INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

THE SARCOPLASMIC RETICULUM (SR) Ca²⁺-ATPase (SERCA) plays a central role in muscle contraction and relaxation by regulating intracellular Ca²⁺ concentration (9). Muscle contraction is activated by the release of Ca²⁺ via the SR Ca²⁺ release channel, whereas muscle relaxation is produced by Ca²⁺ reuptake into the SR by SERCA. The SERCA pump is an ~110-kDa transmembrane protein (34) and belongs to the P-type superfamily of ion transport ATPases, which couple vectorial ion transport processes with the formation and decomposition of a phosphorylated enzyme intermediate (E-P) (12).

Molecular cloning analyses identified three distinct SERCA genes, SERCA1, SERCA2, and SERCA3, which encode at least five Ca²⁺ pump isoforms (4, 7, 33, 47, 48). All these isoforms are produced by alternate mRNA splicing and are expressed in a developmental and tissue-specific manner. The SERCA1 gene encodes two alternatively spliced transcripts, SERCA1a and SERCA1b, which are exclusively expressed in the fast-twitch fibers of skeletal muscle. SERCA1a is predominantly expressed in the adult stages; SERCA1b is mainly expressed in the fetal/neonatal stages (4). The SERCA2 gene encodes SERCA2a and SERCA2b isoforms. SERCA2a is the primary isoform expressed in the heart and in slow-twitch fibers of skeletal muscle (47, 48) and is also expressed in fast-twitch skeletal muscle during fetal/neonatal stages, but it is replaced by the SERCA1a isoform in the adult stage (48). However, SERCA1a is never expressed in the heart. The SERCA2b isoform is ubiquitously expressed but found at high levels in smooth muscle tissues (33), and the SERCA3 isoform is restricted to specialized cell types, such as epithelial and endothelial cells (7).

The functional significance of having various SERCA isoforms is not completely understood. Given the selective expression pattern of SERCA isoforms in a tissue-specific and developmental manner, it seems likely that SERCA1a and SERCA2a pumps have distinct Ca²⁺ transport properties and contribute differently to the contractile function unique to each muscle type. It has been reported that the Ca²⁺ transport capacity is much lower in cardiac than in fast-twitch skeletal muscle SR (8, 43, 46). This is partly due to the difference in SERCA pump density (8, 43). The apparent affinity of cardiac SERCA enzyme for Ca²⁺ is also two to three times lower than that of fast-twitch skeletal muscle SERCA (8, 43, 46). This has been mainly attributed to specific interaction of SERCA2a with phospholamban (PLB), which is a transmembrane phosphoprotein specifically expressed in cardiac, slow-twitch skeletal, and smooth muscles and regulates the apparent Ca²⁺ affinity of SERCA2a by reversible phosphorylation (28).

On the other hand, the primary structure of SERCA1a and SERCA2a protein is ~84% identical (5). As a consequence, the two SERCA isoforms are predicted to have essentially identical topologies and tertiary structure. Recently, the biochemical properties of SERCA isoforms were studied by expressing different SERCA proteins in nonmuscle COS-1 and HEK-293 cells (31, 32). These studies revealed that SERCA1a and SERCA2a isoforms are similar in terms of Ca²⁺ transport capacity, apparent affinity for Ca²⁺ and ATP, and enzyme turnover during ATP hydrolysis (31). However, these in vitro studies may have potential limitations because of the absence of native regulators such as PLB and the native phospholipid environment in the muscle cells, which has been shown to greatly influence the Ca²⁺ transport function (15, 26). Therefore, it remains to be determined whether the observed difference in Ca²⁺ transport capacity between fast-twitch skeletal and cardiac SR is due to the difference in SERCA pump density (level) (43) or the difference in enzymatic properties of SERCA isoforms and their interaction with tissue-specific regulators (e.g., PLB). To determine whether the SERCA1a isoform differs from the SERCA2a isoform in terms of Ca²⁺ transport properties, we utilized a transgenic (TG) mouse model in which the fast-twitch SERCA1a pump is ectopically overexpressed in the heart. Ectopic overexpression of SERCA1a results in remodeling of the cardiac SR containing 80% SERCA1a and 20% endogenous SERCA2a with an ~2.5-fold increase in total SERCA pump level (29a). The TG model provides us with an in vivo system to compare the Ca²⁺ transport properties of SERCA1a with SERCA2a under the same muscle environment.

