Looking back from the present, the genome seems immortal. An unbroken chain of descent links the very first ur-gene with the genes active in your body now — an unbroken chain of perhaps fifty billion copyings over four billion years. There were no breaks or fatal mistakes along the way. But past immortality, a financial adviser might say, is no guarantee of future immortality. Becoming an ancestor is difficult — indeed, natural selection requires it to be difficult.
If it were easy, the competitive edge that causes adaptive evolution would be lost. Even if the human race survives another million years, many of those alive today will contribute no genes to those alive a million years hence: their particular descendants will peter out in childlessness. And if the human race does not survive (most species last only about ten million years and most leave no descendant species behind: we've done five million years and spawned no daughter species so far), none of us alive today will contribute 1 9 6 G E N O M E
anything genetic to the future. Yet so long as the earth exists in something like its present state, some creature somewhere will be an ancestor of future species and the immortal chain will continue.
If the genome is immortal, why does the body die? Four billion years of continuous photocopying has not dulled the message in your genes (partly because it is digital), yet the human skin gradually loses its elasticity as we age. It takes fewer than fifty cell doublings to make a body from a fertilised egg and only a few hundred more to keep the skin in good repair. There is an old story of a king who promised to reward a mathematician for some service with anything he wanted. The mathematician asked for a chessboard with one grain of rice on the first square, two on the second, four on the third, eight on the fourth and so on. By the sixty-fourth square, he would need nearly twenty million million million grains of rice, an impossibly vast number. Thus it is with the human body. The egg divides once, then each daughter cell divides again, and so on. In just forty-seven doublings, the resulting body has more than 100
trillion cells. Because some cells cease doubling early and others continue, many tissues are created by more than fifty doublings, and because some tissues continue repairing themselves throughout life, certain cell lines may have doubled several hundred times during a long life. That means their chromosomes have been 'photocopied'
several hundred times, enough to blur the message they contain.
Yet fifty billion copyings since the dawn of life did not blur the genes you inherited. What is the difference?
Part of the answer lies on chromosome 14, in the shape of a gene called
The story starts with a chance observation in 1972 by James Watson, D N A ' s co-discoverer. Watson noticed that the biochemical machines that copy D N A , called polymerases, cannot start at the very tip of a D N A strand. They need to start several 'words' into the text. Therefore the text gets a little shorter every time it is I M M O R T A L I T Y 1 9 7
duplicated. Imagine a photocopier that makes perfect copies of your text but always starts with the second line of each page and ends with the penultimate line. The way to cope with such a maddening machine would be to start and end each page with a line of repeated nonsense that you do not mind losing. This is exactly what chromosomes do. Each chromosome is just a giant, supercoiled, foot-long D N A molecule, so it can all be copied except the very tip of each end. And at the end of the chromosome there occurs a repeated stretch of meaningless 'text': the 'word' T T A G G G repeated again and again about two thousand times. This stretch of terminal tedium is known as a telomere. Its presence enables the DNA-copying devices to get started without cutting short any sense-containing
'text'. Like an aglet, the little plastic bit on the end of a shoelace, it stops the end of the chromosome from fraying.
But every time the chromosome is copied, a little bit of the telomere is left off. After a few hundred copyings, the chromosome is getting so short at the end that meaningful genes are in danger of being left off. In your body the telomeres are shortening at the rate of about thirty-one 'letters' a year - more in some tissues. That is why cells grow old and cease to thrive beyond a certain age. It may be why bodies, too, grow old — though there is fierce disagreement on this point. In an eighty-year-old person, telomeres are on average about five-eighths as long as they were at birth.1
The reason that genes do not get left off in egg cells and sperm cells, the direct ancestors of the next generation, is the presence of telomerase, whose job is to repair the frayed ends of chromosomes, re-lengthening the telomeres. Telomerase, discovered in 1984 by Carol Greider and Elizabeth Blackburn, is a curious beast. It contains R N A , which it uses as a template from which to rebuild telomeres, and its protein component bears a striking resemblance to reverse transcriptase, the enzyme that makes retroviruses and transposons multiply within the genome (see the chapter on chromosome 8).
Some think it is the ancestor of all retroviruses and transposons, the original inventor of R N A - t o - D N A transcription. Some think that because it uses R N A , it is a relic of the ancient R N A world.
1 9 8 G E N O M E
In this context, note that the 'phrase' T T A G G G , which is repeated a few thousand times in each telomere, is exactly the same in the telomeres of all mammals. Indeed, it is the same in most animals, and even in protozoans, such as the trypanosome that causes sleeping sickness, and in fungi such as
Perhaps ironically, the complete telomerase machine has been isolated only in ciliates, not in human beings. We do not yet know for sure what proteins are brought together to make up human telomerase and it may prove very different from that in ciliates.
Some sceptics refer to telomerase as 'that mythical enzyme', because it is so hard to find in human cells. In ciliates, which keep their working genes in thousands of tiny chromosomes each capped with two telomeres, telomerase is much easier to find. But by searching a library of mouse D N A for sequences that resemble those used in the ciliate telomerase, a group of Canadian scientists found a mouse gene that resembled one of the ciliate genes; they then quickly found a human gene that matched the mouse gene. A team of Japanese scientists mapped the gene to chromosome 14; it produces a protein with the grand, if uncertain title of telomerase-associated I M M
