* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Print - Circulation
Survey
Document related concepts
Magnesium transporter wikipedia , lookup
Cell encapsulation wikipedia , lookup
SNARE (protein) wikipedia , lookup
Organ-on-a-chip wikipedia , lookup
P-type ATPase wikipedia , lookup
Chemical synapse wikipedia , lookup
Cytokinesis wikipedia , lookup
Cell membrane wikipedia , lookup
Action potential wikipedia , lookup
Mechanosensitive channels wikipedia , lookup
Endomembrane system wikipedia , lookup
List of types of proteins wikipedia , lookup
Signal transduction wikipedia , lookup
Transcript
1806 FResearch Advances Series Intracellular Calcium Homeostasis in Cardiac Myocytes William H. Barry, MD, and John H.B. Bridge, PhD Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017 C alcium homeostasis in cardiac myocytes is of functional importance for at least three reasons. First, cardiac myocytes must achieve a resting cytosolic calcium ion concentration ([Ca2+]) of <200 nmol/L if the contractile elements are to relax. With extracellular [Ca2"] at 1 mmol/L, this low [Ca2"] must be maintained in the presence of a 5,000-fold gradient for Ca2' across the sarcolemma. Second, coupling of excitation to contraction (E-C coupling) in heart involves a complex interaction of membrane electric events mediated by specific ion channels in the sarcolemma. This results in calcium influx, the release of calcium from intracellular stores in the sarcoplasmic reticulum (SR) via Ca2+-specific channels in the SR membrane, and subsequent extrusion of the calcium. This in turn is coordinated with reuptake of calcium into the SR stores. A fundamental principle of this process is that to maintain steady-state calcium homeostasis, the amount of calcium entering the cell with each contraction must be extruded before the subsequent contraction. Third, the force of contraction in cardiac myocytes is modulated by variations in the magnitude of the calcium transient. Drugs that modify calcium homeostasis may significantly alter the force of contraction of an individual myocyte and thus of the intact heart. In this discussion, we review recent work that has increased our understanding of the structure and function of the ion channels and transport proteins that are most critically involved in [Ca 2+]i homeostasis. We consider how they interact during the processes of E-C coupling, contraction, and relaxation and briefly discuss how a number of positively inotropic drugs produce an increase in force. The reader is also referred to several other recent reviews that deal with these issues.1-3 Control of Resting [Ca2"Ji Various ion channels and transport proteins involved in calcium homeostasis in the cardiac myocyte are shown schematically in Figure 1. In the resting ventricular myocyte (Figure 1A), the [Ca2+]i is determined by a Ca2+ leak that is compensated for by an ATP-dependent sarcolemmal Ca 2+ pump (sarcolemmal Ca2+-ATPase) and the sarcolemmal Na+-Ca2' exchanger. This exchanger has recently been cloned.4 The complementary DNA encodes From the Division of Cardiology and the Nora Eccles Harrison Cardiovascular Research and Training Institute, University of Utah School of Medicine, Salt Lake City. Address for correspondence: William H. Barry, MD, Division of Cardiology, University of Utah Medical Center, 50 N. Medical Drive, Salt Lake City, UT 84132. Received June 19, 1992; revision accepted February 26, 1993. a protein of 970 amino acids with a molecular mass of 108 kd. The native exchanger protein has a maximal apparent molecular mass of about 160 kd, the discrepancy possibly a result of glycosylation of the native protein. Hydropathy analysis indicates that the protein can be divided into three regions: a hydrophobic NH2 terminal portion containing six potential membrane-spanning segments; a long hydrophilic region that is modeled as a large cytoplasmic loop; and a hydrophobic COOH terminal portion comprising six potential membrane-spanning segments. At this time, little is known about relations between structure and function of this molecule. However, deletion mutants in which the putative cytoplasmic loop has been removed have been expressed in Xenopus oocytes. Whereas the native molecules have a requirement for activating Ca2 ,5 these mutant exchangers lack this requirement (K Philipson, personal communication). Some information is also available on the distribution of this exchanger. Recent experiments investigating the binding of antibodies to the Na+-Ca2' exchanger with immunofluorescence and immunoelectron microscopy indicate that although the exchanger is detectable over the entire myocyte surface, antibody binding sites appear to be concentrated in the T tubule region of the myocyte.6 The functional significance of this arrangement is not yet clear. The exchanger is a counterion transporter on which three Na+ are exchanged for each Ca2+.7 This results in positive charge movement that opposes the direction of calcium transport. The net electrochemical force producing the exchange is given by the following equation: A=3A/Na A/lCa (1) where A is the total electrochemical driving force (free energy), 'A/.Na is the free energy available in the electrochemical gradient for Na+, and 1A/lCa is the free energy in the electrochemical gradient for Ca2`. Thus, if three times the energy in the inward Na+ gradient exceeds that in the Ca2' gradient, Ca2' will be extruded. If, on the other hand, the Na+ gradient is collapsed relative to the Ca2' gradient, Ca2' entry and Na+ extrusion takes place. Equation 1 can be stated in a more familiar form as A=3ENa-2ECa-Em (2) where Em is membrane potential and EN. and ECa are the equilibrium potentials for Na+ and Ca2`. If the exchange is at equilibrium, A=O, and we obtain Eeq =3ENa-c2Ea (3) The reversal potential is the membrane potential at which exchange changes direction, and in the case of the Barry and Bridge Calcium Homeostasis in Myocytes Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017 Na+-Ca ` exchange, it is identical to the equilibrium potential, Eeq (around -40 mV) in resting ventricular cells. The Na+-Ca`+ exchanger is known to be voltage sensitive.8 If the membrane potential is more negative than the reversal potential, as is normally the case in the resting myocyte, the exchanger functions in a Na+-in/ Ca`+-out ("forward") mode and thus produces calcium