Heart and Circulatory Physiology

11,12,20-Trihydroxy-eicosa-8(Z)-enoic acid: a selective inhibitor of 11,12-EET-induced relaxations of bovine coronary and rat mesenteric arteries

Ishfaq A. Bukhari, Abdul Jabbar Shah, Kathryn M. Gauthier, Katherine A. Walsh, Sreenivasulu Reddy Koduru, John D. Imig, John R. Falck, William B. Campbell


Arachidonic acid is metabolized to four regioisomeric epoxyeicosatrienoic acids (EETs) by cytochrome P-450. 5,6-, 8,9-, 11,12-, and 14,15-EET are equipotent in relaxing bovine coronary arteries (BCAs). Vasorelaxant effects of EETs are nonselectively antagonized by 14,15-epoxyeicosa-5(Z)-enoic acid. The 11,12-EET analogs, 20-hydroxy-11,12-epoxyeicosa-8(Z)-enoic acid (20-H-11,12-EE8ZE) and 11,12,20-trihydroxyeicosa-8(Z)-enoic acid (11,12,20-THE8ZE) were synthesized and tested for antagonist activity against EET-induced relaxations in BCAs. In U-46619-preconstricted arterial rings, 5,6-, 8,9-, 11,12-, and 14,15-EET caused concentration-dependent relaxations with maximal relaxations ranging from 80 to 96%. Preincubation of arteries with 20-H-11,12-EE8ZE (10−5 M) inhibited relaxations to 14,15- and 11,12-EET, but not 5,6- and 8,9-EET; however, greatest inhibitory effect was against 11,12-EET (maximal relaxation = 80.6 ± 4.6 vs. 26.7 ± 7.4% without and with 20-H-11,12-EE8ZE, respectively). Preincubation with the soluble epoxide hydrolase inhibitor (tAUCB, 10−6 M) significantly enhanced the antagonist effect of 20-H-11,12-EE8ZE against 14,15-EET-induced relaxations (maximal relaxation = 86.6 ± 4.4 vs. 27.8 ± 3.3%, without and with 20-H-11,12-EE8ZE and tAUCB) without any change in its effect against 11,12-EET-induced relaxations. In contrast to the parent compound, the metabolite, 11,12,20-THE8ZE (10−5 M), significantly inhibited relaxations to 11,12-EET and was without effect on other EET regioisomers. Mass spectrometric analysis revealed conversion of 20-H-11,12-EE8ZE to 11,12,20-THE8ZE by incubation with BCA. The conversion was blocked by tAUCB. 14,15-Dihydroxy-eicosa-5Z-enoic acid (a 14,15-EET antagonist), but not 11,12,20-THE8ZE (an 11,12-EET antagonist), inhibited BCA relaxations to arachidonic acid and flow-induced dilation in rat mesenteric arteries. These results indicate that 11,12,20-THE8ZE is a selective antagonist of 11,12-EET relaxations and a useful pharmacological tool to elucidate the function of 11,12-EET in the cardiovascular system.

  • epoxyeicosatrienoic acid
  • arteries
  • vascular relaxation
  • soluble epoxide hydrolase
  • endothelium-derived hyperpolarizing factor

coronary endothelial cells metabolize arachidonic acid by the cyclooxygenase, lipoxygenase, and cytochrome P-450 (CYP450) pathways (29, 30). CYP450 converts arachidonic acid to 5,6-, 8,9-, 11,12-, and 14,15-epoxyeicosatrienoic acids (EETs). In bovine coronary arteries, EETs function as endothelium-derived hyperpolarizing factors and activate membrane large-conductance calcium-activated (BKCa) potassium channels to induce hyperpolarization and vascular relaxation (3, 5, 6, 12, 20). These vascular effects of EETs are selectively blocked by 14,15-epoxyeicosa-5Z-enoic acid (14,15-EE5ZE) (13). EETs are hydrolyzed to dihydroxyeicosatrienoic acids (DHETs) by soluble epoxide hydrolase (sEH) (17, 31, 34). sEH is a cytosolic enzyme responsible for the metabolism of EETs in arterial endothelial and smooth muscle cells (2, 10, 11). Changes in the EET structure through metabolism by sEH or chemical modifications have significant impact on their biological activity (2, 8, 18). For example, 14,15-DHET is fivefold less potent than 14,15-EET (2). In rat renal arterioles, 11,12-EET causes relaxation, while 11,12-DHET is without effect (18). Conversion of 14,15-EE5ZE to 14,15-dihydroxy-eicosa-5(Z)-enoic acid (14,15-DHE5ZE) by sEH also changes it pharmacological properties (1). Whereas 14,15-EE5ZE is a nonselective EET antagonist, 14,15-DHE5ZE is a selective inhibitor of the relaxations to 14,15-EET, but not other EET regioisomers. 14,15-Epoxyeicosa-5Z-enoic-methylsulfonylimide selectively inhibits relaxations to 5,6- and 14,15-EET, but not to 8,9- and 11,12-EET (15). These findings suggest that EETs produce vascular effects by interacting with receptors or binding sites that are specific for each regioisomer. However, the identity of these receptors/binding sites has not been forthcoming. The development of EET regioisomer-selective antagonists is essential to elucidate the physiological function of individual EET regioisomers and their downstream mechanisms.

