The major core protein of messenger ribonucleoprotein particles (p50) promotes initiation of protein biosynthesis in vitro.

The major core protein of cytoplasmic messenger ribonucleoprotein particles (p50) has been shown previously to inhibit protein synthesis in vitro and in vivo. Furthermore, p50 is highly homologous to the Y-box-binding transcription factor family of proteins, binds DNA containing the Y-box motif, and thus may have a dual function in cells as a regulator of both transcription and translation. Here we show that binding or removal of p50 from rabbit reticulocyte lysate by monospecific antibodies to p50 strongly inhibits translation of endogenous and exogenous globin mRNAs as well as prokaryotic beta-galactosidase mRNA in a rabbit reticulocyte cell-free system. Thus, depending on the conditions, p50 not only may act as a translational repressor, but may also be required for protein synthesis. Translation inhibition with anti-p50 antibodies is not a result of mRNA degradation or its functional inactivation. The inhibition does not change the ribosome transit time, and therefore, it does not affect elongation/termination of polypeptide chains. The inhibition with anti-p50 antibodies is followed by a decay of polysomes and accumulation of the 48 S preinitiation complex. These results suggest that p50 participates in initiation of protein biosynthesis. Although uninvolved in the formation of the 48 S preinitiation complex, p50 is necessary either for attachment of the 60 S ribosomal subunit or for previous 5'-untranslated region scanning by the 43 S preinitiation complex.

The major core protein of cytoplasmic messenger ribonucleoprotein particles (p50) has been shown previously to inhibit protein synthesis in vitro and in vivo. Furthermore, p50 is highly homologous to the Y-boxbinding transcription factor family of proteins, binds DNA containing the Y-box motif, and thus may have a dual function in cells as a regulator of both transcription and translation. Here we show that binding or removal of p50 from rabbit reticulocyte lysate by monospecific antibodies to p50 strongly inhibits translation of endogenous and exogenous globin mRNAs as well as prokaryotic ␤-galactosidase mRNA in a rabbit reticulocyte cell-free system. Thus, depending on the conditions, p50 not only may act as a translational repressor, but may also be required for protein synthesis. Translation inhibition with anti-p50 antibodies is not a result of mRNA degradation or its functional inactivation. The inhibition does not change the ribosome transit time, and therefore, it does not affect elongation/termination of polypeptide chains. The inhibition with anti-p50 antibodies is followed by a decay of polysomes and accumulation of the 48 S preinitiation complex. These results suggest that p50 participates in initiation of protein biosynthesis. Although uninvolved in the formation of the 48 S preinitiation complex, p50 is necessary either for attachment of the 60 S ribosomal subunit or for previous 5-untranslated region scanning by the 43 S preinitiation complex.
Regulation of gene expression in eukaryotes often involves modulating the rate of initiation of protein synthesis (for reviews, see Refs. [1][2][3]. Translation is controlled by proteins that are associated with mRNA and/or ribosomes and that either promote or repress the protein synthesis rate. One type of mechanism involves modulation of initiation factor activities (4) that usually affect global translation, whereas the second type involves specific protein repressors that bind to one or a small set of mRNAs (5), leading to specific control. The third type of mechanism may involve proteins associated with many or even all mRNA species, such as proteins present in messen-ger ribonucleoprotein particles (mRNPs) 1 (6,7). These proteins also may contribute to global control of protein synthesis. mRNPs isolated from the cytoplasm of different cells and tissues contain two major proteins with molecular masses of 70 and 50 kDa (8). The 50-kDa protein (p50) is the most abundant and the most strongly bound protein within free nontranslatable mRNPs (9) and therefore can be regarded as the core mRNP protein of these particles in mammalian somatic cells. It is also present in mRNPs derived from polysomes, but in lower amounts (9,10). Regardless of the fact that p50 displayed little or no specificity for RNA in in vitro binding experiments (10,11), in cell extracts, p50 was found only in association with mRNA (12). This protein is located mostly in the cytoplasm, and its amount is ϳ0.1% of the total protein, which corresponds to ϳ5-10 molecules of p50/molecule of mRNA (12). Among all mRNP proteins, p50 is the most basic, with a pI of ϳ9.5 (10). In the absence of RNA, p50 forms large multimeric complexes with a molecular mass of ϳ800 kDa (11). When binding to mRNA, p50 melts up to 60% of the RNA secondary structure (11). We have shown earlier that p50 is responsible for the repressed nonactive state of globin mRNA within free mRNPs (13) and that it strongly inhibits translation of exogenous mRNA in cell-free translation systems (9,14) as well as in vivo translation of mRNA expressed from a reporter gene (12).
