Myristic Acid Stimulates Endothelial Nitric-oxide Synthase in a CD36- and an AMP Kinase-dependent Manner*

Dietary free fatty acids have been reported to have various effects on the endothelium including the generation of nitric oxide. The goal of the current study was to determine the mechanism whereby free fatty acid causes an increase in nitric oxide synthesis. The specific hypothesis tested was that free fatty acid association with CD36, a class B scavenger receptor, induces the activation of endothelial nitric-oxide synthase (eNOS). A human microvascular endothelial cell line and a transfected Chinese hamster ovary cell system were used to determine which free fatty acids stimulate eNOS. Surprisingly, only myristic acid, and to a lesser extent palmitic acid, stimulated eNOS. The stimulation of eNOS was dose- and time-dependent. Competition experiments with other free fatty acids and with a CD36-blocking antibody demonstrated that the effects of myristic acid on eNOS required association with CD36. Further mechanistic studies demonstrated that the effects of myristic acid on eNOS function were not dependent on PI 3-kinase, Akt kinase, or calcium. Pharmacological studies and dominant negative constructs were used to demonstrate that myristic acid/CD36 stimulation of eNOS activity was dependent on the activation of AMP kinase. These data demonstrate an unexpected link among myristic acid, CD36, AMP kinase, and eNOS activity.

Dietary fatty acids are involved in the development of numerous cardiovascular diseases; however, the reported molecular mechanisms and the physiological end points are often controversial. The controversy is in part because dietary fatty acids affect multiple systems (e.g. liver, lipoproteins, endothelium, macrophages) and often have synergistic and/or antagonistic influences on the development of disease and because different fatty acids have different effects (1)(2)(3)(4). In the present study we focused on the effects of free fatty acids on endothelial nitric-oxide synthase (eNOS) 1 activity. Few studies have exam-ined the link between free fatty acids and eNOS activity. However, numerous studies have demonstrated that acylation of eNOS with myristic acid and palmitic acid can alter the subcellular localization of eNOS with subsequent profound effects on eNOS activity (5)(6)(7)(8). A recent study by Esenabhalu et al. (9) showed that overloading endothelial cells with oleic acid caused an attenuation of calcium signaling, which resulted in a decrease in nitric oxide production. However, Lynn et al. (10) demonstrated that lower concentrations of oleic acid did not affect nitric oxide production. Interestingly, Lynn et al. also demonstrated that eicosapentaenoic acid did increase insulinstimulated nitric oxide production (10). These findings suggest that both the concentration and type of fatty acid can influence the activity state of eNOS and the subsequent production of nitric oxide.
The mechanism(s) of how extracellular free fatty acid influences eNOS activity is not understood. One possibility is through interaction with CD36, a class B scavenger receptor that is expressed on the surface of numerous cells, including macrophage, adipocytes, platelets, cardiac myocytes, and endothelial cells (11)(12)(13)(14). CD36 performs different functions in different cells. For instance, CD36 mediates long chain fatty acid uptake in muscle and fat (15)(16)(17), whereas in macrophage CD36 is in involved in the accumulation of modified lipoproteins (12) and in signal transduction, particularly activation of Src-like kinases (18,19). The role of CD36 in endothelial cells is less well defined, although it appears to be involved in angiogenesis (20), a process known to involve nitric oxide generation (21)(22)(23). Interestingly, CD36 and eNOS are both localized to caveolae in endothelial cells, which may provide a localized and specific signaling environment for fatty acids.
Endothelial nitric-oxide synthase has a surprisingly large number of regulatory mechanisms for controlling the synthesis of nitric oxide. An increase in intracellular calcium followed by calcium-calmodulin binding to eNOS is a major mechanism for stimulating nitric oxide generation. Ceramide (24), subcellular localization (5), acylation (7,8), chaperone proteins (25), cofactor availability (26), and cholesterol (27) have all been shown to modulate eNOS activity. Numerous studies have also demonstrated that eNOS can be activated by direct phosphorylation at serine 1179 (28). Activation of the PI 3-kinase/Akt kinase pathway will induce eNOS phosphorylation and activation (28). In addition, activation of AMP kinase can result in eNOS phosphorylation and activation (29,30).
