Processive enzymes like polymerases or ribosomes are often studied in bulk experiments by monitoring time-dependent signals, such as fluorescence time traces. However, due to biomolecular process stochasticity, ensemble signals may lack the distinct features of single-molecule signals. Here, we demonstrate that, under certain conditions, bulk signals from processive reactions can be decomposed to unveil hidden information about individual reaction steps. Using mRNA translation as a case study, we show that decomposing a noisy ensemble signal generated by the translation of mRNAs with more than a few codons is an ill-posed problem, addressable through Tikhonov regularization. We apply our method to the fluorescence signatures of in-vitro translated LepB mRNA and determine codon-position dependent translation rates and corresponding state-specific fluorescence intensities. We find a significant change in fluorescence intensity after the fourth and the fifth peptide bond formation, and show that both codon position and encoded amino acid have an effect on the elongation rate. This demonstrates that our approach enhances the information content extracted from bulk experiments, thereby expanding the range of these time- and cost-efficient methods.

Alternative ribosome-rescue factor B (ArfB) rescues ribosomes stalled on non-stop mRNAs by releasing the nascent polypeptide from the peptidyl-tRNA. By rapid kinetics we show that ArfB selects ribosomes stalled on short truncated mRNAs, rather than on longer mRNAs mimicking pausing on rare codon clusters. In combination with cryo-electron microscopy we dissect the multistep rescue pathway of ArfB, which first binds to ribosomes very rapidly regardless of the mRNA length. The selectivity for shorter mRNAs arises from the subsequent slow engagement step, as it requires longer mRNA to shift to enable ArfB binding. Engagement results in specific interactions of the ArfB C-terminal domain with the mRNA entry channel, which activates peptidyl-tRNA hydrolysis by the N-terminal domain. These data reveal how protein dynamics translate into specificity of substrate recognition and provide insights into the action of a putative rescue factor in mitochondria.

Many cellular proteins begin folding while the peptide is still attached to the translating ribosome. Contrary to well-studied folding of isolated proteins in vitro, co-translational folding is vectorial. The ribosome interacts with the emerging peptide, protects it from aggregation and misfolding, and may stabilize intermediates that do not exist in solution but are important for vectorial folding. Currently, only few methods can address the dynamic properties or the folding pathways of the emerging nascent chains. Here we investigate co-translational folding and nascent chain dynamics of the N-terminal HemK domain (HemK NTD) in a minimal in vitro translation system consisting of pre-assembled initiation complexes, elongation factors, and aminoacylated tRNAs. It was previously reported that the HemK NTD folds on the ribosome through the formation of compact intermediate states. We used a single molecule equilibrium technique combining photoinduced electron transfer-based quenching and fluorescence correlation spectroscopy (PET-FCS) to probe the dynamics of these compact intermediate states on the ribosome. We developed a PET reporter between a native tryptophan and fluorescently labeled N-terminal methionine of HemK. PET-FCS data of purified ribosome nascent chain complexes resulted in multi-exponential curves reporting on nascent peptide dynamics in the nanosecond to microsecond timescale. By analyzing the data, we obtained rates of interactions between the N-terminal dye and the internal tryptophan or the quencher residues of the ribosome surface. By comparing the rates of different length and varied stability nascent chains with those of peptides released from the ribosome, we could further characterize the folding pathway itself and the dynamics of the observed co-translational folding intermediates of the HemK NTD while on the ribosome.