We synthesized and tested a series of 11,12-EET analogs for agonist and antagonist activity. Only 20-hydroxy-11,12-epoxyeicosa-8(Z)-enoic acid (20-H-11,12-EE8ZE) had antagonist activity against EET-induced relaxations. The present studies indicate that bovine coronary arteries and endothelial cells metabolize 20-H-11,12-EE8ZE to 11,12,20-trihydroxy-eicosa-8(Z)-enoic acid (11,12,20-THE8ZE), which inhibits 11,12-EET-induced relaxation, but not the other EET regioisomers (Fig. 1). The 11,12-EET-selective antagonist properties of 11,12,20-THE8ZE will be useful in elucidating physiological function of 11,12-EET.

Fig. 1.

Structures of 20-hydroxy-11,12-epoxyeicosa-8Z-enoic acid (20-H-11,12-EE8ZE) and 11,12,20-trihydroxyeicosa-8Z-enoic acid (11,12,20-THE8ZE). 20-H-11,12-EE8ZE is metabolized to 11,12,20-THE8ZE by soluble epoxide hydrolase (sEH) and inhibited by the sEH inhibitor trans-4-[4-(3-adamantan-1-yl-ureido)-cyclohexyloxy]-benzoic acid (tAUCB).


Synthesis of EET Analogs

The 11,12-EET analogs, including 20-H-11,12-EE8ZE and 11,12,20-THE8ZE, were synthesized as previously described (9) (Figs. 1 and 2).

Fig. 2.

11,12-Epoxyeicosatrienoic acid (EET) analogs screened for agonist and antagonist activity in bovine coronary arteries (BCAs). A: agonist activity of EET analogs compared with 11,12-EET. B: structures and comparisons of agonist and antagonist activity of 11,12-EET analogs. *Analogs with agonist activity in A. Agonist activity is indicated as maximal %relaxation at 10−5 M, with + for 0–20%, ++ for 20–40%, +++ for 40–60%, and ++++ for 60–80%. Antagonist activity was assigned as inhibition of relaxations to 11,12-EET and indicated as + for antagonist activity and − for no antagonist activity. N/A, not attempted.

Vascular Reactivity Studies

Isometric tension in bovine coronary artery rings.

Measurements of isometric tone in bovine coronary arterial rings were conducted as described previously (6, 13, 15). Fresh bovine hearts were obtained from a local slaughterhouse. Sections of the left anterior descending coronary artery were dissected, cleaned, and cut into 1.5- to 2.0-mm-diameter rings (3 mm length). The arterial rings were suspended in a tissue bath containing a Kreb's-bicarbonate buffer equilibrated with 95% O2-5% CO2 and maintained at 37°C. Tension was measured using a model FT-03C force transducer (Grass Instruments), ETH-400 bridge amplifier, and MacLab 8e A/D converter with MacLab software and Macintosh computer. The arterial rings were slowly stretched to a basal tension of 3.5 g and equilibrated for 1.5 h. KCl (40–60 × 10−3 M) was repeatedly added and rinsed until reproducible stable contractions were observed. The thromboxane mimetic, U-46619 (10−8 to 2 × 10−8 M), was added to increase basal tension to ∼50–75% of maximal KCl contraction. Relaxations to cumulative additions of 11,12-EET and its analogs (10−9-10−5 M) were measured. In other studies, arterial rings were pretreated with vehicle (0.095% ethanol), 11,12-EET analogs, 20-H-11,12-EE8ZE (10−5 M) or 11,12,20-THE8ZE (10−5 M), or the 14,15-EET selective antagonist, 14,15-DHE5ZE (10−5 M). Relaxation responses to cumulative additions of the EETs (10−9-10−5 M), arachidonic acid (10−8-10−5 M), NS1619 (a BKCa channel opener; 10−7-10−4 M), or sodium nitroprusside (SNP) (a nitric oxide donor) (10−9-10−5 M) were recorded. In some experiments, the arteries were pretreated for 10 min with vehicle or trans-4-[4-(3-adamantan-1-yl-ureido)-cyclohexyloxy]-benzoic acid (tAUCB) (10−6 M) to inhibit sEH (1, 24). Basal tension represents tension before the addition of U-46619. Results are expressed as percent relaxation of the U-46619-treated rings, with 100% relaxation representing basal tension.

