The data deposited here are part of a study to determine the kinetic features of the elongation cycle of protein synthesis using an in vitro Eukaryotic translation system and single-molecule fluorescence resonance energy transfer (smFRET).
This paper examines the oxidation reaction of tert-amyl methyl ether (TAME), an oxygenated fuel additive, with chlorine radical initiators in the presence of oxygen. Data are collected at 298, 550, and 700 K. Reaction intermediates and products are probed by a multiplexed chemical kinetics synchrotron photoionization mass spectrometer (SPIMS) and characterized on the basis of the mass-to-charge ratio, ionization energy, and photoionization spectra. Branching fractions of primary products are obtained at the different reaction temperatures. CBS-QB3 computations are also carried out to study the potential energy surface of the investigated reactions to validate detected primary products.
A complex interplay between mRNA translation and cellular respiration has been recently unveiled, but its regulation in humans is poorly characterized in either health or disease. Cancer cells radically reshape both biosynthetic and bioenergetic pathways to sustain their aberrant growth rates. In this regard, we have shown that the molecular chaperone TRAP1 not only regulates the activity of respiratory complexes, behaving alternatively as an oncogene or a tumor suppressor, but also plays a concomitant moonlighting function in mRNA translation regulation. Herein we identify the molecular mechanisms involved, demonstrating that TRAP1: i) binds both mitochondrial and cytosolic ribosomes as well as translation elongation factors, ii) slows down translation elongation rate, and iii) favors localized translation in the proximity of mitochondria. We also provide evidence that TRAP1 is coexpressed in human tissues with the mitochondrial translational machinery, which is responsible for the synthesis of respiratory complex proteins. Altogether, our results show an unprecedented level of complexity in the regulation of cancer cell metabolism, strongly suggesting the existence of a tight feedback loop between protein synthesis and energy metabolism, based on the demonstration that a single molecular chaperone plays a role in both mitochondrial and cytosolic translation, as well as in mitochondrial respiration.
Inositol phosphates (IPs) and inositol pyrophosphate play critical roles in many biological processes as signaling molecules in pathways responsible for cellular functions involved in growth and maintenance. The biosynthesis of IPs is carried out by a family of inositol phosphate kinases. In mammals, Inositol tetrakisphosphate kinase-1 (ITPK1) phosphorylates inositol-1,3,4-trisphosphate (Ins(1,3,4)P3) and inositol-3,4,5,6-tetrakisphosphate (IP4), generating inositol-1,3,4,5,6-pentakisphosphate (IP5), which can be further phosphorylated to become inositol hexakisphosphate (IP6). ITPK1 also possesses phosphatase activity that can convert IP5 back to IP4; therefore, ITPK1 may serve as a regulatory step in IP6 production. IP6 utilization has been implicated in processes fundamental to cellular sustainability that are severely perturbed in many disease states including RNA editing, DNA repair, chromatin structure organization, and ubiquitin ligation. Therefore, ITPK1, with no known inhibitors in the literature, is a potential molecular target for modulating important processes in several human diseases. By independently coupling ITPK1 phosphatase and kinase activities to luciferase activity, we have developed and used biochemical high throughput assays to discover eight ITPK1 inhibitors. Further analysis revealed that three of these leads inhibit ITPK1 in an ATP-competitive manner, with low micromolar to nanomolar affinities. We further demonstrate that the most potent ITPK1 inhibitor can regulate cellular ITPK1 activity. We determined the crystal structure of ITPK1 in complex with this inhibitor at a resolution of 2.25 Å. This work provides insight into the design of potential next-generation inhibitors.
The branched C5 alcohol isopentanol (3-methylbutan-1-ol) has shown promise as a potential biofuel both because of new advanced biochemical routes for its production and because of its combustion characteristics, in particular as a fuel for homogeneous-charge compression ignition (HCCI) or related strategies. In the present work, the fundamental autoignition chemistry of isopentanol is investigated by using the technique of pulsed-photolytic Cl-initiated oxidation and by analyzing the reacting mixture by time-resolved tunable synchrotron photoionization mass spectrometry in low-pressure (8 Torr) experiments in the 550–750 K temperature range. The mass-spectrometric experiments reveal a rich chemistry for the initial steps of isopentanol oxidation and give new insight into the low-temperature oxidation mechanism of medium-chain alcohols. Formation of isopentanal (3-methylbutanal) and unsaturated alcohols (including enols) associated with HO2 production was observed. Cyclic ether channels are not observed, although such channels dominate OH formation in alkane oxidation. Rather, products are observed that correspond to formation of OHvia β-C–C bond fission pathways of QOOH species derived from β- and γ-hydroxyisopentylperoxy (RO2) radicals. In these pathways, internal hydrogen abstraction in the RO2 ⇄ QOOH isomerization reaction takes place from either the –OH group or the C–H bond in α-position to the –OH group. These pathways should be broadly characteristic for longer-chain alcohol oxidation. Isomer-resolved branching ratios are deduced, showing evolution of the main products from 550 to 750 K, which can be qualitatively explained by the dominance of RO2 chemistry at lower temperature and hydroxyisopentyl decomposition at higher temperature.
