Dependence of Initiation Factor IF-2 Activity on Proteins L? and L12 from Escherichia coli 50 S Ribosomes*

SUMMARY The activity of initiation factor IF-2 from coli strain MRE600 has been studied as a function of the presence or absence of the 50 S ribosomal proteins L7 and L12. Three specific aspects of initiation factor IF-2 function were examined: factor IF-2 70 S ribosomes; ribosomal N-forrnyl-methionyl triphosphate catalyzed initiation factor IF-2 in presence of 70 S ribosomes. The results show that the binding of initiation factor IF-2 and of N-formylmethionyl transfer RNA are both reduced 50 to 75% when 50 S particles deficient in L7 and L12 are employed. The 5’-guanosine triphosphate hydrolysis activity is far more drastically reduced with particles deficient in L7 and L12. The latter effect is similar to results found with the 5’-guanosine triphosphate hydrolysis activities associated with elongation factors

From the Department of Biological Chemistry, School of Medicine, University of California, Davis, California 95616 SUMMARY The activity of initiation factor IF-2 from Escherichia coli strain MRE600 has been studied as a function of the presence or absence of the 50 S ribosomal proteins L7 and L12. Three specific aspects of initiation factor IF-2 function were examined: (a) the binding of radioactive initiation factor IF-2 to 70 S ribosomes; (b) the ribosomal binding of N-forrnylmethionyl transfer RNA which is dependent upon initiation factor IF-2; (c) 5'-guanosine triphosphate hydrolysis catalyzed by initiation factor IF-2 in the presence of 70 S ribosomes.
The results show that the binding of initiation factor IF-2 and of N-formylmethionyl transfer RNA are both reduced 50 to 75% when 50 S particles deficient in L7 and L12 are employed.
The 5'-guanosine triphosphate hydrolysis activity is far more drastically reduced with particles deficient in L7 and L12.
The latter effect is similar to results found with the 5'-guanosine triphosphate hydrolysis activities associated with elongation factors EF-G and EF-Tu. The results suggest that proteins L7 and L12 are part of the binding site for initiation factor IF-2, but that they are more intimately involved in the 5'-guanosine triphosphate hydrolysis catalyzed by the factor. Furthermore, the results are consistent with a model in which the binding sites for all three factors, IF-2, EF-G, and EF-Tu, share certain common determinants, including the L7 and L12 proteins.
Three different protein factors required for protein synthesis possess GTPasei activity when they are bound to the 70 S ribosome: elongation factors EF-G (1) and EF-Tu (2), and initiation factor . In all three cases, both the protein factor and the ribosome are required.
The function of EF-G is associated with the translocation step of protein synthesis; those of EF-Tu * This work was supported by grants from the American Cancer Society   The GTPase activities of EF-G and EF-Tu have previously been shown to depend on the presence of two closely related acidic 50 S proteins, namely L7 and L122 (5)(6)(7).
In the case of EF-G not only does its activity depend upon L7/L12, but crosslinking experiments have shown that the factor can be covalently attached to L7/L12 in the 50 S particle (8). Furthermore, it has been demonstrated that EF-G and EF-Tu cannot simultaneously bind to the ribosome (9)(10)(11)(12).
These results strongly suggest that the two factors occupy the same or closely overlapping sites on the 50 S ribosome, and that L7/L12 are constituents of the site(s).
Although the presence of both ribosomal subunits is necessary for the GTPase reaction catalyzed by 14), the specific ribosomal proteins which are involved have not been established. The investigation described here was designed to test the hypothesis that all three factors of protein synthesis which are linked to GTP hydrolysis share a common requirement for proteins L7/L12 for their activity.
We describe here experiments which demonstrate that all the known activities of IF-2 associated with the 70 S ribosome, GTP hydrolysis, fMet-tRNA binding and the binding of IF-2 itself, are dependent upon the presence of proteins L7/L12. Nakamoto and co-workers (15) made a preliminary statement indicative of some of the conclusions documented here. During the preparation of this manuscript similar results were published by Kay et al. (16). Moreover, an article by Lockwood et al. (17)  50 S] particles at 0". These reconstituted particles were then added to appropriate reaction mixtures without further preincubation.
The excess L7/L12 proteins had no effect on the various reactions tested.
The concentrations of the [ &L7/L12 50 S] particles used are indicated in the figure legends.
