Activation of Resolution Pathways to Prevent and Fight Chronic Inflammation: Lessons From Asthma and Inflammatory Bowel Disease

Formerly considered as a passive process, the resolution of acute inflammation is now recognized as an active host response, with a cascade of coordinated cellular and molecular events that promotes termination of the inflammatory response and initiates tissue repair and healing. In a state of immune fitness, the resolution of inflammation is contained in time and space enabling the restoration of tissue homeostasis. There is increasing evidence that poor and/or inappropriate resolution of inflammation participates in the pathogenesis of chronic inflammatory diseases, extending in time the actions of pro-inflammatory mechanisms, and responsible in the long run for excessive tissue damage and pathology. In this review, we will focus on how resolution can be the target for therapy in “Th1/Th17 cell-driven” immune diseases and “Th2 cell-driven” immune diseases, with inflammatory bowel diseases (IBD) and asthma, as relevant examples. We describe the main cells and mediators stimulating the resolution of inflammation and discuss how pharmacological and dietary interventions but also life style factors, physical and psychological conditions, might influence the resolution phase. A better understanding of the impact of endogenous and exogenous factors on the resolution of inflammation might open a whole area in the development of personalized therapies in non-resolving chronic inflammatory diseases.

Overall there are two distinct phases in an inflammatory reaction: the initiation of inflammation and the resolution phase ( Figure 1 ). A post-resolution phase of inflammation that links innate and adaptive immune systems has also been described ( 9 , 10 ). For effective resolution of inflamed tissues to occur and to restore tissue homeostasis, specific cellular mechanisms that are under the control of pro-resolving mediators are enlisted to promote termination of the inflammatory response and initiate tissue repair and healing ( Figure 2 ). Better understanding of how the environment can impact on the resolution of inflammation will lead to an improved understanding of why the chronic inflammatory diseases persist. Figure 2 Key cellular actors of resolution. For effective resolution of inflamed tissues to occur and to restore tissue homeostasis, specific cellular mechanisms that are under the control of pro-resolving mediators are enlisted. They promote termination of the inflammatory response and initiate tissue repair and healing. Pro-inflammatory mediators: red circles, pro-resolving mediators: blue circles. Anx, annexin; DCs, Dendritic cells; Eos, eosinophils; Gal, Galectin; IBD, inflammatory bowel disease; IL, interleukin; ILC2, Type 2 innate lymphoid cells; ILC3, Type 3 innate lymphoid cells; MDSCs, Myeloid-derived suppressor cells; MiRs, MicroRNAs; NK, Natural killer; PMN, polymorphonuclear cells; TGF-beta, Transforming growth factor beta; Th1, Type 1 T helper cells; Th2, Type 2 T helper cells; Treg, regulatory T cells; SPMs, specialized pro-resolution lipid mediators; VIP, vasoactive intestinal peptide. Cells Involved in the Resolution of Inflammation Macrophages One of the key events in determining the initiation of the resolution phase is the recruitment of non-phlogistic monocytes and their differentiation into macrophages at sites of inflammation. Indeed, central to the successful resolution of inflammation, is the process of local leukocyte clearance by apoptosis and subsequent phagocytosis of the apoptotic cells by surrounding monocyte-derived phagocytes ( Figure 2 ). Engulfment of apoptotic cells signals to the phagocytosing macrophages that the inflammatory response is ending, and alters macrophage mediator production from a predominantly pro-inflammatory (M1) to an anti-inflammatory and pro-resolving phenotype (M2), that further enhances phagocytosis of apoptotic cells and promotes the return to tissue homeostasis ( 11 , 12 ). This shifting balance between pro-inflammatory M1 and wound-healing M2 macrophages over time is essential for proper resolution of inflammation ( 13 ).