	MATERIAL AND METHODS

Top Abstract Introduction Methods Results Discussion References

SERCA1a TG mice. Construction of the SERCA1a transgene and generation of TG mice are described elsewhere (29a). SERCA1a TG mice were generated using the cardiac -myosin heavy chain promoter (17) linked to rat SERCA1a cDNA. Five independent lines (transgene copies 2-9) were identified to express SERCA1a mRNA and protein in the hearts. TG line 38, which showed the highest level of SERCA1a mRNA and protein, was chosen to study SERCA functional properties.

Preparation of cardiac homogenates and microsomes enriched in SR membranes. Excised hearts from 12- to 13-wk-old mice were rinsed with ice-cold saline, immediately frozen in liquid nitrogen, and stored at 80°C. The frozen hearts were powdered and homogenized at 4°C in a solution containing 10 mM imidazole (pH 7.0), 0.3 M sucrose, 10 mM NaF, 1 mM EDTA, 0.3 mM phenylmethylsulfonyl fluoride, 0.5 mM dithiothreitol, and the other protease inhibitors: 10 µg/ml aprotinin, 10 µg/ml leupeptin, and 2 µg/ml pepstatin A. For assay of SERCA activity, microsomal fractions enriched in SR membranes were obtained using differential centrifugation of the cardiac homogenates, as described previously (25). Protein concentrations were determined by the Bio-Rad method, with BSA as standard.

Measurement of SERCA activity. Ca²⁺-ATPase activity was measured in isolated microsomes enriched in cardiac SR, as described elsewhere (24). The reaction mixture contained 50 mM histidine, pH 7.0, 100 mM KCl, 5 mM MgCl₂, 5 mM azide, 20 µg/ml microsomal protein, 1 mM EGTA, and various CaCl₂ concentrations to yield the desired free Ca²⁺ concentration from 0.1 to 10 µM (40). The Ca²⁺ ionophore A-23187 (2 µM) was added to the assay system to ensure that Ca²⁺ accumulated by SR vesicles during the assay does not inhibit the Ca²⁺ pumps from inner surfaces of the vesicles (23). In addition, 3.33 mM phosphoenolpyruvate and 3 U/ml pyruvate kinase were included to regenerate ATP from ADP during the reaction (24). The reaction was started by addition of 3 mM ATP at 37°C and was stopped after 20 min by addition of colorimetric reagent. Absorbance was read at 660 nm to monitor the liberated P_i, and the rate of ATP hydrolysis was determined by measuring the amount of P_i liberated during 20 min of incubation of microsomes. Ca²⁺-dependent ATPase activity was calculated by subtracting Ca²⁺-independent (basal) ATPase activity (measured in the presence of 2 mM EGTA and the absence of Ca²⁺) from total ATPase (in the presence of Ca²⁺). The SERCA origin of Ca²⁺-ATPase activity in the isolated microsomes was evaluated by thapsigargin inhibition, as described previously (41). It has been established that thapsigargin is the most potent and specific inhibitor of SERCAs, and it inhibits SERCA Ca²⁺ transport and Ca²⁺-ATPase activity (21). Preincubation (2 min at 37°C) of reaction mixture containing microsomal protein with 1 µM thapsigargin resulted in 92-96% inhibition of Ca²⁺-dependent ATP hydrolysis in the microsomes, indicating that the Ca²⁺-dependent ATP hydrolysis is mostly due to SERCAs.

Ca²⁺ transport assays. Oxalate-facilitated Ca²⁺ uptake into SR vesicles in cardiac homogenates was determined by the Millipore filtration technique following an established method (18). Cardiac homogenates (100 µg/ml) were incubated (37°C) in a medium containing 40 mM imidazole (pH 7.0), 100 mM KCl, 5 mM MgCl₂, 5 mM NaN₃, 5 mM potassium oxalate, 0.5 mM EGTA, and various concentrations of CaCl₂ to yield 0.03-3 µM free Ca²⁺ (containing 1 µCi/µmol ⁴⁵Ca²⁺), as determined by the computer program (40). Ruthenium red (1 µM) was added to obtain the maximal stimulation of SR Ca²⁺ uptake by inhibiting the Ca²⁺ release channel (18). The reaction was initiated by the addition of 5 mM ATP. The rate of Ca²⁺ uptake was calculated by least-squares linear regression analysis of uptake at 30, 60, and 90 s. SR specificity of Ca²⁺ uptake in cardiac homogenates was confirmed by thapsigargin inhibition. We observed that preincubation of homogenates with 1 µM thapsigargin resulted in 95 and 94% inhibition of Ca²⁺ uptake activity at saturated Ca²⁺ concentration in hearts from TG and non-TG (NTG) mice, respectively, suggesting that the Ca²⁺ uptake measured with cardiac homogenates is mostly supported by SR Ca²⁺ uptake, as defined previously (29).

