European Journal of Neuroscience, Vol. 9, pp.

1273-1281, 1997 0 European Neuroscience Association

Na+-Ca2+ Exchange Currents in Cortical Neurons:
Concomitant Forward and Reverse Operation and Effect
of Glutamate
Shan Ping Yu and Dennis W. Choi
Center for the Study of Nervous System Injury and Department of Neurology, Washington University School of Medicine,
St Louis, MO 63110, USA
Keywords: mouse, patch clamp, intracellular calcium, glutamate toxicity
Abstract
Na+-Ca2+exchanger-associated membrane currents were studied in cultured murine neocortical neurons, using
whole-cell recording combined with intracellular perfusion. A net inward current specifically associated with
forward (Na+o-Ca2+i) exchange was evoked at -40 mV by switching external 140 mM Li+to 140 mM Na+. The
voltage dependence of this current was consistent with that predicted for 3Na+:I Ca2+exchange. As expected,
the current depended on internal Ca2+, and could be blocked by intracellular application of the exchanger
inhibitory peptide, XIP. Raising internal Na+from 3 to 20 mM or switching the external solution from 140 mM Li+
to 30 mM Na+activated outward currents, consistent with reverse (Na+,-Ca2+o) exchange. An external Ca2+-
sensitive current was also identified as associated with reverse Na+-Ca2+exchange based on its internal Na+
dependence and sensitivity to XIP. Combined application of external Na+and Ca2+in the absence of internal
Na+triggered a 3.3-fold larger inward current than the current activated in the presence of 3 mM internal Na+,
raising the intriguing possibility that Naf-Ca2+ exchangers might concurrently operate in both the forward and
the reverse direction, perhaps in different subcellular locations. With this idea in mind, we examined the effect of
excitotoxic glutamate receptor activation on exchanger operation. After 3-5 min of exposure to 100-200 pM
glutamate, the forward exchanger current was significantly increased even when external Na+was reduced to
100 mM, and the external Ca*+-activated reverse exchanger current was eliminated.
Introduction
Na+-Ca2+exchangers are a specialized class of ion transporters with
11 transmembrane segments, distributed widely in heart, kidney, brain
and other tissues (Rahamimoff, 1990; Philipson and Nicoll, 1992).
In the brain, Na+-Ca2+exchanger mUNA is abundant in the cortex,
hippocampus, dentate gyrus, thalamus and cerebellum (Wroblewski,
1992). The brain exchanger is similar to the heart Na+-Ca2+
exchanger, and can be labelled with antibodies raised against the
latter (Furman et al., 1993; Marlier et al., 1993). Under normal
conditions, Na+-Ca2' exchangers are thought to transport Na+in
and Ca2+out (forward, or Na+o-Ca2+i exchange), a function vital
to the maintenance of intracellular free Ca2+( [Ca2+Ii) homeostasis.
Under some conditions, however, such as when intracellular Na+
([NafIi) is increased or the membrane is depolarized, energetics can
favour reverse operation, moving Ca2+in and Na+out (reverse or
Na+i-Ca2+o exchange) (Mullins, 1977; DiPolo and Beau@, 1986).
The physiology of Na+-Ca2+ exchangers has been extensively
studied in cardiovascular preparations (Reeves and Sutko, 1980;
Kaczorowski, 1992; Hilgemann et al., 1991; Niggli and Lederer,
1993), rod outer segments (Schnetkamp et al., 1989) and squid giant
axons (Allen, 1991). The majority of Na+-Ca2+ exchangers have a
transport stoichiometry of 3Na': 1Ca2+, and therefore are electrogenic,
moving one net positive charge inward with each forward cycle. The
membrane current specifically associated with exchanger activity has
been demonstrated electrophysiologically in mammalian heart cells
(Barcenas-Ruiz et al., 1987; Kimura et al., 1987; Hume and Uehara,
1988; Mechmann and Pott, 1988; Ehara et al., 1989; Niggli and
Lederer, 1993), in retinal cells (Schnetkamp et al., 1989; Gleason
et al., 1995) and in the squid giant axon (Caputo et al., 1989).
Neuronal Na+-Ca2+exchanger may also play important roles
in maintaining [Ca2+Ii in normal and pathological conditions in
mammalian central neurons (Fontana et al., 1995; Reuter and Porzig,
1995). The aim of the present study was to extend the electrophysiolo-
gical characterization of Na+-Ca2+ exchanger current to central
neurons. A specific question was how the exchanger would respond
to excitotoxic overactivation of glutamate receptors, something diffi-
cult to predict a priori. On the one hand, glutamate-induced increase
in [Na+]i might stimulate reverse operation of the exchanger, acting
to augment intracellular Ca2+overload and potentiate injury (Choi,
1988; Kiedrowski et al., 1994). On the other hand, the concurrent
glutamate-induced increase in [Ca2+]i might stimulate the forward
Correspondence to: Dennis Choi, Department of Neurology, Box 8111, Washington University School of Medicine, 660 South Euclid Avenue St Louis,
MO 63110, USA
Received 3 September 1996, revised 30 December 1996, accepted 31 January 1997
1274 Nai-Ca2+ exchanger currents in cortical neurons
Na+-Ca2+exchange, which might serve as an important protective
mechanism, acting to restore Ca2+homeostasis and limit resultant
cell injury (Choi, 1992). By measuring the membrane currents
associated with forward and reverse Na+-Ca2+ exchanges respect-
ively, the present experiments revealed that both exchange modes
may concurrently occur in the same cells, and that glutamate receptor
stimulation particularly enhances the forward Na+-Ca2+ exchange
activity even at reduced external Na'.
Materials and methods
Neurocotfical primary cell cultures
Mixed cortical cultures, containing neurons and confluent glial bed,
were prepared as described previously (Rose et al., 1993). Dissociated
neocortices obtained from fetal mice at 15-17 days of gestation were
plated onto a previously established glial monolayer (see below) at
a density of 0.27 hemispheres per 35 mm culture dish (Falcon,
Primaria), in Eagle's Minimal Essential Medium (MEM, Earle's salts)
supplemented with 20 mh4 glucose (final concentration, 25 mM), 5%
fetal bovine serum and 5% horse serum. Medium was changed after
1 week to MEM containing 20 mM glucose and 10% horse serum,
as well as cytosine arabinoside (final concentration 10 pM) to inhibit
cell division. Subsequently, cultures were fed once weekly with MEM
containing 25 mM glucose plus 2 mM glutamine. Cultures were kept
in a 37"C, humidified incubator in a 5% COz atmosphere. All
experiments were performed on cultures between days 12 and
20 in vitro.
