- Email: [email protected]
Identification of neighbouring proteins by cross-linking of intact 70 S ribosomes from Escherichia coli
Biochimica et Biophysica Acta 869 (1986) 1-7 Elsevier
1
BBA 32403
Identification of neighbouring proteins by cross-linking of intact 70 S ribosomes from Escherichia coli Ute Scheibe and Rolf Wagner * M a x -Planck-lnstitut ft~r Molekulare Genetik, A bteilung Wittmann, lhnestrasse 63 - 73, 1000 Berlin 33 (Germany)
(Received July 18th, 1985)
Key words: Chemical cross-linking; 70 S ribosomes; Structural organization: ( E. coil )
70 S ribosomes from Escherichia coil have been reacted with the bifunctional reagent 1,4-phenyldiglyoxal under near physiological conditions. As a result of the cross-linking reaction a number of high-molecularweight protein fractions with altered electrophoretic mobility could be isolated. A new chemical procedure has been introduced to reverse the cross-links between proteins at least partially. The cleavage reaction did not affect the gel electrophoretic mobility of the proteins. Thus a direct identification of cross-linked proteins using one- or two-dimensional gels was made possible. Two protein trimers, $3-$4-$5 and LI-$4-$5, as well as five protein dimers, $3-$4, L 6 - L 7 / 1 2 , L I 0 - L 7 / 1 2 , $9-L19 and LI8-LI9 could be identified as close neighbours in the E. coil 70 S ribosome. The protein pairs $9-L19 and LI8-LI9 had previously not been identified as near neighbours using cross-linking studies.
Introduction Investigations of the topography of the bacterial ribosome have employed many different techniques including electron microscopy, neutron scattering, fluorescence energy transfer, X-ray crystallography, chemical modification or crosslinking (for reviews see Ref. 1-3). From these studies a rough three-dimensional model of the E. coli ribosome can be constructed. The localisation of a number of ribosomal components by the different methods agrees well, while the position of some others are still the subject of controversy. For the construction of a valid three-dimensional model of the ribosome it is important therefore to obtain firm data by comparison of the results of several independent studies. Since high-resolution methods like X-ray crystallography do not yet provide enough data for a
* To whom correspondence should be addressed.
detailed structural model [4] chemical cross-linking is still one of the most important and direct methods for the elucidation of the ribosomal topography. Possible artefacts of this method are greatly reduced if certain cross-linking results can be repeatedly found using different conditions and reagents. In a recent study we used the cross-linking reagent 1,4-phenyldiglyoxal for a systematic analysis of the RNA-protein neighbourhoods within and across the ribosomal subunits. The same reagent has been used by us and others in a number of studies for R N A - R N A cross-linking as well as for affinity labelling [5-9]. Until now we were unable to use the same reagent for the analysis of protein-protein cross-links. Such products were consistently formed under the reaction conditions but no method for the cleavage of the protein-protein cross-links was available. Recently we have developed a method to reverse the protein crosslinks. This enabled us to identify a number of protein neighbourhoods by direct gel electro-
0167-4838/86/$03.50 © 1986 Elsevier Science Publishers B.V. (Biomedical Division)
phoretic analysis. The cross-links within the 70 S ribosomes were introduced under similar conditions as described for the RNA-protein analysis [7]. Among the cross-links identified are protein trimers and dimers. Some of them are cross-linked within the ribosomal subunits, while others are found to consist of proteins from the small and the large subunit and therefore are considered to be interface components. The results are discussed with respect to the known topography of the 70 S ribosome. Materials and Methods
1,4-Phenyldiglyoxal was synthesized as described [5]. Acrylamide, N,N'-methylenebisacrylamide, sodium dodecyl sulfate (SDS) were from Bio-Rad, U.S.A. Urea (ultra pure) was from BRL Inc., U.S.A. 1,2-Phenylenediamine was from Merck, Darmstadt. The radioiodination system for the radioactive labeling of proteins was from New England Nuclear, U.S.A. Isolation of 70 S ribosomes. Ribosomes were prepared from E. coli MRE 600 cells as described in Ref. 7. Cross-linking of 70 S ribosomes. 70 S ribosomes (0.4 nmol/ml) were incubated with 0.35 mM phenyldiglyoxal at 37°C for 4 h. The reaction was stopped by the addition of 0.7 vol. of ethanol ( - 2 0 ° C ) and the ribosomes were precipitated for 2h.
a model compound. Lysozyme was cross-linked intermolecularly with phenyldiglyoxal, and the protein dimer was isolated by gel electrophoresis according to Ref. 12. The lysozyme dimers were treated under various conditions with 1,2-phenylenediamine. Incubation of cross-linked dimers with 50 mM 1,2-phenylenediamine at 37°C for 16 h was found to be optimal for the reversal of the cross-link bonds. Under these conditions at least 40% of the cross-linked dimers could be converted to single proteins. The cleaved proteins did not show altered electrophoretic mobilities in one- or two-dimensional gel systems. Radioactive labelling of marker proteins. A preparation of total ribosomal proteins not reacted with phenyldiglyoxal was radioactively labelled with 125I using the standard lactoperoxidase labelling procedure. The labelled proteins were used as markers at concentration at which clear autoradiograms could be obtained, but the 125I labelled protein concentration was too small to be stained with Coomassie blue. Gel electrophoretic separation of proteins. Gel electrophoretic separations of proteins were performed in one or two dimensions as described [13,14]. The proteins were stained with Coomassie brilliant blue. In cases where only a limited number of proteins had to be assigned on two-dimensional gels a small amount of 125I-labelled marker proteins were separated simultaneously. The gels were first stained and then autoradiographed after drying.
