Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Ca²⁺ is an important regulator of many cellular functions. In the heart, elevation of Ca²⁺ near the myofilaments initiates generation of force, whereas relaxation is dependent on the removal of cytosolic Ca²⁺. Due to its critical role, cytosolic Ca²⁺ levels are tightly controlled in the heart. Ca²⁺ is pumped into the sarcoplasmic reticulum (SR) by the SR-Ca²⁺-ATPase (SERCA) or extruded through the sarcolemma into the interstitium by the Na⁺-Ca²⁺ exchanger (NCX). In the heart, Na⁺-Ca²⁺ exchange removes up to 20% of the Ca²⁺ from the cytosol (Bers and Bridge, 1989). The remaining Ca²⁺ is mainly transported back into the SR by SERCA. The Ca²⁺ uptake into the SR mediated by SERCA (Lytton and MacLennan, 1991) is regulated by phospholamban (PLB). PLB inhibits activity by lowering the Ca²⁺ affinity of SERCA (Kranias et al., 1985; Kim et al., 1990). Ca²⁺ stored in the SR is primarily bound to calsequestrin (CSQ; Mitchell et al., 1988). CSQ is located in the junctional SR, whereas SERCA and PLB are located in the free SR (Fleischer and Inui, 1989). In the junctional SR, CSQ might be in functional contact via junctin and triadin with the Ca²⁺ release channel (Ikemoto et al., 1989; Zhang et al., 1997). Opening of the Ca²⁺ release channel during systole releases Ca²⁺ from the SR to initiate contraction. Na⁺-Ca²⁺ exchange is the major Ca²⁺ efflux mechanism in the myocardium, although a sarcolemmal Ca²⁺-ATPase also contributes to a minor extent. By use of the Na⁺ gradient generated by Na⁺,K⁺-ATPase, NCX mediates exchange: the exchange of one intracellular Ca²⁺ for three extracellular Na⁺ (forward direction; for a review, see Philipson and Nicoll, 1993). NCX may also work in the reverse mode and thus may also be involved in systolic Ca²⁺ entry and contribute to the Ca²⁺-induced Ca²⁺ release mechanism (Leblanc and Hume, 1990).

Proteins that control Ca²⁺ levels in the myocytes are of clinical relevance, because human heart failure is associated with an altered Ca²⁺ homeostasis (Morgan, 1991). Specifically, diastolic free calcium levels are elevated in isolated preparations from failing hearts compared with nonfailing hearts (Gwathmey et al., 1987; Beuckelmann et al., 1992). The biochemical mechanisms are just beginning to be unraveled (for a review, see Hasenfuss, 1998). Mice overexpressing CSQ developed heart hypertrophy. Hence, elevated CSQ levels and hypertrophy might be correlated. However, in human end-stage heart failure, CSQ mRNA and protein expression are unchanged compared with nonfailing myocardium (Arai et al., 1993; Linck et al., 1996). Data in an animal model of hypertrophy and subsequent heart failure showed a transient increase of CSQ during hypertrophy followed by an unchanged CSQ expression in heart failure (Matsui et al., 1995). Moreover, the expression of NCX was increased in human heart failure (Komuro et al., 1992; Studer et al., 1994). In animals, a decreased activity of NCX has been reported in rat cardiac hypertrophy (Hanf et al., 1988), whereas the activity of NCX was increased in hamster cardiomyopathy (Hatem et al., 1994). Both CSQ and NCX lower cytosolic Ca²⁺ levels. Therefore, reciprocal regulation between the CSQ and NCX expression might be conceivable.