Formation and decay of E-P in the presence of ATP and Ca²⁺. The steady-state levels of Ca²⁺-dependent E-P were measured as described by Anderson et al. (2). Cardiac homogenate protein (20 µg) was added to 0.1 ml of reaction mixture containing 40 mM imidazole, pH 7.0, 100 mM KCl, 5 mM MgCl₂, 5 mM NaN₃, and various concentrations of free Ca²⁺ (0.1-100 µM) or 1 mM EGTA. The components of the reaction mixture were preincubated in ice with vortex mixing. The reaction was initiated by addition of 2 µM [-³²P]ATP (sp. act. 10 µCi/nmol) at 0°C and acid quenched after 15 s by addition of 1 ml of ice-cold stop solution (6% TCA, 0.3 mM ATP, and 5 mM P_i). The samples were placed in ice for 5 min and then vacuum filtered through 0.45-µm Millipore filters. Proteins remaining on the filters were washed three times with 15 ml of stop solution, then the filters were processed for scintillation counting. Ca²⁺-dependent E-P formation was calculated as the difference between the amount of ³²P incorporation into protein in the presence of Ca²⁺ and in the presence of EGTA.

Statistical analysis. Values are means ± SE. Data concerning the dependence of ATP hydrolysis and Ca²⁺ uptake rates on free Ca²⁺ and ATP concentrations were fit to the equation for a general cooperative model for substrate as follows where V_max (maximum velocity), K_0.5 (concentration required for half-maximal activation), and n (equivalent of the Hill coefficient) were calculated by using nonlinear regression analysis with Origin 3.5 (MicroCal Software) and where [S] is substrate concentration.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Generation and characterization of SERCA1a TG mice have been described elsewhere (29a). We have shown that ectopic overexpression of SERCA1a in the hearts results in remodeling of the cardiac SR containing ~80% SERCA1a and ~20% endogenous SERCA2a with an ~2.5-fold increase in total SERCA protein levels. SERCA1a overexpression did not alter the expression levels of other proteins, including PLB. The SERCA1a TG mice are healthy and do not display cardiac pathology. Functional analysis of SERCA1a TG hearts demonstrates a significant enhancement (40-42%) in the cardiac contractile performance (29a).

SERCA1a overexpression increases the rate of ATP hydrolysis without altering the apparent pump affinity for Ca²⁺. To determine the effect of SERCA1a overexpression in SERCA activity in hearts from TG mice, we used isolated microsomes enriched in cardiac SR to measure the Ca²⁺-dependent ATP hydrolysis. The rate of ATP hydrolysis was significantly higher in TG than in control microsomes over a wide range of Ca²⁺ concentrations (0.03-10 µM; Fig. 1A). The maximum rate of ATP hydrolysis was increased ~2.5-fold in TG compared with NTG hearts (Fig. 1A, Table 1). These data indicate that SERCA1a overexpression results in a significant increase in cardiac SERCA activity. However, K_0.5 of Ca²⁺ for ATPase activity was similar in TG and NTG hearts (Fig. 1B). Ca²⁺ activation of Ca²⁺-ATPase activity was best fit by a model requiring two cooperative Ca²⁺-binding sites in TG and NTG hearts (Table 1).

View larger version (16K): [in this window] [in a new window]   View larger version (14K): [in this window] [in a new window]   Fig. 1.   Ca2+ dependence of Ca2+-ATPase activity. Ca2+ activation of ATPase activity was measured in microsomes isolated from transgenic (TG) and non-TG (NTG) hearts. Microsomes (20 µg/ml) enriched in sarcoplasmic reticulum membranes were incubated with reaction mixture containing 2 mM EGTA or various concentrations of Ca2+. Ca2+-dependent ATP hydrolysis was determined by subtracting basal ATPase (in presence of EGTA) from total ATPase activity (in presence of added Ca2+). Average of 4 separate determinations is shown, each performed in duplicate. Curve represents best fit of data to a general cooperative model.

