Abstract
Sepsis-related organ dysfunction remains the most common cause of death in the intensive care unit (ICU), despite advances in healthcare and science. Marked oxidative stress as a result of the inflammatory responses inherent with sepsis initiates changes in mitochondrial function which may result in organ damage. Normally, a complex system of interacting antioxidant defences is able to combat oxidative stress and prevents damage to mitochondria. Despite the accepted role that oxidative stress-mediated injury plays in the development of organ failure, there is still little conclusive evidence of any beneficial effect of systemic antioxidant supplementation in patients with sepsis and organ dysfunction. It has been suggested, however, that antioxidant therapy delivered specifically to mitochondria may be useful.
Sepsis-induced organ failure is associated with oxidative stress and mitochondrial damage.
Reactive oxygen species are produced as a normal consequence of energy production.
Antioxidants protect mitochondria but may become overwhelmed in sepsis.
Antioxidants acting specifically in mitochondria may be of benefit in patients with sepsis.
With a mortality rate of around 25% for uncomplicated sepsis, rising to 80% in those patients who go on to develop multiple organ failure, sepsis is the most common cause of mortality in an intensive care unit and the incidence is increasing.1 During sepsis-induced organ failure, the inflammatory response and ensuing oxidative stress induce changes in mitochondria which result in mitochondrial dysfunction and cell death.
Mitochondrial energy production and release of reactive oxygen species
The inner membrane of mitochondria has a large surface area which is impermeable and contains the enzymes involved in oxidative phosphorylation–energy production from oxygen. Production of energy as ATP takes place via a flow of electrons passed along the five molecular complexes of the electron transport chain. The electron transfer results in reciprocal transfer of protons, creating the mitochondrial membrane potential. As part of this process, reactive oxygen species (ROS) are generated as by-products of the incomplete four-electron reduction of molecular oxygen to water, the final electron acceptor in the process of ATP production (Fig. 1). One by one electron reduction of oxygen from escaped electrons means that as much as 1% of oxygen is converted to ROS. The term ROS includes molecules with an unpaired electron, called free radicals, such as superoxide anion, and also strong oxidizing agents such as hydrogen peroxide. ROS can react avidly with the surrounding molecules in an indiscriminate fashion and so are both highly reactive and short-lived, and can be damaging if not controlled (Fig. 2). The electron transport chain of mitochondria is the major source of intracellular ROS in the cell and mitochondria are also a major target for damage by ROS.2,3 ROS have essential roles in cell signalling and their activity is normally tightly regulated by a collaborative interacting network of antioxidants. Production of ROS by mitochondria is important for normal cellular function and survival and it should be remembered that mitochondria have other roles other than energy production, for example, in various cell signalling pathways and calcium and iron homeostasis.4
Schematic diagram showing the sources of electrons in the electron transport chain of mitochondria (indicated by a red star). The proton flux is denoted by H+. CoQ, coenzyme Q10 (ubiquinone); Cyt c, cytochrome c. See the text for other abbreviations.
Schematic diagram showing the sources of electrons in the electron transport chain of mitochondria (indicated by a red star). The proton flux is denoted by H+. CoQ, coenzyme Q10 (ubiquinone); Cyt c, cytochrome c. See the text for other abbreviations.
Overview of mitochondrial ROS production. ROS production within mitochondria can lead to oxidative damage to mitochondrial proteins, membranes, and mtDNA. Mitochondrial oxidative damage leads to the release of cytochrome c (cyt c) into the cytosol resulting in apoptosis. Increased permeability makes the inner membrane permeable to small molecules. Mitochondrial ROS are also important in cell signalling pathways which modulate several cellular functions. Figure reproduced with permission, from Murphy MP (2009).2 © The Biochemical Society.
Overview of mitochondrial ROS production. ROS production within mitochondria can lead to oxidative damage to mitochondrial proteins, membranes, and mtDNA. Mitochondrial oxidative damage leads to the release of cytochrome c (cyt c) into the cytosol resulting in apoptosis. Increased permeability makes the inner membrane permeable to small molecules. Mitochondrial ROS are also important in cell signalling pathways which modulate several cellular functions. Figure reproduced with permission, from Murphy MP (2009).2 © The Biochemical Society.
