The present study tested the effects of local injection of IL-1 and TNF soluble receptors on a periodontal wound-healing model in nonhuman primates. In this model, periodontal lesions were developed for 16 wk, followed by open flap surgery. Starting at the time of surgery, groups of animals received localized injections of both soluble cytokine receptors or else PBS three times per week for 3, 14, or 35 days. Periodontal wound healing was analyzed for each group at the end of the treatment regimen. Fourteen days after surgery, a significant decrease was observed between the animals treated with soluble receptors and the untreated group with respect to recruitment of inflammatory cells in deep gingival connective tissue. Concurrent apoptosis of inflammatory cells in those tissues increased significantly in treated animals compared with untreated animals. All other outcome parameters of periodontal wound healing were likewise significantly improved in treated animals compared with untreated animals. In marked contrast, however, 35 days after surgery, there was a significant increase in the number of inflammatory cells that had infiltrated into deep gingival connective tissue in treated compared with untreated animals. Outcome parameters of periodontal wound healing worsened in treated animals when compared with untreated. These results indicate that proinflammatory cytokines may play different functional roles in early vs late phases of periodontal wound healing. Short-term blockade of IL-1 and TNF may facilitate periodontal wound healing, whereas prolonged blockade may have adverse effects.

Wound healing involves a complex cascade of events aimed at re-establishing tissue integrity (1). This process has been separated into three overlapping phases: inflammation, granulation tissue formation, and matrix formation and remodeling. These three phases culminate ∼3, 14, and 35 days after injury, respectively (2).

Periodontal wound healing is thought to be far more complex than dermal or mucosal wound healing. Complications inherent to this anatomical site include the presence of microbial plaque, variations in pH and temperature, and often extensive destruction of structural elements in the gingiva and underlying bone, which require a complex and coordinated sequence of regenerative events (3, 4). Published data indicate that the inflammatory response may interfere with proper periodontal tissue regeneration (5), suggesting that limiting periodontal inflammation may be beneficial. The present study was conducted to explore whether local inhibition of the two principal proinflammatory cytokines, IL-1 and TNF (6, 7, 8, 9, 10), would benefit wound healing in an animal model of periodontal disease.

IL-1 (IL-1α and IL-1β) is a pleiotropic proinflammatory cytokine responsible for fundamental functions in wound healing, inflammation, and host antitumor responses (11, 12, 13). IL-1 can induce apoptosis (14) or antagonize apoptosis through activation of NF-κB depending on the cell type (15). A small amount of IL-1 is necessary for host defense and wound healing (16), particularly in a challenging environment (17), whereas overproduction of IL-1 can hinder the early phase of wound healing (18, 19). Two primary IL-1Rs (IL-1RI and IL-1RII) have been identified. After binding IL-1, IL-1RI in association with the receptor accessory protein transduces a signal, whereas IL-1RII binds IL-1, but does not appear to transfer a signal subsequently (20). A recombinant version of the extracellular portion of IL-1RI (rIL-1RI) has been engineered that retains the IL-1-binding properties of the native protein and blocks activity of the ligand (21, 22). This exogenous soluble type I IL-1R (sIL-1RI)3 has been shown to reduce periodontal inflammation (23) and ocular inflammation (24). Finally, IL-1 receptor antagonist (natural antagonist to IL-1) has been shown to improve neurological recovery after traumatic brain injury (25).

TNF (TNF-α and TNF-β) is another central mediator of inflammatory responses, playing important roles in antimicrobial defense, wound healing, and defense against malignant disorders (26). TNF-α can stimulate apoptosis-related events (27) and NF-κB activation (28). Although small amounts of TNF are necessary for host defense against infection, overproduction of TNF can be detrimental. Indeed, TNF was shown to contribute to pathologies such as rheumatoid arthritis, periodontitis, or multiple sclerosis (29, 30, 31), and short-term localized application of TNF-α in rats has been found to inhibit tissue repair (32). TNF biological effects have been shown to be mediated via two distinct cell surface receptors: TNFRI and TNFRII (33). In wound-healing experiments, TNFRI-deficient mice were found to exhibit accelerated wound healing compared with wild-type mice (34). The extracellular domains of the two receptors are similar, while their intracellular domains differ from one another. Most notably, TNFRI contains a death domain that is absent in TNFRII. Soluble TNFRs (sTNFRs) act as inhibitors to TNF by binding directly to TNF, blocking ligand activity (35). Linking sTNFRs to the Fc region of the Ig confers greater stability, resulting in a chimeric molecule with an extended t1/2. One such fusion protein of the p75 sTNFR and the Fc portion of human IgG1 (etanercept) is now commercially available for treating rheumatoid arthritis (36), ankylosing spondylitis (37), and psoriasis (38).

Because inflammation is the initial phase of wound healing, we applied the same logic of interfering with IL-1 and TNF effects in this phase of wound healing (2) in hopes of improving outcomes in a well-defined nonhuman primate model of periodontal wound healing. We administered soluble receptors for IL-1 and TNF to block IL-1 and TNF functions, and then evaluated wound-healing outcomes. Our results suggest that IL-1 and TNF play distinct roles in different phases of periodontal wound healing. We conclude that short-term inhibition of these two cytokines might improve periodontal wound healing, whereas prolonged treatment would most likely delay it.

sIL-1RI (consisting of the extracellular portion of the IL-1RI) and the chimeric soluble receptor to TNF (sTNFR:Fc, consisting of the extracellular domain of TNFRII linked to the Fc portion of a human IgG1) were generously provided by Amgen (Thousand Oaks, CA). Dosing of sIL-1RI (6 μg/injection) and of sTNFR:Fc (6 μg/injection) and administration frequencies were based upon our previous experience (39).