Extraction of ribosomal proteins. After the cross-linking reaction and precipitation of the ribosomes the total protein fraction was extracted with acetic acid according to Hardy et al. [10]. The proteins were dialyzed against decreasing acetic acid concentrations and finally lyophilized. The lyophilized proteins were dissolved in 6 M urea/10 mM dithioerythritol/5 mM EDTA and stored at - 2 0 ° C . Cleavage of the protein-protein cross-links. No cleavage of the protein-protein cross-links could be achieved under conditions where RNA-RNA cross-links produced with phenyldiglyoxal could be reversed completely. We therefore took advantage of the specific reaction of a-dicarbonyl derivatives with 1,2-phenylenediamine [11]. To optimize the cleavage conditions we used lysozyme as
Results
When the total protein fraction of phenyldiglyoxal-reacted 70 S ribosomes was analysed on one- or two-dimensional gel systems, cross-linking between individual proteins could be demonstrated by the appearance of new high-molecularweight protein products. In Fig. 1 the separation of total ribosomal proteins extracted before and after reaction with phenyldiglyoxal is shown on a one-dimensional SDS gel. Protein bands labeled I to IV in Fig. 1 indicate the formation of cross-linked proteins and can clearly be distinguished from the migration positions of the unreacted 70 S proteins. Some of the protein bands are reduced in their intensity compared to the unreacted 70 S
C1
x2
x3
--I
III IV
Fig. 1. One-dimensional separation of ribosomal proteins before and after cross-linking. C 1, C2: Total ribosomal proteins extracted from unreacted ribosomes; 30 and 6 /xg of protein, respectively. X ], X 2, X3: Total proteins extracted from crosslinked 70 S ribosomes. The total amount of 60, 30 and 6 # g proteins was separated, respectively. The ribosomal protein S1 can not be seen on the control tracks as a result of the acetic acid extraction procedure. Its presence on the ribosomes used for the cross-linking was established, however, by the LiCl-urea extraction procedure.
sample. However, under the cross-linking conditions many proteins are cross-linked simultaneously to the ribosomal RNAs [7]. The altered intensities of some of the protein bands in the cross-linked samples visible in Fig. 1 are therefore difficult to interpret with respect to protein-protein cross-linking. When the cross-linked 70 S ribosomal proteins are separated on two-dimensional gels (Fig. 2) the cross-linked products (labeled I to IV in Fig. 1) are resolved into a more complex pattern of individual spots (Fig. 2). Spots containing the cross-linked proteins were cut out and collected from 8-10 two-dimensional gels. The proteins were extracted electrophoretically in the presence of 0.1% SDS and lyophilized. The samples were dissolved in water and divided in two aliquots, one of which was treated with the cleavage reagent. Cleaved and uncleaved protein samples were analyzed on oneor two-dimensional gels. The proteins analyzed on two-dimensional gels were first freed from residual SDS according to [15] and then mixed with radioactive marker proteins prior to separation. The gels were stained and autoradiographed after drying with a gel dryer. Not all the products that appeared as new spots on the two-dimensional gel after cross-linking (Fig. 2) could be analyzed unambiguously. Often material extracted from single spots proved to be heterogeneous. In seven cases, however, a direct assignment of the proteins involved in cross-linking could be achieved. The results of these analyses are shown in Figs. 3-5. In Fig. 3A, B the analyses of cross-linked
TABLE I RELATIVE YIELDS A N D M O L E C U L A R W E I G H T C O M P A R I S O N OF T H E CROSS-LINKS I D E N T I F I E D Yields are given as relative values estimated from the staining intensities of the two-dimensional gels. + + + , strong; + + , medium; + , weak ( + ) very weak.
la Ib II Ilia llIb IVa IVb
Cross-links
Mapp. (kDa)
M~al. (kDa)
Yield
$3-$4-$5 L1-$4-$5 $3-$4 L7/12-L10 L7/12-L6 $9-L19 L18-L19
67 67 47 30 30 26 26
66.5 65.3 48.9 30.0 31.0 27.6 25.7
+ + + + + + + + (+ ) (+ ) + + +
1 st
:
O.
12
S~
SlO~
$12
S17
~k
27
material corresponding to spot I (Fig. 2) is shown before and after hydrolysis of the cross-link. Four protein spots and some remaining uncleaved material are apparent (Fig. 3B). The four proteins were assigned according to the autoradiogram of the 125I-labelled marker proteins as L1, $3, $4 and $5. The relative molar ratios of these four proteins was estimated from the staining intensities to be 1 : 1 : 2 : 2 . The apparent molecular mass of the cross-link fraction I (Fig. 1) can be calculated from the migration position on the SDS gel to be about 67 kDa, which can be explained only with protein trimers. The cross-link fraction I consists therefore most probably of two protein trimers with the common proteins $4 and $5. The trimers can consequently be assigned as: $3-$4-$5 and L1-$4-$5. The cross-link fraction II was very heterogeneous and consisted of many new protein spots as evidenced from Figs. 1 and 2. The spot II in Fig. 2 could, however, be analyzed unambiguously. In Fig. 3C and D the two-dimensional gel-electrophoretic analysis of the products extracted from
Fig. 2. Two-dimensional separation of proteins from 70 S ribosomes after the cross-linking reaction. New protein spots are encircled and numbered I to IV.
