I ntracellular Na regulation in cardiac myocytes

Cardiovascular Research 57 (2003) 897–912www.elsevier.com/ locate/cardiores

Review1I ntracellular Na regulation in cardiac myocytesa , b a*Donald M. Bers , William H. Barry , Sanda Despa

aDepartment of Physiology, Loyola University Chicago, Stritch School of Medicine, 2160 South First Avenue, Maywood, IL 60153,USAbDivision of Cardiology, University of Utah, Health Science Center, Salt Lake City, UT 84132,USA

Received 8 July 2002; accepted 2 September 2002

Abstract

1 1 1Intracellular [Na ] ([Na ] ) is regulated in cardiac myocytes by a balance of Na influx and efflux mechanisms. In the normal celli1there is a large steady state electrochemical gradient favoring Na influx. This potential energy is used by numerous transport

1 1mechanisms, including Na channels and transporters which couple Na influx to either co- or counter-transport of other ions and1 1 1 21solutes. Six sarcolemmal Na influx pathways are discussed in relatively quantitative terms: Na channels, Na /Ca exchange,

1 1 1 21 1 2 1 1 2 1 21Na /H exchange, Na /Mg exchange, Na /HCO cotransport and Na /K /2Cl cotransport. Under normal conditions Na /Ca31 1exchange and Na channels are the dominant Na influx pathways, but other transporters may become increasingly important during

1 21 1 1altered conditions (e.g. acidosis or cell volume stress). Mitochondria also exhibit Na /Ca antiporter and Na /H exchange activity1 21 1 2that are important in mitochondrial function. These coupled fluxes of Na with Ca , H and HCO make the detailed understanding of3

1 1 1[Na ] regulation pivotal to the understanding of both cardiac excitation–contraction coupling and pH regulation. The Na /K -ATPasei1 1 1is the main route for Na extrusion from cells and [Na ] is a primary regulator under physiological conditions. [Na ] is higher in rati i

1 1 1than rabbit ventricular myocytes and the reason appears to be higher Na influx in rat with a consequent rise in Na /K -ATPase activity1 1 1(rather than lower Na /K -ATPase function in rat). This has direct functional consequences. There may also be subcellular [Na ]i

1gradients locally in ventricular myocytes and this may also have important functional implications. Thus, the balance of Na fluxes in1heart cells may be complex, but myocyte Na regulation is functionally important and merits focused attention as in this issue.

2003 European Society of Cardiology. Published by Elsevier Science B.V. All rights reserved.

Keywords: Myocytes; Na/Ca-exchanger; Na/H-exchanger; Na/K-pump; Na-channel

1 11 . General aspects of cardiac Na regulation getically unfavorable transmembrane solute flow to Natransport (secondary active transport). Examples include

1 1 1 1Intact cells maintain a large electrochemical [Na ] Na /neurotransmitter, Na /glucose and Na /amino acids1gradient across the plasma membrane. While extracellular cotransporters as well as various Na / ion co- and counter-

1 1 1 1[Na ] ([Na ] ) is |140 mM, intracellular free [Na ] transporters. The [Na ] also plays an important role ino i1 1 21([Na ] ) is normally 4–16 mM. In excitable cells, the regulating the mitochondrial pH, [Na ] and [Ca ], as ai

1 1 1 1 21energy stored in the transmembrane Na gradient providessubstrate for mitochondrial Na /H and Na /Ca ex-the basis for fast electrical signaling, i.e. propagation of changers.action potentials in neurons and muscles. Many cells Let us consider (Fig. 1A) the free energy of the

1 1 1utilize the electrochemical Na gradient to couple ener- electrochemical Na gradientDG 5RT ln([Na ] /Na i1[Na ] )1 zFE , where the first term is the chemicalo m

energy while the second is the electrical energy (R is the

*Corresponding author. Tel.:11-708-216-1018; fax:11-708-216-6308.

E-mail address: [email protected](D.M. Bers). Time for primary review 28 days.

0008-6363/03/$ – see front matter 2003 European Society of Cardiology. Published by Elsevier Science B.V. All rights reserved.doi:10.1016/S0008-6363(02)00656-9

Dow

nloaded from https://academ

ic.oup.com/cardiovascres/article/57/4/897/347248 by guest on 04 July 2022

mailto:[email protected]

898 D.M. Bers et al. / Cardiovascular Research 57 (2003) 897–912

1Fig. 1. (A) Schematic illustration of the transmembrane electrochemical Na gradient. (B) Main pathways involved in [Na] regulation in cardiac cells.i

universal gas constant,T is temperature in8K, z is valence the cardiac action potential, an increasing fraction of the11 energy from ATP is used for K pumping. At the peak ofand F is Faraday’s constant). For a typical [Na ]510i

the action potential (E |140 mV) most of the energymM andE 5280 mV,DG 5215 kJ/mol (with the two mm Na1 1goes into pumping K in versus Na out.terms being almost equal, at27 and28 kJ/mol, respec-

1This Na energy gradient is used by many processes,tively). A negative value forDG means that Na entry is21 1 211 including extrusion of Ca from the cell via Na /Caenergetically favored, in this case that Na ions tend to

21exchange (NCX). Extrusion of Ca is also very uphillenter the cell. ThisDG is also equivalent to2150 mV inNa

energetically, with an electrochemical gradient of|41electrical potential terms. The energy to build this potential1 1 kJ/mol (where the chemical and electrical terms are 26 andenergy gradient comes from the Na /K -ATPase, which

1 1 15 kJ/mol, respectively). Thus, it requires the energytransports 3 Na out and 2 K in for each ATP hydro-11 obtained from the downhill transport of 3 mol of Nalyzed. The transport of 3 mol of Na against its electro-

21(245 kJ) to remove each mole of Ca (41 kJ), and thischemical gradient requires a free energy of23?DG |2Na

agrees with the generally accepted stoichiometry of the46 kJ. The free energy released by the hydrolysis of 1 mol1 211 1 NCX (3 Na :1 Ca ). The NCX is relatively uniqueof ATP,DG , is 259 kJ. Thus, the Na /K -ATPase hasATP1among the Na exchangers and cotransporters in Fig. 1Ba remarkable energetic efficiency of|77%. At restingEm

in that it may reverse under physiological conditions. Inalmost all of the energy used (.95%) goes into the uphill21 11 1 that case Ca entry can fuel extrusion of 3 Na . This cantransport of Na , because the [K ] gradient is near its

occur because NCX has a reversal potential (analogous toelectrochemical equilibrium (E |290 mV). That is, theK1 ion channels) that is in the physiological range ofE ,uphill chemical [K ] gradient into the cell is almost m

defined asE 53E 22E (whereE andE are theexactly balanced by a downhill electrical gradient for the Na / Ca Na Ca Na Ca1 1 21positive K ion. Of course as the cell depolarizes during equilibriumE for Na and Ca ). Thus the relativem

Dow



D.M. Bers et al. / Cardiovascular Research 57 (2003) 897–912 899

1 21 1energies in the Na and NCX gradients can change with is involved in extruding Mg . Na entering the cells1conditions and this will be addressed in more detail below through these mechanisms is then extruded by the Na /

1 21 1(see Na /Ca exchange section below). K pump. The pump utilizes energy derived from the1 21 1While Na is analogous to Ca in the steep electro- hydrolysis of ATP to exchange three internal Na for two

1chemical gradient favoring ion entry, cells seem to use external K ions and thus maintains the transmembrane1 21 21 1Na and Ca fluxes differently. Ca can carry a major [Na ] gradient. Intracellular buffering is important in the

21depolarizing current in cellular excitation, but in addition regulation of many intracellular ions, including Ca21 21 21free intracellular [Ca ] ([Ca ] ) often changes by a (100:1), Mg (|100:1) and pH (.1000:1). Althoughi

1factor of |10 during activation in many cell types (from data regarding intracellular Na buffering are sparse, we1|100 nM to|1 mM, often in less than 100 ms). The large recently measured very low Na buffering capacity (|1.3)

21 21change in [Ca ] allows [Ca ] to function as a re- in rabbit ventricular myocytes by measuring simultaneous-i i1 1 1markably ubiquitous second messenger in regulating pro- ly [Na ] and integrated Na /K pump current duringi

1cesses such as contraction, secretion and transcription. [Na ] decline upon abrupt pump activation [1]. However,i1 1However, while the Na gradient is used directly for this low global Na buffering capacity does not preclude1 1 1electrical signaling via Na current, changes in [Na ] are local subsarcolemmal Na buffering which may exist (seei

typically small requiring many seconds to minutes to rise below).1 1by even 10%. This makes [Na ] itself a poor candidate Resting [Na ] in heart cells is in the range of 4–8 mMi i

for the sort of rapid dynamic regulation attributed to for most mammalian species [2–7] and much higher (10–21 1[Ca ] . However, the high concentration of Na (in the 15 mM) in rat and mouse [4–9]. This interspecies differ-i

1mM range) allows it to readily participate in quantitatively ence is rather intriguing, as higher resting [Na ] mighti21 21large coupled transport fluxes. This is important for both limit Ca extrusion, altering cellular Ca balance and

the transport of ions across membranes of many cell types,also contributes to rest potentiation, seen in rat, mouse and1but is also important in driving the uphill transport of some other cardiac muscle [7]. [Na ] increases duringi

important cellular substrates such as glucose, amino acidsstimulation in a frequency dependent manner. This is1and neurotransmitters. because Na influx per unit time is higher due to more

1 1In the steady-state, [Na ] is determined by the balance frequent activation of Na channels and NCX. Cook et al.i1between Na influx and efflux. There are several pathways [10] showed that there are also regional differences in the

1 1for Na entry into cardiac cells (Table 1 and Fig. 1B). regulation of [Na ] in rabbit left ventricle (higher ini1Voltage-gated Na channels are responsible for the up- sub-epicardial than sub-endocardial myocytes) both at rest

1stroke of action potentials. NCX generally uses Na influx and during steady-state stimulation at 0.5 Hz. However,21 21 1 1to extrude Ca , and is the main mechanism for Ca the expression of the Na /K pump and the pump current

1 1extrusion during relaxation. Na /H exchange (NHE) density are similar in the two regions [11]. This suggests1 1 1extrudes H in exchange for Na and is critical in cellular that regional differences in [Na ] regulation might arisei

1pH homeostasis. Another important acid extruding trans- from differences in Na influx.1 2 1port is the Na /HCO cotransporter (NBC). There is also Because of NCX, NHE and NBC, [Na ] is centrally3 i

1 1 21a Na /K /2Cl cotransporter (NKCC) important in vol- involved in controlling the [Ca ] and pH . Therefore,i i1 21 1ume regulation, and a Na /Mg exchanger (NMgX) that [Na ] plays an important role in the heart, as both thesei

Table 11Na transport in cardiac cells

Nickname Transport Resting flux Inhibitorsaand stoichiometry rate (mM/min)

Na extrusion pathways1 1 1 1 1 bNa /K -ATPase Na pump,I 3 Na :2 K exchange 0.77 Cardiac glycosidespump

1Na influx pathways1Na channel /current I Uncoupled 0.14 TetrodotoxinNa1 21 c 1 21 d 21Na /Ca exchanger NCX 3 Na :1 Ca exchange 0.28 Ni1 1 1 1Na /H exchanger NHE 1 Na :1 H exchange 0.12 Amiloride, cariporide1 2 1 2 dNa /HCO cotransporter NBC 1 Na :1 HCO cotransport 0.12 DIDS, SITS3 31 1 2 1 1 2Na /K /2Cl cotransporter NKCC 1 Na :1 K :2 Cl cotransport 0.15 Furosemide, bumetanide1 21 1 21Na /Mg antiporter NaMgX 2 Na /1 Mg exchange 0.08 Imipraminea Values are based in part on our measurements in rabbit ventricular myocytes [5] and some extrapolations from guinea-pig data [80,102] (see text for

details).b 1Determined in Ref. [5] for a [Na ] of 5 mM.ic 1 21 1 21 1Depending on the internal and external Na and Ca concentrations and the membrane potential, the Na /Ca exchanger can also function as a Na

extrusion pathway (see text for details).d Stoichiometry is still a matter of debate (see text).

