Thrombin stimulation of human coronary artery endothelial cells (HCAEC) results in activation of a membrane-associated, calcium-independent phospholipase A2 (iPLA2) that selectively hydrolyzes membrane plasmalogen phospholipids. Rupture of an atherosclerotic plaque and occlusion of the coronary vasculature results in a coronary ischemic event in which HCAEC in the ischemic area would be exposed to dramatic decreases in oxygen tension in addition to thrombin exposure. We exposed HCAEC to hypoxia in the presence or absence of thrombin stimulation and measured iPLA2 activation, membrane phospholipid hydrolysis, and the accumulation of biologically active phospholipid metabolites. HCAEC exposed to hypoxia, thrombin stimulation, or a combination of the two conditions demonstrated an increase in iPLA2 activity and an increase in arachidonic acid release from plasmenylcholine. Thrombin stimulation of normoxic HCAEC did not result in an accumulation of choline lysophospholipids, but hypoxia alone and in combination with thrombin stimulation led to a significant accumulation of lysoplasmenylcholine (LPlsCho). We propose that the presence of hypoxia inhibits LPlsCho catabolism, at least in part, as a result of the accumulation of long-chain acylcarnitines. The combination of increased production and decreased catabolism of LPlsCho is necessary for its accumulation. Pretreatment with bromoenol lactone to inhibit iPLA2 blocked membrane phospholipid hydrolysis and production of membrane phospholipid-derived metabolites. The increase in iPLA2 activity and the subsequent accumulation of membrane phospholipid-derived metabolites in HCAEC exposed to hypoxia or thrombin stimulation alone, and particularly in combination, have important implications in inflammation and arrhythmogenesis in atherosclerosis/thrombosis and subsequent myocardial ischemia.
- myocardial ischemia
sudden cardiac death in humans invariably results from malignant ventricular arrhythmias secondary to acute myocardial ischemia precipitated by the evolution of an intracoronary thrombus (3, 7, 32). We have demonstrated previously that thrombin stimulation of human coronary artery endothelial cells (HCAEC) results in hydrolysis of membrane plasmalogen phospholipids by a Ca2+-independent phospholipase A2 (iPLA2) that leads to the generation of several phospholipid metabolites that may play an important role in inflammation or arrhythmogenesis in the heart (21).
Once blood flow is interrupted or severely reduced by increased or complete occlusion of a coronary artery precipitating an ischemic event, the endothelial cells in the ischemic area would be exposed to dramatic decreases in oxygen tension in addition to exposure to thrombin. Results from studies involving release of arachidonic acid from hypoxic endothelial cells are conflicting and appear to depend on both the species and site from where endothelial cells are isolated originally (9, 27). In human umbilical vein endothelial cells, hypoxia and reoxygenation result in increased platelet-activating factor (PAF) production, indicating increased phospholipase A2 (PLA2) activity (9). Additionally, in bovine aortic endothelial cells, hypoxia results in release of free arachidonic acid from plasmenylethanolamine, suggesting that a plasmalogen-selective PLA2 is involved (27).
PLA2 are responsible for hydrolyzing the sn-2 ester bonds in membrane plasmalogen phospholipids, such as plasmenylcholine and plasmenylethanolamine, releasing a free fatty acid and a lysophospholipid. We have shown that the majority of HCAEC PLA2 activity is membrane associated, does not require calcium for activity, and demonstrates a preference for arachidonylated substrates. iPLA2 activity has been demonstrated in both the cytosol and membrane fractions of mammalian cells (2, 11, 12, 19, 28).
iPLA2 plays a critical role in the formation of inflammatory mediators arachidonic acid and PAF, which are essential in both the initiation and propagation of the inflammatory response. Once liberated from phospholipids by iPLA2, arachidonic acid is converted to prostaglandins by the action of cyclooxygenase enzymes. Liberated lysophospholipids, such as lysoplasmenycholine and lysophosphatidylcholine, can be rapidly acetylated at the sn-2 position by lyso-PAF-acetyl-CoA acetyltransferase to produce PAF. To date, no studies have examined the effects of hypoxia on human endothelial cells from an arterial origin with respect to PLA2 activation, phospholipid hydrolysis, and the production of biologically active phospholipid metabolites.
In this study, we determined whether hypoxia alone resulted in increased iPLA2 activity and membrane phospholipid hydrolysis and whether hypoxia potentiates the thrombin-stimulated changes that we have described previously.
MATERIALS AND METHODS
Cell isolation and culture.
