Mitochondria are morphologically distinctive double membrane bound cell organelles, suggested to have evolved from endosymbiosis of α-proteobacteria. Mitochondria are indispensable for diverse forms of eukaryotic cells with ubiquitous distribution throughout the eukaryotic world. Associated with electron transport system coupled with oxidative phosphorylation, mitochondria have been considered as cell’s power house for the generation of energy rich adenosine triphosphate (ATP). However, recent findings have implicated crucial role of mitochondria in biosynthesis of various cellular metabolites, survival, ageing, apoptosis, and development of multiple disease phenotypes. For this reason the past couple of decades, understanding of mitochondrial genome functioning has received considerable attention.
Mitochondria, morphologically distinctive double membrane bound organelles, hypothesized to have evolved from endosymbiotic α-proteobacteria, are indispensable for the vitality of eukaryotic life. They are ubiquitously found throughout the eukaryotic systems. They use electron transport coupled with oxidative phosphorylation to generate ATP (Saraste, 1999). For decades, mitochondria were considered only as cell’s power house. But recent findings are that these organelles play vital roles also in apoptosis and survival, the ageing process, and in several diseases, biosynthesis of amino acids and steroids, β-oxidation of fatty acids, FeS metabolism and physiological adaptations to environmental stress (Birsoy et al., 2015). Due to their endosymbiotic origin mitochondria are the only organelles in animal cells, equivalent to plastid in plant cells which have their own genome (Vothknecht and Soll, 2007). The genome of mitochondria contains a minute subset of genes (Anderson et al., 1981) inherited and retained from their α-proteobacterial ancestors, with other genes of prokaryotic endosymbiotic origin either having been lost or else retained by nuclear chromosomes of the cell (Esser et al., 2004). Genes in mitochondria encode mitochondrial proteins and RNAs (Pesole et al., 2012).
Mitochondrial genome of many eukaryotes had been considered to exist in the form of super-coiled circular DNA. However, there is now strong evidence that many of these circular mapped mtDNA consists primarily of linear, multimeric head-to-tail concatemers (Bendich, 1993, 1996). These, for example, have been found in few unrelated organisms such as ciliates, Apicomplexa (Plasmodium and its relatives), fungi, Chlorophycean green algae (Chlamydomonas and relatives) and several Cnidarian animals. These linear molecules contain various specialized end structures, such as covalently closed single-stranded DNA termini and terminally attached proteins. They also tend to have telomere-like repeats of differing lengths. In fact, the difference in size of mitochondrial genomes is mostly caused by the variations in the length and organization of intergenic regions which, in some cases are consisting of extensive tandem-repeats or stem-loop motifs.
Mitochondrial DNA is compact and less than 20 kilo bases in length (Wolstenholme, 1992) encoding protein subunits of the mitochondrial respiratory chain (Okimoto et al., 1992). There also exist structures called nucleoids, which are mainly composed of 2-8mtDNAcopies (Legros et al., 2004). Nucleoids are associated with the inner mitochondrial membrane and distributed throughout the mitochondrial network at regular spatial intervals (Prachar, 2010). Generally, mitochondrial DNA is multi copied with a 16,569-bp double-stranded circular molecule located within the matrix of the mitochondrion, inherited from the maternal oocyte and is self-replicative.
Mammalian mt-DNA (Fig. 1A) contains 37 genes, out of which 13 encode mitochondrial oxidative phosphorylation complex proteins, 22 encode transfer RNA (tRNA) and others encode ribosomal RNA. In contrast to nuclear genome, mitochondria generally encode two kinds of rRNAs, 16S rRNA and 12S rRNA which participate in the synthesis of 13 mtDNA-encoded proteins. The mRNAs are specific for integral membrane components of the mitochondrial electron transport chain, the rRNAs are involved to produce subunits of mitochondrial ribosome, and the tRNAs mediate in the translation process of the 13 mRNAs by these dedicated mitochondrial ribosomes. Genes are employed using both strands of mtDNA; although they have been classified as heavy (H) and light (L) strands based on their relative buoyant densities in denaturing CsCl gradients.
Transcription initiates at the promoter site using H-strands (called HSP1 andHSP2) and a single L-strand (the LSP). The LSP and HSP1 are located in the major non-coding region of the DNA called the displacement loop (D-loop) regulatory region, whereas HSP2 is located downstream of HSP1 within the tRNA Phe gene. Transcripts derived from HSP2 and LSP are mostly as long as the whole genome and forms polycistronic products, whereas those from HSP1 are terminated at a specific site (in the tRNALeu gene) downstream of the 16S rRNA and produce mainly the two rRNAs (12S and 16S). Because of such unique gene arrangement in mt-DNA i.e., the rRNAs and most mRNAs are immediately flanked by tRNAs, tRNA processing is believed to be the major mechanism that liberates the majority of the 37 mature RNA molecules from the polycistronic primary transcripts.
