Inappropriate, excessive or uncontrolled inflammation contributes to a range of human diseases. Inflammation involves a multitude of cell types, chemical mediators and interactions. The present article will describe nutritional and metabolic aspects of omega-6 (n-6) and omega-3 (n-3) fatty acids and explain the roles of bioactive members of those fatty acid families in inflammatory processes. Eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) are n-3 fatty acids found in oily fish and fish oil supplements. These fatty acids are capable of partly inhibiting many aspects of inflammation including leucocyte chemotaxis, adhesion molecule expression and leucocyte–endothelial adhesive interactions, production of eicosanoids like prostaglandins and leukotrienes from the n-6 fatty acid arachidonic acid and production of pro-inflammatory cytokines. In addition, EPA gives rise to eicosanoids that often have lower biological potency than those produced from arachidonic acid, and EPA and DHA give rise to anti-inflammatory and inflammation resolving mediators called resolvins, protectins and maresins. Mechanisms underlying the anti-inflammatory actions of EPA and DHA include altered cell membrane phospholipid fatty acid composition, disruption of lipid rafts, inhibition of activation of the pro-inflammatory transcription factor nuclear factor κB so reducing expression of inflammatory genes and activation of the anti-inflammatory transcription factor peroxisome proliferator-activated receptor γ. Animal experiments demonstrate benefit from EPA and DHA in a range of models of inflammatory conditions. Human trials demonstrate benefit of oral n-3 fatty acids in rheumatoid arthritis and in stabilizing advanced atherosclerotic plaques. Intravenous n-3 fatty acids may have benefits in critically ill patients through reduced inflammation. The anti-inflammatory and inflammation resolving actions of EPA, DHA and their derivatives are of clinical relevance.
Omega-6 and omega-3 fatty acids: nutritional and metabolic considerations
Omega-6 (n-6) and omega-3 (n-3) are the two major families of polyunsaturated fatty acids (PUFAs) . They can be of plant or animal origin. Plants synthesize linoleic acid (18:2n-6) and α-linolenic acid (18:3n-3), and consequently, these two fatty acids are found in many seeds, nuts, seed oils and products produced from seed oils like margarine. Neither linoleic nor α-linolenic acid can be synthesized in animals, and therefore, these two fatty acids are described as essential. Most plants produce much more linoleic than α-linolenic acid, and therefore, linoleic acid is usually the most prevalent PUFA in the human diet . Approximate daily intakes of linoleic and α-linolenic acids among adults living in Western countries are 10 and 1 g, respectively , although these will be affected by the exact diet consumed, the specific composition of different foods, season and sex among other factors.
Once in the body, linoleic and α-linolenic acids can be converted into other n-6 and n-3 PUFAs, respectively. This conversion involves a series of desaturation and elongation reactions (Figure 1). Desaturation occurs at carbon atoms below carbon number 9 of the acyl chain (counting from the carboxyl carbon). Linoleic acid can be converted into γ-linolenic acid (18:3n-6) by Δ6-desaturase and then γ-linolenic acid can be elongated by the enzyme elongase 5 to dihomo-γ-linolenic acid (20:3n-6; DGLA) (Figure 1). DGLA can be further desaturated by Δ5-desaturase to yield arachidonic acid (20:4n-6; ARA) (Figure 1). Using the same series of enzymes as used to metabolize n-6 PUFAs, α-linolenic acid is converted into timnodonic acid which is more commonly known as eicosapentaenoic acid (20:5n-3; EPA) (Figure 1). In mammals, the pathway of desaturation and elongation mainly occurs mainly in the liver. The desaturation enzymes use NADPH and molecular oxygen, and are located on the endoplasmic reticulum. The genes encoding Δ6- and Δ5-desaturase are known as fatty acid desaturase 2 and 1 (FADS2 and FADS1), respectively. It is evident from the pathway shown in Figure 1 that there is competition between the n-6 and n-3 PUFA families for metabolism. The Δ6-desaturase reaction is rate-limiting in this pathway. The preferred substrate for Δ6-desaturase is α-linolenic acid, but, because linoleic acid is much more prevalent in most human diets than in α-linolenic acid, metabolism of n-6 fatty acids is quantitatively the more important. The activities of Δ6- and Δ5-desaturases are regulated by nutritional status, hormones (e.g. insulin) and feedback inhibition by end products. Polymorphisms in the FADS1 and FADS2 genes have been associated with differences in status of n-6 and n-3 PUFAs , presumably due to the polymorphisms resulting in different activities of the desaturase enzymes.
