INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

POSITIVE INOTROPIC INTERVENTIONS are commonly used to enhance contractile performance in patients with decreased cardiac contractility. -Adrenergic stimulation is one of the positive inotropic strategies that can be employed to reverse this decrease in cardiac contraction in patients. Through an increase in cAMP and a consecutive phosphorylation of phospholamban, the inhibition of the sarcoplasmic reticulum (SR) Ca²⁺ pump is removed (15), resulting in an increase in SR Ca²⁺ load. This increase in SR Ca²⁺ load enhances contractility, despite phosphorylation of troponin I by isoproterenol, which induces a desensitization of the myofilaments for Ca²⁺ (27). An increase in contractile force can also be obtained by increasing the extracellular Ca²⁺ concentration ([Ca²⁺]_o), leading to an increased Ca²⁺ influx and consequent reuptake into the SR.

It has been reported that the glycolytic intermediate pyruvate acts in a positive inotropic manner and can improve contractile function in healthy and diseased animals (16, 21, 29, 32), in vitro in perfused hearts (17, 26), and in isolated myocytes (20), as well as in human patients with heart failure (9). Multiple mechanisms have been postulated to underlie the positive inotropic effect of pyruvate. The main proposed actions by which pyruvate enhances contractility include an increase in phosphorylation potential, a modulation of pH, changes in the cytosolic redox state, and a reduction in inorganic phosphate (P_i) (3, 16, 18, 26). The improvement of the overall energy state of the myocardium in the presence of pyruvate increases the thermodynamic driving force of the SR Ca²⁺ pump, leading to an increased SR Ca²⁺ gradient (4).

Accordingly, we tested the hypothesis that pyruvate can potentiate the positive inotropic response to isoproterenol or increased [Ca²⁺]_o. In isolated multicellular myocardial preparations from rabbit, we investigated changes in contractile parameters after addition of pyruvate, isoproterenol, and Ca²⁺, alone or in combination. Pyruvate can potentiate the isoproterenol-induced increases in contractility as well as the maximal force-generating capacity under force-saturating [Ca²⁺]_o. We conclude that a combination of different inotropic agents may create a more effective inotropic therapy.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Myocardial preparations. Female New Zealand White rabbits weighing 1.5-2.5 kg were anesthetized with thiopental sodium (50 mg/kg) via the ear vein after heparinization (1,000-2,000 IU). All experiments were carried out in accordance with institutional guidelines regarding the care and use of animals. Hearts were rapidly dissected and retrogradely perfused at room temperature through the aorta with a modified Krebs-Henseleit (KH) solution containing (in mM) 120 NaCl, 5 KCl, 2 MgSO₄, 1.2 NaH₂PO₄, 20 NaHCO₃, 10 glucose, and 0.25 CaCl₂; 20 mM 2,3-butanedione monoxime (BDM) was added as a cardioprotective agent (23). This solution was in equilibrium with 95% O₂-5% CO₂, resulting in pH 7.4. From these hearts, small, free-running and unbranched trabeculae were dissected from the right (n = 27) or left (n = 4) ventricle, as previously described (10-12). With the aid of a stereomicroscope, the dimensions of the preparations were measured at ×40 magnification (resolving power ~10 µm). Preparations were mounted in the experimental setup in the BDM-containing KH solution, which was immediately switched to a KH solution without BDM. Average dimensions of the preparations included in the data were 335 ± 33 µm wide, 278 ± 35 µm thick, and 2,127 ± 145 µm long (n = 28).

