The ribosomal protein S1 was modified by reductive methylation of some of its lysyl ammonium groups (SI*). With 6 out of 30 groups methylated the protein lost its capacity to form stable complexes with polyuridylate. Addition of excess polyuridylate inhibited the methylation of the lysyl groups. In equilibrium dialysis experiments it was shown that the binding constant between SI* and Uls was lowered 10-fold as compared to the native protein. The pH-dependence of the complex formation between S1 and U,s confirms a participation of the lysyl residues. When S1 depleted 30-S ribosomes were reconstituted with methylated S1 these ribosomes were inactive in the poly(U) stimulated Phe-tRNA binding. The data are discussed with respect to a grid-like interaction between the lysyl groups of the protein and the phosphodiester bonds of the polynucleotide as a molecular basis of protein nucleic acid interac- tion.
Introduction
The ribosomal protein S1 functions in the initiation step and possibly in the elongation cycle of bacterial protein synthesis [1]. Furthermore it represents one of the subunits of Q~ replicase involved in the selective copying of the plus strand of Q/~ RNA [2]. Purification and physical properties of this protein were described in detail [3,4]. Various, somewhat contradictory suggestions of its molecular function exist: S1 was termed a discrimination factor i.e. it might select mRNAs for binding to the ribosomes [5], or unwinding protein for double stranded RNA [6]. Van Dieijen et al. postulated that it recognizes an initiation type tertiary structure of the mRNA, whereas Dahlberg et al.
477
provided evidence, that it stabilizes the region on the 3'-end of 16 S RNA in a conformation optimized towards double strand formation with the inter- cistronic sequence of a mRNA [7,8].
It is well known, that protein S1 can form stable complexes with poly- nucleotides like polyuridylate [9]. We were able to show that complex forma- tion does not depend directly on the nature of the base moiety. A chain length of 12 to 15 nucleotides and a low Tm-value (<25°C) are optimal for complex formation [10]. This led us to assume that despite the pI value of 4.8 [11], the S1 • polynucleotide complexes could be stabilized by multifold electrostatic interaction between the phosphodiester groups of the polynucleotide and positively charged groups of the protein S1.
It was shown by Moore and Crichton, that protein S1 in the 30 S ribosome may be labeled by reductive methylation using formaldehyde and [3H]sodium borohydride without loss of activity [12].
We used this procedure to investigate the participation of the lysyl groups of S1 in the complex formation with polynucleotides.
In the following we present evidence, that S1 with 6 out 30 modified lysyl groups is still bound to 30-S ribosomes, but is inactive in polyuridylate binding as examined in the nitrocellulose filter assay.
The data are discussed with respect to a screen-like interaction of opposite charges as a model for protein nucleic acid complex formation.
Materials and Methods
tRNA Pae and nucleoside diphosphates were purchased from Boehringer, Mannheim. All radioactivity labeled compounds came from Amersham Buchler, Braunschweig, and Escherichia coli MEE 600 type were from Merck, Darm- stadt. Nitrocellulose filters were procured from Schleicher and Schiill, Dassel, dialysis membranes were a gift from RhSne-Poulenc, Paris.
Protein S1 was extracted from isolated 30-S ribosomal subunits as described by Tal et al. [13]. After removal of contaminating oligonucleotides by DEAE- cellulose chromatography it showed an absorbance ratio A2so,m/A26onm of 1.65.
For functional studies 30-S ribosomal subunits were isolated from '70-S tight couples' by zonal centrifugation [14,15]. They were activated before use according to Zamir et al. [16] 30-S ribosomal subunits lacking S1 (30-S (--1)) were prepared by adsorbance of 30 S to a Sepharose 4B-anti-S1 IgG column followed by elution of the S1 deprived subunits with 1 M NH4C1 [17]. tRNA Phe form yeast was charged with [3H]phenylalanine (spec. act. 1 Ci/mmol) using a partially purified synthetase preparation. Poly(U). [3H]poly(U) and [3H]U14 were prepared by polunucleotide phosphorylase (EC 2.7.7.8) catalysed polymerisation of uridine 5'-diphosphate [ 18].