In mammals, the mature mRNAs lack significant 5ʹ untranslated sequences and thus the mechanism of ribosome binding and mitochondrial translation initiation remains obscure. MtDNA encodes proteins that are structural subunits of complexes I, III, IV and V of the respiratory chain (RC), (Fig. 1B) as detailed below (only complex II is nuclear encoded).
Figure 1 (A) The map of human mitochondrial DNA (16569bp). (OH and OL) Origins of heavy- and light-strand replication, respectively; (ND1-ND6) subunits of NADH dehydrogenase (ETS complex I) subunits 1–6; (COX1-COX3) subunits of cytochrome oxidase (ETC complex IV); (ATP6 and ATP8) subunit 6 and 8 of mitochondrial ATPase (complex V); (Cytb) cytochrome b (complex III). (B) Schematic representation of mitochondrial electron transport chain. The mitochondrial DNA encoded subunits are shown in their respective colors as shown in mitochondrial DNA map (Fig1A). Nuclear DNA encoded subunits are shown in light green color.
Seven components of RC complex I (CI, or NADH dehydrogenase complex), ND1, ND2, ND3, ND4, ND4L, ND5 and ND6 constitute a subset of the 14 catalytic proteins. These proteins are involved in the electron transfer and proton pumping activity of the CI (Guzy et al., 2006) cytochrome b (a component of RC complex III (CIII). Cytochrome b transfers the electrons from the ubisemiquinone at the Qo site of the complex) (outer surface of the inner membrane) to the ubisemiquinone located at the Qi site (inner surface of the inner membrane) (Fontanesi et al., 2008). Catalytic subunits 1, 2and 3 of RC complex IV (CIV) along with 10 additional proteins allow the electron transport process (Hong, 2004).Subunits of complex V (CV, or ATPase complex) include ATPase6 and A6L.The ATPase6 (or “a”) subunit is a trans-membrane protein of the Fo portion of the complex, involved in the passage of protons from the intermembrane space to the matrix. The function of subunit A6L is still unclear. Recent studies by Wittig et al. have indicated that both proteins ATPase6 and A6L play important role in the stabilization of ATP synthase dimers/oligomers.
As the histone proteins has the ability to some extent protect the nuclear genome from various kinds of DNA damaging agents but mitochondrial genome lacks them, hence, mitochondrial DNA is more prone to damage than nuclear DNA. The mitochondrial DNA is constantly exposed to various kinds of exogenous and endogenous DNA damaging agents, some of which can lead to various kinds of diseases including of neurodegenerative form, cancer, cardiomyopathy, diabetes and several aging-related disorders.
One of the major products of mitochondrial electron transport chain is reactive oxygen species (ROS) which frequently induces DNA damage (Fig. 2). Mitochondria accounts for about 90% oxygen consumption, about 1–5% of which is converted to superoxide anion (Papa, 1996) which is the main precursor of other ROS including the most toxic superoxide radicalO2 ‒• . The reduction potential for the conversion of O2 to O2 ‒• is about ‒0.160V and the mitochondrial electron transport chain has several redox centres with standard reduction potential ranging from -0.32V to +0.39V giving mitochondria a fairly reductive environment (Wood, 1987). For this reductive environment various respiratory component such as flavoproteins and Fe-S centres are thermodynamically capable monovalent reduction of molecular oxygen giving rise to the superoxide radicals. Seven potential sites for the generation of O2 ‒• exist in the mitochondrial matrix with complex I and Complex III showing highest rate of O2 ‒• production. These superoxide radicals cannot diffuse through the inner mitochondrial membrane and converted to H2O2 by the action of manganese-superoxide dismutase (Mn-SOD) but if this H2O2 remains within the matrix it will undergo Fenton chemistry with Fe(II) to form hydroxyl radicals (HO•), which show strong reactivity towards DNA and polyunsaturated fatty acids (PUFA).
Figure 2 Sites of superoxide free radical formation in the mitochondrial electron transport chain. These superoxide anions may be converted to hydrogen peroxide by the action of MnSOD in both Intermembrane space and matrix or may reduce cytochrome c to form oxygen.
