Emerging mitochondrial signaling mechanisms in cardio-oncology: beyond oxidative stress

Abstract

Many anticancer therapies (CTx) have cardiotoxic side effects that limit their therapeutic potential and cause long-term cardiovascular complications in cancer survivors. This has given rise to the field of cardio-oncology, which recognizes the need for basic, translational, and clinical research focused on understanding the complex signaling events that drive CTx-induced cardiovascular toxicity. Several CTx agents cause mitochondrial damage in the form of mitochondrial DNA deletions, mutations, and suppression of respiratory function and ATP production. In this review, we provide a brief overview of the cardiovascular complications of clinically used CTx agents and discuss current knowledge of local and systemic secondary signaling events that arise in response to mitochondrial stress/damage. Mitochondrial oxidative stress has long been recognized as a contributor to CTx-induced cardiotoxicity; thus, we focus on emerging roles for mitochondria in epigenetic regulation, innate immunity, and signaling via noncoding RNAs and mitochondrial hormones. Because data exploring mitochondrial secondary signaling in the context of cardio-oncology are limited, we also draw upon clinical and preclinical studies, which have examined these pathways in other relevant pathologies.

INTRODUCTION

Cardiovascular disease (CVD) and cancer remain the leading causes of global morbidity and mortality, and their costs continue to rise (1). By 2030, the annual costs of CVD and cancer have been projected to rise to more than $1 trillion and $8.3 trillion, respectively (2). Although underlying etiologies of the two pathologies differ, CVD and cancer are linked in a bidirectional, causal relationship whereby each disease increases the risk of the other (3). Furthermore, some cancer treatments, such as antibiotic anthracyclines and anti-HER2 therapies, have well-known cardiotoxic effects that limit their clinical utility (4, 5).

The efficacy of anticancer therapy (CTx) is improving; thus, long-term survival of patients with cancer is at an all-time high. With longer survival, adverse cardiovascular side effects of CTx are emerging as a significant contributor to mortality and morbidity. In patients with breast cancer, cardiovascular complications now represent the leading cause of death outside of cancer reoccurrence (6, 7). Although basic cellular mechanisms that lead to CTx-induced cardiovascular toxicity have been studied for some time, most research has focused on primary damage to the heart itself. Few studies have investigated systemic factors, which may indirectly promote cardiovascular toxicity. Despite the prominent role of mitochondrial function in cardiovascular physiology, mitochondrial defects and associated signaling have only recently been explored in the context of cardio-oncology.

In this review, we highlight the contributions of emerging mitochondrial signaling mechanisms to cardiovascular complications of anticancer therapies. We briefly describe pharmacological mechanisms and cardiotoxic side effects of antineoplastic drugs, then discuss molecular and/or cellular mechanisms leading to cardiotoxicity. We focus on secondary signaling events downstream of mitochondrial damage, including changes in epigenetic signaling, innate immunity, and circulating factors such as mitochondrial nucleic acids and mitochondrial hormones.

MITOCHONDRIA AS MEDIATORS OF CARDIOTOXICITY

Mitochondria play crucial roles in energy metabolism, signaling, and cell death. Cardiomyocytes rely heavily on mitochondrial oxidative phosphorylation to meet the high energy demands of initiating contractions and maintaining ion homeostasis in the beating heart. Cardiomyocytes also have relatively low endogenous antioxidant defenses; thus, cardiomyocytes are particularly susceptible to mitochondrial damage and dysfunction. In addition to generating ATP, mitochondria are key signaling organelles that sense and respond to a variety of stressors, such as hypoxia and DNA damage. Mitochondria are major sources of reactive oxygen species (ROS), which serve important functions in signaling cascades. In the vasculature, mitochondrial ROS and mitochondrial Ca²⁺ dynamics regulate vascular tone and atherosclerotic processes. Mitochondria also monitor a variety of signals, such as growth factors and oxidative stress, which can promote apoptosis mediated by opening of the mitochondrial permeability transition pore (mPTP). Despite quality control processes, mitochondrial DNA (mtDNA) and proteins are susceptible to damage, and mitochondrial dysfunction has been implicated in a host of cardiovascular pathologies.

Emerging evidence suggests a role of mitochondrial damage in CTx-induced cardiotoxicity, which includes both local, intracellular effects (e.g., altered cellular metabolism and epigenetic signaling) and systemic signaling (e.g., inflammation and mitochondrial hormones). Several CTx agents promote the production of ROS, which can arise directly from redox cycling of chemotherapeutic drugs themselves, such as the formation of free radicals from anthracycline-iron complexes and semiquinone intermediates (8, 9), or indirectly through mechanisms that promote mitochondrial ROS production or suppress antioxidant mechanisms, such as mtDNA damage. Although oxidative stress has received much attention as a mechanism of CTx-induced cardiotoxicity, other cellular pathways are increasingly being investigated (10), including those involving complex mitochondrial signaling cascades (11, 12). Mitochondria are also involved in CTx-induced cardiotoxicity through their role in intrinsic apoptotic and necrotic cell pathways (12). Moreover, the influence of mitochondria on epigenetic signaling and innate immunity has recently emerged as potential contributors to adverse cardiovascular outcomes related to CTx. Thus, it is becoming clearer that mitochondrial dysfunction can contribute to cardiotoxicity through various pathways that go beyond bioenergetics and apoptosis, although much remains to be explored in the context of CTx.

CARDIOVASCULAR TOXICITY OF COMMON ANTICANCER THERAPIES

The clinical cardiovascular complications of several classes of CTx are well established, though the underlying mechanisms are incompletely understood. Approaches to combat cancer range widely and include classic, cytotoxic chemotherapies, targeted molecular therapies, and immunotherapies, among others. Although the specific mechanisms of cardiotoxicity vary between treatments, many forms of CTx have direct and indirect effects on mitochondria that impact cardiovascular health. Some CTx agents directly modulate mitochondrial function or morphology, which may impair mitochondrial respiration, promote production of mitochondrial ROS, and/or trigger mitochondria-driven apoptosis. Several agents also evoke mtDNA damage, which in turn alters mitochondrial function and promotes systemic secondary signaling mechanisms that provoke cardiovascular inflammation. Though many cardiotoxic mechanisms are treatment specific, other pathways are more generalizable. For example, tumor cell death, which occurs with most forms of CTx, promotes the release of mitochondrial fragments and cellular debris capable of eliciting cardiovascular inflammation.

