The effect of acidosis on Ca2+ uptake and release by the sarcoplasmic reticulum (SR) of rat ventricular myocytes has been investigated. Intracellular Ca2+concentration ([Ca2+]i) was monitored using fura 2; the L-type Ca2+ current (I Ca) was monitored using the perforated patch-clamp technique. Acidosis was produced either by superfusing the cells with an acid solution (intracellular and extracellular acidosis) or by NH4Cl withdrawal (intracellular acidosis). Both types of acidosis increased the amplitude, and slowed the declining phase, of the Ca2+transient. Application of caffeine produced a rise of [Ca2+]i, which declined in the continued presence of caffeine; the declining phase was slowed by the acid solution but was unaffected by NH4Cl withdrawal. Acidosis decreased the fraction of the caffeine-induced release that was released by electrical stimulation but had no effect onI Ca. It is concluded that acidosis inhibits SR Ca2+ uptake and Ca2+-induced Ca2+ release in intact myocytes but that these effects are compensated by an increase in SR Ca2+ content secondary to a rise in cytoplasmic [Ca2+].
- calcium ion
- cardiac muscle
acidosis has marked effects on intracellular [Ca2+] ([Ca2+]i) in cardiac muscle. These include an increase in diastolic [Ca2+]i(5, 23), an increase in the amplitude of the Ca2+ transient (1, 14), and a decrease in the rate of decline of the Ca2+ transient (23).
The increase in diastolic [Ca2+]iappears to be caused by two mechanisms: displacement of Ca2+ from intracellular buffers by H+ (23) and activation of Na+/H+exchange by a rise in intracellular [H+]. The consequent influx of Na+ into the cell in exchange for H+ increases intracellular [Na+], which, via the Na+/Ca2+exchange mechanism, increases [Ca2+]i(6, 14).
The acidosis-induced increase in the amplitude of the Ca2+ transient appears to be caused largely by increased Ca2+release from the sarcoplasmic reticulum (SR; Refs. 14, 23). However, the direct effect of acidosis on the SR, at least in skinned preparations of cardiac muscle and in SR vesicles, is to inhibit Ca2+ uptake by the SR Ca2+-ATPase (11, 20) and Ca2+ release by inhibiting Ca2+-induced Ca2+ release (CICR; Refs. 10, 18). These effects would be expected to decrease SR Ca2+ release and hence to decrease the amplitude of the Ca2+transient. To reconcile this observation with the observed increased Ca2+ release from the SR (14, 23), it has been suggested that the rise in cytoplasmic [Ca2+] described above leads to increased Ca2+ uptake by the SR and hence an increase in SR Ca2+ content and release, which overcomes the direct inhibitory effects of H+ on the SR.
The effect of acidosis on the decline of the Ca2+ transient has previously been characterized by its effect on the time taken for the transient to decline from its peak to half-way back toward its diastolic level (t ½), which is prolonged by acidosis. It has been suggested that this is caused by two factors. First, an acidosis-induced decrease in Ca2+ binding to troponin C, caused by an increase in the off-rate of Ca2+, would tend to slow the rate of decline of the Ca2+ transient (23). Second, acidosis-induced inhibition of Ca2+-removal pathways, particularly the Na+/Ca2+exchange mechanism (24) and the SR Ca2+-ATPase, will slow the rate of removal of Ca2+ from the cell cytoplasm and hence the decline of the Ca2+ transient. However, it has recently been reported that the rate of decline of the Ca2+ transient depends on the range of Ca2+ over which it is monitored (4), so thatt ½ will depend on the amplitude of the Ca2+transient; thus the increase in the amplitude of the Ca2+ transient observed during acidosis may itself altert ½.
Thus current data suggest an important role for the SR in the response of [Ca2+]ito acidosis. However, although acidosis has been shown to have marked inhibitory effects on SR Ca2+uptake and release in skinned cardiac preparations and in isolated vesicles, there is little direct evidence for such effects in intact cardiac cells. Indeed, in the intact cell, acidosis appears to increase Ca2+ uptake and release by the SR, and this has been explained as secondary to changes in cytoplasmic [Ca2+] as described above. The present study was undertaken, therefore, to investigate the effects of acidosis on SR Ca2+uptake and release and their possible role in the changes in [Ca2+]iobserved during acidosis in the intact cardiac cell.
