sign in sign up
<<
Home , Unicellular organism

Medical Hypotheses 127 (2019) 120–128

Contents lists available at ScienceDirect

Medical Hypotheses

journal homepage: www.elsevier.com/locate/mehy

Aging: An evolutionary competition between host cells and mitochondria T Dun-Xian Tan

The Department of System and Anatomy, The University of Texas, Health, San Antonio, TX 78229, USA

ARTICLE INFO ABSTRACT

Keywords: Here, a new theory of aging is proposed. This new theory is referred as the Host-Mitochondria Intracellular Aging Innate Immune Theory of Aging (HMIIITA). The main point of this theory is that the aging is rooted from an Mitochondria evolutionary competition, that is, a never ending coevolutionary race between host cells and mitochondria. Host immunity Mitochondria are the descendants of . The host cells will inevitably sense their bacterial origin, parti- Intracellular innate immunity cularly their circular mtDNA. The host intracellular innate immune pressure (HIIIP) aims to eliminate mtDNA as more as possible while mitochondria have to adapt the HIIIP for survival. Co-evolution is required for both of them. From biological point of view, the larger, the mtDNA, the higher, the chance, it becomes the target of HIIIP. As a result, mitochondria have to reduce their mtDNA size via deletion. This process has last for 1.5–2 billion yeas and the result is that mitochondria have lost excessive 95% of their DNA. This mtDNA deletion is not associated with free radical attack but a unique trait acquired during evolution. In the postmitotic cells, the deletion is passively selected by the mitochondrial fission-fusion cycles. Eventually, the accumulation of deletion will significantly jeopardize the mitochondrial function. The dysfunctional mitochondria no longer provide sufficient ATP to support host cells’ continuous demanding for growth. At this stage, the cell or the aging is inevitable.

Introduction major source of free radicals or other reactive oxygen (ROS) [4,5]. Moreover, researchers had assumed that mitochondrial DNA Aging seems to be an inevitably irreversible process. This is parti- (mtDNA) in contrast to nuclear DNA (nDNA) is naked without protec- cularly valid for higher metazoans which do not possess large numbers tion by histone and, thus, is more vulnerable to ROS attack [6,7].In of stem cells. Some species have a long lifespan, whereas, some exhibit addition, the repair mechanisms for mtDNA are less efficiency than the accelerated aging, thus, live shorter, while others are aging at an those in nDNA [8]. These factors were supposed to cause high intermediate velocity. All of these lifespan variations lead to the mar- rate in mtDNA compared to nNAD. The accumulation of mutant mtDNA velous diversity of histories in creatures inhabiting . However, was concluded to finally result in the dysfunction of mitochondria and represent the only species who care about aging, tries to extend therefore, in phenotypes of aging at cellular and organismal levels life and to maintain healthy state. Starting from our ancestors several [9,10]. thousand years ago, the pursuit of longevity has not been stopped yet. MTA, however, cannot plausibly explain some obviously contra- As a result, at least, 300 plus theories and hypotheses which are related dictory phenomena of aging. For example, several long-lived species to aging have been formulated [1] and many animal studies as well as have higher levels of metabolic rate and generate more ROS than short- computer simulations have been performed in attempts to prove these lived species [11]. Many studies have failed to find accumulated oxi- theories and hypotheses. Even though countless efforts have been made dative damage in mtDNA of old animals or even of humans [12,13]. and considerable amounts of money have been spent, no fundamental Prolonged supplementation with mitochondrially targeted antioxidant breakthrough on aging research has been achieved till now. It appears did not attenuate age-related oxidative damage and rescue the loss of that the aging research will continue forever. muscle mass and function associated with aging [14]. In addition, Among the aging hypotheses, the “Mitochondrial Theory of Aging” mtDNA is not as poorly protected as previously thought. On the con- (MTA) [2] has gained ground. MTA can be considered as an extension trary, mtDNA is well protected by a robust protein coating with mi- of the free radical theory of aging which was proposed by Denham tochondrial transcription factor A (TFAM) and proteins, including an- Harman [3]. He hypothesized that aging might be a result of accumu- tioxidant enzymes [8,15,16]. All these factors let us to reconsider the lated cellular damage inflicted by free radicals. Mitochondria are a merits of MTA.

E-mail address: [email protected]. https://doi.org/10.1016/j.mehy.2019.04.007 Received 18 February 2019; Accepted 11 April 2019 0306-9877/ Published by Elsevier Ltd. D.-X. Tan Medical Hypotheses 127 (2019) 120–128

Here, we propose a new theory of aging which is also associated host cells by cheating. with mitochondrial mutation accumulation, but which is not primarily The easiest and economically most fitted way is camouflage. For the related to the free radical theory. We classify this theory as the Host- bacterium, the most promising procedure might have been to cover Mitochondria Intracellular Immunity Theory of Aging (HMIITA). A itself with a membrane containing, at the surface exposed to the host’s main point of this theory is based on the intracellular innate immune cytosol, components of the host’s membrane, i.e., markers that seem to capacity of . As it will be outlined below, the innate arm of indicate “self”. Traditionally, the outer mitochondrial membrane is the has been shown to not only act in terms of inter- interpreted as a descendant of the endosomal membrane, which con- cellular interactions, but to also play an important intracellular role. I tains host membrane proteins that are specific for the inner face of the hypothesize that aging is rooted in intracellular innate immune pro- plasma membrane. This may be sufficient for preventing detection as cesses which can be understood as a long lasting competition between “nonself” and attack from the cytosol, but it would not yet avoid ly- host cells and mitochondria. sosomal digestion, because the normal fate of an endosome is fusion with a lysosome. However, avoidance of lysosomal digestion has been Co-evolution of host cells and mitochondria described for alphaproteobacteria of the genus Wolbachia [29], which have been regarded by some investigators as genetically close relatives Based on the endosymbiotic theory, the precursor of the mi- of mitochondria [30].InWolbachia, the endosymbiont associates with tochondrion has been an alphaproteobacterium which has been taken Golgi-derived vesicles and is preferably found in the endoplasmic re- up by the protoeukaryotic cell. It is impossible to decide whether the ticulum [29]. Another method of avoiding lysosomal digestion would host cell has actively taken up the bacterium endocytotically by at- be lysis of the endosomal membrane before fusion with the lysosome, a tempting predation of a prey that escaped from becoming digested, or strategy used, e.g., by Listeria [31]. whether the bacterium intruded the host cell by inducing endocytosis, The inner membrane of mitochondria still preserves the trait of a in order to persist as a parasite. Processes of uninvited intrusion as bacterial membrane by containing 20% cardiolipin (CL), which is occurring in recent pathogenic bacteria may serve as models or not. characteristic of bacteria, especially, at this high concentration Endosymbiosis has taken place many times in evolution, in which other [32–36], thus, to cover this bacterial originated inner membrane is bacteria or became endosymbionts of host cells. Well-known necessary. How the outer mitochondrial membrane was formed, may examples are that have transformed into , or the still be debated. The assumption of a derivative of the ancient en- so-called P-symbionts in bacteriocytes of insects [17]. Some eukaryotic dosomal membrane may have some likelihood, but lacks a direct proof. cells, especially those which carried plastids, have also turned into Other intracellular membranes that are able to fuse with endosomal endosymbionts of other uni- or multicellular organisms, which thereby membranes may also be considered, e.g., Golgi-derived vesicles or parts acquired so-called secondary plastids [18,19]. For, instance, the dino- of the endoplasmic reticulum. Lysosomes as one kind of Golgi deriva- flagellate Kryptoperidinium foliaceum carries a second nucleus belonging tives can only be taken into account, if the early proto-endosymbiont is to a -derived symbiont [20]. Even tertiary plastids caused by sufficiently protected against digestion and low pH. This is possible in three successive endocytotic events are known [21]. In these cases of several recent pathogenic bacteria. Some of them such Legionella secondary and tertiary endosymbiosis, a parasitic intrusion into the host pneumophila secrete effector molecules that prevent phagosome ma- appears rather unlikely. The concept of epibiotic , in turation [34], whereas others such as Chlamydia, Anaplasma and My- which epibiotic symbiosis precedes endosymbiosis [22], would also not cobacterium species prefer to hijack host that prevent the endo- be easily compatible with parasitic intrusion. However, such a pre- some/lysosome fusion [35]. Whether an ancestral proto-endosymbiont ceding epibiotic stage seems less likely in cases of endosymbiogenesis of may have been capable of hijacking the host’s membrane synthetic free-swimming partners, such as dinoflagellates and . With re- machinery to de novo produce a membrane according to own require- gard to the uptake of bacterial precursors of endosymbionts, a parasitic ments would only be conceivable in cases of prior endosomal lysis, as in intrusion into host cells is not certain, but cannot be excluded. Close Listeria and may be rather unlikely. For alternative possibilities of associations of , which have been discussed as possible pre- proto-endosymbiont incorporation see Fig. 1. cursors or relatives of eukaryotes, and bacteria are well-known. This After entry of the mitochondrial ancestor into the proto- includes alphaproteobacteria, a from which mitochondria have cell, a remarkable process of mitochondrial remodeling started. In originated. Such associations have been studied in mixed biofilms found morphological terms, the enlargement of the inner mitochondrial in the marine environment [23] in continental subsurface rocks [24], membrane took place, in favor of increased ATP production. The en- and in various other habitats. Multispecies biofilms are characterized by largement and associated folding may not be that much surprising as it intra- and interspecific communication, both quorum and community may appear at first glance. In bacteria including E. coli, experimental sensing, exchange of extracellular membrane vesicles, cheating and overexpression of membrane proteins leads to membrane curvatures, competition with, sometimes, battle-like dimensions under conditions intracellular membrane formation, and shifts in -to-protein ratio of limited resources [25]. Therefore, the situation in an ancient pre- and cardiolipin content [36]. In addition, the most importantly, it was endosymbiotic phase of evolution may have allowed both friendly and the host similarity outer membrane formation. aggressive interactions. For a long term of residents, the membrane modification is not Regardless of the mode of interaction between host and bacterial sufficient to avoid HIIIP. Host cells still can sense clues of the foreign ancestor, the relationship between them has not been free from pro- material covered by the outer membrane. The intruders must funda- blems. During the course of evolution, the proto-endosymbiont, whe- mentally change themselves to eliminate their bacterially originated ther intruder or prey, has evolved into the [26]. This materials as more as possible. Among them, DNA is the only one which event might occur some 1.5–2.0 billion years ago [27]. At that time, the can be evolved and inherited by next generation. Thus, modification of unicellular organisms had already evolved defense mechanism to DNA is the best choice for this purpose. Under the HIIIP, the intruder against unwelcome intruders, such as virus and bacteria. This me- has initiated the longest and most expensive modification on their chanism is referred as the intracellular innate immune response [28]. genes by the process referred as the DNA deletion. In other hand, these Thus, from the appearance of the early true eukaryotes there is a battle deleted DNA segments (genes) have to be integrated and expressed in of host cells and the intruders and this is also applied to the precursor of the host’s genomes, otherwise, the intruder also cannot thrive or sur- mitochondrion. Independently of the mode of uptake as either a prey or vive by malfunctions due to the gene loss. It is known that the extent of an uninvited intruder, the bacterium and its descendants had and still between the endosymbiont and the host. Such have to avoid lysosomal digestion as well as any attack from the cy- processes are generally common in endosymbiotic systems. They are tosolic side of the host. They have to mask themselves as “self” to the well-known from plastids, though to a smaller extent than in