Innate lymphoid cells (ILCs) are a large family of cells with various immunological functions ( 22 ). They can be classified into different subgroups based on their cytokine production and their expression of key transcription factors, similar to T cell subsets. In various mouse models of asthma and IBD, studies suggest a role for ILCs in the induction of inflammation [for recent reviews see ( 23 , 24 )]. Recent evidence suggests a more complex role for these cells, with dual roles in the induction of inflammatory diseases but also the control of chronic inflammation. Type 2 ILC (ILC2) cells demonstrate a flexibility and plasticity dependent on the local microenvironment and can potentially act both as effectors and suppressors [reviewed in ( 25 , 26 )]. ILC2 cells, by producing IL-5 and IL-13, promote the development of type 2 allergic inflammation, independent of Th2 cells ( 9 , 27 ). In contrast, the production of IL-9 by ILC2s was recently reported to mediate resolution of inflammation in a model of chronic arthritis, another chronic non-resolving disease ( 28 ). Also, a potential role for ILC2 has been suggested in tissue repair after acute lung injury in a mouse model of H1N1 influenza virus infection through the production of amphiregulin ( 29 ). The same holds true for type 3 ILC (ILC3) cells, the most abundant ILC subtype in the human intestine at steady state ( 30 ). ILC3 cells are the main contributors to intestinal IL-22 production, which is a tightly regulated mediator for immune homeostasis in the intestinal tract ( 31 ). NK cells are also members of the ILC family with potential roles of SPM-induced resolution of eosinophilic inflammation in Th2 asthma ( 32 , 33 ).

During the inflammatory response, diverse mediators are synthesized in a strict temporal and spatial manner to act on specific receptor targets and to actively prevent the overshooting of acute inflammatory mechanisms, and ultimately restore tissue homeostasis. Functionally, these anti-inflammatory and pro-resolving mediators counter-regulate key events of inflammation. Different from solely anti-inflammatory actions, pro-resolving mediators actions typically target specific pro-resolution mechanisms: limitation and/or cessation of neutrophil recruitment; promotion of non-phlogistic monocyte recruitment; induction of neutrophil apoptosis and their subsequent efferocytosis by macrophages, enhancement of efferocytosis, reprogramming of macrophages from classically activated to alternatively activated cells; return of non-apoptotic cells to the blood or egress via the lymphatic vasculature; stimulation of tissue repair and cellular repopulation of the tissue, leading to “adapted homeostasis” [recently reviewed in ( 43 )]. Pro-resolving mediators are diverse in nature, and include SPMs (lipoxins, resolvins, protectins, and maresins), proteins and peptides [annexin A1 (AnxA1), galectins, adrenocorticotropic hormone (ACTH), and IL-10], gaseous mediators including hydrogen sulfide (H2S) and carbon monoxide (CO), nucleotides (e.g., adenosine), as well as neuromodulators released under the control of the vagus nerve such as acetylcholine and neuropeptides released from non-adrenergic non-cholinergic neurons ( 44 ). As diverse as their nature is their origin, where mediators of resolution can be produced locally, acting in paracrine and autocrine manners, or produced at distant sites, followed by their systemic release and extravasation to sites of inflammation ( 43 ). Below the main pro-resolving mediators will be described. Specialized Pro-resolving Lipid Mediators (SPMs) Recently, a new array of lipid molecules that function in the resolution of inflammation were elucidated and collectively named SPMs ( 3 , 45 , 46 ). These mediators, such as lipoxins (Lx), resolvins (Rv), protectins (PD), and maresins (Mar), are produced during the inflammatory response and derive from polyunsaturated fatty acids (PUFAs). Whereas, Lx derive from the omega-6 PUFA arachidonic acid, the omega-3 PUFAs eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) give rise to Rv, PD, and Mar ( Figure 3 ). More recently SPMs produced from both the omega-6 and omega-3 docosapentaenoic acids have been described ( 47 ). The SPMs are produced via biosynthetic circuits engaged during cell–cell interactions including different innate immune cells, for example macrophages or neutrophils, and structural cells at sites of inflammation. SPMs can also been produced through interactions of platelets with leukocytes ( 48 ). Figure 3 Overview of the pathways for synthesis of resolvins from omega-3 polyunsaturated fatty acids, DHA and EPA. DHA, docosahexaenoic acid; EPA, eicosapentaenoic acid, MaR, maresin; PD, protectin; Rv, resolvin. These bioactive lipids display potencies in the nanomolar range, and signal through cognate G-protein coupled receptors (GPR) such as the N-formyl peptide receptor 2 (ALX/FPR2), GPR32, and GPR18 with many cell type-specific actions ( 49 , 50 ) ( Figure 4 ). Figure 4 Specialized pro-resolving lipid mediators signal through G-protein coupled receptors on a variety of cell involved (deranged) immune response leading to cell specific responses. Akt, protein kinase B; ALX/FPR2, N-formyl peptide receptor 2—LXA 4 receptor; AnxA1, annexin A1; BLT1, leukotriene B 4 receptor 1; CD, cluster domain; CMKLR1, chemokine like receptor 1 or Chemerin Receptor 23; DVR1, RvD1 receptor or G protein coupled receptor (GRP)32; DVR2, RvD2 receptor or GRP18; ERK, extracellular signal regulated kinases; IL, interleukin; INFγ, interferon γ; Mcl-1, anti-apoptotic protein in mast cells; miR, microRNA; mTOR, mammalian target of rapamycin; NK cell, natural killer cell; NFκB, nuclear factor kappa-light-chain-enhancer of activated B cells; P, phosphorylated; PDK1, phosphoinositide-dependent protein kinase 1; PI3K, phosphatidylinositol 3-kinase; PMN, polymorphonuclear cells; Rv, resolvin; LX, lipoxin; S6K, ribosomal protein S6 kinase; Th17 cell: Thelper 17 lympocyte; Treg cell: regulatory T lymphocyte; TNFα, tumor necrosis factor α; Traf6: TNF receptor associated factor 6. LXA 4 binds to the ALX/FPR2. This receptor displays diverse ligand affinities that extend beyond interactions with LXA 4 . Indeed, ALX/FPR2 can interact with over 30 ligands with various affinities, and has been identified as the first receptor to engage both bioactive lipids and peptides/proteins, including annexin A1 ( 50 ). ALX/FPR2 is widely expressed on human leukocytes, including neutrophils, eosinophils, monocyte-macrophages, T cells, NK cells, and ILC2 cells, as well as on tissue resident cells, such as airway epithelial cells and fibroblasts ( 33 , 51 , 52 ). Its expression is u

