ABSTRACT
Periodontal disease (PD) is a chronic inflammatory and alveolar bone destructive disease triggered by oral biofilm-producing microorganisms, such as Aggregatibacter actinomycetemcomitans. The levels of the phospholipid platelet-activating factor (PAF) in the saliva, gingival crevicular fluid, and periodontal tissues are significantly increased during inflammatory conditions, such as PD, but the exact mechanism that links PAF to alveolar bone resorption is not well understood. In the current study, alveolar bone resorption was induced by experimental PD through the oral inoculation of A. actinomycetemcomitans in wild-type (WT) and PAF receptor knockout (Pafr −/−) mice. In vitro experiments using A. actinomycetemcomitans lipopolysaccharide (LPS)-stimulated RAW 264.7 cells treated with a PAF receptor antagonist (UK74505) were also performed. The expression of lyso-PAF acetyltransferase in periodontal tissues was significantly increased 3 h after A. actinomycetemcomitans LPS injection in mice. WT and Pafr −/− mice that were subjected to oral inoculation of A. actinomycetemcomitans presented neutrophil accumulation and increased levels of CXCL-1 and tumor necrosis factor alpha (TNF-α) in periodontal tissues. However, Pafr −/− mice presented less alveolar bone loss than WT mice. The in vitro blockade of the PAF receptor impaired the resorptive activity of A. actinomycetemcomitans LPS-activated osteoclasts. In conclusion, this study shows for the first time that the blockade of PAF receptor may contribute to the progression of PD triggered by A. actinomycetemcomitans by directly affecting the differentiation and activity of osteoclasts.
INTRODUCTION
Periodontal disease (PD) is an inflammatory condition of the tooth-supporting structures caused by oral biofilm-producing bacteria, such as Aggregatibacter actinomycetemcomitans (1). The lipopolysaccharide (LPS) of A. actinomycetemcomitans is a strong virulence factor that exhibits several activities as the induction of the inflammatory mediators' release, which culminates with alveolar bone resorption (2). In this regard, the release of platelet-activating factor (PAF) by inflamed periodontal tissues has been previously demonstrated (3). Indeed, there is a significant positive correlation between periodontal parameters and the levels of PAF in both serum and gingival crevicular fluid (GCF) from patients suffering from periodontitis (4).
PAF is a bioactive phospholipid derived from arachidonic acid that is produced by different cells in response to exogenous stimulation, such as LPS, and rapidly synthesized in response to cell-specific stimuli (5, 6). PAF exerts several biological activities via activation of a G-protein-coupled PAF receptor (PAFR) (5, 7, 8). The biosynthesis of PAF is performed by acetyl-coenzyme A (CoA)-lyso-PAF acetyltransferase (5). PAF has multiple physiological and pathological functions, being implicated in many inflammatory diseases, such as bronchial asthma (9), sepsis (10, 11), graft-versus-host disease (12), Dengue virus infection (13), and intestinal ischemia and reperfusion injury (14), as well as in diseases associated with bone resorption, such as osteoporosis (15).
PAF is expressed in human inflamed gingival tissues (16) and may be associated with bone resorption. It was shown that bacterial LPS can also directly activate PAFR in vitro (17, 18). Another line of evidence linking PAF to bone resorption is that PAF can act directly on osteoclasts (19). In accordance with this, PAF receptor-deficient mice present markedly attenuated bone resorption in a postmenopausal osteoporosis model (15). Nevertheless, the mechanism that links PAF production to alveolar bone loss in experimental PD or in osteoclast activity remains unclear.
Thus, the aim of the present study was to determine the role of PAF receptor in experimental PD. Our results show that PAFR-deficient (Pafr−/−) mice had impaired alveolar bone loss and a reduced number of tartrate-resistant acid phosphatase (TRAP)-positive cells, but the inflammatory parameters (accumulation of neutrophils and production of tumor necrosis factor alpha [TNF-α] and CXCL-1) were similar to those of wild-type (WT) mice. The in vitro blockade of the PAF receptor reduced osteoclast differentiation and activity. These results suggest that the PAF receptor is not important in triggering the A. actinomycetemcomitans-induced inflammatory response but plays an important role in driving bone remodeling in PD by controlling the function of osteoclasts.