extrusion. The importance of the Na+-Ca`+ exchanger in regulating the resting level of calcium within cardiac myocytes was demonstrated most directly by the experiments of Sheu and Fozzard,9 who measured intracellular sodium and calcium activities in cardiac Purkinje cells with ion-sensitive microelectrodes. These experiments revealed a direct relation between [Na+]i and [Ca'+],. Thus, increases in [Na+]i cause dissipation of the electrochemical energy in the Na' gradient available to extrude Ca'+, resulting in an increase in [Ca`+]j. The level of [Na+]l in myocytes is controlled largely by Na+,K+-ATPase, and thus the Na' pump is also an important indirect regulator of Na+-Ca2` exchange. The activity of the Na+-Ca`+ exchanger can also be modulated by a variety of other mechanisms. For example, the exchanger is inhibited by acidosis10 and by severe ATP depletion.1' It appears likely that ATP increases the affinity of the exchanger for Ca`+,12 but it has not yet been demonstrated that protein kinasemediated phosphorylation of the exchanger has any regulatory influence. Calmodulin is a 148-amino acid regulatory protein involved in many Ca 2-dependent signaling pathways.13 In some transport proteins, including the sarcolemmal calcium ATPase,14 there is an autoinhibitory calmodulin-binding domain. Calmodulin can increase the activity of the protein by binding to this amino acid sequence, presumably interfering with the interaction of this domain with other components of the molecule and thus relieving the autoinhibition. Recently, Li et al'5 have synthesized a Na+-Ca2 exchanger inhibitory peptide. This molecule is identical to the calmodulin-binding sequence of the exchanger and effectively inhibits exchange.'5 It is not established, however, that this calmodulin-binding peptide sequence within the exchanger exerts any autoinhibitory effect under normal physiological conditions. Another Ca2 transport system that contributes to maintenance of the low cytosolic concentration in ventricular myocytes is the sarcolemmal calcium ATPase. The structure and function of plasma membrane calcium ATPases have recently been reviewed by Carafoli.16 The plasmalemmal calcium ATPases are found in most cells and use the free energy released by the hydrolysis of one ATP to transport one Ca2 out of the cell against its concentration gradient. General properties of the plasma membrane calcium ATPases include a molecular mass of 134,000 kd; a Km for Ca2 of <0.5 gmol/L in the optimally activated state; and as mentioned above, stimulation by Ca2 calmodulin and possibly kinase-induced phosphorylation. In cultured cardiac myocytes, the rate at which the sarcolemmal calcium ATPase can extrude the Ca2 from the myocyte appears to be about 1/10 that of the Na+-Ca2' exchanger over the range of physiological cytosolic calcium concentrations.'7 Cannell18 has estimated that the Na+-Ca'+ exchanger accounts for as much as 75% of resting Ca'+ efflux. Thus, although the sarcolemmal calcium ATPase probably does contribute to mainte- 1807 nance of [Ca'+], in resting myocytes, its importance in this regard relative to the Na+-Ca'+ exchanger seems to be minor. The intracellular Ca'+ storage sites, such as mitochondria,'9,20 as well as calcium-binding proteins2' can buffer short-term (seconds to minutes) changes in calcium. Steady-state loading of these storage sites is determined by the [Ca'+]j, which in turn is regulated largely by transsarcolemmal calcium entry and extrusion by Na+-Ca'+ exchange and perhaps to a lesser extent by the sarcolemmal calcium ATPase, as discussed above. Excitation-Contraction Coupling In cardiac myocytes, the transition from the resting relaxed state with low [Ca'+], to a contraction occurs because a small quantity of Ca'+ crosses the sarcolemma and induces a much larger release of Ca2+ from the SR. The initial events that couple excitation to contraction are displayed in Figure 1B. This process is initiated by depolarization of the cell membrane, which causes opening of the voltage-gated Na+ and Ca'+ channels. The initial upstroke of the action potential in ventricular cardiac myocytes is caused by sodium influx via the Na+ channel, whereas the subsequent inward current maintaining the plateau of the action potential is caused primarily by Ca'+ influx via the L-type Ca'+ channel. Cumulative Na+ influx via the Na+ channel also contributes to maintenance of the intracellular sodium concentration in myocytes and thus can influence [Ca'+]i via the Na+-Ca2' exchanger. However, it is Ca2 influx via the L-type Ca2 channel that is considered to be of greatest importance in E-C coupling.22-25 An important pharmacological characteristic of Ca2 channels is that they contain a high-affinity receptor for 1,4-dihydropyridine ligands. These ligands can function either as antagonists or agonists. Antagonists such as nifedipine, when bound to channels, effectively block their activity. Bay K 8644 is a well-known dihydropyridine receptor agonist and increases Ca2+ current carried by cardiac Ca2+ channels.26 Because dihydropyridine receptors are found in extremely high density in skeletal muscle T tubules,27 Ca 2+ channels have been isolated from this source. Most of our present knowledge of these channels is based on studies of the skeletal muscle channel. The skeletal muscle channel differs functionally from the heart channel in that it is coupled to the SR Ca2+ release channel by a voltage sensor mechanism.28 However, it is believed that the architecture of the two channels is essentially similar.29 The L-type calcium channel is an oligomeric complex of five subunits. These subunits are designated a,, a2, 13, , and &. The a2 and 8 are linked by disulfide bonds and are encoded by the same gene.29,30 As such, they are designated a2/8. The a, subunit seems to provide the primary structural and functional basis for the assembled channel. For example, it contains the receptor for at least three classes of channel antagonists, including the dihydropyridines; it contains phosphorylation sites for cyclic AMP (cAMP)-dependent protein kinase; and it is known to contain the pore of the channel. It can transmit modest quantities of current and apparently retains the voltage sensitivity of the native channel. The a, subunit from heart is somewhat larger than the skeletal muscle subunit.3' The relation between subunit organization and function