spot II (Fig. 2) is shown before (C) and after (D) cleavage of the cross-link. Two monomeric proteins could be identified in Fig. 3D and were assigned as $3 and $4. The apparent molecular mass of the corresponding cross-link pair (47 kDa) as determined by SDS gel electrophoresis of cross-link fraction II correlates well with the calculated molecular mass, 48.9 kDa. The cross-link fraction III (Fig. 1) was resolved in several spots on the two-dimensional gel (Fig. 2). Spots IIIa and I I I b gave clear analysis results after cleavage of the cross-link (Fig. 4). Spot Ilia consists of the proteins L 7 / 1 2 and L10. Again, the molecular mass of the cross-linked dimer determined from an SDS gel (30 kDa) is in perfect agreement with that calculated (30 kDa). Analysis of material extracted from spot IIIb showed that it was contaminated with products from spot IIIa. The protein L 7 / 1 2 can be found in both spots but with a higher relatiye stoichiometry in I I I b (Fig. 4). The protein L6 is only found in spot IlIb. Spot I I I b consists therefore of the cross-link dimer L6-L7/12 with the contamination
1 st
~_
TP70
III
f~ .-I
Q.
~i~i~i!~'i~i~!~/ ~!~i~i~!~i~!i~!~i~il!i!i~i
Fig. 4. One-dimensional analysis of the cross-link fractions Ilia and lllb. The cross-linked material was isolated from two-dimensional gels and separated on an SDS gel after cleavage of the cross-link. (a, b) Uncleaved protein from cross-link fraction II1.
L1
o
S3 Q
TP 30 50
IV
I S4 S~PL6 i
i
b
Fig. 3. Two-dimensional analysis of cross-linking products. (A) Purified cross-linking product I separated on a two-dimensional gel without cleavage of the cross-link. (B) Same material as in (A) after cleavage of the cross-link. (C) Purified cross-link product II separated without cleavage reaction. (D) Same material as in (C) after cleavage of the cross-link. (E) Separation pattern of ]251-labelled ribosomal marker proteins. (F) Separation of cross-link fraction IVb after cleavage of the cross-link. The small a m o u n t of contaminating L2 was also present in the uncleaved sample. In A, B, C, D and F the gels were stained with Coomassie brilliant blue, while in E an autoradiogram is shown.
J
L18~'-
Fig. 5. One-dimensional analysis of the cross-link fractions IVa and IVb. The material was separated after cleavage of the cross-link as indicated in the legend to Fig. 4.
of the dimer L7/12-L10 from spot IIIa. A similar situation was encountered during the analysis of the cross-link fraction IV. The analysis of material extracted froms pots IVa and IVb (Fig. 2) on a one-dimensional SDS gel after cleavage of the cross-link with 1,2-phenylenediamine is shown in Fig. 5. The two proteins $9 and L19 can be identified from material extracted from spot IVa. Spot IVb yielded the proteins L18 and L19 after the cleavage reaction. Contaminations of spot IVa were, however, frequently apparent in fraction IVb. The ribosomal protein $9 could be seen in addition and the relative stoichiometry of L19 was higher than that of L18. The results were confirmed by two-dimensional gel electrophoresis (see Fig. 3F). The presence of L19 in the cross-link fraction IV was additionally substantiated by immunoblotting using specific antibodies (data not shown). We conclude therefore that the cross-link fraction IV consists of two protein dimers, namely $9-L19 (IVa) and L18-L19 (IVb). The results of the gel electrophoretic analysis are summarized in Table I. Discussion
Among the many products formed under the cross-linking conditions only those fractions were analyzed that appeared reproducibly and in relatively high yields. Some of the cross-linking products, although of high yield (see Fig. 2) were, however, too complex in their protein composition to be analyzed unambiguously. The yields of some other cross-linked protein fractions such as I and II (Fig. 2) seem to be high enough to be isolated in amounts sufficient to allow the analysis not only of the proteins but also of the amino acids involved in cross-linking. Such analysis has been performed to identify the nucleotide and amino acid moieties of RNA-protein cross-links and should be the final goal of all cross-linking studies [16]. Under the cross-linking conditions used in this study the structure of the ribosomes was not altered to a noticeable extent as has been shown in a previous study [7]. From the comparison of the two-dimensional electrophoresis of non-cross-linked versus crosslinked proteins, it is apparent that no major mobil-
ity shifts occur for any of the ribosomal proteins. Furthermore, no change in mobility could be detected for the proteins treated with 1,2-phenylenediamine. Therefore the assignment of the cross-linking proteins by gel electrophoretic methods is valid. When proteins are highly modified with phenyldiglyoxal a small north-west shift on twodimensional gels can be observed. The correct assignment is not affected, however (see also Ref. 7). All the results presented in this study were obtained between two and five times.