Article Figures and data Abstract eLife digest Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract Release factors RF1 and RF2 promote hydrolysis of peptidyl-tRNA during translation termination. The GTPase RF3 promotes recycling of RF1 and RF2. Using single molecule FRET and biochemical assays, we show that ribosome termination complexes that carry two factors, RF1–RF3 or RF2–RF3, are dynamic and fluctuate between non-rotated and rotated states, whereas each factor alone has its distinct signature on ribosome dynamics and conformation. Dissociation of RF1 depends on peptide release and the presence of RF3, whereas RF2 can dissociate spontaneously. RF3 binds in the GTP-bound state and can rapidly dissociate without GTP hydrolysis from termination complex carrying RF1. In the absence of RF1, RF3 is stalled on ribosomes if GTP hydrolysis is blocked. Our data suggest how the assembly of the ribosome–RF1–RF3–GTP complex, peptide release, and ribosome fluctuations promote termination of protein synthesis and recycling of the release factors. https://doi.org/10.7554/eLife.34252.001 eLife digest Inside cells, molecular machines called ribosomes make proteins using messenger RNA as a template. However, the template contains more than just the information needed to create the protein. A ‘stop codon’ in the mRNA marks where the ribosome should stop. When this is reached a group of proteins called release factors removes the newly made protein from the ribosome. Bacteria typically have three types of release factors. RF1 and RF2 recognize the stop codon, and RF3 helps to release RF1 or RF2 from the ribosome so that it can be recycled to produce another protein. It was not fully understood how the release factors interact with the ribosome and how this terminates protein synthesis. Adio et al. used TIRF microscopy to study individual ribosomes from the commonly studied bacteria species Escherichia coli. This technique allows researchers to monitor movements of the ribosome and record how release factors bind to it. The results of the experiments performed by Adio et al. show that although RF1 and RF2 are very similar to each other, they interact with the ribosome in different ways. In addition, only RF1 relies upon RF3 to release it from the ribosome; RF2 can release itself. RF3 releases RF1 by forcing the ribosome to change shape. RF3 then uses energy produced by the breakdown of a molecule called GTP to help release itself from the ribosome. Most importantly, the findings presented by Adio et al. highlight that the movements of ribosomes and release factors during termination are only loosely coupled rather than occur in a set order. Other molecular machines are likely to work in a similar way. The results could also help us to understand the molecular basis of several human diseases, such as Duchenne muscular dystrophy and cystic fibrosis, that result from ribosomes not recognizing stop codons in the mRNA. https://doi.org/10.7554/eLife.34252.002 Introduction Termination of protein synthesis occurs when a translating ribosome encounters one of the three universally conserved stop codons UAA, UAG or UGA. In bacteria, the release of the nascent peptide is promoted by release factors RF1 and RF2 which recognize the stop codons in the A site and hydrolyze the ester bond in the peptidyl-tRNA bound to the P site, allowing the nascent peptide to leave the ribosome through the polypeptide exit tunnel (Dunkle and Cate, 2010; Nakamura et al., 1996). RF1 and RF2 bind to the ribosome in the space between the small and large ribosomal subunits. RF1 and RF2 differ in their stop codon specificity: RF1 utilizes a conserved PET motif to recognize UAG and UAA codons, whereas RF2 uses an SPF motif to recognize UGA and UAA codons. Both RF1 and RF2 have a universally conserved GGQ motif which promotes the catalysis of peptidyl-tRNA hydrolysis (Seit-Nebi et al., 2001); mutations of the GGQ motif to GAQ or GGA inhibit peptide release (Frolova et al., 1999; Mora et al., 2003; Shaw and Green, 2007; Zavialov et al., 2002). After peptide release, RF1 and RF2 dissociate from the post-termination complex to allow for the next steps of translation. The dissociation is accelerated by RF3, a translational GTPase that binds and hydrolyses GTP in the course of termination (Freistroffer et al., 1997; Koutmou et al., 2014; Zavialov et al., 2002). In addition to canonical termination, RF2 takes part in non-canonical termination events such as post-peptidyl transfer quality control (Zaher and Green, 2009) and ribosome rescue on truncated mRNAs (Kurita et al., 2014). There are two different models concerning the sequence of events during termination, including the timing of peptide release, the order of RF1, RF2 and RF3 binding and dissociation, and the role of nucleotide exchange in RF3 and GTP hydrolysis. The first model of translation termination was proposed by Ehrenberg and colleagues (Zavialov et al., 2001; Zavialov et al., 2002). Based on nitrocellulose filtration experiments, the authors reported that free RF3 has a much higher affinity for GDP (Kd = 5.5 nM) than for GTP (Kd = 2.5 µM) or GDPNP (Kd = 8.5 µM) (Zavialov et al., 2001), which would imply that at cellular GTP/GDP concentrations RF3 is expected to be predominantly in the GDP form. The exchange of GDP for GTP occurs only when RF3–GDP binds to the ribosome in complex with RF1 or RF2 (Zavialov et al., 2001). In the absence of the nucleotide, RF3-dependent RF1/2 recycling is slow, which has been interpreted as an indication for a high-affinity complex of apo-RF3 to the ribosome–RF1/2 complex (Zavialov et al., 2001). Furthermore, because RF3-dependent turnover GTPase activity was stimulated by peptidyl-tRNA hydrolysis, the authors suggested that RF3 binds to the ribosome–RF1 complex only after the peptide is released. Based on these results, Ehrenberg et al. suggested the following sequential mechanism of termination: RF1/RF2 bind to the ribosome and hydrolyze peptidyl-tRNA, allowing RF3–GDP to enter the ribosome occupied by RF1 or RF2 to form an unstable encounter complex. Dissociation of GDP leads to a stable high-affinity complex with RF3 in the nucleotide-free state. The subsequent binding of GTP by RF3 promotes RF1/RF2 dissociation. In the final step, RF3 hydrolyses GTP and as a result dissociates from the ribosome (Zavialov et al., 2001; Zavialov et al., 2002). An alternative model was proposed based on the kinetic and thermodynamic analysis of GTP/GDP binding to RF3 by ensemble kinetics and equilibrium methods. The results of those experiments indicated that the affinity of RF3 to GDP and GTP is on the same order of magnitude (5 nM and 20 nM, respectively [Koutmou et al., 2014; Peske et al., 2014]). As the cellular GTP concentration is at least 10 times higher than the GDP concentration (Bennett et al., 2009), these affinities imply that nucleotide exchange in RF3 can occur spontaneously, off the ribosome, and thus RF3 could enter the ribosome in either the GTP- or GDP-bound form. Consistent with previous findings (Zavialov et al., 2001; Zavialov et al., 2002), ribosome–RF1 or ribosome–RF2 complexes accelerate nucleotide exchange in RF3 (Koutmou et al., 2014; Peske et al., 2014); however, this effect is independent of peptide release, because also a catalytically inactive RF2 mutant activates nucleotide exchange in RF3 (Peske et al., 2014; Zavialov et al., 2002). Binding of GTP to RF3 in complex with the ribosome and RF2 is rapid (130 s−1) (Peske et al., 2014), and thus the lifetime of the apo-RF3 state would be too short to assume a tentative physiological role. Peptide release results in the stabilization of the RF3–GTP–ribosome complex, thereby promoting the dissociation of RF1/2, followed by GTP hydrolysis and dissociation of RF3–GDP from the ribosome (Peske et al., 2014). Efficient translation termination not only requires the coordinated action of the release factors, but also entails conformational dynamics of the factors and the ribosome. The key conformational motions of the ribosome during termination and in general in all phases of translation include the rotation of ribosomal subunits relative to each other, the swiveling motion of the body and head domains of the small ribosomal subunit, the movement of the ribosomal protein L1 toward or away from the E-site tRNA, and the movement of tRNAs between classic and hybrid conformation. These motions are loosely coupled and gated by ligands of the ribosome such as translation factors and tRNAs (Adio et al., 2015; Chen et al., 2011; Cornish et al., 2008; Horan and Noller, 2007; Sharma et al., 2016; Shi and Joseph, 2016; Valle et al., 2003; Wasserman et al., 2016). Crystal structures show that termination complexes with RF1 or RF2 are predominantly in the non-rotated (N) state. The P-site tRNA in the complexes is in the classical state and the L1 stalk in an open conformation (Jin et al., 2010; Korostelev et al., 2008; Laurberg et al., 2008; Weixlbaumer et al., 2008). A single molecule fluorescence resonance energy transfer (smFRET) study showed that binding of RF1 to termination complexes stabilizes the open conformation of the L1 stalk, whereas in the absence of RF1 termination complexes make reversible transitions between the open and closed state (Sternberg et al., 2009); the rotation of the ribosomal subunits was not investigated directly in that study. The high sequence similarity between RF1 and RF2 suggests that the two factors interact with the ribosome in the same manner and promote peptide release by a similar mechanism (Freistroffer et al., 1997; Zavialov et al., 2001). However, structures of RF1 or RF2 bound to termination complexes reveal differences regarding the interaction with the L11 region of the 50S subunit (Korostelev et al., 2008; Laurberg et al., 2008; Petry et al., 