Dow




ions are important factors in excitation–contraction cou- [23] heart, although small amounts ofa [27] anda [28],3 21pling (ECC) [12]. An increase in [Na ] would shift the respectively, have also been reported. In the rat, thea andi 1

21balance of fluxes on the NCX to favor more Ca influx a isoforms are expressed in fetal and neonatal hearts and321 21and less Ca efflux, resulting in larger Ca transients thea isoform is replaced by thea isoform early in3 2

and therefore enhanced contractility. This is the likely development [29,30]. Interestingly, the reverse switchmechanism of the inotropic effect of cardiac glycosides, (froma to a ) occurs in hyperthrophied or failing rat2 3

1 1which are specific inhibitors of the Na /K pump and heart [31,32]. Theb isoform is the only b subunit1

have been used to enhance cardiac contractility in the appreciably expressed in the human heart [22], although atreatment of congestive heart failure for more than 200 recent report indicated thatb is also present [33].3

1 1 1years. Indeed, relatively small changes of [Na ] can have Different Na /K -ATPase isoforms have different gly-i1major effects on contractile force. In cardiac Purkinje cosides and [Na ] sensitivity. In the rat,a is |100 timesi 1

1fibers, force can double with|1 mM rise of [Na ] [13]. more resistant to ouabain thana and a . In rabbit, pig,i 2 31In ventricular muscle, small increases in [Na ] produce dog and human, thea isoform is much more sensitive toi 1

less percentage increase in force, probably because of a ouabain than in rat (for a review, see Ref. [16]), thereforeceiling effect. That is, in rabbit ventricle at|29 8C and 0.5 the differences in glycosides affinity are not as marked.Hz, control twitches are already|40% of the maximum Indeed, it has been recently shown [34] that the affinity ofmyofilament force [14], limiting the extent of further all three humana subunit isoforms for ouabain is similar

1 1increase. (|18 nM). Expression of Na /K -ATPase is higher in1 1As Fig. 1B indicates there are many Na entry path- ventricle than atrium [22]. In ventricular myocytes, Na /

1 1 1 1ways, but the Na /K -ATPase is the main Na extrusion K pumps are located in both peripheral sarcolemma and1 1pathway. We will first consider Na /K -ATPase and then T-tubules [23]. In the rat, thea isoform is preferentially1

1Na influx mechanisms. distributed in T-tubules, whereasa and b are homoge-2 1

neously distributed in the T-tubules and peripheral sar-colemma [23]. The level of expression of different iso-

1 12 . Na /K pump forms and/or their cellular localization could have im-portant physiological consequences. James et al. [35]

1 12 .1. Na /K pump structure, isoforms and expression studied the phenotypes of mouse hearts with geneticallyin the heart reduced levels ofa or a isoforms. They found that1 2

21heterozygousa hearts have increased Ca transients and21 1The Na /K pump is a member of the P-type ATPase contractility whereas the opposite happens for thea1

pumps. The reaction mechanism of these ATPases is based heterozygotes. Furthermore, inhibition of thea isoform2

on the formation of a phosphorylated intermediate. Ac- with ouabain increased the contractility of heterozygousa1

companying the phosporylation–dephosphorylation pro- hearts. These results might indicate a specific role for the21cess, P-type ATPases bind, occlude and transport ions bya isoform as a regulator of Ca in the mouse heart, for2

cycling between two different conformations, called E1 example by affecting the function of the Na/Ca exchange.1 1and E2 [15]. The Na /K -ATPase has two major This can happen if thea isoforms are preferentially2

subunits:a and b (for a review, see Ref. [16]). Thea located in the dyads. This is indeed the case in smoothsubunit was first cloned by Shull et al. [17] and Kawakami muscle where the low ouabain-affinitya -isoform is1

et al. [18], has a molecular weight of|110 kDa and ubiquitously distributed, but the higher affinitya - and21 1contains the binding sites for ATP, Na , K and cardiac a -isoforms are preferentially localized in regions overly-3

glycosides (specific inhibitors of the enzyme). The smaller ing the SR [36].b subunit (|50 kDa) modulates the ATPase activity and isimportant in the proper membrane insertion of the pump. A

1 1third, smaller (|12 kDa) protein (g subunit) has also been 2 .2. Ion transport characteristics of the Na /K pumpfound in various tissues [19] but its physiological function

1 1 1is not yet known. Foura (a –a ) and threeb (b –b ) The Na /K -ATPase transports three Na ions out and1 4 1 31 1 1subunits of the Na /K -ATPase have been identified. The two K ions into the cell using the energy of one ATP

a -a isoforms are expressed in a variety of tissues, molecule, and thus moves out one net charge per cycle. Ion1 31 1whereas thea isoform has only been detected in rat testis transport by the Na /K pump has a functional reversal4

[20]. Any ab combination results in a functional pump. potential which depends on both intracellular and extracel-1 1The a isoform is present in the cardiac tissue of all lular [Na ] and [K ], as well as the free energy of1

species studied, but there is marked variation ofa anda intracellular ATP hydrolysis (DG ). Under normal con-2 3 ATP

subunit expression among species. All threea-isoforms are ditions, the reversal potential has been estimated to be1 1present in the human heart [21–23], whereasa and a about 2180 mV [37], such that Na /K pump current1 3

1isoforms are expressed in dog [24,25] and ferret [26]. The (I ) and Na transport are outward over the wholepump

a isoform is predominant in rabbit [11] and guinea-pig physiological range. The ATP concentration for half-maxi-1

Dow




1 1mal activation of the cardiac Na /K pump is in the rangeof 80–150mM [38,39], therefore under control conditionsATP is not rate limiting for the pump (normal ATP levelsin cardiac cells are 5–10 mM). However, this can changeas [ATP] declines during ischemia or metabolic inhibition,and the simultaneous rise in [ADP] and [P ] also contributei

to a reducedDG available for transport. Such aATP1 1reduction inDG would also reduce the [Na ] and [K ]ATP

gradients that the pump can generate. However, duringshort-term metabolic inhibition glycolysis may regenerate

1 1ATP near the Na /K -ATPase, making the pump lessdirectly dependent on oxidative phosphorylation [40].

1 1The main physiological regulators of the Na /K pump1 1 1are internal Na and external K . The [Na ] for half-i

maximal pump activation (K ) in the heart varies widelym

with the internal and external ionic conditions, in the range1of 8–22 mM (for a review, see Ref. [41]). Intracellular K

1competes with Na for binding to the enzyme at the1cytoplasmic surface and results in reduced [Na ] -sen-i

sitivity of the pump [42]. The activatingK for extracellu-m1 1lar K , in the presence of normal external [Na ], is 1–2

mM [41], therefore the pump is|70% saturated with1respect to external K at a normal concentration of 4 mM.

1 1External Na and K compete for common binding sites1 1to the Na /K -ATPase, therefore theK for pumpm

1 1activation by external K is appreciably lower in Na -free1conditions [43]. The absence of external Na also renders

11 1 1 1I voltage insensitive [43]. With normal Na outside,pump Fig. 2. (A) The [Na ] -dependence of Na /K pump-mediated Nai

1the E -dependence ofI in ventricular myocytes is efflux in rat and rabbit ventricular myocytes. Cells were first Na -loadedm pump1 1 1by inhibiting the Na /K pump in the absence of external K . Then, thesigmoidal in shape with a steep positive slope between

1pump was re-activated with 4 mM K and we monitored the time course2100 and 0 mV and nearly no voltage dependence ofIpump 1 1 1of [Na ] decline. The rate of [Na ] decline (2d[Na ] /dt) wasi i iat positive potentials [44] (for a review, see Ref. [45]). 1calculated at each [Na ] . The solid lines represent the fit with a Hill1 1 iAs mentioned above, multiple Na /K -ATPase iso- expression. Replotted from Ref. [5]. (B) Steady-stateE -dependence ofm

forms exist in cardiac cells from most species, and they I activation, inactivation (or availability) and window current.Na

might differ functionally because of differential specificintracellular localization, regulation or kinetic characteris-

1tics. The [Na ] -dependence of the three rata-subuniti1isoforms has been studied in transfected cells [16,46]. activation by Na (K (Na) (was higher in rat) 10 vs. 7m

When transfected into insect cells, the apparent affinity for mM). TheV values may indicate higher pump proteinmax1intracellular Na varies in the ordera .a .a [16] expression levels in rat versus rabbit. We speculate that this2 1 3

whereas in HeLa cells the order seems to bea .a .a might be a cellular adaptation to a chronically higher1 2 31 1[46]. This difference might be due to the different method [Na ] , and there is precedent for such [Na ] -dependenti i

1 1 1 1used to determine the pump activity (Na /K -ATPase upregulation of Na /K -ATPase expression levels [47].1 1activity vs. the rate of Na extrusion from intact cells). Fig. 2A indicates that at rest, when [Na ] is 4.5 mM ini

1 1However, the common point is thata isoform has a much rabbit and 11.1 mM in rat [5], the Na /K pump is only31lower Na -affinity (K |30 mM) thana or a , making it running at|15% ofV in rabbit ventricle versus|60%m 1 2 max

1less active at physiological [Na ] , perhaps serving as a in rat. Therefore rabbit cells might have a greater reservei1 1‘safety’ mechanism in conditions of Na overload. Thea to cope with an enhanced Na influx brought about by3

1isoform also has a lower [K ] affinity [16]. higher stimulation frequency or by pathological conditions.o1 1 1 1How much Na is extruded by the Na /K -ATPase to The rate of Na efflux in rat at the higher measured resting

1 1 1 1 1maintain steady-state [Na ] ? We measured [5] the [Na ] - [Na ] is higher than the resting Na /K -pump rate ini i i1 1 1 1dependence of the Na /K pump-mediated Na efflux in rabbit ventricular myocytes (Fig. 2A). Since Na efflux

1rat and rabbit ventricular myocytes (Fig. 2A). There is must equal Na influx at the steady state, this means that1 1 1 1little difference in the Na /K pump rate for [Na ] resting Na influx must also be higher in rat than rabbiti1 1below 11 mM. However, Na /K -ATPase maximum rate ventricular myocyte. In the following, we will discuss

1(V ) was nearly twice as high in rat and the half- pathways that contribute to this Na influx.max

Dow




1 13 . Pathways for Na influx also be present in cardiac cells [51,52]. This persistent Nacurrent is more sensitive to block by TTX, is activated at

13 .1. Voltage-gated Na channels more negative potentials and has an amplitude of|0.5% ofthe peak transientI . It has been shown thatI has aNa Na,slow

1Voltage-gated Na channels are activated by depolariza- more prominent contribution toI in heart failure andNa

tion and are responsible for the rapid upstroke of the might be partly responsible for the prolongation of the APcardiac action potential (AP). Once threshold depolariza- [52,53].

