Coronary Kinin Generation Mediates Nitric Oxide Release After Angiotensin Receptor Stimulation
Abstract Our goal was to determine whether angiotensin II (Ang II) and its metabolic fragments release nitric oxide and the mechanisms by which this occurs in blood vessels from the canine heart. We incubated 20 mg of microvessels or large coronary arteries in phosphate-buffered saline for 20 minutes and measured nitrite release. Nitrite release increased from 27±2 up to 103±5, 145±17, 84±4, 107±16, and 54±4 pmol/mg (P<.05) in response to 10−5 mol/L of Ang I, II, III, IV, and Ang-(1-7), respectively. The effects of all angiotensins were blocked by Nω-nitro-l-arginine methyl ester (100 μmol/L), indicating that nitrite was a product of nitric oxide metabolism, and by Hoe 140 (10 μmol/L), a specific bradykinin B2 receptor antagonist, indicating a potential role for local kinin formation. The protease inhibitors aprotinin (10 μmol/L) and soybean trypsin inhibitor, which block local kinin formation, inhibited nitrite release by all of the angiotensins. Angiotensin nonselective (saralasin), type 1–specific (losartan), and type 2–specific (PD 123319) receptor antagonists abolished the nitrite released in response to all the fragments. Angiotensin type 1 and type 2 and receptors mediate nitrite release after Ang I, II, III, and Ang-(1-7), whereas only type 2 receptors mediate nitrite release after Ang IV. Similar results were obtained in large coronary arteries. In summary, formation of nitrite from coronary microvessels and large arteries in the normal dog heart in response to angiotensin peptides is due to the activation of local kinin production in the coronary vessel wall.
Angiotensin II (Ang II) is the primary mediator of the renin-angiotensin system and has an important functional role in cardiovascular homeostasis.1 2 Ang I is formed from angiotensinogen by circulating or local renin. Ang I, which is inactive, can be converted by angiotensin-converting enzyme in endothelial cells to Ang II. Whereas Ang II is a potent vasoconstrictor in a variety of vascular beds, angiotensin fragments that arise through metabolism were thought to be inactive. However, a recent study has shown that Ang II may be rapidly degraded to Ang-(1-7) in the brain of rats and dogs and that this fragment may be biologically active.3 Also, Ang IV may interact with endothelial cells to cause endothelial cell–dependent vasodilation,4 although a definitive function has not been assigned to this system. Therefore, in the present study our first aim was to determine the effect of Ang I, Ang II, and the angiotensin fragments (Ang III, Ang IV, and Ang-[1-7]) on nitric oxide (NO) release from the coronary vascular bed of the dog.
Prior studies have indicated the existence of a local kallikrein-kinin system in canine coronary arteries,5 in human cultured endothelial cells,6 and in rat tail arteries and veins.7 In preliminary studies, we found that kinins contribute to the release of nitrite from large coronary arteries and microvessels caused by agonists such as norepinephrine and A23187.8 In addition, recent studies have suggested that a kininlike activity can be released from isolated perfused hearts.9 Therefore, our second goal was to determine the role of local kinins in nitrite release from dog coronary blood vessels after administration of various angiotensins.
Recently, angiotensin receptors have been subcategorized into type 1 (AT1) and type 2 (AT2) based on function and the use of specific antagonists.10 Our final goal in this study was to determine whether stimulation of one of these angiotensin receptor subtypes selectively was responsible for nitrite release in dog coronary blood vessels or whether different fragments activate the same receptor to release nitrite.
Dogs were killed and the hearts removed for in vitro studies. These protocols were approved by the Institutional Animal Care and Use Committee of New York Medical College and conformed to the guiding principles of the American Physiological Society and the National Institutes of Health.
