Diltiazem prevents stress‐induced contractile deficits in cardiomyocytes, but does not reverse the cardiomyopathy phenotype in Mybpc3‐knock‐in mice

Abstract

Key points

• Hypertrophic cardiomyopathy (HCM) is the most common inherited cardiac illness and can lead to diastolic dysfunction, sudden cardiac death and heart failure.

• Treatment of HCM patients is empirical and current pharmacological treatments are unable to stop disease progression or reverse hypertrophy.

• In this study, we tested if the non‐dihydropyridine Ca²⁺ channel blocker diltiazem, which previously showed potential to stop disease progression, can improve the phenotype of a HCM mouse model (Mybpc3‐targeted knock‐in), which is based on a mutation commonly found in patients.

• Diltiazem improved contractile function of isolated ventricular cardiomyocytes acutely, but chronic application did not improve the phenotype of adult mice with a fully developed HCM.

• Our study shows that diltiazem has beneficial effects in HCM, but long‐term treatment success is likely to depend on characteristics and cause of HCM and onset of treatment.

Abstract

Left ventricular hypertrophy, diastolic dysfunction and fibrosis are the main features of hypertrophic cardiomyopathy (HCM). Guidelines recommend β‐adrenoceptor or Ca²⁺ channel antagonists as pharmacological treatment. The Ca²⁺ channel blocker diltiazem recently showed promising beneficial effects in pre‐clinical HCM, particularly in patients carrying MYBPC3 mutations. In the present study we evaluated whether diltiazem could ameliorate or reverse the disease phenotype in cells and in vivo in an Mybpc3‐targeted knock‐in (KI) mouse model of HCM. Sarcomere shortening and Ca²⁺ transients were measured in KI and wild‐type (WT) cardiomyocytes in basal conditions (1‐Hz pacing) and under stress conditions (30 nm isoprenaline, 5‐Hz pacing) with or without pre‐treatment with 1 μm diltiazem. KI cardiomyocytes exhibited lower diastolic sarcomere length (dSL) at baseline, a tendency to a stronger positive inotropic response to isoprenaline than WT, a marked reduction of dSL and a tendency towards arrhythmias under stress conditions. Pre‐treatment of cardiomyocytes with 1 μm diltiazem reduced the drop in dSL and arrhythmia frequency in KI, and attenuated the positive inotropic effect of isoprenaline. Furthermore, diltiazem reduced the contraction amplitude at 5 Hz but did not affect diastolic Ca²⁺ load and Ca²⁺ transient amplitude. Six months of diltiazem treatment of KI mice did not reverse cardiac hypertrophy and dysfunction, activation of the fetal gene program or fibrosis. In conclusion, diltiazem blunted the response to isoprenaline in WT and KI cardiomyocytes and improved diastolic relaxation under stress conditions in KI cardiomyocytes. This beneficial effect of diltiazem in cells did not translate in therapeutic efficacy when applied chronically in KI mice.

Key points

• Hypertrophic cardiomyopathy (HCM) is the most common inherited cardiac illness and can lead to diastolic dysfunction, sudden cardiac death and heart failure.

• Treatment of HCM patients is empirical and current pharmacological treatments are unable to stop disease progression or reverse hypertrophy.

• In this study, we tested if the non‐dihydropyridine Ca²⁺ channel blocker diltiazem, which previously showed potential to stop disease progression, can improve the phenotype of a HCM mouse model (Mybpc3‐targeted knock‐in), which is based on a mutation commonly found in patients.

• Diltiazem improved contractile function of isolated ventricular cardiomyocytes acutely, but chronic application did not improve the phenotype of adult mice with a fully developed HCM.

• Our study shows that diltiazem has beneficial effects in HCM, but long‐term treatment success is likely to depend on characteristics and cause of HCM and onset of treatment.

Abbreviations

AWTh

anterior wall thickness of the left ventricle

BW

body weight

Ct

cycle threshold

Dil

diltiazem

dSL

diastolic sarcomere length

FAS

fractional area shortening

Gnas

G‐protein α‐subunit

dP/dt_min

minimal rate of pressure change

dP/dt_max

maximal rate of pressure change

HCM

hypertrophic cardiomyopathy

ISO

isoprenaline

IVRT

isovolumic relaxation time

KI

Mybpc3‐targeted knock‐in

LV

left ventricle

LVM

left ventricular mass

LW

lung weight

MYBPC3 or Mybpc3

human or mouse cardiac myosin‐binding protein C gene

MV_A

mitral valve velocity in late LV filling

MV_E

mitral valve velocity in early LV filling

MYH7

β‐myosin heavy chain gene

NFT

non‐filling time of the left ventricle

Nppa

atrial natriuretic peptide gene

Nppb

brain natriuretic peptide gene

PWTh

posterior wall thickness of the left ventricle

TL

tibia length

WT

wild type

Introduction

Hypertrophic cardiomyopathy (HCM) is the most common inherited myocardial disease (prevalence 1:500). It is considered a monogenic disease with a predominance of sarcomeric gene mutations. Clinically, HCM patients present with cardiac hypertrophy, diastolic dysfunction, increased risk of sudden cardiac death in the young, and sometimes heart failure in elderly patients (Elliott et al. 2014).

