Innate immunity and tolerance toward mitochondria

Mitochondrion 41 (2018) 14–20

Contents lists available at ScienceDirect

Mitochondrion journal homepage: www.elsevier.com/locate/mito

Review

Innate immunity and tolerance toward mitochondria Anthony Rongvaux a b

T

a,b,⁎

Fred Hutchinson Cancer Research Center, Program in Immunology, Clinical Research Division, Seattle, WA 98109, United States University of Washington School of Medicine, Department of Immunology, Seattle, WA 98109, United States

A B S T R A C T Mitochondria are intracellular organelles that originate from a bacterial symbiont, and they retain multiple features of this bacterial ancestry. The immune system evolved to detect the presence of invading pathogens, including bacteria, to eliminate them by a diversity of antimicrobial mechanisms and to mount long-term protective immunity. Due to their bacterial ancestry, mitochondria are sensed by the innate immune system, and trigger inflammatory responses comparable to those induced by pathogenic bacteria. In both cases, innate sensing mechanisms involve Toll-Like Receptors, Formyl Peptide Receptors, inflammasomes or the cGAS/STING pathway. Stressed mitochondria release mitochondrial molecules, such as cardiolipin and mitochondrial DNA, which are sensed as cellular damage potentially caused by infections. Recent research has identified several conditions in which mitochondrial stress-induced immunity is essential to effective antimicrobial defenses. But, in pathological conditions, the abnormal activation of the innate immune system by damaged mitochondria results in auto-inflammatory or autoimmune diseases. To prevent undesirable mitochondria-targeted responses, immune tolerance toward mitochondria must be established, involving regulation of mitophagy and mitochondrial permeability, as well as activation of specific nucleases and pro-apoptotic caspases. Overall, recent findings identify mitochondria as central in the induction of innate immunity, and provide new insights as to how immune responses to these multi-functional organelles might be exploited therapeutically in various disease states.

1. The bacterial origin of mitochondria Mitochondria's origin traces back to an endosymbiotic event that happened about 1.5 billion years ago, when a protoeukaryotic cell engulfed an α-protobacterium and retained it as an organelle (Archibald, 2015; Zimorski et al., 2014). The newly acquired intracellular symbiont then evolved and specialized in aerobic respiration, by means of the electron transport chain and oxidative phosphorylation. This bioenergetics pathway extracts energy from glucose with high efficiency, enabling the production of 36 molecules of ATP per glucose molecule, compared to only two ATP molecules generated by glycolysis. The highly efficient mitochondrial energy production has fueled the extraordinary evolution of eukaryotic cells, from their initial unicellular state to the complexity of modern multicellular organisms. Present-day mitochondria retain at least four distinctive features of their bacterial ancestry. First, mitochondria replicate autonomously in the cytosol, independently of cell division (Archibald, 2015). Second, the mitochondrial double membrane is composed of specific phospholipids, such as cardiolipin, uniquely found in mitochondrial inner membranes and in prokaryotes, but absent from all other eukaryotic

⁎

membranes (Osman et al., 2011). Third, although the vast majority of mitochondrial proteins is encoded in the nucleus, mitochondria retain their own circular DNA genome with hypomethylated CpG motifs, lacking histones and containing intronless, polycistronic genes that encode 13 mitochondrial proteins, 22 transfer RNAs and two ribosomal RNAs in vertebrates (Aanen et al., 2014; Shadel and Clayton, 1997). Fourth, like in bacteria, mitochondrial protein translation starts with a formylated methionine, an N-terminal modification never found on proteins encoded in the nuclear genome (Dahlgren et al., 2016). Therefore, modern mitochondria can be viewed as vestigial bacteria living in the cytosol of eukaryotic cells (Fig. 1A), and this ancestry has important implications for the immune system. In this review, I detail how the innate immune system detects the presence of bacteria, and how the same immune detection mechanisms are triggered by mitochondria. I then describe the tolerance mechanisms needed to prevent constitutive immunity against mitochondria. Finally, I discuss the roles of mitochondria-triggered immune responses in physiological conditions, as well as the pathological consequences of dysregulations in such responses.

Corresponding author at: Fred Hutchinson Cancer Research Center, Program in Immunology, Clinical Research Division, Seattle, WA 98109, United States. E-mail address: [email protected].

http://dx.doi.org/10.1016/j.mito.2017.10.007 Received 16 August 2017; Received in revised form 5 October 2017; Accepted 9 October 2017 Available online 17 October 2017 1567-7249/ © 2017 Elsevier B.V. and Mitochondria Research Society. All rights reserved.

Mitochondrion 41 (2018) 14–20

A. Rongvaux

Fig. 1. Mitochondria are intracellular organelles derived from the endosymbiosis of an α-protobacterium by a protoeukaryotic cell (A). Tolerance mechanisms are needed to prevent the recognition of mitochondria as non-self (B) and avoid anti-mitochondrial innate immunity (C).

2. The innate immune system detects bacterial infections to induce inflammation and protective immunity

the infection site, amplifying their antimicrobial activity. Specific immune cells migrate to draining lymph nodes, where they present pathogen-derived antigens and stimulate an adaptive immune response by T and B lymphocytes, to confer long term antigen-specific protection against the pathogen.

