The Sarcoplasmic Reticulum and the Na+/Ca2+ Exchanger Both Contribute to the Ca2+ Transient of Failing Human Ventricular Myocytes
Abstract—Our objective was to determine the respective roles of the sarcoplasmic reticulum (SR) and the Na+/Ca2+ exchanger in the small, slowly decaying Ca2+ transients of failing human ventricular myocytes. Left ventricular myocytes were isolated from explanted hearts of patients with severe heart failure (n=18). Cytosolic Ca2+, contraction, and action potentials were measured by using indo-1, edge detection, and patch pipettes, respectively. Selective inhibitors of SR Ca2+ transport (thapsigargin) and reverse-mode Na+/Ca2+ exchange activity (No. 7943, Kanebo Ltd) were used to define the respective contribution of these processes to the Ca2+ transient. Ca2+ transients and contractions induced by action potentials (AP transients) at 0.5 Hz exhibited phasic and tonic components. The duration of the tonic component was determined by the action potential duration. Ca2+ transients induced by caffeine (Caf transients) exhibited only a phasic component with a rapid rate of decay that was dependent on extracellular Na+. The SR Ca2+-ATPase inhibitor thapsigargin abolished the phasic component of the AP Ca2+ transient and of the Caf transient but had no significant effect on the tonic component of the AP transient. The Na+/Ca2+ exchange inhibitor No. 7943 eliminated the tonic component of the AP transient and reduced the magnitude of the phasic component. In failing human myocytes, Ca2+ transients and contractions exhibit an SR-related, phasic component and a slow, reverse-mode Na+/Ca2+ exchange–related tonic component. These findings suggest that Ca2+ influx via reverse-mode Na+/Ca2+ exchange during the action potential may contribute to the slow decay of the Ca2+ transient in failing human myocytes.
Numerous processes contribute to the rise and fall of the intracellular free Ca2+ that induces contraction and relaxation in cardiac myocytes. Ca2+ influx through sarcolemmal, voltage-dependent Ca2+ channels and the Na+/Ca2+ exchanger (reverse mode), together with Ca2+ release from the sarcoplasmic reticulum (SR), is thought to be primarily responsible for the rapid increase in cytosolic free Ca2+ (upstroke of the whole-cell Ca2+ transient). Ca2+ efflux via the Na+/Ca2+ exchanger (forward mode) and the sarcolemmal Ca2+-ATPase, together with Ca2+ uptake by the SR Ca2+-ATPase and the mitochondrial Ca2+ uniporter, is responsible for the time course of the decay of the Ca2+ transient.1 The respective contribution of these Ca2+ transport processes to the Ca2+ transient is species dependent.2 3 4 5 In small mammals with fast heart rates and short-duration action potentials (ie, mice and rats), the size and shape of the Ca2+ transient are almost entirely determined by SR release and uptake.6 In larger mammals with slower heart rates and long action potential durations, transsarcolemmal ionic fluxes, particularly through the Na+/Ca2+ exchanger, appear to play a significantly greater role.6 7 8
In human heart failure Ca2+ homeostasis is disturbed, as indicated by the smaller and more slowly decaying Ca2+ transients recorded from failing versus nonfailing myocytes.9 10 11 12 These Ca2+ transient derangements are thought to be largely responsible for the depressed contractility of the failing heart. The role of specific Ca2+ transport processes to the deranged Ca2+ transients of failing human myocytes has not been well established. There do not appear to be significant changes in the L-type Ca2+ current density in human heart failure.12 13 14 However, several investigations imply that the slow decay of the Ca2+ transients is related to reduced SR Ca2+-ATPase levels (protein and mRNA).15 16 Unfortunately, other studies have been unable to demonstrate any changes in the abundance17 or activity18 19 of SR Ca2+-ATPase proteins in failing human hearts. Therefore, the role of altered SR function in contractile disturbances of the failing heart is still not resolved.
Studies of the Na+/Ca2+ exchanger in human heart failure have produced more consistent results. These studies have primarily shown that the abundance of exchanger mRNA and protein is increased in heart failure.20 21 Because the Na+/Ca2+ exchanger is the principal Ca2+ efflux mechanism in mammalian myocytes, it has been suggested that increased activity of the exchanger in the failing heart could help compensate for the associated decrease in SR Ca2+ uptake.20 However, because the Na+/Ca2+ exchange can also transport Ca2+ into the cell, increased exchanger activity could produce increased Ca2+ influx during the action potential. This notion has not been examined in detail to date.
