1
Fundamentals of RNA and Ribozyme Structure

This chapter is intended to familiarize the reader with the structure of RNAs. Understanding the structural basis of RNAs is a prerequisite for the study of ribozymes and RNA-mediated catalysis. This section shows how nucleotide stereochemistry guides the structuring of helices and consequently the addition of functional motifs that give this polymer its folding and interaction properties, as well as its catalytic properties.

It is important to understand that all biological mechanisms rely on the interaction capabilities of structured molecules. The structure of biomolecules is therefore a fundamental aspect for the understanding of biology. The figures in this book are therefore often developed from experimental structures obtained by radio-crystallography or electron microscopy. Visualizing a biological mechanism through the molecular structures involved allows a better understanding of the actions of the different partners and the domains that compose them. It is then possible to deduce the mechanisms of chemical reactions and also to establish evolutionary relationships between homologous molecules of different organisms that perpetuate these mechanisms while adapting to different selection pressures resulting from distinct ecological constraints.

1.1. Sequences and secondary structures

RNA (ribonucleic acid) is one of the three main biological polymers with DNA (deoxyribonucleic acid) and proteins. RNA adopts complex structures thanks to the physicochemical properties of the four main nucleotides from which it is assembled. The nucleotides are composed of a ribose-phosphate part and an aromatic part, the nucleobase, attached to the ribose. The base is composed of one or two fused aromatic rings containing imines and ethylenic carbons decorated by exocyclic amines and/or carbonyl groups. The ribose is also substituted with a phosphate group (Figure 1.1(a)). The phosphate group gives each nucleotide a negative charge. The "backbone" of the polymer is thus a polyanion. The decorations of the aromatic bases generate an electrostatic profile specific to each one that allows for the local appearance of negative (d-) and/or positive (d+) partial charges. RNAs are therefore not simple polyanions. The electrostatic profile of the bases confers on nucleotides interaction properties between them, as well as with the ions and water molecules that solvate them (Auffinger et al. 2016; D'Ascenzo and Auffinger 2016; Leonarski et al. 2017, 2019). Nucleotides tend to stack and form right-handed double-stranded helices promoted by the establishment of hydrogen bonds between bases. The 5'-3' orientation of the strands is opposite. The strands are therefore antiparallel (Figure 1.1(b)).

Figure 1.1. The four nucleotides of RNA form antiparallel double-stranded helical structures

COMMENTARY ON FIGURE 1.1.- a) Two purines (N) on the left and two pyrimidines on the right (Y). The ribose and the negatively charged phosphate group are shown on the adenine along with the numbering of the atoms according to the IUPAC (International Union of Pure and Applied Chemistry) system. The ribose atoms are numbered from 1' to 5' and the base atoms from 1 to 9 for a purine (R) and from 1 to 6 for a pyrimidine (Y). The oxygen atom of the ribose ring corresponds to the hydroxyl group carried by the C4' and therefore has the same number (O4'). The situation is identical for the hydroxyl groups carried by C2' and C3'. Uracil and cytosine have an O4 or N4 group, respectively. The polynucleic acid parent chains are oriented from 5' to 3'. The present example gives the sequence 5'-ABC-3'. b) Since each phosphate group carries a negative charge, a polynucleotide is a polyanion whose structure mostly forms right-handed antiparallel double-stranded helices. The bases pair up to form plateaus of bases that stack with each other.