Innate immunity is the first line of immune defense against invading pathogens (Medzhitov, 2007, 2010). In contrast to adaptive immunity mediated by T and B lymphocytes, innate immune responses are mounted almost instantaneously after infection, they target a limited set of molecules and generally do not result in memory development. Innate immunity functions based on the recognition by germline-encoded receptors of Pathogen-Associated Molecular Patterns (PAMPs) or Damage-Associated Molecular Patterns (DAMPs). PAMPs are essential microbial molecules never generated by the host cell, such as bacterial cell wall or flagellum components, which constitute a characteristic microorganism molecular signature. Uniquely microbial nucleic acids represent a second important class of PAMPs. Of course, DNA and RNA are also produced by eukaryotic host cells and are not unique to pathogens. But pathogen-derived nucleic acids differ, and are therefore distinguished by innate immune receptors, owing to their specific base composition, chemical modifications, secondary structures and/or their localization in distinct cellular or extra-cellular compartments. In contrast to PAMPs that are uniquely produced by a pathogen, DAMPs (also known as alarmins) are host endogenous molecules that normally reside in a specific cellular compartment and are exposed or released after stress-induced modifications, cellular injury or death. By sensing DAMPs, the immune system is able to detect that cells are undergoing damage potentially caused by an infection, and thereby trigger antimicrobial defenses. Innate immune sensing is performed by two categories of cells (sentinels, and all other cell types), each relying on distinct types of innate immune receptors (Iwasaki and Medzhitov, 2015). Sentinels of the immune systems, such as macrophages, neutrophils and dendritic cells, patrol the organism and survey for pathogens in the extracellular environment using receptors localized at the plasma membrane or in endosomal compartments. In contrast, most other immune and nonimmune cells are equipped with cytosolic receptors for cell-autonomous detection of infection. The signaling pathways downstream of innate immune sensors activate multiple effector mechanisms that contribute to fending off infections. Antimicrobial peptides, phagocytosis, autophagy and the suicide of infected cells all participate in the direct elimination of invading pathogens. A variety of soluble effectors, including cytokines and chemokines, recruit and activate immune cells at

3. Innate immune sensors respond to both bacteria and mitochondria Several classes of innate immune sensors are used to detect diverse types of PAMPs or DAMPs. This section focuses on the main sensors known to be involved in the detection of bacterial infections and that also respond to mitochondrial ligands: Toll-Like Receptors (TLRs), Formyl Peptide Receptors (FPRs), inflammasomes and molecules of the cGMP-AMP Synthase (cGAS)/Stimulator of IFN Genes (STING) pathway (Fig. 2). TLRs are transmembrane proteins localized at the cell surface or in endosomal compartments, specialized in the detection of molecular patterns unique to pathogens (Akira and Takeda, 2004). TLRs involved in the detection of bacteria include TLR4 that recognizes lipopolysaccharide (LPS, a component of gram-negative bacteria membranes), TLR2 that senses different bacterial cell wall components (such as lipopeptides produced by gram-positive bacteria), TLR5 that binds flagellin (a principal component of bacterial flagella) and TLR9 that responds to unmethylated CpG DNA motifs characteristic of bacterial genomes. TLR signaling results in the production of pro-inflammatory cytokines, such as Tumor Necrosis Factor (TNF) α and Interleukin (IL)6, the activation of antigen presenting cells, and ultimately the induction of an adaptive immune response. Peptides with a formylated methionine at their N-terminus, a signature of prokaryotic protein translation, bind with high affinity to FPRs (Dahlgren et al., 2016). Through this FPR interaction, bacterial Nformyl peptides act as chemoattractants, recruiting phagocytic cells, for example neutrophils, to the site of infection. Inflammasomes are multiprotein complexes, typically composed of three components: a specialized sensor protein, the Apoptosis-associated Speck-like Protein containing a C-terminal caspase recruitment domain (ASC) adaptor molecule and caspase-1 (Ogura et al., 2006; Storek and Monack, 2015). Upon activation, caspase-1 induces an inflammatory response via the cleavage and secretion of the IL-1β and IL18 cytokines, and by execution of a pro-inflammatory type of cell death 15

Mitochondrion 41 (2018) 14–20

A. Rongvaux

Fig. 2. Innate immune sensors involved in the recognition and response to pathogens include Toll-Like Receptors, cytosolic RNA and DNA sensors, Formyl Peptide Receptors and inflammasomes.

bacterial ancestors and gradually evolved toward structures sufficiently different from pathogens to avoid immune detection. But, research in the last few years has shown that mitochondria are actually highly immunogenic, capable of triggering immune activation through essentially all the innate sensors involved in bacteria detection (Krysko et al., 2011) (Fig. 3 and Table 1). For example, mitochondrial extracts induce in vitro chemotaxis and cytokine production via FPR and TLR9 signaling, thus identifying mitochondrial N-formyl peptides and mitochondrial DNA (mtDNA) as potential mitochondrial DAMPs (Zhang et al., 2010). When injected in rats, mitochondrial DAMPs induce a neutrophil-mediated inflammatory disease that produces severe lung injury. Mitochondria have also been shown to induce inflammasome activation. Toxins known to induce mitochondrial stress and reactive oxygen species (ROS) induce IL-1β secretion by a mechanism dependent on the NLRP3 inflammasome (Zhou et al., 2011). Two

known as pyroptosis. Multiple sensors, belonging to the families of Nucleotide Oligomerization Domain (NOD)-Like Receptors (NLRs) or Absent in Melanoma 2 (AIM2)-like receptors (ALRs), can trigger the assembly and activation of inflammasomes in response to various infection-related stimuli. Bacterial infections trigger inflammasome activation by two complementary methods, i.e., through PAMPs that are directly sensed, or by indirect physical consequences of infection-induced stress. For example, the NLR Family CARD Domain Containing 4 (NLRC4) inflammasome senses conserved and essential bacterial proteins, such as components of the flagellum or of the type III secretion system (a multi-protein complex used by bacteria to translocate effector molecules into host cells); AIM2 and other ALRs bind DNA in the cytosolic compartment. In contrast, the NLR Family Pyrin Domain Containing 3 (NLPR3) inflammasome generally functions by detecting host modifications induced by an infection, rather than by detecting the pathogen itself. Endogenous signals sensed by NLRP3 include perturbation of ion fluxes, lysosomal rupture or endoplasmic reticulum stress. The cGAS/STING pathway of cytosolic DNA sensing, initially identified as a central mechanism of DNA virus detection, has recently also been recognized as important for the detection of bacterial infections (Cai et al., 2014; Marinho et al., 2017). cGAS is a cytosolic protein that acquires its enzymatic activity upon binding of a DNA ligand, generating the secondary messenger cyclic dinucleotide cGMP-AMP (cGAMP). In turn, cGAMP binds to the adaptor molecule, STING, that activates key transcription factors (IRF3 and NF- κB) and induces the robust secretion of type I interferons (IFNs), cytokines well-known for their antiviral properties and diverse effects on bacterial infections (Stetson and Medzhitov, 2006). In summary, the immune system is equipped with an array of complementary mechanisms to detect various molecules associated with bacterial pathogens or with their effect(s) on the host cells, to minimize the possibility for infectious agents to escape immune detection. Given the structural similarities between mitochondria and bacteria, described above, one wonders whether the innate immune system could mis-identify mitochondria as foreign entities, potentially inducing detrimental auto-inflammatory or auto-immune responses. In principle, mitochondria could have significantly diverged from their

Fig. 3. Mitochondria can trigger innate immunity through multiple innate immune sensing pathways.

