Mol Gen Genet (1981) 184:551-556 © Springer-Verlag 1981

Identification of Proteins at the Peptidyl-tRNA Binding Site of Rat Liver Ribosomes Steven Fabijanski and Maria Pellegrini Molecular Biology Section, University of Southern California, Los Angeles, California 90007, USA

Summary. We have identified proteins involved in the peptidyltRNA-binding site of rat liver ribosomes, using an affinity label designed specifically to probe the P-site in eukaryotic peptidyl transferase. The label is a 3'-terminal pentanucleotide fragment of N-acetylleucyl-tRNA in which mercury atoms have been added at the C-5 position of the three cytosine residues. This mercurated fragment can bind to rat liver peptidyl transferase and function as a donor of N-acetylleucine to puromycin. Concommitant with this binding, the mercury atoms present in the fragment can form a covalent linkage with a small number of ribosomal proteins. The major proteins labeled by this reagent are L5 and L36A. Four protein spots are found labeled to a lesser extent: L10, L7/7a, L3/4 and L25/31. Each of these proteins, therefore, is implicated in the binding of the 3'-terminus of peptidyl-tRNA. The results presented here are correlated with other investigations of the structure-function aspects of rat liver peptidyl transferase. Using these data, we have constructed a model for the arrangement of proteins within this active site.

Introduction

The use of site-specific probes to elucidate the structure-function relationships of eukaryotic ribosomes has intensified recently. In particular, affinity labeling of rat liver ribosomes has provided much valuable information about the peptidyl transferase center. Some of the reagents used in these studies have probed the P-site of this enzymatic center with chemically reactive groups positioned near the peptide linkage of peptidyl-tRNA (Czernilofsky et al. 1977; Stahl et al. 1979). Other reagents directed at the A-site have utilized puromycin derivatives in which the adenosine, aminoacyl, or peptidyl moieties have been modified to allow for the covalent attachment of the reagent to ribosomal components (Stahl et al. 1974; Bohm et al. 1979). Other studies have followed the photoincorporation of unmodified puromycin into ribosomes (Reboud et al. 1981). Here we report the affinity labeling of rat liver ribosomes with a novel reagent which is the mercurated analog of peptidyltRNA, first used to probe Drosophila ribosomes (Fabijanski and Pellegrini 1979). This affinity label consists of the 3'-terminal pentanucleotide fragment of N-acetylleucyl-tRNA to which radioactive mercury atoms have been added at each of the three cytosine residues present. The mercury atoms in this reagent are capable of participating in a covalent reaction with sulfhydOffprint requests to : Dr. Maria Pellegrini

ryl-containing protein side chains at the P-site. Because of the position of the chemically reactive groups within the affinity reagent, we have most likely labeled those proteins that form the binding site for the 3'-terminal nucleotides of peptidyl-tRNA. It is clear from work on E. coli ribosomes that this conserved 3'-terminal sequence, C-C-A, of peptidyl-tRNA is critically important in the interaction of this tRNA with the donor site of peptidyl transferase (Quiggle and Chladek 1980). Our results on the proteins present at the tRNA-binding site can be easily correlated with data on other proteins of this active center. We, therefore, present a model for the arrangement of those proteins involved in the peptidyl transferase of rat liver ribosomes. Materials and Methods

Ribosome and Subunit Isolation. Ribosomes were prepared from fresh rat livers according to the method of Reyes et al. (1976). 80S ribosomes were preincubated with puromycin and separated into 40S and 60S subunits by sucrose gradient centrifugation (Sherton and Wool 1972). 60S subunits were collected and pelleted at 120,000xg for 3 h and stored at - 9 0 ° C . For use in the fragment assay, 60S subunits were resuspended in 10 mM Tris-HC1, pH 7.6, and 100 mM KC1 at 15 to 20 A26o units/ml. Preparation of the Mercurated Fragment I and the Fragment Reaction. The Y-terminal fragment oftRNA, C-A-C-C-A(Ac[SH]Leu) (specific activity 2 Ci/mM) was the generous gift of Dr. Peter Butler. Mercuration of the pyrimidine residues with 2°SHg(NOs)2 (specific activity 305 Ci/M) was carried out as previously described (Fabijanski and Pellegrini 1979). A comparison of the specific activity of [a°SHg] and [3HI leucine present in the purified mercurated fragment indicated quantitative addition of [2°3Hg] to all three cytosine residues present in the fragment. The fragment assay was performed according to Monro (1971). The reaction mixture contained 1.5 A260 units of 60S subunits, 15 nM fragment, 1 mM puromycin, 50 mM Tris-HC1, pH 7.6, 400 mM KCt and 15 mM MgC12 in 0.1 ml. The reactions were initiated by the addition of ethanol to 30% (v/v) and maintained at 0° C for 45 rain. The formation of N-acetyl[SH]leucyl puromycin was determined according to Monro (1971). Affinity Labeling Reaction. Mercurated fragment (15 riM) was incubated with 15 A260 units of rat liver 60S subunits and 1 mM 1 Abbreviations used: Fragment, C-A-C-C-A-(Ac[3H]Leu); mercurated fragment, C(Hg)-A-C(Hg)-C(Hg)-A(Ac[SH]Leu);P-site, donor site of peptidyl transferase; A-site, acceptor site of peptidyl transferase; SDS, sodium dodecyl sulfate

0026-8925/81/0184/0551/$01.20

552 puromycin under the fragment assay conditions. An aliquot was taken to determine the amount of N-acetyl[3H]leucylpuromycin formed (Monro 1971). The remainder of the labeled subunits were diluted with an equal volume of 50 mM Tris-HC1, pH 7.6, 5 mM MgCla, 200 mM KC1 and pelleted through 1 ml of 1 M sucrose containing 50 mM Tris-HC1, pH 7.6, 5 mM MgCI2, 200mM KC1 at 120,000×g for 3 h. This removed >90% of the unbound affinity label.

Isolation and Characterization o f Ribosomal Components. Pellets of affinity-labeled ribosomes were resuspended in 1 ml of 10 mM Tris-HC1, pH 7.6, and ribosomal protein and ribosomal RNA were separated using the Mg 2+-acetic acid procedure of Sherton and Wool (1974). Prior to lyophilization, ribosomal proteins were dialysed against 10% acetic acid in acetylated dialysis tubing (Craig 1967). We found that when non-acetylated dialysis tubing was used, a major portion of the radioactively labeled product remained bound to the tubing. During the isolation and electrophoresis of affinity-labeled protein, all sulfhydryl reducing agents such as 2-mercaptoethanol were omitted. Separation of basic ribosomal proteins was carried out using two different two-dimensional electrophoresis methods: (1) the basic-first dimension, acidic-second dimension method of Kaltschmidt and Wittman (1970) as modified by Lastick and McConkey (1976) and (2) the acidic-first dimension, SDS-second dimension method of Mets and Bogorad (1974) as modified by Gorenstein and Warner (1976). Briefly, for gel system (1), the lyophilized protein extracts were resuspended in the LastickMcConkey sample buffer without 2-mercaptoethanol and subjected to electrophoresis in a 4% polyacrylamide tube gel for 1,200 V-h. After equilibration in the soaking solution without 2-mercaptoethanol the first dimension gel was joined to the second dimension separation gel with a 10% polyacrylamide stacking gel (6 M urea, 0.44 M acetic acid, 0.025 M KOH, 0.30% w/v bisacrylamide, 0.50% v/v TEMED). We found that a polyacrylamide rather than an agarose stacking gel allowed better transfer of labeled proteins from the first to the second dimension gels when 2-mercaptoethanol was omitted. Electrophoresis in the second dimension was for 2,200 V.h. For gel system (2), the lyophilized protein extracts were dissolved in 9 M urea, 5% (v/v) acetic acid and electrophoresed in a 4% polyacrylamide tube gel for 900 V.h. After equilibration for 45 rain with 50 ml of 0.5 M Tris-HC1 (pH 6.8), 1% SDS, the first dimension gel was attached to the second dimension with a 4% polyacrylamide stacking gel. The second dimension was electrophoresed for 1,500 V.h. The positions of radioactively labeled ribosomal proteins in the two-dimensional gels were determined by impregnating the gels with Enhance (New England Nuclear), drying, and exposing them to Kodak XAR-5 X-ray film using an intensifying screen. This allows for simultaneous detection of both the /~ and 7 emissions of [Z°3Hg].

Results Activity of the Mercurated Fragment in the Fragment Reaction with 60S Rat Liver Ribosomal Subunits. The fragment assay is a useful method for determining the activity of peptidyl transferase, which is localized in the 60S subunit. The assay measures the transfer of N-acetylleucine from a 3'-terminal fragment of N-acetylleucyl-tRNA to the antibiotic puromycin. These two substrates are bound in the P-site and A-site, respectively, of peptidyl transferase.

Table 1. Peptidyl transferase activity of the mercurated fragment Fragment concentration (rim)

A26o units of 60S subunits/ml

Fmoles of [3H]Leupuromycin produced

15, non-mercurated 15, mercurated

2.5 2.5

147 201

Either mercurated or non-mercurated fragment was used in the fragment assay as described in Materials and Methods

Table 2. Attachment of [2°3Hg] to ribosomal components

[2°3Hg] mercurated fragment [2°3Hg] mercurated fragment plus 2-mercaptoethanol

Ribosomal protein

Ribosomal RNA

97.8% <0.5%

2.2% <0.5%

Following the affinity-labeling reaction with 60S subunits, pooled fractions ofrRNA and ribosomal protein were either treated or not treated with 5% (v/v) 2-mercaptoethanol and analyzed for radioactivity

In order to determine the specificity of binding of the mercurated fragment affinity label, we tested its ability to participate in the 60S-mediated transfer of N-acetylleucine to puromycin. As shown in Table 1, the mercurated fragment is fully active as a substrate in the peptidyl transferase reaction. It was even more efficient than non-mercurated fragment in the production of N-acetylleucylpuromycin. This higher activity may indicate an increased affinity of the mercurated fragment for the P-site. In any case, the activity of the mercurated fragment in the fragment assay clearly shows that it binds correctly in the P-site of peptidyl transferase.

Covalent Labeling of the 60S Subunit by the Mercurated Fragment. After the radioactive mercurated fragment (15 nM) was incubated with 15 A260 units of 60S rat liver ribosomal subunits and 1 mM puromycin, the labeled subunits were purified from unbound fragment by pelleting them through a 1 M sucrose pad (see Materials and Methods). At this stage, the average yield of 60S subunits which retain a bound mercurated fragment is 3%. In order to determine whether the mereurated fragment had become covalently attached to any ribosomal component(s), we first separated the affinity-labeled 60S subunits into total RNA and protein fractions by the Mg z +-acetic acid extraction procedure of Sherton and Wool (1974). The ribosomal RNA precipitate obtained with this separation method may contain trapped ribosomal proteins. These can be removed by repeated washes of the precipitate with 67% (v/v) acetic acid. These washes were pooled with the original ribosomal protein-containing supernatant to determine the relative amount of radioactivity that copurified with either the protein or RNA fraction. As shown in Table 2, over 97% of[Z°3Hg] radioactivity was found associated with the ribosomal proteins. If affinity-labeled 60S subunits were incubated with 5% (v/v) 2-mercaptoethanol prior to dissociation, [2°3Hg] radioactivity does not remain attached to either ribosomal RNA or ribosomal protein. These results, also seen in Table 2, suggest that the attachment of mercurated fragment to the ribosomal proteins is due to a mercury-sulfur bond which can be reduced by 2-mercaptoethanol.

553

Labeling of the 60S Subunit is Site-Specific. To further determine that both the specific binding and covalent attachment of the mercurated fragment were occurring at the peptidyl transferase center, we performed the affinity-labelingreaction in the presence of increasing amounts of non-mercurated fragment. A fixed amount of mercurated fragment was first mixed with a 2.5to 20-fold molar excess of non-mercurated fragment, then reacted with 60S subunits under the fragment assay conditions. Reacted subunits were collected by precipitation with 2 volumes of ice cold ethanol in the presence of carrier 60S subunits. The ribosomal proteins were extracted and the total [2°3Hg] radioactivity associated with the ribosomal protein fraction was measured. The results of the experiments are shown in Fig. 1. Labeling of the ribosomal proteins by [2°?Hg] is clearly reduced by the presence of non-mercurated fragment. The slope of this line follows that of earlier work (Fabijanski and Pellegrini 1979) using Drosophila ribosomes, where the competition was carried out over a 400-fold range of excess non-mercurated fragment. These results indicate that the site of binding of the mercurated fragment is the same as that of the non-mercurated fragment. This evidence is supported by the ability of ribosome-bound mercurated fragment to participate in the peptidyl transferase reaction as shown above. Identification of the Affinity-Labeled Proteins. Separation of affinity-labeled ribosomal protein mixtures was carried out using two different two-dimensional gel techniques. When affinitylabeled proteins were separated in the basic-first dimension, acidsecond dimension gel system (Kaltschmidt and Wittmann 1970; Lastick and McConkey 1976) a large aggregate of protein was visible at the interface of the stacking and separation gels of

(basic)

1D

®

d" '0 x c

3.C

o

o

2£

o 1.9 I.C n

1

2

I

5

1

10

I

20

40

Molar Excess of Fragment

Fig. 1. Competition of mercurated fragment covalent labeling by nonmercurated fragment. 0.75 Aa60 units of 60S subunits were incubated with 0.30 nM mercurated fragment and the indicated concentrations of non-mercurated fragment under conditions of the fragment assay. After incubation at 00 C for 45 min, the 60S subunits were collected and analyzed for radioactivity and protein content as described in the text

the first dimension. Since these gels must be run in the absence of thiol reducing agents (to prevent cleavage of the mercuryprotein linkage), we assume that these aggregates are due in part to the crosslinking of the numerous sulfhydryl groups present in the ribosomal proteins. These aggregates cannot penetrate the separation gel. The first dimension gel in this system is run at pH 8.6. Many ribosomal proteins are only slightly soluble at this pH. This problem compounds that of not being able to include sulfhydryl reducing agents in the running of these gels.

,®

A 3

0

I

5

6

o~

I

lO

I

12

7,7u Q8

9 I

4

~11

25

25

18,]8A

~23A

I 3o~

] 9 ~ 21

I

3~

52

27A ~ 27

28

29

35A dP35 54 n o,j

56 36A 9D

® Fig. 2A, B. Diagram of two-dimensional gel analysis and autoradioactivity of affinity-labeled 60S ribosomal proteins. A is a diagram of the basic-first dimension acidic-second dimension gel of affinity-labeled 60S ribosomal protein. B is a corresponding autoradiograph. Proteins are numbered according to the system of McConkey et al. (1979)

554

1D (acid)

®

(-!A

-9

O O

3-4 , I D

0

~6

4D8 ,~,14 I0A 1 0 - 1 ~ ~ 1 9 9~ 1 7 ~ ~_._~-,o, 18A ~ 5 ~ ~ 24,~ 36A 11~ 21-23A ,~D 2 ' ~ 1 P ~ 3 5 2~

31

25

27,32

34 0 3 6

35A

(/3 C3 (/3

v

C3 O4

® Fig. 3A-C. Two-dimensional gel analysis and autoradiography of affinity-labeled 60S ribosomal proteins using the acidic-first dimension, SDS-second dimension gel system. A is a diagram of the pattern of 60S affinity-labeled proteins. B and C are autoradiographs of these labeled proteins. Relative amounts of protein loaded on the gels and length of exposure of the autoradiographs are detailed in the text The fluorographic observation of affinity-labeled 60S ribosomal proteins separated in this basic-first dimension, acidicsecond dimension gel system yielded only one area of radioactivity following a 40-day exposure. This spot corresponded to protein LT/Ta (uniform numbering system of McConkey et al. 1979) as shown in Fig. 2. In this and subsequent gel experiments, we found the autoradiographic spot to be slightly displaced to the upper left of the stained spot. This is as expected for a protein covalently bound to a mercurated nucleotide(s). Because of the solubility problems of our labeled ribosomal proteins we chose to separate them in an alternate two-dimensional electrophoresis system, that of Mets and Bogorad (1974). In this system, the first-dimension gel is run at pH 5.0, a pH at which ribosomal proteins are very soluble, and sulfhydryl groups are fully protonated. When affinity-labeled proteins are run in this gel system very little material aggregates at the top of the gel, even in the absence of 2-mercaptoethanol. In the second dimension of these gels which contain 17% (w/v) polyacrylamide and 0.1% (w/v) SDS, again the labeled proteins were found to remain soluble. Fluorographed gels of affinity-labeled protein run in this gel system showed one area of radioactivity (see Fig. 3A). The amount of labeled protein loaded onto this gel was equivalent to that run in the basic-first dimension, acidicsecond dimension gel system discussed above, but the exposure time was only 15 days. However, as shown in Fig. 3 A and B, the protein identified as containing [2°3Hg] residues was L5, not L7/7a as identified

in the previous gel system. In an attempt to resolve this discrepancy, we loaded three times the amount of affinity-labeled ribosomal proteins on an acidic-first dimension, SDS-second dimension gel. A 5-day exposure of this gel is shown in Fig. 3C. In this experiment we were able to identify more affinity-labeled protein species, including L36A, L3/4, L10-13, L25/31, and L7/ 7a. Clearly, the acidic-first dimension, SDS-second dimension gel system allowed us to identify more labeled species and, in the absence of substantial amounts of protein aggregation, to better distinguish the major and minor labeled proteins. The fact that protein LT/7a was the major labeled species found on the basic-first dimension, acidic-second dimension gels and L5 was the major labeled species on the acidic-first dimension, SDS-second dimension gel underscores the problem of solubility in general and possibly selective solubility of our affinity-labeled ribosomal proteins in the former gel system. The ability of the majority of the proteins, both labeled and unlabeled, to enter the acidic-first dimension, SDS-second dimension gels gave us confidence we had identified all of the labeled species. The numerical identification of the labeled proteins was done by correlating our gels with those of Madjar et al. (1979), who have identified the proteins of rat liver ribosomes in four different two-dimensional gel systems. Their basic-first dimension acidicsecond dimension gel was the same as that used in this study, while their acidic-first dimension SDS-second dimension gel was similar enough to ours as to be able to correlate many of the proteins. As a result, the number assigned to each individual

555 A site

protein is retained by that protein in each of the two systems. As stated above, these numbers correspond to the uniform nomenclature of McConkey et al. (1979).

Use of the Acidic-First Dimension, SDS-Second Dimension Gel System. We wish to point out the advantages of the acidic-first dimension gel system (Mets and Bogorad 1974; Gorenstein and Warner 1976) for analysis of ribosomal proteins. Our data strongly suggest that the basic-first dimension, acidic-second dimension gel system (Kaltschmidt and Wittmann 1970; Lastick and McConkey 1976) run here in the absence of 2-mercaptoethanol has excluded several affinity-labeled proteins from the analyses. In addition, the equal inclusion of all labeled proteins in the acidic-first dimension, SDS-second dimension gels allowed both shorter exposure times and easy qualitative comparison of the degree of labeling of the different protein species. Our laboratory has found the acidic-first dimension, SDS-second dimension gel system to be superior for the separation and analysis of Drosophila as well as rat liver ribosomal proteins even when thiol reducing agents are present (data to be presented elsewhere). Discussion

Site-Specific Labeling. We have presented evidence that the mercurated fragment of peptidyl-tRNA functions as an affinity label directed to the P-site of rat liver 60S peptidyl transferase. The ability of the mercurated fragment to function as a substrate for peptidyl transferase indicates that it can bind in a normal fashion to the P-site. The fact that the mercurated fragment and the non-mercurated fragment both compete for the same site also indicates this affinity label binds correctly to the P-site. Once bound in the P-site, the mercurated fragment becomes covalently attached to ribosomal proteins. This linkage is sensitive to a thiol reducing agent as is expected for a mercury-sulfur bond. The limited reaction of this label with two major and four minor protein species also suggests site-specific labeling. Finally, Kruse et al. (1980) have shown that alkylation of the C-C-A and of yeast Phe-tRNA does not impair the biological activity of this tRNA. Model for the Rat Liver Peptidyl Transferase Center. Our identification of two major and four minor affinity-labeled 60S proteins in the tRNA-binding portion of the P-site fits well with the results of other investigators who have probed the site of peptidyl transferase function. Since eukaryotic peptidyl transferases are known to be stimulated by sulfhydryl modifications (Carrasco and Vazquez 1975), the ability of the mercurated fragment to function more efficiently in the fragment assay (Table 1) may stem from the fact that it can modify ribosomes through attachment via sulfhydryl residues at the P-site. Reyes et al. (1977) showed that by selective salt washing-removal of certain proteins from rat liver ribosomes, that this sulfhydryl stimulation of peptidyl transferase could be abolished while maintaining the activity of peptidyl transferase. One of the proteins lost in this procedure was L362, a protein we have shown to be the major labeled species in this study. In addition, Reboud et al. (1980) found that L10 was protected against salt removal by prior fixation of deacylated t R N A onto 60S subunits. Other affinity-labeling studies by Czernilofsky et al. (1977) using p-nitrophenoxycarbamoyl-[3H]phenylalanyl-tRNA identified the following major labeled proteins, L32/33, L362, L21, and L23, as being at or near peptidyl transferase. Their reagent is reactive toward a nmnber of protein groups and the chemically reactive group is bound at a site somewhat distant from the

P site

L4 L6 L24

31 'o

Fig. 4. Model of the arrangement of proteins within the peptidyl transferase center of rat liver 60S subunits. The most frequently labeled proteins are represented as spheres of equal diameter, since their exact shape in the ribosome is unknown. Proteins listed on the side are those found to be labeled to a lesser extent

3' nucleotides of t R N A ; however, L36 appears labeled in both this study and our own. 2 Stahl et al. (1974) using 2-nitro-4azidobenzoic acid-substituted-phenylalanyl-tRNA found LI0 to be the major labeled protein and L17 and L23 to be the minor labeled proteins. Vlasov and Vesterman (1976) found when labeling 80S rat liver particles with N-chloroambucilyl-phenylalanylt R N A that proteins L5, L25, L31, and L32 were reactive, and when using iodoacetyl- or N-bromoacetyl-phenylalanyl-tRNA that L4, L6, L10, L l l , L13, and L30 were labeled. When comparing all of these results with our data, one must bear in mind that the reactive moieties on the labels of Czernilofsky et al. (1977), Vlasov and Vesterman (1976), and Stahl et al. (1974) were located at the amino terminus of the anainoacylt R N A , while the reactive groups in our label were located on the 3'-terminal cytosine residues of the t R N A , a considerable distance from the amino group of leucine in the fragment. Therefore, we have labeled that portion of the P-site involved in binding the C-C-A end of the tRNA, while these investigators have labeled proteins near the peptide linkage of peptidyl-tRNA. We have correlated these data and our own and propose a model of the proteins involved in peptidyl transferase. This model is presented in Fig. 4. The proteins that are labeled most frequently are represented by spheres, and the minor labeled species are listed on either side of the peptidyl transferase center. L28/29 is labeled only by puromycin analogs and in good yield, so it is shown only to reside in the A-site. L10 and L32/33 are labeled by both P-site-specific reagents and puromycin analogs. Since we also find L10 3 as a minor labeled species from the peptidyl-tRNA binding site, we have placed it close to the t R N A binding sites of both the A- and P-sites. L21, L23, and L36A are labeled by P-site reagents, exclusively. L5 is a major labeled species in our studies and minor in the others mentioned 2 In studies other than this one, L36 and L36A are not resolved by the gel systems used. In our work clearly L36A is the reactive protein 3 Our tRNA affinity label reacted with one of two proteins, LI0 13, which are unresolved on the acidic-first dimension, SDS-second dimension gels. Since only L10 in this group of proteins is labeled by other peptidyl transferase-specific reagents we believe that we have probably labeled this protein rather than L13

556 so we propose a role for this protein in binding the C-C-A terminus of the peptidyl-tRNA. This model is also supported by data other t h a n from affinitylabeled studies. Uchiumi et al. (1980) cross-linked L5, L3, L35, L7/7a and L21/23/23a in various combinations, indicating that these proteins are close to each other in the 60S particle. Although this model is by no means complete it is based on a n u m b e r of independent reports, and fits much of the available data. Hopefully, further work in identifying proteins at functional sites in rat liver ribosomes will allow a clearer understanding of the arrangement of the macromolecules within this eukaryotic ribosome, much like what has been accomplished for E. coli ribosomes.

Acknowledgements. This work was supported by Grant GM-24572 from the National Institutes of Health, USA. M.P. is a fellow of the Alfred P. Sloan Foundation. We wish to thank Peter Butler for the generous gift of fragment, Robert Traut for helpful discussions, and Sarah Wright for preparation of this manuscript.

References B6hm H, Stahl J, Bielka H (1979) Photoaffinity labelling of rat liver ribosomes by N-(2-Nitro-4-azidobenzoyl) pnromycin. Acta Biol Med Germ 38:Secte 1447 i452 Carrasco L, Vazquez D (1975) The involvement of sulphydryl groups in the peptidyI transferase centre of eukaryotie ribosomes. Eur J Biochem 50 : 317-323 Craig LC (1967) Techniques for the study of peptides and proteins by dialysis and diffusion. Methods Enzymol 11 : 870 905 Czernilofsky AP, Collatz E, Gressner AM, Ktichler E, Wool IG (1977) Identification of the tRNA-binding sites on rat liver ribosomes by affinity labelling. Mol Gen Genet 153:231-235 Fabijanski S, Pellegrini M (1979) Affinity labeling of a reactive sulfhydryl residue at the peptidyl transferase P site in Drosophila ribosomes. Biochemistry 18:5674 5679 Gorenstein C, Warner JR (1976) Coordinate regulation of the synthesis of eukaryotic ribosomal proteins. Proc Natl Acad Sci USA 73:1547 1551 Kaltschmidt E, Wittmann HG (I970) Ribosomal proteins. VII. Two dimensional polyacrylamide gel electrophoresis for fingerprinting of ribosomal proteins. Anal Biochem 36:401-412 Kruse TA, Siboska GE, Sprinzl M, Clark BFC (1980) The effect of chemical modification of the CCA end of yeast tRNA on its biological activity on ribosomes. Eur J Biochem 107:1-6 Lastick SM, McConkey EH (1976) Exchange and stability of HeLa ribosomal proteins in vivo. J Biol Chem 250:2867-2875

Madjar JJ, Arpin M, Bursson M, Reboud JP (1979) Spot position of rat liver ribosomal proteins by four different two-dimensional electrophoreses in polyacrylamide gel. Moi Gen Genet 171 : 121-134 McConkey EH, Bielka H, Gordon J, Lastick SM, Lin A, Ogata K, Reboud JP, Traugh JA, Traut RR, Warner JR, Welfle H, Wool IG (1979) Proposed uniform nomenclature for mammalian ribosomal proteins. Mol Gen Genet 169: i 6 Mets LJ, Bogorad L (1974) Two-dimensional polyacrylamide gel electrophoresis: an improved method for ribosomal proteins. Anal Biochem 57:200 213 Monro RE (I971) Ribosome peptidyl transferase: the fragment reaction. Methods Enzymol 20 : 472-481 Quiggle K, Chladek S (1980) The role of the cytidine residues of the tRNA 3'-terminus at the peptidyl transferase A- and P-sites. FEBS Lett 118:172-175 Reboud AM, Dubost S, Buisson M, Reboud JP (1980) tRNA binding stabilizes rat liver 60S ribosomal subunits during treatment with LiC1. J Biol Chem 255:6954-6961 Reboud AM, Dubost S, Buisson M, Reboud JP (1981) Photoincorporation of puromycin into rat liver ribosomes and subunits. Biochemistry 20 : 5281-5288 Reyes R, Vfizquez D, Ballesta JPG (1976) Structure and function of rat-liver ribosomes. Modification by 2-methoxy-6-nitropone treatment. Eur J Biochem 67 : 267 274 Reyes R, Vfizquez D, Ballesta JPG (1974) Peptidyl transfel:ase center of rat-liver ribosome cores. Eur J Biochem 73:25 31 Sherton CC, Wool IG (1972) Determination of the number of proteins in liver ribosomes and ribosomal subunits by two-dimensional polyacrylamide gel electrophoresis. J Biol Chem 247:4460-4467 Sherton CC, Wool IG (1974) The extraction of proteins from eukaryotic ribosomes and ribosomal subunits. Mol Gen Genet 135:97-112 Stahl J, Dressler K, Bielka H (1974) Studies on proteins of animal ribosomes. Affinity labelling of rat liver ribosomes by N-bromoacetylpuromycin. FEBS Lett 47 : 167-170 Stahl J, B6hm H, Pozdnjakov VA, Grishovich AS (1979) Photoaffinity labelling of rat liver ribosomes by phenylalanine tRNA N-acetylated by 2-nitro-4-azidobenzoic acid. FEBS Lett 102:273-276 Uchiumi T, Terao K, Ogata K (1980) Identification of neighboring protein pairs in rat liver 60S ribosomal subunits cross-linked with dimethyl suberimidate or dimethyl 3,3' dithiobis propionimidate. J Biochem 88:1033 1044 Vlasov VV, Vesterman P (1976) Modification of rat liver ribosomes with alkylating derivatives of tRNA. Mol Biol (MOSC) 10:670-674

C o m m u n i c a t e d by K. Isono Received May 5 / September 28, 1981