Nitroxyl (HNO) displays pharmacological and therapeutic actions distinct from those of its redox sibling nitric oxide (NO∙). It remains unclear, however, whether the vasoprotective actions of HNO are preserved in disease. The ability of the HNO donor isopropylamine NONOate (IPA/NO) to induce vasorelaxation, its susceptibility to tolerance development, and antiaggregatory actions were compared with those of a clinically used NO∙ donor, glyceryl trinitrate (GTN), in hypercholesterolemic mice. The vasorelaxant and antiaggregatory properties of IPA/NO and GTN were examined in isolated carotid arteries and washed platelets, respectively, from male C57BL/6J mice [wild-type (WT)] maintained on either a normal diet (WT-ND) or high fat diet (WT-HFD; 7 wk) as well as apolipoprotein E-deficient mice maintained on a HFD (ApoE−/−-HFD; 7 wk). In WT-ND mice, IPA/NO (0.1–30 μmol/l) induced concentration-dependent vasorelaxation and inhibition of collagen (30 μg/ml)-stimulated platelet aggregation, which was predominantly soluble guanylyl cyclase/cGMP dependent. Compared with WT-HFD mice, ApoE−/−-HFD mice displayed an increase in total plasma cholesterol levels (P < 0.001), vascular (P < 0.05) and platelet (P < 0.05) superoxide (O2·−) production, and reduced endogenous NO∙ bioavailability (P < 0.001). Vasorelaxant responses to both IPA/NO and GTN were preserved in hypercholesterolemia, whereas vascular tolerance developed to GTN (P < 0.001) but not to IPA/NO. The ability of IPA/NO (3 μmol/l) to inhibit platelet aggregation was preserved in hypercholesterolemia, whereas the actions of GTN (100 μmol/l) were abolished. In conclusion, the vasoprotective effects of IPA/NO were maintained in hypercholesterolemia and, thus, HNO donors may represent future novel treatments for vascular diseases.
- nitric oxide
- platelet aggregation
the therapeutic utility of the nitric oxide (NO∙)-soluble guanylyl cyclase (sGC) signaling pathway has long been recognized, with organic nitrates, such as glyceryl trinitrate (GTN), used in the treatment of angina pectoris and heart failure for over 100 yr. However, the clinical efficacy of such traditional nitrovasodilators is limited due to their susceptibility to tolerance development with continued used (4) and their impaired antiplatelet (3) and vasodilatory efficacy (26) during oxidative stress. Thus, there is a need for a new generation of novel NO∙-like drugs.
Nitroxyl (HNO), the one electron reduced and protonated congener of NO∙, displays distinct pharmacological actions and may have therapeutic advantages over its redox sibling NO∙ (2, 13). Unlike NO∙, HNO appears to be resistant to scavenging by superoxide (O2·−) (18, 20, 27), does not develop tolerance with continued use (10, 13), and serves as a positive cardiac inotrope (via thiol interaction) (22, 28). As such, HNO donors confer protection in the setting of acute experimental heart failure (22), where NO∙ donors have negligible effects.
In addition to its cardioprotective properties, HNO displays vasoprotective actions (2). Using HNO donors, such as Angeli's salt and isopropylamine NONOate (IPA/NO), HNO has been shown to serve as a potent vasodilator in vitro (7, 9, 11, 12) and in vivo (10, 22, 23). Like NO∙, HNO elicits vasorelaxation, at least in vitro, predominantly via the activation of sGC (8, 11) and subsequent generation of cGMP (9, 12). However, HNO can also signal through distinct mechanisms such as via the release of calcitonin gene-related peptide (CGRP) (8, 23) or activation of vascular voltage-dependent K+ (Kv) channels (7, 11) and ATP-sensitive K+ (KATP) channels (8). Moreover, Angeli's salt can also serve as an antiaggregatory agent, inhibiting human platelet aggregation in response to a variety of stimuli (1, 21), via a sGC-dependent mechanism (1). Thus, under nondisease conditions, HNO donors serve as potent vasodilators and antiaggregatory agents.
