β2-Adrenergic Dilation of Resistance Coronary Vessels Involves KATP Channels and Nitric Oxide in Conscious Dogs
Background We considered that β2-adrenergic stimulation may dilate resistance coronary vessels by opening ATP-sensitive potassium (KATP) channels, thereby triggering NO formation.
Methods and Results In conscious instrumented dogs after β1-adrenergic blockade, intracoronary (IC) injections of acetylcholine (ACh), nitroglycerin (NTG), and pirbuterol (PIR), a selective β2-adrenergic agonist, were performed before and after blockade of NO formation with IC Nω-nitro-l-arginine methyl ester (L-NAME, 50 μg·kg−1·min−1×12 minutes) or blockade of KATP channels with IC glibenclamide (25 μg·kg−1·min−1×12 minutes followed by 2 μg·kg−1·min−1). PIR (50.0 ng/kg) increased coronary blood flow (CBF) by 32±6 from 43±7 mL/min and by only 11±2 (P<.01) from 40±7 mL/min after L-NAME. Increases in CBF to ACh were also reduced by L-NAME, but NTG responses were not. Before glibenclamide, PIR increased CBF by 33±5 from 45±7 mL/min and by only 14±3 (P<.01) from 36±5 mL/min thereafter. CBF responses to ACh and NTG were maintained after glibenclamide. Lemakalim, a selective opener of KATP channels, caused dose-dependent increases in CBF that were partially inhibited by L-NAME. In experiments in which CBF was controlled, the fall in distal coronary pressure caused by PIR was less after L-NAME or glibenclamide than before.
Conclusions β2-Adrenergic dilation of resistance coronary vessels involves both the opening of KATP channels and NO formation. L-NAME antagonized lemakalim responses consistent with a link between the opening of KATP channels and NO formation in canine resistance coronary vessels.
There is growing evidence that vasodilators acting through cAMP-dependent mechanisms involve hyperpolarization of vascular smooth muscle triggered by the opening of KATP channels. In this connection, blockade of KATP channels has been reported to blunt CBF increases to β1-adrenergic stimulation in canine resistance vessels,1 to block β2-adrenergic dilation of isolated canine saphenous veins,2 and to antagonize β1- and β2-adrenergic dilation of the rat isolated superior mesenteric arterial bed.3 KATP channels also intervene in coronary vasodilator responses to other cAMP-dependent agents, such as prostacyclin4 and adenosine.5 6 Endothelial cell hyperpolarization may also contribute to vascular relaxation. Because endothelial cells lack voltage-gated Ca2+ channels, the membrane potential is primarily involved in controlling Ca2+ movements across the cellular membrane by adjusting the electrochemical gradient driving Ca2+ entry.7 8 9 10 By increasing intracellular Ca2+ levels, hyperpolarization could therefore influence the activity of the constitutive NO synthase,11 12 a calcium/calmodulin–dependent enzyme.13 Several studies have reported the existence of KATP channels on endothelial cells.7 8 9 10 In fact, openers of the KATP channels, such as cromakalim and pinacidil, hyperpolarize endothelial cells and elevate intracellular calcium levels, which promotes NO formation in vitro.7 8 This may explain why in the pulmonary circulation in vivo, blockade of NO formation reduced pinacidil-induced vasodilation.14 The possibility that hyperpolarization of endothelial cells secondary to the opening of KATP channels triggers endothelial NO formation in resistance coronary vessels has not been studied in vivo.
The fact that cAMP-dependent vasodilation involves KATP channels led us to consider that opening of these channels may intervene in the pathway triggering NO formation during β2-adrenergic stimulation of resistance coronary vessels.15 Therefore, the effects of selective blockade of NO formation or KATP channels on CBF responses to β2-adrenergic stimulation were studied in conscious dogs. The consequences of blockade of NO formation for coronary responses elicited by opening of KATP channels with lemakalim were also investigated.
