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

J. Biol. Chem., Vol. 275, Issue 29, 21817-21826, July 21, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/29/21817    most recent M002412200v1 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 Winstall, E. Articles by Sachs, A. B. Search for Related Content PubMed PubMed Citation Articles by Winstall, E. Articles by Sachs, A. B. Social Bookmarking                      What's this?

The Saccharomyces cerevisiae RNA-binding Protein Rbp29 Functions in Cytoplasmic mRNA Metabolism^*

Eric Winstall^§, Martin Sadowski^¶, Uwe Kühn^¶**, Elmar Wahle^¶**, and Alan B. Sachs

From the  Department of Molecular and Cell Biology, University of California, Berkeley, California 94720 and ^¶ Institut für Biochemie, Justus-Liebig-Universität Giessen, 35392 Giessen, Germany

Received for publication, March 22, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Here we report that the Saccharomyces cerevisiae RBP29 (SGN1, YIR001C) gene encodes a 29-kDa cytoplasmic protein that binds to mRNA in vivo. Rbp29p can be co-immunoprecipitated with the poly(A) tail-binding protein Pab1p from crude yeast extracts in a dosage- and RNA-dependent manner. In addition, recombinant Rbp29p binds preferentially to poly(A) with nanomolar binding affinity in vitro. Although RBP29 is not essential for cell viability, its deletion exacerbates the slow growth phenotype of yeast strains harboring mutations in the eIF4G genes TIF4631 and TIF4632. Furthermore, overexpression of RBP29 suppresses the temperature-sensitive growth phenotype of specific tif4631, tif4632, and pab1 alleles. These data suggest that Rbp29p is an mRNA-binding protein that plays a role in modulating the expression of cytoplasmic mRNA.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

From the time of their synthesis until their degradation, RNA polymerase II transcripts are associated with proteins. More than just participating in the packaging of mature or immature mRNAs, RNA-binding proteins are often actively involved in a series of post-transcriptional events, such as pre-mRNA 3'-end processing, pre-mRNA splicing, nucleocytoplasmic mRNA export, mRNA translation, and turnover (for a review, see Ref. 1). Therefore, RNA-binding proteins are thought to play central roles in the post-transcriptional regulation of gene expression.

The functions of RNA-binding proteins depend on domains that bind to RNA and on auxiliary domains involved in protein-protein interactions. One of the most common RNA-binding domains is the RNA recognition motif (RRM).¹ This motif, conserved from yeast to human, consists of 80-100 amino acids encompassing a highly conserved octapeptide known as the ribonucleoprotein consensus sequence (RNP1) and a hexapeptide termed RNP2 (2-4). Crystal structures and NMR studies have revealed that RRMs from different proteins adopt a similar globular tertiary structure in which RNA binding is primarily supported by -sheet surfaces (5-9). Some of the RRM-containing proteins have a basal level of nonspecific binding affinity for RNA, while others, like the poly(A)-binding protein, show high affinity for specific RNA sequences.

One important cellular pathway that relies upon RNA-binding proteins is translation initiation. Translation initiation in eucaryotic cells is directed by a set of protein-protein and protein-RNA interactions that promote the recruitment of the 43 S preinitiation complex to the mRNA (reviewed in Ref. 10). Two key structural elements of most cellular mRNA, the 5' cap and the 3' poly(A) tail, are important determinants of these interactions. The translation initiation factor eIF4E binds to the cap structure, whereas the poly(A) tail-binding protein (Pab1p) is bound to the poly(A) tail. These two proteins interact with the translation initiation factor eIF4G. This complex, along with other factors, is thought to promote the binding of the 43 S preinitiation complex to mRNAs via the physical interaction between eIF4G and the ribosome-associated complex eIF3 (reviewed in Ref. 11).

We identified in the Saccharomyces cerevisiae genome data base a gene (SGN1, YIR001C) encoding an RRM-containing protein that was highly homologous to poly(A)-binding protein 2 (PABP2), a protein that stimulates the polyadenylation reaction in vitro by binding to the poly(A) tail and stabilizing the interaction of poly(A) polymerase with the poly(A) tail (12). Because of our interests in poly(A) tail-binding proteins, we chose to investigate the function of this uncharacterized ORF more closely. Here we report that RBP29 (SGN1, YIR001C)(RNA-binding protein of 29 kDa) is a cytoplasmic protein that binds to poly(A) with high affinity. Although it is a nonessential gene, we have discovered that RBP29 displays genetic interactions with the translation initiation apparatus. These results suggest that Rbp29p functions to stimulate the expression of mRNA, possibly through enhancing the efficiency of translation initiation.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Yeast Techniques-- Yeast strains (listed in Table I) were grown on standard YPD or YM medium supplemented with 2% glucose as the carbon source and the appropriate amino acids and nucleotides (13). Yeast cells were transformed using the lithium acetate method (14), and transformants were selected on YM plates containing the appropriate additives.

RBP29 Subcloning and Chromosomal Disruption-- A 2.5-kilobase pair BclI genomic fragment of chromosome IX containing the complete ORF of YIR001C (RBP29) was introduced into the BamHI site of the plasmid pAS246 (URA3 2µ) or of the plasmid pAS417 (URA3 CEN4) to produce plasmid pRBP29 URA3 2µ (BAS3582) and plasmid pRBP29 URA3 CEN (BAS3583), respectively. To generate a double hemagglutinin-tagged RBP29 fusion gene driven by a GAL1-10 promotor, RBP29 was amplified by PCR using the primer 5'-SalI-P29 (5'-TGATGAGTCGACATGTCCCAGGAAGAGAAAGT; SalI site underlined) and 3'-HindIII-P29 (5'-TCATCAAAGCTTTTATTTCGAATCTTTTTCTT; HindIII site underlined). The SalI-HindIII-digested PCR product was cloned into high copy vector pSEYC68, yielding pHA-RBP29 (BAS3585). Cloning experiments were performed according to established protocols (15).

One chromosomal copy of the RBP29 gene was replaced by the one-step gene transplacement method (16) in the diploid yeast strain YAS2251 (Table I) with the loxP-kanMX-loxP disruption cassette (17) flanked by short sequences homologous to position 90 to 50 upstream of the initiation codon and 40 base pairs following the stop codon of the RBP29-coding region, respectively. Cells possessing the integrated loxP-kanMX-loxP cassette were selected on YPD containing a 200 µg/ml concentration of the aminoglycoside antibiotic G418 (Roche Molecular Biochemicals). Correct integration and disruption of one RBP29 copy in G418-resistant cells was confirmed by PCR. The diploid strain bearing one deleted copy of RBP29 was sporulated, and tetrads were dissected to obtain the haploid strain YAS2252.

