Electrophysiological and Contractile Effects of Disopyramide in Patients With Obstructive Hypertrophic Cardiomyopathy

Visual Abstract


HIGHLIGHTS
In patients with HCM and symptomatic LVOT-obstruction, first treatment with disopyramide leads to a marked reduction of LVOT gradients, with a slight decrease of resting ejection fraction and a modest increase of corrected QT interval, highlighting high efficacy and safety.
In single cardiomyocytes and intact trabeculae from surgical samples of patients with obstructive HCM, in vitro treatment with 5 mmol/l disopyramide lowered force and Ca2 D transients while reducing action potential duration and the rate of arrhythmic afterdepolarizations.
These effects are mediated by the combined inhibition of peak and late Na D currents, L-type Ca 2D current, delayed-rectifier K D current, and ryanodine receptors.
In addition to the negative inotropic effect of disopyramide, in vitro results suggest additional antiarrhythmic actions.

SUMMARY
Disopyramide is effective and safe in patients with obstructive hypertrophic cardiomyopathy. However, its cellular and molecular mechanisms of action are unknown. We tested disopyramide in cardiomyocytes from the septum of surgical myectomy patients: disopyramide inhibits multiple ion channels, leading to lower Ca transients and force, and shortens action potentials, thus reducing cellular arrhythmias. The electrophysiological profile of disopyramide explains the efficient reduction of outflow gradients but also the limited prolongation of the QT interval and the absence of arrhythmic side effects observed in 39 disopyramide-treated patients. In conclusion, our results support the idea that disopyramide is safe for outpatient use in obstructive patients. since the first reports in the early 1980s and after confirmatory studies of its efficacy and safety (2)(3)(4)(5)(6).
In patients with obstructive HCM and limiting symptoms, disopyramide in addition to a beta blocker has a Class I recommendation by the 2014 European Society of Cardiology guidelines and a Class IIa recommendation by the 2011 American Heart Association/American College of Cardiology Foundation guidelines (7,8). Most recently, safe initiation in outpatients has been retrospectively demonstrated (9).
Its practical use and current place in the armamentarium for obstructive HCM has been described and reviewed elsewhere (10,11).
Despite its active use since the 1980s, there has been little work on the intracellular mechanism of therapeutic effects of disopyramide beyond its categorization as a type Ia antiarrhythmic, that is, as Na channel blocker with action potential (AP)-prolonging effects (12,13). A paper in the late 1980s suggested an effect on sarcolemmal Ca influx and efflux mediated by the Na þ /Ca 2þ exchanger (NCX) (14). In light of its potent negative inotropic effects, it is not known whether the drug has additional direct effects on Ca 2þ current, Ca 2þ release from the sarcoplasmic reticulum (SR) or on the actin-myosin interaction and the effects of disopyramide in human HCM cardiomyocytes have not been characterized.
We previously analyzed the electromechanical profile of cardiomyocytes isolated from myectomy samples of patients with obstructive HCM (15,16).
When compared with control cells, HCM cardiomyocytes showed prolonged AP, frequent afterdepolarizations, slower Ca 2þ transients, and elevated diastolic Ca 2þ concentration, largely determined by overexpression of the late Na þ current (I NaL ). Indeed, these electromechanical abnormalities were reversed by the I NaL inhibitor ranolazine, with beneficial effects on diastolic function and cellular arrhythmias (15,16).
In the present study, we sought to uncover mechanistic insights by applying electrophysiological and biophysical techniques to evaluate the effects of disopyramide on ion fluxes, afterdepolarizations, and twitch tension in isolated HCM cardiomyocytes and intact trabeculae harvested from patients undergoing surgical septal myectomy. In a translational approach, the in vitro study was combined with the first prospective characterization of the electrocardiographic and echocardiographic changes in patients with obstructive HCM started on disopyramide treatment.

Detailed methods are available in the Supplemental
Appendix data supplement. PROSPECTIVE  were dissected and the remaining tissue was minced and subjected to enzymatic dissociation to obtain viable single myocytes, as previously described (17).

SINGLE CELL STUDIES.
A perforated patch wholecell current clamp was used to measure membrane potential, as previously described (15). [Ca 2þ ] variations were simultaneously monitored using the Ca 2þ -sensitive fluorescent dye FluoForte (Enzo Life Sciences, Farmingdale, New York). A whole-cell ruptured patch voltage clamp was used to record peak and late Na þ current, L-type Ca 2þ current (I Ca-L ), and delayed rectifier K þ current (I K ), using appropriate protocols and solutions (15). protocols. In brief, we evaluated the inotropic responses to increased pacing frequencies and the kinetics of isometric twitches. Resting sarcomere length was 1.9 AE 0.1 mm.
Myocytes were permeabilized with saponin (20) and resuspended in an intracellular buffer containing 150 nmol/l free [Ca 2þ ] and 5 mmol/l of the Ca 2þsensitive dye Asante Na þ -green K þ -salt (Teflabs, Austin, Texas). The frequency of spontaneous Ca 2þ sparks was evaluated with a confocal microscope through line scan along the longitudinal cell axis (20).
Myocytes incubated with vehicle were compared with cells exposed to 5 mmol/l of disopyramide. The rate of sparks was calculated from confocal line-scan recordings using the SparkMaster ImageJ plugin (National Institutes of Health, Bethesda, Maryland) for automated analysis (21).

STATISTICS (STUDIES ON CELLS AND TRABECULAE).
None of the 20 consecutively collected patient samples was excluded from the final analysis. However, we were unable to perform all the different experi- Results from each dataset are expressed as mean AE SEM. Statistical analysis, taking into account non-Gaussian distribution, inequality of variances and within-subject correlation, was performed as previously described (15,18). In brief, to reduce the risk of type I errors resulting from the stronger interrelationship among cells/trabeculae isolated from the same patient sample, we used hierarchical statistics including 2 nested levels (patients and       Values are mean AE SD. The p values were calculated using paired Student's t-test. Comparison between echocardiographic parameters obtained before disopyramide initiation (pre-disopyramide) and at the end of study (post-disopyramide).

RESULTS
LVEDV ¼ left ventricular end-diastolic volume; LVEF¼ ejection fraction; LVESV ¼ end-systolic volume; Lateral TDI e 0 (e 0 L) ¼ early diastolic downward velocity of the lateral (free wall) mitral annulus measured at tissue Doppler; Septal TDI e 0 (e 0 S) ¼ early diastolic downward velocity of the medial (septal) mitral annulus measured at tissue Doppler; Transmitral A ¼ late diastolic transmitral flow velocity (during atrial systole); Transmitral E ¼ early diastolic flow velocity through the mitral valve.  As expected, the drug displayed a consistent negative inotropic effect ( Figure 1A). To assess the concentration dependency of this effect, we exposed the muscles to different concentrations of disopyramide ( Figure 1B). Calculated disopyramide concentration at 50% of maximal effect on isometric twitch amplitude was 5.29 AE 1.55 mmol/l. We therefore decided to employ the drug at 5 mmol/l for all the following experiments; notably, 5 mmol/l corresponds to the average plasma concentration of disopyramide measured in patients under a standard treatment regimen (25). Importantly, 5 mmol/l disopyramide hastened isometric twitch kinetics in HCM trabeculae: both time to peak and relaxation time were reversibly shortened by the application of the drug ( Figures 1A and 1C). We tested the effects of disopyramide at different stimulation frequencies ( Figure 1D): the reduction of steady-state isometric twitch force was more pronounced at higher pacing rates as compared with lower rates. Isometric twitch force was reduced by 33 AE 5% at 0.5 Hz (30 beats/min) and by 62 AE 10% at 1.5 Hz (90 beats/min; p ¼ 0.015 vs. 0.5 Hz; data from 13 trabeculae in 10 patients, calculated using linearmixed models).

Disopyramide Has No Direct Effects on Myofilament
Contraction. We tested the effects of disopyramide on demembranated trabeculae from 3 patients with HCM ( Figure 1E). Disopyramide (5 mmol/l) did not affect maximal force obtained when exposing trabeculae to an activating solution with pCa 4.5 ( Figure 1F). Force generation at lower [Ca 2þ ], determining submaximal tension development, was also unaffected by disopyramide; we concluded that disopyramide does not modify myofilament Ca sensitivity ( Figure 1F).
Moreover, disopyramide did not alter the energy cost of tension generation (Supplemental Figure 1). The   Figure 3B) and the effect was more pronounced at lower rates. For instance, average AP shortening at 0.1 Hz was 27 AE 5% and is comparable to that observed in human HCM cardiomyocytes with 10 mmol/l of ranolazine (17). In addition, disopyramide reduced AP amplitude and upstroke speed (Table 3). Disopyramide Inhibits Peak and Late Na D , Ca 2D , and K D Currents. We previously have shown that in HCM cardiomyocytes there is markedly enhanced I NaL , slightly increased I Ca-L , and lower I K, leading to prolonged AP duration (15). We assessed Na þ currents in HCM cardiomyocytes during voltage clamp on depolarization to À10 mV. Peak current was measured in the first 10 ms of depolarization ( Figure 4A), whereas I NaL was estimated by integrating the residual inward current (50 to 800 ms after onset) ( Figure 4B). In HCM myocytes, disopyramide (5 mmol/l) reduced peak Na þ current by 22 AE 4% and greatly decreased I NaL integral by 45 AE 6% (21 myocytes, n ¼ 5).
We then assessed the effects of disopyramide on I K , measured at steady state during depolarization at different potentials ( Figure 4C). Disopyramide (5 mmol/l) exerted a small but significant inhibitory effect on I K currents in HCM cardiomyocytes ( Figure 4D). Notably, the density of steady-state I K at baseline is reduced in our HCM myocytes as compared with control myocardium (Supplemental  Coppini et al.  Figures 6A and 6B). The magnitude of AP duration at 90% of repolarization shortening in HCM endocardial cells was inversely correlated with basal AP duration at 90% of repolarization ( Figure 6C). Further analysis of responders to drug action indicates that I NaL density is the primary determinant of the extent of AP shortening due to disopyramide ( Figure 6D).
In the reconstructed ventricles of a patient with obstructive HCM, 5 mmol/l disopyramide reduced conduction velocity, increasing total activation time by 11 ms (Supplemental Figure 4), comparable to clinical data (QRS prolongation at ECG, see Table 1). In the region of septal hypertrophy, 5 mmol/l of disopyramide markedly decreased dispersion of repolarization times (baseline: 529 AE 62 ms, 625 ms maximum; 5 mmol/l of disopyramide: 504 AE 50 ms, 585 ms maximum) and slightly prolonged repolarization in the nonhypertrophic epicardium ( Figure 6E).

