Chasing The Seeds Of Life : 13.7: Cosmos And Culture How did life originate? This seemingly eternal question was recently the focus of an unusual gathering at CERN in Switzerland. Commentator Stuart Kauffman was at the center of the action and takes us on a journey through the ideas that led up to this meeting of the minds.

Chasing The Seeds Of Life

A digital representation of the human genome at the American Museum of Natural History in New York City. Mario Tama/Getty Images hide caption

toggle caption
Mario Tama/Getty Images

A digital representation of the human genome at the American Museum of Natural History in New York City.

Mario Tama/Getty Images

This past February, I took part in a meeting at CERN to discuss and debate the origin of life. Organized by Günter von Kiedrowski and Eors Szathmary, it is possible that much may come of it. But first, let's start with a little of the history that led up to this moment.

Until Louis Pasteur, there was no origin of life problem: maggots just sprang spontaneously from dank wood after every rain. Pasteur showed us that life only comes from life. But where did life come from in the first place?

The problem rested until the early 20th century, when the concept emerged that a "primitive soup" of organic molecules had given birth to life.

The field leapt forward in the famous experiments of Stanley Miller. He showed that a retort filled with the gases presumed to have been present in the primitive Earth's atmosphere could produce amino acids, the building blocks of proteins, when stimulated by electric sparks that mimicked lightning. For some 40 years since, work along these lines has repeatedly demonstrated the prebiotic synthesis of many of life's organic molecules.

Following the discovery of the famous double-helix structure of DNA, and its cousin RNA, many researchers — Leslie Orgel, among them — adopted the view that molecular reproduction must be based on what is called "template replication" of single-stranded RNA polymers (polymers are made of many linked nucleotide monomers), or its cousins.

Here the "Watson" single-stranded polymer of a Watson-Crick double-stranded RNA helix, was to line up the free building blocks of RNA, A, U, C, G, the nucleotides, each lined up by the Watson-strand polymer, say AAUUCCGG with free U, U, A, A, G, G, C, C "base paired" to the Watson template in proper order. Then, without an enzyme, these free nucleotides were to be linked together to create the second, Crick strand. Then the two strands were to melt apart, i.e., "unzip", creating two single-stranded RNA template sequences, and this replicative process was to iterate to create a growing population of Watson-and-Crick strands.

In roughly 60 years of work since, efforts to confirm template replication have persistently failed. But it may yet succeed.

The next, now dominant, origin-of-life theory is the "RNA world." It was discovered that, in addition to proteins that could act as enzymes and catalyze, or speed up, chemical reactions, RNA molecules called ribozymes could do so as well. Here was the dream: the same class of molecules that carry genetic information, e.g., messenger RNA, could catalyze reactions. Perhaps one class of molecules, RNA, could both perform template replication and carry genetic information.

The RNA world view has morphed in at least two ways. First people have tried to evolve a ribozyme that can function as an enzyme able to link the free nucleotides above together, a so called "ribozyme polymerase." This approach has seen a bit of success, but seems stalled.

The second is the growing view of a "dirty RNA world" that might include amino acids and short sequences of linked amino acids called peptides. There is evidence that forming short peptides without enzymes is relatively easy.

In summary of the RNA world, so far no one has achieved template replication of RNA or its cousins. But they may well do so in the future.

In opposition to the view that life must be based on template replication of DNA, RNA, or similar polymers, is the idea that life can instead be based on a single molecule that catalyzes its own formation from two or more building blocks, e.g., A + B - > C, with C catalyzing the conversion of A + B to more C.

Much more generally, with a set of polymers, or other molecules, each might catalyze the formation of the next polymer or molecule from its fragments, around a "circle" of such polymers or molecules. Such a system would be "collectively autocatalytic" in that no molecule catalyzes its own formation, but the formation of another member of the set. The set as a whole catalyzes its formation from the building blocks of its members.

This approach is making considerable progress. First, firm theory and theorems have recently been developed to show that such collectively autocatalytic sets are expected to emerge spontaneously in sufficiently diverse chemical-reaction soups.

Experimentally, von Kiedrowski demonstrated that a single DNA molecule, six nucleotides long, can bind two "trimers" and glue them together into a second copy of itself. Done in 1987, this was the first molecular reproducing system. Note that this is not template replication, nucleotide by nucleotide. The DNA hexamer is a simple "ligase" enzyme, ligating the two preformed trimers. Shortly thereafter, von Kiedrowski made the first collectively autocatalytic DNA set. Here hexamer A catalyzed the formation of hexamer B from two B fragments, and B catalyzed the formation of A from two A fragments.

Remarkably, since then, peptide autocatalytic molecules and collectively autocatalytic sets have been created. One is a nine-peptide collectively autocatalytic set, busily reproducing. This result demonstrates conclusively that molecular reproduction need not be based on template replication.

Most recently, using a collection of evolved ribozymes — each cut in half and placed in a magnesium-rich buffer — single autocatalytic ribozymes formed, then two-, three-, five- and seven-membered collectively autocatalytic sets, CAS, emerged spontaneously and the larger CAS outcompeted the single autocatalytic molecule.

The way is now open to see if, for example, unevolved collections of stochastic RNA sequences, peptide sequences or both can spontaneously form CAS. Then, if so, can we get them into budding lipid-bilayer vesicles called liposomes, forming protocells, perhaps able to gather energy and evolve. Limited evolvability has been demonstrated, unlimited evolvability may be possible. Or it may require DNA storage, like modern collectively autocatalytic cells.

The workshop at the CERN meeting focused attention on the metabolism first approach. Both it and the RNA world need exploration. The meeting ended with a proposal to get the research community organized behind a common effort, hopefully benefiting from the experience of CERN in fostering international collaboration.