INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

RECENT EVIDENCE INDICATES that neuropeptides in the autonomic nervous system may play significant roles in cardiovascular regulation (4). Calcitonin gene-related peptide (CGRP) is broadly distributed in peripheral autonomic neurons innervating cardiovascular structures (4, 6, 17, 26, 28), including human heart (6). Expression of mRNA of CGRP and CGRP₁ receptors has been localized in rat ventricular myocardial tissue (23) and in in situ mouse ventricular myocytes (1), respectively.

The physiological role of CGRP on the cardiovascular system has not been fully established. A relatively high amount of CGRP has been detected in rat (31) and human plasma (13). Circulating CGRP is largely derived from perivascular sensory nerve endings (28, 32). Increase in circulating CGRP occurs during diseased states such as congestive heart failure (7) and acute myocardial infarction (18). CGRP is an extremely potent coronary and peripheral vasodilator (25, 26, 28). CGRP exerts positive inotropic effects on human atrial tissue (10). It has also been reported that CGRP exerts positive inotropic effects on isolated porcine ventricular muscle strips (20, 29). Intravenous infusion of CGRP dramatically improves ventricular contractile function in heart failure patients (11). However, some studies have failed to demonstrate a CGRP-induced positive ventricular inotropism on isolated canine (26) and rat (19) hearts. Furthermore, mechanisms underlying the inotropic effects of CGRP on ventricular muscle are unclear. We have hypothesized that CGRP exerts positive inotropic effect on ventricular muscle through the activation of L-type Ca²⁺ channels, resulting in Ca²⁺-induced intracellular Ca²⁺ mobilization.

To determine the effect and mechanisms of CGRP on mammalian ventricular muscle and species variability, we examined the effect of CGRP on contractile function of isolated ventricular myocytes from both adult rat and mouse hearts and compared its effects with those elicited by a -adrenoreceptor agonist, isoproterenol (Iso). We used the -adrenoreceptor antagonist atenolol and the CGRP₁ receptor antagonist CGRP-(737) to investigate CGRP receptor effects. To investigate cellular mechanisms, we studied CGRP effect on intracellular [Ca²⁺] transients in the presence and absence of L-type Ca²⁺ channel blockade. Furthermore, CGRP was tested after inhibition of protein kinase A in ventricular myocytes to determine the role of cAMP in CGRP signal transduction.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Preparations of ventricular myocytes. The procedure for obtaining rod-shaped and Ca²⁺-tolerant ventricular myocytes was based on a method previously described (14). In brief, mice (n = 7, 4-6 mo old) and rats (n = 27, 3-6 mo old) were killed using pentobarbital sodium anesthesia. The heart was excised, cannulated through the aorta, and then mounted on a Langendorff perfusion apparatus. The coronary arteries were perfused with Ca²⁺-free Tyrode containing (in mM) 132 NaCl, 4.8 KCl, 1.2 MgCl₂, 5 glucose, and 10 HEPES (pH 7.4) for 5 min. The heart was then perfused with collagenase II (0.4 mg/ml, Worthington Biochemical, Lakewood, NJ) in Ca²⁺-free Tyrode for 10 min. All solutions were continuously oxygenated with a perfusion temperature of 37°C. The ventricles were cut from the atria, minced, and shaken in a waterbath. Dissociated myocytes were collected by centrifugation. The cell pellets were resuspended in Tyrode solution (0.5 mM Ca²⁺, 1% bovine serum albumin, and 1% taurine) before the experiment.

Measurement of cell shortening. Freshly isolated myocytes were placed in a cell chamber that was continuously perfused (37°C) on the stage of an inverted microscope. Experimental solution containing (in mM) 120 NaCl, 2.6 KCl, 1.2 MgCl₂, 1.2 KH₂PO₄, 5 glucose, 5 HEPES, 20 NaHCO₃, and 1 CaCl₂ was circulated through the cell chamber (Biotechs FCS2, Biotechs, Butler, PA). Platinum electrodes connected to a stimulator (Grass S48, Grass Instruments) were used to stimulate cells at a frequency of 1 Hz. A PTI system (Photon Technology International, South Brunswick, NJ) was used for the study. Cell image was continuously acquired through an objective lens and transmitted to a camera. The output from the camera was displayed on a video monitor. Myocyte contraction was measured using a video motion edge detector system (Crescent Electronics, Sandy, UT), and data were acquired, stored, and analyzed by a computer. Cells (90-120 µm in length) were chosen for study based on the criteria that rod-shaped myocytes with clear striations should be quiescent before electrical pacing. On electrical stimulation, myocytes should contract rhythmically and persistently.

