RNA structure. Ribozyme evolution at the crossroads.

Molecules that have similar sequences usually adopt the same structures and have the same functions. In his Perspective, Joyce explains that this is not always the case. In a remarkable study (Schultes and Bartel), an RNA sequence has been designed that can adopt two different structures, each with a different catalytic function. Joyce details how this study sheds light on the evolution of enzymes.

Directed evolution of an RNA enzyme.

An in vitro evolution procedure was used to obtain RNA enzymes with a particular catalytic function. A population of 10(13) variants of the Tetrahymena ribozyme, a group I ribozyme that catalyzes sequence-specific cleavage of RNA via a phosphoester transfer mechanism, was generated. This enzyme has a limited ability to cleave DNA under conditions of high temperature or high MgCl2 concentration, or both. A selection constraint was imposed on the population of ribozyme variants such that only those individuals that carried out DNA cleavage under physiologic conditions were amplified to produce "progeny" ribozymes. Mutations were introduced during amplification to maintain heterogeneity in the population. This process was repeated for ten successive generations, resulting in enhanced (100 times) DNA cleavage activity.

Continuous in vitro evolution of catalytic function.

A population of RNA molecules that catalyze the template-directed ligation of RNA substrates was made to evolve in a continuous manner in the test tube. A simple serial transfer procedure was used to achieve approximately 300 successive rounds of catalysis and selective amplification in 52 hours. During this time, the population size was maintained against an overall dilution of 3 x 10(298). Both the catalytic rate and amplification rate of the RNAs improved substantially as a consequence of mutations that accumulated during the evolution process. Continuous in vitro evolution makes it possible to maintain laboratory "cultures" of catalytic molecules that can be perpetuated indefinitely.

Cleavage of an amide bond by a ribozyme.

A variant form of a group I ribozyme, optimized by in vitro evolution for its ability to catalyze magnesium-dependent phosphoester transfer reactions involving DNA substrates, also catalyzes the cleavage of an unactivated alkyl amide when that linkage is presented in the context of an oligodeoxynucleotide analog. Substrates containing an amide bond that joins either two DNA oligos, or a DNA oligo and a short peptide, are cleaved in a magnesium-dependent fashion to generate the expected products. The first-order rate constant, kcat, is 0.1 x 10(-5) min-1 to 1 x 10(-5) min-1 for the DNA-flanked substrates, which corresponds to a rate acceleration of more than 10(3) as compared with the uncatalyzed reaction.

Climbing Darwin's ladder.

NASA: The work of Bartel and Szostak, in which RNA molecules were selected to enhance the ability to catalyze a reaction similar to a step in protein-catalyzed RNA replication, is discussed. An important aspect of this experiment was the ability to reach a high level of functional organization in ten evolutionary steps. Further steps necessary to obtain an RNA enzyme with RNA replicase activity include performing the reaction with mononucleoside 5'-triphosphates, generalizing the reaction to include a variety of sequences without loss of template-dependent specificity, and overcoming template self-structure that could prevent some regions from being copied efficiently.^ieng

Building the RNA world. Ribozymes.

The isolation of an RNA enzyme with RNA replicase activity, at present only a hypothetical molecule, is considerably closer following the recent demonstration of RNA-catalyzed polymerization of nucleoside triphosphates.

Evolution in vitro: analysis of a lineage of ribozymes.

BACKGROUND: Catalytic RNAs, or ribozymes, possessing both a genotype and a phenotype, are ideal molecules for evolution experiments in vitro. A large, heterogeneous pool of RNAs can be subjected to multiple rounds of selection, amplification and mutation, leading to the development of variants that have some desired phenotype. Such experiments allow the investigator to correlate specific genetic changes with quantifiable alterations of the catalytic properties of the RNA. In addition, patterns of evolutionary change can be discerned through a detailed examination of the genotypic composition of the evolving RNA population. RESULTS: Beginning with a pool of 10(13) variants of the Tetrahymena ribozyme, we carried out in vitro evolution experiments that led to the generation of ribozymes with the ability to cleave an RNA substrate in the presence of Ca2+ ions, an activity that does not exist for the wild-type molecule. Over the course of 12 generations, a seven-error variant emerged that has substantial Ca(2+)-dependent RNA-cleavage activity. Advantageous mutations increased in frequency in the population according to three distinct dynamics--logarithmic, linear and transient. Through a comparative analysis of 31 individual variants, we infer how certain mutations influence the catalytic properties of the ribozyme. CONCLUSIONS: In vitro evolution experiments make it possible to elucidate important aspects of both evolutionary biology and structural biochemistry on a reasonable short time scale.