The strain YAS2620 carrying a disruption in RBP29, tif4631, and tif4632 was used for the substitution of the wild type allele of TIF4631 for tif4631-459, TIF4632, or tif4632-430 on TRP1CEN plasmids by plasmid shuffling on 5-fluoro-orotic acid (18).

Production of Recombinant Rbp29p and Antibodies-- The RBP29-coding region was amplified by PCR from genomic DNA and subcloned into NdeI-BamHI-digested pGM10 (19), yielding plasmid pGM10-H6RBP29 (BAS3584) to allow the expression of RBP29 under the control of the phage T7 promotor with a peptide tag (Met-Ala-(His)₆) fused to its N terminus. Escherichia coli strain BL21(DE3)pLysS (20) was transformed and grown in 0.5 liters of SOB (15) at 37 °C. After induction with 1 mM isopropyl-D-thiogalactopyranoside at a cell density of A₆₀₀ = 0.8, the culture was incubated for 3 h at 37 °C. Cells were harvested; washed in 20 ml of 50 mM Tris-HCl (pH 8.0), 8% sucrose, 5 mM EDTA; and resuspended in buffer SB (50 mM Tris-HCl (pH 8.0), 300 mM KCl, 10% glycerol, 1 mM PMSF, 1 mM -mercaptoethanol, 0.1 mM MgCl₂, 0.01% Nonidet P-40, 10 µg/ml DNase I, 2 µg/ml leupeptin, 2 µg/ml pepstatin). Cells were lysed by sonication, and the lysate was centrifuged at 20,000 × g for 30 min. The supernatant was mixed with 1 ml of a 50% slurry of Ni²⁺-nitrilotriacetic acid-agarose (Qiagen) equilibrated in buffer A (50 mM Tris-HCl (pH 6.3), 300 mM KCl, 10% glycerol, 1 mM -mercaptoethanol, 0.01% Nonidet P-40) and incubated on a shaker for 3 h. The suspension was packed into a column and washed with 40 ml of buffer SB, and bound protein was step-eluted with buffer A containing 10, 50, 150, and 500 mM imidazole, respectively. Fractions containing a polypeptide of ~35 kDa, eluted between 150 and 500 mM imidazole, were combined, dialyzed against buffer D (20 mM Hepes (pH 7.9), 50 mM KCl, 10% glycerol, 0.01% Nonidet P-40, 1 mM dithiothreitol, 2 µg/ml pepstatin, 2 µg/ml leupeptin, 1 mM PMSF), and loaded onto a Mono S HR 5/5 FPLC column (Amersham Pharmacia Biotech) equilibrated in the same buffer. The column was developed with a 75-column volume gradient of 50-1000 mM KCl in buffer D. Rbp29p eluted at 325-375 mM salt. Approximately 1 mg of pure Rbp29p was recovered.

For immunization of rabbits, Rbp29p was purified under denaturing conditions as described (21). Immunization of two rabbits and preparation of serum (AK62 and AK63) was performed by Eurogentec. Antibodies were affinity-purified on a N-hydroxysuccinimide-activated HiTrap affinity column (Amersham Pharmacia Biotech) according to the manufacturer's instructions. Despite affinity purification, both sera cross-reacted with different additional polypeptides of S. cerevisiae.

Filter Binding Assay-- The poly(A) binding activity of Rbp29p was defined by nitrocellulose filter-binding assays adapted from Ref. 22. The standard reaction contained, in 100 ml, 50 mM Tris-HCl (pH 8.0), 100 mM KCl, 10% glycerol, 0.2 mg/ml bovine serum albumin, 0.01% Nonidet P-40, 1 mM EDTA, 0.5 mM dithiothreitol, 0.5 nM end-labeled A₃₄ or U₃₄. Protein was added to the reaction mixture on ice and incubated for 15 min at room temperature. The nitrocellulose filter was washed with 500 ml of buffer W (50 mM Tris-HCl (pH 8.0), 100 mM KCl, 1 mM EDTA) containing 5 µg/ml E. coli rRNA. 80 µl of the reaction mixture was filtered, and the filter was then washed with 5 µl of ice-cold buffer W. Radioactivity was measured in a liquid scintillation counter. For competition assays, poly(A) was used at 2.5 ng/reaction, and other polynucleotides were added as indicated.

Measurement of Poly(A) Tail Lengths-- The experiment was carried out according to Preker et al. (23).