Reconstitution of S1 deprived 30-S ribosomal subunits with purified S1 was performed at a concentration of 2 A2eo units of 30-S (--1) in 100 ~1 reconstitu- tion buffer (16 mM MgC12, 200 mM NI-LCI, 50 mM Tris-HC1, pH 7.5, and 2 mM dithioerythritol) for 25 min at 37°C. The molar ratio of 30 S (--1) to S1 was 1 : 2. [3H]poly(U) binding to S1 was measured by the adsorption of the [3H]-
478
poly(U)" $1 complex to alkaline pretreated nitrocellulose filters [9]. [3H]U14 binding was studied by equilibrium dialysis as described in detail [10]. The poly(U) directed Phe-tRNA binding activity of ribosomes was measured by the method of Nirenberg and Leder [19]. Reductive methylation of protein $1 was carried out in principle, similar to Moore and Crichton [12]. 10 nmol $1 were dissolved in 100 pl buffer (100 mM sodium borate, pH 9.2, 10 mM magnesium acetate, 20 mM KC1) and 50 /~1 (0.075 M) formaldehyde solution were added under stirring. After 30 s, 50 pl NaB3H4 (0.1 M in 0.01 M NaOH spec. act. 500 CI/mol) were added and stirring was continued for additional 60 s. After 10 min the whole procedure was repeated. The amount of protein labeled was determined by two methods:
1. Adsorption of the protein to nitrocellulose filters which were pretreated with 0.5 M NaOH for 45 min at 20°C.
2. The reaction mixture was spotted onto glass fiber filters and the protein S1 was precipitated in 10% trichloroacetic acid. After successive washings in 5% trichloroacetic acid, ethanol, ethanol/ether (1 : 1, v/v) and ether, the filters were dried and counted. Both methods gave identical results.
The number of repetitions are indicated in Fig. 1. Finally the modified S1 (SI*) was separated from excess reagents by chromatography over Sephadex G-50 (1 cm × 50 cm) and concentrated by lyophilisation 22% of the lysines were converted to e-N-methyllysine as shown as by amino acid analysis. The pH-dependence of the SI" [3H]U14 complex formation was followed by equilibrium dialysis. In the pH range from 2--8, 25 mM sodium phosphate + 200 mM NaC1 was used as buffer, and 25 mM glycine/ NaOH + 200 mM NaC1
f r o m 8--12. Solubility was determined by incubation of the protein at the apropriate pH value followed by centrifugation at 20 000 rev./min for 30 min at 0°C. An aliquot of the supernatant was readjusted to pH 7.5 and the amount of protein S1 was measured by the [3H]poly(U) binding assay and in addition by the Lowry method.
Results
5--6 lysines are modified after three repetitions of the reductive methyla- tion with no further increase upon continuation of the methylated cycles (Fig. 1). This experiment was repeated using the same procedure but with non- radioactive NaBI-I4 and the methylated protein S1 was examined in its capacity to form a complex with [3H]poly(U) (Fig. 2). When protein Sl was maintained at 0°C a linear decrease in the polyuridylate binding capacity was observed. When, however, the temperature was raised to 37°C for 10 rain [3H]poly(U) binding decreased linearly until 5--6 lysyl residues were modified with no further decrease in complex formation.
In order to prove that this inactivation of the ribosomal protein S1 is not due to formaldehyde fixation the following control experiment was done. 100 pmol S1 were incubated with formaldehyde at 0°C. When the reaction mixture was assayed directly in the [3H]poly(U) binding reaction no activity was found (Table I). Heating to 37°C for 15 rain, however, completely restores the S1 activity. Even after a 5-fold repetition only a minor inactivation of the protein
4 7 9
loo
O
N 5O
-10
~2-
w
L--
E
3
?
J
~6
E
• AS I
7o 700 protein S 1
," ^$1 • , , , , ,
1 3 5 repetit ions of reductive methy[(~tion [pmol]
Fig. 1. I n a c t i v a t i o n o f p r o t e i n S1 b y r e d u c t i v e m e t h y l a t i o n . A f t e r t h ree m e t h y l a t i o n c y c l e s 6 - - 7 ly s ines a re m o d i f i e d ( . . . . . ) a n d p o l y ( U ) b i n d i n g ac t iv i t y d r o p s t o 2 0 - - 3 0 % ( ). In the p r e s e n c e o f excess p o l y u r i d y l a t e r e d u c t i v e m e t h y l a t i n n is l o w (e) . The n u m b e r o f l y s y l r e s idues m o d i f i e d w a s c o n f i r m e d b y a m i n o ac id ana lys i s .
