Rat Coronary Endothelial Cell Membrane Potential Responses During Hypertension
Abstract—The purpose of this study was to provide the first membrane potential profile in coronary endothelial cells from normotensive sham-operated control and 1-kidney, 1-clip renal hypertensive rats. Dilator responses were assessed in cannulated coronary arteries from control and 1-kidney, 1-clip rats, and the perforated patch-clamp method was used to compare membrane potential responses between the intact endothelial cells. Under these conditions, acetylcholine (100 pmol/L to 10 μmol/L) induced similar large dilations of coronary arteries from control and 1-kidney, 1-clip rats that were associated with endothelial cell hyperpolarizing responses of 16±3 and 18±2 mV, respectively. Substance P (10 fmol/L to 1 nmol/L) and bradykinin (100 fmol/L to 10 nmol/L) also substantially dilated coronary arteries from control rats but only induced small (2 to 4 mV) endothelial cell hyperpolarizing responses. These dilations, which appeared independent of membrane potential changes, were highly blunted or absent in arteries from 1-kidney, 1-clip rats. Thus, dilator responses to acetylcholine that are associated with large endothelial hyperpolarizing responses are normal in the small coronary arteries of 1-kidney, 1-clip rats. However, dilator response to substance P and bradykinin, which apparently are not heavily dependent on endothelial cell hyperpolarizations, are selectively targeted for impairment in the coronary arteries of this model of hypertension
- hypertension, renal
- coronary artery disease
- muscle, smooth, vascular
- potassium channels
Endothelium-mediated dilations are often blunted in isolated arteries from different rat models of experimental hypertension and in arteries of human subjects with essential hypertension.1 2 3 4 5 6 7 8 Although the mechanisms that underlie this dilator dysfunction are complex, a reduced release of endothelium-derived relaxing factors appears to be a contributing factor.9 In this regard, the signaling pathways by which dilator substances release relaxing factors from endothelial cells appear to involve changes in membrane potential (Em).10 11 Specifically, dilator molecules bind to endothelial membrane receptors to activate plasmalemmal Ca2+-dependent K+ channels, resulting in K+ efflux and membrane hyperpolarization. This enhanced electrical gradient for Ca2+ influx, combined with the release of Ca2+ from intracellular stores, elevates cytosolic Ca2+ in the endothelial cell to activate the Ca2+-dependent enzymes required for the synthesis of dilator factors.10 11 12 Hence, hyperpolarization of the endothelial cell membrane is linked to the release of relaxing factors, and the failure of dilator substances to trigger hyperpolarizing responses may promote endothelial dilator dysfunction during hypertension.
In this study, we directly investigated this hypothesis by characterizing vasodilator and endothelial cell Em responses to acetylcholine (ACH), substance P (SP), and bradykinin (BK) in isolated coronary arteries from sham-operated normotensive control rats and 1-kidney, 1-clip (1K1C) hypertensive rats. Notably, small coronary arteries from 1K1C hypertensive rats show normal dilator responses to ACH but blunted endothelium-dependent dilations to SP and BK. Thus, they represent a unique model in which to examine if blunted endothelium-mediated dilations are associated with reduced endothelial cell hyperpolarizing responses. By adapting the perforated-patch, current-clamp technique to measure Em in intact endothelial cells of small coronary arteries, we were able to compare for the first time recordings of Em responses to ACH, SP, and BK in endothelial cells exposed to normal and high levels of blood pressure in vivo.