Primary damage to cardiomyocytes fails to explain some of the clinical phenotypes observed in cancer survivors (e.g., increased risk of coronary artery disease and hypertension); thus, vascular toxicity has been increasingly recognized as a potential contributor to long-term adverse cardiovascular outcomes. Several classes of CTx impact endothelial control of vasomotor tone (13–15), which has implications for regulation of tissue blood flow to vital organs including the heart, brain, and kidney [for review, see Terwoord et al. (16)]. The heart relies heavily on changes in microvascular tone to ensure blood flow is adequate to meet cardiac metabolic demand, and poor microvascular function is the among the best predictors of major adverse cardiovascular events (MACE; 17). Recent evidence suggests that vasodilation mediated by endothelial production of nitric oxide (NO) is severely impaired in patients after neoadjuvant treatment with docetaxel, doxorubicin (DOX), and cyclophosphamide (15). The authors demonstrate that endothelial NO synthase (eNOS) inhibition via phosphorylation of threonine 495 augments vascular ROS production and increases the expression of the NADPH oxidases NOX2 and NOX4. In the human coronary microcirculation, NO suppresses mitochondrial ROS (18); thus, CTx-induced inhibition of NO production may unleash mitochondrial ROS production. Moreover, ROS such as superoxide scavenges NO and can cause uncoupling of eNOS promoting ROS-induced ROS release to futher reduce NO levels. In addition, high levels of ROS induce oxidative damage to mitochondrial proteins and DNA, which further exacerbates mitochondrial ROS production. In sum, CTx may promote ROS-induced ROS production in the vasculature involving mitochondrial ROS production and uncoupling of eNOS. The resulting vascular dysregulation has direct implications for cardiac blood flow in addition to altering paracrine signaling between the heart and endothelium. Some CTx agents (e.g., anthracyclines, cyclophosphamide, carmustine, and carboplatin) also modify membrane permeability and disrupt endothelial barrier function (19–21). This causes vascular damage that in turn enhances the synthesis of inflammatory cytokines, including TNF-α, IL-1β, and IL-6 (19), which are similar to the inflammatory pathways implicated in the atherosclerosis (22). Disrupted endothelial integrity also permits augmented extravasation of CTx drugs (20), which increases the exposure of underlying tissues (e.g., cardiomyocytes) to CTx agents. Together, these complex vascular effects may contribute to adverse cardiovascular outcomes associated with CTx exposure.

In the following section, we summarize key cardiotoxic mechanisms associated with several classes of CTx. The mechanisms underlying CTx-induced cardiotoxicity are complex and multifactorial, and a detailed description is beyond the scope of this review. Thus, we provide a general overview of major known mechanisms associated with cardiotoxicity and highlight mitochondrial involvement. Table 1 summarizes the mitochondrial pathophysiological signaling pathways and associated cardiotoxic outcomes for selected CTx classes. Figure 1 is a visual illustration of pathways to cardiotoxicity using examples summarized in Table 1.

Table 1.

Sample studies highlighting mitochondrial signaling pathways and anticancer therapy-induced cardiotoxicity

Sample Studies	Antineoplastic Drugs	Mitochondrial Signaling Pathways	Cardiotoxic Outcomes
Montaigne et al. (12)	Anthracyclines and Anticancer-targeted therapies, i.e., tyrosine kinase inhibitors (TKI)	Increased mitochondrial oxidative stress and disruption intracellular Ca2+ levels leading to increase in mitochondrial Ca2+ levels The resulting Ca2+ overload pathogenic metabolic pathways lead to mitochondrial ROS production (23)
		Mitochondrial swelling and increased permeability of its outer membrane to apoptotic factors such as cytochrome-c	Starting with left ventricular dysfunction, multiple MACE can occur
		Modulation of mitochondrial membrane permeability and other transcriptional factors that influence cell fate
Sawyer et al. (24)	Anthracyclines	DNA damage through suppressing expression and/or activity GATA4	Disruption of cardiac sarcomeric structure starting with degradation of the titin protein, which is involved in the pathogenesis of progressive ventricular dysfunction leading to other MACE
Sahu et al. (19)	Alkylating agents	Vascular damage and alteration of blood-brain barrier (BBB)	Inducing inflammation and creating a cardiotoxic environment
	Cyclophosphamide, carmustine, and carboplatin
Gorini et al. (25)	Anthracyclines (doxorubicin)	Binding to phospholipid cardiolipin in the inner mitochondrial membrane and disruption of the electron transport chain (ETC)
		Mitochondrial respiratory capacity impairment	Increase in ROS production, excessive oxidative stress, cell damage, cardiac muscle, and cellular membrane damage
		Reduced cellular antioxidative defense [reduction of glutathione (GSH), superoxide dismutase (SOD), and catalase content or activity] leading to prolongation and/or stabilization of mitochondrial damage
Mehrotra et al. (26)	BH3 mimetics	Disruption of mitochondrial morphology and dynamics	Cardiomyopathy
Rasmussen et al. (27)	BH3 mimetics	Disruption of mitochondrial dynamics	Loss of viability and functionality of human cardiomyocytes
Manouchehri et al. (28)	Nilotinib, an oral TKI	Upregulates proatherogenic adhesion proteins on human endothelial cells	Promotion of vascular events
Rodríguez-Hernández et al. (29)	Ponatinib, sorafenib, and regorafenib (TKIs)	Induced mitochondrial dysfunction, uncoupling components of electron transport chain (including mitochondrial Ca2+ overload) and promoted ROS generation	General cardiovascular toxicity
Dempsey et al. (5)	Trastuzumab, an anti-HER2 drug	Blocking the function of neuregulin
		Promotion of damaging effects of oxidative stress	Myocytes attrition over time to heart failure
		DNA breakage and induction of mitochondrial apoptosis
Oun et al. (30)	Cisplatin (a platinum containing agent)	Not fully understood. Either as a secondary effect from nephrotoxicity or as a direct ROS attack the heart	Arrhythmias
			Cardiac ischemia, diastolic disturbances myocardial infarction angina
			Pericarditis
			Thromboembolic events
			Chronic heart failure
Yuan et al. (31)	5-fluorouracil (antimetabolite)	Disruption of RNA synthesis and inhibition of thymidylate synthase and incorporates its metabolites into RNA and DNA Vascular dysfunction with microthrombi formation	Chest pain related to coronary vasospasm,
			Dilated cardiomyopathy,
		Vascular dysfunction with microthrombi formation	Ventricular arrhythmia, Sudden cardiac death

MACE, major adverse cardiovascular events; ROS, reactive oxygen species.

Figure 1.

Graphical illustration of selected anticancer therapy (CTx) drugs illustrating their mitochondrial pathophysiological signaling pathways and some of the associated cardiotoxic outcomes. CTx agents work through different mechanisms and induce mitochondrial dysfunction through various pathophysiological pathways. This is graphical illustration of complex pathogenic pathways using selected examples from literature identified in our article. It is not meant to be an exhaustive depiction of all existing pathways but rather presents some key examples. Image created with Rawgraphs.io and published with permission. ETC, electron transport chain; MACE, major adverse cardiovascular events; ROS, reactive oxygen species; TKI, tyrosine kinase inhibitor.

Anthracyclines

Anthracycline antibiotics, including doxorubicin, epirubicin, and daunorubicin, are among the most effective treatments for many solid tumors, leukemias, and lymphomas. Anthracyclines intercalate into nucleic acids and inhibit the activity of topoisomerases, which disrupts DNA replication and translation. Although interfering with DNA is central to their cytotoxic and antitumor effects, anthracyclines also readily undergo redox reactions to produce free radicals (32), including nuclear and mtDNA damage. In some cells, these events can trigger apoptosis (33, 34) or cellular senescence (35), both of which can impair physiological processes and lead to organ failure (33, 34). Although anthracyclines are highly effective at combatting tumor growth, anthracycline therapy leads to irreversible, dose-dependent cardiotoxicity, which is the primary challenge that limits the cumulative dose patients receive.