Isolation of Coronary Microvessels and Large Coronary Arteries
The heart was obtained immediately from pentobarbital-anesthetized dogs and kept in ice-cold phosphate-buffered saline (PBS) containing 0.1% bovine serum albumin at pH 7.4. All subsequent steps were performed in ice-cold PBS unless indicated. Large coronary arteries were freed of fat and adhering connective tissue and saved in ice-cold PBS. These procedures have been published in preliminary form.8 11
For isolation of coronary microvessels, the left ventricular free wall was freed from fat, connective tissue, epicardium, endocardium, and large coronary vessels. Coronary microvessels were obtained from the remaining myocardium by a series of steps involving sequential dissection, mincing, homogenization, sieving, and glass bead purification with the use of the method of Gerritsen and Printz.12
Briefly, the myocardium was cut with scissors into small pieces, chopped with a tissue chopper (McIlwain), and suspended in ice-cold PBS. The resulting suspension was homogenized for 50 seconds at the maximum speed. The homogenates were poured over a 100-μm nylon mesh sieve (Tetko), and the reddish-brown material that adhered to the nylon mesh was collected. Collected tissue was washed thoroughly and diluted with buffer. With the use of a clean sieve each time, the sieving steps were repeated several times. Phase-contrast microscopy showed that the sieved material was a mixture of myocytes, connective tissue, collagen fibers, and small branching vessels. The final suspension resulting from this step was gently homogenized with a glass homogenizer (Kontes, Dounce, 40-mL capacity). Care was taken not to damage the cells and was used to free the myocytes from the microvessels. This step was repeated two or three times. The resulting homogenates were poured through a column of glass beads (2- to 3-mm diameter, Corning) and washed carefully with PBS. The beads and adherent microvessels were poured into a beaker that contained ice-cold PBS and were stirred. Suspended microvessels were sieved through 80-μm nylon mesh and collected. A portion of each preparation was stained and observed under the microscope to ensure the purity of the preparation.
Studies With Microvessels
Microvessels were placed in a small package of 80-μm nylon mesh, transferred into a tissue bath containing PBS, and oxygenated with 95% O2 and 5% CO2 for 30 minutes. Tissue (20 mg wet weight) was placed in 5-mL plastic tubes that contained 500 μL of PBS as control or 50 μL of each drug, which was added to buffer and resulted in a total volume of 500 μL.
Preparation and Study of Large Coronary Arteries
Circumflex and left anterior descending coronary arteries were isolated and freed of fat and connective tissue. During isolation and dissection, care was taken to prevent tissue damage, and the entire procedure was carried out in ice-cold PBS. Arteries were cut into small rings, collected in a small package as for microvessels, and transferred to the tissue bath for 30 minutes of oxygenation. The same procedure was carried out for measurement of nitrite from large coronary arteries as was described for microvessels.
Ang II (Ang-[1-8])
Ang II (10−8 to 10−5 mol/L) was incubated with 20 mg of coronary microvessels or large arteries for 20 minutes, and nitrite formation was measured. Nω-Nitro-l-arginine methyl ester (L-NAME, 100 μmol/L), a competitive inhibitor of NO synthase; Hoe 140 (10 μmol/L), a bradykinin B2 receptor antagonist; the protease inhibitor aprotinin (500 kallikrein inhibiting units [KIU]/mL); or soybean trypsin inhibitor (100 μmol/L) was added to the tubes before incubation of the tissue with 10−5 mol/L Ang II. Nitrite formation was measured. In another group of vessels before addition of the highest dose of Ang II, 10 μmol/L of DuP 753 (losartan), an AT1 receptor antagonist; saralasin, an AT1 and AT2 antagonist; or PD 123319, an AT2 antagonist, was added to the tubes, and nitrite release was measured. Similar doses of angiotensins and antagonists were studied for other angiotensin fragments; each protocol will be discussed briefly below.
Ang I (Ang-[1-10])
Ang I (10−8 to 10−5 mol/L) was incubated with the PBS containing coronary microvessels or large arteries, and nitrite release was measured. L-NAME, Hoe 140, aprotinin, soybean trypsin inhibitor, DuP 753, saralasin or PD 123319 was incubated with the tissue before addition of 10−5 mol/L Ang I and measurement of nitrite. For determination of the potential conversion of Ang I to Ang II, ramiprilat (10 μmol/L) was incubated with microvessels alone and in the presence of 10−5 mol/L Ang I, and nitrite was measured.
Ang III (Ang-[2-8])
Ang III (10−8 to 10−5 mol/L) was incubated with microvessels and large arteries, and nitrite was measured. L-NAME, Hoe 140, aprotinin, soybean trypsin inhibitor, DuP 753, saralasin, or PD 123319 was added to tissue before the highest dose of Ang III, and nitrite was measured.
Ang IV (Ang-[3-8])
Ang IV (10−8 to 10−5 mol/L) was added to microvessels and large arteries, and nitrite was measured. To block nitrite formation, L-NAME, Hoe 140, aprotinin, soybean trypsin inhibitor, DuP 753, saralasin, or PD 123319 was used.
Ang-(1-7) (10−8 to 10−5 mol/L) was added to microvessels and large arteries, and nitrite formation was measured. L-NAME, Hoe 140, DuP 753, or saralasin was added to the tissue before addition of the highest dose of Ang-(1-7), and nitrite formation was measured.
Drugs and Chemicals
All angiotensin fragments (Ang I, III, IV, and Ang-[1-7]) were purchased from Peninsula Laboratories, Inc. Hoe 140 was generously supplied by Hoechst-Roussell Inc and PD 123319 by Parke-Davis Inc. Losartan (DuP 753) was generously supplied by EI Du Pont de Nemours & Co. Ang II and all other drugs and chemicals were purchased from Sigma Chemical Co and were prepared fresh each day and used immediately.
Statistical Analysis and Calculation of Nitrite Release
Nitrite was measured with the use of the Griess reaction. For each experiment two sets of standard curves were prepared using tubes containing 500 μL of 0, 1, 2.5, 5, 7.5, and 10 μmol/L NaNO2. N-(1-Naphthyl)ethylenediamine (50 μL of 0.2%) and sulfanilamide (450 μL of 0.1%) were added to each tube containing standard or sample and vortexed for a few seconds. The tubes were kept at room temperature for 5 to 10 minutes for development of full pink color. Absorbance was measured at 540 nm with a spectrophotometer (model 930, Uvikon) that was calibrated to zero with the blank solution. Nitrite absorbance was computed with the use of regression analysis (y=a+bx) and converted to a straight line. Only curves with a correlation coefficient >.95 were used. Nitrite absorbance produced by microvessels or large coronary arteries was calculated with the least-squares equation derived from the standard curve. Data are expressed as mean±SEM. All of the graphs were produced with sigma plot. Differences from control were determined with Student’s t test. A value of P<.05 was considered statistically significant.
Coronary microvessels were collected from normal dogs (n=16) with a mean body weight of 20.5±1.0 kg, heart weight of 190±11 g, left ventricular weight of 75±5 g, and microvessel weight of 1.65±0.20 g.
Fig 1⇓ shows changes in nitrite production from control in coronary microvessels in response to increasing doses of Ang I, II, III, IV, and Ang-(1-7). Nitrite formation was increased significantly (P<.05) by 103±5, 145±17, 84±4, 107±16, and 70±3 pmol/mg after incubation of tissue for 20 minutes (37°C) with the highest dose of Ang I, II, III, IV, and Ang-(1-7), respectively. A small basal release of nitrite occurred during the 20-minute incubation with no agonist present, amounting to 25 to 40 pmol/mg per 20 minutes, essentially the equivalent of 10−8 mol/L of angiotensins as shown in Fig 1⇓. Some of this release may be caused by nitrite present in the water used to prepare the buffer, because the antagonists such as L-NAME and Hoe 140 never lowered nitrite completely to zero (Fig 2⇓).
Nitrite production was increased by 145±17 pmol/mg per 20 minutes in response to 10−5 mol/L Ang II (Fig 2⇑, top). Nitrite release was reduced significantly (P<.05) from 145±17 to 43±2, 39±9, 55±6, and 39±4 pmol/mg per 20 minutes by L-NAME, Hoe 140, aprotinin, and soybean trypsin inhibitor, respectively. Nitrite production was reduced from 145±17 to 23±2, 25±2, and 39±4 pmol/mg by DuP 753, saralasin, and PD 123319, respectively (Fig 2⇑, bottom). We did not determine whether there were additive effects of combining PD 123319 and DuP 753 because saralasin had effects that were similar to those of these two antagonists. It is curious that all of the receptor antagonists had the same effect regardless of whether the antagonist was AT1 selective, AT2 selective, or nonselective. We have no explanation for this observation unless the antagonists are not selective or there is some interaction between the receptors.
Ang I (Ang-[1-10])
Fig 3⇓ (top) shows nitrite release in response to 10−5 mol/L Ang I. Nitrite production with 10−5 mol/L Ang I was reduced from 103±5 pmol/mg per 20 minutes (P<.05) to 31±4 with L-NAME, 40±4 with Hoe 140, 40±3 with soybean trypsin inhibitor, and 16±2 with aprotinin. Fig 3⇓ (bottom) also shows changes in nitrite release by the highest dose of Ang I alone or in the presence of DuP 753, saralasin, or PD 123319. These antagonists reduced nitrite release from 103±5 to 25±2, 33±3, and 21±2 pmol/mg per 20 minutes (P<.05), respectively.