2005; Pierson et al., 2016; Rawat et al., 2006; Rawat et al., 2003; Weixlbaumer et al., 2008). Thus, it is not clear whether RF1 and RF2 follow the same mechanism and whether they respond in the same way to the recruitment of RF3 to termination complexes. In the absence of RF1/RF2, binding of RF3 with a non-hydrolyzable GTP analog to the ribosomes where the nascent peptide has been released induces formation of the rotated (R) state of the ribosome, with the tRNA in the P/E hybrid state and the closed conformation of the L1 stalk (Gao et al., 2007; Jin et al., 2011; Zhou et al., 2012). A very similar effect of RF3–GDPNP was found by smFRET (Sternberg et al., 2009). However, it is much less clear what happens when RF1/RF2 and RF3 bind to the ribosome together. Modeling of the atomic structures of RF1 and RF2 into the cryo-EM structure of RF3-bound post-release complex suggests that the RF3-induced ribosome rearrangements break the interactions of RF1/RF2 with both the decoding center and the L11 region of the ribosome, leading to the release of RF1/RF2 (Gao et al., 2007). In this model, stable binding of RF1 and RF3 is mutually exclusive. On the other hand, a cryo-EM structure of ribosomes in complex with a deacylated tRNA in the P site, RF1, and RF3 in the apo form, that is, in the absence of added nucleotide, suggest that both factors can bind simultaneously to the ribosome (Pallesen et al., 2013). smFRET measurements carried out with post-release complexes in the presence of excess RF1 showed that the addition of RF3–GTP induced short-lived transitions from the L1-open to the L1-closed state which were not observed in the absence of RF3. This suggests that the two factors can bind to the ribosome simultaneously (Sternberg et al., 2009). No structural studies are available on the interaction of RF3 with RF2-bound complexes. The interaction of RF3 with the ribosomes prior to peptide release has not been studied. Here, we use TIRF microscopy to monitor smFRET signals reporting on subunit rotation to follow changes in ribosome conformation in response to RF1, RF2 and RF3 and the binding of each individual release factor to the ribosome during termination. Our results demonstrate how the recruitment of release factors change the ribosome conformation in termination complexes, how the dissociation of the factors is achieved, show differences in the function of RF1 and RF2, and explain the importance of GTP binding and hydrolysis by RF3. Results RF1 and RF2 have distinct effects on ribosome dynamics To monitor the rotation of the ribosomal subunits during termination, we utilized ribosomes with fluorescent labels attached to the small subunit protein S6 and the large subunit protein L9, S6-Cy5 and L9-Cy3, respectively. This FRET pair has been extensively characterized in both smFRET and ensemble kinetics experiments and reports on the formation of the non-rotated (N) or the rotated (R) state of the ribosome (Cornish et al., 2008; Ermolenko et al., 2013; Sharma et al., 2016). We prepared termination complexes on an mRNA which is translated up to the stop codon UAA recognized by both RF1 and RF2. The complexes contain a peptidyl-tRNA in the P site and have a stop codon in the A site; those complexes are denoted as pre-hydrolysis complexes (PreHC). In the absence of termination factors, PreHC is found predominantly in a state with the FRET efficiency of 0.73 ± 0.01 (denoted as 0.7 FRET state in the following) (Figure 1A, Figure 1—figure supplement 1; Supplementary file 1). Previous work has shown that this state corresponds to the N state of the ribosome (Cornish et al., 2008; Qin et al., 2014; Sharma et al., 2016). A small fraction of complexes shows a FRET state with an efficiency of 0.52 ± 0.02 (0.5 FRET state), which corresponds to the R state of the ribosome. While peptidyl-tRNA generally favors the N state, the ability of ribosomes with peptidyl-tRNA in the P site to adopt the R state at room temperature has been demonstrated previously by smFRET and cryo-EM (Cornish et al., 2008; Fischer et al., 2010; Ling and Ermolenko, 2015). The distribution of FRET efficiencies and thus the ratio between N and R conformations of PreHC is independent of the tRNA in the P site and of the presence of a single N-terminal amino acid (fMet) or a dipeptide (fMetPhe, fMetVal or fMetLys) at the P-site tRNA (Figure 1—figure supplement 1; Supplementary file 1). This finding has prompted us to use the PreHC with fMet-tRNAfMet in the P site and a stop codon in the A site as a minimal model system, following previous publications which used this approach to study termination (Casy et al., 2018; Jin et al., 2010; Koutmou et al., 2014; Kuhlenkoetter et al., 2011; Pallesen et al., 2013; Pierson et al., 2016; Shi and Joseph, 2016; Sternberg et al., 2009). Figure 1 with 3 supplements see all Download asset Open asset Subunit rotation of termination complexes in the presence of release factors. Histograms and Gaussian fits of normalized FRET distributions of S6/L9-labeled termination complexes in the presence of saturating RF1/RF2 concentrations. (A,D) PreHC and PostHC in the absence of RFs. In (D) PostHC* was generated by addition of puromycin to PreHC. (B,E) Same as A,D in the presence of RF1(GAQ) (1 µM) and RF1 (1 µM), respectively. PostHC is formed by the action of RF1. (C,F) Same as B,E in the presence of RF2(GAQ) (1 µM) and RF2 (1 µM), respectively. Cartoons show the complex composition. Grey triangles and brown circles represent the formyl group and the amino acid of fMet, respectively; stars indicate the positions of the Cy3 (green) and Cy5 (red) labels. The red shade of histogram in (D) indicates frequent reversible transitions between N and R states. The grey shade of all other histograms indicates that transitions were observed in less than 20% of traces. FRET values (Supplementary file 1) are calculated from three independent data sets. See also Figure 1—figure supplement 1, Figure 1—figure supplement 2 Figure 1—figure supplement 3 and Supplementary file 1. https://doi.org/10.7554/eLife.34252.003 To probe the effect of RF1/RF2 binding on subunit rotation, we mixed PreHC with RF1 or RF2 at saturating concentrations of the factors (Zavialov et al., 2002). Peptide release was avoided by using RF1(GAQ) or RF2(GAQ) mutants which are catalytically deficient (Frolova et al., 1999; Zavialov et al., 2002) (Figure 1—figure supplement 2A). Binding of RF1(GAQ) to PreHC stabilizes the N state (Figure 1B). The fraction of the PreHC in the R state, albeit small, is somewhat higher with RF2(GAQ) than with RF1(GAQ) (Figure 1B,C). The hydrolysis of the ester bond between the tRNA and the nascent peptide in PreHC leads to the formation of post-hydrolysis complex (PostHC). To prepare PostHC without the use of termination factors, we released nascent peptides with the help of puromycin, an analog of the A-site aminoacyl-tRNA that reacts with the peptidyl-tRNA in the P site to form peptidyl-puromycin (which then dissociates from the ribosome) and a deacylated tRNA in the P site. These complexes are denoted as PostHC*. The FRET histogram of PostHC* in the absence of the factors indicate the presence of two states, the 0.5 FRET (R) state and the 0.7 FRET (N) state (Figure 1D, Supplementary file 1). FRET time courses of individual ribosomes show reversible transitions between the N and R states (Figure 1—figure supplement 1B). The exact distribution of states depends on the P-site tRNA (Figure 1—figure supplement 1; Supplementary file 1) (Cornish et al., 2008) with tRNAfMet behaving similarly to tRNAVal, thus underscoring the suitability of the minimal model system. To test the effect of RF1 and RF2 on subunit rotation of PostHC, we added saturating concentrations of the wild type RF1 or RF2 to PreHC to allow peptide release. RF1 halts PostHC in the N state (Figure 1E, Figure 1—figure supplement 2B,D), in agreement with the previous smFRET study where RF1 binding stabilizes the L1 stalk in the open state (Sternberg et al., 2009). Binding of RF2 to PostHC shifts the equilibrium toward the N state, but not to the same extent as RF1 (Figure 1F, Figure 1—figure supplement 2C,E). Complexes with RF2 make occasional N to R transitions, in particular with RF2(GAQ) bound to PostHC (Figure 1—figure supplement 2C). These initial observations suggest that although both factors favor the N state, RF1 appears more efficient than RF2. To further probe the potential differences between RF1 and RF2, we monitored subunit rotation in response to factor binding in real time. We injected catalytic amounts of release factors to PreHC and PostHC and recorded the time courses of FRET signal changes (Figure 1—figure supplement 3). RF1(GAQ) binding to PreHC does not change the FRET efficiency appreciably, as the complex is predominantly in the N state with or without the factor (Figure 1—figure supplement 3A). Also the binding of wild-type RF1 to PreHC with subsequent peptide release does not change the FRET efficiency (Figure 1—figure supplement 3B), supporting the notion that stabilization of the N state by RF1 is independent of peptide release (Figure 1B,E). PostHC without factor fluctuates between the N and R state; binding of RF1 to PostHC halts fluctuating ribosomes in the N state and prevents excursions to the R state (Figure 1—figure supplement 3C). With RF2 the picture is somewhat different. PreHC–RF2(GAQ) is predominantly in the N state (Figure 1—figure supplement 3D). However, binding of wild type RF2 and peptide release shift the distribution toward the R state (Figure 1—figure supplement 3E). The resulting PostHC fluctuates between N and R states as shown by synchronization of FRET traces to the first N to R transition. PostHC obtained by puromycin treatment also shows reversible N to R transitions which remain undisturbed by the addition