To minimize potential estrogen effects, 21 adult male Macaca mulatta were used in accordance with the protocol and procedures that had been approved by the Institutional Animal Care And Use Committee at the Boston University Medical Center. The animals used in this study were sedated with ketamine hydrochloric acid for all procedures. In addition, i.v. pentobarbital sodium was used during all surgical procedures. The temporal sequence of this model and the experimental design used in the present study are shown in Fig. 1. Maxillary palatal periodontal defects were produced, as described previously by Caton et al. (40), Amar et al. (41), and Karatzas et al. (42). After an 8-wk quarantine period, intrasulcular incisions were made on the palatal aspect of the maxilla and extended from the canine to the second maxillary molar in that quadrant. Mucoperiosteal flaps were reflected to expose the underlying bone, and under sterile saline irrigation osseous defects were surgically created by removing 5 mm alveolar bone from the crown-root junction (CRJ) in the corono-apical direction to expose the upper portion of the palatal roots of the two premolars and the first and second molars, extending well interproximally. Root surfaces were scaled and root planed to remove the exposed periodontal ligament and surface cementum. Identical lesions were produced on the contralateral teeth. To prevent spontaneous tissue regeneration, each of the involved eight teeth was fitted with a stainless steel band with a finger-like projection extending on the root exposure, and cemented to crown of the tooth. To mimic human periodontal disease, we enhanced plaque accumulation by tying dental plaque-retentive silk ligatures around each tooth, and the flaps were sutured. Six weeks after surgery, the bands were removed and the silk ligatures were replaced to enhance bacterial plaque accumulation on the root surface. After another 6 wk, all silk ligatures were removed. Similar to the standard periodontal care and to obtain clinically healthy gingival condition, an oral hygiene program that consisted of tooth brushing with an antimicrobial (2% chlorhexidine) three times per week was started and continued for the whole duration of the study until tissues were harvested from the animals. Four weeks after the onset of oral hygiene procedures, surgical tissue blocks containing two premolars and the first and second molars with all surrounding tissues (referred as a maxillary block) from both sides (left and right) were dissected en bloc from each of three randomly selected animals. These specimens served as pretreatment controls (baseline controls). Open flap periodontal surgery was performed on the remaining 18 animals. First, inflammatory granulation tissue was carefully debrided. The exposed root surfaces were thoroughly root planed, and the apical extent of root planing was marked by the creation of a notch on the tooth surface. This landmark was used in histologic sections to differentiate newly formed tissues (cementum, bone) from pre-existing tissues. The flaps were then secured with 3.0 silk sutures. Animals were randomly assigned to the experimental (9 animals) or untreated groups (9 animals). For each group, 9 animals were randomly assigned to day 3 (3 animals), day 14 (3 animals), or day 35 (3 animals) time group.

FIGURE 1.

Overview of the experimental design. The study was blinded and randomized. After an 8-wk quarantine, the pre-experimental phase was performed to induce experimental periodontitis for 16 wk, as previously shown (43 ). During this pre-experimental phase, all animals were subjected to the same procedures. This was followed by the experimental phase subdivided into three phases to model the process of repair. During the experimental period, animals underwent periodontal surgery, followed by a series of local injections of IL-1/TNF antagonists or control vehicle for 3, 14, and 35 days. The time points were selected to capture critical periods during the three phases of the wound healing. The 3-day time point corresponds to the peak of the inflammatory phase; the 14-day time point to the peak in granulation tissue formation; and the 35-day time point corresponds to the matrix formation phase. All animals were under a stringent oral hygiene program.

FIGURE 1.

Overview of the experimental design. The study was blinded and randomized. After an 8-wk quarantine, the pre-experimental phase was performed to induce experimental periodontitis for 16 wk, as previously shown (43 ). During this pre-experimental phase, all animals were subjected to the same procedures. This was followed by the experimental phase subdivided into three phases to model the process of repair. During the experimental period, animals underwent periodontal surgery, followed by a series of local injections of IL-1/TNF antagonists or control vehicle for 3, 14, and 35 days. The time points were selected to capture critical periods during the three phases of the wound healing. The 3-day time point corresponds to the peak of the inflammatory phase; the 14-day time point to the peak in granulation tissue formation; and the 35-day time point corresponds to the matrix formation phase. All animals were under a stringent oral hygiene program.

Close modal

As shown in Fig. 2, three animals per time point received IL-1/TNF inhibitors, while three received vehicle alone. Treatment with soluble receptors or vehicle was by local injection into the gingiva adjacent to each periodontal defect. Each injection consisted of 12 μg of soluble receptors (6 μg/injection sIL-1RI plus 6 μg/injection sTNFR:Fc) in 50 μl of sterile PBS or sterile PBS alone. Each animal received one set of injections (8 local injections = 4 injections per quadrant × 2 quadrants) three times per week. Thus, the number of sets of injections per animals was 2 for day 3 animals, 6 for day 14 animals, and 15 for day 35 animals. At the end of each time point, maxillary blocks, including the alveolar bone, periodontal ligament, and teeth, were surgically harvested.

FIGURE 2.

Injection schedule for animals receiving either IL-1/TNF soluble receptors or vehicle alone. Three animals were randomly selected as baseline controls and received no injections. Another 18 animals received injections: 9 randomly selected animals (3 per time point) received the IL-1/TNF soluble receptors, and the other 9 animals (3 per time point) received vehicle alone (sterile PBS). Injections were placed in both the right and left quadrant in the midpalatal area, 3 mm below the CRJ. Each animal received a set of 8 injections (4 injections per quadrant × 2 quadrants) three times per week with Monday-Wednesday-Friday schedule. Day 3 animals received 2 sets of injections, day 14 animals received 6 sets of injections, and day 35 animals received 15 sets of injections. At the end of each time point, maxillary blocks, including the alveolar bones, periodontal ligaments, and teeth, were harvested each group.

FIGURE 2.

Injection schedule for animals receiving either IL-1/TNF soluble receptors or vehicle alone. Three animals were randomly selected as baseline controls and received no injections. Another 18 animals received injections: 9 randomly selected animals (3 per time point) received the IL-1/TNF soluble receptors, and the other 9 animals (3 per time point) received vehicle alone (sterile PBS). Injections were placed in both the right and left quadrant in the midpalatal area, 3 mm below the CRJ. Each animal received a set of 8 injections (4 injections per quadrant × 2 quadrants) three times per week with Monday-Wednesday-Friday schedule. Day 3 animals received 2 sets of injections, day 14 animals received 6 sets of injections, and day 35 animals received 15 sets of injections. At the end of each time point, maxillary blocks, including the alveolar bones, periodontal ligaments, and teeth, were harvested each group.

Close modal

Maxillary blocks were immediately fixed in 4% paraformaldehyde in PBS for 2 days at 4°C. After fixation, specimens were washed and decalcified in 14.5% EDTA/glycerol (pH 7.1) for 14 wk with constant stirring at 4°C (39). After decalcification, each maxillary block was further dissected to provide four individual tooth blocks that included all surrounding periodontal tissues. Tooth blocks were paraffin embedded, and sectioning was performed on a buccal-lingual plane parallel to the long axis of the root. Serial 6-μm sections were prepared and stained with H&E, or with tartrate-resistant acid phosphatase (TRAP) (43), or were subjected to TUNEL assay (44). A diagram of histologic structure landmarks used for the analysis is illustrated in Fig. 3.

FIGURE 3.

Schematic representation of dental tissues and landmarks used in the histomorphometric analysis. Linear measurements are represented by epithelial downgrowth (arrow A), new cementum formation (arrow B), and bone resorption (arrow C). Cell analysis was performed in a minimum of five fields located in the depicted box above notch and represented in deep gingival connective tissue.

