Differential Effects of the Ca2+ Sensitizers Caffeine and CGP 48506 on the Relaxation Rate of Rat Skinned Cardiac Trabeculae
Abstract During heart failure, force production by the heart decreases. This may be overcome by Ca2+-sensitizing drugs, which increase myofibril Ca2+ sensitivity without necessarily altering intracellular Ca2+ concentration. However, Ca2+ sensitizers slow the relaxation of intact cardiac muscle. We used diazo-2, a caged chelator of Ca2+, to study the effects of the Ca2+ sensitizers caffeine and CGP 48506 on the intrinsic relaxation rate of cardiac myofibrils. Trabeculae from rat right ventricles were skinned by 1% Triton X-100 and were activated in a 10-μL bath. In steady state experiments, CGP 48506 (10 μmol/L) shifted the force-pCa curve leftward by 0.41±0.03 pCa units (mean±SEM, n=6). An identical shift was induced by caffeine (20 mmol/L). Photolysis of diazo-2 by a flash of light (160 mJ, 310 to 400 nm) caused an immediate decrease in Ca2+-activated force produced by the trabeculae. Relaxation was fitted by a double-exponential decay, and the rate constants were found to be independent of force and preflash Ca2+ concentration. The initial fast rate, corresponding to myofibrillar relaxation, was increased from 17.3±2.0 to 30.9±3.7 s−1 (n=4) by caffeine but was unaffected by CGP 48506 (16.6±1.7 and 14.4±2.3 s−1 in the absence and presence of drug, respectively; n=5). Thus, myofibril relaxation need not be slowed by Ca2+-sensitizing agents but can even be accelerated. Despite similarities in their effects on myofibril Ca2+ sensitivity, caffeine and CGP 48506 affect the myofibrils at least partly via different mechanisms.
During heart failure, the decrease in myocardial force production may be overcome through two approaches: either the amount of Ca2+ available to the myofibrils can be increased, or the responsiveness of the myofibrils to the available Ca2+ can be augmented by “Ca2+-sensitizing” drugs. The former approach has the disadvantage that the elevation of [Ca2+]i can cause arrhythmias.1 Thus, attention has recently switched to a search for suitable Ca2+-sensitizing compounds. Unfortunately, almost all of the compounds identified to date have also been inhibitors of phospho-diesterase (PDE) III activity, eg, caffeine2 and EMD 57033,3 and thus tend to increase [Ca2+]i in addition to increasing Ca2+ sensitivity. However, recently, a novel Ca2+ sensitizer, CGP 48506 (5-methyl-6-phenyl-1,3,5,6-tetrahydro-3,6-methano-1,5-benzodiazocine-2,4-dione), with no PDE III inhibitory activity, was synthesized by Ciba-Geigy Ltd.4 CGP 48506 increases the amplitude of the twitch in intact muscle with a concomitant prolongation of relaxation.4 5 6 Studies using permeabilized (“skinned”) fibers showed that CGP 48506 acts directly on the myofibrils, shifting the relationship between force and pCa (−log[Ca2+]) toward lower concentrations of Ca2+.4 5 6
A disadvantage of the Ca2+-sensitizing agents is their tendency to slow the rate of relaxation of the cardiac twitch,4 5 6 7 8 which could compromise diastolic filling of the heart. The decreased relaxation rate could be a consequence of the increase in Ca2+ sensitivity, such that force is produced even at levels of Ca2+ where the muscle is normally relaxed. Alternatively, the twitch relaxation rate may be limited by the intrinsic relaxation rate of the myofibrils, which could be slowed in the presence of a Ca2+ sensitizer.
