点击显示 收起
【摘要】
Receptor activator of nuclear factor-B (RANKL)-mediated osteoclastogenesis plays a pivotal role in inflammatory bone resorption. The aim of this study was to identify the cellular source of RANKL in the bone resorptive lesions of periodontal disease. The concentrations of soluble RANKL, but not its decoy receptor osteoprotegerin, measured in diseased tissue homogenates were significantly higher in diseased gingival tissues than in healthy tissues. Double-color confocal microscopic analyses demonstrated less than 20% of both B cells and T cells expressing RANKL in healthy gingival tissues. By contrast, in the abundant mononuclear cells composed of 45% T cells, 50% B cells, and 5% monocytes in diseased gingival tissues, more than 50 and 90% of T cells and B cells, respectively, expressed RANKL. RANKL production by nonlymphoid cells was not distinctly identified. Lymphocytes isolated from gingival tissues of patients induced differentiation of mature osteoclast cells in a RANKL-dependent manner in vitro. However, similarly isolated peripheral blood B and T cells did not induce osteoclast differentiation, unless they were activated in vitro to express RANKL; emphasizing the osteoclastogenic potential of activated RANKL-expressing lymphocytes in periodontal disease tissue. These results suggest that activated T and B cells can be the cellular source of RANKL for bone resorption in periodontal diseased gingival tissue.
--------------------------------------------------------------------------------
Periodontal disease (or periodontitis) is an inflammatory lesion that is accompanied by soft tissue destruction and bone resorption in the tooth-supporting structures. A positive correlation between the occurrence of disease and elevated serum antibody response to the oral bacteria colonizing the gingival crevice1-3 suggests the involvement of an immune response to the multiple bacteria in the onset and development of periodontal disease. In general, immune responses to bacteria are considered to be a host protective mechanism to pathogenic bacteria. However, despite the elevated IgG antibody response to certain disease-associated bacteria colonizing the periodontal crevice, inflammation and/or bone resorption proceed in the periodontitis lesions. The question is posed as to whether immune response to periodontal bacteria is protective, or otherwise pathogenic, in the context of periodontal disease.
The fundamental cytokine system that underlies bone resorption processes is dependent on the osteoclast differentiation, activation, and survival factor, receptor activator of nuclear factor-B (RANKL), and its soluble decoy receptor osteoprotegerin (OPG).4-6 Involvement of immune cells in the course of bone resorption has been demonstrated by the expression of RANKL on activated T cells.7 RANKL expressed by T cells, as well as by osteoblasts and bone marrow stromal cells, triggers signaling in osteoclast precursor cells that elicits the differentiation into their mature form.8 OPG is expressed ubiquitously by many types of cells and tissues, and it counterregulates the excessive bone loss by antagonizing the RANKL-binding to its receptor RANK.9 The paradigm of osteoclast differentiation regulation is based on the RANKL/OPG ratio expressed in the microenvironment surrounding osteoclast precursor cells.10
An active periodontal lesion is characterized by the prominent infiltration of B cells11,12 and T cells.13,14 Although the T cells infiltrating the inflamed gingival tissues express activation markers such as CD45RO15 or CD29,16 functional roles of these activated T cells are not completely clear. We demonstrated that adoptive transfer of RANKL+, antigen-specific T cells can induce bone loss in rat periodontal tissue that received local injection of the T-cell antigen.17,18 Teng and colleagues19 reported that adoptive transfer of an Actinobacillus actimomycetemcomitans-specific human T cell line isolated from patients with aggressive (juvenile) periodontal disease could induce significant periodontal bone loss in NOD/SCID mice orally inoculated with A. actinomycetemcomitans every 3 days.19 Although the latter report showed that systemic administration of OPG-Fc could reduce periodontal bone loss, it is not clear whether the transferred human T cells or bystander cells that might be secondarily stimulated by the transferred human T cells are the source of RANKL. It has been reported that RANKL produced by B cells is responsible for the devastating bone resorption in multiple myeloma.20 Activation of B cells in vitro can induce expression of RANKL, but these cells are deficient in production of OPG.21 Recently, we reported that antigen-specific activated B cells can induce periodontal bone resorption in a rat model.22 However, it is unclear if B cells accumulating in periodontal diseased tissue express RANKL.
To determine the cellular source of RANKL in bone resorptive periodontitis, enzyme-linked immunosorbent assay (ELISA) and double-color confocal microscopic analyses were used. Results of ELISA demonstrated that soluble RANKL (sRANKL) production was significantly elevated in gingival tissues with periodontal disease compared to healthy gingival tissues. Confocal microscopic analyses showed that both T cells and B cells, but not monocytes or fibroblasts, are the cellular source of RANKL in the bone resorptive lesion of periodontal disease. Importantly, RANKL expressed by periodontal T cells and B cells appeared to be the osteoclastogenic functional component, as determined by in vitro RANKL-dependent osteoclast differentiation assays.
【关键词】 lymphocytes resorptive periodontal
Materials and Methods
Patients
Patients diagnosed with chronic periodontitis .
Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR)
Total RNA was extracted from gingival tissues and RT-PCR was performed as previously described.23,24 Primer pairs for human RANKL and OPG were as follows: RANKL forward primer, 5'-TCAGAAGATGGCACTCACTG-3' and RANKL reverse primer, 5'-AACATCTCCCACTGGCTGTA-3' (PCR product size 879 bp),25 OPG forward primer, 5'-GCCCTGACCACTACTACACA-3' and OPG reverse primer, 5'-TCTGCTCCCACTTTCTTTCC-3' (PCR product size 736 bp). Total RNA isolated from gingivae (1 µg) was synthesized to cDNA. The resulting cDNA was subject to PCR by amplifying 30 cycles for RANKL or for OPG along with ß-actin23 as an internal control (94??C for 30 second, 60??C for 1 minute, 72??C for 1 minute, and final elongation at 72??C for 10 minutes). PCR products were separated in 1.7% agarose gels and stained with ethidium bromide.
Preparation of Gingival Tissue Homogenates
Gingival tissues were homogenized with a Dounce glass homogenizer in phosphate-buffered saline supplemented with 0.05% Tween 20, phenylmethyl sulfonyl fluoride (1 mmol/L; Sigma, St. Louis, MO), and protease inhibitor cocktail (Sigma), as published with slight modification.26 For ELISA, the concentration of sRANKL, interleukin (IL)-1ß, and OPG in the tissue homogenates or culture supernatants were measured with ELISA kits for human soluble RANKL (sRANKL) (Peprotech, Rocky Hill, NJ) and DuoSet ELISA for human IL-1ß, OPG, IL-10, IL-12 p70, and GM-CSF (R&D Systems, Minneapolis, MN). The method for detection of IgG response to oral bacteria has been previously published.27,28 To detect anti-bacterial IgG levels in the gingival homogenates, the following formalin-fixed bacteria were used as antigen in ELISA: A. actinomycetemcomitans strain Y4 (American Type Culture Collection no. 43718), Fusobacterium nucleatum str. 25586, Eikenella corrodens str. 23834, Porphyromonas gingivalis str. W83, and Prevotella intermedia str. 25611.
