- Email: [email protected]
The Classically Cardioprotective Agent Diazoxide Elicits Arrhythmias in Type 2 Diabetes Mellitus
JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY
VOL. 66, NO. 10, 2015
ª 2015 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION
ISSN 0735-1097/$36.00
PUBLISHED BY ELSEVIER INC.
http://dx.doi.org/10.1016/j.jacc.2015.06.1329
The Classically Cardioprotective Agent Diazoxide Elicits Arrhythmias in Type 2 Diabetes Mellitus Chaoqin Xie, MD, Jun Hu, PHD, Lukas J. Motloch, MD, PHD, Basil S. Karam, BS, Fadi G. Akar, PHD
ABSTRACT BACKGROUND Type 2 diabetes mellitus (T2DM) is associated with an enhanced propensity for ventricular tachyarrhythmias (VTs) under conditions of metabolic demand. Activation of mitochondrial adenosine triphosphatesensitive potassium (KATP) channels by low-dose diazoxide (DZX) improves hypoglycemia-related complications, metabolic function, and triglyceride and free fatty acid levels and reverses weight gain in T2DM. OBJECTIVES In this study, we hypothesized that DZX prevents ischemia-mediated arrhythmias in T2DM via its putative cardioprotective and antidiabetic property. METHODS Zucker obese diabetic fatty (ZO) rats (n ¼ 43) with T2DM were studied. Controls consisted of Zucker lean (ZL; n ¼ 13) and normal Sprague-Dawley (SprD; n ¼ 30) rats. High-resolution optical action potential mapping was performed before and during challenge with no-flow ischemia for 12 min. RESULTS Electrophysiological properties were relatively stable in T2DM hearts at baseline. In contrast, ischemia uncovered major differences between groups, because action potential duration (APD) in T2DM failed to undergo progressive adaptation to ischemic challenge. DZX promoted the incidence of arrhythmias, because all DZX-treated T2DM hearts exhibited ischemia-induced VTs that persisted on reperfusion. In contrast, untreated T2DM and controls did not exhibit VT during ischemia. Unlike DZX, pinacidil promoted ischemia-mediated arrhythmias in both control and T2DM hearts. Rapid and spatially heterogeneous shortening of APD preceded the onset of arrhythmias in T2DM. DZX-mediated proarrhythmia in T2DM was not related to changes in the messenger ribonucleic acid expression of Kir6.1, Kir6.2, SUR1A, SUR1B, SUR2A, SUR2B, or ROMK (renal outer medullary potassium channel). CONCLUSIONS Ischemia uncovers a paradoxical resistance of T2DM hearts to APD adaptation. DZX reverses this property, resulting in rapid and heterogeneous APD shortening. This promotes reentrant VT during ischemia. DZX should be avoided in diabetic patients at risk of ischemic events. (J Am Coll Cardiol 2015;66:1144–56) © 2015 by the American College of Cardiology Foundation.
T
ype 2 diabetes mellitus (T2DM) is a mul-
Because T2DM affects multiple organ systems, its
tifactorial disease process that encompasses
pharmacological management is fraught with cardio-
hyperglycemia, dyslipidemia, and hyper-
vascular complications (3,4). One promising strategy
insulinemia. Collectively, these factors produce a
is the use of the potassium channel agonist diazoxide
malignant environment marked by oxidative stress
(DZX), which improves the metabolic function of
(1). Although patients with T2DM are more vulnerable
Otsuka Long-Evans Tokushima fatty rats by restoring
to ischemic injury than nondiabetic people (2), the
their triglyceride and free fatty acid levels and
detailed
electrophysiological
substrate
of
T2DM
reversing their weight gain and diabetes status (5).
hearts and their response to ischemia are poorly
The antidiabetic efficacy of DZX is linked to preser-
defined.
vation of pancreatic beta cells through promotion of
From the Cardiac Bioelectricity Research Laboratory, Cardiovascular Institute, Icahn School of Medicine at Mount Sinai, New York, New York. The authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Xie and Hu contributed equally to this work. William Stevenson, MD, served as the Guest Editor for this paper. Listen to this manuscript’s audio summary by JACC Editor-in-Chief Dr. Valentin Fuster. Manuscript received June 26, 2014; revised manuscript received June 6, 2015, accepted June 23, 2015.
JACC VOL. 66, NO. 10, 2015
Xie et al.
SEPTEMBER 8, 2015:1144–56
Diazoxide-Mediated Arrhythmias in Diabetes
1145
beta cell rest and inhibition of apoptosis (5). These
12- to 16-week-old male Zucker obese diabetic
ABBREVIATIONS
mechanistic studies in animal models are being
fatty (fa/fa) rats (ZO) (n ¼ 43) as a standard
AND ACRONYMS
rapidly translated to humans. In a proof-of-principle
model of established insulin resistance and
trial,
DZX
T2DM. This model faithfully recapitulates
(4 mg/kg) exhibited a 30% decrease in glucose pro-
hallmark features of the clinical disease,
duction, with no difference in the rate of glucose
including
uptake compared with placebo-treated counterparts
and
healthy
people
who
received
oral
hyperinsulinemia,
hyperglycemia
(21).
APD = action potential duration
APD75 = action potential
dyslipidemia,
Age-
and
duration at 75% of
sex-
repolarization
(6). Moreover, DZX reversed weight gain in pre-
matched Zucker lean (ZL) (n ¼ 13) and
CV = conduction velocity
diabetic obese, hyperinsulinemic people (7,8). The
normal Sprague-Dawley (SprD) (n ¼ 30) rats
DZX = diazoxide
therapeutic utility of DZX was also tested in patients
were also used. In a subset of experiments
KATP channel = adenosine
with type 1 diabetes mellitus (T1DM) (9) and in those
involving 6 ZL and 7 SprD rats, we found no
triphosphate-sensitive
undergoing coronary bypass surgery (10). Radtke
significant differences in electrophysiological
et al. (9) reported that treatment of newly diagnosed
parameters, including action potential dura-
T1DM patients with DZX for 6 months improved their
tion at 50%, 75% (APD75), and 90% of repo-
glycemic control without affecting their beta cell
larization
function. Finally, in a double-blinded randomized
Marked hyperglycemia and weight gain in the
ribonucleic acid
study, supplementation of cardioplegia with low-
absence of hypertrophy were confirmed in
PCL = pacing cycle length
dose DZX protected cardiac mitochondria, meta-
T2DM compared with control rats (Table 1).
bolism, and function in patients undergoing cardiac
OPTICAL ACTION POTENTIAL MAPPING OF
reaction
EX VIVO PERFUSED HEARTS. Hearts were
ROMK = renal outer medullary
bypass surgery (10).
and
conduction
velocity
potassium channel
mKATP = mitochondrial adenosine triphosphatesensitive potassium
(CV).
mRNA = messenger
PCR = polymerase chain
potassium channel
excised rapidly and Langendorff perfused
SEE PAGE 1157
with
These highly encouraging clinical findings likely stem from the activation of cardioprotective signaling
Tyrode’s
solution
containing
SprD = Sprague-Dawley
121.7
T1DM = type 1 diabetes
mmol/l NaCl, 25.0 mmol/l NaHCO 3, 2.74
mellitus
mmol/l MgSO 4 , 4.81 mmol/l KCl, 5.0 mmol/l
T2DM = type 2 diabetes
dextrose, and 2.5 mmol/l CaCl 2 (pH 7.40; 95%
mellitus
via mitochondrial adenosine triphosphate-sensitive
O2 /5% CO 2). Atria were removed to avoid
potassium (mKATP ) channels by DZX (11–19). Howev-
competitive stimulation of the ventricles.
tachyarrhythmia
Perfusion pressure was maintained at 60 to
ZL = Zucker lean (rat)
er, at high concentrations (300 m mol/l), DZX was shown to promote arrhythmias in coronary-perfused
70 mm Hg by regulating coronary perfusion
atria and ventricles from failing and nonfailing hu-
flow.
man hearts by activating sarcolemmal (as opposed to mitochondrial) KATP channels (20).
VT = ventricular
ZO = Zucker obese diabetic fatty (rat)
Preparations were placed in a custom-built tissue bath and pressed gently against a glass imaging
We hypothesized that owing to its dual nature as
window by a stabilizing piston (22–26). Movement
an antidiabetic and cardioprotective agent, low-dose
was suppressed by perfusion with blebbistatin 10
(30 m mol/l) DZX may be particularly useful in suppressing
ischemia-related
arrhythmias
in
T2DM
mmol/l for 10 to 15 min. To avoid surface cooling, preparations were immersed in the coronary effluent
without altering basal electrophysiology. Contrary
maintained at a physiological temperature by a heat
to our original hypothesis, we uncovered a potent
exchanger assembly (22). Cardiac rhythm was moni-
proarrhythmic effect of DZX exclusively in T2DM
tored continuously via silver electrodes connected to
hearts during challenge with ischemia. Our findings
an electrocardiogram amplifier. Cardiac rhythm,
raise major concerns regarding the safety profile of
perfusion
mK ATP channel activation in T2DM patients at risk of
continuously during each experiment.
pressure,
and
flow
were
monitored
ischemic injury.
METHODS
T A B L E 1 Average BW, HW, Ratio of BW to HW, and
Blood Glucose Levels of T2DM and Control Animals
EXPERIMENTAL RAT MODEL OF T2DM. All pro-
T2DM
Control
410 65
283 29
cedures involving animal handling were approved by
BW (g)
the Animal Care and Use Committee of the Icahn
HW (g)
1.6 0.1
1.2 0.1
School of Medicine at Mount Sinai and adhered to the
HW/BW, mg/g
3.9 0.3
4.3 0.2
Guide for the Care and Use of Laboratory Animals
Glucose, mg/dl
408 119
128 14
published by the U.S. National Institutes of Health.
Age, weeks
13–16
13–14
All animals were purchased from the Charles River Laboratory (Wilmington, Massachusetts). We used
BW ¼ body weight; HW ¼ heart weight; T2DM ¼ type 2 diabetes mellitus.
1146
Xie et al.
JACC VOL. 66, NO. 10, 2015
Diazoxide-Mediated Arrhythmias in Diabetes
SEPTEMBER 8, 2015:1144–56
Optical action potentials were measured with an
amplitude were quantified. Action potential duration
80 80-pixel charge-coupled device camera coupled
(APD) was defined as the temporal difference be-
to an imaging macroscope that contained a high
tween the repolarization and activation times at each
numerical aperture lens, a dichroic mirror, excita-
site.
tion and emission filters, a beam splitter, and a
REAL-TIME
light-collimating tube. Excitation light was provided
MEASUREMENTS. Tissue specimens from SprD (n ¼ 3),
POLYMERASE
CHAIN
REACTION
by a low-noise, high-power tungsten-halogen lamp
ZL (n ¼ 3), and ZO (n ¼ 3) hearts were used to measure
and directed in an epifluorescence configuration onto
messenger
the heart through a side port of the macroscope in a
levels of the alpha (Kir6.1, Kir6.2) and beta (SUR1A,
manner that minimized heterogeneity of incident
SUR1B, SUR2A, SUR2B) subunits that constitute
light across the mapped 8 8 mm 2 region. Emitted light was collected by the front lens of the macroscope, filtered, and directed onto the charge-coupled
ribonucleic
acid
(mRNA)
expression
the K ATP channel, as well as the 2 cardiac isoforms of the
renal
outer
medullary
potassium
channel
(ROMK1 and ROMK2). Total ribonucleic acid was
device detector. High-resolution optical action po-
extracted by acid guanidinium thiocyanate-phenol-
tentials were measured simultaneously from 6,400
chloroform using TRIzol reagent (Life Technologies,
sites during each recording. To improve signal qual-
Carlsbad, California). One microgram of total ribo-
ity, spatial binning was performed, which yielded an
nucleic acid was reverse transcribed into first-strand
array of 400 (20 20) high-fidelity optical action
complementary deoxyribonucleic acid by use of a
potentials that were amenable to accurate automated
high-capacity complementary deoxyribonucleic acid
analyses.
reverse transcription kit with ribonuclease inhibitor. from control
Real-time polymerase chain reactions (PCR) were
(SprD, ZL) and T2DM (ZO) rats underwent steady-
performed with the Applied Biosystems 7500 system
state pacing at a wide range of pacing cycle lengths
and Power SYBR Green PCR Master Mix (Applied
(PCLs) (300 to 100 ms, in 20-ms decrements) during
Biosystems, Waltham Massachusetts) in triplicate.
baseline perfusion. Subsequently, hearts underwent
High-resolution melting curves were generated to
ischemia-reperfusion injury while being paced at a
confirm the specificity of PCR products. Threshold
EXPERIMENTAL
PROTOCOL. Hearts
PCL of 300 ms. We chose a 12-min global no-flow
cycle (CT ) values were determined after rectification
ischemia challenge because we found in a pilot
by the passive reference dye Rox within the master
study that this protocol gave rise to sustained post-
mix. mRNA expression levels were evaluated by the
ischemic arrhythmias within 1 min of reperfusion in
2 - DCT method. Levels of K ATP channel subunits were
w50% of hearts. This rate of post-ischemic arrhyth-
normalized to those of the internal control, beta-
mias allows one to examine whether a given drug
actin.
decreases or increases the incidence of arrhythmias.
