Oleic Acid Inhibits Endothelial Activation
A Direct Vascular Antiatherogenic Mechanism of a Nutritional Component in the Mediterranean Diet
Abstract—Because oleic acid is implicated in the antiatherogenic effects attributed to the Mediterranean diet, we investigated whether this fatty acid can modulate endothelial activation, ie, the concerted expression of gene products involved in leukocyte recruitment and early atherogenesis. We incubated sodium oleate with human umbilical vein endothelial cells for 0 to 72 hours, followed by coincubation of oleate with human recombinant tumor necrosis factor, interleukin (IL)-1α, IL-1β, IL-4, Escherichia coli lipopolysaccharide (LPS), or phorbol 12-myristate 13-acetate for a further 6 to 24 hours. The endothelial expression of vascular cell adhesion molecule-1 (VCAM-1), E-selectin, and intercellular adhesion molecule-1 was monitored by cell surface enzyme immunoassays or flow cytometry, and steady-state levels of VCAM-1 mRNA were assessed by Northern blot analysis. At 10 to 100 μmol/L for >24 hours, oleate inhibited the expression of all adhesion molecules tested. After a 72-hour incubation with oleate and a further 16-hour incubation with oleate plus 1 μg/mL LPS, VCAM-1 expression was reduced by >40% compared with control. Adhesion of monocytoid U937 cells to LPS-treated endothelial cells was reduced concomitantly. Oleate also produced a quantitatively similar reduction of VCAM-1 mRNA levels on Northern blot analysis and inhibited nuclear factor-κB activation on electrophoretic mobility shift assays. Incubation of endothelial cells with oleate for 72 hours decreased the relative proportions of saturated (palmitic and stearic) acids in total cell lipids and increased the proportions of oleate in total cell lipids without significantly changing the relative proportions of polyunsaturated fatty acids. Although less potent than polyunsaturated fatty acids in inhibiting endothelial activation, oleic acid may contribute to the prevention of atherogenesis through selective displacement of saturated fatty acids in cell membrane phospholipids and a consequent modulation of gene expression for molecules involved in monocyte recruitment.
- Received May 6, 1998.
- Accepted June 3, 1998.
There is strong evidence for an association between a Mediterranean-style diet, in which olive oil is the principal source of fat, and protection from cardiovascular disease.1 2 3 Although intervention studies with morbidity and mortality end points are still generally lacking, the association has been recognized as an established fact.4 There are a multitude of effects of components of the Mediterranean diet potentially able to explain a causal association. Such diets lower total and LDL cholesterol compared with diets very rich in saturated fatty acids,5 6 7 thus reducing a dominant risk factor for the development of atherosclerosis. Additional mechanisms may be favorable effects on other coronary risk factors, such as hypertension,8 9 diabetes,10 11 12 and excess body weight.13
Atherosclerosis is thought to be initiated at critical sites of the arterial vasculature by a process of monocyte adhesion to the vessel wall, sustained by the occurrence of active functional changes on the endothelial surface.14 Recent research has shown direct vascular effects of fatty acids, with possible relevance to atherogenesis. In particular, the ω-3 fatty acids docosahexaenoate (DHA) and, to a lesser extent, eicosapentaenoate, which have only minor effects on total and LDL cholesterol15 and yet also appear to be linked to protection from atherosclerosis,16 may act by inhibiting early atherogenic events related to monocyte adhesion to endothelial cells.17 18 19 This process occurs through inhibition of endothelial activation,17 18 ie, the concerted expression of cytokine-inducible endothelial leukocyte adhesion molecules and leukocyte chemoattractants affecting monocyte adhesion. Inhibition of a common signal-transduction pathway involving the transcription factor nuclear factor-κB (NF-κB)20 was therefore hypothesized.17 On the basis of a comparison of the biological activity of several fatty acids, we have also reported preliminary evidence that an important structural feature for this effect appears to be the presence of double bonds on the fatty acid backbone.21
We therefore anticipated that the monounsaturated fatty acid oleate, which is the predominant fatty acid in Mediterranean diets, also possesses some inhibitory activity on endothelial activation. In accordance, the present investigation was undertaken aimed at (1) demonstrating the occurrence, extent, and conditions for this effect; (2) showing its relationship to changes in fatty acid composition in endothelial cell lipids; and (3) offering mechanistic clues for its biological plausibility in explaining the beneficial effects of diets rich in oleate on atherogenesis.
Endothelial Cell Cultures
Human umbilical vein endothelial cells (HUVECs) were isolated from umbilical cords harvested in the Obstetrics and Gynecology Division of “Vito Fazzi” Hospital in Lecce, Italy. HUVECs were harvested enzymatically with 1 mg/mL type IA collagenase as described22 and maintained in medium 199 (Mascia Brunelli) containing 25 mmol/L HEPES, 50 U/mL heparin,23 50 μg/mL endothelial cell growth factor,24 2 mmol/L l-glutamine, antibiotics, and 15% FCS. Once grown to confluence, the cells were replated on 1.5% gelatin–coated flasks at 20 000 cells/cm2. HUVECs isolated by these techniques form a confluent monolayer of polygonal cells and express von Willebrand factor as determined by their content of immunoreactive protein. Cells were used at a time calibrated to allow them reach confluence (60 000 to 80 000 cells/cm2) at the time of stimulation.