The aim of the present study was to investigate the effects of CGP 48506 and caffeine on the intrinsic relaxation rate of the myofibrils in skinned cardiac trabeculae. It is possible to achieve a near-instantaneous drop in [Ca2+] using diazo-2, which chelates Ca2+ more strongly after being photolyzed by near-UV light.9 The rate of relaxation of a skinned muscle after photolysis of diazo-2 is then limited by the relaxation properties of the myofibrils rather than by the rate of fall of [Ca2+]. We used diazo-2 to rapidly reduce [Ca2+] around the myofibrils in skinned cardiac muscles and thereby determined whether the Ca2+ sensitizers CGP 48506 and caffeine alter the intrinsic rate of myofibrillar relaxation. Some of these results were published previously in abstract form.10 11
Materials and Methods
Preparations and Equipment
The procedures followed were in accordance with UK Home Office guidelines (schedule 1) and were based on those described previously.12 13 In brief, male Wistar rats (≈250 g) were rendered unconscious with a blow to the head and then were killed by cervical dislocation. Their hearts were removed, and thin trabeculae (diameter, ≈100 to 200 μm; length, 1 to 3 mm) were dissected out from the right ventricle in Tyrode’s solution containing 25 mmol/L 2,3-butanedione monoxime to minimize cell contracture.14 Each trabecula was then skinned by a 30-minute immersion in relaxing solution (see below) containing 1% Triton X-100 and 0.1 mmol/L phenylmethylsulfonyl fluoride. The ends of the skinned muscle were mounted in nylon snares attached to a servomotor (300S, Cambridge Technology Inc) and to an isometric force transducer (AE801, SensoNor; resonant frequency, 3 kHz). The muscle was then lowered into one of a movable series of wells (each with a volume of 0.5 mL) and was bathed in relaxing solution for >15 minutes to wash out the detergent. The temperature of the muscle baths was maintained at 22.0±0.1°C by circulating water from a thermoregulator. Solutions in the baths were stirred with a jet of N2. Muscle length was adjusted until resting force was just detectable; at this length, the sarcomere length in the resting muscle, measured by diffraction of a HeNe laser beam, was 2.1 μm.
For the photolysis experiments, one of the muscle baths was modified to reduce its volume to 10 μL. This bath (like the others) had sides made from pieces of grade-0 glass coverslips. These coverslips absorb light of <310 nm and so help prevent damage to the muscle resulting from very short wavelength light. A xenon flashlamp (Hi-Tech Ltd), containing a UG5 bandpass filter (300 to 400 nm), was focused horizontally onto the bath by using cylindrical and spherical quartz lenses. The size and position of the focused light from the flashlamp were optimized before each experiment by using the burn pattern on light-sensitive paper. By this means, we ensured that the muscle was positioned completely within the region of maximum light intensity during the flash. PClamp software (Axon Instruments) was used to trigger the flashlamp and to record and analyze the resulting force signals, which were recorded on an AT-compatible computer using a 12-bit A/D board (Digidata 1200, Axon) at a sampling rate of 2 kHz. Slow time-base recordings were also made on a four-channel chart recorder.
For the skinned muscle experiments using diazo-2, all solutions contained (mmol/L) BES 100, potassium propionate 50, Na2H2ATP 6.3, MgCl2 6.4 (Mg2+ 1), Na2 phosphocreatine 10, potassium phosphate 1, glutathione 5, and leupeptin 0.001 (pH 7.1; ionic strength, 200 mmol/L). Leupeptin, a protease inhibitor, reduced the deterioration in force during the experiments. The above solution (ie, with no added EGTA or Ca2+) was used as “preactivating solution” (see below). “Relaxing solution” contained, in addition, 1 mmol/L EGTA, giving a pCa (−log10[Ca2+]) of 8.5. “Maximum activation solution” had 1 mmol/L EGTA and 1.1 mmol/L CaCl2 (pCa 4.5). The diazo-2 solutions contained 0.25 mmol/L diazo-2 (added from a stock solution of 20 mmol/L) and 0 to 0.25 mmol/L CaCl2 (final pCa, ≈7.5 to 4.9). The exact concentration of diazo-2 in the stock solution was measured from its absorption at 370 nm, assuming an extinction coefficient of 22 200 (mol/L)−1·cm−1.9 For measurement of force-pCa relationships, the solutions were similar but contained 10 mmol/L EGTA and 0 to 10.5 mmol/L Ca2+, with the concentration of potassium propionate reduced to maintain the ionic strength. Details of the manufacture of solution and calculation of free ion concentrations have been described previously.13 The dissociation constants for Ca2+ and Mg2+ binding to diazo-2 were taken as 2.2 μmol/L and 5.5 mmol/L, respectively, before photolysis and 0.073 μmol/L and 3.4 mmol/L, respectively, after photolysis.9 Diazo-2 free acid (K+ form) was obtained from Molecular Probes; CGP 48506 was donated by Ciba-Geigy Ltd and was prepared as a stock solution of 2 mmol/L in dimethylsulfoxide (DMSO) because of its limited solubility in aqueous solution. The final concentration of CGP 48506 in the experimental solutions was 10 μmol/L (or occasionally 100 μmol/L), accompanied by 0.5% dimethyl sulfoxide. An equivalent amount of dimethyl sulfoxide was added to the control solutions and had no effect on either force or relaxation rate (results not shown). Caffeine was obtained from Sigma Chemical Co. Other chemicals were obtained from Sigma or Merck Ltd.
