Systems approach to inflammation resolution: identification of novel anti-inflammatory and pro-resolving mediators


  • C. N. SERHAN

    1. Center for Experimental Therapeutics and Reperfusion Injury, Department of Anesthesiology, Perioperative and Pain Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA, USA
    Search for more papers by this author

Charles N. Serhan, Center for Experimental Therapeutics and Reperfusion Injury, Department of Anesthesia, Perioperative and Pain Medicine, Brigham and Women’s Hospital, Thorn 724, 20 Shattuck Street, Boston, MA 02115, USA.
Tel.: +1 617 732 8822; fax: +1 617 582 6141.


Summary.  Using a systems approach to profile self-limited inflammatory exudates, we identified three novel families of lipid-derived mediators, coined the resolvins, protectins and most recently, the maresins that control both the magnitude and duration of inflammation. The mapping of these endogenous resolution circuits provides new avenues to probe the molecular basis of many widely occurring inflammatory diseases. This article focuses on our recent advances on the functional metabolomics of this novel genus of specialized pro-resolving mediators (SPM). SPM include resolvins, protectins and maresins and are biosynthesized from essential omega-3 fatty acid precursors. Each possesses potent multi-pronged actions that proved to be stereoselective with human cells and in animal disease models. Resolvins and protectins are also produced in bone marrow. Together, these findings suggest that defective resolution mechanism(s) may underlie some chronic inflammatory diseases. Moreover, identification of functional SPM biosynthesized during inflammation-resolution indicates that resolution is an active process.


Many widely used anti-inflammatory therapies are directed toward the inhibition of enzymes and/or antagonism of receptors. Both selective cyclooxygenase inhibitors and anti-tumor necrosis factor α (TNF-α) are examples of this approach that are used with the goal of blocking production of pro-inflammatory chemical mediators. Research in the author’s laboratory focusing on profiling self-limited inflammation uncovered novel mechanisms that terminate the local acute inflammatory response as well as stimulate resolution and return of the tissue to homeostasis. Identification of these biochemical and cellular processes indicates that resolution of acute inflammation, once considered a passive process, is actually an active programmed process at the tissue level reviewed in Ref. [1,2]. Therefore, rather than targeting inhibition or antagonism of inflammation, research in this laboratory addresses the potential use of endogenous agonists of resolution to stimulate key regulatory points that naturally resolve inflammation.

This systems approach now opens a new understanding of the mechanisms underlying inflammatory disease as well as a new discipline, namely resolution pharmacology. Essential omega-3 fatty acids, in particular eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), are precursors to a new genus of potent lipid mediators (LM) that are both pro-resolving and anti-inflammatory [specialized pro-resolving mediators (SPM)] and serve a physiologic role defining programmed resolution. This ISTH presentation highlights the biosynthesis and actions of this novel genus of endogenous SPM that govern resolution of acute inflammation.

Complete resolution: the ideal outcome

An acute inflammatory reaction, in general, in response to infection or unwanted tissue damage is characterized at least at the gross level by the classic cardinal signs of inflammation (heat, redness, swelling, and pain), and in experimental settings in vivo the temporal relationships are well established, namely, edema and the accumulation of leukocytes, specifically polymorphonuclear leukocytes (PMN), followed by accumulation of monocytes and macrophages. These events in self-limited or resolving inflammatory reactions are coupled with release of local factors that prevent further or excessive trafficking of leukocytes allowing for resolution [3,4]. Early in the inflammatory response, pro-inflammatory mediators such as prostaglandins and leukotrienes play an important role [5]. The progression from an acute inflammation to chronic inflammation as in many widely occurring human diseases such as arthritis, periodontal disease and cardiovascular disease, to name a few, is widely viewed as an excess of pro-inflammatory mediators. Although mononuclear cells can sometimes contribute to pro-inflammatory responses, they are also critical in wound healing, tissue repair and remodeling in a non-inflammatory, non-phlogistic manner [6]. Hence, it is possible that defects associated with mounting endogenous pro-resolving circuits and local autacoids could underlie some of the aberrant mechanisms in chronic inflammation.

Complete resolution of an acute inflammatory response and the return of the local tissues to homeostasis are necessary for ongoing health. Removal of leukocytes from tissues involved in the inflammatory response without leaving remnants of the host defenses and combat between leukocytes, invading microbes, and/or other initiators of inflammation is an ideal outcome. In my laboratory, we have focused on the question ‘How is the acute inflammatory response regulated?’ It was widely believed that simple dilution of pro-inflammatory mediators is sufficient to ‘burn out’ inflammation, with the subsequent responses ending passively [6].

