Skip to main content
Advertisement
  • Loading metrics

Mitochondrial depolarization promotes calcium alternans: Mechanistic insights from a ventricular myocyte model

  • Vikas Pandey,

    Roles Data curation, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing

    Affiliation Department of Medicine, David Geffen School of Medicine, University of California, Los Angeles, California, United States of America

  • Lai-Hua Xie,

    Roles Formal analysis, Funding acquisition, Investigation, Writing – review & editing

    Affiliation Department of Cell Biology and Molecular Medicine, Rutgers, New Jersey Medical School, Newark, New Jersey, United States of America

  • Zhilin Qu,

    Roles Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Writing – review & editing

    Affiliations Department of Medicine, David Geffen School of Medicine, University of California, Los Angeles, California, United States of America, Department of Computational Medicine, David Geffen School of Medicine, University of California, Los Angeles, California, United States of America

  • Zhen Song

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

    songzh01@pcl.ac.cn

    Affiliations Department of Medicine, David Geffen School of Medicine, University of California, Los Angeles, California, United States of America, Peng Cheng Laboratory, Shenzhen, Guangdong, China

Abstract

Mitochondria are vital organelles inside the cell and contribute to intracellular calcium (Ca2+) dynamics directly and indirectly via calcium exchange, ATP generation, and production of reactive oxygen species (ROS). Arrhythmogenic Ca2+ alternans in cardiac myocytes has been observed in experiments under abnormal mitochondrial depolarization. However, complex signaling pathways and Ca2+ cycling between mitochondria and cytosol make it difficult in experiments to reveal the underlying mechanisms of Ca2+ alternans under abnormal mitochondrial depolarization. In this study, we use a newly developed spatiotemporal ventricular myocyte computer model that integrates mitochondrial Ca2+ cycling and complex signaling pathways to investigate the mechanisms of Ca2+ alternans during mitochondrial depolarization. We find that elevation of ROS in response to mitochondrial depolarization plays a critical role in promoting Ca2+ alternans. Further examination reveals that the redox effect of ROS on ryanodine receptors and sarco/endoplasmic reticulum Ca2+-ATPase synergistically promote alternans. Upregulation of mitochondrial Ca2+ uniporter promotes Ca2+ alternans via Ca2+-dependent mitochondrial permeability transition pore opening. Due to their relatively slow kinetics, oxidized Ca2+/calmodulin-dependent protein kinase II activation and ATP do not play significant roles acutely in the genesis of Ca2+ alternans after mitochondrial depolarization, but their roles can be significant in the long term, mainly through their effects on sarco/endoplasmic reticulum Ca2+-ATPase activity. In conclusion, mitochondrial depolarization promotes Ca2+ alternans acutely via the redox effect of ROS and chronically by ATP reduction. It suppresses Ca2+ alternans chronically through oxidized Ca2+/calmodulin-dependent protein kinase II activation.

Author summary

Mitochondrial dysfunction has been implicated in life-threatening cardiac arrhythmias. Experimentalists have shown that mitochondrial depolarization could cause period-doubling behavior in cytosolic Ca2+ transient (i.e., Ca2+ alternans). However, the underlying mechanisms behind these experimental observations remain unclear. Here, we focus on revealing the underlying mechanisms of cellular Ca2+ alternans in the context of mitochondrial depolarization. Using our newly developed action potential model that incorporates a 3D network of Ca2+ release unit with a 3D network of mitochondria, we dissect out individual roles of mitochondrial Ca2+ and complex signaling pathways (e.g., reactive oxygen species, oxidized Ca2+/calmodulin-dependent protein kinase II activation, and ATP) in the genesis of cellular Ca2+ alternans. The key finding is that mitochondrial depolarization promotes Ca2+ alternans acutely via the redox effect of ROS and chronically by ATP reduction, and oxidized Ca2+/calmodulin-dependent protein kinase II activation may suppress Ca2+ alternans chronically. We believe that this study reveals insightful information for understanding the mechanisms of the genesis of Ca2+ alternans in the context of mitochondrial depolarization and provides valuable simulation results to guide future experiments.

Introduction

Calcium (Ca2+) is a critical regulator of excitation-contraction coupling in cardiac myocytes [1]. Ca2+ enters a cardiac myocyte mostly through the L-type Ca2+ channels (LCCs), which open in response to membrane depolarization. The resulting rise in the intracellular Ca2+ concentration activates ryanodine receptors (RyRs) to release a large amount of Ca2+ from the sarcoplasmic reticulum (SR), an internal Ca2+ store in cardiac myocytes. RyRs are clustered on the membrane of SR, forming discrete Ca2+ release units (CRUs). The SR Ca2+ release causes a transient increase in cytosolic Ca2+, which binds to myofilament to cause contraction. Ca2+ in the cytosol is extruded outside the cell via Na+-Ca2+ exchangers (NCX) and transported back to the SR through the sarco/endoplasmic reticulum Ca2+-ATPase (SERCA) pump. The normal Ca2+ cycling is essential to maintain the normal function of excitation-contraction coupling due to the bi-directional coupling through nonlinear interactions of ion channels, transporters, and pumps. Disturbances in the complex regulations of these components of the cell can lead to various nonlinear dynamics that underlie life-threatening cardiac arrhythmias [2]. Ca2+ alternans is one of the most studied phenomena for cardiac arrhythmias. Ca2+ alternans can cause action potential duration (APD) alternans due to the bi-directional coupling of the membrane voltage and Ca2+ [2,3]. Ca2+ and APD alternans are known to occur in acute myocardial ischemia and other diseased conditions [49]. APD alternans gives rise to T-wave alternans, which is a risk marker for sudden cardiac death [10,11]. Besides SR, mitochondria are also internal Ca2+ stores, and Ca2+ enters into and moves out of the mitochondria during a cardiac cycle, directly affecting intracellular Ca2+ signaling [8,1217]. Moreover, mitochondria produce ATP and reactive oxygen species (ROS), which affect Ca2+ dynamics via their effects on the Ca2+ cycling proteins. Recent studies have shown that mitochondrial Ca2+ cycling plays a key role in cardiac diseases [1822]. The goal of this work is to investigate mitochondrial contributions to the genesis of Ca2+ alternans.