Culture of Peripheral Blood Mononuclear Cells (PBMCs)
PBMCs were collected from healthy patients under informed consent agreement (n = 4; 28- to 40-year-old males; average age, 34.5 years). The mononuclear cell (MC) fraction was separated from blood by gradient centrifugation using Histopaque-1077 (Sigma) and incubated in RPMI 1640 supplemented with 10% fetal bovine serum, 2-mercaptoethanol, L-glutamine, and penicillin and streptomycin. T cells were enriched from PBMCs by glass-wool and nylon-wool column purification.24 B cells were enriched from PBMCs by negative selection using magnetic beads coated with a mixture of anti-CD3 monoclonal antibody (mAb) (OKT3: American Type Culture Collection, Rockville, MD; UCTH1: R&D Systems), CD4 (OKT4), CD8 (OKT8), and CD28 (CD28.2; BD Pharmingen, San Diego, CA). The enriched T cells or B cells were stimulated with immobilized antibodies on a 96-well culture plate with mouse mAb to CD3 (OKT3) and CD28 (CD28.2) or mAb to CD40 (5C3, BD Pharmingen) and goat polyclonal anti-human IgM (Serotec, Oxford, UK), respectively.
Immunofluorescent Laser-Scanning Confocal Microscopy
Basic fluorescent staining technique was performed as previously reported.24 Sectioned tissues mounted on glass slides were fixed with 2% paraformaldehyde. T cells, B cells, and monocytes were stained with the mouse monoclonal antibodies to CD3 (UCTH1), CD20 (H1; BD Pharmingen), and CD14 (3C10; American Type Culture Collection), respectively, followed by fluorescein isothiocyanate-conjugated anti-mouse IgG (Jackson Immunoresearch Laboratories, West Grove, PA) as a secondary reagent. RANKL expression was monitored by biotinylated-OPG-Fc followed by Texas Red-Avidin (Invitrogen, Carlsbad, CA). To block FcRn and FcRII, which can bind to monomeric chain of human IgG-Fc, human IgG was prereacted with some of the gingival tissue sections. OPG-Fc was generously provided by Dr. Colin Dunston (Amgen, Thousand Oaks, CA) under mutual material transfer agreement and was conjugated with biotin using EZ-Link sulfo-NHS-biotin (Pierce, Rockford, IL). A human IgG-Fc fusion protein (L6) conjugated with biotin was used as a negative control. After washing, the cover glass was set on the sample with Fluoromount-G mounting medium (Southern Biotechnology, Birmingham, AL). The staining pattern was analyzed by 0.3 µm sequential optical sectioning at x400 or x1000 magnification with a Leica TCS/SP-2 laser-scanning confocal microscope (Leica, Wetzlar, Germany).
Osteoclast Differentiation Assay
MCs were isolated from patient gingival tissue biopsies as previously described.23 Patient gingival MCs, fresh PBMCs, stimulated PB T cells, or stimulated PB B cells in vitro were fixed with formalin. Human PB CD14+ monocytes were isolated from PBMCs using a magnetic bead-based monocyte-negative isolation kit (Dynal Biotech, Oslo, Norway). In some experiments, patient gingival MCs were incubated in 96-well plates for 3 hours, and nonadherent cells were separated from the adherent cells on the culture plate. The adherent gingival MCs were 80% CD14+ monocytes, whereas nonadherent gingival MCs contained less than 2% CD14+ cells by immunostaining. The resulting adherent cells in the wells and nonadherent gingival MCs were fixed with formalin and were used in the following co-culture system. The fixed human lymphocytes were co-cultured in 96-well plates with the mouse osteoclast precursor cell line MOCP-529 or PB CD14+ monocytes in the presence of M-CSF (10 ng/ml; Peprotech). Recombinant human RANKL (50 ng/ml; R&D Systems) was added to the MOCP-5 culture, as a positive control. To assess the involvement of RANKL in the osteoclast differentiation, OPG-Fc (10 µg/ml) was added to antagonize RANKL in some cultures. The culture medium was changed every 3 days by replacing half of the volume. After 6 to 8 days of culture for MOCP-5, or 14 to 16 days of culture for PB CD14+ monocytes, when multinuclear osteoclast-like cells are observed by phase-contrast microscopy, cells were fixed with 5% formalin-saline. Differentiated osteoclasts were identified as tartrate-resistant acid phosphatase (TRAP)-positive cells with three or more nuclei as described previously.30 The TRAP+ cells with more than three nuclei were counted as osteoclasts using phase contrast microscopy and expressed as cell number/well of 96-well plates. The bone resorption activity of osteoclasts was evaluated by a pit formation assay using a calcium phosphate-coated tissue culture vessel system (Biocoat Osteologic System; BD Biosciences, San Jose, CA) or dentin disks (Alpco Diagnostics, Windham, NH) according to the manufacturers?? instructions.
Results
Detection of sRANKL and OPG Proteins in Gingival Tissue Homogenates
The concentration of sRANKL protein in gingival tissue was significantly elevated (P < 0.02, t-test) in diseased tissues as compared to healthy gingival tissues (Figure 1A) . In addition, there was a positive correlation between the level of sRANKL in tissues and the depth of the gingival crevice where the tissues were sampled (Figure 1B) . Differences in OPG concentration between diseased and healthy tissues were not statistically significant (Figure 1C) . The proinflammatory cytokine IL-1ß was monitored in the same groups of gingival tissues (Figure 1D) , as an indicator of the degree of inflammation in each tissue. All gingival tissues from diseased lesions expressed significantly higher IL-1ß amounts than the healthy gingival tissues (Student??s t-test, P < 0.05). We also evaluated the concentrations of the osteoclastogenesis inhibitory cytokines, IL-10, IL-12 p70, and GM-CSF and the osteoclast precursor chemoattractant factor, MIP-1, compared to healthy tissues. However, although the diseased gingival tissues had higher mean concentrations of each of the four factors, none of the differences were statistically significant. Compared to healthy patients?? antibody levels, the tissue homogenates of patient gingival tissues also demonstrated significantly higher IgG antibody levels to three of five periodontal disease-associated bacteria examined in this study (F. nucleatum, P. gingivalis, and P. intermedia; data not shown).