MATERIALS AND METHODS
The methods used for cell isolation have been described previously (12). In brief, male Wistar rats were killed by stunning followed by cervical dislocation. The heart was removed and Langendorff perfused for 2–3 min with physiological salt solution containing 0.75 mmol/l Ca2+ (seeSolutions and drugs for composition) followed by perfusion with Ca2+-free salt solution supplemented with 1.0 mmol/l EGTA. Finally, the heart was perfused for 8 min with salt solution containing collagenase (1 mg/ml; Worthington type II, Lorne Laboratories), protease (0.1 mg/ml; type XIV, Sigma), and 50 μmol/l Ca2+. The ventricles were then dissected free from the atria, cut open to give maximum exposure to the enzyme, and placed in a conical flask in which they were gently agitated in enzyme solution supplemented with 1% bovine serum albumin (Sigma) for 5 min. This process was repeated four times, and myocytes from each period were harvested by filtration followed by centrifugation at 40 g for 40 s. Myocytes were washed by resuspending them in physiological salt solution containing 0.75 mmol/l Ca2+ and stored until required. Only cells that showed clear striations, were normally quiescent, and contracted rapidly in response to square wave field stimulation (20 ms pulses at 0.5 Hz) were used in the present study.
Measurement of [Ca2+]i.
Myocytes were loaded with the fluorescent Ca2+ indicator fura 2 by incubation with the membrane-permeant (AM) form of the dye (5 μmol/l for 10 min at room temperature). The myocytes were then plated onto the coverslip that formed the base of the experimental chamber (0.1-ml volume), which had been coated with 5–8 μl of mouse laminin to assist cell adhesion, and superfused with Tyrode solution (seeSolutions and drugs). This chamber was mounted on the stage of an inverted microscope (Nikon Diaphot); the cells were electrically field stimulated at 0.5 Hz via platinum electrodes set in the side of the chamber. Fura 2 fluorescence was elicited from an individual cell by alternate (every 2 ms) illumination with 340- and 380-nm light, obtained using a rotating filter wheel (Cairn Research) in front of a xenon excitation lamp (12). The fluorescence emitted at 510 nm was collected using a photomultiplier tube. The ratio of the fluorescence emitted at 510 nm during excitation at 340 nm to that emitted during excitation at 380 nm was obtained using an analog divide circuit and used to monitor [Ca2+]i.
The problems of using fura 2 to monitor [Ca2+]ihave been discussed previously (see, e.g., Ref. 12). However, three problems should be addressed here. The first is compartmentation of the dye when loading with the AM form. However, in previous work we have presented evidence that the degree of compartmentation of fura 2 under our loading conditions is small (see, e.g., Ref. 12). The second problem is the effect of pH on fura 2 fluorescence. In a previous study (14) in which fura 2 was used during acidosis, it was shown that reducing pH from 7.4 to 6.5 slightly decreased in vitro fura 2 fluorescence during excitation at 340 nm and slightly increased fluorescence during excitation at 380 nm. Thus acidosis would tend to decrease the fura 2 fluorescence ratio and hence to minimize the increase in [Ca2+]iobserved during acidosis in the present study. However, the effect was small in vitro and would be expected to be even smaller in the present study, because although extracellular pH (pHo) was reduced from 7.4 to 6.5, intracellular pH (pHi) would be expected to decrease by only 20–40% of the pHo change (6). In addition, no artifactual decrease in fura 2 fluorescence was observed during the time at which pHi was decreasing. This agrees with Grynkiewicz et al. (13), who showed that pH variations between 7.05 and 6.75 had little effect on the Ca2+-free or Ca2+-bound spectra or the in vitro dissociation constant of fura 2. Finally, because 10 mmol/l caffeine was used in the present study to assess the Ca2+ content of the SR, the effects of caffeine on fura 2 fluorescence should be considered. However, we showed previously that 10 mmol/l caffeine has no significant effect on fura 2 fluorescence at the excitation and emission wavelengths used in the present study (14).
Measurement of L-type Ca2+ current.
L-type Ca2+ current (I Ca) was monitored with the perforated patch-clamp technique using an Axopatch 1D (Axon Instruments) patch-clamp amplifier. Microelectrodes were pulled from nonheparinized hematocrit tubes to a resistance of 2–4 MΩ. A 200- to 400-μm column of pipette solution (seeSolutions and drugs for composition) was first drawn into the tip of the pipette, which was then backfilled with the same solution containing 240 μmol/l amphotericin B. The liquid junction potential was corrected after the tip of the electrode was placed in the experimental chamber and before seal formation. After a gigaohm seal was obtained, the pipette potential was set to −40 mV, and 5-mV, 20-ms voltage command pulses were applied to monitor amphotericin incorporation and access resistance. Electrical access was usually obtained within 5–10 min. Depolarizing pulses from −40 to 0 mV were then used to elicit an inward current that increased in amplitude and decreased in time to peak (to <15 ms in the experiments described) within 20 min. Pipette capacitance was compensated, and membrane current was filtered at 1.5 kHz (low-pass Bessel filter).
The patch-clamp amplifier described above was controlled, and command pulses generated, via an analog-to-digital (A/D) interface (model 1401, Cambridge Electronic Design) and a microcomputer (model PC-433, Elonex PLC), which was also used for data acquisition, averaging and measurement of the digitized signals (see Data acquisition and analysis).I Ca was elicited by stepping the membrane potential from a holding potential of −40 mV (to inactivate the Na+ current) to 0 mV for 200 ms or to a series of test potentials between −30 and +60 mV (in 10-mV steps), to determine the current-voltage relationship ofI Ca. The amplitude of I Cawas taken as the difference between the peak inward current during the voltage-clamp pulse and the current remaining at the end of the pulse.