Surgical Induction of Hypertension
Male Sprague-Dawley rats were obtained from Harlan Sprague Dawley, Inc. Protocols for animal use were approved by the Animal Care and Use Committee at our institution. At 8 weeks of age, the rats were fasted overnight and deeply anesthetized (100 mg/kg ketamine, 1 mg/kg acepromazine IM). For preparation of the 1K1C rat model, the right kidney was removed and the renal artery of the left kidney was clipped with a rectangular 0.3-mm silver clip near the aorta.13 Other rats had sham surgery in which the right kidney was manipulated and the left renal artery was cleaned of connective tissue but was left intact. The incisions were closed and the animals were monitored during recovery and allowed free access to food and water. At 10 to 14 days after surgery, the animals were anesthetized (20 mg/kg Ketamine IM; 50 mg/kg inactin IP). Mean arterial pressure, recorded by direct cannulation of the femoral artery, averaged 126±2 mm Hg in sham rats (n=38) and 166±3 mm Hg in 1K1C rats (n=36). Subsequently, the hearts were removed, rinsed, and placed in a dissecting dish in physiological salt solution (PSS).5
Diameter Recording in Cannulated Arteries
Second- and third-order branches of the left anterior descending or septal coronary artery were identified under a dissecting microscope and carefully dissected free. The arteries were cannulated immediately on glass micropipettes in a perfusion-superfusion chamber and pressurized at an intraluminal pressure of 80 mm Hg at 37°C. Internal diameters were monitored by videomicroscopy, and cumulative concentration-response curves to ACH (100 pmol/L to 10 μmol/L), SP (10 fmol/L to 1 nmol/L), and BK (100 fmol/L to 10 nmol/L) were performed as previously described.5 14 In a subset of experiments, the endothelium was removed by the intraluminal perfusion of an air bolus.5 14 At the end of each experiment, all arteries were perfused and superfused with Ca2+-free solution to determine the level of Ca2+-dependent active tone.5
Mounting of Arteries in Silastic Chambers
In preparation for the recording of endothelial cell Em responses, arteries were cannulated and pressurized at 80 mm Hg as described. In a subset of vessels, the endothelium was removed by an air bolus. The vessels were then cut open lengthwise with fine surgical scissors, avoiding contact with the endothelial lining. Subsequently, the arteries were cut from the glass cannulas, transferred to a 1-mL patch-clamp chamber lined with a thin layer of silastic, and carefully positioned lumen side up. The corners were secured with pins finely honed from Teflon-coated tungsten wire with an initial diameter of 76 μm. To confirm the presence of a confluent endothelial cell layer in these pinned vessels, arterial segments from sham and 1K1C rats were fixed with 1% glutaraldehyde in 0.1 mol/L cacodylate buffer and prepared for scanning electron microscopy with methods previously described.15 The micrographs in Figures 1A⇓ and 1B⇓ verify the presence of intact endothelial cells in coronary arteries from sham and 1K1C rats, respectively, that were secured for patch-clamp recordings. In contrast, Figures 1C⇓ and 1D⇓ demonstrate that the endothelium was extensively disrupted in similar arteries exposed to a luminal air bolus. These findings were consistent between micrographs obtained from 12 separate preparations. Thus, the luminal surface of pinned coronary arterial segments retained a confluent sheet of endothelial cells available to patch pipettes for the recording of endothelial Em levels.
Recording of Em in Intact Endothelial Cells
Arteries secured to the silastic base of a patch-clamp chamber were superfused at room temperature with a bath solution composed of (in mmol/L) NaCl 135, KCl 4, CaCl2 2, MgCl2 1, glucose 10, and HEPES 10 (pH 7.4). Maximally effective concentrations of ACH (1 μmol/L), SP (1 nmol/L), or BK (1 nmol/L) were infused into the bath chamber by a plastic 5-mL syringe, and outflow was established simultaneously by a second syringe. With amphotericin B used as a perforating agent, the perforated-patch, current-clamp technique was used to measure the level of Em in intact endothelial cells, as described in detail by Marchenko and Sage.16 Briefly, heat-polished patch pipettes with tip resistances between 2 and 5 MΩ were filled with pipette solution composed of (in mmol/L) KCl 145, MgCl2 1, EGTA 0.5, and HEPES 10 (pH 7). Pipette tips were dipped briefly into drug-free pipette solution, back-filled with the same solution containing 300 μmol/L amphotericin B, and then gently touched to the luminal surface of the pinned arteries. Slight suction was applied to obtain high resistance seals (≥5 GΩ). Typically, electrical access to the cell interior was obtained within several minutes after seal formation, and stable Em values were observed after 10 to 15 minutes. Membrane potentials were recorded with current-clamp methods by a List EPC-7 amplifier (List Instruments) controlled by a microcomputer equipped with a Digidata 1200 interface and pClamp software (Axon Instruments).
All drugs were purchased from Sigma Chemical Co. ACH, SP, and BK were dissolved as 1-mmol/L aqueous stock solutions in PSS. Amphotericin B was dissolved as a 65-μmol/L stock in dimethyl sulfoxide for direct dilution into the pipette solution.
Data are expressed as mean±SEM. Diameter values represent the measurement of internal diameter in microns. A replication factor of 6 to 9 was performed for isolated vessel protocols. Significant differences between rat preparations or between control and drug-induced diameter or Em responses were determined by either Student’s t test or ANOVA with repeated measures, with a subsequent Duncan’s test. Significance was accepted at P<0.05.