Oleic acid (18:1 n-9 cis), palmitic (16:0) and stearic (18:0) acids as negative controls, and DHA (22:6 n-3 all-cis) as a positive control for inhibition of endothelial activation were obtained as >99% pure sodium salts from Nu-Chek, dissolved in water at 10 mmol/L stock concentration, divided into aliquots under an N2 stream, and stored at −80°C for no longer than 6 months before the experiments. For controls, the same fatty acids were also obtained from Sigma. At the time of the experiments, fatty acids were further dissolved in serum-containing medium at the final desired concentration.
The following human recombinant cytokines were used at the reported final concentrations: interleukin (IL)-1α (Hoffmann–La Roche, at 1 to 10 ng/mL), IL-1β (1 to 10 ng/mL), tumor necrosis factor (TNF)-α (1 to 10 ng/mL), and IL-4 (50 to 100 ng/mL) (the latter 3 all from Genzyme). Escherichia coli lipopolysaccharide (LPS, 10 to 10 000 ng/mL) was purchased from Sigma. Phorbol 12-myristate 13-acetate (PMA, 10 nmol/L, equivalent to 6.3 ng/mL) as a stimulus for endothelial activation that bypasses membrane receptors was purchased from Sigma as were all other reagents, when not otherwise specified.
Detection of Cell Surface Molecules
Assay of cell surface molecules was carried out by either cell surface enzyme immunoassays (EIAs) or flow cytometry by using mouse anti-human monoclonal antibodies against vascular cell adhesion molecule-1 (VCAM-1; Ab E1/6), E-selectin (Ab H18/7), intercellular adhesion molecule-1 (ICAM-1; Ab HU5/3), major histocompatibility complex class I (MHC-I; W6/3225 ), or the monoclonal antibody E1/1, recognizing a constitutive and non–cytokine-inducible endothelial cell antigen.26 These last 2 endothelial surface molecules share a high level of basal constitutive expression, although MHC-I antigen, at variance from E1/1 antigen, shows some degree of inducibility by cytokines. EIAs were carried out by incubating monolayers first with saturating concentrations of specific monoclonal antibodies against the target molecule, then with biotinylated goat anti-mouse IgG (Amersham Life Sciences, Amersham), and finally with streptavidin–alkaline phosphatase (Amersham). Layers were washed 3 times with PBS between each incubation step, and the integrity of the monolayer was monitored by phase-contrast microscopy. The surface expression of each protein was quantified spectrophotometrically by reading the optical density of the wells at a wavelength of 450 nm 20 minutes after addition of the chromogenic substrate (3,3′,5,5′-tetramethylbenzidine) and stopping the reaction with 1N H2SO4.
For flow-cytometric analysis, the surface expression of adhesion molecules was assessed by incubating HUVECs, suspended in Hanks’ buffered saline solution with 3 mmol/L EDTA, with the specific primary antibody for 30 minutes at 4°C and subsequently with goat anti-mouse F(ab′)2 from IgG (heavy and light) labeled with fluorescein isothiocyanate (Immunotech) at 4°C. After washing and fixation in 1% formaldehyde, the cell suspension was passed through a Becton-Dickinson FACScan analyzer. Results were plotted as intensity of fluorescence (arbitrary units, on a logarithmic scale as the abscissa) versus cell number (on the ordinate; the total number of cells counted for each condition was 104).
Assessment of Cell Number and Viability
Cell number was assessed by direct cell counting of adherent cells, after trypsin detachment, in a Neubauer hemocytometer (VWR Scientific) and staining with trypan blue. The percentage of cells excluding trypan blue was taken as a measure of cell viability.
Assessment of Total Protein Synthesis
HUVECs were cultured in 96-well plates in the presence or absence of fatty acids for 0 to 72 hours and then for up to 24 hours in the presence or absence of LPS or cytokines. After this time, the total cell-associated protein content was assessed by the Bradford method.27 In parallel experiments, the dye reaction was correlated linearly (R=0.98) with cell number per well, which in turn was determined by hemocytometric counting of trypsinized cells in parallel cultures.
Monocytoid Cell Adhesion Assays
Monocytoid U937 cells were obtained through American Type Culture Collection (Manassas, Va) and grown in RPMI 1640 medium (Gibco BRL, Life Technologies Italia srl) containing 10% FCS. U937 cells were concentrated by centrifugation to 106cells/mL. For the adhesion assays, HUVECs were grown to confluence in 6-well tissue culture plates, after which LPS (10 to 10 000 ng/mL) or IL-1α (10 ng/mL) was added for an additional 16 hours to induce the expression of VCAM-1 in the presence or absence of oleate (10 to 100 μmol/L). For controls, some monolayers were treated with a blocking mouse anti-human monoclonal antibody (E1/6) against VCAM-1. The adhesion assay was performed by adding 1 mL of the concentrated U937 cell suspension to each monolayer under rotating conditions (63 rpm) at 21°C.28 29 After 10 minutes, nonadhering cells were removed by gentle washing with medium 199, and the monolayers were fixed with 1% paraformaldehyde. The number of adherent cells was determined by counting 6 different fields by using an ocular grid and a 20× objective (0.16 mm2/field). Fields for counting adherent leukocytes were randomly located at a half-radius distance from the center of the monolayers.
