Fusobacterium nucleatum enhances invasion of human gingival epithelial and aortic endothelial cells by Porphyromonas gingivalis

Authors


  • Editor: Peter Timms

Correspondence: Atsushi Saito, Department of Clinical Oral Health Science, Tokyo Dental College, 2-9-18 Misaki-cho, Chiyoda-ku, Tokyo 101-0061, Japan. Tel.: +81 3 5275 1721; fax: +81 3 3262 3420; e-mail: atsaito@tdc.ac.jp

Abstract

Invasion by Porphyromonas gingivalis has been proposed as a possible mechanism of pathogenesis in periodontal and cardiovascular diseases. Porphyromonas gingivalis have direct access to the systemic circulation and endothelium in periodontitis patients by transient bacteremia. Periodontitis can be described as one of the predominant polymicrobial infections of humans. In the present study, P. gingivalis strains were tested for their ability to invade a human gingival epithelial cell line (Ca9-22) and human aortic endothelial cells in coinfection with Fusobacterium nucleatum using antibiotic protection assays. Coinfection with F. nucleatum resulted in 2–20-fold increase in the invasion of host cells by P. gingivalis strains. The invasive abilities of P. gingivalis strains were significantly greater when incubated with a F. nucleatum clinical isolate (which possesses strong biofilm-forming ability), than when incubated with a F. nucleatum-type strain. In inhibition assays with metabolic inhibitors, a difference in inhibition profiles was observed between mono- and polymicrobial infections. Collectively, our results suggest that F. nucleatum facilitates invasion of host cells by P. gingivalis. Investigations of polymicrobial infection of host cells should improve our understanding of the role of P. gingivalis in periodontal infection and proatherogenic mechanisms.

Introduction

Epidemiological studies have demonstrated a positive association between periodontitis and cardiovascular diseases (Beck et al., 1996). It has been suggested that one of the linkages between these diseases is directly infectious, i.e. bacteria within the atheroma may be involved in the development of the atherosclerotic plaque (Chiu, 1999; Lalla et al., 2003; Gibson et al., 2004). Periodontal pathogens including Porphyromonas gingivalis have been detected in atherosclerotic plaques in humans using PCR techniques (Haraszthy et al., 2000; Ishihara et al., 2004; Kozarov et al., 2006).

Porphyromonas gingivalis elicits a proatherogenic response in endothelial cells in the form of increased leukocyte adhesion with concomitant upregulation of adhesion molecules, heightened production of proinflammatory cytokines and chemokines, as well as an induction of prothrombotic properties (Kang & Kuramitsu, 2002; Roth et al., 2007b). Interestingly, these effects on endothelial cells cannot be attributed solely to the effect of stimulation by bacterial cell-surface components, but may require the invasion of host cells by viable bacteria (Darveau et al., 2002; Roth et al., 2007a).

Invasion by P. gingivalis has been proposed as a possible mechanism of pathogenesis in periodontal and cardiovascular diseases (Lamont et al., 1995; Deshpande et al., 1998). Porphyromonas gingivalis has direct access to the systemic circulation and the endothelium in periodontitis patients, as transient bacteremias are common (Kinane et al., 2005), and the ability of P. gingivalis, detected at the sites of atherosclerotic disease, to invade host cells has been demonstrated (Kozarov et al., 2005).

Periodontitis can be described as one of the predominant polymicrobial infections of humans (Brogden et al., 2005). Because periodontal diseases result from complex interactions of multiple microorganisms, it is essential to investigate interactions between different periodontal bacteria and host cells. Bacterial species in subgingival plaque have been shown to fall into five distinctive complexes of closely related species (Socransky et al., 1998). Bacteria in one of the complexes referred to as a ‘red complex’ are commonly associated with periodontal lesions. However, bacteria in this complex such as P. gingivalis are usually detected in the presence of bacteria from a closely related complex referred to as an ‘orange complex’, comprising for example Fusobacterium nucleatum (Socransky et al., 1998; Kesavalu et al., 2007). Antagonistic and synergistic physiologic mechanisms, as well as environmental selection, are thought to be involved in such relationships (Kesavalu et al., 2007). Fusobacterium nucleatum initially adheres to early colonizers, including gram-positive cocci, and enhances the adherence of other periodontopathic bacteria including P. gingivalis (Kolenbrander, 2000).

