Skip to main content
NCBI home page
Nucleic Acids Research logoLink to Nucleic Acids Research
. 2026 Aug 10;54(15):gkag778. doi: 10.1093/nar/gkag778

Structural characterization of ribosome recycling and fusidic acid inhibition in Staphylococcus aureus

Adrián González-López 1,2,✉, Maria Selmer 3,4,✉
PMCID: PMC13454839  PMID: 42573073

Abstract

During bacterial ribosome recycling, 70S ribosomes are split into subunits by ribosome recycling factor (RRF) and elongation factor G (EF-G). The antibiotic fusidic acid (FA) inhibits elongation and ribosome recycling by locking EF-G to the ribosome. Yet, no functional ribosome recycling FA complex has been successfully captured. Here, we used single-particle cryo-electron microscopy to resolve multiple FA-stalled intermediates of Staphylococcus aureus ribosomes, including a 70S intermediate with RRF and EF-G in a previously unobserved conformation. Our structures reveal how RRF and EF–G jointly disrupt inter-subunit bridges, promote back–rotation of the small subunit, and move the transfer RNA toward the E site to facilitate ribosome splitting. We further show that FA predominantly inhibits recycling by trapping EF-G on the post-termination complex in the absence of RRF, preventing formation of the active RRF•EF-G complex. These insights advance understanding of the molecular mechanism of bacterial ribosome recycling and the mode of action of FA as an antibiotic.

Graphical Abstract

Graphical Abstract.

For image description, please refer to the figure legend and surrounding text.

Open in a new tab

Introduction

Ribosome recycling is the final step of bacterial protein synthesis, following translation initiation, elongation, and termination. After termination, the ribosome remains as a 70S ribosome with a deacylated P-site transfer RNA (tRNA) and an empty A site, forming what is known as the post-termination complex (postTC). Ribosome recycling factor (RRF) and elongation factor G (EF-G) act together to dissociate this 70S ribosome into 50S and 30S subunits in a GTP-dependent manner [1], allowing the subunits to participate in new rounds of initiation [2]. For clarity, we refer to the subunits as the large and small ribosomal subunits (LSU and SSU) within the 70S ribosome and as 50S and 30S for free subunits. The splitting process requires the disruption of several inter-subunit bridges, where bridge B2a, formed between helix 44 (h44) of 16S RNA and H69 of 23S RNA near the decoding center, has been suggested as central to the mechanism [3, 4]. While the structural mechanism of EF-G-mediated translocation is well understood [5–7], several aspects of ribosome recycling remain unresolved, as recently reviewed [8].

RRF is an L-shaped protein composed of a three-helix bundle (domain I) and an α/β-domain (domain II) [9, 10]. The postTC adopts a rotated state [11], to which RRF binds with domain I spanning the A and P sites [4, 12], while domain II can adopt two defined orientations [13, 14]. This binding is only compatible with a deacylated P-site tRNA [15], while in the absence of tRNA, RRF can also bind to classical-state ribosomes [16, 17]. In some structures, RRF alone induces rearrangements in H69 [16, 18], whereas others detect no such changes [17]. The RRF binding site overlaps with the post-translocational position of EF-G, consistent with the requirement that RRF must bind first during recycling [19]. Upon EF-G binding and GTP hydrolysis, domain II of RRF rotates, a movement proposed to drive ribosome splitting [20]. Low-resolution time-resolved cryo-electron microscopy (cryo-EM) studies suggested that, in the presence of EF-G, domain II of RRF moves toward inter-subunit bridge B2a [13], but the density was of insufficient quality to unambiguously place domain II of RRF. Therefore, there is still no direct structural observation of how RRF together with EF-G breaks essential inter-subunit bridges. It also remains unclear how these structural changes affect the tRNA and messenger RNA (mRNA) to allow their release [8, 21].

The antibiotic fusidic acid (FA) has been successfully used to visualize EF-G on the ribosome [5, 22, 23]. It binds to EF-G and prevents the conformational changes required for EF-G release after GTP hydrolysis and translocation [24, 25]. FA also inhibits ribosome recycling [26, 27], where it can trap EF-G on the postTC in the presence or absence of RRF [27]. No structure has yet captured a functionally engaged EF-G locked by FA together with RRF on the ribosome.

Here, we use cryo-EM in the presence of FA to resolve multiple ribosome recycling complexes from the clinically relevant bacterium Staphylococcus aureus [28]. The structures clarify the fundamental mechanism of bacterial ribosome recycling and reveal how FA perturbs this essential process. We visualize an intermediate containing both EF-G and RRF bound to the ribosome in a previously unseen conformation, providing direct structural evidence for how their coordinated actions destabilize inter–subunit bridges, induce back-rotation of the ribosome and perturb tRNA–SSU interactions.

Materials and methods

Cloning, overexpression, and purification of S. aureus EF-G and RRF

Staphylococcus aureus EF-G was purified as published [23] through immobilized metal chromatography and size exclusion chromatography. This construct includes a 6xHis tag with a TEV protease cleavage site and a linker to the EF-G sequence. After TEV cleavage, only Ser-Leu remains at the N-terminus of EF-G. The final protein sample was stored in 50 mM Tris–HCl (pH 7.5), 300 mM NaCl, 5 mM β-mercaptoethanol at −70°C. 

The S. aureus RRF gene was ordered by gene synthesis from GenScript (Piscataway, New Jersey, United States) in pET28a(+)-TEV. This construct encodes a 6xHis tag with a TEV protease cleavage site and a linker to the RRF sequence. After TEV cleavage, only Gly-His remains at the N-terminus. The plasmid was transformed into Escherichia coli BL21 (DE3) and an overnight culture was inoculated 1:100 into a 2.8 l baffled flask with 800 ml of LB with 50 µg/ml of kanamycin. Protein expression was induced at an OD600 of 0.5–0.6 with 1 mM isopropyl-β-d-thiogalactopyranoside and the culture was incubated for 3 h at 37°C. The cells were harvested at 8000 × g for 15 min in a JLA 9.1000 rotor (Beckman Coulter, Brea, California, USA), washed with 150 mM NaCl, and stored at −20°C until use. Next, the cells were resuspended in lysis buffer (20 mM Tris–HCl, pH 7.5, 500 mM NaCl, 1 mM MgSO4 10 mM imidazole, 10% glycerol) with 0.1% (v/v) Triton X-100, one ethylenediaminetetraacetic-acid-free mini-Complete protease inhibitor tablet (Roche, Basel, Switzerland) and DNase I. The resuspended cells were lysed in a flow cell disruptor (Constant Systems Ltd., Daventry, United Kingdom) and the lysate was centrifuged at 40 000 × g for 30 min in a JA-25.50 rotor (Beckman Coulter).

The supernatant was filtered by 0.45 µm with a polyethersulfone syringe filter (Sarstedt AG & Co, Nümbrecht, Germany) and incubated with 1 ml of Protino® Ni-NTA Agarose (MACHEREY-NAGEL, Düren, Germany) pre-equilibrated with lysis buffer. The column was washed with lysis buffer until the 280 and 260 nm absorbance reached baseline and the protein was eluted with elution buffer (20 mM Tris, pH 7.5, 300 mM NaCl, 1 mM MgSO4, 500 mM imidazole). The eluate was exchanged into GF buffer (20 mM Tris–HCl, pH 7.5, 300 mM NaCl, 1 mM MgSO4) through a dialysis membrane 3.5 kDa cut-off Spectra/Por® (Spectrum Laboratory Products Inc., New Brunswick, New Jersey, USA) in the presence of 1:50 molar ratio of TEV protease for 16–18 h at 4°C. This was followed by reverse immobilized affinity chromatography on 250 µl of Ni-NTA Agarose. RRF was further purified by gel filtration on a Hiload 16/60 Superdex-75 column (GE Healthcare, Uppsala, Sweden) pre-equilibrated with GF buffer. The peak fractions were subjected to additional reverse IMAC with 1 ml of Ni Sepharose Fast Flow to remove any uncleaved protein. The final sample was concentrated to 18 mg/ml with a 10-kDa cutoff Vivaspin Turbo 15 (Sartorius AG, Göttingen, Germany), frozen in liquid nitrogen, and stored at −70°C.