Further conversion of ARA into osbond acid (22:5n-6) and of EPA into cervonic acid more commonly known as docosahexaenoic acid (22:6n-3; DHA) occurs by a complex pathway. This involves chain elongation catalyzed by elongase 5, a second chain elongation catalyzed by elongase 2 or 5, desaturation by Δ6-desaturase and then removal of two carbon atoms as acetyl-CoA by limited β-oxidation in peroxisomes (Figure 1).
ARA is found in red meat, eggs and organ meats. Intakes among adults in Western countries are reported to be ∼0.5 g/day, but this will very much depend on the nature of the diet. EPA and DHA are found in seafood, especially in fatty fish like salmon, tuna, mackerel, herring and sardines, and in omega-3 supplements like fish oil . Both ARA and DHA are important constituents of human breast milk. In most adults eating typical Western diets, intake of EPA and DHA is low and <0.2 g/day, although this will be higher in those who regularly consume fatty fish or use omega-3 supplements.
Arachidonic acid, eicosanoids and inflammation
n-6 and n-3 PUFAs are important constituents of membrane phospholipids, where they are commonly found at the sn-2 position . Linoleic acid, DGLA, ARA, EPA and DHA are all common membrane constituents, with the relative amounts present depending on availability from the diet, metabolic processes, the specific cell-/tissue-type and the nature of the phospholipids present, since different types of phospholipid have different fatty acid compositions. In most cell types, ARA is the most prevalent PUFA present. For example, in mononuclear cells taken from the blood of healthy volunteers consuming a typical Western diet, the mean proportions of linoleic acid, DGLA, ARA, EPA and DHA were 10, 2, 20, 0.5 and 2.5% of total fatty acids . The high content of ARA provides a direct link with inflammation, since ARA released from cell membrane phospholipids acts as a substrate for cyclooxygenase (COX), lipoxygenase (LOX) and cytochrome P450 enzymes to yield the eicosanoid family of mediators (Figure 2). Eicosanoids are important regulators and mediators of inflammatory processes [7–9] and include prostaglandins (PGs), thromboxanes and leukotrienes (LTs). The process of eicosanoid synthesis is up-regulated by inflammatory stimuli through both enzyme activation (e.g. of phospholipase A2 which releases ARA from membrane phospholipids) and up-regulation of the expression of the genes encoding the enzymes. Many anti-inflammatory therapies, such as non-steroidal anti-inflammatory drugs (NSAIDs) and COX inhibitors, target ARA metabolism, indicating the intimate relation of ARA metabolism to inflammatory processes.
Outline of the pathway of conversion of arachidonic acid into eicosanoids.
Omega-3 fatty acid enrichment of cells involved in inflammation and eicosanoid profiles
Increased intake of EPA and DHA from supplements results in an increased content of EPA and DHA in the phospholipids of cells involved in inflammation [6,10–12]. This increase occurs in a time- [6,11,12] and dose-dependent [11,12] manner. Incorporation of EPA and DHA is partly at the expense of ARA [6,10,11,13], resulting in decreased availability of the usual substrate for the synthesis of eicosanoids. Furthermore, EPA has been shown to inhibit ARA metabolism and to decrease expression of the COX-2 gene (see  for references). Several animal studies have shown that production of ARA-derived eicosanoids like PGE2 is decreased by EPA or DHA feeding [15–17]. Consistent with this, many studies in healthy human volunteers or in patients with chronic inflammatory diseases [e.g. rheumatoid arthritis (RA) or inflammatory bowel disease] have described decreased production of ARA-derived 2-series PGs and 4-series LTs by inflammatory cells following use of supplements providing high doses of EPA and DHA for a period of weeks to months (see  for references). Rees et al.  identified that an EPA intake of 1.35 g/day for 3 months was not sufficient to influence ex vivo PGE2 production by lipopolysaccharide (LPS)-stimulated mononuclear cells, whereas an EPA intake of 2.7 g/day significantly decreased PGE2 production. This suggests a threshold for an anti-inflammatory effect of EPA of somewhere between 1.35 and 2.7 g EPA per day. Rees et al.  explored the relationships between the proportions of ARA and EPA in mononuclear cells and the ability of those cells to produce PGE2 when stimulated by LPS. They identified that there was ‘a significant positive relation between PGE2 production by LPS-stimulated mononuclear cells and mononuclear cell phospholipid ARA content and a significant negative relation between PGE2 production by LPS-stimulated mononuclear cells and mononuclear cell phospholipid EPA content’, ‘a significant positive relation between PGE2 production and the ratio of ARA to EPA in mononuclear cell phospholipids’ and ‘a significant negative relation between the change in PGE2 production and the change in mononuclear cell phospholipid EPA content’. These relations indicate a close link between the presence of ARA and EPA in cell membrane phospholipids and the ability of those cells to produce inflammatory mediators and furthermore that one strategy to regulate inflammation is to modulate the amounts of ARA and EPA in cell membrane phospholipids.