Measurement of [Ca²⁺] transients. Myocytes were loaded with 4 mM of fura 2-acetoxymethyl ester (Sigma) at room temperature for 30 min in Ca²⁺-free Tyrode solution. The loaded myocytes were washed three times and equilibrated in experimental solution for 30 min. A PTI spectrophotometer excited the cell at 340/380 nm alternately at a 240-Hz sampling rate. Measurements of 510-nm emissions were recorded as fluorescence ratios (340/380 nm), which were used to represent intracellular Ca²⁺ changes (9).

Pharmacological interventions. All chemicals were purchased from Research Biochemicals Internationals (Natick, MA). Ventricular myocytes were perfused in Tyrode solution (1 mM Ca²⁺) for at least 10 min, and data for baseline cell shortening and [Ca²⁺] transients were collected thereafter. The effects of CGRP and Iso on contraction and [Ca²⁺] transients of myocytes were tested in a concentration-dependent manner. The effect of CGRP was tested in the presence of the -adrenoreceptor blocker atenolol (100 nM) and the CGRP₁ receptor blocker CGRP-(737) (2 × l0⁷ M) (8). Their effects on myocytes were evaluated immediately after they were added to the tissue bath perfusate. The effect of CGRP on [Ca²⁺] transients of rat ventricular myocytes was tested in the presence of the Ca²⁺ channel blocker nifedipine (5 mM) to determine whether intracellular Ca²⁺ release is dependent on Ca²⁺ influx through sarcolemmal L-type Ca²⁺ channels. Cells were exposed to nifedipine for 5 min followed by perfusion of nifedipine and CGRP (1 nM). Thereafter, CGRP (1 nM) was reapplied after nifedipine and CGRP were washed out for 20 min. To determine the role of cAMP in CGRP signal transduction, the effect of CGRP on [Ca²⁺] transients in myocytes was tested in the absence and presence of the cAMP antagonist Rp-cAMPS, which inhibited protein kinase A (30). After identification of CGRP-induced increase in [Ca²⁺] transients in myocytes, the cells were treated with Rp-cAMPS (100 mM) for 30 min before the reapplication of CGRP. CGRP was further tested in the absence and presence of caffeine (15 mM for 2 min) in beating myocytes.

Data analysis. Cell shortening was recorded during baseline and interventions. The amplitude of cell shortening and [Ca²⁺] transients during contraction and the half-maximum myoplasmic Ca²⁺-uptake time during relaxation were calculated from individual waveforms using Matlab software. Changes in the amplitude of [Ca²⁺] transients before and after drug treatments were expressed as mean percentage changes if the mean diastolic baseline fluorescence ratio was not significantly changed. To calculate the velocity of myocyte shortening and relaxation, the derivative was calculated as a two-point slope looking back one point on individual contraction waveforms. Maximum values of negative slope (dL/dt_Max) and positive slope (+dL/dt_Max) during contraction and relaxation were used to represent the velocity of cell shortening and relaxation, respectively. Paired Student's t-test and analysis of variance were used for statistical analyses. A value of P < 0.05 was considered significant.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Cell contraction. In rat ventricular myocytes, the increase in cell shortening after CGRP application was 50 ± 10 to 140 ± 40% greater than their respective baselines at concentrations of 10 pM-10 nM (Table 1 and Fig. 1D). Increase in CGRP concentrations beyond 100 nM led to myocyte fibrillation or contracture. Peak effect was reached 2-4 min after CGRP exposure. The EC₅₀ for CGRP and Iso was 44.1 pM and 9.8 nM, respectively. CGRP (1 nM) increased maximum velocity of rat myocyte shortening (dL/dt_Max) and relaxation (+dL/dt_Max) by 98 and 106%, respectively (12 cells, P < 0.01). In mice, CGRP-induced augmentation in ventricular myocyte shortening was 70 ± 24 to 200 ± 54% greater than respective baselines at concentrations of 10 pM-10 nM (Table 1 and Fig. 1B), with a peak effect at 2-3 min after CGRP exposure. The EC₅₀ for CGRP and Iso was 55.8 pM and 8.4 nM, respectively, in mouse myocytes. CGRP (1 nM) increased dL/dt_Max and +dL/dt_Max by 150 and 213%, respectively (9 cells, P < 0.01). CGRP did not produce any effect in the presence of the CGRP₁ receptor blocker CGRP-(737) but still elicited a 92 ± 23% increase in cell shortening in the presence atenolol at a concentration that abolished the Iso-induced effect (Table 2).

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Dose-dependent effect of calcitonin gene-related peptide (CGRP; A and C) and isoproterenol (Iso; B and D) on ventricular myocyte shortening in mouse (A and B) and rat (C and D). Increase in amplitude of cell shortening is expressed as %change compared with baseline contraction. CGRP increases cell shortening by 50 ± 10, 95 ± 18, 110 ± 46, and 140 ± 40% at the doses of 10 pM-10 nM, respectively, in rat. CGRP increases cell shortening by 70 ± 24, 181 ± 48, 198 ± 44, and 200 ± 54% at the doses of 10 pM-10 nM, respectively, in mice. Inset shows superimposed tracings of a cell responding to CGRP in a dose-dependent manner. ** P < 0.0l (comparison is made between drug treatment and baseline, n = cell number). Vertical and horizontal bars represent 5 µm and 250 ms, respectively.

View this table: [in this window] [in a new window]   Table 2.   Effects of CGRP on rat ventricular myocyte contraction in the presence of CGRP1-receptor blocker CGRP-(737) and -adrenoreceptor blocker atenolol

[Ca²⁺] transients. CGRP did not significantly alter the baseline diastolic [Ca²⁺] transients at all doses tested. The systolic [Ca²⁺] transients in rat ventricular myocytes were increased by 120 ± 36% (P < 0.01) after 1 nM CGRP administration, an effect that was 54% greater than that produced by 1 nM of Iso (Fig. 2). Neither CGRP nor Iso changed diastolic Ca²⁺ levels. The half-maximum myoplasmic Ca²⁺ clearance time during relaxation was shortened by 30 ± 5% (141 ± 24 vs. 96 ± 17 ms; n = 10, P < 0.01) after the cells were exposed to CGRP. The exposure of myocytes to nifedipine (5 mM) for 3-5 min decreased the amplitude of basal [Ca²⁺] transients of rat ventricular myocytes. CGRP-induced increase in [Ca²⁺] transients was abolished in the myocytes in the presence nifedipine (Fig. 3).

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Effects of CGRP and Iso on [Ca2+] transients in rat ventricular myocytes. Increase in [Ca2+] transients is expressed as %change in 340/380 ratio from baseline values. Inset is a typical tracing of a dose-dependent increase in [Ca2+] transients in a rat ventricular myocyte during CGRP administration (vertical calibration bar represents [Ca2+] transient ratio of 1 and time calibration bar is 1 s; ** P < 0.01).

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Effect of CGRP (1 nM) on [Ca2+] transients of ventricular myocytes (8 cells, 4 rats) with and without the Ca2+ channel blocker nifedipine (5 µM). Response is shown as %change in [Ca2+] transients (340/380 ratio) from the baseline value of individual cells. Inset is a typical cell showing that the CGRP-induced increase in [Ca2+] transient (CGRP) is abolished in the presence of nifedipine (nifedipine+CGRP). Experimental order is nifedipine, nifedipine+CGRP, and CGRP. Vertical and horizontal bars represent Ca2+ transient ratio of 1 and 200 ms, respectively. ANOVA was used; ** P < 0.01 between CGRP and baseline, Nifedipine, and Nifedipine + CGRP.

Application of caffeine (15 mM for 2 min) to beating rat ventricular myocytes quickly reduced the amplitude of [Ca²⁺] transients from a level of 65 ± 8 to 20 ± 3% above baseline. In the presence of caffeine, CGRP induced an additional 18 ± 2% increase in [Ca²⁺] transients (Fig. 4). Incubation of rat ventricular myocytes with Rp-cAMPS (100 mM) did not affect the amplitude of basal [Ca²⁺] transients. CGRP no longer increased [Ca²⁺] transients in the myocytes after Rp-cAMPS treatment (Fig. 5).