of RF2 (Figure 1—figure supplement 3F). Although the binding of the factors is not directly monitored in these experiments, the differences in the rotation pattern suggest that RF1 and RF2 have distinct effects on ribosome dynamics. Such differences may result from a shorter residence time of RF2 compared to RF1 on the ribosome, which we tested in the following experiments. Binding of RF1 and RF2 to the ribosome To measure how long the factors remain bound to the ribosome, we prepared Cy5-labeled RF1 and RF2, as well as the respective RF1/2(GAQ) mutants and ribosomes containing Cy3-labeled protein L11 (Adio et al., 2015; Chen et al., 2011; Geggier et al., 2010; Holmberg and Noller, 1999; Stöffler et al., 1980) (Figure 2—figure supplement 1A). The biochemical activity of labeled release factors was indistinguishable from that of the unlabeled counterparts (Figure 2—figure supplement 1B,C) and the factors were fully methylated (Figure 2—figure supplement 2). L11 constitutes part of the factor binding site (Pallesen et al., 2013; Petry et al., 2005; Rawat et al., 2006; Rawat et al., 2003). Recruitment of the factors to the ribosome is expected to result in high FRET efficiency. Binding of RF1 or RF1(GAQ) to either PreHC or PostHC results in a single FRET population centered at 0.72 ± 0.02 (0.7 FRET) (Figure 2A–C, Supplementary file 1). RF1 and RF1(GAQ) are stably bound to the ribosome, in agreement with previous biochemical reports on dissociation rates of RF1 and RF1(GAQ) (0.005–0.1 s−1; [Koutmou et al., 2014; Shi and Joseph, 2016]). The estimated upper limit of the dissociation rate in our experiments is 0.2 s−1 (Supplementary file 1), defined by the photobleaching rate of the FRET dye pair with kphotobleaching = 0.07–0.19 s−1 at the given imaging conditions (Materials and methods). Binding of RF2 or RF2(GAQ) to PreHC or PostHC leads to single FRET populations with efficiencies between 0.6 and 0.7 (Figure 2D–F, Supplementary file 1). However, the residence time of RF2 is much shorter compared to RF1 or RF1(GAQ), with the koff values in the range from 0.8 ± 0.1 s−1 to 1.3 ± 0.2 s−1 (Figure 2D–F, Supplementary file 1). Peptide hydrolysis has no visible effect on RF1 and only a minor effect on RF2 dissociation (in the absence of RF3). Figure 2 with 3 supplements see all Download asset Open asset Residence times of RF1 and RF2 on PreHC and PostHC. (A–C) smFRET upon addition of RF1-Cy5 or RF1(GAQ)-Cy5 to PreHC or PostHC labeled at protein L11 with Cy3. FRET values (mean ± sd) center at 0.72 ± 0.02 (A), 0.71 ± 0.01 (B), and 0.71 ± 0.01 (C). (D–F) smFRET upon addition of RF2-Cy5 or RF2(GAQ)-Cy5 to PreHC or PostHC labeled at protein L11 with Cy3. FRET values (mean ± sd) center at 0.65 ± 0.03 (D), 0.56 ± 0.05 (E), and 0.70 ± 0.04 (F). Experiments were carried out with catalytic amounts of labeled release factors (10 nM). Individual traces were combined in contour plots. FRET histograms are plotted to the right of the contour plots. In (B,E) the peptide was released using puromycin. In the (C,F) peptide was released by RF1 and RF2, respectively. FRET signals were synchronized to the beginning of the FRET signal. koff is the rate of RF1 or RF2 dissociation. Mean FRET values and rate constants with sd were calculated from three independent data sets. See also Figure 2—figure supplement 1, Figure 2—figure supplement 2, Figure 2—figure supplement 3 and Supplementary file 1. https://doi.org/10.7554/eLife.34252.007 The difference in the dissociation rates of RF1 and RF2 implies that RF1 needs an auxiliary factor, RF3, to help it to dissociate from the ribosome, whereas RF2 may be able to recycle independent of RF3. This notion is consistent with previous reports (Petropoulos et al., 2014; Zavialov et al., 2002) and is further supported by our peptide hydrolysis turnover assay (Figure 2—figure supplement 3A). With catalytic amounts of RF1 in the absence of RF3, that is, when RF1 turnover depends on its intrinsic dissociation rate from the ribosome, termination is essentially blocked, whereas in the presence of RF3 RF1-mediated peptide release is very efficient. In contrast, even catalytic amounts of RF2 are sufficient to complete peptide release from PreHC, although RF3 accelerates the reaction by a factor of 10. Thus, RF3 is essential for RF1, but not for RF2 recycling. Addition of RF3 to PreHC–RF2(GAQ) or PostHC–RF2 complexes makes the complexes more dynamic (Figure 2—figure supplement 3B–E). Our results demonstrate that during canonical termination RF1 and RF2 interact with termination complexes in somewhat different ways, as they have different residence times on the ribosome and respond differently to the presence of RF3. Interaction with RF3–GTP Next, we studied how RF3 affects ribosome dynamics and promotes the dissociation of RF1/RF2. To investigate the effect of RF3 on subunit rotation in the absence of RF1 or RF2, we added saturating concentrations of RF3 to S6/L9-labeled PreHC (Figure 3A, Figure 2—figure supplement 1). Binding of RF3 to PreHC, which in the absence of the factor is in the N state, strongly shifts the equilibrium toward the R state (Figure 3A), that is, RF3 has the opposite effect on subunit rotation than RF1 or RF2. The traces are now highly dynamic and show reversible N to R transitions (Supplementary file 1). This finding seems unexpected as ribosomes with peptidyl-tRNA in the P site favor the N state. However, previous cryo-EM and smFRET studies have indicated that those complexes can in fact adopt the R state (Cornish et al., 2008; Fischer et al., 2010; Ling and Ermolenko, 2015). Thus, RF3 appears to bias spontaneous fluctuations of peptidyl-tRNA in the PreHC, rather than induce a previously disallowed conformation. Figure 3 with 1 supplement see all Download asset Open asset Interaction of RF3–GTP with termination complexes. (A,D) Subunit rotation of S6/L9-labeled PreHC and PostHC* in the presence of excess RF3 (1 µM) with GTP (1 mM). (B,E) Distribution of L1-open and L1-closed states in PreHC and PostHC* labeled at tRNAfMet and protein L1 in the presence of excess RF3 (1 µM) with GTP (1 mM). Smooth red and gray lines show distributions of states without RF3. Color code is the same as in histograms (red, frequent transitions between the states; gray, transitions in less than 20% of traces). (C,F) Contour plots representing the residence time of RF3-Cy5 (10 nM) on PreHC and PostHC* labeled at L11 by Cy3. FRET time courses were synchronized to the beginning of the first FRET event. FRET values (mean ±sd) 0.62 ± 0.02 (C) and 0.64 ± 0.04 (F) are from three independent data sets and plotted to the right of the contour plots. koff is the rate of RF3 dissociation. See also Figure 3—figure supplement 1 and Supplementary file 1. https://doi.org/10.7554/eLife.34252.011 To further characterize the conformational changes of the ribosome induced by RF3, we probed the position of the P-site tRNA relative to protein L1. We used a FRET pair with the donor label at the tRNA (fMet-tRNAfMet-Cy3) and the acceptor label on ribosomal protein L1 (L1-Cy5). The two labels are close together and give a high FRET signal when ribosomes are in the L1-closed state and move apart to give a low FRET signal when ribosomes are in the L1-open state (Fei et al., 2009; Fei et al., 2008; Munro et al., 2010a; Munro et al., 2010b; Munro et al., 2010c; Sternberg et al., 2009). FRET histograms of PreHC in the absence of RF3 are dominated by a low-FRET population (0.32 ± 0.01) and do not show transitions to other states (Figure 3—figure supplement 1A, Supplementary file 1). This indicates that ribosomes are predominantly in the L1-open state with fMet-tRNAfMet in the classic P/P state, in agreement with previous studies (Cornish et al., 2009; Fei et al., 2008; Sternberg et al., 2009). RF3 induces dynamic transitions from the low FRET state to a high FRET state (0.74 ± 0.02), which suggests that ribosomes transiently sample the L1-closed state with fMet-tRNAfMet in a hybrid-like P/E state (Figure 3B). This state is short lived (kclosed→open = 6.0 ± 0.8 s−1; Supplementary file 1). The transition rate is faster than subunit rotation (kR→N = 2.2 ± 0.4 s−1) suggesting that the two processes are not tightly coupled, consistent with the previous smFRET work (Munro et al., 2010a; Wasserman et al., 2016) and cryo-EM reconstructions (Fischer et al., 2010). We then monitored the dissociation of RF3 from PreHC using FRET between RF3–Cy5 and L11–Cy3 (Figure 3C). Labeling of RF3 did not change its catalytic properties (Figure 2—figure supplement 1A,D). The dissociation rate of RF3 from PreHC is koff = 5.9 ± 1.1 s−1 (Figure 3C; Supplementary file 1). Thus, RF3-GTP can bind to PreHC and alter its conformation as shown by the rotation of subunits and movement of the peptidyl-tRNA into a P/E-like state, but the residence time of the factor on the ribosome is short. To test whether the interaction of RF3 with termination complexes depends on peptide release, we then studied the effect of RF3 on subunit rotation of PostHC prepared by puromycin treatment (PostHC*) (Figure 3D–F). S6/L9-labeled PostHC* fluctuates between 0.5 and 0.7 FRET states (Figure 1—figure supplement 1). RF3 binding shifts the distribution toward the 0.5 FRET state, indicating that the R state is stabilized (Figure 3D). The L1–tRNA FRET pair shows an enrichment of the high FRET state corresponding to the P/E state of the tRNA (Figure 3E). While subunits are stabilized in the R state and do not fluctuate toward N state, the L1-tRNA label shows reversible transitions between P/E and P/P conformations (Supplementary file 1). This suggests that also in PostHC subunit rotation and the formation of a hybrid-like state are not tightly coupled. Dissociation of RF3 from PostHC is as rapid as from PreHC, koff = 5.4 ± 1.3 s−1 (Figure 3F, Supplementary file 1