In polymicrobial infections by bacterial enteropathogens, it has been shown that the ability of Campylobacter jejuni to invade cultured epithelial cells is significantly enhanced by the presence of other enteropathogens as coinfectants (Bukholm & Kapperud, 1987). Whether a similar interaction occurs in periodontopathogens is unknown. Data on the potential of P. gingivalis invasion into host cells in polymicrobial infection are scarce; the present study therefore sought to investigate the capacity of P. gingivalis to invade human gingival epithelial and aortic endothelial cells in coinfection with F. nucleatum.

Materials and methods

Bacterial strains and growth conditions

Porphyromonas gingivalis ATCC 33277, P. gingivalis W83 (ATCC BAA-308), F. nucleatum TDC100 [a clinical isolate and working strain in our laboratory (Saito et al., 2008)] and F. nucleatum ATCC 25586 were routinely maintained on tryptic soy agar (Difco Laboratories, Detroit, MI) supplemented with 10% defibrinated horse blood, hemin (5 μg mL−1) and menadione (0.5 μg mL−1) at 37 °C under anaerobic conditions. Escherichia coli SCS110 and DH5α strains were used as a control in antibiotic protection assays.

Cells and culture conditions

The human gingival epithelial cell line, Ca9-22, was purchased from the Health Science Research Resources Bank (Osaka, Japan). Ca9-22 is an established transformed human gingival cell line that has been used in previous studies as a culture model of oral epithelial cells (Hirose et al., 1996; Ohshima et al., 2001; Takeuchi et al., 2008). The Ca9-22 cells were maintained in Eagle's minimal essential medium (MEM) supplemented with glutamine (0.6 g L−1), heat-inactivated 10% fetal calf serum and gentamicin (50 μg mL−1)/amphotericin B (50 ng mL−1) (Cascade Biologics, Portland, OR) at 37 °C in 5% CO2.

Human aorta endothelial cells (HAEC) were supplied by Kurabo Inc. (Osaka, Japan) and maintained in HuMedia-EG2 (Kurabo) in an atmosphere of 5% CO2 and 95% air at 37 °C. Cells from passages 4 through 9 were tested for viability and morphology before to seeding in appropriate tissue culture plates and allowed to reach near confluency before assay.

Invasion procedures

Invasion of bacteria was quantitated by the standard antibiotic protection assay (Lamont et al., 1995; Deshpande et al., 1998). Briefly, epithelial cells were seeded in a 12-well flat-bottom culture plate (Iwaki, Chiba, Japan) at a cell density of 2.0 × 105 cells per well. Before infection, the cells were washed twice with phosphate-buffered saline (PBS, pH 7.4) and incubated further for 2 h in MEM without antibiotics. The multiplicity of infection (MOI) was calculated based on the number of cells per well at confluence. Porphyromonas gingivalis and F. nucleatum strains were inoculated into brain heart infusion broth (Becton Dickinson, Sparks, MD) supplemented with 0.5% of yeast extract, hemin (5 μg mL−1) and menadione (0.5 μg mL−1), and grown for 2 days until the OD660 nm reached 1.0. After washing with PBS, bacterial cells were resuspended in MEM. Bacterial suspensions (2.0 × 107 cells per well) were added to confluent Ca9-22 monolayers (MOI=100) and incubated at 37 °C in 5% CO2 for 2 h. After incubation, unattached bacteria were removed following washing of the monolayers three times with PBS. External adherent cells were then killed by incubating the infected monolayers with MEM containing 200 μg mL−1 of metronizazole and 300 μg mL−1 of gentamicin for 1 h. This concentration of antibiotic was sufficient to completely kill 108 bacteria mL−1 in 1 h. Controls for antibiotic killing of bacteria were included in all experiments. After exposure to antibiotic, monolayers were washed twice with PBS, and lysed in 1 mL of sterile distilled water per well. Cells were incubated aerobically for 30 min, during which they were disrupted by repeated pipetting. Lysates were diluted and plated on blood agar plates, and incubated anaerobically at 37 °C for 10 days. CFU of invasive organisms were then enumerated. Invasion efficiency was expressed as the percentage of the initial inoculum recovered after antibiotic treatment and Ca9-22 lysis.

