Cardiomyocyte-Restricted Overexpression of Endothelial Nitric Oxide Synthase (NOS3) Attenuates β-Adrenergic Stimulation and Reinforces Vagal Inhibition of Cardiac Contraction
Background— In the heart, nitric oxide synthases (NOS) modulate cardiac contraction in an isoform-specific manner, which is critically dependent on their cellular and subcellular localization. Defective NO production by NOS3 (endothelial NOS [eNOS]) in the failing heart may precipitate cardiac failure, which could be reversed by overexpression of NOS3 in the myocardium.
Methods and Results— We studied the influence of NOS3 in relation to its subcellular localization on the function of cardiomyocytes isolated from transgenic mice overexpressing NOS3 under the α-myosin heavy chain promoter (NOS3-TG). Immunoblot analysis demonstrated moderate (5-fold) NOS3 overexpression in cardiomyocytes from NOS3-TG heterozygotes. Caveolar localization of transgenic eNOS was demonstrated by immunofluorescence, coimmunoprecipitation with caveolin-3, sucrose gradient fractionation, and immunogold staining revealed by electron microscopy. Compared with wild-type littermate, contractility of NOS3-TG cardiomyocytes analyzed by videomicroscopy revealed a lower incidence of spontaneous arrhythmic contractions (n=32, P<0.001); an attenuation of the β-adrenergic positive inotropic response (isoproterenol, 10−7 mol/L: 62.1±7.8% versus 90.8±8.0% of maximum Ca2+ response; n=10 to 17; P<0.05); a potentiation of the muscarinic negative chronotropic response (carbamylcholine, 3.10−8 mol/L: −63.9±14% versus −27.7±5.6% of basal rate; n=8 to 10; P<0.05), confirmed by telemetry in vivo; and an attenuation of the accentuated antagonism of β-adrenergically stimulated contraction (−14.6±1.5% versus −3.5±1.5; n=7 to 11; P<0.05). Cardiomyocyte NOS inhibition reversed all 4 effects (P<0.05).
Conclusions— Moderate overexpression of NOS3, targeted to caveolae in murine cardiomyocytes, potentiates the postsynaptic muscarinic response and attenuates the effect of high concentrations of catecholamines. Cardiomyocyte NOS3 may represent a promising therapeutic target to restore the sympathovagal balance and protect the heart against arrhythmia.
Received March 15, 2004; de novo received May 14, 2004; revision received June 10, 2004; accepted June 17, 2004.
Nitric oxide (produced by the endothelial isoform of nitric oxide synthase, eNOS, encoded by the NOS3 gene) is the prototypical endothelium-derived relaxing factor in vessels. As such, NO modulates cardiac function through its regulation of hemodynamics and coronary reserve, but it also affects cardiac muscle function directly. This may result from paracrine NO produced by endothelial cells surrounding the myocytes or from autocrine regulation, because eNOS and neuronal NOS (nNOS) are constitutively expressed in cardiomyocytes themselves.1–3 The influence of endothelial versus cardiomyocyte NO on contractility has been difficult to delineate in whole-organ experiments because NOS inhibitors and genetic deletion of NOS isoforms would affect both cell types, obscuring the specific contribution of cardiomyocyte NOS in the overall phenotype. Also, nonconditional genetic deletions produce chronic adaptive mechanisms (eg, cardiac hypertrophy), which have confounded some interpretations, even in single myocyte experiments (reviewed in Massion et al2,4). Despite conflicting results based on these approaches, cardiomyocyte nNOS5 and eNOS1,4 are thought to regulate several aspects of excitation-contraction coupling, as well as the response of cardiac muscle to autonomic regulation. This was confirmed in a recent “knock-in” experiment6 in which adenoviral transfection of NOS3 targeted to cardiomyocytes in NOS3−/− mice restored most aspects of NOS-dependent autocrine regulation previously described in isolated cells.1,7 This observation, however, was exclusively obtained in vivo, which leaves potential doubts about indirect, paracrine effects (eg, from overexpressed cardiomyocyte eNOS on the surrounding vasculature or nerves).
Importantly, the influence of each cardiomyocyte NOS depends on its relative abundance and subcellular localization,8 as well as the stimulus considered.9 In particular, the functional coupling of eNOS to specific receptors critically depends on their colocalization in cardiomyocyte membrane caveolae,10,11 where many signaling molecules are concentrated. Such colocalization and functional interaction was shown to be altered in the aging, hypertrophic, and ischemic heart, with resultant impairment of eNOS activity that could contribute to myocardial dysfunction.12,13 Accordingly, any attempt to restore eNOS signaling and cardiac function with heterologous eNOS expression should take into account the proper subcellular targeting of the enzyme.