HCAEC (Cambrex Bio Science, Walkersville, MD) were grown to confluence in MCDB 131 medium with 5% FCS, 10 ng/ml epidermal growth factor, 1 μg/mg hydrocortisone, 200 μg/ml endothelial cell growth supplement, and 90 μg/ml heparin. Cells were allowed to grow to confluence, achieving a contact-inhibited monolayer of flattened, closely apposed endothelial cells in 4–5 days.
Induction of hypoxia.
A glucose-free 1.2 mM Ca2+-HEPES buffer, pH 7.4, was degassed under vacuum for 1 h and then bubbled with 100% prepurified N2 for at least 2 h to attain a Po2 of <15 mmHg. The medium surrounding confluent HCAEC was removed and replaced with glucose-free 1.2 mM Ca2+-HEPES in a hypoxic environment in which room air had been exchanged with 100% N2. The 100% N2 atmosphere was maintained above the hypoxic solution and cells for the entire hypoxic interval.
Measurement of ATP content.
High energy phosphates were separated and quantified using an HPLC system (Waters Chromatography, Milford, MA) consisting of a model 510 HPLC pump, an injector (Rheodyne, Cotati, CA), and a model 484 tunable absorbance detector, set at a wavelength of 214 nm, interfaced to an IBM PC-AT computer using a 900 series interface and 3000 series chromatography data systems software (Nelson Analytical, Cupertino, CA). Samples (10–50 μl) were injected on a Hi-Pore RP-18 reverse-phase HPLC column (Bio-Rad) and eluted with an isocratic mobile phase of 0.1 mol/l ammonium phosphate, pH 5.5, at a flow rate of 0.5 ml/min. Under these conditions, ATP eluted from the column within 20 min. Quantification of ATP was determined by comparing the integrated peak area with a linear regression curve constructed from the integrated peak areas obtained from injecting 0.2–1.0 nmol of ATP standard (Sigma Chemical, St. Louis, MO). ATP levels were corrected for protein content for each cell culture and expressed as nanomoles per milligram of protein.
Confluent HCAEC were subjected to hypoxia with or without thrombin stimulation or thrombin stimulation alone for the allocated time intervals. For iPLA2 inhibition experiments, HCAEC were pretreated with 5 μM bromoenol lactone (BEL) before exposure to hypoxia or thrombin stimulation. Additionally, in selected experiments, 10 mM ATP was added to HCAEC before exposure to hypoxic conditions. At the end of the stimulation period, the surrounding buffer was removed and immediately replaced with ice-cold buffer containing (in mmol/l): 250 sucrose, 10 KCl, 10 imidazole, 5 EDTA, and 2 dithiothreitol with 10% glycerol, pH = 7.8. Cells were removed from the tissue culture well using a cell scraper, and the suspension was sonicated on ice. PLA2 activity was assessed by incubating 50 μg cellular protein with 100 μM (16:0, [3H]18:1) plasmenylcholine in assay buffer containing 100 mM Tris, 4 mM EGTA, and 10% glycerol, pH = 7.0 at 37°C for 5 min in a total volume of 200 μl. Specific enzyme activity was expressed as the rate of radiolabeled fatty acid production determined following separation from the labeled phospholipid substrate using TLC and liquid scintillation spectrometry with activity normalized to protein content.
Separation and quantification of individual choline and ethanolamine glycerophospholipid molecular species.
Cellular phospholipids were extracted from HCAEC by the method of Bligh and Dyer (4). The chloroform layer was dried under N2, and the lipid residue was resuspended in 1 ml of chloroform-methanol (1:1, vol/vol). Phospholipids were separated into different classes by HPLC using gradient elution with a mobile phase comprised of hexane-isopropanol-water (20). Individual choline and ethanolamine glycerophospholipid molecular species were separated by reverse-phase HPLC using a gradient elution system with a mobile phase comprised of acetonitrile-methanol-water with 20 mM choline chloride (20). Quantification of individual phospholipid molecular species was achieved by determination of lipid phosphorus in reverse-phase HPLC column effluents (20).
Arachidonic acid release.
The extent of arachidonic acid release was determined by measuring the amount of [3H]arachidonic acid released in the surrounding medium from HCAEC prelabeled with 3 μCi of [3H]arachidonic acid/35 mm culture dish for 18 h. Following incubation, HCAEC were washed three times with Tyrode solution containing 3.6% BSA to remove unincorporated [3H]arachidonic acid. Endothelial cells were incubated at 37°C for 15 min before implementation of the experimental conditions. At the end of the stimulation period, radioactivity in both the surrounding medium and endothelial cells was quantified by liquid scintillation spectrometry.
Measurement of phosphatidate phosphohydrolase activity.
Magnesium-dependent phosphatidate phosphohydrolase (PAPH) activity in the cytosolic fraction of HCAEC was measured by incubating cytosolic protein with 600 μM [14C]glycerol-labeled phosphatidic acid (sp act 1 mCi/mmol) at 37°C for 10 min in a final volume of 100 μl containing 100 mM Tris·HCl buffer (pH 7.0), 1 mM dithiothreitol, 1 mM EDTA, 1 mM EGTA, 2 mM MgCl2, and 0.5% Triton X-100. The reaction was stopped by the addition of CHCl3-CH3OH (2:1, vol/vol) and 0.5 ml of 0.1 M KCl. The tubes were mixed vigorously for 2 min, and the phases were separated by centrifugation. [14C]phosphatidic acid and released [14C]diacylglycerol were separated by TLC on silica gel plates using petroleum ether-ether-acetic acid (60:40:1). Released [14C]diacylglycerol was quantified by liquid scintillation spectrometry.
Choline lysophospholipid production.
Lysophosphatidylcholine (LPtdCho) and lysoplasmenylcholine (LPlsCho) were measured using a modification of a radiometric assay method published previously (22). Lipids were extracted from HCAEC and the surrounding medium by the method of Bligh and Dyer (4), and lysophospholipids were separated from other phospholipids by HPLC. The purified LPtdCho and LPlsCho fractions were acetylated with [3H]acetic anhydride, the acetylated lysophospholipids were separated by TLC, and radioactivity was quantified by liquid scintillation spectrometry. Standard curves were constructed, and the LPtdCho and LPlsCho content was derived for all samples and normalized according to the protein content of HCAEC. [14C]LPtdCho was added as an internal standard to all samples to correct for the loss of sample that occurred during extraction, purification, and acetylation (22).
To determine the most appropriate conditions for thrombin stimulation and hypoxia exposure, a time course of iPLA2 activation following each condition was performed. As shown in Fig. 1A, maximal iPLA2 activity occurs after 2 min of thrombin stimulation (0.1 IU/ml). Pretreatment with BEL, a selective iPLA2 inhibitor (12), completely inhibited the increase in iPLA2 activity measured in thrombin-stimulated HCAEC (Fig. 1A).
Confluent monolayers of HCAEC were also incubated under hypoxic conditions for increasing time with or without 30 min reoxygenation, and iPLA2 activity was measured. Increases in iPLA2 activity are maximal after 10 min of hypoxia (Fig. 1B) and stay elevated for up to 40 min. Hypoxic conditioning for up to 20 min followed by 30 min of reoxygenation returns iPLA2 activity to normoxic control levels (Fig. 1B). Because 10 min of hypoxia resulted in maximal and reversible PLA2 activity, we chose this time point for all further studies.
Previously, our laboratory has shown that the presence of ATP significantly inhibits membrane-associated PLA2 activity in both normoxic and hypoxic myocytes (22). Thus we examined whether decreased ATP levels, which occur early after the onset of oxygen depletion, serve as a potential mechanism for the activation of iPLA2 in hypoxic HCAEC. ATP content was measured in HCAEC after exposure to hypoxic conditions for up to 60 min (Fig. 2). ATP levels decrease significantly over time (Fig. 2). Additionally, PLA2 activity was measured in normoxic HCAEC and HCAEC exposed to hypoxic conditions in the presence or absence of ATP (10 nm). Hypoxic conditions in the absence of ATP lead to significant increases in PLA2 activity (Fig. 3). The presence of ATP significantly inhibits the hypoxia-induced increase in PLA2 activity (Fig. 3). These results suggest that the decrease in ATP content observed early in hypoxia may, at least in part, be responsible for the increase in iPLA2 activity.
To determine whether the effects of hypoxia and thrombin stimulation on PLA2 activity were additive, the effects of these conditions alone or in combination in either the absence or presence of BEL pretreatment were compared. Incubation of confluent monolayers of HCAEC under either hypoxic conditions (10 min) or thrombin stimulation (0.1 IU/ml, 2 min) alone led to a significant increase in iPLA2 activity (Fig. 4). These increases in iPLA2 activity were completely inhibited by BEL pretreatment (Fig. 4). Thrombin stimulation in the presence of hypoxia increased iPLA2 activity significantly more than either treatment alone. This increase in iPLA2 activity was also inhibited by BEL pretreatment (Fig. 4).
