Peculiar Features of the Pumping Function of the Heart in Three Types of Cardiomyopathy of Various Genesis

Abstract

The review considers changes in the pumping and contractile function of the heart in three types of cardiomyopathies. Isoproterenol cardiomyopathy is closest to ischemic cardiomyopathy, which is most commonly observed in the clinic. Cardiomyopathy caused by chronic administration of doxorubicin represents the closest to the clinic variant of toxic cardiomyopathy. Diabetic cardiomyopathy is increasingly common in our time; the review will consider information about type 1 diabetes. The greatest attention in the review is paid to diastolic dysfunction of the heart, the main causes of its occurrence and compensatory mechanisms are analyzed. The earliest changes in diastolic dysfunction in these types of cardiomyopathies are a slowdown in myocardial relaxation and endothelial dysfunction. Information is given showing that the basis of delayed relaxation is two reasons—impaired transport of Ca++ in cardiomyocytes and altered properties of connectin (titin). The ability of mitochondrial oriented antioxidants to prevent cardiac dysfunction caused by doxorubicin has been demonstrated.

1. Introduction

Chronic heart failure (CHF) is the end result of many diseases of the circulatory system. Cardiomyopathy, defined as primary myocardial weakness, is one of the common causes of CHF. Its etiology is very diverse—it can occur due to ischemia, impaired energy formation due to the action of toxic factors, hypertension, diabetes, genetic defects associated with impaired synthesis of myofibrillar proteins, and other factors. Despite the commonality of the final effect, each type of cardiomyopathy has its own special features due to specifics of damaging factors. Consideration of this specificity is the subject of this review, which will mainly present the works of the author’s team on the study of the myocardial contractile function and arterial tone in three types of cardiomyopathy. Adrenomimetic isoproterenol in high doses causes ischemic micronecroses of the myocardium, the antibiotic doxorubicin, effective in oncology, selectively damages the function of myocardial mitochondria, and another antibiotic streptozotocin, which damages insulin production and deprives cardiomyocytes of the ability to use glucose to the desired extent.

All experiments have been carried out in male Wistar rats aged 4−6 months accordingly to the principles of the 2000 declaration of Helsinki and the 1985 International Guiding Principles for Biomedical Research Involving Animals (1985). Results are represented as M±SE.

2. Isoproterenol cardiomyopathy

Cardiomyopathy caused by chronic myocardial ischemia occurs most often in the clinic. Studying the pathogenesis of this cardiomyopathy requires an adequate model, which has not yet been created. The model of cardiac ischemia caused by isoproterenol, a nonselective agonist of myocardial beta-adrenergic receptors, is considered a classic model for studying the cardioprotective effect of various compounds [1]. This model was proposed by Rona more than half a century ago, and its main consequence was multiple micronecroses of the myocardium. Subsequent studies have shown the multifactorial genesis of these changes − a violation of microcirculation with the development of local zones of ischemia, the effect of catecholamine oxidation products [2], and the overload of cardiomyocytes with Ca⁺⁺ ions with the development of local cardiomyocyte contractures [3]. Within 2−4 weeks, cardiomyopathy occurs with lower ejection fraction [4, 5].

However, insufficient attention was paid to the study of the pumping and contractile function of the heart in this pathology. In this regard, we conducted a comprehensive study of the action of isoproterenol in various doses, from 85 to 180 mg/kg [6]. The contractile function of the heart was investigated using echocardiography (iE33, Philips Ultrasound, Bothell WA, USA) using the S12–4 sensor (12–4 MHz) and catheterization of the left ventricle (LV) using the Millar precision micromanometer inserted through the carotid artery (SciSense Instruments, Canada) and the strain gauge amplifier Hugo Sachs Elektronik (Germany). Traditional contractility indicators were measured − LV maximal pressure development rate (+dP/dtmax) and the contractility index (+dP/dtmax/P − pressure at the time of reaching the maximum +dP/dt). To characterize the relaxation process, the maximum rate of pressure fall (− dP/dtmin) was used, as well as the time constant of isovolumic relaxation [7]. In total, 4 series of experiments were performed, in which isoproterenol in various doses was administered twice at a daily interval. Mortality in these series ranged from 15 to 40%. The use of isoproterenol, even in a minimal dose (85 mg/kg twice), caused mosaic changes in cardiomyocytes, characterizing diffuse micronecroses of the myocardium. Extensive infiltration by cellular elements and fibroblasts, the organizers of fibrosis was noted [6].

Surviving animals and controls were examined after 2 weeks using echocardiography. In a series of experiments using a cumulative dose of 240 mg/kg, both the cardiac output and LV ejection fraction differed slightly from the control values, but the use of a cumulative dose of 300 mg/kg was accompanied by decreased ejection fraction from 84 ± 1% to 72 ± 3% (p < 0.01). This was combined with increased LV end-diastolic volume from 0.84 ± 0.05 to 1.1 ± 0.1 (p < 0.05). The heart rate and the thickness of posterior LV wall in diastole and systole did not change, but the structure of the cardiocycle was changed − the duration of isometric phases increased from 21 ± 1.0% to 26 ± 1.4% (+26%, p < 0.01), and the duration of the period of LV filling decreased from 36 ± 1.1% to 31 ± 1.1% (− 14%, p < 0.05).

Isoproterenol in a dose of 240 mg/kg did not violate the cardiac hemodynamics and LV contractility but was accompanied by a significant, almost twice, slowdown in myocardial relaxation − the time constant of isovolumic relaxation increased from 6.3 ± 0.8 to 12.3 ± 1.8 s⁻¹ (p < 0.05). But at a dose of 360 mg/kg, delayed myocardial relaxation was combined with decreased myocardial contractility index by 20%. As a result of the deterioration of myocardial contractility and relaxation, the LV end-diastolic pressure was clearly increased from 3 ± 1 to 9 ± 2 mm Hg (p < 0.05).

These results, characterizing the dose-dependent occurrence of diastolic and systolic dysfunction after the isoproterenol administration, have been obtained earlier [8] and confirmed recently [9]. Echocardiography of the heart of mice revealed the development of diastolic dysfunction after administration of isoproterenol at a dose of 150 mg/kg and systolic dysfunction after administration of higher doses.

The reasons for CHF development at high doses of isoproterenol are obviously due to the development of oxidative stress that occurs in the myocardium with prolonged action of isoproterenol [10]. A slowdown in myocardial relaxation occurred within 3−7 days after the administration of isoproterenol, but cardiac work [5] or the pressure developed by isolated heart remained unchanged [11]. In this case, the authors found a significant decrease in the expression of proteins involved in calcium transport—Ca⁺⁺ -ATPase of the sarcoplasmic reticulum (SR) and phospholamban, as well as Na⁺-K⁺-ATPase. These data suggested that calcium transport in cardiomyocytes at higher doses of isoproterenol should be disturbed. This assumption was confirmed in our experiments in cardiomyocytes isolated from rat hearts [6].

The intracellular free Ca⁺⁺ level was measured using fluorescent Ca⁺⁺ indicator Fluo-4 (Invitrogen F-14201), and the signal was recorded using a high-speed digital camera AxioCam HS (Zeiss). To excite the cells, electrical stimulation was used with a rate of 1 Hz and a voltage of 38 V for 10 seconds. In cardiomyocytes from hearts of rats that received isoproterenol, various forms of Ca⁺⁺ signal distortion have been observed − either with an unchanged rhythm of contractions but elevation of Ca⁺⁺ diastolic level with steadily decreased magnitude of subsequent peaks or with a decreased number of signals and the appearance of additional peaks, reflecting a slowdown in Ca⁺⁺ excretion from mycoplasma. At the same time, the ascending arm of the signal (an increase in myoplasmic Ca⁺⁺, reflecting the work of calcium channels) was less altered, but the peak amplitude decreased by about 1.4 times. Thus, judging by the nature of altered calcium peaks, the process of Ca⁺⁺ removal from the mycoplasma suffered predominantly, while the process of Ca⁺⁺ entry into the mycoplasma was less disturbed.

These results can briefly be summarized in the form of three main points: 1) when using smaller isoproterenol dosages, the diastolic dysfunction arose, manifested by a relaxation slowdown and increased LV end-diastolic pressure; 2) at using higher isoproterenol dosages, the systolic dysfunction arose, manifested by decreased LV ejection fraction and LV cavity dilatation; and 3) in all series, regardless of isoproterenol dosage, the myocardial relaxation suffered to a greater extent than contractility, and it has been based on weakened ability of cardiomyocytes to remove Ca⁺⁺ from myofibrils.

3. Doxorubicin cardiomyopathy

Anthracycline cardiomyopathy is a frequent complication of the treatment of cancer patients. The most effective representative of this group is doxorubicin. This type of cardiomyopathy is characterized by the same symptom complex in humans and animals [12, 13], manifested in the form of LV systolic dysfunction and CHF with decreased LV ejection fraction and animal survival [14].

According to most researchers, the damaging effect of doxorubicin is realized by activating free radical oxidation [15, 16]. Mitochondria, characterized by a high intensity of oxidative processes, are the main cellular target of doxorubicin [16, 17, 18]. The abundance of unsaturated fatty acids in the inner membrane of mitochondria favors peroxidation with disruption of the electronic transport chain [19]. Using laser-scanning confocal microscopy with oxidation-sensitive fluorescent tag, it has been shown that the addition of doxorubicin (40–160 μM) to the cardiomyocyte incubation medium enhances oxidative reactions near mitochondria [17]. An increased vulnerability of myocardial mitochondria compared to liver mitochondria is due to the fact that cardiac mitochondria contain a large amount of cardiolipin, which has an increased affinity for doxorubicin [20]. Mitochondrial dysfunction is observed even at nanomolar concentrations of doxorubicin [21] − the formation of superoxide increases rapidly with the subsequent release of cytochrome “c”, especially in the presence of aglycone derivatives of doxorubicin [18], later the synthesis of ATP and Ca⁺⁺ transport are disturbed [22]. The biochemical study of mitochondria in skinned fibers after 2 weeks of doxorubicin injection revealed unchanged maximal level of oxygen consumption, but lower respiratory index by 30% and apparent affinity constant for ADP by 35% suggesting an increased permeability of mitochondrial membranes [23].

3.1 Studies in vivo

In recent years, we have tried to trace changes in the cardiodynamics and LV contractile function at various times after doxorubicin administration (2 mg/kg). Invasive study of the cardiac contractile function was performed using a standard PV catheter FTH-1912B-8018 and an ADV500 amplifier (Transonic, Canada), as well as the PowerLab 4/35 ADC with the LabChart 8.1 program (ADInstruments, Australia), which automatically calculated more than 20 parameters of the contractile function. Initially, experiments were performed, in which the cardiac function was studied after 8 weekly injections of doxorubicin (cumulative dose of 16 mg/kg). After 10 weeks, the mortality rate was 27% [24].

According to the results of echocardiography (n = 59), a clear decrease in LV ejection fraction was found in 54%, boundary values in 26%, and no changes in 20%. Experiments with catheterization of LV and aorta, performed after 9−13 weeks, showed a progressive diminution in LV contractility index, LV + dP/dt, and an extension of time to its peak by 1.5 times. A slowdown in LV relaxation was altered to even greater extent, the time constant of isovolumic relaxation was more than doubled. There was a progressive decrease in the heart rate and blood pressure, and respectively, LV systolic pressure decreased and LV diastolic pressure increased. Thus, the progressive deterioration of myocardial contractility and systolic dysfunction development was partially compensated by lower peripheral resistance and prolongation of the diastolic pause.

In the next series of experiments (n = 20), only 4 injections of doxorubicin were used to reduce the degree of damage [25]. In echocardiographic study performed 4 weeks after the start of administration, 73% of rats had diastolic dysfunction, and 27% had systolic dysfunction, but after another 4 weeks, during which doxorubicin was not administered, the ratio of rats with diastolic and systolic dysfunction changed in the opposite direction of 43: 57%. In 7 experiments of the last series, after 8 weeks, it was possible to trace changes in LV ejection fraction − after 4 weeks it was 76 ± 2%, and after 8 weeks it decreased to 57 ± 4% (p < 0.01). Rats with systolic dysfunction had the same signs of CHF—reduced myocardial contractility and relaxation, lower heart rate, LV systolic pressure, and heart function. Nevertheless, the cardiac output per unit of body weight was within the normal range. An important factor facilitating the LV ejection was a diminution of average aortic pressure to 94 ± 8 mmHg from the control level of 116 ± 4 mmHg, as well as a decrease in arterial elasticity index by 35%. The group of rats with normal ejection fraction was characterized by the same signs, including delayed relaxation, but they had LV contractility index and the maximum rate of pressure development at a normal level, while the relaxation time constant was significantly increased by 15%, as well as LV end-diastolic pressure from 2.4 ± 1.3 to 6.6 ± 1.0 mmHg (p < 0.05).

The measurement of energy metabolism in the myocardium of rats with systolic dysfunction in situ [26] showed a constant content of ATP and the number of adenine nucleotides, but a significant decrease in phosphocreatine content from 27.4 ± 0.3 to 16.4 ± 4.7 μmol/g. Accordingly, the phosphocreatine/ATP ratio was reduced from 2.01 ± 0.12 to 1.26 ± 0.21 (−37%, p < 0.05). It is thought that a low FCr/ATP ratio may be a predictor of cardiovascular disease mortality [27]. It should also be noted that a large increase in the myocardial lactate content from 1.7 ± 0.4 to 12.0 ± 1.8 μmol/g (p < 0.01) indicates a significant activation of anaerobic glycolysis. In rats with diastolic dysfunction, the contents of ATP and phosphocreatine were close to the control values. However, there was a significant increase in lactate compared to controls (20.0 ± 4.6 and 3.6 ± 0.8 μmol/L, respectively, p < 0.05), indicating significant activation of anaerobic glycolysis.

The results of this series showed that the formation of systolic dysfunction with decreased LV ejection fraction passes through the phase of diastolic dysfunction. Delayed myocardial relaxation is an indispensable sign of both diastolic and systolic dysfunction. Two reasons may underlie the delayed relaxation − a weakened absorption of Ca⁺⁺ ions, which activated the act of myofibrillar contraction, in the structures of the sarcoplasmic reticulum (SR) and a change in the elastic properties of connectin (titin).

Experiments with the registration of calcium signals in isolated cardiomyocytes showed changes similar to those observed under the action of isoproterenol − the cleavage of signals with the formation of a delayed Ca⁺⁺ recession. These data are consistent with the results of the study of the myocardium of patients with dilated cardiomyopathy, they showed a significantly reduced expression of SERCA2a not only in relation to the total protein but also in relation to the ryanodine receptor [28]. This means that in systolic dysfunction, the process of Ca⁺⁺ absorption into sarcoplasmic reticulum suffers to a greater extent than Ca⁺⁺ release from the reticulum.

A delayed relaxation may also be associated with a change in the properties of connectin (titin), the largest protein not only in cardiomyocytes but also in the body [29, 30]. It is known that within the sarcomere, one end of connectin is associated with myosin in A-disc region and the other end with the Z-line. Thus, it “anchors” the myosin filaments in the center of the sarcomere, contributing to the maintenance of a highly ordered sarcomere structure. With the development of contraction, myosin filaments, forming bonds with actin ones, shift to the Z line, compressing the connectin spring, and when Ca⁺⁺ is eliminated from myofibrils, the connectin spring straightens, returning the ends of the myosin filaments to their previous position. It is thanks to this that contracting isolated cardiomyocytes that do not experience external stress always return to their original length. Connectin stiffness is the main regulator of contractile activity of striated muscles [31, 32, 33]. It can change under the influence of phosphorylation of its isoforms − a more extensible N2BA (having a longer tensile part in the I-disc of the sarcomere) and a more rigid N2B. The ratio of N2BA/N2B varies in the mammalian heart: from the smallest in the myocardium of small animals to the largest in the myocardium of large animals and humans [32].

In our experiments, the properties of connectin were investigated in rats treated with doxorubicin (2 mg/kg) once a week for 4 weeks [34]. The ratio of the content of more extensible connectin N2BA isoform to elastic N2B isoform in control experiments was 14/86%. In the myocardium of doxorubicin rats, it was changed in the direction of predominance of more extensible N2BA isoform; its content was 26 ± 2%. The overall level of connectin phosphorylation was increased by 60 ± 18% compared to controls. The degree of phosphorylation of connectin was inversely correlated (r = − 0.94) with the content of rigid isoform N2B, and as is known, phosphorylation of this isoform reduces its stiffness, therefore, the myocardial extensibility should increase, contributing to better filling of the ventricular chamber [35].

Results showing increased content of more extensible N2BA isoform and increased connectin phosphorylation, suggesting raised myocardial extensibility, have been obtained for the first time. They allow us to think that in diastolic dysfunction, a delayed relaxation, combined with increased LV diastolic pressure, may be due not only to a violation of Ca⁺⁺ transport but also to increased extensibility of connectin, which slows down the relaxation process.

The analysis of these data shows that the main changes occur in energy-dependent systems that require either ATP production or phosphorylation. These are calcium ATPase of sarcoplasmic reticulum − SERCA2a, phosphorylation of phospholamban, and connectin. This suggests the primacy of disorders of the energy formationsystem in cardiomyocytes as the basis for the subsequent development of diastolic and systolic dysfunction. This is supported by the activation of anaerobic glycolysis in this pathology.

3.2 Studies in the isolated heart

To understand the pathogenesis of diastolic dysfunction, it is important to determine the initial links to myocardial dysfunction. For this purpose, the technique of retrograde perfusion of the isolated heart was used with the registration of pressure in a latex balloon introduced into LV cavity.

In connection with the widespread opinion about the primary role of oxidative stress in alteration of the contractile function of the heart, we aimed to investigate the resistance of the isolated heart to oxidative stress induced by hydrogen peroxide after administration of doxorubicin in vivo [36]. Since a peak of reactive oxygen species formation under the influence of doxorubicin was observed after 2 hours [37], in the first series of experiments, the heart was isolated 2 hours after the administration of doxorubicin in vivo (2.2 mg/kg). A rise in the perfusion rate from 10 to 20 ml/min/g made it possible to reach the maximal LV systolic pressure up to 200 mmHg. The hearts of rats that received doxorubicin showed a normal level of LV systolic pressure and the maximal rate of its development, but a rise in perfusion pressure by 26% was noted, which indicated an increased tone of coronary vessels. This corresponds to previously obtained results [38, 39]. Our data showed that this phenomenon persisted steadily for 2 hours and even 2 weeks after the administration of doxorubicin in vivo. This is probably due to an increased affinity of endothelial NO synthase for doxorubicin, which is the basis of its dose-dependent inhibition [40] with a subsequent decrease in NO formation. Myocardial relaxation indices 2 hours after doxorubicin administration remained at the same level, but after 2 weeks were significantly reduced by 11−14%, and LV minimal diastolic pressure increased from 2 ± 1 to 7 ± 1 mmHg, i.e. signs characteristic of diastolic dysfunction were observed.

The introduction of hydrogen peroxide (100 μM) into the perfusate as an inducer of reactive oxygen species (ROS) in the control group was accompanied by a steadily fall in LV developed pressure, as well as + dP/dt, after 40 minutes these values decreased to 60 ± 3% of the initial level. Rat hearts isolated 2 hours after doxorubicin administration showed a greater fall—to 40 ± 2% (p < 0.001), and this was combined with a significant rise in LV minimal diastolic pressure to 34 mmHg. However, after 2 weeks LV developed pressure response at H₂O₂ introduction in hearts of rats received doxorubicin practically did not differ from the control values, and LV minimal diastolic pressure did not rise [39].

Thus, the elevated negative inotropic effect in response to oxidative stress was observed only in the early stage after the doxorubicin administration, at the peak of ROS formation in the myocardium. It could be assumed that it is associated with a decrease in the activity of antioxidant enzymes, but this activity, measured in myocardial samples taken after the completion of the experiment, did not change either after 2 hours or after 2 weeks, although the concentration of MDA in both series after doxorubicin use was significantly increased by 23−34%. This coincides with the results of the work [41], in which unchanged activity of GSH-Px, catalase, and MnSOD was also found in the interval from 1 to 24 hours after doxorubicin administration (2.5 mg/kg). The activity of antioxidant enzymes increased 4 weeks after the start of doxorubicin [42].

Another series of experiments was aimed to clarify the state of the calcium transport system in the myocardium under the action of doxorubicin [43]. The amplified signals of the pressure sensors were transmitted to the computer via the NI USB-6210 analog-to-digital converter (ADC) from National Instruments (USA), using a digitization frequency of 1000 Hz. A sudden increase in the rate of excitations from a base rate of 5 Hz to 10 Hz was used with a return to 5 Hz after 10 s. The addition of doxorubicin (3 μM) to the perfusate after 30 minutes was accompanied by a decrease in LV developing pressure by 15−20%. The high-frequency load immediately increased LV developed pressure and the maximal rate of its development, as well as the constant of the relaxation rate (the value inversed to the relaxation time constant) and diastolic pressure (Figure 1 in [43]). After doxorubicin addition to the perfusate, similar phenomena were observed, and the magnitude of studied indices coincided. Maintaining an adequate response to the high-frequency load allows us to suggest that the calcium transport system in these conditions was not disturbed, despite a decrease in the initially developed pressure. It is known that the function of ion pumps is supplied by ATP from glycolysis, while the function of myofibrils is from oxidative phosphorylation. The results of these experiments showed that after a single injection of doxorubicin, the earliest changes are an increased tone of the coronary vessels and a gradual slowdown in myocardial relaxation.

3.3 Prevention of heart dysfunction at doxorubicin cardiomyopathy

The increased vulnerability of the cardiac contractile function to H₂O₂ precisely in the initial period of doxorubicin action suggested that reducing the intensity of oxidative stress in the myocardium during this period may delay or prevent the CHF development at prolonged action of doxorubicin. This assumption is based on the information according to which isolated cardiomyocytes with reduced superoxide dismutase (SOD) activity induced by diethyldithiocarbamate were less resistant to the action of doxorubicin [44], and the preliminary use of SOD, like glutathione peroxidase (GSH-Px), prevented the development of apoptosis in embryonic cardiomyocytes under the influence of doxorubicin [45].

To reduce the degree of myocardial damage, the mitochondrial-oriented antioxidant SkQ1 synthesized at Moscow State University was used [46]. Its active ingredient is plastoquinone, a powerful plant antioxidant that performs the same function in plant mitochondria as coenzyme Q in the myocardium. It penetrates into the cells and then into the mitochondria due to a weak positive charge of triphenylphosphonium.

Two series of in vivo experiments were performed. In the first series, doxorubicin was administered for 5 weeks when systolic dysfunction was observed in most experiments [47]. Echocardiographic examination of rats followed by LV catheterization after another 3 weeks, during which the rats received neither doxorubicin nor plastomitin, showed the presence of LV systolic dysfunction with lower LV ejection fraction and contractility index by 32−34% and a rise in the relaxation time constant by 74%. At the same time, LV maximal developed pressure was reduced by 18%, which closely coincided with lower blood pressure by 23% and diminished arterial stiffness by 35%. A significantly decreased peripheral resistance and prolongation of diastolic pause due to 23% lower heart rate can contribute to maintaining LV maximal ejection rate and cardiac output at normal level. The hearts of rats that were administered plastomitin simultaneously with doxorubicin were characterized by normalization of the ejection fraction, all indices of contractility, heart rate, blood pressure, and arterial stiffness.

In another series of experiments [48], doxorubicin (2 mg/kg) was administered twice in 2 weeks, and half of the rats received plastomitin daily along with it. The use of doxorubicin did not cause significant hemodynamic disturbances, but LV maximal rate of pressure development was reduced by 36% and the contractility index by 23%. Even deeper, by 39−43%, the indices of myocardial relaxation were reduced. In the group of rats that received doxorubicin together with plastomitin, all indices of myocardial contractility and relaxation were significantly higher than in the doxorubicin group, and they did not differ from control values. These changes were combined with reduced LV diastolic pressure.

Thus, the use of plastomitin together with doxorubicin completely eliminated the signs of both systolic and diastolic dysfunction caused by doxorubicin. The positive effect of plastomitin is consistent with the idea of the important role of oxidative stress in the initial period of chronic use of doxorubicin. This can serve as the basis for a preclinical trial of plastomitin in doxorubicin therapy for cancer patients.

The results of our study of the pathogenesis of doxorubicin cardiomyopathy showed: a) the earliest changes in the cardiac contractile function are caused by the action of oxidative stress, which impairs the function of mitochondria; at the same time, the calcium transport system in cardiomyocytes does not seem to suffer; b) the initial changes in LV contractile function are manifested in the form of delayed relaxation, as well as increased myocardial extensibility, which is based on connectin phosphorylation, as well as an increase in the content of its more extensible isoform N2BA; c) with smaller doses of doxorubicin, diastolic dysfunction occurs, the obligatory manifestation of which is delayed relaxation, the pumping function of the heart remains at a normal level; d) with higher doses of doxorubicin or increased duration of the drug, systolic dysfunction develops with a violation of the cardiac pump function; and e) both forms of CHF can be successfully prevented if the effective mitochondrial oriented antioxidant plastomitin (SkQ1) is administered to rats concomitantly with the administration of doxorubicin.

4. Diabetic cardiomyopathy

Diabetes is a special disease in which the consumption of glucose by cells is disturbed. There are two types of diabetes. The first type occurs after a significant decrease or complete cessation of insulin production in the pancreas, the second is characterized by normal insulin production, but with a difficulty in glucose transport into cells. The first type is reproduced through the use of streptozotocin or alloxan, which damages the pancreas. Reproduction of the second type is much more difficult. In our experiments, a type 1 model was used.

It is known that in type 1 diabetes the energy metabolism of cardiomyocytes switches almost exclusively to fatty acids as an energy source due to a deterioration of glucose transport into cells [49]. This is combined with the development of diastolic or systolic dysfunction in both animals and humans [50, 51, 52]. Reduced strength and rate of contraction were manifested both in the isolated mouse heart and in isolated papillary muscles [53, 54]. But the causes of the deterioration of myocardial contractility have not been identified.

In our experiments performed 2 weeks after the administration of streptozotocin (60 mg/kg), the level of glucose in the blood increased from 5.4 ± 0.1 mmol/l to 31.0 ± 1.4 mmol/l [55]. Echocardiographic and invasive LV studies showed a decrease in LV ejection fraction by 26% with unchanged LV end-diastolic volume and heart rate. However, all indicators of LV systole − LV systolic pressure, the maximal rate of its development, and the contractility index remained within the normal range, and a decrease in LV ejection fraction was apparently associated with a reduced LV maximal ejection rate by 28%. It is known that cardiac output is determined not only by the force of LV contraction but also by vascular resistance, the components of which are blood pressure and arterial rigidity. The latter is manifested through the elastic expansion of the ascending aorta at the time of LV ejection. In our experiments, blood pressure was normal, but arterial rigidity was significantly increased by almost 1.5 times (from 0.27 ± 0.01 to 0.42 ± 0.07 mmHg/μl, p < 0.01). An inverse correlation was established between LV maximal rate of ejection and arterial rigidity (r = − 0.69). Also specific for cardiomyopathy features, namely delayed relaxation and increased LV minimal diastolic pressure, have been observed.

The results of this work, which showed a deterioration in hemodynamics while maintaining normal myocardial contractility, are in contradiction with the data on reduced contractility in experiments in the isolated heart and papillary muscles [53, 54]. However, it should be kept in mind that these experiments were performed with standard Krebs solution in which glucose was the only energetic source while it is well known that at I-type diabetes the cardiomyocytes almost exclusively used fatty acids as an energetic source [49].

The change in the spectrum of energy substrates in type 1 diabetes raises the question of the state of energy metabolism in the myocardium of diabetic rats. In our experiments, the relationship between LV end-diastolic volume and the area covered by the volume-pressure curve for the cardiac cycle (PVA) reflecting the amount of oxygen consumption was studied with a restriction of the flow to the heart created by short-term clamping of the inferior vena cava [56]. In all experiments, the relationship between these parameters was close to linear, but in experiments in diabetic rat hearts, a reduction in LV end-diastolic volume was accompanied by a steeper PVA drop. This suggests that the myocardium of diabetic rats is more sensitive to the restriction of oxygen flow.

In the study of myocardial energy metabolism in diabetic rats, a reduction in the amount of adenine nucleotides by 21% and ATP content by 29% was found [56]. The formation of phosphocreatine was also impaired, as evidenced by the more than halved ratio of phosphocreatine to free creatine − in diabetes, it was 9 ± 3% and in the control 22 ± 4% (p < 0.05). The lactate content in the diabetic myocardium was increased (5.9 ± 0.8 versus 2.1 ± 1.0 μmol/g dry weight, p < 0.001) showing that some glycolytic processes are taking place. A negative correlation was found between the content of lactate and phosphocreatine (r = − 0.70). A reduction in phosphocreatine/ATP ratio was noted by 31P magnetic resonance spectroscopy in the hearts of patients with diabetes [57].

Thus, despite the sufficient supply of oxygen and energy substrates in the form of free fatty acids to the myocardium, the energetic metabolism of cardiomyocytes is altered and not always sufficient. The reason is undoubtedly related to the state of mitochondria in cardiomyocytes, this aspect is discussed in detail in the review [58]. Mitochondria are less efficient and produce less ATP with sufficient oxygen supply [59, 60] and are characterized by higher production of ROS and MDA [61]. This is especially pronounced in the interfibrillary mitochondria in which lower number of Krebs cycle proteins [62] and decreased cardiolipin content in their membranes [63] result in violation of ATP synthesis.

Relaxation slowing and increased LV diastolic pressure are indispensable companions of cardiomyopathy. Recently, it has been shown that myocardial relaxation can be accelerated and LV diastolic pressure is reduced with a restriction of inflow to the heart through a short-term (2–3 s) clamping of the inferior vena cava [64]. The use of this technique was accompanied by a reduction in blood pressure and LV diastolic volume. With the same values of LV volume (0.3 ml) and blood pressure (60 mm Hg), LV ejection fraction, maximal rate of LV ejection, and the index of arterial stiffness in diabetic hearts were already slightly different from control values [56]. But the relaxation time constant was still increased by 36%, and LV minimal diastolic pressure still remained elevated. These data once again confirm that in type 1 diabetes, myocardial contractility is not the main cause of the resulting systolic dysfunction.

In addition to contractility and relaxation, in these experiments, we also investigated myocardial extensibility. It is known that LV wall stretching during diastole occurs unevenly—at first, quickly then more slowly. Since the assessment of myocardial extensibility at each moment of diastole is still a very difficult task, we used an integrative indicator − the ratio of an increase in LV pressure during diastole, i.e. the difference between the final and minimal LV diastolic pressure, to the value of LV filling equally to the stroke volume. This is an indicator of LV diastolic stiffness; its value is inversely proportional to extensibility. The measurement of this index in the hearts of control animals showed its dependence on the heart rate and LV diastolic volume. It turned out that there is a high negative correlation between them (r = −0.89 and − 0.91, respectively). This means that increased LV diastolic volume or the increased heart rate are combined with a reduced value of the diastolic stiffness index, and, therefore, with increased myocardial distensibility. In this regard, a comparison of this indicator between control and diabetic hearts was carried out in a limited and equal range of LV diastolic volume and heart rate. The diastolic stiffness index in the diabetes group was 9.3 ± 1.0 mmHg/ml, and in the control group 13.8 ± 1.0 (p < 0.01). A lower index in the diabetes group means increased myocardial extensibility.

In this regard, in separate experiments of this series, a study of connectin (titin) was performed. Its content in the control and diabetic groups was the same and amounted to 15−16% of the content of heavy chains of myosin. However, the ratio of isoforms changed in the direction of increasing more distensible N2BA isoform: its content increased from 21 ± 1.1% in the control to 27 ± 1.0% in the diabetes group (p < 0.001). The results of real-time PCR showed a significant increase in the level of mRNA N2BA isoform by 38%, while for the N2B isoform it changed little. The degree of phosphorylation of connectin increased by 30 ± 7.5% (p < 0.05) mainly due to isoforms N2B (+ 15%) and NT (+ 12%). The predominance of the more distensible N2BA isoform, as well as an increase in its transcription, creates a lesser resistance during LV filling, but at the same time its contribution to relaxation decreases − after removing Ca⁺⁺ from myofibrils, less stiff connectin, due to its spring-like structure, may exert a lesser force for myosin filaments to return them to initial position.

To detect diastolic dysfunction in this model of diabetes, two series were performed (10 rats each), in one of which, after administering a dose of streptozotocin of 60 mg/kg, the experiments were performed after 1 week, and in the other, the dose was halved (30 mg/kg) but the period was increased to 2 weeks [56]. At the same time, in most experiments (2/3) with echocardiography, diastolic dysfunction was detected. In an invasive study of this group, it was found that LV ejection fraction was normal, despite a reduced maximum ejection rate of 34%, probably due to increased arterial elasticity. But LV end-diastolic volume was significantly smaller compared to the systolic dysfunction group and the control group (0.36+ 0.02 ml, 0.46 + 0.03 ml, and 0.43 + 0.02 ml, respectively, p < 0.05). Reducing the LV diastolic volume with a lower stroke volume ensures the preservation of the normal ejection fraction even under conditions of a reduced LV maximal ejection rate. This circumstance allows us to think that the ejection fraction does not always characterize myocardial contractility, but rather reflects the relationship between the ventricle and arterial resistance. This opinion coincides with the recently expressed idea [65, 66], according to which the reduction of LV chamber in combination with moderate hypertrophy is the main structural characteristic of diastolic dysfunction in this type of diabetes.

Since myocardial contractility was maintained at a normal level in both forms of heart dysfunction, the question arises − can LV dysfunction in type 1 diabetes be considered cardiomyopathy? After all, a decrease in the pumping function of the heart can be observed with preserved myocardial contractility with various valvular defects, limited venous inflow to the heart, etc., but these conditions are not considered cardiomyopathy. There is currently no universally accepted definition of diabetic cardiomyopathy, with several definitions used that cover the entire spectrum of diabetic heart disease from subclinical changes to outright heart failure [67].

5. Conclusion

The presented data indicate that in all types of cardiomyopathy considered, the CHF formation passes through the stage of diastolic dysfunction. Its main signs are delayed myocardial relaxation and endothelial dysfunction. Myocardial relaxation is a complex process that has active and passive components. The active component is due to volatile transport of Ca⁺⁺ from myofibrils into SR against a high concentration gradient. Both Ca⁺⁺ uptake into SR and exit from SR to myoplasma are under constant mitochondrial control due to the fact that part of Ca⁺⁺ ions enters mitochondria, where Ca⁺⁺ acts as an activator of some key enzymes of the Krebs cycle [68]. Since Ca⁺⁺ transport is energy dependent, any violation in ATP synthesis system will inevitably reduce Ca⁺⁺ inflow into SR, and hence a Ca⁺⁺ fraction released at next excitation. This process is carried out through a release of active oxygen species, a moderate amount of them activates Ca⁺⁺ transport and an excessive one disrupts it [69, 70].

The passive component of relaxation is represented by connectin (titin). Its spring-like structure is stretched when the ventricle is filling, and after contraction completion, it participates in mechanical return of sarcomere length to the original value. The weightiest argument in favor of connectin participation in relaxation process is the phenomenon of elastic recoil, observed in isolated hearts [71, 72, 73, 74]. A detailed study of connectin function made it possible to substantiate the idea of the “restorative force” [75, 76, 77], mobilized at enhanced compression of myofibrils. According to calculations, in rat cardiomyocytes within the length of sarcomeres of 1.6−2.1 μm, connectin is responsible for 90% of passive force and at least 60% of restorative force [75]. This is especially important for the hearts of small animals with a high heart rate and shortened diastolic pause. Our data showed that a change in the ratio of connectin isoforms to the direction of predominance of N2BA isoform favoring myocardial distensibility can reduce the degree of connectin participation in the relaxation process.

All types of cardiomyopathies considered are combined with an early change in the arterial tone. Isoproterenol and doxorubicin, the action of which is largely combined with the development of oxidative stress, suppressing the activity of endothelial NO synthase and increasing vascular tone, including coronary vessels. In diabetic cardiomyopathy, accompanied by hyperglycemia, the active factors of which are glucose glycation products, also increase the formation of reactive oxygen species and inhibit the activity of NO synthases, which leads to a stable increase in arterial rigidity.

The decreased myocardial contractility observed in isoproterenol and doxorubicin cardiomyopathy can be largely compensated by the inclusion of myocardial and systemic mechanisms. At the level of cardiomyocytes, the Starling mechanism is primarily included − an increase in myofibrillar stretching, largely realized by increasing the content of connectin N2BA isoform and phosphorylation of more elastic N2B isoform. An increased myofibrillar sensitivity to Ca++, usually occurring at decreased effective concentration, can also be mobilized. The systemic level of regulation aims to reduce peripheral resistance to facilitate LV ejection. As a result, at doxorubicin and isoproterenol cardiomyopathy, associated with decreased LV contraction, blood pressure decreases moderately, while at diabetic one it can remain normal. At the same time, to increase the inflow to the heart, the Parin reflex is mobilized − (“a decrease in peripheral resistance in the large circle is accompanied by a reciprocal increase in pressure in the small circle”). Systolic dysfunction is characterized by the same compensation mechanisms, but their mobilization is no longer sufficient to preserve the ejection fraction, and then the last compensation mechanisms are mobilized to maintain the cardiac output at an acceptable level − bradycardia, which prolongs the diastolic pause, and cardiac dilatation, which can further stretch the myofibrils. Each of these factors has a limited resource, used mainly by the autonomic nervous system. The degree of its mobilization, as well as the general condition of the body, largely predetermines the development of diastolic or systolic dysfunction in CHF.

Acknowledgments

This work has been supported by Russian Fund for Basic Research grant № 20-015-00027.

Conflict of interest

The author declares no conflict of interest.