by Nick Lane
The stress response, too, is characteristic of virtually all organisms from bacteria to humans. We met a few of the proteins involved in Chapter 10. These proteins are equivalent to the emergency services, helping to repair damaged DNA, degrade damaged proteins and prevent free-radical chain reactions getting out of hand. Cells that successfully respond in this way to the altered conditions have a selective advantage.
They will survive and reproduce, while their less well endowed cousins, even those with extra chromosomes, are more likely to accumulate damage and finally die.
By subjecting a bacterial population to radiation, therefore, we will select, over many generations, for a population able to withstand the effects of radiation. Now imagine that we switch off the radiation source.
We have selected for a battle-scarred population of bacteria, laden with as much armour as a mediaeval knight. We drop them back into peace time.
Suddenly, the extra protection is an unnecessary burden, costly to repro-
220 • SEX AND THE ART OF BODILY MAINTENANCE
duce. Each time a stress-resistant cell reproduces itself, it must replicate multiple copies of its genes, and it is also funnelling a substantial proportion of its energy into the production of more stress proteins. Any bacterium that loses a few chromosomes and switches off its stress response is more likely to replicate quickly. Our armour-clad knight has to get rid of his armour before he can father children. In just a few generations, the stress-resistant bacteria may be but a distant memory. From an anthropocentric point of view, seeking purpose in nature, this endless cycling seems as blind as it is futile, but such is evolution. That is why bacteria are still bacteria.
What does all this have to do with ageing? Bacteria, for the most part, do not age. There is no reason for them to do so. They maintain the integrity of their genes by rapid reproduction. They can produce a new generation every 30 minutes. They protect themselves by hoarding multiple chromosomes, by exchanging genes through conjugation and lateral gene transfer (see Chapter 8, page 154), and by fixing damaged DNA wherever possible. Disabling errors in DNA are quickly eliminated by natural selection. Any selective advantages are embraced just as quickly. Bacteria have behaved in this way for nearly 4 billion years. Certainly they have evolved, and in this sense they have aged, but in every other respect they are as youthful now as they were all those countless generations ago.
The critical point is that the survival of bacterial life involves death on a massive scale. In 24 hours, a single bacterium can produce 248, or 1016, cells, with a total biomass of about 30 kilograms. Clearly, such exponential growth cannot be sustained. In most natural environments, the size of bacterial populations remains roughly constant. Bacteria die from prolonged starvation or dehydration, or are food for other organisms, such as nematode worms — or fail to divide because of cellular damage.
When death outweighs life on such a massive scale, natural selection cleanses the population of genetic damage. Only the fittest survive. Perverse as it may sound, the main criterion for immortality is death.
Shifting the perspective, we are left with the following conclusions. If senescence is not necessary to life as such, and has not always been with us, then presumably it has evolved. If senescence evolved, then it must be determined at least partly by genes, as only traits that are genetically determined can evolve and be passed on to the next generation. And if it has persisted, it must confer some kind of selective advantage.
Trade-offs in the Evolution of Ageing • 221
Ageing evolved at the same time as sex. By sex (I should come clean) I mean the production of sex cells such as sperm and egg and their fusion to form a new organism. The terms ‘sex’ and ‘reproduction’ are often used interchangeably, but technically they have completely different mean-ings. As John Maynard Smith and Eörs Szathmáry put it, “the sexual process is in fact the opposite of reproduction. In reproduction, one cell divides into two: in sex, two cells fuse to form one.” This poses a puzzle, which we shall see applies as much to ageing as to sex: what is the benefit to the individual?
Sex is thought to benefit populations in numerous ways. Perhaps the most important of these benefits is the swift dispersal of a new version of a gene through an entire population, thereby promoting genetic vari-ability. Most genes exist in several different forms, and sexual reproduction brings about new and continuously shifting combinations of these versions. This variety is clear in the human population, in which out of 6
billion people, you’d be hard put to find anyone genetically identical to you, apart from an identical twin. This is important because genetically variable populations are more adaptable to changing environmental conditions or selection pressures.
But these benefits can only be enjoyed after the evolution of sex, and as we saw in Chapter 7, we cannot use hindsight to explain evolution. For a trait to become established in a population, it must first be beneficial to individuals, who then thrive at the expense of other individuals lacking the trait. The advantage to individuals of recombining genes is not immediately obvious. In sexual reproduction, two robust individuals, who have succeeded against the odds in surviving until sexual maturity, and then mating, have their robust genetic constitutions shuffled and reconstituted into new combinations in the offspring that, in statistical terms, are likely to be less robust. The reasons for the evolution of sex are, in fact, still much debated, and we shall look into them in a moment. First, we should just note that the same considerations apply to senescence. If senescence evolved, then it must have been beneficial to individuals or it could never have become such an integral feature of a large part of the living world.
Senescence is, in fact, even more widespread than sex, affecting essentially all plants and animals, implying that the advantage must be very pervasive.
