Resumen de Role of the sarcoplasmic reticulum in the regulation of contraction in isolated cardiac myocytes from the teleost fish rainbow trout (oncorhynchus mykiss)
In lower vertebrates, activation of contraction is considered to depend strongly on transsarcolemmal calcium fluxes. In the amphibian heart, it has been described that L-type calcium channel (ICa) account for the major part of the calcium activator of the contraction (54, 55, 121), and it is often assumed that this is true for other lower vertebrates. This assumption is based on two main observations. First, cardiac myocytes from the teleost heart have similar dimensions to those from the amphibian heart, i.e., cell diameters between 2 and 8 µm, and ultrastructural studies have shown that the myocytes are thin long cells without transverse tubules (T tubules) and a relatively poorly developed sarcoplasmic reticulum (SR) (130, 163, 165). Thus diffusion in myocytes from lower vertebrate hearts may not limit a direct activation of the myofilaments by Ca2+ fluxes across the sarcolemma. Secondly, inhibition of the sarcoplasmic reticulum (SR) function with ryanodine has a small or no effect on cardiac contraction is situ (103) or in multicellular preparations at physiological temperatures and heart rates (39, 40, 71, 75, 120), suggesting that the SR does not play a dominant role in the activation of contraction under these conditions.
Contrary to the general assumption that contraction in lower vertebrates depends largely on transsarcolemmal Ca2+ fluxes, previous results have shown that the SR inhibition with ryanodine do inhibit contraction at room temperature (71) and results from our laboratory have shown that calcium influx through ICa is still only one quarter of the total Ca2+ contributing to the activation of contraction (87). Thus, it seems that under our experimental conditions the trout represents a first exception to the general model of excitation-contraction coupling (E-C coupling) in the lower vertebrate heart and the purpose of this thesis was to examine quantitatively the contribution of the SR to the regulation of contraction in isolated cardiac myocytes from the teleost fish Rainbow trout (Oncorhynchus mykiss). The thesis is structured in seven chapters that investigate the contribution of the SR to the activation of contraction (Chapters 1-3); the capacity and contribution of the SR to remove calcium from the cytosol during relaxation (Chapter 4); modulation of the SR function by temperature and ß-adrenergic stimulation (Chapters 5&6); and Chapter 7 integrates the methodology and findins of chapters 1-6 to characterize calcium handling in the zebrafish heart, an emerging vertebrate model to study genetic variation and defects in the cardiovascular system.
In the first chapter of this thesis, we quantified the calcium release from the SR elicited by as the time integral of the caffeine sensitive tail current (ICICR) elicited by short depolarization pulses before (3-30ms). The amount of Ca2+ carried by ICICR reached a maximum with a depolarization pulse duration of 6-8 ms. Cells with prominent ICICR, had a larger SR Ca2+ content and required a shorter stimulation pulse to elicit contraction. Furthermore, the charge carried by ICICR was up to 10 times larger than the amount of calcium carried by the ICa. Moreover, calcium depletion of the SR by CAF exposure slowed the inactivation of ICa suggesting that there is a cross-talk between calcium release from the SR and ICa inactivation. The maximal cell shortening induced by a short depolarization pulse was 40 ± 6% of the corresponding cell shortening elicited by 200-ms depolarization, and the amount of Ca2+ released from the SR during the short depolarization pulse amounted to 48 ± 10% of the total Ca2+ required to activate contraction.
Thus, contrary to the general assumption in lower vertebrates, the results in Chapter 1 shows that SR Ca2+ release in trout myocytes may account for up to 50% of the total Ca2+ transient at room temperature.
