REVIEW ARTICLE Synthesis and function of ribosomal proteins – fading models and new perspectives Sara Caldarola, Maria Chiara De Stefano, Francesco Amaldi and Fabrizio Loreni Department of Biology, University ‘Tor Vergata’, Roma, Italy Introduction Ribosomal proteins (RPs) are fundamental compo- nents of ribosomes. They assemble with four rRNA molecules in a complex process that takes place sequentially in the nucleolus, in the nucleoplasm, and in the cytoplasm. Nearly 200 nonribosomal factors are required for the synthesis, maturation and export of the two ribosomal subunits [1]. Most of the constitu- ents of the preribosomal particles have been identified in yeast by exploiting the potent combination of genetic and biochemical approaches [2]. More recently, advances in MS techniques have also led to the identi- fication of the nucleolar proteome in human cells [3]. The role of RPs in the assembly of ribosomes has been studied for many years. Reconstitution experiments in prokaryotes have shown a specific order of addition of RPs for self-assembly of ribosomal subunits [4,5]. The greater complexity in the assembly of the eukaryotic ribosome has until now prevented in vitro reconstitu- tion. However, a recent analysis of the in vivo assembly pathway of the 40S ribosomal subunit showed that the formation of distinct structural intermediates may be similar to what occurs in the prokaryotic counterpart [6]. The structure and function of the ribosome appear to be generally conserved in all organisms. The small subunit (30S or 40S) contains the decoding center, whereas the large subunit (50S or 60S) is responsible for the catalysis of the peptide bond formation, due pri- marily to rRNA. However, the initiation, termination Keywords mTOR signaling; nucleolus; protein synthesis; protein turnover; ribosomal pathology; ribosomal stress; ribosome biogenesis; TOP mRNA; translational control Correspondence F. Loreni, Department of Biology, University ‘Tor Vergata’, Via Ricerca Scientifica, 00133 Roma, Italy Fax: +39 062023500 Tel: +39 0672594317 E-mail: loreni@uniroma2.it (Received 16 February 2009, revised 18 March 2009, accepted 2 April 2009) doi:10.1111/j.1742-4658.2009.07036.x The synthesis of ribosomal proteins (RPs) has long been known to be a process strongly linked to the growth status of the cell. In vertebrates, this coordination is dependent on RP mRNA translational efficiency, which changes according to physiological circumstances. Despite many years of investigation, the trans-acting factors and the signaling pathways involved in this regulation are still elusive. At the same time, however, new tech- niques and classic approaches have opened up new perspectives as regards RP regulation and function. In fact, the proteasome seems to play a crucial and unpredicted role in regulating the availability of RPs for subunit assembly. In addition, the study of human ribosomal pathologies and animal models for these diseases has revealed that perturbation in the syn- thesis and ⁄ or function of an RP activates a p53-dependent stress response. Surprisingly, the effect of the ribosomal stress is more dramatic in specific physiological processes: hemopoiesis in humans, and pigmentation in mice. Moreover, alteration of each RP impacts differently on the development of an organism. Abbreviations Atg7, autophagy-related gene 7; CNBP, cellular nucleic acid-binding protein; DBA, Diamond–Blackfan anemia; E, embryonic day; eIF, eukaryotic initiation factor; mTOR, mammalian target of rapamycin; NEDD8, neural-precursor-cell-expressed developmentally downregulated 8; PI3K, phosphoinositide-3-kinase; RP, ribosomal protein; S6K, S6 kinase; TOP, terminal oligopyrimidine; TSC, tuberous sclerosis complex; TSS, transcription start site; USP10, ubiquitin-specific protease 10; ZNF9, zinc finger protein 9. FEBS Journal 276 (2009) 3199–3210 ª 2009 The Authors Journal compilation ª 2009 FEBS 3199 and recycling phases of translation show differences between prokaryotic and eukaryotic ribosomes [7]. Consistent with this observation, there are differences in the protein composition of the ribosomes from the different kingdoms. Of the 80 mammalian RPs, 49 are related to archeal RPs, and 32 are homologous to bac- terial proteins [8]. The remaining 11 RPs are specific for eukaryotic ribosomes and may be involved in addi- tional particular functions, such as intracellular trans- port. Alternatively, they may be required for the more complex regulation of eukaryotic protein synthesis [9]. The identification of the role of a specific RP is com- plicated by the high level of cooperativity among ribo- somal components and by the fact that the ribosome is essential for the cell. Accordingly, most of the analyzed eukaryotic RPs have been reported as being essential for growth [9,10]. It can be postulated that RPs are required for different steps of ribosome biogenesis and ⁄ or ribosome function. Indeed, a systematic study of the incorporation of RPs into preribosomes led to the identification of the in vivo assembly pathway of the eukaryotic small ribosomal subunit [6]. In some cases, specific RPs have been shown to play a role in ribosomal functions such as interaction with transla- tion initiation factors, translation accuracy, and pep- tide bond formation [9,11]. Although trans-acting factors involved in ribosome biogenesis as well as pre- rRNA processing are well conserved among eukary- otes, the synthesis of RPs appears to be regulated quite differently in yeast and in mammals. In fact, the more than 130 yeast RP genes behave as a precisely coordi- nated transcriptional cluster under a variety of envi- ronmental conditions [12]. This is because almost all RP gene promoters in Saccharomyces cerevisiae con- tain one or two sites for the factor Rap1 [13]. Tran- scriptional activation or repression is obtained through the Rap1-dependent recruitment of different additional factors that combine to determine the correct level of transcription [14]. By contrast, early studies of mam- malian RP gene promoters showed that there are no shared elements, but transcriptional activity is approxi- mately equivalent [15]. More recent in silico analyses found some recurring motifs in the transcriptional con- trol regions [16,17]. However, besides some variation of RP transcript