This heterogeneous ventricular action resulted in
QT shortening in septal precordial ECG leads along with modest QT prolongation in the lateral ones ( Figure 6E), hence explaining the apparent discrepancy between AP shortening and the clinical QTc prolongation observed in patients with HCM following disopyramide intake.

DISCUSSION
In the present work, we have investigated the effects of disopyramide in patients with obstructive HCM and on isolated HCM cardiomyocytes harvested from patients undergoing surgical septal myectomy. The   In the current study, disopyramide lowered resting LVOT gradients from mean 58 to 25 mm Hg after 3 months. This was similar to a prior report in 221 patients using disopyramide with an average dose of  Na þ -current inhibition) (35,36). The ratio of negative inotropic efficacy versus Na þ -channel blocking potency was the lowest for tetrodotoxin (0.23) and the highest for disopyramide (2.2). Disopyramide has the most negative inotropic effect of all Class I antiarrhythmics, higher than mexiletine, procainamide, and quinidine (35). As tetrodotoxin is a pure Na þ -channel blocker, these results suggest that Na þ -channel inhibition only slightly contributes to the negative inotropic action of disopyramide, and that disopyramide has additional Na þ -channel-independent mechanisms that make it the most potent negative inotropic agent among class I antiarrhythmics (35).
Nonetheless, Na þ -channel inhibition may independently contribute to the reduction of Ca 2þtransient amplitude by disopyramide through modification of the activity of the NCX (37). When membrane potential is positive (peak and plateau of the AP) and subsarcolemmal [Na þ ] is high (due to large Na þ influx by peak I Na ), NCX works in reverse mode, in other words, letting Ca 2þ enter the cell in exchange for Na þ (38). The contribution of reversemode NCX to Ca 2þ transients is particularly high in HCM cardiomyocytes (16). In the presence of peak and late I Na inhibition by disopyramide, maximal AP voltage is reduced (  (15,16).
Therefore, we can conclude that simultaneous inhibition of peak and late I Na is required for the negative inotropic action of disopyramide.   therefore, its reduction by disopyramide could be antiarrhythmic.

EFFECT ON EARLY AND LATE AFTERDEPOLARIZATIONS:
ANTIARRHYTHMIC POTENTIAL. We observed a reduction of the cellular triggers of arrhythmias in HCM cardiomyocytes treated with disopyramide, that is, EAD and DAD (Figure 3). The risk of EAD is directly associated with AP prolongation (49). The shortening of AP by disopyramide is therefore the main mechanism behind the reduction of EAD (15   (Top) In HCM cardiomyocytes, I Ca-L and I NaL are increased, while I K is markedly decreased, leading to prolonged APs; Na overload impairs NCX, contributing to cytosolic Ca-overload. (Bottom) Disopyramide inhibits I Na-peak (I NaP ), I NaL , I Ca-L and I K , while also stabilizing ryanodine receptors. These effects lead to shortening of APs. Moreover, normalization of NCX function and I Ca-L inhibition and RyR stabilization contribute to reduce diastolic Ca and systolic Ca-release, determining negative inotropic effects. APs ¼ action potentials; HCM ¼ hypertrophic cardiomyopathy; I Ca-L ¼ L-type Ca current; I K ¼ delayed-rectifier K current; I NaL ¼ Late Na current; NCX ¼ Na þ /Ca 2þ exchanger; RyR ¼ ryanodine receptor.
Coppini et al. incomplete genetic data (Supplemental Table 1), we were unable to correlate specific parameters of drug effectiveness with the different disease-causing mutated genes. As disopyramide does not directly interacts with sarcomeres ( Figure 1)