Volume 6 Number 7 1979

Nucleic Acids Research

Studies on the RNA and protein binding sites of the E.coli ribosomal protein

L1O

Ingvar Pettersson

Department of Molecular Biology, University of Uppsala, Box 562, S-751 22 Uppsala, Sweden Received 21 February 1979 ABSTRACT We have used modification of specific amino acid residues in the E. coli ribosomal protein LlO as a tool to study its interactions with another ribosomal protein, L7/L12, as well as with ribosomal core particles and with 23S RNA. The ribosome and RNA binding capability of L10 was found to be inhibited by modification of one more more of its arginine residues. This treatment does not affect the ability of L10 to bind four molecules of L7/L12 in a L7/L12-LIO complex. Our results support the view that LIO's role in promoting the L7/L12ribosome association is due primarily to its ability to bind to both 23S RNA and L7/L12 simultaneously.

INTRODUCTION The E. coli ribosomal proteins L7/L12 and LIO are near neighbors in the 50S ribosomal subunit (1). The interaction between these two proteins and the ribosome has been elucidated through studies using core particles depleted of either L7/L12 or of both L7/L12 and LIO (2-5). These experiments demonstrated that LIO was required for the rebinding of L7/L12 to the ribosome. An L10-L7/ L12 interaction can also be detected when the two proteins have been removed from the ribosome by a simple ammonium chloride-ethanol extraction (6,7). The extracts contain a stable complex of four molecules of L7/L12 bound to one molecule of LIO (8,9). Complex formation can even be induced in vitro with mixtures containing purified L7/L12 and L10 (7,10,11). No other ribosomal com-

ponents are necessary for the in vitro association of these two proteins. The principal concern of the present experiments is to explore the relationship between the two observed binding functions of L10: the capacity to associate with multiple copies of L7/L12 in the absence of ribosomes and the capacity to stimulate L7/L12 binding to ribosomes. It was found that LIO may interact with L7/L12 and with 23S RNA at the same time. The parallel expression of the two functions can be uncoupled after chemical modification of LIO, with a reagent specific for arginine. Here we report an analysis of the ) Information Retrieval Limited 1 Falconberg Court London Wl V 5FG England

2637

Nucleic Acids Research functional effects of this treatment without attempting to give a full protein chemical description of the modified structure. Thus, modification of arginine residues in L10 abolishes its ability to associate with 23S RNA or with ribosomal core particles. The L10 so modified is still able to interact with L7/L12. Our data suggest that L10 is bifunctional, in the sense that its stimulates the ribosome binding of L7/L12 by virtue of its ability to interact simultaneously with both L7/L12 and 23S RNA.

MATERIALS AND METHODS

50S ribosomal subunits were prepared from E. coli MRE 600 as described by Hardy et al. (12). The ribosomes were extracted in an ammonium chloride-ethanol solution as previously described (2,13,14).This treatment yielded core particles and supernatants, which were used for preparing purified proteins (11). The proteins were labelled in vitro by reductive methylation with 3H-HCHO and 14C-HCHO (15). 23S RNA was prepared by 3-fold phenol extraction followed by Sephadex G-100 chromatography and ethanol precipitation. Reconstitution of core particles with L10 and the L7/L12-LlO complex was done using the conditions of Schrier et al. (3). Here, the 100 PI aliquots contained 100 pmoles of core particles and 25-500 pmoles of the test protein. The reconstituted ribosomes were separated from non-bound protein by chromatography over Sepharose 6B columns equilibrated with the reconstitution buffer: 100 mM Tris HC1 pH 7.4, 20 mM Mg (Ac)2, 100 mM NH4C1 and 14 mM S-mercaptoethanol. Fractions were collected and divided into two aliquots. One was used for determining the amount of ribosomes by measuring the absorbance at 260 nm. The other aliquot was used to estimate the amount of 3H- or 14C-label by liquid scintillation coun-

ting. The RNA-binding of LIO, L7/L12 and L7/L12-LlO complex was assayed using the conditions of Dijk et al. (16). Binding mixtures of 100-150 pl contained 70 pmoles 23S RNA and 30-400 pmoles test protein in 30 mM Tris-HCl pH 7.4, 20 mM MgCl2, 350 mM KCI and 6 mM a-mercaptoethanol. These mixtures were incubated 1 hr at 420 and then cooled on ice. The RNA-binding was analyzed :IS described for the ribosome reconstitution. The conditions used for modifying L7/L12, LIO, and the L7/L12-LlO complex with different reagents are suimmarized in Table 1. The reactivity of the residues under the conditions used for modification was studied using 3H-N-

ethylmaleimide

2638

degree of modification of cysteines and 125I probe for tyrosine modification. In both cases

to estimate the

incorporation was used as

a

Nucleic Acids Research Table 1. Summary of reaction conditions

Modified amino acid residue

Buffer

Temp.

