Caloric Restriction

Caloric Restriction (CR), or calorie restriction, is by far the anti-aging intervention that, with our current knowledge, holds the greatest potential to delay human aging. In this essay, I review the data supporting this hypothesis, and the case for CR as a potential anti-aging therapy, as well as evidence against it. I also briefly discuss the most popular products trying to emulate the beneficial effects of CR, in particular resveratrol.


CR in Animal Models
CR in Humans
Mechanisms of CR
Potential CR Mimetics and Resveratrol

Keywords: ageing, anti-aging diet, dietary restriction, eternal youth, old age, rejuvenation

The only known method that might be able to delay the human aging process as defined in is CR. CR, as the name implies, consists of diminishing caloric-intake while maintaining a normal diet regarding other nutrients such as vitamins and minerals. There are different implementations and variants of CR such as protein restriction, food restriction, and intermittent fasting. Since reducing energy-intake is the best-studied intervention, the focus of this essay is on caloric restriction.

CR in Animal Models

Several studies in animals have demonstrated the beneficial effects of CR, for example in lowering cholesterol and blood pressure, and its potential to delay the rate of aging. In particular, studies in mice and rats support the idea that CR delays the aging process and certain age-related diseases like type 2 diabetes and cancer. Life-extension in mice by CR can be higher than 40% and even greater increases have been reported in non-mammalian models (Weindruch et al., 1986; Weindruch and Walford, 1988; Fontana et al., 2010). CR can delay the entire aging phenotype, reducing the frequency of age-related diseases and decelerating aging. CR animals look younger and some tests suggest that they are physiologically younger than their age-matched controls. Demographic analyses also suggest that CR delays aging, at least in rats, by delaying the rate at which mortality increases with age (de Magalhaes et al., 2005). Though some results in rats suggest there may be a threshold, an age after which CR no longer significantly extends lifespan (Lipman et al., 1995), most results in fruit flies and mice suggest that CR quickly exerts its beneficial effects, even in old animals (Spindler, 2005).

In addition to rodents, CR has been shown to extend lifespan in other animals. For example, CR appears to extend lifespan and delay signs of aging in some dog breeds (Kealy et al., 2002). There are also ongoing studies in rhesus monkeys. Thus far, it appears that CR monkeys are shorter, smaller, have a lower body temperature, lower insulin and glucose levels, a normal intelligence, a decreased sexual appetite, and a lower mortality rate (Weindruch and Walford, 1988; Duffy et al., 1990; Ramsey et al., 2000; Bodkin et al., 2003). In one group studied--so far--for 25 years, CR monkeys lived on average 32 years while controls lived 25 years (Bodkin et al., 2003), which suggests a potential life-extension slightly over 20% and hence lower than that witnessed in rodents but still significant. A more recent study suggests that CR in rhesus monkeys delays mortality and onset of age-related diseases (Colman et al., 2009). Another study in rhesus monkeys conducted at a different facility reported health benefits from CR, e.g., in reducing cancer incidence, but no effects on lifespan (Mattison et al., 2012). One possible reason for the discrepancy between these two studies is that control animals in the latter were fed a healthier, more controlled diet. Both studies are still ongoing as some of the animals are still alive, but a strong message from the studies in monkeys is that CR may have health benefits but will likely not significantly extend lifespan in individuals already on a moderately healthy diet.

Although CR extends lifespan in several species, it does not extend lifespan in all species. For example, CR does not appear to extend the lifespan of the housefly (Cooper et al., 2004). Moreover, some animals undergo torpor under CR and thus studying CR's long-term effects in such species is probably impossible. Certainly, it may be argued that perhaps the right conditions for studying CR in these species have not been determined yet, but the fact remains that CR does not extend lifespan or delay aging in some species.

It should also be noted that while CR can delay aging in mice, that is not true for all mouse strains (Forster et al., 2003). In particular, CR does not appear to extend average lifespan in wild-derived mice, even though it somewhat protects against cancer as observed in other mouse strains (Harper et al., 2006). This finding is relevant because, as detailed elsewhere, laboratory mice were adapted to laboratory conditions and thus may not be representative of the species. As such, the inability of CR to extend lifespan of wild-derived mice may suggest that CR is in part an artifact of breeding animals specifically for laboratory studies. Even in wild-derived mice, however, CR extends maximum lifespan (Harper et al., 2006). Because wild-derived mice are genetically heterogeneous, these results suggest that CR may be beneficial to some individuals--hence extending maximum lifespan--but detrimental to others--thus explaining the lack of increase in average lifespan.

CR in Humans

In humans, there are no conclusive studies but some results suggest that CR may be beneficial, at least in some groups of people. One study found that CR has a protective effect against atherosclerosis in people (Fontana et al., 2004). Another study reported beneficial effects on cardiac function (Meyer et al., 2006) and yet another study found some benefits of CR in reducing weight and adiposity, though the benefits were similar to those obtained by exercising (Racette et al., 2006). CR may also improve memory in the elderly (Witte et al., 2009). In addition, CR appears to have beneficial effects on some biomarkers of longevity in overweight individuals (Heilbronn et al., 2006), though the observation that reducing calories is beneficial to overweight patients is not surprising. Finally, one 2-year study in nonobese humans found evidence of potential benefits in survival and disease risk factors (Ravussin et al., 2015).

As I hope everyone is aware, a high-calorie diet is unhealthy for the vast majority of people, increasing the probability of developing age-related diseases like atherosclerosis and type 2 diabetes. Likewise, being overweight is associated with reduced longevity (Whitlock et al., 2009). So maybe CR, in part by reducing body fat which tends to increase with aging (Das et al., 2004), diminishes the chance of developing certain age-related diseases and increases average lifespan but does not truly delay aging in humans. Alternatively, perhaps CR does not delay aging but instead a ad libitum diet--i.e., one in which you eat everything you want, which is the diet used for the control rodents and monkeys used in CR studies--is harmful.

One study compared the high-calorie diet of Japanese Sumo wrestlers, who live about 56 years, with that of male Okinawans, who have a low-calorie diet (though are not calorie-restricted) and live about 77 years. Clearly, a high-calorie diet is unhealthy and decreases lifespan. By extrapolating from the male Okinawans to CR, the study estimated that the life-extension provided by CR for individuals already with a balanced nutrition would be less than 10% (Phelan and Rose, 2005). Others, too, have argued that CR will not significantly extend human lifespan in people who already live a healthy life (de Grey, 2005a). Overall, at the present time, there is no way of telling if CR works in humans or not and, if so, to what extent. It is arguable, however, that individuals with a healthy lifestyle will not greatly benefit from CR. In fact it has been suggested that while CR may have beneficial effects in adults this may not extend to lean people (Fontana and Klein, 2007).

If it is possible that CR works in humans, then why is not CR a widespread treatment? The answer are side-effects (Hopkin, 2003). Personally I have never tried CR, but the following have been documented (Dirks and Leeuwenburgh, 2006), and told to me by individuals undergoing CR. First of all, there is the mental stress for being hungry all the time, which can lead to depression. Even in food restricted rats, depression- and anxiety-like behaviors have been observed (Jahng et al., 2007). Then, if you undergo CR, exercise may become impossible, though studies in mice indicate that exercise while on CR has only mild negative effects on longevity (Holloszy and Schechtman, 1991; Holloszy, 1997). CR also makes people feel less energetic, less alive. And finally there are the sexual problems: diminished libido is a common side-effect in people under CR and infertility is also possible. Please note, of course, that these problems need not occur all at the same time. Still, one can say CR might add some years to life but it will not be a life worth living. Lastly, CR has been reported in mice to hinder their ability to fight infection (Sun et al., 2001; Gardner, 2005), and some evidence suggests that in patients with amyotrophic lateral sclerosis CR accelerates the onset of the disease (Mattson et al., 2007).

Mechanisms of CR

Although CR was discovered in the first half of last century (McCay et al., 1935), its mechanisms are largely unknown. Many hypotheses have been proposed and it is impossible to review them all. Briefly, some argue that the diminished energy intake forces an optimization of the metabolism. Since CR also delays development in mice, others say it slows down the entire genetic program, indirectly affecting aging (de Magalhaes and Church, 2005). CR's effects on several forms of cellular damage have been reported but the results are inconclusive as far as the underlying mechanism is concerned. Interestingly, there have been experiments in mice that seem to mimic CR by disrupting certain hormonal levels. Basically, by diminishing certain hormones or their receptors scientists observed changes in animals similar to those observed under CR, as described elsewhere in more detail. Therefore, hormonal changes may play a role in CR. Nonetheless, many other genes have been related to CR in model organisms (Figure 1), including sirtuins which are discussed below. At present, it is safe to say that we do not know the mechanisms by which CR extends lifespan.

Figure 1: Overview of CR-associated signaling and some of its key players. On one hand, endocrine changes, including those controlled by the brain, are crucial for CR-mediated effects. In turn, systemic changes trigger signaling cascades that involve several mechanisms, pathways and genes. Homologs of genes directly linked to CR life-extending effects in model organisms, from our GenDR database, are highlighted with a blue halo. (Adapted from de Magalhaes et al., 2012).

Potential CR Mimetics and Resveratrol

Even though we do know the mechanisms of CR or if CR delays human aging, several scientists and companies are trying to develop products that mimic the effects of CR without the side-effects. It has been argued that maybe we will never know whether CR delays aging in humans or not, but we can know if CR mimetics improve health which would already be a significant breakthrough (Ingram et al., 2006). Many genes have been associated with the life-extending effects of CR (Figure 1), and our lab has developed GenDR, the first database of such CR-related genes. Moreover, in the case of many drugs we do not need to know exactly how they work for them to work. Therefore, this is a growing area of research and also because obesity is a major health problem and since CR lowers body fat there is an overlap between therapies based on CR and therapies aimed at obesity and the metabolic syndrome (reviewed in de Magalhaes et al., 2012). In fact, some companies are trying to develop CR mimetics by developing products that curb appetite.

Potential CR mimetics include inhibitors of glycolysis (e.g., 2-deoxy-D-glucose) and an anti-diabetic drug called metformin (de Magalhaes et al., 2012). Metformin treatment has been shown to increase the average lifespan of tumor-prone mice by 8% (Anisimov et al., 2005). In normal mice, metformin increases lifespan by a modest 6% (Martin-Montalvo et al., 2013). Nonetheless, one study found that a change in lifestyle is more effective than metformin in preventing diabetes (Knowler et al., 2002), which again argues that no drug or supplement is a substitute for a balanced diet and overall healthy lifestyle, as discussed elsewhere.

Probably, the most well-known molecules being tested as CR mimetics are a group called sirtuin-activating compounds (STACs) that include plant polyphenols such as resveratrol and fisetin. As mentioned elsewhere, resveratrol has antioxidant properties and it also appears to be a cancer chemopreventive agent (Jang et al., 1997). While resveratrol has been studied for some time now and might have beneficial effects, such as preventing cardiovascular disease (Mizutani et al., 2000), what mostly brought resveratrol into the spotlight of anti-aging therapies was its ability to extend lifespan in animal models.

Resveratrol is produced by several plants and is found in relatively high concentrations in red wine. While resveratrol has been touted by some as the reason behind the beneficial effects of red wine (Kopp, 1998), other components of red wine have been shown to have cardioprotective properties (Kiviniemi et al., 2007), so resveratrol may not be whole story. Resveratrol structure

The rationale behind STACs is based on studies in yeast, flies, and worms showing that increasing the levels of sirtuins, which are genes involved in silencing other genes, increases lifespan and may be necessary for CR (Baur and Sinclair, 2006). Initially, sirtuins, and one sirtuin in particular called Sir2, were shown to play a role in yeast senescence. By screening for compounds that activate Sir2, resveratrol was found to be one of the strongest candidates and shown to extend yeast lifespan (Howitz et al., 2003). In flies and worms resveratrol was later reported to increase lifespan by, respectively, about 20 and 10% (Wood et al., 2004), though more recent results from other groups have questioned the ability of resveratrol to increase lifespan in flies and worms (Bass et al., 2007). Whether increased levels of Sir2 increase lifespan in worms and flies, in fact, has also recently been hotly debated (Burnett et al., 2011; Viswanathan and Guarente, 2011).