Although many scholars have implicated anthracycline-induced mitochondrial ROS in cardiomyocyte death, other novel mitochondrial signaling pathways are also involved in anthracycline cardiotoxicity (11). For example, oxidative stress induced by anthracyclines also damages mtDNA (36, 37), which can further exacerbate mitochondrial ROS production and promote innate immune activation and inflammation by the release of mtDNA (discussed further in local mitochondrial signaling and systemic mitochondrial signaling below; 12). Mounting evidence suggests anthracyclines both promote and inhibit autophagic processes in a time-dependent manner (38, 39) and manipulating autophagy can ameliorate cardiotoxicity in preclinical models (40–42). These effects appear to be related to autophagic regulation of mitochondrial dynamics (i.e., fission and fusion) and respiratory function (for review, see Ref. 43). Anthracyclines also modulate mitochondrial membrane permeability and transcriptional factors that influence cell fate (12). Moreover, anthracyclines can modulate intracellular Ca²⁺ levels, including in the mitochondria (23). Although physiological levels of mitochondrial Ca²⁺ stimulate metabolic rate by increasing activity of tricarboxylic acid (TCA) cycle dehydrogenases and ATP output (44, 45), mitochondrial Ca²⁺ overload leads to the opening of the mPTP, release of cytochrome-c, and apoptosis (for a recent review, see Ref. 46). Ca²⁺-calmodulin can also promote ROS production through activation of protein kinases (12, 23), which in turn promotes peroxidation of cardiolipin, a major lipid component of mitochondrial membranes (47). Finally, mitochondrial membrane structural changes can lead to cardiomyocyte dysfunction and death, and consequently, contribute to development of MACE (12).

BH₃ Mimetics

Intrinsic apoptosis is regulated by the B-cell lymphoma 2 (BCL-2) apoptotic pathway. The BCL-2 family comprises three groups of proteins regulated by different BCL-2 homology (BH) regions: prosurvival BCL-2-like proteins, proapoptotic BH₃-only proteins, and proapoptotic Bcl-2-like protein 4 (BAX)/BRI1-associated receptor kinase 1 (BAK) proteins. BH₃-only proteins bind to and inactive prosurvival BCL-2-like proteins to initiate apoptosis. Ultimately, BH₃-only proteins promote activation of BAX/BAK, which causes mitochondrial outer membrane permeabilization and release of mitochondrial apoptotic factors such as cytochrome-c and Smac/DIABLO to induce caspase activation.

A hallmark of cancer cells is their ability to evade apoptosis by manipulating pro- and antiapoptotic genes and proteins. Many cancers overexpress prosurvival BCL-2-like proteins, thereby promoting cancer cell survival and proliferation (48). BH₃ mimetics replicate the antagonistic function of BH₃-only proteins on BCL-2-like proteins to promote apoptosis of cancer cells; thus, their primary mechanism of action targets mitochondrial apoptosis pathways (26, 48). Thus, it is unsurprising that BH₃ mimetics disrupt mitochondrial morphology and dynamics (26).

One mechanism that has received attention as a mediator of the adverse effects of BH₃ mimetics on mitochondrial morphology is inhibition of myeloid cell leukemia-1 (MCL-1), a BCL-2 family protein that is involved in mitochondrial homeostasis in diverse tissues, including the heart (49, 50). The effect of BH₃ mimetics on MCL-1 may contribute to the unintended cardiotoxicity of these drugs. MCL-1 proteins are upregulated in several human hematological and solid tumors (50); indeed, MCL-1 is among the most frequently amplified genes in human cancers (51, 52). High expression of MCL-1 in tumor cells thus underlies the high efficacy of BH₃ mimetics. However, MCL-1 expression in the myocardium differentially exposes the heart as tissue vulnerable to the toxic effects of BH₃ mimetics. A study with cardiomyocyte-specific MCL-1 deletion found that cardiac ablation of MCL-1 in adult mice led to rapid cardiomyopathy and death (50). Mechanistically, the authors showed that myocytes from knockout animals had disorganized sarcomeres and swollen mitochondria. These changes were accompanied by reduced mitochondrial respiration and opening of the mPTP, which is consistent with increased apoptosis (50). Another independent study found that inhibition of MCL-1 by BH₃ mimetics disrupted mitochondrial dynamics and led to a loss of viability and functionality in human cardiomyocytes (27). Due to the complexity of the BCL-2 apoptosis pathway and the variety of BH₃ mimetics available, it is important to note that many additional mechanisms are likely involved in the mitochondrial and cardiovascular effects of BH₃ mimetics.

Tyrosine Kinase Inhibitors

Tyrosine kinase inhibitors (TKIs) exert their antineoplastic effects by inhibiting kinases involved in specific signaling pathways. Kinases activate proteins through phosphorylation and are, therefore, fundamental regulators of cell signaling (53). Inhibition of specific kinases can be harnessed to combat various processes involved in tumor progression (e.g., angiogenesis). Conversely, overactivation of specific kinases can induce numerous pathologic consequences, including cancer (28); thus, TKIs which inhibit cancer-promoting kinases can be effectively targeted to combat cancer growth (54).

Kinases play a critical role in cardiac, vascular, and metabolic homeostasis, and altered kinase activity can lead to various adverse effects on the heart and the vasculature (28). The mechanism of action of some TKIs directly targets the vasculature, such as VEGF-TKIs. TKIs can also exert off-target effects on the vasculature. For example, nilotinib, an oral TKI used in the treatment of chronic myeloid leukemia (CML), impairs endothelial function, reduces nitric oxide bioavailability, and increases ROS generation in the human microcirculation (28). In human endothelial cells, nilotinib also upregulates proatherogenic adhesion proteins (28), including vascular cell adhesion protein 1 (VCAM1), intercellular adhesion molecule 1 (ICAM1), and E-selectin, all of which can recruit inflammatory cells and platelets thereby provoking vascular events (28). Other examples of potentially cardiotoxic side effects of TKIs (such as ponatinib, sorafenib, and regorafenib) include mitochondrial dysfunction, uncoupled respiration, mitochondrial Ca²⁺ overload, and increased ROS generation (29).

Anti-HER2 Drugs

Anti-HER2 drugs, such as the humanized monoclonal antibody trastuzumab, are targeted molecular therapies that work by blocking HER2 receptors (55). HER2 receptors have roles in cell proliferation not only in breast cancer cells but also in developing and adult hearts. Trastuzumab binds to an extracellular domain of HER2 receptors and inhibits tumor cell growth in vitro and in vivo via several mechanisms (56). Trastuzumab is the standard treatment for both early and metastatic HER2-positive breast cancer, although other novel anti-HER2 therapies are currently being used in cancer treatment (57).

Although anti-HER2 therapies are highly effective in treating cancer, they are also associated with cardiotoxic side effects that often manifest as reversible left ventricular dysfunction and heart failure (5, 55, 57). The cardiotoxic effects of trastuzumab are in part mediated by its action of blocking the function of the HER2 ligand neuregulin (5). Neuregulin is a cell-cell signaling protein secreted by coronary microvascular endothelial cells and the endocardium (58). Neuregulin is required for normal cardiac growth and maintenance and is critical in protecting cells against oxidative stress-induced cell death and promoting myocardial cell survival (5, 59). By blocking neuregulin function, trastuzumab promotes the damaging effects of oxidative stress, leading to DNA breakage and induction of mitochondrial apoptosis (5). Specifically, trastuzumab alters the balance of pro- and antiapoptotic BCL-2 proteins, which promotes mitochondrial release of cytochrome-c and initiates caspase activation and apoptosis in cardiomyocytes (60). Activation of mitochondria-mediated apoptosis occurs in conjunction with mitochondrial respiratory dysfunction, reduced ATP levels, impaired redox capacity, and loss of mitochondrial membrane potential (60). Attrition of myocytes over time is the most important mechanism leading to heart failure (59).