Diameter changes in perfused rat mesenteric arteries.

The Medical College of Wisconsin Animal Care and Use Committee approved all experimental procedures. Male Sprague-Dawley rats (250–400 g) were anesthetized with pentobarbital sodium (50 mg/kg ip). Two mesenteric resistance arteries were obtained from each rat, and a different experiment was conducted on each artery. Therefore, the n size represents the number of mesenteric resistance arteries and the number of rats used for each protocol. Mesenteric artery segments were mounted between two cannulas in a pressure myograph system (Danish Myo Technology model 111P). The interior and exterior of the vessel were oxygenated in 95% O2/5% CO2 Krebs physiological salt solution (in 10−3 M: 119.0 NaCl, 25.0 NaHCO3, 4.6 KCl, 1.2 KH2PO4, 1.2 MgSO4, 1.8 CaCl2, 11.0 glucose) at pH 7.4 and 37°C. Under no-flow conditions, the pressure within the vessel was increased in 10-mmHg increments from 20 to 65 mmHg. The vessel was then equilibrated at 65 mmHg for 30 min and remained at that pressure for the duration of the experiment. Lumen diameter measurements were acquired using the MyoView 1.2P user interface (DMT, Aarhus, Denmark). The control lumen diameter was calculated as the mean diameter during the last 1 min of the 30-min equilibration. Vessels were constricted with the thromboxane mimetic U-46619 (10−7 to 2 × 10−6 M), and the diameter of the constricted vessel was calculated as the mean during the last 1 min of 15 min. Vehicle, 20-H-11,12-EE8ZE (10−5 M) plus tAUCB (10−6 M) or 11,12,20-THE8ZE (10−5 M) were added to the bath solution for 10 min before EET addition. EETs (10−9-10−5 M) were added to the bathing solution every 5 min, and the change in diameter to each concentration was measured. For the flow-induced dilation studies, vessels were constricted with U-46619 (10−7 to 2 × 10−6 M), and the mesenteric artery diameter responses to flow (50 μl/min) were determined.

Whole cell Patch-clamp Measurement of K Currents in Bovine Arterial Smooth Muscle Cells

Conventional whole cell patch-clamp electrophysiology was used to measure BKCa currents in smooth muscle cells. Bovine coronary smooth muscle cells were enzymatically dissociated as previously described (4, 6). Coronary smooth muscle cells were placed in a recording bath and perfused with a whole cell solution containing the following (in mM): 140 NaCl, 4 KCl, 1 MgCl2, 2 CaCl2, 10 HEPES, and 10 glucose, pH 7.4. A recording electrode was pulled from borosilicate glass (resistance 3–6 MΩ) (P-87; Sutter Instrument, Novato, CA) and heat-polished with use of a microforge (MF-90, Narishige, China). A glass pipette that was filled with the internal pipette solution (in 10−3 M): 140 NaCl, 4 KCl, 1 MgCl2, 10 HEPES, 1 EGTA, 1 NaATP, and 1 NaGTP and 100 nM CaCl2, pH 7.2, was gently lowered onto a smooth muscle cell for successful cell-attached configuration, and a gigaohm seal was obtained. A negative pressure was briefly applied to rupture the membrane. Membrane current was recorded on an Axopatch 200B amplifier (Axon Instruments, Union City, CA) and saved on a computer for subsequent analysis with Clampfit 10.2. A 2 M KCl-agar salt bridge between the bath and the Ag-AgCl reference electrode was used to minimize offset potentials. All electrical recordings were performed at room temperature. After whole cell configuration, current recordings were obtained for at least 5 min. To determine the effect of 11,12-EET on K channel activity, channel recordings (5–6 min) were obtained in cells that were perfused and incubated with either vehicle or 11,12,20-THE8ZE (10−5 M). Subsequently, 11,12-EET (10−7 M) was added, and, after 3 min of incubation, K channel activity was again recorded.

Metabolism of 20-H-11,12-EE8ZE by Bovine Coronary Arteries

Bovine coronary arteries were prepared as described above (1, 29). 20-H-11,12-EE8ZE (10−5 M) was incubated for 0, 10, or 30 min at 37°C in HEPES buffer (in 10−3 M: 10 HEPES, 150 NaCl, 5 KCl, 2 CaCl2, 1 MgCl2, and 6 glucose, pH 7.4) in the presence and absence of coronary artery rings. The 20-H-11,12-EE8ZE metabolism was repeated in the presence and absence of the sEH inhibitor, tAUCB (10−6 M) (24). The conversion of 20-H-11,12-EE8ZE to 11,12,20-THE8ZE represents sEH activity.