Apparent pump affinity for Ca²⁺ is not altered in SERCA1a TG hearts, but the V_max of SR Ca²⁺ uptake is increased. To investigate the Ca²⁺ transport properties of SERCA1a hearts, we used cardiac homogenates to perform the ATP-dependent oxalate-facilitated SR Ca²⁺ uptake. Our results showed that the velocity of SR Ca²⁺ uptake was significantly higher in TG than in NTG hearts over a wide range of Ca²⁺ concentrations (0.03-3 µM; Fig. 2A). The V_max of SR Ca²⁺ uptake was ~1.9-fold higher in TG than in NTG hearts (Table 1).

View larger version (26K): [in this window] [in a new window]   View larger version (18K): [in this window] [in a new window]   Fig. 2.   Effect of phospholamban (PLB) monoclonal antibody (Ab) on sarcoplasmic reticulum Ca2+-ATPase (SERCA) uptake. Initial rates of Ca2+ uptake were determined in cardiac homogenates. Cardiac homogenates were incubated at 0°C for 15 min in absence (Ab) or presence (+Ab) of PLB antibody, then Ca2+ dependence of Ca2+ uptake was assayed. PLB monoclonal antibody shifts Ca2+ activation curve to left in TG and NTG hearts. Data are expressed as nmol Ca2+ · mg1 · min1 (A) and as percentage of Ca2+ uptake rate at pCa 6.0 in A (B). Average of 4 separate determinations is shown.

Anti-PLB antibody increases the apparent SERCA affinity for Ca²⁺ in TG hearts. Despite SERCA1a overexpression, we did not observe an increase in the apparent SERCA affinity for Ca²⁺ in TG hearts compared with the affinity for Ca²⁺ determined in NTG hearts. This result suggests that PLB may interact with SERCA1a in TG hearts. To further determine whether PLB interacts with SERCA1a, we tested the effect of anti-PLB antibody 2D12 on the SR Ca²⁺ uptake function in TG and NTG hearts. This antibody was previously shown to remove the inhibitory effect of PLB on the Ca²⁺ pump (6, 8). Addition of PLB antibody increased the rate of SR Ca²⁺ uptake 1.5- to 2.5-fold at lower Ca²⁺ concentrations (<1 µM) in TG and NTG hearts (Fig. 2A). PLB antibody treatment increased the pump affinity for Ca²⁺ to a similar extent (2.4-fold decrease in K_0.5) in TG and NTG hearts (Table 1, Fig. 2B). These data are comparable to the previously reported data showing the effect of anti-PLB antibody on cardiac SERCA (8, 39). At the same time, PLB antibody treatment did not stimulate the Ca²⁺ uptake rates at saturating Ca²⁺ concentrations (>1 µM) in TG and NTG hearts (Table 1), indicating that the V_max of Ca²⁺ uptake was not affected by PLB antibody, which is quite consistent with the results reported earlier (8, 30, 39).

pH dependence of SR Ca²⁺ uptake is similar in TG and NTG hearts. It has been proposed that SERCA is also an H⁺ pump that generates an H⁺ gradient to sustain Ca²⁺ transport and exchanges H⁺ for Ca²⁺ (10, 35). It is known that the SERCA Ca²⁺ transport function is pH dependent. TG and NTG hearts showed similar pH dependence for Ca²⁺ uptake rate when pH was varied from 6.0 to 8.0 (Fig. 3A). The optimal pH for Ca²⁺ uptake was ~7.0 for TG and NTG cardiac tissues. Because Ca²⁺ binding to the Ca²⁺-ATPase is pH dependent (10), we measured the effect of acidic pH on the apparent affinity of SERCA for Ca²⁺. In TG and NTG hearts the apparent affinity of the enzyme for Ca²⁺ decreased ~10 times when the pH was lowered from 7.0 to 6.3 and decreased ~40 times when the pH was lowered from 7.0 to 6.0 (Fig. 3B). TG and NTG hearts showed similar apparent affinity for Ca²⁺ at pH 6.3 (2.32 ± 0.18 vs. 2.06 ± 0.20 µM, n = 3, P > 0.05) and pH 6.0 (9.87 ± 0.42 vs. 10.35 ± 1.35 µM, n = 3, P > 0.05; Fig. 3B). In addition to its effect on Ca²⁺ affinity, lowering the pH also caused a decrease in the V_max of Ca²⁺ uptake. However, at pH 6.3, TG hearts still showed an ~1.9-fold increase in the V_max for Ca²⁺ uptake compared with NTG hearts, similar to the increase observed at neutral pH (data not shown). At pH 6.0 the V_max of Ca²⁺ uptake could not be reached, even in the presence of 100 µM Ca²⁺ in TG and NTG hearts. Acidic pH decreased the apparent cooperativity of Ca²⁺ uptake (Hill coefficient) in TG and NTG hearts, which is consistent with the results reported previously (13). There is no difference in the magnitude of decrease between TG and NTG cardiac tissues (Fig. 2B).