The invasion assay with HAEC was performed using the same procedure as above with EG2 medium.

Polymicrobial infection

Following demonstration of monomicrobial infections with P. gingivalis strains, we performed experiments to develop a model of polymicrobial periodontal infection using P. gingivalis and F. nucleatum as members of a prototype consortium and examined the invasion characteristics and interactions of these organisms. For polymicrobial infection, P. gingivalis (1 × 107 cells mL−1) was gently mixed with an equal volume of F. nucleatum (1 × 107 cells mL−1) and the organisms were allowed to interact for 5 min. For the monomicrobial control infection, P. gingivalis was mixed with an equal volume of MEM or EG2 medium. As a control for polymicrobial infection, E. coli SCS110 or DH5α was preincubated with P. gingivalis. The poly- and monomicrobial inocula were added to Ca9-22 or HAEC monolayers.

The bacterial culture growth phase, viability, counts, interaction times, suspension medium, infection dose and infection procedures were all standardized; i.e. the same preparation and infection protocols were used for all invasion assays throughout the study.

Inhibition of bacterial invasion

For inhibition assays with antiserum, P. gingivalis or F. nucleatum cells were preincubated with the indicated dilution of rabbit polyclonal anti-P. gingivalis serum for 30 min at room temperature (RT) before use in assays.

Porphyromonas gingivalis or F. nucleatum cells were also preincubated with indicated concentrations of d-galactose (inhibitor of F. nucleatum adhesion/invasion) for 15 min at RT before use in assays.

To dissect the biochemical pathways involved, the effect of a group of metabolic inhibitors on invasion was investigated. The following inhibitors, in the solvent and at the final concentration indicated, were used. Cytochalasin D, 1 μg mL−1 in dimethyl sulfoxide (DMSO); nocodazole, 10 μg mL−1 in DMSO; staurosporine, 0.5 μM in DMSO; and cycloheximide, 100 μg mL−1 in ethanol. The inhibitors were preincubated with the host cells for 60 min before addition of the bacteria and remained present throughout the invasion assay. All potential inhibitors were tested at the concentration used for possible adverse effects on the host cells, through comparison with cells without inhibitors, by examining the morphology of the cells and the confluency of the monolayer.

Data and statistical analysis

All experiments were performed in duplicate or triplicate for each condition and repeated at least three times. Statistical comparisons were performed using a software package (instat 3.0, GraphPad Software Inc., La Jolla, CA). The Mann–Whitney U-test or anova with the Bonferroni post-test (for multiple comparisons) was used.

Results

Monomicrobial infection

There was strain-to-strain variability in the ability of P. gingivalis to invade HAEC and Ca9-22 (Fig. 1). For both HAEC and Ca9-22, P. gingivalis strain 33277 had greater invasive abilities than strain W83. For HAEC, mean invasion efficiencies for P. gingivalis 33277 and W83 were 1.58% and 0.52%, respectively (Fig. 1a). The ability of P. gingivalis 33277 to invade Ca9-22 was essentially equal to its ability to invade HAEC (Fig. 1b); however, P. gingivalis W83 invaded HAEC approximately threefold over Ca9-22.

Figure 1.