UV Cross-linking in Vivo-- The preparation of cross-linked RNPs was carried out by modification of a previously described procedure (see Ref. 24 and references therein). Cells were grown in 1300 ml of YPD medium to an A₆₀₀ of 10, harvested by centrifugation for 10 min at 5000 × g at 4 °C, washed with ice-cold phosphate-buffered saline (PBS), and repelleted. The cells were resuspended in 100 ml of ice-cold PBS and distributed between 6-8 Petri dishes (diameter of 6.5 cm). UV cross-linking was performed on ice (2 × 2.5 min, UV Stratalinker 1800, Stratagene). The irradiated and nonirradiated (control) cells were harvested by centrifugation, frozen in liquid nitrogen, and stored at 80 °C. Cells were lysed in 10 ml of lysis buffer (20 mM Tris-HCl, pH 7.4, 1 mM EDTA, 50 mM LiCl, 1% SDS, 1% -mercaptoethanol, 5 µg/ml pepstatin, 5 µg/ml leupeptin, 1 mM PMSF) by extensive vortexing with glass beads (Sigma). After centrifugation for 10 min at 9000 × g, the cleared lysate was mixed with volume 5 M LiCl and heat-denatured for 10 min at 65 °C. The lysate was brought to room temperature in a water bath and again centrifuged for 10 min at 12,500 × g. The supernatant was batch-incubated for 45 min at room temperature with 400 mg of oligo(dT)-Sepharose (Roche Molecular Biochemicals) equilibrated in binding buffer (lysis buffer containing 0.5 M LiCl). The oligo(dT)-Sepharose was then packed into a column. The flow-through fractions were heat-denatured again and reapplied twice to the column. The column was washed with 30 ml of binding buffer. Poly(A)⁺ material was eluted with 10 ml of lysis buffer lacking LiCl. RNA-containing fractions (as judged by ethidium bromide staining) were pooled and brought to 0.5 M LiCl, and the oligo(dT)-Sepharose column chromatography was repeated without batch incubation. Eluted poly(A)⁺ material from the second chromatography step was recovered by ethanol precipitation. One-half of the precipitated material was incubated for 2 h at 37 °C in 100 µl of buffer (10 mM Tris-HCl, pH 7.4, 5 mM CaCl₂, leupeptin, pepstatin, and PMSF as above) with 10 µg of RNase A (Sigma) and 300 units of micrococcal nuclease (MBI Fermentas). After precipitation with 3 volumes of ethanol at 20 °C overnight, proteins were resolved by SDS-polyacrylamide gel electrophoresis. Western blot analysis was performed as described (25). Primary antibodies were diluted 1:1000, and secondary antibodies (Dako) were diluted 1:5000. For reprobing, the nitrocellulose membrane was stripped by 2 × 15-min incubation in 200 mM glycine, pH 2.0, 0.1% SDS, and washed twice with water. For control lanes, yeast cells from 1-ml culture volume (A₆₀₀ = 10) were harvested by centrifugation, washed with 1 ml of PBS, and then resuspended in 1 ml of PBS. Cells were lysed by the addition of 150 µl of 1.85 M NaOH, 7.4% -mercaptoethanol, and incubation for 10 min on ice. Proteins were precipitated with 150 µl of 50% trichloroacetic acid and 10-min incubation on ice. After centrifugation for 2 min in a microcentrifuge, the pellet was washed once with 1.5 ml of cold 80% acetone, air-dried, and resuspended in 50 µl of SDS sample buffer containing 100 mM Tris. 7 µl per lane were used for SDS gel electrophoresis.

Immunofluorescence and Microscopy-- Cells (YAS2252) containing the pHA-RBP29 plasmid (BAS3585) were grown in liquid medium containing 3% galactose (to induce expression of the HA-RBP29 fusion gene regulated by the GAL1-10 promoter) to an A₆₀₀ of 0.5-1.0 at 30 °C. To 25 ml of culture, 2.5 ml of 37% formaldehyde were added, and incubation was continued for 1 h at 30 °C. The cells were put on ice for at least 60 min, harvested, and resuspended in 100 mM Tris-HCl (pH 9.4), 10 mM dithiothreitol. The fixed cells were shaken for 10 min and centrifuged. The pellets were resuspended in 1 ml of buffer A (50 mM potassium phosphate (pH 7.4), 1.2 M Sorbitol), washed, and resuspended in the same buffer containing 0.5 mg/ml Zymolyase 100T (Seigakaku Co.). Following incubation for 30 min at 30 °C, the spheroplasts were centrifuged, resuspended in 1 ml of buffer A, and centrifuged again at room temperature. The pellet was resuspended in 50-200 µl of buffer A, and the viscous suspension was added to poly-L-lysine-coated coverslips and allowed to dry for 5 min. All subsequent steps were carried out in buffer B (1× PBS, 0.2% bovine serum albumin, 0.5% Tween 20) at room temperature. The coverslips were washed four times for 5 min with buffer B and incubated upside down in 20 µl of buffer B containing the purified monoclonal mouse antibody 16B12 (Babco) directed against the HA epitope (dilution was 1:50) for 45 min in a moist chamber. The coverslips were washed as above and overlaid with 20 µl of Texas Red-conjugated antibodies directed against mouse IgG (Vector; dilutions 1:100) for 30 min. The coverslips were washed as above, and 4',6-diamidino-2-phenylindole was added at a concentration of 0.5 µg/ml in buffer B and incubated for 5 min. The coverslips were washed twice as above, and cells were mounted in Moviol.

Texas Red and 4',6-diamidino-2-phenylindole were viewed by standard fluorescence microscopy through a TRITC or UV filter, respectively, on a Nikon Microphot-FXA.

Immunoprecipitation Experiments-- Yeast extracts for immunoprecipitation experiments were prepared using a small scale version of the previously published liquid nitrogen lysis method (26). Briefly, 100 ml of YPD were inoculated with the appropriate yeast strain and grown to an A₆₀₀ of 1.8. Cells were harvested and washed twice in 50 ml of buffer A (30 mM Hepes, pH 7.4, 100 mM KOAc, 2 mM MgOAc, 2 mM dithiothreitol). The resulting cell pellet was weighed and resuspended in volume of buffer A containing 1 mM PMSF. The suspension was frozen by dripping into liquid nitrogen. Frozen yeast were crushed in liquid nitrogen using a pestle and mortar. About 200 mg of frozen yeast powder were transferred to an Eppendorf tube and allowed to thaw on ice. The lysate was cleared by two successive centrifugations of 5 min each at 13,000 rpm in an Eppendorf microcentrifuge at 4 °C. Protein concentration was determined by the method of Bradford (27) using the Bio-Rad Protein Assay reagent (Bio-Rad).

For the experiments described in Fig. 2, the anti-Pab1p monoclonal 1G1 antibody (24) was coupled to protein A-Sepharose beads (Repligen) as described previously (28). 50 µl of a 20% suspension of the antibody-protein A-Sepharose resin were incubated for 1 h at 4 °C with 500 µg of the extract. The beads were then washed five times with 1 ml of PBS containing 0.1% Triton X-100 and 0.01% SDS, and proteins in the immunocomplex were eluted by boiling for 5 min in 30 µl of 2× SDS-polyacrylamide gel electrophoresis loading buffer. 30% of the proteins of the eluate were resolved on a 10% polyacrylamide gel and analyzed by Western blot analysis as described previously (28). The polyclonal antibody used to detect Rbp29p was diluted 1:1000. The monoclonal antibody 1G1 used to detect Pab1p was diluted 1:2000.

For the experiments involving treatment of the immunoprecipitate with micrococcal nuclease, beads were first incubated with yeast extracts as described above. After splitting the bead suspension into two tubes, the beads were washed three times with PBS containing 1.5 mM CaCl₂ (PBS plus CaCl₂). The beads were then incubated 30 min at 4 °C in 200 µl of PBS plus CaCl₂ in the presence or absence of 30 units of micrococcal nuclease (Pharmacia). Following five washes with PBS containing 0.1% Triton X-100 and 0.01% SDS, proteins were eluted from the beads as described above.