levels in different tissues and in neu- ronal differentiation [18,19], transcriptional regulation does not appear to play a major role in the control of RP synthesis. In fact, it is now well documented that there are signaling pathways that regulate the transla- tional activity of RP mRNA for adjustment of the biosynthesis of ribosomes to the requirements of cell growth and differentiation. In addition, a relevant contribution of protein turnover to the regulation of RP synthesis and accumulation has been proposed by recent studies [20]. Therefore, this review will focus on the different aspects of translational and post-transla- tional regulation of RP metabolism. We will also high- light the role that studies on putative ribosome pathologies have had in our understanding of regula- tory mechanisms of RP synthesis. Translational regulation of RP synthesis Sequence comparison of some vertebrate RP genes cloned in the early 1980s revealed that these genes share a characteristic and distinctive structure of the transcription start site (TSS), which is always posi- tioned within a pyrimidine stretch (about 10–25 nucle- otides long), so that the transcribed mRNAs always start with a C followed by a stretch of 5–15 pyrimi- dines. Later, it was found that this TSS structure char- acterizes all vertebrate RP genes, including all of the 80 human RP genes. This structure is rather peculiar, given that the vast majority of mRNAs start with a purine, most often an A. There are a number of other genes, implicated directly or indirectly in translation, that share this peculiar TSS structure. Among these, we find all the translation elongation factors, but only a few of the numerous translation initiation factors, i.e. eukaryotic initiation factor (eIF) 3e, eIF3f, and eIF3h [21]. The genes whose corresponding mRNAs begin with a5¢-terminal oligopyrimidine (TOP) sequence and are translationally regulated have been named ‘TOP genes’. In fact, external signals, such as stress or the availability of growth factors, hormones, and nutrients, result in the activation of signaling pathways that rap- idly and reversibly modulate the translation of RP mRNAs and the other TOP mRNAs (Fig. 1). It has been observed that, besides the TOP sequence, RP genes are characterized by short UTRs. For instance the 5¢-UTRs of the 80 human RP mRNAs have an average length of 40 nuceotides (range 12–125), which is rather shorter than the average human 5¢-UTR. Even more striking are the 3¢-UTRs, which, in the 80 human RP mRNAs, have an average length of 35 nucleotides, in contrast to almost 1000 nucleotides for the average human 3¢-UTR. To understand the molecular mechanism involved in the growth-associated translational regulation of RP mRNAs, several studies have sought to identify the cis-acting elements and the trans-acting factors that might be responsible for this specific control. Studies on various vertebrate systems (Xenopus, mouse, and human) have amply demonstrated that the Ribosomal protein synthesis S. Caldarola et al. 3200 FEBS Journal 276 (2009) 3199–3210 ª 2009 The Authors Journal compilation ª 2009 FEBS TOP sequence present at the 5¢-end of all RP mRNAs represents the major cis-acting element [22]. However, the 3¢-UTR of RP mRNA may also play a role in translational regulation. In fact, although this short region does not confer translational regulation on a reporter mRNA without the TOP sequence, it does contribute to the stringency of the regulation of a TOP containing RP mRNA [23]. Putative trans-acting factor(s) that might be involved in the growth-dependent translational regulation of RP mRNAs have remained more elusive. In Xenopus , two proteins have been identified, La and cellular nucleic acid-binding protein (CNBP) ⁄ zinc finger protein 9 (ZNF9), which bind the 5¢-UTRs of RP mRNAs in vitro. La interacts with the TOP sequence, whereas CNBP binds a sequence element located closely downstream [24,25]. The mutually exclusive binding of these two proteins on the 5¢-UTRs of TOP mRNAs led Pellizzoni to propose that La may increase translation, whereas CNBP ⁄ ZNF9 could act as a translational repressor. The interaction of La with RP mRNA has also been confirmed in human cells, where La has been shown to exist in two distinct states that differ in sub- cellular localization [26]. When La is phosphorylated on serine 366, it is localized in the nucleus, where it has a role in polymerase III gene transcription. In con- trast, nonphosphorylated La is found in the cytoplasm, where it binds TOP mRNAs. Moreover, immunocom- plex precipitation of La from HeLa cellular extracts yields a number of mRNAs, including TOP mRNAs, thus supporting the conclusion that La protein binds TOP mRNAs in vivo. More recently, it has been shown that La can also be phosphorylated by AKT, which is a component of a signaling pathway involved in TOP mRNA regulation [27] (see below). Several studies have been set up to verify whether La is actually impli- cated in translational regulation. Unfortunately, the results lack coherence, and make it difficult to draw a Fig. 1. Synthesis and turnover of ribosomes. RPs are translated into the cytoplasm and imported into the nucleolus, where they are degraded by proteasomes or assembled with rRNAs into ribosomal subunits. 40S and 60S subunits are then exported into the cytoplasm to form mature ribosomes that are able to initiate translation or are degraded via autophagy, probably through the USP10–G3BP1 complex. S. Caldarola et al. Ribosomal protein synthesis FEBS Journal 276 (2009) 3199–3210 ª 2009 The Authors Journal compilation ª 2009 FEBS 3201 final conclusion. For instance, inducible overexpression of La in stably transfected Xenopus cell lines had a positive effect on translation of RP mRNAs [28]. This is consistent with the positive role of La observed in the internal ribosome entry site-mediated translation of picornaviruses [29]. On the other hand, opposite results were reported by Schwartz et al. [26]. In addition, recent experiments carried out in our laboratory in human cells showed that neither La overexpression nor downregulation by RNA interference had any signifi- cant effect on translation of RP mRNAs (M. C. De Stefano, unpublished results). Finally, the binding of La to a chimeric human TOP containing 5¢-UTR reporter mRNA