FIGURE 3.

Schematic representation of dental tissues and landmarks used in the histomorphometric analysis. Linear measurements are represented by epithelial downgrowth (arrow A), new cementum formation (arrow B), and bone resorption (arrow C). Cell analysis was performed in a minimum of five fields located in the depicted box above notch and represented in deep gingival connective tissue.

Close modal

All data were analyzed in a double-blind fashion using coded specimens. Sections from different specimens were evaluated in a random sequence to prevent bias. Two independent examiners performed the analysis of the tissue sections. The interexaminer and intraexaminer variations were <10%. All measurements were performed with the help of computer-assisted image analysis using ImageProPlus software (Media Cybernetics, Silver Spring, MD). Images were captured, coded, stored on disk, and analyzed randomly at a later time. Both of the observers used the same sets of coded images to perform analysis. The number of apoptotic cells was measured in situ by the TUNEL assay using a TdT-blue label detection kit (Trivegen, Gaithersburg, MD). Nuclei were counterstained with Fast Red. All of the reagents used for TRAP staining were from Sigma-Aldrich (St. Louis, MO). Slides used for orientation, linear measurement, and inflammatory cell analysis were stained with H&E.

Two sections 120 μm apart, representing the central part of the defect from each tooth block, were analyzed. Measurements from two tissue sections were averaged to provide values for each tooth block. These measurements included bone resorption, extend of the apical migration of the epithelium (epithelial downgrowth), and new cementum formation, as depicted in Fig. 3. Bone resorption was determined by quantifying the distance between CRJ and the coronal level of alveolar crest. Epithelial downgrowth was determined by measuring the distance between CRJ and the apical extend of the junctional epithelium. New cementum formation was determined by quantifying new cementum above notch.

The same sections used in linear measurements were analyzed for inflammatory cell numbers. Images were captured at ×400 magnification. An area of interest was defined as a 1-mm2 box in deep gingival connective tissue in close proximity to bone on the palatal side of each tooth. Based on our experience, this area is critical for the destructive and repair process of the tooth-supporting apparatus in periodontal wound healing, as it is the location in which cellular changes are mostly expected. The apical boundary was set at the level of the notch, and the coronal boundary 1 mm above notch. The lateral boundary was set 1 mm away from the palatal root on one side and extended 1 mm palatal into the gingival space. All cell analyses were performed in this box drawn in all specimens. Specimens with unidentifiable landmarks were not analyzed. Based on a fixed and reproducible landmark (notch), such cell analysis provided means to examine temporal changes occurring in the same tissue area on all the sections. A minimum of five fields (i.e., upper left corner, upper right corner, lower left corner, lower right corner, and center) from each area of interest was quantified. Inflammatory cells included mononuclear and polymorphonuclear leukocytes. Mononuclear leukocytes included lymphocytes, macrophages, monocytes, or plasma cells. Polymorphonuclear leukocytes were identified by their multilobed nucleus. Inflammatory cell numbers were determined by measuring inflammatory cell numbers per mm2 connective tissue.

Osteoclasts were detected using TRAP (43) approach and recognized as positive staining on large multinucleated cells directly lining the bone surface in Howship’s lacunae. Osteoclast numbers were determined by quantifying number of osteoclasts per mm of bone surface.

Inflammatory cell apoptosis was detected by the TUNEL technique (44) using a TdT-blue label detection kit, performed according to the manufacturer’s recommendations. The data are presented as number of apoptotic cells per 100 inflammatory cells.

For each parameter, groups injected with vehicle alone were compared with those injected with soluble receptors at each indicated time point using Student’s t test with SPSS software (SPSS, Chicago, IL). Statistical significance was defined as p < 0.05.

The wound-healing parameters (epithelial downgrowth, bone resorption, and new cementum formation), deep gingival tissue inflammatory cell recruitment, number of osteoclasts, and inflammatory cell apoptosis were examined at days 0, 3, 14, and 35. It should be noted that the larger the value of the epithelial attachment, the worse the outcome of periodontal wound healing, because by migrating on the root surface the epithelium prevents the connective tissue from newly attaching on the root surface. As shown in Fig. 4,A, epithelial attachment did not improve considerably during healing. There was no obvious bone resorption observed at days 0 and 3, consistent with virtually no osteoclast at days 0 and 3. Bone resorption peaked at day 14, consistent with an elevated number of osteoclasts observed at the same time point. The net bone resorption was found improved at day 35 (Fig. 4,B), consistent with fewer osteoclast numbers at the same time point (Fig. 5,C). New cementum formation was observed as early as day 14 and peaked at day 35 (Fig. 4,C). Given that before the periodontal treatment all animals were left to accumulate microorganisms to develop periodontal disease, the number of inflammatory cells in deep gingival connective tissue evaluated at day 0 was relatively high (Fig. 5,A), consistent with a chronic inflammation typically found in periodontitis with low inflammatory cell apoptosis at day 0. Peak number of inflammatory cells in the same compartment was observed at day 3, consistent with an acute inflammation observed after the inception of wound healing. Elevated inflammatory cell apoptosis was observed at this time point (Fig. 5,D). Thereafter, a time-dependent decrease in the number of inflammatory cells was observed (Fig. 5,A). The number of mononuclear inflammatory cells in deep gingival connective tissue was elevated at day 0 (Fig. 5 B), and time-dependent decrease of this parameter was observed.

FIGURE 4.

Periodontal wound healing causes time-dependent changes in epithelial downgrowth (A), bone resorption (B), and new cementum formation (C). Epithelial downgrowth and bone resorption as well as new cementum formation were measured with the aid of an image analysis system on H&E-stained sections. Epithelial downgrowth was obtained by quantifying the distance between CRJ and the apical level of junctional epithelium. Bone resorption was obtained by quantifying the distance between CRJ and the coronal level of alveolar crest. New cementum formation represents the levels of new cementum above notch. Maxillary blocks were harvested at days 0, 3, 14, and 35. Epithelial downgrowth and bone resorption peaked at day 14; new cementum formation was observed as early as day 14 and peaked at day 35.

FIGURE 4.

Periodontal wound healing causes time-dependent changes in epithelial downgrowth (A), bone resorption (B), and new cementum formation (C). Epithelial downgrowth and bone resorption as well as new cementum formation were measured with the aid of an image analysis system on H&E-stained sections. Epithelial downgrowth was obtained by quantifying the distance between CRJ and the apical level of junctional epithelium. Bone resorption was obtained by quantifying the distance between CRJ and the coronal level of alveolar crest. New cementum formation represents the levels of new cementum above notch. Maxillary blocks were harvested at days 0, 3, 14, and 35. Epithelial downgrowth and bone resorption peaked at day 14; new cementum formation was observed as early as day 14 and peaked at day 35.