Isolation of RNA and Northern Blot Analysis
Total cellular RNA was isolated by a single extraction using an acid guanidinium thiocyanate-phenol-chloroform method.30 RNA concentration and quality were determined from the absorbance at 260 nm and the absorbance ratio 260 nm/280 nm, respectively. RNA quality was confirmed by gel electrophoresis before Northern blot analysis. For this analysis, 20 μg of cellular RNA was applied to each lane, separated on a 1% agarose-formaldehyde gel, transferred to a nylon membrane (Amersham Hybond-N), and immobilized by short-wave UV illumination. The membranes were prehybridized for at least 2 hours before hybridization with a 32P-labeled DNA probe for VCAM-1 (labeled by random hexanucleotide priming; Pharmacia) to specific activities >108 counts per minute per μg DNA and autoradiographed. 18S and 28S rRNA fluorescence intensity of the ethidium bromide–stained membranes served as a control of equal loading of different lanes in the Northern blot analyses.
Quantification of densities of autoradiographic bands for Northern blot hybridization was performed with the aid of the National Institutes of Health (Bethesda, Md) Image 1.60 software on a MacIntosh Quadra 800 computer (Apple).
Preparation of Nuclear Extracts and Electrophoretic Mobility Shift Assay (EMSA)
Confluent HUVECs were pretreated for 0 to 72 hours with 50 μmol/L and 100 μmol/L sodium oleate and then exposed to LPS (1 μg/mL) for 1 hour. Cells were scraped mechanically and collected into chilled microfuge tubes. Nuclear and cytosolic extracts were prepared according to Dignam et al,31 with the additional step of washing the nuclear pellets in a low-salt buffer before high-salt extraction of nuclear proteins to remove any residual cytosolic contamination. Aliquots were then assayed for protein concentration.27 To minimize proteolysis, all buffers included 1.0 mmol/L PMSF, and extracts were stored at −80°C until analysis. The wild-type VCAM-1 promoter oligonucleotide was synthesized to encompass the 2 NF-κB consensus repeats (underlined) described at coordinates −77 and −63 of the human VCAM-1 promoter,32 according to the sequence: 5′-CTGCCCTGGGTTTCCCCTTGA-AGGGATTTCCCTCC-3′, together with its complementary oligonucleotide, 5′-GAGGCGGAGGGAAATCCCTTCAAGGGG-AAACCCAGG-3′.
The mutant VCAM-1 promoter oligonucleotide was identical to the wild-type VCAM-1 promoter oligonucleotide except for 4 nucleotide mutations (boldface) in each of the 2 NF-κB binding sites: 5′-CTGCCCTGAGTCACGCCTTGAAGAGACATCACTCC-3′, together with the corresponding complementary oligonucleotide, 5′-GAGGCGGAGGGAAATCCCTTCAAGGGGAAACCCAGG-3′. All oligonucleotides were synthesized by GIBCO-BRL Life Technologies, and double-stranded oligonucleotides were radiolabeled by Klenow filling-in with 50 μCi of [α-32P]dCTP (3000 Ci/mmol) and unlabeled dATP, dGTP, and dTTP (each at 100 μol/L). Unincorporated nucleotides were removed by column chromatography over a Sephadex G-50 column. The DNA binding reaction was performed at 30°C for 15 minutes in a volume of 20 μL containing 8 μg of nuclear extract, 225 μg/mL BSA, 10 000 cpm of 32P-labeled probe, 2 μg poly(dI-dC) (Boehringer Mannheim Italia), 15 μL of binding buffer (12 mmol/L HEPES, pH 7.9, 4 mmol/L Tris, 60 mmol/L KCl, 1 mol/L EDTA, 12% glycerol, 1 mol/L DTT, and 1 mol/L PMSF) with or without a 100-fold molar excess of cold competitor. Samples were subjected to electrophoresis on native 5% 0.5× Tris-borate-polyacrylamide gels.
Fatty Acid Analysis
Relative amounts of the individual chemical species of the most abundant fatty acids present in cell lipids were quantified by integration of peaks obtained by gas-liquid chromatographic separation of fatty acids from HUVEC monolayers. After chloroform-methanol (2:1, vol/vol) extraction, total cell lipids were subjected to acidic trans-methylation,33 and fatty acid methyl esters were analyzed by flame-ionization capillary gas-liquid chromatography on a Hewlett-Packard GC system, HP 6890 series, using a Supelco Omegawax 250 column (30 m×0.25-mm ID, 0.25-mm film thickness). The oven temperature was programmed at 150°C for 15 minutes and then increased to 250°C at 4°C/min, then to 250°C for 35 minutes. The H2 and air flow were 30 and 330 mL/min, respectively; the carrier (helium) pressure was 17 psi. Individual peaks were integrated automatically with a Hewlett-Packard integrator and identified by comparison with retention times of fatty acid standards.
