INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

MYOCARDIAL RELAXATION IS ACHIEVED through the combined effects of a number of cellular proteins including the sarco(endo)plasmic reticulum (SR) phospholamban (PLB)-regulated Ca²⁺-ATPase (SERCA) (10, 23) and the sarcolemmal Na⁺/Ca²⁺ exchanger (26, 29, 35). In cardiac hypertrophy, SERCA2a (the predominant isoform in cardiac myocytes) mRNA and protein have been shown to be decreased in animal models (11, 13, 18); and following human heart failure, some studies have shown reduced SR Ca²⁺-ATPase and PLB mRNA levels (1). The diminished function of the SR and particularly the decrease in SR Ca²⁺-ATPase expression and activity provide a molecular basis for the prolonged cytosolic Ca²⁺ transients and slowed decline of Ca²⁺ during relaxation in isolated myocytes from hypertrophied and failing myocardium. In contrast, the amount of Na⁺/Ca²⁺ exchanger in a cardiac myocyte generally increases with cardiac hypertrophy and failure, the result of which may compensate for a reduction in SR Ca²⁺-ATPase to preserve diastolic function (15, 37).

ANG II and ANG-converting enzyme (ACE) inhibitors contribute to the development and regression of cardiac hypertrophy, respectively, both in vitro and in vivo (22, 27, 33, 44). Release of ANG II, for example, mediates stretch-induced hypertrophy of cardiac myocytes in vitro (32). ACE inhibitors improve diastolic function and alter the abundance of several transcripts implicated in calcium cycling, including those of the SR Ca²⁺-ATPase (9, 43). We previously demonstrated, however, that the amount of SERCA2 mRNA is not an accurate predictor of SR Ca²⁺-ATPase protein abundance (6, 31) and that administration of ramipril alone to normal rats could modify cardiac SERCA2 and PLB mRNA expression without affecting the abundance of either protein (6). In guinea pigs, high-dose ACE inhibition prevents a decrease in SR calcium cycling protein abundance in the pressure-overloaded heart (38). In compensated cardiac hypertrophy, we previously showed that subantihypertensive doses of ramipril similarly increase SR Ca²⁺-ATPase protein abundance and improve uptake in crude cardiac homogenates (6). It remains unclear how low-dose ACE inhibition in established cardiac hypertrophy actually improves myocyte relaxation (6). We therefore performed the present study to test whether rats with established cardiac hypertrophy and treated with ramipril show improved relaxation in isolated myocytes. Specifically we hypothesized that subantihypertensive doses of ramipril would improve myocyte relaxation through 1) enhanced SR Ca²⁺-ATPase activity and 2) reduced Ca²⁺ extrusion via the Na⁺/Ca²⁺ exchanger. Because the Na⁺/Ca²⁺ exchanger is regulated transcriptionally during the perinatal period (19, 31), we also predicted that any decrease in exchanger activity would reflect diminished mRNA and protein expression.

We demonstrate that in compensated cardiac hypertrophy the slowing of myocyte relaxation after rapid cooling is in fact reversible with ramipril treatment. Specifically, ramipril treatment improves myocyte time to peak (TTP) contraction and normalizes myocyte relaxation and Ca²⁺ handling. The improvement in relaxation is mediated primarily through improved SR uptake by the SR Ca²⁺-ATPase and through enhanced Ca²⁺ efflux from the cell independent of measurable changes in Na⁺/Ca²⁺ exchanger mRNA and protein expression.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Model of Cardiac Hypertrophy

