Skip to main content
NCBI home page
The EMBO Journal logoLink to The EMBO Journal
. 2010 Oct 15;29(22):3810–3818. doi: 10.1038/emboj.2010.252

tmRNA–SmpB: a journey to the centre of the bacterial ribosome

Félix Weis 1,2, Patrick Bron 3, Emmanuel Giudice 1, Jean-Paul Rolland 1, Daniel Thomas 1, Brice Felden 2,a, Reynald Gillet 1,b
PMCID: PMC2989106  PMID: 20953161

tmRNA–SmpB: a journey to the centre of the bacterial ribosome

Bacterial ribosomes stalled on incomplete mRNAs can be resolved by the ‘trans-translation' quality control mechanism. This cryo-EM study, and the co-published study by Fu et al, explains how the ribonucleoprotein complex tmRNA–SmpB translocates through the ribosome to allow resumed translation on tmRNA.

Keywords: cryo-electron microscopy, ribosome, SmpB, tmRNA, trans-translation

Abstract

Ribosomes mediate protein synthesis by decoding the information carried by messenger RNAs (mRNAs) and catalysing peptide bond formation between amino acids. When bacterial ribosomes stall on incomplete messages, the trans-translation quality control mechanism is activated by the transfer-messenger RNA bound to small protein B (tmRNA–SmpB ribonucleoprotein complex). Trans-translation liberates the stalled ribosomes and triggers degradation of the incomplete proteins. Here, we present the cryo-electron microscopy structures of tmRNA–SmpB accommodated or translocated into stalled ribosomes. Two atomic models for each state are proposed. This study reveals how tmRNA–SmpB crosses the ribosome and how, as the problematic mRNA is ejected, the tmRNA resume codon is placed onto the ribosomal decoding site by new contacts between SmpB and the nucleotides upstream of the tag-encoding sequence. This provides a structural basis for the transit of the large tmRNA–SmpB complex through the ribosome and for the means by which the tmRNA internal frame is set for translation to resume.

Introduction

Protein synthesis is an elaborate process that is conducted by ribosomes in all living organisms. By using aminoacyl-transfer RNAs (tRNAs) as substrates, protein synthesis allows the decoding of messenger RNAs (mRNAs) into a corresponding sequence of amino acids (Schmeing and Ramakrishnan, 2009). When eubacterial ribosomes stall on an mRNA molecule lacking a stop codon, the trans-translation quality control mechanism is carried out by the transfer-messenger RNA bound to small protein B (tmRNA–SmpB ribonucleoprotein complex). tmRNA is a chimeric molecule that acts as both a tRNA and an mRNA; depending on the species, tmRNA ranges from 230 to 400 nucleotides in length. The modular architecture of tmRNA includes a tRNA-like domain (TLD), a string of pseudoknots (PKs), a long and disrupted helix H2 connecting the TLD to the PKs, and a short internal open reading frame (ORF) (Figure 1) (Felden et al, 1997). The TLD includes an acceptor stem that is always charged by an alanine, as well as a T arm and a D loop, but no D stem. SmpB is a small basic protein (∼18 kDa) that binds both the tmRNA and the ribosome with high affinity (Hallier et al, 2006) and is essential to trans-translation (Karzai et al, 1999). With the help of elongation factor EF-Tu and SmpB, tmRNA targets stalled translational complexes and, despite the absence of codon–anticodon interactions, enters the vacant ribosomal A site. The incomplete polypeptide is then transferred to the alanyl-tmRNA, and translation shifts from the problematic mRNA to the tmRNA internal ORF that encodes a peptide tag targeted by several proteases. The stalled ribosomes are recycled because of the presence of the stop codon within the tmRNA, while the incomplete protein and the truncated mRNA are released and degraded specifically (Moore and Sauer, 2007).

Figure 1.

Figure 1

Open in a new tab

Secondary structure diagram of Thermus thermophilus tmRNA. Base pairs are linked by lines, and GU pairs are represented by dots. The nucleotides within the internal ORF are underlined and shown in a larger font, and the resume codon is highlighted in yellow. Domains are shown in the colours used in Figure 3. This figure was adapted from the tmRDB web site (Andersen et al, 2006).

Together with a wealth of genetic and biochemical data, the first cryo-electron microscopy (cryo-EM) reconstructions, combined with X-ray and NMR structures, provided insight into the mechanism by which the tmRNA recognizes and binds stalled ribosomes. First, EF-Tu places the TLD into the A site as for canonical tRNAs, except that SmpB overcomes the absence of codon–anticodon interactions (Shimizu and Ueda, 2002; Kurita et al, 2007; Nonin-Lecomte et al, 2009). The PK ring wraps around the beak of the 30S small ribosomal subunit, placing the internal ORF close to the entrance of the mRNA path (Kaur et al, 2006; Gillet et al, 2007). Then, as the accommodation takes place, the TLD swings into the A site (Bugaeva et al, 2008; Cheng et al, 2010; Weis et al, 2010). Although this working model reveals the initial positions of tmRNA–SmpB in and around the stalled ribosome, it also raises questions about the topological rearrangements required to move a complex approximately six times larger than a tRNA from the A to the P site, and to swap the problematic mRNA with the internal ORF of the tmRNA.