Because sex and senescence are closely linked, we will examine the problem of sex first, from the point of view of individual cells. The fundamental problem with sex is the rate of reproduction. If an asexual
222 • SEX AND THE ART OF BODILY MAINTENANCE
microorganism, such as a bacterium, reproduces by simply dividing into two ( binary fission), then one cell produces two, two cells produce four and so on. The population as a whole expands at an exponential rate. The rate of sexual propagation is necessarily much slower: two cells conjoin to produce one, and this must, at the very least, divide in two before it can produce daughter cells that can fuse with other cells to produce more offspring. Not only is the rate of reproduction slower, but the sex cells also have to find each other, and determine that they are right for each other, before they can fuse and then reproduce. The process is fraught with danger and is energetically costly. An asexual population should outnumber a sexual population in a handful of generations. Alternatively, if one individual in a sexual population were to revert to asexual reproduction, the asexual progeny should swiftly outnumber the sexual progeny. On the basis of simple arithmetic, sex should never have got started. Having evolved against all the odds, it ought to have been weeded out long ago.
Why did this not happen?
To resolve the dilemma, we must return to the central problem in biology: how to maintain the integrity of the genetic instructions from generation to generation. We have seen that bacteria succeed by combining a high rate of reproduction with heavy selection. The cost of genetic cleansing in asexual reproduction is thus a high death rate, which translates into a lot of pain for little gain. This might not worry the blind watchmaker (the evolutionary biologist Richard Dawkins’s famous coinage for natural selection), but it is wasteful of resources and so might cede advantage to more efficient ways of cleansing, or at least masking, genetic damage. Sex is certainly more efficient at masking damage. This is the basis of that mysterious property known as hybrid vigour in plant and animal breeding, in which the offspring of unrelated parents have qualities superior to either parent. Conversely, too much inbreeding has the opposite effect. To understand why, we need to conside
r the mechanics of sex, in particular the form of cell division known as meiosis, in a little more detail.
In most sexual species, including humans, the sex cells, or gametes, each contain half the genetic material of their progenitor cells. They are said to be haploid, meaning that they have a single set of chromosomes, selected more or less at random from the two parental sets. When the gametes fuse together, the two haploid cells each contribute a single set of chromosomes to the fertilized egg, thereby restoring the full complement of
Trade-offs in the Evolution of Ageing • 223
chromosomes. The fertilized egg thus has two equivalent sets of chromosomes, and so is diploid. Now if this cell were simply to fuse with another, the result would be tetraploid, with four sets of chromosomes. If we continued in this way, we would not have to wait long before the situation became untenable. The usual solution in sexual species is to regenerate more sex cells through meiosis. This type of cell division halves the number of chromosomes, to regenerate more haploid sex cells for the next generation.
As with all types of cell division, even where the ultimate aim is to halve the number of chromosomes, each chromosome is first replicated, with the two new daughter chromosomes remaining joined together.
Then, in the first step of meiosis, the twin sets of duplicated chromosomes pair up and are shuffled like a pack of cards. During this process, corresponding pieces of paired duplicated chromosomes are exchanged with each other. It is as if the top half of a red queen was grafted on to the bottom half of a black queen, to produce a mixed-queen card (or the trousers and jacket of two suits swapped around). This process is called recombination, and it means that the different versions of genes are rearranged in new combinations on a chromosome in the next generation. This is why it is possible for a child to resemble his great-grandfather even though nobody else has done in the intervening generations. In the next step, the pairs of chromosomes separate to produce two nuclei, which are still technically diploid. In the final step of meiosis, the daughter chromosomes of each duplicated chromosome are separated, thus finally producing four new haploid cells, all with different gene combinations, from the original diploid cell. (This is a very much simplified picture but conveys the essential concepts.)
There are, then, two cardinal features of organisms that undergo meiosis and sexual reproduction. First, genes and chromosomes from the two parents can be inherited in various combinations by the offspring.
Second, sexual reproduction results in organisms that alternate between haploid and diploid states, even if one or other of these states is much reduced in the life history.
The advantage of bringing chromosomes together from two different parents is fairly easy to see. When the two haploid cells fuse to produce a new organism, the two copies of every gene are derived from organisms with differing genetic inheritance and life histories. This means that any genetic damage or mutations are unlikely to overlap. Recombination randomizes this allocation process in the same way that shuffling a pack
224 • SEX AND THE ART OF BODILY MAINTENANCE
of cards equalizes the odds of a good hand. Any mutations in a gene from one partner will almost certainly be counterbalanced in the offspring by an unaltered copy of the same gene from the other partner. In the unlikely event that two bad copies of the same gene are inherited, the offspring is eliminated by natural selection, ridding the population of detrimental mutations in the same way as in bacteria. At an individual level, sex is thus a more efficient mechanism than binary fission for maintaining the integrity of the genetic instructions, without the need for quite such heavy mortality.
The advantage of cycling between haploid and diploid states is more enigmatic. Diploidy clearly makes sense. Diploid cells, with their two equivalent sets of chromosomes, are analogous to the stress-resistant bacteria that stockpile multiple sets of identical chromosomes. The diploid state is a trade-off between the cost of producing multiple chromosomes and the danger of only having one copy of a gene. With two equivalent chromosomes, DNA breaks or deletions affecting one chromosome can be repaired using the other as a template. But the haploid state is perplexing.