16

Mitochondrion 41 (2018) 14–20

A. Rongvaux

mitochondrial membrane permeability, and caspase activity (Fig. 4 and Table 1). mtDNA-induced inflammation plays an important role in a mouse model of hemodynamic stress-induced heart failure (Nakai et al., 2007; Oka et al., 2012). Successive studies in wild-type mice suggest that after hemodynamic stress, damaged mitochondria are cleared by autophagy and mtDNA is degraded in the autophagosome by the DNase II nuclease. Deficiency of the gene encoding Autophagy Related 5 (ATG5), a central regulator of autophagy, induces aberrant aggregation of mitochondria and exacerbated disease (Nakai et al., 2007). Furthermore, in mice lacking DNase II, mtDNA accumulates in the autolysosome and triggers an inflammatory response mediated by TLR9 (Oka et al., 2012). Therefore, the maintenance of mitochondrial homeostasis via autophagy/mitophagy, and degradation of mtDNA by nucleases, suppresses inflammatory responses triggered by mitochondria in an injury model. Mitophagy also plays a central role in the regulation of NLRP3 inflammasome activation, and the oxidized form of mtDNA has been identified as the main ligand in this context. Perturbation of the autophagy machinery, using a class III phosphatidylinositol-4,5-bisphosphate 3-kinase and autophagy inhibitor (3-methyladenine), or genetic deletion of autophagy regulating factors (Beclin 1, ATG5, MAP1LC3B or p62), induces the accumulation of damaged ROS-producing mitochondria and is sufficient to trigger inflammasome activation in LPSprimed macrophages (Nakahira et al., 2011; Zhong et al., 2016; Zhou et al., 2011). Together, these observations demonstrate that if stressed mitochondria are not cleared by mitophagy, they release oxidized mtDNA to the cytosol where it triggers NLRP3 inflammasome activation. Mitochondrial permeability and the exchange of molecules between mitochondria and the cytosol are tightly regulated and can induce drastic instantaneous cellular responses such as cell death. Maintaining strict mitochondrial permeability is also required to prevent activation of immune responses by mitochondrial molecules. At least two protein complexes regulate mitochondrial permeability and have been linked to the regulation of immune activation: the mitochondrial permeability transition pore (MPTP), and the Bax/Bak channel regulating mitochondrial outer membrane permeability (MOMP). MPTP opens the inner mitochondrial membrane in stress conditions, generally associated with ROS production, preventing mitochondrial bioenergetic function, resulting in ATP depletion and eventually necrotic cell death (Halestrap and Richardson, 2015). Pharmacological inhibition of the MPTP prevents activation of the inflammasome, suggesting that proinflammatory mitochondrial ligands such as oxidized mtDNA are released via the MPTP (Nakahira et al., 2011; Patrushev et al., 2004; Shimada et al., 2012; Zhou et al., 2011). MOMP is regulated by proteins of the Bcl-2 family of apoptotic regulators. The balance between the pro- and anti-apoptotic Bcl-2 family members regulates the formation of the Bax/Bak pore (Tait and Green, 2010). MOMP results in the engagement the cGAS/STING pathway by mtDNA and elicits a type I IFNmediated inflammatory response (Rongvaux et al., 2014; White et al., 2014). Therefore, pharmacologic and genetic evidence demonstrate that innate immune tolerance toward mitochondria requires a tight regulation of their permeability. But, the exact nature of the released mtDNA molecules remains unknown, as are the precise mechanisms by which MPTP and MOMP regulate their escape to the cytosol (Kanneganti et al., 2015; West and Shadel, 2017). The final reported mechanism to prevent mtDNA-induced immunity is mediated by the caspase-9 pathway (Rongvaux et al., 2014; White et al., 2014). The primary function of MOMP, recognized two decades before its role in controlling mtDNA release was appreciated, is to induce caspase-dependent apoptotic cell death (Jiang and Wang, 2004). Following MOMP, cytochrome c is released from the mitochondrial intermembrane space to the cytosol, where it triggers the assembly of the apoptosome in which caspase-9 is activated. Active caspase-9 in turn activates caspases-3 and -7, the downstream effectors of apoptotic cell demolition. These caspases also prevent MOMP-dependent

Table 1 List of mitochondrial DAMP, their sensors and known tolerance mechanisms. Mitochondrial DAMP

Sensor(s)

Tolerance mechanism(s)

mtDNA

TLR9 cGAS NLRP3

Mitophagy DNase II nuclease Mitochondrial membranes permeability Caspases-9 and -3/7

