I study evolution of aging because I think it is a scientific conundrum. But frequently, people tell me that the evolution of aging is easy to understand. After the individual is done reproducing, its job is finished, and there is no more evolutionary force to keep it alive. It can be pushed over with a feather (if natural selection had feathers to push with).
It’s not just naive people who make this statement. People who should know better sometimes say the same thing. I have heard this explanation from the mouth of a Nobel laureate in medicine.
The reasoning is correct, of course, but the problem is that it begs the question, how did the loss of fertility evolve? Sure, it’s true that once the individual has lost the capacity to reproduce (and is no longer caring for its offspring) there is no evolutionary reason for it to remain alive. But why, in the first place, does evolution put up with the loss of fertility? Why aren’t we all like oak trees that grow larger, stronger and more fertile with every passing year?
The mystery is not that we die after we lose our fertility; the mystery is why natural selection has permitted us to lose our fertility.
Semelparous organisms are those that have evolved to reproduce in one big burst. All of them, to my knowledge, are also evolved to die promptly after reproduction. Familiar examples include Pacific salmon, the octopus, many insects and annual plants. Usually, it is clear that death is internally orchestrated, triggered by hormonal signals, behaviors and anatomical changes. Mayflies, gnats and cicadas have no mouths or digestive capacity as adults. Octopus mothers stop eating and starve to death. Salmon poison themselves with steroid hormones. Pansies shrivel up as soon as they “go to seed”, but you can keep them alive all summer as long as you remember to clip off the flowers when they begin to wilt.
Evolutionary biologists thought they had a good understanding of this dynamic, with the exception of human females. Women lose their fertility in their thirties or forties, but can go on to live into their seventies or eighties. What was evolution thinking, in cutting off female fertility? In classical neo-Darwinian thinking, fitness is maximized by natural selection. There’s no such thing as “enough offspring—ok to stop now,” because if someone else has genes that permit more reproduction, then those genes will be the ones that spread through the next generation and grow to dominate the gene pool. Perhaps there is some physiological limit, and nature has tried and tried to extend female fertility, but it’s just a difficult problem… But that’s just not plausible. Women are born with millions of eggs, but over the course of a lifetime, only a few hundred ripen and descend the fallopian tube where they might be fertilized. The rest die in a process called atresia, akin to programmed cell death. Perhaps the eggs just become damaged over time, and they can no longer develop into a viable foetus. It is true that birth defects increase with a woman’s age, but this hardly seems to be a hard-and-fast physiological limit.
The explanation that was offered until a few years ago was the grandmother hypothesis. Perhaps a woman in her 50s can contribute more to her genetic legacy by helping out rearing her grandchildren. Perhaps infertility frees her from raising new babies of her own, so that she can devote full time to grandmothering, and in this way the number of surviving grandchildren is actually increased over what it would be if she remained fertile for an extra decade or two.
The Grandmother Hypothesis was never so plausible on its face, but it survived by default, having no competition. But a few years ago, some Exeter University researchers did a computation confirming what should have been obvious: If your goal is to have as many children as possible, cutting off your fertility is not a good choice. New children to whom you might be able to give birth if you carry on will all bear your gene, whereas grandchildren have only a 1 in 2 chance of carrying the gene, and you can only indirectly help their survival. Michael Cant and Rufus Johnstone estimated the net cost of menopause with real survival statistics from several primitive human cultures, and it wasn’t even close.
There’s more. Humans are not the only animals to lose their fertility and go on living, as it turns out. Whales and elephants are social mammals, and you might imagine that the grandmother hypothesis applies to them. But birds don’t care for their grandchildren, and several kinds of fowl become infertile long before they die. Fish don’t even care for their children, and post-reproductive life has been discovered in guppies.
Lab worms, C. elegans quite drive the point home as they are hermaphrodites, born with more eggs than sperm. They simply run out of sperm, and stop reproducing after about 4 days, but they can go on to live for 10 days. This is a striking example of “programmed death” because sperm is an itty-bitty, stripped-down cell. Metabolically, the cost of producing sperm is negligible, even for a creature 1/10 mm long. So running out of sperm while there are still plenty of eggs to be fertilized can only be seen as a purposeful curtailment of fertility.
Measuring life span and fertility in the wild is not so easy, and to date there are only a dozen or so examples. But wherever they look, field biologists are finding post-reproductive life span.
This is indeed a mystery. If natural selection is optimizing something called “fitness” which depends most directly on the number of offspring left by an individual, why is fertility being curtailed? If there are resources that could be invested in reproduction or traded in for chips that help to keep the body in good repair, then why are so many chips mis-invested in longevity, when fertility would pay a dividend, and longevity, none?
A couple of years ago, Charles Goodnight and I proposed a solution to this conundrum, but it comes at a steep price. We left behind the neo-Darwinian framework in which maximizing fertility is the essence of fitness. We offered an answer in terms of the Demographic Theory of Senescence. This is a theory that says that nature does not maximize individual fitness only, but also concerns herself with stable and robust communities. The key point is that for animals (but not plants), over-consumption and over-population are ever-present, looming threats to the community. It is all too easy to “win” the game of reproducing, using a strategy of beggar-thy-neighbor. Animals in an area share a communal pool of food, and the easy way to reproduce more is to eat more, to gather more biomass and convert it to babies. This strategy offers a very short-lived victory, because once the food species is depleted, it may take many generations to grow back.
I may have more children than you, and get more of my genes into the gene pool of the next generation. But the result is that my children are greedy, like me, and there are more of them, and it doesn’t take long before they outgrow the food supply that was once sufficient to support our community, and everyone starves.
The classical wisdom is that this is a form of “group selection”, and it’s bound to be slow and inefficient compared to “individual selection”. But as Michael Gilpin first demonstrated 40 years ago, population overshoot can be the basis of a swift and lethal form of group selection. With judicious use of computer modeling and careful mathematical logic, he demonstrated that, in this case, it is easy to understand how group selection tempers individual selection.
The radical conclusion of Gilpin is that, in animals, reproduction can never be maximized without destroying the population. This undermines the assumption at the foundation of classical neo-Darwinism, and introduces a new, more communal picture of how evolution works.
It was in the context of this picture that Charles and I were able to demonstrate the evolutionary benefit of post-reproductive life span. Sustainability is a target of natural selection. The goal is to stabilize populations in good times and bad, to avoid population overshoot that leads to collapse and extinction. The mechanism of natural selection couldn’t be clearer: communities that are not sustainable collapse to extinction.
How does post-reproductive life span help? An infertile, older population acts as a kind of buffer during times when the population might otherwise be expanding too fast. When there is plenty of food, the post-reproductive segment eats some of it, but they do not add to population growth in the next generation. Then, when times become more difficult and food is scarce, the older, weaker segment of the population is the first to die off, and this is no real loss to the population’s reproductive potential.
Remember, all this works in animals but not plants. In fact, post-reproductive lifespan is not found in plants, and indeed most plants appear to follow the expectations of neo-Darwinian theory far more closely than animals; which is to say that plants do appear to be maximizing their reproduction, while animals do not.
This demographic perspective may or may not be the solution to the conundrum of post-reproductive life span. Clearly it doesn’t explain everything–for example, why do pansies die promptly after they go to seed? But at present the Demographic Theory has few competitors, and we think it is a good beginning, and should be a fruitful basis for exploring a new understanding of evolution’s basic machinery.
More details of the Demographic Theory of Aging in my blog post from last year. Here is a link to the journal article by Mitteldorf and Goodnight.
It’s an interesting idea, and what would it mean for treating aging? How much of aging is an “active” program playing out, and how much of it is just random damage occurring due to weaker-than-ideal error-correction, etc?
Welcome, Brian! As you read more, you’ll discover more evidence that aging is programmed death, and not wearing out. http://joshmitteldorf.scienceblog.com/2014/04/07/no-the-body-doesnt-just-wear-out-as-we-get-older/
The parts of aging that appear to be deterioration or damage are all the result of the body shutting down repair mechanisms that were perfectly adequate in our youth.
And yes, this is very positive for anti-aging medicine. It means that we don’t have to “fix” what goes wrong, but only to signal the body to take better care of itself. http://joshmitteldorf.scienceblog.com/2013/10/15/how-young-blood-differs-from-old/
Good thinking Josh…..
additional facts to consider for your model…..as food supplies get smaller and individuals enter a phase of semi starvation hormonal changes both slow the aging process AND inhibit fertility
For example when they experimented with monkeys at Madison with caloric restriction they noticed they had elevated levels of melatonin. Melatonin at 75mg per night can and has been used in human females for birth control. Melatonin also has side effects in men that could also likened to a male form of birth control. Reduced sex drive , reduced semen volume.
So your model says that the elderly provide a non reproducing buffer…but that only applies to females . Remember Ramses had something like more than 100 children when he died at age 90 -ish.
However given that a woman’s womb is required to have kids….maybe in this case the fertility of males is irrelevant..
I came up with the opposite of the grandmother hypothesis years ago (which is a ridiculous idea)
It just takes a little thought experiment…Like most other animals ancient humans likely had maximum lifespans pretty much equal to the age of menopause in women…( as if today we all died around age 50)
So the real question is why did women evolve such long postreproductive lifespans?
Simple..if her sons inherit her aging system….but her sons have no shut off of fertility with age…
.A woman who evolves a longer lifespan will have sons with longer lifespans who can then father that many more children.
I call it the “Son-King Hypotjesis” and there are some indications it is correct….as they found there is a huge number of east asian men that have some DNA that comes from Genghis Kahn…
I never published the idea…..just used it to bash the ridiculous grandmother hypothesis.
Another possible factor driving the evolution of lifespan in animals, (proposed by, I think it was, Michael Rose), is predation and accident. The idea being that there is no adaptive value in spending resources to extend lifespans beyond the point where most individuals would be dead from predation or accident anyway. If you buy that, then the question becomes “why cut off reproduction well in advance of that age?” A possible answer might be that you want to keep mothers around long enough after birth to raise offspring to maturity — which would very roughly coincide with the observed fertility period and lifespans of human females, and also perhaps explain why male fertility does not cut off early (if you assume that fathers’ contribution to raising offspring is less essential).
