Tài liệu Báo cáo khoa học: The social life of ribosomal proteins doc

MINIREVIEW The social life of ribosomal proteins Ditlev E. Brodersen and Poul Nissen Centre for Structural Biology, Department of Molecular Biology, University of Aarhus, Denmark Introduction Ribosomes are complex macromolecular machines that are responsible for the production of every protein in every living cell [1]. Ribosomes are themselves built from the very molecules of life; protein and RNA, and ribosomal composition and structure and the inter- action between the two types of building blocks within them have always fascinated researchers. In recent lit- erature there has been renewed focus on rRNA as the main, and perhaps only, catalyst in the ribosome – a development which in the minds of many in the field has left ribosomal proteins in the dark as ‘merely glue’. In this review we highlight some of the many import- ant biological roles of ribosomal proteins, apart from being ‘RNA-glue’, and show that they indeed seem to have a social life after all. Until the year 2000, ribosomal protein structure and interaction with rRNA were mainly studied in a ‘dissect- ing’ fashion, focusing on each individual protein in turn [2]. Many individual protein structures were determined in isolation and their interactions with rRNA were mapped by various biochemical techniques, such as hyd- roxy-radical probing, protein footprinting, mutational analysis and cross-linking [3]. Though these experiments created a wealth of useful information about the struc- tural and functional organization of the ribosome, the information was very ‘local’ in the sense that it focused on the close surroundings of each ribosomal protein. The overall structure and inner workings of the ribo- some therefore remained elusive. A unified understanding of the ribosome was not possible until complete atomic structures of the two subunits that make up the bacterial 70S ribosome, the 50S and 30S subunits, were published in the summer of 2000 (Fig. 1) [4–6]. Not only did these structures (1.5 MDa and 850 kDa, respectively) represent the lar- gest nonsymmetric crystal structures ever determined, they also increased the size of the nucleic acid database (NDB; http://ndbserver.rutgers.edu/) by several orders of magnitude. The structures contained nothing short of a goldmine of information about RNA structure and immediately suggested several important new RNA folds and rationales of RNA tertiary and quater- nary structure that had not hitherto been appreciated [5,7,8]. A wealth of new information about protein– RNA interactions was likewise deduced from analysis of the 50 or more proteins in the two subunits, in a Keywords crystallography; protein synthesis; ribosomal proteins structure; ribosome; rRNA; translation Correspondence D. E. Brodersen or P. Nissen, Centre for Structural Biology, Department of Molecular Biology, University of Aarhus, Gustav Wieds Vej 10c, DK-8000 A ˚ rhus C, Denmark E-mail: deb@mb.au.dk or pn@mb.au.dk (Received 25 January 2005, accepted 7 March 2005) doi:10.1111/j.1742-4658.2005.04651.x Ribosomal proteins hold a unique position in biology because their func- tion is so closely tied to the large rRNAs of the ribosomes in all kingdoms of life. Following the determination of the complete crystal structures of both the large and small ribosomal subunits from bacteria, the functional role of the proteins has often been overlooked when focusing on rRNAs as the catalysts of translation. In this review we highlight some of the many known and important functions of ribosomal proteins, both during trans- lation on the ribosome and in a wider context. Abbreviations EF-Tu, elongation factor Tu; hnRNP, heteronuclear ribonucleoparticle; IF1, initiation factor 1; IRES, internal ribosome entry site; OB-fold, oligonucleotide-binding fold; PNPase, polynucleotide phosphorylase; RACK1, receptor of activated C kinase; SRP, signal recognition particle. 2098 FEBS Journal 272 (2005) 2098–2108 ª 2005 FEBS field which had previously been dominated by more specialized complexes such as the synthetase–transfer RNA complexes [9,10]. Ribosomal proteins in the presubunit structure era From about 1990 until the complete subunit structures were published in 2000, several research groups invested significant efforts in determining the structures of individual ribosomal protein structures [2,11,12]. With 21 proteins in the bacterial 30S and more than 30 in the 50S subunit, this was not only a gigantic task, but it also proved exceedingly difficult in many cases. Obviously, due to their tight interaction with rRNA in vivo, ribosomal proteins could not always be crystal- lised in isolation, but in some cases, such as for example small ribosomal subunit proteins S12 and S4, simply handling the isolated proteins in vitro proved extremely difficult (V. Ramakrishnan, MRC-LMB, Cambridge, UK, personal communication). Today we know that S9 / S13 tRNA binding at the P site S4/S5 ram mutations S7 tRNA binding at the E site S12 tRNA decoding at the A site A B L5 P site tRNA 5S RNA L2 / L3 Peptidyl transferase L16 (L10e) tRNA binding at the A site L23 / L24 Exit tunnel end L4 / L22 Line peptide exit tunnel L1 L1 S1 S11 mRNA binding at the E site A P E E A P Fig. 1. An overview of ribosomal proteins with known functional roles. (A) The bacterial 30S subunit from Thermus thermophilus in back (left) and front (50S-facing, right) views [5]. The approximate location and extent of ribosomal protein S1 has been indicated by a transparent green area and is based on [67]. The three tRNA-binding sites, A (aminoacyl), P (peptidyl) and E (exit) are likewise indicated. (B) The archaeal 50S subunit from Haloarcula marismortui in front (30S-facing, left) and back (right) views [4]. The approximate shape and extent of the L1 stalk has been indicated in blue. Figure prepared with PYMOL [68]. D. E. Brodersen and P. Nissen The social life of ribosomal proteins FEBS Journal 272 (2005) 2098–2108 ª 2005 FEBS 2099 these problems were caused by long peptide