At the start of the experiment, the skinned trabecula was activated several times with maximal activating solution to check the reproducibility of the force response and of the resting sarcomere length pattern. These initial activations consisted of the following sequence of solution changes: relaxing solution, preactivating solution (used to wash the EGTA out of the muscle and so minimize its Ca2+-buffering capacity), maximum activating solution, and relaxing solution. For the part of the experiment involving diazo-2 photolysis, the procedure was the same, but the muscle was placed not in activation solution but in a diazo-2 solution in the 10-μL photolysis bath. Fig 1⇓ shows that, as with activating solution, the Ca2+ in the diazo-2 solution activated the myofibrils and the muscle developed force. When force was steady, the flashlamp was triggered to deliver a pulse of near-UV light (≈160 mJ in ≈1 millisecond) to the muscle. The photolyzed diazo-2 bound some of the free Ca2+ in the solution and produced a sudden drop in [Ca2+], causing the muscle to relax rapidly (Fig 1⇓). The muscle was then returned to relaxing solution. This was repeated for a range of Ca2+ concentrations in the diazo-2 solution. Control runs (no drug) were carried out before and after a test run in which either 10 μmol/L CGP 48506 or 20 mmol/L caffeine was present in all solutions. By varying the initial [Ca2+], it was possible to check for any dependence of relaxation rate on the preflash [Ca2+] or force. Frequent application of control maximal activating solution to the muscle allowed correction for the decline (assumed linear) in maximum force during an experiment; this amounted to a decrease of only 21.2±4.9% (mean±SEM, n=6) over 4 to 5 hours. The force-pCa relationship was determined in the absence and presence of caffeine or CGP 48506, using six muscles for each (four of which were common to both drugs). Control and test activating solutions were applied alternately, in order of increasing [Ca2+], with relaxation of the fiber between each activation. Flashing the trabecula in the presence of caffeine or CGP 48506 in these activation solutions (containing EGTA rather than diazo-2) did not produce any change in force.
Forces are expressed relative to maximal Ca2+-activated force under control conditions unless stated otherwise. The relationship between steady state force and pCa was fitted by the Hill equation as follows: relative force=maximum force×[Ca2+]nH/(KnH+[Ca2+]nH), where nH is the Hill coefficient and K is a dissociation constant. The relaxation of force after diazo-2 photolysis was fitted to a double-exponential decay using Clampfit software (Axon Instruments). Values are given as mean±SEM of n experiments. Paired t tests were applied where appropriate, and a significant difference was taken as P<.05.
The skinned trabeculae (mean diameter, 167±32 μm; n=6) produced a force of 30.5±7.2 mN·mm−2 during control maximal activation. As shown in Fig 2A⇓, the addition of 10 μmol/L CGP 48506 shifted the force-pCa relationship toward lower concentrations of Ca2+. The pCa for 50% activation (pCa50) increased from 5.51±0.05 (n=6) under control conditions to 5.92±0.04 in the presence of CGP 48506, a rise of 0.41±0.03 pCa units. These results are similar to those obtained using porcine skinned cardiac trabeculae.4 5 Thus, CGP 48506 is a potent Ca2+ sensitizer of rat cardiac myofibrils. An identical increase in myofibril Ca2+ sensitivity was produced by 20 mmol/L caffeine, which raised pCa50 from 5.51±0.05 (n=6) under control conditions to 5.92±0.04, a shift of 0.41±0.03 pCa units (Fig 2B⇓). This result was similar to those obtained previously.12 15 In addition, whereas 20 mmol/L caffeine decreased maximum force by 10.5±2.1% (n=6), 10 μmol/L CGP 48506 did not significantly affect maximum force.