From a therapeutic perspective, it is important to determine if such vasoprotective actions of HNO are preserved in disease states associated with oxidative stress and compromised NO∙ signaling. Given that HNO, unlike NO∙, has been postulated to be resistant to scavenging by O2·−, (18, 20, 27), does not develop tolerance under physiological conditions (10, 12), can target distinct signaling pathways in the vasculature (Kv channels, KATP channels, and CGRP), and its bioavailability may be augmented in the face of disease-associated thiol depletion, it may be anticipated that HNO donor efficacy will be preserved and/or enhanced under pathophysiological conditions. Indeed, a handful of studies to date have indicated, albeit indirectly, that the vasoprotective properties of HNO may be sustained in disease. Thus, Angeli's salt has been shown to lower blood pressure in the setting of acute experimental heart failure (22) and inhibit the aggregation of platelets from patients with sickle cell disease (21). In addition, IPA/NO has been reported to limit neointimal hyperplasia in rat carotid arteries after balloon injury (29).
The present study sought to fully explore the concept of preserved vasoactivity to HNO in disease and compare, for the first time, the vasoprotective actions of HNO with the clinically used nitrovasodilator GTN under conditions associated with oxidative stress. Using a model of hypercholesterolemia in mice [apolipoprotein E-deficient (ApoE−/−) mice placed on a high-fat diet (HFD) for 7 wk], which is characterized by increased plasma cholesterol, elevated vascular O2·− generation, and reduced endogenous NO∙ bioavailablity (14), the vasorelaxant and antiaggregatory actions of the HNO donor IPA/NO and its susceptibility to tolerance development were compared with GTN.
This study was approved by the School of Biomedical Sciences Animal Ethics Committee of Monash University (Clayton, Victoria, Australia) and conformed with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.
Wild-type (WT) and ApoE−/− mice were obtained from the Animal Resources Centre (Canning Vale, WA, Australia). All mice studied were male and fully backcrossed to the C57BL/6J background. From 5 wk of age, C57BL6/J mice were either maintained on a normal diet of standard chow (WT-ND) or a HFD (WT-HFD; 21% fat and 0.15% cholesterol, Speciality Feeds) for 7 wk. ApoE−/− mice were also maintained on a HFD for 7 wk from 5 wk of age. For all experiments, mice were deeply anesthetized by isoflurane inhalation (Baxter Healthcare) before being euthanized by decapitation.
Measurement of total plasma cholesterol levels.
Blood from WT-HFD and ApoE−/−-HFD mice was collected from the inferior vena cava into heparinized tubes, and plasma was isolated via centrifugation (4,000 g, 4°C, 10 min). Plasma total cholesterol levels were then determined using a Roche MODULAR 917 (Roche Diagnostics, Castle Hill, NSW, Australia) enzymatic colorimetric array.
Internal and common carotid arteries from WT-HFD and ApoE−/−-HFD mice were examined for the presence of atherosclerotic lesions. Arteries were isolated, cut into 2- to 3-mm-long segments, mounted in an OCT Tissue-Tek mould, snap frozen in liquid nitrogen, and stored at −80°C. Arteries were sectioned (10 μm) and thaw mounted onto poly-l-lysine-coated microscope slides. Slides were fixed with paraformaldehyde (4%, 10 min) before being washed in 60% isopropyl alcohol (1.5 min) and left to dry. Sections were then stained with oil red O (0.5% in 60% isopropyl alcohol, 60 min). Excess stain was removed with 60% isopropyl alcohol (5 min) and distilled water (1 min), and sections were then counterstained with hematoxylin (25%, 2 min) as previously described (14, 17). Sections were viewed using an Olympus BX51 microscope (Olympus, Tokyo, Japan) equipped with a ×40 oil immersion lens. Images were digitized using a color OP70 Peltier cooled digital camera and captured with a data-acquisition system (Analysis LS Starter version 3.0, Olympus Soft Imaging Solutions, Munster, Germany).
L-012 (100 μmol/l)-enhanced chemiluminescence was used to measure basal and phorbol 12,13-dibutyrate (PDB; 10 μmol/l)-stimulated O2·− production by isolated common carotid artery segments or platelets from WT-HFD and ApoE−/−-HFD mice, as previously described (14).