Fig. 2. R e d u c t i v e m e t h y l a t i o n o f S1 b l o c k s p o l y ( U ) b i n d i n g ac t iv i ty . Wi th 6 - - 7 lys ines m o d i f i e d t h e p o l y ( U ) • S I * c o m p l e x e s are n o m o r e r e t a i n e d o n n i t r o c e l l u l o s e f i l ters . C o m p l e x f o r m a t i o n w a s a s s a y e d in 1 m l 1 0 m M Tris-HC1, p H 7 .6 , 1 5 m M MgC12 w i t h 1 0 0 p m o l S1 a n d excess [ 3 H ] p o l y u r i d y l a t e [ 1 0 ] .
S1 was found. This may be taken as evidence that the protein was neither cross- linked nor denatured by formaldehyde treatment.
Since the poly(U) • S1 complex dissociates so slowly, that it can be retained on nitrocellulose filters, polyuridylate should protect the lysines in the binding site from methylat ion. Thus methylat ion was repeated in the presence of excess polyuridylate. The experiment was done in the dark to avoid reduction of the uracil 5,6 double bond [20] . A 10-fold molar excess of polyuridylate protects the lysyl groups of S1 from methylat ion by more than 80% (Fig. 1).
If the association constant between S1 and Uls (KAss =2.5 • 107 M -1) is com- pared to the SI* • Uls complex (KAss = 1.0 • 106 M -1) at one hand and the S1 • U9 complex (KAss = 1.1 • 106 M -l) on the other hand, in both complexes the association is reduced 10-fold as compared to S1 • Uls. This may be taken
T A B L E I
F O R M A L D E H Y D E T R E A T M E N T D O E S N O T I N A C T I V E $1 IN P O L Y ( U ) B I N D I N G
1 n m o l S1 in 2 1 5 ~1 in p H 9 . 2 b u f f e r (see M e t h o d s ) w a s t r e a t e d w i t h 1 0 0 #1 f o r m a l d e h y d e s o l u t i o n . 3 0 /J1 (90 p m o l $ 1 ) w e r e a s s a y e d in t he p o l y ( U ) b i n d i n g t e s t ( n i t roce l lu lo se f i l te r b i n d i n g ) . F o r t he 37°C t r e a t m e n t 3 0 /J1 w e r e d i l u t e d w i t h 1 .0 m l o f b i n d i n g b u f f e r , m a i n t a i n e d a t 37°C f o r 1 5 m i n , c o o l e d t o O°C a n d a s s a y e d f o r p o l y u r i d y l a t e b i n d i n g . 2 6 0 0 c p m c o r r e s p o n d to 1 0 0 % $1 b i n d i n g ac t i v i t y .
R e p e t i t i o n s F o r m a l d e h y d e B i n d i n g b u f f e r (0°C, c p m ) ( p H 7 .6 , 1 5 m i n , c p m )
C o n t r o l 2 6 0 0 1 4 5 0 2 7 1 0 2 3 0 0 2 0 4 8 3 2 9 0 2 1 0 0 4 3 2 0 2 2 5 0 5 4 0 0 2 0 8 0
100
6(
2(
cD o 0.~.
4 8 0
A~A
"100
2O
! 1'3 pH
Fig. 3. p H - d e p e n d e n c e o f c o m p l e x f o r m a t i o n as assayed by equi l ibr ium dialysis. , a m o u n t o f c o m - Plex f o r m e d w i t h U I S . Th e first m i n i m u m at pH 4 . 5 c o r r e s p o n d s to the p l = 4.6 and a so lubi l i ty m i n i m u m o f the prote in ( . . . . . ). T h e decrease leading to the s e c o n d m i n i m u m has a po in t o f in f l ec t ion at 10.5 w h i c h cou l d correspond to the pK o f th e e - a m m o n i u m groups o f lystnes .
as evidence that the grid-like interaction between S1 and polynucleotides is reduced in number from 14 to about 8 lysyl-phosphate bonds.
Since the preceeding experiments point towards a participation of e-ammo- nium groups of the lysyl moiety in the complex formation we examined the pH-dependence of this binding reaction (Fig. 3). Equilibrium dialysis was used instead of the nitrocellulose filter technique, because [3H]poly(U) tends to stick to the filters at pH-values above 9. Two minima in the binding capacity are found. The one at a lower pH-value represents a solubility minimum near the pI-value of the protein. The transition point of the slope leading to the second minimum could coincide with the pK of a lysyl group (Fig. 3). This pH dependence of complex formation again points towards a functional role of the lysyl moieties of protein S1.