Interestingly, transgenic mice overexpressing CSQ selectively in the heart exhibited severe cardiac hypertrophy and enhanced mortality, whereas transgenic mice overexpressing NCX had no cardiac hypertrophy and did not die prematurely (Adachi-Akahane et al., 1997; Jones et al., 1998). From these findings, one can speculate that the elevated NCX levels in end-stage human heart failure may be a compensatory effect to prevent further Ca²⁺ overload of the cell. Hence, we hypothesize that the parallel overexpression of CSQ and NCX could alter the phenotype of CSQ-overexpressing mice. Therefore, we generated mice overexpressing both CSQ and NCX. Hence, we tested the hypothesis that overexpression of NCX should lead to removal of Ca²⁺ from the cytosol, reduce the excessive storage of Ca²⁺ in the SR, and thus might normalize the biochemical and physiological parameters of heart function in mice overexpressing CSQ.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Generation and Analysis of Transgenic Mice. Transgenic mice overexpressing CSQ or NCX were generated as described previously (Adachi-Akahane et al., 1997; Jones et al., 1998). Double transgenic mice were generated by mating NCX-positive female animals with CSQ-positive male animals. Transgene-positive mice were identified by polymerase chain reaction analysis using specific primers for detection of the corresponding transgene. Mice (6-7 weeks old) were used in accordance with institutional guidelines. The mice were sacrificed at an age of 6 weeks by a blow to the neck and bleeding from the carotid arteries. Hearts were briefly rinsed in Tyrode's solution (see later) to remove blood. Subsequently, the hearts were placed on Kimwipes to remove buffer and weighed. Hearts were quickly frozen in liquid nitrogen and stored at 80°C for biochemical analyses or were rapidly used for physiological experimentation as described later.

Histological Examinations. The hearts of wild-type and transgenic mice were immediately immersed into 4% buffered formaldehyde for a maximum of 12 h. Cross sections were paraffin-embedded, cut at a thickness of 4 µm, and mounted on Silane-coated glass slides. For routine histological examination, slides were stained with hematoxylin-eosin and elastica van Gieson.

Contraction Experiments. Experiments were performed as described recently (Bokník et al., 1997). In brief, force of contraction was measured in electrically driven (frequency, 1 Hz; duration, 5 ms; intensity, 20% greater than threshold) left atrial muscles. The preparations were isolated, mounted, and suspended individually in glass tissue chambers for recording isometric contractions. The bathing solution (10 ml) was a modified Tyrode's solution containing 119.8 mM NaCl, 5.4 mM KCl, 1.8 mM CaCl₂, 1.05 mM MgCl₂, 0.42 mM NaH₂PO₄, 22.6 mM NaHCO₃, 0.05 mM Na₂EDTA, 0.28 mM ascorbic acid, and 5.0 mM glucose. The chamber was continuously gassed with 95% O₂, 5% CO₂ and maintained at 35°C. Force of contraction was measured with an inductive force transducer. Preparations were allowed to equilibrate for 30 min. Time from 10% contraction to peak contraction [time to peak tension (TPT)] and time from peak to 90% relaxation [relaxation time (RT)] were calculated from recordings at high chart speed. Concentration-response curves were obtained cumulatively, and the developed force (systolic and diastolic) and diastolic tension were expressed in mN.

Immunoassay. Immunoblotting was conducted as described previously (Linck et al., 1996). Aliquots of homogenate protein from control or the corresponding transgenic ventricles were subjected to SDS-polyacrylamide gel electrophoresis in 10% polyacrylamide, and proteins were transferred to nitrocellulose membranes. The membranes were cut into horizontal sections, and the appropriate sections were incubated with antibodies specific for CSQ, NCX, SERCA, PLB, and troponin inhibitor (TnI). The CSQ and NCX antibodies were raised against canine proteins and primarily detected the corresponding transgenic canine proteins, whereas the native mouse proteins produced only very weak signals. Bound antibodies were visualized using ¹²⁵I-protein A. Bound radioactivity was quantified with use of a Bio-Rad (Hercules, CA) GS-250 Molecular Imager.

