Search

In Escherichia coli, 4.5S RNA is found in complexes with both protein translocation protein, Ffh (a bacterial homolog of mammalian SRP54) and protein synthesis elongation factor G (EF-G). To analyze the function of 4.5S RNA in translation, we initially assessed the sensitivity of the association of 4.5S RNA with the ribosome after treatment with antibiotics that affect various stages of protein synthesis. Fusidic acid and viomycin caused 4.5S RNA to cosediment with the 70S ribosomal fraction, indicating that 4.5S RNA enters the ribosome before ribosomal translocation and release of EF-G–GDP from the ribosome. On the other hand, depletion of 4.5S RNA led to the retention of a significant amount of EF-G on 70S ribosomes. In addition, 4.5S RNA shares a conserved decanucleotide sequence (58GAAGCAGCCA67) motif with the characterized EF-G-binding site at positions 1068–1077 on 23S RNA. We therefore examined by gel mobility-shift assay whether or not mutations in the domain-IV region of 4.5S RNA, including this conserved motif, disturb the binding of EF-G to 23S RNA. Any mutation at the C62, G64 or A67 residues within this motif abolished competition activity. Therefore, we propose that 4.5S RNA is concerned with the mode of association of EF-G with the ribosomes. Moreover, this function depends on the secondary structure of 4.5S RNA as well as a ten-base sequence conserved between the two RNAs.

The turnover of complexes containing elongation factor (EF) G, the ribosome and either GDP plus fusidic acid or guanyl-5′-yl methylene diphosphonate or guanyl-5′-yl imidodiphosphate has been measured by chasing the complexed, radioactive forms of these guanosine nucleotides with either an excess of unlabelled guanosine nucleotide or the antibiotic thiostrepton. With the three different complexes, the decay promoted by thiostrepton was substantially faster than that induced by the guanosine nucleotide and, under both decay conditions, no simple first-order kinetics were observed. Assuming that the dissociation of the complexes proceeds via a guanosine-nucleotide · EF-G binary complex and a free ribosome, this result suggests that dissociation was reversible and that after it the inactivation of free ribosomes by thiostrepton was faster than the decay of the newly formed binary complex. Increasing the NH4+ concentration from 10 to 180 mM or decreasing the Mg2+ concentration from 30 to 10 mM during measurement of turnover accelerated the chase of [3H]GDP from a [3H]GDP · EF-G · ribosome · fusidic-acid complex. Since these variations did not affect the amount of complex present in solution, they evidently accelerated the reactions leading to formation of the complex. In contrast, increasing the NH4+ concentration of the wash buffer used to isolate the complexes on nitrocellulose membranes sharply decreased their recovery. The antibiotics tetracycline, viomycin and several aminoglycosides that induce misreading (streptomycin, neomycin, kanamycin, tobramycin, gentamicin and paromomycin) inhibited the decay of the GDP · EF-G · ribosome · fusidic acid complex, but other aminoglycosides like spectinomycin and kasugamycin had no effect. Several peptidyl transferase inhibitors (chloramphenicol, lincomycin, viridogrisein, griseoviridin, spiramycin, gougerotin and althiomycin), the EF-Tu-inactivating kirromycin and other antibiotics of less well defined mode of action (erythromycin, bottromycin and micrococcin) were also ineffective. Streptomycin and neomycin were effective at concentrations lower than 10 μM thus indicating that they were probably acting specifically and not as a result of a Mg2+ -like action caused by their polycationic nature. The antibiotics did not completely block turnover and their effect could be explained by an increase in the affinity of the ribosome for the binary GDP · EF-G complex. Since all the effective antibiotics bind to the 30-S subunit and affect functions related to the ribosomal A-site, the results emphasize the important role of this subunit for EF-G interaction and suggest that EF-G binds to the ribosome in a region near the 30-S portion of the A-site. A ternary complex containing Phe-tRNA, EF-Tu and GTP did not modify the chase rate of [14C]GDP from a [14C]GDP · EF-G · ribosome · poly(uridylic acid) · fusidic-acid complex.

