Sources of ROS involved in atherogenesis include NADPH oxidases, nitric oxide synthases (NOS), lipoxygenases (LO), cyclooxygenases (COX), and the mitochondrial respiratory chain 27. Native LDL is modified by ROS generated by these enzymes in vascular tissues. NADPH oxidase is composed of a number of different subunits. There are seven homologues of the gp91phox (Nox-2) subunit. Nox-4 is predominantly expressed in ECs, though the expressions of Nox-1 and Nox-2 are also detected 28. During the respiratory burst in phagocytes, NADPH oxidase converts oxygen molecules to superoxide, which is a microbicidal oxygen metabolite. Then, superoxide is converted by superoxide dismutase (SOD) to hydrogen peroxide, which also kills microorganisms. ROS are also produced in ECs by endothelial NADPH oxidase 2627. Phagocytic NADPH oxidase and endothelial NADPH oxidase is one of the major ROS sources in the vasculature 27. Monocyte differentiation to macrophage is associated with the production and the release of ROS possibly through the induction of NADPH oxidase, resulting in further LDL oxidation 2930. NADPH oxidase generates superoxide on the extracellular side of the plasma membrane, and the enzyme can trigger intracellular signaling by superoxide transport via chloride channel-3 31. Stepp et al. 32 reported that native LDL and MM-LDL differentially increase vascular endothelial superoxide generation in canine carotid arteries, leading to vascular dysfunction and atherogenesis. Native LDL increases superoxide by an endothelial nitric oxide synthase (eNOS)-dependent mechanism whereas MM-LDL induces greater superoxide by the mechanisms dependent on eNOS, xanthine oxidase, and NADPH oxidase. Superoxide production by vascular tissues and its interaction with nitric oxide (NO) play important roles in vascular pathophysiology. Superoxide reacts rapidly with NO, reducing NO bioavailability and producing the oxidative peroxynitrite radical 33. Endothelial activation via LOX-1 produces additional ROS, generating a positive feedback loop for further LDL oxidation 34. ROS generated in a NADPH oxidase-dependent pathway mediate TNF-α-induced MCP-1 expression in ECs, and the induction of MCP-1 expression is suppressed by the antioxidant enzymes, SOD and catalase 35. Since phagocytic NADPH oxidase is the first line of the host defense system, selective suppression of vascular NADPH oxidase in local inflammatory lesions might be one of the therapeutic strategies 36.

Cyclooxygenases (COX-1 or COX-2) and lipoxygenases (5-, 12-, or 15-LO) also contribute to ROS generation during arachidonic acid (AA) metabolism shown in Figure 32737. The initial products in AA metabolism are highly reactive peroxides. Overexpression or induction of COX-2 increases ROS in certain cell types 3839 and the effects of overexpression of COX-2 are cell-/tissue-specific 40. Constitutive COX-1 and inducible COX-2 catalyze the conversion of free PUFAs to prostanoids (prostaglandins and thromboxanes), while LO generates the leukotrienes. Prostanoids and leukotrienes comprise a large and complex family of biologically active lipids derived from PUFAs by insertion of molecular oxygen. Collectively, these compounds are termed eicosanoids. Both prostanoids and leukotrienes play important roles in inflammation. Non-esterified PUFA released from the sn-2 position (middle carbon of glycerol) of the membrane phospholipids by the action of specific phospholipases are substrates for COX, LO, or cytochrome P450 monooxygenases (CYP) 4142. The metabolism of AA has two main pathways: the cyclic pathway leading to prostanoid formation and the linear pathway resulting in leukotriene formation. Two molecules of oxygen are added in the first step for the generation of prostaglandin G (PGG). The second step, via the peroxidase activity of COX, converts PGG2 into PGH2, the precursor for either thromboxanes or other prostaglandins. The peroxidase step is inhibited by aspirin or ibuprofen. LO is also a major source of extracellular superoxide release in a certain cell type during AA metabolism 37. The LO pathway is responsible for the formation of leukotrienes and hydroxyeicosatetraenoic acids (HETEs). LO isoforms act upon arachidonic acid to form 5-, 8-,12-, or 15-hydroperoxy eicosatetraenoic acids (HPETEs), which are unstable and can be reduced to the hydroxyl derivatives (HETE) in vivo. The range of HPETEs with biological activity is known as the leukotrienes. Both COX and LO products diffuse from cells and act locally at nanomolar levels on cell surface receptors linked to G-proteins. Activation of G-protein-associated receptors leads to changes in intracellular cAMP or calcium, which serve as second messengers that activate signaling mechanisms influencing various cellular functions. The COX products are modulators of thromboregulatory and chemotaxic responses, and inflammation. The LO products are involved in vascular permeability, vasoconstriction, and bronchoconstriction. The third route for eicosanoid production is via the CYP, in particular CYP4 family 43, including epoxy derivatives of 20:4 ω-6 that can modulate calcium signaling, channel activity, transporter function, and mitosis. This mechanism seems to be more consequential in cells when COX and LO activities are minimal.