We used the caged chelator of Ca2+, diazo-2, to investigate the effects of CGP 48506 and caffeine on the relaxation rate of skinned trabeculae. Our protocol differs from that in previous studies16 17 in that we used 0.25 mmol/L diazo-2 rather than 2 mmol/L and flashed the muscles in the bath at 22°C rather than in air at lower temperatures. We found that a diazo-2 concentration of ≥1 mmol/L produced only a small relaxation of Ca2+-activated force using photolysis in the bath, probably because the high absorbance of diazo-2 resulted in less light from the flashlamp reaching the muscle itself. However, the relaxation rates measured in the presence of 0.25 or 0.5 mmol/L diazo-2 were not significantly different (n=4), indicating that 0.25 mmol/L was adequate for the present experiments. We also found no difference between relaxation rates obtained on flashing trabeculae in air with 0.25 or 2 mmol/L diazo-2 (results not shown).
In the muscle shown in Fig 1⇑, the diazo-2 after photolysis bound sufficient Ca2+ to cause complete relaxation of the trabecula. However, we found that the extent of the relaxation depended on the initial [Ca2+]. The greatest drop in force following a flash was achieved when the trabecula initially produced 70% to 80% of maximum force. Because of the sigmoidal shape of the force-pCa curve, decreases in [Ca2+] produced proportionally smaller falls in force as the level of activation approached maximum. The magnitude of the decrease in force after photolysis varied between 0% and 75% of control maximum force (depending on the initial [Ca2+]) and was not altered by either caffeine or CGP 48506 (results not shown).
Fig 3A⇓ illustrates the effects of caffeine and CGP 48506 on relaxation rate using a faster time scale than that in Fig 1⇑. The relaxation of force after diazo-2 photolysis was fitted well by a double-exponential decay. Since [Ca2+] falls in ≈1 millisecond, the resulting fall in force reflects the intrinsic relaxation rate of the myofibrils. It was assumed that the first, faster decay phase represented the maximum rate of myofibrillar relaxation (see “Discussion”). Its rate constant was therefore used to analyze the effects of CGP 48506 and caffeine on relaxation rate. The origin of the slower phase of relaxation is unclear (eg, there was no evidence that it was due to sarcomere rearrangement). In the example shown in Fig 3A⇓, the initial, fast rate constant was increased from 16.1 to 31.1 s−1 by caffeine (20 mmol/L) and was unchanged at 16.0 s−1 by CGP 48506 (10 μmol/L). Thus, caffeine clearly speeded up the rate of relaxation, whereas CGP 48506 had no effect. A summary of a complete experiment (Fig 3B⇓) illustrates that these relaxation rates were consistent, irrespective of [Ca2+], and also demonstrates that relaxation rates in control solutions were similar before and after the addition of the Ca2+ sensitizers. The data from all the experiments are combined in the Table⇓. Caffeine almost doubled the fast rate constant of relaxation from 17.3±2.0 to 30.9±3.7 s−1. On the other hand, CGP 48506 did not significantly alter the rate (14.4±2.3 s−1 compared with a control value of 16.6±1.7 s−1). Thus, despite their qualitatively similar effects on the pCa50 of the force-pCa relationship of skinned cardiac muscles, caffeine and CGP 48506 had different effects on the rate of relaxation of the myofibrils. In addition, caffeine (but not CGP 48506) decreased the rate of the slow phase of relaxation. The amplitudes of the fast or slow exponential phases of relaxation (see Table⇓) were not altered by these Ca2+ sensitizers.