Release factors RF1 and RF2 promote hydrolysis of peptidyl-tRNA during translation termination. The GTPase RF3 promotes recycling of RF1 and RF2. Using single molecule FRET together with ensemble kinetics, we show that ribosome termination complexes that carry two factors, RF1–RF3 or RF2–RF3, are dynamic and fluctuate between non-rotated and rotated states, while each factor alone has its distinct signature on the ribosome dynamics and conformation. Dissociation of RF1 depends on peptide release and the presence of RF3, whereas RF2 can dissociate spontaneously. RF3 binds in the GTP-bound state and can rapidly dissociate without GTP hydrolysis from termination complex carrying RF1. GTP cleavage helps RF3 release from ribosomes stalled in the rotated state in the absence of RF1. Our data suggest how the stochastic assembly of the ribosome–RF1–RF3–GTP complex, peptide release, and ribosome fluctuations promote termination of protein synthesis and recycling of the release factors.

During protein synthesis, the nascent peptide chain traverses the peptide exit tunnel of the ribosome. We monitor the co-translational movement of the nascent peptide using a fluorescent probe attached to the N-terminus of the nascent chain. Due to fluorophore quenching, the time-dependent fluorescence signal emitted by an individual peptide is determined by co-translational events, such as secondary structure formation and peptide-tunnel interactions. To obtain information on these individual events, the measured ensemble fluorescence signal has to be decomposed into position-dependent intensities. Here, we describe mRNA translation as a Markov process with specific fluorescence intensities assigned to the different states of the process. Combining the computed stochastic time evolution of the translation process with a sequence of observed ensemble fluorescence time courses, we compute the unknown position-specific intensities and obtain detailed information on the kinetics of the translation process. In particular, we find that translation of poly(U) mRNAs dramatically slows down at the fourth UUU codon. The method presented here detects subtle differences in the position-specific fluorescence intensities and thus provides a novel approach to study translation kinetics in ensemble experiments.

