Vasostatins (VSs) are vasoactive peptides derived from chromogranin A (CgA), a protein contained in secretory granules of chromaffin and other cells. The negative inotropic effect and the reduction of isoproterenol (Iso)-dependent inotropism induced by VSs in the heart suggest that they have an antiadrenergic function. However, further investigation of the mechanisms of action of VSs is needed. The aim of the present study was to define the signaling pathways activated by VS-1 in mammalian ventricular myocardium and cultured endothelial cells that lead to the modulation of cardiac contractility. Ca2+ and nitric oxide (NO) fluorometric confocal imaging was used to study the effects induced by recombinant human VS-1 [STA-CgA-(1-76)] on contractile force, L-type Ca2+ current, and Ca2+ transients under basal conditions and after β-adrenergic stimulation in rat papillary muscles and ventricular cells and the effects on intracellular Ca2+ concentration and NO production in cultured bovine aortic endothelial (BAE-1) cells. VS-1 had no effect on basal contractility of papillary muscle, but the effect of Iso stimulation was reduced by 27%. Removal of endocardial endothelium and inhibition of NO synthesis and phosphatidylinositol 3-kinase (PI3K) activity abolished the antiadrenergic effect of VS-1 on papillary muscle. In cardiomyocytes, 10 nM VS-1 was ineffective on basal and Iso (1 μM)-stimulated L-type Ca2+ current and Ca2+ transients. In BAE-1 cells, VS-1 induced a Ca2+-independent increase in NO production that was blocked by the PI3K inhibitor wortmannin. Our results suggest that the antiadrenergic effect of VS-1 is mainly due to a PI3K-dependent NO release by endothelial cells, rather than a direct action on cardiomyocytes.
- calcium channel
- myocardial contractility
- peptide hormones
- endothelial cell
chromogranin A (CgA) and chromogranin B have long been proposed to control the physiological process of secretory granule formation because of their pH-, Ca2+-, and catecholamine-dependent aggregation properties (18). In particular, it has been suggested that the 49-kDa acidic protein CgA, co-stored and co-released in the secretory granules of the chromaffin cells, together with catecholamines and other neuromodulators, acts as a prohormone, giving rise to peptides such as vasostatins (VSs), chromacin, pancreastatin, WE-14, catestatin, and parastatin, which are involved in homeostatic processes, including tissue assembly and repair, inflammatory responses, and the first line of defense against invading microorganisms (18, 23).
Recent studies, however, strongly suggest a role for CgA-derived peptides also in the physiological mechanism controlling the cardiovascular system. In CgA-null mice, Mahapatra et al. (20) found extreme cardiovascular changes, including elevated systolic and diastolic blood pressure, as well as loss of diurnal blood pressure variation, together with significant ECG changes. Interestingly, Mahapatra et al. were able to restore these parameters to normal levels by exogenous catestatin replacement or “humanization” of CgA-null mice with transgenic insertion of a human CgA haplotype (20). It has been suggested that CgA and CgA-derived peptides are also involved in cardiovascular pathology. Ceconi et al. (8) documented increased levels of circulating CgA in patients with heart failure, depending on the severity of the disease.
Among the CgA-derived peptides, the NH2-terminal fragments, termed VSs for their vasoinhibitory action in conduit and resistance vessels (1, 2, 3), including bovine coronary resistance arteries (5), appear most likely to be novel cardioregulatory peptides in mammals. VS-1 and VS-2, the CgA-(1-76) and CgA-(1-113) polypeptides, respectively, are generated by cleavage at the first and second pair of basic amino acid residues of the NH2-terminal domain of CgA (10).
Recently, VS-1 and VS-2 have been shown to be important protective modulators of cardiac activity because of their ability to counteract the adrenergic signals (9, 18). The negative inotropism exerted by VS-1, even with different mechanisms, has been observed in several vertebrate species. VS-1 and VS-2 induce a negative inotropic effect on the isolated heart of the eel (19), frog (11, 12, 26), and rat (9) under basal conditions and after β-adrenergic stimulation. Interestingly, although the action of VSs on frog cardiac muscle appears direct, i.e., independent of endocardial endothelial (EE) cells and muscarinic/β-adrenergic-related pathways (12, 26), in the heart of the eel, it depends on EE-dependent activation of M1 muscarinic receptors and Gi/o, as well as synthesis of nitric oxide (NO) and cGMP (19). Moreover, in eel and frog heart, changes in cytoskeletal dynamics seem to play a crucial role in the negative inotropic effect of VS-1 (22). In the isolated rat heart, VS-1, but not VS-2, elicits a dose-dependent negative inotropism under basal conditions, but both peptides abolish the isoproterenol (Iso)-induced positive inotropism without modifying the β-adrenergic-dependent coronary dilatation or the ouabain-induced positive inotropism (9).
