1. Myostatin and Heart Morphology
The results on a relationship between myostatin expression and heart weight are not uniform. Myostatin overexpression decreased the whole heart and left ventricular size in mice, whereas its removal had an opposite effect
[1]. Interestingly, myostatin overexpression, specifically in skeletal muscles, decreased the heart mass only in male mice
[2]. According to Rodgers et al.
[3], hearts from adult male and female myostatin-null mice are larger than in wild-type mice. Both the heart and body weight of the neonates were heavier in the wild-type mice than in the myostatin-null mice, making the heart/body weight ratios similar in both groups. In myostatin-null mice, the left ventricular mass, internal diameters, and diastolic and systolic volume were larger than the respective values in the wild-type mice. These data would seem to indicate that lack of myostatin produces eccentric cardiac hypertrophy. In opposition to the above results, Cohn et al.
[4] did not find any differences in body and heart weight or heart/body ratio between myostatin-deprived and wild-type mice. Thus, the question arises whether distinct compensatory mechanisms can account for the variations in results seen across studies. Gaining a deeper comprehension of myostatin biology is essential in order to characterize its physiological importance in the heart.
2. Regulation of Heart Myostatin Expression
The available data clearly indicate that myostatin is involved in the regulation of cardiomyocyte growth, proliferation, and functioning (
Table 1). The presence of myostatin mRNA and protein have been shown for the first time in cardiomyocytes and Purkinje fibers of foetal and adult sheep, being significantly higher in the former
[5]. Similar observations were noticed in rat myocardium, wherein myostatin expression depended on the age of development. The mRNA and protein myostatin expression in foetal cardiomyocytes was low, subsequently increased markedly up to the tenth day of life, and then decreased in the adults to the control level. More specifically, myostatin mRNA expression in the adults was fifteen-fold lower and myostatin protein expression was two-fold lower as compared to the expression on the tenth day
[6]. Additionally, differences in myostatin expression between the heart ventricles were reported in growing piglets, which may reflect chamber-specific alterations in myocardium in response to workload. The expression in the left ventricle of 20-day-old piglets was three times higher than in newborn ones, whereas the expression in the right ventricle remained stable. The myostatin expression in the left ventricle of the newborn piglets was two times higher, and in the 20-day-old piglets was six times higher than in the right ventricle
[7]. These latter data are in opposition to the data obtained in sheep
[5]. It is difficult to say whether the difference in the direction of changes between sheep and the piglets should be ascribed to differences in the species or the time of sampling (i.e., adult sheep in
[5] and 20 day old piglets in
[7]). It seems likely that the growing level of myostatin in the left ventricle inhibits excessive activity of the genes responsible for the ventricle enlargement. To address the role of myostatin at early levels of cardiac development, exogenous myostatin was administered to cultured embryonic and early neonate cardiomyocytes. The inhibition of proliferation occurred in both cell lines, mainly via blockade in the G1-S phase of the cell cycle. The same study showed that myostatin inhibits protein synthesis in isolated cardiomyocytes
[6]. Several underlying pathways through which myostatin interferes with cardiac growth have already been described. These involve the inhibition of phenylephrine (α1 adrenergic receptor activator)-dependent protein synthesis and heart growth as well as phenylephrine-induced growth of cultured cardiomyocytes
[8] through the depression of p38 and the serine-threonine kinase Akt activity
[9]. In accordance with the above, in myostatin knockout mice this phenylephrine-induced heart growth was potentiated as compared to the wild-type mice
[8]. It was further proved that phenylephrine stimulates production of myostatin by cardiomyocytes
[8]. Exogenous myostatin inhibited the action of IGF-1 and its analogue LR3 on proliferation of the cultured cells
[3]. Altogether, these results clearly indicate that myostatin is responsible for inhibition of the hyperplastic growth, cardiomyocyte proliferation, and reduction in the rate of protein synthesis in growing heart.
Table 1. Factors affecting myostatin expression in myocardium and serum.
Cyclic stretch. Cyclic stretching resulted in several-fold elevation in mRNA and protein expression of myostatin in neonatal rat cultured cardiomyocytes, with a concurrent increase in IGF-1 secretion. In the next step, the stretch-induced secretion of myostatin was prevented by the inhibition of IGF-1. Therefore, it was concluded that IGF-1 was responsible for induction of myostatin expression in the cardiomyocytes by cyclic stress
[10].
Exercise. Eight weeks of either resistance or moderate intensity aerobic training did not affect the heart myostatin mRNA expression in rats. However, it markedly elevated heart mRNA follistatin expression. As a result, the ratio of follistatin mRNA/myostatin mRNA increased considerably
[11]. Because follistatin is an inhibitor of myostatin, such a shift in follistatin/myostatin balance would facilitate the heart hypertrophy in the course of training. On the other hand, four weeks of swimming training was shown to increase mRNA myostatin expression in the heart by 74%
[12]. Such an elevation in the myocardial level of myostatin could exert a direct inhibitory effect on development of heart hypertrophy during training. However, it is difficult to explain the reasons for the difference between the two reports.
Angiotensin II. Angiotensin II exerts powerful effects on the cardiovascular system, for instance by promoting the development of cardiac hypertrophy. Initially, it serves as a compensatory mechanism to maintain hemodynamic homeostasis, although long-term angiotensin II upregulation becomes maladaptive and results in pathological cardiac hypertrophy
[29]. It has been shown that angiotensin II promotes expression of myostatin in isolated rat neonatal cardiomyocytes. The process was mediated by the activation of the p38 MAP kinase and MF-2 pathways. This would suggest that myostatin is a negative feedback regulator of angiotensin II-induced heart hypertrophy
[17].
Insulin-like growth factor 1 (IGF-1). IGF-1 increases the expression of myostatin in cultured cardiomyocytes
[10]. Gaussin and Depre
[30] suggested that because IGF-1 stimulates both cardiac growth and cardiac myostatin production, myostatin is a cardiac chalone of IGF-1.
miRNA-208a. miRNA-208a is present in the heart and is involved in the pathogenesis of multiple cardiovascular diseases. Its overexpression in mouse heart reduces the expression of myocardial myostatin protein. Deletion of miRNA-208a leads to the elevation of myostatin expression
[18]. This would seem to indicate that myostatin is likely to be involved in the action of miRNA-208a. On the other hand, removal of myostatin increases the expression of miRNA-208a both in vivo and in vitro. Myostatin treatment reduces the expression of miR-128 induced by the aorta coarctation and by angiotensin II in isolated cardiomyocytes
[31].
3. Myostatin and Heart Function
The available data provide convincing proof that myostatin is needed for normal heart function. In experiments by Rodgers et al.
[3], stroke volume, fractional shortening, and ejection fraction were all lower in myostatin-null mice than in wild-type animals. Reduction in end systolic volume, elevation in fractional shortening, and ejection fraction after treatment with isoproterenol (β-adrenergic agonist) were much greater in the myostatin-null mice than in their wild-type counterparts. These differences would be due to increased calcium release from sarcoplasmic reticulum in the myostatin-null mice. This establishes a mechanism for a myostatin-related maintenance in cardiac output, which otherwise could be expected to decline with high heart rate. On the contrary, Butcher et al.
[32] did not find a difference in terms of the response of fractional shortening and ejection fraction after stimulation with isoproterenol or blockade of β-adrenergic receptors (with propranolol) between control and myostatin knockout mice.
Selective deletion of myostatin in cardiomyocytes of adult mice increases lethality, and leads to heart failure and ventricular hypertrophy. However, in this model the skeletal muscle mass, myocyte area, and myostatin expression remained intact
[33]. Long-term overexpression of myostatin in mouse heart reduces the ejection fraction and stroke volume, elevates the end systolic volume and end diastolic volume, and induces development of fibrosis
[34]. Further studies of this group revealed that myostatin activates the TAK1-MKK3/p38 signaling pathway. Activation of this pathway increases production of collagen 1. A very important observation of the two studies was that within 6 weeks after the removal of myostatin from cardiomyocytes, its level in the heart and serum returned to normal. It was further shown that this adaptive mechanism resulted from the increased production of myostatin by non-cardiomyocyte cells present in the heart
[33][34]. Other models of adaptation to prevent undesirable consequences of myostatin loss can be noticed in germ-line mutants unable to synthesize myostatin via non-cardiomyocytes. One possible mechanism could be a rise in GDF11 due to functional overlap between the two proteins (myostatin and GDF11)
[35], meaning that only limited effects on cardiac function were observed. There are data indicating that removal of myostatin does not result in heart hypertrophy or affect cardiomyocyte size. Furthermore, it does not attenuate cardiac fibrosis in the dystrophin-deficient mdx mouse model of Duchenne muscular dystrophy. The major conclusion of the latter study was that myostatin does not function as a crucial regulator of myocardial growth and regeneration in cardiac muscle in vivo. It was claimed that the difference in the results with other reports was due to the use of better equipment to evaluate heart function, specifically, a high-resolution echocardiography apparatus
[4].
4. Myostatin and Heart Metabolism
The foetal heart mostly uses carbohydrates as an energy fuel, whereas adult hearts preferably use free fatty acids. Hypertrophied adult hearts are overly reliant on glucose as an energy source. This manifests in acceleration of glycolysis, increased uptake of glucose, and increased production of lactate
[36]. APMK (AMP-activated protein kinase) is the key regulator of energy homeostasis in the heart
[37]. Myostatin has been shown to modulate cardiac energy substrate reliance and limit the risk of heart failure. Inactivation of myostatin in adult mouse cardiomyocytes resulted in a nearly two-fold elevation of phosphorylation and activation of AMPK. As a consequence, an elevation in glucose uptake and glycolysis in the myocytes was noted. On the contrary, myostatin overexpression resulted in strong inhibition of AMPK phosphorylation and ultimately prevented a metabolic switch in the direction of glycolysis and glycogen accumulation
[33]. Based on the above, a potential cardioprotective role of myostatin can be ascribed to the suppression of metabolic reprograming towards the fetal metabolic pattern that occurs during cardiac hypertrophy and maintenance of a mature aerobic energy metabolism.
Myostatin can counteract the pathological hypertrophic effects through the modulation of protein turnover in the myocardium. Exogenous myostatin increases proteolysis and reduces protein synthesis in cultured cardiomyocytes. A closer look at the mechanism of myostatin-activated proteolysis revealed that it phosphorylates Smad2 and Smad3 proteins, leading to the activation of proteolysis and the process of autophagy. Decreased protein synthesis was mediated by inhibition of Akt, Foxo3a, and P70S6K phosphorylation
[38]. These data clearly indicate that myostatin is involved in regulation of the myocardial metabolism.
5. Myostatin and Heart Pathology
There are data available from animal and human studies indicating changes in the myocardial metabolism of myostatin in heart pathology as well effects of these changes on distant tissues, mainly skeletal muscle. The most investigated heart disorders are myocardial infarction, myocardial hypertrophy, and chronic heart failure.
6. Myocardial Infarction
Experiments on animals. Coronary circulation is very poorly equipped in anastomoses; thus, collateral circulation does not play a protective role against cardiac ischemia. As a result, occlusion of a coronary artery results in acute ischemia and necrosis of the area fed by the occluded artery. In light of the data on the role of myostatin in heart physiology, it could be expected that myostatin might affect different parameters of the heart function after myocardial infarction. Indeed, the data obtained thus far both in animal experiments and from human beings confirms this assumption. In the reports cited below, the experimental myocardial infarction was produced by ligation of the left anterior descending coronary artery (LADA). Shrama was the first to show that the expression of myostatin protein in sheep was upregulated in the myocardium around the infarcted area
[5]. In rats, 8 weeks after LADA ligation
[19] myocardial myostatin mRNA expression increased only insignificantly; however, the expression of myostatin protein increased more than four times compared to the respective control values. Interestingly, four weeks of training prevented this elevation in the protein expression. Thus, the suppression of myostatin can be one of the factors of anti-catabolic effects of exercise training in chronic heart failure. Another protocol was employed by Rodgers et al.
[3]. They studied myostatin expression in different parts of left ventricle 4 weeks after ischemia/reperfusion procedure. The ischemia lasted for 30 min, and afterwards the samples were harvested from infarcted, border-infarcted, and healthy regions of the myocardium. The expression of myostatin in different locations was similar, and was not affected by ischemia. However, the expression of follistatin was elevated only in the infarcted area. Such a shift in the myostatin/follistatin balance might indicate a reduction in the role of myostatin during recovery from the post-ischemia/reperfusion injury. Furthermore, a role of myostatin in post-infarction recovery was studied in myostatin-null mice 1 and 28 days after the infarction
[39]. The infarct size in myostatin-null mice was similar to the infarct size in wild-type mice. However, deletion of myostatin reduced the post-infarct mortality rate by 20%. The heart rate in the myostatin-null group 28 days after the infarct was higher than in the wild-type mice. Among the parameters characterizing left ventricular function, fractional shortening and ejection fraction were reduced to a similar degree in both groups in one day after the infarct. However, 28 days after the infarction, both parameters normalized in the myostatin-deprived group only. Moreover, lack of myostatin partially prevented deposition of collagen and scarring, and preserved a greater viable area of the myocardium than in the wild-type mice. These results indicate that lack of myostatin is beneficial, as it partially reduced the consequences of myocardial infarction.
A large amount of data on the level of myostatin and other compounds functionally related to myostatin in rat hearts and serum after myocardial infarction were presented by Castillero et al.
[20]. These are the only data examining expression of these factors at several time points beginning from 10 min up to 2 months after myocardial infarction. As these data are unique, more details are presented here. It was found that the myostatin level in the heart increased as early as 10 min after the infarction, and partially normalized over the next hours. It markedly increased again 24 h after the infarction and remained elevated thereafter. The heart level of follistatin remained stable. Cardiac IGF-1 was elevated in the examined period of time, though this elevation was significant only one month after the infarct. Cardiac pAkt/Akt ratio increased after 24 h, remained elevated for 1 month, and then returned to the level in the sham-operated mice. Collectively, the upregulation of cardiac myostatin expression would counteract the pro-hypertrophic action of IGF-1 and pAkt/Akt ratio after infarction. Moreover, MAP kinase and P-p38/p38 ratio were elevated at 10 min and one hour, and then from one week onwards. The cardiac pSmad2,3/Smad2,3 ratio fluctuated, being significantly elevated at 2, 6, and 12 h as well as at one month after the infarct, indicative of the activation of muscle atrophy. Therefore, this research showed that infarction changes the balance between pro- and anti-hypertrophic factors in the myocardium. Interestingly, these alterations are not confined to the heart, as an elevation in the serum myostatin level was observed as well. Most likely, this was a consequence of activation of its production in the myocardium. Serum levels of myostatin increased already only ten minutes after the infarct, and remained elevated until the end of observation. The serum follistatin level was elevated from 24 h onwards, while the circulating IGF-1 concentration was not affected by the infarction. These data clearly indicate that changes in the levels of the examined parameters, and as such their effects on the heart’s post-infarct recovery, appear very early and are long-lasting.
Human data. Data on the myostatin serum concentration after myocardial infarction are available in humans as well. In a study by Oliveira et al.
[21], 102 patients with STEMI were included. STEMI was defined as ischemic chest pain with elevation of the ST segment of the electrocardiogram. Myocardial infarction reduced serum myostatin concentration as compared to the healthy control group. The mortality rate among patients with lower serum myostatin concentration was higher than among those with less reduced levels. No association between serum myostatin and creatine phosphokinase (CK-MB) peak concentration or ejection fraction was shown. The reason for the difference in the direction of the change in the post-infarct serum myostatin concentration between rats (increase) and humans (reduction) remains an open question.
The serum troponin I peak is regarded as a reliable indicator of infarction size. Meloux et al.
