Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

There are several classes of positive inotropic agents. Digitalis has long been used to treat congestive heart failure, but it produces a variety of side effects, including arrhythmias (Capeller et al., 1959). Various phosphodiesterase inhibitors have been developed (Alousi et al., 1979; Maskin et al., 1983; Colucci et al., 1986), and they are more potent at enhancing contractile force than digitalis. However, large-scale clinical investigations have shown that the long-term use of these inhibitors worsens the prognosis of congestive heart failure in patients (Packer et al., 1984, 1991). This may be partly due to an increase in cyclic AMP (cAMP) in heart muscles because it has been documented that cAMP (Martorana, 1971; Lee and Downing, 1980) and -adrenoceptor stimulants (Méhes et al., 1966; Dorigotti et al., 1969), which activate adenylate cyclase, both induce pathological changes in the myocardium. Recently, calcium sensitizers were studied (Lues et al., 1993; Herold et al., 1995; Neumann et al., 1996; Zimmermann et al., 1996). These agents cause muscles to contract more effectively without causing an additional increase in intracellular calcium concentrations (Lee and Allen, 1991). Unfortunately, these agents also slow the relaxing rate of contractions, which leads to a rise in the resting tone of heart muscles (Hajjar et al., 1997).

Cyclic peptides and cyclic depsipeptides (CDPs) are produced by microorganisms; they are ring compounds composed of amino acids and/or hydroxic acids and have been shown to exhibit a large spectrum of biological activities, including immunosuppressant (Borel, 1981; Andrus and Lafferty, 1982), antibacterial (Tanaka et al., 1959), antiviral (Yeh et al., 1996), and insecticidal (Kodaira, 1961; Tamura and Takahashi, 1971; Gupta et al., 1989) activities, as well as cytotoxicity toward leukemic cells (Morel et al., 1983). It has recently been reported that destruxin E induces gene expression and marked secretion of erythropoietin in cultured cells of the epo-3 line (Cai et al., 1998) and inhibits accumulation of lipid droplets in macrophage J774 cells (Naganuma et al., 1992). However, little is known about the effects of these peptides on cardiac muscles. We have found that destruxin A (DA), destruxin B (DB), roseotoxin B (RB), and roseocardin (RC), which is a newly discovered CDP (Tsunoo et al., 1997), have positive inotropic effects on heart muscles. Furthermore, these effects are independent of cAMP levels in heart muscle tissue. This inotropic effect is accompanied by a negative chronotropic effect on the right atrium. Bradycardia may reduce the oxygen consumption of heart muscles and may decrease the incidence of arrhythmia during the treatment of heart failure. Therefore, the negative chronotropic effect of these CDPs would likely be beneficial to any therapy for heart failure patients. In addition, CDPs would not lead to an increase in the level of cAMP as happens with the use of either phosphodiesterase inhibitors or calcium sensitizers, nor would phosphodiesterase activity be affected as occurs with the use of either phosphodiesterase inhibitors or calcium sensitizers. Preliminary results have been published elsewhere (Tsunoo and Kamijo, 1997).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Contraction Experiments. The effects of CDPs on heart muscles were studied using the same methods previously reported (Tsunoo et al., 1991). Male rats (Sprague-Dawley strain; 350-450 g) were decapitated, and the right atria, papillary muscles, and trabecular muscles of the right ventricles were immediately isolated. The isolated muscles were set in a horizontal perfusion bath (0.8 ml). One end of the preparation was pinned onto a Sylgard base, and the other end was connected to a strain-gauge mechanotransducer (IM-300; Physioteck, Tokyo, Japan). A resting tension of 0.5 g was applied to the preparation. Muscle tension was recorded isometrically. The perfusion solution was a Krebs-Henseleit solution composed of 119 mM NaCl, 4.8 mM KCl, 2.5 mM CaCl₂, 1.2 mM MgSO₄, 1.2 mM KH₂PO₄, 24.9 mM NaHCO₃, and 10 mM glucose. The solution was bubbled with 95% O₂/5% CO₂, maintained at 36-37°C, and flowed in and out of the bath at a rate of 4 to 5 ml/min. The pH of the solution was 7.3 to 7.4. Isolated right atria contracted automatically. Isolated papillary and trabecular muscles from the right ventricle were stimulated with rectangular pulses of 0.5- to 1-ms duration at 1 or 2 Hz through a pair of platinum wires in the bath. The signals obtained from the mechanotransducer were fed into and stored in a digital oscilloscope (4094A; Nicolet, Madison, WI) or a personal computer with AcqKnowledge software through an interface (MP100; BIOPAC Systems, Inc., Goleta, CA).