IL-10 is a cytokine important in controlling excessive inflammation. It mediates its major functions through inhibition of cytokine production and down-regulating antigen presentation by macrophages, monocytes, and dendritic cells (DCs) and thereby inhibiting adaptive immune cells such as Th2 and Tregs ( 66 – 68 ). IL-10 can also inhibit eosinophilia, by suppression of IL-5 and GM-CSF and by direct effects on eosinophil apoptosis [( 69 ) #1607; ( 70 ) #1622].

Melanocortins, including adrenocorticotrophic hormone (ACTH) and the α, β and γ-melanocyte-stimulating hormone (MSH) are derived from a larger precursor molecule known as the pro-opiomelanocortin (POMC) protein. They exert their numerous biological effects by activating 7 transmembrane GPCR ( 83 ). ACTH does not only induce cortisol production, as previously assumed, but also exerts anti-inflammatory actions by targeting melanocortin receptors present on immune cells ( 84 ). The protective actions of melanocortins include inhibition of leukocyte transmigration and reduction of pro-inflammatory cytokine production ( 85 , 86 ). Melanocortins also promote clearance of apoptotic cells ( 86 ) and cutaneous wound healing ( 87 ).

Adenosine, is a purine nucleoside generated by the dephosphorylation of adenine nucleotides. In addition to being a potent endogenous physiologic and pharmacologic regulator of many functions, adenosine has pro-resolving mechanisms including inhibition of neutrophil and T cell functions, efferocytosis and macrophage reprogramming [reviewed in ( 93 )].