Significance Nonsense mutations giving rise to premature stop codons (PSCs) cause many diseases, creating the need to develop safe and effective translational read-through–inducing drugs (TRIDs). The current best-characterized TRIDs are ataluren and aminoglycosides. Only ataluren has been approved for clinical use, albeit in a limited context. Here, we provide rate measurements of elementary steps in a single eukaryotic translation elongation cycle, allowing us to demonstrate that ataluren and the aminoglycoside G418 employ orthogonal mechanisms in stimulating PSC read-through: ataluren by inhibiting release factor-dependent termination of protein synthesis and G418 by increasing functional near-cognate transfer RNA mispairing, which permits continuation of synthesis. We conclude that development of new TRIDs combatting PSC diseases should prioritize those directed toward inhibiting termination.
Article Figures and data Abstract eLife digest Introduction Results Discussion Materials and methods References Decision letter Author response Article and author information Metrics Abstract The intergenic IRES of Cricket Paralysis Virus (CrPV-IRES) forms a tight complex with 80S ribosomes capable of initiating the cell-free synthesis of complete proteins in the absence of initiation factors. Such synthesis raises the question of what effect the necessary IRES dissociation from the tRNA binding sites, and ultimately from all of the ribosome, has on the rates of initial peptide elongation steps as nascent peptide is formed. Here we report the first results measuring rates of reaction for the initial cycles of IRES-dependent elongation. Our results demonstrate that 1) the first two cycles of elongation proceed much more slowly than subsequent cycles, 2) these reduced rates arise from slow pseudo-translocation and translocation steps, and 3) the retarding effect of ribosome-bound IRES on protein synthesis is largely overcome following translocation of tripeptidyl-tRNA. Our results also provide a straightforward approach to detailed mechanistic characterization of many aspects of eukaryotic polypeptide elongation. https://doi.org/10.7554/eLife.13429.001 eLife digest Inside cells, machines called ribosomes make proteins using instructions carried by molecules of messenger RNA (or mRNA). The ribosomes bind to the mRNA and then move along it to assemble the proteins in a process called translation. The first step of translation – when the ribosome binds to the mRNA – is known as initiation. In human and other eukaryotic cells, initiation mainly occurs through a mechanism that requires many proteins called initiation factors to recruit the ribosome to a cap structure formed at one end of the mRNA. When viruses infect cells, they hijack the ribosomes of the host cell to produce large quantities of viral proteins. However, unlike their host cells, many viruses use a different pathway to initiate translation of their mRNAs. The mRNAs of these viruses have regions known an internal ribosome entry sites (IRESs) that host cell ribosomes can bind to instead. After initiation, the ribosome progressively assembles the building blocks of proteins (amino acids) into a chain until the new protein is complete. Molecules called transfer RNAs bind to individual amino acids and bring them to the ribosome. Previous research has shown that, prior to initiation, IRESs on Cricket Paralysis Virus mRNAs bind to the ribosome and occupy sites where transfer RNAs would normally bind. However, it was not clear how this affects the elongation process. Zhang et al. now address this question using a cell-free system that allowed them to recreate and observe translation outside of the normal cell environment. Zhang et al. found that the binding of an IRES to a ribosome slows down the early steps of elongation. A likely explanation for this is that the IRES elements have to be displaced from the ribosome before the incoming transfer RNAs can occupy the three tRNA sites. However, as elongation progresses, the effects of the IRES elements are overcome and the pace of elongation increases significantly. Zhang et al.’s findings provide a convenient approach that could be used for future studies of elongation. This approach could also help researchers find out how abnormalities in translation contribute to human diseases, including muscle-wasting