Will calorie restriction work in humans?

Accumulating data suggest that protein intake and dietary aminoacid composition play an important role in regulating mTOR activity, serum IGF-1 concentrations, and longevity [ 2 , 38 , 49 - 54 ]. Our data show that in humans, unlike in rodents, severe CR does not reduce serum IGF-1 concentration unless protein intake is also reduced close to the USDA recommended intake (i.e. 10% calories from protein, or 0.8 g/kg/day) [ 38 ]. Data from genetic animal and human studies indicate that serum IGF-1 concentration is an important regulator of aging [ 2 , 55 - 56 ], and has been found to be inversely correlated with median lifespan in 31 genetically diverse inbred mouse strains [ 57 ]. Interestingly, back in 1950 the Okinawan centenarians were consuming a CR diet (approximately 1800 cal/day) with only 9% of calories coming from protein [ 58 ]. In contrast, both the WNPRC and NIA monkey diets contain 15% or more calorie from protein, which is similar to the average protein intake of US men and women [ 59 ], and may explain why serum IGF-1 concentrations were similar between CR and control monkeys. More studies are needed to understand the role of protein intake (and aminoacid composition of foods) with and without CR in regulating the pro-longevity PI3K/AKT and mTOR pathways in rodents and primates. The old dogma that only calorie intake, and not macronutrient composition (and in particular protein intake), is an important regulator of lifespan is based on a flawed interpretation of a study published by the Masoro's group in 1985. In this experiment the authors restricted calorie intake in one group of rats by 40% and compared them to rats given free access to a diet in which protein content was reduced by 40%. The control group ate a usual protein content diet ad libitum [ 60 ]. In this experiment the 40% CR diet increased maximal longevity ~35%, while the reduced protein diet had no effect. The control diet provided 20% of calories from protein, compared to 12% of calories from protein for the “protein restricted” diet. The problem the authors failed to recognize is that, despite the 40% reduction, protein intake was still more than adequate, i.e. was above the threshold needed to cause an inhibition of the IGF/mTOR pathway, as evidenced by the finding that weight gain for these young, growing animals was the same in the 12% and the 20% protein diet groups [ 60 ]. In contrast, the CR animals gained little weight and were markedly stunted as adults, suggesting that only 40% CR and not 40% protein restriction, inhibits the PI3K/AKT/mTOR pathways. Despite Rhesus monkeys (Macaca mulatta) representing one of the most closely related species to human primates, sharing a ~93% DNA sequence identity with the human genome [ 61 ], major differences in longevity exist between monkeys and humans. Even under the best husbandry and dietary conditions (i.e NIA CR monkey study), average and maximal lifespan of rhesus monkeys is ~31 and ~40 years, respectively. In contrast, average and maximal lifespan in humans is ~80 and ~120 years, respectively. The reason why Rhesus monkeys lifespan is much shorter than in humans is not known, and may involve a different rate of accumulation of unrepaired molecular and cellular damage with time. Therefore, it is extremely important to study the health and longevity effects of CR without malnutrition in humans. Whether or not CR without malnutrition will extend lifespan in humans is not known yet, but accumulating data indicate that moderate CR with adequate nutrition has a powerful protective effect against the development of obesity, type 2 diabetes, inflammation, hypertension and cardiovascular disease, which are major causes of morbidity, disability and mortality [ 37 ]. Accordingly, Lloyd-Jones and colleagues found that in men and women from the Framingham Heart Study with normal cardiovascular risk profile at age 50 (i.e. total glycemia <125 mg/dl, blood pressure <120/80 mmHg, cholesterol <180 mg/dl, BMI <25 kg/m 2 and no smoke) the lifetime probability of developing an atherosclerotic cardiovascular disease was very low (i.e., 6.7% versus 59.5% in participants with ≥2 cardiometabolic risk factors) and average lifespan markedly longer (i.e. >39 versus 29.5 years in participants with ≥2 cardiometabolic risk factors) [ 62 ]. In humans calorie restriction without malnutrition also results in a consistent reduction in circulating levels of growth factors, anabolic hormones, adipokines and inflammatory cytokines, which are associated with an increased risk of some of the most common types of cancer [ 63 ]. It is important to note that none of the 50 men and women (age range 30-82 yrs) practicing long-term CR with adequate nutrition is taking any medication or has developed any chronic disease so far. Moreover, CR in these individuals resulted in an amelioration of two well-accepted markers of cardiovascular aging, i.e. left ventricular diastolic function and heart rate variability [ 64 - 65 ]. These data indicate that CR exerts direct systemic effects that counter the expected age-associated changes in myocardial stiffness and autonomic function so that LV diastolic function and heart rate variability indexes in CR individuals are similar to those of individuals 20 years younger on a typical Western diet. Consistently, we recently found that CR without malnutrition results in dramatic changes of the human skeletal muscle transcriptional profile that resemble those of younger individuals, including a down-regulation of the PI3K/Akt/FOXO pathway, suggesting that CR in humans can slow the age-associated transcriptional modifications in skeletal muscle [ 66 ]. More studies are needed to understand how macro- and micro-nutrients, endurance exercise, and other environmental and psychological factors interact with CR in modulating metabolic and molecular pathways that regulate health and longevity. Both excessive dietary restriction and overnutrition are different forms of malnutrition that lead to organ dysfunction and increased mortality. Even in rodents, excessive CR imposed on some strains of mice increases mortality. For example, it has been shown that wild-caught mice undergoing 40% CR do not live longer than ad-libitum fed mice, despite a much lower cancer incidence [ 67 ]. This may be caused by excessive CR during the developmental age, because 40% CR caused higher mortality early in life, but lower mortality late in life, with the longest lived 8% of mice all coming from the CR wild-caught rodents. Furthermore, in C57BL/6J mice 40% CR increases mortality when started just after weaning (i.e. 4 weeks of age), but increase lifespan when started in middle age [ 68 - 69 ]. The problem is that the rate of physiologic development and sexual maturation varies among different strains of rodents, so that the lifespan response to CR may be different. Forty percent CR may be optimal in some strains of mice, but can cause severe starvation and increased mortality in others, which would benefit from a lower degree of CR. The same applies to humans, in which severe CR could be detrimental in some populations (e.g. children, older adults, pregnant women, etc.). Additional studies are warranted to identify the precise CR-induced metabolic and molecular adaptations associated with healthy longevity, so that dietary energy content and macro-nutrient composition can be tailored based on age, sex, disease predisposition and biological/genetic phenotype of each individual. Randomized, CR-controlled, long-term survival studies in humans will never be performed because of obvious problems with long-term compliance and costs of such a long study. Nonetheless, we hope that by following the health status of individuals practicing long-term CR without malnutrition (i.e. the CRONies), in particular of those who are now in their 70s and 80s, we could gain soon some information about the effects of CR on successful aging and healthy longevity in humans as well. Because we have detailed information about their close relatives' disease and survival histories, if we observe that as the CRONies age, they not develop any of the metabolic abnormalities and/or chronic diseases typical of their parents/siblings, and live substantially longer than their relatives, this will be the best available proof that CR works in humans.