INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

HEART RATE IS A POWERFUL DETERMINANT of myocardial contractility (for review see Ref. 7). In men with normal left ventricular function, hemodynamic parameters of left ventricular performance increase with increasing heart rates. This positive force-frequency relation has also been shown in isolated ventricular muscle strips from nonfailing human hearts, and the increase in twitch force was related to parallel increases in intracellular Ca²⁺ transients and sarcoplasmic reticulum (SR) Ca²⁺ content (21-23). In addition, human left ventricular myocardium is characterized by powerful postrest potentiation of contractile strength, underlining the predominance of SR Ca²⁺ handling for normal excitation-contraction (E-C) coupling (24).

Under physiological conditions, ventricular depolarization is controlled by atrial pacemaker cells, but little is known about the effects of rate and rhythm on human atrial compared with ventricular myocardial contractile performance and Ca²⁺ handling. Depending on experimental conditions, Brixius et al. (10) demonstrated a positive or a negative force-frequency relation in human atrial myocardium from nonfailing hearts, but the underlying mechanisms remained unclear. Furthermore, substantial differences in microarchitecture and E-C coupling processes between human atrial and ventricular myocardium were recently described (9, 13, 29), but measurements of intracellular Ca²⁺ transients and SR Ca²⁺ content have not been described in multicellular atrial vs. ventricular preparations under physiological conditions.

Accordingly, the major goal of the present study was to characterize the influence of stimulation frequency and rest intervals on contractile behavior, intracellular Ca²⁺ transients, and SR Ca²⁺ content [by use of rapid cooling contractures (RCCs)] in nonfailing human atrial compared with ventricular myocardium. In addition, the relative contributions of SR Ca²⁺-ATPase and Na⁺/Ca²⁺ exchange to cytosolic Ca²⁺ removal in human atrial myocardium were assessed by paired RCCs (15, 23).

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Myocardial tissue. Experiments were performed in 19 right atrial trabeculae obtained from 15 patients undergoing aortocoronary bypass operation (mean age 59 ± 3 yr, ejection fraction 59 ± 5%) and in 13 ventricular trabeculae obtained from 9 nonfailing donor hearts that could not be transplanted for technical reasons (mean age 49 ± 5 yr). None of the patients had a history of heart failure, and all had normal left ventricular function. The study was reviewed and approved by the ethical committee.

Muscle strip preparation. Muscle strips were prepared as described previously (22). Briefly, immediately after excision of the myocardium, the tissue was stored in modified Krebs-Henseleit buffer (KHB) containing (in mmol/l) 127 NaCl, 2.3 KCl, 25 NaHCO₃, 1.3 KH₂PO₄, 0.6 MgSO₄, 2.5 CaCl₂, and 11 glucose and 5 IU/l insulin, continuously bubbled with 95% O₂-5% CO₂ (pH 7.4). In addition, the KHB contained 30 mmol/l 2,3-butanedione monoxime as cardioplegic agent (20). Thin atrial or ventricular trabeculae were excised, transferred to an organ chamber, and fixed to an isometric force transducer. After washout of the cardioplegic solution, the muscles were superfused with standard KHB (without 2,3-butanedione monoxime) at 37°C and stimulated electrically at a basal stimulation frequency of 1 Hz (voltage 25% above threshold, 5-ms duration). After an equilibration period of 15-30 min, the muscles were gradually stretched (0.05- to 0.1-mm steps) to the length at which maximum force was reached. The diameter of the muscles was between 0.3 and 0.8 mm (mean 0.63 ± 0.05 for atrial and 0.60 ± 0.03 mm for ventricular muscles, not significant).

Aequorin measurements. Intracellular Ca²⁺ transients were assessed using the Ca²⁺-regulated bioluminescent photoprotein aequorin, which was macroinjected into the quiescent muscle, as described in detail previously (22, 24). Aequorin light emission was analyzed as the amplitude of the aequorin light signal (mV of amplifier output). Aequorin light signals were recorded simultaneously with twitch force on an oscilloscope (model PRO 10C, Nicolet Instrument) and on a strip chart recorder (model WR 3310, Graphtec).