MATERIALS AND METHODS
Microorganism.The bacterium isolate Aggregatibacter actinomycetemcomitans strain FDC Y4, from the American Type Culture Collection (ATCC), was used throughout the experiments. A. actinomycetemcomitans was grown microaerobically at 37°C, under conditions of 5 to 10% CO2 using a glass jar in a biochemical oxygen demand incubator (Thermo Scientific, Waltham, MA), in Trypticase soy broth (TSB; Difco Laboratories, Detroit, MI) supplemented with 0.5% yeast extract (Difco Laboratories, Detroit, MI) for 24 h, after which the suspension was centrifuged (355 × g for 5 min). Thereafter the pellet was suspended in phosphate-buffered saline (PBS) to obtain an A560 of 1.02, which corresponded to 1 × 109 CFU/ml.
Mice.C57BL/6 WT and PAF receptor-deficient (Pafr−/−) mice derived from a WT background were obtained from the Bioscience Unit of the Biological Sciences Institute (Brazil) and were housed under standard conditions in separated cages with free access to commercial chow and water. All of the animals were 6 to 8 weeks old. The experimental protocol used in this study was approved by the local Institutional Animal Ethics Committee under protocol number 256/2008.
Periodontal infection.As previously described (20, 21), the experimental groups received a direct injection of A. actinomycetemcomitans into the palatal gingival tissue at the midline near the second molar. Each mouse was injected with 10 μl of a suspension of A. actinomycetemcomitans containing 1 × 109 CFU/ml in phosphate-buffered saline (PBS). Immediately after the injection, 100 μl of the suspension of A. actinomycetemcomitans containing 1 × 109 CFU/ml in PBS plus 1.5% carboxymethylcellulose was inoculated in the oral cavity with a micropipette. After 48 and 96 h, the protocol was repeated. The experimental and control groups were evaluated at different time points after the infection (five mice of each strain at each time point per group). The negative controls included sham-infected mice, which received an injection of 10 μl of PBS into the palatal gingival tissue and 100 μl of PBS with 1.5% carboxymethylcellulose, and noninfected animals. Each group was housed in separate and appropriate animal cages under standard conditions.
Purification of A. actinomycetemcomitans LPS.The purification of A. actinomycetemcomitans LPS was conducted using the LPS extraction kit (iNtRON Biotechnology, Seoul, South Korea) according to the manufacturer's instructions. LPS extract was dissolved in 10 mM Tris-HCl buffer (pH 7.5), 1 mg/ml DNase I, and 1 mg/ml RNase, incubated at 37°C overnight, and treated with proteinase K (final concentration, 2 mg/ml) at 37°C overnight. LPS was collected by ethanol precipitation (20,000 × g, 15 min, 4°C) using a 0.2 volume of 5 M NaCl and 2 volumes of 95% ethanol. After a wash with 70% ethanol, LPS was dried, retreated with an LPS extraction kit (iNtRON Biotechnology, Seoul, South Korea), and applied to a polymyxin B immobilized column (Detoxi-Gel AffinityPak prepacked columns; Pierce, Rockford, IL) for further purification. LPS eluted from the column was collected by ethanol precipitation (20,000 g, 15 min, 4°C) using a 0.4 volume of 5 M NaCl and 4 volumes of 95% ethanol. After a wash with 70% ethanol, pellets were dried and dissolved in 1 ml of distilled water. LPS concentrations were determined by the malondialdehydethiobarbituric acid reaction (22). After lyophilization, the obtained sediment was stored at −20°C.
Alveolar bone loss induced by A. actinomycetemcomitans LPS.The induction of alveolar bone loss by A. actinomycetemcomitans LPS was performed as previously described (2). Briefly, 5 μg of A. actinomycetemcomitans LPS in 3 μl of phosphate-buffered saline (PBS) was injected every 48 h for 20 days using a microsyringe (Hamilton, Reno, NV) into the palatal papilla between the first and the second molar of the right hemimaxilla under ketamine and xylazine intraperitoneal anesthesia. In addition, PBS was similarly injected in the left hemimaxilla, and this side was used as a control. Groups of five mice were euthanized 3 h after one injection of A. actinomycetemcomitans LPS or PBS to assess lyso-PAF acetyltransferase production. Another group of mice was euthanized 24 h after the tenth A. actinomycetemcomitans LPS injection, and alveolar bone loss was evaluated.