Total RNA Preparation. Total RNA was isolated from ventricular tissue (frozen immediately after sacrifice) according to a modified protocol of Chomczynski and Sacchi (1987). The frozen tissue was homogenized using a microdismembrator (Braun, Melsungen, Germany) in 1 ml TriStar Reagent (AGS, Heidelberg, Germany) containing guanidinium thiocyanate and phenol. RNA was extracted according to the instructions of the manufacturer. In brief, 160 µl chloroform was added to 800 µl homogenate, and the resulting phases were separated by centrifugation. The RNA was precipitated and washed, and the dried pellet dissolved in diethylpyrocarbonate-treated water. RNA was transferred to Hybond N nylon membranes overnight (Amersham Buchler, Braunschweig, Germany) by Northern blot capillary transfer using 20× standard saline citrate (SSC; 3 M NaCl and 0.3 M sodium citrate, pH 7.0) as the transfer medium. Transfer was assessed on an ultraviolet transilluminator. Hybond N nylon membranes were prehybridized for 2-6 h at 42°C in a prehybridization solution containing 50% formamide, 5× Denhardt's solution (1 mg/ml Ficoll, polyvinylpyrrolidone, and BSA), 0.9 M NaCl, 0.06 M NaH₂PO₄, 0.006 M EDTA, 0.1% SDS, and 400 µg/ml tRNA from yeast. For hybridization, the corresponding cDNA probes for CSQ, NCX, SERCA, and atrial natriuretic factor (ANF) were labeled (Mega Prime Kit; Amersham Buchler) with [³²P]dCTP (3.000 Ci/mmol; DuPont-New England Nuclear; Bad Homburg, Germany) to a specific activity of 3.5-8.0 × 10⁸ dpm/µg. Unbound radioactivity was separated by gel filtration with Sephadex G-50 DNA grade (Pharmacia Fine Chemicals, Uppsala, Sweden). Hybridization of the membrane was performed at 42°C for 16 to 20 h. The membranes were washed twice in 2× SSC, 0.1% SDS at room temperature followed by a 15-min washing in 2× SSC, 0.1% SDS at 65°C. The blots were washed three times to a final stringency in 0.2× SSC, 0.1% SDS at 65°C. Wet blots were sealed in plastic wrap and exposed using a PhosphorImager (Molecular Dynamics, Sunnyvale, CA). For further hybridizations with other radiolabeled probes, the blot membrane was washed by boiling in 0.1% SDS. After a control exposure in the PhosphorImager to assess loss of label, membranes were used for subsequent additional hybridizations. Hybridization intensity of autoradiographic signals on Northern blots was measured quantitatively by the PhosphorImager system (ImageQuant; Molecular Dynamics).

Isolation of Cardiomyocytes. Mouse ventricular myocytes were isolated by collagenase digestion of Langendorff-perfused hearts (37°C) obtained from wild-type and transgenic animals using a modification of an earlier protocol (Bokník et al., 1997). At 30 min before preparation, mice were treated with heparin (5000 U/kg). Then, mice were placed under a CO₂ atmosphere for narcosis. Hearts were rapidly excised, mounted onto a modified Langendorff perfusion system, and perfused retrogradely via the cannulated aorta in a nonrecirculating manner at a constant rate of 2 ml/min for 5 min with a Ca²⁺-free solution (solution A: 140 mM NaCl, 5.8 mM KCl, 0.4 mM NaH₂PO₄, 0.5 mM KH₂HPO₄, 0.9 mM MgSO₄, 10 mM HEPES, 10 mM dextrose, pH 7.1). Then, hearts were perfused for 30 min in a recirculating manner with solution A supplemented with 0.22 mg/ml collagenase (type D; Boehringer Mannheim, Mannheim, Germany). During digestion, the Ca²⁺ concentration was increased every 5 min (20, 40, 60, 80, and 100 µM). Then, the hearts were perfused for 10 min with enzyme-free solution A containing 0.1 mM Ca²⁺. Cells were harvested after mincing the hearts with fine scissors, gentle agitation of the tissue, and filtering through a nylon mesh. Cells were incubated for 60 min in a solution containing 50 mM glutamic acid, 20 mM taurine, 20 mM HEPES, 10 mM dextrose, 3 mM MgSO₄, 0.5 mM EGTA, 30 mM KCl, and 30 mM KH₂PO₄, adjusted to pH 7.3 with KOH. Cells used for measurement of Ca²⁺ currents were further stored in this solution, whereas cells used for measurement of Ca²⁺ transients were stored in a low-sodium/high-sucrose solution containing 52.5 mM NaCl, 4.8 mM KCl, 1.19 mM KH₂PO₄, 1.2 mM MgSO₄, 11.1 mM dextrose, 145 mM sucrose, 10 mM HEPES, and 0.2 mM CaCl₂, adjusted to pH 7.3 with NaOH.