Comparison with other topographical data The spatial neighbourhoods of the cross-linked proteins L1-$4-$5, $3-$4-$5, $3-$4, L6-L7/12 and L10-L7/12 identified in this study have been reported in similar or identical arrangement by other investigators using different cross-linking reagents [17-19]. For instance, the protein dimers $3-$4, $4-$5 and also $3-$5 could be identified as crosslinked. The large subunit protein L1 and the small subunit protein $4 known as interface components have also been cross-linked [20]. The two protein trimers L1-$4-$5 and $3-$4-$5 are therefore well substantiated by these studies and their proximity in the ribosome can be taken as firmly established by cross-linking studies. The protein dimers L6-L7/12 and L10-L7/12 have also been cross-linked with different reagents [18,19]. Furthermore, the complex L10-L7/12 can be extracted from unreacted ribosomes [21]. On the other hand, the two protein dimers $9-L19 and L18-L19 have to our knowledge not yet been reported as being cross-linked. The proximity of these proteins in the 70 S ribosome is, however, in perfect agreement with existing topographical data based on cross-linking studies. The two proteins L18 and El9 have both been shown to form dimers with the protein $13 and can therefore be considered as interface proteins and near neighbours [20]. The protein $9 is also an interface protein and has been cross-linked to L5 [22]. The distances of the proteins L5 and L18 can be calculated to be very short because they are known as direct binding proteins to the ribosomal 5 S RNA. Consequently the three proteins $9, L18 and L19 may all be close in the 70 S ribosome. Our crosslinking results confirm such a spatial relationship. There are some differences in the distances of
p r o t e i n s in situ b a s e d o n c r o s s - l i n k i n g results c o m p a r e d to i m m u n e e l e c t r o n m i c r o s c o p i c studies. T h e sites for p r o t e i n s L1 on the o n e h a n d a n d $4, $5 o n the o t h e r h a n d as l o c a l i z e d b y i m m u n e e l e c t r o n m i c r o s c o p y are r a t h e r r e m o t e to e a c h other. Similarly, the d i s t a n c e o f the sites d e t e r m i n e d for L18 a n d L19 s e e m to be t o o far to be b r i d g e d by a c r o s s - l i n k i n g r e a g e n t [23]. T h e p r o x i m i t y of L18 a n d $9 is s u p p o r t e d , h o w e v e r , b y n e u t r o n s c a t t e r i n g results a c c o r d i n g to w h i c h p r o t e i n $9 is l o c a t e d in the h e a d of the 30 S s u b u n i t [24]. T h i s l o c a t i o n is fairly close to the p o s i t i o n of L18 in the 70 S r i b o s o m e [23].
Acknowledgements W e are g r a t e f u l to P r o f e s s o r H . G . W i t t m a n n for his s u p p o r t . W e like to t h a n k Dr. H.J. S c h 6 n f e l d for m a n y h e l p f u l d i s c u s s i o n s a n d Dr. M. St6fflerM e i l i c k e for p r o v i d i n g us w i t h r i b o s o m a l p r o t e i n antisera.
References 1 Chambliss, G., Craven, G.R., Davies, J., Davis, K., Kahan, L. and Nomura, M., (eds.) (1980) Ribosomes: Structure, Function and Genetics University Park Press, Baltimore, MD 2 Liljas, A. (1982) Progr. Biophys. Mol. Biol. 40, 161-228 3 Winmann, H.G. (1983) Annu. Rev. Biochem. 52, 35-65 4 Yonath, A. (1984) Trends Biochem. Sci. 9, 227-230 5 Wagner, R. and Garrett, R.A. (1978) Nucleic Acids Res. 5, 4065-4075 6 Hancock, J. and Wagner, R. (1982) Nucleic Acids Res. 10, 1257-1269
7 Chiam, C.L. and Wagner, R. (1983) Biochemistry 22, 1193-1200 8 Szymkowiak, C. and Wagner, R. (1985) Nucleic Acids Res. 13, 3953-3968 9 Melancon, P., Boileau, G. and Brakier-Gringas, L. (1984) Biochemistry 23, 6697-6703 10 Hardy, S.J.S., Kurland, C.G., Voynow, P. and Mora, G. (1969) Biochemistry 8, 2897-2905 11 Glass, L. and Pelzig, M. (1978) Biochem. Biophys. Res. Commun. 81,527-531 12 Mendel-Hartwig, I. (1982) Anal. Biochem. 121,215-217 13 Laemmli, U.K. (1970) Nature 227, 680-685 14 Geyl, D., B6ck, A. and Isono, K. (1981) Mol. Gen. Genet. 181,309-312 15 Henderson, L.E., Oroszlau, S. and Konigsberg, W. (1979) Analytical Biochem. 93, 153-157 16 Brimacombe, R., Maly, P. and Zwieb, C. (1983) Progr. Nucleic Acids Res. 28, 1-48 17 Barritault, D., Expert-Bezancon, A., Milet, M. and Hayes, D. (1975) FEBS Lett. 50, 114-120 18 Traut, R.R., Lambert, I.M., Boileau, G. and Kenny, J.W. (1980) in Ribosomes: Structure, Function and Genetics (Chambliss, G., Craven, G.R., Davies, J., Davis, K., Kahan, L. and Nomura, M., eds.), pp. 89-110, University Park Press, Baltimore, MD 19 Expert-Bezancon, A., Barritault, D., Milet, M. and Hayes, D.H. (1976) J. Mol. Biol. 108, 781-787 20 Lambert, J.M. and Traut, R.R. (1981) J. Mol. Biol. 149, 451-479 21 Osterberg, R., SjOberg, B., Pettersson, I., Liljas, A. and Kurland, C.G. (1977) FEBS Lett. 73, 22-24 22 Kenny, J.W., Fanning, T.G., Lambert, J.M. and Traut, R.R. (1979) J. Mol. Biol. 135, 151-170 23 StOffler-Meilicke, M., Noah, M. and StUffier, G. (1983) Proc. Natl. Acad. Sci. USA 80, 6780-6784 24 Ramakrishnan, V., Capel, M., Kjeldgaard, M., Engelman, D.M. and Moore, P.B. (1984) J. Mol. Biol. 174, 265-284
1
BBA 32403
Identification of neighbouring proteins by cross-linking of intact 70 S ribosomes from Escherichia coli Ute Scheibe and Rolf Wagner * M a x -Planck-lnstitut ft~r Molekulare Genetik, A bteilung Wittmann, lhnestrasse 63 - 73, 1000 Berlin 33 (Germany)
(Received July 18th, 1985)
Key words: Chemical cross-linking; 70 S ribosomes; Structural organization: ( E. coil )
70 S ribosomes from Escherichia coil have been reacted with the bifunctional reagent 1,4-phenyldiglyoxal under near physiological conditions. As a result of the cross-linking reaction a number of high-molecularweight protein fractions with altered electrophoretic mobility could be isolated. A new chemical procedure has been introduced to reverse the cross-links between proteins at least partially. The cleavage reaction did not affect the gel electrophoretic mobility of the proteins. Thus a direct identification of cross-linked proteins using one- or two-dimensional gels was made possible. Two protein trimers, $3-$4-$5 and LI-$4-$5, as well as five protein dimers, $3-$4, L 6 - L 7 / 1 2 , L I 0 - L 7 / 1 2 , $9-L19 and LI8-LI9 could be identified as close neighbours in the E. coil 70 S ribosome. The protein pairs $9-L19 and LI8-LI9 had previously not been identified as near neighbours using cross-linking studies.
