C-Type Natriuretic Peptide
An Endogenous Inhibitor of Vascular Angiotensin-Converting Enzyme Activity
Background Atrial and B-type natriuretic peptide are both known to be antagonists of the renin-angiotensin system. C-type natriuretic peptide (CNP) is a new member of this family except that its principal source is the vascular endothelium. This study tested the hypothesis that CNP is a local inhibitor of vascular angiotensin-converting enzyme (ACE) activity.
Methods and Results Vascular ACE activity was assessed by the differential vascular response to angiotensin I and angiotensin II. Healthy male volunteers were studied with the use of brachial artery infusions of angiotensin I and angiotensin II at two doses, with and without coinfusion of CNP at 500 pmol/min (n=8) and hydralazine at 10 μg/min (n=8) (as a nonspecific vasodilator control). CNP alone and hydralazine alone caused similar increases in forearm blood flow (CNP+, 93.0±14.8%; hydralazine+, 84.2±22.6%). CNP inhibited the vasoconstrictive effect of angiotensin I (reduction in overall effect with CNP, 56.8±12.9%; P<.001) but not that of angiotensin II. Hydralazine did not significantly inhibit the effect of either angiotensin I or angiotensin II.
Conclusions This evidence of a differential effect of CNP on the vascular response to angiotensin I but not to angiotensin II suggests that CNP acts as a local endogenous regulator of vascular ACE activity in the human forearm resistance vessels.
C-type natriuretic peptide was first discovered in 1990.1 It shares a similar structure and common receptors with the cardiac natriuretic peptides ANP and BNP, but in contrast with these other peptides, CNP appears to be secreted predominantly from the vascular endothelium.2 The physiological role of CNP has yet to be clearly defined: it has been shown to have vasodilatory effects on both arteries and veins3 and also has an antiproliferative effect on vascular smooth muscle cells in culture.4 In contrast, the diuretic and natriuretic effects of CNP are limited in comparison with ANP and BNP.5
One important feature of the actions of natriuretic peptides is their antagonism of the renin-angiotensin-aldosterone system. ANP inhibits renin release from juxtaglomerular cells in the kidney6 ; it has an inhibitory effect on the release of aldosterone7 ; and, in endothelial cell culture, ANP reduces the activity of ACE.8 We hypothesized that CNP, which is released from endothelial cells, might be involved in the local regulation of the vascular renin-angiotensin system, and in particular that it might be acting as an endogenous inhibitor of the activity of endothelial ACE.
The measurement of ACE activity is problematic for two reasons. First, the measurement of angiotensin II concentrations is difficult in the presence of high concentrations of angiotensin I that occur when ACE activity is inhibited.9 This problem is compounded by the very short tissue half-life of angiotensin II.10 Second, ACE can be assessed by using a variety of synthetic substrates to obtain in vitro measurements of plasma ACE activity, but this technique ignores the large amount of ACE that is bound to the luminal surface of the endothelial cell. A better approach is to assess changes in ACE activity in vivo by measuring the differential biological response to angiotensin I (which is inactive until it is metabolized to angiotensin II, predominantly by ACE) and angiotensin II (which acts independent of ACE). This has the advantage of using the natural substrate for the enzyme and avoids the problem of rapid tissue degeneration of angiotensins. The use of brachial artery infusion and measurement of forearm muscle blood flow by venous occlusion strain-gauge plethysmography, with the noninfused arm as a control, allows the study of local vascular effects of the infused substances while minimizing systemic influences. With this method, we sought to determine whether CNP acts locally to inhibit the biological consequences of ACE activity in the human forearm.
To ensure that nonspecific vasodilation did not produce a differential vascular response to angiotensin I and angiotensin II, we performed a parallel control experiment using hydralazine at a dose that produced a similar degree of vasodilatation as CNP. To see whether the effects of CNP occurred within the bloodstream or within vascular tissue, we performed an additional experiment in vitro to assess the direct effect of CNP on plasma ACE activity.
The study consisted of three clinical experiments and one in vitro experiment. All protocols were approved by the local committee on medical research ethics. Fourteen healthy male volunteers, aged 21 to 36 years, with no history of cardiovascular disease were used in the clinical experiments, which were performed at least 3 hours after the last meal after a 24-hour period of abstinence from cigarette smoking, alcohol, and caffeine, in an air-conditioned room with temperature controlled at 23°C to 25°C.