The sophisticated antioxidant system of mitochondrial matrix restricts the steady-state concentration of O2 ‒• within 10–10 M (Cadenas and Davies, 2000). The free radicals escaping from this detoxification process can cause oxidative damage to various biological components. But because mitochondrial DNA is in close proximity to the mitochondrial respiratory chain, it is very much susceptible to oxidative damage by various free radicals. The high rate of oxidative stress operating within the mitochondrial matrix causes a broad spectrum of mitochondrial DNA damage including modifications to bases such as 8-oxo-2′-deoxyguanosine (8-oxodG), 8,5′-cyclo-2′deoxynucleosides, thymine glycol, 5,6-dihydroxycytosine, 2,6-diamino-4-hydroxy-5-formamidopyrimidine. Along with this sugar break down products (such as 2-deoxypentose-4-ulose, 2-deoxypentonic acid lactones and erythrose), base free sites or abasic sites, strand breaks and chemical adducts of bases such as aldehyde modifications are also the consequences of the oxidative stress.
Among all these above modifications, oxidatively modified guanine or 8-oxodG is the most extensively studied and it is one of the biomarkers of oxidative stress. 8-oxodG is more important site in DNA than normal deoxyguanosine (dG) to induce a variety of mutations, if not repaired under normal physiological condition and the production increases with aging. It is also found that 8-oxodG accumulates more rapidly in mitochondrial genome than in the nuclear genome.
The metabolic capacity and membrane composition of mitochondria make it a suitable place for generation of reactive aldehydes which are potent to form various DNA adducts. Among various aldehydes, formaldehyde produces N2-hydroxymethyl-dG, the α, β-unsaturated fatty acids resulting from the oxidation of PUFA and DNA inset a five membered or six-membered ring to the DNA base. LPO also induces several ethno adducts (N6-ethanodeoxyadenosine and N4-ethenodeoxycitidine) and MDA creates six membered, exocyclic adducts (M1 dG, M1dC and M1dA). M1dG is present in higher amount in the mitochondrial genome than in nuclear genome (Jeong et al., 2005). It is because the inner mitochondrial membrane is enriched in PUFA, α, β-unsaturated aldehydes are the major threat to the mitochondrial genome stability and its expression during the oxidative stress.
Along with the endogenously produced DNA damaging agents, exogenous agents such as various chemicals, metabolites in the dietary components, UV radiation, ozone, pesticides, pharmaceuticals etc. can also cause damage to the mitochondrial DNA.
Also exogenous agents such as the acrolein components of the cigarette, fungal toxin aflatoxin B1, platinum based chemotherapeutic agent like cisplatin, ultraviolet radiation, ozone induce adducts on the DNA and interfere with the mitochondrial gene expression.
Aldehydes are produced in increasing amount during alcohol metabolism and they can also take entry in the body from external environment (Swenberg et al., 1999, 2011). Acetaldehyde is produced in large amount after the ingestion of alcohol which is converted to acetate in mitochondria by the action of aldehydes dehydrogenase 2 or ALDH2, failure of which can form several adducts including N2-ethylidene-dG, 1, N2-propano-2′-dG and their derivatives (Wang et al., 2000; Inagaki et al., 2004; Matsuda et al., 2006). Industrial pollutants contain formaldehyde, which together with several alkylating reagents can combine to form N2 -hydroxymethyl-d G (Lu et al., 2011).
Several anticancer drugs are able to bind to DNA causing damage leading to inhibition of tumour growth and also promote apoptosis. These anticancer drugs may have some effects on the mitochondrial genome. Various anticancer drugs such as bleomycin and neocarzinostatin can cause oxidative DNA damage through ROS and reactive aldehydes generation (Dedon et al., 1992, 1998).Another important group of anticancer drugs is platinum (Pt) based chemotherapeutic agents such as cisplatin, oxaliplatin, carboplatin etc which binds to DNA forming single base adducts and intra and inter-strand cross links (Wang et al., 2005). Cisplatin causes intra-strand cross links known as l,2-d(GpG) cross links after drug exposure and this even at low concentrations can cause inhibition to transcription and translation as they prevent strand separation (Ayala-Torres et al., 2000; Todd et al., 2009).
Mitochondria, a semiautonomous organelle probably evolved by endosymbiotically insertion of an unknown whole primitive organism, having their own genome. The mitochondrial genome is the second important element in eukaryotic cells after nuclear genome where most genetic information is being stored. Unlike nuclear DNA, mitochondrial DNA inheritance is almost exclusively maternal and the mitochondrial DNA is redundant with many copies are present within cells.