disorders. https://doi.org/10.7554/eLife.13429.002 Introduction Initiation of protein synthesis in eukaryotic cells proceeds via two well-established pathways. The cap-dependent pathway involves recognition of 7-methyl-guanosine at the 5’-terminus of mRNA by a preinitiation complex of 40S ribosomal subunit and a host of initiation factors prior to a scanning step that results in initiator aminoacyl-tRNA(aa-tRNA) pairing with a cognate start codon, followed by 60S binding to form the 80S initiator complex (Jackson et al., 2010; Aitken and Lorsch, 2012). The second pathway involves binding of the ribosome to an internal ribosome entry site (IRES), a structure that is present in many virus-encoded mRNAs, as well as in some cellular mRNAs (Fitzgerald and Semmler, 2009). Initiation of protein synthesis from an 80S·IRES complex can take place in the absence of some or even all of the initiation factors required in the cap-dependent pathway (Filbin and Kieft, 2009), depending on the IRES source. The intergenic IRES of Cricket Paralysis Virus (CrPV-IRES) forms a complex with 80S ribosomes that is capable of initiating the synthesis of complete proteins in cell-free assays completely lacking initiation factors (Jan et al. 2003; Pestova and Hillen, 2003). More recently, high resolution structural studieshave shown that, prior to polypeptide chain initiation, the closely related Dicistroviridae IRES structures from CrPV (Fernandez et al., 2014; Muhs et al., 2015) and Taura syndrome virus (Koh et al., 2014) occupy all three tRNA binding sites (E, P, and A) on the ribosome, with the protein coding region beginning immediately downstream from IRES segment occupying the A-site (Figure 1). Figure 1 with 1 supplement see all Download asset Open asset Structure of CrPV-IRES bound to the 80S ribosome superposed on A, P, and E tRNA binding sites. The position of the first codon is indicated. Adapted from Fernandez et al. (2014). https://doi.org/10.7554/eLife.13429.003 CrPV-IRES binds with high affinity (Kd ~ 10 nM) to the 80S ribosome (Jang and Jan, 2010), raising the question of what effect the necessary IRES dissociation from the tRNA binding sites, and ultimately from all of the ribosome as well, has on the rates of initial peptide elongation steps as nascent peptide is formed (Muhs et al., 2015). Since prior to the work reported in this paper nothing had been published concerning the rate of initial oligopeptide synthesis by an 80S·CrPV-IRES complex, it has been unclear whether there is a retarding effect due to the presence of IRES on the ribosome, and, if so, how many cycles of peptide elongation are required before the ribosome begins to form peptide bonds at a higher rate. In considering this question, we make use of the simplified 12-step scheme of initial tetrapeptide synthesis shown in Figure 2, which provides a useful framework for presenting the results described in this paper. In this scheme Steps 1–3 show the reactions required for initial binding of the first tRNA to the A site followed by translocation to the P-site, and reactions 4–6, 7–9, and 10–12 represent three elongation cycles, ending with P-site bound tetrapeptidyl-tRNA, completing the third cycle of polypeptide synthesis. This model makes the reasonable assumption that binding of successive aminoacyl-tRNAs (aa-tRNAs) cognate to the mRNA requires the progressive removal of IRES structures from each of the tRNA binding sites, such that translocation of dipeptidyl tRNA to the P-site (structure 7) requires removal of the IRES from the last of the three tRNA binding sites. In the work reported below, we demonstrate first, that the initial elongation steps are indeed quite slow and are limited by the translocation step of the elongation cycle, and second, that the rate of elongation accelerates following translocation of tripeptidyl-tRNA to the P-site. Figure 2 Download asset Open asset Proposed scheme for initial tetrapeptide synthesis on CrPV IRES-programmed ribosomes. This simplified scheme neglects the several substeps, including GTP hydrolysis, Pi release, and elongation factor release, that accompany both productive binding of ternary complex to the ribosome (Steps 2, 4, 7, 10) and tRNA translocation (Steps 3, 6, 9, 12). https://doi.org/10.7554/eLife.13429.005 Results In our experiments, eukaryotic ribosomes are prepared from shrimp cysts (Iwasaki and