Platinum-Containing Agents

Platinum-based drugs such as cisplatin, carboplatin, and oxaliplatin are some of the most effective and widely used cancer chemotherapeutics (61). These drugs prevent DNA transcription and replication and induce DNA damage (62), which ultimately initiates mitochondria-mediated apoptosis (30). In addition to binding nuclear DNA, cisplatin affects mitochondrial DNA and increases mitochondrial ROS production, which is involved in the cytotoxic effects (62). Cancer cell lines that respond to cisplatin have greater mitochondrial content and higher levels of mitochondrial ROS than cisplatin-resistant cancer cells (63).

The mechanisms underlying cisplatin-induced cardiotoxicity are not completely understood, though many of the consequences are thought to be secondary to nephrotoxicity (30). There is also some evidence that cisplatin cardiotoxicity is linked to oxidative damage. Clinically, adverse cardiac outcomes associated with platinum-based therapies include silent and symptomatic arrhythmias, cardiac ischemia, diastolic disturbances without proper relaxation of ventricles before the next contraction, myocardial infarction, angina, pericarditis, thromboembolic events, and chronic heart failure (30).

Antimetabolite Agents

Antimetabolites inhibit essential biosynthetic processes or are incorporated into macromolecules, such as DNA and RNA, and inhibit their normal function (64). Antimetabolites are used in treating a variety of leukemias and solid tumors (31). Fluoropyrimidine 5-fluorouracil (5-FU) and its prodrug capecitabine are widely used antimetabolite agents. 5-FU is converted intracellularly to several active metabolites that disrupt DNA and RNA synthesis by inhibiting the action of thymidylate synthase (TS; 64). 5-FU promotes ROS production in cardiomyocytes and endothelial cells (65) and induces mitochondrial dysfunction characterized by low ATP levels, loss of mitochondrial membrane potential, and excessive mitophagy, which culminate in mitochondria-mediated apoptosis (66, 67).

Although 5-FU has been used for years in cancer treatment, it has severe side effects including cardiotoxicity (31). One potential mechanism associated with its cardiotoxicity is through vascular dysfunction with microthrombi formation (31). Cardiotoxicity caused by 5-FU often presents with chest pain related to coronary vasospasm (31). Other more serious cardiotoxic outcomes include dilated cardiomyopathy, ventricular arrhythmia, and sudden cardiac death (31).

Immunotherapies

In the past decade, immunotherapies, such as immune checkpoint inhibitors and cell-based therapies, have gained momentum as powerful anticancer tools. Under normal circumstances, immune checkpoint proteins on T cells (e.g., programmed death 1; PD-1) engage with corresponding checkpoint proteins on healthy cells (e.g., programmed death-ligand 1; PD-L1) to prevent T cells from attacking healthy cells and rein in immune responses. Thus, tumor cells that express immune checkpoint inhibitors can evade the immune system. By blocking checkpoint proteins, immune checkpoint inhibitors enable T cells to kill cancer cells and thereby promote immunogenic cell death. T cells can also be engineered to target tumor-specific antigens, which is the premise of chimeric antigen receptor (CAR) T-cell therapy.

Because of their relative novelty and the fact that latent cardiotoxicity takes years to develop, the cardiovascular consequences of immunotherapies have not been fully established. Immune-related adverse events such as myocarditis are rare but serious complications of immune checkpoint inhibitors (68). CAR-T cell therapy is associated with cytokine release syndrome, which is frequently followed by major adverse cardiovascular events and development of cardiomyopathy, arrhythmias, and/or circulatory collapse (69, 70). Preclinical evidence suggests that combining immunotherapies with other forms of CTx may potentiate cardiotoxic effects (71), though further research is needed to establish whether this occurs clinically with a range of CTx combinations. Mitochondrial respiration and dynamics are key regulators of T-cell function (72, 73), and oxidative stress generated by proinflammatory cytokines could induce mitochondrial dysfunction; however, sparse data exist examining whether mitochondria play a role in immunotherapy-induced cardiotoxicity.

LOCAL MITOCHONDRIAL SIGNALING

The following section focuses on known mechanism of intracellular communication in response to mitochondrial damage. Existing evidence, though limited, is reviewed, and put in perspective with known mechanisms of cardiovascular defects that originate in mitochondria.

Intracellular Mitochondrial Signaling

Mitochondria are no longer simply considered energy producers within the cells but have been increasingly recognized as intracellular signaling organelles. Classically, retrograde signaling reflects communication initiated from the mitochondria to the nucleus under conditions of stress (74). Nevertheless, in the past decades, it has become clear that mitochondria are constantly communicating to the nucleus, irrespective of stress, through signaling molecules. For example, the role of physiological mitochondrial reactive oxygen species (ROS) in cell proliferation, response to hypoxia and interorganellar communication is well established [for reviews, see Hamanaka and Chandel (75), Shadel and Horvath (76); Joseph et al. (77)]. More recently, the impact of metabolites primarily derived from the tricarboxylic acid (TCA) cycle or impacted by changes in its flux in remodeling the epigenome and influencing gene expression has been documented [for review, see Santos (78)]. Likewise, the ability of mitochondrial nucleic acids (DNA and RNA) to activate innate immunity and influence health outcomes has been increasingly recognized (79). In the context of cardiomyopathy, mice overexpressing a mutant mitochondrial DNA polymerase gamma (POLG) specifically in the heart showed altered nuclear DNA methylation (80). Similarly, inactivation of innate immune signaling ameliorated cardiomyopathy identified in a mouse model that expresses an exonucleolytic deficient POLG (81). Thus, it seems that mitochondrial dysfunction per se can contribute to cardiovascular outcomes through signaling mechanisms that go beyond classic stress responses.

In the below sections, we will focus on studies that provide support to the role of mitochondria in these emerging fields, namely epigenetics and innate immunity, as potential contributors to chemotherapy-associated cardiomyopathy (Fig. 2). The role of mitochondria in the context of ROS/oxidative stress and cardiotoxicity driven by cancer treatment has been reviewed elsewhere (82–86) and therefore will not be discussed here.

Figure 2.

Anticancer therapy (CTx) alters intracellular mitochondrial signaling. Several CTx agents induce mitochondrial dysfunction by promoting mitochondrial DNA (mtDNA) damage, suppressing mitochondrial metabolism, altering mitochondrial Ca²⁺ homeostasis, increasing production of reactive oxygen species (ROS), altering mitochondrial-nuclear communication, and/or inducing epigenetic and transcriptional remodeling. Mitochondria also play a role in CTx-induced apoptosis through release of cytochrome-c (Cyt c) and second mitochondria-derived activator of caspases (SMAC) and activation of apoptosomes. Nucleic acids in the cytosol, including _mtRNA and _mtDNA, activate innate immune cascades that ultimately promote inflammation. Image created with Biorender.com and published with permission. AIM2, absent-in melanoma 2; cGAS, GMP-AMP synthase; IRF3, interferon regulatory factor 3; RIG-I, retinoic acid-inducible gene I; STING, stimulator of interferon genes.

