OUP user menu

Inhibition of epithelial cell apoptosis by Porphyromonas gingivalis

Simin F. Nakhjiri, Yoonsuk Park, Ozlem Yilmaz, Whasun O. Chung, Kiyoko Watanabe, Azza El-Sabaeny, Kyewhan Park, Richard J. Lamont
DOI: http://dx.doi.org/10.1111/j.1574-6968.2001.tb10706.x 145-149 First published online: 1 June 2001


Porphyromonas gingivalis is periodontal pathogen that is capable of invading gingival epithelial cells (GECs). Apoptotic responses of primary cultures of GECs to P. gingivalis were investigated with a DNA fragmentation ELISA assay. P. gingivalis induced a transient increase in GEC DNA fragmentation; however, after prolonged incubation GECs did not undergo apoptosis. Furthermore, P. gingivalis blocked apoptosis in GECs following stimulation with camptothecin. Immunoblotting of GECs with Bcl-2 or Bax antibodies showed that P. gingivalis up-regulated Bcl-2 levels in GECs, whereas Bax levels were transiently elevated and declined after 24 h stimulation. Streptococcus gordonii did not affect levels of either molecule. RT-PCR demonstrated that induction of Bcl-2 occurs at the transcriptional level. The results suggest that P. gingivalis can inhibit apoptosis in GECs by up-regulation of the anti-apoptotic molecule Bcl-2. The prevention of host cell apoptosis may represent a strategy for P. gingivalis survival within invaded GECs.

  • Invasion
  • Apoptosis
  • Bcl-2
  • Bax
  • Porphyromonas gingivalis

1 Introduction

Porphyromonas gingivalis, a Gram-negative anaerobe, is an aggressive pathogen in adult and severe periodontal diseases that are characterized by chronic inflammation of the gingiva, destruction of periodontal tissues and loss of the alveolar bone [1]. P. gingivalis elaborates a vast array of virulence factors including proteinases that can degrade host immune effectors and tissue components, along with toxic metabolites and cellular constituents [2]. In addition to these overtly destructive factors, P. gingivalis engages in an intricate interaction with the epithelial cells that line the gingival crevice. Association between P. gingivalis and these gingival epithelial cells (GECs) results in epithelial cell cytoskeletal rearrangements and calcium ion fluxes that culminate in bacterial internalization within the GECs [2,3]. P. gingivalis cells rapidly accumulate and replicate in the cytoplasm, predominantly in the perinuclear area [4].

Intracellular invasion is a common feature of bacteria that initiate infections at mucosal membranes [5]. In many cases intracellular bacteria can induce or inhibit host cell apoptosis. For example, Salmonella and Shigella can stimulate apoptosis in macrophages, and Legionella can induce apoptosis in both macrophages and epithelial cells [6,7]. In contrast, chronic Helicobacter pylori infection induces an apoptosis-resistant phenotype in gastric epithelial cells [8]. These organisms can impinge upon a number of components of the tightly regulated, multistep intracellular apoptotic pathway, including caspases, nuclear factor κB and the Bcl-2 family of proteins [68]. As apoptosis allows removal of unwanted or damaged cells, epithelial cell apoptosis can be viewed as a means by which the host eliminates infected cells, thus promoting normal epithelial cell growth and integrity [7]. Disruption of apoptotic control pathways, therefore, has an important bearing on pathogen–host interactions.

The outcome of the interaction between P. gingivalis and host cells with regard to apoptosis is uncertain. P. gingivalis products such as butyric acid and proteases can induce apoptosis in T- and B-cells and in fibroblasts, respectively [9,10]. However, P. gingivalis does not induce apoptosis in peripheral blood mononuclear cells [11] and P. gingivalis lipopolysaccharide may inhibit apoptosis in polymorphonuclear leukocytes [12]. Moreover, there are several reports of epithelial cells surviving for up to 8 days after exposure to P. gingivalis [4,1315]. In this study, we investigated the influence of P. gingivalis on apoptosis in primary cultures of GECs.

2 Materials and methods

2.1 Bacteria and culture conditions

P. gingivalis 33277 cells were cultured in trypticase soy broth, supplemented with yeast extract (1 mg ml−1), hemin (5 μg ml−1) and menadione (1 μg ml−1), under anaerobic conditions at 37°C overnight. Streptococcus gordonii DL1 cells were grown in trypticase peptone broth supplemented with yeast extract (5 mg ml−1) and 0.5% glucose at 37°C under static conditions. Bacteria were harvested and washed in phosphate-buffered saline by centrifugation prior to assay. Bacterial numbers were determined in a Klett–Summerson photometer.