In the second chapter of the thesis we examined how the SR Ca2+ release depends on the Ca2+ trigger source. In that work, we used simultaneous recordings of electrophysiological measurements and intracellular Ca2+ transients. Cells were stimulated with consecutive short depolarization pulses (2-20 ms) to elicit calcium-induced calcium release (CICR) from the SR. In accordance with results from Chapter 1, a maximum SR Ca2+ release was reached with a 10-ms depolarization. To determine how calcium release from the SR depended on the membrane potential, myocytes were depolarized for 10 ms to different membrane potentials and the resulting SR Ca2+ release, intracellular Ca2+ transient and cell shortening were recorded. All three parameters showed a biphasic response when the membrane potential during the 10-ms depolarization was increased. Between -40 and +30 mV the relationship was bell-shaped with a peak near 0 mV, after which calcium release declined, coinciding with the voltage dependency of ICa. However at potentials above +40 mV, calcium release increased monophasically with membrane potential. In the presence of 100 µM CdCl2 (that abolished ICa), the SR Ca2+ release elicited at +10 mV was reduced to 40 ± 17% of control, but had no effect between +70 and +150 mV. Furthermore, the SR Ca2+ release was proportional to the membrane potential in the range -50 to +150 mV in the presence of 100 µM CdCl2. Increasing the intracellular Na+ concentration ([Na+]) from 10 to 16 mM enhanced SR Ca2+ release but reduced cell shortening at all membrane potentials examined, whereas the SR Ca2+ content was not significantly different. In the absence of tetrodotoxin (TTX), SR Ca2+ release was potentiated with 16 mM but not 10 mM pipette [Na+]. Comparison of the total sarcolemmal Ca2+ entry and the Ca2+ released from the SR gave a gain factor of 18.6 ± 7.7. Nifedipine (Nif) at 10 µM inhibited ICa and reduced the time integral of the tail current by 61%. The gain of the Nif-sensitive SR Ca2+ release was 16.0 ± 4.7. A 2-ms depolarization still elicited a contraction in the presence of Nif that was abolished by addition of 10 mM NiCl2. The gain of the Nif-insensitive but NiCl2-sensitive SR Ca2+ release was 14.8 ± 7.1.
Thus the results of Chapter 2 shows that both reverse-mode Na+/Ca2+ exchange (NCX) and ICa can elicit Ca2+ release from the SR, but that ICa is more efficient than reverse-mode NCX in activating contraction. This difference may be due to extrusion of a larger fraction of the Ca2+ released from the SR by reverse-mode NCX rather than a smaller gain for NCX-induced Ca2+ release.
After studying the role of the SR in the E-C coupling in trout cardiomyocytes, we attempted in Chapter 3 to measure the sarcolemmal Ca2+ transport by the Na+/Ca2+ exchanger (NCX) and determine how it modulates calcium handling by the SR. Using a standard 200-ms depolarization to 0 mV every 8 s from a holding potential of -80 mV, we could measure both the ICa and INCX currents amplitudes using a pharmacological dissection. With this approach, 5 µM nifedipine abolished ICa and reduced Itail whereas the current at the end of the depolarization pulse was unaffected. Moreover, nifedipine reduced cell shortening by about 35 %. On the other hand, addition of 10 mM NiCl2 together with Nif strongly reduced Itail and abolished both the current at the end of the depolarization pulse and the contraction. Comparison of the sarcolemmal Ca2+ entry and extrusion gave similar results for the amount of Ca2+ carried by ICa (Nif-sensitive current) and INCX (Nif-insensitive but NiCl2-sensitive current) when the cell was depolarized from -80 to 0 mV. Furthermore, the sum of the amount of Ca2+ entering through ICa and INCX was not significantly different from the total amount of calcium extruded by the NCX during Itail (measured as the difference between Itail in control and with Nif+NiCl2). With this pharmacological dissection ICa and INCX could account for 41 and 59 % of the sarcolemmal Ca2+ influx respectively (at 0 mV and with 16 mM Na+ in the patch pipette). Because NCX is an electrogenic pump, we expected that the relative contribution of the NCX to the activation of contraction depended on both the membrane potential and the [Na+] in the patch pipette. To test this, we first used an intracellular perfusion system that allows changes in the pipette [Na+] during an experiment, and found that the amount of Ca2+ carried by Itail was reduced to 44 ± 5% of the initial after changing the pipette solution from 16 mM Na+ to a nominally Na+-free solution. Similarly, the caffeine releasable SR Ca2+ content was reduced to 57 ± 9% of the initial value. Secondly, we tested the effect of different membrane potentials, and found that ICa was the same at -10 or +10 mV whereas Ca2+ extrusion from the cell and the SR Ca2+ content at -10 mV were 65 ± 7% and 80 ± 4% of that at +10 mV. The relative amount of Ca2+ extruded by the NCX (about 55%) and taken up by the SR (about 45%) was, however, similar with depolarizations to -10 and +10 mV.
From the results in Chapter 3, we conclude that modulation of the NCX activity critically determines Ca2+ entry and cell shortening in trout atrial myocytes. This is due to both an alteration of the transsarcolemmal Ca2+ transport and a modulation of the SR Ca2+ content.