Cells were usually superfused at 2–3 ml/min with control, acid, or NH4Cl-containing solution (seeSolutions and drugs). These solutions were pumped to a series of solenoid valves (Lee Products) placed close to the chamber containing the cells. The solenoid valves were used to select which of the solutions flowed to the chamber; the other solutions flowed directly to waste. Solution was continuously removed from the chamber, to waste, by suction.
Rapid application of caffeine-containing solutions was achieved using a flow rate of 15 ml/min from a reservoir placed 1.7 m above the level of the chamber, from which the caffeine-containing solution was gravity fed, via a solenoid valve, to the experimental chamber.
This rapid-flow system was also used to decrease the temperature of the solution in the experimental chamber from the normal temperature used for these experiments (22°C) to 1°C, by using solutions precooled in a glycol-water bath placed 1.7 m above the level of the experimental chamber. This allowed cells to be cooled to 1°C rapidly (<1 s) to elicit Ca2+release from the SR while inhibiting Ca2+-efflux pathways.
Data acquisition and analysis.
Fura 2 fluorescence elicited by excitation by 340- and 380-nm light, fluorescence ratio, membrane potential, and membrane current were displayed on a six-channel pen recorder (model RS3600, Gould Electronics) and were also digitized and recorded on videotape via a digitizer (model D-890, Neuro Data Instrument) for later off-line analysis. The signals were also digitized and recorded directly to the computer’s hard disk via a CED A/D interface (Cambridge Electronic Design) running under the control of CED Chart software. The digitized records of the fura 2 fluorescence ratio were used to calculate the amplitude of the transients elicited by electrical stimulation and of the rise of [Ca2+]ielicited by caffeine, the time course of the decline of [Ca2+]iduring the transient and during the application of caffeine, and the differential of the declining phase of the Ca2+ transient.
Solutions and drugs.
The physiological salt solution used for isolation and storage of myocytes contained (in mmol/l) 130 NaCl, 5.4 KCl, 1.4 MgCl2 ⋅ 6H2O, 0.4 NaH2PO4, 10.0 creatine, 20.0 taurine, 10.0 glucose, and 10.0 HEPES, titrated to pH 7.3 with 2 mol/l NaOH. This was supplemented with EGTA or Ca2+ for the different stages of the isolation procedure described in Cell isolation. The control HEPES-based physiological salt (Tyrode) solution used during the experiments contained (in mmol/l) 130 NaCl, 5 KCl, 1 NaH2PO2 ⋅ 12H2O, 1 MgSO4 ⋅ 7H2O, 1 CaCl2, 20 Na+-acetate, 10 glucose, and 10 HEPES with 5 U/l insulin, pH set to 7.4 using 2 mol/l NaOH. For the measurement ofI Ca, 20 mmol/l CsCl was added to this solution to inhibit contaminating K+ currents. The pipette solution used for the measurement ofI Ca contained (in mmol/l) 130 CsCl, 10 NaCl, 1 MgCl2, 5 HEPES, and 1 CaCl2, pH set to 7.2 using 1 mol/l CsOH. For the rapid cooling experiments, the aliquot of Tyrode solution that was used as control superfusate was set to pH 7.4 or 6.5 at 22°C using 2 mol/l NaOH; the aliquot used as the cold solution was set to pH 7.4 or 6.5 at 1°C using 2 mol/l NaOH. Caffeine was added directly to the Tyrode solution to produce a final concentration of 10 mmol/l.
Two types of acidosis were used in the present study. For the first (HEPES based) acidosis, the pH of the HEPES solution was set to 6.5. This solution will decrease both pHo and pHi (6); we have previously shown (15) that pHi has decreased to a new steady state after 2-min exposure to this type of acidosis in our experimental conditions. The cells were, therefore, superfused with this solution for at least 5 min before the measurements reported in the present paper were obtained. To produce an intracellular acidosis only, the cells were superfused with HEPES solution (pH 7.4) containing 15 mmol/l NH4Cl for 5 min. Subsequent withdrawal of NH4Cl results in a transient decrease of pHi (e.g., Ref. 6), which rapidly recovers back to control values. Because this technique produces a rapid, but transient, decrease of pHi, measurements were taken ∼1 min after withdrawal of NH4Cl.
All data are expressed as means ± SE ofn cells. For each experiment thesen cells were obtained from a minimum of three hearts. Statistical comparisons were made using an unpaired or paired t-test as appropriate. Values of P < 0.05 were taken to indicate statistical significance.
Effect of acidosis on Ca2+ transient amplitude and contraction.
