Introduction
Periodontitis is a chronic inflammatory disease that induces periodontal tissue destruction. Proteolytic enzymes, such as the matrix metalloproteinases (MMPs), play important roles in inflammatory diseases, including chronic periodontitis. Periodontal tissue is a complex and highly specialized structure consisting of both hard and soft tissues.
Gram-negative bacterial infection is the most frequent cause of periodontal diseases [1]. Lipopolysaccharide (LPS) is present on the outer membrane of gram-negative anaerobic microorganisms such as Porphyromonas gingivalis, which causes periodontitis [2]. LPS stimulates the secretion of inflammatory mediators, such as interleukin (IL)-1, prostaglandin E2, and nitric oxide (NO) [3,4]. These mediators play important roles in osteoclastogenesis and connective tissue destruction [5,6]. metalloproteinases (MMPs), which degrade extracellular matrix components, including collagens, fibronectin, and proteoglycan core proteins [8]. Excess formation of MMPs stimulates the decomposition of the extracellular matrix in the periodontium, which results in the loss of connective tissue adhesion and of supporting bone, which in turn causes tooth loss [8,9].
These proteolytic enzymes are classified into subgroups based on their substrate specificities. Collagenase (MMP-1, MMP-8, and MMP-13), gelatinase (MMP-2 and MMP-9), and stromelysin (MMP-3, MMP-10, and MMP-11) are the primary MMPs that participate in the degradation of matrix components [10,11]. In a study of patients with periodontitis, increased MMP-1 and MMP-2 levels were detected in gingival crevicular fluid, periodontal tissues, and periodontal ligament cells (PDLs) [12].
Based on available reports, we know that LPS stimulates macrophages, activates immune-related mediators such as NO and triggers activity of the signal system for mitogenactivated protein kinase (MAPK). It also plays an important role in the activation of nuclear factor kappa-B (NFκB), a protein belonging to the Rel protein group that regulates the transcription of target genes by the formation of dimers. Among in the NFκB transcription factors, RelA/p50 dimers are the most common form. When the signal is transmitted to the cell, inhibitory kappa B (IκB) is phosphorylated by IκB kinase (IKK) and subsequently degrades, and the NFκB that is separated from the IκB regulates transcription. RelA is also capable of phosphorylation, and it is known that RelA phosphorylation displays a regulatory effect on transcription activity and protein stability [13,14].
The periodontal ligament is a cellular connective tissue that links the cementum to the surrounding bone. Periodontal ligament fibroblasts (PDLFs), the primary cell type in the periodontal ligament, act as immune-responsive and supportive cells for the periodontal tissues. The location of these cells in periodontal pathological conditions suggests that they play key roles in regulating the inflammatory response and in amplifying inflammatory signals [15].
Osteoclasts are bone-resorbing cells derived from the monocyte/macrophage lineage. Two major cytokines— macrophage colony-stimulating factor (M-CSF) [16] and receptor activator of nuclear factor kappa-B (RANK) ligand (RANKL)—are essential for osteoclast formation [17]. RANKL binds to RANK and affects adaptor molecules, such as tumor necrosis factor receptor–associated factor (TRAF) 6. This binding induces several downstream signaling pathways, including Jun N-terminal kinase (JNK), extracellular signal– regulated kinase (ERK), p38, NFκB, and phosphatidylinositol 3-kinase/Akt [18,19]. Osteoclasts have important physiological functions in bone resorption and pathological functions in diseases that lead to bone-destroying inflammatory diseases [20].
Natural compounds, such as flavonoids, are found in many plants, including edible plants [21,22]. Myricetin, a hexahydroxyflavone, is a flavonoid usually found in grape seeds, tea, fruits, and medicinal plants [23] that has antiproliferative, antioxidant, and anti-inflammatory effects [24]. Although myricetin is known to exert its antiinflammatory effect by decreasing IL-1 transcription and NO secretion in LPS-stimulated monocytes [25,26], the effect of modulating periodontal inflammation via suppression of MMP production and activity, as well as the basic mechanisms underlying these activities, remains largely unknown.
