Effect of Carbenoxolone on Arrhythmogenesis in Rat Ventricular Muscle

Abstract

Background: Connexin43 (Cx43) is a major connexin that forms gap junction (GJ) channels in the heart and is also present in the cell membrane as unopposed/non-junctional hemichannels and in the inner mitochondrial membrane. By using carbenoxolone (CBX), a blocker of Cx43, the effect of the blockade of Cx43 on Ca²⁺ waves and triggered arrhythmias in the myocardium with non-uniform contraction was examined.

Methods and Results: Trabeculae were obtained from rat hearts. Force, [Ca²⁺]_i, and the diffusion coefficient were measured. Non-uniform contraction was produced with a 2,3-butanedione monoxime jet. Ca²⁺ waves were induced by electrical stimulation. Inducibility of arrhythmias was estimated based on the minimal [Ca²⁺]_o at which arrhythmias were induced. The Ca²⁺ spark rate was measured in isolated single rat ventricular myocytes. CBX reduced the GJ permeability, whereas it did not change force and [Ca²⁺]_i transients. CBX increased the Ca²⁺ leak from the sarcoplasmic reticulum in trabeculae and increased the Ca²⁺ spark rate in isolated single myocytes. CBX increased the velocity of Ca²⁺ waves and further increased the inducibility of arrhythmias. Modulation of mitochondrial K_ATP channels by diazoxide, cromakalim and 5-hydroxydecanoic acid affected the inducibility of arrhythmias increased by CBX.

Conclusions: These results suggest that in diseased hearts, Cx43 plays an important role in the occurrence of triggered arrhythmias, probably under the modulation of mitochondrial K_ATP channels. (Circ J 2016; 80: 76–84)

Life-threatening ventricular arrhythmias are likely to occur in diseased hearts, such as hearts with cardiac hypertrophy and failure.¹ In diseased hearts, non-uniform muscle contraction commonly occurs² as a result of muscle damage,² heterogeneous adrenergic activation, heterogeneous protein expression or non-uniform electrical activation. In the myocardium with non-uniform contraction, regional differences in contractile strength may cause paradoxical stretching of the impaired muscle by contractions of the more viable neighboring muscle.³ During the relaxation of the more viable muscle, the paradoxical shortening of the impaired muscle dissociates Ca²⁺ from the myofilaments and then initiates Ca²⁺ waves.³ These Ca²⁺ waves cause delayed afterdepolarizations (DADs) through the activation of Ca²⁺-activated transient inward currents across the sarcolemma, and their propagation features such as velocity and amplitude are deeply involved with the formation of DADs.⁴ Because DADs cause arrhythmias associated with catecholamine excess,⁵ heart failure,⁶ and mutations of the ryanodine receptor or calsequestrin in the sarcoplasmic reticulum (SR),⁷ the velocity of Ca²⁺ waves is one of the determinants of arrhythmogenesis.⁴ In trabeculae, the propagation velocity of Ca²⁺ waves is determined by cellular Ca²⁺ loading, the open probability of the SR-Ca²⁺ release channels,⁸ and gap junction (GJ) communication.⁹

Connexin43 (Cx43) is a major connexin that forms GJ channels in the heart and is especially abundant in ventricular muscle.¹⁰ Not only present as GJ channels, Cx43 is also present as unopposed/non-junctional hemichannels in the cell membrane^11,12 and is also found in the inner mitochondrial membrane.¹³ Interestingly, the presence of Cx43 is needed for cardioprotection^14–16 by ischemic preconditioning and pharmacological preconditioning with diazoxide,¹⁷ a mitochondrial K_ATP channel opener.^18,19 Although activation of mitochondrial K_ATP channels by diazoxide reduces ventricular arrhythmias induced by ischemia,²⁰ it has not yet been established why Cx43 is needed when activation of mitochondrial K_ATP channels reduces such arrhythmias.

Carbenoxolone (CBX) has been used as a blocker of GJ channels in rabbit,²¹ guinea pig,²² dog,²³ and rat hearts.²⁴ In addition, it blocks Cx43 hemichannels²⁵ and mitochondrial Cx43.²⁶ Therefore, to determine what role Cx43 plays when mitochondrial K_ATP channels are involved in arrhythmogenesis in diseased hearts, we investigated how the blockade of Cx43 by CBX and the modulation of mitochondrial K_ATP channels affect the inducibility of triggered arrhythmias in the myocardium with non-uniform contraction. The results in the present study suggest that Cx43 plays an important role in arrhythmogenesis, probably under the modulation of mitochondrial K_ATP channels in diseased hearts.

Methods

Measurements of Force, [Ca²⁺]_i, and Equivalent Diffusion Coefficient for Fura-2 in Rat Trabeculae

All animal procedures were performed according to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996) and were approved by the Ethics Review Board of Tohoku University (approval reference number: 2011-33). After the animals had been adequately anesthetized, trabeculae were dissected from the right ventricles of rat hearts. Force, [Ca²⁺]_i, and the equivalent diffusion coefficient for fura-2 were measured as previously reported.^8,27

