(Hypertension. 2002;39:51.)
© 2002 American Heart Association, Inc.
Scientific Contributions |
From the Departments of Cardiovascular Medicine (S.C., J.N.T.), Clinical Pharmacology (S.L.N.), and Physiology (J.H.C.), University of Birmingham, Birmingham, United Kingdom.
Correspondence to Dr S. Chowdhary, Department of Cardiovascular Medicine, Queen Elizabeth Hospital, Birmingham B15 2TH, United Kingdom. E-mail S.Chowdhary{at}bham.ac.uk
| Abstract |
|---|
|
|
|---|
Key Words: L-arginine nitric oxide baroreflex autonomic nervous system heart rate variability
| Introduction |
|---|
|
|
|---|
NO is synthesized by a family of NO synthases from the amino acid L-arginine, administration of which has been shown to have significant physiological effects related to enhanced NO synthesis, including improved endothelial function and vasodilatation.513 These effects are present despite the intracellular concentrations of L-arginine that greatly exceed those required for maximal enzyme kinetics,14 an inconsistency that has been called the arginine paradox. Although it is clear that some of the effects of L-arginine are not specific to NO, the impact of this amino acid on NO-mediated effects does appear to be of importance and constitutes a simple, clinically applicable method of enhancing NO pathway activity.
We have studied the effects of intravenous administration of L-arginine on HRV and BRS in healthy volunteers. The confounding influence of baroreflex unloading was controlled for by comparison with the effects of the nonNO-dependent vasodilator hydralazine given at a dose resulting in an equal drop in blood pressure. The specificity of effects for the NO pathway was determined by comparison with an infusion of the stereoisomer D-arginine, which shares many of the nonspecific effects of L-arginine but is not a substrate for NOS. Excretion of breakdown products of NO was also measured as a biochemical estimate of changes in NO pathway activity.
| Methods |
|---|
|
|
|---|
2 hours before the study and from caffeine and alcohol for 24 hours. Experimental protocols were approved by our local research ethics committee, and individual written consent was obtained.
Experimental Protocol
Protocols were of a single-blind, random-order, crossover design. All subjects had a preliminary acclimatization visit when they were trained to breathe to an audio signal set close to the individuals resting respiratory rate. Each patient was then randomly assigned to receive a 30-minute intravenous infusion of either L-arginine (1 g/min) or the control vasodilator hydralazine (loading with 20 mg over 10 minutes, followed by a maintenance infusion of 0.15 mg/min for 20 minutes) during the first of 2 studies. The remaining agent was given during a separate study visit 7 to 14 days later (n=11). This protocol was repeated in a separate set of control experiments with D-arginine (1 g/min up to 10 g, followed by saline infusion for 30 minutes; n=6) or saline infusion alone (n=9).
Studies were performed at a uniform time of day and ambient temperature (24±1°C). Subjects rested in the semisupine position, and signals from a 3-lead ECG, a Portapres (TNO Biomedical Instrumentation) recording continuous arterial pressure, and a strain gauge attached around the subjects chest quantifying respiratory excursion were digitally recorded at 500 Hz. A venous cannula was inserted into an antecubital vein for drug administration, after which subjects rested for 30 minutes before selected periods (each
257 beats) from all 3 signals were stored to disk during breathing at the predetermined frequency. Two recordings were acquired at baseline before each infusion; further recordings were then taken after 10 and 30 minutes of infusion in each study. Thus, data were obtained after administration of 10 g (10 minutes) and 30 g (30 minutes) L-arginine with control data for hydralazine at matched time points. Changes in arterial pressure documented by Portapres were confirmed with intermittent automated arm cuff sphygmomanometry.
Data Analysis
Analysis of HRV and BRS was performed by a blinded investigator according to previously published methods.4 In brief, ECG series containing <1% of ectopic beats were manually edited to remove artifact and nonsinus intervals. Standard time domain measures of the SD of RR interval values (SDNN), root mean square of successive RR interval differences (RMSDD), and percentage of successive RR interval differences exceeding 50 ms (pNN50) were calculated. Overall variability is expressed by SDNN, whereas those indexes based on successive differences in RR intervals, ie, pNN50 and RMSSD, assess high-frequency (HF; "beat to beat") variation mediated principally by the vagus nerve.15 Frequency domain analysis was performed with autoregressive modeling to determine spectral powers at low frequency (LF; centred at
0.1 Hz) and at HF (corresponding to the observed respiratory frequency). Spontaneous BRS was assessed by cross-spectral analysis of the RR interval and systolic blood pressure to derive the
-index, the transfer function of systolic power into coherent RR interval power, for both the HF (
-HF) and LF (
-LF) bands. Although both indexes reflect predominantly parasympathetic cardiac control,
-HF is thought to be most specific to the vagal limb of the cardiac baroreflex because it is determined at a frequency at which sympathetic influence is ineffective.