qPCR.RNA was extracted using TRIzol (Life Technologies, Grand Island, NY), and reverse transcription (RT) was performed to obtain cDNA. RT was carried out with a reaction mixture containing 2 μg total RNA, 50 μM oligo(dT) primer, 10 mM deoxynucleoside triphosphates (dNTPs), and 200 U/μl of a SuperScript III reverse transcription set (Life Technologies, Grand Island, NY) according to the manufacturer's instructions. Quantitative PCR (qPCR) analyses were performed using a 7500 sequence detection instrument and SYBR green chemistry (Life Technologies, Grand Island, NY). SYBR green PCR master mix (Life Technologies, Grand Island, NY), 1 μM specific primers, and 2 μl of template cDNA were used in each reaction. The primer sequences, the predicted amplicon sizes, and the annealing and melting temperatures were designed using the Primer Express software (Life Technologies, Grand Island, NY) along with the Primer BLAST algorithm. The PCR conditions were 95°C (10 min) followed by 40 cycles of 95°C (1 min), 60°C (30 s), and the standard melting curve program. For cDNA analysis, gene expression levels were determined using the 2−ΔΔCT method (23) and normalized to the expression of the 18S rRNA gene, and data are shown as fold increases over the negative-control (noninfected) group. No-template controls were also included. The following specific primer sequences were used: 18S rRNA gene forward, CGT TCC ACC AAC TAA GAA CG; 18S rRNA gene reverse, CTC AAC ACG GGA AAC CTC AC; lyso-PAF acetyltransferase forward, GCC CAG GTG GCT TTC ATG ACG T; and lyso-PAF acetyltransferase reverse, CCA GCA AAC CAC ATG GTG CGC. DNA extraction from the A. actinomycetemcomitans Y4 suspension and from mouse periodontal tissues was performed as previously described (24). After extraction, the DNA was suspended in sterile water and quantified by measuring the optical density at 260 nm. To quantify the A. actinomycetemcomitans load in the maxillae of the infected mice, quantitative real-time PCR DNA analyses were performed as previously described (21). SYBR green PCR master mix (Life Technologies, Grand Island, NY), 100 nM specific primers (forward, ATGCCAACTTGACGTTAAAT; reverse, AAACCCATCTCTGAGTTCTTCTTC), and 5 ng of DNA were used in each reaction. No-template controls were also included. The standard PCR conditions were 95°C (10 min) followed by 40 cycles of 94°C (1 min), 56°C (1 min), and 72°C (2 min), and the standard denaturation curve was used. A. actinomycetemcomitans load was determined by qPCR using a relative quantitation curve based on known amounts of cultured bacteria as determined by optical density measurements. A total of 1 × 104 CFU/ml was extracted, and the curve was built using 10-fold serial dilutions of this template DNA.
Quantification of alveolar bone loss.Quantification of alveolar bone loss was performed as previously described (20, 25). The maxillae were hemisected, exposed overnight to 3% hydrogen peroxide, and mechanically defleshed. To distinguish the cement-enamel junction (CEJ), mouse maxilla jaws were stained with 0.3% methylene blue (26). The palatal faces of the molars were photographed at ×20 magnification using a stereomicroscope (Metrimpex Hungary/PZO; Labimex, Hungary) and a digital camera (Kodak EasyShare C743; Rochester, NY). The images were analyzed using Image J software (http://rsbweb.nih.gov/ij). Quantitative analysis was used for the measurement of the palatal area between the CEJ and the alveolar bone crest (ABC) in the first molar in millimeters squared. At each time point five animals were analyzed. For each animal injected with A. actinomycetemcomitans LPS, the alveolar bone loss was defined as the difference between the CEJ-ABC palatal areas of the left and right arches.
Histological analysis.Five mice selected at random from each group were euthanized, and the maxilla tissues were obtained and fixed in 10% formalin at pH 7.4 for 24 h at room temperature. The specimens were demineralized in 14% EDTA for 2 weeks, dehydrated in graded ethanol, and embedded in paraffin. Serial sections (5 μm) were stained for tartrate-resistant acid phosphatase (TRAP; Sigma-Aldrich, St. Louis, MO). The alveolar bone crest between the first and second molars was used for the osteoclast counts on five fields per section. For each animal, three maxilla sections were analyzed. All of the slides were counted in a blinded manner by a single examiner.
