INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Fusobacterium nucleatum, a gram-negative anaerobic organism, has been implicated in the pathogenesis of pulpal infection, alveolar bone abscesses, and periodontal disease (7). The pathogenic properties of this organism have also been described in urinary tract infection (25), bacteremia (14), pericarditis (22), peritonsillar abscesses (20), and septic arthritis (16). The human mouth contains one of the most complex bacterial floras. Neither the mechanisms of interaction among these bacteria nor their roles in the induction of pathologies in the host are well understood. The interaction between bacterial species and the host defense mechanisms is considered to be the key element in determining the status of health and disease in the mouth. Increased colonization by pathogenic bacteria and subsequent modulation of host defense mechanisms in the oral cavity have been proposed to result in the initiation and progression of periodontal disease (7, 27). The immunosuppressive nature of certain invasive pathogenic oral bacteria, e.g., F. nucleatum, has been reported previously (7, 26, 27). Inhibition of both B- and T-cell functions have also been reported in the presence of F. nucleatum (11, 21, 27). However, the detailed mechanisms of F. nucleatum-mediated immunosuppression have yet to be established. In this paper we show that F. nucleatum activates cell death machinery in the peripheral blood mononuclear cells (PBMCs) and polymorphonuclear cells (PMNs). Activation of death in immune cells by the bacteria may represent one mechanism by which F. nucleatum mediates immunosuppression and inactivation of immune cells. Furthermore, the putative bacterial apoptosis-inducing agent(s) is likely a heat-labile protein on the cell surface of F. nucleatum. Moreover, we demonstrate that both the NF-B and interleukin-converting enzyme (ICE) pathways are involved in F. nucleatum-mediated lymphocyte death.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Cell lines, bacterial strains, and reagents. Jurkat and YT cells and their transfectants were maintained in RPMI 1640 supplemented with 1% sodium pyruvate, 1% nonessential amino acids, 1% penicillin-streptomycin (purchased from Life Technologies, Grand Island, N.Y.), and 10% fetal calf serum (Irvine Scientific, Santa Ana, Calif.). Porphymonas gingivalis (ATCC 33277), Actinobacillus actinomycetemcomitans (ATCC 33384), Treponema denticola (ATCC 33521), and Prevotella intermedia (ATCC 49046) were obtained from the American Type Culture Collection. F. nucleatum (PK1594) was obtained from Paul Kolenbrander at the National Institutes of Health. P. gingivalis, A. actinomycetemcomitans, P. intermedia, and F. nucleatum were grown in brain heart infusion medium (Difco, Detroit, Mich.). T. denticola was grown in TYGVS medium (3). All anaerobic bacteria were grown in an atmosphere of 80% N₂, 10% CO₂, and 10% H₂ at 37°C.

Bacterial treatments. Oral bacteria were treated with and without 1% paraformaldehyde for 1 h at room temperature. The bacteria were then washed three times with 1× phosphate-buffered saline (PBS) and used in the experiments. The heat treatment of the oral bacteria was conducted by boiling the strains at 100°C for 10 min before they were added to the cell cultures. Finally, the oral bacteria were incubated with 10 mg of pronase/ml for 12 to 18 h prior to their addition to Jurkat cells. Coincubation of bacterial strains with either immune cells derived from the peripheral blood or lymphocytic cell lines was carried out in the presence of RPMI 1640 containing 10% fetal calf serum.

Preparation of Jurkat and YT stable transfectants. The pRcCMV-IB_(32A,36A) construct was described previously (12). The mutant IB contains substitutions of alanine for serines 32 and 36. Twenty micrograms of DNA was electroporated into the 10⁷ Jurkat and YT cells, and the stable transfectants were selected by growing the cells in G₄₁₈ selection medium. Jurkat and YT cell lines transfected with pRc/CMV vector alone were used as control samples.

Isolation of PBMCs and PMNs. Venous blood was obtained from healthy individuals by standard forearm venipuncture, following guidelines of the University of California at Los Angeles human subject protection committee. PBMCs were obtained after Ficoll-Hypaque centrifugation as described by Jewett et al. (18, 19). The PBMCs were washed twice and incubated with the bacteria as described below. After the removal of PBMCs, the layer immediately above the red blood cells that was rich in PMNs was collected and subjected to amonium chloride lysis. The lysis of red blood cells was repeated twice, and the PMNs were then layered on fetal bovine serum to remove the residual red blood cell debris. PMNs at a concentration of 2 × 10⁶ per ml were used for further treatment with F. nucleatum.

