Epoxyeicosatrienoic Acids Activate K+ Channels in Coronary Smooth Muscle Through a Guanine

Epoxyeicosatrienoic Acids Activate K+ Channels in Coronary Smooth Muscle Through a Guanine Nucleotide Binding Protein

Pin-Lan Li, , William B. Campbell

From the Department of Pharmacology and Toxicology, Medical College of Wisconsin, Milwaukee.

/ Abstract
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References

Abstract Epoxyeicosatrienoic acids (EETs) are endothelium-derivedarachidonic acid metabolites of cytochrome P450. They dilatecoronary arteries, open K+ channels, and hyperpolarize vascularsmooth muscles. However, the mechanisms of these smooth muscleactions remain unknown. This study examined the effects of EETson the large-conductance Ca2+-activated K+ channel (KCa) insmooth muscle cells of small bovine coronary arteries. In cell-attachedpatch-clamp experiments, 11,12-EET produced a 0.5- to 10-foldincrease in the activity of the KCa channels when added in concentrationsof 1, 10, and 100 nmol/L. In the inside-out excised membranepatch mode, 11,12-EET was without effect on the activity ofthe KCa channel unless GTP (0.5 mmol/L) or GTP and ATP (1 mmol/L)were added to the bath solution. In the presence of GTP andATP, the increase in the KCa channel activity with 11,12-EETin inside-out patches was comparable to that in cell-attachedpatches. This effect of 11,12-EET in inside-out patches wasblocked by the addition of GDP-ß-S (100 µmol/L).In outside-out patches, 11,12-EET also increased the KCa channelactivity when GTP and ATP were added to the pipette solution.The addition of a specific anti-GS antibody (100 nmol/L) inthe pipette solution completely blocked the activation of theKCa channels induced by 11,12-EET. An anti-Gß or anti-Gi antibodywas without effect. We conclude that 11,12-EET activates theKCa channels by a GS-mediated mechanism. This mechanism contributesto the effects of EETs as endothelium-derived hyperpolarizing factorsto hyperpolarize and relax arterial smooth muscle.

Key Words: endothelium-derived hyperpolarizing factor • patch clamp • K+ channel • eicosanoid • G protein

/ Introduction
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References

Recent studies have indicated that coronary endothelial cells synthesizeEETs, a family of cytochrome P450 epoxygenase metabolites ofarachidonic acid.12345 Likewise, the intima of coronaryblood vessels possesses cytochrome P450 monooxygenase activity.67 In vitro studies have demonstrated that EETs dilate coronary arteries4589 as well as renal, cerebral, pial, and caudal arteries.10111213 We have found that all four regioisomeric EETs relaxcoronary arteries in nanomolar concentrations in vitro.458EETs activate K+ channels and hyperpolarize vascular smoothmuscle in similar concentrations.8101114 These studies suggest thatEETs are excellent candidates to serve as endothelium-dependentvasodilators that hyperpolarize vascular smooth muscle. In thisregard, we and other investigators have reported that inhibitionof cytochrome P450 blocks the endothelium-dependent vasorelaxationto arachidonic acid, whereas induction of cytochrome P450 monooxygenaseenhances the vasodilation to arachidonic acid.3789 EETsappear to mediate a portion of the endothelium-dependent relaxationto acetylcholine and bradykinin in coronary arteries.7891516 Cytochrome P450 inhibitors blocked acetylcholine-inducedendothelium-dependent hyperpolarization and relaxation of coronaryvascular smooth muscle.81617 In addition, acetylcholine stimulatesEET release.8 These studies led us to propose that EETs serveas EDHFs in coronary arteries.8

The mechanism by which EETs dilate coronary arteries and hyperpolarizevascular smooth muscle remains unknown. Recent studies haveindicated that EETs activate a KCa channel in vascular smoothmuscle cells.81014 These results further support the hypothesisthat EETs serve as EDHFs, since the KCa channels are thoughtto mediate the effect of EDHF.1819 The purpose of the presentstudy was to examine the effect of 11,12-EET on the activityof large-conductance KCa channels in vascular smooth musclecells isolated from small bovine coronary arteries and to determinethe mechanism by which 11,12-EET activates these channels.

