Although dietary fat has been associated with inflammation and cardiovascular diseases (CVD), most studies have focused on individuals with preexisting diseases. However, the role of dietary fatty acids on inflammatory pathways before the onset of any abnormality may be more relevant for identifying initiating factors and interventions for CVD prevention. We fed young male pigs one of three diets differing in n-6 and n-3 polyunsaturated fatty acids (PUFA) linoleic acid (LA, 18:2n-6) and α-linolenic acid (ALA, 18:3n-3) for 30 days. Cardiac membrane phospholipid fatty acids, phospholipase A2 (PLA2) isoform activities, and cyclooxygenase (COX)-1 and -2 and 5-lipoxygenase (5-LO) expression were measured. The low PUFA diet (% energy, 1.2% LA+0.06% ALA) increased arachidonic acid (AA) and decreased eicosapentaenoic acid (EPA) in heart membranes and increased Ca2+-independent iPLA2 activity, COX-2 expression, and activation of 5-LO. Increasing dietary ALA while keeping LA constant (1.4% LA+1.2% ALA) decreased the heart membrane AA, increased EPA, and prevented proinflammatory enzyme activation. However, regardless of high ALA, high dietary LA (11.6% LA and 1.2% ALA) decreased EPA and led to a high heart membrane AA, and Ca2+-dependent cPLA2 with a marked increase in nitrosative stress. Our results suggest that the potential cardiovascular benefit of ALA is achieved only when dietary LA is reduced concomitantly rather than fed with high LA diet. The increased nitrosative stress in the unstressed heart with high dietary LA suggests that biomarkers of nitrosative stress may offer a useful early marker of the effects of dietary fat on oxidative tissue stress.
- oxidative stress
- arachidonic acid
a growing body of information is combining to support the concept that chronic low-grade systemic inflammation is an important predisposing risk factor contributing to cardiovascular diseases (CVD) such as hypertension, coronary artery disease, and heart failure (46). Although high dietary fat intakes are believed to promote increased synthesis and release of proinflammatory eicosanoids in the heart (55), skeletal muscle, liver, and macrophages (5) in both humans and animals (53), dietary fatty acids are becoming increasingly recognized as important in modulating disease risk (8, 55). For example, some populations such as those consuming a Mediterranean-style diet high in monounsaturated fatty acids but low in n-6 polyunsaturated fatty acids (PUFA) have a lower CVD morbidity and mortality than populations consuming a Western diet, which is higher in saturated and n-6 PUFA but low in n-3 PUFA (14).
Phospholipase A2 (PLA2) is a multienzyme family that plays a pivotal role in several inflammatory processes because of their role in releasing free fatty acids (FFA) from the sn-2 position of membrane phospholipids (56). Arachidonic acid (AA, 20:4n-6) is esterified at the sn-2 position of phospholipids and after release by PLA2 is further metabolized by cyclooxygenases, lipoxygenases, and monooxygenases to different eicosanoids, many of which are potent proinflammatory mediators (3, 56). Indeed, PLA2 is often rate-limiting for eicosanoid production (22). Mammalian PLA2 is classified into cytosolic Ca2+-dependent PLA2 (cPLA2), Ca2+-independent PLA2 (iPLA2), and secretory PLA2 (sPLA2) families, with several isoforms of each (9). Although high-fat diets have been associated with cardiac inflammation (55), the effect of the dietary fatty acid composition in activating specific PLA2 isoforms in the nondiseased myocardium is unknown. Moreover, most studies addressing the effects of fatty acid-mediated regulation of PLA2 activity have been conducted in vitro with single fatty acids and under acute conditions (12, 48, 51).
Among multiple initiators of proinflammatory eicosanoid signaling, cyclooxygenase-1 and -2 (COX-1 and COX-2, respectively) and 5-lipoxygenase (5-LO) are key players (16, 34). COX-1 is constitutively expressed and is responsible for the “physiological” levels of prostaglandins. Although COX-2 is induced primarily by proinflammatory stimuli, COX-1 and COX-2 activity leads to the same metabolic intermediates with the formation of prostanoids with both pro- and anti-inflammatory properties (44). 5-LO is a nonheme iron-containing enzyme that translocates from the cytosol to membrane upon activation. 5-LO converts unesterifed AA to 5-hydroperoxyeicosatetraenoic acid and then to leukotriene A4, a key intermediate in biosynthesis of all leukotrienes responsible for allergy and inflammation (54).
Over the last century, human diets have become increasingly enriched in the n-6 PUFA linoleic acid (LA, 18:2n-6), which is metabolized in many cells by desaturation and elongation to AA (52). A concurrent decrease in the intake of n-3 α-linolenic acid (ALA, 18:3n-3) has led to a dramatic shift in the balance of LA/ALA in Western diets from a traditional 2:1 to about 10:1. Desaturation and elongation of ALA leads to synthesis of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). However, because the same desaturase enzyme is involved in the synthesis of AA from LA and EPA from ALA metabolism, high dietary intakes of LA may interfere with the conversion of ALA to EPA (36). An increase in the dietary LA/ALA balance has occurred in parallel with the increase in CVD and other inflammatory diseases (21, 32). Although higher intakes of fish, the major dietary source of EPA and DHA, are associated with a decreased risk of CVD, the importance of dietary ALA in CVD development remains unclear (41). However, some observational and secondary intervention trials suggest cardioprotective effects of diets high in ALA (38, 41, 42). The objective of the present study was to determine the effect of the dietary LA and ALA intake and their ratio on the initiating pathways of cardiac inflammation in normal pigs.
MATERIALS AND METHODS
Experimental animals and diets.
Male Yorkshire pigs were fed one of three diets containing 50% energy from fat designed to resemble the usual fat content of sow milk and differing only in the composition of fatty acids (23). For reference, human milk also provides the young infant with ∼50% energy from fat (22) and also varies in LA and ALA depending on the maternal dietary intake of these fatty acids. Pigs were fed one of three diets from postnatal day 1 to 30 (n = 6–9 pigs/diet) that contained, as a percent energy, 1.2% LA and 0.06% ALA (low LA + ALA), 1.4% LA and 1.2% ALA (low LA + high ALA), or 11.6% LA and 1.2% ALA (high LA + ALA). Oleic acid (18:1n-9) varied inversely with LA at ∼18.2, 17.2, and 3.5% in the low LA + ALA, low LA + high ALA, and high LA + ALA diets, respectively. To elucidate the specific effects of the dietary essential fatty acids LA and ALA and their ratio on the heart levels of LA, ALA and their metabolites AA and EPA and DHA, respectively, all of the diets were all devoid of preformed AA, EPA, and DHA. Pigs were used as an appropriate species for these studies because pigs have low rates of desaturation of ALA to DHA, similar to that in humans (24), and share many other physiological and anatomical features relevant to CVD (28).
After 30 days feeding the assigned diet, the pigs were anesthetized with 37.5 mg/kg ketamine hydrochloride (MTC Pharmaceuticals, Cambridge, Canada) and 3.75 mg/kg xylazine hydrochloride (Bayvet Division, Chenango, Etobicoke, Canada) and then killed by intracardiac injection of 200 mg/kg pentobarbital sodium. The hearts were isolated, flash-frozen in liquid nitrogen, and stored at −80°C for later lipid and biochemical analysis. All protocols and procedures involving the animals were approved by the University of British Columbia Animal Care Committee and conformed to the guidelines of the Canadian Council of Animal Care.
Total cardiac lipids were extracted and solubilized in degassed chloroform-methanol-acetone-hexane (4:6:1:1 vol/vol/vol/vol). Subsequently, polar and nonpolar lipids, including phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylserine (PS), phosphatidylinositol (PI), lysophosphatidylcholine (LPC), and FFA, were separated using HPLC (2690 Alliance HPLC; Waters, Milford MA), quantified with an evaporative light-scattering detector, and recovered using a fraction collector (26). Fatty acids in the recovered lipids were separated and quantified as their respective methyl esters using heptadecanoic acid (17:0) as the internal standard with a Varian 3400 GLC equipped with a flame ionization detector, a Varian Star data system, and a SP-2330 capillary column (30 m × 0.25 mm internal diameter; Supelco, Bellefonte, PA). Separated lipid classes are expressed relative to tissue protein, and individual fatty acids are expressed as a weight percentage of the total fatty acids in each lipid.
Separation of cytosolic and particulate fractions.
