We administered ghrelin, a novel growth hormone-releasing hormone, to isolated perfused rat hearts, coronary arterioles, and cultured neonatal cardiomyocytes to determine its effects on coronary vascular tone, contractility, and natriuretic peptide secretion and gene expression. We also determined cardiac levels of ghrelin and whether the heart is a source of the circulating peptide. Ghrelin dose dependently increased coronary perfusion pressure (44 ± 9%, P < 0.01), constricted isolated coronary arterioles (12 ± 2%, P < 0.05), and significantly enhanced the pressure-induced myogenic tone of arterioles. These effects were blocked by diltiazem, an L-type Ca2+ channel blocker, and bisindolylmaleimide (Bis), a protein kinase C (PKC) inhibitor. Interestingly, coinfusion of ghrelin with diltiazem completely restored myocardial contractile function that was decreased 30 ± 3% (P < 0.01) by diltiazem alone. In contrast, combination of ghrelin with diltiazem or Bis did not significantly alter atrial natriuretic peptide (ANP) secretion, which was decreased 40% (P < 0.01) and 50% (P < 0.05) by these agents alone, respectively. Administration of ghrelin to cultured cardiomyocytes had no effect on ANP or B-type natriuretic peptide secretion or gene expression. Detectable amounts of low-molecular-weight ghrelin were present in cardiac tissue extracts but not in isolated heart perfusate. Thus we provide the first evidence that ghrelin has a coronary vasoconstrictor action that is dependent on Ca2+ and PKC. Furthermore, the data obtained from diltiazem infusion suggest that ghrelin has a role in regulation of contractility when L-type Ca2+ channels are blocked. Finally, the observation that immunoreactive ghrelin is found in cardiac tissue suggests the presence of a local cardiac ghrelin system.
- calcium channels
- protein kinases
- natriuretic peptides
ghrelin is a recently discovered 28-amino acid peptide containing an n-octanoyl modification at Ser3 and is the endogenous ligand for a previously identified orphan growth hormone (GH) secretagogue receptor (GHS-R) (14). The evidence to date suggests that ghrelin is synthesized, produced, and released from the stomach and circulates at reasonable concentrations (∼100 pmol/l) to act on the pituitary promoting GH release (14, 20). On the other hand, accumulating evidence suggests that GH has actions on the myocardium to affect growth and contractility (9) and that long-term GH therapy can augment intracellular systolic Ca2+ levels in myocytes from rats with postinfarction heart failure (34). This leads to the possibility that agents that act through the GHS-R family of receptors may also act on the myocardium, because these receptors are present in the rat and human heart and vasculature (14), and they appear to be upregulated in atherosclerosis (12). Indeed, GHS-R populations have been detected in the heart, and these are thought to mediate the coronary vasoconstrictor actions of the synthetic GH-releasing hexapeptide hexarelin in isolated heart preparations (3). Hexarelin improves cardiac function and decreases peripheral resistance in rats with myocardial infarction (35) and also protects the isolated heart from ventricular dysfunction associated with Ca2+ depletion (29). Ghrelin may also have a role in the cardiovascular system, because it is expressed in the heart at both the RNA (14) and peptide (10) levels and it decreases mean arterial pressures and increases cardiac index and stroke volume index without any effect on heart rate in humans (21). In rats with heart failure and cachexia, chronic ghrelin administration improves left ventricular dysfunction and attenuates ventricular remodelling (23). However, it is unknown whether ghrelin has direct actions on cardiac contractility or coronary vascular tone or whether it can alter the gene expression and/or peptide release of cardiac-specific molecules such as atrial natriuretic peptide (ANP) or B-type natriuretic peptide (BNP). Therefore, we sought to determine 1) whether ghrelin possesses the ability to modulate hemodynamics in isolated perfused heart preparations and in isolated coronary arteriole preparations, 2) whether ghrelin could have a role in modulating cardiac endocrine function by studying its effects on ANP and BNP secretion and gene expression in perfused heart preparations as well as in cultured neonatal rat ventricular myocytes, and 3) whether ghrelin is present in cardiac tissue and, if so, is the heart a possible source of circulating ghrelin.