[³⁵S]Methionine Incorporation Assays-- This experiment was performed essentially as described previously (29). Briefly, cells were grown at 30 °C in selective YMD medium to an A₆₀₀ of 0.4. Cells were then incubated at 30 °C or 37 °C for an additional 2-h period. 1.5 ml of culture were then mixed with 3.5 ml of prewarmed YMD medium containing 60 ng/ml methionine and 0.5 ng/ml [³⁵S]methionine (cell labeling grade, 1175 Ci/mmol; NEN Life Science Products). The resulting mixture was incubated at 30 or 37 °C for a 10-min period during which 1-ml samples (corresponding to each time point) were removed and mixed with 1 ml of 20% trichloroacetic acid. Trichloroacetic acid solutions were then heated at 95 °C for 20 min, and the protein precipitate was collected on GFC filters (Whatman). Filters were washed with 10 ml of 10% trichloroacetic acid and then with 10 ml of 95% ethanol and counted in Scintiverse (Fisher) scintillation fluid. For each experiment, radioactive counts were corrected for the number of cells present in the labeling mix. Cell counting was done using a hemocytometer.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Rbp29p Is Associated with Polyadenylated mRNA in Vivo-- RBP29 encodes a basic protein of 250 amino acid residues with a predicted molecular mass of 29 kDa. The protein sequence contains one putative RNA-binding domain of the RRM type (4). A BLAST search carried out with the predicted amino acid sequence of Rbp29p over all indexed protein sequences revealed that it was most similar to the bovine PABP2. PABP2 is a mammalian protein that functions as a stimulatory factor for poly(A) polymerase and is involved in poly(A) tail length control in vitro (12, 30) and in vivo (31). Rbp29p has an overall sequence identity of 33% (46% similarity) with bovine PABP2. This is due to strong sequence similarity (43% identity, 62% similarity) in the putative RNA binding domain, whereas the sequence identity outside this region is insignificant (below 20%).

UV cross-linking was used to test whether Rbp29p binds to poly(A)⁺ mRNA in vivo. Irradiation of living yeast cells with UV light results in the covalent cross-linking of a variety of RNA-binding proteins, including Pab1p, Pub1p, Nab2, and Nab3, to polyadenylated RNAs (3, 24, 32, 33). UV cross-linking experiments were performed on wild type or rbp29 cells. RNA was recovered from irradiated or nonirradiated cells and purified over two rounds of selection on oligo(dT) cellulose. Following RNase digestion of the bound material, proteins in this mixture were separated on a gel, transferred to a nitrocellulose membrane, and detected by Western analysis using antibodies directed against Rbp29p (see "Experimental Procedures") or the snRNA-associated protein Snp1 (34). As shown in Fig. 1, Rbp29p was detected in the poly(A)⁺ RNA fraction, while Snp1p was not (compare lanes 2 and 7). In addition, the isolation of Rbp29p with poly(A)⁺ RNA was cross-linking-dependent, since no Rbp29p co-purified with poly(A)⁺ RNA from nonirradiated cells (lane 1). As expected, no signal was detected in the poly(A)⁺ RNA fraction prepared from irradiated rbp29 cells (lane 3). These data show that Rbp29p is associated with mRNA in vivo.

View larger version (20K): [in this window] [in a new window]   Fig. 1.   UV cross-linking of Rbp29p to polyadenylated cellular RNA. UV cross-linking experiments were performed on wild type (wt) or rbp29 () cells. RNA was recovered from irradiated (+) or nonirradiated () cells and purified over two rounds of selection on oligo(dT)-cellulose. Following RNase digestion of the bound material, proteins in this mixture and crude yeast lysates (L) were separated by SDS-polyacrylamide gel electrophoresis, transferred to a nitrocellulose membrane, and detected by Western analysis using antibodies directed against Rbp29p or Snp1p, a snRNA-associated protein. The filled and empty arrowheads indicate the positions of Rbp29p- and Snp1p-specific bands, respectively. The marker position and sizes (in kDa) are indicated on the left.

The majority of the poly(A) tail-binding protein Pab1p associates with cytoplasmic mRNA. To determine whether Rbp29p and Pab1p are bound to the same mRNAs, co-immunoprecipitation experiments were performed using the anti-Pab1p monoclonal antibody 1G1. Fig. 2 shows that immunoprecipitation of Pab1p resulted in the co-immunoprecipitation of Rbp29p (Fig. 2A, lanes 1 and 2). Importantly, higher levels of Rbp29p in extracts resulted in a concomitant increase in the amount of Rbp29p co-immunoprecipitating with Pab1p (Fig. 2A, lanes 4 and 5). The specificity of these assays was confirmed with the observation that Rbp29p was not detected in immunoprecipitates from extracts containing overexpressed Rbp29p and a truncated Pab1 protein (pab1-RRM2) that fails to interact with the Pab1p antibody (35) (Fig. 2A, lane 3). Treatment of Pab1p immunocomplexes with micrococcal nuclease prior to their elution resulted in nearly complete loss of Rbp29p (Fig. 2B, compare lanes 1 and 2 with lanes 4 and 5). This indicates that the association of Rbp29p with Pab1p requires an RNA component. Overall, these data support the hypothesis that Pab1p and Rbp29p can simultaneously bind to the same mRNA.

View larger version (37K): [in this window] [in a new window]   Fig. 2.   Co-immunoprecipitation of Rbp29p with Pab1p is dosage- and RNA-dependent. A, co-immunoprecipitation of Rbp29p and Pab1p. Extracts prepared from strains containing plasmids bearing the indicated genes were immunoprecipitated with Pab1p monoclonal antibodies and resolved by SDS-polyacrylamide gel electrophoresis. Pab1p and Rbp29p in 3% of the extract and in 30% of the immunoprecipitate (IP) were detected by Western analysis with Pab1p- or Rbp29p-specific antibodies. The positions of the Rbp29p and Pab1p proteins on the blots are indicated. Strains used, lane 1, YAS1956; lane 2, YAS2136; lane 3, YAS2629; lane 4, YAS2630; lane 5, YAS2631; lane 6, YAS2620. B, RNA-dependent interaction between Pab1p and Rbp29p. Pab1p immunoprecipitates from yeast bearing the indicated genes were incubated with (+ nuclease) or without micrococcal nuclease (see "Experimental Procedures") prior to elution from the Protein A-Sepharose beads. Proteins were resolved by SDS-polyacrylamide gel electrophoresis, and Rbp29p was detected by Western analysis.