This body of research has verified that multiple enzymes are involved in ROS generation that leads to atherogenesis. Hence, it is speculated that suppressing ROS generation may be a therapeutic target for preventing and alleviating atherosclerosis.

ROS play a role in LDL oxidation. LDLs are rich in polyunsaturated fatty acids (PUFAs), which are susceptible to oxidation 12. Lipid peroxidation can commence by ROS and other mechanisms that result in abstraction of an electron from a PUFA 44. The sequential carbon-centered radical undergoes rearrangement, and in the presence of oxygen, will add oxygen to form a peroxyl radical (ROO•). Propagation of the free radical reaction can occur by reaction of the peroxyl radical with another PUFA, generating the corresponding fatty acid hydroperoxide (ROOH) and another carbon-centered radical. Other factors, such as Fe⁺², and other oxidants can result in an amplification of the free radical process. Vitamin E is a nonenzymatic chain-breaking scavenger of lipid radicals generated in cell membranes; it protects against further lipid peroxidation. Vitamin C is an important antioxidant against lipid peroxidation because it has a high reactivity with the oxygen-centered radical. Oxygen-centered radicals have sufficient polarity to be accessible to the aqueous soluble vitamin C. Also, when vitamin E reacts with a radical, vitamin E is converted to its radical form which can be recycled to reduced vitamin E by reacting with vitamin C. Other reducing compounds such as glutathione and NADPH act in concert with vitamins E and C in an antioxidant cascade.

LDL oxidation also results in changes in apolipoprotein B epitope. The oxidized apolipoprotein portion of LDL is subsequently recognized and internalized by SR-A, whereas the oxidized lipid moiety of LDL is bound to CD36 on macrophages 45.

The relationship between fatty acids and atherosclerosis and other inflammatory diseases has been suggested by epidemiological, clinical, and in-/ex-vivo studies. Increased intake of saturated fatty acids is positively associated with development of atherosclerosis and inflammation. In contrast, omega-3 (ω-3) fatty acids, such as eicosapentaenoic acid (EPA, C20:5) and docosahexaenoic acid (DHA, C22:6), have shown protective effects against CVD. EPA and DHA are major components of dietary fish and fish oils. Like EPA and DHA, CLA isomers, exhibit protective effects against atherogenesis 4647 and inflammatory bowel disease (IBD) 54849 and antioxidant effects 50 in in-/ex-vivo studies; however, clinical studies have been inconclusive.

Most of the fatty acids synthesized or ingested have one of two fates: incorporation into triglycerides for the storage of metabolic energy or incorporation into the phospholipid components of membranes. The selection between the alternative fates depends on the need (i.e. growth and starvation) 42. Fatty acids may differentially affect inflammatory processes and ultimately the etiology of atherosclerosis in three ways as:

1) the components of membrane; fatty acids may serve as precursors of pro- or anti- inflammatory eicosanoids;

2) the components of LDL; fatty acids may differentially modulate recognition of macrophage receptors and subsequent inflammatory processes and atherogenesis; and

3) the regulators of gene expression; fatty acids may differentially regulate inflammatory gene expression, serving as ligands for transcription factors (e.g. PPARs).

Almost all mammalian cells, except red blood cells, produce eicosanoids, which play a role in inflammation. Arachidonic acid (AA, C22:4) is the most important precursor of eicosanoid, and AA is synthesized from linoleic acid (LA, C18:2, ω-6) by enlongation and desaturation. In response to hormonal or other stimuli, phospholipase A2, present in most types of mammalian cells, attacks membrane phospholipids, releasing AA from the middle carbon of glycerol. Enzymes of the smooth endoplasmic reticulum then convert AA into eicosanoids, potent biological signaling molecules 42.

LA (18:2, ω-6) and α-linolenic acid (ALA, 18:3, ω-3) serve as the precursors for longer-chain ω-6 (e.g., arachidonic acid: AA) and ω-3 fatty acids (e.g., EPA, DHA), respectively. Neither ω-3 nor ω-6 fatty acids can be synthesized in mammals due to the lack of certain types of desaturases. ω-3 Fatty acids cannot be generated from ω-6 fatty acids in mammals. Hence, the source of these PUFAs is limited to dietary intake 51. Dietary EPA, DHA, and CLA can partially replace AA derived from LA in the cell membrane 5253. Usually the plentiful LA may exclude these fatty acids from incorporation into membrane phospholipids 54 and/or LDL. However, EPA, DHA and CLA may influence eicosanoid production from AA and subsequent immune and inflammatory processes. For example EPA and DHA decrease the synthesis of pro-inflammatory eicosanoids, such as leukotriene-4 and prostaglandin-2 by replacing AA in phospholipid bilayers and by inhibiting cycloxygenase activity 555657. As components of LDL and/or membrane, these fatty acids may affect the inflammatory gene expression by altering signaling pathways. The unusual conformation structures (kinks) in unsaturated fatty acids interfere with the membrane motion 42 and possibly signal transduction. In addition, EPA and DHA reduce the expression of adhesion molecules induced by oxidized LDL in endothelial cells 585960. ω-3 PUFAs suppress inflammatory gene expression by inhibiting TLR4 signaling pathway, whereas saturated fatty acids exhibit the opposite effect 61, Thus, subsequently decreasing risk for CVDs.

