Calories or protein? The effect of dietary restriction on lifespan in rodents is explained by calories alone

Almost exactly 100years ago Osborne and colleagues demonstrated that restricting the food intake of a small number of female rats extended their lifespan. In the 1930s experiments on the impact of diet on lifespan were extended by Slonaker, and subsequently McCay. Slonaker concluded that there was a strong impact of protein intake on lifespan, while McCay concluded that calories are the main factor causing differences in lifespan when animals are restricted (Calorie restriction or CR). Hence from the very beginning the question of whether food restriction acts on lifespan via reduced calorie intake or reduced protein intake was disputed. Subsequent work supported the idea that calories were the dominant factor. More recently, however, this role has again been questioned, particularly in studies of insects. Here we review the data regarding previous studies of protein and calorie restriction in rodents. We show that increasing CR (with simultaneous protein restriction: PR) increases lifespan, and that CR with no PR generates an identical effect. None of the residual variation in the impact of CR (with PR) on lifespan could be traced to variation in macronutrient content of the diet. Other studies show that low protein content in the diet does increase median lifespan, but the effect is smaller than the CR effect. We conclude that CR is a valid phenomenon in rodents that cannot be explained by changes in protein intake, but that there is a separate phenomenon linking protein intake to lifespan, which acts over a different range of protein intakes than is typical in CR studies. This suggests there may be a fundamental difference in the responses of insects and rodents to CR. This may be traced to differences in the physiology of these groups, or reflect a major methodological difference between 'restriction' studies performed on rodents and insects. We suggest that studies where the diet is supplied ad libitum, but diluted with inert components, should perhaps be called dietary or caloric dilution, rather than dietary or caloric restriction, to distinguish these potentially important methodological differences.