extrusions (colloquially known as ‘tails’) that many ribosomal proteins possess (Fig. 2A). Within the ribosome, these tails often extend away from the main core of the indi- vidual protein and anchor it to the rRNA. In fact, some of the more extended ribosomal proteins (such as S13 or S14) were never crystallised in isolation presum- ably due to their complete lack of a globular protein fold (Fig. 2B). Towards the end of the ‘presubunit era’ the first examples of structures of isolated protein– rRNA complexes representing important subdomains of the subunits emerged (such as the S15–S6–S18 [13] and L11–RNA complexes [14]). This strategy, however, became more difficult with increasing complexity and finally was made obsolete with the completion of the entire subunit structures. The L1–RNA complex repre- sents an important exception [15,16], because this region was disordered and not determined from the 50S structure [4]. Likewise, the complete structure of the S1 protein in the 30S remains unknown. From the increasing set of ribosomal protein struc- tures it was tempting to try to deduce general ideas about how ribosomal RNA is recognized in relation to the variation in protein folds [17]; however, this task proved very difficult and remained so even after the complete subunit structures had been determined [9,10]. Ribosomal proteins in the postsubunit structure era Upon the determination of both the complete 30S and 50S ribosomal subunit structures, it immediately became apparent that ribosomal proteins possessed features unlike those seen in any other protein struc- ture to date. Neutron scattering and immunoelectron microscopy experiments carried out in the 1980s had already established that most ribosomal proteins are located at the surface of the particle [18,19], while the rRNA component seemed to make up the central core. The atomic resolution crystal structures of the two subunits were indeed able to confirm most of these results but they also demonstrated that many proteins contain long peptide tails, either in their termini or present as internal, extended loop structures, which apparently function to anchor each protein to the RNA core and increase the total interaction surface with the rRNA (Fig. 2A) [4,5,9,10]. The presence of the extended tails of ribosomal pro- teins was noticed immediately by researchers working on both the 50S and 30S subunits and does seem to be a general feature of ribosomal architecture [4,5]. How- ever, focus was gathering on the question of how the ribosome performed its many functions; binding tRNA, catalysing peptidyl transfer, and the complex process of translocation. A central point was whether these functions were carried out by RNA or protein components. As catalysis was assumed to take place at internal tRNA-binding sites on the ribosome, the localization of the proteins on the surface of the parti- cle could easily mean that they were purely architec- tural, i.e. being present to shape the rRNA into the correct tertiary fold for it to carry out its catalytic function – simply ‘RNA glue’. Contrary to this, mutational analysis had for many years ascribed significant functional relevance to sev- eral ribosomal proteins, such as small ribosomal pro- tein S12, which was known to be important for correct decoding of tRNA in the ribosomal A site (Fig. 1A, Table 1) [20]. Likewise, mutations located in proteins S4 and S5 in the small subunit appeared to confer resistance to the antibiotic streptomycin, and be related to the accuracy of the ribosome and a Core domain A B Tail Zn 2+ Fig. 2. Examples of ribosomal protein structure. (A) L44e from the Haloarcula marismortui 50S subunit has a zinc-binding domain structure in one end and a long tail in the other. The protein is rain- bow-coloured from the N- (blue) to C-terminus (red). (B) S14 from the Thermus thermophilus 30S has no globular protein structure at all. Figure prepared with PYMOL [68]. The social life of ribosomal proteins D. E. Brodersen and P. Nissen 2100 FEBS Journal 272 (2005) 2098–2108 ª 2005 FEBS switch between two internal states known as restrict- ive and ribosome ambiguity (ram) [21,22]. So the question was really whether the functional effects of these mutations were due to the architectural and sta- bilizing role of the proteins alone, i.e. that perturba- tion of protein structure would affect the catalytic activity of the rRNA in an allosteric way, or that the proteins themselves were somehow involved in cata- lysis, which certainly had been the dominating hypo- thesis earlier. Even with most proteins on the surface of the ribosome, the long-ranging tails observed did allow in many cases for the latter scenario, in that they would be able to interact with functional centres directly. In fact, it was found upon determination of the subunit structures that many proteins did have residues rather close to the sites of action. Examples are S9 and S13, which have tails that come very close to the P site tRNA in the 30S; L2 and L3, which sta- bilize the rRNA surrounding the peptidyl transferase center in the 50S; and L4 and L22 that line the pep- tide exit tunnel (Fig. 1A). Furthermore, small ribo- somal subunit protein S12 was found not at the surface of the ribosome but right at the interface between the two subunits and hence very close to the tRNA-binding sites [5]. To improve the understanding of ribosomal func- tion further, the Steitz group focused on how the large ribosomal subunit carried out its catalytic func- tion, the peptidyl transfer reaction [23,24], while the Ramakrishnan group focused on how the small sub- unit would bind and decode tRNA in the A site [25–27]. From the structural analysis of the 50S sub- unit it immediately became clear that there were no protein residues near any of the rRNA bases implica- ted in peptidyl transfer, and the ribosome was quickly pronounced a ‘ribozyme’ [23]. In the small subunit the situation was more complicated because protein S12 was found very close to the decoding site at the A site. In fact, several