Synthetic rat Ser3-(n-octanoyl)ghrelin was obtained from Phoenix Pharmaceuticals (Belmont, CA) and dissolved in perfusion buffer immediately before infusion or in DMEM/F-12 cell culture medium supplemented with 0.1% BSA to give a 10 μmol/l stock solution. Diltiazem (Orion Pharma) was initially dissolved in 0.9% saline, whereas 2-[1-(3-dimethylaminopropyl)-1H-indol-3-yl]-3-(1H-indol-3-yl)maleimide [GF-109203X, bisindolylmaleimide I (Bis), Calbiochem] was dissolved in DMSO. The final concentration of each solvent was <0.004%. Phenylephrine (PE; Sigma) was dissolved in DMEM/F-12 to give a 10 mmol/l stock solution. BNP peptide and antiserum as well as a 390-bp fragment of rat BNP cDNA probe (24) were provided by Dr. Kazuwa Nakao, Kyoto University School of Medicine (Kyoto, Japan). The rat ANP cDNA probe was provided by Dr Peter L. Davies, Queen's University (Kingston, Ontario, Canada) (6).
Isolated heart preparation.
Sprague-Dawley rats (n = 64, 280–320 g) obtained from the Center for Experimental Animals, University of Oulu (Oulu, Finland), underwent a modified isolated heart procedure similar to that described previously (8). The Animal Use and Care Committee of the University of Oulu approved the experimental design. Briefly, the animals were anesthetized with CO2, the chest was quickly opened, and cannulation above the aortic valve allowed perfusion at 37°C in a retrograde fashion (Langendorff) at constant flow (13 ml/min) using an oxygenated (95% O2-5% CO2) modified Krebs-Henseleit buffer (8). Perfusion pressure, which reflects coronary resistance, was measured by using a pressure transducer (Isotec, Hugo Sachs Elektronik) situated on a side arm of the aortic cannula. The left atrium was removed to allow insertion of a liquid-filled balloon into the left ventricle, which was connected to a pressure transducer to allow measurement of heart rate and contractile parameters. Hearts beat spontaneously and were allowed to equilibrate for 30 min. After a 10-min control period, hearts were perfused at 0.5 ml/min in a random protocol with either vehicle or ghrelin for 60 min. To study the intracellular mechanisms underlying ghrelin-induced vasoconstriction, we infused the L-type Ca2+ channel blocker diltiazem (1 μmol/l) and the specific PKC inhibitor Bis (90 nmol/l) either alone or in combination with 1 nmol/l ghrelin. The concentrations of diltiazem (32) and Bis (33) used here were calculated and titrated to suitable doses based on previous reports indicating their effectiveness in isolated heart preparations. Changes in heart rate, perfusion pressure, and parameters of cardiac function (left ventricular end-diastolic pressure, developed pressure, systolic pressure, and ±dP/dt) were recorded and analyzed using Ponemah data-acquisition software (Gould Instrument Systems). A 15-ml volume of perfusate was collected at 10-min intervals for hormone measurements by specific radioimmunoassay (RIA).
In a separate set of experiments, we sought to determine whether ventricular stretch could release immunoreactive ghrelin from the myocardium. Six hearts were prepared as described above and allowed to equilibrate for 30 min. After a 10-min control period, hearts were then stretched for 2 h by filling the intraventricular balloon to achieve an end-diastolic pressure of ∼25 mmHg, which we have previously shown to stimulate BNP gene expression and release (8). Hemodynamic variables were recorded on computerized software, and perfusate was collected every 10 min for the analysis of ghrelin immunoreactivity.
Responses of isolated coronary arterioles.
