INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THERE HAS BEEN A GREAT DEAL of interest recently in the effects of halides and other anions on the function of the sarcoplasmic reticulum (SR). It has been reported, for example, that some anions including Cl and I can increase the Ca²⁺ permeability of the skeletal muscle ryanodine receptor over that measured in the presence of organic anions such as propionate (2, 8, 25, 26, 29). A Cl-sensitive nonselective channel that allows Ca²⁺ to permeate the SR membrane has also been postulated to exist in skeletal muscle (39). Other investigators (37) suggested that the SR Ca²⁺ pump itself can be directly inhibited by I and SCN. In addition to affecting these mechanisms, the anionic environment of the SR could also influence the movement of compensatory charge across the SR membrane during Ca²⁺ uptake or release. That such movement occurs has been inferred primarily from the following two types of observations (reviewed in Refs. 3, 7, 22, 23, 30, and 40): 1) although the Ca²⁺-ATPase of the SR of muscle cells is believed to be electrogenic, a sustained membrane potential has not been detected across the SR membrane; and 2) both anion and cation channels are found in SR membranes. Thus other ions are thought to cross the membrane to balance charge as Ca²⁺ is released from the SR or actively transported into the SR. In support of this mechanism in smooth muscle, we (33) have shown that SR Ca²⁺ uptake is inhibited in the presence of the Cl channel inhibitors 5-nitro-2(3-phenylpropylamino)benzoic acid (NPPB) or indanyloxyacetic acid 94 [R(+)-IAA-94] and by the substitution of SO for Cl in the medium outside the SR. In a recent work (14), we have shown that cardiac SR Ca²⁺ uptake is inhibited by the Cl channel blocker tamoxifen. The latter effect occurs without an alteration in the ATPase activity of the SR Ca²⁺ pump and without a significant change in the permeability of the SR membrane to Ca²⁺.

In the work reported here, we studied Ca²⁺ uptake into cardiac SR vesicles when I, Br, or SO were substituted for extravesicular Cl. SR Ca²⁺ uptake was not inhibited by the substitution of SO for Cl. Net Ca²⁺ uptake rate was significantly reduced, however, when Br or I was substituted for Cl. This occurred without a detectable effect of I or Br on the ATPase activity of the SR Ca²⁺ pump and without an increase in the Ca²⁺ permeability of the SR membrane.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Preparation of SR vesicles. Animals were euthanized with an overdose of pentobarbital sodium in accordance with procedures approved by the Canadian Council on Animal Care and the University of Calgary Faculty of Medicine Animal Care Committee. SR vesicles were prepared from canine ventricular tissue as previously described (15) according to the method of Chamberlain et al. (1) omitting the sucrose density gradient centrifugation step.

Measurement of SR Ca²+ uptake. The fluorometric method for measuring Ca²⁺ uptake into SR vesicles using fura 2 was described previously (15, 16, 18). Briefly, SR vesicles were first diluted to a total protein concentration of 1 mg/ml in uptake buffer containing 100 mM KX (where X is Cl, I, or Br), 4 mM MgX₂, and 20 mM HEPES (pH 7.0); or 50 mM K₂SO₄, 4 mM MgSO₄, and 20 mM HEPES. Vesicle samples were then added to a 4-ml cuvette containing 2 ml of uptake buffer. For most experiments, oxalate was included in uptake buffer to act as a trapping ion for Ca²⁺ within the SR (28). When oxalate was not included, a higher vesicle protein concentration and other adjustments to the uptake buffers were required to increase measurable SR Ca²⁺ uptake. Differences in the uptake buffers that were used in experiments done in the presence and absence of oxalate are shown in Table 1. When tetraethylammonium (TEA; 10 mM) was added to either KI or KCl uptake buffer, the K⁺ concentration of the buffers was reduced by 10 mM to maintain ionic strength. Uptake was initiated by the addition of Ca²⁺ to the cuvette. Fluorescence was measured with a SPEX fluorimeter (CMX model; Edison, NJ). Fura 2 was excited at 340- and 380-nm wavelengths, and fluorescence emission was measured through a 510-nm band-pass filter (10-nm bandwidth). Fluorescence ratios (340-380 nm) were obtained every 1 s. The contents of the cuvette were continuously stirred during the course of an experiment.

Determination of velocity of Ca²+ uptake. For experiments done in the presence of oxalate, the free Ca²⁺ concentration ([Ca²⁺]_free), total Ca²⁺ concentration ([Ca²⁺]_total), and instantaneous uptake velocity were determined from the fura 2 fluorescence ratio values after correction for background fluorescence and light scatter as described previously (9, 14-18). [Ca²⁺]_total was determined from [Ca²⁺]_free as described previously (14-18) using published binding constants (4, 12, 27, 36). The protein concentration of the vesicle preparations was determined by the Bradford protein assay (kit purchased from Bio-Rad; Hercules, CA). The maximum velocity of Ca²⁺ uptake was determined from plots of uptake velocity versus [Ca²⁺]_free as described by Kargacin and Kargacin (15).

