Mitochondrial free radical theory of aging – at what stage are we?

The mitochondrial free radical theory of ageing – Where do we stand? Frontiers in Bioscience 13, 6554-6579, May 1, 2008

Jan Gruber, Sebastian Schaffer, Barry Halliwell National University of Singapore, Department of Biochemistry, Centre for Life Sciences, 28 Medical Drive, 117456 Singapore

Correspondence address: Barry Halliwell, Department of Biochemistry, National University of Singapore, University Hall, Lee Kong Chian Wing, UHL #05-02G , 21 Lower Kent Ridge Road, Singapore 119077, Tel: 656563247, Fax: 6567752207, E-mail: bchbh@nus.edu.sgKeywords: mitochondria, aging, reactive oxygen species, DNA, antioxidants, mitochondrial nutrients, mtDNA, deletions, point mutations, review

content 1. Summary

2. Introduction
2.1. The role of mitochondrial DNA
2.1.1. Mitochondrial DNA, dysfunction and etiology of age-related degenerative processes: assumptions and questions
2.1.2. Is the mitochondrial matrix an unfavorable environment for DNA?
2.1.3. Do mitochondria have the ability to repair DNA damage and is mtDNA deproteinized?
2.1.4. Do oxidative damage to mtDNA occur frequently in vivo?
2.1.5. Is there evidence of a "vicious circle"?
2.1.6. Do mitochondria matter?
2.1.7. Mosaic is a key aspect of pathology associated with age–related accumulation of mtDNA mutations
2.2. How can mtDNA mutations serve as a causal factor of age-associated tissue dysfunction?
3. Arguments in favor of directed determinants of the basal rate of aging, rather than individual age-related diseases
4. Strategies
4.1. Mitochondrial nutrients
4.1.1. L-carnitine/Acetyl-L-carnitine
4.1.1.1. Basic information
4.1.1.2. Possible mechanisms of action
4.1.1.2.1. Influence on mitochondrial bioenergetics
4.1.1.2.2. Effect on antioxidant protection
4.1.1.3. Human impact studies on aging and age-related diseases
4.1.2. Coenzyme Q10
4.1.2.1. Basic information
4.1.2.2. Possible mechanisms of action
4.1.2.2.1. Influence on mitochondrial bioenergetics
4.1.2.2.2. Effect on antioxidant protection
4.1.2.3. Human impact studies on aging and age-related diseases
4.1.3. Alpha-lipoic acid
4.1.3.1. Basic information
4.1.3.2. Possible mechanisms of action
4.1.3.2.1. Influence on mitochondrial bioenergetics
4.1.3.2.2. Effect on antioxidant protection
4.1.3.3. Human impact studies on aging and age-related diseases
4.2. Comparison of combined and separate intake of mitochondrial nutrients
4.3. Antioxidants – some general information
5. Conclusions and prospects
6. Literature

1. SummaryUnderstanding the molecular mechanisms underlying the aging process can provide the best strategy for solving the problems associated with the aging of the world's population.

One of the theories suggesting such molecular mechanisms was formulated 50 years ago. Harman et al. It has been suggested that aging may be mediated by macromolecular damage resulting from reactions involving reactive oxygen species (ROS). To date, one of the most popular theories of aging is a version of the free radical theory, according to which mitochondria are both a source and a target of ROS. We present a critical overview of the state of the key principles and concepts on which this theory is based. We believe that the evidence currently available indicates that the initial assumptions are controversial, while in order to study a number of critical points, it is necessary to improve existing methods. Even under these conditions, it is becoming increasingly obvious that the integrity of mitochondria and mitochondrial DNA (mtDNA) can indeed be factors determining the aging of the body. We discuss possible prospects for successful interventions in this process, as well as evidence for and against the effectiveness of existing therapeutic approaches.

2. Introduction

In 1954, Gerschman et al. It was suggested that oxygen free radicals are toxic agents responsible for both oxygen poisoning and damage caused by gamma irradiation (1). Two years later, Denham Harman first suggested that endogenous free radicals can cause macromolecular damage in vivo, as well as that mediated by free radical damage is a causal factor in the accumulation of mutations, the development of cancer and aging (2).

For more than 15 years of its existence, the free-radical theory of aging has undergone modifications and additions. The most important of them is the statement according to which the most significant source of reactive oxygen species in vivo is a superoxide radical synthesized as a by-product of the normal process of oxidative phosphorylation, and that mitochondria themselves can be the main target of accumulation of ROS-induced damage [3-5]. This statement gave rise to the mitochondrial free radical theory of aging, the variations of which we will discuss in more detail below. Various versions of the free radical theory of aging have already been actively discussed by other authors (6-10), therefore, the purpose of this article is to critically review a number of evidence considered to confirm this theory, as well as to assess the prospects for reducing mitochondrial damage caused by free radicals.

2.1. The role of mitochondrial DNAMitochondrial DNA, due to its location in the mitochondrial matrix, has been given special attention as the intended target of ROS-mediated damage, since at least part of the ROS produced in the mitochondria is released into the matrix (3,4,11-13).

The original version of the free radical theory of aging is based on four principles (6-13). Firstly, due to its location in the mitochondrial matrix, mtDNA is strongly affected by free radicals. Secondly, unlike nuclear DNA (yDNA), mtDNA is deproteinized, that is, it is not covered with protective structural proteins such as histones. According to the third assumption, mitochondria lack the ability to repair DNA damage caused by oxidation. Based on these three assumptions, the constant baseline load of oxidative damage on mtDNA is significantly higher than on nucleic acid.

The effect of ROS on mtDNA contributes to the formation of products of oxidative damage to DNA bases, such as 8-hydroxy-2’-dioxyguanosine (8OHdG), as well as chain breaks (14-16). Some of these DNA damages are mutagenic, for example, 8-hydroxyguanine can enter into complementary interaction with adenine instead of cytosine, which often leads to transversions (replacement of purine bases in DNA with pyrimidine and vice versa) GC to TA(17-19). On the other hand, oxidative damage to cytosine more often leads to transitions (simple substitutions of one purine or pyrimidine DNA base to the corresponding other base) GC to AT(20). In general, GC transitions to AT and GC transversions to TA are the most common mutations caused by oxidative stress (analyzed in 21). Some components of the mitochondrial electron transfer chain are encoded in mtDNA.

Thus, the fourth assumption underlying the mitochondrial free radical theory of aging is that an increased level of mtDNA mutations further increases the number of ROS, thereby contributing to further accumulation of mutations by disrupting the integrity of the electron transfer chain. This mechanism, in which a small amount of initial damage can trigger an exponential increase in the number of new damage, creates a positive feedback mechanism called a "vicious circle" (4, 22).

2.1.1. Mitochondrial DNA, dysfunction and etiology of age-related degenerative processes: assumptions and questionsAs mentioned above, the essence of the mitochondrial free radical theory of aging is the exponential acceleration of mitochondrial degeneration.

In this process, ROS are both the initiator and the key mediator of the positive feedback mechanism driving the whole process. The central role of free radicals in this model directly suggests the existence of intervention methods based on the use of antioxidants that neutralize mitochondrial ROS, or on the modulation of mitochondrial synthesis of ROS. However, before discussing the merits of some of these methods, it is necessary to study the currently existing experimental data (or lack thereof) confirming the assumptions underlying the mitochondrial free radical theory of aging.

2.1.2. Is the mitochondrial matrix an unfavorable environment for DNA?

Experiments on isolated mitochondria have shown that the most important source of ROS released by mitochondria is the so-called complex III. However, there is no evidence that these ROS are released into the intramembrane space and, accordingly, away from the mitochondrial matrix [23]. However, other components of the mitochondrial electron transfer chain, such as the flavin mononucleotide group (FMN) of complex I, do release free radicals into the mitochondrial matrix during normal respiration (24-26). Most of the data indicating that mitochondria are the main source of free radicals were obtained by studying isolated mitochondria at various stages of the respiration process. The reliability of these data is questionable, especially due to the fact that during experiments, the activity of ROS formation was often determined under non-physiological conditions [27]. At the same time, two recent studies have shown that under physiological conditions, superoxide and hydrogen peroxide are produced in isolated mitochondria of skeletal muscle and brain cells, respectively (28,29). In both studies, the release of ROS occurred mainly from complex I. Similarly, Kozlov et al. (30) found increased ROS production in the mitochondria of the heart muscle isolated from cells of aging animals, compared with the mitochondria of cells of young animals of the control group. However, the authors of this work concluded that the most likely source of ROS is complex III, which, according to the above, is not an immediate threat to biomolecules of the mitochondrial matrix, such as mtDNA.

While the superoxide radical itself is not active enough to damage DNA (14), it can contribute to its damage by interacting with free iron and hydrogen peroxide (H2O2) with the formation of a more reactive hydroxyl radical, or by stimulating the release of iron from iron-sulfur groups, for example, which are part of aconitase (15,31,32). Therefore, mtDNA located in the mitochondrial matrix is a potential target of the oxidative action of endogenous ROS generated by complex I. The activity of complex I, in turn, is most predisposed to the influence of mtDNA mutations, since 7 of the 13 mtDNA encoded polypeptides are part of this complex [33]. Despite extensive knowledge concerning the production of ROS associated with mitochondria in vitro, as well as evidence of their importance in vivo (this is discussed in Chapter 2.1.6.), it is still unclear what the real amount of superoxide and hydroxyl radicals is formed in mitochondria in vivo. Developments in the field of analytical methods, such as electron paramagnetic resonance spectroscopy, can help in finding an answer to this question [34].

Despite the existence of evidence of an increase in iron levels, for example, in the aging brain or in diseases associated with aging (35), our knowledge of iron homeostasis in mitochondria and its possible accumulation is still in its infancy. An interesting new hypothesis concerning the importance of iron homeostasis disorders in cellular stress is based on the identification of the role of ferrihydrite as the main form of deposited iron in the mitochondria of cells of patients with hereditary Friedreich's ataxia. Despite the fact that patients with this diagnosis usually do not suffer from a general excess of iron, and given the relatively low reactivity of ferrihydrite, the authors of one of the works declare the presence of potentially toxic divalent iron in mitochondria, capable of causing moderate but chronic oxidative stress [36].