On the other hand, mitochondrial DNA is continuously replicated even in the terminally differentiated cells such as neurons. Hence somatic mitochondrial DNA damage has more adverse effect in comparison to somatic nuclear DNA damage. And it is not surprising that mutation in mitochondrial genome often results in alterations in mitochondrial functions and ultimately leads to several mitochondrial diseases (Arpa et al., 2003). Mitochondrial diseases can be severe and/or fatal. Along with the mitochondrial diseases, mitochondrial DNA damage is linked to some common diseases including age-related disorders, cancer and diabetes, neurodegenerative disorders and several other symptoms (Coskun et al., 2004; Howell et al., 2005; Taylor and Turnbull, 2005; Wallace, 2005; Chatterjee et al., 2006).
Thus maintaining the integrity of the mitochondrial genome is crucial for the survival of an organism. Mitochondria had long been considered to lack any kind of sophisticated DNA damage repair system like that of nuclear genome because of the discovery that mitochondria is unable to repair UV-induced pyrimidine dimers (Clayton et al., 1974, 1975) and some kinds of alkylation damage (Miyaki et al., 1977). Additionally another idea that mitochondrial fusion and fission events and mitochondrial DNA degradation may allow the organelle to minimize the toxic effect of DNA damage left the mitochondrial DNA damage repair pathways unresolved (Shokolenko et al., 2009). Over the PAST two to three decades the notion of mitochondrial DNA damage repair systems has evolved, and now it has been documented that almost every DNA damage repair system is present in mitochondria like that of in the nucleus (Gredilla, 2011). Important to be noted that for the mitochondria DNA repair this organelle has to depend on the repair enzymes synthesized by the nuclear genome.
Presence of novel DNA damage repair pathway is confirmed by analysing the different components of DNA damage repair pathways in the mitochondrial sub cellular fractionation. Other techniques employed were sub cellular localization of fluorescent fusion protein, immune gold labelling combined with electron microscopy and silico techniques. Overall the understanding of DNA damage repair systems in mitochondria emerged through successive stages. From the preliminary concept of no repair system to recent discoveries of different robust DNA damage repair pathways such as Base excision repair (BER), Nucleotide excision repair (NER), mismatch repair (MMR) and double strand break repair systems. Along with all these repair systems, the elimination of pre-mutagenic damaged dNTPs and selective degradation of severely damaged DNA complement the repair system to maintain the mitochondrial genome integrity.
Direct reversal repair represents the simplest pathway and also an alternative to photoreactivation of cyclobutane pyrimidine dimers or CPD by photolyase in E. coli without cleaving the phosphodiester chain (Sancar, 2008). Although mammalian mitochondria contain homologue of photolyase, transcriptional repressors, cryptochrome1 and 2 (CRY1 and CRY2) (Kobayashi et al., 1998), they are mainly involved in regulating the circadian rhythm. In contrast to yeast photolyase activity, which revert the UV induced damage to DNA (Pasupathy et al., 1992; Yasui et al., 1992), no direct evidence for reversal of UV induced damage could be detected in mitochondria. Along with the yeast, photolyase activity is evidenced in plant mitochondria (Takahashi et al., 2011) and Xenopus mitochondria (Ryoji et al., 1996). It is generally seen that higher eukaryote mitochondria do not possess the photolyase activity.
The O6-methylguanine-DNA methyltransferase (MGMT) is an important enzyme found in mammalian nucleus which is involved in the reversal of alkylated damage to DNA such as O6-alkylguanines. Because of the presence of a MGMT variant in mitochondria, it is able to repair O6-methyl-2ʹ-deoxyguanosine and O6-ethyl-2ʹ-deoxyguanosine (Myers et al., 1988; Satoh et al., 1988). However, the mitochondrial localization of mammalian MGMT is not confirmed yet by Western blotting (Cai et al., 2005). Thus it is likely that this kind of damage in mitochondria is repaired by protein synthesized by nuclear genome.
As mitochondria are the major sources of O2 ‒•, the mitochondrial genome remains under continuous oxidative threat. ROS induced oxidative DNA damage, such as 8-doxo-dG, is preferentially repaired by a robust Base Excision Repair (BER) system (Fig. 3). Existence of mitochondrial BER was first discovered after the identification of mitochondrial uracil-DNA glycosylase (Anderson and Friedberg 1980). The mitochondrial BER pathway that utilizes both mitochondrial and nuclear proteins is even more efficient than nuclear BER pathway because mitochondria remain as the major source of oxidative threat with 8-oxodG is the common DNA lesion (Thorslund et al., 2002; Murphy, 2009).This repair pathway has now been reported in yeast, animal cells and in plant mitochondria also(Mecocci et al., 1993; Higuchi et al., 1995).
