HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 47, 45034-45040, November 22, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/47/45034    most recent M208551200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gueydan, C. Articles by Kruys, V. Search for Related Content PubMed PubMed Citation Articles by Gueydan, C. Articles by Kruys, V. Social Bookmarking                      What's this?

Identification of Ribosomal Proteins Specific to Higher Eukaryotic Organisms^*

Cyril Gueydan^§, Corinne Wauquier, Christelle De Mees^¶, Georges Huez, and Véronique Kruys

From the  Laboratoire de Chimie Biologique and the ^¶ Laboratoire de Biologie du Développement, Institut de Biologie et de Médecine Moléculaires, Université Libre de Bruxelles, 12 rue des Profs. Jeener et Brachet, 6041 Gosselies, Belgium

Received for publication, August 21, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

This report describes the identification of a novel protein named PS1D (Genbank^TM accession number AJ272345), which is composed of an S1-like RNA-binding domain, a (cysteine)×3-(histidine) CCCH-zinc finger, and a very basic carboxyl domain. PS1D is expressed as two isoforms, probably resulting from the alternative splicing of mRNA. The long PS1D isoform differs from the short one by the presence of 48 additional amino acids at its amino-terminal extremity. Analysis of PS1D subcellular distribution by cell fractionation reveals that this protein belongs to the core of the eukaryotic 60S ribosomal subunit. Interestingly, PS1D protein is a highly conserved protein among mammalians as murine, human, and simian PS1D homologues share more than 95% identity. In contrast, no homologous protein is found in lower eukaryotes such as yeast and Caenorhabditis elegans. These observations indicate that PS1D is the first eukaryotic ribosomal protein that is specific to higher eukaryotes.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In recent years the molecular comprehension of the synthesis, structure, and function of the ribosome has progressed tremendously (see Ref. 1 for review). Hence, the three-dimensional structure of the archaeal and bacterial ribosomal subunits determined by x-ray crystallography confirmed that the catalytic activity of ribosome mainly relies on the rRNA, implying that the ribosome is a ribozyme (2-5). The ribosomal proteins (rps)¹ are mostly distributed on the subunit surface, except for the active cleft and the interface between the two subunits. At present, the main function of rps seems to be the stabilization of highly compact rRNA structures by filling the gaps between RNA domains. Most rps interact with RNA. However, whereas some of them are globular, others are composed of globular bodies with flexible extensions devoid of tertiary structure. These extensions penetrate into the interior, filling the gaps between neighboring RNA structures (2).

Despite a high degree of sequence conservation, the ribosomal RNA present in the large ribosomal subunit has undergone a significant increase in size during the course of evolution (Escherichia coli 23 S rRNA, 2904 kb; Yeast 26 S rRNA, 3393 kb; Mouse 28 S rRNA, 4712 kb) (6). Because one of the primary functions of rps is to stabilize the highly compact rRNA structure, it is not surprising to notice that the transition from prokaryotes to lower eukaryotes has also resulted in an increase in the total number of rps (54 in E. coli and 78 in yeast) (7, 8). Although the transition from low to high eukaryotes resulted also in a significant increase in size of 28 S rRNA, mammalian ribosomes count only one additional rp, L28, which is absent in Saccharomyces cerevisiae. However, a homologue of rat L28 is found in Saccharomyces pombe (see "Discussion").

Here we describe the identification and characterization of two protein isoforms present in the large ribosomal subunit of the mammalian ribosome that have no identified equivalent in yeast.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Material-- Enzymes were purchased from Invitrogen and Roche. Oligonucleotides were purchased from Invitrogen and Genset. The synthetic peptide use to immunize rabbits was purchased from Eurogentec. The anti-P0 human serum was purchased from Immunovision. The anti-L28 antibody (kind gift of Dr. D. Nadano) is described in (9).

