6  Polyunsaturated fatty acids in the inflammation hallmark
Literature data mentions two main groups of polyunsaturated fatty acids (PUFAs), of plant or animal origin: omega-3 (n-3) and omega-6 (n-6). The most discussed molecules of each group are α-linolenic acid (ALA) and linoleic acid (LA), respectively. Functionally, notable among the omega-3 fatty acids are the eicosapentaenoic acid (EPA) and the docosahexaenoic acid (DHA) (Calder, 2015), although docosapentaenoic acid (DPA) is materializing as an important member of the n-3 family (Kaur et al., 2011). In human diets, the two essential PUFAs, LA, and ALA are usually derived from plant sources (such as seeds, nuts, seed oils and derived products), as only plants synthesize them (Calder, 2011). Linseeds and their oil habitually contain 45–55% of n-3 PUFAs, mainly ALA, whereas soybean or rapeseed oil, and walnuts only about 10% ALA (Kris-Etherton et al., 2002).
Another important food source of n-3 PUFAs are fatty fish and also other seafood or derived products, known as “fish oils”. All these products contain significant quantities of DPA, EPA and DHA, as a result of plankton and algae consumption and not as a result of endogenous synthesis (Poudyal et al., 2011). Noteworthy, the dominant PUFAs in fish, as well as fish oils, vary between species. For instance, cod liver oil contains more EPA than DHA, whereas tuna oil has a higher content of DHA (Calder, 2012). One portion of fish could bring into diet between 200 and 300 mg n-3 PUFAs, while 1 g of fish oil contains almost 30% EPA and DHA, highlighting the importance of diets including seafood. In the absence of fish or derived products, the daily intake of n-3 PUFAs in most adults is below 200 mg/day (Meyer et al., 2003). As plants produce much more LA than ALA, the former is the most customary PUFA in Western diets (Blasbalg et al., 2011). The daily dietary intake of ALA is 0.5–2 g, while of LA greatly increased in the last 50–60 years, along with a significant change of n-6/n-3 ratio (Calder, 2017), reaching as high as 20 in some Western-type diets (Calder, 2011).
LA and ALA share a common metabolic pathway in animals and humans (Poudyal et al., 2011). LA is metabolised by Δ6-desaturase to γ-linolenic acid (GLA), and later by elongase and Δ5-desaturase to arachidonic acid (ARA), while ALA is converted by the same enzymes into EPA, subsequently to DPA then, finally, to DHA. The rate of transformation from ALA to EPA and later to DHA is affected by several factors such as age, sex, or genetics (Baker et al., 2016). In addition, levels of ALA, LA and DHA are affected by chronic diseases as shown in a study comparing lipid levels between patients with autoimmune diseases and healthy controls (Tsoukalas et al., 2019b).
The conversion of dihomo-gamma-linolenic acid (DGLA) to ARA is regulated by insulin and ARA/EPA ratio is a sensitive marker for insulin resistance, a common denominator of most chronic inflammatory diseases. In a study of healthy adult volunteers AA to EPA ratio and the ARA precursor – DGLA, significantly changed with age (Tsoukalas et al., 2019a). The metabolism of n-6 PUFAs is more prevalent compared with n-3 PUFA, due to the fact that LA is much more common in actual diets (Blasbalg et al., 2011), although ALA, and not LA, is the preferred substrate for Δ6-desaturase (Calder, 2015).
Noteworthy, LA, ARA, EPA and DHA are important constituents of membrane phospholipids and have important roles in membrane function, which greatly influences cell activity (Burdge and Calder, 2015). The proportion of PUFAs found in the membrane is dependent on cell type, dietary intake and metabolism (Calder, 2017). For example, in healthy subjects receiving a typical Western diet, the percentages of DHA, EPA, and ARA, in mononuclear cells were 2.5, 0.5 and, respectively, 20% of total fatty acids (Calder, 2011).
The effects exerted by n-3 PUFAs are arbitrated either by the fatty acid molecules or by their bioactive metabolites from one of the three categories: protectins, resolvins, and maresins. The E-resolvins (such as E1, E2, and E3), are produced from EPA, while the D-series resolvins including D1, D2, D3, D4 and D5, neuroprotectins/protectins (NPD1/PD1), and the maresins (MaR1), are biosynthesized from DHA (Dalli et al., 2013; Serhan and Petasis, 2011). The biotransformation of n-3 PUFAs involves the COX and LOX pathways. S-resolvins, S-protectins, and S-maresins are produced from DHA and EPA via LOX pathway, while R-resolvins and R-protectins are derived from aspirin-activated COX-2 or cytochrome P450 metabolic transformation of DHA and EPA (Calder, 2015). In most cell types, ARA is highly predominant and exhibits direct links to inflammatory pathways, since it constitutes the substrate for enzymes such as cytochrome P450, cyclooxygenase or lipoxygenase, yielding mediators from the eicosanoid family. Indeed, the synthesis of pro-resolving mediators is increased when the diet is rich in n-3 PUFAs. Thus, in healthy subjects, a daily supplement for 3 weeks of 1 g DHA and 1.5 g EPA, quantified measurable levels of resolvins D1, D2, and 17-R resolvin D1 (>25 pg/ml plasma) (Mas et al., 2012). In obese women treated for 3 months with a daily supplement of 1.8 g EPA and DHA, DHA-derived pro-resolvins mediators were measured and found to be increased (RvD1 and RvD2 > 60 pg/ml plasma) (Polus et al., 2016).

6.1  Inflammatory-resolving effects induced by PUFAs
Nowadays, the mechanisms of action by which n-3 PUFAs regulate the inflammatory processes are widely investigated. The suppression of inflammation by n-3 PUFAs is associated with one of the following mechanisms (1) competitive inhibition of n-6 PUFA pathway; (2) modification of cell membrane composition; (3) affecting the formation of rafts or (4) direct anti-inflammatory effect of their bioactive metabolites (resolvins, protectins, and maresins) (Poudyal et al., 2011).
It has been demonstrated that the dietary supplementation with DHA and EPA from fish increases in a dose-response manner the content of DHA and EPA in the cell inflammation-responsible phospholipids (Calder, 2017); increased content of DHA or EPA in different tissues, such as adipose tissue or heart was also observed in correlation with their intake (Calder, 2015) (Fig. 3 ).
Fig. 3 The effect of omega-3 fatty acids and polyphenols in the regulation of the inflammatory response. Omega-3 fatty acids inhibit the inflammatory response by inhibiting PGE2 which promotes inflammation and NF-κB either directly, via the interaction with the transcriptional factors PPARs, or by inhibiting TLR2/4 which normally activates NF-κB. Moreover, omega-3 fatty acids regulate inflammation by activating MAPK and GPR120 which in turn inhibits inflammation. Polyphenols inhibit the inflammatory response by directly inhibiting NF-κB, or via the PPARs. They also promote fatty acid b-oxidation and inhibit VCAM-1, ICAM-1, MAPK pathway, PGE2 and COX-2 that all promote chronic inflammation (PGE2 –prostaglandin 2; NF-κB – nuclear factor kappa-light-chain-enhancer of activated B cells; PPARs – peroxisome proliferator-activated receptors; TLR2/4– toll-like receptor; MAPK – mitogen-activated protein kinase; GPR120 – G-protein coupled receptor 120; VCAM-1 – vascular cell adhesion molecule 1; ICAM-1 – intracellular cell adhesion molecule 1; COX-2 – cyclooxygenase 2; TNF-α – tumour necrosis factor alpha; MCP-1 – monocyte chemoattractant protein 1; AMPK – AMP kinase).
These n-3 PUFAs frequently substitute n-6 PUFAs like ARA, resulting in decreased availability of ARA for eicosanoid synthesis. EPA also inhibits ARA metabolism as a competitive substrate for COX-2, decreasing prostaglandin E2 (PGE2) production. In rats, dietary supplementation with ALA inhibits PG biosynthesis from ARA, while equivalent quantities of ALA and LA decreased up to 40% n-6 PUFAs incorporation in phospholipids (Calder, 2017). Furthermore, Rees et al. observed that a daily EPA intake of 2.7 g or 4.05 g for 3 months decreases the PGE2 production by lipopolysaccharide-stimulated mononuclear cells, while a lower dose of 1.35 g did not. EPA was integrated in a linear dose-dependent manner into mononuclear cell phospholipids and plasma. This study suggested a daily threshold in the range of 1.35–2.7 g EPA for the anti-inflammatory action (Rees et al., 2006).
