Role of Nitric Oxide in the Coronary Microvascular Responses to Adenosine and Increased Metabolic Demand
Background The purpose of this study was to test the hypothesis that endothelium-derived nitric oxide (NO) participates in coronary microvascular responses to adenosine and pacing-induced increases in metabolic demand by maintaining an optimal distribution of coronary resistance.
Methods and Results Coronary microvascular diameters were measured by stroboscopic epi-illumination and intravital microscopy in open-chest dogs (n=20). Epicardial coronary blood velocity (CBV) was measured by Doppler flowmetry. Responses to adenosine (1 and 10 μg · kg−1 · min−1 IC) and left atrial pacing (180 beats per minute) were recorded before and after inhibition of NO synthesis by NG-nitro-l-arginine methyl ester (L-NAME, 30 μg · kg−1 · min−1 IC). At baseline, adenosine dilated arterioles (<100 μm) (11±4% and 25±3% diameter changes, P<.05) more than small arteries (>100 μm) (−4±6% and 7±3%, P<.05 for the higher dose) and increased CBV (43±31% and 118±25%, P<.05). Left atrial pacing dilated arterioles (12±2%, P<.05) and small arteries (8±3%, P<.05) and also increased CBV (68±9%, P<.05). L-NAME abolished CBV increases caused by acetylcholine (10 and 100 ng · kg−1 · min−1 IC; 53±33% and 168±82% versus −12±15% and −1±14%, P<.05) but not papaverine. Small arteries were constricted by L-NAME (−8±2%, P<.05), arterioles were dilated (10±4%, P<.05), and CBV was unchanged. After L-NAME, adenosine failed to dilate arterioles further (3±3% and 2±2%; P<.05 versus prior responses), and CBV changes were attenuated (14±16% and 8±13%; P<.05 versus prior responses). Pacing also failed to dilate arterioles (−4±2%, P<.05 versus prior response), resulting in an attenuated CBV change (34±13%, P<.05 versus prior response). The possibility that adenosine stimulates NO release in canine coronary arterioles was investigated in isolated arterioles (diameters, 81±4 μm; n=8). Adenosine caused dose-dependent dilation to maximal diameter, which was unaffected by inhibition of NO synthesis by L-NAME.
Conclusions Inhibition of NO synthesis attenuates coronary dilation during adenosine infusions and during pacing-induced increases in metabolic demand. Inhibition of NO synthesis may shift the major site of coronary resistance into small arteries through autoregulatory adjustments in arterioles. These data therefore suggest that NO, by dilating predominantly small coronary arteries, promotes metabolic coronary dilation by preserving the tone and vasodilator reserve of arterioles.
Coronary blood flow and myocardial oxygen consumption are closely matched in the normal heart via changes in coronary microvascular resistance in response to the metabolic demands of myocardium.1 The precise mechanism of this matching is unknown, but it is likely to involve metabolic, myogenic, and endothelial (nitroxidergic) influences on the tone of coronary microvascular smooth muscle.2 Furthermore, the role of endothelium-derived nitric oxide (NO) in these vasodilatory reactions has yet to be elucidated. NO may contribute to metabolic vasodilation secondarily to an increase in shear stress in arteries due to downstream arteriolar dilation. This could occur because metabolic vasodilation of arterioles would lower total resistance of the coronary vasculature, thereby increasing flow through upstream vessels and inducing flow-dependent vasodilation of these vessels. These issues have not been unequivocally resolved, but an important role for the endothelium in a number of coronary vasomotor adjustments has been well documented. Specifically, NO is reported to participate in coronary microvascular adjustments during autoregulation,3 hypoxia,4 reactive hyperemia,5 6 adenosine administration,6 phasic coronary flow,7 and the response to an increase in myocardial oxygen consumption.8
We previously postulated that the control of coronary blood flow is due to a multifactorial, series-coupled vascular scheme in which the primary influences of different regulatory elements are exerted at different points in the coronary circulation.2 Metabolic regulation of coronary blood flow is thought to depend on the levels of vasodilator metabolites in the myocardial interstitium.9 Myogenic regulation depends on the variation of resistance with pressure.10 In addition to these mechanisms, endothelium modifies coronary resistance by dilating microvessels in response to flow and agonists.11 Metabolite- and pressure-induced changes in tone predominate in coronary microvessels <150 μm in diameter,12 whereas flow-induced changes in tone appear to be greatest in larger microvessels.2 13 This is compatible with the idea that endothelium-dependent dilation of larger microvessels occurs as a consequence of metabolically induced increases in flow. A further consequence may be that, for any given flow and net resistance, flow-dependent dilation of larger microvessels allows there to be more tone in the more metabolically sensitive smaller microvessels. An endothelium-mediated decrease in the tone and resistance of larger microvessels will increase the flow and pressure transmitted to the smaller microvessels and so increase their tone and resistance by metabolic and myogenic mechanisms. Inhibition of NO synthesis might then be expected to redistribute the resistance away from the smaller microvessels, with overall net reduction of vasodilator reserve, due specifically to reduction of dilator reserve and responsiveness of the smaller microvessels, which are most sensitive to metabolites and pressure. For example, coronary dilatation during increased myocardial oxygen consumption or during infusions of adenosine may be attenuated by inhibition of NO synthesis.