Experimental apparatus and protocol. Muscles were mounted using two blocks of ventricular or valvular tissue in the experimental setup between a basket-shaped extension (10-11, 12, 30) of a force transducer (model KG-4, Scientific Instruments, Heidelberg, Germany) and a hook connected to a microdisplacement device. After the muscle was mounted, superfusion with KH solution (at 37°C) was started, and [Ca²⁺]_o was raised from 0.25 to 1 mM in steps of 0.25 mM every 2-5 min. At 1.0 mM, stimulation was started through 3- to 5-ms asymmetric pulses at 20% above threshold voltage (typically 2-4 V) at 1.0 Hz. [Ca²⁺]_o was then further increased to 1.25 or 2.5 mM, depending on the protocol. Next, the muscle was carefully stretched in several small steps until active developed force (F_dev) did not or only slightly further increased on lengthening or until diastolic force exceeded 5 mN/mm². This muscle length reflects a sarcomere length of ~2.1-2.2 µm (14, 30). Under these conditions, time-dependent deterioration of contractility could be minimized. The muscles were left contracting under these conditions for at least an additional 45 min to equilibrate. After equilibration of the muscle, a concentration-response curve of isoproterenol was measured. From a 10⁴ M stock solution (containing 0.3 mM ascorbic acid to prevent oxidation of isoproterenol), the isoproterenol concentration was sequentially set (after stabilization at the last given concentration) to 10⁹, 10⁸, 10⁷, and finally 10⁶ M. At this concentration, 10 mM pyruvate was given, and after contractile parameters had stabilized, the preparation was superfused with fresh KH solution, without isoproterenol or pyruvate at the same [Ca²⁺]_o, to wash out these compounds. After completion of the washout (stabilization of contractile parameters), 10 mM pyruvate was given. Under pyruvate, the isoproterenol concentration-response curve was repeated. To obtain the maximal twitch-force capacity, at the highest isoproterenol concentration (1 µM), [Ca²⁺]_o was increased to 10 mM. The dose of 10 mM pyruvate was chosen because it has been shown to induce a sizable effect and to allow for comparison with other studies (9, 20, 29). Also the pH of the solution remained unaltered after addition of 10 mM pyruvate. Pilot experiments revealed that isoproterenol did not change force if the baseline protocol was repeated several times; i.e., a change observed in the second dose-response curve is due to the intervention applied.

Data analysis and statistics. Intact rabbit muscles were discarded (n = 3 of 31) when maximal F_dev during the protocol did not reach 20 mN/mm² or time-dependent loss of force (rundown) during the experiment exceeded 15%/h. Data were collected (1 kHz/channel) and analyzed off-line with a custom-designed data acquisition program written in LabView (National Instruments). From twitch contractions the following parameters were analyzed: diastolic force (in mN/mm²), F_dev (in mN/mm²), time from stimulation to peak tension (in ms) to 50% (in ms) and 90% relaxation (in ms), time from peak tension to 50% relaxation (in ms), and the maximal and minimal derivatives of force generation (in mN · mm² · s¹). The program contained an on-line analysis mode to quantify these contractile parameters during the experiment. Statistical significance was determined by Student's t-test for paired or unpaired data where applicable. For analyzing dose-response measurements (for isoproterenol or Ca²⁺) in the presence and absence of pyruvate, repeated-measures (multifactorial) ANOVA (MANOVA) was performed. Values are means ± SE unless stated otherwise. Two-tailed P < 0.05 was accepted as significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The contractile parameters of the preparations were sufficiently stable over time to permit measurement of consecutive concentration-response curves; time-dependent deterioration of the preparations was measured by comparing contractile parameters under identical conditions throughout the protocol. During a protocol at [Ca²⁺]_o of 1.25 mM, 10 mM pyruvate, and 10⁶ M isoproterenol, F_dev was measured at an interval of 1.5-2 h and remained nearly unchanged (from 30.3 ± 3.4 to 33.3 ± 3.2 mN/mm², not significant). To explore potential artifacts due to prolonged isoproterenol exposure, in several pilot experiments we repeated the isoproterenol dose-response curve under baseline conditions and observed no changes between the second and the first run.

We first investigated the impact of pyruvate on basic contractility. Figure 1 depicts the time course of application of pyruvate (10 mM) in the presence of [Ca²⁺]_o of 1.25 mM and in the same preparation in the presence of 10⁶ M isoproterenol. An increase to 10⁵ M isoproterenol did not further increase force development. Under both conditions, pyruvate induced a biphasic behavior of force development. First, a transient decrease in F_dev was observed, reaching a minimum of ~50-70% of the starting force after ~5 min. Then, force increased until a new steady state was reached (after 10-30 min) on a higher level, as before addition of pyruvate. Time courses of this biphasic behavior were similar in all experiments and were slightly slower in thicker preparations.

View larger version (10K): [in this window] [in a new window]   Fig. 1.   Original traces of the time course of force development after addition of 10 mM pyruvate (Py) under different inotropic conditions: 1.25 mM Ca2+ (Ca, left) and 1.25 mM Ca2+ + 106 M isoproterenol (Ca + Iso, right). Arrows, addition of pyruvate. Dimensions of the preparation were 250 µm wide, 175 µm thick, and 3.0 mm long.

Next we investigated the influence of pyruvate on isoproterenol-induced enhancement of contractility. Figure 2 depicts original twitch recordings obtained during an isoproterenol concentration-response curve with and without 10 mM pyruvate. The response to isoproterenol was potentiated by pyruvate, i.e., force in the presence of both pyruvate and isoproterenol was higher than force with isoproterenol plus the increase in force with pyruvate in the absence of isoproterenol (MANOVA, P < 0.001). Figure 3A shows the isoproterenol concentration-response curve in the absence and presence of 10 mM pyruvate. At each isoproterenol concentration, F_dev was higher in the presence of pyruvate. Pyruvate alone increased F_dev by 1.7 ± 0.3 mN/mm² (P < 0.002) and isoproterenol (10⁶ M) by 23.0 ± 2.3 mN/mm² (P < 0.001), whereas the combination of isoproterenol and pyruvate resulted in an increase of 31.4 ± 3.3 mN/mm², which was higher than the addition of the individual effects (MANOVA, P < 0.001). In the presence of [Ca²⁺]_o of 1.25 mM, at saturating isoproterenol concentrations, the addition of 10 mM pyruvate led to a significant increase in F_dev from 25.1 ± 2.1 to 30.3 ± 3.4 mN/mm² (P < 0.05; Fig. 3B). Furthermore, the entire isoproterenol concentration-response protocol was repeated in a different set of muscles at a basal Ca²⁺ concentration of 2.5 mM (n = 11; not shown). In this increased basal inotropic state (F_dev = 7.9 ± 1.2 vs. 1.9 ± 0.4 mN/mm² at 1.25 mM Ca²⁺), the potentiating effect of 10 mM pyruvate was slightly smaller but significant at 10⁷ and 10⁶ M isoproterenol. The attenuation of the potentiating effect is most likely due to the fact that, under an increased inotropic baseline, the potential for absolute improvement is lowered, diminishing the absolute increases in F_dev.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Representative recordings of the positive inotropic effect of isoproterenol on twitch contractions in the absence and presence of 10 mM pyruvate. Extracellular Ca2+ concentration was 1.25 mM. Dimensions of the preparation were 225 µm wide, 175 µm thick, and 2.0 mm long.

View larger version (29K): [in this window] [in a new window]   Fig. 3.   Potentiating effect of pyruvate on isoproterenol-induced positive inotropy. A: average data of developed force during an isoproterenol concentration-response curve before and after addition of 10 mM pyruvate at 1.25 mM Ca2+ (n = 8). The isoproterenol response in the presence of pyruvate was significantly potentiated at all concentrations; i.e., isoproterenol-induced increase in force development was larger in the presence of pyruvate. B: under saturating isoproterenol concentrations (106 M), 10 mM pyruvate could further increase force development. * Significant (P < 0.05) increase compared with the same data in the absence of pyruvate.

The contractile and twitch timing parameters are given in Table 1. Isoproterenol decreased time to peak tension and relaxation times significantly. In contrast, pyruvate increased twitch timing parameters, whereas the cumulative application of pyruvate and isoproterenol counterbalanced the effects on twitch timing to a certain extent. No significant decreases or increases in diastolic tension were observed by any of the interventions alone or by a combination of the interventions.

Sensitivity of the preparations to isoproterenol was calculated from the concentration-response curves of the individual experiments (Fig. 4). At a [Ca²⁺]_o of 1.25 mM, pyruvate significantly sensitized the effect of isoproterenol on F_dev. The EC₅₀ (concentration at which isoproterenol exerts 50% of its maximal response) was 3.59 ± 0.99 × 10⁸ M in the absence of pyruvate and shifted to the left in the presence of pyruvate to 2.54 ± 0.35 × 10⁸ M (P < 0.05), indicating that pyruvate-induced potentiation of the isoproterenol response was associated with changes in isoproterenol sensitivity.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Fraction of maximal force increase by isoproterenol at 1.25 mM Ca2+ (control) and in the presence of 10 mM pyruvate (n = 8). The half-maximal effect (EC50) was reached at significantly lower concentrations in the presence of 10 mM pyruvate, whereas the slope of the curve at EC50 remained unchanged. * Significant difference between interventions (P < 0.05).

To investigate whether the effect of pyruvate was specific for pyruvate rather than simply because pyruvate is another "fuel," in two additional preparations we conducted the following protocol. After the preparation was mounted and equilibrated, 10 mM acetate was applied. Then the acetate was washed out, and force returned to baseline. Pyruvate (10 mM) was added, and after the preparation reached steady state, pyruvate was washed out and force again returned to baseline. Next, 10 mM lactate was added. The comparison of the three fuels was very clear: acetate increased force by 17% (muscle 1) and 15% (muscle 2), pyruvate increased force by 115 and 136%, and lactate decreased force by 2% in one muscle and increased force by 3% in the other. Thus the inotropic effect was much more pronounced for pyruvate than for the other fuels.

