Nitric Oxide, Peroxynitrite and cGMP in Atherosclerosis-Induced Hypertension in Rabbits: Beneficial Effects of CicletanineSzilvássy Z.a · Csont T.b · Páli T.c · Droy-Lefaix M.-T.d · Ferdinandy P.b
aDepartment of Pharmacology, University of Debrecen, Debrecen, bCardiovascular Research Group, Department of Biochemistry, University of Szeged, cDepartment of Biophysics, Biological Research Center, Szeged, Hungary, and dIPSEN-Beaufour, Paris, France
We studied the effect of the furopyridine derivative antihypertensive drug, cicletanine, on blood pressure, vascular nitric oxide (NO) and cyclic guanosine 3′:5′-monophosphate (cGMP) content in the aorta and the renal and carotid arteries, aortic superoxide production, and serum nitrotyrosine level in hypertensive/atherosclerotic rabbits. The effect of cicletanine was compared to that of furosemide. Rabbits were fed a normal or a cholesterol-enriched (1.5%) diet over 8 weeks. On the 8th week, the rabbits were treated per os with 2 × 50 mg/kg daily doses of cicletanine, furosemide, or vehicle for 5 days (n = 5–6 in each groups). The cholesterol diet increased mean arterial blood pressure (MABP) from 86 ± 1 to 94 ± 2 mm Hg (p < 0.05). Cicletanine decreased MABP in atherosclerotic rabbits to 85 ± 1 mm Hg (p < 0.05), but it did not affect MABP in normal animals. Furosemide was without effect in both groups. In normal animals, NO content (assessed by electron spin resonance after in vivo spin trapping) in the aorta and the renal and carotid arteries was increased by cicletanine, and the drug increased cGMP in the renal artery as measured by radioimmunoassay. The cholesterol-enriched diet decreased both vascular NO and cGMP and increased aortic superoxide production assessed by lucigenin-enhanced chemiluminescence and serum nitrotyrosine determined by ELISA. In atherosclerotic animals, cicletanine increased NO and cGMP content in the aorta and the renal and carotid arteries and decreased aortic superoxide production and serum nitrotyrosine. Furosemide did not influence these parameters. We conclude that cicletanine lowers blood pressure in hypertensive/atherosclerotic rabbits. The antihypertensive effect of the drug in atherosclerosis may be based on its beneficial effects on the vascular NO-cGMP system and on the formation of reactive oxygen species.
Copyright © 2001 S. Karger AG, Basel
The furopyridine-derivative cicletanine is an antihypertensive drug with a vasorelaxant effect in addition to its diuretic property [1, 2]. The mechanism by which cicletanine decreases vascular tone is not clearly understood. The drug has been shown to inhibit low-Km Ca2+-calmodulin-dependent cyclic guanosine 3′:5′-monophosphate (cGMP) phosphodiesterase (PDE) and cGMP-selective PDE [3, 4] and to increase cGMP content of the heart . Inhibition of cGMP-PDE has been proposed to serve as an important mechanism of vasorelaxation promoted by cicletanine. cGMP was shown to modulate several potassium channels [6, 7]. Accordingly, cicletanine has also been shown to stimulate K+ efflux in human red blood cells  and to open ATP-sensitive K+ channels in human epigastric artery rings  and in isolated working rat hearts . Nitric oxide (NO) regulates potassium channels through both cGMP-dependent and independent pathways [7, 11] [for a review; see ref. 12], which suggests that the NO-cGMP-potassium channel pathway plays an important role in the regulation of vascular tone. The vascular effect of cicletanine was previously shown to involve endothelium-dependent vasodilation . We have recently shown that cicletanine increases vascular NO content in isolated rabbit aortic rings , which suggests that NO is involved in the antihypertensive effect of the drug.
Atherosclerosis is a well-known ‘NO-deficient state’ in the vasculature, which leads to sustained arterial hypertension [for reviews, see ref. 15, 16] and reduced cardiovascular tolerance to stress [for a review, see ref. 17]. Increasing evidence has accumulated in recent years showing that a high-cholesterol diet impairs NO-cGMP signaling in both endothelial and nonendothelial cells [18, 19, 20, 21], possibly due to enhanced formation of peroxynitrite, a reaction product of NO and superoxide [22, 23]. Therefore, antihypertensive drugs which restore normal NO-cGMP signaling in the vasculature might be especially advantageous in hypertension primarily induced by atherosclerosis/hyperlipidemia [for reviews, see ref. 24, 25].
In the present study we examined the effects of cicletanine on blood pressure and vascular NO and cGMP contents in the aorta and the renal and carotid arteries, superoxide production in the aorta, and serum nitrotyrosine level, a marker for peroxynitrite formation, in normotensive and in experimental atherosclerosis-induced hypertension in rabbits. The effect of cicletanine was compared to that of the structurally similar diuretic drug, furosemide. Here we show that cicletanine lowers blood pressure and increases vascular NO and cGMP content and attenuates the formation of superoxide and peroxynitrite in atherosclerotic rabbits.
The investigation conforms with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH publication No. 85–23, revised 1985).
The experiments were carried out with adult, male New Zealand white rabbits (3.0–3.5 kg) housed as described  and fed either standard laboratory chow or chow enriched with 1.5% cholesterol over a period of 8 weeks, which leads to the development of hyperlipidemia, atherosclerosis, and hypertension as described earlier [27, 28]. On the last week of the diet period, the animals were treated orally with 2 × 50 mg/kg daily doses of either cicletanine, furosemide, or their vehicle for 5 days (n = 5–6 in each group). A polyethylene cannula connected to a pressure transducer (B. Braun-Melsungen, Melsungen, Germany) was introduced into the distal third of the central ear artery for recording mean arterial blood pressure (MABP) in the conscious animals . Thereafter, blood samples were taken for the assay of serum lipids and nitrotyrosine. The animals were then sacrificed and vessel samples were taken for cGMP and superoxide assays. Vascular NO was measured in separate experiments (n = 5 in each group). Vascular NO and cGMP contents were studied in the thoracic aorta and the carotid and renal arteries. We have decided to study these vessels since they are important in the development of human hypertension induced by atherosclerosis. Superoxide production was studied in the aorta.
Serum cholesterol level was determined from blood samples taken prior to and after ingestion of cholesterol-enriched diet for 7 or 8 weeks. The samples were assayed by means of an automatic analyzer (Beckman Model 700, Chemistry System, Miami, Fla., USA) using Boehringer cholesterol kits (Ingelheim, Germany) as described .
Vascular NO content was determined in separate experiments (n = 5 in each group) by electron spin resonance (ESR) after in vivo spin trapping as described previously [29, 30, 31]. The spin-trap diethyldithio-carbamate (DETC, 200 mg/kg), 50 mg/kg FeSO4 and 200 mg/kg sodium citrate were slowly administered intravenously into an ear vein under ether anesthesia. DETC was dissolved in distilled water and was injected separately from FeSO4 and sodium citrate in a 0.5-ml volume to avoid precipitation of Fe2+-(DETC)3. FeSO4 and sodium citrate were dissolved in distilled water and the pH was set to 7.4 with NaOH and the volume of the solution was brought to 1 ml before injection. Five minutes after DETC and FeSO4/citrate treatment (i.e. after 5 min of in vivo spin trapping), the animals were sacrificed and the vessels were instantly isolated and washed for 30 s in oxygenated Krebs buffer (37°C, pH = 7.4) to eliminate blood. Tissue samples (100 mg) were transferred into quartz tubes and frozen in liquid nitrogen until assayed for the specific spectra of the NO-Fe2+-(DETC)2 complex. ESR spectra were recorded with a Bruker ECS106 spectrometer (Rheinstetten, Germany) operating at X band with 100 kHz of modulation frequency at a temperature of 160 K using 10 mW microwave power to avoid saturation. Scans were traced with 2.85 G modulation amplitude, 340 G sweep width, and 3356 G central field as described . After subtraction of the background signal, analysis of the NO signal intensity was performed with double integration, and NO content was expressed in arbitrary units. The detection limit of NO by this highly selective ESR method is approximately 0.05 nmol/g wet tissue weight .
Tissue cGMP content was determined in separate experiments as described [26, 30]. In brief, the vessel samples were instantly frozen and pulverized in liquid nitrogen at the end of the drug treatments. The samples were then homogenized in 10 vol of 6% (v/v) trichloroacetic acid in a mortar cooled to –70°C. After thawing, the samples were centrifuged at 15,000 g for 10 min at 4°C. Supernatant was extracted with 5 ml water- saturated ether over 5 min. The ether fraction was discarded and the extraction was then repeated five times. Samples were then evaporated under nitrogen and assayed for cGMP content using Amersham radioimmunoassay kits. Values were expressed as pmol/mg wet tissue weight.