Anti-inflammatory Cytokines TGF-β is a potent inhibitor of classical pro-inflammatory macrophage activation ( 105 ). TGF-β is also a mediator in critical processes in wound healing, stimulating angiogenesis, fibroblast proliferation, collagen synthesis and deposition and remodeling of extracellular matrix ( 106 , 107 ). Additionally, TGF-β regulates immune responses through the development and differentiation of Th17 cells and FoxP3 + Treg cells ( 108 , 109 ). TGF-β inhibits the differentiation of T helper subsets as it inhibits the expression of Tbet and GATA3, thereby blocking the differentiation of Th1 and Th2 cells respectively ( 110 , 111 ). IL-22 primarily targets non-hematopoietic cells and plays a role in host defense at barrier surfaces where it promotes tissue regeneration ( 112 ). IL-22 is produced by Th17 and Th22 cells, ILCs and NKT cells ( 113 ). It has different roles in the gastrointestinal tract including tissue regeneration, maintenance of the intestinal barrier and intestinal defense against pathogens ( 113 – 116 ). IL-22 levels are enhanced in the lungs of patients with asthma ( 117 ). However, in inducible lung-specific IL-22 transgenic mice, a significant decrease in allergic airway hyperresponsiveness and allergic inflammation occurred indicating an immune modulating effect of IL-22 ( 118 ). IL-1RA (receptor antagonist) is a natural inhibitor of the pro-inflammatory cytokine IL-1 as it functions as an IL-1 receptor competitor ( 119 ). It is produced by CD163 + wound healing M2 macrophages ( 120 ). In IBD, polymorphism in the IL-1RA gene have been demonstrated and an imbalance of IL-1 and IL1RA has been suggested to induce mucosal inflammation associated with IBD ( 121 , 122 ). More recently, IL-4 has been reported to induce macrophage proliferation and activation with reduced pulmonary injury after infection with a lung-migrating helminth ( 123 ).

The paracrine manner of the cellular communication in resolution may be achieved not only by secretion of immune mediators but also through extracellular vesicles. Extracellular vesicles are small membrane vesicles (exosomes, microvesicles, and apoptotic bodies) secreted by all cell types including immune cells in a controlled manner. Extracellular vesicles have recently been reported both as immune activators and immune suppressors as they contain for example MHC class I and II and T cell co-stimulatory molecules ( 128 , 129 ). However, the most described function of extracellular vesicles is triggering of the immune system, and very recently involvement of extracellular vesicles in inflammation resolution, tissue repair and regeneration was reported ( 130 , 131 ).