Several epidemiological studies have reported an inverse relationship between intake of vegetables and fruits (in particular those rich in antioxidant vitamins including vitamins C and E and β-carotene), and risk for CVD 62636465. The protective effects of these antioxidant vitamins on atherosclerosis have been intensively investigated in animal and human studies. According to the oxidative modification hypothesis, oxidized LDL is immunogenic and atherogenic and LDL oxidation triggers atherosclerotic processes. Therefore, the protection of LDL from oxidation may be crucial to the prevention of atherosclerosis; the antioxidant components of LDL may prevent LDL oxidation.

Vitamin E is the generic term for all tocopherol and tocotrienol derivatives that exhibit the biological activity of α-tocopherol. There are eight naturally occurring isoforms synthesized in plants. α-Tocopherol is the most biologically and chemically active form of vitamin E. The hydroxyl groups at the C-6 position of tocopherols enable them to scavenge free radicals and superoxide. Although γ-tocopherol is predominant in the American diet, its plasma levels are only 10% of plasma α-tocopehrol levels (about 25 μmol/L α-tocopherol) 66.

α-Tocopherol is the major antioxidant in LDL and one LDL particle contains approximately six molecules of α-tocopherol. α-Tocopherol in LDL plays a role in preventing LDL oxidation. Vitamin E depletion in LDL may trigger LDL oxidation; and the addition of micromolar concentrations of vitamin E inhibits LDL oxidation. All other antioxidants, such as γ-tocopherol, carotinoids, and ubiquinol-10, are present in much smaller amounts than α-tocopherol. In contrast to α-tocopherol, carotenoids play only a minor or no role in LDL protection 67. However, many clinical studies have failed to demonstrate the protective effects of vitamin E. One explanation may be that vitamin E exhibits prooxidant activity in the absence of co-antioxidant compounds capable of reducing the tocopherol radical 6869. A similar situation may occur with other antioxidants, such as β-carotene 707172. Depending on the concentrations, environmental conditions and presence of oxygen or other oxidants, compounds with antioxidant properties may exhibit prooxidant or other non-antioxidant properties.

Polyphenolic compounds, such as resveratrol and catechins, are derived from plants, and the compounds have shown anti-atherogenic and anti-inflammatory effects. The beneficial effects of the compounds are attributed to their abilities to function as antioxidants by: 1) inhibition of prooxidant enzymes, such as lipoxygenases, cyclooxygenases, and xanthine oxidase, possibly through suppressing the activation of redox-sensitive transcription factors, NF-κB and activator protein-1 (AP-1), and 2) induction of antioxidant enzymes such as glutathione S-transferase, glutathione peroxidase (Gpx), superoxide dismutase (SOD), and catalase 7374.

Antioxidant enzymes are involved in the maintenance of intracellular and extracellular reducing reactions 75 and suppress the generation of free radicals as the first line of antioxidant defense 12. Antioxidant enzymes include superoxide dismutase (SOD) and catalase. SOD is expressed in most cell types, and converts harmful superoxide to less harmful hydrogen peroxide and oxygen. Catalase catalyzes the dismutation of hydrogen peroxide to oxygen and water. Catalase has an iron redox center. Catalase is located predominantly within peroxisomes to protect from hydrogen peroxide generated during fatty acid β-oxidation within the cellular organelles 76.

The antioxidant enzymes play a role in preventing atherogenesis. Increased expression of GR in macrophages reduces atherosclerotic lesion formation in LDL receptor-deficient mice 77. Over-expression and/or induction of CuZn-SOD and catalase can be beneficial because of: 1) decreases in superoxide levels in ECs; 2) suppression of oxidative stress, e.g., age related; 3) protection against inflammatory events by inhibiting NF-κB activation; and 4) suppression of low-density lipoprotein (LDL) oxidation by ECs 507678798081.

NF-κB is a redox-sensitive transcription factor expressed in all cell types; it recognizes and binds to specific DNA sequences (5'-GGGRNNYYCC-3'). NF-κB activation is triggered by the IκB kinase (IKK)-mediated degradation of inhibitor κB (IκB), which regulates NF-κB. NF-κB is activated by intra-/extra-cellular ROS and/or ROS-modified target biomolecules, and is involved in regulating immune and inflammatory responses. NF-κB-mediated target genes include: inflammatory cytokines (e.g., TNF-a, IL-1, IL-2, M-CSF), chemokines (e.g., MCP-1), adhesion molecules (e.g., ICAM-1, VCAM-1), inflammatory enzymes (e.g., iNOS, COX-2), and apoptotic regulators (e.g., Fas ligand, Fas, p53) 85.