1

1

Study of Cardiac Myocytes

Myocyte cell isolation. Myocytes were isolated by enzymatic dissociation with retrograde perfusion of the whole heart as previously described (28, 39) except that the low-Ca²⁺ medium was composed of the following (in mM): 120 NaCl, 5.4 KCl, 5 MgSO₄, 5 pyruvate, 20 glucose, 20 taurine, 10 HEPES, and 5 nitroloacetic acid (NTA) and 40 µM free Ca²⁺. After 5 min in low-Ca²⁺ solution, the heart was perfused with 0.3 mg/ml collagenase (Worthington) and 0.6 mg/ml hyaluronidase (Sigma). Enzymes were dissolved in the low-Ca²⁺ medium without NTA and with a free Ca²⁺ concentration ([Ca²⁺]) of 200 µM. The left ventricle was chopped and shaken in fresh collagenase mixture for two periods of 5 min each. After filtering was completed, the cells were placed in DMEM solution buffered with 25 mM HEPES (GIBCO-BRL) at room temperature. In this medium the cells remained viable and could be used over a period of 3-4 h.

Cell shortening and Ca²⁺ measurements. Cell shortening was measured using a video-based edge-detection system described by Steadman and colleagues (36). Fluorescence measurements indicating changes in intracellular [Ca²⁺] ([Ca²⁺]_i) were made on the same cells employed in the cell-shortening measurements using the dual-emission fluorescent dye indo 1 as previously described (28, 39). Cells were placed in a Plexiglas superfusion chamber (60-µl vol) situated on the stage of an epifluorescence microscope. Cell adhesion to the floor of the chamber was improved by a thin coating of mouse laminin (GIBCO-BRL). Cells were field stimulated at 0.5 Hz via a pair of platinum electrodes placed on either side of the experimental chamber and continuously superfused at a rate of 2-3 ml/min with a Tyrode solution containing (in mM) 140 NaCl, 6 KCl, 1 MgCl₂, 2 CaCl₂, 10 glucose, and 10 HEPES, with pH 7.4 ± 0.01. All experiments were carried out at room temperature (22°C) except during the cooling periods that brought about rapid cooling contractures (RCCs). When Na⁺-free/Ca²⁺-free solution was used, Na⁺ was substituted by 140 mM Li⁺, Ca²⁺ was omitted, and 0.75 mM EGTA was added, at pH 7.4 ± 0.01. When caffeine was used, it was added to normal Tyrode for a final concentration of 10 mM.

RCCs. The methods used to invoke RCCs have been well described (3, 4, 8, 21, 28, 40). Briefly, RCCs were generated 2 s after cessation of field stimulation. The temperature of the solution superfusing the cells was changed from 22°C to 1°C in <1 s (39). This produced an increase in cytoplasmic [Ca²⁺] and a contracture, which have been described in detail (4). Rapid switching of the solutions was achieved by a system of solenoid valves placed near the cell chamber. Solutions were kept cold by placing them in a water-ethylene glycol mixture (4:1 ratio) kept at 3.5°C in a circulating water bath. The tubing carrying the solutions was also jacketed with cold water/ethylene glycol solution. During cooling, solutions flowed through the cell chamber at a rate of 12-15 ml/min. After the cooling contracture reached a plateau, the cell was rewarmed by switching the solutions back to room temperature allowing the cell to relax.

Data acquisition and statistics. Signals from the video edge-detection system and the fluorescence apparatus were recorded simultaneously on tape and computer using a digitization rate of 0.1 kHz and Axotape 2.0 software (Axon Instruments, Foster City, CA). For the measurement of the myocyte twitch and fluorescence transient parameters, four consecutive twitches and transients were signal averaged. TTP was measured between the point before the initial increase of the signal and the peak of the signal. Time to 50% relaxation or decline (R₅₀) was measured between the peak of the signal to the point on the declining phase corresponding to half of the total size of the twitch or transient. Time to 90% relaxation or decline (R₉₀) was measured between the peak of the signal to the point where the declining phase had recovered by 90% of the total size of the twitch or transient.

Measurement of myocyte cell size. Video hard copies of a representative population (n  30) of myocytes were taken from each of the heart preparations using a Video Graphic printer (UP 701, Sony, Japan). Two-dimensional surface areas of the cells were obtained from these hard copies using a digitizing tablet and its associated software (VIDS III, Analytical Measuring Systems, Cambridge). Cell width and cell length were measured from each myocyte and averaged for each heart.