Measurement of L-Type Ca²⁺ Currents. The measurement is slightly modified from our previous reports on guinea pig cardiomyocytes (Bokník et al., 1997). Cells were plated in a Petri dish, which served as recording chamber (volume, approximately 2 ml) on the stage of an inverted microscope (Leica, Köln, Germany). Whole-cell patch-clamp recordings were performed with a modified Tyrode's solution as extracellular solution [solution B: 130 mM tetraethylammonium (TEA)-Cl, 1 mM MgCl₂, 4 mM 4-aminopyridine, 10 mM HEPES, 10 mM dextrose, 2 mM CaCl₂, titrated to pH 7.3 with TEA-OH). TEA-Cl and 4-aminopyridine were used to suppress the large K⁺ currents in mice (Thierfelder et al., 1994). Recording pipettes (soft glass, 1.5-2.5 M) were filled with 80 mM K-aspartate, 50 mM KCl, 10 mM KH₂PO₄, 0.5 mM MgCl₂, 3 mM MgATP, 10 mM HEPES, and 1 mM EGTA, titrated to pH 7.4 with KOH.

Measurement of Intracellular Ca²⁺ Concentration ([Ca]_i). Freshly isolated myocytes were placed onto a Petri dish (volume 300 µl) on the stage of a modified inverted microscope (Diaphot 200; Nikon, Tokyo, Japan). The cells were incubated with solution A (pH 7.3) containing cell-permeant Indo-1 AM (25 µM; Molecular Probes, Eugene, OR) and 1% of the nonionic surfactant Pluronic F-127 (20% in dimethyl sulfoxide; Molecular Probes). After 3 min, the cells were superfused with solution A (pH 7.3; 0.8 ml/min) for at least 30 min to allow the washout of the extracellular dye and intracellular Indo-1 deesterification. The myocytes were electrically stimulated via platinum wire electrodes with a frequency of 0.5 Hz. The experiments were performed at room temperature to minimize loss of the Ca²⁺ indicator from the cells (Spurgeon et al., 1990). [Ca]_i was recorded from a single myocyte using a dual-emission microfluorescence system (PTI, Princeton, NJ). The dual-emission wavelength ratio 405/495 was used as an index of [Ca]_i. The fluorescence data were acquired at 20 Hz. Data acquisition and processing were supported by software (FeliX, Version 1.1; PTI, Princeton, NJ) for intracellular calcium measurement.