ELISA. Wells of ELISA plates were coated with 50 µl of a mixture of B154.9.1 and B154.7.1 monoclonal antibodies, each specific for a different epitope on the TNF molecule. The plates coated with monoclonal antibodies were kept for at least 1 day before use, washed three to four times, and blocked with ELISA PBS containing 1% bovine serum albumin for 30 min. Then the plates were washed twice, and 50 µl of supernatants from treated NK samples was added to each well. After overnight incubation at 37°C, the plates were washed four times, 50 µl of polyclonal anti-TNF- antibody at 1:1,000 dilution was added, and the incubation was continued for 2 h at 37°C. Alkaline phosphatase-conjugated goat anti-rabbit immunoglobulin G (Caltag) at a dilution of 1:2,000 was added to the plates, and the plates were incubated for an additional 2 h at 37°C. Finally, the plates were washed and incubated with the alkaline phosphatase substrate (Sigma 104) and read after 2 h in a titrated Multiscan MCC/240 ELISA reader using the 405-nm-pore-size filter. Monoclonal antibodies for IFN- ELISA were purchased from Genzyme, and polyclonal rabbit antibodies specific for each cytokine were generated in our laboratory.

DNA staining and apoptosis. Staining was performed by labeling the cells with propidium iodide as described previously (18, 19). Briefly, samples of 2 × 10⁵ cells were washed twice with PBS and incubated in 70% ethanol on ice. After 30 min of incubation, the cells were washed twice with PBS and 70 µl of RNase (1 mg/ml; Sigma) and 140 µl of propidium iodide (100 µg/ml; Sigma) were added to each sample. After 1 h of incubation in the dark, DNA analysis was performed using a flow cytometer (Coulter Elite).

DNA gel electrophoresis. Jurkat cells were maintained in RPMI 1640 containing 10% fetal bovine serum and 1% penicillin-streptomycin at a density between 5 × 10⁵ and 1 × 10⁶ per ml; 5 × 10⁶ cells were used for each sample.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Effect of oral bacteria on human PBMCs. To investigate the possible modes of interaction between oral microorganisms and the host's immune cells, we studied the human PBMCs after they were treated with various strains of oral bacteria. Five oral bacteria associated with periodontal disease, P. gingivalis, A. actinomycetemcomitans, T. denticola, F. nucleatum, and P. intermedia, were used in this study. Treatment of PBMCs with oral bacteria triggered significant levels of TNF- and IFN- release in the supernatants. Similar levels of TNF- release were induced by all five oral bacteria tested (Table 1). A. actinomycetemcomitans induced the highest levels of IFN- secretion (Table 1). In addition, F. nucleatum and A. actinomycetemcomitans but not the other three bacterial strains induced significant levels of apoptotic cell death as determined by flow cytometric analysis of propidium iodide-stained cells (Table 1). F. nucleatum was found to induce the highest levels of apoptotic cell death of PBMCs, whereas A. actinomycetemcomitans induced moderate levels (Table 1). Similar results were obtained with several other strains of F. nucleatum (e.g., ATCC 10953 and ATCC 25586) and A. actinomycetemcomitans (SUNY75 and SUNY465) (data not shown). Escherichia coli cells (HB101) were tested and found to have no effect on inducing apoptotic cell death of PBMCs (Table 2). There was no correlation between the induction of TNF- and IFN- cytokine release and the levels of apoptotic cell death in PBMCs (Table 1). Thus, we further analyzed the mechanism of apoptosis using F. nucleatum as a model bacterium.

View this table: [in this window] [in a new window]   TABLE 1.   Induction of TNF-, IFN-, and apoptosis by oral bacteriaa

Characterization of apoptotic cell death induced by F. nucleatum. Addition of F. nucleatum to PBMCs induced significant levels of apoptotic cell death in a great majority of the cells, as determined by flow cytometric analysis of propidium iodide-stained cells (Table 2). Induction of apoptotic cell death in PBMCs was dose dependent, as shown in Fig. 1. Interestingly only the levels of cell death but not the levels of cells in either the S or G₂/M phase of the cell cycle were elevated (data not shown). Increase in the apoptotic cell death in PBMCs was paralleled by a significant decrease in cells in the G₀/G₁ phase of the cell cycle (Fig. 1).