Results and discussion

Cryo-EM reconstruction of the accommodated state of tmRNA–SmpB

In the present cryo-EM study, we first revisited the accommodation of tmRNA–SmpB. This investigation resulted in a new 3D structure at 13 Å resolution, as well as a much better definition of the densities attributable to both counterparts (Figure 2A and C). A portion of tmRNA–SmpB enters deep into the A site, while the rest of the molecule forms an arc around the beak of the small ribosomal subunit. To facilitate interpretation at the molecular level, we docked atomic models into the electron densities. Using our earlier model (Kaur et al, 2006) as the initial starting point, the tmRNA domains (see Figure 1) were manually docked as rigid bodies into the cryo-EM-density map and, as a final step, linked to each other and optimized in the context of the cryo-EM data (see Materials and methods). According to our data, the TLD contacts to the large subunit resemble an accommodated canonical tRNA: the D loop interacts with helix H38, while the acceptor branch brings the CCA 3′ end into the peptidyl transfer centre (PTC) (Figures 3A, 4A, and C; Table I). The lack of a D stem is compensated for by the packing of SmpB into the minor groove of helix H69 of the 23S RNA. This positioning of SmpB realigns helix H2 towards the large subunit, in which it forms extended contacts with protein L11. SmpB interacts with most of the small-subunit regions normally in contact with the anticodon stem loop of the accommodated tRNA, including helix h18 and the top of helix h44 of the 16S RNA and proteins S12/S13 (Figures 3C, 4A, and C; Table I). The 3′ end of the internal ORF is still embedded in helix H5 at the hinge between the head and the body of the 30S subunit. The ring of PKs wraps around the beak of the small subunit, with PK1 in front of the beak of 30S and PK2 lying on the surface of S3. However, PK3/PK4 do not contact the ribosome. Compared with the structure of the pre-accommodated state (Valle et al, 2003a; Kaur et al, 2006), accommodation does not trigger large tmRNA movements, but mainly allows the swing of the acceptor branch into the PTC, allowing transpeptidation with the polypeptide stalled on the P-site tRNA. During this movement, SmpB stays at the same place into the decoding site, but is slightly rotated by ∼30° towards the TLD.

Figure 2.

Figure 2

Open in a new tab

Cryo-EM map of the accommodated tmRNA–SmpB complex on a stalled ribosome, and a side-by-side comparison with the map of the translocated complex. (A, C) Two views of the trans-translational ribosome complex after accommodation. (B, D) Two views of the trans-translational ribosome complex after translocation. The density attributable to tmRNA–SmpB is depicted in red, the 50S subunit is in blue, the 30S subunit is in yellow, and the P-site tRNA in (A) and (C) and E-site tRNA in (B) and (D) are shown in purple. Semi-transparent ribosomal subunits in (C) and (D) emphasize the relative position of tmRNA–SmpB and tRNA in the three active sites of the ribosome before and after translocation. Landmarks on the 50S subunit: L7/L12, stalk of proteins L7/L12; L1, stalk of protein L1; and PTC, peptidyl transfer centre. Landmarks on the 30S subunit: b, beak; sh, shoulder; sp, spur; dc, decoding site; and h, head.

Figure 3.

Figure 3

Open in a new tab

Models for tmRNA and SmpB in the cryo-EM maps of the accommodated and translocated states. (A, B) Interactions of SmpB and tmRNA with the 50S subunit. (C, D) Interactions of SmpB and tmRNA with the 30S subunit. The densities of tmRNA–SmpB are semi-transparent grey. The orientations with respect to the ribosome are defined by the thumbnails. Landmarks on the 50S subunit: L7/L12, stalk of proteins L7/L12; L1, stalk of protein L1; PTC, peptidyl transfer centre; L11, protein L11; and H69, helix 69 of the 23S rRNA. Landmarks on the 30S subunit: b, beak; sh, shoulder; sp, spur; dc, decoding site; and h44, helix 44 of the 16S rRNA. TLD, blue; H2b–c, red; H2d, grey; PK1, orange; single strand between PK1 and PK2, white (resume codon highlighted in yellow); H5, brown; PK2, green; PK3, pink; and PK4, light blue. SmpB is in purple. This colour scheme is also used in Figure 1. For clarity, the 30S subunit and the tRNA are not shown in (A) and (B), and the 50S subunit and the tRNA are not shown in (C) and (D). Insets: orientation of the tmRNA–SmpB complex (in red) with respect to the ribosome. 30S and 50S subunits are in yellow and blue, respectively.

Figure 4.

Figure 4

Open in a new tab

Molecular environment of tmRNA–SmpB complex. Cryo-EM-density map of tmRNA–SmpB (in red) with its neighbouring proteins in the accommodated (A) and translocated (B) states and neighbouring RNA helices of tmRNA–SmpB in the accommodated (C) and translocated (D) states. For clarity, only the proteins and the ribosomal subunits within 5 Å of the tmRNA–SmpB complex are presented. Landmarks: h18, h30–34, h42, and h44 are 16S rRNA helices and H38, H69–71, H81, H84, and H89–92 are 23S rRNA helices. All atomic coordinates were taken from Selmer et al (2006) (PDB code 2J00 and 2J01), except for protein L11, taken from Yusupova et al (2006) (PDB code 2HGJ), and were rigid body fitted into the cryo-EM maps. Insets: orientation of the tmRNA–SmpB complex (in red) with respect to the ribosome. 30S and 50S subunits are in yellow and blue, respectively. The tRNA present in the P site during the accommodation and in the E site after the translocation is in purple.

Table 1. Molecular environment of the A- and P-site tmRNA on both ribosomal subunits.

  A-site tmRNA P-site tmRNA
  30S 50S 30S 50S
TLD   H38/H43/H71/H89/H90/H92 L16/L25 S13 H38/H69/H71/H80/H81/H84/H90/ H92/H93 L5/L16
SmpB h18/h30/h44 S12/S13 H69 h30 S13 H69
H2b–d h32 S19 L11 h42 S13/S19 H38 L5/L11/L31
PK1     h30/h31/h32 S13/S19 H38
MLD S3   h18/h34/h44 S12  
H5 S3/S4/S5   S3/S4/S5  
PK2 S3   S3  
PK3     S10  
PK4        
H, large subunit helix; h, small subunit helix; S, small subunit protein; L, large subunit protein.
Open in a new tab

The tmRNA ring crosses the ribosome

To understand how such a large RNA molecule navigates through the ribosome, we determined the structure of tmRNA–SmpB translocated into the P site of the ribosome at 13.5 Å resolution (Figure 2B and D). This new cryo-EM structure shows that the overall density of the translocated tmRNA–SmpB complex is as large and structured as in the accommodated step (Figure 2A and C; Supplementary Figure S1). While some portions of the tmRNA may unfold as elongation continues (Wower et al, 2005), the present structure shows that, contrary to current belief, the overall tmRNA structure is conserved during P-site translocation.