Reactive nitrogen species
In addition to producing ROS, the mitochondrial respiratory chain can produce nitric oxide, which itself has an unpaired electron and is therefore a free radical, and other nitric oxide by-products called reactive nitrogen species (RNS). For example, the highly toxic molecule peroxynitrite is formed from the reaction of nitric oxide with superoxide anion. Production of nitric oxide is increased during sepsis by de novo synthesis of the inducible form of type II nitric oxide synthase (NOS). Although superoxide anion is quickly converted to hydrogen peroxide and then water under the action of the endogenous antioxidant enzyme systems, the reaction between superoxide anion and nitric oxide to form peroxynitrite is far more rapid. Peroxynitrite is thought to account for most of the cytotoxic actions of nitric oxide. Nitric oxide has several vital physiological roles, but RNS can have detrimental effects through oxidation, nitrosylation, or nitration of various cellular targets, including proteins, nucleic acids, and endogenous antioxidants such as glutathione.5 It has been proposed that a mitochondrial form of NOS exists, although this is by no means certain6,7 and so this review will specifically address mitochondrial ROS and the mitochondrial antioxidant protection mechanisms.
Oxidative stress-induced mitochondrial damage
When antioxidant defences are overwhelmed, oxidative stress results, which can cause significant damage to lipids, proteins, and nucleic acids, both within mitochondria and cells (Fig. 2). For example, peroxidation of the mitochondrial lipid cardiolipin, which is present in the inner mitochondrial membrane and is important for energy metabolism, leads to dissociation of cytochrome c and causes reduced ATP production, and even more ROS production.3,4,8,9 The endogenous antioxidant systems themselves can also be affected by oxidative stress via oxidation and peroxidation of the component enzymes. In addition, mitochondrial DNA (mtDNA) is also a target for damage since it is close to the electron transport chain.10 mtDNA encodes for several polypeptides plus transfer RNA species and ribosomal RNA species that are vital for electron transport and energy generation. All of the mtDNA encodes expressed genes, unlike genomic DNA which contains many non-coding sequences and thus the potential is higher for functional mutations.11 Oxidative stress-mediated damage to mtDNA can lead to a cycle of ROS production and further mtDNA damage; in other words, a perpetual cycle of ROS production facilitated by ROS-induced ROS release. This cycle of mtDNA damage with loss of function of electron transport enzymes and more ROS generation where the antioxidant systems are completely overwhelmed and eventually cell death occurs is known as the ‘mitochondrial catastrophe hypothesis’ or ‘toxic oxidative stress’.12 Although normally the inner membrane is impermeable, mitochondria can undergo the so-called permeability transition resulting in activation of the caspase cascade, via the release of cytochrome c and apoptosis-inducing factor, which ultimately results in apoptosis or programmed cell death. Mitochondrial membrane permeability transition is a critical point for apoptosis. The permeability transition can be initiated by oxidative stress, nitric oxide, calcium overload, or apoptotic protein upregulation, but regardless of the initiating mechanism, increased mitochondrial permeability is the endpoint for the cell since there is no return after the release of caspase activators such as cytochrome c. See Figure 2 for an overview of mitochondrial ROS production and its consequences.