Carotid arteries were excised, cleaned, and cut into segments of ∼2 mm in length. Background chemiluminescence signals were obtained over a 30-min period using a Plate Chameleon Luminescence Reader (Hidex) with separate wells of a white 96-well Opti-plate containing Krebs-HEPES [composed of (in mmol/l) 99 NaCl, 4.7 KCl, 1.2 MgSO4·7H2O, 1.0 KH2PO4, 19.6 Na2HCO3, 20 Na-HEPES, 11.1 d-glucose, 2.5 CaCl2, and 0.026 EDTA; pH 7.4] and L-012 (100 μmol/l). In semi-darkness, arteries were placed in separate wells, and L-012-enhanced chemiluminescence was measured every 2 min over a 30-min period. Experiments were performed in duplicate. O2·− production for each ring segment was calculated by subtracting the background chemiluminescence signal (in relative light units/s) from the signal in the presence of the artery and then normalized to dry tissue weight (in mg).
Platelets were isolated from whole blood, resuspended in Tyrode buffer [composed of (in mmol/l) 12 NaHCO3, 10 HEPES, 137 NaCl, 2.7 KCl, and 5.5 d-glucose with 0.026 EDTA; pH 7.2–7.4, 5 × 107 platelets/ml] and equilibrated for 30 min at 37°C as described in Platelet aggregation. Background counts were obtained over a 20-min period in a 96-well OptiPlate containing Krebs-HEPES and L-012 (100 μmol/l). In semi-darkness, platelets (5 × 108 platelets/ml) were placed in separate wells, and basal chemiluminescence was measured every 2 min over a 20-min period. PDB (10 μmol/l) was then added to the wells, and chemiluminescence was measured every 2 min over a 40-min period. Experiments were performed in quintuplicate. Basal chemiluminescence was calculated by subtracting the background chemiluminescence signal from the basal platelet chemiluminescence signal. PDB-stimulated chemiluminescence was calculated by subtracting the basal platelet chemiluminescence signal from the PDB-stimulated platelet chemiluminescence signal.
Vascular function experiments.
Common carotid arteries from WT-ND, WT-HFD, and ApoE−/−-HFD mice were isolated, cleaned of fat and connective tissue, and cut into ∼2-mm-long ring segments. Arteries were mounted in a Mulvany-style small vessel myograph (Danish Myo Technology, Skejbyparken, Denmark) for the measurement of isometric tension as previously described (11). Vessels were maintained in Krebs-bicarbonate solution [composed of (in mmol/l) 118 NaCl, 4.5 KCl, 0.5 MgSO4, 1.0 KH2PO4, 25 NaHCO3, 11.1 glucose, 2.5 CaCl2 and 0.026 EDTA; pH 7.4] at 37°C and bubbled with carbogen (95% O2-5% CO2). After a 30-min equilibration period, arteries were stretched to a resting tension of 5 mN before being maximally contracted with U-46619 (1 μmol/l; Fmax). To examine vasorelaxation responses, arteries were precontracted to ∼50% Fmax with titrated concentrations of U-46619 (3–30 nmol/l). In carotid arteries from WT-ND mice, cumulative concentration-response curves to the HNO donors IPA/NO (1 nmol/l–30 μmol/l) and Angeli's salt (1 nmol/l–30 μmol/l) were constructed in the absence or presence of the HNO scavenger l-cysteine (3 or 10 mmol/l, 5-min preincubation), the NO∙ scavenger carboxy-[2-(4-carboxyphenyl)-4,4,5,5-tetramethyl-imidazoline-1-oxy-3-oxide] (carboxy-PTIO; 200 μmol/l, 10-min preincubation), and the sGC inhibitor 1H-[1,2,4]oxadiazolo-[4,3-a]quinoxaline-1-one (ODQ; 10 μmol/l, 30-min preincubation). Maximal relaxation was achieved using diethylamine NONOate (DEA/NO; 1 μmol/l) or levcromakalim (10 μmol/l) at the conclusion of each concentration-response curve. Only one concentration-response curve to any vasodilator was obtained for each vessel.
To assess the effect of hypercholesterolemia on vasorelaxant responses, cumulative concentration-response curves to IPA/NO (1 nmol/l–30 μmol/l), the NO∙ donor GTN (1 nmol/l–30 μmol/l), and the endothelium-independent vasodilator papaverine (10 nmol/l–30 μmol/l) were constructed in external common carotid arteries from WT-HFD and ApoE−/−-HFD mice as described above.