According to its amino acid sequence and its affinity for DNA, p50 was identified as a member of the most evolutionarily conserved family of Y-box-binding proteins from bacteria to man (11). Some proteins of this family are known as transcription factors affecting transcription of genes containing Y-box sequence elements in their promoters (15)(16)(17). The prokaryotic Y-box-binding protein is also known as a major cold shock protein stimulating gene expression under cold shock conditions (18). Two homologous core mRNP proteins (p54/p56) from Xenopus oocytes have also identified as members of the Y-boxbinding protein family and are closely related, if not identical, to FRG Y2, the germ cell-specific form of the transcription factor (19). They were reported to be responsible for the masked state of mRNAs in Xenopus oocytes (20 -24). One can therefore conclude that the somatic mRNP protein p50 and the germ cell mRNP proteins p54/p56 have a similar function in the cytoplasm, namely that of preventing mRNA translation. Thus, the Y-box-binding proteins can influence protein synthesis both by affecting transcription of Y-box-containing genes and by inhibiting translation of a wide variety of mRNAs. By interacting with DNA in the nucleus and with mRNA in the cytoplasm, these proteins may establish a balance between gene activity and mRNA content in cells.
In a detailed study of the effect of p50 on exogenous mRNA translation in reticulocyte lysates, we showed that p50 inhibits mRNA translation when added at a high p50/mRNA ratio. However, with high inhibitory amounts of purified mRNA, addition of low amounts of p50 increases the efficiency of mRNA translation (9). This intriguing result suggests that p50 may also play a positive role in translation. Here we ask if depletion of p50 affects the rate of protein synthesis in reticulocyte lysates. Reduced p50 activity was accomplished by adding p50-specific antibodies. In this instance, translation is inhibited at the level of initiation, thereby providing evidence for a positive role of p50 in promoting protein synthesis.

EXPERIMENTAL PROCEDURES
Preparation of p50 -p50 from free mRNPs of rabbit reticulocytes was obtained as described earlier (11), except that the step of protein precipitation with (NH 4 ) 2 SO 4 was omitted. Recombinant p50 was expressed in Escherichia coli BL21(DE3) cells, and the protein was purified by chromatography on heparin-Sepharose 4B and Superose 12 HR 10/30 columns (Pharmacia Biotechnology, Inc.) as described (25). The p50 preparations were dialyzed against buffer containing 10 mM Hepes-KOH, pH 7.8, and 250 mM KCl and stored at Ϫ70°C in aliquots. Protein concentration was determined by staining with a Micro-BCA kit (Pierce).
Generation of Polyclonal Antibodies against p50 and Western Blot Analysis-Polyclonal antibodies were produced either in mice against p50 from rabbit reticulocytes (11) or in rabbits against recombinant p50 synthesized in E. coli (12). Anti-p50 IgG and preimmune IgG were purified on a protein A-Sepharose column (Sigma) (26), dialyzed, and stored in buffer containing 10 mM Hepes-KOH, pH 7.6, and 100 mM KOAc; therefore, the same buffer was added to parallel control incubations throughout the experiments described. For Western blot analysis, proteins were separated by SDS-PAGE and transferred to a nitrocellulose membrane. The membrane was blocked overnight at 20°C with 1% bovine serum albumin, 1% polyvinylpyrrolidone, and 0.05% Tween 20 in 10 mM Tris-HCl, pH 7.6, and 150 mM NaCl and probed with anti-p50 antibodies at 1:500 dilution. Immunocomplexes were detected with alkaline phosphatase-coupled secondary antibodies (Cappel) according to the manufacturer's recommendation.