To investigate the potential of pyruvate to increase SR load, we measured RCCs at different Ca²⁺ concentrations before and after addition of pyruvate. Therefore, a Ca²⁺ concentration-response curve was measured in the absence and then in the presence of 10 mM pyruvate. The absolute increases in F_dev with increased Ca²⁺ were higher (MANOVA, P < 0.05) after addition of 10 mM pyruvate than before addition of pyruvate (Fig. 5A). The effect of pyruvate on Ca²⁺-activated force seems to be additive; a significant statistical interaction between Ca²⁺ and pyruvate was found (P < 0.05). However, RCC amplitude was not dependent on pyruvate over the Ca²⁺ dose-response curve (P = 0.074). Interestingly, although no further increase in F_dev could be evoked by increasing [Ca²⁺]_o from 8 to 16 mM, 10 mM pyruvate could increase F_dev further (Fig. 5C), from 26.4 ± 6.8 to 29.7 ± 7.1 mN/mm² (P < 0.05, n = 9). Stability of the preparations during the experimental protocols was confirmed by the observation that, under pyruvate, at [Ca²⁺]_o of 16.0 mM, F_dev was 31.1 ± 7.1 mN/mm², which was unchanged from 29.7 ± 7.1 mN/mm² in the first concentration-response curve after addition of pyruvate.

View larger version (26K): [in this window] [in a new window]   Fig. 5.   Effect of Ca2+ and pyruvate on force development and rapid cooling contractures. A: force development at different Ca2+ concentrations in the absence and presence of 10 mM pyruvate (n = 8). Addition of pyruvate before Ca2+ concentration-response curve shifted the entire curve upward. B: rapid cooling contractures (RCC) revealed that, at increasing Ca2+ concentration, SR load increased. Pyruvate increased sarcoplasmic reticulum load at 1.25 mM Ca2+, but no significant increases in sarcoplasmic reticulum load were observed at higher Ca2+ concentrations. C: although no further increase in force development was observed by increasing Ca2+ from 12 to 16 mM (A), addition of pyruvate (10 mM) led to a significant further increase in developed force at 16 mM Ca2+. *Significant increase (P<0.05) compared with control data.

RCCs (Fig. 5B) revealed that at increased [Ca²⁺]_o the SR content was increased (P < 0.001). At higher Ca²⁺ concentrations (8-16 mM), although F_dev did not further increase (Fig. 5A, open bars), RCC amplitude continuously increased. Addition of pyruvate (Fig. 5B, open bars) at low Ca²⁺ concentrations increased SR load, as indicated by an increase in RCC amplitude from 7.0 ± 1.6 to 9.0 ± 1.9 mN/mm² (P < 0.05, n = 9). Because pyruvate impacts mainly on F_dev, it may indicate augmentation of fractional release of Ca²⁺ from the SR. Alternatively, this could suggest that not just the amplitude of the Ca²⁺ transient but also kinetics of Ca²⁺ cycling and/or cross-bridge cycling are affected by pyruvate. More evidence to support this possibility was obtained by analysis of twitch timing parameters, which were also affected by pyruvate (Table 2). Time from stimulation to peak tension and time from peak tension to half-relaxation were not altered by increasing Ca²⁺ concentrations but were significantly increased in the presence of pyruvate.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present data show that the addition of pyruvate to other inotropic interventions, i.e., isoproterenol and increased [Ca²⁺]_o, results in a potentiation of their individual effects in nonfailing rabbit myocardium. At concentrations of isoproterenol (10⁶ M) or Ca²⁺ (16 mM) that were maximal, pyruvate (10 mM) was able to further increase F_dev. In addition, pyruvate further increased SR Ca²⁺ load at saturating Ca²⁺ concentrations (16 mM). This indicates that 1) the potentiating effect of pyruvate may partly result from stimulation of SR Ca²⁺ accumulation and 2) pyruvate and comparable interventions may be useful clinically to reduce concentrations of inotropic interventions in the treatment of acute heart failure.