Mitochondria, Epigenetic Signaling

Anthracyclines are the best example of chemotherapy-induced cardiomyopathy, with doxorubicin (DOX) one of the most-studied drugs. Unique to this class of agents, and in particular to DOX, is not only the acute cardiac effects that present at the time of exposure but the cardiovascular outcomes that can arise years after treatment withdrawal (87). The mechanisms associated with the late-arising effects of DOX remain poorly understood, but speculation of epigenetic changes exists given the observations of chronic transcriptional remodeling detected weeks or months after DOX withdrawal (88, 89). Another long-term outcome of anthracycline treatment is mitochondrial dysfunction, which in animal models has been reported even 30 wk after cessation of DOX treatment (90). Nevertheless, it remains unclear whether mitochondrial dysfunction and transcriptional remodeling are parallel events subsequent to DOX exposure or whether mitochondrial changes actively contribute to the transcriptional outcomes through long-term epigenetic remodeling. DOX affects mitochondrial function through different mechanisms, including inhibition of complex I, mitochondrial DNA (mtDNA) damage and depletion, increased iron levels, as well as ROS (for a recent review, see Ref. 91), all of which have been shown to impinge on the status of the epigenome (92–96). Thus, it is conceivable that mitochondrial-driven epigenetic remodeling contributes to the long-term and insidious effects attributed to DOX. However, studies directly probing this connection are still emerging, and the exact cause of the long-term mitochondrial impairments observed after DOX treatment remains unclear as this drug, like all other chemotherapeutic agents, is not a specific mitochondrial inhibitor.

Short-term subchronic studies probing global epigenetic changes indicate that mitochondrial dysfunction induced by DOX might underlie epigenomic remodeling that persists after treatment ends. Ferreira and coworkers (97) treated male rats for 7 days with DOX and analyzed several biochemical and genomics parameters 2 wk after treatment ceased. At the time of analyses, the authors found persistent decrease in mtDNA content, decreased transcription of mtDNA-encoded genes and of several nuclear genes. Using biochemical assays to gauge changes in activities of enzymes associated with epigenetic modifications, the authors detected decreased histone deacetylase (HDAC) activity. Judged by the analysis of bulk 5-methyl-cytosine (5meC) levels using an ELISA assay, they also found that nuclear DNA methylation was diminished in the animals previously exposed to DOX. Previous work had demonstrated changes in levels of mitochondrial TCA cycle metabolites after DOX treatment (98), leading the authors to propose that DOX disturbed mitochondrial production of key metabolites involved in epigenetic maintenance that in turn influenced the epigenome and phenotypes of the heart (97). Notably, mtDNA depletion had been shown to alter the epigenetic landscape (92) and affect TCA cycle flux (94). Nevertheless, TCA metabolites were not directly measured in the study. Also, decreased levels of 5meC could be due to a decrease in the methyl donor s-adenosyl-methionine (SAM) and of DNA methyltransferase activity (DNMT) or reflect increased demethylation of DNA by ten eleven translocation enzymes (TETs). Alternatively, increased oxidative stress caused by DOX could directly oxidize 5meC triggering DNA repair and cytosine demethylation (99).

Despite these remaining questions, to the best of our knowledge, the work from Ferreira et al. (97) is the only to directly correlate epigenetic marks, mitochondrial and transcriptional changes after DOX withdrawal in the heart. Moving forward, it will be necessary to perform more in-depth epigenetic analysis, using deep sequencing technologies, in the hearts of animals treated with DOX, both during the treatment and after drug withdrawal. Genomics analyses that report on the epigenome status at the locus level can help to understand the relationship with gene expression. Likewise, metabolomics analyses can inform about tissue-wide changes in metabolites that can help identify contributors to the phenotype; this is particularly relevant given the wide range off-targets of DOX. It has been estimated that the half-life of mitochondria in the rodent heart is ∼14–17 days (100). Thus, allowing longer “recovery” periods after ceasing DOX treatment is necessary to assure that the protracted effects associated with DOX do not reflect leftover damaged mitochondria from the original exposure that have yet to be recycled. Finally, experiments so far involved males but including females in the analysis will help understand the relative influence, if any, of sex hormones. This is relevant as DOX-induced cardiotoxicity has been shown to be impacted by sex (101).

Mitoxantrone is a DNA topoisomerase inhibitor that causes long-term cardiomyopathy. Although an anthracycline, the cardiomyopathy associated with mitoxantrone seems to derive from cumulative mitochondrial damage (102). In one study looking at long-term cardiomyopathy associated with this exposure, rats were treated with three cycles of mitoxantrone for a total of 20 days. Animals were then evaluated 2 (D22) and 28 (D48) days after the last cycle when several parameters of mitochondria and cardiac function were followed. The authors did not observe effects on antioxidant levels in the heart at either timepoint, whereas aberrant mitochondria, changes in mitochondrial complex IV and V activities, and depletion of ATP levels were observed at D48. Plasma lactate levels, a hallmark of mitochondrial dysfunction, were also increased at this time point as was heart mass, suggesting hypertrophy likely as an effort of the muscle to compensate for bioenergetic deficit. Interestingly, these effects were not observed at D22. Instead, at this time the authors found that complex IV and V activities were increased although this did not affect ATP levels (102). Taken together, these results suggested that mitochondrial dysfunction progressed in the rat heart over time after mitoxantrone was removed. It also suggested that the initial treatment had early effects on the organelle that could not be overcome by mitochondria turnover in the 4 wk after exposure ceased. The mechanisms responsible for these protracted effects remain unclear. It is possible that the initial changes in mitochondrial function altered metabolism and established a new epigenetic landscape, which then influenced mitochondrial function after drug removal. Mitoxantrone has been shown to have a high affinity for histones in vitro (103), it is possible that it could also do so in vivo, although they would be expected to recede upon removal of the drug. More studies are needed to better understand the protracted mitochondrial and cardiovascular effects of mitoxantrone treatment.

In addition to anthracyclines, other chemotherapeutic drugs show varying degrees of cardiotoxicity, with cisplatin, trastuzumab, and sunitinib also affecting mitochondrial function (25, 104, 105). Although generally, their withdrawal eliminates the effects on the heart, cisplatin can show longer-term cardiovascular outcomes that seem to stem from its effects on kidney function (106). In renal cells, cisplatin was shown to alter TCA cycle metabolites (107), but whether it does the same in the heart and/or affects the epigenome is unknown. Many of the cardiovascular effects of cisplatin, trastuzumab, and sunitinib seem to track with previous anthracycline exposure (108), suggesting that they might precipitate but not necessarily cause the cardiac changes. Thus, if altered mitochondrial signaling plays a role in the cardiotoxicity of nonanthracycline chemotherapeutic agents, it is likely that it involves different pathways than those associated with DOX. For example, cisplatin damages nuclear and mtDNA. Previous work has shown that targeting cisplatin to mitochondria alone suffices to cause apoptosis (109). Thus, cisplatin-induced mitochondrial dysfunction may cause cardiac deficit by killing cardiomyocytes. DOX can also cause apoptosis but its high affinity for cardiolipin, a major lipid component of mitochondrial membranes, is unique to this drug (91). As cardiolipin dysfunction has broad physiological effects (110), the effects of DOX on cardiolipin may have significant consequences for long-term cardiovascular health. Interestingly, intrinsic differences in mitochondrial respiratory reserve among tissues of animals chronically treated with DOX were proposed to derive from the depletion of cardiolipin and not be associated with impaired mitochondrial respiration (111).