Figure 1. Protein concentrations of RANKL, OPG, and IL-1ß in gingival tissues. Protein concentrations of RANKL and OPG in gingival tissues were measured by ELISA. Gingival tissue samples were homogenized in the presence of proteinase inhibitors (healthy, n = 3; disease, n = 11). The concentration of each protein, RANKL (A and B), OPG (C), and IL-1ß (D), is expressed as pg per mg of gingival tissue (pg/mg). *Significantly higher than healthy control by Student??s t-test (P < 0.05). B: The concentrations of RANKL were positively correlated with the gingival pocket depth where the biopsy was sampled. **Positive correlation (n = 14, P < 0.01).
Expression of RANKL and OPG mRNA in Gingival Tissue
RT-PCR was performed to detect mRNA expression of RANKL and OPG in the gingival tissue RNA samples (Figure 2) . RANKL mRNA expression was detected from 67% of the diseased gingival tissues (four of six) examined, whereas no expression of RANKL mRNA was observed from the healthy gingival tissues tested (zero of four). OPG mRNA message was observed in tissue from periodontally diseased patients (two of six) and one healthy subject (one of four).
Figure 2. RANKL and OPG mRNA expression in gingival tissues. RT-PCR was performed to detect mRNA expression of RANKL and OPG in the total RNA extracted from whole gingival tissue samples. Expression of RANKL mRNA was detected in 67% of diseased periodontal tissues (solid bars, four of six), whereas no expression of RANKL mRNA was observed in the healthy gingival tissues tested (open bars, zero of four), whereas OPG mRNA message was observed from healthy patients (one of four) and from periodontal disease tissues (two of six). The relative intensity of each mRNA message in the acrylamide gel was scanned by densitometer using the AlphaImage analysis system and expressed as relative signal intensity (RSI) in the histograms (B, RANKL; C, OPG).
RANKL Expression by T and B Lymphocytes in Gingival Tissue
To identify the cell types that express RANKL in inflamed gingival tissue, double-color confocal microscopy (Figure 3) was used after staining for RANKL (red) and lymphocyte-specific CD makers (green). Images of RANKL and CD marker-positive cells were merged in the computer system, and double-positive cells were displayed as yellow staining. Gingival tissues from periodontitis lesions demonstrated marked expression of RANKL (Figure 3, ACC) . The cellular infiltrates in the diseased gingival tissue were predominantly T (Figure 3A) and B lymphocytes (Figure 3B) , with few CD14+ monocytes (Figure 3C) . RANKL was expressed in these cellular infiltrates, especially by T cells and by B cells, and to a much lesser extent by monocytes. Although most B cells were positive for RANKL staining, not all T cells expressed RANKL. Very few lymphocytes expressing RANKL were present in healthy gingival tissues (Figure 3D) . RANKL expression by other nonlymphoid cell types, such as fibroblasts, was not distinctly observed.
Figure 3. Confocal microscopic analyses of expression of RANKL in gingival tissues. Using double-color confocal microscopy, RANKL-expressing cells were identified in the gingival tissue of a periodontal disease patient (ACC) and in healthy gingival tissue (D). CD3 (A and D), CD20 (B), or CD14 (C) expression was indicated by staining with various specific mouse mAbs followed by fluorescein isothiocyanate anti-mouse IgG (left column). RANKL expression was detected by OPG-biotin followed by streptavidin-Texas Red (middle column). Both specific CD marker-stained cells and RANKL+-stained cells were doubly exposed and expressed as yellow (merged; right column). CD20+ and CD14+ cells in healthy tissue (not shown) were as few as CD3+ cells in healthy gingival tissue (D). Scale bars = 20 µm.
The numbers of CD3+ and CD20+ lymphocytes infiltrating the gingival tissues were significantly higher than those in the healthy gingival tissues, whereas the number of CD14+ cells did not show a remarkable increase (Figure 4) . Importantly, the percentage of CD14+ cells to the number of total lymphocytes remained low irrespective of the gingival condition of disease/healthy (19% in healthy versus 5% in disease gingival tissue; Figure 4 ). These data strongly indicate that T cells and B cells are a much more significant source of RANKL than CD14+ cells in the diseased gingival tissues.
Figure 4. Numbers of lymphocytes in the gingival tissues. The total number of CD3+, CD20+, and CD14+ cells in a confocal microscopic field of each tissue section was counted (x400, average cell number of at least three sections was enumerated as the number of cells per sample) and compared between healthy gingival tissues (open column) and disease gingival tissues (filled column). Mean cell number ?? SD is shown. *Significantly higher than healthy gingival tissue by Student??s t-test (P < 0.01).
The percentage of RANKL+ cells in the population of T cells or B cells was also determined (Figure 5) based on counts of the images obtained in the confocal microscopic analyses (Figure 3) . The percentage of RANKL-expressing T cells (Figure 5A) or B cells (Figure 5B) demonstrated a positive correlation with the depth of the gingival crevice. Importantly, in the diseased gingival tissues, RANKL+ cells were greatly elevated compared to the healthy tissues, and a significantly higher percentage of B cells expressed RANKL (94 ?? 5% SD) than T cells (51 ?? 18% SD) (Figure 5C) .
Figure 5. Expression of percentage of RANKL+ T cells or B cells in the gingival tissues. After computer capture of the confocal microscopic image of each gingival section (x400), the percentage of yellow cells (double positive for CD3/RANKL or CD20/RANKL) of the total number of green cells (CD3+ or CD20+) were calculated and are shown on scatter plots (A, CD3+ T cells; B, CD20+ B cells), according to the depth of gingival crevice from which the samples were collected (open triangles, healthy patients; filled triangles, periodontal disease patients). Both RANKL+ T cells and RANKL+ B cells showed positive correlation with the depth of gingival crevice (T cells: n = 34, r = 0.507, P < 0.01; B cells: n = 15, r = 0.702, P < 0.01). C: Percentage of RANKL+ T cells or B cells of the total T cells or B cells in the gingival tissues classified by the depth of the gingival crevice (GC depth) is shown. The data shown in A and B were converted to histograms, presented as mean ?? SD of the percentage of RANKL+ T cells or B cells. *Statistically significant by Student??s t-test (P < 0.0001). Note: the percentage of RANKL+ CD14+ monocytes/total CD14+ monocytes (not shown) was 15.3 ?? 16.8 and 59.5 ?? 39.9 in healthy and disease gingival tissues, respectively, showing no statistical differences by t-test.