Experimental studies have shown that metabolic stress or impaired mitochondrial function promotes Ca2+ alternans [14,2326]. However, the complex effects of mitochondria on Ca2+ cycling, energy metabolism, and signaling make it difficult to dissect out the underlying mechanisms in experiments. Ca2+ enters mitochondria via the mitochondrial Ca2+ uniporter (MCU) [27,28], and exits mitochondria via mitochondrial sodium-calcium exchanger (mNCX) [29] or the mitochondrial permeability transition pore (mPTP). The mPTP open probability is very low under normal conditions but can be high under pathophysiological conditions. In addition to the direct mitochondrial Ca2+ cycling, mitochondria are coupled with intracellular Ca2+ cycling in several other ways. ROS signaling and the redox regulation affect the opening of RyRs and the activity of the SERCA pump [30,31]. Mitochondria provide ATP to the SERCA pump and other sarcolemmal ion pumps, and a shortage of ATP may impair the function of these pumps [32,33]. Moreover, there is a positive feedback loop between mitochondrial ROS production and SR Ca2+ release, i.e., leaky RyRs trigger more Ca2+ release resulting in more Ca2+ sequestered in mitochondria, which can trigger the opening of mPTP, leading to mitochondria depolarizations [34,35]. Mitochondrial depolarization produces ROS, which regulates SR Ca2+ uptake and release. The oxidized Ca2+/calmodulin-dependent protein kinase II (CaMKII) activation enhances the sensitivity of the SERCA pump through phosphorylation of phospholamban [36]. Concurrently, CaMKII also affects the activation of LCCs [37].

Therefore, mitochondria play complex roles in intracellular Ca2+ cycling via different pathways. These pathways can either promote or suppress alternans depending on their properties [38,39], making it very difficult to experimentally dissect out the key pathways responsible for alternans. We recently developed a physiologically detailed ventricular myocyte model that incorporated mitochondrial Ca2+ cycling, mPTP, ROS production, and oxidized CaMKII signaling [40]. We showed that this model could exhibit Ca2+ alternans and spontaneous Ca2+ release mediated delayed afterdepolarizations (DADs) under mitochondrial depolarization. In the present study, we use computer simulations of this model to investigate the underlying mechanisms for the genesis of Ca2+ alternans in response to mitochondrial depolarization induced by mPTP opening. Taking advantage of computer simulations, we can differentiate the effects of direct mitochondrial Ca2+ cycling, the redox regulation of RyRs and SERCA by ROS, oxidized CaMKII signaling, and ATP reduction on Ca2+ alternans. We demonstrate that mitochondrial depolarization promotes Ca2+ alternans acutely via the redox effect of ROS and chronically by ATP reduction, and may suppress Ca2+ alternans chronically by oxidized signaling, mainly via their effects on SERCA activity.

Results

Ca2+ alternans caused by mitochondrial depolarization due to mPTP opening

To show the effects of mPTP opening on Ca2+ alternans, we carried out simulations under the control condition (where the mPTP open probability is very low) and a high mPTP open probability condition (by increasing the transition rate from the closed state, C1, to the open state, O, to 60-fold of the control value, i.e., αmPTP = 60) for PCL = 500 ms (Fig 1A). Under the control condition (black lines), no alternans occurred. Under the condition of high mPTP open probability (red lines), Ca2+ alternans occurred. And the Ca2+ alternans was abolished when all the mPTPs within the cell were commanded to be in the closed state, which also recovered Δψ (S1 Fig). These observations were consistent with our previous experimental results using either mPTP inhibitor or cyclophilin D knockout mouse model [41] in which mitochondrial depolarization was prevented. Note that the corresponding APD alternans appears small (see the enlarged one in S2A Fig), since the Ca2+ alternans amplitude is small (~0.2 μM) at PCL = 500 ms. At a faster pacing rate (PCL = 300 ms), the APD alternans becomes more significant due to greater Ca2+ alternans amplitude (S2B Fig). Note that for this αmPTP value, the open probability of mPTP was about 30%. Also, note that at t = 30 s CaMKII activity and cytosolic ATP are still changing slowly. When we ran the simulations for a much longer time (e.g., 1000 s), the CaMKII activity and the cytosolic ATP became ~74% and ~2 mM, respectively. Since these two variables change extremely slowly over time, we treated them as quasi-steady state variables for the half-minute long simulation. For the same reason, we investigated the acute effect of mitochondrial depolarization on the genesis of Ca2+alternans in this study by performing free-running simulations for 30 s. Fig 1B shows a bifurcation diagram plotting the peak values of cytosolic Ca2+ transient against the pacing cycle length (PCL) for both conditions. The high mPTP open probability condition changed the onset of Ca2+ alternans to a longer PCL, from 450 ms to 550 ms.

thumbnail
Fig 1. Mitochondrial depolarization promotes Ca2+ alternans.

A. Time traces of membrane potential V, whole-cell averaged cytosolic Ca2+ concentration , SR Ca2+concentration , SERCA uptake flux, SR Ca2+ release flux via RyRs, mitochondrial free Ca2+ , open probability of mPTP , mitochondrial membrane potential , cytosolic ROS concentration , cytosolic CaMKII activation , cytosolic ATP concentration for normal control (αmPTP = 1) in black and mitochondrial depolarization (αmPTP = 60) conditions in red, respectively. PCL is 500 ms. Note that the horizontal bar above each variable means that the quantity is an averaged value over all the CRUs or mitochondria within the myocyte. B. Bifurcation diagrams of peak values of in the last two consecutive beats vs. PCL for control (black) and mitochondrial-depolarization (red) cases.

https://doi.org/10.1371/journal.pcbi.1008624.g001

Since mPTP opening and mitochondrial depolarization can affect the intracellular Ca2+ dynamics directly or indirectly via changing the properties of Ca2+ handling proteins (e.g., RyRs and SERCA) or altering CaMKII and other signaling pathways [30,31], it becomes difficult to reveal the roles of each process experimentally. In the following sections, we take advantage of computer simulation to examine the individual roles of cytosolic ROS, mitochondrial Ca2+, CaMKII activation, and cytosolic ATP in the genesis of Ca2+ alternans.

Effects of ROS on the genesis of Ca2+ alternans

To determine the effect of ROS on Ca2+ alternans, we carried out simulations in the conditions of the free-running ROS (i.e., the ROS dynamics obeys the differential equations) and a clamped ROS (i.e., the ROS was fixed to a constant) at PCL = 500 ms, shown in Fig 2A as bifurcation diagrams. Under the free-running case, increasing αmPTP above certain threshold promotes alternans, but when ROS was clamped at a low level (0.1 μM), no alternans occurs. Therefore, our simulation results here suggest that Ca2+ alternans was predominantly mediated by the cytosolic ROS. Also, note that increasing αmPTP increases the open probability of mPTP, and under the setting of the free-running ROS, we found that Ca2+ alternans became obvious when the open probability of mPTP is above ~20% (S3 Fig).

thumbnail
Fig 2. Effect of cytosolic ROS signaling on the genesis of Ca2+ alternans.