Recently, we developed an improved technique to isolate high-quality myocytes from failing human hearts. These myocytes were used in the present experiments to study the Ca2+ transport processes responsible for the small magnitude and the slow decay of Ca2+ transients in failing human left ventricular myocytes. We specifically explored the roles of the SR Ca2+-ATPase and the Na+/Ca2+ exchanger in the decay of the Ca2+ transient in these cells. Our experiments suggest that the small size of the Ca2+ transient results from reduced SR Ca2+ stores and that the slow decay of the transient is in part due to Ca2+ influx via reverse-mode Na+/Ca2+ exchange during the action potential.
Materials and Methods
Human ventricular myocardium was obtained from 18 patients with severe heart failure undergoing orthotopic cardiac transplantation. Heart failure was secondary to ischemic (n=12) or nonischemic (n=6) cardiomyopathy. Our protocol was reviewed by the Temple University Institutional Review Board and was determined to be exempt in accordance with paragraph 4 pertaining to research involving pathological specimens. Patient characteristics are presented in Table 1⇓.
To ensure myocyte preservation during surgery, the coronary arteries were perfused with cold, blood-containing cardioplegic solution in vivo at the time of aortic cross-clamping. This technique was used to minimize subsequent ischemia/reperfusion damage during myocyte isolation. Explanted hearts were then transported (within 5 minutes) from the operating suite to the laboratory in cold, Ca2+-free Krebs-Henseleit (KH) solution containing (in mmol/L) glucose 12.5, KCl 5.4, lactic acid 1, MgSO4 1.2, NaCl 130, NaH2PO4 1.2, NaHCO3 25, and sodium pyruvate 2 (pH 7.4, with NaOH).
On arrival in the laboratory (<5 minutes), the heart was weighed and rinsed in KH, and a small catheter was placed into the lumen of an artery that supplied a noninfarcted free-wall region of the left ventricle. The perfused myocardial segment was cut from the heart and rinsed for 30 minutes with a nonrecirculating KH solution containing 10 mmol/L taurine. The tissue was then perfused for 30 minutes with 200 mL of KH containing 180 U/mL collagenase, 20 mmol/L 2,3-butanedione monoxime (BDM), 20 mmol/L taurine, and 0.05 mmol/L CaCl2. This solution was recirculated. The collagenase-containing solution was washed from the tissue for 10 minutes with 500 mL KH containing 10 mmol/L taurine, 20 mmol/L BDM, and 0.2 mmol/L CaCl2. The cardiac tissue was then removed from the cannula, and only midmyocardial tissue was minced. The resulting cell suspension was filtered, centrifuged (25g for 1 minute), and resuspended in a KH solution (200 mL KH, 1% weight/volume BSA, 10 mmol/L taurine, and 0.25 mmol/L CaCl2). All solutions were equilibrated with 95% O2 and 5% CO2. The temperature was kept at 37°C throughout the isolation. Initial yields of rod-shaped cells were between 7% and 70%. All experiments were conducted within 12 hours of cell isolation.
Myocyte Contraction and Ca2+ Transient Measurements
Myocytes were placed in a chamber mounted on an inverted microscope. The chamber (0.5 mL) was superfused at 1 to 2 mL/min with Tyrode’s solution (in mmol/L): NaCl 150, KCl 5.4, CaCl2 1, MgCl2 1.2, glucose 10, sodium pyruvate 2, and HEPES 5, pH 7.4 at 37°C. Myocytes were chosen on the basis of their morphology (rod shape, no hypercontracted areas) and absence of spontaneous contractions in 1 mmol/L Ca2+. Myocyte contractions were measured by using an edge-detection (Crescent Electronics) technique. Data were recorded and analyzed by using Axotape software (Axon Instruments). The maximal magnitude of contraction was normalized to resting cell length and expressed as percent shortening.
Indo 1 fluorescence was recorded from single myocytes as described previously.11 The excitation was at 350 nm, and the emission light was split through a 460-nm diachronic mirror. The emitted indo 1 fluorescence at 410 nm/480 nm (Ca2+ bound/Ca2+ free) was recorded to represent the cytosolic Ca2+ transient. Myocytes were placed in KH solution containing 4.8 mmol/L indo 1-AM for 2 minutes and then rinsed in Tyrode’s solution for 10 minutes. With this loading technique, cytosolic indo 1-AM is almost completely hydrolyzed and there is minimal cellular compartmentalization. The amount of indo 1 in cellular compartments, such as mitochondria and the SR, was determined by “quenching” the cytosolic indo 1 signal with 50 μmol/L Mn2+. Exposure to Mn2+ abolished the indo 1 transient, and the remaining fluorescence was not significantly different from cellular autofluorescence (n=5 myocytes). These data therefore show that indo 1 is located primarily in the cytoplasm. As an additional control, twitch contractions were measured in myocytes with and without indo 1 to ensure that the loading procedure did not cause any significant buffering of the Ca2+ transient and contraction. Only those myocytes without any significant buffering were used. Although our technique uses small amounts of cytosolic indo 1 to minimally perturb cellular Ca2+ buffering, the signal-to-noise characteristics are acceptable only when photon counting techniques are used.22
Myocytes were field stimulated at 0.5 Hz with electrodes on the sides of the experimental chamber until the Ca2+ transients and contractions reached a steady state. The quantity of the Ca2+ stored in the SR was determined by rapidly applying caffeine (termed a caffeine “spritz”) to the myocyte for either 100 ms or 10 seconds through a large-bore micropipette placed just above the cell. Caffeine-induced SR Ca2+ release was induced in place of normal electrical stimulation. Caffeine (10 mmol/L) was used because it induces full SR Ca2+ release, as verified by failure of a subsequent caffeine spritz to cause a Ca2+ transient.