1tion for an AP is reached, the opening of Na channels As shown in Fig. 2B, there is a very small window atleads to further depolarization, which activates additional about252 mV where both channel availability and activa-

1Na channels and depolarization, etc. This positive feed- tion are|3.5%. This would produce a steady stateback creates a very rapid depolarization toward the electro- ‘window current’ with a conductance of|0.1% of maxi-

1chemical potential of Na . Depolarization stops, in part mum. Considering a peakI of 360 pA/pF and a surfaceNa1because Na channels inactivate, i.e. under sustained to volume ratio of 6.4 pF/pL [54], this steady currentcyt

1 1depolarization, the amplitude of Na current (I ) de- (0.3 A/F) would result in a significant Na influx of 1.4Na1creases.I inactivation effectively switches off Na mM/min. At a restingE of 280 mV this window INa m Na

conductance within a few milliseconds or less (fast in- would be 10–30 times smaller, but would be consistentactivation). with our measurements of tetrodotoxin (TTX) sensitive

1 1Voltage-gated Na channels are composed of a pore- resting Na influx in rabbit ventricular myocytes at 23 andforming a subunit and auxiliaryb subunits. Ten genes 378C (0.14–0.18 mM/min [5,55]). This would require aencodinga subunits have been identified and nine have windowI of |0.01% of maximum. This and theNa

been functionally expressed. Thea subunit consists of four steepness of the activation and availability curves empha-repeat domains (I–IV), each containing six transmembrane sizes that small shifts in activation or inactivation prop-

1segments (S1–S6) and one membrane reentrant domain, erties could lead to substantial sustained Na influx viaconnected by internal and external polypeptides loops [48]. windowI at relatively negativeE .Na m

The S4 segments contain four to eight positively charged During the cardiac AP upstrokeI is very large, butNa1residues, which serve as voltage sensors. An outward very brief. This results in an additional Na influx

movement of these charges under depolarization underlies associated with each AP. Fig. 3C and D showsI and theNa1 1the activation of the channel [49]. Inactivation is mediated Na influx via Na channels during a typical AP, respec-

1by the short intracellular loop connecting domains III and tively. Fig. 3D indicates that|8 mM Na enters the cellIV. Onea subunit is associated with one or two auxiliaryb via I during each AP (6–15mM is probably a reasonableNa

subunits,b1, b2 or b3. These auxiliary subunits modulate range). For a cell contracting at 1 Hz, this means that1channel gating, interact with extracellular matrix and play phasic Na channel activation contributes|0.5 mM/min

1a role as cell adhesion molecules [50]. The primary cardiac to the rate of Na influx. This is surprisingly only about aa subunit isoform (Na 1.5) is much less sensitive to 3-fold increase over the restingI influx.v Na

inhibition by TTX (K |2 mM) and is more sensitive to0.521 1 1 21Cd block than neuronal or skeletal muscle Na chan- 3 .2. Na /Ca exchange

nels.1 1 21 21Fig. 2B shows theE -dependence of Na channels Na /Ca exchange is the major pathway for Cam

activation and availability (or steady-state inactivation).I extrusion from cardiac myocytes. The mammalian NCXNa

activation and availability curves overlap such that there is forms a multigene family of homologous proteins compris-almost no steady state availability whenE is held at ing three isoforms: NCX1, NCX2 and NCX3 (for reviews,m

values where activation is appreciable. This ensures that see Refs. [56–58]). These isoforms share|70% amino1activated Na channels inactivate nearly completely when- acid identity in the overall sequences and thus presumably

1ever activated by depolarization. Na channels require have a very similar molecular structure. NCX1 is the firstrepolarization to recover from inactivation before another NCX cloned and is highly expressed in cardiac muscle andAP can occur. This is a main cause of the electrical brain and to a lesser extent in many other tissues. NCX2refractory period. Recovery ofI requires that the mem- and NCX3 are expressed in a few tissues, such as brain,Na

brane be repolarized to near the diastolic level for a finite but their molecular properties and functions have been lessperiod of time (t|10 ms at2100 mV, 30 ms at280 mV well studied.or 100 ms at272 mV) before the cell is able to fire The cardiac NCX1 consists of 970 amino acids (110

1another AP. Thus Na channels do not recover very kDa). Approximately half of the NCX1 protein constitutesrapidly until repolarization is nearly complete. In this way transmembrane domains, whereas the remaining halfthe long AP contributes to limiting the ability of the cell to (|550 amino acids) forms a large domain exposed on therespond to an early depolarization. Besides these fast cytoplasm. The latter domain does not appear to be

1inactivating Na channels, there are indications that a required for the transport function of NCX1, because a1slowly inactivating, persistent Na current (I ) might mutant lacking most of it still retains exchange activityNa,slow

Dow




ated with ion transport. The density of NCX1 in thesarcolemma of guinea pig ventricular myocytes has been

2estimated to be 250–400 NCX/mm [62].1NCX is generally believed to transport three Na ions in

21exchange for one Ca (for a review, see Ref. [56]), thusone net charge is transported per cycle. However, somerecent studies indicate a 4:1 [63,64] or, under extremelyacidic conditions, a 2:1 stoichiometry [65]. The directionand amount of ion fluxes through NCX depend on the

1 21[Na ] and [Ca ] on both sides of the membrane, as well1as on E . If there is more energy in the inward Nam

1 21gradient (for three Na ions) than in the inward Ca21 21gradient (for one Ca ion), Ca extrusion will be

favored. That is:n(E 2E ). 2(E 2E ), where n isNa m Ca m

the coupling ratio andE andE are the thermodynamicCa Na21 1equilibrium potentials for Ca and Na . Then, theEm

value at which the gradients are exactly equal is thereversal potential (E ) of the I . Hence,E 5Na / Ca NCX Na / Ca

(nE 22E ) /(n22). When E is more positive thanNa Ca m21E , outward I is favored. In other words, CaNa / Ca NCX

1entry /Na exit via the exchanger is favored forE .m21 1E and Ca extrusion/Na entry is favored (inwardNa / Ca

I ) when E ,E . For the normally acceptedn53NCX m Na / Ca

and typical values ofE 570 mV and E 5125 mV,Na Ca

E 5240 mV. Because the resting membrane potentialNa / Ca

(|280 mV) is more negative thanE , the exchangerNa / Ca21 1will work in Ca extrusion/Na influx mode in resting

myocytes. During the action potential, NCX briefly works21 1in Ca entry /Na exit mode, i.e.E .E (see Fig. 3Cm Na / Ca

and discussion below).21

1 Besides being transport substrates, intracellular CaFig. 3. CalculatedI and I mediated Na influx in a ventricularNa NCX 1and Na exert important modulatory effects on NCXmyocyte contracting at 1 Hz. (A) Typical action potential for rabbit2121 21ventricular myocytes. (B) [Ca ] and [Ca ] transients during an AP. activity [57,58]. Binding of intracellular Ca to a highi sm

21 21[Ca ] was calculated from the measured [Ca ] transient as describedsm i affinity site located in the central hydrophilic loop activates1 1

21by Weber et al. [72]. (C) Calculated density of Na current via Na the exchanger ([Ca ] -dependent activation). It should beichannels (I ) and density of NCX current (I ) during an AP using the 21Na NCX1 1 appreciated that this allosteric Ca -regulatory site isI equation of Weber et al. [73]. (D) Net Na fluxes via Na channelsNCX 21

1 distinct from the catalytic or Ca -transport site. While(e I ) and NCX (3e I ), as well as the total Na flux (e I ) duringNa NCX total

the AP. K (Ca) values from 22 to 600 nM have been reported form21this allosteric regulation by [Ca ] , a recent study byi

Weber et al. [66] found a physiologically relevantK (Ca)m

of 125 nM in intact ferret ventricular myocyte during[59]. There were originally thought to be 11 transmem-21 21dynamic [Ca ] changes. This allosteric Ca -activationbrane domains and a very large cytoplasmic hydrophilic i

was rapid (within tens of ms) and so it is probablydomain [60]. However, new topological data suggest thatimportant physiologically during beat-to-beat changes inthere are only nine transmembrane spans [61] and a large

21cellular Ca . There is also an inhibitory regulation at highhydrophilic loop between TMs 5 and 6, with N- and1 1[Na ] [67]. NCX inactivates in a time and [Na ] -depen-C-termini located on the external and internal sides, i i

1dent manner. This [Na ] -dependent inactivation couldrespectively. The N-terminal and C-terminal halves of the i21 21membrane domain contain two internal repeat sequences of prevent excess Ca influx and cellular Ca overload

1 21|40 amino acids, which are designated thea and a under conditions of high [Na ] , where net Ca influx1 2 i1repeats. These repeat sequences are conserved in all might be strongly favored. However, the very high [Na ]i

1members of the NCX family as well as in related cation required to observe Na -dependent inactivation ofINCX

exchangers, and are now thought to be at or near the ion makes this regulation unlikely to be very important in thetranslocation pore. Many details of the NCX1 structure are physiological modulation of NCX.unknown, including the ion transport pathway, the require- NCX exhibits relatively low activity in intact myocytes

21 1ment of the oligomeric protein structure for the function, at rest (partly because [Ca ] is low). Therefore, Nai

and changes in the conformation of the exchanger associ- influx via NCX is rather low in quiescent cells. We

Dow




1measured the rate of Ni-sensitive [Na ] influx upon the same order of magnitude [70]. Of course NCX uses thei1 1 1abrupt Na /K pump inhibition in resting rabbit and rat [Na ] electrochemical gradient (rather than ATP) to

1 21ventricular myocytes. Resting Na influx via NCX was extrude Ca , but the energy still comes from ATP via1 10.28 mM/min in rabbit ventricular myocyte (Fig. 4A) and Na /K -ATPase.

