Hormesis does not make sense except in the light of TOR-driven aging.

Weak stresses (including weak oxidative stress, cytostatic agents, heat shock, hypoxia, calorie restriction) may extend lifespan. Known as hormesis, this is the most controversial notion in gerontology. For one, it is believed that aging is caused by accumulation of molecular damage. If so, hormetic stresses (by causing damage) must shorten lifespan. To solve the paradox, it was suggested that, by activating repair, hormetic stresses eventually decrease damage. Similarly, Baron Munchausen escaped from a swamp by pulling himself up by his own hair. Instead, I discuss that aging is not caused by accumulation of molecular damage. Although molecular damage accumulates, organisms do not live long enough to age from this accumulation. Instead, aging is driven by overactivated signal-transduction pathways including the TOR (Target of Rapamycin) pathway. A diverse group of hormetic conditions can be divided into two groups. "Hormesis A" inhibits the TOR pathway. "Hormesis B" increases aging-tolerance, defined as the ability to survive catastrophic complications of aging. Hormesis A includes calorie restriction, resveratrol, rapamycin, p53-inducing agents and, in part, physical exercise, heat shock and hypoxia. Hormesis B includes ischemic preconditioning and, in part, physical exercise, heat shock, hypoxia and medical interventions.


ROS
In some studies, an increased production of reactive oxygen species (ROS) correlated with extended life span in different species. To explain such paradoxical results it was suggested that an increased ROS in turn increases resistance to ROS, thus extending life span [ 1,12,14]. It was suggested that ROS leads to a condition of mild stress, which in turn enhances vitality [12]. In C. elegans reduced glucose availability promoted formation of ROS, induces catalase activity, and increased oxidative stress resistance, cumulating in extension of life span [15].

Two noticeable problems
First, it is paradoxical to decrease damage by causing damage ( Figure 1B). There is no similar example in medicine. If one wishes to prevent stroke due to high blood pressure, one needs to decrease blood pressure not to increase it. Examples are endless including weight control to prevent diabetes and quitting smoking to prevent lung cancer. No one will advocate "mild and repeated" smoking to prevent lung cancer even though it might activate the defenses. The simple reason is that DNA damage is actually involved in cancer initiation. But even cancer-promoting damage is not random: mutations activate growth-promoting pathways including PI3K/mTOR, the most universal alteration in cancer [42][43][44][45]. And cancer-initiating damage does not cause cellular decline (in contrast, cancer cells are very www.impactaging.com robust and hyper-functional), is not sufficient to cause cancer, requiring rounds of cell replication, selection [46,47] and organismal aging [48]. By slowing down aging, CR and rapamycin delay cancer (without affecting mutations). The notion that aging promotes cancer is beyond the topic of this article and cannot be discussed here. The point here is that since DNA damage contributes to cancer, no one will suggest hormetic smoking or radioactivity (at any doses) to delay cancer. In analogy, if damaging hormesis may delay aging, aging cannot be possibly be caused by damage.
Second problem is the suggestion that mild hormetic stresses protect against severe stresses. What are these severe stresses? Even according to the conventional view, aging is not caused by accidental injures that are stronger than hormetic damage. It is caused by 'everyday' ROS and other background stresses. Let's ask a straightforward question. Are hormetic stresses stronger or weaker than those that cause aging? And there is no answer. If damage that drives aging does not sufficiently induce protective response, then hormesis is stronger than aging-causing stresses. Then why is the purpose of hormesis to protect from stronger stresses? This question will be easily answered in the light of TOR-driven aging.

