Fish Oil

Fish oil is a term used to refer to a certain solution of fatty acids (a component of dietary fat) called eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA); they are referred to as oils from 'fish' (despite a small presence in poultry and presence in neural tissue of all species) as serum EPA and DHA concentrations tend to scale with fish intake, with Americans^[30] usally having lower serum levels of these two fatty acids than Japanese^[31] and Inuit (Greenland)^[32] persons. Fish oil is the bioactive of Cod Liver Oil (alongside Vitamin A and Vitamin D) and Krill Oil (in the form of phospholipids rather than triglycerides).

Any fish oil product may contain more omega-3 fatty acids that are neither EPA nor DHA (for example, the intermediate called DPA^[33]) and may contain fatty acids that do not belong to the omega-3 class; exact levels of fatty acids and omega-3 fatty acids depend on the source of fatty acids and processing, and tend to be stated on the label.

Fish oil refers to two fatty acids, called eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). These are both omega-3 fatty acids. They are found in very high levels in fish, as compared to terrestrial animals.

With other additives that may or may not be included, depening on processing:

• Methylmercury contamination^[34][35][36] although this is dependent on initial source of the oil (fish) and company dependent processing. More info in the safety and toxicology section on this topic
• PCB/dioxin contamination^[37][38]
• Organochlorine contamination^[34][39].

Generally any toxin that is released into the water and is fat-soluble in nature (and thus can be stored in the tissues of fish) has potential to be found in fish oil supplementation. If possible, fish oil supplements from non-predatory and non-bottom feeding fish (such as sardines, herring, or mackerel) should be used, as mercury levels (used as a standard by which to assess 'contaminants' in general) typically are elevated in fish that consume other fish and build up stores of mercury and PCBs^[40][41], and bottom-feeders that feed on carcasses of fish and accumulate toxins and minerals.^[42] Depth of forage may also be correlated with mercury levels, making surface fish safer.^[43]

Fish oil can have the same contaminants as fish, but this is dependent on the quality of the processing and the source of oil. Low predatory fish like sardine, cod and prawn (Krill Oil) are safer and should be used, if the higher cost of oil is not a concern.

The active components of fish oil are generally considered to be the two omega-3 (also written as n3 or ω3) fatty acids eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). Both fatty acids are similar in structure, although DHA a tad longer (eico- refers to a fatty acid with 20 carbons in its chain, while docosa- refers to a carbon chain 22 in length). The term Omega (ω) is used to refer to the 'end' of the fatty acid, and the omega designation of any fatty acid is the distance from the end of the chain where the first double bond occurs.

Any fatty acid that has a double bond is unsaturated (if only once, monounsaturated; if many times like both fish oils, polyunsaturated or PUFA) and thus has an omega designation; saturated fatty acids lack double bonds and thus have no omega designation. The diagram below indicates double bonds via parallel lines

The shortened nomenclature for EPA is 20:5n3 while 22:6n3 is for DHA; the first number refers to the carbons in the sidechain, while the second refers to the overall count of double bonds (and the final number referring to the omega designation).

Eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) are omega-3 polyunsaturated fatty acids, that are 20 and 22 carbons in length, respectively.

For dietary ingestion of EPA and DHA from fish products, there are a few choices; triglycerides, reesterified triglycerides, ethyl ester (the pharmaceutical Lovaza), and phospholipid (crustacean sources such as Krill oil). These four all confer dietary EPA and DHA, but krill oil is approximately a third better absorbed than triglyceride form^[44][45] and ethyl ester the reverse (if fish oil triglycerides are standardized to 100% absorption, ethyl esters reach 73%^[46]), which seem to confer less benefits relative to triglycerides on a gram per gram basis.^[47][48] Re-esterification of triglycerides appears to enhance their absorption (124% of triglycerides^[46]), but although this is somewhat comparable to phospholipid formation, the two have not been directly compared.

Due to fish oil supplements being derived from fish, they are not classified as vegan. Currently, the only significant vegan source of DHA is microalgae (phytoplankton)^[49] and its supplement referred to as 'algae oil'.^[50][51] The DHA component is equivalent to fish sourced DHA in cardiovacular health^[52][53] and seems to have comparable safety^[54] suggesting that they are interchangeable.

Other vegetarian sources of omega-3 fatty acids tend to have the parent structure of alpha-linolenic acid (ALA; not to be confused with alpha-lipoic acid which shares the acronym) and significant plant sources of ALA include hemp protein and flaxseed, while supplements with a smaller ALA quantity include Spirulina and Chlorella.

Vegan sources of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) are limited to supplements derived from phytoplankton.

As all fish oil supplementation is derived from fish, these products are animal byproducts. Their usage would thus not be vegan. It should be noted that absolute depletion of EPA and DHA in the blood of vegans does not occur,^[55] which is thought to be due to adaptation mechanisms^[56] such as increased hepatic production of DHA from the omega-3 precursor ALA^{[57] [58][59]} with no influence on cerebral synthesis^[60] and downregulation of enzymes that consume DHA including cyclooxygenase-1 (COX1)^[61] and phospholipase A₂^[62], which prolongs the half-life of DHA.^[63] It should be noted that DHA synthesis is decreased with dietary DHA surplus^[64] and this adaptive effect is thought to be the reason why there is little to no clinically relevant omega-3 deficiencies in society.^[65][66]

Although vegans and vegetarians tend to have reduced circulating EPA and DHA concentrations, they also possess adaptive mechanisms to attenuate the decline. An absolute depletion of these fatty acids is not seen in one living system. An absolute depletion requires generational deprivation.

Fatty acids (polyunsaturated) are converted from one another in the body via an enzyme class known as desaturases, and the dietary requirement for omega-3 and omega-6 (the essential fatty acids) are due to a lack of the delta (Δ) 15 and Δ12 desaturase, respectively.^[67] The enzyme known as Δ6 desaturase is the rate limiting step for producing DHA in the body,^[68][69] and supplementation of fish oil tends to circumvent this rate limit; alternatively, the enzyme itself can be targeted to increase DHA concentrations in the body (seen in the Fat-1 mouse line, which naturally has all the benefits of fish oil supplementation without requiring ingestion thereof^[70][71][72]) and this is known to be induced by fucoxanthin.^[73][74] Without any modifications and assuming average fish intake, the rate of ALA conversion into DHA tends to be in the 2-10% range.^[75][76]

After ingestion, some polyunsaturated fatty acids can be converted into EPA and, via EPA, are turned into DHA via a chain of enzymes, where Δ6 desaturase is the rate limiting step. This enzyme is an active regulator of bodily DHA levels. Increasing the activity of this enzyme will increase DHA stores in the body and is likely synergistic with dietary alpha-lipoic acid (ALA) intake in increasing bodily DHA stores.

In animal models (rats^[77] and primates^[78]) a DHA deficiency in critical tissue (retina and brain) only occurs after restricting multiple generations but does result in functional impairment of the eyes and brain.^[79][80]

True omega-3 deficiencies (which are related to DHA deficiency) can only be induced through multi-generational depletion of dietary fish oils. Issues with visual processing are the most adverse of the cognitive effects that can result from this depletion.

An enzyme located on the cellular membrane called Phospholipase A₂ is able to hydrolyze (free up) a fatty acid from the middle of a glycerol backbone upon activation, and due to both DHA and arachidonic acid being in the middle of a triglyceride frequently they are frequently mobilized by Phospholipase A₂.

Phospholipase A₂ is stimulated by seizures,^[81][82] ischemia,^[83] and NMDA-receptor stimulation^[84] as well as various inflammatory cytokines (IL-1β,^[85] TNF-α,^[86] and PMA^[87]) and oxidation products.^[88] The molecules that stimulate phospholipase A₂ tend to be associated with cellular and metabolic damage, and thus eicosanoids are thought to be hormetically induced.

The enzyme phospholipase A₂ releases stored polyunsaturated fatty acids in response to stress. Eicosanoids from docosahexaenoic acid (DHA) and arachidonic acid (AA) are released after stressors influence the cell. Stressors include inflammation and oxidation.

The cellular membrane ratio of omega 3 to 6 fatty acids is important as phospholipase A₂ is not discriminatory as to which polyunsaturated fatty acids it releases, and the eicosanoids that are produced when a cell is stressed correlate directly with the polyunsaturated fats that make up the membrane.

The standard western diet (mostly in reference to the US) tends to have a ratio highly favoring omega-6 fatty acids in the range of 15-20:1 approximately (varies depending on source).^[89][90][91] A modern European diet (using data from the UK and Britian) is not significantly better at around 15:1^[92] modern day Japan has a more desirable ratio at around 4:1.^[93] Interestingly, India has a large disparity between rural areas (5-6.1:1) and urban areas (38–50:1).^[94]

It has been hypothesized that paleolithic humans had an omega ratio around 0.79 (slightly more omega-3 than omega-6) due to low dietary omega-6 intake^[89][95] and it has been reported that the diet of Grecians prior to 1960 was in the 1-2 range;^[96] interestingly wild animals tend to also have something similar to a 1:1 ratio.^[97][98] due to these low ratios and the evidence that omega-6 consumption has increased in the past 150 years due to technological development^[96][89] there does appear to be a paleolithic argument for a normalized ratio.

The ratio of omega 3 to omega 6 fatty acids in the cellular membrane (either as an EPA:AA ratio, a DHA:AA ratio, or an EPA+DHA:AA ratio) is a quantifiable way of predicting which eicosanoids are produced in response to stress.

Resolvins (resolution-phase interaction products^[99]) are potent signalling molecules involved in inflammation^[100] derived from omega-3 fatty acids, and those that are derived directly from EPA (without requiring metabolism into DHA) are referred to as the E series, while those derived from DHA are called the D series.^[101][102] The E series of resolvins involves one of two pathways, either the lipoxygenase pathway (favoring R isomers) or the aspirin-inducable COX2/P450 pathway (favoring S isomers); regardless of the pathway, the first intermediate after EPA is known as 15-HEPE.^{[101][102][103]} DHA and the D series follows a similar motif (having a LOX pathway and an aspirin inducable pathway^{[104][105][99][106]}) although its first intermediate is 15-H(p)DHA^[102], which has been confirmed to increase in plasma following supplementation of fish oil.^[102]

The aspirin-inducable conversion involves aspirin modifying the COX2 enzyme (as aspirin modifies the actions of COX2^[107][108], which appears to convert DHA^[106] and EPA^[109] to R-resolvins. COX1 is inactive in this regard,^[110] acetominophen and indomethacin are unable to exert the same effects.^[106] Treatment with aspirin alone without supplementing DHA also seems to increase resolvin levels, since some DHA is present in the body.^[106]

Resolvins are molecules named after their ability to 'resolve' inflammation. They are produced by eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). These are the molecules responsible for the synergy between fish oil and aspirin.

Resolvins of the E series include:

• Resolvin E1 (RvE1 or 5S,12R,18Rtrihydroxy-6Z,8E,10E,14Z,16E-eicosapentaenoic acid^[99])
• Resolvin E2 (RvE2 or 15S,18R-dihydroxyeicosapentaenoic acid^[99])

Resolvins of the D series include:

• Resolvin D1 (RvD1 or 7S,8R,17R-trihydroxy-docosa-4Z,9E,11E,13Z,15E,19Z-hexaenoic acid and induced by aspirin treatment)
• Resolvin D2 (RvD2 or 7S,16R,17S-trihydroxy-docosa-4Z,8E,10Z,12E,14E,19Z-hexaenoic acid also induced by aspirin)

E series resolvins are derived from EPA, while D series resolvins are derived from DHA.

RvE1 has been identified to be a ligand of the ChemR23 GPRC receptor,^{[111][112][113]} which is analgesic.^[100] The resolvin of the D series of similar structure (RvD1) also appears to be a ligand.^[100]

Resolvins of the D series have antiinflammatory properties by inhibiting TNF-α induced cytokine expression in microglia in the picomolar range (IC₅₀ of 50pM).^[106] They are deactivated by oxidation, where they are converted to oxidized resolvins (at either the 8 or 17 position).^[99]

Protectins are molecules that are produced from DHA and are structurally docosatrienes, alongside resolvins they mediate many benefits associated with fish oil ingestion.^[99]

• Neuroprotectin D1 (10,17S-docosatriene of the lipoxygenase pathway^[114])
• Marisen 1 (7,14S-dihydroxy-docosa-4Z,8,10,12,16Z,19Z-hexaenoic acid of the lipoxygenase pathway^[99])

Protectins are molecules derived from docosahexaenoic acid (DHA), but they do not belong to the D series of resolvins.

Marisen 1 is named after macrophage mediators in resolving inflammation and is present in macrophages and platelets and formed via the actions of 12-lipoxygenase,^[99] but beyond that is not fairly well researched.

Neuroprotectin D1 (NPD1) is more well understood. Produced by a 15-lipoxygenase like action^[115] after cleavage from phospholipase A₂^[114] where the metabolite 17_S_H(p)DHA (same as the D series of resolvins) is converted into a 16(17)-epoxide and then reconfigured to NPD1.^[116]

NPD1 appears to have potency antiinflammatory properties (by dysregulating IL-1β induced COX2 induction^[116]) in the brain in response to stroke and ischemia^[117] and anti-Alzheimer's actions by preventing the inflammatory response to the protein agggregates seen in the disease state with an IC₅₀ of 50nM (β-amyloid,^[118] may also reduce levels of the pigmentation^[119]). This protection is via a PPARγ dependent mechanism^[118] and as DHA has been noted to reduce amyloid buildup in animals and in vitro^{[120][121][122]} it is thought that NPD1 mediates these effects.

The first metabolite of DHA can be converted into either D series resolvins or protectins. Neuroprotectin D1 (NPD1) is a notable protectin with potent anti-Alzheimer's actions and neuroprotective properties.

While both resolvins and protectins are fatty acid chains derived from EPA or DHA, prostaglandins are characterized by having a pentacyclic ring in their structure (ie. a pentagon in the fatty acid sidechain similar to furan fatty acids).

Prostaglandins include:

• Prostaglandin J2 (15-deoxy-Δ12,14 prostaglandin J2)^[123][124]

Prostaglandins are eicosanoid metabolites with a pentacyclic structure (pentagon) in their side-chain. They are also bioactive metabolites of polyunsaturated fatty acids.

15-deoxy-Δ12,14 prostaglandin J2 (15d-PGJ2) is a derivative of prostaglandin J2 that can activate PPARγ with an EC₅₀ of around 20μM^{[125][123][124]} and inhibit platelet aggregation with an IC₅₀ in the 5-10nM range.^[126][127]

Prostaglandin J2 is related to PPARγ activation and platelet aggregation inhibition.