AMP kinase is a major regulator of cellular energy metabolism. AMP kinase is activated by phosphorylation when the ATP levels in the cell decrease and the AMP levels rise. Upon phosphorylation and activation, AMP kinase phosphorylates numerous downstream targets, many of which are involved in restoring ATP levels, such as acetyl-CoA carboxylase (ACC), hydroxymethylglutaryl-CoA reductase, glycogen synthase, and eNOS (31). The exact mechanisms responsible for AMP kinase phosphorylation are not clear. A recent study by Thors et al. (32) demonstrated that thrombin and histamine stimulation of AMP kinase and subsequent activation of eNOS was independent of the PI 3-kinase/Akt kinase pathway. In contrast, Ouchi et al. (33) demonstrated that adiponectin-induced eNOS activation required cross-talk between AMP kinase and the PI 3-kinase/Akt kinase pathway. Interestingly, Zou et al. (34) have shown that peroxynitrite can stimulate AMP kinase and eNOS phosphorylation via a Src kinase-mediated mechanism.
In this study, we investigated the effect of free fatty acids on eNOS activity. In addition, we defined a novel signaling pathway involving free fatty acid association with CD36 which resulted in the activation of AMP kinase and the subsequent activation of eNOS.
Cell Culture-Human microvascular endothelial cells were cultured in M199 medium supplemented with 100 units/ml penicillin/streptomycin, 0.5% (v/v) L-glutamine, BME vitamin mix (1 ml/100 ml of M199), BME amino acid mix (1 ml/100 ml of M199), and 10% (v/v) fetal bovine serum. On day 0, 5,000 cells were placed into 12-well plates and used on day 3 at ϳ60% confluence. For CHO cell lines, human CD36 cDNA and human eNOS cDNA were subcloned into pLNCX2. The constructs were transfected into CHO cells, and cell lines stably expressing the mutants were obtained by G418 selection as we described previously (24). CHO cell lines stably expressing vector (pLNCX2), CD36, eNOS, or both CD36 and eNOS were established as described previously (24). The cells were cultured in Ham's F-12 medium containing 5% fetal bovine serum, 2 mM L-glutamine, 100 units/ml penicillin, 100 g/ml streptomycin, and 0.3 mg/ml G418.
Preparation of Fatty Acid-Albumin Complex-Fatty acids were delivered to endothelial cells in a complex with fatty acid-free sheep serum albumin. To generate the fatty acid-albumin complex, fatty acids were added slowly to Kreb-Ringer buffer containing 10 mM HEPES, pH 7.4, and sheep albumin as described previously (35). Fatty acids were added from concentrated stock (200 mM in ethanol). Previous studies have demonstrated that the ratio of 0.5 fatty acids to albumin is sufficient to deliver free fatty acids to cells without causing nonspecific perturbation of the plasma membrane.
