Cardiac Biomarkers and Autoantibodies in Endurance Athletes: Potential Similarities with Arrhythmogenic Cardiomyopathy Pathogenic Mechanisms

he “Extreme Exercise Hypothesis” states that when individuals perform training beyond the ideal exercise dose, a decline in the bene cial effects of physical activity occurs. This is due to signi cant changes in myocardial structure and function, such as hemodynamic alterations, cardiac chamber enlargement and hypertrophy, myocardial in ammation, oxidative stress, brosis, and conduction changes. In addition, an increased amount of circulating biomarkers of exercise- induced damage has been reported. Although these changes are often reversible, long-lasting cardiac damage

cial effects of physical activity occurs. This is due to signi cant changes in myocardial structure and function, such as hemodynamic alterations, cardiac chamber enlargement and hypertrophy, myocardial in ammation, oxidative stress, brosis, and conduction changes. In addition, an increased amount of circulating biomarkers of exercise- induced damage has been reported. Although these changes are often reversible, long-lasting cardiac damage may develop after years of intense physical exercise. Since several features of the athlete's heart overlap with arrhythmogenic cardiomyopathy (ACM), the syndrome of “exercise-induced ACM” has been postulated. Thus, the distinction between ACM and the athlete's heart may be challenging. Recently, an autoimmune mechanism has been discovered in ACM patients linked to their characteristic junctional impairment. Since cardiac junctions are similarly impaired by intense physical activity due to the strong myocardial stretching, we propose in the present work the novel hypothesis of an autoimmune response in endurance athletes. This investigation may deepen the knowledge about the pathological remodeling and relative activated mechanisms induced by intense endurance exercise, potentially improving the early recognition of whom is actually at risk. Keywords: arrhythmogenic cardiomyopathy; athletes; autoantibodies; physical exercise; desmo- somes 1. Introduction Although physical exercise is recommended for the maintenance of a healthy lifestyle and the reduction of cardiovascular disease incidence [1,2], prolonged and intense activ- ity can be deleterious for cardiac structure and function. It can acutely and transiently increase sudden cardiac death (SCD) and myocardial infarction risk in susceptible indi- viduals [3]. Increased myocardial brosis [4,5], coronary artery calci cation [6], and atrial brillation [7,8] have been reported in endurance athletes. Since endurance athletes exceed the usual recommendations for exercise by 15-fold to 20-fold, the “Extreme Exercise Hypothesis” has been proposed to explain how the bene cial effects of physical activity may plateau or decline when individuals perform training be- yond the ideal exercise dose [9,10]. As depicted in Figure, the dose–response relationship Int. J. Mol. Sci.2021,22, 6500.

Int. J. Mol. Sci.2021,22, 6500 2 of 15 between exercise training volumes and health risk is described by a J-shaped (or U-shaped) curve [2,9]. To date, the exact amount of exercise able to impair the cardiovascular system has not been de ned. The metabolic equivalents of task (METs) method is recognized as useful to evaluate the functional capacity or exercise tolerance of an individual [11]. One MET is the amount of oxygen consumed at rest and is equal to 3.5 mL of oxygen per kilogram per minute [11]. Most reports have de ned “vigorous exercise” as needing at least six METs, although the maximal individual capacity could in uence this threshold [3].Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 2 of 15 Since endurance athletes exceed the usual recommendations for exercise by 15-fold to 20-fold, the “Extreme Exercise Hypothesis” has been proposed to explain how the ben- eficial effects of physical activity may plateau or decline when individuals perform train- ing beyond the ideal exercise dose [9,10]. As depicted in Figure 1, the dose–response rela- tionship between exercise training volumes and health risk is described by a J-shaped (or U-shaped) curve [2,9]. To date, the exact amount of exercise able to impair the cardiovas- cular system has not been defined. The metabolic equivalents of task (METs) method is recognized as useful to evaluate the functional capacity or exercise tolerance of an indi- vidual [11]. One MET is the amount of oxygen consumed at rest and is equal to 3.5 mL of oxygen per kilogram per minute [11]. Most reports have defined “vigorous exercise” as needing at least six METs, although the maximal individual capacity could influence this threshold [3]. Figure 1. Schematic representation of the “Extreme Exercise Hypothesis”. A J-shaped (or U- shaped) curve describes the dose–response relationship between exercise training volumes and health risk. When the threshold of optimal exercise dose (red point) is exceeded, the health bene- fits of training can be reduced. Adapted from [9]. Most exercise-associated adverse effects often occur in subjects with occult or diag- nosed structural cardiac diseases [3,12]. Among young individuals,

A J-shaped (or U- shaped) curve describes the dose–response relationship between exercise training volumes and health risk. When the threshold of optimal exercise dose (red point) is exceeded, the health bene- fits of training can be reduced. Adapted from [9]. Most exercise-associated adverse effects often occur in subjects with occult or diag- nosed structural cardiac diseases [3,12]. Among young individuals, the concomitant pathological conditions are often hereditary or congenital cardiovascular abnormalities, such as arrhythmogenic cardiomyopathy, hypertrophic cardiomyopathy, coronary artery anomalies, and bicuspid aortic valve [13–15]. Among older subjects who die during phys- ical exercise, coronary artery disease is the most frequent pathological finding [16,17]. SCD overall incidence during exercise is estimated at 1:50,000 [18–20]. Exercise-re- lated SCD seems to depend on the interaction between the physical activity acute trigger and an underlying disease, but it can be further elicited by other concomitant processes, including emotional stress, hemodynamic changes, and impaired parasympathetic tone [3]. It has been reported that physical activity may increase the risk of SCD by 2.5 times [21]. In the present review, we summarize the current knowledge about exercise-induced cardiac alterations and circulating biomarkers of damage. In addition, we propose a novel hypothesis of an autoimmune response in endurance athletes, based on the analogies with arrhythmogenic cardiomyopathy patients. Figure 1. Schematic representation of the “Extreme Exercise Hypothesis”. A J-shaped (or U-shaped) curve describes the dose –response relationship between exercise training volumes and health risk. When the threshold of optimal exercise dose (red point) is exceeded, the health bene ts of training can be reduced. Adapted from [9]. Most exercise-associated adverse effects often occur in subjects with occult or di- agnosed structural cardiac diseases [3,12]. Among young individuals, the concomitant pathological conditions are often hereditary or congenital cardiovascular abnormalities, such as arrhythmogenic cardiomyopathy, hypertrophic cardiomyopathy, coronary artery anomalies, and bicuspid aortic valve [13–15]. Among older subjects who die during physical exercise, coronary artery disease is the most frequent pathological nding [16,17]. SCD overall incidence during exercise is estimated at 1:50,000 [18–20]. Exercise-related SCD seems to depend on the interaction between the physical activity acute trigger and an underlying disease,

such as arrhythmogenic cardiomyopathy, hypertrophic cardiomyopathy, coronary artery anomalies, and bicuspid aortic valve [13–15]. Among older subjects who die during physical exercise, coronary artery disease is the most frequent pathological nding [16,17]. SCD overall incidence during exercise is estimated at 1:50,000 [18–20]. Exercise-related SCD seems to depend on the interaction between the physical activity acute trigger and an underlying disease, but it can be further elicited by other concomitant processes, including emotional stress, hemodynamic changes, and impaired parasympathetic tone [3]. It has been reported that physical activity may increase the risk of SCD by 2.5 times [21]. In the present review, we summarize the current knowledge about exercise-induced cardiac alterations and circulating biomarkers of damage. In addition, we propose a novel hypothesis of an autoimmune response in endurance athletes, based on the analogies with arrhythmogenic cardiomyopathy patients. 1.1. The Athlete's Heart Vigorous physical exercise is associated with signi cant changes in myocardial struc- ture and function. The athlete's heart has to sustain a higher cardiac output during maximal effort than untrained hearts. Thus, it is subjected to a physiological remodeling that allows its greater resistance during intense activity and suf cient oxygen delivery to exercising muscles. This physiological response is known as the Frank–Starling Mechanism or “law of the heart” [22].

Int. J. Mol. Sci.2021,22, 6500 3 of 15 Sympathetic activation is responsible for the augmented cardiac output through heart rate modulation. The heart rate spans from <40 bpm at rest to >200 bpm in a young maximally exercising individual. The stroke volume may signi cantly increase with sustained training because of the higher ventricular end-diastolic volume and sympatheti- cally mediated end-systolic volume reduction [23]. The hemodynamic changes parallel cardiac chamber enlargement and hypertrophy [24–26]. These cardiac adaptations may mimic those of a diseased heart, but in most cases systolic and diastolic functions are preserved [27,28], although a transient reduction in left ventricular (LV) ejection fraction (EF) has been reported after more than 6 h of continuous exercise [29,30]. In many cases, the right ventricular (RV) function seems to be more compromised by prolonged exercise than left ventricular one, possibly for the thinner-walled structure of RV [31–33]. Indeed, RV wall stress increases more than in the LV during exercise, producing higher pressure load on the RV, not compensated by a suf cient volume increase and myocardial thickening [34]. This depends on the increase in pulmonary artery pressure relative to systemic vascular pressure, necessary to guarantee the requested cardiac output [35]. However, at equal exercise loads, the RV response shows high interindividual variability due to differential adaptations of the pulmonary circulation. Indeed, a higher vascular reserve corresponds to enhanced maximal exercise capacity [36]. In addition, conduction alterations are com- mon, and they are usually mediated by parasympathetic activity and/or sinoatrial node slowing [37]. Endurance athletes are often affected by bradyarrhythmias, such as sinus bradycardia, junctional bradycardia, and rst-degree atrioventricular block [38]. Trained athletes can experience premature beats and non-sustained ventricular tachycardia, usu- ally of benign etiology and without long-term consequences [39,40], although no higher prevalence if compared to sedentary individuals has been assessed [41]. The main prob- lematic effect of intense physical activity is atrial brillation, which has been reported more frequently among athletes than in sedentary individuals [42,43]. Syncope often manifests in the immediate post-exercise period due to neurocardiogenic mechanisms based on a sudden reduction in venous return. When syncope manifests

no higher prevalence if compared to sedentary individuals has been assessed [41]. The main prob- lematic effect of intense physical activity is atrial brillation, which has been reported more frequently among athletes than in sedentary individuals [42,43]. Syncope often manifests in the immediate post-exercise period due to neurocardiogenic mechanisms based on a sudden reduction in venous return. When syncope manifests during exercise, it can be due to malignant arrhythmias, structural cardiac disease, or myocardial ischemia, which have to be thoroughly evaluated [44]. Although these alterations are often reversible, long-lasting cardiac damage may develop after years of intense physical exercise [45,46]. Furthermore, after prolonged endurance exercise, myocardial in ammation, oxidative stress, and brosis have often been reported, representing a substrate for life-threatening arrhythmias [5,10,47]. During exercise, increased metabolic processes with an augmented oxygen uptake may induce a mitochondrial electron “leakage” and the consequent production of reactive oxygen species (ROS) [48,49]. Moreover, the activation of immune and in ammatory responses due to exercise-induced muscle injury may generate high amounts of ROS [50,51]. To possibly counterbalance the oxidative damage, an increase in antioxidant defenses, through the activation of antioxidant enzymes [52], has been reported in response to high volumes of exercise [48]. Thus, it seems that oxidative stress does not occur below a certain threshold of intensity, but only when exercise is strenuous [53]. The higher amount of oxidative stress can increase oxidation of different molecules, causing their damage. For example, acute bouts of exercise can increase LDL oxidation [ The oxidative radicals could impair cardiomyocyte membrane permeability, concur- ring with mechanical stress in cardiomyocyte remodeling [58,59]. In addition, ROS may inhibit glycolysis, producing a perturbation in calcium homeostasis, which could lead to myocardial dysfunction [60]. Exercise-induced myocardial brosis patterns are various and differ according to the age of the athletes [61–63]. Fibrosis is often found near the interventricular septum, especially in middle-aged and older athletes, and near the right ventricular insertion points, mainly in young athletes [4,5,61]. More rarely, a sub-endocardial ischemic pattern, a sub- epicardial pattern, and extensive mid-wall and diffuse brosis could be detected. Since only

differ according to the age of the athletes [61–63]. Fibrosis is often found near the interventricular septum, especially in middle-aged and older athletes, and near the right ventricular insertion points, mainly in young athletes [4,5,61]. More rarely, a sub-endocardial ischemic pattern, a sub- epicardial pattern, and extensive mid-wall and diffuse brosis could be detected. Since only

Int. J. Mol. Sci.2021,22, 6500 4 of 15 speci c brotic patterns have been associated with ventricular arrhythmias and adverse cardiac events, the clinical and prognostic signi cance of myocardial brosis in athletes is yet to be determined [63]. The dose of exercise has been reported to be associated with the extent of brotic substitution in few studies [5,9,64]. The impact of gender on the ventricular response to exercise constitutes a relevant open issue. Although male versus female differences have a strong impact on cardiovascular disease pathogenesis [65–67], female athletes are often underrepresented in studies of cardiac adaptation to exercise. Few studies on this topic demonstrated a similar cardiac remodeling and prevalence of arrhythmic events in both male and female athletes [68,69]. However, RV performance during exercise seems to be enhanced in women when compared to men [68]. Conversely, a recent study reported that LV remodeling is more common in males, whereas RV remodeling mainly concerns females [70]. Thus, further investigations are needed to clarify the effective impact of gender. 1.2. Effects of Myocardial Stretching During intense exercise, the stretching of the myocardium activates different intrinsic physiologic mechanisms to adequately respond to this stimulus [71], through the so-called mechanoelectric feedback, which is able to transduce the mechanical stimulus into an electrical signal [72,73]. Each component of the heart seems to perceive mechanical stimuli, activating in- tracellular pathways that mediate several responses [74,75]. These pathways are often activated without binding of extracellular mediators [76]. An example is stretch-activated channels, able to modulate their permeability to ions and, consequently, electrical and mechanical properties of the myocardium [77]. This response often depends on protein phosphorylation. For example, calcium channel phosphorylation, by intensifying calcium transient, improves the contractile function [78]. Besides ion channels, other cardiomyocyte proteins concur in stretch-activated mechanisms: troponin I phosphorylation increases contractility, owing to the reduction of myo lament calcium sensitivity [79]; titin, both functioning as a mechanosensor and a molecular target, can trigger downstream signaling and modulate myocardial tension and sarcomeric length [80–82]. A central role in mechanosensing is played by intercalated discs, required to maintain mechanical and electric coupling

cardiomyocyte proteins concur in stretch-activated mechanisms: troponin I phosphorylation increases contractility, owing to the reduction of myo lament calcium sensitivity [79]; titin, both functioning as a mechanosensor and a molecular target, can trigger downstream signaling and modulate myocardial tension and sarcomeric length [80–82]. A central role in mechanosensing is played by intercalated discs, required to maintain mechanical and electric coupling between cardiomyocytes [83]. The two main structures of intercalated discs, fascia adherens junctions and desmosomes, contribute to adaptive responses to stretch [84] due to their connection with cytoskeletal actin and intermediate laments, respectively [83]. In volume overload conditions, as during intense physical activity, the intercalated discs undergo dynamic changes [85]. For example, N-cadherin (N-CAD), one of the main proteins of fascia adherens junctions, is upregulated following mechanical stretch and elicits changes in cardiomyocyte shape, myo brillar organization, and function [86]. On the contrary, N-CAD downregulation precludes the correct formation of intercalated discs, provoking cardiac morphological and functional defects [87]. Similarly, desmosomal protein loss impairs mechanotransduction responses [83]. For example, the deletion of the desmosomal protein desmoglein 2 (DSG2), necessary to assembly the extracellular domain, alters cell adhesion and signaling. It provokes the upregulation of heart failure markers, brosis, biventricular dilation and dysfunction, and death [88–90]. In general, desmosome de ciency leads to cardiomyocyte inability to appropriately face high mechanical stress, resulting in myocyte detachment and tissue remodeling [84]. Moreover, gap junctions, prominently localized at intercalated discs, mediate electrical propagation and are thus crucial to excitation and contraction [91]. They are composed of connexins, among which connexin 43 (CX43), the most important in the myocardium [91]. Physical exercise may affect gap junction remodeling, leading to CX43 expression down- regulation during acute exercise, as demonstrated in a murine model [92], and to a possible consequent impairment of electrical conduction [93].