Kaziro, 1979), elongation factors are prepared from yeast, and charged tRNAs are prepared from yeast and E. coli. In addition, the peptide coding sequence attached to the 3’-end of the CrPV-IRES (Figure 1) has been mutated for ease in detection of peptide synthesis via 35S-Met incorporation. In all such mutants the initial wt-codon triplet GCU encoding Ala has been replaced by UUC, encoding Phe, a change that has little effect on the expression of active luciferase in a cell-free protein synthesis assay (Figure 1—figure supplement 1). The initial coding sequences of the mutants used in this work are presented in Supplementary file 1. Collectively, they allow monitoring of the rates of PheMet, PheLysMet, PheValLysMet and PheLyValArgGlnTrpLeuMet synthesis. In presenting the results below, Steps 1–12 and structures 1 – 13 are as described in the scheme for initial tetrapeptide synthesis proposed in Figure 2. Values of t1/2 for Steps 1–12, determined as described, are summarized in Table 2. Rates of Phe-TC binding to the 80S·IRES complex: Steps 1–3, structures 1–4 We previously have utilized two assays to measure binding of the ternary complex Phe-tRNAPhe·eEF1A·GTP (Phe-TC) to the 80S·CrPV-IRES (80S·IRES) complex (Ruehle et al., 2015). The increase in proflavin-labeled Phe-tRNAPhe fluorescence anisotropy measures binding to either the A- or P-site (structures 3 and 4, respectively, Figure 2). [3H]-Phe-tRNAPhe cosedimentation with the 80S·IRES complex measures accumulation of 4 only, since A-site binding is too labile to survive the ultracentrifugation step (Yamamoto et al., 2007). In Figure 3 we present time-resolved application of the anisotropy assay that allows us to measure the rates of Phe-TC binding to form Structure 3 from 1. These resultswere fit to the scheme shown in Figure 2, giving values for k1, k-1, and k2 in both the presence and absence of eEF2·GTP that are summarized in Table 1. In the absence of eEF2 (blue trace), the equilibrium position of Step 1, a so-called pseudo-translocation step (Muhs et al., 2015) in which the IRES vacates the A-site, favors Structure 1 over Structure 2 by approximately 20-fold, consistent with recent structural studies (Fernandez et al., 2014; Koh et al., 2014; Muhs et al., 2015). Phe-TC binds to Structure 2 yielding Structure 3, in a process where the rate-limiting step is the conversion of Structure 1 to Structure 2. Preincubation of 80S·IRES complex with 1 µM or 3 µM eEF2·GTP leads to clear biphasic binding of Phe-TC, with the more rapid and slower phases each accounting for ~50% of binding, respectively (red and black traces). These results indicate that, consistent with recent results of Petrov et al. (2016), the equilibrium between Structures 1 and 2 is shifted in the presence of eEF2·GTP, such that approximately half of 80S·IRES is present as 2.Phe-TC binding to 2, resulting in the formation of Structure 3, accounts for the rapid phase in the red and black traces. Further formation of 3 is limited by the slower rate of 1 to 2 conversion. Although added eEF2·GTP decreases all three apparent rate constants, the effect is much greater on k-1 (~50-fold reduction) than on either k1 (~twofold reduction) or k2 (~fourfold reduction). The near identity of the red and black traces, performed at different eEF2·GTP concentrations, suggests that this factor interacts with both 1 and 2, with a dissociation constant significantly less than 1 µM. The large inhibitory effect of eEF2·GTP on k-1 is consistent with its role as a translocase, and with recent results demonstrating that a principal role of EF-G, the prokaryotic equivalent of eEF2, is to inhibit back-translocation (Adio et al., 2015). eEF2·GTP inhibition of k2 may be due, at least in part, to a requirement for eEF2·GDP dissociation prior to Phe-TC binding. Figure 3 with 2 supplements see all Download asset Open asset Rates of initial Phe-tRNAPhe binding measured by fluorescence anisotropy or Phe-tRNAPhe cosedimentation. Fluorescence anisotropy changes were monitored after rapid mixing of Phe-tRNAPhe (Prf) ternary complex (0.1 µM final concentration, containing 1 mM GTP)with 80S·FVKM-IRES complex (0.1 µM final concentration) either in the absence of eEF2 (blue line) or with 80S·FVKM-IRES complex that was pre-incubated with either 3 µM (black line) or 1 µM eEF2·GTP (red line) for 1–2 hr. These long times ensured full