Decreased mitochondrial Ca²⁺ has been shown to alter TCA cycle metabolites, histone methylation, and the expression of genes associated with myofibroblasts differentiation (and thus fibrosis) in murine cardiac cells (112). In rats treated with cisplatin, decreased Ca²⁺ uptake was identified in isolated mitochondria from kidney (113). More recently, altered epigenetic status in the form of DNA methylation, the abundance of histone marks, and dysregulated expression of noncoding RNAs were associated with cisplatin exposure in renal cells (114). It is feasible to speculate that similar effects may occur in hearts exposed to cisplatin, but this remains to be determined. Interestingly, a recent study that followed testicular cancer survivors (TCS) reported mild decreased diastolic function 30 years after treatment with cisplatin (115). In a separate study, epigenetic changes were identified in the blood of an independent cohort of TCS 16 years after their treatment with cisplatin (116). Conceivably, mitochondrial changes, such as altered Ca²⁺ homeostasis, could drive these epigenetic outcomes in the heart or in the kidneys, or both, ultimately influencing cardiovascular health. Clearly, more studies are needed to define whether a link between mitochondrial dysfunction, epigenetic remodeling, and cisplatin-induced cardiomyopathy exist.

Trastuzumab and sunitinib are targeted therapies that have also been shown to cause cardiotoxicity. In addition to playing a role in cell proliferation in the breast tissue, HER2 has functions in the developing and adult heart. Specifically, HER2 signaling is essential for growth and survival of cardiomyocytes under stress conditions such as mechanical strain (25, 117, 118). In addition, HER2 is expressed on the vascular endothelium, and exposure to anti-HER2 therapy could promote endothelial dysfunction as a possible contributor to the development of MACE. Thus, it is easy to envision how inhibition of Her2 in the tumor can inadvertently lead to off-target cardiovascular effects. Nevertheless, the exact mechanisms involved in trastuzumab-induced cardiotoxicity remain unclear, as this drug has been shown to induce oxidative stress, mitochondrial dysfunction, and apoptosis (119–121)—all of which contribute to myocardial dysfunction. Likewise, sunitinib has been shown to affect mitochondrial function through different mechanisms, including through inhibition of AMPK (AMP-activated protein kinase) that by virtue of sensing changes in the AMP:ATP ratio is a central regulator of energy/mitochondrial homeostasis (122). In cardiac cell culture and animal models, sunitinib was shown to inhibit mitochondrial β oxidation, impair electron transport chain (ETC) activity, increase mitochondrial ROS production, and remodel energy metabolism (123–125). Based on these findings, it would not be surprising to find that chronic but low-grade alterations in mitochondrial metabolism in the heart induced by these drugs could impact the epigenome and contribute to their cardiac effects. However, studies to address this possibility remain needed. Even if they do affect the epigenome through the mitochondria, as withdrawal of these drugs generally resolve the cardiac events, one could argue that such effects might be secondary in the context of their cardiotoxicity.

Finally, it is worth noting that chemotherapeutic drugs, by virtue of their systemic delivery, can also exert their effects by affecting mitochondrial function in immune cells (such as macrophages, monocytes, etc.) as well at vascular cells (mainly endothelial and smooth muscle cells), in turn indirectly impacting cardiovascular health. Studies have shown that epigenetic remodeling in cells from the immune system drive dysregulated overaction of those cells in the heart and vasculature (126). In macrophages, the transition from M1 to M2 polarization is driven by mitochondria (127) and epigenetic remodeling (128), dysregulation of which alters their ability to promote inflammation or an anti-inflammatory response. Recently, both DOX and ionizing radiation (IR) were shown to cause mitochondrial dysfunction in myeloid cells and a persistent proinflammatory phenotype that accelerated coronary atherosclerosis (129, 130). Likewise, proper function of a specific macrophage lineage in the heart has been shown to be necessary to remove and recycle damaged mitochondria to maintain cardiovascular health in animal models (131). Thus, conceptually mitochondrial dysfunction caused by chemotherapeutics in myeloid cells could be involved in their epigenetic remodeling and contribute to their persistent/dysregulated activation in the context of cardio-oncology. The extent to which this might contribute to the cardiac events associated with acute and/or chronic exposures to cancer treatments is unknown, and studies to specifically address this question are warranted.

Innate Immunity and Chemotherapy-Induced Cardiotoxicity: Is There a Role for Mitochondria?

Activation of innate immunity is another area in which mitochondria are now recognized players. Given the increasing evidence demonstrating the significant impact that innate immunity plays in cardiovascular outcomes (132), mitochondrial dysfunction induced by chemotherapeutics may also affect the heart through dysregulating the immune system. Seminal work from Gerald Shadel’s laboratory has demonstrated that depletion of transcription factor A mitochondrial (TFAM), a mitochondrial protein involved in mtDNA replication/transcription and nucleoid structure, leads to mtDNA leakage into the cytosol activating cyclic GMP-AMP synthase (cGAS) and stimulator of interferon genes (STING). Together cGAS and STING are responsible for sensing cytoplasmic DNA and activating a signaling cascade that leads to a series of phosphorylation reactions that promote the dimerization and nuclear translocation of interferon regulatory factor 3 (IRF3). This in turn leads to the expression of interferon-stimulated genes (ISGs), whose products effectively handle pathogens (133). In their initial work, West and colleagues showed that activation of ISGs by depletion of TFAM potentiated type I interferon (IFN-I) responses and conferred broad viral resistance (134). More recently, they showed that genotoxin-damaged mtDNA effectively activated innate immunity. Using primary fibroblasts and cultured cancer cells, authors found that DOX treatment damaged mtDNA and engaged cGAS-STING and ISGs. They also showed that irradiation of TFAM^+/− mice with IR led to enhanced nuclear DNA repair in the spleen. Given these data, these authors concluded that the release of mtDNA in the cytosol elicited a protective signaling response that enhanced nuclear DNA repair in cells and tissues. A similar conclusion was obtained by Tigano and coworkers who, by inducing double-strand breaks on the mtDNA using mitochondrial-targeted transcription activator-like effector nucleases (TALENs), also identified increased nuclear DNA repair in cultured cells exposed to IR (135). Notably, mtRNAs (and not mtDNA) induced the innate immunity cascade, which in this case involved retinoic acid-inducible gene I (RIG-I), an RNA sensing protein that also activates ISGs (135).

Together, these studies demonstrate that both mtDNA and mtRNAs, which can be released upon genotoxin exposures, engage innate immunity. This is relevant in the context of chemotherapy as many drugs used to kill cancer cells are genotoxins, which often damage the nuclear and mitochondrial genomes. Although activation of innate immunity by genotoxins might counterintuitively benefit the tumor by increasing DNA repair and thus providing a means of resistance, it might also have effects in different organs. For example, this could potentially be a means through which direct damage to the nuclear DNA of cardiomyocytes is more efficiently repaired, sparing apoptosis of these terminally differentiated cells overall protecting the heart. Conversely, hyperactivation of innate immunity upon chemotherapy could be rather detrimental to cardiovascular health. A recent study using the POLG exonucleolytic mutator mouse showed that activation of innate immunity through cGAS-STING contributed to the cardiovascular phenotypes of old animals. Specifically, they showed that elevated IFN-I signaling suppressed activation of the nuclear factor erythroid 2-related factor 2 (NRF2), a transcription factor that is a master regulator of the cellular antioxidant response. This in turn increased oxidative stress and enhanced proinflammatory cytokine responses. Ablation of the IFN-I signaling ameliorated the hyperinflammatory phenotype and cardiovascular dysfunction in aged mutator mice (81). Robust IFN-1 signaling driven by mtDNA stress and STING also appears to contribute to DOX-induced cardiotoxicity via autocrine or paracrine signaling downstream of the IFN-1 receptor (136). Mice lacking STING or the IFN-1 receptor are protected from DOX-induced left ventricular dysfunction and cardiac fibrosis, suggesting that overactivation of innate immunity driven by the release of damaged mtDNA can contribute to cardiomyopathy (136).