Introduction Investigations of the topography of the bacterial ribosome have employed many different techniques including electron microscopy, neutron scattering, fluorescence energy transfer, X-ray crystallography, chemical modification or crosslinking (for reviews see Ref. 1-3). From these studies a rough three-dimensional model of the E. coli ribosome can be constructed. The localisation of a number of ribosomal components by the different methods agrees well, while the position of some others are still the subject of controversy. For the construction of a valid three-dimensional model of the ribosome it is important therefore to obtain firm data by comparison of the results of several independent studies. Since high-resolution methods like X-ray crystallography do not yet provide enough data for a
* To whom correspondence should be addressed.
detailed structural model [4] chemical cross-linking is still one of the most important and direct methods for the elucidation of the ribosomal topography. Possible artefacts of this method are greatly reduced if certain cross-linking results can be repeatedly found using different conditions and reagents. In a recent study we used the cross-linking reagent 1,4-phenyldiglyoxal for a systematic analysis of the RNA-protein neighbourhoods within and across the ribosomal subunits. The same reagent has been used by us and others in a number of studies for R N A - R N A cross-linking as well as for affinity labelling [5-9]. Until now we were unable to use the same reagent for the analysis of protein-protein cross-links. Such products were consistently formed under the reaction conditions but no method for the cleavage of the protein-protein cross-links was available. Recently we have developed a method to reverse the protein crosslinks. This enabled us to identify a number of protein neighbourhoods by direct gel electro-
0167-4838/86/$03.50 © 1986 Elsevier Science Publishers B.V. (Biomedical Division)
phoretic analysis. The cross-links within the 70 S ribosomes were introduced under similar conditions as described for the RNA-protein analysis [7]. Among the cross-links identified are protein trimers and dimers. Some of them are cross-linked within the ribosomal subunits, while others are found to consist of proteins from the small and the large subunit and therefore are considered to be interface components. The results are discussed with respect to the known topography of the 70 S ribosome. Materials and Methods
1,4-Phenyldiglyoxal was synthesized as described [5]. Acrylamide, N,N'-methylenebisacrylamide, sodium dodecyl sulfate (SDS) were from Bio-Rad, U.S.A. Urea (ultra pure) was from BRL Inc., U.S.A. 1,2-Phenylenediamine was from Merck, Darmstadt. The radioiodination system for the radioactive labeling of proteins was from New England Nuclear, U.S.A. Isolation of 70 S ribosomes. Ribosomes were prepared from E. coli MRE 600 cells as described in Ref. 7. Cross-linking of 70 S ribosomes. 70 S ribosomes (0.4 nmol/ml) were incubated with 0.35 mM phenyldiglyoxal at 37°C for 4 h. The reaction was stopped by the addition of 0.7 vol. of ethanol ( - 2 0 ° C ) and the ribosomes were precipitated for 2h.
a model compound. Lysozyme was cross-linked intermolecularly with phenyldiglyoxal, and the protein dimer was isolated by gel electrophoresis according to Ref. 12. The lysozyme dimers were treated under various conditions with 1,2-phenylenediamine. Incubation of cross-linked dimers with 50 mM 1,2-phenylenediamine at 37°C for 16 h was found to be optimal for the reversal of the cross-link bonds. Under these conditions at least 40% of the cross-linked dimers could be converted to single proteins. The cleaved proteins did not show altered electrophoretic mobilities in one- or two-dimensional gel systems. Radioactive labelling of marker proteins. A preparation of total ribosomal proteins not reacted with phenyldiglyoxal was radioactively labelled with 125I using the standard lactoperoxidase labelling procedure. The labelled proteins were used as markers at concentration at which clear autoradiograms could be obtained, but the 125I labelled protein concentration was too small to be stained with Coomassie blue. Gel electrophoretic separation of proteins. Gel electrophoretic separations of proteins were performed in one or two dimensions as described [13,14]. The proteins were stained with Coomassie brilliant blue. In cases where only a limited number of proteins had to be assigned on two-dimensional gels a small amount of 125I-labelled marker proteins were separated simultaneously. The gels were first stained and then autoradiographed after drying.