2.1.3. Do mitochondria have the ability to repair DNA damage and is mtDNA deproteinized?The hypothesis that mitochondria are unable to repair DNA damage was born on the basis of the observation that the damage caused by ultraviolet radiation to mtDNA is not effectively repaired.

This indicates that there is no recovery of excision (loss) of nucleotides (VEN) in mtDNA (37,38). However, unlike VEN, the presence of mechanisms for restoring excision of nucleotide bases (VEN) in mitochondria has been proven (37,39,40). In particular, in the mitochondria of rat liver cells, the enzyme OGG1, responsible for ENO 8OHdG(41), was found. Moreover, the mitochondrial DNA damage repair system repairs 8OHdG and mtDNA chain breaks at least as efficiently as the equivalent core system (39,42). Thus, it is obvious that, contrary to the oft-repeated initial assumption, mitochondria are capable of rapid and effective recovery of oxidative DNA damage, especially 8OHdG.

There is also evidence that mtDNA is not as deproteinized as previously thought. For example, it has long been known that mtDNA is not able to move freely in the mitochondrial matrix, but is bound to proteins and anchored on the inner membrane [43]. Moreover, there is evidence that mtDNA can be associated with several molecules of mitochondrial transcription factor A, and it has been suggested that this factor plays the role of a histone-like protein in mitochondria (described in (44)). In accordance with this idea, the results of footprinting (a method for determining the sites of nucleic acids forming complexes with proteins) indicate that mtDNA is not completely deproteinized, but is associated with protective proteins [45].

2.1.4. Do oxidative damage to mtDNA occur frequently in vivo?The mitochondrial free radical theory of aging experienced a significant upsurge in 1988, when Richter et al. suggested that the content of 8OhdG in mtDNA is 16 times higher than in nuclear DNA (13).

However, since then, experts have come to the conclusion that an accurate quantitative assessment of oxidative DNA damage is extremely difficult. Possible problems include unintentional oxidation of DNA during isolation and sample preparation, as well as systematic discrepancies between the methods used [46-48]. The scale of the problem, especially in relation to earlier methods, is illustrated in a study conducted by the European Standards Committee on Oxidative DNA Damage – European Standards Committee on Oxidative DNA damage, ESCODD (49,50). As part of this study, standardized tissue samples were distributed among 28 participating laboratories specializing in the quantitative analysis of oxidative DNA damage. Despite all the efforts spent on standardizing methods and minimizing artifacts, the obtained levels of 8OhdG were in a range whose extreme values differed by more than two orders of magnitude (49.50). In 1999, Beckman and Ames worked through the literature data on the basic levels of 8OhdG in DNA and mtDNA and found that the boundaries of the obtained range differ by more than by 4 orders of magnitude (48), which indicates a significant influence of artifacts on some published data.

These problems mostly relate to mtDNA, since its quantity is usually very small and ex vivo it is more susceptible to artifacts than DNA (48,51,52). Given these difficulties, it is not surprising that the actual data on the scale of oxidative modifications of mtDNA in vivo, both in absolute and relative values, are contradictory.

Thus, some authors reveal higher levels of oxidative damage to mtDNA compared to nucleic acid (13, 53-55), while others do not observe similar differences (46,48,56). Given the known artifacts, as well as the fact that the differences between the results obtained by different methods and different laboratories are many times greater than the expected differences between mtDNA and yDNA, it can be concluded that the information available today is insufficient to prove that the level of oxidative damage to mtDNA is higher than the same indicator for yDNA. A similar situation is observed with the results of attempts to prove that the level of oxidative damage to mtDNA increases with age. At the same time, some authors note an age-associated increase in the level of damage (55.57-60), while others do not detect such an increase (61.62).

On the other hand, indirect evidence of the involvement of ROS-mediated DNA damage in age-dependent mutagenesis has recently been obtained. Using a highly sensitive mutation detection kit (63) Vermulst et al. It has been demonstrated that the accumulation of point mutations in the mtDNA of brain and heart cells of wild-type mice changes exponentially with age [64]. This corresponds to earlier data concerning mtDNA deletions, according to which the load of mtDNA mutations increases with age according to an exponential relationship (65). Exponential kinetics can be interpreted as a positive feedback mechanism, where the frequency of mutations at any given time is a kind of function of the mutation load accumulated by that time. Moreover, mutations identified by Vermulst et al. in wild–type mice, 80% were GC transits to AT, while most of the remaining 20% of mutations were GC transversions to TA(64). This spectrum of mutations corresponds to the consequences observed in oxidative DNA damage, and thus indicates the causal role of ROS-mediated mtDNA damage [21,66]. However, this does not prove that the mechanism involved is mediated by free radicals. In contrast to these observations made in experiments on mice, it has recently been observed that the spectrum of mutations observed in mutational "hot spots" of human mtDNA is more consistent with polymerase errors [67].

In general, upon closer examination, it is obvious that most of the initial factors supporting the assumption that mtDNA is more susceptible to oxidative damage require at least a detailed assessment.

2.1.5. Is there evidence of a "vicious circle"?The validity of the "vicious circle" hypothesis is confirmed by experiments on neuron cultures, where disruption of the electron transfer chain by RNA interference (RNAi) can lead to a significant increase in oxidative damage to DNA and ROS production [68]. Similarly, temporary chemical inhibition of complex I by rotenone in vivo leads to irreversible damage to rat brain mitochondria associated with strengthening of AFC production (69).

Even though the absolute levels of mtDNA point mutations identified by Vermulst et al. (64) are very low (<0.2 per mtDNA molecule), the observed exponential kinetics corresponds to the vicious circle hypothesis.

To directly study the functional consequences of the accumulation of point mutations of mtDNA in vivo, including the effect on ROS production and oxidative stress, scientists have created a number of knockout mouse models with error-prone mtDNA polymerase (70-72). In the first of such transgenic mouse models, selectively expressing mtDNA polymerase incapable of correcting errors in heart cells, accumulation of both point mutations and deletions in these cells occurred faster than in wild-type mice [72]. As a consequence of increased cardiomyocyte death, cardiomyopathy developed in transgenic mice, however, there was no increase in the level of protein carbonyls in animal heart muscle cells, which is a measure of the assessment of oxidative damage to proteins (73.74), as well as a decrease in aconitase activity and glutathione levels (75.76). When performing a strictly qualitative analysis based on southern blotting performed after mtDNA treatment with formamidopyrimidine-DNA glycosylase, Mott et al. No age-dependent growth of the mtDNA oxidative damage load was detected in the heart cells of transgenic mice (75).

Two different models of mutator mice (characterized by an increased frequency of somatic point mutations of mtDNA) carrying homozygous mutations of mtDNA and expressing mtDNA polymerase incapable of correcting errors throughout the body, demonstrated a significant increase in the load of mtDNA mutations and a number of pathologies usually associated with aging, as well as a significantly reduced lifespan (70.71). Kujoth et al. (70) compared the content of 8OHDG in DNA and RNA of liver cells and F2-isoprostanes – a widely used biomarker of oxidative damage to lipids (77.78) – in tissues (skeletal muscles, liver) of mutator mice and wild-type mice and found no evidence of a total increase in the level of oxidative damage. Within the framework of the same work, no significant differences were found in the content of protein carbonyls in the mitochondria of the heart and liver, as well as the production of H2 O2 by isolated liver mitochondria in mutator mice and wild-type mice (70).

In addition to a slight increase in the level of protein carbonyls and not very significant evidence of activation of glutathione peroxidase expression, Trifunovic et al. also, no signs of increased oxidative stress were detected in the tissues of the second mutator mouse model (79). However, an interesting fact is that they revealed comparable levels of free radical production in mouse embryonic fibroblasts isolated from embryos of transgenic mice and wild-type mice, despite the fact that the cells of the first cell line are practically incapable of respiration [79]. The authors noted that such results indicate that the amount of ROS, produced by mutator mouse cells per absorbed oxygen molecule, in reality would be significantly higher compared to this indicator for control cells [44]. However, given the known difficulties and artifacts associated with the evaluation of ROS production in cell cultures (80 and will be discussed below), it is also possible that this result mainly reflects the methodological difficulties characteristic of determining the level of ROS by in vitro methods.

In general, none of the mouse models of an increased number of mtDNA mutations provides evidence of increased oxidative stress. Instead, such animals show signs of increased cell death (70,79). There is practically no evidence that an increased load of mtDNA point mutations actually leads to an overall increase in oxidative stress, even though a significantly increased load of mutations leads to severe pathology associated with activation of cell death mechanisms.

On the other hand, mutator mice demonstrate that a high level of mtDNA mutations is associated with aging, or at least with phenotypes corresponding to some aspects of aging. Does this have to do with normal aging? The question naturally arises: how numerous are mtDNA mutations in vivo with normal aging? The age-associated accumulation of mtDNA mutations has been actively studied in both animal and human tissues. The number of specific point mutations of mtDNA appearing in non-coding mtDNA control regions increases significantly with aging. In some cases, more than 50% of the entire DNA molecule may consist of such mutations (81,82). Such mutations, which have high tissue specificity (for example, specific point mutations usually occur in muscle cells, but are not registered in fibroblasts and vice versa), do not reduce the performance of the mitochondrial oxidative phosphorylation system, and some of them may even provide positive functional adaptation [83].

On the contrary, the data obtained in the study of wild-type mice in the framework of a study using mutator mice (described above) indicate that the number of point mutations of mtDNA affecting the coding regions of molecules and potentially leading to disruption of the normal functioning of the electron transport chain remains at an exceptionally low level even in very old organisms (64). For comparison, mice with a heterozygous mutation used in one of the studies suffered a 500-fold increase in the load of mtDNA point mutations without showing any signs of pathology or premature physiological aging (Fig. 1) (64). The latter result is interpreted as proof that point mutations themselves do not accumulate in the process of normal aging in quantities so significant as to limit life expectancy [64].

