Transport and Regulation of the Cardiac Na⁺-Ca²⁺ Exchanger, NCX1

Abstract

Cardiac muscle fails to relax upon replacement of extracellular Ca²⁺ with Ba²⁺. Among the manifold consequences of this intervention, one major possibility is that Na⁺-Ba²⁺ exchange is inadequate to support normal relaxation. This could occur due to reduced transport rates of Na⁺-Ba²⁺ exchange and/or by failure of Ba²⁺ to activate the exchanger molecule at the high affinity regulatory Ca²⁺ binding site. In this study, we examined transport and regulatory properties for Na⁺-Ca²⁺ and Na⁺-Ba²⁺ exchange. Inward and outward Na⁺-Ca²⁺ or Na⁺-Ba²⁺ exchange currents were examined at 30°C in giant membrane patches excised from Xenopus oocytes expressing the cloned cardiac Na⁺-Ca²⁺ exchanger, NCX1. When excised patches were exposed to either cytoplasmic Ca²⁺ or Ba²⁺, robust inward Na⁺-Ca²⁺ exchange currents were observed, whereas Na⁺-Ba²⁺ currents were absent or barely detectable. Similarly, outward currents were greatly reduced when pipette solutions contained Ba²⁺ rather than Ca²⁺. However, when solution temperature was elevated from 30°C to 37°C, a substantial increase in outward Na⁺-Ba²⁺ exchange currents was observed, but not so for inward currents. We also compared the relative abilities of Ca²⁺ and Ba²⁺ to activate outward Na⁺-Ca²⁺ exchange currents at the high affinity regulatory Ca²⁺ binding site. While Ba²⁺ was capable of activating the exchanger, it did so with a much lower affinity (K _D ∼ 10 μM) compared with Ca²⁺ (K _D ∼ 0.3 μM). Moreover, the efficiency of Ba²⁺ regulation of Na⁺-Ca²⁺ exchange is also diminished relative to Ca²⁺, supporting ∼60% of maximal currents obtainable with Ca²⁺. Ba²⁺ is also much less effective at alleviating Na⁺ _i-induced inactivation of NCX1. These results indicate that the reduced ability of NCX1 to adequately exchange Na⁺ and Ba²⁺ contributes to failure of the relaxation process in cardiac muscle.

introduction

Na⁺-Ca²⁺ exchange plays a major role in Ca²⁺ homeostasis in cardiac muscle. Removal of myoplasmic Ca²⁺ by this mechanism is essential for physiological cardiac relaxation (Bers, 1991). In general, removal of Ca²⁺ by Na⁺-Ca²⁺ exchange is equivalent to Ca²⁺ entry through L-type Ca²⁺ channels on a beat-to-beat basis (Bridge et al., 1990). Na⁺-Ca²⁺ exchange may also serve an important role as a Ca²⁺ entry mechanism during cardiac excitation. Several studies have demonstrated that this “reverse mode” of Na⁺-Ca²⁺ exchange can trigger sarcoplasmic reticulum Ca²⁺ release (LeBlanc and Hume, 1990; Kohmoto et al., 1994; Levi et al., 1994; Vornanen et al., 1994; Wasserstrom and Vites, 1996). Consequently, alterations in Na⁺-Ca²⁺ exchange function, exemplified by digitalis treatment (Lee and Dagastino, 1982) or alterations in the intracellular Na⁺ concentration (Harrison and Boyett, 1995), produce major effects on cardiac contractility.

Recently, several regulatory properties have been characterized for the cardiac Na⁺-Ca²⁺ exchanger, NCX1. Detailed characterization of these regulatory mechanisms has been accomplished for the native and cloned cardiac Na⁺-Ca²⁺ exchanger using the giant excised patch technique. Regarding ionic regulation, both Na⁺ and Ca²⁺ regulate exchange activity in addition to serving as the transport substrates (Hilgemann, 1990). Examination of outward (reverse) Na⁺-Ca²⁺ exchange currents reveals a complex waveform. The application of Na⁺ _i induces an outward current which undergoes a time-dependent inactivation. The extent of this inactivation is governed by both cytoplasmic Na⁺ and Ca²⁺ levels. However, in the presence of a constant level of cytoplasmic Ca²⁺ (e.g., 1 μM), both outward currents and the extent of inactivation increase as Na⁺ _i levels are increased. This behavior is referred to as Na⁺ _i-induced or I₁ inactivation (Hilgemann et al., 1992b ).

Cytoplasmic Ca²⁺ levels regulate exchange activity by influencing the extent of Na⁺ _i-induced inactivation and through an apparent direct activation of the exchange molecule (Hilgemann et al., 1992a , b ). This direct pathway is referred to as I₂ inactivation where removal of cytosolic Ca²⁺ favors entry into an inactive (I₂) state. Both forward and reverse modes of Na⁺-Ca²⁺ exchange are regulated by cytoplasmic Ca²⁺ (Matsuoka et al., 1995), and the high affinity regulatory Ca²⁺ binding site has been identified for the cardiac exchanger, NCX1 (Levitsky et al., 1994; Matsuoka et al., 1995). The ability of other divalent cations to substitute for Ca²⁺ at the regulatory site has not been examined in detail.

Na⁺-Ca²⁺ exchange has a strict specificity for Na⁺ as the transported monovalent cation (Philipson and Nicoll, 1993). Less stringency is observed for transport of other divalent cations with Ba²⁺ and Sr²⁺ being transported to varying degrees. Earlier studies of this nature have used cardiac sarcolemmal vesicles to examine radioisotope fluxes for different divalent cations. From these reports, it appeared that Ca²⁺ and Sr²⁺ were transported at nearly equal rates, whereas Ba²⁺ transport was ∼20 times slower and exhibited a two- to threefold reduction in affinity for the exchanger (Trosper and Philipson, 1983; Tibbits and Philipson, 1985). More recent electrophysiological and fluorescence measurements yield conflicting results. In one instance, outward Na⁺-Ba²⁺ exchange currents were not detectable from whole cell patch clamp experiments using guinea-pig ventricular myocytes (Kimura et al., 1987). However, Ba²⁺ uptake through the exchanger was readily detected in CHO cells expressing the bovine cardiac Na⁺-Ca²⁺ exchanger (Chernaya et al., 1996). An experimental limitation of all the above studies is the difficulty in discriminating between transport and regulatory consequences for this ionic substitution. However, this limitation can be circumvented by using the giant excised patch clamp. In this study, we have compared the effects of Ca²⁺ and Ba²⁺ on properties of the cloned, cardiac Na⁺-Ca²⁺ exchanger, NCX1. The specific goals were to examine how Ba²⁺ substitution alters the transport and regulatory properties of NCX1 and to determine if these differences can provide a reasonable account for why cardiac muscle fails to relax in Ba²⁺-containing media.

methods

Myocyte Preparation and Shortening Measurements

Canine ventricular myocytes were provided by Dr. A . Lukas (University of Manitoba, Winnipeg, Canada) and were prepared as described previously (Lukas and Antzelowitch, 1993). Myocytes were superfused with (in mM): 136 NaCl, 10 HEPES, 8.33 NaH₂PO₄, 5.4 KCl, 1 MgCl₂, 1 CaCl₂ or BaCl₂, pH 7.4 (using NaOH) at 30°C. Field stimulation of myocytes at 0.5 Hz (1.1–1.5 times threshold) was performed using a Grass SD-9 Stimulator (Grass Instrument Co., Warwick, RI). Shortening was monitored with a video edge detection system (Crescent Electronics, Sandy, UT) and recorded using Axon Instruments (Foster City, CA) hardware and software as described previously (Hryshko et al., 1989).

Molecular Biological Techniques

Oocytes were obtained from Xenopus laevis as described previously (Hryshko et al., 1996). Oocytes were treated with collagenase (20 mg/ml) for 1 h, washed in Barth's solution, treated with 100 mM K₂HPO₄ for 15 min, washed, and stored overnight in fresh Barth's solution. NCX1 cRNA was prepared using T3 mMessage mMachine (Ambion Inc., Austin, TX) according to the manufacturer's instructions. Oocytes were injected with cRNA (∼5 ng/oocyte), and activity was measured 3–6 d later.

Electrophysiological Techniques

Na⁺-Ca²⁺ exchange activity was measured using the giant excised patch clamp technique of Hilgemann (1989) as described previously (Matsuoka et al., 1995; Hryshko et al., 1996). Pipettes were pulled from borosilicate glass and polished to a final diameter of 20–40 μm. Pipettes were coated with a parafilm/mineral oil mixture to enhance patch stability and reduce electrical noise. For seal formation, oocytes were placed in a solution containing (in mM): 100 KOH, 100 MES, 20 HEPES, 5 EGTA, 5 MgCl₂, pH 7.0 with MES. Gigaohm seals were formed by gentle suction and patches were excised by progressive movements of the pipette tip. Excised patches were in the inside-out configuration. For outward Na⁺-Ca²⁺ exchange current measurements, pipettes were filled with (in mM): 100 NMG-MES, 30 HEPES, 30 TEA-OH, 16 sulfamic acid, 8 CaCO₃, 6 KOH, 0.25 ouabain, 0.1 niflumic acid, 0.1 flufenamic acid, pH 7.0 (using MES). Outward Na⁺-Ca²⁺ exchange currents were elicited by switching from a Li⁺ to Na⁺-based superfusate containing (in mM): 100 Na- or Li-aspartate, 20 MOPS, 20 TEA-OH, 20 CsOH, 10 EGTA, 0–2.30 CaCO₃ or 0–7.55 Ba(OH)₂, 1–1.5 Mg(OH)₂, pH 7.0 (using MES or LiOH). The amounts of Ca²⁺ and Mg²⁺ were adjusted to yield free Mg²⁺ concentrations of 1 mM and various free Ca²⁺ concentrations as indicated. MAXC software was used to calculate free Ca²⁺ and Mg²⁺ concentrations (Bers et al., 1994). For inward current measurements, pipettes contained (in mM): 100 Na-MES, 20 TEA-MES, 20 Cs-MES, 10 HEPES, 10 EGTA, 4 Mg(OH)₂, 0.2 ouabain, 0.1 niflumic acid, 0.1 flufenamic acid, 0.002 verapamil, pH 7.0. Inward Na⁺-Ca²⁺ exchange currents were activated by switching to the Li⁺-based Ca²⁺ containing superfusates described above for outward current measurements. Current data were acquired and analyzed using Axon Instruments (Foster City, CA) hardware and software. Solution changes (∼200 ms) were accomplished using a custom-built 20-channel computer-controlled solution switcher. All experiments were conducted at 30 ± 1°C unless indicated otherwise. The rationale and design of the different types of experiments are summarized in Fig. 1.

Figure 1.

The underlying rationale and different experimental conditions used in the present study are shown. For outward current measurements, extracellular solutions (pipette) contained 8 mM Ca²⁺ or Ba²⁺, and currents were activated by switching from Li⁺-based to Na⁺-based intracellular (bath) solutions. For inward current measurements, extracellular (pipette) solutions contained 100 mM Na⁺ and currents were activated by switching to intracellular solutions with various concentrations of either Ca²⁺ or Ba²⁺.

results

Fig. 2 illustrates the effects of replacing 1 mM extracellular Ca²⁺ with 1 mM Ba²⁺ on electrically evoked shortening in a canine ventricular myocyte. Almost immediately, contraction fails and a contracture develops. Upon restoration of Ca²⁺ to the superfusate, the contracture gradually wanes and near normal resting length and contractions resume. We observed this pattern of contracture in ten cells from three different myocyte preparations. Initial shortening averaged 10.9 ± 0.8% (mean ± SD, n = 10) in the Ca²⁺-containing superfusate. After recovery from the Ba²⁺-induced contracture, shortening averaged 8.1 ± 1.0%. While numerous cellular processes are affected by this intervention, our specific goal was to determine how Ba²⁺ substitution affects the operation of the Na⁺-Ca²⁺ exchanger. In particular, we sought to distinguish between the effects of Ba²⁺ substitution on transport vs. regulatory aspects of Na⁺-Ca²⁺ exchange function.

Figure 2.

The effects of replacing 1 mM extracellular Ca²⁺ with 1 mM Ba²⁺ on electrically stimulated shortening for a canine ventricular myocyte are shown. Upon substituting Ba²⁺ for Ca²⁺, shortening rapidly fails, and a sustained contracture develops. Restoration of extracellular Ca²⁺ leads to near full recovery of resting length and shortening.

Fig. 3 A shows inward Na⁺-Ca²⁺ and Na⁺-Ba²⁺ exchange currents from a single excised patch. Physiologically, this “forward” mode of transport corresponds to Ca²⁺ extrusion. Currents were activated by the application of Li⁺-based Ca²⁺- or Ba²⁺-containing solutions to the cytoplasmic surface of the patch at the concentrations indicated. The pipette solution contained 100 mM Na⁺, and Ca²⁺- and Ba²⁺-containing solutions were applied in a random order. For Ca²⁺-activation, note the progressive increase of inward currents observed up to concentrations between 30–300 μM. In contrast, at Ba²⁺ concentrations up to 300 μM, inward currents were barely detectable. In eight patches showing robust Ca²⁺-activated inward currents at 30°C, Ba²⁺-activated currents were absent or barely detectable. Three patches were also examined at 37°C and did not exhibit substantial Ba²⁺-activated inward exchange currents, whereas Ca²⁺-activated currents were readily observed (not shown). We did not routinely examine higher Ba²⁺ concentrations as these led to a small outward current at concentrations of 1 mM and above. This small outward current was also observed using pipette solutions containing Li⁺ rather than Na⁺ and may represent residual Ca²⁺ activated Cl⁻ conductance (see discussion).

Figure 3.

(A) Illustrates typical inward currents activated by the application of cytoplasmic Ca²⁺ or Ba²⁺. The pipette solution contained 100 mM Na⁺. Pooled results from three to nine patches (normalized to the current value obtained at 3 μM Ca²⁺ _i in nine patches) (means ± SE) are shown in B.

Fig. 3 B shows pooled data from nine patches for inward current measurements activated by Ca²⁺ or Ba²⁺. Each point represents the average from between three and nine determinations with data normalized to the inward current obtained at 3 μM Ca²⁺. K _D values are not reported as not all divalent concentrations could be examined in a single patch. For comparative purposes, currents obtained at 3 μM Ca²⁺ (n = 9) exceeded currents obtained with 100 μM Ba²⁺ (n = 4) by 17-fold and 300 μM Ba²⁺ (n = 6) by 10-fold. As shown below (see Fig. 5), the absence of inward current is not a consequence of failure to activate the exchanger with Ba²⁺.

Figure 5.

The effects of different concentrations of regulatory Ca²⁺ or Ba²⁺ on outward Na⁺-Ca²⁺ exchange currents are shown for a single patch in A and B. The pipette solution contained 8 mM Ca²⁺. The different concentrations of regulatory Ca²⁺ (A) or Ba²⁺ (B) were present before and during the application of 100 mM Na⁺ to activate the current. Typical concentration dependencies of peak and steady-state (SS) outward currents are shown for regulation by Ca²⁺ (C) and Ba²⁺ (D) from two separate patches. Note the difference in concentration range between these graphs.

Fig. 4 illustrates outward currents activated by the application of 100 mM Na⁺ to the cytoplasmic surface of the patch. Physiologically, this “reverse” mode of transport corresponds to Ca²⁺ entry. Different concentrations of regulatory Ca²⁺ were present before and during Na⁺ application, as indicated. Pipettes contained either 8 mM Ba²⁺ (Fig. 4, traces A and B) or Ca²⁺ (traces C and D). Several features are noteworthy. For Ba²⁺-containing pipettes, we observed very small currents at 30°C (<20 pA in six separate patches). Although regulated by Ca²⁺, the extent of I₂ regulation appeared blunted and Na⁺ _i-induced (I₁) inactivation was not observed. At 37°C, six of six patches exhibited Na⁺-Ba²⁺ exchange activity with regulatory properties more typical of those observed with Ca²⁺-containing pipette solutions. In three patches for which both temperatures could be examined, a three to fourfold increase in current magnitude was observed for this 7°C increase in temperature, as shown in Fig. 4, A and B. Recall that this striking increase in outward Na⁺-Ba²⁺ exchange currents was not evident for inward currents at 37°C. Comparatively, for Ca²⁺-containing pipettes (Fig. 4, C and D), normal I₁ and I₂ regulation were observed at both temperatures. Typical for three patches where currents were obtained at both temperatures, Fig. 4, C and D, shows less augmentation of current (30–50% increase) at 37°C than that observed for Na⁺-Ba²⁺ exchange. For both Ba²⁺- and Ca²⁺- containing pipettes, it is likely that we have underestimated the true temperature sensitivity as current rundown likely occurs during the 5–10 min period required to increase perfusate temperature.

Figure 4.

Typical outward Na⁺-Ba²⁺ (A and B) and Na⁺-Ca²⁺ (C and D) exchange currents examined at 30°C (left) and 37°C (right). The pipette solution contained 8 mM Ba²⁺ or Ca²⁺ and currents were activated by the application of 100 mM Na⁺ at the indicated concentrations of regulatory Ca²⁺ _i. Data were obtained by making measurements at 30°C followed by increasing bath temperature and repeating measurements at 37°C (5–7 min later).

The activity of Na⁺-Ca²⁺ exchangers is regulated by the occupancy status of a high affinity Ca²⁺ binding site on the cytoplasmic surface of the molecule (Levitsky et al., 1994; Matsuoka et al., 1995; Hryshko et al., 1996). Consequently, alterations in the activation of Na⁺-Ca²⁺ exchange activity by Ba²⁺ could also contribute to the reduced ability of NCX1 to transport Ba²⁺. To investigate this possibility, we compared regulation of outward Na⁺-Ca²⁺ exchange currents by Ca²⁺ and Ba²⁺. Fig. 5, A and B, shows current traces obtained from a single patch at different concentrations of regulatory Ca²⁺ or Ba²⁺, respectively. At each concentration indicated, regulatory Ca²⁺ or Ba²⁺ was present before and during the application of 100 mM Na⁺ to activate the outward current. Pipette Ca²⁺ was constant at 8 mM. Note that Ba²⁺ was considerably less effective at activating Na⁺-Ca²⁺ exchange compared to Ca²⁺. For example, virtually no current is observed at 3 μM Ba²⁺ whereas this level of Ca²⁺ leads to near maximal currents for Ca²⁺. In addition, the progressive loss of I₁ inactivation with increasing Ca²⁺ is much less apparent for Ba²⁺-regulated currents.

Fig. 5, C and D, shows typical concentration dependencies for Ca²⁺ and Ba²⁺ regulation, respectively, for outward exchange currents in two separate patches. The concentration ranges examined were selected to determine K _D and illustrate differences between Ca²⁺ and Ba²⁺ on steady-state current properties. Current levels were similar in these two different patches. The K _D for Ca²⁺ regulated peak I_NaCa was 0.35 μM. The complex relationship of regulatory Ca²⁺ with I₁ and I₂ inactivation leads to a near linear relation for steady state currents up to 10 μM Ca²⁺, followed by a progressive decline as Ca²⁺ begins to compete with cytoplasmic Na⁺ (as seen in Fig. 5 A). Ba²⁺ regulated peak I_NaCa and steady-state I_NaCa exhibited K _D's of 9.3 and 10.3 μM, respectively, and I₁ inactivation was not dramatically changed (as seen in Fig. 5 B). We observed similar differences in regulation of outward currents between Ca²⁺ vs. Ba²⁺ in 12 different patches.

Pooled data from 9 patches is shown in Fig. 6 for peak outward currents regulated by Ba²⁺ or Ca²⁺. Current values were normalized to that obtained for peak outward current in the presence of 3 μM regulatory Ca²⁺, allowing direct comparison of all data. Two features are obvious from this graph. First, the affinity of regulatory Ba²⁺ (K _D = 8.7 μM) is ∼30 times lower than that for Ca²⁺ (K _D = 260 nM). Second, the efficiency of Ba²⁺ regulation is substantially lower than that observed with Ca²⁺. Even at the highest Ba²⁺ concentrations examined, peak current was substantially lower than that observed for Ca²⁺.

Figure 6.

Regulation of outward Na⁺-Ca²⁺ exchange currents by Ba²⁺ and Ca²⁺ (inset). Pooled results (mean ± SD) from three to nine determinations in nine separate patches. Currents were normalized to the value of current obtained at 3 μM regulatory Ca²⁺ _i (in all 9 patches).

Ba²⁺ appeared to be much less effective at alleviating Na⁺-induced inactivation compared with Ca²⁺. This behavior is illustrated in Fig. 7 for pooled results from nine patches (3–9 determinations at each concentration). As regulatory Ca²⁺ was progressively increased, steady-state current approached the same level as peak current. In contrast, this behavior was not observed when Ba²⁺ served as the regulatory ion. In two patches examined at 1 mM regulatory Ba²⁺ (not shown), a small reduction in peak to steady state ratio was observed.

Figure 7.

The relationship between peak and steady-state outward Na⁺-Ca²⁺ exchange currents regulated by Ca²⁺ _i (filled squares) and Ba²⁺ _i (open squares) is shown for pooled results from three to nine patches (mean ± SD). For Ca²⁺ _i regulation, steady-state current approaches peak current levels reflecting the progressive reduction in I₁ inactivation. In contrast, I₁ inactivation is not attenuated by regulatory Ba²⁺ _i.

Finally, we compared the ability of Ca²⁺ _i and Ba²⁺ _i to compete for Na⁺ _i at the intracellular transport site. To accomplish this, outward Na⁺-Ca²⁺ exchange currents were examined in excised patches after treatment with 1–2 mg/ml α-chymotrypsin for 1–2 min. After this treatment, both I₁ inactivation and divalent regulatory effects (I₂ inactivation) were eliminated. Thus, competition between Na⁺ _i and divalent cations could be observed by examining outward current with different concentrations of divalents present. Divalent vs. Na⁺ _i competition appears as a reduction of outward current due primarily to: (a) increasing electroneutral Ca²⁺-Ca²⁺ or Ca²⁺-Ba²⁺ exchange, (b) simple competition between Na⁺ and divalent cation occupancy of the transport site, and/or (c) the progressive reduction in driving force for the exchange reaction. Typical and pooled results (mean ± SD, n = 4) are shown in Fig. 8. Outward currents were activated by 100 mM Na⁺, and pipette Ca²⁺ was constant at 8 mM. For Ca²⁺ competition, substantial inhibition of outward exchange current is evident at concentrations greater than 3 μM Ca²⁺. In contrast, the inhibitory effects of Ba²⁺ are greatly reduced and only become evident between 100 and 300 μM Ba²⁺.

Figure 8.

Typical deregulated outward Na⁺-Ca²⁺ exchange currents from a single patch. The pipette contained 8 mM Ca²⁺ and currents were activated by the application of 100 mM Na⁺. Both I₁ and I₂ regulation are absent after deregulation and the inhibitory effects of different concentrations of Ca²⁺ and Ba²⁺ are evident. Pooled results from four patches (mean ± SD) are shown in B. Currents were normalized to the value of current obtained in the absence of regulatory divalent cations.

discussion

We examined the ability of Ba²⁺ to substitute for Ca²⁺ on several aspects of Na⁺-Ca²⁺ exchange function. We show that Na⁺-Ba²⁺ exchange is substantially reduced in the forward direction (i.e., Na⁺ _o-Ba²⁺ _i exchange), appears to be reduced in the reverse direction (Na⁺ _i-Ba²⁺ _o exchange) and is considerably less effective as an activator of the exchanger at the high affinity regulatory Ca²⁺ binding site. These alterations in Na⁺-Ca²⁺ exchange function are likely to contribute to the failure of cardiac relaxation during superfusion with Ba²⁺-containing media.

For inward current measurements, substantial Na⁺-Ca²⁺ exchange currents were observed at Ca²⁺ concentrations of 1 μM and above. In contrast, inward currents due to Na⁺-Ba²⁺ exchange were barely detectable even at 100 and 300 μM Ba²⁺ (Fig. 3). As both Na⁺-Ca²⁺ and Na⁺-Ba²⁺ currents were measured in the same patches, our results indicate that Na⁺-Ba²⁺ exchange is orders of magnitude less effective than Na⁺-Ca²⁺ exchange. Our data indicate that both the affinity for transport and the efficiency of transport are greatly reduced for Na⁺-Ba²⁺ exchange. For example, less inward current was observed at 300 μM Ba²⁺ than was produced by 3 μM Ca²⁺. Comparatively, with equal concentrations of Ca²⁺ and Ba²⁺ (e.g., 100 μM), inward Na⁺-Ca²⁺ exchange current exceeded Na⁺-Ba²⁺ exchange by nearly 300-fold.

A K _D for Ba²⁺-activation of inward Na⁺-Ba²⁺ exchange was not determined as higher concentrations of Ba²⁺ (e.g., 1 mM) invariably led to the appearance of a small outward current. Although the induction of this unidentified outward current may partially mask inward currents at lower Ba²⁺ concentrations (e.g., 100– 300 μM), it seems unlikely that we have grossly under-estimated Na⁺-Ba²⁺ exchange. One possibility is that the outward current represents an endogenous Ca²⁺-activated Cl⁻ conductance in oocyte membranes. Despite using Cl⁻-free pipette and perfusing solution, residual Cl⁻ from the sealing solution invariably contaminates the pipette. However, if present, inward Na⁺-Ca²⁺ exchange currents would also be underestimated, presumably by a similar or greater amount. In addition, we observed this pattern of large Na⁺-Ca²⁺ vs. small Na⁺-Ba²⁺ exchange currents during several long recordings (>10 min) in single patches. Over this time course, nearly complete run-down of the Ca²⁺-activated Cl⁻ conductance occurs due to diffusion of Cl⁻ from the pipette tip and genuine current rundown.

Both forward and reverse transport modes of Na⁺-Ca²⁺ exchange are regulated by cytoplasmic Ca²⁺ (Matsuoka et al., 1995). Therefore, the possibility exists that reduced inward Na⁺-Ba²⁺ exchange is a consequence of reduced exchanger activation by Ba²⁺. However, the large differences we observed between inward Na⁺-Ca²⁺ and Na⁺-Ba²⁺ exchange currents cannot be attributed to the failure of Ba²⁺ to activate the exchanger. As shown in Figs. 5 and 6, 300 μM Ba is sufficient to activate ∼60% of the current obtainable with Ca²⁺ activation. Therefore, even though Ba²⁺ activation is less effective than Ca²⁺, substantial activation of inward Na⁺-Ba²⁺ exchange currents would be expected at the concentrations examined. Thus, alterations of V_max and/or the apparent K _D for transport appear to be causal for reduced inward Na⁺-Ba²⁺ exchange. The results obtained from patches after deregulation by α-chymotrypsin (Fig. 8) showed relatively weak competition between Ba²⁺ and Na⁺, supporting the idea of lower affinity at the intracellular transport site.

It is of interest to compare our results with those obtained from other experimental systems. For example, in cardiac sarcolemmal vesicles, V_max is reduced by 21-fold and the K _m is 2.4-fold larger when Ba²⁺ is substituted for Ca²⁺ (Tibbits and Philipson, 1983). As vesicular uptake is now considered to be mediated almost exclusively by inside-out oriented vesicles (Li et al., 1991), this result is analogous to inward current measurements from giant excised patches. Our results show a similar or greater reduction in transport capacity and a much larger shift in affinity for Na⁺-Ba²⁺ exchange. As an example, a 100 pA inward current for Na⁺-Ca²⁺ exchange would result in a comparatively small Na⁺-Ba²⁺ exchange current (i.e., 5 pA). However, at Ba²⁺ concentrations 30 μM and below, inward currents were never observed. This may be due to our inability to reliably measure currents less than a few picoamps. In CHO cells expressing the bovine cardiac Na⁺-Ca²⁺ exchanger, extracellular Na⁺-dependent ¹³³Ba²⁺ efflux could be measured, consistent with forward Na⁺-Ba²⁺ exchange. However, a reduction in Ba²⁺ concentration due to forward Na⁺-Ba²⁺ exchange was not observed in these same cells based on fura-2 measurements (Condrescu et al., 1997). The reasons for these discrepancies remain unknown but may reflect loss of resolution by the various techniques.

Outward Na⁺-Ba²⁺ exchange currents appear to be considerably smaller than those observed using Ca²⁺ as the transported cation. However, since we cannot measure outward Na⁺-Ca²⁺ and Na⁺-Ba²⁺ exchange currents in the same patch, this comparison is strictly qualitative. Notwithstanding this limitation, comparison with other published data shows that Ba²⁺ substitution for Ca²⁺ can eliminate or greatly diminish currents associated with the exchanger based on whole cell measurements in guinea pig (Kimura et al., 1987) and rabbit ventricular cells (Shimoni and Giles, 1987). In contrast, Na⁺-Ba²⁺ exchange activity could be readily measured in CHO cells expressing the cardiac exchanger using either ¹³³Ba²⁺ or fura-2 measurements under conditions analogous to our outward exchange measurements (Chernaya et al., 1996; Condrescu et al., 1997). Exchange activity was also regulated by cytosolic Ca²⁺ in this preparation (Chernaya et al., 1996; Condrescu et al., 1997).

One surprising result from the present study is the marked effects of temperature on the appearance of outward Na⁺-Ba²⁺ exchange currents (Fig. 4). At 30°C, the small outward currents are regulated by Ca²⁺ _i (I₂) but do not exhibit Na⁺ _i-dependent (I₁) inactivation. Substantially greater currents are observed at 37°C, with near normal I₁ and I₂ regulation. This suggests a high energy barrier for Ba²⁺ translocation. This condition or a greatly reduced affinity for extracellular Ba²⁺ would reduce the fraction of exchangers with intracellularly oriented ion binding sites and consequently would alleviate I₁ inactivation. Such behavior is analogous to lowering extracellular Ca²⁺ which reduces I₁ inactivation (Hilgemann et al., 1992a ). This strong temperature dependence must also be considered when comparing results from other studies examining Na⁺-Ba²⁺ exchange.

Finally, we examined the ability of Ba²⁺ to substitute for Ca²⁺ as an activator of the exchanger at the high affinity regulatory Ca²⁺ binding site. Three major differences are observed for Ba²⁺. First, the affinity for Ba²⁺ regulation is reduced nearly 30-fold. Second, the efficiency of Ba²⁺ activation is considerably less than that observed with Ca²⁺ over the concentration range examined. At 300 μM regulatory Ba²⁺, maximal outward Na⁺-Ca²⁺ exchange currents are only ∼60% of those observed for regulatory Ca²⁺. Third, unlike Ca²⁺, Ba²⁺ does not appear to strongly influence I₁ inactivation. Raising cytoplasmic Ca²⁺ progressively alleviates I₁ inactivation (Hilgemann et al., 1992a ) whereas differences in the profile of Ba²⁺ regulated currents are unremarkable. This feature may prove useful in future studies of the mechanism(s) of I₁ and I₂ regulation.

In conclusion, we have characterized the effects of substituting Ba²⁺ for Ca²⁺ on inward and outward exchange currents and on regulatory properties of the cloned canine cardiac Na⁺-Ca²⁺ exchanger, NCX1. By all accounts, the transport and regulatory consequences of this substitution should severely impair the ability of the exchange mechanism to maintain ionic homeostasis. Other contributory factors include the reduction or absence of SR Ba²⁺ uptake (Palade, 1987; Chernaya et al., 1996) and prolonged depolarization due to effects on L-type Ca²⁺ channels (Lee et al., 1985) and K⁺ channels (Imoto et al., 1987). However, while Ba²⁺ substitution alters numerous other homeostatic processes, the demonstrated changes in Na⁺-Ca²⁺ exchange function alone would appear sufficient to prevent cardiac relaxation.