Ramiprilat by itself increased nitrite production from 62±9 to 100±17 pmol/mg per 20 minutes (P<.05), which was partially blocked by Hoe 140 (to 80±10 pmol/mg per 20 minutes). Ang I increased nitrite to 164±27 pmol/mg per 20 minutes and in the presence of ramiprilat to only 123±14 pmol/mg per 20 minutes (P<.05 compared with Ang I alone). Thus, a significant portion of the increase in nitrite with Ang I depends on Ang I conversion to Ang II.
Ang III (Ang-[2-8])
Fig 4⇓, top, shows nitrite formation from coronary microvessels in response to 10−5 mol/L Ang III. Ang III increased nitrite release by 84±4 pmol/mg per 20 minutes, and this release was reduced to 28±3, 9±1, 30±5, and 23±4 by L-NAME, Hoe 140, aprotinin, and soybean trypsin inhibitor, respectively. DuP 753, saralasin, and PD 123319 also reduced nitrite release significantly (P<.05) from 74±6 to 20±2, 40±3, and 44±4 pmol/mg per 20 minutes, respectively.
Ang IV (Ang-[3-8])
The effects of 10−5 mol/L Ang IV on the nitrite production from coronary microvessels are shown in Fig 5⇓ (top). Ang IV increased nitrite release by 107±16 pmol/mg per 20 minutes. L-NAME, Hoe 140, aprotinin, and soybean trypsin inhibitor reduced nitrite release to 25±4, 18±3, 32±8, and 15±3 pmol/mg, respectively, but DuP 753 had no effect. Saralasin and PD 123319 (Fig 5⇓, bottom) reduced nitrite release to 30±3 and 22±4 pmol/mg (P<.05), respectively, and DuP 753 again had no effect.
Fig 6⇓ shows that nitrite release increased by 54±4 pmol/mg in response to 10−5 mol/L Ang-(1-7). Release was reduced significantly (P<.05) to 25±2, 24±2, 19±1, and 23±3 pmol/mg by L-NAME, Hoe 140, DuP 753, and saralasin, respectively.
Effects of Angiotensin Fragments on Large Coronary Artery Nitrite Production
The Table⇓ summarizes the effects of Ang II and fragments on nitrite release in large coronary arteries. Nitrite production from large coronary arteries was dose related and reduced when L-NAME, Hoe 140, aprotinin, soybean trypsin inhibitor, DuP 753, or PD 123319 was incubated with the tissue before addition of 10−5 mol/L of angiotensins. All of the results were qualitatively similar to those obtained with microvessels.
The results of the present study indicate that Ang I, Ang II, and their fragments (Ang III, Ang IV, and Ang-[1-7]) increased nitrite release in isolated coronary microvessels and large coronary arteries from the normal dog heart. Nitrite production was dose related and attenuated by L-NAME, a competitive inhibitor of NO synthase. Surprisingly, Hoe 140, a specific bradykinin B2 receptor antagonist, and the protease inhibitors soybean trypsin inhibitor and aprotinin also inhibited angiotensin-stimulated nitrite release, suggesting that the NO release depended on the formation of local kinins and stimulation of bradykinin B2 receptors.
Nitrite formation in response to Ang I, II, III, and Ang-(1-7) was reduced significantly by DuP 753 (losartan), an AT1 receptor antagonist; saralasin, an AT1 and AT2 receptor antagonist; and PD 123319, a selective AT2 receptor antagonist. Ang IV–stimulated nitrite release was reduced by both saralasin and PD 123319 but not by DuP 753. Therefore, the results of the present study lead us to speculate that there is link between angiotensin receptors—possibly on the membrane of endothelial cells in the coronary vascular bed—the production of kinins through activation of a local kallikrein-kinin system in coronary blood vessels, and NO synthesis and release.
