Cardiac Sodium-Calcium Exchanger Is Regulated by Allosteric Calcium and Exchanger Inhibitory Peptide at Distinct Sites
http://www.100md.com
《循环研究杂志》
From the Johns Hopkins University, Institute of Molecular Cardiobiology, Division of Cardiology, Baltimore, Md.
The sarcolemmal Na+-Ca2+ exchanger (NCX) is the main Ca2+ extrusion mechanism in cardiac myocytes and is thus essential for the regulation of Ca2+ homeostasis and contractile function. A cytosolic region (f-loop) of the protein mediates regulation of NCX function by intracellular factors including inhibition by exchanger inhibitory peptide (XIP), a 20 amino acid peptide matching the sequence of an autoinhibitory region involved in allosteric regulation of NCX by intracellular Na+, Ca2+, and phosphatidylinositol-4,5-biphosphate (PIP2). Previous evidence indicates that the XIP interaction domain can be eliminated by large deletions of the f-loop that also remove activation of NCX by intracellular Ca2+. By whole-cell voltage clamping experiments, we demonstrate that deletion of residues 562–679, but not 440– 456, 498–510, or 680–685 of the f-loop selectively eliminates XIP-mediated inhibition of NCX expressed either heterologously (HEK293 and A549 cells) or in guinea pig cardiac myocytes. In contrast, by plotting INCX against reverse-mode NCX-mediated Ca2+ transients in myocytes, we demonstrate that Ca2+-dependent regulation of NCX is preserved in 562–679, but significantly reduced in the other three deletion mutants. The findings indicate that f-loop residues 562–679 may contain the regulatory site for endogenous XIP, but this site is distinct from the Ca2+-regulatory domains of the NCX. Because regulation of the NCX by Na+ and PIP2 involves the endogenous XIP region, the 562–679 mutant NCX may be a useful tool to investigate this regulation in the context of the whole cardiac myocyte.
Key Words: cardiac myocytes ， adenovirus ， mutation
The sarcolemmal cardiac Na+-Ca2+ exchanger (NCX) catalyzes the exchange of three1 or possibly four2 Na+ ions for one Ca2+ ion across the membrane. Depending on electrochemical gradients and membrane potential, the NCX functions in either the Ca2+-efflux (forward) or Ca2+-influx (reverse) mode. In larger animals and in man, the NCX removes 30% of the total systolic Ca2+ during an intracellular Ca2+ (Ca2+i) transient to the extracellular space during diastole, matching the systolic influx via the L-type Ca2+ channel at steady-state.3 Under physiological conditions, reverse-mode NCX contributes only a small part to the Ca2+i transient during the early phase of the action potential (AP). In chronic heart failure, however, increased [Na+]i, decreased submembrane [Ca2+] and prolonged AP duration may facilitate NCX-mediated Ca2+ influx, which partially compensates for decreased sarcoplasmic reticulum (SR) Ca2+ release.4–9 On the other hand, enhanced forward mode NCX activity in heart failure may contribute to SR Ca2+ depletion. In this regard, inhibition of the NCX has recently been shown to increase Ca2+i transients in failing myocytes by improving SR Ca2+ load.10 Thus, it is critically important to understand the molecular basis underlying NCX regulation in the native myocyte.
The NCX consists of nine transmembrane segments that catalyze ion translocation across the membrane, and a large cytoplasmic loop (f-loop) that is not involved in ion translocation, but is essential for regulation of NCX activity by intracellular factors (see reviews11,12). An autoinhibitory region of 20 amino acids has been identified at the N-terminal domain of the f-loop, termed the endogenous exchanger inhibitory peptide (XIP) region.13 When applied to the intracellular surface of the NCX, a synthetic peptide with the same sequence as the endogenous XIP region potently inhibits NCX function.13
The endogenous XIP region plays an important role in the regulation of the NCX by intracellular Na+, Ca2+, and phosphatidylinositol-4,5-biphosphate (PIP2).14–16 Despite evidence that the endogenous XIP region interacts with a distinct region on the f-loop,17,18 this XIP interaction site has not been well characterized, and deletions that eliminated the XIP inhibitory effect also abolished allosteric Ca2+ activation of NCX,18 leaving open the question of whether these regulatory domains are functionally distinct. In the present study, we use adenoviral gene transfer technology and mutational analysis of NCX to identify a region on the C-terminal domain of the f-loop that is critically involved in regulation of the NCX by XIP, but not allosteric Ca2+, in the context of the native cardiomyocyte. Vice versa, we demonstrate that regions previously identified to be involved in allosteric Ca2+ activation19–22 are not involved in XIP-induced inhibition of NCX.
Viral Vectors
The coding sequence for the canine wild-type NCX (NCX1.1) was a gift of Dr Kenneth Philipson (University of California, Los Angeles). The deletion mutant 562–679 was designed to correspond to the overlapping region of two previously published mutations (240–679 and 562–685), which were shown to eliminate allosteric Ca2+ activation and XIP-induced inhibition of NCX current (INCX) in excised patch experiments in Xenopus oocytes.18 Furthermore, we created mutants with deletions of amino acids 440–456, 498–510, and 680–685. These f-loop sequences contain acidic residues that are functionally important Ca2+-binding19,20 and/or regulatory domains of NCX.19–22
The expression of NCX wild-type and deletion mutant genes was performed by transient transfection of human embryonic kidney 293 (HEK) cells or by adenoviral gene transfer to a human lung carcinoma cell line (A549 cells, American Type Culture Collection, Rockville, Md, respectively) or cultured guinea pig cardiomyocytes. A detailed description of viral vectors and gene transfer protocols can be found in the online data supplement available at http://circres.ahajournals.org.
Whole-Cell Patch-Clamp and Fluorescence Recording
For INCX measurements (Figure 1 to 4), cover slips were mounted in a heated recording chamber (37°C) on the stage of a fluorescence microscope (Nikon Eclipse TE300) and superfused with External Solution 1 (Table 1). The solution was K+-free to block inward rectifier K+- and Na+-K+-ATPase currents. Furthermore, L-type Ca2+ channels and Na+-K+-ATPase were blocked by nitrendipine (10 μmol/L) and strophanthidin (10 μmol/L). [Ca2+]i was buffered to 200 nmol/L (Pipette Solution 1, Table 1; calculated by Maxchelator program23). The pipette-to-bath liquid junction potential was –2.7 mV and was corrected. Where indicated, the pipette solution contained XIP (RRLLFYKYVYKRYRAGKQRG), synthesized and purified by the protein synthesis core facility of Johns Hopkins University, at 30 or 100 μmol/L. Cells were voltage-clamped at a holding potential (EH) of –40 mV, with families of pulses applied from +80 to –80 mV (in 20 mV steps) for 300 ms at 0.5 Hz (online Figure S1E in the online data supplement). After each pulse, voltage returned to EH. To block INCX, 5 mmol/L (cardiac myocytes) or 10 mmol/L (HEK and A549 cells) of NiCl2 were added temporarily to the superfusing solution at steady state INCX. After establishing a sufficient block of INCX, the external solution was switched back to Ni2+-free solution, and currents recovered within several minutes (Figures 1A and 3D). INCX was calculated as the current blocked by Ni2+ (online Figure S1). Mean capacitances of cells were 22±1 pF for HEK cells (n=82), 46±3 pF for A549 cells (n=52), and 68±3 pF for cardiac myocytes (n=48). Membrane currents were recorded, as described previously,24 in whole-cell voltage clamp mode (Axopatch 200A amplifier, Digidata 1200B interface, Axon Instruments) with 2 to 4 M pipets, to give typical total series resistances of <10 M. Electrophysiological signals were digitized, stored, and analyzed using custom-written software (IonView, B. O’Rourke).
Figure 1. Representative I vs t traces of mean currents of pulses from +80 to –80 mV at 0.5 Hz in HEK cells transfected with wild-type (A), 440–456 (B), or 562–679 mutant NCX (C) compared with nontransfected cells (D). A, left, Control pipette; after wash-in and -out of Ni2+ (10 mmol/L), the patch was excised and whole cell recording was reestablished in the same cell with a second pipette containing XIP (100 μmol/L; right). B and C, Pipettes contained 100 (B) or 30 μmol/L of XIP (C). Gray areas indicate the presence of Ni2+.
Figure 2. Left, Schematic illustrations of the NCX molecule, highlighting the f-loop regions deleted in the respective mutants of the NCX. Middles, Original recordings of Ni2+-sensitive currents induced by pulses from +80 to –80 mV, respectively. Right, Cumulative I/V plots of wild-type (A, Control pipette: n=8/ XIP, 100 μmol/L, pipette: n=7), 440–456 (B, n=5/5), 498–510 (C, n=6/5), 562–679 (D, n=10/7), and 680–685 (E, n=4/4) mutant NCX transiently transfected to HEK cells, respectively. INCX was significantly smaller in the presence of XIP compared with Control conditions at +80 mV for wild-type (P<0.001), 440–456 (P<0.05), 498–510 (P<0.05), and 680–685 (P<0.005), respectively.
Figure 3. Adenoviral gene transfer of NCX mutants in guinea pig cardiac myocytes. A, B, D, and E, Original tracings from experiments in myocytes infected with either the wild-type (A and B) or the mutant 562–679 NCX (D and E). A and D, Control pipettes; B and E, XIP (100 μmol/L) in the pipette. Shaded time-spans indicate the presence of Ni2+ (5 mmol/L) in the superfusate. Dotted horizontal lines indicate the reverse mode current at the first pulse (to +80 mV) in each experiment. In B, instead of repetitive families of pulses (from +80 to –80 mV), the cell was pulsed only to +80 mV (same frequency [0.5 Hz] and pulse duration [300 ms] as families). C and F, Cumulative comparison of the very first (Baseline) and maximal (Activated) reverse mode INCX at +80 mV (see arrows in A), in the absence (Control) and presence of 100 μmol/L of XIP, in wild-type (C; control, n=9/ XIP, n=6) and 562–679 infected myocytes (F; n=9/8), respectively. *P<0.005, **P<0.001 vs Baseline (paired t test).