 Invasion of (a) HAEC (b) Ca9-22 by Porphyromonas gingivalis 33277 or P. gingivalis W83 in mono- or polymicrobial infection with Fusobacterium nucleatum strains (MOI=100, infected for 2 h). Invasion efficiency (%) was expressed as percentage of the inoculum (P. gingivalis) protected from the antibiotic killing for 1 h. Values are the means±SDs of triplicate independent determinations from a typical experiment. *Statistically significantly different (P<0.01, anova with Bonferroni multiple comparisons test).

Polymicrobial infection with P. gingivalis and F. nucleatum

In polymicrobial infection experiments with P. gingivalis and F. nucleatum, viable counts of P. gingivalis strains recovered from HAEC or Ca9-22 cells were used to determine bacterial invasion. Coincubation of P. gingivalis 33277 with F. nucleatum significantly boosted P. gingivalis invasion of host cells, resulting in a 2–20-fold increase in invasion efficiencies (Fig. 1). Coincubation with E. coli SCS110 or DH5α exerted no effect on P. gingivalis invasion of HAEC or Ca9-22.

Invasion of host cells by P. gingivalis 33277 was statistically significantly higher in the presence of F. nucleatum TDC100 (invasion efficiency: 12%) than F. nucleatum 25586 (6%) (P<0.01) (Fig. 1a). In the presence of F. nucleatum, P. gingivalis W83, which showed a low invasive ability in monomicrobial infection, demonstrated a dramatic increase in invasion. Invasion of HAEC with P. gingivalis W83 was significantly more enhanced in the presence of F. nucleatum TDC100 (invasion efficiency: 8%) than F. nucleatum 25586 (2%) (P<0.01). Similarly, F. nucleatum significantly enhanced invasion of P. gingivalis strains to Ca9-22 cells (Fig. 1b).

Inhibition of bacterial invasion

Porphyromonas gingivalis invasion of HAEC was inhibited by anti-P. gingivalis serum (diluted 1 : 100) by c. 70% (Table 1). The inhibition was significant, but relatively low when coincubated with F. nucleatum. A similar trend was observed with Ca9-22 cells (data not shown).

Table 1.   Effect of anti-Porphyromonas gingivalis serum on P. gingivalis invasion of HAEC by mono- or polymicrobial infection
PreincubationInvasion of HAEC (% of control)
Mono-infection
(Pg)
Poly-infection
(Pg+Fn)§
  • *

    Statistically significantly different from control (P<0.01) by anova with Bonferroni multiple comparisons test.

  • Invasion of Porphyromonas gingivalis 33277 relative to the level obtained in the absence of antiserum (preimmune serum control). Values given as means ± SDs of triplicate independent determinations from a typical experiment.

  • Cells were infected by P. gingivalis 33277.

  • §

    Cells were infected by P. gingivalis 33277 and Fusobacterium nucleatum TDC100.

Control (preimmune serum)100 ± 12100 ± 21
Anti-Pg serum (dilution 1 : 1000)79 ± 12*89 ± 18
Anti-Pg serum (dilution 1 : 100)29 ± 20*58 ± 11*

To determine whether the enhanced invasion of P. gingivalis in polymicrobial infection involves a lectin-like adhesin(s), a sugar inhibition assay was performed. Incubation with d-galactose resulted in decreased invasion by P. gingivalis in polymicrobial infection experiments (Fig. 2).

Figure 2.

 Inhibition of fusobacterial enhanced Porphyromonas gingivalis invasion by d-galactose. The effects of increasing concentrations of d-galactose on P. gingivalis invasion of HAEC were assessed for monomicrobial (P. gingivalis only; ○), or polymicrobial (P. gingivalis and Fusobacterium nucleatum together; ▪) infections. Values are the means±SDs of triplicate independent determinations from a typical experiment. *Statistically significantly different from control (P<0.01, Mann–Whitney U-test).