Formylated peptides Cardiolipin ATP

FPR NLRP3 P2X7/NLRP3

mitochondrial ligands have been shown to mediate this inflammasome activation in a cell autonomous manner, i.e., mtDNA, particularly its oxidized form (Collins et al., 2004; Nakahira et al., 2011; Shimada et al., 2012), and the mitochondrial inner membrane phospholipid, cardiolipin (Iyer et al., 2013). Furthermore, necrotic cells can release mitochondrial molecules, such as ATP that triggers NLRP3 inflammasome activation through the purinergic P2X7 receptor (Ghiringhelli et al., 2009; Iyer et al., 2009). Finally, the cGAS/STING pathway of cytosolic DNA detection is also sensitive to activation by mitochondrial ligands. In specific experimental conditions producing mitochondrial stress or mitochondrial membrane permeabilization, mtDNA is likely released from the mitochondrial matrix to the cytosol, where it interacts with cGAS and triggers the activation of an effective inflammatory response mediated by type I IFNs (Rongvaux et al., 2014; West et al., 2015; White et al., 2014). Together, these recent findings demonstrate that mitochondria retain a sufficient bacterial identity to trigger immune activation. 4. Innate immune tolerance can be generated to mtDNA The capacity of the immune system to induce potent mitochondriadirected responses implies that active mechanisms are needed to prevent auto-inflammatory or auto-immune responses, thus establishing a state of immune tolerance toward mitochondria (Fig. 1B). In fact, the experimental conditions that led to the identification of anti-mitochondria innate immunity provide important clues regarding the relevant tolerance mechanisms. As detailed below, they fall into four categories: clearance of damaged mitochondria by autophagy (also known as mitophagy), nuclease activity, tight regulation of

Fig. 4. Several tolerance mechanisms prevent the activation of innate immunity by healthy mitochondria.

17

Mitochondrion 41 (2018) 14–20

A. Rongvaux

induced mitochondrial damage appears crucial in the establishment of an effective innate immune response to HSV-1 infection. Like most other viruses, Dengue Virus (DV) has evolved mechanisms to escape innate immune recognition that would otherwise induce an IFN-mediated antiviral response (Munoz-Jordan et al., 2003). DV is a positive-stranded RNA virus; as such, it encodes non-structural (NS) proteins that antagonize viral RNA recognition pathway (Dalrymple et al., 2015). However, the NS proteins also degrade cGAS and STING, thus eliminating an essential cytosolic DNA sensing mechanism (Aguirre et al., 2017; Aguirre et al., 2012). What is the evolutionary relevance, for an RNA virus, to antagonize a DNA sensing pathway? As with HSV-1, DV infection induces mitochondrial stress and mtDNAdependent activation of cGAS/STING and secretion of type I IFNs, needed to establish resistance to DV infection (Aguirre et al., 2017; Sun et al., 2017). It is likely that post-infection mitochondrial induction of type I IFNs can be generalized to other types of viruses, thus providing a plausible mechanism for the previously reported role of the DNA recognition pathway in the response to RNA viruses (Ishikawa et al., 2009). Furthermore, mtDNA-dependent induction of type I IFN responses is not limited to viral infections. Upon leakage of the bacterial genome into the cytosol, intracellular bacteria, such as Mycobacterium tuberculosis (Mtb), engage cGAS/STING and type I IFN expression (Watson et al., 2015; Watson et al., 2012). But, a recent report suggested that for some strains of Mtb, mtDNA release to the cytosol is responsible for cGAS activation (Wiens and Ernst, 2016). Chitosan, a natural chitin-derived polysaccharide that is a component of crustacean shells and fungi mycelium, has emerged as an attractive adjuvant for human vaccination, owing to its capacity to induce antigen-specific innate immunity, particularly in mucosa (Xia et al., 2015). Like most other adjuvants, the molecular mechanisms underlying the immunostimulatory properties of chitosan are poorly understood. But, a recent report implicated mitochondrial stress, cytosolic DNA and cGAS/STING in chitosan's adjuvant activity, thus highlighting mtDNA as a likely mediator (Carroll et al., 2016). A possible role of mtDNA and cGAS/STING in the immune response to chitin-containing pathogens remains to be established. These examples establish cGAS/STING and inflammasomes as sensors of mitochondrial damage, used by the immune system as a proxy to detect the consequences of an infection or another form of cellular damage. The advantages of such an indirect method of immune activation are twofold. First, it integrates diverse types of cellular stress to converge on a handful of signaling pathways leading to immune activation. Second, this system has the potential to detect microbes that have evolved strategies to escape direct recognition by innate immune sensors. It is currently unknown whether different types of mitochondrial stress trigger inflammasome versus cGAS/STING signaling, or whether both pathways can be activated simultaneously. Interestingly, a recent report demonstrated that NLRP3 inflammasome-activated caspase-1 has the capacity to cleave and inactivate cGAS, thus favoring an IL-1β/ IL-18-mediated immune response over a type I IFN response (Wang et al., 2017). It is also known that type I IFN responses have modulatory functions on inflammasome-dependent immunity (Mayer-Barber et al., 2011; Mayer-Barber and Yan, 2017). Therefore, it is possible that different types of innate responses are induced, depending on the type of mitochondrial signals received.