Despite increasing understanding of the genetic causes of HCM, drug treatment remains empirical and no treatment has been shown to prevent or attenuate disease development, to reverse established manifestations or to impact the prognosis. In order to reduce left ventricular (LV) diastolic pressures and improve LV filling, guidelines recommend β‐adrenoceptor antagonists (β‐blockers) or Ca²⁺ channel blockers in symptomatic patients. β‐Blockers in particular are effective in patients with outflow tract obstruction under exercise and can relieve them from angina and dyspnoea by lowering the gradient and increasing time for diastolic filling (Spirito et al. 1997; Marian, 2009). Ca²⁺ channel blockers such as verapamil or diltiazem are primarily applied in patients with non‐obstructive HCM and have been shown to improve LV function in early diastole and prolong LV filling time (Hanrath et al. 1980; Choudhury et al. 1999). However, they were unable to stop disease progression or even reverse hypertrophy (Frey et al. 2012; Hamada et al. 2014). Diltiazem is recommended as an alternative for patients who are intolerant of verapamil (Elliott et al. 2014). A recent clinical trial showed promising beneficial effects of diltiazem in pre‐clinical HCM, particularly in patients carrying MYBPC3 mutations (Ho et al. 2015). In addition, chronic diltiazem treatment prevented the development of pathology in αMHC^403/+ HCM mice (Semsarian et al. 2002) and heart failure and sudden death induced by acute isoprenaline (ISO) application in TnT‐I79N HCM mice (Westermann et al. 2006).

The goal of this study was to test whether diltiazem would exert beneficial effects in a Mybpc3‐targeted knock‐in (KI) mouse model that we generated previously (Vignier et al. 2009). This model carries a Mybpc3 point mutation (c.772G > A), which is a founder mutation in Tuscany (Girolami et al. 2006) and is associated with a severe phenotype and poor prognosis in humans (Richard et al. 2003). KI mice exhibit systolic dysfunction followed by cardiac hypertrophy right after birth (Gedicke‐Hornung et al. 2013; Mearini et al. 2013, 2014), increased myofilament Ca²⁺ sensitivity and diastolic dysfunction (Fraysse et al. 2012). We evaluated the acute effect of diltiazem on the sarcomere function and Ca²⁺ transient of KI and wild‐type (WT) cardiomyocytes and the effect of chronic (6 months) diltiazem application on the cardiac phenotype of KI and WT mice.

Methods

Ethical approval

This study was performed in accordance with the Guide for the Care and Use of Laboratory Animals published by the NIH (Publication No. 85–23, revised 2011 published by National Research Council). All experimental procedures were in accordance with German Law for the Protection of Animals, and the protocol was ratified by the Ministry of Science and Public Health of the City State of Hamburg, Germany (No. 13/10). The investigators understand the ethical principles under which The Journal of Physiology operates and state that their work complies with the animal ethics checklist and the principles and regulations, as described in the Editorial by Grundy (Grundy, 2015).

Animals

The Mybpc3 KI cardiomyopathy mouse model was generated by the targeted insertion of a G > A transition on the last nucleotide of exon 6 (c.772G > A) and kept on the Black Swiss background (Vignier et al. 2009; Fraysse et al. 2012; Schlossarek et al. 2012, 2014; Gedicke‐Hornung et al. 2013; Mearini et al. 2013, 2014; Stohr et al. 2013; Friedrich et al. 2014; Najafi et al. 2014; Thottakara et al. 2015; Flenner et al. 2016). Mice were kept in the animal facility of the University Medical Centre Hamburg‐Eppendorf in conventional cages with sufficient nesting material at room temperature between 20 and 24°C and humidity between 45 and 65%. Mice received feed and water ad libitum.

Ventricular myocyte preparation

The isolation of cardiomyocytes from WT and KI mouse heart ventricles was performed as formerly described (Pohlmann et al. 2007; Flenner et al. 2016; Friedrich et al. 2016). Mice were sedated with CO₂ and killed by cervical dislocation. Hearts were cut out, cannulated via the aorta and mounted on a temperature‐controlled (37°C) perfusion system. After retrograde perfusion with Ca²⁺‐free buffer solution (113 mm NaCl, 4.7 mm KCl, 0.6 mm KH₂PO₄, 0.6 mm Na₂HPO₄, 1.2 mm MgSO₄, 12 mm NaHCO₃, 10 mm KHCO₃, 30 mm taurine, 5.55 mm glucose, 10 mm 2,3‐butanedione monoxime 10 mm Hepes, pH 7.46) for 6.5 min, hearts were digested with 0.075 mg ml⁻¹ Liberase TM (Roche Diagnostics, Mannheim, Germany) dissolved in buffer solution containing 12.5 μm CaCl₂ for 7–8 min. Ventricles were separated from the atria and minced with forceps to dissociate single cardiomyocytes. Subsequently Ca²⁺ was introduced stepwise up to a concentration of 1 mm.

Sarcomere shortening and Ca²⁺ transient measurements in intact ventricular myocytes

For contractile evaluation of cardiac myocytes the IonOptix system monitoring sarcomere movement and intracellular Ca²⁺ levels during contraction was used. Only rod‐shaped cells without membrane blebs, hypercontractile zones and spontaneous activity showing a stable contraction amplitude and rhythm at 1‐Hz pacing frequency were measured. For details on buffer composition, investigation of sarcomere shortening and Ca²⁺ transients, see Flenner et al. (2016). Measurements of contraction and Ca²⁺ transients were performed with or without diltiazem (1 μm, 5 min incubation, Sigma‐Aldrich).

Long‐term diltiazem treatment, echocardiography and haemodynamic measurements

To test the effects of long‐term diltiazem treatment on the HCM phenotype in mice, WT and KI mice (n = 10) received diltiazem via their drinking water. Treatment started at 6–8 weeks of age and was maintained for 6 months. Diltiazem (250 mg l⁻¹) was dissolved in drinking water, while control groups received normal water. Based on their water consumption, mice received a dose of 25 mg kg⁻¹ day⁻¹ diltiazem (Westermann et al. 2006). Transthoracic echocardiography was executed using the Vevo 2100 System (VisualSonics, Toronto, Canada) every 6–8 weeks over a period of 6 months as described previously (Flenner et al. 2016). Mice were sedated with isoflurane (3.5% for induction, 2% during the recording). B‐mode recordings were performed using a MS 400 transducer (18–38 MHz) with a frame rate of 230–400 frames s⁻¹ to assess LV dimensions and fractional area shortening (FAS).

Haemodynamic measurements were performed using an open‐chest approach in 34‐week‐old KI and WT mice, treated or not for 6 months with diltiazem. Mice were anesthetized with isoflurane (3.5% for induction, 2% during the recording). For analgesia, 0.5 mg (kg BW)⁻¹ buprenorphine was administered. Animals were fixed to a warming platform in a supine position and abdomen and anterior neck were shaved. Tracheotomy was performed and mice were artificially ventilated with a rodent ventilator (MiniVent Type 845, Hugo Sachs, March‐Hugstetten, Germany). The abdomen was opened subxiphoidally. The diaphragm was incised via a transversal subcostal approach and the pericardium was opened. The left ventricle was entered via an apical stab with a 25‐gauge needle, followed by a 1.2F transonic catheter. After a stabilization period of 5 min, heart rate, LV end‐diastolic and end‐systolic pressure and LV contractility (dP/dt _max) and relaxation (dP/dt _min) were recorded with the Scisense ADVANTAGE System (Scisense Inc., London, Ontario, Canada).

Expression analysis

At the end of the diltiazem treatment, WT and KI mice were killed by cervical dislocation; for further analysis hearts were extracted and cut in three parts (base, middle part and apex). The middle part was used for histological analysis, the base and apex parts were frozen in liquid nitrogen for subsequent molecular analysis. RNA was isolated as reported earlier (Friedrich et al. 2015) and 100 ng transcribed into cDNA using the SuperScript III Reverse Transcriptase kit (Life Technologies; (Friedrich et al. 2012, 2014; Thottakara et al. 2015). Quantitative determination of atrial natriuretic peptide (Nppa) and brain natriuretic peptide (Nppb) mRNA levels was achieved by real‐time PCR using the Maxima SYBR Green/Rox qPCR Master Mix (Thermo Scientific), and primers specific for every sequence (Friedrich et al. 2015). Cycle threshold (Ct) values were normalized to the stimulatory G‐protein α‐subunit (Gnas). ∆∆Ct values were related to control WT.

Histology

For the histological analysis of collagen I and III fibres the middle parts (including left and right ventricle) of extracted hearts were fixed (Histofix, Carl Roth) for 24 h, embedded in paraffin and cut into transverse sections containing both ventricles and the septum. Subsequently, sections were stained with Sirius Red to visualize collagen I and III fibres (7–9 mice per group). Sections were scanned and the extent of Sirius Red‐positive area was quantified by ImageJ and related to total cardiac area.

Statistical analysis

Data were expressed as means ± SEM. Comparisons were performed by Student's t test, one‐way or two‐way ANOVA followed by Dunnett's or Bonferroni's post hoc tests, as indicated in the figure legends (GraphPad, Prism 6). A value of P < 0.05 was considered statistically significant.

Results

Diltiazem ameliorates contractile deficits under stress conditions in isolated cardiomyocytes from KI mice

We previously reported that KI cardiomyocytes demonstrate a lack of tolerance to stress conditions (30 nm ISO plus 5‐Hz pacing; Flenner et al. 2016). We therefore applied this stress protocol in the presence or absence of 1 μm diltiazem (Fig. 1).

Figure 1. Contraction and Ca²⁺ transient parameters of WT and KI cells with an increased workload protocol and the influence of 1 μm diltiazem.

A, representative traces of an untreated and a diltiazem‐treated WT and KI cardiomyocyte paced at 1 and 5 Hz and stimulated with 30 nm ISO. The change of sarcomere length upon electrical field stimulation (marks at the bottom) is recorded over time. The grey line indicates the period of ISO stimulation. Graph to the right shows the number of arrhythmic cardiomyocytes. B, diastolic sarcomere lengths, contraction, contraction time and relaxation time. C, diastolic Ca²⁺ level, Ca²⁺ transient amplitude, Ca²⁺ transient rise and Ca²⁺ transient decay. WT (black) and KI (grey) cardiomyocytes stimulated at 1 Hz (−), 1 Hz and 30 nm ISO (+) and 5 Hz and 30 nm ISO (+/5 Hz). Open symbols and dashed lines indicate the presence of 1 μm diltiazem during the measurement: n = 12 for contraction analysis, n = 6–9 for Ca²⁺ transient analysis. Two‐way ANOVA with Bonferroni's post hoc test; ^* P < 0.05, ^** P < 0.01 and ^*** P < 0.001 vs. WT value in the same condition; ⁺ P < 0.05, ⁺⁺ P < 0.01, ⁺⁺⁺ P < 0.001 vs. KI value in the same condition; ^### P < 0.001 vs. WT untreated.

In the absence of diltiazem, diastolic sarcomere length (dSL) was lower in KI than WT cardiomyocytes at 1‐Hz pacing, and stimulation with 30 nm ISO led to a stronger positive inotropic response in KI than in WT (3‐fold vs 2‐fold increase in contraction amplitude, respectively; Fig. 1). The second phase of the stress condition protocol (= 5‐Hz pacing) led to a further shortening of dSL in KI cardiomyocytes (from 1.80 ± 0.05 μm to 1.70 ± 0.11 μm). Additionally, contraction time was longer in KI than WT cells under stress conditions. Throughout the whole recording time (up to 5 min), most WT cells were able to follow the high pacing rate and contracted in a regular manner, whereas almost 40% of KI cardiomyocytes developed arrhythmias or did not maintain stable contraction amplitudes (Fig. 1 A). Neither ISO (30 nm) stimulation nor 5‐Hz pacing alone induced this detrimental effect in KI cardiomyocytes (data not shown). Except for a longer Ca²⁺ transient rise, Ca²⁺ transient parameters were not different between KI and WT cells in all tested conditions. Since diastolic Ca²⁺ level did not differ between KI and WT in all conditions, the significant drop in dSL in KI cardiomyocytes under stress conditions was likely not the consequence of intracellular Ca²⁺ overload.

We then investigated whether a 5‐min pre‐incubation with diltiazem could improve the disease phenotype of KI cardiomyocytes (Fig. 1). We established concentration–response curves for contraction and Ca²⁺ transient amplitude with diltiazem concentrations ranging from 100 nm to 1 mm diltiazem in Mybpc3 WT cells. The full (inhibitory) effect of diltiazem was visible within 3 min of application. The IC₅₀ values of diltiazem for contraction and Ca²⁺ transient amplitudes were 7.2 and 246 μm, respectively (data not shown). We therefore used 1 μm, which was the maximal concentration that did not significantly impair sarcomere contraction. At baseline (1‐Hz pacing, no ISO stimulation), diltiazem did not influence any parameter in both genotypes (Fig. 1). Diltiazem attenuated the ISO‐induced increase in amplitudes of contraction and Ca²⁺ transient in WT and KI cardiomyocytes, and prolonged Ca²⁺ transient rise in KI. Furthermore, diltiazem blunted the ISO‐induced difference in contraction time between KI and WT cells (Fig. 1 B) and influenced Ca²⁺ transient rise time in KI, but did not affect the relaxation and Ca²⁺ transient decay time in both genotypes (Fig. 1 C). Further increase in pacing frequency from 1 to 5 Hz induced a reduction in sarcomere shortening in diltiazem‐pretreated cells. This effect also appeared in the absence of ISO stimulation (data not shown). Despite the negative inotropic effect, diastolic Ca²⁺ levels and Ca²⁺ transient amplitude did not decrease in diltiazem‐treated cells when pacing was increased. Under stress conditions, KI cells treated with diltiazem showed no decline in dSL (Fig. 1 B). Pretreatment with diltiazem also reduced the occurrence of arrhythmic behaviour under stress conditions in KI cardiomyocytes (Fig. 1 A).

Taken together, these data indicate that diltiazem protects KI cells under the stress condition protocol by stabilizing dSL, by reducing the ISO effect and by decreasing the occurrence of arrhythmias.

Long‐term diltiazem treatment does not improve the cardiac disease phenotype of KI mice

Since diltiazem had a protective effect when acutely applied in the stress protocol (Fig. 1), we further evaluated the therapeutic potential of a 6‐month diltiazem application (25 mg kg⁻¹ day⁻¹) in KI mice and compared to WT mice. Cardiac function was measured by echocardiography before, during and at the end of the diltiazem treatment. Echocardiographic recordings (every 8 weeks) of mitral blood flow and movement of the LV were used to determine cardiac function and its changes upon pharmacological treatment. During the study none of the mice died. Body weights (BW) did not significantly differ between WT and KI groups and increased from ∼20 g at the start of the study to up to 30 g at the end of the study (Table 1). The left ventricular mass‐to‐body weight ratio (LVM/BW) was higher in all KI than WT groups before the start of the treatment (Fig. 2 A). Age‐dependent increases in LVM and BW were not influenced by any treatment (Table 1), and, accordingly, LVM/BW did not differ between treated and non‐treated WT and KI mice, except in 6‐ to 8‐week‐old KI mice due to a higher LVM (Fig. 2 A, Table 1). KI mice presented larger left anterior (AWTh) and posterior wall thicknesses (PWTh) in diastole and in systole than WT mice, and both parameters increased over time (Fig. 2 C and D, Table 1). In the course of the study, KI mice displayed larger left ventricular internal diameters in diastole and systole (dLVID and sLVID, respectively) than WT mice, indicating a dilated phenotype (Table 1). Left ventricular wall thickness and chamber dimensions were not affected by diltiazem treatment in both genotypes (Fig. 2 C and D, Table 1). Fractional area shortening (FAS) tended to be lower in KI than in WT mice at the beginning of the study (Fig. 2 B) and decreased further in KI controls, leading to significant differences between WT and KI from the second echocardiography on. Diltiazem treatment did not influence FAS in WT or KI mice.

Table 1.

Echocardiographic parameters of the long‐term diltiazem study

Parameter	Age (weeks)	WT control	KI control	WT diltiazem	KI diltiazem
BW (g)	6–8	20.7 ± 1.3	21.5 ± 1.2	20.8 ± 1.1	20.2 ± 0.8
	14–17	24.5 ± 1.8	25.3 ± 1.8	24.8 ± 1.4	24.0 ± 1.1
	22–25	26.7 ± 2.3	27.2 ± 2.0	28.0 ± 2.1	26.4 ± 1.4
	32–34	27.7 ± 2.6	27.7 ± 2.1	28.4 ± 2.1	27.7 ± 1.9
LVM (mg)	6–8	66.3 ± 3.6	114 ± 7‡	70 ± 3	127 ± 9‡/§
	14–17	68.3 ± 3.5	126 ± 11‡	74 ± 6	123 ± 7‡
	22–25	73.3 ± 5.3	125 ± 9‡	79 ± 5	132 ± 11‡
	32–34	70.8 ± 3.9	130 ± 8‡	88 ± 6	157 ± 16‡
sLVID (mm)	6–8	3.4 ± 0.1	4.3 ± 0.1‡	3.6 ± 0.1	4.5 ± 0.1‡
	14–17	3.6 ± 0.1	4.5 ± 0.1‡	3.7 ± 0.1	4.5 ± 0.1‡
	22–25	3.8 ± 0.1	4.6 ± 0.1‡	3.8 ± 0.1	4.6 ± 0.1‡
	32–34	3.6 ± 0.1	4.5 ± 0.1‡	3.7 ± 0.1	4.7 ± 0.1‡
dLVID (mm)	6–8	4.3 ± 0.1	4.8 ± 0.1*	4.3 ± 0.1	4.8 ± 0.1†
	14–17	4.3 ± 0.1	5.0 ± 0.2†	4.4 ± 0.1	4.9 ± 0.1†
	22–25	4.4 ± 0.1	5.1 ± 0.1‡	4.5 ± 0.1	5.0 ± 0.1*
	32–34	4.3 ± 0.1	4.9 ± 0.1†	4.5 ± 0.1	5.1 ± 0.1‡
sAWTh (mm)	6–8	0.7 ± 0.1	0.8 ± 0.1	0.7 ± 0.1	0.8 ± 0.1
	14–17	0.7 ± 0.1	0.8 ± 0.1	0.7 ± 0.1	0.8 ± 0.1
	22–25	0.7 ± 0.1	0.8 ± 0.1	0.7 ± 0.1	0.9 ± 0.1*
	32–34	0.7 ± 0.1	0.8 ± 0.1	0.8 ± 0.1	0.9 ± 0.1
sPWTh (mm)	6–8	0.7 ± 0.1	0.7 ± 0.1	0.6 ± 0.1	0.7 ± 0.1
	14–17	0.7 ± 0.1	0.8 ± 0.1	0.6 ± 0.1	0.8 ± 0.1†
	22–25	0.6 ± 0.1	0.7 ± 0.1	0.6 ± 0.1	0.8 ± 0.1†
	32–34	0.7 ± 0.1	0.8 ± 0.1	0.7 ± 0.1	0.9 ± 0.1

Values are expressed as means ± SEM. Statistical analyses were done with the two‐way ANOVA with Bonferroni's post hoc test; ^* P < 0.05, ^† P < 0.01, ^‡ P < 0.001 vs. WT value in the same condition; ^§ P < 0.05 vs. respective KI value; n = 8–10 mice. Abbreviations used are: BW, body weight; KI, Mybpc3‐targeted knock‐in mice; LVM, left ventricular mass; sLVID, left ventricular inner dimension in systole; dLVID, left ventricular inner dimension in diastole; sAWTh, anterior wall thickness in systole; sPWTh, posterior wall thickness in systole; WT, wild‐type mice.

Figure 2. Echocardiography of cardiac parameters and function in the long‐term treatment study.

Echocardiographies were performed before the start of treatment at the age of 6–8 weeks, followed by measurements in intervals of 8 weeks (14–17, 22–25 and 32–34 weeks of age). A, ratio of left ventricular mass to body weight (LVM/BW) of control and diltiazem‐treated (hatched bars) WT (black) and KI (grey) mice. B, fractional area shortening of the left ventricle of the same mice. C and D, anterior and posterior wall thicknesses of the left ventricles in diastole (dAWTh and dPWTh, respectively) of the same mice. Two‐way ANOVA with Bonferroni's post hoc test; ^*** P < 0.001 vs. respective WT control group with the same treatment. ^# P < 0.05 vs untreated KI group; n = 7–10.

Pulsed‐wave Doppler images revealed that the time between diastolic filling phases of the LV (non‐filling time, NFT) did not differ between untreated groups during the entire study. On the other hand, the aortic ejection time was shorter in KI than in WT mice (Table 2). This resulted in up to 2‐fold higher myocardial performance index values, a parameter which calculates the fraction of cardiac output during the NFT. Diltiazem treatment did not influence these parameters in WT and KI mice (Table 2).

Table 2.