121 D.-X. Tan Medical Hypotheses 127 (2019) 120–128

Fig. 1. A proposed simple process how does the mitochondrial precursor obtain the host plasma membrane when it initially enters inside the host cells. 1: A cell encounters a bacterium; 2: the process of endocytosis; 3: bacterium enters the cell and is targeted by the lysosome and resisted to lysis; 4. bacterium enters the Fig. 2. A proposed simple process of how deleted mtDNA segments integrate cell and de novo synthesizes a host membrane; 5: in all processes, the bacterium into nuclear DNA. This is a conserved process occurring probably at the time is covered by a host membrane [either as a derivative of the plasma membrane that the mitochondrial precursors enter into host cells. mtDNA has been re- (endosomal membrane) or of the Golgi apparatus (e.g., lysosomal membrane)]; ported to insert into nDNA currently [46,47]. Oval structure: mitochondria; 6: the bacterium evolves into mitochondrion. Large empty cycle: cell; small cycle with black dots: lysosome; red cycle: mtDNA; red pieces: deleted mtDNA solid cycle: nucleus; oval cycle in 3: lysosome; black curves in 4: membrane segments; cycle with DNA structure: nucleus. (For interpretation of the refer- lipids; blue structure: bacterium; oval structure in 6: mitochondrion. (For in- ences to color in this figure legend, the reader is referred to the web version of terpretation of the references to color in this figure legend, the reader is referred this article.) to the web version of this article.)

depend on the mechanism of mobilization. As soon as a gene copy has mitochondria. However, this is not only a matter of time periods in the been transferred to the nucleus, by whatever mechanism, it becomes billion year range, but can be even observed nowadays at the laboratory superfluous in the mitochondrion and can be deleted without dis- scale in real-time [37]. mtDNA transfers to the nucleus, which had advantage (Fig. 2). presumably started at the beginning of endosymbiosis, still continues After uptake of the proto-endosymbiont, one of the driving forces and has been documented as an also recent process [38,39] both in leading to mitochondrial gene erosion may be the host intracellular unicellular organism [40] or in [41] and this mtDNA in- innate immune pressure (HIIIP). serting into nDNA is believed to associate with aging [42]. Genomic comparisons revealed, in nuclear DNA, 630 orthologs of alphaproteo- bacterial origin, presumably derived from the proto-mitochondrial Host intracellular innate immune pressure (HIIIP) genome [43]. However, the loss of proto-mitochondrial genes has presumably been higher than the incorporation of genes into nuclear The terminology of intracellular innate immunity used here is dif- DNA. In , double-strand breaks of nuclear DNA were shown to be ferent from the adaptive innate immune activity which is performed fi repaired by means of mitochondrial DNA [44]. In phylogeny, primarily by the speci c immune cells. The terminology here is to refer nuclear DNA sequences of mitochondrial origin, numts (nuclear mi- the immune activity that occurs virtually in all cells and it is tochondrial DNA segments), were reported to have contributed to also referred as cell-autonomous immunity by some scientists [48]. This double-strand break repair [45]. In other words, DNA transfer can, at intracellular innate immune response has been present in the uni- least partially, be regarded as an aspect of host-endosymbiont co- cellular organisms [49] and it has been acquired before the appearance operation. An intriguing question is that of mechanisms that favor this of the adaptive immunity in the multicellular organisms. This cellular process, e.g, by using transposable elements. This cannot be answered self-defense has the potential to confer antimicrobial protection on for the proto-mitochondrial genome, and it is also difficult to deduce most, if not all, cells. this from recent bacteria with pathogenic or facultative endosymbiotic Since Intracellular defense has to have existed beginning from the ffi properties, because gene transfer to the host should be disfavorable for most ancient times of evolution, it is di cult to judge which of the early them. A hint may have come from Wolbachia, being members of Rick- defense strategies are still at work or may have been transformed into ettsiaceans and endosymbionts as well. These bacteria possess a high modern cell-protective mechanisms. It also seems important to distin- ff abundance of so-called group II introns, which possess transposable guish between di erent scenarios, such as (1) invasion of a bacterial properties [30]. At least, this bacterial family, from which mitochondria parasite that actively leaves the endosome; (2) the presence of an in- fi may have originated, has the potential for enhanced gene transfer by tracellular bacterium within host membrane structures, e.g., modi ed transposable elements. endosomes prevented from fusion with lysosomes, and Golgi- or ER- mtDNA erosion can have been associated with two different fates of derived vesicles; (3) the appearance of bacterial proteins or nucleic the removed segments, (a) transfer into the nucleus and incorporation acids in the host cytosol; and (4) dysfunctional endosymbionts in- into nuclear or (b) uptake into lysosomes via splice var- cluding mitochondria. Under condition (1), surface molecules can be iants of LAMP2 proteins and degradation, a process known as more easily detected by the host and, additionally, components released DNautophagy [46,47]. The decision between these possibilities may from the parasite may induce alarm mechanisms. This may be more difficult in case (2), but is highly likely in (3). Under condition (4),

