changed or abolished by chymotrypsin treatment, including cytoplasmic Na+-dependent inactivation, secondary regulation by free cytoplasmic Ca2+, and ...
Proc. Natl. Acad. Sci. USA Vol. 90, pp. 3870-3874, May 1993 Biochemistry
Initial localization of regulatory regions of the cardiac sarcolemmal Na+-Ca2+ exchanger (exchanger inhibitory peptide/giant excised patch/oocytes)
SATOSHI MATSUOKA*, DEBORA A. NICOLLt, ROBERT F. REILLYt, DONALD W. HILGEMANN*, AND KENNETH D. PHILIPSONt *Department of Physiology, University of Texas Southwestern Medical Center, Dallas, TX 75235; tDepartments of Medicine and Physiology and the Cardiovascular Research Laboratory, University of California at Los Angeles School of Medicine, Los Angeles, CA 90024-1760; and tDepartment of Internal Medicine, Yale University School of Medicine, New Haven, CT 06510
Communicated by Isidore S. Edelman, December 21, 1992 (received for review August 17, 1992)
We have analyzed the regulatory properties ABSTRACT of the wild-type cardiac Na+-Ca2+ exchanger expressed in Xenopus laevis oocytes using the giant excised patch technique. The exchanger is activated by cytoplasmic application of chymotrypsin and exhibits a number of properties that can be changed or abolished by chymotrypsin treatment, including cytoplasmic Na+-dependent inactivation, secondary regulation by free cytoplasmic Ca2+, and inhibition by exchanger inhibitory peptide. Thus, the cloned exchanger expressed in oocytes exhibits regulatory properties similar to those of the native sarcolemmal exchanger. The exchanger protein contains a large (520 amino acids) hydrophilic domain modeled to be intracellular. The role of this region in exchanger function and regulation was examined by deletion mutagenesis. Mutants with residues 240-679 and 562-685 deleted exhibited exchange activity, indicating that this extensive region is not essential for transport function. Both mutants were stimulated by chymotrypsin treatment. Neither mutant demonstrated regulation by free cytoplasmic Ca2+ (Ca;+) or inhibition by exchanger inhibitory peptide (XIP). However, mutant A562-685 but not A240-679 displayed Na+-dependent inactivation. The data suggest that the binding sites for XIP and regulatory Ca2+ may reside in the region encompassed by residues 562-685. A chimera made from renal and cardiac exchangers has normal regulatory characteristics and helps to further derme these sites.
transmembrane helices (7). Five amino-terminal transmembrane helices are separated from six carboxyl-terminal transmembrane helices by a large hydrophilic segment of 520 amino acids (Fig. 1). To define the role of the large hydophilic loop, we have prepared deletion mutants in which portions of this segment are removed, and we have examined the effect of these deletions on the regulatory properties of the exchanger. Recently, a Na+-Ca2+ exchanger from rabbit kidney has been cloned (9). The deduced amino acid sequence of this exchanger is about 95% identical to the canine cardiac exchanger. The kidney exchanger, however, lacks a 28amino acid segment, and an adjoining segment of 33 amino acids is only 33% identical. This region is located on the large hydrophilic loop. We analyzed the functional significance of these amino acid differences by constructing a chimeric exchanger. We report that the deletion mutants maintain Na+-Ca2+ exchange activity and have modified regulatory characteristics, whereas the chimeric exchanger exhibits wild-type regulatory properties. In addition, the deletion-mutant exchangers have greatly altered sensitivity to XIP (8).