After the incubation, the samples were subjected to solid-phase extraction using C18 Bond Elut columns. The samples were dried under a stream of nitrogen and analyzed by liquid chromatography-electrospray ionization-mass spectrometry (Agilent 1100 LC/MSD, SL model), as previously described (1, 25). 20-H-11,12-EE8ZE and 11,12,20-THE8ZE were measured in the selected ion monitoring mode by detecting the mass-hydrogen ions of m/z 340 and m/z 358, respectively. In parallel analyses, the migration times of known standards (20-H-11,12-EE8ZE to 11,12,20-THE8ZE) were determined.

Statistical Analysis

The data are expressed as means ± SE. Statistical analysis was performed by a one-way analysis of variance, followed by the Student-Newman-Keuls multiple-comparison test when significant differences were present. P < 0.05 was considered statistically significant.


Effect of 11,12-EET Analogs on Vascular Tone of Bovine Coronary Arteries

Fifteen 11,12-EET analogs were tested for agonist activity on bovine coronary arterial rings (Fig. 2). The analogs with low agonist activity were also tested for their ability to inhibit relaxations to 11,12-EET. Compared with 11,12-EET, removal of one or two olefins reduced the agonist activity to varying extents. The analogs with a single Δ8 or Δ5 olefin were better agonists than analogs with a single Δ14 olefin or no olefins (Fig. 2). Replacement of the epoxide oxygen with an ether oxygen had little effect on agonist activity; however, activity was greatly reduced by replacement of the epoxide oxygen with a carbon or a sulfur, a thiirane group. None of the epoxyeicosanoic acids, epoxyeicosadienoic acid, or the thiirane altered the relaxations to 11,12-EET, so they were not antagonists. Only 20-H-11,12-EE8ZE and 11,12,20-THE8ZE had low agonist activity and inhibited the relaxations to 11,12-EET. As a result, these analogs were studied further.

Effect of 20-H-11,12-EE8ZE on EET-induced Relaxations of Bovine Coronary Arteries

In the U-46619 preconstricted arteries, 14,15-, 11,12-, 8,9-, and 5,6-EET caused concentration-dependent relaxations, ranging from 80 to 96%. (Fig. 3, A–D, respectively). Preincubation of arteries with 20-H-11,12-EE8ZE (10−5 M) inhibited relaxations to 14,15- and 11,12-EET, but not 5,6- and 8,9-EET; however, the inhibitory effect was greatest against 11,12-EET (maximum relaxation of 80.6 ± 4.6% with 11,12-EET alone and 26.7 ± 7.4% with 20-H-11,12-EE8ZE plus 11,12-EET). SNP (10−9-10−5 M), a nitric oxide donor, and NS1619 (10−7-10−4 M), a BKCa channel opener, caused concentration-related relaxations that were unchanged by pretreatment with 20-H-11,12-EE8ZE (10−5 M) (Fig. 4).

Fig. 3.

Effect of 20-H-11,12-EE8ZE on EET-induced relaxations in BCAs. Arteries were pretreated with vehicle or 20-H-11,12-EE8ZE (10−5 M) before preconstriction with U-46619 (10−8 to 2 × 10−8 M). Increasing concentrations of 14,15- (A), 11,12- (B), 8,9- (C), and 5,6-EET (D) were added, and changes in isometric tension measured. Values are means ± SE. *P < 0.05. **P < 0.01, and ***P < 0.001 (significant difference compared with control).

Fig. 4.

Effect of 20-H-11,12-EE8ZE (A and C) and 11,12,20-THE8ZE (B and D) on the relaxations to sodium nitroprusside (SNP), a nitric oxide (NO) donor (C and D), and NS1619, a large-conductance calcium-activated (BKCa) channel opener (A and B), in BCAs. Arteries were pretreated with vehicle, 20-H-11,12-EE8ZE (10−5 M), or 11,12,20-THE8ZE (10−5 M) before preconstriction with U-46619 (10−8 to 2 × 10−8 M). Increasing concentrations of SNP or NS1619 were added, and changes in isometric tension were measured. Values are means ± SE.

