Structure and dynamics of a processive Brownian motor: the translating ribosome

Abstract

There is mounting evidence indicating that protein synthesis is driven and regulated by mechanisms that direct stochastic, large-scale conformational fluctuations of the translational apparatus. This mechanistic paradigm implies that a free-energy landscape governs the conformational states that are accessible to and sampled by the translating ribosome. This scenario presents interdependent opportunities and challenges for structural and dynamic studies of protein synthesis. Indeed, the synergism between cryogenic electron microscopic and X-ray crystallographic structural studies, on the one hand, and single-molecule fluorescence resonance energy transfer (smFRET) dynamic studies, on the other, is emerging as a powerful means for investigating the complex free-energy landscape of the translating ribosome and uncovering the mechanisms that direct the stochastic conformational fluctuations of the translational machinery. In this review, we highlight the principal insights obtained from cryogenic electron microscopic, X-ray crystallographic, and smFRET studies of the elongation stage of protein synthesis and outline the emerging themes, questions, and challenges that lie ahead in mechanistic studies of translation.

Figure 1

(a) Structure and dynamic features of the ribosome. Cryogenic electron microscopic map of the 70S ribosome, the 30S subunit, and the 50S subunit. The 30S and 50S subunits are shown with their intersubunit space facing the reader. The P- and A-site tRNAs are depicted in green and magenta, respectively, and their positions are denoted on the 70S ribosome. Major landmarks and mobile elements of the 30S subunit are the head (h), shoulder (s), platform (p), and spur (sp). The location of the decoding center (DC) active site is also denoted. Major landmarks and mobile elements of the 50S subunit are the L1 stalk (L1) and the L7/L12 stalk (L7/L12). The locations of the GTPase-associated center (GAC) and peptidyltransferase center (PTC) active sites are also denoted. The locations of all donor (D) and acceptor (A) fluorophore pairs that have thus far been used in single-molecule fluorescence resonance energy transfer investigations of translation elongation are labeled in green (D) and red (A). Details regarding each A-D pair are given in Table 1. (b) The elongation cycle of protein synthesis. The main steps of the translation elongation cycle, (i) aminoacyl-tRNA (aa-tRNA) selection, (ii) peptidyl transfer, and (iii) messenger RNA (mRNA)-tRNA translocation, are shown. The E, P, and A tRNA binding sites run vertically along both subunits. Further details regarding the mechanism of aa-tRNA selection and mRNA-tRNA translocation are provided in the captions for Figures 2 and 3, respectively. Abbreviations: EF, elongation factor.

Figure 2

Distinct states and reversible/irreversible steps of the decoding and peptidyl transfer processes, and corresponding cryogenic electron microscopy (cryo-EM) maps, where available. In this schematic, tRNAs are colored according to their positions in the canonical (A, P, E) scheme, consistent with color choices in previous work [e.g. Reference (63)]. State 0: The posttranslocational state in which the A site is unoccupied, the P site contains a peptidyl-tRNA bound in the classical P/P (denoting the 30S P/50S P sites, respectively) configuration, and the E site contains a deacylated tRNA bound in the classical E/E configuration and in direct contact with the open L1 stalk. (Note: The E-site tRNA contacts the L1 stalk through its central fold, or elbow, domain.) Cryo-EM map I from Valle and coworkers (63). Step 0 → 1 (reversible; k₁/k₋₁): The binding of elongation factior Tu (EF-Tu) in a ternary complex with aminoacyl-tRNA (aa-tRNA) and GTP to the ribosome via the L7/L12 stalk. State 1: The same as state 0 but with ternary complex bound to ribosome. Step 1 → 2 (reversible; k₂/k₋₂): The probing of the mRNA codon by the aa-tRNA anticodon at the decoding center (DC). State 2: The same as state 1 but the aa-tRNA anticodon is engaged with the codon at the DC. Step 2 → 3 (reversible; k₃/k₋₃): The cognate and a fraction of near-cognate ternary complexes are bound sufficiently long to induce GTPase activation of EF-Tu. Step 2 → 3 (irreversible; k_3′): The noncognate and a fraction of near-cognate ternary complexes are rejected as their binding to the ribosome fails to stabilize. State 3: The same as state 2 but with EF-Tu activated for GTP hydrolysis. Cryo-EM map II: The use of guanylyl iminodiphosphate prevents GTP hydrolysis (119; J. Sengupta, O. Kristensen, F. Fabiola, H. Gao, M. Valle, et al., in preparation). Step 3 → 4: (reversible; k₄/k₋₄) GTP hydrolysis on EF-Tu. State 4: The same as state 3 but with EF-Tu bound in the GDP-P_i state. Cryo-EM map III: After GTP hydrolysis, kirromycin prevents conformational change of EF-Tu and locks the ternary complex in the A/T configuration (64). Step 4 → 5 (irreversible; k₅): The departure of P_i. State 5: The same as state 4 but with EF-Tu bound in the GDP only state. Step 5 → 6 (irreversible; k₆): The conformational change of EF-Tu, departure of EF-Tu·GDP, and accommodation of cognate aa-tRNA. (irreversible; k_6′): The conformational change of EF-Tu, departure of EF-Tu·GDP, and departure of near-cognate aa-tRNA. State 6: The same as state 5, but with aa-tRNA accommodated in the classical A/A configuration within the A site. Step 6 → 7 (irreversible; k₇): The departure of E-site tRNA and peptidyl transfer. (Note that the precise timing of the E-site tRNA departure has not been established, so this placement is tentative.) State 7: The macrostate I form of the pretranslocational complex is the same as state 6 but the nascent polypeptide is now covalently linked to the A-site tRNA, whereas the P-site tRNA is deacylated, and the E site is unoccupied. The ribosome is in its nonrotated conformation, the tRNAs are in their classical A/A and P/P positions, and the L1 stalk is in its open conformation. Cryo-EM map IV from Agirrezabala and coworkers (68).