MATERIALS AND METHODS Preparation of Plasmid Giving Enhanced Exchanger Expression. cRNA synthesized from the exchanger clone pTB11 (3) does not always express well in oocytes. To enhance expression, we replaced the 3' untranslated region of pTB11 with that of the Na+/glucose cotransporter clone pMC424 (10). The coding region of pTB11 was removed by digestion with Sal I and SnaBI and ligated into pMC424 from which the coding region had been removed by digestion with Mlu I, treatment with Klenow fragment, and then digestion with Sal I. RNA, synthesized from the resultant plasmid, pMC/SS, using T3 RNA polymerase, reproducibly induces high-level expression of Na+-Ca2+ exchange activity in Xenopus oocytes. All manipulations of nucleic acids were performed as described (11). Preparation of Deletion Mutants and Chimera. Bases 8402159 were removed from pMC/SS by restriction digestion with Bcl I and gel purification of the fragment containing pBluescript plus the 5' and 3' ends of the cDNA insert. The resultant plasmid was recircularized with ligase. This plasmid, pA240-679, codes for the Na+-Ca2+ exchanger with amino acids 240-679 deleted (see Fig. 1). To generate mutant A562685, residues 1805-2180 were removed by partial digestion
The Na+-Ca2+ exchanger of the cardiac sarcolemmal membrane is a highly active transporter that mediates the countermovement of three Na+ ions for one Ca2+ ion. The Na+-Ca2+ exchanger is responsible for movement of substantial amounts of Ca2+ across the cardiac sarcolemma during excitation-contraction coupling processes (1, 2). The cardiac exchanger has been cloned and expressed in Xenopus laevis oocytes (3). Activity of the expressed exchanger can be measured directly by following 45Ca2+ fluxes (3) or electrophysiologically by using the giant excised patch technique (4-6). Regulatory properties of the Na+-Ca2+ exchanger have previously been characterized using giant excised patches from myocytes (4). These properties include activation by Ca2+ (secondary Ca2+ regulation), modulation by ATP, and a cytoplasmic Na+ (Na+)-dependent inactivation process. All forms of regulation can be removed by treating the cytoplasmic surface of the excised patch with chymotrypsin. Accordingly, the regulatory properties of the exchanger are predicted to be mediated by cytoplasmic domains of the protein. By hydropathy analysis and in vitro translation analysis, the mature protein of 938 amino acids is modeled to have 11
Abbreviations: XIP, exchanger inhibitory peptide; Mes, 2-(Nmorpholino)ethanesulfonic acid; NMG, N-methylglucamine; TEAOH, tetraethylammonium hydroxide; Kh, half-maximal concentration; nh, Hill coefficient; CaF', free cytoplasmic Ca2+; Na+, free cytoplasmic Na+; I-V, current-voltage.
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact. 3870
Biochemistry: Matsuoka et al.
Proc. Natl. Acad. Sci. USA 90 (1993)
0
678 9 1011
ID 123E 45 01
2345
mm mr_
'-^
240-679 _ 562-685
B
Chimera
Ca Regulation, XIP Interaction Na Inactivation Ion
C
Binding and Transport
FIG. 1. (A) Linear representation of the Na+-Ca2+ exchanger. Putative transmembrane segments 1-11 are indicated by dark crosshatching. The endogenous exchanger inhibitory peptide (XIP) site (8) is stippled. (B) Deleted regions ofthe exchanger. For the chimera, the region corresponding to residues 570-644 is indicated. (C) Summary of functionally important regions of the exchanger. Amino acids are numbered taking into account the cleaved leader peptide (7) and not as initially published (3).