Prostaglandin F2α has been confirmed to be increased in young men following supplementation of fish oil, and has been noted to increase concentrations of Thromboxane B2 (inactive metabolite of the omega-6 eicosanoid thromboxane A2).^[128]

Fish oil fatty acids interact with the peroxisome proliferator-activated receptor (PPAR) system, which are a class of receptors (PPARα, PPARβ/δ, and PPARγ) that seem to respond to dietary lipids and similarly structured molecules. They are highly involved in the treatment of diabetes and metabolic syndrome (via the drug classes of fibrates and thiazolidinediones), with varying effects on fat mass (PPARα increases beta-oxidation of fatty acids,^[129] while PPARγ promotes fat storage but improves insulin tolerance;^[130] PPARδ appears to be similar to PPARα in this regard^[131]).

PPARs are fairly general receptors, as their binding site is 3-4 times larger than other receptors^[132], which means they have a very general binding capacity.

The PPAR system is a receptor class activated by dietary fatty acids and pharmaceuticals highly involved in lipid and glucose metabolism. It is a druggable target for diabetes and obesity (via PPARγ and PPARα/δ respectively).

15-deoxy-Δ12,14 prostaglandin J2 (15d-PGJ2) is an endogenously produced highly effective PPARγ ligand with an EC₅₀ of 20μM^{[125][123][124]} and its induction (Fuligocandin B from Fuligo candida^[133]) carries the benefits of targeting PPARγ.

18(S)HETE from aradchidonic acid metabolism activates PPARβ and γ with moderate affinity (20μM and greater than 50μM)^[134] and cellular incubation with parent arachidonic acid (without controlling for metabolites) appears to have an EC₅₀ of 1.6μM.^[135]

Metabolism of both omega-3 and omega-6 fatty acids yields molecules that act as PPARγ activators.

Although a few eicosanoids of both omega-3 and omega-6 origin appear to activate PPARα^[134][136] it appears that Leukotriene B4 (from arachidonic acid metabolism) is the most biologically potent and relevant with an EC₅₀ of around 100nM.^[137] The arachidonic acid (AA) metabolite 18(S)HETE also appears to be 10-fold weaker (1μM; R isomer much weaker^[134]) but is better able to recruit the coactivator of SRC-1 (30% the concentration required for 18(S)HETE relative to LKB4^[125]). Parent AA is a weak activator per se (50μM or more^[125]) but past studies that did not control for metabolites noted more efficacy due to its metabolites (1.2μM^[135]).

PPARα is activated by metabolites of the arachidonic acid (omega-6) pathway. Although the significance of this information is unknown, normalizing a ratio of omega 3:6 may reduce overall PPARα stimulation.

Adenosine monophosphate kinase (AMPK) is a nutrient signalling molecule that is antagonistic of mTOR and activated in periods of nutrient deprivation; it is also the molecular target of various supplements like Berberine or the pharamceutical Metformin. The activation of AMPK (seen with both EPA^[138] and DHA^[139]) is partly downstream of PPARγ (noted in general^[140] and with EPA^[141]) and likely mediated by the eicosanoids mentioned in the PPARγ section.

EPA has been found to activate AMPK in adipocytes via insulin-independent means (not requiring activation of PI3K)^[138] and in macrophages.^[139] The α1 subset of AMPK appears to be mostly affected^[139] and both Thr172 phosphorylation (on AMPK) and Ser431 (on LKB1) noted,^[141][142] and DHA has also shown efficacy in isolation in promoting AMPK activity.^{[139][143][144]}

Fish oil activates AMP-activated protein kinase (AMPK). This is partly due to the eicosanoids activating PPRγ, which increases the activity of AMPK.

Via AMPKα1 activation, DHA may increase SIRT1 expression^[139] and suppress inflammation by hindering (via deacetylation^[145]) NF-kB signalling;^[146] this is also a plausible antiinflammatory pathway of fish oil. EPA has been noted to dysregulate inflammatory signalling in adipocytes (usually by suppressing the actions of TNF-α),^[147][148]

Activation of AMPK by EPA may underlie release of adipokines,^[149][138] some inflammation (macrophages can hinder AMPK activation via SIRT1^[146] and EPA preserves this^[139]), improved endothelial function,^[141] hepatoprotection,^[142] insulin sensitivity (related to the liver),^[150] and autophagy (seen with DHA^[143] but a normal consequence of AMPK activation as seen via p53, which induces AMPK^[151][152]).

AMPK activation has been confirmed in mice given 500mg/kg EPA (no reference drug, but just over a doubling in vasculature^[141]) and other studies using fish oil at 1mL/kg^[142] or 15% of the diet.^[150]

AMPK activation has been noted to occur in rodents following oral ingestion of EPA. Since knockout mice (those without AMPK) have failed to show benefits from omega-3 supplementation, it appears that this pathway is very relevant to the benefits of omega-3 fatty acids.

The free fatty acid receptor (FFA), also known as GRP120, is a G-protein coupled receptor (rhodopsin-like^[153]) with a short (361 amino acids; 97.5% homologous between rodents and primates^[154]) and long (377 amino acids; possibly only in humans^[154]) variation expressed mostly in enteroendocrine L cells.^[155] This receptor is named the free fatty acid receptor as it responds to a variety of fatty acids^[156] and some omega-3 fatty acids (ALA and DHA) have been confirmed to be agonists with DHA stimulating activity to 276+/-25% (short variant) and 177+/-13% (long) at a concentration of 100μM.^[157] Elsewhere, mixtures of omega-3 fatty acids have been able to signal via the receptor.^[155]

Activation of this receptor by omega-3 fatty acids is known to secrete some gut hormones (Glucagon-like peptide 1^[158][159] and cholecystokinin^[160]) and is involved in insulin sensitization secondary to antiinflammatory effects^[155] and possible anti-obese effects (as loss of GRP120 is a risk factor for obesity and GRP120 knockout mice are obese^[161]).

It has not yet been confirmed whether DHA and EPA are direct agonists of the receptor or whether they work via eicosanoids, although the ability of alpha-linolenic acid to activate the GRP120 suggests the former.

Activation of GRP120 (free fatty acid receptor) through fish oil supplementation and/or derivative eicosanoids has been confirmed in animals. This may underlie some of the biooactivity of fish oil supplementation.

Other fatty acid receptors exist (all with a GPR designation) including FFAR1 (GPR40), FFAR2 (GPR43), FFAR3 (GPR41), and GPR84.^[162] FFAR1 responds mostly to medium chain fatty acids (palmitic acid and linoleic acid^[163][164]) and both FFAR2 and FFAR3 respond to shorter chain fatty acids (acetate and butyrate^[165]) with GPR84 being the sensor for medium chain fatty acids (lauric acid^[166]). Although these are free fatty acid receptors, they are not seen as molecular targets of EPA nor DHA due to the long length of fish oil fatty acids.

Other free fatty acid receptors are not likely molecular targets of fish oil supplementation, as they respond to shorter chain fatty acids.

EPA and DHA tend to be digested and taken up as normal dietary fats, by getting packaged into micelles in the intestines and being subsequently dropped off at fat cells and muscle cells by chylomicrons (a transport molecule) before the chylomicron remnant goes to the liver.

If the fish oils are microencapsulated (which occurs in some functional foods to avoid a fishy taste) they tend to be absorbed in the upper small intestines^[167] although a large bit is incorporated into the intestinal wall as well.^[168]

Bodily loading of fish oil seems to be maximal after approximately 3 weeks of supplementation with no significant difference between doses tested (210-630mg EPA and 150-450mg DHA).^[169]

It has been hypothesized that dietary DHA is required for optimal neural functioning due to some evidence suggesting a diet low in DHA reduces brain phospholipid content by 32% (high dietary ALA) or 53% (low ALA), relative to a diet high in DHA,^[56] although losses of DHA from the brain are somewhat difficult over the short term in non-neonates and seem to be regulated^[170][171] and neonates seem more sensitive, with DHA deprivation from young mice resulting in DHA losses in the brain within 2 weeks.^[172] Losses have been reported in adult primates on dietary restriction of DHA (and are likely to apply to huamns), but this requires a time period of 18 months^[173] or 5 years (70% reduction)^[174] with the inclusion of alcohol to deplete DHA stores.^[173][174]

Short term dietary restrictions, in adults, are unlikely to modify cerebral concentrations of docosahexaenoic acid (DHA). However, prolonged absence of DHA in the diet seems to be able to progressively reduce DHA concentrations in the brain.

It has been hypothesized that aggression is a symptom of DHA deficiency,^[175] in which case supplemental DHA would alleviate this deficiency.

In otherwise healthy persons, DHA appears to prevent excessive aggression in times of stress.^[176][177] In contrast, another study using a similar model to study agression found that DHA failed to cause a change in aggression levels under non-stressful conditions.^[178] Of note, the authors of this particular double-blinded study found that in contrast to the placebo group, which showed decreased aggression levels over the 3 month supplementation period, subjects taking DHA exhibited the same agression levels before/ after 3 months supplementation. The mechanism for the apparent aggression-stabilizing effects of DHA are unknown, but warrant futher studies since low levels of omega-3 fats in general have been linked to increased incidence of depression.^[179] It is possible that the anti-depressive effects of omega 3 fats in general may also manifest themselves through stabilized agression levels.

The effects of DHA on depression and stress are observed in the range of 1.5g daily, and do not appear to occur with lesser doses of 150 mg DHA daily.^[177] A reduction in aggression was also seen in a group of young men without a previous history of agression at a dose of 672mg/day DHA for 3 months.^[180]

An exception may be schoolchildren, who showed benefit at 3.6g per week in one study.^[181] Another study in children aged 8-16 years old given a drink containing 1g of omega-3 fatty acids a day (consisting of 300 mg of DHA, 200 mg of EPA, 400 mg of alpha-linolenic acid, and 100 mg of DPA) for 6 months also showed a reduction in several measures of aggression compared to a placebo drink of similar flavor and consistency.^[182]

DHA supplementation has been found to prevent excessive agression under stressful conditions. It seems to have a similar aggression-stabilizing effect under non-stressful conditions (i.e. maintaining aggression levels in contrast to a decrease in agression with a placebo). These results may be confounded with influences on noradrenaline, however. See the 'Stress' section for more information.

DHA is investigated for its role in memory formation as higher serum DHA concentrations are correlated with greater verbal fluency skills in older humans^[183] and a deficiency of DHA is known to damage rat memory processing.

For animal models, administration of DHA (300mg/kg is the standard dose) appears to promote cognition with improvements noted in reference memory (without affecting working memory) in otherwise healthy rats.^[184]

There is animal evidence to support docosahexaenoic acid (DHA) improving memory in otherwise healthy rats.

One study conducted in otherwise healthy young (18-25) persons noted that, after 6 months of supplementation with 2g Lovaza (750mg DHA; 930mg EPA) that working memory was enhanced as assessed by a verbal n-back test (with improvements in 3-back but not 1 or 2-back); this was deemed to be independent of dopamine metabolism but instead correlated with the EPA and DHA content or Red Blood Cells (RBCs; up 75% from 2.9+/-1.0% for DHA, and up 350% from 0.4+/-0.1 for EPA) when looking at the RBCs of persons low in DHA.^[185] Elsewhere, youth who have low dietary intake of fish have experienced improved memory retention and reaction time with supplemental DHA at 1,160mg over 6 months.^[186] Some isolated studies not reporting dietary intake of fish oil also report benefits to memory formation (as well as attention and reaction time).^[187] Interventions in older adults tend to note benefit in persons with^[188] and without^[189] apparent cognitive decline.

Some studies fail to find a significant benefit of fish oil supplementation on cognition in healthy adults (400-1,800mg EPA+DHA,^[190] 1,400mg EPA and 800mg DHA,^[191] or 1g EPA+DHA over 12 weeks^[192]) with one of these studies^[191] having an exclusion criteria of 'no more than once weekly' for fish intake but otherwise diet was not reported.

There is human evidence to support an increase in memory (working and episodic) associated with DHA supplementation. Although it cannot be ruled out, it is plausible that this will only occur in people with low dietary DHA intake.

A meta-analysis on fish oil supplementation and depression (inclusive of disease states such as schizophrenia or bipolar disorder)^[193] able to assess 28 studies with a dosage range of 136-6,200mg EPA and 88-3,400mg DHA (Postpartum or perinatal depression,^{[194][195][196][197][198]} major depression disorder or depression without other cognitive diseases,^{[199][200][201][202][203][204]} depression associated with fatigue,^[205][206] bipolar disorder,^{[207][208][209][210]} schizophrenia,^[211] Parkinson's disease,^[212] self-harm,^[213] personality disorders,^[214] or no significant disease state or just mild depression^{[215][216][217][218]}) noted that a higher EPA:DHA ratio was predictive of anti-depressive effects and that the three studies using pure DHA^{[196][200][197]} outright failed to exert antidepressive effects. Oddly, 1g of EPA supplementation (as ethyl ester) appears to be more effective than 2-4g in one trial^[201] or at least 2g fails to outperform 1g^[209] and the meta-analysis came to the same conclusion where supplemental EPA dose was inversely related to efficacy (with higher doses being less effective).^[193]

Beyond identifying the efficacy of lower doses, this meta-analysis found evidence for publicity bias (was unsure if this was due to heterogeneity in the studies assessed as it was very inclusive or other influences) and suggested that persons with higher baseline depression experienced more benefit, with the group with low depressive symptoms having a reduction of −0.074pts on mean depression (95% CI of −0.317 to 0.169) that was not statistically significance but the higher depressive group having a reduction of −0.605 (95% CI −0.871 to −0.339).^[193] The robustness of the studies (assessed by Jadad) was similarly hetereogeneous with some poor quality studies included.^[193]

Supplementation of 1g eicosapentaenoic acid (EPA) appears to benefit depression most. Docosahexaenoic acid (DHA) is not associated with any significant antidepressive effects. The antidepressant effects of EPA are only significant in severely depressed people.

It is plausible that fish oil is an augmentor of antidepressants, as the above meta-analysis did note that most trials conducted in major depression or disease states were also mediated with standard antidepressant drugs (such as lithium).^[193] At least one study has noted that fish oil (in isolation at 1,000mg ethyl-EPA) is comparable to the SSRI drug fluoxetine (20mg) but also supports the idea that combination therapy is best, as response rates were in the range of 50-56% (monotherpaties) and rose to 81% with both.^[219] This synergism has been noted in animal studies with traditional anti-depressants (fluoxetine and mirtazapine)^[220] and with nutraceutical anti-depressants such as uridine.^[221]

There is evidence to support fish oil acting as an antidepressant by itself, but there is also evidence to support significant effects to come from combination therapy. Combination therapy means taking fish oil alongside a proven antidepressant.

In otherwise healthy elderly adults with minor depressive symptoms, fish oil supplementation does not appear to further increase well-being and reduce depression at 1800mg EPA+DHA,^[215] and has failed to reduce depressive symptoms (or increase well being) in adults who do not report depression.^[222][223]

There is insufficient evidence to support a mild antidepressive effect in people who report mild but not debilitating depression. Fish oil supplementation appears to be ineffective for mild depression in the studies that investigate it.