Figure 3 Schematic representation of short and long-patch base excision repair (BER) pathways in mammalian system. All the important steps and factors are highlighted. In addition to FEN1, DNA2 is required for efficient flap processing during long-patch BER in mitochondria. Three sub-pathways of short-patch repair are converged at the gap filling step. Short-patch BER (left) and long-patch BER (right) pathways meet at the nick sealing step.
Mitochondrial BER pathway follows similar three-step mechanism like that of seen in nuclear BER pathway, i.e., recognition and elimination of damaged base, gap tailoring and DNA synthesis/ligation. Depending on the DNA polymerase mediated insertion of either single nucleotide or short sequence, BER pathway is of two types, i.e., short-patch BER and long-patch BER. Although animal mitochondria show both of them, plant mitochondria are restricted to short-patch BER only (Mecocci et al., 1993).The first step in Short patch BER is recognition of the damaged base and its elimination. The damaged base is recognized by two types of DNA glycosylase i.e., monofunctional and bifunctional.
After recognizing the DNA lesion, the monofunctional DNA glycosylase hydrolyzes the N-glycosidic bond at the oxidized bases forming an apurinic/apyrimidinic (AP) site. Two monofunctional glycosylase have been reported in mitochondria till date i.e., uracil-N-glycosylase1 (UNG1) (Anderson and Friedberg 1980) and MutY homolog glycosylase, MUTYH (Ohtsubo et al., 2000). After the action of glycosylase, AP endonuclease (APE1) cleaves the immediate 5ʹ side of the AP site, leaving a 3ʹ-hydroxyl and 5ʹ-deoxyribose-5-phosphate (5ʹ-dRP) residue. The 5ʹ-deoxyribose-5-phosphate residue is then removed by the 5ʹDRP lyase activity of DNA polymerase γ for the subsequent gap-filling polymerization reaction.
Besides two monofunctional glycosylases, four bifunctional glycosylases have also been reported present in mitochondria. These are 8-oxoguanine DNA glycosylase (OGG1) (Takao et al., 1998; de Souza-Pinto et al., 2009), nth (homologous to E. coli endonuclease III)-like 1 (NTHL1, or NTH1) (Karahalil et al., 2003) and Nei (E. coli endonuclease VIII)-like-1 and -2 (NEIL1 and NEIL2) (Hu et al., 2005; Mandal et al., 2012). Unlike monofunctional glycosylase, bifunctional glycosylases have additional AP-lyase activity which incise the apyrimidinic and apurinic (AP) sites by means of β- elimination (OGG1, NTLH1) or β,δ- elimination (NEIL1, NEIL2) (Dodson and Lloyd 2002,).β- elimination reaction results in formation of AP site with 3ʹ phospho-α,β-unsaturated aldehydes (3ʹPUA) and 5ʹP and β,δ- elimination reaction results in single nucleotide gap with terminal 5ʹP and 3ʹP.OGG1, NEIL1 and NEIL2 enzymes exhibit overlapping specificity for 8-oxoG but differs in their recognition of other DNA lesions caused by oxidative stress. Though N-methylpurine DNA glycosylase (MPG) is still not reported but mammalian mitochondria able to repair the lesions that are regular substrates of MPG (Chakravarti et al., 1991; Hang et al., 1996).
Removal of blocking groups at 5ʹ and 3ʹ end of the products of the first step of BER pathway is known as gap tailoring. The 5ʹ-dRP groups created by the action of UNG1, MUTYH1 and APE1 at the damaged base are excised by the dRPase activity of the Polymerase ɣ which also has polymerase and 5ʹ-3ʹ exonuclease activity. As a result, a single strand gap with 5ʹP and 3ʹ-OH is generated. On the other hand AP site with 3ʹPUA end generated by the action of OGG1 and NTLH1 is removed by the phosphodiesterase activity of APE1 and the 3ʹP generated at the site of DNA damage by the action of NEIL1 and NEIL2 and converted to 3ʹ-OH by the phosphatising activity of polynucleotide kinase 3ʹ-phosphatase (Mandal et al., 2012; Tahbaz et al., 2012).The product of the gap tailoring phase of short-patch BER pathway is always a single nucleotide ends with 5ʹ-P and 3ʹ-OH groups. This gap is filled by the activity of polymerase ɣ and the remaining nick is sealed by the action of DNA ligase IIIα. Importantly mitochondria do not contain XRCC1 which acts as a scaffold protein for the assembly of the nuclear BER components and also brings in ligase III to complete the BER pathway repair (Simsek et al., 2011).