inhibits its translation in vitro [30]. Similarly, experiments on CNBP ⁄ ZNF9 showed that overexpression of this protein can inhibit the transla- tion of a chimeric TOP–green fluorescent protein mRNA, but that its downregulation by RNA interfer- ence does not interfere with the growth-associated translational activity of TOP messengers (S. Caldarola, unpublished results). A possible explanation for the inconsistencies could be that additional factors contrib- ute to the regulation. For instance, Ro60 is known to interact with La and CNBP ⁄ ZNF9, whereas small RNAs (Y) form a complex with La. If all of these factors play a role in the regulation, the overexpression or downregulation of only one of them could produce apparently contradictory results in different experimen- tal systems and conditions. A different situation is pre- sented in a recent report by Orom et al. These authors indicate microRNA-10a to be a trans-acting element implicated in the translational regulation of RP mRNAs [31]. The pairing of microRNA-10a with the 5¢-UTRs of three RP mRNAs stimulates RP mRNA translation. This mechanism is unusual for microRNAs because, in general, they have a negative effect on mRNA translation by interacting with their 3¢-UTRs [32], and it is not known whether it can be extended to other TOP mRNAs. Signaling pathways to RP mRNA translation As most of the reports addressing signaling consider TOP mRNA as a homogeneous group, in this section we will refer to RP mRNA as TOP mRNA. In the last 15 years, various research groups have studied the sig- nal transduction pathways involved in TOP mRNA translational control. Polysome separation on sucrose gradients, which allows analysis of the polysome ⁄ sub- polysome distribution of a messenger, has been used to monitor the translation efficiency of TOP messengers in different growth conditions. A variety of signals, such as stress or the availability of growth factors, hormones, and nutrients, can induce a change in the percentage of TOP mRNA associated with polysomes from 30–40% to 65–75%, and vice versa [33]. Several lines of evidence converge in indicating phosphoinosi- tide-3-kinase (PI3K) as a key modulator of TOP mRNA translation after mitogenic stimulation [34]. PI3K activates a signaling pathway that includes: phosphoinositide-dependent kinase 1, protein kinase B (also called AKT), tuberous sclerosis complex (TSC)1– TSC2, and the mammalian target of rapamycin (mTOR) C1 complex (composed of raptor, mLst8, and mTOR). The role of TSC1–TSC2 in the translation of TOP mRNAs has been recently investigated by Bilanges et al. [35], using microarray analysis. The authors analyzed the translational efficiency of many cellular messengers in wild-type, TSC1 ) ⁄ ) or TSC2 ) ⁄ ) mouse embryo fibroblasts, during serum starvation and ⁄ or treatment with the mTORC1 inhibitor rapamycin. They found that translation of most TOP mRNAs is regulated by mitogen-induced signal transduction path- ways acting through TSC1–TSC2 and involving mTORC1, as suggested by the rapamycin effect. Rapamycin, which inhibits mTORC1 by binding to mTOR in a complex with the immunophilin FKBP12, has a variable effect on TOP mRNA translation. In HeLa cells, it totally blocks the recruitment of TOP messengers on polysomes during serum stimulation [33]. In other cell lines, however, this inhibitory effect is only partial [34,36]. Recent data from the Meyuhas group indicate that mTOR is indispensable for the translational activation of TOP mRNAs [37]. How- ever, these authors showed that decreasing the expres- sion of the raptor or rictor genes (partners of mTORC1 and mTORC2 respectively) has only a slight effect on the translation efficiency of TOP mRNAs. This result implies that mTOR regulates TOP mRNA translation through a novel rapamycin-insensitive pathway with a minor, if any, contribution of the canonical mTOR complexes mTORC1 and mTORC2. A further downstream target of the PI3K pathway is RPS6, which is phosphorylated after mitogenic stimu- lation by two closely related kinases, S6 kinase (S6K) 1 and S6K2. The strong correlation between the trans- lational activation of TOP mRNAs and the hyper- phosphorylation of RPS6 [36] led to the assumption that RPS6 phosphorylation was necessary for the recruitment of TOP messengers to the polysomes [38]. For years, RPS6 has been considered to be the key protein responsible for the selective translation of TOP mRNAs able to increase the affinity of ribosomes for this class of messengers. However, this model was initially questioned by the observation that in cells Ribosomal protein synthesis S. Caldarola et al. 3202 FEBS Journal 276 (2009) 3199–3210 ª 2009 The Authors Journal compilation ª 2009 FEBS from S6K1 ) ⁄ ) ⁄ S6K2 ) ⁄ ) mice, the translation of TOP mRNAs is still modulated by mitogens in a rapa- mycin-dependent manner [39]. Unexpectedly, RPS6 phosphorylation at serine 235 and serine 236 persisted in the absence of both S6K1 and S6K2, revealing the presence of another S6K, most likely p90 ribosomal S6 kinase. More recently, to abolish any residual phos- phorylation on RPS6, Meyuhas et al. produced a via- ble and fertile knock-in mouse with mutated unphosphorylatable RPS6 (RPS6P ) ⁄ ) ). Mouse embryo fibroblasts isolated from RPS6P ) ⁄ ) mice still show serum-dependent translational activation of TOP mes- sengers. This indicates that complete abrogation of RPS6 phosphorylation does not affect the translation of TOP mRNAs, definitely disproving the model [40] (shown schematically in Fig. 2). Although TOP mRNAs have been generally consid- ered to be a homogeneous group regulated in a coordi- nated way, a recent report from Sonenberg’s laboratory identified a subset of TOP mRNAs whose translation is influenced by eIF4E overexpression [41]. eIF4E is the limiting component of the eIF4F initia- tion complex, and a key player in the regulation of translation in eukaryotic cells. It is thought to enhance the translation of mRNAs with highly structured 5¢-UTRs [42], and to play an important role in cell growth and proliferation [43,44]. Moreover, eIF4E is overexpressed in many kinds of cancer, and its abun- dance is correlated with the progression of malignan- cies [45]. In order to identify messengers regulated by eIF4E, Sonenberg et al. performed a microarray analy- sis of