Intracellular Na+-concentration, [Na+]i modulates excitation-contraction coupling of cardiac myocytes via the Na+/Ca2+ exchanger (NCX). In cardiomyocytes from rainbow trout (Oncorhyncus mykiss), whole cell patch-clamp studies have shown that Ca2+ influx via reverse-mode NCX contributes significantly to contraction when [Na+]i is 16 mM but not 10 mM. However, physiological [Na+]i has never been measured. We recorded [Na+]i using the fluorescent indicator sodium-binding benzofuran isophthalate in freshly isolated atrial and ventricular myocytes from rainbow trout. We examined [Na+]i at rest and during increases in contraction frequency across three temperatures that span those trout experience in nature (7, 14, and 21°C). Surprisingly, we found that [Na+]i was not different between atrial and ventricular cells. Furthermore, acute temperature changes did not affect [Na+]i in resting cells. Thus, we report a resting in vivo [Na+]i of 13.4 mM for rainbow trout cardiomyocytes. [Na+]i increased from rest with increases in contraction frequency by 3.2, 4.7, and 6.5% at 0.2, 0.5, and 0.8 Hz, respectively. This corresponds to an increase of 0.4, 0.6, and 0.9 mM at 0.2, 0.5, and 0.8 Hz, respectively. Acute temperature change did not significantly affect the contraction-induced increase in [Na+]i. Our results provide the first measurement of [Na+]i in rainbow trout cardiomyocytes. This surprisingly high [Na+]i is likely to result in physiologically significant Ca2+ influx via reverse-mode NCX during excitation-contraction coupling. We calculate that this Ca2+-source will decrease with the action potential duration as temperature and contraction frequency increases.
- atrial and ventricular myocytes
- excitation-contraction coupling
- intracellular sodium
- reverse-mode sodium-calcium exchange
in cardiomyocytes, the intracellular Na+ concentration ([Na+]i) is an important modulator of excitation-contraction coupling. It regulates Ca2+ efflux and influx via the Na+/Ca2+ exchanger (NCX), which passively exchanges three Na+ for one Ca2+ across the sarcolemma. In terms of Ca2+ efflux, the NCX is the main mechanism for Ca2+ extrusion from the cell during relaxation and competes with the sarcoplasmic reticulum (SR) Ca2+ ATPase for the removal of Ca2+ from the cytosol. An increase in [Na+]i lowers the driving force for NCX Ca2+ extrusion and thus increases the SR Ca2+ load. Large increases in [Na+]i impairs diastolic relaxation. In terms of Ca2+ influx, the NCX may switch to a reverse mode and transport Ca2+ into the cell in the initial part of the action potential (AP) when the membrane potential (Em) overshoots the reversal potential for NCX (ENCX). An increase in [Na+]i lowers ENCX and increases the driving force for Ca2+ import so that reverse-mode NCX Ca2+ influx may supplement Ca2+ influx via L-type Ca2+ channels to trigger SR Ca2+ release and contribute to the Ca2+ transient.
In mammalian cardiomyocytes, the NCX mainly functions in the forward-mode extruding Ca2+. Ca2+ influx via reverse-mode NCX is negligible (4, 37). Thus, although the NCX can potentially function as a pathway for Ca2+ influx, this does not occur in vivo. In contrast, in some fish species reverse-mode NCX contributes significantly to the Ca2+ transient. For example, in cardiomyocytes from crucian carp (Carassius carassius) and burbot (Lota lota), it has been estimated to account for ∼50 and 75% of the Ca2+ influx, respectively (24, 33). However, the importance of reverse-mode NCX Ca2+ influx is highly dependent on [Na+]i. In whole cell patch-clamp of rainbow trout atrial myocytes, it is able to trigger Ca2+-induced Ca2+ release from the SR and full contraction at a [Na+] pipette of 16 mM but not 10 mM (19). Therefore, the aim of the present study was to measure the in vivo [Na+]i to understand the physiological role of reverse-mode NCX Ca2+ influx in rainbow trout cardiomyocytes.
In mammalian hearts, [Na+]i varies between species and tissues and with status of health and level of activity. Between species, [Na+]i is highest in rat followed by ferret, then guinea pig, and then rabbit ventricular myocytes (8, 15, 22). Between tissues, [Na+]i is higher in atrial than ventricular cells (35). [Na+]i increases in heart failure as a result of increased diastolic Na+ influx (9, 22). Furthermore, [Na+]i increases with contraction frequency because of increased Na+ influx via Na+-channels and via the NCX during relaxation and Ca2+ extrusion (2, 5, 17), and an increase in [Na+]i is also induced by stretch and acidosis (16, 31). Thus, [Na+]i is highly dynamic and regulated. The present study focused on the heart from healthy rainbow trout in which [Na+]i may vary between atrial and ventricular myocytes and with contraction frequency. Furthermore, because rainbow trout are ectothermic and experience acute changes in body temperature, we tested whether this also modifies [Na+]i and the contraction-induced increase in [Na+]i. Indeed, in resting mammalian cardiomyocytes, cooling induces a significant increase in [Na+]i (29). Therefore, using the fluorescent Na+ indicator sodium-binding benzofuran isophthalate (SBFI), we measured [Na+]i in freshly isolated atrial and ventricular cardiomyocytes at 7, 14, and 21°C. These temperatures were chosen because they are physiologically relevant and allow us to compare with several previous studies (25–27). To assess the contraction-induced increase in [Na+]i, we recorded the fractional change in fluorescence in atrial and ventricular myocytes contracting from 0.2 to 0.8 Hz at 7, 14, and 21°C.
MATERIALS AND METHODS
Rainbow trout (Oncorhyncus mykiss) were obtained from Chirk Fish Farm (Wrexham, UK). They were kept in freshwater tanks at 13 ± 1°C with a 12:12-h light-dark cycle and fed with commercial trout pellets. All procedures adhere to the guidelines from the United Kingdom Home Office Animals Scientific Procedures Act of 1986 as approved by the Animals Procedures Committee. The fish was stunned by a blow to the head and the spinal cord was severed. The heart was excised and immediately transferred to a nominally Ca2+-free isolation solution.
Isolation of cardiomyocytes.
Cardiomyocytes were isolated according to previously described methods (32). Briefly, the heart was perfused in a retrograde manner first with isolation solution for 8–10 min and then with enzymatic solution for 14–18 min. The atrium and ventricle were separated and transferred to isolation solution. The tissue was cut into small pieces that were triturated with a Pasteur pipette. The cells were kept at 4°C in isolation solution until use.
Loading of cells.
The AM-ester of SBFI (Molecular Probes) was dissolved in DMSO to a concentration of 5 mM. Just before loading of the cells, this stock solution of SBFI was mixed with an equal volume of 25% wt/vol pluronic-127 in DMSO to facilitate loading. Cardiomyocytes were incubated with a final concentration of 10 μM SBFI for at least 90 min at 4°C. After this they were washed by dilution and left for at least 30 min for complete deesterification before use.
Cells were mounted in an RC-24 fast exchange bath chamber (Warner Instruments) on the stage of a Nikon epifluorescence microscope. They were allowed to settle before starting perfusion of the bath with external Ringer solution. Light from a xenon lamp (model XPS-100; Nikon) was passed through a monochromator (Cairn Research, Faversham, UK) that alternately excited the cells at 340 and 380 nm through a Fluor 40×/1.30 oil objective (Nikon). Fluorescence emitted from the cells was passed back via the objective through a band pass filter of 510 ± 10 nm to the photomultiplier (model Thorn EMI 9124B; Thales, Surrey, UK). Data were acquired by the computer program pClamp 9.2 (Axon Instruments, Molecular Devices). In parallel with the fluorescence recordings, the cells were illuminated with light passed through a far-red filter that allowed passage of light >800 nm. This light was captured by a high resolution charge-coupled device camera (model WAT-902B; Watec) to monitor cells during recording. Preliminary experiments (not shown) showed that this additional light did not influence the fluorescence signal. The bath was perfused by a six-channel perfusion system run by gravity. For temperature control, the perfusion solution was directed through an SC-20 inline cooler (Warner Instruments) immediately before flowing into the bath. Contraction of cells was provoked by field stimulation (20–80 V square pulses; duration 10 ms).
Isolation solution consisted of (in mM): 100 NaCl, 10 KCl, 1.2 KH2PO4, 4 MgSO4, 50 taurine, 20 glucose, and 10 HEPES. The pH was adjusted to 6.9 with KOH. Enzymatic solution consisted of isolation solution plus 0.5 mg/ml trypsin, 0.75 mg/ml collagenase 1A, and 0.75 mg/ml BSA. External Ringer solution consisted of (in mM): 150 NaCl, 5.4 KCl, 1.5 MgSO4, 0.4 NaH2PO4, 2 CaCl2, 10 glucose, and 10 HEPES. The pH was adjusted to 7.7 with KOH. For in vivo calibration of SBFI, the cells were superfused with a solution free from divalent ions containing (in mM): 140 NaCl/KCl, 1 EGTA, 10 glucose, 10 HEPES, and 10 μg/ml gramicidin D. The latter is an ionophore that allows equilibration of extracellular and intracellular [Na+]. The proportion of Na+ ions in solution was varied by mixing two stock solutions containing 140 mM NaCl and 140 mM KCl, respectively.
