Am J Physiol Heart Circ Physiol 292: H820 –H829, 2007;
doi:10.1152/ajpheart.00366.2006.
Downmodulation of mitochondrial F0F1 ATP synthase by diazoxide in cardiac
myoblasts: a dual effect of the drug
Marina Comelli,1,2,3 Giuliana Metelli,1 and Irene Mavelli1,2,3
1
Department of Biomedical Sciences and Technologies, 2Microgravity, Ageing, Training and Immobility Center of Excellence,
and 3Interdepartmental Center for Regenerative Medicine, University of Udine, Udine, Italy
Submitted 7 April 2006; accepted in final form 22 September 2006
(IPC) induces a state of cardioprotection against subsequent prolonged ischemia-reperfusion injury. The cardioprotective mechanism of IPC is complex and
involves a variety of triggers that activate multiple signaling
pathways, most converging on mitochondria (37). The cellular
survival program activated by IPC integrates several processes,
including opening of ATP-sensitive potassium channels (20,
56), regulation of fatty acid metabolism (28), production of
reactive oxygen species (41), and regulation of mitochondrial
permeability transition (25). Moreover, IPC elicits protective
effects on cellular bioenergetic status by downmodulating the
ATPase activity of mitochondrial ATP synthase, thereby promoting cell survival after ischemic insult (2, 6, 14, 24, 43, 55).
Pharmacological preconditioning is a cardioprotective state
similar to IPC. Diazoxide, an antihypertensive and antihypoglycemic drug, is commonly used to induce preconditioning in
animal models of ischemia-reperfusion (17, 20). This hydrophobic compound passively enters cells and organelles like
mitochondria. Functionally, diazoxide is commonly considered
as a selective opener of mitochondrial ATP-sensitive potassium (mitoKATP) channels (18, 19, 20), but at relatively high
concentrations (⬎100 ␮M) the drug has other effects, including the inhibition of succinate dehydrogenase (22, 48). Interestingly, similar to IPC, diazoxide has beneficial bioenergetic
consequences in perfused rat hearts and isolated rat heart
mitochondria (2, 6, 33).
Activation of mitoKATP channels is considered to be a major
trigger or end effector of IPC (19, 20, 40), even if some authors
have provided renewed support for a role of sarcolemmal KATP
channels in ischemic cardioprotection (47, 51). Evidence for
the existence of mitoKATP channels and for their involvement
in IPC derives from pharmacological studies, since KATP
channel openers (e.g., diazoxide) mimic IPC and KATP channel
blockers (e.g., glibenclamide) inhibit IPC (17, 19, 20, 23).
However, the molecular structure of mitoKATP channels is not
clear yet (12, 34), despite recent advances in understanding its
structure (3, 30). Moreover, alternative mechanisms have been
recently overviewed (4, 23).
Mitochondrial F0F1 ATP synthase is the major producer of
ATP for contractile function and ionic homeostasis in cardiomyocytes (13). When oxygen deprivation collapses the mitochondrial electrochemical gradient, F0F1 ATP synthase
switches from ATP synthesis to ATP hydrolysis and thus,
during severe ischemia, is a major consumer of ATP (13).
Therefore, inhibition of F0F1 ATP synthase during ischemia,
which should conserve myocardial ATP, may play a key role in
the energy-sparing effect elicited by ischemic and pharmacological preconditioning (2, 6, 24, 55). Numerous authors have
suggested that inhibition of the ATPase activity of F0F1 ATP
synthase is carried out by the enzyme’s natural inhibitor
protein, IF1 (2, 7, 24, 55). IF1 is a noncompetitive inhibitor that
reversibly binds with 1:1 stoichiometry to the ␤-subunit of F1,
the catalytic sector of F0F1 ATP synthase; binding requires
ATP hydrolysis associated with loss of proton motive force and
is favored by a low mitochondrial electrochemical gradient
(32). We recently documented (14, 43) that IF1 binding to F0F1
ATP synthase increases during IPC in goat heart in vivo. In
addition, by studying both the purified enzyme and isolated
mitochondrial membranes from beef heart, we demonstrated
Address for reprint requests and other correspondence: I. Mavelli, Dipartimento di Scienze e Tecnologie Biomediche, Università di Udine, Piazzale
Kolbe 4, I-33100 Udine, Italy (e-mail: [email protected]).
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked “advertisement”
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
energy metabolism; H9c2 myoblasts; inhibitor protein IF1
ISCHEMIC PRECONDITIONING
H820
0363-6135/07 $8.00 Copyright © 2007 the American Physiological Society
http://www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.32.246 on June 15, 2017
Comelli M, Metelli G, Mavelli I. Downmodulation of mitochondrial F0F1 ATP synthase by diazoxide in cardiac myoblasts: a dual
effect of the drug. Am J Physiol Heart Circ Physiol 292: H820 –H829,
2007; doi:10.1152/ajpheart.00366.2006.—Similar to ischemic preconditioning, diazoxide was documented to elicit beneficial bioenergetic consequences linked to cardioprotection. Inhibition of ATPase
activity of mitochondrial F0F1 ATP synthase may have a role in such
effect and may involve the natural inhibitor protein IF1. We recently
documented, using purified enzyme and isolated mitochondrial membranes from beef heart, that diazoxide interacts with the F1 sector of
F0F1 ATP synthase by promoting IF1 binding and reversibly inhibiting ATP hydrolysis. Here we investigated the effects of diazoxide on
the enzyme in cultured myoblasts. Specifically, embryonic heartderived H9c2 cells were exposed to diazoxide and mitochondrial
ATPase was assayed in conditions maintaining steady-state IF1 binding (basal ATPase activity) or detaching bound IF1 at alkaline pH.
Mitochondrial transmembrane potential and uncoupling were also
investigated, as well as ATP synthesis flux and ATP content. Diazoxide at a cardioprotective concentration (40 ␮M cell-associated concentration) transiently downmodulated basal ATPase activity, concomitant with mild mitochondria uncoupling and depolarization, without affecting ATP synthesis and ATP content. Alkaline stripping of IF1
from F0F1 ATP synthase was less in diazoxide-treated than in untreated
cells. Pretreatment with glibenclamide prevented, together with mitochondria depolarization, inhibition of ATPase activity under basal but not
under IF1-stripping conditions, indicating that diazoxide alters alkaline
IF1 release. Diazoxide inhibition of ATPase activity in IF1-stripping
conditions was observed even when mitochondrial transmembrane potential was reduced by FCCP. The results suggest that diazoxide in a
model of normoxic intact cells directly promotes binding of inhibitor
protein IF1 to F0F1 ATP synthase and enhances IF1 binding indirectly by
mildly uncoupling and depolarizing mitochondria.
DUAL EFFECT OF DIAZOXIDE ON MITOCHONDRIAL F0F1 ATP SYNTHASE
(9) that interaction of diazoxide with the ␤-subunit of the F1
sector promotes binding with IF1 and reversibly inhibits ATP
hydrolysis. In the present study, we continued our investigation
of diazoxide’s effect on F0F1 ATP synthase, using a biological
model of embryonic heart-derived myoblasts (H9c2). We provided evidence for a dual effect of diazoxide resulting in
enhancement of IF1 binding to the enzyme and downmodulation of ATPase activity, being a mild but detectable mitochondrial depolarization and uncoupling factor triggering such
binding.