1

1

Measurement of SR Ca²+ release. For Ca²⁺ release experiments, the uptake buffers described above were used except that 5 mM D-glucose was included as a substrate for hexokinase and creatine phosphate (CP) and creatine phosphokinase (CPK) were omitted from the uptake buffer. After the SR was loaded and net Ca²⁺ movement into the SR stopped, hexokinase (4.7 U/ml) was added to the cuvette to deplete the buffer of ATP to allow passive Ca²⁺ release to be measured.

Measurement of ATPase activity of SR Ca²+ pump using NADH fluorescence. An enzyme-coupled assay (19) in which ATP hydrolysis by the SR Ca²⁺ pump is coupled to the conversion of NADH to NAD⁺ was adapted to determine whether substitution of Br or I for Cl directly inhibited the SR Ca²⁺ pump. The ATPase activity of the pump was measured in a 4-ml cuvette in buffers containing 100 mM KX (where X is Cl, I, or Br), 4 mM MgX₂, 0.15 mM NADH, 0.21 U/ml pyruvate kinase, 0.46 mM phospho(enol)pyruvate, 2.2 U/ml lactate dehydrogenase (LDH), 1.1 mM ATP, 3.3 µM 4-Br-A23187, and 20 mM HEPES (pH 7.0; [Ca²⁺]_free 3 µM). The Ca²⁺ ionophore 4-Br-A23187 was included in the buffer to prevent the ATPase activity of the SR Ca²⁺ pump from being influenced by the reduction of extravesicular Ca²⁺ or by the establishment of a Ca²⁺ gradient across the SR membrane (discussed in Ref. 5). Solutions in the cuvette were continuously stirred throughout an experiment. Background fluorescence measurements (made with vesicles and all components except NADH in the cuvette) and measurements of the change in NADH fluorescence (made with all components except ATP in the cuvette) were used to correct the measurements of the changes in NADH fluorescence due to the ATPase activity of the SR Ca²⁺ pump. Calibration curves for NADH fluorescence were determined (Fig. 1A) by adding known amounts of NADH to KCl, KBr, or KI buffer. There were no effects of I or Br on NADH fluorescence. The calibration curves (which deviate from linearity due to the inner filter effect; see Ref. 24) were fit by the equation

(1)

where F is the fluorescence intensity in counts per second and a and b are constants. For SR vesicles, the change in NADH content in the sample was linear (see Fig. 5); the rate of change of NADH was therefore determined from linear regression for a 100- to 200-point (50-100 s) segment of the data. For the assay, the sample sizes ranged from 30-190 µg vesicle protein depending on the SR vesicle preparation used. To rule out the possibility that the ATPase assay itself was rate limiting, control experiments were done to show that doubling the amounts of pyruvate kinase and LDH in the assay did not change the Ca²⁺-ATPase rates measured by the assay. The ATPase activity of the preparation (when expressed as µmol · min¹ · mg¹ vesicle protein) was also not altered when the amount of vesicle protein was doubled.

View larger version (15K): [in this window] [in a new window]   Fig. 1.   NADH-based ATPase assay. A: calibration of NADH fluorescence. Fluorescence intensity as a function of NADH concentration ([NADH]) in KCl () and KI () buffer (excitation 350 nm; emission 450 nm) after subtraction of background fluorescence. Lines were drawn according to Eq. 1 with the constants a = 18.3 × 104 and b = 48.46 for KCl buffer (solid line) and a = 17.8 × 104 and b = 45.74 KI buffer (dotted line). B and C: inhibition of the NADH fluorescence ATPase assay by Br or I. The response of the NADH fluorescence assay to the addition of ADP to KCl, KBr, or KI buffer was determined. B: NADH as a function of time after the addition of 145 µM ADP to KCl buffer (circles) and KI buffer (squares). Experiments were done in a 4-ml cuvette containing 2 ml of buffer containing the reagents for the assay (see text). The amount of NADH (in nmol) was determined from fluorescence using the calibration curves in A and the sample volumes in the cuvette. In these experiments, the maximum rates of response of the ATPase assay (solid lines) were determined from line fits to the first 100 data points (50 s; solid symbols) after the addition of ADP. The maximum rate of response of the assay to the addition of ADP in KI buffer was 31.3% of that measured in KCl buffer, indicating that I inhibited the assay itself. Note: only every 4th data point is plotted for clarity. C: summary of results of experiments similar to that shown in B for KCl (n = 12), KBr (n = 6), and KI (n = 6) buffers. Response rates, expressed as percentage of control, were 65.2% in KBr and 29.4% in KI and were significantly different from the rate in KCl (100%) at P = 0.0009 and P = 2.3 × 107, respectively. Error bars are +1 SD.