Fig. 1. (A) Mutational load in the mtDNA of the brain tissue of wild-type mice (DT) compared with heterozygous (Polg^/mut) and homozygous (Polg ^mut/mut) mutant mice. While the age-associated increase in the number of mtDNA point mutations in DT mice is obvious, heterozygous mutant mice have a significant increase in the number of point mutations, reaching 10 times higher values compared to DT mice, without visible signs of premature aging and mitochondrial pathologies. (C) Reduced mutational load in the mtDNA of heart cells of MCAT mice characterized by catalase overexpression in mitochondria compared to DT mice of the corresponding age. All data is borrowed from (64).

However, there are a number of open questions that need to be considered before accepting this conclusion with full confidence. Firstly, mutation occurs in mutator mice in the early stages of life due to errors in the polymerase and there is practically no data on further accumulation of mutations associated with age. In wild-type mice, on the contrary, mtDNA mutations rarely occur in the early stages of life, but as the body ages, they accumulate according to an exponential dependence [64]. This is based on significant differences in the mechanisms and dynamics of accumulation of mtDNA mutations in wild-type mice and mutator mice. Possible causal mechanisms of this will be discussed further.

Secondly, it has been observed that some phenotypes observed in mutator mice indicate a greater involvement of tissues with high mitotic activity (spleen, testes, epidermis) than postmitotic tissues, which is characteristic of present premature aging [84]. At the present stage, it remains unclear whether mutator mice are an adequate model for studying the functional effects of normal age-associated loss of mtDNA integrity (84-86).

The use of polymerase chain reaction (PCR) techniques also revealed mtDNA deletions in various tissues of different organisms (87-92). It is believed that a large number of inherited mtDNA deletions underlie a number of rare diseases, including myopathies and encephalomyopathies (93). Moreover, the level of mtDNA deletions in heart cells increases significantly with age (92), the same is observed in the brain (87,94,95). It is suggested that there is a correlation between the number of deletions and the number of 8OHdG for the mtDNA of heart cells. This indicates that oxidative damage to mtDNA can initiate the formation of de novo mtDNA deletions (96). A possible mechanism for this is oxidative damage to DNA polymerase, since ROS-mediated DNA damage itself usually does not lead to the formation of deletions [97]. However, as in the case of point mutations of mtDNA, the occurrence of deletions in the tissue mass is very low and usually, even in very old animals, remains at a level significantly lower than 1% of the total amount of mtDNA (88.95.98). Accordingly, taking into account the relatively low levels of point mutations and deletions in tissues, most studies on the general ability to oxidative phosphorylation and enzymatic activity in primate and rodent cells and tissues revealed a certain degree of age-associated functional decline, but this decline is tissue-specific, in most cases relatively small and has a high degree of individual variability (99-105).

In general, many initial assumptions leading to the assumption that mtDNA is subject to severe oxidative damage require clarification and to date there is no convincing evidence that oxidative damage to mtDNA is really very pronounced in vivo. It is unclear what functional significance in vivo mtDNA damage has at the scale of one 8OHdG damage per 105 nucleotide bases. According to existing data, oxidative damage to mtDNA may be a causal factor in the appearance of mtDNA point mutations and may also be involved in the formation of de novo mtDNA deletions. According to the conclusion that in the equilibrium state, oxidative damage to mtDNA remains at a low level, the number of deletions and point mutations in the coding regions of mtDNA in most solid tissues is small, even in very old animals. While nonphysiologically high levels of mtDNA mutations are obviously pathogenic, their rare occurrence in normal animals raises the question of their functional significance in the context of normal aging.

2.1.6. Do mitochondria matter?Does all this mean that mitochondrial ROS production and mtDNA integrity are irrelevant for aging?

It is not necessary at all, since there is strong evidence that mitochondrial ROS are an important determining parameter of the rate of aging of the body. A large amount of literature data demonstrates that the levels of mitochondrial superoxide production correlate with the aging rate of various species (106-115). According to recently obtained data, this conclusion remains valid even when taking into account such factors interfering with the analysis as body size and phylogeny (116,117). These observations correspond to the data according to which the load of oxidative damage (estimated by the amount of 8OHdG) of mtDNA, but not nuclear dna, is inversely correlated with life expectancy (118,119); however, in this case, the question of the accuracy of the methodology again arises. Direct evidence of the significance of mitochondrial ROS production for the mammalian aging process was provided by animal models – genetically modified mice expressing human catalase selectively acting in peroxisomes, nuclei or mitochondria. Only the expression of catalase acting directly on mitochondria (MCAT mice) led to a decrease in the level of oxidative damage and caused a very significant increase in both average and maximum life expectancy (up to 21%) (120). In accordance with the causal role of ROS in the formation of de novo point mutations in vivo, 50% fewer point mutations of mtDNA accumulated in the heart cells of MCAT mice than in wild-type cells of the corresponding age (Fig. 1) (64). Contrary to the evidence of the assumption obtained in experiments on MCAT mice, heterozygous knockout MnSOD^/–mice did not show phenotypes corresponding to premature aging, despite elevated levels of 8OHdG in both nucleic acid and mtDNA (121). In these animals, however, there was an increased incidence of malignant diseases and to date it is unclear whether they have an increased number of mtDNA mutations. To resolve these contradictions, it is necessary to conduct further studies of mitochondrial ROS production, ROS detoxification and redox mechanisms, mtDNA damage repair mechanisms and their effect on age-associated mtDNA integrity.

A more correlative proof is based on the observation that significantly less mitochondrial superoxide is synthesized and fewer mtDNA deletions accumulate in the body of animals under conditions of calorie restriction. In addition, such animals are less likely to develop degenerative diseases, and their average and maximum life expectancy is higher than in animals that feed ad libitum ("for their own pleasure") (122,123). However, the superoxide itself does not damage DNA (14), and the increase in life expectancy associated with calorie restriction and the prevention of age–related diseases, in addition to a decrease in superoxide production and oxidative damage (124), may involve a large number of hormonal, physiological and biochemical changes - for example, possibly changes in iron metabolism (15). The exact sequence of events and the role of ROS in these processes in the causal context is still unclear.

2.1.7. Mosaic is a key aspect of pathology associated with age–related accumulation of mtDNA mutations Important details concerning the possible functional role of low levels of mtDNA mutations in the aging process were discovered when studying individual cells of a number of postmitotic tissues, especially skeletal muscles.

Sarcopenia (age-related atrophy and loss of muscle fibers) in humans by the age of 80 leads to a loss of up to 40% of muscle mass (125,126). Immunohistochemical analysis of aging muscle tissue revealed a clear age-associated increase in the number of fibers showing abnormalities in the mitochondrial electron transport chain. These include the so-called COX fibers, which give a negative reaction when coloring complex IV (cytochrome C oxidase, COX) (127-129). In situ hybridization demonstrates that these electron transfer chain anomalies are associated with mtDNA deletions (130-132). Even though the total number of deletions in the tissue as a whole remains at a low level (88), analysis of muscle fiber bundles revealed foci containing a large number of deletions in fiber segments characterized by anomalies of the electron transport chain and atrophy (133-135).

With the help of laser microdissection and quantitative PCR of successive longitudinal sections of muscle fibers, it is possible to observe the progression of the disease along individual muscle fibers and understand the sequence of events that ultimately lead to cell death and loss of muscle fibers. According to the results of such an analysis, the number of mtDNA deletions clonally increases within individual muscle fibers, which first leads to an aggravation of anomalies in the electron transport chain, after which muscle fibers split, atrophy and die (135,136). Based on the observations made, the researchers estimated that about 15% of the muscle fibers of aging rats have regions with significant disturbances in the functioning of the electron transport chain (135).

It has also been shown that a clonal increase in the number of mtDNA point mutations that damage tRNA genes has a pronounced correlation with COX deficiency in human muscle fibers [137]. In the case of skeletal musculature, a high degree of mosaic and clonal increase in mtDNA mutations provide a mechanism by which relatively low levels of mtDNA mutations in the tissue mass can cause age-related deterioration of the functional and structural condition usually observed in certain regions of this tissue. The fact that the fibers most susceptible to this process eventually die off explains the preservation of the level of mtDNA deletions in the total tissue mass below a certain level. Mutant mtDNAs that negatively affect the electron transfer chain are simply eliminated from the tissue as the damaged cells die off.

Focal spread of mtDNA mutations is also found in other postmitotic tissues. According to the results of two independent studies of individual human cardiomyocytes, 15% and 25%, respectively, of the cells of the most elderly patients contained mtDNA deletions (138,139). The same studies revealed the absence of mtDNA deletions in cardiomyocytes of young individuals.

It is believed that in the brain, neurons of the substantia nigra are subject to increased levels of oxidative damage and accumulate tens of times more deletions than neurons in other regions of the brain (87,98,140). It is possible that the high oxidative load in the substantia nigra is a consequence of high-cost energy metabolism in dopamine neurons [141]. In elderly individuals, up to 30% of substantia nigra neurons exhibit defects in mitochondrial respiratory activity. Such cells are characterized by high levels of clonally increased number of deletions (87,142,143). In addition to the great need for oxygen, the destruction of the "neuromelanin-iron" system is fraught with a further increase in the levels of oxidative stress. ROS can cause both the release of a soluble derivative of melanin (the so-called "melanin-free acid") and iron from the melanin framework. In addition to the presence of dopamine, complexes of iron and a soluble melanin derivative make dopaminergic neurons of the substantia nigra particularly susceptible to the development of oxidative stress [144]. An interesting hypothesis, but not proven to date, is the assumption that in vivo this pathological mechanism involves increased mobilization of iron in the mitochondria. The neurotransmitter L-dopa (L-dopa) is considered the gold standard for the treatment of Parkinson's disease (PD) (145). At the same time, it is believed that L-dopa has a neurotoxic effect. This assumption is confirmed by the data of Alam et al., demonstrating an increase in oxidative damage of proteins in the substantia nigra and brain regions in patients with Parkinson's disease, which are usually unchanged in Parkinson's disease (146). This increase in the level of oxidative damage may be partly due to the formation of cytotoxic adducts (attachment products) of L-dopa-cysteinyl, the level of which increases in the substantia nigra of patients with Parkinson's disease [147]. On the other hand, the results of the ELLDOPA study, which for the first time analyzed the dose-response effect caused by L-dopa in patients with Parkinson's disease, indicate in favor of the neuroprotective nature of the action of L-dopa (148).

