2000 Annual Science Report
Scripps Research Institute Reporting | JUL 1999 – JUN 2000
The Ellington lab is carrying out work in three areas related to the evolution of self-replicating species: the development of hybrid or chimeric replicators in which different biopolymers each contribute to a replication cycle, the development of a deoxyribozyme that is capable of self-replication from oligonucleotide substrates, and the evolution of modified ribozymes that may have augmented catalytic activities.
1. RNA Ligations on Peptide Templates. Self replicating systems based on nucleic acids as well as a-helical peptides have previously been demonstrated. In order to further understand self-replicating molecular ensembles we have been investigating a hybrid system in which we are attempting to generate an interdependent set of self-replicating nucleic acids and peptides. To do this we have been using specific RNA sequences (aptamers) that have been evolved in vitro to bind specific peptide targets. As a stating point in our replication cycle we have been investigating the ability of these peptides to serve as templates for the ligation of an aptamer that has been broken into two fragments (Fig. 1).
We have found using the activating agent cyanogen bromide that ligation of two RNA aptamer fragments is specifically enhanced up to ~10 fold in the presence of the aptamers cognate peptide. While nonspecific reaction between peptide and RNA can also occur this can be suppressed by the addition of a non-specific competitor RNA such as tRNA. Our overall yield for the ligation reactions however are low (~1-5%), and so we are currently investigating alternative ligation schemes such as using preformed phosphorimidazolides and N-hydroxybenzoic esters.
2. DNA Ligator That Forms An Unnatural Internucleotide Linkage We have been investigating the ability of DNA to catalyze a ligation reaction between an oligonucleotide bearing a 5’ iodine and another containing a 3’ phosphorothioate (Fig. 2). The resulting 5’phosphorothioether linkage is analogous to the normal phosphodiester linkage and behaves similarly both chemically and structurally.
The general ligation selection scheme shown in Fig 3 is based on the selection scheme used by Shoztak and Bartel. The pool was constructed as shown in Fig 3 (top) and contains a random region of 90 nucleotides. The course of the selection is shown in Fig 4. After 13 rounds of selection the rate of the ligation reaction was enhanced ~500 fold.
Sequencing after round 10 showed two dominant clones which each appeared 3 times out of 20 sequences (class Ia and class II; Fig 5). An additional round was conducted to see if these two clones would take over the population. Round 11 yielded one dominant clone from class II, and no members of class Ia. Surprisingly this clone was the slower of the two clones identified in the previous round. This may be a result of a preference of certain clones for specific substrates (see below). After two additional rounds of selection, the pool was again sequenced. This time the dominant clones from class Ia, the faster of the two dominant clone families from round 10. Although five different substrates were used during the course of the selection, individual clones show a substrate preference. The substrate preferences of clones isolated from round 11 are shown in Fig 6. The switch between dominant clones in the latter rounds of the selection may be related to their relative substrate preferences.
In order to begin to understand what regions of the selected DNA enzyme were necessary for the ligation reaction, a ladder of 3’ deletion mutants was constructed for both class Ia and class II DNA ligators. The deletion mutants were then assayed for their ability to catalyze the ligation reaction. As can be see from Fig. 7, the 3’ ends of the cannot be deleted much past the position of the 3’ primer binding site (underlined region) indicating that the 3’ end of the molecule is important for catalysis. It is not surprising that the primer binding region itself is nonessential as it is doubled stranded during the selection an therefore less available for bonding interactions.
We are currently trying to address other structural characteristics of the class Ia and class II DNA ligators by performing a doped selection. This will allow for identification of regions of co-variation and an better understanding of the structural details of these classes of ligators.
The relatively low rate enhancement for the ligation reaction may be a result of the inability of DNA to stabilize the iodine leaving group. This may also be a factor for RNA enzymes. Both the uncatalyzed and catalyzed ligation reactions appear to be Mg2+ independent. The catalyzed reaction shows only a ~2 fold difference in its absence (data not shown). The inability to use this metal may be the result of the poor affinity the hard Mg2+ for sulfur containing compounds. Replacing Mg2+ with the more thiophilic Mn2+ has little effect on the reaction rate. Other metals which may affect catalysis are Ag and Hg, which can be used specifically cleave the 5’phosphorothioether linkage. We are currently planning another selection using this same pool and a variety of metal ions in hope of identifying faster ligators. A comparison of metal dependent and metal independent ligators will allow us to begin to understand some of the limitations of nucleic acid catalysis as well as address the role of metal ions in enhancing the chemical catalysis necessary to support early metabolism and the first replicating systems
In the next few months we hope to further characterize the structure and catalytic function of the class Ia and class II ligators. Essential to this is a doped sequence selection which is now in progress. Our new knowledge of structural information will be used to redesign the ligators so that they are capable of catalyzing a ligation reaction in trans (Fig. 8 top). We will then begin to look for ways to break the ligators so that they become substrates for this trans ligation reaction (Fig. 8 bottom). In this way we hope to be able to design a novel self-replicating ligase enzyme in which two or more nonfunctional fragments are joined to form an active trans ligator capable of performing additional ligation reactions.
3. Evolution of Ribozymes with Modified Nucleotides. We have synthesized modified nucleoside triphosphates, including 5-hydroxymethyl uridine triphosphate, 5-amidizolemethyl uridine triphosphate, and 5-phenolmethyl uridine triphosphate. We now have available 10-20 micromole amounts of these nucleoside triphosphates that can be used for initial experiments. Each of these modified nucleotides can be readily incorporated into transcripts and pools by T7 RNA polymerase. We have recently synthesized multi milligram amounts of the MN90 pool in which each of these nucleotides has been substituted for uridine to carry out selection experiments. The modified transcription products have slightly altered mobilities by gel electrophoresis, as expected. We will now carry out RNase T2 digestions and isolate products by HPLC to confirm the incorporation of the modified nucleotides and to estimate the level of contamination by uridine, if any. We will then begin selection experiments for ligators. These experiments will be carried out in parallel with a pool containing unmodified ribotides in order to directly compare the catalytic augmentation of RNA by modified nucleotides.
PROJECT MEMBERS:Andrew Ellington
RELATED OBJECTIVES:Objective 2.0
Develop and test plausible pathways by which ancient counterparts of membrane systems, proteins and nucleic acids were synthesized from simpler precursors and assembled into protocells.