Ribosome and tRNA purification

Staphylococcus aureus NCTC 8325-4 ribosomes were purified as published [23]. In short, cells were harvested at OD600 of 1.0, lysed with lysostaphin, and pelleted. The supernatant was centrifuged through a 1.1 M sucrose cushion twice. The resulting crude ribosome pellet was resuspended and 70S ribosomes were isolated through rate-zonal centrifugation in a preformed 10–40% sucrose gradient. The 70S ribosomes were pooled and pelleted. The final 70S sample was stored in HEPES polymix buffer [5 mM HEPES, pH 7.5, 5 mM NH4Cl, 5 mM Mg(OAc)2, 100 mM KCl, 0.5 mM CaCl2, 8 mM putrescine, 1 mM spermidine, and 1 mM dithioerythritol] at −70°C. mRNA Z4AUGAGC (5′- GGCAAGGAGGUAAAAAUGAGCAAA -3′) was produced by chemical synthesis (GenScript) and E. coli tRNASer was overexpressed and purified as published [29].

Sample preparation

A sample mimicking a postTC was prepared by mixing 3 µM (final concentrations) 70S S. aureus ribosomes in HEPES polymix buffer (with 20 mM HEPES, pH 7.5, and 5 mM BME as reducing agent) with 6 µM mRNA Z4AUGAGC and incubated for 10 min at 37°C. Then, 6 µM E. coli tRNASer was added and the mix was incubated for 10 min at 37°C. Next, 40 µM S. aureus RRF was added, incubated for 10 min at 37°C and followed by adding 1 mM GTP. The postTC mimic was kept on ice until vitrification.

Cryo-EM grid preparation

2.4 µl of 40 µM EF-G with 2 mM GTP and 500 µM FA was mixed with 6 µl of postTC mimic on ice, incubated 45 s and 3 µl were deposited on a glow-discharged (15 s at 20 mA) QuantiFoil 200-mesh R 2/2 copper grid with 3 nm continuous carbon, incubated 10 s, blotted for 3 s, and plunge-frozen in a Vitrobot Mark IV (Thermo Fisher Scientific, Waltham, MA, USA) at 4°C and 95% humidity. The final concentrations of the components on the grid were 2.1 µM 70S ribosomes, 4.3 µM mRNA, 4.3 µM tRNASer, 28.6 µM RRF, 11.4 µM EF-G, 143 µM FA, and 1.3 mM GTP.

Cryo-EM data collection

All the grids were screened on a Glacios TEM operated at 200 kV equipped with a Falcon-4i direct electron detector with Selectris energy filter (Thermo Fisher Scientific). The final dataset was collected on a Titan Krios (Thermo Fisher Scientific) operated at 300 kV and equipped with a K3 BioQuantum direct electron detector and energy filter (Gatan, Inc, AMETEK, Berwyn, PA, USA) using 20 eV slit. The data were acquired at 165 000× nominal magnification with a calibrated pixel size of 0.504 Å. A total of 56 478 movies were collected in 33 frames with a total dose of 30 e−/Å2 (17 e−/pixel/s) over 0.45 s with a set defocus between −0.7 and −1.0 µm.

Cryo-EM data processing

All processing was done on CryoSPARC v4.7.1-cuda12 +250814 [30] following the workflow depicted in Supplementary Fig. S1. Default parameters were used unless otherwise stated. The movies were motion-corrected using Patch Motion Correction, the contrast transfer functions (CTFs) were estimated using Patch CTF Estimation. Blob picker was used to produce an initial 3D reconstruction and produce templates for template picking. After template picking a particle stack was created by extracting a 800-pixel box around each particle binned to 600 pixels (0.672 Å/pix). 2D classification was used to identify obvious picking artifacts (e.g. carbon edges) and generate baits for sorting with heterogeneous refinement. Accepted and rejected particles from 2D classification were separately used to generate ab-initio models followed by heterogeneous refinement with all the particles. Classes corresponding to ribosomes were used in a homogeneous refinement performed simultaneously with higher-order CTF estimation (including spherical aberration, tetrafoil, anisotropic magnification) and per-particle defocus refinement, with Ewald’s sphere correction (negative sign) and estimation of per-particle scale factors. The resulting reconstruction was used for referenced-based motion correction, and a new homogeneous refinement was performed with the same settings, resulting in a final consensus reconstruction with 1 823 753 particles. The 70S ribosome was subtracted using a mask that covered the ribosome and tRNA with a 20-pixel soft-edge (0.672 Å/pix). The resulting subtracted particles were then down-sampled to 128 pixels and a focused 3D classification was performed using a mask around the ribosomal A site that covers the RRF•EF-G region with a 20-pixel soft-edge (0.672 Å/pix). The 3D classification was performed using the RRF•EF-G mask, keeping the input per-particle scales, filtering resolution of 6 Å, and RMS density convergence set to false. The number of 3D classes was iteratively increased to identify the optimum. For each job, the resulting classes were homogeneously refined using the original 600 pixel binned non-subtracted particles and the maps were examined to identify the job that produced classes with the best density for RRF and EF-G. The classification with 110 classes resulted in the highest-quality maps capturing the heterogeneity of the data. The particles containing RRF were additionally subjected to focused classification with a 40 Å radius spherical mask around the ribosomal A site with 20 classes, keeping the input per-particle scales, filtering resolution of 6 Å, and RMS density convergence set to false. The particles in the classes containing RRF in the different conformations, EF-G, or RRF•EF-G were subjected to focused classification inside a SSU-mask (with 20-pixel soft edge) with 10 classes, while keeping the input per-particle scales. Homogeneous or non-uniform refinement using the original particles (600-pixel binned, 0.672 Å/pix) resulted in the final structures. Local-resolution estimation, Fourier shell correlation and angular distribution of the particles for the modeled structures are shown in Supplementary Fig. S2. No 30S particles were identified during 2D or 3D classification.

All the final maps were post-processed using phenix.auto_sharpen [31] using the map and half-maps with local sharpening, or low-pass filtered using RELION [32]. The global resolution of maps was estimated within an auto-tightened mask created by CryoSPARC. For all maps used for model refinement, the resolution within a mask generated from the model was also calculated and used as the reported resolution, since the default auto-tightened masks from CryoSPARC did not cover the whole model in all the structures. These masks were created by calculating a synthetic map from the model using the molmap command in ChimeraX [33] at 20 Å, thresholded at a level where the map covers the whole model, and then adding a 20-pixel soft edge.

Model building

The starting models used were 9GHG [34] (S. aureus 70S ribosome), 8P2G [23] (S. aureus EF-G) and an AlphaFold prediction of S. aureus RRF and S. aureus uL11 using the AlphaFold server [35]. Each chain was first rigid-body fitted and refined using Coot [36] against the post-processed map, followed by global refinement using the sharpened maps with Servalcat [37]. Afterward, the structures were revised manually, and validation issues were corrected when possible. Validation was performed using Phenix [31]. Refinement and validation information is found in Table 1.

Table 1.