with EcoRI, and the 5.8-kb fragment was ligated to form the deleted mutant. To generate the cardiac/kidney chimera, the EcoRI (at position 1777)-Apa I (at position 2719) fragment of the kidney exchanger clone was subcloned into the EcoRI (at position 1805)-Apa I (at position 2836) site. This substituted amino acid residues 593-907 of the kidney exchanger for residues 561-904 of the cardiac exchanger and introduced the region of the kidney exchanger containing significant differences (amino acids 570-644 ofthe cardiac exchanger) between the two exchangers. In addition to this region ofdifference, the chimera also had Leu substituted for Ile-776 and His substituted for Gln-864 of the cardiac exchanger. Measurement of Na+_Ca2+ Exchange Activity. RNA (5 ng) was injected into oocytes and exchange activity was measured 3-5 days later. 45Ca2+ fluxes were measured as described (3). Exchange currents were measured by the giant membrane patch technique (6). The oocyte vitellin layer was removed by dissection after shrinking the cells in a hypertonic solution [65 mM KOH/50 mM 2-(N-morpholino)ethanesulfonic acid (Mes)/9 mM Mg(OH)2/5 mM EGTA/10 mM Hepes/10 mM glucose/250 mM sucrose, adjusted to pH 7.0 with Mes]. Glass pipettes were prepared with inner diameters of 20-35 ,Am and a bullet-like shape. Pipettes were coated with a Parafilm/heavy mineral oil mixture (6), modified for oocyte experiments by addition of 1-5% by weight of n-decane (Sigma) or light white oil (Sigma). The pipette (extracellular) solution contained 100 mM N-methylglucamine (NMG), 100 mM Mes, 8 mM CaCO3, 2 mM Mg(OH)2, 30 mM Hepes, 30 mM tetraethylammonium hydroxide (TEA-OH), 2 mM Ba(OH)2, 0.1 mM (each) niflumic and flufenamic acids (Sigma), and 0.25 mM ouabain (pH adjusted to 7.0 with Mes). Mg(OH)2 and CaCO3 were prepared as 0.5 M stock solutions in sulfamic acid. The bath (cytoplasmic) solutions contained 100 mM LiOH or NaOH, 100 mM Mes, 1 mM Mg(OH)2, 30 mM Hepes, 30 mM TEA-OH, and 10 mM EGTA (pH 7.0 with Mes). Ca2+containing solutions were made by addition of appropriate amounts of CaCO3. Xenopus oocyte membrane contains a large Caj+-activated Cl- conductance (12) that must be rigorously eliminated. Preliminary findings of a Caj+-sensitive outward exchange current with the A240-679 mutant (4) were found to be contaminated by this conductance. For the present study, we used the buffer Mes, Cl--free solutions, and the Cl- channel blockers niflumic and flufenamic acids (13) to effectively eliminate all Ca2+-activated Cl- current. Seals were made directly on shrunken oocytes. Seal formation was usually difficult in the complete absence of C1ions, and two protocols overcame this problem. First, a
3871
pipette solution could be employed for sealing in which CaCl2 and MgCl2 were used instead of CaCO3 and Mg(OH)2. After seal formation, the solution in the pipette was replaced with the Cl--free solution by the use of an intrapipette perfusion device (14). Alternatively, seals were made in a bath solution containing 10-40 mM Cl- with the Cl--free pipette solution. In both cases, a 3 M KCl-agar bridge was used in the perfusion device or pipette. Current-voltage (I-V) relations were measured by changing potential in 20-mV steps, of 20-ms duration, from 0 to -120 mV, then from -120 to +60 mV, and finally from +60 to 0 mV. Current values were determined at the end of each step, and the I-V relation of the outward current was determined by subtracting currents in the absence and presence of Na+. Experiments were carried out at 30°C.
