SCIENCE / Battle of the replicants: DNA may not, after all, be the only basis for life. Philip Ball reports on the artificial self- replicating molecules whose test-tube struggles could give new clues to how evolution began

FIRST there were the Original Replicants, who had learnt how to make copies of themselves just like the living organisms of Earth. Then came the Mutants, deformed by harsh ultraviolet rays, whose mutations made them more fit to fight for survival. Now there are the New Replicants, who have interbred with the Originals to form Hybrids which reproduce most prolifically of all.

These are not the characters of a science fiction novel, but the denizens of test-tubes in the laboratory of Dr Julius Rebek, a chemist at the Massachusetts Institute of Technology. They are not, moreover, living organisms but artificial molecules designed and constructed by Dr Rebek and his team. Nevertheless, in their ability to copy themselves - to replicate - Dr Rebek claims that these molecules are exhibiting 'a primitive form of life'.

If 'life' it is, then it is of a sort never seen before on Earth. All living organisms - bacteria, trees or humans - are based on the gene-bearing molecule deoxyribonucleic acid, known as DNA. But scientists such as Dr Rebek are now proposing that life doesn't have to be that way - that other forms of life are conceivable which don't need DNA. What's more, Richard Dawkins, a leading evolutionary biologist at Oxford, says these artificial replicating molecules 'raise the possibility of other worlds having a parallel evolution (to Earth's) but with a fundamentally different chemical basis'.

Scientists have debated the origins of life for decades and constructed several elaborate theories to explain how the simple chemicals in the 'primordial soup' of about four billion years ago eventually evolved into the complex life-forms we see today. The most intriguing and difficult chapter in the story is a mystery: how did replication of the primitive chemicals come about?

But there is more to life than reproduction: to be 'alive' an organism must be able to carry out additional functions such as metabolism - the processing of available raw materials to obtain the energy needed for development and growth - and self-repair. These are not things that an organism can learn from scratch; rather, it must be 'pre-programmed' with genetic information inherited from its progenitors. The DNA molecule plays a central role in life as we know it because of its apparently unique capability to store genetic information.

DNA replicates itself every time a cell divides, so that both new cells have identical copies of the DNA molecule. DNA does this by 'unzipping' itself so that its famous double helix unwinds into two strands; each acts as a template on which another strand forms. So the two original strands in the DNA double helix become four. Dr Rebek sees this template concept as the key to self-replicating molecules. But whereas the copying act of DNA has long been considered a remarkable feat, Dr Rebek suggests that replication in itself is no big deal.

All it takes, he says, is a pair of molecules that are 'complementary', meaning that they fit snugly together like pieces in a jigsaw puzzle. The two strands of the DNA double helix fit together in this way to form complementary pairs: each strand being a mirror image of the other. Dr Rebek believed that it should be possible to create such replicators in the laboratory by aligning complementary molecules.

His group demonstrated in 1990 that this was more than conjecture, when they reported the synthesis of a molecule that builds replicas of itself from a solution of its component parts. Earlier this year, Dr Rebek described new versions of his replicating molecule which have slightly different component parts from the earlier forms. These new replicators can be 'mutated' into yet another variety by shining ultraviolet light on them (UV light causes mutations in DNA in much the same way). The mutants can act as templates for assembly of both the unmutated and the mutant varieties. A fight for pre- eminence then ensues as the two replicators compete for the parts needed to make copies. This test-tube world in which the fittest struggle for survival and replication is an exact analogy of what took place billions of years ago when the first replicating molecules, presumably subjected to similar mutation from the sun's ultraviolet radiation, began competing with each other. In Dr Rebek's microcosm, he found that the mutants were 'fitter' and rapidly took over the available resources.

Whereas all of these replicators are variations on a theme, like different kinds of dog, Dr Rebek recently came across an entirely new 'species' of replicator. This bolsters his claim that a whole menagerie of these 'almost living' systems can be devised. When Dr Rebek tried cross-breeding the old replicators with the new, he obtained two very different types of half-breed: one was the best replicator so far, while the other was unable to reproduce at all. The sterility of the latter is the result of its unfortunate molecular shape, which prevents the molecule from being able to act as a template.

Dr Leslie Orgel, a biochemist at the Salk Institute, San Diego, envisages that these artificial replicating systems might generate a new kind of chemistry in which the desired properties of the products are achieved not by the ingenuity of chemists but by the forces of natural selection and evolution. Better plastics and drugs might be made by basing them on replicators which battle it out until the best design wins.

Where Dr Rebek's work may be of most immediate value, however, is in providing clues about how life on Earth began. 'Once', he says, 'there was physics and chemistry, but there was no biology.' He hopes that his molecules will improve our understanding of how chemistry gave birth to biology in the form of primitive replicating molecules with the capacity to undergo evolution.

Previously, all we knew about replicators was based on DNA and a near relative called RNA. In 1989, the American biochemists Thomas Cech and Sidney Altman won a Nobel prize for their discovery that RNA has the ability to assist in its own replication. Because RNA is a considerably simpler molecule than DNA, this led researchers to suggest that a form of replicating RNA preceded DNA at the beginning of life on Earth. But even RNA is immensely sophisticated by any standards - expecting it to have formed spontaneously from its component building blocks is akin to trying to make a radio by shaking together a box full of transistors.

The problem is that the biochemical synthesis of RNA requires a host of protein enzymes, to organise and direct the assembly process; yet proteins cannot themselves be made without RNA. This chicken-

and-egg puzzle has presented origin-of-life researchers with a very sticky problem. Dr Orgel and his colleague Dr Gerald Joyce, of the Research Institute of Scripps Clinic in California, have shown that under certain conditions random, short chains of RNA can act as templates on which new chains can be assembled without enzymes to trigger the process.

What Rebek's replicators now demonstrate is that even very simple molecules can develop the ability not only to replicate but - most importantly - to evolve. If mutations can enhance a replicator's survival chances, these systems have the opportunity to become increasingly complex and 'clever' through the effects of random mutation accompanied by natural selection.

Rebek's 'creatures' have a long way to go, however, before they can be considered to mimic anything resembling real life. What they currently lack, Dr Joyce says, is the ability to carry genetic information and pass it on accurately from generation to generation. But Joyce and Orgel predict that artificial replicators which can carry information will be produced by the end of the decade.