Pro-resolving lipid mediators and anti-inflammation

SPM are a recently uncovered genus of endogenous chemical mediators identified in exudates that include three distinct new chemical families: resolvins, protectins and the recently identified maresins, which are involved in acute inflammation. Each of these families is actively biosynthesized in the resolution phase of acute inflammation and the mediators are potent agonists that control the duration and magnitude of inflammation [4,7,8]. They are also potent chemo-attractants, but via a non-inflammatory mechanism: for example, lipoxins from arachidonate activate mononuclear cell recruitment without stimulating release of pro-inflammatory chemokines or activation of pro-inflammatory gene pathways [1]. They also stimulate the uptake of apoptotic PMNs [9] and activate endogenous anti-microbial defense mechanisms [10] as well as clearance on mucosal surfaces [11]. These actions are agonistic in that they stimulate receptors; via acting on separate cell populations, they stimulate overall resolution of inflammation.

Lipoxin A4 (LXA4) and LXB4 are anti-inflammatory and were the first pro-resolving mediators identified, as their appearance signals the resolution of acute contained inflammation [12]. Lipoxins are lipoxygenase-derived eicosanoids, derived enzymatically from arachidonic acid (AA), an omega-6 fatty acid that is released and mobilized during inflammation [5]. In human systems, they are biosynthesized, for example, during cell–cell interactions involving mucosal, that is, epithelial cells of the gastrointestinal tract or bronchial tissue, interactions with leukocytes; within the vasculature, platelet–leukocyte interactions are a main source [1]. Aspirin has an unexpected impact within resolution. It ‘jump-starts’ the process via triggering endogenous biosynthesis of lipid mediators [13,14].

During local contained inflammation, the first line of host defense, namely the neutrophils, die at the site and can undergo apoptosis as well as necrotic cell death. As part of resolution, LX signal macrophages to enhance their uptake of the remains of these cells [9]. LX are highly potent anti-inflammatory mediators that are formed and act in picogram to nanogram amounts with human tissues and in animal disease models [12]. LX have the specific pro-resolution actions of limiting PMN recruitment and adhesion. They are essentially braking/stop signals for PMN-mediated tissue injury (see Ref. [12] and references within).

Novel local mediators biosynthesized from omega-3 precursors


Resolvins and protectins are two families of local mediators identified in the resolving exudates of inflammation. They were initially identified using a systems approach with LC-MS-MS-based lipidomics and informatics and then complete structural elucidation of the bioactive mediators and related compounds was achieved [3,4,14–16]. The term resolvins or resolution–phase interaction products refers to endogenous bioactive mediators biosynthesized from the major omega-3 fatty acids eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). Hence, there are E series (RvE) and D series (RvD) resolvins [4]. Resolvins are also produced by a COX-2-dependent pathway in the presence of aspirin generating their aspirin triggered form. A growing body of evidence indicates that resolvins possess potent anti-inflammatory and immunoregulatory actions that include blocking the production of pro-inflammatory mediators and regulating trafficking of leukocytes (Table 1) (reviewed in Ref. [2]). Specifically, resolvins stop PMN infiltration in vivo and transmigration [4,17]. They also reduce cytokine expression by isolated microglia cells [15]. We recently established the stereochemistry and actions of RvD1 and aspirin-triggered (AT)-RvD1, RvE1 and PD1 (vide infra) as well as demonstrated the stereoselective basis for their potent actions [14,16,17]. We compared RvD1 and its 17R-AT form (0–1000 nm) on human PMN transendothelial migration. Both stop PMN transmigration in a concentration-dependent manner (< 0.001) [17]. The potency of these compounds is noteworthy, with ∼50% reduction in PMN transmigration at 10 nm. The actions of RvD1 were studied in several inflammatory disease models and confirmed as potent and stereoselective (Table 1).

Table 1.   Resolvins and protectins in complex disease animal models*
Disease modelSpeciesAction(s)References
  1. *The actions of each of the main resolvins and protectins, that is, RvE1, RvD1 and PD1, were confirmed with compounds prepared by total organic synthesis (see text and cited references for further details).