STATISTICAL ANALYSES. Summary data are pre-
Untreated hearts were perfused with normal Tyrode’s
sented as mean SD. Welch’s unpaired t test was
solution throughout the entire protocol, except for
used to detect differences in blood glucose levels
the 12-min ischemia phase. DZX-treated hearts were
between T2DM and controls. The Student t test was
initially perfused with normal Tyrode’s solution, fol-
used to compare electrophysiological parameters be-
lowed by Tyrode’s containing DZX 30 m mol/l.
tween T2DM and control hearts. The paired t test was
ELECTROPHYSIOLOGICAL MEASUREMENTS. C o n d u c t i o n
used to compare groups before and after drug treat-
v e l o c i t y . As detailed previously (23,24), conduction delays were assessed by recording action potentials during steady-state pacing. Local activation time at each site was defined as the maximum first derivative
ment. One-way analysis of variance was used to compare differences between more than 2 groups. A Fisher exact test was used to compare differences in susceptibility to arrhythmias. Differences were
during the upstroke of the action potential. Velocity
considered significant at p < 0.05.
vectors (magnitude and direction) were derived from
RESULTS
the activation times of each pixel relative to those of its neighbors. CV was measured by averaging the
BASAL ELECTROPHYSIOLOGICAL PROPERTIES OF
magnitude of the velocity vectors along the longitu-
T2DM AND CONTROL ANIMALS. Figure 1 shows
dinal and transverse axes. The anisotropic ratio was
representative optical action potentials recorded
defined as the ratio of fast (longitudinal) divided by
from control (SprD, ZL) and T2DM (ZO) rats during
slow (transverse) CV.
steady-state pacing at a PCL of 300 ms. On average,
A c t i o n p o t e n t i a l d u r a t i o n . Repolarization times at
ZO rat hearts exhibited a trend toward prolonged
75% and 90% relative to the action potential
(by 18.5%; p ¼ 0.052) APD 75 compared with controls
JACC VOL. 66, NO. 10, 2015
Xie et al.
SEPTEMBER 8, 2015:1144–56
Diazoxide-Mediated Arrhythmias in Diabetes
F I G U R E 1 Baseline Electrophysiological Properties of T2DM and Control Hearts
Control
A SprD
T2DM
ZL
Average APD75 100
(PCL=300ms)
200ms
C
APD Heterogeneity (PCL=300ms)
40
Control
20
50ms
p = 0.052
0
n=6
n=6
Control
T2DM
20 10 0
n=6
n=6
Control
T2DM
E SprD 2
SprD 4
Base
Control
SprD 3
CV (cm/sec)
D SprD 1
ZO 2
ZO 3
ZO 4
T2DM
ZO 1
p = 0.863
6
30
60 40
APD75 Std Dev
8
p = 0.632
(ms)
T2DM
(ms)
80
APD75 Range
(ms)
B
ZO
4 2 0
150
n=6
n=6
Control
T2DM
Longitudinal
100
50 n=6 0
n=4
15 ms
RV
Anisotropic Ratio
Apex 0
8mm
n=6
n=4
Control T2DM Control T2DM 4
LV
Transverse
Depolarization Time (ms)
Ratio
3 2 1 0
n=6
n=4
Control
T2DM
(A) Representative optical action potentials recorded during steady-state pacing of hearts from control rats (Sprague-Dawley [SprD] and Zucker lean [ZL]) and rats with type 2 diabetes mellitus (T2DM) (Zucker obese diabetic fatty [ZO]). (B) Average action potential duration measured at 75% of repolarization (APD75) at pacing cycle length (PCL) of 300 ms. Analysis is based on 6 ZL and 6 ZO rats. (C) Range and standard deviation (Std Dev) of APD75 values measured across the mapping field as independent indices of APD heterogeneity. (D) Representative isochrone maps illustrating the elliptical spread of the impulse across the epicardial surface. (E) Average longitudinal and transverse conduction velocities and the anisotropic ratio of control and T2DM hearts. CV ¼ conduction velocity; LV ¼ left ventricle; RV ¼ right ventricle.
(Figure 1B). No significant differences (p ¼ 0.65) in
or their ratio (p ¼ 0.13) between T2DM and control
APD 75 were found between ZL (46.1 7.0 ms) and
hearts (Figure 1E). Finally, programmed electrical
SprD (47.9 3.0 ms) hearts. APD heterogeneity
stimulation with single premature stimuli (S2) was
indexed by the range and standard deviation of APD 75
performed in a subset of 4 ZO hearts. S2 coupling
values over an 8 8 mm 2 region of each heart
intervals were initially varied in 10-ms decrements,
revealed no significant differences between T2DM
followed by 2-ms increments to resolve local refrac-
and controls for either metric (Figure 1C).
toriness. No arrhythmias were observed when this
Figure 1D shows representative isochrone maps
protocol was used (not shown).
depicting the elliptical spread of the action potential
ELECTROPHYSIOLOGICAL RESPONSE OF T2DM HEARTS
wave front across the epicardium in control and
TO ISCHEMIA. Patients with T2DM are highly suscep-
T2DM hearts. There were no significant differences in
tible to ischemic injury. We therefore tested whether
longitudinal CV (p ¼ 0.205), transverse CV (p ¼ 0.156),
ischemia-mediated electrophysiological remodeling
1147
Xie et al.
JACC VOL. 66, NO. 10, 2015
Diazoxide-Mediated Arrhythmias in Diabetes
SEPTEMBER 8, 2015:1144–56
F I G U R E 2 Differential APD Response to Ischemia in Control Versus T2DM Hearts
APD Response to Ischemia
SprD
T2DM ZL
Control
B
SprD
T2DM ZO
ZL
ZO
75 65
Baseline Ischemia (12 min)
Base
55
35 25
200ms
Control
C
T2DM
LV
D 120
7 min
90
APD90 (ms)
6 min
9 min
8mm
RV
Ischemia (12 min)
Baseline
8 min
45
Apex
Ischemia (12 min)
APD75 (ms)
Control
Baseline
A
Ischemia
1148
50ms
T2DM (n=4)
60 30
Control (n=9)
10 min 0 11 min
0
6
8
12
10
Time of Ischemia (min)
12 min 200ms
(A) Representative action potential traces before and 12 min after ischemia in control hearts (purple) and hearts of rats with T2DM (red). (B) Representative APD75 contour maps measured before and after 12 min of ischemia from control and T2DM hearts. (C) Representative action potential traces recorded at baseline and during no-flow ischemia. (D) Average action potential duration measured at 90% of repolarization (APD90) in control and T2DM hearts showing sustained shortening in the former group (purple) and a biphasic response in the latter (red). Superimposed action potential traces from representative T2DM (red) and control (purple) hearts at 12 min of ischemia. Controls consisted of SprD (n ¼ 3) and ZL (n ¼ 6) animals; T2DM was ZO animals (n ¼ 4). APD ¼ action potential duration; other abbreviations as in Figure 1.
was more severe in hearts from T2DM rats compared
We next investigated potential differences in
with control animals. Figures 2A and 2B show repre-
ischemia-mediated
sentative action potential traces and APD contour
groups. Figure 3A shows isopotential contour maps
conduction
slowing
between
maps measured from control (SprD, ZL) and T2DM
recorded at 1-ms intervals before (baseline) and 12 min
(ZO) hearts before and 12 min after no-flow ischemia.
after ischemia in control (SprD, ZL), and T2DM (ZO)
As expected, no-flow ischemia resulted in significant
hearts. Figure 3B shows representative action poten-
APD shortening in hearts from control animals
tial upstrokes (first derivatives) measured from equi-
(Figures 2A to 2C). In sharp contrast, hearts from
distant sites. Figure 3C presents average CV during
T2DM animals did not undergo progressive APD
baseline perfusion and after ischemic challenge. As
adaptation in response to an identical ischemic
expected, ischemia resulted in significant CV slowing
challenge (Figures 2A to 2C). Instead, they exhibited a
in all groups. Interestingly, a trend toward a greater
biphasic response, because early APD shortening was
percentage reduction of CV by ischemia was noted for
fully reversed during the latter phase (9 to 12 min) of
T2DM rats than for controls (Figure 3D); however, this
ischemia (Figure 2D).
trend did not reach statistical significance (p ¼ 0.243).
JACC VOL. 66, NO. 10, 2015
Xie et al.
SEPTEMBER 8, 2015:1144–56
Diazoxide-Mediated Arrhythmias in Diabetes
F I G U R E 3 Conduction Slowing in Response to Ischemia in Control and T2DM Hearts
Control
A
T2DM
SprD
ZO
ZL
baseline
Ischemia
baseline
Ischemia
baseline
0
0
Ischemia 0
48ms
48ms
48ms
B
SprD
C
P = 0.0019
80
ZO
20ms
D % Change In CV
P = 0.0004 P = 0.0002
60
0
40
20
20
n=3
n=3
n=6
n=6
n=4
n=4
Is (1 che 2 m m ia in ) Ba se lin e Is (1 che 2 m m ia in )
e m mia in ) Ba se lin e
(1
2
ch Is
Ba
se
lin
e
0
%
CV (cm/sec)
ZL
SprD
ZL
ZO
n=3
n=6
n=4
40 60
P=0.243
80
(A) Representative series of isopotential contour maps recorded at 1-ms intervals indicate the sequential spread of depolarization before and after the ischemic challenge in SprD, ZL, and ZO hearts. Each frame is 8 8 mm2. (B) Representative action potential first derivatives recorded from equidistant sites across the mapping field. (C) Average CV measurements at baseline and after 12 min of ischemia in control and T2DM hearts. (D) Percentage change in CV after 12 min of ischemia relative to baseline in control and T2DM (ZO) hearts. Controls were SprD (n ¼ 3) and ZL (n ¼ 6) animals; T2DM was ZO animals (n ¼ 4). Abbreviations as in Figure 1.
Finally, after the 12-min ischemia protocol, 4 of
which discounts a role for nonischemic activation of
9 control and 3 of 5 T2DM hearts underwent sus-
sarcolemmal KATP channels.
tained arrhythmias within 1 min of reperfusion
We next investigated potential differences in the
(not shown).
response of untreated and DZX-treated T2DM hearts
DZX PARADOXICALLY PROMOTES ARRHYTHMIAS IN
to ischemia. Although APD 75 of untreated T2DM
T2DM. The main purpose of our study was to deter-
hearts failed to shorten progressively in response to
mine whether low-dose DZX, which improves meta-
ischemia (Figure 4B), DZX treatment caused rapid and
bolic
in
sustained APD shortening during the course of
humans, would prevent arrhythmias in T2DM. We
ischemia (Figure 4B). On average, APD 75 of DZX-
therefore examined its effects on key electrophysio-
treated T2DM hearts was shorter by 32.6% (p < 0.01)
logical properties before and during challenge of
after 6 min of ischemia compared with their un-
T2DM hearts with ischemia.
treated counterparts (Figure 4B).
function
in
experimental
models
and
Figure 4A shows the average APD75 rate relation-
The heightened sensitivity of DZX-treated T2DM
ships in untreated and DZX-treated T2DM hearts
hearts to ischemia-induced APD shortening was likely
during baseline perfusion. Clearly, DZX 30 m mol/l did
significant, because all 7 DZX-treated but none of the
not alter baseline APD values at any PCL (Figure 4A),
5 untreated T2DM hearts exhibited onset of sustained
1149
1150
Xie et al.
JACC VOL. 66, NO. 10, 2015
Diazoxide-Mediated Arrhythmias in Diabetes
SEPTEMBER 8, 2015:1144–56
F I G U R E 4 Proarrhythmic Effect of DZX in T2DM
A
C
80
Incidence of VT during ischemia
T2DM (untreated, n=4)
Control
APD75 (ms)
VT (+)
T2DM (DZX, n=6)
60 40 20 0
160
200
240
280
T2DM
120
PCL (ms)
80
DZX
Untreated DZX
*
* * *
T2DM (untreated)
60
*
40 T2DM (DZX, n=6)
20
APD75 (ms)
*
APD before onset of VT or end of ischemia
80
T2DM (untreated, n=4)
60 APD75 (ms)
Untreated
D
B
VT (-)
40
T2DM (DZX)
20
0
0 0
6
8
10
12
0
Ischemia (min)
8
9 10 Ischemia (min)
11
12
(A) Treatment of hearts with 30 mmol/l diazoxide (DZX) did not alter baseline APD75 in T2DM hearts. (B) Average APD75 at PCL of 300 ms in untreated (red) and DZX-treated (blue) T2DM hearts during ischemia. (C) Summary of ventricular tachyarrhythmia (VT) propensity during ischemia in untreated and DZX-treated control and T2DM hearts. Only DZX-treated T2DM hearts exhibited VT during 12 min of no-flow ischemia. (D) APD75 at end of ischemia or before onset of VT in untreated (red) and DZX-treated (blue) T2DM hearts, respectively. Each asterisk indicates that the red dot is significantly different (greater) than the blue dot. APD ¼ action potential duration; other abbreviations as in Figure 1.
ventricular tachyarrhythmia (VT) during the latter
hearts. Whereas on average, DZX-treated T2DM hearts
phase (10 to 12 min) of ischemia (Figure 4C). More-
exhibited a minor trend toward longer APD75 values
over, none of the 9 untreated or 7 DZX-treated control
during ischemia than DZX-treated controls, differ-
hearts exhibited VT during ischemia (Figure 4C).
ences were not statistically significant (Figure 5A). In
As such, DZX provoked VT exclusively in T2DM hearts
contrast, APD heterogeneity, as indexed by the stan-
during ischemia (Figure 4C). Figure 4D shows indi-
dard deviation and range of APD75 values across the
vidual measurements of APD 75, either before the
mapping field, was more than 2-fold greater in DZX-
onset of VT in DZX-treated T2DM hearts or at the end
treated T2DM than in DZX-treated control hearts at
of the 12-min ischemic challenge in untreated T2DM
8 and 10 min of ischemia (Figure 5A). Figure 5B shows
hearts. In all cases, APD 75 of untreated T2DM hearts
representative APD contour maps that illustrate the
was longer after 12 min of ischemia than that of DZX-
greater APD heterogeneity at 8 min of ischemia in
treated T2DM hearts, which underwent a 26% to 56%
DZX-treated T2DM than in control hearts. Figure 5C
reduction before the onset of VT.
shows representative histograms that illustrate the
We investigated the potential mechanism underly-
wider distribution of APD 75 values in individual DZX-
ing the selective proarrhythmic effect of DZX on T2DM
treated T2DM hearts than in their control counterparts.