Close modal
FIGURE 5.

Periodontal wound healing causes time-dependent changes in deep gingival connective tissue inflammatory cell recruitment (A), deep gingival connective tissue mononuclear inflammatory cell recruitment (B), osteoclast formation (C), and deep gingival connective tissue inflammatory cell apoptosis (D). All cell counts were quantified with the aid of an image analysis system. Inflammatory cells are reported as number of inflammatory cells per mm2 of connective tissue on H&E-stained sections in the box area, as described in Materials and Methods. Osteoclasts were counted on TRAP-stained sections. The data are reported as number of osteoclasts per millimeter length of bone. Inflammatory cell apoptosis was quantified on TUNEL-stained sections. Nuclei of the positive cells were stained dark blue. The data are presented as percentage of apoptotic cells in deep gingival connective tissue. Peak number of inflammatory cells in deep gingival connective tissue was observed at day 3, along with a peak in inflammatory cell apoptosis. Thereafter, a time-dependent decrease in the number of inflammatory cells was observed. The number of mononuclear inflammatory cells in deep gingival connective tissue was elevated at day 0, and time-dependent decrease of this parameter was observed. Note the elevated number of osteoclasts at day 14.

FIGURE 5.

Periodontal wound healing causes time-dependent changes in deep gingival connective tissue inflammatory cell recruitment (A), deep gingival connective tissue mononuclear inflammatory cell recruitment (B), osteoclast formation (C), and deep gingival connective tissue inflammatory cell apoptosis (D). All cell counts were quantified with the aid of an image analysis system. Inflammatory cells are reported as number of inflammatory cells per mm2 of connective tissue on H&E-stained sections in the box area, as described in Materials and Methods. Osteoclasts were counted on TRAP-stained sections. The data are reported as number of osteoclasts per millimeter length of bone. Inflammatory cell apoptosis was quantified on TUNEL-stained sections. Nuclei of the positive cells were stained dark blue. The data are presented as percentage of apoptotic cells in deep gingival connective tissue. Peak number of inflammatory cells in deep gingival connective tissue was observed at day 3, along with a peak in inflammatory cell apoptosis. Thereafter, a time-dependent decrease in the number of inflammatory cells was observed. The number of mononuclear inflammatory cells in deep gingival connective tissue was elevated at day 0, and time-dependent decrease of this parameter was observed. Note the elevated number of osteoclasts at day 14.

Close modal

All changes in epithelial attachment, bone height, and cementum formation were determined relative to baseline control (day 0). For epithelial attachment, positive value represents decrease of epithelial downgrowth (improvement in wound healing outcome), while negative value represents increase of epithelial downgrowth (deterioration of wound-healing outcome). For bone height, positive value represents bone formation, while negative value represents bone resorption. For cementum formation, positive value represents cementum deposition. Fourteen days after surgery, the epithelial attachment was found 1.11 mm shorter in the group injected with soluble receptors compared with the untreated group (p < 0.05) (Fig. 6,A). Moreover, bone height was ≅1 mm greater in the treated group compared with the untreated group at day 14 (p < 0.05) (Fig. 6,B). Consistent with this finding, a 90% reduction of osteoclast numbers in animals that received local injection of IL-1 and TNF soluble receptors compared with the untreated animals (p < 0.05) (Fig. 7,C). Finally, more new cementum formation was observed in the treated group compared with the untreated group, but here the difference was not statistically significant (Fig. 6 C).

FIGURE 6.

Treatment with IL-1 and TNF antagonists reduced epithelial attachment at day 14, but increased it at day 35 (A). IL-1 and TNF antagonists inhibited periodontal bone loss at day 14 (B). Treatment with IL-1 and TNF antagonists decreased new cementum formation at day 35 (C). All changes in epithelial attachment, bone height, and cementum formation were determined relative to baseline control (day 0). The reduction of the epithelial attachment at day 14 and the increase of this parameter at day 35 were statistically significant (p < 0.05). Statistically significant reduction in bone loss was observed (p < 0.05) at day 14. The reduction in new cementum formation was statistically significant (p < 0.05) at day 35. Error bars represent SEM. ∗, Statistically significant at p < 0.05.

FIGURE 6.

Treatment with IL-1 and TNF antagonists reduced epithelial attachment at day 14, but increased it at day 35 (A). IL-1 and TNF antagonists inhibited periodontal bone loss at day 14 (B). Treatment with IL-1 and TNF antagonists decreased new cementum formation at day 35 (C). All changes in epithelial attachment, bone height, and cementum formation were determined relative to baseline control (day 0). The reduction of the epithelial attachment at day 14 and the increase of this parameter at day 35 were statistically significant (p < 0.05). Statistically significant reduction in bone loss was observed (p < 0.05) at day 14. The reduction in new cementum formation was statistically significant (p < 0.05) at day 35. Error bars represent SEM. ∗, Statistically significant at p < 0.05.

Close modal
FIGURE 7.

IL-1 and TNF antagonists reduced inflammatory cell infiltrate in deep gingival connective tissue at day 14, but a substantial increase in inflammatory cells was observed at day 35. IL-1 and TNF antagonists inhibited osteoclast formation at day 14, and increased inflammatory cell apoptosis in deep gingival connective tissue at day 14. Deep gingival connective tissue mononuclear inflammatory cell and total inflammatory cell recruitment are reported relative to the baseline control (day 0) as number of inflammatory cells per mm2 of connective tissue in the box described in Materials and Methods. Number of osteoclasts and deep gingival connective tissue inflammatory cell apoptosis were quantified, as described in Materials and Methods. Differences in total inflammatory cells (A) and mononuclear cells (B) between treated and untreated group were statistically significant at p < 0.05 for both the day 14 and the day 35 time points. Statistically significant reduction in osteoclast number (C) was observed (p < 0.05) at day 14. Statistically significant increase in inflammatory cell apoptosis (D) was observed (p < 0.05) between treated and untreated animals at day 14. Error bars represent SEM. ∗, Statistically significant at p < 0.05.

FIGURE 7.

