Scientific Report 2006

Molecular Biology

Directed Evolution of Nucleic Acid Enzymes

G.F. Joyce, S.E. Hamilton, D.P. Horning, T.A. Jackson, G.C. Johns, B.J. Lam, B.M. Paegel, G.G. Springsteen, S.B. Voytek

All life known to exist on Earth today is based on DNA genomes and protein enzymes, but most likely it was preceded by a simpler form of life based on RNA. This earlier era is referred to as the “RNA world.” During that time, genetic information resided in the sequence of RNA molecules and phenotype was derived from the catalytic behavior of RNA. By studying the properties of RNA in the laboratory, especially with regard to the evolution of catalytic function, we can gain insight into the RNA world. In addition, we can develop novel nucleic acid enzymes that have applications in biology and medicine.

Converting an RNA Enzyme to a DNA Enzyme

The transfer of sequence information between 2 different classes of nucleic acid–like molecules, for example between RNA and DNA, is straightforward because it relies on the 1-to-1 correspondence of Watson-Crick pairing. Nearly 50 years ago, in articulating the central dogma of molecular biology, Francis Crick referred to this property as “sequentialization.” Sequentialization also applies to the transfer of information from RNA to protein via the genetic code. The transfer of function, however, is more difficult because function is an overall property of a macromolecule and cannot be conveyed in a sequential manner. There is no known example of an RNA enzyme that retains catalytic activity when prepared as the corresponding DNA molecule, and vice versa.

We used in vitro darwinian evolution to convert an RNA enzyme to a DNA enzyme of the same function, after the acquisition of a few critical mutations. The starting RNA had the ability to join 2 RNA substrates in a template-directed manner, with a catalytic rate of 0.14 min^–1 (Fig. 1A). A corresponding DNA molecule in which ribose was replaced by deoxyribose and uracil was replaced by thymine had no detectable activity. The DNA molecule was used as a starting point to generate trillions of randomized variants, which were selected for the ability to catalyze the RNA-joining reaction. After 10 generations of evolution, we obtained a population of DNA enzymes with the desired activity. A typical example contains 10 mutations relative to the starting sequence and has a catalytic rate of 0.052 min^–1 (Fig. 1B). When this DNA enzyme was prepared as the corresponding RNA enzyme, it had no detectable activity. Thus, the evolutionary transition from an RNA enzyme to a DNA enzyme represents a switch in the chemical basis of catalytic function.

Fig. 1. Composition of an RNA enzyme (A) and a DNA enzyme (B) related by evolutionary descent. Both enzymes contain about 50 nucleotides and catalyze the joining of 2 RNA substrates (S1 and S2). The evolved DNA enzyme contains 10 mutations relative to the starting RNA enzyme (highlighted with black circles).

Evolutionary pathways such as this one for conversion of an RNA enzyme to a DNA enzyme may exist between other classes of nucleic acid–like molecules. The RNA world may have been preceded by a simpler “pre-RNA world” based on a nucleic acid–like molecule that would have occurred more readily on the primitive Earth. Our findings suggest that the catalytic function of a pre-RNA molecule might have been transferred to a corresponding RNA enzyme through darwinian evolution.

Continuous Evolution of RNA Enzymes

Processes of darwinian evolution are fundamental to understanding biological form and function but are difficult to appreciate on the human timescale. During the past decade, we have developed methods for evolving molecules rapidly and under controlled laboratory conditions. One of the most powerful of these methods, and the one that most closely resembles biological evolution, is a system for the continuous in vitro evolution of RNA enzymes. It involves a population of RNA enzymes that catalyze an RNA-joining reaction. Any molecule in the population that performs the reaction becomes amplified to produce “progeny” molecules, which then have the opportunity to perform the reaction again. The entire process takes place within a common reaction mixture and can be continued indefinitely, so long as an adequate supply of reaction materials is maintained.

Recently, we developed a novel approach for the continuous evolution of RNA enzymes that uses microfluidic technology. With this approach, evolution can be carried out in an automated fashion under computer control, with continuous monitoring of the population size and precise control over critical parameters such as mutation frequency and selection pressure. We have used the microfluidic device to conduct evolution experiments, beginning with a reaction mixture containing about 1 billion RNA enzymes and carrying out repeated rounds of RNA catalysis and selective amplification in an automated fashion. The amount of RNA is monitored continuously by using a confocal laser fluorescence microscope. When a predetermined threshold concentration is reached, the computer initiates an automated dilution and provides a fresh supply of reagents. This process of selective amplification and dilution was carried out for 70 successive dilutions of 10-fold each during a period of 6.5 hours. The microfluidic system is now being used to address fundamental questions of macromolecular evolution, such as the role of genetic diversity in escaping evolutionary bottlenecks and the maximum frequency of mutation that can be tolerated by an evolving population.

Publications

Joyce, G.F., Orgel, L.E. Progress toward understanding the origin of the RNA world. In: The RNA World, 3rd ed. Gesteland, R.F., Cech, T.R., Atkins, J.F. (Eds.). Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 2006, p. 23.

Oberhuber, M., Joyce, G.F. A DNA-templated aldol reaction as a model for the formation of pentose sugars in the RNA world. Angew. Chem. Int. Ed. 44:7580, 2005.

Paul, N., Springsteen, G., Joyce, G.F. Conversion of a ribozyme to a deoxyribozyme through in vitro evolution. Chem. Biol. 13:329, 2006.