Figure 1 shows slow time-base records of the fura 2 fluorescence ratio (Fr) and changes in cell length from a representative myocyte before, during, and after 5-min exposure to a HEPES-based acidosis. On exposure to the acid solution, there was a rapid decrease in contraction that was followed by a slower partial recovery. There was little change in the size of the Ca2+ transient during the initial decrease in contraction. However, the partial recovery of contraction was accompanied by an increase in both diastolic Ca2+ and the amplitude of the Ca2+ transient. These data agree with previous studies (see, e.g., Ref. 1) that suggested that the initial decrease in contraction is caused principally by a decreased response of the myofilaments to Ca2+, whereas the slower partial recovery is caused by the accompanying increase in the size of the Ca2+ transient, which appears to be caused by increased Ca2+release from the SR (see introduction).
Effect of acidosis on rate of decline of Ca2+ transient.
Figure 2 Ashows that, in addition to the increase in amplitude shown in Fig. 1, acidosis slowed the rate of decline of the Ca2+ transient; the traces show normalized and superimposed Ca2+transients recorded at control pH and after 5-min exposure to a HEPES-based acidosis. The declining phase of the Ca2+ transient could be fitted by a single exponential (Fig. 2 A), in agreement with previous studies (see, e.g., Ref. 4). Figure2 B shows the mean (± SE) time constants (τ) of the decline of the Ca2+ transient obtained from such monoexponential fits, which show that τ increased significantly (P < 0.05) during acidosis (from 247.71 ± 6.64 ms in control to 365.57 ± 22.9 ms during acidosis; n = 7 cells) and that this prolongation was reversed on returning to control pH.
To investigate the effect of pHialone on the decline of the Ca2+transient, intracellular acidosis at constant pHo was produced by the application and subsequent withdrawal of 15 mmol/l NH4Cl (seematerials and methods). Figure2 D shows normalized and superimposed Ca2+ transients obtained at control pH and 1 min after withdrawal of NH4Cl. Figure2 E shows that, as for the HEPES-based acidosis, the Ca2+ transient was prolonged during intracellular acidosis; τ increased significantly (P < 0.05) from 245.17 ± 15.11 ms in control to 355.0 ± 26.84 ms during acidosis (n = 6 cells). These changes were reversible and were not significantly different from those observed during the HEPES-induced acidosis.
These data agree with previous reports that acidosis prolongs thet ½ (and hence, by implication, τ) of the Ca2+transient and that this is caused primarily by a decrease of pHi. However, it has been suggested that the rate of decline of the Ca2+ transient depends on the range of free [Ca2+]iover which it is measured, and hence the amplitude of the transient (4). Because the amplitude of the Ca2+ transient increases during acidosis (see Fig. 1 and introduction), the observed changes int ½ or τ may reflect the change in amplitude rather than changes in the activity of the different Ca2+-flux pathways responsible for Ca2+ removal from the cytoplasm. The effect of acidosis on the rate of decline of [Ca2+]iat a given [Ca2+]iwas therefore investigated. The exponential curve fitted to the declining phase of the Ca2+transient was differentiated to give the rate of change of Ca2+(δFr/δt) and plotted against the undifferentiated curve data (Fr).
Figure 2 C shows such a plot for the declining phase of representative Ca2+ transients at control pH and during a HEPES-based acidosis (extracellular + intracellular acidosis), and Fig. 2 F shows similar plots from Ca2+ transients obtained at control pH and during an NH4Cl-induced acidosis (intracellular acidosis). The exponential curves fitted to the original data are shown. During both types of acidosis, δFr/δtwas smaller at any given Fr, showing that acidosis slows the decline of [Ca2+]iat a given [Ca2+]i. Similar results were obtained in five cells. Thus the slower decline of the Ca2+ transient observed during acidosis does not appear to depend on the increased [Ca2+]i.
Because both types of acidosis had a similar effect, this suggests that the slower decline of [Ca2+]imay be caused predominantly by a mechanism affected by pHi. Because the SR appears to be the major Ca2+-sequestration pathway in rat ventricular myocytes (3) and would be expected to be influenced by pHi (see introduction), the role of the SR in the observed slowing of Ca2+ removal from the cell cytoplasm was investigated by studying the effect of acidosis on Ca2+ removal from the cell cytoplasm in the absence of a functional SR.
Effect of acidosis on decline of [Ca2+]iin presence of caffeine.
Application of 10 mmol/l caffeine evokes a large Ca2+ transient because of Ca2+ release from the SR. Under these conditions, the calcium released by the SR appears to be removed from the cytoplasm by non-SR pathways (see, e.g., Ref. 22).