Oxidized LDL may affect atherogenesis in part via the NF-κB activation pathway. Oxidized LDL activates NF-κB as well as C. pneumoniae 86. Resveratrol, an antioxidant polyphenol derived from plants, attenuates TNF-α-induced inflammatory gene expression and monocyte adhesion to human coronary arterial endothelial cells (HCAECs) by inhibiting NF-κB activation, suggesting that the anti-inflammatory actions of resveratrol are responsible for anti-atherogenic effects 78. Oxidized LDL exerts biphasic effects on NF-κB: 1) inflammatory effects by up-regulating inflammatory gene expression via NF-κB activation at lower concentrations of oxidized LDL; and 2) immunosuppressive effects by inhibiting NF-κB activation triggered by inflammatory agents such as lipopolysaccharide (LPS) at higher concentrations of oxidized LDL 85. HUVECs incubated with LPS which causes inflammatory gene expression via TLR4 activation, induce the expressions of TLR4, LOX-1, ICAM-I, and E-selectin, and increase monocyte adhesion to endothelium and NF-κB activation levels, suggesting the atherogenic process is mediated through TLR4/NF-kB pathways 87. There are two types of TLR4/NF-κB pathways identified: MyD88-dependent and independent pathways. MyD88 is a common downstream adaptor molecule for most TLRs, and recruits other molecules required to activate NF-κB. Saturated fatty acids trigger TLR4 and downstream NF-κB activations, resulting in inflammatory gene expression (i.e. COX2 or iNOS). In contrast, unsaturated fatty acids inhibit TLR4/NF-κB activation. This inhibition may be due to the alteration of fatty acid components in membrane lipid rafts, which may lead to the disruption of the recruitment of the downstream signaling components 61.

PPARs (PPARα, PPARβ, and PPARγ), a group of nuclear receptors, belong to the steroid hormone receptor superfamily 88. PPARs heterodimerize with the 9-cis retinoic acid receptor (RXR) and bind to peroxisome proliferator response elements (PPREs: 5'-

AGGTCA

AGGTCA

PPARγ activation may be involved in oxidized LDL-induced inflammatory gene expression and macrophage lipid metabolism 459091. Although CD36 is up-regulated by oxidized LDL via PPARγ activation, PPARγ activation suppresses oxidized LDL-induced inflammatory effects by inhibiting inflammatory gene expression. Kunsch and Medforld 82 suggest that PPARγ participates in a positive feedback loop and that alternative or downstream pathways may trigger PPARγ activation resulting in anti-inflammatory effects. For example, 15d-PGJ2 is known to be an endogenous PPARγ activator. 15d-PGJ2 may be a possible anti-inflammatory mediator, though the physiological levels of 15d-PGJ2 may be insufficient to modulate PPARγ activation 92. PPAR activators are negative regulators of macrophage activation and antagonize the activities of the transcription factors, AP-1, STAT, and NF-κB, involved in inflammatory gene expression 9394.

Lipoprotein lipase (LPL), a lipolytic enzyme, may play an important role in regulating early atherogenesis. LPL neither acts on nor binds to oxidized LDL. LDL(-) is a form of native LDL containing intermediately modified subfractions with higher electronegative charge and is taken up by LDL receptors. Lipid peroxidation is greater in LDL(-) than in native LDL; LDL(-) exhibits inflammatory effects. Ziouzenkova et al. 95 demonstrated that LPL-treated LDL(-) reduced inflammatory gene expression in human ECs by suppressing NF-κB and AP-1 activations and by increasing the expression of IκB, a target gene for PPARα, via PPARα activation. In contrast, LDL(-) alone increased the inflammatory responses. 9- and 13-HODEs, both known as PPARγ activators, are released during the hydrolysis of both native LDL and LDL(-), resulting in PPARα activation and anti-inflammatory effects.

Expression of antioxidant enzymes, Cu/Zn SOD and catalase, may be modulated through both PPARγ and NF-κB activations. Possible multiple binding sites for PPARγ and NF-κB have been identified: 1) within the promoter region of SOD1, and 2) one binding site for NF-κB within the promoter region of catalase. NF-κB activation is associated with the induction of proinflammatory gene expression. Although Cu/Zn SOD is an antioxidant enzyme, it is induced to convert superoxide to hydrogen peroxide that is still microbicidal; it also serves as a host defense with NADPH oxidase in phagocytes. Furthermore, the treatments of possible PPARγ activators increase both PPARγ and NF-κB DNA binding activities, indicating that these two redox-sensitive transcription factors coordinate and propagate feedback loops between each other.