View larger version (15K): [in this window] [in a new window]   View larger version (22K): [in this window] [in a new window]   Fig. 3.   pH dependence of sarcoplasmic reticulum Ca2+ uptake. A: initial rate of sarcoplasmic reticulum Ca2+ uptake determined in cardiac homogenates at different pH values from pH 6.0 to 8.0. Free Ca2+ concentration was 1 µM. B: effect of acidic pH on Ca2+ dependence of Ca2+ uptake. Rate of Ca2+ uptake was measured in reaction media containing 0.1-10 µM free Ca2+ at pH 7.0, 6.3, and 6.0, as determined by computer program. Values are representative of 3 separate determinations.

Increased V_max of Ca²⁺ transport in TG hearts is due to the increased active pump (E-P) level. Ca²⁺ translocation is tightly coupled with the formation of SERCA E-P, in which the terminal phosphate of ATP is incorporated into the aspartyl residue of the ATPase enzyme, so that the steady-state level of E-P can be used to quantitate the amount of active enzyme (36). To address whether the increase in SR Ca²⁺ transport and Ca²⁺-ATPase activity in TG hearts is due to an increase in active pump fraction, we measured the amount of E-P formed during enzyme turnover. Autoradiography demonstrated that the Ca²⁺-dependent incorporation of ³²P from ATP into a single ~110-kDa band corresponds to SERCA. Radiation incorporation was totally Ca²⁺ dependent and completely blocked by 500 nM thapsigargin (Fig. 4A). This specificity allows us to quantitate the amount of E-P. TG cardiac homogenates displayed a 2.5-fold increase in the steady-state level of E-P compared with NTG homogenates (141.11 ± 4.43 vs. 55.11 ± 2.02 pmol/mg, n = 6, P < 0.01; Fig. 4B). This is quite consistent with the magnitude of the increase in ATP hydrolysis and SERCA protein level. The E-P data indicate that the ectopically expressed SERCA1a pump in the hearts is functional and the increase in the V_max of SR Ca²⁺ uptake and ATP hydrolysis is due to increase in the total amount of SERCA pumps.

View larger version (44K): [in this window] [in a new window]   View larger version (15K): [in this window] [in a new window]   View larger version (15K): [in this window] [in a new window]   Fig. 4.   Formation of SERCA phosphorylated intermediate (E-P) in presence of ATP and Ca2+. A: autoradiograph showing Ca2+ dependence and thapsigargin (Thg) sensitivity of phosphoenzyme formation. E-P protein (10 µg/well) was run in 7% acidic SDS-PAGE (pH 6.3), and autoradiography was carried out as described in MATERIALS AND METHODS. B: Ca2+ dependence of E-P formation determined by filtration technique. Cardiac homogenates (0.2 mg/ml) were phosphorylated on ice in presence of 2 µM [-32P]ATP and various concentrations of Ca2+. Reaction was stopped at 15 s with 1 ml of TCA-ATP-Pi solution, then subjected to vacuum filtration. C: ATP dependence of E-P formation. Reaction medium contained 0.1-2 µM [-32P]ATP. Data are expressed as percentage of E-P formation obtained at 2 µM [-32P]ATP. Average of 6 separate determinations is shown.