IL-1 and TNF antagonists reduced inflammatory cell infiltrate in deep gingival connective tissue at day 14, but a substantial increase in inflammatory cells was observed at day 35. IL-1 and TNF antagonists inhibited osteoclast formation at day 14, and increased inflammatory cell apoptosis in deep gingival connective tissue at day 14. Deep gingival connective tissue mononuclear inflammatory cell and total inflammatory cell recruitment are reported relative to the baseline control (day 0) as number of inflammatory cells per mm2 of connective tissue in the box described in Materials and Methods. Number of osteoclasts and deep gingival connective tissue inflammatory cell apoptosis were quantified, as described in Materials and Methods. Differences in total inflammatory cells (A) and mononuclear cells (B) between treated and untreated group were statistically significant at p < 0.05 for both the day 14 and the day 35 time points. Statistically significant reduction in osteoclast number (C) was observed (p < 0.05) at day 14. Statistically significant increase in inflammatory cell apoptosis (D) was observed (p < 0.05) between treated and untreated animals at day 14. Error bars represent SEM. ∗, Statistically significant at p < 0.05.

Close modal

Although the group treated with soluble IL-1/TNF receptors clearly exhibited better healing at day 14, this benefit was entirely lost by day 35. At this time point, epithelial attachment was found 1.10 mm longer in the treated group compared with the placebo-treated group (p < 0.05) (Fig. 6,A). Moreover, bone height was 0.47 mm lesser in the treated group compared with the untreated group (Fig. 6,B). Consistent with this finding, the osteoclast numbers were greater in animals that received IL-1 and TNF soluble receptors compared with untreated animals, but the difference was not statistically significant (Fig. 7,C). Finally, a 98% decrease in new cementum formation was observed in the group treated with IL-1 and TNF soluble receptors when compared with the group treated with vehicle alone (p < 0.05) (Fig. 6 C).

Three days after surgery, the animals that had been treated briefly with soluble receptors had similar numbers of inflammatory cells in deep gingival connective tissue compared with the untreated animals. Fourteen days after surgery, a 47% reduction in inflammatory cells was observed in the group injected with soluble receptors compared with the untreated group (p < 0.05). This reduction was observed mostly in the mononuclear cell segment (Fig. 7, A and B). However, 35 days after surgery, there was a striking increase in inflammatory cell counts in deep gingival connective tissue in the treated group, whereas treated animals exhibited 2.65-fold increased inflammatory cells compared with untreated animals (p < 0.05) at the same time point. This shift in inflammatory cell recruitment resulted mostly from mononuclear cells (Fig. 7, A and B). Comparison of polymorphonuclear cells from treated and untreated animals in each time group reflected similar change as seen for mononuclear cells, but here the differences were not statistically significant (data not shown). Fig. 8 depicts representative histologic sections used for the analysis.

FIGURE 8.

Histologic pictures representative of the tissue sections used in the analysis. Baseline control animals showed inflammatory cell infiltrate in deep gingival tissue on H&E-stained sections (A and B), consistent with experimental periodontitis. Treatment with IL-1 and TNF antagonists reduced inflammatory cell infiltration in deep gingival tissue at day 14, but increased it at day 35 after surgery. At day 14, sections from animals receiving vehicle alone showed substantial increase in inflammatory cell infiltrate in deep gingival connective tissue on H&E-stained sections (C and D), while sections from animals receiving IL-1/TNF soluble receptors showed a reduced inflammatory cell infiltrate (E and F). This pattern was reversed at day 35, with treated animals (I and J) exhibiting greater inflammatory cell infiltrate than untreated animals (G and H) or earlier time point animals (A–H). Insets drawn on A, C, E, G, and I show the areas magnified in B, D, F, H, and J, respectively. B, bone; CT, connective tissue. Original magnification: A, C, E, G, and I, ×100; B, D, F, H, and J, ×400.

FIGURE 8.

Histologic pictures representative of the tissue sections used in the analysis. Baseline control animals showed inflammatory cell infiltrate in deep gingival tissue on H&E-stained sections (A and B), consistent with experimental periodontitis. Treatment with IL-1 and TNF antagonists reduced inflammatory cell infiltration in deep gingival tissue at day 14, but increased it at day 35 after surgery. At day 14, sections from animals receiving vehicle alone showed substantial increase in inflammatory cell infiltrate in deep gingival connective tissue on H&E-stained sections (C and D), while sections from animals receiving IL-1/TNF soluble receptors showed a reduced inflammatory cell infiltrate (E and F). This pattern was reversed at day 35, with treated animals (I and J) exhibiting greater inflammatory cell infiltrate than untreated animals (G and H) or earlier time point animals (A–H). Insets drawn on A, C, E, G, and I show the areas magnified in B, D, F, H, and J, respectively. B, bone; CT, connective tissue. Original magnification: A, C, E, G, and I, ×100; B, D, F, H, and J, ×400.

Close modal

The number of apoptotic inflammatory cells in deep gingival connective tissue is shown in Fig. 7,D. Representative histologic sections from these animals are shown in Fig. 9. Three days after surgery, there was no discernible difference between treated and untreated animals in inflammatory cell apoptosis. Fourteen days after surgery, a 5.7-fold increase in inflammatory cell apoptosis was observed in the group injected with soluble receptors compared with the untreated group (p < 0.05) (Figs. 7,D and 9). Thirty-five days after surgery, inflammatory cell apoptosis was observed to be 38% less in the group injected with soluble receptors compared with the untreated group (Figs. 7,D and 9).

FIGURE 9.

Representative histologic sections illustrating that treatment with IL-1 and TNF antagonists increased inflammatory cell apoptosis in deep gingival connective tissue on periodontal wound healing at day 14 after surgery. TUNEL-stained sections are shown from treated (A) and untreated animals (B). Original magnification: ×400.

FIGURE 9.

Representative histologic sections illustrating that treatment with IL-1 and TNF antagonists increased inflammatory cell apoptosis in deep gingival connective tissue on periodontal wound healing at day 14 after surgery. TUNEL-stained sections are shown from treated (A) and untreated animals (B). Original magnification: ×400.

Close modal

In this study, a nonhuman primate model was used to assess the effect of jointly inhibiting IL-1 and TNF on the periodontal wound-healing process. Our study demonstrates that short-term application (14 days) of inhibitors of IL-1/TNF activity led to a significant improvement of periodontal wound healing, while longer-term (35 days) application of these agents had deleterious effects on the wound-healing outcomes.

In some human clinical trials, IL-1 or TNF antagonists alone do not appear to be as efficacious as expected, based on findings from animal studies. One potential explanation for this is that IL-1 and TNF are strongly synergistic (45, 46, 47). Therefore, we have used a combination of TNF and IL-1 inhibitors in an attempt to interfere with both cytokines and to reduce their potential redundancies of activity. We have previously reported that in a nonhuman primate model, the use of soluble receptors for IL-1 and TNF prevented the spread of inflammation, which in turn reduced loss of connective tissue and alveolar bone (10, 29, 39, 48). Others have shown that inhibiting both IL-1 and TNF results in a synergistic down-regulation of inflammation and bone loss in the context of rheumatoid arthritis (49). Furthermore, the utility of administering both IL-1 and TNF antagonists has been demonstrated for protection against endotoxemia (50).