On each study day, the subject lay supine, and a mercury-in-silastic strain gauge (Medasonics) was applied to each forearm at the point of maximal muscle bulk. The position of the gauge was determined by measuring the distance from the olecranon process and was kept constant for each individual between study days. Cuffs were placed around each wrist and upper arm and were attached to a rapid cuff inflator (Hokanson). FBF measurements were taken from both arms over a 2-minute period at the end of each dose interval, during which the wrist cuffs were inflated to 200 mm Hg to exclude the hand circulation. Each measurement was taken as the mean of five readings, which were obtained during periodic inflation of the upper arm cuffs to 40 mm Hg (to occlude venous outflow) for 10 seconds in every 15 seconds. Data from the strain gauges were processed by a plethysmograph (Medasonics) and analyzed using MacLab computer hardware and software. Heart rate and blood pressure were measured by a semiautomated sphygmomanometer (Dinamap) after each infusion.
A 27-gauge needle was inserted into the brachial artery of the nondominant arm under local anesthesia, and 0.9% saline was infused for at least 30 minutes prior to infusion of peptides. Strain gauge measurements were taken at 10-minute intervals until stable readings, defined by three consecutive measurements with less than 10% variability, were obtained. The mean of the ratio of measurements from both arms at these three time points was taken as the baseline ratio of forearm blood flow. The peptides were dissolved in 0.9% saline, and the infusion rate was kept constant at 1 mL/min throughout the study. All peptides were supplied by Calbiochem-Novabiochem.
Eight subjects were studied on two separate days. On day 1, angiotensin I was infused at 16 pmol/min and 64 pmol/min for 7 minutes each. After a 40-minute washout period, CNP was infused at a rate of 100 pmol/min for 34 minutes. For the last 14 minutes of the CNP infusion, angiotensin I was coinfused at the above rates, using a Y-connector to minimize ex vivo mixing of infusates. On a separate study day, angiotensin II was infused in place of angiotensin I at doses of 4 pmol/min and 16 pmol/min (these doses were chosen to give reductions in forearm blood flow of approximately 50% and 70%, respectively). Fig 1⇓ demonstrates the study design.
Eight subjects were studied on two separate study days according to the above protocol, with the dose of CNP increased to 500 pmol/min.
Eight subjects were studied on two separate days according to a similar protocol to experiment 1 (see Fig 1⇑). Hydralazine was given at a dose of 10 μg/min for 10 minutes in place of CNP. This dose was selected after pilot studies to produce a similar degree of vasodilation to CNP. The second infusions of angiotensin I and angiotensin II were performed 50 minutes after the hydralazine infusion was stopped, by which time the increase in blood flow had reached a plateau.
All infusions were administered on a double-blind basis with balanced randomization of the order of administration of angiotensin I and II. Analysis was performed separately by an observer who was blinded to the infusate used. In experiment 2, venous blood samples were taken from the noninfused arm at baseline, after the initial angiotensin infusion and after the combined CNP and angiotensin infusion. Plasma renin activity was measured by radioimmunoassay (Sorin Biomedica).
In Vitro Study
Venous plasma samples from 6 of the volunteers were divided into aliquots that were incubated at 37°C with vehicle or increasing concentrations of ANP, CNP, and lisinopril (a specific inhibitor of ACE) from 10−7 to 10−11 mol/L. ANP and CNP were supplied by Calbiochem-Novabiochem. Lisinopril was supplied by Zeneca Pharmaceuticals. Plasma ACE activity was then measured immediately with a spectrophotometric assay using the synthetic substrate N-3-(2-furyl) acryloyl-l-phenylalanylglycylglycine.11
Measurements of FBF at each time point were expressed as changes from baseline. For statistical analysis, the ratio of strain gauge measurements from the infused arm over the measurements from the control arm were log-transformed as the assumption of a normal distribution for a ratio of two variables is generally poor. On the transformed scale a ratio becomes a difference, and the assumptions required for parametric tests are more likely to be satisfied.12 All transformed values were converted to percentage changes from baseline for presentation. The effects of angiotensin I and angiotensin II, alone and in combination with CNP, on the FBF in all of the studies were compared with the use of ANOVA. For the in vitro study, the plasma ACE activity with different concentrations of CNP was compared with the use of ANOVA. Results are expressed as mean±SEM. Statistical significance was accepted for values of P<.05.
Angiotensin I and angiotensin II, when given alone, produced similar reductions in FBF at both doses: angiotensin I dose 1, −53.1±17.4%; angiotensin II dose 1, −42.1±10.0% (P=NS); angiotensin I dose 2, −68.2±16.9%; angiotensin II dose 2, −61.8±13.2% (P=NS). CNP given at 100 pmol/min increased FBF by 2.7±11.3% (P=NS). The vasoconstrictor responses to angiotensin I and angiotensin II were not significantly altered by coinfusion of CNP at 100 pmol/min.