SR Ca²⁺ Uptake Measurement. Frozen hearts were homogenized in 250 mM sucrose, 10 µM cantharidin, and 30 mM histidine (pH 7.0; Lindemann et al., 1983). Cantharidin was added to inhibit protein phosphatases. Cantharidin (10 µM) completely inhibits type 1 and type 2A phosphatases (Neumann et al., 1995). Ca²⁺ uptake in homogenates was measured by the microfiltration technique (Martonosi and Feretos, 1964). The reaction buffer contained 50 mM MOPS (pH 7.0), 3 mM MgCl₂, 100 mM KCl, 5 mM NaN₃, 10 mM potassium oxalate, 0.5 mM EGTA, 10 µM cantharidin, and different CaCl₂ concentrations to give pCa values of 7.49 or 6.71. Free Ca²⁺ concentrations were calculated according to the method of Briggs et al. (1992). Ca²⁺ uptake was measured by preincubation of homogenates with the antibody for 20 min on ice. Ca²⁺ uptake could be stimulated by preincubation of homogenates with an anti-PLB antibody (2D12) for 20 min on ice. Ca²⁺ uptake was initiated by the addition of 3 mM ATP and then performed at 37°C. Aliquots of 100 µl were filtered at various time points on 0.22-µm filters (GS type; Millipore) and washed twice with 5 ml of 150 mM NaCl. ⁴⁵Ca²⁺ was measured by scintillation counting.

Na⁺ Gradient-Dependent Ca²⁺ Uptake. Hearts of wild-type and transgenic mice were homogenized for 10 s with a Polytron in 1.4 ml of 560 mM NaCl, 40 mM MOPS/Tris, pH 7.4 and spun for 5 min at 11,000 rpm at 4°C. The pellet was resuspended in 1 ml of 140 mM NaCl, 10 mM MOPS/Tris, pH 7.4, and homogenized with 20 strokes in a Teflon-glass homogenizer. After centrifugation for 5 min at 11,000 rpm at 4°C, the pellet was washed twice with 140 mM NaCl, 10 mM MOPS/Tris, pH 7.4. The resuspended pellet was centrifuged for 5 min at 2000 rpm at 4°C to remove aggregated particles. The supernatant was collected and centrifuged for 5 min at 11,000 rpm at 4°C to pellet the membrane vesicles. Vesicles were resuspended in the same solution to 3 mg of protein/ml. The Na⁺ gradient-dependent Ca²⁺ uptake was measured by diluting 10 µl of membrane vesicles into 240 µl of Ca²⁺ uptake medium containing 140 mM KCl or NaCl, 10 mM MOPS/Tris, pH 7.4, 10 µM CaCl₂, 0.3 µCi of ⁴⁵CaCl₂, and 0.36 µM valinomycin (37°C). The reaction was terminated after 1 s by the addition of 30 µl of 140 KCl and 10 mM EGTA and the subsequent addition of 1 ml of 140 mM KCl and 1 mM EGTA at 4°C. The filter was washed twice with 3 ml of 140 mM KCl and 1 mM EGTA and then subjected to scintillation counting. Data are presented as the Na⁺ gradient-dependent uptake of ⁴⁵Ca²⁺ (uptake in the NaCl medium is subtracted from uptake in the KCl medium).

Data Analysis. Data shown are mean ± S.E. Statistical significance was estimated with Student's t test for paired and unpaired observations as appropriate. Statistical significance was calculated between all groups. Histograms were compared using ANOVA followed by Bonferroni's t test. P < .05 was considered significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Heart and Body Weight. CSQ overexpression alone was accompanied by an increase in heart weight of approximately 86% in comparison with control and NCX transgenic mice (Fig. 1A). This is in accordance with a previous report (Jones et al., 1998). Co-overexpression of CSQ and NCX further augmented heart weight by approximately 119% of wild-type values. Thus, the heart weight in the double transgenic mice was even higher than that in mice that overexpressed only CSQ. The body weight was not different between CSQ, NCX transgenic, and control mice (Fig. 1B). However, the body weight was decreased (by 28%) in the double transgenic mice. The resulting heart/body weight ratio (Fig. 1C) exhibited an increased ratio for CSQ and double transgenic mice in comparison with control and NCX mice. The extreme increase of heart/body weight ratio in the double transgenic mice is due to the excessive elevated heart weight but also due to a diminished body weight.

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Comparison of heart and body weight from wild-type (WT) and transgenic animals. Shown are ventricular heart weight (A), body weight (B), and heart/body weight ratio (C). *, significant difference versus wild type. +, significant differences versus CSQ and NCX/CSQ.