CLA is being sold as a panacea with several alleged benefits including altering body composition, i.e., to reducing obesity and building lean body mass 132. Safety of long-term (≥ 12 months) CLA supplementation was examined in several clinical trials. A randomized, double-blind study was conducted, in which obese individuals were given 6 g/day of either CLA isomer mixture (cis-9, trans-11: trans-10, cis-12 = 50:50) or placebo (high oleic sunflower oil) for 12 months. Although body composition did not differ between the CLA-supplemented group (n = 27) or the placebo group (n = 23), lower levels of adverse effects (alterations in the liver function, glucose and insulin levels, insulin resistance, and white blood cell counts) were observed in the CLA-supplemented group than in the control group. The investigators concluded that CLA isomer mixture as Clarinol™ is safe for use in obese humans for up to one year at the recommended dosage 133. Another long-term (one year) CLA isomer mixture supplementation study was performed in a double-blind fashion 134. Healthy overweight humans (n = 180) were randomly divided into three groups: 1) CLA free fatty acid (cis-9,trans-11: trans-10, cis-12 = 50:50; 3.6 g CLA isomers/day as FFA forms), 2) CLA-triacylglycerol group (cis-9, trans-11: trans-10, cis-12 = 50:50; 3.4 g CLA isomers/day as TAG forms), and 3) placebo (olive oil). The CLA isomer mixture supplementation decreased body fat mass in healthy overweight adult humans. However, there were significant increases in: LDL levels in the CLA-FFA group, HDL levels in the CLA-TAG group, and lipoprotein levels in both CLA groups. Adverse effects, mostly gastrointestinal, were reported by 11.4% of the subjects, and likely resulted from the daily ingestion of oil or of the gelatin capsule alone. Overall, the adverse effects did not differ significantly between the CLA groups and the placebo group, indicating that CLA isomer mixture was tolerated as well as olive oil as the control. One hundred twenty five of 180 subjects who finished this study, continuously participated in the CLA isomer mixture supplementation study for an additional year, thus, total 2 years 135. Two-year-CLA isomer mixture supplementation groups significantly reduced body weight, BMI, body fat mass, energy intake and serum leptin levels, compared with the baselines at month 0. However, serum lipoprotein and aspartate amino transferase levels, and whole blood leukocyte and thrombocyte counts were significantly increased in the CLA groups. Gaullier et al. concluded that CLA isomer mixture supplementation for 24 months in healthy, overweight adults was well-tolerated, and that CLA isomer mixture may be beneficial as a weight loss supplement. Another one-year CLA isomer mixture supplementation study 136 was conducted in a randomized, double-blind, placebo-controlled fashion. No significant differences in body weight or body fat regain were observed between the CLA group (cis-9, trans-11: trans-10, cis-12 = 50:50; 3.4 g/day as TAG forms; n = 40) and placebo (4.5 g olive oil; n = 43). No significant differences in adverse effects or indexes of insulin resistance were observed between the groups. However, a significant increase in the number of leukocytes was observed in the CLA group. Although the investigators did not obtain a perfect group match for body weight at randomization, they concluded that the CLA isomer mixture supplementation for one year has no preventative effect on body weight and body fat regain after the weight loss induced by a low calorie diet for eight weeks in obese subjects.

Many studies have investigated the effects of short-term (mostly 12 weeks or 8 weeks) CLA supplementation. In a six-month double-blind CLA isomer mixture supplementation study 137, 118 healthy overweight and obese adult humans were randomized into two groups supplemented with either 3.4 g/day CLA isomer mixture (cis-9, trans-11: trans-10, cis-12 = 50:50) or placebo. CLA significantly decreased body fat mass, in particular in legs of both males and females and in females with BMI >30 kg/m², at either month 3 or 6, compared with placebo. Lean body mass increased in the CLA supplemented group. The safety parameters including blood lipids, inflammatory and diabetogenic markers remained within the normal range, and adverse events did not differ between the groups in the study. It was concluded that the CLA isomer mixture supplementation in healthy, overweight, and obese subjects decreases body fat mass in specific regions and was well tolerated. The dose-dependent effects of CLA were reported in a CLA isomer mixture supplementation study of 12 weeks 138. Forty eight obese subjects were divided into three groups: 1) 3.2 g/day CLA (cis-9, trans-11: trans-10, cis-12 = 50:50), 2) 6.4 g/day CLA, and 3) placebo (8 g safflower oil). CLA isomer mixture supplementation at the higher dose increased inflammatory markers, IL-6 and C-reactive protein (CRP), however, remained within normal ranges. A significant increase in lean body mass was also found in the same treatment group. No severe adverse effects were reported. The authors concluded that the CLA isomer mixture intervention was well-tolerated. Beneficial effects on immune functions have been reported in a double-blind, randomized CLA isomer mixture supplementation study 139. Twenty-eight healthy adults received either high oleic sunflower oil (placebo) or 3.0 g/day CLA (cis-9, trans-11: trans-10, cis-12 = 50:50; triglyceride form) for 12 weeks. The CLA group showed significantly reduced levels of the proinflammatory cytokines, TNF-α and IL-β, and increased levels of the anti-inflammatory cytokine, IL-10. Immunoglobulin levels were also altered: CLA isomer mixture decreased Ig E levels, and increased both Ig M and Ig A levels. Another CLA isomer mixture supplementation study (2.2 g/day; cis-9, trans-11: trans-10, cis-12 = 50:50; 8 weeks) investigated the effects of CLA isomer mixture on inflammation in a double-blind, randomized, placebo-controlled model using healthy middle-aged males 140. The CLA isomer mixture supplementation significantly reduced concanavalin A-stimulated peripheral blood mononuclear cell IL-2 secretion, suggesting anti-inflammatory and anti-atherogenic effects of CLA isomer mixture. Other inflammatory markers, IL-6, CRP, and fibrinogen, were not affected in this study. Moloney et al. 141 demonstrated that the CLA isomer mixture supplementation (3.0 g/day; cis-9, trans-11: trans-10, cis-12 = 50:50; 8 weeks) increased total HDL cholesterol concentrations and decreased the ratio of LDL cholesterol to HDL cholesterol without changes in inflammatory markers of CVD in subjects with type 2 diabetes. However, this CLA isomer mixture intervention did not show positive effects on insulin and glucose concentrations among the diabetic patients. In a Swedish study, 53 healthy humans were randomly assigned to CLA isomer mixture supplementation (4.2 g/day; cis-9, trans-11: trans-10, cis-12 = 50:50; 12 weeks) in a double-blind fashion. Supplementation with a CLA isomer mixture reduced the proportion of body fat and affected fatty acid metabolism. However, no effects were found for CLA isomer mixture on body weight, serum lipids, glucose metabolism or plasminogen activator inhibitor 1 142. In Noone et al.'s double-blind placebo-controlled study 143, the CLA isomer mixture treatment (3 g/day; cis-9, trans-11:trans-10, cis-12-CLA = 50:50; 8 weeks) showed reduced plasma triacylglycerol levels in normolipaemic human subjects.

