Heart and Circulatory Physiology

Regulation of potassium channels in coronary smooth muscle by adenoviral expression of cytochrome P-450 epoxygenase

William B. Campbell, Blythe B. Holmes, John R. Falck, Jorge H. Capdevila, Kathryn M. Gauthier


Epoxyeicosatrienoic acids (EETs) are endothelium-derived cytochrome P-450 (CYP) metabolites of arachidonic acid that relax vascular smooth muscle by large-conductance calcium-activated potassium (BKCa) channel activation and membrane hyperpolarization. We hypothesized that if smooth muscle cells (SMCs) had the capacity to synthesize EETs, endogenous EET production would increase BKCa channel activity. Bovine coronary SMCs were transduced with adenovirus coding the CYP Bacillus megaterium -3 (F87V) (CYP BM-3) epoxygenase that metabolizes arachidonic acid exclusively to 14(S),15(R)-EET. Adenovirus containing the cytomegalovirus promoter-Escherichia coli β-galactosidase was used as a control. With the use of an anti-CYP BM-3 (F87V) antibody, a 124-kDa immunoreactive protein was detected only in CYP BM-3-transduced cells. Protein expression increased with increasing amounts of virus. When CYP BM-3-transduced cells were incubated with [14C]arachidonic acid, HPLC analysis detected 14,15-dihydroxyeicosatrienoic acid (14,15-DHET) and 14,15-EET. The identity of 14,15-EET and 14,15-DHET was confirmed by mass spectrometry. In CYP BM-3-transduced cells, methacholine (10−5 M) increased 14,15-EET release twofold and BKCa channel activity fourfold in cell-attached patches. Methacholine-induced increases in BKCa channel activity were blocked by the CYP inhibitor 17-octadecynoic acid (10−5 M). 14(S),15(R)-EET was more potent than 14(R),15(S)-EET in relaxing bovine coronary arteries and activating BKCa channels. Thus CYP BM-3 adenoviral transduction confers SMCs with epoxygenase activity. These cells acquire the capacity to respond to the vasodilator agonist by synthesizing 14(S),15(R)-EET from endogenous arachidonic acid to activate BKCa channels. These studies indicate that 14(S),15(R)-EET is a sufficient endogenous activator of BKCa channels in coronary SMCs.

  • endothelium-derived hyperpolarizing factor
  • arachidonic acid
  • vascular relaxation

endothelial cells regulate vascular tone through the release of soluble mediators, including prostacyclin, nitric oxide, and epoxyeicosatrienoic acids (EETs). EETs are cytochrome P-450 (CYP) metabolites of arachidonic acid (4, 33). The four EET regioisomers (14,15-, 11,12-, 8,9-, and 5,6-EET) are made by the endothelium and are released in response to agonists such as bradykinin and acetylcholine (3, 7, 12, 18, 30). With the use of several bioassay methods, bradykinin has been shown to release a transferable factor that relaxes and hyperpolarizes vascular smooth muscle (13, 16, 31). The release of this endothelium-derived hyperpolarizing factor (EDHF) is blocked by inhibitors of CYP (16). Like EDHF, EETs exert their action on vascular smooth muscle. They hyperpolarize the smooth muscle membrane and cause relaxation (32). For these reasons, EETs represent EDHFs.

There are strict structural requirements for the vascular activity of the 14,15-EET molecule. For full agonist activity, a carbon-1 carboxyl, a 8,9-double bond, a 14(S),15(R)-cis epoxide, and 14 carbons between the carboxyl and epoxide are required (6, 14). Changes in these key features reduce agonist potency and/or efficacy. EETs exert their action by promoting the opening of large-conductance, calcium-activated potassium (BKCa) channels in the vascular smooth muscle (3, 11). This potassium channel activation requires a guanine nucleotide binding protein, possibly Gs (24, 25). Potassium channel opening mediates the hyperpolarization and is blocked by potassium channel blockers such as tetraethylammonium, charybdotoxin, and iberiotoxin (3, 24, 25, 32, 37).

