Calcitonin Gene-Related Peptide In Vivo Positive Inotropy Is Attributable to Regional Sympatho-Stimulation and Is Blunted in Congestive Hear
http://www.100md.com
循环研究杂志 2005年第2期
Division of Cardiology (T.K., R.H.H., D.F.B., C.G.T., P.R.F., D.A.K., N.P.), Department of Medicine, Johns Hopkins Medical Institutions
Baltimore, Md
the Department of Pharmacology (D.B.H., J.L.A.), College of Medicine, East Tennessee State University, Johnson City.
Abstract
Calcitonin gene-related peptide (CGRP) is a nonadrenergic/noncholinergic (NANC) peptide with vasodilatative/inotropic action that may benefit the failing heart. However, precise mechanisms for its in vivo inotropic action remain unclear. To assess this, dogs with normal or failing (sustained tachypacing) hearts were instrumented for pressureeCdimension analysis. In control hearts, CGRP (20 pmol/kg per minute) enhanced cardiac contractility (eg, +33±4.2% in end-systolic elastance) and lowered afterload (eC14.2±2% in systemic resistance, both P<0.001). The inotropic response was markedly blunted by heart failure (+6.5±2%; P<0.001 versus control), whereas arterial dilation remained unaltered (eC19.3±5%). CGRP-positive inotropy was not attributable to reflex activation because similar changes were observed in the presence of a ganglionic blocker. However, it was fully prevented by the -receptor antagonist (timolol), identifying a dominant role of sympatho-stimulatory signaling. In control hearts, myocardial interstitial norepinephrine assessed by microdialysis almost doubled in response to CGRP infusion, whereas systemic plasma levels were unchanged. In addition, CGRP receptors were not observed in ventricular myocardium but were prominent in coronary arteries and the stellate ganglia. Ventricular myocytes isolated from normal and failing hearts displayed no inotropic response to CGRP, further supporting indirect sympatho-stimulation as the primary in vivo mechanism. In contrast, the peripheral vasodilatative capacity of CGRP was similar in femoral vascular rings from normal and failing hearts in dogs. Thus, CGRP-mediated positive inotropy is load-independent but indirect and attributable to myocardial sympathetic activation rather than receptor-coupled stimulation in canine hearts. This mechanism is suppressed in heart failure, so that afterload reduction accounts for CGRP-enhanced function in this setting.
Key Words: calcitonin gene-related peptide sympathetic efferent fibers heart failure norepinephrine contractility
Introduction
Calcitonin gene-related peptide (CGRP) is a nonadrenergic/noncholinergic (NANC) peptide with potent vasodilator activity.1eC3 Its role as a counterbalance to vascular sympathetic nerve discharge is supported by the presence of hypertension in mice lacking CGRP.4,5 Exogenous administration of CGRP to patients with congestive heart failure (CHF) increases cardiac output,6eC8 although whether this reflects cardiac inotropy or peripheral vasodilation is unknown. CGRP receptor components (receptor activity-modifying protein 1 and calcitonin receptor-like receptor, RAMP1 and CRLR, respectively) are upregulated in failure models,9 whereas circulating plasma levels are decreased,10,11 and this might suggest a potential efficacy for exogenously administered CGRP. Recently, this notion received further attention with the discovery that the reduced form of nitric oxide (HNO/NOeC) is a potent cardiac stimulant in normal and failing hearts, which appears in part coupled to CGRP signaling.11,12
Direct evidence for cardiotropic effects of CGRP have only been obtained in isolated hearts and tissues.13 CGRP increases atrial contractility in various species including humans by stimulating specific myocardial CGRP receptors coupled to adenylate cyclase.14eC17 CGRP effect on ventricular contractility is less clear, because recent studies of isolated human trabeculae reported little direct response to CGRP18 or no effect at all.19 Furthermore, although mRNAs for all components of the CGRP receptor have been detected in human ventricle,18 definitive evidence for functional myocyte receptors remains lacking. One alternative mechanism suggested by studies performed in isolated ventricle from guinea pig is that CGRP can stimulate catecholamine release from distal sympathetic nerve terminals20 to enhance contractility. However, relevance of this mechanism in vivo, its translation to other species, and whether such release is organ-specific or coupled to local sympathetic stimulation are all unknown.
This study tested the hypotheses that in vivo cardiotonic effects of CGRP are indirect, mediated by -adrenergic neural-dependent mechanisms rather than direct myocyte interaction, and decreased in failing hearts. Studies were conducted in conscious dogs chronically instrumented for pressureeCdimension analyses, and data were measured in normal hearts and in those with CHF induced by sustained tachypacing. We report for the first time to our knowledge that in vivo CGRP-mediated inotropy is primarily attributable to local cardiac -stimulation rather than direct CGRP receptor/agonist signaling on myocytes, and is markedly downregulated in CHF. This contrasts to a direct CGRP capacity to dilate, in vivo arteries and ex vivo vascular rings, that is preserved in heart failure (HF). These data provide important new insights regarding the nature of CGRP modulation of the intact heart and its influence in late-stage CHF.
Materials and Methods
In Vivo Experimental Preparation and Protocol
Adult male mongrel dogs (22 to 25 kg) were chronically instrumented for pressureeCdimension analysis as described.11,21 Animals were anesthetized with 1% to 2% halothane after induction with sodium thiopental (10 to 20 mg/kg, intravenous). The surgical/experimental animal protocol was approved by the Johns Hopkins University Animal Care and Use Committee. The surgical preparation involved placement of an left ventricle (LV) micromanometer (P22; Konigsberg Instruments, Pasadena, Calif), sonomicrometers to measure anteroposterior LV dimension, an inferior vena caval perivascular occluder to alter cardiac preload, aortic pressure catheter, ultrasound coronary flow probe (proximal circumflex artery), and epicardial-pacing electrodes for atrial pacing. Cardiac failure was induced by rapid ventricular pacing for 3 weeks as described.11,21
Cardiovascular effects of CGRP were assessed in 14 normal control dogs and 6 with CHF. Data were acquired in conscious animals, standing quietly in a sling, with ventricular pacing suspended at least 30 minutes before the study in dogs with HF. CGRP (4 to 40 pmol/kg/min x 10 to 20 minutes) was infused intravenously, with heart rate (HR) maintained constant by atrial pacing (140 to 150 beats per minute). These rates were needed to assure matching before and after CGRP, and between control and failing animals. To test the role of baroreflex activation on CGRP effects, studies were performed in the presence of ganglionic blockade (hexamethonium chloride 5 mg/kg every 15 minutes intravenous; n=5). To assess the role of -receptor stimulation, studies were also performed in animals treated with timolol (1 to 2 mg/kg every 30 minutes intravenous; n=6).
Hemodynamic data were digitized at 250 Hz. Steady-state parameters were measured from data averaged from 10 to 20 consecutive beats, whereas data collected during transient inferior vena cava occlusion were used to determine pressureeCdimension relations. These relations strongly correlate with results from pressureeCvolume data in normal and failing hearts, as previously validated.21 Cardiovascular function was assessed by stroke dimension, fractional shortening (stroke dimension/end-diastolic dimension [EDD]), estimated cardiac output (stroke dimensionxHR), peak rate of pressure rise (dP/dtmax), end-systolic elastance (Ees, slope of end-systolic pressureeCdimension relation), the slope of dP/dtmaxeCEDD relation (DEDD),22 prerecruitable stroke work (based on dimension-data), estimated arterial elastance (end systolic pressure/stroke dimension), and estimated total resistance (stroke dimensionxHR/mean aortic pressure). Ees, DEDD, and prerecruitable stroke work provide load-insensitive contractility measures.
Isolated Myocyte Studies
Adult canine ventricular myocytes were isolated from freshly excised LVs of normal (n=3) or CHF (n=3) dogs. Hearts were removed under ice-cold cardioplegia (100 mEq K+; Plegisol, Abbott Labs), and a section of the LV was dissected and perfused at constant flow (25 mL/min) and pressure (90 mm Hg) with warmed (37°C) calcium-free Krebs-Henseleit (K-H) solution, followed by EGTA-free K-H containing collagenase (type I, 178 U/mL; Worthington Biochem) and protease (type XIV, 0.12 mg/mL). Perfusate was then switched to a modified Tyrode solution containing 125 eol/L Ca2+ for 10 minutes, and then heart tissue was mechanically disaggregated. All solutions were oxygenated with 95%O2eC5%CO2 and warmed to 37°C. Cells were imaged with an inverted microscope equipped for simultaneous assessment of sarcomere shortening (IonOptix) and INDO-1AM fluorescence to measure the calcium transient.