122 D.-X. Tan Medical Hypotheses 127 (2019) 120–128 dysfunctionality must lead to signaling towards the host, in order to hand, cGAS appears as a mediator of cell stress reactions, but on the remove the problem. Under these perspectives, it seems necessary to other hand, it gives rise to the assumption to be involved in aging-re- distinguish between pathogen-associated molecular patterns (PAMPs) lated diseases [56]. This conclusion is well in accordance with recent and damage-associated molecular patterns (DAMPs) [50]. The most findings on a Sting gene that is associated with low-risk typical PAMPs are surface molecules of the bacterial or com- of aging-related diseases, presumably by reducing inflammaging [57]. ponents of bacterial flagella, i.e., molecules not present in mitochon- Another question is that of the conditions under which mtDNA is dria. However, nucleic acids can be also identified as PAMPs because of released to the cytosol. Unfortunately, low-rate release as has to be differences to their eukaryotic counterparts, concerning base composi- expected in relation to mitochondrial gene erosion has not been suffi- tion, secondary structures and enzymatic modification [50]. Recogni- ciently studied. Instead, the current knowledge is almost exclusively tion of non-self nucleic acids is known already from bacteria, e.g., by based on viral or bacterial infections or on oxidative mtDNA damage using restriction endonucleases, but this rather reflects antiviral stra- that induce the cytosolic appearance of mtDNA, as recently reviewed tegies, whereas recognition of foreign DNA or RNA in eukaryotic cells is [58]. As soon as mtDNA or, even more, oxidatively modified mtDNA based on other molecules. This may also be a matter of antiviral pro- appears in the cytosol, this causes NLRP3 inflammasome activation tection, e.g., in the detection of double-stranded RNA (dsRNA), but can [58–60]. In other words, released mtDNA, whether oxidized or not, extend to DAMPs. Therefore, mitochondrial nucleic acids are of parti- appears as PAMP or DAMP signals, which fuel HIIIP, with consequences cular interest in the context of the intracellular innate immune system. to the expression of proinflammatory cytokines, to mitophagy or to The innate immune system is able to detect signals from both out- apoptosis. At first glance, mitophagy may be perceived as an aggressive side and inside the cell. With regard to the topic of this article, we shall action of the host towards mitochondria. However, the purpose of this focus on the respective intracellular signals. However, the intracellular process is rather to maintain a functional mitochondrial status in the presence and detection of a foreign molecule does not necessarily mean host cell by removing those or parts of the mitochondrial that it has derived in an intracellular process. For instance, several toll- network that release DAMP molecules. Mitophagy has, in fact, been like receptors (TLRs) are able to detect bacterial nucleic acids: TLR3 interpreted as a means for warranting the host’s tolerance towards well- recognizes dsRNA, TLR7 and TLR8 detect purine-rich ssRNAs, whereas functioning mitochondria [50]. This conclusion contrasts with the de- TLR9 recognizes dsDNA [51]. Contrary to other TLRs, which are pre- fense against intracellular bacteria, which are removed by autophagy in dominantly located in the plasma membrane, they are mainly acting in order to eliminate the uninvited intruder [61,62]. The processes men- the endosomal compartment. Therefore, they come into contact with tioned above occur intracellularly but in outside of mitochondria. foreign molecules after these have been taken up by endocytosis. Thus, The discovery of widespread mtDNA deletions and of competition they detect external signals after internalization. One might argue that between wildtype and others carrying deletions of variable the TLRs have to reach the endosomes by trafficking through other lengths [63] indicate that these events occur inside mitochondria. If as intracellular membranous structures, from their site of synthesis (ER) we have hypothesized that mtDNA alternations is the result of HIIIP, via coat protein complex II (COPII)-coated vesicles to the Golgi. How- the question is how and which of the intracellular innate immune ef- ever, TLRs present in these other compartments are not necessarily fectors enter inside mitochondria active. As shown for TLR7, this receptor is activated in the endosome, One of the complications for appropriately judging the role of mu- by dimerization and proteolytic loss of its N-terminal domain [51]. tant mitochondrial chromosomes results from the fact that a typical However, there seems to be a functional difference in TLR9, which is single cell contains thousands of mt nucleoids [63], which are either also present in intracellular compartments and endosomally active. distributed over many mitochondria or are present in large mitochon- TLR9 recognizes DNA that is devoid of methyl groups in CpG islands, drial networks, which may, in the extreme of skeletal muscle fibers, such as mtDNA. If mtDNA has been liberated as a DAMP component extend as a single, connected structure over a syncytium derived from and appears in the endosome, TLR9 can induce mitophagy [50]. several cells. These numerous nucleoids are not identical, but rather In addition to DAMP recognition in endosomes or other vesicular represent mixed populations of wild-type and mutated chromosomes. structures, cytosolic sensors exist. Foreign RNAs can be detected in the may concern single genes, but can also comprise deletions of cytosol by RIG-1 (retinoic acid inducible gene I) and MDA5 (melanoma different extension [63]. Although increased numbers of mutations do differentiation-associated protein 5), which both activate via the MAVS not necessarily lead to enhanced free radical formation, as shown in (mitochondrial antiviral-signaling protein)/IRF3/NF-κB pathway IFNα/ aging mitochondrial mutator mice [64], age-dependent increases in β and CCR5 expression [50]. These routes are primarily seen as anti- mutated variants of mt chromosomes should have an effect on mi- viral defense mechanisms. The mitochondrial role of MAVS is only in- tochondrial functionality. Although the nucleoids can be densely completely understood. Evidence has been presented for stimulation of packed with proteins, especially TFAM, mitochondrial mutation rates mitophagy under conditions of infections with RNA viruses [52].An are by a factor of 10–17 fold higher than those in the nucleus [65]. example for DNA detection in the cytosol is AIM2 (absent in melanoma TFAM-labeled nucleoids were shown to be located adjacent to or sur- 2), a component of the respective inflammasome and, thereby, inducer rounded by the intracristal space, but were also found in cristae-free of proinflammatory cytokines [53]. Again, the function of AIM2 is matrix space [65]. Moreover, nucleoids vary in their compaction by mainly seen in the context of antiviral, but also antibacterial defense. TFAM binding, depending on their state of activity [65,66]. Therefore, The relationship to mitochondria may be restricted to mitophagy [54]. their vulnerability should be assumed to vary with their functional Another cytosolic detector of foreign DNA is cGAS (cyclic GMP-AMP state. Nucleoids may also appear as larger clusters, which were found to synthase). The second messenger 2′,3′-cGAMP activates the ER protein be usually not homogeneously covered by TFAM [65]. Studies on nu- STING (stimulator of IFN genes), which leads to the formation of cleoid localization indicate binding to the inner mitochondrial mem- proinflammatory cytokines, especially interferons. Cytokine release brane, but it has remained uncertain as to whether this concerns all of indicates extracellular actions, findings that are compatible with a them [65]. primary role in antiviral defense, which is in accordance with ob- The heterogeneity of mitochondrial nucleoids raises the questions of servations in cGas or Sting definciencies. However, cGAS also detects possible competition between variants and driving forces for clonal mtDNA when appearing in the cytosol [50]. In this context, a surprising selection. According to recent studies, clonal expansion of variants aspect of aging was detected. In murine and fibroblasts, deple- takes place, may be influenced by aging and diseases, and does not tion of cGAS and STING counteracted cell senescence and prevented generally favor wild-type mtDNA [63]. For instance, mt chromosomes SASP (senescence-associated secretory phenotype) [55]. In turn, this carrying deletions may be favored by shorter replication times. A re- indicates that mtDNA detection in the cytosol by cGAS, i.e., an action of cently proposed model assumed product inhibition by mitochondrial the intracellular immune system, promotes senescence. On the one proteins on mRNA primers required for mtDNA replication and a