The extra inotropic effect of pyruvate may result from 1) augmentation of SR Ca²⁺ release, 2) increased myofilament sensitivity, and 3) altered cross-bridge kinetics. SR Ca²⁺ release could be augmented by an increased Ca²⁺ accumulation or by an increase in the fractional release of SR Ca²⁺. To test the possibility of increased SR Ca²⁺ accumulation, we used RCCs. By rapidly cooling the preparation, the SR Ca²⁺ is dumped into the cytosol and the subsequent developed contracture is an index of SR Ca²⁺ content (1, 2). Accordingly, the fact that pyruvate induced a further increase in RCC under low inotropic baseline conditions and increased isometric force at maximal inotropic Ca²⁺ concentration may indicate that pyruvate increases the maximal velocity of the SR Ca²⁺ pump and the maximal trans-SR Ca²⁺ gradient. This is consistent with data from Chen et al. (4) using an NMR technique to evaluate SR Ca²⁺ load. Furthermore, the finding of an increase in isometric force at all [Ca²⁺]_o shows that, apart from the Ca²⁺ effect per se (19), the effect of pyruvate on Ca²⁺ cycling was additional and resulted from a different mechanism. Previous work has suggested that pyruvate may act in part through thermodynamic stimulation of the SR Ca²⁺ pump by an increase in the phosphorylation potential and free energy available from ATP hydrolysis (3, 16, 18). However, the results indicate that pyruvate must mainly cause an increase in fractional SR Ca²⁺ release or other sources of Ca²⁺ or points to a downstream mechanism of action, i.e., changes in myofilament Ca²⁺ sensitivity and/or cross-bridge cycling kinetics. Thus, although SR Ca²⁺ content may be, under certain conditions, elevated by pyruvate, the results indicate that at least one other mechanism must underlie the positive inotropic effect of pyruvate.

In isoproterenol experiments, pretreatment with pyruvate not only augmented contractile force but also resulted in a potentiation and increased sensitivity of force to isoproterenol. This may indicate that kinetic stimulation of the SR pump by phosphorylation of phospholamban (15), combined with thermodynamic stimulation by pyruvate (3, 4, 16, 18) and/or secondary effects of pyruvate, resulted in an overproportional effect on SR pump activity.

Our observation that pyruvate increased twitch timing parameters independent of [Ca²⁺]_o, provides more information regarding pyruvate's possible mechanisms. Changes in levels of P_i and ADP have been shown to have an impact on SR Ca²⁺ handling (31) and on force development (13). Because pyruvate has been reported to increase the phosphocreatine-to-P_i ratio and, thereby, to decrease P_i levels (18, 33), part of its inotropic effect could be mediated via a modulation of Ca²⁺ kinetics and/or cross-bridge kinetics and remains to be investigated. Also, it is known that the level of force development of isometric contractions per se prolongs relaxation in these preparations (10). Thus whether the prolongation of the twitch is solely due to the increased force levels or results from a specific effect of pyruvate is unknown. Additionally, that the effect of pyruvate on force development is altered through at least a second mechanism could also be concluded from the time course of force development after application of pyruvate. In all experiments (n = 28), after administration of 10 mM pyruvate, F_dev decreased significantly (30-50%) over several minutes and only then slowly (5-25 min) increased to a constant level higher than the initial F_dev (positive inotropic effect). The negative inotropic effect could be explained by the H⁺-pyruvate cotransport by the monocarboxylate symporter (25), transiently acidifying the cytoplasm and reducing force development (5). The transient aspect of this response might be partially explained by reversal of the intracellular acidosis through other processes (e.g., via the sodium/proton exchanger) and partially by a slower time course of the positive inotropic effect through the increase in phosphorylation potential. During washout of pyruvate, the opposite effect on force development was observed. F_dev increased by an additional 10-40% before declining to pre-pyruvate levels, most likely by the reversal of the mechanism responsible for the wash-in effect.

Increased inotropy comes at the cost of more ATP, reflected by an increase in oxygen consumption (22), and has been reported for various positive inotropic agents, e.g., isoproterenol, Ca²⁺, or Ca²⁺-sensitizing agents (6, 28). When this extra force/pressure production by the myocardium is generated by an overproportional amount of ATP consumption, economy of contraction worsens. A decreased economy is a major disadvantage in the application of positive inotropic agents in the treatment of heart failure. -Adrenergic stimulation induces a desensitization of the myofilaments for Ca²⁺ through phosphorylation of troponin I (27) and removal of the SR Ca²⁺-ATPase inhibition through phosphorylation of phospholamban (15). Thus a substantial fraction of the increased Ca²⁺ is "wasted" to overcome the desensitization of the myofilaments. Additionally, increased futile cycling may occur under increased cAMP levels, adding to the "wasted energy" under catecholamine stimulation (24). These processes are achieved at the cost of extra ATP, reflected by an overproportional increase in oxygen consumption and/or heat production compared with the increase in F_dev/pressure, as has been shown in studies on animal (7, 28) and human myocardium (8). Because pyruvate does not seem to decrease Ca²⁺ sensitivity, low doses of -adrenergic agonists in combination with low doses of pyruvate might reduce the amount of "wasted" ATP and result in a better economy of contraction than is the case with equivalent increases in inotropy induced by -adrenergic agonists alone. Thus addition of pyruvate may be useful to reduce catecholamines, and thus side effects of those drugs, in the treatment of patients with acute heart failure.