Eur. J. Biochem. 73, 25-31 (1977)
Peptidyl Transferase Center of Rat-Liver Ribosome Cores Ramon REYES, David VAZQUEZ, and Juan P. G. BALLESTA Instituto de Bioquimica de Macromolbculas, Centro de Biologia Molecular, Consejo Superior de Investigaciones Cientificas and Universidad Autonoma de Madrid (Received July 13/November 2, 1976)
Protein-deficient particles have been obtained by treating rat liver 80-S ribosomes or their 60-S subunits with 1 M NH4Cl in the presence of 50 % ethanol at 0 "C (Po-cores) and 37 "C (P37-cores). The Po-cores from 8 0 4 ribosomes are totally inactive in polyphenylalanine synthesis but fully active in the 'fragment assay' to test peptidyl transferase activity. The polymerizing activity of the cores is restored up to 40-50% of control activity by incubation in the presence of the split proteins. Three proteins are totally lost in the treatment, namely proteins L12, L40/41 and S25. A series of up to nine different spots in the region of the L40/41 proteins are detected when the split fraction is analyzed by two-dimensional electrophoresis. This series of spots is, however, reduced to only two proteins when the second dimension is carried out in the presence of sodium dodecylsulphate. 80-S ribosome-derived P37-cores are about 80 % active in the 'fragment reaction' while 60-Ssubunit-derived particles are inactive in this assay. The inhibitory effect of a number of antibiotics is differentially affected by the treatment suggesting different localization of their binding sites. A comparative study of the proteins released by treatment in the two types of particles suggests the involvement in the peptidyl transferase center of the ribosome of one or more of the following proteins: L21, L24, L27, L28-and L36. The treatment of bacterial ribosomes with 1 M N&Cl in the presence of 50% ethanol specifically removes a few proteins. Thus only proteins L7 and L12 are released when the treatment is performed at 0 "C, yielding particles inactive in elongation-factordependent functions [l 1. In addition higher temperatures (37 "C) almost totally remove proteins L10, L11 and partially proteins L1, L5, L8/9 and L25 [2- 51. In the case of mammalian organisms the presence of proteins of similar characteristics to L7 and L12 has been detected in the ribosomes of rat liver by using immunological techniques [6]. However, no attempt has been reported for preparing protein-deficient particles derived from mammalian ribosomes using the ammonium chloride/ethanol method. We have prepared core particles from rat liver ribosomes using the latter technique at 0 "C and 37 "C and their functional and structural characteristics are reported here. Abbreviations. Po-proteins, split proteins released at 0 "C; P v proteins, split proteins released at 37 "C.
MATERIALS AND METHODS Ribosomes and Ribosomal Subunits
Ribosomes from Wistar male rats (100- 120 g) were prepared as previously described [7]. Ribosomal subunits were separated by centrifugation on convex exponential sucrose gradients using 8 0 3 ribosomes pretreated with puromycin [8]. All ribosomal preparations were kept in buffer made of 25 mM KCl, 5 mM MgC12, 50 mM Tris-HC1 pH 7.4 and 7 mM 2-mercaptoethanol in liquid nitrogen.
Preparation of Core Particles by Treatment with NH4C1 and Ethanol
A procedure similar to the one described by Hamel et al. [l] was followed. Either 80-S ribosomes or 60-S subunits at 4 mg/ml final concentration in 1 M NH4C1,lO mM imidazole-HC1pH 7.4,20 mM MgC12 and 1 mM 2-mercaptoethanol kept at 0°C or 37°C with continuous shaking were supplemented with
26
0.5 vol. of ethanol. After 10 min another 0.5 vol. of ethanol was added and the solution shaken for a further 5 min. The particles were then centrifuged at 27000 x g for 10 min. The supernatant was kept for recovering the split proteins and the pellet was treated in the same way for a second time. The particles were finally resuspended in 20 mM Tris-HC1 pH 7.4,11 mM KC1, 1 mM MgC12, 7 mM 2-mercaptoethanol, dialyzed against the same buffer for 2 h at 0 "C and stored in liquid nitrogen. Split Protein Fractions
The proteins were precipitated from the supernatants by addition of 2.5 vol. of acetone at - 20 "C, centrifuged at 15000 rev./min for 20 min and the pellets resuspended in 10 mM imidazole-HC1 pH 7.4, 2 mM 2-mercaptoethanol, dialyzed for 2 h against the same buffer and stored in liquid nitrogen. The concentration of proteins in the preparations was estimated by the procedure of Lowry et al. [9] using bovine serum albumin as a standard.
Peptidyl Transferase of Rat-Liver Ribosome Cores
RESULTS Activities of Po-Cores
Cores derived from rat liver 80-S ribosomes treated with 1 M NH4Cl and 50% ethanol at 0° C (Po-cores) are inactive in a phenylalanine-polymerizing system but can be reactivated upon addition of the split protein fraction (Po-proteins) and particles with up to 50% of the control activity obtained (Fig. 1). However the particles are quickly inactivated upon storage even in liquid nitrogen. However, the peptidyl transferase activity of the treated ribosomes or 60-S subunits is not diminished by the treatment. In fact some stimulation that varies from preparation to preparation has always been observed, in the particles derived from 80-S ribosomes (Table 1).
Activity Assays
All the activity tests used have been previously described in detail [7,10]. [3H]Phe-tRNA prepared from commercial yeast tRNA from General Biochemical and [3H]phenylalanine (7.8 Ci/mmol) from the Radiochemical Center (Amersham) was used for the polyphenylalanine synthesis assay. The 'fragment reaction' and the sparsomycin-dependent binding of (U)C-AC-C-A-[3H]Leu-Acwas carried out following standard methods [I 11 using (U)C-A-C-C-A-[3H]Leu-Ac prepared from [3H]Leu-tRNA (53 Ci/mmol) treated with ribonuclease T1 [ l l ] . Two-Dimensional Polyacrylamide Gel Electrophoresis
Standard two-dimensional electrophoresis of proteins was carried out in the conditions of Sherton and Wool [12]. We have also used the nomenclature of proteins adopted by these authors since the pattern of proteins obtained in our gel plates essentially coincides with that reported by them with the small variations described elsewhere [7]. When dodecylsulphate was used in the second dimension of the electrophoresis system the procedure of Martini and Gould [13] was followed. However, the conditions for the electrophoresis were modified by running the gels at 25 V for 14h at room temperature since a small-size apparatus was used [14]. rRNA electrophoresis was carried out in the conditions described by Loening [15] in tubes of 4 mm diameter and 10 cm length, run at constant power (1 Wjtube) until the bromophenol blue marker reached the bottom of the tubes.
PO-proteins (kg)
Fig. 1. Polyphenylalanine synthesis by Po-cores derived from 80-S rihosomes supplemented with increasing amounts of Po-protein. Core particles in 50p1 of 6 0 m M KCI, 1 5 m M MgC12 and 2 0 m M Tris-HCI buffer, p H 7.4 were incubated in the presence of these split proteins at 37 "C during 30 min. The samples were then placed o n ice and completed with the components required for the polyphenylalanine synthesis assay. The activity test was carried out at 37 "C during 20 min
Table 1. Peptidyl transferase activity of Po-cores The reaction mixture in the standard ionic conditions (Methods) contained 1 mg/ml of ribosomes and 8 nM of the (U)C-A-C-C-A[3H]Leu-Ac fragment. The values shown in the table have been calculated after subtraction of 330 counts/min of background obtained in the absence of particles. This background represent between 5 and 10 of the radioactivity in the samples Particle
Core particle derived from
80-S ribosomes
60-S subunits
counts min-' (pmol ribosomes)-' _ _ 145 255 184 252 -
Control Po-cores
R. Reyes, D. Vazquez, and J. P. G. Ballesta
21
Fig. 2. Two-dimensionul gel electrophoresis of Po-proteinsfrom 60-S subunits
Fig. 3 A . Acidic proteins in the Po-proteinsfrom 80-S ribosomes. The electrophoresis was run half of the usual time in the first dimension. In (B) dodecylsulphate was present in the second dimension
Proteins Released from Po-Cores The analysis of the Po-proteins obtained from 60-S subunits by two-dimensional gel electrophoresis shows the presence of protein L12 and a series of up to nine spots in the region of proteins L40/41 (Fig.2). In addition protein S25 was also present in Po-proteins from 80-S ribosomes (not shown). When the second dimension of the gel electrophoresis was run in the presence of dodecylsulphate two spots were detected in the region corresponding to the L40/41 proteins (Fig. 3). Four acidic proteins named A1 to A4 moving to a position similar to that described here for L40/41 also giving two spots in dodecylsulphate have recently been reported by Wooletal. [16]. However these
authors do not identify them with L40/41, which according to them have a less acidic character and move behind the A1 -A4 series. In our case the release by 1 M ammonium chloride and ethanol, indicates that these proteins might be similar to L7/12 of E. coli ribosome and if so, equivalent to those originally named L40/41 [ 6 ] . The proteins present in the split protein fraction are totally absent from the core particles indicating a high specificity of the treatment in their release. Activities of Ps,-Cores
Cores derived from both 80-S ribosomes and 60-S subunits treated with 1 M NH4Cl and 50% ethanol
Peptidyl Transferase of Rat-Liver Ribosome Cores
28
0’ 0
I
1
I
I 2 3 103.Fragment /ribosomes (molImol)
I
I
4
t ’
0.01
I
I
.o
0.1 1 N - Ethylmaleirnide (mM)
I
10
Fig. 4. Peptidyl transferase activity of P3-r-coresfrom80-S ribosome (a) and 60-S subunits (A). The activity is represented as a percentage of the untreated control particles as a function of the ratio of (U)C-A-C-C-A-[3H]Leu-Ac fragment to particles in the reaction mixture
Fig. 5. Effect of N-ethylmaleimide on the peptidyl transferase of so-S ribosomes ( O ) , 60-S subunits (0) and P3,-cores derivedfrom 80-S ribosomes ( 0 ) .The ‘fragment reaction’ was carried out as in Table 1
Table 2. Binding of ( U)C-A-C-C-A-[3H]Leu-Ac to P37-cores particles in the presence of sparsomycin Assay carried out by filtration in the ionic conditions of the ‘fragment reaction’ in the presence of 10pM sparsomycin and 20 or 40 pmol ribosomes
Stimulation of Peptidyl Transferase Center by N-Ethylmaleimide
Particle
Fragment bound with ribosomes 20 pmol 40 pmol counts/min
80-S ribosomes P37-cores
3301 162
5141 1237
(P37-cores) are unable to polymerize polyphenylalanine as is the case with Po-cores but unlike these particles they do not reconstitute their polymerizing activity by addition of the A7-protein fraction in any of a variety of ionic and temperature conditions that we have tested. Inactivation of the particles is not due to degradation of rRNA during the treatment since the RNA extracted from the cores presented the same band pattern in polyacrylamide gel electrophoresis as controls (not shown). When tested in the ‘fragment reaction’ for peptidyl transferase the P37-cores derived from 80-5 ribosomes displayed up to 80 % control activity while the particles derived from 60-S subunits were inactive (Fig. 4). Surprisingly the activity of the 80-S-ribosome derived P37-cores in binding the (U)C-A-C-C-A-[3H]Leu-Ac fragment in the presence of sparsomycin is drastically inhibited (Table 2).
- SH group reagents strongly stimulate the peptidyl transferase activity of mammalian ribosomes [17]. Since the effect caused by these reagents seems to be at the level of the peptidyl transferase center of the ribosome, we were interested in studying the behaviour of the protein-deficient particles treated with N-ethylmaleimide. Fig. 5 clearly shows that N-ethylmaleimide indeed stimulates the activity of 80-S ribosomes and 60-S subunits in the ‘fragment reaction’, but has a slightly inhibitory effect on the activity of the P37cores
Eflect of Peptidyl Transferase Inhibitors on P37-Core Activity
A number of peptidyl transferase inhibitors are known that directly interact with the ribosome active center [18]. Their binding sites are closely related and in some cases it is difficult to discern independent sites by binding studies. The study of the inhibitory effect of these compounds in the peptidyl transferase of the P37-cores may be of great value in understanding the structure of the antibiotic binding sites and the peptidyl transferase center as well. Table 3 summarizes the results obtained with nine different compounds used that according to their inhibitory effect fall in three groups : actinobolin, anisomycin, griseoviridin, blasticidin S and gougerotin inhibited the activity of the P37-cores to the same extent as the control 80-S ribosomes. Trichodermin inhibited the core particles less strongly, while sparsomycin, narciclasine and
R. Reyes, D. Vazquez, and J. P. G. Ballesta
29
Table 3. Effect of peptidyl transferase inhibitors on Ps7-core activity Fragment assay was carried out as in Table 1 in the presence of the indicated drug concentrations and measured as a percentage of the activity of the controls without drug Antibiotic
Concentration Activity in 'fragment reaction' of 80-S ribosomes
P37-cores
1000
32
34
Anisomycin
100 10
8 32
6 32
Trichodermin
100 10
4 11
18 43
100 1
13 93
7 58
21 52
10 30
Narciclasine
0.2 0.1
+
Protein
+,
80-S ribosome
60-S subunit -
Actinobolin
Sparsomycin
Table 4. Proteins released from ribosomes and 6 0 4 subunits by the treatment at 37°C For P~7-proteins,+ or - represents proteins present or absent in the two-dimensional electrophoresis of split fractions. For P37cores, represents protein present in similar amounts to those in control particles; protein in smaller amounts; tr, protein in trace amounts; - , protein absent
Amicetin
1000
80
50
Griseoviridin
1000 100
8 34
10 38
Gougerotin
1000 100
7 43
3 33
Blasticidin S
100 10
10 49
9 49
amicetin were approximately 50 % more inhibitory in the core particles. Proteins Released in the Treatment at 37°C Table4 shows the proteins found in the split fraction obtained by the treatment of 80-S ribosomes and 60-S subunits as well as the quantities of these proteins in the treated particles. Contrary to what occurs at O'C, the release of some proteins at 37°C is only partial. Proteins L6, L10, L12, L14, L24, L28, L35, L36, L38, L39, L40/41, XI and Xz are present in the Pj7protein fraction from 80-S ribosomes. In addition L21 and L27 are also found in the Pj7-protein fraction from 60-S subunits. Proteins L6, L12, L14, L35, L38, L39, L40/41 are totally absent or present only in trace amounts in the core particles from 80-S ribosomes. The same proteins plus L24, L28 and L36 are missing in core particles from the 60-S subunits. DISCUSSION Treatment of rat liver ribosomes with 1 M ammonium chloride in the presence of 50% ethanol at 0 "C releases the acidic proteins equivalent to proteins L7 and L12 split off the ribosomes of E. coli by a
L6 L10 L12 L14 L21 L24 L21 L28 L3 5 L36 L38 L39 L40 L4 1
x1 x2
P37-protein Pw-cores
P37-protein
P37-cores
+ + + + + + + + + + + + + +
+ + + + + + + + + + + + + + + +
+
f -
-
f
*
similar procedure [l]. Findings were similar with yeast [19] and Artemia salina [20] ribosomes. The functional equivalence of these proteins has been proven by cross-reactivation and immunochemical methods [6,19,20] and in the case of rat liver ribosomes were named L40 and L41 [6]. Wool and coworkers [16] also identified up to four different proteins in a region of the gel plates similar to that in which we found a series of up to nine spots. However, they do not consider these proteins to be the original L40 and L41 which according to their interpretation of the gels move behind in the first dimension. Our results seem to indicate that the proteins equivalent to the bacterial L7 and L12 might be in fact the ones that move to the most acidic position. The proteins that Wool et al. previously named L40/41 and now name Axl and Ax2 might correspond to our protein XI [7] that occasionally appears as a double band. The significance of the presence of many spots in the L40 and L41 region is at present obscure. It might be due to different forms of the two molecules detected in dodecylsulphate, modified in some way (possibly phosphorylation or acetylation) and in this sense it is interesting to note that Warner et al. [21] have recently reported the phosphorylation in vivo of the equivalent proteins in yeast ribosomes. Nevertheless due to the probably high mutual affinity of these proteins, already reported for proteins L7 and L12 of E. coli [21], we cannot discard the possibility that they represent aggregates with different proportions of the two proteins detected in dodecylsulphate, a
30
process favoured by overloading of the gels in order to make the proteins visible. Besides L40 and L41 two more proteins L12 and S25 are totally removed from the ribosome by the treatment at 0°C. These proteins are able to reincorporate to the cores since ribosomes active in a polymerization system are reconstituted (Fig. 1). In fact these proteins are present in the reconstituted particles when analyzed by two-dimensional gel electrophoresis (data not shown). At present we do not know the functional or structural role of these proteins but the position in gel plates and the easy removal of protein L12 makes it tempting to suggest its equivalence to protein L10 of E. coli. In any case it is clear that its release does not affect the peptidyl transferase center of the ribosome. The results obtained with the P37-cores are interesting in relation to the differences detected in the peptidyl transferase activity of the particles derived from 80-S ribosomes and 60-S subunits. These dissimilarities can be related to the different protein removed by the treatment in each case. Thus for instance proteins L21 and L27 while not affected in 80-S ribosomes are partially released in 60-S subunits and proteins L24, L28 and L36 which are totally removed from subunits are only partially missing in the 80-S-ribosome-derived cores. One or more of these proteins could be implicated in the loss of activity of the peptidyl transferase center in 60-S subunits. It is interesting to note that protein L21 among others, has also been pointed out as possibly implicated in the peptidyl transferase of rat liver ribosomes by the use of different experimental approaches, such as treatment with KCL [7] and chemical modification with 2-methoxy-5-nitrotropone [lo]. This protein among others is also labeled by a chemically reactive analoge of peptidyl-tRNA that reacts with ribosomal components close to the peptidyl transferase (Czernilafsky, Collatz, Kuechler, Gressner, personal communication). In spite of the activity that P37-cores from 80-S ribosomes display in the ‘fragment reaction’ (70 - 80 % of control) and its high sensitivity to sparsomycin inhibition, their peptidyl transferase center seems to have low activity in the sparsomycin-dependent binding of (U)C-A-C-C-A-[3H]Leu-Ac. The apparently contradictory results could be a consequence of the different conditions used in the two tests. However they might also imply that the stabilization of the (U)C-A-C-C-A-[3H]Leu-Ac fragment binding to the ribosome is an independent effect of the inhibition caused by sparsomycin. The drug probably inhibits peptidyl transferase by binding to the A-site [22] inducing at the same time a change of conformation that stabilizes the substrate bound to the P-site. In the core particles this change of conformation probably
Peptidyl Trdnsferase of Rat-Liver Ribosome Cores
can not take place in spite of the binding of the drug that inhibits the active center. The stimulation of peptidyl transferase by chemical modification using -SH [17] or -NH2 group reagents [lo] is a phenomenon typical of higher cell ribosomes. This effect may either be due to a direct action on the peptidyl transferase center or indirectly induced through a change of conformation in the ribosome structure. The importance of the conformation of the 60-S subunits for peptidyl transferase has been discussed elsewhere in relation to the effect of the 40-S subunit in this activity [lo]. In the case of the -SH group reagent N-ethylmaleimide, however, an effect through the modification of the interaction of the two subunits must be excluded since the stimulation is also detected in isolated 60-S subunits (Fig. 5). On the other hand, the fact that P37-cores totally or partially deficient in some 14 proteins but having a highly peptidyl transferase are insensitive to N-ethylmaleimide, leads us to believe that the stimulatory effect caused by the -SH group reagents acts through a general change of conformation rather than by direct action at the level of the peptidyl transferase center. The interaction of peptidyl transferase with several inhibitors is also affected in the P37-cores. Out of nine drug tested, sparsomycin, narciclasine and amicetin are more effective inhibitors of the enzyme in core particles than in control ribosomes ; trichodermin is less effective in the cores and the remainder; actinobolin, anisomycin, griseoviridin, gougerotin and blasticidin S, are equally effective in both types of particles. In principle, the three groups of drug must have different binding sites in the ribosome, but these sites need not necessarily be the same for drugs in the same group. Thus gougerotin and blasticidin S bind to mutually exclusive sites on the ribosome [23,24] and these sites differ from the binding site of anisomycin [24-261, although the three antibiotics have in common their similar pattern of inhibition on the ‘fragment reaction’ carried out by ribosomes and P37-cores. On the other hand, anisomycin, trichodermin and narciclasine bind to mutually exclusive sites on the 60-S subunit [25-281 and the actions of the three compounds appear to be differentially affected in the derived P37-cores. This work was supported by Institutional Grants from Comisibn Administradora del Descuento Complementario (Instituto Nacional de Previsibn) and Direccibn General de Sanidad (Spain) and by Grants from Lilly Indiana ofspain, and Essex Laboratories of Spain. We are grateful to Miss Pilar Ochoa for expert technical assistance.
REFERENCES 1. Hamel, E., Koka, M. & Nakamoto, T. (1972) J . B i d . Cham. 247, 805-814. 2. Howard, G . A . & Gordon, J. (1974) FEBS Lett. 48, 271-274. 3 . Ballesta, J. P. G. & Vazquez, D. (1974) FEBS Lett. 48, 266270.
Peptidyl Transferase of Rat-Liver Ribosome Cores 4. Highland, J. H. & Howard, G. A. (1975) J . B i d . Chem. 250, 831 - 834. 5. Bernabeu, C., Vizquez, D. & Ballesta, J. P. G. (1976) Eur. J . Biochem. 69,233 - 241, 6. Stoffler, G., Wool, I . G., Lin, A. & Rak, K. H. (1974) Proc. Nail Acud. Sci. U.S.A. 71, 4723-4726. 7. Reyes, R., Vizquez, D. & Ballesta, J. P. G. (1976) Biochim.
Biophys. Aria, in press. 8. Blobel, G. & Sabatini, D. (1971) Proc. Nail Acad. Sci. U.S.A. 68, 390 - 394. 9. Lowry, 0. H., Rossenbrough, N. J., Farr, A. L. & Randall, R. J. (1951) J . Bid. Chem. 193, 265-275. 10. Reyes, R., Vazquez, D. & Ballesta, J . P. G. (1976) Eur. J . Biochem. 67, 267 - 274. 11. Monro, R. E. (1971) Methods Enzymol. 20, 472-481. 12. Sherton, C. C. & Wool, I. G. (1972) J . B i d . Chem. 247, 44604467. 13. Martini, 0. H. W. & Could, H. J. (1971) J . Mol. Biol. 62, 403 - 405. 14. Howard, G. A. & Traut, R. R. (1974) Methods Enzymol. 30, 526- 539. 15. Loening, U. E. (1967) Biochem. J . 102, 251 -257. 16. Collatz, E., Lin, A,, Stoffler, G., Tsurugi, K. & Wool, 1. G. (1976) J . Biol. Chem. 251, 1808- 1816.
31 17. Carrasco, L. & Vazquez, G. (1975) Eur. J . Biochem. 50, 317323. 18. Vizquez, D. (1974) FEBS Lett. 40, S63-SX4. 19. Richter, D. & Moller, W. (1974) in Lipman Symposium: Energy
20. 21. 22. 23. 24.
Biosynthesis and Regulation in Molecular Biology (Richter, D., ed.) pp. 524-533, Walter de Gruyter Verlag, West Ber I in. Moller, W., Slobin, L. I., Amons, R. & Richter, D. (1975) Proc. Nail Acad. Sci. U.S.A. 72, 4744-4748. Osterberg, R., Sjoberg, B., Liljas, A. & Petterson, 1. (1977) FEBS Lett. in press. Fernandez-Mufioz, R. R. & VLzquez, D. (1973) J . Airtihior. (Tokyo) Ser. A , 26, 107- 108. Kinoshita, T., Tanaka, N. & Umezawa, H. (1970) J . Antihiof. (Tokyo) Ser. A , 23, 147-150. Barbacid, M . & Vizquez, D. (1974) Eur. J . Biochem. 44,
445 -453. 25. Barbacid, M. & Vazquez, D. (1974) J . Mol. Biol. 84, 603-623. 26. Jimknez, A. & Vazquez, D. (1975) Eur. J . Biochern. 54, 483492. 27. Barbacid, M. & Vazquez, D. (1974) Eur. J . Biochern. 44,437444. 28. Carrasco, L., Fresno, M. & Vazquez, D. (1975) FEBS Leit. 52, 236- 239.
R. Reyes, D. Vazquez, and J. P. G. Ballesta, Instituto de Bioquimica de'Macromoleculas, Centro de Biologia Molecular, Centro de Investigaciones Biologicas, C.S.I.C. and U.A.M., Velizquez 144. Madrid-6, Spain
Peptidyl Transferase Center of Rat-Liver Ribosome Cores Ramon REYES, David VAZQUEZ, and Juan P. G. BALLESTA Instituto de Bioquimica de Macromolbculas, Centro de Biologia Molecular, Consejo Superior de Investigaciones Cientificas and Universidad Autonoma de Madrid (Received July 13/November 2, 1976)
Protein-deficient particles have been obtained by treating rat liver 80-S ribosomes or their 60-S subunits with 1 M NH4Cl in the presence of 50 % ethanol at 0 "C (Po-cores) and 37 "C (P37-cores). The Po-cores from 8 0 4 ribosomes are totally inactive in polyphenylalanine synthesis but fully active in the 'fragment assay' to test peptidyl transferase activity. The polymerizing activity of the cores is restored up to 40-50% of control activity by incubation in the presence of the split proteins. Three proteins are totally lost in the treatment, namely proteins L12, L40/41 and S25. A series of up to nine different spots in the region of the L40/41 proteins are detected when the split fraction is analyzed by two-dimensional electrophoresis. This series of spots is, however, reduced to only two proteins when the second dimension is carried out in the presence of sodium dodecylsulphate. 80-S ribosome-derived P37-cores are about 80 % active in the 'fragment reaction' while 60-Ssubunit-derived particles are inactive in this assay. The inhibitory effect of a number of antibiotics is differentially affected by the treatment suggesting different localization of their binding sites. A comparative study of the proteins released by treatment in the two types of particles suggests the involvement in the peptidyl transferase center of the ribosome of one or more of the following proteins: L21, L24, L27, L28-and L36. The treatment of bacterial ribosomes with 1 M N&Cl in the presence of 50% ethanol specifically removes a few proteins. Thus only proteins L7 and L12 are released when the treatment is performed at 0 "C, yielding particles inactive in elongation-factordependent functions [l 1. In addition higher temperatures (37 "C) almost totally remove proteins L10, L11 and partially proteins L1, L5, L8/9 and L25 [2- 51. In the case of mammalian organisms the presence of proteins of similar characteristics to L7 and L12 has been detected in the ribosomes of rat liver by using immunological techniques [6]. However, no attempt has been reported for preparing protein-deficient particles derived from mammalian ribosomes using the ammonium chloride/ethanol method. We have prepared core particles from rat liver ribosomes using the latter technique at 0 "C and 37 "C and their functional and structural characteristics are reported here. Abbreviations. Po-proteins, split proteins released at 0 "C; P v proteins, split proteins released at 37 "C.
MATERIALS AND METHODS Ribosomes and Ribosomal Subunits
Ribosomes from Wistar male rats (100- 120 g) were prepared as previously described [7]. Ribosomal subunits were separated by centrifugation on convex exponential sucrose gradients using 8 0 3 ribosomes pretreated with puromycin [8]. All ribosomal preparations were kept in buffer made of 25 mM KCl, 5 mM MgC12, 50 mM Tris-HC1 pH 7.4 and 7 mM 2-mercaptoethanol in liquid nitrogen.
Preparation of Core Particles by Treatment with NH4C1 and Ethanol
A procedure similar to the one described by Hamel et al. [l] was followed. Either 80-S ribosomes or 60-S subunits at 4 mg/ml final concentration in 1 M NH4C1,lO mM imidazole-HC1pH 7.4,20 mM MgC12 and 1 mM 2-mercaptoethanol kept at 0°C or 37°C with continuous shaking were supplemented with
26
0.5 vol. of ethanol. After 10 min another 0.5 vol. of ethanol was added and the solution shaken for a further 5 min. The particles were then centrifuged at 27000 x g for 10 min. The supernatant was kept for recovering the split proteins and the pellet was treated in the same way for a second time. The particles were finally resuspended in 20 mM Tris-HC1 pH 7.4,11 mM KC1, 1 mM MgC12, 7 mM 2-mercaptoethanol, dialyzed against the same buffer for 2 h at 0 "C and stored in liquid nitrogen. Split Protein Fractions
The proteins were precipitated from the supernatants by addition of 2.5 vol. of acetone at - 20 "C, centrifuged at 15000 rev./min for 20 min and the pellets resuspended in 10 mM imidazole-HC1 pH 7.4, 2 mM 2-mercaptoethanol, dialyzed for 2 h against the same buffer and stored in liquid nitrogen. The concentration of proteins in the preparations was estimated by the procedure of Lowry et al. [9] using bovine serum albumin as a standard.
Peptidyl Transferase of Rat-Liver Ribosome Cores
RESULTS Activities of Po-Cores
Cores derived from rat liver 80-S ribosomes treated with 1 M NH4Cl and 50% ethanol at 0° C (Po-cores) are inactive in a phenylalanine-polymerizing system but can be reactivated upon addition of the split protein fraction (Po-proteins) and particles with up to 50% of the control activity obtained (Fig. 1). However the particles are quickly inactivated upon storage even in liquid nitrogen. However, the peptidyl transferase activity of the treated ribosomes or 60-S subunits is not diminished by the treatment. In fact some stimulation that varies from preparation to preparation has always been observed, in the particles derived from 80-S ribosomes (Table 1).
Activity Assays
All the activity tests used have been previously described in detail [7,10]. [3H]Phe-tRNA prepared from commercial yeast tRNA from General Biochemical and [3H]phenylalanine (7.8 Ci/mmol) from the Radiochemical Center (Amersham) was used for the polyphenylalanine synthesis assay. The 'fragment reaction' and the sparsomycin-dependent binding of (U)C-AC-C-A-[3H]Leu-Acwas carried out following standard methods [I 11 using (U)C-A-C-C-A-[3H]Leu-Ac prepared from [3H]Leu-tRNA (53 Ci/mmol) treated with ribonuclease T1 [ l l ] . Two-Dimensional Polyacrylamide Gel Electrophoresis
Standard two-dimensional electrophoresis of proteins was carried out in the conditions of Sherton and Wool [12]. We have also used the nomenclature of proteins adopted by these authors since the pattern of proteins obtained in our gel plates essentially coincides with that reported by them with the small variations described elsewhere [7]. When dodecylsulphate was used in the second dimension of the electrophoresis system the procedure of Martini and Gould [13] was followed. However, the conditions for the electrophoresis were modified by running the gels at 25 V for 14h at room temperature since a small-size apparatus was used [14]. rRNA electrophoresis was carried out in the conditions described by Loening [15] in tubes of 4 mm diameter and 10 cm length, run at constant power (1 Wjtube) until the bromophenol blue marker reached the bottom of the tubes.
PO-proteins (kg)
Fig. 1. Polyphenylalanine synthesis by Po-cores derived from 80-S rihosomes supplemented with increasing amounts of Po-protein. Core particles in 50p1 of 6 0 m M KCI, 1 5 m M MgC12 and 2 0 m M Tris-HCI buffer, p H 7.4 were incubated in the presence of these split proteins at 37 "C during 30 min. The samples were then placed o n ice and completed with the components required for the polyphenylalanine synthesis assay. The activity test was carried out at 37 "C during 20 min
Table 1. Peptidyl transferase activity of Po-cores The reaction mixture in the standard ionic conditions (Methods) contained 1 mg/ml of ribosomes and 8 nM of the (U)C-A-C-C-A[3H]Leu-Ac fragment. The values shown in the table have been calculated after subtraction of 330 counts/min of background obtained in the absence of particles. This background represent between 5 and 10 of the radioactivity in the samples Particle
Core particle derived from
80-S ribosomes
60-S subunits
counts min-' (pmol ribosomes)-' _ _ 145 255 184 252 -
Control Po-cores
R. Reyes, D. Vazquez, and J. P. G. Ballesta
21
Fig. 2. Two-dimensionul gel electrophoresis of Po-proteinsfrom 60-S subunits
Fig. 3 A . Acidic proteins in the Po-proteinsfrom 80-S ribosomes. The electrophoresis was run half of the usual time in the first dimension. In (B) dodecylsulphate was present in the second dimension
Proteins Released from Po-Cores The analysis of the Po-proteins obtained from 60-S subunits by two-dimensional gel electrophoresis shows the presence of protein L12 and a series of up to nine spots in the region of proteins L40/41 (Fig.2). In addition protein S25 was also present in Po-proteins from 80-S ribosomes (not shown). When the second dimension of the gel electrophoresis was run in the presence of dodecylsulphate two spots were detected in the region corresponding to the L40/41 proteins (Fig. 3). Four acidic proteins named A1 to A4 moving to a position similar to that described here for L40/41 also giving two spots in dodecylsulphate have recently been reported by Wooletal. [16]. However these
authors do not identify them with L40/41, which according to them have a less acidic character and move behind the A1 -A4 series. In our case the release by 1 M ammonium chloride and ethanol, indicates that these proteins might be similar to L7/12 of E. coli ribosome and if so, equivalent to those originally named L40/41 [ 6 ] . The proteins present in the split protein fraction are totally absent from the core particles indicating a high specificity of the treatment in their release. Activities of Ps,-Cores
Cores derived from both 80-S ribosomes and 60-S subunits treated with 1 M NH4Cl and 50% ethanol
Peptidyl Transferase of Rat-Liver Ribosome Cores
28
0’ 0
I
1
I
I 2 3 103.Fragment /ribosomes (molImol)
I
I
4
t ’
0.01
I
I
.o
0.1 1 N - Ethylmaleirnide (mM)
I
10
Fig. 4. Peptidyl transferase activity of P3-r-coresfrom80-S ribosome (a) and 60-S subunits (A). The activity is represented as a percentage of the untreated control particles as a function of the ratio of (U)C-A-C-C-A-[3H]Leu-Ac fragment to particles in the reaction mixture
Fig. 5. Effect of N-ethylmaleimide on the peptidyl transferase of so-S ribosomes ( O ) , 60-S subunits (0) and P3,-cores derivedfrom 80-S ribosomes ( 0 ) .The ‘fragment reaction’ was carried out as in Table 1
Table 2. Binding of ( U)C-A-C-C-A-[3H]Leu-Ac to P37-cores particles in the presence of sparsomycin Assay carried out by filtration in the ionic conditions of the ‘fragment reaction’ in the presence of 10pM sparsomycin and 20 or 40 pmol ribosomes
Stimulation of Peptidyl Transferase Center by N-Ethylmaleimide
Particle
Fragment bound with ribosomes 20 pmol 40 pmol counts/min
80-S ribosomes P37-cores
3301 162
5141 1237
(P37-cores) are unable to polymerize polyphenylalanine as is the case with Po-cores but unlike these particles they do not reconstitute their polymerizing activity by addition of the A7-protein fraction in any of a variety of ionic and temperature conditions that we have tested. Inactivation of the particles is not due to degradation of rRNA during the treatment since the RNA extracted from the cores presented the same band pattern in polyacrylamide gel electrophoresis as controls (not shown). When tested in the ‘fragment reaction’ for peptidyl transferase the P37-cores derived from 80-5 ribosomes displayed up to 80 % control activity while the particles derived from 60-S subunits were inactive (Fig. 4). Surprisingly the activity of the 80-S-ribosome derived P37-cores in binding the (U)C-A-C-C-A-[3H]Leu-Ac fragment in the presence of sparsomycin is drastically inhibited (Table 2).
- SH group reagents strongly stimulate the peptidyl transferase activity of mammalian ribosomes [17]. Since the effect caused by these reagents seems to be at the level of the peptidyl transferase center of the ribosome, we were interested in studying the behaviour of the protein-deficient particles treated with N-ethylmaleimide. Fig. 5 clearly shows that N-ethylmaleimide indeed stimulates the activity of 80-S ribosomes and 60-S subunits in the ‘fragment reaction’, but has a slightly inhibitory effect on the activity of the P37cores
Eflect of Peptidyl Transferase Inhibitors on P37-Core Activity
A number of peptidyl transferase inhibitors are known that directly interact with the ribosome active center [18]. Their binding sites are closely related and in some cases it is difficult to discern independent sites by binding studies. The study of the inhibitory effect of these compounds in the peptidyl transferase of the P37-cores may be of great value in understanding the structure of the antibiotic binding sites and the peptidyl transferase center as well. Table 3 summarizes the results obtained with nine different compounds used that according to their inhibitory effect fall in three groups : actinobolin, anisomycin, griseoviridin, blasticidin S and gougerotin inhibited the activity of the P37-cores to the same extent as the control 80-S ribosomes. Trichodermin inhibited the core particles less strongly, while sparsomycin, narciclasine and
R. Reyes, D. Vazquez, and J. P. G. Ballesta
29
Table 3. Effect of peptidyl transferase inhibitors on Ps7-core activity Fragment assay was carried out as in Table 1 in the presence of the indicated drug concentrations and measured as a percentage of the activity of the controls without drug Antibiotic
Concentration Activity in 'fragment reaction' of 80-S ribosomes
P37-cores
1000
32
34
Anisomycin
100 10
8 32
6 32
Trichodermin
100 10
4 11
18 43
100 1
13 93
7 58
21 52
10 30
Narciclasine
0.2 0.1
+
Protein
+,
80-S ribosome
60-S subunit -
Actinobolin
Sparsomycin
Table 4. Proteins released from ribosomes and 6 0 4 subunits by the treatment at 37°C For P~7-proteins,+ or - represents proteins present or absent in the two-dimensional electrophoresis of split fractions. For P37cores, represents protein present in similar amounts to those in control particles; protein in smaller amounts; tr, protein in trace amounts; - , protein absent
Amicetin
1000
80
50
Griseoviridin
1000 100
8 34
10 38
Gougerotin
1000 100
7 43
3 33
Blasticidin S
100 10
10 49
9 49
amicetin were approximately 50 % more inhibitory in the core particles. Proteins Released in the Treatment at 37°C Table4 shows the proteins found in the split fraction obtained by the treatment of 80-S ribosomes and 60-S subunits as well as the quantities of these proteins in the treated particles. Contrary to what occurs at O'C, the release of some proteins at 37°C is only partial. Proteins L6, L10, L12, L14, L24, L28, L35, L36, L38, L39, L40/41, XI and Xz are present in the Pj7protein fraction from 80-S ribosomes. In addition L21 and L27 are also found in the Pj7-protein fraction from 60-S subunits. Proteins L6, L12, L14, L35, L38, L39, L40/41 are totally absent or present only in trace amounts in the core particles from 80-S ribosomes. The same proteins plus L24, L28 and L36 are missing in core particles from the 60-S subunits. DISCUSSION Treatment of rat liver ribosomes with 1 M ammonium chloride in the presence of 50% ethanol at 0 "C releases the acidic proteins equivalent to proteins L7 and L12 split off the ribosomes of E. coli by a
L6 L10 L12 L14 L21 L24 L21 L28 L3 5 L36 L38 L39 L40 L4 1
x1 x2
P37-protein Pw-cores
P37-protein
P37-cores
+ + + + + + + + + + + + + +
+ + + + + + + + + + + + + + + +
+
f -
-
f
*
similar procedure [l]. Findings were similar with yeast [19] and Artemia salina [20] ribosomes. The functional equivalence of these proteins has been proven by cross-reactivation and immunochemical methods [6,19,20] and in the case of rat liver ribosomes were named L40 and L41 [6]. Wool and coworkers [16] also identified up to four different proteins in a region of the gel plates similar to that in which we found a series of up to nine spots. However, they do not consider these proteins to be the original L40 and L41 which according to their interpretation of the gels move behind in the first dimension. Our results seem to indicate that the proteins equivalent to the bacterial L7 and L12 might be in fact the ones that move to the most acidic position. The proteins that Wool et al. previously named L40/41 and now name Axl and Ax2 might correspond to our protein XI [7] that occasionally appears as a double band. The significance of the presence of many spots in the L40 and L41 region is at present obscure. It might be due to different forms of the two molecules detected in dodecylsulphate, modified in some way (possibly phosphorylation or acetylation) and in this sense it is interesting to note that Warner et al. [21] have recently reported the phosphorylation in vivo of the equivalent proteins in yeast ribosomes. Nevertheless due to the probably high mutual affinity of these proteins, already reported for proteins L7 and L12 of E. coli [21], we cannot discard the possibility that they represent aggregates with different proportions of the two proteins detected in dodecylsulphate, a
30
process favoured by overloading of the gels in order to make the proteins visible. Besides L40 and L41 two more proteins L12 and S25 are totally removed from the ribosome by the treatment at 0°C. These proteins are able to reincorporate to the cores since ribosomes active in a polymerization system are reconstituted (Fig. 1). In fact these proteins are present in the reconstituted particles when analyzed by two-dimensional gel electrophoresis (data not shown). At present we do not know the functional or structural role of these proteins but the position in gel plates and the easy removal of protein L12 makes it tempting to suggest its equivalence to protein L10 of E. coli. In any case it is clear that its release does not affect the peptidyl transferase center of the ribosome. The results obtained with the P37-cores are interesting in relation to the differences detected in the peptidyl transferase activity of the particles derived from 80-S ribosomes and 60-S subunits. These dissimilarities can be related to the different protein removed by the treatment in each case. Thus for instance proteins L21 and L27 while not affected in 80-S ribosomes are partially released in 60-S subunits and proteins L24, L28 and L36 which are totally removed from subunits are only partially missing in the 80-S-ribosome-derived cores. One or more of these proteins could be implicated in the loss of activity of the peptidyl transferase center in 60-S subunits. It is interesting to note that protein L21 among others, has also been pointed out as possibly implicated in the peptidyl transferase of rat liver ribosomes by the use of different experimental approaches, such as treatment with KCL [7] and chemical modification with 2-methoxy-5-nitrotropone [lo]. This protein among others is also labeled by a chemically reactive analoge of peptidyl-tRNA that reacts with ribosomal components close to the peptidyl transferase (Czernilafsky, Collatz, Kuechler, Gressner, personal communication). In spite of the activity that P37-cores from 80-S ribosomes display in the ‘fragment reaction’ (70 - 80 % of control) and its high sensitivity to sparsomycin inhibition, their peptidyl transferase center seems to have low activity in the sparsomycin-dependent binding of (U)C-A-C-C-A-[3H]Leu-Ac. The apparently contradictory results could be a consequence of the different conditions used in the two tests. However they might also imply that the stabilization of the (U)C-A-C-C-A-[3H]Leu-Ac fragment binding to the ribosome is an independent effect of the inhibition caused by sparsomycin. The drug probably inhibits peptidyl transferase by binding to the A-site [22] inducing at the same time a change of conformation that stabilizes the substrate bound to the P-site. In the core particles this change of conformation probably
Peptidyl Trdnsferase of Rat-Liver Ribosome Cores
can not take place in spite of the binding of the drug that inhibits the active center. The stimulation of peptidyl transferase by chemical modification using -SH [17] or -NH2 group reagents [lo] is a phenomenon typical of higher cell ribosomes. This effect may either be due to a direct action on the peptidyl transferase center or indirectly induced through a change of conformation in the ribosome structure. The importance of the conformation of the 60-S subunits for peptidyl transferase has been discussed elsewhere in relation to the effect of the 40-S subunit in this activity [lo]. In the case of the -SH group reagent N-ethylmaleimide, however, an effect through the modification of the interaction of the two subunits must be excluded since the stimulation is also detected in isolated 60-S subunits (Fig. 5). On the other hand, the fact that P37-cores totally or partially deficient in some 14 proteins but having a highly peptidyl transferase are insensitive to N-ethylmaleimide, leads us to believe that the stimulatory effect caused by the -SH group reagents acts through a general change of conformation rather than by direct action at the level of the peptidyl transferase center. The interaction of peptidyl transferase with several inhibitors is also affected in the P37-cores. Out of nine drug tested, sparsomycin, narciclasine and amicetin are more effective inhibitors of the enzyme in core particles than in control ribosomes ; trichodermin is less effective in the cores and the remainder; actinobolin, anisomycin, griseoviridin, gougerotin and blasticidin S, are equally effective in both types of particles. In principle, the three groups of drug must have different binding sites in the ribosome, but these sites need not necessarily be the same for drugs in the same group. Thus gougerotin and blasticidin S bind to mutually exclusive sites on the ribosome [23,24] and these sites differ from the binding site of anisomycin [24-261, although the three antibiotics have in common their similar pattern of inhibition on the ‘fragment reaction’ carried out by ribosomes and P37-cores. On the other hand, anisomycin, trichodermin and narciclasine bind to mutually exclusive sites on the 60-S subunit [25-281 and the actions of the three compounds appear to be differentially affected in the derived P37-cores. This work was supported by Institutional Grants from Comisibn Administradora del Descuento Complementario (Instituto Nacional de Previsibn) and Direccibn General de Sanidad (Spain) and by Grants from Lilly Indiana ofspain, and Essex Laboratories of Spain. We are grateful to Miss Pilar Ochoa for expert technical assistance.
REFERENCES 1. Hamel, E., Koka, M. & Nakamoto, T. (1972) J . B i d . Cham. 247, 805-814. 2. Howard, G . A . & Gordon, J. (1974) FEBS Lett. 48, 271-274. 3 . Ballesta, J. P. G. & Vazquez, D. (1974) FEBS Lett. 48, 266270.
Peptidyl Transferase of Rat-Liver Ribosome Cores 4. Highland, J. H. & Howard, G. A. (1975) J . B i d . Chem. 250, 831 - 834. 5. Bernabeu, C., Vizquez, D. & Ballesta, J. P. G. (1976) Eur. J . Biochem. 69,233 - 241, 6. Stoffler, G., Wool, I . G., Lin, A. & Rak, K. H. (1974) Proc. Nail Acud. Sci. U.S.A. 71, 4723-4726. 7. Reyes, R., Vizquez, D. & Ballesta, J. P. G. (1976) Biochim.
Biophys. Aria, in press. 8. Blobel, G. & Sabatini, D. (1971) Proc. Nail Acad. Sci. U.S.A. 68, 390 - 394. 9. Lowry, 0. H., Rossenbrough, N. J., Farr, A. L. & Randall, R. J. (1951) J . Bid. Chem. 193, 265-275. 10. Reyes, R., Vazquez, D. & Ballesta, J . P. G. (1976) Eur. J . Biochem. 67, 267 - 274. 11. Monro, R. E. (1971) Methods Enzymol. 20, 472-481. 12. Sherton, C. C. & Wool, I. G. (1972) J . B i d . Chem. 247, 44604467. 13. Martini, 0. H. W. & Could, H. J. (1971) J . Mol. Biol. 62, 403 - 405. 14. Howard, G. A. & Traut, R. R. (1974) Methods Enzymol. 30, 526- 539. 15. Loening, U. E. (1967) Biochem. J . 102, 251 -257. 16. Collatz, E., Lin, A,, Stoffler, G., Tsurugi, K. & Wool, 1. G. (1976) J . Biol. Chem. 251, 1808- 1816.
31 17. Carrasco, L. & Vazquez, G. (1975) Eur. J . Biochem. 50, 317323. 18. Vizquez, D. (1974) FEBS Lett. 40, S63-SX4. 19. Richter, D. & Moller, W. (1974) in Lipman Symposium: Energy
20. 21. 22. 23. 24.
Biosynthesis and Regulation in Molecular Biology (Richter, D., ed.) pp. 524-533, Walter de Gruyter Verlag, West Ber I in. Moller, W., Slobin, L. I., Amons, R. & Richter, D. (1975) Proc. Nail Acad. Sci. U.S.A. 72, 4744-4748. Osterberg, R., Sjoberg, B., Liljas, A. & Petterson, 1. (1977) FEBS Lett. in press. Fernandez-Mufioz, R. R. & VLzquez, D. (1973) J . Airtihior. (Tokyo) Ser. A , 26, 107- 108. Kinoshita, T., Tanaka, N. & Umezawa, H. (1970) J . Antihiof. (Tokyo) Ser. A , 23, 147-150. Barbacid, M . & Vizquez, D. (1974) Eur. J . Biochem. 44,
445 -453. 25. Barbacid, M. & Vazquez, D. (1974) J . Mol. Biol. 84, 603-623. 26. Jimknez, A. & Vazquez, D. (1975) Eur. J . Biochern. 54, 483492. 27. Barbacid, M. & Vazquez, D. (1974) Eur. J . Biochern. 44,437444. 28. Carrasco, L., Fresno, M. & Vazquez, D. (1975) FEBS Leit. 52, 236- 239.
R. Reyes, D. Vazquez, and J. P. G. Ballesta, Instituto de Bioquimica de'Macromoleculas, Centro de Biologia Molecular, Centro de Investigaciones Biologicas, C.S.I.C. and U.A.M., Velizquez 144. Madrid-6, Spain