During protein synthesis, the nascent peptide chain traverses the ribosomal exit tunnel. We monitor this co-translational movement via fluorescent probes attached to the N-termini of the nascent chains. Due to fluorophore quenching, the time-dependent fluorescence signal emitted by an individual peptide is determined by co-translational events, such as secondary structure formation and peptide-tunnel interactions. To obtain information on these local events, the measured total fluorescence signal has to be decomposed into position-dependent intensities. To this end, we describe mRNA translation as a Markov process and assign a specific fluorescence intensity to each position of the ribosome. Combining the stochastic time evolution of the ribosomal position with the observed time course of the total fluorescence, we obtain a system of linear equations. We solve this system for the unknown position-specific intensities. Our results indicate that the N-terminus of poly-phenylalanine experiences major environmental changes during elongation of the first five amino acids.

Proteins are synthesized as linear polymers and have to fold into their native structure to fulfil various functions in the cell. Folding can start co-translationally when the emerging peptide is still attached to the ribosome and is guided by the environment of the polypeptide exit tunnel and the kinetics of translation. Major questions are: When does co-translational folding begin? What is the role of the ribosome in guiding the nascent peptide towards its native structure? How does translation elongation kinetics modulate protein folding? Here we suggest how novel structural and biophysical approaches can help to probe the interplay between the ribosome and the emerging peptide and present future challenges in understanding co-translational folding.

