Resolvin E1 Regulates Inflammation at the Cellular and Tissue Level and Restores Tissue Homeostasis In Vivo¹

Inflammation is the normal host tissue response to infection and injury. However, uncontrolled and unresolved inflammation contributes to a range of acute and chronic human diseases such as arthritis and cardiovascular diseases. Chronic inflammation is characterized by the production of inflammatory cytokines, arachidonic acid-derived eicosanoids (prostaglandins, thromboxanes, leukotrienes, and other oxidative derivates), reactive oxygen species, and adhesion molecules (1, 2, 3). Periodontitis is a similar progressive inflammatory disease in which microbial etiologic factors induce an inflammatory cascade that leads to destruction of the organ supporting the teeth (periodontium), including soft tissues and bone (4, 5, 6, 7). The very nature of periodontitis, being in the oral cavity and therefore easily observable, led to the use of periodontal diseases as a model system for other inflammatory diseases (8, 9, 10). As with all inflammatory diseases, the arachidonate-derived eicosanoids play a key role in the initiation and pathogenesis of the inflammatory lesion in periodontitis (11, 12, 13). In the case of periodontitis, the etiologic Gram-negative bacteria, such as Porphyromonas gingivalis, initiate an influx of neutrophils and neutrophil cyclooxygenase-2 activation leading to increased PGE₂ in situ (14). Indeed, many of the early pathophysiologic events in periodontal diseases and its chronicity can be attributed to lipid mediators (15). Leukotriene B₄ (LTB₄),³ produced mainly by activated leukocytes, initiates accumulation and superoxide generation by neutrophils within inflamed sites, stimulating the release of granule-associated enzymes and bone resorption (16, 17). PGE₂ is a potent activator of osteoclast-mediated bone resorption, the hallmark of periodontal disease (13, 18) and, with other eicosanoids, mediates inflammation and periodontal tissue destruction (19, 20).

Resolution of inflammation is an actively regulated program rather than the passive termination of inflammation (21, 22, 23, 24). The crucial identification of the cellular events and molecular signals that determine the end of inflammation and beginning of resolution has lead to a new appreciation of pathogenesis in inflammatory diseases (25, 26, 27, 28). Neutrophils are present mainly in inflamed or injured tissues and their effective elimination is a prerequisite for complete resolution of an inflammatory response (29). Most current therapeutic approaches attempt to block activation of inflammation using anti-inflammatory drugs (nonsteroidal anti-inflammatory drugs, TNF inhibitors), or to promote healing with agents such as TGF-β1, bridging molecules, and phagocyte receptors (30). Prostaglandins and leukotrienes play essential roles in orchestrating inflammation and are well appreciated autacoids or local-acting mediators (17). Cyclooxygenase inhibitors are widely used examples of anti-inflammatory drugs that act by blocking prostaglandin biosynthesis (31) but can be toxic to resolution programs in vivo because their development anteceded the recognition of resolution as an active process (21, 22, 32).

A rapidly emerging body of evidence demonstrates that endogenous mediators actively participate in dampening host responses to orchestrate resolution of inflammation (33, 34). Lipoxins, the product of lipoxygenase: lipoxygenase interactions, actively drive resolution of inflammation. In addition, the role of previously unappreciated aspirin-triggered transformation circuits has led to a better understanding of proresolution signaling networks, including a series of complex cellular and chemical reactions, and tissue trafficking events (35). For example, lipoxins not only reduce influx of neutrophils, but also stimulate the nonphlogistic uptake of apoptotic neutrophils by tissue macrophages (21, 36).