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Effect of CGRP on Ca2+ transients in the absence and presence of caffeine. Caffeine (15 mM) markedly reduces the amplitude of [Ca2+] transient to a value 20 ± 3% above diastolic baseline, a level reflecting the amount of Ca2+ influx during depolarization, because sarcoplasmic reticulum (SR) Ca2+ is depleted by caffeine. In the presence of caffeine, application of CGRP induces an additional 18 ± 2% increase in [Ca2+], a value reflecting CGRP-induced Ca2+ influx. Accordingly, SR Ca2+ release in the presence and absence of CGRP during each depolarization is 69 ± 6 (CGRP) and 45 ± 5% (baseline), respectively, after subtracting their Ca2+ influx amount, suggesting increased Ca2+ release from SR after CGRP stimulation. ** P < 0.01 between bracketed columns.

View larger version (14K): [in this window] [in a new window]   Fig. 5.   Effects of CGRP on [Ca2+] transients in rat ventricular myocytes in the absence and presence of a protein kinase A inhibitor Rp-cAMPS. CGRP (100 pM) increases [Ca2+] transients by 86 ± 31%. Treatment of myocytes (n = 6) with Rp-cAMPS (100 mM for 30 min) abolishes CGRP-induced increase in [Ca2+] transients. (t-test and ANVOA are used for baseline CGRP- and Rp-cAMPS-treated groups, respectively; ** P < 0.01).

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

This study demonstrates that the neurotransmitter CGRP increases the amplitude and velocity of cell contraction and increases the rate of cell relaxation in adult rodent ventricular myocytes (Table 1 and Fig. 1). This stimulatory effect is mediated through the activation of myocardial CGRP₁ receptors, which leads to Ca²⁺-induced Ca²⁺ release. Contractility was assessed in terms of cell shortening profile. Evidence shows that measurements of cell length shortening in single myocytes are compatible with the systolic and diastolic contractile properties of intact cardiac muscle (5). These results are further confirmed when compared with those obtained from the -adrenoreceptor agonist Iso.

In addition to its blood-borne hormonal property, CGRP might function as an intrinsic cardiac peptide, because abundant CGRP mRNA has been identified in rat ventricular tissue, implying a local synthesis and release mechanism independent of extracardiac innervation (23). Present data demonstrate a twofold increase in ventricular myocyte contractility compared with baseline when CGRP levels are raised to 100 pM, which is similar to serum CGRP levels in rat and human (13, 31). Thus it is possible that CGRP may exert an endocrine or a paracrine influence on ventricular myocardial performance. Peptide release from cardiac sensory nerves involves the stimulation of mechano- and chemosensitive receptors in the sensory nerve endings (15, 16). Peptides and purines such as CGRP, substance P, and adenosine are released in excessive amounts during cardiac-diseased states such as myocardial ischemia (16, 18). Inhibition of Ca²⁺-dependent K⁺ current in nerve endings may be responsible for sensory receptor activation and neuropeptide release (16, 24). Thus the sensory nervous system may exert an important regulatory function in the cardiovascular system.

There have been inconsistent results regarding the physiological roles of CGRP in regulating ventricular contractile function. It has been demonstrated that CGRP exerts a positive inotropic effect on isolated porcine ventricular muscle strips (20, 29) and on isolated adult rat ventricular myocytes (2). However, some studies have failed to show a CGRP-mediated positive ventricular inotropic effect on isolated blood-perfused canine (26) and isolated rat hearts (19). Many reasons may account for these different observations. A species difference in terms of CGRP receptor expression and density might exist between canine and rodent ventricular muscle. Assessment of the direct inotropic effect of CGRP in vascular-intact isolated hearts may not provide information about whether CGRP has a direct action on ventricle or an indirect action secondary to coronary vasodilation, because CGRP is a potent coronary vasodilator. The influence of coronary vasculature is minimized in strips of cardiac muscle perfused in vitro and absent in suspensions of isolated myocytes.

