Clinical Science (2001) 101, 243–251 (Printed in Great Britain)
Importance of glycolytically derived ATP for
Na+ loading via Na+/H+ exchange during
metabolic inhibition in guinea pig
ventricular myocytes
Hiroshi SATOH, Shiho SUGIYAMA, Noriyuki NOMURA, Hajime TERADA
and Hideharu HAYASHI
Division of Cardiology, Internal Medicine III, Hamamatsu University School of Medicine, 1-20-1 Handayama,
Hamamatsu 431-3192, Japan
A
B
S
T
R
A
C
T
The increase in the intracellular Na+ concentration ([Na+]i) during myocardial ischaemia is
crucial for ischaemia/reperfusion cell injury, and the cardiac subtype of the Na+/H+ exchanger
(NHE-1) has been shown to be a major pathway for Na+ loading. While the importance of
glycolytically derived ATP for the optimal functioning of membrane transporters and channels
has been suggested, whether NHE-1 is actually activated during myocardial ischaemia remains
controversial. Here we examined whether the activity of NHE-1 is predominantly dependent on
intracellular ATP generated by glycolysis, and whether the additional inhibition of glycolysis can
affect the increase in [Na+]i during the inhibition of oxidative phosphorylation in intact guinea
pig ventricular myocytes. The selective inhibition of glycolysis by 2-deoxyglucose prevented the
recovery of intracellular pH and the transient increase in [Na+]i following intracellular acidosis
induced by a NH4Cl pre-pulse. During severe metabolic inhibition (SMI ; induced by amobarbital
and carbonyl cyanide m-chlorophenylhydrazone in a glucose-free perfusate), most myocytes
changed from rod-shaped to contracted forms by " 15 min. [Na+]i increased linearly until rigor
contracture occurred, but after rigor contracture the rate of increase was blunted. The increase
in [Na+]i during SMI was suppressed significantly by an inhibitor of NHE-1, hexamethylene
amiloride. The increase in the intracellular Mg2+ concentration, which can reciprocally indicate
depletion of intracellular ATP, was small during the initial 10 min of SMI, but became larger from
just a few minutes before rigor contracture. In the presence of 2-deoxyglucose, the time to rigor
during SMI was shortened, but the increase in [Na+]i before rigor contracture was not significant,
and was much less than that in the absence of 2-deoxyglucose. It is concluded that ATP generated
by glycolysis is essential to activate NHE-1, and that the dependence of NHE-1 on glycolysis
might affect the increase in [Na+]i observed during myocardial ischaemia.
Key words : glycolysis, metabolic inhibition, myocytes, Na+\H+ exchange, sodium.
Abbreviations : AM, acetoxymethyl ester ; [ATP] , intracellular ATP concentration ; BCECF, 2h,7h-bis(carboxyl)-5h,6h-carboxyi
fluorescein ; [Ca#+] , intracellular free Ca#+ concentration ; CCCP, carbonyl cyanide m-chlorophenylhydrazone ; 2-DG, 2-deoxyi
glucose ; HMA, hexamethylene amiloride ; J , net acid efflux ; [Mg#+] , intracellular Mg#+ concentration ; [Na+] , intracellular Na+
H
i
i
concentration ; NHE-1, cardiac subtype of the Na+\H+ exchanger ; pH , intracellular pH ; SBFI, sodium-binding benzofuran
i
isophthalate ; SMI, severe metabolic inhibition.
Correspondence : Dr Hiroshi Satoh (e-mail satoh36!hama-med.ac.jp).
# 2001 The Biochemical Society and the Medical Research Society
243
244
H. Satoh and others
INTRODUCTION
MATERIALS AND METHODS
Myocardial ischaemia has profound effects on the
functioning and viability of cardiac myocytes. Although
reperfusion of the coronary artery is essential for the prevention of progressive myocardial necrosis and\or the
recovery of heart function, it is not always beneficial to
the damaged myocardium, and causes additional myocardial injury. Among several mechanisms implicated in
ischaemia\reperfusion injury, abnormal handling of
intracellular Ca#+ is thought to represent a final pathway
of cell injury. Sarcolemmal Na+\Ca#+ exchange has been
considered, at least in part, as a mechanism of Ca#+
overload in reperfusion injury [1,2]. This concept requires that the intracellular Na+ concentration ([Na+]i)
increases during the ischaemic period or on reperfusion.
[Na+]i has been reported to increase significantly after
prolonged exposure to hypoxic solutions [3,4], and we
have also demonstrated an elevation in [Na+]i during
severe metabolic inhibition (SMI) in isolated cardiac
myocytes [5]. An increase in [Na+]i could also be
detrimental because of effects on K+ loss [6]. The
mechanism responsible for the increase in [Na+]i is
debated, but several lines of evidence have suggested the
importance of Na+\H+ exchange (a cardiac isoenzyme :
NHE-1) [7,8]. However, it remains controversial
whether [Na+]i actually increases during the ischaemic
period, or whether it does so only after reperfusion [7,9].
The reason why measured changes in [Na+]i differ in
different studies remains unclear, but one possible
explanation is that, during ischaemia, protons will
accumulate in the cell and in the extracellular space, and
the low extracellular pH would inhibit NHE-1 [10].
The activity of NHE-1 is regulated by numerous intraand extra-cellular factors, such as chemical gradients of
Na+ and H+, protein kinase C, protein kinase A,
Ca#+\calmodulin and mitogen-activated protein kinase
[11–16]. In addition, ATP appears to be required for
optimal functioning of NHE-1 in several cell types
[17–20]. It is also proposed that local ATP, close to
sarcolemmal membrane, may be governed primarily by
local glycolytic activity, and may differ significantly from
global ATP [21]. Such a mechanism could have many
implications for ATP-dependent membrane processes
[21–23]. Thus the idea of ATP ‘ compartmentation ’
proposes the second possibility that the activity of
NHE-1 in cardiac myocytes relies on ATP generated by
glycolysis rather than by oxidative phosphorylation. In
this regard, we and others have provided evidence for the
importance of glycolysis in regulation of the exchanger
[17,20,21,24].
Therefore the aims of the present study were to
examine (1) whether Na+ influx via NHE-1 is dependent
on intracellular ATP generated by glycolysis, and (2)
whether the inhibition of glycolysis can affect the increase
in [Na+]i during SMI.
Preparation of ventricular myocytes
# 2001 The Biochemical Society and the Medical Research Society
This investigation conformed with the guidelines in the
Guide for the Care and Use of Laboratory Animals,
published by the U.S. National Institutes of Health.
Ventricular myocytes were isolated from female guinea
pigs (body weight 400–600 g) by the method reported
previously [5]. A small aliquot of myocytes placed in
a bath (volume 500 µl) was mounted on the stage of a
Nikon TMD inverted microscope and perfused with
Hepes-buffered solution containing (in mM) : NaCl 137,
KCl 4, MgSO 1.2, glucose 10, Hepes 10, CaCl 2.45, and
%
#
adjusted to pH 7.4 by NaOH at room temperature. The
Hepes-buffered solution was used to inhibit bicarbonatedependent mechanisms involved in the regulation of
intracellular pH (pHi), especially Na+–HCO − co$
transport.
Measurement of [Na+]i, intracellular free
Ca2+ concentration ([Ca2+]i) and pHi
For measurement of [Na+]i and pHi, the myocytes were
loaded with sodium-binding benzofuran isophthalate
(SBFI)\acetoxymethyl ester (AM) (10 µM) or 2h,7hbis(carboxyl)-5h,6h-carboxyfluorescein (BCECF)\AM
(1 µM) for 30 min at room temperature. Fluorescent
signals were imaged using a silicon-intensified target
camera (model C2400 ; Hamamatsu Photonics, Hamamatsu, Japan), with the output digitized to a resolution of 512i512 pixels by an image analysis system
(Argus 50 ; Hamamatsu Photonics). The cells were
excited via an epifluorescence illuminator from a 100 W
xenon lamp equipped with an interference filter.
For SBFI, the excitation wavelengths were 340 nm and
380 nm, and the emission signal at 520 nm was detected
using a SIT camera and stored in a digital imaging
processor. Each image was an accumulation of eight
(30\s) successive video frames, and images were obtained
every 5–60 s. The fluorescence intensity was measured in
a defined cell area, avoiding nuclei. Fluorescence ratios
were obtained by dividing, pixel by pixel, the image at
340 nm by the image at 380 nm after each background
subtraction. Exposure to excitation light was limited to
the time of actual data collection (0.27 s\each collection)
by an electrically controlled shutter, and neutral-density
filters were placed in the excitation light paths to prevent
photobleaching. The 340\380 nm ratio of dye fluorescence was converted into a value of [Na+]i using in
vivo calibration. For SBFI, in vivo calibration was
performed using calibration solutions containing various
Na+ concentrations (5, 10, 30 and 50 mM). The calibration solutions were prepared from appropriate mixtures of high-Na+ and high-K+ solutions containing
10 µM gramicidin and 100 µM strophanthidin [5].
Glycolysis-dependence of Na+/H+ exchange and intracellular Na+ loading
For BCECF, the excitation wavelengths were 490 nm
and 450 nm, and the emission wavelength was at
505–560 nm. BCECF-loaded myocytes were perfused
with calibration solutions containing 10 µM nigericin,
and the fluorescence ratios were linearly related to pHi
values from 6.5 to 7.5, as reported previously [24].
Estimation of intracellular ATP
concentration ([ATP]i)
We monitored the intracellular Mg#+ concentration
([Mg#+]i), which may reflect reciprocal changes in [ATP]i
due to release of bound Mg on hydrolysis of the
Mg#+–ATP complex [25,26]. For measurement of
[Mg#+]i, myocytes were loaded with 5 µM mag-fura
2\AM for 20 min. [Mg#+]i was calculated using the
following equation :
[Mg#+]i l Kdi(Sf\Sb)i(RkRmin)\(RmaxkR)
where R is the measured 340\380 nm excitation ratio,
Rmin is the ratio at [Mg#+]i l 0 mM, Rmax is the ratio at
saturating [Mg#+]i, Kd is the dissociation constant, and
Sf\Sb is the ratio of excitation efficiencies of free to Mg#+bound mag-fura 2 at 380 nm. In order to obtain the three
constants (i.e. KdiSf\Sb, Rmax and Rmin), we performed
in vivo calibration according to the method described by
Murphy et al. [27]. The dye-loaded cells were perfused
with calibration solutions containing 30 µM ionomycin
and various concentrations of Mg#+ (1, 5, 10 and 25 mM).