Turnover rate of SERCA enzyme (E-P decay) is similar in TG and NTG hearts. An increase in the velocity of SR Ca²⁺ transport can occur as a result of an increase in total pump number and/or faster turnover rate of the SERCA pump. To determine the turnover rate of SERCA enzyme in TG hearts, we measured the rate of E-P decay by performing isotopic chase experiments, as described below. These experiments were carried out at low ATP concentrations (2 µM) and at 0°C to slow the phosphoenzyme decomposition. The E-P turnover rate was determined by first obtaining steady-state levels of radioactive phosphoenzyme in the presence of 100 µM Ca²⁺ and 2 µM [-³²P]ATP and then adding an excess of nonradioactive ATP. After addition of cold ATP, nearly all the newly formed phosphoenzyme is nonradioactive, so the rate of the disappearance of radioactive phosphoenzyme represents phosphoenzyme turnover rate (16). We found that the [³²P]phosphoenzyme decayed in a few seconds and exhibited a first-order exponential curve (Fig. 5), which is in agreement with previous reports (8, 45). As a function of time, the rate of E-P decomposition was similar in TG and NTG hearts, with rate constants of 0.39 ± 0.01 and 0.37 ± 0.02 s¹, respectively (n = 10, P > 0.05). Because the turnover rate of phosphoenzyme (rate constant) measured under this condition is independent of the amount of E-P, our results demonstrate that the turnover rates of SERCA1a and SERCA2a enzymes in the cardiac environment are similar.

View larger version (15K): [in this window] [in a new window]   Fig. 5.   Decomposition of E-P in presence of ATP and Ca2+. Cardiac homogenates were phosphorylated for 15 s on ice in presence of 100 µM Ca2+ and 2 µM [-32P]ATP to obtain steady-state level of E-P. Isotopic chase experiments were conducted by adding 1 mM nonradioactive ATP to measure decay of E-P, and reaction was stopped at different times, then subjected to vacuum filtration. Data are expressed as percentage of E-P value at time-0 baseline from 10 separate preparations.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

We have utilized a TG mouse model that ectopically overexpresses SERCA1a in the hearts to investigate the Ca²⁺ transport properties of the SERCA1a pump. Our findings indicate that 1) SERCA1a overexpression results in a significant increase in the SR Ca²⁺ transport function because of the increase in SERCA pump level, 2) PLB interacts with the SERCA1a pump in the heart, and 3) the SERCA1a pump is similar to SERCA2a in terms of kinetic properties. These findings suggest that SERCA1a can functionally substitute for SERCA2a in the heart. Recently, using adenoviral gene transfer, Inesi et al. (20) showed that overexpression of SERCA1a in chick neonatal cardiac myocytes enhanced SR Ca²⁺ transport function and myocyte shortening. These results also suggest that SERCA1a can functionally substitute for SERCA2a in cardiac myocytes.

In the present study we found that, despite SERCA1a overexpression, the apparent SERCA affinity for Ca²⁺ was similar in TG and NTG hearts, suggesting that PLB regulates SERCA1a in TG hearts. Previous in vitro studies showed that PLB can regulate SERCA1a when reconstituted together in lipid bilayers (15, 26) or when coexpressed with SERCA1a in COS-1 cells (19). Similarly, studies in TG mice showed that ectopic expression of PLB in mouse fast-twitch skeletal muscle decreased the affinity of SERCA1a for Ca²⁺ (44). To further understand whether SERCA1a is regulated by native PLB in TG hearts, we tested the effect of anti-PLB antibody 2D12 on SR Ca²⁺ uptake, which has been proven to relieve PLB inhibition on cardiac SERCA (6, 8). We found that addition of PLB antibody increased the apparent Ca²⁺ affinity 2.4-fold in TG hearts, suggesting that PLB can interact with SERCA1a in the hearts. In TG hearts, 80% of total SERCA pumps are represented by SERCA1a and only 20% by SERCA2a (29a). Therefore, the shift in the apparent affinity for Ca²⁺ by PLB antibody is largely determined by SERCA1a in TG hearts. Furthermore, we observed that the magnitude of the shift in SERCA affinity for Ca²⁺ by PLB antibody was very similar in TG and NTG hearts. Thus PLB appears to interact with SERCA1a in a manner similar to SERCA2a. Consistent with this, a major binding site for PLB has been identified between Asp³⁷⁰ and Lys⁴⁰⁰ in SERCA2a, a region that is highly conserved between SERCA1 and SERCA2 isoforms (22). Taken together, these results suggest that the difference in apparent Ca²⁺ affinity observed in skeletal and cardiac SR is not due to differences in Ca²⁺-binding properties of SERCA isoforms but mainly to selective regulation by PLB. This is also supported by the finding that Ca²⁺ binding to the ATPase under equilibrium conditions in the absence of ATP was identical in cardiac and skeletal SR (8). In addition, we found that the V_max of Ca²⁺ uptake remained unchanged after PLB antibody treatment, providing additional evidence that the main regulatory effect of PLB is on the apparent affinity for Ca²⁺, but not on the V_max of the Ca²⁺ transport, which is in agreement with reports from other investigators (8, 30, 39).