View larger version (28K): [in this window] [in a new window]   FIG. 1.   Dose-dependent induction of apoptotic cell death of PBMCs by F. nucleatum. PBMCs were cocultured in the presence of F. nucleatum at the indicated ratios. The levels of apoptotic cell death were determined by using flow cytometric analysis of propidium iodide-stained cells.

View larger version (84K): [in this window] [in a new window]   FIG. 2.   Inhibition of F. nucleatum-mediated apoptotic cell death of PBMCs by the ICE inhibitor. Jurkat cells were cocultured in the presence of F. nucleatum and P. intermedia (30:1 bacterium-cell ratio) overnight. Equal amounts of DNA (2 µg) extracted from equal numbers of cells for each sample were loaded onto a 2% agarose gel and run on a gel electrophoresis assay. Lanes: 1, molecular weight marker; 2, untreated Jurkat cells; 3, Jurkat cells treated with viable P. intermedia; 4, Jurkat cells treated with viable F. nucleatum; 5, Jurkat cells treated with 1% formaldehyde-treated F. nucleatum; 6, Jurkat cells treated with viable F. nucleatum and 500 µM ICE inhibitor YVAD; 7, Jurkat cells treated with 1% formaldehyde-treated F. nucleatum and YVAD; 8, Jurkat cells treated with anti-FAS antibody; 9, Jurkat cells treated with Anti-FAS antibody and YVAD.

Putative bacterial apoptosis-inducing molecule(s) is likely a heat-labile protein on surfaces of cells. Similar to the effect of viable bacteria, paraformaldehyde-treated F. nucleatum cells were capable of mediating the apoptotic cell death of PBMCs, suggesting that a bacterial surface component(s) is responsible for inducing apoptosis in PBMCs (Fig. 2). Interestingly, heat-killed F. nucleatum was no longer able to induce apoptosis in PBMCs, indicating that the putative bacterial apoptosis-inducing molecule(s) is heat labile (Table 3). To further investigate the nature of the putative bacterial apoptosis-inducing molecule(s), we treated F. nucleatum cells with a protease (EC 3.4.24.31) and found that the resulting bacteria were no longer able to induce apoptosis of PBMCs (Fig. 3). The protease-treated bacteria were viable, since they regained the ability to induce apoptosis several hours after the removal of the protease (data not shown). Therefore, it is likely that the putative bacterial apoptosis-inducing molecule(s) is a heat-labile protein on the surface of the bacterium.

View larger version (42K): [in this window] [in a new window]   FIG. 3.   Inhibition of F. nucleatum-induced apoptotic cell death of PBMCs by pronase. F. nucleatum was cultured for 18 h in the presence or absence of pronase (10 mg/ml) (protease type XIV; EC 3.4.24.31) prior to its addition to Jurkat cells. The Jurkat cells were then cocultured for 18 h with either the pronase-treated F. nucleatum (+ Pronase) or control untreated F. nucleatum ( Pronase). Jurkat cell apoptosis was determined by flow cytometric analysis of propidium iodide-stained cells.

NF-B and ICE pathways are involved in F. nucleatum-mediated apoptotic cell death. F. nucleatum induced the apoptotic cell death of Jurkat T cell and YT NK cell lines (Table 4) but had minimal or no effect on Cal27 and SCC4 squamous cell carcinoma lines (data not shown). IB mutant transfected T and NK cell lines exhibited significantly higher levels of apoptotic cell death compared to cell lines transfected by vector alone in the presence of F. nucleatum (Table 4).

View this table: [in this window] [in a new window]   TABLE 4.   NF-B is involved in bacterium-induced apoptosisa

F. nucleatum-mediated cell death of peripheral blood PMNs as well as mononuclear cells. Incubation of peripheral blood PMNs with F. nucleatum induced significantly higher levels of cell death in PMNs than in PBMCs (Fig. 4). At a concentration of 30:1 (bacterium-to-cell ratio) a complete elimination of PMNs was observed compared to 17 to 59% cell death levels obtained for PBMCs at a 50:1 bacterium-to-cell ratio (Fig. 4 and Table 2). A dose-dependent increase in the levels of cell death was observed in PMNs when they were cocultured in the presence of F. nucleatum (Fig. 4).