DNA Constructs-- The pcDNA3.1-PS1D was generated by PCR amplification of the PS1D coding sequence using the following oligonucleotides: CCGCTCGAGATGTCGTCCTGTCGAGTGGATAAG (forward, short isoform); GCCGCTCGAGATGGAGAACTTGCCTGCACTG (forward, long isoform); GCTCTAGATTCCTTGTGCTTCTTCTTGTGCTTC (reverse, both isoforms). The PCR products were digested by XhoI and XbaI and ligated in the pCDNA3.1/Myc-His plasmid (Invitrogen) cut with the same enzymes. The PtrcHis-PS1D was generated by PCR amplification using the following oligonucleotides: CGGGATCCCGAGAACTTGCCTGCACTGTATAC (forward), and GCAAGCTTTTCCTTGTGCTTCTTCTTGTGC (reverse). The PCR product was digested by BamHI and HindIII and ligated in the pTrcHIS2A plasmid (Invitrogen) cut with the same enzymes. The PS1D-GFP constructs were generated by inserting PCR-amplified fragments encoding different domains of the PS1D protein (Fig. 4C) into the pEGFP-C3 vector (Clontech Laboratories, Palo Alto, CA). The following oligonucleotides were used in the PCRs: GCCGCTCGAGATGGAGAACTTGCCTGCACTG (forward, eGFP-PS1D  Nterm, eGFP-PS1D1); CTCGAGCCTGAAGAGGAAGAGGAAAAAG (forward, eGFP-PS1D2); GCTCTAGATTCCTTGTGCTTCTTCTTGTGCTTC (reverse, eGFP-PS1D, eGFP-PS1D2); CCTCTAGATCAGAGAATACTTGGTTCC (reverse, eGFP-PS1D1); CGGGATCCGGCCTGAGACCATGGAGAAC (forward, eGFP-PS1D  Nterm); GCGGATCCTCCTTGTGCTTCTTCTTGTGCT (reverse, eGFP-PS1D Nterm). All of the DNA constructs were verified by DNA sequencing.

RNA Analysis-- Tissue RNA blot was purchased from Clontech and was first hybridized with an antisense PS1D riboprobe. For normalization, the blot was then hybridized with an actin cDNA probe synthesized with a Rediprime kit (Amersham Biosciences).

DNA Transfection and Fluorescence Microscopy-- Cos-7 cells were transfected using the Fugene (Roche) according to the manufacturer's instructions. Fluorescence microscopy was performed as described in (10).

Antibody Preparation and Western Blotting-- Rabbits were immunized with a 15-mer peptide (Fig. 1) coupled to keyhole limpet hemocyanin carrier protein and complemented with complete Freund adjuvant (Difco Laboratories, Detroit, MI). Two boost injections were performed, respectively, 2 and 4 weeks after the first immunization using the same peptide complemented with incomplete Freund adjuvant. Blood was harvested 2 weeks after the last injection and conserved overnight at 4 °C for clotting. Serum was harvested by low-speed centrifugation of the clotted blood. Immunoglobulins were precipitated with ammonium sulfate, and anti-PS1D antibodies were further purified by affinity chromatography on a PS1D-peptide-Sepharose column prepared as described in (11).

Purified anti-PS1D antibodies were dialyzed overnight against phosphate-buffered saline, concentrated to 0.5 mg/ml with polyethylene-glycol, and stored at 20 °C in 50% glycerol, 0.01% azide. This antibody solution was used in Western blot experiments (dilution 1/100) according to a protocol described elsewhere (12).

Cell Fractionation and Ribosome Isolation-- Cytoplasmic and nuclear extracts from NIH3T3 cells were prepared as follows: cells were incubated in lysis buffer (10 mM Hepes, pH 7.6, 60 mM KCl, 1 mM EDTA, 0.075% Nonidet P-40, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride) for 3 min on ice and centrifuged at 300 × g. The cytoplasmic supernatant was harvested. The nuclear fraction was recovered by vigorous vortexing in 20 mM Tris, 400 mM NaCl, 1 mM EDTA, 25% glycerol, incubated on ice for 10 min, and further centrifuged at 13,000 × g for 10 min to pellet genomic DNA. To isolate the ribosomal fraction, the cytoplasmic extract was first centrifuged at 13,000 × g to pellet mitochondria. The supernatant was then layered on top of a 1 M sucrose cushion and centrifuged at 200,000 × g to pellet ribosomes as described by Madjar in (11).

Ribosomal Subunit Fractionation-- Ribosomes were disrupted by incubation in 60 mM EDTA on ice for 30 min. Ribosomes were then layered on top of a 15-30% sucrose gradient (gradient length is 9 cm) and were ultra-centrifuged at 100,000 × g for 17 h in a SW41 rotor. The 30-ml gradient was harvested in 1-ml aliquots, and the OD at 260 nm was determined for each fraction. Ribosomal RNA was extracted by the Trizol method (200 µl of each fraction), and the samples were analyzed by agarose gel electrophoresis. Ribosomal proteins were recovered by ethanol precipitation (90% final, 1 h at 80 °C), dissolved in Laemmli buffer, and analyzed by Western blot.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Identification and Characterization of PS1D-- A cDNA encoding PS1D was isolated in the course of a cDNA library screening to identify murine RNA-binding proteins (12). The cDNA was entirely sequenced and appeared to be identical to EST clone W88172. The largest open reading frame present in the sequence encoded a putative protein of 193 amino acids. An InterProScan of this polypeptide sequence against the Prosite database revealed the presence of three different domains (13). The amino-terminal sequence contains an RNA-binding domain highly homologous to that of bacterial ribosomal protein S1 and which is also found in other RNA-binding proteins (14). A (cysteine)×2-(histidine)-(cysteine) (CCHC)-type zinc finger is present in the middle of the open reading frame (ORF), and two bipartite nuclear localization signals are identified in the basic carboxyl-terminal end of the sequence. Because of the presence of this S1 RNA-binding domain, the putative protein was named PS1D.

To identify the 5' end of PS1D mRNA, PCR reactions were performed with the cDNA library using a forward primer hybridizing in the cloning vector and a reverse primer hybridizing near the 5' end of the first isolated cDNA. The first 24 nucleotides of the longest PCR product obtained with these primers was identical to the cDNA previously isolated, but they contained a 179-nucleotide insertion at position 25. This insertion results in a 42-amino acid lengthening of the ORF at the amino-terminal side (Fig. 1). A Blast comparison (15) of the long and short isolated cDNA against the EST database revealed the existence of two EST populations with 5' ends corresponding to the ends of the long and the short cDNAs. Taken together, these observations indicate that PS1D can be transcribed as two alternatively spliced mRNAs, probably encoding two PS1D isoforms.

View larger version (49K): [in this window] [in a new window]   Fig. 1.   The nucleotide sequence of PS1D cDNA. The predicted amino acid sequence is given under the nucleotide sequence in single-letter code. The underlined sequence is not present in the PS1D mRNA encoding the short isoform. The initiation codons of the long and the short isoforms and the sequence corresponding to a putative poly A signal are boxed. The peptide sequence used for raising antibodies is dot underlined. The putative nuclear bipartite signal sequences are underlined in brackets. The S1 domain is stipple shaded, and the zinc finger sequence is hatch shaded. The sequence corresponding to the PS1D2 mutant is gray shaded. (See Fig. 4C.)

PS1D mRNA Tissue Distribution-- The expression of PS1D mRNA was examined by Northern blot analysis using poly(A)+ RNA isolated from various mouse tissues and PS1D antisense riboprobe (see "Experimental Procedures"). As shown in Fig. 2 (upper panel), mRNA of ~1.5 kb was detected in all tissues with significant differences in abundance. PS1D mRNA mostly accumulates in the liver, brain, and heart. PS1D mRNA expression is moderate in the kidney and testis and very weak in the spleen, lung, and skeletal muscle. Normalization of the Northern blot with an actin probe confirmed the significant variations in PS1D expression between tissues (Fig. 2, lower panel). It should be mentioned that resolution of the Northern blot did not allow the detection of two mRNA species because these differ by only 179 nucleotides. However, reverse transcription-PCR experiments performed on total RNA extracted from NIH3T3 cells confirmed the existence of these two mRNA and suggested that the mRNA encoding the long isoform was the most abundant in this cell line (data not shown).

View larger version (47K): [in this window] [in a new window]   Fig. 2.   Differential expression of PS1D mRNA in various tissues. Multiple tissue Northern blot was hybridized with a PS1D riboprobe (upper panel). After autoradiography, the probe was stripped off and the membrane was rehybridized with an actin cDNA probe (lower panel).

Expression of PS1D Isoforms-- To confirm that PS1D ORFs can be decoded into proteins, the long and short coding sequences were subcloned in the pcDNA3.1 expression vector in phase with a c-myc tag. These DNA constructs were transfected into Cos-7 cells, and the expression of PS1D isoforms was analyzed by Western blot using either 9E10 or anti-PS1D antibodies (see "Experimental Procedures"). As shown in Fig. 3, both constructs lead to the synthesis of proteins detectable with the two antibodies. However, the long and the short isoforms migrated on SDS-PAGE with apparent molecular masses of 40 and 35 kDa, respectively. Taking into account the 3-kDa mass of the myc tag, PS1D proteins are thus resolved as 37- and 32-kDa proteins, the 5-kDa difference between the two isoforms corresponding to the 48-amino-acid sequence separating the two ATG initiation codons (Fig. 1). The expression of PS1D cDNA encoding the long isoform in bacteria leads to the synthesis of a protein migrating at a molecular mass of 42-43 kDa, which is slightly higher than the eukaryotic form (Fig. 3B). The discrepancy between the theoretical masses of PS1D isoforms (26.5 and 21.2 kDa for the long and short isoforms, respectively) and their apparent molecular masses on SDS-PAGE most probably results from the very basic nature of PS1D proteins. Moreover, the higher electrophoretic mobility of the eukaryotic form as compared with the bacterial one suggests that PS1D proteins are post-translationally modified, probably by methylation and/or acetylation of lysine residues. Indeed, such modifications are commonly observed on several ribosomal proteins (16, 17) and are known to increase mobility in SDS gels.

View larger version (29K): [in this window] [in a new window]   Fig. 3.   Expression of epitope-tagged PS1D protein in COS-7 cells and E. coli. Eukaryotic expression vectors encoding the myc-tagged PS1D short or long isoforms were transfected in COS-7 cells by the Fugene method. A, the proteins were detected by Western blot using either the 9E10 anti-myc tag antibody (left panel) or a specific anti-PS1D antibody (right panel). B, expression of epitope-tagged PS1D protein large isoform in COS-7 cells and E. coli. The proteins were detected by Western blot using an anti-PS1D antibody.

PS1D Protein Sub-cellular Localization-- COS-7 cells were transfected with chimeric PS1D/GFP expression plasmids in which a PS1D sequence was inserted in 5' or 3' of the GFP sequence (PS1D/GFP C-ter and PS1D/GFP N-ter, respectively). The localization of PS1D/GFP fusion proteins was determined by fluorescent microscopy. As shown in Fig. 4A, PS1D-GFP C-ter accumulates both in the cytoplasm and the nucleus, predominantly at the level of nucleoli (upper panel, right), whereas the unmodified GFP protein remains in the cytoplasm (upper panel, left). Surprisingly, the PS1D/GFP N-ter fusion protein is only detectable in the nucleus, mostly in the nucleoli (lower panel, left), suggesting that the GFP extension placed at the amino-terminal side of PS1D protein precludes its association to ribosomal subparticles and its transfer to the cytoplasm. Treatment of the cells with saponin and ethidium bromide significantly altered the fluorescent signal in the nucleus except at the level of the nucleoli, further confirming the nucleolar accumulation of PS1D/GFP (lower right).

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Localization of GFP/PS1D fusion proteins in COS-7 cells. A, eukaryotic expression vector encoding different GFP fusion proteins were transfected in COS-7 cells. GFP fluorescence analysis was performed using a standard fluorescein isothiocyanate (FITC) filter. Nuclear localization was confirmed by treating the cells with saponin (0.2%) and ethidium bromide (lower right panel). B, COS-7 cells were transfected with different GFP expression vectors. Two days after transfection, cells were treated with actinomycin-D (1 µg/ml, 3.5 h; middle panels), -amanitin (10 µg/ml, 3.5 h; right panels), or left untreated (left panels). C, schematic representation of the PS1D sequences fused to GFP. Numbers indicate the amino acid positions in the long PS1D isoform peptidic sequence. The S1 domain is stipple shaded.

As ribosomal RNA synthesis and ribosome assembly constitute the main metabolic activities of nucleoli, we determined whether GFP/PS1D nucleolar localization was dependent upon ongoing rRNA transcription. COS-7 cells transfected with PS1D/GFP N-ter construct were treated with actinomycin-D or -amanitin, two transcriptional inhibitors differentially blocking RNA polymerase I activity. Whereas PS1D/GFP nucleolar localization was not altered by RNA polII-inhibiting concentrations of -amanitin (Fig. 4B, compare left and right panels), doses of actinomycin D, known to inhibit RNA polymerase I activity, completely abolished GFP/PS1D accumulation in the nucleoli but not in other parts of the nucleus (Fig. 4B, middle panels). These results indicated that the nucleolar localization of the GFP/PS1D fusion is coupled to rRNA synthesis.

Sequence determinants specifying PS1D nucleolar localization were mapped by analyzing the distribution of truncated forms of PS1D/GFP N-ter protein (Fig. 4C). These experiments revealed that the 82 carboxyl-terminal amino acids fused to GFP (PS1D2) accumulated in the nucleoli as the full-length protein (Fig. 4B, compare left upper and middle panels). In contrast, the PSD11-GFP protein containing only the 159 amino-terminal amino acids remained mostly cytoplasmic (left lower panel). This observation confirmed the presence of a nuclear localization signal in the basic domain of the protein as predicted with the InterProScan algorithm and also demonstrated that the nucleolar accumulation is dependent on an element present in the carboxyl-terminal portion of PS1D.

Nucleolar accumulation of proteins is a common feature of either ribosomal proteins or proteins involved in the maturation of rRNA (18). These latter proteins usually remain localized in the nucleus, whereas the ribosomal proteins are re-imported in the cytoplasm as assembled ribosomal subunits. To determine whether PS1D belongs to one or the other class of proteins, the distribution of native PS1D was analyzed by cell fractionation. NIH3T3 cells were fractionated into cytoplasmic and nuclear extracts, and distribution of PS1D protein was analyzed by Western blot. Moreover, the fractionated extracts were verified by Western blot analysis of nuclear and cytoplasmic proteins, c-myc and IkB, respectively. As shown in Fig. 5A, PS1D is predominantly found in the cytoplasmic fraction. Further fractionation of the cytoplasmic extract by ultracentrifugation at 200,000 × g through a sucrose cushion revealed that PS1D segregated into the same ribosomal pellet as the L28 ribosomal protein (Fig. 5B). Moreover, PS1D proteins remain associated with ribosomes upon high salt wash (600 mM KCl) in the presence or the absence of EDTA (50 mM) (Fig. 5C), indicating that PS1D isoforms, like P0 protein, are core components of the ribosome.

View larger version (33K): [in this window] [in a new window]   Fig. 5.   Subcellular localization of PS1D proteins determined by cell fractionation. A, nuclear and cytoplasmic extracts of NIH3T3 cells were prepared as described under "Experimental Procedures." Nuclear and cytoplasmic extracts corresponding to a same number of cells were resolved by SDS page and submitted to Western blotting using either a polyclonal anti-PS1D antibody, anti-myc (Santa Cruz Biotechnology), or anti-IkB antibodies (Santa Cruz Biotechnology). Arrows indicate the proteins revealed in the Western blots: upper panel, PS1D; middle panel, c-myc; and lower panel, IkB. B, ribosomes were isolated from NIH3T3 cytoplasmic extracts by centrifugation at 200,000 × g through a 1 M sucrose cushion. The ribosomal pellet was dissolved in a volume of Laemmli buffer equivalent to the volume of the supernatant. The same volume of supernatant and ribosomal fractions was analyzed by Western blot using anti-PS1D antibody or an anti-ribosomal L28 antibody. C, ultracentifuged ribosomal pellet was dissolved in high-salt buffer in the presence or absence of EDTA (50 mM), incubated for 30 min on ice, and ultracentrifuged through a 1 M sucrose cushion. PS1D was assayed by Western blot analysis of equal volumes of supernatant and resuspended pellet. As a control, P0 ribosomal protein was analyzed in the different fractions.

To determine whether PS1D belongs to the 40 or 60s ribosomal subunit, the ribosomal pellet was dissociated with EDTA and fractionated on 15-30% linear sucrose gradient. Separation of the large and small ribosomal subunits was monitored by OD measurement at 260 nm (Fig. 6A) and analysis of total RNA in each fraction by agarose electrophoresis (Fig. 6B). Western blot analysis of each fraction with anti-PS1D and anti-P0 antibodies revealed that PS1D isoforms, the same as P0, is exclusively found in fractions containing 60S subunits (Fig. 6C). Altogether, these data indicate that PS1D proteins participate to the formation of the core of the 60 S ribosomal subunit.

View larger version (53K): [in this window] [in a new window]   Fig. 6.   Association of PS1D proteins with the 60S ribosomal subunit. Ribosomes isolated from NIH3T3 cells were dissociated with EDTA and centrifuged on a linear 15-30% sucrose gradient. A, OD at 260 nm of the different fractions after centrifugation. B, RNA from the different fractions was extracted with Trizol and visualized by agarose gel electrophoresis and ethidium bromide staining. The positions of the ribosomal 18 S and 28 S RNA are indicated. C and D, Western blot analysis of the fractions using either the anti-PS1D polyclonal antibody or an anti-ribosomal P0 antibody.

Analysis of PS1D Distribution in Different Species-- A database search was performed to investigate the existence of PS1D in other species. Surprisingly, Blast analysis of the GenBank EMBL database with the PS1D cDNA sequence revealed only the existence of human and simian PS1D homologs (accession numbers AK022506 and AB060897) having high degree of similarity with the murine sequence (data not shown). The predicted protein sequence encoded by these cDNAs (accession numbers Q9NP64 and BAB46900) has more than 96% identity with the murine PS1D long or short isoforms. Furthermore, a Blast search of the human cDNA nucleotide sequence against the human ESTs database revealed a second subset of human cDNA with a shorter 5' region (for example, see accession numbers AW005427, BE858870, and BI160080). The predicted translation of this group of ESTs defines an open reading frame in which the first AUG corresponds to the initiation codon for the short PS1D isoform (data not shown). The human PS1D gene might thus lead to the synthesis of two alternatively spliced transcripts as in mouse.

Interestingly, a Pblast search of PS1D protein against the S. cerevisiae ORF database (19) or the C. elegans wormpep database (release 61; 20,040 sequences) failed to identify any homologous protein, suggesting that the 60 S ribosomal subunit of these organisms does not contain equivalents of PS1D proteins.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