Three major therapeutic approaches have been investigated to interfere with the proinflammatory effects of IL-1 or TNF. Neutralizing Abs to IL-1/TNF were found to be beneficial in diseases associated with excessive IL-1/TNF production (51, 52). Clinical trials with mAb against TNF-α that focused on rheumatoid arthritis and Crohn’s disease have demonstrated promising results (53). Second, administration of IL-1 receptor antagonist, the natural antagonist to IL-1, was shown to improve recovery after traumatic injury (25, 54). Third, exogenous soluble receptors to IL-1/TNF can bind to IL-1 and TNF to prevent IL-1/TNF from binding to their receptors on host cells.

These results suggest that IL-1 and TNF have a detrimental effect in the early phase of periodontal wound healing, while they may be beneficial in the long-term. Furthermore, short-term limitation of their effects in vivo allows for an improvement of wound-healing parameter, while long-term application worsens wound healing. This interpretation is consistent with the findings that overproduction of IL-1 and TNF has been demonstrated in periodontal disease etiologies (55). Soluble receptors to IL-1/TNF similar to those used in this study have been shown to significantly reduce tissue destruction and inflammation in experimental periodontitis (10, 29, 39, 48). TNF and IL-1 were shown to sustain, amplify, and prolong inflammation and tissue destruction. This is achieved through activating inflammatory cells to release cytokines such as IL-1 and TNF, and small molecule mediators such as NO (56). In our study, interfering with IL-1 and TNF activity resulted in a significant improvement of periodontal healing outcomes in the animals treated for 14 days with soluble receptors compared with untreated animals.

However, by day 35, this benefit was entirely lost. As shown in Fig. 7, inflammatory cell infiltration in deep gingival connective tissue was significantly greater in the animals treated with soluble receptors than in the untreated animals observed 35 days after surgery. Also at day 35, inflammatory cell apoptosis in deep gingival connective tissue was reduced in animals treated with soluble receptors. These results suggest that IL-1/TNF might have a role in resolving inflammation at a later stage of wound healing. This interpretation is consistent with recent reports that demonstrated that mice could not resolve late-phase inflammation if they were deficient in TNF (57). Also, treatment with anti-TNF-α mAbs was shown to compromise inflammatory cell clearance in the lungs of wild-type mice after infection with Rhodococcus equi (58). Nathan (9) showed that TNF was involved in the late phase of wound healing, in switching from the tissue-damaging mode to the tissue repair-promoting mode after debris and dead cells were no longer present. Another explanation is that IL-1 might play an essential role in antibacterial defense in a challenging environment in which microflora are deleterious to the wound-healing process (17). As a result, the early improvement of periodontal healing has been lost in animals treated long-term with IL-1 and TNF soluble receptors compared with the untreated animals. It is unlikely that long-term injection of IL-1/TNF soluble receptors would stimulate animals to generate Ab and thus obviate the benefits of the inhibitors observed at 14 days (59). The relatively short time frame of this experiment relative to Ab generation rules out this possibility particularly in light of our previous study showing that soluble receptors to IL-1/TNF were still able to reduce inflammatory responses even at the end of a longer study (6 wk) (39). In another published study, etanercept (TNF antagonist) were still very effective at reducing peritoneal lesions of spontaneously occurring active endometriosis in the baboon after 8 wk (60).

TNF-α and IL-1α are capable of inducing osteoclastogenesis (61). As anticipated, short-term applications of the cytokine inhibitors did reduce postsurgical osteoclastogenesis when compared with untreated animals, although longer-term applications showed a trend in the opposite direction, toward increased numbers of osteoclasts in the inhibitor-treated animals. This increase in osteoclasts at day 35 seemed to parallel the increase in mononuclear cell infiltration observed at the same time point, consistent with the observation that mononuclear cell products stimulate osteoclastogenesis (62). The resurgence of the inflammatory cell infiltrate and the deterioration of wound-healing outcomes observed in animals treated for a long-term with the cytokine inhibitors did not appear to be a result of increased bacterial growth, as all the strains tested from oral cultures remained unchanged between treated vs untreated animals (data not shown).

Multiple time points are necessary for studying the periodontal wound-healing process, due to the complexity and dynamic nature of this process. In this study, to capture periodontal wound-healing events, we used three different time points that are important in wound-healing processes: day 3 for the peak of inflammation; day 14 for maximal formation of granulation tissue; and day 35 for tissue remodeling (2). Our data indicate a series of effects of the soluble receptors over the course of periodontal wound healing: within 14 days after the wound was incurred, the inhibition of proinflammatory cytokines was beneficial overall in our model. However, prolonged application of these inhibitors appears not to favor the healing process, rather to be detrimental to the overall outcome. This study supports the concept that modulation of inflammatory process through interfering with IL-1/TNF activity can be potentially beneficial to improve periodontal wound healing if it is used only for a short time, while longer-term usage may not be appropriate, particularly in a challenging environment with a high commensal flora.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by National Institute of Dental Research Grants DE12482 and DE11254.

3

Abbreviations used in this paper: sIL, soluble IL; CRJ, crown-root junction; sTNFR, soluble TNFR; TRAP, tartrate-resistant acid phosphatase.