activation of cGAS/STING by mtDNA (Rongvaux et al., 2014; White et al., 2014). During apoptotic cell death, MOMP simultaneously triggers a mtDNA-mediated pro-inflammatory signal and a caspase-dependent tolerance mechanism, thus enforcing the immunologically silent nature of apoptotic cell death. Since caspase-dependent apoptosis is crucial for the maintenance of tissue homeostasis, controlling the death of over 1 million cells per second in the human body, caspase inhibition in the absence of any other stimulus results in the constitutive activation of cGAS/STING and an in vivo state of viral resistance mediated by type I IFNs. The target proteins cleaved by caspase-3 and -7 to silence cGAS/STING signaling have not been identified to date. In summary, several safety measures prevent the recognition of intact mitochondria as non-self: (i) senescent mitochondria are cleared by mitophagy and their mtDNA content is degraded by specific nuclease (s); (ii) mitochondrial membranes permeability is tightly regulated to prevent the release of mitochondrial molecules to other compartments, where they could engage innate immune sensors; and (iii) in physiological conditions of mitochondrial membrane permeabilization, simultaneous mechanisms of tolerance are triggered, such as the caspasedependent silencing of cGAS/STING. 5. mtDNA sensing plays roles in normal physiology The capacity of the immune system to respond to mitochondrial molecules, particularly mtDNA, raises the question as to whether this is a maladaptation inherited from the bacterial ancestry, or if mitochondrial stress sensing has been co-opted as an indirect mechanism to detect infection-induced damage. In fact, accumulating evidence suggests that mitochondrial DAMPs are required to induce innate immune responses to diverse pro-inflammatory stimuli, as described below. The role of mitochondria in inflammasome-mediated inflammation is not limited to responses that specifically target mitochondria (Zhou et al., 2011), but represents a key intermediate in multiple types of inflammasome activation (Iyer et al., 2013; Nakahira et al., 2011; Shimada et al., 2012). Indeed, canonical inflammasome activation in vitro, using standard protocols such as LPS + ATP or nigericin stimulation, is associated with mitochondrial ROS formation and the accumulation of mtDNA in the cytosol of macrophages. Depletion of mtDNA, inhibition of mitochondrial ROS formation and pharmacological inhibition of mitochondrial permeabilization by the MPTP, all abrogate inflammasome activation in these conditions (Nakahira et al., 2011; Shimada et al., 2012). The NLRP3 inflammasome is well known to respond to a broad range of infectious and stress stimuli, but the mechanism by which it can integrate diverse signals to induce a similar response (i.e., caspase-1-dependent cytokine production and pyroptosis) is incompletely understood. A possible explanation is that these stimuli all induce some form of mitochondrial stress, which could converge on mitochondrial components, such as oxidized mtDNA or cardiolipin, acting as secondary messengers that engage the NLRP3 inflammasome (Elliott and Sutterwala, 2015). A comparable situation is found in the engagement of the cGAS/ STING pathway. Mitochondrial DNA stress can be experimentally induced by deletion of the mtDNA-binding transcription factor, Mitochondrial Transcription Factor A (TFAM). This results in aberrant packaging of mtDNA and its release into the cytosol, where it engages cGAS/STING to establish a type I IFN-dependent anti-viral innate immune response (West et al., 2015). Similar conditions of mitochondrial stress and mtDNA-dependent cGAS/STING activation are replicated in a number of infectious and immunostimulatory conditions. The DNA virus Herpes Simplex Virus-1 (HSV-1) encodes the UL12.5 protein, which localizes to mitochondria and induces mtDNA stress, similar to that after TFAM deletion (Corcoran et al., 2009). An HSV-1 strain that expresses a UL12 mutant protein, which does not induce mtDNA stress and does not affect viral replication, fails to induce an optimal type I IFN response; enhanced viral loads and pathogenesis result in vitro and in vivo (West et al., 2015). Thus, sensing infection-

6. mtDNA immunity can have pathological consequences The flip side of sensing mtDNA as an indication of infection-induced stress is that mitochondrial damage in non-infectious situations could lead to inflammatory or autoimmune diseases (Fig. 1C). Such conditions have been reported after traumatic injuries and in systemic lupus erythematosus (SLE). Sterile inflammation, caused by insults such as traumatic injuries, 18

Mitochondrion 41 (2018) 14–20

A. Rongvaux

7. Conclusions and perspectives

triggers a detrimental response comparable to that observed during septic shock caused by uncontrolled infection (Keel and Trentz, 2005). A role for mitochondria in the induction of the inflammatory response has been suggested in this context. Indeed, the serum concentrations of mtDNA in patients with major trauma are highly increased in comparison to serum concentrations in healthy donors (Zhang et al., 2010). The fact that in vivo injection of mitochondrial extracts induces a disease comparable to a systemic inflammatory response syndrome (as described above) (Zhang et al., 2010), and that these mitochondrial extracts trigger immunity via FPR, TLR9 and NLRP3 signaling, suggests that mitochondrial DAMPs, likely released passively from necrotic cells, are a major contributor to the sterile inflammatory response triggered by traumatic injury (Iyer et al., 2009; Zhang et al., 2010). SLE is an autoimmune disease characterized by autoantibodies and the deposition of immune complexes; it affects many organs, including the skin, joints, the central nervous system and the kidneys (Kaul et al., 2016). The induction of a chronic type I IFN response by nucleic acids is known to play a central role in the etiology of SLE, but the nature of the nucleic acids involved is not clear (Banchereau and Pascual, 2006). Two recent studies identified oxidized mtDNA, extruded from neutrophils in complex with TFAM, as highly interferonogenic in SLE patients (Caielli et al., 2016; Lood et al., 2016). Treatment with a mitochondria-targeted anti-oxidant attenuated disease symptoms in an in vivo mouse model of SLE. The two studies identified TLR9 (Caielli et al., 2016) and cGAS/ STING (Lood et al., 2016), respectively, as the IFN-inducing pathways triggered by extruded oxidized mtDNA; this discrepancy is likely due to the use of different experimental models and requires further investigation. But, it also suggests that different IFN-inducing pathways could be involved in initiating autoimmunity in different patients. In support of this possibility, cGAS enzymatic activity was detected in peripheral blood mononuclear cells in approximately 15% of SLE patients (An et al., 2017), although the authors did not determine the nature of the cGAS-activating interferonogenic DNA (mitochondrial or nuclear), or whether TLR9 is driving autoimmunity in the patients in which cGAS is inactive. Further supporting a role for mitochondria in the etiology of the disease, mitochondrial stress has been associated with SLE, particularly oxidative stress (Perl, 2013; Perl et al., 2012). Therefore, restoring mitochondrial function and integrity, for example using mitochondria-targeted anti-oxidant molecules, could provide novel therapeutic opportunities in SLE. Autoantibodies in SLE target a diversity of auto-antigens, such as double stranded DNA and ribonucleoproteins. Other autoimmune diseases, such as primary biliary cirrhosis, are characterized by the presence of anti-mitochondrial antibodies (Webb et al., 2015). This further highlights the immunogenicity of mitochondria and raises the question as to whether an innate mitochondria-directed immune response induces the preferential production of anti-mitochondrial auto-antibodies. Notably, in both sterile inflammation and SLE, extracellular mtDNA causes pathology (Caielli et al., 2016; Lood et al., 2016; Zhang et al., 2010). As cytosolic mtDNA can also trigger an IFN response (Rongvaux et al., 2014; West et al., 2015; White et al., 2014), it is possible that it could induce a cell-intrinsic form of autoimmunity. In fact, the chronic presence of endogenous nucleic acids is known to cause human diseases, collectively known as interferonopathies, mediated by dysregulated type I IFN responses and characterized by a variety of neurological and dermatological symptoms (Crow and Manel, 2015; Crowl et al., 2017; Rodero and Crow, 2016). Monogenic mutations that alter either the production of endogenous nucleic acids, such as retroelements, the sensitivity of nucleic acid sensors, their downstream signaling pathways or associated regulatory mechanisms, are known to be associated with such interferonopathies (Rodero and Crow, 2016). It will be interesting to determine whether defects in mitochondria and/or mtDNA homeostasis can cause human pathologies with similar symptoms.