amino acids of the protein were shown to be involved in the recognition process that leads to acceptance of the correct tRNA at the site [25]. This challenged the ribozyme idea to some extent; however, it could be shown that it was pri- marily the least significant codon–anticodon inter- action (the ‘wobble’) that was affected by protein interactions, while the predominant decoding inter- actions essentially were carried out by universally conserved bases near the 3¢ end of 16S rRNA (A1492 and A1493 in combination with G530, using Escheri- chia coli numbering). So the possibility remained that the ribosome had started its life as an entirely RNA- catalysed enzyme and only later evolved more special- ized functions that were protein-dependent. Ribosomal proteins implicated in ribosome function mRNA recognition In the translating state, the mRNA is tightly wrapped around the upper part of the 30S subunit (the ‘head’) and bends twice away from the 70S ribosome, presum- ably to avoid interference with movements required during translation (Fig. 3A) [28]. Several ribosomal proteins, primarily on the small subunit, are respon- sible for tethering mRNA to the ribosome, most noticeably S1, S7 and S11. S1 is a highly unusual ribo- somal protein being more than twice as large as the second largest protein (S2) and consisting of up to six repeats of the oligonucleotide-binding fold (OB-fold), each similar in sequence to translation initiation factor 1 (IF1) and several other RNA-binding proteins such as transcription factors (reviewed in [29]). S1 is located on the back of the 30S where it presumably has several functions, including raising the affinity of the ribosome for single-stranded RNA in a nonsequence-specific fashion as well as keeping ‘unused’ parts of the mRNA away from functionally active parts of the ribosomal Table 1. Examples of ribosomal proteins with a known function. Protein Subunit Function Prokaryotes S1 30S Non-specific mRNA binding S4 30S Functional mutations (streptomycin) S5 30S Functional mutations (streptomycin) S7 30S mRNA and tRNA binding at the E site S9 30S Interaction with P site tRNA S11 30S mRNA and tRNA binding at the E site S12 30S tRNA decoding at the A site S13 30S Interaction with P site tRNA L2 50S Required for peptidyl transferase L3 50S Required for peptidyl transferase L4 50S Lines peptide exit tunnel L5 50S Interaction with P site tRNA L10 (L7 ⁄ L12) 2 50S Factor-binding stalk L11 50S Factor binding L16 (L10e) 50S A site tRNA binding L22 50S Lines peptide tunnel L23 50S At tunnel exit, interacts with chaperones and SRP L24 50S At tunnel exit, interacts with chaperones and SRP Archaea and eukaryotes L44e 50S Interacts with E site tRNA Eukaryotes only RACK1 40S Signalling, scaffold protein D. E. Brodersen and P. Nissen The social life of ribosomal proteins FEBS Journal 272 (2005) 2098–2108 ª 2005 FEBS 2101 surface during translation. Even though no crystal structure has yet been determined for S1 (the protein was absent from the crystals of the small ribosomal subunit [30,31]), it appears likely from sequence con- siderations that each domain folds as a five-stranded b-barrel similar to the OB-fold proteins. Such domains are commonly seen in proteins that bind RNA nonspe- cifically and would allow S1 to tether a long stretch of mRNA to the ribosome using its consecutive domains. In fact, when comparing with similar proteins for which the structure is known, such as the S1-like domain from polynucleotide phosphorylase (PNPase [32]) and other homologues, it appears that there are conserved amino acids on one face of the protein that would allow for the nonspecific binding of single- stranded nucleotides [11]. However, the most important specific recognition of mRNA, at least in bacteria, works by pairing of the well-known Shine–Dalgarno sequence just upstream of the translation initiation codon with the ‘anti-Shine– Dalgarno’ sequence, a stretch of complementary poly- nucleotides located in the 3¢ end of ribosomal 16S rRNA [33]. It thus appears that even though the S1 protein may be responsible for high affinity binding of mRNA to the ribosome, the ribosomal RNA still plays an important role in this process. Whereas bacterial ribosomes seem well-tuned to translate mRNA as soon as it emerges from the tran- scriptional machinery, the association of mRNA with the eukaryotic ribosome is under complex regulation. Numerous RNA-binding proteins cover the mRNA as it is packaged and exported from the nucleus as a heteronuclear ribonucleoparticle (hnRNP) complex. These proteins are targeted by signalling pathways that link to the initiation machinery. A key player in this process is the receptor of activated C kinase (RACK1). This protein was recently shown to be in fact a ribo- somal protein [34], and localized as a seven-bladed b-propeller structure near the mRNA exit site on the 40S subunit [35]. RACK1 is a typical scaffold protein and it binds kinases such as protein kinase C (which has been shown to activate translation) and the Src kinase as well as mRNA-binding proteins such as Scp160p (reviewed in [36]). These together suggest that RACK1 orchestrates specific mRNA binding and acti- vation of protein synthesis directly on the ribosome. Interestingly, RACK1 also interacts with integrin b and other receptors, and it may further serve as a plat- form to recruit ribosomes for local translation of speci- fic mRNAs, for example in focal adhesions [36]. In eukaryotes, a bypass of the canonical factor-based initiation mechanism is possible, whereby secondary structures on the mRNA can play a major role in guiding the ribosome to internal ribosome entry sites (IRES). The mechanism is exploited by many viruses for efficient expression of viral genes, but is also used A B C RACK1 D rpL11 (L5) L1 S7 S11 rpS5 (S7) rpS0 S1 Fig. 3. mRNA binding to the ribosome. (A) A back view of the Thermus thermophilus 30S subunit showing (with purple spheres) the location of mRNA as deduced by X-ray crystallography [28]. Proteins that are known