Superoxide production in fresh aortic rings was assessed by lucigenin-enhanced chemiluminescence . Aortic rings (approximately 150 mg) were placed in 1 ml air-equilibrated Krebs-Henseleit solution containing 10 mmol/l HEPES-NaOH (pH 7.4) and 0.25 mmol/l lucigenin. Chemiluminescence was measured at room temperature in a liquid scintillation counter using a single active photomultiplier positioned in out-of-coincidence mode.
The concentration of free nitrotyrosine in serum samples, a marker for peroxynitrite generation , was assayed by ELISA. ELISA was conducted as described by us  and the supplier (Cayman Chemical, Ann Arbor, Mich., USA). Briefly, 210-μl serum samples were added to 4× volume 4°C ethanol, vortexed and spinned at 3,000 g for 10 min. Supernatant was evaporated under nitrogen and redissolved in 105 μl ultra-pure water. Samples were then incubated overnight with antinitrotyrosine rabbit IgG and nitrotyrosine acetylcholinesterase tracer in precoated (mouse anti-rabbit IgG) microplates followed by development with Ellman’s reagent for 60 min. Serum nitrotyrosine concentration is expressed as nmol/l.
The reaction of peroxynitrite with tyrosine induces nitration of its aromatic ring via intermediate formation of tyrosyl radicals, thereby yielding nitrotyrosine and dityrosine . It has been previously shown that addition of peroxynitrite to Krebs-Henseleit buffer containing 0.3 mML-tyrosine and 95% O2/5% CO2 at 37°C (pH 7.4) results in rapid (<1 min) formation of dityrosine which can be detected by both fluorescence spectroscopy and high-performance liquid chromatography (HPLC) with an excellent linear correlation (r2 = 0.98) between the HPLC and fluorescence-based determinations of dityrosine in the same sample [36, 37].
Reactions were performed in Krebs-Henseleit buffer containing 0.3 mML-tyrosine and 95% O2/5% CO2 at 37°C (pH 7.4). Stock solutions (100 μl) of cicletanine, furosemide, and urate (a well-known peroxynitrite scavenger which served as positive control [38, 39]) were added to microcentrifuge tubes to give a final concentration in 1,800 μl of 0 (vehicle), 30, 100, 300, or 1,000 μM, respectively. An aliquot of the buffer (1,700 μl) was then added to each tube. The tubes were immediately capped, vortexed, and placed into a 37°C waterbath for 2 min. Each tube was briefly opened and peroxynitrite (30 μl; peroxynitrite was prepared as described ) was added to give a final concentration of 3 or 30 μM, and then the tube was recapped. Tubes were immediately vortexed for 15 s, incubated for 10 min at 37°C, and then placed on ice. The entire volume of the reaction mixture was transferred into a plastic cuvette for fluorometric assay. In order to detect dityrosine, samples were excited at a wavelength of 320 nm and scanned between the emission wavelengths of 360 and 500 nm (scan time: 3 s) in a spectrofluorometer (Shimadzu, Model RF 5000).
The fluorescent signal amplitude was measured at 411.2 nm and any background fluorescent signal derived from Krebs-Henseleit buffer containing L-tyrosine was subtracted from this. Inhibition of dityrosine formation by a test compound was calculated as the percentage of the peak amplitude compared to that in the presence of the vehicle.
Data presented as means ± standard error (SEM) were analyzed by one-way analysis of variance (ANOVA). If a difference was established, a modified t test according to Bonferroni was applied between groups.
After 7 and 8 weeks of the cholesterol-enriched diet, serum total cholesterol increased from the pre-diet value of 1.6 ± 0.3 to 22.9 ± 3.1 and 21.8 ± 2.2 mmol/l in animals treated with the vehicle for cicletanine. Neither cicletanine nor furosemide influenced the diet-induced increase in serum cholesterol level and neither of them produced any effect on serum cholesterol level in time-matched controls.
Exposure to a cholesterol-enriched diet over 8 weeks significantly increased MABP in conscious rabbits (fig. 1). MABP did not change in the time-matched control group. Neither cicletanine nor furosemide influenced MABP in normal animals. In atherosclerotic animals, cicletanine significantly decreased MABP, whereas furosemide was without effect.
Fig. 1. Effect of cicletanine and furosemide on mean arterial blood pressure in normal and atherosclerotic rabbits. Results are means ± SEM, n = 6 in each group. * p < 0.05 atherosclerotic vs. normal; # p < 0.05 cicletanine vs. vehicle.