Time

Ref.

370

30'

(17)

50 mM Borate pH 9.0 10 mM Mg(Ac)2 2,3 butanedione

Arg

50 n

KCl

6 mM ,-mercaptoethanol N-ethylmaleimide (NEM)

2-nitro-5-thio cyanobenzoic acid Tetranitromethane

Diethylpyro-

Cys

10 mM Phosphate pH 7.4 150 mM NaCl

370

30'

(18)

Cys

200 mM Tris-Ac pH 8.0

210

1 hr

(19)

Tyr

50 mM Tris-HCl pH 8.0 100 mM KCI

210

30'

(20)

His

100 mM Phosphate pH 6.0 100 mM KC1

210

1 hr

(21)

Met

20

carbonate

NaIO 4

mM

Phosphate pH 7.0

40

15 hrs (22)

100 mM NaCl

stoichiometric modification occurred, 70-100%. The conditions used for the diethylpyrocarbonate reaction gave incorporation of reagent in 50S total protein corresponding to the calculated histidine content in the sample. The reactivity of the histidine in L10 was not determined directly due to the large amounts of protein needed for the spectroscopic assay. The integrity of the L7/L12-LIO complex after modification was analysed by chromatography of reacted protein samples on Sephadex G-100 Superfine columns. Reconstitution of the L7/L12-LlO complex was done as described previously (11).

RESULTS Effects of arginine modification on the binding of LIO and the L7/L12-LlO complex to 50S core particles L10 and the L7/L12-LIO complex were reacted with 2,3-butanedione as described in Methods. When the reagent was present at a concentration of 1 mM, 2639

Nucleic Acids Research this corresponds to approximately a ten-fold excess over the protein. When the ability of the modified protein samples to bind to 50S core particles was tested, both LlO and the L7/Ll2-LlO complex were found to be inhibited in their binding, as shown in Fig. 1. The extent of inhibition was the same for both of them and dependent on the concentration of reagent used for the modification reaction. An exacting chemical analysis of the modified protein will be required to identify which of the arginines in LlO have been modified. Still more difficult will be the identification of those modified arginines which are responsible for the loss of the binding capacity. Nevertheless, as we show in the following, analysisof the functional effects of this modification yields a simple interpretation of the functions of LlO in the association of the complex to the 50S subunit. The reduced binding can be regarded as a consequence of a lower average affinity for the binding site on the ribosome.This may be compensated for by increasing the concentration of the components. Such an effect is seen when the core particle-protein ratio is changed in the reconstitution mixture. With unmodified protein the plateau level of binding has been reached with a twofold excess of protein over core particles. Using modified proteins we find the drastically inhibited binding depicted in Fig. 1. If instead the modified proteins are reconstituted in a mixture with a 2-3 fold excess of ribosomal core particles, the inhibition if less pronounced. Here, the effect of incubating the protein sample with 1 mM reagent is a 30% inhibition compared with 70% inhibition with a 2-fold excess of protein over core particles. Similarly, with 5 mM reagent 55% inhibition is obtained in excess core particles and 85% when the proteins are in 2-fold excess in the reconstitution mixtures. The decreased binding could be due to modification of either one or both of the components in the complex. There is one arginine residue in each of the 1. Binding of 2,3-butanedione treated LlO and L7/Li2-LlO complex u to 50S core particles. The proteins | . were modified with reagent of the indicated concentration, and subsequently Aa. 8. W reconstituted into 50S core particles at a 2:1 protein/ribosome molar ratio. 0 50 The reconstitution mixture was passed O. through a Sepharose 6B column. The ribosome peak was collected and the proz FS .tein/ribosome molar ratio determined. 0 L7/L2-LlO complex; 9 LlO. m . X 100L

1

'

'

'

'

'

'

'

'

REAGENT CONCENTRATION

2640

.

Fig.