RNA Isolation-Globin mRNA was obtained from polysomes of rabbit reticulocytes by phenol/chloroform extraction followed by oligo(dT)cellulose chromatography as described (27). E. coli ␤-galactosidase mRNA was obtained by in vitro transcription with SP6 polymerase from pJCS-␤-galactosidase linearized with HindIII according to standard procedures (28).
For ribosome transit time measurements, the volume of the translation mixtures was increased to 150 l, and translations were carried out as described above. At the indicated intervals, 25-l aliquots were diluted with 125 l of ice-cold stop buffer containing 10 mM Tris-HCl, pH 7.6, 100 mM KCl, 5 mM MgCl 2 , and 0.1 mM cycloheximide (Sigma) and placed immediately on ice. The mixtures were layered onto a 30% FIG. 1. Monospecific anti-p50 IgG inhibits translation of endogenous or exogenous rabbit globin mRNA in rabbit reticulocyte cell-free systems. A, purified p50 (5 g) from free mRNPs of rabbit reticulocytes (lane 1) and recombinant p50 (8 g) from E. coli (lane 2) were subjected to SDS-PAGE and visualized by Coomassie Blue R-250 staining. B, Western blotting of 0.1 g of purified rabbit p50 (lane 1) and 2 l (60 g of total protein) of rabbit reticulocyte lysate (lanes 2 and 3) was performed with mouse anti-p50 IgG (lanes 1 and 2) or with preimmune IgG (lane 3). C and D, rabbit reticulocyte lysates (15 l) with endogenous and exogenous mRNAs, respectively, were preincubated with the indicated amounts of antibodies for 10 min at 0°C and assayed for [ 14 C]Leu incorporation in the standard translation system as described under "Experimental Procedures." glycerol cushion prepared with 10 mM Tris-HCl, pH 7.6, 100 mM KCl, and 5 mM MgCl 2 and centrifuged at 100,000 rpm in a TLA-100.3 rotor (Beckman Instruments) for 40 min. This procedure deposits polyribosomes and monomeric ribosomes in the pellet. The ribosomal pellet was dissolved in 1% SDS, and this fraction and the supernatant were assayed for radioactivity as described above.
Depletion of p50 Activity in Cell Lysates-For neutralization of p50 activity, reticulocyte lysates (15 l) were preincubated with the indicated amounts of antibodies for 10 min at 0°C. For immunodepletion experiments, nuclease-treated reticulocyte lysates (200 l) were preincubated with various amounts of antibodies for 10 min at 0°C with gentle agitation and passed through a 200-l column of protein A-Sepharose equilibrated with 10 mM Hepes-KOH, pH 7.6, 100 mM KOAc, and 1 mM MgCl 2 . The first 250 l of flow-through material from the column was collected and frozen in liquid nitrogen in 15-l aliquots. Neutralized or immunodepleted reticulocyte lysates were reconstituted by adding the indicated amounts of p50 directly to the lysate and incubating for 3 min at 0°C. The neutralized, depleted, and reconstituted lysates were then assayed for protein synthesis activity as described above.
Sucrose Gradient Analyses and Dot Blot Assays-Cell-free translation mixtures (30 l) were cooled and immediately layered onto 15-30% (w/v) linear sucrose gradients in buffer containing 10 mM Tris-HCl, pH 7.6, 100 mM NaCl, and 1 mM MgCl 2 . Centrifugation was carried out at 45,000 rpm in an SW 60 rotor (Beckman Instruments) for the times indicated in the figure legends. All gradients were monitored for absorbance at 254 nm during their collection from the bottom. Fractions of 300 l were diluted with 700 l of ice-cold water and treated with deacylation solution, and [ 14 C]Leu incorporation into trichloroacetic acid-precipitable material was measured as described above. For dot hybridization analysis of globin mRNA, total RNA was extracted from the fractions by an equal volume of phenol/chloroform, filtered onto a nitrocellulose Hybond-N membrane (Amersham Corp.), and probed with a full-length ␣-globin cDNA 32 P-labeled by the multiprime DNA labeling system (Amersham Corp.). For quantitation of hybridized RNA, each dot was excised, and radioactivity was measured by Cherenkov counting in a Beckman LS-100C scintillation counter.