In vitro evolution of nucleic acids.

NASA: The author reviews recent published reports of in vitro selection and evolution of nucleic acids. These nucleic acids will bind to a target ligand or catalyze a specific chemical reaction. The terms aptamers and systematic evolution of ligands by exponential enrichment (SELEX) are explained. The review focuses on protein binders, small molecule binders, and ribozymes obtained by directed evolution. The reference list identifies articles of special or outstanding interest.^ieng

Non-enzymatic template-directed synthesis on RNA random copolymers. Poly(C,A) templates.

Poly(C,A) random copolymer templates direct the oligomerization of 2-MeImpG (2-MeImpX is the 5'-phospho-2-methylimidazolide of the nucleoside X) and 2-MeImpU, resulting in the production of a variety of oligo (G,U)s. This reaction is less efficient than comparable reactions involving poly(C,U) or poly(C,G) templates. The efficiency of monomer incorporation into newly synthesized oligomers is lower for 2-MeImpU than 2-MeImpG, and cannot be improved by increasing the concentration of 2-MeImpU relative to 2-MeImpG. This suggests that RNA templates containing runs of consecutive adenine residues would not be suitable for use in a chemical self-replicating system. The distribution of oligomeric products can be characterized in detail using high-pressure liquid chromatography on an RPC-5 column. Oligomers are separated on the basis of chain length, base composition, and phosphodiester-linkage isomerism. Oligomers up to about the 13-mer, with base composition Gn, Gn-1, U, and Gn-2, U2, have been identified.

Non-enzymic template-directed synthesis on RNA random copolymers. Poly(C, G) templates.

Poly(C, G) random copolymer templates direct the oligomerization of 2-Me-ImpG and 2-MeImpC, resulting in the production of a variety of oligo(G, C)s. The efficiency of monomer incorporation into newly synthesized oligomers is greater for 2-MeImpG than for 2-MeImpC, and decreases for both monomers as the guanine content of the template increases. The relatively low efficiency of oligomerization on guanine-rich templates is largely a consequence of intra- and intermolecular template self-structure. The problem of template self-structure is clearly a major obstacle to the development of a system of self-replicating polynucleotides. The distribution of oligomeric products can be characterized in detail using high-pressure liquid chromatography on an RPC-5 column. Oligomers are separated on the basis of chain length, base composition and phosphodiester-linkage isomerism. Oligomers up to about the 12-mer, with base composition Gn, Gn-1C and Gn-2C2, have been identified. The 3' to 5' regiospecificity of the products is high, particularly for oligomers with base composition Gn.

Specialization of the DNA-cleaving activity of a group I ribozyme through in vitro evolution.

In an earlier study, an in vitro evolution procedure was applied to a large population of variants of the Tetrahymena group I ribozyme to obtain individuals with a 10(5)-fold improved ability to cleave a target single-stranded DNA substrate under simulated physiological conditions. The evolved ribozymes also showed a twofold improvement, compared to the wild-type, in their ability to cleave a single-stranded RNA substrate. Here, we report continuation of the in vitro evolution process using a new selection strategy to achieve both enhanced DNA and diminished RNA-cleavage activity. Our strategy combines a positive selection for DNA cleavage with a negative selection against RNA binding. After 36 "generations" of in vitro evolution, the evolved population showed an approximately 100-fold increase in the ratio of DNA to RNA-cleavage activity. Site-directed mutagenesis experiments confirmed the selective advantage of two covarying mutations within the catalytic core of the ribozyme that are largely responsible for this modified behavior. The population of ribozymes has now undergone a total of 63 successive generations of evolution, resulting in an average of 28 mutations relative to the wild-type that are responsible for the altered phenotype.

Non-enzymatic template-directed synthesis on RNA random copolymers. Poly(C, U) templates.

Poly(C, U) random copolymer templates direct the oligomerization of 2-MeImpG and 2-MeImpA, resulting in the production of a variety of oligo/(G,A)s. The efficiency of monomer incorporation into newly synthesized oligomers is greater for 2-MeImpG than for 2-MeImpA, and decreases for both monomers as the uracil content of the template increases. The relatively poor incorporation of adenine is partly due to an intrinsically less efficient incorporation reaction, and partly due to the masking of uracil sites by G X U non-complementary pairing. The efficiency of adenine incorporation can be improved by decreasing the concentration of 2-MeImpG and increasing the concentration of 2-MeImpA in the reaction mixture. The oligomeric product distribution can be characterized in detail using high-pressure liquid chromatography on an RPC-5 column. Oligomers are separated on the basis of chain length, base composition, and phospho-diester-linkage isomerism. The 3'----5' regiospecificity of monomer addition to template-bound oligomers is lower for 2-MeImpA than for 2-MeImpG. The presence of an adenine residue at the 2'(3') terminus of the acceptor strand lowers the regiospecificity of 2-MeImpA addition even further.

Alternative conformations of a nucleic acid four-way junction.

A crystal structure of a 108 nucleotide RNA-DNA complex containing a four-way junction was solved at 3.1 A resolution. The structure of the junction differs substantially from the "stacked-X" conformation observed previously, due to a 135 degrees rotation of the branches. Comparison of the two conformers provides insight into the factors contributing to the flexibility of four-way junctions. The stacked-X conformation maximizes base-stacking but causes unfavorable repulsion between phosphate groups, whereas the 135 degrees -rotated "crossed" conformation minimizes electrostatic clashes at the expense of reduced base-stacking. Despite the large rotation of the branches, both junction structures exhibit an antiparallel arrangement of the continuous strands and opposite polarity of the crossover strands.

Template-directed synthesis on the pentanucleotide CpCpGpCpC.

The pentanucleotide CpCpGpCpC facilitates the synthesis of oligomers containing G and C from a mixture of the two activated mononucleotides (guanosine 5'-phosphor)-2-methylimidazolide and (cytidine 5'-phosphor)-2-methylimidazolide. The major pentameric product of the template-directed reaction is all 3' to 5'-linked and has the sequence pGpGpCpGpG, which is complementary to that of the template. It can be obtained in a yield of up to 17%, based on the input of the template. The 3' to 5' isomer of GpG is elongated on the template to give GpGpC, GpGpCpG and GpGpCpGpG, while the 2' to 5' isomer does not initiate the synthesis of detectable amounts of longer oligomers.

A molecular description of the evolution of resistance.

BACKGROUND: In vitro evolution has been used to obtain nucleic acid molecules with interesting functional properties. The evolution process usually is carried out in a stepwise manner, involving successive rounds of selection, amplification and mutation. Recently, a continuous in vitro evolution system was devised for RNAs that catalyze the ligation of oligonucleotide substrates, allowing the evolution of catalytic function to be studied in real time. RESULTS: Continuous in vitro evolution of an RNA ligase ribozyme was carried out in the presence of a DNA enzyme that was capable of cleaving, and thereby inactivating, the ribozyme. The DNA concentration was increased steadily over 33.5 hours of evolution, reaching a final concentration that would have been sufficient to inactivate the starting population in one second. The evolved population of ribozymes developed resistance to the DNA enzyme, reducing their vulnerability to cleavage by 2000-fold but retaining their own catalytic function. Based on sequencing and kinetic analysis of the ribozymes, two mechanisms are proposed for this resistance. One involves three nucleotide substitutions, together with two compensatory mutations, that alter the site at which the DNA enzyme binds the ribozyme. The other involves enhancement of the ribozyme's ability to bind its own substrate in a way that protects it from cleavage by the DNA enzyme. CONCLUSIONS: The ability to direct the evolution of an enzyme's biochemical properties in response to the behavior of another macromolecule provides insight into the evolution of resistance and may be useful in developing enzymes with novel or enhanced function.

A DNA enzyme that cleaves RNA.

BACKGROUND: Several types of RNA enzymes (ribozymes) have been identified in biological systems and generated in the laboratory. Considering the variety of known RNA enzymes and the similarity of DNA and RNA, it is reasonable to imagine that DNA might be able to function as an enzyme as well. No such DNA enzyme has been found in nature, however. We set out to identify a metal-dependent DNA enzyme using in vitro selection methodology. RESULTS: Beginning with a population of 10(14) DNAs containing 50 random nucleotides, we carried out five successive rounds of selective amplification, enriching for individuals that best promote the Pb(2+)-dependent cleavage of a target ribonucleoside 3'-O-P bond embedded within an otherwise all-DNA sequence. By the fifth round, the population as a whole carried out this reaction at a rate of 0.2 min-1. Based on the sequence of 20 individuals isolated from this population, we designed a simplified version of the catalytic domain that operates in an intermolecular context with a turnover rate of 1 min-1. This rate is about 10(5)-fold increased compared to the uncatalyzed reaction. CONCLUSIONS: Using in vitro selection techniques, we obtained a DNA enzyme that catalyzes the Pb(2+)-dependent cleavage of an RNA phosphoester in a reaction that proceeds with rapid turnover. The catalytic rate compares favorably to that of known RNA enzymes. We expect that other examples of DNA enzymes will soon be forthcoming.

