Association of Protein Biogenesis Factors at the Yeast Ribosomal Tunnel Exit Is Affected by the Translational Status and Nascent Polypeptide Sequence*

Ribosome-associated protein biogenesis factors (RPBs) act during a short but critical period of protein biogenesis. The action of RPBs starts as soon as a nascent polypeptide becomes accessible from the outside of the ribosome and ends upon termination of translation. In yeast, RPBs include the chaperones Ssb1/2 and ribosome-associated complex, signal recognition particle, nascent polypeptide-associated complex (NAC), the aminopeptidases Map1 and Map2, and the Nα-terminal acetyltransferase NatA. Here, we provide the first comprehensive analysis of RPB binding at the yeast ribosomal tunnel exit as a function of translational status and polypeptide sequence. We measured the ratios of RPBs to ribosomes in yeast cells and determined RPB occupation of translating and non-translating ribosomes. The combined results imply a requirement for dynamic and coordinated interactions at the tunnel exit. Exclusively, NAC was associated with the majority of ribosomes regardless of their translational status. All other RPBs occupied only ribosomal subpopulations, binding with increased apparent affinity to randomly translating ribosomes as compared with non-translating ones. Analysis of RPB interaction with homogenous ribosome populations engaged in the translation of specific nascent polypeptides revealed that the affinities of Ssb1/2, NAC, and, as expected, signal recognition particle, were influenced by the amino acid sequence of the nascent polypeptide. Complementary cross-linking data suggest that not only affinity of RPBs to the ribosome but also positioning can be influenced in a nascent polypeptide-dependent manner.

Newly synthesized polypeptides exit the ribosome through a tunnel in the large ribosomal subunit. As soon as the polypeptides reach the tunnel exit, important decisions are required to direct subsequent steps of protein biogenesis. In all kingdoms of life a specific set of ribosome-associated protein biogenesis fac-tors (termed RPBs 2 hereafter) is mandatory for the process. However, RPBs differ significantly between bacterial and eukaryotic cells. Eubacteria possess trigger factor, a chaperone involved in cotranslational protein folding, which is restricted to eubacteria and signal recognition particle (SRP), a targeting factor involved in the translocation of membrane proteins with a hydrophobic signal-anchor sequence (1,2). Consistent with the function of a general chaperone, trigger factor and ribosomes form 1:1 complexes whereas bacterial SRP, which is required for the biogenesis of only a subset of newly synthesized proteins, is present at ϳ1 molecule/100 ribosomes (1,3). Notably, trigger factor and SRP bind to the same region close to the exit of the ribosomal tunnel (1,2). The current view is that trigger factor and SRP can bind simultaneously to a single ribosome (1); however, it was suggested that only one at a time contacts a nascent polypeptide (4). According to this model, the decision-making process at the eubacterial tunnel exit would be straightforward: Whether trigger factor or SRP act on a nascent polypeptide depends on their relative affinities to the exposed stretches of amino acids (Refs. 4, 5 and references therein).
For sterical considerations it is difficult to envisage that the full set of eukaryotic RPBs interacts simultaneously with one ribosome. In addition, the time frame for the action of RPBs on nascent polypeptides is only short. Logarithmically growing yeast cells translate with a speed of ϳ10 amino acids/second (26), and thus for the majority of polypeptides cotranslational actions have to be completed in significantly less than a minute. How the arrangement of RPBs at the ribosomal tunnel exit is functionally coordinated in time and space is one of the challenging questions. A prerequisite to understanding the dynamics is information about the interaction of RPBs with ribosomes as a function of translational status and polypeptide sequence. Although the problem is straightforward, the methods to pinpoint RPB dynamics are not. Analysis requires a uniform population of non-translating ribosomes, as well as defined ribosome nascent chain complexes (RNCs) in quantities that allow for immunodetection of RPBs. Moreover, the concentration of a significant number of proteins has to be analyzed in complex mixtures. We have developed the tools and have performed the first thorough investigation of RPB-ribosome interaction under physiological conditions. To that end, we have employed a homologous system in which all RPBs, ribosomes, and RNCs including nascent polypeptides were derived from yeast. Experimental conditions were chosen such that the ratios between different RPBs and between RPBs and ribosomes were the same as in intact cells. We regard this as important, as it was shown in the Escherichia coli system that the normal ratio of cytosolic components is critical for the delicate balance of nascent polypeptide interactions (5). The approach allowed us to study the interaction of the whole set of RPBs with non-translating ribosomes, randomly translating ribosomes, and specific RNCs under conditions that resemble, as closely as possible, an intact cell.