Some experiments were also carried out with 100 μmol/L CGP 48506, which shifted the force-pCa curve leftward by 1.16±0.03 units (n=5, data not shown). CGP 48506 (100 μmol/L) produced a small but significant decrease of the fast rate constant of relaxation from 15.6±0.9 to 12.5±1.4 s−1 (n=5) with no effect on the slow rate constant or the amplitudes of the phases.10 11 However, interpretation of these results was complicated by the possibility that such a high concentration of CGP 48506 may have multiple effects on myofibrillar proteins (see “Discussion”).
Since caffeine and CGP 48506 increased the Ca2+ sensitivity of the myofibrils, Ca2+ activations in the presence of these drugs either produced higher forces than the control (at the same [Ca2+]) or were done at lower Ca2+ concentrations (when we tried to match the forces). Therefore, we checked to see if the drug-induced changes in relaxation rate were due to the difference in [Ca2+] or force from the control. As shown in Fig 4⇓, there was no obvious dependence of relaxation rate on either preflash [Ca2+] or force, either under control conditions or in the presence of CGP 48506 (panels A and B) or caffeine (panels C and D).
In these experiments, we used diazo-2 photolysis to produce a near-instantaneous decrease in [Ca2+], so that the rate of the rapid decline in force (Figs 1⇑ and 3⇑) was limited by the intrinsic rate of relaxation of cardiac myofibrils rather than by the rate of fall of [Ca2+]. Under control conditions, the initial, fast rate of relaxation was ≈17 s−1 at 22°C. This value compares well with the rate of 15.8 s−1 given in a preliminary report18 using single rat myocytes under very similar conditions. Other studies have used diazo-2 to investigate relaxation of skinned cardiac muscle,16 17 but in these, the muscles were flashed in air. Once in air, the muscle cools down to ≈12°C (the dew point), and Pi, from ATP hydrolysis, builds up within the myoplasm; both changes would by themselves cause some relaxation of force. To overcome these problems, we photolyzed diazo-2 with the muscle in solution at 22°C. A potential drawback was the ability of diazo-2 in the solution around the muscle to absorb light of the wavelengths needed for photolysis of diazo-2 within the muscle. By using 0.25 mmol/L diazo-2, however, its absorbance was decreased to an acceptable level. Relaxation rates were not limited by the diazo-2 concentration, since (1) similar rates were obtained using 0.25 or 0.5 mmol/L diazo-2 (not shown), and (2) caffeine increased relaxation rate 2-fold by an action on the myofibrils.
In theory, the rate at which force declines after photolysis of diazo-2 could be limited by the rate of Ca2+ release from troponin C (TnC), by the rate of conformational changes before actomyosin dissociation, or by the crossbridge detachment rate per se. There are several lines of evidence indicating that Ca2+ dissociation from TnC is not the rate-limiting step in relaxation. First, from the measured affinity of TnC for Ca2+ [KCa, ≈4×105 (mol/L)−1]12 19 and the apparent rate constant for Ca2+ binding to TnC [kon, 1.4×108 (mol/L)−1·s−1],19 the off rate constant (koff=kon/KCa) should be ≈350 s−1 (as observed19), which is much faster than the relaxation rate we measured. Second, if the relaxation rate was limited by koff, caffeine, which does not alter the Ca2+ affinity of TnC,12 20 should not affect myofibril relaxation rate. This is clearly not the case (present results). Third, we previously found the rates of relaxation were similar after myofibril deactivation caused by photolysis of diazo-2 or of the photosensitive Ca2+ sensitizer EMD 57033.21 Since EMD 57033 acts directly on the crossbridges without altering Ca2+ binding to TnC,22 this similarity in relaxation rates suggests that the rate is limited by the crossbridges rather than by koff. Finally, we found that in rat skinned trabeculae (containing myosin isoform V1), the rate of relaxation after diazo-2 photolysis was six times faster than that in guinea pig (myosin V3) trabeculae (authors’ unpublished data, 1996). Since cardiac TnC is highly conserved between species, these different rates of relaxation were most likely determined by the myosin moiety. This evidence suggests that the rate of relaxation is likely to be limited by the rate of crossbridge detachment. It is possible, however, that the absolute rates of relaxation we measured were influenced by sarcomere length changes during relaxation, which would occur because of the compliance of the muscle ends. Nevertheless, this would not have affected our conclusions concerning the effects of the drugs on relaxation rate, since the control and drug were tested on the same muscle.