To investigate the potential development of tolerance to GTN or IPA/NO in hypercholesterolemia, carotid arteries from WT-HFD and ApoE−/−-HFD mice were incubated in either the absence or presence of GTN (30 μmol/l) or IPA/NO (30 μmol/l) for a period of 60 min (12). Vessels were then washed thoroughly every 15 min for 60 min. Subsequently, arteries were precontracted to ∼50% Fmax with titrated concentrations of U-46619 (3–30 nmol/l) before concentration-response curves to GTN or IPA/NO were obtained.
Contractile responses to the NO synthase inhibitor Nω-nitro-l-arginine methyl ester (l-NAME; 100 μmol/l) were measured as an indicator of NO bioavailability in external common carotid artery segments from WT-HFD and ApoE−/−-HFD mice. Vessels were precontracted to ∼20–30% Fmax with titrated concentrations of U-46619 (4–20 nmol/l). Once the U-46619 contraction was stable, l-NAME (100 μmol/l) was added, and the contractile response was recorded once it reached a plateau.
Platelets were isolated from whole blood collected from the inferior vena cava of WT-ND, WT-HFD, and ApoE−/−-HFD mice and stimulated for aggregation experiments as previously described (6). To prevent coagulation, blood was treated with a combination of low-molecular-weight heparin (clexane; 400 U/ml) and acid citrate dextrose buffer [composed of (in mmol/l) 85 trisodium citrate, 72.9 citric acid, and 110 d-glucose] immediately after collection. Blood was then washed with platelet wash buffer [composed of (in mmol/l) 4.3 K2HPO4, 4.3 Na2HPO4, 24.3 NaH2PO4, 113 NaCl, and 5.5 d-glucose with 10% BSA and 20 U/ml clexane; pH 6.5] and centrifuged at 200 g (37°C, 2 min). This was performed three times, and each time the platelet rich plasma was removed. After centrifugation of the pooled platelet-rich plasma (2,000 g, 1 min), the pellet was resuspended in Tyrode buffer to give a concentration of 5 × 107 platelets/ml. Platelets were added to siliconized cuvettes containing fibrinogen (2 mg/ml) and allowed to equilibrate at 37°C for 30 min before stimulation with collagen (30 μg/ml). Platelet aggregation was measured using a four-chamber turbidometric platelet aggregometer (AggRAM, Helena Laboratories) as a change in light transmission over 30 min under continuous stirring (600 rpm, 37°C).
To characterise the antiaggregatory actions of IPA/NO, platelets from WT-ND mice were treated with either vehicle (1 mmol/l NaOH, 2 min) or IPA/NO (0.1–3 μmol/l, 2 min) before stimulation with collagen (30 μg/ml). The effects of ODQ (10 μmol/l, 30 min) and the cGMP-dependent protein kinase (cGK) inhibitor Rp-8-pCPT-cGMPS (200 μmol/l, 10 min) on the ability of IPA/NO (1 μmol/l) to inhibit collagen-stimulated platelet aggregation were also examined.
To compare the antiaggregatory actions of HNO and NO∙ in hypercholesterolemia, platelets from WT-HFD and ApoE−/−-HFD mice were preincubated in the presence of vehicle (1 mmol/l NaOH or 0.5% ethanol, 2 min), IPA/NO (3 μmol/l, 2 min), or GTN (100 μmol/l, 2 min) before stimulation with collagen (30 μg/ml).
Data and statistical analysis.
Vasorelaxation responses were expressed as the percent reversal of U-46619 precontraction. Contractile responses to l-NAME were expressed as a percentage of the maximum response to U-46619 (1 μmol/l; Fmax). Individual relaxation curves were fitted to a sigmoidal logistical equation (GraphPad Prism, version 5.0) to provide an estimate of the concentration of agonist causing a 50% relaxation (pEC50 value; in −log mol/l). Differences between mean pEC50 and maximum relaxation values were tested using either a Student's unpaired t-test or one-way ANOVA with either a Dunnett's or Bonferroni post hoc test. Where pEC50 values could not be obtained, concentration-response curves were compared by means of two-way ANOVA with a Tukey post hoc test (Sigma Stat 3.5).
Platelet aggregation responses were expressed as either total aggregation area under the curve (AUC; in arbitrary units) in percent increments of light transmission over the 30-min period after the addition of collagen or time to half-maximal aggregation (in s). Statistical comparisons were performed using one-way ANOVA with a Bonferroni post hoc test (GraphPad Prism, version 5.0). All results are expressed as means ± SE, with the number of animals denoted by n. P values of <0.05 were considered statistically significant.