Biochemical Markers of NO Pathway Activity (Urinary Nitrate/Nitrite and cGMP Concentrations)
Combined urinary concentrations of nitrite and nitrate (NOx), stable breakdown products of NO, and cGMP were measured as surrogate markers of NO formation and activity. Urine samples were collected immediately before and after the study. Nitrate was first reduced to nitrite by enzymatic conversion with nitrate reductase, and NOx concentrations were subsequently quantified with the Griess reaction.16 Urinary concentrations of cGMP were assessed by ELISA (R&D systems kit assay). In each case, results are expressed as a ratio to the urinary creatinine concentration.
Statistical Analysis
Baseline hemodynamic, HRV, and BRS data were expressed as the mean of the 2 recording periods. Data for mean arterial pressure, RR interval, and urinary NOx were compared by a 2-tailed paired Students t test. Differences between groups for indexes of HRV and BRS were determined by use of the Wilcoxon signed-rank test for paired data and the Mann-Whitney test for unpaired data. Statistical significance was taken as P<0.05, and values are expressed as mean±SE.
An expanded Methods section can be found in an online data supplement available at http://www.hypertensionaha.org.
| Results |
|---|
|
|
|---|
|
|
In contrast to the matched blood pressure responses, the changes in RR interval during infusion of L-arginine and the control vasodilator agent hydralazine were discrepant. L-Arginine resulted in a reduction in heart rate, causing a statistically significant increase in mean RR interval length, whereas hydralazine produced cardioacceleration, considerably shortening the RR interval (Table 2).
|
Changes in HRV and BRS were also dissimilar between the 2 vasodilators. Despite baroreflex unloading, indexes of overall HRV (SDNN and total power) and vagally determined HF indexes of HRV (RMSSD, pNN50. and HF power) were unchanged during L-arginine infusion. In contrast, there were substantial reductions in all these indexes with hydralazine. Similarly, the ratio of LF to HF was unchanged with L-arginine, which is in contrast to the significant rise seen with hydralazine. As a further index of vagal cardiac regulation, cross-spectral BRS (both
-HF and
-LF) also showed significant falls with hydralazine but no change or even slight rises with L-arginine (Table 2).
Saline had no significant effect on HRV or BRS. Compared with saline infusion, D-arginine had no effect on most of the autonomic indexes but resulted in a small but significant reduction in measures of HF HRV, namely RMSSD and HF power, as well as in
-HF (Table 3).
|
Measures of whole-body NO pathway activity (urinary excretion of both NOx and cGMP) were significantly increased by L-arginine infusion but not by hydralazine, saline, or D-arginine (Figure 2).
|
| Discussion |
|---|
|
|
|---|
2 mmol/L) can release vasodilator hormones such as insulin and prolactin, the magnitude of their release at this dose has been shown to be insufficient to cause hemodynamic effects. In contrast, higher doses of arginine (resulting in plasma levels of 5 mmol/L) can produce a nonstereospecific vascular response as a result of greater release of these vasodilator hormones.17 Second, the stereospecific nature of our results was confirmed directly. D-Arginine produced no measurable hemodynamic effects with slight falls in HF measures of HRV (RMSSD and HF power) and BRS (
-HF) compared with saline (in contrast to the effects of L-arginine). Because D-arginine shares the same hormonal effects as L-arginine but does not act as a substrate for NOS, this strongly suggests specificity for NO-mediated actions in our protocol. We hypothesize that the small falls in vagally mediated HRV and BRS seen with D-arginine may have been reflex effects consequent to the release of insulin and other vasodilator hormones, which, although causing a mild vasodilatation, did not result in a measurable drop in blood pressure because of baroreflex compensation. Although these nonspecific vasodilator effects would also be present with L-arginine, together with specific NO-mediated vasodilatation, the normal reflex vagal withdrawal appears to have been overridden by the vagotonic effects of increased NO synthesis. Finally, in agreement with previous studies,5,6,8 the urinary excretion of NOx over the time course of the experiment was increased by L-arginine but not by D-arginine, saline, or hydralazine, providing further evidence of a stereospecific increase in NO generation. The concept that the L-arginine supply can be rate limiting to NO synthesis is controversial because biochemical considerations dictate that intracellular concentrations of L-arginine greatly exceed those required for maximal NOS activity.14 Although some studies have failed to show a significant vasomotor influence of systemic L-arginine infusion in healthy volunteers,18,19 these are outnumbered by the studies that do show falls in blood pressure and total peripheral resistance in healthy volunteers58 and hypertensive subjects.6 Furthermore, L-arginine administration has been shown consistently to improve endothelial NO-mediated vasodilator responses in patients with hypercholesterolemia9,11,13 and in the atheromatous coronary circulation.9,10,12 In agreement with our data, many healthy volunteer studies have demonstrated hemodynamic actions after only 10 g L-arginine (reducing the likelihood of nonstereospecific actions) and, importantly, showed biochemical evidence of increased NO production.58 One possible answer to this arginine paradox may lie in the intracellular compartmentalization of constitutive NOS within caveolae, which, in conjunction with certain colocalized arginine transporters (eg, cationic amino acid transporter-1), may serve to regulate the supply of L-arginine to NOS.20