Whether mtDNA damage caused by chemotherapeutic agents such as DOX, cisplatin, and etoposide among others could cause overactivation of innate immunity and negatively contribute to the cardiovascular effects of these drugs remains to be explored. Nevertheless, studies using some of these agents suggest that they can activate innate immunity through mtDNA leakage in other organs, granting studies about whether similar effects are at play in the heart. For example, cisplatin treatment induced mtDNA leakage and activation of cGAS/STING in an animal model of acute kidney injury. Both knockdown of STING and removal of mtDNA ameliorated the progressive nature of the kidney injury, suggesting that mtDNA-driven engagement of cGAS-STING as critical regulator of kidney injury in the context of cisplatin exposure (137). Other chemotherapeutic agents have been shown to activate cGAS-STING (138) although it remains to be defined the extent to which such response involves sensing the mitochondrial genome in the cytosol. In this context, a high-throughput screening assay developed to identify compounds that either inhibit or bypass the Ebola virus protein VP35 IFN-antagonist function, identified five DNA intercalators as hits from a library of bioactive compounds. Most of them were anthracyclines, including the cardiotoxic drug DOX, which along the other agents triggered the DNA-sensing cGAS-STING pathway of IFN induction (139).

Another means through which mitochondria might contribute to chemotherapy-induced cardiotoxicity is through activation of the inflammasome. NLRP3 (nucleotide-binding oligomerization domain leucine-rich repeat and pyrin domain-containing protein 3) inflammasomes are multiprotein complexes of the innate immune system that recruit pro-caspase-1 via the adaptor molecule apoptosis-associated speck-like protein (ASC). This ultimately leads to the maturation of the proinflammatory cytokines IL-1β and IL-18 (140). There is increasing evidence of the relevance of the NLRP3 inflammasome in the pathogenesis of cardiovascular diseases (CVD; 141). There are also reports that chemotherapeutic agents induce NLRP3 activation, including DOX (142) and, most notably, that the cardiotoxic effects of DOX, specifically, can be mitigated by inhibition of the NLRP3 (143, 144). Thus, although it is clear that DOX can activate NLRP3, it remains unknown whether such effects are associated with mitochondrial dysfunction or the other cellular perturbations derived from anthracycline treatment that can also activate this pathway, including modulation of ion flux, extracellular ATP, lysosomal degradation, ROS, etc. (145). Recently, oxidized mtDNA (ox-mtDNA) was shown to effectively turn on the inflammasome (146). Given that DOX can induce both increased mitochondrial ROS production and oxidation of the mtDNA (91), it is feasible that mitochondria might signal to the NLRP3 inflammasome through different mechanisms after DOX exposure. Clearly, more studies are required to determine the extent to which mitochondrial nucleic acids might signal to the immune system in the context of chemotherapy and whether these events play a role in the associated cardiovascular outcomes.

SYSTEMIC MITOCHONDRIAL SIGNALING

Many cancer therapies cause mitochondrial damage in cancer and noncancer cells alike, and the resulting mitochondrial stress can initiate secondary, systemic signaling that may contribute to cardiovascular outcomes. Nevertheless, mitochondrial-derived extracellular signaling is still an underexplored mechanism in the context of adverse cardiovascular events in patients with cancer. Secondary signaling events triggered by mitochondrial damage are well characterized in several cardiovascular conditions, including hypertension, heart failure, and endothelial dysfunction, among others [reviewed by Kadlec et al. (11) and Nakayama et al. (147)]. However, this concept has not been explored in chemotherapy-induced cardiovascular toxicity. This section provides a brief overview of mitochondria-derived signals in cardio-oncology, including circulating mitochondrial fragments, RNAs, and hormones.

mtDNA Damage, DAMPs, PAMPs

In addition to detecting bacteria and other pathogens, the innate immune system senses endogenous signals released from damaged cells and coordinates an inflammatory response to promote tissue repair. Although this is initially protective, chronic activation of inflammatory pathways has been proposed to contribute to cardiovascular dysfunction in a variety of circumstances, including cardio-oncology. Mitochondria share many evolutionarily conserved molecular and structural motifs with bacteria; thus, the innate immune system is primed to recognize and respond to mitochondrial contents released from dying cells. Cell rupture causes intracellular contents to diffuse to surrounding tissues and enter the circulation, where they serve as damage-associated molecular patterns (DAMPs) that activate innate immune signaling cascades (Fig. 3). Likewise, cellular stress, such as mtDNA damage, promotes regulated release of DAMPs (i.e., mtDNA fragments) from intact cells. Damaged mitochondria are subject to quality control processes via autophagy (i.e., mitophagy), and recent evidence suggests that perturbations to mitophagy promote extracellular release of free mitochondria (148). Several mitochondrial DAMPs are known to stimulate innate immune responses, including cell-free mtDNA (cf-mtDNA) fragments, mitochondrial membrane fractions, ATP, and formyl peptides. Circulating DAMPs are detected by a range of intracellular and cell-surface receptors on immune cells, including Toll-like receptors (TLRs), NOD-like receptors (NLRs), formyl peptide receptors (FPRs), and purinergic receptors. Of note, mtDNA damage induced by many anticancer therapies promotes the formation and release of cf-mtDNA, which is recognized by TLRs, stimulates cGAS-STING signaling, and activates both NLRP3 and absent-in melanoma 2 (AIM2) inflammasomes (149–151). The vascular endothelium also expresses innate immune receptors and acts as an interface that senses DAMPs and orchestrates an inflammatory response by expressing adhesion molecules, releasing cytokines, and increasing vascular permeability to facilitate leukocyte migration (152, 153).

Figure 3.

Systemic mitochondrial signaling may promote cardiovascular inflammation following anticancer therapy (CTx). CTx-induced cellular stress and death promote the release of damage-associated molecular patterns (DAMPs). DAMPs released from mitochondria include cell-free mitochondrial DNA (mtDNA), ATP, formyl peptides, and mitochondrial membrane fractions, among others. Circulating mitochondrial DAMPs are recognized by pattern recognition receptors of the innate immune system, including Toll-like receptors (TLRs), purinergic receptors, formyl peptide receptors (FPRs), and nucleotide-binding and oligomerization domain (NOD)-like receptors (NLRs). Activation of innate immune receptors on immune cells, cardiomyocytes, and vascular cells promotes cardiovascular inflammation that may play a role in CTx-induced cardiotoxicity. Image created with Biorender.com and published with permission.

