Pur-alpha functionally interacts with FUS carrying ALS-associated mutations.

Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disorder due to motor neuron loss. Fused in sarcoma (FUS) protein carrying ALS-associated mutations localizes to stress granules and causes their coalescence into larger aggregates. Here we show that Pur-alpha physically interacts with mutated FUS in an RNA-dependent manner. Pur-alpha colocalizes with FUS carrying mutations in stress granules of motoneuronal cells differentiated from induced pluripotent stem cells and that are derived from ALS patients. We observe that both Pur-alpha and mutated FUS upregulate phosphorylation of the translation initiation factor eukaryotic translation initiation factor 2 alpha and consistently inhibit global protein synthesis. In vivo expression of Pur-alpha in different Drosophila tissues significatively exacerbates the neurodegeneration caused by mutated FUS. Conversely, the downregulation of Pur-alpha in neurons expressing mutated FUS significatively improves fly climbing activity. All these findings suggest that Pur-alpha, through the control of mRNA translation, might be involved in the pathogenesis of ALS associated with the mutation of FUS, and that an alteration of protein synthesis may be directly implicated in the disease. Finally, in vivo RNAi-mediated ablation of Pur-alpha produced locomotion defects in Drosophila, indicating a pivotal role for this protein in the motoneuronal function.

Translation factors and ribosomal proteins control tumor onset and progression: how?

Gene expression is shaped by translational control. The modalities and the extent by which translation factors modify gene expression have revealed therapeutic scenarios. For instance, eukaryotic initiation factor (eIF)4E activity is controlled by the signaling cascade of growth factors, and drives tumorigenesis by favoring the translation of specific mRNAs. Highly specific drugs target the activity of eIF4E. Indeed, the antitumor action of mTOR complex 1 (mTORc1) blockers like rapamycin relies on their capability to inhibit eIF4E assembly into functional eIF4F complexes. eIF4E biology, from its inception to recent pharmacological targeting, is proof-of-principle that translational control is druggable. The case for eIF4E is not isolated. The translational machinery is involved in the biology of cancer through many other mechanisms. First, untranslated sequences on mRNAs as well as noncoding RNAs regulate the translational efficiency of mRNAs that are central for tumor progression. Second, other initiation factors like eIF6 show a tumorigenic potential by acting downstream of oncogenic pathways. Third, genetic alterations in components of the translational apparatus underlie an entire class of inherited syndromes known as 'ribosomopathies' that are associated with increased cancer risk. Taken together, data suggest that in spite of their evolutionary conservation and ubiquitous nature, variations in the activity and levels of ribosomal proteins and translation factors generate highly specific effects. Beside, as the structures and biochemical activities of several noncoding RNAs and initiation factors are known, these factors may be amenable to rational pharmacological targeting. The future is to design highly specific drugs targeting the translational apparatus.

Translational control in the stress adaptive response of cancer cells: a novel role for the heat shock protein TRAP1.

TNF receptor-associated protein 1 (TRAP1), the main mitochondrial member of the heat shock protein (HSP) 90 family, is induced in most tumor types and is involved in the regulation of proteostasis in the mitochondria of tumor cells through the control of folding and stability of selective proteins, such as Cyclophilin D and Sorcin. Notably, we have recently demonstrated that TRAP1 also interacts with the regulatory protein particle TBP7 in the endoplasmic reticulum (ER), where it is involved in a further extra-mitochondrial quality control of nuclear-encoded mitochondrial proteins through the regulation of their ubiquitination/degradation. Here we show that TRAP1 is involved in the translational control of cancer cells through an attenuation of global protein synthesis, as evidenced by an inverse correlation between TRAP1 expression and ubiquitination/degradation of nascent stress-protective client proteins. This study demonstrates for the first time that TRAP1 is associated with ribosomes and with several translation factors in colon carcinoma cells and, remarkably, is found co-upregulated with some components of the translational apparatus (eIF4A, eIF4E, eEF1A and eEF1G) in human colorectal cancers, with potential new opportunities for therapeutic intervention in humans. Moreover, TRAP1 regulates the rate of protein synthesis through the eIF2α pathway either under basal conditions or under stress, favoring the activation of GCN2 and PERK kinases, with consequent phosphorylation of eIF2α and attenuation of cap-dependent translation. This enhances the synthesis of selective stress-responsive proteins, such as the transcription factor ATF4 and its downstream effectors BiP/Grp78, and the cystine antiporter system xCT, thereby providing protection against ER stress, oxidative damage and nutrient deprivation. Accordingly, TRAP1 silencing sensitizes cells to apoptosis induced by novel antitumoral drugs that inhibit cap-dependent translation, such as ribavirin or 4EGI-1, and reduces the ability of cells to migrate through the pores of transwell filters. These new findings target the TRAP1 network in the development of novel anti-cancer strategies.