Fish oil supplementation is hypothesized to aid bipolar symptoms as reference drugs (lithium carbonate and valproate) reduce neuronal signal transductance systems^[224][225] which appears to have been noted with omega-3 fatty acids in regards to arachidonic and phospholipid signalling.^[210] Bipolar disorder does appear to also be associated with alterations in membrane lipids, but this is more related to GLA and the omega-6 fatty acids than it is to fish oil.^[226]

However, despite a preliminary study noting that fish oil supplementation may aid symptoms of bipolar disorder at 9.6g daily (6,200mg EPA and 3,400mg DHA)^[210] and (albeit cautious) promise for depressive symptoms mentioned by one meta-analysis on the topic,^[227] a followup study using smaller doses (2g twice a day, totalling 1,680mg EPA and 1,120mg DHA) with or without 1g cytidine delivered via CDP-choline (twice daily; forms uridine in the body^[228]) failed to replicate the results over a similar time period of 16 weeks, and actually reported a nonsignificant trend to worsen symptoms.^[229]

There is mixed evidence regarding fish oil supplementation and bipolar disorder. The limited studies on the topic are too different to compare, as they use very different dosages. Although a beneficial effect cannot be ruled out, a worsening effect is also possible.

A Cochrane meta-analysis on the effect of omega-3 PUFAs in general on people with major depressive disorder found a small benefit (equivalent to a 2.2 drop on the Hamilton Depression Rating Scale), although there was a lot of uncertainty in the estimate of effect size; in addition, the overall quality of trials was rated as low and the risk of bias high.^[230]

Omega-3 polyunsaturated fatty acids may have a small effect on depression in people with major depressive disorder, but the quality of studies used in estimating this effect was low overall.

Fish oil supplementation in rats (at 5% food intake) was found to normalize the stress response after being subject to footshock, a research test used to mimic long-term environmental stress.^[231] This has been replicated in other models on animals and in humans with DHA supplementation at high doses (1.5-1.8g DHA) daily in which adrenaline response to stress is attenuated.^[175][232] In longer term models to assess cortisol (long term stress hormone), students taking 20 exams had a similar decrease in noradrenaline (-31%) at 1.5g DHA daily, although no changes in cortisol were noted.^[233]

In regards to stress, both EPA and DHA seem to have implications. EPA from modulating some immune functions associated with stress^[234] and DHA is tied in with aggressive increases during stressful times.^[235]

Interestingly, a low dose of 762mg EPA+DHA daily can reduce noradrenaline levels even in healthy non-stressed persons.^[236]

A deficiency of omega-3 fatty acids has been noted to reduce glucose metabolism in the brain in rats^[237] thought to be related to reduced GLUT1 transporters^[238] which is amendable with supplemental omega-3 fatty acids (in vitro).^[239][240]

In primates, DHA supplementation (150mg/kg) has been noted to improve neuronal glucose uptake^[241] although this does not appear to occur in humans per se (independent of age);^[242] there does appear to be an inverse correlation between cerebral glucose utilization and serum triglycerides^[242] suggesting that treatment of dyslipidemia may be beneficial.

It is unlikely that fish oil supplementation increases cerebral glucose metabolism in healthy individuals. People with metabolic syndrome or high triglyceride levels might experience a benefit secondary to reducing triglycerides, but further study is needed to confirm this effect.

Isolated EPA has been noted to modulate cerebral blood flow in spontaneously hypertensive rats at 100mg/kg for 8 weeks.^[243]

As assessed by haemodynamics in functional near IR spectroscopy (NIRS) where total blood hemoglobin is closely correlated to blood volume and oxygenation rates can be measured^[244] it has been noted that in otherwise healthy young adults not consuming more than one fatty fish product per week who then recieved 450mg DHA and 90mg EPA for 12 weeks was able to increase cerebral oxygenation during cognitive testing (without affecting deoxygenated hemoglobin as total hemoglobin increased);^[245] this study is duplicated in Medline.^[246]

Fish oil appears to promote blood flow to the brain in otherwise healthy adults, who have low dietary fish intake.

Most epidemiological evidence,^{[247][248][249][250]} but not all,^[251] suggest a reduced risk of stroke associated with high dietary fish consumption particularly in elderly persons^[250] and following standard dose-response;^[247] this is thought to be related to the omega-3 fatty acids^[252] and particularly DHA.^[253][56]

Intravenous administration of fish oil triglycerides to mice given a stroke (hypoxia/reperfusion; 100-375mg/kg fish oil) noted that a 43-47% reduced infarct size resulted with both pre-treatment and up to 2 hours post-stroke,^[254] and 100mg/kg of EPA (as ethyl ester) for 8 weeks in spontaneously hypertensive stroke-prone rats has noted a modulation of cerebral blood flow.^[243]

It is possible that fish oil supplementation has both a therapeutic and preventative role for stroke victims.

The amount of DHA complexed with phosphatidylcholine in plasma appears to be negatively correlated with the risk of dementia in humans^[255] and high dietary intakes of fish have been noted to be protective against the rate of developing dementia and related cognitive decline disease states in older and middle aged persons.^{[256][257][258]}

When assessing older adults, higher cerebral levels of EPA (surprisingly, not DHA) are related to less atrophy of select brain regions (hippocampus, right amygdala) when followed over 4 years^[259] and lower erythrocytic EPA and DHA are associated with reduced brain mass in older adults in cohort studies.^[260] These results suggest a protective role of EPA, and 2,000mg of supplemental DHA for 18 months in older adults has failed to reduce the rate of atrophy.^[261]

Higher fish intake is associated with a lower risk of developing dementia, and reduced docosahexaenoic acid (DHA) levels in serum are associated with a higher risk of dementia. Eicosapentaenoic acid (EPA) appears to be associated with preserving brain mass over time.

A higher intake of omega-6 relative to omega-3 can be associated with increased activity of COX enzymes as they are competing substrates^[262] although the subsequent production of PGE(2) (known to increase amyloid secretion^[263] via inducing γ-seceratase^[264]) which is known to be reduced with fish oil^[265] appears unrelated to DHA-induced suppression of amyloid secretion.^[266]

As mentioned in the protectin section, it is plausible that the mechanism of protection against amyloid buildup is mediated via Neuroprotectin D1 as it potently (EC₅₀ of 50nM) suppresses amyloid formation^[118] and promotes its removal^[119] via PPARγ-dependent mechanisms.^[118]

Prostaglandin E2 (PGE2) induced amyloid buildup does not appear to be related to neuroprotection, in regard to Alzheimer's, but it is likely related to neuroprotectin D1.

In animals, provision of DHA at 300mg/kg (rat dose) for 12 weeks appears to promote cognition, memory, and reduce the rate of cognitive decline in a model of Alzheimer's disease characterized by β-amyloid pigmentation.^{[267][268][269]} Furthermore, the dendritic/synaptic decay seen in Alzheimer's disease^[270][271] appears to be reduced with provision of DHA^[272] which is thought to be the mechanisms of neuroprotection and further thought to be synergistic with Uridine supplementation.^[273][274]

Animal evidence appears to support the role of DHA in high doses (300mg/kg in rats, which is an estimated human dose of 48mg/kg or 3.2g DHA for a 150lb person).

In people with Alzheimer's, 2,000mg DHA (no EPA supplementation) from algae over 18 months has failed to benefit cognitive decline^[261] and another study using 1,700mg DHA and 800mg EPA in persons with Alzheimer's for 6 months has failed to significantly attenuate the rate of cognitive decline as assessed by MMSE^[275] and failed to alter neuropsychiatric symptoms (both citations are the same trial).^[276]

In human studies on age-related cognitive decline that are not Alzheimer's patients, one study found that 900mg DHA may reduce the rate of cognitive decline^[277] but another study using 500mg DHA and 200mg EPA over 24 months found no benefit.^[278] The latter study, however, failed to find any rate of cognitive decline in either group and may not accurately represent fish oil. A third study in people with age-related macular degeneration (AMD) found that 350 mg DHA and 600 mg EPA daily for 4-6 years failed to prevent cognitive decline as measured by a phone-delivered composite of 8 tests for cognitive decline.^[279] However, the mechanism of AMD and cognitive decline may share some common pathways and risk factors,^[280] and so this population may not be representative of the general population.

Studies using practical and high levels of fish oil supplementation in people with Alzheimer's disease and age-related macular degeneration have failed to find practical benefits, but one study in older adults without AD or AMD found that DHA may be able to attenuate the rate of cognitive decline. The reason for this disparity, if it exists, is not known at this time.

Resolvin E1 (RvE1) signals for analgesia via the Chem23 receptor, and is active independent of fish oil supplementation (injections of 0.3-20ng RvE1; more potent than 10mcg of the COX2 inhibitor NS-398);^[100] its metabolic inactivation by oxidation (oxo-RvE1^[281]) is metabolically inactive.^[100] It appears that activation of the Chem23 receptor via RvE1 inhibits ERK^[282] and more specifically TNF-α induced activation of ERK^[100] which would signal to TRPV1^[283] to mediate the perception of pain. Overall, it is hypotheiszed that RvE1 siganllings for pain releif primarily by dysregulating TNF-α and its normal induction of glutamate release form neurons, which activates ERK to induce pain via TRPV1.^[100]

Although RvE1 was researched above, RvD1 (from DHA) appears to also activate the Chem23 receptor^[100] and the parent fatty acids of EPA and DHA were also tested but were approximately 10,000-fold weaker.^[100]

Resolvin E1 (from EPA supplementation) and Resolvin D2 (from DHA) appear to act via a novel receptor to prevent the pro-inflammatory cytokine known as TNF-α from inducing pain. On a molecular level (how much of a molecule is required to induce effects), RvE1 and RvE2 are remarkably potent.

For human studies, supplemental fish oil has been found to reduce pain in persons with inflammatory joint pain as assessed by patient reports and NSAID consumption (rescue medication, which was reduced) but not by physician reports.^[284]

CETP is a transport protein that transports cholesterol from HDL to either vLDL or LDL (apolipoprotein B containing lipoproteins^[285]) in exchange for triglycerides,^[285][286] and reducing the activity of CETP increases HDL-C^[285] possibly in hyperlipidemics only^[287] while CETP activity itself is positively correlated with LDL-C.^[286][288] Due to this and associations between CETP and high cholesterol^[289][290] increasing the activity of CETP is seen as pro-artherogenic (increasing LDL-C while reducing HDL-C, opposite of the desired therapeutic intervention) and its inhibition seen as desirable.

The inhibition of CETP is further desired as a triglyceride reducing treatment^[291] as the exchange for cholesterol from vLDL/LDL towards HDL-C is a 1:1 exchange with triglycerides;^[292][285] a depletion of triglycerides from LDL and vLDL cholesterol (rather than placebo per se) has been confirmed in humans following supplementation of fish oil.^{[293][294][295]}

Fish oil has been noted to increase CETP activity in animals^[296] due to DHA and not EPA,^[297] which is thought to underlie the selective induction of LDL-C seen with DHA and not EPA although it does not explain the increase in HDL-C seen with DHA.^[298]

It is plausible that one of the mechanisms underlying the benefits of fish oil is cholesteryl ester transfer protein (CETP) inhibition, which is able to reduce triglycerides and increase high-density lipoprotein (HDL-C). This pathway alone, however, does not explain the effects on low-density lipoprotein (LDL-C).

In regards to fasting triglycerides (TGs; risk factor for cardiovascular disease when elevated), fish oil appears to be both a potent and reliable triglyceride reducing agent for persons with hyperlipidemic (high blood TGs).

The meta-analyses that have been published indicate that fish oil is effective for general dyslipidemia (0.34mmol/L reduction^[299]) HIV related dyslipidemia (7 trials of 372 persons reducing TGs 1.12 mmol/L^[300]), dyslipidemia associated with renal failure (10 trials of 337 persons reducing TGs by 0.78mmol/L^[301]), and diabetic dyslipidemia (24 trials of 1530 persons reducing TGs by 0.17mmol/L or 7%^[302]). The range of reduction has been cited to be as high as 25-30% with 4g of EPA ethyl ester daily,^[303] but more recent evidence suggests that the reduction is a tad more modest (estimated 15-20% reduction, as the magnitude does depend on baseline triglycerides).

Both EPA and DHA are able to reduce triglycerides^[298] and these benefits extend to other sources of DHA including algae oil (meta-analysis^[52]) and krill oil with comparable potency. When comparing the effects of EPA against DHA when they are used in isolation (separate trials), EPA appears to be a tad more effective (meta-analysis^[298] concluding a 25.1mg/dL reduction with DHA in isolation^{[304][305][306][307][53][308][309][310][311]} as average but a 45.8mg/dL reduction with EPA in isolation^{[312][313][314][315]} are similar doses and baseline TGs). However, studies that assess direct comparisons between DHA and EPA^{[316][317][318][319][320]} note superiority with isolated DHA (when doses are similar in weight; ie. 4g versus 4g^[298]). It is plausible that this difference is due to EPA traditionally being dosed at a higher quantity than DHA, with fish oils typically following dose-dependence.^[299]

Fish oil reliably and potently reduces triglyceride levels after several weeks of supplementation. The reduction in triglycerides is based on dosage, and it is more significant in people with high baseline triglycerides, quantified in the 15-30% range. This is pharmaceutical grade potency, which is why fish oil was patented as Lovaza (ethyl ester).

Supplemental fish oil does not appear to influence postprandial triglycerides when acutely supplemented (one dose at mealtime)^[321] but has been noted to reduce postprandial triglycerides in hyperlipidemics following prolonged supplementation (27% with 4g ethyl EPA^[322] or 19% with 4g fish oil^[323]).

The reduction in postprandial triglycerides occurs after repeated supplementation and may be more reflective of the triglyceride reducing effect per se. Single dosages of fish oil fail to benefit postprandial triglycerides.

It is thought that DHA mediates the cholesterol increasing effects of fish oil supplementation^[298] as supplementation with DHA increases HDL-C by 4.49mg/dL (95% CI of 3.50-5.48mg/dL) and LDL-C by 7.23mg/dL (95% CI of 3.98-10.5) whereas EPA nonsignificantly increases HDL-C by 0.20 mg/dL (95% CI of -0.82 to 0.41) and insignificantly influences LDL-C by 1.85mg/dL (95% CI of −3.01 to 6.71) with the sporadic instances of EPA increasing HDL-C and LDL-C^{[324][312][316][320][318][319]} possibly being explained by conversion of EPA into DHA in the body.^[298]

Fish oil supplementation, secondary to the DHA component, can increase both HDL-C and LDL-C lipoproteins in the body. It appears to increase LDL-C a bit more than it does HDL-C, and thus large doses of fish oil may not be advisable for people with high LDL-C cholesterol, unless the reduction in triglycerides is seen ias more important, or if a statin drug or similar cholesterol reducing agent like berberine is taken alongside the fish oil.

It is possible for fish oil to reduce LDL cholesterol, but infrequent; it requires the user to not have disturbances in LDL cholesterol in the first place (normocholesterolemic)^[325][326] or otherwise augments the LDL-C reducing effects of statin drugs (Nutrient-Nutrient Interactions section).

There are possible LDL-C reducing effects when fish oil is combined with statin drugs, where fish oil and statins appear synergistic in reducing LDL-C, which appears to negate the adverse LDL increase of fish oil that is possible in hyperlipidemics.