In the industrialized world, millions of individuals suffer from inappropriate activation and dysregulation of Th2 cell immune responses responsible for allergic asthma and rhinitis, food allergies and atopic dermatitis (also known as eczema), being part of a process called the atopic march. These disorders are increasingly prevalent and are a major public health problem ( 180 ). Th2 cell mediated immune responses are characterized by the release of type 2 signature cytokines (i.e., IL-4, IL-5, IL-9, and IL-13) from cells of both the innate and adaptive immune systems ( 134 , 181 ). Current therapeutic strategies for chronic Th2 immune disorders are mainly anti-inflammatory, and aim at controlling symptoms. In chronic persistent asthma, inhaled corticosteroids are the main anti-inflammatory treatment effective in most patients, causing relatively minor adverse effects ( 182 ). A subset of asthma patients (~10%) experience persistent symptoms and/or frequent exacerbations despite high doses of inhaled corticosteroids and are often treated with prolonged systemic corticotherapy having many potential side-effects ( 183 ). Monoclonal antibodies targeting inflammatory pathways that activate immune responses leading to airway inflammation have been developed to help broaden the current arsenal of asthma treatment options ( 184 ). The first anti-body based biological therapy approved for treatment of asthma was omalizumab, targeting IgE, a component of the allergic cascade ( 185 ). More recently, monoclonal antibodies have been approved, targeting IL-5 or its receptor (mepolizumab, reslizumab, benralizumab), a key cytokine promoting eosinophil inflammation ( 186 ). Other monoclonal antibodies targeting a wide variety of intermediaries in the pro-inflammatory cascade are currently being tested for their effectiveness in the treatment of asthma ( 184 ). These biological therapies can reduce exacerbations and have glucocorticoid-sparing effects, but the clinical responses to these antibody-therapies are variable, with at least 30% of severe asthmatic patients being non-responders ( 187 ). These therapeutic strategies can be combined with allergen-specific immunotherapies in chronic allergic diseases that are able to improve symptoms but they do not cure allergic disorders ( 188 , 189 ). As for IBD, there is increasing evidence that chronic and uncontrolled inflammation in Th2 immune disorders, might result not just from an excessive uncontrolled pro-inflammatory response but also from uncontrolled and insufficient engagement of pro-resolving pathways and thus impaired resolution of exacerbations ( 5 , 33 , 190 ). First, pro-resolving mediators have proved efficient at improving disease and inflammatory outcomes in a variety of asthma models. Treatment with SPMs decreases key features of asthma pathobiology, including airway hyperresponsiveness, mucus metaplasia, and Th2 cell bronchial inflammation ( 53 , 191 – 193 ). AnxA1 deficient mice exhibit spontaneous airway hyperresponsiveness and exacerbated allergen responses ( 194 ) and AnxA1 mimetics inhibit eosinophil recruitment ( 195 ). Ablation of IL-10 signaling in Th2 cells leads to exacerbated pulmonary inflammation ( 196 ). In murine models, IL10 knock-out mice develop enhanced allergic airway responses ( 197 , 198 ) [( 199 ) #1421]. IL-10 also inhibits pro-inflammatory cytokine production by Th2 cells and down-regulates mast cell and eosinophil function ( 200 ). Administration of recombinant Gal-9 or α-MSH diminish allergic airway inflammation [( 201 ) #1453; ( 202 )]. Low H 2 S production in ovalbumin sensitized and challenged mice results in aggravated AHR and increased airway inflammation ( 203 ). In a rat model of asthma, exogenous administration of H 2 S reduces airway inflammation and airway remodeling ( 204 ). VIP can inhibit eosinophil migration ( 205 ) and airway remodeling in asthmatic mice ( 206 ). Intestinal epithelial cell-derived Gal-9 is involved in the resolution of allergic responses through the induction of tolerogenic DCs and associated Treg cell response ( 207 , 208 ). In addition, some of these galectins can block IgE binding on mast cells and as such, inhibit allergic inflammation ( 209 , 210 ). The role of adenosine in the resolution of inflammation of chronic asthma is not yet elucidated. Based on the evidence for the functional importance of pro-resolving mediators in allergic asthma mouse models, defects in the production or the activity of pro-resolving mediators might therefore participate in the chronicity and severity of human asthma. Several studies in distinct populations have reported that SPMs are underproduced in more severe asthma together with a defect in the expression of their related receptors [for a recent review see ( 6 )]. Annexin A1 (AnxA1) levels are also decreased in patients with asthma ( 211 ), and in wheezy infants ( 212 ). Moreover, plasma and bronchoalveolar AnxA1 levels are correlated with lung function (FEV1 %)

Omega-3 PUFAs The main bioactive omega−3 PUFAs, eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), are poorly synthesized in humans. They are components of seafood, especially oily fish, of fish oil, liver oil, krill oil and algal oil supplements, and of a small number of highly concentrated pseudo-pharmaceutical products. EPA and DHA have long been known to have beneficial health effects including anti-inflammatory, anti-thrombotic, and immuno-regulatory properties ( 242 – 244 ). These n−3 LCPUFAs are substrates for biosynthesis of potent SPMs such as resolvins, protectins, and maresins ( Figure 3 ). DHA is concentrated in neural tissues including brain and retina and in sperm; EPA and DHA are found in membranes of all other cells and tissues and in human milk ( 245 – 247 ). Increased dietary intake of EPA and DHA results in their enrichment in blood and in many cells and tissues. Omega-3 PUFAs can exert significant effects on the intestinal environment and modulate the gut microbiota composition ( 248 ). The airway mucosa is also enriched with DHA in healthy individuals ( 249 ). Interestingly, airway mucosal levels of n−3 PUFAs are lower in patients with asthma than in people without asthma ( 249 ). Population surveys report that diets rich in n-3 fatty acids are associated with lower asthma prevalence ( 250 ). It is noteworthy that SPMs are present at significant levels in placenta and human milk ( 251 – 253 ), which suggests an important role for SPMs in health maintenance during a particularly vulnerable period of infant development. In a recent randomized placebo-controlled study from a Danish birth cohort, supplementation with a high dose of n-3-LCPUFAs (a dose corresponding to a 10–20-times increase of the normal intake) during the third trimester of pregnancy was associated with a significantly lower risk of asthma symptoms and fewer respiratory infections in children at 3 years ( 254 ). This effect was most prominent among children of women who had low pre-intervention EPA and DHA blood levels ( 255 ). Recent human studies have shown that increased intake of EPA and DHA results in higher concentrations of selected SPMs in the bloodstream ( 256 – 258 ). High doses of n-3 PUFAs reduce pain and other symptoms in patients with rheumatoid arthritis ( 259 , 260 ). Many of the mechanisms of action of EPA and DHA suggest that they reduce the pro-inflammatory response ( 242 , 243 ). However, the discovery of SPMs derived from EPA and DHA and the potency of those SPMs in animal models (see earlier) hints that their main action might be promotion of resolution.