To determine whether the increase in iPLA2 activity resulted in accelerated membrane phospholipid hydrolysis, we measured the individual membrane phospholipid molecular species in choline and ethanolamine glycerophospholipid classes. A significant decrease in plasmenylcholine molecular species with arachidonic (20:4) or linoleic (18:2) acid at the sn-2 position was observed following exposure to hypoxia that was potentiated by thrombin stimulation (Fig. 5). There was no significant change in phosphatidylcholine (data not shown), suggesting preferential plasmalogen phospholipid hydrolysis. A significant decrease in arachidonylated plasmenyethanolamine was observed in hypoxic HCAEC that were stimulated with thrombin (Fig. 6), but no change in ethanolamine phospholipids was observed with hypoxia or thrombin stimulation alone. Pretreatment with BEL completely inhibited iPLA2-catalyzed hydrolysis of all membrane phospholipids tested (Figs. 5 and 6).
Arachidonic acid release from HCAEC was significantly increased by exposure to hypoxia or thrombin stimulation and was further potentiated when hypoxia was combined with thrombin stimulation (Fig. 7). Pretreatment with BEL completely inhibited any increase in arachidonic acid release (Fig. 7).
Studies by other laboratories have shown that, in addition to inhibiting iPLA2, BEL inhibits PAPH activity, an enzyme whose activation also results in arachidonic acid release. PAPH converts phosphatidic acid, released from membrane phospholipids following hydrolysis by phospholipase D, to diacylglycerol (DAG). Subsequent activation of protein kinase C by DAG can lead to the activation of iPLA2, releasing arachidonic acid as a product of plasmalogen phospholipid hydrolysis. To ensure the decrease in arachidonic acid seen upon BEL pretreatment of thrombin-stimulated cells is not a result of inhibition of PAPH, we examined PAPH activity in the presence and absence of BEL. PAPH activity is only significantly decreased at concentrations of BEL >10 μM (Fig. 8), twofold higher than the concentration used to directly inhibit iPLA2 activity. These results indicate that HCAEC PAPH activity is not inhibited by 5 μM BEL pretreatment.
The increased iPLA2-catalyzed hydrolysis of membrane phospholipids was accompanied by an increase in choline lysophospholipid accumulation that was significant when hypoxic HCAEC were stimulated with thrombin (Fig. 9). HCAEC exposed to hypoxia alone demonstrated a significant increase in LPlsCho accumulation (Fig. 9, left). Although hypoxia led to an increase in LPtdCho (Fig. 9, right), this increase was not significant. BEL pretreatment completely inhibited any increase in choline lysophospholipid production.
We propose that, under hypoxic conditions, the majority of lysoplasmalogen that is hydrolyzed by iPLA2 is further catabolized, thus precluding large increases in choline lysophospholipids. Under normoxic conditions, the capacity of the myocardium for LPtdCho catabolism is ∼100-fold its capacity for LPtdCho production via PLA2 activity (23). However, in the ischemic myocardium, LPtdCho catabolism is significantly inhibited by both the accumulation of LCAC and the presence of acidosis. To determine whether the increase in choline lysophospholipids in hypoxic HCAEC may be a result of decreased catabolism, we measured LCAC content in HCAEC subjected to increased time intervals of hypoxia. As demonstrated in Fig. 10, LCAC content was significantly increased in HCAEC exposed to 10 min hypoxia and was maximal after 20 min. These data suggest that the accumulation of LCAC could contribute to a decrease in the rate of catabolism of choline lysophospholipids under hypoxic conditions and may be responsible for their accumulation in hypoxic HCAEC.
Thrombotic coronary occlusion has been shown to contribute directly to the incidence of malignant ventricular arrhythmias (5, 10), suggesting that products released from, or associated with, a thrombotic occlusion either directly or indirectly influence the electrophysiological behavior of ischemic cardiac myocytes. In previous studies, we have demonstrated that thrombin stimulation of endothelial cells results in hydrolysis of membrane phospholipids by iPLA2 that results in the generation of several phospholipid metabolites that may play an important role in arrhythmogenesis either directly or indirectly (14, 21). An evolving thrombus in a coronary vessel is associated with increased amounts of thrombin (26), some of which will be carried downstream by residual blood flow and come in contact with endothelial cells distal to the thrombus, activating endothelial cell thrombin receptors. Once blood flow is interrupted or severely reduced by increased or complete occlusion of the coronary vessel, precipitating an ischemic event, the endothelial cells in the ischemic area would be exposed to dramatic decreases in oxygen tension in addition to exposure to thrombin. In this study, we demonstrate that a decrease in oxygen such as would occur in myocardial ischemia can potentiate the thrombin-stimulated production of membrane phospholipid metabolites via activation of iPLA2 in HCAEC.