In secondary prevention of cardiovascular disease, a membrane ratio of 4:1 omega6:3 (average with a standard western diet is 15-16.7:1^[89]) or lower is associated with a 70% decrease in total mortality^[89] as assessed by a single blind prospective study.^[327]

Normalizing the omega 3:6 ratio appears to be associated with cardioprotection.

The mechanistic basis for the improved endothelium-triggered relaxation with n - 3 PUFAs may include the suppression of thromboxane A2 or cyclic endoperoxides, a reduced production of cytokines, the augmented endothelial synthesis of nitric oxide, an improvement of vascular smooth muscle cell sensitivity to nitric oxide, and a reduced expression of endothelial adhesion molecules^[328].

Fish oil supplementation (following studies using ethyl ester EPA) note that fish oil consumption can somewhat reliably increase fasting glucose in the range of 2–6mg/dL, this applying to both diabetics and nondiabetics and conclusions being drawn from multiple meta-analyses(most positive^{[329][330][331]} and some counter evidence^[332]) and a trend to increase glucose is still seen in hyperlipidemic patients.^[322] This increased glucose concentration is not usually met with increases in biomarkers of diabetes seen as adverse (HbA1c^{[330][319][331]} and fructosamine^[305][333]) suggesting that it may not further pathology of diabetes.

It should be noted that the increases in glucose are not observed at low doses of fish oil supplementation (6g of fish oil conferring 1,080mg EPA and 720mg DHA^[334]) and appear to follow dose-dependence, with one study failing to find an influence of 4g fish oil finding an adverse increase with 7.5g (2,600mg EPA and 1.4mg DHA).^[335]

There appears to be a small increase in fasting blood glucose seen with fish oil supplementation that is independent of disease state (affects diabetics, hyperlipidemics, and healthy controls), but is small (2-6mg/dL) and not associated with an increase in diabetic parameters HbA1c and fructosamine. The increase in glucose appears to occur at higher fish oil doses. Diabetics wanting to use fish oil should begin by using the lowest effective dose. .

In healthy persons, fish oil may not increase insulin sensitivity with a high fat (37%) diet when weight loss or gain prevented, at a dose of 3.6 EPA+DHA daily.^[336] This study did note nonsignificant trends of improved sensitivity in individuals who had higher 6:3 ratios at baseline. Other studies note similar results in healthy persons^[337] but did not record phospholipid ratios.

In otherwise healthy males, even pairing exercise with fish oil did not yeild any changes to insulin sensitivity that were attributable to the fish oil.^[338] Fish oil seems to be additive to but not synergistic with exercise.

Other studies suggest improvement in insulin sensitivity in populations who typically have worse 3:6 ratios, such as the elderly^[339] the metabolically unhealthy,^[322][340] and the obese.^[341] It should be noted that this body of evidence is not bullet-proof, and notable studies do detect no changes in insulin sensitivity even in the above populations.^{[342][343][344][345]} Furthermore, information from our rubric show that a large amount of systematic meta-analysis' show no significant ability for fish oil to change fasting glucose or fasting insulin in type II diabetics.^{[346][332][329][330]}

The above mechanism of increasing insulin sensitivity may be by preserving cell fluidity and rheology, or bringing an aberrant omega3:6 ratio back to a normal range (or preventing aberration in the first place) with no therapeutic benefit beyond that. This is supported by Haugaard et al. who demonstrate a correlation between membrane PUFA content (independent of being omega 3 or 6) but additionally the 3:6 ratio, and insulin sensitivity.^[347] Finally; in those who develop insulin resistance from fructose overfeeding, fish oil appears to be ineffective at alleviating the insulin resistance (although it still reduces triglycerides).^[348] This lends credence to the notion that fish oil's insulin sensitizing effects are at the level of the cell, as fructose causes insulin resistance at the level of the liver and pancreas.^[349]

Another possible mechanism is merely negating the negative effects on some saturated fatty acids on insulin sensitivity. Palmitic acid is known to induce muscular insulin resistance,^[350] and polyunsaturated fats (either omega-3 or 6) can reduce the negative effects of palmitic acid.^[351]

There is some evidence that suggests fish oil might increase insulin sensitivity, but these studies are isolated and are dependent on pre-existing conditions that hinder insulin sensitivity. Fish oil supplementation does not increase insulin sensitivity or reduce fasting_ glucose (a long-term marker of glucose metabolism). Fish oil can, however, reduce blood glucose acutely.

There appears to be a reduced risk of diabetes associated with a somewhat normalized omega3:6 ratio.^{[352][353][354]}

In mice, high doses (15% of diet) of fish oil have been noted to induce CPT-1, Nrf1, and PPAR-α expression in adipose tissue;^[355] the latter of which tends to be antiobese when activated^[356] and the former the enzyme that mediates carnitine transport into the mitochondria and serves as the rate-limiting step of fat oxidation.^[357]

PGC1α has also been noted to be induced in adipose of mice^[355] but although activation of PGC1α may increase energy expenditure via UCP2 expression it is possible this is just downstream of PPARα activation as there are similarities in this regard between fish oil and fibrates (pharmaceutical PPARα activators).^[358][359]

Beyond that, DHA is thought to be the active ingredient as it is better correlated with weight loss in humans.^[360]

Fish oil can technically activate the PPARα receptor to induce mitochondrial biogenesis and increase the metabolic rate. Although this effect is confirmed in rodent models, it requires a high oral dose. Rodent models have been noted to be genetically different when it pertains to PPAR metabolism (see: Conjugated linoleic acid).

In studies that assess metabolic rate, it is found to not be significantly influenced (despite increased fat oxidation) in otherwise lean men.^[361]

Fish oil does not reliably increase metabolic rate.

There may be antiinflammatoy effects at the level of the adipocyte as evidence by less cytokine secretion under the influence of fish oil fatty acids.^[362]

Fish oil may exert a localized anti-inflammatory effect, which could indirectly aid fat metabolism in people characterized by excessive inflammation (ie. those with metabolic syndrome).

Through these agonisms, fish oil can increase adiponectin secretion from fat cells.^[363][364] Surprisingly, though, it seems to take up to 6 weeks for this effect to be physiologically relevant in humans at a dose of 2g fish oil daily.^[365] Higher levels of circulating adiponectin are seen with diets higher in fish oil omega-3s.^[366]

EPA is more potent at increasing adiponectin relative to DHA, and this increase in mediated mostly through PPARy activation.^[367][363] Fish oils can also positively regulate leptin in the same manner.^[368]

It has been noted that the anti-obese effects of fish oil on body weight gain in high-fat fed rats is reversed in diabetic rats^[369] and that inclusion of large amounts of sucrose into the diet reduced the anti-inflammatory and weight loss effects of fish oil.^[370]

In rats that are fed a high fat obesogenic diet, fish oil ingestion (without exceeding caloric intake) appears to attenuate the rate of fat gain over time when consisting of 20-40% of overall calories^{[371][372][373][374][375]} with some studies using more reasonable doses (1-12%) noted the same result but to a smaller degree, with some indication that DHA was more relevant than EPA.^[376] It is not sure if this effect is similar in diabetics, as one study has found an augmentation of weight gain (without measuring fat mass)^[369] but appears to still persist in already obese animals who are subject to further weight gain.^[377]

In general, high doses of fish oil in rats (to a level impractical for human consumption) are able to reliably reduce weight gain if the rats are concurrently fed an obesogenic diet. This appears to occur to a smaller extent with smaller doses, which may be more relevant to human ingestion.

Fish oil is known to suppress the adipogenic receptor PPARy^[378] and its mRNA levels.^[379]

Fish oil has been noted to increase the expression of the Carnitine Palmitoyltransferase-1 (CPT-1) enzyme in muscle cells^[380] (possible secondary to PPAR activation^[381]) and increased both peroxisomal acyl-CoA oxidase and UCP3 expression at very high dietary intake in rats (40% of dietary intake).^[371]

Skeletal muscle interactions may mediate fat loss through caloric expenditure.

When looking at survey research, there appears to be an inverse correlation with dietary fish oil intake and obesity rates (suggesting a protective effect)^[382] or no significant relation at all.^[383]

When looking at interventions, one study in otherwise healthy lean men has noted that replacing 6g of fatty acids with fish oil for 3 weeks resulted in a fat loss of -0.88+/-0.16 (relative to placebo with -0.3 +/- 0.34kg) which was associated with an increase in fat oxidation but not metabolic rate.^[361]

Fish oil may increase fat loss in otherwise lean individuals, but further research is needed to confirm this effect.

In obese sedentary persons, 6g of fish oil has failed to outperform 6g of control oil (sunflower) either with or without aerobic exercise although there was evidence of slight synergism with fish oil and exercise.^[384] The lack of benefit of fish oil to reduce fat mass in obese persons has been noted elsewhere with insulin resistant women given 2.9g DHA and 1.3g EPA for 24 weeks.^[343]

Elsewhere, it has been noted (hyperlipidemics) that the decrease in body fat seen with fish oil supplementation is only significant when paired with exercise with no inherent loss being noted at similar doses^[385]

In studies that enforce a caloric restriction, fish oil (2,800mg omega-3) ingestion in obese women is associated with greater losses of body fat and weight (24%) than is control oil (saline) which is thought to be related to higher ketone levels detected in the blood.^[360] Some degree of fat loss has also been noted with fish oil when it is paired with the Zone diet (Zone diet alone in this study also effective) while it alone was ineffective.^[218]

Fish oil may play a role in reducing fat mass in obese people, but it does not have an inherent fat loss effect. This effect is dependent on other weight loss habits, like exercise or caloric restriction. The effect itself is not very strong.

Fish oil supplementation has been noted to augment muscle protein synthesis that is induced by hyperinsulinemia-hyperaminoacidemia in both young^[386] and old adults.^[387]

Although the mechanism by which this occurs is not clear, fish oil has been shown to affect SMAD signaling, a downstream regulator of myostatin. In certain experimental models, fish oil has been shown to suppress SMAD2 (a SMAD that promotes transcriptional activation) and induce SMAD7, an inhibitory SMAD which suppresses SMAD2 and SMAD3 signaling (noted in renal cells^[388] and cardiac cells^[389]). Since fish oil has been noted in one study to block SMAD2/3 nuclear translocation in cardiac fibroblasts,^[389] it is theoretically possible that it could affect myostatin activity via a similar mechanism. Importantly, this has not been experimentally validated and more research is needed to examine this possible connection.

Fish oil supplementation has been shown to augment hyperinsulinemia-hyperaminoacidemia induced muscle protein synthesis in young and old adults. It is not clear whether fish oil may also augment muscle protein synthesis under more physiologic conditions, such as after protein ingestion through normal feeding.

Alhough fish oil has been shown to augment muscle protein synthesis (MPS) induced by hyperinsulinemia-hyperaminoacidemia, another study suggests that it has no effect on MPS in response to protein ingestion or protein ingestion in combination with resistance exercise.^[390] In this randomized, controlled trial, 20 healthy male subjects 24 years of age received either 5 g/day fish oil or a coconut oil control for 8 weeks. After the supplementation period, subjects performed resistance exercise followed by the ingestion of 30g whey protein. Although MPS increased after protein ingestion in the presence or absence of resistance exercise, fish oil supplementation failed to have an effect. Fish oil also suppressed anabolic signaling activity relative to the coconut oil controls as measured by a decrease in p70S6K1 activity, an important signaling molecule for protein synthesis.^[390]

Fish oil was shown in one study to reduce protein and resistance exercise-induced anabolic signaling without affecting muscle protein synthesis. This is in contrast to other studies noting that fish oil augments muscle protein synthesis under experimentally-induced hyperinsulinemic-hyperaminoacidemic conditions.

A very high dose (1g/kg bodyweight; about 28% EPA+DHA content) fish oil in rats shows increased glycogen resynthesis rates and increased glucose oxidation independent of insulin,^[391] and 14% increase lactate concentration that was dependent of insulin stimulation. The increased glucose oxidation and uptake may be downstream effects of increasing transcription of AMPK.^[392] Activation of AMPK has been noted in other tissues by DHA, such as the intestines^[144] and can do so vicariously though adiponectin.^[366] This increase in glucose oxidation (possibly by AMPK) was also noted at intakes of 1.8g omega3 (1.1 EPA, 0.7 DHA), which quantified the same glucose oxidation rates despite a 17% lesser AUC for insulin.^[393]

Fish oil seems to upregulate mRNA for uncoupling proteins (heat production) in mouse muscle UCP3, brown adipose tissue UCP2, and liver UCP2^[359][394] although a drop is seen in white adipose tissue UCP2. Muscle upregulation is seen in bovine as well.^[395] Although increased UCP expression is correlated with decreased energy efficiency, a dose of 7.2g fish oil (1.1g EPA, 0.7g DHA) does not significantly impair energy efficiency in otherwise healthy males.^[338] This study did note a trend of increasing fat utilization over carbohydrates, however; uncoupling proteins were not measured. This may be a dose issue, as metabolic efficiency is greatly reduced when fish oil is superloaded at 40% of energy intake in rats.^[371]

Fish oil (technically, EPA) incubation in muscle cells is associated with a greater ability for the muscle cell to switch from glucose to fat as primary substrate for oxidation, a phenomena known as 'bioenergetic flexibility' of 'metabolic switching'.^[396]

In mice subject to immobilization, fish oil supplementation has been implicated in decreasing the rate of muscle degeneration.^[397] However, it also hinders recovery after the fact for a few days via the same pathway.^[398]

Some studies in post-surgery situations note increased retention of lean body mass when EPA is added to enteral nutrition.^[399][400] Its still under investigation, however, as some studies note no difference.^[401]

Fish oil significantly influences glucose and fat metabolism in muscle cells, as well as makes the process more flexible. In moderate doses, fish oil appears to beneficially influence bioenergetics through a combination of nutrient uptake and mitochondrial enzymes. Many of fish oil's anti-diabetic effects can be indirectly linked to increased uptake of glucose into the muscle, as well as increased insulin sensitivity. Preliminary evidence suggests fish oil may increase hypertrophy. This evidence is promising, but limited.

In mice and over longer periods of time, fish oil can preserve the effects of some hormones (insulin, adiponectin) on muscle cells when normally exposed to an obesogenic diet.^[402][366] This may be due to the fish oil component DHA being able to partially reverse the reduction seen in muscle glucose uptake with Palmitic Acid, a saturated fatty acid.^[392] Normalizing the phospholipid ratio (independent of fish oil) does seem to increase adiponectin secretion, however.^[403] It is not clear whether the means to the end (fish oil) or the end (ratio) are the cause of health benefits seen.