Drugs and their sources.
Drugs and their sources were as follows: clexane (enoxaparin sodium, Aventos), collagen (bovine, Helena Laboratories), GTN (Mayne), aquatex, isopentane, isopropyl alcohol, and paraformaldehyde (Merck), IPA/NO (a kind gift from K. Miranda, University of Arizona), l-012 (Wako), levcromakalim (Tocris Bioscience), Angeli's salt (sodium trioxodinitrate), carboxy-PTIO potassium salt, ODQ, and U-46619 (Sapphire Bioscience), and DEA/NO, fibrinogen-bovine serum type I, l-cysteine hydrochloride, l-NAME, oil red O, papaverine, and Rp-8-pCPT-cGMPS (Sigma). Clexane was prepared at 400 U/ml in saline (100%). Fibrinogen was prepared at 20 mg/ml in 0.9% saline, and all subsequent dilutions were in distilled water. GTN was prepared as a 10 mmol/l stock in 50% ethanol, and all subsequent dilutions were in distilled water. Angeli's salt, DEA/NO, and IPA/NO were prepared at 10 mmol/l in 0.01 mol/l NaOH, and all subsequent dilutions were in 0.01 mol/l NaOH. L-012 was prepared at 100 mmol/l in DMSO, and all subsequent dilutions were in Krebs-HEPES (pH 7.4) solution. Levcromakalim was prepared at 10 mmol/l in 100% methanol. ODQ was prepared at 10 mmol/l in 100% ethanol. Oil red O was prepared as a 0.5% stock solution in 100% isopropyl alcohol and then diluted to a 60% isopropyl alcohol working solution. All other drugs were dissolved and diluted in distilled water.
Vasorelaxant responses to HNO donors.
The HNO donors Angeli's salt and IPA/NO elicited concentration-dependent relaxation in endothelium-intact carotid arteries from WT-ND mice that were of similar potency and efficacy (Fig. 1, A and B, and Table 1). Vasorelaxation to these donors was unchanged in the presence of the NO∙ scavenger carboxy-PTIO, but their potencies were decreased by up to 10-fold (P < 0.001) by the HNO scavenger l-cysteine (3 mmol/l; Fig. 1, A and B, and Table 1). Increasing the concentration of l-cysteine to 10 mmol/l further attenuated IPA/NO-mediated vasorelaxation such that the response to 10 μmol/l was inhibited by 28.3 ± 7.7% (n = 8) versus 42.4 ± 6.5% (n = 5) in the presence of 3 and 10 mmol/l l-cysteine, respectively. In contrast, vasorelaxation to the NO∙ donor GTN was unchanged in the presence of l-cysteine (3 mmol/l; data not shown). The sGC inhibitor ODQ markedly attenuated vasorelaxation to both Angeli's salt and IPA/NO (P < 0.001; Fig. 1, A and B, and Table 1). Given that Angeli's salt and IPA/NO displayed similar vasorelaxant potency, and because IPA/NO (unlike Angeli's salt) does not decompose to generate nitrite (13, 19), all subsequent experiments used IPA/NO.
Antiaggregatory actions of IPA/NO.
IPA/NO caused concentration-dependent inhibition of collagen-stimulated aggregation of washed platelets from WT-ND mice (Fig. 2A), with concentrations of 1 and 3 μmol/l IPA/NO decreasing total aggregation by ∼50% (P < 0.05; Fig. 2B). This was accompanied by a delay in the onset of platelet aggregation, which reached statistical significance at 3 μmol/l IPA/NO (P < 0.05; Fig. 2C).
Mechanisms underlying the antiaggregatory actions of IPA/NO.
ODQ reversed the ability of IPA/NO (1 μmol/l) to inhibit collagen-stimulated aggregation of platelets from WT-ND mice (P < 0.01; Fig. 3, A and B). Similarly, the cGK inhibitor Rp-8-pCPT-cGMPS partially reversed (P < 0.05) the antiaggregatory effects of IPA/NO (1 μmol/l; Fig. 3, C and D). In platelets from a subset of mice, we confirmed that neither ODQ (total aggregation AUC: control 62,563 ± 4,178 vs. ODQ 50,353 ± 6,250, n = 6) nor Rp-8-pCPT-cGMPS (total aggregation AUC: control 47,468 ± 3,263 vs Rp-8-pCPT-cGMPS 48,728 ± 6,591, n = 3) alone significantly altered collagen-stimulated aggregation.