Cultured HUVECs were preincubated with oleate or other fatty acids for 0 to 72 hours, followed by stimulation with cytokines (IL-1α, IL-1β, TNF, IL-4), PMA, or LPS for an additional 16 to 24 hours. At the end of this period, the expression of endothelial surface molecules (cell surface EIA and/or flow cytometry) of VCAM-1 mRNA (Northern blot analysis) or of parameters of cell viability (morphology, number, trypan blue exclusion, total protein synthesis) were assessed, or the adhesion assays were performed. Each experiment was performed at least in triplicate. In EIAs, a minimum number of 8 repeats was run in each experimental condition.
Multiple comparisons were performed by 1-way ANOVA and individual differences tested by Fisher’s protected least significance difference test after demonstration of significant intergroup differences by ANOVA. Two-group comparisons were performed by unpaired Student’s t test. Comparisons of distribution of fluorescence intensities for flow cytometry were performed with Kolmogorov-Smirnov’s statistics with the aid of the Becton-Dickinson statistical software package.
Toxicity Profile of Oleate Effects on HUVECs
The absence of any detectable toxic effect of sodium oleate at the concentrations and times used on cell number or parameters of cell viability is shown in Table 1⇓. Sodium oleate did not reduce cell counts and did not affect either trypan blue exclusion or the estimate of total cell-associated protein for concentrations up to 100 μmol/L and for preincubation times up to 72 hours before addition of the stimulating cytokines, ruling out the possibility that the effects described below could be due to a general inhibition of cell metabolism or any general “toxic” effect. Also, no change in MHC-I or E1/1 antigen expression was detected under any circumstance up to 100 μmol/L oleate concentrations (not shown).
Oleate Reduces LPS-Stimulated VCAM-1 Expression in HUVECs
Despite any general effect on total protein synthesis, preincubation of HUVECs with oleate for 72 hours and then for a further 16-hour incubation with oleate plus LPS caused a concentration-dependent inhibition of stimulated VCAM-1 expression, as assessed by both cell-surface EIA (Figure 1⇓) and flow cytometry (Figure 2⇓) when compared with the unsupplemented medium. Under optimal conditions (72-hour preincubation), the IC50 for this oleate effect was ≈50 μmol/L, between 5 and 10 times higher than what was obtained in comparative experiments with DHA (C22:6 n-3; not shown). No effects were obtained in our experimental system by the saturated fatty acids palmitate (C16:0) or stearate (C18:0), which served as alternative controls to the unsupplemented medium in selected experiments.
Oleate Effects Require Prolonged (Hours) Preincubation and Are Absent on Coincubation With Stimuli for VCAM-1 Expression
Parallel time courses of HUVEC preincubation with 2 different concentrations of oleate (10 and 100 μmol/L) before addition of the stimulating cytokine are shown in Figure 3⇓. Oleate effects were totally absent in the absence of any incubation or after very short preincubations (<6 hours), even with a relatively high oleate concentration (100 μmol/L; Figure 3⇓, upper part). Also, addition of oleate after the stimulating cytokine was devoid of any effect (not shown). On the other hand, significant effects became apparent with much lower concentrations (≈10 μmol/L), provided that the preincubation had been carried out for sufficiently long times (>24 hours; Figure 3⇓, lower part).
Oleate Effects on VCAM-1 Expression Are Independent of the Stimuli Used
A comparison of the effects of a high oleate concentration (100 μmol/L) with a 72-hour preincubation before the addition of either LPS or IL-1α is shown in Figure 4⇓. The magnitude of the effect was remarkably similar with the 2 stimuli. Indeed, the magnitude of the effect was independent of the stimulus used to elicit VCAM-1 expression; this was also the case for IL-1β, IL-4, or TNFα (not shown). To assess whether endothelial activation by a stimulus bypassing cell surface receptors was also inhibited by oleate preincubation, we performed experiments with the protein-kinase C activator PMA. Again, oleate inhibited VCAM-1 expression induced by PMA to an extent similar to cytokines or LPS (Figure 4⇓).
Oleate Is a Global Inhibitor of Endothelial Activation
We used VCAM-1 expression as a paradigm for oleate effects on HUVECs activated with LPS or cytokines in most experiments. However, the effect of oleate, similar to what had been previously reported for DHA, was not restricted to VCAM-1. E-selectin, an adhesion molecule with a different time course of surface appearance (peaking at 8 to 12 hours and subsequently quickly decreasing to near zero expression at 24 hours), and ICAM-1, which is constitutively expressed on the endothelial cell surface and further induced by cytokines, were also inhibited by oleate, indicating a generalized effect on induced adhesion molecules (Figure 5⇓). The effect was similar in magnitude for VCAM-1 and E-selectin and lesser for ICAM-1. Of note, constitutive expression of ICAM-1 was not significantly affected by oleate (not shown).