123 D.-X. Tan Medical Hypotheses 127 (2019) 120–128 selection advantage by diminished feedback in variants with deletions circadian oscillators [74]. Moreover, the fission/fusion balance is concerning the genes of feedback protein [63]. The ultimate of mi- altered by mitochondrial proliferation as stimulated by metabolic tochondrial heteroplasmy can be remarkably high and the extent of sensing [75]. deletions can concern substantial sections of the mt [63]. Whatever the selection advantage may be, clonal expansion of mtDNA Interestingly, similar sizes of mtDNA mutants have the trend to variants carrying deletions lead to a decline in ATP production, mi- segregate together, respectively [76]. This characteristic of mtDNA tochondrial dysfunctionality or, alternately, apoptosis. segregation is an important factor to enhance the deleted mutant se- In author’s opinion, the HIIIP may act as a major selection force for lection during the course of aging. Due to the highly heterogeneric mitochondria deletion and the clonal expansion of shortened mtDNA property of the multiple mtDNA copies (wild type and mutants) in a mutants (will discussed in Section “The speculated aging process based on mitochondrion, these selected mutants initially will not significantly the HMIIITA”). However, as emphasized above, reduction of mi- impact the mitochondrial function. However, as the selection process is tochondrial functionality can finally lead to DAMP signals appear. going on, the accumulation of the deleted variants finally will jeo- For host cells, the selection force is the increased ATP demand for pardize mitochondrial function and result in cell incompetence. thriving and for competing with their counterparts. Under the selection, host cells and intruders had to coevolve. The results are that (4). Autophagy is a typical intracellular innate immunome process that the intruders have lost more than 95% of their original bacterial genetic the host cells use to clean up the “foreigners”(bacteria, virus) and material. Many but by far not all of these bacterial genes have been damaged compartments in almost all cell types [62,77]. This pro- adopted by the host cells and integrated into their genomes. The host cess is triggered either by PAMP or DAMP [78,79]. Mitophagy is cells have become highly dependent on the abundant ATP supply pro- the autophagy that specifically targets the damaged mitochondrial vided by the endosymbionts that evolved to mitochondria. The outcome segments [71,80,81], usually generated by the mitochondrial fis- appears to be perfect. However, one should remember that there is no sion in postmitotic cells. Mitophagy is the direct evidence to in- free lunch in . The seemingly perfect endosymbiotic relationship dicate that mitochondria are under the HIIIP. The mtDNA, espe- has a bad by-product, i.e., aging which appears only in the cells after cially the wild type, and the bacterially originated cardiolipin, they obtaining mitochondria, particularly in the multicellular - when appearing at the OMM, are the likely triggers of mitophagy isms. I hypothesize that this aging process has originated from the in- [82–85]. Additional evidence of the HIIIP against mitochondria is nate immune competition between host cells and the mitochondria in the inflammasome formation which is another indicator of host the postmitotic stage of organisms. intracellular innate immune response [86]. The inflammasome initiator, NLRP3, is also localized in the damaged mitochondrial Several issues related to this hypothesis membrane [87]. The chronic low-grade inflammation is often as- sociated with aging process and it is termed as inflammaging (1). Aging in organisms or at the cellular level mainly occurs in the [88–90]. The mtDNA deletion and intracellular chronic low-grade postmitotic cells. This new role of HIIIP only applies to postmitotic inflammation triggered by mtDNA or mitochondrial cardiolipin are cells and not to meiotic and mitotic cells due to the potentially the twins resulting from the continuously HIIIP against the mi- different mechanisms related to mitochondrial fission (as will dis- tochondria. Mitochondria also play a critical role involving in- cuss later). tracellular innate immune activity [50]. (2). Mitochondria still contain materials of bacterial origin which can (5). Free radical attack on the mtDNA primarily induces point muta- become targets of HIIIP. These mainly include mtDNA, and cardi- tions. Several studies have failed to support that olipin. Cardiolipin, a typical lipid of the inner mitochondrial accumulated in the mitochondria is a substantial cause of aging, membrane (IMM) of bacterial origination. Insofar, it cannot re- but data support the accumulation of deleted mutations being the present a general HIIIP target, although its scramblase-mediated culprit of aging in mice or in human [91,92]. Thus, free radical appearance in the OMM serves as a mitophagic signal. Only attack may not be the primary causative factor of aging, but it mtDNA has been subject to evolution and its variants can be se- causes some diseases that will lead to accelerate the aging process. lected. Thus, I speculated that the HIIIP is the selection force of the Under the normal condition, the free radical formation and the mtDNA modification and the principle driving force of aging. antioxidant system are in balance. The free radical and antioxidant (3). Various of cellular events trigger mitochondrial fission. For in- imbalance often occurs under mitochondrial stress [93–95]. Thus, stance, the mitochondrial fission mechanisms acting during it is possible that excessive free radical production may be an ac- or differ from those in the postmitotic cells. During tive mechanism of the mitochondrial response to the unfavorable the meiosis or mitosis, mitochondrial fission serves as homogenous environments they have experienced. Imbalanced free radical for- distribution of the genetic material (mtDNA) to the daughter mi- mation causes the cascade reaction including mitochondrial tochondria to keeps the mitochondrial population relatively uni- permeable transient pore (mptp) opening, cytochrome C release form and healthy [67–70]. This type of fission is obvious not re- and host cell apoptosis. This is probably another side of the con- lated to the aging process in the but may be associated tinuous battle of host cells and mitochondria. The overwhelming with the different lifespans among species (not to be discussed in generation of the free radicals by mitochondria may mainly as- the current paper). In postmitotic cells, mitochondrial fission is sociate with some diseases and has little to do with the nature mainly related to the elimination of damaged mitochondrial frag- aging per se. ments [71]. The processes of fission are well documented. Under (6). Adaptive immune response against mitochondria has been fre- the guidance of several elements which anchor in the mitochon- quently reported [96–98]. The antibodies against mitochondrially drial outer membrane including mitochondrial fission 1 protein encoded proteins, mtDNA, cardiolipin, etc. have been identified in (FIS1), the dynamin-related protein 1 (Drp1) is recruited from the patients [99–101]. Theses adaptive immune responses are collec- cytosol to the mitochondrial outer membrane [72]. The active tively classified as an autoimmune reaction. The reason is that the Drp1 then initiates mitochondria fi ssion. While mitochondrial fu- mitochondria are considered as part of ourselves by most of phy- sion is assumed to improve mitochondrial function [73]. However, sicians. Actually, our body recognize many mitochondrial elements fission and fusion are not exclusively related to mitophagy, but also as the foreign antigens when the immune surveillance cells en- occur independently, in both mitotic and postmitotic cells. This counter with these elements. This evidence indicates that the im- may be controlled by hormonal stimuli and/or calcium levels. The mune response between hosts and intruders occurs at a different fission/fusion cycle has been shown to be also controlled by level (the level of intercellularly). The adaptive immune response