Plasma cholesterol, vascular morphology, vessel and platelet function in WT-HFD and ApoE−/−-HFD mice.
Total plasma cholesterol levels were approximately fourfold higher (P < 0.001) in ApoE−/−-HFD than in WT-HFD mice (Table 2). Atherosclerotic lesions were apparent in the internal, but not common, carotid arteries of ApoE−/−-HFD mice (Fig. 4, B and D). No lesions were detected in WT-HFD carotid arteries (Fig. 4, A and C). Despite the absence of lesions, basal O2·− production by common carotid arteries from ApoE−/−-HFD mice was ∼60% greater than levels generated by arteries from WT-HFD mice (P < 0.05; Table 2). Contractile responses of common carotid arteries from ApoE−/−-HFD mice to l-NAME (100 μmol/l) were markedly attenuated versus WT-HFD mice (P < 0.001; Table 2), indicative of decreased endogenous NO bioavailability. Basal O2·− production by platelets from WT-HFD and ApoE−/−-HFD mice did not differ; however, PDB-stimulated O2·− production was ∼80% greater in ApoE−/−-HFD mice (P < 0.05; Table 2). Whereas total aggregation to collagen was similar between platelets from WT-HFD and ApoE−/−-HFD mice, platelets from ApoE−/−-HFD mice aggregated more rapidly (P < 0.05; Table 2).
Vasorelaxant responses to IPA/NO in ApoE−/−-HFD mice.
IPA/NO-induced vasorelaxation was preserved in carotid arteries from ApoE−/−-HFD mice, with a small but significant 2.5-fold (P < 0.01) increase in sensitivity compared with WT-HFD arteries (Fig. 5A). Relaxant responses to the NO∙ donor GTN and the NO-independent vasodilator papaverine were similar in carotid arteries from ApoE−/−-HFD and WT-HFD mice (Fig. 5, B and C).
Pretreatment of carotid arteries from WT-HFD and ApoE−/−-HFD mice with 30 μmo/l GTN for 60 min caused up to an ∼16-fold (P < 0.001) decrease in sensitivity to the subsequent application of GTN (Fig. 6, B and D). In contrast, pretreatment of carotid arteries from WT-HFD and ApoE−/−-HFD mice with IPA/NO (30 μmol/l, 60 min) had no effect on their subsequent vasorelaxation to IPA/NO (Fig. 6, A and C).
Antiaggregatory actions of IPA/NO in ApoE−/−-HFD platelets.
IPA/NO (3 μmo/l) inhibited collagen-induced aggregation in ApoE−/−-HFD mice (32 ± 7% inhibition) to a similar extent as in WT-HFD platelets (39 ± 14% inhibition; Fig. 7A). In addition, IPA/NO caused a comparable increase in the time to half-maximal aggregation in ApoE−/−-HFD and WT-HFD platelets (Fig. 7B). In contrast, the antiaggregatory actions of GTN (100 μmol/l, 40 ± 10% inhibition) were abolished in platelets from ApoE−/−-HFD mice (Fig. 7, C and D).
This study demonstrated, for the first time, that the vasoprotective actions of HNO are equally effective in nondisease and disease conditions associated with elevated vascular and platelet O2·− generation and reduced endogenous NO∙ bioavailability. Specifically, we found that the ability of the HNO donor IPA/NO to induce vasorelaxation, exhibit resistance to vascular tolerance development, and inhibit platelet aggregation was preserved in hypercholesterolemic mice. Furthermore, these findings contrast with those observed with the clinically used NO∙ donor GTN, which developed vascular tolerance and exhibited an impaired ability to inhibit platelet aggregation in the setting of hypercholesterolemia. Together, these findings suggest that HNO donors may offer a therapeutic alternative to the currently used organic nitrates in the treatment of vascular disorders.