Activation of NOS in Intact Cells-NOS activation was determined in intact cells as described previously (36). Briefly, the cells were plated into 12-well plates at 5,000 cell/well and grown to 60% confluence. The medium was replaced with serum-free medium for 16 h and then placed for 2 h in phosphate-buffered saline, at 37°C. After the preincubation period, the phosphate-buffered saline was removed from the wells and replaced with 400 l of phosphate-buffered saline containing 0.75 Ci/ml L-[ 3 H]arginine and the indicated treatments. The cells were incubated at 37°C for 30 min. The NOS reaction was terminated by adding 500 l of ice-cold 1 N trichloroacetic acid to each well. The cells were freeze-fractured twice in liquid nitrogen for 2 min with thawing at 37°C for 5 min and scraped with a rubber spatula. The contents of each well were transferred to ice-cold glass test tubes. Ether extraction was performed three times with water-saturated ether to remove the trichloroacetic acid. The samples were neutralized with 1.5 ml of 25 mM HEPES, pH 8, applied to Dowex AG 50WX-8 (Tris form) columns, and eluted with 1 ml of 40 mM HEPES buffer, pH 5.5, containing 2 mM EDTA and 2 mM EGTA. L-[ 3 H]Citrulline was collected in scintillation vials and quantified by liquid scintillation counting. In individual experiments performed in 12-well plates, 3 wells were used for each treatment group. Findings were confirmed in at least six independent experiments. NOS activation in the intact cells was inhibited by 1 mM N-nitro-L-arginine (L-NNA). To ensure that the treatments did not affect the loading of the cells with L-[ 3 H]arginine, the amount of L-[ 3 H]arginine associated with the cells was determined. Electrophoresis and Immunoblots-Samples were concentrated by trichloroacetic acid precipitation and washed in acetone. Pellets were suspended in sample buffer that contained 1.2% (v/v) ␤-mercaptoethanol and heated at 95°C for 3 min before being loaded onto gels. Proteins were separated in a 12.5% SDS-polyacrylamide gel using the method of Laemmli (37). The separated proteins were then transferred to PVDF membrane. The membrane was blocked in TBS that contained 5% dry milk for 1 h at room temperature. Primary antibodies were diluted in TBS that contained 1% dry milk and incubated with the membrane for 1 h at room temperature. The membrane was washed four times, 10 min each in TBS ϩ 1% dry milk. The secondary antibodies (all conjugated to horseradish peroxidase) were diluted 1/20,000 in TBS ϩ 1% dry milk and incubated with the membrane for 1 h at room temperature. The membrane was then washed and the bands visualized by chemiluminescence.
Statistical Analysis-Least squares analysis of variance was used to evaluate the data with respect to treatment, time, and their interaction using the analysis of variance procedure of Statistica. When appropriate, treatments were compared within a given time using the Tukey's honestly significant difference test. Means were considered significant at p Ͻ 0.01.

Effect of Free Fatty Acids on eNOS Activity-Human microvascular endothelial (HME) cells that endogenously express CD36 and eNOS and CHO cells expressing exogenous CD36
and eNOS (Fig. 1A) were used to determine which free fatty acids stimulated the generation of nitric oxide. Confluent HME cells and CHO cells were incubated with 10 M concentrations of the indicated free fatty acids and 0.75 Ci of [ 3 H]arginine for 30 min at room temperature ( Fig. 1, B-F). Additional sets of cells were treated with buffer only (negative control) or with 2 g/ml ionomycin, a calcium ionophore (positive control). Finally, each experimental treatment also included duplicate samples containing 1 mM L-NNA, an eNOS inhibitor, to determine whether the generated nitric oxide was the result of eNOS activity. Fig. 1, C, D, and E, show that control cells lacking eNOS, CD36, or both proteins did not generate nitric oxide in response to fatty acids. Surprisingly, both HME cells ( Fig. 1B) and CHO cells ( Fig. 1F) that contained eNOS and CD36 responded to myristic acid with an 8 -10-fold increase in nitric oxide production. The positive control, ionomycin, caused a 2-3-fold increase in nitric oxide production. In addition, palmitic acid induced a small but significant increase in nitric oxide. Incubation of cells with shorter (lauric acid) or longer (stearic acid, oleic acid, linoleic acid) chain fatty acids did not stimulate eNOS above basal levels.
To determine whether the eNOS stimulatory effect of myristic acid was dose-dependent, different concentrations of myristic acid (0.5 fatty acid:albumin ratio maintained) were added to HME cells along with 0.75 Ci of [ 3 H]arginine for 30 min at room temperature. Fig. 2A demonstrates that myristic acid caused a dose-dependent increase in eNOS activity with a maximum stimulatory dose of ϳ25 M. Importantly, increasing concentrations of other fatty acids did not significantly increase eNOS activity above basal levels (data not shown). To determine whether the effect of myristic acid was time-dependent, 10 M myristic acid was added to HME cells along with 0.75 Ci of [ 3 H]arginine for 0 -30 min at room temperature. Fig. 2B shows that maximal eNOS activity was achieved ϳ20 min after the addition of myristic acid. Finally, control experiments were conducted to determine whether myristic acid affected the uptake of [ 3 H]arginine. HME cells were treated with 0.75 Ci of [ 3 H]arginine and 10 M myristic acid or buffer for 2 min or 20 min in the presence of 1 mM L-NNA. The cells were washed, and the amount of [ 3 H]arginine associated with the cells was determined. Fig. 2C demonstrates that the amount of [ 3 H]arginine associated with the cells was maximal by 2 min and was not affected by the presence of myristic acid.
We next determined whether the large increase in eNOS activity, compared with the ionomycin control, was unique to myristic acid or whether other eNOS agonists also induced the same magnitude of activation. HME cells were incubated with [ 3 H]arginine and myristic acid as described above. Additional sets of cells were treated with 1 M acetylcholine, 1 M bradykinin, or 2 g/ml ionomycin. Fig. 3 demonstrates that acetylcholine, bradykinin, and ionomycin all stimulated eNOS to the same extent. The addition of myristic acid to acetylcholine, bradykinin, and ionomycin did not enhance the stimulation caused by myristic acid alone. These data suggest that myristic acid maximally stimulates eNOS.
Association with CD36 Is Required for Myristic Acid to Increase eNOS Activity-The CHO cell lines used in Fig. 1 strongly suggest that CD36 is necessary for myristic acid to stimulate eNOS activity. To determine further whether the effects of myristic acid were dependent upon CD36, HME cells were pretreated for 30 min at 37°C with 20 g/ml of an established CD36-blocking antibody or 20 g/ml of a nonrelevant, isotyped matched antibody (38,39). The cells were then incubated with 10 M myristic acid and 0.75 Ci of [ 3 H]arginine for 30 min at room temperature. Control cells that were incubated with myristic acid but did not receive any antibodies were stimulated 8 -10-fold over cells not receiving myristic acid (Fig.  4A). The addition of a nonrelevant antibody had no effect on the ability of myristic acid to stimulate eNOS (Fig. 4A). In contrast, preincubation of cells with a CD36-blocking antibody inhibited myristic acid-induced eNOS activity (Fig. 4A).

FIG. 1. Myristic acid stimulates the production of nitric oxide in human microvascular endothelial cells and CHO cells.
A, Western blots demonstrating that eNOS and CD36 are expressed in HME cells but not CHO cells. CHO cells transfected with cDNAs encoding CD36 and/or eNOS were screened for clones expressing levels of CD36 and eNOS similar to HME cells. Protein (10 g) from the selected cells was resolved by SDS-PAGE, transferred to PVDF, and probed with the indicated antibodies. The Western blots were developed by the method of chemiluminescence (1-2-min exposures). The data are representative of three independent experiments. B-F, effect of free fatty acids on eNOS activity in the different cell lines. The cells were incubated with 0.75 Ci/ml [ 3 H]arginine, a 10 M concentration of the indicated free fatty acid, or 2 g/ml ionomycin for 30 min at 25°C. The cells were processed to quantify the amount of citrulline generated. Each experiment included controls using 1 mM L-NNA to demonstrate that the generated citrulline was the result of eNOS activity. The data are from six independent experiments, with triplicate measurements in each experiment, mean Ϯ S.E., p Ͻ 0.01. Note, fatty acid concentrations up to 100 M and fatty acid:albumin ratios up to 1. To confirm further that the effect of myristic acid on eNOS activity requires association with CD36, a competition assay was performed using stearic acid. Stearic acid did not stimulate eNOS activity (Figs. 1 and 2) but will compete with myristic acid for association with CD36. Myristic acid (10 M) was added to HME cells along with various concentrations of stearic acid. [ 3 H]Arginine was then added and the cells incubated for 30 min before quantifying the production of [ 3 H]citrulline. Fig. 4B demonstrates that increasing concentrations of stearic acid compete with the ability of myristic acid to stimulate eNOS.
Mechanism of Myristic Acid-induced Stimulation of eNOS-We next focused on the mechanism whereby myristic acid association with CD36 stimulated eNOS. Endothelial nitricoxide synthase can be stimulated by several mechanisms. The most common mechanism is by an increase in intracellular calcium (40). To determine whether myristic acid stimulated eNOS via a calcium-dependent mechanism we first loaded the cells with the calcium chelator BAPTA-AM (10 M, 30 min) and then added 10 M myristic acid to the cells. Fig. 5A illustrates that loading the cells with BAPTA-AM did not affect the ability of myristic acid to stimulate the production of nitric oxide. However, BAPTA-AM completely inhibited the stimulatory effects of the calcium ionophore, ionomycin (Fig. 5A). To confirm further the activity data we examined the phosphorylation state of eNOS. Phosphorylation of serine 1179 in eNOS increases eNOS activity (28). Fig. 5B demonstrates that treatment with myristic acid did not change the total amount of eNOS in HME cells, but it did dramatically increase the phosphorylation of eNOS. Importantly, and consistent with the activity data, BAPTA-AM did not affect the ability of myristic acid to induce the phosphorylation of eNOS.
Because myristic acid induced the phosphorylation of eNOS we investigated the possible involvement of the PI 3-kinase/Akt kinase pathway, which has been demonstrated to cause the phosphorylation of eNOS (28). Two methods were used to determine whether Akt kinase was involved in myristic acidinduced stimulation of eNOS. First, HME cells were pretreated with 100 nM wortmannin, a PI 3-kinase inhibitor, and then incubated with [ 3 H]arginine and myristic acid as described above. Fig. 6A demonstrates that wortmannin did not inhibit myristic acid-induced stimulation of eNOS. Active Akt kinase is phosphorylated, so to test further the possible involvement of Akt kinase, we used commercially available antibodies to determine whether myristic acid induced Akt kinase phosphorylation. Fig. 6B shows that the positive control, platelet-derived growth factor, stimulated Akt kinase phosphorylation and that wortmannin prevented platelet-derived growth factor-induced phosphorylation of Akt kinase. Myristic acid did not induce Akt kinase phosphorylation even at high concentrations.
We next determined whether myristic acid stimulates AMP kinase because AMP kinase has been demonstrated to activate eNOS by direct phosphorylation of the enzyme (29). To test the potential involvement of AMP kinase, HME cells were treated with myristic acid, myristic acid plus H89, and myristic acid plus R p -cAMPS. H89 is an inhibitor of protein kinase A and AMP kinase, whereas R p -cAMPS inhibits protein kinase A but not AMP kinase (32). The effect of the pharmacological reagents on eNOS activity was measured as described above. In addition, the effects of the inhibitors on eNOS phosphorylation, AMP kinase phosphorylation, and ACC phosphorylation (an AMP kinase substrate) were determined by Western blots with phospho-specific antibodies. Myristic acid induced eNOS activity, phosphorylation of eNOS, phosphorylation of AMP kinase, and phosphorylation of ACC (Fig. 7). H89 prevented eNOS activity, eNOS phosphorylation, and ACC phosphorylation (Fig. 7). H89 did not affect AMP kinase phosphorylation, as shown previously (32). Importantly, R p -cAMPS, which inhibits protein kinase A but not AMP kinase, did not inhibit eNOS activity, eNOS phosphorylation, or ACC phosphorylation (Fig. 7).
To establish further the role of AMP kinase, a dominant negative AMP kinase adenoviral construct and a control adenoviral construct were used with the HME cells. The cells were treated with the adenoviral constructs as described previously (41). The cells were then loaded with [ 3 H]arginine, treated with myristic acid, and processed to measure eNOS activity as described above. Cells that did not receive virus and cells that received control virus both responded to treatment with myristic acid by increasing the production of nitric oxide (Fig. 8). However, cells that received the dominant negative AMP kinase virus did not generate nitric oxide in response to myristic acid (Fig. 8). Importantly, cells treated with the dominant negative AMP kinase virus did respond to ionomycin (Fig. 8), indicating that eNOS was still responsive to calciuminduced stimulation.
Because CD36 has been shown to stimulate Src kinases (19) and because AMP kinase can be activated by Src kinases (42), we determined whether myristic acid stimulates AMP kinase via a Src kinase. HME cells were pretreated for 15 min at 37°C with different Src kinase inhibitors: 10 M genistein, 0.1 g/ml herbimycin A, or 10 M 4-amino-5-(4-chlorophenyl)-7-(t-butyl)pyrazolo [3,4-d]pyrimidine. The cells were then incubated with 10 M myristic acid and 0.75 Ci of [ 3 H]arginine for 30 min at room temperature. Additional sets of cells were treated with buffer only or with 2 g/ml ionomycin. Each experiment also included duplicate samples containing 1 mM L-NNA. Fig. 9 demonstrates that the three different Src kinase inhibitors did not prevent myristic acid-induced stimulation of eNOS. Higher concentrations of the Src kinase inhibitors and longer or shorter pretreatment times did not affect myristic acid-induced stimulation of eNOS (data not shown). Finally, the addition of myristic acid did not alter the AMP:ATP ratio in the cells, suggesting that myristic acid did not stimulate AMP kinase through alterations in energy metabolism (data not shown). DISCUSSION We have defined a novel signaling pathway that couples exogenous free fatty acid to eNOS stimulation. Free fatty acid stimulation of eNOS required association with CD36 and activation of AMP kinase. One of the most striking findings in the current study is the remarkable specificity with regard to which free fatty acids stimulated eNOS. Of the fatty acids examined, myristic acid, was the most effective at stimulating eNOS, whereas palmitic acid caused a small but statistically significant increase in eNOS activity. Surprisingly, lauric acid, oleic acid, stearic acid, and linoleic acid were incapable of stimulating eNOS. Experiments with blocking antibodies, CD36-minus cell lines, and competition assays clearly demonstrated that the effect of myristic acid depended on the presence of CD36.
CD36 is a fatty acid-binding protein and is involved in the uptake of long chain fatty acids, although long chain fatty acid uptake is not absolutely dependent on the presence of CD36 (15)(16)(17). In addition, CD36 can interact with and facilitate the uptake of short chain fatty acids; however, short chain fatty acids cross the plasma membrane at a sufficiently high rate that CD36 does not appear to contribute significantly to this process (43). The molecular mechanisms and regulation involved in CD36-mediated fatty acid uptake are not completely understood. Recently, Bonen et al. (44) demonstrated that in myocytes, contraction and insulin caused an intracellular pool of CD36 to translocate to the cell surface, which caused an increase in fatty acid uptake without synthesis of new CD36. In addition, Pohl et al. (17) have shown that CD36-dependent uptake of long chain fatty acids in 3T3-L1 adipocytes requires the presence of intact lipid rafts. In endothelial cells, caveolae (a type of lipid raft) contain CD36 and eNOS and may form a functional signaling complex that couples exogenous fatty acids to eNOS activation.
Our current data do not distinguish conclusively between myristic acid having an effect after binding to CD36 or after uptake into cells. We speculate that the effects of myristic acid are the result of association with CD36 and not CD36-mediated uptake. This speculation is based on the following. First, myristic acid does not require CD36 for maximal uptake into the cells (43); that is, the presence or absence of CD36 does not significantly alter the amount of myristic acid associated with the cells. Therefore, the effects of myristic acid are absolutely dependent on the presence of CD36. Second, CD36-blocking antibodies and competition studies with stearic acid prevented the stimulatory effects of myristic acid. These reagents will not affect the amount of myristic acid taken up by the cell, again because myristic acid moves across the membrane independently of CD36 (43). Third, activation of eNOS is specific for myristic acid, which argues against membrane perturbations and detergent-like effects of excessive fatty acid accumulation. However, it is feasible that CD36 mediates the uptake and delivery of a small amount of myristic acid to caveolae where it may possibly have direct effects on AMP kinase and/or eNOS.