The dRP-lyase activity of polymerase ɣ cannot efficiently remove certain 5ʹ blocking groups e.g. 5ʹ deoxyribonolactone (5ʹdL) which is about 72% of the total oxidative sugar damage in DNA (Roginskaya et al., 2005). Removal of this kind of lesions in nuclear DNA takes place through long-patch BER pathway and the mitochondrial long-patch BER pathway has been reported few years back (Akbari et al., 2008; Copeland and Longley 2008). In long-patch BER system, DNA polymerase γ displaces the 5ʹ blocking group extending from 3ʹ-OH forming a “flap” of about 6–9 nucleotides. This “flap” is subsequently removed by the joint action of two endonucleases, i.e., FEN1 and DNA2 (Copeland and Longley 2008; Zheng et al., 2008; Duxin et al., 2009; Kalifa et al., 2009) forming a nicked DNA duplex. The nick is then sealed by the activity of DNA Lig3 to recover the damaged part of the DNA double helix.
Nucleotide excision repair (NER) pathway involves the elimination of and re-synthesis of short DNA fragment on the damaged strand. However, there is no evidence of any typical NER pathway in mitochondria like that of nuclear NER pathway. Though several studies have confirmed that mitochondria do not have any NER mechanism for the removal of UV-induced pyrimidine dimers (Clayton et al., 1974; LeDoux et al. 1992; Pascucci et al., 1997), showed that in yeasts certain NER substrates are repaired by an alternative mechanism; for instance, in Schizosaccharomyces pombe mitochondrial cyclobutane pyrimidine dimers (CPDs) and (6–4) pyrimidine-pyrimidone photoproducts [(6–4) PP] are repaired by UV damaged DNA endonuclease-dependent excision repair (UVER) pathway.
Mitochondria also exhibit poor repair of cisplatin induced intra-strand cross links but efficiently repair the inter-strand cross links, which are repaired by Recombination repair system (RRS). Since, both types of cisplatin induced adducts can efficiently be removed by nuclear repair system, mitochondrial damage remains one of the major issues of cisplatin-induced toxicity. Because mitochondria is deficient of a proper NER system, 8,5′-cyclo-2′-deoxypurines (cPu), M1dG and BaPDE adducts, this causes problems in mitochondria to suffer in many tissues of the body (Wei et al., 1995; Hess et al., 1997; Johnson et al., 1997; Reardon et al., 1999; Brooks et al., 2000; Kuraoka et al., 2000; Lloyd and Hanawalt, 2000). The only evidence of NER-like mechanism in mitochondria is in Cockayne syndrome proteins, CSA and CSB, which are required for transcription-coupled NER (TC-NER) repair system (Aamann et al., 2010; Kamenisch et al., 2010; Lagerwerf et al., 2011).
The occurrence of mismatch repair system in mammalian mitochondria was established in-vitro by M-13 based assay in which the rat liver mitochondrial extracts showed small but significant mismatch repair (MMR) activity (Mason et al., 2003). Mismatch repair activity in mammalian mitochondria is efficient in cleaving the G:T and G: G mismatches, without showing any activity for discrimination of the parental strand from the sister strand, thus creating the situation more susceptible to several kind of mutations (Mason et al., 2003). Mammalian mitochondrial mismatch repair system is very different from that of nuclear mismatch repair and is independent of MSH2, one of the master regulators of the nuclear mismatch repair system (de Souza-Pinto et al., 2009). Confocal microscopy also confirmed the absence of other classical MMR members viz. MSH3, MSH6 and MLH1 in the mammalian mitochondria (de Souza-Pinto et al., 2009). In yeast and in E. coli mismatch repair component MutS homologue MSH1 is present which repairs the G: A mismatches generated by replication past 8-oxoG (Dzierzbicki et al., 2004). The Y-box binding protein YB-1 protein (NSEP1 and YBX1), previously known to play a major role in nuclear base excision repair pathway and also in repair of cross-linked DNA, is a key component of the mammalian MMR and is mainly involved in mitochondrial mismatch binding activity. RNA interference study also supported the role of YB-1 protein in mitochondrial MMR and YB-1 depletion showed increase in mitochondrial DNA mutagenesis. All these studies strongly suggest the presence of a mismatch repair system in certain mitochondria which is distinct from that of the nuclear mismatch repair, and this pathway is needed to be further characterized.