In this study, we attempted to verify the effects of myricetin on the expression and activity of MMPs and their signaling mechanisms by inducing an inflammatory response in human PDLFs and in mouse mononuclear cells (RAW264.7 cells).
Materials and Methods
Reagents
LPS from Escherichia coli was purchased from Sigma- Aldrich (St. Louis, MO, USA). Myricetin (Sigma-Aldrich) was dissolved in dimethyl sulfoxide (DMSO) (1,000×). Antibodies against MMP-1, MMP-2, JNK, p38, ERK, IκBα, IKK, and actin were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA), and antibodies against MMP-8 were purchased from Abcam (Cambridge, UK). The antibodies against phospho-JNK, phospho-p38, phospho-ERK, phospho- IKK, p65RelA, phospho-p65RelA, AKT, and phospho-AKT were purchased from Cell Signaling (Danvers, MA, USA). Customized primers for RT-PCR were obtained from Bioneer (Daejeon, Korea).
Cell cultures
Human periodontal ligament tissue was harvested from healthy teeth extracted from three individuals during orthodontic treatments. The Ethics Committee of the School of Dentistry at Dankook University in Korea approved all protocols, and informed consent was obtained prior to the extractions (IRB No. H-1009/006/002). Periodontal ligament (PDL) tissue was carefully removed from the middle third of the root surface with a surgical blade and was placed in a culture dish. The tissues samples were grown in Dulbecco’s modified Eagle’s medium (DMEM) with 10% fetal bovine serum (FBS) and antibiotics (10,000 U/mL of penicillin G and 10,000 μg/mL of streptomycin) (HyClone, South Logan, Utah, USA). After the explanted cells from the tissue reached confluence, the cells were detached by means of 0.25% trypsin and 0.2% EDTA and were subcultured at a ratio of 1:5. Cell cultures between the 5th and 9th passages were used.
Mouse monocyte/macrophage RAW264.7 cells were obtained from American Type Culture Collection (ATCC) (Manassas, VA, USA) and were cultured and maintained at 37°C in a 5% CO2 atmosphere in DMEM with 10% FBS. The medium was changed every 3 days.
Cell viability assays
The cells were seeded in 96-well plates at a density of 104cells/well in DMEM with 10% FBS. After 24 hours, the cells were treated with1 μg/mL of LPS and serially diluted myricetin for 2 days for the PDLFs or 4 days for the RAW264.7 cells. Cell viability was tested with an MTT assay according to the manufacturer’s protocol (Sigma M5655) (Sigma-Aldrich). After the supernatant media were removed, the formazan granules were solubilized with isopropyl alcohol, and the absorbance was then measured using a microplate absorbance reader.
MMP activity assay
To test the effect of myricetin on MMP activity, the PDLFs were seeded in 96-well plates at a density of 104cells/well in DMEM with 10% FBS. After 24 hours, the cells were treated with 1 μg/mL of LPS and serially diluted myricetin for 24 hours. The levels of MMP-1, MMP-2, and MMP-8 activity in the cell culture supernatants were tested using the fluorometric MMP Assay Kit (AnaSpec, Fremont, CA, USA) according to the manufacturer’s protocol. Fluorescence intensity was measured at excitation/emission rates of 340/490 nm using a multi-detection microplate reader (SpectraMax M2e) (Molecular Devices, Sunnyvale, CA, USA).
Induction of multinucleated osteoclasts and tartrateresistant acid phosphatase–positive [TRAP(+)] staining
To differentiate the RAW264.7 monocyte/macrophage cells into osteoclasts, the cells were pre-incubated with 100 ng/mL of RANKL (PeproTech, Rocky Hill, NJ, USA) for 24 hours and stimulated with 1 μg/mL of LPS for 4 or 7 days. After culturing, the cells were fixed with citrate-acetoneformaldehyde for 5 minutes, and a TRAP staining kit was used to TRAP expression. The TRAP(+) multinucleated cells were observed to be osteoclasts by means of an optical microscope.