Rbp29p Is Localized in the Cytoplasm-- The co-immunoprecipitation of Rbp29p and Pab1p suggested that Rbp29p is a cytoplasmic protein. To assess the intracellular localization and distribution of Rbp29p more directly, fluorescence microscopy analyses were performed. A yeast strain expressing a gene fusion encoding an N-terminal hemagglutinin-tagged Rbp29p (HA-RBP29) driven by the GAL1-10 promoter on a high copy plasmid was examined. The HA-Rbp29p was strongly overproduced when compared with the level of wild type Rbp29p (data not shown). Indirect immunofluorescence microscopy revealed that HA-Rbp29p was localized in the cytoplasm (Fig. 3). Cells containing the RBP29-HA fusion gene did not react with anti-HA antibody when synthesis of the fusion protein was repressed by growth on glucose (data not shown). Similar localization data were obtained using Rbp29-green fluorescent protein and Rbp29-protein A fusions (data not shown). Furthermore, cell fractionation studies revealed that Rbp29p, Hxk1p (hexokinase 1), and Cmd1p (calmodulin 1) were found exclusively in the cytosolic fraction, whereas Pap1p (poly(A) polymerase) was detected both in the nuclear and the cytosolic fraction (breakage or leakage of nuclei can account for Pap1p in the cytoplasmic fraction) (data not shown). These immunofluorescence and cell fractionation data suggest that Rbp29p is a cytoplasmic protein. In contrast, mammalian PABP2 is localized primarily in the nucleus (36, 37).

View larger version (86K): [in this window] [in a new window]   Fig. 3.   Rbp29p is localized in the cytoplasm of cells. Cells expressing a gene fusion encoding an N-terminal hemagglutinin-tagged Rbp29p (HA-RBP29) driven by the GAL1-10 promotor were grown in liquid medium containing 3% galactose. Following fixation and permeabilization, the cells were incubated with the antibody 16B12 directed against the HA epitope as described under "Experimental Procedures." A, the HA-Rbp29 fusion protein was detected using a Texas Red-conjugated anti-mouse IgG antibody. B, simultaneous staining with 4',6-diamidino-2-phenylindole was utilized to localize the DNA found in the nucleus and in the mitochondria.

Rbp29p Binds Preferentially to Poly(A) RNA with High Affinity-- The in vitro RNA binding properties of Rbp29p were examined. Rbp29p was expressed in E. coli as a fusion gene with an N-terminal Met-Ala-His₆ tag. A two-step purification by Ni²⁺-nitrilotriacetic acid chromatography followed by anion exchange chromatography (see "Experimental Procedures") resulted in a homogeneous preparation (data not shown). Mass spectroscopy of several column fractions gave molecular masses of 29,781 and 29,831 Da, in reasonable agreement with the expected molecular mass of 29,791 Da for the tagged protein (data not shown). Western analysis using the anti-Rbp29p antibody and known amounts of purified recombinant Rbp29p allowed us to approximate that Rbp29p is present at about 0.03% of total proteins in an S30 yeast lysate (data not shown). This is about 15 times less than the amount of Pab1p, which is estimated to be present at about 0.5% of total proteins in an S30 lysate (38).

Nitrocellulose filter-binding assays with purified recombinant Rbp29p were performed. The homopolymers A₃₄ and U₃₄ were used as ligands. As shown in Fig. 4A, in the presence of 100 mM KCl, recombinant Rbp29p bound most tightly to poly(A) (apparent K_D = 1 nM). The affinity for poly(U) was lower by approximately 500-fold. With decreasing salt concentration, Rbp29 binding to RNA became progressively nonspecific (data not shown).

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Rbp29p exhibits specific, high affinity binding to poly(A). A, binding of Rbp29p to RNA homopolymers. Nitrocellulose filter-binding assays were carried out with the indicated amounts of Rbp29p and radiolabeled RNA as described under "Experimental Procedures." RNA homopolymers were as follows: A34 () and U34 (). B, specificity of poly(A) binding by Rbp29p. Rbp29p (1.5 nM) was incubated with radiolabeled poly(A) (2.5 ng) and unlabeled competitor RNA present at the excess over labeled poly(A) as indicated (0.4, 1, 4, 20, 40, 200). Rbp29p-poly(A) complexes were detected by nitrocellulose filter binding as described under "Experimental Procedures." Competitors were A34 (), C34 (), G34 (), dA (), and rRNA ().

The binding specificity of Rbp29p for poly(A) was confirmed by competition experiments using A₃₄ as the radiolabeled ligand (Fig. 4B). E. coli ribosomal RNA and poly(G) were relatively efficient competitors, with binding to poly(A) being reduced to 50% at a 5-10-fold excess of competitor. In contrast, a 200-fold excess of poly(C) as competitor reduced binding of radiolabeled poly(A) by only 20%. As a control, unlabeled poly(A) competed for binding approximately as expected. Poly(dA) caused no detectable reduction in poly(A) binding even at a 200-fold excess. Therefore, Rbp29p has no significant affinity for single-stranded DNA. Taken together with the data shown in Fig. 4A, these results show that Rbp29p is an RNA-binding protein that has a high affinity for poly(A) in vitro.

Deletion of RBP29 Does Not Change Poly(A) Tail Length Distributions in Vivo-- The significant homology of Rbp29p to PABP2 and its preference for binding to poly(A) RNA prompted us to investigate whether it functioned in some aspect of poly(A) tail metabolism. Temperature-sensitive mutations in yeast genes encoding proteins involved in mRNA 3'-end processing result in a rapid loss of polyadenylated RNA at the restrictive temperature in vivo (23, 39-42). In contrast, mutations in Pab1p and the Pab1p-dependent poly(A) nuclease, consisting of the subunits Pan2p and Pan3p, result in abnormally long poly(A) tails in vivo (43-45) and in vitro (40, 45, 46). We thus investigated whether deletion of RBP29 resulted in aberrant poly(A) tail lengths on mRNA.

