Evolution at Warp Speed

The Faculty of 1000 is aWeb-based literature awareness tool published by BioMed Central. For more information visit www.facultyof1000.com. In Sol Spiegelman's classic "Extracellular Darwinian Experiment,"1 the developer of nucleic acid hybridization allowed a miniversion of the Qß genome that replicated with super efficiency to evolve in a test tube. While in vitro evolution has blossomed in the intervening 35 years, one breakthrough in particular has hastened discovery, exemplified by two recent papers.2,3 Niles Lehman, associate professor of chemistry, Portland State University, Ore., says the early evolution experiments had no amplification methods. But now, Lehman says, "PCR is king. It allows amplification of rare sequences out of a large pool." Two groups, both at the Scripps Research Institute, evolved molecules that can repeatedly perform novel catalytic functions. In the past, researchers evolved molecules that perform a single, novel act, such as adding one nucleotide onto a substrate. This discovery brings investigators closer to creating enzymes that will have utility beyond proof of principal, while recreating what might have been a crucial step in the history of life on Earth. Researchers have now "cut across 3.6 billion years of evolutionary history," says Lehman. As part of a project to expand the genetic code, Peter Schultz, Floyd Romesberg, and colleagues evolved a DNA polymerase that can incorporate ribonucleotides. Creating new bases--decorating them with various functional groups--has been relatively easy, but for the bases to be useful, polymerases must be able to work with them. According to Romesberg, getting to that cooperative point was a stumbling block. "We've had trouble designing bases that allow continued extension. [So] we asked 'What else could we do.'" Nature, he says, has two ways to approach this point: from the base side and from the polymerase itself. So investigators decided to copy nature and evolve polymerases that efficiently incorporate unnatural base pairs. With molecular evolution, the trick is finding the molecule with the desired properties. Romesberg and coworkers came up with a modified phage display system that selects according to function rather than binding properties. The team co-displayed a collection of mutated DNA polymerases and a substrate on the same phage particle. By sticking the substrate right in the face of the polymerase, Romesberg says, a reaction between polymerase and substrate on the same phage is favored over reactions between phage. Adding biotinylated ribonucleotides to the mix provided a hook for pulling out those phage with the polymerases incorporating ribonucleotides. Using this approach, Romesberg not only evolved DNA polymerases with expanded functionality, but also provided new insights into how polymerases maintain selectivity for particular bases. His evolved polymerases replaced the glutamate at position 615 with glycine, which, Romesberg hypothesizes, creates the space for ribonucleotides (which are bulkier than deoxyribonucleotides) where it might not have existed before. In the second paper, Gerald Joyce and colleagues developed a system for continuous evolution,3 or, as Joyce says, "pedal-to-the-metal, test tube evolution." Unlike the more typical step-wise approach, Joyce's system links new functionality to selection and amplification, rather than manually selecting and propagating desired molecules between rounds of evolution. This system is closer to capturing the look and feel of Darwinian evolution as it occurs in nature, Joyce says. In addition, it adds a multiplier not present in step-wise approaches. It is not just having the right phenotype that makes it reproduce; it is the quality of the phenotype. "The better the phenotype, the faster you march through generations." This work harkens back to Spiegelman's Qß experiment, which was the first and only other continuous in vitro evolution system. The difference, though, is that in 1967, catalytic RNA had not been discovered, leaving Spiegelman to tinker only with replication. Now, catalysis has been added to the mix, and Joyce and others have evolved ribozymes that initially performed a single joining reaction,4 and now ribozymes that do three or four additions. Joyce says that this step lends further credence to the "RNA World" hypotheses, which says that RNA was the first informational macromolecule. What has been missing up until now is an RNA molecule that can catalyze its own polymerization. While Joyce's system is not that advanced (a protein polymerase amplifies the ribozymes that perform nucleotide additions), having a ribozyme that can do multiple nucleotide additions brings him one step closer. Achieving the complexity of biochemistry in nature is still a distant goal, but continuous evolution, which runs on its own, compressing hundreds of rounds of evolution into a few hours, may provide the best shot. Says Joyce, "All of directed evolution is giving up control to the large numbers [of molecules] in the test tube and hoping there's a good answer out there. With continuous evolution, it's that in spades." Laura DeFrancesco is a freelance writer in Pasadena, Calif. References 1. D.R. Mills et al., "An extracellular Darwinian experiment with a self-duplicating nucleic acid molecule," Proceedings of the National Academy of Sciences ( ), 58:217-24, 1967. 2. G.X. Zia et al., "Directed evolution of novel polymerase activities: Mutation of a DNA polymerase into an efficient RNA polymerase," PNAS, 99:6597-602, May 14, 2002. 3. K.E. McGinness et al., "Continuous in vitro evolution of a ribozyme that catalyzes three successive nucleotidyl addition reactions," Chemistry and Biology, 9:585-96, May 2002. 4. M.C. Wright et al., "Continuous in vitro evolution of catalytic function," Science, 276:614-7, 1997.

Evolution at Warp Speed

Interested in reading more?

Become a Member of