Research in the past few years has established that mitochondria are highly immunogenic, owing to their bacterial ancestry. Sensing mitochondrial damage by inflammasomes and cGAS/STING serves as a proxy for the indirect immune detection of an infection or other types of cellular stress. But, the abnormal response to mitochondrial molecules is also associated with human pathologies, such as injury-induced sterile inflammation and autoimmune disorders. Several important questions remain to be addressed, to more completely decipher the mitochondrial role in triggering innate immunity. What are the precise molecular mechanisms that regulate the activation and restriction of anti-mitochondria innate immunity? What is the complete spectrum of human pathologies in which mitochondria-induced inflammation is involved? For example, since neurodegeneration and aging are both associated with mitochondrial dysfunctions, do mitochondrial signals contribute to the inflammation found in these conditions? Or do mutations in the mitochondrial genome induce mitochondrial stress sufficient to trigger innate immunity? Finally, does immune tolerance toward mitochondria create immune system vulnerabilities that can be exploited by intracellular pathogenic bacteria? Mitochondria are well known for their roles in cellular metabolism, in cell death induction, and as a platform for multiple signaling pathways (West et al., 2011). The recent recognition of their role in triggering innate immune responses provides an additional perspective on the physiological and pathological roles of these multi-functional organelles. Acknowledgments I thank Arnaud Marlier for figure preparation, Oberdan Leo and Ruaidhrí Jackson for comments on the manuscript, and Deborah Banker for manuscript editing. Research in my lab is supported by a generous gift from the Bezos family, by the Lupus Research Alliance, the Safeway Foundation and the Emerson Collective. References Aanen, D.K., Spelbrink, J.N., Beekman, M., 2014. What cost mitochondria? The maintenance of functional mitochondrial DNA within and across generations. Philos. Trans. R. Soc. B Biol. Sci. 369, 20130438. Aguirre, S., Maestre, A.M., Pagni, S., Patel, J.R., Savage, T., Gutman, D., Maringer, K., Bernal-Rubio, D., Shabman, R.S., Simon, V., et al., 2012. DENV inhibits type I IFN production in infected cells by cleaving human STING. PLoS Pathog. 8, e1002934. Aguirre, S., Luthra, P., Sanchez-Aparicio, M.T., Maestre, A.M., Patel, J., Lamothe, F., Fredericks, A.C., Tripathi, S., Zhu, T., Pintado-Silva, J., et al., 2017. Dengue virus NS2B protein targets cGAS for degradation and prevents mitochondrial DNA sensing during infection. Nat. Microbiol. 2 (17), 037. Akira, S., Takeda, K., 2004. Toll-like receptor signalling. Nat. Rev. Immunol. 4, 499–511. An, J., Durcan, L., Karr, R.M., Briggs, T.A., Rice, G.I., Teal, T.H., Woodward, J.J., Elkon, K.B., 2017. Expression of cyclic GMP-AMP synthase in patients with systemic lupus erythematosus. Arthritis Rheumatol. 69, 800–807. Archibald, J.M., 2015. Endosymbiosis and eukaryotic cell evolution. Curr. Biol. 25, R911–921. Banchereau, J., Pascual, V., 2006. Type I interferon in systemic lupus erythematosus and other autoimmune diseases. Immunity 25, 383–392. Cai, X., Chiu, Y.H., Chen, Z.J., 2014. The cGAS-cGAMP-STING pathway of cytosolic DNA sensing and signaling. Mol. Cell 54, 289–296. Caielli, S., Athale, S., Domic, B., Murat, E., Chandra, M., Banchereau, R., Baisch, J., Phelps, K., Clayton, S., Gong, M., et al., 2016. Oxidized mitochondrial nucleoids released by neutrophils drive type I interferon production in human lupus. J. Exp. Med. 213, 697–713. Carroll, E.C., Jin, L., Mori, A., Munoz-Wolf, N., Oleszycka, E., Moran, H.B., Mansouri, S., McEntee, C.P., Lambe, E., Agger, E.M., et al., 2016. The vaccine adjuvant chitosan promotes cellular immunity via DNA sensor cGAS-STING-dependent induction of type I interferons. Immunity 44, 597–608. Collins, L.V., Hajizadeh, S., Holme, E., Jonsson, I.M., Tarkowski, A., 2004. Endogenously oxidized mitochondrial DNA induces in vivo and in vitro inflammatory responses. J. Leukoc. Biol. 75, 995–1000. Corcoran, J.A., Saffran, H.A., Duguay, B.A., Smiley, J.R., 2009. Herpes simplex virus UL12.5 targets mitochondria through a mitochondrial localization sequence proximal to the N terminus. J. Virol. 83, 2601–2610. Crow, Y.J., Manel, N., 2015. Aicardi-goutieres syndrome and the type I interferonopathies. Nat. Rev. Immunol. 15, 429–440.