PIM1 kinase is destabilized by ribosomal stress causing inhibition of cell cycle progression.

PIM1 is a constitutively active serine/threonine kinase regulated by cytokines, growth factors and hormones. It has been implicated in the control of cell cycle progression and apoptosis and its overexpression has been associated with various kinds of lymphoid and hematopoietic malignancies. The activity of PIM1 is dependent on the phosphorylation of several targets involved in transcription, cell cycle and apoptosis. We have recently observed that PIM1 interacts with ribosomal protein (RP)S19 and cosediments with ribosomes. Defects in ribosome synthesis (ribosomal stress) have been shown to activate a p53-dependent growth arrest response. To investigate if PIM1 could have a role in the response to ribosomal stress, we induced ribosome synthesis alterations in TF-1 and K562 erythroid cell lines. We found that RP deficiency, induced by RNA interference or treatment with inhibitor of nucleolar functions, causes a drastic destabilization of PIM1. The lower level of PIM1 induces an increase in the cell cycle inhibitor p27(Kip1) and blocks cell proliferation even in the absence of p53. Notably, restoring PIM1 level by transfection causes a recovery of cell growth. Our data indicate that PIM1 may act as a sensor for ribosomal stress independently of or in concert with the known p53-dependent mechanisms.

The myotonic dystrophy type 2 (DM2) gene product zinc finger protein 9 (ZNF9) is associated with sarcomeres and normally localized in DM2 patients' muscles.

AIMS: Myotonic dystrophy type 2 (DM2) is caused by a [CCTG]n intronic expansion in the zinc finger protein 9 (ZNF9) gene. As for DM1, sharing with DM2 a similar phenotype, the pathogenic mutation involves a transcribed but untranslated genomic region, suggesting that RNA toxicity may have a role in the pathogenesis of these multisystem disorders by interfering with common cellular mechanisms. However, haploinsufficiency has been described in DM1 and DM2 animal models, and might contribute to pathogenesis. The aim of the present work was therefore to assess ZNF9 protein expression in rat tissues and in human muscle, and ZNF9 subcellular distribution in normal and DM2 human muscles. METHODS: Polyclonal anti-ZNF9 antibodies were obtained in rabbit, high pressure liquid chromatography-purified, and used for Western blot, standard and confocal immunofluorescence and immunogold labelling electron microscopy on a panel of normal rat tissues and on normal and DM2 human muscles. RESULTS: Western blot analysis showed that ZNF9 is ubiquitously expressed in mammalian tissues, and that its signal is not substantially modified in DM2 muscles. Immunofluorescence studies showed a myofibrillar distribution of ZNF9, and double staining with two non-repetitive epitopes of titin located it in the I bands. This finding was confirmed by the visualization of ZNF9 in close relation with sarcomeric thin filaments by immunogold labelling electron microscopy. ZNF9 distribution was unaltered in DM2 muscle fibres. CONCLUSIONS: ZNF9 is abundantly expressed in human myofibres, where it is located in the sarcomeric I bands, and no modification of this pattern is observed in DM2 muscles.

Transcription inhibitors stimulate translation of 5' TOP mRNAs through activation of S6 kinase and the mTOR/FRAP signalling pathway.