Metabolism of 20-H-11,12-EE8ZE in Bovine Coronary Arteries

To determine whether 20-H-11,12-EE8ZE is a substrate for sEH and metabolized by the bovine coronary arteries, segments of arteries were incubated with 20-H-11,12-EE8ZE, and its conversion to 11,12,20-THE8ZE was measured by liquid chromatography/mass spectrometry. As shown in Fig. 5A, no conversion of 20-H-11,12-EE8ZE to 11,12,20-THE8ZE occurred at 0, 10, or 30 min in the absence of arteries. However, incubation of arteries with 20-H-11,12-EE8ZE resulted in a time-dependent increase in 11,12,20-THE8ZE formation (Fig. 5B). The metabolism of 20-H-11,12-EE8ZE to 11,12,20-THE8ZE was completely blocked by treatment of arteries with the sEH inhibitor tAUCB (10−6 M) (Fig. 5C) (24). Interestingly, when 20-H-11,12-EE8ZE was incubated with coronary artery rings (Fig. 5B), the increase of the peak corresponding to 11,12,20-THE8ZE was less than the decrease in the size of the 20-H-11,12-EE8ZE peak. The reason for this discrepancy is not known. However, it is possible that 11,12,20-THE8ZE is further metabolized or incorporated into membrane phospholipids, thus reducing its availability or the ionization of the two eicosanoids in the mass spectrometer differ. These data indicate that sEH is responsible for the conversion of 20-H-11,12-EE8ZE to 11,12,20-THE8ZE in coronary arteries.

Fig. 5.

Metabolism of 20-H-11,12-EE8ZE by BCA rings and the effect of sEH inhibition with tAUCB. 20-H-11,12-EE8ZE (10−5 M) was incubated for 0, 10, and 30 min in the absence (A) and presence (B and C) of coronary arteries. C: arteries were preincubated for 30 min with tAUCB (10−6 M) before their incubation with 20-H-11,12-EE8ZE. Medium was extracted and analyzed by liquid chromatography-electrospray ionization-mass spectrometry. Conversion of 20-H-11,12-EE8ZE to 11,12,20-THE8ZE is an index of sEH activity.

Effect of tAUCB on the Inhibition of EET-induced Relaxations by 20-H-11,12-EE8ZE

Since sEH converts 20-H-11,12-EE8ZE to 11,12,20-THE8ZE, the influence of sEH inhibition by tAUCB on the antagonist properties of 20-H-11,12-EE8ZE was determined. In the U-46619-preconstricted arteries, 11,12- and 14,15-EET caused concentration-dependent relaxation of arterial rings with maximum relaxation of 74.4 ± 5.8 and 86.6 ± 4.4%, respectively (Fig. 6, A and B, respectively). Pretreatment with tAUCB (10−6 M) alone had no significant effect on 11,12- and 14,15-EET-induced relaxations. Pretreatment of arteries with tAUCB markedly enhanced the antagonist effect of 20-H-11,12-EE8ZE against 14,15-EET-induced relaxations, with maximal relaxations reduced from 64.7 ± 5.2% without tAUCB to 27.8 ± 3.3% with tAUCB (Figs. 3A and 6A). 11,12-EET-induced relaxations were inhibited to a similar extent by 20-H-11,12-EE8ZE in the presence and absence of tAUCB (Figs. 3B and 6B). These results suggest that 20-H-11,12-EE8ZE is a better antagonist of 14,15-EET, if its metabolism by sEH is inhibited.

Fig. 6.

Effect of 20-H-11,12-EE8ZE on 11,12- (A) and 14,15-EET-induced relaxations (B) in the presence of the sEH inhibitor tAUCB in BCAs. Arteries were preincubated for 30 min with tAUCB (10−6 M) before addition of 20-H-11,12-EE8ZE (10−5 M) and constriction with U-46619 (10−8 to 2 × 10−8 M). Relaxations to 11,12- and 14,15-EET were measured. Values are means ± SE. ***P < 0.001 (significant difference compared with control).

Effect of 11,12,20-THE8ZE on EET-induced Relaxations in Bovine Coronary Arteries

The effect and specificity of 11,12,20-THE8ZE was tested on relaxations to the four EET regioisomers. The four EETs relaxed the preconstricted arteries to a similar extent (Fig. 7). 11,12,20-THE8ZE (10−5 M) inhibited the relaxations to 11,12-EET (Fig. 7B), but did not inhibit relaxations to 5,6-, 8,9-, and 14,15-EET (Fig. 7, A, C, and D). 11,12,20-THE8ZE and 20-H-11,12-EE8ZE inhibited the relaxations to 11,12-EET to a similar extent (Figs. 7B and 3B). 11,12,20-THE8ZE, unlike 20-H-11,12-EE8ZE, was a selective inhibitor of 11,12-EET. 11,12,20-THE8ZE did not affect the relaxations to SNP or NS1619 (Fig. 4).

Fig. 7.

Effect of 11,12,20-THE8ZE on EET-induced relaxations in BCAs. Arteries were pretreated with vehicle or 11,12,20-THE8ZE (10−5 M) before their constriction with U-46619 (10−8 to 2 × 10−8 M). 14,15- (A), 11,12- (B), 8,9- (C), and 5,6-EET (D) were added in increasing concentrations, and changes in isometric tension measured. Values are means ± SE. ***P < 0.001 (significant difference compared with control).