/ Materials and Methods
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References

Isolation of Vascular Smooth Muscle Cells From Small Coronary Arteries
Bovine hearts were obtained from a local slaughterhouse. A branchof the coronary artery was cannulated and filled with 10 to20 mL of ice-cold 3% Evan's blue in 50 mmol/L sodium phosphatecontaining 0.9% sodium chloride at pH 7.4 (PSS) and 6% albumin.Then the heart was dissected into 2x3x1-cm pieces and slicedinto 300-µm-thick tissue sections. Small coronary arteriesstained with Evans blue were identified under a dissecting stereomicroscope.These arteries were microdissected, pooled, and stored in ice-coldPSS. The dissected small coronary arteries were first incubatedfor 30 minutes at 37°C with collagenase type II (340 U/mL)(Worthington), elastase (15 U/mL) (Worthington), dithiothreitol(1 mg/mL), and soybean trypsin inhibitor (1 mg/mL) in HEPESbuffer consisting of (mmol/L) NaCl 119, KCl 4.7, CaCl2 0.05,MgCl2 1, glucose 5, and HEPES 10 (pH 7.4). The digested tissuewas then agitated with a glass pipette to free the vascularsmooth muscle cells, and the supernatant was collected. Remainingtissue was further digested with fresh enzyme solution, andthe supernatant was collected at 5-minute intervals for an additional15 minutes. The supernatants were pooled and diluted 1:10 withHEPES buffer and stored at 4°C until used.

Current Recordings
Single-channel K+ currents were recorded using the patch-clamptechnique as described by Hamill et al.20 Cell-attached, inside-out,and outside-out configurations were used to identify the KCachannels and to determine the effect of 11,12-EET on the K+ currents in vascular smooth muscle cells. Patch pipettes weremade from borosilicate glass capillaries that were pulled witha two-stage micropipette puller (PC-87, Sutter) and heat-polishedwith a microforge (MF-90, Narishige). The pipettes had tip resistancesof 8 to 10 M for single-channel recordings when filled with145 mmol/L KCl solution. Smooth muscle cells were placed ina 1-mL perfusion chamber mounted on the stage of a Nikon invertedmicroscope. After the tip of the pipette was positioned on acell, a high-resistance seal (5 to 15 G) was formed betweenthe pipette tip and the cell membrane by applying a light suction.The activity of K+ channels in the membrane spanning the pipettetip was recorded. These measurements represented the cell-attachedmode. Inside-out membrane patches were excised by lifting thepipette membrane complex to the air/solution interface. Outside-outmembrane patches were obtained by withdrawing the pipette tipfrom the cell after establishment of the whole-cell configuration,in which the membrane within the pipette was disrupted by alarge pulse of suction.

An EPC-7 patch-clamp amplifier (List Biological Laboratories,Inc) was used to record single-channel currents. The amplifieroutput signals were filtered at 1 KHz with an eight-pole Besselfilter (Frequency Devices Inc). Currents were digitized at asampling rate of 3 kHz and stored on the hard disk of a Gateway486 DS66 computer for off-line analysis. Data acquisition andanalysis were performed with pClamp software (version 5.7.1,Axon Instruments). Average channel activity (NPo) in patcheswas determined from recordings of several minutes by the following:

where N is the maximal number of channels observed in conditions producinghigh levels of Po, T is the duration of the recording, and tjis the time with j=1,2 ... N channels opening.

Solutions
For single-channel recordings in the cell-attached mode, thebath solution contained (mmol/L) KCl 145, CaCl2 1.8, MgCl2 1.1,glucose 10, and HEPES 5 (pH 7.4), and the pipette solution contained(mmol/L) KCl 145, CaCl2 1.8, MgCl2 1.1, and HEPES 5 (pH 7.4).For single-channel recordings using the inside-out excised membranepatch, the bath solution contained (mmol/L) KCl 145, MgCl2 1.1,HEPES 10, and EGTA 2, along with 300 nmol/L ionized Ca2+ (pH 7.2).To determine the sensitivity of the channels to cytosolic Ca2+,the concentration of ionized Ca2+ in the bath solution was variedfrom 10-7 to 10-6 and then to 10-5 mol/L. Ca2+ concentrationwas estimated by a computer program21 and was confirmed bymeasuring the free Ca2+ concentration in the solution usingfura 2 (Molecular Probes Co) with a dual-wavelength spectrofluorometer(Perkin-Elmer). The pipette solution contained (mmol/L) KCl145, CaCl2 1.8, MgCl2 1.1, and HEPES 10 (pH 7.4). For single-channel recordingsin the outside-out configuration, the bath solution contained(mmol/L) KCl 145, CaCl2 1.8, MgCl2 1.1, glucose 10, and HEPES10 (pH 7.4), and the pipette solution contained (mmol/L) KCl145, MgCl2 1.1, HEPES 10, and EGTA 2, along with 100 nmol/Lionized Ca2+ (pH 7.2). All patch-clamp experiments were performedat room temperature, 20°C.

Identification of the KCa Channel in Small Bovine Coronary Arteries
To establish current-voltage relations of the KCa channel, inside-outpatches were exposed to symmetrical KCl (145 mmol/L) solutions,and single-channel currents were recorded while membrane potentialwas varied from -60 to +60 mV in steps of 20 mV. K+ selectivityof the single-channel current was determined by reducing K+ concentration in the pipette solution to 5.4 mmol/L (n=5). Bychanging the concentration of ionized Ca2+ from 10-7 to 10-5 mol/L on the cytosolic side of inside-out patches, the sensitivityof this KCa channel to intracellular Ca2+ concentration wasexamined (n=8). The effect of TEA (Sigma) (n=4) and IBX (ResearchBiochemicals Inc) (n=5) on single K+ channels was examined usingoutside-out excised membrane patches. TEA was added to bathsolution at concentrations of 0.1, 0.3, and 1 mmol/L. IBX wasadded to bath solution at a concentration of 100 nmol/L.

Patch-Clamp Studies on the Effect of 11,12-EET
In cell-attached patches, symmetrical KCl (145 mmol/L) solutionswere used to null the membrane potential of the single smooth musclecell to near 0 mV. A 3-minute control recording at a membranepotential of +40 mV was obtained after a tight seal was established.Then the bath solution was rapidly changed by flushing the perfusionchamber with 10 mL of the same solution containing 11,12-EET (1,10, or 100 nmol/L, n=7), 12-HETE (10 or 100 nmol/L, n=5), or 20-HETE(10 or 100 nmol/L, n=6), and a series of 3-minute recordingswas obtained. To examine the interaction of cholera toxin and11,12-EET on the activity of the KCa channel, 100 ng/mL choleratoxin was included in the pipette solution (n=8).

The excised-patch modes were used to further determine the mechanisms forthe effect of 11,12-EET on the activity of the KCa channels.After inside-out patches were established, a 3-minute control recordingwas obtained at a membrane potential of +40 mV. Then the bathsolution was rapidly changed by flushing the perfusion chamber with5 to 10 mL of the same solution containing 1, 10, or 100 nmol/L 11,12-EET(n=6) with 0.5 mmol/L GTP and 1 mmol/L ATP, and a second successive3-minute recording was obtained.

In some experiments, the concentration of ionized Ca2+ on thecytosolic side of inside-out patches was changed from 10-7 to10-5 mol/L in the presence and absence of 11,12-EET (100 nmol/L),and the KCa channel current was recorded for 3 minutes at each Ca2+ concentration (n=5).

The excised inside-out patch mode was used to determine theeffect of GDP-ß-S (100 µmol/L) on 11,12-EET–inducedactivation of the K+ channel (n=6). GDP-ß-S (100 µmol/L)and 11,12-EET were added to the GTP/ATP bath solution. The outside-out patchmode was used to examine the effects of anti-GS (n=7), anti-Gß(n=8), and anti-Gi (n=4) antibody (New England Nuclear and SignalTransduction, Inc) and rabbit IgG (n=4). Antibodies at concentrationsof 10 or 100 nmol/L were added to the pipette solution containingGTP/ATP.22

Western Blots of GS Protein
The dissected coronary arteries were cut into very small piecesand homogenized with a glass homogenizer in ice-cold HEPES buffercontaining 25 mmol/L sodium HEPES, 1 mmol/L EDTA, and 100 µmol/L phenylmethylsulfonylfluoride. The homogenate containing 30 µg protein wasincubated with 11,12-EET at concentrations of 1 nmol/L to 10µmol/L for 30 minutes and then subjected to 12% SDS-PAGEat 200 V for 65 minutes (Bio-Rad).23 The proteins were electrophoretically transferredonto a nitrocellulose membrane. The membrane was washed and probedwith a 1:1000 dilution of a specific anti-GS antibody (New EnglandNuclear). The ECL detection kit (Amersham) was used to detectthe specific GS protein bands as described by the manufacturer.

Statistics
Data are presented as mean±SEM. The significance of the differencesin mean values between and within multiple groups was examinedusing an ANOVA for repeated measures, followed by a Duncan's multiple-rangetest. Student's t test was used to evaluate statistical significanceof differences between two paired observations. Single-channelconductances were fit by least-squares linear regression orby using the Goldman-Hodgkin-Katz constant field equation. Avalue of P<.05 was considered statistically significant.

/ Results
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References

Characterization of KCa Currents in Small Bovine Coronary Arteries
The K+ channel activity was characterized using inside-out andoutside-out excised membrane patches that were exposed to symmetricalKCl solutions (145 mmol/L) to enhance single-channel conductances.Unitary K+ currents were detected predominantly at membranepotentials from -60 to +60 mV in inside-out patches (Fig 1A).The current-voltage relationship for this channel was linearbetween -60 to +60 mV, and mean slope conductance was 256.3±5pS, with a reversal potential of 0 mV. When the K+ concentrationin the pipette was reduced to 5.4 mmol/L, the reversal potentialshifted in a manner predicted by the Nernst equation for K+.This shift in reversal potential in response to changes in K+ gradient across the membrane suggests that this channel is selectivefor K+ (Fig 1B). This large-conductance K+ current was activatedby membrane depolarization. At the resting membrane potential,the activity of this K+ channel was low, with a mean open probability(NPo) of 0.002±0.0001. When the membrane patch was depolarizedto +60 mV, NPo was increased to 0.15±0.02.

View larger version (17K):
[in this window]
[in a new window]
/ Figure 1. Characterization of the KCa channel in vascular smooth muscle cells isolated from small bovine coronary arteries. A, A representative recording of a large-conductance K+ channel recorded from an inside-out excised membrane patch at membrane potential ranging from -60 to +60 mV. c indicates closed. B, Current-voltage relations for large-conductance K+ channel current recorded from inside-out patches of smooth muscle cells using symmetrical 145 mmol/L KCl solution and after lowering KCl in the pipette to 5.4 mmol/L.

The effects of changes in the cytosolic Ca2+ concentration onthe activity of this channel were examined using inside-outpatches. When the Ca2+ concentration on the cytosolic side ofthe membrane patch was increased from 10-7 to 10-5 mol/L, theactivity of K+ channel increased markedly. At a cytosolic Ca2+ concentrationof 10-7 mol/L and membrane potential of +40 mV, the NPo of thisK+ channel was 0.02±0.0012. When cytosolic Ca2+ concentrationwas increased to 10-6 and then to 10-5 mol/L, the NPo of this K+ channel was increased to 0.04±0.003 and 1.88±0.06, respectively.

TEA and IBX, inhibitors of the KCa channel, were studied usingthe outside-out membrane patch mode. Representative tracingsdepicting the results of these experiments are presented inFig 2A. Addition of TEA to the bath produced a concentration-dependentreversible flickery-type blockade of the K+ channel. The meanunitary current amplitude of this channel fell from 9.89 pAunder control conditions to 6.8, 4.7, and 2.24 pA after 0.1,0.3, and 1 mmol/L TEA, respectively, was added to bath (Fig2B). NPo of this K+ channel was not altered by the additionof TEA. Fig 2C presents representative tracings depicting the effectof IBX on this K+ channel. IBX (100 nmol/L) decreased NPo ofthe channel by 93% when added to the bath (Fig 2D), but it hadno effect on the current amplitude of this channel. These dataare consistent with this being a large-conductance KCa channel.