19

Mitochondrion 41 (2018) 14–20

A. Rongvaux

Oka, T., Hikoso, S., Yamaguchi, O., Taneike, M., Takeda, T., Tamai, T., Oyabu, J., Murakawa, T., Nakayama, H., Nishida, K., et al., 2012. Mitochondrial DNA that escapes from autophagy causes inflammation and heart failure. Nature 485, 251–255. Osman, C., Voelker, D.R., Langer, T., 2011. Making heads or tails of phospholipids in mitochondria. J. Cell Biol. 192, 7–16. Patrushev, M., Kasymov, V., Patrusheva, V., Ushakova, T., Gogvadze, V., Gaziev, A., 2004. Mitochondrial permeability transition triggers the release of mtDNA fragments. Cell. Mol. Life Sci. 61, 3100–3103. Perl, A., 2013. Oxidative stress in the pathology and treatment of systemic lupus erythematosus. Nat.Rev. Rheumatol. 9, 674–686. Perl, A., Hanczko, R., Doherty, E., 2012. Assessment of mitochondrial dysfunction in lymphocytes of patients with systemic lupus erythematosus. Methods Mol. Biol. 900, 61–89. Rodero, M.P., Crow, Y.J., 2016. Type I interferon-mediated monogenic autoinflammation: the type I interferonopathies, a conceptual overview. J. Exp. Med. 213, 2527–2538. Rongvaux, A., Jackson, R., Harman, C.C., Li, T., West, A.P., de Zoete, M.R., Wu, Y., Yordy, B., Lakhani, S.A., Kuan, C.Y., et al., 2014. Apoptotic caspases prevent the induction of type I interferons by mitochondrial DNA. Cell 159, 1563–1577. Shadel, G.S., Clayton, D.A., 1997. Mitochondrial DNA maintenance in vertebrates. Ann. Rev. Biochem. 66, 409–435. Shimada, K., Crother, T.R., Karlin, J., Dagvadorj, J., Chiba, N., Chen, S., Ramanujan, V.K., Wolf, A.J., Vergnes, L., Ojcius, D.M., et al., 2012. Oxidized mitochondrial DNA activates the NLRP3 inflammasome during apoptosis. Immunity 36, 401–414. Stetson, D.B., Medzhitov, R., 2006. Type I interferons in host defense. Immunity 25, 373–381. Storek, K.M., Monack, D.M., 2015. Bacterial recognition pathways that lead to inflammasome activation. Immunol. Rev. 265, 112–129. Sun, B., Sundstrom, K.B., Chew, J.J., Bist, P., Gan, E.S., Tan, H.C., Goh, K.C., Chawla, T., Tang, C.K., Ooi, E.E., 2017. Dengue virus activates cGAS through the release of mitochondrial DNA. Sci. Rep. 7, 3594. Tait, S.W., Green, D.R., 2010. Mitochondria and cell death: outer membrane permeabilization and beyond. Nat. Rev. Mol. Cell Biol. 11, 621–632. Wang, Y., Ning, X., Gao, P., Wu, S., Sha, M., Lv, M., Zhou, X., Gao, J., Fang, R., Meng, G., et al., 2017. Inflammasome activation triggers caspase-1-mediated cleavage of cGAS to regulate responses to DNA virus infection. Immunity 46, 393–404. Watson, R.O., Manzanillo, P.S., Cox, J.S., 2012. Extracellular M. tuberculosis DNA targets bacteria for autophagy by activating the host DNA-sensing pathway. Cell 150, 803–815. Watson, R.O., Bell, S.L., MacDuff, D.A., Kimmey, J.M., Diner, E.J., Olivas, J., Vance, R.E., Stallings, C.L., Virgin, H.W., Cox, J.S., 2015. The cytosolic sensor cGAS detects mycobacterium tuberculosis DNA to induce type I interferons and activate autophagy. Cell Host Microbe 17, 811–819. Webb, G.J., Siminovitch, K.A., Hirschfield, G.M., 2015. The immunogenetics of primary biliary cirrhosis: a comprehensive review. J. Autoimmun. 64, 42–52. West, A.P., Shadel, G.S., 2017. Mitochondrial DNA in innate immune responses and inflammatory pathology. Nat. Rev. Immunol. 17, 363–375. West, A.P., Shadel, G.S., Ghosh, S., 2011. Mitochondria in innate immune responses. Nat. Rev. Immunol. 11, 389–402. West, A.P., Khoury-Hanold, W., Staron, M., Tal, M.C., Pineda, C.M., Lang, S.M., Bestwick, M., Duguay, B.A., Raimundo, N., MacDuff, D.A., et al., 2015. Mitochondrial DNA stress primes the antiviral innate immune response. Nature 520, 553–557. White, M.J., McArthur, K., Metcalf, D., Lane, R.M., Cambier, J.C., Herold, M.J., van Delft, M.F., Bedoui, S., Lessene, G., Ritchie, M.E., et al., 2014. Apoptotic caspases suppress mtDNA-induced STING-mediated type I IFN production. Cell 159, 1549–1562. Wiens, K.E., Ernst, J.D., 2016. The mechanism for type I interferon induction by Mycobacterium tuberculosis is bacterial strain-dependent. PLoS Pathog. 12, e1005809. Xia, Y., Fan, Q., Hao, D., Wu, J., Ma, G., Su, Z., 2015. Chitosan-based mucosal adjuvants: sunrise on the ocean. Vaccine 33, 5997–6010. Zhang, Q., Raoof, M., Chen, Y., Sumi, Y., Sursal, T., Junger, W., Brohi, K., Itagaki, K., Hauser, C.J., 2010. Circulating mitochondrial DAMPs cause inflammatory responses to injury. Nature 464, 104–107. Zhong, Z., Umemura, A., Sanchez-Lopez, E., Liang, S., Shalapour, S., Wong, J., He, F., Boassa, D., Perkins, G., Ali, S.R., et al., 2016. NF-kappaB restricts inflammasome activation via elimination of damaged mitochondria. Cell 164, 896–910. Zhou, R., Yazdi, A.S., Menu, P., Tschopp, J., 2011. A role for mitochondria in NLRP3 inflammasome activation. Nature 469, 221–225. Zimorski, V., Ku, C., Martin, W.F., Gould, S.B., 2014. Endosymbiotic theory for organelle origins. Curr. Opin. Microbiol. 22, 38–48.

