Arrangement of tRNAs in Pre- and Posttranslocational Ribosomes Revealed by Electron Cryomicroscopy

Abstract

The three-dimensional structure of the translating 70S E. coli ribosome is presented in its two main conformations: the pretranslocational and the posttranslocational states. Using electron cryomicroscopy and angular reconstitution, structures at 20 A resolution were obtained, which, when compared with our earlier reconstruction of "empty" ribosomes, showed densities corresponding to tRNA molecules--at the P and E sites for posttranslocational ribosomes and at the A and P sites for pretranslocational ribosomes. The P-site tRNA lies directly above the bridge connecting the two ribosomal subunits, with the A-site tRNA fitted snugly against it at an angle of approximately 50 degrees, toward the L7/L12 side of the ribosome. The E-site tRNA appears to lie between the side lobe of the 30S subunit and the L1 protuberance.

Calculation of the relative geometry of tRNAs in the ribosome from directed hydroxyl-radical probing data

“…It is important to note that in our first computation (Fig+ 5A), the anticodons are brought together in space, as are the 39-CCA ends without any explicit constraints on either position-only the probing data were used in these computations+ The terminal 39 phosphates are 23+5 Å apart, when we model the 39 terminal phosphate chain (bases 73-76) as they appear in the tRNA crystal structures+ It is likely that for the two 39 ends to bring together the growing polypeptide and the next amino acid, in preparation for the peptide bond formation step, the detailed structures of these two termini have to change+ The conformational freedom of the RNA chains from bases 73 to 76 is substantial, and capable of bringing the two amino acids into the correct approximation within the context of our modeled relative tRNA locations+ The introduction of constraints to ensure interaction of both anticodons with a single mRNA molecule and to ensure that the 39-CCA ends are close to one another does not significantly change the relative positions of the two tRNA molecules+ These observations provide strong evidence that the probing constraints provide abundant and biologically relevant information+ Our models appear to be in agreement visually with the cryoelectron microscopy model of van Heel and coworkers (Stark et al+, 1997a) although their data are completely independent of the approach used here+ Also, our model A has an RMSD to the A-and P-site tRNAs in the 70S crystal structure (Cate et al+, 1999) of 5+8 Å+ The angles between the A-and P-site tRNAs in our models fall consistently between 43 and 538 (Fig+ 5)+ Our models are most compatible with the model proposed by Easterwood and coworkers (1994), which is in the S configuration+ Models in the R configuration do not satisfy the constraints as well as those in the S+ Table 3 summarizes the average and maximum constraint errors for two of our models (A and C), the Easterwood model, and the Lim/Mueller model (Lim et al+, 1992; Mueller et al+, 1997), as well as the RMSD of all these models from our model A+ Our full atomic model of the crystal structure of A-site and P-site tRNA docked to mRNA is one possible structure compatible with these data (Fig+ 5C)+ This structure was built with the crystallographically determined positions of the anticodon loops, and without modifying the tRNA structures at all+ No energy minimization was used+ We were able to build mRNA models in other cases by allowing small deviations in the tRNA structures, with RMS deviations from the crystal structure ranging from 1 to 3 Å (Fig+ 5B)+ In these models, the anticodon side chains remained stacked, but deviated from their crystallographic positions+ It is likely that the anticodon bases actually do move slightly upon binding mRNA, and so the model we show in Figure 5C may be somewhat overconstrained+ It thus represents a proof of concept for these tRNA orientations+ Some of the intermolecular contacts in the tRNA crystal structures also suggest ways in which the codon-to-anticodon interactions can affect the conformation of the anticodon loop+ For example, the anticodon-anticodon interaction between tRNA molecules within the crystallographic asymmetric unit in the crystal structure of tRNA Asp shows one way in which codon-anticodon pairing might be accommodated while retaining the 39 stacking geometry of the anticodon loop (Westhof et al+, 1988)+ In our model building, we have classified the distances implied by our experiments into the three categories of strong, medium, and weak+ Figure 8 shows that the structures we have built are not very sensitive t...…”

“…Cryoelectron microscopy and three-dimensional image reconstruction of vacant ribosomes was used as an envelope to constrain the arrangement of tRNAs in the R orientation with a 608 angle between the A-and P-site tRNAs (Stark et al+, 1997a)+ Recently, cryoelectron microscopy was used to directly visualize the arrangement of tRNAs in the ribosome+ In one study (Malhotra et al+, 1998), three tRNAs bound to poly (U)-programmed 70S ribosomes were visualized at 25 Å resolution+ The angle between the A-site and P-site tRNAs was 1608 and the orientation of the tRNAs was intermediate between the R and S orientations+ In contrast, another cryoelectron microscopy study (Spirin, 1983) compared pre-and posttranslocational states of the ribosome and found the angle between the tRNAs to be 508 and in the S orientation+ Nierhaus et al+ (1998) used a proton-spin contrast variation technique to compare the pre-and posttranslocational states and concluded that the angle between the A-and P-site tRNAs is 110 6 108+ They were unable to distinguish between the R and S orientations+ Most recently, the crystal structure of the 70S ribosome (Cate et al+, 1999) seems to show the A-and P-site tRNAs to be almost parallel to each other in the S configuration+ Our results show that the 59-Fe(II)-ASL and 59-Fe(II)-tRNA probing data alone are sufficient to build a reliable model for the arrangement of the tRNAs in the ribosome (Fig+ 5A)+ In fact, even though the FRET data were not used as constraints for this model, they are nevertheless well satisfied by it (Table 2)+ Incorporation of the FRET data results in very little change in the structure+ However, we were unable to build a structure using the FRET data alone that satisfied the 59-Fe(II)-ASL and 59-Fe(II)-tRNA probing data, indicating that the latter have a higher information content+…”

Dissociation of Intact Escherichia coli Ribosomes in a Mass Spectrometer

“…By using cryo-EM, it has been established that complexes formed between ribosomes and EF-G can undergo major conformational changes depending on the particular translocation stage that is inhibited by a given antibiotic (32,33). These complexes therefore provide excellent model systems to test our hypothesis that gas phase dissociation is a sensitive probe of protein-RNA interactions.…”

Functional aspects of ribosomal architecture: symmetry, chirality and regulation

“…Global movements of sizable ribosomal features were detected by cryo-electron microscopy during translocation of the mRNA/tRNA molecules, a process assisted by several non-ribosomal factors. 17,18 More accurate crystallographic investigations, exploiting comparative studies performed on the available high-resolution structures of the free and the complexes of ribosomal particles, also indicated motions in both subunits. The mobile architecture of the small ribosomal subunit seems to be designed for its tasks in controlling the fidelity of the decoding process.…”