METHODS
AJP-Heart Circ Physiol • VOL
and is reported as units (micromoles of ATP per minute) per milligram of protein. Furthermore, the buffer composition (containing
EGTA and ⬍5 mM Na⫹ concentration) was chosen to minimize
interference from Ca2⫹- and Na⫹-K⫹-ATPases (8); this was verified
experimentally by performing the assay with 10 ␮M sodium orthovanadate and 2 mM ouabain, respectively. Oligomycin-sensitive
ATPase activity, indicative of correct coupling of F0 and F1 moieties
of the F0F1 ATP synthase complex, was assayed in the presence of 4
␮M oligomycin (8). Oligomycin concentration was chosen on the
basis of a dose-dependence study of the ATPase activity inhibitory
effect, which permitted us to avoid using submaximal concentrations.
Cell-associated diazoxide concentration. H9c2 cells were incubated with 100 ␮M diazoxide in culture medium for 0 –30 min and
harvested by trypsinization. Cell pellets were extracted with 0.33 N
perchloric acid (500 ␮l per 10-cm dish).
Cell-associated diazoxide concentration was determined by an
HPLC method (54) using a LiChrosorb RP-8 column (10 cm ⫻ 4.6
mm internal diameter; particle size, 5 ␮m; Chrompack, Middelburg,
The Netherlands). Calibration curves, obtained by adding known
amounts of diazoxide to the culture medium immediately before
trypsinization, were linear for the range 0.02–10 ␮g (r ⫽ 0.966); the
sensitivity limit was 0.3 ␮M. The recovery of diazoxide from perchloric acid extracts was 85.2% (SD 0.9) (n ⫽ 4), based on a standard
curve. Diazoxide for standard solutions was dissolved in dimethyl
sulfoxide and diluted in water or culture medium to a final concentration of 0.01– 0.05%.
Mitochondrial respiration assays. Mitochondrial basal respiration
in intact (nonpermeabilized) H9c2 cells, treated with 100 ␮M diazoxide or vehicle for 10 and 30 min, was measured in 3 ⫻ 106 cells
suspended in 2 ml of culture medium. O2 consumption was measured
in an oxygraphic cell at 37°C with a Clark-type O2 electrode (Yellow
Springs Instruments, Yellow Springs, OH). The solubility of O2 was
taken to be 200 nmol O2/ml. Addition of 1 mM KCN completely
suppressed O2 consumption, indicating that mitochondrial respiration
was being measured.
To assess the rate of respiration uncoupled from ATP synthesis in
nonpermeabilized cells, FCCP (5 ␮M) was added directly to the
oxygraphic cell. The ratio of FCCP-stimulated respiration to basal
respiration was considered as the respiratory control ratio (RCR). To
assess the rate of respiration coupled to ATP synthesis (ATP synthesis
flux), 10 ␮M oligomycin was added directly to the oxygraphic
chamber containing nontreated or diazoxide-treated cells; ATP synthesis flux was calculated as the difference between basal respiration
rate and that measured in presence of oligomycin. FCCP and oligomycin concentrations were chosen on the basis of dose dependence
studies of the respective effects on the basal respiration rate, which
permitted us to avoid using submaximal concentrations.
ATP content. H9c2 cells were incubated for 10 –30 min with 100
␮M diazoxide, 10 min with 25 ␮M FCCP, or with the respective
vehicles in culture medium. Cells were harvested by trypsinization
and extracted with 0.6 M trichloracetic acid (500 ␮l per 10-cm dish).
ATP content was determined by HPLC according to Pogolotti and
Santi (44), using a strong anion exchange column (Partisil 10 SAX; 25
cm ⫻ 4.6 mm; Farmitalia, Milan, Italy).
Mitochondrial transmembrane potential. Mitochondrial transmembrane potential (⌬␺m) was measured by confocal microscopy and flow
cytometry, using the fluorescent probe 5,5⬘,6,6⬘-tetrachloro-1,1⬘,3,3⬘tetraethylbenzimidazolylcarbocyanine iodine (JC-1; Molecular
Probes, Eugene, OR). Subconfluent cultures (in 2-cm glass or 10-cm
plastic culture plates) were incubated with 100 ␮M diazoxide, 25 ␮M
FCCP, or vehicle for 10 or 30 min in culture medium. In some
experiments, cells were treated with 10 ␮M glibenclamide for 1 h. To
avoid cells being differently deprived of nutrients because of the
different incubation times required for the treatments, the medium was
replaced with fresh complete medium containing 10 ␮g/ml JC-1
before labeling for 5 min at 37°C in the dark.
292 • FEBRUARY 2007 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.32.246 on June 15, 2017
Chemicals and reagents. All materials were purchased from Sigma
(St. Louis, MO), unless otherwise stated.
Cell cultures. H9c2, a clonal line of rat embryonic heart-derived
myoblasts, was obtained from American Type Culture Collection
(CRL-1446; Rockville, MD). The cells were maintained in culture
medium consisting of Dulbecco’s modified Eagle’s medium (EuroClone, Devon, United Kingdom), 10% fetal bovine serum (Biochrom,
Berlin, Germany), penicillin (100 U/ml), streptomycin (100 ␮g/ml),
and glutamine (4 mM). To prevent loss of differentiation potential,
cells were not allowed to become confluent and were monitored for
the absence of tubular structure formation and myogenin expression.
Passages 12–25 were used for all experiments described in this
article. Cell density and viability were determined by Trypan blue dye
exclusion test (10).
Mitochondrial ATPase assays. H9c2 cells, grown in 10-cm tissue
culture plates, were incubated in culture medium for 0 –30 min in the
presence of 10 – 400 ␮M diazoxide or vehicle (0.05% dimethyl sulfoxide). In some experiments, cells were treated with carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP) for 10 min at 25
␮M or with vehicle (0.1% ethanol). In some experiments, cells were
treated with 10 ␮M glibenclamide for 1 h. After treatments, cells were
harvested by trypsinization (0.05% trypsin, 0.02% EDTA in phosphate-buffered saline), pelleted by centrifugation, resuspended in a
diazoxide-free, ice-cold solution of buffer A (mM: 1 MgCl2, 2 EGTA,
20 HEPES pH 7.0) or buffer B (mM: 1 MgCl2, 2 EGTA, 1 NaATP, 20
HEPES pH 9.2), and homogenized by sonication (UP5OG sonifier;
Hielscher, Berlin, Germany). Conditions were controlled to expose
catalytic sector F1 to solvent and to avoid uncoupling from the F0
domain (8). Specifically, ATPase activity was monitored during
ultrasonic treatment in the absence and in the presence of FCCP to
evaluate breaking of mitochondria and formation of submitochondrial
particles (SMP) and in the presence of oligomycin to ensure minimized damage to the enzyme coupling subunits during the treatment
(7). Homogenization at pH 7.0 maintains the binding equilibrium
between IF1 and F0F1 ATP synthase, permitting us to measure “basal”
ATPase activity, while homogenization at pH 9.2 detaches 95% of
bound IF1 (46), permitting us to measure “potential” (uninhibited)
ATPase activity. Cell homogenates were assayed for ATPase immediately. SMP deprived of IF1 were prepared accordingly to Vadineanu
et al. (53). Briefly, cells were homogenized by sonication in buffer B
and centrifuged at 150,000 g for 20 min at 4°C. The resulting pellet,
washed in buffer B, was then exposed to 100 ␮M diazoxide for 10 min
in buffer A or B, and ATPase activity was assayed immediately. Total
protein content was determined according to Lowry et al. (36), with
bovine serum albumin as standard.