Figure 4. Cumulative I/V plots from all experiments in noninfected (A; control, n=9/ XIP, 100 μmol/L, n=8), wild-type (B, n=9/6), and 562–679–infected myocytes (C, n=9/8). *P<0.05, **P<0.01, ***P<0.005, #P<0.001 vs control. Inset in C, Data of a "paired pipette" experiment in a 562–679 mutant–infected myocyte. After the routine protocol with a control pipette, the patch was excised and whole cell recording was re-established in the same cell with a second pipette containing XIP (see also Figure 1A).
[Ca2+]i measurements were performed as described previously24 using the K+-salt of indo-1 (100 μmol/L; Molecular Probes). Cellular autofluorescence was recorded before rupturing the cell-attached patch and subtracted before determining R (ratio of 405/495 nm emission). [Ca2+]i was calculated according to the equation [Ca2+]i=Kd*?*[(R–Rmin)/(Rmax–R)],25 using a Kd of 844 nmol/L,26 and experimentally determined Rmin=0.5, Rmax=4.8, and ?=1.83. Bath solution was identical to External Solution 1 (Table 1) except that [Ca2+]o was 2 mmol/L, and thapsigargin (1 μmol/L) and nitrendipine (10 μmol/L) were present to inhibit SR Ca2+-ATPase and L-type Ca2+-channels, respectively. The composition of the intracellular solution is given in Table 1 (Internal Solution 2).
In the experiments shown in Figure 5, according to the protocol of Weber et al,22 cells were held at EH of –100 mV for 3 to 5 minutes after break in to promote forward mode INCX (Ca2+ efflux) and achieve low initial [Ca2+]i. Thereafter, repetitive pulses from –100 to +80 mV were applied at 2 Hz. At +80 mV, reverse mode INCX (Ca2+ influx) was favored, whereas at –100 mV, the opposite held true. The duration of each pulse was 100 ms for noninfected, wild-type and 562–679 cells. Because 440–456, 498–510, and 680–685 mutants required higher [Ca2+]i for allosteric Ca2+ activation, pulse duration was set to 280 ms in these myocytes. During each pulse, instantaneous INCX (excluding the capacitive transient) was plotted against [Ca2+]i at a sampling interval of 0.3 ms, and half-maximal Ca2+-induced activation of INCX (KmCaAct) was calculated by nonlinear regression analysis (GraphPad Prism v3.00; GraphPad Software; Figure 5B). Mean capacitance of cardiac myocytes was 61±2 pF (n=46) with no differences between groups.
Figure 5. Simultaneous recordings of INCX and [Ca2+]i in cardiac myocytes. A, Original recordings from a myocyte infected with 562–679. Pulses 1, 4, 7, 14, and 25 are displayed. B, Instantaneous INCX of pulses 1, 4, and 7, plotted against instantaneous [Ca2+]i of the myocyte infected with 562–679 in A, compared with a different representative experiment with a myocyte infected with the 498–510 mutant NCX. C, Baseline and maximal Ca2+-activated reverse-mode INCX in noninfected (n=10), wild-type (n=7), 440–456 (n=6), 498–510 (n=6), 562–679 (n=9), and 680–685–infected myocytes (n=8), respectively. Ca2+-activated INCX P<0.05 vs respective baseline INCX in all groups, respectively (paired t test); *P<0.05 vs noninfected baseline or activated INCX, respectively.
Statistics
Values are given as mean±SEM. Statistical analysis was performed using Student t test for either paired or unpaired samples. For multiple comparisons, ANOVA analysis was performed.
To study the properties of the wild-type NCX and its deletion mutants in a heterologous cell system without native background NCX, transient transfection was performed in HEK cells. In Figure 1A, the average currents recorded during repeated test pulses for families of voltage steps from +80 to –80 mV are plotted versus the time of the experiment (I versus t) for the wild-type NCX. With control pipette solution ([Ca2+]i buffered at 200 nmol/L), currents were reversibly suppressed in the presence of Ni2+. The Ni2+-sensitive currents were regarded as NCX-specific currents and will be referred to as INCX. In this particular cell (unlike in most other experiments which compared unpaired cells), we excised the patch and reestablished whole-cell recording with a second pipette containing 100 μmol/L XIP in the same cell ("paired-pipette" method5). Dialysis of the cell with XIP continuously reduced INCX (Figure 1A).
In Figure 2A, Ni2+-sensitive current characteristics of a representative HEK cell transfected with the wild-type NCX (middle panel) and the cumulative current/voltage (I/V) plots of INCX are illustrated (right panel). With 30 μmol/L (not shown) or 100 μmol/L of XIP in the pipette, INCX was inhibited by 88% in the reverse mode and 73 (30 μmol/L) to 77% (100 μmol/L) in the forward mode, respectively (Figure 2A, right panel).
The left panels of Figure 2 illustrate the positions of the f-loop sequences that were deleted in the respective mutants of the NCX. All deletion mutants expressed INCX. NCX 562– 679 currents had properties similar to wild-type, with outward INCX showing activation during the pulse for depolarizations to high positive potentials (middle panels of Figure 2A and 2D). In contrast, 440–456, 498–510, and 680–685 mutants displayed different current kinetics, with constant or decreasing currents within single pulses at higher voltages (middle panels of Figure 2B, 2C, and 2E).
XIP (100 μmol/L) inhibited INCX in cells transfected with NCX 440–456 (Figures 1B and 2B), 498–510 (Figure 2C), or 680–685 mutants (Figure 2E). In contrast, in cells transfected with 562–679, INCX was not inhibited by XIP (Figure 2D). In Figure 1C, a representative I versus t plot for a 562–679 cell with XIP in the pipette is displayed. Though initially small, INCX continuously increased and reached steady state after approximately 3 minutes. In nontransfected HEK cells, no INCX could be detected (Figure 1D).
In the reverse mode (at +80 mV), INCX was 3.6-fold greater in wild-type compared with 562–679 mutant NCX (Figure 2A and 2D). Similar results were obtained when adenoviral gene transfer was performed in A549 cells (see online Figure S2). Reverse mode INCX was 3.4-fold greater in wild-type (21.0±5.0 pA/pF, n=8) compared with the 562–679 mutant (6.2±0.9 pA/pF, n=18; P<0.001). In contrast to wild-type, INCX in the 562–679 mutant was not inhibited by XIP (online Figure S2).
Because NCX function is regulated by a number of intracellular factors, it was important to study the cardiac NCX in its native environment. Thus, in vitro adenoviral gene transfer of wild-type and 562–679 mutant NCX to adult guinea pig cardiac myocytes was performed. In Figure 3, representative I versus t traces are displayed. With [Ca2+]i buffered to 200 nmol/L, myocytes infected with the wild-type NCX revealed a gradual pulse-dependent increase of INCX under control conditions (Figure 3A). With XIP in the pipette, however, this increase was abolished (Figure 3B). Figure 3C illustrates the mean baseline reverse mode INCX (first pulse of each experiment) compared with the maximum INCX in the respective experiments (see arrows in Figure 3A). A similar result was obtained in noninfected myocytes (Control, INCX increased from 3.4±0.9 to 6.3±0.5 pA/pF, P<0.001; XIP, 2.0±0.8 to 3.8±0.4 pA/pF; P=NS). Myocytes infected with the 562–679 mutant displayed similar activation of INCX compared with wild-type NCX (Figure 3D). However, the presence of XIP did not inhibit the activation of INCX in this mutant (Figure 3E and 3F).
The cumulative I/V plots of all groups are shown in Figure 4A through 4C. Compared with noninfected cells, INCX density at +80 mV was 1.5-fold higher in myocytes infected with wild-type NCX (P<0.01). In contrast, myocytes expressing the 562–679 mutant had slightly lower INCX density compared with noninfected myocytes (P<0.05). Although XIP inhibited INCX in noninfected (Figure 4A) and wild-type infected cells (Figure 4B), myocytes expressing the 562–679 mutant were insensitive to XIP (Figure 4C). The inset in Figure 4C illustrates a "paired pipette" experiment in which a second patch (XIP) could be reestablished after excision of the first patch (control) in the same cell.
The pulse-dependent activation of INCX (Figure 3) suggested that allosteric Ca2+ activation may be present in both wild-type and 562–679–infected myocytes. To analyze this more thoroughly, [Ca2+]i transients elicited by reverse mode INCX were measured in cardiac myocytes without intracellular Ca2+ buffering.22 By blocking L-type Ca2+ channels and the SR and pulsing cells from a holding potential of –100 mV to +80 mV at 2 Hz, reverse mode INCX increased [Ca2+]i from 186±14 to 1444±161 nmol/L (n=46). In Figure 5A, representative recordings of INCX and [Ca2+]i from a myocyte infected with the 562–679 mutant are displayed. At pulse 1, diastolic [Ca2+]i was very low (30 to 50 nmol/L). With increasing [Ca2+]i, INCX also increased dynamically (pulses 4 and 7). In pulse 14, despite continuously rising [Ca2+]i, INCX saturated and even decreased during the late phase of the pulse. In pulse 25, INCX decreased during most of the pulse duration. According to the model of Weber et al,22 we attributed the Ca2+i-induced increase of reverse-mode INCX to allosteric activation by Ca2+i. When plotting instantaneous [Ca2+]i against INCX during these pulses, the data could be fit by the Hill equation (Figure 5B). In noninfected and wild-type infected myocytes, KmCaAct values were 298 and 341 nmol/L, respectively. For the 562–679 mutant, KmCaAct values were shifted only slightly (and statistically insignificant) to 375 nmol/L (Table 2, Figure 5B). In contrast, in 440–456, 498–510, and 680–685 mutant-infected myocytes, allosteric Ca2+ activation of INCX was significantly shifted toward higher [Ca2+]i (710 to 825 nmol/L; Table 2 and Figure 5B). The slope factors of the fits were not different between groups (mean, 4.3±0.2; n=45). Baseline INCX in 440–456 and 680–685 mutants were reduced compared with noninfected myocytes, whereas the maximum INCX was reduced only in 440–456 mutant-infected cells (Figure 5C).
The present study for the first time functionally distinguishes the XIP-regulatory from Ca2+-regulatory domains of the cardiac NCX. Based on mutational analysis of NCX in intact cardiac myocytes, we propose that, in contrast to previous results,18 amino acids 562–679 are not involved in regulation of INCX by Ca2+, but are required for regulation by XIP. In contrast, residues 440–456, 498–510, and 680–685 of the f-loop are involved in regulation of INCX by Ca2+, but not by XIP.
Regulation of NCX by XIP
The NCX is regulated by intracellular Na+ and Ca2+. In excised giant sarcolemmal patches, application of Na+i in the presence of Ca2+i initially activates reverse mode INCX, followed by a process of partial inactivation that reaches steady state within several seconds (Na+-dependent inactivation, I1).27 On the other hand, increasing [Ca2+]i activates INCX, which in the absence of Ca2+i is in the Na+-independent inactive state (I2).
The endogenous XIP region plays an important role in the physiological regulation of INCX by intracellular factors. Na+-dependent inactivation is mediated by the XIP region.14 PIP2 activates INCX by reducing Na+-dependent inactivation,16 presumably by binding to the endogenous XIP region and thus inhibiting the interaction with its putative binding site on the f-loop.15 This binding site, however, is not well characterized. Matsuoka et al18 found that mutants of the NCX lacking residues 240–679 or 562–685 of the f-loop were both insensitive to XIP. They proposed that the overlapping region of these mutants (ie, residues 562–679) may contain the XIP binding site. Conversely, Hale et al28 observed that XIP directly bound to peptides corresponding to residues 445– 455 of the central f-loop. In the present study, the deletion of amino acids 562–679 eliminated XIP-induced inhibition of INCX in heterologous expression systems and cardiac myocytes. In contrast, deletion of residues 440–456 or two other negatively charged regions (498–510 and 680–685) did not affect XIP-regulation of INCX.
In most experimental systems, the inhibitory effect of exogenous XIP saturated between 10 and 30 μmol/L.10,13,17,18,29,30 In the present study, both 30 and 100 μmol/L of XIP inhibited reverse mode INCX by 88% in HEK cells, respectively. In cardiac myocytes, we used 100 μmol/L to achieve maximum inhibition of INCX. The inhibitory efficacy of 100 μmol/L of XIP on INCX was 88% in HEK cells, 82% in A549 cells, but only 40% to 60% in cardiac myocytes.
The reason for this difference is unclear. The cardiac NCX associates with caveolin-3, and caveolin-binding motifs are located in the endogenous XIP region (residues 223–231), but also within the herein identified XIP-regulatory domain (residues 620–627).31 It was suggested that by its association with the endogenous XIP region (or possibly also its regulatory domain), caveolin-3 may activate the NCX by inhibiting the interaction of the endogenous XIP region with its binding site.31 If this holds true, then higher membrane content of caveolin-3 in cardiac myocytes may reduce the inhibitory efficacy of exogenous XIP compared with HEK cells, where caveolin-3 is virtually absent.32 In a similar way, ATP- and PIP2-induced activation of INCX are supposedly attributable to interaction of PIP2 with the endogenous XIP region, thus inhibiting its interaction with the XIP regulatory site.15,16 Because exogenous XIP also avidly binds to PIP2,15 it may be speculated that differences in the native membrane content of PIP2 between cardiac myocytes, HEK, and A549 cells may also account for the different inhibitory efficacy of XIP in these cells. Future studies will be directed to elucidate regulation of NCX by PIP2 and caveolin-3 in its native cellular environment.
The inhibitory efficacy of XIP in whole-cell patch-clamped cardiac myocytes is within the range reported in previous studies.10,30 The results from these studies suggest that [Ca2+]i may have an impact on the efficacy of XIP. In guinea pig30 or canine cardiac myocytes,10 XIP inhibited INCX by 65% when [Ca2+]i was buffered to 100 nmol/L. In contrast, in nonbuffered conditions with [Ca2+]i between 300 and 500 nmol/L, XIP inhibited INCX by only 27%.10 In the present study, [Ca2+]i was buffered to 200 nmol/L, and XIP inhibited INCX by 40% in noninfected guinea pig cardiac myocytes and by 60% in myocytes expressing the canine wild-type NCX. Although the protein sequence of the endogenous XIP region is identical between the guinea pig33 and the canine34 NCX, the XIP-interaction domain (residues 562–679) is 95% identical between the two species. It is unclear whether this contributed to the different inhibitory efficacy of XIP in cardiac myocytes expressing either guinea pig or canine NCX.
Regulation of NCX by Ca2+
In HEK cells, the wild-type and the 562–679 mutant displayed activation of reverse-mode INCX during pulses to higher voltages. This may be attributable to allosteric activation of INCX by Ca[2+]i.22 In contrast, activation of INCX was absent in NCX 440–456, 498–510, or 680–685 under these conditions (Figure 2). In cardiac myocytes with [Ca2+]i buffered to 200 nmol/L, INCX remained unchanged within single pulses; however, the average INCX was monotonically activated within several seconds to minutes. This activation pattern of INCX was similar for wild-type and 562–679 mutant NCX (Figure 3). To quantify allosteric activation of INCX by [Ca2+]i in cardiac myocytes under nonbuffered [Ca2+]i conditions, we used a protocol similar to that of Weber et al,22 who determined Ca2+ activation by averaging INCX during each pulse and observed an increase in mean reverse mode INCX and mean [Ca2+]i from pulse to pulse. Within each pulse, INCX tended to decrease despite rising [Ca2+]i in most of their experiments,22 probably as a result of decreased thermodynamic driving forces for ion transport. In contrast, we observed increasing INCX within each of the first few pulses as well as from pulse to pulse (Figure 5A), as predicted by computational modeling.22
By plotting the instantaneous INCX against [Ca2+]i during each pulse (Figure 5B), we calculated KmCaAct values of 298 and 341 nmol/L for the native guinea pig NCX and the (canine) wild-type NCX expressed in guinea pig myocytes, respectively. These values are in agreement with previous results for the canine, guinea pig, or bovine NCX.20,22,35–37 Nevertheless, there may be species-dependent differences regarding allosteric Ca2+ activation. In human7 and ferret myocytes,22 KmCaAct values were only 150 and 125 nmol/L, respectively. In mouse myocytes, Ca2+ activation was observed in inside-out patches,21 but not in whole-cell patches.22 In the 562–679 mutant NCX, KmCaAct values were not significantly different from wild-type- and noninfected myocytes (375 nmol/L). Even in the presence of XIP, allosteric Ca2+ activation was preserved in this mutant (Figure 3).
Regulation of INCX by [Ca2+]i is complex and involves multiple regions of the NCX molecule.
Biochemical19,38 and electrophysiological studies20 have identified two regions on the central f-loop (residues 440–456 and 498–510) containing aspartate residues that bind Ca2+ with high affinity and in a cooperative manner. Mobility shifts during SDS-PAGE in the presence of Ca2+ suggested that the binding of Ca2+ to these residues induces a conformational change of the NCX molecule.19 Point mutations or deletions of short stretches within these two regions shifted the Ca2+-binding or activation curves from 300 nmol/L to the micromolar range.19,20 Accordingly, in the present study, the deletion of either amino acids 440–456 or 498–510 shifted the KmCaAct values of allosteric Ca2+ activation from 300 nmol/L to 710 and 765 nmol/L, respectively. The slope factors of 4 for Ca2+ activation in our experiments are in agreement with (although not a proof of) the concept of allosteric Ca2+ activation of INCX as a cooperative process.
Similarly, deletion of amino acids 680–685 shifted the Ca2+ activation curves of INCX to 825 nmol/L, confirming that this C-terminal f-loop region plays an important role in Ca2+ regulation.21,22 The functional role of this region, however, is not entirely clear. No direct binding of Ca2+ to these regions could be observed using the 45Ca2+ overlay technique.19 When transgenically overexpressed in mice, excised patch measurements of myocyte blebs indicated that the 680–685 mutant was constitutively activated already at low [Ca2+]i.21 In our experiments, however, allosteric Ca2+ activation was not absent, but rather shifted to higher [Ca2+]i. The reason for this discrepancy is unclear. Different experimental conditions (whole cell versus excised patches of myocyte blebs) may create different microenvironments of NCX regarding membrane-associated regulatory factors (eg, PIP215,16 or caveolin-331). However, the mechanisms of these interactions are presently not fully understood.
Impact of the Native NCX in Cardiac Myocytes After Adenoviral Gene Transfer
In the present study, adenoviral gene transfer of the wild-type and deletion mutants of NCX was performed in guinea pig cardiac myocytes that already express native NCX. It is therefore important to consider how much of INCX is related to native and transferred NCX, respectively. At least two observations indicate that the majority of native NCX appears to have been replaced by the virally-expressed genes, probably as a result of competition between native and mutant mRNAs for the translation machinery. First, in 562–679 mutant-expressing myocytes, INCX was completely insensitive to XIP. If native NCX was still active, at least a small fraction of INCX should have been inhibited by XIP. Second, in myocytes expressing 440–456, 498–510, and 680–685 mutants, the allosteric Ca2+-activation curves were shifted to higher [Ca2+]i, however, without a change in the slope factor. If native NCX had still contributed to INCX, its lower KmCaAct should have decreased the slope factor, and two distinct Ca2+-activation sites should have been observed. Hence, we conclude that native NCX was largely replaced with the mutant NCX.
In cardiac myocytes, wild-type NCX-infected myocytes showed 2-fold higher INCX than 562–679 infected and 1.5-fold higher INCX than noninfected myocytes. It is difficult to differentiate whether this is attributable to overexpression of the canine NCX alone or to coexpression of canine and native (guinea pig) NCX. However, the fact that in HEK and A549 cells, the 562–679 mutant consistently exerted 3.5-fold lower INCX compared with the wild-type NCX (despite different gene transduction techniques) suggests that the deletion mutant may exert intrinsically lower exchange activity. If this is true, it may be speculated that both wild-type and 562–679 NCX proteins were modestly overexpressed compared with noninfected myocytes, with the deletion mutant exerting intrinsically lower INCX.
Conclusions
In summary, we propose that residues 562–679 of the f-loop contain the binding site for the endogenous XIP domain, but are not involved in allosteric Ca2+ regulation of INCX. In contrast, residues 440–456, 498–510, and 680–685 are critical domains for regulation of INCX by Ca2+, but not XIP. The fact that neither deletion mutant used by Matsuoka et al18 was regulated by Ca2+ may reflect deletion of either the central Ca2+-binding domains (residues 440–510, deleted in 240–679) or the C-terminal regulatory domain (residues 680–685, deleted in 562–685).
The 562–679 mutant may represent a novel tool to further investigate regulation of INCX by intracellular factors that possibly involve the interaction of the endogenous XIP region and its binding domain in the environment of the native cardiac myocyte (eg, PIP215,16 and caveolin-331). Furthermore, because selective NCX-inhibitors have been postulated to improve Ca2+ homeostasis in chronic heart failure,10 the development of such drugs may be facilitated by a better understanding of XIP-mediated regulation of the NCX.
References
Kimura J, Miyamae S, Noma A. Identification of sodium-calcium exchange current in single ventricular cells of guinea-pig. J Physiol. 1987; 384: 199–222.
Fujioka Y, Komeda M, Matsuoka S. Stoichiometry of Na+-Ca2+ exchange in inside-out patches excised from guinea-pig ventricular myocytes. J Physiol. 2000; 523: 339–351.
Bers DM. Calcium fluxes involved in control of cardiac myocyte contraction. Circ Res. 2000; 87: 275–281.
Despa S, Islam MA, Weber CR, Pogwizd SM, Bers DM. Intracellular Na+ concentration is elevated in heart failure but Na/K pump function is unchanged. Circulation. 2002; 105: 2543–2548.
Armoundas AA, Hobai IA, Tomaselli GF, Winslow RL, O’Rourke B. Role of sodium-calcium exchanger in modulating the action potential of ventricular myocytes from normal and failing hearts. Circ Res. 2003; 93: 46–53.
Pieske B, Maier LS, Piacentino V, 3rd, Weisser J, Hasenfuss G, Houser S. Rate dependence of [Na+]i and contractility in nonfailing and failing human myocardium. Circulation. 2002; 106: 447–453.
Weber CR, Piacentino V, 3rd, Houser SR, Bers DM. Dynamic regulation of sodium/calcium exchange function in human heart failure. Circulation. 2003; 108: 2224–2229.
Weber CR, Piacentino V, 3rd, Ginsburg KS, Houser SR, Bers DM. Na+-Ca2+ exchange current and submembrane [Ca2+] during the cardiac action potential. Circ Res. 2002; 90: 182–189.
Dipla K, Mattiello JA, Margulies KB, Jeevanandam V, Houser SR. The sarcoplasmic reticulum and the Na+/Ca2+ exchanger both contribute to the Ca2+ transient of failing human ventricular myocytes. Circ Res. 1999; 84: 435–444.
Hobai IA, Maack C, O’Rourke B. Partial inhibition of sodium/calcium exchange restores cellular calcium handling in canine heart failure. Circ Res. 2004; 95: 292–299.
Philipson KD, Nicoll DA. Sodium-calcium exchange: a molecular perspective. Annu Rev Physiol. 2000; 62: 111–133.
Shigekawa M, Iwamoto T. Cardiac Na+-Ca2+ exchange: molecular and pharmacological aspects. Circ Res. 2001; 88: 864–876.
Li Z, Nicoll DA, Collins A, Hilgemann DW, Filoteo AG, Penniston JT, Weiss JN, Tomich JM, Philipson KD. Identification of a peptide inhibitor of the cardiac sarcolemmal Na+-Ca2+ exchanger. J Biol Chem. 1991; 266: 1014–1020.
Matsuoka S, Nicoll DA, He Z, Philipson KD. Regulation of cardiac Na+-Ca2+ exchanger by the endogenous XIP region. J Gen Physiol. 1997; 109: 273–286.
He Z, Feng S, Tong Q, Hilgemann DW, Philipson KD. Interaction of PIP(2) with the XIP region of the cardiac Na/Ca exchanger. Am J Physiol Cell Physiol. 2000; 278: C661–6.
Hilgemann DW, Ball R. Regulation of cardiac Na+,Ca2+ exchange and KATP potassium channels by PIP2. Science. 1996; 273: 956–959.
Kleiboeker SB, Milanick MA, Hale CC. Interactions of the exchange inhibitory peptide with Na-Ca exchange in bovine cardiac sarcolemmal vesicles and ferret red cells. J Biol Chem. 1992; 267: 17836–17841.
Matsuoka S, Nicoll DA, Reilly RF, Hilgemann DW, Philipson KD. Initial localization of regulatory regions of the cardiac sarcolemmal Na+-Ca2+ exchanger. Proc Natl Acad Sci U S A. 1993; 90: 3870–3874.
Levitsky DO, Nicoll DA, Philipson KD. Identification of the high affinity Ca2+-binding domain of the cardiac Na+-Ca2+ exchanger. J Biol Chem. 1994; 269: 22847–22852.
Matsuoka S, Nicoll DA, Hryshko LV, Levitsky DO, Weiss JN, Philipson KD. Regulation of the cardiac Na+-Ca2+ exchanger by Ca2+: mutational analysis of the Ca2+-binding domain. J Gen Physiol. 1995; 105: 403–420.
Maxwell K, Scott J, Omelchenko A, Lukas A, Lu L, Lu Y, Hnatowich M, Philipson KD, Hryshko LV. Functional role of ionic regulation of Na+/Ca2+ exchange assessed in transgenic mouse hearts. Am J Physiol. 1999; 277: H2212–21.
Weber CR, Ginsburg KS, Philipson KD, Shannon TR, Bers DM. Allosteric regulation of Na/Ca exchange current by cytosolic Ca in intact cardiac myocytes. J Gen Physiol. 2001; 117: 119–131.
Patton C, Thompson S, Epel D. Some precaustions in using chelators to buffer metals in biological solutions. Cell Calcium. 2004; 35: 427–431.
O’Rourke B, Kass DA, Tomaselli GF, Kaab S, Tunin R, Marban E. Mechanisms of altered excitation-contraction coupling in canine tachycardia-induced heart failure, I: experimental studies. Circ Res. 1999; 84: 562–570.
Grynkiewicz G, Poenie M, Tsien RY. A new generation of Ca2+ indicators with greatly improved fluorescence properties. J Biol Chem. 1985; 260: 3440–3450.
Bassani JW, Bassani RA, Bers DM. Calibration of indo-1 and resting intracellular [Ca]i in intact rabbit cardiac myocytes. Biophys J. 1995; 68: 1453–1460.
Hilgemann DW. Regulation and deregulation of cardiac Na+-Ca2+ exchange in giant excised sarcolemmal membrane patches. Nature. 1990; 344: 242–245.
Hale CC, Bliler S, Quinn TP, Peletskaya EN. Localization of an exchange inhibitory peptide (XIP) binding site on the cardiac sodium-calcium exchanger. Biochem Biophys Res Commun. 1997; 236: 113–117.
Xu W, Denison H, Hale CC, Gatto C, Milanick MA. Identification of critical positive charges in XIP, the Na/Ca exchange inhibitory peptide. Arch Biochem Biophys. 1997; 341: 273–279.
Chin TK, Spitzer KW, Philipson KD, Bridge JH. The effect of exchanger inhibitory peptide (XIP) on sodium-calcium exchange current in guinea pig ventricular cells. Circ Res. 1993; 72: 497–503.
Bossuyt J, Taylor BE, James-Kracke M, Hale CC. The cardiac sodium-calcium exchanger associates with caveolin-3. Ann NY Acad Sci. 2002; 976: 197–204.
Cha SH, Jung NH, Kim BR, Kim HW, Kwak JO. Evidence for cyclooxygenase-1 association with caveolin-1 and -2 in cultured human embryonic kidney (HEK 293) cells. IUBMB Life. 2004; 56: 221–227.
Tsuruya Y, Bersohn MM, Li Z, Nicoll DA, Philipson KD. Molecular cloning and functional expression of the guinea pig cardiac Na+-Ca2+ exchanger. Biochim Biophys Acta. 1994; 1196: 97–99.
Nicoll DA, Longoni S, Philipson KD. Molecular cloning and functional expression of the cardiac sarcolemmal Na+-Ca2+ exchanger. Science. 1990; 250: 562–565.
Reeves JP, Condrescu M. Allosteric activation of sodium-calcium exchange activity by calcium: persistence at low calcium concentrations. J Gen Physiol. 2003; 122: 621–639.
Trac M, Dyck C, Hnatowich M, Omelchenko A, Hryshko LV. Transport and regulation of the cardiac Na+-Ca2+ exchanger, NCX1. Comparison between Ca2+ and Ba2+ J Gen Physiol. 1997; 109: 361–369.
Hilgemann DW, Matsuoka S, Nagel GA, Collins A. Steady-state and dynamic properties of cardiac sodium-calcium exchange. Sodium-dependent inactivation. J Gen Physiol. 1992; 100: 905–932.
Levitsky DO, Fraysse B, Leoty C, Nicoll DA, Philipson KD. Cooperative interaction between Ca2+ binding sites in the hydrophilic loop of the Na+-Ca2+ exchanger. Mol Cell Biochem. 1996; 160–161:27–32.(Christoph Maack, Anand Ga)
The sarcolemmal Na+-Ca2+ exchanger (NCX) is the main Ca2+ extrusion mechanism in cardiac myocytes and is thus essential for the regulation of Ca2+ homeostasis and contractile function. A cytosolic region (f-loop) of the protein mediates regulation of NCX function by intracellular factors including inhibition by exchanger inhibitory peptide (XIP), a 20 amino acid peptide matching the sequence of an autoinhibitory region involved in allosteric regulation of NCX by intracellular Na+, Ca2+, and phosphatidylinositol-4,5-biphosphate (PIP2). Previous evidence indicates that the XIP interaction domain can be eliminated by large deletions of the f-loop that also remove activation of NCX by intracellular Ca2+. By whole-cell voltage clamping experiments, we demonstrate that deletion of residues 562–679, but not 440– 456, 498–510, or 680–685 of the f-loop selectively eliminates XIP-mediated inhibition of NCX expressed either heterologously (HEK293 and A549 cells) or in guinea pig cardiac myocytes. In contrast, by plotting INCX against reverse-mode NCX-mediated Ca2+ transients in myocytes, we demonstrate that Ca2+-dependent regulation of NCX is preserved in 562–679, but significantly reduced in the other three deletion mutants. The findings indicate that f-loop residues 562–679 may contain the regulatory site for endogenous XIP, but this site is distinct from the Ca2+-regulatory domains of the NCX. Because regulation of the NCX by Na+ and PIP2 involves the endogenous XIP region, the 562–679 mutant NCX may be a useful tool to investigate this regulation in the context of the whole cardiac myocyte.
Key Words: cardiac myocytes ， adenovirus ， mutation
The sarcolemmal cardiac Na+-Ca2+ exchanger (NCX) catalyzes the exchange of three1 or possibly four2 Na+ ions for one Ca2+ ion across the membrane. Depending on electrochemical gradients and membrane potential, the NCX functions in either the Ca2+-efflux (forward) or Ca2+-influx (reverse) mode. In larger animals and in man, the NCX removes 30% of the total systolic Ca2+ during an intracellular Ca2+ (Ca2+i) transient to the extracellular space during diastole, matching the systolic influx via the L-type Ca2+ channel at steady-state.3 Under physiological conditions, reverse-mode NCX contributes only a small part to the Ca2+i transient during the early phase of the action potential (AP). In chronic heart failure, however, increased [Na+]i, decreased submembrane [Ca2+] and prolonged AP duration may facilitate NCX-mediated Ca2+ influx, which partially compensates for decreased sarcoplasmic reticulum (SR) Ca2+ release.4–9 On the other hand, enhanced forward mode NCX activity in heart failure may contribute to SR Ca2+ depletion. In this regard, inhibition of the NCX has recently been shown to increase Ca2+i transients in failing myocytes by improving SR Ca2+ load.10 Thus, it is critically important to understand the molecular basis underlying NCX regulation in the native myocyte.
The NCX consists of nine transmembrane segments that catalyze ion translocation across the membrane, and a large cytoplasmic loop (f-loop) that is not involved in ion translocation, but is essential for regulation of NCX activity by intracellular factors (see reviews11,12). An autoinhibitory region of 20 amino acids has been identified at the N-terminal domain of the f-loop, termed the endogenous exchanger inhibitory peptide (XIP) region.13 When applied to the intracellular surface of the NCX, a synthetic peptide with the same sequence as the endogenous XIP region potently inhibits NCX function.13
The endogenous XIP region plays an important role in the regulation of the NCX by intracellular Na+, Ca2+, and phosphatidylinositol-4,5-biphosphate (PIP2).14–16 Despite evidence that the endogenous XIP region interacts with a distinct region on the f-loop,17,18 this XIP interaction site has not been well characterized, and deletions that eliminated the XIP inhibitory effect also abolished allosteric Ca2+ activation of NCX,18 leaving open the question of whether these regulatory domains are functionally distinct. In the present study, we use adenoviral gene transfer technology and mutational analysis of NCX to identify a region on the C-terminal domain of the f-loop that is critically involved in regulation of the NCX by XIP, but not allosteric Ca2+, in the context of the native cardiomyocyte. Vice versa, we demonstrate that regions previously identified to be involved in allosteric Ca2+ activation19–22 are not involved in XIP-induced inhibition of NCX.
Viral Vectors
The coding sequence for the canine wild-type NCX (NCX1.1) was a gift of Dr Kenneth Philipson (University of California, Los Angeles). The deletion mutant 562–679 was designed to correspond to the overlapping region of two previously published mutations (240–679 and 562–685), which were shown to eliminate allosteric Ca2+ activation and XIP-induced inhibition of NCX current (INCX) in excised patch experiments in Xenopus oocytes.18 Furthermore, we created mutants with deletions of amino acids 440–456, 498–510, and 680–685. These f-loop sequences contain acidic residues that are functionally important Ca2+-binding19,20 and/or regulatory domains of NCX.19–22
The expression of NCX wild-type and deletion mutant genes was performed by transient transfection of human embryonic kidney 293 (HEK) cells or by adenoviral gene transfer to a human lung carcinoma cell line (A549 cells, American Type Culture Collection, Rockville, Md, respectively) or cultured guinea pig cardiomyocytes. A detailed description of viral vectors and gene transfer protocols can be found in the online data supplement available at http://circres.ahajournals.org.
Whole-Cell Patch-Clamp and Fluorescence Recording
For INCX measurements (Figure 1 to 4), cover slips were mounted in a heated recording chamber (37°C) on the stage of a fluorescence microscope (Nikon Eclipse TE300) and superfused with External Solution 1 (Table 1). The solution was K+-free to block inward rectifier K+- and Na+-K+-ATPase currents. Furthermore, L-type Ca2+ channels and Na+-K+-ATPase were blocked by nitrendipine (10 μmol/L) and strophanthidin (10 μmol/L). [Ca2+]i was buffered to 200 nmol/L (Pipette Solution 1, Table 1; calculated by Maxchelator program23). The pipette-to-bath liquid junction potential was –2.7 mV and was corrected. Where indicated, the pipette solution contained XIP (RRLLFYKYVYKRYRAGKQRG), synthesized and purified by the protein synthesis core facility of Johns Hopkins University, at 30 or 100 μmol/L. Cells were voltage-clamped at a holding potential (EH) of –40 mV, with families of pulses applied from +80 to –80 mV (in 20 mV steps) for 300 ms at 0.5 Hz (online Figure S1E in the online data supplement). After each pulse, voltage returned to EH. To block INCX, 5 mmol/L (cardiac myocytes) or 10 mmol/L (HEK and A549 cells) of NiCl2 were added temporarily to the superfusing solution at steady state INCX. After establishing a sufficient block of INCX, the external solution was switched back to Ni2+-free solution, and currents recovered within several minutes (Figures 1A and 3D). INCX was calculated as the current blocked by Ni2+ (online Figure S1). Mean capacitances of cells were 22±1 pF for HEK cells (n=82), 46±3 pF for A549 cells (n=52), and 68±3 pF for cardiac myocytes (n=48). Membrane currents were recorded, as described previously,24 in whole-cell voltage clamp mode (Axopatch 200A amplifier, Digidata 1200B interface, Axon Instruments) with 2 to 4 M pipets, to give typical total series resistances of <10 M. Electrophysiological signals were digitized, stored, and analyzed using custom-written software (IonView, B. O’Rourke).
Figure 1. Representative I vs t traces of mean currents of pulses from +80 to –80 mV at 0.5 Hz in HEK cells transfected with wild-type (A), 440–456 (B), or 562–679 mutant NCX (C) compared with nontransfected cells (D). A, left, Control pipette; after wash-in and -out of Ni2+ (10 mmol/L), the patch was excised and whole cell recording was reestablished in the same cell with a second pipette containing XIP (100 μmol/L; right). B and C, Pipettes contained 100 (B) or 30 μmol/L of XIP (C). Gray areas indicate the presence of Ni2+.
Figure 2. Left, Schematic illustrations of the NCX molecule, highlighting the f-loop regions deleted in the respective mutants of the NCX. Middles, Original recordings of Ni2+-sensitive currents induced by pulses from +80 to –80 mV, respectively. Right, Cumulative I/V plots of wild-type (A, Control pipette: n=8/ XIP, 100 μmol/L, pipette: n=7), 440–456 (B, n=5/5), 498–510 (C, n=6/5), 562–679 (D, n=10/7), and 680–685 (E, n=4/4) mutant NCX transiently transfected to HEK cells, respectively. INCX was significantly smaller in the presence of XIP compared with Control conditions at +80 mV for wild-type (P<0.001), 440–456 (P<0.05), 498–510 (P<0.05), and 680–685 (P<0.005), respectively.
Figure 3. Adenoviral gene transfer of NCX mutants in guinea pig cardiac myocytes. A, B, D, and E, Original tracings from experiments in myocytes infected with either the wild-type (A and B) or the mutant 562–679 NCX (D and E). A and D, Control pipettes; B and E, XIP (100 μmol/L) in the pipette. Shaded time-spans indicate the presence of Ni2+ (5 mmol/L) in the superfusate. Dotted horizontal lines indicate the reverse mode current at the first pulse (to +80 mV) in each experiment. In B, instead of repetitive families of pulses (from +80 to –80 mV), the cell was pulsed only to +80 mV (same frequency [0.5 Hz] and pulse duration [300 ms] as families). C and F, Cumulative comparison of the very first (Baseline) and maximal (Activated) reverse mode INCX at +80 mV (see arrows in A), in the absence (Control) and presence of 100 μmol/L of XIP, in wild-type (C; control, n=9/ XIP, n=6) and 562–679 infected myocytes (F; n=9/8), respectively. *P<0.005, **P<0.001 vs Baseline (paired t test).
Figure 4. Cumulative I/V plots from all experiments in noninfected (A; control, n=9/ XIP, 100 μmol/L, n=8), wild-type (B, n=9/6), and 562–679–infected myocytes (C, n=9/8). *P<0.05, **P<0.01, ***P<0.005, #P<0.001 vs control. Inset in C, Data of a "paired pipette" experiment in a 562–679 mutant–infected myocyte. After the routine protocol with a control pipette, the patch was excised and whole cell recording was re-established in the same cell with a second pipette containing XIP (see also Figure 1A).
[Ca2+]i measurements were performed as described previously24 using the K+-salt of indo-1 (100 μmol/L; Molecular Probes). Cellular autofluorescence was recorded before rupturing the cell-attached patch and subtracted before determining R (ratio of 405/495 nm emission). [Ca2+]i was calculated according to the equation [Ca2+]i=Kd*?*[(R–Rmin)/(Rmax–R)],25 using a Kd of 844 nmol/L,26 and experimentally determined Rmin=0.5, Rmax=4.8, and ?=1.83. Bath solution was identical to External Solution 1 (Table 1) except that [Ca2+]o was 2 mmol/L, and thapsigargin (1 μmol/L) and nitrendipine (10 μmol/L) were present to inhibit SR Ca2+-ATPase and L-type Ca2+-channels, respectively. The composition of the intracellular solution is given in Table 1 (Internal Solution 2).
In the experiments shown in Figure 5, according to the protocol of Weber et al,22 cells were held at EH of –100 mV for 3 to 5 minutes after break in to promote forward mode INCX (Ca2+ efflux) and achieve low initial [Ca2+]i. Thereafter, repetitive pulses from –100 to +80 mV were applied at 2 Hz. At +80 mV, reverse mode INCX (Ca2+ influx) was favored, whereas at –100 mV, the opposite held true. The duration of each pulse was 100 ms for noninfected, wild-type and 562–679 cells. Because 440–456, 498–510, and 680–685 mutants required higher [Ca2+]i for allosteric Ca2+ activation, pulse duration was set to 280 ms in these myocytes. During each pulse, instantaneous INCX (excluding the capacitive transient) was plotted against [Ca2+]i at a sampling interval of 0.3 ms, and half-maximal Ca2+-induced activation of INCX (KmCaAct) was calculated by nonlinear regression analysis (GraphPad Prism v3.00; GraphPad Software; Figure 5B). Mean capacitance of cardiac myocytes was 61±2 pF (n=46) with no differences between groups.
Figure 5. Simultaneous recordings of INCX and [Ca2+]i in cardiac myocytes. A, Original recordings from a myocyte infected with 562–679. Pulses 1, 4, 7, 14, and 25 are displayed. B, Instantaneous INCX of pulses 1, 4, and 7, plotted against instantaneous [Ca2+]i of the myocyte infected with 562–679 in A, compared with a different representative experiment with a myocyte infected with the 498–510 mutant NCX. C, Baseline and maximal Ca2+-activated reverse-mode INCX in noninfected (n=10), wild-type (n=7), 440–456 (n=6), 498–510 (n=6), 562–679 (n=9), and 680–685–infected myocytes (n=8), respectively. Ca2+-activated INCX P<0.05 vs respective baseline INCX in all groups, respectively (paired t test); *P<0.05 vs noninfected baseline or activated INCX, respectively.
Statistics
Values are given as mean±SEM. Statistical analysis was performed using Student t test for either paired or unpaired samples. For multiple comparisons, ANOVA analysis was performed.
To study the properties of the wild-type NCX and its deletion mutants in a heterologous cell system without native background NCX, transient transfection was performed in HEK cells. In Figure 1A, the average currents recorded during repeated test pulses for families of voltage steps from +80 to –80 mV are plotted versus the time of the experiment (I versus t) for the wild-type NCX. With control pipette solution ([Ca2+]i buffered at 200 nmol/L), currents were reversibly suppressed in the presence of Ni2+. The Ni2+-sensitive currents were regarded as NCX-specific currents and will be referred to as INCX. In this particular cell (unlike in most other experiments which compared unpaired cells), we excised the patch and reestablished whole-cell recording with a second pipette containing 100 μmol/L XIP in the same cell ("paired-pipette" method5). Dialysis of the cell with XIP continuously reduced INCX (Figure 1A).
In Figure 2A, Ni2+-sensitive current characteristics of a representative HEK cell transfected with the wild-type NCX (middle panel) and the cumulative current/voltage (I/V) plots of INCX are illustrated (right panel). With 30 μmol/L (not shown) or 100 μmol/L of XIP in the pipette, INCX was inhibited by 88% in the reverse mode and 73 (30 μmol/L) to 77% (100 μmol/L) in the forward mode, respectively (Figure 2A, right panel).
The left panels of Figure 2 illustrate the positions of the f-loop sequences that were deleted in the respective mutants of the NCX. All deletion mutants expressed INCX. NCX 562– 679 currents had properties similar to wild-type, with outward INCX showing activation during the pulse for depolarizations to high positive potentials (middle panels of Figure 2A and 2D). In contrast, 440–456, 498–510, and 680–685 mutants displayed different current kinetics, with constant or decreasing currents within single pulses at higher voltages (middle panels of Figure 2B, 2C, and 2E).
XIP (100 μmol/L) inhibited INCX in cells transfected with NCX 440–456 (Figures 1B and 2B), 498–510 (Figure 2C), or 680–685 mutants (Figure 2E). In contrast, in cells transfected with 562–679, INCX was not inhibited by XIP (Figure 2D). In Figure 1C, a representative I versus t plot for a 562–679 cell with XIP in the pipette is displayed. Though initially small, INCX continuously increased and reached steady state after approximately 3 minutes. In nontransfected HEK cells, no INCX could be detected (Figure 1D).
In the reverse mode (at +80 mV), INCX was 3.6-fold greater in wild-type compared with 562–679 mutant NCX (Figure 2A and 2D). Similar results were obtained when adenoviral gene transfer was performed in A549 cells (see online Figure S2). Reverse mode INCX was 3.4-fold greater in wild-type (21.0±5.0 pA/pF, n=8) compared with the 562–679 mutant (6.2±0.9 pA/pF, n=18; P<0.001). In contrast to wild-type, INCX in the 562–679 mutant was not inhibited by XIP (online Figure S2).
Because NCX function is regulated by a number of intracellular factors, it was important to study the cardiac NCX in its native environment. Thus, in vitro adenoviral gene transfer of wild-type and 562–679 mutant NCX to adult guinea pig cardiac myocytes was performed. In Figure 3, representative I versus t traces are displayed. With [Ca2+]i buffered to 200 nmol/L, myocytes infected with the wild-type NCX revealed a gradual pulse-dependent increase of INCX under control conditions (Figure 3A). With XIP in the pipette, however, this increase was abolished (Figure 3B). Figure 3C illustrates the mean baseline reverse mode INCX (first pulse of each experiment) compared with the maximum INCX in the respective experiments (see arrows in Figure 3A). A similar result was obtained in noninfected myocytes (Control, INCX increased from 3.4±0.9 to 6.3±0.5 pA/pF, P<0.001; XIP, 2.0±0.8 to 3.8±0.4 pA/pF; P=NS). Myocytes infected with the 562–679 mutant displayed similar activation of INCX compared with wild-type NCX (Figure 3D). However, the presence of XIP did not inhibit the activation of INCX in this mutant (Figure 3E and 3F).
The cumulative I/V plots of all groups are shown in Figure 4A through 4C. Compared with noninfected cells, INCX density at +80 mV was 1.5-fold higher in myocytes infected with wild-type NCX (P<0.01). In contrast, myocytes expressing the 562–679 mutant had slightly lower INCX density compared with noninfected myocytes (P<0.05). Although XIP inhibited INCX in noninfected (Figure 4A) and wild-type infected cells (Figure 4B), myocytes expressing the 562–679 mutant were insensitive to XIP (Figure 4C). The inset in Figure 4C illustrates a "paired pipette" experiment in which a second patch (XIP) could be reestablished after excision of the first patch (control) in the same cell.
The pulse-dependent activation of INCX (Figure 3) suggested that allosteric Ca2+ activation may be present in both wild-type and 562–679–infected myocytes. To analyze this more thoroughly, [Ca2+]i transients elicited by reverse mode INCX were measured in cardiac myocytes without intracellular Ca2+ buffering.22 By blocking L-type Ca2+ channels and the SR and pulsing cells from a holding potential of –100 mV to +80 mV at 2 Hz, reverse mode INCX increased [Ca2+]i from 186±14 to 1444±161 nmol/L (n=46). In Figure 5A, representative recordings of INCX and [Ca2+]i from a myocyte infected with the 562–679 mutant are displayed. At pulse 1, diastolic [Ca2+]i was very low (30 to 50 nmol/L). With increasing [Ca2+]i, INCX also increased dynamically (pulses 4 and 7). In pulse 14, despite continuously rising [Ca2+]i, INCX saturated and even decreased during the late phase of the pulse. In pulse 25, INCX decreased during most of the pulse duration. According to the model of Weber et al,22 we attributed the Ca2+i-induced increase of reverse-mode INCX to allosteric activation by Ca2+i. When plotting instantaneous [Ca2+]i against INCX during these pulses, the data could be fit by the Hill equation (Figure 5B). In noninfected and wild-type infected myocytes, KmCaAct values were 298 and 341 nmol/L, respectively. For the 562–679 mutant, KmCaAct values were shifted only slightly (and statistically insignificant) to 375 nmol/L (Table 2, Figure 5B). In contrast, in 440–456, 498–510, and 680–685 mutant-infected myocytes, allosteric Ca2+ activation of INCX was significantly shifted toward higher [Ca2+]i (710 to 825 nmol/L; Table 2 and Figure 5B). The slope factors of the fits were not different between groups (mean, 4.3±0.2; n=45). Baseline INCX in 440–456 and 680–685 mutants were reduced compared with noninfected myocytes, whereas the maximum INCX was reduced only in 440–456 mutant-infected cells (Figure 5C).
The present study for the first time functionally distinguishes the XIP-regulatory from Ca2+-regulatory domains of the cardiac NCX. Based on mutational analysis of NCX in intact cardiac myocytes, we propose that, in contrast to previous results,18 amino acids 562–679 are not involved in regulation of INCX by Ca2+, but are required for regulation by XIP. In contrast, residues 440–456, 498–510, and 680–685 of the f-loop are involved in regulation of INCX by Ca2+, but not by XIP.
Regulation of NCX by XIP
The NCX is regulated by intracellular Na+ and Ca2+. In excised giant sarcolemmal patches, application of Na+i in the presence of Ca2+i initially activates reverse mode INCX, followed by a process of partial inactivation that reaches steady state within several seconds (Na+-dependent inactivation, I1).27 On the other hand, increasing [Ca2+]i activates INCX, which in the absence of Ca2+i is in the Na+-independent inactive state (I2).
The endogenous XIP region plays an important role in the physiological regulation of INCX by intracellular factors. Na+-dependent inactivation is mediated by the XIP region.14 PIP2 activates INCX by reducing Na+-dependent inactivation,16 presumably by binding to the endogenous XIP region and thus inhibiting the interaction with its putative binding site on the f-loop.15 This binding site, however, is not well characterized. Matsuoka et al18 found that mutants of the NCX lacking residues 240–679 or 562–685 of the f-loop were both insensitive to XIP. They proposed that the overlapping region of these mutants (ie, residues 562–679) may contain the XIP binding site. Conversely, Hale et al28 observed that XIP directly bound to peptides corresponding to residues 445– 455 of the central f-loop. In the present study, the deletion of amino acids 562–679 eliminated XIP-induced inhibition of INCX in heterologous expression systems and cardiac myocytes. In contrast, deletion of residues 440–456 or two other negatively charged regions (498–510 and 680–685) did not affect XIP-regulation of INCX.
In most experimental systems, the inhibitory effect of exogenous XIP saturated between 10 and 30 μmol/L.10,13,17,18,29,30 In the present study, both 30 and 100 μmol/L of XIP inhibited reverse mode INCX by 88% in HEK cells, respectively. In cardiac myocytes, we used 100 μmol/L to achieve maximum inhibition of INCX. The inhibitory efficacy of 100 μmol/L of XIP on INCX was 88% in HEK cells, 82% in A549 cells, but only 40% to 60% in cardiac myocytes.
The reason for this difference is unclear. The cardiac NCX associates with caveolin-3, and caveolin-binding motifs are located in the endogenous XIP region (residues 223–231), but also within the herein identified XIP-regulatory domain (residues 620–627).31 It was suggested that by its association with the endogenous XIP region (or possibly also its regulatory domain), caveolin-3 may activate the NCX by inhibiting the interaction of the endogenous XIP region with its binding site.31 If this holds true, then higher membrane content of caveolin-3 in cardiac myocytes may reduce the inhibitory efficacy of exogenous XIP compared with HEK cells, where caveolin-3 is virtually absent.32 In a similar way, ATP- and PIP2-induced activation of INCX are supposedly attributable to interaction of PIP2 with the endogenous XIP region, thus inhibiting its interaction with the XIP regulatory site.15,16 Because exogenous XIP also avidly binds to PIP2,15 it may be speculated that differences in the native membrane content of PIP2 between cardiac myocytes, HEK, and A549 cells may also account for the different inhibitory efficacy of XIP in these cells. Future studies will be directed to elucidate regulation of NCX by PIP2 and caveolin-3 in its native cellular environment.
The inhibitory efficacy of XIP in whole-cell patch-clamped cardiac myocytes is within the range reported in previous studies.10,30 The results from these studies suggest that [Ca2+]i may have an impact on the efficacy of XIP. In guinea pig30 or canine cardiac myocytes,10 XIP inhibited INCX by 65% when [Ca2+]i was buffered to 100 nmol/L. In contrast, in nonbuffered conditions with [Ca2+]i between 300 and 500 nmol/L, XIP inhibited INCX by only 27%.10 In the present study, [Ca2+]i was buffered to 200 nmol/L, and XIP inhibited INCX by 40% in noninfected guinea pig cardiac myocytes and by 60% in myocytes expressing the canine wild-type NCX. Although the protein sequence of the endogenous XIP region is identical between the guinea pig33 and the canine34 NCX, the XIP-interaction domain (residues 562–679) is 95% identical between the two species. It is unclear whether this contributed to the different inhibitory efficacy of XIP in cardiac myocytes expressing either guinea pig or canine NCX.
Regulation of NCX by Ca2+
In HEK cells, the wild-type and the 562–679 mutant displayed activation of reverse-mode INCX during pulses to higher voltages. This may be attributable to allosteric activation of INCX by Ca[2+]i.22 In contrast, activation of INCX was absent in NCX 440–456, 498–510, or 680–685 under these conditions (Figure 2). In cardiac myocytes with [Ca2+]i buffered to 200 nmol/L, INCX remained unchanged within single pulses; however, the average INCX was monotonically activated within several seconds to minutes. This activation pattern of INCX was similar for wild-type and 562–679 mutant NCX (Figure 3). To quantify allosteric activation of INCX by [Ca2+]i in cardiac myocytes under nonbuffered [Ca2+]i conditions, we used a protocol similar to that of Weber et al,22 who determined Ca2+ activation by averaging INCX during each pulse and observed an increase in mean reverse mode INCX and mean [Ca2+]i from pulse to pulse. Within each pulse, INCX tended to decrease despite rising [Ca2+]i in most of their experiments,22 probably as a result of decreased thermodynamic driving forces for ion transport. In contrast, we observed increasing INCX within each of the first few pulses as well as from pulse to pulse (Figure 5A), as predicted by computational modeling.22
By plotting the instantaneous INCX against [Ca2+]i during each pulse (Figure 5B), we calculated KmCaAct values of 298 and 341 nmol/L for the native guinea pig NCX and the (canine) wild-type NCX expressed in guinea pig myocytes, respectively. These values are in agreement with previous results for the canine, guinea pig, or bovine NCX.20,22,35–37 Nevertheless, there may be species-dependent differences regarding allosteric Ca2+ activation. In human7 and ferret myocytes,22 KmCaAct values were only 150 and 125 nmol/L, respectively. In mouse myocytes, Ca2+ activation was observed in inside-out patches,21 but not in whole-cell patches.22 In the 562–679 mutant NCX, KmCaAct values were not significantly different from wild-type- and noninfected myocytes (375 nmol/L). Even in the presence of XIP, allosteric Ca2+ activation was preserved in this mutant (Figure 3).
Regulation of INCX by [Ca2+]i is complex and involves multiple regions of the NCX molecule.
Biochemical19,38 and electrophysiological studies20 have identified two regions on the central f-loop (residues 440–456 and 498–510) containing aspartate residues that bind Ca2+ with high affinity and in a cooperative manner. Mobility shifts during SDS-PAGE in the presence of Ca2+ suggested that the binding of Ca2+ to these residues induces a conformational change of the NCX molecule.19 Point mutations or deletions of short stretches within these two regions shifted the Ca2+-binding or activation curves from 300 nmol/L to the micromolar range.19,20 Accordingly, in the present study, the deletion of either amino acids 440–456 or 498–510 shifted the KmCaAct values of allosteric Ca2+ activation from 300 nmol/L to 710 and 765 nmol/L, respectively. The slope factors of 4 for Ca2+ activation in our experiments are in agreement with (although not a proof of) the concept of allosteric Ca2+ activation of INCX as a cooperative process.
Similarly, deletion of amino acids 680–685 shifted the Ca2+ activation curves of INCX to 825 nmol/L, confirming that this C-terminal f-loop region plays an important role in Ca2+ regulation.21,22 The functional role of this region, however, is not entirely clear. No direct binding of Ca2+ to these regions could be observed using the 45Ca2+ overlay technique.19 When transgenically overexpressed in mice, excised patch measurements of myocyte blebs indicated that the 680–685 mutant was constitutively activated already at low [Ca2+]i.21 In our experiments, however, allosteric Ca2+ activation was not absent, but rather shifted to higher [Ca2+]i. The reason for this discrepancy is unclear. Different experimental conditions (whole cell versus excised patches of myocyte blebs) may create different microenvironments of NCX regarding membrane-associated regulatory factors (eg, PIP215,16 or caveolin-331). However, the mechanisms of these interactions are presently not fully understood.
Impact of the Native NCX in Cardiac Myocytes After Adenoviral Gene Transfer
In the present study, adenoviral gene transfer of the wild-type and deletion mutants of NCX was performed in guinea pig cardiac myocytes that already express native NCX. It is therefore important to consider how much of INCX is related to native and transferred NCX, respectively. At least two observations indicate that the majority of native NCX appears to have been replaced by the virally-expressed genes, probably as a result of competition between native and mutant mRNAs for the translation machinery. First, in 562–679 mutant-expressing myocytes, INCX was completely insensitive to XIP. If native NCX was still active, at least a small fraction of INCX should have been inhibited by XIP. Second, in myocytes expressing 440–456, 498–510, and 680–685 mutants, the allosteric Ca2+-activation curves were shifted to higher [Ca2+]i, however, without a change in the slope factor. If native NCX had still contributed to INCX, its lower KmCaAct should have decreased the slope factor, and two distinct Ca2+-activation sites should have been observed. Hence, we conclude that native NCX was largely replaced with the mutant NCX.
In cardiac myocytes, wild-type NCX-infected myocytes showed 2-fold higher INCX than 562–679 infected and 1.5-fold higher INCX than noninfected myocytes. It is difficult to differentiate whether this is attributable to overexpression of the canine NCX alone or to coexpression of canine and native (guinea pig) NCX. However, the fact that in HEK and A549 cells, the 562–679 mutant consistently exerted 3.5-fold lower INCX compared with the wild-type NCX (despite different gene transduction techniques) suggests that the deletion mutant may exert intrinsically lower exchange activity. If this is true, it may be speculated that both wild-type and 562–679 NCX proteins were modestly overexpressed compared with noninfected myocytes, with the deletion mutant exerting intrinsically lower INCX.
Conclusions
In summary, we propose that residues 562–679 of the f-loop contain the binding site for the endogenous XIP domain, but are not involved in allosteric Ca2+ regulation of INCX. In contrast, residues 440–456, 498–510, and 680–685 are critical domains for regulation of INCX by Ca2+, but not XIP. The fact that neither deletion mutant used by Matsuoka et al18 was regulated by Ca2+ may reflect deletion of either the central Ca2+-binding domains (residues 440–510, deleted in 240–679) or the C-terminal regulatory domain (residues 680–685, deleted in 562–685).
The 562–679 mutant may represent a novel tool to further investigate regulation of INCX by intracellular factors that possibly involve the interaction of the endogenous XIP region and its binding domain in the environment of the native cardiac myocyte (eg, PIP215,16 and caveolin-331). Furthermore, because selective NCX-inhibitors have been postulated to improve Ca2+ homeostasis in chronic heart failure,10 the development of such drugs may be facilitated by a better understanding of XIP-mediated regulation of the NCX.
References
Kimura J, Miyamae S, Noma A. Identification of sodium-calcium exchange current in single ventricular cells of guinea-pig. J Physiol. 1987; 384: 199–222.
Fujioka Y, Komeda M, Matsuoka S. Stoichiometry of Na+-Ca2+ exchange in inside-out patches excised from guinea-pig ventricular myocytes. J Physiol. 2000; 523: 339–351.
Bers DM. Calcium fluxes involved in control of cardiac myocyte contraction. Circ Res. 2000; 87: 275–281.
Despa S, Islam MA, Weber CR, Pogwizd SM, Bers DM. Intracellular Na+ concentration is elevated in heart failure but Na/K pump function is unchanged. Circulation. 2002; 105: 2543–2548.
Armoundas AA, Hobai IA, Tomaselli GF, Winslow RL, O’Rourke B. Role of sodium-calcium exchanger in modulating the action potential of ventricular myocytes from normal and failing hearts. Circ Res. 2003; 93: 46–53.
Pieske B, Maier LS, Piacentino V, 3rd, Weisser J, Hasenfuss G, Houser S. Rate dependence of [Na+]i and contractility in nonfailing and failing human myocardium. Circulation. 2002; 106: 447–453.
Weber CR, Piacentino V, 3rd, Houser SR, Bers DM. Dynamic regulation of sodium/calcium exchange function in human heart failure. Circulation. 2003; 108: 2224–2229.
Weber CR, Piacentino V, 3rd, Ginsburg KS, Houser SR, Bers DM. Na+-Ca2+ exchange current and submembrane [Ca2+] during the cardiac action potential. Circ Res. 2002; 90: 182–189.
Dipla K, Mattiello JA, Margulies KB, Jeevanandam V, Houser SR. The sarcoplasmic reticulum and the Na+/Ca2+ exchanger both contribute to the Ca2+ transient of failing human ventricular myocytes. Circ Res. 1999; 84: 435–444.
Hobai IA, Maack C, O’Rourke B. Partial inhibition of sodium/calcium exchange restores cellular calcium handling in canine heart failure. Circ Res. 2004; 95: 292–299.
Philipson KD, Nicoll DA. Sodium-calcium exchange: a molecular perspective. Annu Rev Physiol. 2000; 62: 111–133.
Shigekawa M, Iwamoto T. Cardiac Na+-Ca2+ exchange: molecular and pharmacological aspects. Circ Res. 2001; 88: 864–876.
Li Z, Nicoll DA, Collins A, Hilgemann DW, Filoteo AG, Penniston JT, Weiss JN, Tomich JM, Philipson KD. Identification of a peptide inhibitor of the cardiac sarcolemmal Na+-Ca2+ exchanger. J Biol Chem. 1991; 266: 1014–1020.
Matsuoka S, Nicoll DA, He Z, Philipson KD. Regulation of cardiac Na+-Ca2+ exchanger by the endogenous XIP region. J Gen Physiol. 1997; 109: 273–286.
He Z, Feng S, Tong Q, Hilgemann DW, Philipson KD. Interaction of PIP(2) with the XIP region of the cardiac Na/Ca exchanger. Am J Physiol Cell Physiol. 2000; 278: C661–6.
Hilgemann DW, Ball R. Regulation of cardiac Na+,Ca2+ exchange and KATP potassium channels by PIP2. Science. 1996; 273: 956–959.
Kleiboeker SB, Milanick MA, Hale CC. Interactions of the exchange inhibitory peptide with Na-Ca exchange in bovine cardiac sarcolemmal vesicles and ferret red cells. J Biol Chem. 1992; 267: 17836–17841.
Matsuoka S, Nicoll DA, Reilly RF, Hilgemann DW, Philipson KD. Initial localization of regulatory regions of the cardiac sarcolemmal Na+-Ca2+ exchanger. Proc Natl Acad Sci U S A. 1993; 90: 3870–3874.
Levitsky DO, Nicoll DA, Philipson KD. Identification of the high affinity Ca2+-binding domain of the cardiac Na+-Ca2+ exchanger. J Biol Chem. 1994; 269: 22847–22852.
Matsuoka S, Nicoll DA, Hryshko LV, Levitsky DO, Weiss JN, Philipson KD. Regulation of the cardiac Na+-Ca2+ exchanger by Ca2+: mutational analysis of the Ca2+-binding domain. J Gen Physiol. 1995; 105: 403–420.
Maxwell K, Scott J, Omelchenko A, Lukas A, Lu L, Lu Y, Hnatowich M, Philipson KD, Hryshko LV. Functional role of ionic regulation of Na+/Ca2+ exchange assessed in transgenic mouse hearts. Am J Physiol. 1999; 277: H2212–21.
Weber CR, Ginsburg KS, Philipson KD, Shannon TR, Bers DM. Allosteric regulation of Na/Ca exchange current by cytosolic Ca in intact cardiac myocytes. J Gen Physiol. 2001; 117: 119–131.
Patton C, Thompson S, Epel D. Some precaustions in using chelators to buffer metals in biological solutions. Cell Calcium. 2004; 35: 427–431.
O’Rourke B, Kass DA, Tomaselli GF, Kaab S, Tunin R, Marban E. Mechanisms of altered excitation-contraction coupling in canine tachycardia-induced heart failure, I: experimental studies. Circ Res. 1999; 84: 562–570.
Grynkiewicz G, Poenie M, Tsien RY. A new generation of Ca2+ indicators with greatly improved fluorescence properties. J Biol Chem. 1985; 260: 3440–3450.
Bassani JW, Bassani RA, Bers DM. Calibration of indo-1 and resting intracellular [Ca]i in intact rabbit cardiac myocytes. Biophys J. 1995; 68: 1453–1460.
Hilgemann DW. Regulation and deregulation of cardiac Na+-Ca2+ exchange in giant excised sarcolemmal membrane patches. Nature. 1990; 344: 242–245.
Hale CC, Bliler S, Quinn TP, Peletskaya EN. Localization of an exchange inhibitory peptide (XIP) binding site on the cardiac sodium-calcium exchanger. Biochem Biophys Res Commun. 1997; 236: 113–117.
Xu W, Denison H, Hale CC, Gatto C, Milanick MA. Identification of critical positive charges in XIP, the Na/Ca exchange inhibitory peptide. Arch Biochem Biophys. 1997; 341: 273–279.
Chin TK, Spitzer KW, Philipson KD, Bridge JH. The effect of exchanger inhibitory peptide (XIP) on sodium-calcium exchange current in guinea pig ventricular cells. Circ Res. 1993; 72: 497–503.
Bossuyt J, Taylor BE, James-Kracke M, Hale CC. The cardiac sodium-calcium exchanger associates with caveolin-3. Ann NY Acad Sci. 2002; 976: 197–204.
Cha SH, Jung NH, Kim BR, Kim HW, Kwak JO. Evidence for cyclooxygenase-1 association with caveolin-1 and -2 in cultured human embryonic kidney (HEK 293) cells. IUBMB Life. 2004; 56: 221–227.
Tsuruya Y, Bersohn MM, Li Z, Nicoll DA, Philipson KD. Molecular cloning and functional expression of the guinea pig cardiac Na+-Ca2+ exchanger. Biochim Biophys Acta. 1994; 1196: 97–99.
Nicoll DA, Longoni S, Philipson KD. Molecular cloning and functional expression of the cardiac sarcolemmal Na+-Ca2+ exchanger. Science. 1990; 250: 562–565.
Reeves JP, Condrescu M. Allosteric activation of sodium-calcium exchange activity by calcium: persistence at low calcium concentrations. J Gen Physiol. 2003; 122: 621–639.
Trac M, Dyck C, Hnatowich M, Omelchenko A, Hryshko LV. Transport and regulation of the cardiac Na+-Ca2+ exchanger, NCX1. Comparison between Ca2+ and Ba2+ J Gen Physiol. 1997; 109: 361–369.
Hilgemann DW, Matsuoka S, Nagel GA, Collins A. Steady-state and dynamic properties of cardiac sodium-calcium exchange. Sodium-dependent inactivation. J Gen Physiol. 1992; 100: 905–932.
Levitsky DO, Fraysse B, Leoty C, Nicoll DA, Philipson KD. Cooperative interaction between Ca2+ binding sites in the hydrophilic loop of the Na+-Ca2+ exchanger. Mol Cell Biochem. 1996; 160–161:27–32.(Christoph Maack, Anand Ga)