The vasoconstrictor effects of angiotensin I and angiotensin II, with and without CNP at 500 pmol/min, are shown in Fig 2⇓. CNP produced an increase in FBF from baseline of 93.0±14.8% (P<.00001). CNP significantly reduced the vasoconstrictor response to angiotensin I and had no significant effect on the vasoconstrictor response to angiotensin II. The difference between the overall vasoconstrictor effect of angiotensin I alone and angiotensin I coinfused with CNP was 56.8±12.9% (P<.001). The difference between the overall vasoconstrictor effect of angiotensin II alone and angiotensin II coinfused with CNP was 15.7±8.9% (P=NS). CNP had a significantly greater effect on the vasoconstriction after angiotensin I than on the vasoconstriction after angiotensin II (P<.05).
Plasma renin activity in venous blood from the noninfused arm during experiment 2 showed no significant changes after infusion of angiotensin I or angiotensin II, with or without CNP. Heart rate and blood pressure did not change significantly during any of the infusions in any of the experiments.
Hydralazine produced an increase in FBF of 84.2±22.6%, which was maximal at 50 minutes. The effects of angiotensin I and angiotensin II, with and without pretreatment with hydralazine, on FBF are shown in Fig 2⇑. Hydralazine produced a nonsignificant reduction in the vasoconstrictive effect of both angiotensin I and angiotensin II. There was no significant difference between the effects of hydralazine on either peptide.
Intraindividual reproducibility was assessed in 6 subjects who were given infusions of angiotensin I and angiotensin II at the above doses on two separate occasions. The intraindividual coefficients of variability for changes in FBF were angiotensin I: dose 1, 9.4%; dose 2, 4.8%; angiotensin II: dose 1, 11.9%; dose 2, 5.8%.
In Vitro Experiment
Plasma ACE activities in samples incubated with different concentrations of ANP, CNP, and lisinopril are shown in Fig 3⇓.
The main finding of this study is that CNP inhibits the vasoconstrictor effect of angiotensin I more than it inhibits the vasoconstrictor effect of angiotensin II in the human forearm. From experiment 1 it can be seen that the vasoconstrictor effect of intra-arterial angiotensin I and angiotensin II is immediately reproducible on a single study day, while the results from individuals studied on separate days demonstrate longer-term reproducibility of the acute vascular changes produced by these peptides. The lower dose of CNP used in this study was apparently ineffective both as a vasodilator, which is in contrast with similar doses of ANP,13 and as a modulator of the renin-angiotensin system.
With the higher dose of CNP there was a nonsignificant trend toward a reduced vasoconstrictor effect of angiotensin II (Fig 2⇑); a similar result was also evident when hydralazine was coinfused with angiotensin I and angiotensin II. This trend is not surprising in view of the vasodilatatory properties of CNP and hydralazine, although the lack of a marked effect of CNP on the vascular response to angiotensin II is consistent with a previous study that included systemic coinfusion of these peptides in humans.5 With regard to the current study, the important observation is that CNP clearly had a differential effect on angiotensin I and angiotensin II, whereas hydralazine had an identical effect on both of the angiotensin peptides.
The relevance of the higher dose of CNP that was used in this study to the physiological regulation of vascular tone is uncertain. With total FBF in the range of 12 to 40 mL/min, the higher dose of CNP in this study would produce a concentration of approximately 12.5 to 42 pmol/mL in the arterial blood, far greater than the relatively low concentrations at which CNP has been detected in venous plasma in humans.14 However, the release of CNP from vascular endothelial cells implies that local concentrations in the vicinity of vascular ACE may be considerably higher. To confirm a physiological role for CNP in the control of the vascular renin-angiotensin system would require blockade of its effects with a monoclonal antibody or a specific receptor antagonist, neither of which is currently available for use in humans. The doses of angiotensin I and angiotensin II that were used in the present study are considerably below the lowest doses that have been shown to have systemic effects on vascular tone.15 During the intra-arterial infusions of angiotensins and CNP, there was no change in heart rate, blood pressure, or plasma renin activity, confirming that the infusions did not have marked systemic effects.
There are several possible explanations, in addition to inhibition of ACE, for the differential effect of CNP on the response to angiotensin I and angiotensin II. First, CNP could theoretically inhibit non-ACE pathways of angiotensin II generation in the forearm. While such pathways certainly exist, a previous study demonstrated that enalaprilat (a specific inhibitor of ACE) almost completely inhibited the response to infused angiotensin I in the forearm,16 suggesting that these non-ACE pathways are not of major physiological importance in the forearm of normal humans at rest. Second, CNP might not affect the activity of ACE directly but could in theory reduce the delivery of angiotensin I to the enzymatic sites due to increased blood flow. However, the lack of a differential effect on angiotensin I and angiotensin II of hydralazine at a dose that produced comparable vasodilatation to CNP suggests that this mechanism does not explain the observed effects of CNP.