Glial cultures were prepared from dissociated neocortices of
postnatal day 1-3 mice. Cells were plated at a density of 0.4
hemispheres per 35 mmdish, in Eagle's MEM containing 25 mM
glucose, 10% fetal bovine serum, 10% horse serum and 10 ng/ml
epidermal growth factor; a confluent glial bed was formed in
1-2 weeks.
Whole-cell recording of Na+-C$+ exchanger currents
The 35 mm culture dish was used as the recording chamber on the
stage of an inverted microscope (Nikon). Neurons of 15-20 pm
diameter, with phase-bright cell bodies on top of the glial bed,
were chosen for recordings. Neuronal identity has been previously
confirmed by Nissl staining and electrophysiological characterization,
whereas the glial bed is immunoreactive for glial fibrillary acidic
protein (Choi et aL, 1987; Rose et al., 1993).
The patch electrodes had tip resistance of 5-10 Mi2 (fire-polished).
Voltage clamp was achieved by using an EPC-7 amplifier (List
Electronic, Darmstadt, Germany). Series resistance compensation and
capacitance compensation were routinely applied. Whole-cell current
was digitally sampled at 100 or 300 ps (10 or 3.3 kHz). The current
signals were filtered by a 3 kHz, three-pole Bessel filter. Current
traces were displayed and stored on a Macintosh computer (Quatra
950, Apple Computer) using a data acquisitiodanalysis program
PULSE (HEKA Electronik, LambrechtPfalz, Germany). Some data
were simultaneously stored on video tape via a digital data recorder
(WR-lOB, Instrutech, Great Neck, USA).
lntracellular perfusion
Manipulation of intracellular ion concentrations was achieved by
perfusion of the patch pipette and cellular dialysis as described before
(Lopez 1992; Yu et al., 1994), using a modified recording electrode
holder (A058-C; E. W. Wright, Guildford, CT). The holder had two
outlets; one of them was connected with fine internal tubing of
200 pm external diameter inserted close to the tip of the electrode
and the other outlet was connected to solution reservoirs. Perfusion
of the internal capillary, generated by negative pressure of 4-5 p.s.i.
applied to the inside of the electrode holder, replaced internal solution
in the recording electrode. J udged by coloured solution replacements,
the solution in the electrode tip could be changed in <1 s. By
following the reversal potential of the voltage-gated fast Na+current,
ZNa, triggered by voltage steps from -70 to 0 mV for 50 ms in
tetrodotoxin-free external solution, it was determined that intracellular
Na' manipulation could be completed in <2 min.
Extracellular perfusion
The bathing medium was normally superfused by gravity at a flow
rate of 1-2 mumin. Glutamate was bath-applied and washed out via
the superfusion system at the higher rate of 5-7 ml/min; a bath
change could be accomplished in -30 s
To activate Na+-Ca2+ exchanger currents, test solutions were
delivered to the selected neuron body via pressure ejection from a
delivery pipette of 100 pm internal diameter placed -100 pm from
the cell body, using the DAD-12 superfusion system (Adams & List
Associates, Westbury, NY). The DAD-12 system allowed programmed
switches between several solutions; the pressure, sequence and
duration of solution applications were computer-controlled. To limit
activation of Naf-CaZf exchangers to those near the cell body (and
hence likely to be under good voltage control), the applied solution
was simultaneously removed by a suction pipette placed behind the
targeted cell. As a control, we demonstrated that voltage-activated
Na+currents were abolished by such push-pull pipette delivery of a
Na+-free solution near the cell body. Concentration jumps could be
reached in 10-50 ms as measured by changes in the liquid junction
potential on the electrode tip.
Solution composition
For manipulation of the forward Na+-Ca2+exchanger, either a Na+-
containing solution or a Li'-containing solution (Table 1) was
delivered via the local superfusion system described above. If the
Na' concentration was altered, equimolar tetraethylammonium (TEA)
or sucrose was used to maintain osmolarity. The Li+ion cannot
substitute for Na+ in forward Na+-Cazt exchange, but it does
not affect reverse Naf-Ca2+ exchange (Gadsby et aL, 1991). To
manipulate reverse Na'-Ca2+ exchange, Ca2+-containing and Ca2+-
free extracellular solutions were used (Table 1). The internal solutions
had no K'. To control [Ca2+Ii the Ca2' chelator 1,2-bis-(o-amino-
phenoxy)ethane-N,N,N'N-tetraacetic acid (BAF'TA), was usually
added into the solution. Tetrodotoxin (0.5 pM), nifedipine (10 pM)
and ouabain (20 pM) were regularly included in extracellular solutions
to block Na+ channels, dihydropyridine-sensitive Ca2+channels
and Na'/K+ ATPase respectively. In experiments where glutamate
receptors were activated, the extracellular solution during glutamate
exposure contained both Ca2+and Na', and the internal solution
contained no BAPTA throughout the experiments (Table 1).
Statistics
Significant difference between experimental data was detected by
Student's two-tailed t-test when P <0.05 (paired test for self-
controlled experiments and unpaired test for compatison of independ-
ent groups). The .average values reported in this study are mean
? SEM.
Chemicals
Glutamate, BAF'TA, TEA and tetrodotoxin were from Sigma. Nifedip-
ine and ouabain were from RBI (Natick, USA). The polypeptide XIP
Na+-Ca2+exchanger currents in cortical neurons 1275
TABLE 1. Experimental solutions
External solution (mM) NaCl LiCl CaC12 HEPES BAPTA Glucose
Naf/Ca2+140
Li+/Ca*+-
Na+/O Ca2+140
Li+/O Ca2+-
Internal solution (mM)
3 mM Na+3
10 mM Na+
20 mM Na' 20
0 Na+-
o Ca2+3
0 BAF'TA
-
140
140
NaCl
130
10
110
133
130
3
-
2
2
-
-
Cs-acetate
120
4.2
4.2
4.2
-
130
10
10
10
10
CaC12
10
4.2
10
10
10
-
-
-
0.5
0.5
HEPES
10
2.0
2.0
2.0
2.0
10
BAPTA Mg-ATP
10
10
10
10
2.0 10
2.0 -
Voltage-activated channel blockers TEA ( 5 mM), nifedipine (10 pM), tetrodotoxin (0.5 pM) and Na+-K+-ATPase inhibitor ouabain (20 pM) were included in
all external solutions. Different Na+concentrations were attained by substituting equimolar TEA or sucrose for NaCl. Free Ca2+in the internal 3 mM or 20
mM Na+solution was estimated as 120 nM (Yu et al., 1994).
pH =7.3 for all solutions.
(kindly provided by Dr R. A Colvin, Ohio State University) is
a selective inhibitor of the Na+-Ca2+ exchanger, acting on an
autoinhibitory site of a calmodulin binding region of the exchanger
(Li et al., 1991). It inhibits both Na+i-dependent Ca2+uptake (Ki -
1.5FM ) and Na+,-dependent Ca2+efflux in sarcolemmal vesicles,
as well as the Na+-Ca2+ exchange current recorded from excised
sarcolemmal patches (4 - 0.1 pM). It does not affect sarcolemmal
Ca2+ binding, Ca2+permeability, Na+-K+ ATPase function or
sarcoplasmic reticulum Ca2+-ATPase function (Li et aZ., 1991).
Results
Membrane current linked to forward Na+-Ca2+ exchange
To block forward Na+-Ca2+exchange, neurons were immersed in
Li+/Ca2+external solution (140 mM Li+, 2 mM Ca2+, and no Naf ;
Table l), with 3 mM Na+internal solution ( 3 mM Na', 130 mM
Cs; Table 1). Cell capacity was 38.6 2 0.5 pF (n =12). At a holding
potential of -40 mV, and in the presence of 0.5 pM tetrodotoxin
and 10 FM nifedipine, no spontaneous currents were observed.
Replacement of the Li+/Ca2+solution with Na+/Ca2+solution (140
mM Na+, 2 mM Ca2+; Table 1) via local application around the cell
body activated an inward current of -75.8 ? 14.9 pA (n =24) (Table
2); the current density was 1.9 ? 0.3 pNpF (n =12), similar to the
current density in heart cells (Kimura et al., 1987; Miura and Kimura,
1989). This external Na+-activated inward current showed no time-
dependent inactivation during the 0.6 s of Na+ application, and
ceased when Na+application stopped (Fig. 1A). There was no gross
'run-down' of this current during 10-15 min recordings (Fig. lA,
inset). The current exhibited the expected voltage-dependence: larger
currents were activated at hyperpolarized potentials (Fig. lC), consist-
ent with the known voltage-sensitivity of forward Na+-Ca2+
exchange. Based on the exchange model of 3Na+:1Ca2+(Mullins,
1977) the equilibrium potential of the Na+-Ca2+ exchanger, E N ~ , c ~ ,
was calculated according to the equation:
where EN^ and Eca respectively are the equilibrium potentials of Naf
and Ca2+across the membrane given by the Nernst equation. The
calculated E N ~ , C ~ in our experimental condition (140 mM external
Na+and 120 nM [Ca2+Ii) was +47 mV, which coincided with the
experimentally determined reversal potential of +47 2 2.3 mV
( n =4).
TABLE 2. Concurrent forward and reverse Naf-Ca2+exchange
~~
External solution change Internal solution Activated current n
Fomard operation of the
Na+-Ca2+ exchanger
Li+/Ca2+to Na+/Ca2'
Li'/Caz+ to Na+/Ca2+ 0 Ca2+
Lif/Ca2+to Na+/Ca*+
(Na' reduced to 100 mM)
3 mM Na'
3 mM Na+
Reverse operation of the
Nu'-Ca2+ exchanger
Li'/O Ca2+to Li+/Ca2+
Li+/O Ca2+to Li'/Ca2+
Li+/O Ca2+to Li+/Ca2+
Concurrent operation of the
Na'-Cd+ exchanger
Li+/O Ca2+to Naf/Ca2+
Li+/O Ca2+to Na+/Ca2+
3 mM Na+
0 Na'
o Ca*+
3 mM Na+
0 Na+
(PA)
-75.8 2 14.9
-28.0 ? 8.0*
-34.6 ? 8.6*
+78.9 Z 10.2
+12.0 ? 5.5*
+22.7 ? 7.0*
-26.0 ? 7.9
-84.7 ? 10.8*
24
3
7
20
10
4
5
3
~ ~ ___ __ ~ ~
*Significant difference ( P <0.05 by Student's t-test) from respective control
(the first row in each category).
When [Na+Ii was raised from 3 mM towards 10 mM (Table 1) by
intracellular perfusion, the exchanger reversal potential shifted towards
negative potentials, as predicted by the calculated equilibrium potential
of 4.5 mV (Fig. 1C). The current was also expectedly sensitive to
external Na+and internal Ca2+. Reducing external Na' to 100 mM
decreased the size of the inward current to -34.6 2 8.6 pA (Table 2).
When 10 mM BAPTA was included in the recording pipette, the
current evoked by external Na+application was also reduced (Table 2).
Further supporting the identification of this external Na+-activated
inward current as the current associated with the Na+-Ca2+exchanger,
the current was largely blocked by intracellular application of the
selective Na+-Ca2+ exchanger blocker, XIP (5 pM; -24 2 15 pA
with XIP, n =5; Li et d., 1991; Wu and Colvin, 1994) (Fig. IB).
Membrane current linked to reverse Na+-Ca2+ exchange
To examine the reverse Na+-Ca2+ exchanger operation, we used
intracellular perfusion to manipulate [Na+Ii while blocking forward
exchange by substituting external Na' with Li+as above. Raising
[Na+Ii from 3 to 20 mM by intracellular perfusion reversibly evoked
1276 Nat-Ca2t exchanger currents in cortical neurons
A
External
Li+/Ca2+
180 PA
0.2 sec
... .... ... .. . . .. . ......... ........ . ... .. . ... .................. W A
/ u u u u LI V'"I
I 1 min 3min 5min 8min 10min 15min
1
Li+/Ca2+
+J a+/Ca2+ Na+/Ca2+
I - 5 0 1
Control
5 -200
-250
-300
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
Second
C
Current (PA)
200
100
--
--
-400 1 1 1 1 1 1 1 1 \
-30 -10 I 0 30 50
Voltage (mV)
Li+/Ca2+
- Na+/Ca2+
2
5 prvl XIP
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
Second
Current (PA)
-400 ' - i
-50 -30 -10 10 30
Voltage (mV)
3 mM [NaIi 10 mM [NaIi
FIG. 1. Forward exchanger operation. (A) In a bath mediumcontaining 0 Na', 140 mM LiCl and 2 mh4 CaCI2 (Li+/Ca2+solution), local application to the
cell body of a solution containing 140 mM NaCl and 2 mM CaC12 but no Li+(Na+/Ca2+solution) reversibly activated an inward current response. Holding
potenbal here and subsequently was 4 0 mV unless stated otherwise. (Inset) The Na+-activated inward current was fairly stable during a 15 min control
recorchng, a dotted line shows the baseline level. (B) Switching from Na+/Ca2+solution to a Li+/Caz+solution reversibly evoked an outward current shift that
could beblocked by intracellular perfusion of 5 pM XIP. Notice that in the presence of XIP the baseline current moved from --125 to --55 PA, presumably
due to block of a tonic inward exchanger activity. (C) The current-voltage relationship for the Na+-activated inward current shift. The current shift was triggered
by ramp voltage pulses in Na+-free solution (Li+/Ca2+solution) and in Na'-containing solution (Na+/Ca2+solution); internal Na+was 3 mM. The Na+-
activated current disappeared at -40 mV (shown by intersection of the Na' relationship with the Li+relationship, marked by an arrow). If intracellular Na'
was increased by perfusion to 10 mM, this Na+-activated current shift was attenuated and the intersection point with the Li+relationship shifted to more
negative potentials. Determination of the exact intersection point was hampered by an increase in the noise level, due to the higher capacitance of the special
electrode holder needed for intracellular perfusion.
an outward membrane current (Fig. 2A). As this current was opposite
in direction to that expected from an alteration of membrane surface
charge, it probably reflected reverse Na+-Ca2+ exchange.
Alternatively, reverse Na+-Ca2+ exchange current could be trig-
gered by switching the external solution from Lif/Ca2+to a low
level of Na+/Ca2+(30 mM) (Table 1). Instead of the inward current
Na+-Ca2+exchanger currents in cortical neurons 1277
was due to block of tonically active reverse Naf-Ca2+ exchange, it
was blocked by prior intracellular dialysis with Na+-free solution
(Fig. 2B). Such internal Na'-dependence argues against the alternative
possibility that the current shift reflects activation of a Ca2+-sensitive
Cl- channel by Ca2+removal (Taleb et al., 1992). In addition, the
current could be blocked by intracellularly applied 10 pM XIP (-2.9
2 4.7 pA with XIP, n =5).
Yet another explanation for the current shift produced by application
of external Naf/O Ca2+solution might be an increase in forward
exchange. External Ca2+might compete with Na+for the external
binding site of the Na+-Ca2+exchanger (Miura and Kimura, 1989),
so removing external Ca2+ might enhance forward Na+-Ca2+
exchange due to increased binding of external Na+to the exchanger.
To eliminate this possibility, experiments were performed in external
Na' -free solutions which block forward exchange. Withdrawal of
external Ca2+(from Li+/Ca2' to Li+/0 Ca2+solution) induced a
similar inward current shift, and application of Ca2+(from Li+/0
Ca2+ to Li+/Ca2+ solution) triggered an outward current shift
(Table 2). Again, prior intracellular dialysis with Na' -free solution
suppressed these current shifts (Table 2), consistent with the idea that
these currents reflected reverse Na+-Ca2+ exchange. Lastly, since
Na+-Ca2+exchanger activation depends on the availability of [Ca2+],
for an internal regulatory site (DiPolo and BeaugC, 1986; Kimura
et al., 1986), we verified that reduction of [Ca2+], (with 10mM
BAPTA in Ca2+-free pipette solution) reduced the outward current
shift evoked by the external solution switch from Li+/0 Ca2+to Li'/
Ca2+(Table 2).
A
20 mM Na' 20 mM Na'
3 mM Na'
-
Solution -
-#
I- PA
40 sec
B
Na'lO Caz+
Na+/O C&
Na*/Ca2+ N a 7 E Na?Ca2'
- -
............. ...................................
3 mM Internal Na' No Internal Nat
( M P A
-
0.3 SBC
FIG. 2. Reverseexchanger operation. (A) An outward current was generated
by raising intracellular Na'. Theextracellular solution contained 0 Na', 140
mM Li+and 2 mM Ca2+, conditions disabling forward operation of theNa+-
Ca" exchanger. Raising internal Na' to 20mM by intracellular perfusion
reversibly activated an outward current shift. Similar results wereobtained in
threeex eriments. (B) In the presenceof 3 mM internal Na', switching from
current shift (left), that wasdependent upon thepresenceof intracellular Na'
(right). Thesameresults wereobtained in threeexperiments. A dotted line
shows thezero current baseline.
Na+/Ca l3+ solution to Na+/O Ca2+solution reversibly induced an inward
TABLE 3. Effects of glutamateon Nat-Ca2+ exchanger currents
External solution change Control Glutamate n
(PA) (PA)
Forward operation
Li+/Ca2+to Naf/Ca2+ -83.8 2 19.2 -196.2 2 60.8* 5
Li+/Ca2+to Na+/Ca2+
(Na' reduced to 100 mM) -37.0 2 7.2 -105.8 ? 24.9* 12
Reverse operation
Li'/Ca*+ to Na+/Ca2+
(Na' reduced to 30mM) +63.8 2 13.7 +94.0 2 22.4 4
Li'IO Ca2' to Li'/Ca2+ +76.4 2 16.5 -18.9 2 6.2* 9
Forward and reverse exchange currents were activated by the specified
extracellular solutions before(control) and 1-3 min after glutamatetreatment
(100-200 pM in Na+/Ca2+solution for 3-5 min). BAPTA was omitted from
thepipettesolution to permit [Ca2+Ii to increasenormally. Negativecurrent
reflects inward current (forward exchange) and positive current reflects outward
current (reverse exchange).
*Significant differencefromcontrol (P <0.01).
triggered by shifting from Lif/Ca2+to 140 mM Na+/Ca2+(see
above), an outward current appeared (Table 3). This apparent reversal
of exchange current is consistent with negative shift in exchange
reversal potential predicted by the 3Na': 1Ca2+exchange model
(Mullins, 1977; Ehara et al., 1989).
Wealso identified a membrane current sensitive to external Ca2+
as the reverse Na+-Ca2+exchange current. When Na+/Ca2+external
solution was switched to a Ca2+-free solution (Na+/O Ca2+solution;
Table l), an inward current shift was reversibly evoked (-84.0 ?
20.7 PA, n =4; Fig. 2B). Supporting the idea that this current shift
Concomitant forward and reverse Na+-C$+ exchanges in the
same cells
Both external Na+- and external Ca2+-activated membrane currents
were observed in the same cells. In each of six cells tested, we observed
both a baseline reverse exchange current (outward current blocked by
removal of external solution switch from Li+/Ca2+to Li+/0 Ca2+) and
a subsequent forward exchange current (inward current immediately
activated by a second switch from Li+/O Ca2+to Nai/O Ca2+) (Fig. 3).
These data led us to the hypothesis that both forward and reverse
Naf-Ca2+ exchange activities might exist simultaneously, perhaps in
different parts of same cells. To test this hypothesis, we examined
whether forward exchange activation combined with reverse exchange
blockage would evoke a larger inward current than the activation with-
out blocking reverse exchange. Indeed, activation of forward exchange
by external solution switch from Li/0 Ca2+to Na+/Ca2+in the absence
of Na+in the recording pipette evoked a 3.3-fold larger inward current
than that activated by the same solution switch in cells where the pipette
solution contained 3 mM Na' (Table 2) .
Effect of glutamate on forward and reverse Na+-Ca2+
exchange currents
With baseline characterization accomplished, we examined whether
toxic exposure to glutamate would reduce, or even reverse, normal
forward Na+-Ca2+exchange. Cells were initially immersed in Li/Ca2+
external solution and the inward current associated with forward Na+-
Ca2+exchange was activated by local application of Na+/Ca2+solution
as described above. Glutamate (100-200 pM for 3-5 min) exposure
was carried out in Na+/Ca2+solution, zero internal BAFTA, and no
voltage clamp, to approximate physiological conditions (Table 1). The
membrane potential depolarized from a baseline value of -5 1 ? 2 mV
( n =45) to -14 ? 2 mV (n =16), measured during a 5 min application
of 200 pM glutamate; it returned to -35 % 2 mV measured 1-10 min
after washing out glutamate (n =16). Brief checking under voltage
1278 Nat-Ca2' exchanger currents in cortical neurons
Li+/O Ca2+
Li+/Ca2+ Na'lO Ca2+
c
-100
-1 :::i 20
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
Time (second)
FIG. 3. Both forward and reverseoperation of the exchanger can bemeasured
in the same cell. This experiment was started with an external solution
containing Li' (140 mM) and Ca2+(2 mM), and an internal solution
containing 3 mM Na' and 120 nM Ca2+. Withdrawal of external Ca2+
produced aninward current shift, probably representing blockadeof ongoing
reverse exchanger operation. When this Ca2+-free external solution was
switched fromLi+- to Na+-containing solution, an additional inward current
shift appeared, due to activation of forward exchanger current. Similar results
wereobserved in anadditional fiveexperiments.
clamp (-70 mV) during glutamate exposure revealed that glutamate
induced an inward current of -806 2 108 pA ( n =13).
When the Naf -activated (140 mM) forward Na+-Ca2+exchange
current was reassessed 1-3 min after glutamate exposure (at 110 mV
holding potential), it was observed to be more than doubled (2.4-fold
increase; Fig. 4A, Table 3), despite a negative shift in exchanger reversal
potential to +20 ? 2.9 mV (n =3; Fig. 4C). We considered the
possibility that failure to detect reductionheversa1 of forward Na+-
Ca2+exchange was due to artificially high levels of external Na'.
After intense glutamate receptor activation in vivo, one might expect a
reduction in external Na+concentration due to massive influx of Na+
(Arens et al., 1992). However, even when the above experiment was
repeated with external Na+reduced to 100 mM (external Na+during
glutamate exposure was maintained at 140 mM), the same basic result
was obtained (2.8-fold increase; Table 3). Reduction of external Na'
to 30 mM, a condition which produced reversal of forward exchange
current under normal conditions (see above) still triggered reverse
exchange current shifts (Table 3).
Accompanying the increased forward exchange current, glutamate
exposure eliminated the external Ca2+-sensitive current shift associated
with reverse exchange (Fig. 4B), suggesting that the normal baseline
reverse Na+-Ca2+ exchanger activity had been eliminated. Indeed,
after glutamate exposure a small opposite membrane current was noted
in several experiments upon removal of external Ca2+(Fig. 4B). Similar
results were obtained when the reverse Na'-Ca2+ exchanger was
probed with the external solution switch from Li+/O Ca2+to Lif/Ca2+
(Table 3). Even when tested at 0 mV holding potential, further favouring
reverse Naf-Ca2+exchange, no Ca2+-sensitive reverse exchanger cur-
rent was observed after glutamate exposure (-3.6 ? 2.2 PA; n =5).
Discussion
Identification of neuronal Na+-Ca2+ exchanger currents
To our knowledge, the present study is the first to describe the membrane
current associated with Na+-Ca2+exchange activity in brain neurons.
As reviewed above, previous reports have described this current in
squid axons, retinal cells and cardiac muscles.
Our identification of the external Na+-activated inward current as
generated by forward Na+,,-Ca2+i exchange is supported by its depend-
ence on Na+gradient and internal Ca2+, characteristic voltage depend-
ence, and its sensitivity to the specific Nat-Ca2+ exchanger inhibitor
XIP. As expected, the current persists in the presence of Ca2+, Na+and
K+ channel blockers, plus a Na+-K+-ATPase inhibitor. In addition,
the measured reversal potential of this inward current matches the value
calculated from the 3Na': 1Ca2+exchanger model. The external Ca2+-
activated membrane current is internal Na+-dependent, consistent with
reverse Na+-Ca2+exchange; its XIP-sensitivity and intracellular Ca2+
dependence further support its association with the Naf-Ca2+
exchanger.
Concomitant forward and reverse Na+-C$+ exchange
activities
The present study also provides preliminary evidence suggesting that
both forward and reverse exchanger operations can take place concur-
rently in a single neuron, most likely in different parts of the cell. Na+-
Ca2+exchangers have been detected on neuronal cell bodies as well as
on dendrites (J ung et al., 1994; Reuter and Porzig, 1995). Whole-cell
current can be expected to reflect a summation of currents from these
different cellular locations. Concurrent forward and backward Na+-
Ca2+exchange might be caused by differences in the distribution of
intracellular Ca2+in different submembrane areas (Muller and Connor,
1992; Niggli and Lipp, 1993). Forward exchange might occur in loca-
tions where local [Ca2+Ii is high, such as in dendrites, while reverse
exchange might occur where local [Ca2+]i is low, such as in the cell
body. We acknowledge that our holding potential of -40 mV may
have accentuated reverse exchange over what would occur at more
hyperpolarized resting conditions, but in any case our data provide
support for the idea that the exchanger can achieve a high degree of
dynamic flexibility, responding differently to different conditions
within the same cell. Further investigation into the possibility of concur-
rent bidirectional Na+-Ca2+exchanger operation in both neurons and
non-neuronal cell types is warranted.
Effect of toxic glutamate exposure on Na+-C$+ exchange
currents
After exposure to toxic concentrations of glutamate (Choi, 1988),
the inward current associated with forward Na+-Ca2+ exchange
increased, and the outward current associated with reverse Na+-Ca2+
exchange was eliminated or even shifted to a small net inward current
of uncertain aetiology. This was true even when the membrane
holding potential was shifted positively to 0 mV (above the level
attained during the glutamate exposure itself), suggesting that the
Naf-Ca2+exchanger favours restoration of Ca2+homeostasis follow-
ing glutamate-induced toxic Ca2+overload.
Favouring Ca2+over Na' homeostasis may in part reflect the
known asymmetry of Na+-Ca2+ exchanger activation, as Ca2+can
activate the exchanger with higher affinity (>10-fold) from inside
the membrane than from outside (DiPolo and BeaugB, 1990, 1991).
Further increasing the impact of glutamate-induced [Ca2+], elevation
upon the exchanger, [Ca2+]i may rise transiently higher in the
immediate submembrane region than in the cytoplasm as a whole,
due to hindered diffusion (Zucker and Stockbridge, 1983; Nowycky
and Pinter, 1993). In contrast, entering Na' probably diffuses away
from the submembrane region more rapidly than Ca2+, and thus may
have a smaller potentiating effect on reverse exchange than entering
Ca2+does on forward exchange.
Na+-Ca2+exchanger currents in cortical neurons
1279
Na+/Ca2+ Na+/Ca2+
Li+/Ca2+ Li+/Ca2+
Li+/Ca*+ -
Control
Li+/O Ca2+
Li+/Ca2+
-
.......................................
300 QA
0.3 sec
After Glutamate
Li+/O Ca2+
Li+/Ca2+
-
Li+/Ca2+
-
.. ...................................................
B v
Control After Glutamate
C
100
50
n
a 0
2
Q
W
w
-50
-100
3
0
-150
-200
Control w
-60 -40 -20 0 20 40 60
Voltage (mV)
FIG. 4. Glutamate exposure enhanced the forward exchanger current and blocked the reverse exchanger current. (A) Changing from Li+-containing solution to
Naf-containing solution activated the forward exchanger current. After 3 min of exposure to 200 FM glutamate in Na''Ca2+ external solution this forward
exchanger current was markedly increased. BAPTA was omitted from the intracellular solution to permit increases in [Ca2'Ii (as in B). A dotted line shows the
zero current baseline (as in B). (B) The external Ca2+-sensitive current linked to the reverse Naf-Ca2' exchange was blocked by glutamate exposure. Unlike
in Figure 2B, the current was triggered in 0 Na' extracellular solution, showing its independence of external Na'. The inward current due to block of the
reverse Na'-Ca2+ exchange was totally abolished after exposure to 200 yM glutamate for 3 min. Instead, a small outward current appeared. (C) The current-
voltage relationship of the Na+-activated forward exchanger current before and after glutamate exposure. Glutamate exposure increased the slope of the current-
voltage relationship and shifted the reversal potential to the left.
The glutamate-induced changes in exchanger current observed in
the present experiments are unlikely to be due to direct effects of
glutamate upon the exchanger, as current measurements were per-
formed after washout of applied glutamate. However, it is quite
possible that the glutamate exposure produced a lasting alteration in
exchanger behaviour mediated by intracellular signalling pathways,
for example an overall up-regulation of exchanger function due to
protein phosphorylation (Blaustein et al., 1996).
The idea that glutamate exposure preferentially enhances forward
exchange is consistent with findings that the Na'-Ca2' exchanger
in hippocampal neurons contributes to recovery of [Ca2+]i following
K+ depolarization and glutamate receptor activation (Koch and
Barish, 1994), and that inhibition of Na+-Ca2' exchange following
toxic glutamate exposure increases the incidence of death of cerebellar
granule cells (Andreeva et al., 1991). However, the alternative
possibility, that toxic glutamate exposure per se increases [Na+]i
1280 Naf-Ca2+ exchanger currents in cortical neurons
sufficiently to overcome the [Ca2']i increase and shift the exchanger
towards injury-promoting reverse operation (Choi, 1988; Kiedrowski
et al., 1994), may occur under certain conditions. For example, in
the ischaemic brain in vivo, massive cellular Na' influx may lower
"a'], to 38-50 mM (Hansen, 1985; Siesjo, 1992; Xie et al., 1994),
a level near that which alone increased reverse exchange in the
present experiments. Reversal of the Naf-Ca2+ exchanger has been
implicated in Ca2+-loading myocardial cells during ischaemia (Allen
et al. 1993) and in anoxic injury to optic nerve (Stys et al., 1992) or
hepatocytes (Gasbanini et al., 1993). However, upon restoration of
blood flow to the ischaemic brain, cellular Na+ gradients recover
quickly (Hansen and Nedergaard, 1988; Eleff et al., 1991), perhaps
shifting the Na+-Ca2+ exchanger back to cytoprotective forward
operation.
Acknowledgements
The authors express their gratitude to Dr R. A. Colvin for kindly providing
the Na+-Ca2+exchanger inhibitory peptide XIP. The study was supported in
part by National Institutes of Health (USA) grant 50788 (to D. W. C).
Abbreviations
BAF'TA 1,2-bis-(O-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid
MEM Minimal Essential Medium
TEA tetraethylammonium
References
Allen, D. G., Cairns, S. P., Turvey, S. E. and Lee, J . A. (1993) Intracellular
calciumand myocardial function during ischemia. Adv. Exp. Med. Biol.,
Allen, T. J. A. (1991) The exchange in intact squid axons. Ann. N.Z Acad.
Sci., 639, 71-84.
Andreeva, N., Khodorov, B., Stelmashook, E., Cragoe, E., J r and Victorov, I.
(1991) Inhibition of Nat-Ca2+ exchange enhances delayed neuronal death
elicited by glutamate in cerebellar granule cell cultures. Brain Res., 548,
322-325.
Arens, J ., Stabel, J . and Heinemann, U. (1992) Pharmacological properties of
excitatory amino acid induced changes in extracellular calciumconcentration
in rat hippocampal slices. Can. J. Physiol. Pharmacol., 70, S1944205.
Barcenas-Ruiz, L., Beuckelmann, D. J . and Wier, W. G. (1987) Sodium-
calciumexchange in heart: membrane currents and changes in [Ca2+],.
Science, 238, 1720-1723.
Blaustein, M. P., Fontana, G. and Rogowski, R. S. (1996). The Na+-Ca2+
exchanger in rat brain synaptosomes-kinetics and regulation. Ann. N. Z
Acad. Sci., 779, 300-317.
Caputo, C., Bezanilla, F. and DiPolo, R. (1989) Currents related to the
sodium-calcium exchange in squid giant axon. Biochim. Biophys. Ada,
986, 250-256.
Choi, D. W. (1988) Calcium-mediated neurotoxicity: relationship to specific
channel types and role in ischemic damage. Trends Neurosci., 11, 465469.
Choi, D. W. (1992) Excitotoxic cell death. J. Neurobiol., 23, 1261-1276.
Choi, D. W., Maulucci-Gedde, M. and Kriegstein, A. R. (1987) Glutamate
neurotoxicity in cortical cell culture. J. Neurosci., 7, 357-368.
DiPolo, R. and BeaugB, L. (1986) Reverse Na-Ca exchange requires internal
Ca and/or ATP in squid axons. Biochim. Biophys. Acta, 854, 298-306.
DiPolo, R. and BeaugB, L. (1990) Asymmetrical properties of the Na-
Ca exchanger in voltage-clamped, internally dialyzed squid axons under
symmetrical ionic conditions. J. Gen. Physiol., 95, 819-885.
DiPolo, R. and BeaugB, L . (1991) Regulation of Na-Ca exchange. Ann. N.Z
Acad. Sci., 639, 100-111.
Ehara, T., Matsuoka, S. and Noma, A. (1989) Measurement of reversal
potential of Na+-Ca2+exchange current in single guinea-pig ventricular
cells. J. Physiol. (Lond.), 410, 227-249.
Eleff, S. M., Maruki, Y., Monsein, L. H., Traystman, R. J., Bryan, R. N. and
Koehler R. C. (1991) Sodium, ATP, and intracellular pH transients during
reversible complete ischemia of dog cerebrum. Stroke, 22, 233-241.
Fontana, G., Rogowski, R. S. and Blaustein, M. P. (1995) Kinetic properties
346, 19-29.
of the sodium-calcium exchanger in rat brain synaptosomes. J. Physiol.
(Lond.), 485, 349-364.
Furman, I., Cook, O., Kasir, J. and Rahamimoff, H. (1993) Cloning of two
isoforms of the rat brain Na+-Cazf exchanger gene and their functional
expression in HeLa cells. FEBS Lett., 319, 105-109.
Gadsby, D. C., Noda, M., Shepherd, R. N. and Nakao, M. (1991) Influence
of external monovalent cations on Na-Ca exchange current-voltage
relationships in cardiac myocytes. Ann. N.I: Acad. Sci., 639, 140-146.
Gasbarrini, A,, Borle, A. B. and VanThiel, D. H. (1993) Ca2+antagonists do
not protect isolated perfused rat hepatocytes from anoxic injury. Biochinz.
Biophys. Acba, 1177, 1-7.
Gleason, E., Borges, S. and Wilson, M. (1995) Electrogenic Na-Ca exchange
clears Ca2+loads fromretinal amacrine cells in culture. J. Neurosci., 15,
Hansen, A. J . (1985) Effect of anoxia on ion distribution in the brain. Physiol.
Rev., 65, 101-148.
Hansen, A. J. and Nedergaard, M. (1988) Brain ion homeostasis in cerebral
ischemia. Neurochem. Pathol., 9, 195-209.
Hilgemann, D. W., Nicoll, D. A. and Philipson, K. D. (1991) Charge movement
during Na+translocation by native and cloned cardiac Na+/Ca2+exchanger.
Nature, 352, 715-718.
Hume, J . R and Uehara, A. (1988) 'Creep currents' in single frog atrial cells
may be generated by electrogenic NdCa exchanger. J. Gen. Physiol., 87,
857-884.