2.2 Epithelial cell culture

Primary cultures of GECs were generated as described previously [3]. Briefly, healthy gingival tissue was collected from patients undergoing surgery for removal of impacted third molars. Surface epithelium was separated by overnight incubation with 0.4% dispase. Single epithelial cells were recovered by centrifugation following digestion with 0.05% trypsin and 0.53 mM EDTA. Cells were cultured as monolayers in serum-free keratinocyte growth medium (Clonetics) at 37°C in 5% CO2. GECs were used for experimentation at 70–80% confluence and reacted with bacterial cells or other test reagents in growth medium under normal culture conditions.

2.3 Assessment of epithelial cell apoptosis

To detect fragmentation of DNA in apoptotic epithelial cells, histone associated DNA fragments were examined in a cell death detection ELISA kit (Boehringer Mannheim). GEC cytoplasmic extracts were added to wells of ELISA plates coated with monoclonal antibodies against histones. The presence of histone-associated DNA fragments was then detected in a sandwich ELISA using anti-DNA peroxidase with 2,2′-azino-di-[3-ethylbenzthiazoline-sulfonate] substrate. Absorbance was measured at 405 nm and background at 490 nm. As a positive control for apoptosis GECs were incubated with camptothecin (2 μg ml−1) for 4 h.

2.4 Immunoblotting

Epithelial cells were lysed by sonication (30 s) on ice and cell debris removed by centrifugation (5000×g 10 min). Protein was determined by the Bio-Rad protein assay. Proteins (2 μg) were separated by 10% SDS–PAGE and transferred to nitrocellulose. After blocking with 0.05% Tween 20–3% bovine serum albumin, blotted proteins were reacted with primary antibodies to Bcl-2 family proteins (Bcl-2, Bax, Bcl-xS, Bcl-xL; Santa Cruz) diluted 1:1000. Antigen–antibody binding was visualized with peroxidase-conjugated secondary antibodies and chemiluminescence detection (Amersham Pharmacia). Duplicate blots were probed with keratin 5/6 antibodies (Chemicon) to ensure equal protein loading.

2.5 RT-PCR

RNA was prepared from GECs with Totally RNA™ (Ambion). Bcl-2 specific primers were: Bcl2-1, 5′-CACCTGTGGTCCACCTGAC-3′; Bcl2-2, 5′-AGGGCCAAACTGAGCAGAG-3′. Reverse transcription was performed using RETROscript™ (Ambion), following the protocol recommended by the manufacturer. Briefly, 2 μg of RNA was denatured at 85°C for 3 min in 12 μl solution containing 100 pmol of oligo(dT). The enzyme mixture (2 μl of 10×RT buffer, 4 μl of dNTP mix, 1 μl of RNase inhibitor, and 1 μl of reverse transcriptase) was added and incubated at 42°C for 1 h. The enzyme was inactivated by incubation at 92°C for 10 min. The resulting cDNA was amplified in a 20 μl PCR mixture containing 1×PCR buffer, 1 μl of cDNA, 2.5 mM MgCl2, 10 mM dNTP mix, 50 pmol of each primer, and 1 U of Taq DNA polymerase (Promega). The amplification conditions were as follows: denaturation at 94°C for 30 s, annealing at 65°C for 30 s and elongation at 72°C for 30 s, for 30 cycles. As a control, cDNA was also amplified with primers for actin (Act-1, 5′-AGAGCTACGAGCTGCCTGC-3′; Act-2, 5′-AAAGCCATGCCAATCTCATC-3′).

3 Results

3.1 GEC apoptosis

Our previous results [4,13] and laboratory observations show that exposure of GECs to P. gingivalis for up to 48 h, at a multiplicity of infection (MOI) ranging from 10 to 1000, does not affect epithelial cell viability as evidenced by intracellular hydrolysis of calcein-acetomethyl ester and physiologic intracellular calcium ion concentrations. These qualitative viability observations were quantitated with regard to apoptotic cell death by a cell death detection ELISA (Fig. 1). After 2 h stimulation of GECs by P. gingivalis (MOI 100) DNA fragmentation in GECs was initiated. As DNA fragmentation precedes some of the more visible markers of apoptosis such as plasma membrane breakdown, this finding is not inconsistent with results from visual inspection of GECs after 2 h exposure to P. gingivalis that show phenotypically normal cells [4]. After 24 h P. gingivalis GEC incubation, DNA fragmentation levels had returned to normal and the GECs were not apoptotic. In contrast, S. gordonii, a non-pathogenic, non-invasive oral organism, induced apoptosis in GECs. Moreover, a 24-h preincubation with P. gingivalis blocked GEC apoptosis induced by camptothecin. Identical results were obtained at a bacterial MOI of 1000 (not shown). Thus, P. gingivalis appears to actively suppress apoptosis in GECs.