Not only DNA is subjected to damage by different chemical agents but the free deoxynucleotide triphosphate (dNTPs) pool can be damaged too and may be severely, mainly by ROS oxidation. These damaged (oxidized) dNTPs are the major sources of mismatches during the replication of mitochondrial DNA. On the other hand the fidelity of mitochondrial DNA polymerase ɣ (Pol ɣ) is reduced when the enzyme finds oxidized dGTP (8-oxodGTP), 0.06–0.6% of the total dGTP pool (Pursell et al., 2008). This information indicates that oxidation of dNTPs pool is a greater threat to the stability of mitochondrial DNA. Mitochondria have developed specialized sanitizing enzymes to eliminate the oxidized dNTPsand replace it by dUTPs or 8-oxodGTPS; these are incorrectly incorporated into the mitochondrial DNA during replication in that 8-oxodGTPs is placed opposite to the adenine in the template DNA forming 8-oxodG:dA base pairs which escapes the proof reading activity of Pol ɣ leading to a AT to CG transversions (Hanes et al., 2006).This can lead to mutated mitochondrion and can become dysfunctional
The major mechanism in mitochondria to counter against premutagenic8-oxodGTPs includes the removal of 8-oxodGTPsfrom the dNTPs pool by the activity of mitochondrial MTH1. MTH1 is a mammalian homologue of E. Coli MutT protein (Kang et al., 1995; Nakabeppu, 2001). MTH1 actually hydrolyzes the 8-oxo-2ʹ-deoxyguanosine triphosphate to dGMP, the corresponding monophosphate form thus minimizing the chances of incorporation of 8-oxodGTPsby Pol ɣ during mitochondrial DNA replication.MTH1 also protects cell from cytotoxicity of sodium nitroprusside by preventing the build-up of 8-oxoG (Ichikawa et al., 2008). Although sanitation of oxidized dNTPs is not bonafide DNA damage repair mechanism but elimination of damaged dNTPs reduces the chances of mismatches in the mitochondrial DNA which helps in mitochondrial genome maintenance.
A well-developed double-strand break repair system is also present in mitochondria as reported in both, yeast and Drosophila melanogaster (Contamine and Picard 2000; Morel et al., 2008). In-vitro studies demonstrated the presence of both homologous recombination (HR) and non-homologous end joining (NHEJ) pathways in mammalian mitochondria; also it has been reported that in mitochondria, recombination repair is carried out in both homology dependent and independent manner (Thyagarajan et al., 1996; Coffey et al., 1999; Lakshmipathy et al., 1999; Bacman et al., 2009; Fukui et al., 2009). Presence of both intra-molecular and inter-molecular recombination products in heteroplasmic mice after generating double strand breaks with the help of Pst1 and Sca1 confirmed the presence of both the HR and NHEJ pathways in mammalian mitochondria though the frequency of NHEJ products is rare . The key regulator of nuclear HR pathway in mammal, i.e., Rad51 is also present in mitochondria and thought to be involved in the mitochondrial double strand break repair (Sage et al., 2010). Homologous recombination is also reported in plant mitochondria and later the presence of Rec A homologue in the moss Physcomitrella patens is also reported (Manchekar et al., 2006; Odahara et al., 2007). In higher plants the Rec A homologues are mainly involved in the surveillance of recombination pathways.
Despite all these DNA repair systems, degradation of damaged part of DNA plays an important role in maintaining the genome integrity in mitochondria to avert mitochondrial related diseases. Unlike nuclear genome, mitochondrial genome is present in high copy number which keeps them out of the “repair or die” constraint. Though certain yeast mitochondria have a photolyase that may repair pyrimidine dimers (Yasuhira and Yasui 2000), but mammalian mitochondria cannot repair UV-induced pyrimidine dimers which spurs the concept of mitochondrial DNA degradation in response to DNA damage (Clayton et al., 1974). However, Mita et al. (1988) showed that mitochondrial genome of HeLa cells accumulate very few number of mutations when exposed to various carcinogens such as ethyl methane sulphonate (EMS) suggested that mitochondrial genome is not replicated when large amounts of damage or irreparable damage is accumulated (Mita et al., 1988). Some other studies also showed that many cell lines show resistance to the loss of mitochondrial DNA and can survive gradual loss of mitochondrial DNA after treatment with some intercalating agents such as ethidium bromide (King and Attardi 1989). If the mitochondrial DNA is induced to extensive or persistent double strand breaks by means of site-specific restriction endonuclease enzymes, mitochondrial DNA shows depletion (Alexeyev et al., 2008; Kukat et al., 2008) with a very low amount of DSB leading to recombination (Bacman et al., 2009). This is contrasting feature of the nuclear genome in which persistent DSB can leads to apoptosis. Given the high copy number of mitochondrial DNA, the hypothesis that degradation of the damaged mitochondrial DNA can be a mechanism adopted by mitochondria to cope up with extensive DNA damage. Lack of direct experimental evidence and also discovery of mitochondrial BER pathway (Pettepher et al., 1991) puts doubt on this hypothesis.