124 D.-X. Tan Medical Hypotheses 127 (2019) 120–128

against mitochondria is probably not associated with the aging but If the functions of the mitochondria with the selected mutants are the diseases. not substantially impeded, the cells can still maintain their minimal requirements for survival and growth; however, the growth for the The speculated aging process based on the HMIIITA individuals will be slow down compared to the young ages. After re- moval of the wild type of mtDNA, the relatively large mutants will Immediately after birth, the newborn has mitochondria being become the next favorable targets of the HIIIP. This process continues dominated with wild type mtDNA populations. These mitochondria are until to the point that the selected mtDNA mutant is too small to obtained from mitochondrial fission during the process of meiosis from maintain the functions of mitochondria. The dysfunctional mitochon- maternal and then mitosis in embryonal development. In the early life, dria can no longer provide sufficient ATP for the host cell’s growth, but most of the cells are still going through mitosis, as we mentioned that to generate more free radicals to further damage the cell functions. The the mitochondrial fission in mitosis generates symmetric daughter mi- aging becomes obvious at this point. tochondria. After majority of the cells enter the postmitotic status, the asymmetric mitochondrial fission takes place. Mitochondrial fission at The testable feature and perspective postmitotic cell separates the damaged portion of mitochondria from their relatively healthy potion. The purpose of this type fission serves as HMIIITA is testable. For example, some of indirect evident support a method to eliminate damaged mitochondrial segments; however, this the HMIIITA. In murine and human fibroblasts, depletion of cGAS and process also becomes a passive selection force for mitochondrial evo- STING which are the host intracellular innate immune effectors pri- lution in individual postmitotic cell. Many factors can cause mi- marily to attack mtDNA counteracted cell senescence and prevented tochondrial damage. Here, the attention is given to the potential da- senescence-associated secretory phenotype [55]. In addition, when mage initiated by the HIIIP. Due to the bacterial origin with circular depleting the mtDNA in unicellular organism, rho0 yeast cells, it sig- structure and the unmethylated CpG islands [102], mtDNA seems the nificantly increased their longevity [108]. The association between most favorite target for the HIIIP. From an immunological point of aged phenotypes and increased resistance against cell has also view, the larger the mtDNA, the higher, the chance that it becomes the been observed in SK-Hep1 rho0 cells [109] which are mammalian cells target of HIIIP due to the larger mtDNA possessing more attacking sites being lack of mtDNA. The well-known immunosuppressive agents, cy- than that of the smaller one for HIIIP. Wild type probably is the largest closporin A and rapamycin, both reduced aging-related disorders and mtDNA among the mutants of deletion. Thus, most likely, the majority prolong the life span of mammals [110–113] even the additional me- of the immune actions will occur on the wild type mtDNA. To avoid this chanisms were involved for their anti-aging effects, the suppression of attack, mitochondria have a trait to reduce their mtDNA sizes by de- intracellular innate immunoresponse cannot be ignored. letion. This is the trait that has resulted in mitochondria having omitted If the new types of specific intracellular innate immunosuppressive more than 95% of their genome during evolution. Even through the molecules have been identified wich act at different sites or pathways, mutants of deletion in postmitotic cells cannot pass over to next gen- judging from the current evidence the combination of these molecules eration and contribute to evolution, they have still possessed this trait might significantly slow down the aging process and prolong the life acquired from maternal inheritance. As a result, mtDNA deletion occurs span of organisms. This may have the trade-off effects for more vul- in a regular base in postmitotic cells. It is not unlikely that the host nerable to certain infectious diseases. If the adaptive imminoresponse intracellular innate immune effectors, probably the or RNAs, was not impeded by these agents the infectious diseases could still be enter into mitochondria. If the size of these effectors are at the range of manageable due to the nature that the majority of infectious diseases 1.5 kD or less they can enter mitochondria via the mPTP which allows are finally controled by the adaptive immunity. Thus, this opportunity such sizes of substances to across [103]. The large molecular immune was worthy of investigation. effectors can also be translocated into the mitochondria by specificor non-specific translocators. As we mentioned above that TLR9 or cGAS are present intracellularly and also endosomally active. There is no Conclusion remarks reason to exclude their mitochondrially active. Both TLR9 and cGAS recognize mtDNA as PAMP and/or DAMP. If these host intracellular The HMIIITA hypothesizes that the mitochondrial gene erosion innate immune effectors enter mitochondria they will interact with the during evolution is mainly made by the HIIIP via mtDNA deletion and mtDNA to form an immune complex and/or damage it. mtDNA loca- the aging is the by-product of the of the host cell and mi- lizes close to the mitochondrial inner membrane and this immune at- tochondria in the postmitotic stage under the continuously HIIIP. tack will impact the intact of inner membrane and the ETC, reduce the Importantly, this theory is readily testable. membrane potential and significantly impede the mitochondrial func- There is no doubt that aging is caused by multiple internal and tion. Reduced membrane potential is the decisive factor to initiate the external factors which have been reviewed by countless publications. In mitochondrial asymmetric fission [104,105]. The specific mitochon- the current paper, the attention is only given to the host intracellular drial outer membrane proteins including FIS1 will first sense the altered innate immune competition between host cells and the mitochondria. mitochondrial membrane potential and recruit Drp1 to the mitochon- HIIIP may be one of the many factors which lead to cell as well as drial outer membrane from cytosol. The active Drp1, then, recruits organism aging. This HMIIITA is not aimed to replace any existed aging other effectors to initiate the mitochondrial fission process [106,107] theory but is only the complimentary to the 300 plus aging theories. (Fig. 3). This process leads to a mitochondrion dividing into two Hopefully, the HMIIITA will stimulate the enthusiasm in the fi eld of asymmetric fragments. One contains more wild type mtDNA which is aging research. heavily attacked by HIIIP and it will be subjected to mitophagy to be cleaned up [105]. The remaining fragment contains more of the deleted Conflict of interests mtDNA mutants which are less attacked by the host innate immune effectors and it is relatively healthy compared to the wild type. It then The author claims no conflict interest. will be fused with other mitochondrion or mitochondrial fragment via a process of mitochondrial fusion. For several fission-fusion cycles, the mtDNA deleted mutants are selected (Fig. 3). This selection lacks any Appendix A. Supplementary data long term of survival benefit for cells and it is considered as a passive procedure accompanied with elimination of the substantially damaged Supplementary data to this article can be found online at https:// mtDNA. It just selects the bad from the worst. doi.org/10.1016/j.mehy.2019.04.007.

125 D.-X. Tan Medical Hypotheses 127 (2019) 120–128

Fig. 3. The proposed selection process of the mtDNA deletion mutants associated with aging. 1: mitochondrion with wild type mtDNA (red cycles); 2; mtDNA deletion (small purple cycles) evolves and segregates together with age; 3: the immune effectors (empty cross) attack the wild type mtDNA to form the immune complex which reduces mi- tochondrial membrane potential, the mitochondrial outer membrane proteins FIS1(black arrows) sense this alteration of membrane potential and recruits the DRP1(yellow arrows) to relocate them from cytosol to mitochondrial outer membrane; 4: DRP1 initiates the mitochondrial fission process and se- parate the mitochondrion into two fragments; 5B: the dysfunctional fragment with more wild type mtDNA; 8: 5B is mitophaged and digested by the process of lysosome enzymes (black dots); 5A: the relatively healthy fragment with more deleted mutation of mtDNA; 5C: another relatively healthy fragment; 6: 5A and 5C enter the mitochondrial fusion process; 7: a new mitochondrion is formed with selection of deleted mutation of mtDNA. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