Measurement of ATPase activity of SR Ca²+ pump using purine-nucleoside phosphorylase. A second spectrophotometric assay (43, 44) in which the inorganic phosphate (P_i) generated by the ATPase activity of the SR Ca²⁺ pump was measured was also used to determine whether substitution of I for Cl had a direct inhibitory effect on the SR Ca²⁺-ATPase. In the assay, purine-nucleoside phosphorylase (PNP) converts 2-amino-6-mercapto-7-methylpurine riboside (MESG) and P_i to ribose 1-phosphate (R1-P) and 2-amino-6-mercapto-7-methypurine (AMM)

(2)

As the SR Ca²⁺ pump hydrolyzes ATP and generates P_i, conversion of MESG (maximum absorbance at 330 nm) to AMM (maximum absorbance at 360 nm) results in an absorbance increase at 360 nm that is proportional to the change in P_i. The generation of P_i by turnover of the cardiac SR Ca²⁺ pump was measured in buffer containing 100 mM KX (where X is Cl or I), 4 mM MgX₂, 1.1 mM ATP, 1 U/ml PNP, 150 µM MESG, and 20 mM HEPES (pH 7.0; [Ca²⁺]_free 3 µM). 4-Br-A23187 (3.3 µM) was included in the buffer for the reasons discussed above. Measurements were made in a 4-ml cuvette with an ultraviolet/visible spectrophotometer (model Lambda 3B, Perkin-Elmer; Norwalk, CT); solutions in the cuvette were continuously stirred during the experiments. The assay system was calibrated by adding known amounts of P_i to the buffers. The calibration curves in Fig. 2A show that the magnitude of the change in absorbance for a given change in P_i was slightly less in KI buffer than it was in KCl buffer. The rate at which the assay responded to step changes in P_i was also compared in KCl and KI buffers. Figure 2B shows that the response of the assay to step changes in P_i was slightly slower in the KI buffer. The mean initial rate of response in KI buffer (determined as shown in Fig. 2B) for a variety of step changes and starting P_i concentrations was 79.4 ± 4.6% (n = 7) of that measured in KCl buffer. Figure 2C shows that the rate at which the enzyme system could respond to changes in P_i was not limited by the rate at which solutions could be stirred into the cuvette.

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Calibration of the purine-nucleoside absorbance assay. A: absorbance at 360 nm as a function of Pi in KCl () and KI (). Slopes of the lines drawn from line fits to the data were 0.01277 absorbance units/[Pi] ([Pi] in µM) in KCl and 0.01225 absorbance units/[Pi] ([Pi] in µM) in KI. B: rate of response of the assay after step changes in Pi in KCl (circles) or KI (squares) buffer. Equal amounts of Pi in H2O were added to a 4-ml cuvette containing 3.3 ml KCl or KI buffer and 137 µM 2-amino-6-mercapto-7-methylpurine riboside (MESG), 1 U/ml purine-nucleoside phosphorylase (PNP), and 3.0 µM 4-Br-A23187. Pi as a function of time was determined from the calibration curves in A. The maximum rate of response (determined from line fits to the points shown by closed symbols) in KI buffer was 80.8% of that measured in KCl buffer. Note: data collected during the first 4 s (the time required for mixing; see C) after the addition of Pi to the cuvette was not included in the calibration. C: time required for mixing. The time required for solutions added to the cuvette to mix was determined by adding fixed amounts of reaction product (2-amino-6-mercapto-7-methylpurine) to a cuvette containing 3 ml of KCl buffer. It required a mean time of 3.5 s for the assay system to completely respond to a step change in absorbance resulting from the addition of reaction product. D: generation of Pi by the cardiac muscle sarcoplasmic reticulum (SR) vesicle preparation. Pi generation (circles) was measured in a 4-ml cuvette containing 2 ml of KCl buffer, 30 µg of SR vesicles, 3.3 µM 4-Br-A23187, 150 µM MESG, 1 U/ml PNP, and 1.5 mM ATP; free Ca2+ concentration ([Ca2+]free) was ~3 µM. The solid line is the line fit to a 100-point segment () of the data. The slope of the line (0.199 nmol Pi/s) corresponds to an ATPase rate of 0.398 µmol Pi · min1 · mg1. Note: only every 4th data point is plotted for clarity.