Data were analyzed in Clampfit 9.0 (Axon Instruments, Molecular Devices) and are presented as means ± SE. For resting [Na+]i, statistical significance of the effect of tissue (ventricular vs. atrial cells) and temperature (7, 14 and 21°C) were evaluated by a two-way ANOVA. For the fractional increase in 340/380 ratio during contraction, the results from atrial and ventricular cells were pooled, and statistical significance of the effect of temperature (7, 14, and 21°C) and frequency (0.2, 0.5, and 0.8 Hz) was evaluated by a two-way ANOVA followed by a Tukey posttest analysis. P < 0.05 was considered statistically significant.
In vivo spectrum of SBFI.
The excitation spectrum of SBFI exhibits a shift from in vitro to in vivo conditions (10–12). The in vivo spectrum is not likely to vary between species, because the shift is mainly due to protein binding (3). Indeed, the in vivo spectrum of SBFI in rainbow trout cardiomyocytes (Fig. 1) is similar to that in rat cardiomyocytes (10) and shows that when recording at 510 ± 10 nm, the largest Na+-dependent shift occurs at 380 nm, whereas 340 nm is the isosbestic point at which there is no Na+-dependent shift. This confirmed that our experimental settings were correct.
Mitochondrial autofluorescence may interfere with the signal from SBFI because light with a wavelength of 340 nm excites NADH. The emission spectrum of NADH peaks near 460 nm but may also emit some light above 500 nm, which could affect measurements of [Na+]i (12), especially in rainbow trout cardiomyocytes, which have a larger fractional mitochondrial volume (∼45%) (32) than rat cardiomyocytes (30–35%) (4). We tested for this in resting trout cardiomyocytes and show that autofluorescence does not interfere with the signal from SBFI (Fig. 1). To exclude artifacts from autofluorescence in contracting cells, only cells in which fluorescence at 340 nm remained constant were taken into account.
In vivo calibrations of SBFI at different temperatures.
SBFI was calibrated in vivo by superfusing the cell with a series of solutions containing the ionophore gramicidin D (10 μg/ml) and varying Na+ concentrations. In many studies, an inhibitor of the Na+-K+-ATPase, such as strophanthidin or ouabain, is included in the calibration solution to facilitate the equilibration between extracellular and intracellular [Na+]. However, many cells died during calibration in the presence of 10 μM ouabain as has been used for isolated toad cardiomyocytes (23). We are uncertain whether this was due to Ca2+ overload. As it should not affect the steady-state recordings (12), we omitted ouabain from our calibration solutions. A representative example of a calibration is shown in Fig. 2.
Because the affinity of ion indicators may change with pH and temperature [e.g., the Kd of fura-2 for Ca2+ decreases with increasing temperature (21)] we performed in vivo calibrations of SBFI at 7, 14, and 21°C to determine the Kd for Na+ by a Hanes plot of the background-subtracted signal. Fig. 3 shows representative examples of a Hanes plot at each temperature. Table 1 shows the average Kd calculated from a Hanes plot for each cell. The SBFI Kd obtained in trout cardiomyocytes at 21°C is the same (29.3 mM) as that found by Donoso et al. (12) in rat cardiomyocytes at 27°C. We found that the Kd of SBFI tended to decrease with increasing temperature, but this was not statistically resolvable (Table 1). However, a potential effect of temperature on the Kd may be concealed within the high standard errors associated with SBFI recordings. Therefore, we did not use the mean Kd for each temperature in our calculations. Rather, each recording of [Na+]i in a resting cell was followed by a full calibration, and the resting [Na+]i was calculated from the cell-specific calibration curve.
[Na+]i in resting cardiomyocytes.
Resting [Na+]i was recorded in atrial and ventricular cardiomyocytes at 7, 14, and 21°C. The 340/380 ratio from SBFI-fluorescence as a function of [Na+] can be fit by a hyperbolic curve when [Na+] ranges from 0 to 140 mM (11) or by a straight line when [Na+] ranges from 0 to 16 mM (17). In the present experiments, the data was best fit by a linear equation to the calibration points at 0, 7, and 14 mM. The results are shown in Table 2. Interestingly, there was no significant difference in [Na+]i between atrial and ventricular cardiomyocytes. Therefore, we pooled the values from both tissues, and these results are shown in the last column for each temperature. Furthermore, there was no significant effect of temperature on resting [Na+]i in either cell type. Therefore, we pooled the values from all temperatures, and these results are shown in the last row for each tissue. Our combined data reveal a resting [Na+]i of 13.4 ± 0.7 mM (n = 43) for trout cardiomyocytes.
[Na+]i in contracting cardiomyocytes.
Contractile activity increases Na+ influx via Na+ channels during excitation and via NCX during Ca2+-extrusion (5). As a consequence, [Na+]i will increase and stimulate the Na+-K+-ATPase until the increased influx is balanced by an increased efflux. Indeed, our results showed that the 340/380 ratio increased with contraction frequency. Fig. 4 shows the recording from a single cell; the ratio increases when the cell is stimulated to contract at increasing frequencies and returns to baseline when stimulation is stopped and the cell is allowed to rest. This increase in [Na+]i with contraction frequency was not affected by temperature (Table 3). Therefore, we pooled the results from all temperatures and this is shown in Fig. 5. The 340/380 ratio increased significantly (P < 0.001) in a frequency-dependent manner by 3.2, 4.7, and 6.5% at 0.2, 0.5, and 0.8 Hz, respectively (Fig. 5). Assuming a resting [Na+]i of 13.4 mM, the fractional increase corresponds to a total increase of 0.4, 0.6, and 0.9 mM at 0.2, 0.5, and 0.8 Hz, respectively. This increase in [Na+]i with frequency is relatively modest. In comparison, [Na+]i increases 3.7 mM from rest to 0.5 Hz in rabbit cardiomyocytes (7). However, this is in agreement with the fact that the frequency-dependent increase in [Na+]i is smaller in cells showing a frequency-dependent reduction in force (as it is the case in trout cardiomyocytes) (14).
The present study has three main findings. First, [Na+]i is higher in cardiomyocytes from rainbow trout than any healthy mammalian species. Second, in contrast to mammalian hearts, [Na+]i does not differ between atrial and ventricular cardiomyocytes either at rest or during stimulation. Third, temperature has no significant effect on either resting [Na+]i or the contraction-induced increase in [Na+]i. These findings have important implications for our understanding of the role of NCX in rainbow trout cardiac excitation-contraction coupling, and how temperature and contraction frequency affect cardiac performance.
This is the first study on [Na+]i in intact rainbow trout cardiomyocytes. We found a [Na+]i of 13.4 mM. This is higher than that reported for any mammalian species. For comparison, [Na+]i is 4.5 mM in rabbit (8), 6.5 mM in guinea pig (15), 9 mM in ferret (22), and 11 mM in rat ventricular myocytes (8). Interestingly, when combined with the relatively long AP in trout cardiomyocytes (34) such a high [Na+]i may allow for significant Ca2+ influx via reverse-mode NCX when the cell is depolarized during excitation-contraction coupling. Thus, in contrast to healthy mammalian cardiomyocytes, reverse-mode NCX Ca2+ influx may contribute significantly to the Ca2+ transient in rainbow trout cardiomyocytes. Support for this hypothesis comes from a comparison with mammalian heart failure cardiomyocytes, which are also characterized by long AP durations and high [Na+]i (2, 6). [Na+]i in ferret heart failure myocytes is 13.3 mM (22), which is strikingly similar to what we have found in trout. Furthermore, in mammalian heart failure cardiomyocytes, Ca2+ influx via reverse-mode NCX provides a significant amount of Ca2+ for contraction (2, 6), and we hypothesize that this is also the case in trout cardiomyocytes.
In rainbow trout, we found no difference between [Na+]i in atrial and ventricular myocytes. In contrast, [Na+]i is significantly higher in atrial than ventricular guinea pig myocytes (35). The lack of regional differences in [Na+]i in trout heart is interesting, because atrial and ventricular myocytes otherwise differ in contractile characteristics. Atrial cells have a shorter AP duration (34) and faster rate of contraction and recovery, and this is associated with a higher myosin ATPase activity and SR Ca2+ uptake rate and greater contribution of the SR to contraction (1). As [Na+]i is the same in atrial and ventricular cells, regional differences in the functional importance of NCX will be governed by the shape of the APs and Ca2+ transients. It is likely that the Ca2+ influx via reverse-mode NCX is larger in ventricular than atrial cells due to their longer AP duration (34).
Acute temperature change did not affect either resting [Na+]i or its contraction-induced increase in rainbow trout cardiomyocytes. This is in contrast to the situation in mammalian cardiomyocytes in which cooling increases [Na+]i (29). However, the temperature sensitivity of ion-regulating mechanisms varies between species. Indeed, the trout cardiac NCX is relatively temperature insensitive compared with mammalian cardiac NCX (30). Furthermore, trout heart ryanodine receptors remain functional during cold temperatures (20).
Our results for [Na+]i highlight the potential efficacy of reverse-mode NCX in rainbow trout cardiac myocytes. Indeed, previous work using whole cell patch-patch has shown that when intracellular Na+ is 16 mM, Ca2+ influx via reverse mode NCX can trigger sizeable contractions and can cause Ca2+-induced Ca2+ release from the SR (19). Thus, the level of intracellular Na+ reported here suggests that Ca2+ influx via reverse-mode NCX is physiologically important in trout excitation-contraction coupling. It is therefore interesting to consider that the small or no effect of contraction frequency and acute temperature change on [Na+]i may be important in the context of heart function. In trout heart preparations, an increase in temperature and/or stimulation frequency is associated with a decrease in contractile force (28), and in isolated cells, this is associated with a decrease in the Ca2+ transient ([Ca2+]i) (18, 26). Previous studies have primarily focused on the role of Ca2+ influx via L-type Ca2+ channels, calcium current (ICa), and Ca2+-induced Ca2+ release from the SR in causing this decrease in [Ca2+]i. Interestingly, these mechanisms only provide a partial explanation for the decrease in [Ca2+]i with increased temperature and frequency (26, 28). The present study suggests that reverse-mode NCX Ca2+ influx is also an important Ca2+-source for contraction, and our calculations (see Fig. 6) suggest that it too decreases with an increase in temperature and frequency because it depends on the AP duration. Shiels et al. (26) recorded APs and Ca2+ transients in rainbow trout atrial myocytes at three different temperatures at stimulation frequencies that corresponded to the in vivo heart rate at each temperature. Using this data (26), we calculated the subsarcolemmal Ca2+ transient by the method of Weber et al. (36) and used our present data on [Na+]i to calculate ENCX. ([Na+]i was calculated by linear regression of the contraction-induced increase in 340/380 ratio to be 14.1 mM at 7°C and 0.6 Hz, 14.4 mM at 14°C and 1.0 Hz, and 14.7 mM at 21°C and 1.4 Hz). Figure 6 shows a plot of ENCX and the corresponding APs [from Shiels et al. (26)] at 7, 14, and 21°C. The area between the curves when Em > ENCX, which can be taken as a measure of reverse-mode NCX driving force, decreases from 100% at 7°C and 0.6 Hz to 83% at 14°C and 1.0 Hz and 57% at 21°C and 1.4 Hz. Similar frequency-dependent decreases in Ca2+ influx on reverse-mode NCX were calculated in ventricular myocytes at a constant temperature using data from Harwood et al. (18), which stated that the area between Em and ENCX decreased from 100% at 0.2 Hz to 89% at 0.8 Hz and 56% at 1.4 Hz.
Thus, our calculations show that the temperature- and frequency-dependent decrease in [Ca2+]i can be explained by the decrease in reverse-mode NCX Ca2+ influx supplementing the decrease in ICa. The negative force-frequency relationship in trout and rat cardiomyocytes are thus of different origin. Namely, while the situation in rat is still unclear (for a review, see Ref. 13), the AP duration is too short for reverse-mode NCX to play a significant role in excitation-contraction coupling. Interestingly, we also calculated that temperature and frequency do not decrease reverse-mode NCX Ca2+ influx if [Na+]i is 10 mM. This, together with the fact that the effect of temperature and frequency on [Ca2+]i cannot be fully explained by ICa or SR Ca2+-release, indirectly supports the high [Na+]i found in this study.
In conclusion, we found a high [Na+]i in rainbow trout cardiomyocytes, which suggests that reverse-mode NCX Ca2+ influx is an important physiological Ca2+ source during contraction. On the basis of our present study and our calculations, we predict that Ca2+ influx via NCX will decrease with an increase in temperature and contraction frequency. Thus, changes in [Na+]i are important to consider when investigating the effects of temperature and contraction frequency on the Ca2+ transient and contractile force in trout hearts.
This study was funded by the Biotechnology and Biological Sciences Research Council and the Danish Natural Science Research Council.
The authors thank Ed White for providing the data from Harwood et al. (18), and Fabien Brette for useful comments on an earlier version of the manuscript.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society