is now under investigation. It appears that the 1808 Circulation Vol 87, No 6 June 1993 A T TUBULE ATPas 2I MOT30N [csa2*]jL 0 e 0- MEMBRAME POTENfL.mY Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017 -so- 100 Wc i MY08N 1 ACTIN B 3Na T TUBULE 2K~ 2C~~~~~~~ MOTION *00- rC.2.]L.M 200- MEUMRANE POTET9AL~V Z MYOM I Z ACTN FIGURE 1. Thispage and facing page. Schematic illustration ofthe sequence ofE-C coupling in an adult mammalian ventricular myocyte. The intracellular ion transport events are shown in the main figures; the corresponding changes in membrane potential, cytosolic [Ca2+], and contractile force (shortening) from an isolated adult rabbit ventricular myocyte loaded with the Ca 2`-sensitive dye indo-1 stimulated at 1 Hz at 360C are shown in the insets on the left. Panel A: The processes ongoing in the resting myocyte or at end diastole. Panel B.: The events associated with early E-C coupling before a rise in cytosolic [Ca 2+]. Barry and Bridge Calcium Homeostasis in Myocytes 1809 MOTION [c2 ]i.LO Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017 MEMRANE POTENTIALV MOTION STU boo- 21m - MEMBRANE POTENTIAL.mV FIGURE 1. Continued. Panel C: Processes associated with the rise in [Ca 2`Ji and contraction. Panel D: The events associated with a fall in [Ca 2`Ji and relaxation. See text for discussion. SR sarcoplasmic reticulum; SL, sarcolemma. 1810 Circulation Vol 87, No 6 June 1993 Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017 various subunits of the calcium channel regulate and enhance many of the properties exhibited by the a1 subunit. It has recently become possible to express cardiac a, subunits in Xenopus oocytes by injecting the appropriate messenger RNA.30 By expressing the skeletal muscle subunits a2/8 and f3 alone or in combination, it has become possible to demonstrate interactions between these subunits and a. The current caused by the Ca24-channel agonist Bay K through a, alone is small but is enlarged if a1 is coexpressed with a2/8 and is somewhat reduced by coexpression with f3+a2/8.30 Experiments of this type reveal that the various subunits modulate sensitivity to dihydropyridine agonists, kinetics, and voltage dependence of activation and inactivation. The broad conclusion is that all subunits are required before the channel can fully express its native properties, although it should be emphasized that no combination of subunits yet reported completely mimics the native channel. It appears that Ca24 influx via the Na+-Ca2+ exchanger ("reverse" mode) can also occur during initial depolarization (See Figure 1B). First, theoretical considerations (see Equation 3) indicate that with abrupt membrane depolarization positive to the reversal potential for the Na+-Ca2+ exchanger, Ca21 influx should occur as the exchange reverses. This process may be further stimulated by subsarcolemmal rises in sodium concentration caused by the sodium influx via the cardiac sodium channel.32'33 However, whether or not this occurs during normal E-C coupling remains controversial.m The Ca21 that enters the cell early after depolarization via the L-type Ca21 channel (and possibly on the Na+-Ca2' exchanger) can bind to the Ca2' release channel of the SR (Figure 1B) and activate this channel,35 causing release of calcium from internal stores within the SR (calcium-induced calcium release36) (Figure 1C). As the [Ca2+]i concentration subsequently begins to rise as a result of Ca24 release from the SR, the reversal potential for the Na4-Ca2' exchanger becomes greater than the plateau of the membrane potential because of the rise in the [Ca21]i. Thus, after an initial period, which may be as brief as 10 msec,18 the Na4-Ca2' exchanger begins a Na4-in/Ca24-out "forward" mode of operation and thus contributes to Ca24 efflux, as we discuss below. The Ca24 release channel in the SR has been cloned37 and has been shown to be a 4,969-amino acid protein with a molecular mass of 564,711 d. Experiments by Anderson et a135 and Rousseau et a138 have demonstrated that the probability of opening of this channel is markedly increased by exposure to micromolar concentrations of Ca24. This channel is also opened by methylxanthines such as caffeine39 and has a high affinity for the plant alkaloid ryanodine. In low concentrations (<10 ytmol/L), ryanodine induces an open, low-conductance configuration of the channel38; at higher concentrations, ryanodine completely blocks the SR calcium release channel.40 The SR calcium release channel has phosphorylation, ATP-binding, and calmodulin-binding sites. The extent and mechanisms of its regulation are not well understood at present, although calmodulin appears to directly reduce the duration of single-channel opening without having an effect on single-channel conductance.38 Low concentrations of ryanodine open the SR calcium release channel and thus deplete the SR calcium stores. This effect has been used to examine the extent to which calcium release from SR by the calciuminduced calcium release mechanism is involved in cardiac E-C coupling. The magnitude of the decrease in contractile force induced by exposure to ryanodine ranges from 10% to 90%, depending on the species studied,41 with rat being the most sensitive. However, this may not accurately reflect the contribution of Ca24 released from the SR to force development, because in the presence of continued stimulation of myocardial cells, it appears that SR Ca2' stores can be maintained even in the presence of ryanodine.42'43 Thus, with continued stimulation of myocytes, ryanodine does not completely deplete the SR. In rested cells exposed to ryanodine,43 the SR becomes completely depleted. Under these conditions, Ca 2+ influx via the Ca 2+ channel (and perhaps on the Na4-Ca2' exchanger) causes little or no contraction directly. Interpretation of this result is complicated by the possibility that some of this influx of Ca24 could be directly taken up by an empty SR and thus prevented from binding to contractile elements. These observations, however, suggest that most of the Ca causing contraction in mammalian myocytes during normal E-C coupling is Ca24 released from the SR. The primary roles of the quantitatively relatively small Ca24 influx across the sarcolemma44 are to trigger SR Ca24 release in mammalian myocardium and to maintain SR Ca24 loading. There is good evidence that not all the Ca24 within