Neutrophils rescue gingival epithelial cells from bacterial-induced apoptosis

Abstract

In the pathogenesis of chronic inflammatory periodontal disease, neutrophils are recognized as a major cellular component from the histopathology of the periodontal lesion around teeth and from clinical cases where absence or dysfunction of neutrophils results in major periodontal destruction. Neutrophils are recruited in vast numbers into the gingival crevice during periodontal inflammation, attracted by microbial plaque chemoattractants and chemokines released following microbial perturbation of gingival epithelial cells. Porphyromonas gingivalis, a major periodontopathogen, triggers a vast array of cellular responses in gingival epithelial cells but also induces apoptosis. We demonstrate here that neutrophils, when combined in a P. gingivalis challenge assay of epithelial cells, prevent epithelial cell apoptosis by phagocytosing P. gingivalis and later undergoing apoptosis themselves. By removing P. gingivalis by phagocytosis, neutrophils also protect the host from the harmful effects of its microbial proteases, which degrade inflammatory cytokines and other host molecules.

Neutrophils are the most abundant leukocyte within the periodontal tissues in early and chronic periodontal lesions [1]. The concentration of neutrophils in the periodontal tissues exceeds that of blood. In minimally inflamed gingiva, 2.5 × 10⁷ PMN/cm³ infiltrate the connective tissue, and 1.7 × 10⁸ PMN/cm³ are found at the junctional epithelium [2]. The nature of lymphocytic responses, particularly that of the neutrophils, is an important determinant of destructive periodontal lesions such as periodontitis [3], a multifactorial inflammatory disease linked with atheroma and type II diabetes [4]. An impaired balance between the host’s defense mechanisms and the invading microorganisms results in periodontitis, the most common cause of tooth loss in developed countries [5].

Bacteria and their products interact with gingival epithelia, the first physical barrier to microbial pathogens, to induce an inflammatory response consisting of cytokine, chemokine, and antimicrobial protein production and subsequent recruitment of neutrophils [3, 6, 7]. Among the periodontopathic bacteria, Porphyromonas gingivalis, a Gram-negative, anaerobic bacteria, has been studied extensively over the years. P. gingivalis infection triggers a wide range of host cellular responses [7]. It also dampens the inflammatory response directly through degradation of the cytokines produced by host cells [8].

Literature describing the interactions between P. gingivalis and epithelial or immune cells is numerous [7, 9,10,11]; however, interactions among P. gingivalis, neutrophils, and epithelial cells in combination have not been reported. Describing the kinetics and interactions of bacteria, epithelial cells and neutrophils will better elucidate the innate immune responses relevant in the early stages of infection and specifically in the gingival crevice during periodontal disease, where these three components are histopathologically noted as predominant. Thus, we developed an in vitro model of bacteria-host interaction using primary HGECs, which were challenged with P. gingivalis with neutrophils. The innate responses in terms of cytokine production and cell death were analyzed.

Gingival tissue collection, HGEC isolation, and subculture were performed as described previously [12]. Individual informed consent was collected with Institutional Review Board approval. Antibiotic-free medium was used in all experiments involving the use of bacteria. P. gingivalis 33277 was purchased from American Type Culture Collection (Manassas, VA, USA). After cultivation [8], bacteria were harvested by centrifugation, washed in PBS, and used immediately. For FITC labeling, 2 × 10⁹/mL P. gingivalis were incubated with 1 mg/ml FITC at 37°C, washed, and resuspended in PBS. For neutrophil isolation, human blood was collected from healthy donors by venipuncture and anticoagulated with acid citrate dextrose. Neutrophils were isolated using a method described previously [13]. Following isolation, neutrophils were resuspended in the media used for the experiments. Neutrophil preparations were routinely >98% pure and >99% viable.

Bacterial challenge was carried out as described by Stathopoulou et al. [8]. When HGECs reached confluence (∼10⁶ cells/well), the cells were washed and were incubated first with viable neutrophils at a HGEC:neutrophil ratio of 1:1 (neutrophils=1×10⁶) or 1:10 (1×10⁷) for 30 min and then challenged with live P. gingivalis 33277 at a MOI of 100 (1×10⁸). Bacterial viability was determined at various time-points under the aerobic conditions of the challenge assay and indicated that bacterial viability was diminished but not quenched entirely over the 24-h time course (Supplemental Fig. 1). The 1:1 HGEC:neutrophil ratio was adapted from a previous study [14], and the 1:10 ratio was included in this study as a log difference comparison and to closely approximate the amount of neutrophils in the peridontium [2]. HGECs in plain media only served as controls. At 4 h and 24 h, supernatants were collected and measured for IL-1β, IL-6, IL-8, and IL-10 by ELISA (BD Biosciences, San Diego, CA, USA). The adherent cells were fixed for direct TUNEL assay (Roche Biosciences, Indianapolis, IN, USA). Analysis was done by confocal microscope (Fluoview-500, Olympus, Center Valley, PA, USA). We have noted previously that P. gingivalis at a MOI of 100 induces apoptosis in HGECs at 24 h after challenge, predominantly through gingipain-dependent mechanisms (unpublished results). Furthermore, O’Brien-Simpson and co-workers [15] published similar findings in human periodontitis tissues. In this study, we extend these findings with the striking addition that when neutrophils were coincubated with epithelial cells, no apoptosis occurred even at 24 h after P. gingivalis challenge (Fig. 1D), preserving HGEC integrity as evidenced by intact cytokeratin (Fig. 1F). In addition, neutrophils coincubated with P. gingivalis-stimulated HGECs showed delayed apoptosis (Fig. 1E). However, higher rates of apoptosis occurring early in the challenge are observed in neutrophils incubated in plain media only (Fig. 1C) and with P. gingivalis (Fig. 1D). The spontaneous apoptosis of cultured neutrophils in plain media observed in this study is consistent with a previous study [16], and the promotion of apoptosis seen in live P. gingivalis-stimulated neutrophils is in accordance with the study in PBMCs by Geatch et al. [17], who also noted that heat-killed P. gingivalis had no effect on lymphocyte apoptosis. This could explain the delay in apoptosis seen by Preshaw et al. [18] in neutrophils stimulated by LPS only.

Figure 1.

FITC-conjugated TUNEL staining on primary HGECs and neutrophils (arrows). Neutrophils and confluent HGECs (10⁶), with or without neutrophil coincubation, were challenged with live P. gingivalis (Pg) at 4 h and 24 h. Pan-cytokeratin staining for HGECs (F) is from a 24-h challenge of HGEC + neutrophils with P. gingivalis. Images are shown at the original ×600 magnification, except the confocal clip in G.

Live P. gingivalis are able to produce potent proteases that cause cellular damage and induce apoptosis [7]. The delayed apoptosis of neutrophils coincubated with P. gingivalis-stimulated HGECs observed in this study can be attributed to several factors, including the suppression of neutrophil apoptosis and potent bactericidal effect by human β-defensins [19], secreted by HGECs [20] and phagocytosis (Fig. 2C), resulting in a lowered dose of bacteria. The same bactericidal activity of β-defensins may account for the delayed apoptosis in P. gingivalis-infected HGECs without neutrophil coincubation (Fig. 1B). Quantitative measurements of apoptosis are provided in Supplemental Figures 2 and 3. This finding is an important addition to the current understanding about how phagocytic cells modulate bacterial-oral epithelial cell interaction and assist epithelial cells actively in dealing with bacterial insults. This work is unique in that previous P. gingivalis and gingival epithelial cell reports, including our own [21,22,23,24], have not included neutrophils in the challenge, despite the histopathological and clinical assay evidence that neutrophils are present in high numbers in the gingival sulcus and are capable of phagocytosis [25, 26]. The apoptotic death of activated neutrophils is proposed to be a critical component in the resolution of the inflammatory process [27].

Figure 2.

Neutrophil phagocytosis and its effects on cytokine levels. HGECs were exposed to live FITC-labeled P. gingivalis at a MOI of 100 or coincubated first with neutrophils (N) at a HGEC:neutrophil ratio of 1:10 and then exposed to labeled or unlabeled P. gingivalis. (A) Confocal images of HGECs with neutrophils (arrows) and FITC-labeled P. gingivalis. Some neutrophils clearly show internalized, FITC-labeled P. gingivalis (circles). At 15 min, bacterial clearance is evident already. (B) Confocal x-z scans of one microscopic field confirms labeled P. gingivalis internalized by neutrophils. (C, left) Flow cytometry analyses of the supernatants collected from A. ^†, Mean fluorescence intensity. (right) Scatter dot plot representing one of the phagocytosis assays analyzed in the left figure. Control (neutrophils+unlabeled P. gingivalis) shows two populations: neutrophils only (left) and neutrophils with internalized, unlabeled P. gingivalis (right). This has been compared with P. gingivalis-free neutrophils and other controls (figure not shown). Succeeding figures show neutrophils with internalized, FITC-labeled P. gingivalis shifting to the lower right quadrant in a time-dependent manner. (D) HGECs, with or without coincubation with neutrophils and neutrophils alone (no HGECs), were challenged with live P. gingivalis for 4 h and 24 h. Unstimulated HGECs in plain media were used as controls. The supernatants were collected and then analyzed for IL-1β at 4 h (P<0.0001) and 24 h (P<0.0001; i); IL-8 at 4 h (P<0.0001) and 24 h (P<0.0001; ii); IL-6 at 4 h (P<0.0001) and 24 h (P=0.028; iii); and IL-10 at 4 h (P<0.0001) and 24 h (P<0.0001; iv). All of the values above represent the mean of at least three assays ± sd and analyzed by repeated measures of ANOVA. All images are the original ×600 magnification.

For bacterial challenge for phagocytosis assay, confluent HGECs were coincubated first with neutrophils at a HGEC:neutrophil ratio of 1:10 (neutrophils=1×10⁷). P. gingivalis (1×10⁸), with or without FITC labeling, were then added and incubated for 15, 30, 60, and 90 min. The supernatants were collected at the specified time-points and centrifuged (18 g, 5 min, 4°C) to pellet neutrophils and to discard unphagocytosed bacteria. Quantification of phagocytosis was performed using a method described previously [28]. Analysis was done by flow cytometry (FACSCalibur, BD Biosciences). The following were used as flow cytometry references or controls: neutrophils only, neutrophils with unlabeled P. gingivalis, and neutrophils with unquenched, FITC-labeled P. gingivalis. Statistical significance in all of the experiments was calculated by repeated measures of ANOVA. A P value of <0.05 was considered significant. As expected, neutrophils phagocytose bacteria efficiently to initiate microbial killing. Phagocytosis occurred immediately at 15 min (mean uptake, 31.73%) and continued progressively until the 90-min time-point (76%; Fig. 2, A and C), resulting in efficient clearing of bacteria (Fig. 2A). This explains the absence of bacterial-induced apoptosis in HGECs coincubated with neutrophils in Figure 1 and the reduced or inhibited cytokine degradation in Figure 2D.

P. gingivalis stimulation of neutrophil-incubated HGECs results in the release of pro- and anti-inflammatory and pro- and antiapoptotic cytokines (Fig. 2D). The levels of these cytokines, particularly IL-1β, IL-8, and IL-6, are significantly higher compared with the combined levels of HGECs and neutrophils stimulated separately, suggesting a synergistic response between HGECs and neutrophils upon bacterial challenge. IL-1 has been reported as generally proapoptotic [29], and IL-10 has been shown to induce neutrophil apoptosis [30] but is protective to epithelial cells [31]. The levels of both cytokines are increased at 24 h (Fig. 2D, i and iv), where delayed neutrophil apoptosis is observed, and HGECs remain viable (Fig. 1E). IL-6 and IL-8 attenuate apoptosis [32, 33]. In this study, the level of IL-6 is increased at 4 h (Fig. 2D, iii) when neutrophil apoptosis is not yet evident (Fig. 1E) and drops at 24 h when apoptosis occurs. The IL-8 level is increased at 4 h and 24 h (Fig. 2D, ii) and likely relates to its role in neutrophil chemotaxis and neutrophil function; hence, it essentially delays apoptosis and is produced in large amounts [32].

From the apoptosis, phagocytosis, and cytokine assays, it is evident that HGECs and neutrophils exploit several mechanisms to preserve cellular integrity and resolve inflammation. Phagocytosis reduces the dose of bacteria, resulting in the attenuation of their deleterious effects. Neutrophil apoptosis releases an “eat-me” signal for successful uptake by macrophages [34], and the preservation of extravascular cells such as epithelial cells maintains their proinflammatory function. The review by Middleton et al. [35] underlines the importance of extravascular cells in the release of chemotactic molecules that direct leukocyte extravasations and chemotaxis. Proper neutrophil function is important in the pathogenesis of periodontal disease, as neutropenia and impaired leukocyte function result in generalized periodontitis [36]. Certain conditions, such as periodontitis, are known to feature hyperinflammatory peripheral neutrophils [37], and thus, we will incorporate neutrophils from healthy and periodontitis patients in future challenge assays.

The current findings underscore the importance of neutrophils in modifying the responses of primary gingival epithelial cells to bacterial insults. Epithelial cells constitute the first line of defenses and are bombarded constantly by a vast array of stimuli. Their ability to respond and to release signals for the “professional” defense cells is crucial in setting up an inflammatory response to protect the periodontal tissues. At the same time, the neutrophils, as professional phagocytes, clearly protect epithelial cells from bacterial damage and induction of apoptosis. These results clarify the roles of epithelial cells and neutrophils in mucosal defense and question the translational relevance of reductionist experiments, including our own previous reports, which fail to incorporate cell types that histopathology indicates are relevant to the disease.

AUTHORSHIP

J. C. G. conceived, designed, and performed experiments, analyzed and interpreted data, and drafted the paper. M. R. B. and P. G. S. analyzed and interpreted data. D. F. K. conceived and designed the experiments and drafted the paper.

ACKNOWLEDGMENTS

This research was supported by National Institutes of Health, National Institute of Dental and Craniofacial Research, DE017384.

DISCLOSURE

The authors declare no conflicts of interest.