Results from mammals are far from clear, in part because of the increased complexity of mammals. The gene in mammals that most strongly resembles Sir2 is called SIRT1 and CR has been shown to activate SIRT1 at least in some tissues (Cohen et al., 2004). So far, whether SIRT1 plays a role in mammalian aging remains unknown, though there are ongoing studies. Interestingly, sirtuin inhibition rather than activation has been shown to protect neurons against cell death and may have applications in certain neurodegenerative diseases like Parkinson's (Outeiro et al., 2007). It has also been reported that while moderate activation of SIRT1 is beneficial in the heart of mice, a large activation can be detrimental (Alcendor et al., 2007). Strikingly, mice with high levels of SIRT1 do not have a longer lifespan, even if some health benefits, such as protecting against some types of cancer, have been reported (Herranz et al., 2010). Taken together these results suggest that SIRT1 does not play a major role in mammalian aging.

As for resveratrol, it has been reported to improve health and survival in obese mice (Baur et al., 2006) and some evidence suggests that it can increase the aerobic capacity of mice (Lagouge et al., 2006). More recently, however, resveratrol failed to increase the lifespan of normal mice (Miller et al., 2011), even though it may ameliorate age-related changes (Pearson et al., 2008). Much more powerful STACs that activate sirtuins to a larger extent than resveratrol have been synthesized and some have been shown to extend yeast lifespan (Yang et al., 2007). They have been shown to improve insulin sensitivity and lower plasma glucose in obese mice (Milne et al., 2007), and slightly increase mouse lifespan (Mercken et al., 2014; Mitchell et al., 2014). One study reported that resveratrol consumption for 30 days in obese humans resulted in beneficial metabolic changes (Timmers et al., 2011). Therefore, we have only scratched the surface regarding the effects of sirtuins, STACs and resveratrol in particular on mammalian health, longevity, and aging. At present, however, there is no evidence to suggest that resveratrol is a viable anti-aging treatment.

Although serious negative side-effects of resveratrol have not been reported, it is difficult, if not impossible, to predict any negative long-term effects of this or virtually any other compound. For example, while eating grapefruit has been reported to help lose weight and reduce insulin levels (Fujioka et al., 2006), recent results suggest that grapefruit intake may increase the risk of breast cancer among postmenopausal women (Monroe et al., 2007). Therefore, caution is advised.

Overall, like for many other products that can extend lifespan in animal models, even though resveratrol and the aforementioned compounds might have beneficial effects, there is no evidence that these products delay the process of aging in animal models and certainly not in humans. Although it is possible that some of these compounds are healthy, just like vegetables and fruits (like grapes) are healthy, or may have a positive effect on certain age-related diseases such as diabetes, there is at present no evidence they can delay, even if slightly, the human aging process.


If you are serious about trying to delay aging then CR is, with our current knowledge, your only potential chance of success. It will likely not have such a marked effect as in rodents, particularly if you already have a healthy diet, and there are side-effects to consider. Probably, CR will not delay human aging even if it may have some health benefits and protect against some age-related diseases, in particular cancer (de Magalhaes, 2013), and it might slightly extend lifespan. Personally, I would only consider undergoing CR if I had cancer, and even this would have to be carefully considered and discussed with my doctor. As for CR mimetics and resveratrol, like I mention elsewhere for other dietary supplements, there is no evidence to suggest that they extend lifespan of model organisms, much less delay human aging. Since we also do not know anything about any long-term effects of taking such products, my opinion based on our present knowledge is that the potential benefits (and there may be some) do not outweigh the potential risks. Recapitulating my conclusion for other similar products, no therapy is a replacement for a healthy lifestyle.

Having said all that, targeting genes and pathways that mediate CR life-extension effects remains one of the most promising avenues for targeting aging-related mechanisms with the view of at least improving human health and fighting age-related diseases. For a review of drug discovery in the context of aging, in particular via CR-related pathways, see (de Magalhaes et al., 2012).

Disclaimer: The information on this page is for informational purposes only and should not be used as medical advice. If you wish to undergo any of these treatments or take any of the products described in this page, you should talk to your physician. I often make generalizations that might not apply to your particular medical history. As with, use this information at your own risk.

Up to the Visitor's Resources

Back to

Thank you for visiting my website. Please feel free to contact me if you have any questions, ideas, comments or suggestions.

Copyright © 1997 - 2001, 2004, 2005, 2007, 2008, 2012 - 2014, 2016, 2018 by João Pedro de Magalhães. All rights reserved.