Tricks to make DNA beget DNA.

For scientists interested in how life came about, the chicken-and-egg controversy boils down to a question of molecular replication. Modern DNA molecules -- the stuff of genes -- encode information about other molecules, including enzymes that enable DNA to replicate, mutate, and evolve as conditions change. But how did DNA -- or perhaps RNA -- replicate before there were enzymes?

Several research groups already have mimicked many of the necessary steps for molecular evolution (SN: 8/7/93, p.91) in their attempt to re-create conditions leading to the origin of life. But in their experiments they make new copies of these molecules articifically, with enzyme helping.

Now, two groups have tricked small pieces of DNA into making copies by themselves, without enzymatic assistance. Both teams report their results in the May 19 NATURE.

As a result of this work, "We are a step closer to understanding possible pathways to life," comments James Ferris of Rensselaer Polytechnic Institution in Troy, N.Y.

DNA and RNA are made up of long chains of nucleotides. In cells, each link in the chain readily pairs off with its complement: puries attach to pyrimidines and vice versa.

These connections give rise to DNA's typical structure -- a double-stranded helix -- which enzymes help split apart during cell division. The newly created single strands then act as templates. Each nucleotide seeks out a new partner, and these partners align to form a complementary strand, thereby creating two new double helices.

In test tubes, single purine nucleotides readily assemble on a pyrimidine template, but the reverse doesn't occur, replication comes to a halt with mixed template. Also, even when scientists could get molecules to replicate, those molecules could not make copies of their complements.

However, using DNA fragments with three nucleotides overcomes this obstacle, leading to the formation of complements on an ongoing basis, says Gunther von Kiedrowski from Albert-Ludwigs University in Freiburg, Germany.

For their experiments, von Kiedrowski and a colleague put nucleotide threesomes into a solution that also contained a six-nucleotide strand. The matching threesomes then lined up to make a complementary six-nucleotide strand. This strand, too, began serving as a template for new strands.

Von Kiedrowski thinks that life's earliest molecules arose when small DNA fragments came together and served as templates for longer ones. Such fragments could have formed on clay substrates, adds Ferris.

Bigger nucleotide fragments also work, report Tianhu Li and Kyriacou C. Nicolaou, chemists at the Scripps Research Institute in La Jolla, Calif. They started with a palindromic sequence of 24 nucleotides: The order of purines and pyrimidines reads the same from either end of the strand.

In a slightly acidic solution, a double-stranded DNA fragment attracted two shorter, 12-nucleotide fragments, which assembled into a third 24-nucleotide strand upon the addition of a chemical reagent, the scientists report. Making the test-tube solution less acidic or adding more of the 12-nucleotide fragments causes that third strand to separate from the original double strand and to act as a template for a second strand complementary to itself, they add.

"We're not saying that we've created life," says Nicolaou, "but this is perhaps the first example that molecules can replicate themselves without the help of any enzymes."

Living systems expand exponentially: Two DNA strands beget four, which beget eight, then 16, then 32, and so on. Chemical systems increase incrementally, from one to two to three copies and so on. These two processes yield molecules at an in-between rates, say the scientists.