Cryo-EM refinement parameters and model validation

postTC•RRF•LSU (EMDB-55945) (PDB-9TI6) postTC•RRF•LSU-2 (EMDB-55946) (PDB-9TI7) postTC•RRF-SSU (EMDB-55944) (PDB-9TI5) 70S•RRF•EF-G•FA (EMDB-55948) (PDB-9TI9) 50S•RRF•EF-G•FA (EMDB-55947) (PDB-9TI8)
Data collection and processing
Magnification 165 000 165 000 165 000 165 000 165 000
Voltage (kV) 300 300 300 300 300
Electron exposure (e–/Å2) 30 30 30 30 30
Defocus range (μm) −0.7 to −1.0 −0.7 to −1.0 −0.7 to −1.0 −0.7 to −1.0 −0.7 to −1.0
Pixel size (Å) 0.504 0.504 0.504 0.504 0.504
Symmetry imposed C1 C1 C1 C1 C1
Initial particle images (no.) 2 556 350 2 556 350 2 556 350 2 556 350 2 556 350
Final particle images (no.) 25 462 24 569 123 631 9322 77 333
Map resolution (Å) 2.62 2.76 2.2 3.06 2.18
 FSC threshold 0.143 0.143 0.143 0.143 0.143
Refinement
Initial model used (PDB code) 9GHG, AlphaFold 9GHG, AlphaFold 9GHG, AlphaFold 9GHG, AlphaFold, 8P2H 9GHG, AlphaFold, 8P2H
Model resolution (Å) 2.5 2.6 2.1 3.1 2.1
 FSC threshold 0.5 0.5 0.5 0.5 0.5
Post-processing Phenix auto-sharpen Phenix auto-sharpen Phenix auto-sharpen Phenix auto-sharpen Phenix auto-sharpen
 Map sharpening B factor (Å2) 15.87 14.73 6.65 48.03 15.02
Model composition
 Non-hydrogen atoms 142 252 142 254 140 211 147 288 93 901
 Protein residues 5737 5739 5602 6266 3976
 RNA residues 4522 4536 4473 4564 2922
 Waters 0 0 0 0 0
 Magnesium atoms 122 113 127 90 115
B factors (Å2)
 Protein 65 64 59 78 57
 RNA 50 48 44 60 41
 Ligand 26 23 26 35 38
 Water – – – – –
R.m.s. deviations
 Bond lengths (Å) 0.008 0.008 0.008 0.008 0.008
 Bond angles (°) 1.277 1.304 1.315 1.294 1.323
Validation
 MolProbity score 1 1.13 1.03 1.14 0.96
 Clashscore 1.41 2.15 1.8 2.68 1.93
 Poor rotamers (%) 0.35 0.49 0.38 0.55 0.66
Ramachandran plot
 Favored (%) 97.32 97.22 97.58 97.62 98.26
 Allowed (%) 2.61 2.73 2.38 2.32 1.74
 Disallowed (%) 0.07 0.05 0.04 0.06 0
 Rama-Z score −1.41 −1.26 −1.03 −1.16 −0.68
Open in a new tab

SSU body rotation angles were calculated similarly as done by others [6]. For SSU body rotation, the structures were superimposed by 23S rRNA and the angle between the axes formed by the 16S body residues (1–933 and 1409–1553) was calculated in ChimeraX. PDB entry 7NHM [38] was used as a classical-state reference. To assess the conformational change in RRF, the structures were aligned by domain I of RRF (1–29, 104–184) and the angle between the axes formed by domain II (30–103) was calculated in ChimeraX.

Figures

Figures were rendered using UCSF ChimeraX, and assembled in Affinity Designer (Serif Europe Ltd, Nottinghamsire, United Kingdom).

Results

Ribosome recycling captured by cryo-EM and fusidic acid inhibition

To study ribosome recycling, we assembled a postTC mimic (hereafter just postTC) containing S. aureus 70S ribosomes, a short synthetic mRNA, deacylated E. coli tRNASer and S. aureus RRF to which S. aureus EF-G, GTP and FA were added shortly before sample vitrification for cryo-EM analysis. A high concentration of FA was used to ensure effective inhibition. Image processing identified 1 823 753 ribosomal particles, which were first subjected to 3D classification using a mask focused on the RRF and EF-G binding sites. This was followed by focused classification on the SSU to obtain homogenous structural states (Supplementary Fig. S1). This resulted in several complexes containing RRF, EF-G or both solved at 2.2–3.1 Å global resolution (Fig. 1 and Supplementary Fig. S2). The majority of ribosome particles corresponded to 70S postTC complexes carrying either EF-G locked by FA (33.8%) or RRF alone (25.8%). Notably, a small but well–resolved fraction of 70S particles contained both EF–G and RRF (1.1%). Additional, less abundant complexes included 50S subunits bound to EF-G (2.7%), RRF (3.6%) or both factors (4.2%). Because the 70S complexes with RRF displayed variability in SSU rotation, the highest-population states were modeled and analyzed in detail (Supplementary Fig. S1). All complexes with EF-G showed density for FA.

Figure 1.

Cryo-EM segmented density maps of the main ribosomal states with density for RRF or EF-G. For each subset, the percentage of particles is indicated.

Open in a new tab

Cryo-EM analysis of FA inhibition of ribosome recycling. Cryo-EM reconstructions of ribosomal states containing RRF or EF-G identified in the dataset. Maps are segmented to show the LSU (light blue), SSU (light yellow), tRNA (yellow), EF-G (green), and RRF (salmon). The percentage of particles assigned to each state is indicated. States containing an empty A-site or A-site tRNA are included in Supplementary Fig. S1.

These populations show that a considerably larger proportion of the 70S ribosomes are locked to only EF-G (33.8%) than to both RRF and EF-G (1.1%). Moreover, the majority of complexes harboring both factors are 50S subunits (4.2%) rather than intact 70S ribosomes.

We observe EF-G trapped to the postTC in four main states with one or two tRNA molecules. The chimeric (CHI), rotated (ROT) and E-site states each have a single tRNA in either pe/E (CHI), p/E (ROT) or E-site conformation. The POST state contains two tRNAs in the P and E sites. The populations of the different states are comparable (Fig. 1), with the state containing only E–site tRNA being the least abundant. We further found EF-G locked to 50S subunits (2.7%), which likely reflects binding of EF–G to 50S rather than EF–G remaining locked after RRF departure.

Staphylococcus aureus RRF binds to the 70S ribosome in two main conformations

In agreement with previous observations [13, 14], RRF binds to the postTC with domain I anchored to the LSU and domain II in two distinct conformations. A main conformation shows domain II oriented toward the SSU (postTC•RRF-SSU, 17.7%), while in a less abundant conformation, domain II faces the LSU (postTC•RRF-LSU, 3.7%) (Figs 1 and 2A–C, and Supplementary Fig. S1). In general, domain II of RRF appears more dynamic and reconstructs to lower resolution (Supplementary Fig. S2) and in one set of particles domain II could not be clearly resolved (4.5%, Fig. 1).

Figure 2.

Structure models of RRF bound to the postTC in cartoon representation.

Open in a new tab

Structures of S. aureus RRF bound to the postTC. (A) Overview of the postTC•RRF-SSU structure showing the LSU (light blue), SSU (light yellow), tRNA (yellow), mRNA (brown), and RRF (salmon). (B) Enlarged view of panel (A) highlighting domain II of RRF contacting uS12. (C) Same view as in panel (B) for the postTC•RRF-LSU, in which domain II of RRF engages the LSU. (D) Same view as in panel (B) showing the postTC•RRF-LSU-2 state, with the tRNA positioned in the SSU E site and H69 partially disengaged from h44.

In both conformations, domain I of RRF contacts only the LSU, interacting with the hairpin loop of H80 and extensively with H70 and H71 (Fig. 2B and C, and Supplementary Fig. S3B and C). In the postTC•RRF-LSU conformation (Fig. 2C and Supplementary Fig. S3C), domain II extends toward the hairpin loops of H43, H44, and H95 (the sarcin-ricin loop, SRL) and uL11, without contacting the SSU. In contrast, in postTC•RRF-SSU, domain II of RRF interacts with uS12 (Fig. 2A and B, and Supplementary Fig. S3A and B). For both complexes, the most populated classes show the SSU in a rotated state and tRNA in a p/E hybrid state (Supplementary Fig. S4). In the RRF–50S complexes, domain II is generally unresolved, but in a low-population class, domain II is observed interacting with the 50S subunit (Fig. 1).