Activation of Peripheral Blood T Cells and B Cells Induces sRANKL Production
To investigate whether naïve or activated lymphocytes express RANKL, peripheral blood T cells or B cells from healthy patients were stimulated with immobilized anti-CD3 and anti-CD28 mAbs or with immobilized anti-CD40 and/or anti-IgM, respectively (Figure 6) . The stimulation via CD3/CD28 or CD40/BCR (B-cell receptor; anti-IgM) induced proliferation and sRANKL expression by the purified T cells or B cells . Although production of OPG by the activated T cells or B cells was also monitored (by ELISA), the level of OPG production was lower than the detection limit (4 pg/ml). An oral bacterial stimulus (fixed A. actinomycetemcomitans) also activated peripheral blood lymphocytes to produce both sRANKL and cellular RANKL from T cells and B cells (examined by confocal microscopy, not shown). The activated T cells and B cells isolated from peripheral blood were able to induce osteoclast (TRAP+, multinucleated) cell differentiation in a RANKL-dependent manner (Figure 6E) , whereas activated CD14+ monocytes did not induce such osteoclastogenesis (Figure 6E) . Of note, most published literature also supports the finding that CD14+ cells express of RANK but not RANKL.31,32 These data suggested that T cells and B cells that express RANKL in gingival tissues are probably in the activated form and that these activated T and B cells are the major source of RANKL.
Figure 6. Activation of peripheral blood T cells or B cells induces sRANKL expression. Purified T cells isolated from a healthy subject PBMCs (2 x 105/well) were activated for 7 days with immobilized anti-CD3 (xCD3) and anti-CD28 (xCD28) mAbs, and proliferation (A) and sRANKL production (C) were measured by thymidine incorporation assay and by ELISA, respectively. Purified B cells isolated from the PBMCs (2 x 105/well) were also incubated in a 96-well plate for 7 days with immobilized anti-CD40 (xCD40), anti-IgM (xIgM), and anti-CD40 and anti-IgM antibodies, and proliferation (B) and sRANKL production (D) were measured by the same method as for T cells. E: Osteoclastogenesis induced by the activated peripheral blood T cells, B cells, and CD14+ monocytes is shown. The cultured T cells or B cells under the conditions shown in A or B were harvested and fixed on day 7. After activation of these CD14+ monocytes with fixed A. actinomycetemcomitans for 4 days, CD14+ monocytes produced IL-12 (143 + 23 pg/ml), but not RANKL or OPG. The fixed T cells, B cells, or CD14+ monocytes were co-cultured with MOCP-5 (104/well) in the presence of M-CSF (10 ng/ml) as described under Materials and Methods. The number of TRAP-positive multinuclear cells induced in each well was counted and expressed as mean ?? SD of triplicate determinants. *Significantly higher than medium alone by Student??s t-test (P < 0.05). Other PBMC samples isolated from one periodontal diseased and two orally healthy patients showed similar results.
Evaluation of the Osteoclast Differentiation Function of RANKL Expressed by Gingival Lymphocytes
To evaluate the functional aspects of RANKL expressed by MCs isolated from the diseased gingival tissue, an in vitro differentiation assay was outperformed using the osteoclast precursor cell line MOCP-5 (Figure 7A) or peripheral blood CD14+ monocytes (Figure 7B) . Both MOCP-5 cells and CD14+ monocytes differentiated into TRAP+ multinucleated cells in response to recombinant human sRANKL in the presence of M-CSF. The isolated diseased gingival MCs as well as recombinant RANKL induced TRAP+ multinucleated cells (Figure 7A) . However, only nonadherent diseased gingival MCs, but not adherent gingival MC cells, induced RANKL-dependent osteoclastogenesis (Figure 7B) , indicating that any RANKL expression that might have been detected on CD14+ cells (Figure 3C) is not functionally active. The morphological appearance of the differentiated MOCP-5 cell line is shown , demonstrating that RANKL expressed by MCs from diseased tissues can induce RANKL-dependent osteoclast differentiation. The pit formation assay was performed to assess the functional capabilities of differentiated multinucleated cells (Figure 8, ECJ) . Resorption pits shown as positively stained by toluidine blue (indicated by arrows) were observed in the disks from wells in which MOCP-5 cells were co-cultured with sRANKL or MCs from diseased gingival tissue (Figure 8, F and G , respectively). The formation of resorption pits induced by MOCP-5 co-cultured with MCs from the diseased tissue was inhibited by the presence of OPG (not shown). The pit formation assay using a calcium-phosphate-coated tissue culture vessel system (Figure 8, HCJ) showed similar results to the assay using the dentin disks (Figure 8, ECG) . The MOCP-5 co-cultures with MCs from diseased tissue showed resorption pits (Figure 8J , white clear area, indicated by arrows) as well as MOCP-5 incubated with sRANKL (Figure 8I , arrows), compared to the negative control of MOCP-5 cultured in medium alone, which did not show such resorption pits (Figure 8H) .
Figure 7. Osteoclast differentiation assay using the MOCP-5 osteoclast precursor cell line or human peripheral blood CD14+ monocytes. A: MOCP-5 osteoclast precursor cells were co-cultured with human recombinant RANKL, fresh nonstimulated PBMCs, or gingival MCs isolated from periodontal disease patients. PBMCs and gingival MCs were prefixed with formalin before co-culture with MOCP-5 (104/well). B: CD14+ monocytes isolated from PBMCs of a healthy subject (2 x 104/well) were co-cultured with human recombinant RANKL or with fixed adherent or nonadherent gingival MCs isolated from periodontal disease patient. All culture medium contained recombinant human M-CSF (10 ng/ml) in the presence or absence of OPG-Fc (10 µg/ml) (both A and B). Half of the total culture medium was exchanged every 3 days. On day 8 (A) or on day 16 (B), the cultures were fixed with formalin, and TRAP staining was outperformed. The numbers of cells with more than three nuclei demonstrating TRAP+ staining were counted in each well under phase contrast microscopy. Similar results of TRAP+ cell induction were obtained using the MCs isolated from two different patients?? gingival tissues. *Significantly elevated compared to the control MOCP-5 (A); significantly elevated compared to human CD14+ monocytes cultured in medium alone without OPG-Fc (B) by Student??s t-test (P < 0.01).