Extraction of ribosomal proteins. After the cross-linking reaction and precipitation of the ribosomes the total protein fraction was extracted with acetic acid according to Hardy et al. [10]. The proteins were dialyzed against decreasing acetic acid concentrations and finally lyophilized. The lyophilized proteins were dissolved in 6 M urea/10 mM dithioerythritol/5 mM EDTA and stored at - 2 0 ° C . Cleavage of the protein-protein cross-links. No cleavage of the protein-protein cross-links could be achieved under conditions where RNA-RNA cross-links produced with phenyldiglyoxal could be reversed completely. We therefore took advantage of the specific reaction of a-dicarbonyl derivatives with 1,2-phenylenediamine [11]. To optimize the cleavage conditions we used lysozyme as
Results
When the total protein fraction of phenyldiglyoxal-reacted 70 S ribosomes was analysed on one- or two-dimensional gel systems, cross-linking between individual proteins could be demonstrated by the appearance of new high-molecularweight protein products. In Fig. 1 the separation of total ribosomal proteins extracted before and after reaction with phenyldiglyoxal is shown on a one-dimensional SDS gel. Protein bands labeled I to IV in Fig. 1 indicate the formation of cross-linked proteins and can clearly be distinguished from the migration positions of the unreacted 70 S proteins. Some of the protein bands are reduced in their intensity compared to the unreacted 70 S
C1
x2
x3
--I
III IV
Fig. 1. One-dimensional separation of ribosomal proteins before and after cross-linking. C 1, C2: Total ribosomal proteins extracted from unreacted ribosomes; 30 and 6 /xg of protein, respectively. X ], X 2, X3: Total proteins extracted from crosslinked 70 S ribosomes. The total amount of 60, 30 and 6 # g proteins was separated, respectively. The ribosomal protein S1 can not be seen on the control tracks as a result of the acetic acid extraction procedure. Its presence on the ribosomes used for the cross-linking was established, however, by the LiCl-urea extraction procedure.
sample. However, under the cross-linking conditions many proteins are cross-linked simultaneously to the ribosomal RNAs [7]. The altered intensities of some of the protein bands in the cross-linked samples visible in Fig. 1 are therefore difficult to interpret with respect to protein-protein cross-linking. When the cross-linked 70 S ribosomal proteins are separated on two-dimensional gels (Fig. 2) the cross-linked products (labeled I to IV in Fig. 1) are resolved into a more complex pattern of individual spots (Fig. 2). Spots containing the cross-linked proteins were cut out and collected from 8-10 two-dimensional gels. The proteins were extracted electrophoretically in the presence of 0.1% SDS and lyophilized. The samples were dissolved in water and divided in two aliquots, one of which was treated with the cleavage reagent. Cleaved and uncleaved protein samples were analyzed on oneor two-dimensional gels. The proteins analyzed on two-dimensional gels were first freed from residual SDS according to [15] and then mixed with radioactive marker proteins prior to separation. The gels were stained and autoradiographed after drying with a gel dryer. Not all the products that appeared as new spots on the two-dimensional gel after cross-linking (Fig. 2) could be analyzed unambiguously. Often material extracted from single spots proved to be heterogeneous. In seven cases, however, a direct assignment of the proteins involved in cross-linking could be achieved. The results of these analyses are shown in Figs. 3-5. In Fig. 3A, B the analyses of cross-linked
TABLE I RELATIVE YIELDS A N D M O L E C U L A R W E I G H T C O M P A R I S O N OF T H E CROSS-LINKS I D E N T I F I E D Yields are given as relative values estimated from the staining intensities of the two-dimensional gels. + + + , strong; + + , medium; + , weak ( + ) very weak.
la Ib II Ilia llIb IVa IVb
Cross-links
Mapp. (kDa)
M~al. (kDa)
Yield
$3-$4-$5 L1-$4-$5 $3-$4 L7/12-L10 L7/12-L6 $9-L19 L18-L19
67 67 47 30 30 26 26
66.5 65.3 48.9 30.0 31.0 27.6 25.7
+ + + + + + + + (+ ) (+ ) + + +
1 st
:
O.
12
S~
SlO~
$12
S17
~k
27
material corresponding to spot I (Fig. 2) is shown before and after hydrolysis of the cross-link. Four protein spots and some remaining uncleaved material are apparent (Fig. 3B). The four proteins were assigned according to the autoradiogram of the 125I-labelled marker proteins as L1, $3, $4 and $5. The relative molar ratios of these four proteins was estimated from the staining intensities to be 1 : 1 : 2 : 2 . The apparent molecular mass of the cross-link fraction I (Fig. 1) can be calculated from the migration position on the SDS gel to be about 67 kDa, which can be explained only with protein trimers. The cross-link fraction I consists therefore most probably of two protein trimers with the common proteins $4 and $5. The trimers can consequently be assigned as: $3-$4-$5 and L1-$4-$5. The cross-link fraction II was very heterogeneous and consisted of many new protein spots as evidenced from Figs. 1 and 2. The spot II in Fig. 2 could, however, be analyzed unambiguously. In Fig. 3C and D the two-dimensional gel-electrophoretic analysis of the products extracted from
Fig. 2. Two-dimensional separation of proteins from 70 S ribosomes after the cross-linking reaction. New protein spots are encircled and numbered I to IV.