RCCs. To investigate SR Ca²⁺ content, RCCs were elicited by a rapid decrease in the temperature of the muscle chamber from 37 to 1°C, as described previously (6, 19, 23). On cooling, all Ca²⁺ is released from the SR, leading to a stable contracture of the muscles, since all Ca²⁺ transport systems are blocked at the low temperature. In addition, paired RCCs were elicited during the force-frequency experiments to investigate the competition between the SR Ca²⁺ pump and Na⁺/Ca²⁺ exchange for cytosolic Ca²⁺ elimination at each stimulation frequency, as previously described (15, 23). Specifically, a first RCC (RCC1) was elicited 1 s after the last electrical stimulus, and the muscle was rewarmed after ~10 s. A second RCC (RCC2) was then elicited in the unstimulated muscle 2-4 s after rewarming of RCC1 (at the moment when the cooling contracture had completely relaxed). RCC1 releases all the Ca²⁺ from the SR and inhibits Ca²⁺ transport systems. On rewarming, the Ca²⁺ transport systems are reactivated and compete for cytosolic Ca²⁺. The fraction of Ca²⁺ taken up by the SR is available for further release at RCC2, whereas the Ca²⁺ extruded from the cell by the Na⁺/Ca²⁺ exchange system at the end of RCC1 is not. Thus the ratio of the RCC amplitudes (RCC2/RCC1) is an index of the fraction of Ca²⁺ taken up by the SR during relaxation of RCC1 (relative to that extruded by Na⁺/Ca²⁺ exchange). Twitch force, RCCs, and temperature were recorded on a strip chart recorder (model WR 3310, Graphtec).

Experimental protocol. Force-frequency relations were tested by a stepwise increase in stimulation rate from 0.25 to 3 Hz. Recordings of isometric twitch force were obtained at steady-state conditions at each frequency. To investigate the influence of rest, stepwise rest intervals between 1 and 240 s were instituted from a basal stimulation rate of 1 Hz. Twitch force, RCC amplitude, and aequorin light signal were compared with the steady-state values under control conditions (i.e., at 0.25 Hz for force-frequency relations and after 1 s of rest for postrest behavior).

Statistics. Values are means ± SE. Statistical analysis was performed with one- or two-way repeated-measurements ANOVA followed by Student-Newman-Keuls test where appropriate. Statistical significance was taken as P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Effects of stimulation frequency on force and Ca²⁺ handling in human atrial and ventricular myocardium. Increasing stimulation frequency from 0.5 to 2 Hz in human atrial myocardium resulted in an increase in isometric twitch force, aequorin light signals as a measure of intracellular Ca²⁺ transients, and RCC amplitude as a measure of SR Ca²⁺ content (Fig. 1).

View larger version (9K): [in this window] [in a new window]   Fig. 1.   Effects of increasing stimulation frequency from 0.5 to 2 Hz. Isometric twitches (A), aequorin light signals as a measure for intracellular Ca2+ transients (B), and rapid cooling contractures (RCCs) as a measure of sarcoplasmic reticulum Ca2+ content (C) in isolated human atrial trabeculae are shown.

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Frequency-dependent changes in twitch force (filled symbols), aequorin light signals (gray symbols), and RCCs (open symbols) in 19 atrial muscles from 15 human hearts (A) and 13 ventricular muscles from 9 human hearts (B). Values represent percent change in force and aequorin light signals from the basal values at 0.25 Hz. * Significantly different (P < 0.05) from steady-state values at 0.25 Hz.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Frequency-dependent changes in the ratio of the second RCC (RCC2) to the first RCC (RCC1) as a measure of the relative contribution of sarcoplasmic reticulum Ca2+-ATPase and Na+/Ca2+ exchange to cytosolic Ca2+ removal in 8 atrial muscles from 7 human hearts and 7 ventricular muscles from 4 human hearts. Values represent percent change from the basal value at 0.25 Hz. * Significantly different (P < 0.05) from steady-state values at 0.25 Hz.