The free Mg#+ concentration in the solution was set and
determined with EGTA buffers as described by Murphy
et al. [27]. We obtained values of KdiSf\Sb l 3.675 mM,
Rmax l 3.665 and Rmin l 2.03.
Experimental protocol
For SMI, the perfusate contained 3.3 mM amobarbital
and 5 µM carbonyl cyanide m-chlorophenylhydrazone
(CCCP), and lacked glucose [5]. Thus ATP production
was limited to intrinsic glycolysis, resulting in rapid and
profound ATP depletion. For the selective inhibition
of glycolysis, glucose was replaced with 20 mM
2-deoxyglucose (2-DG). These inhibitors have little
effect on cell autofluorescence, presumably because
CCCP decreases the NADH level while amobarbital
increases it [5].
The activity of NHE-1 was estimated as the rate of pHi
recovery and the extent of the transient increase in [Na+]i
following intracellular acidosis, which was induced by
washout of 15 mM NH Cl after its transient application
%
(5 min). The changes in pHi were monitored every 2 min
during a 60 min preincubation in the metabolic inhibitors
or control Hepes-buffered solution, and every 30 s on
application of NH Cl. The data acquisition interval after
%
the removal of NH Cl was shortened to 5 s to improve
%
temporal resolution.
For the quantitative analysis of NHE-1 activity, net
acid efflux (JH) was estimated during recovery from
intracellular acidosis using the equation :
JH (mM\min) l βid(pHi)\dt
where β is the intrinsic buffering power obtained from
the stepwise reduction in NH Cl, and d(pHi)\dt is the
%
pHi recovery rate at any given pHi (calculated by a
single-exponential fit) [24].
Reagents
Strophanthidin, gramicidin, 2-DG and CCCP were
purchased from Sigma. Amobarbital was obtained from
Tokyo Kasei Inc. (Tokyo, Japan). Hexamethylene amiloride was obtained from Research Biochemicals Inc.
SBFI\AM, BCECF\AM and mag-fura 2\AM were supplied by Molecular Probes Inc. These reagents were
used from stock solutions in ethanol or DMSO. None of
the reagents by themselves produced either a change in
cellular autofluorescence or fluorescence artifacts.
Statistical analyses
Results are expressed as meanspS.E.M. for the indicated
number (n) of myocytes from at least three guinea pigs.
Student’s t test and one- or two-way ANOVA were used
for statistical analyses. Multiple-group comparison was
carried out using the Bonferroni-modified t test, and the
difference was considered significant at P 0.05.
RESULTS
Changes in [Na+]i and cell morphology
during SMI
In our experiments, 50–60 % of all cells were rod-shaped,
with clear sarcomeres. As reported previously, most of
the myocytes changed from an elongated rod-shaped
form to contracted (75–90 % of initial length) and
hypercontracted ( 75 % of initial length) forms during
SMI [5]. The time at which cells had contracted varied
between 12 and 21 min, and the mean rigor time was
15 min (Figure 1b, n l 17). Figure 1(a) shows a typical
record of the changes in longitudinal cell length
and [Na+]i during SMI. During the initial 14 min of SMI,
[Na+]i increased linearly, but after rigor contracture
occurred the rate of increase was blunted (18–40 min).
Figures 1(b) and 1(c) summarize the changes in the percentage of rod cells and in [Na+]i during SMI. The
application of 1 µM hexamethylene amiloride (HMA), an
inhibitor of NHE-1, did not influence the morphological
changes, but suppressed the increase in [Na+]i significantly. The mean values of [Na+]i at 16 min of SMI were
15.8p2.9 mM in the absence and 9.8p1.1 mM in the
presence of HMA (P 0.05 by two-way ANOVA). It
thus appeared that the increase in [Na+]i during SMI was
# 2001 The Biochemical Society and the Medical Research Society
245
246
H. Satoh and others
Figure 2
Changes in [Mg2+]i during SMI
ATP exhaustion during SMI was estimated by monitoring [Mg2+]i, which is thought
to reflect reciprocal changes in [ATP]i. During the initial 10 min of SMI, [Mg2+]i
increased slightly and slowly, which might indicate ATP production by activated
glycolysis. Then [Mg2+]i began to increase rapidly just a few minutes before rigor
contracture, and continued to increase after completion of the contracture.
S2–S10, min of SMI ; BR4 and BR2, min before rigor ; R, rigor onset, R2–R4, min
after rigor. Values are meanspS.E.M. Significance : *P 0.05 compared with
values at 0 min.
During the initial 10 min of SMI, [Mg#+]i increased
slightly and slowly ; it then increased rapidly just a few
minutes before rigor contracture, and continued to
increase after completion of the contracture.
Dependence on glycolysis of Na+ influx via
NHE-1
Figure 1
Changes in cell morphology and [Na+]i during SMI
(a) Typical record of the changes in longitudinal cell length () and [Na+]i (
)
during SMI. During the initial 14 min of SMI, [Na+]i increased linearly, but after
rigor contracture had occurred the rate of increase was blunted. (b, c) Summarized
data for changes in the percentage of rod cells and in [Na+]i during SMI in the
absence ($ ; n l 17) and presence (# ; n l 18) of 1 µM HMA. HMA did not
influence the time to rigor onset, but suppressed the increase in [Na+]i
significantly. Values are meanspS.E.M. Significance : *P 0.05 compared with
values at 0 min.
due mainly to Na+ influx via NHE-1, and was blunted
after energy depletion. On the basis of these results, we
proposed a hypothesis that the Na+ influx via NHE-1 is
energy dependent, and that after energy depletion Na+
influx is reduced because of inhibition of NHE-1.
To examine this possibility, the time course of changes
in [Mg#+]i was monitored during SMI (Figure 2). The
control level of [Mg#+]i was 0.95p0.13 mM, and this
remained constant during a 60 min perfusion (n l 12).
# 2001 The Biochemical Society and the Medical Research Society
The simplest explanation for the inhibition of the increase
in [Na+]i after energy depletion is that the activity of
NHE-1 is dependent on global [ATP]i. However, several
studies have shown the importance of glycolysis in the
regulation of NHE-1 [17,20,21]. Thus we next asked
whether Na+ influx via NHE-1 is dependent on glycolysis. We examined the effects of the selective inhibition
of glycolysis on the recovery of pHi and the transient
increase in [Na+]i after acidification by an NH Cl pre%
pulse. During the 60 min perfusion of 2-DG, [Mg#+]i
increased to " 1.3 mM, and the percentage of rod-shaped
cells decreased to 89 % (16\18 cells). The contracted cells
were omitted from the following analyses.
Figure 3(a) shows a representative recording of pHi in
control Hepes solution. pHi rose from 7.16 to 7.67 on
exposure of the cell to 15 mM NH Cl ; it then fell rapidly
%
to 6.68 after the removal of NH Cl, and had recovered to
%
7.11 by 10 min after the beginning of the NH Cl pre%
pulse. This recovery of pHi was almost completely
inhibited by 1 µM HMA (results not shown). Figure 3(b)
shows that a small transient rise in [Na+]i (∆[Na+]i-c)
occurred after the removal of NH Cl. The second NH Cl
%
%
pre-pulse in the presence of 100 µM strophanthidin, a
cardiac glycoside, produced a larger and sustained increase in [Na+]i (∆[Na+]i-s). In the presence of 2-DG, the
Glycolysis-dependence of Na+/H+ exchange and intracellular Na+ loading
Figure 3
Representative recordings of the changes in pHi and [Na+]i during an NH4Cl pre-pulse
Panels (a) and (c) show recordings of pHi during the NH4Cl pre-pulse protocol in control Hepes solution (a) and in the presence of 15 mM 2-DG (c). Panels (b) and
(d) show recordings of [Na+]i during the NH4Cl pre-pulse protocol in control Hepes solution (b) and in the presence of 2-DG (d). ∆[Na+]i-c, the small increase in [Na+]i
during the first NH4Cl pre-pulse ; ∆[Na+]i-s, the larger and sustained increase in [Na+]i during the second NH4Cl pre-pulse in the presence of strophanthidin (100 µM).
Effects of inhibition of glycolysis on JH and transient increases in [Na+]i during an NH4Cl pre-pulse
(a) JH values at an identical pHi of 6.8 in control Hepes solution (n l 7) and in 2-DG (n l 6). Panels (b) and (c) show the increases in [Na+]i during the first NH4Cl
pre-pulse (b ; ∆[Na+]i-c), and in the presence of 100 µM strophanthidin (c ; ∆[Na+]i-s) in control Hepes solution (n l 10) and in 2-DG (n l 15). Data are
meanspS.E.M. Significance : *P 0.05 compared with control (unpaired t test).
Figure 4
change in pHi (" 0.6 unit) during the NH Cl pre-pulse
%
was similar to that in the control Hepes solution,
although the pHi level prior to the NH Cl pre-pulse was
%
low because of a gradual decrease during the 60 min
preincubation. The recovery of pHi from intracellular
acidosis was obviously slowed, and the pHi did not
recover fully to the level before the NH Cl pre-pulse
%
during the protocol (Figure 3c). The transient increase in
# 2001 The Biochemical Society and the Medical Research Society
247
248
H. Satoh and others
increases in [Na+]i following the removal of NH Cl. All
%
of the JH values and increases in [Na+]i in the absence
(∆[Na+]i-c) and presence (∆[Na+]i-s) of strophanthidin
were significantly suppressed in the presence of 2-DG
(P 0.05 by unpaired t test). It is concluded that the
inhibition of glycolysis can actually prevent the increase
in [Na+]i that occurs due to Na+ influx via NHE-1.
Dependence on glycolysis of the increase in
[Na+]i during SMI
Finally, we applied the results obtained to the changes in
[Na+]i that occur when both glycolysis and oxidative
phosphorylation are blocked. For this purpose, the time
course of [Na+]i during SMI was monitored in the
presence of 2-DG. The time to rigor contracture in
the presence of 2-DG was shortened to 6 min (P 0.01
compared with value in the absence of 2-DG ; unpaired
t test) (Figure 5b). [Na+]i did not change significantly
during the 60 min preincubation with 2-DG. The representative example of the changes in [Na+]i demonstrates
that, in the presence of 2-DG, the initial increase in [Na+]i
during SMI was suppressed (until 6 min), but that
[Na+]i increased greatly after rigor contracture occurred
(Figure 5a). Figures 5(b) and 5(c) summarize the time
courses of the percentage of rod cells and of [Na+]i respectively in the presence or absence of 2-DG. The
increase in [Na+]i before rigor contracture was not
significant in the presence of 2-DG, and the levels of
[Na+]i were much lower than those in the absence of
2-DG (7.5p1.4 mM and 12.2p1.0 mM respectively at
6 min of SMI ; P 0.05 by two-way ANOVA). These
results indicate that intrinsic glycolysis is crucial for the
increase in [Na+]i that occurs during SMI.
DISCUSSION
Figure 5 Effects of the inhibition of glycolysis on the changes
in cell morphology and in [Na+]i during SMI
(a) A typical recording of the changes in longitudinal cell length () and [Na+]i
(
) during SMI in the presence of 2-DG. The initial rise in [Na+]i during SMI was
suppressed, but a larger increase in [Na+]i occurred after rigor contracture. Panels
(b) and (c) show summarized data for changes in the percentage of rod cells and
in [Na+]i during SMI in the absence ($ ; n l 17) and the presence
(# and = ; n l 14) of 2-DG. The data in the absence of 2-DG were same as
in Figure 1. The time to rigor onset was shortened in the presence of 2-DG, but
the increase in [Na+]i before rigor contracture was not significant, and the levels
of [Na+]i were much lower compared with those in the absence of 2-DG. Values
are meanspS.E.M. Significance : *P 0.05 compared with values at 0 min.
[Na+]
following the removal of NH Cl was also
i
%
eliminated by 2-DG (Figure 3d). The activation of Na+
extrusion via the Na+\K+ pump (by the accumulation of
K+ in transverse tubules) is negligible, because a transient
rise in [Na+]i was not observed even in the presence of
strophanthidin. Figure 4 summarizes the JH values and
# 2001 The Biochemical Society and the Medical Research Society
In this study, we demonstrate the importance of glycolysis for the activation of NHE-1 and of Na+ loading
during SMI in intact ventricular myocytes. The major
findings are that : (1) the increase in [Na+]i during SMI
was due mainly to Na+ influx via NHE-1, but this influx was reduced after energy depletion ; (2) the selective
inhibition of glycolysis prevented both the recovery of
pHi from intracellular acidosis and the transient increase
in [Na+]i during an NH Cl pre-pulse protocol ; and (3)
%
the additional blockade of intrinsic glycolysis prevented the increase in [Na+]i during SMI.
Measurement of intracellular ion
concentrations
First, we have to consider possible factors that may
perturb the measurement of ion concentrations ; these
include : (1) compartmentalization of dye, (2) incomplete
hydrolysis of intracellular accumulated dye, and (3)
photobleaching of dye fluorescence. The residual fluor-
Glycolysis-dependence of Na+/H+ exchange and intracellular Na+ loading
escence intensities of the dyes after treatment with
20 µM digitonin were " 10 % of the pre-exposed control
levels [5,24]. Myocytes were incubated for a further
30 min after dye loading to allow formation of the
hydrolysed fluorescent dye forms, and then in vivo
calibration was applied. Illumination from the xenon
lamp was eliminated with neutral-density filters. Secondly, an important Na+ gradient exists close to the cell
membrane under normal conditions [28]. Finally, the
intracellular acidification that occurs during SMI could
affect the Kd values of fluorescent indicators [29]. We
showed previously that, while the fluorescences excited
at 340 and 380 nm were smaller at lower pH, the change
in the fluorescence ratio was less than 10 % [30].
Estimation of global [ATP]i by changes in
[Mg2+]i
Direct measurements of [ATP]i are feasible in single cells,
using the chemiluminescent couple luciferin\luciferase.
However, this approach is technically difficult and has
some problems [25,31]. Alternatively, several previous
studies have shown that [Mg#+]i is very sensitive to
changes in the energy state in metabolically compromised
myocytes [25,26]. However, the use of [Mg#+]i as an
index of the reciprocal changes in global [ATP]i has some
limitations. First, in the presence of 2-DG, Pi is sequestered as 2-DG 6-phosphate, and free ADP should
increase. Therefore the binding of free ADP to Mg#+
could blunt the increase in [Mg#+]i. Secondly, there is
cell-to-cell heterogeneity in the loss of ATP during
hypoxia [31]. Since we did not monitor [Mg#+]i and
[Na+]i (or pHi) in the same cells, such asynchronous
responses would affect the result. Thirdly, mag-fura 2 is
known to compartmentalize into subcellular organelles
and to interact with Ca#+. Subcellular Ca#+ accumulation
during SMI would interfere with the accurate measurement of [Mg#+]i. However, the affinity of ADP for Mg#+
is much lower than that of ATP [25], and the changes in
[Mg#+]i were relatively homogeneous among cells under
our experimental conditions. Further, the [Ca#+]i level
needed to cause direct interference by Ca#+ with magfura 2 was reported to be 1 µM [32], and we verified
that electrical stimulation at high frequency (10 Hz) did
not alter the ratio of the dye (results not shown).
Preferential dependence of NHE-1 on ATP
from glycolysis
In cardiac Purkinje fibres, which contain more glycogen
than ventricular muscle, Wu and Vaughan-Jones [20]
reported that the recovery in pHi and the rise in
intracellular Na+ activity during an NH Cl pre-pulse
%
were both significantly slowed in the presence of 2-DG
or iodoacetate. In the present study, we verified their
result in ‘ ventricular myocytes ’ more precisely using
pHi- and [Na+]i-sensitive fluorescent dyes. Wu and
Vaughan-Jones [20] also demonstrated that pHi acidified
slowly on the application of 2-DG, as we observed (see
also Figure 3c). The acidification by 2-DG could not be
explained solely by the decreased acid extrusion via
NHE-1, because the fall in pHi during the perfusion of
HMA was not significant [24]. Bountra et al. [33]
suggested that an inhibition of glycolysis would stimulate
aerobic respiration, which leads to the excessive generation of carbonic acid. This carbonic acid would be
hydrated intracellularly, and would lower pHi.
Before confirming the idea that the activity of NHE-1
depends predominantly on [ATP]i generated by glycolysis, it was necessary to rule out several other
possibilities. First, the inhibition of NHE-1 by 2-DG
may simply result from a decrease in global [ATP]i. After
60 min of perfusion with 20 mM 2-DG, [Mg#+]i reached
a value of " 1.3 mM (" 1.5-fold increase compared with
the control value). Demaurex et al. [34] studied the ATPdependence of NHE-1 using the whole-cell patch–clamp
technique, and showed that half-maximal activation of
the exchanger occurred at " 5 mM ATP. Although it is
impossible to define a precise quantitative relationship
between [Mg#+]i and [ATP]i, an NMR study indicated
that a 1.5-fold increase in [Mg#+]i is equivalent to an
" 15 % decrease in [ATP]i [27]. If the basal [ATP]i was
assumed to be 5–10 mM [25], then the level of [ATP]i in
the presence of 2-DG would be 4.3–8.5 mM, and this fall
in [ATP]i would be insufficient to inhibit NHE-1
significantly. Therefore the inhibitory effect of 2-DG on
NHE-1 cannot be explained solely by the decrease in
global [ATP]i. Secondly, the rise in [Na+]i or [Ca#+]i
could affect the activity of NHE-1 by altering the transsarcolemmal Na+ gradient, and by activating certain
protein kinases [11,15,20]. We demonstrated previously
that no changes occurred in [Na+]i or [Ca#+]i during a
60 min perfusion with 2-DG [24]. Finally, we could not
rule out the possibility that some metabolites produced
by the inhibitory effect of 2-DG on the glycolytic cascade
inhibited NHE-1 directly. However, a previous study
showed that the depressed activity of NHE-1 after
inhibition with 2-DG and oligomycin could be readily
restored by addition of ATP to the patch pipette [34].
The predominant effects of ATP depletion on NHE-1
are to alter the sensitivity of the exchanger to intracellular
H+ and to lower the maximal rate of transport [34]. The
cytosolic tail seems to be involved in the regulation of
NHE-1, as it contains numerous potential sites for
phosphorylation by various protein kinases, and a
Ca#+\calmodulin-binding site. The decrease in phosphocreatine (or increase in creatine and Pi) might also relate
directly to the inhibition of NHE-1.
Dependence on glycolysis of Na+ loading
during myocardial ischaemia
An increase in [Na+]i during myocardial ischaemia is
crucial for ischaemia\reperfusion cell injury because it
causes Ca#+ overload through Na+\Ca#+ exchange [2,3].
# 2001 The Biochemical Society and the Medical Research Society
249
250
H. Satoh and others
One of the main pathways for Na+ influx during
ischaemia is expected to be NHE-1. Indeed, extensive
studies using various NHE-1 inhibitors such as amiloride
and its analogues, and more potent inhibitors such as
HOE694 and cariporide, have shown protective effects
against reperfusion myocardial injury in a variety of
experimental models and in some clinical trials [35,36].
However, it remains controversial as to whether [Na+]i
actually increases during the ischaemic period or does so
only after reperfusion [7,9].
During hypoxia (or SMI), glycolysis continues until
the available intrinsic glycogen is exhausted, even if
glucose is removed from the perfusate (see Figure 2). This
sustained anaerobic glycolysis will produce a rapid
decline in pHi. When the extracellular pH is kept
constant, the H+ gradient is expected to increase in a
time-dependent fashion and to activate Na+ influx via
NHE-1 [35]. During ischaemia, Lazdunski et al. [10]
originally hypothesized that extracellular acidosis, which
would develop gradually following intracellular acidosis,
could suppress the activity of NHE-1 by reducing the
trans-sarcolemmal H+ gradient. However, NHE-1 can
still operate under conditions of low extracellular pH
[12,30]. During ischaemia, the high rate of glycolysis is
not sustained, and decreases despite the presence of
glycogen stores, since excessive intracellular acidosis and
the accumulation of some metabolic products would
lower glycolytic activity [37]. In the present study, the
inhibition of glycolysis suppressed the increase in [Na+]i
during SMI. On the basis of our results, the increase in
[Na+]i during ischaemia is expected to depend on the
residual activity of glycolysis, and this concept could
partly explain the inconsistency in the changes in [Na+]i
observed in many experimental models.