Since the mechanisms underlying the VS-dependent vasorelaxing and cardioinhibitory effects remain to be fully elucidated, the aim of the present study was to define the signaling pathways activated by VS-1 in mammalian cardiac and endothelial cells that lead to modulation of cardiac contractility. We used different preparations, isolated ventricular cells and papillary muscle, to study the effects of human recombinant VS-1 on L-type Ca2+ current (ICa,L), Ca2+ transients, and contractile force under basal and Iso-stimulated conditions and cultured endothelial cells to study NO production and its relative modulation. We have shown that the VS-1-induced antiadrenergic inotropic effect is due to a Ca2+-independent phosphatidylinositol 3-kinase (PI3K)-dependent endothelial release of NO, rather than a direct action on cardiac cells.
Female rats (n = 55, 200–300 g body wt) were allowed ad libitum access to tap water and standard rodent diet. The animals received humane care in compliance with the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals (NIH Publication No. 85-23, revised 1996) and in accordance with Italian law (DL-116, 27 January 1992). The scientific project was supervised and approved by the local ethical committee.
Ventricular cell isolation.
Young adult rats were anesthetized with pentobarbital sodium (Nembutal; 1 mg/g ip) and killed by stunning and cervical dislocation. The hearts were explanted, washed in modified Ca2+-free Tyrode solution (see Solutions and drugs), and cannulated via the aorta. All operations were carried out under a laminar flow hood. The heart was perfused at a constant flow rate of 10 ml/min with Ca2+-free Tyrode solution with a peristaltic pump for ∼5 min at 37°C to wash away the blood and then with 10 ml of Ca2+-free Tyrode solution supplemented with collagenase (0.3 mg/ml) and protease (0.02 mg/ml). Hearts were then perfused and enzymatically dissociated with 30 ml of the Ca2+-free Tyrode solution containing 50 μM CaCl2. Atria and ventricles were separated, and the ventricles were cut into small pieces and shaken for 10 min in 20 ml of Ca2+-free Tyrode solution containing 50 μM CaCl2, collagenase, and protease.
For measurement of ICa,L, the cell was depolarized every 5 s from −100 to −40 mV for 50 ms to inactivate fast Na+ current (INa) and T-type Ca2+ current and for 300 ms from −40 to +10 mV, which is the potential at which ICa,L is maximal in most cells. Warm cesium-containing Tyrode solution was constantly perfused into the bath with a peristaltic pump. Ca2+ current was measured as the difference between the peak of the inward current and the current at the end of the 300-ms pulse. The steady-state current at −100 mV and INa were monitored to control the stability of the experiment. For exchange of external solutions, the pipette with the cell attached was moved into the tip of one of a series of six silicone plastic tubes (250 μm diameter) connected to gravity-fed reservoirs of the desired test solutions. All control and analysis of the experiments were performed with Pulse software (Heka) on PowerMac computers (Apple Computer, Cupertino, CA). All experiments were performed at ∼30°C under continuous thermostatic control.
Measurement of Ca2+ transients.
Cardiomyocytes were loaded with 2 μM indo 1-AM (Invitrogen, Carlsbad, CA) and incubated for 40 min at 37°C. Cells were then washed in control Tyrode solution and placed on the stage of an inverted microscope. Cells were electrically stimulated at a frequency of 2 Hz by a two-platinum-electrode insert (model RC-37W, Warner Instruments) connected to a bipolar stimulator (model SIU-102, Warner Instruments). Ca2+ transients were evaluated as the ratio of fluorescence at 400 nm to 490 nm emitted by the cells excited at 350 nm. During experiments, cardiomyocytes were maintained in standard Tyrode solution, and solution changes were performed with a microperfusion system. The experiments were recorded by Axograph software on an PowerMac computer and analyzed with Igor software.
Isolated papillary muscle.
Papillary muscles were dissected free from the left ventricle under a stereomicroscope and superfused with oxygenated Tyrode solution at 36°C. Papillary muscles were driven at constant frequency (120 beats/min) with a pair of electrodes connected to a stimulator (302 T Anapulse, W. P. Instruments, New Haven, CT) via a stimulus isolator (model 305-R, W. P. Instruments) operating in constant-current mode. Isometric twitches were evaluated by a transducer (model 60-2997, Harvard Instruments) and continuously acquired and recorded by a PowerMac computer using Labview software (National Instruments).
Before each experiment, papillary muscles were equilibrated in oxygenated (100% O2) Tyrode solution for ≥30 min. In preliminary experiments, we tested the effects of different concentrations (2–100 nM) of VS-1 and found that 10 nM VS-1 was the highest concentration without effect on basal contractility. Therefore, we chose to use 10 nM VS-1 in the subsequent experiments. The muscles were treated with VS-1 for 20 min; then the perfusion was switched to Tyrode solution alone to study the reversibility of the effects. All solutions containing drugs were prepared immediately before the experiments. To investigate the role of EE in cardiac alterations induced by VS-1, additional experiments were performed in papillary muscle treated with 0.5% Triton X-100 for 1-2 s, followed by 20–30 min of washout with Tyrode solution to remove EE (6). NG-nitro-l-arginine methyl ester (l-NAME, 1 mM) was used to block NO synthesis. Wortmannin (100 nM) was used to block PI3K activity.
Bovine aortic endothelial cells.
The effects of VS-1 on intracellular Ca2+ concentration and NO production were studied in bovine aortic endothelial (BAE-1) cells by confocal microscopy. BAE-1 cells (European Collection of Cell Cultures, Salisbury, Wiltshire, UK) were maintained in DMEM (Sigma, St. Louis, MO) supplemented with 10% heat-inactivated FCS (lot no. 1SB0019, BioWhittaker, Verviers, Belgium), 100 U/ml penicillin, 100 μg/ml streptomycin, 2 mM glutamine, and 0.25 mg/ml amphotericin B (Fungizone) at 37°C in a humidified atmosphere of 5% CO2 in air. Cells were used at passages 2–6.
Confocal fluorometric measurements were performed using a laser scanning confocal system (Fluoview 200, Olympus America, Melville, NY) mounted on an inverted microscope (model IX70, Olympus) equipped with a ×60 oil-immersion objective (NA 0.17). Cells were seeded on glass-bottom dishes (35 × 22 mm, Willco Wells, Amsterdam, The Netherlands) at a density of 5,000 cells/cm2. For simultaneous Ca2+ and NO measurements, cells were loaded simultaneously with 2 μM fluo 3-AM (Invitrogen) and 2 μM DAR 4M-AM (Invitrogen) for 30 min at 37°C and excited at 488 and 568 nm. Emission signals were filtered with 515-nm (for fluo 3-AM) and 610-nm (for DAR 4M-AM) band-pass filters. During experiments, BAE-1 cells were maintained in standard Tyrode solution. Solutions were applied with a microperfusion system (200-μm-ID pipette). x-y Plane images (resolution 512 × 512 pixels) were acquired every 3.3 s and subsequently analyzed with ImageJ, a public domain Java image-processing software tool (Rasband W. ImageJ, version 1.36, NIH, Bethesda, MD). Changes in intracellular Ca2+ concentration were represented as (F − F0)/F0 (where F is fluorescence) to normalize the traces. Inasmuch as the reaction of DAR 4M-AM with NO forming the corresponding fluorescent triazole compound is not reversible, changes in intracellular NO synthesis were expressed as (F − F0)/F0 and as the first derivative of the fluorescence signal to highlight accumulating NO synthesis over changes in slope.
Solutions and drugs.
Ca2+-free Tyrode solution contained (mM) 135 NaCl, 4 KCl, 1 MgCl2, 2 HEPES, 10 glucose, 10 butanedione monoxime, and 5 taurine, with pH adjusted to 7.40 with NaOH. The whole cell patch-clamp standard solution contained (mM) 133 CsCl, 5 EGTA-free acid, 5 Na2ATP, 5 Na2phosphocreatine, 5 HEPES, 3 MgCl2, and 0.4 Na2GTP, with pH adjusted to 7.3 with CsOH. Tyrode solution contained (mM) 154 NaCl, 4 KCl, 2 CaCl2, 1 MgCl2, 5.5 d-glucose, and 5 HEPES, with pH adjusted to 7.35 with NaOH. Cesium Tyrode solution contained (mM) 138 NaCl, 20 CsCl, 2 CaCl2, 1 MgCl2, 5.5 d-glucose, 1 4-aminopyridine, and 5 HEPES, with pH adjusted to 7.35 with NaOH. All drug-containing solutions were prepared fresh before the experiments. We used recombinant human VS-1 [STA-CgA-(1-78)] containing the VS-1 peptide (1-76) previously described (9). The rat CgA sequence is identical in that region to the human sequence. If not otherwise indicated, chemicals were purchased from Sigma.
Values are means ± SE. All data were subjected to ANOVA followed by Bonferroni's correction for post hoc t-tests. Significance was accepted at P < 0.05.
Isolated papillary muscle.
Previous work has shown that VS-1 exerts a negative inotropic effect under basal conditions and on β-adrenergic stimulation in isolated rat heart (9). In agreement with the latter result, we observed that VS-1 reduces the inotropic effect of 1 μM Iso in papillary muscle. As shown in Fig. 1, the enhancement of contractile force induced by 1 μM Iso was blunted by VS-1 in a concentration-dependent manner. Although 2 nM VS-1 had no significant effect (+1.0 ± 6.9% of Iso alone, n = 4), 5 and 10 nM VS-1 reduced the effect of Iso by 11.1 ± 1.5 and 26.8 ± 5.2% (n = 4 and 5, respectively). Higher (25–100 nM) concentrations of VS-1 exerted a comparable antiadrenergic effect (23.6 ± 4.8% of baseline at 100 nM, n = 5). Under basal conditions, low concentrations (≤10 nM) of VS-1 had no effect on myocardial contractility (98.5 ± 1.5 and 101.0 ± 1.0% of baseline at 5 and 10 nM, respectively), whereas higher concentrations of VS-1 enhanced developed tension in a dose-dependent manner (17.0 ± 1.5, 29.7 ± 1.8, and 48.0 ± 15.0% over control at 25, 50, and 100 nM, respectively). The positive inotropic effect of 100 nM VS-1 was abrogated by pretreatment of papillary muscles with the β-adrenergic blocker propranolol (2.7 ± 5.4% over control, P < 0.05).
On the basis of previous experiments suggesting that, at least in certain animal species, the antiadrenergic effect of VS-1 on cardiac muscle depends on NO release from endothelial cells (19), we studied the effect of 10 nM VS-1 after removal of EE or after pretreatment with 1 mM l-NAME. We observed that removal of EE with Triton X-100 did not modify the inotropic effects induced by Iso under basal conditions (165.7 ± 19.6% over control), whereas the antiadrenergic effect of VS-1 was completely blocked (99.5 ± 2.4% of Iso alone, n = 6). Similarly, pretreatment of papillary muscles with 1 mM l-NAME significantly reduced the antiadrenergic effect of VS-1 after Iso (97.3 ± 2.5% of Iso alone), whereas the inotropic effect induced by Iso was not modified (106.3 ± 17.1% over control, n = 6).
Inasmuch as a mechanism of activation of endothelial NO synthase (eNOS) might involve protein kinase B (Akt), which has recently been shown to phosphorylate eNOS and increase its sensitivity to Ca2+, so that it is active at subphysiological concentration (15, 17), we tested the role of the PI3K-Akt-NO pathway in the endothelium-dependent antiadrenergic effect induced by VS-1. Papillary muscles were treated with 100 nM wortmannin before administration of Iso and VS-1. As shown in Fig. 1, the antiadrenergic effect of VS-1 in the presence of Iso was significantly reduced (97.4 ± 1.2% of Iso alone, n = 6), suggesting that PI3K plays an important role in the inotropic effects induced by VS-1.
Isolated ventricular cells.
To exclude a direct inotropic effect of VS-1 on cardiomyocytes, we measured ICa,L by whole cell patch-clamp experiments in basal and Iso-stimulated conditions. VS-1 (10 nM) was ineffective on basal and Iso-stimulated ICa,L (Fig. 2). In agreement with the effects on papillary muscle previously shown, 100 nM VS-1 induced a significant increase of the basal current (34.8 ± 7.6%, n = 8) that was completely abrogated by 1 μM propranolol (3.6 ± 3.9%, n = 4).
To investigate the effect of VS-1 on other voltage-dependent ionic currents involved in the action potential and, therefore, the Ca2+ transients, we measured Ca2+ transients in electric-field-stimulated cardiomyocytes loaded with the fluorescent Ca2+ probe indo 1. As shown in Fig. 3, 10 nM VS-1 was ineffective on basal and Iso-stimulated Ca2+ transients.
NO production by BAE-1 cells.
Inasmuch as the previous experiments on isolated papillary muscle and isolated ventricular myocytes suggest an indirect, endothelium-derived NO-dependent antiadrenergic effect of VS-1, we simultaneously measured NO and intracellular Ca2+ in BAE-1 cells with fluorescent probes. The cells were observed under confocal microscopy after they were loaded with fluo 3-AM and DAR 4M-AM. As a positive control, 100 μM ATP was used to induce Ca2+-dependent NO production.
In contrast with ATP, VS-1 enhanced NO production with a mechanism that was independent of intracellular Ca2+ concentration. To distinguish the increase in DAR 4M-AM fluorescence intensity from the accumulating basal NO synthesis, we measured the slope changes by calculating the derivative of the fluorescence intensity (1.13 ± 0.28 in control and 5.57 ± 0.7 after VS-1, n = 41). Figure 4 shows a representative experiment of NO and Ca2+ measurement on a single cell from a selected field.
Inasmuch as PI3K blockade abolished the NO-mediated antiadrenergic effect of VS-1 in isolated papillary muscle, we investigated the VS-1-dependent NO increase in BAE-1 cells pretreated with 100 nM wortmannin. VS-1 failed to induce NO synthesis stimulation (derivative of fluorescence intensity was 1.03 ± 0.14 after wortmannin pretreatment and 2.15 ± 0.57 after wortmannin + VS-1, n = 20; Fig. 5). As shown in Fig. 5C, slope changes only in the presence of VS-1 alone.
The present study shows that the cardiac antiadrenergic action of VS-1 does not directly affect cardiomyocytes but rather, is mediated by a Ca2+-independent/PI3K-dependent NO release in endothelial cells.
Previous studies have shown that CgA-derived VSs, in particular VS-1, at nanomolar concentrations, induce a negative inotropic effect on isolated heart preparations from different animal species under basal conditions and after β-adrenergic stimulation. Therefore, it has been suggested that VSs, in addition to their vasoinhibitory properties, are cardioregulatory mediators; i.e., they are able to exert an antiadrenergic action that could protect the heart against intense excitatory stimulations, such as those elicited by the stress response (9, 11, 19, 26).
Our experiments on isolated papillary muscle confirmed the antiadrenergic action of VS-1 previously observed in the isolated heart. The negative antiadrenergic effect of VS-1 was present at low concentrations (5–10 nM), which were comparable to the serum concentrations of CgA (0.5–2 nM) detected radiologically in healthy humans (18) and plasma concentrations of CgA (5–20 nM) in patients with severe heart failure (8). Conceivably, the intracardiac concentrations of CgA and its derived fragments may be even higher under certain conditions, e.g., in response to stress and/or secretagogues, promoting their secretion by various cell types. In fact, immunodetection studies have documented the presence of CgA, colocalized with atrial natriuretic peptides, in the secretory granules of rat atrial myoendocrine cells (25), as well as in rat Purkinje cells of the conducting system and in H9c2 rat cardiomyocytes (27). These data, together with the recent biochemical identification of CgA-derived peptides containing the vasostatin motif in the rat heart (16), strongly suggest that CgA-derived peptides play an important autocrine/paracrine role in cardiovascular homeostasis in physiological and pathophysiological conditions.
To investigate the mechanism involved in the inotropic role of VS-1, we used whole cell patch-clamp experiments to test the effect of VS-1 on ICa,L in rat ventricular cardiomyocytes. VS-1 was ineffective on ICa,L in the basal condition and after Iso stimulation.
It has indeed been shown that the negative inotropism induced by VS-1 in isolated frog (12) and eel (19) hearts is, at least in part, dependent on the availability of K+ channels, since it was abolished by pretreatment with the K+ channel inhibitors Ba2+, 4-aminopyridine, tetraethylammonium chloride, and glibenclamide (12, 19). To exclude the possibility that, in our model, VS-1 could affect ionic channels other than the L-type Ca2+ channel, we measured Ca2+ transients on electric-field-stimulated cardiomyocytes. The lack of effect of VS-1 on basal and Iso-stimulated Ca2+ transient amplitude represents an indirect but strong demonstration that the ionic currents involved in the action potential are not affected by VS-1.
The absence of an inhibitory effect on isolated ventricular cells suggests that, in the rat heart, VS-1 does not act directly on cardiomyocytes but, rather, on other cell types that are present alongside them in cardiac tissue. We performed additional experiments to study the role of EE and NO as mediators of the effects of VS-1 in papillary muscle. These experiments strongly suggested that the antiadrenergic effect of VS-1 is mainly due to stimulation of NO production in EE. Indeed, removal of EE and inhibition of NO synthesis did not modify the inotropic effect induced by Iso under basal conditions, whereas the antiadrenergic effect of VS-1 was completely blocked.
Results of experiments performed on BAE-1 cells further support the role of NO released by EE in the effects exerted by VS-1. These experiments indirectly indicate that VS-1 may promote the release of NO also from the EE by means of a Ca2+-independent mechanism (Fig. 4).
Ca2+-independent activation of eNOS is not unknown: there have been reports that insulin, insulin-like growth factor-1, and estrogens can increase NO production by endothelium without increasing intracellular free Ca2+ (13, 15, 17, 24). As proposed by Hartell et al. (17) and Shaul (24), such a mechanism might involve Akt-dependent NO synthase phosphorylation.
Our experiments performed on papillary muscles and on BAE-1 cells treated with wortmannin strongly suggested that VS-1-induced NO synthesis depends on PI3K activation.
Interestingly, a recent report proposes a novel Ca2+-independent mechanism of eNOS activation involving caveolae-mediated endocytosis induced by the albumin-binding protein gp60 and activation of downstream Src, Akt, and PI3K pathways (21). Inasmuch as it has been observed that VSs may interact with the caveolar domain (4) and that endothelial cells internalize CgA-(1-78) (14), we could suggest a similar mechanism to explain VS-1-dependent NO synthase activation in BAE-1 cells.
Even if the experiments on BAE-1 cells clarify the mechanism of the vasosuppressive and cardiotropic effects of VS-1, we acknowledge several limitations in our investigation. Although the VS-1 concentrations (5–10 nM) are comparable to plasma CgA concentrations in patients with severe heart failure, the pharmacological conditions may be totally different given the generalized severe endothelial dysfunction in these patients. Also, the functional properties of aortic endothelial cells differ significantly from those of EE cells (7). However, this limitation is diminished by the agreement between the experiments on BAE-1 cells and the experiments on papillary muscle: both concur that the NO-PI3K pathway is involved in the VS-1 effect.
The present results, demonstrating that the antiadrenergic effect induced by VS-1 is due to a Ca2+-independent PI3K-dependent endothelial release of NO, rather than a direct action on cardiac cells, provide an unifying working hypothesis to explain at the cellular level the mechanism of the cardiac inotropic effect of VS-1 and the vasodilator effects observed in some vascular preparations (1-3, 5).
This work was supported by grants (ex-60% 2005 and 2006 to G. Alloatti) from the Italian Ministry of University and Research. R. Ramella was the recipient of research funds from Regione Piemonte (2003 and 2004).
The authors thank Prof. A. Corti (Dept. of Biological and Technological Research, San Raffaele H Scientific Institute, Milan, Italy) for generously providing VS-1, Dr. S. Antoniotti and Prof. L. Munaron for technical assistance and useful discussion of the BAE-1 cell results, and Dr. Mary Ann McIntosh for proofreading the English manuscript.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 by the American Physiological Society