Plasma CGRP Assay
Arterial, venous, and coronary sinus blood plasma were sampled and analyzed for CGRP concentration by RIA following manufacturer’s instructions (Peninsula Labs). CGRP antiserum, code RAS 6012, was used, with a dynamic range of 1 to 128 pg per 300 e蘈.23
Cardiac Interstitial Fluid Norepinephrine
Interstitial myocardial norepinephrine in response to CGRP was determined by microdialysis in 4 normal anesthetized dogs. Eight additional animals served as time controls. All the procedures and experimental protocols were reviewed and approved by the East Tennessee State University Institutional Committee on Animal Care and conformed to the Animal Welfare Act according to the Public Health Policy on Humane Care and Use of Laboratory Animals. Anesthesia was induced by sodium thiopental (15 mg/kg intravenous) and maintained with isofluorane (2% inhalation). Hearts were exposed by median sternotomy, and three microdialysis probes (Clirans; Terumo, Tokyo, Japan) were inserted into the anterior LV myocardium at base, middle, and apical regions.24 The inflow capillary tube for each probe was connected via a larger deactivated silica tube to a gas-tight glass syringe filled with normal saline and perfused at 2.5 e蘈/min. Effluent (dialysate) was collected from the outflow tube in EGTA and reduced glutathione and immediately frozen (eC80°C) until analyzed.24 Data were collected at least 2 hours after surgical instrumentation to assure stable basal norepinephrine levels. CGRP (20 pmol/kg per minute) was infused into a central vein for 15 minutes, and interstitial fluid and blood from aorta and coronary sinus were collected before, during, and after drug infusion. Norepinephrine levels were determined by radio-enzymatic assay (Amersham Pharmacia Biotech).
CGRP Receptor Autoradiography
Fresh ventricular full-thickness myocardium and stellate ganglia were frozen and cut into 20-e sections, thaw-mounted onto separate chrome alum-gelatineCcoated slides, dried, and stored at eC80°C. CGRP receptors were labeled using [125I]hCGRP (2200 Ci/mmol; PerkinElmer Life Science Products, Boston, Mass) as described.25 Autoradiograhy films were processed after exposure for 3 or 7 days at 4°C, and digitized images were analyzed using image quantitation software (MCID; Imaging Research, Ontario, Canada) to convert relative optical density values to amounts of radioligand bound as fmol/mg tissue. Several measurements were made for each tissue region and averaged to yield total, nonspecific, and specific (difference between first two) binding for each animal.
Ex Vivo Arterial Response to CGRP
Studies of the vascular responsiveness to CGRP were performed in fresh canine femoral arterial segments obtained from failing and normal animal hearts at the time of euthanization. A 2-cm segment of femoral artery was dissected free of fat and connective tissue, cut into 2-mm vascular rings, and placed in ice-cold Krebs buffer (concentrations in mM: 118 NaCl, 4.7 KCL, 1.6 CaCl2, 1.2 KH2PO4, 25 NaHCO3, 1.2 MgSO4, and 11.1 glucose). Rings were suspended between two wire stirrups and immersed in organ chambers containing Krebs buffer maintained at 37°C, pH 7.4, bubbled with 95% O2eC5% CO2.
Rings were stretched to 3 grams of developed tension over a 1-hour period to optimize contractile response to KCl. A single concentration of KCl (60 mmol/L) was used to assess vascular smooth muscle viability. Rings were then washed, preconstricted with 10eC6 M phenylephrine, and exposed to increasing concentrations of CGRP (10eC11 to 10eC7 M).26 Data were collected using the MacLab system and analyzed using Dose Response Software (AD Instruments).
Chemicals
-CGRP, CGRP8eC37, timolol, hexamethonium, and protease were purchased from Sigma (St. Louis, Mo) and dissolved in saline just before use.
Statistical Analysis
Data are presented as mean±SEM. Analysis was performed by paired t test, one-way analysis of variance, or repeated measures ANOVA with a Tukey test for post-hoc comparisons.
Results
Hemodynamic Effects of CGRP in Control and Failing Hearts
Figure 1A shows an example of pressureeCdimension data before and during CGRP infusion. End-systolic pressureeCdimension relation shifted leftward and had a steeper slope, indicating a positive inotropic effect. At 20 pmol/kg per minute, this amounted to a 34.2±4.1% Ees increase and 40.9±5.0% increase in DEDD (both P<0.001; Table 1). CGRP also reduced arterial resistance (eC16.3±1.6%; P<0.001) but with little net decline in systolic pressure. For comparison, the extent of CGRP-induced arterial dilation in failing preparations (eC19±4.5% in total resistance; P<0.05; n=6) is not far from the response to a maximal dose (limited by arrhythmias) of a nitric oxide (NO) donor (diethylamine NONOate, 2.0 e/kg per minute, eC29.5±11%, n=5, P<0.05 versus base, unpublished data). Diethylamine NONOate also reduces preload by eC4%, P<0.05 versus base), whereas CGRP had minimal effect on cardiac filling (ie, venodilatation).
The response to CGRP was linear over a 10-fold dosing range (Figure 1B), with a basal arterial CGRP concentration of 7.1±0.2 pmol/L rising to 21.5±2.0 pmol/L at the mid-dose (see online data supplement available at http://circres.ahajournals.org). Near-identical concentrations were measured in coronary sinus plasma. At 20 pmol/kg per minute CGRP, mean coronary flow did not change. However, this occurred despite reduction in mean and diastolic arterial pressure (Table 1), suggesting that mean coronary flow was maintained by a CGRP-mediated coronary relaxation. This was enhanced at 40 pmol/kg per minute, with a 13±3.6% increase in coronary flow (P=0.024, n=5) despite 20% decline in mean and diastolic arterial pressures.
Basal plasma CGRP levels were lower in dogs with cardiac failure (3.5±0.1 pmol/L, P<0.001 versus control). Furthermore, failing hearts displayed a reduced cardiac inotropic response to exogenous CGRP, whereas peripheral vasodilation was preserved (Figure 1A lower panels, Figure 1D, and Table 1). Thus, improved cardiac output with exogenous CGRP reflected peripheral vasodilation and inotropy in controls but principally vasodilation in failing hearts.
Role of Reflex and -Adrenergic Receptor Stimulation
Given the potential impact of hypotension-induced reflex activation of sympathetic efferents on cardiac inotropy, we tested whether the response to exogenous CGRP was prevented by ganglionic blockade in control. The adequacy of blockade was confirmed using a previously reported method.27 Under control conditions, Ees increased after a preload decline (+baroreflex, +52.7±18.9; P<0.05), and this was eliminated by hexamethonium (+1.7±3.2%; P<0.03 versus baseline). Despite reflex blockade, CGRP-mediated inotropy and vasodilation were unchanged (Figure 2). However, blockade of downstream -adrenergic receptors by timolol eliminated the inotropic response to CGRP (Ees: +3.2±4.3%, DEDD: eC3.6±2.8%, P=NS versus timolol alone) but still had no impact on CGRP-mediated systemic vasodilation (Figure 2). The timolol dose fully blocked inotropic (Ees, +71.6±12.5% versus eC8.1±1.1%) and chronotropic (HR, +39.9±35.7% versus eC0.5±0.5%) effects of high-dose isoproterenol infusion (0.4 e/kg per minute, n=2).
To test whether CGRP triggered cardiac-specific versus diffuse sympathetic efferent activation, norepinephrine content was measured in LV myocardial interstitial fluid by microdialysis. Norepinephrine nearly doubled from 5.08±0.55 to 9.94±1.56 nmol/L with CGRP infusion (Figure 3), peaking after 10 minutes and persisting 10 minutes after the infusion was terminated. In contrast, systemic plasma norepinephrine concentration was unchanged (1.31±0.07 versus 1.49±0.10 nmol/L).
CGRP Receptor Binding Distribution
To better-define the potential tissue targets for CGRP stimulation, we performed autoradiography using [125I]CGRP (Figure 4). Binding was undetectable in left ventricular myocardium of hearts of normal dogs and those with HF. Conversely, binding was very abundant in coronary arteries (Figure 4A to 4C, Table 2) and arteries in and around the stellate ganglia in both groups (Figure 4D to 4F, Table 2). Specific CGRP binding was also present in regions of stellate ganglia that contain sympathetic efferent neurons.
CGRP Does Not Alter Inotropy of Isolated Ventricular Myocytes
The preceding findings suggested that CGRP-mediated positive inotropy was coupled to regional myocardial catecholamine release rather than direct interaction of the peptide with receptors on cardiomyocytes. Canine ventricular myocytes were isolated from normal and failing hearts and sarcomere shortening, as well as whole-cell calcium transients, were recorded in response to 10 nmol/L CGRP. Such doses had been previously reported to enhance contractility in isolated rat ventricular myocytes28 and to induce relaxation of isolated coronary arteries.29 CGRP did not alter sarcomere shortening or calcium transient in myocytes from hearts of either condition (Figure 5). Positive control data with 10 nmol/L isoproterenol are provided to confirm inotropic reserve (P<0.001 versus base, n=15).
Preserved CGRP Vasodilation in Ex Vivo Vascular Rings
The similarity of CGRP-induced systemic vasodilation in controls, HF animals, and dogs treated with ganglionic or -blockade suggested a direct and unaltered CGRP dilator effect. To test this, we exposed preconstricted femoral artery rings to incremental doses of CGRP. CGRP vasorelaxation was dose-dependent and was similar in rings from failing or control animals (Figure 6).
Discussion
This study provides the first in vivo evidence that exogenous infusion of the NANC-peptide CGRP induces a load-independent increase in cardiac contractility and reveals that this response is attributable to localized cardiac sympatho-stimulation mediated by -receptors rather than to a direct myocyte CGRP effect. This mechanism is blunted in animals with dilated cardiac failure characterized by -adrenergic downregulation. The study provides the first demonstration that CGRP induces selective arterial but no apparent venodilation, and that it does so similarly in animals with or without -blockade or cardiac failure.