Reagents. ADP (potassium salt), K₂ATP, CP, CPK, NADH, phosphoenol pyruvate, pyruvate kinase, dithiothreitol, 4-aminopyridine, D-glucose, and histidine were purchased from Sigma Chemical (St. Louis, MO). Aristar grade KCl, KOH, and sucrose, Suprapur KI, and AnalaR KBr were purchased from BDH (Edmonton, AB, Canada). Hexokinase was purchased from Roche Diagnostics (Laval, QC, Canada). Microselect MgCl₂ · 6H₂O and MgBr₂ · 6H₂O, puriss grade oxalic acid and MgSO₄ · 7H₂O, purum grade TEA hydroxide, and MgI₂, ruthenium red, and HEPES (potassium salt) were purchased from Fluka (Ronkonkoma, NY). LDH was purchased from Worthington (Freehold, NJ). Fura 2 free acid, 4-Br-A23187, and the reagents for the purine-nucleoside phosphorylase assay (ENZCheck kit) were purchased from Molecular Probes (Eugene, OR).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Effect of anion substitutions on Ca²+ uptake in SR vesicles in presence of oxalate. Substitution of K₂SO₄ for KCl in the uptake buffer did not significantly change the rate at which Ca²⁺ was taken up into the SR in the presence of oxalate. A comparison of uptake rates in K₂SO₄ and KCl buffers is shown in Fig. 3A. The uptake rate in KBr uptake buffer was less than that measured in KCl uptake buffer (Fig. 3B) and was slower still in KI uptake buffer (Fig. 3C). For the experiments shown, the maximum uptake rate was 71.9% of control in KBr and 44.4% of control in KI buffer. Consistent with the results shown in Fig. 3, A-C, the results of several similar paired experiments showed that the maximum velocity was significantly lower in KBr (69.8 ± 4.6% in KBr, n = 7; 100.0 ± 7.8% in KCl, n = 7) and KI (39.3 ± 2.6% in KI, n = 6; 100.0 ± 13.4% in KCl, n = 5) buffers. Uptake velocity was not significantly different from control in K₂SO₄ buffer (94.4 ± 8.8% in K₂SO₄, n = 5; 100.0 ± 13.4% in KCl, n = 8). Figure 3D shows that the inhibitory effects of Br and I on uptake velocity were concentration dependent.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Anion substitution experiments. A: SR Ca2+ uptake is shown as [Ca2+]free vs. time for 25-µg cardiac vesicle samples in KCl uptake buffer () and K2SO4 uptake buffer (). The maximum uptake rates (determined as described in text) in the Cl and SO buffers were 0.354 and 0.339 µmol · min1 · mg1, respectively. B: as in A, showing uptake in KCl uptake buffer () and KBr uptake buffer (). Maximum uptake rates for the curves shown were 0.384 and 0.276 µmol · min1 · mg1 in the presence of Cl and Br, respectively. C: as in A, showing uptake in KCl uptake buffer () and KI uptake buffer (). Maximum uptake rates for the curves shown were 0.396 and 0.176 µmol · min1 · mg1 in Cl and I, respectively. Note: in A-C, only every 4th data point was plotted for clarity. D: maximum Ca2+ uptake rate as a function of Br or I concentration. Maximum Ca2+ uptake rates for cardiac SR vesicles in uptake buffers with different relative amounts of KBr and KCl () and maximum uptake rates obtained with different relative amounts of KI and KCl () are shown. Maximum uptake rates are given as percentage of the average value for maximum uptake rate measured in the control buffer (108 mM Cl:0 mM Br or I). Relative to the control uptake rate (100 ± 2.8%; n = 12), maximum uptake rate in the presence of 54 mM Br:54 mM Cl was 96.1 ± 5.0% (n = 6) of control; maximum uptake rate in the presence of 108 mM Br:0 mM Cl was 69.8 ± 4.6% (n = 7) of control. Relative to control uptake rate (100 ± 7.6%; n = 15), maximum uptake rate in the presence of 50 mM I:58 mM Cl was 66.0 ± 5.9% (n = 6) of control; maximum uptake rate in the presence of 100 mM I:8 mM Cl was 43.4 ± 2.2% (n = 5) of control; and maximum uptake rate in the presence of 108 mM I:0 mM Cl was 39.3 ± 2.6% (n = 6) of control. Note: the open and solid symbols for the control experiments were slightly offset from one another horizontally to increase clarity. Error bars are ±1 SD; *significantly different from control at P  0.01. E: uptake in KCl and KI buffers in the absence of oxalate. Ca2+ uptake is shown as [Ca2+]free vs. time for 0.25 mg of SR protein in KCl buffer and 0.36 mg of SR protein in KI buffer. Uptake rates for these experiments were 27.8 nmol · min1 · mg1 in KCl and 11.9 nmol · min1 · mg1 in KI.