Maximal ATPase activity (Vmax) was determined by coupling the
production of ADP to the oxidation of NADH via pyruvate kinase and
lactic dehydrogenase reactions with a diazoxide-free assay buffer
containing an ATP-regenerating system (mM: 50 Tris 䡠 HCl pH 7.4, 30
sucrose, 50 KCl, 4 MgCl2, 2 NaATP, 2 EGTA, 2 phosphoenolpyruvate, and 300 NADH with 3 IU lactate dehydrogenase and 4 IU
pyruvate kinase). Activity was measured by following NADH oxidation at 340 nm on a Perkin-Elmer Vis-UV Spectrometer Lambda14
H821
H822
DUAL EFFECT OF DIAZOXIDE ON MITOCHONDRIAL F0F1 ATP SYNTHASE
For confocal microscopy, labeled cells were washed three times
with culture medium and examined under a laser scanning confocal
microscope (TCS NT; Leica, San Jose, CA). Fluorescence was excited
by the 488-nm line of an argon laser, and emission was recorded at
525 and 590 nm. For quantitative analyses, pixel intensities were
analyzed with MetaMorph software (Crisel Instruments, Rome, Italy).
For flow cytometry, labeled cells were harvested by trypsinization,
washed three times in culture medium, resuspended at 106 cells/ml,
and analyzed (10,000 cells/sample) with a FACScan cytometer (Becton Dickinson, Mannheim, Germany). The excitation wavelength was
488 nm, and emission was monitored at 582 nm. Data were analyzed
with Cell Quest Software (Becton Dickinson, New York, NY).
Statistical analysis. Data are reported as means (SD) unless otherwise indicated. Intergroup comparisons were made with Student’s
t-test for two groups and by one-way ANOVA followed by post hoc
Tukey’s test for multiple groups. A value of P ⬍ 0.05 was considered
to be statistically significant.
RESULTS
We first examined the effects elicited on the ATPase activity
of mitochondrial F0F1 ATP synthase by the exposure of cultured H9c2 myoblasts to diazoxide. When subconfluent cultures were homogenized at pH 7.0, ATPase activity was 0.126
U/mg (SD 0.006) (n ⫽ 20). The presence of 10 ␮M Na
orthovanadate and 2 mM ouabain reduced ATPase activity by
2% and 7%, respectively, confirming that the assay conditions
were specific for mitochondrial ATPase. Nevertheless, the
oligomycin-sensitive ATPase activity, indicative of correctly
assembled enzyme complex, was low [0.063 U/mg (SD
0.002)], although still in the range reported for cell lines
assayed under experimental conditions avoiding mitochondrial
isolation (5). This may suggest that disturbed assembly/stability of F0F1 ATP synthase complex occurred in H9c2 cells,
considering that the homogenization conditions were carefully
controlled to minimize damage to the coupling subunits of the
enzyme.
When H9c2 cells were treated with 0 – 400 ␮M diazoxide for
10 min before homogenization at pH 7.0, a mild but significant
dose-dependent decrease in both oligomycin-sensitive and -inAJP-Heart Circ Physiol • VOL
Fig. 2. Time courses of diazoxide inhibition of mitochondrial ATPase and
cell-associated diazoxide concentration. Cells were treated with 100 ␮M
diazoxide for 0 –30 min. A: ATPase activity in homogenates prepared at pH
7.0. B: cell-associated diazoxide concentration. Data are means (SD) from at
least 4 experiments performed in duplicate. *P ⬍ 0.001 vs. time 0 control
(Student’s t-test); data marked with the same superscript letters are significantly different (P ⬍ 0.05, Tukey’s test).
292 • FEBRUARY 2007 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.32.246 on June 15, 2017
Fig. 1. Dose-dependent effects of diazoxide on mitochondrial ATPase activity
(open bars) and oligomycin-sensitive ATPase activity (gray bars) in H9c2
myoblasts. Cells were exposed to vehicle or 10 – 400 ␮M diazoxide for 10 min
before homogenization at pH 7.0, in the presence or in the absence of
oligomycin (see METHODS). Data are means (SD) of 3 experiments made in
duplicate. *P ⬍ 0.001 vs. control (Student’s t-test).
sensitive ATPase activity was observed at concentrations
known to have cardioprotective effects (50 –200 ␮M) (Fig. 1).
The maximum inhibitory effect (26% and 27% in the two
cases) was achieved at 100 ␮M. Diazoxide had no effect on
cell viability (data not shown).
The time course analysis of the inhibitory effects of 100 ␮M
diazoxide revealed maximal reduction in ATPase activity over
10 min (Fig. 2A). Thereafter, activity remained relatively
constant until it returned to the control value at 30 min. The
profile of cell-associated diazoxide during the time course
paralleled that of the inhibitory effects, with a maximum
concentration at 10 min (Fig. 2B). Concentrations of cellassociated diazoxide at all time points were less than the 100
␮M added to the medium and below those at which the drug
exerts pharmacological effects other than mitoKATP activation
(e.g., inhibition of succinate dehydrogenase).
The fact that diazoxide inhibition of ATPase was transient
pointed to a reversible downmodulation of F0F1 ATP synthase,
probably by naturally occurring IF1, similar to that we previously observed on IPC (14, 43). To test this hypothesis, taking
into account that IF1 binding to F0F1 ATP synthase is optimal
at pH 6.7 and that sonication at pH 9.2 releases 95% of bound
DUAL EFFECT OF DIAZOXIDE ON MITOCHONDRIAL F0F1 ATP SYNTHASE
H823
Fig. 3. Effect of diazoxide exposure of H9c2 cells (A) and of IF1-depleted
submitochondrial particles (SMP; B) on mitochondrial ATPase. A: cell homogenates were prepared at pH 7.0 (physiological conditions of IF1 binding;
open bars) and at pH 9.2 (IF1-stripping conditions; filled bars). Cells were
exposed to vehicle (control), 100 ␮M diazoxide, or 25 ␮M carbonyl cyanide
p-trifluoromethoxyphenylhydrazone (FCCP) for 10 min before homogenization at pH 7.0 or 9.2. B: IF1-depleted SMP from H9c2 were exposed to vehicle
(control) or 100 ␮M diazoxide for 10 min at pH 7.0 (open bars) and pH 9.2
(filled bars). Data are means (SD) from at least 4 experiments performed in
duplicate. *,§P ⬍ 0.001 vs. the pH-specific control (Student’s t-test).
IF1 (46), we compared mitochondrial ATPase activity in homogenates prepared at pH 7.0 and 9.2 (Fig. 3A). In control cells
(treated with vehicle), ATPase activity was only 20% higher
(P ⬍ 0.05) in homogenates made at pH 9.2 (potential ATPase
activity) than at pH 7.0 (basal ATPase activity), indicating a
low level of bound IF1 under physiological conditions. In cells
treated with 100 ␮M diazoxide for 10 min and homogenized at
physiological pH, ATPase activity was 26% less than in
nontreated cells, as shown earlier; homogenization at pH 9.2
increased ATPase activity by only 21%, similarly to nontreated
cells and therefore was unable to overcome the inhibitory
effects of the drug. Treatment with 25 ␮M FCCP (to increase
IF1 binding through the loss of ⌬␺m) also inhibited ATPase by
AJP-Heart Circ Physiol • VOL
Fig. 4. Time course of diazoxide inhibition of mitochondrial ATPase in
homogenates prepared at pH 7.0 and 9.2. A: ATPase activity in homogenates
prepared at pH 7.0 (䊐) and pH 9.2 (E). B: %inhibition of mitochondrial ATPase
by bound IF1 during diazoxide treatment. Calculations are as in RESULTS. Data
are means (SD) from at least 4 experiments performed in duplicate. *P ⬍
0.001 vs. time 0 control (Student’s t-test).