A DNA enzyme with Mg(2+)-dependent RNA phosphoesterase activity.

BACKGROUND: Previously we demonstrated that DNA can act as an enzyme in the Pb(2+)-dependent cleavage of an RNA phosphoester. This is a facile reaction, with an uncatalyzed rate for a typical RNA phosphoester of approximately 10(-4) min-1 in the presence of 1 mM Pb(OAc)2 at pH 7.0 and 23 degrees C. The Mg(2+)-dependent reaction is more difficult, with an uncatalyzed rate of approximately 10(-7) min-1 under comparable conditions. Mg(2+)-dependent cleavage has special relevance to biology because it is compatible with intracellular conditions. Using in vitro selection, we sought to develop a family of phosphoester-cleaving DNA enzymes that operate in the presence of various divalent metals, focusing particularly on the Mg(2+)-dependent reaction. RESULTS: We generated a population of > 10(13) DNAs containing 40 random nucleotides and carried out repeated rounds of selective amplification, enriching for molecules that cleave a target RNA phosphoester in the presence of 1 mM Mg2+, Mn2+, Zn2+ or Pb2+. Examination of individual clones from the Mg2+ lineage after the sixth round revealed a catalytic motif comprised of a three-stem junction. This motif was partially randomized and subjected to seven additional rounds of selective amplification, yielding catalysts with a rate of 0.01 min-1. The optimized DNA catalyst was divided into separate substrate and enzyme domains and shown to have a similar level of activity under multiple turnover conditions. CONCLUSIONS: We have generated a Mg(2+)-dependent DNA enzyme that cleaves a target RNA phosphoester with a catalytic rate approximately 10(5)-fold greater than that of the uncatalyzed reaction. This activity is compatible with intracellular conditions, raising the possibility that DNA enzymes might be made to operate in vivo.

Template-directed ligation of peptides to oligonucleotides.

BACKGROUND: Oligonucleotide-peptide conjugates have several applications, including their potential use as therapeutic agents. We developed a strategy for the chemical ligation of unprotected peptides to oligonucleotides in aqueous solution. The two compounds are joined via a stable amide bond in a template-directed reaction. RESULTS: Peptides, ending in a carboxy-terminal thioester, were converted to thioester-linked oligonucleotide-peptide intermediates. The oligonucleotide portion of the intermediate binds to a complementary oligonucleotide template, placing the peptide in close proximity to an adjacent template-bound oligonucleotide that terminates in a 3' amine. The ensuing reaction results in the efficient formation of an amide-linked oligonucleotide-peptide conjugate. CONCLUSIONS: An oligonucleotide template can be used to direct the ligation of peptides to oligonucleotides via a highly stable amide linkage. The ligation reaction is sequence-specific, allowing the simultaneous ligation of multiple oligonucleotide-peptide pairs.

Inventing and improving ribozyme function: rational design versus iterative selection methods.

Two major strategies for generating novel biological catalysts exist. One relies on our knowledge of biopolymer structure and function to aid in the 'rational design' of new enzymes. The other, often called 'irrational design', aims to generate new catalysts, in the absence of detailed physicochemical knowledge, by using selection methods to search a library of molecules for functional variants. Both strategies have been applied, with considerable success, to the remodeling of existing ribozymes and the development of ribozymes with novel catalytic function. The two strategies are by no means mutually exclusive, and are best applied in a complementary fashion to obtain ribozymes with the desired catalytic properties.

Amplification, mutation and selection of catalytic RNA.

RNA, by virtue of its genotypic and phenotypic properties, is a suitable substrate for molecular evolution in the laboratory. We have developed techniques for the rapid amplification, mutation and selection of catalytic RNA. By combining these techniques in an iterative fashion, we are attempting to construct an RNA-based evolving system. Such a system could be used to explore the catalytic potential of RNA.