Various metabolic inhibitors reported previously to reduce P. gingivalis or F. nucleatum invasion were assessed for the ability to inhibit Fusobacterium-enhanced P. gingivalis invasion. In mono- and polymicrobial infection experiments, invasion of the host cells by P. gingivalis required multiple components of the host including actin, microtubule and protein kinases (Table 2). One notable difference in the inhibition profiles was observed between mono- and polymicrobial infections. Cycloheximide (which targets host cell protein synthesis) significantly reduced invasion by P. gingivalis in polymicrobial infection experiments. This inhibitor has previously been shown to inhibit F. nucleatum invasion but not P. gingivalis invasion.

Table 2.   Effects of metabolic inhibitors on Porphyromonas gingivalis invasion of HAEC by mono- or polymicrobial infection
InhibitorTargetInvasion of HAEC (% of untreated
control)
Monoinfection
(Pg)§
Polyinfection
(Pg+Fn)
  • *

    Statistically significantly different from control (P<0.01) by anova with Bonferroni multiple comparisons test.

  • Statistically significantly different from monomicrobial infection (P<0.01) by Mann–Whitney U-test.

  • Invasion of Porphyromonas gingivalis 33277 relative to the level obtained in the absence of inhibitor (medium control). Values given as means ± SDs of triplicate independent determinations from a typical experiment.

  • §

    Cells were infected by P. gingivalis 33277.

  • Cells were infected by P. gingivalis 33277 and Fusobacterium nucleatum TDC100.

Cytochalasin DActin11.5 ± 3.5*23.0 ± 11.1*
NocodazoleMicrotubule14.8 ± 5.0*18.2 ± 16.4*
SaturosporineProtein kinase54.4 ± 5.5*22.7 ± 17.9*,†
CycloheximideProtein synthesis91.7 ± 5.525.7 ± 8.3*,†

Discussion

In the present study, we first explored the abilities of different P. gingivalis strains to invade gingival epithelial and endothelial cells. Porphyromonas gingivalis W83 has been shown to be highly virulent in experimental animal models (Neiders et al., 1989; Genco et al., 1991). In our experimental setup, the invasive ability of P. gingivalis W83 into host cells was relatively low when compared with P. gingivalis 33277, which has been shown to be highly fimbriated but less virulent. Furthermore, P. gingivalis W83 displayed an invasive ability that differed in the cell types tested. Fimbriae are considered important in adherence and invasion by P. gingivalis. However, it has been shown that the presence and expression of fimA is not sufficient for P. gingivalis invasion of endothelial and epithelial cells (Dorn et al., 2000; Umeda et al., 2006). The differential invasion efficiency observed for different cell types is likely due to different interactions between P. gingivalis and the types of cell surface receptors present on the different cell types that are involved in the invasion process.

There is increasing evidence in the literature for the importance of polymicrobial infections in which selected microorganisms interact in a synergistic or an antagonistic fashion, impacting on the pathogenesis of periodontal disease (Chen et al., 1996; Feuille et al., 1996; Kesavalu et al., 2007). Synergistic interactions in virulence between F. nucleatum and P. gingivalis have been observed in vitro and in animal models (Feuille et al., 1996; Ebersole et al., 1997). In this study, we demonstrated that F. nucleatum enhances the ability of P. gingivalis to invade host cells. To the best of our knowledge, this study is the first to demonstrate such an interaction between F. nucleatum and P. gingivalis.

In polymicrobial infection experiments, F. nucleatum TDC 100 enhanced P. gingivalis invasion of host cells significantly more than the F. nucleatum-type strain. A previous study from our group has demonstrated that F. nucleatum TDC 100 has a synergistic relationship with P. gingivalis and a strong biofilm forming ability (Saito et al., 2008). Han et al. (2000) reported that a spontaneous lam mutant F. nucleatum, defective in aggregation with human lymphocytes and coaggregation with P. gingivalis, was defective in attachment to and invasion of human gingival epithelial cells, suggesting that the same bacterial determinants are involved in aggregation properties and the ability to invade host cells. Fusobacterium nucleatum and P. gingivalis are strong coaggregating pairs, and the coaggregation may have the capacity to alter the expression of virulence factors in individual microorganisms (Feuille et al., 1996). Coaggregation between P. gingivalis and F. nucleatum is mediated by a galactoside moiety on the P. gingivalis surface and a lectin on the F. nucleatum, and inhibited by lactose, galactose and related monosaccharides (Kolenbrander & Andersen, 1989; Kinder & Holt, 1993). We have observed a strong coaggregation reaction between P. gingivalis 33277 and F. nucleatum TDC 100 that is inhibitable by galactose (data not shown). The synergistic interactions between F. nucleatum and P. gingivalis observed in the present study could be partly explained by the coaggregating effect between these organisms, as galactose also inhibited F. nucleatum-enhanced P. gingivalis invasion.