It was reported by Moore et al. that protein S1 in solution exchanges with S1 bound to the 30 S ribosome. No such behaviour is found with methylated
0,2 -
^
1() 20
• 1S x
.10
.5
\ 30 [ ml ]
Fig. 4. R e a s s o c i a t i o n o f $1" w i t h 30 S ( - -1 ) r ibosomes . $1 wa s r e m o v e d f r o m 30-S r i b o s o m e s w i t h Sepha ros e 4B-ant i $1 IgG. Pro te in $1 was labe led by reduct ive m e t h y l a t i o n w i t h N a B 3 H 4 . R e c o n s t i t u t e d 30 S ( - -1 ) + S I * r i b o s o m e s w e r e separated f r o m e x c e s s $1" b y Sepharose 6B c h r o m a t o g r a p h y . (~, A 2 6 0 units , A cpm) .
481
t,9 o
-6
°~0
~ s r~ i
o-
-6 E
o 70S
• 70 S (-1)
• 70S ( - 1 ) + S 1
10 20
b0 o
CL 0 ~ I 0 '
8 o
~z s. t
o 30S b
'~ 30 S(-1)
• 3 0 S ( - I ) , S I
10 20
poly(U) [nrnot pO]
Fig. 5. P o l y ( U ) d e p e n d e n t P h e - t R N A binding to S I * reconst i tuted 30-S and 70-S r i b o s o m e s . S l depleted 30-S and 70-S as wel l as S I * r e c o n s t i t u t e d r i b o s o m e s are inac t ive in P h e - t R N A b ind ing , Activ i ty can be restored by a d d i t i o n of t w o f o l d excess of $1 .
$1. If, however, S1 depleted ribosomes are incubated with excess $1" in 16 mM Mg 2+ at 37°C, 40--50% of the subunits become reconsti tuted with SI* (Fig. 4).
30 S (--1) and 70 S reconst i tuted from 30 S (--1) and 50 S show a reduced activity in the poly(U) stimulated binding of Phe-tRNA. Both regain full activ- ity when they are reconst i tuted with stoichiometric amounts of protein $1. Ribosomes, however, which were reconst i tuted with SI* are inactive in Phe-tRNA binding (Fig. 5). This cannot be explained by the poly(U) binding capacity of (30 S ( - 1 ) + SI*) ribosomes, since 30 S and 30 S (--1) show no dif- ference in poly(U) binding.
Discussion
Although protein nucleic acid complexes play a dominant role in the life cycle of a cell very little is known about the molecular details of this complex formation. Some progress has been made in the gene 5 protein DNA complex, the interaction of histones with DNA and the molecular structure of tobacco mosaic virus. The ribosomal protein S1 plays an interesting double functional role as outl ined in the introduct ion. In our opinion the complex between the protein S1 and homopolynucleot ides is a suitable model system to s tudy the molecular details o f protein nucleic acid recognition. It forms a 1 : 1 complex with Uls and we have no indication of a strong and a weak binding site in this system [21]. Since this interaction shows no base specificity, but requires a defined number of phosphate groups and a high flexibility of the nucleic acid (Tin < 25°C) for optimal complex formation, it seems to be a plausible assump- tion that a grid of either opposite charges (Lys, Arg, His) or hydrogen donating groups to the .'>P = O bond of the nucleic acid represents the major force for complex formation. The observed pH-dependence of the U14" S1 complex strongly points towards the participation of the lysyl group. One has to men- tion, however, that the acidic pK of the uracil moieties of poly(U) (PKa = 9.2) is very close to the point of inflection found in the S1 • Uls ti tration experiment [22]. The role of the lysines is further supported by inactivation of the protein
4 8 2
$1 by the methylat ion of about six lysines. Since this modification is blocked by excess polyuridylate, the lysines should be located in the polynucleotide binding center.
From the optimal complex formation between $1 and U15 one is tempted to conclude that about 14 charge interactions stabilize this complex. Therefore the question arises why the modification of only six lysyl residues blocks com- plex formation. If one compares the asociation constants for Uls and U9 than this value drops from Knss = 107 to 106 M -1. Since complex formation was assayed by nitrocellulose filtration the rate dissociation constant of the com- plex may be too high to withstand the necessary washing procedures. In equilib- rium dialysis experiments the association constant between modified SI* and U~5 drops to KAss = 106 M -~ which is similar to the $1 • U9 value. In both cases the loss of about six charge interactions is reflected in a 10-fold lower associa- tion constant.
Naturally it cannot be ruled out that the modification of the lysines results in a conformational change of the protein. Since, however, the methylated protein still reconstitutes with S1 depleted 30 S ribosomal subunit, it is unlikely that a drastic change occurs.
In conclusion our data support multifold electrostatic inteaction between lysyl residues of the protein and the phosphodiester bonds of the nucleic acid as a molecular basis for complex formation. Since a low Tm value of the nucleic acid favours complex stability, it is likely that a defined arrangement of positive charges exists in the polynucleot ide binding site of this protein. This should not implicate that neighbouring lysyl residues are responsible for this interaction. One has to wait for the three dimensional structure of protein S1 to find out whether lysyl residues presumedly scattered within the primary sequence may form a lysyl grid within the tertiary structure of the protein. In general our data support a model of S1 function as outlined by Dahlberg et al. and Argetsinger-Steitz et al. [7,8].
Acknowledgements
We thank Mrs. E. RSnnfeldt and Mr. O.E. Beck for their help in preparing the manuscript. Supported by grants from the Deutsche Forschungsgemein- schaft and the Fonds der Chemischen Industrie.
References
1 Van Duin, J. and Van Knippenberg, P.H. (1974) J. Mol. Biol. 84, 1 8 5 - - 1 9 5 2 Kamen , R., Kondo , M., Romer , W. and Weismann, C. (1972) Eur. J. Biol. 31, 44--51 3 Sub raman ian , A.R. , Haase, C. and Giesen, M. (1976) Eur. J. Biochem 67, 591- -601 4 Carrniehael, G.C. (1975) J. Biol. Chem. 250, 6 1 6 0 - - 6 1 6 7 5 Revel, M., Greenshpan , H. and Herzberg, M. (1970) Eur. J. Bioehern. 16, 1 1 7 - - 1 2 2 6 Szer, W., Herrnoso, J.M. and Boublik, M. (1976) Biochem. Biophys. Res. C o m m u n . 70, 9 5 7 - - 9 6 4 7 Dahlberg, A.E. and Dahlberg, J.E. (1975) Proe. Natl. Acad. Sci. U.S. 72, 2 9 4 0 - - 2 9 4 4 8 Stcitz, J .A. and Jakes, K. (1975) Proc. Natl. Acad. Sci. U.S. 72, 4 7 3 4 - - 4 7 3 8 9 Smolsxsky, M. and Tal, M. (1970) Biochim. Biophys. Acta 213, 4 0 1 - - 4 1 6
10 Lipecky , R., Kohlschein , J. and Gassen, H.G. (1977) Nucl. Acids Res. 4, 3 6 2 7 - - 3 6 4 1 11 Laughrea , M. and Moore, P.B. (1977) J. Mol. Biol. 112, 3 9 9 - - 4 2 9 12 Moore, G. and Cr ich ton , R.R. (1973) FEBS Lett . 37, 74 - -78 13 Tal, M., Aviram, M., Kanarek , A. and Weiss, A. (1972) Bioehim. Biophys. Ac ta 281, 3 8 1 - - 3 9 2
4 8 3
14 Sypherd, P.S. and Wireman, J.W. (1974) Methods Enzymo]. XXX, 349, 354 15 Noll, M., Hapke, B., Schreier, M. and NoH, H. (1973) J. Mol. Biol. 75, 281--294 16 Zarnir, A., Miskin, R. and Elson, D. (1971) J. MoL Biol. 60, 347--364 17 Linde, R., Khanh, N.Q. and Gassen, H.G. (1978) Methods in Enzymology, Nucleic Acid and Protein
Synthesis Part G, in press 18 Simpkin, H. and Richards, E.G. (1967) J. Mol. Biol. 29, 349--356 19 Nirenberg, M. and Leder, P. (1964) Science 145, 1399--1407 20 Cerutti , P., Skeda, K. and Witkop, B. (1965) J. Am. Chem. Soc 87, 2505--2506 21 Draper, D.E. and von Hippel, P.H. (1976) In: Molecular Mechanisms in the Control of Gene Expres-
sion (Nierlich, D.P., Rut ter , W.J. and Fox, C.F. eds.) ICN-UCLA symposia on molecular and cellular biology, Vol. 5, pp. 421--426, Academic Press, New York
22 Fox, J.J. and Shugar, D. (1952) Biochim. Biophys. Acta 9, 369