References two dinotoms has generated little change in the mitochondrial genomes of their dinoflagellate hosts and diatom endosymbionts. PLoS One 2012;7:e43763https:// doi.org/10.1371/journal.pone.0043763. [1] Diaconeasa AG, Rachita M, Stefan-van Staden R-I. A new hypothesis of aging. Med [19] Kamikawa R, Tanifuji G, Kawachi M, Miyashita H, Hashimoto T, Inagaki Y. – Hypotheses 2015;84:252 7. https://doi.org/10.1016/j.mehy.2015.01.006. genome-based phylogeny pinpointed the origin of the green-colored plastid in the [2] Miquel J, Economos AC, Fleming J, Johnson JE. Mitochondrial role in cell aging. dinoflagellate Lepidodinium chlorophorum. Genome Biol Evol 2015;7:1133–40. – Exp Gerontol 1980;15:575 91. https://doi.org/10.1093/gbe/evv060. [3] Harman D. Aging: a theory based on free radical and radiation chemistry. J [20] Figueroa RI, Bravo I, Fraga S, Garcés E, Llaveria G. The life history and of – Gerontol 1956;11:298 300. Kryptoperidinium foliaceum, a dinoflagellate with two eukaryotic nuclei. [4] Panigrahy SK, Bhatt R, Kumar A. Reactive oxygen species: sources, consequences 2009;160:285–300. https://doi.org/10.1016/j.protis.2008.12.003. – and targeted therapy in type 2 diabetes. J Drug Target 2017;25:93 101. https:// [21] Archibald JM. The puzzle of plastid evolution. Curr Biol 2009;19:R81–8. https:// doi.org/10.1080/1061186X.2016.1207650. doi.org/10.1016/j.cub.2008.11.067. [5] Munro D, Treberg JR. A radical shift in perspective: mitochondria as regulators of [22] Kooijman SALM, Auger P, Poggiale JC, Kooi BW. Quantitative steps in symbio- – reactive oxygen species. J Exp Biol 2017;220:1170 80. https://doi.org/10.1242/ genesis and the evolution of homeostasis. Biol Rev Camb Philos Soc jeb.132142. 2003;78:435–63. [6] Kim H-R, Won SJ, Fabian C, Kang M-G, Szardenings M, Shin M-G. Mitochondrial [23] Woodall LC, Jungblut AD, Hopkins K, Hall A, Robinson LF, Gwinnett C, et al. Deep- DNA aberrations and pathophysiological implications in hematopoietic diseases, sea anthropogenic macrodebris harbours rich and diverse communities of bacteria fl chronic in ammatory diseases, and . Ann Lab Med 2015;35:1. https://doi. and archaea. PLoS One 2018;13:e0206220https://doi.org/10.1371/journal.pone. org/10.3343/alm.2015.35.1.1. 0206220. [7] Akhmedov AT, Marín-García J. Mitochondrial DNA maintenance: an appraisal. [24] Escudero C, Vera M, Oggerin M, Amils R. Active microbial biofilms in deep poor – Mol Cell Biochem 2015;409:283 305. https://doi.org/10.1007/s11010-015- porous continental subsurface rocks. Sci Rep 2018;8:1538. https://doi.org/10. 2532-x. 1038/s41598-018-19903-z. [8] Szczepanowska K, Trifunovic A. Origins of mtDNA mutations in . Essays [25] Dang H, Lovell CR. Microbial surface colonization and biofilm development in – Biochem 2017;61:325 37. https://doi.org/10.1042/EBC20160090. marine environments. Microbiol Mol Biol Rev 2016;80:91–138. https://doi.org/ – [9] Kowald A. The mitochondrial theory of aging. Neurosignals 2001;10:162 75. 10.1128/MMBR.00037-15. https://doi.org/10.1159/000046885. [26] Sagan L. On the origin of mitosing cells. J Theor Biol 1967;14:255–74. [10] Kalous M, Drahota Z. The role of mitochondria in aging. Physiol Res [27] Kurland CG, Andersson SG. Origin and evolution of the mitochondrial proteome. – 1996;45:351 9. Microbiol Mol Biol Rev 2000;64:786–820. ff [11] Bu enstein R, Edrey YH, Yang T, Mele J. The oxidative stress theory of aging: [28] Tam JCH, Jacques DA. Intracellular immunity: finding the enemy within–how cells embattled or invincible? Insights from non-traditional model organisms. Age recognize and respond to intracellular pathogens. J Leukoc Biol 2014;96:233–44. – (Omaha) 2008;30:99 109. https://doi.org/10.1007/s11357-008-9058-z. https://doi.org/10.1189/jlb.4RI0214-090R. [12] Kennedy SR, Salk JJ, Schmitt MW, Loeb LA. Ultra-sensitive sequencing reveals an [29] Lindsey ARI, Bhattacharya T, Newton ILG, Hardy RW. Conflict in the intracellular age-related increase in somatic mitochondrial mutations that are inconsistent with of endosymbionts and viruses: a mechanistic look at wolbachia-mediated oxidative damage. PLoS Genet 2013;9:e1003794https://doi.org/10.1371/journal. pathogen-blocking. Viruses 2018;10:141. https://doi.org/10.3390/v10040141. pgen.1003794. [30] Leclercq S, Giraud I, Cordaux R. Remarkable abundance and evolution of mobile [13] Itsara LS, Kennedy SR, Fox EJ, Yu S, Hewitt JJ, Sanchez-Contreras M, et al. group II introns in Wolbachia bacterial endosymbionts. Mol Biol Evol Oxidative stress is not a major contributor to somatic mitochondrial DNA muta- 2011;28:685–97. https://doi.org/10.1093/molbev/msq238. tions. PLoS Genet 2014;10:e1003974https://doi.org/10.1371/journal.pgen. [31] Tattoli I, Sorbara MT, Philpott DJ, Girardin SE. Stalling autophagy: a new function 1003974. for Listeria phospholipases. Microb Cell (Graz, Austria) 2014;1:48–50. https://doi. [14] Sakellariou GK, Pearson T, Lightfoot AP, Nye GA, Wells N, Giakoumaki II, et al. org/10.15698/mic2014.01.124. Long-term administration of the mitochondria-targeted antioxidant mitoquinone [32] Mileykovskaya E, Dowhan W. Cardiolipin membrane domains in and mesylate fails to attenuate age-related oxidative damage or rescue the loss of eukaryotes. Biochim Biophys Acta – Biomembr 2009;1788:2084–91. https://doi. muscle mass and function associated with aging of skeletal muscle. FASEB J org/10.1016/j.bbamem.2009.04.003. – 2016;30:3771 85. https://doi.org/10.1096/fj.201600450R. [33] Arechaga I. Membrane invaginations in bacteria and mitochondria: common fea- [15] Kang D, Hamasaki N. Mitochondrial transcription factor A in the maintenance of tures and evolutionary scenarios. J Mol Microbiol Biotechnol 2013;23:13–23. – mitochondrial DNA. Ann N Y Acad Sci 2005;1042:101 8. https://doi.org/10. https://doi.org/10.1159/000346515. 1196/annals.1338.010. [34] Prashar A, Terebiznik MR. Legionella pneumophila: homeward bound away from [16] Kienhöfer J, Häussler DJF, Ruckelshausen F, Muessig E, Weber K, Pimentel D, et al. the phagosome. Curr Opin Microbiol 2015;23:86–93. https://doi.org/10.1016/j. Association of mitochondrial antioxidant enzymes with mitochondrial DNA as mib.2014.11.008. – integral constituents. FASEB J 2009;23:2034 44. https://doi.org/10. [35] Toledo A, Benach JL. Hijacking and Use of Host Lipids by Intracellular Pathogens. 1096/fj.08-113571. Virulence Mech. Bact. Pathog. Fifth Ed.American Society of ; 2015. p. [17] Wilson ACC, Duncan RP. Signatures of host/symbiont genome coevolution in in- 637–66. 10.1128/microbiolspec.VMBF-0001-2014. – sect nutritional endosymbioses. Proc Natl Acad Sci 2015;112:10255 61. https:// [36] Jamin N, Garrigos M, Jaxel C, Frelet-Barrand A, Orlowski S. Ectopic neo-formed doi.org/10.1073/pnas.1423305112. intracellular membranes in escherichia coli: a response to membrane protein- [18] Imanian B, Pombert J-F, Dorrell RG, Burki F, Keeling PJ. Tertiary endosymbiosis in