Resting Diameter and Ca2+-Dependent Tone
Small coronary arteries from sham and 1K1C rats showed similar resting diameters averaging 121±2 μm (n=7) and 139±16 μm (n=9), respectively. At the end of each experiment, arteries were perfused and superfused with Ca2+-free PSS to assess their level of active tone. Under these conditions, the diameter of arteries from sham and 1K1C animals increased comparably by 57% and 52% to 190±6 μm and 211±17 μm, respectively.
Dilator and Endothelial Em Responses to ACH
To confirm earlier findings showing that dilator responses to ACH are normal in small coronary arteries of hypertensive rats,5 we performed concentration-response curves to evaluate endothelial dilator responses to ACH. Figure 2A⇓ shows that ACH (100 pmol/L to 10 μmol/L) progressively and similarly dilated coronary arteries from sham and 1K1C rats by a maximum of 59±9 μm and 49±8 μm, respectively (n=7, 8). Removal of the vascular endothelium by an air bolus abolished these dilator responses (n=6 each; not shown). Arteries from 1K1C rats showed slightly blunted dilations in response to high ACH concentrations, but this effect was not statistically different from sham responses.
Subsequently, current-clamp Em responses to 1 μmol/L ACH were compared between the intact coronary endothelial cells of sham and 1K1C rats. In all endothelial cells studied, resting Em averaged −42±1 mV (n=41) and −38±1 mV (n=31) in sham and 1K1C rats, respectively, showing a small but significant depolarization in the endothelium exposed to hypertension. However, the original traces in Figures 2B⇑ and 2C⇑, respectively, show that ACH (1 μmol/L) induced large and similar endothelial cell hyperpolarizing responses of 11 mV and 12 mV in arteries from sham and 1K1C rats. In all preparations, the average amplitude of the hyperpolarizing responses to 1 μmol/L ACH was 16±3 mV and 18±2 mV in coronary endothelial cells from sham and 1K1C rats, respectively (n=13, 14). Thus, the pronounced dilator responses to ACH observed in the isolated coronary arteries from sham and 1K1C rats were associated with large endothelial hyperpolarizing responses in both vascular preparations.
Dilator and Endothelial Em Responses to SP
Although small coronary arteries from 1K1C rats dilated normally in response to ACH, they showed severely impaired dilator responses to SP. Figure 3A⇓ shows that SP (10 fmol/L to 1 nmol/L) maximally dilated the arteries from sham animals by 29±4 μm, whereas arteries from 1K1C rats only dilated by 10±3 μm. Removal of the endothelium abolished these dilations (n=6 each; data not shown). Surprisingly, Figure 3B⇓ shows that 1 nmol/L SP only slightly hyperpolarized the coronary endothelial cells from sham rats, although it substantially dilated the isolated arteries by 29±4 μm. Figure 3C⇓ demonstrates that 1 nmol/L SP also slightly hyperpolarized the coronary endothelial cells of 1K1C rats, even though the dilator response to the same SP concentration was highly blunted in these isolated arteries. Overall, SP (1 nmol/L) induced small but significant hyperpolarizing responses averaging 2±1 mV in endothelial cells from both sham and 1K1C rats (n=7, 10). Perfusion of drug-free PSS did not alter resting Em values, providing assurance that the perfusion process per se did not induce artifactual changes in Em levels (n=5, data not shown).
Dilator and Endothelial Cell Em Responses to BK
Dilator responses to BK also were highly blunted in coronary arteries of 1K1C rats. Figure 4A⇓ shows that BK (100 fmol/L to 10 nmol/L) maximally dilated arteries from sham rats by 22±5 μm, and this response was abolished by endothelium removal (n=6; not shown). In contrast, coronary arteries from 1K1C rats failed to dilate during the application of BK (n=7, 9). Subsequently, current-clamp experiments revealed only slight hyperpolarizing responses to BK, even in the normal endothelial cells from the sham arteries. The original recording in Figure 4B⇓ shows that 1 nmol/L BK induced a hyperpolarizing response of only 2 mV in a coronary endothelial cell from a sham rat, although the cannulated arteries from these animals dilated substantially to this stimulus. Figure 4C⇓ shows that 1 nmol/L BK also slightly hyperpolarized the endothelial cell of a coronary artery from a 1K1C rat, although these arteries failed to dilate to BK. Overall, coronary endothelial cells from sham and 1K1C rats hyperpolarized by 4±2 mV and 3±1 mV in response to 1 nmol/L BK, respectively (n=7, 13).