According to the classic view,1 2 the renin-angiotensin system has a variety of physiological actions in the cardiovascular system. Circulating renin acts on angiotensinogen in blood or locally to form Ang I, a biologically inactive decapeptide. Ang I is converted to the active octapeptide Ang II, which is among the most potent of vasoconstrictors, by angiotensin-converting enzyme located on the luminal surface of vascular endothelial cells. Ang II can be converted to Ang III by the action of an aminopeptidase. Ang II and III are converted enzymatically to a series of fragments that were thought to be biologically inactive. However, recent studies have challenged this notion. Swanson et al4 have reported the existence of a specific binding site for Ang IV distinct from either Ang II or III receptors. Ang-(1-7) is another biologically active peptide that can be formed from either Ang II or Ang I.3
Although Ang II is a vasoconstrictor, recent studies suggest that Ang II and various angiotensin fragments may also stimulate the production of endothelium-derived relaxing factors. Hasegawa et al13 have found in domestic fowl that Ang II reduced blood pressure in vivo and caused endothelium-dependent relaxation of aortic smooth muscle in vitro that resembled the effects of acetylcholine. L-NAME decreased the vasorelaxant effect of both Ang II and acetylcholine. Stallone et al14 suggested that Ang II has specific receptors on the membrane of vascular endothelial cells that mediate a vasodepressor response in vivo and endothelium-dependent relaxation of aortic rings in vitro. Haberl et al15 have shown that Ang III and IV cause endothelium-dependent dilation of rabbit cerebral arterioles independent of prostaglandin synthesis. Ang IV, which lacks the amino-terminal l-arginine residue, had no effect on cerebral arterioles when applied topically, but when it was used after application of l-arginine, it induced dilation. The vasodilator effect of angiotensin fragments was eliminated by methylene blue, a known inhibitor of guanylate cyclase. Osei et al16 have reported a selective regional vascular response to Ang-(1-7). This angiotensin fragment produced dose- and region-dependent vasodilation and vasoconstriction in the mesenteric and hindquarter vascular beds of the cat. The vasoconstriction was eliminated by DuP 753, an AT1 receptor antagonist. The vasodilation was partially blocked by L-NAME but not by the cyclooxygenase inhibitor meclofenamate. Therefore, this study also suggested that the vasodilator response in cat mesentery and hindquarters was mediated through NO.
Receptor binding assays and molecular cloning have shown that Ang II has multiple receptors. The vasoconstrictor effect of Ang II in blood vessels is mediated through the AT1 receptor and can be inhibited by DuP 753.17 18 Our study indicates that nitrite formation from isolated coronary microvessels and large arteries in response to Ang I, Ang II, and their fragments Ang III and Ang-(1-7) was attenuated by DuP 753. These results lead us to speculate that activation of the AT1 receptor by angiotensins is also linked to the local production of kinins and stimulation of NO synthase. Hence, NO release may buffer the vasoconstriction or even cause vasodilation of the blood vessels in response to angiotensins. Inhibition of nitrite release by L-NAME, a competitive NO synthase inhibitor, Hoe 140, and the protease inhibitors soybean trypsin inhibitor and aprotinin, which block bradykinin B2 receptors and the enzymic formation of bradykinin locally, also supports our conclusion that angiotensins indirectly release NO.
Several lines of evidence suggest that AT1 receptors are divided into two different subtypes, AT1a and AT1b.17 These receptor subtypes are structurally if not functionally different. Since DuP 753 blocks both receptor subtypes, it was difficult to determine the receptor subtype that stimulated endothelium-derived relaxing factor release in the present study. It seems that the AT1b receptor subtype has a higher affinity for PD 123319 than for DuP 753. The AT2 receptor has been cloned recently.19 20 Wiemer et al21 have found in bovine aortic endothelial cells and isolated ischemic rat hearts that Ang II was able to increase NO synthesis and release. Their data indicated the existence of Ang II receptors on the membrane of vascular endothelial cells. NO release was blocked with the use of L-NAME and Hoe 140, and this blockage was concomitant with complete inhibition of the increase in endothelial cell cGMP. Their data are consistent with our own. Taken together, the results of these experiments indicate that release of NO/nitrite in response to Ang II is most likely mediated through the local production of kinins.