View larger version (26K): [in this window] [in a new window]   FIG. 4.   Induction of cell death in peripheral blood PMNs by F. nucleatum. PMNs at a concentration of 2 × 106 per ml were cocultured in the presence of F. nucleatum for 16 h. The numbers of viable cells were counted by trypan blue staining.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Apoptosis, programmed cell death, is a regulated physiological and pathological process involved in cell deletion during normal tissue homeostasis and embryological development, as well as during viral and bacterial infections (4, 15, 24, 31, 34). Apoptosis can also be induced by a variety of environmental factors, such as ionizing radiation, stress-related hormones (e.g., glucocorticoids), viral infection (e.g., human immunodeficiency virus), and oncogenes (4, 15, 24, 31, 34). In these cases, there are apparent benefits for the organisms in having the damaged or infected cells destroy themselves. By this logic, apoptosis may also occur in cells infected by pathogenic bacteria. In fact, there are a number of reports of the death of immune cells induced by bacterial pathogens (8, 9, 10, 23). In this paper, we have demonstrated for the first time that the apoptosis of PBMCs and PMNs was induced by an oral bacterial pathogen, F. nucleatum. Our studies indicated that the apoptotic cell death of PBMCs was induced by certain bacterial surface protein(s) and regulated through signaling proteins, such as ICE and NF-B in the target cells.

The oral immune network prevents, in general, the invasion of the host by the oral microflora. However, virtually all individuals exhibit some degree of periodontal disease, a localized destructive inflammatory response to the microflora. Periodontal disease varies widely in severity among individuals, and it is reasonable to propose that this variability is due to differences in microfloras as well as to differences in the immune responses against the oral microfloras. There is prior evidence that bacterial species can suppress immune responses (11, 22, 28). In particular, F. nucleatum has been shown to inhibit many immunological functions (7, 26, 27). Initial observations by Shenker and Dirienzo (27) indicated significant inhibition of peripheral blood lymphocyte function by cytoplasmic extracts obtained from F. nucleatum. Since then, the authors have purified and characterized FIP (F. nucleatum inhibitory protein) as the putative protein responsible for the immunosuppressive effect mediated by F. nucleatum. FIP was shown to elicit arrest at the G₀/G₁ phase of the cell cycle (11). Monocyte suppression of human polyclonal B lymphocyte activation was also observed in the presence of F. nucleatum (21). Shenker and Dirienzo (27) have hypothesized that immunosuppression caused by F. nucleatum is a relatively temporary phenomenon, since many patients eventually develop a detectable humoral and cellular response to periodontal pathogens.

Preceding the induction of apoptotic cell death by F. nucleatum, significant aggregation of PBMCs was observed within a few minutes of the addition of the oral bacterium (data not shown). Such aggregation was only observed in the presence of F. nucleatum and not the other oral bacterial species tested (data not shown). Only viable or formaldehyde-treated F. nucleatum cells were able to induce aggregation and the induction of death in PBMCs. The ability of F. nucleatum to induce aggregation and apoptotic cell death was lost when the bacterium was killed by heat treatment. A close relationship was observed between the ability of F. nucleatum to induce aggregation of the PBMCs and its ability to cause apoptotic cell death. Thus, it is possible that aggregation is a necessary step for the induction of death in PBMCs. Therefore, the bacteria might upregulate the Fas and TNF receptor-mediated death of PBMCs and the aggregation might serve to bring Fas and Fas ligand and TNF receptor and TNF into close proximity to each other for optimal signaling in PBMCs. F. nucleatum-mediated aggregation of peripheral blood lymphocytes has also been observed by Haake and Lindmann (17). The aggregation of PBMCs was inhibited by L-arginine, L-lysine, and heat treatment (17). Phytohemagglutinin-stimulated DNA synthesis and interleukin 2R expression of PBMCs were also inhibited in the presence of F. nucleatum (17).

Alternatively, F. nucleatum directly delivers death signals to PBMCs through the binding of a putative surface protein. It is likely that F. nucleatum delivers a direct death signal through its surface component as well as aiding in the upregulation of cell death machinery (Fas- and TNF receptor-mediated signaling) in PBMCs. These possibilities are under investigation in our laboratory.