10.64 mM/min in rat. From a functional standpoint, this It is not clear why there is higher Na influx via NCX in1basal rate of Na influx via NCX would extrude 1.5 and rat (vs. rabbit) myocytes, especially considering that

21 13.5 mM Ca /s in rabbit and rat, respectively. This is on [Na ] is higher in rat. The density of outward NCXi21 1the order of estimates of resting Ca leak into ventricular current upon [Na ] removal is higher in rat than rabbito

myocytes (0.5–5mM/s; [68–70]). There is also a sar- [71], so if that holds for inward NCX current as well that21 21colemmal Ca -ATPase, and it is reasonable to infer that may provide a partial explanation. Ca leak into rat

21 21resting Ca extrusion (to match this inward resting Ca myocytes might also be higher (which would require more21 1 21leak) via NCX and the sarcolemmal Ca -ATPase are on Na influx via NCX). Resting SR Ca spark frequency is

also higher in rat versus rabbit [72], and the resulting high21 21 1local [Ca ] may drive local Ca extrusion and Nai

influx via NCX.NCX function during the AP is complicated because

21[Ca ] and E change dramatically. This changes thei m

reversal potential, kinetic and allosteric factors. There are21also spatial gradients of [Ca ] near the sarcolemmai

which complicate this picture [73]. Fig. 3 showsINCX1(panel C) and the Na flux via NCX (derived by inte-

grating I and assuming a 3:1 stoichiometry, panel D)NCX

expected during an AP. Throughout most of the AP there is1 21net Na influx (Ca efflux) via NCX, except for a brief

21period early in the AP before [Ca ] has risen enough toi21 21cause Ca extrusion. During each AP and Ca transient

1 1NCX brings in|32 mM [Na ] . Again, this Na influx isi21required to extrude the|10mM Ca which enters the cell

21 1 21via Ca current during each AP (on a 3 Na :1 CaNCX). For a cell contracting at 1 Hz, this means that NCX

1contributes|1.9 mM/min to the rate of Na influx (about1seven times higher than the resting Na influx rate via

1NCX). As can be seen from Fig. 3, this Na entry is moredistributed over the course of the AP versusI . Figs. 3Na

1and 4 also show that, quantitatively, Na influx via NCXis considerably higher (three to four times) than that which

1occurs via Na channels.As mentioned above, NCX is reversible and can also be

1a Na extrusion mechanism (using the large energy in the21Ca electrochemical gradient). However, this is unlikely

to be a quantitatively important function under normalconditions. This can be appreciated by Fig. 3C where only

1a very small Na efflux occurs very early during the AP.This situation can change during heart failure where lower

21 1Ca transient amplitude, higher [Na ] and prolonged APi1 21can all favor greater Na extrusion (and Ca influx) via

21NCX [55]. On the other hand, whatever Ca enters viaNCX during the AP must also be extruded by NCX during

1the diastolic interval, making net Na extrusion via NCXnegligible. Thus, NCX is probably not very important as a

1Na extrusion mechanism.1Fig. 4. The rate of Na influx in rabbit ventricular myocytes via different

1 1pathways at rest (A), during steady-state stimulation at 1 Hz (B, where 3 .3. Na /H exchange1only Na current and NCX were increased whereas the values measuredin resting myocytes were used for all the other pathways) and in

1Another Na influx pathway in cardiac cells is theconditions of mild acidosis (C, where only NHE and NBC were increased1 1

according to Fig. 5A). Na /H exchanger (NHE), which is important in the

Dow




regulation of pH and cell volume. NHE functions mainlyi1 1as a proton extruder in a 1 Na :1 H stoichiometry

rendering the process electroneutral. There are at least sixNHE isoforms thus far identified, with the NHE1–NHE5apparently restricted to the plasma membrane whereasNHE6 might be the mitochondrial isoform [74] (but seeRef. [75]). NHE1, the so-called housekeeping isoform, isubiquitously distributed in most tissues and is the primaryNHE subtype found in mammalian cardiac cells [76,77].NHE1 is a 110-kDa glycoprotein and consists of twoprincipal domains: a 500-amino acid transmembrane do-main and a 315-amino acid highly hydrophilic carboxyl-terminus cytoplasmic domain (for a review, see Ref. [76]).NHE1 contains 12 membrane-spanning regions that arecritical for the maintenance of NHE1 function in terms ofproton extrusion. The hydrophilic cytoplasmic region playsan important role in modulation of the exchanger, especial-ly through phosphorylation-dependent reactions [78].

It has recently been suggested that NHE1 is localizedprimarily in the intercalated disk region of atrial andventricular myocytes in close proximity to connexin 43,and, to a lesser degree, at the transverse tubular systems[79]. Such NHE1 localization could make NHE especiallyimportant in modulating cell-to-cell communication viagap junctions, and local signaling in T-tubules (althoughthere is little functional data in this regard).

1NHE activity is extremely sensitive to pH . The ex- Fig. 5. (A) The pH -dependence of Na fluxes through cardiac NHE andi i

NBC, as described in Ref. [80]. At physiological pH (7.0–7.2) thechanger is allosterically activated by cytosolic protons, a1contribution of these transporters to Na influx is comparable (Fig. 2A).drop in pH below a threshold level promoting the rapidi 1(B) The rate of Na influx through different pathways in resting rabbitextrusion of acid. Fig. 5A shows the pH -dependence of 1i and rat ventricular myocytes. Na influx was calculated as the initial1 1

1 1 1the rate of ion fluxes (Na influx, H efflux) through NHE slope of [Na ] increase following abrupt Na /K pump inhibition [5].i21 1in guinea-pig ventricular myocytes [80]. NHE is nearly TTX-, Hoe-642- and Ni -sensitive Na influx was calculated as the

difference between the total influx and the influx measured in theinactive at a physiological pH of|7.2, but it activatesi21presence of 30mM TTX, 2 mM Hoe-642, or 5 mM Ni , respectively.quickly when pH decreases. NHE activity in the hearti

seems to be species dependent, with a higher activity(determined as the rate of acid efflux at pH 6.9) in rat (2.8 tive agent with respect to reperfusion injury [84,85]. Whilei

mM/min) versus human (1.1 mM/min) or guinea-pig the events surrounding ischemia/ reperfusion are complex(0.93 mM/min) ventricular myocytes [80,81]. We also and incompletely understood, it seems likely that NHE and

1found that at physiological pH (and 238C) the rate of Na NCX are importantly involved.i

entry via NHE is higher in rat (0.43 mM/min) versus1 2rabbit (0.12 mM/min) ventricle [5]. Although intracellular 3 .4. Na /HCO cotransporter3

1Na can also affect NHE, it is unlikely to be a major1 2regulator within a physiological range (4–16 mM) of Na /HCO cotransport (NBC) also serves as an acid3

1 1 2[Na ] [82]. Proton extrusion through NHE is also in- extruder in cardiac myocytes (Na and HCO influx).i 3

hibited by extracellular acidosis [82]. This may be partly a NBC was first described in the renal proximal tubule of the1 1 2thermodynamic effect (i.e. reducing the [H ] gradient for salamander [86], where it mediates Na and HCO efflux3

proton extrusion), it may also be due to a competition of across the basolateral membrane. Functionally related1 1 2H with Na for binding to the external transport site of cotransporters, which mediate both HCO influx and3

the exchanger [82]. The inhibition of NHE by extracellular efflux, have now been described in many tissues, includingacidosis may limit proton extrusion during ischemia, and the heart (reviewed in Ref. [87]). Cardiac NBC wasrelief of this inhibition upon reperfusion may contribute to originally suggested to have a stoichiometry ranging

1a rapid activation of proton extrusion and Na entry [83]. between 1:1 and 1:2 [88] and thus an equilibrium pHi1A consequent gain in [Na ] during reperfusion could also which, under most conditions, would favor acid extrusion.i

21contribute to Ca overload, diastolic dysfunction and In guinea-pig ventricular myocytes [88] and sheep Purkinjearrhythmogenesis (due to NCX function). This may be fibers [89], acid extrusion via NBC appears to be voltage-why the NHE inhibitor cariporide (Hoe-642) is a protec- insensitive, consistent with an electroneutral cotransporter.

Dow




1 1 1Reports in cat papillary muscle [90] and rat ventricular accounts for 17% of the resting Na influx [97]. Na /K /2myocytes [91] indicate a strong voltage-sensitivity, con- 2Cl cotransport may be important in cell volume regula-

sistent with an electrogenic NBC, probably with a stoi- tion [101].1 2chiometry of 1 Na :2 HCO . Two classes of electrogenic3

(NBCe [92] and NBC4 [93]) as well as one electroneutral 1 213 .5.2. Na /Mg antiporter[94] NBC carrier have been cloned and all three are 21The concentration of free Mg in the cytoplasmpresent in the human heart at the mRNA level. However, 21([Mg ] , 0.5–1 mM) is several hundred times lower thanithe molecular identity of NBCs that mediate acid extrusion 21expected if Mg ions were at electrochemical equilib-in heart cells from different species is not known. NBC is 21rium. Since Mg is a permeant ion across the plas-generally inhibited by stilbenes (DIDS and SITS), al-

malemma, it must be constantly extruded from the cells.though the electroneutral isoform NBCn1 cloned from rat 21The main mechanism that removes Mg from the cells isaorta is only very weakly inhibited by DIDS [95]. 1 21the Na /Mg antiporter. However, its presence in the

Leem et al. [80] measured the pH -dependence of NBCi heart is controversial. Some studies reported the functionalactivity in guinea-pig ventricular myocytes and found that 1 21presence of Na /Mg exchange in guinea-pig and ferretNBC activates roughly linearly with decreasing pH .i ventricle [102] whereas others found no evidence for it inWhether or not allosteric stimulation is involved is not

guinea-pig [103] and chicken heart [104]. Such studies are1known. Fig. 5 shows the pH -dependence of Na influx 21 1i generally based on changes in [Mg ] when [Na ] isi ovia NBC using the data from Ref. [80] and assuming a 1:1 21varied. However, Mg is heavily buffered in cells and itsstoichiometry of the cotransport (as reported for guinea-pig

membrane permeability is rather low, therefore theventricular myocytes). It is notable that for relatively 21 1changes in [Mg ] are minimal. Also, changes in [Na ]i omodest acid loads (i.e. to pH 6.9), both NHE and NBC are 21i affect [Ca ] and pH and these in turn will affecti istimulated about equally. Thus, at pH values within the 21 21i intracellular Mg buffering. Thus, the changes in [Mg ]inormal physiological range, both transporters are about

in such protocols might not be mediated solely by theequally important for mediating acid extrusion and they 1 21Na /Mg exchanger. Tashiro and Konishi [105] used rat1contribute equally to Na influx. For larger intracellular 21ventricular myocytes rendered Mg -permeable withacid challenges, NHE removes more acid than does NBC. 21ionomycin in high [Mg ] conditions to facilitate passiveo