Solution: a new view on aging
If aging is driven by damage, then damage must accelerate aging. If hormesis induces damage and slows down aging, then aging is not driven by damage. So a straightforward explanation is that aging is not caused by accumulation of molecular damage [36]. It was predicted "that five years from now, current opponents will take the TOR-centric model for granted, which then will become new dogma (ironically)" [35].
While rejecting ROS-driven aging, scientists still do not dare to reject the view that aging is a decline due to accumulation of random damage. Yet data exclude not only ROS but also damage as a cause of aging. For example, in C elegans, CR did not decrease the accumulation of spontaneous mutations with age but nevertheless extended life span [54].
Yes, perhaps, molecular damage accumulates but organisms do not live long enough to age from this accumulation. Even humans, long-living organisms, do not die from a decline due to such an accumulation. (And of course worm that lives just 5 days [67] cannot possibly accumulate deadly levels of molecular damage). Instead any human being has died from agerelated diseases, which are caused by active cellular processes initiated by hyper-activation of signaling pathways including mTOR [36,68]. The pathogenesis of diseases is well known. In contrast, a mysterious cellular decline due to accumulation of molecular damage is the fiction of gerontology, unknown in medical science.
Instead of trying to adopt the phenomenon of hormesis to the view on aging as accumulation of random molecular damage, we will reconsider the view on aging. hormones activate the nutrient-sensing TOR (Target of Rapamycin) pathway, which promotes growth and then aging, causing age-related diseases. In turn, diseases cause non-random organ damage and death. Hormesis type A inhibits TOR thus slowing down aging. Hormesis type B increases aging-tolerance and tolerance to complications of age-related diseases.
The TOR pathway drives cellular mass growth. In proliferating cells, cellular growth is balanced by cell division [80]. In quiescent (resting) cells, growthpromoting pathways drive senescence [80,81]. When the cell cycle is blocked but mTOR is still active, it causes hypertrophic, hyperactive, hyper-functional (for example, hyper-secretory) phenotype, with compensatory resistance to signals such as insulin and growth factors [82,83] and compensatory lysosomal activation. In other words, mTOR converted quiescence into senescence, the process that was named gerogenic conversion or geroconversion or gerogenesis [83].

Longevity: (a) slow aging and (b) aging-tolerance
Life span can be extended by either (a) slowing down aging and (b) by increasing aging tolerance, defined as the ability to survive complications (catastrophes) of aging [36].
a. Obviously, inhibition of aging should extend life span and delay age-related diseases. For example, calorie restriction (CR) slows aging. CR delays age-related diseases such as cancer and atherosclerosis, thus extending life span. In other words, inhibition of mTOR-driven aging delays catastrophic complications of aging: namely, complications of age-related diseases such as stroke, myocardial infarction ( Figure 2). These non-random catastrophes actually cause death.
b. But inhibition of aging (and delaying diseases) is not the only way to extend life span. The second way is to increase aging tolerance, which allows an organism to survive catastrophes caused by age-related diseases.

Why organisms age
The existence of aging is well understood from the evolutionary perspective and was discussed in detail. Roughly speaking, in the wild, organisms do not live long enough to experience aging. Therefore forces of natural selection against aging are weak. Only in protected environment (humans, domestic and laboratory animals) die from aging.
Why organisms have low aging-tolerance From the evolutionary perspective, organisms do not tolerate aging for the same reason why they age in the first place. In the wild, organisms do not live long enough to experience aging and therefore organisms are not naturally shaped to experience complications of aging. Organisms are not selected by nature for agingtolerance. For example, parts of myocardium depend on a single coronary artery. The occlusion of a coronary artery causes life-threatening ischemia. Collateral arteries would prevent ischemia. Natural selection would favor such anatomical re-design, if it will extend reproductive life span. If humans were routinely reproducing after the age of 70, then variations with additional branches of coronary arteries would be selected.
Thus myocardial ischemia, due to artery occlusion is one of the most common causes of death. The occlusion may result from thrombosis of atherosclerotic coronary arteries. But if the ischemic zone would receive blood www.impactaging.com supply from a collaterally artery, the organism would survive the catastrophe. Thus, anatomical modifications of a myocardial blood supply would increase agingtolerance without affecting aging itself. Noteworthy, this is how coronary stents extend lifespan without affecting aging. Most medical treatments increase aging tolerance, thus extending an average life span despite chronic age-related diseases. In contrast, pharmacological suppression of aging would increase healthy lifespan by postponing diseases [108].

Two types of hormesis: (a) slowing down aging and (b) increasing aging-tolerance
Hormetic stresses include two groups of agents that (a) slow down aging by inhibiting the TOR pathway and (b) increase aging tolerance, without affecting the aging process ( Figure 2). We will call them hormesis A and hormesis B. Examples of hormesis A are calorie restricttion, rapamycin, resveratrol and p53-inducing agents. Examples of hormesis B are adaptive preconditioning to ischemia and coronary bypass. Heat shock, hypoxia and physical exercise belong to both groups.

Chemical hormesis
Plants, microorganisms and sea animals produce toxic agents that inhibit or damage microtubules, DNA and many other vital targets. Due to their toxicity, some of them are used as anti-cancer drugs, although nature did not created them for that purpose. Nature of course created these poisons to hurt predators and competitors [125,126]. Similarly, rapamycin is an antifungal antibiotic produced by bacteria. TOR stimulates growth in response to nutrients. Therefore, soil bacteria produce rapamycin to inhibit yeast growth. While inhibiting TOR-dependent growth, rapamycin slows down TORdependent aging in older yeast [93,94]. Given that cancer (like aging) is "a form of growth", the mTOR pathway is activated in cancer. And, although not created for that purpose by nature, inhibitors of mTOR are used as anticancer agents [42,43,127,128]. I wish to emphasize again that bacteria produce rapamycin neither as a medicine for longevity nor as an anticancer drug, but as an antifungal antibiotic. Simply the same signaling pathways that are involved in growth also are involved in cancer and aging [88]. Growth suppressants may suppress aging because aging is a continuation of growth, driven by the same TOR/S6K pathway [129]. To extend lifespan, they either should inhibit the TOR pathway or increase aging tolerance (Figure 2).

Rapamycin
Rapamycin extends life span in yeast, drosophila and mice [94,98,99,130,101,103,130]. It is indicated for almost all age-related diseases [31, 48, 68, 131]. Rapamycin is not toxic for normal cells at concentrations that exceed therapeutic levels 1000 fold [132], [133]. There are no side effects of high dose rapamycin in healthy volunteers [123,134]. Rapamycin has been used in children [135] and in pregnant women [136]. Despite common misconception that rapamycin is an immunosuppressant, rapamycin improves immunity in mice when used appropriately [137][138][139].
As an anti-aging modality, rapamycin could be used in doses and schedules that do not cause side effects [132].

Physical exercise
As an example of hormesis A, chronic increase in physical activity inhibits mTOR/S6K1 in rat skeletal muscle [175]. Physical activity can also increase agingtolerance, acting as hormesis B.
Heat shock The TOR pathway stimulates Cap-dependent protein synthesis. Elevated temperature inhibits cap-dependent protein synthesis. Thus, heat shock blocks TORstimulated protein synthesis. For example, heat shock protein Hsp27 inhibits translation during heat shock by binding eIF4G [176]. Therefore, heat shock acts as "hormesis A" by imitating TOR inhibition. The small heat-shock proteins also delay the onset of polyglutamine-expansion protein aggregation, suggesting that these proteins couple the normal aging process to this type of age-related disease [177]. Also, HSPs and chaperones can increase resistance or tolerance to catastrophic complications of aging, defining them also as hormesis B.

Hormesis B
Hormesis B extends life span by increasing aging tolerance. Mild stresses prepare organism to catastrophes caused by diseases of aging. Examples of catastrophes include stroke and myocardial ischemia. The occlusion of a cerebral artery for 60 min (injurious ischemia) damages the brain. The occlusion of the same cerebral artery for 15 min (preconditioning) protects from the damage caused by injurious ischemia [208]. Similarly, severe myocardial ischemia causes irreversible injury. Mild ischemia protects the heart from severe ischemia. Similarly, by inducing HSPs, heat shock may protect the myocardium from severe ischemia. Repeated, transient ischemic episodes or heat shock augment the ischemic tolerance of affected myocardium. Upregulation of immediate early genes and heat shock genes plays an important role in myocardial adaptation to acute ischemic stress [209]. Also, hormetic stresses can cause growth of collateral arteries. This coronary collateral function can preserve ischemic myocardium [210].
The cardioprotection against myocardial injury by regular exercise may include the development of collateral coronary arteries and induction of myocardial heat shock proteins [211].
Similarly, coronary bypass protected heart from ischemia. Although we do not call such medical procedures hormesis, there is no strict borderline between them and hormesis B. For example, reconditioning, hypoxia, and stresses may "train" cardiomyocutes to survive acute episode of coronary thrombosis. Also, it develops small blood vessels that could compensate for the occlusion of main artery. Now we can solve the second problem of hormesis (see "two noticeable problems"). The answer is: hormetic stresses protect from stronger stresses. But these stronger stresses are not those that cause aging (aging is not caused by any stresses). These are complications of aging or age-related diseases. We call them catastrophes. Hormesis B protects from lethal catastrophes.

Conclusion
The hypothesis that aging is NOT driven by accumulation of random damage allows us to explain hormesis. Type A hormesis antagonizes the TOR pathway ( Figure 2). Hormesis B causes stresses including damaging stresses. Since aging is not caused by damage, this does not contribute to aging but instead may cause aging-tolerance, thus protecting organisms from lethal consequences of aging-induced catastrophes.

CONFLICT OF INTERESTS STATEMENT
The author of this manuscript has no conflict of interest to declare.