RNA and Protein

RNA isolation. Total RNA was prepared from freshly isolated rat left ventricles as previously described (6). RNA was prepared from eight animals from each experimental group and stored as an ethanol precipitate at 20°C.

RNase protection assays. Na⁺/Ca²⁺ exchanger (NCX1) mRNA was measured by RNase protection assay (RPA) as previously described (19). cRNA probes were prepared using T₇ RNA polymerase in the presence of 1.85 MBq ³²UTP and 1.0 µg of linearized plasmid [BamHI-digested plasmid pRCNaCa10 and rat glyceraldehyde-3-phosphate dehydrogenase (GAPDH) plasmid (Ambion)]. Full-length probes were 699 and 383 bp and had specific activities of 3.2 × 10⁸ and 1.8 × 10⁸ cpm/mg, respectively. RPAs were performed on 5 µg of total RNA using 5 × 10⁴ counts per minute (cpm) Na⁺/Ca²⁺ exchanger probe and 5 × 10⁴ cpm GAPDH probe per reaction. After separation on a 6% denaturing polyacrylamide gel, dried gels were exposed to X-ray film for between 2 h (GAPDH) and 90 h (Na⁺/Ca²⁺ exchanger) and quantitated using a phosphoimager. Results were expressed as the densitometric ratio of Na⁺/Ca²⁺ exchanger to GAPDH.

Protein analysis by Western blotting. Protein preparation and Western blots were performed as previously described (6, 19). Frozen tissue (50-150 mg) was homogenized in seven volumes of lysis buffer [in mM: 20 HEPES, 4 EGTA, and 1 dithiothreitol (pH 7.5) containing 1 phenylmethylsulfonyl fluoride and 0.3 leupeptin], and 100 µg of total protein in one volume of loading buffer were separated on 4% SDS-7.5% polyacrylamide gels and transferred to nitrocellulose membranes (Hybond-C super, Amersham). Protein was detected using a 1:1,000 dilution of a polyclonal Na⁺/Ca²⁺ primary antibody (Swant) and a 1:15,000 dilution of a horseradish-peroxidase-coupled secondary antibody (goat anti-rabbit, Dako). Enhanced chemiluminescence films were scanned at 176 µm of resolution using a Molecular Dynamics laser densitometer and analyzed by the One dimensional software package (12). To ensure equal protein loading, membranes were reprobed with a monoclonal antibody to myosin heavy chain (MHC, Novocastra Laboratories) and subsequently stained with 0.15% amido black as previously described (6).

Statistics

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Cardiac and myocyte hypertrophy. Nine weeks after aortic constriction, heart wt increased by 36% (P < 0.001). At this time, body wt and tibial length were unchanged between each of the experimental groups (P > 0.05). Heart weight-to-tibial length ratio increased by 43% (P < 0.001) after cardiac hypertrophy but was not significantly reduced after ramipril treatment. Aortic constriction also led to a significant increase in myocyte cell length (P < 0.01) and width (P < 0.05). Four weeks of low-dose ramipril treatment did not significantly decrease cell size or reduce cardiac mass (P > 0.05).

Myocyte contractile properties. With myocyte hypertrophy, the TTP of contraction slowed and the time of relaxation lengthened (Fig. 1). After ramipril treatment these twitch characteristics improved; however, slowing of both contraction and relaxation in hypertrophied myocytes was not accompanied by similar changes in intracellular Ca²⁺ transients. Specifically, the TTP of contraction slowed by 36% (control + vehicle and hypertrophy + vehicle; P < 0.001) after 9 wk of aortic constriction (Fig. 2A). Ramipril did not change the twitch characteristics of the controls; however, after 4 wk of ramipril treatment the TTP of contraction was significantly improved (P < 0.05) in the hypertrophied group. The TTP of the intracellular Ca²⁺ transients (as measured by indo 1 fluorescence) were not significantly different between any groups, suggesting that myofilament sensitivity had been altered.