Few other studies have examined the influence of L-arginine on HRV in humans. Nomura et al21 infused 30 g L-arginine (1 g/min) into 8 healthy volunteers and found significant increases in indexes of vagal cardiac control at as early as 5 to 10 minutes of infusion despite a fall in systolic blood pressure. Thus, the vagotonic effects of L-arginine in their study were even more pronounced than those observed in our population, a difference perhaps explained by ethnic variation. These investigators did not directly study the specificity of autonomic effects to the L-arginineNO pathway. Instead, they indirectly concluded that the autonomic influence of L-arginine was independent of NO, despite a rise in L-citrulline, a byproduct of NO production, because it was not reproduced by infusion of the NO donor isosorbide dinitrate. We suggest that the vagotonic effects of isosorbide dinitrate in the study of Nomura et al may have been blunted by a greater degree of baroreflex unloading and the observed activation of humoral factors such as noradrenaline, renin, and angiotensin, effects not seen with L-arginine. We have previously demonstrated that when the fall in blood pressure produced by an NO donor (sodium nitroprusside) is matched with that resulting from a nonNO-dependent vasodilator (hydralazine), levels of vagally mediated HRV are higher with the NO donor.4 L-Arginine has also been shown to increase HF power in patients with liver cirrhosis.22 Our protocol cannot localize the site of action of L-arginine in its modification of cardiac autonomic activity, but animal data suggest that L-arginine may act as a substrate in a number of central and peripheral neuronal populations within the cardiac baroreflex arc that exhibit a discrete localization of neuronal NOS.3 Functional activity for L-arginine has been demonstrated within the primary relay for baroreceptor afferents in the brainstem, the nucleus tractus solitarii. Microinjection of L-arginine but not D-arginine into the nucleus tractus solitarii of rats caused an increase in nucleus tractus solitarii neuronal activity, leading to a fall in systemic blood pressure and bradycardia. Both the neuronal and hemodynamic effects were blocked by a neuronal NOS inhibitor and an NO scavenger and reproduced by NO donors.23,24 Furthermore, NO has been shown to increase neuronal activity within other central sites regulating parasympathetic outflow to the heart.25 Peripherally, NO potentiates the bradycardic effects of parasympathetic stimulation2628 and enhances the ability of the efferent vagus to antagonize sympathetic cardiac responses,29 with both presynaptic and postsynaptic mechanisms postulated.26,28
Study Limitations
We cannot exclude a sympathetic nervous influence of L-arginine on our results. Although the ratio of LF to HF and
-LF contain some information on sympathetic cardiac control, they lack specificity because they are also influenced by vagal activity.15 L-Arginine may have acted to inhibit sympathetic nervous activity either through the sympatholytic actions of NO3 or via stimulation of central
2-receptors by its metabolite agmatine.30 However, L-arginine infusion has been shown not to alter levels of noradrenaline spillover in healthy human subjects.21
Establishing the specificity of observed effects to activation of NOS may have been aided by coadministration of NOS inhibitors. However, opposing effects would have extended to hemodynamic and autonomic influence, rendering the final results difficult to interpret. Finally, we did not study the influence of other cofactors for NOS activity such as tetrahydrobiopterin. The supply of L-arginine relative to BH4 may be critical in determining whether NOS generates NO or O2-, and it is possible that the observed autonomic activity of L-arginine may have been even greater if coadministered with tetrahydrobiopterin.
In conclusion, we have demonstrated that a vasodepressor dose of L-arginine did not result in heart rate acceleration or depression of vagally mediated autonomic indexes, in contrast to a control vasodilator. These findings were not reproduced by D-arginine, which, in conjunction with biochemical evidence of increased NO generation, suggests that these effects were mediated by the L-arginineNO pathway. Extensive study has been made of the effects of L-arginine supplementation to correct endothelial dysfunction observed in various cardiac disease states. Our results now suggest that the potential for L-arginine therapy to correct the impairment of vagal control associated with mortality in cardiac disease also merits investigation.
| Acknowledgments |
|---|
Received June 7, 2001; first decision July 5, 2001; accepted August 17, 2001.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
F. Iellamo, M. Tesauro, S. Rizza, S. Aquilani, C. Cardillo, M. Iantorno, M. Turriziani, and R. Lauro Concomitant Impairment in Endothelial Function and Neural Cardiovascular Regulation in Offspring of Type 2 Diabetic Subjects Hypertension, September 1, 2006; 48(3): 418 - 423. [Abstract] [Full Text] [PDF] |
||||
![]() |
C J Charles, D L Jardine, M G Nicholls, and A M Richards Adrenomedullin increases cardiac sympathetic nerve activity in normal conscious sheep J. Endocrinol., November 1, 2005; 187(2): 275 - 281. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Chowdhary, A. M. Marsh, J. H. Coote, and J. N. Townend Nitric Oxide and Cardiac Muscarinic Control in Humans Hypertension, May 1, 2004; 43(5): 1023 - 1028. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Piccirillo, M. Nocco, A. Moise, M. Lionetti, C. Naso, S. di Carlo, and V. Marigliano Influence of Vitamin C on Baroreflex Sensitivity in Chronic Heart Failure Hypertension, June 1, 2003; 41(6): 1240 - 1245. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2002 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |