Effects of EMD 57033 on Contraction and Relaxation in Isolated Rabbit Hearts
Background Ca2+ sensitizers are reported to enhance contractility with modest effects on energy utilization. In the present study we assessed the effects of the relatively “pure” Ca2+ sensitizer EMD 57033 on mechanical performance and energy consumption in the beating heart.
Methods and Results In 10 isolated, red blood cell–perfused rabbit hearts the effects of EMD 57033 (5.0 to 5.8 μmol/L) on left ventricular (LV) pressure and O2 consumption (V̇o2) were examined at heart rates of 100 and 150 beats per minute (bpm) and perfusate [Ca2+] ([Ca2+]o) of 2.5 and 1.0 mmol/L (isovolumic contractions). LV developed pressure and maximum dP/dt increased, but less so at 150 bpm or 1.0 mmol/L [Ca2+]o. End-diastolic pressure also increased, more so at 150 bpm or 1.0 mmol/L [Ca2+]o. EMD 57033 decreased time to peak isovolumic pressure (Tmax) and prolonged time to 50% pressure decline (T1/2). These changes were greater at slower heart rate or lower [Ca2+]o. The magnitude of increased V̇o2 with EMD 57033 was greater at 100 bpm than 150 bpm but unaffected by [Ca2+]o. We then investigated the influence of ejection on the response to EMD 57033 (n=7). The increase in developed pressure with EMD 57033 was greater for ejecting than isovolumic beats (25.5±10.2 versus 14.7±7.5 mm Hg at 100 bpm, P<.01), while the increase in end-diastolic pressure was less (P=NS). The increase in V̇o2 was significantly greater for ejecting than isovolumic beats (0.027±0.013 versus 0.020±0.009 mL O2/beat per 100 g at 100 bpm, P<.01).
Conclusions EMD 57033 enhances contractility and prolongs relaxation. Its effects are modulated by heart rate, [Ca2+]o, and contraction mode, with positive inotropic effects being more prominent for ejecting beats.
Recent advances in the development of inotropic agents have resulted in the introduction of thiadiazinone derivatives. The compound EMD 53998 is one such compound. It has a dual mechanism of action1 : an increase in the Ca2+ sensitivity of the contractile proteins and inhibition of phosphodiesterase III. EMD 53998 is a racemic, equimolar mixture of two optical enantiomers, namely, (+)-enantiomer EMD 57033 and (−)-enantiomer EMD 57439.1 Recent studies in skinned cardiac fibers2 3 and isolated papillary muscles4 have shown that the Ca2+-sensitizing activity is predominantly attributable to the (+)-enantiomer EMD 57033, while inhibition of phosphodiesterase III is mainly due to the (−)-enantiomer EMD 57439. The Ca2+-sensitizing activity of EMD 57033 was demonstrated in skinned cardiac trabeculae by showing that 10 μmol/L of this drug shifted the relation between isometric tension and Ca2+ concentration to the left.1 In intact canine ventricular myocytes, EMD 57033 (up to 10 μmol/L) exerted a significant positive inotropic effect accompanied by a progressive reduction in diastolic cell length without a clear change in the indo-1 intracellular Ca2+ transient.2 3 EMD 57033 also is reported to decrease the energy cost of active tension development in isolated ventricular trabeculae.5 Based on these observations, in the beating heart EMD 57033 would be anticipated to enhance contractility while possibly having relatively modest effects on energy utilization. However, a possibly adverse effect of EMD 57033 would be slowing of relaxation and elevation of diastolic pressure.
Previous studies of the effects of EMD 57033 have been undertaken in skinned cardiac fibers, isolated myocytes,1 2 3 and isolated papillary muscle preparations.4 Although a few studies of other Ca2+ sensitizers have been performed in excised canine heart preparations,6 7 the latter agents have had important effects besides Ca2+ sensitization. The effects of a drug such as EMD 57033, whose main mechanism of action is thought to be Ca2+ sensitization, have not been reported in a beating heart preparation to date. The purpose of the present study was to investigate the effects of EMD 57033 on mechanical performance and myocardial energy consumption in a red blood cell–perfused, isolated rabbit heart preparation. We performed two series of experiments. In the first, the effects of the drug were examined at normal and relatively low extracellular Ca2+ concentration ([Ca2+]o) and at varying heart rates during isovolumic contractions (Ca2+ series). In the second, we investigated whether ejection modifies cardiac mechanoenergetic responses to the drug (ejecting contraction series).
Heart Preparation for Ca2+ Series
We used an isolated rabbit heart preparation perfused with bovine red blood cells suspended in Krebs-Henseleit buffer.8 Ten adult New Zealand White rabbits (body weight, 2.87±0.26 kg [mean±SD]) were used. After premedication with fentanyl (0.044 mg/kg IM) and droperidol (2.2 mg/kg IM), each rabbit was anesthetized with ketamine hydrochloride (20 mg/kg IM) and xylazine (1 mg/kg IM). A tracheotomy was performed, and the rabbit was ventilated with a Harvard respirator. The chest was opened at the median line of the sternum, and the pericardium was incised widely. After administration of heparin (1000 U/kg IV), both pulmonary hili and the superior and inferior venae cavae were ligated simultaneously. A perfusion tube was inserted into the ascending aorta. The coronary arteries were immediately perfused via the aortic root with the perfusate containing red blood cells, and the heart was removed from the chest. Interruption of coronary perfusion lasted no longer than 20 seconds.
After the heart was excised and hung from the coronary perfusion tube, a small vent was inserted into the left ventricular (LV) apex to drain Thebesian effluent. To allow subsequent control of heart rate by electrical pacing, the right atrium (RA) was then opened and atrioventricular block was induced by electrical coagulation of the atrioventricular nodal region of the RA. Destruction of the atrioventricular node was confirmed by the appearance of atrioventricular dissociation. A flexible tube was inserted through the RA into the right ventricle (RV), and the RA and the pulmonary artery were ligated. The left atrium then was opened widely and the chordae tendineae were cut. A collapsed, thin Latex balloon (unstressed volume, 3 mL) was placed in the LV through the mitral orifice. The balloon was connected to a Gould P23 XL pressure transducer and a 2-mL graduated syringe with a rigid tube. Pacing electrodes connected to an electronic stimulator (model S9, Grass Instruments Co) were attached to the LV surface. The RV was kept collapsed by continuous hydrostatic drainage to minimize right ventricular O2 consumption. The heart was placed in a chamber with a heating jacket, and the temperature of the heart was maintained at 35° to 37°C. Coronary blood flow was measured by timed collections in a graduated cylinder of the coronary venous drainage from the RV. This neglects LV Thebesian flow, which is a very small fraction of total flow.9 Coronary arteriovenous O2 content difference (Avo2Δ) was measured continuously with an AVOX system, which was calibrated with a Lex-O2-Con oximeter (Lexington).
Heart Preparation for Ejecting Contraction Series
Seven male New Zealand White rabbits (body weight, 2.89±0.18 kg) were used. The rabbits were anesthetized, and the hearts were excised from the chest in the same manner as in the Ca2+ series. After the left atrium was opened and chordae tendineae were cut, the heart was positioned such that the tip of the cylinder of a servo motor system (see below) could be positioned at the mitral annulus. An empty, thin Latex balloon (unstressed volume, 3 mL) attached to a flared end of the cylinder of the servo motor was placed in the LV through the mitral orifice. A string was placed around the mitral annulus and tightened to secure the heart at the end of the cylinder. The atrioventricular nodal region was coagulated, and pacing electrodes were attached to the LV surface.
LV volume was controlled by a volume servo system composed of a linear motor, a piston-cylinder device, a linear variable displacement transducer (LVDT, model 1000-0012, Trans-Tek Inc), a custom-designed analog position controller, and a custom-designed high current amplifier. The piston-cylinder device was composed of a modified 3-mL glass syringe (Popper & Sons Inc) and an acrylic supporting apparatus with two side ports. The piston barrel was connected to one end of the metal armature shaft of the linear motor. A 5F micromanometer catheter (model MPC-500, Millar Instruments) was introduced into the center of the balloon via a side port. The LVDT mounted on the back of the linear motor had a frequency response of 1 kHz and a resolution of ±0.015 mm. It allowed precise measurement of the piston position and therefore instantaneous LV volume with proper calibration.
The piston position was controlled by a classic analog proportional-integral-differential (PID) compensator, which received a volume command from a control computer. The output signal from the PID compensator was sent to the high current amplifier and was then delivered to the linear motor armature to change the piston position to match the volume command. Thus, the volume servo system controlled the volume in the LV balloon according to the volume command from the control computer.
A personal computer (Gateway 2000) was used to control the volume servo system. We used an analog-to-digital convertor to sample the instantaneous LV pressure and volume and a digital-to-analog convertor to send the volume command to the analog PID controller of the volume servo system. A digital intake/output was used to trigger a stimulator (model S9, Grass Instruments Co) to pace the heart. Software developed in our laboratory controlled stroke volume, timing for start/stop of ejection/filling, and stimulation frequency.
Preparation of Perfusate
We used a perfusate consisting of bovine red blood cells suspended in Krebs-Henseleit buffer (hematocrit, 35%). The preparation of the red blood cells and composition of the buffer has been described in detail previously.8 For these studies, buffer with Ca2+ concentration of both 2.5 and 1.0 mmol/L was prepared. The perfusate was transported to the perfusion tubing by a variable-flow pump (Masterflex, Cole-Palmer) and equilibrated in an oxygenator with 98% O2 and 2% CO2 to achieve a Po2 of over 100 mm Hg and a Pco2 of approximately 40 mm Hg. Coronary perfusion pressure was controlled by a pressurized arterial reservoir connected to a pressure regulator and compressed air. The temperature of the perfusate was maintained at 35° to 37°C with water jackets around the oxygenator and the pressurized arterial reservoir in the perfusion circuit. The perfusate was not recirculated.
Preparation and Concentration of EMD 57033
EMD 57033 was provided by Pharmaceutical Research, E. Merck. Since this compound is almost insoluble in water, 1 mmol/L stock solution was prepared in dimethyl sulfoxide (DMSO) before each experiment. The perfusate containing red blood cells was prepared in two containers, and the 1-mmol/L stock solution was added to one of those containers to a 10 μmol/L final concentration of EMD 57033. This resulted in 1% DMSO in the final perfusate. We added 1% DMSO alone to the other container. The perfusate containing DMSO only was used during control contractile state experiments. This perfusate concentration of EMD 57033 was selected after pilot experiments disclosed that at [Ca2+]o of 2.5 mmol/L concentrations significantly lower resulted in very small to nonexistent increases in LV peak systolic pressure under isovolumic contraction conditions at constant LV volume, while significantly higher concentrations resulted in marked increases in LV end-diastolic pressure. Thus, this concentration resulted in an “effect” on peak systolic pressure comparable to what might be used as an end point clinically without producing effects on end-diastolic pressure that would be unacceptable. The concentration of EMD 57033 in perfusate supernatant was measured in three of the experiments (see “Acknowledgments”). Samples were taken from the coronary perfusion tube just above the heart. The actual concentration of EMD 57033 in the supernate ranged from 5.0 to 5.8 μmol/L. Thus, it would appear that the red blood cells adsorbed or absorbed a significant amount of the drug.
The 10 hearts were divided into two groups, 5 hearts perfused with perfusate containing 2.5 mmol/L Ca2+ (2.5 mM [Ca2+]o group) and the remaining 5 hearts with perfusate containing 1.0 mmol/L Ca2+ (1.0 mmol/L [Ca2+]o group). In both groups, measurements were made during steady state isovolumic contractions. Coronary perfusion pressure was maintained constant at about 75 mm Hg throughout each experiment. The heart was stimulated via the electrodes attached to the LV surface. LV volume was initially adjusted to obtain an LV peak systolic pressure of approximately 100 mm Hg at a heart rate of 100 beats per minute (bpm). LV volume was kept constant at this value throughout the experiment. We then waited 4 to 5 minutes to attain stable conditions before beginning the procedures outlined below.
In each heart, data were first collected before EMD 57033 was administered (control contractile state). LV pressure, coronary flow, coronary perfusion pressure, and Avo2Δ were recorded at 100 bpm. In 2 of 10 hearts we used a heart rate of 120 bpm instead of 100 bpm because stable contractions could not be obtained at 100 bpm. For convenience, we will continue to refer to a heart rate of 100 bpm for all hearts. Stimulation frequency then was increased to 150 bpm, and measurements were repeated during steady state conditions at this heart rate. After measurements were completed under control conditions, the perfusate was switched to that containing EMD 57033. Heart rate was kept at 150 bpm; 10 to 15 minutes were required to obtain stable enhancement of contractile state. When the LV peak and minimum pressure reached a plateau, measurements were repeated at heart rates of 100 and 150 bpm. After the recordings with enhanced contractile state, the perfusate was switched to that without EMD 57033. We waited until LV peak pressure declined to about the same level as before EMD 57033 was administered. This required 10 to 15 minutes. Heart rate was kept at 150 bpm during this period. We then repeated the same measurements as in the control and enhanced contractile states at the same two heart rates (repeat control state). At the end of the experiment, the LV and RV weights were measured. LV weight was 5.29±0.77 g and RV weight was 1.76±0.29 g for the Ca2+ series.
Ejecting Contraction Series
All hearts were perfused with 2.5 mmol/L [Ca2+]o. The heart was paced at 100 bpm, and LV end-diastolic volume was adjusted to obtain a LV peak isovolumic pressure of approximately 100 mm Hg. LV end-diastolic volume was kept constant at this value throughout the experiment. After the preparation had stabilized, LV pressure, coronary flow, coronary perfusion pressure, and Avo2Δ were recorded during isovolumic and ejecting contractions at 100 and 150 bpm. We waited 2 to 3 minutes to obtain steady state conditions after each change of contraction mode or heart rate. The stroke volume was chosen such that ejection fraction was 50% to 60%. Starting and stopping of ejection and filling were timed so that the LV pressure waveform appeared as physiological as possible. After the measurements under control conditions were completed, the perfusate was switched to that containing EMD 57033. The heart rate was kept at 150 bpm. We waited 10 to 15 minutes to obtain a stable enhancement of contractile state. Measurements then were repeated under isovolumic and ejecting contractions at the same two heart rates as in the control contractile state. After the recordings with enhanced contractile state, the heart rate was set at 150 bpm and the perfusate was switched to that without EMD 57033. After the LV peak pressure declined to about the same level as before EMD 57033 was administered, we repeated the same measurements as in the control and enhanced contractile states at the same two heart rates as in the control contractile state (repeat control state). The LV weight was 5.44±0.41 g and the RV weight was 2.00±0.31 g for the ejecting contraction series.
LV pressure, coronary perfusion pressure, and Avo2Δ were recorded on a pen recorder and stored on a hard disk at 5-ms sampling intervals for off-line data analysis with a personal computer (Gateway 2000). O2 consumption (V̇o2) per minute was calculated as the product of coronary flow (mL/min) and Avo2Δ (vol%) and was divided by heart rate to yield total V̇o2 per beat (in mL O2/beat). V̇o2 was normalized per 100 g weight to give V̇o2 in mL O2/beat per 100 g. LV volume was determined as the sum of the volume of water within the LV balloon and the volume of the balloon walls and connector within the left ventricle.10 LV developed pressure was defined as the difference between peak and minimum LV pressures during one cardiac cycle. End diastole was defined as the time when LV positive dP/dt increased to 10% of its peak value. Duration of contraction was estimated by Tmax, the time from end diastole to LV peak pressure. Relaxation duration was quantified as T1/2, which was defined as the time it took the LV pressure to fall to a value half of the peak pressure.
Data are presented as mean±SD. Three-way or two-way ANOVA for repeated measures was used to detect differences for all data.11 When the F test indicated a significant difference among the conditions, the Student-Newman-Keuls test was used to test the significance of difference between specific conditions. A value of P<.05 was accepted as the level of significance.
LV volume was 1.33±0.18 mL and 1.59±0.18 mL for the 2.5 mmol/L [Ca2+]o group and the 1.0 mmol/L [Ca2+]o group, respectively. Although LV volume for the 1.0 mmol/L [Ca2+]o group was slightly larger, this difference was not statistically significant (P=.06).
Effect of EMD 57033 in the 2.5 mmol/L [Ca2+]o Group
Fig 1⇓ shows recordings of LV pressure and LV dP/dt before and during administration of EMD 57033 in one heart of the 2.5 mmol/L [Ca2+]o group at 100 bpm. There was a marked increase in LV peak pressure and maximum LV dP/dt (LV dP/dtmax) at a constant LV volume of 1.65 mL with EMD 57033. LV end-diastolic pressure increased from 2 to 7 mm Hg. In this example LV minimum dP/dt (LV dP/dtmin) decreased from −555 to −723 mm Hg/s, and its timing was slightly delayed. However, T1/2 increased from 125 ms under the control condition to 163 ms during EMD 57033 administration. V̇o2 was 0.036 mL O2/beat per 100 g before and 0.065 mL O2/beat per 100 g during administration of EMD 57033. Other hearts in the 2.5 mmol/L [Ca2+]o group showed similar results. Figs 2A through 2G show changes in LV mechanical variables before, during, and after administration of EMD 57033 for the 2.5 mmol/L [Ca2+]o group at 100 and 150 bpm. Under initial control conditions, none of the measured variables were significantly different between rates of 100 and 150 bpm. LV peak pressure significantly increased with EMD 57033 at both heart rates (Fig 2A⇓). Although LV end-diastolic pressure tended to increase, the change was statistically insignificant at both heart rates (Fig 2B⇓). LV developed pressure and LV dP/dtmax significantly increased with EMD 57033 at both heart rates (Figs 2C⇓ and 2E⇓). Interestingly, LV dP/dtmin significantly decreased (that is, peak rate of pressure fall increased) with EMD 57033 at both rates (Fig 2D⇓), but T1/2 was significantly prolonged, indicating slowed relaxation (Fig 2F⇓). Tmax was shortened significantly with EMD 57033 at 100 bpm. Although Tmax tended to be abbreviated with EMD 57033 at 150 bpm, this change did not reach statistical significance (P=.07) (Fig 2G⇓). During the repeat control state, all mechanical variables except Tmax were not significantly different from those during the initial control state. Tmax remained significantly shortened at 100 bpm even after EMD 57033 was discontinued.
Table 1⇓ summarizes energetic variables before, during, and after administration of EMD 57033 for both the 2.5 mmol/L [Ca2+]o and 1.0 mmol/L [Ca2+]o groups. For the 2.5 mmol/L [Ca2+]o group, coronary blood flow significantly increased with EMD 57033 at both heart rates and remained significantly greater under the repeat control state at 150 bpm (P<.05). Avo2Δ showed no significant change in response to EMD 57033. Myocardial V̇o2 significantly increased with EMD 57033 at both heart rates and returned toward its initial level after discontinuation of EMD 57033 at both heart rates.
Effect of EMD 57033 in 1.0 mmol/L [Ca2+]o Group
Recordings of LV pressure and LV dP/dt at 100 bpm in one of the 1.0 mmol/L [Ca2+]o group hearts before and during EMD 57033 administration are shown in Fig 3⇓. LV peak pressure and dP/dtmax markedly increased with EMD 57033. Tmax was abbreviated from 219 to 190 ms. In this example, LV end-diastolic pressure increased from 4 to 16 mm Hg and T1/2 prolonged from 131 to 231 ms. During EMD 57033 administration in this group there was a consistent inflection point during LV pressure decay, and the time of maximum relaxation rate occurred much later than in the control beat. In contrast to the 2.5 mmol/L [Ca2+]o group, LV dP/dtmin increased slightly, from −617 to −524 mm Hg/s, with EMD 57033. Figs 4A through 4G show variables of LV mechanics before, during, and after administration of EMD 57033 for the 1.0 mmol/L [Ca2+]o group. Figs 5A through 5G compare the magnitude of changes in mechanical variables in response to EMD 57033 between the two heart rates at each [Ca2+]o. (Differences related specifically to [Ca2+]o are described in the next section.) All measured variables were not significantly different in the 1.0 mmol/L [Ca2+]o group between the two heart rates under initial control conditions except for LV developed pressure, which was significantly lower at 150 bpm than at 100 bpm (P<.05) (Fig 4⇓). As in the 2.5 mmol/L [Ca2+]o group, LV peak pressure significantly increased with EMD 57033 at both heart rates in the 1.0 mmol/L [Ca2+]o group (Fig 4A⇓). In contrast to the 2.5 mmol/L [Ca2+]o group, LV end-diastolic pressure significantly increased with EMD 57033 (Fig 4B⇓). The increase in LV end-diastolic pressure at 150 bpm was significantly greater than that at 100 bpm (Fig 5B⇓). LV developed pressure increased significantly at 100 bpm but did not significantly change at 150 bpm (Fig 4E⇓). LV dP/dtmax also increased significantly only at 100 bpm (Fig 4C⇓). The magnitude of increase in LV developed pressure and LV dP/dtmax was significantly greater at 100 bpm than at 150 bpm (Figs 5C⇓ and 5E⇓). LV dP/dtmin did not change significantly at either heart rate with administration and discontinuation of EMD 57033 (Fig 4D⇓). T1/2 was significantly prolonged with EMD 57033 at both heart rates (Fig 4F⇓), and the magnitude of prolongation was significantly greater at 100 bpm than at 150 bpm (Fig 5F⇓). There was an abbreviation in Tmax at 100 and 150 bpm, and its magnitude was significantly greater at 100 bpm than at 150 bpm (Fig 5G⇓). During the repeat control state, all mechanical variables except Tmax were not significantly different from those under initial control condition at both heart rates. Under these conditions, Tmax remained significantly shorter compared with the initial control contractile state at both heart rates (P<.01 for 100 bpm, P<.05 for 150 bpm) (Fig 4G⇓).
The response of energetics variables to EMD 57033 in the 1.0 mmol/L [Ca2+]o group was similar to that in the 2.5 mmol/L [Ca2+]o group (Table 1⇑), that is, an insignificant change in Avo2Δ and a significant increase in coronary blood flow and V̇o2. However, V̇o2 during EMD 57033 administration was significantly lower in the 1.0 mmol/L [Ca2+]o group than in the 2.5 mmol/L [Ca2+]o group at both heart rates (both P<.01). Although coronary blood flow remained at a significantly higher level, V̇o2 declined to its initial level after discontinuation of EMD 57033.
Comparison of Results at the Different [Ca2+]o
Comparisons of the magnitude of changes in mechanical parameters with EMD 57033 between the 2.5 mmol/L and the 1.0 mmol/L [Ca2+]o groups at a given heart rate are presented in Figs 5A through 5G. ANOVA did not indicate any significant effect of the different [Ca2+]o on the change in LV peak pressure (Fig 5A⇑). The absolute value for LV end-diastolic pressure during EMD 57033 administration was significantly greater in the 1.0 mmol/L [Ca2+]o group than in the 2.5 mmol/L [Ca2+]o group at both heart rates (both P<.01). The increase in LV end-diastolic pressure with EMD 57033 was also markedly and significantly greater in the 1.0 mmol/L [Ca2+]o group than in the 2.5 mmol/L [Ca2+]o group at both heart rates (Fig 5B⇑). The increase in LV dP/dtmax was of similar magnitude in the 2.5 mmol/L and 1.0 mmol/L [Ca2+]o groups at 100 bpm. At 150 bpm, the magnitude of the increase in LV dP/dtmax was significantly greater in the 2.5 mmol/L [Ca2+]o group than in the 1.0 mmol/L [Ca2+]o group (Fig 5C⇑). The LV developed pressure results were similar to the LV dP/dtmax results (Fig 5E⇑). Thus, based on changes in developed pressure and LV dP/dtmax, a positive inotropic effect of EMD 57033 was not detectable at higher heart rate and lower [Ca2+]o. The effect of EMD 57033 on LV end-diastolic pressure was significantly more prominent at lower [Ca2+]o (Fig 5B⇑). Thus, in contrast to the positive inotropic effect, the increase in LV end-diastolic pressure was greatest at higher heart rate and lower [Ca2+]o.
EMD 57033 produced complex effects on the duration of contraction and relaxation. Duration of contraction expressed as Tmax was significantly abbreviated in both [Ca2+]o groups at both heart rates. Abbreviation of Tmax was significantly greater in the 1.0 mmol/L [Ca2+]o group than in the 2.5 mmol/L [Ca2+]o group at 100 bpm but was not significantly different at 150 bpm (P=.23) (Fig 5G⇑). LV dP/dtmin significantly decreased with EMD 57033 at both heart rates in the 2.5 mmol/L [Ca2+]o group but was unaffected in the 1.0 mmol/L [Ca2+]o group in conjunction with marked delay of the timing of LV dP/dtmin and changes in the morphology of the pressure fall tracing. The magnitude of LV dP/dtmin changes was significantly different only at 150 bpm when compared with the 1.0 mmol/L [Ca2+]o group (Fig 5D⇑). Despite an increase in the peak rate of LV pressure decay, relaxation duration quantified as T1/2 was prolonged with EMD 57033 in both [Ca2+]o groups. The magnitude of the prolongation in T1/2 was significantly greater in the 1.0 mmol/L [Ca2+]o group than in the 2.5 mmol/L [Ca2+]o group at both heart rates (Fig 5F⇑). Thus, T1/2 changed in a reciprocal fashion with contraction duration, that is, more prolongation at lower heart rates and, as with Tmax, effects were more prominent at low [Ca2+]o. However, changes in T1/2 were consistently larger than changes in Tmax, and significant prolongation occurred even with normal [Ca2+]o.
Table 2⇓ presents magnitude of changes in energetic variables with EMD 57033 administration in the Ca2+ series. The magnitude of increase in V̇o2 was significantly greater at 100 bpm than at 150 bpm in both [Ca2+]o groups (Table 2⇓). However, the increase in V̇o2 with EMD 57033 was similar in the 2.5 mmol/L [Ca2+]o and the 1.0 mmol/L [Ca2+]o groups at a given heart rate, and ANOVA did not show a significant effect of [Ca2+]o on the change in V̇o2 with EMD 57033 (P=.48).
Ejecting Contraction Series
LV end-diastolic volume (EDV) averaged 1.49±0.24 mL. For ejecting beats, stroke volume was 0.81±0.18 mL and ejection fraction was 54.6±7.2% for both heart rates. Ejection rate was 2.3±0.4 EDV/s and 3.1±0.4 EDV/s for 100 and 150 bpm, respectively. LV filling was completed at an average of 89±2% of the cardiac cycle, beginning at the stimulation pulse of the electronic pacemaker.
Fig 6⇓ shows representative tracings of LV volume, LV pressure, and LV dP/dt before and during administration of EMD 57033 at a constant heart rate of 100 bpm during steady state ejecting contractions. LV end-diastolic volume and stroke volume were fixed before and during EMD 57033 administration. EMD 57033 increased LV peak pressure from 61 to 88 mm Hg at the same end-systolic volume of 0.55 mL and increased LV dP/dtmax from 760 to 1250 mm Hg/s. LV end-diastolic pressure increased from 9 to 15 mm Hg with EMD 57033. In this example, dP/dtmin was slightly decreased, that is, peak relaxation rate slightly increased and occurred slightly earlier during EMD 57033. Administration of EMD 57033 resulted in an increase in coronary flow from 4.6 to 7.6 mL/min and a slight decrease in Avo2Δ from 4.8 to 4.1 vol%. Accordingly, V̇o2 increased from 0.036 to 0.052 mL O2/beat per 100 g with EMD 57033.
Comparison of the Effect of EMD 57033 on Cardiac Mechanics During Isovolumic and Ejecting Contractions
Cardiac mechanics results during isovolumic contractions in this series were similar to those obtained in the Ca2+ series except that there was a significant increase in LV end-diastolic pressure at both heart rates (Fig 7B⇓). Figs 7E through 7H show the changes in LV peak pressure, LV end-diastolic pressure, LV developed pressure, and LV dP/dtmax with EMD 57033 during ejecting contractions. LV peak pressure was significantly increased with EMD 57033 at both heart rates (Fig 7E⇓). LV end-diastolic pressure also showed a significant increase at both heart rates (Fig 7F⇓). Both LV developed pressure and LV dP/dtmax showed a significant increase at 100 and 150 bpm (Figs 7G⇓ and 7H⇓). No significant effect of EMD 57033 on LV dP/dtmin was detected by ANOVA (data not shown, P=.18). After discontinuation of EMD 57033, all the mechanical variables returned toward their initial control values, and in each case there was no statistically significant difference between the two controls.
Figs 8A through 8D compare the effect of EMD 57033 on LV mechanics for ejecting versus isovolumic contraction mode at the two heart rates. The increase in LV peak pressure was significantly greater with ejecting than isovolumic contractions at both heart rates (Fig 8A⇓). Similar findings were observed for LV developed pressure and LV dP/dtmax (Figs 8C⇓ and 8D⇓). There was a trend toward a smaller increase in LV end-diastolic pressure for ejecting contractions than for isovolumic contractions at each heart rate. This effect of contraction mode did not reach statistical significance (P=.09) (Fig 8B⇓). Thus, EMD 57033–related changes in conventional indexes of contractility were more prominent with ejecting contractions, while effects on the extent of relaxation were somewhat smaller but not significantly different between isovolumic and ejecting beats.
Effect of EMD 57033 on Cardiac Energetics During Ejecting Contractions
Coronary perfusion pressure was maintained virtually constant throughout each experiment (Table 3⇓). Coronary blood flow was increased with EMD 57033 under all conditions, as was the case in the Ca2+ series. No significant effect of EMD 57033 on Avo2Δ was detected with ANOVA (P=.65). As observed in the Ca2+ series, myocardial V̇o2 was significantly increased with EMD 57033 under all conditions, and this increase was larger at 100 bpm than at 150 bpm for each contraction mode. The magnitude of increase in V̇o2 was significantly greater for ejecting compared with isovolumic contractions at a given heart rate (Table 4⇓). Under the repeat control state myocardial V̇o2 became similar to its initial control level. However, coronary blood flow remained at significantly higher levels than under initial control conditions even after EMD 57033 was discontinued, indicating a persisting influence of the drug on coronary vascular resistance.
The present study constitutes the first report of the mechanoenergetic effects in the beating heart of a drug whose predominant mechanism of action is Ca2+ sensitization. EMD 57033 enhanced contractility and increased LV end-diastolic pressure, the latter being more prominent at faster heart rate and with lower [Ca2+]o. The increase in LV contractility with EMD 57033 was more prominent under ejecting conditions, while the effects of the drug on LV end-diastolic pressure tended to be less. EMD 57033 produced an abbreviation of time to peak pressure and a prolongation of relaxation duration along with a delay of the timing of dP/dtmin and changes in the morphology of the pressure fall tracing. However, peak relaxation rate was either increased or unaffected. These mechanical effects are very different from those observed in the beating heart with other Ca2+ sensitizers.6 7 The latter drugs do not impede ventricular relaxation, most likely because they have important effects besides Ca2+ sensitization, especially phosphodiesterase III inhibition. Although myocardial O2 consumption per beat increased with EMD 57033, the magnitude of the increase was inversely related to heart rate but not different between low and normal [Ca2+]o at a given heart rate. Moreover, the increase in O2 consumption was significantly greater for ejecting than isovolumic contractions.
Effects of EMD 57033 on Mechanical Performance
EMD 57033 has been reported to exhibit a positive inotropic effect in single cardiac myocytes and isolated papillary muscles by increasing the Ca2+ sensitivity of the myofilaments.1 2 3 4 Prior studies of single myocytes2 3 showed that EMD 57033 increased the extent of shortening without increasing the peak amplitude of the intracellular Ca2+ transient. Ca2+ sensitization appears to be the predominant mechanism of action of EMD 57033 at concentrations up to about 5 μmol/L.2 3 4 At higher concentrations, the drug may exert phosphodiesterase III inhibition, which is reported as a major mechanism of action of EMD 57439.2 3 4 A recent study by Solaro et al2 showed that there was no effect of EMD 57033 on Ca2+ binding to myofilament troponin C and that the drug increased the actomyosin ATPase activity of myofilament preparations in which either troponin or tropomyosin had been extracted. Other studies12 13 also confirm a direct effect on crossbridge turnover rate. Based on these observations, Solaro et al2 speculated that Ca2+ sensitization by EMD 57033 may involve a myofilament domain other than troponin C. They proposed that EMD 57033 binds to a “receptor” at the actin-myosin interface, either on actin or the crossbridge itself. They suggested that binding of EMD 57033 at the actin-myosin interface reverses the usual inhibition of formation of strong crossbridges by troponin-tropomyosin and in turn promotes spread along the thin filament to immediately adjacent functional units that are not activated by attachment of Ca2+ to troponin C (nearest neighbor interactions in the terminology of Solaro et al2 ). In addition, binding of EMD 57033 to a site at the actin-myosin interface would increase actomyosin ATPase activity. Accordingly, the mechanism of action of EMD 57033 could be a combination of (1) a disinhibition of formation of strong crossbridges that facilitate nearest neighbor interactions and (2) a direct increase in actomyosin ATPase activity. More recently, Kraft et al14 and Pan and Barsotti15 have reported results of skinned fiber studies that suggest that the drug increases force per crossbridge rather than the number of strongly bound crossbridges. However, their studies were performed at much higher concentrations of drug than our own or those of Solaro et al.2
In the present study, an increase in contractility occurred (increased LV developed pressure and LV dP/dtmax) with a decrease in contraction duration. The latter effect is associated with an increase in actomyosin ATPase activity in the beating heart.16 17 The increase in LV end-diastolic pressure and prolonged relaxation that we observed are in agreement with the findings of prior studies that have shown a reduction in diastolic cell length of cardiac myocytes2 3 or an increase in the time to 50% relaxation in isolated heart muscle.4 The prolonged time course of relaxation is consistent with prolongation of the time required for strongly bound crossbridges to detach and for tension to return to its initial level. Despite prolongation of relaxation, peak relaxation rate actually increased under conditions of 2.5 mmol/L [Ca2+]o. This can be ascribed to the fact that developed pressure increased, that is, the total pressure fall during relaxation was greater. Under conditions of 1.0 mmol/L [Ca2+]o, with even more pronounced prolongation of relaxation, the increase in peak prelaxation rate was no longer apparent. It is also possible that EMD 57033 increases diastolic crossbridge formation, resulting in a true increase in resting tension. Arguing against this is the fact that end-diastolic pressure increased less at the lower heart rate at both [Ca2+]o conditions.
Influence of Heart Rate and [Ca2+]o on the Effect of EMD 57033
The present study demonstrates that enhancement of ventricular contractility by EMD 57033 is least prominent at higher heart rate with low [Ca2+]o. Attenuated enhancement of contractility at higher heart rates is also consistent with the idea that more time is required for strongly bound crossbridges to detach after EMD 57033 administration, whether the drug enhances nearest neighbor interactions or force per crossbridge. Thus, as heart rate increases during EMD 57033 administration, fewer strongly bound crossbridges have time to detach and therefore are not available to reform after the next electrical depolarization. This also would result in reciprocal contractility-relaxation effects and in reciprocal relations between contraction and relaxation duration in relation to heart rate. Accordingly, to achieve a maximum increase in “contractility” would require slowing heart rate sufficient to allow all strongly bound crossbridges to detach.
The present study also shows that lower [Ca2+]o resulted in much larger increases in LV end-diastolic pressure and a smaller increase in contractility with EMD 57033 (the latter was significant only at 150 bpm). These effects were predicted by Solaro et al2 based on their model of drug effect because at low [Ca2+]o, with less underlying activation, the contractile proteins are considered to possess a greater capacity to augment nearest neighbor effects with administration of EMD 57033. Based on this mechanism, EMD 57033 would be expected to result in both greater enhancement of contractility and greater retardation of relaxation at low [Ca2+]o. In fact, EMD 57033 administration at low [Ca2+]o resulted in a consistently exaggerated effect on relaxation but similar or smaller effects on contractility, depending on heart rate. As shown in Figs 5C⇑ and 5E⇑, when the heart rate was decreased to 100 bpm at low [Ca2+]o, effects on contractility became very similar to those present at normal [Ca2+]o. Thus, the net effects of EMD 57033 on contractility are a function of the interplay between heart rate and [Ca2+]o. With lower [Ca2+]o and perhaps a greater capacity for EMD 57033 to increase disinhibition of formation of strong crossbridges, heart rate determines to what extent this mechanism is manifested as contraction versus relaxation effects.
Influence of Ejection on the Response to EMD 57033
There is no corollary information available in isolated muscle studies in which the effects of EMD 57033 were compared in isometric twitches versus shortening contractions. In the present study, the increase in ventricular contractility was greater for ejecting than isovolumic beats. A possible explanation for this result is that ejection physically disrupts nearest neighbor interactions. This would result in more crossbridges available to reform on the next beat and greater developed force. Based on this idea, LV end-diastolic pressure during EMD 57033 administration might be anticipated to be lower for ejecting versus isovolumic contractions. We found that while there was no statistically significant difference between contraction modes, there were trends toward a smaller increase in LV end-diastolic pressure during ejecting beats at both heart rates (Fig 8B⇑). It is possible that the smaller absolute values of LV end-diastolic pressure (versus developed pressure or dP/dtmax) made it difficult to detect a difference related to ejection. Another admittedly speculative explanation for this result is that ejection somehow increases the effects of EMD 57033 on actomyosin ATPase activity. An increase in actomyosin ATPase activity would lead to increased ATP splitting and increased myocardial V̇o2. Our results do indicate a significantly greater increase in V̇o2 with EMD 57033 for ejecting compared with isovolumic beats (Table 4⇑), supporting a possible modulating influence of ejection on the effect of EMD 57033 on actomyosin ATPase activity.
Response of Myocardial Energy Metabolism to EMD 57033
We found that EMD 57033 administration resulted in a similar increase in myocardial V̇o2 at a given heart rate at the different levels of [Ca2+]o despite the fact that our results and those of Solaro et al2 are consistent with the idea that disinhibition of strong crossbridge formation and promotion of nearest neighbor effects are more prominent at low [Ca2+]o. This result could be a manifestation of the effect of EMD 57033 on actomyosin ATPase activity, which is not necessarily dependent on [Ca2+]o, and could be a more important determinant of changes in energy utilization in response to the drug than disinhibition of strong crossbridge formation. Alternatively, this finding could be a manifestation of phosphodiesterase III inhibition. At a perfusate concentration of 5 to 6 μmol/L, it is likely that this effect was modest at best.2 3 4 Furthermore, the mechanical effects that we have documented, especially those related to relaxation, are not reminiscent of phosphodiesterase III inhibition1 4 or those reported for other Ca2+ sensitizers, which are known to be more active than EMD 57033 as phosphodiesterase III inhibitors.6 7 Thus, while we cannot exclude a component of phosphodiesterase III inhibition, we believe this is an unlikely explanation for the energetic results.