21 21Mg influx. The rate of [Mg ] rise was significantly1 i3 .5. Other Na influx pathways 1smaller in the presence of 140 mM external Na than in itsabsence. Washout of ionomycin and lowering extracellular1 1 23 .5.1. Na /K /2Cl cotransporter 21 21 1Mg caused rapid decline of [Mg ] when [Na ] is 1401 1 2 i oThe Na /K /2Cl cotransporter (NKCC) was first 21mM. This Mg efflux was completely inhibited by

detected in Ehrlich ascites tumor cells [96]. Since then, the 1withdrawal of extracellular Na and was largely attenuatedcotransporter has been found in a variety of tissues, 1 21by imipramine, a known inhibitor of Na /Mg exchange.including the heart [97–99]. Two NKCC isoforms have 1 21At 140 mM external Na , the rate of Mg transportbeen cloned: NKCC1, found in the membrane of a variety 21through the exchanger was 25–50mM/min. Mg trans-of cell types, and NKCC2, identified thus far only in the 1port depended on [Na ] according to a Hill-type curve,omedullary regions of the kidney [99]. NKCC is inhibited

with the mid-point at|80 mM and a Hill coefficient of 2.by loop diuretics, such as furosemide and bumetanide. In

This suggests that the exchanger has a stoichiometry of 21 1 2all tissues investigated, Na /K /2Cl cotransport is 1 21Na :1 Mg and therefore is electroneutral. A moreelectroneutral, with a stoichiometry of 1:1:2 and with an 1 1 2complicated stoichiometry (2 Na12 K 12 Cl :11 1 2absolute requirement that Na , K and Cl are all present 21Mg ) has been proposed for the exchanger in squid gianton the same side of the membrane. In chick cardiac cells,

axons and barnacle muscle cells [106].2 1the [Cl ] and [Na ] for half-maximal stimulation of the 1There may also be other Na flux pathways that wecotransporter are 59 and 40 mM, respectively [97]. Net 1have not discussed (e.g. background Na channels), but wetransport may occur into or out of the cells, the magnitude

feel that these are likely to be minor quantitatively. We canand direction of this transport being determined by the sum 1account for most of the measured Na fluxes by theof the chemical potential gradients of the transported ions.

mechanisms mentioned above (Fig. 4).1 1 2In physiological conditions, Na /K /2Cl cotransport is2inwardly directed and thus maintains the intracellular [Cl ]

2([Cl ] ) above its electrochemical equilibrium (|15 mM).i2 1Intracellular Cl is an important regulator of the co- 4 . Balance of Na influx and efflux

2transporter, where high [Cl ] inhibits ion fluxes in bothi

directions. Gillen and Forbush [100] found a very steep 4 .1. Quiescencerelationship between bumetanide-sensitive ion flux and

2 1[Cl ] in HEK-293 cells transfected with human NKCC1. At steady state, Na influx and efflux must be equal.i1 1 2 1In chick cardiac cells, the Na /K /2Cl cotransporter Fig. 4A shows the rate of resting Na influx through the

Dow




pathways discussed above. The values are based in part on (see below). It will be valuable for additional quantitativeour measurements in rabbit ventricular myocytes [5] and data to be measured for these estimates to be further

1some extrapolations from guinea-pig data [80,102] and refined. In conclusion, while the various Na influxcorrespond to mammalian species with lower resting pathways differ little at rest with only NCX being much

1 1[Na ] (i.e. other than rat and mouse). Adding the contri- larger than the others (and only by|2-fold), Na channelsi1bution of each pathway, the total Na influx in resting and NCX are likely to become much more dominant

1myocytes is 0.88 mM/min. Na influx pathways un- during normal electrical and contractile activity (in this1accounted for in Fig. 4A might increase this rate by|0.1 example providing 85% of the total Na influx).

1 1mM/min [5]. Fig. 2A indicates that the Na /K pump1 1will extrude this amount of Na if resting [Na ] is 5.3i

1 4 .4. AcidosismM (5.5 mM if the additional 0.1-mM/min Na influx isconsidered). So, this principle of flux balance is also met

1 Acidosis is a major consequence of myocardial ischemiawith these quantitative resting Na flux estimates.and contributes to the ischemic decline in force. The major

1 negative inotropic effect is mainly due to intracellular4 .2. Why is resting [Na ] higher in rat versus rabbit?i

versus extracellular acidosis and is caused in large part by21 211 decreased myofilament Ca sensitivity. Ca transientNow we can revisit the question of why resting [Na ]i

amplitude during abrupt acidosis can be initially increased,in rat is higher than in rabbit ventricular myocytes. In1 1 unchanged or decreased [107–109]. However, in almost allreference to Fig. 2A we concluded that resting Na /K -

1 reports there is then a progressive increase in twitchATPase-mediated Na efflux was higher in rat than rabbit211 D[Ca ] and this causes a partial recovery of contractions.and we inferred that this must be matched by higher Na i

211 1 Acidosis can also gradually increase diastolic [Ca ]influx. We evaluated resting Na influx by abrupt Na / i1 [110].K -ATPase block and measurement of the initial rate of

1 Low pH stimulates proton extrusion via NHE and NBCrise of [Na ] [5]. Indeed, as Fig. 5B shows, we found the ii1 (Fig. 5A), especially when pH is relatively normal. Thus,resting Na influx to be 2.4 times higher in rat versus o

11 resting Na influx is enhanced during acidosis. Fig. 4Crabbit ventricular myocytes. Several Na influx pathways1shows estimations of the rate of resting Na influx throughappear to contribute to different extents. For example,

1 various pathways in the case of a mild acidosis (pH 6.9).TTX-sensitive Na influx was almost five times higher in i1 With respect to Fig. 4A only NHE and NBC wererat, while the Ni-sensitive Na influx (presumably mainly

increased as predicted from Fig. 5A. Of course there mayNCX) was only about two times higher in rat. These11 also be inhibition of other Na influx pathways as wellresults confirm that the likely reason [Na ] is higher in rati

1 (e.g. NCX andI ). Our expectation is that NHE and NBCis because resting Na influx is higher. This would Na1 1 1 increase greatly and in this case account for 75% of restingstimulate higher Na /K -ATPase-mediated Na efflux

1 1 11 1 Na influx. Furthermore, the Na /K pump is partially(and possibly Na /K -ATPase expression) until a newinhibited at low pH [111]. These two factors lead to ansteady state balance is achieved at higher levels of both i

11 increase in [Na ] [112], which may contribute to the slowNa influx and efflux. This scenario might also hold for i1 recovery of contractility via a shift in NCX and increase inother species where [Na ] is high (where mouse is likei

211 [Ca ] . NCX is also inhibited by low pH [113]. This mayrat) and where [Na ] is low (where guinea-pig is like i ii21rabbit). limit the ability of NCX to extrude Ca (especially at

1high [Na ] ). As with cardiac glycosides, this may alsoi21 214 .3. Stimulation contribute to increased SR Ca content and larger Ca

21transients seen during acidosis, but could also lead to Ca1As described above Na influx is increased in contract- overload and consequent arrhythmias.

1ing myocytes, at least because of increased Na entry via During severe hypoxia (associated with ischemia) there1 1I and NCX (Fig. 3). Fig. 4B shows estimates of Na is a large rise in [Na ] which can reach 40 mM inNa i1influx in myocytes contracting at 1 Hz. The total Na prolonged anoxia. Eigel and Hadley [114] evaluated the

1influx in contracting cells is increased by|3.7-fold (vs. cellular basis of the Na gain. The main mechanisms of1 1rest) to |3.5 mM/min. We only increasedI and NCX net Na gain during anoxia were TTX sensitive NaNa

1 1(by |3 and 7-fold as discussed above), but certainly cannot channels and Hoe-642-sensitive Na /H exchange. TTX1 1exclude the possibility of increased Na influx by other and Hoe-642 each blocked|50% of the Na gain during

1pathways (e.g. NHE and NBC). This higher Na influx simple anoxia. For anoxia plus extracellular acidosis the1 1would require a matching increase in Na efflux via the Na gain was blocked by TTX, but not by Hoe-642.

1 1 1Na /K pump and this would occur at|11 mM [Na ] However, when anoxia or acidosis (pH 6.85) was coupledi1 1 1(Fig. 2A). This is higher than the [Na ] we measure in with high [K ] (10 mM) to simulate ischemia, the Nai o

rabbit at 1-Hz stimulation (|8 mM). Part of this difference gain was suppressed almost completely by Hoe-642, but1 1could reflect restricted subsarcolemmal diffusion of Na not by TTX. It is unclear why Na channel influx would

Dow




1 1occur in the cases with normal [K ] (while cells are [Na ] regulation is unclear. While in heart, mitochondriao i

quiescent), but anoxia and ischemic metabolites have been occupy|35% of the total cell volume it should be kept in1shown to produce persistent Na channel opening in mind that mitochondria are a totally intracellular compart-

ventricular myocytes [115,116]. Since depolarization pre- ment and as such are a limited source or sink. However,1 1 1 1vented the TTX-sensitive Na gain, it is possible that since [Na ] rises with [Na ] , Na entry into mito-m i

1anoxia allows a tiny windowI at resting E , but chondria may serve as a buffer for [Na ] , and could limitNa m i1 1inactivation prevails upon depolarization in high [K ] . changes in [Na ] during pathophysiological conditions.o i

1Thus, during ischemia [Na ] rise may be mediated byi1 1 1Na /H exchange and Na channel entry [117].

16 . Intracellular Na gradients15 . Mitochondrial Na transport

Several studies have provided evidence for the presence11 1 of a [Na ] gradient between the subsarcolemmal spaceFig. 1B shows mitochondrial Na movements. [Na ]i

121 ([Na ] ) and bulk cytosol in arterial smooth muscle [127]can importantly regulate mitochondrial [Ca ], because sm21 1 and cardiac myocytes [128–132]. The existence of suchmitochondrial Ca extrusion occurs mainly via a Na /

121 1 intracellular [Na ] gradients implies restricted diffusionCa antiporter, which may be electroneutral (2 Na :121 1 with respect to transport rate. Wendt-Gallitelli et al. [133]Ca ), but might also be.2:1 [118–120]. The [Na ]i

11 21 used electron probe microanalysis to measure [Na ] in adependence of this Na /Ca antiporter is sigmoidal with1 volume within 20 nm of the inner side of the sarcolemmahalf-maximal activity at |5–8 mM Na , making this

11 of guinea-pig ventricular myocytes. [Na ] increased upsystem sensitive to physiological [Na ] changes smi1 to 40 mM during stimulation of the cells, with no change[118,121]. While variations in bulk cytoplasmic [Na ]

1 1in bulk [Na ] . This steep subsarcolemmal [Na ] gradientduring the cardiac cycle are probably insufficient to cause i21 dissipated, but only within minutes of the end of stimula-rapid release of mitochondrial Ca , large changes in

11 21 tion. This is surprisingly slow dissipation of such a [Na ][Na ] can induce substantial mitochondrial Ca release in21 1 gradient, and this method measures total rather than freevitro [118]. TheV value for Ca extrusion (and Namax

1 11 21 [Na ]. This could reflect increased binding of Na in aentry) via mitochondrial Na /Ca exchange in the heart1subsarcolemmal space, which could in turn alter Nais |15 nmol Ca/mg protein per min [122] or|0.9 mM/

1 21 diffusion.min. This gives the mitochondrial Na /Ca antiport1 11 A transient peak in the Na /K pump current has beencomparable flux capacity to some sarcolemmal Na trans-

observed in voltage-clamped ventricular myocytes, whenporters.1 11 1 21 the Na /K pump was abruptly reactivated after a periodNa entering mitochondria through the Na /Ca