Drugs that increase the Ca2+ sensitivity of the cardiac myofibrils also tend to prolong the twitch in intact muscle, eg, EMD 57033,7 caffeine,8 and CGP 48506.4 5 6 In the present study, our aim was to test whether Ca2+-sensitizing agents slow the rate of relaxation of the heart through a direct reduction in the intrinsic rate of myofibrillar relaxation. In a previous study, Simnett et al17 (1993) reported that the Ca2+ sensitizer EMD 57033 (10 μmol/L) increased the rate constant for the fast phase of relaxation after diazo-2 photolysis by 37% (from 15.4 to 21.1 s−1) in guinea pig trabeculae in air. However, we recently discovered21 that EMD 57033 is partially destroyed by light of wavelengths similar to those used for diazo-2 photolysis. Since EMD 57033 potentiates force at any given [Ca2+], the flash-induced reduction in EMD 57033 concentration itself causes a decline in tension, the rate of which is similar to that achieved after photolysis of diazo-2 alone.21 Thus, it is difficult to determine the contribution of any effect of EMD 57033 directly on the myofibril relaxation rate when diazo-2 and EMD 57033 are photolyzed concurrently.
We investigated the effects on relaxation rate of two Ca2+ sensitizers, caffeine and CGP 48506, that were not affected by the flash. It is difficult to explain the observed effects of caffeine and CGP 48506 on the myofibrils, but clearly, several mechanisms are involved. The suppression of maximum force by caffeine has been discussed in detail previously.12 The compounds had similar effects on the pCa50 of the force-pCa relationship and might have been expected to have the same mechanism of action. It is known that neither drug affects Ca2+ binding to TnC.5 6 12 20 Surprisingly, the two compounds had different effects on the rate of relaxation after diazo-2 photolysis: 20 mmol/L caffeine increased the rate by 80%, whereas 10 μmol/L CGP 48506 did not alter it. The changes in rate were not a consequence of the lowered [Ca2+] used in the presence of a Ca2+ sensitizer (where force was matched to the control level) or a consequence of the potentiation of preflash force induced by the Ca2+ sensitizers (when [Ca2+] was kept at the control level), since relaxation rate was independent of the initial [Ca2+] or force (Fig 4⇑). Therefore, caffeine and CGP 48506 exert their effects on cardiac myofibrils at least partly through different mechanisms. Since we argue above that relaxation rate is largely determined by crossbridge kinetics, we suggest that caffeine increases crossbridge kinetics, although the mechanism of this action remains unclear. In contrast to the results with caffeine, the Ca2+ sensitization produced by CGP 48506 occurred without any significant change in relaxation rate, suggesting no major alteration in crossbridge kinetics. Since CGP 48506 does not affect the Ca2+ affinity of TnC,6 the increase in Ca2+ sensitivity may be due to an action on thin filament Ca2+ regulation at a step after Ca2+ binding to TnC. A high concentration (100 μmol/L) of CGP 48506 did slow relaxation by 20%, which could be due to a slowing of crossbridge kinetics. The present study focused on a lower (10 μmol/L) concentration of CGP 48506 to avoid such multiple actions. In addition, lower concentrations are more likely to be useful therapeutically.
In conclusion, the present results indicate that these Ca2+-sensitizing agents do not slow the rate of twitch relaxation through a direct reduction in the intrinsic rate of myofibrillar relaxation. In fact, we found that caffeine increased the intrinsic relaxation rate nearly 2-fold. Similarly, Simnett et al17 (1993) reported that EMD 57033 increased relaxation rate (although this may have been affected by EMD 57033 photolysis). With 10 μmol/L CGP 48506, which causes pronounced slowing of the twitch in intact muscles and cells,4 5 6 there was no change in the intrinsic rate of myofibrillar relaxation. Thus, it seems likely that the prolongation of the twitch by Ca2+ sensitizers is an unavoidable consequence of the enhanced Ca2+ sensitivity, which would cause force to be developed even at the low concentrations of intracellular Ca2+ found during the later stages of relaxation. This problem may be exacerbated by certain drugs, such as EMD 57033, which can cause activation even in the complete absence of Ca2+,5 23 although caffeine and CGP 48506 do not show this undesirable property. The fact that low concentrations (≤10 μmol/L) of CGP 48506 slow relaxation in intact cardiac muscles4 5 and cells6 without slowing the intrinsic rate of myofibrillar relaxation (present results) suggests that under these conditions the rate of relaxation during the twitch is limited by the rate of fall of [Ca2+]i rather than by the kinetic properties of the myofibrils. A similar conclusion can be made for the slowing of intact muscle relaxation by caffeine.8
This work was supported by the British Heart Foundation. We are grateful to Dr J. Layland and D. McCloskey for commenting on the manuscript.