This report describes the identification of a novel protein belonging to the 60 S ribosomal subunit. Indeed, cell fractionation experiments revealed that PS1D is a cytoplasmic protein co-sedimenting with ribosomes upon ultracentrifugation (Fig. 5, A and B). Moreover, PS1D was exclusively found in 60 S-enriched fractions after ultracentrifugation of dissociated ribosomes on 15-30% sucrose gradient (Fig. 6) and remained associated to ribosomes upon high salt wash (Fig. 5C). Sherton and Wool (Ref. 20 and references therein) showed that mammalian ribosomal subunits dissociated in 0.88 M KCl and separated by centrifugation can re-associate to form translationally active ribosomes once they are returned to physiological salt concentration. However, it is clear that increasing the ionic strength to the equivalent of 0.5 M KCl induces the dissociation of a number of accessory factors from ribosomal subunits (21, 22). Altogether, these observations indicate that PS1D is a protein belonging to the core of the 60 S ribosomal subunit. This is further supported by the structural features of PS1D. Like several ribosomal proteins, PS1D contains a very basic domain, a glutamic acid-rich repeat, and a zinc finger (7). In addition, its amino-terminal domain contains a RNA binding motif originally identified in bacterial rp S1 (23). This S1 motif most probably confers to PS1D the RNA binding activity that allowed its identification in our screening experiment. The carboxyl-terminal part of PS1D, which is very basic, confers a nucleolar localization to PS1D/GFP fusion proteins (Fig. 4). This subcellular localization is another common feature of ribosomal proteins that transit to the nucleolus to be assembled with nascent rRNA. Many rps contain a bipartite nuclear localization signal that is characterized by a sequence started by two basic residues followed by a 10-amino-acid non-conserved spacer and an element of five residues, of which three are basic (24). It should be noted that the PS1D carboxyl-terminal domain contains two such motifs that might mediate PS1D nucleolar translocation (see Fig. 1).

In contrast to PS1D/GFP C-ter fusion protein, the PS1D/GFP N-ter protein is sequestered in the nucleus, most probably because of its inability to be assembled into ribosomal subunits.

It should be noticed that analysis of native PS1D distribution by cell fractionation failed to reveal any PS1D protein in the nucleus (Fig. 5A), thereby indicating that in natural conditions the nuclear transit of PS1D proteins is very brief.

Two distinct populations of murine PS1D cDNAs or ESTs can be isolated or found in the databases. These cDNAs differ from each other by a 179-nucleotide-long insertion in PS1D 5' untranslated region (Fig. 1), leading to the synthesis of two isoforms. The long isoform contains a amino-terminal extension of 48 amino acids modifying the S1 motif and is significantly more abundant than the short one in all tissues and cell lines examined (Figs. 5 and 6 and data not shown). At this point, it is not known whether both isoforms can be assembled at different ratios in a 60 S subunit or whether only one of the two isoforms is incorporated in one subunit. In the latter case, ribosomes would be heterogeneous for PS1D protein, most of them containing a PS1D long isoform. So far, only two rps have been shown to be encoded as two isoforms: human S 24, which is encoded by alternatively spliced mRNAs (25) and human S 4, which results from two transcribed genes, one on the X and a second on the Y chromosome (26). Whereas S 24 isoforms differ from each other only by the presence of three additional amino acids at the C terminus of the long isoform, S 4Y and S 4X differ at 19 of 263 positions. Both S4X and S4Y genes are transcribed in human males, and S 4 isoforms appear to be functionally interchangeable (27). PS1D is the first rp so far described that is expressed as two isoforms in several mammalian species. Moreover, the amino-terminal extremity of the two isoforms differ from each other by the presence or the absence of 48 amino acids in the S1 domain. Although the functional importance of these 48 additional amino acids remains to be investigated, we speculate that PS1D is the first mammalian rp described so far that may result in ribosome functional heterogeneity.

Analysis of PS1D mRNA accumulation in different tissues revealed a significant variability of expression. This variability might reflect the abundance of ribosomes in each tissue, because a similar profile of expression was found for another rp mRNA, L28 (data not shown).

So far, 80 rps have been identified in the mammalian ribosome, and 47 belong to the large 60 S subunit. Most of these proteins were first identified and purified in rat by Wool and co-workers (20, 28, 29). The sequence of these proteins was established either by complete edman degradation of overlapping peptides or partial peptide sequencing and subsequent cDNA cloning. Although the reason why they did not identify PS1D is not obvious, it might result from their extraction procedure to isolate ribosomal proteins. Indeed, several protocols have been described to isolate such components, and the one they used in their study might not have been efficient to extract PS1D proteins.

Among the 80 eukaryotic ribosomal proteins identified so far, 31 have no homologues in the prokaryotic ribosome. However, all eukaryotic rps are largely conserved between yeast and rat, two evolutionarily distant eukaryotic species. Indeed, the average identity of related rps from rat and S. cerevisiae is 60%, the range going from 40% for rp P1 to 88% for L41 (7). Until now, only the mammalian L28 protein had no equivalent in the S. cerevisiae protein database. However, a L28 homologue exists in S. pombe (accession number T37749). PS1D is thus the unique rp for which no homologue can be found in lower eukaryotes such as yeast and C. elegans. Indeed, the only proteins presenting similarities to PS1D is the C. elegans protein mog-5, a 1200-amino-acid RNA helicase homologous to the human protein HRH1 and yeast Prp22p. However, the alignment of PS1D with these proteins revealed that these similarities cover only their S1 domains. Moreover, both mog-5 and Prp22P proteins are much larger than PS1D (Prp22p, 1145 amino acids; Rrp5p, 1729 amino acids; PS1D, 241 amino acids), and other features distinguish these proteins from PS1D. Prp22p contains a DEAD helicace and a helicase carboxyl domain not found in PS1D; Rrp5p contains 8 repeats of the S1 motif, whereas PS1D contains only 1 such motif. Finally, both proteins have identified human homologues, HRH1 for Prp22p and BAA11502 for Rrp5p. It is worth mentioning that no PS1D homologue could be found in the translated subset of Xenopus ESTs. However, the lower number of EST entries existing for this species (220,182 sequences released in August 2002) does not allow formal exclusion of the presence of PS1D homologue in that species. Altogether, these data indicate that PS1D is rp specific to higher eukaryotes and that ribosomes from lower and higher eukaryotes display structural differences that might be exploited in the design of new ribosomal antibiotics (30).

It should be mentioned that the primary and secondary structure of 28 S rRNA has been modified in the course of evolution of eukaryotes. Indeed, 28 S rRNA of higher eukaryotes is significantly longer than in lower eukaryotes because of the insertion of 12 distinct domains within an otherwise very conserved 28 S molecule. Furthermore, comparison of 28 S rRNA from two vertebrates, Xenopus and mouse, reveals the extension of only three of these domains (6). This rRNA elongation has been accompanied by the synthesis of additional proteins. Whereas some of these eukaryote-specific rps might be involved in the stabilization of the 28 S rRNA structure, others might be involved in other processes such as translation initiation or translational regulation. The structural analysis of higher eukaryotic ribosomal subunits should provide more information about the role of these rps including PS1D.

	ACKNOWLEDGEMENTS

We thank Drs. Denis Lafontaine and Bruce Beutler for helpful discussions and Dr. D. Christophe and C. Christophe-Hobertus for a plasmid gift and advice for the fluorescence microscopy experiments.

	FOOTNOTES

^* This work was funded by the European Community (Contract QLK3-2000-00721), the Fund for Medical Scientific Research (Belgium, Grant 3.4618.01), and the Actions de Recherches Concertées (Grant 00-05/250).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in this paper has been submitted to the GenBank^TM/EBI Data Bank with accession number(s) AJ272345.

^§ To whom correspondence should be addressed. Tel.: 322-6509806; Fax: 322-6509800; E-mail: cgueydan@ulb.ac.be.

Published, JBC Papers in Press, August 28, 2002, DOI 10.1074/jbc.M208551200

	ABBREVIATIONS

The abbreviations used are: rps, ribosomal proteins; rp, ribosomal protein; EMBL, European Molecular Biology Laboratory; EST, expressed sequence tag; GFP, green fluorescent protein.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				T. Zhang, N. Delestienne, G. Huez, V. Kruys, and C. Gueydan Identification of the sequence determinants mediating the nucleo-cytoplasmic shuttling of TIAR and TIA-1 RNA-binding proteins J. Cell Sci., December 1, 2005; 118(23): 5453 - 5463. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/47/45034    most recent M208551200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gueydan, C. Articles by Kruys, V. Search for Related Content PubMed PubMed Citation Articles by Gueydan, C. Articles by Kruys, V. Social Bookmarking                      What's this?