The bacterial signal recognition particle (SRP) is part of the machinery that targets ribosomes synthesizing membrane proteins to membrane-embedded translocons co-translationally. Recognition of nascent membrane proteins occurs by virtue of a hydrophobic signal-anchor sequence (SAS) contained in the nascent chain, usually at the N terminus. Here we use fluorescence-based stopped-flow to monitor SRP-ribosome interactions with actively translating ribosomes while an SRP substrate is synthesized and emerges from the peptide exit tunnel. The kinetic analysis reveals that, at cellular concentrations of ribosomes and SRP, SRP rapidly binds to translating ribosomes prior to the emergence of an SAS and forms an initial complex that rapidly rearranges to a more stable engaged complex. When the growing peptide reaches a length of ∼50 amino acids and the SAS is partially exposed, SRP undergoes another conformational change which further stabilizes the complex and initiates targeting of the translating ribosome to the translocon. These results provide a reconciled view on the timing of high-affinity targeting complex formation, while emphasizing the existence of preceding SRP recruitment steps under conditions of ongoing translation.

Protein domains can fold into stable tertiary structures while they are synthesized on the ribosome. We used a high-performance, reconstituted in vitro translation system to investigate the folding of a small five-helix protein domain—the N-terminal domain of Escherichia coli N5-glutamine methyltransferase HemK—in real time. Our observations show that cotranslational folding of the protein, which folds autonomously and rapidly in solution, proceeds through a compact, non-native conformation that forms within the peptide tunnel of the ribosome. The compact state rearranges into a native-like structure immediately after the full domain sequence has emerged from the ribosome. Both folding transitions are rate-limited by translation, allowing for quasi-equilibrium sampling of the conformational space restricted by the ribosome. Cotranslational folding may be typical of small, intrinsically rapidly folding protein domains.

Nascent proteins emerging from translating ribosomes in bacteria are screened by ribosome‐associated protein biogenesis factors (RPBs), among them the signal recognition particle (SRP), trigger factor (TF), peptide deformylase (PDF), and methionine aminopeptidase (MAP). Whereas SRP and TF can bind to translating ribosomes simultaneously, the time of action for the modifying enzymes, PDF and MAP, is not clear. Here we examine the RPB choreography by monitoring FRET between labeled ribosomes and either TF or SRP. The competition binding analysis reveals how the RPB binding pattern changes depending on the state of the ribosome.We observe that TF, like SRP, rapidly scans all ribosomes until it is stabilized on ribosomes presenting TF‐specific nascent chains. PDF and TF compete for binding to those complexes, whereas MAP can bind together with TF. The same competition pattern we observe with SRP and PDF or SRP and MAP on ribosomes exposing an SRP‐specific nascent chain. These results suggest that PDF and TF act sequentially in that PDF deformylates short nascent chains prior to stable TF binding to ribosomes carrying longer nascent chains. By contrast, competition between PDF and SRP, which is recruited to ribosomes carrying rather short nascent chains, may interfere with N‐terminal processing of nascent membrane proteins that enter the membrane co‐translationally in an SRP‐dependent fashion.

Nascent proteins emerging from translating ribosomes in bacteria are screened by a number of ribosome-associated protein biogenesis factors, among them the chaperone trigger factor (TF), the signal recognition particle (SRP) that targets ribosomes synthesizing membrane proteins to the membrane and the modifying enzymes, peptide deformylase (PDF) and methionine aminopeptidase (MAP). Here, we examine the interplay between these factors both kinetically and at equilibrium. TF rapidly scans the ribosomes until it is stabilized on ribosomes presenting TF-specific nascent chains. SRP binding to those complexes is strongly impaired. Thus, TF in effect prevents SRP binding to the majority of ribosomes, except those presenting SRP-specific signal sequences, explaining how the small amount of SRP in the cell can be effective in membrane targeting. PDF and MAP do not interfere with TF or SRP binding to translating ribosomes, indicating that nascent-chain processing can take place before or in parallel with TF or SRP binding.