polysome-associated mRNAs from NIH3T3 cells overexpressing eIF4E. They identified messengers cod- ing for proteins involved in cell proliferation (MIF and cenpA), survival (i.e. survivin, BI-1, and dad1), and ribosome biogenesis (members of the small and large ribosomal subunits). Interestingly, not all RP mRNAs respond to eIF4E overexpression, suggesting the existence of subclasses of TOP mRNAs with different regulatory mechanisms. RP turnover Ribosome production is strongly linked to the rate of cellular growth. The construction of ribosomes is among the most energy-consuming events that occur in a cell. A growing HeLa cell synthesizes about 7500 ribosomal subunits per minute, using up some 300 000 RPs, accounting for almost 50% of all cellular proteins in growing cells [46]. Mature ribosomes are very stable complexes, with an estimated half-life of about 5 days for both RPs and rRNA [47]. Several laboratories have tried to understand how ribosomes are recycled and whether there is a specific mechanism of degradation to adjust their number. In a recent report by the Andersen group [20], quantitative analyses of RP trafficking in HeLa cells revealed a prominent role for the protea- some in regulating their turnover. Using fluorescence recovery after photobleaching and MS analysis, the researchers measured the turnover of RPs within the cell, and observed that newly produced RPs accumulate in the nucleolus much faster than do other nucleolar proteins. Despite this, only about one-quarter of the synthesized RPs are assembled into ribosomes and exported to the cytoplasm, a large number of them being degraded via proteasomes. These results indicate that the nuclear export of RPs assembled in ribosomal subunits is slower than the import of free RPs, and that most RPs are produced in excess with respect to the amount needed for ribosome production. Thus, degra- dation of nucleolar RPs could be a general mechanism by which mammalian cells control ribosome produc- tion, adjusting it according to cellular needs. It has been observed that ribosomes are abundantly ubiquiti- nated, suggesting a role of the proteasome in RP turn- over. Ubiquitination occurs on RPS2, RPS3, RPS20 [48], and RPL27a [49]. The last of these modifications, identified in HEK293 cells, is conserved in the yeast homolog L28. RPL27a ubiquitination is reversible and Fig. 2. Signal transduction pathways involved in TOP mRNA trans- lational control. Black arrows indicate activation, and bars indicate inhibition. Gray arrows refer to signaling pathways not involved in TOP mRNA translational control (see text for details). S. Caldarola et al. Ribosomal protein synthesis FEBS Journal 276 (2009) 3199–3210 ª 2009 The Authors Journal compilation ª 2009 FEBS 3203 cell cycle-regulated, and increases the translational effi- ciency of ribosomes, indicating that addition of ubiqu- itin molecules to RPs can also have a nonproteolytic role (as previously shown for histones [52]). In addi- tion, the molecular chaperone Hsp90 has been shown to interact with RPS3 and RPS6, protecting them from ubiquitination and proteasome-dependent degradation [50]. Ubiquitination also has a role in ribosome biogen- esis. In fact, it has been shown that proteasome inhibi- tion alters both rRNA gene transcription and maturation of the 90S preribosome complex; it also leads to the depletion of 18S and 28S [51]. Moreover, ubiquitin molecules on RPs can promote ribosome assembly. In fact, in eukaryotes, RPL40, RPS27a and RPP1 are synthesized as ubiquitin fusions, although the ubiquitin part is then removed by post-translational modification [53,54]. The transient association between ubiquitin and RPs can promote their incorporation into mature ribosomes, and is required for efficient ribosome biogenesis. Another post-translational modifi- cation of RPs has been shown by Hay et al., who, in the search for novel proteins modified by neural-pre- cursor-cell-expressed developmentally downregulated 8 (NEDD8) conjugation, identified 36 RPs from both small and large subunits [55]. NEDD8 is a ubiquitin- like molecule involved in the regulation of protein sta- bility that can modulate cell proliferation and survival. Its best characterized substrates are members of the cullin family of proteins [56]. NEDDylation can have opposite effects on the stability of its molecular targets: it stimulates cullin degradation [57], but increases RP stability. An additional mechanism of ribosome degra- dation that involves autophagy has been characterized in a recent report from the Peter laboratory [58]. Auto- phagy is a highly conserved catabolic mechanism for degrading proteins and organelles such as mitochondria (mitophagy), peroxisomes (pexophagy), and endoplas- mic reticulum (reticulophagy). Kraft et al. have identi- fied, together with nonselective processes, a novel type of selective autophagy that they term ‘ribophagy’, and that occurs in S. cerevisiae upon nutrient starvation. Ribophagy requires an intact autophagy machinery [cells deficient in autophagy-related gene 7 (Atg7) fail to degrade ribosomes] and the ubiquitin protease Ubp3p together with its cofactor Bre5 [whose mamma- lian homologs are ubiquitin-specific protease 10 (USP10) and G3BP respectively]. Ubiquitination plays an important role in this kind of selective autophagy, because ribosomes need to be ubiquitinated in the early steps of ribophagy for the recognition of the autopha- gic membranes. Subsequently, ubiquitin molecules have to be removed for the completion of the autophagic process. Interestingly, even if both ribosomal subunits are degraded by ribophagy, only 60S requires the ubiquitin protease complex ubiquitin-specific protease 3 (Ubp3p)–Bre5p. This ribosome-specific autophagic mechanism could also be involved in regulating the amount of ribosomes according to cellular growth con- ditions, or could act as a quality control mechanism able to remove damaged or wrongly assembled ribo- somes (summarized in Fig. 1). RPs in human pathologies and animal models Ribosome deficiencies due to mutations in the genes coding for RPs or for rRNA have been known for many years in Drosophila and Xenopus [59–61]. In both cases, the main phenotype is slow growth, as expected in the case of protein synthesis impairment. It was quite surprising, therefore, that mutations were identi- fied in the RPS19 gene as being the cause of Dia- mond–Blackfan anemia (DBA) [62]. In fact, this syndrome is characterized principally by defective erythropoiesis associated with a variable degree of growth retardation and malformations. Most RPS19 mutations are whole gene deletions, translocations, or truncating mutations (nonsense or frameshift), suggest- ing that haploinsufficiency is the basis of DBA patho- logy. However, several missense mutations have also been described [63]. The recent finding that mutations in other RPs are also involved in DBA strongly sug- gests that a ribosomal failure is responsible for the clinical phenotype. Among DBA patients, mutations have been found in RPS19 (25%), RPL5 (9%), RPL11 (6%), RPL35a (3%), RPS24 (2%), RPS17 (1%), and RPS7 (< 1%) [62,64–67]. At present, these mutations account for about 50% of DBA cases, and other mutated RPs could therefore be found. Although an additional tissue-specific role for the involved RPs [68,69] cannot be ruled out, the most likely hypothesis is that erythropoiesis is the human developmental pro- cess that is most sensitive to ribosomal defects. Consis- tent with this model, it has been recently shown that 5q- syndrome is caused by a defect in RPS14 [70]. The hematological phenotype of this syndrome (macrocyto- sis, erythroid hypoplasia, increased risk of leukemia) is strikingly similar to DBA, thus confirming the impor- tance of ribosome function in erythropoiesis. Further support for this hypothesis can be obtained by the analysis of other human pathologies, such as dyskera- tosis, cartilage-hair hypoplasia, and Shwachman– Diamond syndrome [71]. All three diseases depend on alterations of some aspects of ribosome biogenesis and, besides specific clinical phenotypes, they all share defective hematopoiesis. Analysis of the molecular Ribosomal protein synthesis S. Caldarola et al. 3204 FEBS Journal 276 (2009) 3199–3210 ª 2009 The Authors Journal compilation ª 2009 FEBS mechanism of DBA in cultured cells showed that alter- ation of any of the involved RPs can affect the matu- ration of rRNA [72]. Moreover, investigations on the effect of mutations on the synthesis of RPS19 showed that: (a) mutations that affect mRNA structure cause a decrease in RPS19 mRNA level [73,74]; and (b) mis- sense mutations affect the stability of the protein more or less severely according to the position within the amino acid sequence [75,76]. An explanation for the hematological phenotype of ribosome pathologies could be that, because erythroid progenitor cells prolif- erate extraordinarily rapidly and need to accumulate high concentrations of globin proteins, they require a high level of ribosome biogenesis. The failure to meet such requirements would trigger apoptosis, possibly through specific mechanisms (ribosomal stress). The several animal models with RP deficiency reported in the literature only partially support this hypothesis. The first alteration of an RP in mice was an inducible deletion of both copies of the RPS6 gene in the liver of adult mice [77]. In this study, the altered response to partial hepatectomy suggested the existence of a novel checkpoint preventing cell cycle progression as a consequence of a defect in ribosome biogenesis. Subse- quently, the same research group showed that genetic inactivation of p53 in RPS6-haploinsufficient mouse embryos bypassed the observed blocking of the cell cycle at gastrulation [embryonic day (E) 5.5]. The res- cued embryos developed until E12.5, when they died with diminished fetal liver erythropoiesis and placental defects [78]. A less severe phenotype was observed in the belly spot and tail mouse mutation, which is a deletion in the RPL24 gene causing a splicing defect. Belly spot and tail homozygotes die before E9.5, but the heterozygotes reach adulthood, although they are smaller than wild-type littermates [79]. More specific phenotypes of Bst ⁄ + mice include alterations in pig- mentation (white ventral midline spot, white hind feet), skeletal abnormalities (kinked tail), and defects in reti- nal development. An even less drastic phenotype is observed in the case of mutations of the RPL29 gene. In fact, mice lacking one of the two alleles develop normally, and even RPL29-null animals are viable. A delay in global growth is, however, observed in null embryos around mid-gestation [80]. This results in pro- portionally smaller organs and smaller stature. In addi- tion, fibroblasts from RPL29-null embryos show decreased rates of proliferation and protein synthesis. Therefore, RPL29 is dispensable for embryonic develop- ment, although ribosomes without this protein may work with reduced efficiency. Alteration of RPL22 also has a mild effect on the organism. Relative to control littermates, RPL22 ) ⁄ ) mice show no evident differ- ences in growth rate and size [81]. RPL22 deficiency, however, selectively arrested development of a specific T-cell lineage by inducing cell death. It is noteworthy that knockdown of p53 blocked cell death and restored thymocyte development. This suggests that, in addi- tion to RPS6, RPL22 deficiency can also activate a p53-dependent checkpoint, albeit, in this case, only in specific cell types. A role of p53 in mediating the effect of RP deficiency was also shown in a recent publica- tion by McGowan et al. [82]. In a chemical mutagenesis screen in mice for pigmental abnormalities, missense alterations of RPS19 and RPS20 were identified in two mutants with dominantly inherited dark skin in ears, footpads, and tail (Dsk3 and Dsk4). In addition, Dsk3 ⁄ + mice showed a slightly reduced erythrocyte level, increased apoptosis of erythroid precursors, and reduced body weight. Pigmentation alteration could be reproduced by conditional deletion of one copy of RPS6 in keratinocytes. All phenotypes (pigmentation, red cells, growth) are dependent on the increase in p53. Hyperpigmentation is therefore due to stimulation of the production of Kit ligand in keratinocytes, which in turn causes melanocytosis. Another mouse knockout model for RPS19 produced results partially in contrast with this last report. In fact, the RPS19 ) ⁄ ) animals die prior to implantation, whereas heterozygous mice have a normal phenotype, including the hematopoietic sys- tem [83]. Finally, interesting new information has also been obtained from zebrafish models. Amsterdam et al. [84] reported that many RP genes may act as tumor suppressors. Moreover, tumors due to RP hap- loinsufficiency show defects in p53 synthesis, suggest- ing that appropriate amounts of RPs are required for p53 protein production in vivo, and that disruption of this regulation could contribute to tumorigenesis [85]. In other studies, RP deficiency was induced by inject- ing antisense oligonucleotide analogs (morpholinos) into one-cell-stage zebrafish embryos. The reduced amounts of RPS19 and several other RPs caused hematopoietic and developmental abnormalities similar to DBA [86,87]. Interestingly RPL11-deficient embryos display abnormalities mostly in the brain [88]. Simi- larly to some mouse models, RP deficiency in zebrafish seems to activate a p53-dependent checkpoint that induces developmental abnormalities [86,88]. The affected tissues, however, could be different according to the RP involved. Vertebrate animal models for RP deficiency are summarized in Table 1. Conclusions In the last few years, the study of the synthesis and function of RPs has both expanded our knowledge S. Caldarola et al. Ribosomal protein synthesis FEBS Journal 276 (2009) 3199–3210 ª 2009 The Authors Journal compilation ª 2009 FEBS 3205 and highlighted the issues that still need to be solved. Translational regulation of RP synthesis associated with the growth status of the cell has been known for more than 20 years. The cis-acting sequences responsi- ble for the regulation were identified in the early 1990s, but the trans-acting factors involved are still unknown, and the few hypotheses proposed remain unconvincing. Further disappointment came from research that disproved the widespread model of the role of S6Ks and phosphorylated RPS6 in TOP mRNA translational regulation. Nevertheless, stimulat- ing results were obtained from the analysis of RP turn- over and investigations into the effects of RP mutations in animal models and human pathologies. A role for protein turnover in RP gene expression was proposed in early studies on ribosome biogenesis [89]. However, the observation that RPs are produced in excess and then rapidly degraded in the nucleolus [20] is surprising. A rationalization of this apparent waste of energy could be that the ribosomes are so important that they justify a certain degree of redundancy in their synthesis. This idea, however, conflicts with the finely tuned regulation at the translational level observed in response to growth factors, nutrient sufficiency, etc. New studies on RP turnover have opened up a scenario of additional regulatory mechanisms in RP Table 1. Vertebrate animal models with RP alterations. RP Organism Alteration Phenotype p53 inhibition References RPS6 Mouse Conditional deletion (liver) Cell cycle block Not done [77] RPS6 Mouse Deletion ) ⁄ +: embryonic lethal Partial rescue [78] RPS19 Mouse Deletion ) ⁄ +: no phenotype ) ⁄ ): lethal Not done [83] RPS19 Zebrafish Knock-down Hematopoietic and developmental abnormalities Rescue [86,87] RPS19, RPS20 Mouse Missense mutations (Dsk3 and Dsk4) Dsk ⁄ +: alteration of pigmentation, erythrocyte development Dsk ⁄ Dsk: lethal Rescue [82] RPL11 Zebrafish Knock-down Brain abnormalities, lethal Rescue [88] RPL22 Mouse Deletion ) ⁄ +: no phenotype ) ⁄ ): viable, defect in alpha–beta T-cells Rescue [81] RPL24 Mouse Missense mutation (Bst) Bst ⁄ +: alteration of pigmentation, skeleton and retinal development Bst ⁄ Bst: lethal Not done [79] RPL29 Mouse Deletion ) ⁄ +: no phenotype ) ⁄ ): viable, mild growth retardation Not done [80] Fig. 3. p53-dependent ribosomal stress. Defects of ribosome biogenesis at any step lead to the activation of p53 and conse- quently to block of the cell cycle or apop- tosis. Red crosses indicate steps that may be affected. Ribosomal protein synthesis S. Caldarola et al. 3206 FEBS Journal 276 (2009) 3199–3210 ª 2009 The Authors Journal compilation ª 2009 FEBS synthesis involving small modifying peptides (ubiquitin and NEDD8) affecting protein stability. The most intriguing recent finding is the phenotype resulting from mutations in RPs in zebrafish, mice, and humans. The studies of mutations in various RPs in the different organisms identified both a common effect and species- specific and RP-specific alterations. The general consequence of RP alteration is the activation of a p53- dependent ‘ribosomal checkpoint’. This is the first response of the cell to a ribosome defect, and consists of blocking of the cell cycle and ⁄ or activation of apoptosis mediated by an increase in p53 levels (Fig. 3). The pre- valent effect downstream of this checkpoint appears to be an alteration of hemopoiesis, especially in humans. The same phenotype is also partially observed in mice; however, here, an alteration of pigmentation seems to prevail. Why erythroid differentiation and melanocyto- sis are more sensitive to ribosome defects remains unclear, and the difference between mice and humans is puzzling. Similarly unexpected are the RP-specific effects observed in mice. The explanation of a supplementary role of a few RPs, although demonstrated in some cases, is not entirely convincing. A more intriguing interpreta- tion is a possible specific functional role of the various RPs within the ribosome, as recently observed in yeast [90]. As a consequence, RPs could be more or less important for ribosome functioning, consistent with the variable impact of mutations in different RPs observed in mice (e.g. RPS6 > RPS19 > RPL22 > RPL29; see also Table 1). A further extension of this hypothesis could be heterogeneity in the composition of the ribo- some, as shown in Ascaris [91], although there is no evi- dence for this in vertebrates. Another possibility that could partially explain the different impacts of muta- tions in diverse RPs is a variable basal level (of both mRNA and ⁄ or protein) in different tissues and ⁄ or species. Despite some evidence for variability in the amounts of RPs in different tissues, this aspect has not yet been thoroughly analyzed. A final remark is that the identification of human pathologies dependent on RP mutations has stimulated interest in this group of basic cell components. This has already helped to step up research in this field, and will hopefully clarify issues that remain unsolved. Acknowledgements We thank V. Iadevaia for the artwork. The financial support of Telethon–Italy (Grant no. GGP07241A to F. Loreni) is gratefully acknowledged. This work was also supported by the Diamond Blackfan Anemia Foundation, Inc. and the Italian Ministry of Univer- sity and Research (FIRB and PRIN grants). References 1 Fatica A & Tollervey D (2002) Making ribosomes. Curr Opin Cell Biol 14, 313–318. 