REFERENCES
1
2
3
4
5
6
7
8
9
10
11
12
Conclusions
In conclusion, our results suggest that ATP generated by
glycolysis is essential to activate NHE-1, and support the
idea of ‘ ATP compartmentation ’ in the regulation of
NHE-1 in isolated ventricular myocytes. The dependence of NHE-1 on glycolysis could also affect the
changes in [Na+]i during myocardial ischaemia. More
detailed studies should be carried out in order to clarify
the structural and functional features of NHE-1.
13
14
15
16
ACKNOWLEDGMENTS
This study was supported by a Japan Heart Foundation
and Pfizer Pharmaceuticals Grant for Research on
Coronary Artery Disease, and by Grant-in-Aid 11670670
from the Ministry of Education, Culture, and Science of
Japan (H. S.).
# 2001 The Biochemical Society and the Medical Research Society
17
18
Allen, D. G. and Orchard, C. H. (1983) Intracellular
calcium concentration during hypoxia and metabolic
inhibition in mammalian ventricular muscle. J. Physiol.
(Cambridge, U.K.) 339, 107–122
Tani, M. and Neely, J. R. (1989) Role of intracellular Na+
in Ca#+ overload and depressed recovery of ventricular
function of reperfused ischemic rat heart. Circ. Res. 65,
1045–1056
Anderson, S. E., Murphy, E., Steenbergen, C., London,
R. E. and Cala, P. M. (1990) Na-H exchange in
myocardium : effects of hypoxia and acidification on Na
and Ca. Am. J. Physiol. 259, C940–C948
Haigney, M. C. P., Miyata, H., Lakatta, E. G., Stern,
M. D. and Silverman, H. S. (1992) Dependence of hypoxic
cellular calcium loading on Na+-Ca#+ exchange. Circ.
Res. 71, 547–557
Satoh, H., Hayashi, H., Katoh, H., Terada, H. and
Kobayashi, A. (1995) Na+\H+ and Na+\Ca#+ exchange in
regulation of [Na+] and [Ca#+] during metabolic
i
i
inhibition. Am. J. Physiol. 268, H1239–H1248
Shivkumar, K., Deutsch, N. A., Lamp, S. T., Khuu, K.,
Goldhaber, J. I. and Weiss, J. N. (1997) Mechanism of
hypoxic K loss in rabbit ventricle. J. Clin. Invest. 100,
1782–1788
Pike, M. M., Luo, C. S., Clark, M. D. et al. (1993) NMR
measurements of Na+ and cellular energy in ischemic rat
heart : role of Na+-H+ exchange. Am. J. Physiol. 265,
H2017–H2026
Imahashi, K., Hashimoto, K., Yamaguchi, H., Nishimura,
T. and Kusuoka, H. (1998) Alteration of intracellular Na+
during ischemia in diabetic rat hearts : the role of reduced
activity in Na+\H+ exchange against stunning. J. Mol.
Cell. Cardiol. 30, 509–517
Vanheel, B., DeHemptinne, A. and Leusen, I. (1990)
Acidification and intracellular sodium ion activity during
simulated myocardial ischemia. Am. J. Physiol. 259,
C169–C179
Lazdunski, M., Frelin, C. and Vigne, P. (1985) The
sodium\hydrogen exchange system in cardiac cells : its
biochemical and pharmacological properties and its role
in regulating internal concentrations of sodium and
internal pH. J. Mol. Cell. Cardiol. 17, 1029–1042
Moolenaar, W. H. (1986) Effects of growth factors on
intracellular pH regulation. Annu. Rev. Physiol. 48,
363–376
Vaughan-Jones, R. D. and Wu, M.-L. (1990) Extracellular
H+ inactivation of Na+-H+ exchange in the sheep cardiac
Purkinje fibre. J. Physiol. (Cambridge, U.K.) 428,
441–466
Sardet, C., Counillon, L., Franchi, A. and Pouysse! gur, J.
(1990) Growth factors induce phosphorylation of the
Na+\H+ antiporter, a glycoprotein of 110 kD. Science
247, 723–725
Bertrand, B., Wakabayashi, S., Ikeda, T., Pouysse! gur, J.
and Shigekawa, M. (1994) The Na+\H+ exchanger
isoform 1 (NHE1) is a novel member of the calmodulinbinding proteins. J. Biol. Chem. 269, 13703–13709
Fliegel, L. and Wang, H. (1997) Regulation of the
Na+\H+ exchange in the mammalian myocardium.
J. Mol. Cell. Cardiol. 29, 1991–1999
Sabri, A., Byron, K. L., Samarel, A. M., Bell, J. and
Lucchesi, P. A. (1998) Hydrogen peroxide activates
mitogen-activated protein kinases and Na+-H+ exchange
in neonatal rat cardiac myocytes. Circ. Res. 82,
1053–1062
Cassel, D., Katz, M. and Rotman, M. (1986) Depletion of
cellular ATP inhibits Na+\H+ antiport in cultured human
cells. J. Biol. Chem. 261, 5460–5466
Little, P. J., Weissberg, P. L., Cragoe, E. J. and Bobik, A.