MPO activity.The activity of myeloperoxidase (MPO) in the maxillae was measured as previously described (20). Briefly, a posterior portion of the maxilla, including the teeth, periodontal soft tissues, and alveolar bone, were removed, weighed, and processed. Upon processing, the tissue was assayed for myeloperoxidase activity by measuring the change in optical density at 450 nm (OD450) using tetramethylbenzidine (TMB). The results are expressed as neutrophil relative units, which denote the myeloperoxidase activity of casein-elicited murine peritoneal neutrophils processed using the same method. One experiment, representative of three, is presented in the results.
ELISA.The levels of cytokines in the maxillae, including the periodontal tissues, teeth, and alveolar bone, were measured as previously described (20). For protein extraction, the samples were homogenized in phosphate-buffered saline (PBS) containing protease inhibitors and 0.05% Tween 20, pH 7.4, using a Power Gen 1000 homogenizer (Fisher Scientific, Pittsburgh, PA) and then centrifuged (9,000 × g, 10 min), and the supernatants were stored at −20°C. The cytokine concentrations in the periodontal extracts were determined using the mouse TNF-α or CXCL-1 DuoSet enzyme-linked immunosorbent assay (ELISA) development kits (R&D Systems, Minneapolis, MN). All assays were carried out according to the manufacturer's instructions. The lower limit of the assay was 31.2 pg/ml for TNF-α and 15.6 pg/ml for CXCL-1. The results are expressed as picograms of cytokine per 100 mg of maxilla tissues for one experiment (representative of three experiments).
Osteoclast culture.The murine monocyte/macrophage cell line RAW 264.7 was purchased from the ATCC (Manassas, VA) and grown in Dulbecco's modified Eagle medium (DMEM; GIBCO, Invitrogen, Carlsbad, CA) supplemented with 10% heat-inactivated fetal bovine serum (FBS), penicillin (40 U/ml), and gentamicin (40 μg/ml). All cells were grown in a humidified atmosphere containing 5% CO2 at 37°C. For osteoclast differentiation, RAW 264.7 cells were suspended in DMEM containing 10% FBS, seeded at 1 × 105 cells/well in 24-well culture plates over 13-mm glass coverslips, and cultured with 50 ng/ml of soluble RANKL (PeproTech Inc., Rocky Hill, NJ) for 4 days. Afterwards, the cells were treated with PAF receptor antagonist UK74505 (10−7 M for 10 min) or PAF (10−7 M, 30 min; Calbiochem, Merck KGaA, Darmstadt, Germany) and then stimulated with A. actinomycetemcomitans LPS (2.5 μg/ml) or DMEM for 48 h. The supernatant was collected for ELISA-based analysis, and after 7 days the cells were fixed and stained for TRAP (Sigma-Aldrich, St. Louis, MO). For the pit formation assays, RAW 264.7 murine macrophages (1 × 103 cells/well) were seeded in 16-well plates of the BD Biocoat osteologic bone cell culture system (BD Biosciences, Ontario, Canada) in DMEM according to the manufacturer's instructions. The resorption pits were assessed according to the manufacturer's instructions. The data are expressed as the number of pits/field.
Statistical analysis.Quantitative data were analyzed using the statistical features of GraphPad Prism, version 4.0 (GraphPad Inc., San Diego, CA). A one-way analysis of variance (ANOVA) followed by a Kruskal-Wallis test and Dunn's posttest were used to determine significance, with P < 0.05 considered statistically significant.
RESULTS
Lyso-PAF acetyltransferase expression is increased after A. actinomycetemcomitans LPS injection in WT mice.The injection of LPS into periodontal tissues induces the production of inflammatory mediators (27, 28). In the present study, we used A. actinomycetemcomitans LPS to induce an experimental PD. Results showed a significant expression of lyso-PAF acetyltransferase 3 h after A. actinomycetemcomitans LPS injection in the periodontal tissues of WT mice (Fig. 1A). It has already been reported that injection of A. actinomycetemcomitans LPS into periodontal tissues induces alveolar bone loss in mice (29); thus, this study assessed whether the absence of PAFR could influence alveolar bone loss after A. actinomycetemcomitans LPS injection. Injection of A. actinomycetemcomitans LPS resulted in significant alveolar bone loss in WT mice. In contrast, no significant alveolar bone loss was observed in Pafr−/− mice (Fig. 1B).