Previous studies by our laboratory have focused on determining the mechanism of iPLA2 activation by thrombin (24), with results suggesting that thrombin-stimulated iPLA2 activation is mediated by a novel protein kinase C isoform. Studies detailed in this manuscript suggest the decrease in ATP levels that accompany ischemia may be a possible mechanism for the activation of iPLA2 seen in hypoxic conditions.
The accumulation of amphiphilic metabolites, such as lysophospholipids, in the ischemic myocardium can elicit electrophysiological derangements in normoxic myocytes as a result of incorporation into the phospholipid bilayer of the sarcolemma (23). More importantly, prevention of sarcolemmal accumulation of amphiphilic metabolites markedly reduces the incidence of ischemia-induced ventricular arrhythmias (6). LPtdCho has been shown to associate preferentially with the outer leaflet of the myocyte sarcolemma, and it has been suggested that this type of association may be responsible for LPtdCho-induced electrophysiological alterations in the heart (16). Akita et al. (1) have demonstrated that intracellular microinjection of LPtdCho in isolated myocytes at concentrations as high as 500 μM did not alter electrophysiological properties of the myocytes, whereas extracellular concentrations as low as 5–10 μM have been shown to have profound effects (1, 23). These data, together with the finding that LPtdCho is elevated in the venous and lymphatic effluent from the ischemic myocardium of both animals and humans (29, 30) with no evidence of irreversible myocyte damage, suggest that LPtdCho from an extramyocytic source plays an important role in arrhythmogenesis.
We have more recently examined the effect of including LPlsCho in the perfusate of isolated normoxic cardiac myocytes and have observed profound alterations in the action potential of these myocytes, leading to the production of afterdepolarizations (15, 22). These changes were reversible upon removal of LPlsCho from the perfusate. Although the changes in cardiac myocyte action potential parameters were similar to those observed previously using other amphiphilic compounds such as LPtdCho and LCAC (see Ref. 23 for review), they occurred at much lower concentrations, suggesting that LPlsCho may interact specifically with ion channel proteins in the membrane in addition to its ability to compromise the biophysical properties of the phospholipid bilayer. The action potential derangements caused by LPlsCho occur as a result of the action of this amphiphilic compound on multiple membrane currents and on myocytic calcium handling (15).
In addition to the production of choline lysophospholipids, activation of HCAEC iPLA2 results in the production of a free fatty acid. When the membrane phospholipid that is hydrolyzed is arachidonylated, the resultant free arachidonic acid can be further metabolized to eicosanoids. The synthesis of arachidonic acid and eicosanoids in the heart has been implicated in several pathological conditions and may have direct inotropic or chronotropic effects (13). Finally, acetylation of choline lysophospholipids to produce PAF in HCAEC may also play an important role in heart disease. Although PAF remains endothelial cell-associated (21), increased PAF production results in increased binding of neutrophils to the endothelial cell surface (25, 31) and may play a role in the adhesive interaction of other cells such as eosinophils (8), thus contributing to progressing atherothrombosis and inflammatory processes in the coronary vasculature.
The increased LPtdCho observed in HCAEC stimulated with thrombin in the presence of hypoxia was not accompanied by a significant decrease in phosphatidylcholine. Under ischemic conditions in the myocardium, LPtdCho catabolism is significantly inhibited by the presence of acidosis and LCAC accumulation (see Ref. 23 for review). Because we detected significant LCAC accumulation in hypoxic HCAEC, we propose that there may be little, if any, LPtdCho catabolism in these cells and that the PLA2-catalyzed hydrolysis of phosphatidylcholine was too small to determine a significant decrease in mass. In contrast, we detected a significant decrease in plasmenylcholine (∼8 nmol PO4/mg protein) in hypoxic HCAEC stimulated with thrombin that was accompanied by an increase in LPlsCho of only 0.3 nmol/mg protein. These data suggest that, even under hypoxic conditions, the majority of LPlsCho can be further catabolized. We did not detect an increase in choline phospholipids that would indicate reacylation of LPlsCho, suggesting that the majority of LPlsCho may be catabolized via lysoplasmalogenase or lysophospholipase D. To date, there are no data available regarding the catabolic pathways for LPlsCho in endothelial cells, however, these may the subject of future studies.
These findings suggest that the presence of decreased oxygen tension and thrombin activated HCAEC iPLA2, resulting in increased membrane plasmalogen hydrolysis and the accumulation of phospholipid metabolites that may have important implications in arrhythmogenesis and inflammation following myocardial ischemia.
This work was supported in part by National Heart, Lung, and Blood Institute Grant HL-68588 and by the American Heart Association (Heartland Affiliate).
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 the American Physiological Society