A study using 49 healthy male subjects showed the isomer-/dose-dependent (0.59, 1.19, 2.38 g/day of the cis-9, trans-11 CLA isomer; 0.63, 1.26, 2.52 g/day of the trans-10, cis-12 CLA isomer; 8 weeks) opposite effects of CLA on plasma total cholesterol and LDL-cholesterol levels: hypolipidemic properties of the cis-9, trans-11 CLA isomer and hyperlipidemic properties of the trans-10, cis-12 CLA isomer. However, neither CLA isomer supplementation affected insulin resistance 144. Using the same healthy male subjects and the same supplementation design, Tricon et al. investigated the effects of two CLA isomers on immune cell functions 145. The results showed a dose-dependent reduction in the mitogen-induced activation of T lymphocytes and a negative relationship between the mitogen-induced T lymphocyte activation and the contents of each CLA isomer in mononuclear cells, suggesting beneficial effects in inflammatory diseases such as atherosclerosis.

Additional adverse effects of CLA supplementation, in particular trans-10, cis-12 CLA, were reported. Riserus et al. demonstrated that the purified trans-10, cis-12 CLA isomer supplementation (3.4 g/day, 3 months), but not CLA mixture supplementation (3.4 g/day; cis-9, trans-11: trans-10, cis-12 = 50:50; FFA form, 3 months), increased oxidative stress, CRP, and proinsulin levels, and decreased insulin sensitivity in non-diabetic abdominally obese males 146 and in males with metabolic syndrome 147 in two double-blind, randomized, placebo- (3.4 g/day, olive oil) controlled studies. Like the study by Tricon et al. 144, Riserus et al.'s two studies also suggest the isomer-dependent detrimental effects of CLA. Another unfavorable effect of the trans-10, cis-12 CLA isomer was also reported in a human study examining non-enzymatic and enzymatic lipid peroxidation (8-iso-PGF_2α and 15-keto-dihydro-PGF_2α, respectively) in human plasma and urine. Sixty healthy subjects were divided into six groups: three CLA isomer mixture groups, (3.5 g/day, cis-9, trans-11: trans-10, cis-12 = 50:50, 4 weeks) and three trans-10, cis-12 CLA isomer groups (4.0 g of the trans-10, cis-12 CLA isomer/day, 4 weeks): 1) the CLA supplement alone, 2) with vitamin E (D-α-tocopherol acetate), and 3) with COX-2 inhibitor (refecoxib). Although both CLA isomer mixture and the trans-10, cis-12 CLA isomer supplementations increased the eicosanoid levels in the urine, the trans-10, cis-12 CLA isomer supplementation with the COX-2 inhibitor suppressed the increase in urinary 15-keto-dihydro-PGF_2α levels. This result suggests that increased lipid peroxidation in eicosanoid synthesis may be due to induced COX-2 expression by the CLA supplementations, in particular the trans-10, cis-12 CLA isomer 148. Taylor et al. 149 documented that CLA isomer mixture supplementation (4.5 g/day; cis-9, trans-11: trans-10, cis-12 = 50:50; 12 weeks) impaired endothelial function and increased markers of oxidative stress in 40 healthy white males, suggesting caution in the use of CLA isomers as an aid for weight loss.