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Typical twitch characteristics of myocytes from three experimental groups. A: change in cell length, with data normalized to maximum contraction. B: changes in indo 1 fluorescence normalized to maximum fluorescence. Individual curves are as follows: line a, control + vehicle; line b, hypertrophy + ramipril; and line c, hypertrophy + vehicle.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Summed data from the twitch and fluorescence analyses described in Fig. 1. A: time to peak of contraction and indo 1 fluorescence. B: time to 50% relaxation (R50) of cell length and indo 1 fluorescence. C: time to 90% relaxation (R90) of cell length and indo 1 fluoresecence. *P < 0.05, hypertrophy + vehicle vs. hypertrophy + ramipril; **P < 0.01, control + vehicle vs. hypertrophy + vehicle; and ***P < 0.001, control + vehicle vs. hypertrophy + vehicle. From 5-7 hearts: control + vehicle, n = 19 cells; control + ramipril, n = 13 cells; hypertrophy + vehicle, n = 18 cells; and hypertrophy + ramipril, n = 15 cells.

RCC experiments. RCCs provide a useful means of assessing SR Ca²⁺ content (4, 8, 21, 28, 40). Rapidly cooling cardiac cells or tissue (to about 1°C in <1 s) induces Ca²⁺ release from the SR and a resultant contracture. The SR releases essentially all stored Ca²⁺ by a mechanism that bypasses Ca²⁺ influx and Ca²⁺-induced Ca²⁺ release. Rewarming the tissue produces a relaxation brought about by the same mechanisms operating during relaxation of a normal twitch. The rewarming can be carried out in Na⁺-free/Ca²⁺-free solution to inhibit the Na⁺/Ca²⁺ exchanger or in the presence of 10 mM caffeine to inhibit the ability of the SR to retain Ca²⁺.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   A: typical relaxation profiles of myocytes following rapid cooling contractures (RCCs) in normal Tyrode solution (line a), Na+-free/Ca2+-free solution (line b), and 10 mM caffeine (line c). B: decline in indo 1 fluorescence. Typical relaxation profiles in Tyrode solution: control + vehicle (line a), hypertrophy + ramipril (line b), and hypertrophy + vehicle (line c).

View larger version (18K): [in this window] [in a new window]   Fig. 4.   R50 of cell length after RCCs in normal Tyrode solution; n = 13-19 cells from 5-7 hearts for each analysis. A: relaxation of cell length. **P < 0.01, hypertrophy + vehicle vs. hypertrophy + ramipril; and ***P < 0.001, control + vehicle vs. hypertrophy + vehicle. B: R50 for the decline of indo 1 fluorescence. *P < 0.05, hypertrophy + vehicle vs. hypertrophy + ramipril; and **P < 0.01, control + vehicle vs. hypertrophy + vehicle.

Relaxation in Na-free/Ca-free solution. Relaxation in Na⁺-free/Ca²⁺-free solution permitted the examination of changes in the absence of a functional Na⁺/Ca²⁺ exchanger (Fig. 5). Figure 5A shows that in Na⁺-free/Ca²⁺-free solution, cardiac hypertrophy resulted in a 63% slowing of the R₅₀ of cell length after RCCs. After ramipril treatment, relaxation was faster (hypertrophy + vehicle vs. hypertrophy + ramipril; P < 0.05). Figure 5B shows the corresponding results for indo 1 fluorescence changes under these conditions. In cardiac hypertrophy a 60% slowing in the decline of fluorescence was observed during relaxation. Fluorescence declined more rapidly with ramipril treatment (hypertrophy + vehicle vs. hypertrophy + ramipril; P < 0.05). Ramipril did not alter the R₅₀ of cell length or the R₅₀ of the decline in fluorescence of control myocytes. These experiments suggest that SR Ca²⁺ uptake in the intact cardiac myocyte is reduced in cardiac hypertrophy, but uptake significantly improves after ramipril treatment. SR Ca²⁺-ATPase activity thus appears to be improved. That myocytes in Na⁺-free/Ca²⁺-free solution showed a similar slowing of relaxation in hypertrophy that improved with ramipril treatment suggests that the Na⁺/Ca²⁺ exchanger may not contribute greatly to the improved relaxation.