Figure 1

ELISA of histone-associated DNA fragments in the cytoplasm of GECs. Control represents GECs under normal culture conditions. P. gingivalis or S. gordonii were incubated with GECs at a MOI of 100 for the times indicated. Camptothecin (2 μg ml−1) was incubated with GECs for 4 h, either with (+Pg) or without (−Pg) prior exposure to P. gingivalis for 24 h. Error bars represent standard deviation, n=3.

3.2 Levels of expression of proteins associated with apoptosis

Immunoblots of GEC lysates showed no detectable expression of the anti-apoptotic protein Bcl-2 in unstimulated GECs or in cells stimulated with P. gingivalis for 2 h at a MOI of 100 (Fig. 2). In concordance with the cell death assay however, after 24 h incubation P. gingivalis induced expression of both the ∼26- and ∼29-kDa isoforms of Bcl-2. Exposure of GECs to S. gordonii did not result in detectable Bcl-2 expression under any of the test conditions, a result also consistent with the apoptosis assay. Expression of the anti-apoptotic protein Bcl-xL was not affected by P. gingivalis (not shown). Identical results were obtained at a bacterial MOI of 1000 and with cells derived from two different individuals. Heat-killed P. gingivalis did not induce expression of Bcl-2 or Bcl-xL.

Figure 2

Immunoblot analysis of Bcl-2 expression in GECs after exposure to P. gingivalis or S. gordonii at a MOI of 100. Control represents GECs under normal culture conditions. Tumor necrosis factor α (TNFα, 10 ng ml−1) is a positive control for Bcl-2 expression. Specificity of Bcl-2 antibodies (1:1000) was confirmed by pretreatment with Bcl-2 peptide (Santa Cruz) prior to probing GEC extract stimulated with P. gingivalis for 24 h.

As infection by P. gingivalis induced DNA fragmentation at the initial stage, expression of the pro-apoptotic proteins Bax and Bcl-xS was investigated. Resting and S. gordonii-stimulated cells did not exhibit detectable Bax expression (Fig. 3). However, P. gingivalis induced Bax expression in GECs after 2 h exposure. After 24 h exposure Bax returned to undetectable levels. Induction of Bax by P. gingivalis required viable cells as heat-killed P. gingivalis did not affect Bax levels. The finding that S. gordonii did not cause an increase in Bax expression suggests that cell death in response to this organism is Bax-independent. Expression of Bcl-xS by GECs was not altered by P. gingivalis infection (not shown). Identical results were obtained with cells derived from two different individuals and with a bacterial MOI of 1000.

Figure 3

Immunoblot analysis of Bax expression in GECs after exposure to P. gingivalis or S. gordonii at a MOI of 100. Control represents GECs under normal culture conditions. Heat-killed P. gingivalis were treated at 80°C for 30 min.

3.3 Correlation of Bcl-2 protein and mRNA levels

To investigate if the observed elevation of Bcl-2 expression is associated with an increase in gene transcription, steady-state levels of mRNA were determined by RT-PCR (Fig. 4). P. gingivalis caused a significant increase in Bcl-2 mRNA levels compared to normal or S. gordonii-stimulated GECs.

Figure 4

RT-PCR of Bcl-2 or actin (control) mRNA in control or GECs exposed to P. gingivalis or to S. gordonii at a MOI of 100 for 24 h.

4 Discussion

Apoptosis, or programmed cell death, is a fundamentally important biological process that is required by multicellular organisms to eliminate redundant, damaged or infected cells and so maintain normal development, integrity and homeostasis of the organism [16]. Apoptosis is characterized by cell shrinkage and condensation of the nucleus along with specific protein degradation and DNA fragmentation. There are at least six surface receptors on mammalian cells that transduce apoptotic signals subsequent to ligand binding [6]. Intracellularly, the major components of the apoptotic machinery include a family of Bcl-2 proteins [17] that integrate diverse upstream signals to control the cellular commitment to apoptosis, and a family of cysteine proteases (caspases) that act downstream of Bcl-2 and cleave critical target proteins to effect cell death [16,17].