While a basal NO signal was detected in vessels from normal animals, NO was nondetectable in cholesterol-fed rabbits (table 1). In normal animals, cicletanine produced a significant increase in vascular NO content in all samples studied. Cicletanine-induced NO accumulation was the most pronounced in the renal artery. In atherosclerotic animals, cicletanine attenuated the decrease in NO content in all the three different vessels studied, however, NO levels remained significantly lower than that in normal animals. Cicletanine-induced increase in vascular NO level was the most significant in the renal artery. Furosemide did not change vascular NO signal.
Table 1. Effect of cicletanine and furosemide on vascular NO signal intensity and cGMP content in the aorta and the carotid and renal arteries of normal and atherosclerotic rabbits
In normal rabbits, cicletanine significantly increased vascular cGMP concentration in the renal artery, whereas it did not significantly influence cGMP content in the aorta and the carotid artery (table 1). cGMP content was significantly decreased by the 8-week cholesterol-enriched diet in samples from either the renal and carotid arteries or the aorta. In atherosclerotic rabbits, cicletanine significantly attenuated the decrease in cGMP level in all the vascular tissues studied. Furosemide did not change vascular cGMP in normal or atherosclerotic rabbits.
Superoxide production assessed by lucigenin-enhanced chemiluminescence was markedly enhanced in aortic rings obtained from cholesterol-fed rabbits. Cicletanine significantly reduced superoxide production while furosemide had no effect (fig. 2).
Fig. 2. Lucigenin-enhanced chemiluminescence in aortic rings obtained from normal and atherosclerotic rabbits. Results are means ± SEM, n = 5–6 in each group. * p < 0.05 atherosclerotic vs. normal.
High-cholesterol diet increased serum free nitrotyrosine content when compared to controls. Cicletanine decreased nitrotyrosine level in hypercholesterolemic rabbits while it did not significantly decrease this parameter in normal animals. Furosemide remained ineffective in both normal and hyperlipidemic rabbits (fig. 3).
Fig. 3. Serum nitrotyrosine concentration in normal and atherosclerotic rabbits. Results are means ± SEM, n = 6 in each group. * p < 0.05 atherosclerotic vs. normal.
When 3 μM (fig. 4) or 30 μM (data not shown) peroxynitrite was added to Krebs-Henseleit buffer containing 0.3 mML-tyrosine at physiological pH and temperature, neither cicletanine, nor furosemide inhibited peroxynitrite-induced dityrosine formation, however, it was concentration-dependently inhibited by the peroxynitrite scavenger urate, which served as positive control for these in vitro studies (fig. 4).
Fig. 4. Inhibitory effects of cicletanine, furosemide and the positive control urate on dityrosine formation induced by 3 μM peroxynitrite in Krebs-Henseleit buffer containing bicarbonate/CO2 and L-tyrosine at physiological temperature and pH. Results are means ± SEM, n ≥3.
The results show that cicletanine lowers MABP in conscious rabbits with experimental hypercholesterolemia/atherosclerosis without affecting normal blood pressure in healthy, normotensive conscious rabbits. The antihypertensive effect of cicletanine in atherosclerotic rabbits is accompanied with an increase in vascular NO and cGMP content and a decrease in vascular superoxide production and in serum nitrotyrosine level, a marker for peroxynitrite production. Furosemide, a diuretic drug which is structurally similar to cicletanine, did not affect blood pressure and vascular NO and cGMP contents, superoxide production, and serum nitrotyrosine in control or hypertensive rabbits. This suggests that the antihypertensive effect of cicletanine is not related to its diuretic effect, however, the drug improves the disturbed NO-cGMP signaling in atherosclerosis, thereby normalizing blood pressure in atherosclerotic/hypertensive animals.
It has been previously shown that cicletanine does not significantly affect normal blood pressure [40, 41, 42], however, its antihypertensive effect is well documented in spontaneously hypertensive rats and hypoxia-induced pulmonary hypertension [43, 44]. The drug also reverses vascular hyperactivity due to sympathetic overactivity  and pregnancy . Similarly to these results, cicletanine influenced blood pressure only in hypertensive but not in normotensive rabbits. This is the first demonstration that cicletanine is able to reduce blood pressure in hyperlipidemic/atherosclerotic animals.
Our present results showed that although the drug increased vascular NO level by approximately 100% in the aorta and the carotid and renal arteries, it increased vascular cGMP level significantly (approximately 50%) only in the renal artery in normal animals. Cicletanine produced a moderate decrease in aortic superoxide production and serum nitrotyrosine level which may explain the indirect elevation of vascular NO levels. These findings may show that activation of the normal NO-cGMP system does not lead to a hypotensive effect, possibly due to the normal function of other local regulatory mechanisms of the vascular tone in healthy normotensive rabbits.
Atherosclerosis is a well known ‘NO-deficient state’ in the vasculature which, among other factors, leads to hypertension [15, 16]. Numerous studies have demonstrated that a dysfunction in the release of EDRF/NO occurs already at an early stage of the development of atherosclerosis in animals and humans [45, 46]. A high-cholesterol diet impairs NO-cGMP signaling in both endothelial and non-endothelial cells [18, 19, 20, 21]. It has been shown that hypercholesterolemia leads to enhanced superoxide production in the vasculature [23, 47, 48]. This suggests increased formation of peroxynitrite in hyperlipidemia/atherosclerosis [22, 23, 49]. Accordingly, we have shown here that hyperlipidemia/atherosclerosis decreased vascular NO content to a level nondetectable by ESR and decreased cGMP by approximately 60% in the aorta and the carotid and renal arteries, vascular beds which have a high impact on the development of atherosclerosis-induced hypertension. We have also shown that both aortic superoxide formation and serum peroxynitrite concentration increased in cholesterol-fed rabbits. To the best of our knowledge, this is the first demonstration that serum free nitrotyrosine level, a marker for peroxynitrite formation, increases in hyperlipidemic/atherosclerotic animals.
Cicletanine restored NO and cGMP levels close to their normal level, decreased superoxide production and serum nitrotyrosine concentration, and lowered blood pressure in atherosclerotic rabbits. This suggests that restoration of impaired NO-cGMP signaling and a decreased in the formation of superoxide and peroxynitrite leads to normalization of blood pressure in atherosclerotic animals.
The mechanism by which cicletanine increases the ratio of NO and superoxide, thereby decreasing peroxynitrite formation and increasing cGMP level, is not precisely known. Our in vitro studies show that cicletanine does not directly scavenge peroxynitrite. Therefore, stimulation of NO synthase, inhibition of superoxide formation, or enhancement of detoxification mechanisms against peroxynitrite might play a role [22, 50, 51]. It should be noted that cicletanine may also increase the vascular cGMP level via its well-defined cGMP phosphodiesterase inhibitory effect as well as through increasing vascular NO content.
The diuretic drug furosemide failed to modify vascular NO, cGMP, superoxide, or serum nitrotyrosine and failed to decrease blood pressure in atherosclerotic animals. Since the hypertension model used in the present study [27, 28] is not characterized by volume expansion or sodium retention, the ineffectiveness of diuretics on blood pressure is not surprising. This also suggests that the antihypertensive effect of cicletanine is based on its possible direct vascular effects detailed above rather than its diuretic property in atherosclerotic conscious rabbits.
In summary, atherosclerosis induced by a high cholesterol diet leads to a decrease in vascular NO and cGMP content and an increase in superoxide and peroxynitrite production which results in the development of hypertension. Cicletanine increases vascular NO and cGMP content, decreases superoxide and peroxynitrite formation and lowers blood pressure in hypertensive/atherosclerotic conscious rabbits. This suggests that cicletanine might be advantageous in the treatment of hypertension induced by the disturbed NO-cGMP pathway due to atherosclerosis.
This work was supported by grants from IPSEN-Beaufour, Paris, France; Hungarian Scientific Research Found (T029843), Hungarian Ministries of Education (FKFP-0340/2000 and FKFP-0485/2000) and Health (ETT 51/2000), and Hungarian Space Research Office, Budapest, Hungary.
Dr. Péter Ferdinandy, Associate Professor
Cardiovascular Research Group, Department of Biochemistry
University of Szeged, Dóm tér 9, H–6720 Szeged (Hungary)
Tel. +36 62 545096, Fax +36 62 455097
E-Mail firstname.lastname@example.org, http://www.cardiovasc.com/
Received: Received: April 14, 1999
Accepted after revision: August 15, 2000
Number of Print Pages : 8
Number of Figures : 4, Number of Tables : 1, Number of References : 51
Journal of Vascular Research
Founded 1964 as Angiologica by M. Comèl and L. Laszt (1964–1973) continued as Blood Vessels by J.A. Bevan (1974–1991)
Vol. 38, No. 1, Year 2001 (Cover Date: January-February 2001)
Journal Editor: M.J. Mulvany, Aarhus
ISSN: 1018–1172 (print), 1423–0135 (Online)
For additional information: http://www.karger.com/journals/jvr