Besides decreasing production of PGE2, DHA and EPA are also substrates for the biosynthesis of lipid derivatives, but the EPA-derived mediators as series-3 prostaglandins (PGD3) or series-5 leukotrienes are typically less biologically potent, having a lower ability to interact with relevant eicosanoid receptors (Calder, 2015). For example, EPA-derived leukotriene B5 (LTB5) is almost 100 times less active as a leukocyte chemoattractant than ARA-derived LTB4 (Calder, 2017). However, in some cases, EPA-derived mediators have similar potency with ARA-derived mediators. It appears that EPA-derived PGD3 inhibits the effect of the ARA-derived PGD2, due to a stronger interaction with the DP1 receptor compared to PGD2 (Wada et al., 2007). In other cases, EPA-derived mediators exhibited a similar magnitude of effect (e.g. inhibition of TNF-α production by blood monocytes) (Dooper et al., 2002).

6.2  Molecular targets of PUFAs
There are several pharmacological studies suggesting molecular targets for the anti-inflammatory effects of n-3 PUFAs and their metabolites: PPAR-γ, GPR120, CMKLR1 (known as ChemR23), BLT1(leukotriene B4 receptor 1), GPR32 and ALX/FPR2 (Im, 2012). Thus, resolvins E1 and D1 exhibited a higher affinity for these receptors compared to EPA or DHA. Chem R23 and BLT1 are receptors of resolvin E1, while GPR32 and ALX/FPR2, bind to lipoxin A4 and resolvin D1 with high affinity. GPR120 was reported to be a receptor of EPA and DHA (EC50 ~ 1–10 μM), while ALX/FPR2 to annexin I and lipoxin A4 (Serhan and Petasis, 2011). Furthermore, some studies on GPR120 KO mice suggest that n-3 PUFAs that activate GPR120, interact with β-arrestin 2, and suppress NF-κB activation and macrophage-mediated inflammatory responses (Oh et al., 2010). However, it is important to highlight that the in vivo anti-inflammatory effects of n-3 PUFAs in humans are minor and might only occur at high n-3 PUFA levels, it was demonstrated in vitro that BSA-conjugated n-3 PUFA are incapable of activating GPR120 (Im, 2012).
DHA and EPA are weaker agonists of PPAR-γ (EC50 ~ 10–100 μM), while their oxidized metabolites (such as protectin D1) are much more potent (Yamamoto et al., 2005). Also, ALA or ARA has a similar potency to DHA or EPA for on PPAR-γ, and higher for PPAR-α (Calder, 2015). As PPAR-γ activation reduces inflammatory responses, via the NF-κB pathway, this mechanism could partially explain the anti-inflammatory effects of n-3 PUFAs. Furthermore, n-3 PUFAs were reported to suppress NF-κB activation in a PPAR-γ-independent manner by binding to TLR-4 under certain conditions (Im, 2012). Taking into account all these reports, it looks like three mechanisms are employed by n-3 PUFAs to suppress inflammatory signalling via NF-κB: (1) preventing NF-κB nuclear translocation via PPAR-γ activation, (2) interfering with membrane activation of NF-κB via TLR4 and (3) interaction with GPR120 initiating an anti-inflammatory signalling cascade (Calder, 2015).
Resolvin D1 is a potent agonist to GPR32 and ALX/FPR2 (EC50 = 8.8 pM and 1.2 pM), while Resolvin E1 strongly binds to Chem R23 (Kd = 4.5 nM), reducing IL-12 production (Krishnamoorthy et al., 2010) and is a partial agonist to BLT1, so it induces NF-κB activation via BLT1, inhibiting neutrophil migration (Arita et al., 2007). For the other resolvins, protectins or maresins, the molecular targets are not yet identified.
Isolated n-3 PUFAs and their bioactive mediators were extensively examined in animal models of colitis or arthritis or using specific transgenic models. n-3 PUFAs and RvD1, RvD5, PD1 and MaR1 administration proved effective in animal models of colitis, decreasing inflammation and chemically induced colonic damage. The beneficial effects are, in all cases, correlated with the reduction of ARA-derived mediators in the colonic mucosa (Bosco et al., 2013; Charpentier et al., 2018; Gobbetti et al., 2017; Marcon et al., 2013). Surprisingly, aspirin-triggered resolvin D1 (AT-RvD1) displayed a stronger anti-inflammatory effect than RvD2 in experimental colitis, through lipoxin A4 receptor (ALX) activation (Bento et al., 2011). Furthermore, n-3 PUFAs as fish oil has shown not just anti-inflammatory effects in peripheral tissues, but several beneficial effects in obesity-induced animal models, such as improved lipid profile, decreased hepatic steatosis and insulin resistance (Bargut et al., 2015; Pimentel et al., 2013). Indeed, beneficial effects of DHA and EPA in adipose tissue were reported in mice fed a high-fructose diet, including modulating pro- and anti-inflammatory markers and ameliorating adipocyte abnormalities. The effects were significantly higher for DHA compared to EPA (Bargut et al., 2017).
Additionally to anti-inflammatory effects, correlated with down-regulation of IL-6 and TNF-α expression in liver, n-3 PUFAs also exhibited triglyceridemia lowering effects in diabetic rats via modulation of PPAR-α (Devarshi et al., 2013; Ghadge et al., 2016). Additionally, Lee et al. demonstrated that a diet with a high n-6/n-3 PUFAs ratio (~9) induced dysbacteriosis of the gut microbiota in obesity-induced T2DM or high-fat-diet treated rats, while a low ratio (~3) enhanced blood glucose homeostasis (Lee et al., 2019). The outcomes of the most recent animal studies are summarized in Table 5 .
Table 5 Mittigating inflammation in animal model studies – effects of PUFAs.
Tested compound(s) Animal model Main anti-inflammatory findings References
n-3 PUFA (fish oil or mix fish and olive oil or flaxseed oil) TNBS colitis ↓IL-1β; IL-12p70; ↓IL-6; ↓TNFα; ↑PGE3,↑ TXB3; ↑ LTB5 Bosco et al. (2013)
TNBS colitis ↓colon iNOS, ↓COX-2 expression, ↓IL-6, ↓LTB4, ↓TNFα production Charpentier et al. (2018)
DSS colitis ↓TNF-α; ↓ COX-2; ↑anti-inflammatory PG; Sharma et al. (2019)
Carrageenan induced inflammation ↓TNF-α; ↓ IL-6 Zadeh-Ardabili and Rad (2019)
STZ- diabetic rats ↑ gene expression PPRγ; ↓ NF-κB activity Ghadge et al. (2016)
STZ- diabetic rats ↓TNF-α; ↓ IL-6 Lee et al. (2019)
STZ-NIC diabetic rats ↑ PPAR-α only by flaxseed oil; both (flaxseed oil and fish oil):↑ D5 and D6 desaturases; ↓TNF-α; ↓ IL-6; Devarshi et al. (2013)
STZ-NIC diabetic rats ↑ renal SOD-1; ↑ GPx-1 expression; ↑ CAT; ↓ renal AGEs formation ↓AGE protein expression; ↓ IL-6; ↓ NF-κB expression Jangale et al. (2016)
STZ-NIC diabetic rats ↓ IL-1β; ↓TNFα; ↓IL-6; ↓IL-17 A; ↓MDA Zhu et al. (2020)
Wistar rats ↓ IL-6; ↓ TNF-α; ↓IL- 10 receptor Pimentel et al. (2013)
C57BL/6 mice ↓ NF-κB expression; ↓ IL-6; ↓ TNF-α Bargut et al. (2015)
EPA monogliceride DSS colitis ↓PMN infiltration; ↓ NF-κB activity; ↓IL-1β; ↓TNF-α; ↓ IL-6; ↓expression of COX2 in colon Morin et al. (2016)
ALA TNBS colitis ↓ IP-8, ↓ LTB4, ↓ colon NF-κB DNA binding activity Hassan et al. (2010)
EPA vs. DHA high-fructose fed C57BL/6J mice ↓ TNF-alpha and IL-6 gene expressions; ↓MCP-1 pERK and NFkB protein expressions Bargut et al. (2017)
EPA free fatty acid APCMin/+ FAP model ↓ COX-2 expression; ↑ EPA tissue uptake; ↓ lipid peroxidation Fini et al. (2010)
CAC model C57BL/6J mouse ↓PGE2; ↑ EPA tissue uptake Piazzi et al. (2014)
Endogenous conversion n-6 into n-3 PUFA CAC modelFat-1 mouse ↓ COX-2 expression; ↓ NF-κB activity; ↓PGE2 Han et al. (2016)
Chronic arthritisFat-1 mouse vs WT mouse ↓ IL-17; ↑mRNA expression of Foxp3 (in Fat-1 mouse) Kim et al. (2018)
AT-RvD1 DSS colitis/TNBS colitis ↓PMN infiltration; ↓ NF-κB activity and mRNA expression; ↓IL-1β; ↓MIP-2; ↓mRNA expression of VCAM-1, ICAM-1 Bento et al. (2011)
Adjuvant-induced arthritis ↓TNF-α; IL-1β Lima-Garcia et al. (2011)
RvD2 DSS colitis ↓IL-1β; ↓ murine KC (IL-8 human homolog) Campbell et al. (2010)
TNBS colitis ↓PMN infiltration; ↓ NF-κB activity and mRNA expression; ↓IL-1β; ↓MIP-2; ↓mRNA expression of VCAM-1, ICAM-1 Bento et al. (2011)
RvE1 DSS colitis ↓PMN infiltration; ↓TNF-α; ↓mRNA expression of IL-6, TNFα, IL-1β Ishida et al. (2010)
Collagen-induced arthritis No statistical significant effect on TNF-α de Molon et al. (2019)
RvD5 DSS colitis ↓PMN infiltration; ↓TNF-α; ↓ IL-6; ↓IL-1β; Gobbetti et al. (2017)
MaR1 DSS colitis/TNBS colitis ↓PMN infiltration; ↓ NF-κB activity; ↓IL-1β; ↓TNF-α; ↓ IL-6; ↓mRNA expression of ICAM-1 Marcon et al. (2013)
PD1 DSS colitis ↓PMN infiltration; ↓IL-1β only partially Gobbetti et al. (2017)
CAC – colitis associated cancer; CAT – catalase DSS – dextran sulfate sodium; GPx – glutathione peroxidase NIC – Nicotinamide; PD – protectin; SOD – superoxide dismutase; STZ – Streptozotocin; TNBS – trinitrobenzene sulphonic acid; EPA – Eicosapentaenoic Acid; DHA – Docosahexaenoic Acid; TNF – α-Tumour Necrosis Factor alpha; LTB – Leukotriene, PPAR – peroxisome proliferator-activated receptor; COX – cyclooxygenase; AGE – advanced glycation end products; TXB – Thromboxane; PG – prostaglandins; ICAM – Intercellular Adhesion Molecule; VCAM – Vascular Cell Adhesion molecule; IL – Interleukin; NF-κB, nuclear factor kappa B; PMN – polymorphonuclear leukocyte; MIP-2 –macrophage inflammatory protein 2; IP-8 – Isoprostane-8; Foxp3 – Forkhead box P3; pERK – protein kinase RNA-like endoplasmic reticulum kinase; MCP-1 – Monocyte chemoattractant protein-1; murine KC – murine chemokine.

6.3  n-3 PUFAs effects in humans
In healthy subjects, different daily doses of EPA and DHA up 1800 mg, administered up to 5 months, showed no significant effects on the CPR, IL-6, and TNF-α (Asztalos et al., 2016; Flock et al., 2014; Muldoon et al., 2016). A comparable conclusion was drawn by other authors. According to Rangel-Huerta's meta-analysis, consumption of 900 mg–2000 mg n-3 PUFAs does not change inflammatory biomarkers in healthy subjects (Rangel-Huerta et al., 2012). On the other hand, doses between 1250 and 2400 mg n-3 PUFA for 4 months lowered inflammation in sedentary and overweight middle-aged and older adults (Kiecolt-Glaser et al., 2012). Lp-PLA2, another anti-inflammatory marker was significantly reduced by a high dose of EPA (1800 mg) but not by DHA (Asztalos et al., 2016). Interestingly, only DHA modified the lipid profile by decreasing postprandial triglyceride concentrations and significantly increasing low-density lipoprotein cholesterol, with no significant changes in inflammatory biomarkers (Asztalos et al., 2016). In elderly subjects, daily supplementation with 2500 mg EPA and DHA, for 8 weeks, significantly reduced the plasma levels of fatty acids, IL-1β, IL-6, and TNF-α (Tan et al., 2018). Similarly, in obese patients who received different doses of combined n-3 PUFAs (380–1290 mg DHA and 360–460 mg EPA) for 2–3 months, the intervention reduced the expression of proinflammatory genes in adipocytes and systemic inflammatory markers sVCAM-1, CRP, IL-6, and TNF-α (Itariu et al., 2012; Polus et al., 2016). Furthermore, other relevant metabolic findings connected with n-3 PUFA treatment were reported, such as decreasing fasting triglycerides and insulin (Allaire et al., 2016; Polus et al., 2016) or decreasing fasting blood glucose in obese diabetics (Ellulu et al., 2016). Partially, these results are in line with the modest reduction in waist circumference and body-weight found in a meta-analysis (Bender et al., 2014). The authors indicated that the effect regarding waist circumference produced by fish intake or fish oil supplementation and might be greater in men than in women. The beneficial effect in overweight and obese adults concerning waist circumference and triglyceridemia was confirmed by two other meta-analyses (Du et al., 2015; Zhang et al., 2017).
As most of the cited trials utilize a mixture of DHA and EPA which may mask the effects of each compound, the individual effects of DHA or EPA in obese patients were also investigated, but the results were inconclusive. A significant reduction in serum IL-18 and adiponectin with DHA than with EPA was observed in one study (Allaire et al., 2016), while no differences between DHA and EPA in the expression of pro-inflammatory genes were observed in another study (Vors et al., 2017).
In patients with impaired glucose metabolism and T2DM in almost all studies, none of the combined EPA and DHA doses had any effect on IL-6, IL-1β, CRP, VCAM, and sICAM (Clark et al., 2016; Mocking et al., 2012; Sawada et al., 2016). Further, no effects on glycated haemoglobin (HbA1c) (Sawada et al., 2016; Wong et al., 2010), insulin (Clark et al., 2016) or lipid profile (Veleba et al., 2015) were found. Only Veleba et al. (2015), observed that HbA1c decreased significantly, and fasting blood glucose increased after n-3 PUFAs treatment for 24 weeks, but no other key findings. An overview of the most recent human studies where the inflammatory biomarkers as a result of PUFAs treatments, were assessed as the main outcome, is summarized in Table 6 .
Table 6 Clinical effects induced by PUFAs.
Intervention (n) Main anti-inflammatory findings Other relevant findings References
Healthy subjects
4-month intervention:➢ 1.5 g fish oil (1042.5 mg EPA and 174 DHA daily)
➢ 2.5 g fish oil (2085 mg EPA and 348 mg DHA daily)
➢ placebo 138 ↓TNF-α and ↓ IL-6 for both low and high dose groups ↓ n-6:n-3 ratio for both low and high dose groups Kiecolt-Glaser et al. (2012)
6-week intervention:➢ 600 mg EPA/day
➢ 1800 mg EPA/day
➢ 600 mg DHA/day
➢ placebo 121 No effect on hsCRP, TNF-α, IL-6, VCAM-1, ICAM-1 and fibrinogen Only High dose EPA ↓ Lp-PLA2;DHA: ↓ TG; ↑ LDLNo effect of low dose EPA Asztalos et al. (2016)
18-week intervention:1000 mg EPA + 400 mg DHA/day vs. placebo 261 No effect on serum CRP and IL-6 – Muldoon et al. (2016)
5-month intervention:➢ 300 mg EPA + DHA/day
➢ 600 mg EPA + DHA/day
➢ 900 mg EPA + DHA/day
➢ 1800 mg EPA + DHA/day
➢ placebo 125 No significant effect on IL-6 or CPR;Marginal effect on TNF-α observed at the highest dose (1800 mg) Higher RBC DHA was associated with lower TNF-α concentrations. Flock et al. (2014)
8-week intervention:2500 mg EPA + DHA/day vs. placebo 35 ↓IL-6, IL-1β and TNFα – Tan et al. (2018)
Obese patients
2-month intervention:➢ 460 mg EPA and 380 mg DHA/day
➢ control (butter fat) 55 ↓IL-6↓ inflammatory gene expression in adipose tissue↑release of anti-inflammatory eicosanoids in adipose tissue ↓TG Itariu et al. (2012)
3-month intervention: 360 mg EPA and 1290 mg DHA/day vs. placebo 59 ↓ VCAM-1; ↓ PECAM-1; ↓ hsCRP No effect on IL-6 ↓ TG; ↓insulin.No effect on TC, HDL, LDL, NEFA, FBG Polus et al. (2016)
10-week intervention:➢ 2700 mg EPA/day
➢ 2700 mg DHA/day
➢ placebo 154 ↓ IL-18 and ↑ adiponectin (DHA > EPA)No difference between EPA and DHA regarding effect on CRP, IL-6, TNFα ↓ TG; ↑ HDL (DHA > EPA)↑ LDL by DHA only in men Allaire et al. (2016)
10-week intervention:➢ 2700 mg EPA/day
➢ 2700 mg DHA/day
➢ control (corn oil) 154 EPA: ↑TRAF3 and PPARα expressionDHA: ↑ PPARα and TNFα expression both ↓CD14 expression No significant difference between EPA and DHA. Vors et al. (2017)
12-week intervention:➢ LSM + 600 mg EPA + DHA/day
➢ LSM
➢ placebo 29 LSM & n-3 PUFA ↑ adiponectin in comparison to LSMNo effect on IL6 No effect on leptin, LIF, follistatin, BDNF, and fasting triacylglycerol Sedláček et al. (2018)
Hypertensive and/or diabetic obese patients
8-week intervention:300 mg EPA + 200 mg DHA/day vs. control 64 ↓CRP ↓ FBG; ↓TG Ellulu et al. (2016)
Impaired glucose metabolism patients
6-month intervention:1800 mg EPA/day vs. placebo 107 ↓ CRP but similar effects in placebo ↑ HDL and ↓fasting TG; No effect on HbA1c and FBG Sawada et al. (2016)
9-month intervention:2388 mg EPA +1530 mg DHA/day vs placebo 36 No effect in IL-1B, IL-6, hsCRP, ICAM and VCAM No effect on FBG, insulin, HOMA-IR. Clark et al. (2016)
Type 2 diabetes mellitus
12-week intervention: 4000 mg (42% EPA + 25%DHA)/day vs. placebo 91 No significant effect on CPR ↓ TG; No effect on LDL, HDL, HbA1c Wong et al. (2010)
12-week intervention: 900 mg EPA/day vs. placebo 24 No effect on CRP, IL-6 and TNFα ↑ HDL and ↑ total cholesterol Mocking et al. (2012)
8-week intervention: 2700 mg EPA + DHA/day 84 ↓ IL-2 and ↓ TNFαNo effect on CRP None tested Malekshahi Moghadam et al. (2012)
12-weeks intervention:➢ 1000 mg EPA/day
➢ 1000 mg DHA/day
➢ placebo 60 No effect on serum CRP and MDA No effect on body weight, BMI or fat mass Azizi-Soleiman et al. (2013)
24-week intervention:➢ 2800 mg EPA + DHA + 15 mg pioglitazone/day
➢ 2800 mg EPA + DHA + placebo/day
➢ 5 mg pioglitazone/day
➢ placebo 60 No effect on SOD, TBARS, GSSG/GSH ↑ HbA1c; ↑FBGNo effect on TG, TC, HDL, LDL, NEFA, Leptin, Adiponectin Veleba et al. (2015)
3-month intervention: 1000 mg EPA + 1000 mg DHA/day vs. placebo 74 No effect on hsCRP, IL-6, TNF-α, ICAM-1, VCAM-1 No effect on insulin, HbA1c, adiponectin, leptin, and lipid levels Poreba et al. (2017)
Metabolic syndrome
90-day intervention:➢ 1800 mg EPA+ 1200 mg DHA + 10 mL extra virgin oil/day
➢ 1800 mg EPA + 1200 mg DHA + placebo/day
➢ 10 mL extra virgin oil/day
➢ placebo 102 No effect on CPR No effect on TG, TC, HDL, LDL, FBG, insulin, HOMA-IR Venturini et al. (2015)
Inflammatory bowel disease
8-week intervention: 3400–3600 mg n-3 PUFA (as salmon)/day 12 ↓ CPR, ↑ anti-inflammatory fatty acid indexNo effect TNF- α, MDA ↑ n-3 PUFAs,↑ n-3/n-6 ratio in plasma and rectal biopsies; No effect on the fecal calprotectine Grimstad et al. (2011)
90-day intervention: 2000 mg EPA/day 20 ↑ IL-10 expression; HES1, SOCS3, and KLF4 ↓ fecal calprotectinePartially redressed microbiota composition Prossomariti et al. (2017)
6-month intervention: 1000 mg EPA/day vs.placebo 60 No effect on CPR ↓ fecal calprotectine Scaioli et al. (2018)
EPA – Eicosapentaenoic Acid; DHA – Docosahexaenoic Acid; PG – Prostaglandins; LTB – Leukotriene; TNF-α – Tumour Necrosis Factor alpha; IL – Interleukin; CRP – C-Reactive Protein; ICAM – Intercellular Adhesion Molecule; VCAM – Vascular Cell Adhesion molecule; hsCRP – high sensitive C reactive protein; IL1RN – interleukin-1 receptor antagonist protein; LSM – lifestyle modification; NF-κB – nuclear factor-kappa B); PPAR – peroxisome proliferator-activated receptor; TRAF3 – TNF Receptor Associated Factor 3; Lp-PLA2 – lipoprotein-associated phospholipase A2; PECAM – platelet and endothelial cell adhesion molecule; COX – cyclooxygenase; LOX – lypoooxigenase; TBARS –thiobarbituric acid substances; SOD – superoxide dismutase; GSH – glutathione peroxidase; MDA – malonyldialdehyde; HOMA-IR – homeostasis model assessment of insulin resistance index; HES1 – transcription factor HES1; SOCCS3 – suppressor of cytokine signaling 3; KLF4 – Kruppel-like factor 4; HbA1c – Glycated haemoglobin; RBC – red blood cell; BDNF – Brain-derived neurotrophic factor; CD14 – cluster of differentiation 14; NEFA – Non-esterified fatty acids; TC – total cholesterol; TG – Triglycerides; LDL – low-density lipoproteins; HDL – high-density lipoproteins; FBG – fasting blood glucose; LIF – leukocyte inhibitory factor.

6.4  Dietary recommendations of n-3 LC-PUFA
Currently, the recommended daily intake of n-3 PUFAs varies as regarding expert committees, ranging from 250 to 500 mg for healthy adults (Flock et al., 2013). The European Food Safety Authority's (EFSA) recommendation is 250 mg/day n-3 PUFAs (EPA and DHA) for adult males and non-pregnant females, and supplementation of 100–200 mg/day of DHA in pregnancy (EFSA Panel on Dietetic Products, 2010). For children less than 2 years old, the daily recommendation corresponds to 100 mg DHA, while for children more than 2 years and in adolescents, the daily recommendation is similar to that for adults (EFSA Panel on Dietetic Products, 2010). The upper daily value of acceptable for EPA and DHA consumption has been set at 2000 mg, but higher doses used in clinical trials did not induce adverse effects (Rangel-Huerta et al., 2012).
There are also some additional recommendations in patients with high and extremely high fasting-triglyceride levels. The American Heart Association (AHA) recommends a daily intake of 500 mg–1000 mg DHA and EPA in patients with borderline levels (150–199 mg/dl), 1000–2000 mg in patients with high levels, and 2000–4000 mg in patients with very high triglyceride levels (>500 mg/dl) (Kris-Etherton et al., 2002). Importantly, further research should be conducted before any definitive daily recommendation or on the routine use of n-3 PUFAs in other chronic inflammatory diseases.
Studies show that n-3 PUFAs do not impact in a positive manner the metabolic profile of healthy subjects and induce variable effect on diabetes mellitus cases (no important effects on inflammation but a constant ability to increase the HDL level); on the other hand, a reduction of inflammation is obese and ageing patients as well as in those with metabolic syndrome.