Interestingly, for postTC•RRF-LSU, one reconstruction is in close-to-classical state (1.8° body rotation and 1.4° swivel) and only carries an E-site tRNA (postTC•RRF-LSU-2, Fig. 2D and Supplementary Figs 3D and 5A). In this state, H69 is partially disengaged from the inter-subunit bridge B2a and instead interacts with domain I of RRF (Fig. 2D and Supplementary Figs 3D and 5B). This also occurs when EF-G is trapped on the ribosome with only an E-site tRNA (Supplementary Fig. S5C), suggesting that B2a is stabilized by the P-site tRNA.

Structures of RRF and EF-G bound to the 70S and 50S ribosome

We observe two well-resolved states of RRF and EF-G together on the ribosome: 50S•RRF•EF-G•FA and 70S•RRF•EF-G•FA. An additional, 70S•RRF•EF-G•FA-2 class was not interpreted due to low resolution and residual heterogeneity. Domain II of RRF is in all states oriented toward the SSU.

The 2.18 Å structure 50S•RRF•EF-G•FA allowed detailed interpretation of the interactions between RRF and EF-G on the ribosome (Fig. 3 and Supplementary Fig. S6). As expected, FA is bound next to GDP and the SRL (Fig. 3B and Supplementary Fig. S6B). RRF mainly interacts with domains III-IV of EF-G through its domain II (Fig. 3C and Supplementary Fig. S6C), while domain I of RRF contacts domain III (Fig. 3D and Supplementary Fig. S6D) and the C-terminal helix of domain V of EF-G (Fig. 3E and Supplementary Fig. S6E). EF-G maintains its canonical contacts with the LSU at H43, H44, and H95.

Figure 3.

Structure models of RRF•EF-G bound to 50S in cartoon representation and sticks for labeled residues.

Open in a new tab

Structure of the 50S•RRF•EF-G•FA complex. (A) Structure overview, showing the LSU (light blue), RRF (salmon), and EF-G (green). (B) Close-up of the FA binding site showing EF-G residues within 4 Å of FA (dark blue), and GDP (orange) and magnesium ion (green). (C) Interaction between domain II of RRF and domain IV of EF-G. (D) Interaction between the C-terminus of RRF and domain II of EF-G. (E) Interaction between domain I of RRF and the C-terminus of EF-G.

The 70S•RRF•EF-G•FA state shows both RRF and EF-G bound to the 70S, indicating that this represents a ribosome recycling intermediate prior to subunit splitting (Fig. 4A). Although this reconstruction is of lower resolution (Supplementary Fig. S2), the maps allowed confident modelling of each domain of RRF and EF-G (Supplementary Figs 7A and B, and 8), guided by the higher-resolution 50S•RRF•EF-G•FA structure. Notably, unlike during translocation, EF-G makes almost no contact with the SSU in this state (Fig. 4 C).

Figure 4.

Structure models of RRF•EF-G bound to 70S in cartoon representation and comparison to the models of postTC•RRF-SSU and 50S•RRF•EF-G.

Open in a new tab

Structure of the 70S•RRF•EF-G•FA complex. (A) Structure overview, showing the LSU (light blue), SSU (light yellow), tRNA (yellow), mRNA (brown), RRF (salmon), and EF-G (green). (B) Conformational changes, indicated by arrows, between postTC•RRF-SSU (gray) and 70S•RRF•EF-G•FA [colored as in panel (A)] aligned by 23S rRNA. EF-G is shown semi-transparent. (C) H69 is disengaged from h44, breaking inter-subunit bridge B2a in 70S•RRF•EF-G•FA. The 6 Å low-pass filtered map is shown in gray. (D) The tRNA in 70S•RRF•EF-G•FA (gray map) is shifted from the P site toward the SSU E site relative to postTC•RRF-SSU (blue map). Maps are low-pass filtered to 6 Å. (E–G) Conformational changes, indicated by arrows, from 70S•RRF•EF-G•FA [colored as in panel (A)] to 50S•RRF•EF-G•FA (gray) aligned by 23S rRNA. (E) Conformational changes of EF-G. (F, G) Changes at the EF-G-RRF interface, showing how RRF domain II would clash with h44 of the 30S subunit.

The combined action of RRF and EF-G destabilizes inter-subunit bridges and moves the tRNA from the P site

The three RRF-bound structures along the recycling reaction pathway, postTC•RRF-SSU (EF-G substrate), 70S•RRF•EF-G•FA (proposed recycling intermediate), and 50S•RRF•EF-G•FA (proposed post-splitting product) allow direct comparison of the conformational states during recycling (showcased in Supplementary Movie S1). Binding of EF-G to postTC•RRF-SSU leads to several major structural changes (Fig. 4B). In the 70S•RRF•EF-G•FA complex, domains III and IV of EF-G would sterically clash with domain II of RRF as positioned in postTC•RRF-SSU. Consequently, upon EF-G binding and GTP hydrolysis, domain II of RRF rotates by 32° relative to domain I, moving away from uS12 and toward inter-subunit bridge B2a, where it wedges between H69 and h44 (Fig. 4C), also causing rearrangement of h45. This action breaks bridge B2a, displacing H69 toward domain I of RRF and away from h44 of the SSU, which now interacts with domain II of RRF. These rearrangements collectively cause the SSU to back-rotate to a classical-like state (1.7° rotation, Supplementary Fig. S7C) and render the head dynamic (Supplementary Fig. S2E). Additionally, inter-subunit bridges B1b and B7b are broken, and bridges B6 and B8 are altered (Supplementary Figs S9 and S10). Simultaneously, the tRNA anticodon stem shifts ~6–7 Å from the SSU P site toward the E site (Fig. 4D). The density for the anticodon stem-loop is weaker than for the rest of the tRNA and surrounding rRNA, suggesting a more dynamic structure (Supplementary Fig. S7D). Due to weak density, the mRNA was only traced from the SD to the AUG codon before the cognate AGC codon, which would be too far from the tRNA anticodon to allow codon-anticodon interaction (Supplementary Fig. S7E and F). Thus, there is no support for preserved codon–anticodon interaction (Supplementary Fig. S7D–F). Although FA is present in both 70S•RRF•EF-G•FA states (Supplementary Fig. S7G and H), the low population of these complexes (1.1%) compared to 50S•RRF•EF-G•FA (4.2%) and postTC•EF-G•FA complexes (33.8%) indicate that these do not represent the main FA-inhibited states under these conditions. The observed loss of inter-subunit bridges and resulting SSU heterogeneity support the interpretation that this is a state on the path to subunit dissociation.

In the 50S•RRF•EF-G•FA complex, the interactions between RRF and EF-G are identical to in the 70S•RRF•EF-G•FA complex, and H69 remains in contact with RRF. However, both RRF and EF-G are shifted with respect to 23S rRNA (Fig. 4E–G). In particular, in 50S•RRF•EF-G•FA, EF-G domains IV-V and RRF domain II are positioned ~4–6 Å toward where the SSU would be (Fig. 4F and G), while domains I-III of EF-G and domain I of RRF are situated similarly in both structures (Fig. 4E). This shows that RRF and EF-G in the presence of FA adopt a structure on the 50S subunit that would be incompatible with a bound 30S subunit.

There is weak density for the E-site tRNA (Fig. 1), possibly due to re-binding. The high similarity of 50S•RRF•EF-G•FA to 70S•RRF•EF-G•FA, including RRF contacts with H69, suggests that this complex is the product of ribosome splitting in the presence of FA. However, we cannot exclude that an identical complex could form by binding and FA-locking of free EF-G to a 50S•RRF complex.