Oleate Decreases Monocytoid Cell Adhesion to HUVECs
In accordance with data on the expression of endothelial leukocyte adhesion molecules, oleate decreased the adhesion of U937 monocytoid cells to HUVECs stimulated with LPS under rotating conditions (Figure 6⇓). Of note, U937 cell adhesion is mostly due to VCAM-1 expression under the experimental conditions used, as shown by the large extent of inhibition observed with the blocking monoclonal antibody E1/6. The effect of 100 μmol/L oleate preincubation for 72 hours on U937 cell adhesion stimulated by LPS was actually greater than what was expected on the basis of its effects on VCAM-1 expression.
Oleate Decreases VCAM-1 mRNA Steady-State Levels
To obtain some preliminary insight into the mechanism(s) by which oleate may affect endothelial activation, we investigated the VCAM-1 steady-state mRNA levels by Northern blot analysis. Oleate preincubation (100 μmol/L for 72 hours) was accompanied by an unequivocal reduction of the VCAM-1 mRNA bands on Northern blot analysis (Figure 7⇓) under experimental conditions not affecting the amount of total RNA or rRNA obtained. Quantitative videodensitometric analysis showed a 65% reduction of the VCAM-1 mRNA band (Figure 7⇓).
Oleate Decreases NF-κB Activation
We assessed the extent of NF-κB activation as a possible mechanistic explanation of the pretranslational effect of oleate on VCAM-1 mRNA detected by Northern blot analysis. To this purpose, an EMSA was performed on nuclear extracts from cells treated with or without oleate and then stimulated with LPS. LPS stimulation was associated with evidence of NF-κB activation, and this was markedly attenuated by oleate pretreatment (Figure 8⇓).
Oleate Effects on Endothelial Activation Are Accompanied by a Relative Selective Displacement of Saturated Fatty Acids in Total Cell Lipids and an Increase in the Unsaturation Index
To gain further insight on the relationship of oleate incorporation in total cell lipids with oleate effects on endothelial activation and of the changes in the relative proportions of saturated, monounsaturated, and polyunsaturated fatty acids on incubation with exogenous oleate, we assessed the fatty acid composition of total cell lipids under some of the experimental conditions described above. Fatty acid analysis was carried out for the most abundant saturated, monounsaturated and polyunsaturated fatty acids accounting for >90% of the total fatty acids reported to be present in endothelial cell membranes. HUVECs, under the experimental conditions used (passage 3 or 4 in serum-containing medium), contained 19% oleate in their total cell lipids at baseline. Supplementation of the medium with 100 μmol/L exogenous oleate caused an enrichment of oleate in total cell lipids, reaching 42% at 72 hours. There was a simultaneous decrease in saturated fatty acids, while the relative proportions of polyunsaturated fatty acids (both of the n-3 and n-6 classes) remained virtually unchanged (Table⇑ ). This result suggests that the effects of oleate are related to its increased incorporation in total cell lipids and that oleate incorporation proceeds almost totally at the expense of saturated fatty acids. As a result of this relative replacement selectivity, there was an increase in the unsaturation index as a consequence of exposure to oleate. Such an index, calculated as the sum of the relative percent of unsaturated fatty acids measured and multiplied by the number of double bonds in each fatty acid, changed from 65.6 before oleate supplementation to 96.0 after supplementation.
This study shows that oleic acid, when added to endothelial cells in culture and under conditions yielding significant incorporation in total cell lipids, is able to reduce stimulated surface expression of endothelial leukocyte adhesion molecules in a concentration- and time-dependent fashion. The effect appears to be a generalized “damping” of endothelial activation, is specific for cytokine-inducible molecules, is independent of the stimulus used to elicit endothelial activation, and is accompanied by a functional counterpart in the inhibition of monocytoid cell adhesion to endothelium. The effect of oleate on protein expression and on monocyte adhesion is accounted for, on a quantitative basis, by an inhibition of corresponding mRNA and, upstream, by inhibition of NF-κB activation after endothelial cell stimulation. To the best of our knowledge, this is the first report of a direct, vascular effect of oleate on vascular cells and is a candidate explanation for direct antiatherogenic properties of diets rich in this compound, adding to other beneficial effects of oleate on the lipid profile or other cardiovascular risk factors.