Endothelial Cell Hyperpolarizing Responses to ACH Persist in Hypertension
Reports from several laboratories indicate that ACH activates K+ channels in the endothelial cell membrane to induce hyperpolarization.10 16 17 18 19 This hyperpolarizing response has been linked to the elevation of cytosolic Ca2+ levels in the endothelial cell and appears to be obligatory for the release of dilator factors.12 18 19 For example, in bioassay experiments of rabbit aorta, Demirel et al10 have shown that perfusion of ACH through endothelium-intact donor rings results in the relaxation of recipient rings denuded of endothelium. In this preparation, pharmacological block of K+ channels in the ACH-stimulated donor segment inhibited the downstream relaxation of the recipient ring, inferring that ACH activates endothelial K+ channels to mediate the release of dilator factors. Our finding that the large hyperpolarizing response to ACH in the coronary endothelial cells of sham and 1K1C rats was correlated with a pronounced dilation of the cannulated arteries also links the activation of endothelial K+ channels to dilator factor release and provides the first direct evidence that the intricate signaling cascade that links membrane hyperpolarization to the release of vasodilator factors is not impaired in the coronary endothelial cells of 1K1C rats.
Notably, the preservation of endothelial cell hyperpolarizing responses to ACH during hypertension may represent a unique feature of the coronary circulation. The evidence for this includes our finding that small coronary arteries from 1K1C rats dilate normally to ACH and earlier studies that also demonstrated normal dilator responses to ACH in isolated coronary arteries from spontaneously hypertensive rats and salt-sensitive Dahl hypertensive rats.5 20 In contrast, blunted ACH-induced dilations, consistent with the presence of an endothelial dilator defect, have been reported in aortic vascular segments of the same animals and in small arteries from the cerebral and mesenteric beds.1 2 6 It is unknown if these dilator defects in other vascular beds are related to an inability of ACH to initiate normal endothelial cell hyperpolarizing responses.
Impaired Dilations to SP and BK Are Not Linked to Impaired Electrical Responses
Although SP and BK substantially dilated the small coronary arteries of sham rats, only weak hyperpolarizing responses (2 to 4 mV) were recorded in intact coronary endothelial cells. These data imply that membrane hyperpolarization is not the main signaling mechanism by which SP and BK mediate the release of dilator factors from the rat coronary endothelium. This concept also is supported by our finding that although small hyperpolarizing responses to SP and BK persisted in the coronary endothelial cells of 1K1C rats, the dilator responses to SP and BK were highly impaired or absent. Thus, these vasodilator peptides apparently induce the release of relaxing factors from the coronary endothelium by transduction pathways independent of Em. This alternative signaling pathway may involve receptor-associated GTP-binding regulatory proteins, and the downregulation of their expression or function in the endothelial cell membrane may be a mechanism that contributes to vasodilator defects in animals and humans with chronic hypertension.12 21 22
Depolarized Resting Em in Intact Endothelium of 1K1C Hypertensive Rats
The resting Em level in coronary endothelial cells from 1K1C rats was depolarized by 4 mV compared with cells from sham animals. This finding raised the possibility that the loss of SP and BK dilations in the coronary arteries from the 1K1C rats was caused by an inability to hyperpolarize these endothelial cells to the negative Em level required for dilator factor release. However, in several arteries from 1K1C rats, we added a “priming dose” of ACH to slightly dilate the vessel before the addition of SP and BK, and this maneuver did not restore normal dilator responses to the peptides (data not shown). This observation, combined with the normal dilator responses of the coronary arteries from 1K1C rats to ACH, suggests that the depolarized Em level in the endothelial cells of the hypertensive animals did not impair the release of dilator factors under our conditions. Thus, although the endothelial cell membrane may be susceptible to electrical remodeling during hypertension, whereby the ion channel profile that normally regulates the resting Em level is altered, the functional significance of this observation is unknown.
