Ad hoc construct design to slow down functional dynamics enabled the capturing of group I intron splicing and folding intermediates.

(A) Schematic representation of group I introns highlighting the engineering strategies that enabled cryo-EM investigation of catalytic intermediates, i.e. insertion of IGS, addition of 5’- and 3’-splice analogues, and active site mutations (G264A). Folding intermediates have been obtained by modulating the refolding conditions, as indicated in the inset (concentration of magnesium, Mg2+, and temperature, T). (B) Representative analytical PAGE gel of the G264A group I intron sample (TETG264A) used for cryo-EM. (C) PAGE gels confirming activity of the IGS-engineered group I intron. The top gel displays the first step of splicing reaction, the bottom gel the second step of splicing reaction. S, S1 and S2 indicate fluorescently labeled substrates, P indicates the product of the reaction, TET indicates the intron. (D) Superposition of co-transcriptionally folded group I intron structures formed in the first step of splicing, and displaying the P1 helix (depicted in orange, for the 5’-exon moiety, and blue, for the intron 5’-end moiety) in the relaxed (PDB id: 8HD6), intermediate (abbreviated as Int, PDB id: 8HD7), and docked (PDB id: 8I7N) states. The distance between the 5’-splice site and the G-binding site (in green) in depicted with dashed black line for the relaxed (62 Å) and docked (4 Å) state. The left and right subpanels are rotated 180° degree with respect to one another. (E) Superposition of the misfolded (M, PDB id: 7XSK) intermediate and native (N, PDB id: 7XSN, in grey) states of group I intron. For the M-state, the P7 helix is in green, the P8 helix is in salmon, the P3 helix is in yellow, the J7/8 helix is in dark blue, the J7/3 helix is in purple and the P9 helix is in grey. Panel B has been adapted from Extended Data Figure 7 panel A from (Luo et al., 2023). Bottom panel C is reproduced from Supplementary Figure 1A from Li et al., 2023.

© 2023, Zhang et al. Top panel C was reprinted with permission from Supplementary Figure 1 panel A from Zhang et al., 2023, Nucleic Acids Research, which was published under a Creative Commons Attribution-Non-Commercial License. Further reproductions must adhere to the terms of this license.

Construct engineering and optimization of vitrification conditions enabled the visualization of group II intron folding intermediates.

(A) Secondary structure map of the group II intron used for visualizing folding intermediates. The map is colored by domains. Black dashed squares indicate tertiary structure motifs (θ-θ’, κ, ζ, λ) that are flexible in both the cryo-EM maps and in previous chemical probing studies of folding intermediates. (B) Constructs used for cryo-EM investigation. The color code of each domain is the same as in panel A. (C) PAGE gel showing the purity and homogeneity of a representative construct. (D) PAGE gels comparing the activity of wild type and mutant introns, used to validate the folding model. (E) Micrographs of the D1-3 construct acquired in suboptimal grid vitrification conditions (31 μM concentration, -3 force, 1 sec blotting) and displaying unfolded, aggregated particles. (F) Micrographs of the D1-3 construct acquired in optimal grid vitrification conditions (30 μM concentration, -10 force, 2 sec blotting) and displaying homogenously distributed particles. (G) 2D classes showing comet-like particles formed by dynamic group II intron constructs. (H) Superposition of different conformations of the D1-3 folding intermediate, highlighting the D1c helix and the hinge motif that control its dynamics. The θ and θ’ nucleotides are also labeled, showing that the θ-θ’ interaction is dynamic during group II intron folding in line with the chemical probing data (see panel A and the main text). The two subpanels are rotated by 98° with respect to one another. Panel C is reproduced from Figure 1B from Jadhav & Marcia, 2025a. Panel D is reproduced from Figure 8B of Jadhav et al., 2025. Panel G is reproduced from Figure 6A from Jadhav & Marcia, 2025a.

Integration of cryo-EM and AFM data enabled the visualization of cobalamin riboswitch dynamics between its apo- and its holo-states.

(A) High resolution AFM images of rCbl in monomeric and dimeric states. Dimeric and monomeric states have been classified on the basis of shape. (B) Superposition of dimer 1 (PDB ID: 8SA2, shown as surface representation in grey) and dimer 2 (PDB ID:8SA3, in olive). The black dotted arrow indicates the motion of chain B in dimer 2 with respect to dimer 1. (C) Comparison of the apo (PDB ID: 8SA6, shown as surface representation on the left) and the holo dimeric forms of the riboswitch (dimer 1, PDB ID: 8SA2, shown as surface representation on the right). The distances between phosphorus atoms of G74 (in the P5 domain; in cyan) and C165 (in the P3 domain; in orange) in both structures are displayed as black lines and labeled (in Å) to quantify the movement of the P6 domain (in yellow) between the two states. Panel A is reproduced from Figure 1C and D of Ding, Lee, et al., 2023.

Strategic construct design enabled modeling dynamic viral domains at medium resolution.