View larger version (19K): [in this window] [in a new window]   Fig. 5.   R50 of cell length after RCCs in Na+-free/Ca2+-solution; n = 13-19 cells from 5-7 hearts. A: relaxation of cell length. *P < 0.01, hypertrophy + vehicle vs. hypertrophy + ramipril; and **P < 0.01, control + vehicle vs. hypertrophy + vehicle. B: changes in indo 1 fluorescence. *P < 0.05, control + vehicle vs. hypertrophy + vehicle; and *P < 0.05, hypertrophy + vehicle vs. hypertrophy + ramipril.

Relaxation in caffeine. In caffeine, the open probability of the SR Ca²⁺-release channels is greatly increased so that the SR is no longer able to retain Ca²⁺. Relaxation in caffeine is then mainly a function of the Na⁺/Ca²⁺ exchanger. To test any potential role of the exchanger, experiments were performed in the presence of caffeine. Figure 6 shows pooled data of the R₅₀ myocyte relaxation and the decline in fluorescence in caffeine during rewarming after rapid cooling. No changes in relaxation rate of cell length were observed in any groups when the relaxations were carried out in the presence of 10 mM caffeine (Fig. 6A). Caffeine is, however, well known to alter myofilament sensitivity making interpretation of the cell length changes difficult. No significant difference in the rates of indo 1 fluorescence decline between control + vehicle, control + ramipril, or hypertrophy + vehicle could be demonstrated; however, the rate of decline of the hypertrophy + ramipril group was 30% faster than the hypertrophy + vehicle group; P < 0.05 (Fig. 6B). It is unclear why the rate of decline of indo 1 fluorescence did not improve in the hypertrophy + vehicle group given that we find a threefold increase in the presence of the exchanger protein (see RNA and protein measurements). The data show that ramipril treatment of the hypertrophied heart improves Ca²⁺ extrusion from the cells, probably via the Na⁺/Ca²⁺ exchanger.

View larger version (19K): [in this window] [in a new window]   Fig. 6.   R50 of cell length after rapid cooling in 10 mM caffeine solution; n = 13-19 cells from 5-7 hearts. A: relaxation of cell length; B: change in indo 1 fluorescence. *P < 0.05, hypertrophy + vehicle vs. hypertrophy + ramipril.

RNA and protein measurements. To demonstrate that improved myocyte relaxation in caffeine after ramipril treatment was due to increased Na⁺/Ca²⁺ exchanger abundance, exchanger mRNA and protein were measured. RNA abundance was quantified using the RPA, and GAPDH mRNA abundance was used to normalize data (Fig. 7). No significant difference in GAPDH transcript abundance could be demonstrated between the four groups. After cardiac hypertrophy, Na⁺/Ca²⁺ exchanger mRNA abundance almost doubled when compared with controls on vehicle (*P < 0.05). When treated with ramipril, the hypertrophy group had even higher quantities of exchanger mRNA albeit not a significant increase above the hypertrophied group. Unexpectedly, exchanger mRNA expression was most abundant in the control + ramipril groups (Fig. 7B), suggesting posttranscriptional control of exchanger protein abundance.