Oleic acid is a predominant component of olive oil and, in turn, of the Mediterranean diet,1 2 3 which is currently thought to exert atheroprotective effects, mostly through a lowering of total and LDL cholesterol5 6 7 or through a decrease in coronary risk factors such as hypertension,8 9 diabetes,10 11 12 or obesity.13 The data presented here make the alternative or additional explanation plausible that at least some of oleate’s beneficial effects are exerted through inhibition of the very early phenomena in atherogenesis, leading to monocyte recruitment and the development of the fatty streak. Fatty acids in general have recently emerged as possible physiological regulators of the endothelial responsiveness to activating stimuli. In particular, the highly unsaturated n-3 fatty acid DHA, abundant in fish and fish-derived products, has been shown by us17 18 as well as others19 to be a potent inhibitor of endothelial responses to cytokines, thus decreasing the expression of VCAM-1, E-selectin, and ICAM-1, as well as of the soluble products of endothelial activation, such as IL-6 and IL-8. Functional consequences of these biochemical effects have also been shown.17 18 In an effort to understand the structural component of the fatty acid structure required for these effects, we have recently compared several fatty acids, differing in chain length and the number, position, and configuration of the double bonds. While saturated fatty acids were always devoid of any ability to interfere with endothelial activation, the presence of double bonds appeared to be the minimum essential and sufficient requirement for the inhibition of endothelial activation, with no apparent relevance of the position (n-3 versus n-6) and even of the configuration (cis versus trans) of the double bond. The original fatty acid investigated, DHA, appeared to be the most potent because of the high number of double bonds, the highest able to be accommodated in the fatty acid chain of that length.21 This finding led to the hypothesis that the monounsaturated fatty acid oleate, in which the single double bond is located in the n-9 position but which is much more abundant nutritionally than many polyunsaturates, could also possess some of these properties. We therefore investigated the occurrence, extent, and conditions for this effect in detail and its relationship to changes in fatty acid composition in endothelial cell lipids and sought to reach conclusions on its biological plausibility in explaining its benefits on atherogenesis reduction.
In line with our predictions, oleate exerted a concentration-dependent inhibitory effect on stimulated VCAM-1 expression. The magnitude of inhibition, an ability to reduce VCAM-1 expression to ≈50% of the control response, depended on both the concentration and the duration of incubation. At any given concentration, a plateau of response to oleate was reached by prolonging the incubation time to 72 hours before cytokine stimulation, whereas no inhibition occurred by coincubation or administration of oleate after LPS or cytokines. This kinetics closely resembles that of DHA.17 Also, similar to what had been previously shown for DHA, the effect is totally independent of the stimulus used: cytokines acting on totally different receptors, such as IL-1 (α or β), TNF-α, IL-4, or LPS, and inducing different responses, both in terms of selectivity (IL-4 selectively inducing VCAM-1 but not E-selectin on endothelial cells) and relative potency (maximum response to IL-4 being, for example, <1/2 that induced by a maximal concentration of LPS), had their responses inhibited to the same extent by any given concentration of oleate administered with the same preincubation time. Furthermore, oleate also inhibited VCAM-1 expression induced by PMA, an agent able to induce endothelial activation that bypasses cell surface receptors and likely acting through direct activation of protein kinase C, possibly even bypassing NF-κB activation.34 We also confirm, in line with what we reported for other more unsaturated fatty acids,17 but contrary to what has been reported by others for DHA,19 that the effect of oleate is not restricted to VCAM-1 expression but also pertains to other inducible adhesion molecules, suggesting a mechanism on a common pathogenetic transducer of endothelial activation. Together with the data on the inhibition of VCAM-1 mRNA accumulation obtained herein by Northern blot analysis and the notion that the regulation of adhesion molecule expression appears to be to largely transcriptional,20 we postulate transcriptional interference by oleate, possibly mediated by inhibition of the activation of the transcription factor NF-κB. The same concentrations of oleate that exerted effects on adhesion molecule expression were indeed also able to decrease the nuclear translocation and binding of NF-κB to its consensus sequences, showing for the first time an effect of a monounsaturated fatty acid on a common pathway of cell activation by a variety of cytokine and noncytokine stimuli. Thus, the effect of oleate on NF-κB is at least a partial explanation for its pleiotropy, ie, effects on multiple adhesion molecules and, possibly, on other mediators of endothelial activation implicated in monocyte and monocytoid cell adhesion (eg, monocyte chemoattractant protein-135 ). This pleiotropy would also account for the noteworthy inhibition of monocytoid cell adhesion, which we herein report being actually larger than expected on the basis of VCAM-1 inhibition data alone. Our results do not rule out additional effects of oleate on adhesion molecule expression independent of NF-κB, as actually suspected on the basis of proven effects on PMA- or IL-4–induced activation. Such effects may be modifications of specific receptor-agonist interactions or on protein kinase C–dependent signal transduction pathways that have not been investigated further in the current study. However, results on NF-κB activation may account for most of the effects seen with the activating cytokines used.