CGRP Receptor Distribution and Signaling
Previous studies have established the existence of CGRP receptors in cardiac tissue, and in rodents, agonist-receptor binding triggers cAMP and positive inotropy in isolated muscle and myocytes.18,19,30 However, controversy remains whether this is species-specific and how such data translate to in vivo CGRP contractility modulation. In vitro contractile CGRP responses have been inconsistent even in the same species (eg, human) and tissue (ventricular trabeculae).18,19 The latter is paralleled by disparate data regarding the presence of CGRP receptors in human ventricular myocytes showing moderate CRLR-like immunoreactivity31 but virtually no CGRP binding sites by autoradiography.32 Co-expression of CRLR and RAMP1 leads to the formation of functional heterodimeric receptors for CGRP. Along the same line, Sugiyama et al29 first reported that immunoreactive CGRP fibers primarily targeted coronary arteries and not myocardial tissue in the canine heart, and they found minimal inotropy or chronotropy in isolated papillary muscle and sinoatrial node preparations from canine hearts in response to CGRP. Given the very similar distribution of CGRP receptors and fibers in humans and dogs, it is likely the current results are more pertinent to humans32 and, importantly, provide a novel mechanism for in vivo CGRP inotropy that does not require direct myocyte receptor/CGRP interaction. Our results further show the necessity of taking an integrative approach to properly identify neurohumoral and myocyte interactions.
CGRP and Sympatho-Stimulation
Several lines of evidence support localized sympatho-stimulation by CGRP; norepinephrine increased in cardiac interstitial fluid but not systemic plasma, and CGRP inotropy persisted despite ganglionic blockade but was inhibited by -blockade. The notion that CGRP could stimulate release of norepinephrine from isolated sympathetic nerves was first raised by Seyedi et al20 in isolated guinea pig hearts, and they also demonstrated this could be inhibited by the selective CGRP blocking peptide CGRP8eC37. The present study is the first to our knowledge to show this as the primary mechanism for CGRP inotropy in vivo.
The fibers involved with CGRP sympatho-stimulation are likely those innervating the ventricles directly, given the localized myocardial norepinephrine response. Norepinephrine can also act on sensory cardiac nerve endings and resistance vessels to attenuate CGRP release33,34; whether such a feedback loop played a role in the current study is unclear. A similar circuit has been proposed for CGRP and histamine release from local mast cells to activate H3 receptors on C-fiber endings, leading eventually to CGRP release inhibition.35 However, it would seem consistent with the anatomic localization of CGRP receptors in the stellate ganglia and with presence of CGRP-immunoreactive cells and nerves processes,36 where sympathetic neurons that innervate the heart are also found. CGRP-labeled neurons are abundant in human stellate ganglia as well, and CGRP peptide abundance is enhanced after myocardial infarction37 or in patients with congenital heart disease.38 Whether CGRP release is altered in these conditions and/or whether CGRP has alternative actions at the neuronal level beyond triggering norepinephrine release from sympathetic efferent fibers remain unknown.
CGRP and the Failing Heart
CGRP enhanced cardiac function in failing hearts largely because of systemic vasodilation, not inotropy. This is consistent with a primarily sympatho-stimulatory mechanism, because cardiac failure (and in particular the tachypacing model) is accompanied by -adrenergic downregulation.39 CGRP triggered release of interstitial norepinephrine, and norepinephrine interstitial level directly correlates with ventricular contractility.40 This CGRP paracrine effect is likely blunted in failing hearts, because previous studies have shown a reduced capacity for norepinephrine release in response to electrical stimulation in the same failure model.41 In this setting, alterations of norepinephrine-releasing mechanisms rather than reduced norepinephrine stores are likely responsible for this deficiency. Norepinephrine overflow occurring in response to tyramine was of a similar magnitude in failing and healthy preparations.41 The overall effect we obtained with exogenous CGRP in failing hearts is congruent with previous data showing attenuated LV dP/dtmax to exogenous CGRP in the same setting42 but, unlike the earlier study, rule out HR or differential loading as major explanations for this response.
Unlike CGRP-mediated sympatho-stimulation and positive inotropy, dilation of vascular smooth muscle was preserved in HF, and with ganglionic and -blockade, suggesting a direct and independent mechanism.3,43,44 As recently reviewed,3 vasorelaxant effects of CGRP are attributable to multiple mechanisms. The vascular ring results indicate direct effects independent of sympathetic nerve release, and this was similar between normal and failing hearts. Furthermore, systemic norepinephrine levels were similar before and after CGRP infusion, so substantial systemic sympatho-stimulation was not observed. The decline in basal CGRP levels in the failing heart is consistent with previous reports10,11 and may contribute to increased resting vascular tone and resistance in this disorder.
CGRP signaling has received renewed interest because of recent discoveries that nitroxyl anion (HNO/NOeC), the one electron-reduced form of NO, triggers load-independent and reflex-independent inotropy associated with elevated plasma CGRP.11,12,45 In controls, this response was inhibited by co-infusion of the CGRP receptor blocker (CGRP8eC37), but not -blockade,11,12 and nitroxyl-induced more selective venodilation. The present findings that CGRP inotropy is prevented by -blockade, blunted by CHF, and associated principally with arterial dilation all suggest that CGRP signaling does not explain the in vivo HNO/NOeC response. Additional studies are needed to test if HNO/NOeC stimulates more generalized NANC fiber signaling, involves release of alternate peptides that share CGRP cardiovascular properties, or has direct myocardial effects.
Limitations
First, the tachypacing model of cardiac failure recapitulates many important abnormalities observed in the human disease. It has admitted differences as well. More direct analysis of the present findings to human CHF would be required to establish this. Second, we did not directly assess cardiac interstitial norepinephrine release after CGRP infusion in failing preparations. This was performed given concerns over the fragility of the preparation and need for open-chest interventions to obtain these data. Further, there are existing data demonstrating that impairment of norepinephrine release from sympathetic efferents in this CHF model, supporting our proposed mechanism.41
Conclusion
Exogenous CGRP exerts dose-dependent load-independent positive inotropy in the normal in vivo canine heart attributable to local sympatho-stimulation. This "CGRPeCnorepinephrine axis" is likely to be relevant in species, such as dogs (and humans), that use -adrenergic signaling as an important "reserve" mechanism to increased demand. CGRP levels increase during exercise in humans,46 and our results not only provide novel coupling of this response to adrenergic inotropic responses in the heart but also highlight how this would be impacted by -receptor downregulation (blockade or failure). In the light of these data, afterload reduction is the primary mechanism for improved cardiac output to exogenous CGRP in HF, and increased CGRP levels are unlikely to augment contractility in this disease in which myocardial -receptor signaling is downregulated.
Acknowledgments
This study was supported by American Heart Association Beginning Grant-in-Aid 0265435U (to N.P.), a University of Tokyo fellowship grant (to T.K.), NIH grant HL54633 (to D.B.H), American Heart Association Southeast Affiliate Grant-in-Aid (to J.L.A.), and NIH grants HL-47511 and P50-HL52307 (to D.A.K.). We thank Richard S. Tunin for surgical/technical assistance and Hunter C. Champion for help with CGRP measurements.
This manuscript was sent to Richard Walsh, Consulting Editor, for review by expert referees, editorial decision, and final disposition.
References
Franco-Cereceda A, Gennari C, Nami R, Agnusdei D, Pernow J, Lundberg JM, Fischer JA. Cardiovascular effects of calcitonin gene-related peptides I and II in man. Circ Res. 1987; 60: 393eC397.
Rubino A. Non-adrenergic non-cholinergic (NANC) neural control of the atrial myocardium. Gen Pharmacol. 1993; 24: 539eC545.
Brain SD, Grant AD. Vascular actions of calcitonin gene-related peptide and adrenomedullin. Physiol Rev. 2004; 84: 903eC934.
Gangula PR, Zhao H, Supowit SC, Wimalawansa SJ, Dipette DJ, Westlund KN, Gagel RF, Yallampalli C. Increased blood pressure in alpha-calcitonin gene-related peptide/calcitonin gene knockout mice. Hypertension. 2000; 35: 470eC475.
Oh-hashi Y, Shindo T, Kurihara Y, Imai T, Wang Y, Morita H, Imai Y, Kayaba Y, Nishimatsu H, Suematsu Y, Hirata Y, Yazaki Y, Nagai R, Kuwaki T, Kurihara H. Elevated sympathetic nervous activity in mice deficient in alphaCGRP. Circ Res. 2001; 89: 983eC990.
Gennari C, Fischer JA. Cardiovascular action of calcitonin gene-related peptide in humans. Calcif Tissue Int. 1985; 37: 581eC584.
Gennari C, Nami R, Agnusdei D, Fischer JA. Improved cardiac performance with human calcitonin gene related peptide in patients with congestive heart failure. Cardiovasc Res. 1990; 24: 239eC241.
Stevenson RN, Roberts RH, Timmis AD. Calcitonin gene-related peptide: a haemodynamic study of a novel vasodilator in patients with severe chronic heart failure. Int J Cardiol. 1992; 37: 407eC414.
Cueille C, Pidoux E, de Vernejoul MC, Ventura-Clapier R, Garel JM. Increased myocardial expression of RAMP1 and RAMP3 in rats with chronic heart failure. Biochem Biophys Res Commun. 2002; 294: 340eC346.
Taquet H, Komajda M, Grenier O, Belas F, Landault C, Carayon A, Lechat P, Grosgogeat Y, Legrand JC. Plasma calcitonin gene-related peptide decreases in chronic congestive heart failure. Eur Heart J. 1992; 13: 1473eC1476.
Paolocci N, Katori T, Champion HC, St John ME, Miranda KM, Fukuto JM, Wink DA, Kass DA. Positive inotropic and lusitropic effects of HNO/NO- in failing hearts: independence from beta-adrenergic signaling. Proc Natl Acad Sci U S A. 2003; 100: 5537eC5542.
Paolocci N, Saavedra WF, Miranda KM, Martignani C, Isoda T, Hare JM, Espey MG, Fukuto JM, Feelisch M, Wink DA, Kass DA. Nitroxyl anion exerts redox-sensitive positive cardiac inotropy in vivo by calcitonin gene-related peptide signaling. Proc Natl Acad Sci U S A. 2001; 98: 10463eC10468.
Bell D, McDermott BJ. Calcitonin gene-related peptide in the cardiovascular system: characterization of receptor populations and their (patho)physiological significance. Pharmacol Rev. 1996; 48: 253eC288.
Satoh M, Oku R, Maeda A, Fujii N, Otaka A, Funakoshi S, Yajima H, Takagi H. Possible mechanisms of positive inotropic action of synthetic human calcitonin gene-related peptide in isolated rat atrium. Peptides. 1986; 7: 631eC635.
Wang X, Fiscus RR. Calcitonin gene-related peptide increases cAMP, tension, and rate in rat atria. Am J Physiol. 1989; 256: R421eCR428.
Ishikawa T, Okamura N, Saito A, Goto K. Effects of calcitonin gene-related peptide (CGRP) and isoproterenol on the contractility and adenylate cyclase activity in the rat heart. J Mol Cell Cardiol. 1987; 19: 723eC727.
Ishikawa T, Okamura N, Saito A, Masaki T, Goto K. Positive inotropic effect of calcitonin gene-related peptide mediated by cyclic AMP in guinea pig heart. Circ Res. 1988; 63: 726eC734.
Saetrum OO, Hasbak P, de Vries R, Saxena PR, Edvinsson L. Positive inotropy mediated via CGRP receptors in isolated human myocardial trabeculae. Eur J Pharmacol. 2000; 397: 373eC382.
Du XY, Schoemaker RG, Bos E, Saxena PR. Different pharmacological responses of atrium and ventricle: studies with human cardiac tissue. Eur J Pharmacol. 1994; 259: 173eC180.
Seyedi N, Maruyama R, Levi R. Bradykinin activates a cross-signaling pathway between sensory and adrenergic nerve endings in the heart: a novel mechanism of ischemic norepinephrine release J Pharmacol Exp Ther. 1999; 290: 656eC663.
Senzaki H, Isoda T, Paolocci N, Ekelund U, Hare JM, Kass DA. Improved mechanoenergetics and cardiac rest and reserve function of in vivo failing heart by calcium sensitizer EMD-57033. Circulation. 2000; 101: 1040eC1048.
Little WC. The left ventricular dP/dtmax-end-diastolic volume relation in closed-chest dogs. Circ Res. 1985; 56: 808eC815.
Onuoha GN, Nugent AM, Hunter SJ, Alpar EK, McEneaney DJ, Campbell NP, Shaw C, Buchanan KD, Nicholls DP. Neuropeptide variability in man. Eur J Clin Invest. 2000; 30: 570eC577.
Farrell DM, Wei CC, Tallaj J, Ardell JL, Armour JA, Hageman GR, Bradley WE, Dell’Italia LJ. Angiotensin II modulates catecholamine release into interstitial fluid of canine myocardium in vivo. Am J Physiol Heart Circ Physiol. 2001; 281: H813eCH822.
Chang Y, Stover SR, Hoover DB. Regional localization and abundance of calcitonin gene-related peptide receptors in guinea pig heart. J Mol Cell Cardiol. 2001; 33: 745eC754.
Quebbeman BB, Dulas D, Altman J, Homans DC, Bache RJ. Effect of calcitonin gene-related peptide on well-developed canine coronary collateral vasculature. J Cardiovasc Pharmacol. 1993; 21: 774eC780.
Kass DA, Yamazaki T, Burkhoff D, Maughan WL, Sagawa K. Determination of left ventricular end-systolic pressure-volume relationships by the conductance (volume) catheter technique. Circulation. 1986; 73: 586eC595.
Huang MH, Knight PR, III, Izzo JL, Jr. Ca2+-induced Ca2+ release involved in positive inotropic effect mediated by CGRP in ventricular myocytes. Am J Physiol. 1999; 276: R259eCR264.
Sugiyama A, Kobayashi M, Tsujimoto G, Motomura S, Hashimoto K. The first demonstration of CGRP-immunoreactive fibers in canine hearts: coronary vasodilator, inotropic and chronotropic effects of CGRP in canine isolated, blood-perfused heart preparations. Jpn J Pharmacol. 1989; 50: 421eC427.
Nakajima T, Takikawa R, Sugimoto T, Kurachi Y. Effects of calcitonin gene-related peptide on membrane currents in mammalian cardiac myocytes. Pflugers Arch. 1991; 419: 644eC650.
Hagner S, Stahl U, Knoblauch B, McGregor GP, Lang RE. Calcitonin receptor-like receptor: identification and distribution in human peripheral tissues. Cell Tissue Res. 2002; 310: 41eC50.
Coupe MO, Mak JC, Yacoub M, Oldershaw PJ, Barnes PJ. Autoradiographic mapping of calcitonin gene-related peptide receptors in human and guinea pig hearts. Circulation. 1990; 81: 741eC747.
Kawasaki H, Nuki C, Saito A, Takasaki K. Adrenergic modulation of calcitonin gene-related peptide (CGRP)-containing nerve-mediated vasodilation in the rat mesenteric resistance vessel. Brain Res. 1990; 506: 287eC290.
Amerini S, Rubino A, Mantelli L, Ledda F. Alpha-adrenoceptor modulation of the efferent function of capsaicin-sensitive sensory neurones in guinea-pig isolated atria. Br J Pharmacol. 1992; 105: 947eC953.
Imamura M, Smith NC, Garbarg M, Levi R. Histamine H3-receptor-mediated inhibition of calcitonin gene-related peptide release from cardiac C fibers. A regulatory negative-feedback loop. Circ Res. 1996; 78: 863eC869.
Ursell PC, Ren CL, Albala A, Danilo P, Jr. Nonadrenergic noncholinergic innervation. Anatomic distribution of calcitonin gene-related peptide-immunoreactive tissue in the dog heart. Circ Res. 1991; 68: 131eC140.
Roudenok V, Gutjar L, Antipova V, Rogov Y. Expression of vasoactive intestinal polypeptide and calcitonin gene-related peptide in human stellate ganglia after acute myocardial infarction. Ann Anat. 2001; 183: 341eC344.
Roudenok V, Schmitt O. Upregulation of vasoactive intestinal polypeptide (VIP) and calcitonin gene-related peptide (CGRP) expression in stellate ganglia of children with congenital cardiovascular lesions. Ann Anat. 2001; 183: 209eC212.
Vatner DE, Sato N, Ishikawa Y, Kiuchi K, Shannon RP, Vatner SF. Beta-adrenoceptor desensitization during the development of canine pacing-induced heart failure. Clin Exp Pharmacol Physiol. 1996; 23: 688eC692.
Kawada T, Yamazaki T, Akiyama T, Shishido T, Miyano H, Sato T, Sugimachi M, Alexander J, Jr., Sunagawa K. Interstitial norepinephrine level by cardiac microdialysis correlates with ventricular contractility. Am J Physiol. 1997; 273: H1107eCH1112.
Cardinal R, Nadeau R, Laurent C, Boudreau G, Armour JA. Reduced capacity of cardiac efferent sympathetic neurons to release noradrenaline and modify cardiac function in tachycardia-induced canine heart failure. Can J Physiol Pharmacol. 1996; 74: 1070eC1078.
Shen YT, Mallee JJ, Handt LK, Gilberto DB, Lynch JJ, Jr., Hargreaves RJ, Koblan KS, Gould RJ, Kane SA. Effects of inhibition of alpha-CGRP receptors on cardiac and peripheral vascular dynamics in conscious dogs with chronic heart failure. J Cardiovasc Pharmacol. 2003; 42: 656eC661.
Quayle JM, Bonev AD, Brayden JE, Nelson MT. Calcitonin gene-related peptide activated ATP-sensitive K+ currents in rabbit arterial smooth muscle via protein kinase A. J Physiol. 1994; 475: 9eC13.
Zhang X, Hintze TH. cAMP signal transduction cascade, a novel pathway for the regulation of endothelial nitric oxide production in coronary blood vessels. Arterioscler Thromb Vasc Biol. 2001; 21: 797eC803.
Wink DA, Miranda KM, Katori T, Mancardi D, Thomas DD, Ridnour L, Espey MG, Feelisch M, Colton CA, Fukuto JM, Pagliaro P, Kass DA, Paolocci N. Orthogonal properties of the redox siblings nitroxyl and nitric oxide in the cardiovascular system: a novel redox paradigm. Am J Physiol Heart Circ Physiol. 2003; 285: H2264eCH2276.
Lind H, Brudin L, Lindholm L, Edvinsson L. Different levels of sensory neuropeptides (calcitonin gene-related peptide and substance P) during and after exercise in man. Clin Physiol. 1996; 16: 73eC82.(Tatsuo Katori, Donald B. )
Baltimore, Md
the Department of Pharmacology (D.B.H., J.L.A.), College of Medicine, East Tennessee State University, Johnson City.
Abstract
Calcitonin gene-related peptide (CGRP) is a nonadrenergic/noncholinergic (NANC) peptide with vasodilatative/inotropic action that may benefit the failing heart. However, precise mechanisms for its in vivo inotropic action remain unclear. To assess this, dogs with normal or failing (sustained tachypacing) hearts were instrumented for pressureeCdimension analysis. In control hearts, CGRP (20 pmol/kg per minute) enhanced cardiac contractility (eg, +33±4.2% in end-systolic elastance) and lowered afterload (eC14.2±2% in systemic resistance, both P<0.001). The inotropic response was markedly blunted by heart failure (+6.5±2%; P<0.001 versus control), whereas arterial dilation remained unaltered (eC19.3±5%). CGRP-positive inotropy was not attributable to reflex activation because similar changes were observed in the presence of a ganglionic blocker. However, it was fully prevented by the -receptor antagonist (timolol), identifying a dominant role of sympatho-stimulatory signaling. In control hearts, myocardial interstitial norepinephrine assessed by microdialysis almost doubled in response to CGRP infusion, whereas systemic plasma levels were unchanged. In addition, CGRP receptors were not observed in ventricular myocardium but were prominent in coronary arteries and the stellate ganglia. Ventricular myocytes isolated from normal and failing hearts displayed no inotropic response to CGRP, further supporting indirect sympatho-stimulation as the primary in vivo mechanism. In contrast, the peripheral vasodilatative capacity of CGRP was similar in femoral vascular rings from normal and failing hearts in dogs. Thus, CGRP-mediated positive inotropy is load-independent but indirect and attributable to myocardial sympathetic activation rather than receptor-coupled stimulation in canine hearts. This mechanism is suppressed in heart failure, so that afterload reduction accounts for CGRP-enhanced function in this setting.
Key Words: calcitonin gene-related peptide sympathetic efferent fibers heart failure norepinephrine contractility
Introduction
Calcitonin gene-related peptide (CGRP) is a nonadrenergic/noncholinergic (NANC) peptide with potent vasodilator activity.1eC3 Its role as a counterbalance to vascular sympathetic nerve discharge is supported by the presence of hypertension in mice lacking CGRP.4,5 Exogenous administration of CGRP to patients with congestive heart failure (CHF) increases cardiac output,6eC8 although whether this reflects cardiac inotropy or peripheral vasodilation is unknown. CGRP receptor components (receptor activity-modifying protein 1 and calcitonin receptor-like receptor, RAMP1 and CRLR, respectively) are upregulated in failure models,9 whereas circulating plasma levels are decreased,10,11 and this might suggest a potential efficacy for exogenously administered CGRP. Recently, this notion received further attention with the discovery that the reduced form of nitric oxide (HNO/NOeC) is a potent cardiac stimulant in normal and failing hearts, which appears in part coupled to CGRP signaling.11,12
Direct evidence for cardiotropic effects of CGRP have only been obtained in isolated hearts and tissues.13 CGRP increases atrial contractility in various species including humans by stimulating specific myocardial CGRP receptors coupled to adenylate cyclase.14eC17 CGRP effect on ventricular contractility is less clear, because recent studies of isolated human trabeculae reported little direct response to CGRP18 or no effect at all.19 Furthermore, although mRNAs for all components of the CGRP receptor have been detected in human ventricle,18 definitive evidence for functional myocyte receptors remains lacking. One alternative mechanism suggested by studies performed in isolated ventricle from guinea pig is that CGRP can stimulate catecholamine release from distal sympathetic nerve terminals20 to enhance contractility. However, relevance of this mechanism in vivo, its translation to other species, and whether such release is organ-specific or coupled to local sympathetic stimulation are all unknown.
This study tested the hypotheses that in vivo cardiotonic effects of CGRP are indirect, mediated by -adrenergic neural-dependent mechanisms rather than direct myocyte interaction, and decreased in failing hearts. Studies were conducted in conscious dogs chronically instrumented for pressureeCdimension analyses, and data were measured in normal hearts and in those with CHF induced by sustained tachypacing. We report for the first time to our knowledge that in vivo CGRP-mediated inotropy is primarily attributable to local cardiac -stimulation rather than direct CGRP receptor/agonist signaling on myocytes, and is markedly downregulated in CHF. This contrasts to a direct CGRP capacity to dilate, in vivo arteries and ex vivo vascular rings, that is preserved in heart failure (HF). These data provide important new insights regarding the nature of CGRP modulation of the intact heart and its influence in late-stage CHF.
Materials and Methods
In Vivo Experimental Preparation and Protocol
Adult male mongrel dogs (22 to 25 kg) were chronically instrumented for pressureeCdimension analysis as described.11,21 Animals were anesthetized with 1% to 2% halothane after induction with sodium thiopental (10 to 20 mg/kg, intravenous). The surgical/experimental animal protocol was approved by the Johns Hopkins University Animal Care and Use Committee. The surgical preparation involved placement of an left ventricle (LV) micromanometer (P22; Konigsberg Instruments, Pasadena, Calif), sonomicrometers to measure anteroposterior LV dimension, an inferior vena caval perivascular occluder to alter cardiac preload, aortic pressure catheter, ultrasound coronary flow probe (proximal circumflex artery), and epicardial-pacing electrodes for atrial pacing. Cardiac failure was induced by rapid ventricular pacing for 3 weeks as described.11,21
Cardiovascular effects of CGRP were assessed in 14 normal control dogs and 6 with CHF. Data were acquired in conscious animals, standing quietly in a sling, with ventricular pacing suspended at least 30 minutes before the study in dogs with HF. CGRP (4 to 40 pmol/kg/min x 10 to 20 minutes) was infused intravenously, with heart rate (HR) maintained constant by atrial pacing (140 to 150 beats per minute). These rates were needed to assure matching before and after CGRP, and between control and failing animals. To test the role of baroreflex activation on CGRP effects, studies were performed in the presence of ganglionic blockade (hexamethonium chloride 5 mg/kg every 15 minutes intravenous; n=5). To assess the role of -receptor stimulation, studies were also performed in animals treated with timolol (1 to 2 mg/kg every 30 minutes intravenous; n=6).
Hemodynamic data were digitized at 250 Hz. Steady-state parameters were measured from data averaged from 10 to 20 consecutive beats, whereas data collected during transient inferior vena cava occlusion were used to determine pressureeCdimension relations. These relations strongly correlate with results from pressureeCvolume data in normal and failing hearts, as previously validated.21 Cardiovascular function was assessed by stroke dimension, fractional shortening (stroke dimension/end-diastolic dimension [EDD]), estimated cardiac output (stroke dimensionxHR), peak rate of pressure rise (dP/dtmax), end-systolic elastance (Ees, slope of end-systolic pressureeCdimension relation), the slope of dP/dtmaxeCEDD relation (DEDD),22 prerecruitable stroke work (based on dimension-data), estimated arterial elastance (end systolic pressure/stroke dimension), and estimated total resistance (stroke dimensionxHR/mean aortic pressure). Ees, DEDD, and prerecruitable stroke work provide load-insensitive contractility measures.
Isolated Myocyte Studies
Adult canine ventricular myocytes were isolated from freshly excised LVs of normal (n=3) or CHF (n=3) dogs. Hearts were removed under ice-cold cardioplegia (100 mEq K+; Plegisol, Abbott Labs), and a section of the LV was dissected and perfused at constant flow (25 mL/min) and pressure (90 mm Hg) with warmed (37°C) calcium-free Krebs-Henseleit (K-H) solution, followed by EGTA-free K-H containing collagenase (type I, 178 U/mL; Worthington Biochem) and protease (type XIV, 0.12 mg/mL). Perfusate was then switched to a modified Tyrode solution containing 125 eol/L Ca2+ for 10 minutes, and then heart tissue was mechanically disaggregated. All solutions were oxygenated with 95%O2eC5%CO2 and warmed to 37°C. Cells were imaged with an inverted microscope equipped for simultaneous assessment of sarcomere shortening (IonOptix) and INDO-1AM fluorescence to measure the calcium transient.
Plasma CGRP Assay
Arterial, venous, and coronary sinus blood plasma were sampled and analyzed for CGRP concentration by RIA following manufacturer’s instructions (Peninsula Labs). CGRP antiserum, code RAS 6012, was used, with a dynamic range of 1 to 128 pg per 300 e蘈.23
Cardiac Interstitial Fluid Norepinephrine
Interstitial myocardial norepinephrine in response to CGRP was determined by microdialysis in 4 normal anesthetized dogs. Eight additional animals served as time controls. All the procedures and experimental protocols were reviewed and approved by the East Tennessee State University Institutional Committee on Animal Care and conformed to the Animal Welfare Act according to the Public Health Policy on Humane Care and Use of Laboratory Animals. Anesthesia was induced by sodium thiopental (15 mg/kg intravenous) and maintained with isofluorane (2% inhalation). Hearts were exposed by median sternotomy, and three microdialysis probes (Clirans; Terumo, Tokyo, Japan) were inserted into the anterior LV myocardium at base, middle, and apical regions.24 The inflow capillary tube for each probe was connected via a larger deactivated silica tube to a gas-tight glass syringe filled with normal saline and perfused at 2.5 e蘈/min. Effluent (dialysate) was collected from the outflow tube in EGTA and reduced glutathione and immediately frozen (eC80°C) until analyzed.24 Data were collected at least 2 hours after surgical instrumentation to assure stable basal norepinephrine levels. CGRP (20 pmol/kg per minute) was infused into a central vein for 15 minutes, and interstitial fluid and blood from aorta and coronary sinus were collected before, during, and after drug infusion. Norepinephrine levels were determined by radio-enzymatic assay (Amersham Pharmacia Biotech).
CGRP Receptor Autoradiography
Fresh ventricular full-thickness myocardium and stellate ganglia were frozen and cut into 20-e sections, thaw-mounted onto separate chrome alum-gelatineCcoated slides, dried, and stored at eC80°C. CGRP receptors were labeled using [125I]hCGRP (2200 Ci/mmol; PerkinElmer Life Science Products, Boston, Mass) as described.25 Autoradiograhy films were processed after exposure for 3 or 7 days at 4°C, and digitized images were analyzed using image quantitation software (MCID; Imaging Research, Ontario, Canada) to convert relative optical density values to amounts of radioligand bound as fmol/mg tissue. Several measurements were made for each tissue region and averaged to yield total, nonspecific, and specific (difference between first two) binding for each animal.
Ex Vivo Arterial Response to CGRP
Studies of the vascular responsiveness to CGRP were performed in fresh canine femoral arterial segments obtained from failing and normal animal hearts at the time of euthanization. A 2-cm segment of femoral artery was dissected free of fat and connective tissue, cut into 2-mm vascular rings, and placed in ice-cold Krebs buffer (concentrations in mM: 118 NaCl, 4.7 KCL, 1.6 CaCl2, 1.2 KH2PO4, 25 NaHCO3, 1.2 MgSO4, and 11.1 glucose). Rings were suspended between two wire stirrups and immersed in organ chambers containing Krebs buffer maintained at 37°C, pH 7.4, bubbled with 95% O2eC5% CO2.
Rings were stretched to 3 grams of developed tension over a 1-hour period to optimize contractile response to KCl. A single concentration of KCl (60 mmol/L) was used to assess vascular smooth muscle viability. Rings were then washed, preconstricted with 10eC6 M phenylephrine, and exposed to increasing concentrations of CGRP (10eC11 to 10eC7 M).26 Data were collected using the MacLab system and analyzed using Dose Response Software (AD Instruments).
Chemicals
-CGRP, CGRP8eC37, timolol, hexamethonium, and protease were purchased from Sigma (St. Louis, Mo) and dissolved in saline just before use.
Statistical Analysis
Data are presented as mean±SEM. Analysis was performed by paired t test, one-way analysis of variance, or repeated measures ANOVA with a Tukey test for post-hoc comparisons.
Results
Hemodynamic Effects of CGRP in Control and Failing Hearts
Figure 1A shows an example of pressureeCdimension data before and during CGRP infusion. End-systolic pressureeCdimension relation shifted leftward and had a steeper slope, indicating a positive inotropic effect. At 20 pmol/kg per minute, this amounted to a 34.2±4.1% Ees increase and 40.9±5.0% increase in DEDD (both P<0.001; Table 1). CGRP also reduced arterial resistance (eC16.3±1.6%; P<0.001) but with little net decline in systolic pressure. For comparison, the extent of CGRP-induced arterial dilation in failing preparations (eC19±4.5% in total resistance; P<0.05; n=6) is not far from the response to a maximal dose (limited by arrhythmias) of a nitric oxide (NO) donor (diethylamine NONOate, 2.0 e/kg per minute, eC29.5±11%, n=5, P<0.05 versus base, unpublished data). Diethylamine NONOate also reduces preload by eC4%, P<0.05 versus base), whereas CGRP had minimal effect on cardiac filling (ie, venodilatation).
The response to CGRP was linear over a 10-fold dosing range (Figure 1B), with a basal arterial CGRP concentration of 7.1±0.2 pmol/L rising to 21.5±2.0 pmol/L at the mid-dose (see online data supplement available at http://circres.ahajournals.org). Near-identical concentrations were measured in coronary sinus plasma. At 20 pmol/kg per minute CGRP, mean coronary flow did not change. However, this occurred despite reduction in mean and diastolic arterial pressure (Table 1), suggesting that mean coronary flow was maintained by a CGRP-mediated coronary relaxation. This was enhanced at 40 pmol/kg per minute, with a 13±3.6% increase in coronary flow (P=0.024, n=5) despite 20% decline in mean and diastolic arterial pressures.
Basal plasma CGRP levels were lower in dogs with cardiac failure (3.5±0.1 pmol/L, P<0.001 versus control). Furthermore, failing hearts displayed a reduced cardiac inotropic response to exogenous CGRP, whereas peripheral vasodilation was preserved (Figure 1A lower panels, Figure 1D, and Table 1). Thus, improved cardiac output with exogenous CGRP reflected peripheral vasodilation and inotropy in controls but principally vasodilation in failing hearts.
Role of Reflex and -Adrenergic Receptor Stimulation
Given the potential impact of hypotension-induced reflex activation of sympathetic efferents on cardiac inotropy, we tested whether the response to exogenous CGRP was prevented by ganglionic blockade in control. The adequacy of blockade was confirmed using a previously reported method.27 Under control conditions, Ees increased after a preload decline (+baroreflex, +52.7±18.9; P<0.05), and this was eliminated by hexamethonium (+1.7±3.2%; P<0.03 versus baseline). Despite reflex blockade, CGRP-mediated inotropy and vasodilation were unchanged (Figure 2). However, blockade of downstream -adrenergic receptors by timolol eliminated the inotropic response to CGRP (Ees: +3.2±4.3%, DEDD: eC3.6±2.8%, P=NS versus timolol alone) but still had no impact on CGRP-mediated systemic vasodilation (Figure 2). The timolol dose fully blocked inotropic (Ees, +71.6±12.5% versus eC8.1±1.1%) and chronotropic (HR, +39.9±35.7% versus eC0.5±0.5%) effects of high-dose isoproterenol infusion (0.4 e/kg per minute, n=2).
To test whether CGRP triggered cardiac-specific versus diffuse sympathetic efferent activation, norepinephrine content was measured in LV myocardial interstitial fluid by microdialysis. Norepinephrine nearly doubled from 5.08±0.55 to 9.94±1.56 nmol/L with CGRP infusion (Figure 3), peaking after 10 minutes and persisting 10 minutes after the infusion was terminated. In contrast, systemic plasma norepinephrine concentration was unchanged (1.31±0.07 versus 1.49±0.10 nmol/L).
CGRP Receptor Binding Distribution
To better-define the potential tissue targets for CGRP stimulation, we performed autoradiography using [125I]CGRP (Figure 4). Binding was undetectable in left ventricular myocardium of hearts of normal dogs and those with HF. Conversely, binding was very abundant in coronary arteries (Figure 4A to 4C, Table 2) and arteries in and around the stellate ganglia in both groups (Figure 4D to 4F, Table 2). Specific CGRP binding was also present in regions of stellate ganglia that contain sympathetic efferent neurons.
CGRP Does Not Alter Inotropy of Isolated Ventricular Myocytes
The preceding findings suggested that CGRP-mediated positive inotropy was coupled to regional myocardial catecholamine release rather than direct interaction of the peptide with receptors on cardiomyocytes. Canine ventricular myocytes were isolated from normal and failing hearts and sarcomere shortening, as well as whole-cell calcium transients, were recorded in response to 10 nmol/L CGRP. Such doses had been previously reported to enhance contractility in isolated rat ventricular myocytes28 and to induce relaxation of isolated coronary arteries.29 CGRP did not alter sarcomere shortening or calcium transient in myocytes from hearts of either condition (Figure 5). Positive control data with 10 nmol/L isoproterenol are provided to confirm inotropic reserve (P<0.001 versus base, n=15).
Preserved CGRP Vasodilation in Ex Vivo Vascular Rings
The similarity of CGRP-induced systemic vasodilation in controls, HF animals, and dogs treated with ganglionic or -blockade suggested a direct and unaltered CGRP dilator effect. To test this, we exposed preconstricted femoral artery rings to incremental doses of CGRP. CGRP vasorelaxation was dose-dependent and was similar in rings from failing or control animals (Figure 6).
Discussion
This study provides the first in vivo evidence that exogenous infusion of the NANC-peptide CGRP induces a load-independent increase in cardiac contractility and reveals that this response is attributable to localized cardiac sympatho-stimulation mediated by -receptors rather than to a direct myocyte CGRP effect. This mechanism is blunted in animals with dilated cardiac failure characterized by -adrenergic downregulation. The study provides the first demonstration that CGRP induces selective arterial but no apparent venodilation, and that it does so similarly in animals with or without -blockade or cardiac failure.
CGRP Receptor Distribution and Signaling
Previous studies have established the existence of CGRP receptors in cardiac tissue, and in rodents, agonist-receptor binding triggers cAMP and positive inotropy in isolated muscle and myocytes.18,19,30 However, controversy remains whether this is species-specific and how such data translate to in vivo CGRP contractility modulation. In vitro contractile CGRP responses have been inconsistent even in the same species (eg, human) and tissue (ventricular trabeculae).18,19 The latter is paralleled by disparate data regarding the presence of CGRP receptors in human ventricular myocytes showing moderate CRLR-like immunoreactivity31 but virtually no CGRP binding sites by autoradiography.32 Co-expression of CRLR and RAMP1 leads to the formation of functional heterodimeric receptors for CGRP. Along the same line, Sugiyama et al29 first reported that immunoreactive CGRP fibers primarily targeted coronary arteries and not myocardial tissue in the canine heart, and they found minimal inotropy or chronotropy in isolated papillary muscle and sinoatrial node preparations from canine hearts in response to CGRP. Given the very similar distribution of CGRP receptors and fibers in humans and dogs, it is likely the current results are more pertinent to humans32 and, importantly, provide a novel mechanism for in vivo CGRP inotropy that does not require direct myocyte receptor/CGRP interaction. Our results further show the necessity of taking an integrative approach to properly identify neurohumoral and myocyte interactions.
CGRP and Sympatho-Stimulation
Several lines of evidence support localized sympatho-stimulation by CGRP; norepinephrine increased in cardiac interstitial fluid but not systemic plasma, and CGRP inotropy persisted despite ganglionic blockade but was inhibited by -blockade. The notion that CGRP could stimulate release of norepinephrine from isolated sympathetic nerves was first raised by Seyedi et al20 in isolated guinea pig hearts, and they also demonstrated this could be inhibited by the selective CGRP blocking peptide CGRP8eC37. The present study is the first to our knowledge to show this as the primary mechanism for CGRP inotropy in vivo.
The fibers involved with CGRP sympatho-stimulation are likely those innervating the ventricles directly, given the localized myocardial norepinephrine response. Norepinephrine can also act on sensory cardiac nerve endings and resistance vessels to attenuate CGRP release33,34; whether such a feedback loop played a role in the current study is unclear. A similar circuit has been proposed for CGRP and histamine release from local mast cells to activate H3 receptors on C-fiber endings, leading eventually to CGRP release inhibition.35 However, it would seem consistent with the anatomic localization of CGRP receptors in the stellate ganglia and with presence of CGRP-immunoreactive cells and nerves processes,36 where sympathetic neurons that innervate the heart are also found. CGRP-labeled neurons are abundant in human stellate ganglia as well, and CGRP peptide abundance is enhanced after myocardial infarction37 or in patients with congenital heart disease.38 Whether CGRP release is altered in these conditions and/or whether CGRP has alternative actions at the neuronal level beyond triggering norepinephrine release from sympathetic efferent fibers remain unknown.
CGRP and the Failing Heart
CGRP enhanced cardiac function in failing hearts largely because of systemic vasodilation, not inotropy. This is consistent with a primarily sympatho-stimulatory mechanism, because cardiac failure (and in particular the tachypacing model) is accompanied by -adrenergic downregulation.39 CGRP triggered release of interstitial norepinephrine, and norepinephrine interstitial level directly correlates with ventricular contractility.40 This CGRP paracrine effect is likely blunted in failing hearts, because previous studies have shown a reduced capacity for norepinephrine release in response to electrical stimulation in the same failure model.41 In this setting, alterations of norepinephrine-releasing mechanisms rather than reduced norepinephrine stores are likely responsible for this deficiency. Norepinephrine overflow occurring in response to tyramine was of a similar magnitude in failing and healthy preparations.41 The overall effect we obtained with exogenous CGRP in failing hearts is congruent with previous data showing attenuated LV dP/dtmax to exogenous CGRP in the same setting42 but, unlike the earlier study, rule out HR or differential loading as major explanations for this response.
Unlike CGRP-mediated sympatho-stimulation and positive inotropy, dilation of vascular smooth muscle was preserved in HF, and with ganglionic and -blockade, suggesting a direct and independent mechanism.3,43,44 As recently reviewed,3 vasorelaxant effects of CGRP are attributable to multiple mechanisms. The vascular ring results indicate direct effects independent of sympathetic nerve release, and this was similar between normal and failing hearts. Furthermore, systemic norepinephrine levels were similar before and after CGRP infusion, so substantial systemic sympatho-stimulation was not observed. The decline in basal CGRP levels in the failing heart is consistent with previous reports10,11 and may contribute to increased resting vascular tone and resistance in this disorder.
CGRP signaling has received renewed interest because of recent discoveries that nitroxyl anion (HNO/NOeC), the one electron-reduced form of NO, triggers load-independent and reflex-independent inotropy associated with elevated plasma CGRP.11,12,45 In controls, this response was inhibited by co-infusion of the CGRP receptor blocker (CGRP8eC37), but not -blockade,11,12 and nitroxyl-induced more selective venodilation. The present findings that CGRP inotropy is prevented by -blockade, blunted by CHF, and associated principally with arterial dilation all suggest that CGRP signaling does not explain the in vivo HNO/NOeC response. Additional studies are needed to test if HNO/NOeC stimulates more generalized NANC fiber signaling, involves release of alternate peptides that share CGRP cardiovascular properties, or has direct myocardial effects.
Limitations
First, the tachypacing model of cardiac failure recapitulates many important abnormalities observed in the human disease. It has admitted differences as well. More direct analysis of the present findings to human CHF would be required to establish this. Second, we did not directly assess cardiac interstitial norepinephrine release after CGRP infusion in failing preparations. This was performed given concerns over the fragility of the preparation and need for open-chest interventions to obtain these data. Further, there are existing data demonstrating that impairment of norepinephrine release from sympathetic efferents in this CHF model, supporting our proposed mechanism.41
Conclusion
Exogenous CGRP exerts dose-dependent load-independent positive inotropy in the normal in vivo canine heart attributable to local sympatho-stimulation. This "CGRPeCnorepinephrine axis" is likely to be relevant in species, such as dogs (and humans), that use -adrenergic signaling as an important "reserve" mechanism to increased demand. CGRP levels increase during exercise in humans,46 and our results not only provide novel coupling of this response to adrenergic inotropic responses in the heart but also highlight how this would be impacted by -receptor downregulation (blockade or failure). In the light of these data, afterload reduction is the primary mechanism for improved cardiac output to exogenous CGRP in HF, and increased CGRP levels are unlikely to augment contractility in this disease in which myocardial -receptor signaling is downregulated.
Acknowledgments
This study was supported by American Heart Association Beginning Grant-in-Aid 0265435U (to N.P.), a University of Tokyo fellowship grant (to T.K.), NIH grant HL54633 (to D.B.H), American Heart Association Southeast Affiliate Grant-in-Aid (to J.L.A.), and NIH grants HL-47511 and P50-HL52307 (to D.A.K.). We thank Richard S. Tunin for surgical/technical assistance and Hunter C. Champion for help with CGRP measurements.
This manuscript was sent to Richard Walsh, Consulting Editor, for review by expert referees, editorial decision, and final disposition.
References
Franco-Cereceda A, Gennari C, Nami R, Agnusdei D, Pernow J, Lundberg JM, Fischer JA. Cardiovascular effects of calcitonin gene-related peptides I and II in man. Circ Res. 1987; 60: 393eC397.
Rubino A. Non-adrenergic non-cholinergic (NANC) neural control of the atrial myocardium. Gen Pharmacol. 1993; 24: 539eC545.
Brain SD, Grant AD. Vascular actions of calcitonin gene-related peptide and adrenomedullin. Physiol Rev. 2004; 84: 903eC934.
Gangula PR, Zhao H, Supowit SC, Wimalawansa SJ, Dipette DJ, Westlund KN, Gagel RF, Yallampalli C. Increased blood pressure in alpha-calcitonin gene-related peptide/calcitonin gene knockout mice. Hypertension. 2000; 35: 470eC475.
Oh-hashi Y, Shindo T, Kurihara Y, Imai T, Wang Y, Morita H, Imai Y, Kayaba Y, Nishimatsu H, Suematsu Y, Hirata Y, Yazaki Y, Nagai R, Kuwaki T, Kurihara H. Elevated sympathetic nervous activity in mice deficient in alphaCGRP. Circ Res. 2001; 89: 983eC990.
Gennari C, Fischer JA. Cardiovascular action of calcitonin gene-related peptide in humans. Calcif Tissue Int. 1985; 37: 581eC584.
Gennari C, Nami R, Agnusdei D, Fischer JA. Improved cardiac performance with human calcitonin gene related peptide in patients with congestive heart failure. Cardiovasc Res. 1990; 24: 239eC241.
Stevenson RN, Roberts RH, Timmis AD. Calcitonin gene-related peptide: a haemodynamic study of a novel vasodilator in patients with severe chronic heart failure. Int J Cardiol. 1992; 37: 407eC414.
Cueille C, Pidoux E, de Vernejoul MC, Ventura-Clapier R, Garel JM. Increased myocardial expression of RAMP1 and RAMP3 in rats with chronic heart failure. Biochem Biophys Res Commun. 2002; 294: 340eC346.
Taquet H, Komajda M, Grenier O, Belas F, Landault C, Carayon A, Lechat P, Grosgogeat Y, Legrand JC. Plasma calcitonin gene-related peptide decreases in chronic congestive heart failure. Eur Heart J. 1992; 13: 1473eC1476.
Paolocci N, Katori T, Champion HC, St John ME, Miranda KM, Fukuto JM, Wink DA, Kass DA. Positive inotropic and lusitropic effects of HNO/NO- in failing hearts: independence from beta-adrenergic signaling. Proc Natl Acad Sci U S A. 2003; 100: 5537eC5542.
Paolocci N, Saavedra WF, Miranda KM, Martignani C, Isoda T, Hare JM, Espey MG, Fukuto JM, Feelisch M, Wink DA, Kass DA. Nitroxyl anion exerts redox-sensitive positive cardiac inotropy in vivo by calcitonin gene-related peptide signaling. Proc Natl Acad Sci U S A. 2001; 98: 10463eC10468.
Bell D, McDermott BJ. Calcitonin gene-related peptide in the cardiovascular system: characterization of receptor populations and their (patho)physiological significance. Pharmacol Rev. 1996; 48: 253eC288.
Satoh M, Oku R, Maeda A, Fujii N, Otaka A, Funakoshi S, Yajima H, Takagi H. Possible mechanisms of positive inotropic action of synthetic human calcitonin gene-related peptide in isolated rat atrium. Peptides. 1986; 7: 631eC635.
Wang X, Fiscus RR. Calcitonin gene-related peptide increases cAMP, tension, and rate in rat atria. Am J Physiol. 1989; 256: R421eCR428.
Ishikawa T, Okamura N, Saito A, Goto K. Effects of calcitonin gene-related peptide (CGRP) and isoproterenol on the contractility and adenylate cyclase activity in the rat heart. J Mol Cell Cardiol. 1987; 19: 723eC727.
Ishikawa T, Okamura N, Saito A, Masaki T, Goto K. Positive inotropic effect of calcitonin gene-related peptide mediated by cyclic AMP in guinea pig heart. Circ Res. 1988; 63: 726eC734.
Saetrum OO, Hasbak P, de Vries R, Saxena PR, Edvinsson L. Positive inotropy mediated via CGRP receptors in isolated human myocardial trabeculae. Eur J Pharmacol. 2000; 397: 373eC382.
Du XY, Schoemaker RG, Bos E, Saxena PR. Different pharmacological responses of atrium and ventricle: studies with human cardiac tissue. Eur J Pharmacol. 1994; 259: 173eC180.
Seyedi N, Maruyama R, Levi R. Bradykinin activates a cross-signaling pathway between sensory and adrenergic nerve endings in the heart: a novel mechanism of ischemic norepinephrine release J Pharmacol Exp Ther. 1999; 290: 656eC663.
Senzaki H, Isoda T, Paolocci N, Ekelund U, Hare JM, Kass DA. Improved mechanoenergetics and cardiac rest and reserve function of in vivo failing heart by calcium sensitizer EMD-57033. Circulation. 2000; 101: 1040eC1048.
Little WC. The left ventricular dP/dtmax-end-diastolic volume relation in closed-chest dogs. Circ Res. 1985; 56: 808eC815.
Onuoha GN, Nugent AM, Hunter SJ, Alpar EK, McEneaney DJ, Campbell NP, Shaw C, Buchanan KD, Nicholls DP. Neuropeptide variability in man. Eur J Clin Invest. 2000; 30: 570eC577.
Farrell DM, Wei CC, Tallaj J, Ardell JL, Armour JA, Hageman GR, Bradley WE, Dell’Italia LJ. Angiotensin II modulates catecholamine release into interstitial fluid of canine myocardium in vivo. Am J Physiol Heart Circ Physiol. 2001; 281: H813eCH822.
Chang Y, Stover SR, Hoover DB. Regional localization and abundance of calcitonin gene-related peptide receptors in guinea pig heart. J Mol Cell Cardiol. 2001; 33: 745eC754.
Quebbeman BB, Dulas D, Altman J, Homans DC, Bache RJ. Effect of calcitonin gene-related peptide on well-developed canine coronary collateral vasculature. J Cardiovasc Pharmacol. 1993; 21: 774eC780.
Kass DA, Yamazaki T, Burkhoff D, Maughan WL, Sagawa K. Determination of left ventricular end-systolic pressure-volume relationships by the conductance (volume) catheter technique. Circulation. 1986; 73: 586eC595.
Huang MH, Knight PR, III, Izzo JL, Jr. Ca2+-induced Ca2+ release involved in positive inotropic effect mediated by CGRP in ventricular myocytes. Am J Physiol. 1999; 276: R259eCR264.
Sugiyama A, Kobayashi M, Tsujimoto G, Motomura S, Hashimoto K. The first demonstration of CGRP-immunoreactive fibers in canine hearts: coronary vasodilator, inotropic and chronotropic effects of CGRP in canine isolated, blood-perfused heart preparations. Jpn J Pharmacol. 1989; 50: 421eC427.
Nakajima T, Takikawa R, Sugimoto T, Kurachi Y. Effects of calcitonin gene-related peptide on membrane currents in mammalian cardiac myocytes. Pflugers Arch. 1991; 419: 644eC650.
Hagner S, Stahl U, Knoblauch B, McGregor GP, Lang RE. Calcitonin receptor-like receptor: identification and distribution in human peripheral tissues. Cell Tissue Res. 2002; 310: 41eC50.
Coupe MO, Mak JC, Yacoub M, Oldershaw PJ, Barnes PJ. Autoradiographic mapping of calcitonin gene-related peptide receptors in human and guinea pig hearts. Circulation. 1990; 81: 741eC747.
Kawasaki H, Nuki C, Saito A, Takasaki K. Adrenergic modulation of calcitonin gene-related peptide (CGRP)-containing nerve-mediated vasodilation in the rat mesenteric resistance vessel. Brain Res. 1990; 506: 287eC290.
Amerini S, Rubino A, Mantelli L, Ledda F. Alpha-adrenoceptor modulation of the efferent function of capsaicin-sensitive sensory neurones in guinea-pig isolated atria. Br J Pharmacol. 1992; 105: 947eC953.
Imamura M, Smith NC, Garbarg M, Levi R. Histamine H3-receptor-mediated inhibition of calcitonin gene-related peptide release from cardiac C fibers. A regulatory negative-feedback loop. Circ Res. 1996; 78: 863eC869.
Ursell PC, Ren CL, Albala A, Danilo P, Jr. Nonadrenergic noncholinergic innervation. Anatomic distribution of calcitonin gene-related peptide-immunoreactive tissue in the dog heart. Circ Res. 1991; 68: 131eC140.
Roudenok V, Gutjar L, Antipova V, Rogov Y. Expression of vasoactive intestinal polypeptide and calcitonin gene-related peptide in human stellate ganglia after acute myocardial infarction. Ann Anat. 2001; 183: 341eC344.
Roudenok V, Schmitt O. Upregulation of vasoactive intestinal polypeptide (VIP) and calcitonin gene-related peptide (CGRP) expression in stellate ganglia of children with congenital cardiovascular lesions. Ann Anat. 2001; 183: 209eC212.
Vatner DE, Sato N, Ishikawa Y, Kiuchi K, Shannon RP, Vatner SF. Beta-adrenoceptor desensitization during the development of canine pacing-induced heart failure. Clin Exp Pharmacol Physiol. 1996; 23: 688eC692.
Kawada T, Yamazaki T, Akiyama T, Shishido T, Miyano H, Sato T, Sugimachi M, Alexander J, Jr., Sunagawa K. Interstitial norepinephrine level by cardiac microdialysis correlates with ventricular contractility. Am J Physiol. 1997; 273: H1107eCH1112.
Cardinal R, Nadeau R, Laurent C, Boudreau G, Armour JA. Reduced capacity of cardiac efferent sympathetic neurons to release noradrenaline and modify cardiac function in tachycardia-induced canine heart failure. Can J Physiol Pharmacol. 1996; 74: 1070eC1078.
Shen YT, Mallee JJ, Handt LK, Gilberto DB, Lynch JJ, Jr., Hargreaves RJ, Koblan KS, Gould RJ, Kane SA. Effects of inhibition of alpha-CGRP receptors on cardiac and peripheral vascular dynamics in conscious dogs with chronic heart failure. J Cardiovasc Pharmacol. 2003; 42: 656eC661.
Quayle JM, Bonev AD, Brayden JE, Nelson MT. Calcitonin gene-related peptide activated ATP-sensitive K+ currents in rabbit arterial smooth muscle via protein kinase A. J Physiol. 1994; 475: 9eC13.
Zhang X, Hintze TH. cAMP signal transduction cascade, a novel pathway for the regulation of endothelial nitric oxide production in coronary blood vessels. Arterioscler Thromb Vasc Biol. 2001; 21: 797eC803.
Wink DA, Miranda KM, Katori T, Mancardi D, Thomas DD, Ridnour L, Espey MG, Feelisch M, Colton CA, Fukuto JM, Pagliaro P, Kass DA, Paolocci N. Orthogonal properties of the redox siblings nitroxyl and nitric oxide in the cardiovascular system: a novel redox paradigm. Am J Physiol Heart Circ Physiol. 2003; 285: H2264eCH2276.
Lind H, Brudin L, Lindholm L, Edvinsson L. Different levels of sensory neuropeptides (calcitonin gene-related peptide and substance P) during and after exercise in man. Clin Physiol. 1996; 16: 73eC82.(Tatsuo Katori, Donald B. )