EETs are made by the vascular endothelium but not vascular smooth muscle (33). As a result, we hypothesized that if smooth muscle cells were given the ability to synthesize EETs, the endogenous EETs would act in an autocrine manner to activate BKCa channels. To test this possibility, bovine coronary smooth muscle cells were transduced with an adenovirus coding the CYP Bacillus megaterium (BM)-3 (CYP 102) with phenylalanine (F) substituted for a valine (V) at amino acid 87 (CYP BM-3) (17). CYP BM-3 is a 124-kDa fusion protein containing CYP epoxygenase domain and CYP reductase domain (26, 35). The epoxygenase is regio- and stereoselective, producing only 14(S),15(R)-EET (17). We have shown that EET-induced vasorelaxation is stereospecific with the 14(S),15(R)-EET being more potent and more efficacious than the 14(R),15(S)-EET (14). Thus the CYP BM-3 enzyme produces exclusively the active 14,15-EET enantiomer. These studies demonstrate that infection of smooth muscle cells with a CYP BM-3 adenovirus conferred the ability to synthesize 14,15-EET. The endogenous release of 14,15-EET by smooth muscle was stimulated by methacholine and resulted in the activation of BKCa channels. These studies further substantiate a role for EETs in the regulation of potassium channel activity and vascular tone.


Culture of coronary arterial smooth muscle cells.

Culture of coronary arterial smooth muscle cells was performed as described previously (33). Briefly, after separate enzymatic dispersion of arterial endothelial cells, the coronary artery tissue was laid lumen side down, and the smooth muscle cells were allowed to explant from the artery onto a 60-mm tissue culture dish. Cells were grown in medium 199 containing 10% bovine serum, 1% antibiotics, and 1% l-glutamine (growth media). The flasks were placed in a 5% CO2 air incubator at 37°C. The culture media were replaced every other day. After 4 days, the artery was removed and the cells were allowed to grow to 50% confluence. Cells were then transferred to the appropriate flask or dish and used at a confluency of 80%. Cells were used before five passages.

Viral infection of smooth muscle cells with CYP BM-3 (F87V) adenovirus.

Before transduction, smooth muscle cells were plated at 4.3 × 105 cells/100 mm dish for Western blot and liquid chromatography (LC)/mass spectrometry (MS) studies. For immunofluorescence and patch-clamping studies, cells were grown in six-well plates at a plating density of 5 × 104 cells/well. The plating density of the cells for [14C]arachidonic acid metabolism was 9.5 × 105 cells/75 cm2 flask. The cells were transfected with a replication defective cytomegalovirus promoter-Escherichia coli β-galactosidase (CMV-β-Gal) adenovirus or CMV promoter-CYP BM-3 F87V mutant (CYP BM-3) adenovirus (5). The cells were incubated at 37°C in 95% air-5% CO2 with 2 ml of RPMI containing 2% BSA per well containing various amounts of the virus.

During the initial transduction period, the plates were gently rocked every 10 min for 1 h to allow optimal distribution of viral particles. After 1 h of incubation, growth medium was added to a final volume to 1 ml. Cells were then incubated for various times. Optimal multiplicity of infection (MOI) was determined using various amounts (MOI of 0–1,000) of the CMV-β-Gal adenovirus. For determination of transduction efficiency, transduced cells were washed three times in PBS (pH 7.4) and fixed in PBS containing 1% glutaraldehyde and 1 mM MgCl2 for 15 min at room temperature. Cells expressing the β-Gal gene product were incubated with 5 mM K4Fe(CN)6, 5 mM K3Fe(CN)6, 2 mM MgCl2, and 0.2% 5-bromo-4-chloro-3-indoyl-β-d-galactopyranoside. The β-Gal-expressing cells were identified by the generation of blue color within the cytosol. Optimum blue staining of all cells was used to determine the optimal MOI and incubation time. These data were used as a starting point for further experiments with the CYP BM-3 adenovirus using Western blot to determine the expression of CYP BM-3 enzyme. Optimum transfection conditions were 24 h incubation with an MOI of 750–1,000.

Western blot method for CYP BM-3 (F87V) and BKCa channel proteins.

Protein (20 μg) from lysed cells was boiled for 5 min in a water bath and loaded onto 10% SDS-PAGE mini gels, and the proteins were resolved after 1 h at 180 V. The proteins were transferred electrophoretically to nitrocellulose for 1 h at 100 V in transfer buffer (25 mM Tris base, 192 mM glycine, and 20% methanol). Membranes were preblocked in Tris base and sodium chloride (TBS) with 10% milk overnight at 4°C. The membranes were washed twice with TBS-T (Tris-buffered saline with 10% Tween 20) for 5 min and three more times with TBS for 5 min. Primary antibodies for either CYP BM-3 (F87V) or BKCa channel α-subunit were applied at a dilution of 1:5,000 or 1:1,000, respectively, for 3 h at room temperature. Membranes were washed two times with TBS-T and three times with TBS and then incubated for 1 h with 1:3,000 horseradish peroxidase-labeled goat anti-rabbit IgG. After washing the membranes five times for 5 min each, 2 ml of Detection Solutions 1 and 2 (1:1) (Amersham, IL) were added directly to the blots on the surface carrying the proteins. After incubation for 1 min at room temperature, the membranes were wrapped in Saran Wrap and exposed to Kodak ML-1 film.

Immunohistological methods.

Briefly, cells were grown in six-well plates at a density of 5 × 104 cells/well. The cells were transduced with CYP BM-3 adenovirus for 24 h. The cells were then washed with PBS buffer and fixed using 50:50 of methanol:ethanol for 10 min at 4°C. Cells were washed three times with PBS/0.05% Pluronic F-127 and then blocked with 5% goat serum in PBS/0.05% Pluronic F-127 for 30 min at room temperature. Cells were washed 3 times as above and incubated with CYP BM-3 (F87V) primary antibody (1:500) for 15 min at room temperature. The cells were washed as above, the secondary antibody (Alexa 594 goat anti-rabbit IgG diluted 1:500) in PBS/0.05% Pluronic F-127 was added, and cell were incubated for 15 min at room temperature. Cells were then washed six times with PBS, and fluorescence was viewed using a Nikon Diaphot fluorescence microscope.

Metabolism of [14C]arachidonic acid by transduced cells.

Cells were used at a density of 9.5 × 105 cells/75 cm2 flask. Cells transduced with either CYP BM-3 adenovirus or the CMV-β-Gal control adenovirus for 24 h were washed two times with HEPES buffer and incubated with [U-14C]arachidonic acid (0.5 μCi, 10−7 M) for 10 min at 37°C. The calcium ionophore A23187 (2 × 10−5 M) was added, and the incubation continued for 20 min. After incubation, the cells were scraped, and the buffer and cells were extracted.

Solid-phase extraction of arachidonic acid metabolites.

The frozen samples were thawed on ice. For LC/electrospray ionization (ESI)/MS, internal standards, 1 ng each for [2H8]EETs, 0.8 ng each for [2H8]DHETs, and 0.8 ng for 20-[2H2]HETE were added. Ethanol was added to the final concentration of 25%, and the samples were vortexed, sonicated, and centrifuged at 1,500 rpm for 3 min (27). The supernatant was applied to the C18 Bond Elut SPE columns that had been preconditioned with 5 ml of ethanol and 15 ml of water. The columns were washed with 20 ml of water and allowed to run dry. The eicosanoids were eluted from the column with 5 ml of ethyl acetate and dried under the stream of nitrogen. For LC/ESI/MS, the samples were dissolved in 20 μl of acetonitrile, transferred to an insert in a sample vial, and frozen (−80°C). The samples for HPLC analysis were dried under nitrogen and frozen (−80°C).

Separation of [14C]arachidonic acid metabolites by HPLC.

Samples were resolved by reverse-phase (Nucleosil-C18 column, 5 μm, 4.6 × 250 mm) HPLC using solvent system I. Solvent A was water, and solvent B was acetonitrile containing 0.1% glacial acetic acid. The program was a 40-min linear gradient from 50% B in A to 100% solvent B. Flow rate was 1 ml/min. The column eluate was collected in 0.2 ml fractions and analyzed by liquid scintillation counting.

Incubation of smooth muscle cells with methacholine.

Cells were grown to a confluency of 4.3 × 105 cells/100 mm dish and transduced with CYP BM-3 or CMV-β-Gal control adenovirus for 24 h. Cells were washed three times in HEPES buffer (in mM: 10 HEPES, 155 NaCl, 5 KCl, 1.8 CaCl2, 1.0 MgCl2, 5.5 glucose, pH 7.4) and incubated with 5 ml of HEPES buffer containing methacholine (10−5 M) for 15 min at 37°C. After the incubation, the CYP inhibitor miconazole (10−5 M) was added to stop arachidonic acid metabolism. The cells were scraped, and the cells and media were transferred to a 50-ml conical tube, sonicated, vortexed, and extracted for LC/ESI/MS analysis.

LC/MS measurements.