JACC VOL. 66, NO. 10, 2015
Xie et al.
SEPTEMBER 8, 2015:1144–56
Diazoxide-Mediated Arrhythmias in Diabetes
1151
F I G U R E 5 APD Response to DZX Treatment in Control and T2DM Hearts
Average APD
APD Heterogeneity
A
DZX-treated DZX-treated
P = 0.0039
Ischemia (10 min)
6
P = 0.0008
5
10
n=5
n=6
n=5
n=6
1
n=5
0
Co
T2
nt
ro
l
DM
l ro Co
T2
nt
DM
l ro nt Co
T2
n=5
n=6
n=5
0
DM
l ro nt Co
DM T2
Co
nt
ro
l
0
n=6
DM
n=6
2
T2
n=5
3
l
20
20
4
ro
40
n=6
P = 0.0049
nt
P = 0.087
APD Std. Dev. (ms)
APD Range (ms)
30 P = 0.106
P = 0.045
Co
60 APD75 (ms)
Ischemia (8 min)
40
DM
80
Ischemia (10 min)
DZX-treated
Ischemia (10 min)
T2
Ischemia (8 min)
Ischemia (8 min)
DZX-treated
C
Ischemia (8 min) 40
25
30 35 40 APD75 (ms)
45
Count 45
0 20
50
0 20
10
25
30 35 40 APD75 (ms)
45
50
30 35 40 APD75 (ms)
45
50
APD75
30 Count
20
25
40
APD75
Count
Count
Count
T2DM
30 35 40 APD75 (ms)
30
10
2 mm
25
40
30
20
10
0 20
50
APD75
30
20
20
10
40
20
30
Count
Count
Count
20
0 20
40
APD75
30
10
50 ms
T2DM
40
APD75
30
Control
Control
40
APD75
Count
DZX-treated Ischemia (8 min)
Count
B
20
10
0 20
25
30 35 40 APD75 (ms)
45
50
0 20
25
30 35 40 APD75 (ms)
45
50
(A) Average APD75, APD75 range, and APD75 standard deviation at 8 min and 10 min of ischemia in DZX-treated control and T2DM hearts. (B) Representative APD75 contour maps from control and T2DM hearts at 8 min of ischemia. (C) Representative histograms showing the distribution of APD75 values across the mapping field in individual DZX-treated control and DZX-treated T2DM hearts at 8 min of ischemia. Controls were 3 ZL and 3 SprD rats; T2DM was 5 ZO rats. APD ¼ action potential duration; other abbreviations as in Figures 1 and 3.
We next compared the electrophysiological ef-
control and T2DM hearts (Figure 6B), whereas DZX
fects of DZX with those of the classic SUR2A activator
only
pinacidil. Unlike DZX, which did not affect base-
(Figure 5). Profound action potential shortening was
exerted
a
proarrhythmic
effect
in
T2DM
line APD 75 in either group, pinacidil 100 m mol/l
evident by 5 min of ischemia in pinacidil-treated
caused significant APD75 shortening in T2DM hearts
control and T2DM hearts (Figure 5C).
(p ¼ 0.047) and a strong trend toward shorter levels in
EXPRESSION
controls (p ¼ 0.069) during baseline perfusion
determine whether the proarrhythmic effect of DZX
(Figure 6A). Pinacidil also promoted the rapid (within
in T2DM hearts was related to altered expression of
6 to 7 min) onset of VT during ischemia in both
KATP channels, we measured the mRNA expression
OF
KATP
CHANNEL
SUBUNITS. To
Xie et al.
JACC VOL. 66, NO. 10, 2015
Diazoxide-Mediated Arrhythmias in Diabetes
SEPTEMBER 8, 2015:1144–56
F I G U R E 6 Effects of Pinacidil on Control and T2DM Hearts
A
DZX Control
T2DM
60 APD75 (ms)
APD75 (ms)
60 40 20
n=6
n=6
n=4
n=4
Baseline
DZX
Baseline
DZX
20
Control
VT (+) Untreated Pinacidil
Untreated
T2DM
P = 0.069
P = 0.047
n=4
n=4
n=4
n=4
Baseline
PIN
Baseline
PIN
0
C
Incidence of VT during ischemia
Control
40
0
B
Pinacidil
80
80
T2DM
1152
Control
T2DM
VT (-) Baseline
Pinacidil
Ischemia 5 min
Pinacidil 200ms (A) Average APD75 values for control (n ¼ 4) and T2DM (n ¼ 4) hearts before and after treatment with DZX (30 mmol/l) or pinacidil (100 mmol/l). (B) Summary of VT propensity during ischemia in untreated and pinacidil-treated control and T2DM hearts. Arrhythmogenesis during early ischemia was provoked by pinacidil treatment in both control and T2DM hearts. (C) Representative action potential traces revealing profound action potential shortening at 5 min of ischemia in pinacidil-treated control and T2DM hearts. PIN ¼ pinacidil; other abbreviations as in Figures 1 and 3.
levels of all known KATP channel subunits using real-
that are expressed in the heart. We found no sig-
time PCR. Figure 7A shows the normalized (to beta-
nificant differences in the expression levels of
actin) mRNA expression of the alpha (Kir6.1 and
either isoform between groups (Figure 7B). Primers
Kir6.2) and beta (SUR1A, SUR1B, SUR2A, and SUR2B)
used
subunits of the KATP channel in SprD, ZL, and ZO
Figure 7C.
for
mRNA
measurements
are
shown
in
hearts. We found that samples from ZL and ZO animals exhibited a trend toward increased expres-
DISCUSSION
sion of most K ATP channel subunits compared with their SprD counterparts (Figure 7A); however,
We
this apparent up-regulation only reached statistical
quences of ischemia in T2DM and tested whether low-
significance in the case of SUR2B between ZO and
dose DZX, a potent activator of cardioprotective
SprD animals (Figure 7A). Importantly, there were no
signaling (11–19), could improve electrical function in
significant differences in the expression of any of
that setting. Our major findings were as follows:
these subunits between ZL and ZO hearts (Figure 7A).
1) during normoxic perfusion, T2DM hearts are stable,
As such, none of the observed changes could explain
exhibiting only minor electrophysiological changes;
the proarrhythmic effect that was exclusive to ZO
2) in response to ischemia, T2DM hearts are charac-
animals.
terized by repolarization resistance, because their
A recent study highlighted the potential impor-
investigated
the
electrophysiological
conse-
average APD fails to exhibit progressive shortening;
tance of ROMK (KCNJ1) in formation of the mK ATP
3)
channel (27). We therefore measured the mRNA
clusively in T2DM hearts by promoting rapid and
DZX
treatment
levels of the 2 ROMK isoforms (ROMK1 and ROMK2)
heterogeneous
APD
exacerbates shortening
arrhythmias during
ex-
ischemia
JACC VOL. 66, NO. 10, 2015
Xie et al.
SEPTEMBER 8, 2015:1144–56
Diazoxide-Mediated Arrhythmias in Diabetes
F I G U R E 7 K ATP Channel Subunit Expression in Control and T2DM Hearts
B SprD (n=3)
4
ZL (n=3) ZO (n=3)
3
4
P < 0.05 mRNA Expression Level
mRNA Expression Level
A
2 1
3 2 1 0
0 Kir6.1
C
Kir6.2
Sur1A
Sur1B
Sur2A
ROMK1
Sur2B
ROMK2
Gene name
Forward primer (5’-3’)
Reverse primer (5’-3’)
beta-Actin Kir6.1 Kir6.2 SUR1A SUR1B SUR2A SUR2B ROMK1 ROMK2
CCTCTATGCCAACACAGTGCTGTCT AGCTGGCTGCTCTTCGCTATCA CAACGTCGCCCACAAGAACATC AGCATGCGTGGTACTCATC TGCCTCTCTCTTCCTCACA GGAGTGCGATACTGGTCCAAACCT CATAGCTCATCGGGTTCACACCATT AGTGAGCCAGCCAATCAGTGACA CCAAGGCACTGGGCACCTTTAGC
GCTCAGGAGGAGCAATGATCTTGA CCCTCCAAACCCAATGGTCACT CCAGCTGCACAGGAAGGACATG GCTGAGAGCTCCCTGTGTAG GTTGGATCTCCATGTCTGC CCCGATGCAGAGAACGAGACACT GCATCGAGACACAGGTGCTGTTGT TGGACCCAGACTCAGACAAAGGC TGCCGGGAACGCCCAAATATGTG
(A) Messenger RNA (mRNA) expression of Kir6.1, Kir6.2, SUR1A, SUR1B, SUR2A, and SUR2B measured by real-time polymerase chain reaction (RT-PCR) in SprD (n ¼ 3), ZL (n ¼ 3), and ZO (n ¼ 3) animals. (B) mRNA expression of ROMK1 and ROMK2 measured by RT-PCR in SprD (n ¼ 3), ZL (n ¼ 3), and ZO (n ¼ 3) animals. (C) Primers used for RT-PCR measurements. Beta-actin was used as a normalizing control. KATP ¼ adenosine triphosphate-sensitive potassium channel; other abbreviations as in Figure 1.
(Central Illustration); and 4) changes in the mRNA
speculated that acute ischemia might unmask the
expression of KATP channel subunits do not cause
electrophysiological vulnerability of these hearts. A
these phenotypic differences.
major finding of the present report was the failure of
STABILITY
ELECTROPHYSIOLOGICAL
T2DM hearts to exhibit progressive APD adaptation
PROPERTIES IN T2DM. Our initial objective was to
(shortening) in response to 12 min of no-flow
define the electrophysiological substrate of hearts
ischemia. The mechanism underlying this repolari-
from obese T2DM animals. We focused on conduction
zation resistance will require additional investiga-
slowing and APD prolongation as potentially impor-
tion,
tant features of pathological electrical remodeling.
sarcolemmal K ATP channels, resulting from either
Surprisingly, we found only modest changes in APD
reduced expression or function of these channels.
and CV in this established model of obesity and
Interestingly, Ren et al. (29) found a marked reduc-
T2DM. As such, our present findings discount a major
tion in SUR2A mRNA expression in a model of T1DM.
role for these basal electrophysiological properties in
We measured the transcript levels of all known alpha
OF
BASAL
although
it
suggests
down-regulation
of
the proarrhythmic vulnerability of the diabetic heart.
and beta subunits of K ATP channels, including Kir6.1,
Our present findings are reminiscent of our previous
Kir6.2, SUR1A, SUR1B, SUR2A, SUR2B, ROMK1, and
findings in a guinea pig model of streptozotocin-
ROMK2. An important advantage of our experimental
induced T1DM, in which arrhythmias were only
design was the use of 2 different control groups that
encountered on depletion of the scavenging capacity
were functionally identical but genetically distinct.
of the heart (28).
Specifically, ZL controls share a common genetic
ALTERED APD ADAPTATION DURING ISCHEMIA:
background with their ZO counterparts but are func-
REPOLARIZATION RESISTANCE IN T2DM? Because
tionally identical (in terms of electrophysiology, heart
patients with T2DM are highly susceptible to the
weight, body weight, and blood glucose levels) to
complications
age-matched SprD animals. By measuring changes in
of
coronary
artery
disease,
we
1153
1154
Xie et al.
JACC VOL. 66, NO. 10, 2015
Diazoxide-Mediated Arrhythmias in Diabetes
SEPTEMBER 8, 2015:1144–56
C EN T RA L IL LUSTR AT I ON
DZX-Mediated Arrhythmias in T2DM: Proarrhythmic Effect of DZX in Rats
Xie, C. et al. J Am Coll Cardiol. 2015; 66(10):1144–56.