Cells were electrically stimulated at 0.5 Hz while being perfused with either control or acid solution. Stimulation was then stopped for 2 s before the same Tyrode solution (at either pH 7.4 for control or 6.5 for acidosis), but containing 10 mmol/l caffeine, was rapidly applied to the cell (see materials and methods); application of caffeine was continued until the transient increase in [Ca2+]ievoked by caffeine returned to baseline (∼20 s). The solution was then switched back to control, and stimulation was restarted. Figure3 shows results from a typical experiment, showing both electrically stimulated and caffeine-induced Ca2+ transients recorded at control pH (Fig. 3 A) and during a HEPES-based acidosis (Fig. 3 B). The caffeine-induced Ca2+ transient declined more slowly than the electrically stimulated Ca2+ transients, presumably because SR Ca2+ uptake was inhibited by caffeine. Acidosis significantly increased the amplitude of the caffeine-induced rise of [Ca2+]i,from 1.34 ± 0.078 ratio units in control to 2.67 ± 0.29 ratio units during acidosis (P < 0.001; n = 6 cells) and significantly slowed its decline (Fig. 3 C). The time constant of decline of the caffeine-induced Ca2+ transient increased from 1.84 ± 0.17 s at control pH to 3.23 ± 0.39 s during acidosis (n = 6 cells,P < 0.05); this slowing was reversed on returning to control pH.
Because acidosis altered the amplitude, as well as the time course, of the caffeine-induced rise of [Ca2+]i, it seemed possible that the measured rate of decline of [Ca2+]iwas being modulated by the change in amplitude (see Effect of acidosis on rate of decline of Ca2+ transient). Therefore, the rate of decline of [Ca2+]iat a given [Ca2+]iin the presence of caffeine was determined as described above. Figure3 D shows such plots from a representative experiment at control pH and during a HEPES-based acidosis, showing that δFr/δtwas consistently smaller for a given Fr during a HEPES-induced acidosis, consistent with acidosis slowing the rate of decline of [Ca2+]iat a given [Ca2+]iin the presence of caffeine. This suggests that acidosis can slow the rate of removal of Ca2+ from the cell cytoplasm by a route other than the SR. To determine whether the site of action for this mechanism was extracellular or intracellular, the effect of NH4Cl on the rate of decline of the caffeine-induced Ca2+ transient was investigated.
Figure 4 shows data from a representative experiment, showing both electrically stimulated and caffeine-induced Ca2+ transients recorded at control pH (Fig. 4 A) and during an NH4Cl-induced acidosis (Fig.4 B). Acidosis significantly (P < 0.05) increased the amplitude of the caffeine-induced Ca2+release from 1.52 ± 0.14 ratio units in control to 2.13 ± 0.31 ratio units during acidosis (n = 6 cells). However, although this type of acidosis slowed the time course of the electrically stimulated Ca2+ transients (compareleft panels of Fig. 4,A andB), intracellular acidosis had no significant effect on the rate of decline of the caffeine-induced Ca2+ transient, as shown in Fig.4 C, which shows normalized and superimposed records of the caffeine-induced Ca2+ transient at control pH and during acidosis: τ was 2.1 ± 0.12 s in control and 1.96 ± 0.21 s during acidosis (n = 6 cells; not significant). Figure 4 D shows that intracellular acidosis had little effect on the relationship between δFr/δtand Fr. Thus, although intracellular plus extracellular acidosis decreased the rate of decline of Ca2+ in the presence of caffeine, intracellular acidosis alone had no significant effect. This difference does not appear to be caused by differences in the amplitude of the caffeine-induced release, because the difference is observed at a given [Ca2+]i(cf. Figs. 3 D and4 D).
These data suggest that the effect of intracellular acidosis on the decline of [Ca2+]i(Fig. 2) is brought about via the SR, because in the absence of a functional SR intracellular acidosis has little effect on the decline of [Ca2+]i(Fig. 4). It appears that acidosis can also slow the decline of [Ca2+]iby acting at an extracellular site, because in the presence of caffeine intracellular plus extracellular acidosis slowed the rate of decline of [Ca2+]i(Fig. 3), whereas intracellular acidosis alone had no effect on the rate of decline of [Ca2+]i(Fig. 4).
Effect of acidosis on Ca2+ release from SR.
The data reported above suggest that acidosis can slow the decline of [Ca2+]iby inhibiting Ca2+ sequestration by the SR. However, inhibition of Ca2+ uptake will not only affect the decline of the Ca2+ transient but may also have important consequences on the Ca2+ load of the SR and subsequent release. To investigate whether acidosis also affects SR Ca2+ release, the size of the electrically stimulated Ca2+transient was compared with the amplitude of the caffeine-induced release at control pH and during acidosis, as an index of the fraction of the SR Ca2+ content that is released in response to electrical stimulation (2).
The amplitude of the caffeine-induced Ca2+ transient was significantly (P < 0.05,n = 6 cells) larger than the preceding electrically stimulated Ca2+transient, suggesting that not all of the Ca2+ stored in the SR is released during a twitch (see Figs. 3 and 4). Figure5 shows that at control pH, the ratio of Ca2+ transient amplitude to caffeine-induced release amplitude (fractional release) was 0.74 ± 0.082, suggesting that ∼74% of the Ca2+ in the SR was released during each Ca2+ transient [a similar value was obtained when rapid cooling was used to elicit Ca2+ release from the SR (not shown) when a ratio of 0.77 ± 0.079 (n = 6 cells) was obtained].