Histological Examinations. Light microscopical examination of the hearts showed regularly orientated cardiomyocytes in all lineages without any evidence of necrotic changes or inflammatory alterations (Fig. 2, A-D). In the transgenic hearts, the sections stained with elastica van Gieson revealed neither enhanced interstitial fibrosis nor tissue scars compared with wild-type hearts. The mice used in this study were 6 weeks old and showed no histological alterations. This is in contrast to results of Jones et al. (1998), in which 3-month-old mice overexpressing CSQ had histological signs of hypertrophy. However, it is known that the -MHC promoter that drives the CSQ overexpression can be more active in older mice than in younger, and hence the histological alterations might aggravate at a higher age.

View larger version (134K): [in this window] [in a new window]   Fig. 2.   Histological examinations of cardiac sections from wild-type (A), CSQ (B), NCX (C), and NCX/CSQ (D) hearts. Cardiac tissue was stained as described under Materials and Methods.

Effects on Time Parameters. The basal time parameters of contraction [TPT, RT, and total contraction time (TCT)] were unchanged in heart muscle strips from NCX in comparison with wild-type mice. However, basal TPT, RT, and TCT were prolonged in CSQ and double transgenic mice (Table 1). The RT was prolonged after stimulation by caffeine (5 mM) in all four groups, but the effect was less pronounced in CSQ mice. However, stimulation with 12.6 mM Ca²⁺ did not further increase RT (Table 1).

Effects on Force of Contraction. The concentration-response curve for the positive inotropic effect of Ca²⁺ was shifted to higher concentrations of Ca²⁺ in the CSQ (EC₅₀ = 4.1 ± 0.4; P < .05) and double transgenic (EC₅₀ = 4.3 ± 0.5; P < .05) mice compared with wild-type mice (EC₅₀ = 3.2 ± 0.1; Fig. 3). However, the EC₅₀ in NCX transgenic mice was slightly decreased (2.7 ± 0.6) in comparison with wild-type mice.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Concentration-dependent effect of Ca2+ on force of contraction in isolated heart muscle strips of wild-type (WT) and transgenic mice. These were significant increased EC50 values for CSQ and double transgenic mice compared with wild-type mice.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Concentration-dependent effects of caffeine on force of contraction and diastolic tension in isolated heart muscle strips of wild-type (WT) and transgenic mice. A, original recording. B, maximum inotropic effect of caffeine in percentage. C, increase in diastolic tension (in mN). *, significant difference versus wild type. +, significant differences versus control. ×, significant differences versus NCX and NCX/CSQ.

View larger version (21K): [in this window] [in a new window]   Fig. 5.   Concentration-dependent effect of isoproterenol (0.001-10 µM) on force of contraction in isolated heart muscle strips of wild-type (WT) and transgenic mice. *, significant difference versus wild type. +, significant differences versus NCX and NCX/CSQ.

Overexpression-Induced Alterations in Protein and mRNA Expression. To test the hypothesis that alterations in protein and mRNA expression are responsible for the functional alterations and excessive Ca²⁺ storage in CSQ mice (Jones et al., 1998) and that co-overexpression of CSQ and NCX influences these expression patterns, we determined the expression levels of proteins responsible for intracellular Ca²⁺ handling. The protein and mRNA expression of CSQ was increased in both CSQ-overexpressing mice and the double transgenic mice, and the protein and mRNA level of the NCX was increased as expected only in NCX and double transgenic mice (data not shown). Both the CSQ and NCX antibody detected the transgenic protein with a strong signal but were not able to detect the endogenous mouse protein. The expression for SERCA and PLB, proteins important for the Ca²⁺ uptake into the SR, was elevated in CSQ mice by 83% for SERCA (Fig. 6A) and by 80% for PLB (Fig. 6B). The protein levels for SERCA and PLB were unchanged in NCX mice. Strikingly, double transgenic mice revealed protein levels of SERCA and PLB similar to those of wild type. The protein expression for the TnI, which regulates the Ca²⁺ sensitivity of the myofilaments, was unaltered between all groups (Fig. 6C). Parallel results were obtained for mRNA expression. The mRNA expression of ANF as a marker for heart hypertrophy was increased in CSQ and double transgenic mice (Fig. 7B). Interestingly, the double transgenic mice exhibited even higher levels of ANF mRNA (162%) than CSQ-overexpressing mice alone, which is consistent with the hypertrophy data described above (Fig. 1).