Experimental Protocol With Trabeculae

Non-uniform muscle contraction was produced with a 2,3-butanedione monoxime (BDM) jet as previously reported.^3,28 The equivalent diffusion coefficient for fura-2 was first calculated without electrical stimulation before, just after, and 1 h after superfusion with 5 and 10 μmol/L CBX for 15 min ([Ca²⁺]_o=0.7 mmol/L). In the muscle model with non-uniform contraction, force, [Ca²⁺]_i, the velocity of Ca²⁺ waves, and the minimal [Ca²⁺]_o ([Ca²⁺]_o,min) at which arrhythmias were induced by rapid electrical stimulation were then examined before and after superfusion with 5 μmol/L CBX for 15 min (24℃). Additionally, the [Ca²⁺]_o,min was again determined in the presence of 100 μmol/L diazoxide, 10 μmol/L cromakalim, and 200 μmol/L 5-hydroxydecanoic acid (5-HD) (24℃).²⁹ The mitochondrial membrane potential (∆Ψ_m) was estimated using tetramethylrhodamine methylester (TMRM) fluorescence.³⁰

Measurements of Ca²⁺ Sparks in Isolated Rat Cardiac Myocytes

Myocytes were enzymatically isolated from ventricles of rat hearts, and Ca²⁺ sparks were measured using fluo-4 and fast 2D confocal Ca²⁺ imaging.³¹ At different CBX concentrations (0, 5, 10, and 50 μmol/L), Ca²⁺ sparks were counted between stimulated transients at a steady basal Ca²⁺ level ([Ca²⁺]_o=2 mmol/L, 35℃).

Statistical Analysis

All measurements were expressed as mean±SEM. Statistical analysis was performed with 1-way repeated-measures ANOVA, with Tukey-Kramer for multiple comparisons and a paired t-test for 2-group comparisons when the data were normally distributed. Otherwise, the Friedman test with Scheffe was used for multiple comparisons, and the Wilcoxon signed-rank test was used for 2-group comparisons unless otherwise mentioned. These analyses were performed using software for statistical analysis (Ekuseru-Toukei 2012, Social Survey Research Information Co, Ltd, Tokyo, Japan). Values of P<0.05 were considered to be significant.

The authors had full access to and take full responsibility for the integrity of the data. All authors have read and agreed to the manuscript as written.

See expanded ‘Methods’ section in Supplementary File 1.

Results

The effect of CBX on GJ communication was first investigated in rat ventricular trabeculae. The diffusion of fura-2 within trabeculae was recorded after a 15-min superfusion with 5 and 10 μmol/L CBX; and the diffusion coefficients for fura-2 were calculated. As shown in Table, the diffusion coefficients were significantly decreased after superfusion with both 5 and 10 μmol/L CBX, and these decreases in the diffusion coefficients lasted for at least 1 h. When the GJ permeability was calculated using the previously reported diffusion coefficient in the myoplasm (42.5×10^–8 cm²/s),²⁷ it was 6.13×10^–5 cm/s before superfusion with CBX and was reduced to 1.38×10^–5 and 1.81×10^–5 cm/s just after and 1 h after superfusion with 5 μmol/L CBX, respectively. This means that 5 μmol/L CBX reduces the GJ permeability to 20–30% of its initial value and that this reduction lasts for at least 1 h after superfusion. In addition, this result also suggests that 5 μmol/L CBX actually blocks Cx43 because GJ channels in adult rat ventricular muscle are chiefly formed by Cx43.¹⁰

Table. Effect of CBX on Diffusion Coefficients for Fura-2 in Rat Cardiac Trabeculae

	Control	0 h	1 h
5 μmol/L CBX (n=7)
Diffusion coefficient (×10−8 cm2/s)	25.1±4.3	10.4±2.1#	12.7±1.7*
10 μmol/L CBX (n=6)
Diffusion coefficient (×10−8 cm2/s)	26.1±6.1	5.3±2.4#	1.7±0.7#

Control, 0 h, and 1 h indicate the diffusion coefficients without, just after, and 1 h after 15-min superfusion with CBX, respectively. Superfusion with CBX significantly decreased the diffusion coefficients for fura-2 in rat cardiac trabeculae. Results are expressed as mean±SEM. *P<0.05 vs. control, ^#P<0.01 vs. control. CBX, carbenoxolone.

CBX has been reported to affect neither the action potential in rabbit hearts²¹ nor the electrocardiogram in human hearts,³² and thus the effect of CBX on force and intracellular Ca²⁺ handling was next assessed in rat trabeculae. Force and [Ca²⁺]_i transients were measured during electrical stimulation with 2-s intervals before and after superfusion with 5 μmol/L CBX ([Ca²⁺]_o=2.0 mmol/L). As shown in Figure 1, 5 μmol/L CBX caused no significant changes in the developed force, the peak [Ca²⁺]_i, the diastolic [Ca²⁺]_i, and the time constant of the decline in [Ca²⁺]_i transients during electrical stimulation, suggesting that 5 μmol/L CBX has no significant effect on intracellular Ca²⁺ handling in rat cardiac muscle. To further examine the effect of CBX on intracellular Ca²⁺ handling under conditions of Ca²⁺ overload, force and [Ca²⁺]_i transients were again measured at higher [Ca²⁺]_o. Figure 2A shows force and changes in [Ca²⁺]_i along a trabecula between 1 and 5 s after the stimulus during electrical stimulation at 5-s intervals ([Ca²⁺]_o=6 mmol/L). Note that spontaneous increases in [Ca²⁺]_i did not occur within the trabecula before superfusion with 5 μmol/L CBX, whereas they did occur at least 3 times after superfusion. Summary data show that 5 μmol/L CBX significantly increased the number of spontaneous increases in [Ca²⁺]_i within trabeculae (Figure 2B), whereas it caused no significant change in force during electrical stimulation at 5-s intervals ([Ca²⁺]_o=4.8±0.6 mmol/L, Figure 2C), suggesting that CBX increases diastolic Ca²⁺ leak from the SR without an increase in the Ca²⁺ loading within the SR in rat trabeculae.