TRAP activity assay
RAW264.7 cells were seeded in 96-well plates at a density of 104cells/well in α-MEM with 10% FBS. The cells were pre-treated with 100ng/mL of RANKL and cultured with 1 μ g/mL of LPS and serially diluted myricetin for 4 days. The cells were then harvested with a 0.1% Triton X-100 solution, and TRAP activity was examined using 50 mM of citrate solution with 100 mM of p-nitrophenyl phosphate as a substrate and 10 mM of sodium tartrate using spectrophotometry. The total protein content in the cell lysates was measured using Lowry protein assay reagents (Pierce Biotechnology, Rockford, IL, USA) with bicinchoninic acid (BCA) as the standard.
Western blot analysis
PDLFs or RAW264.7 cells were seeded in 6-well plates at a density of 5×105cells/well. After 24 hours, the cells were serum-starved for 6hours, pre-treated with serially diluted myricetin in media including 10% FBS for 2 hours, and then stimulated with 5 μg/mL of LPS for 15 minutes or 2 hours. Next, the cells were collected, and protein homogenates were prepared using a lysis solution. Proteinlysates (30 μg/lane) were electrophoresed using SDS–PAGE and were transferred to polyvinylidenefluoride (PVDF) membranes. The membranes were blocked with 5% bovine serum albumin in Tris-buffered saline and Tween20 (TBST) and probed with antibodies against MMP-1, MMP-2, MMP-8, phospho-ERK1/2 (Thr202/Tyr204), phospho-SAPK/JNK (Thr183/Tyr185), phospho-p38, phospho- IKK, phospho-p65RelA, phospho-AKT, and IκBα. The blots were then re-probed with the corresponding non–phosphoantibodies for MAPK, IKK, p65RelA, AKT, and IκBα. The membranes were developed using X-ray film (KOREA FUJI Photo Film, Seoul, Korea) exposure and a chemiluminescence system (Amersham Biosciences, Buckinghamshire, UK).
Interleukin-6 (IL-6) ELISA assay
To test the effect of myricetin on IL-6 release, RAW264.7 cells were seeded in 96-well plates at 104cells/well in α-MEM with 10% FBS. After 1day, the cells were incubated with1 μg/mL of LPS and serially diluted myricetin for 2 days. The amount of IL-6 released into the culture supernatants was determined using ELISA kits (R&D Systems, Minneapolis, MN, USA) according to the manufacturer’s protocols. The absorbance was detected at 450 nm using a spectrophotometer.
Nitrite assay
NO secretion was evaluated by measuring the levels of nitrite, a stable oxidized form of NO, in the culture supernatants. To test the effect of myricetin on inducible nitric oxide synthase (iNOS) activity, RAW264.7 cells were seeded in 96-well plates at 104cells/well in α-MEM with 10% FBS for 24 hours. Next, the cells were cultured with1 μg/mL of LPS and serially diluted myricetin in phenol red–free media for 72 hours. After incubation, 100 μL of culture supernatant was reacted with 100 μL of Griess reagent (Sigma-Aldrich) in a 96-well reading plate, and the absorbance was detected at 550 nm using a spectrophotometer. Nitrite levels measured were compared with a sodium nitrite standard.
Statistical analysis
Significance was determined using Student’s t-test. All tests were performed in triplicate. Each value represents the mean ± standard error (SE). Differences were considered statistically significant at p<0.05 and p<0.01.