Figure 8. TRAP+ multinuclear cells induced from MOCP-5 osteoclast precursors by recombinant RANKL and patient??s gingival MCs. MOCP-5 osteoclast precursor cells were co-cultured with human recombinant RANKL or with gingival MCs isolated from the patient??s periodontal disease lesion. Patient??s gingival MCs fixed with formalin (104/well) were applied to the MOCP-5 culture. MOCP-5 cells were cultured on the oval-shape coverslip in a 96-well plate. Medium that contained M-CSF (10 ng/ml) was exchanged (50% of total culture volume) every 3 days. On day 8, the cultures were stopped, and the cells were fixed with formalin. The MOCP-5 cells adherent on the coverslip were reacted with TRAP reagent and mounted on glass slides. The figures demonstrate the appearance of TRAP staining of MOCP-5 co-cultured for 8 days with recombinant human RANKL (50 ng/ml) (A), recombinant human RANKL plus OPG-Fc (10 µg/ml) (B), fixed patient??s gingival MCs isolated from a diseased lesion (C), and fixed patient??s gingival MCs isolated from a diseased lesion plus OPG-Fc (10 µg/ml) (D). The bone-resorptive activities of TRAP+ multinuclear cells were examined using dentin disks (ECG) and the Osteologic system (HCJ). The MOCP-5 was cultured in a medium supplemented with M-CSF (10 ng/ml) for 8 days in the following different conditions; medium alone (E, H), in the presence of recombinant RANKL (50 ng/ml) (F, I), in the presence of patient??s fixed gingival MCs (G, J). The arrows in A and C indicate the TRAP+ multinuclear cells. The arrows in ECJ indicate the resorption pits formed by TRAP+ multinuclear cells. Of note, toluidine blue stains resportion pits dark brown (F, G), while the resorption pits in the Osteologic system are demonstrated as white clear areas (I, J).
Discussion
The present study of the bone resorptive lesions of periodontal disease tissues demonstrated that, not only T cells, but also B cells, are major sources of RANKL. The concentrations of sRANKL and IL-1ß examined in the gingival tissue homogenates were significantly elevated in the diseased gingival tissues compared to healthy tissues, whereas OPG protein production was not significantly higher in diseased tissues than healthy tissues. Overexpression of OPG in transgenic mice and RANK gene knockout mice develop severe osteopetrosis33,34 and OPG gene knockout mice demonstrate osteopenia.35 The ratio of RANKL to OPG is normally increased in a bone resorptive lesion characterized by extensive osteoclastic activity.36 Therefore nearly exclusive production of RANKL and the lack of production of OPG by lymphocytes seem to contribute to the increased RANKL to OPG ratio in the bone resorptive periodontitis lesions. Key studies showed that RANKL gene knockout mice not only develop osteopetrosis but are also deficient in both T and B lymphocytes,10,37 implying a RANKL-associated linkage between T and B cells and bone resorption. Another study using a SCID-Hu mouse model indicated that human T cells can express RANKL in response to certain oral bacteria.19 RANKL protein expression has been demonstrated by T cells isolated from periodontitis gingival tissue by flow cytometry.38 Consistent with our findings that diseased gingival tissue contains much more sRANKL than healthy tissue (Figure 1A) and that the concentration of sRANKL is greater in deeper gingival crevices (Figure 1B) is the recent report that the ratio of sRANKL to OPG in gingival crevice fluid (GCF) is elevated in periodontitis patients compared to healthy patients.39 Therefore, the findings presented herein that RANKL is expressed by T cells and also B cells and that a higher percentage of B cells express RANKL than T cells in periodontitis tissue provides information quite relevant to the elucidation of bone resorptive factors in periodontal disease.
We have recently demonstrated that activated antigen-specific B cells can induce periodontal bone resorption in a RANKL-dependent manner using a rat periodontitis model.22 B cells primed in vivo with specific antigen (A. actinomycetemcomitans) up-regulated RANKL expression in response to in vitro stimulation with antigen. The adoptive transfer of these donor antigen-primed B cells to congenitally athymic recipient rats that received simultaneous gingival injection with antigen resulted in periodontal bone resorption in conjunction with elevated IgG antibody response to A. actinomycetemcomitans.22 This periodontal bone resorption was abrogated by OPG-Fc local injection.22 Accumulation of plasma cells and B cells in the disease lesion11,12 and elevated IgG antibodies to periodontal bacteria in the serum1,2,28 and gingival crevice fluid40 are distinct features of periodontitis. However, the nature of B cells in the context of periodontal bone resorption has not been clearly elucidated until our previous study.22 Herein, we revealed that B cells, in addition to T cells, are primary cellular sources of bone destructive factor by immunohistological studies of human periodontal disease lesions.
In our previous studies of the rat periodontal disease model using adoptive transfer of T-clone cells, antigen-specific stimulation of the T-clone cells enhanced expression of RANKL mRNA to a greater degree than expression of OPG mRNA as determined by RT-PCR, and resulted in local periodontal bone resorption.17 Sakata and colleagues41 reported that dental mesenchymal cells produce OPG and enhance OPG production in response to proinflammatory factors such as IL-1ß or tumor necrosis factor (TNF)-. Nagasawa and colleagues38 showed that lipopolysaccharide stimulation of gingival fibroblasts induced OPG expression and inhibited differentiation of monocytes to osteoclast cells. Although the ELISA system in the present study detected OPG protein production in both healthy and diseased gingival tissues, mRNA message for OPG was not detected in healthy gingival tissue as measured by RT-PCR. Because the same RT-PCR system demonstrated the mRNA message of ß-actin and RANKL, nonspecific mRNA degradation by RNases during the sampling process was excluded as the cause of such discrepancy between OPG mRNA and OPG protein expression. Furthermore, the difference between in vivo (present study) and in vitro cultures of dental mesenchymal cells41 may cause the discrepancy in OPG mRNA detection. In the present study, stimulation of peripheral blood T cells and B cells in vitro did not produce detectable levels of OPG irrespective of stimulation (data not shown). Therefore, the OPG production in both healthy and diseased gingival tissues may be attributed, at least, to fibroblasts in periodontal tissue that are capable of producing physiologically relevant amounts of OPG, maintaining the homeostasis of healthy periodontal bone remodeling.
Although few studies investigated the mechanism underlying osteoclast recruitment to the bone resorption lesion, MIP-1 and other CCR1 ß-chemokines seem to be associated with the osteoclast migration as well as osteoclast formation and bone resorption.42,43 With respect to the chemotaxis of osteoclast precursors, the homing mechanism of such cells to gingival tissues remains unclear. In the present study, MIP-1 concentration in diseased gingival tissues was not significantly different from healthy tissues.