Hyperpolarizing Responses to SP and BK Differ Between Cultured and Intact Endothelial Cells
In cultured endothelial cells from porcine or guinea pig coronary arteries, SP and BK induce large hyperpolarizing responses between 23 and 37 mV, as measured by intracellular microelectrodes or whole-cell patch-clamp methods.23 24 25 These large Em changes to SP and BK clearly contrast with the very slight hyperpolarizing responses to these peptides that we observed in current-clamped intact endothelial cells. They also contrast with the findings of Wang et al19 in freshly isolated current-clamped endothelial cells of rat aorta studied by the perforated-patch method. In this endothelial preparation, BK (400 nmol/L) also elicited only a slight hyperpolarizing response, whereas ACH (10 μmol/L) induced pronounced hyperpolarizations averaging 51 mV in the intact endothelial cells.19 Notably, the perforated-patch method as used in this study provides several distinct advantages over the use of microelectrode or standard whole-cell patch-clamp recording for monitoring Em levels in endothelial cells.16 First, in contrast to standard whole-cell approaches, the perforated patch method permits electrical access to the cell interior for the dynamic recording of Em responses without interrupting the structural integrity and cytosolic signaling of the endothelial cell. Second, when applied to the intact endothelium, it circumvents damage incurred from enzymatic dispersion and allows endothelial cell-to-cell contacts and endothelial cell–vascular smooth muscle interactions to remain intact. Third, by measuring Em responses in intact endothelial cells, phenotypic changes observed in cultured endothelial cells are avoided. Thus, the perforated-patch method is uniquely suited to compare dynamic Em responses between normal and diseased endothelial cells that can retain their full complement of cytosolic signaling molecules and intercellular contacts.
Coronary Endothelium-Derived Relaxing Factors
Several relaxing factors are released by the coronary endothelium including nitric oxide, prostacyclin, and cytochrome P450 products derived from arachidonic acid.5 26 27 28 The regulation of the production and release of these relaxing factors by ACH, SP, and BK is complex and is initiated by the activation of membrane receptors linked to differing G-protein–signaling cascades, which ultimately results in the stimulation of enzymes specific for the production of each factor.9 In this regard, our previous study of coronary microvessels from Dahl rats has shown that pharmacological block of nitric oxide and prostacyclin synthesis had little effect on ACH-, SP-, or BK-induced relaxations, whereas the addition of octadecynoic acid, an inhibitor of cytochrome P450, markedly attenuated these relaxations.5 Thus, a cytochrome P450 product rather than nitric oxide or prostacyclin appears to mediate the dilator responses to ACH, SP, and BK in small rat coronary arteries. These findings, together with those of the present study, imply that the release of the same dilator factor can be achieved by either hyperpolarization of the endothelial cell membrane (ACH) or by pathways that operate independent of Em change to mediate dilator factor release (SP and BK).
The results of this study provide the first evidence that hyperpolarizing responses are intact in rat coronary endothelial cells exposed to increased blood pressure in vivo. Additionally, the lack of correlation between vasodilator responses and endothelial cell Em responses to SP and BK suggests that SP and BK mediate dilation of small coronary arteries by endothelial cell transduction pathways that are distinct from membrane hyperpolarization. This pathway rather than electrical signaling may be vulnerable to damage during hypertension. In this regard, it is possible that the impaired dilator responses to the native coronary peptides SP and BK may eliminate important dilator influences in the coronary microvasculature and thereby favor increased vascular resistance, whereas the persistent responsiveness of the endothelium to ACH may preserve dilator function during increased parasympathetic drive.29 The importance of these findings in the human coronary circulation is unclear at present, although SP and BK appear to be involved in the regulation of resting blood flow and flow-dependent dilation, and a reduced dilation to SP has been described in the coronary microcirculation of hypertensive humans.8 30 31 Therefore, further characterization of the signaling pathways in endothelial cells that are involved in the release of dilator factors may clarify the complex patterns of endothelial dilator defects observed in the heart during hypertension and may provide a conceptual framework for designing drug therapies to restore normal endothelium-dependent dilations to the coronary circulation in this disease.
This research was supported by National Heart, Lung, and Blood Institute grant R01-HL-59238 from the National Institutes of Health (N.J. Rusch). K. Gauthier was a predoctoral fellow of the American Heart Association–Wisconsin Affiliate. We gratefully acknowledge the surgical expertise of Anne Runnells in the preparation of the rat models and the technical support of Janet De Bruin with scanning electron microscopy. We thank Carol Gross for preparing the manuscript.
- Received May 2, 2000.
- Revision received May 12, 2000.
- Accepted July 31, 2000.
Myhan WG. Role of prostaglandin H2-thromboxane A2 in responses of cerebral arterioles during chronic hypertension. Am J Physiol. 1992;262:H539–H543.
Lüscher TF, Raij L, Vanhoutte PM. Endothelium-dependent vascular responses in normotensive and hypertensive Dahl rats. Hypertension. 1987;9:157–163.
Wirth KJ, Linz W, Wiemer G, Scholkens BA. Differences in acetylcholine- and bradykinin-induced vasorelaxation of the mesenteric vascular bed in spontaneously hypertensive rats of different ages. Arch Pharmacol. 1996;354:38–43.
Gauthier-Rein KM, Rusch NJ. Distinct endothelial impairment in coronary microvessels from hypertensive Dahl rats. Hypertension. 1998;31:328–334.
Fujii K, Tominaga M, Ohmori S, Kobayashi K, Koga T, Takata Y, Fujishima M. Decreased endothelium-dependent hyperpolarization to acetylcholine in smooth muscle of the mesenteric artery of spontaneously hypertensive rats. Circ Res. 1992;70:660–669.
Kelm M, Preik M, Hafner DJ, Strauer BE. Evidence for a multifactorial process involved in the impaired flow response to nitric oxide in hypertensive patients with endothelial dysfunction. Hypertension. 1996;27:346–353.
Egashira K, Suzuki S, Hirooka Y, Kai H, Sugimachi M, Imaizumi T, Takeshite A. Impaired endothelium-dependent vasodilation of large epicardial and resistance coronary arteries in patients with essential hypertension: different responses to acetylcholine and substance P. Hypertension. 1995;25:201–206.
Vanhoutte PM. Endothelial dysfunction in hypertension. J Hypertens. 1996;14:S83–S93.
Demirel E, Rusko J, Laskey RE, Adams DJ, van Breemen C. TEA inhibits ACh-induced EDRF release: endothelial Ca2+-dependent K+ channels contribute to vascular tone. Am J Physiol. 1994;267:H1135–H1141.
Champion HC, Kadowitz PJ. Vasodilator responses to acetylcholine, bradykinin and substance P are mediated by a TEA-sensitive mechanism. Am J Physiol. 1997;273:R414–R422.
Himmel HM, Whorton AR, Strauss HC. Intracellular calcium, currents, and stimulus-response coupling in endothelial cells. Hypertension. 1993;21:112–127.
Liu Y, Harder DR, Lombard JH. Myogenic activation of canine small renal arteries after nonchemical removal of the endothelium. Am J Physiol. 1994;267:H302–H307.
Chen G, Cheung DW. Characterization of acetylcholine-induced membrane hyperpolarization in endothelial cells. Circ Res. 1992;70:257–263.
Busse R, Fichtner H, Lückhoff A, Kohlhardt M. Hyperpolarization and increased free calcium in acetylcholine-stimulated endothelial cells. Am J Physiol. 1988;255:H965–H969.
Tschudi MR, Lüscher TF. Age and hypertension differently affect coronary contractions to endothelin-1, serotonin and angiotensins. Circulation. 1995;91:2415–2422.
Tsutsui M, Shimokawa H, Tanaka S, Kuwaoka I, Hase K, Nogami N, Nakanishi K, Okamatsu S. Endothelial Gi protein in human coronary arteries. Eur Heart J. 1994;15:1261–1266.
Pratt PF, Rosolowsky M, Campbell WB. Mediators of arachidonic acid-induced relaxation of bovine coronary artery. Hypertension. 1996;28:76–82.
Duffy SJ, Castle SF, Harper RW, Meredith IT. Contribution of vasodilator prostanoids and nitric oxide to resting flow, metabolic vasodilation, and flow-mediated dilation in human coronary circulation. Circulation. 1999;100:1951–1957.
Broten TP, Miyashiro JK, Moncada S, Feigl EO. Role of endothelium-derived relaxing factor in parasympathetic coronary vasodilation. Am J Physiol. 1992;262:H1579–H1584.
Zeitlin IJ, Parratt JR, Fagbemii O. Coronary vessels contain enzymes that liberate kinin-like vasodilator substances. Eur Heart J. 1989;10:73–77.
Groves P, Kurz S, Just H, Drexler H. Role of endogenous bradykinin in human coronary vasomotor control. Circulation. 1995;92:3424–3430.