Overall, the effects of CLA isomers (or mixture) on atherogenic and/or inflammatory parameters in humans have not been definitive. Although a meta-analysis of 18 CLA human studies (including three single isomer studies) suggests the beneficial use of CLA isomers only as a body fat reducing supplement 150, the therapeutic potentials of CLA isomers in inflammatory diseases including atherosclerosis remain to be determined. Differences in purity and content of CLA isomers may cause these conflicting results. Impurities might induce undesirable side effects 151. The variety of CLA isomer content in supplements and/or the differences in CLA dose might cause inconsistent results due to the dose- and/or isomer-dependent effects of CLA suggested by some other studies 138144146147. Human subjects were not limited in diet (therefore, dietary fat intake, excluding supplemental fat intake, is of concern) and/or physical activities in some study designs. In addition, CLA isomer (or mixture) supplementations contained other fatty acids, including PUFAs and saturated fatty acids, at up to 20% in some study designs. The details of supplemental contents, other than CLA isomers, are not even provided in some other studies. Since CLA isomers are incorporated into membrane phospholipids, they may compete in enlongation and desaturation steps with other PUFAs that are precursors of arachidonic acid (AA, 20:3, ω-6). The competition in the incorporation may alter eicosanoid biosynthesis 152, therefore, subsequent immune and inflammatory processes. An experiment by Brown et al. 153 found that the presence of linoleic acid (LA) may affect the possible benefits of CLA isomers. LA, one of the PUFAs present in Western diets and the human body is a possible antagonist to CLA isomers. The plentiful LA may exclude CLA isomers from incorporation into phospholipids and drive it into storage as a component of neutral lipid 52. LA also inhibits EPA incorporation in membrane phospholipids from fish-oil supplements 54. Like CLA isomers, dietary EPA and DHA partially replace AA derived from LA in the cell membrane. These ω-3 fatty acids are associated with decreased risk for CVD, and reduce the formation of pro-inflammatory eicosanoids 53. The replacement of AA by ω-3 fatty acids may cause the alteration of fatty acid composition in membrane lipid bilayers influencing signaling pathways and subsequent immune and inflammatory processes. The anti-inflammatory effects of CLA isomers observed in a limited number of studies may be attributed to the replacement of AA generated by LA. In addition, saturated fatty acids have been reported to provoke inflammation by inducing pro-inflammatory gene expression through innate immune receptor (TLRs) activation 61. Thus, the coexistence of other fatty acids might potentially affect the results of any human studies with CLA isomers. Moreover, there may be divergent effects of CLA isomers in obese or diabetic subjects compared to the normal-weight or healthy subjects as well as differences determined by gender and/or genetics, i.e., SNPs in related genes. Further studies are needed to investigate the effectiveness and safety of CLA supplementation and to elucidate these confounding factors.

Adiposity plays an important role in fatty acid mobilization, fat storage, and formation of pro-and anti-inflammatory cytokines and chemokines 128. As mentioned above, atherosclerosis is viewed as a chronic inflammatory disease affecting lipid profiles. CLA isomers have been shown to influence lipid metabolism associated with inflammation and atherogenesis in in vitro studies. One possible mechanism by which CLA isomers could modulate atherogenesis is regulating the production of lipoproteins in the liver. Sterol element binding proteins (SREBP) are a group of membrane-bound transcription factors that bind to their specific DNA binding sites (SRE-1) to activate the expression of target genes that encode enzymes necessary for lipid synthesis, including the LDL receptor (LDLR) gene in sterol-depleted cells 154. A study by Ringseis et al. 155 reported that the trans-10, cis-12 CLA isomer, not the cis-9, trans-11 CLA isomer, induced LDLR gene expression via SREBP-2 in human hepatoma cells (HepG2). They concluded that the enhanced uptake of VLDL and LDL cholesterols by hepatic LDLR may account for the decreased plasma cholesterol levels in response to CLA isomer (or mixture) supplementations in a limited number of human and animal studies. An alternative pathway for CLA-mediated LDLR expression is also suggested. Yu-Poth et al. 156 demonstrated that a CLA isomer mixture (50:50, 400 μmol/L final concentration) up-regulated LDL receptor (LDLR) mRNA and protein expression at three- to five-fold in HepG2 cells. The results of the study suggest the upregulation of the LDLR gene by CLA through a mechanism that is independent of SREBP-1 and acyl CoA: cholesterol acyltransferase (ACAT).

Monocyte-endothalial interaction is a key step of atherogenesis. However, CLA isomers showed no effects on TNFα-induced adhesion molecule expression, monocyte adhesion, and chemokine release or on the molecular mechanisms regulating these processes in human aortic endothelial cells 21. This suggests that the anti-atherogenic effects of CLA isomers may not be associated with the reduction of monocyte-endothelial interactions.