Membrane and cytosol fractions of heart ventricles were prepared based on the method of Morabito et al. (40). Briefly, ventricles were homogenized (20 mM Tris base, 2 mM EDTA, 10 mM EGTA, 250 mM sucrose, 10 mM DTT, and 55 μM leupeptin, pH 7.45) and centrifuged (100,000 g, 1 h), and the resulting supernatant was collected as the cytosolic fraction. The pellet was resuspended in 200 μl buffer containing 1% (vol/vol) Triton X-100, incubated at room temperature for 30 min, and then diluted with buffer to a final concentration of 0.2% Triton X-100. The homogenate was centrifuged (100,000 g, 1 h), and the solubilized membrane proteins were collected. The purity of the fractions was estimated by immunoblotting for manganese superoxide dismutase (MnSOD), a mitochondrial protein (anti-sheep MnSOD; Calbiochem, San Diego, CA), and copper/zinc SOD (Cu/Zn SOD), a cytosolic protein (anti-sheep Cu/Zn SOD; Calbiochem) as described previously (20). No detectable signals for MnSOD or Cu/Zn SOD were present in the cytosolic or particulate fractions, respectively. Protein was quantified using a Bradford protein assay, with BSA as a standard.
Heart tissue homogenates or membrane and cystosolic fractions were diluted and boiled with sample loading dye; samples corresponding to 50 μg protein were used in SDS-PAGE, transferred to nitrocellulose membranes, and blocked in 5% skim milk overnight in Tris-buffered saline containing 0.1% Tween 20 (TBS-T). Membranes were incubated with antibodies raised in either rabbit [COX-1 (BioVision, Mountain View, CA) and -2 (Cayman Chemical, 4, MI), 5-LO, inducible nitric oxide synthase (iNOS), endothelial nitric oxide synthase (eNOS; Santa Cruz Biotechnology, Santa Cruz, CA), polyclonal nitrotyrosine (Cayman Chemical)], mouse [monoclonal nitrotyrosine (Cayman Chemical)], or sheep [Cu/Zn SOD, Mn superoxide dismutase (MnSOD)] for 2 h at room temperature (20). β-Actin antibody was used as an additional loading control. After three washes in TBS-T, the membranes were incubated for 2 h at room temperature with secondary goat anti-rabbit, anti-mouse, or donkey anti-sheep alkaline phosphatase-conjugated antibodies and visualized using CDP-Star substrate (Perkin-Elmer, Woodbridge, Ontario) with the Chemigenius detection system using GeneTools software (Syngene, Frederick, MD) for densitometric analysis.
Thiobarbituric acid-reactive substances assay.
Lipid peroxidation in cardiac tissue was estimated by thiobarbituric acid-reactive substances (TBARS) assay. Frozen heart tissue was ground in liquid nitrogen and homogenized. After acidification of the homogenate with 1% phosphoric acid and treatment with 0.6% thiobarbituric acid (TBA) solution, the mixture was heated in a boiling water bath for 1 h in the presence of 0.4% butylated hydroxytoluene to prevent further oxidation of the tissue. After cooling, 1:2 adduct of malondialdehyde and TBA was extracted in 4 ml of n-butanol, and the absorbance was measured at 540 nm against 1,1,3,3-tetramethoxypropane used as the standard (19).
Assay for PLA2 isoforms.
Frozen heart sections were homogenized in cold buffer containing 50 mM HEPES and 1 mM EDTA (pH 7.4). Samples were centrifuged at 10,000 g for 15 min, and the supernatant was subjected to total PLA2 assay with a commercially available kit (Cayman Chemicals, Ann Arbor, MI). Briefly, PLA2 cleaves fatty acids from the sn-2 position of glycerophospholipids to yield FFA and lysophospholipid. Arachidonyl thio-phosphatidylcholine (thio-PC) is a synthetic substrate that contains an arachidonyl thioester bond at the sn-2 position. Hydrolysis of this bond releases free thiol that is detected by 5,5′-dithiobis(2-dinitrobenzoic acid) (DTNB) to give yellow-colored 5-thio-2-nitrobenzoic acid (λmax 412 nm; see Ref. 47). The sample was reacted with assay buffer (160 mM HEPES, 300 mM NaCl, 20 mM CaCl2, 8 mM Triton X, 60% glycerol, and 2 mg/ml BSA) and 1.5 mM arachidonyl thio-PC for 60 min at room temperature. iPLA2 and sPLA2 activity were blocked by incubating samples for 30 min with either 1 mM bromoenol lactone or 1 mM thioetheramide-phosphatidylcholine, respectively. Bee venom PLA2 was used a positive control. Finally, 10 μl of 25 mM DTNB/EDTA was added to develop color, which was measured spectrophotometrically at 405 nm and calculated as described previously (20).
Results are represented as means ± SE. Differences among the groups were determined using one-way ANOVA followed by Fisher's test. The level of statistical significance was set at P < 0.05.
Characteristics of animals.
There were no differences in body weight (low LA + ALA, 7,350 ± 515; low LA + high ALA, 6,872 ± 588; high LA + ALA, 7,038 ± 654 g) or heart weight (low LA + ALA, 34.9 ± 2.5; low LA + high ALA, 35.5 ± 2.5; high LA + ALA, 38.5 ± 2.6, g) among the three groups of pigs. Furthermore, we found no significant differences in the heart concentrations of total cholesterol, phospholipids, or individual phospholipids PC, PE, PI, and PS among the diet groups (Table 1). However, the concentrations of FFA and LPC were both significantly higher in the heart of the high LA + ALA-fed pig hearts (Table 1).
Membrane phospholipid fatty acids.
Analysis of the fatty acid composition of membrane phospholipid (PL), PE, PC, PS, and PI shows marked effects of the dietary fatty acid composition on the heart membrane n-6 and n-3 PUFA (Figs. 1 and 2). The high LA + ALA-fed pigs had high levels of AA in PE, PC, PS and PI. Although this might be anticipated because the high LA + ALA diet provided 11.6% energy from LA, the results in Fig. 1A also emphasize high levels of AA in the heart of pigs fed the low LA + ALA diet with only 1.2% dietary energy from LA. Remarkably, while maintaining a similar low LA intake of 1.2% energy, addition of 1.2% ALA to the low LA diet resulted in a marked decrease in the heart PL of AA and significantly higher EPA (Fig. 1B), ALA (Fig. 2B), and DHA (Fig. 2D) in pigs fed the low LA + high ALA diet. Of note, the heart membrane phospholipid levels of dihommo γ-linolenic acid (20:3n-6; Fig. 1C) were lower, whereas AA was higher in the high LA + ALA compared with low LA + high ALA-fed pigs. The concomitant higher EPA in the heart phospholipids of the low LA + high ALA compared with high LA + ALA-fed pigs suggests competition at the levels of the Δ5-desaturase, which in presence of a low dietary LA/ALA balance facilitates greater metabolism of the n-3 series fatty acids (40). Important to emphasize, the amount of ALA was similar in the low LA + high ALA and high LA + ALA diets (1.2% energy from ALA in each), yet increasing LA to 11.6% energy in the high LA + ALA diet decreased the heart membrane PL ALA, EPA, and DHA in the high LA + ALA-fed pigs. The higher AA and lower EPA in heart PL of both the high LA + ALA- and low LA + ALA-fed pigs resulted in AA-to-EPA ratios between 18 and 60 compared with <3 in the low LA + high ALA-fed pigs (Fig. 1D).
Activity of PLA2 isoforms, AA, and EPA levels.
We found no statistically significant differences in the total cardiac PLA2 activity among the groups (1.68 ± 0.2, 1.88 ± 0.3, and 1.90 ± 0.3 μmol·min−1·mg protein−1 for the low LA + ALA, low LA + high ALA, and high LA + ALA groups, respectively). However, the dietary fatty acid composition did influence the activity of the PLA2 isoforms. The heart cPLA2 activity was the highest in the high LA + ALA-fed pigs, with the lowest activity in the low LA + low ALA-fed pigs and intermediate activity in the low LA + high ALA-fed pigs (Fig. 3A). iPLA2 activity showed an opposite effect, with the highest activity in the low LA + ALA-fed pigs and lower activities in the LA + high ALA- and high LA + ALA-fed groups (Fig. 3B). sPLA2 activity was negligible, and we could discern no differences in activity among the three groups (low LA + ALA, 0.04 ± 0.01; low LA + high ALA, 0.06 ± 0.0; high LA + ALA, 0.07 ± 0.0 μmol·min−1·mg protein−1).
Both cPLA2 and iPLA2 hydrolyze the sn-2 position of PL, leading to release of unesterified AA and EPA. Because the levels of AA and EPA differed in the heart PL of pigs fed the different diets, we also analyzed AA and EPA in heart FFA. As shown in Fig. 3C, unesterified AA was significantly higher in the heart of the low LA + ALA- or high LA + ALA-fed pigs than in the low LA + high ALA-fed pigs. Unesterified EPA, on the other hand, was significantly higher in the heart of the low LA + high LA-fed pigs than in the other two diet groups (Fig. 3D). We also note that the concentration of FFA was higher in the heart of the high LA + ALA (Table 1), which together with the higher percent AA and lower EPA would increase the concentration of unesterified AA/EPA in the heart of the high LA + high ALA compared with low LA + ALA- or low LA + ALA-fed pigs (Table 1).
Activation of proinflammatory enzymes in the high fat-fed pig hearts.
Analysis of protein expression of COX-1 by Western blotting showed no significant differences in expression among the groups (data not shown). However, regardless of the absence of experimentally induced stress or disease, we found marked differences in the heart COX-2 and 5-LO expression among the groups. Notably, the inducible proinflammatory COX-2 protein expression was significantly higher in the heart of the low LA + ALA-fed pigs when compared with the other two diets (Fig. 4A). Because activation of 5-LO involves translocation from the cytosol to the membrane (54), we determined 5-LO protein expression in both the cytosolic and membrane fractions. Using this approach, we show selective upregulation of 5-LO protein in the membrane fractions (Fig. 4B) from the heart of pigs fed the low LA + ALA diet only.
Estimation of cardiac nitrosative stress.
Both COX-2 and 5-LO activation can be modified by nitrosative stress (10, 17). Nitrosative stress is the oxidative damage induced by reactive nitrogen species like peroxynitrite. Nitric oxide (NO) is predominantly generated by nitric oxide synthases, including iNOS and eNOS. iNOS protein expression was significantly higher in the heart of high LA + ALA-fed pigs than in pigs fed the other two diets, whereas eNOS protein expression was not different among the groups (Fig. 5, A and B). NO, when combined with superoxide, can form peroxynitrite, a particularly damaging free radical (4). Cu/Zn SOD and MnSOD neutralize excess superoxide to prevent the formation of superoxide (4). Although Cu/Zn SOD expression was not different among the groups (data not shown), MnSOD protein expression was significantly lower in the heart of the high LA + ALA-fed pigs than in the low LA + ALA- or low LA + high ALA-fed pigs (Fig. 5C). Cardiac nitrotyrosine levels were estimated using both polyclonal (directed against total cell lysate treated with peroxynitrite) and monoclonal (BSA treated with peroxynitrite) antibodies for qualitative and quantitative analysis. Analysis of nitrotyrosine by monoclonal antibody demonstrated approximately threefold higher levels of nitrated BSA in the heart of the high LA + ALA-fed pigs than in either the low LA + ALA- or low LA + high ALA-fed pigs (Fig. 6A). Measurement of lipid peroxidation using the TBARS assay showed significantly lower stress in the low LA + high ALA-fed pigs than in the low LA + ALA-fed pigs, with the highest stress in the high LA + ALA-fed animals (Fig. 6B).
To date, most studies relating dietary fat to inflammatory processes in the heart have focused on individuals with preexisting diabetes, atherosclerosis, or heart failure rather than healthy individuals, which is more relevant to the initiation of cardiovascular inflammation and interventions for disease prevention. Although considerable emphasis has been given to the role of EPA and DHA in CVD (31), the metabolic effects of dietary ALA are less well understood and yet are of significant practical and physiological relevance to Western diets. In this study, we present novel evidence that, in normal healthy pigs, 1 mo of feeding a diet differing in the LA-to-ALA ratio has a dramatic effect on cardiac membrane PL fatty acids, activities of PLA2 isoforms, the expression of proinflammatory enzymes central to the initiation of cardiac inflammation, and measures of nitrosative stress and lipid peroxidation.
Using diets differing in fatty acid composition, but not total fat, and in the absence of differences in body weight, dietary energy density, or other nutrients, we show that a diet high in ALA with low LA decreases the level of AA in cardiac membranes, an effect that is evident only when the diet also contains low levels of the n-6 LA. Although LA provided only 1.2% total energy in the low LA + ALA diet, the heart levels of AA exceeded 30% fatty acids in PE and PI, and growth was not different from that of animals fed 11.6% energy from LA. Our results show that LA is further desaturated to AA and efficiently acylated into membrane lipids, thus reaffirming that dietary requirements for LA are very low, in the range of 1% dietary energy (24, 25). On the other hand, the dietary intakes of LA have increased over the last century from 3% to an average of 7% dietary energy in the United States and other Western nations, with half of the population consuming higher levels of LA. However, our results to show a marked effect of dietary LA at 11.2% dietary energy, which is within the range in human diets (1), in reducing the incorporation of all n-3 fatty acids including ALA, EPA, and DHA in heart membrane PL raises the question of whether high intakes of n-6 fatty acids antagonize n-3 fatty acid incorporation into heart lipids in humans. Possibly, differences in n-6 fatty acid intakes could contribute to the lack of consistency among studies on the beneficial effect of dietary ALA in reducing risk of CVD. Our results also clearly raise the possibility that any positive effects of ALA are only realized when the diet is also low in LA (18, 45), such as those found in intervention trials incorporating a Mediterranean diet high in monounsaturated fatty acids (13, 20a).
In agreement with previous studies, our results show that differing dietary LA-to-ALA ratios do not alter heart total PLA2 activity (2). However, to our knowledge, this is the first report demonstrating that dietary n-6 and n-3 fatty acids modulate PLA2 isoforms in the healthy myocardium. PLA2 does not discriminate between EPA- and AA-containing PL, thus explaining the higher unesterified AA in the heart of the low LA + ALA- and high LA + ALA-fed pigs and higher unesterified EPA in the heart of the low LA + high ALA-fed pigs. Notably, iPLA2 activity was highest in the heart of the low LA + ALA-fed pigs, whereas Ca2+-dependent cPLA2 activity was highest in the heart of the high LA + ALA-fed group. The reasons for the difference in iPLA2 or cPLA2 is not readily apparent from our studies. However, because Ca2+ and substrate concentrations were not limiting under our assay conditions, it is reasonable to consider possible changes in the protein expression of these enzymes may be involved.
The effects of membrane PL n-6 and n-3 fatty acids on inflammatory signaling are complex. In the cardiovascular system, AA is metabolized to eicosanoids responsible for smooth muscle constriction, platelet aggregation, and other inflammatory responses via COX and 5-LO enzymatic pathways, although some studies have reported that EPA can directly inhibit both COX and 5-LO activities (1, 27, 30). The higher levels of EPA in the heart membrane PL and FFA in the heart PL and FFA of the low LA + ALA-fed pigs may explain at least in part our results to show lower COX-2 protein expression and 5-LO membrane translocation in this group. LA and ALA compete for the desaturase enzyme required for synthesis of AA and EPA from LA and ALA, respectively. As expected then, the low LA + ALA-fed pigs had lower AA and higher EPA in their heart PL than pigs fed a similar amount of ALA but with high LA. PLA2, however, lacks substrate specificity; thus, diet-induced changes in membrane PL AA and EPA were accompanied by the same changes in AA and EPA in the unesterifed fatty acids. However, the present study involved unstressed, nondiseased animals and did not include measures of eicosanoids or cytokines. Also of importance, AA plays important roles in the initiation of inflammation, and, through its role as a precursor of lipoxins, AA also plays an important role in the resolution of inflammation (50). Thus dietary or disease conditions that lower AA could contribute to an inappropriate prolongation of inflammatory responses. However, our work focuses on the potential role of tissue fatty acids in contributing to a proinflammatory or prooxidative state. In this regard, recent studies have suggested that LA also has proinflammatory properties that may have important cardiovascular implications (43, 49). Thus the high membrane LA in high LA + ALA-fed animals could contribute to a prooxidative and proinflammatory state within this group.
Interestingly, we found no evidence that increased membrane PL or free AA or LA led to increased 5-LO translocation or COX-2 expression in the heart of the high LA + ALA-fed pigs. Thus we explored other possible factors that could inhibit these proinflammatory enzymes. Previous studies have demonstrated that COX-2 and 5-LO expression can be inhibited by nitrosative stress (10, 17), a factor often ignored while investigating the effect of dietary fat on proinflammatory enzyme regulation. Furthermore, recent studies have suggested that peroxynitrite, a reactive nitrogen species, can inhibit COX-2 through several mechanisms, including inhibition of COX-2 gene expression and nitrosylation of tyrosine residues in the catalytic domain that prevents COX-2 activation (7, 11, 17, 39). Of relevance, peroxynitrite can nitrosylate 5-LO directly or can inhibit 5-LO translocation and activity by oxidation of its iron from ferrous to ferric form (10). In the heart, NO, the precursor for peroxynitrite, is generated predominantly by eNOS, which is a physiological donor of NO, and iNOS, which can generate increased amounts of NO over extended periods of time (4). Previous studies have shown that increased dietary LA can induce the expression of iNOS in endothelial cells and cardiomyocytes (4, 20). In the present study, we show that, although eNOS expression did not change, iNOS expression was upregulated in the heart of the high LA + ALA-fed pigs, providing a possible mechanism for chronically increased cardiac NO production. NO in the presence of excess superoxide can form the oxidant peroxynitrite. Besides inactivation of proinflammatory enzymes, peroxynitrite can also nitrate tyrosine residues of proteins to produce nitrotyrosine (4). Peroxynitrite can also damage mitochondria and DNA and precipitate long-term complications in postmitotic tissues such as the heart. Importantly, we found nitrotyrosine levels that were almost threefold higher in the hearts of pigs fed the high LA + ALA diet than in pigs fed the two diets low in LA.
Notably, we also provide evidence of increased oxidative stress via lipid peroxidation in the heart when low LA + ALA was fed. However, the low ALA diet provided only 0.06% of energy from ALA, raising the possibility of increased oxidative stress because of inadequate n-3 PUFA, independent of peroxynitrite generation or nitrosative stress. In this regard, in vitro and in vivo studies have provided evidence that low n-3 PUFA is associated with increased reactive oxygen species like hydroxyl radicals and hydrogen peroxide, which can promote mild oxidative stress and can stimulate COX-2 expression and 5-LO translocation (33, 35, 37). Further studies will be needed to determine whether the low dietary n-3 PUFA was responsible for the increased COX-2 and 5-LO found in the heart of pigs fed the low LA + ALA diet. MnSOD, the chief mitochondrial antioxidant, is a stress-responsive gene that increases in response to superoxide within the cell (29). Interestingly, the present study shows higher MnSOD, probably as a consequence of higher oxidative stress in the heart of low LA + ALA-fed pigs. However, on the other hand, decreased oxidative stress in the low LA + high ALA-fed pig hearts provides a reasonable explanation for a decreased expression of MnSOD. Alternately, in the presence of excess peroxynitrite, MnSOD protein may be modified via nitration (15), which could also explain why MnSOD was lower in the hearts of pigs fed a high LA diet.
In summary, we demonstrate that diets differing in n-6 LA and n-3 ALA and their ratios, representing the broad range of LA and ALA in the diet of human populations (52), do influence PLA2 isoforms and the proinflammatory enzymes COX-2 and 5-LO in the heart of healthy pigs. We also provide evidence that the potential cardiovascular benefit of high dietary ALA and membrane incorporation of other ALA, EPA, and DHA is not achieved when ALA is fed in diets rich in LA. Lowering the dietary LA-to-ALA ratio closer to 1:1 decreases 5-LO and COX activation and reduces lipid peroxidaton and nitrotyrosine production in the heart. Many clinical trials have shown that, when compared with saturated fatty acids, increasing LA is effective in reducing serum total and LDL cholesterol. However, we provide strong evidence from in vivo studies that LA can increase nitrosative stress, especially nitrotyrosine levels, a measure not usually estimated in clinical trials (20, 49). Importantly, although chronic nitrosative stress can prevent inflammatory eicosanoid production, peroxynitrite can damage mitochondria and DNA and precipitate long-term complications in postmitotic tissues such as the heart. In this regard, the incidence of CVD has indeed increased in parallel with the increase in LA intakes in many nations (31, 32). The present study showing increased cardiac nitrosative stress in the unstressed, nondiseased heart with short-term, high LA intake, representing 11.6% dietary energy suggests the need to consider measures of nitrotyrosine as a biomarker in future studies on the role of dietary PUFA in oxidative stress.
The studies described in this paper were supported in part by an operating grant from the Canadian Institute for Health Research and an Unrestricted Freedom-to-Discover Research Award from the Bristol Myers Squibb Foundation. Graduate studentship awards from the Child and Family Research Institute and National Sciences and Engineering Research Council (Elizabeth Novak) and a postdoctoral fellowship from the Child and Family Research Institute (Sanjoy Ghosh) are also gratefully acknowledged.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 by the American Physiological Society