Crowl, J.T., Gray, E.E., Pestal, K., Volkman, H.E., Stetson, D.B., 2017. Intracellular nucleic acid detection in autoimmunity. Ann. Revi. Immunol. 35, 313–336. Dahlgren, C., Gabl, M., Holdfeldt, A., Winther, M., Forsman, H., 2016. Basic characteristics of the neutrophil receptors that recognize formylated peptides, a danger-associated molecular pattern generated by bacteria and mitochondria. Biochem. Pharmacol. 114, 22–39. Dalrymple, N.A., Cimica, V., Mackow, E.R., 2015. Dengue virus NS proteins inhibit RIG-I/ MAVS signaling by blocking TBK1/IRF3 phosphorylation: dengue virus serotype 1 NS4A is a unique interferon-regulating virulence determinant. mBio 6, e00553–00515. Elliott, E.I., Sutterwala, F.S., 2015. Initiation and perpetuation of NLRP3 inflammasome activation and assembly. Immunol. Rev. 265, 35–52. Ghiringhelli, F., Apetoh, L., Tesniere, A., Aymeric, L., Ma, Y., Ortiz, C., Vermaelen, K., Panaretakis, T., Mignot, G., Ullrich, E., et al., 2009. Activation of the NLRP3 inflammasome in dendritic cells induces IL-1beta-dependent adaptive immunity against tumors. Nat. Med. 15, 1170–1178. Halestrap, A.P., Richardson, A.P., 2015. The mitochondrial permeability transition: a current perspective on its identity and role in ischaemia/reperfusion injury. J. Mol. Cell. Cardiol. 78, 129–141. Ishikawa, H., Ma, Z., Barber, G.N., 2009. STING regulates intracellular DNA-mediated, type I interferon-dependent innate immunity. Nature 461, 788–792. Iwasaki, A., Medzhitov, R., 2015. Control of adaptive immunity by the innate immune system. Nat. Immunol. 16, 343–353. Iyer, S.S., Pulskens, W.P., Sadler, J.J., Butter, L.M., Teske, G.J., Ulland, T.K., Eisenbarth, S.C., Florquin, S., Flavell, R.A., Leemans, J.C., et al., 2009. Necrotic cells trigger a sterile inflammatory response through the Nlrp3 inflammasome. Proc. Natl. Acad. Sci. U. S. A. 106, 20388–20393. Iyer, S.S., He, Q., Janczy, J.R., Elliott, E.I., Zhong, Z., Olivier, A.K., Sadler, J.J., KnepperAdrian, V., Han, R., Qiao, L., et al., 2013. Mitochondrial cardiolipin is required for Nlrp3 inflammasome activation. Immunity 39, 311–323. Jiang, X., Wang, X., 2004. Cytochrome C-mediated apoptosis. Ann. Rev. Biochem. 73, 87–106. Kanneganti, T.D., Kundu, M., Green, D.R., 2015. Innate immune recognition of mtDNA–an undercover signal? Cell Metab. 21, 793–794. Kaul, A., Gordon, C., Crow, M.K., Touma, Z., Urowitz, M.B., van Vollenhoven, R., RuizIrastorza, G., Hughes, G., 2016. Systemic lupus erythematosus. Nat. Rev. Dis. Prim. 2, 16039. Keel, M., Trentz, O., 2005. Pathophysiology of polytrauma. Injury 36, 691–709. Krysko, D.V., Agostinis, P., Krysko, O., Garg, A.D., Bachert, C., Lambrecht, B.N., Vandenabeele, P., 2011. Emerging role of damage-associated molecular patterns derived from mitochondria in inflammation. Trends Immunol. 32, 157–164. Lood, C., Blanco, L.P., Purmalek, M.M., Carmona-Rivera, C., De Ravin, S.S., Smith, C.K., Malech, H.L., Ledbetter, J.A., Elkon, K.B., Kaplan, M.J., 2016. Neutrophil extracellular traps enriched in oxidized mitochondrial DNA are interferogenic and contribute to lupus-like disease. Nat. Med. 22, 146–153. Marinho, F.V., Benmerzoug, S., Oliveira, S.C., Ryffel, B., Quesniaux, V.F.J., 2017. The emerging roles of STING in bacterial infections. Trends Microbiol. Mayer-Barber, K.D., Yan, B., 2017. Clash of the Cytokine Titans: counter-regulation of interleukin-1 and type I interferon-mediated inflammatory responses. Cell. Mol. Immunol. 14, 22–35. Mayer-Barber, K.D., Andrade, B.B., Barber, D.L., Hieny, S., Feng, C.G., Caspar, P., Oland, S., Gordon, S., Sher, A., 2011. Innate and adaptive interferons suppress IL-1alpha and IL-1beta production by distinct pulmonary myeloid subsets during Mycobacterium tuberculosis infection. Immunity 35, 1023–1034. Medzhitov, R., 2007. Recognition of microorganisms and activation of the immune response. Nature 449, 819–826. Medzhitov, R., 2010. Innate immunity: quo vadis? Nat. Immunol. 11, 551–553. Munoz-Jordan, J.L., Sanchez-Burgos, G.G., Laurent-Rolle, M., Garcia-Sastre, A., 2003. Inhibition of interferon signaling by dengue virus. Proc. Natl. Acad. Sci. U. S. A. 100, 14333–14338. Nakahira, K., Haspel, J.A., Rathinam, V.A., Lee, S.J., Dolinay, T., Lam, H.C., Englert, J.A., Rabinovitch, M., Cernadas, M., Kim, H.P., et al., 2011. Autophagy proteins regulate innate immune responses by inhibiting the release of mitochondrial DNA mediated by the NALP3 inflammasome. Nat. Immunol. 12, 222–230. Nakai, A., Yamaguchi, O., Takeda, T., Higuchi, Y., Hikoso, S., Taniike, M., Omiya, S., Mizote, I., Matsumura, Y., Asahi, M., et al., 2007. The role of autophagy in cardiomyocytes in the basal state and in response to hemodynamic stress. Nat. Med. 13, 619–624. Ogura, Y., Sutterwala, F.S., Flavell, R.A., 2006. The inflammasome: first line of the immune response to cell stress. Cell 126, 659–662.

20