Endogenous antioxidant protection in mitochondria
Several interacting endogenous antioxidant systems exist within mitochondria to protect against damage by ROS. This network of antioxidant defence systems is tightly regulated and consists of a combination of enzyme and non-enzyme pathways. These include manganese superoxide dismutase (MnSOD), the glutathione (GSH) and thioredoxin (TSH) systems, peroxiredoxins, sulphiredoxins, cytochrome c, peroxidase, and catalase.13,14 Superoxide anion generated within mitochondria is unable to cross the mitochondrial membrane very easily and so is converted to hydrogen peroxide by the action of MnSOD. Although hydrogen peroxide is able to diffuse out of mitochondria for metabolism by catalase, in all but heart mitochondria, this occurs within peroxisomes, not mitochondria. The main system of removal is through the oxidation of reduced mitochondrial GSH (mGSH), catalysed by mGSH peroxidase-1 with recycling back to reduced glutathione catalysed by glutathione reductase, which all take place within mitochondria.15 Oxidation of mitochondrial thioredoxin-2 (TRX-2) with its associated enzymes peroxiredoxin-3 and thioredoxin reductase-2 is also important.16
Although GSH is the most abundant antioxidant in mitochondria, the TRX-2 system has been reported to be more efficient at maintaining mitochondrial proteins in a reduced state compared with mGSH.13 However, when oxidative stress is at a relatively low level, both systems keep hydrogen peroxide levels in check and have well-defined roles. However, under conditions of more severe oxidative stress, such as those seen during sepsis, the TRX-2 system may become more important. In human endothelial cells cultured under conditions of sepsis, the proteins of the TRX-2 system were more resistant to the effects of oxidative stress and appeared to have a more important role in protecting against mitochondrial dysfunction than the mGSH system.17
Oxidative stress in sepsis
Oxidative stress results when ROS production and the antioxidant protection mechanisms are imbalanced. Over the last 10–15 yr, there has been a wealth of studies describing oxidative stress in patients with sepsis, with evidence of ROS production and associated damage, and antioxidant depletion.18–24 Inflammatory responses initiated by oxidative stress occur via the activation of redox pathways for transcriptional activation, for example, increased activation of nuclear factor κB (NFκB) and increased circulating inflammatory mediators including cytokines and pentraxin-3 have been reported in patients with sepsis.25–27
Mitochondrial dysfunction and organ damage in sepsis
Sepsis-induced organ dysfunction has been suggested to be at least in part due to mitochondrial dysfunction as a result of oxidative stress and which results in failure of energy production.28,29 The pathogenesis of mitochondrial damage as a result of sepsis is probably a complex series of events. Both nitric oxide and ROS combined with the release of a variety of exacerbating inflammatory mediators can act to directly or indirectly influence mitochondrial function and energy production. It remains unclear if the self-amplifying cycle of ROS generation and mitochondrial damage occurs with mitochondrial dysfunction leading to oxidative stress and more mitochondrial impairment as the primary event, or if oxidative stress initiates mitochondrial dysfunction and further ROS release. There is a suggestion that progressive improvement in mitochondrial respiration is associated with recovery of organ function in patients who survive sepsis.30
Mitochondrial dysfunction has been shown in animal models of sepsis.31–33 Ultrastructural damage was seen in mitochondria from livers of patients who had died of severe sepsis34 and altered mitochondrial redox state35 and antioxidant depletion associated with mitochondrial dysfunction and related to severity of organ failure and eventual outcome, were reported.36 The link between sepsis and mitochondrial damage has been described in several reviews.37,38 Oxidative stress-mediated damage to mitochondria therefore appears to be fundamental to the pathophysiology of organ failure in sepsis, suggesting a therapeutic role for antioxidants. Despite this, antioxidant supplementation has not been particularly successful in critically ill patients,39–41 perhaps because these are not specifically targeting intracellular location where most oxidative damage occurs, mitochondria. It has therefore been suggested by several experts that antioxidants targeted to mitochondria may be of more benefit.42,43
Reliable delivery and activity of a targeted antioxidant to the mitochondria is central to successful and consistent effects. Targeting has been achieved by a variety of approaches and has been tested in vitro and in animal models of oxidative stress.
Targeting antioxidants to mitochondria
Targeting of antioxidants to mitochondria can be achieved by delivering the antioxidant specifically to mitochondria using a carrier molecule of some sort, by administering an antioxidant which naturally acts or accumulates in mitochondria or by augmenting the body's own mitochondrial antioxidant defences by pharmacological means or by increasing endogenous expression of antioxidant enzymes genetically.
Lipophilic cations
Antioxidant molecules can be covalently attached to lipophilic cations which accumulate in mitochondria as a result of the mitochondrial membrane potential. MitoQ consists of the antioxidant ubiquinone (or co-enzyme Q10) attached to a lipophilic triphenylphosphonium (TPP) cation (Fig. 3).44 The negative charge inside the mitochondrial inner membrane results in the MitoQ accumulating within mitochondria to about 500 times the levels in the cytoplasm (recently reviewed).45 Once inside the mitochondria, the MitoQ is adsorbed onto the inner membrane and is recycled to active ubiquinol in the respiratory chain.