TLD–SmpB contacts the ribosome in the same way as a canonical P-site tRNA (Selmer et al, 2006). On the large 50S subunit, SmpB still compensates for the lack of a D stem by contacting H69 (Figures 3B, 4B, and D; Table I), while on the small 30S subunit, it still mimics an anticodon stem loop by interacting with helix h30 of the 16S RNA and protein S13 (Figures 3D, 4B, and D; Table I). Nevertheless, the transit of the large ring to the P site requires a new gap to allow it to progress between the two ribosomal subunits. Twenty-eight per cent of the initial data set for the accommodated state accounts for the ribosome in the ratcheted state, an intermediate known to affect the intersubunit bridge connections on the pretranslocational ribosomes (Frank and Agrawal, 2000). The corresponding map (Figure 5B) shows an intersubunit rotation and a characteristic movement of the L1 stalk towards the E site (Valle et al, 2003b). Following ratcheting, no changes were observed in the tmRNA–SmpB position, except for the acceptor arm of the peptidyl-tmRNA, the weak density of which accounts for its movement into the A/P hybrid position. The deacylated tRNA is recovered in a P/E hybrid state. Further conformational changes include the movement of the tip of helix H38 of the 23S RNA (also known as the A-site finger or ASF, part of bridge B1a) from protein S13 to protein S19 (Reblova et al, 2010). It is known that the ASF is involved in translocation by interacting directly with the A-site tRNA and maintaining the reading frame of canonical mRNAs (Komoda et al, 2006). In the accommodated state, the tip of H38 physically prevents the transit of the tmRNA (Figure 5A). An up-and-down motion of the ASF during ratcheting opens the gate (Figure 5B), allowing helix H2 of the tmRNA to pass to the other side of the bridge (Figure 5C). There, H2 makes additional contacts with the H38 tip of the 23S RNA and with proteins S13/S19 (Figure 4B and D). Interestingly, it was shown recently that mutations in the ASF can inhibit tmRNA tagging in some contexts in vivo (Crandall et al, 2010).

Figure 5.

Figure 5

Open in a new tab

A-site finger motion as the tmRNA–SmpB complex translocates to the P site. (A) Cryo-EM map of the 70S ribosome with tmRNA–SmpB in the accommodated state. (B) Cryo-EM map of the 70S ribosome in the ratcheted state prior to the translocation of tmRNA–SmpB. Note the motion of the H38 tip, which permits the transit of tmRNA helix H2 (white arrow). (C) Cryo-EM map of the 70S ribosome with tmRNA–SmpB in a translocated state. Helix H2 of the tmRNA moves to the other side of bridge B1a, while PK1 remains at the entrance of the gate. Landmarks are as follows: H38, helix 38 of the 23S rRNA; CP, central protuberance; and b, beak. PK1, PK4, and H2b–c refer to tmRNA domains described in Figure 1.

Based on our current results, we propose that trans-translation could be facilitated by this disruption, favouring the passage of H2 from the A to the P site. At the same time, PK1 stays at the entrance of the gate, approximately where H2 was positioned in the accommodated state (Figure 5A and C). At this stage, the function of PK1 is mainly structural, as it can be replaced with a variety of hairpin structures (Tanner et al, 2006). As helix H2d connects both PK1 and PK4, the 3′ end of PK4 also follows the dynamics of H2, although it does not yet make any contacts with the ribosome. In contrast, the 5′ region of PK4, together with H5/PK2/PK3, forms a static block packed against the hinge between the head and the body of the small subunit. Accordingly, replacing each of these PKs, except PK1, with a single-stranded RNA does not affect the functions of tmRNA, confirming that PK2, PK3, and PK4 do not interact in a functional manner with the ribosome during the trans-translation process (Nameki et al, 2000). However, the tension between the large movements of TLD/H2/PK1/PK4 and the immobility of H5/PK2/PK3 stretches the ring of PKs to follow the translocation movement, causing it to become ellipsoidal around the 30S subunit beak and triggering the unfolding of the internal ORF (Figure 3C and D; Supplementary Movie S1).

Contributions of SmpB to decoding

Fitting the atomic model into the cryo-EM-density map confirmed the standard interactions of SmpB with the TLD (Gutmann et al, 2003; Bessho et al, 2007), but also revealed new contacts between the protein and the nucleotides upstream of the tmRNA tag-encoding sequence (Figures 3D and 6). Indeed, in the crowded environment of the ribosome, placing this region in close contact with SmpB is the only solution to position the resume codon into the decoding site, while maintaining enough space for the next incoming tRNA. The new frame has to be determined in a unique manner, because the resume codon is not specified by an initiator tRNA or the canonical initiation factors. The five nucleotides just upstream of the tag-encoding sequence are particularly important for resuming translation onto tmRNA internal ORF (Williams et al, 1999; Lee et al, 2001; Miller et al, 2008). Based on several mutations introduced in the upstream of the tag-encoding region on tmRNA (Konno et al, 2007) and on a genetic selection for tmRNA activity in vivo (Watts et al, 2009), it was shown that SmpB and the sequence upstream of the tmRNA template are functionally linked to direct frame selection.

Figure 6.

Figure 6

Open in a new tab

Docking of SmpB into the cryo-EM density of the translocated complex (close-up view). Interactions of SmpB with tmRNA indicate two different binding sites. While the standard binding site links part of the OB (oligonucleotide/oligosaccharide binding) fold of SmpB with the TLD, the second binding site links the hydrophobic pocket at the bottom of the protein and the C-terminal tail of SmpB to the upstream region of the tmRNA resume codon. The resume codon is in yellow; helices h30 and h34 of the 16S rRNA are in pink; the TLD is in grey; and SmpB is in purple. Helices α1 and α3, strands β8 and β9, and the C-terminal helix are indicated. Locations of SmpB mutations known to affect frame choice are highlighted in orange (Watts et al, 2009).