Figure 1.

Effect of 5 µmol/L carbenoxolone (CBX) on the developed force and [Ca²⁺]_i transient. (A) Representative recordings of force (Upper panel) and [Ca²⁺]_i (Lower panel). Black and red lines indicate the recordings before and after superfusion with 5 µmol/L CBX, respectively. The trabecula was stimulated electrically at 2-s intervals. An arrow with ST indicates the moment of electrical stimulation ([Ca²⁺]_o 2.0 mmol/L, 24.3℃, Experiment number 110810). (B) Summary data concerning the effect of CBX on the developed force. CBX did not significantly change the developed force (P=0.23, n=12, [Ca²⁺]_o 2.0 mmol/L). (C) Summary data concerning the effect of CBX on [Ca²⁺]_i. CBX did not significantly change the peak [Ca²⁺]_i (P=0.43), the diastolic [Ca²⁺]_i (P=0.46), or the time constant of [Ca²⁺]_i decline (P=0.34, n=7, [Ca²⁺]_o 2.0 mmol/L).

Figure 2.

Effect of carbenoxolone (CBX) on spontaneous increases in [Ca²⁺]_i in trabeculae and on Ca²⁺ sparks in isolated single myocytes. (A) Representative recordings of force (Top) and spatial changes in [Ca²⁺]_i (Bottom) between 1 and 5 s after a stimulus during electrical stimulation at 5-s intervals before (Left panel) and after superfusion with 5 µmol/L CBX (Right panel). After superfusion, spontaneous increases in [Ca²⁺]_i did occur at least 3 times (yellow arrows) within trabeculae ([Ca²⁺]_o 6.0 mmol/L, 23.9℃, Experiment number 121003). (B) Summary data concerning the effect of 5 µmol/L CBX on the number of spontaneous increases in [Ca²⁺]_i within trabeculae. CBX significantly increased the number (n=7, [Ca²⁺]_o 4.8±0.6 mmol/L). *P<0.05 vs. CBX (–). (C) Summary data concerning the effect of 5 µmol/L CBX on the developed force of trabeculae during electrical stimulation with 5-s intervals. CBX did not change the developed force (P=0.97, n=7, [Ca²⁺]_o 4.8±0.6 mmol/L). (D) Effect of CBX on Ca²⁺ sparks in isolated single myocytes during stimulation at 5-s intervals. CBX increased the number of Ca²⁺ sparks depending on the concentrations of 0 (n=12), 5 (n=15), 10 (n=23) and 50 µmol/L (n=13) (P<0.0001, Jonckheere-Terpstra trend test). *P<0.05 vs. CBX (–), ^#P<0.01 vs. CBX (–) (Mann-Whitney U-test; P-values being adjusted for multiple comparisons by using Holm’s method). ([Ca²⁺]_o 2 mmol/L, 35℃).

It is possible that CBX causes such an increase in diastolic Ca²⁺ leak from the SR by blocking GJ channels within trabeculae. Thus, using isolated single myocytes, Ca²⁺ sparks were counted before and after superfusion with CBX (Figure 2D). Even in isolated single myocytes, 5 μmol/L CBX significantly increased the number of Ca²⁺ sparks. In addition, the number of Ca²⁺ sparks increased depending on the concentration of CBX. These results suggest that, depending on the concentration, CBX does increase diastolic Ca²⁺ leak from the SR, but not by blocking GJ channels.

It has been previously reported that the propagation velocity of Ca²⁺ waves is decreased by the reduction of GJ communication,⁹ and is increased by the increase in Ca²⁺ leak from the SR.⁸ To investigate which effect is more prominent in the case of CBX, the velocity of Ca²⁺ waves was measured before and after superfusion with 5 μmol/L CBX. In a trabecula with non-uniform contraction created by a regional jet of BDM, a Ca²⁺ wave occurred within the jet-exposed region after rapid electrical stimulation and was propagated through GJs within the trabecula (Figure 3A Left panel). Superfusion with 5 μmol/L CBX increased the velocity of the Ca²⁺ wave (Figure 3A Right panel). Summary data indicate that 5 μmol/L CBX did not change the developed force by the last stimulus of the train (Figure 3B), but significantly increased both the velocity of Ca²⁺ waves (Figure 3C) and the amplitude of aftercontractions during the wave propagation (Figure 3D). These results suggest that after superfusion with 5 μmol/L CBX, an increase in Ca²⁺ leak from the SR more prominently affects the velocity of Ca²⁺ waves than the reduction of GJ communication.

Figure 3.

Effect of 5 µmol/L carbenoxolone (CBX) on the velocity of Ca²⁺ waves. (A) Representative recordings of force (Top) and spatial changes in [Ca²⁺]_i (Bottom) during the last 3 electrical stimuli (400-ms stimulus interval) before (black line, Left panel) and after superfusion with CBX (red line, Right panel). Before superfusion with CBX, Ca²⁺ surged around the 2,3-butanedionemonoxime (BDM) jet-exposed region and propagated as a wave (white arrow, 2.1 mm/s) through the trabeculae after the trains of electrical stimuli. After superfusion, the Ca²⁺ wave propagated faster (white arrow, 4.6 mm/s). Arrows with ST indicate the moments of electrical stimulation ([Ca²⁺]_o 2.0 mmol/L, 24.6℃, Experiment number 110801). (B) Summary data concerning the effect of CBX on the developed force by the last electrical stimulus of the train (400-ms stimulus interval, [Ca²⁺]_o 2.0 mmol/L). CBX did not change the developed force (P=0.43, n=11). (C) Summary data concerning the effect of CBX on the velocity of Ca²⁺ waves. CBX significantly increased the velocity of Ca²⁺ waves (n=11, [Ca²⁺]_o 2.0 mmol/L). ^#P<0.01 vs. CBX (–). (D) Summary data concerning the effect of CBX on the amplitude of aftercontractions (AC force) during Ca²⁺ wave propagation. CBX significantly increased the AC force (n=11, [Ca²⁺]_o 2.0 mmol/L). ^#P<0.01 vs. CBX (–).

As an increase in the velocity of Ca²⁺ waves is deeply involved with arrhythmogenesis,⁴ the effect of 5 μmol/L CBX on the inducibility of triggered arrhythmias was then investigated. As shown in Figure 4A, before superfusion with 5 μmol/L CBX, arrhythmias were not induced even at 7 mmol/L [Ca²⁺]_o by electrical stimulation at 300 ms intervals for 30 s, whereas they were induced at 1 mmol/L [Ca²⁺]_o after superfusion. The minimal [Ca²⁺]_o ([Ca²⁺]_o,min) at which arrhythmias were induced by electrical stimulation was significantly decreased after superfusion with 5 μmol/L CBX (Figure 4B). It is well known that ouabain increases arrhythmogenesis, and in this study, it (5 μmol/L) actually decreased the [Ca²⁺]_o,min, as shown in Figure 4C, supporting the concept that the changes in the [Ca²⁺]_o,min are deeply involved with arrhythmogenesis.

Figure 4.

Effect of 5 µmol/L carbenoxolone (CBX) on the inducibility of triggered arrhythmias. (A) Representative recordings of force before (Upper panel) and after superfusion with 5 µmol/L CBX (Lower panel). Before superfusion, no spontaneous contractions were induced by electrical stimulation at 7 mmol/L [Ca²⁺]_o. In contrast, after superfusion, spontaneous contractions were induced by electrical stimulation even at 1 mmol/L [Ca²⁺]_o and sustained for more than 10 s. Arrows with ST indicate the moments of electrical stimulation (23.7℃, Experiment number 111013). (B) Summary data concerning the effect of CBX on the minimal [Ca²⁺]_o ([Ca²⁺]_o,min), at which spontaneous contractions were induced by electrical stimulation. After superfusion with CBX, spontaneous twitch contractions were induced by electrical stimulation at significantly lower [Ca²⁺]_o (n=6). *P<0.05 vs. CBX (–). (C) Summary data concerning the effect of 5 µmol/L ouabain on the [Ca²⁺]_o,min. In the presence of ouabain, spontaneous twitch contractions were induced by electrical stimulation at significantly lower [Ca²⁺]_o (n=6). *P<0.05 vs. CBX (–).

It is still unclear, however, how CBX increases Ca²⁺ leak from the SR and increases arrhythmogenesis. It has been reported that CBX blocks mitochondrial Cx43 and that the blockade of mitochondrial Cx43 reduces mitochondrial K_ATP channel currents.²⁶ Thus, as the next step, we focused on the effect of mitochondrial K_ATP channels on arrhythmogenesis. In the presence of 100 μmol/L diazoxide, 5 μmol/L CBX no longer changed the amplitude of aftercontractions during Ca²⁺ wave propagation (Figure 5A) nor the [Ca²⁺]_o,min (Figures 5B,C). Because it is probable that diazoxide may also affect sarcolemmal K_ATP channels, the [Ca²⁺]_o,min was measured again in the presence of 10 μmol/L cromakalim. In the presence of cromakalim, 5 μmol/L CBX did not change the [Ca²⁺]_o,min (Figure 5D), as is the case in the presence of diazoxide. In addition, in the presence of cromakalim, 200 μmol/L 5-HD also did not change the [Ca²⁺]_o,min (Figure 5E), suggesting that the blockade of mitochondrial K_ATP channels cannot increase the inducibility of arrhythmias by itself. Interestingly, CBX decreased the [Ca²⁺]_o,min at last in the presence of 5-HD (Figure 5E). These results suggest that CBX increases the inducibility of arrhythmias only when mitochondrial K_ATP channels, rather than sarcolemmal K_ATP channels, are closed.

Figure 5.

Inducibility of triggered arrhythmias under the modulation of K_ATP channels. (A) Summary data concerning the effect of 5 µmol/L carbenoxolone (CBX) on the amplitude of aftercontractions (AC force) in the presence of 100 µmol/L diazoxide. CBX did not change the amplitude of the AC force (n=10, P=0.17). (B) Representative recordings of force before (Upper panel) and after superfusion with 5 µmol/L CBX (Lower panel) in the presence of 100 µmol/L diazoxide. Spontaneous contractions were not induced by electrical stimulation, even after superfusion with CBX. Arrows with ST indicate the moments of electrical stimulation (23.8℃, Experiment number 140303). (C) Summary data concerning the effect of CBX on the minimal [Ca²⁺]_o ([Ca²⁺]_o,min) at which spontaneous contractions were induced by electrical stimulation in the presence of 100 µmol/L diazoxide. CBX did not change the [Ca²⁺]_o,min (n=7, P=0.18). (D) Summary data concerning the effect of CBX on [Ca²⁺]_o,min in the presence of 10 µmol/L cromakalim. CBX did not change the [Ca²⁺]_o,min (n=7, P=1.00). (E) Summary data concerning the effect of CBX on [Ca²⁺]_o,min in the presence of 10 µmol/L cromakalim and 200 µmol/L 5-hydroxydecanoic acid (5-HD). CBX decreased the [Ca²⁺]_o,min in the presence of both cromakalim and 5-HD (n=4). *P<0.05 vs. 5-HD (–) CBX (–) and 5-HD (+) CBX (–). (F) Recordings of tetramethylrhodamine methylester (TMRM) fluorescence before and after superfusion of CBX in the presence (blue circles) and absence of 100 µmol/L diazoxide (red circles). The Y-axis shows the changes (%) in the TMRM fluorescence ([Ca²⁺]_o 2.0 mmol/L, 24.1℃, Experiment numbers 150629, 150701).

Finally, the effect of CBX on ∆Ψ_m was observed using TMRM fluorescence. As shown in Figure 5F, TMRM fluorescence was decreased after superfusion with 5 μmol/L CBX, and this decrease was suppressed in the presence of 100 μmol/L diazoxide, suggesting that CBX depolarizes ∆Ψ_m under the modulation of mitochondrial K_ATP channels.

Discussion

The present study characterized how CBX affects arrhythmogenesis in the myocardium with non-uniform contraction. To the best of our knowledge, it shows for the first time that regardless of the reduction of the GJ permeability, CBX increases the inducibility of arrhythmias and that activation of mitochondrial K_ATP channels suppresses this increase. These results suggest that Cx43 plays an important role in the occurrence of triggered arrhythmias, probably under the modulation of mitochondrial K_ATP channels in diseased hearts.

CBX and GJ Communication

It has been reported that the common logarithm of GJ permeability inversely correlates with molecular weight; the values of the permeability range from 7×10^–5 cm/s (mol wt 559, lissamine rhodamine) to 770×10^–5 cm/s (mol wt 39, K⁺).³³ In the present study, the calculated GJ permeability for fura-2 was 6.13×10^–5 cm/s (mol wt 832), which is similar to the GJ permeability determined for molecules of a similar size. Although it is well known that high Ca²⁺ levels close GJs, we assume that the [Ca²⁺]_i level during the measurement of the diffusion coefficients was not high enough to affect the GJ permeability because they were measured without electrical stimulation at 0.7 mmol/L [Ca²⁺]_o.

CBX has been used as a GJ blocker at concentrations ranging from 15 to 100 μmol/L in rabbit,²¹ guinea pig,²² and dog hearts.²³ In guinea pig hearts, CBX at concentrations of less than 13 μmol/L did not change the conduction velocity of excitation, whereas in rat hearts, CBX even at a concentration of 10 μmol/L, affected it.²⁴ The present study showed that in rat trabeculae, 5 μmol/L CBX reduced the GJ permeability to 20–30% of its initial value. It has been reported that in adult rat ventricular muscle, GJ channels are formed by Cx43¹⁰ and that a reduction of Cx43 using transgenic methodology slows the conduction velocity of excitation in ventricular muscle,^34,35 suggesting that CBX reduces the GJ permeability in rat trabeculae by blocking Cx43, although CBX has also been reported to block other Cx families.³⁶

In multicellular cardiac muscle, the Ca²⁺ movement through GJs determines the velocity of Ca²⁺ waves because Ca²⁺ waves propagate within the cardiac muscle by Ca²⁺-induced Ca²⁺ release from the SR mediated by Ca²⁺ diffusion through GJs.²⁷ Actually, it has been previously reported that in rat trabeculae, the velocity of Ca²⁺ waves is decreased by the reduction of GJ communication.⁹ As for cardiac muscle excitation, ventricular conduction of excitation was not altered by a 50% reduction of Cx43 in some studies,³⁷ although a similar 50% reduction of Cx43 slowed the conduction velocity in other studies,³⁵ and a reduction of Cx43 below 20% of control levels prominently slowed ventricular conduction and caused re-entrant arrhythmias.³⁴ In the present study, 5 μmol/L CBX reduced the GJ permeability to 20–30% of its initial value, but increased the velocity of Ca²⁺ waves (Figure 3), suggesting that the reduction of GJ communication to such a level is not sufficient to decrease the velocity of Ca²⁺ waves under the condition that Ca²⁺ leak from the SR is increased.

CBX and Ca²⁺ Leak From the SR

In the present study, 5 μmol/L CBX increased the number of spontaneous increases in [Ca²⁺]_i within trabeculae and increased the number of Ca²⁺ sparks depending on its concentration in isolated single myocytes, although it did not change the developed force and [Ca²⁺]_i transients by electrical stimulation (Figures 1,2). These results suggest that CBX increases Ca²⁺ leak from the SR not by increasing the Ca²⁺ loading in the SR but probably by lowering the SR threshold for Ca²⁺ leak. Such an increase in Ca²⁺ leak from the SR accelerates Ca²⁺ waves⁸ because Ca²⁺ waves result from the spatial and temporal summation of Ca²⁺ sparks. As shown in Figure 4, CBX also increased the inducibility of arrhythmias, which is consistent with past reports using other experimental models.²³ Acceleration of Ca²⁺ waves increases arrhythmogenesis through enhancement of DADs,⁴ suggesting that CBX may have increased it by the acceleration of Ca²⁺ waves, as shown in Figures 3. Therefore, these results suggest that CBX accelerates Ca²⁺ waves by an increase in Ca²⁺ leak from the SR and thereby increases arrhythmogenesis.

Diazoxide and cromakalim have been used as K_ATP channel openers,^18,19,29 and diazoxide is 2000-fold more potent at the mitochondrial channel than at the surface channel.³⁸ In the present study, CBX did not increase the inducibility of arrhythmias in the presence of diazoxide (Figures 5B,C) or cromakalim (Figure 5D), but increased it in the presence of 5-HD (Figure 5E), suggesting that CBX increases the inducibility only when mitochondrial K_ATP channels are closed. In addition, as shown in Figure 5E, 5-HD did not increase the inducibility of arrhythmias by itself, suggesting that CBX increases the inducibility not by blocking mitochondrial K_ATP channels.

As shown in Figure 5F, a decrease in TMRM fluorescence after the addition of CBX was suppressed by diazoxide, suggesting that the blockade of Cx43 depolarizes ∆Ψ_m and that activation of mitochondrial K_ATP channels suppresses this depolarization. It is still controversial as to whether diazoxide affects ∆Ψ_m.³⁹ Some results indicate no effect, and others indicate depolarization of ∆Ψ_m. Furthermore, it has also been reported that ∆Ψ_m depolarization caused by anoxia/reoxygenation or reactive oxygen species is prevented by diazoxide.³⁹ One possible explanation is that activation of mitochondrial K_ATP channels by diazoxide may partially depolarize ∆Ψ_m and cause a compensatory increase in proton pumping and cellular respiration to maintain ∆Ψ_m and oxidative phosphorylation,²⁹ although the underlying mechanism has not yet been established.⁴⁰ Additionally, it has been reported that ∆Ψ_m depolarization may be involved with the occurrence of arrhythmias,⁴¹ further supporting the possibility that CBX may have increased arrhythmogenesis by depolarizing ∆Ψ_m in this study. As discussed above, CBX blocked Cx43 in the present study and has been reported to block mitochondrial Cx43.^26,27 Taken together, the results in the present study suggest that the blockage of Cx43 increases arrhythmogenesis, probably through ∆Ψ_m depolarization under the modulation of mitochondrial K_ATP channels.

In diseased hearts,^42,43 such as those suffering from myocardial infarction and heart failure, impaired muscle is occasionally distributed within hearts, causing regional differences in contractile strength around the muscle. Such regional differences in contractile strength may produce paradoxical stretching and shortening of the impaired muscle by contractions of the more viable neighboring muscle, thereby inducing Ca²⁺ waves and triggered arrhythmias.^3,44 The findings of the present study show that in the myocardium with non-uniform contraction, Cx43 plays some role in Ca²⁺ wave propagation and the occurrence of arrhythmias, probably under the modulation of mitochondrial K_ATP channels, making Cx43 an attractive therapeutic target for suppressing triggered arrhythmias in such diseased hearts.^45,46

In conclusion, the present study suggests that Cx43 plays an important role in the occurrence of triggered arrhythmias, probably under the modulation of the mitochondrial K_ATP channels in diseased hearts.

Study Limitations

Mouse models with chronic reduction of Cx43 GJ channels^34,35 or mitochondrial Cx43 by transgenic methodology would be useful to investigate their role in arrhythmogenesis. In the present study, however, we did not use such mouse models because we could not create non-uniform muscle contraction in mouse cardiac muscle due to its smaller size. In addition, we could not find a knock-out model of Cx43 hemichannels or mitochondrial Cx43. Furthermore, we could not find drugs that have been widely recognized as selective GJ channel blockers or selective mitochondrial Cx43 blockers. Therefore, in the present study we had to use CBX, which has the ability to block not only GJ channels but also Cx43 hemichannels²⁵ and mitochondrial Cx43.²⁶

Acknowledgments

This work was supported by a Grant-in-Aid for Scientific Research (C) from the Japan Society for the Promotion of Science (M.M., No. 26460288).

Disclosures

None.

Supplementary Files

Supplementary File 1

Methods

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-15-0401

References

• 1.    Packer  M. Sudden unexpected death in patients with congestive heart failure: A second frontier. Circulation 1985; 72: 681–685.
• 2.    Smalling  RW,  Ekas  RD,  Felli  PR,  Binion  L,  Desmond  J. Reciprocal functional interaction of adjacent myocardial segments during regional ischemia: An intraventricular loading phenomenon affecting apparent regional contractile function in the intact heart. J Am Coll Cardiol 1986; 7: 1335–1346.
• 3.    Wakayama  Y,  Miura  M,  Stuyvers  BD,  Boyden  PA,  ter Keurs  HEDJ. Spatial nonuniformity of excitation-contraction coupling causes arrhythmogenic Ca²⁺ waves in rat cardiac muscle. Circ Res 2005; 96: 1266–1273.
• 4.    Sugai  Y,  Miura  M,  Hirose  M,  Wakayama  Y,  Endoh  H,  Nishio  T, et al. Contribution of Na⁺/Ca²⁺ exchange current to the formation of delayed afterdepolarizations in intact rat ventricular muscle. J Cardiovasc Pharmacol 2009; 53: 517–522.
• 5.    Remo  BF,  Qu  J,  Volpicelli  FM,  Giovannone  S,  Shin  D,  Lader  J, et al. Phosphatase-resistant gap junctions inhibit pathological remodeling and prevent arrhythmias. Circ Res 2011; 108: 1459–1466.
• 6.    Pogwizd  SM,  McKenzie  JP,  Cain  ME. Mechanisms underlying spontaneous and induced ventricular arrhythmias in patients with idiopathic dilated cardiomyopathy. Circulation 1998; 98: 2404–2414.
• 7.    Liu  N,  Rizzi  N,  Boveri  L,  Priori  SG. Ryanodine receptor and calsequestrin in arrhythmogenesis: What we have learnt from genetic diseases and transgenic mice. J Mol Cell Cardiol 2009; 46: 149–159.
• 8.    Miura  M,  Boyden  PA,  ter Keurs  HEDJ. Ca²⁺ waves during triggered propagated contractions in intact trabeculae: Determinants of the velocity of propagation. Circ Res 1999; 84: 1459–1468.
• 9.    Zhang  YM,  Miura  M,  ter Keurs  HEDJ. Triggered propagated contractions in rat cardiac trabeculae: Inhibition by octanol and heptanol. Circ Res 1996; 79: 1077–1085.
• 10.    van Kempen  MJ,  Fromaget  C,  Gros  D,  Moorman  AF,  Lamers  WH. Spatial distribution of connexin43, the major cardiac gap junction protein, in the developing and adult rat heart. Circ Res 1991; 68: 1638–1651.
• 11.    Sáez  JC,  Schalper  KA,  Retamal  MA,  Orellana  JA,  Shoji  KF,  Bennett  MV. Cell membrane permeabilization via connexin hemichannels in living and dying cells. Exp Cell Res 2010; 316: 2377–2389.
• 12.    De Vuyst  E,  Wang  N,  Decrock  E,  de Bock  M,  Vinken  M,  van Moorhem  M, et al. Ca²⁺ regulation of connexin 43 hemichannels in C6 glioma and glial cells. Cell Calcium 2009; 46: 176–187.
• 13.    Rodriguez-Sinovas  A,  Boengler  K,  Cabestrero  A,  Gres  P,  Morente  M,  Ruiz-Meana  M, et al. Translocation of connexin 43 to the inner mitochondrial membrane of cardiomyocytes through the heat shock protein 90-dependent TOM pathway and its importance for cardioprotection. Circ Res 2006; 99: 93–101.
• 14.    Ruiz-Meana  M,  Rodríguez-Sinovas  A,  Cabestrero  A,  Boengler  K,  Heusch  G,  Garcia-Dorado  D. Mitochondrial connexin43 as a new player in the pathophysiology of myocardial ischaemia-reperfusion injury. Cardiovasc Res 2008; 77: 325–333.
• 15.    Heinzel  FR,  Luo  Y,  Li  X,  Boengler  K,  Buechert  A,  García-Dorado  D, et al. Impairment of diazoxide-induced formation of reactive oxygen species and loss of cardioprotection in connexin 43 deficient mice. Circ Res 2005; 97: 583–586.
• 16.    Schwanke  U,  Konietzka  I,  Duschin  A,  Li  X,  Schulz  R,  Heusch  G. No ischemic preconditioning in heterozygous connexin43-deficient mice. Am J Physiol Heart Circ Physiol 2002; 283: H1740–H1742.
• 17.    Hausenloy  D,  Wynne  A,  Duchen  M,  Yellon  D. Transient mitochondrial permeability transition pore opening mediates preconditioning-induced protection. Circulation 2004; 109: 1714–1717.
• 18.    Chen  YR,  Zweier  JL. Cardiac mitochondria and reactive oxygen species generation. Circ Res 2014; 114: 524–537.
• 19.    Nichols  CG,  Singh  GK,  Grange  DK. K_ATP channels and cardiovascular disease: Suddenly a syndrome. Circ Res 2013; 112: 1059–1072.
• 20.    Matejíková  J,  Kucharská  J,  Pintérová  M,  Pancza  D,  Ravingerová  T. Protection against ischemia-induced ventricular arrhythmias and myocardial dysfunction conferred by preconditioning in the rat heart: Involvement of mitochondrial K_ATP channels and reactive oxygen species. Physiol Res 2009; 58: 9–19.
• 21.    de Groot  JR,  Veenstra  T,  Verkerk  AO,  Wilders  R,  Smits  JPP,  Wilms-Schopman  FJG, et al. Conduction slowing by the gap junctional uncoupler carbenoxolone. Cardiovasc Res 2003; 60: 288–297.
• 22.    Plummer  BN,  Cutler  MJ,  Wan  X,  Laurita  KR. Spontaneous calcium oscillations during diastole in the whole heart: The influence of ryanodine reception function and gap junction coupling. Am J Physiol Heart Circ Physiol 2011; 300: H1822–H1828.
• 23.    Strom  M,  Wan  X,  Poelzing  S,  Ficker  E,  Rosenbaum  DS. Gap junction heterogeneity as mechanism for electrophysiologically distinct properties across the ventricular wall. Am J Physiol Heart Circ Physiol 2010; 298: H787–H794.
• 24.    Howarth  FC,  Qureshi  MA. Effects of carbenoxolone on heart rhythm, contractility and intracellular calcium in streptozotocin-induced diabetic rat. Mol Cell Biochem 2006; 289: 21–29.
• 25.    Ye  ZC,  Wyeth  MS,  Baltan-Tekkok  S,  Ransom  BR. Functional hemichannels in astrocytes: A novel mechanism of glutamate release. J Neurosci 2003; 23: 3588–3596.
• 26.    Miro-Casas  E,  Ruiz-Meana  M,  Agullo  E,  Stahlhofen  S,  Rodríguez-Sinovas  A,  Cabestrero  A, et al. Connexin43 in cardiomyocyte mitochondria contributes to mitochondrial potassium uptake. Cardiovasc Res 2009; 83: 747–756.
• 27.    Miura  M,  Boyden  PA,  ter Keurs  HEDJ. Ca²⁺ waves during triggered propagated contractions in intact trabeculae. Am J Physiol 1998; 274: H266–H276.
• 28.    Miura  M,  Taguchi  Y,  Nagano  T,  Sasaki  M,  Handoh  T,  Shindoh  C. Effect of myofilament Ca²⁺ sensitivity on Ca²⁺ wave propagation in rat ventricular muscle. J Mol Cell Cardiol 2015; 84: 162–169.
• 29.    O’Rourke  B. Evidence for mitochondrial K⁺ channels and their role in cardioprotection. Circ Res 2004; 94: 420–432.
• 30.    Lyon  AR,  Joudrey  PJ,  Jin  D,  Nass  RD,  Aon  MA,  O’Rourke  B, et al. Optical imaging of mitochondrial function uncovers actively propagating waves of mitochondrial membrane potential collapse across intact heart. J Mol Cell Cardiol 2010; 49: 565–575.
• 31.    Stuyvers  BD,  Dun  W,  Matkovich  S,  Sorrentino  V,  Boyden  PA,  ter Keurs  HEDJ. Ca²⁺ sparks and waves in canine Purkinje cells: A triple layered system of Ca²⁺ activation. Circ Res 2005; 97: 35–43.
• 32.    Kojodjojo  P,  Kanagaratnam  P,  Segal  OR,  Hussain  W,  Peters  NS. The effects of carbenoxolone on human myocardial conduction: A tool to investigate the role of gap junctional uncoupling in human arrhythmogenesis. J Am Coll Cardiol 2006; 48: 1242–1249.
• 33.    Imanaga  I,  Kameyama  M,  Irisawa  H. Cell-to-cell diffusion of fluorescent dyes in paired ventricular cells. Am J Physiol 1987; 252: H223–H232.
• 34.    Danik  SB,  Liu  F,  Zhang  J,  Suk  HJ,  Morley  GE,  Fishman  GI, et al. Modulation of cardiac gap junction expression and arrhythmic susceptibility. Circ Res 2004; 95: 1035–1041.
• 35.    Guerrero  PA,  Schuessler  RB,  Davis  LM,  Beyer  EC,  Johnson  CM,  Yamada  KA, et al. Slow ventricular conduction in mice heterozygous for a connexin43 null mutation. J Clin Invest 1997; 99: 1991–1998.
• 36.    Szente  M,  Gajda  Z,  Said Ali  K,  Hermesz  E. Involvement of electrical coupling in the in vivo ictal epileptiform activity induced by 4-aminopyridine in the neocortex. Neuroscience 2002; 115: 1067–1078.
• 37.    Morley  GE,  Vaidya  D,  Samie  FH,  Lo  C,  Delmar  M,  Jalife  J. Characterization of conduction in the ventricles of normal and heterozygous Cx43 knockout mice using optical mapping. J Cardiovasc Electrophysiol 1999; 10: 1361–1375.
• 38.    Garlid  KD,  Paucek  P,  Yarov-Yarovoy  V,  Sun  X,  Schindler  PA. The mitochondrial KATP channel as a receptor for potassium channel openers. J Biol Chem 1996; 271: 8796–8799.
• 39.    Coetzee  WA. Multiplicity of effectors of the cardioprotective agent, diazoxide. Pharmacol Ther 2013; 140: 167–175.
• 40.    Schulz  R,  Görge  PM,  Görbe  A,  Ferdinandy  P,  Lampe  PD,  Leybaert  L. Connexin 43 is an emerging therapeutic target in ischemia/reperfusion injury, cardioprotection and neuroprotection. Pharmacol Ther 2015; 153: 90–106.
• 41.    Akar  FG,  Aon  MA,  Tomaselli  GF,  O’Rourke  B. The mitochondrial origin of postischemic arrhythmias. J Clin Invest 2005; 115: 3527–3535.
• 42.    Liew  R,  Khairunnisa  K,  Gu  Y,  Tee  N,  Yin  NO,  Naylynn  TM, et al. Role of tumor necrosis factor-α in the pathogenesis of atrial fibrosis and development of an arrhythmogenic substrate. Circ J 2013; 77: 1171–1179.
• 43.    Kurokawa  S,  Niwano  S,  Niwano  H,  Murakami  M,  Ishikawa  S,  Masaki  Y, et al. Cardiomyocyte-derived mitochondrial superoxide causes myocardial electrical remodeling by downregulating potassium channels and related molecules. Circ J 2014; 78: 1950–1959.
• 44.    Miura  M,  Nishio  T,  Hattori  T,  Murai  N,  Stuyvers  BD,  Shindoh  C, et al. Effect of nonuniform muscle contraction on sustainability and frequency of triggered arrhythmias in rat cardiac muscle. Circulation 2010; 121: 2711–2717.
• 45.    Park  H,  Park  H,  Lee  D,  Oh  S,  Lim  J,  Hwang  HJ, et al. Increased phosphorylation of Ca²⁺ handling proteins as a proarrhythmic mechanism in myocarditis. Circ J 2014; 78: 2292–2301.
• 46.    Wen  JY,  Wei  CY,  Shah  K,  Wong  J,  Wang  C,  Chen  HSV. Maturation-based model of arrhythmogenic right ventricular dysplasia using patient-specific induced pluripotent stem cells. Circ J 2015; 79: 1402–1408.