Computational and Mathematical Methods in Medicine

Computational and Mathematical Methods in Medicine / 2016 / Article
Special Issue

Computational and Mathematical Methods in Cardiovascular Diseases

View this Special Issue

Research Article | Open Access

Volume 2016 |Article ID 7861653 | 17 pages | https://doi.org/10.1155/2016/7861653

In Silico Evaluation of the Potential Antiarrhythmic Effect of Epigallocatechin-3-Gallate on Cardiac Channelopathies

Academic Editor: Sharon Zlochiver
Received11 Aug 2016
Revised21 Sep 2016
Accepted29 Sep 2016
Published02 Nov 2016

Abstract

Ion channels are transmembrane proteins that allow the passage of ions according to the direction of their electrochemical gradients. Mutations in more than 30 genes encoding ion channels have been associated with an increasingly wide range of inherited cardiac arrhythmias. In this line, ion channels become one of the most important molecular targets for several classes of drugs, including antiarrhythmics. Nevertheless, antiarrhythmic drugs are usually accompanied by some serious side effects. Thus, developing new approaches could offer added values to prevent and treat the episodes of arrhythmia. In this sense, green tea catechins seem to be a promising alternative because of the significant effect of Epigallocatechin-3-Gallate (E3G) on the electrocardiographic wave forms of guinea pig hearts. Thus, the aim of this study was to evaluate the benefits-risks balance of E3G consumption in the setting of ion channel mutations linked with aberrant cardiac excitability phenotypes. Two gain-of-function mutations, -p.R222Q and -p.I141V, which are linked with cardiac hyperexcitability phenotypes were studied. Computer simulations of action potentials (APs) show that 30 μM E3G reduces and suppresses AP abnormalities characteristics of these phenotypes. These results suggest that E3G may have a beneficial effect in the setting of cardiac sodium channelopathies displaying a hyperexcitability phenotype.

1. Introduction

Ion channels are transmembrane proteins that allow the passage of ions according to the direction of their electrochemical gradients across cell membranes. They are pore-forming membrane proteins whose normal function is critical for several physiological processes in cells. In excitable cells, such as cardiac cells, the activity of these proteins maintains the resting membrane potential and generates action potentials that are essential for excitation-contraction coupling process.

Ion channel dysfunction is the principal pathophysiological mechanism underlying various inherited forms of arrhythmic disorders, also called channelopathies [1]. In cardiac cells, mutations in more than 30 genes encoding ion channels have been associated with an increasingly wide range of inherited cardiac arrhythmias [1]. Examples of genetic cardiac disorders include congenital ectopic Purkinje-related premature contractions (MEPPC) and exercise induced polymorphic ventricular tachycardia (EPVT) which have been linked to the presence of the -p.R222Q and -p.I141V mutations [24].

The ECG of the -p.R222Q carriers displayed atrial fibrillation, narrow junctional, and rare sinus beats competing with numerous premature ventricular contractions. The observed arrhythmia disappears under exercise [2, 3]. For -p.I141V carriers, the ECG is characterized by an increased sinus rate, atrial tachyarrhythmias, and an increased number of ventricular complexes during exercise [4]. On the molecular level, these mutations affect the biophysical properties of by shifting its voltage dependence of steady state of activation towards more negative potentials and accelerating its activation and inactivation kinetics [26].

Ion channels become one of the most important molecular targets for several classes of drugs including antiarrhythmics and local anesthetic molecules [7]. In this sense, green tea flavonoids could offer a natural promising alternative. Indeed, Epigallocatechin-3-Gallate (E3G), as a major flavanol of green tea, has shown significant effects on the electrocardiographic wave forms in guinea pig. This compound has been demonstrated to exhibit inhibitory action on several cardiac ion channels [8].

Tea is manufactured from the dried leaves of Camellia sinensis in three basic forms of nonoxidized (green), semioxidized (oolong), and oxidized (black). Green tea is one of the most widely consumed beverages in North Africa and exhibits high content in polyphenolic flavanols known as catechins which may constitute up to 36% of the dry leaf weight [9, 10]. Catechins represent 80% to 90% of green tea total flavonoids, where epigallocatechin gallate appears to be the major predominant catechin (48–55%) followed by epigallocatechin (9–12%), epicatechin gallate (9–12%), epicatechin (5–7%), and a small proportion of catechin (0.3–0.6%) [11].