Previously published in abstract form (J Mol Cell Cardiol. 1996;28:A227; J Physiol [Lond]. 1996;491:162P).
- Received December 2, 1996.
- Accepted January 14, 1997.
- © 1997 American Heart Association, Inc.
Lee JA, Allen DG, eds. Modulation of Cardiac Calcium Sensitivity: A New Approach to Increasing the Strength of the Heart. Oxford, England: Oxford University Press; 1993.
Butcher RW, Sutherland EW. Adenosine 3′,5′-phosphate in biological materials, 1: purification and properties of cyclic 3′,5′-nucleotide phosphodiesterase and use of this enzyme to characterize adenosine 3′,5′-phosphate in human urine. J Biol Chem. 1962;237:1244-1250.
Herold P, Herzig JW, Wenk P, Leutert T, Zbinden P, Fuhrer W, Stutz S, Schenker K, Meier M, Rihs G. 5-Methyl-6-phenyl-1,3,5,6-tetrahydro-3,6-methano-1,5-benzodiazocine-2,4-dione (BA 41899): representative of a novel class of purely calcium-sensitizing agents. J Med Chem. 1995;38:2946-2954.
Palmer S, Di Bello S, Davenport SL, Herzig JW. The novel inotropic agent CGP 48506 alters force primarily by Ca2+-independent mechanisms in porcine skinned trabeculae. Cardiovasc Res. 1996;32:411-421.
Wolska BM, Kitada Y, Palmiter KA, Westfall MV, Johnson MD, Solaro RJ. CGP-48506 increases contractility of ventricular myocytes and myofilaments by effects on actin-myosin reaction. Am J Physiol. 1996;39:H24-H32.
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 Res. 1993;73:61-70.
Allen DG, Kurihara S. Calcium transients in mammalian ventricular muscle. Eur Heart J. 1980;1:(suppl A):5-15.
Palmer S, Kentish JC. The effects of caffeine and CGP 48506 on the intrinsic relaxation rate of cardiac myofibrils. J Mol Cell Cardiol. 1996;28:A227. Abstract.
Palmer S, Kentish JC. The effects of caffeine and CGP 48506 on the relaxation rate of skinned trabeculae isolated from rat ventricles. J Physiol (Lond). 1996;491:162P. Abstract.
Mulieri LA, Hasenfuss G, Ittleman F, Blanchard EM, Alpert NR. Protection of human left ventricular myocardium from cutting injury with 2,3-butanedione monoxime. Circ Res. 1989;65:1441-1444.
Zhang R, Zhao J, Mandveno A, Potter JD. Cardiac troponin I phosphorylation increases the rate of cardiac muscle relaxation. Circ Res. 1995;76:1028-1035.
Walker JW, Hofmann PA, Lu Z, Moss RL. Rapid relaxation of single cardiac myocytes by photolysis of the Ca2+ chelator diazo-2. Biophys J. 1992;61:A20. Abstract.
Dong WJ, Rosenfeld SS, Wang CK, Gordon AM, Cheung HC. Kinetic studies of calcium binding to the regulatory site of troponin C from cardiac muscle. J Biol Chem. 1996;271:688-694.
Powers FM, Solaro RJ. Caffeine alters cardiac myofilament activity and regulation independently of Ca2+ binding to troponin C. Am J Physiol. 1995;37:C1348-C1353.
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.