Before we examined the vasoprotective actions of HNO in hypercholesterolemia, it was important to fully characterise the vasorelaxant and antiaggregatory effects of this nitrogen oxide under nondisease conditions. This was achieved using the HNO donors Angeli's salt and IPA/NO, both of which caused relaxation of mouse common carotid arteries with a similar potency. Using the well-established HNO and NO∙ scavengers l-cysteine (11, 12) and carboxy-PTIO (11, 12), respectively, we confirmed that HNO, and not NO∙, mediates vasorelaxation in response to both Angeli's salt and IPA/NO. Moreover, in accordance with previous studies (7, 8, 11, 12) in large conduit and small resistance arteries, the sGC inhibitor ODQ markedly attenuated vasorelaxant responses to both Angeli's salt and IPA/NO in carotid arteries from nondiseased mice, indicating a predominant role for the sGC/cGMP pathway in HNO-mediated vasorelaxation. While both IPA/NO and Angeli's salt spontaneously decompose at physiological pH to donate HNO with similar half-lives (∼2.5 min) (13, 19), subsequent experiments used IPA/NO to avoid the potentially confounding effect of nitrite, which is concomitantly released with HNO by Angeli's salt (13).
In addition to characterizing the vasodilator capacity of HNO, we demonstrated for the first time in mice that IPA/NO induces concentration-dependent inhibition of collagen-induced platelet aggregation. These findings are consistent with studies in human and mouse platelets, where Angeli's salt was shown to exert antiaggregatory properties in response to arachidonic acid, ADP, collagen, and thrombin (1, 21) as well as collagen (5), respectively. Furthermore, in accordance with these studies (1, 5), we found that HNO primarily inhibits platelet aggregation via a sGC/cGMP-dependent signaling pathway with the antiaggregatory effects of IPA/NO reversed by ODQ and attenuated, in part, by the cGK inhibitor Rp-8-pCPT-cGMPS.
While numerous studies, including the present one, have shown that HNO has vasoprotective actions under nondisease conditions, little is known about its efficacy during disease. Studies in support of preserved vascular actions of HNO under pathophysiological conditions include the finding that Angeli's salt-mediated inhibition of ADP-stimulated platelet aggregation was sustained in patients with sickle cell disease (21). IPA/NO has also been shown to limit neointimal hyperplasia in rats (29), and Paolocci and colleagues (22) have indirectly shown that the hemodynamic actions of HNO are sustained in acute experimental heart failure. However, it remains unclear whether HNO donors offer advantages over traditional nitrovasodilators, such as GTN, in the treatment of vascular diseases where there is increased generation of ROS such as O2·− and reduced NO∙ signaling. These are clinically relevant questions as the therapeutic efficacy of GTN is limited due to its propensity to induce tolerance to subsequent drug exposures (4), resistance of platelets to its antiaggregatory effects (3), and its reduced efficacy under conditions of elevated ROS generation (26) due to scavenging of NO∙ by O2·− and dysfunction of sGC itself (16).
To begin to address these issues, this study used ApoE−/− mice maintained on a HFD for 7 wk as a model of hypercholesterolemia and early stage of disease. In agreement with previous studies (14, 15, 17), we found that these ApoE−/−-HFD mice displayed elevated plasma cholesterol levels, increased vascular O2·− production, reduced vascular NO∙ bioavailability, and atherosclerotic lesions in the internal carotid artery.
In addition to the vascular changes, we found that platelets from ApoE−/−-HFD mice were able to aggregate more rapidly compared with those from WT-HFD. Moreover, consistent with findings of elevated O2·− generation in hypercholesterolemic patients (24, 25), O2·− production in response to PDB, an activator of PKC, was elevated in platelets from ApoE−/−-HFD mice. As PKC can activate the ROS-generating enzyme NADPH oxidase, we hypothesize that increased activity of NADPH oxidase may account for the greater O2·− production in hypercholesterolemic mice. Taken together, these findings indicate that ApoE−/− mice maintained on a HFD for 7 wk display the hallmark characteristics associated with hypercholesterolemia.
Interestingly, in the setting of hypercholesterolemia, the vasodilator capacity of HNO was preserved with a small but significant increase in vasorelaxant potency to IPA/NO observed in carotid arteries from ApoE−/−-HFD compared with normocholesterolemic WT-HFD mice. Similarly, vasorelaxant responses to the NO∙ donor GTN and the sGC-independent vasodilator papaverine were unchanged in carotid arteries from ApoE−/−-HFD versus normocholesterolemic mice. Our finding that vasorelaxant responses to GTN were maintained in hypercholesterolemia is not surprising given the relatively early stage of vascular disease. Indeed, oxidative stress and dysfunction of the NO-sGC-cGMP signaling pathway are more prominent in the later stages of hypercholesterolemia/atherosclerosis (16). Thus, future studies using ApoE−/− mice maintained on a HFD for a more prolonged period of time may unmask differential vasodilator effects of HNO versus NO∙ donors.