1 1 of pump blockade [1,128,131].I then decayed over aantiporter is then extruded via a Na /H exchange pump

few minutes to a steady-state level. ThisI sag might besystem, driven by the electrochemical gradient of protons pump1the result of lowering [Na ] because of the pump(which is created by the electron transport chain). The sm

11 1 activity (even when global [Na ] was presumed ormitochondrial Na /H exchanger is reversible and seems i1 measured to be little changed). PeakI also increased into be symmetrical in its interaction with Na [123]. The pump

1 11 response to a train of depolarizations prior to Na /KK for Na transport is|26 mM at an intramitochondrialm1pump reactivation [134], suggesting that activation of NapH of 8.0 [124] and decreases with more alkaline mito-

11 channels increases [Na ] . Lipp and Niggli [135] havechondrial pH. Matrix protons compete with Na for sm21also observed thatI can activate Ca influx via NCX,binding to a common site and thus inhibit the exchanger Na

1inferring that I raised [Na ] . While some preliminarywith a mid-point of |30 nM (pH|7.5) [125]. The mito- Na sm11 1 data imply thatI may not modify local [Na ] suffi-chondrial Na /H antiport shares many properties with Na sm

1 1ciently to alter NCX or Na /K -ATPase [136,137], thethe sarcolemmal NHE, however it differs in some aspects1whole issue of [Na ] gradients requires further study.such as the sensitivity to amiloride analogues and relative i

1 1Su et al. [134] showed that abrupt Na /K -ATPaselack of regulation [123].1 1 inhibition in mouse ventricular myocytes increases theMitochondrial [Na ] ([Na ] ) appears to be lower thanm

211 1 1 efficacy of a givenI to trigger SR Ca release (Fig. 6).[Na ] [8,126]. [Na ] increases linearly with [Na ] and Cai m i11 This was interpreted as a prevention of Na extrusion inthe [Na ] gradient across the mitochondrial inner mem-

1the SR junctional cleft allowing local cleft [Na ] to risebrane depends on the energetic state of the mitochondria sm1 21 1 1[126]. In isolated mitochondria, the Na gradient was and favor Ca entry via NCX. Abrupt Na /K pump

1equal to the electrochemical H gradient over a wide range inhibition was also shown to slow the decline of caffeine-1 21of external [Na ] and changed rapidly in response to induced Ca transients andI [132]. Goldhaber et al.NCX

1 1changes in matrix pH [126], suggesting that Na dis- [138] also found that abrupt removal of [Na ] caused ano21tribution across the inner mitochondrial membrane is increase in resting Ca spark frequency, and they attribu-

1 1 21effectively determined by the equilibrium of the Na /H ted this to a blockade of tonic Ca removal from the1 21antiporter. The quantitative role of mitochondria in overall junctional cleft by Na /Ca exchange. All of these data

Dow




[2] Gray RP, McIntyre H, Sheridan DS, Fry CH. Intracellular sodiumand contractile function in hypertrophied human and guinea-pig

¨myocardium. Pflugers Arch 2001;442:117–123.21[3] Yao A, Su Z, Nonaka A et al. Abnormal myocyte Ca homeostasis

in rabbits with pacing-induced heart failure. Am J Physiol1998;275:H1441–H1448.

[4] Harrison SM, McCall E, Boyet MR. The relationship betweencontraction and intracellular sodium in rat and guinea-pig ventricularmyocytes. J Physiol 1992;449:517–550.

1[5] Despa S, Islam MA, Pogwizd SM, Bers DM. Intracellular [Na ]i1and Na -pump rate in rat and rabbit ventricular myocytes. J Physiol

2002;539:133–143.[6] Levi AJ, Lee CO, Brooksby P. Properties of the fluorescent sodium

indicator SBFI in rat and rabbit cardiac myocytes. J CardiovascElectrophysiol 1994;5:241–257.

[7] Shattock MJ, Bers DM. Rat versus rabbit ventricle: Ca flux andintracellular Na assessed by ion-selective microelectrodes. Am J

1 1Fig. 6. Effects of abrupt inhibition of the Na /K pump (Pump Inhib.) Physiol 1989;256:C813–C822.21on [Ca ] transients and membrane currents. The left panel shows the [8] Donoso P, Mill JG, O’Neil SC, Eisner DA. Fluorescence measure-i

last of a series of eight voltage-clamp steps (280 to 0 mV) to induce a ments of cytoplasmic and mitochondrial sodium concentration in rat21stable [Ca ] transient magnitude (a), membrane currents (c, including ventricular myocytes. J Physiol 1992;448:493–509.i

21 1I and I ) and [Ca ] transients (d). Abrupt Pump Inhib. (by [K ] [9] Yao AS, Su Z, Nonaka A et al. Effects of overexpression of theNa Ca i o1 21 21removal using a rapid solution switcher, b) for 1.5 s immediately before Na –Ca exchanger on [Ca ] transients in murine ventricular

and during the test pulse was evidenced by the inward shift in holding myocytes. Circ Res 1998;82:657–665.21 [10] Cook SJ, Chamunorwa JP, Lancaster MK, O’Neill SC. Regionalcurrent (c). With Pump Inhib. peak [Ca ] was significantly higher (d andi

1 differences in the regulation of intracellular sodium and in actionbar graph). Results are mean6S.E.M. of 16 cells. Pipette [Na ] was 15¨potential configuration in rabbit left ventricle. Pflugers ArchmM. (From Su et al. [134])

1997;433:515–522.[11] Lancaster MK, Bennet DL, Cook SJ, O’Neill SC. Na/K pumpa

subunit expression in rabbit ventricle and regional variations of1suggest that local or junctional microdomain [Na ] andsm ¨intracellular Na regulation. Pflugers Arch 2000;440:735–739.21 1 1 1[Ca ] may be controlled by local Na /K -ATPase, Na [12] Bers DM. Excitation–contraction coupling and cardiac contractilei

channels and NCX, such that, when the system is not at force. Dordrecht, The Netherlands: Kluwer, 2001.1 [13] Lee CO, Dagostino M. Effect of strophanthidin on intracellular Nasteady-state, they differ significantly from bulk [Na ] andi

21 1 1 ion activity and twitch tension on constantly driven canine cardiac[Ca ] . Thus, a normally functioning Na /K -ATPasei Purkinje fibers. Biophys J 1982;40:185–198.1 21may maintain [Na ] and [Ca ] in the junctional cleft ati i [14] Harrison SM, Bers DM. The influence of temperature on thelower values than bulk cytosolic, and this may limit CICR. calcium sensitivity of the myofilaments of skinned ventricularThese issues, however, merit further study and quantitative muscle from the rabbit. J Gen Physiol 1989;93:411–427.

[15] Lauger P. Electrogenic ion pumps. Sunderland, Massachusetts:characterization. Some of these results are difficult toSinauer Associates, Inc, 1991.reconcile with expected ion diffusional characteristics in

[16] Blanco G, Mercer RW. Isozymes of the Na-K-ATPase: hetero-cells or are counterbalanced by other results. For example, geneity in structure, diversity in function. Am J Physiol

1ultrastructural studies also indicate that Na channels, 1998;275:F633–F650.21NCX and SR Ca release channels in rat ventricular [17] Shull GE, Schwartz A, Lingrel JB. Amino-acid sequence of the

1 1catalytic subunit of the (Na ,K )-ATPase deduced from a com-myocytes are not co-localized with each other [139]. On1 1 plementary DNA. Nature 1985;316:691–695.the other hand, the Na /K pump and NCX are co-

[18] Kawakami K, Noguchi S, Noda M et al. Primary structure of thelocalized in smooth muscle [140]. It appears that there is 1 1

a-subunit ofTorpedo californica (Na -K )-ATPase deduced from1much still to learn about cellular Na regulation. We hope cDNA sequence. Nature 1985;316:733–736.

1that this integrative overview of cardiac Na regulation [19] Mercer RW, Biemesderfer D, Bliss DP, Collins JH, Forbush B.Molecular cloning and immunological characterization of theg-helps to provide an initial semi-quantitative framework forpolypeptide, a small protein associated with the Na,K-ATPase. Jfuture work and subsequent articles in this focused issue.Cell Biol 1993;121:579–586.

1 1[20] Shamraj OI, Lingrel JB. A putative fourth Na ,K -ATPaseasubunit gene is expressed in testis. Proc Natl Acad Sci USA1994;91:12952–12956.A cknowledgements

[21] Zahler R, Gilmore-Hebert M, Baldwin JC, Franco K, Benz Jr. EJ.Expression ofa isoforms of the Na,K-ATPase in human heart.

This work was supported by NIH grants HL64098 and Biochim Biophys Acta 1993;1149:1890194.HL64724 (D.M.B.) and HL53773 (W.H.B.). [22] Wang J, Schwinger RH, Frank K et al. Regional expression of

1 21sodium pump subunit isoforms and Na –Ca exchanger in thehuman heart. J Clin Invest 1996;98:1650–1658.

[23] McDonough AA, Zhang Y, Shin V, Frank JS. Subcellular distributionR eferences of Na pump isoform subunits in mammalian cardiac myocytes. Am J

Physiol 1996;270:C1221–C1227.[1] Despa S, Bers DM. Simultaneous measurements of [Na] and Na/K [24] Kim CH, Fan TH, Kelly PF et al. Isoform-specific regulation ofi

pump current in rabbit ventricular myocytes. Biophys J 2002;82:68, myocardial Na,K-ATPase alpha-subunit in congestive heart failure.(abstract). Role of norepinephrine. Circulation 1994;89:313–320.

Dow




[25] Zahler R, Gilmore-Hebert M, Sun W, Benz Jr. EJ. Na,K-ATPase expression and Na pump activity. Eur J Pharmacol 1998;351:113–isoform gene expression in normal and hypertrophied dog heart. 119.Basic Res Cardiol 1996;91:256–266. [48] Catterall WA. Cellular and molecular biology of voltage-gated Na

1 1[26] Ng YC, Book CB. Expression of Na -K -ATPasea1 and a3- channels. Physiol Rev 1992;72:S15–S48.isoforms in adult and neonatal ferret hearts. Am J Physiol ¨[49] Stuhmer W, Conti F, Suzuki H et al. Structural parts involved in1992;263:H1430–H1436. activation and inactivation of the sodium channel. Nature

[27] Baba A, Yoshikawa T, Nakamura I, Iwata M, Wainai Y, Ogawa S. 1989;339:597–603.1 1Isoform-specific alterations in cardiac and erythrocyte Na ,K - [50] Isom LL, De Jongh KS, Catterall WA. Auxiliary subunits of voltage-

ATPase activity induced by norepinephrine. J Card Fail gated ion channels. Neuron 1994;12:1183–1194.1998;4:333–341. [51] Saint DA, Ju YK, Gage PW. A persistent sodium current in rat

[28] Gao J, Wymore R, Wymore RT et al. Isoform-specific regulation of ventricular myocytes. J Physiol 1992;453:219–231.the sodium pump by alpha- and beta-adrenergic agonists in the [52] Maltsev VA, Sabbah HN, Higgins RS, Silverman N, Lesch M,guinea-pig ventricle. J Physiol 1999;516:377–383. Undrovinas AI. Novel, ultraslow inactivating sodium current in

[29] Orlowski J, Lingrel JB. Tissue specific and developmental regula- human ventricular cardiomyocytes. Circulation 1998;98:2545–2552.tion of rat Na,K-ATPase catalytica isoform andb subunit mRNAs. [53] Undrovinas AI, Maltsev VA, Sabbah HN. Repolarization abnor-J Biol Chem 1988;263:10436–10442. malities in cardiomyocytes of dogs with chronic heart failure: role of

[30] Lucchesi PA, Sweadner KJ. Postnatal changes in Na,K-ATPase sustained inward current. Cell Mol Life Sci 1999;55:494–505.isoform expression in rat cardiac ventricle. J Biol Chem [54] Satoh H, Delbridge LM, Blatter LA, Bers DM. Surface:volume1991;266:9327–9331. relationship in cardiac myocytes studied with confocal microscopy

[31] Charlemagne D, Orlowski J, Oliviero P et al. Alteration of Na,K- and membrane capacitance measurements: species-dependence andATPase subunit mRNA and protein levels in hypertrophied rat heart. developmental effects. Biophys J 1996;70:1494–1504.