the SR is released with each beat45 and that SR Ca24 release can be "graded" by the amount of trigger Ca2+ entering the cell. The evidence indicating a central role of Ca2+ influx via the Ca2+ channel in this process has recently been summarized by Fabiato.46 However, a graded mechanism requires that Ca2+ released from a Ca2+ release channel does not open adjacent Ca24 release channels not initially opened by the trigger Ca2+. This might occur if the rate constant of Ca2+ binding to an inactivating site of the SR Ca2+ release channel were lower than to an activating site.47 However, studies of the behavior of SR Ca2+ release channels incorporated into lipid bilayer membranes35,40 have failed to show evidence of Ca2+-induced inactivation of these channels. It is also possible that diffusion in the Ca2+ release channel environment is restricted,33 so that Ca2+ released from an SR release channel might not have easy access to other channels. However, Ca2+ spontaneously released from a Ca2+-overloaded SR may induce adjacent SR Ca2+ release, resulting in a propagating "wave" of increasing [Ca2+4] and contraction.48-50 A difference in system "gain" may account for the fact that Ca2+ released from SR does not activate adjacent SR Ca2+ release channels under conditions of normal E-C coupling,25 but this issue is still not satisfactorily resolved. The Ca2+ that is released from the SR initiates contraction by binding to the contractile proteins (Figure 1C). In the resting state, the interaction of actin and myosin is inhibited by the troponin-tropomyosin complex, which is bound to actin. When Ca24 binds to troponin C, a conformational change is induced that results in relief of this inhibition, with cross-bridge interaction and contractile element shortening. This Barry and Bridge Calcium Homeostasis in Myocytes interaction can be modulated by other contractile element proteins, such as C protein and myosin light chain 2 (see Reference 51 for review). Troponin T can also influence the affinity of troponin C for Ca2' and thus alter the sensitivity of the contractile elements for Ca2+.52 The affinity of troponin C for Ca2+ is also decreased by decreased intracellular pH.53 cAMP-dependent protein kinase-induced phosphorylation of troponin 1s4 decreases calcium sensitivity.55 Thus, changes in the affinity of the contractile elements for Ca2' as well as the magnitude of the Ca21 transient achieved during a single beat can regulate force development by a myocyte. Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017 Relaxation of the Ventricular Myocyte The decay of the calcium transient (Figure 1D) occurs because of reuptake of Ca21 into the SR mediated by the SR calcium ATPase and extrusion of Ca2+ from the myocyte as mentioned above primarily by the Na+-Ca21 exchanger. The SR calcium ATPase calcium transport enzyme is concentrated in the longitudinal component of the SR.56 The cardiac SR calcium ATPase, SERCA2, is a 100-115-kd protein that has been cloned.57 The Ca21 transport characteristics of this enzyme have been well characterized by studies of calcium uptake by SR membrane vesicles, including vesicles from human myocardium.58 The KCa for calcium transport is 300-400 nmol/L. Thus, the SR Ca2' ATPase transport is strongly activated in the range of physiological calcium concentrations that occur during the normal contraction-relaxation cycle. Regulation of the SR calcium ATPase occurs primarily by phosphorylation of phospholamban. Phospholamban is a 6,080-d protein59 that binds to and inhibits Ca21 transport by the SR calcium ATPase. When it is phosphorylated by cAMP-dependent protein kinase, this inhibition is removed, resulting in a greater Vmax and a lower Kca for calcium transport.60 Phosphorylation of phospholamban can also occur via action of Ca21 calmodulin-dependent and Ca21 phospholipid-dependent protein kinases at distinct sites.61 Dephosphorylation of phospholamban occurs via an SR-associated type 1 phosphatase.62 After uptake by the SR calcium ATPase in the longitudinal SR, Ca21 is bound to calsequestrin, which is located primarily in the junctional SR. Calsequestrin is the major Ca2+-binding protein in cardiac muscle (-50 mol Ca21 per mol protein.) Binding and release of Ca2' by calsequestrin is thought to be important in E-C coupling, but its exact role is not understood.63 Ca2' extrusion from the cells via the Na+-Ca2' exchanger also occurs during the declining phase of the Ca21 transient. At steady state, a balance exists between the amount of Ca2' entering the cell as the Ca21 current, ICa, and the amount of Ca2+ extruded by the Na+-Ca2' exchanger.64 Measurements of the amounts of current generated by these opposite Ca21 transport pathways show a 2:1 relation (Figure 2). For each 2+ charge entering the cell as a Ca 2+, a 3 + charge leaves the cell as Na+, transported with a 3 Na+:1 Ca2+ exchange stoichiometry. As previously mentioned, this electrogenic Na+-Ca2+ exchange is sensitive to membrane potential,8 and therefore the time course with which calcium is extruded is influenced by the trajectory of membrane potential. It has also been established that inward exchange current (corresponding to Ca2+ extru- 1811 sion) is linearly related to [Ca2+],.65 Thus, as the membrane potential changes during the Ca2' transient, the trajectory of Na+-Ca2' exchange current will reflect both the time courses of the Ca2' transient and the membrane potential. The striking effect of membrane potential on [Ca2+]i can be inferred from work by Bridge et a166 on voltageclamped myocytes. Ventricular myocytes in which the SR was disabled with caffeine were unable to relax from a contraction (induced by an imposed voltage clamp) until the membrane potential was restored toward resting value (Figure 3, left panel). Moreover, the rate of relaxation depended on the membrane potential at which relaxation occurred and was also Na+ dependent (see Figure 2). These experiments are consistent with a significant effect of membrane repolarization on the rate of Ca2+ extrusion via a voltage-sensitive Na+-Ca2+ exchange. There is not a great deal of information on the relation between Na+-Ca2+ exchange and voltage at various [Ca2+]i levels. Recent results indicate that at elevated [Ca 2+], the relation between exchange current and voltage is steep and is sigmoid67 (see Figure 3, right panel). Miura and Kimura68 have obtained