View larger version (15K): [in this window] [in a new window]   Fig. 6.   Expression of cardiac proteins in homogenates from wild-type (WT) and transgenic ventricles; 50 µg of homogenate was subjected to SDS-polyacrylamide gel electrophoresis, blotted to nitrocellulose, and incubated with antibodies specific for SERCA (A), PLB (B), and TnI (C). *, significant difference versus wild type.

View larger version (16K): [in this window] [in a new window]   Fig. 7.   Expression of mRNA in total RNA isolated from wild-type (WT) and transgenic ventricles. Membranes were hybridized with 32P-labeled specific probes for SERCA (A) and ANF (B). *, significant difference versus wild type. +, significant differences versus CSQ and NCX/CSQ.

Alterations in SERCA- and NCX-Mediated Ca²⁺ Uptake. Next, we studied the SERCA-mediated Ca²⁺ uptake into the SR (Fig. 8A). The maximum rate of Ca²⁺ uptake was increased in CSQ mice by approximately 152% compared with wild type under linear uptake rate conditions at pCa 7.5. The Ca²⁺ uptake in double transgenic and NCX mice was similar to that of wild type. SR-Ca²⁺ uptake activity in the presence of an anti-PLB antibody was greatly stimulated in all groups.

View larger version (35K): [in this window] [in a new window]   Fig. 8.   A, Ca2+ uptake properties of the SR in ventricles from wild-type (WT) and transgenic mice. Tissue was homogenized and SR-Ca2+ uptake was determined in the absence (Ab) or presence (+Ab) of an anti-PLB antibody. B, Na+-dependent Ca2+ uptake of NCX assayed in membrane vesicles. *, significant difference versus Ab. +, significant differences versus wild type.

Caffeine-Induced Ca²⁺ Release in Unstimulated Cells. Isolated myocytes were labeled with Indo-1 and superfused with buffer in the absence or presence of 5 mM caffeine. Even in the absence of caffeine, spontaneous Ca²⁺ transients were noted in CSQ and double transgenic mice but not in wild-type and NCX mice. After the application of 5 mM caffeine, Ca²⁺ release occurred in CSQ mice as demonstrated by a representative tracing in Fig. 9A. The data are summarized in Fig. 9B and indicate 42% increase in indo fluorescence induced by caffeine in CSQ mice. The caffeine-induced Ca²⁺ release in double transgenic mice increased fluorescence by approximately 19%. However, no Ca²⁺ release in the presence of caffeine was detectable in NCX and wild-type mice.

View larger version (31K): [in this window] [in a new window]   Fig. 9.   Caffeine-induced Ca2+ transients in isolated myocytes from wild-type (WT) and transgenic mice. A, original tracings for CSQ in the absence or presence of 5 mM caffeine. B, data are summarized. *, significant difference versus corresponding basal Ca2+ transients.