[40] examined an association between serum myostatin and troponin I peak concentration in 296 patients with acute myocardial infarction. The serum myostatin data were positively correlated with troponin I peak data. Researchers are not aware of any other results comparing the serum myostatin level and troponin I peak as markers of the infarct size. The data presented by this research are encouraging. Undoubtedly, further studies are needed to confirm the usefulness of serum myostatin levels in evaluating infarct size. In the same study
[40], the serum myostatin concentrations were higher in patients with ventricular tachycardia or fibrillation acquired in the hospital. This would seem to indicate involvement of myostatin in cardiac pathologies. Obviously, more data are needed on the relationship between myostatin and myocardial infarction in humans.
7. Myocardial Hypertrophy
Usually, two models of heart hypertrophy are used in experimental animals. In one model, the abdominal aorta is transversely constricted. This forces the left ventricle muscle to exert more work to pump blood through the narrowed vessel. As a result, left ventricle hypertrophy develops. In another model, an aortal–caval shunt is made. It increases the blood volume moving to the heart, which finally reaches the left ventricle. In turn, more work on the part of the left ventricle is required to pump the blood out to the aorta. As such conditions last for longer periods of time, the left ventricle muscle mass increases gradually, i.e., hypertrophy develops. Persistent increased workload finally leads to chronic cardiac failure. Shyu at al.
[23] produced cardiac hypertrophy in rats using the volume-overload model for 4 weeks. The shunt increased heart weight and heart rate, left ventricular end-diastolic pressure, left ventricular end-diastolic dimension, and end-systolic dimension. At the same time, myostatin protein and mRNA expression in the myocardium increased more than twofold. Treatment with carvedilol (a nonselective blocker of β-adrenergic receptors and α1-adrenergic receptors) blocked the elevation of myostatin mRNA expression in the heart and skeletal muscle produced by the shunt. These latter data suggest that the adrenergic system is involved in the activation of myostatin expression in the muscles. No further data on this topic are available.
Hypertension. Hypertension is another cause of the heart hypertrophy. Hypertension requires more work by the left ventricular muscle to pump blood out to the aorta against the increased blood pressure in the vessel, leading to hypertrophy of the muscle. Spontaneously hypertensive rats with heart failure showed a hypertrophied and dilated left atrium and left ventricle chamber, increased myocyte diameter, and increased myocardial interstitial collagen content. Left ventricular dysfunction manifested with decreased midwall fractional shortening. Concomitantly, a reduction in myocardial protein myostatin and follistatin expression in the left ventricle was demonstrated. The expression of both myostatin and follistatin was correlated with the functional parameters of the left ventricle
[22]. These results are at odds with data reporting elevated myostatin expression in hypertrophied hearts
[23]. The only cause for the difference between the two reports would different pathomechanisms of spontaneous hypertension compared to the experimental hypertrophy. It is likely that other factors accompanying spontaneous hypertension/hypertrophy could reverse the direction of the myostatin expression.
As indicated above, experimentally hypertrophied myocardium produces more myostatin. This raises the question of how prolonged exposure to elevated endogenous myostatin levels influences the myocardium itself as well as the skeletal muscles. An extensive study on this topic was performed by Qi et al.
[31]. Cardiac hypertrophy was produced in rats by a reduction in the abdominal aorta diameter. The function of the heart was evaluated ten weeks later. Myostatin deletion augmented hypertrophy and autophagy produced by aorta coarctation or angiotensin II. This was accompanied by a reduction in the parameters of the heart function. Mechanistic studies on isolated cardiomyocytes revealed that myostatin silencing augmented the hypertrophic effect of angiotensin II. On the contrary, treatment with myostatin dramatically reduced both hypertrophy and autophagy of the cells, partially via inactivation of the AMP-activated kinase (AMPK)/mammalian target of rapamycin (mTOR) pathway and activation of the peroxisome proliferator activated receptor gamma (PPARγ)/nuclear factor κB (NF-κB) pathway. Myostatin treatment reduced expression of miR-128 induced by aorta coarctation and angiotensin II in isolated cardiomyocytes. The final conclusion was that increased production of myostatin is a self-defence mechanism in the hypertrophic heart that leads to the attenuation of hypertrophy, heart dysfunction, and autophagy.