Lyso-PAF acetyltransferase expression and alveolar bone loss induced by A. actinomycetemcomitans LPS. (A) Levels of lyso-PAF acetyltransferase in maxilla samples of WT mice after injection of A. actinomycetemcomitans LPS. (B) Quantification of alveolar bone loss in WT and Pafr−/− mice after injection of A. actinomycetemcomitans LPS, performed through the measurement of the CEJ-ABC area in the palatal face of the first molar using Image J software. The values (means ± SEM) were obtained from five animals at each point in two independent experiments. *, P < 0.05 versus the control side by one-way ANOVA with Dunn's posttest.
The absence of PAF receptor impairs alveolar bone loss induced by A. actinomycetemcomitans infection in mice.To evaluate whether PAF was involved in A. actinomycetemcomitans-induced bone loss, WT and Pafr−/− mice were infected with A. actinomycetemcomitans. At days 30 and 60 after initial infection, the mice were euthanized and alveolar bone loss was measured. Analysis of maxilla samples at day 60 after A. actinomycetemcomitans initial infection indicated the presence of alveolar bone loss in WT mice (Fig. 2A) but not in Pafr−/− mice (Fig. 2B). At the same time point, maxillary histological sections from noninfected WT mice (Fig. 2C) showed a distance between CEJ and ABC similar to that observed in noninfected Pafr−/− mice (Fig. 2D). However, histological sections from infected WT mice (Fig. 2E) showed a greater distance between CEJ and ABC, which was not observed in infected Pafr−/− mice (Fig. 2F). These findings were confirmed by the analysis of the area between CEJ and ABC in the maxillary samples. We observed that significant alveolar bone loss started at day 30 and presented a greater extent of bone loss at 60 days after initial infection in WT mice, while alveolar bone loss was not observed in the absence of PAFR (Fig. 2G). Moreover, the number of TRAP-positive cells was also significantly higher in WT mice than Pafr−/− mice at day 60 of initial infection (Fig. 2H).
Experimental periodontal disease in WT and Pafr−/− mice. Maxilla palatal aspects of alveolar bone loss at 60 days postinfection in WT mice (A) and Pafr−/− mice (B). White bars, 1 mm. Histological sections of periodontal tissues in noninfected WT (C) and Pafr−/− (D) mice and infected WT (E) and Pafr−/− (F) mice. Black bars indicate the distance between CEJ and ABC. Scale, 50 μm. (G) Alveolar bone loss quantification was performed through measurements of the CEJ-ABC area in the palatal face of the first molar using ImageJ software. (H) The number of TRAP-positive cells associated with periodontal bone. The values (means ± SEM) were obtained from five animals at each point in two independent experiments. P < 0.05 (* and #) and P < 0.01 (**) versus the control by one-way ANOVA with Dunn's posttest. ns, not significant.
The absence of PAF receptor does not modify bacterial load, neutrophil influx, and production of cytokines in periodontal tissues after infection.The bacterial load recovery was significant in WT mice at days 3 and 60 after initial infection, and a similar profile was observed in Pafr−/− mice (Fig. 3A). To evaluate neutrophil influx, mice submitted to oral inoculation of A. actinomycetemcomitans were euthanized at days 30 or 60 after initial infection, and periodontal tissues were processed for MPO quantification. Neutrophil influx was significantly increased at day 60 after initial infection in WT mice, while a significant neutrophil influx was observed in Pafr−/− mice at days 30 and 60 after initial infection (Fig. 3B). Additionally, there was a significant production of TNF-α at day 30 after initial infection in WT and Pafr−/− mice, although no difference was observed at day 60 in both groups (Fig. 3C). The production of the chemokine CXCL-1 was increased at day 60, as observed in WT and Pafr−/− mice (Fig. 3D); this increase was concomitant with neutrophil influx. Thus, although there was significant neutrophil influx at day 30 after initial infection in the absence of PAFR, the production of TNF-α and CXCL-1 in periodontal tissues after A. actinomycetemcomitans initial infection showed a similar profile in WT and Pafr−/− mice.