Inhibition of Ca²+ uptake by I in absence of oxalate. As noted above (see METHODS), measurement of SR Ca²⁺ uptake in the presence of oxalate allows one to determine uptake rate for a single sample over a range of physiological [Ca²⁺]_free. However, although the relative concentration of oxalate to the other anions used in the experiments described above was low (10 mM oxalate:108 mM other anions), the presence of oxalate in the KBr and KI buffers may have partially masked the full inhibitory effect of these anions on Ca²⁺ uptake. To test this possibility, the magnitude of the inhibitory effect of I (the anion that inhibited SR uptake to the greatest extent) on Ca²⁺ uptake was measured in the absence of oxalate. Figure 3E shows that in the absence of oxalate, SR Ca²⁺ uptake was still inhibited by I to approximately the same extent as it was in the presence of oxalate (Fig. 3, C and D). In experiments similar to those shown in Fig. 3E, the maximum rate of SR Ca²⁺ uptake in KI buffer was 32.0 ± 7.9% (n = 6) of the maximum rate of uptake seen in KCl. This inhibitory effect of I was not significantly different (P = 0.11) from that measured in oxalate buffer, where the rate in KI buffer was 39.3 ± 2.6% of the rate in KCl (n = 6).

Inhibitory effect of I on SR Ca²+ uptake in presence of ruthenium red. It is possible that SR Ca²⁺ uptake could be inhibited in the presence of Br or I if these anions increased the permeability of the SR membrane to Ca²⁺. The increased leak of Ca²⁺ out of the SR would reduce the net amount of Ca²⁺ moved into the SR by the Ca²⁺ pump over any time interval. This possibility is consistent with recent results indicating that the permeability of the skeletal muscle SR membrane to Ca²⁺ is greater in the presence of some inorganic anions than it is in the presence of organic anions such as propionate (2, 8, 25, 26, 29, 31, 39). These reported increases in permeability, whether they occur through the ryanodine receptor (2, 8, 25, 26, 29) or through a different SR channel (39), are blocked by ruthenium red (see Refs. 8, 26, and 39). Therefore, if Br or I inhibited net SR Ca²⁺ uptake in our experiments by increasing the permeability of the cardiac SR membrane through one of these mechanisms, one would expect ruthenium red to block this inhibition. Because the Ca²⁺ sensitivity of the cardiac SR Ca²⁺-ATPase is reduced by ruthenium red (17), Ca²⁺ uptake measurements made in KI uptake buffer containing ruthenium red were compared with control experiments done in KCl buffer with the same concentration of ruthenium red. As shown in Fig. 4A, SR Ca²⁺ uptake was reduced by I even in the presence of ruthenium red. The extent of inhibition by I was the same when ruthenium red (20 µM) was present in the uptake buffers (summarized in Fig. 4B) as it was in the absence of ruthenium red (mean uptake rate in KI with ruthenium red was 32.0 ± 6.1% of the rate in KCl with ruthenium red; mean uptake rate in KI without ruthenium red was 39.3 ± 2.6% of the rate in KCl buffer without ruthenium red).

View larger version (26K): [in this window] [in a new window]   Fig. 4.   Substitution of Cl with I does not alter the permeability of the SR membrane to Ca2+. A and B: inhibition of SR Ca2+ uptake by I in the presence of ruthenium red. A: SR Ca2+ uptake as a function of time in KCl buffer with 20 µM ruthenium red (bottom trace) and KI buffer with 20 µM ruthenium red (top trace). For this experiment, maximum uptake rate was 0.248 µmol · min1 · mg1 in KCl buffer and 0.098 µmol · min1 · mg1 in KI buffer. B: summary of results with ruthenium red. Maximum rate of SR Ca2+ uptake in KI buffer + 20 µM ruthenium red was 32 ± 6% (n = 8; crosshatched bar) of the maximum uptake rate in KCl buffer + 20 µM ruthenium red (100 ± 18%, n = 8; open bar). Error bars are +1 SD. C and D: Ca2+ release in the presence of Cl and I. C: cardiac SR samples (0.145 mg) were loaded with Ca2+ in the presence of ATP (K2ATP = 0.76 mM) in KCl () and KI () uptake buffer. Ca2+ uptake was allowed to proceed until [Ca2+]free declined to 60 nM in the cuvette. Hexokinase was then added (at 335 s for KCl; 500 s for KI) to rapidly deplete the buffer of ATP, and the rates of the resulting Ca2+ release were compared in the two buffers. D: Ca2+ release segments from the curves in C are shown for KCl () and KI () aligned on the time axis for comparison. Release rates were determined by a linear regression fit to the first 100 points of the release curves. Release rates were 2.28 × 103 µM/s in KCl (regression line shown by dashed line) and 1.86 × 103 µM/s for KI (solid line). These rates, normalized to total protein, were 1.97 nmol · min1 · mg1 in KCl buffer and 1.61 nmol · min1 · mg1 in KI buffer.