126 D.-X. Tan Medical Hypotheses 127 (2019) 120–128

induced stress involving membrane curvature and domains. Biomolecules 1016/j.biocel.2018.10.012. 2018;8:88. https://doi.org/10.3390/biom8030088. [66] Kukat C, Davies KM, Wurm CA, Spåhr H, Bonekamp NA, Kühl I, et al. Cross-strand [37] Shapiro JA. Constraint and opportunity in genome innovation. RNA Biol binding of TFAM to a single mtDNA molecule forms the mitochondrial nucleoid. 2014;11:186–96. https://doi.org/10.4161/rna.27506. Proc Natl Acad Sci USA 2015;112:11288–93. https://doi.org/10.1073/pnas. [38] Thorsness PE, Fox TD. Escape of DNA from mitochondria to the nucleus in 1512131112. . Nature 1990;346:376–9. https://doi.org/10.1038/ [67] Lee S, Kim S, Sun X, Lee J-H, Cho H. Cell cycle-dependent mitochondrial bio- 346376a0. genesis and dynamics in mammalian cells. Biochem Biophys Res Commun [39] Ricchetti M, Tekaia F, Dujon B. Continued colonization of the human genome by 2007;357:111–7. https://doi.org/10.1016/j.bbrc.2007.03.091. mitochondrial DNA. PLoS Biol 2004;2:E273. https://doi.org/10.1371/journal. [68] Wang F, Liu P, Zhang Q, Zhu J, Chen T, Arimura S, et al. Phosphorylation and pbio.0020273. ubiquitination of dynamin-related proteins (AtDRP3A/3B) synergically regulate [40] Cheng X, Ivessa AS. Accumulation of linear mitochondrial DNA fragments in the mitochondrial proliferation during mitosis. Plant J 2012;72:43–56. https://doi. nucleus shortens the chronological life span of yeast. Eur J Cell Biol org/10.1111/j.1365-313X.2012.05052.x. 2012;91:782–8. https://doi.org/10.1016/j.ejcb.2012.06.005. [69] Jin J-X, Lee S, Taweechaipaisankul A, Kim GA, Lee BC. Melatonin regulates lipid [41] Caro P, Gómez J, Arduini A, González-Sánchez M, González-García M, Borrás C, in porcine oocytes. J Pineal Res 2017;62:e12388https://doi.org/10. et al. Mitochondrial DNA sequences are present inside nuclear DNA in rat tissues 1111/jpi.12388. and increase with age. Mitochondrion 2010;10:479–86. https://doi.org/10.1016/ [70] Zhao X-M, Wang N, Hao H-S, Li C-Y, Zhao Y-H, Yan C-L, et al. Melatonin improves j.mito.2010.05.004. the fertilization capacity and developmental ability of bovine oocytes by reg- [42] Martínez-Cisuelo V, Gómez J, García-Junceda I, Naudí A, Cabré R, Mota-Martorell ulating cytoplasmic maturation events. J Pineal Res 2018;64:e12445https://doi. N, et al. Rapamycin reverses age-related increases in mitochondrial ROS produc- org/10.1111/jpi.12445. tion at complex I, oxidative stress, accumulation of mtDNA fragments inside nu- [71] El-Hattab AW, Suleiman J, Almannai M, Scaglia F. Mitochondrial dynamics: bio- clear DNA, and lipofuscin level, and increases autophagy, in the liver of middle- logical roles, molecular machinery, and related diseases. Mol Genet Metab 2018. aged mice. Exp Gerontol 2016;83:130–8. https://doi.org/10.1016/j.exger.2016. https://doi.org/10.1016/j.ymgme.2018.10.003. 08.002. [72] Tilokani L, Nagashima S, Paupe V, Prudent J. Mitochondrial dynamics: overview [43] Gabaldón T, Huynen MA. Reconstruction of the proto-mitochondrial metabolism. of molecular mechanisms. Essays Biochem 2018;62:341–60. https://doi.org/10. Science 2003;301:609. https://doi.org/10.1126/science.1085463. 1042/EBC20170104. [44] Ricchetti M, Fairhead C, Dujon B. Mitochondrial DNA repairs double-strand breaks [73] Tan D-X, Manchester LC, Qin L, Reiter RJ. Melatonin: a mitochondrial targeting in yeast chromosomes. Nature 1999;402:96–100. https://doi.org/10.1038/47076. molecule involving mitochondrial protection and dynamics. Int J Mol Sci [45] Hazkani-Covo E, Covo S. Numt-mediated double-strand break repair mitigates 2016;17:2124. https://doi.org/10.3390/ijms17122124. deletions during primate genome evolution. PLoS Genet 2008;4:e1000237https:// [74] de Goede P, Wefers J, Brombacher EC, Schrauwen P, Kalsbeek A. Circadian doi.org/10.1371/journal.pgen.1000237. rhythms in mitochondrial respiration. J Mol Endocrinol 2018;60:R115–30. [46] Fujiwara Y, Kikuchi H, Aizawa S, Furuta A, Hatanaka Y, Konya C, et al. Direct https://doi.org/10.1530/JME-17-0196. uptake and degradation of DNA by lysosomes. Autophagy 2013;9:1167–71. [75] Halling JF, Ringholm S, Olesen J, Prats C, Pilegaard H. Exercise training protects https://doi.org/10.4161/auto.24880. against aging-induced mitochondrial fragmentation in mouse skeletal muscle in a [47] Fujiwara Y, Hase K, Wada K, Kabuta T. An RNautophagy/DNautophagy receptor, PGC-1α dependent manner. Exp Gerontol 2017;96:1–6. https://doi.org/10.1016/ LAMP2C, possesses an arginine-rich motif that mediates RNA/DNA-binding. j.exger.2017.05.020. Biochem Biophys Res Commun 2015;460:281–6. https://doi.org/10.1016/j.bbrc. [76] Ling F, Niu R, Hatakeyama H, Goto Y-I, Shibata T, Yoshida M. Reactive oxygen 2015.03.025. species stimulate mitochondrial allele segregation toward homoplasmy in human [48] Randow F, MacMicking JD, James LC. Cellular self-defense: how cell-autonomous cells. Mol Biol Cell 2016;27:1684–93. https://doi.org/10.1091/mbc.E15-10-0690. immunity protects against pathogens. Science (80) 2013(340):701–6. https://doi. [77] Evans RJ, Sundaramurthy V, Frickel E-M. The interplay of host autophagy and org/10.1126/science.1233028. eukaryotic pathogens. Front Cell Dev Biol 2018;6:118. https://doi.org/10.3389/ [49] Uehling J, Deveau A, Paoletti M. Do fungi have an innate immune response? An fcell.2018.00118. NLR-based comparison to plant and animal immune systems. PLOS Pathog [78] Zhang Q, Kang R, Zeh III HJ, Lotze MT, Tang D. DAMPs and autophagy. Autophagy 2017;13:e1006578https://doi.org/10.1371/journal.ppat.1006578. 2013;9:451–8. https://doi.org/10.4161/auto.23691. [50] Rongvaux A. Innate immunity and tolerance toward mitochondria. Mitochondrion [79] Deretic V. Multiple regulatory and effector roles of autophagy in immunity. Curr 2018;41:14–20. https://doi.org/10.1016/j.mito.2017.10.007. Opin Immunol 2009;21:53–62. https://doi.org/10.1016/j.coi.2009.02.002. [51] Petes C, Odoardi N, Gee K. The toll for trafficking: toll-like receptor 7 delivery to [80] Chu CT. Mechanisms of selective autophagy and mitophagy: implications for the endosome. Front Immunol 2017;8:1075. https://doi.org/10.3389/fimmu. neurodegenerative diseases. Neurobiol Dis 2018. https://doi.org/10.1016/j.nbd. 2017.01075. 2018.07.015. [52] Sun X, Sun L, Zhao Y, Li Y, Lin W, Chen D, et al. MAVS maintains mitochondrial [81] Zimmermann M, Reichert AS. How to get rid of mitochondria: crosstalk and reg- homeostasis via autophagy. Cell Discov 2016;2:16024. https://doi.org/10.1038/ ulation of multiple mitophagy pathways. Biol Chem 2017;399:29–45. https://doi. celldisc.2016.24. org/10.1515/hsz-2017-0206. [53] Ma Z, Ni G, Damania B. Innate sensing of DNA virus genomes. Annu Rev Virol [82] Van Houten B, Hunter SE, Meyer JN. Mitochondrial DNA damage induced au- 2018;5:341–62. https://doi.org/10.1146/annurev-virology-092917-043244. tophagy, cell death, and disease. Front Biosci (Landmark Ed) 2016;21:42–54. [54] Harris J, Lang T, Thomas JPW, Sukkar MB, Nabar NR, Kehrl JH. Autophagy and [83] Liu S, Feng M, Guan W. Mitochondrial DNA sensing by STING signaling partici- inflammasomes. Mol Immunol 2017;86:10–5. https://doi.org/10.1016/j.molimm. pates in inflammation, and beyond. Int J Cancer 2016;139:736–41. https:// 2017.02.013. doi.org/10.1002/ijc.30074. [55] Glück S, Guey B, Gulen MF, Wolter K, Kang T-W, Schmacke NA, et al. Innate [84] Ryan T, Bamm VV, Stykel MG, Coackley CL, Humphries KM, Jamieson-Williams R, immune sensing of cytosolic chromatin fragments through cGAS promotes senes- et al. Cardiolipin exposure on the outer mitochondrial membrane modulates α- cence. Nat Cell Biol 2017;19:1061–70. https://doi.org/10.1038/ncb3586. synuclein. Nat Commun 2018;9:817. https://doi.org/10.1038/s41467-018- [56] Glück S, Ablasser A. Innate immunosensing of DNA in cellular senescence. Curr 03241-9. Opin Immunol 2018;56:31–6. https://doi.org/10.1016/j.coi.2018.09.013. [85] Dudek J. Role of cardiolipin in mitochondrial signaling pathways. Front Cell Dev [57] Hamann L, Ruiz-Moreno JS, Szwed M, Mossakowska M, Lundvall L, Schumann RR, Biol 2017;5:90. https://doi.org/10.3389/fcell.2017.00090. et al. STING SNP R293Q Is associated with a decreased risk of aging-related dis- [86] Próchnicki T, Latz E. Inflammasomes on the crossroads of innate immune re- eases. Gerontology 2018:1–10. https://doi.org/10.1159/000492972. cognition and metabolic control. Cell Metab 2017;26:71–93. https://doi.org/10. [58] Fang C, Wei X, Wei Y. Mitochondrial DNA in the regulation of innate immune 1016/j.cmet.2017.06.018. responses. Protein Cell 2016;7:11–6. https://doi.org/10.1007/s13238-015- [87] Shi C-S, Kehrl JH. Cytochrome c negatively regulates NLRP3 inflammasomes. PLoS 0222-9. One 2016;11:e0167636https://doi.org/10.1371/journal.pone.0167636. [59] Nakahira K, Haspel JA, Rathinam VAK, Lee S-J, Dolinay T, Lam HC, et al. [88] Cardinali DP, Hardeland R. Inflammaging, metabolic syndrome and melatonin: a Autophagy proteins regulate innate immune responses by inhibiting the release of call for treatment studies. Neuroendocrinology 2017;104:382–97. https://doi.org/ mitochondrial DNA mediated by the NALP3 inflammasome. Nat Immunol 10.1159/000446543. 2011;12:222–30. https://doi.org/10.1038/ni.1980. [89] Müller-Werdan U, Nuding S, Ost M. Assessing inflammageing. Curr Opin Clin Nutr [60] Shimada K, Crother TR, Karlin J, Dagvadorj J, Chiba N, Chen S, et al. Oxidized Metab Care 2017;20:346–8. https://doi.org/10.1097/MCO.0000000000000391. mitochondrial DNA activates the NLRP3 inflammasome during apoptosis. [90] Giuliani A, Prattichizzo F, Micolucci L, Ceriello A, Procopio AD, Rippo MR. Immunity 2012;36:401–14. https://doi.org/10.1016/j.immuni.2012.01.009. Mitochondrial (Dys) function in inflammaging: do mitomirs influence the en- [61] Amano A, Nakagawa I, Yoshimori T. Autophagy in Innate Immunity against ergetic, oxidative, and inflammatory status of senescent cells? Mediators Inflamm Intracellular Bacteria. J Biochem 2006;140:161–6. https://doi.org/10.1093/jb/ 2017;2017:1–11. https://doi.org/10.1155/2017/2309034. mvj162. [91] Edgar D, Trifunovic A. The mtDNA mutator mouse: dissecting mitochondrial in- [62] da Siqueira MS, de Ribeiro RM, Travassos LH. Autophagy and its interaction with volvement in aging. Aging (Albany NY) 2009;1:1028–32. https://doi.org/10. intracellular bacterial pathogens. Front Immunol 2018;9:935. https://doi.org/10. 18632/aging.100109. 3389/fimmu.2018.00935. [92] Kazachkova N, Ramos A, Santos C, Lima M. Mitochondrial DNA damage patterns [63] Kowald A, Kirkwood TBL. Resolving the Enigma of the clonal expansion of mtDNA and aging: revising the evidences for humans and mice. Aging Dis 2013;4:337–50. deletions. Genes (Basel) 2018;9:126. https://doi.org/10.3390/genes9030126. https://doi.org/10.14336/AD.2013.0400337. [64] Thompson LV. Oxidative stress, mitochondria and mtDNA-mutator mice. Exp [93] Blajszczak C, Bonini MG. Mitochondria targeting by environmental stressors: im- Gerontol 2006;41:1220–2. https://doi.org/10.1016/j.exger.2006.10.018. plications for redox cellular signaling. Toxicology 2017;391:84–9. https://doi.org/ [65] Ježek P, Špaček T, Tauber J, Pavluch V. Mitochondrial Nucleoids: superresolution 10.1016/j.tox.2017.07.013. microscopy analysis. Int J Biochem Cell Biol 2019;106:21–5. https://doi.org/10. [94] Georgieva E, Ivanova D, Zhelev Z, Bakalova R, Gulubova M, Aoki I. Mitochondrial

