Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Oct 6.
Published in final edited form as: Structure. 2015 Aug 20;23(10):1858–1865. doi: 10.1016/j.str.2015.07.014

Structure of Ribosomal Silencing Factor Bound to Mycobacterium tuberculosis Ribosome

Xiaojun Li 1,3, Qingan Sun 1,3, Cai Jiang 1,3, Kailu Yang 1,3, Li-Wei Hung 2, Junjie Zhang 1,*, James C Sacchettini 1,*
PMCID: PMC4718548  NIHMSID: NIHMS713748  PMID: 26299947

SUMMARY

The ribosomal silencing factor RsfS slows cell growth by inhibiting protein synthesis during periods of diminished nutrient availability. The crystal structure of Mycobacterium tuberculosis (Mtb) RsfS, together with the cryo-electron microscopy (EM) structure of the large subunit 50S of Mtb ribosome, reveals how inhibition of protein synthesis by RsfS occurs. RsfS binds to the 50S at L14, which, when occupied, blocks the association of the small subunit 30S. Although Mtb RsfS is a dimer in solution, only a single subunit binds to 50S. The overlap between the dimer interface and the L14 binding interface confirms that the RsfS dimer must first dissociate to a monomer in order to bind to L14. RsfS interacts primarily through electrostatic and hydrogen bonding to L14. The EM structure shows extended rRNA density that it is not found in the Escherichia coli ribosome, the most striking of these being the extended RNA helix of H54a.

Graphical Abstract

graphic file with name nihms713748u1.jpg

INTRODUCTION

A wealth of structural and biochemical studies on the ribosome published over the last two decades have led to a deeper understanding of the mechanism of translation (Ban et al., 2000; Gao et al., 2009; Harms et al., 2001; Korostelev et al., 2006; Moore, 2012; Schuwirth et al., 2005; Selmer et al., 2006; Shasmal and Sengupta, 2012; Yusupov et al., 2001). Ribosome structures have provided molecular details for the machinery of translation as well as information on how drugs bind to the ribosome and interfere with protein synthesis (Wilson, 2014). In bacteria, the functional ribosome consists of two subunits, a 50S and a 30S. The 50S subunit is composed of the 23S rRNA, 5S rRNA, and over 30 proteins (35 in Mycobacterium tuberculosis [Mtb]) while the 30S subunit contains a 16S rRNA and over 20 proteins (22 in Mtb). Given their fundamental importance to cell viability, it is not surprising that more than half of all clinically prescribed antibiotics target the ribosome (Wilson, 2014). However, the detailed mechanisms underlying the regulation of translation are, in some cases, poorly understood. This is especially true for the pathogenic bacterium Mtb, and it is most likely due to its complex growth characteristics, including the ability to enter a non-replicating state in which protein synthesis is presumed to be greatly diminished.

Tuberculosis (TB) remains a major global health problem and is linked to high morbidity and mortality worldwide (Anastasia et al., 2014; Harries and Dye, 2006; Russell, 2007). The World Health Organization estimates that one-third of the global population is infected with latent Mtb while more than eight million people develop active TB. Roughly one million people die from the disease annually (http://www.who.int/mediacentre/factsheets/fs104/en/) and co-infection with HIV markedly increases mortality associated with TB (Kwan and Ernst, 2011).

The success of Mtb as a pathogen is largely attributed to its ability to persist in host tissues (Flynn and Chan, 2001; Wayne and Sohaskey, 2001). In chronic or persistent infections, Mtb shows reduced growth, which is thought to be related to the host’s immune system or to the lack of susceptibility to antibiotics (Harries and Dye, 2006; Wayne and Sohaskey, 2001). Persistent Mtb has been shown to exist in a non-replicating state equivalent to dormancy in which many metabolic processes, including protein synthesis, are assumed to be greatly reduced in order to conserve cellular resources (Kumar et al., 2012; Trauner et al., 2012). As a consequence, drug treatment must be dramatically extended (Gomez and McKinney, 2004). Mtb’s emergence from dormancy is not understood but it can lead to the reactivation of the disease.

A recently characterized protein from Escherichia coli, named ribosomal silencing factor during starvation or stationary phase (RsfS), has been shown to slow down or block translation entirely (Hauser et al., 2012). It appears to play an important role in the maintenance of sustainable energy levels during nutritional shortages. Disruption of the RsfS gene slows the adaptation from rich to poor media and impairs the viability of the cells during the stationary phase. Previous mutagenesis studies in E. coli suggest that RsfS binds to the 50S large ribosomal subunit (Jiang et al., 2007), preventing the normal association of the 50S and 30S into a functional 70S complex (Hauser et al., 2012). In order to improve our understanding of the molecular details of the inhibition of translation by RsfS, biochemical and structural studies have been completed on the 50S ribosome from Mtb. The crystal structure of Mtb RsfS and the cryo-EM structure of the Mtb 50S, with and without Mtb RsfS bound, provide a detailed understanding of the molecular basis of RsfS function.

RESULTS

Mtb RsfS Inhibits Translation by Preventing 70S Ribosome Association

We have developed a translation assay that uses Mtb ribosome to translate GFP mRNA in an Mtb-derived cell-free extract. When purified recombinant Mtb RsfS was added to our cell-free assay, we observed a significant reduction in translation of GFP. Specifically, RsfS (0.28 μM) was able to block GFP synthesis by the ribosome (0.14 μM) by 84%, a level comparable with 0.7 μM streptomycin and 0.14 μM chloramphenicol (Figure 1A). Interestingly, when Mtb 50S and 30S ribosome were mixed with a 15-fold molar excess of RsfS (2.1 μM final concentration) and applied to a sucrose gradient (Figure 1B), only 10% of the large subunit was in the 70S form. However, in the absence of Mtb RsfS, more than 70% of 50S subunits were associated into the 70S ribosome. When the same concentration of RsfS was incubated with preformed 70S ribosome for 2 hr at room temperature, no significant dissociation of 70S was observed. These results indicated that, while RsfS was able to block the association of 50S and 30S subunits, it was unable to significantly dissociate the 70S ribosome like E. coli RsfS (Hauser et al., 2012).

Figure 1. RsfS Prevents Ribosomal Subunit Association Resulting in Inhibition of Protein Translation.

Figure 1

Open in a new tab

(A) GFP translation by Mtb ribosome. The assay was performed with Mtb ribosome 70S at 0.14 μM and Mtb RsfS at 0.14 μM and 0.28 μM. Chloramphenicol at 0.14 μM and streptomycin at 0.7 μM provided approximately the same inhibition. Data are represented as means ± SEM.

(B) Sucrose gradient separation of 70S ribosome association.

(C) The results of the nickel pull-down assay of Mtb ribosome using His6-tagged RsfS. RsfS was mixed with purified 70S, 50S, and 30S, respectively, and incubated for 1 hr before the nickel beads were added. The beads were washed and the eluted fractions were loaded on SDS polyacrylamide gel. Only ribosome 50S can be pulled down by RsfS.