spot II (Fig. 2) is shown before (C) and after (D) cleavage of the cross-link. Two monomeric proteins could be identified in Fig. 3D and were assigned as $3 and $4. The apparent molecular mass of the corresponding cross-link pair (47 kDa) as determined by SDS gel electrophoresis of cross-link fraction II correlates well with the calculated molecular mass, 48.9 kDa. The cross-link fraction III (Fig. 1) was resolved in several spots on the two-dimensional gel (Fig. 2). Spots IIIa and I I I b gave clear analysis results after cleavage of the cross-link (Fig. 4). Spot Ilia consists of the proteins L 7 / 1 2 and L10. Again, the molecular mass of the cross-linked dimer determined from an SDS gel (30 kDa) is in perfect agreement with that calculated (30 kDa). Analysis of material extracted from spot IIIb showed that it was contaminated with products from spot IIIa. The protein L 7 / 1 2 can be found in both spots but with a higher relatiye stoichiometry in I I I b (Fig. 4). The protein L6 is only found in spot IlIb. Spot I I I b consists therefore of the cross-link dimer L6-L7/12 with the contamination
1 st
~_
TP70
III
f~ .-I
Q.
~i~i~i!~'i~i~!~/ ~!~i~i~!~i~!i~!~i~il!i!i~i
Fig. 4. One-dimensional analysis of the cross-link fractions Ilia and lllb. The cross-linked material was isolated from two-dimensional gels and separated on an SDS gel after cleavage of the cross-link. (a, b) Uncleaved protein from cross-link fraction II1.
L1
o
S3 Q
TP 30 50
IV
I S4 S~PL6 i
i
b
Fig. 3. Two-dimensional analysis of cross-linking products. (A) Purified cross-linking product I separated on a two-dimensional gel without cleavage of the cross-link. (B) Same material as in (A) after cleavage of the cross-link. (C) Purified cross-link product II separated without cleavage reaction. (D) Same material as in (C) after cleavage of the cross-link. (E) Separation pattern of ]251-labelled ribosomal marker proteins. (F) Separation of cross-link fraction IVb after cleavage of the cross-link. The small a m o u n t of contaminating L2 was also present in the uncleaved sample. In A, B, C, D and F the gels were stained with Coomassie brilliant blue, while in E an autoradiogram is shown.
J
L18~'-
Fig. 5. One-dimensional analysis of the cross-link fractions IVa and IVb. The material was separated after cleavage of the cross-link as indicated in the legend to Fig. 4.
of the dimer L7/12-L10 from spot IIIa. A similar situation was encountered during the analysis of the cross-link fraction IV. The analysis of material extracted froms pots IVa and IVb (Fig. 2) on a one-dimensional SDS gel after cleavage of the cross-link with 1,2-phenylenediamine is shown in Fig. 5. The two proteins $9 and L19 can be identified from material extracted from spot IVa. Spot IVb yielded the proteins L18 and L19 after the cleavage reaction. Contaminations of spot IVa were, however, frequently apparent in fraction IVb. The ribosomal protein $9 could be seen in addition and the relative stoichiometry of L19 was higher than that of L18. The results were confirmed by two-dimensional gel electrophoresis (see Fig. 3F). The presence of L19 in the cross-link fraction IV was additionally substantiated by immunoblotting using specific antibodies (data not shown). We conclude therefore that the cross-link fraction IV consists of two protein dimers, namely $9-L19 (IVa) and L18-L19 (IVb). The results of the gel electrophoretic analysis are summarized in Table I. Discussion
Among the many products formed under the cross-linking conditions only those fractions were analyzed that appeared reproducibly and in relatively high yields. Some of the cross-linking products, although of high yield (see Fig. 2) were, however, too complex in their protein composition to be analyzed unambiguously. The yields of some other cross-linked protein fractions such as I and II (Fig. 2) seem to be high enough to be isolated in amounts sufficient to allow the analysis not only of the proteins but also of the amino acids involved in cross-linking. Such analysis has been performed to identify the nucleotide and amino acid moieties of RNA-protein cross-links and should be the final goal of all cross-linking studies [16]. Under the cross-linking conditions used in this study the structure of the ribosomes was not altered to a noticeable extent as has been shown in a previous study [7]. From the comparison of the two-dimensional electrophoresis of non-cross-linked versus crosslinked proteins, it is apparent that no major mobil-
ity shifts occur for any of the ribosomal proteins. Furthermore, no change in mobility could be detected for the proteins treated with 1,2-phenylenediamine. Therefore the assignment of the cross-linking proteins by gel electrophoretic methods is valid. When proteins are highly modified with phenyldiglyoxal a small north-west shift on twodimensional gels can be observed. The correct assignment is not affected, however (see also Ref. 7). All the results presented in this study were obtained between two and five times.
