Ligand-induced stabilization of the aptamer terminal helix in the add adenine riboswitch

Abstract

Riboswitches are structured mRNA elements that modulate gene expression. They undergo conformational changes triggered by highly specific interactions with sensed metabolites. Among the structural rearrangements engaged by riboswitches, the forming and melting of the aptamer terminal helix, the so-called P1 stem, is essential for genetic control. The structural mechanisms by which this conformational change is modulated upon ligand binding mostly remain to be elucidated. Here, we used pulling molecular dynamics simulations to study the thermodynamics of the P1 stem in the add adenine riboswitch. The P1 ligand-dependent stabilization was quantified in terms of free energy and compared with thermodynamic data. This comparison suggests a model for the aptamer folding in which direct P1-ligand interactions play a minor role on the conformational switch when compared with those related to the ligand-induced aptamer preorganization.

Keywords: P1 stem; RNA aptamer; adenine riboswitch; free energy calculation; molecular dynamics simulation.

FIGURE 1.

Adenine riboswitch aptamer and binding site. (A) Secondary structure elements and (B) 3-dimensional structure with bound adenine. The P1 stem is gray; the other stems and loops are black. (C) Cartoon representation of the binding site; the two dotted lines represent the hydrogen bonds of the WC pairing between the U62 and the ligand.

FIGURE 2.

Secondary structure representation of the add riboswitch in the ON-state (A) and OFF-state (B). The ligand, the initiation codon, and the Shine-Dalgarno sequence are labeled.

FIGURE 3.

Initial (A) and final (B) configuration of the SMD simulation opening the P1 stem shown here for the Holo form. The backbone of the aptamer is in blue except for the P1 stem, which is in light blue. The ligand and the 18 bases forming the helix are shown. The P1 stem is formed in A and disrupted in B.

FIGURE 4.

Representative structures of the Holo binding pocket at the beginning (RMSD = 0 nm) (A) and at the end (RMSD = 0.35 nm) (B) of the SMD. The portion of the P1 stem removed in our simulations is in light blue. Bases forming the binding pocket are labeled, ligand is shown in red. A9-U63 pair is formed in A and disrupted in B.

FIGURE 5.

RMSD from native structure. (A) Holo form during 48-nsec equilibration, computed on the whole aptamer (black) and on the bases from 9 to 63 (gray). (B) Same as A done on the Apo form (whole aptamer, black; bases from 9 to 63, gray). The difference between black and gray profiles, in both panels, indicates that the P1 stem is less stable than the rest of the aptamer. (C) Δ1–8/64–71 Holo RMSD along the 48-nsec equilibration at constant temperature. (D) Same as C for the Δ1–8/64–71 Apo form.

FIGURE 6.

Base-pair ruptures during P1 pulling. In the pulling simulations, the 9 bp forming the P1 stem were unpaired. Here, we monitored the RMSD of each bp (gray scale) from their native conformation: (A) Holo; (B) Apo. A9-U63 bp (in black) was disrupted (RMSD ≈0.5 nm, arrows) later in the Holo form (≈19 nsec) than in the Apo form (≈16 nsec).

FIGURE 7.

Unbinding process of the A9-U63 bp. (A) Mechanical work performed as a function of the value of the steered RMSD, or equivalently of time, for 512 simulations for Apo (pink) and Holo form (light blue). The respective free energies resulting from the Jarzynski equality (Jarzynski 1997a) are shown in thicker red and blue lines. The initial free-energy decrease related to the entropy gain induced by the restraint movement has no consequence on the final result. (B) Hydrogen bonds occurrence for both the systems Apo (red) and Holo (blue). (C) Snapshots of the Holo form (ligand in black) with two, one, or zero hydrogen bonds (dotted lines) formed between A9 and U63.

FIGURE 8.

Schematic representation of the aptamer secondary structure in its folding intermediates with and without the ligand. (A) Stems P2 and P3, loops L2 and L3 are folded but not stable. (B₁) The junctions (J1-2, J2-3, J3-1) arrange around the ligand (ADE) and the interloop pairings occur (L2–L3), stabilizing also the stems. (B₂) Alternative possible intermediate in which all three stems are not completely and stably folded before ligand binding. The junctions and the tertiary interaction between the loops are not stable. (C) P1 stem is fully folded and stabilized by the ligand.