Vascular responses to ghrelin were investigated in isolated rat coronary arterioles, as described previously (1, 15, 16). Briefly, with the use of microsurgery instruments and an operating microscope, a branch of the septal coronary artery (∼1 mm in length) was isolated and transferred into an organ chamber containing two glass micropipettes filled with Krebs solution equilibrated with a gas mixture of 95% O2 and 5% CO2
In separate experiments, in the presence of 80-mmHg intraluminal pressure, arterioles were incubated with Bis (90 nmol/l) for 30 min (before the administration of ghrelin), and ghrelin-induced changes in arteriolar diameter as a function of time were then obtained. Next, changes in diameter to increases in intraluminal pressure were measured, and each pressure step was maintained for 5–8 min to allow the vessel to reach a steady-state diameter. Arteriolar dilations to the Ca2+ antagonist diltiazem (10−9–10−5 mol/l) were also obtained before and after incubation of arterioles with ghrelin and ghrelin plus Bis. Ghrelin-induced arteriolar responses as a function of time are shown as changes in arteriolar diameter. Myogenic constriction was calculated at each pressure step as the percent change in diameter compared with the corresponding passive diameter in Ca2+-free Krebs solution. Diltiazem-induced arteriolar responses were expressed as a percentage of the maximal dilation of the vessel, defined as the passive diameter at 80 mmHg of intraluminal pressure in Ca2+-free Krebs solution. Data are expressed as means ± SE.
Ventricular myocytes were prepared from 2- to 4-day-old neonatal rat hearts (28, 36). Cells were plated at the density of 2 × 105 cells/cm2 onto Falcon wells 15–35 mm in diameter. After 24 h, the serum-containing medium was replaced with complete serum-free medium (CSFM). After 48-h incubation in CSFM, the wells were divided into test groups, and the medium was replaced with CSFM or CSFM supplemented with 1, 10, or 100 nmol/l ghrelin and incubated up to 48 h at 37°C. Medium was replenished every 24 h. After experiments, cells were washed twice with PBS and quickly frozen at −70°C before RNA extraction.
Isolation and analysis of RNA.
Total RNA from cultured cardiac myocytes was isolated using the guanidine thiocyanate-CsCl method (5). For RNA Northern blot analyses, 1.8- to 6-μg samples were separated by electrophoresis and transferred to nylon membranes (Osmonics). The cDNA probes complementary to rat ANP or BNP mRNA or 18S rRNA were random prime labeled with Rediprime II (Amersham Biosciences). The membranes were hybridized and washed three times for 20 s at 62°C as previously described (28, 36). Thereafter, the membranes were exposed with PhosphorImager screens (Amersham Biosciences), which were scanned with Molecular Imager FX Pro Plus and quantitated using Quantity One software (Bio-Rad). Hybridization signals of ANP and BNP were normalized to that of 18S RNA.
Cardiac tissue extraction and HPLC.
Cardiac tissue extracts from ventricular stretch experiments were prepared from atrial and ventricular tissue samples as previously described (27). Extracted supernatants were subjected to a specific RIA for ghrelin to calculate tissue concentrations. Supernatants were then dried under air, reconstituted in 20% acetonitrile-0.1% trifluoroacetic acid (TFA), and subjected to reverse-phase HPLC (RP-HPLC). Immunoreactive ghrelin fractions from RP-HPLC were then pooled, concentrated, dried under air, and reconstituted in 10% acetonitrile-0.15 mol/l NaCl for size exclusion HPLC on a Pharmacia HR10/30 Superdex high-resolution peptide column. Each immunoreactive fraction from RP-HPLC was run separately on size exclusion HPLC using an isocratic gradient of 10% acetonitrile-0.1% TFA and 0.15 mol/l NaCl, with a flow rate of 0.25 ml/min. Fractions were collected at 1-min intervals and subjected to further ghrelin RIA to establish molecular size.