Selective inhibitors of SR Ca2+ transport (thapsigargin, 0.1 μmol/L)23 and reverse-mode Na+/Ca2+ exchange [No. 7943; (2-(2-4-(4-nitrobenzyloxy)phenyl)ethyl)isothiourea methanesulfonate), 1 μmol/L]24 activity were used to define their respective contributions to the Ca2+ transient. In the thapsigargin experiments, failure of a caffeine spritz to induce SR Ca2+ release was used as an indicator of SR unloading/emptying. Previous studies showed that 3 to 10 μmol/L No. 7943 is sufficient to block Na+/Ca2+ exchange–mediated Ca2+ influx without significantly altering the other ion transporters, such as the Na+/H+ exchanger, L-type Ca2+ channels, sarcolemmal and SR Ca2+-ATPases, Na+,K+-ATPase, and passive Na+ permeability.24 The β-adrenergic agonist isoproterenol (1 μmol/L) was used to modify the SR component of the Ca2+ transient. Preliminary experiments showed that this concentration of isoproterenol induces maximal effects on failing myocytes without signs of toxicity.
Action Potential Measurements
Action potentials were recorded as described in detail in numerous previous studies from this laboratory. In brief, an aliquot of cells was placed in a heated (37°C) chamber on the stage of an inverted microscope (Zeiss Axiovert 10). Normal Tyrode’s bath solution contained (in mmol/L) glucose 10, HEPES 5, KCl 5.4, MgCl2 1.2, NaCl 150, and sodium pyruvate 2 (pH 7.4 with NaOH). Low-resistance (1.5 to 4 MΩ) glass patch pipettes were used to gain electrical access to the cell interior. Fire-polished patch pipettes were filled with a standard solution that contained (in mmol/L) HEPES 20, KCl 130, NaCl 10, MgCl2 5, and K2ATP 5 (pH 7.2, with KOH). Membrane potentials were recorded with an amplifier (Axoclamp 2A, Axon Instruments) connected to a personal computer. Passing current through the recording pipette induced action potentials. Data were analyzed with pclamp 6.0 (Axon Instruments) software.
Taurine, BDM, isoproterenol, thapsigargin, and albumin were obtained from Sigma Chemical Co. Collagenase (type II) was from Worthington Biochemical Co. Indo 1-AM was obtained from Calbiochem. Compound No. 7943 [2-(2-4-(4-nitrobenzyloxy)phenyl)-ethyl)isothiourea methanesulfonate] was kindly provided by the New Drug Discovery Research Drug Laboratory, Kanebo Ltd, Osaka, Japan.
All data in the text and tables are reported as mean±SEM. Differences between treatments on the same cell were assessed by 2-tailed, paired t tests. A probability level of P<0.05 was set as significant.
Field Stimulation–Induced Ca2+ Transients and Contractions
Steady-state contractions (n=21) and Ca2+ transients (n=36) induced by electrical (field) stimulation were measured in failing human ventricular myocytes. Data from a representative myocyte are shown in Figure 1⇓, and the average data are listed in Table 2⇓. In these experiments, Ca2+ transients and contractions exhibited both phasic and tonic components in every cell studied (0.5-Hz stimulation rate). These findings suggest that 2 processes may make significant contributions to the Ca2+ transient of failing human ventricular myocytes.
Relationship of Contraction and Action Potential Duration
In a recent report from this laboratory, we showed that when contractions are induced with depolarizing voltage-clamp steps, the phasic portion of the contraction decayed during depolarization while the secondary tonic component was maintained until the membrane potential was repolarized.25 The results of this previous study25 suggested that the level and duration of the plateau phase of the action potential may determine the magnitude and duration of the tonic component of contraction. To examine this idea further, we measured action potentials and associated contractions in 15 failing human ventricular myocytes. These experiments showed that at a slow rate of stimulation (0.5 Hz), the phasic component of contraction occurs during the action potential plateau and that the tonic component is maintained until final repolarization occurs (Figure 2A⇓). Fluctuations in action potential duration at a fixed, slow rate of stimulation were associated with simultaneous changes in the tonic component of the contraction while the phasic component was not markedly changed. In every myocyte studied, we found a linear relationship between contraction and action potential duration (Figure 2B⇓).