RESULTS The Wild-Type Na+-Ca2+ Exchanger. Fig. 2A shows outward Na+-Ca2+ exchange currents of an excised inside-out patch from an oocyte injected with wild-type Na+-Ca2+ exchanger cRNA. In the presence of 8 mM Ca2+ in the pipette ("extracellular"), a rapid replacement of 100 mM Li+ with equimolar Na+ in a bathing solution ("cytoplasmic") that contains 1 AM free Ca2+ induces an outward current (arrow a). The current partially inactivates over a period of several seconds. As detailed in patches from cardiac myocytes (4), the inactivation is a specific response to exchanger activation by cytoplasmic Na+ and is termed Na,+-dependent inactivation. The exchanger is also regulated by Ca2+. When Ca2+ is removed from the bath (arrow b), exchange current declines toward baseline. A transient component of the outward exchange current was typically insensitive to removal of Ca?+ from the oocyte membrane (Fig. 2B). Similar Ca2+-independent currents were observed in myocyte membranes under certain conditions (15). The data demonstrate that Na,+-dependent inactivation can be seen in the absence of regulatory Ca2+. The magnitude of exchange currents varied about 3-fold in different batches of oocytes (Fig. 2 A and B). Stimulatory effects of MgATP were either small or absent in the oocyte membrane (not shown), consistent with the idea that additional proteins are involved in the MgATP mechanism (6). The I-V relations of the exchanger current at peak (a-e) or steady state (b-e, d-e) and in the absence of Ca2+ (c-e) are shown in Fig. 2D. The voltage dependence is very similar for the three cases. Currents are simply scaled down in the steady state as compared to the transient peak. The current increases 2.7-fold in about 190 mV, similar to results with myocyte membranes under the same conditions (5). Proteolysis of the cytoplasmic surface of sarcolemma with a-chymotrypsin increases the exchange current and deregulates the exchanger (16). The oocyte-expressed exchanger responds in a similar manner (Fig. 2C). The outward exchange current is increased, and the current transient and the requirement for intracellular Ca2+ are eliminated. Figs. 2 E and F show the Na+ and Ca2+ dependencies of the outward exchange current. The half-maximal concentration (Kh) for Na+ is 18.3 mM, and the Hill coefficient is 2.7. The Kh for regulatory Ca2+ is 3.8 ,uM. The values are similar to results with cardiac myocytes (6). Mutant A240-679. Although most of the hydrophilic domain is deleted in mutant A240-679 (Fig. 1), this mutant is still capable of Na+-Ca2+ exchange activity (Fig. 3). Activity is decreased compared to the wild-type exchanger. Maximal currents were typically about 10% of those for the wild-type exchanger. When the oocyte Na+-Ca2+ exchange activity was quantitated by measuring Na+-dependent 45Ca2+ fluxes, the activity in oocytes expressing A240-679 was 19% ± 4% (n = 10) of that in oocytes expressing the wild-type ex-
3872
Biochemistry: Matsuoka et al.
Proc. Natl. Acad. Sci. USA 90 (1993)
A 150PA a I
100
B
b
c I
I
I
I a
1 min
\
C
d e pA 125 1 min
10 pA 30 s
CHYMO.
+Ca 2+ Ca2+ i
50
.FL~~
0 -50 1lpM Ca,
1
L
D
0
0mMNaU9L 1
E 1
.
%M
,
1.0
F
-
1.0
100 b-e d-e
0.5
.' r,~c-e
o
,9,#
0.5
'- 50
100 mV
0
-100
0
0 0
100
50
80
40 b)2+ (PM)
0
C:a, Ca i
Na+(mM) Na+ (mM)
FIG. 2. Outward Na+-Ca2+ exchange current from the wild-type exchanger in a giant inside-out patch of oocyte membrane. (A) Current trace at 0 mV. The current was evoked by rapid replacement of Li:' with Na,+ at the cytoplasmic (bath) surface. Na,+ and Ca?+ concentrations are shown below the current trace. I-V relations were measured at the points indicated by arrows. (B) Typical current transient in the presence and absence of (C) Activation of the exchange current with 2 mg of a-chymotrypsin (CHYMO.) per ml. (D) I-V relations of the exchange current at the peak (a-e), at steady state (b-e, d-e), and in the absence of Ca?+ (c-e). Background currents (e) were subtracted from exchange currents in each case, as indicated. (E) Na,+ dependence of the exchange current. Current amplitudes at steady state were normalized to the amplitude at 100 mM Na,+ and fit to the Hill equation [half-maximal concentration (Kh) = 18.3 mM; Hill coefficient (nh) = 2.7]. Na+ was replaced with Li+ (o) or Cs+ (-). (F) Secondary Ca2+ dependence of the outward current. Steady-state current amplitudes were normalized with a maximal value (at 80 mM) and fit to a rectangular hyperbola. Kh of Ca2+ is 3.8 ,uM.
