Ageing – why, how and what happens next?
This essay was written by Michael G. Sargent and was first published in the 2009 Mill Hill Essays.
Ageing, one of nature’s most familiar but most mysterious phenomena, always invites the question: “Why can’t I stay like this forever?” Probably every generation in recorded history has wondered why human physical and mental powers slip away with the bloom of youth but a convincing answer only emerged quite late in the twentieth century, in the form of an evolutionary theory of ageing. After the publication of Charles Darwin’s On the Origin of Species, ageing and death were usually regarded as essential facilitators of natural selection that made space on earth for future generations. The idea was not challenged until the 1950s, when the fundamental flaw was exposed by Peter Medawar, the Nobel Laureate and Director of NIMR from 1962-71, when he pointed out that wild animals rarely survive beyond their reproductive years, but die from predation, starvation and hostile environments and not through diseases of old age. Consequently, natural selection could not favour the development of such a programme and, in any case, was intrinsically incapable of operating for “the good of the species” only for the benefit of the individual.
Another important contribution to answering our timeless question came in the late 1970s from Tom Kirkwood and Robin Holliday, in the Genetics Division at NIMR. They argued that all the well known phenomena of ageing reflected a failure of bodily maintenance beyond the reproductive years and that maintenance of the body was entirely secondary to the reproductive imperative. For small animals whose existence was constantly threatened by predators this meant focusing resources on rapid reproduction, while animals whose lives were less precarious could invest in better parental care for fewer progeny. Their theory, called the disposable soma hypothesis, is in a sense a tribute to the nineteenth century German zoologist, August Weismann, who laid the foundations for modern thinking about ageing. His important insight was to distinguish between the germ-line (the lineage of cells that make eggs and sperm) that was potentially immortal, and cells of the soma (the body) that he predicted had limited division potential. Seventy years later, this was confirmed experimentally with the discovery of the “Hayflick limit” the limited capacity to divide that was characteristic of cells in culture.
For Kirkwood and Holliday “disposable” was an allusion to the throw-away products of modern manufacturing, which they believed were analogous to the fate of those whose reproductive duties had been discharged. The hypothesis gave a physically tangible answer to our question; physiological and cellular processes are damaged by the ageing process and longevity could only be secured by investment in systems of maintenance. We can appreciate this point of view if we examine ageing phenomena in nature, but first we need a definition of ageing. Most opinion defines ageing broadly, as the accretion of all the bodily changes that make an organism more likely to die, including the circumstances that initiate the diseases of late middle age but not accidents or infections.
If we look at the evolution of reptiles, birds and mammals, we can see the link between evolution and longevity. In the earliest period of mammalian history, when dinosaurs ruled the earth, mammals were tiny animals at the bottom of the food chain that survived as a species only by reproducing rapidly. Their small modern successors still make the most of their short lives by rapid growth, early sexual maturity and prolific reproduction. Bigger terrestrial animals emerging later in evolution were less vulnerable to predators, starvation or climatic extreme and could evolve towards having smaller families with better prospects of reaching maturity. They grew more slowly than smaller animals and their longevity increased in rough proportion to their size.
Four groups of animals deviate intriguingly from these general rules. Flying birds, bats, sea mammals and primates all have notably greater longevity than land mammals of similar size. Each group emerged when their ancestors discovered previously unexploited habitats teeming with food that improved their chance of survival. This meant that fast early growth and high fecundity were no longer crucial. Resources could be engaged profitably in developing elaborate maternal care for fewer offspring. For bats and birds, the ability to fly was a liberation that helped them evade earth-bound predators, to locate new sources of food, to discover safe nesting sites and to turn their backs on flooded burrows and similar catastrophes. The success of the airborne lifestyle is evident in that most flying birds live three times longer than terrestrial animals of the same size: a robin lives for about seventeen years after it leaves the nest, while some albatrosses live for more than fifty years. Species of bat, half the mass of a mouse, live more than 30 years, a capacity unconnected with hibernation, as non-hibernating tropical varieties are also long-lived. The development of an extraordinarily robust cardio-vascular system allowed sea mammals to exploit the oceans and live very long lives, perhaps two centuries for Bow-head whales. Primates with highly developed cognitive abilities took to the forest canopy where they were relatively safe from predators and could develop arrangements of unparalleled sophistication for nurturing progeny. Many developed a long lifespan and, most remarkably, humans of the industrial world have a lifespan that far exceeds what we might expect based on weight. Enthusiasm for parental care was such that the juvenile state now lasts for almost two decades.
While all known mammals and birds undergo senescence before they die, the class of animal from which they evolved, the reptiles, include species that did not. Turtles, tortoises and crocodiles show few or no signs of senescence in their long lives, growing continuously, remaining fertile and usually dying by accident rather than through a disease of old age. Lizards and snakes however clearly do age. Several reptilian features have a permanency not seen in mammals: eggs retain their viability in the bodies of older females; damaged limbs can regenerate; teeth develop throughout their lives, unlike mammals that usually have two sets of teeth in a lifetime. Nobody understands why some reptiles age so slowly, but it is a trait shared with some deep water fish and lobsters. Even more extreme examples of negligible senescence are found amongst some classes of invertebrates and in plants, which can often develop new individuals from somatic cells.