Circulating DAMPs are elevated in patients with cancer (154), which could arise from a number of sources, including tumor growth and remodeling, tumor cell death, and carcinogenic environmental toxins (155). As the goal of most cancer therapies is to induce cancer cell death, dying tumor cells represent a major potential source of mitochondrial DAMPs, which could provoke cardiovascular inflammation. Indeed, Klee et al. (154) postulated a direct relationship of tumor size to levels of circulating DAMPs. However, nontumor cells, such as cardiomyocytes, are also susceptible to CTx-induced damage (e.g., mtDNA damage), which could lead to release of DAMPs. Thus, the origin of circulating DAMPs that trigger cardiovascular inflammation in patients with cancer undergoing therapy is not clear. In addition to DAMPs released from host cells, alterations in the microbiome and death of microbiota trigger the release of pathogen-associated molecular patterns (PAMPs; reviewed in Refs. 156, 157), which activate similar innate immune pathways and may also play a role in anticancer therapy complications.

Elevated DAMPs following anticancer therapy may have important implications for cardiovascular health. Chronic elevation of DAMPs is observed in patients with a range of CVD, including hypertension, atherosclerosis, cerebrovascular disease, and heart failure (158–160). In both rodent models (161) and large patient cohorts (162), a direct correlation between DAMPs and adverse cardiovascular events has been identified. Moreover, preclinical evidence suggests that DAMPs are involved in cardiovascular disease pathogenesis (163–167). Considering the established link between high levels of DAMPs and CVD pathology, a contribution of DAMPs to elevated cardiovascular risk in patients with cancer is plausible but has not been explored to date. Genetic downregulation or pharmacological blockade of TLRs protects against vascular endothelial dysfunction and ameliorates doxorubicin-induced cardiomyopathy (159, 168, 169). Given the well-documented role of DAMPs in exacerbating inflammation in CVD, we speculate that DAMPs, irrespective of their impact on cancer treatment and progression, may facilitate cardiotoxicity during and following cancer therapy.

Although DAMPs play a key role in innate immune-mediated inflammation, they have a range of effects that can both promote and suppress inflammation (reviewed in Refs. 170, 171). In cancer, DAMPs play a dual role with both beneficial and deleterious consequences. There is growing interest in using DAMPs and associated signaling to promote immunogenic tumor cell death and improve treatment efficacy (172–174). Conversely, DAMPs are also associated with development of treatment resistance and may promote tumor progression in some instances (175–178). These paradoxical effects likely depend upon the level and type of DAMPs, mechanisms of action of antineoplastic therapies, and other systemic and local interactions. Ultimately, novel strategies should aim to limit the off-target effects of unrestrained inflammation while preserving treatment efficacy of anticancer therapies.

Mitochondria-Derived Noncoding RNAs

There is an unmet clinical need to better understand the factors that drive pathological cardiac remodeling and subsequent heart failure. Several genetic modifiers have been identified which contribute to cardiac hypertrophy; however, the precise underlying molecular mechanisms are complex, multifactorial, and incompletely understood. In addition to coding RNAs that produce proteins, long noncoding RNAs (lncRNAs) are emerging as key molecular regulators of cardiac physiology and pathophysiology. Recent studies have established the predictive value of lncRNA profile and validity for therapeutic targets (179–181; reviewed in Ref. 182). However, few studies have connected changes in lncRNA profile with cardiovascular events in relation to anticancer therapy, and to the best of our knowledge, no study has evaluated chronological changes in relation to treatment-induced cardiotoxicity in humans or in a longitudinal manner. Several regulatory lncRNAs that govern basic cellular mechanisms such as autophagy and mitochondrial metabolism have been described and represent promising treatment targets (183–185). Our group has described a critical contribution of many of these pathways to the regulation of microvascular function in health and disease as well as during acute stress (13, 14, 186–192).

The prognostic value of using noncoding RNAs (ncRNAs) and circulating microRNAs (miRNAs) as biomarkers for disease progression has been explored in CVD such as coronary syndrome (193) [reviewed by Viereck and Thum (194)], but their value as biomarkers of anticancer therapy-induced cardiotoxicity has not been investigated. The nuclear and mitochondrial genome alike encode for miRNAs that can be observed in the circulation. Mitochondria contain miRNAs, termed mitomiRs, which have been identified in several species and cell types (195). Although our knowledge of mitochondrial ncRNAs is still incomplete, they appear to be key players in the regulation of the nuclear and mitochondrial gene expression. Noncoding mitochondria-derived miRNAs have been linked to hypertension, cardiac hypertrophy, myocardial infarction, and heart failure and have been proposed as a biomarker for cardiovascular disease progression (for review, see Ref. 196). In addition, several mitochondrial miRNAs have been characterized as key players in lipid metabolism (197, 198), and hyperlipidemia is an established clinical complication in patients receiving anthracycline-based therapies that may promote CVD.

Mitochondrial dysfunction caused by disruption of mitochondrial–nuclear communication may play a role in the development of various diseases, including cardiovascular disease. The investigation of key players involved in the mitochondrial ncRNAs regulatory network may open the path for the identification and development of novel diagnostic and therapeutic approaches for mitochondrial dysfunction-related CVD (for review, see Ref. 196). However, several challenges remain to be addressed before considering the mitochondrial clinical application of ncRNAs. The biology and function of mitochondrial ncRNAs as well as their antero-retrograde translocation need to be experimentally validated and better understood.

Mitochondrial Hormones

In addition to DAMPs and ncRNAs, mitochondria produce signaling peptides that act in a paracrine and endocrine manner. Many of these mitochondria-derived peptides (MDPs) are released in response to stress and have cytoprotective effects that include preserving mitochondrial function and inhibiting initiation of apoptosis (199–201). Emerging evidence suggests that mitochondrial hormones may ameliorate CVD (for review, see Ref. 202). In a preclinical model, administration of an analog of the MDP humanin protected against DOX-induced cardiotoxicity (203). However, the cytoprotective effects of MDPs such as humanin may also promote treatment resistance and accelerate tumor growth in some circumstances (204). Further research in this developing area is warranted to investigate the role and potential therapeutic utility of MDPs in cardio-oncology.

SUMMARY AND OUTLOOK

Cardiotoxicity is a major clinical complication of anticancer therapies. In addition to direct effects of antineoplastic agents on the heart and vasculature, cardiovascular dysfunction may arise secondary to treatment-induced mitochondrial stress and associated signaling. At the intracellular level, mitochondrial damage alters epigenetic signaling, calcium dynamics, and metabolism and activates innate immune pathways. Mitochondrial damage also promotes the release of DAMPs, ncRNAs, and mitochondrial hormones to the circulation, which trigger systemic inflammation that may contribute to treatment-induced cardiotoxicity. Further research investigating the impact of cancer and its treatments on mitochondria and associated secondary signaling pathways will guide efforts to minimize the cardiovascular side effects of anticancer therapies while preserving treatment efficacy. This in turn could be applicable to other environmental exposures that also affect mitochondria, allowing better understanding and integration of knowledge to improve overall health outcomes of patients with cancer.

GRANTS

This work is funded by National Heart, Lung, and Blood Institute (NHLBI) Grant T32HL134643 (to J.D.T. and A.M.B.), American Heart Association Scientific-focused Research Network Diversity Supplement Grant 960133 (to J.C.B.), and NHLBI Grants R01HL157025 and R01HL133029.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

J.D.T. conceived and designed research; J.C.B., J.H.S., and A.M.B. prepared figures; J.C.B., J.D.T., J.H.S., and A.M.B. drafted manuscript; J.C.B., J.D.T., J.H.S., and A.M.B. edited and revised manuscript; J.C.B., J.D.T., J.H.S., and A.M.B. approved final version of manuscript.