The standard technique for determination of the ribosomal site location of bound tRNA, viz. the puromycin reaction, has been analyzed with regard to its applicability under tRNA saturation conditions. The criteria derived have been used to re-examine the exclusion principle for peptidyl-tRNA binding, which states that only one peptidyl-tRNA (AcPhe-tRNA) can be bound per ribosome although in principle two sites (A and P site) are available. The following results were obtained. 1 The puromycin reaction is only appropriate for a site determination if the reaction conditions prevent one ribosome from performing more than one puromycin reaction. With an excess of AcPhe-tRNA over ribosomes, and in the absence of EF-G, this criterion is fulfilled at 0°C, where the P-site-bound material reacts with puromycin (quantitative reaction after 50 h), while the A-site-bound material does not. In contrast, at 37°C the extent of the puromycin reaction can exceed the binding values by 2–4-fold (‘repetitive reaction’). In the presence of EF-G a repetitive puromycin reaction is seen even at 0°C, i.e. EF-G can already promote a translocation reaction at 0°C. However, the extent of translocation becomes negligibly low for short incubation times (up to 60 min) at 0°C, if only catalytic amounts of EF-G are used. 2 Using the criteria outlined above, the validity of the exclusion principle for Escherichia coli ribosomes was confirmed pursuing two different experimental strategies. a) Ribosomes were saturated with AcPhe-tRNA at one molecule per 70S ribosome, and a quantitative puromycin reaction demonstrated the exclusive P-site location of the AcPhe-tRNA. The same result was also found in the presence of viomycin, which blocks the translocation reaction. These findings also indicate that here nearly 100% of the ribosomes participate in AcPhe-tRNA binding to the P site. b) Precharging the P sites of 70S ribosomes with one Ac[14C]Phe-tRNA molecule per ribosome prevented additional Ac[3H]Phe-tRNA binding. In contrast, 70S particles carrying one molecule of [14C]tRNAPhe per ribosome were able to bind up to a further 0.64 molecule Ac[3H]Phe-tRNA per ribosome.

In Escherichia coli cell-free systems, the aminboglycoside antibiotic hygromycin B inhibits (80%) the elogation-factor-(EF)G-plus-GTP-dependent reaction of either AcPhe-tRNA or natural peptideyl-tRNA with puromycin. The antibiotic does not substantially affect the poly(uridylic acid)-directed, EF-Tu-plus-GTP-dependent binding of Phe-tRNA to ribosomes or the reaction of donor-site-bound AcPhe-tRNA with puromycine. Moreover, the nascnet peptide chains of either purified endogenous E. coli polysomes or poly(uridylic acid)-programmed ribosomes, blocked by hygromycin B during polypeptide synthesis, react slowly with puromycin while the chains of uninhibited polysomes or tetracycline-blocked ribosomes react very rapidly. These results indicate that hygromyckin B temporaily retains peptidely-tRNA in the ribosomal acceptor site and, consequently, that the inhibition of polypeptide chain elongation by the antibiotic is, at least in part, due to the impairment of ribosomal translocation. Hygromycin B at concentrations strongly inhibitory for translocation (80 μM) does not affect the EF-G-plus-ribosome-dependent GTP hydrolysis, the formation of the GDP · EF-G · ribosome · fusidic-acid complex or the non-enzyme binding of AcPhe-tRNA to the ribosomal donor site. In contrast, hygromycin B inhibits the non-enzymic translocation and the release of AcPhe-tRNA from the ribosomal acceptor site promoted by depletion of NH+4 ions. It is suggested that hygromycin B inhibits the movement of peptideyl-tRNA associated with translocation by interfering with its exit from the ribosomal acceptor site. Hygromycin B has been previously described as an inducer of misreading [Davies, J., Gorini, L. and Davis, B. D. (1965) Mol. Pharmacol. 1, 93–106]. It is now shown that the antibiotic interfaces with translation and induces misreading in the same poly(uridylic acid)-directed system and in approximately the same range of concentrations. The results suggest a relationship between the ribosomal processes of aminoacyl-tRNA recognition and translocation.