Figure 5 A shows that an extracellular plus intracellular (HEPES) acidosis significantly (P < 0.05) decreased this ratio to 0.53 ± 0.054 (n = 6 cells), suggesting that under these conditions only 53% of the SR Ca2+ content was released during electrical stimulation. Figure 5 Bshows similar data for an intracellular (NH4Cl induced) acidosis, also showing a significant (P < 0.05) decrease in this ratio, to 0.52 ± 0.07 (n = 6), which was not significantly different from that obtained during a HEPES-based acidosis. These changes in fractional release were reversed on returning to control pH. It proved impossible to elicit Ca2+ release from the SR by rapid cooling during either type of acidosis so that similar data could not be obtained by releasing Ca2+ from the SR using this technique.
Thus acidosis appears to decrease the fraction of the SR Ca2+ content that is released in response to electrical stimulation from ∼0.75 to ∼0.5. There are three possible explanations for this result. First, because changes in SR Ca2+ content are known to alter fractional release (2), it is possible that an acidosis-induced change in SR Ca2+ content might underlie the observed change in fractional release. However, it appears that acidosis increases SR Ca2+ content (as assessed using caffeine, see, e.g., Fig. 3; see also introduction), which would be expected to increase fractional release (2). Second, acidosis might decrease the trigger for Ca2+ release from the SR, which in the rat under the conditions of the present experiments, appears to be predominantly Ca2+ entry via the L-type Ca2+ current (9). Third, acidosis might inhibit the coupling betweenI Ca and the amount of Ca2+ released from the SR, so that a smaller fraction of the available Ca2+ is released for a given trigger (10, 18). To distinguish between the latter two possibilities in intact myocytes, the effect of acidosis onI Ca has been investigated in ventricular myocytes using the perforated patch-clamp technique.
Figure 6 Ashows a plot of the amplitude ofI Ca, recorded during 200-ms voltage-clamp pulses to 0 mV from a holding potential of −40 mV, throughout a 5-min exposure to a HEPES-based acidosis. Fast time base records ofI Ca at control pH and after 5-min exposure to acidosis from a representative cell are shown superimposed in Fig. 6 A,right. In contrast to previous studies (see, e.g., Refs. 16, 19), under these conditions acidosis had no significant effect on this current. This is shown in Fig.6 B, which shows mean (±SE;n = 5 cells) current-voltage relationships forI Ca at control pH and during acidosis, demonstrating that intracellular plus extracellular acidosis has no significant effect on the amplitude ofI Ca at any test potential.
It appears unlikely, therefore, that acidosis is decreasing fractional release either by an effect on SR Ca2+ content or by decreasingI Ca. It appears more likely that the decreased fractional release is caused by an acidosis-induced inhibition of the CICR mechanism.
The data in the present study show that acidosis has marked effects on SR Ca2+ uptake and release in isolated ventricular myocytes. However, before these data are considered further, the two types of acidosis used in this study should be considered in more detail. An acidic HEPES-buffered solution was used to decrease both pHo and pHi. The rate of decrease of pHo using this type of acidosis is determined by the bath volume and perfusion rate and is, therefore, rapid (see materials and methods). pHi decreases to a new steady state within 2 min with the use of this type of acidosis under our experimental conditions (15), so that when our measurements were taken, after 5 min, pHi had reached a new steady state (15). The decrease of pHi produced by this method is 20–40% of the pHo change (6), so that in the present study pHi would decrease from its normal value of ∼7.1 to ∼6.8. This decrease of pHi is associated with an increase of intracellular [Na+] of ∼2 mmol/l as a consequence of activation of Na+/H+exchange by the intracellular acidosis (6), although this mechanism will be inhibited to some extent by the decrease of pHo (28). The protocol used to decrease pHi alone produces a similar decrease of pHi to HEPES-based acidosis (6), albeit more rapidly than HEPES-based acidosis, so measurements were taken 1 min after NH4Cl withdrawal. The increase in intracellular [Na+] produced by this type of acidosis is consequently also rapid but greater than that produced by HEPES-based acidosis (6) because the absence of an extracellular acidosis results in no inhibition of the Na+/H+exchange by this route. Thus the two types of acidosis used in this study would be expected to produce similar decreases of pHi, albeit at different times after the intervention used to produce acidosis, but the HEPES-based acidosis also produces an extracellular acidosis and hence a smaller rise of intracellular [Na+].
Effect of acidosis on rate of decline of Ca2+ transient.