It is, in principle, both avoidable and dangerous, as cells could easily cycle between the diploid and tetraploid states (four chromosomes) with much less risk. More perplexing still is the fact that haploid cells are not uncommon in nature, and are by no means restricted to sex cells. Indeed, the males of many species of Hymenoptera, including some wasps, bees and ants, are entirely haploid — they develop from an unfertilized egg. In contrast, the females develop from a fertilized egg, and are diploid. This is no accident. In some circumstances, male haploidy is even maintained by behavioural adaptations. In the honey bee, for example, 8 per cent of newborn males are diploid, the rest being haploid. Within six hours of emergence, the worker bees actually find and eat all of the diploid males.
Why should this be? The answer is far from certain, but Wirt Atmar, an eminent computer scientist with a deep personal interest in ecology and animal behaviour, put forward a strong argument in the journal Animal Behaviour in 1991. His argument seems to have been inexplicably ignored by many molecular biologists. Perhaps too few read Animal Behaviour. Atmar argued that haploid males act as ‘auxiliary defect sieves’, to purge the population of genetic error by exposing latent gene defects to selection. In other words, because genetic errors cannot be concealed in haploid animals, any haploid males with demonstrable vigour must possess a near-perfect complement of genes. In this sense, haploid males are an
Trade-offs in the Evolution of Ageing • 225
extreme version of haploid sperm. But why go the whole hog and produce an entirely haploid male? Sperm, after all, are far from costly — a single millilitre of semen contains 100 million of them.
According to Atmar, the answer lies in the distinction between
‘housekeeping’ and ‘luxury’ genes. Housekeeping genes are concerned with the basic metabolism of cells, and so are active in virtually all cells, including sperm. Luxury genes, on the other hand, code for specialized proteins that are only produced in particular cells, such as haemoglobin in red blood cells in mammals. Defects in haemoglobin would certainly be detected in a haploid human — diseases like sickle-cell anaemia would be eliminated in next to no time — but do not affect haploid sperm, which do not need haemoglobin. Incidentally, Atmar also notes that men are haploid for the X and Y chromosomes, whereas women are diploid for the X chromosome. To the extent that only men suffer from genetic disorders caused by mutations on the Y or X chromosome, such as haemophilia and colour blindness, it may be that we too use a mild form of haploidy to clean up our germ line (the inheritable DNA in sex cells). Atmar goes on to argue that this mild form of haploidy is supplemented in most species by the evolution of aggressive male traits and high mortality rates — only the vigorous, dominant males survive to fertilize the females.
Let us accept, for the sake of argument, that haploid males serve as defect filters. Seen in this light, their existence explains some unexpected experimental data that put paid to a once popular theory of ageing — the so called ‘somatic mutation’ theory (from the ‘soma’, meaning the body).
According to this theory, ageing is caused by the accumulation of spontaneous mutations in somatic DNA during the lifetime of the animal, in much the same way that cancer is caused by spontaneous mutations. The theory is easily testable. If ageing is indeed caused by spontaneous mutations, and having two copies of the same gene masks any errors in one copy, then haploid animals should age more quickly than diploid animals, because they have only one copy of each gene. Further, irradiation should speed up ageing in haploid males more than in diploid females, because a single mutation would destroy function in a haploid animal, but not in a diploid animal. The idea was put to the test in wasps and found to be wanting. In fact, haploid males and diploid females live for similar periods. Although very high doses of radiation kill male wasps faster (as might be expected for an extre
me situation not at all representative of normal ageing), low doses of irradiation do not affect their rate of ageing.
226 • SEX AND THE ART OF BODILY MAINTENANCE
These results do not fulfil the predictions of the somatic mutation theory of ageing. Clearly, spontaneous mutations alone are not responsible for ageing in wasps. From an evolutionary point of view, perhaps this result should have been expected. If the function of haploid males is to eliminate defects in the germ line, then the vigorous males must be robust enough to survive long enough to pass on essentially faultless DNA to the next generation. This implies that the rate of spontaneous mutations cannot be cripplingly high, otherwise the haploid males would be unable to pass on anything approaching ‘faultless’ DNA. Most subsequent studies have confirmed that the rate of spontaneous mutation is not sufficiently high to be the sole cause of ageing in most individuals (although mutations almost certainly contribute).
Sex cleans up the germ line by recombining DNA from different sources, and by directing the full force of natural selection at a section of the population — the haploid sex cells — rather than the whole population. In humans, an average ejaculation contains several hundred million sperm, which are subject to intense competition to fertilize the egg. Perhaps a couple of thousand sperm eventually reach the vicinity of the egg, so 99.9999 per cent perish en route — this is natural selection on a par with bacteria. Inherent in this idea of sex is the concept of redundancy. To preserve an uncorrupted germ line, a redundant part of the population is held out for selection, and only the best bits are ploughed back into the germ line. It seems that the nagging feeling of redundancy experienced by many men runs deep. This is not all. Redundancy is also behind the differentiation of cells into germ-line cells and somatic cells.