In muscle cells, fish oil can also increase AMPK mRNA levels, particularly the AMPKa2 subset.^[392] It has been implicated in doing so in white adipose tissue as well.^[148]

High doses of EPA (500mg/kg) have been shown to reduce PPARd and PPARy expression in muscle cells, and interfere with the production of pro-inflammatory TNF-a and IL-6, which lend credence to its anti-inflammatory claims.^[148] Increased GLUT4 expression was seen at this dose, although lower doses show only increases in GLUT1 translocation.^[404]

Possibly through AMPK, a decreased n6/n3 ratio (more omega 3 relative to omega 6) in muscle cells is associated with increased glucose uptake and better whole-body glucose tolerance independent of mitochondria.^[72] The ratio was approximately 0.5:1-1.5:1 (fish oil) relative to 17.5:1-29.7:1 (control), as measured in the cell membrane.^[72] The muscle cell membrane seems highly response to dietary changes in omega fatty acid intake^[405] and has been associated in vivo with human insulin sensitivity.^[406]

Interleukin-2 (IL-2) secretion from lymphocytes has been noted to be reduced when murine splenic cells are incubated with fish oil fatty acids^[407] which has been noted to affect persons regardless of disease state (lymphocytes in this study isolated from diabetics and controls^[408]). As IL-2 is a cytokine that positively influences T-cell proliferation^[409][410] and is a stimulator of TNF-α^[411] and IL-1 (alpha and beta)^[412] secretion, it is likely the suppression of IL-2 underlies other immunological effects of fish oil supplementation.

Mechanisms are not fully established, as a suppression of IL-2 signalling (assessed by T-cell cycle progression) has been noted with fish oil^[413] and a reduction in Diacylglycerol (DAG) and ceramide has been noted,^[407] with both of those being positive regulators of T-cell proliferation.^[414][415] Although the receptor itself appears unaffected in content,^[416] a decreased signalling potential (as assessed by ERK1/2 phosphorylation) has been noted in T-cells incubated with fish oil^{[417][418][419]} which may be related to reduced recruitment of PKC isomers (alpha and epsilon) to the cell membrane;^[420] this would result in the immunosuppressive effects of fish oil being dependent on membrane rheology and the omega-3:6 ratio.

3.5g fish oil for 12 weeks (otherwise healthy 50-70yrs) has failed to significantly influence IL-2^[421] and failed to influence with 2g daily in persons with isolated hypertriglyceridemia.^[422] However, 18g of fish oil (2,754g EPA and 1854mg DHA) daily in otherwise healthy youth has been noted to reduce secretion of IL-2 in stimulated PBMCs by a variable 23-52%^[409] and in type II diabetics, IL-2 has been reduced following fish oil supplementation (1,548mg EPA and 338mg DHA for 8 weeks) by 17.1%.^[423] The efficacy of fish oil in suppressing T-cell activity and IL-2 does not appear to depend on disease state.

Athletes undergoing exercise given fish oil (6 weeks of 1,300mg EPA and 300mg DHA) have noted an increase in neutrophil (PBMC) produced IL-2 when measured 3 hours post exercise relative to placebo,^[424] which due to IL-2 normally being suppressed after exercise^[425] this was interpreted as a reduction in immunosuppression. This has been noted elsewhere in elite swimmers,^[426]

Interleukin-2 appears to be somewhat unreliably suppressed following supplementation of fish oil, which may be due to the dietary ratio of omega-3 and omega-6. Suppression of IL-2 results from impairing signalling on a T-cell, and the ultimate result is less IL-2, which results in less T-cell proliferation, tumor necrosis factor (TNF-α), and IL-1β.

Tumor necrosis factor alpha (TNF-α) is a pro-inflammatory cytokine that appears to be negatively correlated with omega-3 status.^[427] This cytokine is positively influenced by IL-2 stimulation^[411] and a reduction in IL-2 would result in a reduction in TNF-α. Similar to IL-2, the receptors for TNF-α are unaffected following fish oil supplementation^[428] but unlike IL-2 the stimulation of TNF-α from a stimulated immune cell (in this case, monocyte) does not appear altered.^[429]

Reductions in TNF-α have been noted in otherwise healthy men,^[430][222] and youth (offspring of type II diabetics^[431] or obese youth^[432]) and has been noted with high doses (18g) in young adults^[433] and more moderate doses in persons on hemodialysis.^[434] However, similar to IL-2 there are several null effects suggesting no change^{[421][435][436]} and the demographics of the positive and negative studies overlap including disease states or medical conditions such as hemodialysis^[434][435] and thus it is unlikely that this conditionally works in a certain demographic.

TNF-α concentrations in serum appear to be unreliably reduced following supplementation with fish oil, and due to the high correlation with reductions in IL-2 and TNF-α paired with a plausible mechanism, it is thought that the reductions in TNF-α are due to less circulating IL-2.

C-reactive protein normally is inversely related to serum omega-3 fatty acids^[437] or noted to be positively correlated with omega-6 fatty acids.^[438]

C-Reactive protein has been noted to be reduced at rest in otherwise healthy men following 6 weeks consumption of 2224mg EPA and 2208mg DHA^[430] and women on hormone replacement therapy have experienced a decrease in C-reactive protein with 7-14g fish oil daily (35% and 10.7%, respectively).^[439]

1.5g of fish oil, with or without 800mg Vitamin E has failed to reduce C-Reactive protein.^[429] In stroke recoverers (65+/-10yrs) given 1.2g fish oil daily, C-reactive protein is unaffected,^[440] youth given 0.6g EPA with 0.26g DHA fail to find a reduction,^[428] and in persons with mild hypertriglyceridemia it is also unaffected.^[441]

C-reactive protein appears to be reduced following ingestion of fish oil, although it is somewhat unreliable. Usually, no significant influence is seen. It is possible that very high doses of fish oil can force a reduction in CRP.

3.5g fish oil for 12 weeks (aged 50-70) has failed to influence IL-6 along with 12 other parameters measured (IL-1α, IL-1β, IL-2, IL-4, IL-5, IL-6, IL-8 (or CXCL8), IL-10, IL-12p70, IL-13, IFNγ).^[421] IL-1β and IL-6 have elsewhere been unaffected in persons on dialysis (2,400mg fish oil^[435] and 3,400mg^[441]) as well as healthy persons (775mg EPA^[436]) but IL-6 has been reported to be decreased in otherwise healthy older men given 1.5-2.5g fish oil daily to a magnitude of 10-12%^[222] and in women on hormone replacement therapy with 7-14g fish oil daily.^[439]

The levels of IL-6 secretion in response to LPS stimulation have been seen to not be significantly influenced by 7-14g fish oil in postmenopausal women^[439] but has been noted to be reduced (14%) in medical students in response to LPS.^[442] When assessing neutrophil function, although IL-1β, IL-10, and IL-23 appear to be suppressed (IL-5 and IL-17 trending) IL-6 was not.^[436] IL-1 (both subunits) have elsewhere been noted to be reduced with fish oil supplementation.^[433]

Other interleukins tend to not be significantly affected, although IL-6 appears to be reduced in some instances. The instances where IL-6 are reduced correlate well with instances where C-reactive proteins are reduced.

Chemotaxis is the process by which immune cells are recruited to a specific site in the body in response to secreted cytokines and involves the immune cell rolling along the endothelium until it attaches to cellular adhesion factors (E-Selectin, ICAM-1, and VCAM-1 being most researched) and is pulled into tissue.^[443][444]

The arachidonic acid metabolite known as Leukotriene B4 is a potent chemoattractant (promoting chemotaxis) and it appears that following supplemental fish oil this chemoattractant is reduced and ultimately less chemotaxis occurs for both neutrophils and monocytes in both diseased and healthy populations.^{[445][446][447][448][433]}

There appear to be less chemotaxis associated with fish oil supplementation, which reduces the rate at which immune cells (neutrophils and monocytes) can penetrate tissue. This is an immunosuppressive action, since it lowers levels of the immunosupportive omega-6 fatty acid metabolite (Leukotriene B4) and can reduce immunity and inflammation independent of circulating cytokines.

In animal studies or in vitro, decreased expression of cell adhesion factors have been noted on monocytes,^[449] macrophages,^[450] lymphocytes,^[451] and the endothelium;^{[452][453][454]} This has been noted with both EPA^[454] and DHA^[453] in isolation.

In human studies, supplemental fish oil is able to reduce the ability of isolated immune cells to express cell adhesion factors in response to immunostimulatory agents when tested outside the body^[455][456] and serum levels of soluble adhesion factors have been noted to be decreased^[457][454] There appears to be some associated with age, having this immunosuppressive reduction of adhesion factors occuring in elderly persons but not youth^[457] but even then it is somewhat unreliable.^[421]

Cell adhesion factors are technically reduced by eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). Since omega-6 fatty acids confer the opposite effect, it is plausible that increasing the omega 3:6 ratio confers immunosuppression while the reverse confers immunopotentiation.

Neutrophil EPA and DHA are both reliably increased following supplementation of fish oil.^{[458][459][436]} Supplementation of EPA in isolation is able to increase DHA levels in neutrophil membranes^[436] although very high doses (1600mg) of DHA are required to increase EPA concentration.^[459]

The suppression of Leukotriene B4 seen in vitro^[460] and following oral ingestion in humans^[458][459] appears to be related to a reduced ability of stimulated neutrophils to produce it,^[460] which has been confirmed in humans consuming fish oil (31% reduction with 775mg EPA).^[436] When looking at genes affected by fish oil ingestion in neutrophils, a decrease in protein content of PI3Kα and mRNA content of PI3Kγ (PI3Kβ and PI3Kδ unchanged) has been noted and the reduction in signalling via Akt/NF-kB thought to play a role.^[436]

Fish oil supplementation, due to the interaction of fish oil fatty acids and the neutrophil membrane, appears to reduce the ability of neutrophils to secrete the pro-inflammatory leukotriene B4. This may be related to the suppression of phosphoinositide 3-kinase (PI3K)/Akt.

In regards to oxidative function (neutrophils use oxidation to destroy pathogens via a process known as oxidative burst,^[461] mediated by NADPH^[462]) superoxide production appears to be increased following fish oil ingestion at 2g daily (300mg EPA and 400mg DHA)^[463]

2g fish oil (300mg EPA, 400mg DHA) daily in cancer patients for 8 weeks reversed the chemotherapy-induced reduction in neutrophil count and phagocytosis (neutropenia^[464]) to a 29% increase (all PMBCs) and 14% increase respectively; superoxide production of neutrophils was enhanced 28%.^[463]

Monocytic secretion of IL-1β, IL-6, and TNF-α appears unaffected (MCP-1 activation) following ingestion of fish oil at the dose of 1,500mg daily for 12 weeks with or without 800mg Vitamin E.^[429]

3 weeks of 4g fish oil has failed to significantly influence parameters of monocytic activation or adhesion in healthy persons and those with coronary artery disease.^[323]

Natural Killer (NK) Cells are immune cells that aim to induce cellular death, and are important in cancer prevention. Fish oils at 10% of the diet by weight in rats are able to preserve NK levels closer to pre-operation levels than when compared to a normal feed diet, with no apparent effect in pre-operation (healthy) levels.^[465]

At rest in older individuals (55yr or above), supplemental fish oil (720mg EPA, 280mg DHA) for 12 weeks has noted a 48% reduction in NK cell activity which was not replicated by supplemental DHA (720mg) or other tested fatty acids (arachidonic acid, GLA, alpha-linolenic acid)^[466] despite other studies noting that supplemental DHA may reduce NK cell activity in young men (albeit at 6,000mg daily)^[467] and EPA between 1,350-4,150mg daily for 12 weeks in youth failed to note an increase.^[468]

NK cell activity has been found to be enhanced in otherwise healthy young men subject to exercise (accreddited to an increase in NK cell content that was induced by exercise^[424]) which is thought to be due to decreasing PGE(2) concentrations which naturally retard NK cell activity.^[469] An increase in IL-2 was also noted,^[424] which is known to stimulate NK cell activity.^[470]

There is currently mixed evidence as to the role of fish oil supplementation on Natural Killer (NK) cell activity, with both increases and decreases observed. Further research is needed to determine the cause of the discrepancies.

B cells are a subset of lymphocytes with prominent effects in lung and intestinal tissue that serve to secrete antibodies and some cytokines to act as support cells in defense against pathogens.^[471][472] B-cells have been confirmed to incorporate fish oil lipids into their membranes ex vivo^[473] and following oral ingestion.^[474] It is thought that the mechanism by which fish oil acts is via supporting formation of lipid rafts.^[474][475]

Mice ingesting fish oil (12-14.5% of diet by calories) appear to have increased CD69^[473] and CD40^[474] receptor expression with no influence on CD80, CD86, nor MHCII;^[474] B-cell secretion of IFNγ and IL-6 appears to be enhanced up to 50% ex vivo when stimulated by LPS^[474] and has been noted to occur in vivo.^[473][475] The response of B-cells to antigens appears to not be significantly influenced,^[474] yet higher IgA concentrations have been detected in mice.^[475]

In humans, studies have failed to find alterations in serum concentration of B cells (not activation thereof) with doses of DHA ranging from 720mg to 6,000mg.^{[466][467][468]}

Supplementation of fish oil has not been found to alter the concentration of B-cells in the body, and the binding of B-cells to antigens appear to be unaltered. However, the B-cell appears to be more responsive to inflammatory stimuli and may secrete more antibodies, which would suggests fish oil may enhance adaptive immunity.

B-cell activation may underlie increases in serum interleukin 6 (IL-6) and interferon gamma (IFNγ).

T cells are lymphocytes that can be divided into helper T cells (depending on whether they express the receptor known as CD4, in which case they are referred to as CD4 positive) and cytotoxic T cells (same idea but with a receptor known as CD8). CD4+ and CD8+ are shorthand for helper and cytotoxic T cells, respectively.

CD4+ T-cell activation occurs when T cells are met with antigen presenting cells (dendritic cells) and is mediated by receptors (TCRξ/CD3) and costimulators (ICOS and CD28 as positive modulators, CTLA-4 and CD152 as negative).^[476] It appears that the signalling cascade that results in expression of the T-cell receptor (TCR) is suppressed when fish oils are incorporated into the cell membrane^{[477][478][479]} and the ultimate result is less associated with T-cells and dendritic cells and an immunosuppressive effect^[480][481] which has been confirmed in mice at 1.5% of the diet as omega-3^[482] or 200mg/kg.^[481]

Incorporation of fish oil fatty acids into the T cell membrane appears to be associated with fewer dendritic cells, and thus less activation.

When looking at human interventions fish oil supplementation has repeatedly failed to negatively influence T-cell function with 4,050mg EPA for 12 weeks in young and middle aged men,^[468] 6,000mg of DHA for 90 days,^[467] or their combination at 720mg EPA and 280mg DHA.^[466] However, one trial in healthy older adults with similar doses as before (720mg EPA and 280mg DHA) has noted a suppressive effect associated with EPA (and the omega-6 fatty acid GLA) but not DHA.^[483] This suppressive effect was partially reversed 4 weeks after supplement cessation.^[483]

Interestingly, dietary alpha-linolenic acid (parent omega-3 fatty acid) has been reported to reduce the rate of lymphocyte proliferation at 18g a day^[484] and in vitro arachidonic acid has also shown immunosuppressive effects on lymphocyte proliferation.^[485][486] It is possible that overall polyunsaturated fatty acids play a role (it has been noted elsewhere that T cell suppression is abolished not by COX/LOX inhibitors (eicosanoids) but by lipid antioxidants^[487]), although this does not explain the lack of efficacy of DHA in the same study that noted EPA was immunosuppressive.^[483]

There is mixed evidence as to whether fish oil related immunosuppression is a concern in otherwise healthy people. It is possible that the interaction of fish oil and T lymphocytes is related more to total membrane polyunsaturated fatty acid (PUFA) content than it is to the omega 3:6 ratio, but this claim requires further research.