We have analysed the effect of transcription inhibitors on the polysomal localization of 5' terminal oligopyrimidine (TOP-) mRNAs. It is known that, in vertebrates, the translation of this group of mRNAs is regulated according to the growth status of the cell. Mitogenic stimulation of quiescent cells induces a rapid recruitment of TOP mRNAs from translationally inactive light messenger ribonucleoprotein particles to polysomes. It was found that administration of transcription inhibitors to resting cells causes a similar collective translational activation of TOP mRNAs, without affecting global translation. A number of transcription inhibitors were tested in amphibian and mammalian cultured cells. Actinomycin D (act D), cordycepin, and 5, 6-dichloro-1-beta-D-ribofuranosylbenzimidazole caused a similar activation whereas alpha-amanitin or low doses of act D did not induce the translational response. Concentrations of act D sufficient to induce TOP mRNA translation also induce 40S ribosomal protein S6 kinases 1 (S6K1) activation. Moreover at these concentrations of act D increased phosphorylation of 4E-BP1 was also observed, indicating the involvement of FRAP/mTOR. Consistent with this observation, pretreatment of resting cells with rapamycin suppresses the activation of TOP mRNA translation induced by act D. These results indicate that the effect of act D on translation is mediated by the S6Ks through FRAP/mTOR.

La protein has a positive effect on the translation of TOP mRNAs in vivo.

In vertebrates, the mRNAs encoding ribosomal proteins, as well as other proteins implicated in translation, are characterized by a 5'-untranslated region (5'-UTR), including a stretch of pyrimidines at the 5'-end. The 5'-terminal oligopyrimidine (5'-TOP) sequence, which is involved in the growth-dependent translational regulation characteristic of this class of genes (so-called TOP genes), has been shown to specifically bind the La protein in vitro, suggesting that La might be implicated in translational regulation in vivo. In order to substantiate this hypothesis, we have examined the effect of La on TOP mRNA translational control in both stable and transient transfection experiments. In particular we have constructed and analyzed three stably transfected Xenopus cell lines inducible for overexpression of wild-type La or of putative dominant negative mutated forms. Moreover, La-expressing plasmids have been transiently co-transfected together with a plasmid expressing a reporter TOP mRNA in a human cell line. Our results suggest that in vivo La protein plays a positive role in the translation of TOP mRNA. They also suggest that the function of La is to counteract translational repression exerted by a negative factor, possibly cellular nucleic acid binding protein (CNBP), which has been previously shown to bind the 5'-UTR downstream from the 5'-TOP sequence.

Translational control of terminal oligopyrimidine mRNAs requires a specific regulator.

Terminal oligopyrimidine (TOP) mRNAs are a group of messengers translationally regulated according to the growth status of the cell. Two hypotheses have been proposed for the mechanism of the regulation: (i) there is a specific translational regulator which can reversibly alter TOP-mRNA structure, (ii) a component of the general translational apparatus can specifically affect the translation of TOP-mRNAs. To verify one of the two hypotheses we induced a partial inhibition of translation initiation in Xenopus cultured cells and analyzed the effect on TOP-mRNA translation. Our results suggest that a specific regulator is necessary to explain the translational control of these of mRNAs.

Atrial natriuretic factor (ANF) and ANF receptor C gene expression and localization in the respiratory system: effects induced by hypoxia and hemodynamic overload.

Atrial natriuretic factor (ANF) and ANF receptor C (ANF.RC) expression have been investigated in healthy and cardiomyopathic hamsters (CMPH) with widespread necrosis of the diaphragm and myocardium leading to respiratory and heart failure. ANF- and ANF.RC-producing cells were localized in different structures of the respiratory system, and the regulation of their expression by the individual and/or combined action of hypoxia and hemodynamic overload was analyzed. The study was performed in 20-, 90-, and 150-day-old animals using immunohistochemistry, in situ hybridization, Northern blot, and RIA analyses. ANF was shown to be expressed in the tracheo-bronchial epithelium and muscle and, to a lesser extent, in the alveolar wall and muscular media of the pulmonary arteries and extraparenchymal pulmonary veins in both healthy hamsters and CMPH. In 150-day-old CMPH, hypoxia (PaO2 < 50 mm Hg) induced a 10-fold increase in ANF messenger RNA accumulation and a 6-fold increase in the immunoreactive ANF (IR-ANF) concentration in lungs, as quantitated by RIA. As plasma IR-ANF concentrations were elevated in all CMPH age groups, it was most likely produced by the myocardium. ANF.RC messenger RNA was homogeneously distributed throughout the entire respiratory system and was increased 2-fold in hypoxic 150-day-old CMPH only. These results suggest that ANF originating in the respiratory system exerts only paracrine effects on different structures of the respiratory system in addition to the action of circulating ANF. Hemodynamic overload (left ventricular end-diastolic pressure, 17.20 +/- 3.80 mm Hg) might contribute to enhanced ANF gene expression only in extraparenchymal pulmonary vein walls of 150-day-old CMPH. We also propose that ANF.RC overexpression might be a protective mechanism operated via either ANF clearance or inhibition of adenylate cyclase activity to counteract exaggerated smooth muscle relaxation.

