of incubation. The ability of P. gingivalis strains to attach and form microcolonies on the surface was scored by measuring cell density. Biofilm formation by wild-type strain 33277, strain DFAET (ΔfimA Δmfa1), strain FAT (ΔfimA), and strain MFAE (Δmfa1) was measured by determining OD₆₀₀. Each data point represents the mean ± the standard deviation from at least three independent experiments.

To visualize P. gingivalis biofilms, we examined bacterial attachment and aggregation by using inverted confocal microscopy. P. gingivalis 33277, the ΔfimA mutant FAT, the Δmfa1 mutant MFAE, and the ΔfimA Δmfa1 double mutant DFAET were grown in TSB for 24 h, harvested by centrifugation, and resuspended in PBS. The cells were labeled with fluorescein 5-isothiocyanate (FITC; final concentration, 4 μg/ml; Sigma). P. gingivalis-FITC suspensions were incubated anaerobically at 4°C for 20 min in the dark with gentle shaking. The labeled cells were then washed twice with PBS, and 2 ml of cell suspension (10⁸ cells) was inoculated into MatTek glass bottom culture dishes (1 mm by 3.5 mm; MatTek Corp., Ashland, MA). Each dish was pretreated with human whole saliva. After anaerobic incubation for 1 or 24 h, the unbound cells were removed and the dishes were washed three times with PBS and refilled with 1 ml of sterile PBS. The bacteria attached to the glass bottom culture dishes were examined under a 510 inverted confocal laser scanning microscope (Carl Zeiss MicroImaging, GmbH, Germany). P. gingivalis biofilms were analyzed by using the LSM software, which calculates measurement by the x-y-z pixel dimensions. As showed in Fig. 4, 1 h after inoculation, wild-type strain 33277 formed monolayers punctuated by a few small microcolonies (<1 μm thick) on the surface of a saliva-coated glass bottom culture dish (Fig. 4A). A similar observation was made with the mfa1 mutant (Fig. 4B). The results are consistent with the biofilm assays, suggesting that mfa1 mutation does not affect the ability of P. gingivalis to adhere to saliva-coated surfaces. However, the binding ability of the fimA mutant FAT and the fimA mfa1 double mutant DFAET was abolished. Very few cells were detected on the saliva-coated surfaces (Fig. 4C and D), even after 24 h of exposure to saliva-coated surfaces (Fig. 4G and H). Wild-type strain 33277 continued to develop microcolonies, eventually covering 16% of the surface after 24 h of incubation (Fig. 4E). P. gingivalis 33277 microcolonies were about 800 μm² in size and 5 μm thick. The mfa1 mutant formed a progressively denser monolayer with small dispersed microcolonies after 24 h (Fig. 4F). Statistical analyses of microcolony values were performed with superANOVA (version 1.11; Abicus Software). The MFAE (Δmfa1) microcolonies on a saliva-coated surface were significantly smaller and thinner (25 μm² in area and 3.5 μm thick; P < 0.0001) and only covered 9% of the surface. These data further demonstrate that the minor fimbriae are required for cell-cell interactions and aggregation. The mfa1 mutant was capable of binding to the saliva-coated surface but could not efficiently recruit cells to form microcolonies. Our observations suggest that, in the early stage, the developmental pathway of P. gingivalis biofilms may also includes two steps, from a monolayer to microcolonies. The P. gingivalis major fimbriae are required for initial attachment. However, in the later stages of biofilm formation, the minor fimbriae appear to play an important role in microcolony formation by facilitating cell-cell interactions.

FIG. 4.

Biofilm formation by P. gingivalis on saliva-coated MatTek glass bottom culture dishes. P. gingivalis cells were stained with FITC and incubated in saliva-coated glass bottom culture dishes for 1 (A, B, C, and D) or 24 (E, F, G, and H) h. P. gingivalis biofilms were visualized with a confocal laser scanning microscope.

Conclusions and perspective.

P. gingivalis is considered a major periodontal pathogen and a secondary colonizer of dental plaque. The abilities of P. gingivalis to bind to oral surfaces are essential characteristics of bacterial pathogenicity. Specific binding allows P. gingivalis to remain at periodontal sites and avoid elimination by salivary flow through either expectoration or swallowing. In this study, we report that the minor fimbriae may be responsible for P. gingivalis autoaggregation and microcolony formation. The minor fimbriae could be one of the key elements that allow the organism to survive shearing forces and to establish P. gingivalis colonization. Inhibition of production of minor fimbriae in P. gingivalis may restrain accumulation of the organism, which is required for efficient infection.