equilibration prior to TC addition. In the latter cases, eEF2 concentration was kept constant by including 3 µM or 1 µM eEF2, respectively, in the TC solution. eEF2 displays virtually no GTPase activity when it is not bound to the ribosome (Nygård and Nilsson, 1989). Rates of Phe-tRNAPhe binding to the P site, as determined by cosedimentation, were measured by rapidly mixing Phe-TC (1.6 µM final concentration) with 80S·FVKM-IRES complex (0.8 µM final concentration) pre-incubated for 5’ – 60’ in the presence (1 µM) (□) or absence of eEF2·GTP (○). In both cases, eEF2 final concentration after mixing was adjusted to 1 µM, by including 1 µM or 2 µM eEF2·GTP, respectively, in the TC solution. After quenching with 0.5 M MES buffer (pH 6.0), ribosome bound Phe-tRNAPhe was measured by cosedimentation. In the preincubation experiment, three-fold increases of both eEF2·GTP and Phe-TC concentrations, or of just eEF2·GTP concentration, had little effect on the cosedimentation results. Results in this Figure are corrected for IRES-independent changes in fluorescence anisotropy or Phe-tRNAPhe cosedimentation (Figure 3—figure supplements 1,2). All three solid green lines are best fits of the results obtained to the scheme in Figure 2, using the numerical integration program Scientist. https://doi.org/10.7554/eLife.13429.006 Table 1 Apparent rate constants for Steps 1 and 2. https://doi.org/10.7554/eLife.13429.009 Apparent rate constants (s-1)-eEF2+eEF2k10.0071 ± 0.00330.0033 ± 0.0001k-10.15 ± 0.040.0034 ± 0.0001k2 ([Phe-TC] = 0.1 µM)0.11 ± 0.040.0256 ± 0.0002 Formation of Structure 4 from Structure 1, as measured by the co-sedimentation assay, requires the presence of eEF2·GTP and proceeds at a considerably slower rate than formation of Structure 3 from Structure 1 (Figure 3), allowing estimation of a t1/2 for Step 3, a second pseudo-translocation step involving conversion of 3 to 4, of 210 ± 10 s. It is this further slow step that accounts for the lack of significant effect of preincubation with eEF2·GTP (5’ or 60’) on the rate of formation of 4 from 1 (Figure 3). Rates of oligopeptide formation and Met-tRNAMet cosedimentation Using ribosomes programmed with the appropriate coding sequence mutants (Supplementary file 1) and [35S]-Met-TC, we employ a rapid mixing and quench assay to measure rates of PheMet, PheLysMet, and PheLysValMet synthesis, with detection and quantification of product by thin layer electrophoresis (TLE) (Figure 4A and Figure 4—figure supplement 1). For PheMet synthesis (Figure 4B) we preform Structure 4 and measure its conversion to Structure 6. We measurePheLysMet synthesis, Structure 9, starting from either Structure 4 or Structure 7 (Figure 4C) and PheValLysMet synthesis, Structure 12, starting from either Structure 7 or Structure 10 (Figure 4D). In all three cases, reactions involving only TC binding and a single peptide bond formation (4 to 6; 7 to 9; 10 to 12) proceed in remarkably similar fashion, each showing biphasic behavior with a rapid phase accounting for 65 ± 10% of reaction proceeding with a t1/2 of ~6–9 s and a slower, minor phase proceeding much more slowly (t1/2 ~220–240 s), possibly corresponding to defective ribosomes. Reactions involving formation of two peptide bonds, as in the conversion of 4 to 9 or 7 to 12 are well approximated as single phase reactions with t1/2 values of 90–110 s. Conversion of 4 to 9 proceeds via Steps 4 – 8, allowing the t1/2 value for the translocation Step 6 to be estimated as 84 s, from the difference between the t1/2 value for the 4 to 9 reaction and the sum of the t1/2 values for the 4 to 6 and 7 to 9 reactions (major phases). Similarly, the t1/2 value for the translocation Step 9 can be estimated as 110 s from the difference between the t1/2 value for the 7 to 12 reaction and the sum of the t1/2 values for the 7 to 9 and 10 to 12 reactions. Since the di-, tri- and tetrapeptides synthesized in the results reported in Figure 4 use different coding sequence mutants, these estimates of translocation t1/2 values depend on the not unreasonable assumption that the identities of the tRNAs undergoing translocation do not have a major influence on the translocation rate. With this caveat, the results presented in Figure 4 lead to the clear conclusion that translocation is the rate limiting step in each of the first two cycles of polypeptide elongation, proceeding