Despite different electrostatic properties, the stereochemistry of each nucleotide is identical. The ßD-ribofuranose isomer1, hereafter simply referred to as ribose, adopts an envelope (E) or twist (T) fold depending on whether four or three ring atoms define a plane, respectively. This plane is oriented using the only non-asymmetric carbon atom outside the ribose ring, the C5'. When the ribose is drawn with the C5' above the plane and to the left, the C2' and C3' carbon atoms point forward and the O4' points backward (Figure 1.2(a)). This configuration is the only one found in nature and gives an idea of the intensity of selection pressures that led to the emergence of these stereoisomers in biological nucleic acid synthesis pathways. The base is branched at the C1' position and points to the same edge as the phosphate group. All nucleotides are therefore superimposable to each other by their ribose-phosphate part. The strands (chains) of nucleotides are naturally structured into helices that interact with each other to form double-stranded helices characterized by grooves of different morphologies (Figure 1.2(b)). In fact, the path between the riboses of a base pair is shorter on one edge of the helix than on the other, which gives rise to the notion of major and minor grooves. In RNA, the minor groove is wide and shallow, and the major groove is narrow and deep. Width and depth are not independent. A deep groove is narrow, and a wide groove is shallow. The proteins that interact with RNA therefore tend to interact with the shallow groove.

Figure 1.2. Nucleotides and base pairs generating antiparallel double-stranded helices

COMMENTARY ON FIGURE 1.2.- a) Nucleotides adopt a precise conformation within the helices. The three edges of the nucleobases define the Watson-Crick, Hoogsteen and Sugar edges schematized by the edges of a right triangle. The atoms are numbered as shown. O1P and O2P are not linearly integrated into the 5'-3' ribose-phosphate backbone. The phosphate moiety is prochiral because the tetrahedral phosphorus has two identical moieties (O1P corresponds to the Pro-S stereoisomer and O2P corresponds to the Pro-R with reference to the R and S stereodescriptors). b) RNA helices are formed by the stacking of base pairs formed via Watson-Crick edges. Watson-Crick helices have an antiparallel orientation of the 5'-3' strands. If the 5' end of one strand is behind the plane of the sheet, then the 5' end of the other strand is above the plane. c) The permutations of these base pairs are isosteric, i.e. their riboses are superimposable 2 to 2. The orange dashed line collinear to the hydrogen bonds verifies whether the pairing is in the cis (both riboses are on the same edge) or trans configuration (both riboses are on opposite sides of this line). The O3'-P bonds between two adjacent riboses induce a right-handed rotation of about 33° between two consecutive base pairs. The A-form RNA helices therefore rotate to the right.

Five important properties of RNA follow from this absolute configuration.

The planar structure of nucleobases defines three hydrogen bond acceptor and/or donor edges, Watson-Crick (W or WC), Hoogsteen (H) and sugar (S). The W edge is responsible for the base pairs identified in the DNA double helix model by James Watson and Francis Crick in 1953 (Watson and Crick 1953). Karst Hoogsteen (1963) was the first to observe pairings involving the N7 and N6 positions of adenines, thus giving his name to this edge of the nucleobases:

- The O2' group plays a prominent role in interactions with the sugar edge and, as we will see later, in catalysis. Since DNA is free of the O2' group, its structural repertoire is consequently not as rich as that of RNA, as is its chemical reactivity.

- The nucleotides are linked to each other from the 5' position to the 3' position. The ribose-phosphate backbone is thus polarized. In projection on the axis of the helix, the nucleotide i-1 which precedes the nucleotide considered i is thus "above" the plane of the ribose.

- The plane of the base is perpendicular to the plane of the ribose and to the helix axis (Figure 1.2(b)).

- The grooves of the helix have different widths and depths. The deep groove is less accessible because it is narrow due to the stacking of the base plateaus. On the contrary, the shallow groove is very accessible. The S sides of the bases are therefore more accessible for another macromolecule than the H sides. The W sides are involved in pairing and are therefore less frequently involved in interactions with other molecular partners.

These five properties result in the formation of anti-parallel double-stranded helices. In a multi-nucleotide chain, the bases tend to stack with each other (stacking interactions), providing the edges capable of forming hydrogen bonds with the opportunity to interact with another RNA chain. For a helix to form, the interacting bases must be complementary. The Watson-Crick edges of a purine A or G always form a base pair with the Watson-Crick edges of a pyrimidine U or C, respectively, by establishing hydrogen bonds. In RNAs, G can also interact with U by forming two hydrogen bonds. This particular conformation is called wobble because it induces a shift towards the deep groove of the pyrimidine. This geometry stabilizes helices whose strands are not...