(A) Secondary structure map of BMV-TLS WT and two RNA helix engineered constructs B3ext+Cshort and Dext+B2short. (B) Native PAGE gel image of WT BMV-TLS. (C) Cryo-EM maps of the WT BMV-TLS element wherein the B3+E RNA helix is highlighted with solid line. (D) Structure superposition of BMV-TLS TyrRS bound and unbound states showing 90° rotation of B3+E helix (orange and purple for unbound and bound protein structures respectively). (E) Secondary structure map of SARS-CoV2 SL5 (left) compared to the secondary structure map of SARS-CoV2 SL5 with the SL6 extension (right, in yellow). (F) Flexible nucleotides identified by chemical probing are indicated as red dots over the two secondary structure maps, supporting their modelling as single stranded nucleotides. (G) Superposition of the cryo-EM maps of SARS-CoV2 SL5 (grey) and SL5-6 (yellow). Density for SL6 is indicated by the black circle. (H) Superposition of three conformations of MERS SL5. Stem SL5a in conformation 1, 2, and 3 is displayed in maroon, tomato, and goldenrod, respectively. Its rotation is indicated by the solid lines (in °). The change in distance for hexaloop 1 are indicated by the dotted lines (in Å). (I) Superposition of three conformations of BtCoV-HKU5 SL5. Stem SL5a from conformation 1, 2, and 3 are displayed in maroon, tomato, and goldenrod, respectively. Its rotation is indicated by the solid lines (in). The change in distance for hexaloop 1 are indicated by the dotted lines (in Å). Panel F is reproduced from Supplementary Figure 2 panel “SARS-CoV-2 SL5 domain Scarless M2-seq” from Kretsch et al., 2024.

© 2021 AAAS. Panel B was reprinted from Supplementary Figure 9 of Bonilla et al., 2021, with permission from AAAS. It is not covered by the CC-BY 4.0 license and further reproduction of this panel would need permission from the copyright holder.

In vitro evolution selected an efficient RNA polymerase ribozyme characterized by pronounced catalytically relevant dynamics.

(A) Schematic representation of the in vitro evolution cycle that was used to select the best performing RNA polymerase ribozyme, and specifically its 5TU catalytic subunit. (B) Cryo-EM map (EMD-40984) and structural coordinates (PDB id: 8T2P) of TPR polymerase subunits t1 (in teal) and 5TU (in light blue) in two orientations, rotated by 90° with respect to one another. Kissing loops (KL) 1 and 2, and helices P1 and P3 (from subunit t1), and P7 and P10 (from subunit 5TU) are indicated. The conformational changes of t1:P1 and 5TU:P10 helices are indicated by the dotted lines. The degree of motion is estimated from the figures reported in the original publication (McRae et al., 2024), because only one static set of coordinates is deposited in the PDB. The inset shows the active site of 5TU with all associated nucleotides in stick representation.

Synthetic design of crossovers and fusions enabled the generation of dynamic RNA origami.

(A) Schematics depicting the engineering of kissing loops and junctions that enabled cryo-EM determination of the 6HBC origami. (B) Conformational rearrangement of H6 accompanying the transition of the 6HBC origami from the young to mature state. (C) Conformational rearrangement of the H6 kissing loop accompanying the transition of the 6HBC origami from the young (in blue) to the mature (in orange) state. A30 and A34 are shown in green; A522 and A526 are shown in orange. (D) Representative images of negatively stained 16HS-shaped particles. (E) Superposition of 16HS conformations 1 and 2. Flexibility of 5HT-B is also depicted with a curved arrow. (F) Comparison of the conformations of the iSpinach aptamer derived from the crystal structure (PDB: 5OB3, left) and the iSpinach aptamer model built from the cryo-EM density map (right). (G) Cartoon representations compare the iSpinach binding pocket between the crystal structure and cryo-EM structure. Relative to the crystal structure, the binding pocket of iSpinach in the Traptamer model is markedly contorted. Residues are colored as follows: U22 (crystal structure) and U100 (cryo-EM structure) in red; A45 (crystal structure) and A271 (cryo-EM structure) in orange; U42 (crystal structure) and U268 (cryo-EM structure) in cyan; DFHBI-1T is shown as blue spheres.

© 2023 Springer Nature. Panel D was reprinted from Figure 5B of McRae et al., 2023 with permission from Springer Nature. It is not covered by the CC- BY 4.0 license and further reproduction of this panel would need permission from the copyright holder.

Integration of cryo-EM with SAXS and NMR enabled the visualization of apo- and ligand-bound D43 aptazyme dynamics.

(A) Denaturing gel electrophoresis of purified D43 (left panel) and refolded D43 (right panel). (B) Superposition of the cryo-EM structures of holo-D43 (PDB ID: 8T5O; shown in blue) and apo-D43 (PDB ID: 8TKJ; shown in orange), which differ by an RMSD of 1.4 Å. The three modules of the D43 aptazyme are labelled, as the ribozyme module, aptamer module, and communication module (CM), respectively. The inset shows a zoomed view into the CM in holo-D43 and apo-D43, along with the distances between G48-U100 and G49-C99. The increase in the distance between G48-U100 and loss of planarity in G49-C99 lead to breakage of base pairing in apo-D43. (C) The location and extent of the NMR chemical shift perturbations (left panel) and intensity ratio differences (right panel) as indicated by the size and color of spheres. Large changes in both chemical shift and peak intensity are observed within the CM, most notably the G48*U100 wobble pair. (D) Kratky plot for apo-D43 and holo-D43 from SAXS/WAXS experiments and as back- calculated from the cryo-EM structures (PDB ids 8TKJ and 8T5O, respectively). The plot was generated using BioXTAS RAW 2.4.1 (Hopkins, 2024).

© 2025 Nucleic Acids Research. Panels A and C are reproduced from Supplementary Figure S1B and D and from Figure 3B, respectively, from Stagno et al., 2025, Nucleic Acids Research, with permission from Nucleic Acids Research. They are not covered by the CC-BY 4.0 license and further reproduction of this panel would need permission from the copyright holder.

Key properties of the targets discussed in the case studies.