292 • FEBRUARY 2007 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.32.246 on June 15, 2017
25%, showing that free IF1 is available in cells but in a low
amount. FCCP concentrations lower than 25 ␮M had little
effect in this cell system (data not shown). Homogenization at
pH 9.2 fully overcame the protonophore’s inhibitory effects,
returning ATPase activity to that of control cells and confirming that the inhibitory action of FCCP is due to increased IF1
binding. FCCP treatment had no effect on cell viability, and its
vehicle had no effect on ATPase activity (data not shown).
Control experiments using SMP deprived of IF1 showed that
ATPase activity was not affected by diazoxide at both pH 7.0
and 9.2 (Fig. 3B), suggesting that the drug did not interfere per
se with the catalytic activity of the enzyme and confirming the
role of IF1 in the ATPase activity inhibition afforded by
diazoxide.
The time course of diazoxide inhibition of mitochondrial
ATPase assayed in homogenates prepared at pH 9.2 was
similar to that observed in homogenates prepared at physiological pH, with a rapid and significant decrease between 0 and
10 min and recovery at 30 min (Fig. 4A). At all time points,
ATPase activity in alkaline extracts was greater than that in pH
7.0 extracts. With the consideration that alkaline treatment
removes ⬎95% of bound IF1 (46), the ATPase activity observed in control cells at pH 9.2 was taken as the activity of
IF1-free enzyme. On this basis, we calculated the amount of
IF1-inhibited enzyme as the ratio of the enzyme activity mea-
H824
DUAL EFFECT OF DIAZOXIDE ON MITOCHONDRIAL F0F1 ATP SYNTHASE
Fig. 5. Diazoxide increases mitochondrial respiration in H9c2 cells without
affecting ATP synthesis flux or cellular ATP content. Subconfluent cells were
exposed to diazoxide or FCCP, used as a positive control to uncouple
respiration from ATP synthesis. A: mitochondrial respiration (open bars) and
ATP synthesis flux (gray bars). Mitochondrial O2 consumption was registered
in nonpermeabilized cells suspended in culture medium in the absence and
presence of the 2 agents (100 ␮M diazoxide, 5 ␮M FCCP). ATP synthesis flux
was determined as the difference between respiration rate detected in absence
of oligomycin and that detected in presence of 10 ␮M oligomycin. B: cellular
ATP content. HPLC determination of ATP concentration was performed in
acidic extracts of trypsinized cells (100 ␮M diazoxide, 25 ␮M FCCP). Data are
means (SD) of at least 5 experiments performed in duplicate. *P ⬍ 0.005 vs.
untreated controls; °P ⬍ 0.005 vs. diazoxide-treated cells (Student’s t-test).
AJP-Heart Circ Physiol • VOL
only the oligomycin-sensitive respiration, i.e., respiration coupled to ATP synthesis (ATP synthesis flux), an overall low rate
[2.1 nmol O2 䡠min⫺1 䡠mg protein⫺1 (SD 0.2)] was observed in
control cells. This suggests that the proton gradient driving
ATP formation was not perfectly tight to oxidative phosphorylation and that the ATP synthesis control over respiration was
quite low. The low coupling might depend on both basal proton
leak and disturbed assembly/stability of F0F1 ATP synthase
complex, which we suggested above to occur in control cells
on the basis of the observed low value of oligomycin-sensitive
ATPase activity. When cells were incubated with 100 ␮M
diazoxide for 10 min, ATP synthesis flux increased but not
significantly so. As expected, 10 min of 25 ␮M FCCP blunted
ATP synthesis flux. Finally, in cells treated with 100 ␮M
diazoxide for 10 min, cellular ATP content showed a negligible, nonsignificant increase (Fig. 5B), while in cells treated
with 25 ␮M FCCP for 10 min ATP content was significantly
reduced. Thus, in normoxic cells exposed to diazoxide, mitochondrial F0F1 ATP synthase normally synthesized ATP despite a significantly reduced Vmax of ATP hydrolysis.
The fact that diazoxide did not reduce ATP levels agrees
with its small effect on ATP synthesis flux and suggests that
the transient uncoupling is too moderate to significantly alter
mitochondrial energetics. If this is the case, membrane depolarization should also be moderate. Therefore, we investigated
the effects of diazoxide on ⌬␺m by monitoring the fluorescence
intensity of the lipophilic cation JC-1. At confocal microscopy,
control cells had punctuate red staining indicative of normal
mitochondrial uptake of JC-1 driven by high ⌬␺m (Fig. 6A).
After treatment with diazoxide (10 min, 100 ␮M), cells maintained punctuate staining but were predominantly green with
some yellow, indicative of lower ⌬␺m; quantification of red
fluorescence intensity (590 nm) indicated a 23% (SD 2) decrease in ⌬␺m compared with control cells. Cells treated with
FCCP had blotchy greenish-yellow fluorescence, indicating
lower ⌬␺m but also undefined mitochondrial edges and leakage
of dye. Similar results were observed at flow cytometry:
diazoxide treatment for 10 min altered the peak of the red
fluorescence intensity distribution (Fig. 6B). Mean fluorescence intensity was 26% (SD 2) less than in control cells (P ⬍
0.001), in agreement with confocal microscopy results. The
intensity profile for a 30-min treatment with diazoxide was
similar to that of control cells. FCCP treatment resulted in a
54% (SD 3) reduction in mean red fluorescence intensity (P ⬍
0.001 vs. control cells). These results are in agreement with the
effects of diazoxide on mitochondrial ATPase, and they suggest that diazoxide treatment of H9c2 cells moderately and
transiently decreases ⌬␺m, similar to its inhibition of F0F1 ATP
synthase.
Because of its hydrophobic nature, diazoxide passes through
membranes and may interact with F0F1 ATP synthase irrespectively of its effects on mitoKATP channels or mitochondrial
depolarization. To test this hypothesis, H9c2 cells were preexposed for 1 h to glibenclamide before treatment with 100 ␮M
diazoxide. When compared with no preexposure, glibenclamide treatment prevented the diazoxide-induced loss of
⌬␺m, as indicated by stable JC-1 red fluorescence intensity
measured with flow cytometry (Fig. 7A); glibenclamide alone
had no effect on ⌬␺m. Glibenclamide pretreatment also
blocked diazoxide’s time-dependent inhibition of mitochondrial ATPase activity, assessed in homogenates made at pH 7.0
292 • FEBRUARY 2007 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.32.246 on June 15, 2017
sured after sonication at pH 7.0 to the value of the activity of
IF1-free enzyme. With respect to the 20% inhibition of the
ATPase activity of F0F1 ATP synthase observed in nontreated
cells during diazoxide treatment, we calculated 41% of IF1inhibited enzyme (at 10 and 20 min), which then returned to
20% at 30 min (Fig. 4B).
To determine whether the effects of diazoxide involved
uncoupling of mitochondrial respiration, we assayed the CNsensitive O2 consumption of intact suspended H9c2 cells.
Treatment with 100 ␮M diazoxide for 10 min, but not 30 min,
significantly increased respiration, indicating a transient uncoupling effect (Fig. 5A). Respiration in the presence of 5 ␮M
FCCP was significantly higher than in both control and diazoxide-treated cells and permitted us to confirm, on the basis of
the RCR [1.8 (SD 0.2)], that the measurements were typical of
intact (nonpermeabilized) cells. With the consideration that
DUAL EFFECT OF DIAZOXIDE ON MITOCHONDRIAL F0F1 ATP SYNTHASE
H825
(Fig. 7B). Thus, when mitoKATP channels are blocked, diazoxide is unable to transiently induce mitochondrial depolarization
and inhibit ATPase.
We further compared mitochondrial ATPase activity in
homogenates prepared at pH 7.0 and pH 9.2 from cells
subjected to various pharmacological treatments (Fig. 8).
Figure 8A demonstrates that, as already shown, ATPase
activity from control and diazoxide-treated cells was ⬃20%
higher (P ⬍ 0.05) in homogenates made at pH 9.2 than at pH
7.0, because of alkaline stripping of IF1. Pretreatment with
glibenclamide blocked the inhibitory effects of diazoxide in
homogenates prepared at physiological pH and did not alter
the limited increase in ATPase activity observed at pH 9.2.
This result suggests that diazoxide treatment altered IF1
release from ATP synthase at alkaline pH, irrespective of
the effects elicited on ⌬␺m. When the experiments were
repeated in presence of 25 ␮M FCCP (Fig. 8B), we again
observed FCCP inhibition of ATPase in pH 7.0 homogenates but full recovery of ATPase activity in pH 9.2 homogAJP-Heart Circ Physiol • VOL
enates; these results were not altered by pretreatment with
glibenclamide, indicating that FCCP uncoupling stimulated
IF1-mediated ATPase inhibition independently of the state
of mitoKATP channels. When cells were treated with both
diazoxide and FCCP, there was no additive inhibitory effect
on ATPase assessed in pH 7.0 homogenates, likely because
of limited IF1 availability, and the recovery of ATPase
activity in alkaline homogenates was only partial (as observed with diazoxide alone). Finally, when cells were
treated with both diazoxide and FCCP after preexposure to
glibenclamide, ATPase activity in pH 7.0 homogenates was
reduced to the same extent as without glibenclamide, indicating that the inhibition was due to FCCP uncoupling rather
than to diazoxide. However, when these triply treated cells
were homogenized at pH 9.2 to strip bound IF1, there was
only modest recovery of ATPase, attributable to the inhibitory effects of diazoxide. Notably, in all four diazoxidetreated samples, ATPase activity measured after alkaline
stripping of IF1 was significantly lower than that in the
292 • FEBRUARY 2007 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.32.246 on June 15, 2017
Fig. 6. Effect of diazoxide on mitochondrial transmembrane potential (⌬␺m). A: qualitative assessment of ⌬␺m. H9c2 cells grown on glass were exposed to
vehicle (1), 100 ␮M diazoxide (2), or 25 ␮M FCCP (3) for 10 min in culture medium before replacement with medium containing 10 ␮g/ml JC-1. After labeling
for 5 min at 37°C in the dark, cells were assessed by confocal microscopy. One experiment representative of 3 is shown. B: flow cytometry analysis of ⌬␺m.
Cells were treated with diazoxide (DZO, 1) and FCCP (2) and labeled as described in A before trypsinization and analysis with a FACScan cytometer. One
experiment representative of 3 is shown.
H826
DUAL EFFECT OF DIAZOXIDE ON MITOCHONDRIAL F0F1 ATP SYNTHASE
DISCUSSION
respective nondiazoxide controls, while for all four FCCPtreated samples, there were no significant differences in
such activity compared with the respective non-FCCP controls (Fig. 8). These results show that, even in presence of a
KATP channel blocker, diazoxide interfered with alkaline
release of IF1 from F0F1 ATP synthase. Furthermore, diazoxide had this effect even when IF1 binding was increased
by the depolarizing effects of FCCP.
Fig. 8. Effects of diazoxide in presence of
glibenclamide or FCCP on mitochondrial ATPase under physiological conditions of IF1
binding and under IF1 alkaline stripping conditions. H9c2 cells were preexposed to 10 ␮M
glibenclamide (GLB) or vehicle for 1 h before
treatment for 10 min with 100 ␮M diazoxide
(DZO) in the absence (A) or presence (B) of
25 ␮M FCCP. ATPase activity was measured
in homogenates prepared at pH 7.0 (open
bars) and pH 9.2 (filled bars). Data are means
(SD) from at least 4 experiments performed in
duplicate. Columns marked with the same
letters are significantly different (Tukey’s
test, P ⬍ 0.05).
AJP-Heart Circ Physiol • VOL
292 • FEBRUARY 2007 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.32.246 on June 15, 2017
Fig. 7. Preexposure of H9c2 cells to glibenclamide prevents the effects of
diazoxide treatment on ⌬␺m loss (A) and on basal ATPase activity (B). Cells
were preexposed to vehicle (open bars) or 10 ␮M glibenclamide (gray bars) in
culture medium for 1 h before addition of 100 ␮M diazoxide for 0 –30 min. A:
⌬␺m evaluated by flow cytometry as JC-1 red fluorescence intensity (590 nm).
B: ATPase activity in homogenates prepared at pH 7.0. Data are means (SD)
of 3 experiments performed in duplicate. *P ⬍ 0.001 vs. cells not preexposed
to glibenclamide (Student’s t-test).
Because an improved mitochondrial energy state is recognized to play an important role in the biochemical mechanisms
of IPC and pharmacological preconditioning (2, 24, 43, 55), we
used cultured rat H9c2 myoblasts to characterize the effects
elicited by diazoxide on mitochondrial F0F1 ATP synthase in
intact cells. In this model system, under normoxic conditions,
diazoxide inhibited the ATPase activity of F0F1 ATP synthase
in a transient manner, with a maximal effect of 26% at 100
␮M. This effect of diazoxide coincided with a transient and
mild mitochondria uncoupling and membrane depolarization
but was achieved without reducing ATP synthesis flux or
cellular ATP content. Concerning ATP synthesis flux, the
limitation of the experimental setup consisting of the fact that
oligomycin increases proton motive force by inhibiting its
dissipation by ATP synthesis as well as proton leak proportionally should be considered. Nevertheless, because diazoxide
had no effect on oligomycin sensitivity of basal respiration, we
can expect that the limitation was common to both control and
diazoxide-treated cells, thus not altering the finding that ATP
synthesis rate was not affected by diazoxide. Our finding are
consistent with those of Fryer et al. (17), who found that the
rate of ATP synthesis in isolated mitochondria of diazoxidetreated heart was not affected by the drug. Furthermore, the
unaltered ATP levels were in accordance. The mitochondrial
effects of diazoxide were blocked by pretreating cells with
glibenclamide, a KATP channel blocker. However, even in the
presence of glibenclamide or FCCP (which strongly uncouples
and depolarizes mitochondria), the effect of diazoxide was
observed as a lower ATPase activity in homogenates prepared
by sonication at pH 9.2 (i.e., by an effective way of releasing
bound IF1). These results suggest that diazoxide elicits its
effects through interference with IF1 binding and/or release,
although other pH-dependent pharmacological effects of the
drug cannot be ruled out. Moreover, these results are consistent
with a dual effect of diazoxide leading to ATPase inhibition: by
mildly uncoupling and depolarizing mitochondria, diazoxide
indirectly favors binding of IF1 to F0F1 ATP synthase and
thereby inhibits ATPase activity; in addition, diazoxide directly alters IF1 binding/release to/from F0F1 ATP synthase, as
IF1 alkaline stripping is less effective. This is in accordance
with our previous report (9), in which we described a lower Kd
value of IF1 binding to purified F1 in the presence of diazoxide,
DUAL EFFECT OF DIAZOXIDE ON MITOCHONDRIAL F0F1 ATP SYNTHASE
AJP-Heart Circ Physiol • VOL
phoric” effect of the drug (23, 27, 38, 39). Unfortunately, we
do not know whether or not K⫹ cycling contributes to the
uncoupling and mitochondrial depolarization elicited by diazoxide in our conditions. We did not perform experiments
addressing this question, considering that a number of side
effects, possibly occurring by culturing cells under conditions
of low K⫹ concentration, may challenge data reliability in
intact cells. On the other hand, with regard to the effect
afforded by diazoxide on F0F1 ATP synthase and promoting
IF1 binding to the enzyme, in our previous paper (9) we
reported experiments run with isolated membrane-bound enzyme and documented that IF1 binding to submitochondrial
particles was similarly affected by diazoxide in both the absence and the presence of 75 mM K2SO4. Thus we may suggest
that in cells the effect altering the steady state of IF1 binding/
release to/from the enzyme was also afforded by diazoxide
independently of K⫹ cycling. Conversely, the effect inducing
IF1 binding and mediated by mitochondrial depolarization
possibly depended on energy-diverting K⫹/H⫹ cycling. In the
resting state of our cells, in which O2 consumption is low and
⌬␺m is high, concomitant with the effects observed by us (i.e.,
mild loss of ⌬␺m and inhibition of ATPase activity mediated
by IF1 binding), a modest increase of mitochondrial K⫹ influx
and matrix expansion may be caused by diazoxide, as suggested by Garlid and coworkers (11, 15, 19, 20). In this
hypothesis, the induction of IF1 binding to ATP synthase
consequent to mild loss of ⌬␺m afforded by diazoxide may be
an additional molecular consequence of mitoKATP channel
opening, besides preservation of mitochondrial intermembrane
architecture and of the normal low outer membrane permeability to ADP/ATP (15). Our data concerning the protective effect
of glibenclamide may prompt us to conclude, within a pharmacological approach, that this was the case. Nevertheless,
metabolic effects of glibenclamide not related to the activity of
specific KATP channels have been reported (23) and must be
considered, together with the finding that the sensitivity of
mitochondrial Na⫹ and K⫹ channels to glibenclamide critically depends on factors such as Mg2⫹, ATP, and GTP concentrations (29, 50).
Although mitoKATP channels have been implicated in most
studies regarding IPC and pharmacological preconditioning,
there is no conclusive molecular evidence for their existence
and no definitive information regarding structural properties.
Studies in which mitoKATP channels are pharmacologically
manipulated, including the present study, must be interpreted
with caution because of the multiple metabolic effects of drugs
active at these potassium channels. As already mentioned,
besides the actions on ATPase reported here and by others (2,
6, 55) and those on succinate dehydrogenase (22, 48), diazoxide also acts as a protonophoric uncoupler (26, 27, 33) and
inhibits other nucleotide-requiring cellular ATPases (16). Furthermore, glibenclamide reduces fatty acid oxidation by inhibiting carnitine palmitoyltransferase (35) and blocks ATP-binding cassette transporters (42) and chloride channels (56);
whether these actions, other than KATP channel inhibitors, play
a role in its ability to block diazoxide-induced preconditioning
remains to be tested. Drug potency and targets may be critically dependent on the metabolic state of the cells and on the
experimental conditions chosen (29).
As reported by many authors (2, 6, 24, 55), we believe that
reducing ATP hydrolysis rate in ischemia is a fundamental
292 • FEBRUARY 2007 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.32.246 on June 15, 2017
although it did not influence the energization-dependent IF1
release. The selective action of diazoxide, inhibiting the hydrolase but not the synthase activity of F0F1 ATP synthase,
may be responsible for the beneficial effects of pharmacological preconditioning on bioenergetics and cell survival during
ischemia-reperfusion. A similar action has been reported by
Grover et al. (21) for a series of selective ATPase inhibitors.
In the cell system investigated here, the maximal effect of
diazoxide on F0F1 ATP synthase was observed at 100 ␮M, a
relatively high concentration but nonetheless within the pharmacological range. Although some effects have been observed
at lower concentrations (18, 39), other authors have also used
100 ␮M (1, 22). To our knowledge, no authors have measured
cell-associated concentrations of diazoxide, which may vary in
different cells depending on the efficiency of drug-extruding
machinery. When H9c2 cells were exposed to 100 ␮M diazoxide, the cell-associated drug concentrations were 30 – 40 ␮M
over 30 min. Therefore, in the experiments reported here, low
intracellular concentrations of diazoxide should have helped
avoid undesired effects such as succinate dehydrogenase inhibition (22, 48).
Our results regarding the ability of IF1 in inhibiting ATPase
activity in rat cells in response to stimuli (i.e., diazoxide
exposure) are in agreement with those of Das (13) and Hassinen et al. (24), who observed in hearts of small rodents
IF1-dependent inhibition of ATPase during ischemia and IPC;
conversely, Rouslin and Broge (46) reported no significant
inhibition. The low level (20%) of bound IF1 observed in H9c2
cells under normoxic, physiological conditions is in accordance with Schwerzmann and Pedersen (49) and Rouslin (45),
who both reported a 0.2 ratio of bound IF1 to catalytic sector F1
in rat. Finally, the modest mitochondrial uncoupling achieved
with diazoxide in H9c2 cells is similar to that reported by
Minners et al. (38, 39), who indicated properly in a modest
uncoupling the cellular response of preconditioning-mediated
cardioprotection. The mechanism by which uncoupling of
oxidative phosphorylation in preconditioning results in a beneficial adaptive mitochondrial response to cellular stress has
not been established yet. It may include reduction of excessive
reactive oxygen species generation in the contest of ischemia
(41), uncoupling-induced augmentation of glucose uptake (31),
and preservation of mitochondrial calcium overload (26). Inhibition of ATPase activity of F0F1 ATP synthase, due to the
dual effect of diazoxide on IF1 binding documented here, may
be central to the adaptive response by attenuating ATP depletion. It should be emphasized that diazoxide elicited decrease
of ⌬␺m and enhancement of respiration, but not decrease of
mitochondrial ATP synthesis and enhancement of ATP hydrolysis. Therefore, diazoxide cannot be considered to act as a
“conventional protonophore” under the experimental conditions we used, although we cannot exclude such an effect on a
different timescale. We can speculate that the expected increase of ATP hydrolysis may be counteracted by the effect
altering the steady state of IF1 binding/release to/from the F0F1
ATP synthase that diazoxide afforded concomitantly, resulting
in promotion of IF1 binding and inhibition of ATPase activity.
Conversely, at present we do not have a clear explanation for
the finding that the expected decrease in ATP synthesis flux
was undetectable.
Diazoxide uncoupling may be due to K⫹ influx induced by
the opening of mitoKATP channels and/or to the “protono-
H827
H828
DUAL EFFECT OF DIAZOXIDE ON MITOCHONDRIAL F0F1 ATP SYNTHASE
GRANTS
This work was supported by a grant from Ministero dell’Istruzione,
dell’Università e della Ricerca Scientifica (MIUR) (Progetto di Rilevanza
Nazionale PRIN, 2004 –2006).
REFERENCES
1. Akao M, Ohler A, O’Rourke B, Marban E. Mitochondrial ATPsensitive potassium channels inhibit apoptosis induced by oxidative stress
in cardiac cells. Circ Res 88: 1267–1275, 2001.
2. Ala-Rämi A, Ylitalo KV, Hassinen IE. Ischaemic preconditioning and a
mitochondrial KATP channel opener both produce cardioprotection accompanied by F1F0 ATPase inhibition in early ischaemia. Basic Res Cardiol
98: 250 –258, 2003.
3. Ardehali H, Chen Z, Ko Y, Mejia-Alvarez R, Marban E. Multiprotein
complex containing succinate dehydrogenase confers mitochondrial ATPsensitive K⫹ channel activity. Proc Natl Acad Sci USA 101: 11880 –
11885, 2004.
4. Ardehali H, O’Rourke B. Mitochondrial KATP channels in cell survival
and death. J Mol Cell Cardiol 39: 7–16, 2005.
5. Barrientos A. In vivo and in organello assessment of OXPHOS activities.
Methods 26: 307–316, 2002.
AJP-Heart Circ Physiol • VOL
6. Belisle E, Kowaltowski AJ. Opening of mitochondrial K⫹ channels
increases ischemic ATP levels by preventing hydrolysis. J Bioenerg
Biomembr 34: 285–298, 2002.
7. Bosetti F, Yu G, Zucchi R, Ronca-Testoni S, Solaini G. Myocardial
ischemic preconditioning and mitochondrial F0F1-ATPase activity. Mol
Cell Biochem 215: 31–37, 2000.
8. Comelli M, Lippe G, Mavelli I. Differentiation potentiates oxidant injury
to mitochondria by hydrogen peroxide in Friend’s erythroleukemia cells.
FEBS Lett 352: 71–75, 1994.
9. Contessi S, Metelli G, Mavelli I, Lippe G. Diazoxide affects the IF1
inhibitor protein binding to F1 sector of beef heart F0F1 ATP synthase.
Biochem Pharmacol 67: 1843–1851, 2004.
10. Cook JA, Mitchell JB. Viability measurements in mammalian cell system. Anal Biochem 179: 1–7, 1989.
11. Costa ADT, Quinlan CL, Andrukin A, West IC, Jaburek M, Garlid
KD. The direct physiological effects of mitoKATP opening on heart
mitochondria. Am J Physiol Heart Circ Physiol 290: H406 –H415, 2006.
12. Da Cruz S, Xenarios I, Langridge J, Vilbois F, Parone FA, Martinou
JC. Proteomic analysis of the mouse liver mitochondrial inner membrane.
J Biol Chem 278: 41566 – 41571, 2003.
13. Das AM. Regulation of the mitochondrial ATP-synthase in health and
disease. Mol Genet Metab 79: 71– 82, 2003.
14. Di Pancrazio F, Mavelli I, Isola M, Losano G, Pagliaro P, Harris DA,
Lippe G. In vitro and in vivo studies of F0F1 ATP synthase regulation by
inhibitor protein IF1 in goat heart. Biochim Biophys Acta 1659: 52– 62,
2004.
15. Dos Santos P, Kowaltowski AJ, Laclau MN, Seetharaman S, Paucek
P, Boudina S, Thambo JB, Tariosse L, Garlid KD. Mechanisms by
which opening the mitochondrial ATP-sensitive K⫹ channels protects the
ischemic heart. Am J Physiol Heart Circ Physiol 283: H284 –H295, 2002.
16. Dzeja PP, Bast P, Ozcan C, Valverde A, Holmuhamedov EL, Van
Wylen DGL, Terzic A. Targeting nucleotide-requiring enzymes: implications for diazoxide-induced cardioprotection. Am J Physiol Heart Circ
Physiol 284: H1048 –H1056, 2003.
17. Fryer RM, Eells JT, Hsu AK, Henry MM, Gross GJ. Ischemic
preconditioning in rats: role of mitochondrial KATP channel in preservation of mitochondrial function. Am J Physiol Heart Circ Physiol 278:
H305–H312, 2000.
18. Garlid KD, Paucek P, Yarov-Yarovoy V, Sun X, Schindler PA. The
mitochondrial KATP channel as a receptor for potassium channel openers.
J Biol Chem 271: 8796 – 8799, 1996.
19. Garlid KD, Paucek P, Yarov-Yarovoy V, Murray HN, Darbenzio RB,
D’Alonzo AJ. Cardioprotective effect of diazoxide and its interaction with
mitochondrial ATP-sensitive K⫹ channels: possible mechanism of cardioprotection. Circ Res 81: 1072–1082, 1997.
20. Grover GJ, Garlid KD. ATP-sensitive potassium channels: a review of
their cardioprotective pharmacology. J Mol Cell Cardiol 32: 677– 695,
2000.
21. Grover GJ, Atwal KS, Sleph PG, Wang FL, Monshizadegan H,
Monticello T, Green DW. Excessive ATP hydrolysis in ischemic myocardium by mitochondrial F1F0-ATPase: effect of selective pharmacological inhibition of mitochondrial ATPase hydrolase activity. Am J Physiol
Heart Circ Physiol 287: H1747–H1755, 2004.
22. Hanley PJ, Mickel M, Loffler M, Brandt U, Daut J. KATP channelindependent targets of diazoxide and 5-hydroxydecanoate in the heart.
J Physiol 542: 735–741, 2002.
23. Hanley PJ, Daut J. KATP channels and preconditioning: a re-examination
of the role of mitochondrial KATP channels and an overview of alternative
mechanisms. J Mol Cell Cardiol 39: 17–50, 2005.
24. Hassinen IE, Vuorinen KH, Ylitalo K, Ala-Rami A. Role of cellular
energetics in ischemia-reperfusion and ischemic preconditioning of myocardium. Mol Cell Biochem 184: 393– 400, 1998.
25. Hausenloy D, Wynne A, Duchen M, Yellon D. Transient mitochondrial
permeability transition pore opening mediates preconditioning-induced
protection. Circulation 109: 1714 –1717, 2004.
26. Holmuhamedov EL, Wang L, Terzic A. ATP sensitive K⫹ channel
openers prevent Ca2⫹ overload in rat cardiac mitochondria. J Physiol 519:
347–360, 1999.
27. Holmuhamedov EL, Jahangir A, Oberlin A, Komarov A, Colombini
M, Terzic A. Potassium channel openers are uncoupling protonophores:
implication in cardioprotection. FEBS Lett 568: 167–170, 2004.
28. Hopkins TA, Dyck JR, Lopaschuk GD. AMP-activated protein kinase
regulation of fatty acid oxidation in the ischaemic heart. Biochem Soc
Trans 31: 207–212, 2003.
292 • FEBRUARY 2007 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.32.246 on June 15, 2017
aspect of cardioprotection. In our opinion, both the mechanism
documented by Garlid and coworkers and that suggested by us
and involving a dual effect on ATP synthase may contribute to
diazoxide-mediated reduction of energy waste. We therefore
speculate that the modest dissipation of ⌬␺m caused by diazoxide in the resting state of our cells, whether mediated by
mitoKATP channels or by other mechanisms, could be sufficient
to induce IF1 binding to ATP synthase, thereby inhibiting ATP
hydrolysis catalyzed by the enzyme but not ATP synthesis.
Therefore, mild uncoupling and the dual effect on ATP synthase documented by us during normoxia may represent important effects helping in the search for the mechanisms, still
unknown, by which diazoxide confers cardioprotection during
ischemia.
In conclusion, we observed that in normoxic intact cells,
diazoxide has a dual effect on mitochondrial ATP synthase that
can be summarized as follows. Diazoxide-triggered mitochondrial uncoupling and depolarization inhibit ATPase activity by
promoting binding with the membrane potential-sensitive inhibitor IF1. Concomitantly, diazoxide interferes with the binding/release of IF1 to/from the enzyme, further enhancing the
inhibition. The transient nature of diazoxide effects we observed remains to be explained. Nevertheless, the finding that
cell-associated drug concentration significantly varies over the
same timescale and parallels such effects may suggest a threshold effect of the drug concentration, which appeared to diminish to the initial value at 30 min as a likely consequence of high
efficiency of drug-extruding machinery. Furthermore, it should
be emphasized that in fully energized mitochondria diazoxideinduced reduction of ⌬␺m likely has a relatively small and
transient effect, whereas under conditions of metabolic stress
relatively small decreases in ⌬␺m can translate into large
decreases of matrix Ca2⫹ and large changes in the functional
state of mitochondria and specifically of ATPase/synthase.