A. Bifurcation diagram of peak values of Ca2+ transients vs. αmPTP for free-running ROS (black) and fixed ROS = 0.1 μM (red), respectively. B. Bifurcation diagram of peak values of Ca2+ transient vs. αmPTP with free-running ROS, but the redox effect of ROS only exerted on RyRs (green) or SERCA (blue), respectively. The PCL is 500 ms. The total simulation time for each simulation is 30 sec.

https://doi.org/10.1371/journal.pcbi.1008624.g002

To further dissect out how ROS promotes Ca2+ alternans, we first investigate its redox effect on RyRs and SERCA separately. Fig 2B shows the bifurcation diagrams for the redox effect of ROS on RyRs only (green, by setting fup,ros = 1 in Eq 5) or for the redox effect of ROS on SERCA only (blue, by setting Δkros = 0 in Eq 2). We then performed the same simulations as in Fig 2A for these two cases (Fig 2B). When the redox effect on SERCA was removed, the amplitude of Ca2+ alternans was greatly reduced (Fig 2B, green). When the redox effect on RyRs was removed, the amplitude of Ca2+ alternans was also decreased (Fig 2B, blue) compared to the case of free-running ROS (Fig 2A, black). Note that the onsets of alternans for these two cases are about the same as in the case of free-running ROS (Fig 2A, black). These results suggest that the redox effect of ROS on RyRs and SERCA synergistically promotes Ca2+ alternans.

Effects of mitochondrial Ca2+ on the genesis of Ca2+ alternans

Under certain pathological conditions, such as nonischemic cardiomyopathy, MCU has been reported to dramatically upregulated [17], which may markedly increase the mitochondrial free Ca2+. Although our previous simulation study [40] has shown that the release of mitochondrial Ca2+ to the cytosol may only transiently affect the cytosolic Ca2+, the elevation of mitochondrial free Ca2+ is believed to promote the opening of mPTP and thus the production of ROS. Here we performed simulations to investigate how the MCU activity affects the genesis of Ca2+ alternans. For simplicity, we multiplied a pre-factor, αMCU, to the maximal MCU conductance. Therefore, αMCU = 1 represents the control MCU conductance and increasing αMCU increases the maximal MCU activity. Fig 3 shows the dependence of Ca2+ alternans amplitude on αMCU and αmPTP. When the close-to-open rate of mPTP is low (αmPTP<20), no Ca2+ alternans occurs even if the maximal MCU activity is increased to 50-fold (αMCU = 50). However, as mPTP open probability increases (i.e., as αmPTP increases), the value of αMCU required to generate Ca2+ alternans decreases. Furthermore, when the close-to-open rate of mPTP further increases (αmPTP>80), the effect of αMCU on the genesis of Ca2+ alternans becomes less important. In addition, we found that inhibition of either MCU or mNCX in the setting of these simulations was unable to abolish Ca2+ alternans (S4 Fig). In conclusion, these results suggest that MCU upregulation may play an important role in generating Ca2+ alternans through mitochondrial Ca2+ mediated mPTP opening under certain pathological conditions.

thumbnail
Fig 3. MCU upregulation promotes Ca2+ alternans through Ca2+- dependent opening of mPTP.

Dependence of Ca2+ alternan amplitude on αMCU and αmPTP. is calculated as the difference between the last two Ca2+ transient peaks in a simulation of 30 sec. PCL = 500 ms.

https://doi.org/10.1371/journal.pcbi.1008624.g003

Effects of oxidized CaMKII signaling on Ca2+ alternans

Since mitochondrial depolarization activates oxidized CaMKII signaling via ROS, we then evaluate the importance of CaMKII activation in the genesis of Ca2+ alternans during mitochondrial depolarization. As discussed earlier, CaMKII activation dynamics is a very slow process in the model and thus it can be treated as a quasi-steady state variable. Therefore, we clamped the CaMKII activation at different levels and examined the corresponding effects on the genesis of Ca2+ alternans. Fig 4 shows the bifurcation diagrams of the peak values of cytosolic Ca2+ transient against αmPTP for CaMKII activity clamped at 1%, 10%, and 30% levels. These results show that Ca2+ alternans is suppressed as the CaMKII activation level increases. However, as shown in Fig 1A, the CaMKII activation level is about 1% for both control and the high mPTP open probability conditions within the 30 sec total simulation time. Therefore, our simulation results suggest that for the acute effect of mitochondrial depolarization on the genesis of Ca2+ alternans, CaMKII may not exhibit a big effect. However, for the long-term effect, CaMKII may play a more significant role, suppressing Ca2+ alternans via its regulation of SERCA.

thumbnail
Fig 4. Effect of CaMKII activation on the genesis of the Ca2+ alternans.

Bifurcation diagrams of peak values of Ca2+ transients in the last two beats vs. αmPTP for clamped CaMKII activation level 1% (A), 10% (B) and 30% (C), respectively. PCL = 500 ms.

https://doi.org/10.1371/journal.pcbi.1008624.g004

Effects of ATP depletion on Ca2+ alternans due to mitochondrial depolarization

When a mitochondrion depolarizes, it stops producing ATP, and as more mitochondria depolarize in the cell due to the increased mPTP open probability, the cytosolic ATP decreases. Since ATP is required for the SERCA pump, a low ATP level could impair the SERCA activity to promote Ca2+ alternans. As shown in Fig 1A, in our model, the cytosolic ATP level decays very slowly during mitochondrial depolarization. Similar to what we did for CaMKII, we clamped the cytosolic ATP concentration to different values and examined the consequences of ATP depletion during mitochondrial depolarization on the genesis of Ca2+ alternans.

Fig 5 shows the bifurcation diagram of the peak values of Ca2+ transient vs. the clamped cytosolic ATP concentration. The result shows that reducing the cytosolic ATP concentration promotes Ca2+ alternans during mitochondrial depolarization (αmPTP = 30). This result suggests that the cytosolic ATP indeed has a great impact on the genesis of Ca2+ alternans. However, considering its slow decay during the process of mitochondrial depolarization, ATP may not play a central role in inducing Ca2+ alternans at least at the scale of sub-minute evolution. Therefore, similar to CaMKII, during mitochondrial depolarization, ATP may not exhibit a significant effect on the genesis of Ca2+ alternans acutely, but may promote alternans in a much longer time scale when the cytosolic ATP level reduces dramatically to impair the SERCA activity.

thumbnail
Fig 5. Effect of ATP on the genesis of Ca2+ alternans.