to interact with the mRNA are shown in col- our as in Fig. 1. Figure prepared with PYMOL [68]. (B) A cryoelectron microscopy recon- struction of the Hepatitis C virus IRES bound to the human 40S subunit (back view of the subunit with the IRES in dark purple). Reprinted with permission from [37]. Copy- right 2001 AAAS. (C) The cryoelectron micro- scopy structure of the cricket paralysis virus IRES bound to the human 80S ribosome. The figure shows a back view of the 40S subunit with the IRES in light purple. (D) Top view of the entire 80S ribosome of the same structure as in C. Figure panels C and D are reproduced from [38] with permission from Elsevier. The social life of ribosomal proteins D. E. Brodersen and P. Nissen 2102 FEBS Journal 272 (2005) 2098–2108 ª 2005 FEBS by a substantial number of endogenous genes in the cell. Cryoelectron microscopy studies have revealed the impressive ingenuity with which these RNA elements bind to the human 40S subunit (Fig. 3B,C) [37,38]. Of these, the structure of the human 80S in complex with a cricket paralysis virus IRES element shows that the ribosomal proteins rpL1 (L10A, prokaryotic L1), rpL11 (prokaryotic L5) and rpS5 (prokaryotic S7) all contact the IRES element (Fig. 3D) [38]. With the iden- tification and localization of RACK1 on the ribosome it appears that this protein may also be involved in IRES recognition [35]. Upon 40S IRES complex forma- tion, RACK1 is pushed downwards and forms a con- tact to the rear of the mRNA platform of the 40S subunit, to the region ascribed to the rpS0 protein. The conformational change and interactions, centred on ribosomal proteins, are also of potential importance for the canonical initiation mechanism, which will operate from the same side on the ribosome. In particular, the RACK1-mediated conformational change in the 40S subunit may play a central role during initiation. tRNA recognition and decoding The ribosome contains three binding sites for tRNA, termed the A, P and E sites (Fig. 1). Of these, the A site (where the aminoacyl tRNA initially binds and is selected) and the P site (where the peptidyl tRNA is bound) are essential while the functional involvement of the E (exit) site remains a debated issue [39]. From the subunit crystal structures it can be seen that the cru- cial operations in both the A and P sites are mainly catalysed by the rRNA component of the ribosome. Selection of cognate tRNA at the A site is thus carried out by two universally conserved adenines (A1492 and A1493 in Escherichia coli 16S) that presumably are stacked in the interior of the penultimate helix 44 of 16S rRNA in the absence of tRNA [25,40]. Upon cognate tRNA interaction with the base trip- let on mRNA in the A site, the two adenines have been shown to flip out from their position inside helix 44 to make strong hydrogen bonds with the first two bases of the duplex formed between tRNA and mRNA [25]. This movement leads to small but concerted rear- rangements throughout the 16S rRNA which presuma- bly then trigger GTP hydrolysis on elongation factor Tu (EF-Tu), which in turn signals that the tRNA has been accepted [26]. Along with the tight interactions with the cognate ternary complex a kinked conforma- tion of the tRNA anticodon stem arises which then presumably relaxes as a spring after GTPase-mediated release of EF-Tu, thereby promoting the A site accom- modation process [41]. Further structural work has concluded that only upon cognate tRNA–mRNA interaction can the energy barrier associated with the transition of the adenines be overcome. The mechan- ism thus provides a strong discrimination against near- cognate tRNAs that have only a single mismatched base pair and hence cannot be reliably rejected on the basis of free energy only [26]. Ribosomal proteins are not involved in this process except for a hydrogen bond between a serine in S12 and one of the adenines in position two. However, at the third (or ‘wobble’) position, protein S12 in the small subunit plays a more prominent role in the recognition process, in that it coordinates a magnesium ion that lies at the interface between crucial bases involved in decoding. However, as the wobble position is much less strictly monitored by the ribosome than the first two positions, the real importance of S12 in decoding can be questioned. Again it seems that the RNA has maintained the most important role in the process. The P site, where the peptidyl transfer reaction takes place on the 50S, is also mostly composed of RNA. However, two long C-terminal tails of small ribosomal subunit proteins S9 and S13 that are otherwise located at the top of the 30S, make their way down through the head of the subunit and come very close to the P site tRNA [40]. This might cause speculation as to a possible functional role of the two proteins at the P site, but this idea has recently been dismissed by showing that mutant E. coli cells that have had the C-termini of the two proteins removed are fully viable, indicating that their ribosomes are active [42]. A slower growth rate of these cells was observed, however, indi- cating that the tails might play an architectural or weak functional role. On the large subunit, no proteins make direct contacts to the P site tRNA, yet the L5 protein is close by interacting with the 5S rRNA resides on top of the subunit cleft on the central protuberance. The E site on the 30S subunit is more dominated by protein than any of the two other tRNA binding sites (Fig. 1A). Two proteins, S7 and S11, are both believed to be in contact with E site tRNA, and S7 in particular contains a long hairpin structure that might have a functional role in dislodging the tRNA from the ribosome [40]. Furthermore, the L44e pro- tein (corresponding to the L33 protein in eubacteria [43]) interacts directly with the 3¢ CCA end of E site tRNA [44]. Binding site for GTP-containing translation factors The S4, L6, L14 and L11 proteins and the stalk pro- teins L10 and L7 ⁄ L12 form the factor-binding site at D. E. Brodersen and P. Nissen