The biological relevance of the observations herein reported for oleate is mostly supported by 2 lines of reasoning. The first relates to concentrations of fatty acids required for the reported effects. Although fatty acids mostly circulate in plasma while bound to plasma proteins, it is likely that a small portion of free fatty acids in equilibrium with the bound fatty acid pool is responsible for most biological effects and drives their incorporation into membrane phospholipids by enzymatic esterification mechanisms.36 Plasma concentrations of oleate under conditions of high olive oil consumption are likely to be fully in the range of concentrations exerting biological effects in our system, likely between 10 and 100 μmol/L. We currently report biological effects demonstrable with oleate concentrations as low as 10 μmol/L, provided that incubation is performed for times long enough to ensure adequate incorporation. Of note, our biological system also contained binding proteins (albumin among others) because of the relatively high content of serum used (15% vol/vol). Therefore, the same binding to plasma proteins that occurs in vivo after intake of olive oil–containing meals and the subsequent increase in the concentrations of free oleate in the free fatty acid plasma pool are likely to occur and be reproduced by our in vitro experimental system. The second line of consideration, pointing to the plausibility of the reported effect as an explanation for the benefits of oleate on atherogenesis, is based on the relative selectivity of oleate incorporation in total cellular lipid pools. On prolonged incubation, oleate selectively displaced saturated fatty acids, thereby decreasing their relative proportion to an extent virtually equal to the magnitude of oleate incorporation. Contrary to this result, the polyunsaturated fatty acid pool, taken as the sum of the most abundant representatives of the n-3 and n-6 fatty acid classes, was almost totally unaffected by oleate. As an overall consequence of the selective increase in oleate concentration of the medium, the unsaturation index, calculated from the relative proportions of all main unsaturated and saturated fatty acids in the lipid pool and the number of double bonds present in each fatty acid, increased. This contrasts with what we previously reported for the incorporation of the n-3 polyunsaturated fatty acid DHA, which displaces both saturated and other polyunsaturated fatty acids from membrane lipids.17 The nonrandom distribution of oleate during its incorporation suggests the existence of metabolic mechanisms governing specific fatty acid incorporation and allowing precise directioning of selected fatty acid species into relatively segregated pools. Implications of this selectivity would be that the increase in oleate content in the diet would be reflected in a relatively selective replacement of saturated fatty acids, which are biologically “inert” with respect to endothelial activation, with a biologically active unsaturated fatty acid component. Also, this would imply that coadministration of oleate and of higher unsaturated fatty acid species could be additive and not mutually contrasting with respect to the biological effects herein described. Research on the mechanisms of this selectivity of incorporation and analysis of specific fatty acid phospholipid pools where oleate becomes mostly incorporated therefore appear warranted. In addition, research on the intimate molecular mechanism by which the presence of a single double bond in the structure of oleate may cause—directly or through an as-yet-unknown metabolite(s)—the reported attenuation of NF-κB activation and of the endothelial responses to cytokines (possibly by interference with the generation of reactive oxygen species) appears to be of considerable interest.
The entirety of this work was carried out at the Lecce Section of the CNR Institute of Clinical Physiology and at the Institute of Physiology at the University of Lecce, Italy. Minor support from the National Research Council and Ministero dell’Università e della Ricerca Scientifica (60%) is acknowledged. The authors also express gratitude to Fondazione Rico Semeraro—Lecce for fellowship support to Dr M.A. Carluccio and to the International Association for Post-Graduate Education for the purchase of disposables. The authors also wish to express gratitude to the Department of Obstetrics and Gynecology at the Vito Fazzi Hospital in Lecce for the supply of umbilical cords; Dr Giovan Battista Lobreglio, Laboratory of Clinical Chemistry, Vito Fazzi Hospital, Lecce, for the use of flow-cytometry apparatus; Dr Riccardo Guzzardi, Head, Lecce Section of the CNR Institute of Clinical Physiology; and to those many who in general have supported the establishment of an Atherosclerosis Research Laboratory in Lecce. The authors are also grateful to Dr Michael A. Gimbrone, Jr, Brigham and Women’s Hospital, and Dr Arnold Freedman, Dana Farber Cancer Institute, Boston, Mass, for generous gifts of the antibodies used in these studies, and to Dr Peter Libby, Brigham and Women’s Hospital, Boston, Mass, for generous gifts of IL-1α.
Keys A, Menotti A, Karvonen M, Aravanis C, Blackburn H, Buzina R, Djordjevic B, Dontas A, Fidanza F, Keys M. The diet and 15-year death rate in the Seven Countries Study. Am J Epidemiol. 1986;124:903–915.
Willett W. Diet and coronary heart disease. Monogr Epidemiol Biostat. 1990;15:341–379.
World Health Organization. Diet, nutrition, and the prevention of chronic diseases: report of WHO Study Group. Geneva, Switzerland: World Health Organization; 1990:1–75.
Assmann G, de Backer G, Bagnara S, Betteridge J, Crepaldi G, Fernandez-Cruz A, Godfredsen J, Jacotot B, Paoletti R, Renaud S, Ricci G, Rocha E, Trautwein E, Urbinati GC, Varela G. International consensus statement: olive oil and the Mediterranean diet: implications for health in Europe. Proceedings of an International Meeting of the European Commission. Rome, Italy: April 11, 1997. 1997:1–3.
Mensink R, Katan M. Effect of dietary fatty acids on serum lipids and lipoproteins: a meta-analysis of 27 trials. Arterioscler Thromb. 1992;12:911–919.
Gardner C, Kraemer H. Monounsaturated versus polyunsaturated dietary fat and serum lipids: a meta-analysis. Arterioscler Thromb Vasc Biol. 1995;15:1917–1927.