The mechanisms whereby CNP might regulate vascular ACE are worthy of some discussion. In our in vitro experiment, neither ANP nor CNP had a direct effect on plasma ACE, which was potently inhibited by the specific ACE inhibitor drug lisinopril. Therefore, the effect of CNP on ACE is tissue dependent and is therefore likely to be mediated by specific receptors in the vascular endothelium. CNP is known to act via two types of receptor, natriuretic peptide receptors B and C.17 The former of these is linked to guanylate cyclase, causing accumulation of cGMP, while the latter, which was previously thought to be biologically silent, is now thought to produce several effects including inhibition of cAMP synthesis.18 Both of these receptor groups have been identified in vascular endothelial cells.19 ACE activity in cultured endothelial cells is stimulated by cAMP and by agents that increase its production but is not affected by cGMP,20 suggesting that CNP may act via natriuretic peptide receptor C to reduce cAMP production and inhibit ACE activity. It is also possible that CNP might inhibit the synthesis of ACE by endothelial cells, but this is unlikely to have occurred during the short infusion periods used in this study, and it is notable that in cultured cells inhibition of ACE activity occurs after only 1 hour of incubation with ANP.8 Rather than acting on the enzyme directly, CNP could alter the environment around the enzyme sites: it has been shown to cause activation of voltage-dependent potassium channels,21 and it may affect the concentration of other ions that affect the activity of ACE, for example, chloride.22
We have demonstrated that CNP causes a differential suppression of the vasoconstrictive effects of angiotensin I as opposed to angiotensin II in healthy subjects. This suggests an important role for CNP as a paracrine endogenous regulator of vascular ACE activity. Further studies are required to confirm this finding, to investigate the possible mechanisms, and to establish how this finding relates to atherogenesis.
Selected Abbreviations and Acronyms
|ANP||=||atrial natriuretic peptide|
|BNP||=||B-type natriuretic peptide|
|CNP||=||C-type natriuretic peptide|
|FBF||=||forearm blood flow|
Dr Davidson was supported by a grant from the British Heart Foundation. Lisinopril was a kind gift from Zeneca Pharmaceuticals. We would like to thank Mrs L. McFarlane for expert technical assistance.
- Received August 14, 1995.
- Revision received October 20, 1995.
- Accepted October 23, 1995.
- Copyright © 1996 by American Heart Association
Suga S, Nakao K, Itoh H, Komatsu Y, Ogawa Y, Hama N, Imura H. Endothelial production of C-type natriuretic peptide and its marked augmentation by transforming growth factor-β. J Clin Invest. 1992;90:1145-1149.
Wei CM, Aarhus LL, Miller VM, Burnett JC. The actions of natriuretic peptides on isolated human saphenous veins and internal mammary arteries. J Am Coll Cardiol. 1994;23:177A. Abstract.
Porter JG, Catalano R, McEnroe G, Lewicki JA, Protter AA. C type natriuretic peptide inhibits growth factor-dependent DNA synthesis in smooth muscle cells. Am J Physiol. 1992;263:C1001-C1006.
Kurtz A, Della Bruna R, Pfeilschifter J, Taugner R, Bauer C. Atrial natriuretic peptide inhibits renin release from juxtaglomerular cells by a cGMP mediated process. Proc Natl Acad Sci U S A. 1986;83:4769-4773.
Delkers W, Kleiner S, Bähr V. Effects of incremental infusions of atrial natriuretic factor on aldosterone, renin and blood pressure in humans. Hypertension. 1988;12:462-446.
Nussberger J, Brunner DB, Waeber B, Brunner HR. True versus immunoreactive angiotensin II in human plasma. Hypertension. 1985;7(suppl I):I-1-I-7.
Lyall F, Morton JJ, Lever AF, Cragoe EJ. Angiotensin II activates Na-H exchange and stimulates growth in cultured vascular smooth muscle cells. J Hypertens. 1988;6(suppl 4):S438-S441.
Stingo AJ, Clavell AL, Heublein DM, Wei CM, Pittelkow MR, Burnett JC Jr. Presence of C-type natriuretic peptide in cultured human endothelial cells and plasma. Am J Physiol. 1992;263:H1318-H1321.
Ames RP, Borkowski AJ, Sicinski AM, Laragh JH. Prolonged infusions of angiotensin II and norepinephrine and blood pressure, electrolyte balance, and aldosterone and cortisol secretion in normal man and in cirrhosis with ascites. J Clin Invest. 1965;44:1171-1178.
Levin ER. Natriuretic peptide C-receptor: more than a clearance receptor. Am J Physiol. 1993;264:E483-E489.
Wei CM, Hu S, Kim CH, Miller V, Burnett JC Jr. The in vitro actions of C-type natriuretic peptide in coronary arteries. Circulation. 1993;88(suppl I):I-181. Abstract.