Samples were analyzed by LC/ESI/MS (Agilent 1100 LC/MSD, SL model) as previously described (27). The samples were separated on a reverse-phase C18 column (Kromasil, 5 μm, 2 × 150 mm) using water-acetonitrile with 0.005% acetic acid as the mobile phase (solvent system II) at the flow rate of 0.200 ml/min. Drying gas was nitrogen at a flow of 12 l/min. The drying gas temperature was 350°C, nebulizer pressure was 35 pounds per square inch gauge, vaporizer temperature was 325°C, capillary voltage was 3,000 V, and fragmentor voltage was 120 V. Detection was made in the negative mode. For quantitative measurements, mass/charge (m/z) of 319, 327, 337, 345, 319, and 321 were used for EETs, [2H8]EETs, DHETs, [2H8]DHETs, 20-HETE, and 20-[2H2] HETE, respectively. The standard curves were constructed over the range of 1–100 pg/injection. Concentrations of these eicosanoids were calculated by comparing their ratios of peak areas to the standard curves.

Vascular reactivity of bovine coronary arteries.

Bovine hearts were purchased from a local slaughterhouse. The left anterior descending coronary artery was dissected and cleaned of connective tissue. Arteries with an intact endothelium were cut into 2-mm diameter rings (3 mm width), and isometric tension was measured as previously described (3, 6, 32). Arteries were stored in Krebs buffer (in mM: 119 NaCl, 4.8 KCl, 24 NaHCO3, 1.2 KH2PO4, 3.2 CaCl2, 1.2 MgSO4, 11 glucose, and 0.02 EDTA). Basal tension was set at the length-tension maximum of 3.5 g, and the arteries were equilibrated for 1.5 h. KCl (40–60 mM) was added to the chamber until maximum contractions were maintained. The thromboxane agonist U-46619 (10–20 nM) was used to preconstrict the arterial rings from basal tension to between 50% and 90% of the maximal KCl contraction. Cumulative concentrations of racemic 14,15-EET, 14(S),15(R)-EET, or 14(R),15(S)-EET were added to the chamber. For racemic 14,15-EET, the responses were repeated after the addition of iberiotoxin (10−7 M) 10 min before U46619. Tension is presented as percentage of relaxation, where 100% relaxation is basal pre-U-46619 tension.

Patch-clamp studies.

Cell-attached, single-channel potassium currents were recorded in control cultured bovine coronary smooth muscle cells or smooth muscle cells transduced with CMV-β-Gal or CYP BM-3 adenovirus using patch-clamp procedures as previously described (2, 3, 16, 24, 25, 32). Currents were sampled at 3 kHz and filtered at 1 kHz at a membrane potential of +60 mV. Perfusate and pipette solutions contained (in mM) 145 KCl, 1 MgCl2, 1 EGTA, and 10 HEPES, as well as 100 nM ionized Ca2+ (pH = 7.4). Cells were perfused and incubated with either vehicle or 14(S),15(R)-EET and 14(R),15(S)-EET (10−7–10−5 M), and the potassium channel recordings were obtained for 2–4 min for each treatment. In additional studies, the cells were incubated for 10 min with vehicle or 17-octadecynoic acid (ODYA, 10−5 M). Methacholine (10−5 M) or racemic 14,15-EET (10−5–10−7 M) was added, and potassium channel activity was recorded.

For characterization of channel properties, current-voltage relationship and calcium-sensitivity protocols were performed using the inside-out patch configuration and perfusate and pipette solutions described above. After obtaining the inside-out configuration, channel activity (0.5–2 min) was recorded at membrane potentials of −60, −40, −20, 0, 20, 40, and 60 mV. For calcium sensitivity, channel activity was recorded at +40 mV with control perfusate solutions (10−7 M ionized Ca2+) and after increasing perfusate Ca2+ to 10−6 and 10−5 M. Channel activity is graphed as mean open-state probability (NPo) calculated as previously described (24, 32).

Drugs and chemicals.

Methacholine was purchased from Sigma and prepared as a 10−2 M stock in water. U-46619 and ODYA were purchased from Caymen Chemical. 14,15-EET, 14(S),15(R)-EET, and 14(R),15(S)-EET were synthesized as described (6). ODYA, 14,15-EET, 14(S),15(R)-EET, and 14(R),15(S)-EET were prepared as 10−2 M stocks in 95% ethanol. U-46619 was prepared as a 2 × 10−3 M stock in 95% ethanol. [U-14C]arachidonic acid was purchased from New England Nuclear. All solvents were HPLC grade and purchased from Burdick and Jackson or Sigma.