Interestingly, some of the inter-subunit bridges are to a smaller extent perturbed in the absence of EF-G and P-site tRNA in postTC•RRF-LSU-2 (Supplementary Figs S9 and S10), where bridge B2a is partially broken and B7b is absent. This showcases the relevance of destabilizing the tRNA–SSU interaction during recycling (Fig. 4D and Supplementary Fig. S7D), as the P-site tRNA connects the head and body of the SSU with the LSU.

Discussion

In this study, we have used cryo-EM to visualize the inhibition of ribosome recycling by FA in the clinically relevant bacterium S. aureus. Previous cryo-EM structures of RRF and EF-G bound together to postTCs or 50S subunits [13, 14, 39] were limited by resolution and 3D classification methodology. Here, we have captured a previously unobserved late intermediate state of ribosome recycling prior to subunit splitting (70S•RRF•EF-G•FA) and obtained high-resolution maps of an RRF-EF-G-50S complex. These structures provide mechanistic detail of how the sequential binding of RRF and EF-G leads to ribosome splitting, and how FA inhibits this process.

Most RRF complexes with postTCs show either of two distinct conformations of RRF, the more prevalent conformation in postTC•RRF-SSU and the less abundant conformation in postTC•RRF-LSU (Fig. 1), which were both previously observed in low-resolution cryo-EM structures [13, 14]. Thus, supported by a high number of particles and extensive classification, our data exclude the previous suggestion that the LSU-oriented conformation of RRF was formed only in the absence of P-site tRNA [13], but agree that postTC•RRF-LSU is a “non-authentic intermediate” as this is not the conformation to which productive binding of EF-G occurs.

Based on time-resolved cryo-EM structures, it was hypothesized that EF-G would help domain II of RRF to disrupt inter-subunit bridge B2a [13, 14]. This was supported by experiments showing that ribosomes lacking H69 can be recycled in the absence of RRF [40]. However, there was no direct observation of the process of subunit separation, and the density for domain II of RRF appeared fused to that of EF-G. Thus, our 70S•RRF•EF-G•FA structure, presenting clear density showing how RRF domain II wedges apart bridge B2a and breaks inter-subunit contacts (Fig. 4 and Supplementary Figs S9 and S10) supports the initial hypothesis. Furthermore, it had been unclear whether ribosome recycling indeed involved back-rotation of the SSU [14]. Our structures show that rearrangements of h44 are associated with back-rotation of the SSU, disruption of bridges B1b and B7b, changes in B6 and B8, and movement of the tRNA toward the E site.

A long-standing question is how tRNA is released during recycling [21, 41–43]. Kinetic experiments have demonstrated a correlation between the SD-P-site codon distance and the binding stability of deacylated P-site tRNA [43]. In the absence of an upstream SD, it was shown that mRNA dissociates first, then tRNA leaves and finally splitting occurs, while for SD-containing mRNAs, the three events appear to coincide [21]. In our sample, containing a strong SD sequence, the 70S•RRF•EF-G•FA structure suggests that the tRNA anticodon has lost contact with the P site in the SSU (Fig. 4D and Supplementary Fig. S7D), consistent with tRNA dissociation upon splitting, while the mRNA remains bound to the SSU via SD interactions (Supplementary Fig. S7E). In the absence of an SD sequence, it is possible that the mRNA would dissociate from the 30S subunit. Structures of intermediates of mammalian mitochondrial recycling have suggested that the tRNA stays bound to the 50S after splitting [44]. There is weak density for tRNA present in 50S•RRF•EF-G•FA, but we cannot exclude the possibility that this reflects binding of excess tRNA.

Ribosome splitting by EF-G and RRF is faster in the presence of deacylated P-site tRNA than for vacant ribosomes [45], consistent with previous findings that deacylated tRNA favors the rotated ribosome conformation [11], required for binding of RRF and EF-G [41]. In addition, when P-site tRNA is absent, it is possible (but not required) for H69 to disengage from B2a (this work and [46, 47]), suggesting that upon RRF-mediated B2a breakage, the transmission of the conformational change leading to splitting is less efficient when the P site is empty. Ribosome splitting is three-fold slower with tRNAfMet than with tRNAPhe [45]. This can be explained by the interactions between the conserved G-C base pairs of initiator tRNAs and the SSU P site [48], which according to our structure, would need to break during recycling. In support of the requirement for the displacement of the P-site tRNA to break B2a, previous 70S•RRF structures that observed changes in H69 [16, 18] all lacked P-site tRNA, as in postTC•RRF-LSU-2. In agreement with biochemistry showing no translocation of mRNA during recycling [43, 49], the 70S•RRF•EF-G•FA structure shows a displaced tRNA with disordered anticodon loop that does not appear positioned to make a cognate codon-anticodon interaction with the mRNA (Supplementary Fig. S7E and F).

With these new structural insights, we can propose a model for the mechanism of ribosome recycling and FA inhibition (Fig. 5). Following termination, a deacylated tRNA remains paired to the P-site mRNA codon on the 70S ribosome (postTC). This state allows SSU rotation, bringing the tRNA into the p/E site. RRF binds and stabilizes the rotated ribosome [41, 45], with domain I fixed by LSU interactions and domain II adopting multiple orientations (Fig. 2). In the most populated state, domain II is positioned toward the SSU (postTC•RRF-SSU). Only this state is sterically compatible with initial binding of EF-G•GTP, quickly triggering GTP hydrolysis [19]. In the resulting complex, domains III-IV of EF-G push RRF toward inter-subunit bridge B2a, where RRF interactions cause rearrangement of H69 and h44, leading to disruption of bridge B2a (Fig. 4B and C). In addition, conformational changes of h44 and h45 cause the SSU to back-rotate (Supplementary Fig. S7C), shifting the tRNA toward the SSU E site (Fig. 4D) and rendering the anticodon loop more dynamic (Supplementary Fig. S7D). The resulting conformational changes propagate to bridges B1a, B1b, B7b, B6, and B8, which are broken or altered (Supplementary Figs S9 and S10), consistent with subunit splitting and tRNA release. This would be followed by departure of EF-G•GDP from the 50S, followed by dissociation of RRF.

Figure 5.

Schematic figure of the proposed model for ribosome recycling and FA inhibition. The figure is made with a simplified representation of each structure.

Open in a new tab

Mechanism of ribosome recycling and inhibition by FA. The postTC comprises the ribosome (LSU in light blue and SSU in light yellow), deacylated tRNA (yellow) and mRNA (brown). (A) EF-G (green) can repeatedly bind to the postTC, hydrolyze GTP and dissociate. RRF cannot bind to postTC•EF-G. EF-G dissociation is inhibited by FA. This is likely the main mode of FA inhibition of ribosome recycling. (B) postTC. (C) Upon binding to postTC, RRF (salmon) adopts multiple orientations of domain II. (D–E) EF-G binds to the RRF-bound postTC and hydrolyzes GTP. Their combined action induces SSU back-rotation, disrupts inter-subunit bridges and moves the tRNA out of the SSU P site, leading to splitting of the 70S ribosome. In the absence of a nearby SD-sequence, tRNA and mRNA can be released. FA inhibits splitting, but not tRNA and mRNA release. Upon FA dissociation, splitting can proceed. (E) FA can also inhibit dissociation of RRF and EF-G from the 50S. In the 50S complex, the conformations of RRF and EF-G are not compatible with a bound 30S. (F) Release of RRF and EF-G from the 50S subunit.