1

1

1

1

ATPase activity of SR Ca²+ pump in presence of Br or I. A second mechanism that could account for the inhibitory effects of I and Br on SR Ca²⁺ uptake is a direct effect of these ions on the SR Ca²⁺ pump. This possibility is consistent with some results in the literature (37) but not with others (11). To determine whether I or Br directly inhibited the SR Ca²⁺ pump in our experiments, two methods were used to determine the ATPase activity of the SR Ca²⁺ pump in the presence of Cl, Br, and I.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Hydrolysis of ATP by the SR Ca2+ pump in KCl, KBr, and KI buffers determined with the NADH fluorescence assay. A: NADH as a function of time monitored by the NADH fluorescence assay for 190-µg SR vesicle samples in KCl and KI buffers. Raw data obtained before correction for the effect of I on the ATPase assay are shown (solid symbols). The maximum rates of change in NADH for the raw data (red lines) were determined by line fits to the data segments between the yellow symbols. These slopes were 1.25 nmol/s in KCl buffer and 0.457 nmol/s in KI buffer. These values correspond to ATPase rates of 0.395 µmol Pi · min1 · mg1 in KCl and 0.144 µmol Pi · min1 · mg1 in KI buffer, respectively. The maximum uncorrected rate of change in NADH in KI was 36.8% of that measured in KCl buffer; this difference in rate could be entirely accounted for by the inhibition of the ATPase assay itself by I (response of the assay in KI buffer to a step change in ADP was reduced to 29.4 ± 9.8% of control; see Fig. 1). The change in NADH in KI buffer after correction of the raw data for the inhibition of the ATPase assay itself by I is shown by the blue line (the slope of the corrected KI data = 1.55 nmol/s). B: ATPase activity of the SR Ca2+ pump in KCl, KBr, and KI buffers. ATPase rates, measured for SR vesicle samples, are expressed as a percentage of the maximum rates in KCl buffer (red bars) before (open bars) and after (blue bars) correction for the effects of Br or I on the ATPase assay (n values are shown above the bars). Error bars are +1 SD. Br and I denote uncorrected results; *Br and *I denote corrected results.

1

1

1

1

View larger version (36K): [in this window] [in a new window]   Fig. 6.   Hydrolysis of ATP by the SR Ca2+ pump in KCl and KI buffers determined with the PNP (absorbance) assay. ATPase activity expressed as a percentage of the mean activity in KCl buffer (crosshatched bar) before (open bar) and after (hatched bar) correction for the effect of I on the ATPase assay itself (corrected rate in I = measured rate/0.808). The mean corrected ATPase rate of the SR Ca2+ pump in KI buffer was not significantly different from that measured in KCl (P = 0.68; n = 5 for all experiments). Error bars are +1 SD. I denotes uncorrected results in KI buffer; *I denotes corrected results.

SR uptake in presence of TEA or 4-aminopyridine. Although it has been suggested (see Refs. 22 and 23) that Cl may be the primary ion involved in compensating for the net positive charge moved into the SR by the Ca²⁺-ATPase, movement of K⁺ out of the SR lumen may also occur and could become more important under conditions in which anion movement is inhibited to the greatest extent (i.e., KI buffer). To test the possibility that efflux of K⁺ out of the SR occurred during SR Ca²⁺ uptake, uptake rates in KCl and KI buffers were compared in the presence and absence of 4-aminopyridine or TEA. These blockers were used because they are known to block skeletal muscle SR K⁺ channels from the cytoplasmic side of the SR membrane at the concentrations used in our experiments (6, 7). Figure 7 shows that 1 mM 4-aminopyridine had no significant effect on the maximum rate of SR Ca²⁺ uptake in KCl buffer or on the reduced uptake rate measured in KI buffer. The maximum rates of uptake in KCl or KI buffers were also unaffected by 10 mM TEA (results not shown).

View larger version (25K): [in this window] [in a new window]   Fig. 7.   Maximum uptake velocity in KCl or KI uptake buffers in the absence or presence of 4-aminopyridine (4-AP). Maximum uptake rate in KCl buffer in the presence of 1 mM 4-AP (hatched bar, left) was not significantly different (92.9 ± 3.5%; n = 5) from the maximum rate of uptake (100.0 ± 9.5%; n = 5) in the absence of 4-AP (open bar, left). Maximum uptake rate in buffer containing 108 mM I (open bar, right) was 36.1 ± 1.4% (n = 6) of the maximum uptake rate obtained in KCl buffer. The maximum uptake rate in KI buffer with 1 mM 4-AP (hatched bar, right) was 35.3 ± 1.8% (n = 6) of that measured in KCl buffer and was not significantly different from that in determined in the presence of I without 4-AP (36.1 ± 1.4%) at P  0.05. Error bars are +1 SD.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Table 2 summarizes the results of our study showing that, although the substitution of extravesicular Cl with either Br or I inhibits Ca²⁺ uptake into cardiac SR vesicles, neither ion has a direct inhibitory effect on the SR Ca²⁺ pump nor does the inhibition of uptake appear to be accompanied by an increase in the Ca²⁺ permeability of the SR membrane. As discussed below, our results are consistent with the hypothesis that the anionic environment of the SR can modulate Ca²⁺ uptake by altering the movement of negative charge into the SR that occurs as Ca²⁺ is actively transported by the Ca²⁺ pump.

