Exobiology and Evolutionary Biology

Landscapes in Catalytic Nucleic Acids

PI: Steven Benner

In the decade since Stanley Miller argued that chemical instability precluded the use of ribose or any other carbohydrate as a component of any prebiotic genetic molecule, research has emerged suggesting that RNA might be more accessible prebiotically than was thought in the nadir of the “RNA first” hypothesis in the mid 1990’s. This work discovered minerals that stabilize carbohydrates under conditions where they are formed, laboratory conditions forming the problematic glycosidic bond in one ribonucleoside, meteoritic material that phosphorylates alcohols in water, liposomes that concentrate ribose, alternative solvents such as formamide that both provide RNA building blocks and permit assembly of oligomeric RNA, and oligomeric RNA generated upon evaporation of building blocks in the presence of lipids. This has led to a renaissance, at least for some, of the “RNA first” hypothesis for the origin of life on Earth. At the same time, NextGen “deep sequencing” technologies have emerged that allow us to survey libraries of oligonucleotides that contain selectable catalytic power only weakly, and only sparsely.

This all makes timely a return to the question: How is selectable catalysis distributed within RNA and DNA “sequence space”? Answers to this question define the magnitude of the challenge to be met to make plausible the “RNA first” hypothesis. For example, at the extremes, if sequence space for relatively short fragments (20-30 nucleotides) is rich in catalytic power, that scenario is less grim than if useful catalytic behavior exists only in long RNA oligonucleotides, and then only sparsely. At a second level, the “grimness” of this scenario depends on the ratio of productive and counterproductive catalysis in these spaces, and asks how this ratio depends on the nature of the biopolymer (for example, DNA vs. RNA, the number and kinds of building blocks) and the environment, including the solvent, solutes, and the reaction being catalyzed.

This work will generate quantitative experimental data to determine where two biopolymers (DNA and RNA) known to be able to support both genetics and catalysis lie between these extremes. It will focus on one productive type of nucleic acid catalysis (the Szostak-Bartel-Inouye ligase) and one counterproductive type (Breaker-Joyce nuclease), the first RNA, the second DNA.

NextGen sequencing, as well as preliminary studies that improved the Breaker-Joyce experiment to help set-up and lower leakage, will allow us to quantitate active nucleases after fewer rounds of selection, and to ask direct questions about the comparative richness of nucleases in sequence space under different conditions. Questions to be asked include: (i) How richer is that space for longer vs. shorter catalytic regions, (ii) how much more abundant are active nucleases in the same length space in the presence or absence of Mg(+2), and (iii) how is the richness of that space influenced by denaturants such as formamide?

Parallel experiments with RNA ligases will ask parallel questions, describing the comparative richness of RNA sequence space as a source for ligases (i) when the length of the ligation fragments are shortened, (ii) when the nature of the leaving group is changed, and (iii) when solutes in the environment are changed.

By strategically comparing two related biopolymers, two related substrate structures, two environmental perturbations having opposite effects, and the lengths of the catalytic species, this work will provide broad outlines of the constraints that must be met by prebiotic experiments directed towards RNA. It may also lead towards a “synthetic biology”, where a human-created genetic system supports Darwinian evolution in the laboratory, a “second example” of “life” that will help NASA not only understand the origin of life on Earth, but how life having a genesis independent of life on Earth might be detected by NASA missions.