Results
To examine the protective effects of myricetin on periodontal tissue using LPS-stimulated human PDLFs, we determined the expression and enzyme activity of MMPs and the phosphorylation of MAPK and activation of NFκB-related molecules. In addition, we investigated myricetin’s activity in LPS-stimulated murine RAW264.7 cells under inflammatory conditions, including its effects on the expression and enzyme activity of MMPs, osteoclast generation, and the phosphorylation of MAPK and IL-6 and NO release.
Studies in PDLFs
Myricetin inhibits LPS-stimulated expression of MMPs in PDLFs
Myricetin did not cause any significant changes on viability of the PDLFs (Fig. 1A). We used western blot analysis to examine MMP-1, MMP-2, and MMP-8 protein production. PDLFs were pre-treated with myricetin for 2 hours and activated with LPS for 2 hours. Myricetin inhibited protein production by MMP-1, MMP-2, and MMP-8 (Fig. 1B).
Myricetin inhibits LPS-activated enzyme activity of MMPs in PDLF
To test the inhibitory effect of myricetin on extracellular matrix destruction, the activities of MMP-1, MMP-2, and MMP-8 were assessed in the culture supernatants from the LPS-stimulated PDLFs using a fluorescent assay. The enzyme activities of MMP-1, MMP-2, and MMP-8 were activated by LPS in the PDLFs. Myricetin decreased the MMP-2 and MMP-8 activities at concentrations of 2 or 10 μM, as compared with the LPS-induced control (Fig. 1C).
Myricetin inhibits phosphorylation of MAPK in LPS-stimulated PDLFs
To determine the kind of effects myricetin has on the signal transduction pathway in PDLFs, we first verified the effects on activation of MAPKs by means of western blot analysis. In this way, PDLFs were pre-treated with myricetin for 2 hours and then activated with LPS for 15 minutes. We measured the degree of protein expression for MAPK. Although phosphorylation of JNK and p38 was observed following LPS treatment, no marked activation of ERK was observed. Myricetin suppressed JNK and p38 phosphorylation (Fig. 2).
Myricetin inhibits activation of NFkB-related molecules in LPS-stimulated PDLFs
Since it has been reported that NFκB can act as an important transcription factor in the expression of inflammatory cytokines by LPS, we investigated whether or not myricetin also has an effect on the activation of NFκB in PDLFs. We pre-treated PDLFs with myricetin for 2 hours and then stimulated them with LPS. Subsequently, we prepared protein samples and observed changes in the phosphorylation of IKK. When PDLFs were treated with LPS, an increase in the phosphorylation of IKK was observed after 10 minutes and quantitative changes were observed up to 30 minutes after treatment. In all cases after treatment with LPS and myricetin, the phosphorylation of IKK was suppressed within 15 minutes (Fig. 3). The phosphorylation reaction was observed in IKK, IκBα, p65RelA, and AKT over a 15-minute period following treatment with LPS. As a result, there was an increase in the phosphorylation of IKK, p65RelA, and AKT in the LPS-treated cells; in the case of myricetin treatment, the increased phosphorylation was suppressed (Fig. 4).
Studies in RAW264.7 cells
Myricetin inhibits osteoclast generation in RAW264.7 cells
Pre-treatment with RANKL and LPS significantly induced osteoclast generation in RAW264.7 cell culture on day 7. Myricetin at concentrations of 0.2, 1, and 5 μM suppressed the formation of RANKL-pre-treated, LPS-induced multinucleated osteoclasts (Fig. 5A and 5B). TRAP activity was measured in the cell lysates from LPS-stimulated RAW264.7 cells using spectrophotometry. Myricetin inhibited TRAP activity at concentrations of 1 and 5 μM, as compared with the LPS-induced control (Fig. 5C).