1J Biol Chem 1994;269:1541–1547. [55] Despa S, Islam MA, Pogwizd SM, Bers DM. Intracellular Na[32] Semb SO, Lunde PK, Holt E, Tonnessen T, Christensen G, Sejersted concentration is elevated in heart failure, but Na/K-pump function is

OM. Reduced myocardial Na-K pump capacity in congestive heart unchanged. Circulation 2002;105:2543–2548.failure following myocardial infarction in rats. J Mol Cell Cardiol [56] Blaustein MP, Lederer WJ. Sodium/calcium exchange: its physio-1998;30:1311–1328. logical implications. Physiol Rev 1999;79:763–854.

[33] Pierre S, Paganelli F, Sennoune S et al. RT-PCR detection of the [57] Philipson KD, Nicoll DA. Sodium–calcium exchange: a molecularNa,K-ATPaseb -isoform in human heart. Cell Mol Biol (Noisy-le- perspective. Annu Rev Physiol 2000;62:111–133.3

1 21grand) 2001;47:261–264. [58] Shigekawa M, Iwamoto T. Cardiac Na –Ca exchange: molecular1[34] Wang J, Velotta JB, McDonough AA, Farley RA. All human Na - and pharmacological aspects. Circ Res 2001;88:864–876.

1K -ATPasea subunit isoforms have a similar affinity for cardiac [59] Matsuoka S, Nicoll DA, Reilly RF, Hilgemann DW, Philipson KD.glycosides. Am J Physiol 2001;281:C1336–C1343. Initial localization of regulatory regions of the cardiac sarcolemmal

1 21[35] James PF, Grupp IL, Grupp G et al. Identification of a specific role Na –Ca exchanger. Proc Natl Acad Sci USA 1993;90:3870–for the Na,K-ATPasea2 isoform as a regulator of calcium in the 3874.heart. Mol Cell 1999;3:555–563. [60] Philipson KD, Nicoll DA. Molecular and kinetic aspects of sodium–

1[36] Juhaszova M, Blaustein MP. Na pump low and high ouabain calcium exchange. Int Rev Cytol 1993;137C:199–227.affinity a subunit isoforms are differently distributed in cells. Proc [61] Nicoll DA, Ottolia M, Lu L, Lu Y, Philipson KD. A new topological

1 21Natl Acad Sci USA 1997;94:1800–1805. model of the cardiac sarcolemmal Na –Ca exchanger. J Biol1[37] Glitsch HG, Tappe A. Change of Na pump current reversal Chem 1999;274:910–917.

potential in sheep cardiac Purkinje cells with varying free energy of [62] Hilgemann DW, Nicoll DA, Philipson KD. Charge movement during1 1 21ATP hydrolysis. J Physiol 1995;484:605–616. Na translocation by native and cloned cardiac Na /Ca ex-

[38] Hilgemann DW, Nagel GA, Gadsby DC. Na/K pump current in changer. Nature 1991;352:715–718.1 21giant membrane patches excised from ventricular myocytes. In: [63] Fujioka Y, Komeda M, Matsuoka S. Stoichiometry of Na –Ca

Kaplan JH, De Weer P, editors, The sodium pump: recent develop- exchange in inside-out patches excised from guinea-pig ventricularments, New York: Rockefeller University Press, 1991, pp. 543–547. myocytes. J Physiol 2000;523:339–351.

1 1 1 21[39] Friedrich T, Bamberg E, Nagel G. Na ,K -ATPase pump currents [64] Dong H, Dunn J, Lytton J. Stoichiometry of the cardiac Na /Cain giant excised patches activated by an ATP concentration jump. exchanger NCX1.1 measured in transfected HEK cells. Biophys JBiophys J 1996;71:2486–2500. 2002;82:1943–1952.

1 1 1 21[40] Glitsch HG, Tappe A. The Na /K pump of cardiac Purkinje cells [65] Egger M, Niggli E. Paradoxical block of the Na –Ca exchanger¨is preferentially fuelled by glycolytic ATP production. Pflugers Arch by extracellular protons in guinea-pig ventricular myocytes. J

1993;422:380–385. Physiol 2000;523:353–366.[41] Glitsch HG. Electrophysiology of the sodium-potassium-ATPase in [66] Weber CR, Ginsburg KS, Philipson KD, Shannon TR, Bers DM.

cardiac cells. Physiol Rev 2001;81:1791–1826. Allosteric regulation of Na/Ca exchange current by cytosolic Ca in1 1[42] Therien AG, Blostein R. K /Na antagonism at cytoplasmic sites intact cardiac myocytes. J Gen Physiol 2001;117:119–131.

1 1of Na -K -ATPase: a tissue-specific mechanism of sodium pump [67] Hilgemann DW, Matsuoka S, Nagel GA, Collins A. Steady-state andregulation. Am J Physiol 1999;277:C891–C898. dynamic properties of cardiac sodium–calcium exchange. Sodium-

[43] Nakao M, Gadsby DC. [Na] and [K] dependence of the Na/K pump dependent inactivation. J Gen Physiol 1992;100:905–932.current-voltage relationship in guinea pig ventricular myocytes. J [68] Diaz ME, Trafford AW, O’Neill SC, Eisner DA. Measurement of

21 21Gen Physiol 1989;94:539–565. sarcoplasmic reticulum Ca content and sarcolemmal Ca fluxes21[44] Gadsby DC, Kimura J, Noma A.Voltage dependence of Na/K pump in isolated rat ventricular myocytes during spontaneous Ca

current in isolated heart cells. Nature 1985;315:63–65. release. J Physiol 1997;501:3–16.21[45] Rakowski RF, Gadsby DC, De Weer P. Voltage dependence of the [69] Trafford AW, Diaz ME, Negretti N, Eisner DA. Enhanced Ca

21Na/K pump. J Membr Biol 1997;155:105–113. current and decreased Ca efflux restore sarcoplasmic reticulum21[46] Zahler R, Zhang Z-T, Manor M, Boron WF. Sodium kinetics of Ca content after depletion. Circ Res 1997;81:477–484.

Na,K-ATPasea isoforms in intact transfected HeLa cells. J Gen [70] Su Z, Bridge JHB, Philipson KD, Spitzer KW, Barry WH. Quantita-Physiol 1997;110:201–213. tion of Na/Ca exchanger function in single ventricular myocytes. J

1 1[47] Liu X, Songu-Mize E. Effect of Na on Na ,K -ATPase a subunit Mol Cell Cardiol 1999;31:1125–1135.

Dow




[71] Choi HS, Trafford AW, Eisner DA. Measurement of calcium entry [94] Pushkin A, Abuladze N, Lee I, Newman D, Hwang J, Kurtz I.¨and exit in quiescent rat ventricular myocytes. Pflugers Arch Cloning, tissue distribution, genomic organization, and functional

2000;440:600–608. characterization of NBC3, a new member of the sodium bicarbonate21[72] Satoh H, Blatter LA, Bers DM. Effects of [Ca] , Ca load and rest cotransporter family. J Biol Chem 1999;274:16569–16575.i

21on Ca spark frequency in ventricular myocytes. Am J Physiol [95] Boron WF. Sodium-coupled bicarbonate transporters. J Pancreas1997;272:H657–668. 2001;2:176–181.

[73] Weber CR, Piacentino III V, Ginsburg KS, Houser SR, Bers DM. [96] Geck P, Pietrzyk C, Burckhardt BC, Pfeiffer B, Heinz E. Electrically1 21 21

1 1 2Na –Ca exchange current and submembrane [Ca ] during the silent cotransport on Na , K and Cl in Ehrlich cells. Biochimcardiac action potential. Circ Res 2002;90:182–189. Biophys Acta 1980;600:432–447.

[74] Numuta M, Petrecca K, Lake N, Orlowski J. Identification of a [97] Frelin C, Chassande O, Lazdunski M. Biochemical characterization1 1mitochondrial Na /H exchanger. J Biol Chem 1998;273:6951– of the Na/K/Cl co-transport in chick cardiac cells. Biochem

6959. Biophys Res Commun 1986;134:326–331.1 1[75] Brett CL, Wei Y, Donowitz M, Rao R. Human Na /H exchanger [98] Liu S, Jacob R, Piwnica-Worms D, Lieberman M. (Na1K12Cl)

isoform 6 is found in recycling endosomes of cells, not in mito- cotransport in cultured embryonic chick heart cells. Am J Physiolchondria. Am J Physiol 2002;282:C1031–1041.

1987;253:C721–C730.1 1[76] Orlowski J, Grinstein S. Na /H exchangers of mammalian cells. J[99] Russell JM. Sodium-potassium-chloride cotransport. Physiol Rev

Biol Chem 1997;272:22373–22376.2000;80:211–276.