1
Martin, P..
1997
. Wound healing: aiming for perfect skin regeneration.
Science
276
:
75
.
2
Clark, R..
1995
. Wound repair: overview and general considerations. R. Clark, ed.
The Molecular and Cellular Biology of Wound Repair
1
. Plenum Press, New York.
3
Wikesjo, U. M., K. A. Selvig.
1999
. Periodontal wound healing and regeneration.
Periodontol.
2000 19
:
21
.
4
Karring, T., S. Nyman, J. Gottlow, L. Laurell.
1993
. Development of the biological concept of guided tissue regeneration: animal and human studies.
Periodontol.
2000 1
:
26
.
5
Werfully, S., G. Areibi, M. Toner, J. Bergquist, J. Walker, S. Renvert, N. Claffey.
2002
. Tensile strength, histological and immunohistochemical observations of periodontal wound healing in the dog.
J. Periodontal Res.
37
:
366
.
6
Leibovich, S. J., P. J. Polverini, H. M. Shepard, D. M. Wiseman, V. Shively, N. Nuseir.
1987
. Macrophage-induced angiogenesis is mediated by tumor necrosis factor-α.
Nature
329
:
630
.
7
March, C. J., B. Mosley, A. Larsen, D. P. Cerretti, G. Braedt, V. Price, S. Gillis, C. S. Henney, S. R. Kronheim, K. Grabstein, et al
1985
. Cloning, sequence and expression of two distinct human interleukin-1 complementary DNAs.
Nature
315
:
641
.
8
Beutler, B., A. Cerami.
1986
. Cachectin and tumor necrosis factor as two sides of the same biological coin.
Nature
320
:
584
.
9
Nathan, C..
2002
. Points of control in inflammation.
Nature
420
:
846
.
10
Graves, D. T., A. J. Delima, R. Assuma, S. Amar, T. Oates, D. Cochran.
1998
. Interleukin-1 and tumor necrosis factor antagonists inhibit the progression of inflammatory cell infiltration toward alveolar bone in experimental periodontitis.
J. Periodontol.
69
:
1419
.
11
Meghji, S., W. Qureshi, B. Henderson, M. Harris.
1996
. The role of endotoxin and cytokines in the pathologenesis of odontogenic cysts.
Arch. Oral Biol.
41
:
523
.
12
Dinarello, C. A..
1996
. Biologic basis of interleukin-1 in disease.
Blood
87
:
2095
.
13
Wustrow, T. P..
1994
. Biology of interleukin-1 (IL-1) with respect to otorhinolaryngology: head and neck surgery.
Head Neck
16
:
88
.
14
Fortunato, S. J., R. Menon.
2003
. IL-1β is a better inducer of apoptosis in human fetal membranes than IL-6.
Placenta
24
:
922
.
15
Awane, M., P. G. Andres, D. J. Li, H. C. Reieaker.
1999
. NF-κB-induced kinase is a common mediatior of IL-17, TNF-α, and IL-1β-induced chemokine promoter activation in intestinal epithelial cells.
J. Immunol.
162
:
5337
.
16
Dinarello, C. A..
1993
. Modalities for reducing interleukin-1 activity in disease.
Immunol. Today
14
:
260
.
17
Graves, D. T., N. Nooh, T. Gillen, M. Davey, S. Patel, D. Cottrell, S. Amar.
2001
. IL-1 plays a critical role in oral, but not dermal, wound healing.
J. Immunol.
167
:
5316
.
18
Angele, M. K., M. W. Knoferl, A. Ayala, J. E. Albina, W. G. Cioffi, K. I. Bland, I. H. Chaudry.
1999
. Trauma-Hemorrhage delays wound healing potentially by increasing proinflammatory cytokines at the wound site.
Surgery
126
:
279
.
19
Trengove, N. J., H. Bielefeldt-Ohmann, M. C. Stacey.
2000
. Mitogenic activity and cytokine levels in non-healing and healing chronic leg ulcers.
Wound Repair Regen.
8
:
13
.
20
Colotta, F., S. K. Dower, J. E. Sims, A. Mantovani.
1994
. The type II “decoy” receptor: a novel regulatory pathway for interleukin-1.
Immunol. Today
15
:
562
.
21
Giri, J. G., R. C. Newton, R. Horuk.
1990
. Identifications of soluble interleukin-1 binding protein in cell-free supernatants.
J. Biol. Chem.
265
:
17416
.
22
Dinarello, C. A., S. M. Wolff.
1993
. The role of IL-1 in diseases.
N. Engl. J. Med.
328
:
106
.
23
Delima, A. J., S. Karatzas, S. Amar, D. T. Graves.
2002
. Inflammation and tissue loss caused by periodontal pathogens is reduced by interleukin-1 antagonists.
J. Infect. Dis.
186
:
511
.
24
Rosenbaum, J. T., R. S. Boney.
1991
. Use of a soluble interleukin-1 receptor to inhibit ocular inflammation.
Curr. Eye Res.
10
:
1137
.
25
Tehranian, R., S. Andell-Jonsson, S. M. Beni, I. Yatsiv, E. Shohami, T. Bartfai, J. Lundkvist, K. Iverfeldt.
2002
. Improved recovery and delayed cytokine induction after closed head injury in mice with central overexpression of the secreted isoform of the interleukin-1 receptor antagonist.
J. Neurotrauma
19
:
939
.
26
Kohno, T., M. T. Brewer, S. L. Baker, P. E. Schwartz, M. W. King, K. K. Hale, C. H. Squires, R. C. Thompson, J. L. Vannice.
1990
. A second tumor necrosis factor receptor gene product can shed a naturally occurring tumor necrosis factor inhibitor.
Proc. Natl. Acad. Sci. USA
87
:
8331
.
27
Declercq, W., G. Denecker, W. Fiers, P. Vandenabeele.
1998
. Cooperation of both TNF receptors in inducing apoptosis: involovement of the TNF receptor-associated factor binding domain of the TNF receptor 75.
J. Immunol.
161
:
390
.
28
Beg, A. A., D. Baltimore.
1996
. An essential role for NF-κB in preventing TNF-α induced cell death.
Science
274
:
782
.
29
Graves, D. T..
1999
. The potential role of chemokines and inflammatory cytokines in periodontal disease progression.
Clin. Infect. Dis.
28
:
482
.
30
Feldmann, M., F. M. Brennan, M. J. Elliot, R. O. Williams, R. N. Maini.
1995
. TNFα is a therapeutic target for rheumatoid arthritis.
Ann. NY Acad. Sci.
766
:
272
.
31
Raine, C. S..
1995
. Multiple sclerosis: TNF revisited, with promise.
Nat. Med.
1
:
211
.
32
Rapala, K..
1996
. The effect of tumor necrosis factor-α on wound healing: an experimental study.
Ann. Chir. Gynaecol. Suppl.
211
:
1
.
33
Mallett, S., A. N. Barclay.
1991
. A new superfamily of cell surface proteins related to the nerve growth factor receptor.
Immunol. Today
12
:
220
.
34
Mori, R., T. Kondo, T. Ohshima, Y. Ishida, N. Mukaida.
2002
. Accelerated wound healing in tumor necrosis factor receptor p55-deficient mice with reduced leukocyte infiltration.
FASEB J.
16
:
963
.
35
Zhang, M. H., K. J. Tracey.