127 D.-X. Tan Medical Hypotheses 127 (2019) 120–128

dysfunction and redox imbalance as a diagnostic marker of “free radical diseases”. [104] van der Bliek AM, Shen Q, Kawajiri S. Mechanisms of mitochondrial fission and Anticancer Res 2017;37:5373–81. https://doi.org/10.21873/anticanres.11963. fusion. Cold Spring Harb Perspect Biol 2013;5:a011072https://doi.org/10.1101/ [95] Chen Y, Zhou Z, Min W. Mitochondria, oxidative stress and innate immunity. Front cshperspect.a011072. Physiol 2018;9:1487. https://doi.org/10.3389/fphys.2018.01487. [105] Youle RJ, van der Bliek AM. Mitochondrial fission fusion, and stress. Science (80-) [96] Yamanaka T, Fukatsu T, Ichinohe Y, Hirata Y. Antimitochondrial antibodies-po- 2012;337:1062–5. https://doi.org/10.1126/science.1219855. sitive myositis accompanied by cardiac involvement. BMJ Case Rep 2017;2017. [106] Kalia R, Wang RY-R, Yusuf A, Thomas PV, Agard DA, Shaw JM, et al. Structural https://doi.org/10.1136/bcr-2016-218469. basis of mitochondrial receptor binding and constriction by DRP1. Nature [97] Long SA, Van de Water J, Gershwin ME. Antimitochondrial antibodies in primary 2018;558:401–5. https://doi.org/10.1038/s41586-018-0211-2. biliary cirrhosis: the role of xenobiotics. Autoimmun Rev 2002;1:37–42. [107] Hu C, Huang Y, Li L. Drp1-dependent mitochondrial fission plays critical roles in [98] Ramos-Casals M, Pares A, Jara L-J, Solans R, Vinas O, Vazquez P, et al. physiological and pathological progresses in mammals. Int J Mol Sci 2017:18. Antimitochondrial antibodies in patients with chronic hepatitis C virus infection: https://doi.org/10.3390/ijms18010144. description of 18 cases and review of the literature. J Viral Hepat 2005;12:648–54. [108] Woo DK, Poyton RO. The absence of a mitochondrial genome in rho0 yeast cells https://doi.org/10.1111/j.1365-2893.2005.00642.x. extends lifespan independently of retrograde regulation. Exp Gerontol [99] Toubi E, Halas K, Ben-David J, Sabo E, Kessel A, Luntz M. Immune-mediated 2009;44:390–7. https://doi.org/10.1016/j.exger.2009.03.001. disorders associated with idiopathic sudden sensorineural hearing loss. Ann Otol [109] Park SY, Choi B, Cheon H, Pak YK, Kulawiec M, Singh KK, et al. Cellular aging of Rhinol Laryngol 2004;113:445–9. https://doi.org/10.1177/ mitochondrial DNA-depleted cells. Biochem Biophys Res Commun 000348940411300605. 2004;325:1399–405. https://doi.org/10.1016/j.bbrc.2004.10.182. [100] Aguirre-Villarreal D, Whittall LP, Fragoso-Loyo H, Cantú-Brito C, Crispín JC, [110] Karlsson J, Fong KSK, Hansson MJ, Elmàr E, Csiszar K, Keep MF. Life span ex- Valdés-Ferrer SI. Intrathecal formation of anticardiolipin antibodies in a patient tension and reduced neuronal death after weekly intraventricular cyclosporin in- with SLE-related relapsing longitudinal myelitis: a possible pathogenic connection. jections in the G93A transgenic mouse model of amyotrophic lateral sclerosis. J Lupus 2018. https://doi.org/10.1177/0961203318810430. Neurosurg 2004;101:128–37. https://doi.org/10.3171/jns.2004.101.1.0128. [101] Wang H, Li T, Chen S, Gu Y, Ye S. Neutrophil extracellular trap mitochondrial DNA [111] Jones MG, Harris G. Prolongation of life in female NZB/NZW (F1) hybrid mice by and its autoantibody in systemic lupus erythematosus and a proof-of-concept trial cyclosporin A. Clin Exp Immunol 1985;59:1–9. of metformin. Arthritis Rheumatol (Hoboken, NJ) 2015;67:3190–200. https://doi. [112] Bitto A, Ito TK, Pineda VV, LeTexier NJ, Huang HZ, Sutlief E, et al. Transient org/10.1002/art.39296. rapamycin treatment can increase lifespan and healthspan in middle-aged mice. [102] Krieg AM. CpG Motifs in bacterial DNA and their immune effects. Annu Rev Elife 2016:5. https://doi.org/10.7554/eLife.16351. Immunol 2002;20:709–60. https://doi.org/10.1146/annurev.immunol.20. [113] Arriola Apelo SI, Pumper CP, Baar EL, Cummings NE, Lamming DW. Intermittent 100301.064842. administration of rapamycin extends the life span of female C57BL/6J mice. J [103] Halestrap A, Clarke SJ, Javadov SA. Mitochondrial permeability transition pore Gerontol Ser A Biol Sci Med Sci 2016;71:876–81. https://doi.org/10.1093/ opening during myocardial reperfusion – a target for cardioprotection. Cardiovasc gerona/glw064. Res 2004;61:372–85. https://doi.org/10.1016/S0008-6363(03)00533-9.

128