Total RNA prepared from exponentially growing wild type and rbp29 cells was labeled at the 3'-end and subjected to RNase A and T1 digestion, which leaves only the poly(A) tails intact. As shown in Fig. 5, poly(A) tails on mRNA from wild type (lane 3) and rbp29 cells (lane 4) reached the same maximal length of approximately 70 nucleotides, which corresponds to the tail lengths previously reported (47). In addition, rbp29 cells showed no significant differences in the pattern of the poly(A) tail length distribution as compared with wild type cells. Thus, the absence of Rbp29p does not have a detectable influence on the poly(A) tail length of mRNA in vivo. In addition, extracts prepared from wild type or rbp29 cells were found to cleave and polyadenylate a pre-mRNA substrate with the same efficiency (data not shown). These data strongly suggest that Rbp29p is not involved in the 3'-end processing or subsequent poly(A) metabolism of mRNA. Since RBP29 is the only ORF of S. cerevisiae with significant homology to PABP2, these results may indicate that yeast does not have a protein that functions as PABP2.

View larger version (23K): [in this window] [in a new window]   Fig. 5.   Deletion of Rbp29p does not affect mRNA poly(A) tail lengths in vivo. Poly(A) tails on mRNA derived from either wild type (RBP29) or mutant (rbp29) yeast strains were 3'-end-labeled with [32P]cordycepin, resolved on a polyacrylamide gel, and visualized by autoradiography (see "Experimental Procedures"). A70, size-selected A65-75; L3pre(A)40, polyadenylated in vitro transcribed mRNA that was subject to the same treatments as the total RNA. The marker position and sizes (in nucleotides) are indicated on the left.

RBP29 Overexpression Suppresses the Temperature Sensitivity of Various tif4631, tif4632, and pab1 Alleles-- Although RBP29 encodes a cytoplasmic RNA-binding protein, it is a nonessential gene whose deletion results in very little change in the growth characteristics of the yeast. A yeast strain that depends upon Rbp29p for normal growth would be invaluable for future studies on Rbp29p, since it would allow for an in vivo analysis of its function. Toward this goal, we pursued our observation that overexpression of RBP29 leads to scorable phenotypes in different yeast translation mutants.

S. cerevisiae contains two homologs of eIF4G that are encoded by the genes TIF4631 (eIF4G1) and TIF4632 (eIF4G2) (48). We have previously described an extensive analysis of the in vitro and in vivo consequences of mutations in the eIF4E-binding domain of these two eIF4G homologs (49-51). Briefly, replacement of two highly conserved leucine residues with alanine within the eIF4E-binding region of either eIF4G1 (tif4631-459) or eIF4G2 (tif4632-430) severely impairs the association of eIF4E with eIF4G in vitro. Moreover, yeast strains carrying the tif4631-459 or the tif4631-430 allele as their sole source of eIF4G display a temperature-sensitive growth phenotype at 37 °C that is suppressed by increasing the expression of eIF4E or by removing Caf20p, a negative regulator of eIF4E (50).

We found that overexpression of RBP29 on a yeast 2µ plasmid also suppressed the temperature-sensitive phenotype of the tif4631-459 strain at 37 °C (Fig. 6A). This suppression, although easily scored, was much weaker than that observed when eIF4E is overexpressed or when CAF20 is deleted (data not shown). To test whether suppression by excess Rbp29p was specific for the tif4631-459 mutation, yeast strains bearing temperature-sensitive alleles of different translation initiation factors were transformed with RBP29 on a 2µ plasmid and placed at 37 °C. As seen in Fig. 6A and Table II, suppression of temperature sensitivity by excess RBP29 was observed in strains carrying alleles of eIF4G1 (tif4631-427) and eIF4G2 (tif4632-430) that are deficient in binding to eIF4E (50, 52). In contrast, excess RBP29 did not suppress the temperature sensitivity of conditional alleles of TIF4632 (tif4632-1, tif4632-6, and tif4632-8) that are defective in eIF4A binding (53), of the tif4631-454 allele (49) that more fully destroys eIF4E binding to eIF4G1, or of the eIF4E temperature-sensitive allele cdc33-1 (54). Excess Rbp29p also did not suppress the temperature sensitivity of a tif1-1 strain, which contains a mutation in the eIF4A protein (55). These data suggest that RBP29-mediated suppression of temperature sensitivity is an allele-specific effect.

View larger version (66K): [in this window] [in a new window]   Fig. 6.   Excess RBP29 suppresses the temperature sensitivity of tif4631, tif4632, and pab1 mutants. A multicopy plasmid with no insert (empty 2µ) or the yeast RBP29 gene on either a multicopy (2µ) or a centromeric (CEN) plasmid were transformed into the tif4631-459 (YAS2074) and tif4632-430 (YAS2002) mutants (A) and into the pab1-F364L (YAS120) mutant (B). A photograph of the minimal medium plates after 7 days at 30 or 37 °C is shown.

We have previously shown that a loss-of-function allele of the PAB1 gene (pab1-16) is synthetically lethal with the eIF4E mutation cdc33-1. This suggested that Pab1p and eIF4E have functional redundancy (49). Given our observation that overexpression of RBP29 was able to rescue the temperature sensitivity of specific tif4631 and tif4632 mutations that result in poor binding to eIF4E, we decided to test if excess RBP29 also suppressed the conditional growth defect of a pab1 mutant.

RBP29 on a 2µ vector was introduced into a strain carrying the pab1-F364L allele. This gene encodes a Pab1 protein that lacks its first three RRMs and has a phenylalanine to leucine substitution in the remaining RRM4 (2). Importantly, this mutant form of Pab1p does not contain the RRM2 domain, which is required for the functional interaction of Pab1p with eIF4G (35) and for the Pab1p-mediated transactivation of cap-dependent translation in vitro (26). This form of Pab1p also confers a temperature-sensitive growth phenotype to cells (2). As shown in Fig. 6B, excess RBP29 suppressed the temperature-sensitive phenotype of the pab1-F364L strain. Incubation at the restrictive temperature of serial dilutions of midlog cultures of the pab1-F364L strain bearing RBP29 on a 2µ plasmid confirmed that the observed suppression was due to the presence of excess RBP29 and not the appearance of spontaneous suppressors of pab1-F364L (data not shown).

Together with the ability of RBP29 to suppress the temperature-sensitive phenotype of the tif4631 and tif4632 mutants, our observation that RBP29 overexpression suppresses the temperature-sensitive phenotype of pab1-F364L provides a simple assay to elucidate the functional domains of Rbp29p in the future. It also suggests strongly that Rbp29p is involved in controlling the expression of cytoplasmic mRNA, since its overexpression allows for cell growth in the presence of an inadequate translation initiation apparatus.