The oxidative induced DSB acts as a signal inducing mammalian mitochondrial DNA degradation, presumably generated by stalled DNA or RNA polymerases at the damaged DNA strand. As this kind of DNA breaks and AP sites lack any genetic information, complete elimination of such sites can prevent the mitochondrial mutagenesis and will also protect the integrity of the mitochondrial genome. In these cases, stalled DNA or RNA polymerase may act as signal for mitochondrial DNA degradation.
>A nuclear genome encoded evolutionarily conserved mitochondrial protein Endonuclease G (Endo G) is most abundant and active nuclease in mitochondria and is considered to be primarily involved in generating RNA primer for the mitochondrial DNA replication (Cote and Ruiz-Carrillo, 1993). This Endo G preferentially cleaves the oxidatively damaged mitochondrial DNA (Ikeda and Ozaki, 1997); it is thought to act as an agent for selectively degrading the mitochondrial genome. But in contrary the Endo G null mice or the cells with lower level of Endo G do not show any kinds of defect in mitochondrial genome, its copy number and mutation rate (Zhang et al., 2003; Irvine et al., 2005; David et al., 2006; Huang et al., 2006) making it difficult to understand how Endo G is involved in maintaining the integrity and stability of the mitochondrial genome. Disruption of yeast homologue of EndoG, NUC1 also showed marginal effects on the integrity of mitochondrial DNA (Zassenhaus et al., 1988), which puts a question mark on the role of Endo G and more studies are required to understand the function of Endo G in mitochondrial DNA degradation.
Besides certain distinct functional differences between mitochondrial and nuclear genome, they both are composed of DNA double helix with coding sequences. One healthy living eukaryotic cell contains several mitochondria with each mitochondrion comprising of multiple copies of mitochondrial genome. Recent studies have established that mitochondrial genome has a considerably higher rate of mutation than the nuclear genome, thus generating heterogeneous population mitochondrial genome within the same cell and sometimes even within the same mitochondrion, thus conferring heteroplasmic condition. Although mitochondria are transmitted to the daughter cells during cell division, the segregation occurs in a random manner and irregular fashion as compared to nuclear genome segregation, thus generating additional instability at the cellular level. One relevant question that has been addressed in recent studies includes the high mutation rate in mitochondrial genome. The nuclear gene encoded protein DNA polymerase γ(POLG) is responsible for replicating the mitochondrial genome. The catalytic subunit of POLG contains the polymerase activity, while an exonuclease domain recognizes and removes the DNA base-pair mismatches which originate during DNA replication. More detailed studies have indicated a nucleotide imbalance in mitochondria decreases the fidelity of POLG, thus causing higher mutational rates in mitochondrial genome.
Mitochondrial mutations are known to be responsible for a number of common clinical diseases and symptoms. Some examples include diabetes, predisposition to Alzheimer’s and Parkinson’s disease. Another important mitochondria-associated disorder includes the Leber hereditary optic neuropathy (LHON), which causes loss of vision in both eyes and has been shown to be linked with homoplasmic mitochondrial DNA mutation. However, because of maternal influence in inheritance, males with a mitochondrial disease are generally not considered to be the potential carrier for transmitting the disorder to the next generation. Apart from this, since several mitochondria are present in a cell, carrying potential mutations in the mitochondrial DNA, understanding mitochondrial genome stability in a global scale is important. Since mitochondria function as the power house of the cell, tissues which have high energy demand, such as brain, renal, skeletal and the cardiac muscle tissues are often associated with distinct phenotypes due to mitochondrial mutations. Extensive research has shown that mitochondria affect the stability of the nuclear genome and vice versa. As mentioned before, although mitochondrial genome replicates independently, all classical DNA damage repair pathways participate in repairing mitochondrial genome damage. However, in contrast to nuclear genome stability mechanisms, information is limited about the underlying mechanisms which regulate mitochondrial genome stability. In addition, much less is known about the two-way communication and mutual influence of the nuclear and mitochondrial genomes. Therefore, further future research will provide more insights into the genome stability mechanisms in mitochondria for explanation of various genome instability phenotype associated with the failure of various mitochondrial functions.
The authors gratefully acknowledge The Council of Scientific and Industrial Research, Govt. of India, (Ref. No. 38(1417)/16/EMR-II, dated:17/05/2016 to SR), UGC, Govt. of India Start-Up research grant No.F.30-141/2015(BSR) and SERB, DST, Govt of India (Ref. No. ECR/2016/000539 to SR) for providing financial supports.