Increasingly, it is becoming appreciated that epithelial cell responses to pathogenic bacteria can involve either elevated apoptosis, as in the case of Salmonella dublin and Legionella, or suppression of cell death, as in the case of H. pylori. Most commonly, bacterial modulation of epithelial cell apoptosis has been studied in cell lines [7,8]. Such models have obvious limitations inasmuch as these cells are likely to have limited control over apoptotic pathways. We have developed a culture system to maintain primary GECs in vitro for a limited number of passages (typically five to seven) [3,4]. These primary, short-lived, epithelial cells were utilized to study the apoptotic responses to the periodontal pathogen P. gingivalis. Expression of Bcl-2 family members in unstimulated GECs was low, not detectable by immunoblotting with commercially available antibodies. This contrasts with gastric epithelial cell lines in which Bcl-2, Bax, Bak and Bcl-xL are expressed [8] and may be the result of either tissue-specific differences in expression, or tighter control of Bcl-2 family proteins in non-transformed cells.

As measured by a DNA fragmentation ELISA, the early response of GECs to P. gingivalis infection was induction of apoptosis. However, after 24 h of exposure to P. gingivalis, DNA fragmentation levels had returned to background and the cells were not apoptotic. P. gingivalis cells invade GECs rapidly and in high numbers [3], and antibiotic protection assays indicate that individual GECs can harbor several hundred P. gingivalis, results that are corroborated by direct microscopic observation [4]. Thus, despite the burden of large numbers of intracellular P. gingivalis, GECs do not undergo programmed cell death over at least a 24-h timeframe. Moreover, exposure to P. gingivalis protected GECs from apoptosis induced by camptothecin, a human topoisomerase I inhibitor used as a chemotherapeutic agent in cancer treatment [18]. Similarly, H. pylori, which is associated with gastric cancer, confers resistance to apoptosis through decreased expression of p27kip1[8]. These findings suggest that P. gingivalis may disrupt tissue homeostasis in the periodontal pocket and allow the intriguing speculation that P. gingivalis could be associated with some forms of oral cancer.

GEC apoptotic responses to P. gingivalis were correlated with expression levels of certain Bcl-2 family proteins. Bcl-2-like proteins can be functionally categorized on the basis of their ability to either promote (e.g. Bax, Bcl-xS, Bad) or suppress (e.g. Bcl-2, Bag, Bcl-w and Bcl-xL) apoptosis. Apoptosis may be controlled by interactions with the scaffold proteins necessary for the assembly of certain caspase precursors, or by the formation of ion channels in mitochondrial membranes that affect permeability and release of cytochrome c, thereby also modulating caspase activation [17]. Increased expression of Bcl-2 by P. gingivalis is, therefore, likely to make a major contribution to apoptosis resistance. Moreover, as one outcome of camptothecin treatment is reduced Bcl-2 expression [18], induction of Bcl-2 by P. gingivalis could also account for the ability of the organism to block the activity of camptothecin. While levels of Bcl-2 were elevated, Bcl-xL expression was not modulated by P. gingivalis. This is consistent with earlier studies that have shown that Bcl-xL has a pattern of expression that is distinct from Bcl-2 [19]. In contrast to Bcl-2, expression levels of Bax and Bcl-xS in GECs were not above normal resting levels after 24 h exposure to P. gingivalis. As pro- and anti-apoptotic Bcl-2 family members can heterodimerize and titrate one another's activity [17], low levels of pro-apoptotic proteins will not interfere with the ability of Bcl-2 to function in an anti-apoptotic capacity. Interestingly, levels of Bax were transiently induced by P. gingivalis, increasing after 2 h bacterial exposure before returning to control levels after 24 h. This finding, coupled with the transient increase in DNA fragmentation, suggests that the first response of GECs to P. gingivalis infection is to initiate apoptotic pathways but these are then stalled by P. gingivalis through up-regulation of Bcl-2 and down-regulation of Bax.

The mechanisms by which P. gingivalis impinges on control of Bcl-2 family members remain to be characterized at the molecular level. Heat-treated P. gingivalis were unable to induce Bcl-2. This may be the result of failure of heat-killed P. gingivalis cells to locate intracellularly [4], or the denaturation of a protein necessary for induction of activity. In either event, P. gingivalis is an active participant in the process.

In conclusion, we have established that P. gingivalis does not induce apoptosis in primary cultures of short-term gingival epithelial cells, and furthermore, that P. gingivalis-infected cells are resistant to apoptosis induced by camptothecin. Apoptosis is associated with elevated levels of Bcl-2 and suppression of initial elevation of Bax. The ability to suppress apoptosis may be an important survival strategy of P. gingivalis, facilitating retention within the periodontal tissues. Further definition of the molecular basis of apoptosis may provide insights into the pathogenicity of P. gingivalis and of the full range of diseases related to this invasive organism.


Supported by NIDCR DE11111 and DE07023.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
View Abstract