Effect of Caffeine on L-Type Ca²⁺ Currents. Caffeine affected the L-type Ca²⁺ currents in cells isolated from transgenic hearts. With wild-type mice, only a very weak increase in current was detectable (Fig. 10A), but a large increase (295%) occurred in cells from CSQ mice. Peak current data are summarized in Fig. 10B. NCX and double transgenic mice exhibited caffeine-induced L-type Ca²⁺ currents more similar to wild-type mice. However, a reduced basal current was shown in CSQ mice (to 46%) and double transgenic mice (to 37%) compared with wild type. Interestingly, the inactivation kinetics of the L-type Ca²⁺ current were slowed in CSQ and double transgenic mice. Mouse atrial myocytes exhibited a fast and a slow component of inactivation of L-type Ca²⁺ current as described previously (Masaki et al., 1998). It has been suggested that the fast component of inactivation is Ca²⁺-dependent, whereas the slow component is voltage-dependent. To determine whether parameters of inactivation differed, we analyzed inactivation properties in control and transgenic mice (Table 2). The time course of inactivation was similar in NCX and control mice. However, both components of inactivation were prolonged in CSQ and double transgenic mice (Table 2).

View larger version (26K): [in this window] [in a new window]   Fig. 10.   Ca2+ currents in isolated myocytes from wild-type (WT) and transgenic mice. A, representative recording in the absence or presence of 5 mM caffeine. B, data are summarized. *, significant difference versus corresponding basal Ca2+ current. +, significant differences versus wild type.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

CSQ Mice. The starting point of our study was the observation that mice overexpressing CSQ developed severe cardiac hypertrophy (Jones et al., 1998; Sato et al., 1998; present study). This hypertrophic response in CSQ mice was associated with reinduction of ANF in the heart (Sato et al., 1998; Cho et al., 1999; present study). ANF is expressed in the fetal heart and is shut off in the adult heart but can be reexpressed under pathological conditions. The reexpression of ANF is typically noted in failing human hearts and in animal models of heart failure (Arai et al., 1993; Brunner, 1999). Mice that overexpress CSQ in the heart are considered as an unique model for cardiomyopathy because the primary abnormality lies in intracellular Ca²⁺ regulation and is followed by concentric left ventricular hypertrophy that progresses to dilated cardiomyopathy (Cho et al., 1999). Hence, we wanted to study alterations of the Ca²⁺ homeostasis in this model.

NCX Mice. As expected, the overexpression of NCX resulted in an enhanced Na⁺ gradient-dependent Ca²⁺ uptake, indicating the functional activity of the increased NCX expression in these mice. In contrast to CSQ mice, overexpression of NCX did not lead to cardiac hypertrophy (Adachi-Akahane et al., 1997; present study), confirmed by unchanged heart weight, cell capacitance, and ANF mRNA expression. Moreover, the expression of the SR proteins PLB and SERCA was unchanged and was consistent with a normal SR-Ca²⁺ uptake in biochemical experiments. In agreement with these biochemical findings, the RT was not prolonged, and the inotropic responses to Ca²⁺ and caffeine in preparations from NCX mice were similar to those in wild-type mice. Likewise, the Ca²⁺ transients and Ca²⁺ currents in the absence and presence of caffeine were the same in mice overexpressing NCX compared with wild-type mice. Taken together, the data revealed no obvious defective Ca²⁺ handling in mice overexpressing NCX.

Double Transgenic Mice. The co-overexpression of CSQ and NCX was successful. As expected, the overexpression of NCX resulted in an enhanced Na⁺ gradient-dependent Ca²⁺ uptake, indicating the functional activity of the increased NCX expression in these mice. NCX extrudes Ca²⁺ from the cytosol, and we hypothesized this might prevent the elevated Ca²⁺ storage of the SR and might normalize the cardiac hypertrophy of CSQ mice. Surprisingly, this was not the case. Co-overexpression of CSQ and NCX genes in transgenic mice led to further cardiac hypertrophy in comparison with mice overexpressing CSQ alone as indicated by the increased heart/body weight ratio. ANF mRNA was even further increased in double transgenic mice compared with mice overexpressing CSQ alone. This increase in hypertrophy correlated with mortality data as no double transgenic mice survived beyond 6 months of age, whereas approximately half of the CSQ-overexpressing mice and all NCX and wild-type mice survived for 6 months and longer.