Omega-3 polyunsaturated fatty acids (ω-3 PUFAs) are well known to decrease the production of inflammatory eicosanoids, cytokines and reactive oxygen species, and the expression of adhesion molecules (37). However, the molecular basis remains unknown. An ω-3 PUFA, eicosapentaenoic acid (EPA), is metabolized by aspirin-modified cyclooxygenase-2 to form a novel small molecule that promotes resolution of inflammation termed resolvin E1 (RvE1: 5S,12R,18R-trihydroxy-6Z,8E,10E,14Z,16E-eicosapentaenoic acid) (38). In murine peritonitis, specific molecular circuits/pathways are activated by RvE1 (39), which reduce the maximal number of neutrophils in the exudate and initiate early resolution (40). In rabbit periodontitis, RvE1 prevents the onset and progression of P. gingivalis-induced periodontitis by >95% confirming the inflammatory nature of periodontitis (10). RvE1 signals via interactions with the LTB₄ receptor BLT1 on human polymorphonuclear neutrophils and the ChemR23 receptor on monocytes regulating migration of leukocytes in acute inflammation (41). RvE1 attenuates LTB₄-dependent proinflammatory signals such as mobilization of intracellular calcium and NF-κB activation (39). Taken together, the findings suggest that a rational approach to therapeutics for inflammatory diseases may involve the use of agonists of proresolution in addition to, or instead of, traditional anti-inflammation approaches that block initiation pathways of inflammation and display unwanted side effects.

The study was approved by Boston University Medical Center (BUMC) Institutional Animal Care and Use Committee before study initiation (IACUC protocol no. AN-13948). In addition, BUMC Institutional Biohazard Committee (IBC) approved the use of P. gingivalis in this animal model to induce periodontal disease (IBC protocol no. 06-016).

New Zealand White rabbits (39 males, 3.5–4.0 kg) were purchased from Pine Acre Rabbit Farms (PARF), kept in individual cages, received water ad libitum, and fed standard rabbit chow at the Laboratory Animal Science Center at BUMC.

P. gingivalis (strain A7436) was grown as previously described (9). In brief, bacteria were cultured on agar plates containing trypticase soy agar supplemented with 0.5% (w/v) yeast extract (Invitrogen Life Technologies), 5% defibrinated sheep RBC, 5 μg of hemin, and 1 μg/ml vitamin K (Sigma-Aldrich). Plates were incubated for 3 days at 37°C in jars anaerobically maintained through palladium-catalyzed hydrogen/carbon dioxide envelopes (GasPak Plus; BD Microbiology Systems). Colonies were randomly selected and anaerobically cultured overnight at 37°C in Schaedler’s broth supplemented with vitamin K and hemin. Bacterial numbers were spectrophotometrically determined at 600 nm, adjusted to 10⁹ CFU (0.8 OD) and mixed with carboxymethylcellulose to form a thick slurry, which was applied topically to the ligated teeth every other day for the first 6 wk of the experiment.

Periodontitis in rabbits was used to monitor inflammatory events associated with bone disorders and the actions of the proresolving agonist, RvE1. Periodontitis was induced and established in all animals for a 6-wk period using a previously established protocol (9, 10). A 3-0 silk suture (ligature) was placed around the second premolar of both mandibular quadrants under general anesthesia (40 mg/kg ketamine, Ketaset (Fort Dodge Animal Health) and 5 mg/kg xylazine, Anased (Ben Venue Laboratories) injections). The slurry containing P. gingivalis was topically applied to the ligatures on Monday, Wednesday, and Friday over a 6-wk period to induce periodontal inflammation. The sutures were checked at every application, and lost or loose sutures were replaced.

At the end of the first 6-wk period, 5 randomly selected animals (group 1) were euthanized to determine the level of the periodontal disease obtained. The remaining animals were randomly assigned to four different groups: group 2 = ethanol (vehicle-alone) treatment (10 animals); group 3 = RvE1 treatment, 4 μg per tooth in ethanol (14 animals); group 4 = LTB₄ treatment, 4 μg per tooth in ethanol (5 animals); group 5 = PGE₂ treatment, 4 μg per tooth in ethanol (5 animals). P.gingivalis application was stopped and treatments initiated for a second 6 wk on the same time schedule with the same anesthesia regimen. The animals euthanized at the end of the first 6-wk period confirmed the periodontal inflammation and associated soft tissue and bone destruction at the disease baseline.

The RvE1 preparation was delivered in ethanol (8 μl) and the same volume of ethanol served as placebo. LTB₄ was purchased as a solution in ethanol (Cayman Chemicals) and delivered to the site via a Hamilton syringe in the same volume as RvE1. Concurrently, PGE₂ was purchased in salt form (Cayman Chemicals) and diluted in ethanol to obtain the same dose as RvE1. At the end of the 6-wk period, animals were euthanized using an overdose of pentobarbital (120 mg/kg Euthanasia-5 solution; Veterinary Laboratories) according to the approved protocol of IACUC.

After sacrificing the animals, the mandible was dissected free of muscle and soft tissue, keeping the attached gingiva intact. The mandible was split into two halves from the midline between the central incisors. Half was taken for morphometric analysis of bone loss, and the other half was used for histological evaluation of periodontitis. For morphometric analysis, the sectioned mandible was defleshed by immersion in 10% hydrogen peroxide followed by careful soft tissue removal, washed with distilled water, air-dried and stained with methylene blue for visual distinction between the tooth and bone. The bone level around the second premolar was measured directly using a 0.5 mm calibrated periodontal probe. The bone level was also quantified using Image Analysis (Image-Pro Plus 4.0; Media Cybernetics). Radiographs were taken with a digital x-ray (Schick Technologies). To quantify bone loss, the length of the tooth from cusp tip to the apex of the root was measured, as was the length of the tooth structure outside the bone that was measured from the cusp tip to the coronal extent of the proximal bone. From this, the individual percentages of the teeth within the bone were calculated (42). Bone values are expressed as the percentage of bone loss.

In addition, the soft tissue (pocket) depth and infrabony defect depth were measured in all groups using a 0.5 mm calibrated periodontal probe. The tip of the tooth at the measured site was used as the reference point for these measurements. Furthermore, tooth mobility was also calculated using Muhleman’s mobility index (43) as follows: 0 = no mobility; 1 = >0.5 and <1 mm mobility buccolingually; 2 = >1 mm but <2 mm mobility buccolingually; 3 = >2 mm mobility buccolingually; 3+ = both vertical and buccolingual mobility.

Half of the mandible was immersed in 10 volumes of Immunocal (Decal Corporation) and the solution was replaced every 24 h for 2 wk. Decalcification was confirmed by serial radiographs, which were taken every other day. After the decalcification, the tissues were rinsed for 1–3 min in running water, placed in Cal-Arrest (Decal Corporation) to bring tissues to neutral pH, which enhanced embedding and staining characteristics, and to terminate further decalcification. Tissues were kept in this solution for 2–3 min, rinsed again in flowing deionized water for at least 3 min and placed in formalin for ∼24 h before embedding in paraffin. Thin sections (5 μm) were cut and either stained with H&E for light microscopy and identification of the cellular composition of inflammatory infiltrates or with tartrate-resistant acid phosphatase (TRAP) to examine osteoclastic activity (44). For qualitative assessments, three areas were analyzed on each tooth corresponding to the coronal, middle, and apical third of the root. In addition, some of the sections per animal were stained with Masson’s trichrome to detect whether new collagen depositions and new bone formation were detectable following treatment (45). Furthermore, some of the sections of each of the groups were stained with osteocalcin as a measure of osteoblastic activity and new bone formation.

For un-decalcified sections, some of the harvested blocks were placed in 4% formalin/1% CaCl₂ fixative (46). The specimens were dehydrated and embedded in methylmethacrylate. Eight to ten un-decalcified sections of ∼10 μm in thickness were obtained from each specimen using a slow-speed diamond saw with coolant and stained superficially with Van Gieson’s trichrome. Sections were analyzed for qualitative histological findings.

To quantify the changes in bone, the mean value (±SD) of the linear distance and the area of bone loss were calculated for each group. Previously developed measurement technique (47) was used to calculate the bone changes at three different sections of the root using the ProImage software. The linear measurements were made at three levels each corresponding to one third of the root and alveolar bone interface: crestal, mid, and apical. Linear distance is reported as the distance from the base of the epithelium to the alveolar crest border at the three chosen levels, the apical, middle, and the coronal third of the root and is expressed as the difference between treated and untreated sites. Likewise, area measurements were presented as the difference between the treated and untreated total area.

In addition, osteoclastogenesis was examined in all TRAP-stained sections by calculating the osteoclasts in affected areas. The total number of osteoclasts at the surface of the bone was compared between the groups.

Microbial dental plaque was sampled at baseline at 6 and 12 wk using paper points. The area was isolated to prevent saliva contamination, air-dried and 30 s samples were collected using sterile paper points according to previously reported methods (48). Each sample was placed in an individual Eppendorf tube containing 0.15 ml TE (10 mM Tris-HCl, 1 mM EDTA (pH 7.6)) and 0.5 M NaOH was added for stabilization. Eighteen species representing periodontal organisms including P. gingivalis, Actinobacillus actinomycetemcomitans, Actinomyces odontolyticus, Actinomyces viscosus, Actinomyces israelii, Peptostreptococcus micros, Prevotella intermedia, Prevotella nigrescens, Capnocytophaga curva, Capnocytophaga rectus, Streptococcus oralis, Streptococcus intermedius, Tannerella forsythensis, Treponema denticola, Eikenella corrodens, Fusobacterium nucleatum, Escherichia coli, and Enterococcus faecalis were investigated in each plaque sample using the checkerboard DNA-DNA hybridization technique (49). Evaluation of the chemiluminescent signals is performed by radiographic detection, comparing the obtained signals with the signals generated by pooled standard comparisons of 10⁴, 10⁵, and 10⁶ of each of the 18 species. Chemiluminescence units are transformed into a scale of 0 to 5: 0, no signal; 1, signal density weaker than the low standard (i.e., undetectable <10⁴ bacteria); 2, signal density equal to the next standard (=10⁵ bacteria); 3, signal density higher than the 10⁵ standard but lower than that of the 10⁶ standard (>10⁵ but <10⁶ bacteria); 4, signal density equal to the high standard (=10⁶ bacteria); and 5, signal density higher than the high standard (>10⁶ bacteria).

Blood samples were obtained from the central ear artery using standard venipuncture techniques from each animal at baseline, 6 and 12 wk. Serum samples were collected and stored at 80°C for analysis of IL-1β and C-reactive protein (CRP). CRP was determined using a rabbit-specific ELISA kit following the manufacturer’s instructions (Immunology Consultants Laboratory). IL-1β was analyzed by ELISA specific for human IL-1β according to the manufacturer (Abcam).

Mean values for linear and area measurements were used to determine the changes in bone level. In addition, TRAP plus stained cell counts were calculated to detect the osteoclastogenesis. Multiple comparisons within groups were made using ANOVA with Bonferroni correction (α = 0.05). Mean values for IL-1β and CRP levels were used to detect the changes in inflammatory markers before and after treatment. Statistical analysis was performed using nonparametric tests. Comparisons between the six study groups were performed using the Kruskal-Wallis test. In case of significant differences, post hoc two-group comparisons were made with the χ² test. The statistical significance was set at p ≤ 0.05.

To examine the potential of proresolving molecules such as RvE1 for the treatment of inflammatory diseases, we used a unique in vivo system to assess both soft and hard tissue destruction, namely rabbit periodontitis (9, 10). A common human periodontal pathogen, P. gingivalis, induced periodontitis. Thirty nine male New Zealand White rabbits were used as described in Materials and Methods. Disease was induced for 6 wk in all groups (Fig. 1). One group was sacrificed at 6 wk to determine baseline disease. The other groups then entered treatment for an additional 6 wk. The treatment arms of the study included RvE1 as a monotherapy; vehicle-alone (95% ethanol) as a placebo control; and two structurally related lipid mediators, LTB₄ and PGE₂, as alternative monotherapies. Results revealed that P. gingivalis induced significant periodontal disease (Fig. 2, A1/B1). The progression of disease was unaffected by placebo therapy, and progression was significantly more severe when either LTB₄ or PGE₂ was used (Fig. 2, A2/B2, A4/B4, and A5/B5, respectively). Topical RvE1 treatment resulted in essentially complete resolution of periodontal inflammation and restoration of both soft and hard tissues clinically (Fig. 2, A3/B3). The irregular, edematous and hyperemic appearance of soft tissues was also resolved, and no clinical signs of inflammation were identifiable in RvE1-treated rabbits. Surprisingly, RvE1 treatment resulted in bone regrowth to pre-disease levels. Periodontal disease progressed from baseline disease (6 wk) in each of the other groups (Fig. 2, A2/B2, A4/B4, and A5/B5).

Quantitative morphologic assessments demonstrated that administration of topical RvE1 (4 μg/tooth) resulted in statistically significant bone regrowth (>95%) compared with baseline periodontitis (Fig. 2, A6). Both the infrabony (vertical) defects and horizontal bone loss were completely restored. Teeth in all groups were mobile at 6 wk due to destruction of periodontal attachment and local inflammation. Mobility became quite severe by 12 wk in all groups except the RvE1 treatment group where the teeth exhibited essentially no mobility at 12 wk.

Radiographic bone levels were calculated using the modified Bjorn technique (42) (Fig. 3). Baseline periodontal disease (6 wk) displayed ∼30% bone loss (Fig. 3, inset). Analyses of radiographic images demonstrated that RvE1 treatment restored the lost bone, whereas those treated with vehicle alone, LTB₄ or PGE₂ progressed with ∼13, 9, and 18% more bone loss, respectively, compared with baseline disease.

In addition, clinical periodontal disease parameters including pocket depth (distance between crest of the gingiva and soft tissue attachment on the tooth surface), infrabony defect depth (vertical dimension of bone loss) and tooth mobility (43) were evaluated (Table I). Pocket depths and infrabony defect depths reflected established periodontitis at 6 wk (baseline periodontitis). Evaluation after a 6-wk treatment phase (at 12 wk) revealed that topical RvE1 treatment resulted in statistically significant decreases in pocket depth and infrabony defect depth when compared with baseline periodontitis and all other treatment groups (p < 0.05). At 12 wk, the other treatment groups (i.e., vehicle-alone, LTB₄, and PGE₂) exhibited significantly greater pocket depth, infrabony defect depth and tooth mobility than at 6 wk, indicating progression of the periodontal disease.

Although the pathogenesis of periodontitis appears to be inflammatory, the etiology is the bacterial biofilm that forms on teeth commonly known as dental plaque. These biofilms appear to release chemoattractant, which stimulates recruitment of neutrophils that, without control, can destroy surrounding tissues. To this end, P. gingivalis was introduced to the rabbit oral microflora as an exogenous pathogen to initiate local inflammation and periodontal disease during the first 6 wk of the study. At 6 wk, addition of the pathogen ceased (Fig. 1). Dental plaque samples were collected at baseline and 6 and 12 wk to determine the persistence of the pathogen and the response of the resident microflora. The rabbit oral microflora comprised anaerobic and aerobic bacteria; predominantly A. viscosus, P. micros, and C. curva and C. rectus in health (50) (Table II). When P. gingivalis was introduced, there was a shift to a more anaerobic flora in the biofilm and an overall increase in bacterial load. Previously undetected species including S. intermedius and F. nucleatum were observed. P. gingivalis was detectable throughout the experiment except in the RvE1-treated group (Table II). There is a precedent for the observation of Gram-negative pathogens altering the dynamics of the endogenous biofilm (51) suggesting biofilm behavior in rabbits is similar to that observed in humans.

H&E-stained sections showed significant inflammatory cell infiltration in connective tissue and bone resorption with irregular bone surfaces and resorptive lacunae in all groups except the RvE1-treated group, where inflammatory changes and bone loss were essentially completely reversed (Fig. 4,a). Histomorphometric analysis of H&E-stained sections complemented clinical assessments. Linear and area measurements revealed significant changes in bone levels between treatment groups (Fig. 4 b). The vehicle-alone, PGE₂-, and LTB₄-treated groups progressed significantly by 12 wk to ≥50% bone loss. In sharp contrast, RvE1 treatment resulted in 30% bone gain restoring bone to pretreatment levels.

To differentiate between bone growth and regeneration of the periodontal organ (periodontal ligament, cementum, and bone), un-decalcified sections were stained with Van Gieson’s, and evaluated by light and polarized microscopy (46). The examination of the un-decalcified sections revealed regeneration of tissues including the periodontal ligament, cementum, and bone at the sites treated with RvE1 (Fig. 5). The regenerated tissues were indistinguishable from the native structures found in the apical areas of the same teeth. Adjacent to the cementum on the root surface, newly formed and continuous periodontal ligament with oblique connective tissue fibers was observed (Fig. 5,a). Sharpey’s fibers inserting in newly formed cementum and bone were observed. The lamellar nature of new bone was apparent when viewed under polarized light (Fig. 5 b), and the regenerated bone was generally rather mature with only remnants of woven bone.

Masson’s Trichrome stained sections were also evaluated for evidence of new bone and connective tissue formation (Fig. 6). New collagen deposition (in blue) and new bone as well as cementum formation (in red) were observed in sections treated with RvE1. The periodontal ligament presented densely packed collagen fiber bundles oriented perpendicular to the root surface and inserted into the new cementum, and the presence of functionally oriented collagen fibers appeared to be closely related to the presence or absence of newly formed alveolar bone adjacent to new cementum. In the vehicle-alone, LTB₄-, or PGE₂-treated sections, granulation tissue was observed in place of bone and connective tissue along with loss of the periodontal ligament and cementum. A profound inflammatory infiltrate with soft tissue and collagen destruction was present in subepithelial connective tissue in all groups except RvE1 treatment.

To determine whether the proresolving actions of RvE1 are mediated through the suppression of osteoclastogenesis, TRAP staining was used to identify clastic cells. Resorbing bone lacunae contained large numbers of TRAP-positive cells in vehicle-alone, LTB₄-, and PGE₂-treated animals, whereas RvE1-treated specimens contained few or nondetectable TRAP-positive cells (Fig. 7, a and b).

In addition, osteoblast activity was evaluated by immunohistochemistry in sections stained with osteocalcin as a marker of new bone formation (52, 53). In sections from RvE1-treated specimens, cells consistent with osteoblasts were demonstrable on newly formed alveolar bone (Fig. 7 c). The bone surface was intact and demonstrated the normal characteristics of alveolar bone suggesting that RvE1 treatment regenerated and completely restored the healthy architecture of osseous tissue surrounding the teeth including well organized connective tissue, periodontal ligament, and cementum deposition.

There is substantial epidemiologic evidence and some experimental evidence in humans suggesting that periodontal inflammation can influence the course of systemic disease, in particular, cardiovascular disease, diabetes and low-birth weight premature birth (54, 55, 56, 57). CRP is considered a component of normal serum, occurring in concentrations ranging from 0.07 to 2.9 μg/ml with a median value of 0.6 to 1.9 μg/ml; however, elevated CRP reflects an elevation in systemic inflammation that is associated with increased risk for cardiovascular disease (58). Elevated serum CRP levels in periodontitis subjects were initially reported by Ebersole et al. (59). IL-1β is produced at the initial stages of inflammation primarily by monocytes and macrophages. IL-1β induces capillary endothelial cells to secrete chemokines and to increase the expression of cell adhesion molecules (60). IL-1β also induces the expression of metalloproteinases (61). To evaluate the influence of a local inflammation on the systemic inflammatory response, we evaluated serum IL-1β and CRP levels. Serum samples were collected from peripheral blood obtained at baseline, 6 and 12 wk (end of the treatment period). Local inflammation (periodontitis) induced elevations of IL-1β and CRP in all animals. Treatment with vehicle-alone resulted in no additional changes, whereas PGE₂ treatment induced a marked increase in IL-1β compared with baseline disease. Also, the systemic marker CRP (58) was markedly increased with LTB₄ treatment. Topical RvE1 therapy gave a statistically significant reduction in both systemic IL-1β and CRP levels compared with the other groups (Fig. 8) (p < 0.05). The levels of IL-1β and CRP with RvE1 treatment were comparable with levels associated with health.

No adverse events were apparent throughout the study and no animals were prematurely lost during the study. The rabbits tolerated these procedures well and showed no indication of adverse events related to the pathogen, e.g., lethargy or fever. To examine the potential for local and systemic side effects of RvE1, intraoral soft and hard tissue examinations were performed during the course of the study. In addition, biopsies were obtained from specific internal organs including, esophagus, lung, liver, spleen, and kidney at the end of the study. Clinical assessments of oral soft and hard tissues did not reveal any untoward events including irritation, redness, suppuration or any lesions of oral tissues. Histological evaluations of the tissue biopsies were normal, with the exception of one PGE₂-treated animal that exhibited focal areas of inflammation in the liver.

In this study, we report for the first time the regeneration of hard and soft tissues lost to inflammatory disease by the activation of inflammation resolving pathways with an endogenous mediator, RvE1, recently identified from ω-3 fatty acids, used as a topical treatment. Homeostasis is a fundamental characteristic of living things. It is the maintenance of the internal environment within tolerable limits and is often described as a process of balance. The physiologic resolution of a well orchestrated inflammatory response is essential to maintain homeostasis at the cellular and tissue level generating specific mediators that can dampen the magnitude of the leukocyte infiltrate during inflammation and promote resolution (24, 62). Recently, a new family of local-acting mediators was discovered that are products of PUFA aspirin-triggered transformation circuits (63). These new chemical mediators are endogenously generated from EPA in inflammatory exudates collected during the resolution phase and were termed resolvins because specific members of the family control the magnitude and duration of inflammation in animal models (10, 24, 39, 40). There is a body of evidence that proresolution molecules from the new families derived from ω-3 PUFAs, resolvins, and protectins, counterregulate neutrophil infiltration and promote resolution (40). RvE1 specifically interacts with the LTB₄ receptor BLT1 on neutrophils and ChemR23 on monocytes to regulate leukocytes during inflammation (41). RvE1 also stimulates the uptake and clearance of local cytokines (29).

Periodontitis is characterized by destruction of connective tissue and bone by the host response (64). Lipid mediators of inflammation play an important role throughout the pathogenesis of periodontitis. In particular, PGE₂ and LTB₄ are strongly associated with progressive disease (65, 66) and are, in large part, drivers of the chronic lesion. Conversely, recent new discoveries demonstrate that resolution of inflammation is an active process and that homeostasis cannot be achieved until the lesion is free of neutrophils (67). These principles are supported by these studies where we demonstrate that exogenous PGE₂ and LTB₄ enhance the local inflammatory response leading to neutrophil recruitment and enhanced neutrophil-mediated tissue damage. Monocytes recruited to the chronic lesion enhance the inflammatory response through secretion of more PGE₂, IL-1β, TNF-α, and other proinflammatory molecules. The resolving molecules stop neutrophil infiltration and drive neutrophils to apoptosis (23, 29, 68), while at the same time attracting monocytes to the lesion (23). However, the phenotype of the resolvin-recruited monocyte is nonphlogistic (69), and they phagocytose apoptotic neutrophils without contributing to further inflammation or tissue damage.

Much has been written in recent years about the relationship between periodontitis and systemic disease (for review, see Ref. 70). The reported work in this area suggests that the local inflammatory burden presented by periodontal infection causes a systemic inflammatory burden. CRP and IL-1β are most often reported markers of systemic inflammatory burden. As indicated earlier, RvE1 seems to work through specific receptors on cells, but the systemic effect is almost certainly indirect. RvE1 therapy lowers the inflammatory burden locally, which results in a lower systemic inflammatory burden.

Periodontitis is unique among the inflammatory diseases because the etiology is well known and well characterized: the biofilm, dental plaque. Our work complements the earlier observations of Marsh et al. (51), in chemostat biofilm systems that indicated that Gram-negative pathogens have an impact on the stability and dynamics of the biofilm. They observed, as we have here, that the Gram-negative pathogen causes an overgrowth of the resident flora and a shift to a more pathogenic flora that incorporates other Gram-negative pathogens. Of particular interest in this study is the observation that control of inflammation through proresolution pathways resulted in elimination of the Gram-negative pathogen from the flora and a return to pre-disease homeostasis of both the resident flora and the host. RvE1 has no inherent antibacterial activity (10). The elimination of the pathogen is likely related to the physiology of the organism.

P. gingivalis is a Gram-negative, obligate anaerobic, asaccharolytic organism. Unable to use sugars for energy, it metabolizes essential amino acids. The source of amino acids is collagen breakdown products provided by the host through inflammation. In fact, P. gingivalis possesses an array of proteolytic enzymes, the gingipains (71), to accomplish this. We hypothesize that resolution of inflammation effectively eliminates P. gingivalis from the lesion by removing the food source. We suggest that while the etiology of periodontitis is bacteria, the pathogenesis is inflammatory. The chronic inflammation supports the growth of the pathogen through production of tissue breakdown products. Resolution of the inflammatory lesion removes the ecological niche of the pathogen. This is supported by our earlier observation that prevention of inflammation with resolving molecules precludes establishment of the pathogen and tissue breakdown (10).

Regeneration of tissues lost to disease is problematic in many human diseases, such as arthritis and periodontitis. Therapies that are aimed at regeneration of lost tissues attempt to recapitulate development with the assumption that the ability to recapitulate development is lost as the organism ages. Hence, the therapeutic strategy is to add back growth factors, substrates, and other molecules to mimic development. The present results emphasize the role of local inflammation in tissue regeneration.

In summary, our results are the first to demonstrate RvE1 as a therapeutic agent in an in vivo leukocyte- and osteoclast-mediated inflammatory disease. RvE1 acts as a modulator of the inflammatory response shifting the response to more rapid resolution and effectively preventing the chronic phase. Elimination of inflammation in the healing lesion promotes tissue regeneration. These principles may be applicable to other inflammatory diseases including arthritis and cardiovascular disease due to the similarities between these diseases, such as the neutrophil induced panus formation in arthritis (72, 73) and the inflammatory tissue damage to blood vessels stimulating atherogenesis (74). These observations taken together provide novel evidence that Resolvin E1 not only plays a key role in controlling inflammation but also might be useful for a wide range of complex inflammatory conditions including bone disorders, such as periodontitis and arthritis, by restoration of stem cells (75) thereby promoting regeneration of lost tissues, including connective tissue and bone.

We thank Jennifer Deady, Malika Kohli, Kerry Light, and the veterinary and technical personnel of the Boston University Medical Campus Laboratory Animal Science Center for assistance during animal handling and experimental procedures; and the Organic Synthesis Core of P50-DE016191 (to C.N.S.) led by Dr. Nicos Petasis for preparing synthetic RvE1.

Drs. Hatice Hasturk, Algdogan Kantarci, Emilie Goguet-Surmenian, Amanda Blackwood, and Chris Andry have no conflicting financial interests. Brigham and Women’s Hospital and Boston University are assigned patents on resolvins that are licensed for clinical development and are subject to consultant agreements for Drs. Charles N. Serhan and Thomas E. Van Dyke.