[77] Karmazyn M, Gan XT, Humphreys RA, Yoshida H, Kusumoto K.1 1 [100] Gillen CM, Forbush III B. Functional interaction of the K-ClThe myocardial Na –H exchange: structure, regulation, and its

cotransporter (KCC1) with the Na-K-Cl cotransporter in HEK-293role in heart disease. Circ Res 1999;85:777–786.cells. Am J Physiol 1999;276:C328–C336.´[78] Counillon L, Pouyssegur J. Structure-function studies and molecular

[101] Clemo HF, Feher JJ, Baumgarten CM. Modulation of rabbitregulation of the growth factor activatable sodium–hydrogen ex-1 1 2ventricular cell volume and Na /K /2Cl cotransport by cGMPchanger (NHE-1). Cardiovasc Res 1995;29:147–154.

and atrial natriuretic factor. J Gen Physiol 1992;100:89–114.[79] Petrecca K, Atanasiu R, Grinstein S, Orlowski J, Shrier A. Subcellu-1 1 [102] Fry CH. Measurement and control of intracellular magnesium ionlar localization of the Na /H exchanger NHE1 in rat myocardium.

concentration in guinea pig and ferret ventricular myocardium.Am J Physiol 1999;276:H709–H717.Magnesium 1986;5:306–316.[80] Leem CH, Lagadic-Gossmann D, Vaughan-Jones RD. Characteriza-

21[103] Buri A, Chen S, Fry CH et al. The regulation of intracellular Mgtion of intracellular pH regulation in the guinea-pig ventricular21in guinea-pig heart, studied with Mg -selective microelectrodesmyocytes. J Physiol 1999;509:487–496.

and fluorochromes. Exp Physiol 1993;78:221–233.[81] Yokoyama H, Gunasegaram S, Harding SE, Avkiran M. Sarcolem-1 1 [104] Freudenrich CC, Murphy E, Liu S, Lieberman M. Magnesiummal Na /H exchanger activity and expression in human ventricu-

homeostasis in cardiac cells. Mol Cell Biochem 1992;114:97–103.lar myocardium. J Am Coll Cardiol 2000;36:534–540.1 1[82] Wu ML, Vaughan-Jones RD. Interaction between Na and H ions [105] Tashiro M, Konishi M. Sodium gradient-dependent transport of

on Na–H exchange in sheep cardiac Purkinje fibers. J Mol Cell magnesium in rat ventricular myocytes. Am J PhysiolCardiol 1997;29:1131–1140. 2000;279:C1955–C1962.

[83] Lazdunski M, Frelin C, Vigne P. The sodium/hydrogen exchange [106] Rasgado-Flores H, Gonzalez-Serratos H. Plasmalemmal transportsystem in cardiac cells: its biochemical and pharmacological prop- of magnesium in excitable cells. Front Biosci 2000;5:D866–D879.erties and its role in regulating internal concentrations of sodium and [107] Orchard CH, Kentish JC. Effects of changes of pH on theinternal pH. J Mol Cell Cardiol 1985;17:1029–1042. contractile function of cardiac muscle. Am J Physiol

[84] Xue YX, Aye NN, Hashimoto K. Antiarrhythmic effects of 1990;258:C967–C981.1 1 21HOE642, a novel Na –H exchanger inhibitor, on ventricular [108] Hulme JT, Orchard CH. Effect of acidosis on Ca uptake and

arrhythmias in animal hearts. Eur J Pharmacol 1996;317:309–316. release by sarcoplasmic reticulum of intact rat ventricular1 1[85] Stromer H, de Groot MC, Horn M et al. Na /H exchange myocytes. Am J Physiol 1998;275:H977–H987.

inhibition with HOE642 improves postischemic recovery due to [109] Choi HS, Trafford AW, Orchard CH, Eisner DA. The effect of21 21attenuation of Ca overload and prolonged acidosis on reperfusion. acidosis on systolic Ca and sarcoplasmic reticulum calcium

Circulation 2000;101:2749–2755. content in isolated rat ventricular myocytes. J Physiol[86] Boron WF, Boulpaep EL. Intracellular pH regulation in the renal 2000;529:661–668.

proximal tubule of the salamander. J Gen Physiol 1983;8:53–94. [110] Bers DM, Ellis D. Intracellular calcium and sodium activity in1 2[87] Romero MF, Boron WF. Electrogenic Na /HCO cotransporters: sheep heart Purkinje fibers: effect of changes of external sodium3

¨cloning and physiology. Annu Rev Physiol 1999;61:699–723. and intracellular pH. Pflugers Arch 1982;393:171–178.[88] Lagadic-Gossmann D, Buckler KJ, Vaughan-Jones RD. Role of [111] Bielen FV, Bosteels S, Verdonck F. Consequences of CO acidosis2

1bicarbonate in pH recovery from intracellular acidosis in the guinea- for transmembrane Na transport and membrane current in rabbitpig ventricular myocyte. J Physiol 1992;458:361–384. cardiac Purkinje fibres. J Physiol 1990;427:325–345.

1 2[89] Dart C, Vaughan-Jones RD. Na -HCO symport in the sheep [112] Harrison SM, Frampton JE, McCall E, Boyet MR, Orchard CH.31 21 1cardiac Purkinje fibre. J Physiol 1992;451:365–385. Contraction and intracellular Na , Ca and H during acidosis in

´ ´[90] Camilion de Hurtado MC, Perez NG, Cingolani HE. An electrogenic rat ventricular myocytes. Am J Physiol 1992;262:C348–C357.sodium-bicarbonate cotransport in the regulation of myocardial [113] Philipson KD, Bersohn MM, Nishimoto AY. Effects of pH on

1 21intracellular pH. J Mol Cell Cardiol 1995;27:231–242. Na –Ca exchange in canine cardiac sarcolemmal vesicles. Circ[91] Aiello EA, Vila Petroff MG, Mattiazzi AR, Cingolani HE. Evidence Res 1982;50:287–293.

1 2 1for an electrogenic Na -HCO symport in rat cardiac myocytes. J [114] Eigel BN, Hadley RW. Contribution of the Na channel and31 1 1Physiol 1998;512:137–148. Na /H exchanger to the anoxic rise of [Na ] in ventricular

[92] Choi I, Romero MF, Khandoudi N, Bril A, Boron WF. Cloning and myocytes. Am J Physiol 1999;277:H1817–H1822.1 2characterization of a human electrogenic Na -HCO cotransporter [115] Ju YK, Saint DA, Gage PW. Hypoxia increases persistent sodium3

isoform (hhNBC). Am J Physiol 1999;276:C576–C584. current in rat ventricular myocytes. J Physiol 1996;497:337–347.[93] Pushkin A, Abuladze N, Newman D, Lee I, Xu G, Kurtz I. Cloning, [116] Undrovinas AI, Fleidervish IA, Makielski JC. Inward Na current at

characterization and chromosomal assignment of NBC4, a new resting potentials in single cardiac myocytes induced by themember of the sodium bicarbonate cotransporter family. Biochim ischemic metabolite lysophosphatidylcholine. Circ ResBiophys Acta 2000;1493:215–218. 1992;71:1231–1241.

Dow




1 1 1 1[117] Xiao XH, Allen DG. Role of Na /H exchanger during ischemia [130] Semb SO, Sejersted OM. Fuzzy space and control of Na -Kand preconditioning in the isolated rat heart. Circ Res pump rate in heart and skeletal muscle. Acta Physiol Scand1999;85:723–730. 1996;156:213–224.

[118] Crompton M, Capana M, Carafoli E. The sodium-induced efflux of [131] Su Z, Zou A, Nonaka A, Zubair I, Sanguinetti MC, Barry WH.1 1 21calcium from heart mitochondria. A possible mechanism for Influence of prior Na pump activity on pump and Na /Ca

regulation of mitochondrial calcium. Eur J Biochem 1976;69:453– exchange currents in mouse ventricular myocytes. Am J Physiol462. 1998;275:H1808–H1817.

1 1[119] Crompton M. The regulation of mitochondrial Ca transport in [132] Terracciano CMN. Rapid inhibition of the Na -K pump affects1 21heart. Curr Top Memb Transp 1985;25:231–276. Na –Ca exchanger-mediated relaxation in rabbit ventricular

[120] Jung DW, Baysal K, Brierley GP. The sodium-calcium antiport of myocytes. J Physiol 2001;533:165–173.mitochondria is not electroneutral. J Biol Chem 1995;270:672– [133] Wendt-Gallitelli MF, Voigt T, Isenberg G. Microheterogeneity of678. subsarcolemmal sodium gradients. Electron probe microanalysis in

[121] Fry CH, Powell T, Twist VW, Ward JPT. The effects of sodium, guinea-pig ventricular myocytes. J Physiol 1993;472:33–44.hydrogen and magnesium ions on mitochondrial calcium sequestra- [134] Su Z, Sugishita K, Ritter M, Li F, Spitzer KW, Barry WH. The

21tion in adult rat ventricular myocytes. Proc R Soc Lond sodium pump modulates the influence ofI on [Ca ] transientsNa i

1984;223:239–254. in mouse ventricular myocytes. Biophys J 2001;80:1230–1237.[122] Crompton M, Kunzi M, Carafoli E. The calcium-induced and [135] Lipp P, Niggli E. Sodium current-induced calcium signals in

sodium-induced effluxes of calcium from heart mitochondria. Eur J guinea-pig ventricular myocytes. J Physiol 1994;474:439–446.Biochem 1977;79:549–558. [136] Weber CR, Ginsburg KS, Bers DM. Cardiac Na current (I ) doesNa

[123] Brierley GP, Baysal K, Jung DW. Cation transport systems in not elevate submembrane [Na] ([Na] ) sensed by Na–Ca ex-sm1 1mitochondria: Na and K uniports and exchangers. J Bioenerg change (NCX). Biophys J 2002;82:565, (abstract).

Biomembr 1994;26:519–526. [137] Silverman BZ, Warley A, Miller JIA, James AF, Shattock MJ. Is[124] Rosen BP, Futai M. Sodium/proton antiporter of rat liver mito- there a transient rise in sub-sarcolemmal Na and activation of

chondria. FEBS Lett 1980;117:39–43. Na/K-pump current following activation ofI in ventricularNa

[125] Kapus A, Lukacs GL, Cragoe Jr. EJ, Ligeti E, Fonyo A. Charac- myocardium. Cardiovasc Res 2003;57:1025–1034.1 1terization of the mitochondrial Na –H exchange. The effect of [138] Goldhaber JI, Lamp ST, Walter DO, Garfinkel A, Fukumoto GH,

amiloride analogues. Biochim Biophys Acta 1988;944:383–390. Weiss JN. Local regulation of the threshold for calcium sparks in[126] Jung DW, Apel LM, Brierley GP. Transmembrane gradients of free rat ventricular myocytes: role of sodium–calcium exchange. J

1Na in isolated heart mitochondria estimated using a fluorescent Physiol 1999;520:431–438.probe. Am J Physiol 1992;262:C1047–1055. [139] Scriven DRL, Dan P, Moore EDW. Distribution of proteins

21[127] Arnon A, Hanlyn JM, Blaustein MP. Ouabain augments Ca implicated in excitation–contraction coupling in rat ventricular1transients in arterial smooth muscle without raising cytosolic Na . myocytes. Biophys J 2000;79:2682–2691.

1Am J Physiol 2000;279:H679–H691. [140] Moore ED, Etter EF, Philipson KD et al. Coupling of the Na /21 1 1[128] Bielen FV, Glitsch HG, Verdonck F. Changes of the subsarcolem- Ca exchanger, Na /K pump and sarcoplasmic reticulum in

1mal Na concentration in internally perfused cardiac cells. Bio- smooth muscle. Nature 1993;365:657–660.chim Biophys Acta 1991;1065:269–271.

1[129] Carmeliet E. A fuzzy subsarcolemmal space for intracellular Nain cardiac cells. Cardiovasc Res 1992;26:433–442.

Dow



I ntracellular Na regulation in cardiac myocytes

Documents

Transcript of I ntracellular Na regulation in cardiac myocytes