Most flavonoids affect vascular system insofar to normalize blood pressure by either inhibiting calcium channels or activating potassium channels or both. But contrary to the clear-cut pathophysiological benefit of flavonoids on vascular system, the impact of these compounds on cardiac channelopathies is yet somewhat unclear [12]. Previous investigation using patch clamp technique showed that flavonoids act as multichannel inhibitors, thereby triggering generally unexpected pharmacological effects. Therefore, the reported effects on cardiac ion channels of most flavonoids remain largely unknown whether they are anti- or proarrhythmic [12, 13]. In this sense, voltage gated sodium channel (VGSC) inhibition by polyphenols is well documented as cardioprotective and antiarrhythmic pathways. Catechins, like other polyphenols, share the common structural feature of one or more phenolic rings with several antiarrhythmic VGSC inhibitors such as lidocaine and mexiletine. These polyphenolic compounds may also inhibit peak and/or late , leading to beneficial impact on the parameters associated with arrhythmias.

In this context, the aim of this study was to evaluate the benefits-risks balance of E3G effect on the setting of cardiac channelopathies.

2. Materials and Methods

2.1. Models

The action potentials were simulated using the updated mathematical model of the human atrial action potential of Maleckar-Greenstein-Trayanova-Giles (MGTG) [14], Stewart–Aslanidi–Noble–Noble–Boyett–Zhang (SANNBZ) Purkinje cell model [15], and Tusscher–Noble–Noble–Panfilov (TNNP) human ventricular cell models [16].

2.2. Formulation of Fast Sodium Current

In TNNP and SANNBZ models, the sodium current is represented according to a Hodgkin–Huxley formalism: , where is the maximal conductance of . , , and are the activation gate, fast inactivation gate, and slow inactivation gate, respectively. represents the membrane potential, and is the Nernst potential of sodium.

For the MGTG atrial model, the sodium current is represented according to the following equation:where represents the permeability to the sodium and , h1, and are the activation gate, fast inactivation gate, and slow inactivation gate, respectively. represents the membrane potential, and is the Nernst potential of sodium.

2.3. Computer Modeling of -p.R222Q and -p.I141V Mutants

The same strategy was used for all models of the atrial, human Purkinje cells, and left-ventricular myocytes [1518]. As reported by Mann et al., 2012, and Swan et al., 2014, [3, 4], the equations corresponding to the current were modified to reproduce the relative variation of biophysical properties of the sodium current due to the -p.R222Q and -p.I141V mutations.

The effects of the -p.R222Q and -p.I141V mutations were simulated as previously described by Mann et al., 2012, and Swan et al., 2014, for the ventricular and Purkinje models [3, 4]. , , , , and were modified to reproduce the shift in the voltage dependencies of steady state of activation and inactivation and their kinetics. For the MGTG atrial model, , factor, , and factor were modified to reproduce the shift of the activation and inactivation curves as well as the changes observed in the sodium current kinetics. In all conditions, the heterozygous states were reproduced by the summation of half the WT current and half the mutant current.

2.4. Computer Modeling of Epigallocatechin-3-Gallate Effect on Ion Channels

For all models and conditions (WT and mutants), the effects of E3G on ion channels were reproduced based on the experimental work of Kang et al. [8].

Table 1 summarizes the modifications of the cardiac ion currents that were introduced in all models to match the experimental Data.


TNNP/SANNBZ models, p.R222Q.TNNP/SANNBZ models, p.I141V.TNNP/SANNBZ models, TNNP/SANNB models,

WT, WT () = , WT ()
, WT () = , WT ()
WT () = , WT ()
, WT () = , WT ()
WT () = , WT ()
, WT () = , WT ()
, WT () = , WT ()
WT () = , WT ()
, WT () = , WT ()
WT () = , WT ()
100% of 100% of

WT + 30 µM E3G, WT () = , WT ()
, WT () = , WT ()
WT () = , WT ()
, WT () = , WT ( + 6)
WT () = , WT ()
, WT () = , WT ()
, WT () = , WT ()
WT () = , WT ()
, WT () = , WT ( + 6)
WT () = , WT ()
80% of 50% of