The proarrhythmic effect of mitochondrial adenosine triphosphate-sensitive potassium (mKATP) channel activation by low-dose diazoxide (DZX) in hearts from rats with type 2 diabetes mellitus (T2DM) is illustrated. Electrophysiological properties were stable at baseline but revealed impaired adaptation to increased metabolic demand during acute ischemia in diabetic (red) compared with nondiabetic (blue) rats. The classically cardioprotective agent DZX caused paradoxical exacerbation of arrhythmias in diabetic but not normal animals, owing to a rapid and spatially heterogeneous shortening of action potentials across the heart, which in turn, created a substrate for reentrant ventricular tachyarrhythmia. APD ¼ action potential duration; APD75 ¼ action potential duration at 75% of repolarization.
the mRNA expression of K ATP channel subunits in all 3
underlying impaired APD adaptation of T2DM hearts
groups, we were able to determine whether a given
to ischemia warrants investigation.
change was functionally important. Surprisingly, we
DZX FOR TREATMENT OF DIABETES MELLITUS. In a
only found a significant change in the expression of
recent clinical study, DZX suppressed endogenous
SUR2B between ZO and SprD animals. Therefore, our
glucose production (6). The presumed antidiabetic
data indicate that the unique functional response of
efficacy of DZX has been linked to preservation of
T2DM hearts to ischemia is unrelated to the transcript
pancreatic beta cells through promotion of beta cell
levels of K ATP channel subunits. This suggests
rest and inhibition of apoptosis. Indeed, pilot clinical
involvement of post-transcriptional mechanisms that
trials using DZX have demonstrated improvement of
regulate protein synthesis, trafficking, insertion, sta-
metabolic function in pre-diabetic obese, hyper-
bility, or gating of these channels. Of note, AMP-
insulinemic patients and in those with T1DM (7–9).
activated protein kinase (AMPK) signaling has been
Numerous experimental studies have established
implicated in preconditioning-induced shortening of
DZX as a potent cardioprotective agent that reduces
the action potential through effects on K ATP channel
reperfusion injury by promoting activation of bene-
activity and recruitment (30). Moreover, drugs that
ficial mitochondrial pathways (11–19). Indeed, a pilot
activate AMPK signaling are a mainstay of treatment
study demonstrated that supplementation of car-
of T2DM patients (31). Impaired AMPK in T2DM may
dioplegia with DZX improved the outcome of patients
contribute
Clearly,
undergoing coronary bypass surgery (10); however,
identification of the exact molecular mechanisms
the infarct-sparing efficacy of DZX in diabetes is less
to
repolarization
resistance.
JACC VOL. 66, NO. 10, 2015
Xie et al.
SEPTEMBER 8, 2015:1144–56
Diazoxide-Mediated Arrhythmias in Diabetes
clear, because DZX failed to reduce infarct size after
(Central Illustration). Specifically, a potent proar-
30 min of left anterior descending coronary artery
rhythmic effect during ischemic challenge should
occlusion in T2DM animals (32). This is consistent
caution against the rapid translation of DZX-based
with the notion that cardioprotective stimuli (such as
therapies for the treatment of T2DM patients, who
ischemic preconditioning), which likely depend on
are indeed at high risk of coronary artery disease,
mK ATP channels, are defective in diabetes (33).
acute myocardial infarction, and silent myocardial
Although the infarct-sparing effects of DZX have
ischemia (4,34–36).
been widely studied, there are no reports of its role in the modulation of electrophysiological properties of
REPRINT REQUESTS AND CORRESPONDENCE: Dr.
T2DM hearts. A major finding of the present report
Fadi G. Akar, Cardiovascular Research Center, Icahn
was that DZX elicited a potent proarrhythmic effect
School of Medicine at Mount Sinai, One Gustave L.
exclusively in the setting of diabetes, because all
Levy Place, Box 1030, New York, New York 10029.
DZX-treated T2DM hearts but none of the untreated
E-mail: [email protected].
T2DM or control hearts exhibited sustained ischemiamediated VT. We used a widely accepted DZX concentration (30 mmol/l) that is relatively specific for the
PERSPECTIVES
mitochondrial pool of KATP channels (15–19). At this concentration, DZX did not alter basal electrophysi-
COMPETENCY IN MEDICAL KNOWLEDGE: The mito-
ological properties. The lack of effect of low-dose DZX
chondrial KATP channel agonist DZX exerts salutary myocardial
on baseline APD strongly discounts a role for nonse-
protective effects against ischemic injury, produces vasodila-
lective activation of sarcolemmal K ATP channels in the
tion in hypertensive patients, and reverses hypoglycemic
proarrhythmic effect that we uncovered. Unlike DZX,
complications in patients with T2DM. However, in an animal
the classic SUR2A-specific K ATP channel activator,
model of T2DM with impaired electrophysiological adaptation
pinacidil, provoked ischemia-mediated VT in all
to acute ischemia, DZX elicited proarrhythmic effects because
hearts (T2DM and control), consistent with previous
of the rapid, heterogeneous shortening of the action potential
findings in failing and nonfailing human heart prep-
unrelated to transcript levels of ATP-sensitive potassium
arations (20). Finally, the lack of arrhythmia protec-
channels.
tion by DZX in T2DM hearts on ischemic challenge is consistent
with
the
lack
of
protection
against
TRANSLATIONAL OUTLOOK: Agents considered safe and
myocardial infarction by ischemic preconditioning
effective in structurally normal hearts might be associated with
and DZX (33).
proarrhythmic risks in the setting of diabetes mellitus. Strategies to mitigate the proarrhythmic effects of mitochondrial KATP
CONCLUSIONS
channel agonists should be investigated in patients with T2DM at
Our current study raises important safety concerns
risk of myocardial ischemia.
regarding the use of DZX in diabetic patients
REFERENCES 1. Candido R, Srivastava P, Cooper ME, Burrell LM. Diabetes mellitus: a cardiovascular disease. Curr Opin Investig Drugs 2003;4:1088–94.
rats at prediabetic stage. J Diabetes Complications 2008;22:46–55.
and autoimmunity but improves glycemic control. Diabetes Care 2010;33:589–94.
2. Ansley DM, Wang B. Oxidative stress and myocardial injury in the diabetic heart. J Pathol
6. Kishore P, Boucai L, Zhang K, et al. Activation of KATP channels suppresses glucose production in humans. J Clin Invest 2011;121:
10. Deja MA, Malinowski M, Golba KS, et al. Diazoxide protects myocardial mitochondria, metabolism, and function during cardiac sur-
2013;229:232–41.
4916–20.
3. Buse JB, Ginsberg HN, Bakris GL, et al. Primary
7. Schreuder T, Karreman M, Rennings A, Ruinemans-
prevention of cardiovascular diseases in people with diabetes mellitus: a scientific statement from the American Heart Association and the American Diabetes Association. Circulation 2007;115:114–26.
Koerts J, Jansen M, de Boer H. Diazoxide-mediated insulin suppression in obese men: a dose-response study. Diabetes Obes Metab 2005;7:239–45.
gery: a double-blind randomized feasibility study of diazoxide-supplemented cardioplegia. J Thorac Cardiovasc Surg 2009;137:997–1004, 1004e1–2.
4. Grundy SM, Benjamin IJ, Burke GL, et al. Diabetes and cardiovascular disease: a statement for healthcare professionals from the American Heart Association [published correction appears in Circulation 2000;101:1629–1631]. Circulation 1999; 100:1134–46. 5. Guo Z, Bu S, Yu Y, et al. Diazoxide prevents abdominal adiposity and fatty liver in obese OLETF
8. van Boekel G, Loves S, van Sorge A, RuinemansKoerts J, Rijnders T, de Boer H. Weight loss in obese men by caloric restriction and high-dose diazoxide-mediated insulin suppression. Diabetes Obes Metab 2008;10:1195–203. 9. Radtke MA, Nermoen I, Kollind M, et al. Six months of diazoxide treatment at bedtime in newly diagnosed subjects with type 1 diabetes does not influence parameters of b-cell function
11. Seharaseyon J, Ohler A, Sasaki N, et al. Molecular composition of mitochondrial ATP-sensitive potassium channels probed by viral Kir gene transfer. J Mol Cell Cardiol 2000;32:1923–30. 12. Sato T, Marban E. The role of mitochondrial KATP channels in cardioprotection. Basic Res Cardiol 2000;95:285–9. 13. Sato T, Sasaki N, Seharaseyon J, O’Rourke B, Marbán E. Selective pharmacological agents implicate mitochondrial but not sarcolemmal KATP
1155
1156
Xie et al.
JACC VOL. 66, NO. 10, 2015
Diazoxide-Mediated Arrhythmias in Diabetes
SEPTEMBER 8, 2015:1144–56
channels in ischemic cardioprotection. Circulation 2000;101:2418–23. 14. Hu H, Sato T, Seharaseyon J, et al. Pharmacological and histochemical distinctions between molecularly defined sarcolemmal KATP channels and native cardiac mitochondrial KATP channels. Mol Pharmacol 1999;55:1000–5. 15. Carroll R, Gant VA, Yellon DM. Mitochondrial KATP channel opening protects a human atrialderived cell line by a mechanism involving free radical generation. Cardiovasc Res 2001;51: 691–700. 16. Costa AD, Jakob R, Costa CL, Andrukhiv K, West IC, Garlid KD. The mechanism by which the mitochondrial ATP-sensitive Kþ channel opening and H2O2 inhibit the mitochondrial permeability transition. J Biol Chem 2006;281:20801–8. 17. Facundo HT, de Paula JG, Kowaltowski AJ. Mitochondrial ATP-sensitive Kþ channels prevent oxidative stress, permeability transition and cell death. J Bioenerg Biomembr 2005;37:75–82. 18. Garlid KD, Paucek P, Yarov-Yarovoy V, et al. Cardioprotective effect of diazoxide and its interaction with mitochondrial ATP-sensitive Kþ channels: possible mechanism of cardioprotection. Circ Res 1997;81:1072–82. 19. Quinlan CL, Costa AD, Costa CL, Pierre SV, Dos
21. Sárközy M, Zvara A, Gyémánt N, et al. Metabolic syndrome influences cardiac gene expression pattern at the transcript level in male ZDF rats. Cardiovasc Diabetol 2013;12:16. 22. Akar FG, Aon MA, Tomaselli GF, O’Rourke B. The mitochondrial origin of postischemic arrhythmias. J Clin Invest 2005;115:3527–35. 23. Akar FG, Nass RD, Hahn S, et al. Dynamic changes in conduction velocity and gap junction properties during development of pacing-induced heart failure. Am J Physiol 2007;293:H1223–30. 24. Akar FG, Spragg DD, Tunin RS, Kass DA, Tomaselli GF. Mechanisms underlying conduction slowing and arrhythmogenesis in nonischemic dilated cardiomyopathy. Circ Res 2004;95:717–25. 25. Jin H, Chemaly ER, Lee A, et al. Mechanoelectrical remodeling and arrhythmias during progression of hypertrophy. FASEB J 2010;24: 451–63. 26. Xie C, Kauffman J, Akar FG. Functional crosstalk between the mitochondrial PTP and KATP channels determine arrhythmic vulnerability to oxidative stress. Front Physiol 2014;5:264. 27. Foster DB, Ho AS, Rucker J, et al. Mitochondrial ROMK channel is a molecular component of mitoKATP. Circ Res 2012;111:446–54.
Santos P, Garlid KD. Conditioning the heart induces formation of signalosomes that interact with mitochondria to open mitoKATP channels. Am J Physiol Heart Circ Physiol 2008;295:H953–61.
28. Xie C, Biary N, Tocchetti CG, et al. Glutathione oxidation unmasks proarrhythmic vulnerability of
20. Fedorov VV, Glukhov AV, Ambrosi CM, et al. Effects of KATP channel openers diazoxide and pinacidil in coronary-perfused atria and ventricles from failing and non-failing human hearts. J Mol
29. Ren Y, Xu X, Wang X. Altered mRNA expression of ATP-sensitive and inward rectifier potassium channel subunits in streptozotocin-induced diabetic rat heart and aorta. J Pharmacol Sci 2003;
Cell Cardiol 2011;51:215–25.
93:478–83.
chronically hyperglycemic guinea pigs. Am J Physiol Heart Circ Physiol 2013;304:H916–26.
30. Sukhodub A, Jovanovic S, Du Q, et al. AMPactivated protein kinase mediates preconditioning in cardiomyocytes by regulating activity and trafficking of sarcolemmal ATP-sensitive Kþ channels. J Cell Physiol 2007;210:224–36. 31. Hardie DG. Role of AMP-activated protein kinase in the metabolic syndrome and in heart disease. FEBS Lett 2008;582:81–9. 32. Katakam PV, Jordan JE, Snipes JA, Tulbert CD, Miller AW, Busija DW. Myocardial preconditioning against ischemia-reperfusion injury is abolished in Zucker obese rats with insulin resistance. Am J Physiol Regul Integr Comp Physiol 2007;292: R920–6. 33. Kristiansen SB, Løfgren B, Støttrup NB, et al. Ischaemic preconditioning does not protect the heart in obese and lean animal models of type 2 diabetes. Diabetologia 2004;47:1716–21. 34. Jacoby RM, Nesto RW. Acute myocardial infarction in the diabetic patient: pathophysiology, clinical course and prognosis. J Am Coll Cardiol 1992;20:736–44. 35. Margolis JR, Kannel WS, Feinleib M, Dawber TR, McNamara PM. Clinical features of unrecognized myocardial infarction–silent and symptomatic: eighteen year follow-up: the Framingham study. Am J Cardiol 1973;32:1–7. 36. Scognamiglio R, Negut C, Ramondo A, Tiengo A, Avogaro A. Detection of coronary artery disease in asymptomatic patients with type 2 diabetes mellitus. J Am Coll Cardiol 2006;47:65–71.
KEY WORDS action potentials, ischemia, mitochondria, obesity, repolarization, ventricular tachycardia
VOL. 66, NO. 10, 2015
ª 2015 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION
ISSN 0735-1097/$36.00
PUBLISHED BY ELSEVIER INC.
http://dx.doi.org/10.1016/j.jacc.2015.06.1329
The Classically Cardioprotective Agent Diazoxide Elicits Arrhythmias in Type 2 Diabetes Mellitus Chaoqin Xie, MD, Jun Hu, PHD, Lukas J. Motloch, MD, PHD, Basil S. Karam, BS, Fadi G. Akar, PHD
ABSTRACT BACKGROUND Type 2 diabetes mellitus (T2DM) is associated with an enhanced propensity for ventricular tachyarrhythmias (VTs) under conditions of metabolic demand. Activation of mitochondrial adenosine triphosphatesensitive potassium (KATP) channels by low-dose diazoxide (DZX) improves hypoglycemia-related complications, metabolic function, and triglyceride and free fatty acid levels and reverses weight gain in T2DM. OBJECTIVES In this study, we hypothesized that DZX prevents ischemia-mediated arrhythmias in T2DM via its putative cardioprotective and antidiabetic property. METHODS Zucker obese diabetic fatty (ZO) rats (n ¼ 43) with T2DM were studied. Controls consisted of Zucker lean (ZL; n ¼ 13) and normal Sprague-Dawley (SprD; n ¼ 30) rats. High-resolution optical action potential mapping was performed before and during challenge with no-flow ischemia for 12 min. RESULTS Electrophysiological properties were relatively stable in T2DM hearts at baseline. In contrast, ischemia uncovered major differences between groups, because action potential duration (APD) in T2DM failed to undergo progressive adaptation to ischemic challenge. DZX promoted the incidence of arrhythmias, because all DZX-treated T2DM hearts exhibited ischemia-induced VTs that persisted on reperfusion. In contrast, untreated T2DM and controls did not exhibit VT during ischemia. Unlike DZX, pinacidil promoted ischemia-mediated arrhythmias in both control and T2DM hearts. Rapid and spatially heterogeneous shortening of APD preceded the onset of arrhythmias in T2DM. DZX-mediated proarrhythmia in T2DM was not related to changes in the messenger ribonucleic acid expression of Kir6.1, Kir6.2, SUR1A, SUR1B, SUR2A, SUR2B, or ROMK (renal outer medullary potassium channel). CONCLUSIONS Ischemia uncovers a paradoxical resistance of T2DM hearts to APD adaptation. DZX reverses this property, resulting in rapid and heterogeneous APD shortening. This promotes reentrant VT during ischemia. DZX should be avoided in diabetic patients at risk of ischemic events. (J Am Coll Cardiol 2015;66:1144–56) © 2015 by the American College of Cardiology Foundation.
T
ype 2 diabetes mellitus (T2DM) is a mul-
Because T2DM affects multiple organ systems, its
tifactorial disease process that encompasses
pharmacological management is fraught with cardio-
hyperglycemia, dyslipidemia, and hyper-
vascular complications (3,4). One promising strategy
insulinemia. Collectively, these factors produce a
is the use of the potassium channel agonist diazoxide
malignant environment marked by oxidative stress
(DZX), which improves the metabolic function of
(1). Although patients with T2DM are more vulnerable
Otsuka Long-Evans Tokushima fatty rats by restoring
to ischemic injury than nondiabetic people (2), the
their triglyceride and free fatty acid levels and
detailed
electrophysiological
substrate
of
T2DM
reversing their weight gain and diabetes status (5).
hearts and their response to ischemia are poorly
The antidiabetic efficacy of DZX is linked to preser-
defined.
vation of pancreatic beta cells through promotion of
From the Cardiac Bioelectricity Research Laboratory, Cardiovascular Institute, Icahn School of Medicine at Mount Sinai, New York, New York. The authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Xie and Hu contributed equally to this work. William Stevenson, MD, served as the Guest Editor for this paper. Listen to this manuscript’s audio summary by JACC Editor-in-Chief Dr. Valentin Fuster. Manuscript received June 26, 2014; revised manuscript received June 6, 2015, accepted June 23, 2015.
JACC VOL. 66, NO. 10, 2015
Xie et al.
SEPTEMBER 8, 2015:1144–56
Diazoxide-Mediated Arrhythmias in Diabetes
1145
beta cell rest and inhibition of apoptosis (5). These
12- to 16-week-old male Zucker obese diabetic
ABBREVIATIONS
mechanistic studies in animal models are being
fatty (fa/fa) rats (ZO) (n ¼ 43) as a standard
AND ACRONYMS
rapidly translated to humans. In a proof-of-principle
model of established insulin resistance and
trial,
DZX
T2DM. This model faithfully recapitulates
(4 mg/kg) exhibited a 30% decrease in glucose pro-
hallmark features of the clinical disease,
duction, with no difference in the rate of glucose
including
uptake compared with placebo-treated counterparts
and
healthy
people
who
received
oral
hyperinsulinemia,
hyperglycemia
(21).
APD = action potential duration
APD75 = action potential
dyslipidemia,
Age-
and
duration at 75% of
sex-
repolarization
(6). Moreover, DZX reversed weight gain in pre-
matched Zucker lean (ZL) (n ¼ 13) and
CV = conduction velocity
diabetic obese, hyperinsulinemic people (7,8). The
normal Sprague-Dawley (SprD) (n ¼ 30) rats
DZX = diazoxide
therapeutic utility of DZX was also tested in patients
were also used. In a subset of experiments
KATP channel = adenosine
with type 1 diabetes mellitus (T1DM) (9) and in those
involving 6 ZL and 7 SprD rats, we found no
triphosphate-sensitive
undergoing coronary bypass surgery (10). Radtke
significant differences in electrophysiological
et al. (9) reported that treatment of newly diagnosed
parameters, including action potential dura-
T1DM patients with DZX for 6 months improved their
tion at 50%, 75% (APD75), and 90% of repo-
glycemic control without affecting their beta cell
larization
function. Finally, in a double-blinded randomized
Marked hyperglycemia and weight gain in the
ribonucleic acid
study, supplementation of cardioplegia with low-
absence of hypertrophy were confirmed in
PCL = pacing cycle length
dose DZX protected cardiac mitochondria, meta-
T2DM compared with control rats (Table 1).
bolism, and function in patients undergoing cardiac
OPTICAL ACTION POTENTIAL MAPPING OF
reaction
EX VIVO PERFUSED HEARTS. Hearts were
ROMK = renal outer medullary
bypass surgery (10).
and
conduction
velocity
potassium channel
mKATP = mitochondrial adenosine triphosphatesensitive potassium
(CV).
mRNA = messenger
PCR = polymerase chain
potassium channel
excised rapidly and Langendorff perfused
SEE PAGE 1157
with
These highly encouraging clinical findings likely stem from the activation of cardioprotective signaling
Tyrode’s
solution
containing
SprD = Sprague-Dawley
121.7
T1DM = type 1 diabetes
mmol/l NaCl, 25.0 mmol/l NaHCO 3, 2.74
mellitus
mmol/l MgSO 4 , 4.81 mmol/l KCl, 5.0 mmol/l
T2DM = type 2 diabetes
dextrose, and 2.5 mmol/l CaCl 2 (pH 7.40; 95%
mellitus
via mitochondrial adenosine triphosphate-sensitive
O2 /5% CO 2). Atria were removed to avoid
potassium (mKATP ) channels by DZX (11–19). Howev-
competitive stimulation of the ventricles.
tachyarrhythmia
Perfusion pressure was maintained at 60 to
ZL = Zucker lean (rat)
er, at high concentrations (300 m mol/l), DZX was shown to promote arrhythmias in coronary-perfused
70 mm Hg by regulating coronary perfusion
atria and ventricles from failing and nonfailing hu-
flow.
man hearts by activating sarcolemmal (as opposed to mitochondrial) KATP channels (20).
VT = ventricular
ZO = Zucker obese diabetic fatty (rat)
Preparations were placed in a custom-built tissue bath and pressed gently against a glass imaging
We hypothesized that owing to its dual nature as
window by a stabilizing piston (22–26). Movement
an antidiabetic and cardioprotective agent, low-dose
was suppressed by perfusion with blebbistatin 10
(30 m mol/l) DZX may be particularly useful in suppressing
ischemia-related
arrhythmias
in
T2DM
mmol/l for 10 to 15 min. To avoid surface cooling, preparations were immersed in the coronary effluent
without altering basal electrophysiology. Contrary
maintained at a physiological temperature by a heat
to our original hypothesis, we uncovered a potent
exchanger assembly (22). Cardiac rhythm was moni-
proarrhythmic effect of DZX exclusively in T2DM
tored continuously via silver electrodes connected to
hearts during challenge with ischemia. Our findings
an electrocardiogram amplifier. Cardiac rhythm,
raise major concerns regarding the safety profile of
perfusion
mK ATP channel activation in T2DM patients at risk of
continuously during each experiment.
pressure,
and
flow
were
monitored
ischemic injury.
METHODS
T A B L E 1 Average BW, HW, Ratio of BW to HW, and
Blood Glucose Levels of T2DM and Control Animals
EXPERIMENTAL RAT MODEL OF T2DM. All pro-
T2DM
Control
410 65
283 29
cedures involving animal handling were approved by
BW (g)
the Animal Care and Use Committee of the Icahn
HW (g)
1.6 0.1
1.2 0.1
School of Medicine at Mount Sinai and adhered to the
HW/BW, mg/g
3.9 0.3
4.3 0.2
Guide for the Care and Use of Laboratory Animals
Glucose, mg/dl
408 119
128 14
published by the U.S. National Institutes of Health.
Age, weeks
13–16
13–14
All animals were purchased from the Charles River Laboratory (Wilmington, Massachusetts). We used
BW ¼ body weight; HW ¼ heart weight; T2DM ¼ type 2 diabetes mellitus.
1146
Xie et al.
JACC VOL. 66, NO. 10, 2015
Diazoxide-Mediated Arrhythmias in Diabetes
SEPTEMBER 8, 2015:1144–56
Optical action potentials were measured with an
amplitude were quantified. Action potential duration
80 80-pixel charge-coupled device camera coupled
(APD) was defined as the temporal difference be-
to an imaging macroscope that contained a high
tween the repolarization and activation times at each
numerical aperture lens, a dichroic mirror, excita-
site.
tion and emission filters, a beam splitter, and a
REAL-TIME
light-collimating tube. Excitation light was provided
MEASUREMENTS. Tissue specimens from SprD (n ¼ 3),
POLYMERASE
CHAIN
REACTION
by a low-noise, high-power tungsten-halogen lamp
ZL (n ¼ 3), and ZO (n ¼ 3) hearts were used to measure
and directed in an epifluorescence configuration onto
messenger
the heart through a side port of the macroscope in a
levels of the alpha (Kir6.1, Kir6.2) and beta (SUR1A,
manner that minimized heterogeneity of incident
SUR1B, SUR2A, SUR2B) subunits that constitute
light across the mapped 8 8 mm 2 region. Emitted light was collected by the front lens of the macroscope, filtered, and directed onto the charge-coupled
ribonucleic
acid
(mRNA)
expression
the K ATP channel, as well as the 2 cardiac isoforms of the
renal
outer
medullary
potassium
channel
(ROMK1 and ROMK2). Total ribonucleic acid was
device detector. High-resolution optical action po-
extracted by acid guanidinium thiocyanate-phenol-
tentials were measured simultaneously from 6,400
chloroform using TRIzol reagent (Life Technologies,
sites during each recording. To improve signal qual-
Carlsbad, California). One microgram of total ribo-
ity, spatial binning was performed, which yielded an
nucleic acid was reverse transcribed into first-strand
array of 400 (20 20) high-fidelity optical action
complementary deoxyribonucleic acid by use of a
potentials that were amenable to accurate automated
high-capacity complementary deoxyribonucleic acid
analyses.
reverse transcription kit with ribonuclease inhibitor. from control
Real-time polymerase chain reactions (PCR) were
(SprD, ZL) and T2DM (ZO) rats underwent steady-
performed with the Applied Biosystems 7500 system
state pacing at a wide range of pacing cycle lengths
and Power SYBR Green PCR Master Mix (Applied
(PCLs) (300 to 100 ms, in 20-ms decrements) during
Biosystems, Waltham Massachusetts) in triplicate.
baseline perfusion. Subsequently, hearts underwent
High-resolution melting curves were generated to
ischemia-reperfusion injury while being paced at a
confirm the specificity of PCR products. Threshold
EXPERIMENTAL
PROTOCOL. Hearts
PCL of 300 ms. We chose a 12-min global no-flow
cycle (CT ) values were determined after rectification
ischemia challenge because we found in a pilot
by the passive reference dye Rox within the master
study that this protocol gave rise to sustained post-
mix. mRNA expression levels were evaluated by the
ischemic arrhythmias within 1 min of reperfusion in
2 - DCT method. Levels of K ATP channel subunits were
w50% of hearts. This rate of post-ischemic arrhyth-
normalized to those of the internal control, beta-
mias allows one to examine whether a given drug
actin.
decreases or increases the incidence of arrhythmias.
