Book of Abstracts: Albany 2009

category image Albany 2009
Conversation 16
June 16-20 2009
© Adenine Press (2008)

Modeling of Structure and Dynamics of Eukaryotic Ribosomal Termination Complex of eRF1-mRNA-tRNA

Translation is terminated when the ribosomal A-site is occupied by one of the three stop codons: UAA, UAG, or UGA, which are recognized by the class 1 release factor RF. In eukaryotes, all three stop codons are recognized by the only RF, eRF1, which does not have a sequence and structure similarity with two factors, RF1 and RF2 of prokaryotes. The crystal structure of eRF1 (1) and ribosomal complexes of prokaryotic Thermus thermophilus including 70S ribosome, mRNA, and tRNA in the A-, P-, and E-sites (2) as well the structure of 70S ribosome in complex with RF1 and RF2 (3) are known. The structural data on eukaryotic ribosomal translation termination complexes are indirect and restricted to the result of biochemical studies, crosslinking of the stop codon nucleotides to the eRF1 residues and low-resolution, ∼14 Å, cryoelectron microscopy, because stable eukaryotic ribosomal termination complexes are difficult to obtain in vitro.

3D model of ribosomal triple molecular complex of eRF1-mRNA-tRNA in P-site is reconstructed by computational modeling based on structure of individual molecules, i.e., eRF1, mRNA, tRNA, taking into account essential conformational dynamics of involved molecules and all indirect experimental structural data. In translation termination complex the A-site tRNA is replaced by the eRF1. Therefore, the spatial dimension and general shape of eRF1 molecule in the ribosomal complex should fit the respective dimensions of A-site tRNA. There are two major functional fragment that local structural organization are thought to be similar in prokaryotic and eukaryotic ribosomes: (i) positions of mRNA stop codon and tRNA anticodon triplet and respective position of stop codon binding site of the eRF1, (ii) position of the GGQ tripeptide of eRF1 in eukaryotic complex should fit the position of that tripeptide in the prokaryotic ribosome complex. The stop codon binding site of eRF1 is located around NIKS fragment of eRF1 N-domain according to biochemical data. Exact localization of the stop codon binding site of eRF1 are unknown. Models of stop codon binding sites of eRF1 were calculated by docking of stop codon UAA on the eRF1. Two models of the stop codon binding sites have been obtained. The reference position of GGQ tripeptide of RF1 was taken from the prokaryotic complex RF1-mRNA-tRNA, which is solved to 5.9 Å resolution (PDB ID 2B64) (3). Having in hand the positions of reference fragments of the eRF1, the reference structure of prokaryotic complex RF1-mRNA-tRNA, the computational modeling of the eukaryotic complex eRF1-mRNA-tRNA has been done by the following procedure. (I) the essential conformational movements of eRF1 domains has been determined to perform deformation of the crystal Y-shaped form of N-, M-, and C-domains to the L-shaped, to fit the cavity of ribosomal A-site and to move the distance between GGQ tripeptide and NIKS motif, from ∼110 Å in crystal to ∼75 Å in ribosomal complex, as it is shown in Figure 1; (II) substitute the A-site tRNA in the T. thermophilus 70S ribosome by the eRF1 in the tRNA-like conformation. (III) The mRNA with P-site tRNA was redocked to the eRF1 in the tRNA-like conformation taking into account all atom-atom interactions in the triple complex eRF1-mRNA-tRNA(P-site). (IV) the model of triple complex eRF1-mRNA-tRNA has been relaxed by the MD simulated annealing with constraint for the GQQ tripeptide position. The final two models of the triple ribosomal complex are shown in Figure 2. The main difference between two models is that the mRNA strand interacts with grooves at different sides of N-domain surface of eRF1. Models 1 and 2 are structurally similar in the M-domain GGQ tripeptide region that is responsible for peptidyl transferase activity. To choose between two models, the chemical crosslinks between nucleotides of mRNA in triple complex eRF1-mRNA-tRNA and atoms of eRF1 has been modeled for the both models of triple complex. A computer model of complex with twelve different analogs of mRNA containing a modified nucleotides in positions from +4 to +9, carrying the reactive groups have been build. Free molecular dynamics of each modified mRNA analog was simulated for 10 ns by simulated annealing at T = 250 ? 500 K. The number of short contacts < 7 Å between reactive azido group of mRNA analogs with eRF1 heavy atoms has been analyzed for every 10 ps and statistics of contacts have been accumulated. Since the photoactiable azido group is highly reactive, the probability of crosslinking is proportional to probability of its collision with eRF1 atoms. It was found that the distribution of the crosslinks calculated for the model 1 of the triple complex eRF1-mRNA-tRNA for 12 mRNA analogs and eRF1 residues coincides with the experimental crosslinks distribution. The model 2 has calculated croslinks distribution which is incompatible with experimental data.

Figure 1 (A) Conformational dynamics of eRF1 domains, a set of instantaneous structures with interval 500ps. (B) Movement of M-domain. (C) Movement of C-domain. (D) Fit of eRF1 on the tRNAphe in A-site by virtual fit of GGQ with CCA and of NIKS with AC loop.

Y.N. Vorobjev

Institute of Chemical Biology and Fundamental Medicine of Siberian Branch of Russian Academy of Sciences
Novosibirsk 630090, Russia

Email: ynvorob@niboch.nsc.ru

Figure 2: (a) Model 1 of the eRF1-mRNA-tRNA ribosomal complex. The eRF1 molecule is shown in ribbons: thin ribbon is N-domain (NIKS motif is indicated), medium ribbon is M domain (GGQ tripeptide is shown), thick ribbon is C-domain. The mRNA is shown as thick tube. The P-site tRNA is shown as smooth thin tube. (b) Model 2 of the eRF1-mRNA-tRNA ribosomal complex.

Thus, molecular modeling allowed us to construct a model of the eukaryotic translation termination ribosomal complex eRF1-mRNA-tRNA. The arrangement of eRF1 domains is substantially changes upon eRF1 binding in A-site from Y- to L-shaped to mimic tRNA (4).

The work is supported by RFFI projects No. 05-04-48322, 09-04-00136.

References and Footnotes
  1. Song, H., Mugnier, P., Webb, H. M., Evans, D. R., Tuite, M. F.,Hammings, B. A., Barford, D. Cell 100, 311-321 (2000).
  2. Yusupova, G. Z., Yusupov, M. M., Cate, J. H., Noller, H. F. Cell 106, 233-241 (2001).
  3. Petry, S., Brodersen, D. E., Murphy, IV, F. V., Dunham, C. M., Selmer, M., Tarry, M. J., Kelley, A. C., Ramakrishnan, V. Cell 123, 1255-1266 (2005).
  4. Vorobjev, Yu. N. and Kiselev, L.L. Molecular Biology 42, 341-351 (2008).