Echocardiographic parameters obtained by tissue Doppler imaging

Parameter	Age (weeks)	WT control	KI control	WT diltiazem	KI diltiazem
NFT (ms)	6–8	104 ± 12	115 ± 24	101 ± 7	109 ± 12
	14–17	94 ± 11	94 ± 10	99 ± 12	104 ± 14
	22–25	95 ± 9	100 ± 12	96 ± 13	106 ± 7
	32–34	88 ± 11	96 ± 7	90 ± 5	103 ± 10
AET (ms)	6–8	53 ± 8	35 ± 4‡	50 ± 8	36 ± 6‡
	14–17	44 ± 12	39 ± 8	51 ± 6	37 ± 5‡
	22–25	50 ± 4	36 ± 2‡	52 ± 7	37 ± 2‡
	32–34	47 ± 7	38 ± 10	44 ± 7	34 ± 4*
MPI	6–8	1.0 ± 0.3	2.3 ± 0.5‡	1.2 ± 0.5	2.2 ± 0.5‡
	14–17	1.4 ± 0.7	1.5 ± 0.3	1.1 ± 0.3	1.9 ± 0.4‡
	22–25	1.0 ± 0.2	1.8 ± 0.3‡	0.9 ± 0.2	2.0 ± 0.2‡
	32–34	0.9 ± 0.2	1.6 ± 0.4†	1.2 ± 0.4	2.1 ± 0.2‡
MVE (mm s−1)	6–8	710 ± 131	356 ± 77‡	662 ± 160	427 ± 80†
	14–17	669 ± 137	436 ± 45†	640 ± 110	402 ± 129†
	22–25	675 ± 145	414 ± 69‡	672 ± 87	437 ± 111‡
	32–34	754 ± 73	442 ± 85‡	725 ± 170	461 ± 95‡
MVA (mm s−1)	6–8	485 ± 60	258 ± 67*	359 ± 130	539
	14–17	487 ± 133	350 ± 117	355 ± 101	498
	22–25	483 ± 97	284 ± 182	422 ± 157	356
	32–34	532 ± 53	318 ± 99	447 ± 170	438
E/A	6–8	1.52 ± 0.1	1.31 ± 0.3	1.93 ± 0.3	1.03
	14–17	1.53 ± 0.5	1.37 ± 0.4	1.66 ± 0.4	1.09
	22–25	1.38 ± 0.3	1.88 ± 0.9	1.66 ± 0.4	1.12
	32–34	1.46 ± 0.3	1.35 ± 0.3	1.50 ± 0.4	1.04
IVRT (ms)	6–8	31 ± 2	47 ± 9‡	32 ± 4	45 ± 2*
	14–17	31 ± 3	39 ± 2	29 ± 3	45 ± 5†
	22–25	26 ± 2	39 ± 2*	25 ± 2	42 ± 2‡
	32–34	24 ± 3	33 ± 3	29 ± 2	40 ± 2*

Values are expressed as means ± SEM. Statistical analyses were done with the two‐way ANOVA with Bonferroni's post hoc test; ^* P < 0.05. ^† P < 0.01. ^‡ P < 0.001 vs. WT value in the same condition; n = 8–10 mice. Abbreviations used are: WT, wild‐type (mice); KI, Mybpc3‐targeted knock‐in (mice); NFT, non‐filling time; AET, aortic ejection time; MPI, myocardial performance index (MPI = (NFT − AET)/AET); MV_E, early filling of the left ventricle; MV_A, late filling of the left ventricle; E/A; ratio of the early (E) to late (A) ventricular filling velocities; IVRT, isovolumic relaxation time. Note that only one KI mouse treated with diltiazem showed measurable A‐waves.

Diastolic function was evaluated by measuring blood flow velocities at the mitral valve in the early (MV_E) and late (MV_A) phases of LV filling and isovolumic relaxation time (IVRT). Lower E‐wave values were repeatedly found in KI groups (maximal blood flow velocity ∼400 mm s⁻¹ vs. 700 mm s⁻¹ in WT animals; P < 0.01; Table 2). Despite the high variability caused by the small number of usable A‐wave measurements, this parameter also showed a tendency to be lower in KI mice (200–300 mm s⁻¹ in KI and ≥400 mm s⁻¹ in WT; Table 2). IVRT was significantly longer in untreated KI mice before the start of the study and in the third echocardiography (Table 2). These data suggest impaired diastolic function in KI, even though the E/A ratios did not differ between KI and WT groups. Diltiazem treatment did not influence E‐waves in KI and WT mice (Table 2). The A‐wave peaks were slightly higher in WT and KI mice supplied with diltiazem, which led to a tendency to smaller E/A ratios than in untreated mice (Table 2). However, IVRT remained longer in diltiazem‐treated KI than WT mice throughout the whole study (Table 2).

In a subset of mice, cardiac function was further evaluated by performing intraventricular haemodynamic measurements with a catheter (Fig. 3 A–C). Heart rates were lower in diltiazem‐treated mice, but the difference was not significant, while maximal rate of pressure change (dP/dt _max), used as an indicator of LV systolic function, did not differ between the groups in all conditions (Fig. 3 A and B). While the minimal rate of pressure change (dP/dt _min), used as another indicator of diastolic function, did not differ between untreated KI and WT mice, it was significantly lower in diltiazem‐treated KI than WT mice (Fig. 3 C).

Figure 3. Haemodynamic measurements of cardiac function of selected treatment groups and body parameters at the end of the long‐term treatment study.

A, heart rate of WT (black bars) and KI (grey bars) control mice and diltiazem‐treated KI animals (hatched bars). B, maximal rate of pressure change (dP/dt _max). C, minimal rate of pressure change (dP/dt _min) of the same animal groups. D, body weight (BW). E, heart weight (HW) to body weight ratio. F, lung weight (LW) to body weight ratio. G, tibia length (TL). H, heart weight to tibia length ratio. I, lung weight to tibia length ratio. One‐way ANOVA with Bonferroni's post hoc test; ^* P < 0.05 and ^*** P < 0.001 vs. WT in the same condition, or Student's t test ^# P < 0.05 vs. KI control.

At the end of the study we assessed BW, heart and lung weight (HW, LW, respectively) and tibia length (TL). BW, TL, LW/BW, LW/TL did not differ between the groups (Fig. 3 D, F, G and I). In contrast, HW/BW and HW/TL were higher in KI that in WT mice and not affected by chronic diltiazem application (Fig. 3 E and H).

It has been reported that long‐term diltiazem application partially prevented the reactivation of the fetal gene program of hypertrophy and the development of fibrosis in αMHC^403/+ HCM mice (Semsarian et al. 2002). Despite the absence of beneficial effects of chronic diltiazem treatment on the cardiomyopathy phenotype of KI mice, we evaluated whether diltiazem impacts on these parameters. The level of Nppa mRNA was slightly and the level of Nppb mRNA was significantly higher in untreated KI than WT mice, but diltiazem treatment did not significantly affect these levels in both genotypes (Fig. 4). Histological analysis of Sirius Red‐stained collagen I and III fibres in mouse cardiac sections showed a slight, but not significantly higher percentage of collagen content in untreated KI than in WT mice, and a slight, not significant reduction after diltiazem in both genotypes (Fig. 5).

Figure 4. Levels of mRNA markers for hypertrophy.

Levels of atrial natriuretic peptide mRNA (A) and brain natriuretic peptide mRNA (B) in WT and KI ventricular samples. One‐way‐ANOVA, Tukey post hoc test, ^* P < 0.05 vs. WT in same condition.

Figure 5. Sirius‐Red positive area in Mybpc3 WT and KI hearts.

A, representative mouse heart images of a Mybpc3 WT control mouse, a Mybpc3 KI control mouse, a diltiazem‐treated Mybpc3 WT mouse and a diltiazem‐treated Mybpc3 KI mouse. B and C, quantification of Sirius‐Red positive area, 1 section per heart in 7–9 animals as indicated in the bars. Scale bar 1 mm (A) and 100 μm (B).

Taken together, these in vivo data did not provide evidence for amelioration of the cardiomyopathy phenotype in KI mice by chronic application of diltiazem.

Discussion

Guideline‐recommended treatment strategies of HCM primarily consist of β‐blockers and Ca²⁺ channel blockers, which improve clinical symptoms, help to prevent arrhythmias and ameliorate diastolic dysfunction by prolonging LV filling time and reducing outflow tract obstruction (Spoladore et al. 2012; Elliott et al. 2014; Hamada et al. 2014; Tardiff et al. 2015). Nevertheless, evidence of a long‐term impact on functional capacity or prognosis in HCM patients is missing. A recent clinical trial showed promising beneficial effects of diltiazem in pre‐clinical HCM, particularly in patients carrying MYBPC3 mutations (Ho et al. 2015). A common feature observed in human HCM and animal models is an increased myofilament Ca²⁺ sensitivity. Myofilaments with increased affinity for Ca²⁺ may act as Ca²⁺ buffers leading to an increase in the Ca²⁺ pool at the Z‐disc which could also activate hypertrophic signalling cascades such as calcineurin leading to hypertrophy and fibrosis, which are also prominent HCM characteristics (Frank et al. 2006; Rohini et al. 2010; Shabbir et al. 2011). In the present study we evaluated whether the non‐dihydropyridine Ca²⁺ channel blocker diltiazem would have beneficial effects in an Mybpc3‐targeted knock‐in mouse model of HCM. The main findings of this study are: (i) diltiazem acutely improved KI cardiomyocyte performance under stress conditions; (ii) long‐term application of diltiazem did not reverse the activation of the fetal gene program, fibrosis, cardiac hypertrophy and dysfunction in KI mice. Therefore, our study provides additional evidence that acute diltiazem treatment can prevent stress‐induced contractile abnormalities, whereas chronic diltiazem treatment does not reverse a pre‐existing cardiac disease phenotype.

The combination of ISO and high pacing frequency (= stress conditions) worsened the phenotype of KI cardiomyocytes, showing a drastic decrease in dSL and a trend towards higher frequency of arrhythmic contractions than WT. This supports previous observations that ISO application worsened diastolic function and increased arrhythmia frequency in the TnT‐I79N HCM mouse model (Knollmann et al. 2001; Sirenko et al. 2006; Baudenbacher et al. 2008; Huke et al. 2013). Whereas diltiazem did not have major effects at baseline, it reduced Ca²⁺ transient amplitude and sarcomere shortening under stress conditions. Interestingly in KI, diltiazem appeared to diminish the Ca²⁺ transient more strongly than sarcomere shortening. This might be explained by high myofilament Ca²⁺ sensitivity in KI (Fraysse et al. 2012; Flenner et al. 2016; Friedrich et al. 2016), a condition in which small changes in cytosolic Ca²⁺ can evoke strong changes in contractility. Diltiazem itself neither influenced myofilament Ca²⁺ sensitivity directly nor indirectly via changing phosphorylation of PKA targets (data not shown). Specifically in KI cells, diltiazem improved diastolic function and reduced the occurrence of arrhythmias under stress conditions, suggesting that it protects KI cells under acute stress conditions. The effect could be explained by limiting the ISO‐induced increase in Ca²⁺ influx via the L‐type Ca²⁺ channel (LTCC), leading to less Ca²⁺ available for Ca²⁺ storage in the sarcoplasmic reticulum (SR) than normal after ISO stimulation (Eisner et al. 2013), and consequently, less Ca²⁺ release from the SR. These data support previous findings that diltiazem prevented acute ISO‐induced contractile dysfunction and sudden cardiac death in TnT‐I79N HCM mice (Westermann et al. 2006). Together, the data suggest that the reduction of Ca²⁺ influx via blockade of LTCC enabled HCM cells and mice to better tolerate adrenergic stress.