Fish oil EPA and DHA are both polyunsaturated fatty acids, and each unsaturated bond (double bond) can possibly be oxidized; this would convert the lipid itself into an oxidant capable of producting other oxidants^{[488][489][490]} and is a phenomena common to any unsaturated fatty acid including arachidonic acid.^[491] Although this is sometimes required for bioactivity (first stage in eicosanoid production from DHA requires a free radical), it is desirable to avoid fish oil oxidation ex vivo (prior to supplementation) and is a reason why Vitamin E (reference lipid antioxidant) is almost always included alongside fish oil products (with somewhat lacklustre results, actually^[492]).

Lipid peroxidation can be measured in the blood by either TBARS,^[493] Malondialdehyde (MDA),^[494] 4-hydroxy-2-nonenal,^[495] or oxidative metabolites of eicosanoids (notable 8-iso-PGF2_α);^[496] sometimes serum Vitamin E is also measured, and a decline is thought to be due to being sacrified to prevent lipid peroxidation. Of these measurements, MDA and 4-hydroxy-2-nonenal may be more reliable as 8-iso-PGF2_α has been noted to be decreased in the urine following fish oil consumption^{[497][498][499]} (indicative of antioxidant effects) and fatty acids such as Conjugated Linoleic Acid (CLA) have been noted to interact with 8-iso-PGF2_α indepedent of oxidation before.

Polyunsaturated fatty acids can be oxidized to form lipid peroxides, which can produce a variety of intermediates, like 4-hydroxy-2-nonenal and malondialdehyde. These intermediates are pro-oxidative.

8g of fish oil (1,600mg EPA+DHA) that has already been oxidized prior to consumption in otherwise healthy subjects for 7 weeks has failed to show evidence for lipid peroxidation (Vitamin E, isoprostane, and hydroxy-nonenal measurements)^[500] and makes note of other studies using nonoxidized fish oil failing to find an increase of urinary isoprostane or serum biomarkers (MDA) following fish oil consumption^{[501][497][502]} which seems to be the consensus.^[503]

When selectively looking at evidence that does support a change, the direction is mixed; some studies have reported increases in 4-hydroxy-2-nonenal following DHA consumption in humans^[504] and the combination of fish oil and exercise (although quelled with Vitamin E)^{[505][506][507]} and may increase lipid peroxidation in animals via TBARS.^[508]

There is currently weak evidence to support an increase in lipid peroxidation following fish oil consumption (even if it is oxidized prior to ingestion). There is ample evidence to suggest such an increase does not occur in otherwise healthy adults.

DNA damage can easily be induced by oxidative stress and lipid peroxides are capable of damaging DNA,^[509] and is a mechanism by which oxidation and cancer risk are linked (with inducing damage to DNA being negative).^[510]

Studies in research animals that measure DNA damage note less damage with fish oil supplementation relative to safflower oil as reference^[511][512] (assessed by urinary 8-oxo-7,8-dihydroguanine^[513]) and fail to find induced damage with intakes up to 3,305 and 3,679mg/kg isolated DHA in otherwise healthy rats (male and female respectively; human equivlance of 529mg/kg and 588mg/kg)^[514] although DHA has been noted to increase DNA damage in older rats at 300mg/kg.^[515] One study that confirmed an increase in urinary 8-oxo-7,8-dihydroguanine with a normalized omega-3:6 ratio (relative to high omega-6) also noted enhanced DNA repair enzymes^[516]

Human evidence suggests (epidemiology) that higher serum omega-3 fatty acids are associated with higher rates of DNA damage relative to higher omega-6 fatty acids^[517] but interventions have found no significant influence on DNA fragmentation during a marathon race,^[518] and during pregnancy.^[519]

It is theoretically possible that high doses of fish oil (lowest dose noted being 300mg/kg in rats or 48mg/kg DHA in humans) in a susceptible population, such as elderly people, can enhance the rates of DNA damage, but it is not known how relevant this is to supplementation, since enhanced DNA repair was also noted. Based on human evidence, fish oil supplementation causing DNA damage does not appear to be a concern.

In regards to human studies that measure antioxidant enzymes (notably glutathione peroxidase, catalase, and superoxide dismutase) there do not appear to be significant changes in either a protective nor harmful manner^{[500][520][521]} although limited evidence suggest a small (likely not clinically relevant) increase in glutathione in overweight women.^[522] An increase has been noted in rodent studies when investigating populations that normally have suppressed glutathione (the increase thought to be secondary to preservation of glutathione^[523][524]) but human evidence does not yet replicate this.

Although increases in antioxidant enzymes have been noted sporadically, there is likely no significant effect of fish oil supplementation on the most commonly measured parameters.

Studies that have failed to find a significant influence of fish oil consumption on VO2 max include 3,000mg fish oil (1,300mg EPA; 300mg DHA) for 6 weeks.^[424]

In response to exercise in trained men, high dose fish oil (2224mg EPA and 2208mg DHA) for 6 weeks was able to reduce inflammatory cytokines (CRP and TNF-α) at rest but failed to alter the exercise-induced changes in immune parameters.^[430]

Natural Killer cells appear to have enhanced cytotoxicity for 2 hours after exercise (afterwards they return to baseline)^[525][526] and 3,000mg of fish oil (1,300mg EPA and 300mg DHA) daily for 6 weeks in otherwise healthy men is able to augment the exercise-induced increase in NK cell activity alongside an increase in IL-2 (no changes in IL-4, IL-6, cortisol, or IFN-γ).^[424] This is somewhat different than other anti-inflammatories, as indomethacin has been noted to abolish NK cell activity from exercise^[526] and is thought to be due to increasing NK cell count rather than individual activity (as no evidence for increased activity was noted when cellular concentration was controlled for).^[424] Possible explanations for this include an increase in IL-2 (noted^[424] and known to stimulate NK cell activity^[470]) and a reduction in PGE(2) concentrations, which would be attenuating a negative regulator of NK cell activity.^[469]

Preliminary evidence suggests that fish oil can augment the natural killer cell (NK) cytotoxic response to exercise, but due to the variability seen with natural killer cells in general, it is not known how reliable this effect is.

Both animal^[527] and human^[528][529] data suggest that the reproductive lifespan of females is reflected by levels of follicle-stimulating hormone (FSH), with higher levels suggesting a shortening time of fertility. Mouse data has suggested that fish oil may attenuate reproductive aging and extend the reproductive lifespan.^[530]

One study in humans using 4 g Lovaza (1860 mg EPA and 1500 mg DHA) over two menstrual cycles in 12 women of normal weight and 15 women with BMIs greater than 30 found that FSH levels both before and after an IV infusion of gonadotropin-releasing hormone (GnRH) decreased in normal-weight, but not obese, women by an average of 17%.^[531]

High-dose fish oil supplementation improves markers of reproductive lifespan in women of normal weight, but not in women with BMIs greater than 30.

It is thought that supplemental fish oil can benefit the mothers due to the fetus sequestering EPA and DHA for development, which is thought to underlie the reduction in plasma EPA and DHA seen in pregnant women.^[532]

One study in pregnant women who also had major depressive disorder found that 2,200mg EPA and 1,800mg DHA was able to reduce depressive symptoms during the perinatal period and postpartum^[198] although to counter this intervention is a fairly large amount of trials using a range of EPA or DHA supplementation reporting null effects.^{[533][534][194][195][196][197]} It is possible that fish oil merely acts in pregnant women the same as in all depressed persons (EPA being more anti-depressive in persons with worse depression) and that perinatal and postpartum related depression that is not to the magnitude seen in major depressive disorder is unaffected.

Fish oil, particularly eicosapentaenoic acid (EPA), does not appear to have any special effects in regard to depression associated with the perinatal and postpartum period. EPA is antidepressive in the most depressed cohorts. This has also been noted in pregnant women, who also had major depressive disorder. Many trials with depression of a lesser magntiude have reported a failure of fish oil to benefit depression.

Gestational diabetes is a transient state of diabetes occurring in 3-8% of pregnancies^[535] with mixed survey evidence as to whether dietary fat from fish is associated with gestational diabetes risk.^{[536][537][538]}

800mg DHA daily in pregnant women has failed to significantly alter risk for gestational diabetes.^[539]

Fish oil does not appear to significantly reduce the risk of gestational diabetes.

Pre-eclampsia is a pregnancy complication associated with vasoconstriction and endothelial damage, and its pathology appears to involve prostaglandins.^[540][541] One meta-analysis^[542] has claimed insufficient evidence to support an effect of fish oil (using mostly underpowered studies) and a more robust trial of 2399 women has failed to find a protective effect with 800mg DHA supplementation (1.5g omega-3).^[539]

There is insufficient evidence to support fish oil's role in reducing the risk of pre-eclampsia.

There may be less death in infants associated with maternal DHA consumption, with one study noting that while control experienced 12 and 5 neonatal deaths and convulsions (respectively) 800mg DHA reduced this to 3 and 0.^[539] This trial stated that further research is needed.

Infant weight has been noted to be increased with supplemental fish oil to a moderate degree (47g, 95% CI of 1-93g) and a slightly increased time to birth (2.55 days; 95% CI of 1.03-4.07 days), but despite the increased time to birth this analysis only found a protective effect against premature birth when measuring before 34 weeks (no effect at 37 weeks).^[542] This increased time until birth resulting in greater infant weight has been noted elsewhere with 2,700mg fish oil.^[543]

Fish oil may reduce the risk of birth complications and the risk of premature birth (relative to not consuming any omega-3 fatty acids), with a moderate amount of evidence to support increased birth weights and prolonged time to birth. Very preliminary evidence suggests fish oil supplementation can reduce the risk of neonate death.

Consumption of omega-3 fatty acids (or any polyunsatuated fatty acid) is known to cross the placental barrier^[544][545] via FATP transporters (particularly FATP4)^[546] to regulate nervous system development.^[547] Unlike adults, the fetus in not capable of inherently synthesizing sufficient omega-3 fatty acids and thus parental provision is mandatory^[547] and supplemental DHA has been confirmed (in primates) to be approximately 8-22 times more effective at increasing neural DHA stores in offspring than the parent omega-3 fatty acid (ALA).^[548] It should be noted that arachidonic acid (omega-6 counterpart to EPA) is also vital for cognitive development, but seems to be less responsive to the diet suggestive of better regulation.^[549]

Docosahexaenoic acid (DHA) plays a critical role in the neural development of the fetus during pregnancy. The fetus depends on parental provision of DHA, either through supplementation or the diet. It is for these reasons that supplemental fish oil is thought to increase cognitive development in unborn children, but it is unknown if the omega 3:6 ratio plays any role here, as arachidonic acid is also critical.

One review and meta-analysis (11 trials reviewed with a sample of 5272; 7 in meta-analysis) has been conducted assessing cognitive and visual performance of offspring of mothers who consumed omega-3 fatty acids during pregnancy^[550] assessing the following trials^{[551][552][553][554][555][534][556][557][558]} (two not found online) noted that no significant effect of fish oil on cognitive capacity could be reliably determined while the one statistically significant benefit on Developmental Standard Scores (3.92; 95% CI of 0.77-7.08) had a high risk for bias.^[550] The bias mentioned in the review to be of most concern were high attrition rates (27–86%) and unclear and haphazardous randomization while publication bias could not be ruled out due to some studies not being fully published (but similar null results being reported in the abstracts).^[550]

Further study is needed to determine the effects of maternal fish oil supplementation on the cognitive development of offspring. Although a benefit cannot be ruled out, current evidence does not support any beneficial effects.

-

Dietary DHA intake is critical during the first three months of life, where it correlates greatly with neural DHA levels (as assessed by autopsy reports)^{[559][549][560]} and due to this importance it is a mandatory additive to baby formulation^[549] and provision to preterm infants highly recommended.^[561] DHA is also a component of breast milk (and thus provided during breast feeding)^[562] of which the concentration of DHA in breast milk is correlated with the mother's diet.^{[563][564][565]}

Supplemental ALA (from flaxseed or plant sources of omega-3) is ineffective in raising breast milk concentrations of DHA, despite an increase in breast milk concentrations of ALA.^[566]

It is critical for infants to consume docosahexaenoic acid (DHA) during their first few months of life in order to support cognitive development. DHA is found normally in breast milk, and is a mandatory additive to any infant dietary formulas.

Maternal intake of salmon (3.45g of fish oil per week) during the 20 weeks of pregnancy prior to birth is sufficient to increase breast milk concentrations of EPA (80%) and DHA (90%) when the mothers normally did not consume fish.^[567] Increase in breast milk have been noted with supplementation as well during both pregnancy^[568][569] and during lactation exclusively^[570] or both;^[571] with some manner of dose-dependence being noted and highest levels being reached after 2 weeks supplementation.^{[570][571][568][569]} Daily intake does not appear required, as supplementing fish oil (or consuming fish) repeatedly and then ceasing for 2-4 weeks does not normalize breast milk concentrations of EPA/DHA although a decline does appear present.^{[567][572][573][574]}

One study (salmon twice weekly, giving 3.45g fish oil) has noted a reduction in breast milk IgA concentrations.^[567]

Breast milk DHA concentrations reflect dietary DHA concentration. Both fish intake and supplementation can elevate levels of breast milk DHA. Supplementation does not appear to be required, as up to 90% increases have been detected with 3.45g of fish oil per week (via salmon consumption). Daily ingestion of fish oil supplements or fish products does not appear to be required.

Omega-3 fatty acids, particularly DHA, are known to be highly involved as modulators of retinal capillary integrity, neovascularization and inflammation^[575] related to their protectins and resolvins.^[114]

DHA has been noted to be decreased in the retina of diabetic rodents^[576][577] (as well as plasma of humans^[578]), and dietary provision of fish oil (5% of diet totalling 10.26% DHA and 14.16% EPA) is able to abolish the increase in angiogenesis and grealty attenuate inflammatory biomarkers in type II diabetic rats (relative to soybean oil)^[579] as well as type I.^[580] This protection actually seems to extend beyond diabetes (being noted in a mouse model of retinopathy of prematurity at 2% of the diet^[581]) and seems to be related to increased eicosanoids (Neuroprotectin D1 and Resolvin E1/D1 being detected^[581]) and a normalized omega3:6 ratio, as the Fat-1 mouse line (genetically altered to normalize the ratio) appears to have reduced risk of angiopathy.^[582]

The increase of acid sphingomyelin (ASM) is also fully normalized relative to omega-6 (soybean) controls^[579][580] and is thought to be involved in the pathology of retinal angiogenesis.^[579]

Normalizing the omega 3:6 ratio appears to be very protective against retinal angiogenesis. DHA is also important in this protection. In rodents, this protection is nearly absolute during a reasonable intake of dietary fish oil. This means people supplementing fish oil will most likely experience a similar effect.