2 Henras AK, Soudet J, Gerus M, Lebaron S, Caizer- gues-Ferrer M, Mougin A & Henry Y (2008) The post- transcriptional steps of eukaryotic ribosome biogenesis. Cell Mol Life Sci 65, 2334–2359. 3 Andersen JS, Lam YW, Leung AK, Ong SE, Lyon CE, Lamond AI & Mann M (2005) Nucleolar proteome dynamics. Nature 433, 77–83. 4 Fahnestock S, Erdmann V & Nomura M (1973) Recon- stitution of 50S ribosomal subunits from protein-free ribonucleic acid. Biochemistry 12, 220–224. 5 Mizushima S & Nomura M (1970) Assembly mapping of 30S ribosomal proteins from E. coli. Nature 226, 1214–1218. 6 Ferreira-Cerca S, Poll G, Kuhn H, Neueder A, Jakob S, Tschochner H & Milkereit P (2007) Analysis of the in vivo assembly pathway of eukaryotic 40S ribosomal proteins. Mol Cell 28, 446–457. 7 Budkevich TV, El’skaya AV & Nierhaus KH (2008) Features of 80S mammalian ribosome and its subunits. Nucleic Acids Res 36, 4736–4744. 8 Wool IG, Chan YL & Gluck A (1995) Structure and evolution of mammalian ribosomal proteins. Biochem Cell Biol 73, 933–947. 9 Dresios J, Panopoulos P & Synetos D (2006) Eukary- otic ribosomal proteins lacking a eubacterial counter- part: important players in ribosomal function. Mol Microbiol 59, 1651–1663. 10 Ferreira-Cerca S, Poll G, Gleizes PE, Tschochner H & Milkereit P (2005) Roles of eukaryotic ribosomal pro- teins in maturation and transport of pre-18S rRNA and ribosome function. Mol Cell 20, 263–275. 11 Valasek L, Mathew AA, Shin BS, Nielsen KH, Szamecz B & Hinnebusch AG (2003) The yeast eIF3 subunits TIF32 ⁄ a, NIP1 ⁄ c, and eIF5 make critical connections with the 40S ribosome in vivo. Genes Dev 17, 786–799. 12 Gasch AP, Spellman PT, Kao CM, Carmel-Harel O, Eisen MB, Storz G, Botstein D & Brown PO (2000) Genomic expression programs in the response of yeast cells to environmental changes. Mol Biol Cell 11, 4241–4257. 13 Lieb JD, Liu X, Botstein D & Brown PO (2001) Pro- moter-specific binding of Rap1 revealed by genome- wide maps of protein–DNA association. Nat Genet 28, 327–334. 14 Hu H & Li X (2007) Transcriptional regulation in eukaryotic ribosomal protein genes. Genomics 90, 421–423. 15 Hariharan N, Kelley DE & Perry RP (1989) Equipotent mouse ribosomal protein promoters have a similar architecture that includes internal sequence elements. Genes Dev 3, 1789–1800. S. Caldarola et al. Ribosomal protein synthesis FEBS Journal 276 (2009) 3199–3210 ª 2009 The Authors Journal compilation ª 2009 FEBS 3207 16 Ishii K, Washio T, Uechi T, Yoshihama M, Kenmochi N & Tomita M (2006) Characteristics and clustering of human ribosomal protein genes. BMC Genomics 7, 37, doi:10.1186/1471-2164-7-37. 17 Perry RP (2005) The architecture of mammalian ribosomal protein promoters. BMC Evol Biol 5, 15, doi:10.1186/1471-2148-5-15. 18 Angelastro JM, To ¨ ro ¨ csik B & Greene LA (2002) Nerve growth factor selectively regulates expression of tran- scripts encoding ribosomal proteins. BMC Neurosci 3, 3, doi:10.1186/1471-2202-3-3. 19 Bortoluzzi S, d’Alessi F, Romualdi C & Danieli GA (2001) Differential expression of genes coding for ribo- somal proteins in different human tissues. Bioinformat- ics 17, 1152–1157. 20 Lam YW, Lamond AI, Mann M & Andersen JS (2007) Analysis of nucleolar protein dynamics reveals the nuclear degradation of ribosomal proteins. Curr Biol 17, 749–760. 21 Iadevaia V, Caldarola S, Tino E, Amaldi F & Loreni F (2008) All translation elongation factors and the e, f, and h subunits of translation initiation factor 3 are encoded by 5¢-terminal oligopyrimidine (TOP) mRNAs. RNA 14, 1730–1736. 22 Avni D, Shama S, Loreni F & Meyuhas O (1994) Ver- tebrate mRNAs with a 5¢-terminal pyrimidine tract are candidates for translational repression in quiescent cells: characterization of the translational cis-regulatory element. Mol Cell Biol 14, 3822–3833. 23 Ledda M, Di Croce M, Bedini B, Wannenes F, Corvaro M, Boyl PP, Caldarola S, Loreni F & Amaldi F (2005) Effect of 3¢UTR length on the translational regulation of 5¢-terminal oligopyrimidine mRNAs. Gene 344, 213–220. 24 Pellizzoni L, Cardinali B, Lin-Marq N, Mercanti D & Pierandrei-Amaldi P (1996) A Xenopus laevis homo- logue of the La autoantigen binds the pyrimidine tract of the 5¢UTR of ribosomal protein mRNAs in vitro: implication of a protein factor in complex formation. J Mol Biol 259, 904–915. 25 Pellizzoni L, Lotti F, Maras B & Pierandrei-Amaldi P (1997) Cellular nucleic acid binding protein binds a con- served region of the 5¢ UTR of Xenopus laevis ribo- somal protein mRNAs. J Mol Biol 267, 264–275. 26 Schwartz EI, Intine RV & Maraia RJ (2004) CK2 is responsible for phosphorylation of human La protein serine-366 and can modulate rpL37 5¢-terminal oligo- pyrimidine mRNA metabolism. Mol Cell Biol 24, 9580– 9591. 27 Brenet F, Socci ND, Sonenberg N & Holland EC (2009) Akt phosphorylation of La regulates specific mRNA translation in glial progenitors. Oncogene 28, 128–139. 28 Crosio C, Boyl PP, Loreni F, Pierandrei-Amaldi P & Amaldi F (2000) La protein has a positive effect on the translation of TOP mRNAs in vivo. Nucleic Acids Res 28, 2927–2934. 29 Meerovitch K, Pelletier J & Sonenberg N (1989) A cellular protein that binds to the 5¢-noncoding region of poliovirus RNA: implications for internal translation initiation. Genes Dev 3, 1026–1034. 30 Zhu J, Hayakawa A, Kakegawa T & Kaspar RL (2001) Binding of the La autoantigen to the 5¢ untranslated region of a chimeric human translation elongation fac- tor 1A reporter mRNA inhibits translation in vitro. Biochim Biophys Acta 1521, 19–29. 31 Orom UA, Nielsen FC & Lund AH (2008) MicroRNA- 10a binds the 5¢UTR of ribosomal protein mRNAs and enhances their translation. Mol Cell 30, 460–471. 32 Pillai RS, Bhattacharyya SN & Filipowicz W (2007) Repression of protein synthesis by miRNAs: how many mechanisms? Trends Cell Biol 17, 118–126. 