Under general anesthesia with sodium pentobarbital (30 mg/kg IV) and under sterile conditions, seven mongrel dogs (weight, 32±1 kg) underwent a left thoracotomy at the fifth intercostal space under artificial ventilation. A Tygon catheter (Norton Plastics and Synthetic Division) was implanted in the thoracic aorta. This catheter was connected to an external transducer (model 800, Bentley Trantec) to measure arterial pressure. A solid-state pressure transducer (model P6.5, Konigsberg Instruments) was inserted into the LV cavity to record LVP and to obtain LV dP/dt. A catheter was also implanted in the LV cavity to cross-calibrate the miniature pressure gauge. A cardiotachometer (model 9857, SensorMedics) triggered by the LV pressure pulse was used to monitor HR. An ultrasonic Doppler blood flow transducer was placed around the left circumflex coronary artery 20 to 30 mm from the bifurcation of the left main coronary artery. CBF was monitored with a 10-MHz pulsed Doppler flowmeter.16 At necropsy, the internal circumference of the vessel under the probe was measured to obtain the vessel cross-sectional area and to calculate a calibration factor (in mL·min−1·kHz−1). A Silastic (Dow Corning Co) IC catheter implanted proximal to the flow probe by the approach described by Gwirtz17 was used for IC drug delivery. The portion of the catheter within the coronary vessel had an external diameter of 0.6 mm. Analgesia was provided after surgery with buprenorphine (0.3 mg IM; Temgesic, Reckitt and Colman Pharmaceuticals). Prophylactic procaine penicillin G (300 000 U IM) and benzathine penicillin G (300 000 U IM) were administered for 10 days after the surgery. Enteric coated aspirin (650 mg/d PO) was given on a daily basis.
Hemodynamic variables were recorded on a VHS tape with a PCM recording adaptor (model 4000A, A.R. Vetter Co) and monitored on a direct ink-writing strip-chart recorder (model 2800s, Gould).
β2-Adrenergic Dilation and Blockade of NO Formation
Experiments were performed 2 to 6 weeks after surgery in conscious healthy dogs lying quietly on their right sides in a dimly illuminated laboratory. While HR, LVP, LV dP/dt, phasic arterial pressure and MAP, and phasic and mean CBF were continuously monitored in animals pretreated with 1.0 mg/kg atenolol IV (Sigma Chemical Co) to eliminate β1-adrenergic influences, IC bolus injections of pirbuterol hydrochloride (Pfizer Inc), a selective β2-adrenergic agonist18 ; of acetylcholine (Sigma); and of nitroglycerin (Parke-Davis) were administered. Drugs injected IC were freshly dissolved in 0.2 mL warm saline (38.9°C), slowly injected, and then flushed with an additional 1.0 mL saline delivered with a pump-driven syringe over a 15-second period. The sequence of drug administration was randomly selected. At least 3 to 5 minutes was allowed between injections for the return to a steady-state hemodynamic baseline. The same procedure was repeated 10 minutes after the completion of IC administration of 50 μg·kg−1·min−1 for 12 minutes of L-NAME (Sigma) delivered in 0.5 mL/min saline.
β2-Adrenergic Dilation and Blockade of KATP Channels
At least 48 hours later, and after atenolol (1.0 mg/kg IV), the effects of IC injections of pirbuterol, acetylcholine, and nitroglycerin were studied before and after blockade of KATP channels with glibenclamide (Sigma) delivered as an IC infusion of 25 μg·kg−1·min−1×12 minutes followed by 2 μg·kg−1·min−1 for the duration of the experiment. The adequacy of KATP channel blockade with glibenclamide was confirmed by blunted CBF responses to IC adenosine (Sigma) and to lemakalim (SmithKline Beecham Pharma), a selective KATP channel opener.19 20 Lemakalim was prepared from a stock solution containing 1% DMSO in saline, and the final concentration of DMSO in the injectate was 0.03%. The vehicle alone had no significant effect on CBF. Glibenclamide was freshly dissolved in saline containing 2.0 mmol/L sodium hydroxide and was injected at a rate of 1.0 mL/min IC during 12 minutes. Administration of the vehicle containing sodium hydroxide used to dissolve glibenclamide did not interfere with pirbuterol-induced CBF responses. Blood glucose levels were maintained within the normal range during glibenclamide infusion by intravenous 10% dextrose infusion.
Opening of KATP Channels and Blockade of NO Formation
On a separate day, the effects of IC lemakalim, acetylcholine, and nitroglycerin were studied before and after IC L-NAME.
Controlled CBF and Blockade of NO Formation or KATP Channels
Coronary responses were also studied while CBF increases were prevented, and DCP was used as an index of coronary vasomotion. In six additional dogs, a Doppler flow probe was placed close to the origin of the left circumflex coronary artery, and a hydraulic constrictor placed next to this probe was used for preventing drug-induced increases in CBF. An IC catheter was implanted distal to the hydraulic constrictor to inject the drugs and to measure DCP with an external transducer (model 800, Bentley Trantec). The tip of this catheter was located before the first marginal coronary artery. After atenolol (1.0 mg/kg IV), baseline hemodynamics and DCP were recorded, and bolus doses of drugs (0.2 mL) were injected into the IC catheter and flushed with 1.0 mL saline over a period of 8 seconds. DCP recordings were obtained immediately thereafter. When CBF was controlled, the hydraulic constrictor was inflated at the time CBF started to increase, and care was taken to limit CBF increases to <10% above baseline levels and to avoid even slight decreases below baseline levels. Lemakalim and pirbuterol were injected during normal and controlled CBF conditions before and after IC L-NAME. On a separate day, the effects of pirbuterol before and after glibenclamide were studied with and without controlled CBF. Dextrose was administered as previously described.
Hemodynamic data were read directly from the strip charts during baseline conditions and at peak increases of CBF after boluses injections of the various drugs. With a computer-assisted digitizing tablet, the area under mean CBF recordings in excess of baseline CBF, called volume response, was planimetered from the rise in CBF to the return of CBF to baseline and reported in milliliters. This approach was used to analyze responses to acetylcholine, pirbuterol, nitroglycerin, and adenosine. Because of the long-lasting effects of lemakalim on CBF, volume responses were measured from the rise in CBF to the point of 50% decline from peak CBF. The duration of CBF responses was established by direct measurement on the strip charts of the time corresponding to the measured volume responses. In experiments in which CBF was controlled, mean DCP was reported before and at peak decreases after drug injection. Data are reported as mean±SEM throughout. Overall simultaneous comparisons of baseline levels (before the various drugs) before and after L-NAME or before and after glibenclamide were made with two-way ANOVA for repeated measurements.21 22 A similar approach was used for comparing responses to a specific drug before and after L-NAME or glibenclamide. For protocols involving the injection of a single dose of acetylcholine, nitroglycerin, or lemakalim, paired t tests were used for comparing responses before and after L-NAME or glibenclamide. Comparisons of LVP, LV dP/dt, MAP, and HR at peak CBF with baselines were made with paired t tests.
Statistical significance was reached when P<.05 in all cases. All experimental procedures were approved by an ethical committee on animal care and performed in accordance with the Guide to the Care and Use of Experimental Animals (Canadian Council on Animal Care publication No. [ISBN] 0-919087-18-3, Ottawa, 1993).
β2-Adrenergic Dilation and Blockade of NO Formation
Overall simultaneous comparisons made between baseline values (values before all drugs) revealed slight increases in LVP and MAP and decreases in HR and LV dP/dt after L-NAME. Baseline CBF was not altered after L-NAME, as reported in Table 1⇓. Except for increases in CBF, other hemodynamic effects of pirbuterol, acetylcholine, and nitroglycerin were trivial and did not differ statistically before and after L-NAME.
Pirbuterol. A recording of responses elicited by pirbuterol (100.0 ng/kg) before and after L-NAME is displayed in Fig 1⇓. L-NAME resulted in a smaller CBF response to pirbuterol. Before L-NAME, pirbuterol 50.0 ng/kg increased CBF by 32±6 from 43±7 mL/min. Pirbuterol resulted in smaller (P<.01) CBF responses (11±2 from 40±7 mL/min) after L-NAME. Overall, peak increases in CBF and the volume and duration of CBF responses to pirbuterol were smaller (P<.01) after L-NAME, as reported in Fig 2⇓.
Acetylcholine. Acetylcholine 3.0 ng/kg increased CBF by 49±6 from 42±7 mL/min before L-NAME. After blockade of NO formation, acetylcholine-induced increases in CBF were limited (P<.01) to 28±4 from 44±8 mL/min. Overall, peak increases in CBF and the volume and duration of CBF responses to acetylcholine were smaller (P<.01) after L-NAME.
Nitroglycerin. Nitroglycerin 50.0 ng/kg increased CBF by 18±2 from 42±7 mL/min before L-NAME. Similar responses (22±4 from 46±8 mL/min) were elicited by nitroglycerin after L-NAME. The volume and the duration of CBF responses to nitroglycerin did not differ statistically before and after L-NAME.
β2-Adrenergic Dilation and Blockade of KATP Channels
Overall, simultaneous comparisons made between baseline values (values before all drugs) revealed slight increases in MAP, decreases in LV dP/dt, and no changes in LVP and HR after glibenclamide. Baseline CBF fell (P<.05) after glibenclamide, as reported in Table 2⇓. Except for increases in CBF, other hemodynamic effects of pirbuterol, acetylcholine, adenosine, lemakalim, and nitroglycerin were trivial and did not differ statistically before and after glibenclamide. Lemakalim 100.0 ng/kg increased CBF by 45±3 mL/min before glibenclamide and by only 6±1 mL/min (P<.01) thereafter. Thus, adequate blockade of KATP channels was achieved.
Pirbuterol. A recording of responses elicited by pirbuterol 100.0 ng/kg before and after glibenclamide is displayed in Fig 3⇓. Glibenclamide resulted in a smaller CBF response to pirbuterol.
Before glibenclamide, pirbuterol 50.0 ng/kg increased CBF by 33±5 from 45±7 mL/min. After blockade of KATP channels, pirbuterol resulted in smaller (P<.01) increases in CBF (14±3 from 36±5 mL/min). Overall, peak increases in CBF and the volume and duration of CBF responses to pirbuterol were smaller (P<.01) after glibenclamide, as reported in Fig 4⇓.
Acetylcholine. Acetylcholine 3.0 ng/kg increased CBF by 49±3 from 45±7 mL/min before glibenclamide. After blockade of KATP channels, acetylcholine-induced increases in CBF did not differ (46±6 from 37±6 mL/min). Overall, peak increases in CBF and the volume and duration of CBF responses to acetylcholine did not differ statistically before and after glibenclamide, as reported in Fig 5⇓.
Nitroglycerin. Nitroglycerin 50.0 ng/kg increased CBF by 15±1 from 42±7 mL/min before glibenclamide. Similar responses (15±2 from 39±6 mL/min) were elicited by nitroglycerin after glibenclamide. The volume and duration of CBF responses to nitroglycerin did not differ before and after glibenclamide.
Opening of KATP Channels and Blockade of NO Formation
The effects of lemakalim 50.0 ng/kg before and after L-NAME are illustrated in Fig 6⇓. Lemakalim caused a dose-dependent increase in CBF before L-NAME. Blockade of NO formation resulted in reductions (P<.01) of CBF responses (Fig 7⇓). The percent inhibition of CBF responses created by L-NAME was inversely related to the dose of lemakalim. Both the volume and the duration of CBF responses to lemakalim were smaller (P<.01) after L-NAME.
CBF responses to acetylcholine 3.0 ng/kg were reduced (P<.01) after L-NAME, but responses to nitroglycerin 50.0 ng/kg were maintained after blockade of NO formation, as reported above.
Controlled CBF and Blockade of NO Formation or KATP Channels
The effects of L-NAME on CBF and DCP (when CBF increases were prevented) with pirbuterol and lemakalim are reported in Fig 8⇓. Consistent with responses elicited with normal CBF, decreases in DCP when CBF was controlled were smaller after L-NAME. With arterial constriction, CBF increases caused by pirbuterol and lemakalim were <5% over baseline and did not differ before and after L-NAME.
The effects of glibenclamide on coronary responses elicited by pirbuterol with normal and controlled CBF are reported in Fig 9⇓. After glibenclamide, pirbuterol caused smaller decreases in DCP when CBF was controlled. The residual increases in CBF during arterial constriction did not differ before and after glibenclamide.
The present study demonstrates that β2-adrenergic dilation of resistance coronary vessels involves both the opening of KATP channels and NO formation. Glibenclamide, a blocker of KATP channels, selectively antagonized β2-adrenergic dilation without altering acetylcholine- and nitroglycerin-induced dilation. NO formation was also important in β2-adrenergic dilation of resistance coronary vessels, as indicated by blunted CBF increases after L-NAME, a powerful inhibitor of NO formation. Lemakalim, a selective opener of KATP channels, also triggered NO formation, as demonstrated by blunted CBF responses after L-NAME. Thus, opening of KATP channels could be the primary mechanism involved in β2-adrenergic coronary dilation. The increase in NO formation may be a secondary phenomenon triggered by the hyperpolarization of endothelial cells. Presumably, this led to an augmented electrochemical gradient for Ca2+ entry into endothelial cells, thereby increasing NO formation.7 8 Whether these observations could be extended to other vasodilators acting through a cAMP-dependent mechanism, such as prostacyclin and adenosine, remains to be established. Nevertheless, the present observations could be helpful in understanding the action of blockers of KATP channels given as hypoglycemic agents in the clinical setting and of openers of these channels, such as nicorandil, used as vasodilators in humans.
In the present study, CBF responses were used as an index of changes of vascular tone in resistance vessels. Given that only 20% of the pressure loss across the coronary circulation occurs in vessels >200 μm23 and that large epicardial coronary vessels contribute little to overall coronary resistance (2% to 5%),24 CBF responses reported in the present study should primarily reflect changes of vascular tone in vessels <200 μm, where the highest resistance resides. Dilation of vessels with a higher caliber would be expected to have a minor influence on CBF responses. In the present study, the chronic treatment with aspirin, targeting the blockade of cyclooxygenase and the reduction of platelet aggregation for maintaining the patency of the IC catheters may have influenced our data by changing the level of NO synthase activity. This possibility should be kept in mind when the present data are considered.
The involvement of NO formation in β-adrenergic dilation remains a debated issue. In isolated canine epicardial coronary vessels, β-adrenergic relaxation has been reported to be endothelium-independent25 26 as well as partially endothelium-dependent.27 28 In contrast, in conscious dogs, L-NAME antagonized CBF responses triggered by β2-adrenergic stimulation, consistent with the involvement of an endothelium-dependent process.15 A receptor-operated process rather than a flow-dependent phenomenon was involved, because blockade of β-adrenergic dilation by L-NAME could be demonstrated even when CBF increases were prevented.15 The apparent discrepancy between in vivo and in vitro data may be related to segmental differences in the vessels studied, ie, conductance and resistance coronary vessels.
Our data are in general agreement with several recent reports in which KATP channels intervened in β-adrenergic effects.1 2 3 6 We can only speculate on the reason why our data differed from those reported by Narishige et al,1 in which increases in CBF caused by β1-adrenergic stimulation were blocked by glibenclamide but β2-adrenergic responses were not. Aside from the fact that Narishige et al1 studied anesthetized dogs, our study differed in the use of a higher dose of glibenclamide to block KATP channels. Perhaps a more complete blockade of KATP channels was achieved in our study. In our hands, glibenclamide reduced CBF responses to lemakalim 100.0 ng/kg by >85%, whereas in the study by Narishige et al,1 glibenclamide effects were limited to a 45% attenuation of responses of pinacidil 30 μg/min.
Conceivably, the effects of glibenclamide on β2-adrenergic dilation could be secondary to the blockade of β2-adrenergic receptors by the KATP channel blocker. Our data do not directly allow us to rule out this possibility. However, in an earlier study, Randall and McCulloch3 showed that glibenclamide 1 nmol/L to 100 μmol/L did not displace the specific [3H]dihydroalprenolol binding from rat β-adrenergic receptors. Consistent with this observation, glibenclamide did not prevent increases in contractility caused by isoproterenol in anesthetized dogs,1 as would be expected if β-adrenergic blockade were involved. In the present study, the highest rate of IC glibenclamide delivery created an estimated IC concentration in the range of 10−5 mol/L, which should not have interfered with β-adrenergic receptor activation.3
Lemakalim, the active enantiomer of cromakalim, acts as a hyperpolarizing agent and creates vasodilation by opening KATP channels.19 20 Our data are consistent with a selective action of lemakalim on KATP channels, because glibenclamide, a blocker of KATP channels,29 was a powerful antagonist of lemakalim-induced dilation of resistance coronary vessels. Furthermore, the unaltered CBF responses to nitroglycerin and acetylcholine after glibenclamide suggest a maintained coronary reactivity after blockade of KATP channels. This also rules out the possibilities that a decrease in baseline CBF caused by glibenclamide led to smaller CBF responses or that blockade of KATP channels interfered with processes related to NO formation or its action on vascular smooth muscles.
The contribution of NO formation to lemakalim-induced increases in CBF along with the blockade of β2-adrenergic dilation by either glibenclamide or L-NAME supports the possibility that opening of KATP channels triggers NO formation during β2-adrenergic stimulation. Consistent with this hypothesis, KATP channels have been demonstrated to intervene in the regulation of endothelial cell membrane potentials. Lückhoff and Busse7 8 have further established that the Ca2+ influx into endothelial cells and the formation of endothelium-derived relaxing factor triggered by openers of KATP channels are secondary to cellular hyperpolarization. Openers of KATP channels prolong K+ currents elicited by bradykinin and augment Ca2+ entry into those cells.7 8 Endothelial K+ channels activated by shear stress can also cause membrane hyperpolarization, which promotes Ca2+ influx and augments NO formation.30 31 KATP channels have been reported to account for the release of NO caused by adenosine.32 Our data suggest that a similar mechanism involving KATP channels as the triggering mechanism of NO formation intervenes in the cascade of reaction leading to β2-adrenergic dilation of resistance coronary vessels.
In addition to an action on the endothelium, a direct effect of pirbuterol on vascular smooth muscle could also be involved in glibenclamide-sensitive β2-adrenergic vasodilation. In this connection, Nakashima and Vanhoutte2 reported that smooth muscle relaxation of the canine saphenous vein induced by β2-adrenergic agonists involved KATP channels. Although our data with L-NAME suggest that NO is a major intermediate in β2-adrenergic relaxation, the residual vasodilation after blockade of NO formation may reflect the involvement of a different mechanism, perhaps a direct smooth muscle relaxation caused by β2-adrenergic stimulation. An incomplete blockade of NO formation in resistance coronary vessels may also explain the residual acetylcholine and β2-adrenergic CBF responses after L-NAME, in which case the contribution of a direct smooth muscle effect of β2-adrenergic stimulation could only be overestimated by the present data.
The relative contribution of NO formation to overall lemakalim responses was inversely related to the dose of the agonist used. For example, NO formation accounted for ≈70% of peak increases in CBF caused by 20.0 ng/kg lemakalim but only for 30% of the response to 100.0 ng/kg lemakalim. This may explain why opposite conclusions regarding the role of NO formation were reached when a single dose of an opener of KATP channels was studied in vivo.14 33
The possibility that L-NAME or glibenclamide interfered with pirbuterol and lemakalim effects on CBF by blocking a flow-dependent component was studied directly in separate experiments in which DCP was used as an index of vascular tone of resistance vessels when CBF increases were prevented. Consistent with our initial observations, in which L-NAME reduced CBF responses to pirbuterol and lemakalim when CBF was normal, L-NAME reduced the fall in DCP when CBF was controlled. Glibenclamide also antagonized the CBF responses to pirbuterol in normal conditions and the fall in DCP when CBF was controlled. Thus, L-NAME and glibenclamide did not interfere with pirbuterol and lemakalim responses solely by blocking a flow-dependent component, consistent with our contention that opening of KATP channels may be the mechanism triggering NO formation.
In conclusion, β2-adrenergic dilation of resistance coronary vessels involves both the opening of KATP channels and NO formation. A causal relationship between the opening of KATP channels and NO formation is suggested by the finding of smaller CBF responses to an opener of KATP channels after L-NAME.
Selected Abbreviations and Acronyms
|DCP||=||distal coronary pressure|
|KATP||=||ATP-sensitive potassium (channels)|
|L-NAME||=||Nω-nitro-l-arginine methyl ester|
|LVP||=||LV systolic pressure|
|MAP||=||mean arterial pressure|
This work was supported through grants from the Medical Research Council of Canada, Canadian Heart and Stroke Foundation, Fonds de la Recherche en Santé du Québec, and Fonds de la Recherche de l’Institut de Cardiologie de Montréal. The authors thank Claude Mousseau, Jean-Pierre Turcotte, and Jhésabelle Voyer for skillful assistance in conducting these studies. Pirbuterol was generously supplied by Pfizer Inc, Groton, Conn, and lemakalim by SmithKline Beecham Pharma, Oakville, Ontario, Canada.
- Received August 5, 1996.
- Revision received October 12, 1996.
- Accepted November 4, 1996.
- Copyright © 1997 by American Heart Association
Narishige T, Egashira K, Akatsuka Y, Imamura Y, Takahashi T, Kasuya H, Takeshita A. Glibenclamide prevents coronary vasodilation induced by β1-adrenoceptor stimulation in dogs. Am J Physiol. 1994;266:H84-H92.
Nakashima M, Vanhoutte PM. Isoproterenol causes hyperpolarization through opening of ATP-sensitive potassium channels in vascular smooth muscle of the canine saphenous vein. J Pharmacol Exp Ther. 1995;272:379-384.
Jackson WF, König A, Dambacher T, Busse R. Prostacyclin-induced vasodilation in rabbit heart is mediated by ATP-sensitive potassium channels. Am J Physiol. 1993;264:H238-H243.
Akatsuka Y, Egashira K, Katsuda Y, Narishige T, Ueno H, Shimokawa H, Takeshita A. ATP sensitive potassium channels are involved in adenosine A2 receptor mediated coronary vasodilation in the dog. Cardiovasc Res. 1994;28:906-911.
Jackson WF. Arteriolar tone is determined by activity of ATP-sensitive potassium channels. Am J Physiol. 1993;265:H1797-H1803.
Graier WF, Kukovetz WR, Groschner K. Cyclic AMP enhances agonist-induced Ca2+ entry into endothelial cells by activation of potassium channels and membrane hyperpolarization. Biochem J. 1993;291:263-267.
Himmel HM, Whorton AR, Strauss HC. Intracellular calcium, currents, and stimulus-response coupling in endothelial cells. Hypertension. 1993;21:112-127.
Förstermann U, Pollock JS, Schmidt HHHW, Heller M, Murad F. Calmodulin-dependent endothelium-derived relaxing factor/nitric oxide synthase activity is present in the particulate and cytosolic fractions of bovine aortic endothelial cells. Proc Natl Acad Sci U S A. 1991;88:1788-1792.
Chang J-K, Moore P, Fineman JF, Soifer SJ, Heymann MA. K+ channel pulmonary vasodilation in fetal lambs: role of endothelium-derived nitric oxide. J Appl Physiol. 1992;73:188-194.
Parent R, Al-Obaidi M, Lavallée M. Nitric oxide formation contributes to β-adrenergic dilation of resistance coronary vessels in conscious dogs. Circ Res. 1993;73:241-251.
Hartley CJ, Cole JS. An ultrasonic pulsed Doppler system for measuring blood flow in small vessels. J Appl Physiol. 1974;37:626-629.
Gwirtz PA. Construction and evaluation of a coronary catheter for chronic implantation in dogs. J Appl Physiol. 1986;60:720-726.
Moore PF, Constantine JW, Barth WE. Pirbuterol, a selective β2-adrenergic bronchodilator. J Pharmacol Exp Ther. 1978;207:410-418.
Clapp LH, Davey R, Gurney AM. ATP-sensitive K+ channels mediate vasodilation produced by lemakalim in rabbit pulmonary artery. Am J Physiol. 1993;264:H1907-H1915.
Glantz SA. Primer of Biostatistics. 2nd ed. New York, NY: McGraw-Hill Book Co; 1987:245-277.
Winer BJ. Statistical Principles in Experimental Design. 2nd ed. New York, NY: McGraw-Hill Book Co; 1971:514-603.
Chilian WM, Eastham CL, Marcus ML. Microvascular distribution of coronary vascular resistance in the beating left ventricle. Am J Physiol. 1986;251:H779-H788.
Kelley KO, Feigl EO. Segmental α-receptor mediated vasoconstriction in the canine coronary circulation. Circ Res. 1978;43:908-917.
Nelson MT, Quayle JM. Physiological roles and properties of potassium channels in arterial smooth muscle. Am J Physiol. 1995;268:C799-C822.
Cooke JP, Rossitch E Jr, Andon NA, Loscalzo J, Dzau VJ. Flow activates an endothelial potassium channel to release an endogenous nitrovasodilator. J Clin Invest. 1991;88:1663-1671.
Kuo L, Chancellor JD. Adenosine potentiates flow-induced dilation of coronary arterioles by activating KATP channels in endothelium. Am J Physiol. 1995;269:H541-H549.
Minkes RK, Santiago JA, McMahon TJ, Kadowitz PJ. Role of K+ATP channels and EDRF in reactive hyperemia in the hindquarters vascular bed of cats. Am J Physiol. 1995;269:H1704-H1712.