Exercise enhances functional capacity, through increased aerobic capacity and muscle strength, improves quality of life and has the potential to protect from cardiovascular disease, type 2 diabetes mellitus, and certain types of cancer [reviewed in ( 267 )]. The potential mechanisms underlying exercise-mediated protection toward these disorders, include changes in body composition, neuro-hormonal status, as well as effects on resolution pathways. Indeed, acute increases in intramuscular IL-6 following exercise promote resolution processes by increasing the synthesis of anti-inflammatory cytokines such as IL-1RA and IL-10, and inhibit pro-inflammatory cytokines such as TNF-α ( 268 ). Exercise also modulates the production of the PUFA derived SPMs described earlier. Indeed, maximal physical exertion was found to result in a rapid post-exercise increase in the urinary excretion of arachidonic acid (AA) derived lipoxin A 4 (LXA 4 ) in healthy subjects ( 269 ). Similarly, EPA derived resolvin E1 (RvE1) transiently increases early in human serum following exercise and DHA derived resolvin D1 (RvD1) and protectins increased later during recovery ( 270 ). Interestingly, levels of LXA 4 are found to increase immediately after exercise in exhaled air condensate of asthmatic children with exercise induced bronchoconstriction (EIB) ( 271 ). The authors hypothesized that airway LXA 4 increases to compensate bronchoconstriction and to suppress acute inflammation, and that spontaneous bronchodilation after EIB may be due to LXA 4 . In murine studies in relation to asthma, physical exercise reduced asthma associated bronchial inflammation (IL-4, IL-5 expression and eosinophil infiltrate) which was associated with an increase of IL-10 ( 272 , 273 ). Exercise in animal models of colitis also reduced levels of TNF-α, and decreased markers of oxidative stress and histological damage to the colon in parallel to increased levels of the resolution-promoting and anti-inflammatory cytokine IL-10 ( 274 ).

There are a number of immune-mediated chronic diseases that might, at least in part, be controlled or prevented in an immune fit person, including IBD, allergy, and asthma developed in this review, as well as rheumatoid arthritis, chronic obstructive pulmonary disease (COPD), Parkinson's disease, Alzheimer's disease, multiple sclerosis, diabetes or myalgic encephalomyelitis. The focus of research on resolution of the immune response as a possible therapeutic approach has only been apparent in the twenty-first century and there is now increasing evidence that poor and/or inappropriate resolution of inflammation participates in the pathogenesis of IBD or asthma, being responsible in the long run for excessive tissue damage and pathology. This might now open a whole area in the development of personalized therapeutic options for chronic immune diseases driven in part by maladaptive, non-resolving inflammation. Moreover, since there is a strong link between a compromised immune system and the brain, individuals can experience a reduced quality of life and lack of well-being ( 280 , 281 ). Many chronic inflammatory diseases are associated with depression, anxiety, and reduced cognitive function ( 282 ) and it is becoming apparent that many brain diseases (psychiatric and neurological) are associated with activation of the immune system ( 281 , 283 ). Therefore, deviant immune fitness because of overreaction and poor or defective resolution of the immune system has an enormous impact and is a central issue in these chronic immune diseases.