Nucleic Acids Research L7/L12 molecules (23) and 13 in the L10 molecule (24,25). However, modification of L7/L12 using another arginine-specific reagent, cyclohexanedione, has been reported (26) and it does not influence the dimeric structure of L7/L12 or the ability to bind back to core particles. We have repeated and confirmed these results with 2,3-butanedione (data not shown). Thus we may conclude that the inhibition seen with the L7/L12-LlO complex is due to the modification of one or more of the arginine residues in L10. The decreased ribosome binding of the L7/L12-LIO complex might be connected with a decreased capacity of LIO to bind L7/L12. This possibility was tested by chromatography of treated and nontreated samples of the L7/L12-LlO complex on Sephadex columns. No change in elution volume or the appearance of new peaks was observed, as can be seen from the elution profile in Fig. 2. In order to verify that this method will detect dissociation the elution profile of periodate-oxidized complex is also presented in Fig. 2. Oxidation will disrupt the L7/L12-dimers (22) and lead to dissociation of the L7/L12-LlO complex. Therefore, the gel filtration experiments indicate that arginine modification of the complex does not lead to dissociation. The specificity of the effect of arginine modification was explored by reacting the L7/L12-LIO complex with reagents specific for other amino acid residues present only in LIO, namely, cysteine, histidine and tyrosine. Table 2 summarizes the effects found on the L10-L7/L12 interaction and on

(W

I

I

.

.

.

.

go

x ,, O 3

)t

01

()

Ve-unmodified

L7/L12-L10 compLex

14

U. 2

t)

Q)~a V.-unmodified

-L7/1L12 20

35 30 40 25 FRACTION NUMBER

45

Fig. 2. Elution profiles of 2,3-butanedione treated and periodate oxidized I.7/L12-LIO complex. Samples of 2,3-butanedione treated and periodate oxidized -H labelled L7/L12-LlO complex were chromatographed on a Sephadex G-100 Superfine column. The fractions were collected and the radioactivity determined. The column was calibrated using unmodified L7/L12-LIO complex and L7/L12 as standards. 0 oxidized complex; 0 2,3-butanedione treated complex. 2641

Nucleic Acids Research Table 2.

Effects of chemical modifications of the L7/L12-LlO complex

Reagent

Concentration Modified

Effects

amino acid residue

2,3-butanedione

100 pM-10 mM

N-ethylmaleimide 100 pM-1 mM 2-nitrothiocyano- 1 mM-10 mM benzoic acid Tetranitro500 pM-5 mM

Binding to core particles

L7/L12-LlO interaction

Complex intact

Cysteine Cysteine

Decreased to 12Z with 10 mM Not affected Not affected

Tyrosine

Not affected

Complex intact

Histidine

Not affected

Complex intact

Arginine

Complex intact Complex intact

methane

Diethylpyrocarbonate

100 iM-l mM

the ribosome binding when different residues were modified. These modification reactions were performed with varying reagent to protein molar ratios in the range 10-200. Control experiments described in Methods indicated that these residues were indeed reactive under the conditions used. In no case is the ribosome binding affected to the extent seen with the arginine-specific reagent, nor will these other reagents cause the disruption of the L7/L12-LlO complex.

Binding of unmodified LIO and L7/L12-LlO complex to 23S RNA Binding of the L7/L12-LIO complex to 23S RNA in vitro has been reported by Dijk et al.(16). We used this assay in order to determine if treatment with the arginine-specific reagent would interfere with the protein RNA association. Fig. 3 shows the binding curves for the L7/L12-LlO complex and L10, as well as for denatured-renatured L10 and L7/L12 to 23S RNA. Unlike Dijk et al. we found that LIO could bind to the 23S RNA by itself and not only as part of the L7/L12LIO complex. The different methods used for preparation of the proteins, for the isolation of the RNA and for assaying complex formation can probably explain the different results obtained here and by Dijk et al. The weak binding to 23S RNA displayed by L7/L12 probably reflects a general affinity for nucleic acids, since it will bind even to DNA (27). Thus, we regard as significant only the stoichiometric binding of L10 and of the L7/L12-LlO complex. The possibility that binding to the RNA will induce changes in LlO's ability to interact with L7/L12 was tested by using complexes reconstituted from C

2642

Nucleic Acids Research

3.0 m 0

w

/ /

' .

