Sepsis-induced Cardiac Mitochondrial Damage and Potential Therapeutic Interventions in the Elderly

: The incidence of sepsis and its attendant mortality risk are significantly increased with aging. Thus, severe sepsis in the elderly is likely to become an emerging concern in critical care units. Cardiac dysfunction is an important component of multi-organ failure after sepsis. In our laboratory, utilizing a pneumonia-related sepsis animal model, our research has been focused on the mechanisms underlying sepsis-induced cardiac failure. In this review, based on findings from others and ours, we discussed age-dependent decay in mitochondria and the role of mitochondrial reactive oxygen species (mtROS) in sepsis-induced cardiac inflammation and autophagy. Our recent discovery of a potential signal transduction pathway that triggers myocardial mitochondrial damage is also discussed. Because of the significance of mitochondria damage in the aging process and in sepsis pathogenesis, we hypothesize that specific enhancing mitochondrial antioxidant defense by mitochondria-targeted antioxidants (MTAs) may provide important therapeutic potential in treating elder sepsis patients. In this review, we summarized the categories of currently published MTA molecules and the results of preclinical evaluation of MTAs in sepsis and aging models.


Mitochondrial Damage and Aging Hearts
Mitochondrial dysfunction is a major focus in the study of aging process. The free radial theory of aging proposed by Harman half century ago suggests that aging is a result of deleterious effects of accumulation of harmful reactive oxygen species (ROS) [24]. Multiple intracellular sites produce ROS, such as xanthine oxidase in cytosol [25], NADPH oxidase at membrane [26] and lipid oxidation in peroxisomes [27]. However, the majority of oxidative stress burden comes from mitochondria, where ROS are generated as by products during oxidative phosphorylation and ATP production [28]. Scavenging of mitochondrial ROS (mtROS) is achieved via enzymatic and non-enzymatic antioxidants.
Mitochondrial antioxidant enzymes consist of glutathione peroxidase (GPx), catalase (CAT) and manganese superoxide dismutase (MnSOD) [29][30][31]]. An imbalance between mtROS production and scavenging leads to accumulation of mtROS, which disrupt the function of mitochondrial proteins, lipids, and DNA through structural modifications and therefore alter multiple aspects of mitochondrial function [32,33]. As an extension to the free radical theory, it is proposed that mtROS are the main cause of functional deficiencies associated with aging [34].
However, to date, the correlation between ROS and aging still remains controversial, since published studies have provided evidence in both supporting and against the free radical theory of aging. For example, in yeast, deletion of three mitochondrial antioxidant genes, SOD1 (Cu, Zn superoxide dismutase, CuZnSOD), SOD2 (manganese superoxide dismutase, MnSOD) and CCS1 (Copper chaperone), shortened the life span enormously, suggesting the importance of antioxidant defense in maintaining longevity [35]. Consistently, in Caenorhabditis elegans, giving the wild-type worms small synthetic SOD or catalase mimetics extended their life span by a mean of 44% [36]. It was further shown in mice that mitochondria-specific overexpression of human catalase (mCAT mice) attenuated age-associated mitochondrial dysfunction [37], reduced oxidative damage and significantly increased life span [38]. However, on the contrary, in a transgenic mouse study, overexpression of cytosolic CuZnSOD, catalase, or combinations of either CuZnSOD and catalase, or CuZnSOD and MnSOD failed to provide any longevity benefit [39]. Also, in the yeast study mentioned above, deletion of the other known mitochondrial antioxidant genes (TTR1, CCD1, GLO4, TRR2, TRX3, GRX5, PRX1) had little effect on life span [35]. In according with this, antioxidant supplements so far tested in human did not provide any beneficial effect over a well-balanced diet [40]. Thus, it appears that ROS are not the sole determinant factor but may provide significant impetus to the aging process. Accumulating evidence has indicated the importance of reducing mitochondrial ROS in life span and cardiac health benefits.