Determination of Protein Concentrations-Protein concentrations were determined according to the manufacturers' manuals, with bovine serum albumin as a standard by the Bradford assay (Bio-Rad), the BCA assay (Sigma), and the DC protein assay (Bio-Rad) or were calculated from absorption at 280 nm (supplemental Table S1).
Purification of FLAG-tagged RNCs under Native Conditions-For a typical experiment 75-l translation reactions were performed at 20°C for 80 min and were terminated by the addition of cycloheximide to a final concentration of 200 g/ml. Translation reactions were then added to 40 l of ANTI-FLAG M2 affinity gel (␣FLAG-beads; Sigma) resuspended in 500 l of immunoprecipitation buffer (20 mM HEPES-KOH, pH 7.4, 150 mM potassium acetate acetate, 2 mM magnesium acetate, 50 g/ml trypsin inhibitor, 1 mM phenylmethylsulfonyl fluoride, protease inhibitor mix: 1.25 g/ml leupeptin, 0.75 g/ml antipain, 0.25 g/ml chymostatin, 0.25 g/ml elastinal, 5 g/ml pepstatin A). Native immunoprecipitation reactions were incubated for 4 h at 4°C on a shaker. The beads were separated from the supernatant by centrifugation and were washed twice with 500 l of ice-cold immunoprecipitation buffer. Immunoblotting confirmed that RPBs were not lost during the washes (data not shown). Washed ␣FLAG beads were incubated in SDS-PAGE sample buffer for 10 min at 95°C, and aliquots and standard proteins were run on the same 10% Tris-Tricine gels. Non-tagged versions of each nascent polypeptide were translated and analyzed in parallel reactions to determine the background signal. His 6 -Rps9a was used as a standard for the determination of RNCs. Resulting values for Rps9a/b were divided by the factor 0.7 corresponding to the deviation of Rps9a/b from the mean value of all four ribosomal proteins (Table 1). Each experiment was performed at least in triplicate.
Generation of Non-translating and Translating Ribosomes-Efficient puromycin release requires conditions of high ionic strength (32), which interfere with ribosome association of RPBs. To prevent such release of RPBs from ribosomes we have made use of the observation that non-translating ribosomes are efficiently generated in vivo when glucose is removed from the growth medium (33). For the analysis of RPB interaction with randomly translating ribosomes and non-translating ribosomes, cultures of MH272-3f␣ were grown to an A 600 of 0.4 at 30°C on YPD, collected, and resuspended in YPD or in YP medium lacking glucose. Growth was resumed for 10 min at 30°C. Cells were harvested in the presence of 100 g/ml cycloheximide to stabilize translating ribosomes. Preparation of cell extract was carried out by glass bead disruption in 20 mM HEPES-KOH, pH 7.4, 100 mM potassium acetate acetate, 2 mM magnesium acetate, 100 g/ml cycloheximide, 0.5 mM dithiothreitol as described (33). Of each lysate 10 A 260 units were loaded onto a 10.8-ml 15-55% linear sucrose gradient and centrifuged for 2.5 h at 200 000 ϫ g. Gradients were fractionated from top to bottom with a density gradient fractionator (Teledyne Isco, Inc.) monitoring A 254 .

RPB Concentration in Yeast Cells-Quantification of untagged
RPBs and ribosomes in complex mixtures requires quantitative immunoblotting. We have heterologously expressed and puri-fied His 6 -tagged versions of four ribosomal proteins and at least one subunit of each yeast RPB ( Fig. 1A and Table 1. Protein concentrations of the purified standards were determined by two unrelated quantification methods ("Experimental Procedures" and supplemental Table S1). Total extracts derived from logarithmically growing yeast cells containing 3-5 ϫ 10 7 cells/ml were applied to quantitative immunoblotting with purified RPBs as standards. Representative immunoblots (Fig.  1B) and corresponding standard curves (Fig. 1C) are shown. Further calculations are based on the average concentration of subunits that form stable oligomeric complexes (Table 1). Based on this evaluation, a yeast cell contained ϳ310,000 ribosomal particles, 400,000 molecules of NAC, 280,000 molecules of Ssb1/2, and 90,000 molecules of RAC. SRP and the group of RPBs that modify nascent polypeptides enzymatically were ϳone order of magnitude less abundant. A yeast cell contained 20,000 molecules of Map1, and 6,000 -8,000 molecules of Map2, NatA, and SRP each. The ratio between ribosomes and RPBs excludes that the bulk of ribosomes are occupied by the complete set of RPBs at steady state. Rather, the data suggest FIGURE 1. Quantification of RPBs and ribosomal proteins. A, purified RPBs and ribosomal proteins were used as standard proteins. Each 1 g of the purified protein was separated on a 10% Tris-Tricine gel followed by Coomassie staining. Heterodimeric RAC and NAC were purified from Saccharomyces cerevisiae. The Srp54 subunit of SRP and the Nat1 and Ard1 subunits of NatA, Map1, Map2, and Ssb1 were expressed as His 6 -tagged versions in E. coli. For the quantification of ribosomes two proteins of the small ribosomal subunit (Asc1 and Rps9a) and two proteins of the large subunit (Rpl39 and Rpl17a) were expressed as His 6 -tagged versions in E. coli. For details see "Experimental Procedures." B, quantification via immunoblotting. Total cell extract corresponding to 0.6 -2.4 ϫ 10 7 cells of logarithmically growing wild type yeast was separated on 10% Tris-Tricine gels. Standard proteins were applied to the same gel and were analyzed by immunoblotting using antibodies specifically recognizing the proteins of interest. As an example, immunoblots for the quantification of Ssb1/2, Rpl17, and Srp54 are shown. Note that the purified, His 6 -tagged standard proteins have a slightly higher molecular mass. C, calibration curves. Densitometric analysis was performed to determine the range of linearity for each standard and to quantify protein concentrations in the total cell extracts. A summary of the results is shown in Table 1  dynamic cycling on and off ribosomes for all RPBs with the exception of NAC and Ssb1/2 (see below).
Qualitative Analysis of RPB Interaction with Randomly Translating Ribosomes-Association of RPBs with ribosomes and polysomes in total cell extract can be qualitatively demonstrated by sucrose density centrifugation ( Fig. 2A). Resulting ribosome profiles have been previously employed to demonstrate ribosome association of Ssb1/2 (8, 18, 34 -36), RAC (9,18,36), NAC (18,37), and Map1 (11). Although the studies agree on ribosome association of RPBs, the extent varies significantly (e.g. association of zuotin in Refs. 9 and 36 or NAC in Refs. 37 and 18). The variability most likely reflects differences in extract preparation and buffer composition and complicates a comparative evaluation of existing data. Moreover, ribosome profiles of Map2 and yeast SRP have not previously been published. We have now revisited the issue and have analyzed in parallel the distribution of the complete set of RPBs in a polysome-rich extract (Fig. 2B). Gentle extract preparation and physiological salt concentrations revealed that the bulk of NAC, RAC, NatA, Map1, and Map2 was bound to polysomes and 80 S ribosomes. As reported previously, Ssb1/2 (8) was abundant also in the cytosolic fraction. SRP was the only other RPB detected in the cytosolic fraction in significant amounts (Fig. 2B). After release from ribosomes, RPBs colocalized with soluble cytosolic proteins, confirming that comigration in the profiles was due to ribosome association (data not shown).
Quantitative Analysis of RPB Interaction with Non-translating and Randomly Translating Ribosomes-The ratio of ribosomes to RPB in each fraction is a measure of how many ribosomes are occupied by a particular RPB. To analyze the effect of the general translational status on these interactions, we have determined the ratio between ribosomes and RPBs in polysomal fractions as well as in fractions containing non-translating ribosomes (Fig. 3A). On average, 88% of non-translating ribosomes were occupied by NAC, 19% by RAC, 15% by Ssb1/2, 2% by Map1, 2% by NatA, and 1% by Map2. SRP was not detected in fractions containing non-translating ribosomes. 89% of ran-domly translating ribosomes were occupied by NAC, 35% by RAC, 30% by Ssb1/2, 4% by Map1 and NatA, 2% by Map2, and 1% by SRP (Fig. 3B). In general, RPBs displayed a preference for translating ribosomes over non-translating ribosomes, which is consistent with their function. An exception was NAC, which occupied even non-translating ribosomes to a large extent. Please note that Ssb1/2, which approximately equals the number of ribosomes in total extract, occupied only about one third of ribosomes involved in translation (see also "Discussion").
RPB Interaction with RNCs Carrying Either Cytosolic or ERtargeted Nascent Polypeptides-Polysomes carry an undefined mixture of nascent polypeptides with respect to amino acid sequence. To assess how RPB binding was affected in the presence of specific nascent polypeptides, we generated RNCs engaged in the translation of particular nascent polypeptides (Fig. 4). To that end, in vitro translation reactions were performed using truncated, stop codon-less mRNAs as a template. Under the conditions of the experiment, the bulk of nascent polypeptides remained quantitatively and firmly attached to ribosomes (supplemental Fig. S1A). To purify specific RNCs, nascent polypeptides were fused to an N-terminal FLAG tag FIGURE 2. Interaction of RPBs with polysomes. A, ribosome profile of logarithmically growing wild type yeast. Log-phase yeast grown on a rich glucose medium was supplemented with 100 M cycloheximide prior to harvest in order to stabilize translating ribosomes. Total cell extract was applied to sucrose gradient centrifugation. To localize ribosomal subunits (40 S, 60 S), monosomes (80 S), and polysomes, fractionation was monitored at 254 nm. B, localization of RPBs in a polysome-rich ribosome profile. Aliquots of the 20 fractions were analyzed by immunoblotting using antibodies as indicated. On each gel 1/20 of the total cell extract (T) was loaded as a control. Ubp6 was used as a marker for the localization of cytosolic proteins in the gradient; ribosomal proteins Rps9a (small subunit) and Rpl24a (large subunit) were used as markers for the ribosomal subunits.

TABLE 1 Quantification of RPBs and ribosomes in a logarithmically growing yeast cell
Quantifications were performed as outlined in Fig. 1 and are derived from the analysis of at least three independently grown cultures. Protein/subunit per cell is the number of the respective molecule in a yeast cell. Oligomer per cell is an average of the number of subunits contained in one complex. RPBs per 100 ribosomes is the percentage of RPBs compared to ribosomes in a logarithmically growing yeast cell.  (38). FLAG-tagged RNCs could be quantitatively isolated and contained ϳ1.5-2.5% of ribosomes present in translation reactions; reactions thus contained an excess of non-translating ribosomes (supplemental Fig. S1B). Please also note that, apart from the crude yeast extract, no extra protein components were added. Examples of the pulldown experiments and quantifications are given in Fig. 5, A and B.

Protein
Using this experimental setup we have tested whether and how nascent polypeptides of 87-amino acid length affected RPB-ribosome interaction. The nascent polypeptides represented three specific protein biogenesis pathways: Pgk1 (39), a monomeric soluble protein localized to the cytosol; prepro-␣factor (pp␣-factor), a precursor that matures into the secreted pheromone ␣-factor (40); and Dap2 (41), a type II membrane protein that is finally localized to the vacuole (Fig. 4). The data were evaluated under the assumption that a Ն2-fold difference in RPB binding reflected a significant change in affinity. As a result, the amount of RAC, Map1, and NatA bound to RNCs was not significantly affected by the sequence of the nascent polypeptide (Fig. 5C). With respect to the nascent polypeptide-modifying enzymes Map1 and NatA, one has to bear in mind that due to the experimental design neither of the nascent polypeptides represented a substrate (25). Additional experiments are on the way to determine how the affinity of the aminopeptidases and acetyltransferase are affected by substrate polypeptides. Binding of Ssb1/2, NAC, and SRP was modulated by the sequence of nascent polypeptides (Fig. 5C). The three RPBs distinguished between RNCs carrying nascent Pgk1, pp␣factor, or Dap2. Consistent with the exposure of the signal anchor sequence of Dap2, SRP was strongly enriched on Dap2-RNCs. Remarkably, pp␣-factor, which also exposes a signal sequence, did not recruit more SRP to RNCs than Pgk1. Ssb1/2 and NAC were recruited 2-fold less efficiently to Dap2-RNCs compared with Pgk1-RNCs (Fig. 5C). In comparison to nontranslating ribosomes (Fig. 3B) the Ssb1/2 affinity for Dap2-RNCs was of similar strength, whereas NAC interaction with Dap2-RNC was significantly decreased.
Cross-linking of RPBs to Cytosolic and ER-targeted Nascent Polypeptides-The experiments described above revealed the extent to which each type of RNC attracted RPBs. Based on this information we now asked how recruitment correlated with nascent polypeptide interaction. To that end, RNCs were generated under the same conditions as in the RPB binding studies. After RNC isolation, RPB proximity to nascent polypeptides was assessed using a homobifunctional cross-linker reactive toward primary amino groups as represented by the ⑀-amino group of lysines and the N ␣ -amino group of polypeptides (Fig.  4). RAC and Map1 did not form cross-links to any of the nascent polypeptides (data not shown). Nascent Dap2 formed an efficient cross-link to SRP and a weak cross-link to NAC, but no cross-link to Ssb1/2 and NatA. Nascent Pgk1 and pp␣-factor formed cross-links to NAC, Ssb1/2, and NatA, but not to SRP   (39), yeast pp␣-factor is the precursor of the secreted pheromone ␣-factor (40), and yeast Dap2 is a vacuolar type II membrane protein (41). RNCs containing the N-terminal 87 amino acids of Pgk1, pp␣-factor, or Dap2 as a nascent polypeptide were used for RPB binding and cross-linking experiments. For the purification of RNCs, nascent polypeptides were fused to an N-terminal FLAG tag (DYKDDDDK). The position of lysines (K) that provided primary amino groups for the cross-linking reactions is indicated. For crosslinking experiments untagged versions of the proteins were used in which the amino acid at position 2 was changed to a lysine. Helical regions of Pgk1, the N-terminal signal sequence of pp␣-factor, and the signal anchor sequence of Dap2 are indicated. (Fig. 6). The absence of a cross-link between nascent pp␣-factor and SRP differs from previous results demonstrating an efficient crosslink between yeast pp␣-factor and mammalian SRP (42). Introduction of an additional lysine at position 5 (pp␣-S5K) (42) did not alter the cross-linking pattern of pp␣-factor (supplemental Fig. S2). We conclude that the signal sequence of yeast pp␣-factor does not attract yeast SRP to RNCs (Fig. 5C) nor does it interact with SRP (Fig. 6). In fact, RPBs were either in close proximity to nascent Pgk1 and pp␣-factor (Ssb1/2, Nat1) or nascent Dap2 (SRP). Only NAC formed crosslinks to all three nascent polypeptides; consistent with its less efficient binding to Dap2-RNCs, NAC cross-links to nascent Dap2 were weaker than to nascent Pgk1 or pp␣-factor (Fig. 6).

DISCUSSION
The Ratio of RPBs Versus Ribosomes; Dynamics at the Tunnel Exit-Prior to this study, quantitative immunoblotting had been applied to only few RPBs. The best documented example was Ssb1/2, for which a cellular ratio of 3 Ϯ 2 molecules/ribosome was reported (43). We now find that Ssb1/2 is expressed at lower concentrations, approximately equimolar to ribosomes. An elegant high throughput study had previously determined cellular expression levels by epitope tagging the open reading frames of yeast, such that the fusion proteins were expressed under control of their natural promoters (44). In this study the levels of ribosomal proteins were highly variable, and for some RPBs, e.g. Srp54, the expression levels were not determined. However, expression levels of most RPBs, including Ssb1/2, are in excellent agreement. With the exception of Ssb1/2 and NAC, which will be discussed below, RPBs are expressed at substoichiometric levels compared with ribosomes. This shortage strongly suggests that the interaction with ribosomes cannot be static but  Fig. 4 and "Experimental Procedures"). RNCs carrying FLAG-tagged nascent polypeptides were isolated by native immunoprecipitation using ␣FLAG-covered beads. RNCs carrying the same nascent polypeptide but lacking the tag served as a control in parallel reactions. Aliquots of the material recovered on ␣FLAG beads and standard proteins (Fig. 1) were applied to the same Tris-Tricine gel and were subsequently analyzed by immunoblotting. Signals obtained from nontagged RNCs were subtracted as a background from the signals derived from FLAG-tagged RNCs. Quantification was performed as described in Fig. 1 requires cycling of RPBs. SRP affinity is known to be modulated by substrates containing signal sequences (Ref. 45 and references within). What affects ribosome binding in the case of the other RPBs is not understood. For example, RAC binding was not significantly influenced by the specific nascent polypeptides tested in the course of this study (Fig. 5C); however, translation in general seemed to exert a positive effect on RAC binding (Fig. 3B). As RAC acts in concert with Ssb1/2 (20 -22), these two RPBs very likely associate with a single ribosome, at least transiently. However, our data do not favor a model where RAC is recruited preferentially to ribosome⅐Ssb1/2 complexes: Pgk1-RNCs, which carried significantly more Ssb1/2 than non-translating ribosomes, did not carry more RAC. Moreover, RAC is also quantitatively associated with ribosomes obtained from a yeast strain lacking Ssb1/2 (23). In this context it is also interesting to recall that yeast can tolerate very low cellular RAC concentrations without much effect (22). Although details remain to be established, the combined data suggest that RAC binds to ribosomes with high affinity but also in a highly dynamic fashion that ensures its action on a large number of ribosomes.
It is a unique property of Ssb1/2 that it behaves like an integral component of the translating ribosomal particle (43). This stable association led to the plausible model that Ssb1/2 becomes a static component of the ribosome when translation starts and cycles between two rounds of translation (43). It was therefore unexpected to find that only a fraction of translating ribosomes carried Ssb1/2 (Figs. 3B and 5C). The result would be consistent with a model where only some polypeptides are synthesized on ribosomes stably occupied by Ssb1/2. Alternatively, the interaction of Ssb1/2 with translating ribosomes might be more dynamic than anticipated. Release may be coupled to a specific step of the elongation cycle. In this case, tight binding of Ssb1/2 to RNCs, as observed after breaking cells or in vitro, may reflect the absence of ongoing translation.
Our data also suggest that Ssb1/2 may adopt different conformations on the ribosome or, alternatively, possess more than one ribosomal binding site. This would explain how Ssb1/2 that was bound with similar efficiency to pp␣-RNCs (24% occupation) and Dap2-RNCs (17% occupation) formed an efficient cross-link to nascent pp␣-factor but not to nascent Dap2. Because pp␣-factor contains only a single lysine at position 2 whereas Dap2 contains a lysine at the same position plus additional lysines (Fig. 4), we do not favor the possibility that differences in the availability of primary amino groups account for the lack of a Dap2 cross-link. Interestingly, the same applies to NatA that was bound equally well to all RNCs but formed a cross-link only to nascent Pgk1 and pp␣ but not to nascent Dap2. The failure of nascent Dap2 to cross-link to Ssb1/2 as well as to NatA was not confined to a specific length of the nascent polypeptide but was also observed for shorter and longer versions of Dap2. 3 As cross-linking is suited to reveal even short-lived interactions, it seems unlikely that ribosome-bound Ssb1/2 or NatA are even transiently close to nascent Dap2. It will be interesting to identify the sequence attributes of nascent polypeptides that seemingly affect RPB positioning on the ribosome. Experiments are on the way to determine whether it is a general feature of Ssb1/2 and NatA to discriminate SRP substrates.
The affinity of yeast SRP increased from non-translating ribosomes to pp␣-RNCs Х Pgk1-RNCs to Dap2-RNCs. In a previous study fluorescence techniques have been employed to determine affinities of mammalian SRP to wheat germ ribosomes and RNCs at equilibrium (45). In this experimental system the affinity increased from SRP⅐non-translating ribosomes to SRP⅐RNCs lacking signal sequences to various SRP⅐RNCs bearing a signal sequence (45). Thus, our data are in good agreement with respect to the general preferences of SRP and confirm that SRP distinguishes not only between RNCs bearing a signal sequence or not but also between non-translating and translating ribosomes (45). The complete lack of interaction between SRP and the signal sequence of pp␣-factor was surprising, particularly in light of the earlier data (42). However, it confirms work of Walter and co-workers (46), who have shown in vivo that Dap2 requires SRP to be translocated to the ER whereas pp␣-factor does not. Our data support the idea that the affinity of SRP for a signal sequence determines whether an ER-targeted protein enters the SRP-dependent, cotranslational or the SRP-independent, posttranslational pathway (46,47). Interestingly, it was recently found in the E. coli system that proteins containing signal-anchor sequences are selected for cotranslational targeting by SRP at an early stage during biogenesis, whereas nascent secretory proteins were not (5).
SRP and NAC displayed inverse affinity for the RNCs analyzed in the course of this study (Fig. 5C). The observation is consistent with a previous model suggesting that SRP competes with NAC for the same binding site on the ribosome (48). Recent studies have confirmed that SRP and NAC indeed inter-3 S. Oellerer and S. Rospert, unpublished data. FIGURE 6. Interaction of RPBs with nascent polypeptides. A yeast translation extract was programmed with truncated mRNA encoding the N-terminal 87 amino acids of Pgk1 (Pgk1-RNCs), pp␣-factor (pp␣-RNCs), or Dap2 (Dap2-RNCs) (Fig. 4) in the presence of [ 35 S]methionine. RNCs were isolated by centrifugation through a sucrose cushion and were subsequently incubated either in the absence (TOT Ϫ BS 3 ) or in the presence (TOT ϩ BS 3 ) of the homobifunctional cross-linker BS 3 . Aliquots corresponding to 4 ϫ the material of the TOT ϩ BS 3 were subjected to immunoprecipitations under denaturing conditions (IP) with antibodies directed against Nat1, Ssb1, Srp54, and ␣/␤NAC. Samples were run on Tris-Tricine gels and were subsequently analyzed by autoradiography. act with the same ribosomal protein at the tunnel exit (18,49,50). There is evidence, however, that SRP and NAC can simultaneously occupy a single ribosome (18), and our data do not exclude this possibility. The experimental system described in this report shall facilitate future experiments to address these fundamental questions.