View larger version (28K):
[in this window]
[in a new window]
/ Figure 2. Effect of TEA and iberiotoxin (IBTX) on the KCa channel in outside-out membrane patches of smooth muscle cells isolated from small bovine coronary arteries. TEA was added to the bath so that the external surfaces of outside-out patches were exposed to the indicated concentrations of the drug. A, A representative tracing showing the effect of TEA at various concentrations on the KCa channel current recorded at a membrane potential (Em) of +40 mV. c indicates closed. B, Effect of TEA on current amplitude of the KCa channels. C, A representative recording of the KCa channel before and after IBTX (100 nmol/L) was added to the bath. D, Effect of IBTX on NPo of the KCa channels. *P<.05 vs control (C).

Effect of 11,12-EET on the Activity of the KCa Channel in the Cell-Attached Patch Mode
Representative recordings of single-channel K+ currents thatwere recorded in the cell-attached mode before and after theaddition of 11,12-EET to the bath are presented in Fig 3A. 11,12-EETcaused a concentration-dependent increase of the activity ofthe KCa channel. 11,12-EET at concentrations of 1, 10, and 100nmol/L produced a 0.5- to 10-fold increase in NPo of this KCa channel(Fig 3B). A significant effect was seen even at the lowest concentrationof 11,12-EET studied (1 nmol/L) (P<.05). The amplitude ofthese channels was unaltered by 11,12-EET even at the highestconcentration studied (100 nmol/L) (Fig 3C). When cell membrane potentialwas changed by adjusting the pipette potential from -20 to -40and then to -60 mV, the activity of the KCa channel was significantlyincreased as described above, but the effects of 11,12-EET werenot altered. 11,12-EET at a concentration of 100 nmol/L producedan 10-fold increase in NPo of the KCa channels in spite of changesin membrane potential from 20 to 60 mV. 12-HETE, a structuralanalogue of 11,12-EET, had no effect on the activity of theKCa channel, and 20-HETE decreased the activity of this KCachannel (P<.05) (Table 1).

View larger version (33K):
[in this window]
[in a new window]
/ Figure 3. Effect of 11,12-EET on the KCa channel activity in cell-attached patches of smooth muscle cells isolated from small bovine coronary arteries. A, A representative recording of the KCa channel under control conditions and after addition of 1, 10, and 100 nmol/L 11,12-EET to the bath at a membrane potential (Em) of +40 mV. c indicates closed. B, Effect of 11,12-EET on NPo of the KCa channels in smooth muscle cells isolated from small bovine coronary arteries. C, Effect of 11,12-EET on current amplitude of the KCa channels. W indicates washout of the EET. *P<.05 vs control (C).
View this table:
[in this window]
[in a new window]
/ Table 1. Effect of 12-HETE and 20-HETE on KCa Channel Activity in Cell-Attached Patches of Coronary Smooth Muscle Cells

Effect of 11,12-EET on the Activity of the KCa Channel in the Inside-out Patch Mode
In contrast to the marked effects of 11,12-EET (100 nmol/L)in cell-attached patches, 11,12-EET had no effect on the activityof the KCa channel when applied to the internal surface of inside-outexcised membrane patches (Table 2). The number of channel openings,mean open time, and the amplitude of these KCa channels recordedfrom inside-out excised membrane patches were not significantlyaltered when even a high concentration of 11,12-EET (1 µmol/L)was applied to the internal surface of the patch. NPo was 0.03±0.009for control and 0.0267±0.01 with 11,12-EET. However,when 0.5 mmol/L GTP and 1 mmol/L ATP were included in the bathsolution, 11,12-EET produced a concentration-dependent increasein the KCa activity (Fig 4A). 11,12-EET increased the NPo ofthese channels to an extent comparable to that in the cell-attachedpatch mode and in comparable concentrations. The increase inthe channel activity was reversed by washing out the 11,12-EET(Fig 4B). In addition, addition of 0.5 mmol/L GTP alone alsoincreased the basal activity of the KCa channels by 15% andrestored the effect of 11,12-EET in these excised membrane patchesto a level comparable to that obtained in the cell-attachedpatches (Table 2). However, ATP alone had no effect on the basalactivity of the KCa channels, and 11,12-EET did not stimulatethe KCa channels in the presence of only ATP.