Immunoreactive ANP (38) and BNP (13) concentrations in the perfusate and cell culture medium were determined using specific RIA as previously described. The sensitivities of the BNP and ANP assays were 2 and 1 fmol/tube, respectively. Fifty percent displacements (ED50) of the respective standard curve occurred at 16 and 25 fmol/tube. The intra- and interassay variations were 10% and 15%, respectively. Immunoreactive ghrelin in perfusate and cardiac tissue extracts was measured as previously described (26). The RIA has a mean zero binding of 24 ± 2%, mean sample detection limit of 3.3 fmol/tube, and ED50 of 136.2 ± 10 fmol/ml.
Results are presented as means ± SE. Hemodynamic and peptide RIA time-course data from isolated heart and coronary arteriole experiments were analyzed with two-way ANOVA for repeated measures, followed by a least-significant difference (LSD) post hoc test. For multiple comparisons, data were analyzed with one-way ANOVA, followed by a LSD post hoc test. Gene expression data were analyzed using Student's t-test for unpaired data. A value of P < 0.05 was considered statistically significant.
Infusion of ghrelin in isolated perfused rat hearts.
Basal values for hemodynamic variables in isolated hearts are given in Table 1. One-hour infusion of ghrelin had no significant effect on cardiac contractility at the doses of 0.01–10 nmol/l, although there was a tendency for developed pressure to slowly decrease in the ghrelin group compared with the vehicle group. Perfusion pressure, however, increased significantly in a dose-dependent manner during ghrelin infusion. The effect was most pronounced at the concentration of 1 nmol/l, which resulted in a 44 ± 9% increase in perfusion pressure compared with the vehicle group (P < 0.01; Fig. 1).
Role of L-type Ca2+ channels and PKC.
To study potential signaling mechanisms underlying ghrelin-induced increases in coronary perfusion pressure, we infused the L-type Ca2+ channel blocker diltiazem and the specific PKC inhibitor Bis with 1 nmol/l ghrelin. Ghrelin-induced increases in perfusion pressure were completely abolished by a coinfusion of 1 μmol/l diltiazem (P < 0.01; Fig. 2A) and 90 nmol/l Bis (P < 0.05; Fig. 2B). Diltiazem (Fig. 2A) or Bis (Fig. 2B) had no significant effect on perfusion pressure when infused alone.
As shown in Fig. 2C, 1 nmol/l ghrelin had no significant effect on developed pressure when administered alone. However, the negative inotropic effect of diltiazem (−30 ± 3%, P < 0.01 vs. vehicle) was completely abolished when ghrelin was coinfused (P < 0.05 vs. diltiazem), restoring developed pressure to vehicle control levels (Fig. 2C). Bis had no effect on developed pressure either when infused alone or in combination with ghrelin (data not shown).
Responses of isolated coronary arterioles.
After a 1-h incubation period, the active diameter of coronary arterioles was 103 ± 9 μm in the presence of 80-mmHg pressure. The administration of 1 nmol/l ghrelin resulted in a slow constriction of coronary arterioles (P < 0.05; Fig. 3A). Furthermore, ghrelin significantly enhanced the pressure-induced myogenic tone of arterioles between 40 and 120 mmHg (P < 0.05; Fig. 3B). Both effects were attenuated with 90 nmol/l Bis (Fig. 3, A and B). Finally, diltiazem-induced arteriole vasodilation was not altered by ghrelin nor ghrelin and Bis together (Fig. 3C).
Effect of ghrelin on natriuretic peptide secretion in isolated perfused hearts.
Basal ANP concentrations in the isolated heart perfusate before agent infusions are given in Table 1. Ghrelin infusion at 1 nmol/l had no effect on basal ANP secretion compared with vehicle (P = 0.24) in the isolated heart preparation. Neither did ghrelin modify ANP secretion when coinfused with diltiazem. Diltiazem alone caused a 40–45% reduction (P < 0.01) in ANP secretion (Fig. 4A), an observation consistent with previous reports (27). Bis caused a significant 50% (P < 0.05) decrease in perfusate ANP levels (Fig. 4B), which was not significantly affected by coinfusion with ghrelin (P = 0.37, Bis vs. Bis + ghrelin). Yet, the previous statistically significant difference between Bis and vehicle disappeared when ghrelin was coinfused with Bis (P = 0.07, Bis + ghrelin vs. vehicle). BNP secretion was unaffected in all isolated heart experiments (data not shown).
Ghrelin and natriuretic peptide secretion and gene expression in cultured cardiomyocytes.
Cultured neonatal rat ventricular myocytes were treated with incremental doses (1–100 nmol/l) of ghrelin for up to 48 h. PE, an α-adrenergic agonist known to stimulate ANP and BNP synthesis and secretion in cardiac myocytes (28, 36), was used as a positive control. Administration of 10 μmol/l PE stimulated ANP and BNP secretion by 77 ± 18% (P < 0.001; Fig. 5A) and 93 ± 11% (P < 0.05; Fig. 5B), respectively, and resulted in 70% (P < 0.05) increases in both ANP (Fig. 6B) and BNP (Fig. 6C) mRNA levels. Ghrelin had no significant effect on ANP or BNP peptide secretion (Fig. 5) and gene expression (Fig. 6).
Cardiac tissue ghrelin immunoreactivity and response to ventricular stretch.
Cardiac tissue extracts contained immunoreactive ghrelin, with the right atrium having a slightly higher content compared with the right and left ventricle (4.1 ± 0.5, 2.4 ± 0.1, and 1.0 ± 0.1 fmol/mg wet wt, respectively, n = 6). However, these levels were 450- to 1,800-fold less than those reported for stomach tissue extracts (10, 14). Primary analysis of ghrelin immunoreactivity on RP-HPLC revealed two peaks in both atrial and ventricular extracts (Fig. 7). Peak 1 (Fig. 7, A and C) was consistent in RP-HPLC retention time with synthetic octanoyl ghrelin, whereas peak 2 eluted later. Separate analysis of both atrial and ventricular peaks by size exclusion HPLC revealed each single peak to be of low molecular weight (Mr ∼3,400; Fig. 7, B and D). Despite cardiac tissue extracts containing bona fide ghrelin, immunoreactive ghrelin was not detectable at any time in the perfusate from stretch experiments, arguing against stretch-mediated cardiac secretion (data not shown).
Ghrelin was initially discovered from the stomach and identified as the natural ligand for a particular G protein-coupled orphan receptor, denoted GHR-S (14). Subsequent studies have clearly shown ghrelin to be a potent stimulator of GH release (21, 40) and that it has a significant role in energy balance and carbohydrate metabolism (37). However, more recent work has shown ghrelin to have effects on blood pressure, cardiac function, and energetics (23), particularly with respect to cardiac catabolic-anabolic imbalance in severe congestive heart failure (22). These results are supported by the identified tissue distributions of GHR-S, which include the lung, intestine, pancreas, and adipose tissue (14) and also the heart and coronary vasculature (12). However, although ghrelin has been implicated as a potential cardiovascular peptide, it is unknown whether it can directly influence cardiac function or coronary vasomotor tone or whether it has any effect on the endocrine function of the heart, such as natriuretic peptide secretion. Thus this report provides several notable firsts: 1) we describe a constrictor effect of ghrelin on the coronary vasculature and its dependence on Ca2+ and PKC; 2) there is evidence of a role for ghrelin in modulating cardiac contractile function in relation to Ca2+ status; 3) there is a the lack of effect of ghrelin on ANP and BNP activity in cardiac myocytes; and 4) this is the first description of whether the heart could secrete ghrelin.
The time course of the increases in perfusion pressure and coronary arteriole constriction in response to ghrelin observed here was slow. This effect was significantly inhibited by both diltiazem and Bis, suggesting a dependence on both Ca2+ and PKC, respectively. In this regard, ghrelin has an almost identical profile to previously described hexarelin-induced increases in coronary perfusion pressure (3) but is effective at 1,000-fold lower doses. Such a high potency for activity (observed at ∼1 nmol/l in the present study) is comparable with endothelin-1 (ET-1) (31), suggesting that ghrelin is one of the most potent identified regulators of coronary tone.
Initial dose-ranging experiments have revealed that doses lower and higher than 1 nmol/l ghrelin resulted in (nonsignificant) peak increases in perfusion pressure by 30 min. In contrast, data from human studies suggest that ghrelin has in vivo vasorelaxant activity (20, 25) that may be independent of NO activity (25) and that it is able to antagonize ET-1-induced vasconstriction (39) in vitro in endothelium-denuded internal mammary artery preparations. Naturally, in vivo release of GH from the anterior pituitary (but absent in the isolated perfused heart), differential GHS-R distributions across tissue beds (14), and regulatory and counterregulatory systems may account for these observed differences. Thus Wiley and Davenport (39) employed ghrelin in vitro at doses up to 300 nM, nearly 300 times those employed in our study, whereas Okumura et al. (25) utilized sequential pharmacological boli (2, 5, and 10 μg) of ghrelin and only achieved significant in vivo forearm vasodilation at 5- and 10-μg doses. Given that our vasconstriction was achieved at physiological levels of ghrelin (between 0.3 and 1 nM), this may explain some of the discrepancy. Furthermore, in human studies by Nagaya et al. (20), the decrease in mean arterial pressure was noted in response to a single bolus of ghrelin, which achieved a pharmacological plasma level of ∼45 nmol/l. Taken together, these results suggest that any vasoconstrictor activity attributable to ghrelin may depend on the site of administration, the dosage employed, GHS-R expression profiles, and the species in question.
Our data suggesting that ghrelin has no direct inotropic effect on the heart is consistent with a previous report (23) indicating no effect of the peptide on fractional cell shortening in isolated myocytes. An intriguing aspect of our data is the observation that ghrelin appears to protect cardiac function when cytosolic Ca2+ concentration is decreased and that PKC appears to play no significant role. Thus, when the isolated perfused heart preparation was subjected to L-type Ca2+ channel blockade, the developed pressure significantly decreased (as expected), yet coinfusion of ghrelin with diltiazem restored developed pressure to vehicle control levels. In this context, in vivo subcutaneous administration of hexarelin for 7 days has been reported to precondition and protect subsequent isolated perfused heart preparations against calcium overload-induced increases in left ventricular end-diastolic pressure in normal rats subjected to low Ca2+ perfusion/normal Ca2+ reperfusion (29) or to ischemia-reperfusion injury in hypophysectomized rats (17). Furthermore, in vivo evidence of hexarelin-induced improvements in cardiac function have been reported after bolus injection in humans (2), and in vitro data from isolated rat heart preparations suggest that ghrelin may be protective against ischemia-reperfusion injury, at least partially through reducing myocyte lactate dehydrogenase and myoglobin release (4) and/or via PKC-related mechanisms (7). The mechanism(s) behind the protective effects of hexarelin/ghrelin is unclear, but any improvements in cardiac function need to be weighed up against potential deleterious constrictor actions at higher doses of ghrelin (>0.7 nM) and hexarelin (>0.5 μM). Nevertheless, the possibility that ghrelin and hexarelin may have direct effects on Ca2+ homeostasis and whether this is responsible for the beneficial effects of each peptide in experimental myocardial infarction (35) and congestive heart failure (23) merit further investigation.
In our hands, ghrelin exhibited no effect on basal ANP or BNP peptide secretion from isolated hearts, and it had no effect on diltiazem-induced reductions in ANP secretion. Consistent with the known role of PKC in regulating ANP secretion (30), infusion of the PKC inhibitor Bis (which attenuates α-, β-, γ-, δ-, and ε-isoforms of the enzyme) significantly inhibited ANP secretion in the isolated perfused heart preparation. Although the difference between Bis versus Bis and ghrelin was not significant, the effect of Bis appeared to be attenuated toward the end of the perfusion period by coinfusion of ghrelin. This suggests that mechanisms governing the effects of ghrelin on cardiac hemodynamics and any putative endocrine secretory effects of ghrelin are dissociated and/or have little influence on one another, at least in the 1-h time period of infusion used here. In this regard, the peptide secretion and gene expression results from cultured ventricular cardiomyocytes indicate more clearly the lack of effect of ghrelin on natriuretic peptide secretion during a longer period of administration (up to 48 h). Our results do not rule out that ghrelin may affect both pathways (Ca2+ and PKC) through a common intracellular mechanism. Indeed, ghrelin-induced vasoconstriction was completely abolished by either L-type Ca2+ channel blockade or PKC inhibition. Thus the vasoconstrictive and endocrine effects of ghrelin might be mediated through different GHS-R subtypes (3) as well as different isozymes of PKC (7). In support of this are the observations that different PKC isozymes (some of which are Ca2+-dependent) are differentially distributed within cardiovascular tissues (18) and that specific PKC isozyme anchoring protein receptors for activated C kinase are differentially activated within cardiac myocytes (19).
Cardiac tissue extracts contained measurable amounts of ghrelin, with the atrium containing approximately twice as much as the ventricle on a picomole per wet weight basis, values that are in agreement with those previously reported in rat cardiac tissue (10). RP-HPLC and size exclusion HPLC analysis demonstrated this immunoreactivity to be made up of two low-molecular-weight forms of ghrelin, one of which was consistent with the octanoyl form. This may represent multiple posttranslational products of ghrelin processing (defined by the degree of acylation) as recently described in human plasma and the stomach (11), and it is known that the acyl and des-acyl forms of the peptide have markedly different retention times on RP-HPLC (14). However, despite the presence of quantifiable amounts of ghrelin in cardiac tissue extracts, we could not detect immunoreactive ghrelin in isolated heart perfusates, despite a 10-fold concentration when prepared for RIA. This suggests that if ghrelin has an endogenous role in modulating cardiac function, it might act in a paracrine/autocrine manner similar to angiotensin II and ET-1 (31).
In summary, we provide the first report of a slow-acting, vasoconstrictor action of the novel peptide ghrelin on the coronary vasculature that is dependent on L-type Ca2+ channel and PKC activation. Additionally, the negative inotropic action of diltiazem was effectively blocked by ghrelin, which suggests that ghrelin has a role in the regulation of cardiac Ca2+ homeostasis. Decreases in ANP secretion induced by blocking L-type Ca2+ channels or PKC were not affected by coadministration of ghrelin, suggesting a differential action on hemodynamics versus the endocrine function of the heart. Cardiac myocyte culture demonstrated a lack of effect of ghrelin on natriuretic peptide secretion and gene expression, even after 48 h of administration. We show that bona fide ghrelin exists in cardiac tissue, yet levels are not high enough to suggest that the heart is a significant source of circulating ghrelin, which was confirmed by the observation that ghrelin was undectable in isolated heart perfusate and that cardiac stretch was not sufficient to induce its release. However, when coupled with previous reports identifying 1) the presence of GHS-Rs in cardiac endothelial cells (12, 14) and 2) beneficial effects of ghrelin administered to patients with heart failure (21), our observations suggest that the cardiac ghrelin system is a potential target for future therapeutic strategies in congestive heart failure. The precise cellular distribution of a cardiac ghrelin system and the identification of intracellular signaling pathways underlying these diverging actions are therefore logical targets for further studies.
This study was supported by the Foundation of Research, Science and Technology of New Zealand, Academy of Finland, Hungarian Scientific Research Fund Grants T033117 and F035213, the Sigrid Juselius Foundation, and the Finnish Foundation for Cardiovascular Research. C. J. Pemburton was the recipient of a Postdoctoral Fellowship from the Foundation of Research, Science and Technology of New Zealand.
We thank Marja Arbelius and Tuulikki Kärnä for expert technical assistance.
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 © 2004 by the American Physiological Society