similar results. We may infer from these current-voltage relations that membrane repolarization will have an activating effect on Na+-Ca2+ exchange. It should be appreciated that under certain circumstances, the inactivating effect of a declining Ca2+ could be offset by the activating effect of membrane repolarization if the membrane potential and [Ca2+]i transient coincide appropriately. In particular, this is likely to occur if membrane repolarization is rapid during a slowly declining [Ca2+1] transient. (see Figure 1D). Competition between Ca 2+ extrusion on the Na+Ca2+ exchanger and Ca2+ uptake by the SR69 is physiologically important because calcium that is taken up by the SR is available (after transport to SR Ca2' release sites where it is stored, bound to calsequestrin) to enhance the Ca21 transient during a subsequent beat and thus can contribute to increased force development. Under steady-state conditions, the amount of Ca2+ entering the cell equals that extruded from the cell, and the amount released by the SR equals that sequestered by the SR. If an abrupt change in this balance of fluxes takes place, then the calcium content of the SR will be affected. It is perhaps worth considering, therefore, how competition between the SR Ca21 pump and the Na+Ca2+ exchange might be regulated by membrane potential. Let us imagine that some intervention abruptly shortens the action potential, so that the steep phase of repolarization occurs earlier during the calcium transient. Because repolarization increases forward exchange earlier during the [Ca 2+]i transient, more Ca21 will be extruded than on a beat during which action potential duration was not reduced. Thus, the SR calcium content would be reduced at least by the additional calcium extruded. An action potential that was suddenly prolonged relative to the calcium transient would have the opposite effect, i.e., Ca2+ extrusion would be suppressed by membrane depolarization. Results reported by Spurgeon et a170 indicate that in rat ventricular myocytes, initial repolarization does occur early during the calcium transient. Thus, one would expect Na'-Ca 2+ exchange to be activated and the resulting inward current to contribute to the plateau of Circulation Vol 87, No 6 June 1993 1812 145maM Na0a 0 160 - 120 0- 1.5 nA JI NaCa-dt 80 (nA.ms) '0.3 |nA 40 80 160 120 200 JICa.dt (nA.ms) 0 Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017 I sec FIGURE 2. Tracings and graph demonstrating the quantitative relation between the amount of Ca2+ entering the cell as the calcium current ICa and the amount subsequently extruded by the Nai-Ca 2+ exchanger, INa-Ca. In the leftpanel, an inward calcium current (initial portion of trace A) is elicited by a voltage clamp pulse from -40 to 0 mVin the absence of external sodium (Nao). This is associated with a contraction, trace D. On repolarization of the voltage clamp to -40 mV (lower trace), there is no relaxation in the absence ofNa0. However, when Na0 is abruptly increased to 145 mmolIL (upper trace), there is onset of relaxation (trace E) associated with an inward current caused by electrogenic Na+-Ca 2+ exchange (trace B and with an expanded scale, trace C). The right panel shows the relation of the integrals of ICa and INa Ca from 11 experiments obtained in a similar protocol. The solid lines indicate the expected relations for a stoichiometry of 4 or 3 Na+ :1 Ca 2+. The results indicate that for the expected stoichiometry of 3:1, the amount of Ca 2+ entering as ICa is balanced by the amount of Ca2+ extruded by Nai-Ca 2+ exchange. Reprinted with permission.66 before significant repolarization had taken place, one might expect a reeverse exchange (Ca2' influx) during the late plateau of the action potential. Thus, agents that prolong the cardiac action potential might cause calcium overload by both impairing Ca2' extrusion and the rat action potential, as demonstrated by Schouten and Ter Keurs.71 Conversely, if the Ca transient were short compared with the cardiac action potential duration, different effects might be predicted. For example, were the SR to pump [Ca"2]i down to diastolic levels mV 0 -150 -100 -20 mY -80 -40 -50 0 50 100 0.0 ] -0.1 < 2 2 s*c -0.2 FIGURE 3. Left panel: Tracings showing voltage clamp pulses (top trace) and corresponding shortening records obtained in isolated guinea pig ventricular myocytes. In these experiments, in which the sarcoplasmic reticulum was disabled by exposure to 10 mmolIL caffeine, the extent of relaxation (and presumably decline in [Ca 2+J]) was strongly dependent on membrane voltage. Right panel: Graph showing the dependence of INa-Ca on membrane voltage in clamped guinea pig myocytes in which [Ca 2`Ji was elevated in the absence of Na0 by rapid pacing during exposure to 1 gumol/L ryanodine. Na0 was then rapidly applied at different voltages to activate the Naf-Ca2+ exchange current, INa-Ca. Note the strong dependence of INaCa on membrane voltage over the range of potentials shown in the left panel. Reprinted with permission.64 Barry and Bridge Calcium Homeostasis in Myocytes increasing Ca2` influx via Na+-Ca2` exchange. Calcium overload may cause spontaneous oscillatory release of Ca2` from SR,48 resulting in delayed afterdepolarizations and "triggered" arrhythmias in the intact heart. This mechanism may contribute to arrhythmias seen in the context of a prolonged QT interval caused by antiarrhythmic drug therapy and perhaps in the long-QT syndrome. Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017 Inotropic Drug Effects on Ca2' Homeostasis In the context of the above discussion, we may now briefly consider some of the mechanisms of action of inotropic drugs. The most commonly used of these is digitalis. The cardiac glycosides bind to the a-subunit of the Na+,K+-ATPase and inhibit the translocation of Na' and K+.72 Thus, partial inhibition of the sodium pump in the sarcolemma of a myocyte by cardiac glycosides causes a slight rise in intracellular sodium973,74 and a slight fall in intracellular potassium. An increase in intracellular Na' concentration would be expected to increase the influx of calcium on the Na+Ca2' exchanger initially during the action potential75 and to impair extrusion of Ca2' by Na+-Ca2+ exchange subsequently during the Ca21 transient. Both these effects would be expected to cause the observed increase in diastolic [Ca 2+1]9 and loading of the SR with Ca2+ ,76 providing more calcium available for Ca2+induced Ca2' release and thus an enhanced calcium transient and a positive inotropic effect. A slight prolongation of membrane action potential observed in some studies during inhibition of the Na' pump74 might also be expected to impair calcium extrusion by Na+Ca2' exchange and contribute to an inotropic effect. Catecholamines bind to 8-receptors in the cardiac cell membrane, activating a guanine nucleotide-binding protein (G protein), GQ. This in turn causes stimulation of adenylate cyclase and a resulting increase in the production of cAMP, which causes activation of cAMPdependent protein kinases.77 As mentioned above, this kinase causes phosphorylation of the a, subunit of the calcium channel, which increases calcium influx, and of phospholamban, which increases calcium uptake by the SR. These effects produce both an increase in the "trigger" Ca2+ entering the cell and an increase in releasable SR calcium content. In spite of a marked increase in calcium influx via the calcium channel, the cell is able to maintain its ability to relax (positive lusitropic effect) because of the stimulation of the SR calcium ATPase and shortening of the duration of the action potential. The shortening of the duration of the action potential and the corresponding enhanced extrusion of calcium by the Na4-Ca2+ exchanger may explain why progressive calcium overload, with elevation of diastolic [Ca2+],, does not usually occur in myocytes stimulated with catecholamines. In fact, end-diastolic [Ca24], may actually decline in myocytes exposed to isoproterenol.70 The increased Ca2+ influx per beat can be extruded more effectively by the Na+-Ca2+ exchanger because of the shorter duration of the action potential, which in turn results from a cAMP-induced increase in IK and IC.78,79 Some inotropic agents, including milrinone and amrinone, exert a positive inotropic effect by inhibiting phosphodiesterase, the enzyme that breaks down cAMP. Cyclic nucleotide phosphodiesterases differ with 1813 respect to drug sensitivity and substrate specificity,80 and the extent to which both cAMP-dependent protein kinases and phosphodiesterases are physically located adjacent to their particular substrates is currently under investigation.8' The phosphodiesterase inhibitors also are potent vasodilators, and thus the beneficial hemodynamic effects of these agents are a result of both positive inotropy and afterload reduction. Some inotropic agents increase the sensitivity of the contractile elements to calcium. Endothelin82,83 and angiotensin II84 can exert a potent inotropic effect in cardiac myocytes by inducing an intracellular alkalosis, and this change in intracellular pH sensitizes the contractile elements to calcium. The mechanisms of other agents that appear to sensitize contractile elements to calcium84-86 and thus produce an increase in force development without a corresponding increase in the magnitude of the calcium transient are less well understood. These agents may bind to the contractile element troponin-tropomyosin complex and, by a direct effect on these thin-filament regulatory proteins, alter the calcium sensitivity of the myofilaments. These agents have generated considerable interest because increased Ca24 loading of the SR is not a component of their action, and thus a positive inotropic effect may be achieved without danger of Ca2' overload-induced arrhythmias. Summary Calcium homeostasis in cardiac myocytes results from the integrated function of transsarcolemmal Ca 2+ influx and efflux pathways modulated by membrane potential and from intracellular Ca24 uptake and release caused predominantly by SR function. These processes can be importantly altered in different disease states as well as by pharmacological agents, and the resulting changes in systolic and diastolic [Ca24], can cause clinically significant alterations in contraction and relaxation of the heart. It may be anticipated that a rapid increase in our understanding of the pathophysiology of Ca24 homeostasis in cardiac myocytes will be forthcoming as the powerful new tools of molecular and structural biology are used to investigate the regulation of Ca 2+ transport systems. References [Ca24] in mammalian ventricle: Dynamic control by cellular processes. Annu Rev Physiol 1990;52:467-485 Reiter M: Calcium mobilization and cardiac inotropic mechanisms. Pharmacol Rev 1988;40:189-217 Bers DM: Excitation-contraction coupling and cardiac contractile force. Dev Cardiovasc Med 1991;122:1-258. Dordrecht/Boston/ London, Kluwer Academic Publications Nicoll DA, Longoni S, Philipson KD: Molecular cloning and functional expression of the cardiac sarcolemmal Na+-Ca 2+ exchanger. Science 1990;250:562-565 Hilgeman DW: Regulation and deregulation of cardiac Na+-Ca2+ exchanger in giant excised sarcolemmal membrane patches. Nature 1990;344:242-245 Frank JS, Mottino G, Reid D, Molday RR, Philipson KD: Distribution of the Na-Ca exchange protein in mammalian cardiac myocytes: An immunofluorescence and immunocolloidal gold-labelling study. J Cell Biol 1992;117:337-345 Reeves JP, Hale CC: The stoichiometry of the cardiac sodiumcalcium exchange system. J Biol Chem 1984;259:7733-7739 Kimura J, Noma A, Irisawa H: Na-Ca exchange current in mammalian heart cells. Nature 1986;319:596-599 Sheu SS, Fozzard HA: Transmembrane Na+ and Ca2+ electrochemical gradients in cardiac muscle and their relationship to force development. J Gen Physiol 1982;80:325-351 1. Weir WG: Cytoplasmic 2. 3. 4. 5. 6. 7. 8. 9. 