Bifurcation diagrams of peak values of Ca2+ transients in the last two beats vs. clamped whole-cell average ATP level in the cytosol. αmPTP = 30. The PCL = 500 ms.

https://doi.org/10.1371/journal.pcbi.1008624.g005

Discussion

In this study, we used a physiologically-detailed ventricular myocyte model consisting of a 3D network of coupled CRUs and mitochondria to investigate the roles of mitochondrial depolarization via mPTP opening in the genesis of Ca2+ alternans. We showed that the effects of mitochondrial depolarization on Ca2+ alternans mainly via the ROS induced redox regulation of SERCA and RyRs. ROS induced oxidized CaMKII signaling and ATP decay exhibit little effect acutely but can exhibit large effects chronically via their regulations of SERCA. Mitochondrial Ca2+ cycling alone exhibits little effect on Ca2+ alternans but can exhibit a large effect via the Ca2+-dependent opening of mPTP.

Role of ROS in the genesis of Ca2+ alternans

We have shown that an increase in the cytosolic ROS concentration in response to mitochondrial depolarization plays a key role in generating Ca2+ alternans. By judicious manipulation of the redox effects of ROS on RyRs and SERCA, we have been able to demonstrate that there is a synergy effect of the ROS redox regulation on both SERCA and RyRs in generating Ca2+ alternans. Our results show that only enhancing the RyR activity (Fig 2B, green) or reducing SERCA pump ability (Fig 2B, blue) via the ROS redox regulation causes a much smaller amplitude of Ca2+ alternans compared to the case where the ROS redox regulation effect is turned on for both RyRs and SERCA (Fig 2A, black). A previous study by Belevych et al. [23] has shown that an increase in ROS increases the RyR open probability, producing leaky RyR channels, and in turn resulting in Ca2+ alternans. Our findings here extend those of Belevych et al. [23], suggesting that the ROS redox regulation on SERCA is also crucial for promoting Ca2+ alternans. In fact, these findings well agree with the unified alternans theory developed in our previous study [42], since either hyperactivating RyRs or reducing the SERCA pump ability contributes to the increased steepness of the SR release-load relationship in cardiac myocytes, and therefore it is not surprising that the combined effect of the two factors synergistically promotes Ca2+ alternans.

Besides the redox effect, ROS is known to activate CaMKII, PKA and PKC pathways [30,31], which could affect the genesis of Ca2+ alternans. The model used in this study takes into account the effect of ROS on the activation of CaMKII by incorporating the oxidized CaMKII signaling formulation developed by Foteinou et al. [37]. The effect of CaMKII activation on Ca2+ handling proteins and ion channels has been simulated here following Hund and Rudy [36]. However, the kinetics of CaMKII appears to be very slow as seen in Fig 1A, since the percentage of CaMKII activation merely changes from 0.4% to 1.2%, whereas the mPTP open probability increases from ~0% to ~32% in response to the mitochondrial depolarization. These results therefore indicate that in the case of acute and severe metabolic insults, CaMKII activation may not be the cause of the genesis of Ca2+ alternans. In fact, our simulations of clamping CaMKII activation to different levels even suggest that higher CaMKII activation tends to suppress Ca2+ alternans (Fig 4) due to CaMKII phosphorylation of phospholamban, which reduces the half-maximal value of SERCA to increase SERCA pump activity. This agrees with our previous theoretical prediction that increasing SERCA pump activity can move the cell system out of the alternans regime [42]. However, that does not mean oxidized CaMKII activation during mitochondrial depolarization is beneficial, since CaMKII activation enhances LCC, RyRs and SERCA, which are known factors promoting arrhythmogenic Ca2+ waves, DADs and EADs [4345].

Clinically, ischemia/reperfusion injury of myocardium is associated with both mitochondrial depolarization and repolarization phases. During the repolarization phase, Ca2+ homeostasis is disrupted, which is believed via opening of mPTP because of high ROS and mitochondrial Ca2+ accumulation [4648]. However, some experiment showed that sustained depolarization of mitochondrial membrane potential did not occur even after 10 min of reperfusion [48], which requires more comprehensive future studies to explain. Nevertheless, our simulations suggest that ROS is the main factor for mitochondrial depolarization induced Ca2+ alternans, and targeting mPTP or ROS may prevent Ca2+ alternans in acute myocardial ischemia.

Mitochondrial Ca2+ cycling and intracellular Ca2+ alternans

We have shown that MCU upregulation combined with the increased C1-to-O transition rate of mPTP gating kinetics promotes Ca2+ alternans (Fig 3). In our previous study [40], we found that mitochondrial Ca2+ released into the cytosol due to severe mitochondrial depolarization (100%) can only transiently change the cytosolic Ca2+ dynamics, but not the steady state. Therefore, our findings here indicate that mitochondrial Ca2+ may not directly affect the intracellular Ca2+ homeostasis, but it can enhance mPTP opening, which in turn causes Ca2+ alternans via ROS induced signaling pathways or lowering ATP. However, in nonischemic cardiomyopathy, MCU upregulation, together with other electrophysiological remodeling changes in heart failure conditions, can result in an all-or-none behavior or bistability, corresponding to the no EAD and EAD states in AP without mitochondrial depolarization [17]. Furthermore, the Ca2+ alternans in our simulations is driven by the cytosolic Ca2+, not the mitochondrial Ca2+. As shown in S5 Fig, when the mitochondrial Ca2+ was clamped to fixed values, the cytosolic Ca2+ alternans still exists (S5A Fig), but when cytosolic Ca2+ was clamped to fixed values, the mitochondrial Ca2+ alternans disappeared (S5B Fig).