Bacterial load, neutrophil influx, and production of cytokines in periodontal tissues. (A) Bacterial load in maxilla samples from WT and Pafr−/− mice after 3 and 60 days of infection. P < 0.05 (*) and P < 0.001 (**) by one-way ANOVA with Dunn's posttest (n = 4). (B) Concentrations of myeloperoxidase in periodontal tissues of WT and Pafr−/− mice. Amounts of TNF-α (C) and CXCL-1 (D) in maxilla samples of WT and Pafr−/− mice. P < 0.05 (*) and P < 0.01 (**) versus the control by one-way ANOVA with Dunn's posttest (n = 5).
PAF and PAFR are important for osteoclast activity.To clarify how PAF and PAF receptor impair alveolar bone loss, RAW 264.7 murine macrophages were cultured in the presence of RANKL to induce differentiation into osteoclasts. These osteoclast-like cells were then cultured with A. actinomycetemcomitans LPS in the presence of the PAF receptor antagonist UK74505. As demonstrated in this study, a significant production of TNF-α was detected in WT and Pafr−/− mice during A. actinomycetemcomitans infection. Accordingly, significant production of TNF-α was observed 24 h after A. actinomycetemcomitans LPS stimulation of osteoclast-like cells, even with the blockade of the PAF receptor by UK74505 (Fig. 4A). A. actinomycetemcomitans LPS stimulation also resulted in an increased number of TRAP-positive cells (Fig. 4B). In contrast, in the presence of UK74505, significantly fewer TRAP-positive cells were observed (Fig. 4B). Furthermore, the blockade of PAF receptor by UK74505 impaired the resorptive activity of A. actinomycetemcomitans LPS-activated osteoclasts (Fig. 4C). The addition of PAF increased the number of TRAP-positive cells in the presence of RANKL (Fig. 4D). No modification in the percentage of cell viability was seen in our in vitro conditions (control, 91.5% ± 3.5%; LPS, 92.8% ± 3.2%; UK74505, 93.6% ± 0.85%; UK74505 plus LPS, 93.6% ± 4.3%; PAF, 88.5% ± 0.4%; PAF plus LPS, 84.3% ± 4.9%; means ± standard errors of the means [SEM]).
Involvement of PAF in osteoclast activity. (A) Quantification of TNF-α in osteoclast media detected by ELISA 5 h after A. actinomycetemcomitans LPS stimulation (n = 5). (B) Number of TRAP-positive cells 6 days after UK74505 treatment and A. actinomycetemcomitans LPS stimulation (n = 3). (C) Number of resorption pits/field 10 days after A. actinomycetemcomitans LPS stimulation (n = 3). *, P < 0.05 versus the control; #, P < 0.05 versus LPS by one-way ANOVA with Dunn's posttest. (D) Number of TRAP-positive cells 6 days after PAF and/or A. actinomycetemcomitans LPS stimulation (n = 3). P < 0.05 (*) versus the control and P < 0.05 (#) versus RANKL by one-way ANOVA with Dunn's posttest. The values (means ± SEM) of all in vitro experiments were obtained from two independent experiments.
DISCUSSION
Periodontal disease (PD) is a chronic infectious inflammatory disease, and its progression is related to the inflammatory response triggered by the breakdown of microbial homeostasis. One of the most important Gram-negative bacteria involved in the initiation and progression of aggressive PD is Aggregatibacter actinomycetemcomitans, and one of its major virulence factors is LPS (1). It has been suggested that A. actinomycetemcomitans LPS has more potent bone resorption activity than Porphyromonas gingivalis LPS (27). In this study, A. actinomycetemcomitans LPS was chosen because of the direct correlation with the experimental PD induced by A. actinomycetemcomitans infection. LPS is a potent stimulator of the innate immune response. It activates leukocytes to produce inflammatory mediators involved in the destruction of soft and hard tissues (30). The platelet-activating factor (PAF) is among these mediators. It has been shown that A. actinomycetemcomitans invades endothelial cells via a mechanism dependent on the PAF receptor (31). Although the increase of PAF in gingival crevicular fluid has been positively correlated with periodontal status (4) and has also been implicated in inflammatory bone disorders (19, 32), the specific role of PAF in the pathogenesis of PD is not known. In the current study, there was a significant increase of lyso-PAF-acetyltransferase expression in the periodontal environment in vivo in response to A. actinomycetemcomitans LPS. One study indicates that the Toll-like receptor 4 (TLR4)/MyD88-dependent pathway is responsible for LPS-induced lyso-PAF-acetyltransferase activation (33). Accordingly, we have previously shown that the signaling due to A. actinomycetemcomitans LPS recognition is MyD88 dependent (29). Further, a marked reduction was detected in the severity of experimental alveolar bone loss induced by A. actinomycetemcomitans LPS injection or oral inoculation of A. actinomycetemcomitans in the absence of the PAF receptor, suggesting an important role of PAF and its receptor in the pathogenesis of experimental PD.