In the present study, coinfection with F. nucleatum strains markedly enhanced invasion of host cells by P. gingivalis W83, a strain we have shown to be minimally invasive in monomicrobial infection of host cells. Rudney et al. (2005) showed that intracellular infections of buccal epithelial cells with periodontal pathogens were uniformly polymicrobial, and proposed several scenarios regarding invasion of host cells by a consortium of oral bacteria. Invasiveness might be limited to a subset of oral species that use it as a virulence factor. Alternatively, a wide range of oral bacteria that principally live in a biofilm might be capable of invasion as a means of persisting. Because species interaction appears to be widespread in oral biofilm (Cook et al., 1998; Palmer et al., 2001; McNab et al., 2003), another alternative could be that noninvasive species gain entrance to cells by forming consortia with invasive species. It has been reported that F. nucleatum transports noninvasive Streptococcus cristatus into human epithelial cells (Edwards et al., 2006). In the present study, polymicrobial infection of the host cells by P. gingivalis and F. nucleatum not only facilitated P. gingivalis invasion but also resulted in the invasion by F. nucleatum, although the extent of F. nucleatum invasion was relatively low, when compared with that of P. gingivalis 33277 (data not shown). Although anti-P. gingivalis serum abrogated P. gingivalis invasion of host cells in a monomicrobial infection setting, the extent of inhibition was less in polymicrobial infection. These results suggested that mechanisms other than the adherence signal induced by P. gingivalis are likely to be involved, and that interaction(s) between F. nucleatum and host cells may play a significant role in Fusobacterium-enhanced P. gingivalis invasion.

Also, in the inhibition experiment, cycloheximide reduced invasion significantly by P. gingivalis in polymicrobial infection. As this inhibitor has been shown previously to inhibit F. nucleatum invasion (Han et al., 2000) but not P. gingivalis invasion (Deshpande et al., 1998), it is conceivable that infection by F. nucleatum may pave the way for increased invasion of P. gingivalis.

We cannot yet clarify whether the effect exerted by the coinfectant is directed at the host cell or the P. gingivalis. Invasive bacteria generally gain entry by co-opting and redirecting host cell mechanisms such as endocytosis (Lamont et al., 1995; Sandros et al., 1996; Meyer et al., 1999; Progulske-Fox et al., 1999; Han et al., 2000). Immunomodulating roles of F. nucleatum have been suggested by previous studies (Feuille et al., 1996; Choi et al., 2001). Polymicrobial infections may actually modulate the adaptive host responses, leading to more effective evasion of protective immune responses.

In summary, this report demonstrates that F. nucleatum facilitates P. gingivalis invasion of human gingival epithelial and endothelial cells. The significance of this increased ability to invade cells in progression of periodontal as well as cardiovascular diseases needs to be elucidated in future studies. We are currently investigating the molecular mechanisms involved in this relationship, and whether other periodontopathogens in a consortium are able to induce the synergistic effects.

Acknowledgement

The authors thank David Blette for editing the manuscript. This research was supported by Oral Health Science Center Grant HRC7 from Tokyo Dental College and by a ‘High-Tech Research Center’ Project for Private Universities: matching fund subsidy from MEXT (Ministry of Education, Culture, Sports, Science and Technology) of Japan, 2006–2010.

Ancillary