A number of studies have found an inverse correlation between lifespan and fertility as a result of caloric restriction. This is true in fruit flies, worms, rodents, and mammals, including primates -- we blogged about the recently reported primate study here. The results of a recently reported experiment on flies that takes the question one step further, teasing out the specific dietary components responsible, was reported in Nature last week (Amino-acid imbalance explains extension of lifespan by dietary restriction in Drosophila, Grandison et al., published online 2 Dec 2009).
Although Grandison et al. carefully don't stress a selective aspect to their finding, they do mention an idea that others have proposed, which is that in times of scarcity, organisms can expend their limited supplies of energy on survival or on reproduction, but not on both, and one view can be that it makes sense to delay reproduction until food becomes more abundant. The literature on this subject suggests that this is an evolved response. In their study, Grandison et al. tested the prediction that it's not possible to optimize both lifespan and fecundity with the same dietary conditions.
In laboratories, fruit flies are fed yeast and sugar. Grandison et al. diluted their flies' usual diet, and indeed found that this resulted in longer lifespan and fertility reduction (which they measured as number of eggs laid). After finding the expected effect, they sought to maximize lifespan and fecundity by adding back specific components of the diet and determining whether a nutrient had an effect on both.
Among other things, the investigators found that replacing vitamins, lipids (fats) or carbohydrates had no effect, which suggests that caloric intake in and of itself is not what reduces lifespan in these flies, and that these aren't the components of the diet that affect fecundity. They did find an effect of reintroducing essential amino acids, however. Adding back essential amino acids increased fecundity to the level found at full feeding, specifically methionine, although they also saw increased fecundity with the addition of tryptophan.
And, adding back methionine didn't reduce lifespan, leading the investigators to conclude that "the fact that high fecundity and high lifespan can co-occur is inconsistent with the idea that any aspect of reproduction directly inflicts damage on the soma to shorten lifespan", which has been suggested by some. In addition, they conclude that "the responses of lifespan and fecundity to full feeding are independently mediated by different amino acids."
Finally, they suggest that there is "thus an imbalance in the ratio of amino acids in yeast relative to the ratio the fly requires for the high fecundity from full feeding, and some consequence of this imbalance decreases lifespan."
The mechanisms that influence lifespan are conserved over the large evolutionary distances between invertebrates and mammals, and our results hence imply that in mammals also the benefits of dietary restriction for health and lifespan may be obtained without impaired fecundity and without dietary restriction itself, by a suitable balance of nutrients in the diet.
But is it adaptive?
This is all very interesting, but natural selection need have had nothing to do with it in species in which lifespan extends past the reproductive span, so that natural selection essentially can't 'see' it (there are some technical ways in which there can be a fitness effect, but these are very indirect and often seem quite forced, a way of ensuring that every trait has to have a selective explanation).
The fact that there's so much play in the caloric intake that can sustain life suggests that this is yet one more example of the imprecision of life. But, if one must force an adaptive explanation, it could be something that happened very early in evolution because all organisms have it, and that's the ability to adapt to change, both minute-to-minute, by modulating heart rate or body temperature or increasing bile production after a fatty meal, or over the annual cycle, by dropping leaves when temperatures fall and daylight decreases or going into hibernation, and so forth.
Cells use receptors to detect and capture nutrients that are passing by, in the circulation for example. They can respond to low nutrient levels by increasing the number or density of receptor molecules they put on their surface, or by storing the energy they are able to extract from their environment. This is very generic. Since there are so many more somatic than germ-line cells, one consequence could easily be that in hard times the germ line receives a proportionately lower fraction of what's available. There need be no particular selective competition between these types of cells, though of course it's easy to dream up all kinds of scenarios about this if one feels compelled to do that.
In the face of constantly changing environments, organisms have to be able to respond and keep up. That is why life is organized the way it is, a major feature of our book, in terms of protected, sequestered environments that monitor and respond to what's outside. That is the basic nature of life, or even the reason for life, as a cellular phenomenon.
Responsiveness has its bounds, of course, and some organisms have much wider tolerance for change than others, but the ways in which organisms can adapt is astounding. In addition, almost any physiologic system you look at shows a continuum of 'normal' levels; vitamin D, potassium, calcium, sodium, cholesterol, body mass index, the environmental temperature at which the organism can survive, etc., can all vary, all compatible with life.
And let's expand that continuum beyond 'normal' to survivable -- body mass index, or amount of body fat, is just one measure that can vary greatly and still allow survival, and successful reproduction. So, it's not surprising that the idea of an 'optimal' caloric intake might have little biological importance.
It's common to think in terms of 'survival of the fittest', as though the odds of surviving to reproduce only favor the extremes, but that would suggest that there's one optimal cholesterol or potassium level, or number of eggs to lay and so on. But, life didn't seem to evolve that way, as perhaps there would have been too few survivors for populations to be viable, and perhaps it's just built into the nature of many-component biochemical processes.
We think it's more accurate to think in terms of 'failure of the frail'; for example, an organism whose cholesterol level varies significantly above or below normal may not survive, but otherwise, a wide range of cholesterol levels is compatible with survival and even reproduction. And this is true of most traits. Selection is a lot more tolerant than its reputation in popular culture and even within most of biology.