Mtb RsfS shows selectivity for the 50S subunit, as demonstrated using a pull-down assay with His-tagged RsfS. In this assay, 6.7 μM (0.1 mg/ml) RsfS was mixed with purified 1.2 μM 70S (2.9 mg/ml), 50S (1.9 mg/ml) and 30S (1.0 mg/ml) fractions from a sucrose gradient. After incubation for 1 hr at 4°C, nickel affinity beads were used to pull-down RsfS. Only the 50S subunit associated with RsfS (Figure 1C).

Crystal Structure of Mtb RsfS

Crystals of full-length Mtb RsfS could not be obtained, therefore we resorted to screening single-point mutants of RsfS in order to produce diffraction-quality crystals. Nine distinct single-point mutants were made to hydrophobic amino acids that were predicted to be on the surface. Only one mutant protein of Y102A yielded diffraction-quality crystals. Mtb RsfS Y102A crystallized in the P1 space group with four Mtb RsfS molecules (referred to as A, B, C, and D) in the crystal’s asymmetric unit (ASU) (Figure S1A). The structure was solved by molecular replacement using the Bacillus halodurans ortholog (PDB: 2O5A) as the search model. The structure was refined with diffraction data to 2.1-Å resolution (Table 1). The R factor of the final model was 21.0% (Rfree = 26.0%) with good stereochemistry and 97.0% of the amino acids were in the preferred regions of the Ramachandran plot. The C-terminal 8–13 amino acids were disordered and showed either weak or no corresponding electron density.

Table 1.

Data Collection and Refinement Statistics of Mtb RsfS Crystal Structure

Mtb RsfS
Data Collection
Space group P1
Cell dimensions
a, b, c (Å) 50.98, 50.98, 64.55
α, β, γ (°) 110.27, 96.17, 110.59
Resolution (Å) 46.06–2.1 (2.14–2.1)a
Rsym or Rmerge 0.086 (0.263)
I/σI 6.3 (3.1)
Completeness (%) 93.3 (82.7)
Redundancy 1.7 (1.6)
Refinement
Resolution (Å) 46.06–2.1
No. of reflections 49,832
Rwork/Rfree 0.21/0.26
No. of atoms 3,922
Protein 3,522
Ligand/ion 22
Water 378
B factors 24.8
Protein 24.1
Ligand/ion 19.3
Water 31.6
Rmsd
Bond lengths (Å) 0.005
Bond angles (°) 0.89
Open in a new tab
a

Values in parentheses are for the highest-resolution shell.

RsfS adopts the α1-β1-β2-α2-β3-β4-β5-α3 fold, where five β strands form one β sheet. The first two long α helices (α1 and α2) reside on one side of the β sheet, while the short α3 helix with a long C-terminal tail resides on the edge of β sheet close to β1 (Figure 2A). This is a well-conserved domain referred to as DUF143 (Fung et al., 2013).

Figure 2. Mtb RsfS Structure.

Figure 2

Open in a new tab

(A) Mtb RsfS dimer.

(B) RsfS dimer rotated 90° from the view in (A). The two RsfS molecules are related by ~180° rotation around the pseudo-dyad axis in the center of dimer. The residues involved in hydrogen bonding are highlighted as sticks and the hydrogen bonds are shown as black lines (see also Figures S1A–S1C and S1F).

(C) Size-exclusion chromatography (Superdex 75) of Mtb RsfS showed a solution dimer in agreement with the crystal structure. The standard curve is shown as a dashed line (see also Figures S1D and S1E).

Alignment of the sequence of Mtb RsfS with other bacterial orthologs indicated relatively good sequence similarity for the portion of RsfS that is well ordered in the crystal (Figure S1B). Superimposition of the three ortholog structures (PDB: 2O5A for B. halodurans RsfS, PDB: 3UPS for Zymomonas mobilis RsfS, and PDB: 2ID1 for Chromobacterium violaceum RsfS) onto Mtb RsfS showed that the overall structure is well conserved from the N terminus to the end of β5 (Figure S1C), corresponding to His95 in Mtb RsfS. The mutation Y102A did not appear to change the overall structure of Mtb RsfS when compared with the crystal structures of other orthologs. The Cα root-mean-square difference (rmsd) values (for residues 6–95 in Mtb RsfS) are 1.6 Å, 1.9 Å, and 2.2 Å for PDB: 2ID1, 3UPS, and 2O5A, respectively. The lack of similarity in the structure of the C termini implies that they are not the critical determinant for inhibition of the ribosome. Although E. coli RsfS has an overall sequence identity of 25% with Mtb RsfS, the C-terminal region is much shorter (15 residues) than Mtb RsfS (32 residues) and shares only 16% sequence identity in this region compared with Mtb RsfS. However, full-length recombinant E. coli RsfS showed approximately the same level of inhibition as Mtb RsfS in the Mtb ribosome cell-free translation assay (80% inhibition of E. coli RsfS and 84% inhibition of Mtb RsfS on Mtb translation assay). This supported the notion that the C-terminal extensions of these orthologs were not critical components in the binding to ribosome.

Mtb RsfS Forms a Dimer in Both Crystal and Solution

The structures of the four copies of the Mtb RsfS in the crystal’s ASU are very similar (Thr2 to Pro112), except for the last six residues of C termini. The four subunits are packed into two nearly identical dimers (dimers AB and CD) each with quasi two-fold rotational symmetry for the subunits. The buried surface areas for the two dimers are 1,822 Å2 and 1,706 Å2, respectively (analyzed by PISA; Krissinel and Henrick, 2007), and this represents around 15% of the total surface of each dimer. A high percentage of buried surface area and the conservation of the packing of both dimers in the asymmetric unit indicated that the dimer observed in the crystal was equivalent to what is observed in solution (Krissinel and Henrick, 2007; Nooren and Thornton, 2003). RsfS also eluted from size-exclusion chromatography (Figure 2C) as a dimer based on the calculated molecular weight of 29 kDa (an RsfS subunit is 15 kDa) (Figure S1D). The RsfS dimer is also consistent with native gel electrophoresis and glutaraldehyde crosslinking experiments (Figure S1E).

The dimer interface is predominately formed through interactions between the side chains of amino acids that are in β3 (Arg72-Gly75) and β4 (Trp81-Asp85) from opposing subunits. In addition, loop 2 (Val31-Asp39) and loop 4 (Ala76-Arg80) from each subunit contribute residues that are located at the dimer interface (Figure 2B). Indeed, the two subunits appear to be stitched together through an extensive network of 22 intermolecular hydrogen bonds and electrostatic interactions (Figure S1F). The two dimers are very well conserved with only small conformational differences in loop 4 and the adjacent N terminus of helix α2. The rmsd between the A and C chain for the first 111 Cα is only 0.17 Å; that between B and D is 0.37 Å. In contrast, the rmsd between chains A and B is 0.88 Å, and 0.90 Å between chains C and D. In the crystal lattice, contacts between subunits A and C of each dimer are very close to these loops and it is likely that they influence the observed differences.