Effect of 11,12,20-THE8ZE on BKCa Channel Activation by 11,12-EET in Coronary Smooth Muscle

Since 11,12-EET relaxes coronary arteries by activation of BKCa channels and membrane hyperpolarization, the effect of 11,12,20-THE8ZE on 11,12-EET-induced K channel activation was determined using the patch-clamp method in the whole cell mode. Coronary arterial smooth muscle cells showed an outward current in response to increasing steps in the pipette voltage (Fig. 8). These currents are blocked by iberiotoxin (10−7 M), indicating that these currents are mediated by BKCa channels (data not shown). The magnitude of the K currents increased significantly with the addition of 11,12-EET (10−7 M). In contrast, 11,12,20-THE8ZE (10−5 M) did not alter the K currents. 11,12,20-THE8ZE blocked the increase in K current magnitude by 11,12-EET.

Fig. 8.

Effect of 11,12,20-THE8ZE on coronary smooth muscle cell BKCa channel activation by 11,12-EET. Using the whole cell patch-clamp mode, current changes were monitored to stepwise increases in pipette voltage. The effect of vehicle, 11,12-EET (10−7 M), 11,12,20-THE8ZE (10−5 M), and 11,12-EET plus 11,12,20-THE8ZE were measured. Values are means ± SE. *P < 0.05.

Effect of 20-H-11,12-EE8ZE and 11,12,20-THE8ZE on EET-induced Dilation of Rat Mesenteric Arteries

The effects of 20-H-11,12-EE8ZE and 11,12,20-THE8ZE on EET-induced dilation was determined in resistance arteries from the rat mesenteric circulation. Both 14,15-EET and 11,12-EET dilated rat mesenteric arteries in similar concentrations (10−9-10−5 M) (Fig. 9, A and B, respectively). 20-H-11,12-EE8ZE in the presence of tAUCB inhibited the dilations to both 11,12- and 14,15-EET to a similar extent. In contrast, 11,12,20-THE8ZE inhibited the dilations to 11,12-EET and slightly increased the dilations to 14,15-EET. Thus 11,12,20-THE8ZE is a selective 11,12-EET antagonist in both bovine coronary arteries and rat mesenteric arteries.

Fig. 9.

Effect of EET analogs on dilation of rat mesenteric resistance arteries to 14,15-EET (A) and 11,12-EET (B). Rat mesenteric arteries were cannulated and perfused. Arteries were preconstricted with U-46619 (10−7 to 2 × 10−6 M), and changes in diameter were monitored in response to increasing concentrations of EETs. Arteries were pretreated with vehicle, 20-H-11,12-EE8ZE (10−5 M), and tAUCB (10−6 M) or 11,12,20-THE8ZE (10−5 M). Values are means ± SE.

Contribution of 14,15- and/or 11,12-EET to Arachidonic Acid-induced Relaxations and Flow-induced Dilation Using 11,12,20-THE8ZE and 14,15-DHE5ZE

Arachidonic acid relaxed coronary arteries in a concentration-related manner (Fig. 10A). These relaxations were unchanged by pretreatment with 11,12,20-THE8ZE, an 11,12-EET antagonist, but were inhibited by ∼63% by 14,15-DHE5ZE, the 14,15-EET antagonist (1). Increasing flow from 0 to 50 μl/min increased the diameter of rat mesenteric arteries by 37% (Fig. 10B). This flow-induced dilation was inhibited by ∼50% by 14,15-EE5ZE and 14,15-DHE5ZE, but unchanged by 11,12,20-THE8ZE. Thus 14,15-EET, but not 11,12-EET, mediate relaxations to arachidonic acid and increases in flow.

Fig. 10.

Effect of the 14,15-EET antagonist, 14,15-dihydroxy-eicosa-5(Z)-enoic acid (14,15-DHE5ZE), and the 11,12-EET antagonist, 11,12,20-THE8ZE, on arachidonic-induced relaxations in BCAs (A) and flow-induced dilation in rat mesenteric arteries (B). Arteries were pretreated with vehicle or 14,15-DHE5ZE (10−5 M), or 11,12,20-THE8ZE (10−5 M), preconstricted with U-46619 (10−8 to 2 × 10−8 M for BCAs, 10−7 to 2 × 10−6 M for rat mesenteric arteries), and arachidonic acid (AA) was added, or changes in flow (50 μl/min) produced. Changes in isometric tension (A) or artery diameter (B) were measured. Each value represents means ± SE. *P < 0.05.


