by Nick Lane
After making allowances for historical trends towards smaller families and longer lives, they still found that “the longest-lived aristocrats tended, on the average, to have had the greatest trouble with fertility”. Taking the pattern as a whole, they concluded that a predisposition to above-average longevity is indeed linked to below-average fertility.
I am satisfied that the disposable soma theory applies to us. This is good news, in the sense that an obsolete optimum, which has served its purpose, can in principle be modified to a more congenial optimum with a new purpose — to abolish the misery of old age. The disposable soma
234 • SEX AND THE ART OF BODILY MAINTENANCE
theory argues that the rate of ageing is dictated by the level of resources that are committed to bodily maintenance. The question we must address now is why do these resources become less efficient as we grow older? If we can maintain ourselves perfectly well in youth, when our sexuality is at its height, why do we then decline?
The disposable soma theory does not discriminate between competing mechanistic theories of ageing, such as the programmed and stochastic theories. For example, are we programmed to commit maximum resources to somatic maintenance until reaching sexual maturity, and then decline after we have switched our resources to sex? Such a process could be envisaged easily enough in terms of hormonal changes, which in any case control development, puberty and the menopause.
Why not ageing? Or is ageing not an example of programmed development at all, but a gradual accumulation of damage? If so, why are we not troubled incrementally from childhood? Why do we not ‘feel’ as if we are ageing until we reach middle age? We will explore these issues in the next chapters.
C H A P T E R T W E L V E
Eat! Or You’ll Live Forever
The Triangle of Food, Sex and Longevity
ugust weismann, the nineteenth-century german biologist who first made the distinction between the immortal germ line and Athe mortal body, was also the first to think about ageing from a Darwinian perspective. Since resources are limited, said Weismann, it is imperative that parents do not compete with their offspring. Weismann argued that ageing was a means of ridding the population of worn-out individuals, thereby clearing space for offspring, but not so fast as to lose the social benefits of experience. There are also genetic advantages to having a flux of individuals in the population. The problem is that a genetically static population is a sitting target for pathogens and predators. In the same way, it is much easier to rob a bank if you have memorized the unchanging patrols of security guards. In each generation, old genes are remixed in new combinations and so present a shifting and elusive target for predators and pathogens. According to Weismann and his followers, ageing is an adaptation to reap the benefits of social wisdom, while clearing space for new individuals and thus maintaining the species in a state of genetic flux.
Weismann’s argument is nowadays dismissed by most evolutionary biologists as unworkable, as it places the emphasis on so-called group selection rather than individual selection. If ageing is programmed in the same way as, say, our embryonic development, the benefit is to the group,
236 • EAT! OR YOU’LL LIVE FOREVER
not the individual, who gains nothing from being displaced. As we saw with sex, any theory that seeks to explain the origins of a trait must do so in terms of individual selection. On the other hand, even if group selection does not explain the origins of ageing, it remains possible that group selection could maintain ageing once it had evolved. The idea persistently refuses to go away, and is, in fact, the conceptual bedrock of most theories of programmed ageing. Is there any evidence that group selection maintains a programme for ageing?
In the sense that animals and plants have fixed lifespans, longevity is obviously written in the genes. This does not mean there is a formal genetic programme, any more than a car is programmed to become obsolete over 20 years. In the case of a car, the parts are designed from the beginning to last for only so long, and the fact that they wear out simultaneously is no evidence of the workings of a hidden programme. An apoc-ryphal story tells of Henry Ford looking at a junkyard filled with Model Ts.
“Is there any thing that never goes wrong with any of these cars?” he asked.
Yes, he was told, the steering column never fails. “Then go and redesign it”, he said to his chief engineer. “If it never breaks we must be spending too much on it.”
Natural selection works in the same way. If an organ works well enough for its deficiencies not to constitute an adverse selective pressure, then natural selection has no way to improve on it. Conversely, if an organ works better than required (in new circumstances), the random accumulation of negative mutations over generations will gradually degrade its performance to that required, at which point selection pressure will maintain the standard. For this reason, animals that have adapted recently (in evolutionary terms) to permanent darkness in a cave or at the bottom of the ocean often have vestigial eyes that are no longer functional. Degradation to a common denominator is alone sufficient to explain the apparently synchronous wearing out of organ systems as we age. As John Maynard Smith put it, “synchronous collapse does not imply a single mechanism of senescence.”
The impression that ageing is programmed is strongest in animals that undergo ‘catastrophic’ senescence. The most famous example is the Pacific salmon, though there are several others, including mayflies, marsupial mice ( Antechinus) and the octopus Octopus hummelincki. Pacific salmon hatch
The Triangle of Food, Sex and Longevity • 237
in small freshwater streams and migrate to the sea. When mature, the adults cover long distances to return to the stream of their birth, where they spawn, producing huge numbers of eggs and sperm. Within a few weeks, the adults degenerate and die, their rotting flesh augmenting local food chains, ultimately to the benefit of their own young. These dramatic events are brought about by hormonal changes, and benefit the group —
the offspring — rather than the parents. If this is not an example of group selection leading to programmed ageing, as postulated by Weismann, then it is hard to imagine what might be. The process has even been termed
‘phenoptosis’, or the programmed demise of the phenotype, in sonorous contrast to apoptosis, or the programmed death of a cell.
The first point we should appreciate about the Pacific salmon is that it is an exception to the rule, even among salmon. The Atlantic salmon, for example, migrates shorter distances and is able to breed over several seasons; it does not undergo catastrophic senescence. We would be quite mistaken to consider the Pacific salmon a paradigm for human ageing.
Nonetheless, we cannot dismiss the idea of programmed ageing in people unless we can explain why the Pacific salmon is different. In fact, we do not have to look far. The critical point, once again, is sex. The Pacific salmon, in common with mayflies, marsupial mice and O. hummelincki, are semelparous: they breed only once. In iteroparous organisms, which breed repeatedly, the decline is typically more gradual.
Think again about the disposable soma theory of ageing (Chapter 11, page 228) and the trade-off between sex and survival. If an individual breeds only once, and does not look after its offspring, then its survival afterwards has no effect whatsoever on the genetic make-up of the next generation. Looking at it the other way round, all the selective pressure is squeezed into a short time window early in life, during the reproductive period. To understand why this is important, imagine that one individual puts more effort into breeding, perhaps because it naturally produces a little extra testosterone or oestrogen. The heightened reproductive effort results in more offspring, but has a cost in terms of the survival of the parent. If that parent is engaged in rearing young, natural selection might notice the difference and select against it; but in the case of the semelparous salmon, which has no contact with its offspring, the blind watchmaker cares not a whit. The most fertile, shortest-lived salmon will come to prevail in the population. Almost incidentally
, an intensified rush of reproductive hormones will be selected for before spawning, which ensures maximum breeding success. If we measure the hormonal changes
238 • EAT! OR YOU’LL LIVE FOREVER
in the Pacific salmon, we certainly see evidence of a pattern that looks like programmed senescence; but in fact, the hormonal changes are almost certainly secondary to the evolutionary imperative to breed. They are not the causal mechanism behind ageing. Thus, the catastrophic demise of the Pacific salmon is explained by the disposable soma theory as the total dedication of resources to sex, coupled with the complete decommission-ing of all genes involved in longevity.
A similar argument, derivable from the disposable soma theory, applies to iteroparous species, which breed repeatedly. Rather than a single breeding opportunity, the important parameter in this case is the breeding window, bounded by the likelihood of death. Recall that long-lived animals breed more slowly. If they die by accident or from predation, they will leave fewer offspring behind. Short-lived, fertile animals will therefore predominate, and longer-lived variants will be eliminated by natural selection. This clearly happens in species subject to predation, such as opossums. On the other hand, if the threat of predation is lifted, longer lives will be selected for, if only because the risk of mortality in childbirth is lower with smaller litter sizes. Species that have a naturally low mortality in the wild would be expected to live longer because they are not penalized for their slow reproduction. This is true of opossums living on islands, and seems to be true of birds, bats, tortoises, social insects and humans. All live long lives because all are sheltered from predation, by virtue of aerial flight, hard shells, social organization or intelligence.
Longer lives can also be selected for if the parents affect the survival of their offspring. If, by bringing up our children, we increase the likelihood of their survival, then genes for longevity will be selected for. This is not a gift from the fairy godmother, a reward for altruistic spirit; it is simply the opposite of the Pacific salmon’s catastrophic demise. Given a population of individuals with a normal scatter of longevity genes, then the parents that live the longest will, on balance, offer more support to their children. These children will have a better chance of surviving childhood.
They will, of course, inherit the same longevity genes, so their own children will reap the same benefits. Eventually, a long-lived population will be selected for (as long as ‘accidental’ death is held at bay for long enough).
As Tom Kirkwood and Steven Austad have argued, the power of this effect is exemplified by the existence of the menopause. For older women, the balance between sex and survival adjusts itself towards survival. Older women have more to offer, in biological terms, by bringing up children
The Triangle of Food, Sex and Longevity • 239
than they do by having more children. The risk of childbirth is dangerously high for older women, whereas their prolonged survival benefits existing children or grandchildren; hence the menopause. The same cannot be said of men, who do not give birth or go through the menopause, and generally die earlier. Over evolutionary time, the prolonged survival of fathers has been less important, and they have less to lose from father-ing more children.
In all the cases we have examined, the advantages of a long or a short life always accrue to the individual. We have come full circle. This view is exactly the opposite of Weismann’s theory, with which we opened this chapter. Weismann argued that ageing is somehow programmed into individuals, an enforced act of altruism for the good of the species. In fact, even catastrophic ageing is explained far more believably by the disposable soma theory, in terms of selfish individual selection. If faced with a necessarily short lifespan, individuals pass on more of their genes if they breed quickly. The effort to breed undermines the capacity for survival, as both properties draw on resources from the same pot. The lifespan and the rate of breeding therefore find an optimal trade-off that fits the time-window available. If the time-window is less pressing, then longer lives are selected for, especially in cases where the parents help to rear their offspring.
In all of these cases, the genetic balance resets itself through selection. There is no need for a programme for ageing and no evidence that one exists. Even without a programme, however, these changes are clearly encoded in the genes. Thankfully, people do not senescence catastrophically after sex; a small nap is enough. If ageing is in the genes, but is not programmed, what does happen?
Ironically, the difficulty with Weismann’s Darwinian argument is implicit in Darwin’s own theory of natural selection. Survival of the fittest presup-poses death of the weakest. When set against a backdrop of high mortality, selection pressure falls quickly with time. If our average life expectancy was 20 years, and our reproductive cycle was completed within this time, then there would be little selective pressure to extend life beyond 20. This argument was first put forward by J. B. S. Haldane and Peter Medawar in the 1940s and 1950s, and later developed by the American evolutionary biologist George C. Williams as the antagonistic pleiotropy theory of ageing
240 • EAT! OR YOU’LL LIVE FOREVER
(pleiotropy is from the Greek for ‘many effects’, of which some are opposing, or antagonistic).
The example of low selection pressure originally cited by Haldane in 1942 is still perhaps the most eloquent: Huntington’s disease. The hallmarks of this cruel genetic disorder are a relentlessly progressive chorea (loss of motor control causing repetitive, jerking movements) combined with dementia. Typically starting with mild twitches and stumbling in early middle age, the disease eventually strips away the ability to walk, talk, think and reason. Historically, the lurching madness was mistaken for possession by witchcraft, and many victims were burnt at the stake, including some of the notorious Salem witches in 1693. Despite its severity, Huntington’s disease remains among the most common genetic disorders, afflicting one in 15 000 people worldwide. In some areas, such as the villages lining the shores of Lake Maracaibo in Venezuela, the prevalence is as high as 40 per cent. In these villages, all the cases are thought to be descended from Maria Concepción, who had 20 children early in the nineteenth century. She is said to have had 16 000 descendants (so far).