In the present in vivo and in vitro studies, we investigated the possibility that inhibition of NO synthesis and the consequent constriction of small coronary arteries (diameters >100 μm) is associated with dilation of arterioles (diameters <100 μm). Specifically, we tested the hypothesis that endothelium-derived NO participates in coronary microvascular responses to adenosine and pacing-induced increases in metabolic demand by maintaining an optimal distribution of coronary resistance. To test this, we examined the baseline effects of inhibition of NO on coronary microvascular diameters and the reactions of microvessels before and after inhibition to adenosine- and pacing-induced vasodilation.
In Vivo Measurement of Coronary Microvascular Diameters
Adult mongrel dogs of either sex (n=20) weighing between 4 and 11 kg were sedated (droperidol, 0.5 mg/kg IV), anesthetized (pentobarbital, 30 mg/kg IV), and placed on a homeothermic blanket to maintain body temperature at 37°C. The right femoral artery and vein were cannulated. A 5F fluid-filled catheter was advanced from the right carotid artery to the left ventricular cavity to enable recording of left ventricular pressure, from which the left ventricular dP/dt was derived by an on-line differentiator. This signal was used as a trigger from which illumination and respiration (jet ventilator) were synchronized to the cardiac cycle.14 15 Arterial blood gases and pH were analyzed frequently and were maintained in the following ranges by adjustment of the tracheal cannula or the administration of sodium bicarbonate: Pco2, 25 to 40 mm Hg; Po2, 100 to 200 mm Hg; and pH, 7.34 to 7.44.
The heart was exposed by a left thoracotomy in the fifth left intercostal space and was partially stabilized in a pericardial cradle. A 24-gauge cannula was inserted into the proximal circumflex coronary artery to enable coronary artery pressure to be measured and drugs and fluorochromes to be administered. A pacing wire was attached to the left atrial appendage in some studies (n=7). The heart was restrained by four 22-gauge pins passed through it and attached to an externally fixed rod. Although vertical cardiac motion was eliminated by this maneuver, vigorous myocardial contraction continued in the horizontal plane. Resting blood flow and vasodilator reserve are not affected by this use of myocardial restraint.14 In most studies, a 20-MHz pulsed Doppler ultrasound cuff-type transducer (flowmeter model 545C-3, University of Iowa; Titronics Medical Instruments) was positioned around the proximal circumflex coronary artery to obtain a Doppler frequency shift signal, which was considered proportional to changes in coronary artery blood velocity.
The diameters of epicardial coronary microvessels were measured by stroboscopic intravital fluorescent microscopy (Leitz Ploemopak, Leitz H2 excitation/barrier filter, Wild Leitz, USA, Inc) with a light-intensified CCD video camera (Cohu Inc).15 The microscope objectives used were the Leitz EF4 (×4; numerical aperture, 0.22) and the Leitz L10 (×10; numerical aperture, 0.22), with the resulting magnification on the monitor either ×160 or ×400, respectively.