CYP450 epoxygenases add oxygen across the double bonds of arachidonic acid to produce four regioisomeric cis-epoxides: 5,6-, 8,9-, 11,12-, and 14,15-EETs (7, 33). In bovine coronary arteries, the four EET regiosiomers are equipotent in causing relaxation (29). EETs activate membrane BKCa channels to induce hyperpolarization and vascular relaxation (3, 5, 6, 12). The endothelium-dependent relaxations and hyperpolarizations to acetylcholine and bradykinin are blocked by CYP450 inhibitors, 14,15-EE5ZE, and BKCa channel inhibitors. Thus EETs act as endothelium-derived hyperpolarizing factors. In addition, EETs have other important vascular actions (22, 31).

EETs are often described collectively when discussing their biological activities or roles as mediators of a biological response. In some instances, such as coronary vasorelaxation and endothelial cell proliferation, the EET regioisomers are equipotent (27, 29). However, with other activities, EET regioisomers differ markedly. For example, 11,12-EET is the most potent EET regioisomer in decreasing vascular cell adhesion molecule-1 expression in endothelial cells (26), in inhibiting PDGF-stimulated smooth muscle cell migration (32), in dilating rat renal arterioles (19, 35), and in stimulating platelet membrane hyperpolarization (21). 14,15-EET is more potent than the other isomers in promoting monocyte adherence to the endothelium and enhancing endothelial cell survival (1, 28). Adding to this complexity, CYP450 isozymes differ in the proportion of the EET regioisomers that they synthesize, and cells differ in the CYP450 isozymes that are expressed (33). Thus there is a need to determine the contribution of individual EET regioisomers to biological effects. Development of regioisomer-specific antagonists is one approach.

14,15-EE5ZE inhibits relaxations to the four EET regioisomers (13), while modification at its carboxyl group to a methylsulfonylimide changes its pharmacological properties (15). 14,15-EE5ZE-methysulfonylimide inhibits the relaxations to 14,15- and 5,6-EET but does not change relaxations to 11,12- or 8,9-EET (15). Metabolism of 14,15-EE5ZE alters its antagonist specificity. 14,15-DHE5ZE, the sEH metabolite of 14,15-EE5ZE, inhibits relaxations to 14,15-EET, but does not alter relaxations to other EET regioisomers (1). These findings imply that EET regioisomers may have distinct binding sites/receptors and raise the possibility for development of other regioisomer-specific antagonist that will have great utility in uncovering the biological roles of regioisomeric EETs.

Earlier studies have revealed diverse structural requirements for the agonist and antagonist activities of EETs analogs (9, 1315). 14,15-EET has distinct structural requirements for full agonist activity: a C-1 carboxyl, Δ8 double bond, 20 carbons, and a cis-epoxide (9). While the present study of 11,12-EET analogs was not as comprehensive, 11,12-EET and 14,15-EET share structural features for full agonist activity, including a Δ8 double bond and an epoxide. Of the fifteen 11,12-EET analogs tested, only 20-H-11,12-EE8ZE and 11,12,20-THE8ZE had weak agonist activity and EET antagonist activity. The antagonist activity was described in both bovine coronary and rat mesenteric arteries. 20-H-11,12-EE8ZE significantly inhibited the relaxations to 11,12-EET and 14,15-EET without altering relaxations to 5,6- and 8,9-EET. The antagonist effect of 20-H-11,12-EE8ZE was greater against relaxations to 11,12-EET than 14,15-EET. This suggests that the two regioisomers, 11,12-EET and 14,15-EET, may have different binding sites or receptors, and 20-H-11,12-EE8ZE preferentially binds to the site of 11,12-EET. Preincubation of arteries with tAUCB, a sEH inhibitor, markedly enhanced the antagonist effect of 20-H-11,12-EE8ZE against 14,15-EET relaxations, indicating that 20-H-11,12-EE8ZE is a better inhibitor of 14,15-EET if its metabolism by sEH is blocked (24). Supporting this possibility, bovine coronary arteries converted 20-H-11,12-EE8ZE to 11,12,20-THE8ZE, and the metabolism was blocked in the presence of tAUCB. This finding confirms earlier studies that EETs and their analogs with an intact epoxide group are metabolized to DHETs by sEH in bovine coronary arteries (1, 2, 13). 20-H-11,12-EE8ZE inhibited 11,12-EET relaxations to a similar extent in the presence and absence of tAUCB, suggesting that the intact epoxide group of 20-H-11,12-EE8ZE is necessary for its antagonist effect against 14,15-EET relaxations, but not necessary for inhibition of 11,12-EET relaxations. As a result, we tested the antagonist activity of 11,12,20-THE8ZE, the sEH metabolite of 20-H-11,12-EE8ZE. 11,12,20-THE8ZE specifically inhibited the relaxations to 11,12-EET without affecting relaxations to 14,15-EET. Thus the Δ8 double bond and vicinal diol favor the 11,12-EET antagonist activity of 11,12,20-THE8ZE.

