INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

SODIUM-CALCIUM (Na⁺/Ca²⁺) exchange is the principal means by which intracellular Ca²⁺ (Ca²⁺_i) is removed from cardiac myocytes during diastole (9, 22, 31). Consequently, the activity of this transporter can affect cardiac contractile force by influencing both diastolic Ca²⁺_i levels and the overall cellular Ca²⁺ load. Moreover, alterations in the rate of Ca²⁺_i removal should strongly influence the characteristics of individual contractions by altering the nature of the Ca²⁺ transient. Depending on the existing electrochemical gradients, Na⁺/Ca²⁺ exchange may also contribute to the elevation of Ca²⁺_i (12, 33) and may be involved in Ca²⁺-induced Ca²⁺ release from the sarcoplasmic reticulum (SR) (24, 25). However, our understanding of the regulation of Na⁺/Ca²⁺ exchange activity to accomplish this range of tasks is incomplete.

Perhaps the best illustration of the role of Na⁺/Ca²⁺ exchange in influencing cardiac contractile force is provided by cardiac glycosides (e.g., digoxin). These agents, through inhibition of the Na⁺-K⁺-ATPase, lead to elevations in intracellular Na⁺ (Na⁺_i) levels. This effect mediates inhibition of net Ca²⁺ efflux by Na⁺/Ca²⁺ exchange with a resultant positive inotropic effect (17, 31). Furthermore, elevation of Na⁺_i enhances the ability of reverse Na⁺/Ca²⁺ exchange to operate as a Ca²⁺ entry mechanism, which would also result in positive inotropy (12, 33).

During steady-state stimulation, Na⁺/Ca²⁺ exchange must remove the same amount of Ca²⁺ that enters the myocyte to avoid Ca²⁺_i overload or depletion (14, 30). Because the amount of Ca²⁺ entering the cell via L-type Ca²⁺ channels can vary widely in response to physiological and pharmacological manipulation, it appears that Na⁺/Ca²⁺ exchange is capable of matching these fluctuations. This could be accomplished by a reserve of exchangers and/or by regulating their activity. Existing data suggest that exchanger capacity may be larger than ordinarily required because both the SR and Na⁺/Ca²⁺ exchange can independently mediate rapid relaxation (10). This ability may result from an excess of tonically active exchangers or via recruitment from a pool of regulated exchangers.

Giant patch-clamp studies have established that in addition to transporting Na⁺ and Ca²⁺, the exchanger is also regulated by these ions (19). Under conditions of outward current generation (i.e., reverse Na⁺/Ca²⁺ exchange), application of Na⁺_i not only triggers exchange activity but also induces an inactive state of the transporter (21). This process, termed Na⁺_i-dependent regulation, manifests as a rapid (i.e., ~100 ms) rise of exchange current to a peak, followed by a relatively slow decay (i.e., seconds) to a steady-state level of activity. Ca²⁺_i-dependent regulation, on the other hand, refers to the stimulation of outward exchange current by Ca²⁺_i acting at a site distinct from the transport site (20). Although giant patch-clamp studies have demonstrated this regulatory process to be most responsive within the diastolic/systolic range of Ca²⁺_i concentration ([Ca²⁺]_i) (i.e., EC₅₀  0.3 µM) (19, 20, 27), studies in intact cells point to a much higher potency of Ca²⁺_i to regulate exchange activity (23, 29). The basis(es) for this discrepancy is currently unknown.

Ionic regulation of Na⁺/Ca²⁺ exchange activity has been characterized extensively in giant excised membrane patches from isolated ventricular myocytes and Xenopus laevis oocytes expressing the canine cardiac exchanger NCX1.1 (19-21, 27, 35). Although structure-function studies have identified several regions of the exchanger molecule that play important roles in these regulatory processes (16, 26-28), there is presently little information available concerning the physiological significance of exchanger regulation. To address this issue, we determined the dependence of contractile force on rest interval and stimulation frequency in papillary muscles from two transgenic mouse lines, one of which cardiospecifically overexpressed canine NCX1.1 and the other a deletion mutant of NCX1.1, 680-685, in which both Na⁺- and Ca²⁺-dependent regulatory mechanisms were essentially eliminated.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Production of transgenic mice. The production of mice overexpressing NCX1.1 has been described previously (1, 6, 15, 34, 38). The same transgene construct using the -myosin heavy chain promoter was utilized to produce mice overexpressing 680-685 in cardiac muscle. A DNA cassette encoding for NCX1.1 with amino acids 680-685 deleted was subcloned into the construct used previously. The transgene was then used by the University of California, Los Angeles, Transgenic Core Facility for transgenic mouse production.

Isolation of murine cardiac myocytes. Ventricular myocytes were isolated from adult mouse hearts as described (36), with minor modifications. Briefly, mice were heparinized (10 IU ip) for 10 min, generally anesthetized (5% isoflurane-95% O₂), and euthanized by cervical dislocation. The hearts were rapidly excised and placed in oxygenated Ca²⁺-free Tyrode solution consisting of (in mM) 137 NaCl, 10 D-glucose, 5.4 KCl, 5 HEPES, 0.5 MgCl₂, and 0.3 NaH₂PO₄, pH 7.4 (37°C) with NaOH, at 22°C, and the aortas were cannulated and connected to a perfusion apparatus. Perfusion solutions were equilibrated with 100% O₂ at 37°C and administered at ~2 ml/min as follows: 5 min with Ca²⁺-free Tyrode solution; ~10 min with Ca²⁺-free Tyrode solution containing 1.25 mg/ml collagenase (type 2; Worthington), 0.063 mg/ml protease (type XIV; Sigma), and 0.94 mg/ml fatty acid-free BSA (Sigma); and 10 min with Kraftbrühe solution (KB) consisting of (in mM) 70 L-glutamic acid, 25 KCl, 20 taurine, 10 KH₂PO₄, 10 HEPES, 10 D-glucose, 3 MgCl₂, and 0.5 EGTA, pH 7.4 (37°C) with KOH. After they were removed from the perfusion apparatus, ventricles were teased apart and cells dispersed by trituration. Cells were washed several times in KB and stored at 4°C until use.

Isolation of canine cardiac myocytes. Ventricular myocytes were isolated from adult male mongrel dog hearts as follows. All procedures were conducted at 37°C, and all solutions were equilibrated with 95% O₂-5% CO₂. Dogs were anesthetized with pentobarbital sodium (30 mg/kg) containing heparin (220 IU/kg), and the hearts were rapidly excised through an intercostal incision, fibrillated, and submerged in Tyrode solution consisting of (in mM) 129 NaCl, 20 NaHCO₃, 5.5 D-glucose, 4.0 KCl, 1.8 CaCl₂, 0.9 NaH₂PO₄, and 0.5 MgSO₄, pH 7.4 at 37°C with NaOH. A large wedge of left ventricle was excised around the left anterior descending coronary artery, and the artery was cannulated and flushed with ~50 ml of Ca²⁺-free Krebs solution consisting of (in mM) 118.5 NaCl, 14.5 NaHCO₃, 11.1 D-glucose, 4.8 KCl, 2.7 MgSO₄, and 1.2 KH₂PO₄ containing 0.1% BSA (Sigma), pH 7.4 (37°C) with NaOH. The wedge was mounted on a recirculating pump and perfused at 12 ml/min for ~10-15 min with Ca²⁺-free Krebs solution containing 0.05% collagenase (type 2; Worthington). Epicardial layers were removed, and midmyocardial layers were harvested with a dermatome, minced, and placed in Krebs solution supplemented with 0.5 mM MgSO₄, 0.3 mM CaCl₂, 1.5% BSA, and 0.04% collagenase before they were incubated (with shaking) for 10 min and passed through a nylon mesh (220 µm) for collection of dispersed cells. Tissue fragments were returned to fresh collagenase-containing Krebs solution, and the procedure was repeated four times. Myocytes were centrifuged at 40 g for 2 min, supernatants were discarded, and pellets were resuspended in HEPES-Tyrode solution consisting of (in mM) 132 NaCl, 20 HEPES, 11.1 D-glucose, 5 KCl, 3.2 MgSO₄, and 0.5 CaCl₂ containing 1.5% BSA and 50 mg/ml gentamicin. The myocyte fraction exhibiting the highest ratio of viable to nonviable cells was stored at 4°C until use.

Miscellaneous. Oocytes were prepared from Xenopus laevis. cRNA encoding NCX1.1 and 680-685 were prepared, and ~5 ng cDNA were injected per oocyte exactly as previously described (16). Myocyte membrane "blebbing" was induced by placing cells in a hypotonic solution consisting of (in mM) 67.5 KCl, 9.0 D-glucose, 6.75 HEPES, 4.5 EGTA, and 0.9 MgCl₂, pH 7.2 (22°C) with KOH, at 4°C for several hours before the experiment (21).

Electrophysiological analyses. Outward Na⁺/Ca²⁺ exchange currents were characterized in ~25-µm-diameter membrane patches from oocytes and myocyte "blebs" using the giant excised patch-clamp technique. Outward currents were elicited by rapid (i.e., ~200 ms) application of 100 mM Na⁺_i plus 0-10 µM Ca²⁺_i to the intracellular surface of the patches. Pipette (i.e., extracellular) Ca²⁺ (Ca²⁺_o) to be transported was constant at 8 mM. Cytoplasmic solutions contained (in mM) 100 Li/Na-aspartate, 20 TEA-OH, 20 MOPS, 20 CsOH, 10 EGTA, 1.02-1.5 Mg(OH)₂:2.04-3.0 NH₃SO₃, and 0-9.96 Ca(OH)₂:0-19.92 NH₃SO₃, pH 7.0 (37°C) with MES or LiOH. Extracellular (i.e., pipette) solution contained (in mM) 100 N-methyl-D-glucamine (NMG):MES, 30 HEPES, 30 TEA-OH, 16 NH₃SO₃, 8 CaCO₃, 6 KOH, 0.25 ouabain, 0.1 flufenamic acid, and 0.1 niflumic acid, pH 7.0 (37°C) with MES. All experiments were conducted at 37 ± 1°C.

Contractility measurements. Strain control or transgenic mice were euthanized, and their hearts were rapidly excised and placed in oxygenated Tyrode solution at 22°C containing 30 mM 2,3-butanedione monoxime plus 0.5 mM Ca²⁺. Left ventricular papillary muscles were accessed and tied at both ends using 9.0 nylon suture. Muscles were then placed between a permanently mounted glass hook located on an ~0.25-ml bath and a second hook attached to a capacitive force transducer mounted to a micromanipulator. Muscles were exposed to a constant temperature (37 ± 1°C), regulated flow (~6-7 ml/min) of oxygenated Tyrode solution containing 2 mM Ca²⁺. The muscles were allowed to equilibrate at 3-Hz stimulation with a small preload applied [e.g., ~0.5 maximum length (L_max)]. After ~60 min, the muscles were gradually stretched to L_max, and experiments commenced once steady-state levels of force were attained. For force-rest interval studies, electrical stimulation was interrupted for 1-60 s and then resumed until steady state was reestablished. Statistical analyses were conducted using Student's t-test, and the Bonferroni correction factor was applied to reduce the risk of a type 1 error.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Figure 1 illustrates the salient features of Ca²⁺_i-dependent regulation for cloned canine NCX1.1 and 680-685 expressed in Xenopus laevis oocytes. Outward Na⁺/Ca²⁺ exchange currents are shown where cytoplasmic Na⁺ exchanges for extracellular (i.e., pipette) Ca²⁺. This outward (or reverse) mode of transport is typically used to investigate Ca²⁺_i regulation because the transported (i.e., extracellular) and regulatory (i.e., intracellular) pools of Ca²⁺ are on opposite membrane surfaces. The ability to distinguish between the effects of regulatory and transport Ca²⁺ is obscured for inward (i.e., forward mode) current measurements because both processes occur on the same membrane surface (i.e., intracellular). In the representative records shown, currents were activated by applying 100 mM Na⁺ to the cytoplasmic surface of the patch, which exchanges for 8 mM Ca²⁺ in the pipette. The single variable in these experiments was the concentration of regulatory Ca²⁺_i (0-10 µM) applied to the cytoplasmic surface. Note the progressive stimulation of exchange current for NCX1.1 in response to increasing levels of Ca²⁺_i (Fig. 1). Regulatory Ca²⁺ stimulates peak current and progressively increases steady-state currents by alleviating current inactivation. At higher regulatory Ca²⁺ concentrations (e.g., 10 µM), peak current declines (see below). These properties are largely eliminated for 680-685; that is, exchange current records appear maximally stimulated, similar to those observed for NCX1.1 in the presence of 10 µM regulatory Ca²⁺ or for NCX1.1 that has undergone deregulation after exposure to -chymotrypsin (19).

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Outward Na+/Ca2+ exchange currents mediated by NCX1.1 or 680-685 expressed in Xenopus laevis oocytes. Currents were activated by the rapid (i.e., ~200 ms) application of 100 mM Na+i to cytoplasmic surface of patch. Regulatory Ca2+i, at concentrations indicated by numbers (X µM Ca2+), was present for 32 s before and throughout each current activation event. Pipette (i.e., extracellular) Ca2+ (Ca2+o) was constant at 8 mM. Representative current traces are from single-oocyte giant membrane patches.

The pooled data shown in Fig. 2 illustrate the effects of different concentrations of regulatory Ca²⁺_i on peak and steady-state outward exchange currents. Data were obtained as described for Fig. 1, and currents were normalized to the respective values obtained at 3 µM Ca²⁺_i for either NCX1.1 or 680-685. For NCX1.1, peak current was markedly stimulated by 1 µM Ca²⁺_i but was reduced at the highest Ca²⁺_i concentration examined (i.e., 10 µM), an effect attributable to competition between Na⁺_i and Ca²⁺_i at the intracellular transport site (20, 35). Steady-state currents mediated by NCX1.1 were progressively increased by regulatory Ca²⁺_i. This stimulatory effect occurs because of the progressive saturation of the regulatory Ca²⁺ binding site and the alleviation of Na⁺_i-dependent inactivation, balanced by the inhibition of current caused by the competition between Na⁺_i and Ca²⁺_i at the intracellular transport site (21). In contrast, 680-685 was mainly insensitive to the presence or absence of regulatory Ca²⁺_i except at 10 µM, where, as with NCX1.1, competition between Na⁺_i and Ca²⁺_i led to a reduction in current levels. Because both Na⁺_i- and Ca²⁺_i-dependent regulatory properties are effectively eliminated for 680-685, as shown by near-maximal peak and steady-state currents at 0-10 µM Ca²⁺_i, we believe that the activity of this exchanger would be unaltered by the ionic fluxes it could likely encounter in an intact mammalian cardiomyocyte.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Regulatory Ca2+i dependence of peak and steady-state outward Na+/Ca2+ exchange currents mediated by NCX1.1 or 680-685 expressed in Xenopus laevis oocytes. Data are normalized to respective peak and steady-state current values obtained at 3 µM regulatory Ca2+i for either NCX1.1 or 680-685. Currents were obtained as described for Fig. 1. Data are means ± SE of 4-15 measurements at each Ca2+i concentration ([Ca2+]i) taken from 15 NCX1.1 patches and 3 measurements at each [Ca2+]i taken from 3 680-685 oocyte giant patches.

Because the transgenic mice employed in this study overexpressed canine Na⁺/Ca²⁺ exchangers, we needed to determine whether or not species differences in exchanger regulatory and/or transport properties could potentially confound the interpretation of results from subsequent experiments. Thus we examined the electrophysiological properties of the native transporters in isolated ventricular myocytes from mice and dogs. For these studies, outward Na⁺/Ca²⁺ exchange currents were elicited in giant patches excised from sarcolemmal membrane blebs that were induced by incubating the myocytes in hypotonic buffer for several hours before experimentation (see METHODS). Figure 3 illustrates the ionic regulatory profiles of exchange currents from these two preparations in representative patches, with the use of the same protocol described for Fig. 1. Note that the response to regulatory Ca²⁺_i is qualitatively similar between these two species as well as to that observed for NCX1.1 expressed in Xenopus oocytes (Fig. 1). This suggests that the regulatory properties of canine NCX1.1 expressed in transgenic mice will behave similarly to those of the native murine exchangers, varying primarily in the level of exchanger expression.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Outward Na+/Ca2+ exchange current recordings obtained from sarcolemmal membrane blebs from dog and mouse ventricular myocytes. Membrane blebs were induced by incubating myocytes in hypotonic buffer for several hours at 4°C, as described in METHODS. Currents were obtained as described for Fig. 1. Representative current traces are from single-myocyte giant membrane patches.

This notion is borne out by examination of the pooled data shown in Fig. 4, which illustrate the effects of regulatory Ca²⁺_i on peak and steady-state outward currents mediated by the native dog and mouse Na⁺/Ca²⁺ exchangers. As for oocyte membrane patches (Fig. 2), currents were normalized to the respective values obtained at 3 µM regulatory Ca²⁺_i for either canine or murine myocytes. With respect to the Ca²⁺_i dependence of steady-state currents, native dog and mouse exchangers behaved identically. Similarly, the relationship between peak current and regulatory [Ca²⁺]_i was superimposable for dogs and mice up to 3 µM, although a deviation at the highest [Ca²⁺]_i of 10 µM was observed. We have no explanation for this deviation. Overall, however, it appears reasonable that expression of the canine NCX1.1 transgene in mice will not impart a novel ionic regulatory phenotype to the Na⁺/Ca²⁺ exchange process.

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Regulatory Ca2+i dependence of peak and steady-state outward Na+/Ca2+ exchange currents in sarcolemmal membrane blebs from dog and mouse ventricular myocytes. Data are normalized to respective peak and steady-state current values obtained at 3 µM regulatory Ca2+i for either dog or mouse giant patches. Currents were obtained as described for Fig. 1. Data are means ± SE of 2-5 measurements at each [Ca2+]i taken from 4 canine myocyte patches and 3-5 measurements at each [Ca2+]i taken from 5 control mouse patches.

The transgenic mice overexpressing NCX1.1 have been described in detail elsewhere (1, 6, 15, 34, 38), and a mouse line overexpressing the 680-685 exchanger in cardiac tissue was developed using an identical approach. The level of exchanger overexpression in transgenic 680-685 mice was assessed using membrane vesicles isolated from cardiac tissue, as described previously (1). Na⁺ gradient-dependent ⁴⁵Ca²⁺ uptake was 0.17 ± 0.02 and 0.38 ± 0.04 nmol Ca²⁺ · mg protein¹ · 3 s¹ in vesicles from strain control and 680-685 transgenic mouse hearts, respectively. This 124% increase is comparable to the level of overexpression seen previously in the NCX1.1 transgenic mice (148% increase), as assessed by the same technique (1).

Figure 5 illustrates the effects of regulatory Ca²⁺_i on outward Na⁺/Ca²⁺ exchange currents in giant patches from sarcolemmal membrane blebs derived from transgenic mouse myocytes under conditions identical to those described for Fig. 3. From transgenic NCX1.1 myocyte patches, we observed the typical stimulation of peak and steady-state exchange currents as regulatory Ca²⁺_i was elevated. In contrast, the response to regulatory Ca²⁺_i was largely eliminated in myocyte patches from transgenic mice overexpressing 680-685. Thus overexpression of 680-685 appears to overwhelm the native ionic regulatory phenotype of the mouse and leads to a regulatory profile dominated by Ca²⁺_i insensitivity. The residual Na⁺_i- and Ca²⁺_i-dependent regulation can be attributed to the background profile of native mouse exchangers.

View larger version (19K): [in this window] [in a new window]   Fig. 5.   Outward Na+/Ca2+ exchange current recordings obtained from sarcolemmal membrane blebs from transgenic mouse ventricular myocytes overexpressing NCX1.1 or 680-685 Na+/Ca2+ exchangers. Currents were obtained as described for Fig. 1. Representative current traces are from single-myocyte giant membrane patches.

This is further substantiated by examination of the pooled data shown in Fig. 6, which illustrate the Ca²⁺_i dependence of peak and steady-state exchange currents in myocyte patches derived from transgenic NCX1.1 and 680-685 mice. Ca²⁺_i regulation for transgenic NCX1.1 myocytes was similar to that observed in native dog and mouse myocytes (Fig. 3) as well as that in Xenopus oocytes expressing NCX1.1 (Fig. 1); that is, both peak and steady-state currents are augmented by regulatory Ca²⁺_i to a maximum at ~3 µM. In contrast, outward currents generated in myocyte patches derived from transgenic 680-685 mice were mainly insensitive to the presence or absence of regulatory Ca²⁺_i, a pattern similar to that observed in Xenopus oocytes expressing 680-685 (Fig. 1). These results indicate that ionic regulation has been essentially eliminated in the transgenic 680-685 line, whereas overexpression of NCX1.1 is associated with ionic regulatory properties indistinguishable from those of the native dog and mouse exchangers.

View larger version (20K): [in this window] [in a new window]   Fig. 6.   Regulatory Ca2+i dependence of peak and steady-state outward Na+/Ca2+ exchange currents in sarcolemmal membrane blebs from transgenic mouse ventricular myocytes overexpressing NCX1.1 or 680-685 Na+/Ca2+ exchangers. Data are normalized to respective peak and steady-state current values obtained at 3 µM regulatory Ca2+i for either transgenic NCX1.1 or 680-685 mouse giant patches. Currents were obtained as described for Fig. 1. Data are means ± SE of 7-9 measurements at each [Ca2+]i taken from 9 NCX1.1 patches and 6-11 measurements at each [Ca2+]i taken from 11 680-685 giant patches.

This point is further illustrated by the representative outward current data shown in Fig. 7, in which the effects of application, removal, and reapplication of regulatory Ca²⁺_i on steady-state Na⁺/Ca²⁺ exchange currents are shown in giant patches derived from control, transgenic NCX1.1, and transgenic 680-685 myocytes. The protocol employed was the same as that used to generate the data shown in Figs. 1 and 3, with the exception that once steady-state current levels had been attained in the presence of 1 µM Ca²⁺_i, regulatory Ca²⁺_i was removed for ~32 s and then reapplied until steady-state levels of current were reacquired. Both the control and transgenic NCX1.1 lines exhibited a similar, and substantial, current decrease on removal of regulatory Ca²⁺_i. Note that under these conditions, exchange current is almost completely suppressed despite the huge gradient favoring Na⁺/Ca²⁺ exchange (i.e., 100 mM Na⁺_i vs. 0 mM Na⁺_o; 8 mM Ca²⁺_o vs. 0 µM Ca²⁺_i). This illustrates that Na⁺/Ca²⁺ exchange activity can be tightly regulated by this mechanism. In contrast, transgenic 680-685 exchangers were essentially unresponsive to this maneuver; that is, steady-state current levels were largely insensitive to the presence or absence of regulatory Ca²⁺_i. The slight reduction in steady-state current levels is most likely attributable to the native, regulated exchangers present in the myocyte membrane. Thus, if Ca²⁺_i-dependent regulation plays an important role in activating wild-type Na⁺/Ca²⁺ exchangers under physiological conditions, it is reasonable to expect that transgenic 680-685 exchangers should be constitutively active and their activity virtually independent of these regulatory mechanisms. On the other hand, Na⁺/Ca²⁺ exchange in control and transgenic NCX1.1 mouse lines should be normally regulated according to the cytoplasmic Ca²⁺ fluxes of a myocyte under steady-state stimulation.

View larger version (14K): [in this window] [in a new window]   Fig. 7.   Effect of removal and reapplication of regulatory Ca2+i on outward Na+/Ca2+ exchange currents in sarcolemmal membrane blebs from ventricular myocytes obtained from control and/or transgenic (NCX1.1 or 680-685) mice. Currents were obtained as described for Fig. 1. Patches were perfused with 1 µM regulatory Ca2+i for 32 s before, and for 32 s after, current activation with 100 mM Na+i. Subsequently, and in continuous presence of 100 mM Na+i, regulatory Ca2+i was removed for 32 s and then reapplied for a further 32 s.

Having established that myocytes from the two transgenic mouse lines show the anticipated electrophysiological characteristics, we investigated cardiac contractile properties in electrically stimulated, isolated papillary muscles. Figure 8 illustrates representative force tracings from papillary muscles derived from the two transgenic mouse lines. Muscles were electrically stimulated at 3 Hz, and a rest interval of 5 s was imposed near the middle of the traces. Postrest potentiation was observed in both preparations, followed by a gradual recovery to steady-state force levels. This relationship was characterized over a range of frequencies (2-6 Hz) and rest intervals (1-60 s) to gain insight into the Ca²⁺-handling behavior of the myocytes.

View larger version (56K): [in this window] [in a new window]   Fig. 8.   Postrest potentiation in papillary muscles from transgenic mice overexpressing NCX1.1 or 680-685 Na+/Ca2+ exchangers. Twitch contractions were evoked by electrical stimulation at 3 Hz. Representative data show postrest potentiation after a 3-s rest interval.

Figure 9 illustrates postrest force development in papillary muscles derived from transgenic NCX1.1 and 680-685 mice over a range of stimulation frequencies (3-6 Hz) after rest intervals of 3, 5, and 30 s. Data are expressed in terms of the potentiation fraction, defined as the ratio of force developed by the first postrest beat to that of steady-state beats preceding the rest interval. The duration of rest intervals imposed was randomly applied, and muscles were allowed to return to steady-state force levels between each rest period. At each rest interval and at all frequencies examined, postrest potentiation was greater for muscles obtained from transgenic 680-685 mice than for those obtained from NCX1.1 mice.

View larger version (18K): [in this window] [in a new window]   Fig. 9.   Potentiation fraction, as a function of stimulation frequency, in papillary muscles from transgenic mice overexpressing NCX1.1 or 680-685 Na+/Ca2+ exchangers. Potentiation fraction is defined as ratio of developed force for first postrest beat to force of preceding (i.e., prerest) steady-state beats. Data are shown for rest intervals of 3, 5, and 30 s. Data are means ± SE of 4-9 measurements or 4-8 measurements at each stimulation frequency from 8 papillary muscles each for NCX1.1 or 680-685 mice, respectively. * P < 0.05; ** P < 0.01 vs. NCX1.1.

Figure 10 illustrates postrest potentiation over a wider range of rest intervals (1-60 s) imposed on a steady-state train of stimuli at a frequency of 4 Hz. In general, postrest potentiation increased at shorter intervals (2-10 s) and then gradually declined for both transgenic mouse lines. Frequently, spontaneous beating occurred during the longer rest intervals (e.g., 15-60 s). Consequently, these data were not included in the data shown in Fig. 10, but they raise the possibility that the decline in postrest potentiation at lengthy rest intervals may be the result of asynchronous Ca²⁺ release events below our detection threshold. Nevertheless, postrest potentiation was greater at all rest intervals in muscles from the 680-685 mice than in those from the NCX1.1 transgenic mice.

View larger version (17K): [in this window] [in a new window]   Fig. 10.   Potentiation fraction, as a function of rest interval, in papillary muscles from transgenic mice overexpressing NCX1.1 or 680-685 Na+/Ca2+ exchangers. Muscles were stimulated at a frequency of 4 Hz, and rest intervals between 1 and 60 s were imposed in random order. Data are means ± SE of 3-7 measurements or 3-8 measurements at each rest interval from 8 papillary muscles each for transgenic NCX1.1 and 680-685 mice, respectively.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We examined the consequences of overexpressing the canine cardiac Na⁺/Ca²⁺ exchanger NCX1.1 and a deletion mutant of NCX1.1, 680-685, lacking ionic regulation, in transgenic mouse hearts. The first goal was to determine whether overexpression of the 680-685 mutant led to a phenotype in which ionic regulation of Na⁺/Ca²⁺ exchange was reduced or eliminated. Using electrophysiological techniques, we showed that a marked reduction in ionic regulation of Na⁺/Ca²⁺ exchange occurs with expression of 680-685 and that myocytes acquire a regulatory phenotype similar to that observed for the 680-685 mutant expressed in Xenopus oocytes. Using intact papillary muscles, we then showed that this alteration of ionic regulation of Na⁺/Ca²⁺ exchange leads to alterations of cardiac contractile properties. Specifically, we observed differences in postrest potentiation, a paradigm providing insight into the interplay between Ca²⁺ handling by the sarcolemma and the SR.

Na⁺/Ca²⁺ exchange: role and regulation. Na⁺/Ca²⁺ exchange is the major pathway for transsarcolemmal Ca²⁺ removal, a requisite for cardiac muscle relaxation (2, 9). Ca²⁺ efflux by this mechanism is similar in quantity to that which enters cardiac cells via L-type Ca²⁺ channels (30). On a beat-to-beat basis, the exact magnitude of Ca²⁺ fluxes mediated by Na⁺/Ca²⁺ exchange is not known because several confounding factors prevent accurate assessment. For example, in several species and/or under certain experimental conditions, Na⁺/Ca²⁺ exchange may also serve as a Ca²⁺ entry mechanism (24, 25, 37). If this occurs, then an even greater amount of Ca²⁺ efflux would be required of the Na⁺/Ca²⁺ exchange process to maintain Ca²⁺ homeostasis. On the other hand, although the role of sarcolemmal Ca²⁺-ATPases in diastolic Ca²⁺ removal is generally thought to be of less importance than Na⁺/Ca²⁺ exchange, any contribution by this pathway would reduce the Ca²⁺ load presented to the exchanger (3-5). The lack of specific inhibitors for either of these Ca²⁺ efflux mechanisms hampers our ability to determine their exact contributions. Furthermore, the substantial species differences, in terms of the relative importance of these Ca²⁺ efflux pathways, render the generalization of results from most studies problematic at best. In any event, Ca²⁺ efflux must equal Ca²⁺ influx during regular stimulation to avert Ca²⁺ overload or depletion, and Na⁺/Ca²⁺ exchange plays a major role in this process.

Consequences of NCX1.1 and 680-685 overexpression in transgenic mice. Several recent studies have examined the functional consequences of overexpressing canine NCX1.1 in transgenic mice. Notable differences between the transgenic line and control mice include 1) enhanced Na⁺/Ca²⁺ exchange currents and relaxation rates of Ca²⁺ transients and contractions (1, 34, 38), 2) an increased inotropic responsiveness to the Na⁺-channel agonist BDF-9148 (6), 3) gender-specific increased susceptibility to ischemia-reperfusion injury (15), and 4) the finding that compensatory alterations do not occur in the Ca²⁺-handling proteins of the SR (34). Overall, cardiac function appears to be relatively normal, although experimental manipulations can detect alterations associated with overexpression of NCX1.1.