Antioxidants conjugated to the lipophilic cation TPP, accumulate specifically within mitochondria, and block mitochondrial oxidative damage and redox cell signalling. Antioxidants which have been attached to TPP include co-enzyme Q10 (ubiquinone) to form MitoQ and α-tocopherol to form MitoVitE.
Antioxidants conjugated to the lipophilic cation TPP, accumulate specifically within mitochondria, and block mitochondrial oxidative damage and redox cell signalling. Antioxidants which have been attached to TPP include co-enzyme Q10 (ubiquinone) to form MitoQ and α-tocopherol to form MitoVitE.
It has been shown that MitoQ provides better protection in vitro against oxidative stress-mediated injury than untargeted equivalents.46–49 It has been tested in several animal models of disease.50–52 In addition, MitoQ has been used in humans in both patients with Parkinson's disease and those with hepatitis C.45
MitoVitE is a form of tocopherol (vitamin E) and, like MitoQ, is attached to the TPP cation. It has been shown in vitro to protect mitochondria and whole cells from oxidative stress and is much more effective than non-targeted equivalents.44,47,49 Other compounds have also been conjugated to TPP, such as the peroxidase compound Ebselen, called MitoPeroxidase. In contrast to MitoQ and MitoVitE, MitoPeroxidase was only a little more effective than the non-targeted form in preventing oxidative stress-induced mitochondrial damage, due to the fact that its accumulation in mitochondria is less pronounced.53 Plastoquinone is a plant quinone, involved in photosynthesis, which was attached to the TPP cation to form a molecule called SkQ, as an alternative to MitoQ. SkQ protects cells against oxidative stress in vitro and in vivo.54
Hemigramicidin-TEMPOL conjugates
The compound TEMPOL can catalyse the removal of superoxide anion, limit hydroxyl radical formation from hydrogen peroxide, and accept an electron to form the antioxidant hydroxylamine. Conjugation of TEMPOL to part of the antibiotic, gramicidin-S, results in targeting to mitochondria since gramicidin has high affinity for mitochondrial membranes by an unknown mechanism.55 In a rat model of haemorrhagic shock, hemigramicidin-TEMPOL was more beneficial than non-targeted TEMPOL.56,57
Antioxidant peptides
Small synthetic positively charged basic peptides of less than 10 amino acids freely penetrate by passive diffusion into cells and are taken up into mitochondria.58 Szeto and Schiller designed these peptides and so they were named ‘SS peptides’. The selective targeting of the peptides to mitochondria may be due to an interaction between the cationic peptides and the anionic cardiolipin on the inner mitochondrial membrane. The actual peptide sequences are a closely guarded secret. Some of the SS peptides can scavenge hydrogen peroxide, hydroxyl radical, and peroxynitrite and limit lipid peroxidation in a dose-dependent manner. Although there have been no studies of the peptides in models of sepsis, they have been shown to be effective in ischaemia–reperfusion-mediated oxidative stress models.59
Increasing endogenous mitochondrial antioxidants
A combination of antioxidant mechanisms regulates redox balance in mitochondria including the glutathione, thioredoxin, and peroxiredoxin systems, of which glutathione is the most abundant. Glutathione is synthesized in the cell cytoplasm from its three component amino acids and then transported into the mitochondria. Water-soluble compounds such as glutathione N-acetyl-l-cysteine choline esters increase endogenous glutathione by supplying the relevant amino acids and are able to limit mitochondrial depolarization in vitro,60 although these agents have not been studied in models of sepsis.