1998
. Tumor necrosis factor. A. W. Thomson, ed.
The Cytokine Handbook
517
. Academic Press, San Diego.
36
Kremer, J. M., M. E. Weinblatt, A. D. Bankhurst, K. J. Bulpitt, R. M. Fleischmann, C. G. Jackson, K. M. Atkins, A. Feng, D. J. Burge.
2003
. Etanercept added to background methotrexate therapy in patients with rheumatoid arthritis: continued observations.
Arthritis Rheum.
48
:
1493
.
37
Gorman, J. D., K. E. Sack, J. C. Davis, Jr.
2002
. Treatment of ankylosing spondylitis by inhibition of tumor necrosis factor α.
N. Engl. J. Med.
346
:
1349
.
38
Leonardi, C. L., J. L. Powers, R. T. Matheson, B. S. Goffe, R. Zitnik, A. Wang, A. B. Gottlieb, Etanercept Psoriasis Study Group.
2003
. Etanercept as monotherapy in patients with psoriasis.
N. Engl. J. Med.
349
:
2014
.
39
Assuma, R., T. Oates, D. Cochran, S. Amar, D. T. Graves.
1998
. IL-1 and TNF antagonists inhibit the inflammatory response and bone loss in experimental periodontitis.
J. Immunol.
160
:
403
.
40
Caton, J., L. Mota, L. Gandini, B. Laskaris.
1994
. Non-human primate models for testing efficacy and safety of periodontal regeneration procedures.
J. Periodontol.
65
:
1143
.
41
Amar, S., K. M. Chung, S. H. Nam, S. Karatzas, F. Myokai, T. E. Van Dyke.
1997
. Markers of bone and cementum formation accumulate in tissues regenerated in peirodontal defects treated with expanded polytetrafluoroethylene membranes.
J. Periodontal Res.
32
:
148
.
42
Karatzas, S., A. Zavras, D. Greenspan, S. Amar.
1999
. Histologic observations of periodontal wound healing after treatment with PerioGlas in nonhuman primates.
Int. J. Periodontics Restorative Dent.
19
:
489
.
43
Chiang, C. Y., G. Kyritsis, D. T. Graves, S. Amar.
1999
. Interleukin-1 and tumor necrosis factor activities partially account for calvarial bone resorption induced by local injection of lipopolysaccharide.
Infect. Immun.
67
:
4231
.
44
Gavrieli, Y., Y. Sherman, S. A. Ben-Sasson.
1992
. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation.
J. Cell Biol.
119
:
493
.
45
Kosaka, H., N. Harada, M. Watanabe, H. Yoshihara, Y. Katsuki, T. Shiga.
1992
. Synergistic stimulation of nitric oxide hemoglobin production in rats by recombinant interleukin 1 and tumor necrosis factor.
Biochem. Biophys. Res. Commun.
189
:
392
.
46
Kumar, A., V. Thota, L. Dee, J. Olson, E. Uretz, J. E. Parrillo.
1996
. Tumor necrosis factor α and interleukin 1β are responsible for in vitro myocardial cell depression induced by human septic shock serum.
J. Exp. Med.
183
:
949
.
47
Caldwell, J., S. G. Emerson.
1995
. Interleukin-1α upregulates tumor necrosis factor receptors expressed by a human bone marrow stromal cell strain: implications for cytokine redundancy and synergy.
Blood
86
:
3364
.
48
Delima, A. J., T. Oates, R. Assuma, Z. Schwartz, D. Cochran, S. Amar, D. T. Graves.
2001
. Soluble antagonists to interleukin-1 (IL-1) and tumor necrosis factor (TNF) inhibit loss of tissue attachment in experimental periodontitis.
J. Clin. Periodontol.
28
:
233
.
49
Kim, S. H., E. R. Lechman, S. Kim, J. Nash, T. J. Oligino, P. D. Robbins.
2002
. Ex vivo gene delivery of IL-1Ra and soluble TNF receptor confers a distal synergistic therapeutic effect in antigen-induced arthritis.
Mol. Ther.
6
:
591
.
50
Ulich, T. R., E. S. Yi, S. Yin, C. Smith, D. Remick.
1994
. Intratracheal administration of endotoxin and cytokines. VII. The soluble interleukin-1 receptor and the soluble tumor necrosis factor receptor II (p80) inhibit acute inflammation.
Clin. Immunol. Immunopathol.
72
:
137
.
51
Kagari, T., T. Doi, T. Shimozato.
2002
. The importance of IL-1β and TNF-α, and the noninvolvement of IL-6, in the development of monoclonal antibody-induced arthritis.
J. Immunol.
169
:
1459
.
52
Kadokami, T., C. Frye, B. Lemster, C. L. Wagner, A. M. Feldman, C. F. McTiernan.
2001
. Anti-tumor necrosis factor-α antibody limits heart failure in a transgenic model.
Circulation
104
:
1094
.
53
Nahar, I. K., K. Shojania, C. A. Marra, A. H. Alamgir, A. H. Anis.
2003
. Infliximab treatment of rheumatoid arthritis and Crohn’s disease.
Ann. Pharmacother.
37
:
1256
.
54
Scripter, J. L., J. Ko, K. Kow, A. Arimura, C. F. Ide.
1997
. Regulation by interleukin-1β of formation of a line of delimiting atrocytes following prenatal trauma to the brain of the mouse.
Exp. Neurol.
145
:
329
.
55
Graves, D. T., D. Cochran.
2003
. The contribution of interleukin-1 and tumor necrosis factor to periodontal tissue destruction.
J. Periodontol.
74
:
391
.
56
Tracey, K. J..
2002
. The inflammatory reflex.
Nature
420
:
853
.
57
Marino, M. W., A. Dunn, D. Grail, M. Inglese, Y. Noguchi, E. Richards, A. Jungbluth, H. Wada, M. Moore, B. Williamson, et al
1997
. Characterization of tumor necrosis factor-deficient mice.
Proc. Natl. Acad. Sci. USA
94
:
8093
.
58
Kanaly, S. T., M. Nashleanas, B. Hondowicz, P. Scott.
1999
. TNF receptor p55 is required for elimination of inflammatory cells following control of intracellular pathogens.
J. Immunol.
163
:
3883
.
59
Dayer-Metroz, M. D., D. Duhamel, N. Rufer.
1992
. IL-1Ra delays the spontaneous autoimmune diabetes in the BB rat.
Eur. J. Clin. Invest.
22
:
A50
.
60
Barrier, B. F., G. W. Bates, M. M. Leland, D. A. Leach, R. D. Robinson, A. M. Propst.
2004
. Efficacy of anti-tumor necrosis factor therapy in the treatment of spontaneous endometriosis in baboons.
Fertil. Steril.
81
:
775
.
61
Kudo, O., Y. Fujikawa, I. Itonaga, A. Sabokbar, T. Torisu, N. A. Athanasou.
2002
. Proinflammatory cytokine (TNFα/IL-1α) induction of human osteoclast formation.
J. Pathol.
198
:
220
.
62
Kobayashi, K., N. Takahashi, E. Jimi, N. Udagawa, M. Takami, S. Kotake, N. Nakagawa, M. Kinosaki, K. Yamaguchi, N. Shima, et al
2000
. Tumor necrosis factor α stimulates osteoclast differentiation by a mechanism independent of the ODF/RANKL-RANK interaction.
J. Exp. Med.
191
:
275
.