Electrophysiological Experiments. The papillary muscle was set in a bath as described in the contraction experiments. The muscle fiber was impaled with a glass microelectrode (filamented borosilicate glass, GD-1; OD 1.0 mm; Narishige Co., Ltd., Tokyo, Japan). The tip resistance was 25 to 45 M when it was filled with 3.0 M KCl. An Ag-AgCl pellet was used as a reference electrode. Membrane potentials were measured with a preamplifer (Axoclamp-2A; Axon Instruments, Burlingame, CA), and the signals were analyzed by CAPA software (Physiotech, Tokyo, Japan). To measure the membrane resistance of the muscles, hyperpolarizing current pulses of 0.1-s duration were passed through the recording electrode by means of a bridge circuit in the amplifier.

cAMP Measurements. Trabecular strips were isolated from the right ventricles of the Sprague-Dawley rats. The preparations were set in vertical tubes filled with 10 ml of Krebs-Henseleit solution that was bubbled with 95% O₂/5% CO₂ and maintained at 37°C. The upper ends of the preparations were connected with force transducers, and tensions were recorded isometrically. The preparations were stimulated with rectangular pulses of a 3-ms duration at 1 Hz through electrodes in the tubes. The preparations were exposed to RB and RC at 600 µM for 5 min and to isoproterenol at 1 µM for 2 min. Immediately after the contractile measurements, the preparations were frozen in liquid nitrogen and stored at 30°C. The frozen samples were homogenized in trichloroacetic acid (6%) by a microhomogenizer and centrifuged at 10,000 rpm for 10 min, and the supernatants were isolated. The supernatants were washed five times with ether, the aqueous phases were lyophilized, and the residues were dissolved in methanol. Measurements of cAMP contents were carried out using the cAMP EIA system (Amersham Life Science, Clearbrook, IL).

Measurements of Na⁺,K⁺-ATPase Activities. Na⁺,K⁺-ATPase (EC 3.6.1.3) was prepared according to the method of Hara and Nakao (1981) with the use of rat kidney tissue. The effects of the CDPs on Na⁺,K⁺-ATPase activities were examined according to the method of Lane et al. (1973).

CDPs and Chemicals. The CDPs were purified from the culture broth of Trichothecium roseum TT103 as reported previously (Tsunoo et al., 1997). The CDPs were then dissolved in methanol or propylene glycol. BA 41899, which is a racemic mixture of CGP 48506 and CGP 48508 (1:1), was synthesized according to the procedure of Herold et al. (1995). The purity was >99%, and the structural integrities were verified by melting point, elemental analysis, mass spectroscopy, ultraviolet spectroscopy, and proton NMR. BA 41899 was dissolved in dimethyl sulfoxide. Atrial natriuretic peptide (rat, 1-28), bradykinin, calcitonin gene-related peptide (rat), neurokinin A, and neuropeptide Y (human and rat) were purchased from Peptide Institute (Osaka, Japan). Other chemicals were purchased from Sigma Chemical Co. (St. Louis, MO).

Statistic Analyses. The experimental results are expressed as mean ± S.E. For the contraction experiments, a paired Student's t test was used to examine the statistically significant difference between the values taken before and after drug application. In the cAMP experiments, the variances of the control and test groups were assessed with F test, and the significant differences between the control and test groups were evaluated with the unpaired Student's t test. A level of P < .05 was considered significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

CDPs. The CDPs examined in the present study (Fig. 1) are RB, RC, DA, and DB. All contain a common L-amino acid chain consisting of proline-isoleucine-methylvaline-methylalanine--alanine. In RB and RC, the proline is methylated at the 3-position. The hydroxic acids are of the D-configuration and vary with each compound.

View larger version (27K): [in this window] [in a new window]   Fig. 1.   Structures of CDPs. The hydroxic acid in each compound is of the D-configuration. All amino acids are of the L-configuration.