Fluorescein isothiocyanate–dextran (molecular weight, 2 000 000) was injected in short pulses (50 to 100 μL, 10 mg/mL) through the circumflex coronary cannula so that the IDs of the epicardial microvessels could be measured and arterial and venous vessels could be distinguished. Five to eight images of the vessel during late diastole were obtained over a period of <30 seconds, and diameter measurements over this period typically varied by <±3% from the average value. Control images were obtained at least 15 minutes after each intervention, and vessels in which the microvascular diameters varied from the prior control diameters by more than 10% were excluded. The fluorescent images were digitized from the camera by a frame digitizer (Imaging Technology Inc) and were transferred to a Macintosh IIfx computer (Apple Computer Inc) for diameter measurements (Image 2.18, National Institutes of Health Research Services Branch). Diameters were measured by aligning cursors at the vessel edges, the measurements in pixels being converted to micrometers by use of a conversion factor determined in previously described calibration experiments using microspheres of different sizes.15
In Vitro Measurement of Coronary Arteriolar Diameters
Adult dogs of either sex (n=6) were sedated and anesthetized as described above. Heparin (1000 IU/kg IV) was administered. The animals were intubated and ventilated with room air. A left thoracotomy was performed, and the heart was electrically fibrillated, removed, and placed immediately in cold saline (4°C). The left anterior descending and circumflex coronary arteries were individually cannulated for perfusion with an India ink/gelatin mixture in physiological salt solution (PSS)16 to facilitate microdissection. Coronary arterioles <100 μm in ID were carefully dissected from the subepicardial myocardial tissue at 4°C and were transferred to a Lucite vessel chamber containing PSS-albumin solution at pH 7.4. Each end of each arteriole was cannulated with a glass micropipette with tip OD of approximately 40 μm and secured with 11-0 ophthalmic suture. The India ink/gelatin PSS solution was flushed out at low pressure (20 cm H2O), and the other end of the microvessel was secured to a second micropipette.
After the vessels were cannulated, the chamber was transferred to the stage of an inverted microscope (IM35, Carl Zeiss; objective, Zeiss ×40; numerical aperture, 0.75) with a Dage TV camera (67M Newvicon) and video micrometer. Arterioles with diameters >75 μm were pressurized to 60 cm H2O and those with diameters <75 μm to 40 cm H2O by adjustment of the height of a reservoir connected to each micropipette. These pressures approximate the estimated intraluminal pressures for microvessels of these sizes in vivo.14 With both reservoirs set to the same height, the vessels were pressurized without flow. Leaks were detected by closing off the system to the reservoirs and examining for a decline in intraluminal pressure. Vessels with leaks were excluded from further study. IDs were recorded continuously during experiments. The microvessels were set to their in situ length and were bathed in PSS-albumin solution with the temperature maintained at 36°C to 37°C by an external heat exchanger.
Beating Heart Studies
At the start of each experiment, the cyclooxygenase inhibitor indomethacin (5-mg/kg IV injection) was administered. This drug was given to avoid the potentially confounding influence of endothelial prostanoid synthesis on coronary microvascular tone, although there is overwhelming evidence that coronary flow–dependent dilation is mediated mainly by an NO-like substance in the dog.17 The β-adrenergic antagonist propranolol (1-mg/kg IV injection, with repeat 0.25-mg/kg IV injections at 2- to 3-hour intervals) was administered to limit changes in coronary microvascular diameter caused by reflex neurohumoral sympathetic activation during adenosine administration and pacing.
After stable systemic hemodynamics and coronary microvascular diameters were recorded, acetylcholine, an endothelium-dependent vasodilator in the dog, was administered (100 ng · kg−1 · min−1 intracoronary infusions for 3 to 5 minutes) to confirm that endothelium-dependent coronary microvascular dilation was intact. In some animals, the changes in coronary blood velocity caused by acetylcholine (10 and 100 ng · kg−1 · min−1 for 3 to 5 minutes) were also recorded (n=8).
After the preparation had returned to a stable baseline state, systemic hemodynamics, coronary artery blood velocity, and coronary microvascular diameters were recorded during the intracoronary administration of adenosine (n=12; 1 and 10 μg · kg−1 · min−1 infusions for 3 to 5 minutes).
After further baseline measurements, systemic hemodynamics, coronary blood velocity, and coronary microvascular diameters were obtained 3 to 5 minutes after the onset of left atrial pacing at 180 beats per minute to evaluate the effect of an increase in myocardial oxygen consumption (n=7).
After further baseline measurements, the endothelium-independent vasodilator papaverine was administered in some animals (n=6; 0.25 mg intracoronary bolus).
Inhibition of NO Synthesis
The principal purpose of these studies was to compare the coronary microvascular responses to adenosine and pacing before and after inhibition of NO synthesis. This was achieved in each animal by an intracoronary infusion of nitro-l-arginine methyl ester (L-NAME, 30 μg · kg−1 · min−1 for 15 to 45 minutes) in a dose sufficient to diminish by >50% coronary microvascular dilation by acetylcholine (100 ng · kg−1 · min−1). At least 15 minutes after the administration of L-NAME, systemic hemodynamics, coronary artery blood velocity, and coronary microvascular diameter were recorded for comparison with the previous control measurements and with the subsequent intervention.
After inhibition of NO synthesis by L-NAME, the coronary microvascular responses to repeated adenosine, left atrial pacing, and papaverine were evaluated. The preparations were allowed to return to a stable baseline state between each intervention.
Isolated Coronary Microvessel Studies
In separate studies of isolated coronary arterioles, we investigated the possibility raised by these and other studies5 that adenosine is an endothelium-dependent vasodilator in the coronary microcirculation. Dose-response curves to adenosine (10−10 to 10−4 mol/L) were constructed in isolated coronary arterioles with maximal diameters <100 μm (eight vessels from six animals) before and after inhibition of NO synthesis by L-NAME. Some studies (n=4 vessels) were performed with indomethacin (10−5 mol/L) in the bath to match the experimental conditions of the in vivo studies. Adenosine was added cumulatively to the bath in 50 μL aliquots, and the diameter was recorded during each steady state, which was normally observed after 2 minutes. After adenosine was washed out of the bath and spontaneous tone was recovered, the dilation in response to acetylcholine (10−8 and 10−7 mol/L) was recorded. NO synthesis was inhibited by incubation of the arterioles with L-NAME (10−5 mol/L) for 15 minutes. Acetylcholine was administered to confirm inhibition of endothelium-dependent dilation, and the responses to incremental doses of adenosine were recorded. Maximal arteriolar diameter was determined at the end of the experiments by addition of sodium nitroprusside (10−4 mol/L) to the bath.
Adenosine and propranolol were prepared as 1-mg/mL solutions in 0.9% saline. Acetylcholine was prepared as a 10-μg/mL solution in 0.9% saline. Indomethacin was dissolved in 95% ethanol and made up to a 5-mg/mL solution in 0.9% saline so that the final concentration of ethanol was <20%. L-NAME was prepared as a 1-mg/mL solution in 0.9% saline brought to a physiological pH (between 7.3 and 7.5) by addition of small aliquots of 1 mol/L NaOH immediately before use. All drugs were obtained from Sigma Chemical Co.
Microvascular diameter changes during the administration of drugs and during pacing are expressed as a mean percent change (±SEM) from the control diameters (thus, +% indicates dilation, and −% indicates constriction). The significance of the hemodynamic changes and the diameter changes from baseline induced by acetylcholine, adenosine, pacing, papaverine, and inhibition of NO synthase activity were assessed by factorial ANOVA. Two-group ANOVA with repeated measures followed by Fisher’s least-significant-difference multiple-range tests was used to compare the percent changes in diameter in response to drugs and pacing before and after inhibition of NO synthesis. Data are presented only for vessels in which the dilation by acetylcholine was attenuated by at least 50% after L-NAME. Although microvascular responses were generally graded with respect to vessel size, data for small arteries (>100 μm in diameter) and arterioles (<100 μm in diameter) were analyzed separately in view of the well-recognized differences in physiological behavior between vessels of these size classes. Data were also analyzed for all vessels in each protocol.
In the in vitro studies, microvascular diameters during interventions were expressed as the mean normalized to the maximal diameter (±SEM). Dose-response curves obtained during graded adenosine administration in the presence and absence of L-NAME were compared by Scheffé’s multiple contrasts with ANOVA. A probability level of 95% was used in all studies as the criterion of statistical significance.
Coronary and aortic pressures and heart rate during the studies are shown in the Table⇓. There was no significant change in these hemodynamic parameters during pacing or during the administration of adenosine, acetylcholine, or papaverine.
Coronary Blood Velocity and Microvascular Diameters in the Beating Heart
Inhibition of NO Synthesis
Inhibition of NO synthesis by intracoronary L-NAME did not change coronary blood velocity (−1±20%, P=NS) or coronary perfusion pressure (Table⇑). Small coronary arteries were constricted by L-NAME (−8±2%, P<.05), whereas arterioles were dilated (+10±4%, P<.05, Fig 1⇓).
Before inhibition of NO synthesis, acetylcholine increased coronary blood velocity only at the higher dose (10 ng · kg−1 · min−1, +53±33% change in Doppler frequency shift [mean±SEM], P=NS; 100 ng · kg−1 · min−1, +168±82%, P<.05) by dilating microvessels of all size classes.
After inhibition of NO synthesis, acetylcholine failed to increase blood velocity (10 ng · kg−1 · min−1, −12±15%; 100 ng · kg−1 · min−1, −1±14%; both P=NS versus baseline and P<.05 versus prior response in the absence of L-NAME). The coronary microvascular diameters attained during acetylcholine were diminished (125±11 μm after L-NAME versus 131±12 μm before, P<.05), supporting loss of endothelium-dependent dilation.
Before inhibition of NO synthesis, adenosine increased coronary blood velocity (1 μg · kg−1 · min−1, +43±31%, P=NS; 10 μg · kg−1 · min−1, +118±25 ng · kg−1 · min−1, P<.05). Adenosine dilated arterioles (1 μg · kg−1 · min−1, +11±4% change in diameter; 10 μg · kg−1 · min−1, +25±3%; both P<.05, Fig 2⇓) and small arteries (1 μg · kg−1 · min−1, −4±6%, P=NS; 10 μg · kg−1 · min−1, +7±3%, P<.05), although the arteriolar dilation was of greater magnitude (P<.05 for both doses). The extent of dilation of microvessels was inversely related to the initial diameter, inasmuch as smaller vessels dilated proportionately to a greater extent than those upstream.
After inhibition of NO synthesis, the increase in coronary blood velocity caused by adenosine was attenuated (1 μg · kg−1 · min−1, +14±16%; 10 μg · kg−1 · min−1, +8±13%; both P=NS versus baseline and P<.05 versus prior response). Also, dilation of arteriolar vessels was abolished (1 μg · kg−1 · min−1, +3±3%; 10 ng · kg−1 · min−1, +2±2%; both P=NS versus baseline and P<.05 versus prior response, Fig 2⇑). However, it is worth noting that some arterioles still possessed vasodilator reserve to adenosine. The dilation of small coronary arteries by the higher dose of adenosine also was attenuated (1 μg · kg−1 · min−1, +3±3%, P=NS versus baseline and prior response; 10 μg · kg−1 · min−1, +0±2%, P=NS versus baseline and P<.05 versus prior response). Despite the marked differences in the changes from baseline induced by adenosine before and after L-NAME, the absolute diameters measured during adenosine infusions were similar (124±12 versus 128±11 μm, P=NS), reflecting the importance of the change in baseline.
Before inhibition of NO synthesis, left atrial pacing increased coronary blood velocity (+68±9%, P<.05) and dilated coronary arterioles (+12±2%, P<.05, Fig 3⇓) and small arteries (+8±3%, P<.05). The magnitude of the dilation appeared to be inversely related to microvessel size, but there was no significant difference between the responses of arterioles and those of small arteries.
After inhibition of NO synthesis, the increase in coronary blood velocity during left atrial pacing was diminished (+34±13%, P<.05 versus baseline and previous response). Pacing-induced dilation was abolished in coronary arterioles (−4±2%, P=NS versus baseline and P<.05 versus prior response) and in small arteries (−7±3%, P=NS versus baseline and P<.05 versus prior response). Pacing achieved a lesser overall microvascular diameter than under baseline conditions (105±10 versus 120±13 μm, P<.05), perhaps due to loss of flow-dependent dilation in small arteries.
Before inhibition of NO synthesis, papaverine substantially increased coronary blood velocity (+491± 208%, P<.05) by dilating coronary microvessels by an average of +24±4% (P<.05, Fig 4⇓). The pattern of microvascular dilation by papaverine appeared to be essentially uniform. Yet it should be noted that nearly one half of the arterioles dilated by 30%, but none of the small arteries possessed this degree of vasodilator reserve. This suggests profound heterogeneity of segmental and regional coronary vasodilator reserve.
After inhibition of NO synthesis, coronary microvascular dilation by papaverine was unchanged (+758±243% increase in flow velocity, P<.05 versus baseline and P=NS versus prior response). All size classes of microvessels were dilated and therefore attained a greater absolute diameter after L-NAME than before (113±9 versus 105±9 μm, P<.05).
Coronary Arteriolar Responses to Adenosine In Vitro
Adenosine (10−10 to 10−4 mol/L) caused dose-dependent dilation of isolated coronary arterioles (estimated log [molar] ED50 of 6.5, Fig 5⇓). L-NAME induced slight but significant constriction under baseline conditions (control, 72±2% of maximal diameter; L-NAME, 68±2% of maximal diameter). Inhibition of endothelium-dependent relaxation by L-NAME did not alter the dose-response relation to adenosine (no change in the log [molar] ED50 or maximum response). No difference was observed between the diameter responses to L-NAME and adenosine in the presence and absence of indomethacin. After completion of the dose-response curves to adenosine, diameter recovered to baseline under control conditions and during L-NAME.
In the present studies, we have made the following new observations: (1) Inhibition of NO synthesis by intracoronary L-NAME results in constriction of small coronary arteries but dilation of arterioles in vivo; (2) arteriolar dilation after inhibition of NO synthesis accounts for diminished further coronary microvascular dilation in response to adenosine and to cardiac pacing; and (3) adenosine-mediated dilation in isolated coronary microvessels is not mediated by NO. From these results, we suggest that NO causes heterogeneously distributed dilation in the intact coronary microcirculation and that it aids in the maintenance of a normal distribution of coronary microvascular resistance, ie, the major site of resistance residing in arterioles under baseline conditions. This happens because NO tonically dilates and lessens the resistance of small coronary arteries between 100 and 300 μm in diameter. Thus, NO preserves the vasodilator potential of the arteriolar segment of the coronary microcirculation that is most sensitive to metabolite- and pressure-mediated dilation.
Critique of the Experimental Approach
Studies of coronary microvascular responses in the beating heart are limited by the inescapable need for an open-chest anesthetized animal, which limits the validity of extrapolating the findings to conscious preparations. However, the methodology is validated and characterized so that our findings are relevant to those from other studies of the canine coronary circulation. A further limitation is that we studied only the epicardial coronary microcirculation, which might differ from the microcirculation of deeper myocardial layers in terms of both the intrinsic distribution of resistance18 and the vasoactive stimuli of transmural pressure and flow.19 In the present studies, in which measurements of coronary artery blood velocity were taken to indicate the flow to the whole microvascular bed, unseen changes in microvascular tone in deeper layers may potentially lead to discrepancies between changes in epicardial microvascular diameters and in epicardial coronary blood velocity. Indeed, as shown in the “Results,” such a discrepancy was observed. Specifically, we observed that L-NAME abolished epicardial arteriolar dilation to pacing, but flow velocity increased. Because myocardial metabolic rate is higher in deeper as opposed to superficial layers of the left ventricular wall,20 vessels in these deeper layers are probably under more intense metabolic control than those in the superficial epicardium. This difference probably accounts for the fact that epicardial microvessels failed to dilate during pacing after inhibition of NO synthase, while flow increased, representing metabolic dilation of intramyocardial resistance vessels.
The in vivo coronary microvascular responses to adenosine and pacing were markedly changed by L-NAME. This substance potently inhibits endothelium-dependent dilation in response to agonists in the coronary circulation of the several animal species.3 5 7 21 22 Although flow-dependent dilation has not been demonstrated directly in the intact coronary microcirculation, it has been demonstrated in canine epicardial coronary arteries7 17 23 and is abolished by l-arginine analogues.7 17 Furthermore, l-arginine analogues abolish flow-dependent dilation in isolated porcine coronary microvessels.11 Thus, many investigators have routinely used L-NAME and other arginine analogues to examine nitroxidergic responses in the coronary circulation.
Information concerning the local mechanisms governing resistance in the beating heart is indirectly provided by studies of isolated coronary arterioles removed from the surrounding myocardium and from the microvascular network. Microvascular viability under these conditions was confirmed by the development of spontaneous tone and by endothelium-dependent dilation in response to acetylcholine. These complementary studies were performed in microvessels that were cannulated and pressurized without flow. Given this caveat, the experiments provide information about the behavior of isolated canine coronary microvessels.
Inhibition of NO Synthesis
Inhibition of NO synthesis by L-NAME led to markedly heterogeneous changes in baseline coronary microvascular diameters in the present study. Small coronary arteries constricted, while arterioles dilated. The constriction of small coronary arteries after L-NAME indicates that the basal, probably flow-dependent, activity of endothelium-derived NO normally reduces the tone of these vessels. The dilation of arterioles after L-NAME implies that mechanisms other than endothelium-dependent dilation primarily govern tone in these vessels. Coronary blood velocity and microvascular resistance were unchanged by L-NAME, suggesting that arteriolar dilation is an intrinsic autoregulatory process, perhaps a metabolic and/or myogenic mechanism. It may be mediated partially by increased adenosine release, previously demonstrated in rabbit hearts after L-NAME.5 The dilation of arterioles after inhibition of NO synthesis implies that NO production aids in maintaining the normal distribution of coronary vascular resistance in which the major component resides in arterioles.
The increase in coronary flow caused by adenosine was attenuated after inhibition of NO synthesis. This was due primarily to attenuation of further dilation by adenosine of arterioles already dilated after L-NAME. These data are consistent with the findings of Parent et al,6 who found that the hyperemic response to intracoronary adenosine is attenuated after L-NAME in dogs. The vasodilatory response to pacing was also attenuated by inhibition of NO synthesis, probably reflecting attenuated dilation of vessels sensitive to endogenous metabolites. In contrast to its effects on coronary dilation by adenosine and pacing, L-NAME did not alter the coronary dilation caused by papaverine, perhaps because of the high potency of the drug on all segments of the coronary circulation.24 One aspect of the present results that deserves mention concerns the fact that only submaximal vasodilatory stimuli (pacing and moderate doses of adenosine) were used. This prompts the question: Would vasodilatory reserve to maximal doses of adenosine or intense stimuli, such as reactive hyperemia, also be compromised? Although one might predict that the baseline redistribution of resistance could hamper adenosine-induced dilation, Smith and Canty3 reported that L-NAME did not alter maximal coronary vasodilation to adenosine. We reconcile our results with this observation by arguing that the eventual outcome of inhibition of NO synthesis would not be a diminution of maximal adenosine-induced vasodilation but rather a shift in sensitivity. This would occur because adenosine, like papaverine, can dilate coronary arterioles and small arteries >100 μm in diameter but with far less sensitivity than arterioles.25 26 Thus, extreme doses of adenosine would be necessary to produce maximal vasodilation because of the shift in resistance to upstream vessels with low sensitivity to adenosine.
Attenuation of adenosine-induced dilation after L-NAME raises the possibility, also suggested by others,6 that adenosine stimulates NO release from coronary microvascular endothelium. However, L-NAME did not affect the dilation of isolated arterioles by adenosine, suggesting that adenosine does not act by releasing NO in canine coronary microvessels. Further, it was also reported that inhibition of guanylyl cyclase does not attenuate the dilation of isolated coronary microvessels to adenosine.27 This observation supports our contention that adenosine-induced dilation is independent of NO. The apparent discrepancy between the in vitro and in vivo observations is best explained by the baseline effects of L-NAME in vivo on arteriolar tone. Dilation of this segment, which is normally most responsive to adenosine,12 26 greatly attenuates the in vivo responses.
Pathophysiological Significance of the Results
These data provide clear evidence that endothelium-derived NO activity in small coronary arteries promotes the potential for metabolic coronary dilation by maintaining basal tone and vasodilator reserve in arterioles. By shifting resistance into arterioles, endothelium-derived NO tonically maintains the coronary vasodilator reserve available to metabolic and autoregulatory stimuli operating mainly in arterioles.12 26 This would also explain why inhibition of NO synthase blunts autoregulation in conscious dogs3 ; ie, loss of arteriolar vasodilator reserve after NO inhibition would be expected to reduce the autoregulatory range. This mechanism may account for the positive correlation between the coronary dilator responses to atrial pacing and to acetylcholine in the patients being investigated for chest pain recently described by Quyyumi et al.28 An altered distribution of resistance may account for a reduced coronary flow reserve in other conditions associated with coronary endothelial dysfunction, eg, atherosclerosis29 and hypertension.30 Reduced dilator reserve of arterioles due to endothelial dysfunction is likely to lower the threshold for myocardial ischemia, particularly in the presence of coronary stenoses. Further studies of the intact coronary microcirculation should determine whether endothelial dysfunction reduces the dilator reserve of arterioles exposed to hypoxia and acidosis and to the reduced pressure and flow beyond a coronary stenosis.
This work was done during the tenure of a British-American Research Fellowship of the American Heart Association and the British Heart Foundation (Dr Jones) and was supported by the Sue Washburn Endowment for Cardiovascular Research and grants HL-32788, HL-17669, and HL-46502 from the US Public Health Service, National Institutes of Health.
- Received July 5, 1994.
- Revision received October 17, 1994.
- Accepted October 31, 1994.
- Copyright © 1995 by American Heart Association
Berne RM. Regulation of coronary blood flow. Physiol Rev. 1964;44:1-29.
Kuo L, Davis MJ, Chilian WM. Endothelial modulation of arteriolar tone. NIPS. 1992;7:5-9.
Smith TP Jr, Canty JM Jr. Modulation of coronary autoregulatory responses by nitric oxide: evidence for flow-dependent resistance adjustments in conscious dogs. Circ Res. 1993;73:232-240.
Park KH, Rubin LE, Gross SS, Levi R. Nitric oxide is a mediator of hypoxic coronary vasodilatation: relation to adenosine and cyclooxygenase-derived metabolites. Circ Res. 1992;71:992-1001.
Kostic MM, Schrader J. Role of nitric oxide in reactive hyperemia of the guinea pig heart. Circ Res. 1992;70:208-212.
Parent R, Pare R, Lavallee M. Contribution of nitric oxide to dilation of resistance coronary vessels in conscious dogs. Am J Physiol. 1992;262:H10-H16.
Canty JM Jr, Schwartz JS. Nitric oxide mediates flow-dependent epicardial coronary vasodilation to changes in pulse frequency but not mean flow in conscious dogs. Circulation. 1994;89:375-384.
Ueeda M, Silvia SK, Olsson RA. Nitric oxide modulates coronary autoregulation in the guinea pig. Circ Res. 1992;70:1296-1303.
Feigl EO. Coronary physiology. Physiol Rev. 1983;63:1-206.
Kuo L, Davis MJ, Chilian WM. Myogenic activity in isolated subepicardial and subendocardial coronary arterioles. Am J Physiol. 1988;255:H1558-H1562.
Kuo L, Chilian WM, Davis MJ. Interaction of pressure- and flow-induced responses in porcine coronary resistance vessels. Am J Physiol. 1991;258:H1706-H1715.
Kanatsuka H, Lamping KG, Eastham CL, Dellsperger KC, Marcus ML. Comparison of the effects of increased myocardial oxygen consumption and adenosine on the coronary microvascular resistance. Circ Res. 1989;65:1296-1305.
Chilian WM, Eastham CL, Marcus ML. Microvascular distribution of coronary vascular resistance in beating left ventricle. Am J Physiol. 1986;251:H779-H788.
Chilian WM. Functional distribution of α1- and α2-adrenergic receptors in the coronary microcirculation. Circulation. 1991;84: 2108-2122.
Kuo L, Davis MJ, Chilian WM. Endothelium-dependent, flow-induced dilation of isolated coronary arterioles. Am J Physiol. 1990;259:H1063-H1070.
Chu A, Chambers DE, Lin C-C, Kuehl WD, Palmer RMJ, Moncada S, Cobb FR. Effects of inhibition of nitric oxide formation on basal vasomotion and endothelium-dependent responses of the coronary arteries in awake dogs. J Clin Invest. 1991;87:1964-1968.
Chilian WM. Microvascular pressures and resistances in the left ventricular subepicardium and subendocardium. Circ Res. 1991;69:561-570.
Kuo L, Davis MJ, Chilian WM. Myogenic activity in isolated subepicardial and subendocardial coronary arterioles. Am J Physiol. 1988;255:H1558-H1562.
Weiss HR, Sinha AK. Regional oxygen saturation of small arteries and veins in the canine myocardium. Circ Res. 1978;42:119-126.
Smith REA, Palmer RMJ, Bucknall CA, Moncada S. Role of nitric oxide synthesis in the regulation of coronary vascular tone in the isolated perfused rabbit heart. Cardiovasc Res. 1992;26:508-512.
Jones CJH, DeFily DV, Patterson J, Chilian WM. Endothelium-dependent relaxation competes with α1- and α2-adrenergic constriction in the canine epicardial coronary microcirculation. Circulation. 1993;87:1264-1274.
Antaraccio MJ, ed. Cardiovascular Pharmacology. New York, NY: Raven Press; 1984.
Chilian WM, Layne SM, Klausner EC, Eastham CL, Marcus ML. Redistribution of coronary microvascular resistance produced by dipyridamole. Am J Physiol. 1989;256:H383-H390.
Chilian WM, Layne SM. Coronary microvascular responses to reductions in perfusion pressure: evidence for persistent arteriolar vasomotor tone during coronary hypoperfusion. Circ Res. 1990;66:1227-1238.
Quillen JE, Harrison DG. Vasomotor properties of porcine endocardial and epicardial microvessels. Am J Physiol. 1992;262: H1143-H1148.
Quyyumi AA, Cannon RO, Panza JA, Diodati JG, Epstein SE. Endothelial dysfunction in patients with chest pain and normal coronary arteries. Circulation. 1992;86:1864-1871.
Kuo L, Davis MJ, Cannon MS, Chilian WM. Pathophysiological consequences of atherosclerosis extend into the coronary microcirculation: restoration of endothelium-dependent responses by l-arginine. Circ Res. 1992;70:465-476.
Treasure CB, Klein JL, Vita JA, Manoukian SV, Renwick GH, Selwyn AP, Ganz P, Alexander RW. Hypertension and left ventricular hypertrophy are associated with impaired endothelium-mediated relaxation in human coronary resistance vessels. Circulation. 1993;87:86-93.