In metastatic bone disease, the -chemokine IL-8 has been implicated as a stimulator of osteoclastogenesis and bone resorption.44 However, IL-8 is an important chemoattractant factor to recruit neutrophils to gingival tissues because an innate immune reaction and defect in neutrophil function is thought to be a pathogenic mechanism in periodontal disease.45 It is important to clarify the role of IL-8 in periodontal bone resorption because T cells and B cells, which are found to be the major sources of RANKL, are attracted by ß-chemokines including MIP-1 but not by IL-8.
In addition to RANKL and OPG, other cytokines including M-CSF, IL-1, IL-6, IL-11, IL-17, and TNF- have been reported to support osteoclast formation by up-regulating RANKL expression, promoting osteoclast differentiation in a RANKL-dependent or perhaps in a RANKL-independent manner.5,46-49 By contrast to these osteoclastogenesis-promoting cytokines, IL-10 is reported to inhibit osteoclast differentiation.50 Although the role of GM-CSF is controversial, this cytokine seems to decrease the osteoclastogenesis induced by bone marrow stromal cells.51 IL-12 and IL-18 are indicated as osteoclast inhibitory cytokines through mechanisms dependent on GM-CSF52 or IFN-.53 In the present study, an inflammatory cytokine, IL-1ß, measured in gingival tissues was significantly elevated in the diseased tissues as was RANKL (Figure 1, A and E) , whereas IL-12, IL-10, and GM-CSF levels were not significantly different in the diseased gingival tissues compared to the healthy tissues. Because IL-1 is downstream of TNF--induced osteoclastogenesis,54,55 it is conceivable that elevated IL-1ß may reflect the TNF- produced in periodontal tissues. The involvement of TNF- in the periodontal bone loss was demonstrated in a primate model of periodontal disease; ie, soluble receptors to IL-1 or to TNF-, respectively, can significantly inhibit the inflammatory response and bone loss.56 TNF- can cause osteoclastogenesis in a RANKL-independent manner,46 and also synergistically increase RANKL-dependent osteoclast differentiation.54 However, OPG-Fc mediated nearly complete inhibition of in vitro osteoclastogenesis from MOCP-5 and from human peripheral blood CD14+ monocytes (Figure 7, A and B) . These findings seemed to exclude the possible involvement of membrane-bound TNF- and other secondary osteoclast-stimulatory components expressed on activated T cells and/or B cells.
It is well documented that generic inflammation in the synovial cavity can lead to the secondary expression of RANKL and other osteoclast differentiation factors.55 We, however, hypothesize that the mechanism underlying periodontal bone resorption is different from the mechanism for rheumatoid arthritis, because the healthy oral cavity is constantly exposed to bacterial inflammatory mediators, such as lipopolysaccharide or peptidoglycan,57 whereas the normal synovial cavity is free from bacterial components. Nonetheless, healthy gingival tissue is free from bone resorption, whereas mitogenic stimulation associated with bacteria may cause inflammation in synovial tissues. Our hypothesis is that generic inflammation, per se, is not the cause of bone resorption in the periodontal tissues. This hypothesis is supported by gingivitis, another form of periodontal inflammation,58 that does not exhibit bone resorption. Our rat periodontal disease model also illustrated that antigen-specific T-cell and B-cell activation in the inflamed gingival tissues is necessary for the induction of periodontal bone loss, and simple inflammation induced by lipopolysaccharide alone does not cause periodontal bone resorption.22,30 Therefore RANKL expression by activated lymphocytes may be a prerequisite to develop osteoclast differentiation in the gingival tissues having pre-existing inflammation.
It is conceivable that not only lymphocytes but also osteoblasts may be involved in increases of the RANKL to OPG ratio in the inflamed gingival tissue of periodontitis. The present study did not evaluate osteoblasts, which are also a potential RANKL source in periodontal disease tissue. IL-1 and TNF- appear to be present in the inflamed gingival tissues59 and are able to induce RANKL expression by osteoblast cells. Despite the potential involvement of other factors in the bone destruction process, prominent expression of RANKL by B and T cells in the periodontal disease lesion seems to play a primary role in the augmentation of bone resorption processes in this disease.
Acknowledgements
We thank Dr. Colin Dunstan (Amgen Inc., Thousand Oaks, CA) for OPG-Fc; Dr. Zedonis Skobe, Dr. Rick A. Rogers, Ms. Jean Lai, and Ms. Elke Pravda for confocal microscopy support; and Dr. Takashi Yanaoka and Mr. Tadahiko Kounoike for some support of this study.
【参考文献】
McArthur WP, Clark WB: Specific antibodies and their potential role in periodontal diseases. J Periodontol 1993, 64:807-818
Taubman MA, Ebersole JL, Smith DJ: Genco RJ Mergenhangen SE eds. Association between systemic and local antibody and periodontal diseases. Host Parasite Interactions in Periodontal Diseases. 1982:pp 283-298 American Society of Microbiology, Washington DC
Ebersole JL, Taubman MA, Smith DJ, Frey DE, Haffajee AD, Socransky SS: Human serum antibody responses to oral microorganisms. IV. Correlation with homologous infection. Oral Microbiol Immunol 1987, 2:53-59
Wu X, Pan G, McKenna MA, Zayzafoon M, Xiong WC, McDonald JM: RANKL regulates Fas expression and Fas-mediated apoptosis in osteoclasts. J Bone Miner Res 2005, 20:107-116
Hofbauer LC, Khosla S, Dunstan CR, Lacey DL, Boyle WJ, Riggs BL: The roles of osteoprotegerin and osteoprotegerin ligand in the paracrine regulation of bone resorption. J Bone Miner Res 2000, 15:2-12