Elongation factor G (EF-G) promotes the movement of two tRNAs and the mRNA through the ribosome in each cycle of peptide elongation. During translocation, the tRNAs transiently occupy intermediate positions on both small (30S) and large (50S) ribosomal subunits. How EF-G and GTP hydrolysis control these movements is still unclear. We used fluorescence labels that specifically monitor movements on either 30S or 50S subunits in combination with EF-G mutants and translocation-specific antibiotics to investigate timing and energetics of translocation. We show that EF-G-GTP facilitates synchronous movements of peptidyl-tRNA on the two subunits into an early post-translocation state, which resembles a chimeric state identified by structural studies. EF-G binding without GTP hydrolysis promotes only partial tRNA movement on the 50S subunit. However, rapid 30S translocation and the concomitant completion of 50S translocation require GTP hydrolysis and a functional domain 4 of EF-G. Our results reveal two distinct modes for utilizing the energy of EF-G binding and GTP hydrolysis and suggest that coupling of GTP hydrolysis to translocation is mediated through rearrangements of the 30S subunit.

The ribosome is a molecular machine responsible for protein synthesis and a major target for small-molecule inhibitors. Compared to the wealth of structural information available on ribosome-targeting antibiotics in bacteria, our understanding of the binding mode of ribosome inhibitors in eukaryotes is currently limited. Here we used X-ray crystallography to determine 16 high-resolution structures of 80S ribosomes from Saccharomyces cerevisiae in complexes with 12 eukaryote-specific and 4 broad-spectrum inhibitors. All inhibitors were found associated with messenger RNA and transfer RNA binding sites. In combination with kinetic experiments, the structures suggest a model for the action of cycloheximide and lactimidomycin, which explains why lactimidomycin, the larger compound, specifically targets the first elongation cycle. The study defines common principles of targeting and resistance, provides insights into translation inhibitor mode of action and reveals the structural determinants responsible for species selectivity which could guide future drug development.

The translocation of tRNAs through the ribosome proceeds through numerous small steps in which tRNAs gradually shift their positions on the small and large ribosomal subunits. The most urgent questions are: (i) whether these intermediates are important; (ii) how the ribosomal translocase, the GTPase elongation factor G (EF-G), promotes directed movement; and (iii) how the energy of GTP hydrolysis is coupled to movement. In the light of recent advances in biophysical and structural studies, we argue that intermediate states of translocation are snapshots of dynamic fluctuations that guide the movement. In contrast to current models of stepwise translocation, kinetic evidence shows that the tRNAs move synchronously on the two ribosomal subunits in a rapid reaction orchestrated by EF-G and GTP hydrolysis. EF-G combines the energy regimes of a GTPase and a motor protein and facilitates tRNA movement by a combination of directed Brownian ratchet and power stroke mechanisms.

Elongation factor G (EF-G) is a GTPase that catalyzes tRNA and mRNA translocation during the elongation cycle of protein synthesis. The GTP-bound state of the factor on the ribosome has been studied mainly with non-hydrolyzable analogs of GTP, which led to controversial conclusions about the role of GTP hydrolysis in translocation. Here we describe a mutant of EF-G in which the catalytic His91 is replaced with Ala. The mutant EF-G does not hydrolyze GTP, but binds GTP with unchanged affinity, allowing us to study the function of the authentic GTP-bound form of EF-G in translocation. Utilizing fluorescent reporter groups attached to the tRNAs, mRNA, and the ribosome we compile the velocity map of translocation seen from different perspectives. The data suggest that GTP hydrolysis accelerates translocation up to 30-fold and facilitates conformational rearrangements of both 30S subunit (presumably the backward rotation of the 30S head) and EF-G that lead to the dissociation of the factor. Thus, EF-G combines the energy regime characteristic for motor proteins, accelerating movement by a conformational change induced by GTP hydrolysis, with that of a switch GTPase, which upon Pi release switches the conformations of EF-G and the ribosome to low affinity, allowing the dissociation of the factor.

Inner‐membrane proteins in Escherichia coli are synthesized on ribosomes that are bound to the translocation pore (translocon) and enter the membrane co‐translationally. Ribosomes are targeted to the translocon by the signal recognition particle (SRP) and the SRP receptor, FtsY. Targeting is initiated when SRP is recruited to ribosomes that expose a signal‐anchor sequence (SAS). Here we report the kinetics of the interaction of SRP with ribosomes in different functional states and with FtsY, as followed by stopped‐flow monitoring FRET between fluorophores in SRP and the ribosome or FtsY. Rapid SRP binding to the ribosome is followed by two rearrangements the forward rate constants of which are independent of the functional state of the ribosomes. Conversely, ribosome‐SRP complex dissociation is very slow with ribosomes exposing an SAS, whereas the complex with non‐translating ribosomes dissociates rapidly. SAS‐induced stabilization of SRP on the ribosome is correlated with an acceleration of receptor binding. Thus, SRP can rapidly scan all ribosomes until it encounters a ribosome exposing an SAS, where SRP binding is strongly stabilized and membrane targeting is promoted by accelerating targeting complex formation with FtsY and, presumably, the translocon. The formation of the targeting complex is correlated with the lengthening of the nascent peptide within the exit tunnel of the ribosome.

Ribosomes synthesizing inner membrane proteins in Escherichia coli are targeted to the membrane by the signal recognition particle (SRP) pathway. By rapid kinetic analysis we show that after initial binding to the ribosome, SRP undergoes dynamic fluctuations in search of additional interactions. Non-translating ribosomes, or ribosomes synthesizing non-membrane proteins, do not provide these contacts, allowing SRPs to dissociate rapidly. A nascent peptide in the exit tunnel stabilizes SRPs in a standby state. Binding to the emerging signal-anchor sequence (SAS) of a nascent membrane protein halts the fluctuations of SRP, resulting in complex stabilization and recruitment of the SRP receptor. We propose a kinetic model where SRP rapidly scans all ribosomes until it encounters a ribosome exposing an SAS. Binding to the SAS switches SRP into the targeting mode, in which dissociation is slow and docking of the SRP receptor is accelerated.