Huntington’s disease is caused by a single, dominant gene. Dominance means that only one copy of the gene is needed to cause the disease, unlike most genetic diseases, which are ‘recessive’; that means they only occur when the person carries two copies of the ‘bad’ gene. As we saw in Chapter 11, in diploids, the negative effects of a ‘bad’ gene from one parent are often suppressed by a functional copy of the gene from the other parent. A number of such recessive traits are maintained in the population by hidden benefits. For example, the defective haemoglobin gene responsible for sickle-cell anaemia also protects against malaria, and has been maintained at a high frequency in regions where malaria is endemic, such as West Africa. Each year, hundreds of thousands of children die from sickle-cell anaemia, but the carriers, who have a single bad copy of the gene, rarely suffer from serious anaemia. They benefit by having an almost complete protection against malaria and its consequences. The frequency of the gene for sickle-cell anaemia is therefore determined by the balance of risks and benefits. (J. B. S. Haldane, incidentally, was the first to suggest that there might be a link between sickle-cell anaemia and malaria.)
The same cannot be true of Huntington’s disease, where every carrier succumbs to the disease. The difference here, as Haldane pointed out, is the average age of onset — 35 to 40 years. For most of the history of mankind, most people simply did not live to that age. The selection pres-
The Triangle of Food, Sex and Longevity • 241
sure to remove the Huntington mutation from the population has therefore been weak. Imagine, in contrast, the fate of any variant of the gene that caused this disease at the age of 10 — it would be eliminated from the population, as those possessing it would not have children.
In this light, ageing could be seen as the r
esult of the accumulation of deleterious late-acting mutations over many generations, not within an individual lifetime. Each individual inherits the late-acting mutational baggage of previous generations. Ageing is thus a kind of ‘rubbish bin’ of bad genes. The idea of antagonistic pleiotropy is a development of this concept. The problem with the rubbish-bin theory is that there is no selective force causing negative late-acting mutations to accumulate: there is no selective force that favours degeneration, other than the general tendency towards wear and tear. George C. Williams put forward one positive reason why genes with a detrimental effect might be selected for in evolution. He pointed out that many genes have more than one effect: they are pleiotropic. In the same way, we saw that vitamin C is involved in multiple cellular processes. Similarly, a gene might have some beneficial effects, but we can easily envisage that these benefits might be opposed, or antagonized, by other, detrimental, effects. In the case of vitamin C, we saw that its beneficial antioxidant properties are counterbalanced, in some circumstances, by potentially dangerous pro-oxidant properties. The theory of antagonistic pleiotropy posits that when genes have both ‘good’ and ‘bad’ effects, the outcome is an optimal trade-off between the good and the bad.
The theory of antagonistic pleiotropy assumes that rather than being merely a rubbish bin of late-acting mutations, individual genes involved in ageing would have beneficial actions early in life and detrimental actions later. If the benefits outweigh the disadvantages, then the gene will be selected for by evolution. As Medawar put it, “Even a relatively small advantage conferred early in the life of an individual may outweigh a catastrophic disadvantage withheld until later.” Let us stay with Huntington’s disease. A number of studies have suggested, tantalizingly, that the mutations in the Huntington’s gene do in fact confer a competitive advantage in youth, although the mechanism is unknown. People with the gene for Huntington’s disease, who go on to develop the disease in middle age, tend to have more interest in sex than the rest of us. Studies in Wales, Canada and Australia concur that fertility is enhanced in people who go on to develop Huntington’s disease, compared with either their unaffected siblings or the general population. The slightness of this effect