Myricetin inhibits LPS-activated expression and enzyme activity of MMPs in RAW264.7 cells
Myricetin decreased cell viability at a concentration of 5 μ M in RAW264.7 cells (Fig. 6A). The protein expression of the MMP-8 gene was assessed with western blot analysis. The Protein expression of MMP-8 was stimulated with LPS. Myricetin inhibited the induction of MMP-8 expression in RAW264.7 cells (Fig. 6B). MMP-8 enzyme activity was measured in the culture supernatants from the LPS-stimulated RAW264.7 cells using a fluorescent assay. Myricetin inhibited the MMP-8 activity at 2 and 10 μM, as compared with the LPS-induced control (Fig. 6C).
Myricetin does not suppress phosphorylation of MAPK in LPS-stimulated RAW264.7 cells
The basic mechanisms of action of myricetin during osteoclast generation were evaluated by assessing myricetin’s effect on the phosphorylation of MAPK. RAW264.7 cells were pre-treated with myricetin for 2 hours and then induced with LPS for 15 minutes. The activation of MAPK components JNK, ERK1/2, and p38 was assessed by western blot analysis. LPS induced the phosphorylation of JNK, ERK1/2, and p38 within 15 minutes. LPS-induced phosphorylation of JNK, ERK1/2, and p38 was not changed by myricetin. In addition, LPS induced the degradation of IκBα, whereas myricetin induced a slight recovery from the LPS-activated degradation of IκBα (Fig. 7).
Myricetin suppresses IL-6 and NO release in LPS-stimulated RAW264.7 cells
To determine whether myricetin was capable of reducing the expression of inflammatory cytokines that was increased when RAW264.7 cells were stimulated with LPS, we treated RAW264.7 cells with LPS and compared the expression levels of IL-6 using the ELISA method and measured NO with the Griess reagent assay. The amounts of secreted IL-6 and NO were assessed by ELISA and Griess reagent assay, respectively. Levels of IL-6 and NO secretion were evaluated to determine whether myricetin inhibited the LPS-stimulated release of these inflammatory cytokines. LPS increased IL-6 and NO production in RAW264.7 cells, whereas myricetin significantly decreased LPS-stimulated IL-6 and NO production (Fig. 8A and 8B).
Discussion
In this study, we assessed the effects of the flavonoid myricetin on the regulation of LPS-activated MMP expression and activity in human PDLFs. We also examined the basic mechanism of action of myricetin in LPS-activated osteoclast generation in murine RAW264.7 cells.
Myricetin is present in many plants and medicinal herbal extracts [23]. Although many studies have shown the efficacy of myricetin, such as its anti-inflammatory effects, in certain conditions, its targets in periodontal tissues have not been determined.
MMP-1 and MMP-2 transcription levels have been correlated with destructive gingival tissue in patients with chronic periodontitis [27,28], and the amounts of MMP-1 and MMP-2 proteins in gingival crevicular fluid are higher in patients with periodontitis than in periodontally healthy subjects [29]. Accordingly, the levels of MMP-1 and MMP-2 in PDLFs may also be related to the grade of periodontal tissue destruction. Our results showed that myricetin significantly decreased MMP-1 expression and MMP-2 activity in PDLFs when compared with control cells. Therefore, we hypothesized that myricetin may alleviate periodontal disease in gingival tissue.
Previously, MMP-8 was believed to be produced only by polymorphonuclear leukocytes [30]. However, it was more recently reported that MMP-8 is also produced in cells of non-polymorphonuclear lineage, such as gingival crevicular epithelial cells, gingival fibroblasts, and periodontal ligament (PDL) cells [31~33]. Guan et al. [34] reported that infection with Prevotella intermedia stimulates MMP-8 expression and activates this protein in PDL cells. In patients with chronic periodontitis, the amount of MMP-8 in the gingival crevicular fluid, gingival tissue, and PDL cells was significantly increased. The amount of MMP-8 was significantly inhibited after periodontal surgical treatment, which suggests that MMP-8 is involved in periodontal tissue degradation [35]. In our study, myricetin inhibited MMP-8 expression and activity in PDLFs, which suggests that myricetin may suppress connective tissue degradation by repressing MMP-8 activity as well as mRNA and protein expression. In addition, myricetin decreased MMP-8 expression and activity in the RAW264.7 cells. Therefore, inhibition of MMP-8 activity may help manage the progression of periodontal disease.