Another interesting energetics finding is that EMD 57033 resulted in smaller increases in V̇o2 per beat at higher heart rate (Tables 2⇑ and 4⇑). This is also consistent with the idea put forth previously that under the influence of EMD 57033, fewer crossbridges are available to reform and split ATP at higher heart rate.
Lee et al19 have shown in ferret papillary muscles that a decrease in developed tension induced by acidosis was recovered with EMD 57033. They also reported a similar effect of EMD 53998 on reduced tension in anoxia.20 Although these studies were performed under different experimental conditions than ours, they suggest that Ca2+ sensitizers may have particularly beneficial effects on contractility under hypoxic or acidotic conditions. In view of the effects observed in the present study, it will be important to assess EMD 57033 during hypoxia and/or acidosis in more physiological preparations.
Our results demonstrate that EMD 57033 increases LV end-diastolic pressure and prolongs relaxation, effects that could detract from its clinical use. However, we also found a greater increase in contractility with more physiologically relevant ejecting contractions. Moreover, the magnitude of LV end-diastolic pressure elevation was somewhat less for ejecting contractions, although this difference was not statistically significant. Thus, any unfavorable effects of EMD 57033 on ventricular relaxation might be offset by a greater increase in contractility in a physiologically ejecting heart at a relatively slow rate.
Finally, one of the potential advantages of Ca2+-sensitizing drugs is an increase in contractility achieved with smaller increases in V̇o2 than other positive inotropic drugs, in particular drugs that increase cAMP. Our study was not designed to directly compare the energetics of EMD 57033 with other agents. However, we observed substantial increases in V̇o2 under all experimental conditions. Thus, the notion that significant effects on contractility can be obtained at little or no energetic cost with Ca2+ sensitizers is not borne out by our results. However, in contrast to our preparation, in the intact cardiovascular system this advantage of Ca2+-sensitizing drugs may be magnified in patients with heart failure, in whom reductions in heart size and rate occurring in response to the drug would further decrease energy requirements.
This study was supported by National Institutes of Health Grants HL-45116 and HL-52087 and National American Heart Association Grant-in-Aid 93006340. We thank Stephen P. Bell and Judit Fabian for their excellent technical assistance. We are also grateful to Dr N. Beier and E. Merck for generously supplying EMD 57033 and providing assays of the drug.
- Received April 10, 1995.
- Revision received May 30, 1995.
- Accepted July 5, 1995.
- Copyright © 1995 by American Heart Association
Solaro RJ, Gambassi G, Warshaw DM, Keller MR, Spurgeon HA, Beier N, Lakatta EG. Stereoselective actions of thiadiazinones on canine cardiac myocytes and myofilaments. Circ Res. 1993;73:981-990.
Gambassi G, Capogrossi MC, Klockow M, Lakatta EG. Enantiometric dissection of the effects of the inotropic agent, EMD 53998, in single cardiac myocytes. Am J Physiol. 1993;264:H728-H738.
White J, Lee JA, Shah N, Orchard CH. Differential effects of the optical isomers of EMD 53998 on contraction and cytoplasmic Ca2+ in isolated ferret cardiac muscle. Circ Rec. 1993;73:61-70.
Hata K, Goto Y, Futaki S, Ohgoshi Y, Yaku H, Kawaguchi O, Takasago T, Saeki A, Taylor TW, Nishioka T, Suga H. Mechanoenergetic effects of pimobendan in canine left ventricles: comparison with dobutamine. Circulation. 1992;86:1291-1301.
de Tombe PP, Burkhoff D, Hunter WC. Effects of calcium and EMD-53998 on oxygen consumption in isolated canine hearts. Circulation. 1992;86:1945-1954.
Yaku H, Slinker BK, Myhre ESP, Watkins MW, LeWinter MM. Stability of myocardial O2 consumption-pressure-volume-area relation in red cell-perfused rabbit heart. Am J Physiol. 1991;261:H1630-H1635.
Goto Y, Slinker BK, LeWinter MM. Similar normalized Emax and O2 consumption-pressure-volume area relation in rabbit and dog. Am J Physiol. 1988;255:H366-H374.
Goto Y, Slinker BK, LeWinter MM. Accuracy of volume measurement of rabbit left ventricle by balloon method. Am J Physiol. 1988;255:H394-H396.
Glantz SA, Slinker BK. Primer of Applied Regression and Analysis of Variance. New York, NY: McGraw-Hill; 1990.
Warshaw D, Gambassi G, Beier N, Lakatta EG. A novel thiadiazinone increases the acto-myosin interaction. J Mol Cell Cardiol. 1992;24(suppl III):S42.
Kraft T, Lues I, Brenner B. Increase in force per cross-bridge by EMD 57033, a calcium-sensitizing agent. Biophys J. 1994;66:A188. Abstract.
Goto Y, Slinker BK, LeWinter MM. Effects of amrinone and isoproterenol on mechanoenergetics of blood-perfused rabbit heart. Am J Physiol. 1992;262:H719-H727.
Goto Y, Slinker BK, LeWinter MM. Decreased contractile efficiency and increased nonmechanical energy cost in hyperthyroid rabbit heart: relation between O2 consumption and systolic pressure-volume area or force-time integral. Circ Res. 1990;66:999-1011.
Moss RL. Ca2+ regulation of mechanical properties of striated muscle: mechanistic studies using extraction and replacement of regulatory proteins. Circ Res. 1992;70:865-884.
Lee JA, Allen DG. EMD 53998 sensitizes the contractile proteins to calcium in intact ferret ventricular muscle. Circ Res. 1991;69:927-936.