Cell-free Translation Systems Deficient in p50 Do Not Synthesize Proteins
Efficiently-To verify the suggestion that p50 may play a positive role in protein biosynthesis, we sought to deplete p50 activity in reticulocyte lysates. For selective reduction of p50 activity in the rabbit reticulocyte cell-free translation system, two different polyclonal anti-p50 antibody preparations were used. The antibodies were raised either in mouse against p50 from free rabbit reticulocyte mRNPs or in rabbit against recombinant p50 expressed in E. coli. Both p50 preparations were purified to homogeneity; no other peptide bands were detected by Coomassie Blue R-250 staining of these proteins resolved by SDS-PAGE (Fig. 1A). The two anti-p50 antibodies were purified on protein A-Sepharose, and their specificities were tested by immunoblotting. The antibodies reacted strongly with purified p50 and recognized only a single antigen in reticulocyte lysates corresponding in electrophoretic mobility to p50 (Fig. 1B). Therefore, the antibodies are monospecific and may be suitable for neutralization of p50 activity due to their specific binding to p50 in the rabbit reticulocyte cell-free system and for p50 immunodepletion by affinity adsorption from the lysate. Both antibody preparations gave identical results, so further reported experimental data are on antibodies against rabbit reticulocyte p50, unless indicated otherwise.
The effect of anti-p50 IgG addition on translation of endogenous and exogenous globin mRNAs in the rabbit reticulocyte cell-free system is shown in Fig. 1 (C and D, respectively). In both systems, the antibodies caused a dramatic inhibition of globin synthesis, whereas the control immunoglobulins from a nonimmunized animal (preimmune IgG) affected the activity of the system only slightly. A straightforward interpretation of these results is that anti-p50 IgG binds to p50 and inhibits its putative function in promoting mRNA translation.
To deplete p50 from the cell-free translation system, a mi-crococcal nuclease-treated lysate was mixed with various amounts of anti-p50 IgG and then passed through a protein A-Sepharose column. Increasing amounts of added antibodies against either rabbit or recombinant p50 caused an increasing loss of lysate protein-synthesizing activity in exogenous mRNA (Fig. 2), with the highest amount resulting in ϳ90% inhibition. Since antibodies against recombinant p50 cannot be contaminated with even trace amounts of antibodies against other eukaryotic proteins, we conclude that anti-p50 antibodies inhibit protein synthesis. Furthermore, the p50 content in the depleted lysates was measured by Western blot analysis (Fig.  2, inset). The results show a progressive loss of p50, with barely detectable p50 at the highest antibody concentration. There is a rough linear correlation between the p50 content and translational activity, suggestive of p50 being limiting for protein synthesis in these lysates.
To obtain stronger evidence that the translation inhibition caused by adding specific anti-p50 antibodies results from p50 inactivation, p50 was added back to the cell-free systems that had been neutralized or depleted (Fig. 3). In these experiments, we used p50 isolated from free rabbit reticulocyte mRNPs as well as recombinant p50 synthesized in E. coli and therefore absolutely free of even trace amounts of other eukaryotic proteins that can affect translation. The results of experiments with both preparations proved to be nearly the same. p50 completely restored protein synthesis in the inhibited systems reduced to 40% of their activity (Fig. 3B). With more profound inhibition (reduced to 10 -20% activity), p50 stimulated translation 3-4-fold; the system's activity increased linearly with increasing amounts of p50 and finally reached 45-65% of the initial lysate activity (Fig. 3A). A further increase of the p50 amount caused inhibition of translation (data not shown), which is in good agreement with our earlier observation that p50 acts as a translational repressor at a high p50/mRNA ratio (9). In reticulocyte lysate, after immunodepletion, biosynthesis was restored after addition of 2-3 g of p50/1 g of exogenous globin mRNA (10 -15 pmol of p50/pmol of globin mRNA), which is near the p50/mRNA ratio in COS cells, i.e. 5-10 molecules of p50/molecule of mRNA (12). After profound inhibition of protein biosynthesis by antibodies, incomplete restoration by p50 can probably be explained by improper modification of p50 from nontranslatable free mRNPs or recombinant p50 used for the activity restoration. Another explanation is that other auxiliary factors important for p50 activity were removed from the system by antibodies due to their association with p50. These proteins are the subject of our study at the moment. Nevertheless, the fact that highly purified or recombinant p50 by itself is capable of restoring or stimulating protein-synthesizing activity strongly indicates that the inhibition by antibodies is mostly due to specific recognition and inactivation of p50 rather than some other proteins. These results suggest that p50 is required for efficient protein synthesis.
Inhibition of Protein Synthesis with Anti-p50 IgG Is Not Caused by Acceleration of mRNA Degradation-Antibody binding to p50 may stimulate mRNA decay in the cell-free translation system, thereby causing translation inhibition. To verify this idea, RNAs were isolated from lysates incubated under cell-free translation conditions with and without anti-p50 IgG and tested for messenger activity. Both mRNA preparations possessed equal messenger activity in the cell-free translation system and produced full-length globin chains as determined by SDS-PAGE (Fig. 4). The messenger activity of these RNAs was the same as that of RNA from a nonincubated lysate (data not shown). Thus, anti-p50 IgG-induced protein synthesis inhibition is not caused by mRNA degradation. Fig. 5 shows the effect of anti-p50 IgG on the kinetics of protein synthesis in rabbit reticulocyte lysates with endogenous (panel A) and exogenous (panel B) globin mRNAs. With either type of mRNA, anti-p50 IgG inhibited strongly at high concentrations, whereas preimmune IgG had little or no effect. The inhibition of endogenous mRNA translation with a high anti-p50 IgG concentration resembled the inhibition caused by edeine, a specific inhibitor of the initiation phase of protein synthesis (30,31). Thus, the shape of the curves suggests that the antibodies inhibit mainly translation initiation.

Specific Anti-p50 Antibodies Inhibit Initiation but Not Elongation/Termination of Globin Synthesis in Reticulocyte Lysates-
The phase of translation affected by an inhibitor can be determined more precisely by polysome profile analysis and measurement of the elongation rate. The run-off of ribosomes The 50-kDa Protein from mRNPs from polysomes usually indicates inhibition of initiation, whereas maintenance of or an increase in polysome size suggests inhibition of elongation/termination. We have compared the polysome profiles of reticulocyte lysates incubated in the absence and presence of either anti-p50 IgG or preimmune IgG (Fig. 6). A short incubation (2 min) of the cell-free system without antibodies or with preimmune IgG did not produce any remarkable change in the polysome profiles (Fig. 6, compare A  with B and C). However, incubation with anti-p50 IgG resulted in a complete polysomal decay (Fig. 6D). The process was accompanied by dissociation of the radiolabeled polypeptide chain from the ribosomes. This indicates that polysomal decay is not due to mRNA fragmentation by ribonucleases as such cleavage of polysomes produces ribosomes associated with the growing polypeptide. Rather, the run-off of ribosomes from polysomes implies that anti-p50 IgG inhibits mainly initiation and not elongation or termination.
The average time of elongation ϩ termination of polypeptide chains (transit time) can be quantitatively determined by measuring the kinetics of radioactive amino acid incorporation into total protein and into completed polypeptides released from the ribosome (32). We used this technique to determine the effect of anti-p50 IgG on the elongation ϩ termination rate. From the results shown in Fig. 7, the transit time for globin synthesis in the control uninhibited lysate was 1.4 min. In the experimental system with anti-p50 IgG, the same transit time was obtained, although in this case, protein synthesis was suppressed 2-fold over the studied time range (6 min). Thus, anti-p50 antibodies do not affect the polypeptide elongation ϩ termination rate and inhibit only initiation.
Specific Anti-p50 Antibodies Inhibit Initiation but Not Elongation of ␤-Galactosidase Synthesis in Reticulocyte Lysates-p50 is a universal mRNP protein associated with all or almost all mRNAs of mammalian somatic cells (11). The question arises as to whether p50 promotes only initiation of eukaryotic mRNAs having an m 7 G cap, a characteristic consensus sequence around the initiator AUG codon and 3Ј-poly(A) tail, or whether it would exert the same positive effect on translation of a bacterial mRNA lacking these features. To answer this question, an uncapped mRNA transcript encoding E. coli ␤-ga-lactosidase was translated in the reticulocyte cell-free system without antibodies or with anti-p50 IgG added either at zero time or after 10 min of incubation (when only elongation occurs). The kinetics of ␤-galactosidase synthesis was studied by SDS-PAGE analysis of the product obtained after different incubation times. Synthesis of full-length ␤-galactosidase was completed in 20 -40 min in the system without antibodies (Fig.  8A) (33). When anti-p50 IgG was added before incubation, no synthesis of full-length or even partial-length protein was observed (Fig. 8B), consistent with inhibition at initiation. With anti-p50 IgG added after 10 min of incubation, ␤-galactosidase synthesis occurred and yielded full-length product after 40 min of incubation (Fig. 8C). The pattern of partial proteins seen does not differ detectably from that obtained with the control lysate (Fig. 8A). These results show that p50 is strongly required for initiation of bacterial mRNA translation, and it does not affect elongation even at antibody concentrations providing the complete inhibition of initiation. This is in good agreement with our previous findings on eukaryotic globin mRNA (Figs. 5-7).
Specific Anti-p50 Antibodies Cause Accumulation of 48 S Preinitiation Complexes-To identify the step in the initiation pathway that is inhibited by anti-p50 IgG, the cell-free translation systems were incubated without antibodies, with anti-p50 IgG, or with edeine and then fractionated by centrifugation in sucrose gradients. The distribution of ␣-globin mRNA in the gradient was determined by dot hybridization with 32 P-labeled ␣-globin cDNA. In the control cell-free translation system, mRNA was found mainly on 80 S ribosomes and to a lower extent in the region of free globin mRNPs (ϳ20 S) (Fig. 9A). After incubation in the presence of edeine blocking the joining of the 60 S subunit (30,31), most of the ␣-globin mRNA was found in the region where the small 40 S ribosomal subunit sediments and can be regarded as the 48 S preinitiation complexes (40 S⅐Met-tRNA i ⅐eukaryotic initiation factor-2⅐GTP⅐ mRNA) (Fig. 9C). In the system incubated in the presence of anti-p50 IgG, the bulk of ␣-globin mRNA was also found in 48 S preinitiation complexes (Fig. 9B). Thus, anti-p50 IgG does not inhibit binding of 40 S ribosomes to mRNA, but rather inhibits a subsequent step in the initiation pathway.

DISCUSSION
The results reported here indicate that p50, the major protein of cytoplasmic mRNPs, not only functions as a repressor of mRNA translation, but also is required for protein biosynthesis. Two independent preparations of monospecific anti-p50 antibodies raised against either rabbit or recombinant p50 for 10 min at 0°C, and translations were carried out for the indicated times. [ 14 C]Leu incorporation into trichloroacetic acid-precipitable material was measured as described under "Experimental Procedures." Transit times were obtained by multiplying by 2 the distance along the time axis between the "total" and "post-ribosomal" lines. In B, the relevant portion of the curve is the early time points up to 6 min, where most of the [ 14 C]Leu incorporation is due to elongation on already initiated mRNAs.
FIG. 8. Anti-p50 antibodies inhibit initiation but not the elongation phase of prokaryotic ␤-galactosidase synthesis. Translation reactions were carried out with 20 g/ml E. coli ␤-galactosidase mRNA without antibodies (A) and in the presence of 8 g of anti-p50 IgG added at zero time of incubation (B) or after 10 min of incubation (C). At the indicated intervals, 3-l aliquots were removed, and [ 35 S]Met-labeled translation products were resolved by SDS-PAGE and visualized by autoradiography. The protein bands migrating more rapidly than full-length ␤-galactosidase may represent either incomplete nascent chains or earlier quitters. were used to reduce the activity of endogenous p50 in rabbit reticulocyte lysates or to prepare depleted lysates. Immunoneutralization or immunodepletion of p50 by both antibody preparations caused inhibition of protein synthesis. Addition of highly purified rabbit or recombinant p50 proteins to such systems restored the translational activity completely or nearly completely, which confirms the positive role of p50 in mRNA translation.
How can p50 function both to promote and to repress translation? The in vitro assays for p50 activity suggest that the level of p50 determines whether it stimulates or inhibits translation. With the p50/mRNA weight ratio increasing up to 2 (which is characteristic of polysomal mRNPs), p50 stimulates protein biosynthesis, whereas a further increase of the ratio up to 5-6 (close to the ratios characteristic of free mRNPs) causes a gradual inhibition of translation until it ceases completely (9). Thus, the inactive-to-active transition of mRNA may be connected with a decrease in the amount of p50 attached to an mRNA molecule.
The amount of p50 on mRNA can possibly be regulated by phosphorylation. It is known that p50 is a phosphoprotein (10, 34) and may be phosphorylated by a protein kinase present in mRNPs that resembles casein kinase II. 2 Furthermore, the Xenopus proteins p54/p56 are phosphoproteins, and their binding to RNA is enhanced upon phosphorylation by casein kinase II (35). Phosphorylation of p54/p56 results in inhibition of protein synthesis, whereas inhibitors of casein kinase II activate translation (24,35). Work is in progress to elucidate how phosphorylation of p50 may affect its binding to RNA and its activity in stimulating and inhibiting translation in vitro.
Immunoneutralization of p50 does not change the ribosome transit time for globin synthesis and does not affect elongation of prokaryotic ␤-galactosidase. This means that p50 is not required for elongation and termination, and consequently, it is required only for initiation. The effect of anti-p50 IgG on the kinetics of protein synthesis in the cell-free translation system and on the decay of polysomes also points to inhibition of translation initiation.
Analysis of the distribution of globin mRNA in reticulocyte lysates inhibited by anti-p50 IgG shows that mRNA accumulates in 48 S preinitiation complexes. This means that p50 is not required for attachment of the small subunit of the ribosome to mRNA, although it is necessary for subsequent binding of the 60 S ribosomal subunit to the complex. We suggest that either p50 directly participates in attachment of the 60 S ribosomal subunit to the 48 S preinitiation complex or that it is involved in the previous step of 5Ј-untranslated region mRNA scanning by the 43 S preinitiation complex.
Several mechanisms of p50 participation in translation initiation can be proposed. (i) Nonspecific affinity of p50 for RNA and the presence of many copies of p50 on mRNA can protect mRNA against nonspecific binding of initiation factors along its entire length, thus contributing to their specific binding to the 5Ј-untranslated region (9,36). (ii) Since p50 possesses RNAunwinding activity (11), its direct participation in 5Ј-untranslated region scanning is quite possible. (iii) p50 provides the general mRNA structure favorable for translation initiation. (iv) Finally, we cannot rule out that p50 affects translation initiation by its direct interaction with translation initiation factors. These suggestions are being verified currently.
p50 is not the only mRNP protein implicated in the initiation phase of protein synthesis. Another major mRNP protein, p70 (or poly(A)-binding protein), appears to be involved in protein synthesis initiation, too (37)(38)(39)(40). Whether poly(A)-binding protein and p50 interact directly on mRNAs has not yet been determined.
FIG. 9. Anti-p50 antibodies cause accumulation of 48 S preinitiation complexes. Reticulocyte lysates with endogenous mRNA were incubated under translation conditions for 15 min at 30°C for polysomal decay, and then either 8 g of anti-p50 IgG (B) or 5 M edeine (C) was added. Incubation was continued for 10 min at 0°C and for another 7 min at 30°C. In A, incubation was carried out without any additions (control). Samples were ice-cooled and subjected to 15-30% linear sucrose gradient centrifugation for 4 h at 45,000 rpm in an SW 60 rotor. For globin mRNA determinations, Northern dot hybridization with 32 P-labeled ␣-globin cDNA (4 ϫ 10 5 cpm/ml) was performed as described under "Experimental procedures." UV absorption profiles at 254 nm (OO) and 32 P radioactivity (q) are shown. Sedimentation is from right to left; sedimentation positions of 80 S, 60 S, and 40 S ribosomes are indicated.