Ca?+.
changer. It was not possible to determine if this difference was due to an inherent difference in the two exchangers or to a difference in the amount of exchanger protein that was synthesized since the mutant exchanger does not react with antibodies to the wild-type exchanger on immunoblots. Striking differences between the wild-type and mutant A240-679 exchangers exist. There is no Na,+-dependent inactivation of the mutant exchanger as seen by the lack of a current transient when Na+ is applied to the bath (Fig. 3). The mutant exchanger is also not modulated by Cah'. Removal of Ca2+ from the bathing solution does not affect exchanger 12 r pA
currents. The mutant exchanger is, however, still stimulated by chymotrypsin treatment, indicating that chymotrypsin stimulation can occur in the absence of the Na,+-dependent inactivation and Ca2+-regulation. The Kh (15.8 mM) and Hill coefficient (1.9) for Na1+ and the voltage dependence were similar to those for the wild-type exchanger (not shown). Mutant A562-685. Levels of activity in this mutant, containing a smaller deletion (Fig. 1), were comparable to that of the wild-type exchanger (Fig. 4). Likewise, the apparent
B
A
CHYMO.
30 sec
CHYMO.
1 min
nA
8P
m-
nA 1 min
0.2 0.1
Z
Cal
0.0 J76
0E
100rmMNa, 0
L
-1 ,uM Cao
1
A
0O900LmM
Na
i
~~~~~~2+
~~~~~~~~1 ,UM Ca,i
FIG. 3. Outward exchange current from mutant A240-679. Exchange current at 0 mV. (Left) Presence or absence of 100 mM Naj+, 1 ,M Ca7+. (Right) Activation by 2 mg of a-chymotrypsin (CHYMO.) per ml.
FIG. 4. Exchange current from mutant A562-685. (A) Outward exchange current at 0 mV, in the presence of 90 mM Nai+ and 1 /AM Ca?'. (B) Left, currents activated by Na,+ with or without (dotted trace) Ca?' are superimposed. Right, current after treatment with 2 mg of a-chymotrypsin (CHYMO.) per ml.
Biochemistry: Matsuoka et al. affinity (Kh = 27.2 mM) and nh (2.2) for Na,+ and the voltage dependence of the mutant exchanger were similar to the wild-type exchanger (not shown). Mutant A562-685, like the wild-type exchanger, displayed Na+-dependent inactivation as seen by the current transient when Na+ was first applied (Fig. 4A). Like the wild-type exchanger, the Nai+-dependent inactivation was abolished by chymotrypsin treatment (Fig. 4B). Thus, residues 562-685 are not involved in Na,+-dependent inactivation. Deletion mutant A562-685 was not regulated by Caj2+. Removing Ca2+ from the bath had no effect on exchanger current (Fig. 4A). Also, Car- is not required to initiate exchanger current as the current traces seen upon applying Na+ to the bath are identical in the presence or absence of Car+ (Fig. 4B). These data suggest that residues 562-685 are involved in regulation by Ca2+. Inhibition by XIP. The outward exchange current of the regulated, wild-type exchanger is completely blocked by XIP, with an ICo of 0.15 ,uM (Fig. 5). The potency of XIP diminishes =10-fold after deregulation by chymotrypsin, and complete inhibition is no longer achieved. The deletion mutants, in contrast, are insensitive to XIP at concentrations up to 20 ,uM. Cardiac/Kidney Chimera. The kidney exchanger has recently been cloned (9) but not yet expressed. The differences between the cardiac and kidney exchangers are localized to the cleaved leader peptide region and to residues 570-644 (Fig. 1). In the kidney exchanger, residues 570-602 are only 33% identical to the cardiac exchanger and the kidney exchanger does not contain residues equivalent to residues 617-644 of the cardiac exchanger. We constructed a cardiac/ kidney chimeric exchanger to determine if the altered regions of the renal exchanger were involved in regulation. The chimeric exchanger expressed in oocytes displayed the wild-type characteristics of Na,+-dependent inactivation, Car+ regulation (Fig. 6), and inhibition by XIP (not shown). a-Chymotrypsin activated the chimeric exchanger and removed Na,+-dependent inactivation and secondary Ca2+ dependence (not shown).
DISCUSSION The cloned exchanger, expressed in oocytes, displays the same regulatory properties as the native exchanger of myoA 240-679 \ 100 I 562-685' \ ~~~~~A 75 F _o -
50
F
_Z
Chymotrysn-
treated wild-type 25
Wild-type \ 0
0
3873
Proc. Natl. Acad. Sci. USA 90 (1993)
10 XIP (pM)
FIG. 5. XIP inhibition of wild-type and mutant exchangers: relative amplitude of the exchange current, activated by 100 mM Naj+ with 1 ,uM Car-, versus XIP concentration. Currents from wild-type (A) and chymotrypsin-treated wild-type (v) exchangers. IC"0 values were about 0.15 and 3.3 ,uM, and maximal inhibitions were 100%t and 61%, respectively, as determined in other experiments (not shown). XIP had no effect on deletion mutant A240-679 (o) or A562-685 (o). Data were fit to a rectangular hyperbola.
80 r pA
0 2+ 0 pM Ca.i
1,uM Ca
i
30 sec
60 F 40 F
20
0
L
11 0
1l00- 0-
0 100 mM
Na.
0
FIG. 6. Exchange current from cardiac/kidney chimera. The current evoked by Na,+ in the absence (Left) and presence (Right) of
Ca?+.
cytes (16). These properties include Na,+-dependent inactivation, regulation by Car+, and inhibition by XIP. Cloned and native exchangers are stimulated by chymotrypsin, which also removes all forms of regulation. The similarities between the myocyte and oocyte-expressed exchangers suggest that the observed properties are inherent in the exchanger protein itself and not to interactions with other myocyte proteins. To determine which portions of the exchanger protein are involved in the different aspects of exchanger function, we constructed two deletion mutants, A562-685 and A240-679, and a cardiac/kidney chimera. These modified exchangers have deletions and/or alterations in the large hydrophilic domain between transmembrane segments 5 and 6. A crude structure-function map of the exchanger protein, based on the results and discussed below, is included in Fig. 1. Ion Binding and Translocation Sites. All three modified exchangers are capable of exchange activity. Therefore, the ion binding and transport domain of the exchanger must be contained in the amino and carboxyl termini. These regions contain all 11 putative transmembrane segments. Na1+-Dependent Inactivation. The wild-type and each of the modified exchangers except A240-679 display a Na1+dependent inactivation. In mutant A562-685, Na,+-dependent inactivation but not Car--regulation is seen. These two traits are apparently not linked. In mutant A240-679 stimulation by chymotrypsin but no Na,+-dependent inactivation is seen, suggesting that chymotrypsin stimulation and inactivation are also distinct. We conclude that residues located between amino acids 240 and 561 are involved in Na,+-dependent inactivation. Ca?' Regulation. Mutants A240-679 and A562-685 lack Car- regulation. Since amino acids 562 to 679 were deleted in both mutants, residues in this region may be involved in regulation by Car-. The cardiac/kidney chimera allows further dissection of the Car+ regulatory region. The chimera, which does not contain residues 617-644, is still regulated by Car-, thus ruling out a role for these residues. Likewise, residues 570-602, which are only 33% identical to the cardiac exchanger, may not be involved in Car+ regulation, although this region in the chimeric and cardiac exchangers may still have a similar secondary structure to provide the same function. By this further elimination, the most likely residues to be involved in Car- regulation are 562-569, 603-616, and 645-679. The Car- regulatory region may be directly involved in binding Ca2+ or indirectly involved, for example, in conformational changes elicited by the binding of Ca2+ elsewhere in the protein. Ifthe region directly binds Ca2+, then the binding site is not of the well-defined EF-hand type since this region does not contain an EF-hand consensus sequence (17). However, the region contains a number of acidic residues that might contribute to a Ca2+-binding site. Altematively, a
3874
Biochemistry: Matsuoka et al.
potential EF-hand Ca2+ binding site is located about 100 residues away at positions 498-509 (7). If this is the Ca2+ binding site, perhaps the region we have identified interacts with the binding site to effect Ca2+ regulation. XIP Inhibition. The potency and completeness of inhibition by XIP for the expressed wild-type exchanger are much greater when the exchanger is in the regulated state than following chymotrypsin digestion (Fig. 5). This result explains a discrepancy between XIP effects on sarcolemmal vesicles and excised patches of sarcolemmal membrane (8). Maximal inhibition by XIP in vesicles was 75% with an IC50 of 1.5 ,uM and in patches was 100% with an IC50 of 0. 15 ,uM. If the Na+-Ca2+ exchanger in sarcolemmal vesicles is deregulated, those data are consistent with the present findings. Although the exchange activity of sarcolemmal vesicles can be stimulated by chymotrypsin (18), high-affinity Ca2+ regulation has never been observed, consistent with a deregulated state. XIP sensitivity, like Ca?-+ regulation, is present in the wild-type and chimeric exchangers but is not present in the deletion mutants. Therefore, a XIP interaction region can also be tentatively localized to residues 562-569, 603-616, and 645-679. XIP, a basic peptide, may interact with acidic residues in these regions. The amino acid sequence of XIP corresponds to amino acid residues 219-238, a potential autoinhibitory region (8). If residues 219-238 of the exchanger (the "endogenous" XIP) do function as an autoinhibitory region, then the large hydrophilic domain may fold so that the XIP interaction region comes into contact with the endogenous XIP region. We have tentatively identified functionally important regions of the Na+-Ca2+ exchanger. However, the various defined regions may constitute only part of a functional unit. For example, the XIP interaction site may consist of the region defined in addition to other segments of the exchanger. Alternatively, the deletions may create gross structural changes that have long-range effects. The fact that the modified exchangers are still capable of ion transport argues
against this possibility.
Proc. Natl. Acad Sci. USA 90 (1993) We are grateful to Drs. C. Gunderson and E. Wright for providing oocytes and to Dr. E. Wright for providing the plasmid pMC424. This work was supported by National Institutes of Health Grants RO1HL27821 (K.D.P.) and R29-HL45240 (D.W.H.), by an American Heart Association Grant-in-Aid (D.W.H.), by the American Heart Association, Greater Los Angeles Affiliate (D.A.N.), and by the Laubisch Foundation. D.W.H. is an Established Investigator of the American Heart Association. 1. Hilgemann, D. W. (1987) J. Gen. Physiol. 87, 675-706. 2. Bridge, J. H. B., Smolley, J. R. & Spitzer, K. W. (1990) Science 248, 376-378. 3. Nicoll, D. A., Longoni, S. & Philipson, K. D. (1990) Science 250, 562-565. 4. Hilgemann, D. W., Collins, A., Cash, D. P. & Nagel, G. A.
(1991) Ann. N. Y. Acad. Sci. 639, 126-139. 5. Hilgemann, D. W., Nicoll, D. A. & Philipson, K. D. (1991) Nature (London) 352, 715-718. 6. Collins, A., Somlyo, A. & Hilgemann, D. W. (1992) J. Physiol. (London) 454, 25-57. 7. Nicoll, D. A. & Philipson, K. D. (1991) Ann. N. Y. Acad. Sci. 639, 181-188. 8. Li, Z., Nicoll, D. A., Collins, A., Hilgemann, D. W., Filoteo, A. G., Penniston, J. T., Weiss, J. N., Tomich, J. M. & Philipson, K. D. (1991) J. Biol. Chem. 266, 1014-1020. 9. Reilly, R. F. & Shugrue, C. A. (1992) Am. J. Physiol. 262, F1105-F1109. 10. Hediger, M. A., Coady, M. J., Ikeda, T. S. & Wright, E. M. (1987) Nature (London) 330, 379-381. 11. Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual (Cold Spring Harbor Lab., Plainview, NY), 2nd Ed. 12. Miledi, R. & Parker, I. (1984) J. Physiol. (London) 357, 173-183. 13. White, M. H. & Aylwin, M. (1990) Mol. Pharmacol. 37, 720-724. 14. Soejima, M. & Noma, A. (1984) Pflugers Arch. 400, 424-431. 15. Hilgemann, D. W. & Collins, A. (1992) J. Physiol. (London) 454, 59-82. 16. Hilgemann, D. W. (1990) Nature (London) 344, 242-245. 17. Marsden, B. J., Shaw, G. S. & Sykes, B. D. (1990) Biochem. Cell Biol. 68, 587-601. 18. Philipson, K. D. & Nishimoto, A. Y. (1982) Am. J. Physiol.
243, C191-C195.