Vascular reactivity, patch-clamp, and MS data are expressed as means ± SE. Significance of differences between mean values was evaluated by Student's t-test or ANOVA followed by the Student-Newman-Keuls multiple comparison test. Significance was accepted at P < 0.05.


Smooth muscle cells were transduced for 24 h with various MOIs of the CYP BM-3 or CMV-β-Gal adenovirus, and the expression of CYP BM-3 was determined by immunoblotting (Fig. 1). CYP BM-3 protein expression was not observed in the control or CMV-β-Gal-transduced cells. However, a 124-kDa immunoreactive protein was detected in the CYP BM-3 adenovirus-transduced cells, which increased with increasing MOI. On the basis of these results, a MOI of 1,000 was selected for all subsequent experiments. The CYP BM-3 antibody was used to detect the enzyme in smooth muscle cells by immunofluorescence (Fig. 2). Fluorescent staining was not observed in control cells. However, intense immunofluorescence was detected around the nucleus of CYP BM-3 adenovirus-transduced smooth muscle cells. Nearly all cells showed fluorescent staining. CYP BM-3-transduced cells incubated with the secondary antibody but without the primary antibody did not show fluorescent staining (data not shown). These results confirm the transduction of the CYP BM-3 protein in the CYP BM-3 adenoviral-treated cells.

Fig. 1.

Western immunoblot analysis of cytochrome P-450 Bacillus megaterium-3 (F87V) (CYP BM-3) protein expression in control and cytomegalovirus promoter-Escherichia coli β-galactosidase (CMV-β-Gal)- and CYP BM-3 adenoviral-transduced smooth muscle cells. Cells were transduced with 500, 750, or 1,000 multiplicity of infection (MOI) of the adenovirus; 20 μg protein were loaded in each lane. An immunoreactive band corresponding to the CYP BM-3 protein was detected at 124 kDa.

Fig. 2.

Immunofluorescent staining for CYP BM-3 (F87V) protein expression in control (A) and CYP BM-3 (B) adenoviral-transduced smooth muscle cells. Fluorescent staining was observed in CYP BM-3-transduced cells and not observed in control cells.

CMV-β-Gal- and CYP BM-3-transduced smooth muscle cells were incubated with [14C]arachidonic acid, and the metabolites were resolved by HPLC. In CMV-β-Gal-transduced cells, metabolites comigrated with prostaglandins and 2-arachidonyl glycerol (2-AG) (Fig. 3A). In addition, radioactive metabolites comigrating with 14,15-DHET and 14,15-EET were observed in the CYP BM-3-transduced cells (Fig. 3B). These radioactive peaks were collected and analyzed by LC/ESI/MS. The first metabolite (fraction 71–77, Fig. 3B) comigrated with 14,15-DHET, and the mass spectra gave a major M-1 ion of 335 m/z (Fig. 3C). The second metabolite (fraction 124–129, Fig. 3B) comigrated with 14,15-EET, and the mass spectra gave a major molecular ion (M-1) ion of 319 m/z (Fig. 3D). These data show that smooth muscle cells transduced with CYP BM-3 adenovirus express the active epoxygenase capable of synthesizing 14,15-EET and 14,15-DHET from arachidonic acid.

Fig. 3.

Metabolism of [14C]arachidonic acid by CMV-β-Gal (A) and CYP BM-3 (B) adenoviral-transduced smooth muscle cells. Media were extracted and analyzed by reverse-phase HPLC and liquid chromatography (LC)/electrospray ionization (ESI) mass spectrometry. A and B: HPLC chromatograms. Migration times of known standards are shown above each chromatogram. 14,15-Dihydroxyeicosatrienoic acid (14,15-DHET) and 14,15-epoxyeicosatrienoic acid (14,15-EET) production was observed only from CYP BM-3-transduced smooth muscle cells (B). C and D: LC/ESI mass spectra of 14,15-DHET (C) and 14,15-EET (D) fractions from CYP BM-3-transduced cells. The 14,15-DHET fraction (C) produced a major ion of 335 mass/charge (m/z) (M-1), and the 14,15-EET fraction (D) produced a major ion of 319 m/z (molecular ion-1). 2-AG, 2-arachidonyl glycerol.

This LC/ESI/MS method was used to assay 14,15-EET and 14,15-DHET in the media of cells treated with vehicle or methacholine (10−5 M). In CMV-β-Gal-transduced cells, 14,15-EET and 14,15-DHET production was at the lower limits of detection of our method and was not changed after methacholine treatment (Fig. 4). In contrast, in CYP BM-3-transduced cells, basal 14,15-EET release was detectable and was significantly increased by methacholine (Fig. 4A). The production of 14,15-DHET was ∼30% of 14,15-EET release in these cells. Methacholine did not stimulate the release of 14,15-DHET (Fig. 4B). The release of 20-hydroxyeicosatetraenoic acid (20-HETE) was 0.52 ± 0.13 pg/ml in CMV-β-Gal-transduced cells and was not increased in CYP BM-3-transduced cells or stimulated by methacholine. Thus, in the CYP BM-3-transduced cells, methacholine stimulates the release of 14,15-EET.

Fig. 4.

Effect of methacholine (MeCH; 10−5 M) on 14,15-EET (A) and 14,15-DHET (B) release from CMV-β-Gal- and CYP BM-3 adenoviral-transduced smooth muscle cells. Media were extracted, and 14,15-EET and 14,15-DHET concentrations were determined by LC/ESI mass spectrometry. Values are means ± SE. *Significantly different from control, P < 0.05; n = 7–10.

To determine if CMV-β-Gal or CYP BM-3 transduction alters the expression of BKCa channels, we performed Western immunoblotting. The expression of a 126-kDa protein corresponding to the BKCa channel α-subunit was detected in control cells as well as cells transduced with various MOI of CMV-β-Gal and CYP BM-3 adenovirus (Fig. 5A). Additionally, we performed patch-clamp analyses to determine if CMV-β-Gal or CYP BM-3 transduction altered channel properties. The current-voltage relationships of inside-out patches in equimolar K+ (145 mM) are shown in Fig. 5B. Slope conductances were similar and averaged 269 ± 4 pS, 271 ± 6 pS, and 264 ± 4 pS in control, CMV-β-Gal-transduced, and CYP BM-3-transduced cells, respectively. Increasing bath calcium from 10−7 to 10−5 M caused similar BKCa channel activation in control and CMV-β-Gal-transduced and CYP BM-3-transduced cells (Fig. 5C). Therefore, cultured bovine coronary smooth muscle cells possess BKCa channels, and BKCa channel α-subunit expression and channel electrical properties are not altered by transduction with CMV-β-Gal or CYP BM-3.

Fig. 5.

Large-conductance calcium-activated potassium (BKCa) channel α-subunit expression, current-voltage relationship, and calcium sensitivity of the BKCa channel in control smooth muscle cells (C) or CMV-β-Gal- and CYP BM-3-transduced smooth muscle cells. A: Western immunoblot analysis of BKCa α-subunit expression. Cells were transduced with 500, 750, or 1,000 MOI; 20 μg protein were loaded into each lane. An immunoreactive band corresponding to the BKCa α-subunit was detected at 126 kDa. B: current-voltage relationship of BKCa channels from inside-out patches. Values are means ± SE; n = 3–4. C: effect of increasing Ca2+ concentration on BKCa channel activity from inside-out patches. Values are means ± SE; n = 4–5. NPo, open-state probability.

We next determined the effects of endogenous 14,15-EET on BKCa channel activity of CMV-β-Gal- and CYP BM-3-transduced cells. Using the cell-attached configuration, a large-conductance potassium channel (203 ± 11 pS) was observed in control cells, which is similar to the BKCa channels previously described in cell-attached patches of freshly isolated coronary smooth muscle cells (24, 25, 32). Potassium channel conductance was similar in the patches from CMV-β-Gal-transduced cells (208 ± 8 pS) and CYP BM-3-transduced cells (214 ± 8 pS). Cells were treated with vehicle or methacholine, and potassium channel activity was recorded (Fig. 6, A and B). In CMV-β-Gal transduced cells, basal potassium channel activity was not altered by methacholine. However, in CYP BM-3-transduced cells, methacholine caused a fourfold increase in potassium channel activity compared with vehicle or CMV-β-Gal-transduced cells. Pretreatment of CYP BM-3-transduced cells with the CYP inhibitor ODYA (10−5 M) completely blocked the methacholine-stimulated BKCa channel activity. Treatment of control cells with ODYA did not alter BKCa channel activation by 14,15-EET (Fig. 6C). These results demonstrate that methacholine stimulates BKCa channel activity in CYP BM-3-transduced smooth muscle cells through a CYP-dependent pathway.

Fig. 6.

BKCa channel activity in cell-attached patches. Effect of methacholine (10−5 M) on CMV-β-Gal- and CYP BM-3 adenoviral-transduced smooth muscle cells or CYP BM-3-transduced cells pretreated with 17-octadecynoic acid (ODYA; 10−5 M). A: original tracings of BKCa channel activity from a single CMV-β-Gal- or CYP BM-3-transduced cell before and after methacholine addition. B: averaged data; n = 17–18. C: 14,15-EET activation of BKCa channels in control cells in the presence of vehicle (control) or ODYA (10−5 M); n = 6–8. Values are means ± SE. *Significantly different from control, P < 0.05.

Racemic 14,15-EET relaxed the coronary arteries in a concentration-related manner (Fig. 7A). These relaxations were blocked by pretreatment with the BKCa channel inhibitor iberiotoxin (10−5 M). Because 14(S),15(R)-EET is the only product of CYP BM-3 (17), we tested the effect of the 14,15-EET stereoisomers on vascular tone in isolated coronary arterial rings and on potassium channel activity in coronary arterial smooth muscle cells. Both 14(S),15(R)- and 14(R),15(S)-EET relaxed the preconstricted coronary arteries in a concentration-related manner (Fig. 7B). However, 14(S),15(R)-EET was more potent and had a greater maximal activity at 10−5 M than 14(R),15(S)-EET. In cell-attached patches of the cultured smooth muscle cells, 14(S),15(R)-EET increased K+ channel opening (NPo) in a concentration-related manner (Fig. 7C). 14(R),15(S)-EET did not alter K+ channel activity. These data indicate that 14(S),15(R)-EET is the active isomer in coronary smooth muscle cells.

Fig. 7.

Effect of racemic 14,15-EET and 14,15-EET stereoisomers on vascular tone of bovine coronary arteries and BKCa channel activity in cell-attached patches of isolated smooth muscle cells. A: relaxation to racemic 14,15-EET in the presence and absence of the BKCa channel inhibitor iberiotoxin (IBTX, 10−7M). Arterial segments were preconstricted with U-46619. Changes in isometric tension were measured; n = 5–7. B: 14(S),15(R)- and 14(R),15(S)-EET vascular relaxations; n = 25–32. C: 14(S),15(R)- and 14(R),15(S)-EET activation of BKCa channels; n = 9. Values are means ± SE. *Significantly different from control, P < 0.05.


The vascular endothelium metabolizes arachidonic acid by CYP epoxygenases in a regioisomer and stereospecific manner to produce 5,6-, 8,9-, 11,12- and 14,15-EET (8, 33). Endothelium-derived EETs act in a paracrine fashion and diffuse to the vascular smooth muscle where they activate membrane BKCa channels to cause hyperpolarization and vascular relaxation (3, 32). Therefore, EETs are considered EDHFs. However, the vascular smooth muscle does not produce EETs. This current study shows that transduction of vascular smooth muscle with a CYP BM-3 adenovirus results in epoxygenase expression and 14,15-EET production. Western immunoblot analysis and immunocytochemistry showed that CYP BM-3 protein expression was present only in the smooth muscle cells transduced by the CYP BM-3 adenovirus. EET production was negligible in the vascular smooth muscle transduced with CMV-β-Gal. However, the production of both 14,15-EET and 14,15-DHET, the hydration product of 14,15-EET, was detected in the CYP BM-3-transduced cells. DHET production indicates the presence of epoxide hydrolase activity. This is consistent with our previous finding that coronary smooth muscle cells possess epoxide hydrolase activity (2). Similarly, because 14,15-EET is metabolized to 14,15-DHET, the measurement of 14,15-EET release with methacholine stimulation underestimates 14,15-EET synthesis.

Methacholine stimulated the release of 14,15-EET from CYP BM-3-transduced cells, indicating agonist-induced coupling of the transduced epoxygenase and lipase-released arachidonic acid. Additionally, methacholine increased the activity of BKCa channels in the CYP BM-3-transduced cells. The methacholine-induced increase in channel activity was blocked by the CYP inhibitor ODYA. Therefore, in the CYP BM-3-transduced cells, methacholine stimulates BKCa channel activity via a CYP-dependent pathway and 14,15-EET serves as the mediator of BKCa channel activation. In a previous study, LLCPKc14 renal proximal tubule cells were transduced with CYP BM-3, and epidermal growth factor increased 14,15-EET production and thymidine incorporation (5). Together, these results support a role for 14,15-EET as cellular second messengers of epoxygenase. 14,15-EET and 14,15-DHET were detected in the incubation media of the CYP BM-3-transduced cells, demonstrating that 14,15-EET and 14,15-DHET are released from the cell and, therefore, could activate receptors on the extracellular surface of the membrane. EET activation of coronary smooth muscle BKCa channels requires a guanine nucleotide binding protein (24, 25), and specific structural requirements of the EET molecule are necessary for relaxation (6). Thus a membrane receptor or binding site may initiate EET vascular effects.