Int. J. Mol. Sci.2021,22, 6500 5 of 15 Additionally, costamere proteins, which are responsible for the connection between the contractile apparatus and extracellular matrix, as integrins, are involved in mechan- otransduction and can be compromised when subjected to mechanical stress [94–97]. All these modi cations generally have an adaptive meaning, but, depending on the strength of the stimulus, their nature, and the individual's genetic background, they can result in maladaptive pathological remodeling [98], with mechanoelectric feedback dysfunction [72] and consequent arrhythmias, cardiac hypertrophy, and heart failure [74,84]. Indeed, when the physical activity is prolonged and intense, the impairment of these processes may provoke an altered cellular response and possibly heart disease. 1.3. Circulating Biomarkers of Exercise-Induced Damage The changes induced by endurance exercise are associated with several circulating biomarker increases. These elevations are usually modest and transient, but their clinical implications are not fully elucidated. As for cardiac damage biomarker, cardiac troponin (cTn) levels signi cantly increase after only 30 min of intense physical activity [99], reaching higher levels in younger and untrained individuals [100], concomitant with cardiovascular risk factors [101], greater exercise duration and intensity [32,101–105], and dehydration [106]. cTn release likely depends on exercise-induced cardiomyocyte necrosis, or the changes in membrane perme- ability caused by intense activity could determine the leakage of unbound troponin [107]. Further studies are needed to understand the mechanisms mediating its elevation [59,108]. Usually, cTn levels return to baseline within 72 h [109,110], and any cardiac dysfunction associated with increased cTn has been reported transient [59]. B-type natriuretic peptide (BNP) and its cleaved form NT-proBNP are secreted in re- sponse to cardiomyocyte stress produced by volume or pressure overload [111]. Thus, they can increase after endurance exercise [32,112–117], but return to baseline within72 h [ . Exercise duration [104,115], age [113,118], and poor physicalpreparation [ can impact on BNP and NT-proBNP elevation. Creatine kinase MB (CKMB), belonging to myocardial infarction biomarkers, can also be increased after intense activity, but it possibly originates more from skeletal muscle damage than from myocardial injury [120]. Moreover, typical brosis biomarkers have been associated with intense physical exercise. Soluble

[ . Exercise duration [104,115], age [113,118], and poor physicalpreparation [ can impact on BNP and NT-proBNP elevation. Creatine kinase MB (CKMB), belonging to myocardial infarction biomarkers, can also be increased after intense activity, but it possibly originates more from skeletal muscle damage than from myocardial injury [120]. Moreover, typical brosis biomarkers have been associated with intense physical exercise. Soluble suppression of tumorigenicity 2 (sST-2) concentrations exceed the up- per reference value after endurance activity, reaching higher levels as exercise intensity increases, but its complete normalization occurs within 48 h [121]. Tissue inhibitors of matrix metalloproteinase type I (TIMP-1), carboxy-terminal telopep- tide of collagen type I (CITP), and carboxy-terminal propeptide of collagen type I (PICP) are other circulating markers of collagen synthesis and degradation that are augmented in endurance athletes [59]. Similarly, galectin-3 resting levels are greater in athletes than controls, and further increase after physical activity, possibly produced mainly by skeletal muscle [122]. Indeed, no correlations with cardiac function have been detected [122]. For what concern oxidative stress markers, 13- and 9-hydroxy-octadecadienoic acid (13-HODE and 9-HODE), known oxidized linoleic acid metabolites, signi cantly increase immediately post-exercise, but their plasma concentrations return to baseline levels within 24 h [123]. Their production could be linked to lipoxygenase activation in response to cell injury [124]. Moreover, lipid peroxidation increases after endurance exercise, as demonstrated by higher levels of malondialdehyde (MDA) [117,125] and F(2)-isoprostanes [57]. In both cases, the augmentation is transient. Similarly, the heat shock proteins Hsp70 and Hsp72, known in ammation markers, are upregulated in athletes' serum after physical exercise [126–128]. As for the majority of the exercise-induced circulating biomarkers, the increase is rapid but transient. Higher

This review explores cardiac biomarkers and potential autoimmune responses in endurance athletes.