Introduction

In pharmacotherapy for heart failure, the conventional inotropic drugs such as digitalis, β-agonist, and phosphodiesterase inhibitors were sought to enhance the calcium (Ca2+) handling for activating contractions. Thus, such drugs increased in myocardial energy expenditure and the consequently increased energy expenditure can worsen a patient’s prognosis in heart failure from a long-term perspective (Bers and Harris 2011).

A novel cardiac myosin activator, omecamtiv mecarbil (OM), has been reported to exert a positive inotropic action by accelerating the transition rate of myosin into the strongly actin-bound force-generating state, and decreasing the rate of actin-independent phosphate release, i.e., reducing non-productive and hence wasteful hydrolysis of ATP, without enhancing intracellular Ca2+ load (Teerlink 2009; Malik et al. 2011). Thus, theoretically, this drug can enhance cardiac output without increasing energy expenditure unlike conventional inotropic drugs. Previous studies reported that OM did not change minimum dP/dt, an index of diastolic function in vivo (Shen et al. 2010; Malik et al. 2011). On the other hand, in vitro, OM increased the relaxation time of cardiomyocytes (Horváth et al. 2017). The effects of OM on diastolic function and energy consumption remain controversial. Shen et al. (2010) demonstrated that chronic infusion of OM improved left ventricular (LV) function without increasing myocardial O2 consumption (MVO2) in systolic heart failure canine model. In contrast, Bakkehaug et al. (2015) reported that OM lead to reduced contractile efficiency and increased energy expenditure as mediated by hyperactivity in resting myosin ATPase in both of contractile and non-contractile of the hearts in mice and pigs. In addition, they showed that OM increased MVO2 for basal metabolism in ex vivo mice hearts with unchanged MVO2 for excitation–contraction (E-C) coupling. There are obvious discrepancies among these OM effects on cardiac function and energy expenditure, although these were reported from different animal hearts.

In the present study, we would clarify the effects of OM on the relationship between mechanical work and energy consumption using equivalent Emax (eEmax)-pressure-volume area (PVA)-VO2 framework to analyze LV mechanics and energetics, i.e., mechanoenergetics, in rat normal and failing hearts using an excised, blood-perfused whole heart preparation (Takaki 2004).

Methods

Experimental animals

Isoproterenol-induced heart failure as failing hearts (ISO-HF) made as previously reported (Shibata et al. 2011; Mitsuyama et al. 2013) and normal hearts as control (CTL) were used in the present study.

Excised cross-circulated rat heart model and data analysis

The excised cross-circulated rat heart model was described in detail in the Online supplements (supplemental Fig. 1). OM infused at a final concentration about 200–500 ng/mL into coronary circulation. Data analysis obtained from this experimental model was also described in detail in the Online supplements (supplemental Fig. 2).

Analyses of one-beat LV pressure-time curve by a logistic function

To evaluate LV end-diastolic relaxation rate or lusitropism, we used the logistic time constant derived from a logistic model to analyze LV isovolumic relaxation pressure-time curve at midrange LV volume (mLVV) (Mitsuyama et al. 2013).

Statistics

We compared the VO2–PVA regression lines of the volume runs between pre- and post-OM treatment in each heart by analysis of covariance (ANCOVA). Comparison of paired and unpaired individual values was performed by paired and unpaired t test, respectively. A value of p < 0.05 or p < 0.01 was considered statistically significant. All data are expressed as the mean ± S.D.

Results

Effects of OM on LV contractility in the CTL and ISO-HF

OM showed that both of LV ESPVRs and end-diastolic pressure-volume relations (EDPVRs) did not change in both CTL and ISO-HF (Fig. 1a). In fact, the mean ESPmLVV and PVAmLVV (n = 6, each) in ISO-HF showed significant decreases compared with those in CTL but did not show significant differences between pre- and post-OM treatment (Fig. 1d and e). The coronary blood flows were not significantly different between both CTL and ISO-HF in the absence or presence of OM (data not shown).

Fig. 1
figure 1

Representative data of ESPVRs and EDPVRs (a), and VO2–PVA relations in pre- and post-OM treated same hearts in CTL (b) and ISO-HF (c), respectively. The comparison of ESPmLVV (d), PVAmLVV (e), slope (f), and VO2 intercepts (g), and the means VO2 for E-C coupling (h) and O2 consumption for basal metabolism (i) at the pre- and post-OM treatments in CTL (n = 5–6 each) and ISO-HF (n = 4–6 each). Values are mean ± SD (h and i). *P < 0.01, pre-OM vs. post-OM. #P < 0.05, CTL vs. ISO-HF

Full size image

Effects of OM on VO2–PVA relations in the CTL and ISO-HF

OM treatment decreased the slopes of VO2–PVA linear relations in both of CTL and ISO-HF (Fig. 1b and c) and elevated the VO2 intercepts of VO2–PVA linear relations exclusively in ISO-HF (Fig. 1c). In comparisons by paired t test between pre- and post-OM treatments, the mean slopes were significantly decreased in both CTL and ISO-HF (P < 0.01, Fig. 1f). The mean VO2 intercepts were significantly increased in ISO-HF (P < 0.01, Fig. 1g) but not in CTL. These results indicate that OM markedly improves the cost of PVA or its reciprocal shows contractile efficiency, i.e., the energy transduction rate of mechanical work per ATP.

Effects of OM on O2 consumptions for E–C coupling and basal metabolism in the CTL and ISO-HF

To clarify the mechanisms of the increased VO2 intercepts in OM-treated ISO-HF, we performed cardiac arrest by KCl infusion to obtain O2 consumption for basal metabolism. OM treatment did not change the O2 consumption for basal metabolism in both of CTL and ISO-HF (Fig. 1i) but significantly increased the VO2 for E–C coupling exclusively in ISO-HF (P < 0.01, Fig. 1h). These results indicate that OM increases VO2 intercepts mediated via the increase of VO2 for E–C coupling in ISO-HF.

Effects of OM on lusitropism in the CTL and ISO-HF

We compared the normalized LV pressure-time curves and logistic time constants between pre- and post-OM treatments in the same hearts of each of CTL and ISO-HF to reveal the effects of OM on LV lusitropism. OM prolonged the duration of LV relaxation time in a CTL at 300-bpm pacing (Fig. 2a) and an ISO-HF at 240-bpm pacing (Fig. 2c) but not in a CTL at 240-bpm pacing (Fig. 2b). The mean post-OM logistic time constants elongated in both CTL at 300-bpm pacing (P < 0.05) and ISO-HF at 240-bpm pacing (P < 0.01) but not in CTL at 240-bpm pacing (Fig. 2d).

Fig. 2
figure 2

Representative data of normalized pressure-time curves at mLVV in the CTL at 300 bpm (a), 240 bpm (b) pacing, and ISO-HF (c) at 240-bpm pacing in the absence (closed circle) and the presence (opened circle) of OM (289–316 ng/mL in the blood). The comparison of the mean logistic time constants at mLVV in the CTL at 300-bpm pacing (n = 8), 240-bpm pacing (n = 7), and ISO-HF at 240-bpm pacing (n = 8) in the absence (solid column) and presence of OM (open column) (211–357 ng/mL in the blood) (d). Values are mean ± SD. *P < 0.01, #P < 0.05, OM (+) vs. OM (−)

Full size image

Discussion

In the present study, we found that the administration of OM did not change the ESPVR and EDPVR, ESP, and PVA at mLVV in both of CTL and ISO-HF, indicating that OM does not affect the LV contractility in both normal and failing hearts. Previous in vivo studies showed that OM improved stroke volume (SV) and cardiac output related to the extension of systolic ejection time (SET) in rat or canine failing hearts (Malik et al. 2011; Shen et al. 2010; Malik and Morgan 2011). In addition, Bakkehaug et al. (2015) reported that the OM increased cardiac output through the extension of SET in swine failing hearts. These results suggested that OM would exert positive inotropic effects by extending the SET without changing cardiac contractility. The results from clinical trials showing OM significantly increases SET in a dose-dependent manner in both healthy volunteers and patients with systolic heart failure (Cleland et al. 2011; Teerlink et al. 2011) might support this suggestion.

The most significant effect of OM would be an enhancing cardiac function without increasing intracellular Ca2+ and energy expenditure unlike conventional inotropic drugs. Shen et al. reported that OM treatment increased stroke work over 60% with no increase in O2 demand, resulting in a 25–30% increase in cardiac efficiency (Teerlink 2009; Shen et al. 2010). In contrast, Bakkehaug et al. (2015) reported that OM decreased the contractile efficiency evaluated by the reciprocal of the slope of PVA–MVO2 linear relation with the unchanged stroke work, indicating that OM leads to significant O2 waste. In addition, OM increased basal metabolic VO2. In the present study, our experimental model has an isovolumic contracting hearts without ejection, i.e., no cardiac output, which is a possible reason for giving the discrepancy from previous studies. We found that OM obviously improves the contractile efficiency in both normal and failing hearts with the unchanged ESPVR, indicating that OM leads to apparent O2 saving with the unchanged LV contractility (Fig. 1a, d–f). This discrepancy in the effects of OM on contractile efficiency might be generated by the difference among experimental models. In the present study, the reason why OM decreased O2 consumption in spite of the unchanged PVA at the same preload is still unknown. OM might have decreased the increasing rate of the number of myosin head binding actin filaments and/or reduced non-productive ATP hydrolysis (the rate of actin-independent phosphate release) under increasing LVV in isovolumic contracting hearts. In addition, the effects of OM on contractile efficiency appear to exert in both normal and failing hearts regardless of cardiac myosin isoform. Because it is well known that myosin isoform shift, for example, from the faster alpha-myosin isoform to slower beta-myosin isoform in rodents occurs in failing hearts. Our recent study demonstrated that the slope of VO2–VA relations in the hearts of cardiac sarcoplasmic reticulum Ca2+ ATPase (SERCA) 2a-overexpressed transgenic rats significantly smaller than that in wild-type rats, which indicates that the increased SERCA protein improved the efficiency of chemo-mechanical energy transduction (Zhang et al. 2012). Thus, the interaction between Ca2+ handling in E–C coupling and cross-bridge cycling may be related to the mechanism providing contractile efficiency.

Recently, it was reported that OM increased the open probability of ryanodine receptor channel and the Ca2+ sensitivity of force production (Nánási et al. 2017; Nagy et al. 2015), suggesting that a Ca2+ leak from the sarcoplasmic reticulum (SR) may modify the inotropic effect of OM. On the other hand, OM did not increase the Ca2+ transient though it increased rat and canine cardiac myocyte shortening (Malik et al. 2011; Horváth et al. 2017) and OM did not change the VO2 for Ca2+ handling in E–C coupling in ex vivo mouse hearts (Bakkehaug et al. 2015). In the present study, we found that the VO2 for Ca2+ handling in E–C coupling was increased in ISO-HF but not CTL after OM treatment (Fig. 1h). This increase might be related to the deterioration of Ca2+ uptake due to the SERCA2a downregulation in failing hearts. The increased intracellular Ca2+ with a Ca2+ leak from SR by OM is dominantly removed from the cytosol via Na+/Ca2+ exchanger (NCX), rather than Ca2+ uptake via the SERCA2a to SR. Although NCX per se does not consume ATP to remove cytosolic Ca2+ in exchange with Na+ influx (stoichiometry of 3Na+:1Ca2+), Na+ influx must be pumped out by Na+/K+-ATPase, with a stoichiometry of 3Na+:2 K+:1ATP, resulting in the net stoichiometry of 1Ca2+:1ATP. In contrast, SERCA2a removes cytosolic Ca2+ based on stoichiometry of 2Ca2+:1ATP (Takaki 2004; Mitsuyama et al. 2013). Thus, it is a possible mechanism that OM increased the VO2 in E–C coupling only in failing hearts.

OM significantly increased in LV relaxation time in failing hearts at 240-bpm pacing (higher 240-bpm pacing is impossible in failing hearts) or CTL at 300-bpm pacing, which would be associated with the impaired extrusion of the increased intracellular Ca2+ from the cytosol (Fig. 2). Recently, Horváth et al. (2017) demonstrated that high-frequency pacing in the presence of OM and/or high concentration of OM decreased canine cardiomyocyte shortening and OM dose-dependently prolonged canine cardiomyocyte relaxation time with the unchanged Ca2+ transient amplitude. These results suggest that any hearts with a higher heart rate, especially in diastolic dysfunction, would be susceptible to any detrimental effects of OM.

Finally, we concluded that OM improved the efficiency for converting chemical energy into mechanical work with the unchanged LV contractility in isovolumic contracting rat heart model. In the present failing hearts, OM increased the VO2 for Ca2+ handling in E–C coupling. Therefore, upon the clinical application of OM, we must perform a careful pharmacotherapy for the patients with higher heart rate, i.e., tachycardia, and/or with diastolic dysfunction, to be sufficient for the ventricular filling and the coronary flow.