In the heart, mitochondria comprise about 30% of myocardial volume [41]. Thus, the heart is especially prone to mitochondrial oxidative stress. Studies suggest that age-dependent cardiac mitochondrial damage is caused by over production of mtROS [42,43]. It was shown that, in mCAT mice, mitochondria-specific overexpression of human catalase provided resistance to heart failure induced by pressure overload [44], angiotensin II and Galphaq overexpression [45]. These transgenic mice exhibited improved cardiac performance and decreased age-associated cardiac pathology, such as ventricular fibrosis, and enlargement of myocardial fiber size [46]. Consistently, over expression of another mitochondrial antioxidant enzyme, manganese superoxide dismutase (MnSOD), protected cardiac morphology and normalized contractility of cardiomyocytes in a type 1 diabetes model [47]. In addition, gene knockout of MnSOD impaired left ventricular functions and promoted heart hypertrophy with accompanying fibrosis and necrosis [48]. Taken all Aging and Disease • Volume 5, Number 2, April 2014 139 together, studies from these transgenic and gene knockout animal models have provided direct evidence to support the critical role of mtROS in cardiac dysfunction and aging. These investigations also suggest that targeted defense against mtROS may become an effective therapeutic strategy in dealing with age-associated cardiac malfunction.

Mitochondrial Damage in Septic Hearts
Cardiac dysfunction is a vital component of sepsisassociated multi-organ failure [49][50][51]. Severe sepsis patients with cardiac dysfunction have significantly higher mortality compared with patients without this condition (70 vs. 20%) [52,53]. Clinically, the degree of mitochondrial dysfunction is tightly linked to sepsis outcomes [54,55]. Since heart is a mitochondria-rich organ, the role of mitochondrial damage in sepsis-induced cardiac failure has been receiving a significant attention. Current studies suggest that multiple aspects of mitochondrial dysfunction, such as impaired metabolism, altered energy generation, and elevated production of mtROS, contribute to sepsis-induced heart failure [56][57][58].
These molecules are released from broken mitochondria into circulating system during cell death and organ injury, initiating inflammatory responses through multifactorial pathways. For example, circulating mtDNA fragments isolated from the plasma from trauma patients are capable to trigger peripheral inflammation in animal models [78,79] Our laboratory previously developed a pneumoniarelated sepsis model in rats [91]. In this model, rats were infected with S. pneumoniae and sepsis symptoms were confirmed by positive blood cultures, pulmonary inflammation, lactic acidosis, and a fall in mean arterial blood pressure 24 hours post-infection [92][93][94][95]. Using this model, we demonstrated that sepsis impaired cardiac mitochondria, causing compromised membrane integrity, increased oxidative stress and decreased antioxidant defense [96]. Further, this sepsis-triggered mitochondrual damage occurred prior to cardiac inflammatory responses such as cytokine productions and NF-B activation [96]. Infiltration of neutrophil [89], accumulation of mtDNA fragments and ASC (apoptosis-associated specklike protein containing a carboxy-terminal CARD), an inflammasome component (unpublished results), were also observed in septic myocardium. We further showed that specific suppression of mtROS protected cardiac mitochondria, attenuated inflammation and improved heart function in the same sepsis animal model [89]. We hypothesisze that sepsis-induced mtROS and inflammation in myocardium are linked through a positive feedback-signaling network. In this scenario, in response to septic challenge, mtROS participate in inciting inflammation that further triggers additional increases of mitochondrial damage and mtROS overproduction, leading to downstream exacerbation of inflammatory responses. In fact, myocardial mtROS increase and mitochondrial damage induced through Aging and Disease • Volume 5, Number 2, April 2014 140 inflammatory mediators have been previously reported using sepsis [97] and non-sepsis models [98]. Thus, specific targeting mtROS, such as using mitochondriatargeted antioxidants, in early sepsis stage may have a therapeutic potential to control the progression of mitochondrial dysfunction and inflammation in later severe sepsis stage.

