1. The Two Ways by Which ROS Cause Damage
There are two ways ROS cause cellular distress: dysregulation in redox signalling and oxidative damage. Aberrant redox signalling occurs when signals are not produced or are generated at the incorrect time/place or with the wrong intensity
[1]. presented a few examples of the former above. Damage is provoked by too much signalling or a lack thereof. Ablation of mitochondrial redox signalling in mouse astrocytes causes metabolic and behavioural alterations
[2]. On the other hand, overexpression of the ATPase inhibitory factor 1 (IF1) in mouse neurons increases mtROS generation, boosting the learning skills of mice
[3]. Conversely, knocking out IF1 impairs memorization and performance in different cognitive tests. The other way ROS cause harm is through oxidative damage. High concentrations of ROS trigger oxidative stress
[4]. This happens when the concentration of oxidants exceeds that of antioxidants, leading to oxidative damage
[5]. For instance, animals fed oxidants such as hydrogen peroxide or paraquat exemplify the former scenario and die as a result
[6]. Likewise, in mice, SOD2 knockout is lethal during embryonic development
[7]. Additionally, ischemia–reperfusion provides a more physiological example of oxidative stress culminating in cell death
[8]. ROS-RET triggered by succinate accumulation during ischemia causes oxidative damage, subsequently triggering tissue damage
[9]. Accordingly, preventing RET attenuates oxidative stress and preserves the function of the affected tissue
[9][10][11][12]. Some cancers are caused by mutations in genes encoding CII subunits that increase mtROS levels
[13][14]. Recent studies have revealed that ROS-RET is upregulated in cancer stem cells with aberrant NOTCH signalling, a critical factor in the growth of tumour cells in mouse and fly brain models.
[15]. Tumour necrosis factor (TNF) is critical to protect cells against mycobacteria
[16]. However, excessive TNF signalling triggers ROS-RET, which causes necrosis in macrophages infected by mycobacteria
[17]. Accordingly, metformin inhibits both ROS-RET and macrophage death. These findings open the possibility of using metformin in combination with antituberculosis drugs to prevent the harmful effects of the latter.
Several diseases are associated with excessive CIII ROS. As previously noted, unchecked CIII ROS generation can lead to lung fibrosis
[18]. Similarly, pollution causes thrombosis in the lungs by triggering CIII ROS
[19][20]. Like CI ROS, CIII ROS play a crucial role in the proliferation and metastasis of specific tumour types, such as those regulated by KRAS signalling
[21][22]. Again, it remains unclear whether CI/CIII ROS are responsible for the proliferation and metastasis of specific tumour cells or if the divergent outcomes are attributable to methodological differences in the execution of the experiments. Finally, some unidentified in vivo ROS generators may be crucial for other afflictions, since at least 11 sites produce superoxide in isolated mitochondria
[23].
2. The Role of mtROS in Ageing
Damage caused by endogenously produced mitochondrial free radicals was initially proposed as the cause of ageing by Denham Harman
[24][25]. In his famous “Mitochondrial Free Radical Theory of Ageing” (MFRTA), Harman proposed that mtROS produced as byproducts of metabolism cause the accumulation of oxidative damage responsible for ageing and age-related diseases. Over the years, the MFRTA has been supported by two types of evidence
[26]: (1) observational studies describing the age-related accumulation of oxidative damage
[27][28] and (2) correlative studies showing that mitochondria from long-lived animals produce fewer ROS
[29][30]. While the theory was highly prevalent at the turn of the 21st century, it has since experienced a significant decline in popularity and is now considerably less popular than it once was
[31][32]. The demise of MFRTA began when its forecasts were subjected to experimental verification. The first prediction states that increasing levels of antioxidants will extend lifespan and that, conversely, decreasing their concentration will shorten survival. Supplementing non-enzymatic antioxidants fails to increase longevity in most animal models
[29]. Similarly, overexpression of enzymatic antioxidants reduces oxidative stress but does not extend the lifespan of mice
[33][34]. An important exception is the targeted expression of catalase to mitochondria, significantly improving the median and maximum survival of mice
[35]. However, the same strategy in fruit flies shortens lifespan
[36] and causes deleterious alterations in mouse behaviour
[2]. Unexpectedly, genetic depletion of most enzymatic antioxidants elevates oxidative damage without impacting mouse survival
[37][38][39]. Nevertheless, there are some exceptions, such as knocking out SOD1 in mice or knocking down peroxiredoxins 3 or 5 in fruit flies
[40][41]. However, these three enzymes are instrumental in redox signalling, and therefore, it is impossible to conclude whether the animals die because of the accumulation of oxidative damage, interruption in redox signalling or both.
The second prediction of MFRTA argues that decreasing mtROS levels must extend lifespan and that increasing mtROS generation must shorten survival. In
Caenorhabditis elegans, it is possible to increase mtROS production by using inhibitors of the ETC or knocking down subunits of the respiratory complexes. Both types of interruptions in electron flow increase longevity
[42]. To the best of the knowledge, ETC inhibitors have not been reported to extend the lifespan of fruit flies, although inhibition of CV does (see below). Surprisingly, feeding rotenone to killifish extends longevity by 15%
[43]. In flies, the knockdown of CI, CIII, CIV and CV extends survival
[44]. The most consistent results are obtained by knocking down CI subunits, where lifespan is prolonged by an mtROS-dependant mechanism
[45]. However, a recent report showed a reduction in mtROS associated with CI depletion in
Drosophila flight muscle and extended lifespan
[46]. Extending the lifespan of Drosophila is achieved by expressing the internal alternative NADH dehydrogenase 1 (NDI1) from Saccharomyces cerevisiae
[47][48]. NDI1 promotes longevity, stimulating the generation of mtROS via RET. Similarly, several reports show that chemical or genetic inhibition of ATP synthesis by blocking CV increases mtROS levels and the survival of fruit flies and worms
[49][50][51][52]. In summary, all the former results do not support the role of mtROS limiting lifespan per se. Conversely, they indicate that adequate mitochondrial redox signalling is essential for a healthy lifespan.
3. Accumulation of Damaged Mitochondria during Ageing Causes an Interruption in Mitochondrial Redox Signalling
Mitochondrial redox signalling is a sophisticated communication system that becomes a source of damage during ageing. Controlled and regulated occurrence of ROS-RET and CIII ROS production requires specific conditions
[53]. For example, ROS-RET requires the appropriate levels of reduction of the CoQ pool and a sufficiently high proton motive force to provide the necessary energy for RET
[54]. In the ageing process, a significant concern arises due to the difficulty experienced by old mitochondria in maintaining a high mitochondrial membrane potential
[55]. Thus, the age-related decline in the NAD+-to-NADH ratio can potentially modify how CI generates ROS
[56]. Furthermore, the depletion of NAD+ has the potential to impact the activity of SOD2, as it necessitates deacetylation by SIRT3 for activation, and SIRT3, in turn, is reliant on NAD+
[57]. This may also impede the transformation of superoxide signals into more stable hydrogen peroxide signals, thereby modifying the cellular signalling process. Initiating the CIII signalling process requires an influx of electrons, while a proper efflux is necessary to conclude the process
[58]. Therefore, any process that alters the entry or exit of electrons, such as those associated with ageing, as described below, also modifies CIII redox signalling. In conclusion, it is unsurprising that alterations in the mitochondrial electron flow caused by the disruption in the OXPHOS machinery profoundly affect redox signalling
[59][60]. Accordingly, one of the hallmarks of ageing is the accumulation of dysfunctional mitochondria that generate fewer ATP and higher levels of mtROS
[61].
Reduced mitochondrial oxygen consumption is a common characteristic of ageing found in multiple animal species
[62][63][64][65]. The referred decrease is related to problems with both the entry (due to problems with CI) and the exit of electrons (caused by issues with CIV)
[62][66][67][68][69]. Age-related reduction in the levels of CI has been reported in the human cortex
[70], naked mole rat skeletal muscle
[71]) and fruit fly mitochondria
[72]. However, other studies inform of a preferential decrease in CIV activity correlated with high levels of mtROS in flies and mouse adipocytes
[68][73]. Reducing the entry of electrons can impede the generation of ROS signals
[74], while blocking their exit results in the cessation of ROS signalling due to the ongoing production of uncontrolled ROS
[75]. Interestingly, the opposite also occurs, i.e., increased mitochondrial oxygen consumption that disrupts mitochondrial redox signalling during ageing. The former happens during cellular senescence when an expansion in mitochondrial mass increases both cellular oxygen consumption and levels of ROS
[76]. In senescence cells, mtROS reprogram the cell, causing DNA damage and releasing proinflammatory cytokines such as IL-6. Accordingly, strategies that reduce either the number of mitochondria or the amount of ROS decrease DNA damage and the senescence-associated secretory phenotype
[77]. All the examples presented above indicate that age-related alterations in mitochondrial respiration disrupt physiological redox signalling. Notably, modifications in redox signalling can arise through different mechanisms, which can entail increments and decrements in oxygen consumption. A significant challenge for future research is to conduct tissue-specific analysis of mitochondrial function, investigate how these alterations affect redox signalling and explore the potential consequences for the cellular processes regulated by mtROS.
Numerous examples in the literature demonstrate age-related changes in mitochondrial redox signalling, leading to severe disruptions in cellular homeostasis. In the fly brain and muscle, there is a significant increase in the levels of mtROS during ageing
[46][62][63][75]. However, old and young brain mitochondria produce ROS in different ways. Young brain mitochondria have low levels of mtROS that are increased in response to specific types of stress. In contrast, old mitochondria continuously produce high levels of ROS and are less responsive to external stimuli
[75][78]. In young flies, blocking the exit of electrons by knocking down subunits of CIV increases mtROS and causes interruptions in redox signalling. However, interventions targeting the entry of electrons (by knocking down CI subunits) prevent the generation of mtROS signals without increasing the amount of ROS
[46][75]. A separate study reported an increase in mtROS in the flight muscle of Drosophila, which is crucial for extending lifespan resulting from CI depletion
[45]. Along the same line, the knockout of the CIV assembly factor COX15 activates an ROS-dependent adaptation program in mouse muscle
[60]. Interruption of this program by expressing AOX damages the antistress response and shortens survival. Similarly, in Drosophila flies, blocking mitochondrial redox signalling by expressing a mitochondrially targeted catalase damages the antistress response
[74]. Conversely, boosting it by overexpressing SOD2 extends survival under stress and non-stress conditions
[74][79]. Similarly, ectopic catalase expression in the mitochondria of astrocytes interrupts normal redox signalling, disrupting brain activity
[2]. The former data indicate that young mitochondria produce low basal ROS levels and only increase ROS production in response to specific stimuli. Once the stress situation is over, mtROS levels return to normal. However, old mitochondria continually produce ROS and are unresponsive to stress (
Figure 1). A recent paper showed that old fly mitochondria generate more ROS, triggering RET in CI
[46]. Accordingly, restricting the inhibition of ROS-RET to flies older than 20 days using a novel, specific CI blocker (6-chloro-3-(2,4-dichloro-5-methoxyphenyl)-2-mecapto-7-methoxyquinazolin-4(3H)-one) leads to an extension of longevity. However, other manuscripts reported that ROS-RET extends fly lifespan
[62], is instrumental for stress adaptation
[74] and does not occur in the brain of flies older than 25 days
[75]. Therefore, although different data support the loss of redox signalling during ageing, more studies are required to clarify how the former occurs mechanistically. An essential aspect of elucidating this mechanism is whether identical mechanisms generate ROS in both young and old mitochondria and whether there is a shift in the distribution of low- and highly reactive oxygen species in older individuals. Increased concentrations of hydroxyl radicals, peroxynitrite and other highly reactive free radicals may account for the accumulation of oxidative damage observed in aged individuals. Such an increase in the levels of aggressive ROS could be attributed to the liberation of iron and other metals from the catalytic centres of mitochondrial enzymes due to the amplified intensity of ROS signalling
[80].
Figure 1. Mitochondrial redox signalling is altered during ageing, with young mitochondria producing ROS only in response to specific stimuli, thereby maintaining low total ROS levels. Conversely, ageing mitochondria continually produce high levels of ROS and are unresponsive to external stimuli, resulting in oxidative stress and damage. Antioxidants alone are not effective in increasing lifespan, as, while they can reduce oxidative damage, they cannot restore redox signalling. Effective antiageing therapies require not only a reduction in oxidative damage but also restoration of mitochondrial ROS (mtROS) signalling. The figure illustrates the contrasting differences between young and ageing mitochondria, highlighting the importance of restoring mtROS signalling for healthy ageing.
The proposed model presented in the previous figure (
Figure 1) has not been experimentally verified, and there is indirect supporting and opposing evidence. For example, disruption of redox signalling explains why boosting antioxidants fails to positively impact lifespan in animal models
[81] or human health in clinical trials
[82]. Antioxidants neutralize oxidative damage but cannot restore mitochondrial redox signalling. The transcriptome of old flies presents many similarities with the transcriptome of flies exposed to hyperoxia
[83], indicating an increase in the oxidation state of fly cells during ageing. This observation supports a model in which the interruption of redox signalling induces oxidative stress. However, the redoxome of neither flies nor mice consistently shows an increase in the number of oxidized cysteines in aged individuals.
[6][84]. The former is surprising, considering the age-related decrease in reduced glutathione in insects and rodents
[85][86]. On the one hand, the lack of consistent age-related modifications in the redoxome does not support a generalized interruption of redox signalling caused by mitochondrial dysfunction. On the other hand, specific tissues in mice exhibit age-related modifications in highly reactive cysteines in proteins associated with age-related diseases
[84]. These specific modifications could account for the changes observed in the transcriptome of
Drosophila without significant alterations in the redoxome. The consequences of losing redox signalling at the onset of ageing and age-related diseases deserve further investigation. Future studies must consider ROS as signals, not as simple metabolic byproducts. The majority of ROS-regulated processes discussed earlier have not been investigated in the context of ageing, a process during which mitochondrial function is notably impaired. This is not surprising, given that redox signalling remains a developing area of research, with limited studies dedicated to investigating its role in ageing, particularly in mammals. For example, an appropriate generation of mtROS is required to trigger sleep in fruit flies
[87]. Accordingly, overexpression of catalase or AOX reduces sleep in transgenic flies, likely due to interruptions in redox signalling. However, it remains to be seen whether the age-related increase in mtROS generation is involved in disrupting sleep patterns in old animals
[88]. To employ manipulations in redox signalling as a therapeutic strategy, it is crucial to first determine how the interruption of redox signalling affects ROS-regulated processes in both young and aged individuals. Second, it is essential to understand the underlying mechanisms driving changes in redox signalling and whether restoring redox signalling could potentially promote healthy ageing in humans.