Deletion of RBP29 Results in a Synthetic Slow Growth Phenotype When Combined with tif4631 and tif4632 Mutations-- Having established that RBP29 function could be evaluated by monitoring its ability to suppress various tif4631, tif4632, and pab1 alleles, we next sought to identify conditions where a deletion of RBP29 resulted in a scorable growth phenotype. We found that the growth characteristics of the RBP29 deletion mutant were not significantly different from those of a wild type strain under a variety of conditions, including different temperatures or carbon sources, osmotic stress, and exposure to formamide, ethanol, or caffeine. Sporulation and germination efficiency were also not affected by this deletion. We thus chose to search for a growth phenotype associated with a deletion of RBP29 in the presence of mutations within the eIF4G genes.

As shown in Fig. 7, A and B, deletion of RBP29 in strains having only eIF4G1 or eIF4G2 as their source of eIF4G (see "Experimental Procedures") resulted in viable cells having the same growth rate as the control strain at 30 °C. However, when serial dilutions of cultures were incubated at 37 °C, these strains displayed a growth rate significantly slower than the parental stains containing an intact genomic copy of RBP29. This synthetic slow growth phenotype observed at 37 °C but not at 30 °C shows that Rbp29p is required for optimal growth at higher temperature when only one of the two yeast eIF4G homologues is available.

View larger version (45K): [in this window] [in a new window]   Fig. 7.   Synthetic slow growth phenotypes of rbp29 in combination with tif4631 or tif4632 mutations. Strains carrying a normal or a disrupted allele of RBP29 and either the TIF4631 (A), TIF4632 (B), tif4631-459 (C), or tif4632-430 (D) gene as their sole source of eIF4G were grown to stationary phase in minimal medium. Serial dilutions (1:5) of the cultures were prepared, and equal numbers of cells from each strain were plated on a minimal medium and incubated at 30 or 37 °C for 5 days.

When the wild type version of TIF4631 was replaced with the tif4631-459 allele, cells deleted for RBP29 showed a slower growth at 30 °C when compared with the parental strain containing the tif4631-459 allele and intact RBP29 (Fig. 7C). Deletion of RBP29 in a strain having the tif4632-430 allele also resulted in a subtle slower growth rate at 30 °C (Fig. 7D). This milder effect most probably reflects the weaker phenotype associated with the tif4632-430 allele (50). As expected, the rbp29 tif4631-459 and rbp29 tif4632-430 double mutants exhibited temperature-sensitive phenotypes when incubated at 37 °C (data not shown). Deletion of RBP29 in a strain producing an eIF4G1 protein that fails to interact with Pab1p (tif4631-N300; see Ref. 49) also resulted in a significant slow growth phenotype at 30 °C (data not shown).

In conjunction with the above high copy suppression data, these results support the hypothesis that Rbp29p normally functions to enhance the expression of mRNA. Under conditions where translation initiation is operating with limiting eIF4G levels, the function of Rbp29p appears to become more important for cells to retain their normal growth rates.

Rpb29p Overexpression Does Not Restore Normal Levels of Translation to tif4631-459 Strains-- The above results indicated that RBP29 plays a role in modulating the expression of cytoplasmic RNA. We next chose to determine whether RBP29 overexpression enhanced the ability of tif4631-459 mutants to translate mRNA more efficiently. We thus measured the rate of [³⁵S]methionine incorporation into protein in tif4631-459 cells expressing normal or high levels of Rbp29p. These rates were compared with the rate of incorporation of cells bearing a wild type copy of TIF4631.

Cultures in midlog phase were incubated for 2 h at 30 or 37 °C and labeled with [³⁵S]methionine for 10 min. Fig. 8 shows that, even at the permissive temperature, cells carrying the tif4631-459 allele had [³⁵S]methionine incorporation rates that were significantly lower than that observed for the wild type cells. Surprisingly, following a 2-h incubation period at the restrictive temperature (37 °C), the tif4631-459 cells continued to incorporate [³⁵S]methionine at about the same rate as they had at 30 °C. Longer incubation times at 37 °C (up to 6 h) gave similar results (data not shown). Excess Rbp29p in the tif4631-459 cells led to only a small increase in the rate of [³⁵S]methionine incorporation at both 30 and 37 °C.

View larger version (18K): [in this window] [in a new window]   Fig. 8.   Overexpression of RBP29 does not suppress the translational deficiency of tif4631-459 strains. Strains carrying TIF4631 (YAS1956), tif4631-459 (YAS2074), or tif4631-459 and RBP29 on a 2µ vector (tif4631 + 2µ RBP29) (YAS2618) were grown at 30 °C and then incubated at 30 or 37 °C for an additional 2-h period. The rate of [35S]methionine incorporation into protein was then measured as described under "Experimental Procedures."

These results show that cells carrying the tif4631-459 allele have a reduced rate of protein synthesis at both the permissive and nonpermissive temperatures and that excess Rbp29p does not restore these rates to a significant degree. One important ancillary conclusion that can be drawn from these data is that shifting tif4631-459 cells from the permissive to the restrictive temperature does not induce a general arrest of translation.

Since high copy suppression by excess Rbp29p appeared specific to eIF4G mutants defective in their binding to eIF4E (Table II), we tested the hypothesis that excess Rbp29p may increase the binding of eIF4E to eIF4G1-459. We found that the amount of eIF4E co-immunoprecipitating with eIF4G was the same in extracts prepared from tif4631-459 cells expressing either a normal or high level of Rbp29p (data not shown). We also tested the possibility that a high level of Rbp29p could suppress the temperature-sensitive growth phenotype of the tif4631-459 strain by increasing or decreasing the intracellular level of translation initiation factors or regulators. Western blot analysis using specific antibodies directed against eIF4G1, eIF4E, eIF4A, Caf20p, and Pab1p showed that the levels of these proteins were similar in extracts from tif4631-459 cells having a normal or high level of Rbp29p (data not shown).

Other results did not favor the hypothesis that RBP29 is involved in global control of protein synthesis. First, polysomal profiles of rbp29 cells were identical to those of the parental wild type cells. Second, there was no difference in the rate of translation of luciferase reporter mRNAs added to extracts prepared from rbp29 or wild type cells (data not shown). Third, extracts prepared from tif4631-459 cells expressing a high level of Rbp29p displayed the same in vitro translation characteristics as those observed using tif4631-459 extracts with normal Rbp29p levels. Finally, the addition of recombinant Rbp29p to translation extracts prepared from the tif4631-459 strain did not result in any significant changes in the translation of the different luciferase reporter mRNAs (data not shown). Our results also do not favor the hypothesis that Rbp29p might be involved in the global control of mRNA decay. We found that the half-life and the degradation pattern of the unstable MFA2 mRNA in rbp29 cells were nearly identical to those observed in wild type cells (data not shown).