Resolvin E1
 PeriodontitisRabbitReduces neutrophil infiltration; prevents connective tissue and bone loss; promotes healing of diseased tissues; regenerates lost soft tissue and bone[27]
 PeritonitisMouseStops neutrophil recruitment; regulates chemokine/cytokine production
Promotes lymphatic removal of phagocytes
 Dorsal air pouchMouseStops neutrophil recruitment[3]
 RetinopathyMouseProtects against neovascularization[23]
 ColitisMouseDecreases neutrophil recruitment and proinflammatory gene expression; improves survival; reduces weight loss[26]
Resolvin D1
 PeritonitisMouseStops neutrophil recruitment[15,17]
 Dorsal skin air pouchMouseStops neutrophil recruitment[4,15]
 Kidney ischemia-reperfusionMouseProtects from ischemia-reperfusion-induced kidney damage and loss of function; regulates macrophages and protects from fibrosis[28]
 RetinopathyMouseProtects against neovascularization[23]
Protectin D1
 PeritonitisMouseStops neutrophil recruitment; regulates chemokine/cytokine production
Promotes lymphatic removal of phagocytes
Regulates T-cell migration
 AsthmaMouseProtects from lung damage, airway inflammation and airway hyperresponsiveness[30]
HumanProtectin D1 is generated in humans and appears to be diminished in asthmatics[30]
 Kidney ischemia-reperfusionMouseProtects from ischemia-reperfusion-induced kidney damage and loss of function; regulates macrophages and is anti-fibrotic[28]
 RetinopathyMouseProtects against neovascularization[23]
 Ischemic strokeRatStop leukocyte infiltration, inhibits NF-κB and cyclooxygenase-2 induction[19]
 Alzheimer’s diseaseHumanDiminished protectin D1 production in human Alzheimer’s disease[31]


Protectins are also biosynthesized from DHA via a separate new pathway. Protectins are distinguished by the presence of their conjugated triene containing structure [16]. The name ‘protectins’ was coined from the observed anti-inflammatory [15] and protective actions. The prefix neuroprotectin gives the tissue location of its production, such as neuroprotectin D1 [16,18]. Like resolvins, the protectins also stop PMN infiltration [15,16]. They are biosynthesized by and act on glial cells and reduce cytokine expression [15]. In mouse models, NPD1 reduces retinal and corneal injury [18] and stroke damage [19], and improves corneal wound healing [20] (see Refs. [1,2] and references cited within).

Programmed resolution of inflammation

To study resolution using an unbiased systems approach, the murine dorsal air pouch was ideal because it permitted isolation of contained inflammatory exudate [3,4]. This system also enabled direct LM-lipidomics–informatics (bioactive products, as well as their inactive precursors and further metabolites), proteomics, and cellular composition. Importantly, it was possible to establish by direct comparisons when and where different local mediators were biosynthesized and activated during resolution, namely temporal spatial differential analyses [1,21]. In this regard, we used the murine dorsal air pouch as well as peritonitis to determine the formation and roles of endogenous LXA4 in the resolution of acute inflammation [22]. Upon initiation of inflammation with TNF-α, there was a typical acute-phase response denoted by rapid PMN infiltration preceded by both local prostaglandins and leukotrienes. Unexpectedly, the eicosanoids then underwent what we have termed a ‘class switch’. As the exudate evolved, the eicosanoid profiles switched and the lipid mediators made within this milieu changed with time [22]. Leukotrienes (potent chemo-attractants) were deactivated and the transcriptional regulation of enzymes required for LX and resolvin production was activated. This in turn attracts mononuclear cells and stimulates macrophages to take up apoptotic neutrophils within the contained inflammatory exudate site.

The lipid mediator class switch in exudates was driven in part by COX-derived prostaglandins E2 and D2, which regulate transcription of enzymes involved in lipoxin biosynthesis [22]. Thus, with Sir John Savill we introduced the concept that ‘alpha signals omega,’ namely the beginning signals the end in inflammation [6].

Omega connection: resolvins, protectins and maresins

DHA is well known for its important role in neuronal systems and, along with AA, is also one of the major polyunsaturated fatty acids (PUFA) found in the retina. The other major omega-3 PUFA present in marine fish oils is EPA. To determine whether EPA, DHA, and AA regulate inflammation in vivo, we studied disease models to compare the effects of omega-3 and -6 PUFAs in wild-type mice with mice that overexpress the Caenorhabditis elegansfat-1 gene. This gene converts omega-6 PUFA into omega-3 resulting in elevated tissue levels of omega-3 PUFA within the fat-1 overexpressing mice. A protective action against pathological angiogenesis was found in retina when there was a lower ratio of omega-6:omega-3 PUFA [23]. Wild-type mice lacking the fat-1 transgene had extensive vaso-obliteration and severe retinal neovascularization compared with fat-1 mice. In mice fed omega-3 PUFA, there were markers of neuroprotectin D1 (NPD1) and RvE1 biosynthesis. In mice without omega-3 PUFA supplementation, administration of RvD1, RvE1, or NPD1 gave protection from vaso-obliteration and neovascularization [23]. These fat-1 mice are also protected from colitis [24]. In addition to exudates, murine marrow produces resolvins and protectins, which are enhanced with feeding EPA and DHA [25].