IF-2 GTPase Assay-"Coupled" IF-2 GTPase activity (dependent on fMet-tRNA binding) was assayed in a 50+1 reaction mixture containing: 50 mM Tris-HCl, pH 7.4; 100 mM NH&I; 5 mM magnesium acetate; 1 mM dithiothreitol; 32 pmoles of purified 30 S ribosomal subunits; 25 PM A-U-G; 35 pmoles of [%]fMet-tRNA (specific activity, 260 Ci per mole) ; 0.4 pg of purified IF-1 (45 pmoles) ; and 10 PM [y-32P]GTP (specific activity, 150-300 cpm per pmole) . The amounts of purified IF-2, 50 S ribosomal subunits and any further additions are indicated in the appropriate figure legends or in the text. After incubation at 30" for 5 min, 40 ~1 of the reaction mixture were removed and added to a 15-ml conical centrifuge tube containing 0.2 ml of 1 MIM potassium phosphate, pH 7.2, and 0.2 ml of 1 M perchloric acid. The subsequent steps of the assay for the release of [32Pi] were carried out as described by Kolakofsky et al. (3). The "uncoupled" GTPase activity (GTP hydrolysis independent of fMet-tRNA binding) was assayed in an identical manner with the exception that A-U-G, [ for 15 min at 37", the reaction was stopped by the addition of 1.5 ml of 5% trichloroacetic acid, and the tubes were heated to 90" for 20 min followed by filtration It was not established whether the remaining material represents residual L7/L12, or other minor 50 S proteins.
The extracted protein fraction contained proteins L7/L12 as well as a contaminant of higher molecular weight (approximately 20% of the total protein) which was not considered to be a major ribosomal protein constituent.
The gel patterns of isolated [ +L7/ L12 50 S] particles resembled that of 50 S subunits, indicating the reassociation of L7/L12 with the deficient particle.
Association of [ -L"//LlZ 60 S] Particles with SO Subunits-The function of L7/L12 previously reported as well as the experiments described here could be explained by the failure of [ -L7/L12 50 S] particles to reassociate with 30 S subunits to form 70 S couples (24). We investigated this possibility by sucrose density gradient analysis and found that 50 S] particles were tested for their activity in the synthesis of polyphenylalanine in the presence of 30 S ribosomes and polyuridylic acid, as described by Nirenberg and Matthaei (25). Results similar to those of Hamel et al. (15) were obtained and are shown in Table I EF-G Catalyzed GTPase Activity-In order to confirm that the inhibition of protein synthesis activity was the result of a deficiency in the GTPase activity of the [ -L7/L12 50 S] particle and to confirm that our [ -L7/L12 50 S] particles were similar to those described in other laboratories, the uncoupled GTPase reaction of EF-G was measured (26). A 5-fold diminution in EF-G catalyzed GTPase activity was observed with [ -L7/L12 50 S] particles as compared to [ +L7/L12 50 S] particles. This result, in agreement with the reports of others (5,7), confirms that the presence of proteins L7/L12 are required for the EF-G catalyzed GTP hydrolysis activity.
Since the [ -L7/L12 50 S] particles used in these experiments were similar to those previously reported with respect both to polyphenylalanine~synthesis and EF-G catalyzed GTP hydrolysis, their activity in reactions associated with IF-2 was studied.
Reactions Associated with IF-i? GTP Hydrolysis-The rate of GTP hydrolysis catalyzed by IF-2 in the presence of all the components required for fMet-tRNA binding (i.e. GTPase coupled to fMet-tRNA binding) was assayed with [ -L7/L12 50 S] and [ +L7/L12 50 S] particles. Details of the reaction conditions are given in the legend to Fig. 1. The results show that at concentrations of the 50 S subunit which were limiting with respect to 30 S subunits, [+L7/L12 50 S] particles catalyze the hydrolysis of GTP four to five times more rapidly than [ -L7/L12 50 S] particles. Furthermore, the hydrolysis of GTP is completely dependent on IF-2.
The large stimulation obtained with L7/L12 shows that these proteins are required to obtain the maximum rate of the coupled GTP hy drolysis reaction catalyzed by IF-2.
The coupled and uncoupled IF-2 catalyzed GTPase reactions of [ -L7/L12 50 S] and [ +L7/L12 50 S] particles were compared as a function of the amount of IF-2 added (Fig. 2). Both the coupled and the uncoupled GTPase reactions were substantially greater with the [ +L7/L12 50 S] particle. Although even the uncoupled IF-2 GTPase activity is stimulated by the addition of L7/L12 proteins to [ -L7/L12 50 S] particles, a much larger stimulation is observed under coupled conditions. fMet-tRNA Binding to 70 S Ribosomes-At limiting concentrations of IF-2, extensive fMet-tRNA binding to 70 S ribosomes requires GTP hydrolysis.