Mechanisms underlying CGRP-mediated positive inotropic effect on ventricular myocytes were examined in the present study. We found that CGRP-induced positive ventricular inotropic effect was not mediated through the -adrenoreceptors, because CGRP induced a 92% increase in cell shortening in the presence of atenolol (Table 2). That CGRP no longer elicited a positive inotropic response in the presence of CGRP₁ receptor blockade confirmed that CGRP-induced inotropic effect was mediated specifically through the activation of CGRP₁ receptors. CGRP produced a concentration-dependent increase in the magnitude of the [Ca²⁺] transients in beating rat ventricular myocytes (Fig. 2). This effect was abolished in the presence of blockade of the L-type Ca²⁺ channel by nifedipine (Fig. 3), suggesting that increase in intracellular Ca²⁺ by CGRP is dependent on the activation of L-type Ca²⁺ channel. This finding is consistent with a previous report that CGRP causes a voltage-dependent elevation of intracellular Ca²⁺ in nonbeating guinea pig ventricular myocytes (12). The magnitude of CGRP-induced Ca²⁺ influx and sarcoplasmic reticulum (SR) Ca²⁺ release in beating myocytes was determined in the presence of 15 mM of caffeine, a concentration that could quickly deplete SR Ca²⁺ store. After SR Ca²⁺ depletion by caffeine, depolarization induced a 20 ± 3% increase in [Ca²⁺] transient above baseline, presumably reflecting fractional Ca²⁺ influx during electrical pacing (Fig. 4). Reapplication of CGRP to the cells in the presence of caffeine induced a 38 ± 4% increase in [Ca²⁺] transient above baseline, indicating an additional 18 ± 2% increase in Ca²⁺ influx by CGRP compared with pacing-induced [Ca²⁺] transient alone (Fig. 4), a finding consistent with the reports that CGRP increases Ca²⁺ current in atrial myocytes (21, 22). Accordingly, the fractional SR Ca²⁺ release without and with CGRP was 45 ± 5 and 69 ± 6%, respectively, after subtracting their Ca²⁺ influx amount (Fig. 4). Thus CGRP induced an additional 21 ± 6% SR Ca²⁺ release compared with pacing-induced SR Ca²⁺ release alone. Evidently, the enhanced Ca²⁺ release from SR is induced by enhanced Ca²⁺ influx after CGRP stimulation. In addition to its effect on increasing intracellular Ca²⁺ release during cell contraction, CGRP also accelerated myoplasmic Ca²⁺ uptake during cell relaxation, because the half-maximum Ca²⁺ uptake time was shortened by 30% after CGRP stimulation. It is known that the rate of relaxation of [Ca²⁺] transient is dependent on the systolic Ca²⁺ levels in cardiac myocytes (3). Thus the enhanced myoplasmic Ca²⁺ uptake during cell relaxation is likely due to the fact of increased systolic intracellular Ca²⁺ after CGRP stimulation. Although CGRP significantly shortened myoplasmic Ca²⁺ uptake time during cell relaxation, it did not do so in the presence of L-type Ca²⁺ channel blockade by nifedipine. The signal of the [Ca²⁺] transient is usually small in the presence of L-type Ca²⁺ channel blockade. The resolution of the Ca²⁺ uptake slope of the small [Ca²⁺] transient can be confused with background noise level. This interference due to a small signal-to-noise ratio may reduce our software detection capacity for Ca²⁺ uptake slope assessment.

It is well established that phosphorylation of voltage-dependent L-type Ca²⁺ channels by protein kinase A increases transsarcolemmal Ca²⁺ influx and Ca²⁺-induced Ca²⁺ release (27). Recent biochemical evidence indicates that CGRP₁ receptors belong to a G protein-coupled receptor superfamily, its activation resulting in increased cAMP production (1). In agreement with the biochemical evidence, our results indicate that stimulation of CGRP₁ receptors in rat ventricular myocytes leads to the activation of protein kinase A, because the inhibition of protein kinase A by Rp-cAMPS eliminates CGRP-induced increase in [Ca²⁺] transients (Fig. 5). Protein kinase A phosphorylates voltage-dependent L-type Ca²⁺ channels, which leads to increased open probability of Ca²⁺ channel and enhanced Ca²⁺ influx (27). With the use of a modified buffer solution, a previous study reported that CGRP-induced cell shortening in rat ventricular myocytes was independent of activation of L-type Ca²⁺ channels and protein kinase A (2). The discrepancy between those results (2) and the results reported in this study and by others (12) is not understood. Perhaps the different experimental conditions may account, in part, for the different results. In conclusion, CGRP exerts a potent positive inotropic effect on adult rodent ventricular myocytes. This effect is mediated through the stimulation of CGRP₁ receptors. Activation of ventricular CGRP₁ receptors leads to Ca²⁺-induced Ca²⁺ release during contraction and facilitated Ca²⁺ uptake during relaxation. The mechanism for CGRP signal transduction involves the activation of protein kinase A in ventricular myocytes.

Perspectives