Inflammatory mediators have been largely related to the progression of PD, which involves recruitment of leukocytes and production of bone-resorptive factors by inflammatory cells (34). Here, despite the reduced alveolar bone resorption observed in Pafr−/− mice subjected to A. actinomycetemcomitans oral inoculation, the bacterial load in periodontal tissues was not influenced by the absence of PAFR. Indeed, some previous reports have shown that the presence of a periodontopathogen, such as A. actinomycetemcomitans, in periodontal tissues is important but not sufficient to trigger periodontal breakdown (35, 36). Accordingly, the absence of another inflammatory mediator, macrophage migration inhibitory factor (MIF), impaired alveolar bone loss in mice without changing bacterial load (20).
PAF is synthesized by a variety of proinflammatory cells which participate in the development of inflammation, such as monocytes/macrophages, polymorphonuclear neutrophils (PMN), eosinophils, basophils, and platelets (37, 38). These cells are all targets of PAF effects, as they bear PAF receptors on their cell surfaces (39). Some studies have shown that PAF promotes PMN activation and adhesion to endothelial cells, favoring their recruitment (40, 41). Nevertheless, we observed increased levels of MPO, an indirect marker of the neutrophil influx, in infected WT and Pafr−/− mice compared to noninfected mice. Thus, we demonstrated in the present study that the absence of PAF receptor did not impair neutrophil influx, a key inflammatory cell in the control of periodontal bacteria (42). Accordingly, one study had suggested that PAF, by acting on its receptor, plays a minor role in the local production of chemokines and the recruitment of leukocytes during Klebsiella pneumoniae infection in mice (43). Indeed, we observed that the high levels of MPO were directly correlated with enhanced production of the neutrophil chemotactic factor CXCL-1 in WT and Pafr−/− mice.
Some studies have shown that the proinflammatory cytokine TNF-α induces the secretion of collagenase by fibroblasts and the resorption of cartilage and bone, and TNF-α has been implicated in the destruction of alveolar bone in periodontitis (reviewed in reference 44). Despite this evidence, we observed that, even in the absence of PAFR, the production of TNF-α was significant in periodontal tissues after A. actinomycetemcomitans infection, although Pafr−/− mice did not present significant alveolar bone loss. It has been demonstrated that TNF-α directly promotes osteoclast formation in vitro in the absence of RANKL (45, 46), but the administration of TNF-α does not induce osteoclast formation in Rank-deficient mice (47). In fact, in the in vitro conditions of this study, the blockade of PAFR by UK74505, after A. actinomycetemcomitans LPS stimulation, did not impair increased levels of TNF-α despite the decreased number of TRAP-positive cells. Therefore, these results suggested a direct action of PAF in osteoclasts, as shown previously (19).
A model for PAF action in osteoclasts was proposed previously (15). In the proposed model, increased levels of proinflammatory cytokines, such as TNF-α and interleukin-1β, seem to enhance PAF production by activating lyso-PAF acetyltransferase, thereby affecting osteoclast functions, including bone resorption activity. Moreover, it has been suggested that the response of osteoclasts to PAF is direct and does not involve an intermediate cell type, because PAF acts at specific receptor sites on osteoclasts (19). We could observe that, in vitro, addition of PAF enhanced osteoclast differentiation. Thus, based on the in vivo and in vitro results, we can also support the direct action of PAF in osteoclasts.
In conclusion, the present study showed that PAF, through its receptor, is fundamental for osteoclast activity without affecting the production of TNF-α, CXCL-1, and the neutrophil influx in periodontal tissues after A. actinomycetemcomitans infection.
ACKNOWLEDGMENTS
We thank Ozamu Fujise for providing us with the A. actinomycetemcomitans LPS.
This work was supported by CAPES, CNPq (INCT–Dengue program), and FAPEMIG.
FOOTNOTES
- Received 22 August 2013.
- Accepted 25 August 2013.
- Accepted manuscript posted online 3 September 2013.
- Copyright © 2013, American Society for Microbiology. All Rights Reserved.