(1988) Dependence of Na+-H+ antiport activation in
cultured rat aortic smooth muscle on calmodulin, calcium
and ATP. J. Biol. Chem. 263, 16780–16786
Glycolysis-dependence of Na+/H+ exchange and intracellular Na+ loading
19
20
21
22
23
24
25
26
27
28
Brown, S. E. S., Heming, T. A., Benedict, C. R. and
Bidani, A. (1991) ATP sensitive Na+-H+ antiport in
Type II alveolar epithelial cells. Am. J. Physiol. 261,
C954–C963
Wu, M.-L. and Vaughan-Jones, R. D. (1994) Effect of
metabolic inhibitions and second messengers upon
Na+-H+ exchange in the sheep cardiac Purkinje fibre.
J. Physiol. (Cambridge, U.K.) 478, 301–313
Hilgemann, D. W. (1997) Cytoplasmic ATP-dependent
regulation of ion transporters and channels : Mechanisms
and messengers. Annu. Rev. Physiol. 59, 193–220
Glitsch, H. G. and Tappe, A. (1993) The Na+\K+ pump
of cardiac Purkinje cells is preferentially fuelled by
glycolytic ATP production. Pflu$ gers Arch. 422, 380–385
Dizon, J., Burkhoff, D., Tauskela, J., Whang, J., Cannon,
P. and Katz, J. (1998) Metabolic inhibition in the perfused
rat heart : evidence for glycolytic requirement for normal
sodium homeostasis. Am. J. Physiol. 274, H1082–H1089
Sugiyama, S., Satoh, H., Nomura, N., Terada, H.,
Watanabe, H. and Hayashi, H. (2001) The importance of
glycolysis-derived ATP for the Na+\H+ exchange activity
in guinea pig ventricular myocytes. Mol. Cell. Biochem.
217, 153–161
Leyssens, A., Nowieky, A. V., Patterson, L., Crompton,
M. and Duchen, M. R. (1996) The relationship between
mitochondrial state, ATP hydrolysis, [Mg#+] and [Ca#+]
i
i
studied in isolated rat cardiomyocytes. J. Physiol.
(Cambridge, U.K.) 496, 111–128
Silverman, H. S., Lisa, F. D., Hui, R. C. et al. (1994)
Regulation of intracellular free Mg#+ and contraction in
single adult mammalian cardiac myocytes. Am. J. Physiol.
266, C222–C233
Murphy, E., Steenbergen, C., Levy, L. A., Raju, B. and
London, R. E. (1989) Cytosolic free magnesium levels in
ischemic rat heart. J. Biol. Chem. 264, 5622–5627
Carmeliet, E. (1992) A fuzzy subsarcolemmal space for
intracellular Na+ in cardiac cells. Cardiovasc. Res. 26,
433–442
29
30
31
32
33
34
35
36
37
Lattanzio, Jr, F. A. (1990) The effects of pH and
temperature on fluorescent calcium indicators as
determined with Chelex-100 and EDTA buffered system.
Biochem. Biophys. Res. Commun. 171, 102–108
Nakamura, T., Hayashi, H., Satoh, H., Katoh, H.,
Kaneko, M. and Terada, H. (1999) A single cell model of
myocardial reperfusion injury : changes in intracellular
Na+ and Ca#+ concentrations in guinea pig ventricular
myocytes. Mol. Cell. Biochem. 194, 147–157
Bowers, K. C., Allshire, A. P. and Cobbold, P. H. (1993)
Continuous measurement of cytoplasmic ATP in single
cardiomyocytes during simulation of the ‘ oxygen
paradox ’. Cardiovasc. Res. 27, 1836–1839
Hurley, T. W., Ryan, M. P. and Brinck, R. W. (1992)
Changes of cytosolic Ca#+ interfere with measurements
of cytosolic Mg#+ using mag-fura-2. Am. J. Physiol. 263,
C300–C307
Bountra, C., Powell, T. and Vaughan-Jones, R. D. (1990)
Comparison of intracellular pH transients in single
ventricular myocytes and isolated ventricular muscle of
guinea-pig. J. Physiol. (Cambridge, U.K.) 424, 343–365
Demaurex, N., Romanek, R. R., Orlowski, J. and
Grinstein, S. (1997) ATP dependence of Na+\H+
exchange nucleotide specificity and assessment of the role
of phospholipids. J. Gen. Physiol. 109, 117–128
Karmazyn, M., Gan, X. T., Humphreys, R. A., Yoshida,
H. and Kusumoto, K. (1999) The myocardial Na+-H+
exchange : structure, regulation, and its role in heart
disease. Circ. Res. 85, 777–786
Theroux, P., Chaitman, B. R., Danchin, N. et al. (2000)
Inhibition of the sodium-hydrogen exchanger with
caliporide to prevent myocardial infarction in high-risk
ischemic situations. Main results of the GURDIAN trial.
Circulation 102, 3032–3038
Allen, D. G. and Orchard, C. H. (1987) Myocardial
contractile function during ischemia and hypoxia. Circ.
Res. 60, 153–168
Received 9 January 2001/27 March 2001; accepted 4 May 2001
# 2001 The Biochemical Society and the Medical Research Society
251