Both the action potential duration and the phasic and tonic components of contraction were influenced by the frequency of stimulation. Increasing the stimulation frequency from 0.5 to 1.5 Hz caused the action potential duration to decrease (Figure 3⇓, top). Additionally, this increase in stimulation frequency also caused a decrease in the magnitude of the phasic component of contraction and the shortening and eventual elimination of the tonic portion of the contraction (Figure 3⇓, bottom). These results strongly support the idea that voltage-dependent processes determine the tonic component of contraction and influence the phasic component of contraction.
Caffeine-Induced Ca2+ Transients
The magnitude of the phasic component of the Ca2+ transients in failing human myocytes is smaller than what we have observed in normal myocytes from other species studied under identical experimental conditions.26 The idea that peak systolic Ca2+ is smaller than normal in failing human ventricular myocytes is also well supported by a number of studies from other laboratories.12 27 28 The reduced peak systolic Ca2+ of failing human myocytes could result from defects in Ca2+ influx, excitation-contraction coupling, and/or SR Ca2+ loading. We studied the idea that the amount of Ca2+ in the SR is abnormal by inducing full SR Ca2+ release with a caffeine spritz that was substituted for an action potential.
Previous studies from this and other laboratories8 10 have shown that caffeine-induced Ca2+ transients (Caf transients) usually have a higher peak Ca2+ and a slower rate of rise and decay than do the action potential–induced Ca2+ transients (AP transients). The reason that Caf Ca2+ transients are larger than AP transients appears to be that caffeine induces full SR Ca2+ release, whereas normal excitation-contraction coupling causes only a fraction of the SR Ca2+ stores to be released.3 4 5 8 The reason that the decay rate of Caf Ca2+ transients is slower than the decay rate of AP transients is thought to be that Na+/Ca2+ exchange, which transports cytosolic Ca2+ at a slower rate than does the SR,3 4 5 8 is primarily responsible for the decay of the Caf transient, whereas the SR is primarily responsible for the decay of the AP transient.3 5 8
We measured the differences in Caf and AP Ca2+ transients to determine the amount of Ca2+ stored in the SR, assess the fraction of this store that is released during excitation-contraction coupling, and determine the relative abilities of the SR and the Na+/Ca2+ exchanger to lower cytosolic free Ca2+. Caf and AP Ca2+ transients were measured in 36 failing human myocytes. The average results are reported in Table 2⇑, and a representative example is shown in Figure 4⇓. The most apparent differences between these transients were that the Caf transients had no tonic portion and were shorter in duration than AP transients, as evidenced by a significantly shorter (P<0.05) time to 95% decay in the Caf (1.76 seconds) versus the AP (2.17 seconds) transients. AP and Caf Ca2+ transients had similar peaks (see Table 2⇑), but the rate of rise of Caf transients was slower than those induced by APs (see Table 2⇑). All other features of AP and Caf transients were similar. The similar size of AP and Caf Ca2+ transients suggests that in failing human ventricular myocytes, the action potential induces a large fractional SR Ca2+ release. The fact that Caf transients, which occur in the absence of an action potential, do not have a tonic component suggests that Ca2+ influx during the action potential plateau is responsible for inducing this component of action potential–induced contractions. The similar early decay rates of Caf and AP Ca2+ transients also suggest that under the conditions of our experiments, forward-mode Na+/Ca2+ exchange and SR Ca2+ uptake can reduce cytosolic Ca2+ at similar rates.
Previous studies have shown that the peak of the Caf Ca2+ transients can be blunted by rapid Ca2+ extrusion by forward-mode Na+/Ca2+ exchange activity.29 This can lead to significant underestimation of the SR Ca2+ load and fractional SR Ca2+ release. This effect could be especially relevant in the present study, because Na+/Ca2+ exchanger activity is thought to be increased in failing human ventricular myocytes.20 The possibility that Ca2+ efflux via forward-mode Na+/Ca2+ exchange activity blunts the peak Ca2+ produced when SR release is induced by caffeine was tested by rapidly exposing myocytes to caffeine in an Na+- and Ca2+-free solution. This technique induces SR Ca2+ release and virtually eliminates Ca2+ transport by the sarcolemmal Na+/Ca2+ exchanger. Under these conditions (n=13), the peak of the Caf Ca2+ transient was 29% greater than the AP transient (Figure 5A⇓).
