Is Measurement of Cyclic Guanosine Monophosphate in Plasma or Urine Suitable for Assessing In Vivo Nitric Oxide Production?
To the Editor:
In their recently published article on nitric oxide (NO) synthesis in patients with peripheral arterial occlusive disease Böger et al1 based their conclusions among others on urinary excretion of cyclic guanosine monophosphate (cGMP). The increasing use of urinary and plasma cGMP as a marker of NO production prompts us to critically comment on the basis of doing this. NO stimulates soluble guanylate cyclase and elevates intracellular cGMP.2 The other isoenzyme of guanylate cyclase, particulate guanylate cyclase, is stimulated by natriuretic peptides, which also leads to an increase in intracellular cGMP.3 The induction of cGMP either by natriuretic peptides, NO, or nitrates in target tissues may cause an egression of cGMP into the supernatant.4 5 We observed a release of cGMP into the medium after stimulation of human internal mammary artery grafts with either atrial natriuretic peptide (ANP) or SIN-1 (unpublished results). However, much higher concentrations of SIN-1 were necessary to achieve comparably high cGMP concentrations in the medium.
In humans, ANP injections cause a rapid and pronounced increase in plasma and urinary cGMP,6 7 whereas nitroglycerol infusions or molsidomine injections lead to a nonsignificant increase or no increase in peripheral venous plasma cGMP concentrations.6 8 cGMP is only partly eliminated from plasma by glomerular filtration, and most of plasma cGMP is eliminated by extrarenal clearance. Urinary cGMP is primarily of renal cellular origin and correlated with the natriuresis induced by ANP.8 Therefore urinary cGMP has been proposed as a biologic marker for the renal activities of natriuretic peptides in vivo.8 We found no correlation between plasma and urinary cGMP concentrations in humans.9 Urinary and plasma cGMP concentrations are influenced by renal function. Urinary cGMP per gram of creatinine was significantly lower and plasma cGMP of patients with renal diseases was significantly higher than that of control subjects.6 9 Therefore urinary cGMP is mainly influenced by the biologic activities of natriuretic peptides and also by renal function. As a consequence, urinary cGMP does not appear to be a reliable marker for the in vivo NO production in humans.
However, Tsutamoto et al10 could demonstrate that the arteriovenous cGMP difference may be useful for assessing the local stimulation of the soluble guanylate cyclase in humans. Although venous cGMP concentrations did not change, the decrease in arterial cGMP during nitroglycerol infusion indicated a local cGMP production by stimulation of the soluble guanylate cyclase. A decrease in ANP plasma concentrations during nitroglycerol infusion excluded natriuretic peptides as a cause of the observed cGMP production. cGMP production correlated with hemodynamics and did not occur after captopril administration.
In conclusion, most of the plasma cGMP is derived from the endogenous natriuretic peptides and only a minor part from other pathways, such as soluble guanylate cyclase. The change of plasma cGMP concentrations by nitrates is much smaller than that by ANP, with the same hemodynamic effect.10 In contrast to urinary cGMP, the arteriovenous cGMP production may allow assessment of the activation of soluble guanylatcyclase in vivo. However, it is mandatory to measure natriuretic peptide concentrations simultaneously to exclude changes in natriuretic peptides as the underlying cause of cGMP production.
- Copyright © 1998 by American Heart Association
Böger RH, Bode-Böger SM, Thiele W, Junker W, Alexander K, Frölich JC. Biochemical evidence for impaired nitric oxide synthesis in patients with peripheral arterial occlusive disease. Circulation.. 1997;95:2068–2074.
Hamet P, Pang SC, Tremblay J. Atrial natriuretic factor-induced egression of cyclic guanosine monophosphate in cultured vascular smooth muscle and endothelial cells. J Biol Chem.. 1989;264:12364–12369.
Billiar TR, Curran RD, Harbrecht BG, Stadler J, Williams DL, Ochoa JB, Di Silvio M, Simmons RL, Murray SA. Association between synthesis and release of cGMP and nitric oxide biosynthesis by hepatocytes. Am J Physiol.. 1992;262:C1077–C1082.
Vorderwinkler KP, Artner-Dworzak E, Jakob G, Mair J, Dienstl F, Pichler M, Puschendorf B. Release of cyclic guanosine monophosphate evaluated as a diagnostic tool in cardiac diseases. Clin Chem.. 1991;37:186–190.
Jakob G, Mair J, Vorderwinkler KP, Judmaier G, König P, Zwierzina H, Pichler M, Puschendorf B, Clinical significance of urinary cyclic guanosine monophosphate in diagnosis of heart failure. Clin Chem.. 1994;40:96–100.
Tsutamoto T, Kinoshita M, Ohbayashi Y, Wada A, Maeda Y, Adachi T. Plasma arteriovenous cGMP difference as a useful indicator of nitrate tolerance in patients with heart failure. Circulation.. 1994;90:823–829.
The letter by Drs Mair and Puschendorf gives us an opportunity to comment on the use of plasma and urinary nitrate and cGMP as indicators of in vivo nitric oxide formation.
Nitric oxide is formed in the vascular endothelium and in other tissues. One of its major targets is the soluble guanylyl cyclase (sGC), which leads to the formation of cGMP. NO is rapidly inactivated through oxidation to nitrite and nitrate; under pathophysiologic conditions, oxidative inactivation may occur even before the sGC has been stimulated.R1 Both cGMP and nitrite/nitrate can be found in conditioned endothelial cell media, in plasma, and in urine. Because the chemical half-life of NO is in the range of seconds, NO itself can hardly be measured in vivo. Two main strategies have therefore been followed to assess NO activity in vivo. One is to determine NO-dependent vasodilation and the other is to measure the metabolites and/or second messenger of NO, nitrite/nitrate and cGMP, as biochemical surrogates for NO. However, both of these strategies may be limited by some constraints: NO-dependent vasodilation allows one to assess the biologic activity of NO irrespective of whether decreased elaboration of NO or enhanced oxidative inactivation may underlie this disorder.R1 On the other hand, the simultaneous quantitation of nitrite/nitrate and cGMP allows differentiation between impaired NO synthesis (in which case nitrite/nitrate levels and cGMP levels are expected to be low) and oxidative inactivation (in which case cGMP levels are expected to be low, but nitrite/nitrate levels should be normal or elevated). This approach is curtailed by the potential influence of dietary nitrate intakeR2 and by cGMP formation by the particulate guanylyl cyclase as discussed by Drs Mair and Puschendorf. However, because dietary nitrate would only affect nitrate levels but not those of cGMP, and activation of the particulate GC would only affect cGMP levels but not those of nitrate, we have repeatedly advocated the parallel use of both indicators to estimate NO elaboration.R3 R4 R5 R6 Furthermore, plasma levels of these index molecules may only reflect a momentary situation in a localized area of the circulation, whereas urinary levels reflect systemic NO production rates but may be affected by renal excretory function. We have addressed this latter question in an experimental study and found that correction of urinary nitrate and cGMP concentrations by urinary creatinine concentration (ie, urinary excretion rates of these metabolites instead of urinary concentrations) eliminates the dependency on renal excretory function.R3 Using this approach, we and others have adopted nitrate and cGMP measurements as reliable indicators of NO elaboration during physiologic (eg, physical exercise)R4 R7 and pharmacologic stimulation.R5 R8 R9 In our recent study to which Drs Mair and Puschendorf refer,R6 we have analyzed basal urinary nitrate and cGMP excretion rates in patients with peripheral arterial occlusive disease of the legs. Twenty-three of these patients had impaired renal function as judged by creatinine clearance. We have reanalyzed urinary nitrate and cGMP excretion rates in the subgroups of patients with normal or impaired renal function and found no statistically significant differences in these parameters despite a 50% reduction of creatinine clearance in the patients with impaired renal function (Figure⇑ ). This demonstrates again that urinary nitrate and cGMP excretion rates are independent of renal excretory function.
The in vitro studies cited by Drs Mair and Puschendorf confirm that cGMP levels are influenced both by ANP and NO donors. However, it is difficult to us to extrapolate results obtained with pharmacologic concentrations of ANP and NO donors in isolated arteries in vitro to physiologic situations in vivo. Indeed, Arnal and coworkersR10 found that the in vivo basal aortic cGMP levels in rats were mainly dependent on NO synthase:soluble guanylyl cyclase activity. During chronic NO synthase inhibition, aortic cGMP levels significantly decreased; in this setting cGMP levels were correlated with ANP levels. The reduction of aortic cGMP levels during NO synthase inhibition was reversed by l-arginine. Tolins and coworkersR11 reported similar findings from an in vivo study in which they infused acetylcholine into rats under euvolemic conditions. Acetylcholine induced hypotension with concomitant increased urinary cyclic GMP excretion. These effects were reversed by L-NMMA, they were not paralleled by increased plasma ANP levels. These authors also showed a correlation between the changes in urinary cGMP excretion and the acetylcholine-induced decrease in systemic blood pressure. We have shown that changes in urinary cGMP excretion rates could not be explained by differences in ANP levels in healthy humans at baselineR4 and even after intravenous volume loading.R12 This may be different in patients in whom the ANP system is activated, as in chronic heart failure. The usefulness of urinary nitrate and cGMP as markers for systemic NO elaboration in vivo should therefore be evaluated in any patient group separately.
Green LC, Ruiz de Luzuriaga K, Wagner DA, Rand W, Istfan N, Young VR, Tannenbaum SR: nitrate biosynthesis in man. Proc Natl Acad Sci USA. 1981;78:7764–7768.
Bode-Böger SM, Böger RH, Alfke H, Heinzel D, Tsikas D, Creutzig A, Alexander K, Frölich JC. l-Arginine induces nitric oxide-dependent vasodilation in patients with critical limb ischemia: a randomized, controlled study. Circulation. 1996;93:85–90.
Böger RH, Bode-Böger SM, Thiele W, Junker W, Alexander K, Frölich JC. Biochemical evidence for impaired nitric oxide synthesis in patients with peripheral arterial occlusive disease. Circulation. 1997;95:2068–2074.
Jungersten L, Ambring A, Wall B, Wennmalm A. Both physical fitness and acute exercise regulate nitric oxide formation in healthy humans. J Appl Physiol. 1997;82:760–764.
Suzuki H, Ikenaga H, Hishikawa K, Nakaki T, Kato R, Saruta T. Increases in NO2−/NO3− excretion in the urine as an indicator of the release of endothelium-derived relaxing factor during elevation of blood pressure. Clin Sci. 192;82:631–634.
Arnal JF, Warin L, Michel JB. Determinants of aortic cyclic guanosine monophosphate in hypertension induced by chronic inhibition of NO synthase. J Clin Invest. 1992;90:6647–6652.
Tolins JP, Palmer RMJ, Moncada S, Raij L. Role of endothelium-derived relaxing factor in regulation of renal hemodynamic responses. Am J Physiol. 1990;258:H655–H662.