Potentiation of Endothelium-Dependent Relaxation by Epoxyeicosatrienoic Acids
Abstract Epoxyeicosatrienoic acids (EETs) are potent endothelium-derived vasodilators formed from cytochrome P-450 metabolism of arachidonic acid. EETs and their diol products (DHETs) are also avidly taken up by endothelial cells and incorporated into phospholipids that participate in signal transduction. To investigate the possible functional significance of EET and DHET incorporation into cell lipids, we examined the capacity of EETs and DHETs to relax porcine coronary arterial rings and determined responses to bradykinin (which potently activates endothelial phospholipases) before and after incubating the rings with these eicosanoids. 14,15-EET and 11,12-EET (5 μmol/L) produced 75±9% and 52±4% relaxation, respectively, of U46619-contracted rings, whereas 8,9-EET and 5,6-EET did not produce significant relaxation. The corresponding DHET regioisomers produced comparable relaxation responses. Preincubation with 14,15-EET, 11,12-EET, 14,15-DHET, and 11,12-DHET augmented the magnitude and duration of bradykinin-induced relaxation, whereas endothelium-independent relaxations to aprikalim and sodium nitroprusside were not potentiated. Pretreatment with 2 μmol/L triacsin C (an inhibitor of acyl coenzyme A synthases) inhibited [3H]14,15-EET incorporation into endothelial phospholipids and blocked 11,12-EET– and 14,15-DHET–induced potentiation of relaxation to bradykinin. Exposure of [3H]14,15-EET–labeled endothelial cells to the Ca2+ ionophore A23187 (2 μmol/L) resulted in a 4-fold increased release of EET and DHET into the medium. We conclude that incorporation of EETs and DHETs into cell lipids results in potentiation of bradykinin-induced relaxation in porcine coronary arteries, providing the first evidence that incorporated EETs and DHETs are capable of modulating vascular function.
- epoxyeicosatrienoic acid
- porcine coronary artery
- dihydroxyeicosatrienoic acid
- arachidonic acid
- acyl coenzyme A synthase
Vascular endothelial cells produce several EDRFs that are capable of regulating vascular tone. These include NO, derived from the action of NO synthase on l-arginine; PGI2, a product of cyclooxygenase-mediated arachidonic acid metabolism; and EDHF, an unidentified factor or factors that produce vasorelaxation by activating K+ channels.1 2 3 Recently, EDHF in some blood vessels was proposed to be an EET metabolite of arachidonic acid.4 5 6 7 8 The four EET regioisomers, 5,6-, 8,9-, 11,12-, and 14,15-EET, are formed from arachidonic acid by the action of cytochrome P-450 epoxygenase enzymes.9 10 11 The production of EETs by blood vessels and vascular endothelial cells has recently been demonstrated.7 12 13 14 In several studies, EETs were observed to relax blood vessels by activating KCa channels.4 7 15 16 Moreover, inhibitors of cytochrome P-450 enzyme activity were reported to block EDHF-mediated relaxations in isolated blood vessels from several species and in the perfused rat heart.4 5 6 7 8 17
One property that distinguishes EETs from vasoactive prostaglandins and leukotrienes is their close structural similarity to the precursor arachidonic acid. Like arachidonic acid, EETs occur predominantly esterified to phospholipids in plasma,18 heart,19 liver,20 21 and kidney.22 Moreover, porcine aortic endothelial cells avidly take up EETs; saturation does not occur even at EET concentrations of up to 5 μmol/L.23 The uptake of EET is very rapid, reaching a maximum in just 15 to 30 minutes. Furthermore, a high percentage of the EET is incorporated into PC and PI, phospholipids that participate in endothelial signal transduction. These observations suggest that preformed EETs may be stored in endothelial cells and influence vascular function in several ways. For example, EETs present in cell membrane phospholipids may alter membrane function, ion transport, or lipid-dependent cell signaling pathways.11 5,6-EET was reported to enhance NO production by stimulating Ca2+ influx into vascular endothelial cells24 ; thus, the release of incorporated EETs during endothelial cell activation may also amplify EDRF production. Finally, preformed EETs released from endothelial cells after stimulation of phospholipid hydrolysis by agonists such as bradykinin, and perhaps by other stimuli such as shear stress, might be transferred to underlying smooth muscle cells, thereby promoting vasorelaxation through the activation of KCa channels.4 7 15 16
Besides becoming rapidly incorporated into cell lipids, EETs are rapidly converted by cellular epoxide hydrolases to their respective DHET regioisomers,23 25 which can themselves be taken up by vascular endothelial cells and incorporated into cell lipids.26 In several previous studies, DHETs were found to lack vasoactive properties.27 28 29 However, we recently observed that 11,12-DHET is vasoactive in the porcine coronary artery, producing a similar amount of relaxation as its parent EET.30 Thus, it is possible that the incorporation of DHETs into endothelial cell lipids may also influence vascular function.
The biological relevance of EET incorporation into cell lipids is suggested by the recent identification of EETs in lipid extracts from human platelets31 and cardiac tissue.19 However, whether such incorporated EETs or DHETs affect cellular function has not been determined. In the present study, we investigated the capacity of EETs and their diol products to relax porcine coronary arteries, and we examined responses to bradykinin (a receptor agonist that potently activates endothelial phospholipases) before and after incubating coronary arterial rings with EETs and DHETs.
Materials and Methods
EETs and DHETs were purchased from Cayman Chemical Co. U46619, PGF2α, bradykinin, acetylcholine, sodium nitroprusside, PDBu, L-NAME, indomethacin, and l-glutamine were purchased from Sigma Chemical Co. ODYA and triacsin C were purchased from Biomol, Inc. MEM nonessential amino acids, MEM vitamin solution, HEPES, and trypsin were obtained from GIBCO; FBS from HyClone Laboratories; fatty acid–free bovine serum albumin from Miles Laboratories, Inc; and gentamicin from Schering Corp. Radiolabeled arachidonic acid was obtained from American Radiolabeled Chemicals Inc and Amersham Corp, and phospholipid standards were obtained from Avanti Polar Lipids. Radiolabeled 14,15-EET was obtained from Dupont–New England Nuclear Corp, and Whatman LK5D silica gel TLC plates and silica gel G TLC plates were purchased from Alltech Associates, Inc.
Measurement of Porcine Coronary Arterial Ring Vasoactivity
Coronary arteries were dissected from pig hearts immediately after removal at a local slaughterhouse, placed into ice-cold KRB solution (composed of [mmol/L] NaCl 118.3, KCl 4.7, CaCl2 2.5, MgSO4 1.2, KH2PO4 1.2, NaHCO3 26, and glucose 11.1, aerated with 95% O2/5% CO2), and prepared for experimentation as described previously.30 32 33 34 The arterial rings were contracted several times with KCl (60 mmol/L) until a maximum level of tension was achieved. Each ring was then contracted with a thromboxane mimetic, U46619, to 40% to 80% of the maximal tension obtained with KCl (60 mmol/L). When tension stabilized, bradykinin (0.3 to 100 nmol/L) was administered in a cumulative fashion. The baths were rinsed, and after tension returned to baseline, the rings were contracted again with U46619 to a level of tension similar to that of the previous contraction. When a stable level of tension was achieved, one of the EET or DHET regioisomers (0.05 to 5 μmol/L) was administered in a cumulative fashion. Thirty minutes after introduction of the last EET or DHET dose (final concentration, 5 μmol/L), the organ chambers were rinsed with six exchanges of fresh KRB solution to remove EET or DHET that had not been taken up by the tissue. After an additional 30 minutes, the rings were recontracted with U46619 to a level of tension similar to that of the previous contractions, and when a stable level of tension was achieved, bradykinin (0.3 to 100 nmol/L) was administered in a cumulative fashion. When used, L-NAME (100 μmol/L) and indomethacin (10 μmol/L) were introduced into the organ chambers 45 minutes before the initial contractions and were thereafter included for the duration of the experiments. Triacsin C was added to the organ chambers 20 minutes before contracting the rings with U46619; the compound was not added again after the administration of 11,12-EET or 14,15-DHET and rinsing of the organ chambers. The final concentrations of all vehicles were <0.1%.
In some experiments, protocols were performed exactly as described above, except that rings were contracted with acetylcholine (0.03 to 1.0 μmol/L), PGF2α (0.02 to 2.0 μmol/L), or KCl (22 to 28 mmol/L) instead of U46619. To contract the rings with KCl, a KRB solution was prepared in which 20 mmol/L KCl was isotonically substituted for NaCl. Additional KCl from a 3 mol/L stock solution was added to the organ chambers to achieve 40% to 80% of the tension obtained with KCl (60 mmol/L). When the rings were contracted with PDBu (9 to 100 nmol/L) to test responses to bradykinin, 11,12-EET, or 14,15-EET, repetitive responses could not be obtained because of the failure of rings to return to basal tension after washout of PDBu.
Mechanical disruption of the endothelium was achieved immediately after rings were cut from arteries by inserting the rings onto the tips of watchmaker’s forceps and rolling them back and forth over a gauze wetted with KRB solution. The endothelium was considered to be denuded when maximal relaxation to bradykinin was <10%, as described previously.32 33 Responses to bradykinin and sodium nitroprusside (30 to 300 nmol/L) in endothelium-denuded U46619-contracted rings were determined before and after treatment with 11,12-EET or 14,15-EET, as described above. Relaxation responses were expressed as the percent decrease from the agonist-induced tension, as described previously.30 32 33 34
Cell Culture and Incubations
Porcine coronary artery endothelial cells were isolated as reported previously34 and grown in medium 199 supplemented with MEM nonessential amino acids, MEM vitamin solution, 15 mmol/L HEPES, 2 mmol/L l-glutamate, 50 μmol/L gentamicin, and 10% FBS. The cultures were maintained until confluent at 37°C in a humidified atmosphere containing 5% CO2. Stocks were subcultured weekly by trypsinization, and before experimentation the cells between passages 3 and 7 were grown to confluence on six-well plates. All experiments were conducted with medium 199 supplemented as described above but containing 0.1 μmol/L bovine serum albumin instead of 10% FBS.
Cells were incubated in 1 mL of fresh medium containing 14,15-EET (0.5 μmol/L) and [3H]14,15-EET1 (0.3 μCi) for 1 hour. To investigate EET incorporation into cell lipids, the medium was removed, and after they were washed twice with fresh medium containing no EET, the cells were harvested by scraping into methanol, a process that does not cause appreciable hydrolysis of tissue lipids,35 and transferred to test tubes. In some experiments, cells were pretreated with triacsin C (2 μmol/L) or its vehicle, DMSO, for 30 minutes before incubation with [3H]14,15-EET. To investigate the release of incorporated EET, cells were incubated with EET for 1 hour as described above, after which the cells were washed with fresh medium containing no EET. The medium was then replaced with fresh medium containing either A23187 (2 μmol/L) or vehicle (DMSO). Thirty minutes after the application of A23187 or vehicle, the incubation was terminated by removing the medium and washing the cells twice with 2 mL ice-cold PBS, and the cells were harvested and transferred to test tubes. The amount of radioactivity in the medium was determined by scintillation counting. The final concentrations of DMSO and ethanol in the medium (in all experiments) were <0.1%.
Analyses of Cell Lipids
Cell lipids were extracted and analyzed by TLC as described previously,30 36 37 38 with the following modifications: 6 mL of chloroform/methanol (2:1) followed by 2 mL of acidified saline (4 mmol/L HCl in 0.09% NaCl) was used to extract cell lipids, and neutral lipids were separated on silica gel G plates with hexane/ethyl ether/acetic acid/methanol (85:20:2:2).37 Some of the cell lipid extracts were saponified as described previously30 and assayed by reverse-phase HPLC, as described below.
Assay of Incubation Medium
After incubation of porcine coronary artery endothelial cells with [3H]14,15-EET, the incubation medium was extracted and separated by reverse-phase HPLC as described previously,30 except that a Gilson 302 pump plus 2050 UV detector and a Gilson 715 gradient controller were used, and the acetonitrile gradient was increased from 25% to 100% over 60 minutes at a flow rate of 0.7 mL/min.
Incubation of Porcine Coronary Arterial Rings With [1-14C]Arachidonic Acid
Porcine coronary arterial rings were incubated in 1 mL KRB solution at 37°C in 5% CO2. After 2 hours, the KRB solution was exchanged with fresh solution containing triacsin C (2 μmol/L) or vehicle. Twenty minutes later, the KRB solution was removed, and fresh solution containing triacsin C or its vehicle plus U46619 (10 nmol/L) was applied in order to duplicate, as much as possible, the conditions under which ring vasoactivity was examined. After 20 minutes, the KRB solution was exchanged for fresh solution containing U46619, triacsin C or its vehicle, and 2.6 μmol/L [1-14C]arachidonic acid. After 30 minutes, the KRB solution was removed, and the rings were washed with PBS (4°C) and transferred to test tubes. Six milliliters of chloroform/methanol (2:1) was added, and the tubes were mixed and stored overnight at −20°C. The following day, 2 mL acidified saline was added, and the rings were vortexed and maintained at 4°C for 2 hours. Cell lipids were extracted and separated by TLC as described above, except that phospholipids were separated using chloroform/methanol/glacial acetic acid/Millipore–filtered water (50:50:2:1.5), and neutral lipids were separated using heptane/ethyl ether/glacial acetic acid (75:60:1.5). The values were converted to picomoles, calculated from the specific activity of the [1-14C]arachidonic acid with which the rings were incubated, and normalized to ring wet weight.
All data are expressed as mean±SEM. The effects of treatment with EETs and DHETs on bradykinin-induced relaxation were determined by repeated-measures ANOVA with a Greenhouse-Geisser ε correction (SAS-STAT [SAS Instititutes], procedure GLM). When the interaction term (ie, concentration of bradykinin×treatment with EETs or DHETs) was not significant, analysis of the main effect of treatment (ie, overall relaxation to bradykinin) was performed. The half-maximal effective concentration (EC50) of bradykinin was calculated for each concentration-response curve before and after treatment with the individual EET and DHET regioisomers. The EC50 values (expressed as −log[M]) before and after treatment were analyzed by paired Student’s t tests. All other data were also analyzed by paired Student’s t tests. Values of P≤.05 were considered to be statistically significant.
Vasoactivity of EETs in Porcine Coronary Arteries
The effects of the EET regioisomers (0.05 to 5.0 μmol/L) on endothelium-intact porcine coronary arterial rings submaximally contracted with a thromboxane mimetic, U46619 (0.5 to 30 nmol/L), were examined. 14,15-EET and 11,12-EET produced concentration-dependent relaxations exceeding that produced by the vehicle (ethanol), whereas 8,9- and 5,6-EET produced minimal relaxations (Fig 1⇓).
In some experiments, 11,12-EET or 14,15-EET was administered to endothelium-denuded porcine coronary arterial rings contracted with U46619. Both 11,12-EET and 14,15-EET produced relaxations of denuded rings that were similar in magnitude and duration to those resulting from administration to endothelium-intact rings (71±6% and 65±5%, respectively, at 5 μmol/L; n=4 for each group), as was previously reported for bovine and canine coronary arteries.12 13
In separate experiments, the effects of 14,15-EET and 11,12-EET on endothelium-intact porcine coronary artery rings contracted with agonists other than U46619 were examined. Both compounds relaxed rings contracted with PGF2α, but neither produced significant relaxations of rings contracted with acetylcholine, KCl, or PDBu (Table 1⇓). In fact, small contractions were frequently observed when either compound was administered to rings contracted with KCl or when 14,15-EET was administered to rings contracted with acetylcholine. In contrast to these findings, bradykinin, in the presence of indomethacin (10 μmol/L) and L-NAME (100 μmol/L) to inhibit prostaglandin and NO production, respectively,32 relaxed porcine coronary rings contracted with either acetylcholine or PDBu.
The effects of 17-ODYA, a cytochrome P-450 enzyme inhibitor,17 on bradykinin-induced relaxation of endothelium-intact U46619-contracted rings were examined. These experiments were conducted in the presence of indomethacin (10 μmol/L) and L-NAME (100 μmol/L) to inhibit prostaglandin and NO production, respectively.32 After the initial determination of bradykinin-induced relaxation, rings were treated with 50 μmol/L 17-ODYA for 30 minutes and then recontracted with U46619. When tension stabilized, bradykinin was applied in a concentration-dependent fashion. Under these conditions, bradykinin-induced relaxation was unaltered by treatment with 17-ODYA (73±4% [before 17-ODYA] versus 76±7% [after 17-ODYA] at 100 nmol/L bradykinin, n=4).
Effects of EETs on Endothelium-Dependent Relaxation to Bradykinin
Relaxation responses produced by the EETs were more sustained (Fig 2⇓, top) than the brief relaxations produced by bradykinin before exposure to EETs (Fig 2⇓, middle). After the administration of EETs to endothelium-intact U46619-contracted rings, the organ chambers were rinsed thoroughly to remove any EET that had not been taken up by the tissues. After 30 minutes, the rings were contracted with U46619 to a level of tension similar to that of the preceding contractions, and bradykinin-induced relaxation was determined and compared with the initial (pre-EET) bradykinin-induced response. Contractions elicited by U46619 after exposure to the EETs did not differ from the pre-EET contractions (n=4 to 8 for each regioisomer, data not shown). However, after treatment with 14,15-EET or 11,12-EET, bradykinin-induced relaxation was enhanced in magnitude and duration (EC50 values were 8.4±0.1 [before 14,15-EET] versus 8.8±0.0 [after 14,15-EET] and 8.4±0.2 [before 11,12-EET] versus 8.9±0.0 [after 11,12-EET], P<.03, respectively) (Fig 2⇓, bottom; Fig 3⇓, top and middle). In contrast to these findings, bradykinin-induced relaxation was not potentiated by 8,9-EET (EC50 values were 8.7±0.0 [before 8,9-EET] versus 8.6±0.0 [after 8,9-EET], P=.8) (Fig 3⇓, bottom). Likewise, bradykinin-induced relaxation was not potentiated by 5,6-EET (n=4, not shown) or the EET vehicle (n=4, not shown).
In some experiments, responses to bradykinin and sodium nitroprusside were determined in endothelium-denuded U46619-contracted rings before and after administering 11,12-EET or 14,15-EET. Bradykinin failed to relax endothelium-denuded rings before or after treatment with either 11,12-EET or 14,15-EET (n=6, not shown), and relaxation to sodium nitroprusside was not potentiated after treatment with 14,15-EET (52±10% [before 14,15-EET] versus 46±5% [after 14,15-EET] at 30 nmol/L sodium nitroprusside, n=4).
The effects of treatment with 11,12-EET and 14,15-EET on relaxation to aprikalim (0.03 to 3 μmol/L) in endothelium-intact rings were also determined. These experiments were performed exactly as described previously, except that the rings were relaxed with aprikalim rather than bradykinin before and after treatment with 14,15-EET or 11,12-EET. Unlike relaxation to bradykinin, relaxation to aprikalim (which hyperpolarizes vascular smooth muscle cells by activating ATP-sensitive K+ channels39 ) was not potentiated after treatment with 14,15-EET or 11,12-EET (n=4 for each group, not shown).
Role of NO in EET-Induced Potentiation of Relaxation to Bradykinin
We investigated whether 14,15-EET or 11,12-EET could have potentiated bradykinin-induced relaxation by enhancing NO production. To selectively examine bradykinin-induced NO-mediated relaxation, endothelium-intact rings were preincubated with indomethacin (10 μmol/L) to inhibit prostaglandin production and contracted with KCl (22 to 28 mmol/L) to inhibit EDHF-mediated relaxation, as described previously.32 33 34 Under these conditions, bradykinin-induced relaxation of the porcine coronary artery is mediated solely by NO.32 33 34 40 41 Responses to bradykinin were determined before and after the administration of 14,15-EET or 11,12-EET to indomethacin-pretreated KCl-contracted rings. Under these conditions, treatment with 14,15-EET (n=4, not shown) or 11,12-EET (Fig 4⇓, top) did not potentiate bradykinin-induced relaxation (EC50 values were 8.5±0.1 [before 11,12-EET] versus 8.6±0.1 [after 11,12-EET], P=.4).
To confirm that NO production is not essential for EET-induced potentiation of relaxation to bradykinin, endothelium-intact rings were preincubated with L-NAME (100 μmol/L) to inhibit NO synthase activity32 33 34 and with indomethacin (10 μmol/L). Bradykinin-induced relaxation of U46619-contracted rings was then determined before and after the administration of 14,15-EET or 11,12-EET, as described above. Despite inhibition of NO synthase, 14,15-EET (n=4, not shown) and 11,12-EET (Fig 4⇑, bottom) still potentiated bradykinin-induced relaxation (EC50 values were 8.0±0.1 [before 11,12-EET] versus 8.5±0.2 [after 11,12-EET], P<.01).
Vascular Effects of DHETs in Porcine Coronary Arteries
We examined the effects of DHETs on endothelium-intact U46619-contracted porcine coronary arterial rings and investigated whether exposure to DHETs also results in potentiation of endothelium-dependent relaxation to bradykinin. Similar to the results obtained with the EETs, 14,15-DHET and 11,12-DHET produced relaxations of U46619-contracted porcine coronary arterial rings which exceeded that produced by the vehicle, whereas 8,9-DHET produced minimal relaxation (Fig 5⇓, top). DHET-induced relaxations were sustained in duration, like those produced by EETs. Responses to 5,6-DHET were not examined because of the instability of this compound in aqueous solutions.
After exposure to 14,15-DHET (Fig 5⇑, bottom) or 11,12-DHET (n=6, not shown), bradykinin-induced relaxation was potentiated (EC50 values were 8.4±0.2 [before 14,15-DHET] versus 9.2±0.2 [after 14,15-DHET], P<.05). In contrast, as was observed with 8,9-EET (see Fig 2⇑, bottom), exposure to 8,9-DHET did not potentiate bradykinin-induced relaxation (n=4, not shown).
Importance of EET and DHET Incorporation Into Cell Lipids to the Potentiation of Bradykinin-Induced Relaxation
To investigate whether EETs and DHETs must be incorporated into cell lipids to potentiate bradykinin-induced relaxation, we pretreated endothelium-intact rings with an inhibitor of acyl-CoA synthases, triacsin C,42 before exposure to 11,12-EET and 14,15-DHET. Acyl-CoA synthases catalyze the conversion of free fatty acids to acyl-CoA thioesters, the metabolically active forms of fatty acids that participate in numerous cellular processes, including phospholipid synthesis and remodeling.43 After the initial determination of bradykinin-induced relaxation, triacsin C (2 μmol/L) was introduced into the organ chambers, the rings were contracted with U46619, and 11,12-EET or 14,15-DHET (0.05 to 5.0 μmol/L) was administered as described previously. The organ chambers were then rinsed with six exchanges of KRB solution, and bradykinin-induced relaxation was determined and compared with the initial (pre-EET or -DHET) bradykinin-induced relaxation. Pretreatment with triacsin C prevented the 11,12-EET–induced potentiation (Fig 6⇓, top) and attenuated the 14,15-DHET–induced potentiation (Fig 6⇓, bottom) of relaxation to bradykinin (EC50 values were 8.0±0.2 [before 11,12-EET] versus 8.0±0.1 [after 11,12-EET] and 8.4±0.1 [before 14,15-DHET] versus 8.7±0.1 [after 14,15-DHET]; P=.5 and P=.1, respectively). In separate control experiments, triacsin C was administered as described above to endothelium-intact rings that were contracted with U46619 but not subsequently exposed to 11,12-EET or 14,15-DHET. Under these conditions, triacsin C did not alter bradykinin-induced relaxation (n=5, not shown).
Next, we investigated the efficacy of triacsin C to block [3H]14,15-EET incorporation into porcine coronary artery endothelial cell lipids. Cultured porcine coronary artery endothelial cells were pretreated for 30 minutes with triacsin C (2 μmol/L) or its vehicle and then incubated with [3H]14,15-EET for 1 hour. The medium was removed, and the cell lipids were extracted and analyzed by TLC. Triacsin C inhibited both the total uptake and the amount of [3H]14,15-EET incorporated into the PI, PC, and PE fractions (Fig 7⇓).
To confirm that triacsin C also blocks fatty acid incorporation into freshly harvested arterial tissue, we determined the effects of triacsin C on [1-14C]arachidonic acid incorporation into porcine coronary arterial rings. Similar to the results observed with [3H]14,15-EET in cultured endothelial cells, triacsin C (2 μmol/L) inhibited total [1-14C]arachidonic acid uptake into porcine coronary arterial rings and the amount of [1-14C]arachidonic acid incorporated into phospholipids (Table 2⇓). Triacsin C also inhibited [1-14C]arachidonic acid incorporation into triglycerides. In contrast, the amount of [1-14C]arachidonic acid present as free fatty acids was unaffected by triacsin C, suggesting that the compound did not nonspecifically inhibit fatty acid transport into the tissue.
Release of Incorporated EETs From Vascular Endothelial Cells
To investigate whether incorporated EETs are released from endothelial cells by agonists that stimulate endothelial phospholipase activity, porcine coronary artery endothelial cells were incubated with [3H]14,15-EET for 1 hour and then washed to remove EET that had not been taken up by the cells. Cell lipids were then extracted in some experiments and analyzed by TLC or saponified and analyzed by HPLC. In other experiments, the cells were treated with A23187 (2 μmol/L) or its vehicle for 30 minutes, after which the medium was removed, assayed for radioactivity, and analyzed by HPLC. After treatment with A23187 or vehicle, the cell lipids were also extracted and analyzed by TLC. After a 1-hour incubation with [3H]14,15-EET, over half of the cell-associated radioactivity was contained in the PC fraction (Fig 8⇓, top). The remainder of the incorporated radioactivity was contained in PI and PE. HPLC analysis of the saponified cell lipids indicated that ≈90% of the incorporated radioactivity consisted of [3H]14,15-EET (Fig 8⇓, bottom), whereas <5% consisted of [3H]14,15-DHET.
Treatment of [3H]14,15-EET–labeled cells with A23187 for 30 minutes resulted in a 4-fold increase in the release of radioactive products into the media (Fig 9⇓, top). Calculations indicate that >95% of the released radioactivity was derived from PC and PI (n=3, not shown). HPLC analysis of the media demonstrated that 66% of the released radioactivity coeluted with a peak previously identified by gas chromatography coupled with mass spectrometry as [3H]14,15-DHET23 (Fig 9⇓, bottom), whereas the remainder of the released radioactivity consisted of [3H]14,15-EET.
In the present study, we found that 14,15-EET, 11,12-EET, and their diol products relaxed porcine coronary arterial rings precontracted with U46619 and that relaxation to bradykinin, an endothelium-dependent vasodilator, was potentiated after the rings were exposed to 14,15-EET, 11,12-EET, 14,15-DHET, and 11,12-DHET. Both the magnitude and duration of bradykinin-induced relaxation were enhanced and matched the values observed when EETs or DHETs were directly applied. Moreover, the same order of regiospecificity was observed for potentiation and for the EET- or DHET-induced relaxations. We recently showed that these same doses of bradykinin potently stimulate phospholipase activity in porcine coronary arteries.34 Together, these findings suggest that the potentiation of bradykinin-induced relaxation resulted from the release of EETs and/or DHETs, perhaps after their incorporation into endothelial phospholipids.
Alternatively, the potentiation of bradykinin-induced relaxation may have resulted from an effect of EETs or DHETs on the vascular smooth muscle rather than endothelial cells. Fang et al44 observed that EETs can be incorporated into porcine aortic smooth muscle cell lipids. However, treatment with EETs did not alter U46619-induced contractions, and bradykinin failed to relax endothelium-denuded rings after exposure to EETs. Furthermore, relaxations produced by two endothelium-independent relaxing agonists, aprikalim and sodium nitroprusside, were not potentiated by EETs. We did not investigate whether incorporation of EETs into cell lipids might potentiate endothelium-independent relaxation mediated by activation of KCa channels. Also, our experimental preparation did not enable us to selectively examine the effects of smooth muscle cell EET incorporation on bradykinin-induced relaxation. Thus, further studies are required to definitively address the importance of vascular smooth muscle cell EET and DHET incorporation to the potentiation of bradykinin-induced relaxation.
EETs are rapidly converted by vascular endothelial and smooth muscle cells to DHETs,23 44 which can also become incorporated into endothelial cell lipids.26 In noncoronary blood vessels, DHETs were reported to be devoid of vasoactive properties.27 28 29 However, we recently observed that 11,12-DHET is vasoactive in porcine coronary arteries, producing an amount of relaxation similar to that with 11,12-EET.30 In the present study, we found that 14,15-DHET also produced substantial relaxation of U46619-contracted porcine coronary arterial rings, whereas 8,9-DHET did not; 5,6-DHET was not tested because of the chemical lability of this diol. Moreover, exposure to 14,15-DHET or 11,12-DHET potentiated the endothelium-dependent relaxation to bradykinin. Pretreatment with triacsin C attenuated the 14,15-DHET–induced potentiation of relaxation to bradykinin, suggesting that the potentiation was dependent on incorporation of 14,15-DHET into cell lipids through an acyl-CoA synthase–dependent pathway. These observations extend our previous findings and confirm that DHETs are capable of regulating vascular tone in coronary arteries. Interestingly, both EET- and DHET-induced vasorelaxations were regioisomeric specific and reflected structural polarities; ie, the eicosanoids with the epoxide and hydroxyl moieties farthest removed from the carboxyl group eluted first on a reverse-phase HPLC column45 and possessed the greatest amount of vasoactivity. In contrast, all four EET regioisomers produce equivalent amounts of relaxation of U46619-contracted canine and bovine coronary arteries.12 13 However, 5,6-EET potently dilates rat intestinal microvessels, whereas 14,15-EET produces little effect.27 Likewise, in cat cerebral microvessels, the only EET demonstrated to produce potent dilation was 5,6-EET, an effect that was attributed to cyclooxygenase-dependent free radical production.46 Recently, nanomolar concentrations of 11(R),12(S)-EET were observed to hyperpolarize and relax renal preglomerular arterioles, whereas 11(S),12(R)-EET and 14,15-EET were far less potent.16 Thus, in blood vessels and in platelets,47 48 the effects of EETs may be both regioisomeric and stereoisomeric specific.
The similarities between the vascular effects of the EET and DHET regioisomers observed in the present study suggest that conversion to DHETs could contribute to EET-induced relaxations in porcine coronary arteries. Campbell et al7 recently reported that the conversion of EETs to DHETs in bovine coronary arteries occurs slowly in relation to the rapid activation of KCa channels by EETs, implying that conversion to DHETs is not responsible for the EET-induced KCa activation. However, in the present study, twice as much 14,15-DHET as 14,15-EET accumulated in the medium after stimulation of [3H]14,15-EET–labeled cells with A23187 for 30 minutes. This observation suggests that after the exogenous administration or endogenous release of EET, the intracellular DHET concentration may rise rapidly enough to contribute to the EET-induced response. Further studies are required to determine the role of conversion to DHET in EET-induced relaxation of porcine coronary arteries.
It is unlikely that the EET-induced potentiation of relaxation to bradykinin was due to enhanced production or activity of NO. This possibility was suggested by a recent report that 5,6-EET can activate Ca2+ influx into vascular endothelial cells, leading to stimulation of NO synthase activity and hence NO production.24 However, 14,15-EET and 11,12-EET failed to potentiate bradykinin-induced NO-mediated relaxation of KCl-contracted rings. Moreover, the EET-induced potentiation of U46619-contracted rings was not prevented by treatment with L-NAME (100 μmol/L), an inhibitor of NO synthase. Since the latter experiments were performed in the presence of indomethacin, these results also suggest that the EET-induced potentiation was not due to enhanced production of vasodilator prostaglandins.
In the presence of NO synthase and cyclooxygenase inhibition, bradykinin relaxes porcine coronary arteries by stimulating the release of EDHF.3 40 41 In a recent study, bradykinin-induced EDHF-mediated relaxation of porcine coronary arteries was reported to be blocked by SKF525a (proadifen) and clotrimazole, inhibitors of cytochrome P-450 enzymes, leading the authors to postulate that relaxation attributed to EDHF in porcine coronary artery is mediated by EETs.4 However, in two other studies, bradykinin-induced EDHF-mediated relaxation of porcine coronary arteries was not blocked by proadifen, clotrimazole, or 7-ethoxyresorufin.32 33 Recently, relaxation induced by A23187 in porcine coronary arteries treated with NO synthase and cyclooxygenase inhibitors was not blocked by the cytochrome P-450 inhibitors econazole or protoporphyrin IX.49 In the present study, 17-ODYA (50 μmol/L), a selective inhibitor of cytochrome P-450 enzymes,17 also did not block bradykinin-induced EDHF-mediated relaxation. These results suggest that unlike EDHF in bovine coronary artery7 49 and in perfused rat heart,5 17 EDHF in porcine coronary artery is resistant to inhibitors of cytochrome P-450 enzymes. Furthermore, in the presence of NO synthase and cyclooxygenase inhibition, bradykinin relaxed porcine coronary arterial rings contracted with acetylcholine or PDBu, whereas the EETs failed to produce significant relaxations of rings contracted with either of these agonists. Although the reasons for these differences in relaxant properties are unclear, these observations suggest that the mechanisms of EET- and EDHF-induced vasorelaxations differ in the porcine coronary artery. Taken together, these results suggest that EDHF in the epicardial porcine coronary artery is most likely not an EET.
In summary, incubation of porcine coronary arterial rings with EETs and DHETs resulted in regioisomeric-specific potentiation of relaxation to bradykinin. The capacity of EETs and DHETs to potentiate bradykinin-induced relaxation paralleled the vasorelaxing potency of the individual regioisomers. Inhibition of acyl-CoA synthase activity blocked the incorporation of EET into endothelial cell lipids and attenuated the EET- and DHET-induced potentiation of relaxation to bradykinin, suggesting that the potentiation is dependent on the incorporation of EETs and DHETs into cell lipids. Thus, unlike vasoactive prostaglandins and leukotrienes, EETs and DHETs may be stored in cell lipids to subsequently modulate vascular tone.
Selected Abbreviations and Acronyms
|acyl-CoA||=||acyl coenzyme A|
|EDHF||=||endothelium-derived hyperpolarizing factor|
|EDRF||=||endothelium-derived relaxing factor|
|KCa channel||=||Ca2+-activated K+ channel|
|L-NAME||=||Nω-nitro-l-arginine methyl ester|
|PGF2α, PGI2||=||prostaglandin F2α and I2|
This study was supported by grant HL-49264 from the National Heart, Lung, and Blood Institute, National Institutes of Health, and by an American Heart Association Clinician Scientist Award (No. 96004540) to Dr Weintraub. The authors gratefully acknowledge Ruzicka’s Meat Processing in Solon, Iowa, for supplying the porcine coronary arteries.
↵1 [3H]14,15-EET: the 3H is present at carbons 5, 6, 8, 9, 11, 12, 14, and 15. The 14,15- refers to the location of the epoxide group.
- Received November 25, 1996.
- Accepted May 30, 1997.
- © 1997 American Heart Association, Inc.
Campbell WB, Gebremedhin D, Pratt PF, Harder DR. Identification of epoxyeicosatrienoic acids as endothelium-derived hyperpolarizing factors. Circ Res. 1996;78:415-423.
Chen G, Cheung DW. Modulation of endothelium-dependent hyperpolarization and relaxation to acetylcholine in rat mesenteric artery by cytochrome P450 enzyme activity. Circ Res. 1996;79:827-833.
Capdevila JH, Falck JR, Estabrook RW. Cytochrome P450 and the arachidonate cascade. FASEB J. 1992;6:731-736.
Rosolowsky M, Falck JR, Willerson JT, Campbell WB. Synthesis of lipoxygenase and epoxygenase products of arachidonic acid by normal and stenosed canine coronary arteries. Circ Res. 1990;66:608-621.
Rosolowsky M, Campbell WB. Role of PGI2 and epoxyeicosatrienoic acids in relaxation of bovine coronary arteries to arachidonic acid. Am J Physiol. 1993;264:H327-H335.
Gebremedhin D, Ma Y-H, Falck JR, Roman RJ, VanRollins M, Harder DR. Mechanism of action of cerebral epoxyeicosatrienoic acids on cerebral arterial smooth muscle. Am J Physiol. 1992;263:H519-H525.
Zou A-P, Fleming JT, Falck JR, Jacobs ER, Gebremedhin D, Harder DR, Roman RJ. Stereospecific effects of epoxyeicosatrienoic acids on renal vascular tone and K+-channel activity. Am J Physiol. 1996;270:F822-F832.
Wu S, Moomaw CR, Tomer KB, Falck JR, Zeldin DC. Molecular cloning and expression of CYP2J2, a human cytochrome P450 arachidonic acid epoxygenase highly expressed in human heart. J Biol Chem. 1996;271:3460-3468.
Karara A, Dishman E, Falck JR, Capdevila JH. Endogenous epoxyeicosatrienoyl-phospholipids. J Biol Chem. 1991;266:7561-7569.
VanRollins M, Kaduce TL, Knapp HR, Spector AA. 14,15-Epoxyeicosatrienoic acid metabolism in endothelial cells. J Lipid Res. 1993;34:1931-1942.
Catella F, Lawson JA, Fitzgerald DJ, Fitzgerald GA. Endogenous biosynthesis of arachidonic acid epoxides in humans: increased formation in pregnancy-induced hypertension. Proc Natl Acad Sci U S A. 1990;87:5893-5897.
VanRollins M, Kaduce TL, Fang X, Knapp HR, Spector AA. Arachidonic acid diols produced by cytochrome P-450 monooxygenases are incorporated into phospholipids of vascular endothelial cells. J Biol Chem. 1996;271:14001-14009.
Proctor KG, Falck JR, Capdevila J. Intestinal vasodilation by epoxyeicosatrienoic acids: arachidonic acid metabolites produced by a cytochrome P450 monooxygenase. Circ Res. 1987;60:50-59.
Pfister SL, Falck JR, Campbell WB. Enhanced synthesis of epoxyeicosatrienoic acids by cholesterol-fed rabbit aorta. Am J Physiol. 1991;261:H843-H852.
Fang X, Kaduce TL, Weintraub NL, VanRollins M, Spector AA. Functional implications of a newly characterized pathway of 11,12-epoxyeicosatrienoic acid metabolism in arterial smooth muscle. Circ Res. 1996;79:784-793.
Zhu Y, Brand Schieber E, McGiff JC, Balazy M. Identification of arachidonate P-450 metabolites in human platelet phospholipids. Hypertension. 1995;25:854-859.
Weintraub NL, Stephenson AH, Sprague RS, McMurdo L, Lonigro AJ. Relationship of arachidonic acid release to porcine coronary artery relaxation. Hypertension. 1995;26:684-690.
Spector AA, Kaduce TL, Hoak JC, Fry GL. Utilization of arachidonic and linoleic acids by cultured human endothelial cells. J Clin Invest. 1981;68:1003-1011.
Folch JM, Lees M, Sloane Stanley GH. A simple method for the isolation and purification of total lipids from animal tissues. J Biol Chem. 1957;226:497-509.
Spector AA, Kaduce TL, Hoak JC, Czervionke RL. Arachidonic acid availability and prostacyclin production by cultured human endothelial cells. Arteriosclerosis. 1983;3:323-331.
Nelson MT, Quayle JM. Physiological roles and properties of potassium channels in arterial smooth muscle. Am J Physiol. 1995;268:C799-C822.
Cowan CL, Cohen RA. Two mechanisms mediate relaxation by bradykinin of pig coronary artery: NO-dependent and NO-independent responses. Am J Physiol. 1991;261:H830-H835.
Korchak HM, Kane LH, Rossi MW, Corkey BE. Long chain acyl coenzyme A and signaling in neutrophils. J Biol Chem. 1994;269:30281-30287.
Fang X, VanRollins M, Kaduce TL, Spector AA. Epoxyeicosatrienoic acid metabolism in arterial smooth muscle cells. J Lipid Res. 1995;36:1236-1246.
Ellis EF, Police RJ, Yancey L, McKinney JS, Amruthesh SC. Dilation of cerebral arterioles by cytochrome P-450 metabolites of arachidonic acid. Am J Physiol. 1990;259:H1171-H1177.
Fitzpatrick FA, Ennis MD, Baze ME, Wynalda MA, McGee JE, Liggett WF. Inhibition of cyclooxygenase activity and platelet aggregation by epoxyeicosatrienoic acids: influence of stereochemistry. J Biol Chem. 1986;261:15334-15338.
VanRollins M. Epoxygenase metabolites of docosahexaenoic and eicosapentaenoic acids inhibit platelet aggregation at concentrations below those affecting thromboxane synthesis. J Pharmacol Exp Ther. 1995;274:798-804.