Several studies have investigated the implication of CLA isomers in eicosanoid synthesis and the role of CLA isomers in inflammation and atherogenesis. Dietary CLA may suppress the biosynthetic pathway of AA. CLA may suppress eicosanoid formation via direct action on COX and LOX, i.e., by inhibiting the expression or the activities of these enzymes 56. Constitutive COX-1 and inducible COX-2 catalyze the conversion of free PUFAs to prostanoids, while LOX generates the leukotrienes. The trans-10, cis-12 CLA isomer suppresses COX-2 expression and PGE₂ release in rat macrophages either by inhibiting NF-κB activation in vivo and in vitro or by inhibiting the MAPK/ERK/JNK pathway. The trans-10, cis-12 CLA isomer inhibits the NF-κB p50 and p65 subunits binding to DNA 83. The 50:50 mixture of CLA isomers inhibits the expression of both COX-2 and inducible nitric oxide synthase (iNOS) in LPS activated murine macrophages, resulting in decreases in prostaglandin E₂ and NO synthesis 157. The effects of CLA on prostanoid formation can be either inhibitory or stimulatory, depending on isomer-specificity, chemical forms of CLA isomers (i.e., free fatty acid or esterified forms), or cellular states (i.e., resting or stimulated states) in human endothelial cells and platelets 158. CLA isomers have also been reported to reduce prostaglandin E2 synthesis in certain cell types in both humans and mice 159.

The beneficial effects of CLA isomers may be attributed to their properties as PPARα/γ activators 548155160. Structurally, CLA resembles 13-HODE, as well as 15-HETE and 15d-PGJ2, which were all identified as natural activators of PPARγ 49. Ringseis et al160 demonstrated that either the cis-9, trans-11 or trans-10, cis-12 CLA isomer (50 μmol/L) reduced AA proportions in human vascular smooth muscle cells (SMCs), TNFα-induced NF-κB DNA binding activity, mRNA levels of enzymes involved in eicosanoid synthesis (e.g., COX-2), and production of PGE2 and PGI2. These CLA isomer treatments increased PPARγ DNA binding activity. Furthermore, a PPARγ repressor suppressed the inhibitory actions on the eicosanoid formation and NF-κB DNA binding activity in the SMCs. Synthetic PPAR activators exert their anti-inflammtory actions, at least in part, by negatively regulating NF-κB activation 157. PPARγ and NF-κB may be involved in regulating genes, whose products influence ROS generation that contributes to inflammation and atherogenesis, and these transcription factors coordinate and propagate feedback loops between each other. Thus, anti-inflammatory and anti-atherogenic effects of CLA isomers may be associated with these redox-sensitive transcription factors.

Although CLA isomers and ω-3 PUFAs have shown anti-inflammatory effects, they differ in the nature of their immunomodulatory properties. CLA isomers appear to enhance immune function, while ω-3 PUFAs are immunosuppressive 5. Tian et al.'s study 51 using rats treated with 300 mg/kg/day of clofibrate, a PPARα activator, for up to 14 days, reported an increase in myocardial DHA proportion and a decrease in the portion of AA that is a precursor of pro-inflammatory eicosanoids (e.g., leukotriene B4 and prostaglandin E2). Tian et al. implicate enhanced uptake of ω-3 PUFAs from blood circulation and/or increased biosynthesis of ω-3 PUFAs in rats treated with the PPARα activator, in those results. Similarly, Attar-Bashi et al.'s human study 161 investigated the effects of CLA isomers mixture [3.2 g/day, cis-9, trans-11: trans-10, cis-12 = 50:50, plus 11 g of alpha-linoleic acid (ALA), 8 weeks] as PPARα activators on DHA (22:6, ω-3) and EPA (20:5, ω-3) biosynthesis from ALA (18:3, ω-3) through Δ5- and Δ6-desaturases, both of which are possible PPARα target genes. The study demonstrated that ALA (18:3, ω-3) plus CLA isomer mixture increased EPA and decreased AA. However, the CLA isomer mixture supplementation did not affect DHA biosynthesis in humans. DHA synthesis from ALA needs additional peroxisomal oxidation. Thus, CLA isomers may play a role in PPARα-mediated gene expression and ω-3 PUFA-mediated anti-inflammatory effects.

CLA isomers are readily metabolized in vivo via multiple pathways, and enlongated and desaturated metabolites of CLA have been detected in the liver and mammary tissue of rats and adipose tissue and sera of humans 6. Some studies have suggested the involvement of CLA metabolites in anti-atherogenic and ant-inflammatory processes 5160, though the Δ6-desaturase metabolites of CLA may not be important for the alterations in gene expression induced by CLA 6.

Thus, multiple signaling pathways, such as PPARα, PPARγ, NF-κB, and MAPK/ERK/JNK, may be involved in the anti-inflammatory and anti-atherosclerotic effects of CLA isomers.