View larger version (47K): [in this window] [in a new window]   Fig. 7.   A: example of an RNase protection assay to quantify Na+/Ca2+ exchanger and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA abundance. Top bands represent protected fragments of cardiac Na+/Ca2+ exchanger (NCX1), and bottom bands indicate protected fragment of GAPDH. Protected fragments for NCX1 and GAPDH were 625 and 316 bp, respectively. Samples were from the four experimental groups as indicated, and atrial and liver RNA preparations were used as controls. The minus sign indicates lanes that were loaded with full-length probes for either NCX1 or GAPDH; the plus sign indicates lanes that were loaded with full-length probes for NCX1 or GAPDH and treated with RNases. B: summed data showing relative changes in Na+/Ca2+ exchanger expression among the four groups; n = 6-8 hearts from each group. *P < 0.05, control + vehicle vs. hypertrophy + vehicle; **P < 0.01, control + vehicle vs. control + ramipril; and *P < 0.05, control + vehicle vs. hypertrophy + ramipril.

View larger version (42K): [in this window] [in a new window]   Fig. 8.   A: Western blot of the Na+/Ca2+ exchanger protein. Four specific bands for the exchanger were seen (arrows) consistent with what we and others have previously shown (19). Bands 1 and 2 represent the 160-kDa and 140-kDa Na+/Ca2+ rat exchanger, respectively; bands 3 and 4 represent the 70-kDa and 40-kDa proteolytic fragments of the protein. Total amounts of exchanger protein were quantified by addition of the four specific bands. Control + vehicle (section a); hypertrophy + vehicle (section b); control + ramipril (section c); and hypertrophy + ramipril (section d). B: summed data showing relative changes in Na+/Ca2+ exchanger protein abundance among the four experimental groups; n = 6-8 hearts. Equal protein loading was ensured by amido black staining and use of myosin heavy chain antibodies (as described in METHODS). *P < 0.05, control + ramipril vs. hypertrophy + ramipril; and ***P < 0.001, control + vehicle vs. hypertrophy + vehicle.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Myocyte function as measured with RCCs. The proteins responsible for relaxation in the rat are primarily the SR Ca²⁺-ATPase and secondarily the Na⁺/Ca²⁺ exchanger (2, 23, 26, 42). The relative contribution of these proteins to relaxation varies according to the pathophysiological state (16, 30, 37). Rapid cooling of cardiac cells to +1°C in <1 s induces Ca²⁺ release from the SR resulting in contracture (4, 8, 21). Under such conditions, cytosolic Ca²⁺ extrusion systems are inhibited. After rewarming these systems are reactivated; Ca²⁺ is removed from the cytoplasm, and the cell relaxes. Bers and Bridge (3) used this method to inhibit different relaxation systems selectively. Relaxation after RCCs thus can be used to differentiate between individual components of relaxation. As with relaxation after a twitch, relaxation of hypertrophied cells after rapid cooling in normal Tyrode solution is significantly slower; however, unlike the twitch data, the slowed relaxation is accompanied by a slowing in the decline of intracellular Ca²⁺. This result can be explained by the much larger complete release of SR Ca²⁺ produced by cooling, revealing compromised uptake and extrusion processes, whereas smaller releases of Ca²⁺ normally produced by Ca²⁺-induced Ca²⁺ release can be adequately handled.

Differential regulation. In the hypertrophied rat myocardium, Na⁺/Ca²⁺ exchanger mRNA abundance increased consistent with results from other hypertrophy models (14, 19, 25, 26). In our model, Na⁺/Ca²⁺ exchanger (NCX1) mRNA expression increased significantly relative to GAPDH not only in the hypertrophy + vehicle group but also in the control + ramipril and hypertrophy + ramipril groups. Previously in this same model we demonstrated a relative increase in PLB and SERCA2 mRNA in the control + ramipril and hypertrophy + ramipril groups (6). Neither atrial natriuretic factor nor calsequestrin mRNA abundance were altered significantly in any of the groups after ramipril treatment (6) suggesting that the effects on mRNA expression are gene/transcript specific. Administration of ramipril to control groups thus selectively increases the mRNA of SERCA2, PLB, and NCX1. Although it is unclear what mechanisms (transcriptional or posttranscriptional) are responsible for these selective increases, the subsequent discordance between RNA and protein abundance strongly implicates a degree of post-transcriptional regulation.