In general, the accumulation of sporadic point mutations of mtDNA in the brain increases significantly with age and negatively correlates with the activity of the mitochondrial enzyme cytochrome C oxidase (149,150). In the large intestine, the role of clonally increased mass of point mutations as the cause of the appearance of crypts with cytochrome-C oxidase deficiency has also been proven (151,152).

In all the cases mentioned above, mutant mtDNA, possibly, but not necessarily, the result of indirect damage to mtDNA by reactive oxygen species, undergoes clonal expansion until the entire cell begins to experience a deficit in the normal functioning of the electron transport chain, while in the tissue the number of mtDNA deletions as a whole remains at a low level.

Experts have proposed various models of clonal expansion, but the exact mechanisms that trigger this process are poorly understood at the moment (153-157). Interestingly, the ability of caloric restriction of the diet to reduce the production of ROS in mitochondrial complex I and the formation of de novo mtDNA deletions has been proven. On the other hand, there is evidence that caloric restriction does not suppress the clonal expansion of existing mtDNA deletions (122,123). It is possible that clonal expansion independent of ROS is a mechanism (possibly the main one) by which low levels of mtDNA mutations in certain foci accumulate to values sufficient to disrupt the functioning of cells and, ultimately, tissues.

2.2. How can mtDNA mutations serve as a causal factor of age-associated tissue dysfunction?Recently, experts are increasingly inclined to believe that ROS are not only harmful by-products of metabolism, but also play an important role in signal transmission, especially between mitochondria and the nucleus, as well as in controlling the mechanisms of cell death and proliferation (analyzed in (7,15,158)).

To date, there is no explanation for how ROS avoid meeting with mitochondrial and cytosolic antioxidant defense mechanisms and modulate gene expression in the nucleus.

Cells suffering from clonally "multiplied" mtDNA mutations affecting the electron transfer chain often experience a decrease in the efficiency of electron transfer. Such changes can increase the concentration of oxygen inside the mitochondria due to a decrease in oxygen consumption, and also contribute to a further decrease in the efficiency of the electron transport chain. This should increase the production of ROS, which, in combination with low levels of ATP, can stimulate the mechanisms of cell death (159,160).

Consequently, as in the case of sarcopenia, mtDNA mutations can stimulate age-associated disorders of postmitotic tissue functions by triggering the death of cells that are not capable of recovery (161,162). The increased susceptibility of cells and organisms to stressful stimuli and influences is one of the defining characteristics of the aging process in all species exposed to it.

This model is confirmed by the results of in vitro experiments using cells carrying inherited mtDNA pathology (163), as well as data obtained on transgenic mouse models, such as mutator mice, which often (but not always) demonstrate an increased level of cell death as a phenotype characteristic of genetic interventions that negatively affect the integrity of mtDNA (analyzed in (162)).

It is known that signaling mechanisms mediated by ROS and redox reactions affect various processes, including energy homeostasis, stress resistance, inflammation, mitogenesis and the threshold of cell death (7,15,158). The gradual loss of mtDNA integrity associated with changes in energy homeostasis and redox status can potentially affect a wide range of cellular processes, resulting in a violation of tissue homeostasis by triggering additional mechanisms such as pro-inflammatory reactions.

The existence of alternative mechanisms is also discussed, due to which low levels of focal-widespread mutant mtDNA can cause significant disorders of tissue functioning (133,164-167). Some of these hypotheses imply mechanisms by which cells containing mtDNA mutations can have a pronounced toxic effect on the surrounding tissue (165,166). One of these theories, called the "reducing hotspot hypothesis", is based on the assumption that cells can compensate for the presence of mitochondria with disturbances in the electron transport chain by reducing the level of molecular oxygen using the plasma membrane redox system, producing extracellular superoxide and increasing the oxidative load on the environment. fabric (165,167). An alternative hypothesis proposed by Mott et al. in 2005 (164), states that mitochondrial mutations lead to the transcription of proteins encoded in mtDNA with an abnormal structure of molecules, which subsequently interact with mitochondrial mediators of cell death. According to this hypothesis, due to the fragility of the balance between mediators mediating the initiation or prevention of cell death, even a small number of aberrant transcripts are potentially capable of causing cell death [164].

3. Arguments in favor of directed determinants of the basal rate of aging, rather than individual age-related diseases Currently, the aging populations of industrialized countries live in conditions of low mortality due to external causes.

The leading causes of morbidity and mortality in these countries are chronic degenerative diseases. Naturally, a huge amount of resources is spent on the search for treatment or effective preventive measures for each of these diseases (168). However, since age itself is not the single most important risk factor for most degenerative diseases, treatment or prevention of any of them in an aging population is likely to lead to its rapid replacement by another, possibly no less severe disease. Thus, as the population ages, interventions aimed at treating or preventing certain diseases will increasingly reduce overall mortality, as well as the associated costs of ensuring the operation of health care systems.

For comparison, a 20% increase in life expectancy observed in genetically modified MCAT mice expressing human catalase (see section 2.1.6.) would have a 5 times greater effect than the result predicted if all forms of cancer could be eliminated from the human population (169,170). An interesting fact is that the development of heart pathologies and the formation of cataracts is delayed in MCAT mice, which indicates a decrease in the severity of age-related diseases, simultaneously with a decrease in the basic rate of aging (120). The predominance of the rate of aging over the progression of specific pathological processes was subsequently elegantly demonstrated in 2004 by comparing the time of onset and the rate of progress of neurodegeneration due to the spectrum of functionally equivalent mutations of orthologous genes of five mammalian species with a maximum lifespan of 3.5 years (mice) to 122 years (humans) (161). According to the analysis of the data obtained, the rate of neurodegeneration in each case decreases proportionally with an increase in the maximum life expectancy, despite the fact that the two processes are based on identical molecular mechanisms. In response to functionally identical mutations, aging species progress faster according to identical sequences of neurodegenerative events at a rate proportionally higher than the rate characteristic of slower aging species (161). The loss of cells through the activation of cell death mechanisms is a critical moment in all the diseases under consideration.

As described above, mitochondria and mitochondrial ROS production are intricately interconnected with the threshold of triggering and implementation of cell death mechanisms, and mitochondrial modulation of these mechanisms, thus, may be an important factor explaining the phenomenon described above (161). The possibility that mitochondrial ROS production and mtDNA integrity may be important determinants of cell death and age-dependent inflammatory processes, as well as the basal rate of aging of the body, makes mitochondria the main target for the development of therapies aimed at preventing or mitigating age-associated degenerative processes. When methods for accurately modulating mtDNA integrity in vivo become available, testing the effect of reducing the load of mtDNA mutations on life expectancy will provide more answers to the question of whether mtDNA damage is related to aging.

4. StrategiesThere is evidence that as the body ages, a moderate amount of oxidative damage to proteins, lipids and DNA accumulates in many tissues (171-176), while the increase in their total amount is quite small.

In addition, some of the detected damages demonstrate properties that do not correspond to damages that occur in a random stochastic way. For example, oxidative modifications of the protein are observed mainly in a small number of specific targets (described in (15)). According to one of the assumptions based on the free radical theory of aging, antioxidants should slow down the aging of the body. As described above, ROS can indeed cause at least some of the original mtDNA mutations. However, to date, there is little evidence of a significant and random increase in the amount of oxidative damage, initially assumed by the "vicious circle" theory. Moreover, many age-related diseases are themselves associated with increased oxidative damage (15,35). Thus, when studying ROS-mediated macromolecular damage and aging, it is very difficult to differentiate the causes from the consequences [7]. This is of great importance for any intervention strategy based on modulating mitochondrial ROS production.

Experiments on MCAT mice and mice kept under caloric restriction conditions (64,122,123) provided a number of evidence that reducing the effects of ROS-mediated damage in mitochondria protects mtDNA and increases life expectancy. However, according to the data summarized above, ROS-mediated mtDNA mutagenesis de novo is most likely a relatively rare phenomenon. The mechanism of clonal expansion implies that strategies aimed at modulation or binding of mitochondrial ROS can presumably influence only de novo mutagenesis and do not affect already formed mtDNA mutations. Thus, such strategies require long-term use, and their effectiveness will decrease as the body ages. Modulation of mtDNA damage repair mechanisms is a little-studied approach to affecting the integrity of mtDNA and deserves more attention. Most of the traditional strategies discussed below are aimed at modulating the production of mitochondrial ROS and oxidative phosphorylation reactions.

4.1. Mitochondrial nutrientsThe critical point is to maintain a stable state of mitochondria in the aging process.

The integrity of mitochondria depends on a huge number of biomolecules, some of which, in addition to endogenous biosynthesis, can enter the body directly from food and dietary supplements. Given the importance of compounds such as L-carnitine/acetyl-L-carnitine, coenzyme Q10 (CoQ10) and lipoic acid for mitochondrial bioenergetics, increasing their levels in the body through dietary or pharmaceutical interventions is considered a promising strategy for maintaining health and slowing aging.

Before discussing the effectiveness of these so-called mitochondrial nutrients in preventing aging and age-related diseases, we will provide brief information about: 1) L-carnitine/acetyl-L-carnitine, coenzyme Q10 (CoQ10) and lipoic acid; 2) possible mechanisms of their action and 3) evidence of their potential positive effect on the condition health that can be obtained as a result of human trials.

4.1.1. L-carnitine/Acetyl-L-carnitine 4.1.1.1. Basic information

The four-component ammonium compound L-carnitine is found everywhere.

In the human body, L-carnitine is synthesized from its precursors, the amino acids lysine and methionine, by the kidneys, brain and, especially, liver at an average daily rate of 1.2 micromol/kg of body weight. However, most of the L-carnitine is found in human skeletal muscles. The biosynthesis of L-carnitine is closely interrelated with the outer mitochondrial membrane (177, 178).

L-carnitine is necessary for beta-oxidation of long chains of fatty acids in mitochondria (Fig. 2). Consequently, in a state of L-carnitine deficiency, the body cannot utilize the main energy source (fat) for ATP synthesis (177). L-carnitine is a part of almost all food products. However, compared to vegetable products, animal food contains much more L-carnitine, which exists both in free and in acetylated (fatty acid-bound) form. Acetyl-L-Carnitine accounts for 5-30% of all food L-carnitine (177,179). Despite the lack of information about the needs for L-carnitine throughout the life cycle, the results of some studies indicate an age-associated decrease in the level of L-carnitine in various tissues not only of animals, but also of humans (179,180).

Fig. 2. Localization and key functions of mitochondrial nutrients (details in the text). (A) Coenzyme Q10 (CoQ) transfers electrons from complexes I and II to complex III. Lipoic acid in the form of lipoamide acts as a cofactor of the mitochondrial enzymes alpha-ketoglutarate decarboxylase (a-KG-DC) and pyruvate dehydrogenase (P-DH). Lipoic acid is also necessary for the catalysis of the cycle of alpha-keto acids with branched chains (not shown). (C) Carnitine promotes the transfer of long-chain fatty acids through the external and internal mitochondrial membranes (ACS = acyocoenzyme-A-synthetase; CACAT = carnitine-carnitine-acylcarnitine translocase; CAT = carnitine acetyltransferase; CPT = carnitine palmitoyl transferase). The diagram is borrowed from (23,177,241).

4.1.1.2. Possible mechanisms of action4.1.1.2.1. Influence on mitochondrial bioenergetics

As mentioned earlier, a decrease in the activity of mitochondrial enzymes is a sign of the aging process.

The subchronic addition of 300 mg/kg of L-carnitine to the diet of aging rats (>22 months) prevented an age-related decrease in the activity in brain tissues of both individual enzymes of the tricarboxylic acid cycle (Krebs cycle) and complexes I and IV of the electron transport chain, maintaining their levels within values comparable to similar indicators for brains of young (3-4 months) animals. It should be noted that L-carnitine did not cause any significant changes in the brains of young animals (181). The effects of L-carnitine supplementation on the activity of mitochondrial enzymes are not limited to brain tissues and are observed in other tissues, such as skeletal muscles and the heart (182).

A brain-specific increase in cytochrome oxidase (complex IV) activity in synaptic and non-synaptic mitochondria was also observed when acetyl-L-carnitine was added to the rat diet (183,184). In the same animals, the activity of Krebs cycle enzymes, especially citrate synthetase, decreased in all regions of the brain under consideration, which indicates different modulating effects of acetyl-L-carnitine and L-carnitine on the energy metabolism of the brain. Changes in the status of acetylation of lysine residues are considered as one of the possible mechanisms of action of acetyl-L-carnitine on the activity of mitochondrial enzymes [185].

Moreover, intravenous administration of acetyl-L-carnitine at a dose of 500-750 mg/kg caused a significant regional increase in glucose metabolism in rats, while the simultaneous use of acetate and L-carnitine did not affect glucose utilization in any way [186]. As noted above, acetyl-L-carnitine acts as a carrier of acyl groups, including for the synthesis of acetylcholine, between mitochondria and cytosol. Therefore, it is interesting to note that acetyl-L-carnitine most significantly enhances glucose metabolism in subcortical cholinergic regions. A similar effect of acetyl-L-carnitine on energy reserves was revealed by Al-Majid et al. (187), who noted a 50% increase in ATP levels in the hippocampus of rats treated (intraperitoneally) with acetyl-L-carnitine at a dose of 300 mg/kg. A similar protocol provided a significant degree of protection against ischemia-induced oxidative stress and damage to hippocampal cells. The intake of oral acetyl-L-carnitine supplements by rats at a concentration of 1.5% (weight/volume) for one month led to a significant recovery of age-associated decrease in oxygen consumption and metabolic parameters, such as gluconeogenesis, in the perfused liver model. However, it is believed that the positive regulation of oxygen consumption is due to an increase in the activity of acyl-CoA synthetase rather than an increase in the maximum oxygen capacity [188]. An interesting fact is that under the same experimental conditions, the same group of researchers revealed not only increased respiration of skeletal muscles in old rats who consumed acetyl-L-carnitine, but also a significantly increased mass of mitochondrial proteins (189). Since acetyl-L-carnitine is rapidly hydrolyzed to L-carnitine, it is very difficult to determine whether the effects observed after intravenous and intraperitoneal administration of acetyl-L-carnitine are the result of the action of the compound itself. Data on the activity of orally taken acetyl-L-carnitine should also be interpreted with special attention, since acetyl-L-carnitine absorbed in the small intestine begins to deacetylate to L-carnitine already in intestinal cells, in which L-carnitine, in turn, is partially reacetylated to acetyl-L-carnitine. In general, due to the high metabolic rate of acetyl-L-carnitine, it is very difficult to clearly distinguish the effects associated with L-carnitine and acetyl-L-carnitine (190, 191).

4.1.1.2.2. Effect on antioxidant protectionDespite the fact that L-carnitine and its derivatives (namely acetyl-L-carnitine and propionylcarnitine) do not have pronounced direct antioxidant activity (as can be assumed from the chemical structure of their molecules (15)), their ability to indirectly modulate the antioxidant defense system is periodically mentioned in the literature.

In parallel with the restoration of the activity of Krebs cycle enzymes and mitochondrial complexes in certain regions of the brain of rats subchronically taking L-carnitine at a dose of 300 mg/kg, an improvement in the antioxidant status of the brain was observed, as evidenced by an increased content of glutathione, ascorbic acid and alpha-tocopherol, as well as a reduced accumulation of lipofuscin, protein carbonyls and peroxidation products lipids (active forms of thiobarbituric acid) (192). Despite widespread use, the level of active forms of thiobarbituric acid cannot be considered a quantitative indicator reflecting the damage caused by lipid oxidation, since the formation of active forms of thiobarbituric acid is influenced by a huge number of compounds [15]. The significance of changes in the level of ascorbic acid in animals capable of synthesizing this compound independently is very difficult to assess.

Administration of 400 mg/kg of acetyl-L-carnitine to SAMP8 mice (prone to rapid physiological aging) three times a week until they reached 4 months of age significantly reduced the level of hydroperoxide in the brain. At the same time, the indicators of age-related learning and memory disorders significantly improved in animals (193). Given the different phenotypes of SAMP lines(194), presumably due to mechanisms other than the aging process itself, the SAMP model can be a valuable tool for studying chronic degenerative diseases.

In two studies, researchers compared the effects of propionylcarnitine and L-carnitine on antioxidant status and biomarkers of oxidative stress in the heart and aorta of rats with spontaneous hypertension and normal pressure (195,196). In general, the condition of hypertensive animals improved with the administration of propionylcarnitine and L-carnitine. At the same time, propionylcarnitine had a negligible effect on the aorta of healthy animals, and L-carnitine had a negative effect on metabolism, which was manifested in a deterioration of NADPH-stimulated superoxide production and a decrease in the expression of endothelial NO synthetase (eNOS). Based on the latest observation, it can be assumed that the beneficial effects of L-carnitine are limited to states of dysfunction (such as aging and disease) and may be absolutely useless to prevent normal age-related deterioration of the functions of a healthy body.

On the other hand, L-carnitine has the ability to increase the amount of other antioxidants, such as alpha-tocopherol, in the liver of rats (197). In contrast, short-term administration of acetyl-L-carnitine not only reduces the level of vitamin C in hepatocytes (but not in heart tissue) of both young and aging rats, but also increases the formation of a marker of lipid peroxidation of malondialdehyde, evaluated by gas chromatography/mass spectrometry (an adequate biomarker of the final product of peroxide lipid oxidation), in the liver of old rats (198). However, again, the effect on the biosynthesis of ascorbic acid must be considered taking into account the interpretation of changes in its level.

Despite the great interest in the use of L-carnitine, acetyl-L-carnitine and propionylcarnitine to prevent and/or reduce the consequences of pathologies associated with oxidative stress and, accordingly, with aging, very little is known about the real effect of these compounds on the antioxidant status and the level of oxidative stress in the human body. In accordance with the results of the above-mentioned animal studies, elevated plasma levels of retinol and alpha-tocopherol were observed in women who took L-carnitine orally at a dose of 680 mg/ day for three weeks. At the same time, the number of lipid peroxidation products in plasma was significantly reduced (199), although the significance of the latter observation is unclear due to the use of a method for assessing the level of active compounds of thiobarbituric acid, which often gives erroneous results. When assessing the level of 8OHdG in the rat brain using high-performance liquid chromatography with electrochemical indication, Haripriya et al. a significant decrease in the frequency of mtDNA damage was revealed in old animals treated with L-carnitine (200). High-performance liquid chromatography with electrochemical indication is considered the best method for quantifying 8OHdG, although there is still uncertainty about the exact background levels of 8OHdG(47).

4.1.1.3. Human impact studies on aging and age-related diseasesL-carnitine and its derivatives are effective in the treatment of primary carnitine deficiencies.

The general conclusions drawn on the basis of the results of these studies are exhaustively analyzed in the works of other authors (201,202).

Since the level of carnitine decreases with aging and may probably be insufficient to meet the needs that increase with aging, specialists have paid great attention to the possibilities of using and effectiveness of L-carnitine, acetyl-L-carnitine and propionylcarnitine to suppress pathological manifestations of age-related diseases (203). Propionylcarnitine, highly specific for skeletal and cardiac muscles, was used for the treatment of cardiovascular diseases (204). In primary small-scale trials, propionylcarnitine intake had a significant effect on the ability of patients with cardiovascular diseases to tolerate physical activity (204). The results of the subsequent multicenter study confirm the ability of propionylcarnitine to increase the duration of physical activity tolerated by patients with chronic heart failure who are on stable pharmacotherapy. It should be noted that the best results of taking propionylcarnitine are observed in patients with relatively preserved myocardial function and, accordingly, with moderate deterioration in physical condition (205). When assessing the effect of oral administration of L-carnitine (2 g per day, for 6 months) on physical and mental fatigue of people aged 100 and over, as well as on their cognitive function in a randomized double-blind trial conducted under placebo control, Malagnarnera et al. we found a significant improvement in all three parameters (206). However, based on the assumption that acetyl-L-carnitine enters the human body, and the observation that it moves through the blood-brain barrier faster than L-carnitine using a saturated sodium-dependent process, among carnitine derivatives, acetyl-L-carnitine is the compound of choice when studying the possible positive effect of acetyl-L-carnitine on the cognitive function of the elderly (185,207). At the same time, when treated with acetyl-L-carnitine, there is a tendency to decrease the manifestations of cognitive impairment, especially in patients with initially less pronounced disorders (185). However, due to the small number of patients who participated in these studies, as well as the rather moderate effect of acetyl-L-carnitine on cognitive status, further studies are needed to evaluate the effectiveness of acetyl-L-carnitine in the treatment of dementia.

4.1.2. Coenzyme Q104.1.2.1. Basic information

Coenzyme Q (also known as ubiquinone) consists of a 2,3-dimethoxy-5-methylbenzoquinone core and a side chain consisting of isoprene molecules (208).

Coenzyme Q10 (CoQ10) is the most common ubiquinone derivative that is part of human tissues and biological fluids, where it occurs in various forms, differentiated depending on the redox status of the quinone nucleus (208-210):

1. Fully oxidized form – ubiquinone (Q)

2. partially restored, free radical form – ubisemiquinone (^•QH)

3. Fully recovered form – ubiquinol (QH ₂)The potential positive effect of CoQ10 on health is mainly related to its key role in mitochondrial bioenergetics (Fig. 2) and, possibly, to its direct and indirect antioxidant activity (208,211), although other functions of CoQ10, for example, modulating gene expression, are also considered (212).

Maintenance of mitochondrial function and removal of ROS are considered as mechanisms of the supposed preventive and therapeutic effects of CoQ10 in relation to cardiovascular, neurodegenerative and mitochondrial diseases (202,209,213). Since the consumption of conventional foods normally does not have a significant effect on the level of CoQ10 in plasma, to increase its concentration in the blood, it is necessary to add CoQ10-enriched supplements to the diet (209).

Despite the fact that _the CoQ10H2/CoQ10 ratio is subject to preanalytical variations (such as the time elapsed after blood sampling), this indicator has been proposed as a possible biomarker for assessing the level of oxidative stress in vivo (214,215). When studying plasma samples of healthy men, Wada et al. a small but significant increase in the relative content of the oxidized form of CoQ10 was found in older individuals (20-39 years – 3.1±0.9%; 40-59 years - 3.6± 1.2%; >60 years – 4,7±1,6%) (216), while the level of CoQ10H ₂ remained constant regardless of age. Although the authors of the latest work concluded that there is an association between the redox status of CoQ10 and calendar age, it is still necessary to find out whether these small changes in plasma CoQ10 content are biologically significant in relation to aging and the onset of chronic degenerative diseases.

4.1.2.2. Possible mechanisms of action 4.1.2.2.1. Influence on mitochondrial bioenergetics

Electron transfer through the inner mitochondrial membrane forms the basis of mitochondrial respiration and ATP synthesis.

CoQ10 plays a fundamental role in this cellular process by transferring electrons from complexes I and II to complex III(217).

The heart tissue, due to the increased need for ATP, is especially rich in CoQ10. Myocardial dysfunction is often detected in the elderly and is often aggravated during interventions that induce aerobic or ischemic stress (218). Daily intraperitoneal administration of CoQ10 to old rats (35 months) for 6 weeks (4 mg/kg) significantly improved oxygen uptake by cardiac tissue and contributed to the restoration of heart function compared to control group animals (218). Similarly, tissue samples of patients (>70 years old) who received 300 mg of CoQ10 per day (for 7 days) before heart surgery demonstrated more effective mitochondrial respiration and faster recovery after a period of hypoxia compared to individuals in the control group. Apparently, these effects are directly related to an increase in the level of CoQ10 in the tissue and, accordingly, in the mitochondria [218].

Since mitochondrial dysfunction is a sign of neurodegeneration, specialists have repeatedly tested the effectiveness of CoQ10 treatment in restoring and/or maintaining the mitochondrial bioenergetics of the brain. However, an adequate interpretation of data such as those presented below, obtained in experiments on rodents who used oral CoQ10 supplements, is complicated by the fact that CoQ9 is found in much larger quantities in mice and rats than CoQ10 (209). Nevertheless, the results of animal studies indicate that CoQ10 supplements are able to weaken the breakdown of ATP in the brain when administered neurotoxins such as malonate, 3-nitropropionic acid, 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) or aminooxyacetic acid (described in (210)). However, very little is known about whether CoQ10 treatment has a direct effect on the mitochondria of the brain. According to widespread opinion, the use of CoQ10 does not lead to an increase in the level of endogenous CoQ10, except for its content in plasma and liver. In contrast to this view, elevated levels of CoQ10 were found in rat brain tissue after oral administration of CoQ10 (200 mg/kg, for 2 months) and its water-soluble form (150 mg/kg, for 4 and 13 weeks) (219,220). According to the results of a later work by Sohal et al. (2006), an increase in the level of CoQ10 is observed in liver, heart, kidney and skeletal muscle homogenates, as well as isolated mitochondria of mice that have been taking various concentrations of CoQ10 for a long time (221). Despite the lipophilic nature of CoQ10, the authors found no signs of an increase in the concentration of CoQ10 in the brain. Moreover, in none of the animals did CoQ10 supplements in any way affect the activity of the mitochondrial electron transport chain oxidoreductase and the state of active respiration ("state 3"), evaluated at the 19th and 25th months of life, and did not change the life expectancy in any way (221). These data correspond to earlier results obtained in experiments on mice (222,223) and fruit flies drosophila, which also did not show an increase in average and maximum life expectancy (224). In fact, feeding drosophila CoQ10 at a concentration of 1.25 mg/ml of medium not only increased ROS production, but also reduced life expectancy (225). Given the importance of the pro-oxidant effects of CoQ10 exerted through the synthesis of superoxide and hydrogen peroxide, mainly through the intermediate compound CoQ10-semiquinone, in organisms (over)saturated with CoQ10, it is quite natural to expect enhanced ROS production [226]. The results of studies of the effects of CoQ10 on Caenorhabditis elegans (C. elegans) worms, which are a generally recognized model for studying the aging processes modulated by oxidative stress in postmitotic tissues, also contradict each other (217,227).

4.1.2.2.2. Effect on antioxidant protection Ubiquinol (CoQH 2), which is part of biological membranes and lipoproteins, suppresses lipid peroxidation by absorbing peroxide radicals.

Moreover, CoQH ₂ participates in: 1) binding of active forms of nitrogen, such as nitric oxide and peroxynitrite (228) and 2) in the regeneration of alpha-tocopherol by reducing alpha-tocopheroxyl radicals. From a biological point of view, this effect is considered more important than the direct binding of free radicals (15). While, according to the results of several experiments on cell cultures, CoQ10 has a powerful antioxidant effect (229), the results of a study conducted on rabbit corneal keratinocytes indicate that CoQ10 prevents cell death, regardless of its ability to bind free radicals (230). Another potentially interesting effect of CoQ10 is due to its supposed interaction with mitochondrial uncoupling proteins. It is traditionally believed that the activation of MRB reduces the load of mitochondrial ROS production (210,231).

However, the results of studies on the effect of CoQ10 on antioxidant status and antioxidant protection in animals and, in particular, in humans are not so convincing. Rats fed a diet rich in polyunsaturated fatty acids and additionally treated with CoQ10 had an increased content and/or activity of CoQ10, alpha-tocopherol and catalase in the heart tissue. The use of the CoQ10 supplement also reduced the severity of age-associated deterioration of mitochondrial functions and reduced ex vivo levels of hydroperoxide, assessed using the method of iron oxidation followed by oxidation of xylenol orange (ferrous oxidation in xylenol orange, FOX), in the mitochondria of the heart. These results indicate the ability of CoQ10 in some cases to exert both direct and indirect antioxidant effects in vivo (232). However, as mentioned above, Sohal et al. (2006) could not identify any positive effect of prolonged CoQ10 intake on mitochondrial bioenergetics (221). In fact, despite the accumulation of CoQ10 in tissues (except the central nervous system), such indicators as the content of carbon monoxide, the glutathione redox state of tissues and the activity of supreoxydismutase, catalase and glutathione peroxidase did not change in the liver, kidneys, skeletal muscles, heart and brain [221]. On the other hand, significant antioxidant protection was observed with respect to lipoproteins isolated from animal and human tissues receiving CoQ10 supplementation (214,233), indicating the ability of CoQ10 to have a positive effect on homeostasis of at least one of the putative key compounds triggering the development of atherosclerosis (i.e. oxidized lipoproteins) in vivo. It is impossible to say whether this conclusion extends to the tissues of a living organism, especially the brain, on the basis of existing data.

4.1.2.3. Human impact studies on aging and age-related diseases It has been proven that the myocardial tissue and plasma of patients with cardiovascular diseases contain less CoQ10 than the corresponding tissues of healthy individuals.

Since the 1960s, the effect of CoQ10 on the state of the cardiovascular system has been analyzed in more than 20 studies involving patients with cardiomyopathy and congestive heart failure. Despite the fact that several small-scale studies have not revealed a positive effect of CoQ10, a recent meta-analysis has shown that CoQ10 improves systolic function in chronic heart failure, especially in small heart lesions, possibly due to a larger number of cells capable of recovery (234). To do this, however, it is not enough only to bring the level of CoQ10 in plasma to normal values, it is also necessary to ensure that the concentration of CoQ10 exceeds normal values, presumably to ensure maximum absorption by the tissue (235). There is also published, although less convincing, evidence of the effect of CoQ10 on the condition of patients with coronary heart disease (236).

There are data obtained in animal experiments, according to which, in the case of neurodegeneration, CoQ10 can act as a neuroprotector (237,238). Despite the fact that CoQ10 deficiency has been proven in neurodegenerative diseases, oral administration of CoQ10 drugs to patients with Parkinson's disease and Huntington's chorea caused only a slight improvement in neurological symptoms. The condition of patients with Friedreich's chorea improved when taking CoQ10 (in combination with vitamin E), which manifested itself in improving the bioenergetics of skeletal and cardiac muscles, but no positive effect on neurological parameters was observed [239]. To date, there are no data obtained from clinical trials that would support the use of CoQ10 for the prevention or treatment of Alzheimer's disease (240).

4.1.3. Alpha-lipoic acid4.1.3.1. Basic information

As well as L-carnitine and CoQ10, lipoic acid is a natural compound most often covalently bound to the epsilon-amino group of the lysine residue.

Lipoic acid is usually found in products of vegetable (spinach) and animal (heart) origin. It is read that lipoic acid is synthesized in mitochondria from octanic acid and sulfur compounds (241-243).

Oxidative decarboxylation of pyruvate, alpha-ketoglutarate and branched alpha-keto acids occurs in mitochondria and is critically dependent on lipoic acid (in the form of lipoamide), which in this case acts as an enzymatic cofactor (Fig. 2). During the transfer of acyl groups from the enzymatic complex to another biomolecule, lipoic acid is reduced to dihydrolipoic acid, which in the presence of lipoamide dehydrogenase can be oxidized back to lipoic acid with the formation of NADH (243).

Free lipoic acid is considered to be the most important therapeutic form of lipoic acid. While eating regular food is associated with only insignificant levels of free lipoic acid, taking its purified form causes a significant but short-term increase in the concentration of free lipoic acid in plasma [243]. In addition to participating in the regulation of mitochondrial energy metabolism through electron transfer to NAD, the redox pair lipoic acid/dihydrolipoic acid in a free, non-protein-bound form has a significant effect on the balance of cellular oxidative stress in vitro. The question remains to what extent free lipoic/dihydrolipoic acid acts as an antioxidant in vivo or its supposed effectiveness is the result of maintaining and restoring levels of other ROS scavengers, such as glutathione and vitamin C (242,244).

4.1.3.2. Possible mechanisms of action 4.1.3.2.1. Influence on mitochondrial bioenergetics

The effect of intravenous administration of lipoic acid on the activity of mitochondrial enzymes in young and old rats has been studied in several studies.

With daily administration of lipoic acid for 14 days at 100 mg / kg, elderly animals showed a significant (sometimes insignificant) decrease in the activity of various enzymes of the Krebs cycle, as well as complexes I and IV in the liver and kidneys compared with control group animals (245). Administration of the same dose of lipoic acid for 30 days resulted in a significant increase in the activity of Krebs cycle enzymes and complexes I, II, III and IV in the mitochondria of brain cells of old rats (246). Similar to the previously described mitochondrial nutrients, lipoic acid has only a slight positive effect on young animals.

In addition to alpha-ketoglutarate dehydrogenase, which directly depends on lipoic acid as a cofactor, the activity of other mitochondrial enzymes also increased when lipoic acid was administered to animals. This fact indicates that lipoic acid can perform other functions in the cell, for example, influencing the antioxidant status and thus improving mitochondrial bioenergetics.

In contrast to the results obtained in rats, the analysis of postmortem brain tissue samples of elderly individuals, as well as patients with Alzheimer's disease and multiinfarction dementia did not reveal a decrease in succinate dehydrogenase (SDH) activity, which indicates the existence of species-specific differences in age-associated mitochondrial bioenergetics disorders [247]. However, in postmortem tissue samples of patients with Alzheimer's disease and multiinfarction dementia, the activity of the lipoic acid-dependent pyruvate dehydrogenase (PDG) complex was reduced. Interestingly, it was possible to stimulate the activity of pyruvate dehydrogenase using lipoic acid exclusively in tissue samples of patients with multiinfarction dementia, but not with Alzheimer's disease. Moreover, based on the observation that the activity of succinate dehydrogenase does not differ either under normal conditions or under stimulation in both control samples and in samples of patients with multiinfarction dementia and Alzheimer's disease, the following conclusions can be drawn: 1) the loss of mitochondria does not explain the decrease in the activity of pyruvate dehydrogenase and 2) the selective sensitivity of pyruvate dehydrogenase (and possibly other mitochondrial enzymes) may be due to direct or indirect, disease-related, structural and/or functional impairment of enzymes (248).

Since mitochondrial biogenesis largely depends on the regulatory action of PGC-1-alpha (activated by peroxisome proliferators receptor-gamma-coactivator-1-alpha), the observation according to which lipoic acid (as well as, for example, resveratrol) It is able to stimulate the PGC-1-alpha-mediated signaling mechanism and thereby increase the content of mtDNA and proteins, provides a new promising approach to the detailed study of the possible effects of lipoic acid in preventing mitochondrial dysfunction [249].

4.1.3.2.2. Effect on antioxidant protection Lipoic and dihydrolipoic acids are described as the most versatile antioxidant pair.

This statement is based on the fact that either both or at least one of the compounds can interact with all important free radicals and other ROS (241,250,251). Moreover, lipoic and dihydrolipoic acids can trigger the reuse of other exo- and endogenous antioxidants, such as vitamins C and E, CoQ10 and glutathione (241,252).

The data according to which lipoic acid exhibits antioxidant activity in vivo, for the most part, were obtained as a result of quantitative analysis of biomarkers of oxidative stress. Plasma and organ samples obtained from animals that consumed lipoic acid showed increased levels of glutathione (253,254) and vitamins C and E (255-258), reduced ROS production (255), lower lipid load (257,259) and a lower degree of DNA oxidation (255), as well as less mtDNA damage (246). The methods used in the above-mentioned works include high-performance liquid chromatography with electrochemical detection to assess DNA damage (8-oxo-dG), the use of fluorescent dye 2'-7'-dichlorofluorescein (DCF) to evaluate ROS production, as well as the use of FOX to assess levels of lipid hydroperoxides. Despite the fact that a number of errors may affect the accuracy of all these methods (15), all the data obtained indicate the presence of antioxidant effects.

On the other hand, pro-oxidative activity can be observed in vitro, characteristic mainly of dihydrolipoic acid (250). Very little is known about the potentially harmful effects of lipoic/dihydrolipoic acids on animals and humans in vivo. Whereas subchronic treatment of old rats with lipoic acid caused increased protein oxidation and nitriding in the brain (but not in the muscles) (259), the life-prolonging effect of lipoic acid on drosophila (260) and C.elegans worms (261) indicates positive metabolic changes in lipoic acid in association with the aging process.

4.1.3.3. Human impact studies on aging and age-related diseases In the 60s, lipoic acid was first used as a therapeutic agent in the treatment of patients with cirrhosis of the liver, symptoms of intoxication and, in particular, diabetic polyneuropathy (243).

The use of lipoic acid in diabetes to prevent polyneuropathic complications and damage associated with oxidative stress is justified only by a small amount of clinical data (262-264). The antioxidant effects of lipoic acid in diabetics with albuminuria and poor glycemic control even persisted for some time (265). The effectiveness of lipoic acid for the prevention and/or symptomatic treatment of chronic age-related cardiovascular diseases and dementia, on the contrary, has yet to be proven. In a review of the Cochrane Collaboration published in 2007, the following conclusion was made: lipoic acid cannot be recommended to patients with impaired cognitive function due to the lack of results of scientifically based randomized clinical trials conducted by a double-blind method under placebo control (266).

There is a lot of information about the multifaceted effect of lipoic/dihydrolipoic acids in vitro and on animal models of cardiovascular diseases(267). However, since the effects of taking lipoic acid have been studied only in several small-scale clinical trials, it is currently impossible to draw conclusions about the effectiveness of its use for suppressing symptoms and pathological mechanisms associated with cardiovascular diseases [267].

4.2. Comparison of combined and separate intake of mitochondrial nutrientsSince L-carnitine/acetyl-L-carnitine, CoQ10 and lipoic acid are involved in the management of mitochondrial bioenergetics and antioxidant protection, it is suggested that taking a combination of two or more mitochondrial nutrients may have higher efficiency compared to taking these compounds separately.

Unfortunately, only a few authors have tried to compare the effects of separate and combined intake of mitochondrial nutrients in relation to biomarkers of oxidative stress and, for example, cognitive functions, such as various types of learning. In experiments on rats, Hagen et al. (198) demonstrated that simultaneous administration of L-carnitine and acetyl-L-carnitine suppressed both an age-associated decrease in vitamin C levels in the liver and an increase in lipid peroxidation (evaluation was performed by gas chromatography/mass spectrometry for malondialdehyde modified with pentofluorophenyl), which indicates a possible synergistic effect of these two nutrients (198). When further studying the effect of L-carnitine, acetyl-L-carnitine and their combinations on age-associated changes in rats, the same group of authors revealed a higher activity of taking a combination of compounds, compared with taking L-carnitine and acetyl-L-carnitine separately, with respect to some (but not all) parameters related to oxidative stress and the state of cognitive function (172, 268). However, since taking acetyl-L-carnitine can stimulate oxidative stress in the form of lipid peroxidation and reduce the level of vitamin C with antioxidant properties in both young and old rats (see above (198)), it is very difficult to determine whether the combination of L-carnitine and acetyl-L-carnitine really has enhanced effect due to synergistic action, or part of the effectiveness of L-carnitine is spent on neutralizing the negative metabolic effects of acetyl-L-carnitine.

Similarly, the results of a careful re-analysis of two data arrays (Fig. 3), recently provided by another group of researchers, question the claim that the combined intake of lipoic acid and, in this case, L-carnitine has a more pronounced effect compared to the intake of individual mitochondrial nutrients (182,269).

The use of L-carnitine in both studies slowed down the age-associated decrease in the activity of mitochondrial enzymes in heart cells (Fig. 3), and the authors claim that the administration of L-carnitine and lipoic acid to old rats for 30 days led to a further improvement in the activity of mitochondrial enzymes, significantly exceeding the effects of taking L-carnitine alone. However, the effects exerted by the combination of L-carnitine and lipoic acid in work II do not differ from the effects identified during work I, in which the effects of the use of L-carnitine alone were evaluated (for 21 days). Based on the information provided, the age of the animals, as well as the method of administration and dosage of L-carnitine and lipoic acid did not differ during both studies, which minimizes the effect of differences, except for a slight difference in the duration of studies, on the inconsistency of the data. However, oral administration of L-carnitine in combination with N-acetylcysteine and S-adenosylmethionine improved cognitive function and reduced aggression in adult mice expressing the human gene of the fourth type of apolipoprotein E (ApoE4), which is a risk factor for Alzheimer's disease. In this case, the state of cognitive function of wild-type mice also improved when taking a combination of nutrients (270).

Thus, to date, there is a very limited amount of evidence that the combined intake of mitochondrial nutrients can have a cumulative or even synergistic positive effect on health in vivo, and discrepancies in the results obtained by different laboratories require special attention.

3. The effect of taking L-carnitine (LC) compared with taking a combination of LC and lipoic acid (LA) on the enzymatic activity of alpha-KGD, isocitrate dehydrogenase (ICD), malate dehydrogenase (MD) and succinate dehydrogenase (SD) on the heart of young and old rats. Data are taken from Kumaran et al. (182) and Savitha et al. (269). Study I: taking LC (300 mg/kg) for 21 days; study II: taking LC (300 mg/kg) for 30 days and taking a combination of LC (300 mg/kg) and LA (100 mg/kg) for 30 days

# – significantly differs from young animals of the control group (issl. I);
±±– significantly different from the young animals of the control group (issl. II);
a – significantly differs from the old animals of the control group (issl. I);
b – significantly differs from the old animals of the control group (issl. II);
c – significantly differs from the old animals of the control group (issl. II);
* – significantly different from the old animals that took LC (issl. II).
It should be noted similar effects of LC(issl. I) and combinations of LC LA (exl. II), as well as the estimated most pronounced effect of using only LC(exl. I compared to issl. II) on enzymatic activity in old animals.

4.3. Antioxidants – some general informationA prerequisite for antioxidant therapy to have a pronounced protective effect against oxidative damage to mtDNA is the creation of a sufficiently high concentration of applied antioxidants in the immediate vicinity of mtDNA, which is necessary for the absorption of ROS with high efficiency.

Even though there is some evidence of the protective effect of classical antioxidants (55,60,271) and mitochondrial nutrients (see above) on mtDNA, any strategy for the use of antioxidants will benefit from a "targeted" approach that ensures selective accumulation of antioxidants in mitochondria.

This conclusion is confirmed by the fact that in animal studies, in which scientists tried to slow down the aging of mammals with the help of classical (non-directional) antioxidants, there is no clear evidence of their effect on the maximum life expectancy (10,15,223,272,273). An alternative explanation is that most of the compounds studied may not have a powerful antioxidant effect in vivo, which actually applies to most human-taken antioxidant supplements (274,275). This does not negate the beneficial effects of antioxidants in the context of age-related diseases. Many pathologies developing with aging, including neurodegenerative diseases such as Alzheimer's and Parkinson's diseases, are associated with increased inflammatory processes and oxidative damage (15,35). If scientists are able to identify antioxidants suitable for use in vivo, that is, compounds that really have antioxidant properties in vivo and are able to penetrate into brain tissue, antioxidant therapy will have a beneficial effect on patients with similar diseases.

Specialists are working on the creation of targeted antioxidants acting on mitochondria by conjugation of a lipophilic cation with various antioxidant groups (276). One of these compounds, MitoQ, a ubiquinone derivative selectively acting on mitochondria, demonstrates selective accumulation in mitochondria, with an increase in concentration by several hundred times (276,277). Other strategies for targeted injection of compounds into mitochondria are also described (278). However, it should be remembered that in some cases, depending on environmental conditions, antioxidants can act as potential pro-oxidants [15]. This fact is a potential problem, especially concerning mitochondria, due to the close proximity of numerous electron donors and acceptors. In this context, it is interesting to note that in some systems MitoQ undergoes cyclic redox changes, acting as a pro-oxidant, stimulating superoxide production and cell death (279,280).

And, finally, the picture is further complicated by the fact that ROS are important signaling molecules whose functions, among other things, include participation in cell growth and proliferation, and a total decrease in ROS levels to non-physiologically low values causes (at least theoretically) the possibility of unintended side effects, for example, suppression cell proliferation (7,158).

Despite these problems, the ability to selectively direct the action of antioxidants, probes and modulators of the electron transfer chain on mitochondria opens up new opportunities in manipulating mitochondrial functions and ROS production in vivo, making this area a promising area of research.

5. Conclusions and prospectsGiven that mitochondrial degeneration and loss of mtDNA integrity are involved in the development of age-related degenerative processes, mitochondria are an attractive target for attempts to prevent or slow down the development of age-related pathologies and, possibly, even to modulate the basal rate of aging.

To date, most of these approaches, both theoretical and practical, are based on the assumption that reducing the level or absorption of mitochondrial ROS will protect the functional activity of mitochondria and the integrity of mtDNA. As discussed earlier, attempts to achieve this in practice consisted in the use of mitochondrial nutrients, as well as classical and some targeted antioxidants. A number of in vitro studies have demonstrated that mitochondrial nutrients, as well as targeted antioxidants, are able to prevent or even save the cell from the destructive effects of agents that adversely affect the mitochondria.

However, the data obtained in experiments on cell cultures are very easy to misinterpret (80, 281), since the cultivation of animal and human cells is a very complex process, fraught with the appearance of artifacts. For example, too high or too low efficiency of cell seeding on plastic has pronounced effects on the formation of a healthy population of the same cells and, accordingly, on the results of the experiment. Moreover, the use of various media for cell culture can lead to irrelevant positive or negative results due to the interaction between the culture medium and the test substance, which, in addition to other side effects, can unexpectedly cause the development of oxidative stress (80, 282). The appearance of artifacts is further aggravated by the use of concentrations exceeding physiological values of the studied compound (283). Moreover, the fact that cells are usually cultured under conditions of a strong excess of oxygen can cause disproportionately high ROS production (80).

According to the data obtained in vivo experiments, in animals receiving mitochondrial nutrients, there is often an increase in the bioenergetics of mitochondria and an improvement in the parameters of age-related weakening of functions. In contrast to this proven effectiveness of maintaining and partially restoring mitochondrial energy production, there is little evidence that available mitochondrial nutrients and antioxidants can prevent mtDNA damage and reduce mutational load in vivo (200,246). The results of further work with mitochondrial nutrients and targeted antioxidants should show whether these compounds are really effective mitochondrial antioxidants, i.e. whether any of them are capable of significantly reducing oxidative damage to mitochondria and, in particular, mtDNA in vivo. A more fundamental question is the following: is the suppression of oxidative damage to mtDNA maintained throughout life capable of reproducing the phenotype observed in mice of the IRU line, namely, is it possible to transfer this to a significant decrease in the mutational load of mtDNA in vivo and an increase in life expectancy. The study of this issue will provide important new information about the mechanisms underlying the age-related wear of mitochondria and the body as a whole. Apart from some facts (284), there is currently no convincing data confirming that regular intake of mitochondrial nutrients by a person prevents the development of age-related diseases and, moreover, slows down the aging process.

An alternative approach to modulating mitochondrial ROS production is to manipulate the potential of the mitochondrial membrane, for example, by weak activation of the dissociation mechanism (285,286). A moderate decrease in the mitochondrial membrane potential leads to a significant decrease in ROS penetration, which can be considered as a decrease in the load of mtDNA damage. However, ultimately, any approach aimed at reducing the load of ROS through their neutralization or "metabolic tuning" will be complicated by the intricacy of the feedback systems involved, a clear example of which is the dual role of ROS in signaling mechanisms. Any such attempt to change a limited number of desired parameters without triggering undesirable side effects or compensatory effects is actually a very difficult task in the context of an intricate system of closely interrelated processes. Moreover, the prevention of ROS-mediated damage will never be perfect, and the mechanisms of ROS-independent clonal expansion can lead to an increase in the level of any mtDNA damage to physiologically significant values.

As a possible solution, a direct impact on damage mechanisms is proposed instead of attempts to modulate the rate of their accumulation. Most likely, this approach will not cause harmful effects as a result of unforeseen cellular reactions, which are fraught with interference with metabolism [287]. Another class of strategies aimed at reducing the lifelong accumulation of oxidative damage is based on strengthening endogenous systems of antioxidant protection and detoxification, or activation of mechanisms of renewal and repair of damage. Such activation of endogenous protective reserves has been demonstrated, for example, as a reaction to hormesis (stimulation of any body system by external influences with an intensity insufficient for the manifestation of harmful factors) (described in (288)), as well as in connection with genetic interventions modulating the aging process (289,290). Pharmacological strategies can use these endogenous reserves either through direct interaction with the corresponding signaling mechanisms, or by copying the action of stressors without causing real harm. In the future, new approaches should appear based on a growing understanding of the mechanisms underlying the action of caloric restriction, clonal expansion of mtDNA mutations, mtDNA repair, mitogenesis and mitochondrial renewal.

See the list of references in a separate file.

Translated by Evgenia Ryabtseva
Portal "Eternal youth"

03.10.2008