Inotropic and Chronotropic Effects of CDPs. RB reversibly increased the force of automatic contractions of the right atrium in the rat and prolonged the contraction interval, as shown in Fig. 2. The positive inotropic effect was concentration dependent in the range of 0.6 to 600 µM (Fig. 3). At lower concentrations, RB did not affect the beating rate of automatic atrial contractions. However, at higher concentrations, RB reduced the rate of these contractions. Similarly, RB increased the force of the contractions of the ventricular muscles resulting from electrical stimulation. The concentration-response relationships of papillary and trabecular muscle contractions of the right ventricle are shown in Fig. 4A. Although RB did not produce a strong inotropic effect in trabecular muscles, RB clearly caused a concentration-dependent increase in the contractile force of the papillary muscles.

View larger version (27K): [in this window] [in a new window]   Fig. 2.   Effects of RB (A and B) and RC (C and D) on automatic contractions of an isolated right atrium of the rat. A and C, RB (20 µM) and RC (20 µM) were applied to an isolated right atrium (solid line). The data for the 6-min application period were omitted (dashed line). The traces labeled 3 were obtained 22 min after the start of washing. The recordings of the parts numbered in A and C are shown in an expanded time scale in B and D, respectively. 1, time immediately before application of each agent. 2, steady state of the response to each agent. 3, 22 min after washing with agent-free solution. The calibration for A also applies to C. The calibration for B also applies to D.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Concentration-response relationships for inotropic and chronotropic effects of RB and RC on automatically contracting atria. A, concentration-response relationships for RB-induced effects. B, concentration-response relationships for RC-induced effects. Data are means ± S.E. (n = 4-6) of values relative to those obtained immediately before application of the agents.

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Concentration-response relationships for inotropic effects of RB and RC on papillary and trabecular muscles of the right ventricles. Papillary muscles () and trabecular muscles () were electrically stimulated. A, concentration-response relationships for RB-induced effects. B, concentration-response relationships for RC-induced effects. Data are means ± S.E. (n = 5-8) of values relative to those obtained immediately before application of the agents.

Effects on Time Course of Twitch Contraction. The CDP-induced changes in the time course of twitch contractions are summarized in Table 1. All of the CDPs, as well as isoproterenol, caused a prolongation in the time from rise to peak in a dose-dependent manner. BA 41899, a calcium sensitizer that is a racemic mixture of active CGP48506 and inactive CGP48508 in a ratio of 1:1, increased the time interval between the rise and the contraction peak much more than did the CDPs. Isoproterenol significantly shortened the half-decay time. RB, DA, and DB did not significantly affect the half-decay time. RC at 60 µM significantly prolonged the half-decay time. However, the extent of the RC-induced prolongation was much less than that of BA 41899. CDPs increased the maximal rates of rise and relaxation. Similar to isoproterenol, there was a tendency for the CDPs to increase the relaxation rate rather than the rising rate. On the other hand, BA 41899 affected both rates equally.

Pharmacological Properties of CDP Effects. We examined whether the effects of the CDPs mimicked those of known receptor agonists on the right atria. As mentioned, the CDPs caused an increase in the contractile force, without shortening contraction intervals. The examined agonists were methoxamine (3 µM), clonidine (3 µM), isoproterenol (50 nM), histamine (60 µM), serotonin (60 µM), atrial natriuretic peptide (rat) (0.5 µM), bradykinin (5 µM), calcitonin gene-related peptide (rat, 1-28) (1 µM), endothelin-1 (20 nM), neurokinin A (5 µM), and neuropeptide Y (0.5 µM). Among them, isoproterenol, serotonin, calcitonin gene-related peptide, and endothelin-1 significantly increased the contractile force to 4.75 ± 1.04, 1.65 ± 0.05, 2.49 ± 0.35, and 1.48 ± 0.14 times the control (n = 4), respectively. However, each of them significantly shortened the contraction interval to 0.53 ± 0.02, 0.83 ± 0.01, 0.74 ± 0.04, and 0.89 ± 0.02 times the control (n = 4), respectively.

Non-cAMP Dependence of CDP Effects. Many recently developed cardiotonic agents have a phosphodiesterase-inhibiting activity that increases cAMP levels in heart muscle tissue. Pharmacological experiments were carried out to determine whether the CDPs produce their effects by increasing the cAMP levels in papillary muscles. The positive inotropic effects of RB, RC, and DA were not affected by 3-isobutyl-1-methylxanthine (IBMX), which is a phosphodiesterase inhibitor, as shown in Fig. 5. In contrast, the effect of isoproterenol, which activates adenylate cyclase, was significantly potentiated by IBMX.

View larger version (14K): [in this window] [in a new window]   Fig. 5.   Effects of IBMX on the CDP- and isoproterenol-induced increase in the amplitude of contractions of the papillary muscles. Open bars, responses of the muscles to CDPs (20 µM) or isoproterenol (33 nM). Filled bars, responses to CDPs or isoproterenol in the presence of IBMX (5 µM). Values are mean ± S.E. of the contraction amplitudes relative to those obtained immediately before application of the agents. Six experiments were performed for RB, five experiments were performed for RC and DA, and four experiments were performed for isoproterenol. *P < .05.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Effects of carbachol on CDP- and isoproterenol-induced increase in the amplitude of contractions of the papillary muscles. Open bars, responses of the muscles to CDPs (20 µM) or isoproterenol (33 nM). Filled bars, responses to CDPs or isoproterenol in the presence of carbachol (30 µM). Values are mean ± S.E. (n = 4) of the contraction amplitudes relative to those obtained immediately before application of the agents. **P < .01.

View larger version (19K): [in this window] [in a new window]   Fig. 7.   Relationship between contractile increase and cAMP levels. Trabecular muscles of the right ventricle were stimulated at 1 Hz and exposed to CDPs (600 µM) or isoproterenol (1 µM). A, relative contraction amplitude. Values are mean ± S.E. of the contraction amplitudes relative to those obtained immediately before application of the agents. B, cAMP levels. After termination of incubation of the agents, the muscles were frozen in liquid nitrogen, and cAMP levels were measured. CDPs and isoproterenol were dissolved in methanol. Values are means ± S.E. Eight experiments were performed for RB, six experiments were performed for RC, and seven experiments were performed for isoproterenol and methanol. **P < .01 versus methanol experiment.

Effects of CDPs on Na⁺,K⁺-ATPase. We examined whether the CDPs affect Na⁺,K⁺-ATPase isolated from rat kidneys. RB, RC, and DA at 100 µM did not markedly affect the activity of the enzyme, with values of 97 ± 3% (n = 3), 93 ± 1% (n = 3), and 89 ± 3% (n = 3) of the maximal activity, respectively. On the other hand, ouabain at 100 µM inhibited the enzyme by up to 46 ± 2% (n = 3) of the maximal activity.

Effects on Resting Membrane Potential and Apparent Membrane Resistance. CDPs did not significantly affect the resting membrane potential and apparent input resistance of papillary muscle. The resting membrane potential and the input resistance were 73.2 ± 5.4 mV (n = 5) and 2.43 ± 0.70 M (n = 5), respectively, in the absence of RB and 73.0 ± 4.6 mV and 2.47 ± 0.84 M in the presence of 60 µM RB. The resting membrane potential and the input resistance were 76.1 ± 4.1 mV (n = 5) and 2.23 ± 0.80 M (n = 5), respectively, in the absence of RC and 75.0 ± 4.1 mV and 2.22 ± 0.87 M in the presence of 60 µM RC. The resting membrane potential and the input resistance were 76.1 ± 3.5 mV (n = 5) and 1.28 ± 0.16 M (n = 5), respectively, in the absence of DB and 73.5 ± 3.0 mV and 1.32 ± 0.13 M in the presence of 60 µM DB.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The present study indicates that the CDPs produce complementary effects on cardiac functions, namely, positive inotropic and negative chronotropic effects. As a positive inotrope, the cyclic peptidyl structure is unique among the various cardiotonics discovered and developed to date. Possibly as the result of this, pharmacological experiments suggest that they do not activate classic receptors. It also appears that inhibition of Na⁺,K⁺-ATPase is not involved in the CDP-induced increase in contractile force in ranges of less than 100 µM. Enhancement of contractile force by these peptides is not affected by IBMX or carbachol. Furthermore, cAMP levels in heart muscles are not affected by these peptides. These results strongly suggest that the positive inotropy of these peptides is independent of the cAMP level. Measurements of the contraction parameters indicate that the half-decay time of a relaxation is slightly prolonged by the peptides as revealed by the application of 60 µM RC (Table 1), although this effect in all other cases is not statistically significant. However, compared with the effects of established calcium-sensitizing compounds such as CGP 48506 (Neumann et al., 1996; Zimmermann et al., 1996) and EDM 57033 (Lues et al., 1993), this effect is too weak to explain the CDP-induced positive inotropy. Because they do not affect the resting membrane potential or the input resistance of the papillary muscles, it also seems unlikely that they act as an ionophore. This is supported by the fact that destruxins do not increase transport of Na⁺, K⁺, Ca²⁺, and Mg²⁺ in Pressman cells (Abalis, 1981; Samuels et al., 1988) and that they do not increase the release of Ca²⁺ ions from human red blood cell ghosts (Samuels et al., 1988). Our preliminary experiments imply that the CDPs prolong the action potential duration (data not shown), suggesting that the peptides inhibit outward K⁺ currents and/or increase inward Ca²⁺ currents. Depression of the K⁺ currents as well as increase in the Ca²⁺ currents would lead to an increased flux of Ca²⁺ ions into the cardiac myocyte, which would enhance the contractile force of the myocyte.

Digitalis produces bradycardia by vagal stimulation in a whole animal, and in heart tissue, it directly enhances automaticity (Hoffman and Bigger, 1990). On the contrary, RB and RC slow the beating rate of automatic contractions in the isolated atria, suggesting that the CDP-induced negative chronotropy is due to its direct action on the right atria. Concentration-response relationships in the right atrium (Fig. 3) show that there is a discrepancy between inotropic and chronotropic effects. In the concentration ranges in which the contractile force clearly increases, the peptides do not markedly change the contraction interval. This suggests that the mechanism for the chronotropic effect may differ from that for the inotropic effect.

It has been reported that destruxin E is a potent inhibitor of ATP-dependent acidification of endosomes and lysosomes and that it may act by blocking vacuolar H⁺-ATPase (Naganuma et al., 1992; Togashi et al., 1997). Although the present study shows that RB, RC, DA, and DB at levels up to 100 µM do not markedly affect Na⁺,K⁺-ATPases, there is a possibility that these CDPs inhibit acidification of the organelles due to a blockade of vacuolar H⁺-ATPases, which then brings about changes in strength and interval of contractions. More experiments are needed to elucidate the mechanisms underlying the positive inotropic and negative chronotropic effects of CDPs.

At present, angiotensin-converting enzyme inhibitors have become the first choice in therapy for heart failure. The administration of -adrenergic antagonists is also widely used (Lechat et al., 1998). These agents do not produce positive inotropic action. This trend in therapy may in part stem from the lack of success of positive inotropes for long-term use (Packer et al., 1984, 1991). Early works indicated that cAMP (Lee and Downing, 1980) and isoproterenol (Martorana, 1970; Tanaka, 1981) cause pathological changes in the heart. It was recently reported that -adrenergic stimulation leads to myocardial apoptosis mediated by protein kinase A (Communal et al., 1998). These results suggest that overproduction of this nucleotide in the myocardium may be a major drawback to the long-term use of phosphodiesterase inhibitors. Beneficial effects of -adrenergic antagonists may be partly due to an inhibition of cAMP production evoked by -adrenergic stimulation because basal sympathetic activities sustain background cAMP levels and early work in this area found that the antagonists reduce isoproterenol-induced damage in the heart (Méhes et al., 1966; Dorigotti et al., 1969). Another candidate is a calcium-sensitizing agent. Although many such agents display phosphodiesterase-inhibiting activity, a pure calcium sensitizer, BA 41899, has been developed. However, as shown in this study, BA 41899 slows the relaxing process after twitch contraction. This would lead to the possibility that calcium sensitizers might impair diastolic relaxation in vivo; this would be especially disadvantageous because during heart failure, there already is a relaxation disorder after contraction. However, these possible limitations do not diminish the usefulness of positive inotropes because digitalis is still frequently used clinically and has been recently reevaluated (DiBianco et al., 1989; The Digitalis Investigation Group, 1997). Therefore, if inotropic action could be produced by an alternate mechanism, this would be a clear improvement for the treatment of heart failure and would complement the current angiotensin-converting enzyme inhibitor/-adrenergic antagonist combination therapy. This study shows that the CDPs elicit a positive inotropic action independent of a cAMP increase without causing a slowing in the relaxation rate and elicit a negative chronotropic effect. These characteristics of effects of CDPs may aid the treatment of heart failure without worsening the pathophysiological changes in the failed heart.