View this table: [in this window] [in a new window]   Table 2.   Summary of results with Br and I

Recent interest in the influence that the anionic environment of the SR may have on its ability to regulate intracellular Ca²⁺ in muscle cells has been focused primarily on the modulation of Ca²⁺ channels in the SR that mediate Ca²⁺ release or control the leakiness of the SR membrane to Ca²⁺. In skeletal muscle, Cl and some other inorganic anions (including I; reviewed in Ref. 25) can increase the rate of release of Ca²⁺ through the ryanodine receptor Ca²⁺ channel over the release rates measured in the presence of inorganic anions such as propionate, methanesulfonate, or gluconate (2, 8, 25, 26, 29, 31, 39); however, similar effects on the cardiac muscle ryanodine receptor have not been consistently observed (8, 25). It has also been proposed (39) that a nonselective channel mediates a Cl-induced increase in Ca²⁺ permeability in skeletal muscle SR. This permeability increase was not seen with I or Br (39). The anion-dependent increases in Ca²⁺ release from skeletal muscle SR, whether mediated through the ryanodine receptor or through a Cl-sensitive nonselective channel (39), have been shown to be blocked by ruthenium red (see Refs. 26 and 39). In our experiments, the inhibition of SR uptake by the substitution of I for Cl was the same in the presence and absence of ruthenium red (Table 2 and Fig. 4). This is inconsistent with the hypothesis that inhibitory effects I and Br on net Ca²⁺ uptake are due to an increased Ca²⁺ efflux through an SR Ca²⁺ channel similar to those discussed above. We also showed (Table 2 and Fig. 4) that I did not increase the leakiness of the SR membrane to Ca²⁺, as evidenced by the rate at which Ca²⁺ was released from the SR when the ATP available to the SR Ca²⁺ pump was rapidly removed by hexokinase. From these experiments, we conclude that cardiac SR Ca²⁺ uptake is inhibited by I by a mechanism that does not involve an increase in the Ca²⁺ permeability of the SR membrane.

A second mechanism that might explain our results is a direct inhibitory effect of I or Br on the SR Ca²⁺ pump. Our results, however, do not support this conclusion. Our results showing that I does not directly inhibit the cardiac SR Ca²⁺-ATPase are consistent with those of Highsmith (11), who measured the ATPase activity of the skeletal muscle SR Ca²⁺ pump in the presence of I (at concentrations as high as 40 mM) with the use of molybdate to monitor phosphate release. On the other hand, as noted above (see RESULTS), Stefanova et al. (37) reported an inhibition of the skeletal muscle SR Ca²⁺-ATPase by I when ATPase activity was monitored with the same enzyme-coupled ATPase assay (based on NADH fluorescence) that was used in some of our experiments. Although there may be differences in the influence of I on the activity of the skeletal and cardiac SR Ca²⁺ pumps, we believe that the discrepancy between the results of Highsmith (11) and those of Stefanova et al. (37) and between those of the latter authors and our own results can be accounted for by the effect of I on the NADH-based assay itself. The conclusion that I and Br have no direct inhibitory effects on the cardiac SR Ca²⁺ pump is further supported by our results with the absorbance assay of Ca²⁺-ATPase activity. It might be argued, nevertheless, that I and Br do not affect the ability of the SR Ca²⁺ pump to hydrolyze ATP but uncouple this activity from the binding and transport of Ca²⁺. To our knowledge, this type of uncoupling has not been reported for any known direct inhibitors of SR Ca²⁺ pumps. Our previous work (33) showing that the rate of Ca²⁺ uptake into the SR of smooth muscle is not detectably altered when Cl is replaced by either Br or I also argues against an effect of these ions on the coupling between Ca²⁺ binding and ATPase activity. The SERCA2b Ca²⁺ pump in smooth muscle is identical to the SERCA2a form of the pump of cardiac muscle for most of its amino acid sequence (the 4-amino acid COOH-terminal sequence of SERCA2a is replaced by a 49-amino acid COOH-terminal addition in SERCA2b that is hydrophobic and thought to insert into the smooth muscle SR membrane; reviewed in Ref. 34). Any direct effects of I or Br on the binding of Ca²⁺ to the cardiac SR Ca²⁺ pump would, therefore, have to be mediated by the four COOH-terminal amino acids unique to the cardiac form of the pump or, conversely, the 49-amino acid tail of the SERCA2b form of the pump would have to prevent this from occurring in smooth muscle. It is thus unlikely that Br and I would have this type of a direct inhibitory effect on the SR Ca²⁺ pump in cardiac muscle but not smooth muscle.

The results of our experiments are most consistent with the hypothesis that I and Br inhibit cardiac SR Ca²⁺ uptake by reducing the rate at which negative charge moves into the SR as Ca²⁺ is actively transported by the Ca²⁺-ATPase. The conclusion that such an anion influx occurs during cardiac SR Ca²⁺ uptake is also consistent with our recent work (14) showing that tamoxifen, an agent known to block some types of Cl channels, inhibits cardiac SR uptake without directly inhibiting the SR Ca²⁺ pump and without inducing a significant increase in the permeability of the SR membrane to Ca²⁺.

The fact that Br and I have no effect on smooth muscle SR Ca²⁺ transport suggests that the anion-permeant pathways in the SR are different in the two muscle types. This conclusion is consistent with our findings (33) that SR Ca²⁺ uptake in smooth muscle is inhibited by two Cl channel blockers [NPPB and R(+)-IAA-94] that have no effect on cardiac muscle Ca²⁺ uptake.

We did not see effects of TEA (at a concentration high enough to block many types of K⁺ channels) or of 4-aminopyridine on Ca²⁺ uptake in KI or KCl uptake buffers. If K⁺ movement significantly contributed to charge compensation during SR Ca²⁺ uptake, we might have expected the K⁺ channel blockers to inhibit uptake. The results with TEA and 4-aminopyridine are generally consistent with those of Fink and Stephenson (6; see also Ref. 7) who found that the amount of force that could be produced by Ca²⁺ released from the SR of skinned amphibian skeletal muscle fibers was not decreased but instead increased slightly when the SR was loaded in the presence of TEA (10 mm), 4-aminopyridine (6 µM-2 mM), or other K⁺ channel blockers. Although our results cannot be used to completely rule out the involvement of a K⁺-permeant pathway in the cardiac SR membrane that allows cations to leave the SR during Ca²⁺ uptake, they do indicate that if cation movement does occur during uptake, it occurs through a channel or channels that are TEA and 4-aminopyridine insensitive. It is important to note in this regard that canine cardiac SR K⁺ channels reconstituted into lipid bilayers were found to be insensitive to pharmacological manipulation by lemakalim, glyburide, and charybdotoxin (32). It is also possible that the involvement of K⁺ channels during uptake is more complex than would be expected for a channel that simply allowed K⁺ efflux from the SR (discussed in Refs. 6 and 7). Our results do not address the possibility that K⁺ movement is important during SR Ca²⁺ release.

Comparison of the properties of the striated muscle SR anion channels that have been studied in isolation with the functional evidence from our study suggests some possible candidates for the channel(s) that mediates anion movement during SR Ca²⁺ uptake. Although a number of the SR Cl channels that have been characterized in electrophysiological experiments have permeabilities to I and/or Br that are greater than or equal to that of Cl (see Refs. 10, 13, and 41), a recent report by Kawano et al. (20) describes a cardiac muscle SR anion channel that is less permeable to I than to Cl. This channel is, however, more permeable to Br than to Cl. There are a number of plasma membrane anion channels found in muscle and nonmuscle cells that have permeabilities to Br and/or I that are less than their permeability to Cl (reviewed in Ref. 38). In our experiments, substitution of Cl with SO did not alter the maximum rate of SR Ca²⁺ uptake, a functional property that is consistent with the involvement of a channel with properties similar to those of the small conductance Cl channel identified by Kourie et al. (23) in rabbit skeletal muscle SR. To our knowledge, the permeability of this channel to Br or I has not been determined. In attempting to correlate single-channel studies with functional studies of the intact SR, one must also consider the possibility that different Cl channels may function during SR Ca²⁺ uptake and release or that more than one anion channel type is involved in either or both of these processes in the intact SR. If the latter possibility is the case, the properties of the Cl-permeant pathway that can be deduced from measurements of net Ca²⁺ uptake or release would be a composite of those of two or more channels. The relative importance of these channels during Ca²⁺ uptake may depend on the anionic environment of the SR and/or the involvement of other SR membrane proteins or intracellular agents (see Refs. 21, 22, and 35, for example). The continued study of both the electrophysiological and molecular properties of specific anion channels isolated from SR membranes and studies such as ours, in which the net effects of various ions or molecules on SR function are examined, should help distinguish among these possibilities.