EPA is also a substrate for the COX, LOX and cytochrome P450 enzymes that produce eicosanoids; the metabolism of EPA is analogous to that shown for ARA in Figure 2. However, the mediators produced from EPA have a different structure from those produced from ARA (see  for references), and this structural difference between ARA and EPA-derived eicosanoids renders the latter less biologically potent (see  for references). For example, LTB5 produced from EPA is 10 to 100 times less potent as a leucocyte chemoattractant than LTB4 produced from ARA (see  for references). One reason for this reduced biological potency is that eicosanoid receptors typically have a lower affinity for the EPA-derived mediator than for the ARA-derived one . Thus, EPA generally results in decreased production of potent eicosanoids from ARA and increased production of weak eicosanoids.
Omega-3 fatty acids, inflammatory signalling and inflammatory proteins
In addition to modifying the profile of eicosanoids involved in inflammatory processes, n-3 PUFAs affect production of many inflammatory proteins including cytokines and adhesion molecules. Much work has focused on the effects on the classic pro-inflammatory cytokines tumour necrosis factor (TNF)-α, interleukin (IL)-1 β and IL-6. Many studies using cell culture and animal feeding models have demonstrated that EPA and DHA decrease production of these three cytokines in response to LPS (see  for references). Some studies also report that EPA and DHA increase the concentration of the anti-inflammatory cytokine IL-10 (see  for references). Several studies providing EPA and DHA as supplements to healthy human volunteers have reported decreased production of TNF, IL-1β and IL-6 by LPS-stimulated monocytes or mononuclear cells [19–22], although not all studies report this effect. Some of the studies that fail to show an effect of n-3 PUFAs on cytokine production have provided <2 g EPA + DHA per day, which may be an insufficient dose.
n-3 PUFAs, especially DHA, have been shown to decrease expression of adhesion molecules like vascular cell adhesion molecule (VCAM)-1 and intercellular adhesion molecule (ICAM)-1 on the surface of endothelial cells and monocytes in culture (see  for references). Feeding studies in rats reported that EPA and DHA lower adhesion molecule expression on macrophages  and lymphocytes . The lower adhesion molecule expression has a functional effect, resulting in decreased adhesion between inflammatory cells and endothelial cells [24–26], including under conditions of flow [27,28]. Supplementing the diet of healthy humans with n-3 PUFAs resulted in a lower level of expression of ICAM-1 on the surface of blood monocytes stimulated ex vivo with interferon-γ , while consumption of EPA + DHA by patients with peripheral artery disease decreased the adhesion of their monocytes to endothelial cell monolayers in culture . Supplementation with EPA + DHA decreased circulating concentrations of soluble VCAM-1 in elderly subjects , while supplementation with EPA decreased the concentrations of soluble ICAM-1 and soluble VCAM-1 in the bloodstream of patients with metabolic syndrome .
Effects of n-3 PUFAs on cytokines and other proteins involved in inflammation are related to alterations in the expression of genes encoding those proteins (see ). This suggests that n-3 PUFAs have an effect on the signalling pathways that control gene expression in inflammatory cells. Nuclear factor kappa B (NF-κB) is one of the main transcription factors involved in up-regulation of the genes encoding proteins involved in inflammation including many cytokines, adhesion molecules and COX-2. In its inactive state, NF-κB exists as a trimer in the cytosol; one of the subunits of this trimer is an inhibitory subunit called inhibitory subunit of NF-κB (IκB). NF-κB is activated through a signalling cascade triggered by various extracellular inflammatory stimuli, including LPS binding to Toll-like receptor (TLR) 4. This cascade involves phosphorylation of IκB, which then dissociates from the remaining dimer and is degraded. This allows translocation of the remaining NF-κB dimer to the nucleus where it binds to response elements and up-regulates inflammatory gene expression. As described earlier, EPA and DHA decrease cell surface expression of adhesion molecules and production of inflammatory cytokines and COX-2 metabolites. One common mechanism to explain these effects would be an impact on the NF-κB system. In line with this, EPA or fish oil decreased LPS-induced activation of NF-κB in human monocytes [32–34], and this was associated with decreased IκB phosphorylation [34,35]. Likewise, DHA reduced NF-κB activation in response to LPS in cultured macrophages  and dendritic cells [37,38], an effect that involved decreased IκB phosphorylation . Myeloid differentiation primary response gene 88 (MyD88) is a cell membrane-associated adapter protein used by TLR4 in the early stages of the signalling cascade that eventually activates NF-κB. Inhibition of COX-2 expression by DHA in macrophages bearing constitutively active TLR4 did not occur in macrophages not bearing constitutively active MyD88 , suggesting that DHA acts upstream of MyD88. It is known that TLR4, MyD88 and other signalling proteins associate into lipid rafts in inflammatory cells exposed to endotoxin. Exposure of macrophages to DHA inhibited the ability of LPS to promote recruitment of signalling proteins into rafts . Thus, the effect of DHA, and perhaps also EPA, on inflammatory signalling initiated through TLR4 and affecting NF-κB appears to relate to its ability to disrupt raft formation within the membrane of inflammatory cells.
A second mechanism by which EPA and DHA might influence NF-κB activation involves peroxisome proliferator-activated receptor (PPAR)-γ. This is a transcription factor which acts in an anti-inflammatory manner, probably because it physically interferes with the translocation of NF-κB to the nucleus. PPAR-γ can be activated by n-3 PUFAs [40–43] and DHA induced PPAR-γ in dendritic cells , an effect associated with inhibition of NF-κB activation and reduced production of the pro-inflammatory cytokines TNF-α and IL-6 following LPS stimulation . Furthermore, DHA induced many known PPAR-γ target genes in dendritic cells , suggesting that this as an important anti-inflammatory mechanism of action of DHA and perhaps also of EPA. The EPA derivatives PGD3 and 15-deoxy-PGD3 activate PPAR-γ in adipocytes, a process linked to the induction of the anti-inflammatory adipokine adiponectin . Thus, activation of PPAR-γ may be one of the anti-inflammatory mechanisms of action of EPA and DHA, and this may also link to the inhibition of NF-κB activation described earlier.
Omega-3 fatty acids and resolution of inflammation
A huge advance in the field of PUFAs and inflammation has been the discovery and elucidation of the structures, actions and mechanisms of so-called pro-resolving lipid mediators produced from EPA and DHA. These mediators include resolvins produced from EPA (E-series) and DHA (D-series) and protectins and maresins produced from DHA (see [46–48]); protectins are also referred to as neuroprotectins when generated within neural tissue. The synthesis of resolvins, protectins and maresins involves the COX and LOX pathways (Figure 3), with different epimers being produced in the presence and absence of aspirin (see [46–48]). These pathways operate in a transcellular manner with the early steps occurring in one cell type and the latter in another. More recently, analogous compounds have been shown to be produced from the third seafood-derived n-3 PUFA docosapentaenoic acid (22:5n-3) .
Overview of the pathways of synthesis of specialized pro-resolving mediators from EPA and DHA.
Resolvin synthesis is increased by feeding laboratory rodents a diet rich in EPA and DHA . A few human studies have demonstrated increased levels of some pro-resolving mediators in the blood of humans consuming increased amounts of EPA and DHA [51–53].