When the three orthologs deposited in the PDB were compared to the Mtb RsfS crystal structures, they all showed significant differences in their quaternary structures. Z. mobilis RsfS (PDB: 3UPS) forms the clearest dimer of the three with a buried surface area of 2,840 Å2, which represents 28% of the total surface of the dimer. B. halodurans RsfS (PDB: 2O5A) forms an apparently less stable dimer with a buried surface area of 1,070 Å2, which represents 9% of the total surface of the dimer. The analysis of the crystal packing for C. violaceum RsfS (PDB: 2ID1) does not reveal any higher level oligomerization, suggesting it is a monomer in solution (Krissinel and Henrick, 2007). Comparison of Mtb RsfS dimer with Z. mobilis and B. halodurans RsfS dimers indicated that the intra-dimer interfaces occur at different regions for each of the three proteins. For the Z. mobilis RsfS dimer, the interactions are primarily between α helices (α2 and α3), loop 2 (connecting β1 and β2), and the two bends (connecting α2 and β3 or β4 and β5). In the B. halodurans RsfS dimer, the interactions are between β strands (β3 and β4) and loops (loop 1 connects β1 and β2; loop 4 connects β5 and α3). Although the dimers organize differently between the three proteins, the interfaces of the Z. mobilis and B. halodurans dimers are similarly dominated by hydrogen bonding and electrostatic interactions.

Cryo-Electron Microscopy Structures of Mtb 50S

We used single particle cryo-electron microscopy (EM) to visualize the purified 50S Mtb ribosome mixed with recombinant Mtb RsfS at a molar ratio of 1:15. A density map was generated from 42,138 screened particles (Figure S2A) and was calculated to be at 8.5-Å resolution using the gold standard Fourier shell correlation (Scheres and Chen, 2012) (Figure S2B). We observed weak density for RsfS, suggesting that despite the 15-fold molar excess of RsfS to ribosome, we may have had a mixture of RsfS-bound and RsfS-free 50S particles. A modified supervised classification was used to classify the 42,138 particles into the RsfS-bound (20,851 particles) and RsfS-free (21,287 particles) states, from which two density maps were reconstructed (Figure S2C). After the classification, the RsfS density was strong in the RsfS-bound state and clearly showed the secondary structures of the RsfS. The calculated final resolutions were 9.3 Å for the 50S subunit alone and 9.1 Å for the RsfS-bound 50S (Figures 3A–3D; Figure S2B). To rule out the possibility that the density of RsfS was due to the reference bias, we performed extra steps, described in the Experimental Procedures, to prove the reliability of the classification.

Figure 3. Cryo-EM Maps of Mtb Ribosome 50S in Its RsfS-free and RsfS-Bound States.

Figure 3

Open in a new tab

The cryo-EM density maps of 50S ribosome (A) without RsfS and (B) RsfS-bound Mtb 50S are shown in “crown view” with their densities colored (23S RNA, light gray; 5S RNA, dark gray; ribosomal proteins, blue; RsfS, red; handle, gold).

(C–F) The “front view” of the density map made by rotating the maps 90° along the x axis in (A) and (B). Central protuberance (CP), stalk base (SB), and L1 protuberance (L1) are labeled accordingly. The dashed square regions in (C) and (D) show maps and models of the RsfS-free 50S (E) and RsfS-bound 50S (F). The cryo-EM density maps for 23S RNA are in gray, L14 in blue, and RsfS in red.

(G) The interacting surfaces between L14 and RsfS.

(H) L14 has a net positively charged side while the RsfS is negatively charged on the interface (see also Figures S2 and S3).

At 8–9 Å, the cryo-EM density map of the core regions of the Mtb 50S ribosome agree well with the crystal structure of the E. coli 50S ribosome (Berk et al., 2006). The density was observed for the common structural motifs associated with the ribosome structures including the body, the stalk base (SB) of the L7/L12 arm near the A site (entry for the aminoacyl tRNA), the L1 stalk near the E site (exit site of the uncharged tRNA), and the central protuberance (CP) found between the L1 and L7/L12 stalks (Figure 3). It is known that the two peripheral stalks (L1 and L7/L12) of the 50S subunit are intrinsically dynamic. In fact, the L1 stalk of the Mtb structure has relatively weaker density and the density for L7/L12 was completely missing. Density maps for the ribosomal proteins L9 and L11 were missing, probably due to their loose association with the 50S subunit. Figure S2D shows the local resolution of our density map calculated from ResMap (Kucukelbir et al., 2014). Most regions of our 50S density maps have a resolution better than 9 Å.

The most distinct structural feature of the Mtb 50S ribosome is a 107 nucleotide-long RNA helical extension of H54a (golden density in Figure 3 and Figure S2E) that ends close to the L1 stalk and the E site. Density attributable to the nucleotide (nt) sequence extensions of rRNA (23S) was also observed in Mtb 50S subunit (Figure S2F). Helix 15 and helix 16a are ~40 and 20 nt long in Mtb, observed in the density close to the base of the L1 stalk. Helix 31a, which is 25 nt longer in Mtb compared with E. coli, is visible at the solvent side of the 50S close to the CP. Helix 25 shows the greatest length variability among the three phylogenetic domains (Petrov et al., 2014). It is a short stem loop in E. coli, an ~80 nt bent helix in Archaea, and is longer as one progresses to higher organisms (876 nt in humans). Mtb H25 is 42 nt long, which is 15 nt longer than its counterpart in E. coli.

A Single Subunit of RsfS Directly Binds to the L14 Protein on the Mtb 50S

Cryo-EM density corresponding to a single RsfS protein was clearly identified on the surface of the 50S, interacting directly with the L14 protein. L14 is composed of a five-stranded β barrel, a C-terminal loop region that contains two small α helices, and a β ribbon that projects from the β barrel (Davies et al., 1996). The primary structures of L14 from Mtb, E. coli, Thermus thermophilus, and Haloarcula marismortui are highly conserved. Mtb L14 has between 66% and 78% sequence identities with these orthologs and, as expected, all three of the L14 crystal structures fit nicely into the density map. Our homology model of Mtb L14 was built using SWISS-MODEL (Schwede et al., 2003) based on the E. coli L14 structure and the resulting model was fit into the Mtb density map.

Given that RsfS is a dimer in solution, we were surprised to find that the cryo-EM density for the bound RsfS was only large enough to accommodate a single protein subunit (Figures 3E and 3F). In order to get the structure of the complex, the refined crystal structure of a single subunit of RsfS and the homology model of L14 were optimized to fit their cryo-EM densities by starting from random initial orientations of the RsfS relative to the L14. The two models were refined into density using the real-space refinement routine in PHENIX (Adams et al., 2010). The final refined model of the complex between L14 and RsfS had a cross-correlation score of 0.9 with its cryo-EM density map. Visual inspection showed very good agreement of the secondary structural components of the RsfS crystal structure with the EM density.

Most of the atoms of RsfS that form the dimer interface are also at the interface of RsfS with L14. This is in agreement with the observation of a single subunit of RsfS dimer bound to L14. The structure of the complex indicates that the two small C-terminal α helices (Arg104 to Leu117) of L14 are interacting with the RsfS β sheet (β1, Val26-Asp30; β2, Cys40-Gly46; β3, Arg72-Gly75; β4, Trp81-Asp85; β5, Ile89-His95), loop 2 (Val31-Asp39), and the C-terminal α3 (Phe101-Gly109) of RsfS (Figure 3G). In fact, the buried surface area between L14 and RsfS is 2,212 Å2, which represents about 20% of the total surface area of the RsfS-L14 complex (Krissinel and Henrick, 2007), slightly more than the buried surface area observed in the dimer. While it is not possible to assign hydrogen bonds at this resolution, the binding interface of the RsfS-L14 contains complementary electrostatic surface potentials as well as many potential H-bond donors and acceptors. At the interface of the two proteins, L14 has a net positive electrostatic surface while the RsfS interface has a negative potential (Figure 3H), indicating that H bonds and electrostatic forces are likely the dominant interactions between the RsfS and the L14. We have made 14 mutations to residues on RsfS that are common between the interface of the dimer and the RsfS-L14 complex. Only one mutation, E74A, provided soluble recombinant protein. The mutant was 64% less active in the inhibition for the Mtb translation assay. In addition, we found this mutant could not pull down 50S to the same degree as wild-type (Figure S3A), suggesting that there was a significant reduction in affinity. However, the mutant protein was still a dimer in solution.

We compared our cryo-EM Mtb RsfS-50S structure to a published E. coli RsfS-50S homology model (Hauser et al., 2012). The model was constructed based on alanine scanning mutagenesis. Both Mtb RsfS and the modeled E. coli ortholog interact with L14 through their C-terminal helix α3 and loop 2, as well as the β sheet. However, there is a relatively large difference in the position and orientation of RsfS bound to L14. The position of the E. coli model of RsfS on L14 was rotated about 172° compared with the Mtb cryo-EM structure. While we cannot rule out the possibility that E. coli RsfS could bind differently, Mtb RsfS and L14 share 25% and 66% sequence identities with their E. coli counterparts, respectively, and therefore one would expect the binding to be conserved.

DISCUSSION

Mtb RsfS Regulates Translation by Binding to 50S

The mechanism by which RsfS is able to slow translation when cells transition to the stationary phase is not well understood. Transcriptome analysis in E. coli and Mtb shows that RsfS mRNA levels are not significantly altered during bacterial growth (Hruz et al., 2008). Indeed, RsfS mRNA levels remained relatively constant from the early log phase through the stationary phase, a time when one would expect RsfS levels to increase so that translation would be slowed. The gene expression profiling results indicated that mRNA levels of RsfS and other ribosomal proteins are consistent (Galagan et al., 2010; Reddy et al., 2009). This suggests that RsfS is likely to be regulated at the protein level.

RsfS inhibits translation by directly binding to the L14 protein of the 50S ribosome at a site that, in the functional 70S, is occupied by helix 14 of the rRNA 16S in the 30S subunit. The overlap between the RsfS binding site and that of helix 14 is relatively small, only about 30 Å2, compared with the extremely large buried surface area of the 50S with the 30S. The structures are consistent with the observation that RsfS binding is sufficient to compete with the 30S subunit. However, RsfS was unable to disassociate preformed 70S, as we only observed binding to the free 50S subunit. Yet when RsfS was added to an Mtb cell-free translation assay, it was able to significantly decrease translation, indicating that the monomer was present in the cell-free translation conditions. These results suggest that RsfS does not interfere with normal ribosomal functions during the elongation phase but it has the potential to block the formation of the functional 70S ribosome and to inhibit mRNA translation, both of which are consistent with RsfS’s role as a regulator of translation.

The crystal structure and biochemical analysis clearly show that RsfS is a dimer in solution, and the EM density indicates that a single subunit of the dimer binds to L14 of the 50S subunit. It is highly unlikely that the second subunit of the dimer is bound and not visible due to flexibility, because most of the RsfS dimer interface observed in the crystal structure directly interacts with L14 in the inhibited RsfS-50S complex. Therefore, dissociation of the RsfS dimer must occur before binding to the L14.

A relatively large cavity is found adjacent to the interface of the RsfS dimer. On the periphery, the cavity contains the side chains of mostly polar and charged residues. It is tempting to speculate that binding of a molecule might be responsible for dimer dissociation. Numerous attempts using pull-down experiments have neither identified a molecule bound to RsfS nor conditions where RsfS dimer dissociates. However, other groups have reported that RsfS from E. coli interacted with several hypothetical proteins, such as yehL, yehQ, yihU, and yjcF in E. coli (Butland et al., 2005). Mtb does not have any identifiable orthologs to any of these proteins, again suggesting that Mtb dimer dissociation may be a regulatory event.

E. coli has three other proteins that have been implicated in the regulation of the ribosome in the bacterial transition to the stationary phase (Polikanov et al., 2012). Ribosome modulation factor (RMF) and hibernation promoting factor (HPF) are thought to act by inducing dimerization of the ribosome into a 100S particle, which is incapable of translation. Protein Y (PY) has been shown to reverse this ribosome dimerization, although the resulting 70S ribosome appeared to be inactive. RMF and HPF are thought to induce dimerization by binding to the mRNA and tRNA binding sites on 30S subunits. PY is a paralog of HPF and its binding site overlaps with that of HPF and part of the RMF. We searched for the presence of RMF, HPF, and PY by BLASTP (Altschul et al., 1990) in Mtb using E. coli counterpart sequences and found that only one hypothetical protein (Rv3241c) shares homology with HPF.

RsfS May Have Multiple Roles in Mtb

It is possible that RsfS serves other roles in the bacterium aside from simply silencing the ribosome. Interestingly, the RsfS ortholog in humans, C7orf30, has been shown to participate in the assembly and stability of the large subunit of the mitochondrial ribosome. Inactivation of C7orf30 using RNAi leads to instability and an assembly defect in the large subunit, which results in reduced mitochondrial translation (Fung et al., 2013; Rorbach et al., 2012; Wanschers et al., 2012).

The initiation factor eIF6 is highly conserved from yeast to mammals. It binds to RpL23, the L14 counterpart in yeast. Binding of eIF6 to the large 60S subunit inhibits subunit joining (Gartmann et al., 2010; Klinge et al., 2011). The mechanism of anti-association factor eIF6 is likely to be similar to that of RsfS. However, this protein shares no structural similarity with RsfS. While eIF6 consists of five β sheets that form a barrel to cap rpL23, RsfS contains only one β sheet. It is possible that RsfS is involved in the assembly of the ribosome and prevents the association of a premature 50S.

EXPERIMENTAL PROCEDURES

Expression and Purification of Mtb RsfS and E. coli RsfS

Mtb rsfS wild-type and site-directed mutant (Y102A, E74A) were cloned into the NdeI and HindIII sites of p1602-dest (Life Technologies) vector. The vectors encoded a C-terminal His6-tagged RsfS and that was transformed into M. smegmatis cells MC24517. Colonies containing the plasmid were selected by hygromycin. For large-scale production of recombinant proteins, cells were grown in 6 l of 7H9 broth to a cell density (OD600) of 0.8, and then induced by 0.2% acetamide at 37°C for 8 hr. The Mtb RsfS protein was purified according to a modified protocol (Noens et al., 2011). The culture was centrifuged, separated from the pellet, and the cell pellet was lysed using a French press in lysis buffer composed of 20 mM Tris-HCl (pH 7.5), 150 mM NaCl, and 10 mM imidazole. The suspension was centrifuged at 30,000 × g for 45 min at 4°C and its supernatant was applied to a nickel chromatography column (GE Healthcare). The column was washed extensively and the overexpressed RsfS protein was eluted using elution buffer (lysis buffer with 250 mM imidazole [pH 7.5]). Next, size-exclusion chromatography was performed using a Superdex 75 column (GE Healthcare) in a buffer containing 20 mM Tris-Cl (pH 7.5) and 150 mM NaCl and the protein profile was compared with protein molecular size standards. The Y102A RsfS mutant was only used for crystallization. Both wild-type RsfS and mutant E74A were used in the ribosome function assay and the nickel pull-down assay, and only wild-type RsfS was used for cryo-EM structure determination.

E. coli rsfS wild-type was cloned into the NdeI and XhoI sites of pET22b (Novagen) vector. The final construct encoded a C-terminal His6-tagged RsfS and was transformed into E. coli strain BL21 (DE3). The colonies containing plasmid were selected by ampicillin. For large-scale production, cells were grown in 6 l of LB broth up to a cell density (OD600) of 0.8 and then induced by 0.5 mM imidazole at 37°C for 3 hr. The E. coli RsfS protein was purified with the same method as Mtb RsfS.

Purification of Mtb 70S and 50S

Mtb cells MC27000 (Vilcheze et al., 2011) were grown in 7H9 medium supplemented with 10% oleic albumin dextrose catalase (BD), 0.5% glycerol, 0.05% Tween-80, and 50 μg/ml pantothenic acid at 37°C until an OD600 of 1.0. The following procedures were performed at 4°C. Harvested cells were lysed in a bead beater (BioSpec) in lysis buffer (20 mM Tris-HCl [pH 7.5], 100 mM NH4Cl, 10 mM MgCl2, 0.5 mM EDTA, 6 mM 2-mercaptoethanol). Mtb ribosome 70S and 50S were purified according to modified protocols (Noll et al., 1973; Selmer et al., 2006). Cell lysate was clarified by centrifugation at 30,000 × g for 1 hr. The supernatant was pelleted in sucrose cushion buffer (20 mM HEPES [pH 7.5], 1.1 M sucrose, 10 mM MgCl2, 0.5 M KCl, and 0.5 mM EDTA) at 40,000 rpm in a Beckman Type 45Ti rotor for 20 hr. The pellet was resuspended in the buffer containing 20 mM Tris-HCl (pH 7.5), 1.5 M (NH4)2SO4, 0.4 M KCl, and 10 mM MgCl2. The suspension was then applied to a hydrophobic interaction column (Toyopearl Butyl-650S) and eluted with a reverse ionic strength gradient from 1.5 M to 0 M (NH4)2SO4 in the buffer containing 20 mM Tris-HCl (pH 7.5), 0.4 M KCl, and 10 mM MgCl2. The eluted ribosome peak was changed to re-association buffer (5 mM HEPES-NaOH [pH 7.5], 10 mM NH4Cl, 50 mM KCl, 10 mM MgCl2, and 6 mM 2-mercaptoethanol) or dissociation buffer (20 mM Tris-HCl [pH 7.5], 2 mM MgCl2, 150 mM NH4Cl, 50 mM KCl, and 6 mM 2-mercaptoethanol) and concentrated before loading on top of a 10%–40% linear sucrose gradient centrifuged in a Beckman SW28 rotor at 19,000 rpm for 19 hr. The 70S and 50S fractions were concentrated to about A260 = 300 after removal of the sucrose.

Ribosome Functional Assay

The assay used to measure in vitro ribosome activity relied on the production of GFP in an Mtb-based cell-free system. Mtb S100 cell-free extract was prepared from Mtb MC27000. The supernatant was pelleted in the buffer containing 20 mM HEPES (pH 7.5), 1.1 M sucrose, 10 mM MgCl2, 0.5 M KCl, and 0.5 mM EDTA to remove endogenous ribosome. The assay was carried out in a 96-well plate, which involved incubating the necessary substrates with ribosome-free cell extract in an incubator plate reader. The standard reaction mixture contained 2 mM each of the 20 amino acids, 33 mM phosphoenolpyruvate, 0.33 mM nicotinamide adenine dinucleotide, 0.26 mM coenzyme A, 2 μl of 0.85 μM purified Mtb ribosome, 200 ng of GFP mRNA, and 24 μl of S100 ribosome-free Mtb cell-free extract in certain buffer (Swartz et al., 2004). GFP mRNA was prepared from an in vitro transcription assay (Baugh et al., 2001). Either Mtb RsfS or E. coli RsfS was added to the reaction to final concentrations of 0.14 μM and 0.28 μM. The final concentrations of streptomycin and chloramphenicol were 0.7 μM and 0.14 μM, respectively. The total volume of the assay was 100 μl and samples were incubated at 37° for 40 hr.

Crystallization and Structure Determination of Mtb RsfS

Nine different single-point mutants were made to hydrophobic amino acids that were predicted to be on the surface. Only one mutant protein of Y102A yielded diffraction-quality crystals. The RsfS mutant Y102A was concentrated to ~40 mg/ml and mixed with an equal volume of 0.1 M sodium cacodylate (pH 6.5), 0.2 M magnesium acetate, and 30% (+/−)-2-methyl-2,4-pentanediol. Crystals were produced by vapor diffusion in sitting-drop trays at 20°C and were directly harvested from the drop, flash frozen, and stored in liquid N2. Diffraction data were collected to 2.1 Å at the Advanced Light Source synchrotron at Lawrence Berkeley National Laboratory and were processed by HKL2000 (Otwinowski and Minor, 1997) (Table 1). The structure was solved by molecular replacement with a truncated poly-Ala model (Ala7-Ala103) derived from a B. halodurans ortholog (PDB: 2O5A) as the search model in AutoMR of PHENIX (Adams et al., 2010). After initial refinement, the side chains were rebuilt in AutoBuild of PHENIX. Then iterative cycles of manual rebuilding in COOT (Emsley et al., 2010) and PHENIX refinement led to the final model. Simulated annealing was applied in the early stages of refinement.

Cryo-EM Sample Preparation

RsfS was added to purified 50S ribosome in a molar ratio of 15:1. This mixture was further diluted with 5 mM HEPES-Na (pH 7.5), 10 mM NH4Cl, 50 mM KCl, and 10 mM MgCl2 to final concentrations of 50S at 0.36 μM and RsfS at 5.4 μM. This sample was then applied onto a 200 mesh R2/2 Quantifoil grid (Quantifoil Micro Tools). The grid was previously glow discharged. After applying the sample, the grid was blotted and rapidly frozen in liquid ethane using a Vitrobot (FEI), and then stored in liquid nitrogen before imaging.

Electron Microscopy, Image Processing, Map Segmentation, and Visualization

The grid was imaged on an FEI Tecnai F20 with a field emission gun operated at 200 kV (FEI). One hundred and sixty-five micrographs were recorded at an effective magnification of 81,081×, on a Gatan 4k × 4k charge-coupled device camera (Gatan) with a final image pixel size of 1.85 Å.

Each micrograph was carefully examined for drift and astigmatism. One hundred and fifty-four micrographs with a defocus range of 1–2.5 μm were used for further processing. We carefully boxed the raw particles using EMAN2 (Tang et al., 2007) and manually removed (1) ice contamination, (2) particles that touched each other, and (3) particles that were on the carbon. This gave us 73,103 particles. We then used the unsupervised 3D classification in Relion 1.3 (Scheres, 2012) to classify the particles into ten classes and removed seven bad classes of particles, leaving a total of 42,138 “clean” particles (Figure S2C, step 1). We then processed these selected particles to generate a 3D map in Relion 1.3. This map (named Initial Map 1) already shows relatively weak extra density for RsfS next to the L14 protein. Next, we manually erased the extra density of the RsfS in UCSF Chimera (Pettersen et al., 2004) and generated Initial Map 2. Both the 3D initial maps were subjected to additional low-pass filtering with a Gaussian radius of 30 Å before applied for the supervised classification to separate the raw particles into two classes, 21,287 particles for RsfS-free and 20,851 particles for RsfS bound. This 30-Å radius filtering removed the high-resolution features and minimized the artifacts by manually removing the weak RsfS density, but still showed a noticeable difference in the RsfS binding site. Finally, we swapped the two reference maps to refine against the separated raw particles. The refined maps showed consistent results, with or without RsfS density, even with the contrary reference maps (Figure S2C). This validated the correct separation of the raw particles and ruled out the possibility of model bias. The final resolutions were 9.3 Å for RsfS-free 50S and 9.1 Å for the RsfS-bound 50S (Figure S2B). The local resolution of RsfS in the RsfS-bound state was ~9 Å according to the ResMap result. Map segmentation was done in UCSF Chimera with the reference from an E. coli 50S ribosome PDB structure (PDB: 2I2V; Berk et al., 2006). Figures of the maps and models were produced with UCSF Chimera.

Molecular Modeling, Docking and Flexible Fitting of RsfS within the Density Map

The Mtb RsfS monomer from the RsfS crystal structure was first roughly docked onto the homology model of Mtb L14, which was built with the SWISS-MODEL server (Schwede et al., 2003) in the cryo-EM density map. To avoid the initial model bias in the refinement, the complex of L14 and RsfS was diversified into 1,000 initial models in the following two steps: (1) arbitrarily displacing RsfS away from L14 within a hemisphere of 10-Å radius; (2) randomly applying a rotation on the RsfS with the azimuthal angle between 0° and 360°, an altitude angle between 0° and 180°, and phi angle between 0° and 360°. All the 1,000 initial models were then refined independently with MOSAICS-EM (Zhang et al., 2012) using the cryo-EM density map as a constraint. The best-fit model to the EM density map was further refined using the real-space refinement routine in PHENIX to optimize the protein stereochemistry (Adams et al., 2010).

Supplementary Material

updated supplement

Highlights.

  • Crystal structure of a ribosomal silencing factor, RsfS, from Mtb

  • Cryo-EM structures of the Mtb ribosome’s large subunit with or without RsfS bound

  • RsfS dimer dissociates into monomers in order to bind to L14 of the 50S subunit

  • RsfS inhibits the association of 30S subunit and blocks protein synthesis

Acknowledgments

This work was supported by Welch foundation grants A-0015 (J.C.S.), A-1863 (J.Z.), and NIH TB Structural genomics grant P01AI095208. We would like to thank the staff at beamline 5.0.2 managed by the Berkeley Center for Structural Biology (BCSB) at the Advanced Light Source (ALS) for technical support. The BCSB is supported in part by the NIH, National Institute of General Medical Sciences, and the Howard Hughes Medical Institute. The ALS is supported by the Director, Office of Science, Office of Basic Energy Sciences, of the US Department of Energy under contract no. DE-AC02-05CH11231. J.Z. is grateful to Michael Levitt and Roger Kornberg at Stanford University for their support on the cryo-EM experiments and to BioX3 at Stanford University for the initial cryo-EM data processing. J.Z. would like to acknowledge the Texas A&M Supercomputing Facility for providing computing resources along with the Center for Phage Technology and the Department of Biochemistry and Biophysics at Texas A&M University for providing startup funding. We thank Jeng-Yih Chang for the preparation of Figure 3H and Dr. Matthew Sachs for carefully editing the paper.

Footnotes

ACCESSION NUMBERS

Coordinates and structure factors of the Mtb RsfS have been deposited in PDB: 4WCW. The cryo-EM maps have been deposited in the EMDB: Mtb 50S (EMD-6178) and Mtb RsfS-50S (EMD-6177).

SUPPLEMENTAL INFORMATION

Supplemental Information includes three figures and can be found with this article online at http://dx.doi.org/10.1016/j.str.2015.07.014.

AUTHOR CONTRIBUTIONS

X.L. and C.J. purified Mtb ribosome and crystallized Mtb RsfS. C.J. developed the function assay based on the GFP signal. L.H. collected and analyzed the X-ray diffraction data and Q.S. solved the crystal structure of RsfS. J.Z. and K.Y. solved the cryo-EM structures and interpreted the binding between the RsfS and the Mtb 50S. J.S. and J.Z. are the principal investigators.