Swanson et al4 first discovered Ang IV (Ang-[3-8]) in bovine adrenal cortical cells. This fragment has a high affinity for membrane binding sites in many species and tissues. It binds specifically, reversibly, and saturably and has a low affinity for Ang II and III receptors. Ang IV may cause endothelium-dependent vasodilation. Our data indicate that Ang IV can cause NO release from coronary microvessels and large arteries. Nitrite release was blocked by L-NAME, Hoe 140, saralasin, PD 123319, soybean trypsin inhibitor, and aprotinin but not by DuP 753. Blockade by Hoe 140 and the protease inhibitors indicates that nitrite release in response to Ang IV is mediated through the local generation of kinins. Blockade by saralasin and PD 123319 but not DuP 753 indicates that kinin formation is initiated through the AT2 receptor. Other recent studies by Jarvis et al22 have suggested that Ang IV can compete for a low-affinity binding site with Ang II in the presence of DuP 753 in bovine adrenal cortical cells. This binding was not blocked by AT2 receptor antagonists, and the authors concluded that this may indicate the presence of a unique angiotensin receptor, AT4. In our blood vessel preparations, the effects of Ang IV were selectively blocked by PD 123319, indicating a potential role for an AT2 receptor. The difference in these two studies may be due to differences in tissue type, the concentration of the agonists and antagonists, or some fundamental difference in the expression of angiotensin receptor subtypes in tissues.
Ferrario et al3 have reported the existence of Ang-(1-7), which is produced directly from Ang I by an enzymatic reaction distinct from angiotensin-converting enzyme, in the canine and rat brain. This fragment is derived from the amino terminal of Ang I or II and is biologically active. This fragment is as potent as Ang II as a stimulus for the secretion of hypophyseal hormones, ie, vasopressin, but lacks the direct pressor and dypsogenic effects of Ang II. Our data also revealed that Ang-(1-7) can cause nitrite release from coronary microvessels and large coronary arteries that was blocked by Hoe 140, the protease inhibitors, and both AT1 and AT2 receptor antagonists. Thus, nitrite release also depends on the generation of local kinins but not on any specific angiotensin receptor subtype.
In our studies we attempted to determine whether the release of nitrite by Ang I was due to conversion to Ang II or was a direct effect of this peptide. To answer this question, we incubated Ang I with a converting enzyme inhibitor, ramiprilat, before measuring nitrite. Ramiprilat by itself, and in hindsight we suspect most other angiotensin-converting enzyme inhibitors, released nitrite, which was blocked by both L-NAME and Hoe 140. Ang I increased nitrite production significantly from 62 to 164 pmol/mg per 20 minutes or by 165%. Ramiprilat increased nitrite from 62 to 100 pmol/mg per minute, and in the presence of both ramiprilat and Ang I, nitrite increased to 123 pmol/mg per minute, ie, by only 23%. Thus it appears that a substantial portion of the nitrite release in response to Ang I is blocked by ramiprilat; there is some concern, however, in interpreting this finding because ramiprilat by itself had such dramatic effects on baseline nitrite production. We are more certain that appreciable amounts, certainly not predominant amounts, of Ang IV were not formed in vitro because this was the only peptide whose ability to stimulate nitrite was selectively blocked by an AT2 receptor antagonist and not affected by DuP 753.
Since the release of nitrite by all of the angiotensins was inhibited by both soybean trypsin inhibitor and aprotinin, our data suggest that angiotensins activate a plasma kallikrein-like enzyme. Because our tissues are extensively washed during preparation, it is unlikely that there is enough plasma in our preparation for the kallikrein to be just a contaminant. On the other hand, plasma kallikrein may specifically bind to the plasma membrane of the endothelial cell, or it could be incorporated into vesicles by pinocytosis. This is an area of active investigation currently in our laboratory; the exact location of this plasmalike kallikrein is not yet clear.
Some caution should be exerted when evaluating the role indicated by our data of AT1 and AT2 receptors in NO release. We used high doses of the antagonists, so there may be some nonspecificity, especially for PD 123319. However, there appeared to be some selectivity at these doses, because the response to Ang IV was selectively blocked by the AT2 antagonist (Fig 5⇑) at the doses used. In addition, we used high doses of all agonists in our study to elicit significant nitrite release. This may indicate that high nonphysiological levels of angiotensins are needed to activate local kinin and NO production or more likely that there are enzymes present, as evidenced by the presence of angiotensin-converting enzyme in our study, which metabolizes angiotensins.