Mitochondria and cardiac autophagy
Increase in autophagy, a lysosome-dependent mechanism of removing damaged proteins and organelles [99], associates with failing hearts [100][101][102][103][104]. Autophagy cascade is initiated by Beclin-1 (autophagy-regulated gene 6) [105], which forms complex with class III phosphoinositide 3-kinase (PI3K) to promote the formation of autophagosomes. After subsequent fusion with lysosomes, the materials inside autophagosomes are degraded [106,107]. Autophagy is either protective or detrimental to myocardium, depending on varying disease conditions [101,104,108,109]. It has been proposed that, under physiological responses or mild stress, autophagy provides cellular quality control to promote survival and is therefore adaptive. However, under severe or chronic stress, excessive or inadequate autophagy causes massive self-degradation or accumulation of toxic materials; both are maladaptive and eventually provoke cell death [110,111]. As mentioned earlier, mitochondrial oxidative stress burden increases along with aging. Effective removal of damaged mitochondria and unwanted mitochondrial molecules is essential for maintaining a healthy heart. However, recent studies strongly indicate an ageassociated impairment of cardiac autophagy [112,113]. One possible factor for this autophagy deficiency lies in the enlargement of mitochondria. It was shown that mitochondria in aged cells are often enlarged, showing structural changes such as swelling, loss of cristae, and/or almost complete damage of mitochondrial components [114,115]. A recent in vitro study in cultured neonatal cardiomyocytes suggests that autophagic turnover of small mitochondria is more efficient than that of the large ones [116]. Further, disruption of lysosomal function is another factor contributing to the slow-down autophagy in the aged hearts [117][118][119]. As a result of autophagy deficiency, aged hearts unavoidably accumulate damaged mitochondria, mtROS and other mitochondria-derived DAMPs molecules, which increase the heart vulnerability to deteriorative inflammatory and autophagic responses under trauma and sepsis conditions.
In sepsis, increase in autophagy has been detected in multiple organs, including the heart, in animal models and in clinical samples [56,[120][121][122][123]. However, the mechanism(s) underling its occurrence remains unclear.
Function of mtROS in induction of autophagy has been suggested by studies from other disease conditions. For example, in a hypertensive cardiomyopathy model, angiotensin II-provoked autophagy was inhibited by overexpression of mitochondria-targeted antioxidant enzyme catalase [45]. In Hela cells, starvation-induced autophagy was decreased when mtROS failed to increase [124]. On the other hand, evidence also indicates that autophagy exerts a control over mtROS levels, since autophagy is often initiated in order to remove toxic molecules, including mtROS, under certain stress conditions [125,126]. In our current preliminary investigation, we obtained data suggesting that mtROS may have a stimulatory role in sepsis-induced autophagic responses in the heart, and we hypothesize that, in septic hearts, imbalanced overproduction of mtROS starts an autophagy-promoting feed-forward pathway that leads to pathological progressions.
Current knowledge with regards to the role of cardiac autophagy in sepsis outcomes, adaptive or maladaptive, is still limited and inconclusive.
In septic hearts, pharmacological activation of autophagy in mouse CLP model [127] or in cultured cardiomyocytes [128] suggests that stimulating autophagy is protective to myocardium, and thus autophagy is an adaptive response. However, a recent publication showed that reducing autophagy by an autophagy inhibitor or antioxidants improves cardiac contractility in a mouse lipopolysaccharide (LPS)induced sepsis model [129], suggesting cardiac autophagy as a maladaptive response. The discrepancy of these observations is probably caused by the differences in the level of autophagy, the severity of sepsis and the timing of drug administration in individual experimental settings. Future investigations using autophagy transgenic and knockout models are needed to address the role of autophagy in septic hearts. So far, autophagy status in older sepsis patients remains unclear. Detailed analysis of cardiac autophagy in aging animal models in response to sepsis challenge will help us to understand the pathological conditions in aging septic hearts, assisting further improvement on therapeutic strategies to combat cardiac failure in elder sepsis patients.

Signal transduction of cardiac mitochondrial damage in sepsis
To date, little is known about the intracellular signal transduction pathway(s) that triggers mitochondrial damage in the heart after sepsis. Recent investigation from our laboratory suggests that sepsis alters mitochondrial translocation of tyrosine kinase cSrc and phosphatase SHP2, which may stimulate mitochondrial dysfunction and mtROS production in myocardium [130].
Aging and Disease • Volume 5, Number 2, April 2014 141 During the past several years, a growing body of evidence has suggested that certain well-known intracellular signaling molecules, such Src-family tyrosine kinases [131], tyrosine phosphatases PTP-1B and SHP2 [132], and serine/threonine kinases, protein kinase C (PKC) [133,134] and extracellular-signal-regulated kinases (ERK) [135,136], also provide important functions inside mitochondria. Their intra-mitochondria localization was verified using immune electron microscopy [131,135,137] and western blot analysis [131,132]. However, since these molecules do not possess mitochondria-sorting peptide, the mechanism of their mitochondrial translocation has not been understood yet. In mitochondria, these kinases and phosphatases may play an important part in control of mitochondrial function and structure through reversible phosphorylation and dephosphorylation [132,136,138,139]. Proteomic analysis of healthy mitochondria from rat brains [140] and from mouse hearts [141] captured phosphorylation sites on critical enzymes of mitochondria metabolism, membrane components and biosynthesis molecules. Some key components of oxidative phosphorylation (OXPHOS) complexes, such as subunits of NADHcoenzyme Q oxidoreductase (complex I) [142,143], subunit IV of cytochrome c oxidase (complex IV) [144] and subunit ∂ of F0F1-ATP synthase (complex V) [145], have been identified as targets of phosphorylation. In addition to this category, other mitochondrial functional proteins, such as adenine nucleotide translocator 1 (ANT1) [138], aconitase [146] and telomerase reverse transcriptase (TERT) [147], were also shown regulated via tyrosine phosphorylation and dephosphorylation. It is noteworthy to point out that, since mtROS are generated from the reactions of OXPHOS complexes [148], changes in mitochondrial-localized kinases and phosphotases will inevitably affect the production levels of mtROS.
In a pneumonia-related sepsis animal model, we found that a significant decrease in mitochondrial Src and an increase in mitochondrial SHP2 in myocardium were directly associated with sepsis [130]. Correlated with these changes, tyrosine phosphorylation of mitochondrial proteins, including some essential structural and functional proteins, was dramatically reduced. Both in vitro biochemical analysis and in vivo animal study suggest that OXPHOS complex I and III contain putative substrates of Src and SHP2 [130], consistent with previous findings using small molecule inhibitors that implicated Src family kinases and SHP2 phosphatase as main regulators of tyrosine phosphorylation in mitochondria [132,149,150]. We hypothesize that, during sepsis, certain receptors of pathogen-associated molecular patterns (PAMPs) and/or danger-associated molecular patterns (DAMPs) alter mitochondrial translocation of Src and SHP2 in myocardium. The resulted changes in tyrosine phosphorylation of mitochondrial proteins produce functional deficiency and mtROS overproduction, and thus damaged mitochondria further generate more DAMPs to aggravate inflammatory responses and organ dysfunction [71]. However, several aspects need to be further addressed to support this hypothesis. Mitochondrial substrates of Src and SHP2 remain to be defined, and the upstream receptor(s) that regulates mitochondrial translocation of Src and SHP2 awaits to be identified. Furthermore, whether alteration of mitochondrial Src and SHP2 relates to the production of mitochondrial-derived DAMPs to stimulate inflammation and how this signaling pathway affects cardiomyocyte function deserve further elucidation.
Current studies started to reveal some evidence that correlates the changes in reversible protein phosphorylation/dephosphorylation inside mitochondria with aging. For example, intra-mitochondrial AMPactivated protein kinase (AMPK) activity decreased with age, contributing to reduced mitochondrial biogenesis [151,152]. Mitochondrial translocation of p66-Shc, an adaptor protein to tyrosine kinase receptors, stimulates mtROS production and regulates longevity in mice [153,154].
Given the important role of kinases and phosphatase in the regulation of mitochondrial function, future research to understand the kinases and phosphatases events inside mitochondria will promote the understanding of pathogenesis in the heart of older sepsis patients. Research in this area will also help to identify new therapeutic targets to control cardiac dysfunction.

Therapeutic Approaches Targeting mtROS
For a long time, oxidative stress has been well recognized as a major promoter in sepsis pathogenesis [155,156], and antioxidants are expected to attenuate inflammation and improve survival following sepsis. However, although this expectation has been met in animal sepsis models [157,158], clinical trials of antioxidant therapies have led to inconsistent results [159][160][161]. One limitation of the conventional antioxidants is that they are globally acting agents, and insufficient dosage and/or lower efficacy are very likely to be the reasons for the failures [162].