2.0

Fig. 3. Binding of L10, L7/L12 and L7/L12L10 complex to 23S R1NA in vitro. A fixed amount of 23S RNA was incubated with increasing amounts of protein. The mixture was passed through a Sepharose 6B column and the protein/RNA ratio in the RNA peak determined. Protein in the incubation mixtures:

complex;

0 L7/L12; 0 L10; * L7/L12-LlO 0 L10 denatured-renatured.

z

z

0 0 5010 7.0 12023 PROTEIN/RNA RATIO IN INCUBATION MIXTURE labelled L10 and 3H labelled L7/L12. The reconstituted complex was incubated with 23S RNA and the 3H/14 C ratio was measured for the added complex, for the protein recovered bound to 23S RNA and for the excess non bound protein. The results from two experiments are given in Table 3. The constant 3H/ 14C ratio in each of the two experiments indicated that the added 4:1 L7/L12-LlO complex was bound to the 23S RNA as a 4:1 complex. This result supports the conclusion that the interaction between 23S RNA and LIO does not change the L7/L12 binding capacity of LIO.

Binding of arginine modified LIO and L7/L12-LlO complex to 23S RNA More direct evidence about the involvement of the arginines comes from experiments where arginine modified proteins were used in the 23S RNA binding assays. The binding curves from such experiments are presented in Fig.4. As in the case when the binding to core particles was assayed, the inhibition

Table 3.

3H/ 14C

ratio before and after RNA binding

Reconstituted L7/L12-LlO complex Exp. 1 Exp. 2

.92 1.30

Complex recovered bound to 23S RNA .99 1.40

Excess, non bound complex 1.09 1.42

2643

Nucleic Acids Research %

Fig.4. Binding of 2,3-butanedione treated L10 and L7/L12-LlO complex to 23S RNA in

vitro. The proteins were modified with in0 \ creasing concentrations of reagent and then cc - \ \added at a 2-fold molar excess to 23S RNA. U) The mixture was assayed for protein-RNA complex formation as described for the 50 o unmodified protein. 0 L7/Ll2-LlO complex; S LIO. CD z m

z

z

5 10mm REAGENT CONCENTRATION

was found to be dependent on the concentration of the modifying reagent. However, here the inhibition curve was more shallow than that for the reduction of binding to ribosomal cores. The higher residual L10 binding after arginine modification compared with that of the L7/L12-LIO complex, could be caused by binding of aggregated L10 to the RNA or by multiple binding sites, some of which are inaccessible for the more bulky complex. Unspecific binding of LIO to the RNA under these conditions was found when denatured-renatured protein was used, as is shown in Fig. 3. The effect of the arginine modification on the binding capacity of the treated proteins in the 23S RNA binding assay imply that arginine residues in LIO are important for the protein-RNA interaction. As

demonstrated in the previous section, the protein-protein interaction in the

L7/L12-L10 complex is seemingly undisturbed by this modification.

DISCUSSION

Protein L7/L12 influences several steps in protein biosynthesis, in particular those requiring interactions between the ribosome and protein factors involved in initiation, elongation and termination (28). Previous studies have shown that another protein from the 50S subunit, LlO, carries out the important task of linking the L7/L12 molecules to the ribosome (3-5). The role of L10 does not seem to be to induce indirectly a suitable binding site for L7/L12. A direct interaction between the two proteins is strongly suggested by the identification of a stable L7/Ll2-LlO complex (6,7). This complex can be isolated from the ribosome or reformed from purified proteins in the absence of other ribosomal components (8-11). Experimental evidence for a direct 2644

Nucleic Acids Research interaction between 23S RNA and the L7/L12-LlO complex has been obtained (16). The experiments reported here using an arginine specific reagent show that the association of the L7/L12-LlO complex with the ribosome can be eliminated without seemingly influencing the stability of the complex itself. In addition, the data suggest that one or more arginine residues of L10 play an important role in binding both this protein alone and the whole L7L12-LlO complex to 23S RNA. It could also be shown that the interaction with L7/L12 is not required for the expression of L10's ribosome binding or 23S RNA binding capacity. These results lead us to suggest that the L10 molecule has at least two functionally defined binding sites, one for interaction with the protein L7/L12 and one for interaction with the 23S RNA in the 50S particle. The two sites work independently in the sense that the L10-RNA binding can be inhibited without seemingly influencing the protein-protein recognition in the L7/L12-LlO complex. Whether or not there exist two physically separated sites on the L10 molecule which correspond to the two binding functions cannot be deduced from the present data.

ACKNOWLEDGEMENTS I thank Professor C.G. Kurland and Dr. Anders Liljas for continuous support and interest as well as Dr. Roger Garrett for discussions concerning the

RNA-protein experiments. The technical assistance of Mrs.B.Wiklund, Mr.B.Karlsson and Mr.S.Eriksson is gratefully acknowledged. This work was supported by the Swedish Natural Science Research Council, the Swedish Cancer Society and the Magnus Bergvall Foundation.

REFERENCES 1. Expert-Bezanqon,A., Barritault,D., Millet,M. and Hayes,D.H. (1976) J. Mol.Biol. 108, 781-787 2. Hamel,E., Koka,M. and Nakamoto,T. (1972) J.Biol.Chem. 247, 805-814 3. Schrier,P.I., Maassen,J.A. and M811er,W. (1973) Biochem.Biophys.Res.Comm. 53, 90-98 4. Highland,J.H. and Howard,G.A. (1975) J.Biol.Chem. 250, 831-834 5. Stoffler,G., Hasenbank,R., Bodley,J.W. and Highland,J.H. (1974) J.Mol.Biol. 86, 171-174 6. Pettersson,I., Hardy,S.J.S. and Liljas,A. (1976) FEBS Lett. 64, 135-138 7. Gudkov,A.T., (1976) Doklady Akademii Nauk SSSR 230, 979-980 8. Osterberg,R., Sjoberg,B., Pettersson,I., Liljas,A. and Kurland, C.G. (1977) FEBS Lett. 73, 22-24 9. Gudkov,A.T., Tumanova,L.G., Venyaminov,S.Yu. and Khechinashivili,N.N.

(1978) FEBS Lett. 93, 215-218 2645

Nucleic Acids Research 10. Chu,F., Caldwell,P., Samuels,M., Weissbach,H. and Brot,N. (1977) BBRC 76, no 2, 593-601 11. Pettersson,I. and Liljas,A. (1979) FEBS Lett. 98, 139-144 12. Hardy,S.J.S., Kurland,C.G., Voynow,P. and Mora,G. (1969) Biochemistry 8, 289 7-2905 13. Sander,G,, Marsh,R.C., and Parmeggiani,A. (1972) Biochem.Biophys.Res.Comm. 47, 866-873 14. Glick,B.R. (1977) FEBS Lett. 73, 1-5 15. Rice,R.H, and Means,G.E. (1971) J.Biol.Chem. 246, 831-832 16. Dijk,J., Littlechild,J. and Garrett,R.A. (1977) FEBS Lett. 77, 295-300 17. Rohrbach,M.S. and Bodley,J.W. (1977) Biochemistry 16, 1360-1363 18. Riordan,J.F., Vallee,B.L. (1972) in Methods in Enzymology XXV (Hirs,Ch.W. and Timasheff,S.N. eds.) pp 449-456 19. Jacobson,G.R., Schaffer,M.H., Stark,G.R. and Vanaman,T.C. (1973) J.Biol. Chem. 248, 6583-6591 20. Riordan,J.F. and Vallee,B.L. (1972) in Methods in Enzymology XXV (Hirs, C.H.W. and Timasheff,S.N. eds.) pp 515-521 21. Muhlrad,A., Hegyi,G. and Horanyi,M. (1969) Biochem.Biophys.Acta 181, 184-190 22. Gudkov,A.T. (1977) Mol.Biol. 11, 1201-1205 23. Terhorst,C., Moller,W., Laursen,R. and Wittmann-Liebold,B. (1972) FEBS Lett. 28, 325-328 24. Dovgas,N.V., Vinokurov,L.M., Velmoga,I.S., Alakhov,Yu.B. and Ovchinnikov, Yu.A. (1976) FEBS Lett. 67, 58-61 25. Heiland,I., Brauer,D. and Wittmann-Liebold,B. (1976) Hoppe-Zeyler's Z. Physiol.Chem. 357, 1751-1770 26. Koteliansky,V.E., Domogatsky,S.P. and Gudkov,A.T. (1978) Eur.J.Biochem. 90, 319-323 27. Lathe,R. and Lecocq,J.P. (1977) Molec.gen.Genet. 154, 53-60 28. Moller,W. (1974) in Ribosomes (Nomura,M., Tissieres,A. and Lengyel,R., eds.) pp 711-731 Cold Spring Harbro Press

2646