Several studies have demonstrated the co-evolution of EF-G and RRF to ensure productive interactions during ribosome recycling [50–52]. We compared our 70S•RRF•EF-G•FA complex with a heterologous postTC with Thermus thermophilus RRF, E. coli EF-G, and FA [14]. In the model of the heterologous complex, domain II of TtRRF is positioned between domains IV and V, on the opposite side of domain IV of EcEF-G rather than approaching bridge B2a, providing a structural rationale for why TtRRF and EcEF-G are incompatible and unable to promote subunit splitting [50]. However, the 10 Å resolution of the map limits confidence in such detailed interpretations. Still, that study proposed a recycling mechanism involving back-rotation of the SSU coupled with tRNA movement to the E site [14], in line with our observations. The interactions observed in our structure are also inconsistent with those seen in the crystal structure of a non-rotated 70S ribosome bound to RRF and an engineered EF-G-bL9 fusion (Supplementary Fig. S11), in which EF-G adopts a compact conformation with domain IV oriented away from RRF, while two tRNAs occupy the E site and a so-called p/R (peptidyl/recycling) state [53]. How this conformation, formed only upon addition of GDP, fits into the mechanism of recycling remains unclear. Neither of these structures challenge our interpretation of the S. aureus complexes, in which both the ribosomes and recycling factors are derived from the same species.

Despite its low abundance, the 70S•RRF•EF-G•FA complex offers clues to how FA influences the final stages of ribosome recycling. The 70S•RRF•EF-G•FA complex is considerably less abundant than other EF-G•FA complexes, and displays multiple disrupted inter-subunit bridges (Supplementary Figs S9 and S10), as well as limited contacts between EF-G and the SSU. According to biochemical studies with E. coli components, FA inhibits subunit splitting [21, 27], but not the release of tRNA and mRNA [21]. Thus, the 70S•RRF•EF-G•FA complex is expected to remain stable while FA is bound; on the order of ~5 seconds, but remains primed for ribosome splitting upon FA dissociation [27]. The observed wedging of RRF domain II into bridge B2a and the displacement of the P-site anticodon stem from the SSU P site are consistent with the idea that 70S•RRF•EF-G•FA represents an on-pathway intermediate in recycling that would proceed to splitting upon FA dissociation. The 50S•RRF•EF-G•FA structure shows a further shift of RRF domain II that would severely clash with the SSU (Fig. 4 F and G), suggesting that a similar movement in the absence of FA would complete subunit splitting. In this model, the abundant 50S•RRF•EF-G•FA complex could arise either from FA rebinding after subunit splitting but before EF-G dissociation from 50S•RRF, or from binding of EF-G and FA to a pre-existing 50S•RRF complex.

Our structural data would also be compatible with a model in which the S. aureus 70S•RRF•EF-G•FA complex is splitting-competent, and in which RRF and EF-G remain trapped on the resulting 50S product, but this mechanism would be distinct from that proposed for the E. coli system. However, our structures and population distribution support a model where FA, under these conditions, does not primarily inhibit ribosome recycling by preventing splitting of postTCs bound to RRF and EF-G. Instead, both models support the idea that FA primarily inhibits recycling by trapping EF-G on the postTC in the absence of RRF, thereby preventing RRF binding and blocking the splitting reaction [27].

Measurements of GTP hydrolysis by EF-G per postTC splitting event at varying RRF concentrations showed that, under saturating RRF concentrations, the fraction of non-productive GTP hydrolysis cycles not leading to subunit splitting is negligible [19]. Further, the GTP consumption per ribosome splitting event is not affected by FA [27]. Based on the cellular concentrations of RRF and EF-G in E. coli, it was estimated that five GTP molecules are consumed per ribosome-splitting event in vivo [19]. This suggests that the majority of postTC•EF-G complexes available for FA binding lack RRF and that, during recycling, this mode of inhibition may also be predominant in vivo.

During translocation, GTP hydrolysis and Pi release from EF-G unlocks the decoding center and brings the ribosome into a chimeric SSU state, through back-rotation and head swivel [6, 7]. These changes involve direct interactions between EF-G and the SSU. During recycling, by contrast, EF-G also promotes SSU back-rotation, but this is mediated by RRF, which additionally leads to disruption of inter-subunit bridges. This results in a state with very limited contacts between EF-G and the SSU, but extensive contacts between EF-G and RRF. The requirement for GTP hydrolysis for ribosome splitting was biochemically demonstrated [1, 49] and later confirmed by single-molecule experiments with the non-hydrolysable GTP analogue GDPNP [41]. This indicates that EF-G•GDPNP is incapable of inducing the required conformational changes in RRF [43], stalling the reaction at an earlier state than observed with FA, which binds after Pi release. Interestingly, the earliest structural states of EF-G during translocation, with tRNA in the a/P state, prior to Pi release [6], would severely clash with RRF in postTC•RRF-SSU (Supplementary Fig. S12A), and the GDP state of EF-G would also clash with RRF in 70S•RRF•EF-G•FA (Supplementary Fig. S12B). This indicates that both upon initial binding, prior to Pi release, and at the end of the reaction, the conformation of EF-G must differ during recycling from that during translocation. However, time-resolved cryo-EM suggests that EF-G domains I-II dissociate first from the ribosome after translocation, leading to small conformational changes in EF-G domains III-V [6]. A similar sequence of events might occur during recycling, and the conformational state of domains IV-V observed in the 50S•RRF•EF-G•FA structure may correspond to such a late-stage intermediate.

In summary, this study captured FA inhibition of ribosome recycling in S. aureus, visualizing intermediates in which inter-subunit bridges are disrupted by RRF•EF-G, leading to SSU back-rotation, tRNA movement, and ultimately subunit splitting. The data support a model in which FA primarily inhibits recycling by trapping EF-G to the postTC in the absence of RRF rather than preventing EF-G from completion of RRF-mediated ribosome splitting. Together, these findings provide structural insights into the mechanism of RRF•EF-G-mediated recycling and its inhibition by FA.

Supplementary Material

gkag778_Supplemental_Files
gkag778_supplemental_files.zip (104.3MB, zip)

Acknowledgements

We thank Måns Ehrenberg and Anneli Borg for critical discussion of the results. We thank Daniel Larsson for cryo-EM data collection, discussions, and comments on the manuscript and Shirin Akbar for comments on the manuscript. We acknowledge the use of the Cryo-EM Uppsala facility for grid preparation and screening, funded by the Department of Cell and Molecular Biology, the Disciplinary Domains of Science and Technology and of Medicine and Pharmacy at Uppsala University. We acknowledge access and support of the cryo-EM facilities at the UK national electron Bio-Imaging Centre, proposal BI42543.

Author contributions: Adrián González-López (Conceptualization [equal], Data curation [lead], Formal analysis [lead], Investigation [lead], Validation [lead], Visualization [lead], Writing – original draft [lead], Writing – review & editing [equal]) and Maria Selmer (Conceptualization [equal], Data curation [supporting], Formal analysis [supporting], Funding acquisition [lead], Investigation [supporting], Resources [lead], Supervision [lead], Validation [supporting], Visualization [supporting], Writing – original draft [supporting], Writing – review & editing [equal])

Contributor Information

Adrián González-López, Department of Cell and Molecular Biology, Uppsala University, Uppsala 75124, Sweden; Uppsala Antibiotic Center, Uppsala University, Uppsala 75124, Sweden.

Maria Selmer, Department of Cell and Molecular Biology, Uppsala University, Uppsala 75124, Sweden; Uppsala Antibiotic Center, Uppsala University, Uppsala 75124, Sweden.

Supplementary data

Supplementary data is available at NAR online.

Conflict of interest

None declared.

Funding

This work was supported by Uppsala Antibiotic Center (to M.S.), the Swedish Research Council (2022-04511 to M.S), Sven and Lilly Lawski Foundation (to A.G.L.), and Helge Ax:son Johnsons stiftelse (to A.G.L.). Funding to pay the Open Access publication charges for this article was provided by Uppsala University.