To further resolve the functional impact of eNOS on cardiomyocyte contractility, we therefore chose an alternative approach based on 3 principles: (1) cardiomyocyte-restricted eNOS expression under the α-myosin heavy chain (MHC) promoter; (2) characterization of the phenotype in isolated myocytes from 1- to 2-day-old neonatal mice, at a time when the activity of the α-MHC promoter begins to drive the transgene expression,14 thereby avoiding chronic adaptive phenotypic changes in vivo; and (3) thorough analysis of eNOS subcellular localization. In purified cardiomyocytes, we found that the transgene, targeted to caveolar fractions, modulated the response to β-adrenergic and muscarinic cholinergic stimulation, 2 key aspects of cardiac physiological regulation. These results prefigure the potential benefit of heterologous expression of eNOS through restoration of the sympathovagal balance in the diseased heart.
Isolation and Culture of Ventricular Myocytes
Ventricular myocytes, obtained by enzymatic digestion of neonatal hearts and purified to homogeneity by Percoll gradient sedimentation and differential attachment,7 were cultured in 10% serum-containing medium and analyzed in serum-free balanced salt solution containing (in mmol/L) calcium 1.2, Na-HEPES 10 (final pH 7.5), and l-arginine 0.5. Heterozygote transgenic mice overexpressing NOS3 under the α-myosin heavy chain promoter (NOS3-TG) and wild-type littermate were generated as described previously.15
Cardiomyocyte extracts were prepared in hypotonic solution (in mmol/L: Tris-HCl 50, EGTA 0.1, β-mercaptoethanol 2), with a protease inhibitor cocktail, diluted in CHAPS buffer.10 Preclearing was performed twice with uncoated protein G-sepharose beads, and the supernatant was incubated with rabbit polyclonal anti-eNOS antibodies (Affinity BioReagents). Immune complexes precipitated with protein G-sepharose beads (IP) and 50 μL of supernatant (SNT) were added to Laemmli solution and processed for SDS-PAGE. eNOS and caveolin-3 signals were revealed with monoclonal anti-eNOS (1:1,500; Transduction/BD), polyclonal anti-troponin I (1:2,500; Santa Cruz), and anti-caveolin-3 (1:15,000; Transduction/BD) antibodies.
Subcellular Fractionation by Isopycnic Ultracentrifugation
Extracts were prepared in 1 mL of Na2CO3 0.5 mol/L (pH 11) and added to 1 mL of 80% sucrose (in NaCl, 0.15 mol/L; MES, 25 mmol/L, pH 6.5; final sucrose concentration, 40%).10 A 2-step gradient was loaded on the top of the sample (4 mL of 35% sucrose overlaid with 4 mL of 5% sucrose). After ultracentrifugation at 38 000 rpm (4°C, 16 to 18 hours), fractions were collected from top to bottom, precipitated with trichloroacetic acid (7.2%), washed, and resuspended in Laemmli for discontinuous (8% to 12% SDS-polyacrylamide) electrophoresis and Western blotting.
Immunofluorescence of Caveolin-3 and eNOS
Neonatal cardiomyocytes fixed with 4% paraformaldehyde on Laboratory-Tek slides were permeabilized and incubated with goat anti-mouse caveolin-3 (1:200, Santa Cruz) and mouse anti-human eNOS (1:25, Transduction/BD) antibodies, washed, and incubated with TRITC-labeled rabbit anti-goat (1:100) and FITC-labeled rat anti-mouse (1:40) antibodies. Five-micrometer frozen sections from whole adult hearts were processed similarly. Images were analyzed with an Olympus fluorescence microscope coupled to a digital camera under identical acquisition parameters (eg, time of light exposure).
Small blocks (0.5 mm3) from 4% paraformaldehyde-fixed hearts were dehydrated in alcohol and embedded into soft grade LR White resin (Sigma). Rehydrated ultrathin sections collected on nickel grids were pretreated with glycine 0.05mol/L, blocked with PBS-BSA 3%-milk 2%, and incubated with any of 3 primary polyclonal rabbit anti-eNOS antibodies (1:50, Affinity Bioreagents, CalBiochem or Transduction; omitted in negative controls). Sections were incubated with secondary biotinylated sheep anti-rabbit antibodies (1:200), then with 10-nm gold particles conjugated to streptavidin (1:50; 90 minutes), counterstained with uranyl acetate and lead citrate, and examined with a Zeiss EM109 at 12 000 times original magnification. Particle density was determined in plasma membrane and T-tubules (along Z-lines that included I-bands) or in cytosol (all remaining myocyte surface) by counting gold particles per surface area (determined with image J software) in each location; the results were normalized by the particle density in the myocyte nucleus (taken as background). All results are means from 3 to 4 determinations from random fields from different sections.
Contractility Analysis by Video-Motion Microscopy
Attached cardiomyocytes were put on the heat-controlled stage of an inverted microscope coupled to a video-motion analyzer (Ionoptix) and perfused with HBSS-HEPES-arginine with graded concentrations of agonists.7 Ectopic contractions were detected as beats occurring at either <80% or >120% than average beat interval calculated over the recording. For inotropic analysis, cardiomyocytes were paced at 5 Hz with a platinum electrode (≈5V) to control for independent effects of changes in beating rate.
In Vivo Telemetry
Miniaturized telemetry devices (Datascience Corp) were implanted as previously described,16 and mice were left to recover for 1 week before recordings were done. Agonists were infused in unanesthetized, freely moving mice through an intraperitoneal catheter connected to a syringe above the cage. Heart rate was derived from blood pressure signals sampled at 2000 Hz (Notocord HEM 3.4 software).
Differences were tested by 1- or 2-way ANOVA or Student’s t test for paired or unpaired data, where appropriate. Values of P<0.05 were considered significant.
Transgenic eNOS Interacts and Colocalizes With Caveolin-3 in Caveolae-Enriched Fractions of Mouse Cardiomyocytes
The abundance of eNOS was compared in extracts of cultured neonatal cardiomyocytes from NOS3-TG and wild-type mice (Figure 1A). Densitometric analysis revealed a 5-fold increase in eNOS proteins (corrected for α-sarcomeric actin) in NOS3-TG relative to levels in wild-type mice (P<0.05; n=4 from 3 preparations).
Different posttranslational modifications of heterologously expressed proteins may affect their subcellular targeting and function; therefore, we assessed targeting of transgenic eNOS with 4 different approaches. Because a significant proportion of eNOS is known to interact with the myocyte-specific isoform of caveolin, caveolin-3, we first assessed the interaction between these 2 proteins by a coimmunoprecipitation assay. Figure 1B shows that caveolin-3 is coimmunoprecipitated with eNOS using an anti-eNOS immunoprecipitating antibody. Specificity of this interaction is demonstrated by the negligible caveolin-3 signal in the control lacking the immunoprecipitating antibody and the absence of coimmunoprecipitation of troponin I.
Because caveolin-3 is a resident protein of caveolae, we then verified whether eNOS and caveolin-3 are colocalized in light membrane fractions enriched in caveolae, as obtained by sucrose gradient ultracentrifugation. Figure 1C shows that a significant portion of both caveolin-3 and eNOS was found in the same low-density (light) fractions (1 to 6). The same proteins were also found in heavier fractions, as expected, but given the much lower total protein content in lighter fractions, the signals presented correspond to a substantial enrichment of eNOS in caveolar fractions.
Next, we assessed the colocalization of these 2 proteins by immunofluorescence in intact cells and tissue. Figure 2 shows eNOS immunostaining (in green), caveolin-3 immunostaining (in red), and the superposition of both (in yellow) in cultured neonatal (Figure 2A) and in situ adult (Figure 2B) cardiomyocytes from NOS3-TG mice. Consistent with the data from the cellular fractionation experiments, the combined images indicate colocalization of eNOS and caveolin-3 mostly at the cell periphery. Immunofluorescence analysis of sections of adult ventricular hearts also shows an enrichment of eNOS in the myocardium of NOS3-TG mice (Figure 2C, right) compared with wild-type (Figure 2C, left).
Finally, immunogold labeling of adult heart sections analyzed by electron microscopy showed an ≈3-fold enrichment of eNOS in plasmalemmal and T-tubular membranes compared with cytosolic structures (Figure 3). Similar results were obtained with 3 different polyclonal anti-eNOS antibodies (all P<0.05), whereas no gold particle was observed in the absence of primary antibody.
eNOS Overexpression Potentiates the Postsynaptic, Muscarinic Negative Chronotropic Effect In Vitro and In Vivo
We then tested whether recombinant eNOS was physiologically regulated by physical (ie, beating) and agonist stimulation, as would be expected from its membrane localization, which allows proper coupling to specific receptor pathways. We first examined the basal rhythm of spontaneously beating cardiomyocytes from the 2 strains in culture. Using a specific algorithm for the quantification of ectopic beats under videomicroscopy (see Methods), we observed their incidence to be lower in myocytes from NOS3-TG than from wild-type mice (2.6±1.2% versus 40.8±5.4% ectopic beats; n=33; P<0.0001). Treatment of NOS3-TG cells with a NOS inhibitor, l-nitro arginine (L-NA) in part attenuated the difference (11.9±3.1%; n=28; P<0.005), as well as intracellular cGMP levels (see supplemental Figure online), which confirms that this rhythm stabilization depends in part on a functional enzyme producing NO. We next examined the response of these myocytes to carbamylcholine (Figure 4), an analog of acetylcholine, which we have previously shown to produce a NO-dependent negative chronotropic response in a similar model.7 In wild-type myocytes, carbamylcholine produced a dose-dependent slowing of the spontaneous beating rate that was little influenced by L-NA treatment, in line with the low endogenous expression of eNOS in this strain (from a C57Bl/6J background), as shown in Figures 1A and 2⇑C. In contrast, the negative chronotropic effect of carbamylcholine was clearly potentiated in NOS3-TG mice (n=7 to 10; P<0.05) and was largely inhibited after treatment with L-NA (n=6; P<0.0001).
Similar findings were obtained in vivo after infusion of carbamylcholine in unanesthetized mice under telemetry, with a potentiation of the negative chronotropic effect in NOS3-TG (EC50=19±6 μg/kg) compared with wild-type mice (45±5 μg/kg; n=4 to 5; P<0.01; Figure 5A), despite similar blood pressure responses (Figure 5B). This potentiation was also abrogated under l-nitro-arginine methyl ester (heart rate at 35 μg/kg: 88.7±9% versus 86.6±0.9% of baseline; not shown). This confirms that the transgenic eNOS is functionally coupled to muscarinic stimulation and that cardiomyocyte NO production potentiates the muscarinic chronotropic response.
eNOS Overexpression Attenuates the Contractile Response to Maximal β-Adrenergic Stimulation
Next, we examined modulation of the β-adrenergic response to isoproterenol by transgenic eNOS. Isoproterenol increased the amplitude of cell shortening (an index of contraction force, normalized as percent of maximum response to 4 mmol/L Ca2+) in wild-type myocytes, as expected. By comparison, the increase in shortening was significantly blunted in myocytes from NOS3-TG at high concentrations of isoproterenol (n=10 to 17; P<0.05; Figure 6). Similarly, the isoproterenol-induced increase in contractile velocity (+dl/dt) was attenuated in NOS3-TG myocytes (isoproterenol 10−6 mol/L: 75.8±10.1% versus 100.8±8.1% of maximum Ca2+; n=10 to 17 from 3 preparations; P<0.05; not shown). Again, NOS inhibition with L-NA in NOS3-TG restored the shortening amplitude and velocity (109.5±8.6%; P<0.05) responses back to levels in wild type.
In vivo, the positive chronotropic effect of isoproterenol (25 mg/kg) was also blunted in NOS3-TG (+Δbpm: 41.2±27.1) compared with wild type (159.7±20.6 bpm; n=3 each; P<0.05) despite similar blood pressure responses (−Δmm Hg: 24.3±4.4 and 35.3±7.3; P=NS). This confirmed the efficient coupling of transgenic eNOS to β-adrenergic receptors and confirmed that cardiomyocyte NO production attenuates the inotropic and chronotropic effect of catecholamines.
eNOS Overexpression Potentiates the Accentuated Antagonism
The vagal inhibition of cardiac contraction is accentuated in cardiac preparations prestimulated with a β-adrenergic agonist, a phenomenon classically referred to as “accentuated antagonism.” To examine its modulation by cardiomyocyte eNOS, we compared the accentuated antagonism between the 2 strains. As expected, carbamylcholine had little, if any, effect on the shortening amplitude in unstimulated cells from any group (Figure 7, left) but consistently inhibited β-adrenergically stimulated contraction by 5% to 10% in wild-type cells. NOS3-TG cardiomyocytes were then prestimulated with a concentration of isoproterenol (1 to 10.10−8 mol/L) that was adjusted to obtain a similar level of shortening as in wild-type cells; when carbamylcholine was next applied on these cells, it produced a 3-fold greater inhibition than in wild-type cells (P<0.001; Figure 7, right). This accentuated antagonism was significantly reduced by preincubation with a NOS inhibitor (P<0.005).
We provide evidence at the isolated cardiomyocyte level that autocrine NO produced by eNOS stabilizes the spontaneous beating rate, attenuates the maximal inotropic response to catecholamines, and potentiates the muscarinic negative chronotropic effect in vitro and in vivo and attenuates antagonism of the β-adrenergic inotropic response. Our use of purified myocytes expressing eNOS under the α-MHC promoter enabled us to definitively separate these effects from indirect neural or vascular paracrine influences that have long confounded the interpretation of experiments in whole organ or tissue preparations.
The modulatory role of cardiomyocyte eNOS on muscarinic/adrenergic signaling has not been uniformly observed in previous studies, but technical limitations, as well as chronic phenotypic adaptation in unrestricted, nonconditional genetic deletion experiments, may have explained some of the negative data (reviewed in Massion et al2).The present results nevertheless recapitulate previous paradigms on the modulation of the β-adrenergic and muscarinic response by cardiomyocyte eNOS in isolated mammalian hearts and cardiac cells.1,2,6,7,11,17 Of interest, our observations suggest that the eNOS-dependent modulation of the muscarinic response may prevail over NOS-independent mechanisms in cases of eNOS overexpression, as illustrated by the increased sensitivity of the chronotropic response to NOS inhibition in NOS3-TG cells (Figure 5). Similarly, the muscarinic inhibition of cardiac contraction was potentiated by the NOS substrate l-arginine in hearts from pertussis toxin–treated rats that exhibited a moderate upregulation of eNOS expression. This eNOS-dependent pathway even compensated for Gα-i inhibition by the toxin, which suggests that eNOS can act as a backup muscarinic coupling mechanism.17 Another study of the contractility of whole hearts overexpressing a cardiomyocyte-restricted eNOS transgene18 previously concluded that there was a neutral effect of eNOS on the β-adrenergic or muscarinic response, but contrary to the present study, the strains used in that report had much higher (40- to 90-fold) overexpression, which probably accounted for continuous, nonregulated NO production in all cellular compartments. This may have explained the depressed basal contraction and the absence of selective modulation of agonist response. Conversely, we provide extensive evidence that the moderately expressed transgenic eNOS in the present study was targeted to its proper compartment, ie, a caveolae-enriched membrane fraction, and as such, was amenable to activation on stimulation of membrane receptors. This further emphasizes the importance of colocalization of eNOS and muscarinic cholinergic receptors, in particular, in cardiomyocyte caveolae for efficient NO production and modulation of the muscarinic response.10,11 Our in vivo data with telemetry extend the validity of these paradigms to the adult heart and are in complete agreement with a recent in vivo study in adult NOS3-null mice transfected with an adenovirus encoding a cardiac-restricted NOS3.6 Compared with the latter, our transgenic approach ensures uniform expression in all cardiomyocytes, and our use of purified cardiomyocytes further excludes any indirect effect of NO production on (or from) the surrounding vasculature, which may confound the interpretation with whole-heart preparations. Other studies provided evidence for a distinctive role of nNOS within mouse cardiomyocytes to modulate excitation-contraction coupling, as well as β-adrenergic contractile response.5 The present study does not support or invalidate this additional paradigm but emphasizes the specific potential of eNOS to modulate postsynaptic autonomic regulation of heart function, eg, its parasympathetic limb.
When integrated at the level of whole-organ physiology, however, these effects of cardiomyocyte eNOS may combine with other NO sources to reinforce the vagal inhibition of heart rate (eg, with presynaptic nNOS19) and attenuate β-adrenergic stimulation. Such reequilibration of the sympathovagal balance by cardiomyocyte eNOS would be beneficial in states of excessive adrenergic input likely to damage the myocardium, such as in the failing heart. Indeed, a functional eNOS polymorphism (Asp 298 eNOS, characterized by reduced stability of the mutated enzyme) has been associated with an adverse prognosis in a population of heart failure patients specifically with nonischemic etiology, which further suggests the functional importance of myocardial (versus vascular) eNOS.20 Accordingly, restoration of myocardial eNOS activity, as observed with β-blockers21 or statins16,22 in patients, may contribute to their protective effects, eg, against arrhythmia. In agreement with recent results in vivo,6,15 the present cellular analysis demonstrating proper targeting and receptor coupling of heterologously expressed eNOS emphasizes the potential of such approaches, which now warrant rigorous testing in models of cardiac diseases.
This study was supported by a Fonds de la Recherche Scientifique Médicale grant (3.4738.98F), an Action de Recherche Concertée (ARC 01-06) from the Communauté Française de Belgique, and Pôle d’Attraction Interuniversitaire (P5/02) from the Politique Scientifique Fédérale, Belgium. Dr Massion was supported by the Fond Special de Recherche (UCL) and the Fondation Damman. Dr Janssens is a clinical investigator for the FWO-Vlaanderen and is the Astra-Zeneca cardiology chair holder. Drs Dessy and Feron are Research Associates and Dr Belge and F. Desjardins are Fellows of the Fonds National de la Recherche Scientifique.
The Supplemental figure is available as an online-only Data Supplement at http://www.circulationaha.org.
Balligand JL, Kobzik L, Han X et al. Nitric oxide-dependent parasympathetic signaling is due to activation of constitutive endothelial (type III) nitric oxide synthase in cardiac myocytes. J Biol Chem. 1995; 270: 14582–14586.
Massion PB, Feron O, Dessy C, et al. Nitric oxide and cardiac function: ten years after, and continuing. Circ Res. 2003; 93: 388–398.
Xu KY, Huso DL, Dawson TM, et al. Nitric oxide synthase in cardiac sarcoplasmic reticulum. Proc Natl Acad Sci U S A. 1999; 96: 657–662.
Ashley EA, Sears CE, Bryant SM, et al. Cardiac nitric oxide synthase 1 regulates basal and β-adrenergic contractility in murine ventricular myocytes. Circulation. 2002; 105: 3011–3016.
Champion HC, Georgakopoulos D, Takimoto E, et al. Modulation of in vivo cardiac function by myocyte-specific nitric oxide synthase-3. Circ Res. 2004; 94: 657–663.
Balligand JL, Kelly RA, Marsden PA, et al. Control of cardiac muscle cell function by an endogenous nitric oxide signaling system. Proc Natl Acad Sci U S A. 1993; 90: 347–351.
Feron O, Smith TW, Michel T, et al. Dynamic targeting of the agonist-stimulated m2 muscarinic acetylcholine receptor to caveolae in cardiac myocytes. J Biol Chem. 1997; 272: 17744–17748.
Feron O, Dessy C, Opel DJ, et al. Modulation of the endothelial nitric-oxide synthase-caveolin interaction in cardiac myocytes: implications for the autonomic regulation of heart rate. J Biol Chem. 1998; 273: 30249–30254.
Piech A, Dessy C, Havaux X, et al. Differential regulation of nitric oxide synthases and their allosteric regulators in heart and vessels of hypertensive rats. Cardiovasc Res. 2003; 57: 456–467.
Ratajczak P, Damy T, Heymes C, et al. Caveolin-1 and -3 dissociations from caveolae to cytosol in the heart during aging and after myocardial infarction in rat. Cardiovasc Res. 2003; 57: 358–369.
Lyons GE, Schiaffino S, Sassoon D, et al. Developmental regulation of myosin gene expression in mouse cardiac muscle. J Cell Biol. 1990; 111: 2427–2436.
Janssens S, Pokreisz P, Schoonjans L, et al. Cardiomyocyte-specific overexpression of nitric oxide synthase 3 improves left ventricular performance and reduces compensatory hypertrophy after myocardial infarction. Circ Res. 2004; 94: 1256–1262.
Pelat M, Dessy C, Massion P, et al. Rosuvastatin decreases caveolin-1 and improves NO-dependent heart rate and blood pressure variability in apolipoprotein E(−/−) mice in vivo. Circulation. 2003; 107: 2480–2486.
Brunner F, Andrew P, Wolkart G, et al. Myocardial contractile function and heart rate in mice with myocyte- specific overexpression of endothelial nitric oxide synthase. Circulation. 2001; 104: 3097–3102.
McNamara DM, Holubkov R, Postava L, et al. Effect of the Asp298 variant of endothelial nitric oxide synthase on survival for patients with congestive heart failure. Circulation. 2003; 107: 1598–1602.
Fukuchi M, Hussain SN, Giaid A. Heterogeneous expression and activity of endothelial and inducible nitric oxide synthases in end-stage human heart failure: their relation to lesion site and β-adrenergic receptor therapy. Circulation. 1998; 98: 132–139.
Node K, Fujita M, Kitakaze M, et al. Short-term statin therapy improves cardiac function and symptoms in patients with idiopathic dilated cardiomyopathy. Circulation. 2003; 108: 839–843.