Because ROS are mainly generated via mitochondrial respiration, mitochondria themselves are thought to be the primary target of oxidative damage. Targeting antioxidant defense specifically in mitochondria has been expected to provide more effective mitochondrial protection. Accordingly, strategies for mitochondria-targeted delivery of antioxidants are being developed [163][164][165][166][167][168][169][170]. One such approach covalently links bio-molecules to lipophilic triphenylphosphonium cation (TPP + ). Due to a positive charge, the molecules are driven by the mitochondrial membrane potential to accumulate solely in Aging and Disease • Volume 5, Number 2, April 2014 142 mitochondria [167][168][169]. Another group of targeted antioxidants, Szeto-Schiller (SS)-peptides, are small positively charged peptides that accumulate in mitochondria independent of membrane potential [164,170]. These novel mitochondria-targeted antioxidants (MTAs) have demonstrated their higher capability in various experimental settings to fight oxidative stress and to protect mitochondrial function [171][172][173][174]. Currently, MTAs have not yet been applied clinically. A clinical trial of mitochondrial-targeted ubiquinone (MitoQ) showed its benefit in treating liver inflammation [175], and a phase IIb human trial has been initiated in the U.K. to assess the efficacy of MitoQ in non-alcoholic fatty liver disease [176]. To date, the therapeutic potential of MTAs is under intense investigation using pre-clinical models of mitochondrial abnormalities-associated diseases such as neurodegenerative diseases [177,178], cardiac dysfunction [179], cardiac ischemia-reperfusion injury [163], hypertension [180], diabetes [181], and sepsis [182,183].
In sepsis animal models, MitoQ showed its therapeutic benefits in the improvement of cardiac function and prevention of liver damage [182,183]. In a recent published study, we compared the effects of Mito-Vit-E with untargeted vitamin E in the rat pneumoniarelated sepsis model [89]. Both types of antioxidants exhibited significant inhibition on peripheral and cardiac inflammation. At the same dose, Mito-Vit-E provided higher efficacy to reduce cytokine production and to impede neutrophil infiltration in myocardium. This advantage of Mito-Vit-E over vitamin E is likely caused by the fact that vitamin E is distributed globally and its protection of mitochondria against oxidative damage is less efficient, especially in mitochondria-enriched organs such as the heart. Further study of MTAs effects using different sepsis models will allow us to recommend possible candidate molecules for clinical studies and promote translating the application of these novel antioxidants into significantly improved clinical outcomes.
Studies of antioxidant SkQ1, TPP + -conjugated plastoquinone [184], in animal models have revealed certain evidence to support using MTAs as an anti-aging approach. This compound reversed aging-dependent behavioral trait in rats after a ten-week-treatment [185]. In mice with lifelong treatment, SkQ1 significantly reduced age-related changes of hematopoietic and mesenchymal progenitor cells [186]. SkQ1 also showed effect to prolong lifespan in Drosophila [187], mice and hamsters [188]. It has been suggested that supplementation with low doses MTAs is a promising intervention to achieve a healthy aging. However, evidence from both pre-clinical and clinical research is needed to support this hypothesis.

Conclusion
Sepsis represents as a major threat in critical care units. Treatment for this deadly condition remains to be supportive care such as using intravenous fluids and oxygen [4]. Most current attempts of molecular targetbased treatments have failed clinically [162,189]. Even though older patients account nearly two third of severe sepsis cases [7], elderly population is likely to be excluded when new anti-sepsis and anti-microbial agents are tested in clinical trials. It is now realized that aging-associated decay in mitochondrial function and overproduction of mitochondrial oxidative stress are key elements to cause deficiencies in inflammation and autophagy, which are critical responses to trigger organ failure in severe sepsis stage. We anticipate that protection of mitochondria by mitochondria-targeted antioxidants (MTAs) may provide an effective therapeutic strategy for sepsis patients, especially for the elderly. Thus, future preclinical and clinical assessment of MTAs will have important translational implications to significantly impact patient care quality and clinical outcomes.