New insights into the enzymatic role of EF-G in ribosome recycling

Abstract

During translation, elongation factor G (EF-G) plays a catalytic role in tRNA translocation and a facilitative role in ribosome recycling. By stabilizing the rotated ribosome and interacting with ribosome recycling factor (RRF), EF-G was hypothesized to induce the domain rotations of RRF, which subsequently performs the function of splitting the major intersubunit bridges and thus separates the ribosome into subunits for recycling. Here, with systematic mutagenesis, FRET analysis and cryo-EM single particle approach, we analyzed the interplay between EF-G/RRF and post termination complex (PoTC). Our data reveal that the two conserved loops (loop I and II) at the tip region of EF-G domain IV possess distinct roles in tRNA translocation and ribosome recycling. Specifically, loop II might be directly involved in disrupting the main intersubunit bridge B2a between helix 44 (h44 from the 30S subunit) and helix 69 (H69 from the 50S subunit) in PoTC. Therefore, our data suggest a new ribosome recycling mechanism which requires an active involvement of EF-G. In addition to supporting RRF, EF-G plays an enzymatic role in destabilizing B2a via its loop II.

Figure 1.

Function of EF-G during translocation and ribosome recycling. (A) Brief model of EF-G catalyzed translocation. The process from PRE to POST state is promoted by GTP hydrolysis. (B) Brief model of ribosome recycling process. EF-G and RRF function together to dissociate the PoTC into subunits. (C-E) Interactions of EF-G loop II (blue) and RRF (green) with ribosomal intersubunit bridge B2a. The B2a contains H69 (gray) of 23S rRNA and h44 (lavender) of 16S rRNA. Both EF-G and RRF interact with this bridge. The model of the POST-state ribosome(C, D) was extracted from a previous crystal structure (PDB 4V5F (19)). The model of PoTC•RRF (E) was also from a crystallography study (PDB 4V9D(26)). (F) Effect of loop II deletion and single site mutants on ribosome breakdown. The ribosome sedimentation pattern was analyzed by 15–45% sucrose density gradient ultracentrifugation. Fractions were collected from the bottom to the top of the tube and measured at A₂₅₄. The peaks corresponding to the 70S ribosome, disome, trisome and tetrasome are marked as 70S, 2X, 3X and 4X, respectively. The 70S position is indicated with dashed line throughout. The red colored mutants are effective in both recycling and tRNA translocation, while the blue ones only in tRNA translocation.

Figure 2.

Effect of EF-G single site mutants on total protein synthesis and tRNA translocation. (A) Effect of overexpression of EF-G mutants on the growth of E.coli BL 21. Black arrow indicates the time that IPTG was added. (B and C) Effects of EF-G overexpression on total protein synthesis in vivo. Total soluble proteins from E.coli BL 21 overexpressing EF-G wt or mutants were analyzed by SDS-PAGE (B). After ³⁵S-methionine addition, the ratio of proteins containing ³⁵S-methionine increased over time as new proteins were continued to be synthesized in cells (C). (D) Brief illustration of the pyrene-modified mRNA translocation assay. The translocation action was monitored by tracing the fluorescence of pyrene attached to the 3′ end of the mRNA. The fluorescence emission is high (red star) before translocation and decreases (black star) upon translocation. (E) Relative fluorescence of PRE state ribosomes programmed by the labeled mRNA with incorporation of EF-Gwt or mutants. Error bars, s.e.m. (n = 3 technical replicates). ***P < 0.001.

Figure 3.

Condition optimization to achieve PoTC•EF-G•RRF complex. (A and B) Titration of EF-G wt (A) or mutants (B) to PoTC. Pre: before sucrose cushion ultracentrifugation. Su: supernatant after sucrose cushion ultracentrifugation. Quantification of EF-G occupancy on PoTC is shown beneath each gel. The optimized ratio is colored in red. Error bars, s.e.m. (n = 3 technical replicates). (C) Titration of RRF to PoTC with subsequent addition of EF-G proteins in the presence of GDPNP. (D) Polysomes fractionation and factor detection by western blotting with corresponding antibodies against EF-G, RRF and small ribosomal subunit protein S2. In the fraction of disome (red rectangle), both EF-G and RRF were detected.

Figure 4.

FRET analyses of ribosomal rotation. (A) Locations of fluorescence labeled residues in 30S and 50S subunits. Small ribosomal subunit protein S6 was labeled with AF555 (donor) and large ribosomal subunit protein L9 was labeled with AF647 (acceptor). The black arrow indicates the direction of rotation of PoTC induced by EF-G binding. (B) Fluorescence emission of the S6/L9 ribosome constructs. Black curve is the spectrum of vacant ribosome; blue curve is the PoTC complex in the presence of EF-G·GDPNP. The excitation wavelength was 550 nm. Addition of EF-G wt led to a significant decrease in FRET efficiency, indicated by orange and green rectangles, and enlarged in the right panels. (C) Fluorescence emission of the PoTC in the presence of EF-G mutants Δloop II, Δloop II’ and, S588P.

Figure 5.

Cryo-EM structure of PoTC•EF-G (S588P)•GDPNP. (A) Overview of the PoTC with EF-G (S588P) (red) and a deacylated tRNA (cyan) in the P/E state.To model this complex, the 70S ribosome crystal structure (PDB 4KIX and 4KIY (35)) were used. (B) Zoom-in view of B2a region composed of H69 and h44 (gray), and loop II (blue) of EF-G domain IV. (C) Rotation of the 30S subunit (yellow) relative to the 50S subunit (transparent blue) caused by EF-G (S588P) binding. Compared to the classical state (gray, PDB 4V51 (50)), 30S shows a counter-clockwise rotation by 7°(viewed from the solvent side of 30S subunit). The structures are aligned using the 23S rRNA as reference. (D) Superimposition of the density map of a cryo-EM structure of 70S•EF-G (H91A) with the derived atomic model (PDB 3JA1 (41)). (E) Superimposition of our density map of the PoTC•EF-G (S588P)•GDPNP complex with the atomic model from 3JA1. (F) Superimposition of our density map of the PoTC•EF-G (S588P)• GDPNP complex with the atomic model from 4KIY (35). Compared to (B), the orientation in (D–F) has been turned ∼18° along the long axis of EF-G domain IV.

Figure 6.

Schematic representation of the catalytic role of EF-G loop II in ribosome recycling. At the end of translation, RRF binds to PoTC inducing the conformational change of H69 and h44 to destabilize B2a. Upon EF-G•GTP binding, loop II (blue) locates in the vicinity of B2a. Together with RRF, loop II disassembles the ribosome into the subunits.