Comparison with other topographical data The spatial neighbourhoods of the cross-linked proteins L1-$4-$5, $3-$4-$5, $3-$4, L6-L7/12 and L10-L7/12 identified in this study have been reported in similar or identical arrangement by other investigators using different cross-linking reagents [17-19]. For instance, the protein dimers $3-$4, $4-$5 and also $3-$5 could be identified as crosslinked. The large subunit protein L1 and the small subunit protein $4 known as interface components have also been cross-linked [20]. The two protein trimers L1-$4-$5 and $3-$4-$5 are therefore well substantiated by these studies and their proximity in the ribosome can be taken as firmly established by cross-linking studies. The protein dimers L6-L7/12 and L10-L7/12 have also been cross-linked with different reagents [18,19]. Furthermore, the complex L10-L7/12 can be extracted from unreacted ribosomes [21]. On the other hand, the two protein dimers $9-L19 and L18-L19 have to our knowledge not yet been reported as being cross-linked. The proximity of these proteins in the 70 S ribosome is, however, in perfect agreement with existing topographical data based on cross-linking studies. The two proteins L18 and El9 have both been shown to form dimers with the protein $13 and can therefore be considered as interface proteins and near neighbours [20]. The protein $9 is also an interface protein and has been cross-linked to L5 [22]. The distances of the proteins L5 and L18 can be calculated to be very short because they are known as direct binding proteins to the ribosomal 5 S RNA. Consequently the three proteins $9, L18 and L19 may all be close in the 70 S ribosome. Our crosslinking results confirm such a spatial relationship. There are some differences in the distances of
p r o t e i n s in situ b a s e d o n c r o s s - l i n k i n g results c o m p a r e d to i m m u n e e l e c t r o n m i c r o s c o p i c studies. T h e sites for p r o t e i n s L1 on the o n e h a n d a n d $4, $5 o n the o t h e r h a n d as l o c a l i z e d b y i m m u n e e l e c t r o n m i c r o s c o p y are r a t h e r r e m o t e to e a c h other. Similarly, the d i s t a n c e o f the sites d e t e r m i n e d for L18 a n d L19 s e e m to be t o o far to be b r i d g e d by a c r o s s - l i n k i n g r e a g e n t [23]. T h e p r o x i m i t y of L18 a n d $9 is s u p p o r t e d , h o w e v e r , b y n e u t r o n s c a t t e r i n g results a c c o r d i n g to w h i c h p r o t e i n $9 is l o c a t e d in the h e a d of the 30 S s u b u n i t [24]. T h i s l o c a t i o n is fairly close to the p o s i t i o n of L18 in the 70 S r i b o s o m e [23].
Acknowledgements W e are g r a t e f u l to P r o f e s s o r H . G . W i t t m a n n for his s u p p o r t . W e like to t h a n k Dr. H.J. S c h 6 n f e l d for m a n y h e l p f u l d i s c u s s i o n s a n d Dr. M. St6fflerM e i l i c k e for p r o v i d i n g us w i t h r i b o s o m a l p r o t e i n antisera.
References 1 Chambliss, G., Craven, G.R., Davies, J., Davis, K., Kahan, L. and Nomura, M., (eds.) (1980) Ribosomes: Structure, Function and Genetics University Park Press, Baltimore, MD 2 Liljas, A. (1982) Progr. Biophys. Mol. Biol. 40, 161-228 3 Winmann, H.G. (1983) Annu. Rev. Biochem. 52, 35-65 4 Yonath, A. (1984) Trends Biochem. Sci. 9, 227-230 5 Wagner, R. and Garrett, R.A. (1978) Nucleic Acids Res. 5, 4065-4075 6 Hancock, J. and Wagner, R. (1982) Nucleic Acids Res. 10, 1257-1269
7 Chiam, C.L. and Wagner, R. (1983) Biochemistry 22, 1193-1200 8 Szymkowiak, C. and Wagner, R. (1985) Nucleic Acids Res. 13, 3953-3968 9 Melancon, P., Boileau, G. and Brakier-Gringas, L. (1984) Biochemistry 23, 6697-6703 10 Hardy, S.J.S., Kurland, C.G., Voynow, P. and Mora, G. (1969) Biochemistry 8, 2897-2905 11 Glass, L. and Pelzig, M. (1978) Biochem. Biophys. Res. Commun. 81,527-531 12 Mendel-Hartwig, I. (1982) Anal. Biochem. 121,215-217 13 Laemmli, U.K. (1970) Nature 227, 680-685 14 Geyl, D., B6ck, A. and Isono, K. (1981) Mol. Gen. Genet. 181,309-312 15 Henderson, L.E., Oroszlau, S. and Konigsberg, W. (1979) Analytical Biochem. 93, 153-157 16 Brimacombe, R., Maly, P. and Zwieb, C. (1983) Progr. Nucleic Acids Res. 28, 1-48 17 Barritault, D., Expert-Bezancon, A., Milet, M. and Hayes, D. (1975) FEBS Lett. 50, 114-120 18 Traut, R.R., Lambert, I.M., Boileau, G. and Kenny, J.W. (1980) in Ribosomes: Structure, Function and Genetics (Chambliss, G., Craven, G.R., Davies, J., Davis, K., Kahan, L. and Nomura, M., eds.), pp. 89-110, University Park Press, Baltimore, MD 19 Expert-Bezancon, A., Barritault, D., Milet, M. and Hayes, D.H. (1976) J. Mol. Biol. 108, 781-787 20 Lambert, J.M. and Traut, R.R. (1981) J. Mol. Biol. 149, 451-479 21 Osterberg, R., SjOberg, B., Pettersson, I., Liljas, A. and Kurland, C.G. (1977) FEBS Lett. 73, 22-24 22 Kenny, J.W., Fanning, T.G., Lambert, J.M. and Traut, R.R. (1979) J. Mol. Biol. 135, 151-170 23 StOffler-Meilicke, M., Noah, M. and StUffier, G. (1983) Proc. Natl. Acad. Sci. USA 80, 6780-6784 24 Ramakrishnan, V., Capel, M., Kjeldgaard, M., Engelman, D.M. and Moore, P.B. (1984) J. Mol. Biol. 174, 265-284