Effects of rest intervals on force and SR Ca²⁺ content in human atrial and ventricular myocardium. In Fig. 4, steady-state twitches, postrest twitches, and RCCs in isolated human atrial myocardium are presented at a basal stimulation frequency of 1 Hz. After a rest interval of 5 s (see above), postrest twitch amplitude was measured. After complete equilibration of twitch force, a second rest of 5 s was followed by an RCC. Postrest twitches and RCCs were also measured after rest intervals of 120 s (see below).

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Steady-state twitches, postrest twitches, and RCCs in isolated human atrial myocardium. Basal stimulation frequency is 1 Hz. After a rest interval of 5 s (top), postrest twitch amplitude was measured. After complete equilibration of twitch force, a second 5-s rest interval was followed by an RCC. Postrest twitches and RCCs were also measured after rest intervals of 120 s (bottom).

View larger version (19K): [in this window] [in a new window]   Fig. 5.   Rest-dependent changes in the amplitudes of twitch force (filled symbols) and RCCs (open symbols) in human myocardium. Values represent means for 7 atrial muscles from 6 human hearts (A) and 6 ventricular muscles from 4 human hearts (B). All values are normalized (%) to the values obtained at steady-state conditions (1 s of rest). * Significantly different (P < 0.05) from steady-state values.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The main findings of the present study were that force-frequency behavior was similar but postrest contractile behavior in atrial myocardium was different from that in ventricular myocardium. Furthermore, these changes were related to parallel changes in intracellular Ca²⁺ transients and SR Ca²⁺ content. The data show that substantial differences in E-C coupling processes between atrial and ventricular myocardium can be demonstrated under specific experimental conditions.

Differences between atrial and ventricular myocardium. Human atrial myocytes are substantially smaller and possess a less-developed transverse tubular system than ventricular myocytes. Ultrastructural analysis revealed that atrial myocytes are characterized by a poorly developed longitudinal, but abundant corbular, SR (12, 18). In contrast to ventricular cells, a large proportion of the SR Ca²⁺ release channels is not coupled to L-type Ca²⁺ channels in the subsarcolemmal space in atrial cells but is located in the corbular SR in the central portions of the myocytes. These differences in microarchitecture have recently been related to differences in E-C coupling processes between human atrial and ventricular myocytes (13). Nevertheless, in the present study, myocardial trabeculae were investigated under physiological conditions (37°C, 0.25- to 3-Hz stimulation rate, isometric contractions), whereas Hatem et al. (13) used a much lower temperature (23°C) at very low frequencies (0.1-1 Hz) in unloaded myocytes. These methodological differences may contribute to the prolonged Ca²⁺ transients the authors found in their study, which is in contrast to the observations of short intracellular Ca²⁺ transients compared with ventricular myocardium in the present study. In addition, Hatem et al. do not report on contractile function in their work. In fact, we could show that it is important to use a physiological range of stimulation rates to investigate the relative contributions of SR Ca²⁺-ATPase and Na⁺/Ca²⁺ exchange to cytosolic Ca²⁺ elimination: it could be demonstrated that the relative contribution of SR Ca²⁺-ATPase vs. Na⁺/Ca²⁺ exchange to cytosolic Ca²⁺ elimination changes with stimulation frequency in human atrial myocardium.