The BKCa channel activity in the cultured bovine coronary smooth muscle cells shows similar properties and function as BKCa channels in freshly isolated coronary smooth muscle cells. Western immunoblot of nontransduced and CMV-β-Gal- and CYP BM-3-transduced cells verified the presence of a 126- kDa band that corresponds to the BKCa α-subunit. A similar immunoreactive band was found in freshly isolated bovine coronary arteries (15). Additionally, BKCa channels from the cultured and transduced cells displayed current-voltage profiles and calcium sensitivity similar to those of freshly isolated cells (16). Furthermore, BKCa channels from the cultured bovine coronary smooth muscle cells were activated by 14(S),15(R)-EET (14). This demonstrates that the adenoviral transduction and culture conditions do not alter K+ channel properties. Thus BKCa channels from the cultured cells in this study represent functional targets for EETs.

In bovine coronary arteries, the four EET regioisomers are equipotent in causing vascular relaxation (3). However, in many cases, relaxation to EET regioisomers is stereospecific. The 14(S),15(R)-EET isomer was more potent at inducing relaxations and activating smooth muscle BKCa channels than the 14(R),15(S)-EET isomer (14). Similar stereospecificity has been seen with 11,12-EET stereoisomers in rat renal and cerebral arteries. The 11(R),12(S)-EET isomer induced relaxation of renal arteries and activated cerebral artery smooth muscle cell potassium currents, whereas the 11(S),12(R)-EET isomer was inactive (23, 37). In the rat kidney, 8(S),9(R)-EET caused vasoconstriction while 8(R),9(S)-EET was without effect (21). In contrast, stereoisomers of 8,9-, 11,12-, and 14,15-EET similarly dilated small porcine and canine coronary arteries, and both 11,12-EET stereoisomers activated BKCa channels (36). Therefore, the vascular activity of EET stereoisomers may vary with species and arterial vasculature. CYP BM-3 production of the active 14(S),15(R)-EET stereoisomer provides a unique tool to study the vascular activity of 14,15-EET.

Importantly, the methacholine-induced EET production and BKCa channel activation of the CYP BM-3-transduced cells were independent of the endothelium. This suggests that epoxygenase activity and the resulting EET production are sufficient to provide agonist-stimulated hyperpolarizing activity to the smooth muscle. The loss of endothelial cell function that occurs in numerous vascular diseases decreases EDHF activity. In hypertension and insulin resistance, the EDHF response is reduced (9, 10, 20). In contrast, the impaired endothelium-dependent dilation associated with hypercholesterolemia is associated with a normal EDHF response (1, 19). Therapies designed to enhance epoxygenase expression and EET production even in the absence of functional endothelium could restore agonist-induced hyperpolarization-dependent vascular relaxation in disease states with a reduced contribution of EDHF.

In addition to vascular relaxation, EETs have numerous vascular effects including anti-inflammatory activity, inhibition of smooth muscle migration, ischemic injury protection, enhanced fibrinolysis, and decreased platelet adhesion (22, 28, 29, 34). While all four regioisomers cause vasorelaxation (3), 11,12-EET is more potent than 14,15-EET in anti-inflammatory (28) and fibrinolytic activity (29, 34). Similarly, 11,12-EET was more effective than 14,15-EET or 8,9-EET in altering platelet function (22). Thus the expression of epoxygenases such as CYP BM-3 that are regiospecific and stereospecific may selectively enhance vasorelaxation without enhancing anti-inflammatory or fibrinolytic activities. Design of additional mutated CYP epoxygenases that selectively synthesize a particular EET regioisomer may be therapeutically useful or may provide a biochemical tool to study the action of a specific EET regioisomer.


These studies were supported by grants from the National Institutes of Health (HL-51055, DK-38226, GM-37992, and GM-31278) and the Robert A. Welch Foundation.


We thank Gretchen Barg for secretarial assistance.


  • 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.


View Abstract