STATISTICAL ANALYSES. Summary data are pre-
Untreated hearts were perfused with normal Tyrode’s
sented as mean SD. Welch’s unpaired t test was
solution throughout the entire protocol, except for
used to detect differences in blood glucose levels
the 12-min ischemia phase. DZX-treated hearts were
between T2DM and controls. The Student t test was
initially perfused with normal Tyrode’s solution, fol-
used to compare electrophysiological parameters be-
lowed by Tyrode’s containing DZX 30 m mol/l.
tween T2DM and control hearts. The paired t test was
ELECTROPHYSIOLOGICAL MEASUREMENTS. C o n d u c t i o n
used to compare groups before and after drug treat-
v e l o c i t y . As detailed previously (23,24), conduction delays were assessed by recording action potentials during steady-state pacing. Local activation time at each site was defined as the maximum first derivative
ment. One-way analysis of variance was used to compare differences between more than 2 groups. A Fisher exact test was used to compare differences in susceptibility to arrhythmias. Differences were
during the upstroke of the action potential. Velocity
considered significant at p < 0.05.
vectors (magnitude and direction) were derived from
RESULTS
the activation times of each pixel relative to those of its neighbors. CV was measured by averaging the
BASAL ELECTROPHYSIOLOGICAL PROPERTIES OF
magnitude of the velocity vectors along the longitu-
T2DM AND CONTROL ANIMALS. Figure 1 shows
dinal and transverse axes. The anisotropic ratio was
representative optical action potentials recorded
defined as the ratio of fast (longitudinal) divided by
from control (SprD, ZL) and T2DM (ZO) rats during
slow (transverse) CV.
steady-state pacing at a PCL of 300 ms. On average,
A c t i o n p o t e n t i a l d u r a t i o n . Repolarization times at
ZO rat hearts exhibited a trend toward prolonged
75% and 90% relative to the action potential
(by 18.5%; p ¼ 0.052) APD 75 compared with controls
JACC VOL. 66, NO. 10, 2015
Xie et al.
SEPTEMBER 8, 2015:1144–56
Diazoxide-Mediated Arrhythmias in Diabetes
F I G U R E 1 Baseline Electrophysiological Properties of T2DM and Control Hearts
Control
A SprD
T2DM
ZL
Average APD75 100
(PCL=300ms)
200ms
C
APD Heterogeneity (PCL=300ms)
40
Control
20
50ms
p = 0.052
0
n=6
n=6
Control
T2DM
20 10 0
n=6
n=6
Control
T2DM
E SprD 2
SprD 4
Base
Control
SprD 3
CV (cm/sec)
D SprD 1
ZO 2
ZO 3
ZO 4
T2DM
ZO 1
p = 0.863
6
30
60 40
APD75 Std Dev
8
p = 0.632
(ms)
T2DM
(ms)
80
APD75 Range
(ms)
B
ZO
4 2 0
150
n=6
n=6
Control
T2DM
Longitudinal
100
50 n=6 0
n=4
15 ms
RV
Anisotropic Ratio
Apex 0
8mm
n=6
n=4
Control T2DM Control T2DM 4
LV
Transverse
Depolarization Time (ms)
Ratio
3 2 1 0
n=6
n=4
Control
T2DM
(A) Representative optical action potentials recorded during steady-state pacing of hearts from control rats (Sprague-Dawley [SprD] and Zucker lean [ZL]) and rats with type 2 diabetes mellitus (T2DM) (Zucker obese diabetic fatty [ZO]). (B) Average action potential duration measured at 75% of repolarization (APD75) at pacing cycle length (PCL) of 300 ms. Analysis is based on 6 ZL and 6 ZO rats. (C) Range and standard deviation (Std Dev) of APD75 values measured across the mapping field as independent indices of APD heterogeneity. (D) Representative isochrone maps illustrating the elliptical spread of the impulse across the epicardial surface. (E) Average longitudinal and transverse conduction velocities and the anisotropic ratio of control and T2DM hearts. CV ¼ conduction velocity; LV ¼ left ventricle; RV ¼ right ventricle.
(Figure 1B). No significant differences (p ¼ 0.65) in
or their ratio (p ¼ 0.13) between T2DM and control
APD 75 were found between ZL (46.1 7.0 ms) and
hearts (Figure 1E). Finally, programmed electrical
SprD (47.9 3.0 ms) hearts. APD heterogeneity
stimulation with single premature stimuli (S2) was
indexed by the range and standard deviation of APD 75
performed in a subset of 4 ZO hearts. S2 coupling
values over an 8 8 mm 2 region of each heart
intervals were initially varied in 10-ms decrements,
revealed no significant differences between T2DM
followed by 2-ms increments to resolve local refrac-
and controls for either metric (Figure 1C).
toriness. No arrhythmias were observed when this
Figure 1D shows representative isochrone maps
protocol was used (not shown).
depicting the elliptical spread of the action potential
ELECTROPHYSIOLOGICAL RESPONSE OF T2DM HEARTS
wave front across the epicardium in control and
TO ISCHEMIA. Patients with T2DM are highly suscep-
T2DM hearts. There were no significant differences in
tible to ischemic injury. We therefore tested whether
longitudinal CV (p ¼ 0.205), transverse CV (p ¼ 0.156),
ischemia-mediated electrophysiological remodeling
1147
Xie et al.
JACC VOL. 66, NO. 10, 2015
Diazoxide-Mediated Arrhythmias in Diabetes
SEPTEMBER 8, 2015:1144–56
F I G U R E 2 Differential APD Response to Ischemia in Control Versus T2DM Hearts
APD Response to Ischemia
SprD
T2DM ZL
Control
B
SprD
T2DM ZO
ZL
ZO
75 65
Baseline Ischemia (12 min)
Base
55
35 25
200ms
Control
C
T2DM
LV
D 120
7 min
90
APD90 (ms)
6 min
9 min
8mm
RV
Ischemia (12 min)
Baseline
8 min
45
Apex
Ischemia (12 min)
APD75 (ms)
Control
Baseline
A
Ischemia
1148
50ms
T2DM (n=4)
60 30
Control (n=9)
10 min 0 11 min
0
6
8
12
10
Time of Ischemia (min)
12 min 200ms
(A) Representative action potential traces before and 12 min after ischemia in control hearts (purple) and hearts of rats with T2DM (red). (B) Representative APD75 contour maps measured before and after 12 min of ischemia from control and T2DM hearts. (C) Representative action potential traces recorded at baseline and during no-flow ischemia. (D) Average action potential duration measured at 90% of repolarization (APD90) in control and T2DM hearts showing sustained shortening in the former group (purple) and a biphasic response in the latter (red). Superimposed action potential traces from representative T2DM (red) and control (purple) hearts at 12 min of ischemia. Controls consisted of SprD (n ¼ 3) and ZL (n ¼ 6) animals; T2DM was ZO animals (n ¼ 4). APD ¼ action potential duration; other abbreviations as in Figure 1.
was more severe in hearts from T2DM rats compared
We next investigated potential differences in
with control animals. Figures 2A and 2B show repre-
ischemia-mediated
sentative action potential traces and APD contour
groups. Figure 3A shows isopotential contour maps
conduction
slowing
between
maps measured from control (SprD, ZL) and T2DM
recorded at 1-ms intervals before (baseline) and 12 min
(ZO) hearts before and 12 min after no-flow ischemia.
after ischemia in control (SprD, ZL), and T2DM (ZO)
As expected, no-flow ischemia resulted in significant
hearts. Figure 3B shows representative action poten-
APD shortening in hearts from control animals
tial upstrokes (first derivatives) measured from equi-
(Figures 2A to 2C). In sharp contrast, hearts from
distant sites. Figure 3C presents average CV during
T2DM animals did not undergo progressive APD
baseline perfusion and after ischemic challenge. As
adaptation in response to an identical ischemic
expected, ischemia resulted in significant CV slowing
challenge (Figures 2A to 2C). Instead, they exhibited a
in all groups. Interestingly, a trend toward a greater
biphasic response, because early APD shortening was
percentage reduction of CV by ischemia was noted for
fully reversed during the latter phase (9 to 12 min) of
T2DM rats than for controls (Figure 3D); however, this
ischemia (Figure 2D).
trend did not reach statistical significance (p ¼ 0.243).
JACC VOL. 66, NO. 10, 2015
Xie et al.
SEPTEMBER 8, 2015:1144–56
Diazoxide-Mediated Arrhythmias in Diabetes
F I G U R E 3 Conduction Slowing in Response to Ischemia in Control and T2DM Hearts
Control
A
T2DM
SprD
ZO
ZL
baseline
Ischemia
baseline
Ischemia
baseline
0
0
Ischemia 0
48ms
48ms
48ms
B
SprD
C
P = 0.0019
80
ZO
20ms
D % Change In CV
P = 0.0004 P = 0.0002
60
0
40
20
20
n=3
n=3
n=6
n=6
n=4
n=4
Is (1 che 2 m m ia in ) Ba se lin e Is (1 che 2 m m ia in )
e m mia in ) Ba se lin e
(1
2
ch Is
Ba
se
lin
e
0
%
CV (cm/sec)
ZL
SprD
ZL
ZO
n=3
n=6
n=4
40 60
P=0.243
80
(A) Representative series of isopotential contour maps recorded at 1-ms intervals indicate the sequential spread of depolarization before and after the ischemic challenge in SprD, ZL, and ZO hearts. Each frame is 8 8 mm2. (B) Representative action potential first derivatives recorded from equidistant sites across the mapping field. (C) Average CV measurements at baseline and after 12 min of ischemia in control and T2DM hearts. (D) Percentage change in CV after 12 min of ischemia relative to baseline in control and T2DM (ZO) hearts. Controls were SprD (n ¼ 3) and ZL (n ¼ 6) animals; T2DM was ZO animals (n ¼ 4). Abbreviations as in Figure 1.
Finally, after the 12-min ischemia protocol, 4 of
which discounts a role for nonischemic activation of
9 control and 3 of 5 T2DM hearts underwent sus-
sarcolemmal KATP channels.
tained arrhythmias within 1 min of reperfusion
We next investigated potential differences in the
(not shown).
response of untreated and DZX-treated T2DM hearts
DZX PARADOXICALLY PROMOTES ARRHYTHMIAS IN
to ischemia. Although APD 75 of untreated T2DM
T2DM. The main purpose of our study was to deter-
hearts failed to shorten progressively in response to
mine whether low-dose DZX, which improves meta-
ischemia (Figure 4B), DZX treatment caused rapid and
bolic
in
sustained APD shortening during the course of
humans, would prevent arrhythmias in T2DM. We
ischemia (Figure 4B). On average, APD 75 of DZX-
therefore examined its effects on key electrophysio-
treated T2DM hearts was shorter by 32.6% (p < 0.01)
logical properties before and during challenge of
after 6 min of ischemia compared with their un-
T2DM hearts with ischemia.
treated counterparts (Figure 4B).
function
in
experimental
models
and
Figure 4A shows the average APD75 rate relation-
The heightened sensitivity of DZX-treated T2DM
ships in untreated and DZX-treated T2DM hearts
hearts to ischemia-induced APD shortening was likely
during baseline perfusion. Clearly, DZX 30 m mol/l did
significant, because all 7 DZX-treated but none of the
not alter baseline APD values at any PCL (Figure 4A),
5 untreated T2DM hearts exhibited onset of sustained
1149
1150
Xie et al.
JACC VOL. 66, NO. 10, 2015
Diazoxide-Mediated Arrhythmias in Diabetes
SEPTEMBER 8, 2015:1144–56
F I G U R E 4 Proarrhythmic Effect of DZX in T2DM
A
C
80
Incidence of VT during ischemia
T2DM (untreated, n=4)
Control
APD75 (ms)
VT (+)
T2DM (DZX, n=6)
60 40 20 0
160
200
240
280
T2DM
120
PCL (ms)
80
DZX
Untreated DZX
*
* * *
T2DM (untreated)
60
*
40 T2DM (DZX, n=6)
20
APD75 (ms)
*
APD before onset of VT or end of ischemia
80
T2DM (untreated, n=4)
60 APD75 (ms)
Untreated
D
B
VT (-)
40
T2DM (DZX)
20
0
0 0
6
8
10
12
0
Ischemia (min)
8
9 10 Ischemia (min)
11
12
(A) Treatment of hearts with 30 mmol/l diazoxide (DZX) did not alter baseline APD75 in T2DM hearts. (B) Average APD75 at PCL of 300 ms in untreated (red) and DZX-treated (blue) T2DM hearts during ischemia. (C) Summary of ventricular tachyarrhythmia (VT) propensity during ischemia in untreated and DZX-treated control and T2DM hearts. Only DZX-treated T2DM hearts exhibited VT during 12 min of no-flow ischemia. (D) APD75 at end of ischemia or before onset of VT in untreated (red) and DZX-treated (blue) T2DM hearts, respectively. Each asterisk indicates that the red dot is significantly different (greater) than the blue dot. APD ¼ action potential duration; other abbreviations as in Figure 1.
ventricular tachyarrhythmia (VT) during the latter
hearts. Whereas on average, DZX-treated T2DM hearts
phase (10 to 12 min) of ischemia (Figure 4C). More-
exhibited a minor trend toward longer APD75 values
over, none of the 9 untreated or 7 DZX-treated control
during ischemia than DZX-treated controls, differ-
hearts exhibited VT during ischemia (Figure 4C).
ences were not statistically significant (Figure 5A). In
As such, DZX provoked VT exclusively in T2DM hearts
contrast, APD heterogeneity, as indexed by the stan-
during ischemia (Figure 4C). Figure 4D shows indi-
dard deviation and range of APD75 values across the
vidual measurements of APD 75, either before the
mapping field, was more than 2-fold greater in DZX-
onset of VT in DZX-treated T2DM hearts or at the end
treated T2DM than in DZX-treated control hearts at
of the 12-min ischemic challenge in untreated T2DM
8 and 10 min of ischemia (Figure 5A). Figure 5B shows
hearts. In all cases, APD 75 of untreated T2DM hearts
representative APD contour maps that illustrate the
was longer after 12 min of ischemia than that of DZX-
greater APD heterogeneity at 8 min of ischemia in
treated T2DM hearts, which underwent a 26% to 56%
DZX-treated T2DM than in control hearts. Figure 5C
reduction before the onset of VT.
shows representative histograms that illustrate the
We investigated the potential mechanism underly-
wider distribution of APD 75 values in individual DZX-
ing the selective proarrhythmic effect of DZX on T2DM
treated T2DM hearts than in their control counterparts.
JACC VOL. 66, NO. 10, 2015
Xie et al.
SEPTEMBER 8, 2015:1144–56
Diazoxide-Mediated Arrhythmias in Diabetes
1151
F I G U R E 5 APD Response to DZX Treatment in Control and T2DM Hearts
Average APD
APD Heterogeneity
A
DZX-treated DZX-treated
P = 0.0039
Ischemia (10 min)
6
P = 0.0008
5
10
n=5
n=6
n=5
n=6
1
n=5
0
Co
T2
nt
ro
l
DM
l ro Co
T2
nt
DM
l ro nt Co
T2
n=5
n=6
n=5
0
DM
l ro nt Co
DM T2
Co
nt
ro
l
0
n=6
DM
n=6
2
T2
n=5
3
l
20
20
4
ro
40
n=6
P = 0.0049
nt
P = 0.087
APD Std. Dev. (ms)
APD Range (ms)
30 P = 0.106
P = 0.045
Co
60 APD75 (ms)
Ischemia (8 min)
40
DM
80
Ischemia (10 min)
DZX-treated
Ischemia (10 min)
T2
Ischemia (8 min)
Ischemia (8 min)
DZX-treated
C
Ischemia (8 min) 40
25
30 35 40 APD75 (ms)
45
Count 45
0 20
50
0 20
10
25
30 35 40 APD75 (ms)
45
50
30 35 40 APD75 (ms)
45
50
APD75
30 Count
20
25
40
APD75
Count
Count
Count
T2DM
30 35 40 APD75 (ms)
30
10
2 mm
25
40
30
20
10
0 20
50
APD75
30
20
20
10
40
20
30
Count
Count
Count
20
0 20
40
APD75
30
10
50 ms
T2DM
40
APD75
30
Control
Control
40
APD75
Count
DZX-treated Ischemia (8 min)
Count
B
20
10
0 20
25
30 35 40 APD75 (ms)
45
50
0 20
25
30 35 40 APD75 (ms)
45
50
(A) Average APD75, APD75 range, and APD75 standard deviation at 8 min and 10 min of ischemia in DZX-treated control and T2DM hearts. (B) Representative APD75 contour maps from control and T2DM hearts at 8 min of ischemia. (C) Representative histograms showing the distribution of APD75 values across the mapping field in individual DZX-treated control and DZX-treated T2DM hearts at 8 min of ischemia. Controls were 3 ZL and 3 SprD rats; T2DM was 5 ZO rats. APD ¼ action potential duration; other abbreviations as in Figures 1 and 3.
We next compared the electrophysiological ef-
control and T2DM hearts (Figure 6B), whereas DZX
fects of DZX with those of the classic SUR2A activator
only
pinacidil. Unlike DZX, which did not affect base-
(Figure 5). Profound action potential shortening was
exerted
a
proarrhythmic
effect
in
T2DM
line APD 75 in either group, pinacidil 100 m mol/l
evident by 5 min of ischemia in pinacidil-treated
caused significant APD75 shortening in T2DM hearts
control and T2DM hearts (Figure 5C).
(p ¼ 0.047) and a strong trend toward shorter levels in
EXPRESSION
controls (p ¼ 0.069) during baseline perfusion
determine whether the proarrhythmic effect of DZX
(Figure 6A). Pinacidil also promoted the rapid (within
in T2DM hearts was related to altered expression of
6 to 7 min) onset of VT during ischemia in both
KATP channels, we measured the mRNA expression
OF
KATP
CHANNEL
SUBUNITS. To
Xie et al.
JACC VOL. 66, NO. 10, 2015
Diazoxide-Mediated Arrhythmias in Diabetes
SEPTEMBER 8, 2015:1144–56
F I G U R E 6 Effects of Pinacidil on Control and T2DM Hearts
A
DZX Control
T2DM
60 APD75 (ms)
APD75 (ms)
60 40 20
n=6
n=6
n=4
n=4
Baseline
DZX
Baseline
DZX
20
Control
VT (+) Untreated Pinacidil
Untreated
T2DM
P = 0.069
P = 0.047
n=4
n=4
n=4
n=4
Baseline
PIN
Baseline
PIN
0
C
Incidence of VT during ischemia
Control
40
0
B
Pinacidil
80
80
T2DM
1152
Control
T2DM
VT (-) Baseline
Pinacidil
Ischemia 5 min
Pinacidil 200ms (A) Average APD75 values for control (n ¼ 4) and T2DM (n ¼ 4) hearts before and after treatment with DZX (30 mmol/l) or pinacidil (100 mmol/l). (B) Summary of VT propensity during ischemia in untreated and pinacidil-treated control and T2DM hearts. Arrhythmogenesis during early ischemia was provoked by pinacidil treatment in both control and T2DM hearts. (C) Representative action potential traces revealing profound action potential shortening at 5 min of ischemia in pinacidil-treated control and T2DM hearts. PIN ¼ pinacidil; other abbreviations as in Figures 1 and 3.
levels of all known KATP channel subunits using real-
that are expressed in the heart. We found no sig-
time PCR. Figure 7A shows the normalized (to beta-
nificant differences in the expression levels of
actin) mRNA expression of the alpha (Kir6.1 and
either isoform between groups (Figure 7B). Primers
Kir6.2) and beta (SUR1A, SUR1B, SUR2A, and SUR2B)
used
subunits of the KATP channel in SprD, ZL, and ZO
Figure 7C.
for
mRNA
measurements
are
shown
in
hearts. We found that samples from ZL and ZO animals exhibited a trend toward increased expres-
DISCUSSION
sion of most K ATP channel subunits compared with their SprD counterparts (Figure 7A); however,
We
this apparent up-regulation only reached statistical
quences of ischemia in T2DM and tested whether low-
significance in the case of SUR2B between ZO and
dose DZX, a potent activator of cardioprotective
SprD animals (Figure 7A). Importantly, there were no
signaling (11–19), could improve electrical function in
significant differences in the expression of any of
that setting. Our major findings were as follows:
these subunits between ZL and ZO hearts (Figure 7A).
1) during normoxic perfusion, T2DM hearts are stable,
As such, none of the observed changes could explain
exhibiting only minor electrophysiological changes;
the proarrhythmic effect that was exclusive to ZO
2) in response to ischemia, T2DM hearts are charac-
animals.
terized by repolarization resistance, because their
A recent study highlighted the potential impor-
investigated
the
electrophysiological
conse-
average APD fails to exhibit progressive shortening;
tance of ROMK (KCNJ1) in formation of the mK ATP
3)
channel (27). We therefore measured the mRNA
clusively in T2DM hearts by promoting rapid and
DZX
treatment
levels of the 2 ROMK isoforms (ROMK1 and ROMK2)
heterogeneous
APD
exacerbates shortening
arrhythmias during
ex-
ischemia
JACC VOL. 66, NO. 10, 2015
Xie et al.
SEPTEMBER 8, 2015:1144–56
Diazoxide-Mediated Arrhythmias in Diabetes
F I G U R E 7 K ATP Channel Subunit Expression in Control and T2DM Hearts
B SprD (n=3)
4
ZL (n=3) ZO (n=3)
3
4
P < 0.05 mRNA Expression Level
mRNA Expression Level
A
2 1
3 2 1 0
0 Kir6.1
C
Kir6.2
Sur1A
Sur1B
Sur2A
ROMK1
Sur2B
ROMK2
Gene name
Forward primer (5’-3’)
Reverse primer (5’-3’)
beta-Actin Kir6.1 Kir6.2 SUR1A SUR1B SUR2A SUR2B ROMK1 ROMK2
CCTCTATGCCAACACAGTGCTGTCT AGCTGGCTGCTCTTCGCTATCA CAACGTCGCCCACAAGAACATC AGCATGCGTGGTACTCATC TGCCTCTCTCTTCCTCACA GGAGTGCGATACTGGTCCAAACCT CATAGCTCATCGGGTTCACACCATT AGTGAGCCAGCCAATCAGTGACA CCAAGGCACTGGGCACCTTTAGC
GCTCAGGAGGAGCAATGATCTTGA CCCTCCAAACCCAATGGTCACT CCAGCTGCACAGGAAGGACATG GCTGAGAGCTCCCTGTGTAG GTTGGATCTCCATGTCTGC CCCGATGCAGAGAACGAGACACT GCATCGAGACACAGGTGCTGTTGT TGGACCCAGACTCAGACAAAGGC TGCCGGGAACGCCCAAATATGTG
(A) Messenger RNA (mRNA) expression of Kir6.1, Kir6.2, SUR1A, SUR1B, SUR2A, and SUR2B measured by real-time polymerase chain reaction (RT-PCR) in SprD (n ¼ 3), ZL (n ¼ 3), and ZO (n ¼ 3) animals. (B) mRNA expression of ROMK1 and ROMK2 measured by RT-PCR in SprD (n ¼ 3), ZL (n ¼ 3), and ZO (n ¼ 3) animals. (C) Primers used for RT-PCR measurements. Beta-actin was used as a normalizing control. KATP ¼ adenosine triphosphate-sensitive potassium channel; other abbreviations as in Figure 1.
(Central Illustration); and 4) changes in the mRNA
speculated that acute ischemia might unmask the
expression of KATP channel subunits do not cause
electrophysiological vulnerability of these hearts. A
these phenotypic differences.
major finding of the present report was the failure of
STABILITY
ELECTROPHYSIOLOGICAL
T2DM hearts to exhibit progressive APD adaptation
PROPERTIES IN T2DM. Our initial objective was to
(shortening) in response to 12 min of no-flow
define the electrophysiological substrate of hearts
ischemia. The mechanism underlying this repolari-
from obese T2DM animals. We focused on conduction
zation resistance will require additional investiga-
slowing and APD prolongation as potentially impor-
tion,
tant features of pathological electrical remodeling.
sarcolemmal K ATP channels, resulting from either
Surprisingly, we found only modest changes in APD
reduced expression or function of these channels.
and CV in this established model of obesity and
Interestingly, Ren et al. (29) found a marked reduc-
T2DM. As such, our present findings discount a major
tion in SUR2A mRNA expression in a model of T1DM.
role for these basal electrophysiological properties in
We measured the transcript levels of all known alpha
OF
BASAL
although
it
suggests
down-regulation
of
the proarrhythmic vulnerability of the diabetic heart.
and beta subunits of K ATP channels, including Kir6.1,
Our present findings are reminiscent of our previous
Kir6.2, SUR1A, SUR1B, SUR2A, SUR2B, ROMK1, and
findings in a guinea pig model of streptozotocin-
ROMK2. An important advantage of our experimental
induced T1DM, in which arrhythmias were only
design was the use of 2 different control groups that
encountered on depletion of the scavenging capacity
were functionally identical but genetically distinct.
of the heart (28).
Specifically, ZL controls share a common genetic
ALTERED APD ADAPTATION DURING ISCHEMIA:
background with their ZO counterparts but are func-
REPOLARIZATION RESISTANCE IN T2DM? Because
tionally identical (in terms of electrophysiology, heart
patients with T2DM are highly susceptible to the
weight, body weight, and blood glucose levels) to
complications
age-matched SprD animals. By measuring changes in
of
coronary
artery
disease,
we
1153
1154
Xie et al.
JACC VOL. 66, NO. 10, 2015
Diazoxide-Mediated Arrhythmias in Diabetes
SEPTEMBER 8, 2015:1144–56
C EN T RA L IL LUSTR AT I ON
DZX-Mediated Arrhythmias in T2DM: Proarrhythmic Effect of DZX in Rats
Xie, C. et al. J Am Coll Cardiol. 2015; 66(10):1144–56.
The proarrhythmic effect of mitochondrial adenosine triphosphate-sensitive potassium (mKATP) channel activation by low-dose diazoxide (DZX) in hearts from rats with type 2 diabetes mellitus (T2DM) is illustrated. Electrophysiological properties were stable at baseline but revealed impaired adaptation to increased metabolic demand during acute ischemia in diabetic (red) compared with nondiabetic (blue) rats. The classically cardioprotective agent DZX caused paradoxical exacerbation of arrhythmias in diabetic but not normal animals, owing to a rapid and spatially heterogeneous shortening of action potentials across the heart, which in turn, created a substrate for reentrant ventricular tachyarrhythmia. APD ¼ action potential duration; APD75 ¼ action potential duration at 75% of repolarization.
the mRNA expression of K ATP channel subunits in all 3
underlying impaired APD adaptation of T2DM hearts
groups, we were able to determine whether a given
to ischemia warrants investigation.
change was functionally important. Surprisingly, we
DZX FOR TREATMENT OF DIABETES MELLITUS. In a
only found a significant change in the expression of
recent clinical study, DZX suppressed endogenous
SUR2B between ZO and SprD animals. Therefore, our
glucose production (6). The presumed antidiabetic
data indicate that the unique functional response of
efficacy of DZX has been linked to preservation of
T2DM hearts to ischemia is unrelated to the transcript
pancreatic beta cells through promotion of beta cell
levels of K ATP channel subunits. This suggests
rest and inhibition of apoptosis. Indeed, pilot clinical
involvement of post-transcriptional mechanisms that
trials using DZX have demonstrated improvement of
regulate protein synthesis, trafficking, insertion, sta-
metabolic function in pre-diabetic obese, hyper-
bility, or gating of these channels. Of note, AMP-
insulinemic patients and in those with T1DM (7–9).
activated protein kinase (AMPK) signaling has been
Numerous experimental studies have established
implicated in preconditioning-induced shortening of
DZX as a potent cardioprotective agent that reduces
the action potential through effects on K ATP channel
reperfusion injury by promoting activation of bene-
activity and recruitment (30). Moreover, drugs that
ficial mitochondrial pathways (11–19). Indeed, a pilot
activate AMPK signaling are a mainstay of treatment
study demonstrated that supplementation of car-
of T2DM patients (31). Impaired AMPK in T2DM may
dioplegia with DZX improved the outcome of patients
contribute
Clearly,
undergoing coronary bypass surgery (10); however,
identification of the exact molecular mechanisms
the infarct-sparing efficacy of DZX in diabetes is less
to
repolarization
resistance.
JACC VOL. 66, NO. 10, 2015
Xie et al.
SEPTEMBER 8, 2015:1144–56
Diazoxide-Mediated Arrhythmias in Diabetes
clear, because DZX failed to reduce infarct size after
(Central Illustration). Specifically, a potent proar-
30 min of left anterior descending coronary artery
rhythmic effect during ischemic challenge should
occlusion in T2DM animals (32). This is consistent
caution against the rapid translation of DZX-based
with the notion that cardioprotective stimuli (such as
therapies for the treatment of T2DM patients, who
ischemic preconditioning), which likely depend on
are indeed at high risk of coronary artery disease,
mK ATP channels, are defective in diabetes (33).
acute myocardial infarction, and silent myocardial
Although the infarct-sparing effects of DZX have
ischemia (4,34–36).
been widely studied, there are no reports of its role in the modulation of electrophysiological properties of
REPRINT REQUESTS AND CORRESPONDENCE: Dr.
T2DM hearts. A major finding of the present report
Fadi G. Akar, Cardiovascular Research Center, Icahn
was that DZX elicited a potent proarrhythmic effect
School of Medicine at Mount Sinai, One Gustave L.
exclusively in the setting of diabetes, because all
Levy Place, Box 1030, New York, New York 10029.
DZX-treated T2DM hearts but none of the untreated
E-mail: [email protected].
T2DM or control hearts exhibited sustained ischemiamediated VT. We used a widely accepted DZX concentration (30 mmol/l) that is relatively specific for the
PERSPECTIVES
mitochondrial pool of KATP channels (15–19). At this concentration, DZX did not alter basal electrophysi-
COMPETENCY IN MEDICAL KNOWLEDGE: The mito-
ological properties. The lack of effect of low-dose DZX
chondrial KATP channel agonist DZX exerts salutary myocardial
on baseline APD strongly discounts a role for nonse-
protective effects against ischemic injury, produces vasodila-
lective activation of sarcolemmal K ATP channels in the
tion in hypertensive patients, and reverses hypoglycemic
proarrhythmic effect that we uncovered. Unlike DZX,
complications in patients with T2DM. However, in an animal
the classic SUR2A-specific K ATP channel activator,
model of T2DM with impaired electrophysiological adaptation
pinacidil, provoked ischemia-mediated VT in all
to acute ischemia, DZX elicited proarrhythmic effects because
hearts (T2DM and control), consistent with previous
of the rapid, heterogeneous shortening of the action potential
findings in failing and nonfailing human heart prep-
unrelated to transcript levels of ATP-sensitive potassium
arations (20). Finally, the lack of arrhythmia protec-
channels.
tion by DZX in T2DM hearts on ischemic challenge is consistent
with
the
lack
of
protection
against
TRANSLATIONAL OUTLOOK: Agents considered safe and
myocardial infarction by ischemic preconditioning
effective in structurally normal hearts might be associated with
and DZX (33).
proarrhythmic risks in the setting of diabetes mellitus. Strategies to mitigate the proarrhythmic effects of mitochondrial KATP
CONCLUSIONS
channel agonists should be investigated in patients with T2DM at
Our current study raises important safety concerns
risk of myocardial ischemia.
regarding the use of DZX in diabetic patients
REFERENCES 1. Candido R, Srivastava P, Cooper ME, Burrell LM. Diabetes mellitus: a cardiovascular disease. Curr Opin Investig Drugs 2003;4:1088–94.
rats at prediabetic stage. J Diabetes Complications 2008;22:46–55.
and autoimmunity but improves glycemic control. Diabetes Care 2010;33:589–94.
2. Ansley DM, Wang B. Oxidative stress and myocardial injury in the diabetic heart. J Pathol
6. Kishore P, Boucai L, Zhang K, et al. Activation of KATP channels suppresses glucose production in humans. J Clin Invest 2011;121:
10. Deja MA, Malinowski M, Golba KS, et al. Diazoxide protects myocardial mitochondria, metabolism, and function during cardiac sur-
2013;229:232–41.
4916–20.
3. Buse JB, Ginsberg HN, Bakris GL, et al. Primary
7. Schreuder T, Karreman M, Rennings A, Ruinemans-
prevention of cardiovascular diseases in people with diabetes mellitus: a scientific statement from the American Heart Association and the American Diabetes Association. Circulation 2007;115:114–26.
Koerts J, Jansen M, de Boer H. Diazoxide-mediated insulin suppression in obese men: a dose-response study. Diabetes Obes Metab 2005;7:239–45.
gery: a double-blind randomized feasibility study of diazoxide-supplemented cardioplegia. J Thorac Cardiovasc Surg 2009;137:997–1004, 1004e1–2.
4. Grundy SM, Benjamin IJ, Burke GL, et al. Diabetes and cardiovascular disease: a statement for healthcare professionals from the American Heart Association [published correction appears in Circulation 2000;101:1629–1631]. Circulation 1999; 100:1134–46. 5. Guo Z, Bu S, Yu Y, et al. Diazoxide prevents abdominal adiposity and fatty liver in obese OLETF
8. van Boekel G, Loves S, van Sorge A, RuinemansKoerts J, Rijnders T, de Boer H. Weight loss in obese men by caloric restriction and high-dose diazoxide-mediated insulin suppression. Diabetes Obes Metab 2008;10:1195–203. 9. Radtke MA, Nermoen I, Kollind M, et al. Six months of diazoxide treatment at bedtime in newly diagnosed subjects with type 1 diabetes does not influence parameters of b-cell function
11. Seharaseyon J, Ohler A, Sasaki N, et al. Molecular composition of mitochondrial ATP-sensitive potassium channels probed by viral Kir gene transfer. J Mol Cell Cardiol 2000;32:1923–30. 12. Sato T, Marban E. The role of mitochondrial KATP channels in cardioprotection. Basic Res Cardiol 2000;95:285–9. 13. Sato T, Sasaki N, Seharaseyon J, O’Rourke B, Marbán E. Selective pharmacological agents implicate mitochondrial but not sarcolemmal KATP
1155
1156
Xie et al.
JACC VOL. 66, NO. 10, 2015
Diazoxide-Mediated Arrhythmias in Diabetes
SEPTEMBER 8, 2015:1144–56
channels in ischemic cardioprotection. Circulation 2000;101:2418–23. 14. Hu H, Sato T, Seharaseyon J, et al. Pharmacological and histochemical distinctions between molecularly defined sarcolemmal KATP channels and native cardiac mitochondrial KATP channels. Mol Pharmacol 1999;55:1000–5. 15. Carroll R, Gant VA, Yellon DM. Mitochondrial KATP channel opening protects a human atrialderived cell line by a mechanism involving free radical generation. Cardiovasc Res 2001;51: 691–700. 16. Costa AD, Jakob R, Costa CL, Andrukhiv K, West IC, Garlid KD. The mechanism by which the mitochondrial ATP-sensitive Kþ channel opening and H2O2 inhibit the mitochondrial permeability transition. J Biol Chem 2006;281:20801–8. 17. Facundo HT, de Paula JG, Kowaltowski AJ. Mitochondrial ATP-sensitive Kþ channels prevent oxidative stress, permeability transition and cell death. J Bioenerg Biomembr 2005;37:75–82. 18. Garlid KD, Paucek P, Yarov-Yarovoy V, et al. Cardioprotective effect of diazoxide and its interaction with mitochondrial ATP-sensitive Kþ channels: possible mechanism of cardioprotection. Circ Res 1997;81:1072–82. 19. Quinlan CL, Costa AD, Costa CL, Pierre SV, Dos
21. Sárközy M, Zvara A, Gyémánt N, et al. Metabolic syndrome influences cardiac gene expression pattern at the transcript level in male ZDF rats. Cardiovasc Diabetol 2013;12:16. 22. Akar FG, Aon MA, Tomaselli GF, O’Rourke B. The mitochondrial origin of postischemic arrhythmias. J Clin Invest 2005;115:3527–35. 23. Akar FG, Nass RD, Hahn S, et al. Dynamic changes in conduction velocity and gap junction properties during development of pacing-induced heart failure. Am J Physiol 2007;293:H1223–30. 24. Akar FG, Spragg DD, Tunin RS, Kass DA, Tomaselli GF. Mechanisms underlying conduction slowing and arrhythmogenesis in nonischemic dilated cardiomyopathy. Circ Res 2004;95:717–25. 25. Jin H, Chemaly ER, Lee A, et al. Mechanoelectrical remodeling and arrhythmias during progression of hypertrophy. FASEB J 2010;24: 451–63. 26. Xie C, Kauffman J, Akar FG. Functional crosstalk between the mitochondrial PTP and KATP channels determine arrhythmic vulnerability to oxidative stress. Front Physiol 2014;5:264. 27. Foster DB, Ho AS, Rucker J, et al. Mitochondrial ROMK channel is a molecular component of mitoKATP. Circ Res 2012;111:446–54.
Santos P, Garlid KD. Conditioning the heart induces formation of signalosomes that interact with mitochondria to open mitoKATP channels. Am J Physiol Heart Circ Physiol 2008;295:H953–61.
28. Xie C, Biary N, Tocchetti CG, et al. Glutathione oxidation unmasks proarrhythmic vulnerability of
20. Fedorov VV, Glukhov AV, Ambrosi CM, et al. Effects of KATP channel openers diazoxide and pinacidil in coronary-perfused atria and ventricles from failing and non-failing human hearts. J Mol
29. Ren Y, Xu X, Wang X. Altered mRNA expression of ATP-sensitive and inward rectifier potassium channel subunits in streptozotocin-induced diabetic rat heart and aorta. J Pharmacol Sci 2003;
Cell Cardiol 2011;51:215–25.
93:478–83.
chronically hyperglycemic guinea pigs. Am J Physiol Heart Circ Physiol 2013;304:H916–26.
30. Sukhodub A, Jovanovic S, Du Q, et al. AMPactivated protein kinase mediates preconditioning in cardiomyocytes by regulating activity and trafficking of sarcolemmal ATP-sensitive Kþ channels. J Cell Physiol 2007;210:224–36. 31. Hardie DG. Role of AMP-activated protein kinase in the metabolic syndrome and in heart disease. FEBS Lett 2008;582:81–9. 32. Katakam PV, Jordan JE, Snipes JA, Tulbert CD, Miller AW, Busija DW. Myocardial preconditioning against ischemia-reperfusion injury is abolished in Zucker obese rats with insulin resistance. Am J Physiol Regul Integr Comp Physiol 2007;292: R920–6. 33. Kristiansen SB, Løfgren B, Støttrup NB, et al. Ischaemic preconditioning does not protect the heart in obese and lean animal models of type 2 diabetes. Diabetologia 2004;47:1716–21. 34. Jacoby RM, Nesto RW. Acute myocardial infarction in the diabetic patient: pathophysiology, clinical course and prognosis. J Am Coll Cardiol 1992;20:736–44. 35. Margolis JR, Kannel WS, Feinleib M, Dawber TR, McNamara PM. Clinical features of unrecognized myocardial infarction–silent and symptomatic: eighteen year follow-up: the Framingham study. Am J Cardiol 1973;32:1–7. 36. Scognamiglio R, Negut C, Ramondo A, Tiengo A, Avogaro A. Detection of coronary artery disease in asymptomatic patients with type 2 diabetes mellitus. J Am Coll Cardiol 2006;47:65–71.
KEY WORDS action potentials, ischemia, mitochondria, obesity, repolarization, ventricular tachycardia