The biological effects of resolvins, protectins and maresins have been examined extensively in cell culture and animal models of inflammation (see [46–48]). These models have shown them to be anti-inflammatory and inflammation resolving. For example, resolvin E1, resolvin D1 and protectin D1 all inhibit transendothelial migration of neutrophils, so preventing the infiltration of neutrophils into sites of inflammation; resolvin D1 inhibits IL-1β production and protectin D1 inhibits TNF-α and IL-1β production (see [46–48] for references). Resolvins reduce inflammation and protect experimental animals in models of inflammatory disease including arthritis , colitis  and asthma [56–59]. In common with eicosanoids, the biological activities of resolvins are mediated by specific G-protein coupled receptors (see ).
Translation of the anti-inflammatory effects of omega-3 fatty acids to human disease
RA is a chronic inflammatory autoimmune disease that affects the joints, with infiltration of activated T lymphocytes, macrophages and antibody-secreting B lymphocytes into the synovium (the tissue lining the joints) and proliferation of fibroblast-like synovial cells called synoviocytes. These cells and new blood vessels form a tissue termed pannus which leads to progressive destruction of cartilage and bone. This is most likely due to cytokine- and eicosanoid-mediated induction of destructive enzymes such as matrix metalloproteinases (MMPs). Expression of COX enzymes is increased in the synovium of patients with RA, and synovial fluid from patients with RA contains high levels of pro-inflammatory eicosanoids and cytokines (see  for references). Patients with RA are treated with NSAIDs that inhibit COX metabolism of ARA, indicating the importance of eicosanoids in the disease. Rodent models of arthritis demonstrate benefit from n-3 PUFAs [60–62]. Supplemental EPA + DHA decreased inflammatory cytokines and eicosanoids in the blood of patients with RA (see  for references). These effects should reduce pain and cartilage destruction; if pain is reduced then patients may decrease their use of pain-controlling drugs like NSAIDs. Indeed, randomized controlled trials of n-3 PUFA supplements in RA report improvements in several clinical outcomes including duration of morning stiffness, number of tender or swollen joints, joint pain, time to fatigue, grip strength and use of NSAIDs (see  for references). Meta-analyses of trials of n-3 PUFA supplements in RA identified reduced tender joint count and morning stiffness  and decreased patient assessed joint pain, duration of morning stiffness, number of painful and/or tender joints, and use of NSAIDs . Thus, there is fairly robust evidence of the efficacy of EPA + DHA in RA.
Atherosclerotic plaque rupture
The rupture of an atherosclerotic plaque can initiate a myocardial infarction or stroke, depending on the location of the plaque. Plaque rupture is an inflammatory event involving the release of MMPs from inflammatory macrophages within the blood vessel wall [66,67]. The MMPs degrade the proteins that form the fibrous cap that serves to restrain the plaque within the vessel wall. Indeed, the characteristics of an atherosclerotic plaque that make it vulnerable to rupture include a thin fibrous cap and increased numbers of inflammatory cells such as macrophages [67–69]. EPA and DHA could stabilize atherosclerotic plaques by decreasing infiltration of inflammatory cells into the plaques and/or by decreasing the activity of those cells once in the plaque. A study in patients awaiting surgical removal of plaques from the carotid artery showed that EPA and DHA are incorporated from n-3 PUFA supplements into advanced atherosclerotic plaques and that this incorporation is associated with fewer macrophages in the plaque and with structural changes consistent with increased plaque stability . A second study demonstrated that a higher EPA content of the plaque is associated with lower plaque inflammation and less instability, and that the levels of genes encoding several MMPs in advanced carotid plaques are lower in patients treated with EPA + DHA than in those receiving placebo . A study in apoE- and LDL-deficient mice fed a Western-type diet or the same diet plus EPA reported increased plaque collagen and decreased macrophage numbers in the aortic plaques of the mice fed EPA . Increased plaque stability with n-3 PUFAs could account for fewer cardiovascular events and lower mortality from cardiovascular disease (see  for references).