Kong YY, Boyle WJ, Penninger JM: Osteoprotegerin ligand: a common link between osteoclastogenesis, lymph node formation and lymphocyte development. Immunol Cell Biol 1999, 77:188-193
Kong YY, Feige U, Sarosi I, Bolon B, Tafuri A, Morony S, Capparelli C, Li J, Elliott R, McCabe S, Wong T, Campagnuolo G, Moran E, Bogoch ER, Van G, Nguyen LT, Ohashi PS, Lacey DL, Fish E, Boyle WJ, Penninger JM: Activated T cells regulate bone loss and joint destruction in adjuvant arthritis through osteoprotegerin ligand. Nature 1999, 402:304-309
Boyle WJ, Simonet WS, Lacey DL: Osteoclast differentiation and activation. Nature 2003, 423:337-342
Simonet WS, Lacey DL, Dunstan CR, Kelley M, Chang MS, Luthy R, Nguyen HQ, Wooden S, Bennett L, Boone T, Shimamoto G, DeRose M, Elliott R, Colombero A, Tan HL, Trail G, Sullivan J, Davy E, Bucay N, Renshaw-Gegg L, Hughes TM, Hill D, Pattison W, Campbell P, Boyle WJ: Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell 1997, 89:309-319
Lacey DL, Timms E, Tan HL, Kelley MJ, Dunstan CR, Burgess T, Elliott R, Colombero A, Elliott G, Scully S, Hsu H, Sullivan J, Hawkins N, Davy E, Capparelli C, Eli A, Qian YX, Kaufman S, Sarosi I, Shalhoub V, Senaldi G, Guo J, Delaney J, Boyle WJ: Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 1998, 93:165-176
Mackler BF, Frostad KB, Robertson PB, Levy BM: Immunoglobulin bearing lymphocytes and plasma cells in human periodontal disease. J Periodontal Res 1977, 12:37-45
Seymour GJ, Greenspan JS: The phenotypic characterization of lymphocyte subpopulations in established human periodontal disease. J Periodontal Res 1979, 14:39-46
Taubman MA, Stoufi ED, Ebersole JL, Smith DJ: Phenotypic studies of cells from periodontal disease tissues. J Periodontal Res 1984, 19:587-590
Okada H, Kida T, Yamagami H: Identification and distribution of immunocompetent cells in inflamed gingiva of human chronic periodontitis. Infect Immun 1983, 41:365-374
Yamazaki K, Nakajima T, Aoyagi T, Hara K: Immunohistological analysis of memory T lymphocytes and activated B lymphocytes in tissues with periodontal disease. J Periodontal Res 1993, 28:324-334
Seymour GJ, Taubman MA, Eastcott JW, Gemmell E, Smith DJ: CD29 expression on CD4+ gingival lymphocytes supports migration of activated memory T lymphocytes to diseased periodontal tissue. Oral Microbiol Immunol 1997, 12:129-134
Valverde P, Kawai T, Taubman MA: Selective blockade of voltage-gated potassium channels reduces inflammatory bone resorption in experimental periodontal disease. J Bone Miner Res 2004, 19:155-164
Valverde P, Kawai T, Taubman MA: Potassium channel-blockers as therapeutic agents to interfere with bone resorption of periodontal disease. J Dent Res 2005, 84:488-499
Teng YA, Nguyen H, Gao X, Kong Y, Gorczynski RM, Singh B, Ellen RP, Penninger JM: Functional human T-cell immunity and osteoprotegerin ligand control alveolar bone destruction in periodontal infection. J Clin Invest 2000, 106:R59-R67
Croucher PI, Shipman CM, Lippitt J, Perry M, Asosingh K, Hijzen A, Brabbs AC, van Beek EJ, Holen I, Skerry TM, Dunstan CR, Russell GR, Van Camp B, Vanderkerken K: Osteoprotegerin inhibits the development of osteolytic bone disease in multiple myeloma. Blood 2001, 98:3534-3540
Choi Y, Woo KM, Ko SH, Lee YJ, Park SJ, Kim HM, Kwon BS: Osteoclastogenesis is enhanced by activated B cells but suppressed by activated CD8(+) T cells. Eur J Immunol 2001, 31:2179-2188
Han X, Kawai T, Eastcott JW, Taubman MA: Bacterial-responsive B lymphocytes induce periodontal bone resorption. J Immunol 2006, 176:625-631
Takeichi O, Haber J, Kawai T, Smith DJ, Moro I, Taubman MA: Cytokine profile of T lymphocytes from gingival tissues with pathological pocketing. J Dent Res 2000, 79:1548-1555
Kawai T, Seki M, Hiromatsu K, Eastcott JW, Watts GFM, Sugai M, Smith DJ, Porcelli SA, Taubman MA: Selective diapedesis of Th1 cells induced by endothelial cell RANTES. J Immunol 1999, 163:3269-3278
Hofbauer LC, Gori F, Riggs BL, Lacey DL, Dunstan CR, Spelsberg TC, Khosla S: Stimulation of osteoprotegerin ligand and inhibition of osteoprotegerin production by glucocorticoids in human osteoblastic lineage cells: potential paracrine mechanisms of glucocorticoid-induced osteoporosis. Endocrinology 1999, 140:4382-4389
Smith DJ, Gadalla LM, Ebersole JL, Taubman MA: Gingival crevicular fluid antibody to oral microorganisms. III. Association of gingival homogenate and gingival crevicular fluid antibody levels. J Periodontal Res 1985, 20:357-367
Kawai T, Ito H, Sakado N, Okada H: A novel approach for detecting an immunodominant antigen of Porphyromonas gingivalis in diagnosis of adult periodontitis. Clin Diagn Lab Immun 1998, 5:11-17
Ebersole JL, Frey DE, Taubman MA, Smith DJ: An ELISA for measuring serum antibodies to Actinobacillus actinomycetemcomitans. J Periodontal Res 1980, 15:621-632
Chen W, Li Y-P: Generation of mouse osteoclastogenic cell lines immortalized with SV40 large T antigen. J Bone Miner Res 1998, 13:1112-1123
Kawai T, Eisen-Lev R, Seki M, Eastcott JW, Wilson ME, Taubman MA: Requirement of B7 costimulation for Th1-mediated inflammatory bone resorption in experimental periodontal disease. J Immunol 2000, 164:2102-2109
Boyle WJ: Oppenheim JJ Feldmann M Durum SK Hirano T Vilcek J Nicola NA eds. RANK ligand. Cytokine Reference. 2001:pp 489-497 Academic Press, San Diego
Boyle WJ: Oppenheim JJ Feldmann M Durum SK Hirano T Vilcek J Nicola NA eds. RANK. Cytokine Reference. 2001:pp 1691-1709 Academic Press, London
Min H, Morony S, Sarosi I, Dunstan CR, Capparelli C, Scully S, Van G, Kaufman S, Kostenuik PJ, Lacey DL, Boyle WJ, Simonet WS: Osteoprotegerin reverses osteoporosis by inhibiting endosteal osteoclasts and prevents vascular calcification by blocking a process resembling osteoclastogenesis. J Exp Med 2000, 192:463-474