The decay of Caf Ca2+ transients is faster than that of the AP Ca2+ transient and is thought to be produced by forward-mode Na+/Ca2+ exchange activity.4 To further document the idea that the decay of the Caf Ca2+ transient results from forward-mode exchange activity, myocytes were exposed to caffeine in an Na+- and Ca2+-free solution for 10 seconds (Figure 5B⇑). Under these conditions, the decay of the Ca2+ transient was very slow. These results strongly support the idea that the decay of the Caf Ca2+ transient results from Ca2+ efflux via the Na+/Ca2+ exchanger.
A caffeine spritz causes depletion of SR Ca2+ stores, as evidenced by the fact that a second spritz causes no additional Ca2+ release. Therefore, the first action potential–induced contraction after exposure to caffeine should not have a component mediated by SR Ca2+ release. We induced action potentials after a caffeine spritz in 36 myocytes and found that only the tonic component of the Ca2+ transient was present (Figure 6⇓). With subsequent stimuli, the phasic component of the transient increased and reached a steady state within 20 beats. The tonic component of the Ca2+ transient decreased slightly in beat number and reached a steady state at the same time (Figure 6⇓). These experiments provide additional evidence that the phasic and tonic components of the indo 1 transient of failing human ventricular myocytes result from different processes, ie, SR Ca2+ release and Ca2+ influx, respectively.
Inhibiting the SR Ca2+-ATPase
Thapsigargin, an inhibitor of SR Ca2+-ATPase, eliminated the phasic component of the Ca2+ transient but had no significant effect on the tonic component (n=9 myocytes). Representative results are presented in Figure 7⇓, and average data are listed in Table 3⇓. Failure of a caffeine spritz to elicit a Ca2+ transient after exposure to thapsigargin verified SR Ca2+ depletion. The peak systolic indo 1 ratio of the AP transient was significantly (P<0.05) reduced after thapsigargin treatment. Thapsigargin also delayed the time-to-peak transient, time to 50%, and time to 95% decay; however, these data were not statistically significant. These findings provide further strong support for the idea that the phasic component of the indo 1 transient and contraction result from SR Ca2+ release and reuptake.
Inhibiting Reverse-Mode Na+/Ca2+ Exchange
One possible source of Ca2+ for the tonic component of contraction is Ca2+ influx via reverse-mode Na+/Ca2+ exchange. We tested this idea by measuring indo 1 transients before and after exposure to a putative reverse-mode-exchange inhibitor (No. 7943, Kanebo, Ltd).24 Application of No. 7943 eliminated the tonic component of the Ca2+ transient (n=9 myocytes). Representative raw data are presented in Figure 8⇓, and average data are listed in Table 4⇓. No. 7943 also significantly (P<0.05) reduced the peak of the Ca2+ transient (control, 0.47 versus No. 7943, 0.43) and abbreviated (P<0.05) the time to 95% decay (control, 2.15 seconds versus No. 7943, 1.36 seconds). These results suggest that Ca2+ enters the cell during the action potential through reverse-mode Na+/Ca2+ exchange, loads the SR, and directly elevates cytosolic Ca2+. Ca2+ influx via reverse-mode Na+/Ca2+ exchange activity during the action potential would antagonize SR Ca2+ uptake and contribute to the slow decay of the systolic Ca2+ transients of failing human ventricular myocytes.
Effect of β-Adrenergic Stimulation on the Ca2+ Transients
β-Adrenergic stimulation is known to increase SR Ca2+ uptake and increase SR Ca2+ loading in cardiac myocytes.30 To further explore the basis of the 2 components of the Ca2+ transient, failing human ventricular myocytes were exposed to isoproterenol, a nonspecific β-adrenergic agonist. Isoproterenol decreased the diastolic indo 1 ratio, increased (P<0.05) the peak of the phasic component of the Ca2+ transient, but had no significant effects on the tonic component of the Ca2+ transient (n=8 myocytes; see Figure 9⇓ and Table 5⇓). These experiments provide additional support that the phasic component of contraction in failing human myocyte results from SR Ca2+ release.
The objective of this study was to explore the cellular and molecular bases of the reduced magnitude and slow decay of the systolic Ca2+ transient of ventricular myocytes from patients with severe congestive heart failure. In a recent study of these myocytes with voltage-clamp techniques, we showed that contractions can result from both SR Ca2+ release and, when the duration of the depolarizing voltage step was sufficiently long, from direct elevation of cytosolic Ca2+, apparently via reverse-mode Na+/Ca2+ exchange.25 In the present experiments, we primarily studied Ca2+ transients induced by extracellular field stimulation so that the native intracellular milieu would be undisturbed (no cytoplasmic dialysis via a micropipette). Experiments were performed in this fashion because the intracellular environment might be at least partly responsible for the behavior exhibited by failing human myocytes.
The most important observations of the present research were the following: (1) Action potentials elicited at low rates of stimulation induced contractions and Ca2+ transients with 2 distinct components (phasic and tonic). The peak systolic indo 1 ratio of the phasic component was smaller than those that we have observed in normal and hypertrophied/failing feline ventricular myocytes, but the tonic component was larger.26 (2) The phasic component of contraction and the Ca2+ transient was eliminated by thapsigargin and depletion of SR Ca2+ stores with caffeine but was increased by β-adrenergic stimulation. These findings strongly support the idea that the phasic component of the Ca2+ transient results from SR Ca2+ release and subsequent reuptake. (3) The tonic component of the Ca2+ transient was most apparent at slow rates of stimulation when the action potential duration was longest and was insensitive to thapsigargin or depletion of SR Ca2+ stores with caffeine but was blocked by the putative Na+/Ca2+ exchanger inhibitor No. 7943, suggesting that it is a direct result of Ca2+ influx via reverse-mode Na+/Ca2+ exchange.
Phasic and Tonic Components of Contraction and the Indo 1 Transient
Contractions with both phasic and tonic components have been reported in many previous studies of cardiac myocytes.25 31 They are most often observed together when the duration of either the action potential or the voltage-clamp step is longer than the duration of the SR-mediated contraction. These conditions were present in our experiments because failing human ventricular myocytes have long-duration action potentials.28 In addition, we accentuated this feature by performing our experiments at low rates of stimulation.32 In large mammals, including humans,28 the action potential duration decreases as the beat frequency is increased. Therefore, as stimulation rate was increased, the tonic component of contraction became less obvious (Figure 3⇑).
Ca2+ transients with 2 distinct components (termed L1 and L2) have been observed in failing human myocardium in previous studies.33 However, the kinetics and pharmacological sensitivity of these transients are significantly different from those recorded with indo 1 in the present experiments. The reasons for these differences are not clear. However, the Ca2+ transients measured in voltage-clamped, failing human myocytes12 28 with fluorescent Ca2+ indicators are similar to those we report here.
It is also worth noting that the tonic component of the transient observed in our experiments was not due to secondary SR Ca2+ release resulting from cellular Ca2+ overload. If SR Ca2+ overload were responsible for the tonic component of the Ca2+ transient, then thapsigargin and caffeine spritzes would have abolished it, which they did not.
The weakened contractility of failing human myocytes is thought to result in large part from reduced release of Ca2+ from the SR.34 The small phasic components of the indo 1 transients observed in the present experiments are consistent with this hypothesis. The mechanisms that produce these small, SR-mediated Ca2+ transients need additional study under more controlled experimental conditions. Possible explanations include reduced SR Ca2+ loading resulting from slowed Ca2+ uptake.34 This could be a consequence of reduced expression of SR Ca2+ pumps, as shown in some15 16 but not all17 previous studies of failing human hearts. Another possibility is abnormal triggering of SR Ca2+ release from a normally Ca2+-loaded SR, as suggested by a recent study of failing rat ventricular myocytes.35 We began to explore these issues by inducing full SR Ca2+ release via rapid exposure to caffeine. We found that the caffeine-induced SR Ca2+ release was only slightly greater than the action potential–mediated release, even when caffeine-induced release was measured in the absence of Na+/Ca2+ exchanger activity (Na+- and Ca2+-free extracellular solutions). These observations suggest that abnormal SR Ca2+ loading may be more important than abnormal excitation-contraction coupling in producing reduced peak systolic Ca2+ in failing human ventricular myocytes, at least under our experimental conditions. Similar results have been recently published by another group.27
It is noteworthy that we were able to show that the phasic component of the Ca2+ transient (which is the major component) in our failing human myocytes was blocked by the SR Ca2+-ATPase inhibitor thapsigargin. A previous report on failing human myocytes could not demonstrate significant effects of thapsigargin on contraction.36 The reasons for this discrepancy are not clear at present but could be the result of the differences in cell isolation techniques used in these 2 studies. It is worth pointing out that the magnitude of contraction we report here and in another recent study of failing human ventricular myocytes37 is significantly greater than previously reported in either normal or failing human myocytes. We believe that the improved performance of the myocytes we are using is the result of the in vivo cardioprotective techniques we used (see Methods and Reference 37 ). Our results suggest that many of the myocytes used in previous studies may have been significantly damaged by long periods of warm ischemia and subsequent reperfusion injury during the isolation process.