from 4 to 10. Figure 4 with 2 supplements see all Download asset Open asset Kinetics of peptide synthesis and Met-tRNAMet cosedimenting with ribosomes. Reaction mixtures were quenched at various times after mixing. Peptide synthesis aliquots were quenched with 0.8 M KOH, and the released [35S]-containing peptide was resolved and quantified by TLE and autoradiography (Materials and methods). Cosedimentation assay aliquots were quenched with with 0.5 M MES buffer (pH 6.0) and [35S] cosedimenting with ribosomes was determined. For all the reactions shown, final concentrations of reactants after mixing were: 80S·IRES complexes (0.8 μM); all added TCs (1.6 µM); eEF2·GTP (1 µM). The numbers in blue in parts (B–D) refer to the Structures in Figure 2 whose rates of conversion are measured. For example, the peptide synthesis result in part (B) labeled 4 – 6 measures conversion of Structure 4 to Structure 6. (A) Time course for formation of PheValLysMet tetrapeptide as determined by TLE. 80S·FVKM-IRES complex was mixed with Phe-TC, Val-TC, Lys-TC and [35S]-Met-TC. The migration positions of [35S]-Met and [35S]-PheValLysMet (*) are indicated. (B) 80S·FM-IRES complexes with Phe-tRNAPhe at the P site were mixed with [35S]-Met-TC. Dipeptide synthesis (□); cosedimentation assay (■). (C) Tripeptide synthesis: 80S·FKM-IRES complexes with either Phe-tRNAPhe (O) in the P site (Structure 4) or PheLys-tRNALys (Δ) in the P site (Structure 7) were mixed with either Lys-TC and [35S]-Met-TC or with just [35S]-Met-TC, respectively. Cosedimentation assay: 80S·FKM-IRES complex with PheLys-tRNALys in the P site was mixed with [35S]-Met-TC (■). (D) Tetrapeptide synthesis: 80S·FVKM-IRES complexes with either PheVal-tRNAVal (O) in the P site (Structure 7) or PheValLys-tRNALys (Δ) in the P site (Structure 10) were mixed with either Lys-TC and [35S]-Met-TC or with just [35S]-Met-TC, respectively. Cosedimentation assay: 80S·FKM-IRES complex with PheValLys-tRNALys in the P site was mixed with [35S]-Met-TC (■). Solid lines are best fits using single (B, 4–7; C, 4–9; D, 7–12) or double (B, 4–6; C, 7–8, 7–9; D, 10–11, 10–12) exponentials. https://doi.org/10.7554/eLife.13429.010 Table 2 t1/2 values*. https://doi.org/10.7554/eLife.13429.013 Step (s)t1/2 (s)1† 1 (+eEF2)†230 ± 5 237 ± 52‡ 2 (+eEF2)‡15 ± 9 30 ± 53210 ± 104 + 58 ± 24-898 ± 156 = (4-8) – (4+5) – (7+8)§84 ± 1673 ± 18 = (7+8) – 7§4 ± 27 + 86 ± 27-11128 ± 269 = (7-11) – (7+8) – (10+11)§110 ± 30102 ± 111 = (10 + 11) – 10§7 ± 310 + 119 ± 212<10 * Error ranges shown are based on the variances of fits to single or double exponentials of the results presented in Figure 4, unless otherwise noted. † Calculated as 0.69 (k-1 + k2)/k1k2 (see Table 1). ‡ Calculated as 0.69 (k-1 + k2)/k22 (see Table 1). § Error ranges for these steps, which are not observed directly, are based on the error ranges of the directly observed steps. In an attempt to resolve the TC binding step (reactions 4, 7, and 10) from the peptide formation step (reactions 5, 8, and 11) we also employed a rapid mixing and quench assay to determine the rates with which [35S]-Met-tRNAMet is able to cosediment with the ribosome following mixing of [35S]-Met-TC with structures 4, 7, or 10. This strategy was successful for [35S]-Met–TC reaction with structure 7 (containing P-site bound PheLys-tRNALys, Figure 4C) or structure 10 (containing P-site bound PheValLys-tRNALys Figure 4D), in which the [35S]-Met-TC cosedimentation rates outpace the rates of peptide bond formation with Met-TC. These rate differentials permit estimates to be made for the t1/2 values of TC binding (Step 7, 3 s; Step 10, 2 s) and peptide bond formation (Step 8, 4 s; Step 11, 7 s). They also provide a clear indication that, within Structures 8, 9, 11 and 12, Met-tRNAMet, PheLysMet-tRNAMet, and PheValLysMet- tRNAMet, whenbound to the A-site, efficiently cosediment with ribosomes, which is typical for A-site bound tRNAs in conventional (non-IRES) elongation complexes (Warner and Rich, 1964; Nwagwu, 1975). However, for [35S]-Met–TC reaction with structure 4 (containing P-site bound Phe-tRNAPhe), the [35S]-Met-TC cosedimentation rate is much slower than the dipeptide formation rate (Figure 4B). This indicates that PheMet-tRNAMet, and possibly Met-tRNAMet as well, are not bound stably to the ribosome in Structures 5 and 6, and that only PheMet-tRNAMet bound to the P-site (Structure 7) is fully recovered by cosedimentation. As a result, the cosedimentation assay does not permit estimation of the t1/2 values for Steps 4 and 5. It is possible that the lability of the A-site tRNAs in structures 5 and 6 is due to IRES binding to the E-site, which is absent in structures 8, 9 and 11, 12, and may reflect an allosteric A-site: E-site interaction. Evidence for allosteric A-site/E-site interactions has been presented for both bacterial and eukaryotic ribosomes (Nierhaus 1990; Chen et al., 2011; Ferguson et al., 2015), although the general validity of this interaction has been questioned (Semenkov et al., 1996; Petropoulos and Green, 2012). Translocation of tetrapeptidyl-tRNA (Step 12) is much more rapid than of tripeptidyl-tRNA (Step 9) The results presented in Figure 4 show that translocation proceeds slowly through the first two elongation cycles of nascent protein synthesis, raising the question of how far nascent protein synthesis has to proceed to overcome the retarding effect of ribosome-bound IRES. In Figure 5 we present the results of two experimental approaches demonstrating that translocation of tetrapeptidyl-tRNA proceeds much more rapidly than translocation of tripeptidyl-tRNA. Figure 5 with 1 supplement see all Download asset Open asset Tetrapeptide translocation (Step 12) is faster than tripeptide translocation (Step 9). (A) Puromycin reaction with PheValLys-tRNALys bound either at the A site (D) or at the P-site (O) of the 80S·FVKM-IRES complex or being translocated from the A site to the P site (□). (B) Puromycin reaction with PheValLysMet-tRNAMet either bound at the P-site (O) of the 80S·FVKM-IRES complex or being translocated from the A site to the P site (□). Lines in A. and B. Are fits to single exponentials. (C) Time dependence of PheLysValArgGlnTrpLeuMet octapeptide synthesis from the 80S·FKVRQWLM-IRES complex containing various peptidyl-tRNAs pre-bound at the P site, as indicated. The pre-bound peptidyl tRNAs were prepared using the standard procedure (see Complex Preparations in Materials and methods) by incubating the 80S-IRES complex with the relevant TCs for 15 min. The remaining TCs needed for octapeptide synthesis, including [35S]-Met-TC, were then added, each at a concentration of 1.6 µM, for the indicated times prior to quenching. PheLysValArgGlnTrpLeuMet octapeptide synthesis was measured by [35S]-Met cosedimenting with 80S ribosome. https://doi.org/10.7554/eLife.13429.014 The first approach makes use of the fact that formation of peptidyl-puromycin proceeds more rapidly with peptidyl-tRNA bound to the P-site than to the A-site, permitting puromycin reactivity to distinguish A-site from P-site peptidyl-tRNA. As shown in Figure 5A, puromycin (1 mM) reacts with A-site bound PheValLys-tRNALys, Structure 9, about 20times more slowly (t1/2 1400 ± 300 s) than it reacts with P-site bound PheValLys-tRNALys(t1/2 76 ± 16 s). The corresponding t1/2 value for puromycin reaction with PheValLys-tRNALys undergoing translocation from the A- to P-site is 170 ± 30 s. This increase of approximately 100 s for translocating PheValLys-tRNALysvs. translocated PheValLys-tRNALys closely matches the t1/2 value of 110 ± 30 s estimated above for the translocation of tripeptidyl-tRNA (Table 2) and can be assigned to the translocating step. In contrast, the rates of puromycin reaction with translocating and translocated PheValLysMet-tRNAMet (Structure 13)are indistinguishable from one another (t1/2 values of 37 ± 4 s and 46 ± 7 s, respectively, Figure 5B), a clear demonstration that translocation of PheValLysMet-tRNAMet proceeds rapidly with respect to puromycin reaction. Our results allow us to estimate an upper limit value of t1/2 for the translocation Step 12 of ≤10 s. Puromycin reacts at similar rates with translocated PheValLys-tRNALys (Structure 10, t1/2 76 ± 16 s) and PheValLysMet-tRNAMet (Structure 13, t1/2 46 ± 7 s). These rates, while consistent with those reported by others for puromycin reaction with eukaryotic P-site bound Met-tRNAMet (Lorsch and Herschlag, 1999), N-AcPhe-tRNAPhe (Ioannou et al., 1997), and Cy3-Met-tRNAMet (Ferguson et al., 2015), are several hundred-fold slower than those measured for puromycin reaction with prokaryotic P-site bound peptidyl- or fMet-tRNA. This largely explains why the rate reduction for puromycin reaction with A-site