SPM are protective in peritonitis. For example, zymosan A was injected into mice to induce peritonitis [16] and PD1 was protective [16,17,21]. Rv and PD1 each block > 85% of further PMN migration into the site [16]. When both PD1 and RvE1 were injected together to determine whether their actions were synergistic or additive, RvE1 (10 ng) reduced PMN infiltration, although less than that with PD1 (10 ng). In combination, the reduction was greater, suggesting an additive action. Inflammatory bowel disorders, like colitis, are characterized by relapsing inflammation. In a well-studied experimental colitis model, RvE1 protects against bowel inflammation in mice challenged with an intrarectal antigenic hapten, 2,4,6-trinitrobenzene sulfonic acid (TNBS), to induce colitis [26]. As little as 1 μg RvE1 dramatically reduced mortality, weight loss, and histologic severity of colitis. In rabbits, RvE1 is protective in periodontal disease, where it appears to stimulate regeneration [27]. Thus, pro-resolving mediators display anti-inflammatory and anti-fibrotic actions in several widely used laboratory disease models (see Table 1).

Maresins: macrophage mediators in resolving inflammation

Using self-resolving inflammatory exudates and lipidomics, we identified a new pathway involving biosynthesis of potent anti-inflammatory and pro-resolving mediators from the essential fatty acid docosahexaenoic acid (DHA) by macrophages [8]. During the resolution of murine peritonitis, exudates accumulated both 17-HDHA, a known marker of 17S-D-series resolvin and protectin biosynthesis, and 14S-HDHA from endogenous DHA. Addition of either DHA or 14S-hydroperoxydocosa-4Z,7Z,10Z,12E,16Z,19Z-hexaenoic acid (14S-HpDHA) to activated macrophages converted these substrates to novel dihydroxy-containing products that possessed potent anti-inflammatory and pro-resolving activity with a potency similar to resolvin E1 and protectin D1. Stable isotope incorporation, intermediate trapping, and characterization of physical and biological properties of the products demonstrated a novel 14-lipoxygenase pathway, generating bioactive 7,14-dihydroxy-docosa-4Z,8,10,12,16Z,19Z-hexaenoic acid, coined maresin [macrophage mediator in resolving inflammation (MaR1)], which enhances resolution. These findings suggest that maresins and this new macrophage metabolome may be involved in some of the beneficial actions of DHA when utilized by resolving macrophages in tissue homeostasis, inflammation-resolution, wound healing, and host defense.

In summation, the acute inflammatory response initiated by neutrophils with injury or infection is, in ideal settings, self-limited and protective. Neutrophil-derived pro-inflammatory mediators, including leukotrienes and prostaglandins, can amplify inflammation. We found that neutrophils within contained exudates can change phenotypes to generate protective mediators, derived from fatty acids, to promote resolution. There is an active catabasis to return tissues to a homeostatic healthy state from the battles of host defense [21]. Of interest, SPM, such as protectins, resolvins and maresins identified in resolving exudates, when administered in animal disease models, accelerate the return to homeostasis [7,8]. These protective SPM include the arachidonic acid derived lipoxins as well as omega-3 essential PUFA-derived resolvins, protectins and maresins. These findings provide evidence that the resolution of acute inflammation is not passive, but an actively programmed response to terminate inflammation.


The author thanks Mary H. Small for expert assistance with manuscript preparation and the members of his laboratory and collaborators for their expertise and efforts in the reports referenced herein. The author acknowledges the support of NIH grants P50-DE016191, GM038765 and DK074448.

Disclosure of Conflict of Interest

The author is inventor on patents assigned to Brigham and Women’s Hospital and Partners HealthCare on the composition of matter, uses, and clinical development of anti-inflammatory and proresolving lipid mediators. These are the basis of ongoing consulting agreements with pharmaceutical companies.