1814 Circulation Vol 87, No 6 June 1993 Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017 10. Philipson KD, Bersohn MM, Nishimoto AY: Effects of pH on Na+-Ca2` exchange in cardiac sarcolemmal vesicles. Circ Res 1982; 50:287-293 11. Haworth RA, Goknur AB, Hunter DR, Hegge JO, Berkoff HA: Inhibition of Ca influx in isolated adult rat heart cells by ATP depletion. Circ Res 1987;60:586-594 12. DePolo R, Beauge L: Reverse Na-Ca exchange requires internal Ca and/or ATP in squid axons. Biochim Biophys Acta 1986;854: 298-306 13. O'Neil KT, DeGrado WF: How calmodulin binds to its targets: Sequence independent recognition of amphiphilic a-helices. Trends Biochem Sci 1990;15:59-64 14. Vorherr T, Chiesi M, Schwaller R, Carafoli E: Regulation of the calcium ion pump of sarcoplasmic reticulum: Reversible inhibition by phospholamban and by the calmodulin-binding domain of the plasma membrane calcium ion pump. Biochemistry 1992;31: 371-376 15. Li Z, Nicoll DA, Collins A, Hilgerman DW, Filoteo AG, Penniston JT, Weiss JN, Tomich JM, Philipson KD: Identification of a peptide inhibitor of the cardiac sarcolemma Na+-Ca2+ exchanger. J Biol Chem 1991;266:1014-1020 16. Carafoli E: The Ca2` pump of the plasma membrane. J Biol Chem 1992;267:2115-2118 17. Barry WH, Rasmussen CAF Jr, Ishida H, Bridge JHB: External Na-independent Ca extrusion in cultured ventricular cells. J Gen Physiol 1985;88:394-411 18. Cannell MB: Contribution of sodium-calcium exchange to calcium regulation in cardiac muscle. Ann N YAcad Sci 1991;639:428-443 19. Miyata H, Silverman HS, Sollott SJ, Lakatta EG, Stern MD, Hansford RG: Measurement of mitochondrial free Ca2+ concentration in living single rat cardiac myocytes. Am J Physiol 1991;261: H1123-H1134 20. Wolska BM, Lewartowski B: Calcium in the in situ mitochondria of rested and stimulated myocardium. J Mol Cell Cardiol 1991;23: 217-226 21. Robertson SP, Johnson JD, Patter JD: The time course of Ca2+ exchange with calmodulin, troponin, parvalbumin, and myosin in response to transient increases in Cal2. Biophys J 1981;34:559-569 22. duBell WH, Houser SR: Voltage and beat dependence of the Ca 2+ transient in feline ventricular myocytes. Am J Physiol 1989;257: H746-H759 23. Beukelmann DJ, Weir WG: Mechanism of release of Ca2+ from sarcoplasmic reticulum of guinea-pig cardiac cells. J Physiol (Lond) 1988;405:122-255 24. London B, Kreuger JW: Contraction in voltage-clamped internally perfused single heart cells. J Gen Physiol 1986;88:475-505 25. Niggli E, Lederer WJ: Voltage-independent calcium release in heart muscle. Science 1990;250:565-568 26. Sanguinetti MC, Krafte DS, Kass RS: Voltage-dependent modulation of Ca 2+ channel current in heart cells by Bay K8644. J Gen Physiol 1986;88:369-392 27. Fosset M, Jaimovich E, Delpont E, Lazdunski M: [3H] nitrendipine receptors in skeletal muscle. J Biol Chem 1983;258:6086-6092 28. Rios G, Brum G: Involvement of dihydropyridine receptors in excitation-contraction coupling in skeletal muscle. Nature 1987; 325:717-720 29. Catterall WA: Functional subunit structure of voltage-gated calcium channels. Science 1991;253:1499-1500 30. Singer D, Biel M, Lotan I, Flockerzi V, Hofmann F, Dascal N: The roles of the subunits in the function of the calcium channel. Science 1991;253:1553-1558 31. Mikami A, Imoto K, Tanabe T, Niidome T, More Y, Takeshima H, Narumiya S, Numa S: Primary structure and functional expression of the cardiac dihydropyridine-sensitive calcium channel. Nature 1989;340:230-233 32. Leblanc N, Hume JR: Sodium current induced release of calcium from cardiac sarcoplasmic reticulum. Science 1990;248:372-376 33. Lederer WJ, Niggli E, Hadley RW: Na+-Ca2` exchange in excitable cells: Fuzzy space. Science 1990;248:283 34. Sham JSK, Cleemann L, Morad M: Gating of the cardiac Ca21 release channel: The role of Na+ current and Na+-Ca21 exchange. Science 1992;255:850-853 35. Anderson K, Lai FA, Lin QY, Rousseau E, Erickson HP, Meissner G: Structural and functional characterization of the purified cardiac ryanodine receptor-Ca 2+ release channel complex. J Biol Chem 1989;204:1329-1331 36. Fabiato A: Calcium induced release of calcium from the cardiac sarcoplasmic reticulum. Am J Physiol 1983;245:C1-C14 37. Otsu K, Willard HF, Khanna VK, Zorzato F, Green NM, MacLennan DH: Molecular cloning of cDNA encoding the Ca 2+ release 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. channel (ryanodine receptor) of rabbit cardiac muscle sarcoplasmic reticulum. J Biol Chem 1990;265:13472-13483 Rousseau E, Smith JS, Meissner G: Ryanodine modifies conductance and gating behavior of single Ca 2+ release channels. Am J Physiol 1987;253:C364-C368 Meissner G, Henderson JS: Rapid calcium release from cardiac sarcoplasmic reticulum vesicles is dependent on Ca2+ and is modulated by Mg2', adenine nucleotide and calmodulin. J Biol Chem 1987;262:3065-3073 Meissner G: Ryanodine activation and inhibition of the Ca21 release channel of sarcoplasmic reticulum. J Biol Chem 1986;261: 6300-6306 Sutko JL, Willerson JT: Ryanodine alteration of the contractile state of rat ventricular myocardium: Comparison with dog, cat, and rabbit ventricular tissues. Circ Res 1980;46:332-343 Rasmussen CAF Jr, Sutko JL, Barry WH: Effects of ryanodine and caffeine on contractility, membrane voltage, and calcium exchange in cultured heart cells. Circ Res 1987;60:495-504 Bers DM, Bridge JHB, MacLeod KT: The mechanisms of ryanodine action in rabbit cardiac muscle evaluated with Ca 2k-sensitive microelectrodes and rapid cooling contractures. Can J Physiol Pharmacol 1987;65:610-618 Isenberg G: Ca entry and contraction as studied in isolated bovine ventricular myocytes. Z Naturforsch 1982;37c:502-512 Moravec CS, Bond M: Calcium is released from the junctional sarcoplasmic reticulum during cardiac muscle contraction. Am J Physiol 1991;260:H989-H997 Fabiato A: Appraisal of the physiological relevance of two hypotheses of the mechanism of Ca 2+ release from the cardiac sarcoplasmic reticulum: Calcium-induced release versus charge coupled release. Mol Cell Biochem 1989;89:135-143 Fabiato A: Time and calcium dependence of activation and inactivation of calcium-induced calcium release from the sarcoplasmic reticulum of a skinned canine cardiac Purkinje cell. J Gen Physiol 1985;85:247-289 Berlin JR, Cannell MB, Lederer WJ: Cellular origins of the transient inward current in cardiac myocytes: Role of fluctuations and waves of elevated intracellular calcium. Circ Res 1989;65:115-126 Wier WG, Blatter LA: Ca 2k-oscillations and Ca 2k-waves in mammalian cardiac and vascular smooth muscle cells. Cell Calcium 1991;12:241-251 Williams DA, Delbridge LM, Cody SH, Harris PJ, Morgan TO: Spontaneous and propagated calcium release in isolated cardiac myocytes viewed by confocal microscopy. Am J Physiol 1992;262: C731-C742 Moss RL: Ca 2+ regulation of mechanical properties of striated muscle. Circ Res 1992;70:865-884 Nassar R, Malouf NN, Kelly MD, Oakeley AE, Anderson PAW: Force-pCa relation and troponin T isoforms of rabbit myocardium. Circ Res 1991;69:1470-1475 Blanchard EM, Solaro RJ: Inhibition of the activation and troponin calcium binding of dog cardiac myofibrils by acidic pH. Circ Res 1984;55:382-391 McClellan GB, Winegrad