33 Caldarola S, Amaldi F, Proud CG & Loreni F (2004) Translational regulation of terminal oligopyrimidine mRNAs induced by serum and amino acids involves distinct signaling events. J Biol Chem 279, 13522–13531. 34 Stolovich M, Tang H, Hornstein E, Levy G, Cohen R, Bae SS, Birnbaum MJ & Meyuhas O (2002) Transduc- tion of growth or mitogenic signals into translational activation of TOP mRNAs is fully reliant on the phos- phatidylinositol 3-kinase-mediated pathway but requires neither S6K1 nor rpS6 phosphorylation. Mol Cell Biol 22, 8101–8113. 35 Bilanges B, Argonza-Barrett R, Kolesnichenko M, Skinner C, Nair M, Chen M & Stokoe D (2007) Tuberous sclerosis complex proteins 1 and 2 control serum-dependent translation in a TOP-dependent and -independent manner. Mol Cell Biol 27, 5746–5764. 36 Jefferies HB, Reinhard C, Kozma SC & Thomas G (1994) Rapamycin selectively represses translation of the ‘polypyrimidine tract’ mRNA family. Proc Natl Acad Sci USA 91, 4441–4445. 37 Patursky-Polischuk I, Stolovich-Rain M, Hausner-Han- ochi M, Kasir J, Cybulski N, Avruch J, Ruegg MA, Hall MN & Meyuhas O (2009) The TSC–mTOR path- way mediates translational activation of TOP mRNAs by insulin largely in a raptor- or rictor-independent manner. Mol Cell Biol 29, 640–649. 38 Jefferies HBJ, Fumagalli S, Dennis PB, Reinhard C, Pearson RB & Thomas G (1997) Rapamycin suppresses 5¢TOP mRNA translation through inhibition of p70s6k. EMBO J 16, 3693–3704. 39 Pende M, Um SH, Mieulet V, Sticker M, Goss VL, Mestan J, Mueller M, Fumagalli S, Kozma SC & Thomas G (2004) S6K1() ⁄ )) ⁄ S6K2() ⁄ )) mice exhibit perinatal lethality and rapamycin-sensitive 5¢-terminal oligopyrim- idine mRNA translation and reveal a mitogen-activated protein kinase-dependent S6 kinase pathway. Mol Cell Biol 24, 3112–3124. Ribosomal protein synthesis S. Caldarola et al. 3208 FEBS Journal 276 (2009) 3199–3210 ª 2009 The Authors Journal compilation ª 2009 FEBS [...]... CA, Newburger PE et al (2008) Ribosomal protein L5 and L11 mutations are associated with cleft palate and abnormal thumbs in Diamond–Blackfan anemia patients Am J Hum Genet 83, 76 9–7 80 68 Naora H (1999) Involvement of ribosomal proteins in regulating cell growth and apoptosis: translational modulation or recruitment for extraribosomal activity? Immunol Cell Biol 77, 19 7–2 05 69 Wool IG (1996) Extraribosomal... Rad6-dependent ubiquitination of histone H2B in yeast Science 287, 50 1–5 04 Ribosomal protein synthesis 53 Archibald JM, Teh EM & Keeling PJ (2003) Novel ubiquitin fusion proteins: ribosomal protein P1 and actin J Mol Biol 328, 77 1–7 78 54 Kirschner LS & Stratakis CA (2000) Structure of the human ubiquitin fusion gene Uba80 (RPS27a) and one of its pseudogenes Biochem Biophys Res Commun 270, 110 6–1 110 55 Xirodimas... (2009) Loss of ribosomal protein L11 affects zebrafish embryonic development through a p53dependent apoptotic response PLoS ONE 4, e4152, doi:10.1371/journal.pone.0004152 89 Warner JR (1977) In the absence of ribosomal RNA synthesis, the ribosomal proteins of HeLa cells are synthesized normally and degraded rapidly J Mol Biol 115, 31 5–3 33 90 Komili S, Farny NG, Roth FP & Silver PA (2007) Functional specificity... 10, 60 2–6 10 59 Kongsuwan K, Yu Q, Vincent A, Frisardi MC, Rosbash M, Lengyel JA & Merriam J (1985) A Drosophila Minute gene encodes a ribosomal protein Nature 317, 55 5–5 58 60 Ritossa FM, Atwood KC & Spiegelman S (1966) A molecular explanation of the bobbed mutants of Drosophila as partial deficiencies of ribosomal DNA Genetics 54, 81 9–8 34 61 Wallace H & Birnstiel ML (1966) Ribosomal cistrons and the... Extraribosomal functions of ribosomal proteins Trends Biochem Sci 21, 16 4–1 65 70 Ebert BL, Pretz J, Bosco J, Chang CY, Tamayo P, Galili N, Raza A, Root DE, Attar E, Ellis SR et al (2008) Identification of RPS14 as a 5q- syndrome gene by RNA interference screen Nature 451, 33 5–3 39 71 Liu JM & Ellis SR (2006) Ribosomes and marrow failure: coincidental association or molecular paradigm? Blood 107, 458 3–4 588 72... leads to developmental abnormalities and defective erythropoiesis through activation of p53 protein family Blood 112, 522 8–5 237 87 Uechi T, Nakajima Y, Chakraborty A, Torihara H, Higa S & Kenmochi N (2008) Deficiency of ribosomal protein S19 during early embryogenesis leads to reduction of erythrocytes in a zebrafish model of Diamond– Blackfan anemia Hum Mol Genet 17, 320 4–3 211 88 Chakraborty A, Uechi T,... Diamond–Blackfan anemia Hum Mutat 29, 91 1–9 20 64 Cmejla R, Cmejlova J, Handrkova H, Petrak J & Pospisilova D (2007) Ribosomal protein S17 gene (RPS17) is mutated in Diamond–Blackfan anemia Hum Mutat 28, 117 8–1 182 65 Farrar JE, Nater M, Caywood E, McDevitt MA, Kowalski J, Takemoto CM, Talbot CC Jr, Meltzer P, Esposito D et al (2008) Abnormalities of the large ribosomal subunit protein, Rpl35a, in Diamond–Blackfan... paradigm? Blood 107, 458 3–4 588 72 Dianzani I & Loreni F (2008) Diamond–Blackfan anemia: a ribosomal puzzle Haematologica 93, 160 1–1 604 73 Chatr-Aryamontri A, Angelini M, Garelli E, Tchernia G, Ramenghi U, Dianzani I & Loreni F (2004) Nonsense-mediated and nonstop decay of ribosomal protein S19 mRNA in Diamond–Blackfan anemia Hum Mutat 24, 52 6–5 33 74 Gazda HT, Zhong R, Long L, Niewiadomska E, Lipton JM, Ploszynska... (2008) Study of the effects of proteasome inhibitors on ribosomal protein S19 mutants, identified in patients with Diamond–Blackfan anemia Haematologica 93, 162 7–1 634 77 Volarevic S, Stewart MJ, Ledermann B, Zilberman F, Terracciano L, Montini E, Grompe M, Kozma SC & Thomas G (2000) Proliferation, but not growth, blocked by conditional deletion of 40S ribosomal protein S6 Science 288, 204 5–2 047 78 Panic... DH et al (2004) RNA and protein evidence for haplo-insufficiency in Diamond– Blackfan anaemia patients with RPS19 mutations Br J Haematol 127, 10 5–1 13 75 Angelini M, Cannata S, Mercaldo V, Gibello L, Santoro C, Dianzani I & Loreni F (2007) Missense mutations associated with Diamond–Blackfan anemia affect the assembly of ribosomal protein S19 into the ribosome Hum Mol Genet 16, 172 0–1 727 76 Cretien A, . ARTICLE Synthesis and function of ribosomal proteins – fading models and new perspectives Sara Caldarola, Maria Chiara De Stefano, Francesco Amaldi and Fabrizio. (TSC) 1– TSC2, and the mammalian target of rapamycin (mTOR) C1 complex (composed of raptor, mLst8, and mTOR). The role of TSC1–TSC2 in the translation of TOP