Mutants, p.I141V () = , WT , p.I141V () = , WT ()100% of 100% of
, p.I141V () = , WT (), p.I141V () = , WT
, p.I141V () = , WT (), p.I141V () = , WT ( + 7)
, p.I141V () = , WT ( + 6.2)
WT () = , WT ()
, p.I141V () = , WT ()
WT () = , WT ( + 7)

Mutants + 30 µM E3G, p.I141V () = , WT , p.I141V () = , WT 80% of 50% of
, p.I141V () = , WT (), p.I141V () = , WT
, p.I141V () = , WT (), p.I141V () = , WT ( + 7)
, p.I141V () = , WT ( + 6.2 + 6), p.I141V () = , WT ( + 6)
WT () = , WT ()WT () = , WT ( + 7)

2.5. Conduction Velocity

Conduction velocity was investigated in fibers of MGTG, TNNP, and SANNBZ cell models (pacing rates: 1 Hz for the atrial and ventricular models and 2.5 Hz for the Purkinje model). Table 3 summarizes the parameters used for the calculation of conduction velocity.

All simulations were performed by Myokit v.1.20.5 [19].

3. Results

3.1. Simulated Effect of E3G on the Electrical Activity of Cardiac Cells

To investigate the functional consequences of 30 μM E3G on the electrical activity of atrial, Purkinje, and ventricular cells, we used MGTG, SANNBZ, and human epicardial, midmyocardial, and endocardial ventricular TNNP cell models. Using these models, the observed changes in cardiac ion channels function such as voltage dependencies and current amplitudes were implemented (see Section 2, Figures 13).

In the formulations of MGTG, SANNBZ, and TNNP cell models, the effects of E3G on the sodium channel function were simulated by shifting the voltage dependence of the steady state equilibrium of gate by −6 mV (Figures 1(a), 2(a), and 3(a)). The and gates were left unchanged.

The introduction of a negative shift in the inactivation curve, related to the presence of E3G, allowed us to reproduce the inhibitory effect of this compound on the sodium current amplitude for the ventricular and Purkinje cell models (Figures 2(b), 2(c), 3(b), and 3(c)). Indeed, as reported by Kang et al. [8], the inhibitory effect of E3G is higher at depolarized resting potentials. However, for the atrial cell model, we observed that 30 μM E3G decreases the sodium current amplitude only when the resting potential is maintained at −70 mV (Figures 1(b) and 1(c)). There is no E3G effect when the resting potential is maintained at −90 mV. This is due to the biophysical properties of inactivation at basal condition. In fact, as shown in Figure 1(a), there is no difference in the sodium channel availability with or without E3G at −90 mV. Thus, an equal number of sodium channels were available in both conditions when the membrane was maintained at this potential.

Moreover, by using the same voltage protocols described by Kang et al., the inhibitory effect of 30 μM E3G on and currents was reproduced via decreasing the amplitude of these currents to 80% and 50% of WT amplitude, respectively (Figures 1(d), 1(e), 2(d), 2(e), 3(d), and 3(e)). All used formulations are summarized in Tables 1 and 2 (see Section 2). Simulations were run for 60 s with a cycle length of 1 Hz to stabilize the model. Then, supplementary run was started for another 5 s and then the last AP of each supplementary run was analyzed.


MGTG atrial model, p.R222Q.MGTG atrial model, p.I141V.⁢MGTG atrial model, ⁢MGTG atrial model,

WT, WT () = , WT (), WT () = , WT ()100% of 100% of
factor, WT () = factor, WT ()
factor, WT () = factor, WT ()
, WT () = , WT (), WT () = , WT ()
factor, WT () = factor, WT () factor, WT () = factor, WT ()
= 0.03 (s) + 0.0003 (s) = 0.03 (s) + 0.0003 (s)

WT + 30 µM E3G, WT () = , WT (), WT () = , WT ()80% of 50% of
factor, WT () = factor,
WT ()
factor, WT () = factor, WT ()
, WT () = , WT ( + 6), WT () = , WT ( + 6)
factor, WT () = factor, WT () factor, WT () = factor, WT ()
= 0.03 (s) + 0.0003 (s) = 0.03 (s) + 0.0003 (s)

Mutants, WT () = m, WT ( + 6.3), WT () = , WT ( + 7)100% of 100% of
factor, WT () = factor, WT () factor, WT () = factor, WT ( + 7)
, WT () = , WT ( + 6.2), WT () = , WT ()
factor, WT () = factor, WT () factor, WT () = factor, WT ( + 7)
= 0.03 (s) + 0.0003 (s) = 0.03 (s) + 0.0003 (s)

Mutants + 30 µM E3G, WT () = , WT ( + 6.3), WT () = , WT ( + 7)80% of 50% of
factor, WT () = factor, WT () factor, WT () = factor, WT ( + 7)
, WT () = , WT ( + 6.2 + 6), WT () = , WT ( + 6)
factor, WT () = factor, WT () factor, WT () = factor, WT ( + 7)
= 0.03 (s) + 0.0003 (s) = 0.03 (s) + 0.0003 (s)


ModelCell number
()
Intercellular conductance
(mS/µF)
Step size
(ms)

MGTG100170.01
SANNBZ100170.01
TNNP10070.001

The combined effects of 30 μM E3G on cardiac ion channels slightly decreased the AP amplitudes and maximum upstroke velocities of atrial, Purkinje, and ventricular cells (Figures 1(f), 2(f), and 3(f)). In addition, the plateau phase of atrial AP was reduced (Figure 1(f)). However, E3G increased AP duration in Purkinje cell model (Figure 2(f)).

For midmyocardial cells, E3G shortened the AP duration (Figure 3(f)). On the other hand, the superimposition of the epi-, midmyo-, and endocardial APs predicted a small decrease of the repolarization dispersion across the ventricular wall (Figure 4).

3.2. The p.R222Q and p.I141V Effects on Cardiac Excitability

According to the experimental work of Kang et al. [8], the E3G antiarrhythmic effect, in the sitting of MEPPC and EPVT cardiac disorders, was evaluated. First, we incorporated the biophysical modifications that are induced by the -p.R222Q and -p.I141V mutations and got insight on their effects on atrial, ventricular, and Purkinje APs using MGTG, TNNP, and SANNBZ models (Figures 5, 6, and 7).

Interestingly, the introduction of equations mimicking the heterozygous state into the atrial and ventricular cell models induced minor changes in their AP morphologies (Figures 8(a), 8(b), 9(a), 9(b), 10(a), 10(b), 11(a), and 11(b)). Conversely, Mann et al. and Swan et al. [3, 4] reported a drastic effect when these equations are introduced in the Purkinje cell model. Indeed, for the -p.R222Q and -p.I141V mutants, the model showed an accelerated rate of spontaneous activity of Purkinje cells leading to the occurrence of ectopic beats during the diastolic interval at 1 Hz (Figures 12(a), 12(b), 13(a), and 13(b)). These abnormalities disappeared at higher pacing rates (Data not shown).

On the other hand, using MGTG, TNNP, and SANNBZ models, strength-duration curves were constructed. In the presence of -p.R222Q and -p.I141V mutations, a lower excitation threshold for action potential generation (pacing rates: 1 Hz for atrial and ventricular models and 2.5 Hz for the Purkinje model) was observed in the p.R222Q and p.I141V mutations in homozygous and heterozygous genotypes compared with the WT ((d) in Figures 813).

Conduction velocity was investigated in fibers of MGTG, TNNP, and SANNBZ cell models (pacing rates, as described above). The presence of the -p.I141V mutation, in homozygous and heterozygous states, accelerated atrial and ventricular conduction at 1 Hz and Purkinje conduction at 2.5 Hz. Similar variations were observed in -p.R222Q mutation, whereas the conduction velocity was lower than WT condition in Purkinje cell model at 2.5 Hz pacing rate (Table 4).


ModelAtrial cellsVentricular cellsPurkinje cells
ConditionCV (cm/s)
control
CV (cm/s)
30 µM E3G
CV (cm/s)
control
CV (cm/s)
30 µM E3G
CV (cm/s)
control
CV (cm/s)
30 µM E3G

WT55.0750.8449.9145.7267.6837.20
p.I141V57.4453.9353.6349.1172.4342. 92
p.R222Q60.9356.2351.9345.5662.72

Of note, strength-duration curves and conduction velocities could no