References

  1. Adams PD, Afonine PV, Bunkoczi G, Chen VB, Davis IW, Echols N, Headd JJ, Hung LW, Kapral GJ, Grosse-Kunstleve RW, et al. PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr D Biol Crystallogr. 2010;66:213–221. doi: 10.1107/S0907444909052925. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. J Mol Biol. 1990;215:403–410. doi: 10.1016/S0022-2836(05)80360-2. [DOI] [PubMed] [Google Scholar]
  3. Anastasia K, Valerie M, Digby FW. The impact of drug resistance on Mycobacterium tuberculosis physiology: what can we learn from rifampicin? Emerg Microbes Infect. 2014;3:e17. doi: 10.1038/emi.2014.17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Ban N, Nissen P, Hansen J, Moore PB, Steitz TA. The complete atomic structure of the large ribosomal subunit at 2.4 A resolution. Science. 2000;289:905–920. doi: 10.1126/science.289.5481.905. [DOI] [PubMed] [Google Scholar]
  5. Baugh LR, Hill AA, Brown EL, Hunter CP. Quantitative analysis of mRNA amplification by in vitro transcription. Nucleic Acids Res. 2001;29:E29. doi: 10.1093/nar/29.5.e29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Berk V, Zhang W, Pai RD, Cate JH. Structural basis for mRNA and tRNA positioning on the ribosome. Proc Natl Acad Sci USA. 2006;103:15830–15834. doi: 10.1073/pnas.0607541103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Butland G, Peregrin-Alvarez JM, Li J, Yang W, Yang X, Canadien V, Starostine A, Richards D, Beattie B, Krogan N, et al. Interaction network containing conserved and essential protein complexes in Escherichia coli. Nature. 2005;433:531–537. doi: 10.1038/nature03239. [DOI] [PubMed] [Google Scholar]
  8. Davies C, White SW, Ramakrishnan V. The crystal structure of ribosomal protein L14 reveals an important organizational component of the translational apparatus. Structure. 1996;4:55–66. doi: 10.1016/s0969-2126(96)00009-3. [DOI] [PubMed] [Google Scholar]
  9. Emsley P, Lohkamp B, Scott WG, Cowtan K. Features and development of Coot. Acta Crystallogr D Biol Crystallogr. 2010;66:486–501. doi: 10.1107/S0907444910007493. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Flynn JL, Chan J. Tuberculosis: latency and reactivation. Infect Immun. 2001;69:4195–4201. doi: 10.1128/IAI.69.7.4195-4201.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Fung S, Nishimura T, Sasarman F, Shoubridge EA. The conserved interaction of C7orf30 with MRPL14 promotes biogenesis of the mitochondrial large ribosomal subunit and mitochondrial translation. Mol Biol Cell. 2013;24:184–193. doi: 10.1091/mbc.E12-09-0651. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Galagan JE, Sisk P, Stolte C, Weiner B, Koehrsen M, Wymore F, Reddy TB, Zucker JD, Engels R, Gellesch M, et al. TB database 2010: overview and update. Tuberculosis (Edinb) 2010;90:225–235. doi: 10.1016/j.tube.2010.03.010. [DOI] [PubMed] [Google Scholar]
  13. Gao YG, Selmer M, Dunham CM, Weixlbaumer A, Kelley AC, Ramakrishnan V. The structure of the ribosome with elongation factor G trapped in the posttranslocational state. Science. 2009;326:694–699. doi: 10.1126/science.1179709. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Gartmann M, Blau M, Armache JP, Mielke T, Topf M, Beckmann R. Mechanism of eIF6-mediated inhibition of ribosomal subunit joining. J Biol Chem. 2010;285:14848–14851. doi: 10.1074/jbc.C109.096057. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Gomez JE, McKinney JD. M. tuberculosis persistence, latency, and drug tolerance. Tuberculosis (Edinb) 2004;84:29–44. doi: 10.1016/j.tube.2003.08.003. [DOI] [PubMed] [Google Scholar]
  16. Harms J, Schluenzen F, Zarivach R, Bashan A, Gat S, Agmon I, Bartels H, Franceschi F, Yonath A. High resolution structure of the large ribosomal subunit from a mesophilic eubacterium. Cell. 2001;107:679–688. doi: 10.1016/s0092-8674(01)00546-3. [DOI] [PubMed] [Google Scholar]
  17. Harries AD, Dye C. Tuberculosis. Ann Trop Med Parasitol. 2006;100:415–431. doi: 10.1179/136485906X91477. [DOI] [PubMed] [Google Scholar]
  18. Hauser R, Pech M, Kijek J, Yamamoto H, Titz B, Naeve F, Tovchigrechko A, Yamamoto K, Szaflarski W, Takeuchi N, et al. RsfA (YbeB) proteins are conserved ribosomal silencing factors. PLoS Genet. 2012;8:e1002815. doi: 10.1371/journal.pgen.1002815. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hruz T, Laule O, Szabo G, Wessendorp F, Bleuler S, Oertle L, Widmayer P, Gruissem W, Zimmermann P. Genevestigator v3: a reference expression database for the meta-analysis of transcriptomes. Adv Bioinformatics. 2008;2008:420747. doi: 10.1155/2008/420747. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Jiang M, Sullivan SM, Walker AK, Strahler JR, Andrews PC, Maddock JR. Identification of novel Escherichia coli ribosome-associated proteins using isobaric tags and multidimensional protein identification techniques. J Bacteriol. 2007;189:3434–3444. doi: 10.1128/JB.00090-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Klinge S, Voigts-Hoffmann F, Leibundgut M, Arpagaus S, Ban N. Crystal structure of the eukaryotic 60S ribosomal subunit in complex with initiation factor 6. Science. 2011;334:941–948. doi: 10.1126/science.1211204. [DOI] [PubMed] [Google Scholar]
  22. Korostelev A, Trakhanov S, Laurberg M, Noller HF. Crystal structure of a 70S ribosome-tRNA complex reveals functional interactions and rearrangements. Cell. 2006;126:1065–1077. doi: 10.1016/j.cell.2006.08.032. [DOI] [PubMed] [Google Scholar]
  23. Krissinel E, Henrick K. Inference of macromolecular assemblies from crystalline state. J Mol Biol. 2007;372:774–797. doi: 10.1016/j.jmb.2007.05.022. [DOI] [PubMed] [Google Scholar]
  24. Kucukelbir A, Sigworth FJ, Tagare HD. Quantifying the local resolution of cryo-EM density maps. Nat Methods. 2014;11:63–65. doi: 10.1038/nmeth.2727. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kumar A, Majid M, Kunisch R, Rani PS, Qureshi IA, Lewin A, Hasnain SE, Ahmed N. Mycobacterium tuberculosis DosR regulon gene Rv0079 encodes a putative, ‘dormancy associated translation inhibitor (DATIN)’. PLoS One. 2012;7:e38709. doi: 10.1371/journal.pone.0038709. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kwan CK, Ernst JD. HIV and tuberculosis: a deadly human syndemic. Clin Microbiol Rev. 2011;24:351–376. doi: 10.1128/CMR.00042-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Moore PB. How should we think about the ribosome? Annu Rev Biophys. 2012;41:1–19. doi: 10.1146/annurev-biophys-050511-102314. [DOI] [PubMed] [Google Scholar]
  28. Noens EE, Williams C, Anandhakrishnan M, Poulsen C, Ehebauer MT, Wilmanns M. Improved mycobacterial protein production using a Mycobacterium smegmatis groEL1DeltaC expression strain. BMC Biotechnol. 2011;11:27. doi: 10.1186/1472-6750-11-27. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Noll M, Hapke B, Schreier MH, Noll H. Structural dynamics of bacterial ribosomes. I Characterization of vacant couples and their relation to complexed ribosomes. J Mol Biol. 1973;75:281–294. doi: 10.1016/0022-2836(73)90021-1. [DOI] [PubMed] [Google Scholar]
  30. Nooren IM, Thornton JM. Structural characterisation and functional significance of transient protein-protein interactions. J Mol Biol. 2003;325:991–1018. doi: 10.1016/s0022-2836(02)01281-0. [DOI] [PubMed] [Google Scholar]
  31. Otwinowski Z, Minor W. Processing of X-ray diffraction data collected in oscillation mode. In: Carter CW Jr, Sweet RM, editors. Macromolecular Crystallography, Part A. Academic Press; 1997. pp. 307–326. [DOI] [PubMed] [Google Scholar]
  32. Petrov AS, Bernier CR, Hsiao C, Norris AM, Kovacs NA, Waterbury CC, Stepanov VG, Harvey SC, Fox GE, Wartell RM, et al. Evolution of the ribosome at atomic resolution. Proc Natl Acad Sci USA. 2014;111:10251–10256. doi: 10.1073/pnas.1407205111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, Ferrin TE. UCSF Chimera–a visualization system for exploratory research and analysis. J Comput Chem. 2004;25:1605–1612. doi: 10.1002/jcc.20084. [DOI] [PubMed] [Google Scholar]
  34. Polikanov YS, Blaha GM, Steitz TA. How hibernation factors RMF, HPF, and YfiA turn off protein synthesis. Science. 2012;336:915–918. doi: 10.1126/science.1218538. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Reddy TB, Riley R, Wymore F, Montgomery P, DeCaprio D, Engels R, Gellesch M, Hubble J, Jen D, Jin H, et al. TB database: an integrated platform for tuberculosis research. Nucleic Acids Res. 2009;37:D499–D508. doi: 10.1093/nar/gkn652. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Rorbach J, Gammage PA, Minczuk M. C7orf30 is necessary for biogenesis of the large subunit of the mitochondrial ribosome. Nucleic Acids Res. 2012;40:4097–4109. doi: 10.1093/nar/gkr1282. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Russell DG. Who puts the tubercle in tuberculosis? Nat Rev Microbiol. 2007;5:39–47. doi: 10.1038/nrmicro1538. [DOI] [PubMed] [Google Scholar]
  38. Scheres SH. RELION: Implementation of a Bayesian approach to cryo-EM structure determination. J Struct Biol. 2012;180:519–530. doi: 10.1016/j.jsb.2012.09.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Scheres SH, Chen S. Prevention of overfitting in cryo-EM structure determination. Nat Methods. 2012;9:853–854. doi: 10.1038/nmeth.2115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Schuwirth BS, Borovinskaya MA, Hau CW, Zhang W, Vila-Sanjurjo A, Holton JM, Cate JHD. Structures of the bacterial ribosome at 3.5 A resolution. Science. 2005;310:827–834. doi: 10.1126/science.1117230. [DOI] [PubMed] [Google Scholar]
  41. Schwede T, Kopp J, Guex N, Peitsch MC. SWISS-MODEL: an automated protein homology-modeling server. Nucleic Acids Res. 2003;31:3381–3385. doi: 10.1093/nar/gkg520. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Selmer M, Dunham CM, Murphy FV, 4th, Weixlbaumer A, Petry S, Kelley AC, Weir JR, Ramakrishnan V. Structure of the 70S ribosome complexed with mRNA and tRNA. Science. 2006;313:1935–1942. doi: 10.1126/science.1131127. [DOI] [PubMed] [Google Scholar]
  43. Shasmal M, Sengupta J. Structural diversity in bacterial ribosomes: mycobacterial 70S ribosome structure reveals novel features. PLoS One. 2012;7:e31742. doi: 10.1371/journal.pone.0031742. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Swartz JR, Jewett MC, Woodrow KA. Cell-free protein synthesis with prokaryotic combined transcription-translation. Methods Mol Biol. 2004;267:169–182. doi: 10.1385/1-59259-774-2:169. [DOI] [PubMed] [Google Scholar]
  45. Tang G, Peng L, Baldwin PR, Mann DS, Jiang W, Rees I, Ludtke SJ. EMAN2: an extensible image processing suite for electron microscopy. J Struct Biol. 2007;157:38–46. doi: 10.1016/j.jsb.2006.05.009. [DOI] [PubMed] [Google Scholar]
  46. Trauner A, Lougheed KE, Bennett MH, Hingley-Wilson SM, Williams HD. The dormancy regulator DosR controls ribosome stability in hypoxic mycobacteria. J Biol Chem. 2012;287:24053–24063. doi: 10.1074/jbc.M112.364851. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Vilcheze C, Baughn AD, Tufariello J, Leung LW, Kuo M, Basler CF, Alland D, Sacchettini JC, Freundlich JS, Jacobs WR., Jr Novel inhibitors of InhA efficiently kill Mycobacterium tuberculosis under aerobic and anaerobic conditions. Antimicrob Agents Chemother. 2011;55:3889–3898. doi: 10.1128/AAC.00266-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Wanschers BF, Szklarczyk R, Pajak A, van den Brand MA, Gloerich J, Rodenburg RJ, Lightowlers RN, Nijtmans LG, Huynen MA. C7orf30 specifically associates with the large subunit of the mitochondrial ribosome and is involved in translation. Nucleic Acids Res. 2012;40:4040–4051. doi: 10.1093/nar/gkr1271. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Wayne LG, Sohaskey CD. Nonreplicating persistence of Mycobacterium tuberculosis. Annu Rev Microbiol. 2001;55:139–163. doi: 10.1146/annurev.micro.55.1.139. [DOI] [PubMed] [Google Scholar]
  50. Wilson DN. Ribosome-targeting antibiotics and mechanisms of bacterial resistance. Nat Rev Microbiol. 2014;12:35–48. doi: 10.1038/nrmicro3155. [DOI] [PubMed] [Google Scholar]
  51. Yusupov MM, Yusupova GZ, Baucom A, Lieberman K, Earnest TN, Cate JH, Noller HF. Crystal structure of the ribosome at 5.5 A resolution. Science. 2001;292:883–896. doi: 10.1126/science.1060089. [DOI] [PubMed] [Google Scholar]
  52. Zhang J, Minary P, Levitt M. Multiscale natural moves refine macromolecules using single-particle electron microscopy projection images. Proc Natl Acad Sci USA. 2012;109:9845–9850. doi: 10.1073/pnas.1205945109. [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

updated supplement
NIHMS713748-supplement-updated_supplement.pdf (1.8MB, pdf)

RESOURCES