On the other hand, long‐term treatment of KI mice with diltiazem did not produce salutary structural or functional effects. All the parameters of LV hypertrophy, dilatation and dysfunction did not differ between diltiazem‐treated and untreated mice, and diastolic function (dP/dt _min) obtained by haemodynamic measurements even worsened in diltiazem‐treated KI mice. Consequently, diltiazem did not reduce the expression of the fetal gene program or the extent of fibrosis. Our findings are in agreement with those obtained with the same dose and treatment duration in the TnT‐I79N mice that exhibit hypercontractility and diastolic dysfunction (Westermann et al. 2006). On the other hand, our findings differ from the findings obtained in preclinical HCM patients with MYH7 or MYBPC3 mutations (Ho et al. 2015) and in the αMHC^403/+ mice with or without cyclosporine treatment (Fatkin et al. 2000; Semsarian et al. 2002), showing prevention of the development of hypertrophy and hypercontractility, as well as partial prevention of expression of hypertrophic markers and fibrosis in HCM mice (Semsarian et al. 2002).

The lack of long‐term efficacy of diltiazem in both Mybpc3 KI and TnT‐I79N mice (Westermann et al. 2006) is likely to be due, at least in part, to the pre‐existing cardiac disease phenotype. We previously showed that different pharmacological treatment approaches with ranolazine or the β‐blocker metoprolol did not reverse or improve the disease phenotype in the same KI mice (Friedrich et al. 2015; Flenner et al. 2016). KI mice indeed had already developed cardiac dysfunction followed by hypertrophy at the neonatal age (Gedicke‐Hornung et al. 2013; Mearini et al. 2013). Therefore, preventive therapeutic options should be tested at postnatal day 1, before the development of the disease phenotype. In this condition, Mybcp3‐gene therapy could prevent the development of cardiac dysfunction and hypertrophy over the long‐term in mice (Mearini et al. 2014). Although challenging in neonates, further analyses are needed to validate the preventive efficacy of diltiazem, as has been observed in human MYBPC3–mutation carriers (Ho et al. 2015). Alternatively, since cardiac function in KI mice does not further deteriorate, and their overall development, behaviour, food consumption or longevity are not different to WT mice, it would be advisable to define surrogate endpoints such as exercise performance or adrenergic stress and changes upon treatment. Another limitation is that the Mybpc3 KI mice show many HCM features only in the homozygous state. Mybpc3 KI mice also present a lower ejection fraction than WT. These features are in contrast to the more common characteristics of left ventricular hypertrophy, interstitial fibrosis, diastolic dysfunction and normal or even supra‐normal ejection fraction in HCM patients with heterozygous mutation states. Therefore, it might be valuable to use mice on another genetic background, as the outbred Black Swiss mice of this study seem to be resistant to heart failure‐related death, whereas mice on the congenic C57BL/6j background displayed a more severe phenotype and died earlier than mice on the Black Swiss background (Friedrich et al. 2015); and authors’ unpublished data). Finally, we cannot exclude the possibility that the given dose of diltiazem was not sufficient to improve the phenotype of the KI mice. As we did not evaluate plasma levels of diltiazem, we do not know if the concentration of diltiazem which reached the heart in vivo is comparable with our in vitro experiments. In another study though, the same dose was sufficient to protect mice carrying an HCM mutation from isoprenaline‐induced sudden cardiac death (Westermann et al. 2006).

The results of the clinical trial NCT00319982 that evaluated the potential of diltiazem in the prevention of HCM development emphasized the importance of early onset of treatment and showed diltiazem efficacy particularly in MYBPC3, but not MYH7 mutation carriers (Ho et al. 2015). Still, in contrast to homozygous Mybpc3 KI mice, disease progression in HCM patients often is slow, leaving a larger therapeutic window than in the mouse model used in the present study. The discrepancies between diltiazem effects in different HCM mouse models and patients emphasizes the need for individualized treatment which could be achieved by in vitro models with human cells. By disease modelling in a dish, therapies (drug or gene therapy) could be applied in a mutation‐ or patient‐specific context (Eschenhagen et al. 2015).

Additional information

Competing interests

The authors declare that there are no competing interests or conflict of interest in accordance with the policy of The Journal of Physiology.

Author contributions

The experiments were performed in the Department of Experimental Pharmacology and Toxicology, Cardiovascular Research Centre, University Medical Centre Hamburg‐Eppendorf, Hamburg, Germany. F.F.: conception of the study; acquisition, analysis, and interpretation of data; critical revision of the manuscript and important intellectual contribution. B.G.: acquisition and analysis of data, contribution to writing the manuscript. S.R.D.: acquisition and analysis of data, contribution to writing the manuscript. FW: acquisition and analysis of data, contribution to writing the manuscript. T.E.: critical revision of the manuscript and important intellectual contribution to writing the manuscript. L.C.: conception of the study; data interpretation; critical revision of the manuscript and important intellectual contribution. F.W.F.: conception of the study; acquisition, analysis, and interpretation of data; drafting of the manuscript. All authors approved the final version of the manuscript and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed.

Funding

This work was supported by the Deutsche Stiftung für Herzforschung (F/28/12).