Eicosanoids derived from arachidonic acid (AA) appear to be involved in the destruction of pancreatic β-cell (insulin producing cells of the pancreas^[583][584]) population,^[585] with the PGE₂ produced from COX enzymatic activity being most relevant^{[584][586][587]} and related to COX2 specifically.^[588][589] COX2 is overexpressed in pancreatic β-cell due to transcription factor NF-IL6,^[589] and its increased activity by the cytokine IL-1β appears to link overall inflammation and PGE₂.^[590]

When PGE₂ is elevated in a pancreatic cell, insulin secretion is suppressed^[587][591] and proliferation of β-cells are suppressed.^[592]

12-HETE (catabolite of arachidonic acid) is detected in pancreatic cells^[593] and appears to also contribute to β-cell toxicity via NADPH oxidase-1 dependent mechanisms following conversion by 12-LOX (also expressed in pancreatic cells^[593])^[594] yet does not suppress insulin secretion^[595] like similar structures (5-HPETE^[596] and both LTB4 and 15-HETE^[595]); although dietary inclusion of omega-3 fatty acids can reduce the arachidonic acid metabolite LTB4,^{[458][436][459]} results in humans investigating 5-HETE suggest no suppression.^[458]

There is limited evidence to draw connections between the above mechanisms and supplemental fish oil. One study in Fat-1 mice (with a normalized omega3:6 ratio) noted more protection against inflammation induced cell death^[597] and seem to be resistant to the pancreatic toxin streptozotocin.^[598]

There is no human evidence to suggest a link between fish oil consumption and improved pancreatic cell functioning. This relationship appears to be plausible in mouse models, where it has been observed.

Fish oil supplementation beneficially effects kidney function in those with diabetes (and at risk for diabetic nephropathy) at 4g daily,^[599] whereas animal models with higher doses show more dramatic protection.^[600] The mechanism may be through reducing pro-inflammatory cytokines in the kidney and through eicosanoid production.^[601][602] There isn't the largest body of literature on this function exclusively, and at least one recent review suggests that a final conclusion on fish oil's effects on renal function is preliminary.^[603]

There have been correlations established between dietary PUFA (Polyunsaturated fat) intake of omega3s and prevention of renal disease, suggesting a preventative role may also exist.^[353]

Omega-3 supplementation may play a protective role in the development of renal pathology associated with diabetes.

High intake of dietary fish products are associated with reduced risk of skin cancer^[604][605] (while higher omega-6 intake is associated with increased risk^[606]) which is thought to be related to reduced sunlight-induced immunosuppression (noted in rats^[607] and humans with 4,000mg EPA+DHA^[608]) resulting in reduced tumor multiplicity and increased tumor latency.^{[609][610][611]}

The mechanisms are thought to be related to membrane fatty acid content (and due to that, eicosanoid and prostaglandin signalling) as EPA is known to compete with arachidonic acid in the membrane^[612] and higher dietary intakes of omega-6 in research animals augment solar radiation induced skin carcinogenesis secondary to immunosuppression.^{[607][613][609]}

Fish oil supplementation, and specifically the omega 3:6 ratio in skin membranes, appears to have a protective mechanism in regard to skin cancer. The links between fish oil and skin cancer development are related to immunity, with higher fish oil intake reducing the immunosuppression induced by sunlight.

In pancreatic cancer cells, both EPA and DHA induce apoptosis (not inhibited by COX inhibitors) via inducing reactive oxygen species and subsequent autophagy at a concentration of 10mM, which was abolished (in vitro) with the addition of Vitamin E; when mice are injected with pancreatic tumors (MIA-PaCa-2 cell line) and fed 5% of the diet as fish oil, it appears that fish oil was associated with a reduction of tumor volume to approximatley a third of control.^[614]

Since omega-3 fats tend to suppress inflammation, which is a suspected contributor to carcinogenesis in prostate cancer,^[615] the effect of fish oil supplementation on prostate cancer risk is of interest. One study found a reduction in prostate cancer risk with increased consumption of omega-3 fats. This nested case-control study was derived from blood collected from 14,916 healthy men in 1982. Blood samples were analyzed for fatty acid levels from 476 of these men that were diagnosed with prostate cancer in comparison to age-matched matched controls. The study found that blood levels of omega-3 fats were inversely related to overall prostate cancer risk, with a relative risk (RR) of 0.59 (95% confidence interval = 0.38-0.93).^[616]

One study found that increased omega-3 fats are associated with decreased risk for prostate cancer.

In contrast, a later case-cohort study by Brasky and colleagues found the opposite result, where high blood omega-3 levels were associated with increased risk for prostate cancer. The study was observational by design, and based on the SELECT trials. Of the participants in the SELECT trials Brasky et al got a sample of persons from that trial that were diagnosed with prostate cancer (n = 834) and made note of how many developed advanced prostate cancer (n = 156). These were compared to 1393 age and race-matched participants of the SELECT trial who did not have prostate cancer. The investigators measured serum levels of the omega-3 fats EPA, DHA, and DPA and stratified the groups into quartiles to examine whether there was an association. The results showed that persons who had prostate cancer were more likely to have higher omega-3 fats in their blood. When comparing the quartiles against one another, the highest levels of fish-based omega-3 fatty acids were associated with increased risk as assessed by Hazard Ratio (HR) for total cancer HR = 1.23 (95% confidence interval (CI) = 1.07-1.40), low-grade prostate cancer HR = 1.24 (95% CI = 1.07-1.43), and high grade prostate cancer HR = 1.24 (95% CI = 1.00-1.54) prostate cancer. Looking at individual omega-3s, DHA had a HR showing an association with total prostate cancer HR = 1.21 (95% CI = 1.07-1.37), low-grade prostate cancer HR = 1.21 (95% CI = 1.06-1.38), and high-grade prostate cancer HR = 1.26 (95% CI = 1.03-1.54), while DPA was only associated with total prostate cancer HR = 1.23 (95% CI = 1.03-1.46) and low-grade prostate cancer HR = 1.30 (95% CI = 1.08-1.57). Taking all of the omega 3 fatty acids into account together (EPA + DHA + DPA), the researchers found a 43% increased overall prostate cancer risk.

An observational study based on the SELECT trial found a positive association between increased omega 3 fatty acids in the blood and prostate cancer risk. Taking all of the omega 3 fatty acids into account (EPA + DHA + DPA), there was a 43% increased risk for prostate cancer.

It is important to note association does not mean causation. No studies have yet established a causative relationship between increased omega-3 fatty acid levels and prostate cancer. More research is needed to make recommendations for or against fish oil supplementation as far as prostate cancer is concerned.

Studies on blood levels of omega-3 fats relative to risk for prostate cancer have reported mixed results, with studies showing either an increased or decreased risk for developing prostate cancer depending on the study design and population studied. It is important to note that no causative relationship has been established yet between omega-3 fats and prostate cancer.

A randomized, placebo-controlled trial of 345 mg DHA per day (with no EPA) over 4 months found no effect on ADHD symptoms in children aged 6-12 with ADHD.^[617]

However, other trials which included EPA have yielded positive results. Some of these trials were done in children with ADHD-like symptoms, while others were done in children with established ADHD.

In one such trial in children with ADHD-like symptoms, children age 8-12 who showed features of ADSD who also had other learning disabilities (mainly dyslexia) improved on the CPRS-L rating scale after being supplemented with 480mg DHA and 186mg EPA for 12 weeks versus an olive oil placebo.^[618] Another study using a blend of vitamins and essential fatty acids (480mg DHA, 80mg EPA, 40mg Arachidonic Acid, 96mg gamma-linolenic acid, 24mg Vitamin E) showed a decrease on a hyperactivity rating scale (Disruptive Behavior Disorders (DBD) Rating Scale for Attention) versus olive oil placebo in children with ADHD-like symptoms also reporting thirst and skin problems.^[619]

Other trials using both omega-3s in children with ADHD have also shown some positive results. In children using 558mg EPA and 174mg DHA for 15 weeks in children with aged 7-12 years with scores in the top 2.5% above the population average on the Conners ADHD Index, lower symptom reports were found on parent ratings of core ADHD symptoms, inattention, hyperactivity/impulsivity, and on the Conners Index.^[620] Another 16-week trial using 750mg EPA and 300mg DHA in children with ADHD found a decrease in several Conners subscales as well as DSM-IV symptoms of ADHD.^[621] An additional study in boys aged 8-14 that used 650 mg each of EPA and DHA for 16 weeks found improvements in the Attention Problems subset of the Child Behavior Checklist (CBCL) for children with ADHD (with improvements being seen in the overall CBCL score for children both with and without ADHD).^[622] However, one trial using 600 mg EPA and 120 mg DHA in children with ADHD for 16 weeks found only improved working memory function versus placebo, with no effect parent- and teacher-rated behavior or measures of attention.^[623]

EPA alone has also been found to be effective in one study. That study used 500 mg of EPA for 15 weeks in children aged 7-12 with ADHD, and found an improvement in the inattention/cognitive subscale of the Conners parent/teacher rating scale, although the overall Conners score was unchanged.^[624] However, subgroup analysis of this study found that some children responded strongly to supplementation (greater than 25% improvement) and children with lower baseline blood EPA levels tended to respond better.^[624]

One meta-analysis has also addressed the efficacy of omega-3 fatty acids in children, concluding that supplementation, especially that with high EPA content, was mildly efficacious in reducing symptoms of ADHD, having a much smaller effect size than most pharmaceuticals on the market.^[625] However, a Cochrane review on polyunsaturated fatty acids (PUFAs) concluded that there was little evidence to support the efficacy of PUFAs in ADHD.^[626] This review included many studies that examined PUFAs besided EPA and DHA, however, such as evening primrose oil,^[627] gamma-linolenic acid,^[628], and short-chain PUFAs^[629] in addition to studies that focused primarily on DHA content.^[630] Thus, the Cochrane review may not be entirely relevant to fish oil in particular.

Fish oil supplementation with a high EPA content may be effective in reducing symptoms of ADHD both in children with clinically diagnosed ADHD and in children with only some symptoms related to it.

Lupus erythematosus (Lupus) is a disease state characterized by arthritis, vasculitis, rash, and the involvement of the central nervous system that appears to be associated with reduced omega-3 (EPA and ALA) and GLA content in lipid membranes.^[487] Fish oil is investigated for treatment of lupus as the first pilot study in humans noted full remission from symptoms associated with mixed EPA/DHA supplementation for 8 months or more (162mg EPA and 144mg DHA).^[487]

Other studies note that 3g of omega-3 fatty acids from fish oil^[631] or EPA in isolation^[632] in patients with lupus (for up to 24 weeks of supplementation) appears to reduce general symptoms as assessed by the rating scales of SLAM-R (33-34% reduction^[632][631]), BILAG (51% reduction^[631]), and may benefit blood flow as assessed by flow-mediated vasodilation^[631] and has mixed effects (55% of patients) on treating dyslipoproteinemia.^[633] In rats, the progression of nephtritis is attenuated with fish oil supplementation which promotes lifespan^[634][635] and this effect has been noted in humans (15g for 1 year) to a much lesser degree, where proteinuria is reduced nonsignificantly and glomerular filtration rate is unaffected.^[636]

There is some counter evidence, with at least one study noting that the benefits observed at 3 months was no longer present at 6 months (200mg/kg bodyweight at 18.6% EPA and 12.1% DHA)^[637] and benefit has been noted with superloading elsewhere (20g daily for 12 weeks; study did not proceed towards the 6 month timeframe).^[638] Interestingly, the lone other study to use a prolonged high dose intervention (15g daily for 1 year) also noted lacklustre results in renal symptoms of lupus.^[636]

Fish oil supplementation appears to play a role in controlling the symptoms associated with lupus, but there is evidence to believe that long term dosing of high levels of fish oil eventually abrogates the benefit and that lower doses (the standard 180mg EPA and 120mg DHA dosage) may be more beneficial.

Since gamma-Linolenic acid (GLA) and dihomo-y-linolenic acid(DGLA) appear to be associated with symptoms of lupus, it is plausible that the omega 3:6 ratio in lipid membranes is important, and that high doses skew the ratio too severely in favor of omega-3.

Dietary intake of EPA is known to increase skin levels of EPA following ingestion of 10g fish oil (1,800mg EPA and 1,200mg DHA^[639]), 4g fish oil (95% EPA ethyl esters)^[640] and 10g fish oil (1,800mg EPA and 1,200mg DHA).^[639]

Solar radiation is known to transiently suppress the immune system^[641] in a dose-depedent manner^[642][643] and persons with contact dermatitis (a topical allergic reaction) can be used as research models to assess photoimmunosuppression.^[644] Using this model, 5g fish oil (3,500mg EPA and 500mg DHA) has been demonstrated to reduce photoimmunosuppression by 6.9-11%.^[608]

Sunlight induced erythema (reddenning) appears to be reduced following supplement ingestion for 3 months or so, with the time of exposure required to induce erythema increaasing 37-117%.^[640][639] A reduced sensitivity to sunburns has also been reported.^[640]

Fish oil appears to protect the skin against sunlight, with benefits during sunlight-induced immunosuppression as well as sunlight-induced reddening (erythema). A reduced risk of burns has been reported to be associated with fish oil consumption. These mechanisms appear to be sensitive to the omega 3:6 ratio.

There are mixed reports on how oxidation is influenced in the skin following fish oil consumption, with one reporting no alterations in DNA damage per se but reduced sunlight-induced DNA damage^[640] while elsewhere lipid peroxidation (TBARS) has been noted to be increased in skin tissue.^[639]

It is possible that fish oil supplementation can increase lipid peroxidation in the skin, but this has not yet been linked to adverse toxicological effects, like as DNA damage. These effects are reduced following exposure to sunlight and fish oil, relative to sunlight alone.

Fish oil was first thought to reduce pressure ulcers in critically ill persons in a study that was confounded with GLA and antioxidants^[645] where the benefit was thought to be due to improved blood flow noted with this combination therapy^[646] or changes in immunity;^[647] fish oil in isolation has been noted to confer these properties by reducing pressure ulcer formation by 20-25%^[648] although this study has been criticized for its lack of data.^[649]

It is theoretically possible that fish oil supplementation can reduce pressure ulcer formation in the critically ill, but this claim lacks a strong evidence and requires further investigation.