Dougall WC, Glaccum M, Charrier K, Rohrbach K, Brasel K, De Smedt T, Daro E, Smith J, Tometsko ME, Maliszewski CR, Armstrong A, Shen V, Bain S, Cosman D, Anderson D, Morrissey PJ, Peschon JJ, Schuh J: RANK is essential for osteoclast and lymph node development. Genes Dev 1999, 13:2412-2424
Mizuno A, Amizuka N, Irie K, Murakami A, Fujise N, Kanno T, Sato Y, Nakagawa N, Yasuda H, Mochizuki S, Gomibuchi T, Yano K, Shima N, Washida N, Tsuda E, Morinaga T, Higashio K, Ozawa H: Severe osteoporosis in mice lacking osteoclastogenesis inhibitory factor/osteoprotegerin. Biochem Biophys Res Commun 1998, 247:610-615
Hofbauer LC, Kuhne CA, Viereck V: The OPG/RANKL/RANK system in metabolic bone diseases. J Musculoskel Neuronal Interact 2004, 4:268-275
Kong YY, Yoshida H, Sarosi I, Tan HL, Timms E, Capparelli C, Morony S, Oliveira-dos-Santos AJ, Van G, Itie A, Khoo W, Wakeham A, Dunstan CR, Lacey DL, Mak TW, Boyle WJ, Penninger JM: OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis. Nature 1999, 397:315-323
Nagasawa T, Kobayashi H, Kiji M, Aramaki M, Mahanonda R, Kojima T, Murakami Y, Saito M, Morotome Y, Ishikawa I: LPS-stimulated human gingival fibroblasts inhibit the differentiation of monocytes into osteoclasts through the production of osteoprotegerin. Clin Exp Immunol 2002, 130:338-344
Mogi M, Otogoto J, Ota N, Togari A: Differential expression of RANKL and osteoprotegerin in gingival crevicular fluid of patients with periodontitis. J Dent Res 2004, 83:166-169
Murray PA, Burstein DA, Winkler JR: Antibodies to Bacteroides gingivalis in patients with treated and untreated periodontal disease. J Periodontol 1989, 60:96-103
Sakata M, Shiba H, Komatsuzawa H, Fujita T, Ohta K, Sugai M, Suginaka H, Kurihara H: Expression of osteoprotegerin (osteoclastogenesis inhibitory factor) in cultures of human dental mesenchymal cells and epithelial cells. J Bone Miner Res 1999, 14:1486-1492
Yu X, Huang Y, Collin-Osdoby P, Osdoby P: CCR1 chemokines promote the chemotactic recruitment, RANKL development, and motility of osteoclasts and are induced by inflammatory cytokines in osteoblasts. J Bone Miner Res 2004, 19:2065-2077
Roodman GD, Choi SJ: MIP-1 alpha and myeloma bone disease. Cancer Treat Res 2004, 118:83-100
Bendre MS, Margulies AG, Walser B, Akel NS, Bhattacharrya S, Skinner RA, Swain F, Ramani V, Mohammad KS, Wessner LL, Martinez A, Guise TA, Chirgwin JM, Gaddy D, Suva LJ: Tumor-derived interleukin-8 stimulates osteolysis independent of the receptor activator of nuclear factor-kappaB ligand pathway. Cancer Res 2005, 65:11001-11009
Gainet J, Dang PM, Chollet-Martin S, Brion M, Sixou M, Hakim J, Gougerot-Pocidalo MA, Elbim C: Neutrophil dysfunctions, IL-8, and soluble L-selectin plasma levels in rapidly progressive versus adult and localized juvenile periodontitis: variations according to disease severity and microbial flora. J Immunol 1999, 163:5013-5019
Kobayashi K, Takahashi N, Jimi E, Udagawa N, Takami M, Kotake S, Nakagawa N, Kinosaki M, Yamaguchi K, Shima N, Yasuda H, Morinaga T, Higashio K, Martin TJ, Suda T: Tumor necrosis factor alpha stimulates osteoclast differentiation by a mechanism independent of the ODF/RANKL-RANK interaction. J Exp Med 2000, 191:275-286
Suda T, Takahashi N, Udagawa N, Jimi E, Gillespie MT, Martin TJ: Modulation of osteoclast differentiation and function by the new members of the tumor necrosis factor receptor and ligand families. Endocr Rev 1999, 20:345-357
Teitelbaum SL: Bone resorption by osteoclasts. Science 2000, 289:1504-1508
Goldring SR, Gravallese EM: Pathogenesis of bone erosions in rheumatoid arthritis. Curr Opin Rheumatol 2000, 12:195-199
Xu LX, Kukita T, Kukita A, Otsuka T, Niho Y, Iijima T: Interleukin-10 selectively inhibits osteoclastogenesis by inhibiting differentiation of osteoclast progenitors into preosteoclast-like cells in rat bone marrow culture system. J Cell Physiol 1995, 165:624-629
Gorny G, Shaw A, Oursler MJ: IL-6, LIF, and TNF-alpha regulation of GM-CSF inhibition of osteoclastogenesis in vitro. Exp Cell Res 2004, 294:149-158
Horwood NJ, Elliott J, Martin TJ, Gillespie MT: IL-12 alone and in synergy with IL-18 inhibits osteoclast formation in vitro. J Immunol 2001, 166:4915-4921
Yamada N, Niwa S, Tsujimura T, Iwasaki T, Sugihara A, Futani H, Hayashi S, Okamura H, Akedo H, Terada N: Interleukin-18 and interleukin-12 synergistically inhibit osteoclastic bone-resorbing activity. Bone 2002, 30:901-908
Wei S, Kitaura H, Zhou P, Ross FP, Teitelbaum SL: IL-1 mediates TNF-induced osteoclastogenesis. J Clin Invest 2005, 115:282-290
Walsh NC, Crotti TN, Goldring SR, Gravallese EM: Rheumatic diseases: the effects of inflammation on bone. Immunol Rev 2005, 208:228-251
Assuma R, Oates T, Cochran D, Amar S, Graves DT: IL-1 and TNF antagonists inhibit the inflammatory response and bone loss in experimental periodontitis. J Immunol 1998, 160:403-409
Madianos PN, Bobetsis YA, Kinane DF: Generation of inflammatory stimuli: how bacteria set up inflammatory responses in the gingiva. J Clin Periodontol 2005, 32(Suppl 6):57-71
Page RC: Gingivitis. J Clin Periodontol 1986, 13:345-359
Tanaka Y, Nakayamada S, Okada Y: Osteoblasts and osteoclasts in bone remodeling and inflammation. Curr Drug Targets Inflamm Allergy 2005, 4:325-328
作者单位：From the Departments of Immunology* and Cytokine Biology,** The Forsyth Institute, Boston, Massachusetts; the Department of Oral Medicine, Infection, and Immunity, Harvard School of Dental Medicine, Boston, Massachusetts; the Department of General Dentistry,¶ Tufts University School of Dental M