During an electrically stimulated Ca2+ transient, Ca2+ sequestration by the SR and transsarcolemmal Ca2+ efflux both contribute to the decline of [Ca2+]i. Although the rate of decline of the [Ca2+]itransient decreases with [Ca2+]i(see, e.g., Fig. 2, C andF), so that the declining phase can be fitted by a single exponential, this is the sum of the Ca2+ handling pathways in the cell (4), so that the time constant of this exponential decline depends on the range of [Ca2+] over which it is measured (4). However, the present study shows that although acidosis increased the amplitude of the Ca2+ transient, it slowed its rate of decline, even at a given [Ca2+]. This slowing cannot, therefore, be accounted for simply by the change in the amplitude of the Ca2+ transient (4) and so must reflect changes in cellular Ca2+ handling.
In the present study, an NH4Cl-induced (intracellular) acidosis slowed the rate of decline of the electrically stimulated Ca2+ transient (Fig. 2), suggesting that at least one of the pathways that remove Ca2+ from the cytoplasm was inhibited by intracellular acidosis. When the SR was inhibited by caffeine, the decline of [Ca2+]iwas not significantly altered by an NH4Cl-induced acidosis (Fig. 4), suggesting that the slowed decline of the electrically stimulated Ca2+ transient is caused by a decreased rate of Ca2+ uptake by the SR.
There are two main mechanisms whereby intracellular acidosis may affect the SR to slow the decline of [Ca2+]i. First, acidosis may directly inhibit the SR Ca2+-ATPase, thus decreasing Ca2+ uptake by the SR. In support of this hypothesis, it was previously shown in skinned cardiac preparations and isolated SR vesicles that SR Ca2+ uptake is reduced during acidosis (11, 20). Second, Ca2+uptake into the SR may be altered if the phosphorylation of phospholamban, the protein that regulates the Ca2+-ATPase (26), is affected either directly or indirectly by changes in pHi. However, it was previously shown that acidosis has no net effect on the phosphorylation of phospholamban, making this explanation unlikely (15, 21).
During exposure to a HEPES-based acidosis, which reduces both pHo and pHi, the decline of the electrically stimulated Ca2+transient was slowed to a extent similar to that observed during NH4Cl withdrawal; it seems likely that inhibition of SR Ca2+ uptake as a consequence of the intracellular acidosis played a role in the slower decline. However, unlike NH4Cl withdrawal, HEPES-based acidosis slowed the rate of decline of the caffeine-induced Ca2+ transient. Because this effect is unlikely to be caused by a decrease of pHi (because pHi alone has no effect on the rate of decline of the caffeine-induced Ca2+ transient), this suggests two other possibilities. First, the rise of intracellular [Na+] produced by acidosis could decrease the rate of Na+ influx, and hence Ca2+ efflux, on the Na+/Ca2+exchange. This seems unlikely, however, because the rise of intracellular [Na+] produced by this type of acidosis will be smaller than that produced by intracellular acidosis alone. The second possibility is that acidosis is slowing Ca2+ extrusion by acting at an extracellular site; there are a number of possible sites. Extracellular H+ may inhibit Ca2+ extrusion via Na+/Ca2+exchange, because when the SR is inhibited by caffeine the exchanger appears to be the main mechanism for Ca2+ removal from the cytoplasm (3), and it has recently been reported that extracellular acidosis can inhibit Na+/Ca2+exchange current (8). Another possibility is that extracellular H+ inhibits the sarcolemmal Ca2+-ATPase. However, little is known about the pH sensitivity of this ATPase in cardiac muscle. Because this pump exchanges 1 Ca2+for 2 H+, if extracellular [H+] increases, it might be expected that influx of H+ would be favored, thus extruding more Ca2+ from the cytoplasm.
It was surprising that intracellular acidosis alone did not inhibit the rate of decline of the caffeine-induced Ca2+ transient, because acidosis has been reported to inhibit Na+/Ca2+exchange at an intracellular site on the exchange protein (7) and the rise of intracellular [Na+] produced by acidosis would also be expected to inhibit Ca2+ efflux on the exchanger (27). The reason why such inhibitory effects were not observed is unknown.
Effect of acidosis on Ca2+ transient amplitude.
In agreement with previous reports (see introduction), acidosis increased the amplitude of the Ca2+ transient. This increase was previously shown to be inhibited by inhibitors of the SR (14, 23), suggesting that increased Ca2+release from the SR underlies the increase in the size of the Ca2+ transient (14, 23). Because the direct effects of acidosis on the SR would tend to inhibit Ca2+ uptake and release, it has been suggested that increased SR Ca2+ release occurs because the rise in cytoplasmic Ca2+ that occurs during acidosis increases the Ca2+ available to the Ca2+-ATPase and that this more than compensates for the direct inhibitory effect of acidosis on SR Ca2+ uptake, thus increasing SR Ca2+ content and hence release (14). The present observation that acidosis increases the apparent SR Ca2+ content, assessed using caffeine, is consistent with this suggestion.
Despite an apparent increase in SR Ca2+ content, acidosis decreased the fraction of Ca2+ released from the SR. It is unlikely that this can be explained by changes in cytoplasmic Ca2+ buffering, because an acidosis-induced decrease in buffering power would be expected to increase both the electrically induced and caffeine-induced Ca2+ transients so that the ratio of these two (fractional release) would not be expected to change. There are, however, a number of alternative explanations.
First, because changes ofI Ca alter fractional Ca2+ release from the SR (2), it seemed possible that an acidosis-induced decrease inI Ca (see, e.g., Refs. 16, 19) caused the observed decrease in fractional release. This seems unlikely, however, because in the conditions of the present study, acidosis had no significant effect onI Ca (Fig. 6). This was surprising because several previous studies showed a decrease in I Ca during acidosis (see, e.g., Refs. 16, 19). The reasons for this discrepancy are unknown, although there are several possibilities. The effect of acidosis on I Cawas previously studied predominantly in ventricular myocytes from the guinea pig using the whole-cell patch clamp technique and high concentrations of intracellular Ca2+ buffers to eliminate changes in [Ca2+]i. In contrast, in the present study the effect of acidosis onI Ca was investigated in rat ventricular myocytes using the perforated patch-clamp technique to minimize intracellular dialysis and in the absence of intracellular Ca2+buffers, allowingI Ca to be recorded under more physiological conditions. One possible explanation, therefore, is that the acidosis-induced increase in intracellular [Ca2+] may alterI Ca in addition to effects of acidosis onI Ca mediated by other mechanisms. For example, if the acidosis-induced increase in [Ca2+]iactivates Ca2+-calmodulin-dependent protein kinase (15), this would increaseI Ca (30) and could offset direct inhibitory effects of acidosis onI Ca. Thus, in previous studies in which [Ca2+]iwas buffered, only the direct inhibitory effect of acidosis onI Ca would be observed, whereas in the present study this effect would by offset by a Ca2+-calmodulin-dependent increase in I Ca, resulting in little overall change inI Ca. The present data may, therefore, represent a more physiological response ofI Ca to acidosis.
It also appears unlikely that acidosis decreases fractional release by a decrease in SR Ca2+ content, because it appears that SR Ca2+content increases during acidosis and this would be expected to increase fractional release (2).
A third explanation is that acidosis directly inhibits the coupling between I Ca and SR Ca2+ release, so that a smaller fraction of the SR Ca2+ content is released for a givenI Ca (10, 18). Such an inhibitory effect of acidosis on CICR could explain why a smaller fractional release was observed during acidosis in the present study. The molecular mechanism by which acidosis inhibits CICR release is unknown, but decreased Ca2+binding to the activation site on the ryanodine receptor and/or a direct inhibitory effect of acidosis on the ryanodine receptor protein could be postulated (25).
These data suggest, therefore, that in the intact cell acidosis1) inhibits Ca2+ uptake by the SR by a direct effect on the SR Ca2+-ATPase and2) decreases fractional release of Ca2+ from the SR by an inhibitory effect on CICR, yet 3) increases the size of the Ca2+ transient for a given I Ca, i.e., increases the “gain” of the Ca2+ release process. This is unlikely to be caused by an increase in single-channel current (29), because acidosis decreases single-channel current through the L-type Ca2+ channel (17). To reconcile these changes and to account for the increase in the amplitude of the Ca2+ transient during acidosis, the following hypothesis is proposed. Acidosis increases cytoplasmic Ca2+ sufficiently to overcome the direct inhibitory effects of acidosis on SR Ca2+ uptake, so that the Ca2+ content of the SR increases. This increase in Ca2+ content increases the amount of Ca2+released by the SR, even thoughI Ca does not change and the fractional release decreases as a consequence of the inhibitory effects of acidosis on CICR. This is illustrated in Fig.7, which shows the observed changes in SR Ca2+ content, assessed using caffeine, and fractional release during HEPES-based acidosis. If it is assumed that the caffeine-induced Ca2+ transient reflects the amount of releasable Ca2+ in the SR, ∼75% of this Ca2+ was released in response to electrical stimulation at control pH 7.4, but this fraction was reduced to ∼50% during acidosis. However, the total amount of releasable Ca2+ stored within the SR approximately doubled. Thus, even though the fraction of Ca2+ release during acidosis was decreased, the actual amount of Ca2+ released during a twitch is still greater than control.
In conclusion, these data show that in the intact cell acidosis inhibits SR function by inhibiting Ca2+ uptake and CICR, in agreement with previous data obtained in skinned cells and isolated preparations of the SR. However, in the intact cell these effects appear to be modulated by an increase in SR Ca2+ content brought about by an increase in the Ca2+ available to the Ca2+-ATPase, which tends to increase the amount of Ca2+released from the SR. Thus in the intact cell it appears likely that the amount of Ca2+ released by the SR will depend on the balance between these effects.
The authors are grateful to the British Heart Foundation for financial support. J. T. Hulme is a University of Leeds research scholar.
Address reprint requests to C. H. Orchard.
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. §1734 solely to indicate this fact.
- Copyright © 1998 the American Physiological Society