These results are consistent with an earlier report showing that myricetin decreases LPS-induced MMP-2 and MMP-8 in human gingival fibroblasts (HGFs) [36]. However, a present study showed that myricetin is less effective in decreasing MMP activities in PDLFs, as compared with HGFs. In our report we used western blot analysis to confirm the actions of myricetin on the MMP changes in PDLFs.
Since we were able to verify that myricetin reduced the expression and activity of MMP-1, MMP-2, and MMP-8 in PDLFs, we also investigated its effects on intracellular signal transduction in order to determine what kind of molecular mechanism was responsible for these effects. In observing the changes in the activity of MAPKs in response to LPS, we found that LPS increased the activity of JNK, and these effects were suppressed by myricetin.
In addition, myricetin’s effects were observed with respect to the NFκB-related signal transduction system. The regulation of phosphorylation of RelA is also known to play an important role in activating the NFκB pathway [37]. It is known that kinases such as casein kinase II, AKT, and IKK are capable of inducing the phosphorylation of RelA [37]. In our study as well, RelA phosphorylation was increased by treatment with LPS, and increases in the phosphorylation of IKK and AKT were also observed. In cases where the PDLFs were treated with myricetin, this increase in phosphorylation was suppressed. From these results, it is believed that the suppression of RelA phosphorylation and the resulting decrease in DNA-binding capacity also act as important mechanisms in the decrease in NFκB transcription activity in response to myricetin.
We also tested the effects of myricetin on osteoclast generation in murine RAW264.7 cells. Myricetin suppressed the generation of osteoclast-like cells, and there was no evidence of cytotoxic changes in the RAW264.7 cells after treatment with myricetin at concentrations between 0.2 and 1 μM.
During inflammation, MAPK components participate in the signaling pathways that regulate the synthesis and secretion of pro-inflammatory factors from activated macrophages [38]. In this study, myricetin did not affect the LPS-activated phosphorylation of MAPK proteins such as JNK, ERK and p38. RANKL induces the activation of NFκB via TRAF6 signaling [39]. Activation of NFκB requires the breakdown of IκBα, an inhibitory protein that complexes with NFκB dimers [40]. In this study, myricetin led to recovery of the slightly degraded IκBα.
PDLFs are involved not only in the differentiation of osteoblasts and the formation of bone, but they are also known to be involved in the inflammatory response when periodontal tissue is exposed to bacteria, during which cytokine secretion is promoted [41]. Accordingly, after treatment with myricetin, we observed a reduction in the secretion of cytokines from PDLFs. IL-6 is an activator of bone destruction, and excess production of this cytokine is involved in periodontal tissue destruction [42]. In this study, myricetin suppressed LPS-activated IL-6 release in monocytes.
In addition, inflammatory factors, such as NO, are involved in bone loss [43]. NO is produced by the salivary glands in the mouth and is secreted into the saliva. The microorganisms that induce inflammatory infections are known to further promote the production of NO by oral tissues by stimulating immune cells such as macrophages. Recently, it has been demonstrated that NO is a target for the development of therapies in patients with periodontal disease because NO plays an important role in the tissue degradation associated with such diseases [44]. In the current study, myricetin decreased LPS-activated NO secretion in RAW264.7 cells.
In conclusion, our results showed that myricetin exerts suppressive effects on periodontal tissue degradation and osteoclastogenesis and, in this way, plays a role in the inhibition of NFκB-related protein. Therefore, inhibition of periodontal tissue destruction may be a plausible therapeutic strategy for the management of patients with inflammatory periodontal disease. However, we believe it will be necessary to study animal models of periodontitis in order to verify whether myricetin can actually be of benefit in the clinical setting.