The social life of ribosomal proteins FEBS Journal 272 (2005) 2098–2108 ª 2005 FEBS 2103 the edge of the intersubunit cleft of the ribosome, together with the sarcin-ricin loop and the L11 RNA region (the GTPase-associated center, GAC) (Fig. 1B). These components make direct contacts to, for exam- ple, elongation factor EF-G and the aminoacyl tRNA– EF-Tu ternary complex on the ribosome [41,45,46]. Concerted movements of these contact points relate to decoding events on the small subunit, and associate with conformational changes in factor complexes as seen for aminoacyl tRNA–EF-Tu [41]. It remains unclear how the GTPase activity of the GTP ⁄ GDP binding translation factors is in fact activated, but the the sarcin–ricin loop is the only ribosomal component that comes close to the GTP cofactor in EF-G and EF-Tu bound ribosome complexes. Thus, rRNA again seems to be responsible for a central ribosomal activ- ity, and proteins that have been suggested to activate GTP hydrolysis, in particular L11 and L7 ⁄ L12, must be now ruled out as having a direct role. The peptidyl transferase and peptide exit tunnel The heart of the ribosome, the peptidyl transferase center, is devoid of protein residues, as mentioned in the introduction. The L16 protein (L10e in eukaryo- tes) comes the closest, yet it merely supports the accommodation of aminoacyl tRNA in the A site (Fig. 1B). Further down the polypeptide exit tunnel, proteins L4 and L22 expose loops to the interior tunnel surface and form a narrow constriction [23]. This site may serve as a sensory site, which could monitor the functional state of the ribosome or per- haps also signal sequences for specific targeting of the polypeptide. It remains to be shown whether the L4–L22 constriction has any such function. Point mutations or even deletions in those regions of the L4 and L22 proteins confer resistance to antibiotics such as erythromycin, which otherwise block the tun- nel and thereby inhibit protein synthesis [47,48]. Parts of L22 and L39e also line the tunnel, and are part of what gives the tunnel its ‘Teflon-like’ proper- ties [23]. Signal recognition, secretion and chaperones The tunnel exit area is a highly important platform for external factors that interact with the nascent chain, such as the signal recognition particle (SRP), the membrane-embedded Sec61 and SecYEG com- plexes of eukaryotes and prokaryotes, respectively, as well as the trigger factor chaperone. The exit area is encircled by several ribosomal proteins, including the universally conserved L22, L23, L24 and L29 proteins (Fig. 4). The L23 protein is the central anchoring point for the SRP [49] and the trigger factor [50]. The ring of proteins around the exit area also forms the interac- tion site for the doughnut-shaped Sec61 complex embedded in the endoplasmic reticulum membrane of eukaryotic cells [51]. However, a tight seal is not formed. Instead, the interaction is centred on specific interaction points, again with L23 as a key player, together with L19e, L24 and L29 [51]. Ribosomal proteins involved in nuclear export It has been known for a long time that the nuclear export of the 5S rRNA in eukaryotes depends on a complex formation with the L5 protein and the tran- scription factor TFIIIA [52]. L5 contains the nuclear export signal, which is found in a leucine-rich region in the middle of the protein [53]. Similarly, the yeast ribosomal protein rpS15 is required for nuclear exit of the 40S subunit [54]. A more sophisticated role is played by the 60S L10e protein, which serves as the binding site for the NMD3 protein – a nuclear export factor of the entire 60S subunit [55,56]. In the cyto- plasm, NMD3 bound to L10e is released again from the 60S subunit by interaction with the cytoplasmic GTPase Lsg1p [57]. Fig. 4. The exit of the peptide tunnel on the 50S subunit. A view down the peptide tunnel from its exit at the back of the 50S towards the peptidyl transferase site inside it. The 23S rRNA is shown as a combined sticks and ribbon model and relevant pro- teins on the subunit coloured in surface representation. Figure pre- pared with PYMOL [68]. The social life of ribosomal proteins D. E. Brodersen and P. Nissen 2104 FEBS Journal 272 (2005) 2098–2108 ª 2005 FEBS Ribosomal proteins working off the ribosome Several ribosomal proteins are known to be very loosely attached to the ribosome and thus only spend part of their time working in translation. As men- tioned above, small ribosomal subunit S1 is loosely attached to the 30S subunit, probably by two out of the six repeated domains, and purified subunits are always substoichiometric in this protein [31]. Some ribosomal proteins have biological roles outside their contribution to translation, such as S10, which was shown early on to work as an antiterminator of tran- scription of k phage N protein in E. coli [58]. Fur- thermore, several ribosomal proteins are known to regulate the expression of themselves or other ribosom- al proteins through translational feedback, such as L4 that regulates S10 expression [59], and S8 that regu- lates expression of L5 [60]. RNA sequence analysis revealed that this regulation is based on a similarity between the secondary structures of the ribosomal pro- tein mRNAs and the corresponding rRNA structures [61]. The RNA-binding abilities of ribosomal proteins are thus being exploited to regulate ribosome turnover in the cell. The human RACK1 protein has also been linked with a wealth of soluble proteins and membrane re- ceptors (recently reviewed in [36]). It is possible that some of these interactions involve RACK1 as a sol- uble factor. It has been reported that a pool of free RACK1 is formed by upregulation of the expression level during stationary growth of Saccharomyces cere- visiae [62], however, a later study indicates that there is no significant pool of yeast RACK1 outside the ribosome [63]. Ribosome assembly Assembly of the dozens of components making up each ribosomal subunit is an extremely delicate pro- cess, of which we still understand little. What has been established, however, is that ribosomal assembly is a sequential rather than a concerted operation, requiring that some proteins bind to the rRNA later than others. Early in vitro reconstitution experiments with both the small and large subunit established many of the inter- dependencies of the individual ribosomal proteins dur- ing assembly and showed that in each subunit some proteins functioned as ‘initiators’ of assembly by being able to bind directly to the naked rRNA [64,65]. While the complete subunit structures enable us to confirm many of these dependencies, they still fall short of explaining the details of assembly, mainly because in each case we only know the structural endpoint of the process, namely the fully folded subunits. How- ever, there are a few interesting aspects of assembly that the structures have shed light on. In their paper describing the details of protein–RNA interactions in the 50S subunit, Klein et al. argue that proteins joining the growing complex early during assembly (‘early proteins’) must be those with the lar- gest areas of interaction with rRNA [9]. This seems intuitively right and the authors argue further that it must be true because a strong binding power is needed simply to overcome the energetic and entropic barriers associated with the initial assembly. Their hypothesis can be tested by calculating the areas of interaction for each ribosomal protein based on the crystal structures. In the 30S, for example, the six initiator proteins (S4, S7, S8, S15, S17 and S20) representing roughly 6 ⁄ 21 (29%) of the total number of protein residues (assu- ming roughly equal size) contribute approximately 35% of the total protein–RNA interface area. Thus there seems to be a slightly larger relative protein– RNA interface for early proteins; however, the way this calculation is carried out can be debated. One interesting aspect of the assembly process is how the long extensions of some of the proteins are accom- modated into the growing particle. Clearly, some con- certed action is required if a tail extends far away from a given protein, into another RNA domain, for exam- ple. Klein et al. propose a hypothesis whereby the glob- ular domain of these proteins first binds to a region of rRNA with an intermediate structure similar to its final conformation, thus stabilizing the protein on the RNA before the tail (whether terminal or internal) is placed in the structure [9]. Interestingly, the authors find that extensions are present in four of six initiator proteins in the 50S, whereas for the 30S particle all initiators are globular proteins devoid of tails [10]. The only consen- sus we can derive from this is that a globular domain with strong RNA-binding abilities is probably import- ant for initiators during early assembly. Ribosomal assembly in prokaryotes must be seen in the context of transcription of the ribosomal RNA, in that the process probably begins as soon as the 5¢ end of the nascent RNA protrudes from the polymerase complex. Chemical probing of 30S assembly intermedi- ates carried out in the Noller lab has shown that assembly does indeed proceed in a 5¢)3¢ direction [66], and this seems to correlate well with the observed interactions of proteins and rRNA as a function of each protein’s location in the reconstitution diagram [10]. In other words, ‘early proteins’ are primarily found to interact with the 5¢ end of rRNA and late proteins likewise mainly have interactions near the 3¢ end. In the large subunit, Klein et al. note that pro- D. E. Brodersen and P. Nissen The social life of ribosomal proteins FEBS Journal 272 (2005) 2098–2108 ª 2005 FEBS 2105 teins with extensive areas of rRNA interaction (hence assumed to be ‘early’) primarily bind to domain I, the 5¢ domain of 23S rRNA [9], consistent with the direc- tionality assumption. Conclusion Ribosomal proteins remain at the periphery of transla- tion in more than one sense. First and foremost, they are (with a few noteworthy exceptions) located on the surface of the particle and therefore generally far from the ‘real action’ at the centre. Second, most biochemical and structural evidence now all but exclude the proteins from the inner circle of chemical catalysts involved in the essential processes of transla- tion, in particular peptidyl transfer. However, as we hope to have illustrated above, ribosomal proteins are still much more than merely ‘RNA glue’ that hold the ribosomal RNA in place – they do have a real social life! Being at the surface of the particle, the proteins are in the best possible position to mediate the many inter- actions of the ribosome, particularly in higher organ- isms where the level of organization is so much greater. This is perhaps also visible by the mere fact that ribosomal proteins are larger and more plentiful in the eukaryotes. The field of ribosome research has for several decades been looking at the particle as an isolated protein-synthesizing machine, probably because its sheer size seemed daunting enough. But recent research has begun to expand this horizon, and thus try to understand the ribosome in a larger, cellu- lar context. This has been elegantly demonstrated by the cryoelectron microscopy structures of the yeast ribosome bound to its protein-conducting channel on the endoplasmic reticulum [51] and the recent identifi- cation of RACK1, a scaffold protein involved in signal transduction, as a ribosomal protein [35]. Common to many of these ‘extraterrestrial’ activities of the ribo- some is the strong involvement of ribosomal proteins and we envisage that many new such connections will be discovered in the future. It therefore may well be that the ribosomal proteins, in fact, are the social life of the ribosome. Acknowledgements We appreciate useful discussions with V. Ramakrish- nan, T. A. Steitz, and P. B. Moore on matters presented in this review. We are furthermore grateful to Jakob Nilsson for valuable comments and sugges- tions and Joachim Frank for providing the cryo-EM figures in Fig. 3. References 1 Garrett R, Douthwaite S, Liljas A, Matheson A, Moore P & Noller H (2000) The Ribosome: Structure, Function, Antibiotics, and Cellular Interactions. ASM Press, Washington DC. 2 Ramakrishnan V & White SW (1998) Ribosomal pro- tein structures: insights into the architecture, machinery and evolution of the ribosome. Trends Biochem Sci 23, 208–212. 3 Green R & Noller HF (1997) Ribosomes and transla- tion. Annu Rev Biochem 66, 679–716. 4 Ban N, Nissen P, Hansen J, Moore PB & Steitz TA (2000) The complete atomic structure of the large ribosomal subunit at 2.4 A ˚ resolution. Science 289, 905–920. 5 Wimberly BT, Brodersen DE, Clemons WM Jr, Morgan-Warren RJ, Carter AP, Vonrhein C, Hartsch T & Ramakrishnan V (2000) Structure of the 30S ribo- somal subunit. Nature 407, 327–339. 6 Schluenzen F, Tocilj A, Zarivach R, Harms J, Glueh- mann M, Janell D, Bashan A, Bartels H, Agmon I, Franceschi F & Yonath A (2000) Structure of function- ally activated small ribosomal subunit at 3.3 angstroms resolution. Cell 102, 615–623. 7 Nissen P, Ippolito JA, Ban N, Moore PB & Steitz TA (2001) RNA tertiary interactions in the large ribosomal subunit: the A-minor motif. Proc Natl Acad Sci USA 98, 4899–4903. 8 Klein DJ, Schmeing TM, Moore PB & Steitz TA (2001) The kink-turn: a new RNA secondary structure motif. EMBO J 20, 4214–4221. 9 Klein DJ, Moore PB & Steitz TA (2004) The roles of ribosomal proteins in the structure assembly, and evolu- tion of the large ribosomal subunit. J Mol Biol 340, 141–177. 10 Brodersen DE, Clemons WM Jr, Carter AP, Wimberly BT & Ramakrishnan V (2002) Crystal structure of the 30 S ribosomal subunit from Thermus thermophilus: structure of the proteins and their interactions with 16 S RNA. J Mol Biol 316, 725–768. 11 Draper DE & Reynaldo LP (1999) RNA binding strate- gies of ribosomal proteins. Nucleic Acids Res 27, 381–388. 12 Liljas A & Garber M (1995) Ribosomal proteins and elongation factors. Curr Opin Struct Biol 5, 721–727. 13 Agalarov SC, Sridhar Prasad G, Funke PM, Stout CD & Williamson JR (2000) Structure of the S15,S6,S18- rRNA complex: assembly of the 30S ribosome central domain. Science 288, 107–113. 14 Wimberly BT, Guymon R, McCutcheon JP, White SW & Ramakrishnan V (1999) A detailed view of a riboso- mal active site: the structure of the L11-RNA complex. Cell 97 , 491–502. 15 Nikulin A, Eliseikina I, Tishchenko S, Nevskaya N, Davydova N, Platonova O, Piendl W, Selmer M, The social life of ribosomal proteins D. E. Brodersen and P. Nissen 2106 FEBS Journal 272 (2005) 2098–2108 ª 2005 FEBS Liljas A, Drygin D, et al. (2003) Structure of the L1 protuberance in the ribosome. Nat Struct Biol 10, 104– 108. 16 Nikonov S, Nevskaya N, Eliseikina I, Fomenkova N, Nikulin A, Ossina N, Garber M, Jonsson BH, Briand C, Al-Karadaghi S, et al. (1996) Crystal structure of the RNA binding ribosomal protein L1 from Thermus ther- mophilus. Embo J 15, 1350–1359. 17 Draper DE (1999) Themes in RNA-protein recognition. J Mol Biol 293, 255–270. 18 Capel MS, Kjeldgaard M, Engelman DM & Moore PB (1988) Positions of S2, S13, S16, S17, S19 and S21 in the 30S ribosomal subunit of Escherichia coli. J Mol Biol 200, 65–87. 19 Stoffler G & Stoffler-Meilicke M (1984) Immunoelec- tron microscopy of ribosomes. Annu Rev Biophys Bioeng 13, 303–330. 20 Funatsu G & Wittmann HG (1972) Ribosomal proteins. 33. Location of amino-acid replacements in protein S12 isolated from Escherichia coli mutants resistant to strep- tomycin. J Mol Biol 68, 547–550. 21 Deusser E, Stoffler G & Wittmann HG (1970) Riboso- mal proteins. XVI. Altered S4 proteins in Escherichia coli revertants from streptomycin dependence to inde- pendence. Mol Gen Genet 109, 298–302. 22 Stoffler G, Deusser E, Wittmann HG & Apirion D (1971) Ribosomal proteins. XIX. Altered S5 ribosomal protein in an Escherichia coli revertant from strptomy- cin dependence to independence. Mol Gen Genet 111, 334–341. 23 Nissen P, Hansen J, Ban N, Moore PB & Steitz TA (2000) The structural basis of ribosome activity in pep- tide bond synthesis. Science 289, 920–930. 24 Hansen JL, Schmeing TM, Klein DJ, Ippolito JA, Ban N, Nissen P, Freeborn B, Moore PB & Steitz TA (2001) Progress toward an understanding of the structure and enzymatic mechanism of the large ribosomal subunit. Cold Spring Harb Symp Quant Biol 66, 33–42. 25 Ogle JM, Brodersen DE, Clemons WM Jr, Tarry MJ, Carter AP & Ramakrishnan V (2001) Recognition of cognate transfer RNA by the 30S ribosomal subunit. Science 292, 897–902. 26 Ogle JM, Murphy FV, Tarry MJ & Ramakrishnan V (2002) Selection of tRNA by the ribosome requires a transition from an open to a closed form. Cell 111, 721–732. 27 Brodersen DE, Carter AP, Clemons WM Jr, Morgan- Warren RJ, Murphy FVT, Ogle JM, Tarry MJ, Wimberly BT & Ramakrishnan V (2001) Atomic struc- tures of the 30S subunit and its complexes with ligands and antibiotics. Cold Spring Harb Symp Quant Biol 66, 17–32. 28 Yusupova GZ, Yusupov MM, Cate JH & Noller HF (2001) The path of messenger RNA through the ribo- some. Cell 106, 233–241. 29 Subramanian AR (1983) Structure and functions of ribosomal protein S1. Prog Nucleic Acid Res Mol Biol 28, 101–142. 30 Clemons WM Jr, May JL, Wimberly BT, McCutcheon JP, Capel MS & Ramakrishnan V (1999) Structure of a bacterial 30S ribosomal subunit at 5.5 A ˚ resolution. Nature 400, 833–840. 31 Clemons WM Jr, Brodersen DE, McCutcheon JP, May JL, Carter AP, Morgan-Warren RJ, Wimberly BT & Ramakrishnan V (2001) Crystal structure of the 30S ribosomal subunit from Thermus thermophilus: purifica- tion, crystallization and structure determination. J Mol Biol 310, 827–843. 32 Bycroft M, Hubbard TJ, Proctor M, Freund SM & Murzin AG (1997) The solution structure of the S1 RNA binding domain: a member of an ancient nucleic acid-binding fold. Cell 88, 235–242. 33 Shine J & Dalgarno L (1974) The 3¢-terminal sequence of Escherichia coli 16S ribosomal RNA: complementari- ty to nonsense triplets and ribosome binding sites. Proc Natl Acad Sci USA 71, 1342–1346. 34 Link AJ, Eng J, Schieltz DM, Carmack E, Mize GJ, Morris DR, Garvik BM, Yates JR & 3rd. (1999) Direct analysis of protein complexes using mass spectrometry. Nat Biotechnol 17, 676–682. 35 Sengupta J, Nilsson J, Gursky R, Spahn CM, Nissen P & Frank J (2004) Identification of the versatile scaffold protein RACK1 on the eukaryotic ribosome by cryo- EM. Nat Struct Mol Biol 11, 957–962. 36 Nilsson J, Sengupta J, Frank J & Nissen P (2004) Regu- lation of eukaryotic translation by the RACK1 protein: a platform for signalling molecules on the ribosome. EMBO Report 5, 1137–1141. 37 Spahn CM, Kieft JS, Grassucci RA, Penczek PA, Zhou K, Doudna JA & Frank J (2001) Hepatitis C virus IRES RNA-induced changes in the conformation of the 40s ribosomal subunit. Science 291, 1959–1962. 38 Spahn CM, Jan E, Mulder A, Grassucci RA, Sarnow P & Frank J (2004) Cryo-EM visualization of a viral internal ribosome entry site bound to human ribosomes: the IRES functions as an RNA-based translation factor. Cell 118, 465–475. 39 Blaha G & Nierhaus KH (2001) Features and functions of the ribosomal E site. Cold Spring Harb Symp Quant Biol 66, 135–146. 40 Carter AP, Clemons WM, Brodersen DE, Morgan-War- ren RJ, Wimberly BT & Ramakrishnan V (2000) Func- tional insights from the structure of the 30S ribosomal subunit and its interactions with antibiotics. Nature 407, 340–348. 41 Valle M, Zavialov A, Li W, Stagg SM, Sengupta J, Nielsen RC, Nissen P, Harvey SC, Ehrenberg M & Frank J (2003) Incorporation of aminoacyl-tRNA into the ribosome as seen by cryo-electron microscopy. Nat Struct Biol 10, 899–906. D. E. Brodersen and P. Nissen The social life of ribosomal proteins FEBS Journal 272 (2005) 2098–2108 ª 2005 FEBS 2107 [...].. .The social life of ribosomal proteins 42 Hoang L, Fredrick K & Noller HF (2004) Creating ribosomes with an all-RNA 30S subunit P site Proc Natl Acad Sci USA 101, 12439–12443 43 Harms J, Schluenzen F, Zarivach R, Bashan A, Gat S, Agmon I, Bartels H, Franceschi F & Yonath A (2001) High resolution structure of the large ribosomal subunit from a mesophilic eubacterium... from the 60S ribosomal subunit requires Rpl10p and the cytoplasmic GTPase Lsg1p EMBO J 24, 567–579 Friedman DI, Schauer AT, Baumann MR, Baron LS & Adhya SL (1981) Evidence that ribosomal protein S10 participates in control of transcription termination Proc Natl Acad Sci USA 78, 1115–1118 Olins PO & Nomura M (1981) Regulation of the S10 ribosomal protein operon in E coli: nucleotide sequence at the start... 66 67 68 of the large ribosomal subunit J Cell Biol 151, 1057– 1066 Gadal O, Strauss D, Kessl J, Trumpower B, Tollervey D & Hurt E (2001) Nuclear export of 60s ribosomal subunits depends on Xpo1p and requires a nuclear export sequence-containing factor, Nmd3p, that associates with the large subunit protein Rpl10p Mol Cell Biol 21, 3405–3415 Hedges J, West M & Johnson AW (2005) Release of the export... Mizushima S & Nomura M (1973) Reconstitution of Escherichia coli 30S ribosomal subunits from purified molecular components J Biol Chem 248, 5720– 5730 Powers T, Daubresse G & Noller HF (1993) Dynamics of in vitro assembly of 16S rRNA into 30S ribosomal subunits J Mol Biol 232, 362–374 Sengupta J, Agrawal RK & Frank J (2001) Visualization of protein S1 within the 30S ribosomal subunit and its interaction with... bound to the large ribosomal subunit Mol Cell 10, 117–128 49 Gu SQ, Peske F, Wieden HJ, Rodnina MV & Wintermeyer W (2003) The signal recognition particle binds to protein L23 at the peptide exit of the Escherichia coli ribosome RNA 9, 566–573 50 Kramer G, Rauch T, Rist W, Vorderwulbecke S, Patzelt H, Schulze-Specking A, Ban N, Deuerling E & Bukau B (2002) L23 protein functions as a chaperone docking... is required for the interaction of the RNA-binding protein Scp160p with polysomes Biochem J 380, 823–830 Gerbasi VR, Weaver CM, Hill S, Friedman DB & Link AJ (2004) Yeast Asc1p and mammalian RACK1 are functionally orthologous core 40S ribosomal proteins that repress gene expression Mol Cell Biol 24, 8276– 8287 Nierhaus KH & Dohme F (1974) Total reconstitution of functionally active 50S ribosomal subunits... Structures of deacylated tRNA mimics bound to the E site of the large ribosomal subunit RNA 9, 1345–1352 45 Valle M, Sengupta J, Swami NK, Grassucci RA, Burkhardt N, Nierhaus KH, Agrawal RK & Frank J (2002) Cryo-EM reveals an active role for aminoacyltRNA in the accommodation process EMBO J 21, 3557–3567 46 Stark H, Rodnina MV, Wieden HJ, Zemlin F, Wintermeyer W & van Heel M (2002) Ribosome interactions of. .. elongation factor Tu in the codon-recognition complex Nat Struct Biol 9, 849–854 47 Schlunzen F, Zarivach R, Harms J, Bashan A, Tocilj A, Albrecht R, Yonath A & Franceschi F (2001) Structural basis for the interaction of antibiotics with the peptidyl transferase centre in eubacteria Nature 413, 814–821 48 Hansen JL, Ippolito JA, Ban N, Nissen P, Moore PB & Steitz TA (2002) The structures of four macrolide... ribosomal protein operon in E coli: nucleotide sequence at the start of the operon Cell 26, 205–211 Yates JL, Arfsten AE & Nomura M (1980) In vitro expression of Escherichia coli ribosomal protein genes: autogenous inhibition of translation Proc Natl Acad Sci USA 77, 1837–1841 Olins PO & Nomura M (1981) Translational regulation by ribosomal protein S8 in Escherichia coli: structural homology between... Stauber RH, Hirschmann N, Bevec D & Hauber J (2000) Human ribosomal protein L5 contains defined nuclear localization and export signals J Biol Chem 275, 12061–12068 54 Leger-Silvestre I, Milkereit P, Ferreira-Cerca S, Saveanu C, Rousselle JC, Choesmel V, Guinefoleau C, Gas N & Gleizes PE (2004) The ribosomal protein Rps15p is required for nuclear exit of the 40S subunit precursors in yeast EMBO J 23, 2336–2347 . connections will be discovered in the future. It therefore may well be that the ribosomal proteins, in fact, are the social life of the ribosome. Acknowledgements We. [21,22]. So the question was really whether the functional effects of these mutations were due to the architectural and sta- bilizing role of the proteins