Figure 3

Distinct states and reversible/irreversible steps of the translocation process and corresponding cryo-EM maps, where available. In this schematic, unlike Figure 2, colors mark individual tRNAs, so that the steps of their translocation can be followed. State 7: The macrostate I (MS-I) form of the pretranslocational (PRE) complex is the same as state 7 in Figure 2. The A site contains the newly formed peptidyl-tRNA, the P site contains a deacylated tRNA, and the E site is unoccupied. The ribosome is in its nonrotated conformation, the tRNAs are in their classical A/A (denoting the 30S A/50S A sites, respectively) and P/P positions, and the L1 stalk is in its open conformation. Cryo-EM map IV from Agirrezabala and coworkers (68). Step 7 → 8 (reversible; k₈/k₋₈): The rearrangement of MS-I into an intermediate state of ratcheting (44). State 8: This is the same as state 7 but with the PRE complex in an intermediate state of ratcheting consisting of a ribosome in a semirotated state and tRNAs in an intermediate classical A/A, hybrid P/E configuration that lies somewhere between the classical A/A and P/P configuration and the hybrid A/P and P/E configuration. The L1 stalk is in a “closed” position, where it forms a direct intermolecular contact with the hybrid P/E tRNA (44, 156). Step 8 → 9 (reversible; k₉/k₋₉): The rearrangement of the intermediate state of ratcheting to macrostate II (MS-II) (44). State 9: The MS-II form of the PRE complex is the same as in state 8 but with the ribosome in the rotated state, tRNAs in hybrid A/P and P/E configurations, and the L1 stalk in a closed conformation where it directly contacts the hybrid P/E tRNA elbow. Cryo-EM map V from Agirrezabala and coworkers (68). Note: State 9 can alternatively be reached directly from state 7, bypassing the intermediate state 8. Step 9 → 10 (reversible; k₁₀/k₋₁₀): The binding of EF-G in the GTP state. State 10: The same as in state 9 but with EF-G bound in GTP state, stabilizing MS-II. Cryo-EM map VI from Valle and coworkers (63). Note that EF-G in the presence of guanylyl iminodiphosphate stably binds only to ribosomes with an unoccupied A site (63, 68). Step 10 → 11 (reversible; k₁₁/k₋₁₁): GTP hydrolysis on EF-G. State 11: The same as state 10 but with EF-G bound in the GDP-P_i state. Step 11 → 0 (irreversible; k₀): The ribosome returns to the nonrotated position, the newly formed peptidyl-tRNA and the newly deacylated tRNA move into the classical P/P and E/E configurations, respectively, and the L1 stalk moves into the open position. EF-G·GDP and P_i depart from the ribosome. State 0: The posttranslocation complex. The same as in state 11 but with the ribosome in the nonrotated position, the newly formed peptidyl-tRNA and the newly deacylated tRNA in classical P/P and E/E configurations, respectively, and the L1 stalk in the open position. Cryo-EM map I from Valle and coworkers (63). [Note that using fusidic acid, EF-G has also been trapped on the ribosome in the GDP state, with the ribosome in the nonrotated position, the newly formed peptidyl-tRNA and the newly deacylated tRNA in classical P/P and E/E configurations, respectively, and the L1 stalk in the open position as depicted in state 0 (see References and 86).] For clarity, we have refrained from adding a panel depicting this configuration.

Figure 4

A heuristic schematic of the complex free-energy landscape of the elongation cycle, including macrostate (MS)-I and MS-II of the pretranslocational (PRE) ribosomal complex. Conformational changes of the ribosomal complex can occur along either the reaction coordinate or the conformational space axes. Conformational changes that take place along the reaction coordinate axis correspond to the rearrangements of the ribosomal complex that facilitate the elongation reaction that will ultimately transform posttranslocation (POST)-1 into POST-2. Conformational changes along the conformational space axis, by contrast, correspond to fluctuations among the ensemble of conformers that exist at all points along the reaction coordinate, leading to the availability of numerous parallel reaction pathways, which are the hallmark of a complex free-energy landscape. The energetic barriers separating POST-1 from the MS-I state of the PRE complex and the MS-II state of the PRE complex from POST-2 are large enough such that overcoming these barriers generally requires the energy released from GTP hydrolysis by elongation factor Tu and/or peptidyl transfer (for POST-1 → MS-I transitions) and GTP hydrolysis by EF-G (for MS-II → POST-2 transitions). The energetic barrier separating MS-I from MS-II, however, is small enough such that stochastic, thermally driven fluctuations between MS-I and MS-II are permitted. In addition, the ruggedness of the landscape strongly suggests that the valleys defining POST-1, MS-I, MS-II, and POST-2 are themselves composed of a multiplicity of smaller valleys separated by barriers even smaller than that separating MS-I from MS-II. Thus, POST-1, MS-I, MS-II, and POST-2 are each expected to be composed of an ensemble of conformations, with the population of any one member of the ensemble depending on the exact depth of its valley and heights of the barriers separating it from its neighbors. As experimentally demonstrated in Figure 5, the depth of the valleys within POST-1, MS-I, MS-II, and POST-2, as well as the depths of the POST-1, MS-I, MS-II, and POST-2 valleys themselves, are sensitive functions of environmental conditions (e.g., substrate, cofactor, or allosteric effector binding) or the dissociation of reaction products. The circled numbers listed underneath the POST-1, MS-I, MS-II, and POST-2 valleys refer to the equivalently labeled POST, MS-I, and MS-II complexes depicted in Figure 3. Abbreviation: ΔG^‡, free energy of activation.

Figure 5

Modulating the free-energy landscape of the macrostate (MS)-I ⇄ MS-II equilibrium. (a) An equilibrium single-molecule fluorescence resonance energy (smFRET) versus time trajectory for a pretranslocation (PRE) ribosomal complex sample containing a donor-labeled P-site tRNA and an acceptor-labeled L1 stalk (41), (D₅/A₅, Figure 1a and Table 1) (top). The smFRET trajectory is calculated using I_Cy5/(I_Cy3+I_Cy5), where I_Cy3 is the raw emission intensity of the Cy3 donor fluorophore, and I_Cy5 is the raw emission intensity of the Cy5 acceptor fluorophore. In this labeling scheme, disruption of the L1 stalk-P/E (denoting the 30S P/50S E sites, respectively) tRNA contact (L1 ◦ tRNA, MS-I) generates a FRET value centered at 0.16 FRET, whereas formation of the L1 stalk-P/E tRNA contact (L1 • tRNA, MS-II) generates a FRET value centered at 0.76 FRET. Analysis of the dwell time spent in the MS-I state prior to transitioning to the MS-II state provides the average rate constant governing MS-I → MS-II transitions, and the analogous analysis for the dwell time spent in MS-II provides the average rate constant governing MS-II → MS-I transitions. These average rate constants can be converted to free energies of activation, ΔG^‡, for the two transitions using the equation ΔG^‡ = −RT ln(hk/k_BT), where R is the universal gas constant (8.314 J K⁻¹ mol⁻¹), T is the temperature (in K), h is Planck’s constant (6.626 × 10⁻³⁴ J s), k is the rate constant (in s), and k_B is Boltzmann’s constant (1.381 × 10⁻²³ J K⁻¹). The smFRET trajectory shown here, used as a point of reference, was recorded using a PRE complex containing a phenylalanine-specific tRNA (tRNA^Phe) in the P site, an unoccupied A site, and no addition of elongation factor G·guanylyl iminodiphosphate (EF-G·GDPNP) in a buffer containing 15 mM Mg²⁺. A contour plot of the time evolution of population FRET (bottom) is generated by superimposing the first five of numerous individual smFRET trajectories, binning the data into 20 FRET bins and 30 time bins, and normalizing the resulting data to the most populated bin in the plot. N indicates the number of trajectories used to generate the contour plot. (b) Inspection of the smFRET trajectories (top) and contour plots of the time evolution of population FRET (bottom) reveal that lowering the [Mg²⁺] from 15 mM to 5 mM (left), replacing the P-site tRNA^Phe with formylmethionine specific tRNA (tRNA^fMet) (center), and binding of EF-G·GDPNP (right) all markedly alter the free-energy landscape of the MS-I ⇄ MS-II equilibrium, changing the average rates and corresponding ΔG^‡s for the MS-I → MS-II and MS-II → MS-I transitions and thereby modulating the observed MS-I and MS-II equilibrium populations. (c) Two-dimensional free-energy profile of the MS-I ⇄ MS-II equilibrium in which the conformational space coordinate has been averaged to a single average conformer. The plot summarizes how the ΔG^‡ for the MS-I → MS-II and MS-II → MS-I transitions is altered by changes in [Mg²⁺], P-site tRNA identity, and EF-G·GDPNP binding.