Critical illness and sepsis
Sepsis is a life-threatening condition in which an exaggerated inflammatory response driven by an uncontrolled infection causes injury to the host's organs. Patients with critical illness and sepsis show increased NF-κB activation in blood mononuclear cells  and elevated circulating concentrations of inflammatory cytokines [74–76], the latter being predictive of mortality [75,76], establishing a direct link between inflammation and adverse outcome. The phenomenon can be mimicked, in part, by administration of LPS to experimental animals , although LPS is sterile. In such models of endotoxaemia, n-3 PUFAs lessen the metabolic response , decrease the inflammatory response  and improve survival [80,81]. Intravenous infusion of lipid emulsions containing fish oil as a source of n-3 PUFAs in elderly critically ill patients decreased circulating TNF-α concentration and increased the concentration of the anti-inflammatory cytokine IL-10 . Septic patients receiving intravenous fish oil showed lower circulating IL-6, better gas exchange and shorter hospital stay than those seen in the control group . Meta-analyses of studies of intravenous fish oil in critically ill patients are generally supportive, although there are some inconsistencies . In their meta-analysis, Manzanares et al.  concluded that ‘fish oil-containing lipid emulsions may be associated with a reduction in infections and also could be associated with a reduction in duration of ventilation and hospital length of stay’.
Summary and conclusions
Inappropriate, excessive or uncontrolled inflammation contributes to a range of human diseases. It involves a multitude of cell types, chemical mediators and interactions. EPA and DHA are the major n-3 PUFAs found in oily fish and fish oil supplements. There is substantial evidence that these fatty acids are capable of partly inhibiting many aspects of inflammation including leucocyte chemotaxis, adhesion molecule expression and leucocyte–endothelial adhesive interactions, production of eicosanoids like PGs and LTs from the n-6 fatty acid ARA and production of inflammatory cytokines like TNF-α and IL-1β. EPA and DHA act through a variety of mechanisms summarized in Figure 4. Many effects of EPA and DHA on inflammatory processes appear to be mediated by, or at least are associated with, changes in fatty acid composition of cell membranes. Changes in fatty acid composition can modify lipid raft formation, cell signalling leading to altered gene expression and the pattern of lipid mediators produced (Figure 4). Cells involved in the inflammatory response are typically rich in the n-6 fatty acid ARA, but the contents of ARA and of EPA and DHA can be altered through administration of EPA and DHA. Eicosanoids produced from ARA, like PGE2 and 4-series LTs, have roles in inflammation. EPA also gives rise to eicosanoids, but these are usually less potent than those produced from ARA. EPA and DHA give rise to resolvins and DHA gives to protectins and maresins that are anti-inflammatory and inflammation resolving. Increased membrane content of EPA and DHA (and decreased ARA content) results in a changed pattern of production of eicosanoids, resolvins and protectins. Thus, fatty acid exposure and the fatty acid composition of human inflammatory cells influence the function of those cells, and the contents of ARA, EPA and DHA appear to be especially important (Figure 4). As a result of their anti-inflammatory actions, EPA and DHA may have therapeutic efficacy in inflammatory diseases. Work with animal models of RA, colitis and asthma has demonstrated efficacy of fish oil and of mediators derived from EPA and DHA, like some of the resolvins (see  for references). There have been many clinical trials of fish oil in patients with RA [63–65] or other inflammatory diseases (see ). Many trials in RA report clinical improvements (e.g. improved patient assessed pain, decreased morning stiffness, fewer painful or tender joints and decreased use of NSAIDs), and when the trials have been pooled in meta-analyses, statistically significant clinical benefit has emerged [64,65]. A few human trials, supported by findings from animal models, indicate that EPA and DHA have a role in stabilizing advanced atherosclerotic plaques, which could reduce cardiovascular events and mortality . Intravenous n-3 fatty acids may have benefits in critically ill patients [82–85]. Thus, the anti-inflammatory and inflammation resolving effects of EPA, DHA and their derivatives are clinically relevant acting to improve patient outcome.
Depiction of the key anti-inflammatory actions of n-3 fatty acids.
fatty acid desaturase
intercellular adhesion molecule
inhibitory subunit of NF-κB
myeloid differentiation primary response gene 88
nuclear factor kappa B
non-steroidal anti-inflammatory drug
peroxisome proliferator-activated receptor
polyunsaturated fatty acid
tumour necrosis factor
vascular cell adhesion molecule
The author serves on Scientific Advisory Boards of Pronova BioPharma, Danone/Nutricia, Smartfish and DSM, acts as a consultant to Cargill and Baxter Healthcare, and in the last 3 years has received speaking honoraria from Pronova BioPharma, Fresenius Kabi, Abbott Nutrition, Smartfish and DSM.