Previous studies have shown that alterations in PLB protein level by means of PLB gene ablation or overexpression result in altered apparent pump affinity for Ca²⁺ (30, 25). Because the levels of the cardiac SERCA were not changed in these previous studies, the data suggested that alterations in the relative stoichiometry of PLB to SERCA can alter the apparent SERCA affinity for Ca²⁺ (28). In contrast, in SERCA1a TG hearts the SERCA pump level was increased ~2.5-fold without a change in the PLB protein level. Therefore, because of the preferential increase in the SERCA-to-PLB protein ratio, one would expect a significant increase in the SERCA apparent affinity for Ca²⁺. Surprisingly, we did not find such an effect in TG hearts. Moreover, we obtained similar results when SERCA2a was overexpressed in the heart. In SERCA2a TG hearts the total SERCA2a protein was increased by ~50%, resulting in a 37% increase in the V_max of SR Ca²⁺ uptake, but the PLB level was unchanged, and the apparent affinity for Ca²⁺ was not altered (3a). The molecular mechanisms of SERCA regulation by PLB are not completely understood, and the functional stoichiometry of PLB to SERCA2a is not known. This is complicated by the fact that PLB exists as a pentamer (inactive form) and a monomer (active form), as evaluated using an electron paramagnetic resonance technique (11). The equilibrium between these two states may be regulated by phosphorylation of PLB (11) and SERCA pump level (D. Thomas, personal communication). Therefore, an increase in SERCA pump level may shift the dynamic equilibrium of PLB toward a monomer, which has been reported to be a more effective inhibitor than a PLB pentamer (3, 11, 27). At any rate, our results suggest that there is enough "space" for PLB molecules in the heart to effectively inhibit a 2.5-fold overexpression of Ca²⁺ pumps.

Our studies showed that, regardless of the total SERCA pump level, the enzymatic properties of SERCA pumps from TG and NTG hearts are similar. The TG and NTG hearts revealed similar apparent affinity for Ca²⁺ and ATP and similar Hill coefficients. In addition, the phosphoenzyme turnover was also similar between TG and NTG hearts, suggesting that SERCA1a and SERCA2a have similar kinetic properties in cardiac SR. This is consistent with the previous results showing similar enzymatic properties of SERCA1a and SERCA2a in nonmuscle cell (COS-1 and HEK-293) microsomes (31, 32). On the other hand, it has been found that the effect of acidic pH on K_0.5 and V_max of Ca²⁺ uptake was less pronounced in fast-twitch skeletal muscle SR vesicles than in cardiac SR vesicles, suggesting that SERCA1a is more resistant than SERCA2a to acidic pH (46). In our studies, however, we observed that acidic pH exerted a similar inhibitory effect on SERCA in TG and NTG hearts. Acidic pH lowered apparent Ca²⁺ affinity and Ca²⁺ binding cooperativity. Although there are reports which indicate that acidic pH enhances the Ca²⁺-binding cooperativity (when the free Ca²⁺ concentrations are controlled with buffers other than EGTA) (14), in this study we found that the Ca²⁺-binding isotherms at various pH levels remain the same in TG and NTG cardiac tissues. These data suggest that SERCA1a and SERCA2a have similar pH dependence under a similar cardiac SR environment. The difference in pH dependence of Ca²⁺ uptake observed between fast-twitch skeletal SR and cardiac SR may be due to differences in the SR membrane environments, including phospholipid composition and tissue-specific modulators (such as PLB). Recently, a low-molecular-weight proteolipid, sarcolipin, has been found to be mainly expressed in fast-twitch skeletal muscle to regulate SERCA1 and has structural homology with PLB (38). Therefore, we should not exclude the possibility that the functional properties of SERCA isoforms may be additionally regulated by the other tissue-specific modulators.