New Way to Study the Origins of Organic Life

Thomas Edison lit up his lab with thousands of glowing filaments to quickly locate the materials that worked for light bulbs. Now a modern form of shotgun testing, known as combinatorial testing, helps chemists sort through trillions of molecules to find the few that work for them.

That's how Ronald Breaker came up with the kind of molecules he thinks may have helped organic life arise on Earth. What's more, he expects to shape such molecules into powerful probes to detect elusive toxins in our environment, neutralize harmful viruses, or remove or replace defective genes in an organism's genetic code. "We want to understand the basic biology. But we keep an eye on practical applications," he says.

Now widely used in drug development, combinatorial testing is spreading. Dr. Breaker says that "without question, it will revolutionize" biochemistry, materials science, and many other fields.

In his Yale University laboratory in New Haven, Conn., Breaker is pursuing the fundamental which-came-first question of biology. Earthly life now depends essentially on three types of molecules. DNA carries an organism's genetic blueprint. The closely related RNA reads the DNA instructions and tells the machinery of living cells what proteins to make and when to make them. This third type of molecule - the proteins - enables living cells to do their job. But which came first when Earth's life arose nearly 4 billion years ago - DNA, RNA, or proteins?

DNA and RNA, now need the help of proteins to assemble themselves or rearrange their parts. Proteins carry the chemical "scissors" that snip DNA and RNA molecules.

Proteins also carry the "sewing" machinery that stitches bits of DNA or RNA together. However, biochemists have shown that RNA could have done that by itself. So RNA may have come first.

Breaker and co-worker Alan Roth look forward to making what he calls "highly sophisticated DNA and RNA enzymes," using the various amino acids that give proteins their chemical potency. That's how they hope to make enzymes that can take apart harmful viruses and do other useful work.

The trick is to find the DNA or RNA fragments that combine with the desired amino acids. It's easy to make trillions of random nucleic acid fragments with a computer-controlled DNA synthesizer.

Trying to test these one by one would be even more futile than Edison trying to test thousands of electric bulb filaments one at a time. That's where shotgun testing - or as chemists call it, "combinatorial chemistry" - comes in. Breaker and his colleagues spread DNA fragments on a grid and wash them with the desired amino acid. The ones that combine with the target amino acid wash away and are recovered. The duds are left behind.

Breaker says that being able to test a wide range of chemicals in one go gives unexpected insights: "Sometimes we get exactly what we're looking for.... Sometimes we find out we don't know as much as we need to know."