Our experiments strongly support the idea that the tonic component of the Ca2+ transient and contraction in failing human ventricular myocytes results from Ca2+ influx during the action potential and that this influx occurs via reverse-mode Na+/Ca2+ exchange activity. This conclusion is in large part based on the fact that a putative reverse-mode exchange inhibitor24 abolishes this tonic component (Figure 8⇑). However, it is also supported by the fact that the inhibition of SR function with thapsigargin (Figure 7⇑) and depletion of SR Ca2+ stores with caffeine (Figure 6⇑) do not eliminate the tonic phase of contraction. The idea that the Na+/Ca2+ exchanger might make a larger-than-normal contribution to contraction in failing human ventricular myocytes has been proposed in previous studies in which increased expression of exchanger mRNA and protein was observed.20 In those studies it was hypothesized that the exchanger might make an increased contribution to relaxation in failing human myocytes. Perhaps the most important aspect of the present study is that our results suggest that the exchanger will contribute to relaxation only when decay of the SR-mediated component of the Ca2+ transient coincides with repolarization of the action potential to induce Ca2+ efflux via forward-mode exchange activity. Only under these conditions will the SR and the exchanger work in concert to lower cytosolic free Ca2+. Importantly, our results also suggest that as SR uptake begins to lower the free intracellular Ca2+ concentration during the plateau phase of the action potential, the Na+/Ca2+ exchanger may add Ca2+ to the cytoplasm via reverse-mode exchange activity to slow the fall in cytosolic free Ca2+. As hypothesized above, the critical issue in failing human myocytes appears to be whether the exchanger and the SR are acting in concert to lower cytosolic free Ca2+ or whether their respective actions are opposed. These ideas can only be adequately addressed by using more controlled experimental conditions than applied in the present study.
Our experiments in which SR Ca2+ release was induced with caffeine strongly support the idea that forward-mode Na+/Ca2+ exchange activity is capable of transporting Ca2+ from the failing human myocyte at a rate comparable to SR Ca2+ uptake. This conclusion is based on the fact that the decay of the Caf Ca2+ transients, which is mediated primarily by the Na+/Ca2+ exchanger, is almost identical to the rate of decay of the phasic component of the Ca2+ transient that is eliminated by the SR Ca2+-ATPase inhibitor thapsigargin. Our study did not measure the absolute levels of SR or exchanger function in failing myocytes. They do show that the relative balance of activities by these 2 Ca2+ transport systems is significantly different from what has been observed in normal myocytes in other species (wherein the Caf transient decays much more slowly than does SR-mediated decay; see References 3 and 83 8 ).
There are a number of important limitations to this study. The first is that the myocytes were derived from patients with different diseases and that these patients were taking different medications. Despite these facts, the results were uniformly observed. We have always been concerned about cell damage during myocyte isolation and its influence on the results of studies with single cardiac cells. However, we have developed an in vivo myocyte protection procedure that helps minimize these effects. There is also concern that myocyte behavior is heterogeneous in different regions of the heart. Because of the technically demanding nature of our experiments, we focused only on myocytes from the midmyocardial region of the left ventricular free wall. The idea that myocytes from other regions of the failing human heart have different contractile properties is always a possibility. In addition, we have had limited access to nonfailing human hearts. Therefore, we do not know whether the behavior we observed is limited to the failing human heart or is common to both normal and diseased human myocytes. However, our contention is that this issue has been thoroughly addressed in previous studies. Our key question is why do failing human cells behave the way they do?
The present experiments suggest that the phasic contraction of failing human ventricular myocytes primarily results from SR Ca2+ release and that the reduced magnitude of this contraction may result from abnormally low SR Ca2+ stores. We also show that when the action potential is sufficiently long, a tonic component of contraction results from reverse-mode Na+/Ca2+ exchange. Our most important conclusion is that the slow decay of the Ca2+ transient in failing human myocytes (and thus, slow relaxation) may be related to Ca2+ influx via reverse-mode Na+/Ca2+ exchange during at least the terminal phases of the action potential plateau. If these ideas are correct, then shortening the action potential duration in the failing human heart could augment relaxation and reduce diastolic defects.
This research was supported by grants from the National Institutes of Health, Bethesda, Md (HL39201 and HL61495 to S.R. Houser; HL 03560 and HL61494 to K.B. Margulies) and the Southeastern Pennsylvania Affiliate of the American Heart Association (to K. Dipla).
- Received June 15, 1998.
- Accepted November 23, 1998.
- © 1999 American Heart Association, Inc.
Barry W, Bridge JH. Intracellular calcium homeostasis in cardiac myocytes [Review]. Circulation. 1993;87:1806–1815.
Bassani JWM, Bassani RA, Bers DM. Relaxation in rabbit and rat cardiac cells: species-dependent differences in cellular mechanisms. J Physiol. 1994;476:591–608.
Bassani J, Yuan W, Bers DM. Fractional SR Ca2+ release is regulated by trigger Ca2+ and SR Ca2+ content in cardiac myocytes. Am J Physiol. 1995;268:C1313–C1319.