Strikingly, our model links together these two regions: one area comprising the five nucleotides upstream of the resume codon, and a second region formed by the hydrophobic pocket comprising helix α1, the loop between α3 and β8, β9, and the C-terminal tail of SmpB (Figure 6). Accordingly, this region possesses the highest binding affinity for SmpB (Metzinger et al, 2008). This SmpB position is obviously instrumental in specifying the correct resume codon when translation shifts onto the tmRNA.

mRNA swapping

Switching from trans-peptidation to trans-translation requires a swap between the original truncated mRNA that is stalled in the mRNA path and the internal tmRNA ORF (Ivanova et al, 2005). On the accommodated (Supplementary Figure S2A) and ratcheted (not shown) state maps, the presence of the stalled mRNA is confirmed by the extra density found between the head and the back of the 30S subunit platform, where the Shine–Dalgarno interaction protrudes from the ribosome surface. Accordingly, the translocation of peptidyl-tmRNA to the P site is accompanied by the disappearance of this structural element and a weak E-site tRNA density, indicating a concomitant destabilization of the stalled mRNA from the ribosome when the tmRNA resume codon is moved into the decoding site (Supplementary Figure S2B). When trans-translation triggers stalled polysomes, one can, therefore, presume a sequential delivery of stalled ribosomes one by one at the 3′ end of the problematic mRNAs. After each trans-translation event, the flawed mRNA will then be subject to competition between the polysome moving forward towards its 3′ end, and the 3′-to-5′ exoribonuclease R ready to take it for degradation (Richards et al, 2006).

Navigating through the ribosome

To summarize, the present cryo-EM structures address a number of questions regarding the process of translocation of tmRNA–SmpB. After transpeptidation, the complex moves to the P site as a canonical tRNA, except that codon:anticodon pairing is supported by SmpB. Owing to its large size and complicated topology, the gap in bridge B1a triggered by the ratcheting of the ribosome is necessary for the transit of tmRNA. Helix H2 crosses the bridge while PK1 stays on the other side. PK4 follows this dynamic, but almost all of the rest of the molecule stays in place around the beak of the small subunit. The truncated mRNA is immediately released, while simultaneous binding of SmpB to the TLD and to the upstream region of the tagging sequence sets the resume codon into the decoding site. At this step, the internal ORF is only partially unfolded, as its internal helix H5 is still highly structured. This model lends support to a mechanism in which the next transition of tmRNA–SmpB, from the P site to the E site, would follow the same dynamics. In line with the first translocation, the movement of the TLD to the E site is likely to drag PK1 over bridge B1a. Accordingly, an isolated tmRNA–ribosomal complex blocked with tmRNA in the E site was recently analysed using chemical probing. The model resulting from this study suggests a positioning of PK1 and H2 near the E site (Bugaeva et al, 2009). The dynamics of the other modules are more elusive and mainly depend on when helix H5 will unwind to be correctly translated into the mRNA path. Cryo-EM work is now underway in our laboratory to follow the fate of tmRNA–SmpB during the next steps of trans-translation, inside and outside of the ribosome.

Materials and methods

Preparation of ribosomal complexes

One of the main hurdles in preparing stalled ribosomal complexes is the overall heterogeneity found in the samples (Kaur et al, 2006; Bugaeva et al, 2009; Cheng et al, 2010; Weis et al, 2010). To overcome that problem, the ribosomal complexes were prepared in vitro by using ribosomes, SmpB, EF-G, and Ala-tmRNA from Thermus thermophilus; these molecules are more suitable for structural studies than the analogous components from Escherichia coli. All complexes were prepared in the absence of EF-Tu·GTP that traps some ribosomes in a pre-accommodated state, thus increasing sample heterogeneity (Weis et al, 2010). Indeed, the trans-transfer reaction can proceed without EF-Tu·GTP, albeit at a much lower efficiency (Shimizu and Ueda, 2002; Hallier et al, 2004). Salt-washed T. thermophilus 70S ribosomes free of S1 were prepared as previously described (Clemons et al, 2001). His-tagged AlaRS from E. coli and T. thermophilus EF-G and His-tagged SmpB were overexpressed in E. coli using the T7 expression system, then isolated by Ni2+ precharged HisTrap chelating column, according to the manufacturer's procedures (GE Healthcare, Orsay, France). The T. thermophilus tmRNA gene was directly cloned into a pUC19 plasmid, downstream of a T7 RNA polymerase promoter. Plasmids were linearized by BsmB1 digestion, and the corresponding RNAs were synthesized overnight using the T7 MEGAscript kit, according to the manufacturer's protocol (Ambion, Courtabœuf, France). Purification was performed by denaturing gel electrophoresis, followed by passive elution overnight. The tmRNA was heated at 80°C for 2 min in folding buffer (10 mM Hepes-KOH (pH 7.5), 20 mM NH4Cl, 5 mM MgCl2, 0.1 mM EDTA), then cooled for 30 min at room temperature prior to aminoacylation. Alanylation was performed by incubation at 37°C for 30 min in 50 mM Hepes-KOH (pH 7.5), 60 mM NH4Cl, 7 mM MgCl2, 0.1 mM EDTA, 2 mM ATP, 30 μM alanine, 2 μM SmpB, 1 μM tmRNA, and 1 μM AlaRS. The sample was four-fold concentrated on a 5-kDa cutoff Microcon centrifugal unit (Millipore). Stalled ribosomes were obtained by incubating 1 μM 70S ribosomes with a short mRNA containing a Shine–Dalgarno sequence and a P-site codon: (i) 2 μM of an mRNA consisting of the GGCAAGGAGGUAAAAAUG sequence (bold: SD sequence; underlined: P-site codon) and 2 μM E. coli deacylated tRNAfMet for the accommodated complex or (ii) 2 μM of an mRNA sequence consisting of the AGGAGGUGAGGUUUU sequence (bold: SD sequence; underlined: P-site codon), and 2 μM E. coli acylated Phe-tRNAPhe for the translocated complex, for 30 min at 37°C in 5 mM Hepes-KOH (pH 7.5), 10 mM NH4Cl, 10 mM MgOAc, 50 mM KCl, 0.1 mM EDTA, and 6 mM β-mercaptoethanol. The final complexes were obtained by mixing 1 μM of stalled ribosomes with 2 μM SmpB for 5 min at 37°C, before adding the concentrated solution of Ala–tmRNA–SmpB complex (4:1 tmRNA:stalled ribosome) in the absence (accommodated complex) or in the presence (translocated complex) of 10 μM EF-G and 2 mM GTP for 15 min at 42°C.