We think it is more likely that myristic acid association with CD36 activates a signaling cascade that results in AMP kinase activation and subsequently eNOS activation. Recently, Medeiros et al. (18) have shown that in mouse macrophages apolipoprotein C-II can activate the Src-like kinase, Lyn, in a CD36dependent manner. In addition, the activation of Lyn resulted in p44/p42 mitogen-activated protein kinase activation and ultimately an inhibition in the expression of tumor necrosis factor-␣ (19). Macrophages isolated from CD36 null mice did not respond to apolipoprotein C-II treatment. In a different study, Lee et al. (45) demonstrated that caveolin-1 and c-Src are functionally coupled in caveolae. The available data sug- FIG. 6. Myristic acid stimulation of eNOS is independent of Akt kinase. A, the indicated cells were pretreated with 100 nM wortmannin for 30 min at 25°C. The cells were then incubated with 0.75 Ci/ml [ 3 H]arginine, buffer only, or the indicated concentrations of myristic acid for 30 min at 25°C. The cells were then processed to quantify the amount of citrulline generated. Each experiment included controls using 1 mM L-NNA to demonstrate that the generated citrulline was the result of eNOS activity (data not shown). The data are from three independent experiments, with triplicate measurements in each experiment, mean Ϯ S.E. B, the cells were incubated with 67 ng/ml platelet-derived growth factor (PDGF), 100 nM wortmannin (pretreated 30 min), or 10 M myristic acid for 30 min at 25°C. The cells were washed, lysates generated, and cellular proteins resolved by SDS-PAGE (10 g). The separated proteins were transferred to PVDF and Western blotted with actin and phospho-Akt antibodies. The Western blots were developed by the method of chemiluminescence (1-2-min exposures). The data are representative of three independent experiments.
gest that caveolae and caveolin-1 play a role in coupling CD36dependent stimulation of Src kinases. Intriguingly, a study by Zou et al. (41) demonstrated that peroxynitrite can stimulate AMP kinase via a c-Src-dependent mechanism. The studies described above along with our data provide a plausible mechanism to couple myristic acid/CD36 to AMP kinase activation. Considerable additional studies will be required to address the involvement of caveolae/lipid rafts because methods to disrupt caveolae and lipid rafts, such as filipin, are nonspecific in that they cause major membrane perturbations, relocation of caveolin and eNOS, and may not always disrupt caveolae functions (46). Furthermore, the use of small interfering RNA to eliminate caveolin-1 is problematic because caveolin-1 is known to regulate eNOS function directly, independent of CD36 and myristic acid. Elucidating the possible role of caveolae in this process is important and will require numerous approaches to generate a reliable conclusion.
Studies using the calcium chelator BAPTA-AM demonstrated that myristic acid did not stimulate eNOS by causing an increase in intracellular calcium. In addition, we used the PI 3-kinase inhibitor, wortmannin, and phospho-Akt antibodies to rule out the involvement of the PI 3-kinase/Akt kinase activation pathway for eNOS. Because of the link among CD36, Src, and AMP kinase and because AMP kinase has been shown to activate eNOS through phosphorylation of serine 1179 in eNOS we used two different methods to test the possible involvement of AMP kinase. The first method used pharmacological reagents. The compound H89 is an inhibitor of both protein kinase A and AMP kinase, whereas R p -cAMPS inhibits protein kinase A but not AMP kinase. H89 inhibited myristic acidinduced phosphorylation of eNOS, eNOS activity, and AMP kinase-induced phosphorylation of ACC. ACC is a specific substrate for AMP kinase and was used as a marker for AMP kinase activity in these studies (32). As reported previously, H89 did not alter AMP kinase phosphorylation even though it inhibited AMP kinase enzymatic activity (32). Importantly, R p -cAMPS did not inhibit eNOS phosphorylation, eNOS activity, or ACC phosphorylation, demonstrating that protein kinase A was not responsible for the phosphorylation events. To confirm these findings further we used a dominant negative inhibitor of AMP kinase under an adenovirus promoter (41) to  8. A dominant negative AMP kinase prevents myristic acid stimulation of eNOS. HME cells were infected with an adenovirus construct containing dominant negative AMP kinase or empty control adenovirus at a multiplicity of infection of 50 for 24 h as described previously (41). The cells were then incubated with 0.75 Ci/ml [ 3 H]arginine, buffer only, 10 M myristic acid, or 2 g/ml ionomycin for 30 min at 25°C. The cells were processed to quantify the amount of citrulline generated. Each experiment included controls using 1 mM L-NNA to demonstrate that the generated citrulline was the result of eNOS activity. The data are from three independent experiments, with triplicate measurements in each experiment, mean Ϯ S.E., p Ͻ 0.01. The adenoviral infections did not significantly increase cell death (data not shown). Black bars, the indicated treatment; open bars, the indicated treatment ϩ L-NNA.
FIG. 9. Src kinase inhibitors do not prevent myristic acidinduced stimulation of eNOS. HME cells were pretreated for 15 min at 37°C with different Src kinase inhibitors: 10 M genistein, 0.1 g/ml herbimycin A, or 10 M 4-amino-5-(4-chlorophenyl)-7-(t-butyl)pyrazolo [3,4-d]pyrimide. Cells were then incubated with 10 M myristic acid and 0.75 Ci of [ 3 H]arginine for 30 min at room temperature. Additional sets of cells were treated with 2 g/ml ionomycin or buffer. Duplicate samples also contain L-NNA. The data are from three independent experiments with triplicate measurements in each experiment, mean Ϯ S.E. Black bars, the indicated treatment; open bars, indicated treatment ϩ L-NNA. transfect cells and examine their response to myristic acid. The cells transfected with a control adenovirus, and uninfected cells, both displayed a stimulatory response to myristic acid and ionomycin. However, the cells transfected with the dominant negative AMP kinase construct did not respond to myristic acid, further demonstrating that AMP kinase is in the signaling pathway leading to eNOS activation in these cells. It is important to emphasis that agonist-induced stimulation of AMP kinase and eNOS may vary between different types of cells. For instance, Fleming et al. (47) demonstrated that insulin stimulates eNOS via AMP kinase in platelets but not in endothelial cells. The mechanistic explanation for this difference is not known.
Myristic acid is the third most common saturated fat in the diet with ϳ8 g consumed per day in the United States (48). Saturated fatty acid increases plasma high and low density lipoprotein levels in humans (1). Of the dietary fatty acids, myristic acid causes the largest increase in plasma low density lipoprotein levels. Although an increase in high density lipoprotein levels is considered cardioprotective, an increase in low density lipoprotein levels is considered proatherogenic (1). Comparisons between the incidence of coronary heart disease and saturated fatty acid consumption across countries led to the conclusion that saturated fats were highly pro-disease. However, analysis of data from the Nurses' Health Study indicated that saturated fatty acids were associated separately with only a small increase in risk for coronary heart disease (49). Obviously, the current understanding of the role of dietary saturated fatty acids is incomplete.
Our current data demonstrate that a specific fatty acid, myristic acid, can cause the stimulation of eNOS in a CD36and AMP kinase-dependent fashion. Nitric oxide is generally considered antiatherogenic, and the increase in nitric oxide may offset the myristic acid-induced increase in plasma cholesterol. These data not only demonstrate an unexpected link between myristic acid and nitric oxide but also suggest that the type of dietary fatty acid may be as important as the absolute amount of fatty acid.