Structure and expression of ribosomal protein genes in Xenopus laevis.

In Xenopus laevis, as well as in other vertebrates, ribosomal proteins (r-proteins) are coded by a class of genes that share some organizational and structural features. One of these, also common to genes coding for other proteins involved in the translation apparatus synthesis and function, is the presence within their introns of sequences coding for small nucleolar RNAs. Another feature is the presence of common structures, mainly in the regions surrounding the 5' ends, involved in their coregulated expression. This is attained at various regulatory levels: transcriptional, posttranscriptional, and translational. Particular attention is given here to regulation at the translational level, which has been studied during Xenopus oogenesis and embryogenesis and also during nutritional changes of Xenopus cultured cells. This regulation, which responds to the cellular need for new ribosomes, operates by changing the fraction of rp-mRNA (ribosomal protein mRNA) engaged on polysomes. A typical 5' untranslated region characterizing all vertebrate rp-mRNAs analyzed to date is responsible for this translational behaviour: it is always short and starts with an 8-12 nucleotide polypyrimidine tract. This region binds in vitro some proteins that can represent putative trans-acting factors for this translational regulation.

Xenopus laevis ribosomal protein L22: full-length cDNA sequence and expression analysis.

A cDNA clone was isolated from a Xenopus laevis embryo library and sequenced. Primer extension experiments indicated the full-length nature of the insert and the encoded product was identified on a two dimensional gel as ribosomal protein (r-protein) L22. The 510-bp L22 cDNA sequence presents short untranslated regions and a 5'-end polypyrimidine tract found in all other vertebrate r-protein mRNA (rp mRNA) so far analyzed. Both the nucleotide (nt) and the deduced amino acid (aa) sequences have been compared with the homologous sequences from other species. The L22 nt sequence is about 70% similar to the mammalian L27a rp mRNA and about 60% homologous to the Drosophila, Tetrahymena and yeast corresponding mRNAs. The 148-aa sequence presents a higher conservation, being 90% similar to the mammalian sequence and more than 70% to the other species. Expression analysis showed that, both during X. laevis embryogenesis and in X. laevis cultured cells during growth-rate changes, L22 synthesis is translationally regulated. Therefore X. laevis L22 mRNA is a new example of the correlation between the polypyrimidine terminal tract and the translational regulation observed in other rp mRNAs.

Different forms of U15 snoRNA are encoded in the introns of the ribosomal protein S1 gene of Xenopus laevis.

Recent cloning and sequencing of one of the two Xenopus gene copies (S1b) coding for the ribosomal protein S1 has revealed that its introns III, V and VI carry a region of about 150 nt that shares an identity of 60%. We show here the presence in Xenopus oocytes and cultured cells of a 143-147 nt long RNA species encoded by these three repeated sequences on the same strand as the S1 mRNA and by at least one repeat present in the S1 a copy of the r-protein gene. We identify these RNAs as forms of the small nucleolar RNA U15 (U15 snoRNA) because of their sequence homology with an already described human U15 RNA encoded in the first intron of the human r-protein S3 gene, which is homologous to Xenopus S1. Comparison of the various Xenopus and human U15 RNA forms shows a very high conservation in some regions, but considerable divergence in others. In particular the most conserved sequences include two box C and two box D motifs, typical of most snoRNAs interacting with the nucleolar protein fibrillarin. Adjacent to the two D boxes there are two sequences, 9 and 10 nt in length, which are perfectly complementary to an evolutionary conserved sequence of the 28S rRNA. Modeling the possible secondary structure of Xenopus and human U15 RNAs reveals that, in spite of the noticeable sequence diversity, a high structural conservation in some cases may be maintained by compensatory mutations. We show also that the different Xenopus U15 RNA forms are expressed at comparable levels, localized in the nucleoli and produced by processing of the intronic sequences, as recently described for other snoRNAs.