S: Cyclic nucleotide regulation of contractile proteins in mammalian cardiac muscle. J Gen Physiol 1980; 75:283-295 Ray KP, England PJ: Phosphorylation of the inhibitory subunit of troponin and its effect on the calcium dependence of cardiac myofibril adenosine triphosphatase. FEBS Lett 1976;70:11-18 Jorgensen AO, Shen ACY, Daly P, MacLennan DH: Localization of Ca2++Mg2+-ATPase of the sarcoplasmic reticulum in adult rat papillary muscle. J Cell Biol 1982;93:883-892 Grover AK, Khan I: Calcium pump isoforms: Diversity, selectivity and plasticity. Cell Calcium 1992;13:9-17 Movsesian MA, Colyer J, Wang JH, Krall J: Phospholambanmediated stimulation of Ca 2+ uptake in sarcoplasmic reticulum from normal and failing hearts. J Clin Invest 1990;85:1698-1702 Fujii J, Ueno A, Kitono K, Tanaka S, Kadoma M, Tada M: Complete complementary DNA-derived amino acid sequence of canine cardiac phospholamban. J Clin Invest 1987;79:301-304 Sasaki T, Incis M, Kimura Y, Kuzuya T, Tada M: Molecular mechanism of regulation of Ca2+ pump ATPase by phospholamban in cardiac sarcoplasmic reticulum. J Biol Chem 1992;267: 1674-1677 Simmerman HKB, Theibert JL, Wegener AD, Jana KR: Sequence analysis of phospholamban: Identification of phosphorylation sites and two major structural domains. J Biol Chem 1986;261: 13333-13341 Barry and Bridge Calcium Homeostasis in Myocytes Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017 62. Steenaart NAE, Ganim JR, DiSalvo J, Kranias EG: The phospholamban phosphatase associated with cardiac sarcoplasmic reticulum is a type 1 enzyme. Arch Biochem Biophys 1992;293:17-24 63. Cala SE, Jones LR: Phosphorylation of cardiac and skeletal muscle calsequestrin isoforms by casein kinase II. J Biol Chem 1991;266: 391-398 64. Bridge JHB, Smolley JR, Spitzer KW: The relationship between charge movements associated Ica and INa.C. in cardiac myocytes. Science 1990;248:376-378 65. Barcenas-Ruiz L, Beuckelmann DJ, Weir WG: Sodium-calcium exchange in heart: Membrane currents and changes in [Ca2+j]. Science 1987;238:1720-1722 66. Bridge JHB, Spitzer KW, Ershler PR: Relaxation of isolated ventricular myocytes by a voltage dependent process. Science 1988;241: 823-825 67. Bridge JHB, Smolley J, Spitzer KW, Chin TK: Voltage dependence of sodium-calcium exchange and the control of Ca extrusion in the heart. Ann N YAcad Sci 1991;639:34-47 68. Miura Y, Kimura J: Sodium-calcium exchange current. J Gen Physiol 1989;93:1129-1145 69. Bers DM, Bridge JHB: Relaxation of rabbit ventricular muscle by Na-Ca exchange and sarcoplasmic reticulum. Circ Res 1989;65: 334-342 70. Spurgeon HA, Stern MA, Baartz G, Raffalli S, Hanford RG, Tale A, Lakatta EG, Capogrossi MC: Simultaneous measurement of Ca2 , contraction, and potential in cardiac myocytes. Am J Physiol 1990;258:H574-H586 71. Schouten VJA, Ter Keurs HEDJ: The slow repolarization phase of the action potential in rat heart. J Physiol (Lond) 1985;360:13-25 72. Horisberger JD, Lemas V, Kraehenbuhl JP, Rossier BC: Structure-function relationship of Na,K-ATPase. Annu Rev Physiol 1991; 53:565-584 73. Biedert S, Barry WH, Smith TW: Inotropic effects and changes in sodium and calcium contents associated with inhibition of monovalent cation active transport by ouabain in cultured myocardial cells. J Gen Physiol 1979;74:479-494 74. Lee CO, Levi A: The role of intracellular sodium in the control of cardiac contraction. Ann N YAcad Sci 1991;639:408-427 75. Barry WH, Hasin Y, Smith TW: Sodium pump inhibition, enhanced calcium influx via sodium-calcium exchange, and positive inotropic response in cultured heart cells. Circ Res 1985;56: 231-241 1815 76. Bers DM, Bridge JHB: Effect of acetylstrophanthidin on twitches, microscopic tension fluctuations, and cooling contractures in rabbit ventricle. J Physiol (Lond) 1988;404:53-69 77. Birnhaumer L: G proteins in signal transduction. Annu Rev Phar- macol Toxicol 1990;30:675-705 78. Walsh KB, Begenisich TB, Kass RS: 3-Adrenergic modulation in the heart: Independent regulation of K and Ca channels. Pflugers Arch 1988;411:232-234 79. Hume JR, Harvey RD: Chloride conductance pathways in heart. Am J Physiol 1991;261:C399-C412 80. Lugnier C, Gauthier C, LeBec A, Soustre H: Cyclic nucleotide phosphodiesterases from frog atrial fibers: Isolation and drug sensitivities. Am J Physiol 1992;262:H654-H660 81. Hohl CM, Li Q: Compartmentation of cAMP in adult canine ventricular myocytes. Circ Res 1991;69:1369-1379 82. Kramer BK, Smith TW, Kelly RA: Endothelin and increased contractility in adult rat ventricular myocytes. Circ Res 1991;68: 269-279 83. Kohmoto 0, Ikenouchi H, Hirata Y, Momomura S, Serizawa T, Barry WH: Variable effects of endothelin-1 on [Ca2+]i transients, pH,, and contraction in ventricular myocytes. Am J Physiol (in press) 84. Ikenouchi H, Bridge JHB, Lorell BH, Zhao L, Barry WH: Effects of angiotensin II on [Ca2+]i, motion, Ca2l current, and pHi in adult rabbit ventricular myocytes. (abstract) Circulation 1992;86(suppl I):I-218 85. Bohm M, Morano I, Pieske B, Ruegg JC, Wankerl M, Zimmerman R, Erdmann E: Contribution of cAMP-phosphodiesterase inhibition and sensitization of the contractile proteins for calcium to the inotropic effect of pimobendan in the failing human myocardium. Circ Res 1991;68:689-701 86. Ferroni C, Hano 0, Ventura C, Lakatta EG, Klockow M, Spurgeon H, Capogrossi MC: A novel positive inotropic substance enhances contractility without increasing the Ca2` transient in rat myocardium. J Mol Cell Cardiol 1991;23:325-331 87. Hajjar RJ, Gwathmey JK: Modulation of calcium-activation in control and pressure-overload hypertrophied ferret hearts: Effect of DPI201-106 on myofilament calcium responsiveness. J Mol Cell Cardiol 1991;23:65-75 KE-Y WORDs * calcium * myocytes * Research Advances Series Intracellular calcium homeostasis in cardiac myocytes. W H Barry and J H Bridge Circulation. 1993;87:1806-1815 doi: 10.1161/01.CIR.87.6.1806 Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017 Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 1993 American Heart Association, Inc. All rights reserved. Print ISSN: 0009-7322. Online ISSN: 1524-4539 The online version of this article, along with updated information and services, is located on the World Wide Web at: http://circ.ahajournals.org/content/87/6/1806 Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Circulation is online at: http://circ.ahajournals.org//subscriptions/