Ageing is not always the slow crumble we associate with human experience; a few species undergo senescence with breath-taking speed. The Pacific Salmon mates after an epic migration to the ancestral spawning grounds, on the way mobilising every calorie from fat and muscle through the action of cortisol. With the reproductive imperative fulfilled and spawning complete, nothing more is expended on maintenance and the bodies of these exquisite creatures disintegrate rapidly, dramatically epitomising the disposability of the soma. The male marsupial mouse also dies abruptly after a two-week mating-frenzy fuelled by testosterone and the stress-hormone cortisol while the female survives long enough to wean her babies.
Ageing can be extraordinarily delayed in animals that live in caste-based societies. Members of insect communities are genetically identical but queens and workers are distinguished by massive differences in lifespan. Honey bee queens can live for eight years and queen ants for 28 years, while the lower castes live no more than six months and usually much less. The key to this extraordinary longevity is that queens have access to nutritional benefits not available to the other castes. Naked mole-rats, subterranean inhabitants of the desert wastes of Somalia, are the only mammals that form a caste-based society. With their scaly naked flesh and ugly incisors, they lack the bright-eyed charm of furry rodents, but they fascinate biologists for many reasons. Most remarkably, they can live at least twenty-eight years compared with the mouse’s three, although they are less than twice the size and, it seems, never develop tumours. Each colony has just one queen who is the sole breeder, typically using three male partners over a lifetime, while the rest of the colony live chastely as tunnellers or soldiers for as long as the queen. She becomes bigger than the other castes through an enlargement of the vertebral column that accompanies each pregnancy, allowing the gut to expand to manage its substantial food intake. The queens are astonishingly prolific mothers. Eight hundred offspring in a life time in captivity is common and much more than would be predicted by their life expectancy.
The disposable soma theory and many other important ideas are now subsumed into an “evolutionary theory of ageing” that gives the best answer we have to the great “why” question of ageing. Significant dissent may come from those who believe that the Tree of Life is the work of an “intelligent designer” rather than natural selection. They generally have little to say about the ageing process but may prefer to believe the Almighty assigns each of us to one of the gruesome fates that fill medical text-books rather than accept the evolutionist’s view of the ageing process.
Opportunities to investigate directly the evolution of ageing in vertebrates are hard to find; one notable effort was a study of the opossum (a marsupial of the warmer parts of the Americas). The opossums of mainland USA have many predators and live for no more than a year, but the island of Sapelo on the coast of Georgia, is predator-free. The opossums of Sapelo have evolved to be smaller, slower to develop, with fewer offspring but substantially longer lifespans. Exactly as predicted by evolutionary theory, the lack of predators has eliminated the pressure for rapid development and great fertility. Strikingly, analysis of the collagen in tendons reveals that the chemical signs of ageing that make tendons stiff in older animals, appear more slowly in the opossums of Sapelo than in their mainland cousins.
The shorter lifespans of invertebrates make it possible to conduct the most rigorous investigations of the effect of natural selection on lifespan. Eggs from fruit-fly mating can be selected at the very end of their reproductive lives, in repeated cycles of breeding so that eventually the progeny are all long-lived mutant flies. These flies are less fertile than normal strains, are smaller and slower to develop, all as predicted by the evolutionary theory of ageing. Dog-breeders have inadvertently performed a related kind of experiment that had the opposite effect, when they bred big animals with great strength and endurance for hunting. The contemporary Irish Wolfhound, descendants of animals bred long ago to hunt wolves, reaches seventy kilograms but usually lives less than seven years, while a Chihuahua that weighs less than three kilograms can live up to eighteen years. Evidently, selection for very fast growing animals compromises the maintenance systems that permit a long lifespan.
Mechanistic explanations of the ageing process are really responses to the “how” question. Generally, they can be accommodated within the evolutionary theory of ageing and are not contradictory explanations; all probably contribute to the ageing process in an exceptionally complicated way. Genetics clearly explains some manifestations of ageing. The detrimental effects of pro-ageing genes reveal themselves only when reproductive life is over although some of these genes confer important benefits in early life. Known variants of certain genes predispose people to all the major diseases of old age and other variants confer greater longevity.
In parallel with genetic mechanisms, other processes contribute to ageing. The bodily “wear and tear” arising from life experience is caused by the powerful oxidising activity of “free radicals”, by-products of cellular respiration and anti-microbe defence that can damage proteins and DNA. Oxidation of collagen, a key structural component of tendons and connective tissue, makes tendons stiff and skin less resilient. Similarly, the elastin of arterial walls loses its elasticity and contributes to raised blood pressure, while crystalin, a key material of the eye lens, loses its transparency and results in cataracts.