J ung, A., Lischka, F. W., Engel, J . and Schild, D. (1994) Sodiudcalcium
exchanger in olfactory receptor neurones of Xenopus laevis. NeuroReport,
5, 1741-1744.
Kaczorowski, G. J. (1992) Na+/Ca2+exchange. In Cragoe, E. J., Jr, Kleyman,
T. R. and Simchowitz, L. (eds), Amiloride and its Analogs-Unique Cation
Transport Inhibitors. VCH, New York, pp. 121-143.
Kiedrowski, L., Brooker, G., Costa, E. and Wroblewski, J. T. (1994) Glutamate
impairs neuronal calciumextrusion while reducing sodiumgradient. Neuron,
Kimura, J ., Miyamae, S. and Noma, A. (1987) Identification of sodium-
calcium exchange current in single ventricular cells of guinea-pig. J.
Physiol. (Lond.), 384, 199-222.
Koch, R. A. and Barish, M. E. (1994) Perturbation of intracellular calcium
and hydrogen ion regulation in cultured mouse hippocampal neurons by
reduction of the sodium ion concentration gradient. J. Neurosci., 14,
2585-2593.
Li, Z., Nicoll, D. A., Hilgemann, D. W., Filoteo, A. G., Penniston, J. T.,
Weiss, J. N., Tomich, J . M. and Philipson, K. D. (1991) Identification of a
peptide inhibitor of the cardiac sarcolemmal Na+-Ca2+exchanger. J. Biol.
Chem., 266, 1014-1020.
Lopez, H. S. (1992) Kinetics of G protein-mediated modulation of the
potassium M-current in bullfrog sympathetic neurons. Neuron, 8, 725-736.
Marlier, L. N., Zheng, T., Tang, J . and Grayson, D. R. (1993). Regional
distribution in the rat central nervous system of a mRNA encoding a portion
of the cardiac sodiudcalcium exchanger isolated from cerebellar granule
neurons. Mol. Brain Res., 20, 21-39.
Mechmann, S. and Pott, L. (1988) Identification of Na-Ca exchange current
in single cardiac myocytes. Nature, 319, 597-599.
Miura, Y. and Kimura, J. (1989) Sodium-calcium exchange current-
dependence on internal Ca and Na and competitive binding of external Na
and Ca. J. Gen. Physiol., 93, 1129-1145.
Muller, W. and Connor, J. A. (1992) Ca2+signaling in postsynaptic dendrites
and spines of mammalian neurons in brain slice. J. Physiol. (Paris), 86,
Mullins, L. J. (1977) A mechanismfor NdCa transport. J. Gen. Physiol., 70,
68 1-695.
Niggli, E. and Lederer, W. J. (1993) Activation of Na-Ca exchange current
by photolysis of 'caged calcium'. Biophys. J. , 65, 882-891.
Niggli, E. and Lipp, P. (1993) Subcellular restricted spaces: significance for
cell signaling and excitation-contraction coupling. J. Muscle Res. Cell
Motil., 14, 288-291.
Nowycky, M. C. and Pinter, M. J . (1993) Time courses of calciumand
calcium-bound buffers following calcium influx in a model cell. Biophys.
Philipson, K. D. and Nicoll, D. A. (1992) Sodium-calcium exchange. Cum
Opin. Cell BioL, 4, 678-683.
Rahamimoff, H. (1990) Na+-Ca2+exchanger: the elusive protein. Curr: Top.
Cell. Regul., 31, 241-271.
Reeves, J. P. and Sutko, J. L. (1980) Sodiumxalcium exchange activity
generates a current in cardiac membrane vesicles. Science, 208, 1461-1464.
Reuter, H. and Porzig, H. (1995) Localization of functional significance of
3612-3621.
12, 295-300.
57-66.
J., 64, 77-91.
the Na'lCa'' exchanger in presynaptic boutons of hippocampal cells in
culture. Neuron, 15, 1077-1084.
Rose, K., Goldberg, M. P. and Choi, D. W. (1993) Cytotoxicity in murine
cortical cell culture. In Tyson, C. A. and Frazier, J. M. (eds), In Vitro
Biological Methods. Academic Press, San Diego, CA, pp. 46-60.
Schnetkamp, P. P. M., Basu, D. K. and Szerencsei, R. T. (1989) Na+-Ca2'
exchange in bovine rod outer segments requires and transports K'. Am. J.
Physiol., 251, C1534157.
Siesjo, B. K. (1992) Pathophysiology and treatment of focal cerebral ischemia.
J. Neurosurg., 71, 169-184.
Stys, P. K., Waxman, S. G. and Ransom, B. R. (1992) Ionic mechanisms of
anoxic injury in mammalian CNS white matter: role of Na' channels and
Na'-Ca'' exchanger. J. Neurosci., 12, 430-439.
Taleb, O., Feltz, P., Bossu, J. L. and Feltz, A. (1992) Sensitivity of chloride
channels to changes in intracellular calcium: investigations on spontaneous
and GABA-evoked activity. Epilepsy Res., Suppl. 8, 47-56.
Na+-Ca2+exchanger currents in cortical neurons 1281
Wroblewslu, J . T. (1992). Na+/Ca*' exchange and glutamate neurotoxicity.
Neurosci. Facts, 3, 16.
Wu, A. and Colvin, R. A. (1994) Characterization of exchange inhibitory
peptide effects on Na'lCa'' exchange in rat and human membrane vesicles.
J. Neurochem., 63, 2136-2143.
Xie, Y., Dengler, K., Zacharias, E., Wilffert, B. and Tegtmeier, F. (1994)
Effects of the sodiumchannel blocker tetrodotoxin (TTX) on cellular ion
homeostasis in rat brain subjected to complete ischemia. Bruin Res., 652,
Yu, S. P., OMalley, D. M. and Adams, P. R. (1994) Regulation of M current
by intracellular calcium in bullfrog sympathetic ganglion neurons. J.
Neurosci., 14, 348773499.
Zucker, R. S. and Stockbridge, N. (1983) Presynaptic calciumdiffusion and
the time courses of transmitter release and synaptic facilitation at the squid
giant synapse. J. Neurosci., 3, 1263-1269.
216-224.