Under such conditions diazoxide effects may be thus amplified
and made longlasting. It is also possible that a deeper investigation of the actual relevance of energy-diverting K⫹/H⫹
cycling as a mechanism of the effects documented here (as
discussed above) may provide a contribution to the explanation
of their transient nature. The biological relevance of these
results versus cardioprotection needs further in vivo investigation.
DUAL EFFECT OF DIAZOXIDE ON MITOCHONDRIAL F0F1 ATP SYNTHASE
AJP-Heart Circ Physiol • VOL
43. Penna C, Pagliaro P, Rastaldo R, Di Pancrazio F, Lippe G, Gattullo D,
Mancardi D, SamajaM, Losano G, Mavelli I. F0F1 ATP synthase
activity is differently modulated by coronary reactive hyperemia before
and after ischemic preconditioning in the goat. Am J Physiol Heart Circ
Physiol 287: H2192–H2200, 2004.
44. Pogolotti AL Jr, Santi DV. High-pressure liquid chromatography—
ultraviolet analysis of intracellular nucleotides. Anal Biochem 126: 335–
345, 1982.
45. Rouslin W. The mitochondrial adenosine 5⬘-triphosphatase in slow and
fast heart rate hearts. Am J Physiol Heart Circ Physiol 252: H622–H627,
1987.
46. Rouslin W, Broge CW. IF1 function in situ in uncoupler-challenged
ischemic rabbit, rat, and pigeon hearts. J Biol Chem 271: 23638 –23641,
1996.
47. Sanada S, Kitazake M, Asanuma H. Role of mitochondrial and sarcolemmal KATP channels in ischemic preconditioning of the canine heart.
Am J Physiol Heart Circ Physiol 280: H256 –H263, 2001.
48. Schafer C, Wegener C, Portenhauser R, Bojanovski D. Diazoxide, an
inhibitor of succinate oxidation. Biochem Pharmacol 18: 2678 –2681,
1969.
49. Schwerzmann K, Pedersen P. Regulation of the mitochondrial ATP
synthase/ATPase complex. Arch Biochem Biophys 250: 1–18, 1986.
50. Szewczyk A, Pikula S, Wojcik G, Nalecz MJ. Glibenclamide inhibits
mitochondrial K⫹ and Na⫹ uniports induced by magnesium depletion. Int
J Biochem Cell Biol 47: 863– 871, 1996.
51. Tanno M, Miura T, Tsuchida A. Contribution of both the sarcolemmal
KATP and mitochondrial KATP channels to infarct size limitation by KATP
channel openers: differences from preconditioning in the role of sarcolemmal KATP channels. Naunyn Schmiedebergs Arch Pharmacol 364: 226 –
232, 2001.
52. Tominaga M, Horie M, Sasayama S, Okada Y. Glibenclamide, an
ATP-sensitive K⫹ channel blocker, inhibits cardiac cAMP-activated Cl⫺
conductance. Circ Res 77: 417– 423, 1995.
53. Vadineanu A, Berden JA, Slater EC. Proteins required for the binding
of mitochondrial ATPase to the mitochondrial inner membrane. Biochim
Biophys Acta 449: 468 – 479, 1976.
54. Vree TB, Lenselink B, Huysmans FT, Fleuren HL, Thien TA. Rapid
determination of diazoxide in plasma and urine of man by means of
high-performance liquid chromatography. J Chromatogr 164: 228 –234,
1979.
55. Ylitalo K, Ala-Rami A, Vuorinen K, Peuhkurinen K, Lepojarvi M,
Kaukoranta P. Reversible ischemic inhibition of F1F0-ATPase in rat and
human myocardium. Biochim Biophys Acta 1504: 329 –339, 2001.
56. Zhu Z, Li YL, Li DP, He RR. Effect of anoxic preconditioning on
ATP-sensitive potassium channels in guinea-pig ventricular myocytes.
Pflügers Arch 439: 808 – 813, 2000.
292 • FEBRUARY 2007 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.32.246 on June 15, 2017
29. Jaburek M, Yarov-Yarovoy V, Paucek P, Garlid KD. State-dependent
inhibition of the mitochondrial KATP channel by glyburide and 5-hydroxydecanoate. J Biol Chem 273: 13578 –13582, 1998.
30. Jiang MT, Ljubkovic M, Nakae Y, Shi Y, Kwok W, Stowe DF,
Bosnjak ZJ. Characterization of human cardiac mitochondrial ATPsensitive potassium channel and its regulation by phorbol ester in vitro.
Am J Physiol Heart Circ Physiol 290: H1770 –H1776, 2006.
31. Khayat ZA, Tsakiridis T, Ueyama A, Somwar R, Ebina Y, Klip A.
Rapid stimulation of glucose transport by mitochondrial uncoupling depends in part on cytosolic Ca2⫹ and cPKC. Am J Physiol Cell Physiol 275:
C1487–C1497, 1998.
32. Klein G, Sarastre M, Dianoux AC, Vignais PV. Radiolabeling of natural
adenosine triphosphatase inhibitor with phenyl (14C) isothiocyanate and
study of its interaction with mitochondrial adenosine triphosphatase.
Localization of inhibitor binding sites and stoichiometry of binding.
Biochemistry 19: 2919 –2925, 1980.
33. Kowaltowski AJ, Seetharaman S, Paucek P, Garlid KD. Bioenergetic
consequences of opening the ATP-sensitive K⫹ channel of heart mitochondria. Am J Physiol Heart Circ Physiol 280: H649 –H657, 2001.
34. Lacza Z, Snipes JA, Miller AW, Szabò C, Grover G, Busija D. Heart
mitochondria contain functional ATP-dependent K⫹ channels. J Mol Cell
Cardiol 35: 1339 –1347, 2003.
35. Lehtihet M, Welsh N, Berggren PO, Cook GA, Sjoholm A. Glibenclamide inhibits islet carnitine palmitoyltransferase 1 activity, leading to
PKC-dependent insulin exocytosis. Am J Physiol Endocrinol Metab 285:
E438 –E446, 2003.
36. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 193: 265–275, 1951.
37. Marin-Garcia, Goldenthal MJ. Mitochondria play a critical role in
cardioprotection. J Card Fail 10: 55– 66, 2004.
38. Minners J, van den Bos EJ, Yellon DM, Schwalb H, Opie LH, Sack
MN. Dinitrophenol, cyclosporin A, and trimetazidine modulate preconditioning in the isolated rat heart: support for a mitochondrial role in
cardioprotection. Cardiovasc Res 47: 68 –73, 2000.
39. Minners J, Lacerda L, McCarthy J, Meiring JJ, Yellon DM, Sack
MN. Ischemic and pharmacological preconditioning in Girardi cells and
C2C12 myotubes induce mitochondrial uncoupling. Circ Res 89: 787–
792, 2001.
40. O’Rourke B. Evidence for mitochondrial K⫹ channels and their role in
cardioprotection. Circ Res 94: 420 – 432, 2004.
41. Pain T, Yang XM, Critz SD, Yue Y, Nakano A, Liu GS. Opening of
mitochondrial KATP channels triggers the preconditioned state by generating free radicals. Circ Res 87: 460 – 466, 2000.
42. Payen L, Delugin L, Courtois A, Trinquart Y, Guillouzo A, Fardel O.
The sulphonylurea glibenclamide inhibits multidrug resistance protein
(MRP1) activity in human lung cancer cells. Br J Pharmacol 132:
778 –784, 2001.
H829