Riemersma R, Wood D, Butler S, Elton R, Oliver M, Salo M, Nikkari T, Vartiainen E, Puska P, Gey F. Linoleic acid content in adipose tissue and coronary heart disease. BMJ. 1986;292:1423–1427.
Espino-Montoro A, Lopez-Miranda J, Castro P, et al. Monounsaturated fatty acid enriched diets lower plasma insulin levels and blood pressure in healthy young men. Nutr Metab Cardiovasc Dis. 1996;6:147–154.
Garg A, Grundy S, Unger R. Comparison of effects of high and low carbohydrate diets on plasma lipoprotein and insulin sensitivity in patients with mild NIDDM. Diabetes. 1992;41:1278–1285.
Griffin M, Dimitriadis E, Lenehan K, et al. Non-insulin-dependent diabetes mellitus: dietary monounsaturated fatty acids and low-density lipoprotein composition and function. Q J Med. 1996;89:211–216.
Bosello O, Armellini F, Zamboni M. Obesity. In: Spiller G, ed. The Mediterranean Diets in Health and Disease. New York, NY: Van Nostrand Reinhold; 1991:256–276.
Gimbrone MA Jr, Kume N, Cybulsky MI. Vascular endothelial dysfunction and the pathogenesis of atherosclerosis. In: Weber P, Leaf A, eds. Atherosclerosis Reviews. New York, NY: Raven Press; 1993:1–9.
Harris W. Fish oils and plasma lipid and lipoprotein metabolism in humans: a critical review. J Lipid Res. 1989;30:785–807.
De Caterina R, Cybulsky MI, Clinton SK, Gimbrone MA Jr, Libby P. The ω-3 fatty acid docosahexaenoate reduces cytokine-induced expression of pro-atherogenic and pro-inflammatory proteins in human endothelial cells. Arterioscler Thromb. 1994;14:1829–1836.
De Caterina R, Cybulsky MI, Clinton SK, Gimbrone MA Jr, Libby P. Omega-3 fatty acids and endothelial leukocyte adhesion molecules. Prostaglandins Leukot Essent Fatty Acids. 1995;52:192–195.
Weber C, Erl W, Pietsch A, Danesch U, Weber PC. Docosahexaenoic acid selectively attenuates induction of vascular cell adhesion molecule-1 and subsequent monocytic cell adhesion to human endothelial cells stimulated by tumor necrosis factor-α. Arterioscler Thromb Vasc Biol. 1995;15:622–628.
De Caterina R, Bernini W, Libby P. Fatty acid structural requirements for the inhibition of endothelial activation. Circulation. 1996;94(suppl. I):I-102.
Gimbrone MA Jr. Culture of vascular endothelium. In: Spaet T, ed. Progress in Hemostasis and Thrombosis. Vol 3. New York, NY: Grune & Stratton; 1976:1–28.
Thornton S, Muller S, Levine E. Human endothelial cells: use of heparin in cloning and long term serial cultivation. Science. 1983;222:623–625.
Maciag T, Cerundolo J, Ilsley J, Kelley P, Forand R. An endothelial cell growth factor from bovine hypothalamus: identification and partial characterization. Proc Natl Acad Sci U S A. 1976;76:5674–5678.
Warner SJC, Friedman GB, Libby P. Regulation of major histocompatibility gene expression in cultured human vascular smooth muscle cells. Arteriosclerosis. 1989;9:279–288.
Pober J, Gimbrone MA Jr, Lapierre L, Mendrick D, Fiers W, Rothlein R, Springer T. Overlapping patterns of activation of human endothelial cells by interleukin 1, tumor necrosis factor, and immune interferon. J Immunol. 1986;137:1893–1896.
Spertini O, Luscinskas FW, Gimbrone MA Jr, Tedder T. Monocyte attachment to activated human vascular endothelium in vitro is mediated by leukocyte adhesion molecule-1 (L-selectin) under nonstatic conditions. J Exp Med. 1992;175:1789–1792.
Schleimer R, Sterbinsky S, Kaiser J, Bickel C, Klunk D, Tomioka K, Newman W, Luscinskas FW, Gimbrone MA Jr, McIntyre W, Bochner B. Interleukin-4 induces adherence of human eosinophils and basophils but not neutrophils to endothelium. J Immunol. 1992;148:1086–1092.
Dignam J, Lebovitz R, Roeder R. Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei. Nucleic Acids Res. 1983;11:1475–1489.
Iademarco M, McQuillan J, Rose G, Dean D. Characterization of the promoter for vascular cell adhesion molecule-1 (VCAM-1). J Biol Chem. 1992;267:16323–16329.
Morrison W, Smith L. Preparation of fatty acid methyl esters and dimethylacetals from lipids with boron fluoride-methanol. J Lipid Res. 1964;5:600–608.
Vance DE, Vance JE. Biochemistry of Lipids and Membranes. New York, NY: Benjamin/Cummings; 1985:1–220.