One possible explanation for bacterium-induced apoptosis is that bacterial lipopolysaccharide mediates the induction of cell death by triggering TNF- release by PBMCs. However, our preliminary data are inconsistent with this hypothesis due to the following observations: (i) lipopolysaccharide is heat stable, while the putative bacterial apoptosis-inducing molecule(s) is heat labile; (ii) we tested a group of oral bacteria, and while all were able to induce the production of TNF- at similar levels, only A. actinomycetemcomitans and F. nucleatum were able to induce apoptosis of PBMCs; and (iii) we found that the rate of apoptosis of lymphocytes induced by the addition of exogenous TNF- was much less than that induced by the bacteria.

The immunosuppression by F. nucleatum observed by other investigators could be due to the ability of the bacterium to induce apoptotic cell death of the PBMCs. The implications of this observation for the generation and maintenance of periodontal diseases are speculative at present. By eliminating immune cells that are important for immune defense against oral bacteria, F. nucleatum can contribute to the recruitment of other pathogenic bacteria and subsequently to the initiation and the progression of periodontal disease. Indeed, positive association between F. nucleatum, P. gingivalis, P. intermedia, and Bacteroides forsythus in subgingival-plaque samples have been reported previously (2). More importantly, colonization by P. intermedia was found to be due to F. nucleatum, since P. intermedia was never detected in a site unless F. nucleatum was also present (1). Combinations of F. nucleatum, B. forsythus and Campylobacter rectus have been reported in periodontal sites that had the most attachment loss and the deepest pockets (29). The complex of F. nucleatum, B. forsythus, and C. rectus was also found in patients refractory to treatment (13, 29). Increase in the number of bacteria associated with F. nucleatum might later serve to recruit and activate local immune cells, resulting in tissue destruction and the progression of periodontal disease. Indeed, colonization by other oral bacteria can serve to either compete with or cover the sites on F. nucleatum which are responsible for the induction of death in PBMCs. Therefore, since the generation of periodontal disease has been attributed to the superactivity of the immune cells in terms of the production of cytokines, such as interleukin 1 and TNF, decrease or loss of apoptotic signaling by F. nucleatum in PBMCs could serve as an important step in the progression of periodontal disease. These contrasting hypotheses indicate the complexity of the immune defense needed in the oral cavity in order to ensure a balanced state of oral health. We have just started to address such issues in terms of host-parasite interaction in the maintenance of oral health, and induction of apoptosis by some of the oral bacteria might represent the heart of the complexity with which we are faced in the oral cavity.

Ligation of death signal-transmitting receptors, such as TNFR1 and FAS/APO1, initiates the process of cell death and leads to the activation of ICE proteases, resulting in the degradation of chromosomal DNA (15, 24). The inhibition of cell death mediated in the presence of ICE inhibitor indicates the importance of this pathway in the initiation of apoptotic cell signaling in PBMCs by F. nucleatum. In contrast, a decrease in cellular NF-B results in a significantly higher sensitivity of the cells to F. nucleatum-mediated death, indicating the protective role of this protein in bacterium-mediated cell death as well as in TNF-- and radiation-mediated cell death (5, 6, 33, 35).

F. nucleatum induced significantly higher levels of death in PMNs than in PBMCs when they were cocultured in the presence of similar numbers of the oral bacteria. PMNs are important effector cells in first-line defense against bacterial pathogens. Indeed, induction of death in both PMNs and PBMCs by F. nucleatum indicates the ability of this organism to mediate a generalized paralysis of the immune system. Although we did not observe a significant induction of death by F. nucleatum on Cal 27 and SCC4 oral keratinocyte cell lines, the effect of this bacterium on normal human keratinocytes remains to be elucidated. Indeed, sonic extracts from F. nucleatum and A. actinomycetemcomitans have been shown to have a cytotoxic effect on human gingival fibroblasts (30). Thus, the overall paralysis of immune function by F. nucleatum might play an important role in the initiation and progression of periodontal disease.

Based on our findings, we propose the following model system for the role of F. nucleatum in the induction of periodontal disease. Initial colonization and increase in the number of F. nucleatum cells can cause depletion of immune cells at the site of the infection due to the induction of apoptotic cell death. This immunosuppression will lead to the recruitment and the binding of other pathogenic microorganisms to the sites previously colonized by F. nucleatum. Binding of other bacteria to the active sites on F. nucleatum may in turn compete with the sites on F. nucleatum, which induces death of the lymphocytes. Thus, local immune activation and expansion of lymphocytes to oral bacteria bound to F. nucleatum could initiate hypersensitivity and result in the characteristic tissue injury observed in periodontal diseases.