Data availability

The models and cryo-EM maps in this study have been deposited in the Protein Data Bank under the DOIs https://doi.org/10.2210/pdb9ti5/pdb (postTC•RRF-SSU), https://doi.org/10.2210/pdb9ti6/pdb (postTC•RRF-LSU), https://doi.org/10.2210/pdb9ti7/pdb (postTC•RRFLSU-2), https://doi.org/10.2210/pdb9ti8/pdb (50S•RRF•EF-G•FA), and https://doi.org/10.2210/pdb9ti9/pdb (70S•RRF•EF-G•FA), and Electron Microscopy Data Bank (https://www.ebi.ac.uk/emdb/) under the accession codes EMD-55944 (postTC•RRF-SSU), EMD-55945 (postTC•RRFLSU), EMD-55946 (postTC•RRF-LSU-2), EMD-55947 (50S•RRF•EF-G•FA), and EMD-55948 (70S•RRF•EF-G•FA). Raw cryo-EM movies, particle stacks, and extra maps have been deposited in the Electron Microscopy Image Archive (https://www.ebi.ac.uk/empiar/) under the accession code EMPIAR-13144. Custom python scripts used during modeling and analysis are available on Git Hub (https://github.com/adriangl97/pdb_python_tools) and Zenodo (https://doi.org/10.5281/zenodo.21379912).

References

  • 1. Karimi  R, Pavlov  MY, Buckingham  RH  et al.  Novel roles for classical factors at the interface between translation termination and initiation. Mol Cell. 1999;3:601–9. 10.1016/S1097-2765(00)80353-6. [DOI] [PubMed] [Google Scholar]
  • 2. Hirashima  A, Kaji  A. Role of elongation factor G and a protein factor on the release of ribosomes from messenger ribonucleic acid. J Biol Chem. 1973;248:7580–7. 10.1016/S0021-9258(19)43330-9. [DOI] [PubMed] [Google Scholar]
  • 3. Lancaster  L, Kiel  MC, Kaji  A  et al.  Orientation of ribosome recycling factor in the ribosome from directed hydroxyl radical probing. Cell. 2002;111:129–40. 10.1016/S0092-8674(02)00938-8. [DOI] [PubMed] [Google Scholar]
  • 4. Wilson  DN, Schluenzen  F, Harms  JM  et al.  X-ray crystallography study on ribosome recycling: the mechanism of binding and action of RRF on the 50S ribosomal subunit. EMBO J. 2005;24:251–60. 10.1038/sj.emboj.7600525. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Rundlet  EJ, Holm  M, Schacherl  M  et al.  Structural basis of early translocation events on the ribosome. Nature. 2021;595:741–5. 10.1038/s41586-021-03713-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Carbone  CE, Loveland  AB, Gamper  HB  et al.  Time-resolved cryo-EM visualizes ribosomal translocation with EF-G and GTP. Nat Commun. 2021;12:7236. 10.1038/s41467-021-27415-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Petrychenko  V, Peng  B-Z, de  AP  et al.  Structural mechanism of GTPase-powered ribosome-tRNA movement. Nat Commun. 2021;12:5933. 10.1038/s41467-021-26133-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Seely  SM, Gagnon  MG. Mechanisms of ribosome recycling in bacteria and mitochondria: a structural perspective. RNA Biology. 2022;19:662–77. 10.1080/15476286.2022.2067712. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Selmer  M, Al-Karadaghi  S, Hirokawa  G  et al.  Crystal structure of Thermotoga maritima ribosome recycling factor: a tRNA mimic. Science. 1999;286:2349–52. 10.1126/science.286.5448.2349. [DOI] [PubMed] [Google Scholar]
  • 10. Kim  KK, Min  K, Suh  SW. Crystal structure of the ribosome recycling factor from Escherichia coli. EMBO J. 2000;19:2362–70. 10.1093/emboj/19.10.2362. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Cornish  PV, Ermolenko  DN, Noller  HF  et al.  Spontaneous intersubunit rotation in single ribosomes. Mol Cell. 2008;30:578–88. 10.1016/j.molcel.2008.05.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Agrawal  RK, Sharma  MR, Kiel  MC  et al.  Visualization of ribosome-recycling factor on the Escherichia coli 70S ribosome: functional implications. Proc Natl Acad Sci USA. 2004;101:8900–5. 10.1073/pnas.0401904101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Fu  Z, Kaledhonkar  S, Borg  A  et al.  Key intermediates in ribosome recycling visualized by time-resolved cryoelectron microscopy. Structure. 2016;24:2092–101. 10.1016/j.str.2016.09.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Yokoyama  T, Shaikh  TR, Iwakura  N  et al.  Structural insights into initial and intermediate steps of the ribosome-recycling process: recycling of translational post-termination complex. EMBO J. 2012;31:1836–46. 10.1038/emboj.2012.22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Dunkle  JA, Wang  L, Feldman  MB  et al.  Structures of the bacterial ribosome in classical and hybrid states of tRNA binding. Science. 2011;332:981–4. 10.1126/science.1202692. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Pai  RD, Zhang  W, Schuwirth  BS  et al.  Structural insights into ribosome recycling factor interactions with the 70S ribosome. J Mol Biol. 2008;376:1334–47. 10.1016/j.jmb.2007.12.048. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Weixlbaumer  A, Petry  S, Dunham  CM  et al.  Crystal structure of the ribosome recycling factor bound to the ribosome. Nat Struct Mol Biol. 2007;14:733–7. 10.1038/nsmb1282. [DOI] [PubMed] [Google Scholar]
  • 18. Borovinskaya  MA, Pai  RD, Zhang  W  et al.  Structural basis for aminoglycoside inhibition of bacterial ribosome recycling. Nat Struct Mol Biol. 2007;14:727–32. 10.1038/nsmb1271. [DOI] [PubMed] [Google Scholar]
  • 19. Borg  A, Pavlov  M, Ehrenberg  M. Complete kinetic mechanism for recycling of the bacterial ribosome. RNA. 2016;22:10–21. 10.1261/rna.053157.115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Kiel  MC, Raj  VS, Kaji  H  et al.  Release of ribosome-bound ribosome recycling factor by elongation factor G. J Biol Chem. 2003;278:48041–50. 10.1074/jbc.M304834200. [DOI] [PubMed] [Google Scholar]
  • 21. Chen  Y, Kaji  A, Kaji  H  et al.  The kinetic mechanism of bacterial ribosome recycling. Nucleic Acids Res. 2017;45:10168–77. 10.1093/nar/gkx694. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Gao  Y-G, Selmer  M, Dunham  CM  et al.  The structure of the ribosome with elongation factor G trapped in the posttranslocational state. Science. 2009;326:694–9. 10.1126/science.1179709. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. González-López  A, Larsson  DSD, Koripella  RK  et al.  Structures of the Staphylococcus aureus ribosome inhibited by fusidic acid and fusidic acid cyclopentane. Sci Rep. 2024;14:14253. 10.1038/s41598-024-64868-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Bodley  JW, Zieve  FJ, Lin  L  et al.  Formation of the ribosome-G factor-GDP complex in the presence of fusidic acid. Biochem Biophys Res Commun. 1969;37:437–43. 10.1016/0006-291X(69)90934-6. [DOI] [PubMed] [Google Scholar]
  • 25. Borg  A, Holm  M, Shiroyama  I  et al.  Fusidic acid targets elongation factor G in several stages of translocation on the bacterial ribosome. J Biol Chem. 2015;290:3440–54. 10.1074/jbc.M114.611608. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Savelsbergh  A, Rodnina  MV, Wintermeyer  W. Distinct functions of elongation factor G in ribosome recycling and translocation. RNA. 2009;15:772–80. 10.1261/rna.1592509. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Borg  A, Pavlov  M, Ehrenberg  M. Mechanism of fusidic acid inhibition of RRF- and EF-G-dependent splitting of the bacterial post-termination ribosome. Nucleic Acids Res. 2016;44:3264–75. 10.1093/nar/gkw178. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Godtfredsen  W, Roholt  K, Tybring  L. Fucidin: a new orally active antibiotic. Lancet. 1962;279:928–31. 10.1016/S0140-6736(62)91968-2. [DOI] [PubMed] [Google Scholar]
  • 29. Krishnaswamy  S, Akbar  S, Larsson  DSD  et al.  Doublet decoding of tRNASer3 demonstrates plasticity of ribosomal decoding center. Nat Commun. 2025;16:5402. 10.1038/s41467-025-61016-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Punjani  A, Rubinstein  JL, Fleet  DJ  et al.  cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination. Nat Methods. 2017;14:290–6. 10.1038/nmeth.4169. [DOI] [PubMed] [Google Scholar]
  • 31. Liebschner  D, Afonine  PV, Baker  ML  et al.  Macromolecular structure determination using X-rays, neutrons and electrons: recent developments in Phenix. Acta Crystallogr D Struct Biol. 2019;75:861–77. 10.1107/S2059798319011471. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Scheres  SHW. RELION: implementation of a Bayesian approach to cryo-EM structure determination. J Struct Biol. 2012;180:519–30. 10.1016/j.jsb.2012.09.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Pettersen  EF, Goddard  TD, Huang  CC  et al.  UCSF ChimeraX: structure visualization for researchers, educators, and developers. Protein Sci. 2021;30:70–82. 10.1002/pro.3943. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. González-López  A, Ge  X, Larsson  DSD  et al.  Structural mechanism of FusB-mediated rescue from fusidic acid inhibition of protein synthesis. Nat Commun. 2025;16:3693. 10.1038/s41467-025-58902-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Abramson  J, Adler  J, Dunger  J  et al.  Accurate structure prediction of biomolecular interactions with AlphaFold 3. Nature. 2024;630:493–500. 10.1038/s41586-024-07487-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Emsley  P, Cowtan  K. Coot : model-building tools for molecular graphics. Acta Crystallogr D Biol Crystallogr. 2004;60:2126–32. 10.1107/S0907444904019158. [DOI] [PubMed] [Google Scholar]
  • 37. Yamashita  K, Palmer  CM, Burnley  T  et al.  Cryo-EM single-particle structure refinement and map calculation using Servalcat. Acta Crystallogr D Struct Biol. 2021;77:1282–91. 10.1107/S2059798321009475. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Crowe-McAuliffe  C, Murina  V, Turnbull  KJ  et al.  Structural basis of ABCF-mediated resistance to pleuromutilin, lincosamide, and streptogramin A antibiotics in Gram-positive pathogens. Nat Commun. 2021;12:3577. 10.1038/s41467-021-23753-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Gao  N, Zavialov  AV, Ehrenberg  M  et al.  Specific interaction between EF-G and RRF and its implication for GTP-dependent ribosome splitting into subunits. J Mol Biol. 2007;374:1345–58. 10.1016/j.jmb.2007.10.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Ali  IK, Lancaster  L, Feinberg  J  et al.  Deletion of a conserved, central ribosomal intersubunit RNA bridge. Mol Cell. 2006;23:865–74. 10.1016/j.molcel.2006.08.011. [DOI] [PubMed] [Google Scholar]
  • 41. Prabhakar  A, Capece  MC, Petrov  A  et al.  Post-termination ribosome intermediate acts as the gateway to ribosome recycling. Cell Rep. 2017;20:161–72. 10.1016/j.celrep.2017.06.028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Hirokawa  G. Post-termination complex disassembly by ribosome recycling factor, a functional tRNA mimic. EMBO J. 2002;21:2272–81. 10.1093/emboj/21.9.2272. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Zavialov  AV, Hauryliuk  VV, Ehrenberg  M. Splitting of the posttermination ribosome into subunits by the concerted action of RRF and EF-G. Mol Cell. 2005;18:675–86. 10.1016/j.molcel.2005.05.016. [DOI] [PubMed] [Google Scholar]
  • 44. Koripella  RK, Deep  A, Agrawal  EK  et al.  Distinct mechanisms of the human mitoribosome recycling and antibiotic resistance. Nat Commun. 2021;12:3607. 10.1038/s41467-021-23726-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45. Pavlov  MY, Antoun  A, Lovmar  M  et al.  Complementary roles of initiation factor 1 and ribosome recycling factor in 70S ribosome splitting. EMBO J. 2008;27:1706–17. 10.1038/emboj.2008.99. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Morgan  CE, Huang  W, Rudin  SD  et al.  Cryo-electron microscopy structure of the Acinetobacter baumannii 70S ribosome and implications for new antibiotic development. mBio. 2020;11:e03117–19. 10.1128/mbio.03117-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47. Murphy  EL, Singh  KV, Avila  B  et al.  Cryo-electron microscopy structure of the 70S ribosome from Enterococcus faecalis. Sci Rep. 2020;10:16301. 10.1038/s41598-020-73199-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48. Selmer  M, Dunham  CM, Murphy  FV  et al.  Structure of the 70 S ribosome complexed with mRNA and tRNA. Science. 2006;313:1935–42. 10.1126/science.1131127. [DOI] [PubMed] [Google Scholar]
  • 49. Peske  F, Rodnina  MV, Wintermeyer  W. Sequence of steps in ribosome recycling as defined by kinetic analysis. Mol Cell. 2005;18:403–12. 10.1016/j.molcel.2005.04.009. [DOI] [PubMed] [Google Scholar]
  • 50. Fujiwara  T, Ito  K, Nakayashiki  T  et al.  Amber mutations in ribosome recycling factors of Escherichia coli and Thermus thermophilus : evidence for C-terminal modulator element. FEBS Lett. 1999;447:297–302. 10.1016/S0014-5793(99)00302-6. [DOI] [PubMed] [Google Scholar]
  • 51. Ito  K, Fujiwara  T, Toyoda  T  et al.  Elongation factor G participates in ribosome disassembly by interacting with ribosome recycling factor at their tRNA-mimicry domains. Mol Cell. 2002;9:1263–72. 10.1016/S1097-2765(02)00547-6. [DOI] [PubMed] [Google Scholar]
  • 52. Raj  VS, Kaji  H, Kaji  A. Interaction of RRF and EF-G from E. coli and T. thermophilus with ribosomes from both origins—insight into the mechanism of the ribosome recycling step. RNA. 2005;11:275–84. 10.1261/rna.7201805. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53. Zhou  D, Tanzawa  T, Lin  J  et al.  Structural basis for ribosome recycling by RRF and tRNA. Nat Struct Mol Biol. 2020;27:25–32. 10.1038/s41594-019-0350-7. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

gkag778_Supplemental_Files
gkag778_supplemental_files.zip (104.3MB, zip)

Data Availability Statement

The models and cryo-EM maps in this study have been deposited in the Protein Data Bank under the DOIs https://doi.org/10.2210/pdb9ti5/pdb (postTC•RRF-SSU), https://doi.org/10.2210/pdb9ti6/pdb (postTC•RRF-LSU), https://doi.org/10.2210/pdb9ti7/pdb (postTC•RRFLSU-2), https://doi.org/10.2210/pdb9ti8/pdb (50S•RRF•EF-G•FA), and https://doi.org/10.2210/pdb9ti9/pdb (70S•RRF•EF-G•FA), and Electron Microscopy Data Bank (https://www.ebi.ac.uk/emdb/) under the accession codes EMD-55944 (postTC•RRF-SSU), EMD-55945 (postTC•RRFLSU), EMD-55946 (postTC•RRF-LSU-2), EMD-55947 (50S•RRF•EF-G•FA), and EMD-55948 (70S•RRF•EF-G•FA). Raw cryo-EM movies, particle stacks, and extra maps have been deposited in the Electron Microscopy Image Archive (https://www.ebi.ac.uk/empiar/) under the accession code EMPIAR-13144. Custom python scripts used during modeling and analysis are available on Git Hub (https://github.com/adriangl97/pdb_python_tools) and Zenodo (https://doi.org/10.5281/zenodo.21379912).

Articles from Nucleic Acids Research are provided here courtesy of Oxford University Press

RESOURCES