227 Views of RNA: Is RNA Unique in Its Chemical Isomer Space?

Abstract

Ribonucleic acid (RNA) is one of the two nucleic acids used by extant biochemistry and plays a central role as the intermediary carrier of genetic information in transcription and translation. If RNA was involved in the origin of life, it should have a facile prebiotic synthesis. A wide variety of such syntheses have been explored. However, to date no one-pot reaction has been shown capable of yielding RNA monomers from likely prebiotically abundant starting materials, though this does not rule out the possibility that simpler, more easily prebiotically accessible nucleic acids may have preceded RNA. Given structural constraints, such as the ability to form complementary base pairs and a linear covalent polymer, a variety of structural isomers of RNA could potentially function as genetic platforms. By using structure-generation software, all the potential structural isomers of the ribosides (BC5H9O4, where B is nucleobase), as well as a set of simpler minimal analogues derived from them, that can potentially serve as monomeric building blocks of nucleic acid-like molecules are enumerated. Molecules are selected based on their likely stability under biochemically relevant conditions (e.g., moderate pH and temperature) and the presence of at least two functional groups allowing the monomers to be incorporated into linear polymers. The resulting structures are then evaluated by using molecular descriptors typically applied in quantitative structure-property relationship (QSPR) studies and predicted physicochemical properties. Several databases have been queried to determine whether any of the computed isomers had been synthesized previously. Very few of the molecules that emerge from this structure set have been previously described. We conclude that ribonucleosides may have competed with a multitude of alternative structures whose potential proto-biochemical roles and abiotic syntheses remain to be explored.

FIG. 1.

The molecular structures of RNA and DNA and their components. (A) The sugars ribose and deoxyribose and their atom-numbering conventions. Note the stereochemistry of the bonds between the ring and its substituents. (B) The nitrogen heterocycles used in RNA [adenine (A), uracil (U), guanine (G), and cytosine (C)] and DNA [A, G, C, and thymine (T)] and their ring-atom-numbering conventions. (C and D) The structures of phosphate-linked RNA and DNA. (E) A simplified generic nucleoside analog structure.

FIG. 2.

Some previously explored prebiotic syntheses of nucleosides and nucleotides.

FIG. 3.

(A) Some commercial antiviral nucleoside analogues. (B) Antisense nucleoside-analog-based polymers that have found use in biotechnology or that have been found to be capable of Watson-Crick-type base-pairing.

FIG. 4.

A general work flow for the enumeration of the riboside isomers. The enumeration of good structures can be iterated to produce increasingly “good” sets of output.

FIG. 5.

The final enumerated set of riboside isomers. Structures are ordered from right to left and top to bottom according to the location of the double bond equivalent (DBE), beginning with aldehydes, then ketones, esters, BC(=O)C linkages, BC(= O)O linkages, carboxylic acids, and finally rings. The structure corresponding to the natural ribosides is highlighted by a black cartouche.

FIG. 6.

The number of structures with a given number of stereocenters of the output set. Black bars: number of connectivity isomers. White bars: number of total stereoisomers.

FIG. 7.

Distribution of possible backbone distance monomer repeats from the enumerated set. Black bars: ester linkages. White bars: diol-mediated linkages connected directly as ethers. Light gray bars: diol linkages assuming incorporation of a phosphate linker. Linkage distance repeat frequencies are summed over the total number of occurrences in the set, i.e., a single structure that can be linked three unique ways is counted three times.

FIG. 8.

Computed MOI (top) and cross-sectional molecular shadows (bottom) of the 962 energy-minimized output adenosine-analog stereoisomers, shown in descending order from left to right of the pairwise metrics for the three major axes (i.e., X+Y, Y+Z, Z+X). The stereoisomers of the ribofuranosides are shown as open red circles. D and L β-ribofuranosyladenosine are represented as a solid red circle. Color graphics are available at www.liebertonline.com/ast.

FIG. 9.

Plot of the number of free rotatable bonds to the computed van der Waals volume. The DL-β-ribofuranosyl- and DL-β-ribopyranosyladenosine structures are shown as a white triangle and square, respectively, superposed over the remaining 958 stereoisomers. The structures associated with the three outlying left-most points are also shown for comparison.

FIG. 10.

Computed standard molar Gibbs free energies of formation of the 227 A-substituted isomers. The calculated and measured values for adenosine are shown as vertical red and blue lines, respectively. The shaded area represents the estimated uncertainty of±13.4% in the calculated value. Color graphics are available at www.liebertonline.com/ast.

FIG. 11.

Method for extracting functionally minimal nucleoside motifs from the enumerated riboside isomers.

FIG. 12.

The 68 functionally minimal acyclic substructures derived from the computed library, presented from left to right by the number of carbon atoms they contain.

FIG. 13.

SYLVIA-generated synthetic accessibility scores for the riboside isomers and minimal structures. The riboside isomers are shown as open black circles, the minimal structures as open red circles. Individual isomers are annotated. Color graphics are available at www.liebertonline.com/ast.