Ac[14C]Phe-tRNA (cognate) and Ac[3H]Leu-tRNA (miscognate) were bound as competing peptidyl-tRNA analogues to tightly coupled ribosomes programmed with poly(U). The test system allows the determination of the accuracy of tRNA selection at the A and P site, as well as the measurement of the peptidyltransferase and translocation reactions in the presence of various antibiotics. The following results were obtained. 1 Cognate tRNA could be bound to either the A or P site with nearly 10% specificity. In contrast, miscognate tRNA could be exclusively bound to the P site but poorly to the A site. 2 The accuracy of tRNA selection (extent of binding) was sixfold higher at the A site than at the P site. 3 The extent of A site binding for both cognate and miscognate tRNA was severely inhibited by viomycin, tetracycline and neomycin, and to a lesser extent by streptomycin. No inhibition of binding at the P site was found with all antibiotics tested. This finding, and that of the preceding point, is taken as evidence that acetylated aminoacyl-tRNA (an analogue for a peptidyl-tRNA) has direct access to the P site and does not reach this site via a transient A site binding. 4 In addition to the known effects of streptomycin and neomycin on tRNA selection at the A site, viomycin was found to enhance drastically the error fraction at this site. 5 The peptidyltransferase is inhibited by the drugs in the following order: chloramphenicol > viomycin, neomycin > streptomycin. Equivalent results were found with cognate and miscognate tRNA. The translocation reaction was severely inhibited by viomycin and neomycin and to a lesser extent by streptomycin. 6 A striking correspondence was found between the error-inducing capacity (distortion of codon-anticodon interaction at the A site) and the inhibition of the translocation reaction by various drugs. This finding suggests that codon-anticodon interaction at the ribosomal A site is an important if not essential prerequisite for the translocation reaction.

The basic peptide antibiotic viomycin rapidly inhibits polypeptide chain elongation when added to purified endogenous Escherichia coli polysomes actively engaged in polypeptide synthesis. After inhibition is established, the polysomal nascent peptide chains react only slowly with puromycin, while the chains of uninhibited polysomes or tetracycline-blocked polysomes react very rapidly. Since viomycin does not inhibit the peptidyl transferase reaction, these results indicate that the antibiotic confines peptidyl-tRNA to the ribosomal acceptor site and, consequently, that it blocks polypeptide chain elongation by inhibiting ribosomal translocation. Viomycin at concentrations strongly inhibitory for translocation (10 μM) does not impair the elongation-factor (EF)-G-plus-ribosome-dependent GTP hydrolysis, the formation of GDP · EF-G · ribosome · fusidic-acid complex and guanyl-5′-yl-methylene-diphosphonate · EF-G · ribosome complex, or the non-enzymic binding of AcPhe-tRNA to the ribosomal donor site. In contrast, viomycin strongly inhibits the non-enzymic translocation and the release of AcPhe-tRNA from the ribosomal acceptor site promoted by depletion of NH4+ ions. It is suggested that viomycin inhibits the movement of peptidyl-tRNA associated with translocation by interfering with its exit from the ribosomal acceptor site. Viomycin at high concentrations (10–100 μM) partially inhibits the EF-G-plus-ribosome-dependent GTP hydrolysis and the turnover of the GDP · EF-G · ribosome · fusidic-acid complex. These effects, however, are probably unrelated to the inhibition of ribosomal translocation. The binding of viomycin to AcPhe-tRNA · ribosome · poly(U) complexes is readily reversible, since the antibiotic can be removed by filtration through molecular sieves.