Vertebrate mRNAs with a 5'-terminal pyrimidine tract are candidates for translational repression in quiescent cells: characterization of the translational cis-regulatory element.

The translation of mammalian ribosomal protein (rp) mRNAs is selectively repressed in nongrowing cells. This response is mediated through a regulatory element residing in the 5' untranslated region of these mRNAs and includes a 5' terminal oligopyrimidine tract (5' TOP). To further characterize the translational cis-regulatory element, we monitored the translational behavior of various endogenous and heterologous mRNAs or hybrid transcripts derived from transfected chimeric genes. The translational efficiency of these mRNAs was assessed in cells that either were growing normally or were growth arrested under various physiological conditions. Our experiments have yielded the following results: (i) the translation of mammalian rp mRNAs is properly regulated in amphibian cells, and likewise, amphibian rp mRNA is regulated in mammalian cells, indicating that all of the elements required for translation control of rp mRNAs are conserved among vertebrate classes; (ii) selective translational control is not confined to rp mRNAs, as mRNAs encoding the naturally occurring ubiquitin-rp fusion protein and elongation factor 1 alpha, which contain a 5' TOP, also conform this mode of regulation; (iii) rat rpP2 mRNA contains only five pyrimidines in its 5' TOP, yet this mRNA is translationally controlled in the same fashion as other rp mRNAs with a 5' TOP of eight or more pyrimidines; (iv) full manifestation of this mode of regulation seems to require both the 5' TOP and sequences immediately downstream; and (v) an intact translational regulatory element from rpL32 mRNA fails to exert its regulatory properties even when preceded by a single A residue.

U17XS8, a small nucleolar RNA with a 12 nt complementarity to 18S rRNA and coded by a sequence repeated in the six introns of Xenopus laevis ribosomal protein S8 gene.

U17XS8 RNA is a 220 nt small RNA coded by a sequence repeated in each of the six introns of the gene for ribosomal protein S8 of Xenopus laevis. It is mainly localized in the nucleolus, as shown by in situ hybridization, and it is assembled in a ribonucleoprotein particle (RNP) sedimenting at about 12S, slightly faster than U3 RNP, and with a density of 1.45 g/ml. DNA and RNA microinjections in Xenopus oocytes have shown that U17XS8 RNA is not the product of an independent transcription unit, but is produced by processing of intron sequences of r-protein S8 transcript, as has been recently shown for other small nucleolar RNAs encoded in the introns of other genes. Its accumulation during Xenopus development, oogenesis and embryogenesis, increases in parallel to that of r-protein S8 mRNA. Another interesting feature is the presence in the U17XS8 RNA of a 12 nt sequence complementary to 18S rRNA. The results presented suggest a possible role of this RNA in some step(s) of ribosome assembling in the nucleolus. Some relevant differences between Xenopus U17XS8 RNA and the corresponding human U17 RNA, recently described, have been observed.

Coordinate translational regulation in the syntheses of elongation factor 1 alpha and ribosomal proteins in Xenopus laevis.

The regulation of the synthesis of elongation factor 1 alpha (EF-1 alpha) in Xenopus laevis has been analyzed from the point of view of translational control. The 5' end of EF-1 alpha mRNA, examined by primer extension, revealed the presence of a terminal pyrimidine tract that is characteristic of ribosomal protein mRNAs (rp-mRNAs). We have then compared the translation pattern of EF-1 alpha and rp-mRNAs during Xenopus embryogenesis and in Xenopus cultured cells during growth rate changes. In Xenopus embryos EF-1 alpha transcripts, that appear after midblastula transition, are initially mostly localized on mRNP and translationally inactive. Only later in embryogenesis, together with rp-mRNAs, they are gradually recruited on polysomes. Also in Xenopus cells B 3.2, EF-1 alpha mRNA shows a distribution change similar to an rp-mRNA: part of it moves from polysomes to mRNP during serum deprivation and goes back on polysomes after restitution of serum to the culture. Moreover EF-1 alpha mRNA, similarly to rp-mRNAs, is always localized on mRNP or fully loaded on polysomes but never on small polysomes. Therefore EF-1 alpha mRNA for structural features and translation behavior can be included in the 'regulatory' group of rp-mRNAs.

Sequence of the gene coding for ribosomal protein S8 of Xenopus laevis.

We present here the cloning and the entire sequence of one of the two gene copies coding for ribosomal protein (r-protein) S8 in Xenopus laevis (corresponding to r-protein S7 in rat) and its flanking regions. The S8a gene contains seven exons and six introns for a total length of about 12,700 bp coding for a mRNA of 663 nucleotides (nt) plus a poly(A) tail. Mapping of the 5' end of the gene has shown that the transcription start point is located in a pyrimidine-rich tract, as has been observed for all r-protein-encoding genes of X. laevis and other vertebrates so far characterized. A computer analysis of the S8a sequence has revealed the presence of a 220-nt sequence repeated, with some variations, once in each of the six introns. RNA analysis by hybridization with oligo probes specific for the two gene copies coding for r-protein S8 has demonstrated that the two of them are expressed at similar levels both in oocytes and in embryos.

Translational regulation of ribosomal protein synthesis in Xenopus cultured cells: mRNA relocation between polysomes and RNP during nutritional shifts.

Translational control of ribosomal protein mRNA was analyzed in a Xenopus cell line during growth-rate changes induced by serum deprivation and readdition. After being transferred into serum-free medium, the cells rapidly decrease their DNA, RNA and protein synthesis, while addition of serum to the culture after a few hours of deprivation causes a rapid recovery. During these growth-rate changes, we observed a shift in ribosomal protein mRNA distribution between polysomes and RNP. The proportion of mRNA on polysomes for the four ribosomal proteins analyzed changed from 70-80% during rapid growth to 25-35% during the downshift and back to 70-80% after the upshift. Northern blot analysis showed that ribosomal protein mRNA level was constant during the shifts even in the presence of the transcriptional inhibitor actinomycin D. This indicates that the distribution changes were due to a reversible transfer of ribosomal protein mRNA between polysomes and RNP without altering mRNA stability. We have also compared the kinetics of ribosomal protein mRNA distribution changes with the kinetics of the changes in the partition of ribosomes between free monomers and polysomes. The results obtained show that the change in ribosomal protein mRNA localization is very fast, allowing short-term adjustments of ribosome synthesis rate. Moreover, our observations are consistent with the hypothesis that the amount of free ribosomes present in the cell could affect ribosomal protein mRNA utilization.

Analysis of mRNAs under translational control during Xenopus embryogenesis: isolation of new ribosomal protein clones.

We have analyzed several randomly selected mRNAs, of the relatively abundant category, on the basis of maternal or zygotic origin and translational efficiency at different developmental stages. For this purpose, clones from a Xenopus embryo cDNA library were hybridized with cDNA probes prepared with poly(A)+RNA from polysomes and from mRNPs of embryos at different stages. The results obtained indicate that the majority of the relatively abundant mRNAs (38 out of 61) is subject to some kind of translational regulation during embryogenesis. Moreover, 30 clones have been selected as corresponding to mRNAs that behave, from the point of view of transcriptional and translational regulation, similarly to previously studied ribosomal protein (r-protein) mRNAs. Sequence analysis of 20 of these selected cDNAs has shown that half of them are in fact homologous to already sequenced r-protein mRNAs. Unexpectedly we have found that also the mRNA for alpha-cardiac actin and another mRNA homologous to creatine kinase M mRNA have a similar translational regulation during embryogenesis.

HrpF, a human sequence-specific DNA-binding protein homologous to XrpFI, a Xenopus laevis oocyte transcription factor.

The identification in HeLa nuclei of a novel DNA-binding protein, designated HrpF, is presented. This factor recognizes and binds a sequence of the Xenopus laevis L14 ribosomal protein (r-p) gene promoter bound by the Xenopus r-p transcription factor I (XrpFI). We show here that XrpFI and HrpF share a conserved DNA-binding domain. We also present evidences suggesting that the two factors perform similar functions in the cell. We discuss the hypothesis that closely related factors might be involved in the control of rp-gene transcription in vertebrates.