ATP level and Ca2+ alternans

Previous experimental studies have documented that a reduced cellular ATP level is linked to Ca2+ alternans[14,24,25], presumably via the reduced SERCA pump activity. Here, we have shown that the ATP level during acute mitochondrial depolarization (30%) only changes slightly in our simulation (Fig 1A). If that is the case in experiment, then it is unlikely that the Ca2+ alternans occurring during acute mitochondrial depolarization is associated with the cellular ATP change. However, in experiment, whether the cellular ATP level is reduced has been shown to depend on the concentration of FCCP and treatment time [49]. Because the newly developed experimental technology can now be used to measure the ATP level in living functioning cells [50,51], future work should therefore include the fine-tuning of the ATP computer model to match the dynamics of ATP in individual experiments. Here, in order to evaluate the role of the ATP level in the genesis of Ca2+ alternans, we have performed simulations to clamp the cytosolic ATP concentration at different levels during mitochondrial depolarization. We show that when ATP is low enough, it can promote alternans via its effect on SERCA. However, it could work in synergy with the redox effect of ROS on SERCA even at much higher levels. Note that the effect of ATP on the genesis of Ca2+ alternans under mitochondrial depolarization is also mPTP dependent, since different mPTP open probabilities would cause different levels of cytosolic ROS, thereby affecting the cytosolic Ca2+ dynamical regime, which is on top of the ATP effect on Ca2+ alternans (S6 Fig). Also, the cytosolic ATP concentration itself depends on the mPTP activity; the opening of mPTP depolarizes the mitochondrial membrane potential, malfunctioning the ATP synthases.

Limitations

Several limitations should be noted in this study. The AP model and the 3-dimensional CRU network used in this study can successfully simulate the basic excitation-contraction-metabolism coupling in a ventricular myocyte, but it cannot capture all the aspects of electrophysiology of a real myocyte, such as heterogeneities in T-tubule networks and distribution of ion channels and Ca2+ handling proteins [52,53]. Such heterogeneities may alter the propensity of a cardiac myocyte for alternans [9]. Moreover, in this study the opening of mPTP was only mitochondrial free Ca2+ dependent, and the open probability was increased by increasing the C1 to O transition rate constant. However, mPTP could open via a ROS-induced ROS release mechanism [54,55]. This mechanism is important for modeling mitochondrial depolarization waves [5658], which are not the focus of this study. Furthermore, mitochondria have been found to constantly divide and fuse in cardiac myocytes, and in heart failure conditions, mitochondrial fusion could be depressed, which may contribute to the genesis of cardiac arrhythmias [59]. Therefore, further advanced computer models should be developed in the future to incorporate the feature of mitochondrial fusion and fission. Lastly, mitochondria have a special BK channels (mBKs), which is a voltage-dependent and Ca2+-activated K+ channel with a conductance ~100–300 pS. It has been known that the opening of mBK brings in K+ into the mitochondrion, depolarizing Δψ. It then reduces the driving force of MCU, which in turn attenuates the overload of mitochondrial Ca2+ (Stowe et al. [60]). The cardioprotective role of the mBK channel has been proposed to be similar to the mKATP channel (Stowe et al. [60]). The opening of these channels may prevent mPTP from opening by reducing mitochondrial Ca2+ overload. Therefore, in future studies, models of mBK and mKATP channels will be developed and incorporated in the mitochondrion model.

Methods

The details of this model, including mathematical formulation, values of parameters, and experimental validation, can be found in Song et al. [40]. Here we describe some of the key components of the model that are important for this study.

The overall ventricular myocyte model structure

This rabbit ventricular myocyte model contains a 3-dimensional coupled network of CRUs and mitochondria. There are 21504 (64×28×12) CRUs and 5376 (64×14×6) mitochondria (see Song et al. [40] for the details on the arrangement of these networks.).The membrane potential (V) of the cell is described by (1) where Cm = 1 μF/cm2 is cell membrane capacitance, and Isti is the stimulus pulse with the current density being -80 μA/cm2 and the duration being 0.5 ms. The formulations of the ionic currents are referred to Song et al. [40].

The Gilespie method was used to simulate the random transitions of LCCs, RyRs, and mPTPs. The Euler method was used to solve the differential equations, and an adaptive time step method was used to compute the AP upstroke [61] with a time step 0.001 ms. The time step for computation for the rest of the AP was 0.01 ms. The computer model was programmed in CUDA C++ with double precision on Nvidia Tesla K20c and K80 GPU cards.

ROS and CaMKII regulation of RyRs

Both oxidized CaMKII signaling and the redox regulation of ROS increase the RyRs open probability [30,31,6264]. To model these effects, we formulated the close-to-open rate (k12) of RyRs as follows: (2) where ΔkCaMKII and ΔkROS are the CaMKII-dependent and ROS-dependent components, respectively. The equations of ΔkCaMKII and ΔkROS are formulated in Song et al. [40].

(3)(4)

In Fig 2B, when we examined the effect of ROS via SERCA alone on inducing Ca2+ alternans, we set ΔkROS = 0 to remove the effect of ROS on RyR activity. Other than that, ΔkROS was calculated using Eq 4. kbase and ku are rate constants. [Ca2+]p is the Ca2+ concentration in the dyadic space. k12 represents a closed-to-open rate of the RyR model [40] incorporated in the cell model. Increasing k12 increases the open probability of RyRs.

ROS and CaMKII regulation of SERCA pump

Direct redox regulation slows the SERCA pump activity [31], and CaMKII phosphorylation of the phospholamban reduces the half maximum value [36]. Hence, the CaMKII effect on the activity of SERCA competes with the ROS effect on the SERCA activity. SERCA activity is also influenced by the ATP [65], i.e., reducing ATP impairs the SERCA pump function. Thus, the formulation of SERCA activity incorporating the ATP, CaMKII and ROS dependency is as follows: (5) where fup,ATP, and fup,ROS are ATP and ROS-dependent functions, which are detailed in Song et al. [40].

(6)(7)

Eq 6 was adopted from Cortassa et al. [65]. Eq 7 was formulated in our previous study [40] to account for the redox effect of ROS on SERCA. In Fig 2B, when we examined the effect of ROS via RyR on inducing the Ca2+ alternans, we set fup,ROS = 1 to eliminate the redox effect of ROS on SERCA in our model. Other than that, fup,ROS was calculated using Eq 7 in this study. vup and ki is the maximum SERCA strength and the half maximum value, respectively. PLB([CaMKII]act) is a CaMKII dependent function affecting the half maximum value of SERCA, where [CaMKII]act is the local CaMKII activation in the cytosol.

The mPTP model

A 3-state (two close states C0 and C1, and an open state O) model of the mPTP (S7 Fig) was used. The transition from the C0 state to the C1 state is mitochondrial free Ca2+ dependent as shown below: (8)

Where hmPTP is the Hill coefficient, [Ca2+]m is the mitochondrial free Ca2+ in the corresponding mitochondrion, and [Ca2+]0 is the half maximum value. Other transition rates are assumed to be constant. To simulate different levels of mPTP open probability, we multiplied a pre-factor, αmPTP, to the C1 to O transition rate, , (9)

The relationship between αmPTP and the steady state open probability of mPTP is the following, (10)

Eq 10 suggests that PmPTP~0 when αmPTP~0, and PmPTP~1 when αmPTP~ ∞. Therefore, by simply increasing αmPTP, we were able to increase the level of the open probability of mPTP.

Supporting information

S1 Fig.

From top to bottom, time traces of mitochondrial membrane potential, whole-cell averaged cytosolic ROS level, whole-cell averaged cytosolic Ca2+ concentration. The red bar indicates the time period when all the mPTPs in the cell were commanded to be closed.

https://doi.org/10.1371/journal.pcbi.1008624.s001

(TIF)

S2 Fig.

A. Time traces of voltage and Ca for control and mitochondrial depolarization cases at PCL = 500 ms. B. Time traces of voltage and Ca for mitochondrial depolarization case at PCL = 300 ms. Note that we plot two consecutive beats on top of each other to better observe the alternans.

https://doi.org/10.1371/journal.pcbi.1008624.s002

(TIF)

S3 Fig.

Bifurcation diagram of peak values of Ca2+ transients vs. the whole-cell averaged open probability of mPTP. The open probability of mPTP was measured at the end of each simulation of Fig 2A. We can see that the threshold for mPTP open probability is ~20% in this model. However, this specific value of the threshold depends on many factors, such as what we discussed in the Limitation section.

https://doi.org/10.1371/journal.pcbi.1008624.s003

(TIF)

S4 Fig.

A. Time traces of whole-cell averaged cytosolic Ca2+ transient and mitochondrial Ca2+ for αMCU = 1 (red) and 0.5 (black). αmPTP = 60. B. Same as A, but mitochondrial NCX was at control (red) and 50% reduction (black). αmPTP = 60, PCL = 500 ms.

https://doi.org/10.1371/journal.pcbi.1008624.s004

(TIF)

S5 Fig.

A. Time traces of the whole-cell average cytosolic Ca2+, mitochondrial Ca2+, where the mitochondrial Ca2+ was clamped to 0.4 (black), 0.5 (red), and 0.6 (blue) μM at t = 25 sec. B. Same as A, but the cytosolic Ca2+ was clamped to 0.2 (black), 0.5 (red), and 1 (blue) μM at t = 25 sec.

https://doi.org/10.1371/journal.pcbi.1008624.s005

(TIF)

S6 Fig.

Dependence of Ca2+ alternans amplitude on αmPTP and was calculated as the difference between the last two Ca2+ transient peaks in a simulation of 30 sec. PCL = 500 ms.

https://doi.org/10.1371/journal.pcbi.1008624.s006

(TIF)

S7 Fig.

Three-state mPTP model. C0 and C1 are the two closed states. O represents the open state.

https://doi.org/10.1371/journal.pcbi.1008624.s007

(TIF)

References

  1. 1. Bers DM. Cardiac excitation–contraction coupling. Nature. 2002;415: 198–205. pmid:11805843
  2. 2. Qu Z, Hu G, Garfinkel A, Weiss JN. Nonlinear and stochastic dynamics in the heart. Phys Rep. 2014;543: 61–162. pmid:25267872
  3. 3. Xie LH, Sato D, Garfinkel A, Qu Z, Weiss JN. Intracellular Ca alternans: Coordinated regulation by sarcoplasmic reticulum release, uptake, and leak. Biophys J. 2008;95: 3100–3110. pmid:18539635
  4. 4. Clusin WT. Mechanisms of calcium transient and action potential alternans in cardiac cells and tissues. Am J Physiol—Hear Circ Physiol. 2008;294: 1–10. pmid:17951365
  5. 5. Wu Y, Clusin WT. Calcium transient alternans in blood-perfused ischemic hearts: Observations with fluorescent indicator Fura Red. Am J Physiol—Hear Circ Physiol. 1997. pmid:9374749
  6. 6. Clusin WT. Calcium and cardiac arrhythmias: DADs, EADs, and alternans. Critical Reviews in Clinical Laboratory Sciences. 2003. pmid:12892319
  7. 7. Lau E, Kossidas K, Kim TY, Kunitomo Y, Ziv O, Zhen S, et al. Spatially discordant alternans and arrhythmias in tachypacing-induced cardiac myopathy in transgenic LQT1 rabbits: The importance of IKsand Ca2+ cycling. PLoS One. 2015;10: 1–21. pmid:25970695
  8. 8. Stary V, Puppala D, Scherrer-Crosbie M, Dillmann WH, Armoundas AA. SERCA2a upregulation ameliorates cellular alternans induced by metabolic inhibition. J Appl Physiol. 2016;120: 865–875. pmid:26846549
  9. 9. Song Z, Liu MB, Qu Z. Transverse tubular network structures in the genesis of intracellular calcium alternans and triggered activity in cardiac cells. J Mol Cell Cardiol. 2018;114: 288–299. pmid:29217432
  10. 10. Rosenbaum DS, Jackson LE, Smith JM, Garan H, Ruskin JN, Cohen RJ. Electrical Alternans and Vulnerability to Ventricular Arrhythmias. N Engl J Med. 1994;330: 235–241. pmid:8272084
  11. 11. Verrier RL, Klingenheben T, Malik M, El-Sherif N, Exner D V., Hohnloser SH, et al. Microvolt T-wave alternans: Physiological basis, methods of measurement, and clinical utilityconsensus guideline by international society for Holter and noninvasive Electrocardiology. Journal of the American College of Cardiology. 2011. pmid:21920259
  12. 12. Yan Y, Liu J, Wei C, Li K, Xie W, Wang Y, et al. Bidirectional regulation of Ca2+ sparks by mitochondria-derived reactive oxygen species in cardiac myocytes. Cardiovasc Res. 2008;77: 432–441. pmid:18006452
  13. 13. Zhou L, Aon MA, Liu T, O’Rourke B. Dynamic modulation of Ca2+ sparks by mitochondrial oscillations in isolated guinea pig cardiomyocytes under oxidative stress. J Mol Cell Cardiol. 2011;51: 632–639. pmid:21645518
  14. 14. Florea SM, Blatter LA. The role of mitochondria for the regulation of cardiac alternans. Front Physiol. 2010;1 NOV: 1–9. pmid:21522484
  15. 15. Zhao Z, Gordan R, Wen H, Fefelova N, Zang WJ, Xie LH. Modulation of intracellular calcium waves and triggered activities by mitochondrial Ca flux in mouse cardiomyocytes. PLoS One. 2013;8: 1–17. pmid:24348912
  16. 16. Shimizu H, Schredelseker J, Huang J, Lu K, Naghdi S, Lu F, et al. Mitochondrial Ca2+ uptake by the voltage-dependent anion channel 2 regulates cardiac rhythmicity. Elife. 2015;4: 17347. pmid:25588501
  17. 17. Xie A, Song Z, Liu H, Zhou A, Shi G, Wang Q, et al. Mitochondrial Ca2+ influx contributes to arrhythmic risk in nonischemic cardiomyopathy. J Am Heart Assoc. 2018;7: 1–21. pmid:29627768
  18. 18. Samanta K, Douglas S, Parekh AB. Mitochondrial calcium uniporter MCU supports cytoplasmic Ca2+ oscillations, store-operated Ca2+ entry and Ca2+-dependent gene expression in response to receptor stimulation. PLoS One. 2014;9. pmid:25004162
  19. 19. Pan X, Liu J, Nguyen T, Liu C, Sun J, Teng Y, et al. The physiological role of mitochondrial calcium revealed by mice lacking the mitochondrial calcium uniporter. Nat Cell Biol. 2013;15: 1464–1472. pmid:24212091
  20. 20. Kwong JQ, Lu X, Correll RN, Schwanekamp JA, Vagnozzi RJ, Sargent MA, et al. The mitochondrial calcium uniporter selectively matches metabolic output to acute contractile stress in the heart. Cell Rep. 2015;12: 15–22. pmid:26119742
  21. 21. Li Q, Su D, O’Rourke B, Pogwizd SM, Zhou L. Mitochondria-derived ROS bursts disturb Ca2+ cycling and induce abnormal automaticity in guinea pig cardiomyocytes: A theoretical study. Am J Physiol—Hear Circ Physiol. 2015;308: H623–H636. pmid:25539710
  22. 22. Williams GSB, Boyman L, Chikando AC, Khairallah RJ, Lederer WJ. Mitochondrial calcium uptake. Proc Natl Acad Sci U S A. 2013;110: 10479–10486. pmid:23759742
  23. 23. Belevych AE, Terentyev D, Viatchenko-Karpinski S, Terentyeva R, Sridhar A, Nishijima Y, et al. Redox modification of ryanodine receptors underlies calcium alternans in a canine model of sudden cardiac death. Cardiovasc Res. 2009;84: 387–395. pmid:19617226
  24. 24. Hüser J, Wang YG, Sheehan KA, Cifuentes F, Lipsius SL, Blatter LA. Functional coupling between glycolysis and excitation-contraction coupling underlies alternans in cat heart cells. J Physiol. 2000;524: 795–806. pmid:10790159
  25. 25. Kockskämper J, Zima A V., Blatter LA. Modulation of sarcoplasmic reticulum Ca2+ release by glycolysis in cat atrial myocytes. J Physiol. 2005;564: 697–714. pmid:15695247
  26. 26. Smith RM, Visweswaran R, Talkachova I, Wothe JK, Tolkacheva EG. Uncoupling the mitochondria facilitates alternans formation in the isolated rabbit heart. Am J Physiol—Hear Circ Physiol. 2013;305. pmid:23645464
  27. 27. De Stefani D, Raffaello A, Teardo E, Szabó I, Rizzuto R. A forty-kilodalton protein of the inner membrane is the mitochondrial calcium uniporter. Nature. 2011;476: 336–340. pmid:21685888
  28. 28. Baughman JM, Perocchi F, Girgis HS, Plovanich M, Belcher-Timme CA, Sancak Y, et al. Integrative genomics identifies MCU as an essential component of the mitochondrial calcium uniporter. Nature. 2011;476: 341–345. pmid:21685886
  29. 29. Palty R, Silverman WF, Hershfinkel M, Caporale T, Sensi SL, Parnis J, et al. NCLX is an essential component of mitochondrial Na+/Ca 2+ exchange. Proc Natl Acad Sci U S A. 2010;107: 436–441. pmid:20018762
  30. 30. Wagner S, Rokita AG, Anderson ME, Maier LS. Redox regulation of sodium and calcium handling. Antioxidants Redox Signal. 2013;18: 1063–1077. pmid:22900788
  31. 31. Zima A V., Blatter LA. Redox regulation of cardiac calcium channels and transporters. Cardiovascular Research. 2006. pp. 310–321. pmid:16581043
  32. 32. Kohlhaas M, Nickel AG, Maack C. Mitochondrial energetics and calcium coupling in the heart. 2017 pp. 3753–3763. pmid:28105746
  33. 33. Maack C, O’Rourke B. Excitation-contraction coupling and mitochondrial energetics. Basic Res Cardiol. 2007;102: 369–392. pmid:17657400
  34. 34. Boyman L, Williams GSB, Lederer WJ. The growing importance of mitochondrial calcium in health and disease. Proc Natl Acad Sci U S A. 2015;112: 11150–11151. pmid:26311848
  35. 35. Santulli G, Xie W, Reiken SR, Marks AR. Mitochondrial calcium overload is a key determinant in heart failure. Proc Natl Acad Sci U S A. 2015;112: 11389–11394. pmid:26217001
  36. 36. Hund TJ, Rudy Y. Rate dependence and regulation of action potential and calcium transient in a canine cardiac ventricular cell model. Circulation. 2004;110: 3168–3174. pmid:15505083
  37. 37. Foteinou PT, Greenstein JL, Winslow RL. Mechanistic investigation of the arrhythmogenic role of oxidized CaMKII in the heart. Biophys J. 2015;109: 838–849. pmid:26287635
  38. 38. Nivala M, Qu Z. Calcium alternans in a couplon network model of ventricular myocytes: Role of sarcoplasmic reticulum load. Am J Physiol—Hear Circ Physiol. 2012;303: 341–352. pmid:22661509
  39. 39. Qu Z, Nivala M, Weiss JN. Calcium alternans in cardiac myocytes: Order from disorder. J Mol Cell Cardiol. 2013;58: 100–109. pmid:23104004
  40. 40. Song Z, Xie L-H, Weiss JN, Qu Z. A spatiotemporal ventricular myocyte model incorporating mitochondrial calcium cycling. Biophys J. 2019;117: 2349–2360. pmid:31623883
  41. 41. Gordan R, Fefelova N, Gwathmey JK, Xie L-H. Involvement of mitochondrial permeability transition pore (mPTP) in cardiac arrhythmias: Evidence from cyclophilin D knockout mice. Cell Calcium. 2016;60: 363–372. pmid:27616659
  42. 42. Qu Z, Liu MB, Nivala M. A unified theory of calcium alternans in ventricular myocytes. Sci Rep. 2016;6: 1–14. pmid:28442746
  43. 43. Song Z, Qu Z, Karma A. Stochastic initiation and termination of calcium-mediated triggered activity in cardiac myocytes. Proc Natl Acad Sci U S A. 2017;114: E270–E279. pmid:28049836
  44. 44. Pezhouman A, Singh N, Song Z, Nivala M, Eskandari A, Cao H, et al. Molecular basis of hypokalemia-induced ventricular fibrillation. Circulation. 2015;132: 1528–1537. pmid:26269574
  45. 45. Song Z, Ko CY, Nivala M, Weiss JN, Qu Z. Calcium-voltage coupling in the genesis of early and delayed afterdepolarizations in cardiac myocytes. Biophys J. 2015;108: 1908–1921. pmid:25902431
  46. 46. Korge P, Honda HM, Weiss JN. Regulation of the mitochondrial permeability transition by matrix Ca2+ and voltage during anoxia/reoxygenation. Am J Physiol—Cell Physiol. 2001. pmid:11171571
  47. 47. Halestrap AP, Clarke SJ, Javadov SA. Mitochondrial permeability transition pore opening during myocardial reperfusion—A target for cardioprotection. Cardiovascular Research. 2004. pmid:14962470
  48. 48. Rodrigo GC, Standen NB. Role of mitochondrial re-energization and Ca2+ influx in reperfusion injury of metabolically inhibited cardiac myocytes. Cardiovasc Res. 2005. pmid:15885675
  49. 49. Brennan JP, Southworth R, Medina RA, Davidson SM, Duchen MR, Shattock MJ. Mitochondrial uncoupling, with low concentration FCCP, induces ROS-dependent cardioprotection independent of KATP channel activation. Cardiovasc Res. 2006;72: 313–321. pmid:16950237
  50. 50. Yaginuma H, Kawai S, Tabata K V., Tomiyama K, Kakizuka A, Komatsuzaki T, et al. Diversity in ATP concentrations in a single bacterial cell population revealed by quantitative single-cell imaging. Sci Rep. 2014;4. pmid:25283467
  51. 51. Nakano M, Imamura H, Nagai T, Noji H. Ca 2+ regulation of mitochondrial ATP synthesis visualized at the single cell level. ACS Chem Biol. 2011;6: 709–715. pmid:21488691
  52. 52. Baddeley D, Jayasinghe ID, Lam L, Rossberger S, Cannell MB, Soeller C. Optical single-channel resolution imaging of the ryanodine receptor distribution in rat cardiac myocytes. Proc Natl Acad Sci U S A. 2009;106: 22275–22280. pmid:20018773
  53. 53. Soeller C, Cannell MB. Examination of the transverse tubular system in living cardiac rat myocytes by 2-photon microscopy and digital image-processing techniques. Circ Res. 1999;84: 266–275. pmid:10024300
  54. 54. Zorov DB, Juhaszova M, Sollott SJ. Mitochondrial ROS-induced ROS release: An update and review. Biochim Biophys Acta—Bioenerg. 2006;1757: 509–517. pmid:16829228
  55. 55. Zorov DB, Filburn CR, Klotz LO, Zweier JL, Sollott SJ. Reactive oxygen species (ROS)-induced ROS release: A new phenomenon accompanying induction of the mitochondrial permeability transition in cardiac myocytes. J Exp Med. 2000;192: 1001–1014. pmid:11015441
  56. 56. Yang L, Korge P, Weiss JN, Qu Z. Mitochondrial oscillations and waves in cardiac myocytes: Insights from computational models. Biophys J. 2010;98: 1428–1438. pmid:20409461
  57. 57. Zhou L, Aon MA, Almas T, Cortassa S, Winslow RL, O’Rourke B. A reaction-diffusion model of ROS-induced ROS release in a mitochondrial network. PLoS Comput Biol. 2010;6. pmid:20126535
  58. 58. Nivala M, Korge P, Nivala M, Weiss JN, Qu Z. Linking flickering to waves and whole-cell oscillations in a mitochondrial network model. Biophys J. 2011;101: 2102–2111. pmid:22067147
  59. 59. Knowlton AA, Chen L, Malik ZA. Heart failure and mitochondrial dysfunction: The role of mitochondrial fission/fusion abnormalities and new therapeutic strategies. J Cardiovasc Pharmacol. 2014;63: 196–206. pmid:23884159
  60. 60. Stowe DF, Aldakkak M, Camara AKS, Riess ML, Heinen A, Varadarajan SG, et al. Cardiac mitochondrial preconditioning by Big Ca2+-sensitive K+ channel opening requires superoxide radical generation. Am J Physiol Heart Circ Physiol. 2006;290: H434–40. pmid:16126810
  61. 61. Nivala M, de Lange E, Rovetti R, Qu Z. Computational modeling and numerical methods for spatiotemporal calcium cycling in ventricular myocytes. Frontiers in Physiology. Frontiers; 2012. p. 114. pmid:22586402
  62. 62. Wehrens XHT, Lehnart SE, Reiken SR, Marks AR. Ca2+/calmodulin-dependent protein kinase II phosphorylation regulates the cardiac ryanodine receptor. Circ Res. 2004;94. pmid:15016728
  63. 63. Ai X, Curran JW, Shannon TR, Bers DM, Pogwizd SM. Ca2+/calmodulin-dependent protein kinase modulates cardiac ryanodine receptor phosphorylation and sarcoplasmic reticulum Ca2+ leak in heart failure. Circ Res. 2005;97: 1314–1322. pmid:16269653
  64. 64. Guo T, Zhang T, Mestril R, Bers DM. Ca2+/calmodulin-dependent protein kinase II phosphorylation of ryanodine receptor does affect calcium sparks in mouse ventricular myocytes. Circ Res. 2006;99: 398–406. pmid:16840718
  65. 65. Cortassa S, Aon MA, O’Rourke B, Jacques R, Tseng HJ, Marbán E, et al. A computational model integrating electrophysiology, contraction, and mitochondrial bioenergetics in the ventricular myocyte. Biophysical Journal. 2006. pp. 1564–1589. pmid:16679365