Negretti N, O’ Neil SC, Eisner DA. The relative contributions of different intracellular and sarcolemmal systems to relaxation in rat ventricular myocytes. Cardiovasc Res. 1993;27:1826–1830.
Bers D, Bassani JW, Bassani RA. Competition and redistribution among calcium transport systems in rabbit cardiac myocytes. Cardiovasc Res. 1993;27:1772–1777.
Bailey B, Houser SR. Sarcoplasmic reticulum-related changes in cytosolic calcium in pressure-overload-induced feline left ventricular hypertrophy. Am J Physiol. 1993;265:H2009–H2026.
Beuckelmann D, Erdmann E. Ca++-currents and intracellular Ca++-transients in single ventricular myocytes isolated from terminally failing human myocardium. Basic Res Cardiol. 1991;86(suppl 1):235–243.
Beuckelmann D, Nabauer M, Erdmann E. Intracellular calcium handling in isolated ventricular myocytes from patients with terminal heart failure. Circulation. 1992;84:1046–1055.
Meyer M, Schillinger W, Pieske B, Holubarsch C, Heilmann C, Posival H, Kuwajima G, Mikoshiba K, Just H, Hasenfuss G. Alterations of sarcoplasmic reticulum proteins in failing human dilated cardiomyopathy. Circulation. 1995;92:778–784.
Qi M, Shannon TR, Euler DE, Bers DM, Samarel A. Downregulation of sarcoplasmic reticulum Ca-ATPase during progression of left ventricular hypertrophy. Am Physiol Soc. 1997;275:H2416–H2423.
Schwinger R, Bohm M, Schmidt U, Karczewski P, Bavendiek U, Flesh M, Krause EG, Erdmann E. Unchanged protein levels of SERCA II and phospholamban but reduced Ca2+ uptake and Ca2+-ATPase activity of cardiac SR from dilated cardiomyopathy patients compared with patients with nonfailing hearts. Circulation. 1995;92:3220–3228.
Movsesian M, Bristow MR, Krall J. Ca2+ uptake by cardiac sarcoplasmic reticulum from patients with idiopathic dilated cardiomyopathy. Circ Res. 1989;65:1141–1144.
Studer R, Reinecke H, Bilger J, Eschenhagen T, Bohm M, Hasenfuss G, Just H, Holtz J, Drexler H. Gene expression of the cardiac Na+-Ca2+ exchanger in end-stage human heart failure. Circ Res. 1994;75:443–453.
Janczewski A, Lakatta EG. Thapsigargin inhibits Ca++ uptake, and Ca++ depletes sarcoplasmic reticulum in intact cardiac myocytes. Am J Physiol. 1993;265:H517–H522.
Iwamoto T, Watano T, Shigekawa M. A novel isothiourea derivative selectively inhibits the reverse mode of Na+/Ca2+ exchange in cells expressing NCX1. J Biol Chem. 1996;271:22391–22397.
Mattiello J, Margulies KB, Jeevanandam V, Houser SR. Contribution of reverse-mode sodium-calcium exchange to contractions in failing human left ventricular myocytes. Cardiovasc Res. 1998;37:424–431.
Sipido KR, Stankovicova T, Flameng W, Vanhaecke, Verdonck F. Frequency dependence of Ca2+ release from the sarcoplasmic reticulum in human ventricular myocytes from end-stage heart failure. Cardiovasc Res. 1998;37:478–488.
McCall E, Ginsburg KS, Bassani RA, Shannon TS, Qi M, Samarel AM, Bers DM. Ca2+ flux, contractility, and excitation-contraction coupling in hypertrophied rat ventricular myocytes. Am J Physiol. 1988;274:H1348–H1360.
Houser S, Bahinski A, Silver LH. Passive membrane properties of isolated feline ventricular myocytes. Am J Physiol. 1985;248:H622–H630.
Gwathmey J, Copelas L, MacKinnon R, Schoen FJ, Feldman MD, Grossman W, Morgan JP. Abnormal intracellular calcium handling in myocardium from patients with end-stage heart failure. Circ Res. 1987;61:70–76.
Beuckelmann D. Contributions of Ca++-influx via the L-type Ca++-current and Ca++-release from the sarcoplasmic reticulum to [Ca++]i transients in human myocytes. Basic Res Cardiol. 1997;92(suppl 1):105–110.
Davia K, Davies CH, Harding SE. Effects of inhibition of sarcoplasmic reticulum calcium uptake on contraction in myocytes isolated from failing human ventricle. Cardiovasc Res. 1997;33:88–97.
Dipla K, Mattiello JA, Jeevanadam V, Houser SR, Margulies KB. Myocyte recovery after mechanical circulatory support in humans with end-stage heart failure. Circulation. 1998;97:2316–2322.