Genetic approaches have been used to augment endogenous antioxidant proteins. Adenoviral transfection of human MnSOD into rats resulted in an upregulation of MnSOD activity in the liver, with reduced hepatic oxidative damage induced by either alcohol or ischaemia–reperfusion.61,62 Non-protein mimetics with SOD activity which concentrate in mitochondria in the heart have been shown to reduce mitochondrial ROS production in a rat ischaemia–reperfusion model of injury but again there are no studies in sepsis.63
Other approaches
Melatonin is synthesized in many cells from the amino acid tryptophan, which is first converted to serotonin, then N-acetylserotonin and finally melatonin. Melatonin has profound antioxidant activity, reacting with both oxygen- and nitrogen-derived reactive species, and in addition, several of its metabolites also have antioxidant activity. Melatonin is both lipophilic and hydrophilic, and the highest levels in the cell are found in mitochondria. It has both anti-inflammatory and antioxidant activities, scavenging hydrogen peroxide, augmenting endogenous antioxidant pathways, and decreasing nitric oxide production. Melatonin prevents mitochondrial dysfunction, energy failure, and apoptosis and ameliorates inflammatory cytokine release in cells and animal models of oxidative injury.64
α-Lipoic acid is reduced within mitochondria to dihydrolipoate, a powerful antioxidant. It inhibits the activation of NFκB and beneficial effects in oxidative stress-mediated disease models have been shown.65 Another approach used the β-oxidation pathway in mitochondria to biotransform pro-drugs to their corresponding phenolic or thiol antioxidants.66 This approach protected cells against hypoxia-reoxygenation injury, but biotransformation rates varied depending on the position of methyl groups on the pro-drug. In addition, biotransformation was dependent on the mitochondrial membrane potential, suggesting that in the presence of mitochondrial dysfunction, this method of targeting may not be effective.
Targeting antioxidants to mitochondria in sepsis
A review of mitochondrial targeting of antioxidants in sepsis by Fink and colleagues56 suggested that antioxidants conjugated to lipophilic cations such as TPP were unlikely to be helpful in patients with sepsis, since the mitochondrial damage seen during sepsis may reduce the accumulation of such antioxidants into mitochondria. However, despite these reservations, MitoQ is one of the most studied mitochondrial-targeted antioxidants and has been shown to be useful in models of sepsis, with both antioxidant and anti-inflammatory actions. In human endothelial cells, MitoQ decreased the rate of ROS formation and protected mitochondrial membrane potential under conditions of sepsis.67 Both MitoQ and MitoVitE decreased interleukin-(IL)-6 and IL-8 release induced by lipopolysaccharide (LPS) in vitro.67,68 In rats, sepsis-induced organ dysfunction was less after MitoQ or MitoVitE treatment67,69 and sepsis-induced cardiac dysfunction in rats or mice was ameliorated by MitoQ administration.70
Other targeting approaches have been used in animal models of sepsis. In mice, hemigramicidin-TEMPOL was anti-inflammatory in vitro in endotoxin-treated macrophages, and in vivo, pretreatment of endotoxin exposed mice with hemigramicidin-TEMPOL decreased NOS expression.55 Both melatonin and lipoic acid were also protective against sepsis-induced mitochondrial dysfunction in animal models. Several studies of melatonin treatment before caecal ligation and puncture in rats or mice showed a reduction in mitochondrial dysfunction71,72 and pretreatment with lipoic acid in an LPS model of sepsis in rats was shown to reduce oxidative stress and mitochondrial dysfunction.73
In the in vivo studies described above, hemigramicidin-TEMPOL, melatonin, or lipoic acid was given before the initiation of sepsis.56,71–73 Such an approach is not clinically relevant and delaying administration of the targeted antioxidant until after the onset of sepsis is more representative of what happens in patients. There will always be both advantages and disadvantages of studies in animals,74 but attempts should be made to refine these, such that the findings are more relevant to human sepsis.75,76
Treatment of sepsis remains supportive at present and novel therapeutic approaches to reduce the impact are welcome. Targeting antioxidants to mitochondria may offer a novel therapy in the future but clearly further studies are needed.
Conflict of interest
C.D.H. has received payments and travel funding for overseas lectures from ABZ Pharmaceuticals Inc. Total funding approximately £15 000. E.F.L. has received £25 000 to perform clinical studies from IOLabs Ltd. H.G. is a Member of the Editorial Board of the BJA. She is also an editor of the BJA and has received research funding from BJA.




Comments