Cryo-EM and image processing

The accommodated and translocated state complexes were diluted to a final concentration of 80 nM. Suspensions were applied to Quantifoil holey-carbon grids (Quantifoil, Jena, Germany) previously glow discharged. Excess solution was blotted, and the grids were flash frozen in liquid ethane, using the Vitrobot automat (FEI). This process resulted in ribosomal complexes embedded in a thin film of vitrified ice. Electron micrographs were recorded under low-dose conditions at liquid nitrogen temperature with a JEOL JEM-2200FS operated at 200 kV. Images were collected at × 45 700 magnification with a defocus range of 0.9–2.2 μm. In order to obtain the maximal number of complex ribosomal particles, good micrographs were digitized on an Nikon Coolscan 9000 ED with a step size of 7.5 μm. Approximately 138 000 and 270 000 particles for the accommodated and translocated complexes, respectively, were selected semi-automatically from raw micrographs using the e2boxer subroutine of EMAN2 (Tang et al, 2007) and were coarsened by a factor of 2, resulting in a pixel size corresponding to 3.28 Å. Contrast-transfer function parameters were determined using ctffind3 (Mindell and Grigorieff, 2003) and corrected by phase flipping. Images were processed using the IMAGIC-V software package (van Heel et al, 1996). The final ‘accommodated' and ‘translocated' maps were computed from ∼49 000 and ∼70 000 particles, respectively, after eliminating particles with heterogeneous composition and conformation by a supervised classification method (see Supplementary data). The final resolutions were 13 and 13.5 Å, based on the Fourier shell correlation 0.5 bit criterion (Supplementary Figure S3) (van Heel et al, 2000), and the maps were filtered accordingly. The visualization and the fitting of the cryo-EM densities with the atomic models were performed using the programmes Chimera (Pettersen et al, 2004) and VMD (Humphrey et al, 1996).

Atomic model of the accommodated tmRNA–SmpB complex

Using our earlier model (Kaur et al, 2006) as our initial starting point, we separated the tmRNA–SmpB into small fragments corresponding to the PK and helices, which we subsequently fitted into the density corresponding to the tmRNA–SmpB complex. Owing to its characteristic shape, we started with the TLD–SmpB complex, using the 2CZJ crystal structure (Bessho et al, 2007) for coordinates. This complex was docked as a rigid body into the cryo-EM-density map using CHIMERA (Pettersen et al, 2004). We then simultaneously placed helices H2b and H2c into the dense region following the TLD. We manually positioned PK1 and H5 helix on both sides of the poorly defined domain corresponding to the upstream sequence tag. In order to allow the correct fitting of PK2 and PK3 into the density map, we split them into their respective helical regions. As for helices H2b, H2c, H5, H6a, H6b, H7, H8, and H9, PK2 and PK3 were manually fitted in the denser region of the cryo-EM map. We docked PK4 as a rigid body into its corresponding density map. We then placed the H2d helix in the remaining density between PK1, PK4, and H2c, carefully positioning this last structural element to connect it to the neighbouring structural fragments. To finish, all the single strands (including the long one connecting PK1 to H5) were manually positioned in the density map to connect the different structural domains.

In order to adjust the position of flexible elements, the whole structure was minimized in vacuo for 35 000 steps using NAMD (Phillips et al, 2005) and the CHARMM27 force field (Mackerell et al, 1998, 2004; Foloppe and Mackerell, 2000; MacKerell and Banavali, 2000), while maintaining the protein and the helical region of the RNA. The model was then optimized by molecular dynamic flexible fitting (MDFF) (Trabuco et al, 2008, 2009), a method consisting of the application of forces proportional to the gradient of the EM-density map in molecular dynamic simulations to drive the atomic structure into the high-density regions. The MDFF simulations were performed in vacuo using NAMD (Phillips et al, 2005) and the CHARMM27 force field (Mackerell et al, 1998, 2004; Foloppe and Mackerell, 2000; MacKerell and Banavali, 2000). A multiple time-step integration scheme was used to calculate bonded interactions every femtosecond (fs), and nonbonded interactions every 2 fs. A cutoff distance of 10 Å was used for the nonbonded interactions. A dielectric constant of 80 was applied to adjust electrostatic interactions, and the temperature was maintained at 300 K using a Langevin thermostat with a damping coefficient of 5 ps−1. To avoid the ‘overfitting' artefact (Tama et al, 2004), supplementary restraints had to be applied to both the protein and RNA (Trabuco et al, 2008, 2009). To preserve the secondary structure of the protein, the ψ and ϕ dihedral angles of the amino-acid residues in helices or β-sheets were harmonically restrained using a force constant of 200 kcal mol−1 rad−2. Identically, the helical part of the RNA was preserved using two 200 kcal mol−1 Å−2 restraints to maintain the distance between paired bases, and a set of 200 kcal mol−1 rad−2 restraints to maintain the dihedral angles. The dihedral angle of the last four bases of the tmRNA were also restrained to maintain this fragment in a conformation similar to that of the corresponding base in the tRNA, using the 2J00 (Selmer et al, 2006) crystal structures as our reference. We decided to use a three-step protocol to optimize our model. First, we performed an additional 65 000 steps of minimization on the tmRNA–SmpB complex, using a 0.3-kcal mol–1 scaling factor and a threshold of 0.01 to adjust the strength of the influence of the cryo-EM map on the molecular system. We then slowly heated our system to 300 K over the course of 20 000 simulation steps. In the last step, we performed 1 ns of molecular dynamic simulation, using the same scaling factor and force constant as in the previous MDFF steps. The convergence of the MDFF simulations was assessed during the course of the simulation by monitoring the cross-correlation coefficient between the atomic model and the original cryo-EM-density map. The convergence of the MDFF was usually found after 600 ps (data not shown).

From the accommodated model to the translocated tmRNA–SmpB

We started from the model of the accommodated state in order to create an atomic model of the translocated tmRNA. By comparing the two cryo-EM maps, we noticed that the density region corresponding to the H5 helix and PK2/PK3 did not differ between the two states. These three structural domains did not move significantly during translocation; therefore, we reused the coordinates of the accommodated model for the translocated tmRNA–SmpB. For the remaining sections, we used rigid-body docking to position the TLD–SmpB complex, the H2b, H2c, and H2d helices, and PK1 and PK4 in their respective densities. We then connected the various regions by placing the appropriate single strands in the density map. We took special care of the resume codon that we positioned into the decoding site. We also added a fragment of the 30S subunit, including the h30–34 helices, in order to properly position the long single strand connecting PK1 to H5, leaving enough space for an incoming tRNA. The coordinates of this fragment were once again taken from crystal structure 2J00 (Selmer et al, 2006). NMR studies have shown that the C-terminal 20–30 residues are disordered in solution (Dong et al, 2002; Someya et al, 2003), and the crystal structure of these residues has not yet been shown (Gutmann et al, 2003; Bessho et al, 2007). Nevertheless, as previously predicted, we modelized the C-terminal tail of the SmpB in an α helix to account for the extra density found in the P site of the 30S (Jacob et al, 2005). We decided not to restrain this region during the optimization process, in order to let it relax and adjust freely.

The same minimization and MDFF protocols were used to optimize this second model. The only difference resides in the threshold of 0.03, which was used this time to adjust the strength of the influence of the cryo-EM map on the molecular system. Both the resume codon and the h30–34 helices were held fixed during the minimization and the MDFF simulations.

Supplementary Material

Supplementary Movie S1
Download video file (5MB, avi)
Supplementary Data
emboj2010252s2.pdf (3.6MB, pdf)
Review Process File
emboj2010252s3.pdf (241.3KB, pdf)

Acknowledgments

We thank V Ramakrishnan (MRC-LMB, Cambridge, UK) for providing the T. thermophilus ribosomes. We also thank A Molza, J Lai Kee Him, and A Goulet for their assistance. This work was supported by Grants ANR-08JCJC-0027-01 and ANR-09-MIE. FW was supported by a PhD fellowship from the Ministère de l'Enseignement Supérieur et de la Recherche. Author contributions: RG, DT, PB and BF designed research; FW and RG performed biochemical work; FW, PB, and J-PR performed cryo-EM; FW and PB performed image processing; EG performed the atomic models; all authors contributed to the interpretation of the data, and RG, FW, and EG wrote the paper with important input from others. Coordinates of the cryo-EM maps and atomic models of the accommodated and translocated states have been deposited in the EMDB (accession codes EMD-5188 and EMD-5189, respectively) and in the PDB (accession codes 3IYQ and 3IYR, respectively).

Footnotes

The authors declare that they have no conflict of interest.

References

  1. Andersen ES, Rosenblad MA, Larsen N, Westergaard JC, Burks J, Wower IK, Wower J, Gorodkin J, Samuelsson T, Zwieb C (2006) The tmRDB and SRPDB resources. Nucleic Acids Res 34: D163–D168 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bessho Y, Shibata R, Sekine S, Murayama K, Higashijima K, Hori-Takemoto C, Shirouzu M, Kuramitsu S, Yokoyama S (2007) Structural basis for functional mimicry of long-variable-arm tRNA by transfer-messenger RNA. Proc Natl Acad Sci USA 104: 8293–8298 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bugaeva EY, Shpanchenko OV, Felden B, Isaksson LA, Dontsova OA (2008) One SmpB molecule accompanies tmRNA during its passage through the ribosomes. FEBS Lett 582: 1532–1536 [DOI] [PubMed] [Google Scholar]
  4. Bugaeva EY, Surkov S, Golovin AV, Ofverstedt LG, Skoglund U, Isaksson LA, Bogdanov AA, Shpanchenko OV, Dontsova OA (2009) Structural features of the tmRNA-ribosome interaction. RNA 15: 2312–2320 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Cheng K, Ivanova N, Scheres SH, Pavlov MY, Carazo JM, Hebert H, Ehrenberg M, Lindahl M (2010) tmRNA.SmpB complex mimics native aminoacyl-tRNAs in the A site of stalled ribosomes. J Struct Biol 169: 342–348 [DOI] [PubMed] [Google Scholar]
  6. Clemons WM Jr, Brodersen DE, McCutcheon JP, May JL, Carter AP, Morgan-Warren RJ, Wimberly BT, Ramakrishnan V (2001) Crystal structure of the 30 S ribosomal subunit from Thermus thermophilus: purification, crystallization and structure determination. J Mol Biol 310: 827–843 [DOI] [PubMed] [Google Scholar]
  7. Crandall J, Rodriguez-Lopez M, Pfeiffer M, Mortensen B, Buskirk A (2010) rRNA mutations that inhibit transfer-messenger RNA activity on stalled ribosomes. J Bacteriol 192: 553–559 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Dong G, Nowakowski J, Hoffman DW (2002) Structure of small protein B: the protein component of the tmRNA-SmpB system for ribosome rescue. EMBO J 21: 1845–1854 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Felden B, Himeno H, Muto A, McCutcheon JP, Atkins JF, Gesteland RF (1997) Probing the structure of the Escherichia coli 10Sa RNA (tmRNA). RNA 3: 89–103 [PMC free article] [PubMed] [Google Scholar]
  10. Foloppe N, Mackerell AD Jr (2000) All-atom empirical force field for nucleic acids: 1) parameter optimization based on small molecule and condensed phase macromolecular target data. J Comput Chem 21: 86–104 [Google Scholar]
  11. Frank J, Agrawal RK (2000) A ratchet-like inter-subunit reorganization of the ribosome during translocation. Nature 406: 318–322 [DOI] [PubMed] [Google Scholar]
  12. Gillet R, Kaur S, Li W, Hallier M, Felden B, Frank J (2007) Scaffolding as an organizing principle in trans-translation. The roles of small protein B and ribosomal protein S1. J Biol Chem 282: 6356–6363 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Gutmann S, Haebel PW, Metzinger L, Sutter M, Felden B, Ban N (2003) Crystal structure of the transfer-RNA domain of transfer-messenger RNA in complex with SmpB. Nature 424: 699–703 [DOI] [PubMed] [Google Scholar]
  14. Hallier M, Desreac J, Felden B (2006) Small protein B interacts with the large and the small subunits of a stalled ribosome during trans-translation. Nucleic Acids Res 34: 1935–1943 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hallier M, Ivanova N, Rametti A, Pavlov M, Ehrenberg M, Felden B (2004) Pre-binding of small protein B to a stalled ribosome triggers trans-translation. J Biol Chem 279: 25978–25985 [DOI] [PubMed] [Google Scholar]
  16. Humphrey W, Dalke A, Schulten K (1996) VMD: visual molecular dynamics. J Mol Graph 14: 33–38, 27–38 [DOI] [PubMed] [Google Scholar]
  17. Ivanova N, Pavlov MY, Ehrenberg M (2005) tmRNA-induced release of messenger RNA from stalled ribosomes. J Mol Biol 350: 897–905 [DOI] [PubMed] [Google Scholar]
  18. Jacob Y, Sharkady SM, Bhardwaj K, Sanda A, Williams KP (2005) Function of the SmpB tail in transfer-messenger RNA translation revealed by a nucleus-encoded form. J Biol Chem 280: 5503–5509 [DOI] [PubMed] [Google Scholar]
  19. Karzai AW, Susskind MM, Sauer RT (1999) SmpB, a unique RNA-binding protein essential for the peptide-tagging activity of SsrA (tmRNA). EMBO J 18: 3793–3799 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kaur S, Gillet R, Li W, Gursky R, Frank J (2006) Cryo-EM visualization of transfer messenger RNA with two SmpBs in a stalled ribosome. Proc Natl Acad Sci USA 103: 16484–16489 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Komoda T, Sato NS, Phelps SS, Namba N, Joseph S, Suzuki T (2006) The A-site finger in 23 S rRNA acts as a functional attenuator for translocation. J Biol Chem 281: 32303–32309 [DOI] [PubMed] [Google Scholar]
  22. Konno T, Kurita D, Takada K, Muto A, Himeno H (2007) A functional interaction of SmpB with tmRNA for determination of the resuming point of trans-translation. RNA 13: 1723–1731 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kurita D, Sasaki R, Muto A, Himeno H (2007) Interaction of SmpB with ribosome from directed hydroxyl radical probing. Nucleic Acids Res 35: 7248–7255 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Lee S, Ishii M, Tadaki T, Muto A, Himeno H (2001) Determinants on tmRNA for initiating efficient and precise trans-translation: some mutations upstream of the tag-encoding sequence of Escherichia coli tmRNA shift the initiation point of trans-translation in vitro. RNA 7: 999–1012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. MacKerell AD Jr, Banavali N (2000) All-atom empirical force field for nucleic acids: 2) application to molecular dynamics simulations of DNA and RNA in solution. J Comput Chem 21: 105–120 [Google Scholar]
  26. Mackerell AD Jr, Bashford D, Bellott M, Dunbrack RL Jr, Evanseck J, Field MJ, Fischer S, Gao J, Guo H, Ha S, Joseph D, Kuchnir L, Kuczera K, Lau FTK, Mattos C, Michnick S, Ngo T, Nguyen DT, Prodhom B, Reiher WE III et al. (1998) All-atom empirical potential for molecular modeling and dynamics studies of protein. J Phys Chem B 102: 3586–3615 [DOI] [PubMed] [Google Scholar]
  27. Mackerell AD Jr, Feig M, Brooks CL III (2004) Extending the treatment of backbone energetics in protein force fields: limitations of gas-phase quantum mechanics in reproducing protein conformational distributions in molecular dynamics simulations. J Comput Chem 25: 1400–1415 [DOI] [PubMed] [Google Scholar]
  28. Metzinger L, Hallier M, Felden B (2008) The highest affinity binding site of small protein B on transfer messenger RNA is outside the tRNA domain. RNA 14: 1761–1772 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Miller MR, Healey DW, Robison SG, Dewey JD, Buskirk AR (2008) The role of upstream sequences in selecting the reading frame on tmRNA. BMC Biol 6: 29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Mindell JA, Grigorieff N (2003) Accurate determination of local defocus and specimen tilt in electron microscopy. J Struct Biol 142: 334–347 [DOI] [PubMed] [Google Scholar]
  31. Moore SD, Sauer RT (2007) The tmRNA system for translational surveillance and ribosome rescue. Annu Rev Biochem 76: 101–124 [DOI] [PubMed] [Google Scholar]
  32. Nameki N, Tadaki T, Himeno H, Muto A (2000) Three of four pseudoknots in tmRNA are interchangeable and are substitutable with single-stranded RNAs. FEBS Lett 470: 345–349 [DOI] [PubMed] [Google Scholar]
  33. Nonin-Lecomte S, Germain-Amiot N, Gillet R, Hallier M, Ponchon L, Dardel F, Felden B (2009) Ribosome hijacking: a role for small protein B during trans-translation. EMBO Rep 10: 160–165 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, Ferrin TE (2004) UCSF Chimera—a visualization system for exploratory research and analysis. J Comput Chem 25: 1605–1612 [DOI] [PubMed] [Google Scholar]
  35. Phillips JC, Braun R, Wang W, Gumbart J, Tajkhorshid E, Villa E, Chipot C, Skeel RD, Kale L, Schulten K (2005) Scalable molecular dynamics with NAMD. J Comput Chem 26: 1781–1802 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Reblova K, Razga F, Li W, Gao H, Frank J, Sponer J (2010) Dynamics of the base of ribosomal A-site finger revealed by molecular dynamics simulations and Cryo-EM. Nucleic Acids Res 38: 1325–1340 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Richards J, Mehta P, Karzai AW (2006) RNase R degrades non-stop mRNAs selectively in an SmpB-tmRNA-dependent manner. Mol Microbiol 62: 1700–1712 [DOI] [PubMed] [Google Scholar]
  38. Schmeing TM, Ramakrishnan V (2009) What recent ribosome structures have revealed about the mechanism of translation. Nature 461: 1234–1242 [DOI] [PubMed] [Google Scholar]
  39. Selmer M, Dunham CM, Murphy FVt, Weixlbaumer A, Petry S, Kelley AC, Weir JR, Ramakrishnan V (2006) Structure of the 70S ribosome complexed with mRNA and tRNA. Science 313: 1935–1942 [DOI] [PubMed] [Google Scholar]
  40. Shimizu Y, Ueda T (2002) The role of SmpB protein in trans-translation. FEBS Lett 514: 74–77 [DOI] [PubMed] [Google Scholar]
  41. Someya T, Nameki N, Hosoi H, Suzuki S, Hatanaka H, Fujii M, Terada T, Shirouzu M, Inoue Y, Shibata T, Kuramitsu S, Yokoyama S, Kawai G (2003) Solution structure of a tmRNA-binding protein, SmpB, from Thermus thermophilus. FEBS Lett 535: 94–100 [DOI] [PubMed] [Google Scholar]
  42. Tama F, Miyashita O, Brooks CL III (2004) Normal mode based flexible fitting of high-resolution structure into low-resolution experimental data from cryo-EM. J Struct Biol 147: 315–326 [DOI] [PubMed] [Google Scholar]
  43. Tang G, Peng L, Baldwin PR, Mann DS, Jiang W, Rees I, Ludtke SJ (2007) EMAN2: an extensible image processing suite for electron microscopy. J Struct Biol 157: 38–46 [DOI] [PubMed] [Google Scholar]
  44. Tanner DR, Dewey JD, Miller MR, Buskirk AR (2006) Genetic analysis of the structure and function of transfer messenger RNA pseudoknot 1. J Biol Chem 281: 10561–10566 [DOI] [PubMed] [Google Scholar]
  45. Trabuco LG, Villa E, Mitra K, Frank J, Schulten K (2008) Flexible fitting of atomic structures into electron microscopy maps using molecular dynamics. Structure 16: 673–683 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Trabuco LG, Villa E, Schreiner E, Harrison CB, Schulten K (2009) Molecular dynamics flexible fitting: a practical guide to combine cryo-electron microscopy and X-ray crystallography. Methods 49: 174–180 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Valle M, Gillet R, Kaur S, Henne A, Ramakrishnan V, Frank J (2003a) Visualizing tmRNA entry into a stalled ribosome. Science 300: 127–130 [DOI] [PubMed] [Google Scholar]
  48. Valle M, Zavialov A, Sengupta J, Rawat U, Ehrenberg M, Frank J (2003b) Locking and unlocking of ribosomal motions. Cell 114: 123–134 [DOI] [PubMed] [Google Scholar]
  49. van Heel M, Gowen B, Matadeen R, Orlova EV, Finn R, Pape T, Cohen D, Stark H, Schmidt R, Schatz M, Patwardhan A (2000) Single-particle electron cryo-microscopy: towards atomic resolution. Q Rev Biophys 33: 307–369 [DOI] [PubMed] [Google Scholar]
  50. van Heel M, Harauz G, Orlova EV, Schmidt R, Schatz M (1996) A new generation of the IMAGIC image processing system. J Struct Biol 116: 17–24 [DOI] [PubMed] [Google Scholar]
  51. Watts T, Cazier D, Healey D, Buskirk A (2009) SmpB contributes to reading frame selection in the translation of transfer-messenger RNA. J Mol Biol 391: 275–281 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Weis F, Bron P, Rolland JP, Thomas D, Felden B, Gillet R (2010) Accommodation of tmRNA-SmpB into stalled ribosomes: a cryo-EM study. RNA 16: 299–306 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Williams KP, Martindale KA, Bartel DP (1999) Resuming translation on tmRNA: a unique mode of determining a reading frame. EMBO J 18: 5423–5433 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Wower IK, Zwieb C, Wower J (2005) Transfer-messenger RNA unfolds as it transits the ribosome. RNA 11: 668–673 [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Yusupova G, Jenner L, Rees B, Moras D, Yusupov M (2006) Structural basis for messenger RNA movement on the ribosome. Nature 444: 391–394 [DOI] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

Supplementary Movie S1
Download video file (5MB, avi)
Supplementary Data
emboj2010252s2.pdf (3.6MB, pdf)
Review Process File
emboj2010252s3.pdf (241.3KB, pdf)

Articles from The EMBO Journal are provided here courtesy of Nature Publishing Group

RESOURCES