Structure activity studies of EET analogs have revealed that changing the double bond of 14,15-EEZE from the Δ8 to the Δ5 position converts an EET agonist to a nonselective EET antagonist (9, 14). Hydration of the 14,15-epoxide of 14,15-EE5ZE to the vicinal diol, 14,15-DHE5ZE, makes the antagonist selective for 14,15-EET (1). The structure-activity relationship for 11,12-EET antagonist activity differs from 14,15-EET. Surprisingly, addition of a 20-hydroxy group to 11,12-EE8ZE converts the agonist analog to an antagonist of 11,12- and 14,15-EET. As with 14,15-EE5ZE, hydration of the epoxide of 20-H-11,12-EE8ZE confers specificity for a single regioisomer. 11,12,20-THE8ZE, with Δ8, 11,12-vicinal diol, and 20-hydroxy group, is a selective inhibitor of only 11,12-EET. Interestingly, 20-H-11,12-EE8ZE and its metabolite 11,12,20-THE8ZE exhibited consistent effects in rat mesenteric arterioles and inhibited relaxations to 11,12-EET with almost similar potencies. 11,12-EET is considered the predominant EET in rat renal arterioles (18, 35), and this finding further substantiates the selective inhibitory activity of 11,12,20-THE8ZE in bovine coronary arteries.

EET-regioisomer-specific antagonists are important pharmacological tools that can be used to dissect the contribution(s) of the EET-regioisomers to physiological processes, pathological conditions, and pharmacological treatments. As examples of their utility, we tested the effects of the 14,15-EET antagonist, 14,15-DHE5ZE, and the 11,12-EET antagonist, 11,12,20-THE8ZE, on relaxations to arachidonic acid and flow-induced dilation. In coronary arteries treated with indomethacin, the endothelium-dependent relaxations to arachidonic acid are mediated by the EETs (6, 29). These relaxations were inhibited by 14,15-DHE5ZE, but not by 11,12,20-THE8ZE. Similarly, increases in flow and shear stress dilate arteries by an endothelium-dependent mechanism (16, 23). Endothelial EETs mediate a portion of the dilation. In rat mesenteric arteries, the flow-induced dilation was inhibited in part by 14,15-EE5ZE and 14,15-DHE5ZE, but not by 11,12,20-THE8ZE. These studies indicate that 14,15-EET, but not 11,12-EET, mediate arachidonic acid- and flow-induced dilation and indicate the utility of these EET-regioisomer-specific antagonists in identifying the regioisomer mediating physiological responses.

In summary, the data from our present study show that bovine coronary arteries covert 20-H-11,12-EE8ZE to 11,12,20-THE8ZE. 20-H-11,12-EE8ZE inhibits relaxations to both 11,12- and 14,15-EET. However, it metabolite, 11,12,20-THE8ZE, specifically inhibits 11,12-EET. Furthermore, our present findings imply that EET regioisomers have different binding sites/receptors, and modification of EET structure would be expected to furnish antagonists with EET regioisomeric specificity. The discovery of EET-regioisomeric-specific antagonists will be helpful in establishing the physiological functions of endogenous EETs and development of EET mimetics for the treatment of cardiovascular diseases.


This work was supported by the National Heart, Lung, and Blood Institute (HL-51055 and HL-59699), the National Institute of General Medical Sciences (GM-31278), the Robert Welch Foundation (GL 625910), and the Higher Education Commission, Pakistan (fellowships to I. A. Bukhari and A. J. Shah).


No conflicts of interest, financial or otherwise, are declared by the author(s).


I.A.B., A.J.S., K.M.G., S.R.K., J.D.I., J.R.F., and W.B.C. conception and design of research; I.A.B., A.J.S., K.M.G., K.A.W., and S.R.K. performed experiments; I.A.B., A.J.S., K.M.G., K.A.W., S.R.K., J.D.I., and J.R.F. analyzed data; I.A.B., A.J.S., K.M.G., K.A.W., S.R.K., J.D.I., J.R.F., and W.B.C. interpreted results of experiments; I.A.B., A.J.S., K.M.G., and K.A.W. prepared figures; I.A.B., A.J.S., and W.B.C. drafted manuscript; A.J.S., K.M.G., J.D.I., J.R.F., and W.B.C. edited and revised manuscript; K.M.G., J.D.I., J.R.F., and W.B.C. approved final version of manuscript.


The authors thank Gretchen Barg for secretarial assistance, and Sarah Christian and Marilyn Isbel for technical assistance. Dr. Bruce Hammock of the University of California at Davis kindly provided the sEH inhibitor, tAUCB.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
View Abstract