Time parameters of contractions, Ca²⁺ transients, and RCCs. It was previously reported that contraction and relaxation processes are faster in atrial than in ventricular myocardium (29), but the underlying mechanisms remained unexplained. In the present study we also observed shorter time parameters of contractions in atrial than in ventricular myocardium, which were associated with substantially faster intracellular Ca²⁺ transients. Therefore, faster twitch relaxation in atrial tissue may at least in part result from faster Ca²⁺ removal from the cytosol during diastole. In transgenic mice lacking the phospholamban gene, markedly enhanced relaxation of isolated myocytes was observed (30) and related to the absence of the inhibitory function of phospholamban on SR Ca²⁺ ATPase. Accordingly, Bokník at al. (9) recently described lower protein levels of phospholamban and higher protein levels of SR Ca²⁺-ATPase in human atrial than in ventricular myocardium. In addition, these authors measured twitch force but pointed out that it would be important to compare Ca²⁺ transients in contracting atrial and ventricular preparations. The present work was influenced by their intriguing work, and thus we could show that not only twitch contractions, but also Ca²⁺ transients, are much shorter in human atrial than in ventricular myocardial trabeculae. In addition, one major advantage of our investigation is that it compares, for the first time, simultaneously stimulation frequency-dependent changes in twitch force and intracellular Ca²⁺ handling (Ca²⁺ transients and SR Ca²⁺ content); such a comparison complements previous reports on protein expression in human atrial myocardium. In summary, a higher Ca²⁺ uptake capacity of the SR may contribute to the enhanced rate of decline of Ca²⁺ transients and twitch force in human atrial myocardium. In addition, action potentials are shorter in the atrium than in the ventricle (2). Nevertheless, it cannot be ruled out that the predominant expression of -myosin heavy chain in the atrium with higher ATPase activity than of -myosin heavy chain in the ventricle may contribute to the differences in time parameters of contractions (17).

Force-frequency relation. Human atrial and ventricular myocardium was characterized by a positive force-frequency relation accompanied by a parallel increase in intracellular Ca²⁺ transients and SR Ca²⁺ content. However, for a comparable increase in force or Ca²⁺ transients, the increase in SR Ca²⁺ content was less in atrial than in ventricular myocardium (at 3 Hz ~25% of ventricle). Furthermore, at stimulation frequencies >2 Hz, twitch force and Ca²⁺ transients began to decline in the atrium, although SR Ca²⁺ content further increased in the present study. The reason for the decline in twitch force and Ca²⁺ transients at high stimulation frequencies may be that the relatively small increase in SR Ca²⁺ content cannot overcome the refractoriness of SR Ca²⁺ release processes at high stimulation rates, whereas in the ventricle the large increase in SR Ca²⁺ content more than compensates for the negative effects of increased refractoriness at high stimulation rates (23). This explanation may be supported by findings in failing human myocardium, where a negative force-frequency relation occurs without major changes in SR Ca²⁺ content (23). In addition, paired RCCs in the present study showed that the relative contribution of SR Ca²⁺-ATPase to cytosolic Ca²⁺ removal was higher in atrial (~50%) than in ventricular (~40%) myocardium at low stimulation rates. However, on an increase in stimulation frequency, the dynamic range of SR Ca²⁺-ATPase was substantially larger in ventricular than in atrial myocardium: at 3 Hz, ventricular SR Ca²⁺-ATPase removed ~75% of cytosolic Ca²⁺ during relaxation compared with only ~63% for atrial SR Ca²⁺-ATPase. If ryanodine is used to eliminate the SR from E-C coupling processes, twitch force is reduced by ~50% in atrial (26) and ventricular (27) myocardium at a stimulation rate of 1 Hz. This is in line with the findings of the paired cooling contractures in this study, where relative SR Ca²⁺-ATPase activity was similar between the two tissues at this stimulation rate but markedly differed at higher rates. However, ryanodine experiments at high stimulation frequencies have not been performed in human atrial myocardium.

Why does postrest contractile behavior differ between atrial and ventricular myocardium? Atrial myocardium exhibited clear postrest decay at longer rest intervals because of a decrease in SR Ca²⁺ content, whereas ventricular myocardium showed pronounced postrest potentiation of twitch force and SR Ca²⁺ content.

Clinical implications. Regular atrial activity improves ventricular filling and contributes to normal cardiac pump function. The positive force-frequency behavior in human atria and ventricles ascertains effective atrioventricular coupling at increasing heart rates, and increased atrial contractility may partly counterbalance impaired left ventricular filling during tachycardia. In addition, it is unknown whether the negative force-frequency relation in failing human ventricular myocardium (22) is accompanied by a persisting positive force-frequency relation in atria, possibly contributing to ventricular volume overload in patients with heart failure. In addition, our findings of distinct postrest contractile behavior might indicate that effective atrioventricular coupling is impaired during supraventricular arrhythmias, where the interval between beats may vary.