Current Protein and Peptide Science - Volume 3, Issue 1, 2002

Our understanding of the process of translation has progressed rapidly since the availability of highly resolved structures for the ribosome. A wealth of information has emerged in terms of both RNA and protein structure and the interplay between them. This has prompted us to revisit the astonishing “treasure trove” of functional data regarding the ribosome that has accumulated over the past decades. Here we try a systematic synopsis of these ribosomal functions in light of the cryo-electron microscopic structures (resolution >7 Å) and the atomic x-ray structures (>2.4 Å) of the ribosome.

Analysis of the high resolution structure of the small subunit from Thermus thermophilus shed light on its inherent conformational variability and indicated an interconnected network of features allowing concerted movements during translocation. It also showed that conformational rearrangements may be involved in subunit association and that a latch-like movement guarantees processivity and ensures maximized fidelity. Conformational mobility is associated with the binding and the anti association function of initiation factor 3, and antibiotics interfering with prevent the initiation of the biosynthetic process. Proteins stabilize the structure mainly by their long basic extensions that penetrate into the ribosomal RNA. When pointing into the solution, these extensions may have functional roles in binding of non-ribosomal factors participating in the process of protein biosynthesis. In addition, although the decoding center is formed of RNA, proteins seem to serve ancillary functions such as stabilizing ist required conformation and assisting the directionality of the translocation.

Structural analysis of the recently determined high resolution structures of the small and the large ribosomal subunits from three bacterial sources, assisted by the medium resolution structure of a complex of the entire ribosome with three tRNAs, led to a quantum jump in our understanding of the process of the translation of the genetic code into proteins. Results of these studies highlighted dynamic aspects of protein biosynthesis illuminated the modes of action of several antibiotics indicated strategies adopted by ribosomes for maximizing their functional activity and revealed a wealth of architectural elements, including long tails of proteins penetrating the particle's cores and stabilizing the intricate folds of the RNA chains. Binding of substrate analogues showed that the decoding and the peptide-bond formation are accomplished mainly by RNA. However, several proteins may be functionally relevant in directing the mRNA and in mediating the proper orientation of the tRNA molecules within the ribosomal rRNA frame. Elements involved in intersubunit contacts or in substrate binding are inherently flexible, but maintain well-ordered characteristic conformations in unbound particles. The ribosomes utilize this conformational variability for optimizing their efficiency and minimizing non-productive interactions, hence disorder of functionally relevant features may be linked to less active conformations or to far from physiological conditions. Clinically relevant antibiotics bind almost exclusively to rRNA. In the small subunit they affect the decoding accuracy or limit conformational mobility and in the large subunit they either interfere with substrate binding, by interacting with components of the peptidyl transferase cavity, or hinder the progression of the growing peptide chain.

Single particle electron cryomicroscopy is nowadays routinely used to generate three-dimensional structural information of ribosomal complexes without the need of crystallization. A large number of structures of functional important ribosomal complexes have thus been determined using this technique. In E. coli 70S ribosomes all three tRNA binding sites could be localized. The ternary complex of EF-Tu·tRNA·GTP that delivers the tRNA to the ribosome was directly visualized in a ribosomal complex blocked by the antibiotic kirromycin. Three different functional states of translocation have been studied and the respective EF-G binding sites have been mapped. The level of resolution achievable with electron cryomicroscopy allows conformational changes in the domain structures of elongation factors to be modelled in terms of rigid body movements. Structural information on eukaryotic ribosomes is also available for yeast and mammalian 80S ribosomes. The structural differences between rabbit 80S and E. coli 70S ribosomes could be interpreted in terms of ribosomal RNA expansion segments in the 18S and 23S RNA. The EF-G homologue EF2 was mapped analysing the structure of an 80S·EF2·sodarin complex and most recently the binding of a hepatitis C virus IRES element to a yeast 40S subunit has been studied. The first electron cryomicroscopical 3D reconstructions have further been used to overcome the initial phasing problems in X-ray crystallographic studies of the ribosome facilitating structure determination of the recent atomic resolution structures of the 30S and 50S ribosomal subunits. In turn, the knowledge of the atomic structure of the ribosome makes detailed interpretations of cryo-EM maps possible at ∼20 Å resolution.

The acidic L7 / L12 (prokaryotes) and P1 / P2 (eukaryotes) proteins are the only ribosomal components that occur in more than one, specifically four, copies in the translational machinery. These ribosomal proteins are the only ones that do not directly interact with ribosomal RNA but bind to the particles via a protein, L10 and P0, respectively. They constitute a morphologically distinct feature on the large subunit, the stalk protuberance. Since a long time proteins L7 / L12 have been implicated in translation factor binding and in the stimulation of the factor-dependent GTP-hydrolysis. Recent studies reproduced such activities with the isolated components and L7 / L12 can therefore in retrospect be regarded as the first GTPase activating proteins identified.GTP-hydrolysis induces a drastic conformational change in elongation factor (EF) Tu, which enables it to dissociate from the ribosome after having successfully delivered aminoacylated tRNA into the A-site. It is also used as a driving force for translocation, mediated by EF-G. The in vitro stimulation of translation-uncoupled EF-G-dependent GTP-hydrolysis seems to be an intrinsic property of the ribosome that is dependent on L7 / L12, reaches a maximum with four copies of the proteins per particle, and reflects the in vivo hydrolysis rate during translation. It is much larger than the analogous activity observed for EF-Tu, which is correlated with the in vitro polypeptide synthesis rate. Therefore, at least certain stimulatory activities of L7 / L12 are controlled by the ribosomal environment, which in the case of EF-Tu senses the successful codon-anticodon pairing. Present knowledge is consistent with a picture in which proteins L7 / L12 constitute a ‘landing platform’ for the factors and after rearrangements induce GTP-hydrolysis. The molecular mechanism of the GTPase activation is unknown.While sequence comparisons show a large diversity in the stalk proteins across the kingdoms, a conserved functional domain organization and conserved designs of their genetic units are discernible. Consistently, stalk transplantation experiments suggest that coevolution took place to maintain functional L7 / L12 - EF-G and P-protein - EF-2 couples.The acidic proteins are organized into three distinct functional parts: An N-terminal domain is responsible for oligomerization and ribosome association, a C-terminal domain is implicated in translation factor interactions, and a hinge region allows a flexible relative orientation of the latter two portions. The bacterial L7 / L12 proteins have long been portrayed as highly elongated dimers displaying globular C-terminal domains, helical N-termini, and unstructured hinges. Conversely, recent crystal structures depict a compact hetero-tetrameric assembly with the hinge region adopting either an α-helical or an open conformation. Two different dimerization modes can be discerned in these structures. Models suggest that dimerization via one association mode can lead to elongated dimeric complexes with one helical and one unstructured hinge. The physiological role of the other dimerization mode is unclear and is in apparent contradiction to distances measured by fluorescence resonance energy transfer. The discrepancies between the crystal structures and results from other physico-chemical methods may partly be a consequence of the dynamic functions of the proteins, necessitating a high flexibility.

Bacteria require three initiation factors, IF1, IF2 and IF3, to start protein synthesis. In the last few years the elucidation of both structural and mechanistic aspects pertaining to these proteins has made substantial progress. In this article we outline the translation initiation process in bacteria and review these recent developments giving a summary of the main features of the structure and function of the initiation factors.

Since the pioneering discovery of the inhibitory effects of kirromycin on bacterial elongation factor Tu (EF-Tu) more than 25 years ago [1], a great wealth of biological data has accumulated concerning protein biosynthesis inhibitors specific for EF-Tu. With the subsequent discovery of over two dozen naturally occurring EF-Tu inhibitors belonging to four different subclasses, EF-Tu has blossomed into an appealing antimicrobial target for rational drug discovery efforts. Very recently, independent crystal structure determinations of EF-Tu in complex with two potent antibiotics, aurodox and GE2270A, have provided structural explanations for the mode of action of these two compounds, and have set the foundation for the design of inhibitors with higher bioavailability, broader spectra, and greater efficacy.

tRNA is the adaptor in the translation process. The ribosome has three sites for tRNA, the A-, P-, and E-sites. The tRNAs bridge between the ribosomal subunits with the decoding site and the mRNA on the small or 30S subunit and the peptidyl transfer site on the large or 50S subunit. The possibility that translation release factors could mimic tRNA has been discussed for a long time, since their function is very similar to that of tRNA. They identify stop codons of the mRNA presented in the decoding site and hydrolyse the nascent peptide from the peptidyl tRNA in the peptidyl transfer site. The structures of eubacterial release factors are not yet known, and the first example of tRNA mimicry was discovered when elongation factor G (EF-G) was found to have a closely similar shape to a complex of elongation factor Tu (EF-Tu) with aminoacyl-tRNA. An even closer imitation of the tRNA shape is seen in ribosome recycling factor (RRF). The number of proteins mimicking tRNA is rapidly increasing. This primarily concerns translation factors. It is now evident that in some sense they are either tRNA mimics, GTPases or possibly both.

The amino acid selenocysteine represents the major biological form of selenium. Both the synthesis of selenocysteine and its co-translational incorporation into selenoproteins in response to an in-frame UGA codon, require a complex molecular machinery. To decode the UGA Sec codon in eubacteria, this machinery comprises the tRNASec, the specialized elongation factor SelB and the SECIS hairpin in the selenoprotein mRNAs. SelB conveys the Sec-tRNASec to the A site of the ribosome through binding to the SECIS mRNA hairpin adjacent to the UGA Sec codon. SelB is thus a bifunctional factor, carrying functional homology to elongation factor EF-Tu in its N-terminal domain and SECIS RNA binding activity via its C-terminal extension. In archaea and eukaryotes, selenocysteine incorporation exhibits a higher degree of complexity because the SECIS hairpin is localized in the 3' untranslated region of the mRNA. In the last couple of years, remarkable progress has been made toward understanding the underlying mechanism in mammals. Indeed, the discovery of the SECIS RNA binding protein SBP2, which is not a translation factor, paved the way for the subsequent isolation of mSelB / EFSec, the mammalian homolog of SelB. In contrast to the eubacterial SelB, the specialized elongation factor mSelB / EFSec the SECIS RNA binding function. The role is carried out by SBP2 that also forms a protein-protein complex with mSelB / EFSec. As a consequence, an important difference between the eubacterial and eukaryal selenoprotein synthesis machineries is that the functions of SelB are divided into two proteins in eukaryotes. Obviously, selenoprotein synthesis represents a higher degree of complexity than anticipated, and more needs to be discovered in eukaryotes. In this review, we will focus on the structural and functional aspects of the SelB and SBP2 factors in selenoprotein synthesis.