Importantly, a further limitation of nitrovasodilators, such as GTN, is their susceptibility to tolerance development with continued use. Tolerance is a multifactorial process that may arise as a consequence of neurohumoral counterregulation, reduced mitochondrial biotransformation of organic nitrates, impaired sGC function, increased activity of cGMP-degrading phosphodiesterases, or increased O2·− generation (4). We have previously shown that, in contrast to GTN, the HNO donor Angeli's salt does not develop tolerance either in vitro (12) or when chronically administered in vivo (10). This resistance to tolerance development is likely due to the ability of Angeli's salt to spontaneously donate HNO, whereas GTN requires biotransformation to generate NO∙. Here, we extended these observations, reporting for the first time that in a cardiovascular disease associated with increased O2·− generation, HNO donors, unlike GTN, are resistant to tolerance development in both normocholesterolemic and ApoE−/−-HFD mice. Thus, preservation of the vasodilator efficacy of HNO, together with its lack of tolerance in disease, represents significant advantages over GTN.
Similarly, we found that the antiaggregatory actions of HNO are sustained in hypercholesterolemia with IPA/NO inhibiting collagen-induced platelet aggregation to a similar extent in ApoE−/−-HFD versus normocholesterolemic mice. These findings are consistent with a previous report (21) that showed that the antiaggregatory actions of Angeli's salt were preserved in platelets from patients with sickle cell disease. In contrast to HNO, GTN failed to inhibit platelet aggregation in ApoE−/−-HFD mice. A loss of responsiveness to GTN has also been reported in platelets from patients with stable angina pectoris, acute coronary syndromes, and obesity (3), and this is commonly referred to as NO∙ resistance. This phenomenon of NO∙ resistance at the level of platelet aggregation has been attributed to scavenging of NO∙ by O2·− as well as to dysfunction of sGC (3). As discussed, we found that O2·− generation in response to PDB was elevated in platelets from ApoE−/− mice. Thus, we predict that the hyporesponsiveness to GTN in hypercholesterolemic mice may have arisen due to scavenging of NO∙ by O2·−. In addition, an elevation of ROS may lead to a reduced capacity of platelets to biotransform GTN to NO∙ (4). Based on our preliminary evidence showing that IPA/NO does not scavenge O2·− in a xanthine/xanthine oxidase cell-free assay (data not shown), we anticipate that the resistance of HNO to scavenging by O2·− may underlie its preserved antiaggregatory actions in hypercholesterolemia.
In conclusion, this study has demonstrated, for the first time, that the vasodilator and antiaggregatory capacities of HNO are preserved in the early stages of vascular disease in which increased O2·− generation and compromised NO∙ signaling is evident. Importantly, HNO donors may circumvent the vascular tolerance development and platelet hyporesponsiveness to NO∙ observed with clinically used nitrovasodilators such as GTN. Thus, the novel cardioprotective actions of HNO (22), coupled with its vasoprotective properties (2), may offer significant advantages over traditional nitrovasodilators. Moreover, we predict that the vasoprotective actions of HNO, as demonstrated in this study, will be sustained in more advanced models of cardiovascular disease, a concept that is currently under investigation.
This work was supported by National Health and Medical Research Council of Australia (NHMRC) Grant 606556. M. L. Bullen is the recipient of an Australian Postgraduate Award. A. A. Miller is the recipient of a Biomedical Career Development Award from the NHMRC. C. G. Sobey and G. R. Drummond are Senior Research Fellows of the NHMRC.
B. K. Kemp-Harper has a collaborative research agreement with Cardioxyl Pharmaceuticals, a company that develops novel HNO donors, none of which were used in the present study.
The authors thank Henry Diep (Department of Pharmacology, Monash University) for the assistance with the collection of plasma and assay of plasma cholesterol levels and Dr. Katrina Miranda (Department of Chemistry, University of Arizona) for the synthesis of IPA/NO.
- Copyright © 2011 the American Physiological Society