These results indicate that RBP29 overexpression probably does not suppress the temperature-sensitive growth phenotype of the tif4631-459 strain by stimulating a general increase in translation. Instead, it may do so by increasing the expression of a subset of mRNAs whose limited expression at increased temperature results in the growth arrest phenotype.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

This study presents the characterization of S. cerevisiae Rbp29p, previously known to be a nonessential 29-kDa protein containing one RRM domain (56). Rbp29p was found to be a cytoplasmic RNA-binding protein that is associated with poly(A)⁺ RNAs in vivo. It is also associated with Pab1p in crude yeast extracts in a dosage- and RNA-dependent manner. In vitro, recombinant Rbp29p shows high binding affinity for poly(A). Although cells lacking Rbp29p have no discernible phenotype, we discovered that deletion of the RBP29 gene worsens the slow growth phenotype of tif4631 and tif4632 mutants. Moreover, excess Rbp29p suppresses the temperature-sensitive phenotype associated with tif4631, tif4632, and pab1 alleles. These data suggest that Rbp29p plays a stimulatory role in mRNA expression, possibly by affecting the efficiency of translation of a subset of transcripts.

RBP29 is a member of a family of about 45 genes in S. cerevisiae that encode proteins encompassing at least one RRM domain. Its localization to the cytoplasm, its interaction with cytoplasmic Pab1p, and its genetic interaction with the translation machinery make it a bona fide member of a small group of yeast cytoplasmic RRM-containing proteins. These include translation initiation factors eIF4B (Tif3p), the -subunit of eIF3 (Prt1p), and eIF4G as well as the mRNA-binding proteins Pab1p and Pub1p.

Despite having exhibited a function in mRNA expression, Rbp29p in crude yeast extracts did not co-sediment with polyribosomes in sucrose gradients (data not shown). This might be explained if Rbp29p binds to the coding region of mRNA and has to dissociate from the transcript to allow for the passage of the ribosomes. It is important to note that although Rbp29p has high affinity for poly(A) in vitro (Fig. 4), there is no direct evidence that Rbp29p is bound to the poly(A) tail of mRNAs in vivo. In fact, our results showing that rRNA can efficiently compete with poly(A) for Rbp29p binding (Fig. 4B) could indicate that Rbp29p functions as a more general RNA-binding protein. However, it should also be noted that the poly(A) binding of mammalian PABP2, which is homologous to Rbp29p, is also competed by rRNA (12).

The present study extends the characterization of the tif4631-459 mutant by showing that even at the permissive temperature, this strain has a highly reduced rate of protein synthesis. Shifting the cell to the restrictive temperature (37 °C) does not result in a further significant reduction of these rates. Based on these results, it seems likely that the growth arrest of the tif4631-459 mutant at 37 °C is a consequence of a reduction in the rate of translation of subset of mRNAs whose products are required for progression through the cell cycle. This pool of mRNAs could include transcripts that are normally poorly translated or whose steady state levels are reduced at higher temperature. One possibility is that Rbp29p functions to ensure the optimal level of expression of such a subpopulation of mRNAs. Consistent with this idea are our results showing that Rbp29p is required for optimal growth at higher temperature when only one of the two yeast eIF4G homologues is available (Fig. 7, A and B). This requirement for Rbp29p is also evident at 30 °C when eIF4G cannot interact properly with the cap-binding protein (Fig. 7C) or with Pab1p (data not shown).

Rbp29p could work in dosage suppression of the tif4631-459 temperature sensitivity either by inducing a general increase in mRNA translatability or by increasing the expression of a specific subset of transcripts. Our observation that the amount of Rbp29p co-immunoprecipitating with Pab1p is dependent on the amount of Rbp29p present in the cell is consistent with the hypothesis that it functions as a dosage suppressor as a result of its increased binding to mRNA. Whether this increased binding leads to more efficient mRNA export, mRNA translation, mRNA stabilization, or some other process will only be known when the mRNAs whose expression is stimulated by Rbp29p are identified.

In conclusion, we have characterized a nonessential RNA-binding protein and provided clues about its possible function in cytoplasmic mRNA metabolism. Our identification of genetic interactions between mutations in RBP29 and TIF4631 and TIF4632 now provide useful tools for further investigations of the role of Rbp29p in yeast. Together with the anticipated progress in understanding the cellular roles of other poorly characterized yeast RNA-binding proteins, our work should provide a basis for the development of a more complete model of how mRNA packaging functions to regulate gene expression in eucaryotes.

	ACKNOWLEDGEMENTS

We thank Reinhard Lührmann for the antibody against Snp1p; Anne Nemeth for help with initial experiments; Lionel Minvielle for discussions, advice, and help with experiments; and Walter Keller for support. We also thank members of the Sachs laboratory for help and advice.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grant GM50308 (to A. B. S.) and by funding from the Deutsche Forschungsgemeinschaft, the TMR program of the European Union, and the Fonds der Chemischen Industrie to Elmar Wahle.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.

^§ Supported by a postdoctoral fellowship from The Natural Sciences and Engineering Research Council of Canada.

Present address: Dept. of Cell Biology, Biozentrum, University of Basel, Klingelbergstr. 70, 4056 Basel, Switzerland.

^** Present address: Institut für Biochemie, Martin-Luther-Universität Halle, 06099 Halle, Germany.

To whom correspondence should be addressed: Dept. of Molecular and Cell Biology, 401 Barker Hall, University of California, Berkeley, CA 94720. Tel.: 510-643-7698; Fax: 510-643-5035; E-mail: asachs@uclink4.berkeley.edu.

Published, JBC Papers in Press, April 11, 2000, DOI 10.1074/jbc.M002412200

	ABBREVIATIONS

The abbreviations used are: RRM, RNA recognition motif; RNP, ribonucleoprotein; ORF, open reading frame; PCR, polymerase chain reaction; PMSF, phenylmethylsulfonyl fluoride; PBS, phosphate-buffered saline; HA, hemagglutinin.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit<