Na+-Ca2+ Exchange Activity Is Localized in the T-Tubules of Rat Ventricular Myocytes
Detubulation of rat ventricular myocytes has been used to investigate the role of the t-tubules in Ca2+ cycling during excitation-contraction coupling in rat ventricular myocytes. Ca2+ was monitored using fluo-3 and confocal microscopy. In control myocytes, electrical stimulation caused a spatially uniform increase in intracellular [Ca2+] across the cell width. After detubulation, [Ca2+] rose initially at the cell periphery and then propagated into the center of the cell. Application of caffeine to control myocytes resulted in a rapid and uniform increase of intracellular [Ca2+]; the distribution and amplitude of this increase was the same in detubulated myocytes, although its decline was slower. On application of caffeine to control cells, there was a large, rapid, and transient rise in extracellular [Ca2+] as Ca2+ was extruded from the cell; this rise was significantly smaller in detubulated cells, and the remaining increase was blocked by the sarcolemmal Ca2+ ATPase inhibitor carboxyeosin. The treatment used to produce detubulation had no significant effect on Ca2+ efflux in atrial cells, which lack t-tubules. Detubulation of ventricular myocytes also resulted in loss of Na+-Ca2+ exchange current, although the density of the fast Na+ current was unaltered. It is concluded that Na+-Ca2+ exchange function, and hence Ca2+ efflux by this mechanism, is concentrated in the t-tubules, and that the concentration of Ca2+ flux pathways in the t-tubules is important in producing a uniform increase in intracellular Ca2+ on stimulation.
Transverse (t-) tubules are invaginations of the cell membrane that are found in skeletal and cardiac muscle. In cardiac ventricular myocytes they occur perpendicular to the longitudinal axis of the cell at intervals of ≈2 μm1 and extend into the cell both laterally and longitudinally.2 Immunohistochemical studies have shown that many of the proteins involved in excitation-contraction coupling are concentrated at the t-tubules,3,4⇓ suggesting an important role for the t-tubules in excitation-contraction coupling. For example, L-type Ca2+ channels are concentrated in the t-tubular membrane, in close proximity to the Ca2+ release channels (ryanodine receptors; RyR) of the junctional sarcoplasmic reticulum (SR).5
Detubulation of cardiac cells enables the functional consequences of this protein distribution to be investigated.6 Detubulation of rat ventricular myocytes results in a ≈25% decrease in cell surface area, which is accompanied by a ≈75% decrease in the amplitude of the L-type Ca2+ current, and a decrease in the amplitude of the Ca2+ transient.6 This is consistent with the immunolabeling data that show the L-type Ca2+ channels concentrated in the t-tubular membrane.
In the present study, we have used detubulation to investigate further the role of the t-tubules in Ca2+ cycling during excitation-contraction coupling in rat ventricular myocytes. We have investigated the hypothesis that the t-tubules underlie the rapid and homogeneous Ca2+ release observed in cardiac ventricular myocytes (eg, Kawai et al6). We have also investigated the functional distribution of the Na+-Ca2+ exchanger between the surface and t-tubule membranes. This exchanger is the major route of Ca2+ efflux from cardiac cells,7,8⇓ and can also act as a Ca2+ influx pathway; it has been suggested that Ca2+ entering the cell by this route might, under some circumstances, act as a trigger for Ca2+ release from the SR.9 Determining the location of the exchange protein is, therefore, important for our understanding of Ca2+ flux in the cardiac cell. However the distribution of the exchanger is unclear from immunolabeling studies. Some studies show concentration of the exchange protein in the t-tubule membrane, 3 whereas others show a more uniform distribution of the exchanger between the surface and t-tubular membranes.10
Materials and Methods
Adult Wistar rats (≈250 g; Central Biomedical Services, Leeds, UK) were stunned and then killed by cervical dislocation, in accordance with the UK Home Office Guidance on the Operation of the Animals (Scientific Procedures) Act of 1986. Ventricular myocytes were isolated as described previously.11
Osmotic Shock Treatment of Ventricular Myocytes With Formamide
To induce osmotic shock, and hence detubulation, 1.5 mol/L formamide was added to the cell perfusate for 15 minutes, before returning to control solution, as described previously.6
Confocal imaging was performed using a laser-scanning unit (BioRad Microradiance 2000) attached to a Nikon Diaphot inverted microscope. The X-Y resolution of the system was 0.45 μm, as measured from the point-spread function of fluorescent microspheres (diameter 0.175 μm). The aperture size was set to the size of the Airy disc to optimize z-axis resolution. Fluo-3 was excited with the 488- nm line of an argon ion laser and fluorescence measured at >515 nm. Line scan images were acquired at 6 ms or 2 ms intervals across the width of the cell. To reduce possible laser damage, the position of the line was changed after 2 to 3 scan sequences. Images were analyzed using IDL (Research Systems Inc) and Laserpix (Bio-Rad) software.