In order to determine the capacity of the SR to take up calcium we used in Chapter 4 permeabilized trout ventricular myocytes. First, we titratted the SR Ca2+ pumps with thapsigargin which gave a pump site density of 454 nmol/mg cell protein. Secondly, by adding 10 µM Ca2+ to a cell suspension of permeabilized myocytes at 20ºC we induced SR Ca2+ uptake. This gave a maximal uptake rate (Vmax) of 4.4 ± 0.8 nmol Ca2+¿mg cell protein-1¿min-1, an nH of 2.5 ± 0.2, and a K0.5 of 0.87 ± 0.11 µM at 20ºC. The uptake rate at 1µM free Ca2+ was reduced by 50% at 10ºC and by 63% at 5ºC, giving an average Q10 of 1.6. Contrary to the mammalian heart, passive Ca2+ leak from the SR decreased when temperature was lowered to 10 and 5ºC. In intact single myocytes subjected to voltage clamp, exposure to 10 mM CAF elicited a cell contracture and an inward ionic current. This inward current was due to Ca2+ extrusion from the cytosol by the Na+/Ca2+ exchanger (NCX), and the time integral of the exchange current (INCX) gave a steady-state SR Ca2+ content of 22.5 ± 2.8 amol Ca2+/pF or 750 µmoles/l non-mitochondrial cell volume. Using specific SR Ca2+ loading protocols, we estimated the maximal SR Ca2+ content as 52.0 ± 5.9 amol Ca2+/pF and the maximal Ca2+ uptake rate as 12.2 ± 1.2 amol Ca2+ pF-1 s-1 or 417 µM/s, which is sufficient to remove the entire calcium transient in trout cardiomyocytes at physiological beating rates.
Based on these findings, we conclude that the trout SR has the capacity to participate in both the activation and relaxation of contraction at physiological heart rates.
In Chapter 5, experimental protocols employed in chapters 1, 2 and 4 were used to investigate how the experimental temperature affects SR function in trout cardiomyocytes. The results showed that the relationship between Vm and SR Ca2+ uptake was not significantly changed by lowering the experimental temperature from 21 to 7ºC, and that the relationship between the time integral of the tail current and the time integral of the CAF-induced NCX current was similar at the two experimental temperatures with a pooled Vmax of 66 amol/pF and a K0.5 of 4.3 amol/pF. Quantification of the SR Ca2+ release elicited by 10-ms depolarization to different Vm showed a bell-shaped relationship between SR Ca2+ release and membrane potential between -50 and +50 mV with a peak near +10 mV. Lowering of the experimental temperature did not affect this relationship significantly, giving an SR Ca2+ release at +10 mV of 1.71 and 1.54 amol/pF at 21 and 7ºC, respectively. This SR Ca2+ release corresponds to 42 and 38% of the total Ca2+ required to activate a contraction at 21 and 7ºC, respectively. Quantification of the caffeine releasable SR Ca2+ content revealed that the SR also accumulates and liberates Ca2+ equally to 7 and 21ºC. Furthermore, clearance of the SR Ca2+ content slowed down inactivation of ICa at both experimental temperatures (from 15.0 ± 2.0 ms on the first depolarization to 10.4 ± 1.9 ms on the 20th depolarization at 21ºC, and from 73 ± 24 and 38 ± 15 ms at 7ºC), suggesting that calcium released from the SR speeds up ICa inactivation at both temperatures.
Thus we conclude in Chapter 5 that the SR has the capacity to remove the entire Ca2+ transient at physiologically relevant stimulation frequencies at both 21 and 7ºC, although it is estimated that ~40% of the total Ca2+ transient is liberated from and reuptaken by the SR with continuous stimulation at 0.5 Hz independently of the experimental temperature.
As ß-adrenergic stimulation is a key important modulator of cardiac contraction and rhythm in the vertebrate heart, in Chapter 6 we addressed how ß-adrenergic stimulation modulates calcium handling by the SR. For this purpose, simultaneous measurements of intracellular [Ca2+] ([Ca2+]i) and whole membrane current were performed during long depolarizations of the cardiac myocytes, preeceded and followed by rapid CAF applications at -80 mV. This allowed us to establish the relationship between SR Ca2+ loading, membrane potential, and [Ca2+]i under control and phosphorylating conditions. In the absence of the ß-adrenergic agonist isoproterenol (ISO), maximal SR Ca2+ load was 597 µmol¿l-1 and loading was half-maximal at -12 mV. In the presence of ISO, maximal SR Ca2+ loading was not significantly affected but the potential for half-maximal loading was shifted by -26 mV. In order to study the effect of ISO on the relationship between the SR Ca2+ uptake and the [Ca2+]i, we measured the [Ca2+]i during depolarization to different membrane potentials and during caffeine applications in the absence and the presence of 1µM ISO. In control conditions, there was no significant difference between the average Ca2+ transient during depolarization and the peak Ca2+ transient elicited by CAF after depolarization. However, the CAF induced Ca2+ transient was significantly bigger than the average Ca2+ transient during depolarization in the presence of ISO. Therefore, ISO directly stimulates SR Ca2+ uptake. The maximal SR Ca2+ uptake rate (Vmax) was 418 µmol¿l-1¿s-1 in control conditions. ISO did not affect Vmax, but significantly lowered the average free Ca2+ transient during the depolarization and shifted the K0.5 for the relationship between SR Ca2+ uptake and [Ca2+]i from 1.27 in control to 0.8 µmol¿l-1 with ISO. Following repetitive 200-ms depolarizatios, ISO incresed ICa amplitude by 91 ± 29% and the peak Ca2+ transient by 41 ± 10%., and ISO decreased the half life of the Ca2+ transient significantly from 151 ± 12 in control to 111 ± 6 ms with ISO. Using the relationship between [Ca2+]i and SR Ca2+ uptake to calculate the total SR Ca2+ uptake during a Ca2+ transient elicited by a 200-ms depolarization, a significant increase in the SR Ca2+ uptake from 37 ± 6 µmol¿l-1 in control to 68 ± 4 µmol¿l-1 with ISO was seen. When normalizing the SR Ca2+ uptake to the total Ca2+ transport, the contribution of the SR was not significantly different in the absence (35 ± 6%) or the presence of ISO (41 ± 4%).
We conclude in Chapter 6 that although ISO has a stimulatory effect on the SR Ca2+ pump that may contribute to the faster decay of the Ca2+ transient, the relative contribution of the SR to the Ca2+ removal from the cytosol during relaxation is not alterd by ISO in trout cardiac myocytes.
The zebrafish has been emerged in cardiovascular research as a new model for functional studies of genetic mutations (31, 32, 38, 42, 106). However, little information is available on E-C coupling in the zebrafish heart and the purpose of Chapter 7 was to address this issue. To achieve this goal, adult zebrafissh cardiomyocytes were isolated by enzymatic perfusion of the cannulated ventricle, and subjected to amphotericin-perforated patch-clamp technique, confocal calcium imaging, and/or measurements of cell shortening. Simultaneous recordings of the voltage dependence of the ICa amplitude and cell shortening showed a high peak ICa density (12 pA/pF) with a typical bell-shaped current-voltage relationship (I-V relationship) for ICa, with a maximum at +10 mV. In contrast, calcium transients and cell shortening showed a monophasic increase with membrane depolarization, that reached a plateau at membrane potentials above +20 mV; suggesting that ICa is the major source of Ca2+ activator of contraction at voltages below +10 mV, whereas the contribution of reverse-mode NCX becomes more increasingly reliable at membrane potentials above +10 mV. Comparison of the recovery of ICa from acute and steady-state inactivation showed that reduction of ICa upon elevation of the stimulation frequency (from 0.5 to 3 Hz) is primarily due to calcium dependent ICa inactivation.
In conclusion, we demonstrate in Chapter 7 that a large yield of healthy atrial and ventricular myocytes can be obtained by enzymatic perfusion of the cannulated zebrafish heart. Moreover, zebrafish ventricular myocytes differ from that of large mammals by having larger ICa density and a monophasically increasing contraction-voltage relationship, suggesting that caution should be taken upon extrapolation of the functional impact of mutations on calcium handling and contraction in zebrafish cardiomyocytes.
Together the research undertaken in this thesis document that calcium sequestration by the SR participates actively in calcium handling on a beat-to-beat basis in isolated teleost cardiomyocytes. Specifically, we show that calcium release can be triggered by both L-type calcium current and reverse-mode Na+/Ca2+ exchange current. Moreover, quantification of the amount of calcium released from the SR during depolarization and re-sequestrated during diastole show that calcium released from the SR accounts for 40-50% of the total calcium transient. This is true over the physiological temperature range (7-21º) that the trout heart may experience, even though lowering of the temperature slows down the kinetics of both SR calcium uptake and release. Similarly, ß-adrenergic stimulation speeds up calcium sequestration, but it does not modify the relative contribution of the SR to the activation and relaxation of contraction.
These findings emphasizes the necessity of revising the role of the SR in the excitation-contraction coupling in the lower vertebrate heart, and proposes isolated teleost myocytes as an useful new model to study regulation of intracellular calcium handling under experimentally demanding conditions such as the application of mechanical stress. The quantitative studies of excitation-contraction coupling in teleost myocytes performed in this thesis are also important to emerging new fields that focus on the unique properties of the teleost heart such as its ability to regenerate itself, and the emergence of the zebrafish as a new model to study genetic variation and mutations in cardiovascular disease.