(I am enjoying your posts very much, very glad to have discovered your blog.)
Hi, Jack –
The idea that environments are so tough that no one lives long enough to die of old age came into vogue with Medawar (1952). People believed it by default for a long while, but it has been thoroughly discredited, I believe, by studies of mortality and age distribution in the wild. The fitness cost of aging in the wild is substantial. Even though no one literally keels over from old age, the early stages of senescence cause the mortality rate to be higher for old animals than for young animals, and this occurs at ages that many animals attain in the wild. See for example http://www.jstor.org/discover/10.2307/2410110 and http://www.nature.com/nature/journal/v420/n6914/abs/420377a.html .
This is all very interesting… I like this discussion!
There is another aspect of the demography theory of aging: that is the gender difference in fertility vs age, and the reciprocity of old and new information carried by the eggs vs sperm cells..At birth, in females there are approximately 1 million eggs- already divided; and by the time of puberty, only about 300,000 remain. Of these, only 300 to 400 will be ovulated during a woman’s reproductive lifetime. While a female is born with all the eggs she’ll ever carry, by the time a male turns 40, his gonad cells will have divided 600+ times to make spermatozoa. By the time he’s in his 50s, that number goes up to 800+. Males are reproductive even at their 80ies+..Each time of these cells dividing, mistakes may appear in the DNA, but also the epigenetic learning could also be programmed into the new male sperm cells as a „new learning”- eg new skills, how to fight, how to be more adaptive in the challenging environment etc. This epigenetic upgrade information for males’ skills were probably more important in the ancient times to protect the tribes than for females, whose caring functions were the primary „wired” need of the community, that was less informed via transgenerational epigenetic modifications.
It raises another hypothesis, that cheating the epigenetic information channel that informs the new spermatozoa to include this new learnings of the fathers, could influence how the next generation’s genetic programming will see the age of the new individual, and it will lead to longer or shorter life expectancy…(?) Let me know your thoughts- (sorry if it was already discussed elsewhere in these pages).
I just found this paper very interesting- related to my previous comment: Eisenberg et al. published a study in 2012 which showed that greater paternal age of reproduction in humans is associated with longer telomeres across two generations of descendants. Since telomere length has effects on health and mortality, this may have effects on health and life expectations in these offspring. The authors speculated that this effect may provide a mechanism by which populations have some plasticity in adapting longevity to different social and ecological contexts. Does it mean that the older males’ sperm cells include the epigenetically transcripted biological and social learning on how to live longer into the next generations?
Eisenberg DT, Hayes MG, Kuzawa CW. Delayed paternal age of reproduction in humans is associated with longer telomeres across two generations of descendants. Proc Natl Acad Sci U S A. 2012 Jun 26;109(26):10251-6. http://www.pnas.org/content/109/26/10251.full.pdf
This is a great find. Thank you, Thomas!
Few other interesting research papers, you may find interesting:
Giraldo & Traniello, 2014: This research critically examine selection for worker lifespan in ants and discuss the relationship between functional senescence, longevity, task performance, and colony fitness. The potential benefits of large colony size, achievable in part through extended worker life span, appear to argue against the programming of worker senescence and suggest that colony level selection promotes ant worker longevity under conditions that realistically affect colony fitness. This paper suggests that programmed senescence and functional senescence may not be necessarily directly linked. If worker function does not decline with age, the efficacy of task performance by older workers may even be enhanced by experience through changes in gene expression or synaptic connectivity that maintain sensory perception, integrative processing, and motor output in “elderly” workers:
Giraldo YM, Traniello JF. Worker senescence and the sociobiology of aging in ants. Behav Ecol Sociobiol. 2014 Dec;68(12):1901-1919.
Khalyavkin & Krutko, 2014: This new theoretical paper developed another approach that considers both programmed and stochastic concepts of aging to be incorrect. they argue that aging is a simple deprivation syndrome driven by preventable and even reversible drifts of control systems set points because of an inappropriate “organism-environment” interaction.
Khalyavkin AV, Krutko VN. Aging is a simple deprivation syndrome driven by a quasi-programmed preventable and reversible drift of control system set points due to inappropriate organism-environment interaction. Biochemistry (Mosc). 2014 Oct;79(10):1133-5.
Goldsmith TC: The Evolution of Aging: How New Theories Will Change the Future of Medicine. Azinet Press, 2014.
Full text: http://www.azinet.com/aging/Aging_Book.pdf
Interesting article to confirm programmed aging here.
It actually resembles one year old Nature article.
Here is a link to an aritcle about new research involving heat shock proteins in C. elegans strongly supporting programmed aging, and this mechanism seems to be preserved across species.
The demographic perspective does speak to the issue.
Another perhaps synergistic idea may be the maintenance of the ‘wealth’ of experience (cultural learning) that this post reproductive population can bring to bear for the community. Certainly this would aid to the survival and longer term health, well being and prosperity of the community.
Sure – but accumulating wisdom doesn’t work so well for worms as it does for elephants and humans.