In summary, there is a local kinin-generating system in isolated coronary microvessels and large coronary arteries that stimulates NO synthesis, measured as nitrite in vitro. Nolly et al7 have shown the existence of a kallikrein or kallikrein-like enzyme in the tail arteries and veins of the rat. In our study, Ang I, Ang II, and their fragments Ang III, Ang IV, and Ang-(1-7), acting via specific receptors, caused activation of the vascular kallikrein-kinin system, generation of bradykinin or a bradykinin-like substance, and release of NO. Although the function of vascular kinins is not clear yet, there is already much evidence that angiotensin-converting enzyme inhibitors block the breakdown of kinins in the vascular wall and increase nitrite release, as in our study, and that a significant portion of the antihypertensive effects of these drugs may be due to the local formation of kinins and subsequent NO release.23
This work was supported by PO-HL-43023, HL-18579, HL-50142, and HL-53053 from the National Heart, Lung, and Blood Institute. We would like to thank Hoechst-Roussel for the generous supply of Hoe 140, EI Du Pont de Nemours for the supply of DuP 753, and Parke-Davis for the supply of PD 123319.
- Received March 2, 1995.
- Revision received April 3, 1995.
- Accepted April 3, 1995.
Oparil S, Haber E. The renin-angiotensin system. N Engl J Med. 1974;291:389-401.
Ferrario CM, Barnes KL, Block CH, Brosnihan KB, Diz DI, Khosla MC, Santos RAS. Pathways of angiotensin formation and function in the brain. Hypertension. 1990;15(suppl I):I-13-I-19.
Iwaarden FV, de Groot PG, Sixma JJ, Berrettini M, Bouma BN. High molecular weight kininogen is present in cultured human endothelial cells: localization, isolation and characterization. Blood. 1988;71:1268-1276.
Nolly H, Carretero OA, Scicli G, Madeddu P, Scicli AG. A kallikrein-like enzyme in blood vessels of one kidney, one clip hypertensive rat. Hypertension. 1990;16:436-440.
Seyedi N, Nasjletti A, Xu X, Wolin MS, Hintze TH. NO/nitrite modulation of vasoconstrictor agents norepinephrine and angiotensin II but not endothelin from canine coronary microvessels and large arteries is due to local kinin production. Circulation. 1993;88(suppl I):I-470. Abstract.
Zeitlin IJ, Fagbemi SO, Parratt JR. Enzymes in normally perfused and ischaemic dog hearts which release a substance with kinin like activity. Cardiovasc Res. 1990;23:91-99.
Seyedi N, Shen W, Wang J, Gerritsen ME, Wolin MS, Hintze TH. Reduction in release of nitrite from microvessels and large coronary arteries of the left ventricle during heart failure. Circulation. 1992;86(suppl I):I-763. Abstract.
Gerritsen ME, Printz M. Sites of prostaglandin synthesis in the bovine heart and isolated coronary microvessels. Circ Res. 1981;49:1152-1163.
Hasegawa K, Nishimura H, Khosla MC. Angiotensin II induced endothelium dependent relaxation of fowl aorta. Am J Physiol. 1993;264:R903-R911.
Stallone JN, Nishimura H, Nasjletti A. Angiotensin II binding sites in aortic endothelium of domestic fowl. Am J Physiol. 1990;258:E777-E782.
Haberl RL, Decker PJ, Einhaupl KM. Angiotensin degradation products mediate endothelium dependent dilation of rabbit brain arterioles. Circ Res. 1991;68:1621-1627.
Chiu AT, McCall DE, Ardecky RJ, Duncia JV, Nguyen TT, Timmermans PBMWM. Angiotensin II receptors subtypes and their selective nonpeptide ligands. Receptor. 1990;10:33-40.
Kambayashi Y, Bardhan S, Takahashi K, Tsuzuki S, Inui H, Hamakubo T, Inagami T. Molecular cloning of a novel angiotensin II receptor isoform involved in phosphotyrosine phosphatase inhibition. J Biol Chem. 1993;268:24543-24546.
Mukoyama M, Nakajima M, Horiuchi M, Sasamura H, Pratt R, Dzau VJ. Expression cloning of type 2 angiotensin II receptors reveals a unique class of seven-transmembrane receptors. J Biol Chem. 1993;268:24539-24542.
Wiemer G, Scholkens BA, Busse R, Wagner A, Heitsch H, Linz W. The functional role of angiotensin II subtype AT2 receptors in endothelial cells and isolated ischemic rat hearts. Pharm Pharmacol Lett. 1993;3:24-27.
Cachofeiro V, Sakakibara T, Nasjletti A. Kinins, nitric oxide, and the hypotensive effect of captopril and ramiprilat in hypertension. Hypertension. 1992;19:138-145.