The damage caused by free radicals is not detectable in young animals because a highly efficient housekeeping system identifies and destroys affected proteins. In later life, this system becomes less effective; damaged material accumulates and becomes particularly burdensome to non-dividing cells, such as nerves and heart cells. The start of Parkinson’s and Alzheimer’s disease is usually attributed to cell-death provoked by the pathological accumulation of damaged proteins. Free radicals are not entirely villainous; they play a crucial role in eliminating incorrectly folded proteins, by leaving a characteristic chemical tag on their surface that the housekeeping system recognises. They are then destroyed, eliminating a nuisance that might otherwise interfere in cellular processes as effectively as a “spanner in the works”.
Free radicals are also kept in order by a great variety of anti-oxidants, including the vitamins C and E, in the human body. The fascination of the great longevity of the queens in caste-based animal societies is that they have evolved exquisitely efficient systems for managing free radicals through their food (royal jelly in the case of bees). Bats and the Naked Mole Rat also deal with oxidative damage with notable efficiency.
DNA is constantly assailed by free radicals so repair of damage is a crucial part of cell maintenance systems. If these systems fail, cell suicide may eliminate an irretrievably damaged cell. In later life, these maintenance systems lose efficiency and unrepaired structural damage to DNA may then start a malignancy. As predicted by the evolutionary theory of ageing, DNA repair and maintenance is most powerful in long lived animals and weakest in short-lived animals.
The uterus is an almost perfect nursery for the mammalian foetus, infinitely safer than the arrangements for fish and frog embryos, but it is vulnerable to sub-optimal maternal nutrition or neuro-endocrinological stress that can have far-reaching consequences for adult life. However, these effects are also adaptations to the world inhabited by the mother. A rodent fetus that experiences protein-limited nutrition in utero protects the brain at the expense of other organs, including the liver, kidneys, heart and the insulin-producing cells of the pancreas, an underinvestment that reduces lifespan. In challenging times, this improves an animal’s chance of reproductive success; the brain and the centres controlling reproductive processes are protected at the expense of life expectancy while the resultant smaller body needs less food. The facts are unequivocal for rodents, but more difficult to prove for humans. Nonetheless, low birth weight babies become adults with a tendency to develop diseases of old age prematurely (eg cardiovascular disease or type-2 diabetes).
We now have some understanding of why and how we age but what should we think about the doubling in life expectancy in the industrial world of the last 150 years? Before 1950, improvement in life expectancy was driven by a massive decline in the death rate of infants and young people, but since 1950, it has been propelled by greatly improved survival of older people. An empirical observation by the nineteenth century mathematician Benjamin Gompertz helps understand this continuing improvement in life expectancy. His observation was that once humans reach adulthood, the probability of death (or the rate of ageing) increases exponentially, with a doubling time of about eight years. In other words, the death rate accelerated uniformly until everyone in the cohort was dead. This seemed correct while data for the very old was scarce; now it seems the death rates of people older than eighty years are significantly lower than expected. Evidently, people who reach this age differ from the general population; they could be benefiting from better health-care or from an advantageous genetic pedigree. Another possibility is that this “late-life” phase is qualitatively different to earlier periods, because the risk of initiating diseases of old age has actually declined for those who have passed all the earlier tests of survival.
The incomparably good Swedish State Birth and Death statistics, collected since 1751, are illuminating; the maximum age at death of the oldest Swede alive has increased by 0.44 years per decade from 1861 to 1969 and thereafter, the rate accelerated to about 1.11 years per decade. This seems set to continue, while the death rate of people over seventy years old has steadily declined. The cohort of Swedes born since 1900 have certainly benefited from better health-care but it seems likely that the primary risk factors for age-related disease have been postponed, by better nutrition starting in utero and a decline in chronic inflammation caused by infectious disease. An independent measure of improving health in twentieth century Sweden (and many other countries) is that adult stature has increased markedly as better nutrition permitted a realisation of full genetic capability. As adult stature is generally determined at birth, this points to important improvements in fetal nutrition in this period. Investigations with mice strongly support this idea. Mice reared for many generations on a marginally protein-deficient diet are smaller and have shorter lifespans than control animals reared on an optimal diet; on restoration to the optimal diet, several generations elapse before the progeny grow to normal size and lifespan. This probably involves a modification of the information stored in their genomes, through what is now called an epigenetic code _ essentially reversible changes to DNA organisation that affect the capacity of genes to express themselves. For humans of the industrial world, the trend to greater stature may represent a similar escape from historic malnutrition.
Average human lifespan will probably reach 85 years in the industrial world in 2050 if economic, social and climatic circumstances remain benign, and the biomedical profession’s commitment to improving life expectancy continues. With our understanding of what life experiences might cause premature age-related diseases (smoking, obesity, chronic inflammation, malnutrition, poor foetal experiences) further improvements are within our grasp. We are however, entering a new phase; gene functions critical for the initiation of senescence are becoming apparent and a few biotechnology companies exist that hope to use this information to develop “anti-aging pills”. For an overpopulated world, the implications can only be deeply disquieting but before we face that, there is another surreal problem. Providing the pills are safe and have a plausible benefit – probably to a mouse – they may be on sale soon but nobody will honestly know if they have any significant impact on human ageing for one hundred years.