Monitoring Intracellular Ca2+ (Cai) Distribution
Myocytes were loaded with the fluorescent Ca2+ indicator fluo-3 (Molecular Probes) by incubation with its AM form (5 μmol/L for 6 minutes).
Monitoring Ca2+ Efflux
Myocytes were initially bathed in a solution containing (in mmol/L) NaCl 113, KCl 5, MgSO4 1, Na2HPO4 1, HEPES 10, Na acetate 20, glucose 10, and CaCl2 1 (pH 7.4) at room temperature. After stimulation at 0.5 Hz, bathing [Ca2+] was reduced to ≈100 nmol/L (solution as above except 0 Ca2+ and 0.4 mmol/L EGTA) for 100 seconds. The cell was then perfused for 60 seconds with solution containing 10 μmol/L Fluo-3 and 0.05 mmol/L EGTA before rapid application of, and perfusion with, solution containing 20 mmol/L caffeine. In early experiments, 5 μmol/L thapsigargin was also used during the 60 seconds before application of caffeine, to inhibit Ca2+ reuptake by the SR. However, this had little effect on the response to caffeine and was not used subsequently.
Na+-Ca2+ exchange current (INa/Ca) was monitored using the whole-cell (ruptured patch) voltage clamp technique as described previously.12 The pipette solution contained (in mmol/L) CsCl 110, NaCl 20, TEACl 20, MgCl2 0.4, HEPES 10, BAPTA 5, CaCl2 1, and glucose 5 (pH 7.2, adjusted with CsOH); calculated free [Ca2+] 95 nmol/L. The cells were superfused with a solution containing (in mmol/L) NaCl 140, CsCl 4, MgCl2 1, HEPES 5, Glucose 10, and CaCl2 2.5 (pH 7.4, adjusted with NaOH). This solution also contained 20 μmol/L strophanthidin to block the Na+/K+ pump and 10 μmol/L nifedipine to block L-type calcium current (ICa). The BAPTA within the pipette buffered Cai sufficiently to stop contraction; control experiments showed that Ca2+ and K+ currents were absent in these solutions.
Sodium current (INa) was also monitored using the whole-cell (ruptured patch) voltage clamp technique with low resistance electrodes (1 to 2 MΩ). The pipette solution contained (in mmol/L) CsCl 130, MgCl2 2, Na2ATP 5, Na2GTP 0.5, CaCl2 0.5, HEPES 10, and EGTA 5 (pH 7.2, adjusted with CsOH). The cells were superfused with a solution containing (in mmol/L) NaCl 20, CsCl 110, TEACl 10, MgCl2 2, CaCl2 1, CoCl2 2, HEPES 10, and Glucose 5.5 (pH 7.4, adjusted with CsOH).
Membrane currents were recorded using an Axopatch 1D voltage clamp amplifier, controlled by Vclamp software (Cambridge Electronic Design) running on an Elonex PC433 computer, via a CED1401 A/D interface. Data were digitized (at 2 kHz during recording of INa/Ca and 20 kHz during recording of INa) via this interface and stored on the hard disk of the computer. The protocols are described in the relevant section of the Results. A straight line was fitted to the central, linear portion of the I-V relation of INa/Ca; this line adequately described the majority of the I-V relationship, and its slope was used as a simple method of characterizing INa/Ca. The rapid solution changes made on the confocal microscope could only be performed at room temperature without inducing large temperature transients. To ensure consistency, all experiments were, therefore, performed at room temperature.
All solutions were prepared using ultrapure water supplied by a Milli-Q system (Millipore). All solution constituents were reagent grade and purchased from Sigma.
Data are expressed as mean±SEM of n cells; paired or unpaired t tests were used as appropriate. A value of P<0.05 was taken as statistically significant.
Effect of Formamide Treatment on T-Tubule Structure and Membrane Capacitance
Treatment with formamide resulted in loss of the t-tubules, visualized using the dye di-8-ANNEPS, as reported previously6 (not shown). Membrane capacitance was determined from capacity transients at the beginning of 10-mV hyperpolarizing pulses from a holding potential of −80 mV. Capacitance decreased significantly (P<0.05, n=25) from 193±41 pF in control cells to 143±34 pF in detubulated cells. This decrease (25.9%) is similar to that reported previously.6 Assuming that 80% of the treated cells are detubulated,6 these data suggest that the decrease in cell capacitance in the cells that are detubulated is 32.4%, close to previous estimates of the percentage of the cell membrane in the t-tubules (33%).1
Effect of Detubulation on Intracellular Ca2+ Distribution
Previous work has shown that ICa is concentrated in the t-tubules.6 Detubulation would therefore be expected to disrupt normal synchronous Ca2+ release. Confocal microscopy was used to image the spatial distribution of Cai in control and detubulated cells. Figure 1A shows a line scan image of a control cell, loaded with fluo-3, during electrical stimulation, showing a rapid synchronous rise of Cai across the width of the cell. In contrast, Figure 1B shows that in a detubulated cell Cai initially increases close to the surface membrane and then propagates into the cell, at 48.4±5.2 μm · s−1 (n=8), so that Cai in the center of the cell rises later than at the periphery.
We investigated whether the propagation of Ca2+ into the cell observed in detubulated myocytes was SR dependent. Figure 1 shows that the propagation (Figure 1B) is inhibited by thapsigargin (Figure 1C), although a small rise of Cai close to the cell membrane can still be observed. The SR inhibitor ryanodine had a similar effect (not shown). Thus, it appears that the slow propagation of Ca2+ into the cell after detubulation is due to Ca2+-induced Ca2+ release (CICR) from the SR after Ca2+ influx at the cell surface. It is unlikely that altered RyR function or distribution contributes to the inhomogeneous Ca2+ release observed after detubulation because the rate of propagation of spontaneous Ca2+ waves, which depends on RyR distribution and function, was similar in control and detubulated cells, and the amplitude and distribution of the rise of Cai produced by caffeine, which acts by opening RyR, was the same in control and detubulated cells (see next section).
Effect of Detubulation on Ca2+ Extrusion
Figure 2 shows confocal images of cardiac cells bathed in solution containing fluo-3 to monitor extracellular Ca2+ (Cao; see Materials and Methods). Under these conditions, application of 20 mmol/L caffeine enabled Ca2+ efflux from the cell to be monitored. Figure 2A shows that in a control cell, application of caffeine resulted in a rapid and marked rise of Cao over ≈1 second, as Ca2+ was extruded from the cell. Cao subsequently slowly declined to baseline as the extruded Ca2+ was washed away in the perfusing solution. Figure 2B shows that in a detubulated cell under identical conditions the rise of Cao was markedly reduced: peak fluorescence/basal fluorescence (F/Fo) increased to 5.73±0.46 in control cells (n=8), and to 2.28±0.22 in detubulated cells (n=7; P< 0.05). These data suggest, therefore, that loss of the t-tubules results in a decrease in Ca2+ efflux that is greater than the ≈30% loss of cell membrane (see “Effect of Formamide …”), and therefore, that Ca2+ efflux pathways are concentrated in the t-tubules.
Na+-Ca2+ exchange and sarcolemmal Ca2+ ATPase are the two main extrusion pathways in rat ventricular cells.7,8⇓ We investigated, therefore, the effect of the sarcolemmal Ca2+ ATPase inhibitor carboxyeosin on the residual Ca2+ efflux in detubulated cells. Figure 2C shows Cao during caffeine application in a representative detubulated cell in the presence of carboxyeosin, showing that inhibition of the ATPase abolished the residual rise of Cao (n=3).
Figure 3 shows that electrical stimulation of detubulated cells produced a smaller intracellular Ca2+ transient than in control cells, as reported previously,6 but that the increase of Cai produced by caffeine had a similar amplitude and uniform distribution in control and detubulated cells (F/Fo was 3.01±0.22 in control cells, n=10, 3.03±0.20 in detubulated cells, n=11; NS), but that the decline of the caffeine-induced rise of Cai was prolonged in the detubulated cells: the time for a 75% decrease from peak increased from 2.57±0.23 (n=10) to 5.56±0.04 seconds (n=11; P<0.001). Thus, the decreased Ca2+ efflux observed in detubulated cells is not due to a decrease in the rise in Cai produced by caffeine.
To ensure that the decreased efflux observed after formamide treatment was not due to a direct effect of formamide on the proteins involved, the effect of formamide treatment on Ca2+ extrusion was monitored in atrial cells, which lack t-tubules. Formamide treatment had no significant effect on the rise of Cao that occurred on application of caffeine to these cells: F/Fo increased to 3.44±0.66 in control cells (n=6) and to 3.22±0.21 in formamide-treated cells (n=6; NS; not shown). Thus, it appears that in ventricular cells Na+-Ca2+ exchange flux is concentrated in the t-tubules, and that the residual Ca2+ flux after detubulation is due to Ca2+ ATPase activity on the surface membrane. To investigate this further, INa/Ca was monitored.
Effect of Detubulation on INa/Ca
INa/Ca was measured in control and detubulated cells using a descending voltage clamp ramp protocol as described previously12: from a holding potential of −40 mV, the cell membrane was depolarized to +50 mV and held at this voltage for 100 ms, to allow inactivation of rapidly inactivating currents, before being hyperpolarized to −120 mV over 2 seconds. Ramps were applied at 0.4 Hz. This was repeated in the presence of 5 mmol/L nickel, to block the exchanger. The difference current (between the absence and presence of nickel) was taken as INa/Ca (see Convery and Hancox12 and Discussion).
Figure 4A shows the mean current density-voltage relationship for the current recorded during the descending voltage ramp before and after application of Ni, and the difference current, in control cells. The slope of each current-voltage relation was calculated to enable comparison of the current in different conditions. Nickel significantly decreased the slope of the I-V relation (from 0.304±0.083, n=15 cells, to 0.229±0.087, n=12 cells; P<0.05), suggesting that 5 mmol/L Ni blocks a current, which we take to be INa/Ca,12,13⇓ which is plotted in Figure 4A as the difference current.
Figures 4B shows data obtained in detubulated cells using the same protocol. In these cells, Ni had no significant effect on the current-voltage relationship. The mean values of the slopes in the absence and presence of Ni were not significantly different (0.229±0.087, n=14, in detubulated cells in control conditions, compared with 0.188±0.092, n=11, in detubulated cells in the presence of Ni; NS). ie, in detubulated cells the difference current is almost zero, so that there appears to be no INa/Ca. In addition, the slope of the relationship in detubulated cells in the absence of Ni was not significantly different from that in control cells in the presence of Ni, again compatible with loss of INa/Ca after treatment with formamide. It is worth noting, however, that the density of the Ni insensitive current in the detubulated cells was also not significantly different from the control current density in the normal cells. Although the identity of the Ni-insensitive current is unknown, a possible explanation of this result is that this current is concentrated in the surface membrane, so that its density increases slightly on detubulation, offsetting a decrease in INa/Ca density, although the small current density makes comparison between groups difficult.
It has previously been suggested that Na+ entering the cell via INa can alter the activity of the Na+-Ca2+ exchanger,19 in which case the localized activity of the exchanger could be secondary to localization of INa. We therefore investigated the localization of INa.
Effect of Detubulation on INa
INa was elicited using 20-ms voltage clamp pulses from a holding potential of −110 mV to a series of test potentials between −105 mV and +50 mV at a frequency of 0.2 Hz. Figure 5 shows representative currents recorded using this protocol and the current density-voltage relationship in control and detubulated cells, showing that current density was not significantly altered by formamide treatment at any voltage (ie, the current decreased in proportion to the decrease in cell capacitance and, hence, surface area), nor was time to peak of the current significantly altered (2.15±0.14 ms in control cells at −10 mV [n=13]; 1.93±0.11 ms in detubulated cells [n=9]; NS). Thus, formamide has no significant effect on INa density, suggesting that INa is uniformly distributed between the cell surface and t-tubule membranes.
Treatment of isolated cardiac myocytes with formamide causes loss of the t-tubules,6 consistent with the measured decrease in membrane capacitance. It is unlikely that formamide has direct effects on ion transport proteins6 because formamide treatment of atrial cells (which lack t-tubules) has no significant effect on cell capacitance, ICa,14 or Ca2+ efflux (as seen in this study). The present study also shows that formamide has no apparent effect on INa or RyR function.
Currents are shown normalized to membrane capacitance (ie, to cell surface area, assuming uniform capacitance throughout the cell membrane); a decrease in current density after detubulation suggests that the current was concentrated in the t-tubule membrane.
The technique used to monitor INa/Ca has been described previously.12 Blockers were used to inhibit other currents so that Ni-sensitive current, as in previous studies, is taken to represent INa/Ca. Block of the exchanger might alter local Cai, thus altering Ca2+-activated currents. However, Choi et al15 have shown that even when Cai is increased using caffeine (ie, to levels higher than those in the present study, in which Cai was buffered with BAPTA), no resulting current was observed in rat ventricular myocytes when the exchanger was inhibited by Ni or by the absence of bathing Na+ and Ca2+, making this unlikely.
INa was monitored using the whole-cell patch clamp technique with low-resistance (1 to 2 MΩ) electrodes, and using low extracellular [Na+] (20 mmol/L) and normal/high intracellular [Na+] (11 mmol/L) to reduce the Na+ gradient and, hence, current. However, despite this, and although the time to peak of the current was relatively fast, it seems possible that voltage control was lost during measurement of INa. This problem would be common to control and detubulated cells, and the amplitude and time course of the current were not significantly different in the two groups.
Spatial Distribution of Cai During Electrical Stimulation
In control cells, Ca2+ increased homogeneously across the cell on electrical stimulation. However, in formamide-treated cells, Ca2+ initially rose close to the cell membrane and then propagated into the cell interior, consistent with detubulation: the initial rise close to the surface membrane is similar to that reported in atrial and Purkinje cells, which lack t-tubules.16,17⇓ This propagation was inhibited by thapsigargin and ryanodine, suggesting that it is due to CICR as in atrial cells16 but in contrast to Purkinje cells, in which the inward spread of Ca2+ appears to be due principally to buffered diffusion of Ca2+.17 Thus, it appears that in detubulated cells the initial rise is due to Ca2+ influx across the surface cell membrane, and the propagation to CICR from the SR. Because Na+-Ca2+ exchange activity appears to be concentrated in the t-tubules (see next section), it is likely that Ca2+ influx via Ca2+ channels in the surface membrane causes the initial rise of Cai. Consistent with this idea, the rise was increased by the Ca2+ channel agonist BayK (not shown).
In agreement with previous work showing that Ca2+ sparks arise predominantly at the t-tubules,18 the present data suggest that the t-tubules play an important role in excitation-contraction coupling because the high concentration of Ca2+ influx pathways in the t-tubule membrane, ICa5,6⇓ and INa/Ca (see next section), allows synchronized CICR from the adjacent ryanodine receptors and thus synchronized Ca2+ release throughout the cell.
Distribution of the Na+-Ca2+ Exchanger Between Cell Surface and T-Tubule Membrane
Measurement of Cao during application of caffeine to control cells showed a marked and rapid rise of Cao, as Ca2+ was extruded from the cell. Although the amplitude of the rise of Cai produced by caffeine was not significantly different in control and detubulated cells, the rise of Cao was significantly smaller in detubulated cells, suggesting that the Ca2+ efflux pathways are concentrated in the t-tubules. In the presence of carboxyeosin, even the residual Ca2+ efflux was abolished, suggesting that it was due to Ca2+ ATPase activity. These data suggest, therefore, that Ca2+ extrusion via Na+-Ca2+ exchange occurs exclusively in the t-tubules. Consistent with this, measurement of INa/Ca showed that detubulation abolished INa/Ca.
It is unlikely that the observed effects can be ascribed to direct effects of formamide, because formamide treatment had no effect on Ca2+ efflux from atrial cells, which lack t-tubules but possess Na+-Ca2+ exchange, and which undergo similar size and volume changes as ventricular cells in response to formamide and its withdrawal (Brette and Orchard, unpublished observation, 2001). In addition, measurements were made after removal of formamide, so that osmotic effects on the exchanger would be absent. It is also unlikely that the effects of formamide are secondary to effects on the cytoskeleton, because formamide treatment had no apparent effect on cytoskeleton structure, assessed by labeling with monoclonal antibody to β-tubulin (Brette and Orchard, unpublished observations, 2002).
The simplest explanation of the present data is that the exchange protein is localized to the t-tubules, consistent with a recent immunohistochemical study using rat ventricular myocytes.5 An alternative explanation is that the exchange protein is uniformly distributed, but only the protein located in the t-tubules is active during the protocols used in the present investigation. This might occur, for example, if Na+ entering the cell via INa is used by the exchanger,19 but access of this Na+ to the exchanger is different in the t-tubules and at the surface membrane. This appears unlikely because (1) INa appears to be uniformly distributed (as seen in this study); (2) there appears to be little colocalization of the two proteins5; (3) during measurement of INa/Ca, INa will be triggered on returning to −40 mV from −120 mV at the end of the voltage ramp, so that time is available for intracellular diffusion of Na+ between INa being triggered and INa/Ca being monitored.
Although it has previously been shown that there is a high concentration of Ca2+ channels in the t-tubules,5,6⇓ the Ca2+ channel blocker nifedipine was used in the present study during measurement of INa/Ca to inhibit interference by Ca2+ influx by this route. Thus, ion flux on Na+-Ca2+ exchange is localized in the t-tubules, probably because the exchange protein is concentrated in the t-tubules, although it remains possible that protein function does not mirror protein distribution because of other (unknown) local regulation.
It is also of interest that INa appears to be uniformly distributed across the cell membrane; Scriven et al5 showed, using immunolabeling, that the L-type Ca2+ channel, Na+-Ca2+ exchanger, and Na+ channel appear to be concentrated in the t-tubules. However, only the Na+ channel showed significant presence on the sarcolemma between the z-lines, consistent with a more uniform distribution of this protein.
Functional Consequences of the Observed INa/Ca Distribution
Na+-Ca2+ exchange is the major Ca2+ efflux pathway in rat ventricular myocytes.7,8⇓ Because Ca2+ efflux via the exchanger is localized to the t-tubules, they play an important role in Ca2+ efflux in normal cells. It is, however, worth considering the determinants of the decline of the electrically stimulated, and caffeine-induced, Ca2+ transients in detubulated cells. The decline of the electrically stimulated Ca2+ transient in detubulated cells is presumably due to Ca2+ uptake by the SR, Ca2+ diffusion into adjacent regions of the cell because of the inhomogeneous distribution of Cai, and Ca2+ efflux via the sarcolemmal Ca2+ ATPase.
The amplitude of the caffeine-induced rise of Cai was unchanged by detubulation, although its decline was slower. The maintained amplitude is, perhaps, surprising, given that ICa and INa/Ca are concentrated in the t-tubules. However, Ca2+ efflux was lower in the detubulated cells. Because Ca2+ extrusion can decrease the amplitude of the caffeine-induced Ca2+ transient, this might suggest that the SR Ca2+ content was lower in detubulated cells, consistent with a loss of Ca2+ influx pathways; such a decreased load may also contribute to the smaller rapid upstroke of the Ca2+ transient observed in detubulated cells6 (as seen in this study). In addition, however, the small fraction of ICa remaining in the surface membrane,6 and the loss of the main Ca2+ efflux pathway (INa/Ca) in detubulated cells would result in the SR retaining more Ca2+, thus helping to maintain its content20; the longer diffusion distances for Ca2+ from the center of the cell to extrusion at the cell surface and the relatively slow Ca2+ extrusion rate of the remaining extrusion pathways may also help maintain SR Ca2+ uptake and content.
The prolonged decline of the caffeine-induced increase of Cai in detubulated cells is expected, because Na+-Ca2+ exchange is normally the main pathway removing Ca2+ from the cytoplasm in the presence of caffeine.7,8⇓ However, previous work showed a more modest slowing of the rate of decline.6 It is likely that the more marked effect observed in the present study is a better assessment of the response to caffeine, because (1) in the present study we ensured detubulation of each cell before application of caffeine, by checking that electrical stimulation caused the pattern of Ca2+ release shown in Figure 1B. Because not all cells are detubulated by formamide,6 it is likely that the mean data in the previous study included data from normal cells. (2) A higher concentration of caffeine was used in the present study, to ensure more complete Ca2+ release from the SR and to minimize the contribution of Ca2+ reuptake by the SR to the change of Cai.21 (3) In the present study, caffeine was applied directly and rapidly to the cell being studied, whereas in the previous study the rapidity of application depended on the position of the cell in the bath, which is likely to have influenced the observed response.
In previous studies, the rate constant of decline of the caffeine-induced rise of Cai has been used to determine the relative contributions of different Ca2+ extrusion pathways.7,8⇓ Although detubulation decreased the rate constant of decline (from 0.54 s−1 to 0.25 s− 1), it is difficult to compare the relative contribution of different efflux pathways in control and detubulated cells using this technique, because other factors may influence the rate of decline in detubulated cells. For example, diffusion distances to the cell membrane will be different and the t-tubules, which appear to reseal within the cell although uncoupled from the surface membrane, may play some role in intracellular Ca2+ cycling. However, Figure 2 shows that in detubulated cells, all of the Ca2+ efflux that occurs on application of caffeine is inhibited by carboxyeosin, so that in the absence of t-tubules there appears to be no efflux via Na+-Ca2+ exchange: all the efflux appears to be via the Ca2+ ATPase. This is in contrast to control cells, in which Na+-Ca2+ exchange is the most important efflux pathway.7,8⇓
The Ca2+ extrusion pathways responsible for the decline of the caffeine-induced increase of Cai in detubulated cells are not clear. In the absence of carboxyeosin, the sarcolemmal Ca2+ ATPase will extrude Ca2+ (and a continuing low rate of Ca2+ efflux may be difficult to detect). It is also possible that in both the absence and presence of carboxyeosin, intracellular organelles such as the mitochondria may sequester Ca2+.7,8⇓
The location of the exchanger is important for the normal function of the cell because of its role as (1) a route for Ca2+ influx, which may act as a trigger for Ca2+ release from the SR, and load the SR with Ca2+9; and (2) the main Ca2+ efflux pathway from ventricular myocytes.7,8⇓
Concentration of Na+-Ca2+ exchange activity in the t-tubule membrane of rat myocytes is germane to its trigger function, because Ca2+ influx via the exchanger will occur adjacent to the RyRs, which are concentrated at the t-tubules. However, it is not clear that the exchanger is located sufficiently close to the ryanodine receptors,5 or that the flux rate of Ca2+ through the exchanger is sufficient to make Ca2+ entry via this pathway an effective trigger under normal conditions.22,23⇓ This observation could, however, explain why it has proved so difficult to measure an exchange current accompanying putative Na+-Ca2+ exchange-generated Ca2+ release: a small INa/Ca flowing in the t-tubules might, under some conditions, be sufficient to cause release, because Ca2+ will be delivered close to the ryanodine receptor, but might be difficult to detect. The location of the exchanger may also enable Ca2+ influx via the exchanger to load the SR with Ca2+,9 because SR Ca-ATPase is found adjacent to the t-tubules,24 so that Ca2+ entering the cell via the exchanger may be sequestered by these pumps.
The location of Na+-Ca2+ exchange activity may also be important for its role as the main route for Ca2+ efflux. First, because it will enable a relatively rapid and synchronous removal of Ca2+ throughout the cell. Second, because the exchanger is located close to the site of SR Ca2+ release, Ca2+ efflux via the exchanger may be sensitive to modulation by Ca2+ released from the SR. Because Na+-Ca2+ exchange is electrogenic, current carried by the exchanger, which influences action potential configuration, will also be concentrated in the t-tubules and modulated by the local environment.
This work was funded by the Wellcome Trust and British Heart Foundation.
Original received February 6, 2002; revision received July 8, 2002; accepted July 8, 2002.
- ↵Soeller C, Cannell MB. Examination of the transverse tubular system in living cardiac rat myocytes by 2-photon microscopy and digital image-processing techniques. Circ Res. 1999; 84: 266–275.
- ↵Frank JS, Mottino G, Reid D, Molday RS, Philipson KD. Distribution of the Na+-Ca2+ exchange protein in mammalian cardiac myocytes: an immunofluorescence and immunocolloidal gold-labeling study. J Cell Biol. 1992; 117: 337–345.
- ↵Negretti N, O’Neill SC, Eisner DA. The relative contributions of different intracellular and sarcolemmal systems to relaxation in rat ventricular myocytes. Cardiovasc Res. 1993; 27: 1826–1830.
- ↵Kohmoto O, Levi AJ, Bridge JHB. Relation between reverse sodium-calcium exchange and sarcoplasmic reticulum calcium release in guinea-pig ventricular cells. Circ Res. 1994; 74: 550–554.
- ↵Brette F, Komukai K, Orchard CH. Validation of formamide as a detubulation agent: its effect on isolated rat atrial myocytes. Biophys J. 2001; 80: 647a.Abstract.
- ↵Leblanc N, Hume JR. Sodium-current-induced release of calcium from cardiac sarcoplasmic reticulum. Science. 1990; 248: 372–376.
- ↵Eisner DA, Choi HS, Diaz ME, O’Neill SC, Trafford AW. Integrative analysis of calcium cycling in cardiac muscle. Circ Res. 2000; 87: 1087–1094.
- ↵Evans AM, Cannell MB. The role of L-type Ca2+ current and Na+ current-stimulated Na/Ca exchange in triggering SR calcium release in guinea-pig cardiac ventricular myocytes. Cardiovasc Res. 1997; 35: 294–302.
- ↵Sipido KR, Maes MM, Van De Werf F. Low efficiency of Ca2+ entry through the Na+-Ca2+ exchanger as trigger for Ca2+ release from the sarcoplasmic reticulum. Circ Res. 1997; 81: 1043–1054.
- ↵Musa H, Lei M, Hongo H, Jones SA, Dobrzynski H, Lancaster M, Takagishi Y, Henderson Z, Kodama I, Boyett MR. Heterogenous expression of Ca2+ handling proteins in rabbit sinoatrial node. J Histochem Cytochem. 2002; 50: 311–324.