vs. P-site bound peptidyl-tRNA is so much more modest for eukaryotic ribosomes (~20-fold, Figure 5A) than for prokaryotic ribosomes (103–104-fold, Pan et al., 2007; Semenkov et al., 1992 ; Sharma et al., 2004; Peske et al., 2004 ). Above we have demonstrated that, under our conditions, aa-tRNA binding and peptide bond formation proceed with an overall t1/2of 6 – 9 s for each of the three elongation steps we have studied. This relative constancy, coupled with the much slower translocation of tripeptidyl-tRNA (Step 9) vs. tetrapeptidyl-tRNA (Step 12), leads to the prediction that synthesis of a longer peptide that required the tripeptidyl-tRNA translocation step (Step 9) would proceed significantly more slowly than synthesis not requiring this step. In the second approach we verified this prediction by demonstrating that octapeptide FKVRQWLM formation, as measured by the cosedimentation assay, is much slower when synthesis is initiated with P-site bound PheLys-tRNALys (Structure 7) vs. P-site bound PheLysVal-tRNAVal (Structure 10) (Figure 5C). Indeed, the rates of FKVRQWLM synthesis are only marginally increased when reaction is initiated with P-site bound tetrapeptidyl-tRNA or pentapeptidyl-tRNA as compared with tripeptidyl-tRNA, reinforcing the notion that the retarding effect of ribosome-bound IRES on protein synthesis is largely overcome following translocation of tripeptidyl-tRNA. Discussion The results presented in this paper constitute the first time that rates of reaction have been determined for the initial cycles of IRES-dependent elongation. They demonstrate quite clearly that the first two cycles of elongation proceed much more slowly than subsequent steps, and that these reduced rates arise from slow, rate-determining, pseudo-translocation and translocation steps. Translocation during the first elongation cycle (Step 6) clearly requires displacement of the IRES from the E-site, so it is not unexpected that it would be slow. Less predictable is the slow translocation in the second elongation cycle, (Step 9) after the IRES structure has, presumably, already left the E-site (Figure 1). The slow rate of Step 9 might be due to a full dissociation of IRES from the ribosome during this step, a suggestion that could be tested by appropriately designed structural studies. In any case, our results do clearly demonstrate that, following translocation of tripeptidyl-tRNA from the A- to P-site, the pace of nascent peptide chain elongation picks up dramatically. Further work, comparing quantitatively the rates of successive cycles of nascent peptide elongation following tetrapeptide formation (i.e, cycles 4, 5, 6, 7, etc.) will be required to determine how many cycles are required before any retarding influence of bound CrPV-IRES is completely eliminated. Our results also clarify an aspect of the initial binding of the first aa-tRNA to the 80S·CrPV-IRES complex. Prior results have shown that initial aa-tRNA binding, in the form of a ternary complex, to an 80S·IRES complex, as measured either by cosedimentation (Fernandez et al., 2014), or by filter binding and toeprinting (Yamamoto et al., 2007), requires eEF2·GTP, leading to the conclusion that initial aa-tRNA binding can only bind to the 80S·IRES complex after an eEF2-dependent translocation event (Fernandez et al., 2014). While we agree with the experimental results, and have in fact reproduced the cosedimentation result in our own work, we disagree with the conclusion. This is because these earlier experiments only measured stable aa-tRNA binding, corresponding to formation of Structure 4 in which aa-tRNA binds to the P-site. However, it is clear from the anisotropy experiment conducted in the absence of added eEF2·GTP (Figure 3, blue trace) that ternary complex binding measured in situ, which can monitor labile binding to the A-site (Structure 3)does not require eEF2·GTP. This is easily understood as an example of Le Chaltelier’s principle, in which the equilibrium between Structure 1 (clos
'%3e%3cpath fill='%23012169' d='M0 0v30h60V0z'/%3e%3cpath stroke='%23fff' stroke-width='6' d='m0 0 60 30m0-30L0 30'/%3e%3cpath stroke='%23C8102E' stroke-width='4' d='m0 0 60 30m0-30L0 30' clip-path='url(%23b)'/%3e%3cpath stroke='%23fff' stroke-width='10' d='M30 0v30M0 15h60'/%3e%3cpath stroke='%23C8102E' stroke-width='6' d='M30 0v30M0 15h60'/%3e%3c/g%3e%3c/svg%3e)