There is a surprisingly lack of literature investigating the link between fish oil and hair, despite the knowledge that prostaglandins are involved in hair growth regulation.^[650] The machinery appears to be present, as phospholipase A₂ has been detected in hair follicles on the outermost epithelium^[651] and receptors for PGE(2) (EP3 and EP4) have been detected on the dermal papillae.^[652][653] Drugs that act like PGE₂ and PGE₂α (Viprostol and Latanoprost) have been noted to induce hair growth^[654][655] associated with promotion of hair follicles to anagen phase^[656] while COX2 overexpression induces hair loss (restored with COX2 inhibitors^[657][658]); both overexpressing and abolishing phospholipase A₂ activity reduces hair growth.^[651]

Both PGE₂ and PGE₂α have been found to be produced locally in hair cells^[650] and theoretically can induce hair growth via their receptors (EP3 and EP4), although all prostaglandin receptors appear to be expressed in various areas of the hair follicle.^[659]

There is furthermore a plausible link between prostaglandin receptors and androgen metabolism, via the receptors that respond to both classes of molecules (AKR1C1 to a lesser degree, and both CBR1 and AKR1C3^[660] are expressed in hair follicles^[650]); PGE(2) may not increase testosterone per se, however.^[661] Additionally, the enzyme known as prostaglandin D2 synthase (which converts arachidonic acid metabolite PGH2 to PGD2) is known to be induced by androgenic signalling^[662][663] and the higher PGD2 is correlated with and has a causative role in suppressing hair growth in man;^[664] EPA has been noted to suppress PGD2 in mast cells (immune cells) by competing with arachidonic acid at the COX enzymes^[665] and has been noted to suppress PGD2 elsewhere in macrophages.^[666]

Prostaglandin E2 (PGE2) production from eicosanoids appears to be a positive modulator of hair growth, while inflammation appears to negatively influence hair growth via prostaglandins (and COX2). The connection between this relationship and fish oil supplementation is not known at this time.

Supplemental polyunsaturated fats to dogs (9.3g linoleic acid with either 3.3g or 0.42g of ALA per 1000kcal) has been noted to increase hair softness and glossiness but also greasiness and scaliness, which was thought to be associated with increased cholesterol esters found in hair cells.^[667] This has been noted elsewhere, again in dogs.^[668]

Further study is needed to determine the effects of fish oil supplementation on hair.

In cellular membranes phosphatide subunits bind to fatty acids, uridine, choline, and some other molecules such as amino acids to form the components of the membrane.^[669] Uridine is important in the body as it confers a pool of substrate^[228][670] to make the molecule Cytidine-5'-triphosphate, the availability of which is the rate-limiting step^[671] in transfering a Cytidine monophosphate from Cytidine-5'-triphosphate to phosphocholine, which would result in the production of CDP-choline. As CDP-choline highly interacts with DHA to form phosphatidylcholine phospholipids in membranes,^[672] supplemental Uridine (via CDP-choline) is thought to be synergistic.

Uridine provision is the rate limiting step in the body when producing phosphatidylcholine (via CDP-choline), associated with docosahexaenoic acid (DHA). Provision of uridine accelerates production.

Ingestion of choline, uridine, and DHA (constituents of phosphatidylcholine synthesis) in rats appears to increased brain membrane concentrations of phosphatidylcholine by 50% or more^[273] and appears to be synergistic as any agent alone increased phosphatidylcholine by 13-22% in this study where combination of all increased PC concentration by 45%.^[673] This positive influence on brain phospholipids also applies to other phosphatamides such as phosphatidylethanolamine, phosphatidylinositol, and phosphatidylserine^[673][274] and while DHA is less effective than combination therapy uridine in isolation tends to be inactive which rules out mere additive effects.^[274]

Uridine and docosahexaenoic acid (DHA) appear to be synergistic in increasing phospholipid concentrations in the brain of research animals (uridine augments DHA's ability to increase). There may be further synergism with choline. This suggests potency between Krill Oil, which is DHA complexed, phosphatidylcholine, and uridine supplementation.

Combination therapy with DHA (300mg/kg) and Uridine as 5’-monophosphate (0.5%) has been found to increase dendritic spine density (36%) in the adult gerbil hippocampus over 4 weeks; combination therapy appeared to be more effective than DHA alone (18%) and since uridine was inactive in isolation it appeared to be synergistic.^[274] There was no influence on spine size, only density.^[274]

Uridine has been implicated in augmenting dendritic spine density in the hippocampus, induced by docosahexaenoic acid (DHA) provision, which is thought to underlie possible synergism in memory enhancement.

Linoleic acid (LA) is the parent omega-6 fatty acid, which is bioconverted into arachidonic acid in the body and tends to antagonize the effects of fish oil supplementation.

One study that measured the triglyceride lowering effects of fish oil noted that while fish oil (3.1g daily) reduced triglycerides by 51% when paired with a low LA food product, but by increasing linoleic acid by 7.3g per day with a high LA food product the reduction in triglycerides was attenuated to 19%.^[674] This possibly extends to food consumption as well, with fish having a higher omega 6:3 ratio having less benefits to artherogenesis (plaque in arteries) than do a similar amount of omega-3 with less omega-6.^[675]

The effects of fish oil on immune cells may not be affected by linoleic acid consumption.^[676]

The parent omega-6 fatty acid may antagonize the triglyceride-lowering effects of fish oil, even when supplemented or eaten. If fish oil is being used for that purpose, coingesting the two is not advised.

Astaxanthin is a carotenoid that serves as a lipid antioxidant, and is thought to in part contribute to the health benefits of consuming red fish (salmon) or krill oil.

A decrease in proliferative capacity of lymphocytes has been noted with fish oil (17mg/kg bodyweight) paired with astaxanthin (1mg/kg) which is thought to be either additive or synergistic since no reduction was seen with either agent alone;^[677] said reductions have indeed been noted with astaxanthin (5µM^[678]) and with PUFAs elsewhere,^[679] but in higher concentrations/dosages.

Curcumin is the main bioactive of the curcuminoids derived from some spices (usually turmeric but with a small content in ginger).

The combination of DHA and Curcmin in isolated breast cancer cells (MDA-MB-231, MCF-7, and three others) appears to be synergistically antiproliferative, where although 30μM of either compound in isolation failed to act the combination (18μM DHA and 12μM curcumin) suppressed proliferation.^[680] There was less genomic activity with the combination than with isolated curcumin, and proteins that were unaffected by either compound in isolation but affected by the combination include CXCR4 (suppressed), aromatase (induced), SERPINB5 (suppressed), PPARγ, and p53 phosphorylation.^[680]

Fish oil and curcumin appear to be synergistic in suppressing breast cancer cell proliferation, which is associated with inducing some proteins not seen with either isolated compound.

There may be synergistic anti-inflammatory effects in macrophages with curcumin and both fish oil fatty acids as assessed by LPS-induced PGE(2) production^[681] and in a rat model of colitis (inflammatory bowel disorder) curcumin and fish oil have been noted to be synergistic.^[682]

Brain derived neurotrophic factor (BDNF) is a protein that positively regulates synaptic growth and neuronal growth^[683] and due to its positive influence on long term potentiation and synaptic growth^[684] it is thought to be a molecular target of cognitive enhancement.

DHA is known to support neuronal membrane fluidity^[685] and increase BDNF concentrations^[686] and DHA (1.25% of diet) has been noted to augment the BDNF-induced learning that occurs from exercise in rats.^[686] Due to the synergism present with DHA and exercise and the ability of curcumin to also support BDNF levels^[687] it has been hypothesized that they are synergistic.^[688]

Theoretically, curcumin may be synergistic with fish oil in increasing brain-derived neurotrophic facto (BDNF), and thus learning. This effect has not been demonstrated in practice.

The carotenoid from seaweed, fucoxanthin, has been found to be slightly synergistic with fish oil for attenuating weight gain in obese and diabetic mice.^[689] The addition of fish oil at 6.9% of the diet (quite a high dose) was found to make 0.1% dietary fucoxanthin as effective at suppressing fat gain as double the dose.^[689]

Fucoxanthin and fish oil may be synergistic in their anti-obesity effects. Further study is needed to confirm this relationship.

Interestingly, fucoxanthin can increase liver levels of DHA independent of fish oil consumption.^[74]

When a fenugreek oil (formulated with 15% fish oil by weight) was given to diabetic rats at 5% of food intake, it resulted in a 51% decrease in blood glucose levels after a meal due to decrease the activity of carbohydrate digesting enzymes in the pancreas (46% reduction in α-amylase, 37% reduction in maltase) and plasma (52% α-amylase, 35% maltase).^[690] The combination 5% group was slightly more potent than the 5% fenugreek group and much more than the 5% fish oil group.

A protective effect on pancreatic beta-cells was also noted with this combination^[690] as well as decreases in triglycerides attributed to the fish oil component. Said infusion also normalized the increase in ACE that diabetic rats experience.^[690]

Taurine is a sulfur-containing amino acid which is seen as anti-diabetic, it was investigated alongside Fish Oil fatty acids due to both being present in high amounts in seafood. The expected increases in body fat seen in diabetic mice subject to both taurine (at 4% of the diet) and fish oil was lesser with the combination than either nutrient alone, and the levels of insulin and glucose were lower with the combination than either molecule in isolation.^[691]

Taurine potentially has additive benefits when taken with fish oil.

Among statin-treated patients, a more normalized ratio is still associated with slower rates of atheroma progression when compared to a high omega-6 ratio^[692] and intervening to reduce the ratio via delivering dietary omega-3 fatty acids reduces cardiometabolic risk factors.^[693] Due to the benefit of a normalized ratio even in statin patients and the high level of safety seen with fish oil during statin therapy,^[694] they are thought to be worthy combinations.

Several studies have suggested that the combination of a statin drug (usually simvastatin) and omega-3 fatty acids (2-4g EPA+DHA) is complementary on improving HDL-C while it can reduce triglycerides (inherent property of fish oil) and appears to synergistically reduce LDL-C (which is notable, as fish oil alone may increase LDL-C).^{[305][333][695][694]} Potential synergism or additive benefit has also been noted in regards to vascular function in diabetics^[696] and this synergism appears to apply to lovastatin (via its nutraceutical form of Red Yeast Rice^[697]).

The synergism between omega-3 fatty acids and statin drugs appeared to be similar in result to but either outperformed^[698][699] or similar in potency^[700] to a statin and fibrate combination therapy (rosuvastatin and fenofibrate).

Fish oil supplementation appears to be synergistic with statin drugs, in terms of improving cardiometabolic parameters, since both inherently lower triglyceride count and augment the low-density lipoprotein reduction of statin drugs. It is unclear at this time if this synergy has a role to play in people using statins and fibrates already.

Although there are numerous toxins associated with fish consumption, mercury is the one at the forefront of concern due to its correlation with omega-3 intake in fish^[701][249] and its adverse effects on child cognition when consumed by pregnant mothers, as mercury can pass the placental barrier^[702] and reach the child; as assessed by umbilical cord exposure.^[703] Other toxins do not have as strong a correlation in children, such as PCBs and Dioxins^[704] and although a concern, are less of a concern relative to mercury.

Additionally, mercury just has an adverse pharmacokinetic profile. When fish is cooked, the methylmercury binds to meat proteins^[705] and 95% of ingested mercury is absorbed within 2 days^[706] where it persists in the body for 70-90 days.^[707]

In some epidemiological research, high consumption of mercury is related to heart disease risk, mostly with whale meat^[708] but related to the mercury intake itself.^[709][710] The omega3s offer a protective effect though, and avoiding the highest sources of mercury reduces a lot of risk on cardiovascular disease.^[711][712] Only the highest sources of mercury (shark and whale) seem to cause enough of an effect for significance to arise in this epidemiological research, although the effect of mercury per se may be dose dependent.

In food, one recent review noted that the safest fish in terms of "High omega-3, low mercury" were salmon, trout and shrimp.^[710] They examined those three, as well as other common fish (cod, halibut, shark, three forms of tuna, mackerel, seabass, snapper, tilapia and swordfish) for mercury content. Their results were:

• Mackerel, Cod, Trout, Catfish, Farm Raised and Canned Salmon, Shrimp, and Tilapia were all under 0.1mcg/g (0.044, 0.026, 0.020, 0.014-0.015, 0.027-0.076, 0.012, and 0.020; respectively)
• Halibut and Canned Light Tuna crossed over 0.1mcg/g (0.069-0.160, 0.030-0.102)
• Albacore Tuna, Snapper, Ahi Tuna, Chilean Sea Bass, Swordfish and Shark all were above 0.1mcg/g (0.148-0.259, 0.465, 0.291, 0.194, 0.293 and 0.541; respectively)

They^[710] also averaged the omega-3 content of said species by doing a literature review of a few studies^{[713][714][715][716][717]}, their averages were:

• Tilapia and Snapper had less than 0.2g/3oz (0.115 and 0.170)
• Cod, Light Tuna, Catfish, and Shrimp had between 0.2-0.4g/3oz (0.204, 0.238, 0.260, 0.301)
• Seabass and Swordfish had beteen 0.4-0.6g/3oz (0.417, 0.493)
• Shark, Ahi Tuna, and Albacore Tuna had between 0.6-0.8g/3oz (0.711, 0.716, 0.732)
• Halibut and Trout were between 0.8-1g/3oz (0.800, 0.818)
• Salmon and Mackerel were above 1g/3oz (1.090-1.582, with farm raised salmon having more; canned mackerel at 1.251)

This may not extend to all contaminants in whole fish, however. One worldwide study of farmed versus wild salmon found higher levels of 13 out of 14 contaminants measured in farmed salmon on average.^[718] When broken down by region, European and North American farmed salmon was higher than wild salmon in all 14 contaminants, whereas South American farmed salmon was higher in six contaminants and lower in two.^[718] Based on the levels of dioxin-like contaminants (DLC) found, the same research group estimated that to achieve the World Health Organization's tolerable daily intake of DLCs, farmed salmon would have to be consumed less than 10 times per month generally, with farmed salmon from Northern Europe, where contamination is higher, limited to less than 4 meals per months. The group also estimated that wild Pacific salmon could be consumed at levels greater than once per day and still lie below the tolerable daily DLC intake.^[719]

In supplements, fish oil capsules and cod liver oil seem to be relatively low in mercury. Although products will vary in concentrations (depending on the fish used), one study noted a range of 0.013ng/g-2.03ng/g Mercury and no detectable methylmercury in capsules and 0.233ng/g in cod liver oil.^[35] A study conducted in the US looking at three (unnamed) brands noted values of 9.89ng/g, 38.8ng/g, and 123ng/g in one salmon oil product.^[720]

A letter to the Editors in which independent testing was done^[721] mentioned that many popular fish oil products sold in North America have below 0.1mcg/g; TwinLab, Kyolic, Nature's Way, Natrol, Health from the Sun, and Nordic were cited in this letter.^[721]

Organochlorines and PCBs are at a minute level in supplementation, below the detection limit of many studies looking at them.^[722] Some studies do note detection, however, and tend to be by far highest in predatory oils like shark oil (usually supplemented for the squalene content).^[34]

Supplements made from cod, sardines and mackerel (non-predatory cold water fish) are the safest in terms of mercury level. Krill Oil is another option, if a lower dose is used.

Farm-raised salmon, mackerel, cod, trout and shrimp have slightly higher levels of omega-3 fatty acids and lower mercury. However, farm-raised salmon also tends to have significantly higher levels of many other contaminants.