The hydrolysis of GTP appears to stimulate the release of IF-2 from the 70 S initiation complex and thus allows the factor to act catalytically (18,27,28). Since [ -L7/L12 70 S] particles have a greatly reduced GTPase activity catalyzed by IF-2, experiments were carried out to test the effect of L7/L12 on fMet-tRNA binding at limiting concentrations of IF-2.

8557
One standard method for determining fMet-tRNA binding to ribosomes is measurement of the fMet-tRNA retained on Millipore filters due .to the tRNA-ribosome interaction (29). However, this assay does not distinguish between the ,binding of fMet-tRNA to 30 S subunits  The stimulation with L7/L12 proteins occurred both under conditions of GTP hydrolysis and also when GTP hydrolysis was prevented (i.e. with GMP-PCP).
It is therefore clear that the L7/L12 proteins affect more than just the GTP hydrolysis functions of the 70 S ribosome.
In the presence of Gpp(CHz)p the binding of 32P-IF-2 and [3H]fMet-tRNA was approximately equimolar (Table II), in agreement with our previous results (18). Therefore, the binding of both fMet-tRNA and IF-2 is comparably dependent on the presence of L7/L12 proteins.

DISCUSSION
The question posed in the experiments reported here is whether initiation factor IF-2 shares with elongation factors EF-G and  The mixtures were incubated for 5 min at 30", xe volume of 5% glutaraldehyde was added and the mixtures were cooled to 5". Each was then applied to a 7 to 2570 sucrose gradient (4.5 ml) containing: 10 mM Tris-HCl, pH 7.4; 50 mM KCl; 5 mM magnesium acetate; and 2 mM dithiothreitol.
The gradients were centrifuged for 145 min at 4" at 40,000 rpm in a Beckman SW 56 rotor. Approximately 0.16-ml fractions were collected; 0.5 ml of water and 5 ml of Tritontoluene scintillation fluid (1:2) were added, and the samples were counted in a Beckman LS-200 scintillation counter. See Table II for the components used for gradients A, B, C, D.
EF-Tu a requirement for the 50 S ribosomal proteins L7/L12 for full expression of its activity.
Three functional properties of IF-2 involved in the formation of the 70 S initiation complex were examined in the presence or absence of proteins L7/L12: (a) the binding of radioactive IF-2 to the 70 S initiation complex; (b) the binding of fMet-tRNA catalyzed by IF-2; (c) the GTPase activity catalyzed by IF-2.
The results show that the functional activities of IF-2, like the two elongation factors, depend upon the presence of L7/L12.
Since, as previously demonstrated in this (18) and other laboratories (27,28), GTP hydrolysis accompanies the release of IF-2 from the 70 S initiation complex, it might be anticipated that all conditions which blocked GTP hydrolysis would thereby enhance or stabilize the binding of IF-2.
The results reported here (see Fig. 3) are contrary to this interpretation. When GTPase is inhibited by a deficiency in proteins L7/L12, a slight decrease, not an increase, in IF-2 binding is observed.
The result is consistent with a dual function of L7/L12 : (a) as participants in the Obviously if the absence of L7/L12 resulted in a complete loss of IF-2 binding to the 70 S ribosome, it would follow that the other related activities would also be abolished.
However, we find that the absence of L7/L12 results in a reduction in, but not a complete loss of capacity for IF-2 to bind in the 70 S initiation complex.
We estimate that the [ -L7/L12 50 S] particles employed in these experiments are at least 80 ye to 90 To deficient in L7/L12.
Thus, the fact that the inhibition of IF-2 binding is only 50% cannot be explained by the presence of intact 50 S subunits in our preparation of [ -L7/L12 50 S] particles. A more likely explanation is that, although L7/L12 is part of the binding site for IF-2, other proteins are also cooperatively involved in the site, and therefore the depletion of L7/L12 alone does not completely abolish IF-2 binding.
The results with the binding of fMet-tRNA are consistent with the above interpretation.
Again fMet-tRNA binding is only partially reduced in the [ -L7/L12 50 S] particles; moreover, this reduction is of the same order as that for IF-2 binding.
Only when IF-2 is allowed to act catalytically does the amount of fMet-tRNA bound exceed the amount of bound IF-2. By contrast to the results on IF-2 and fMet-tRNA binding, the experiments with IF-2 catalyzed GTPase show a much greater involvement of L7/L12; i.e. IF-2 catalyzed GTPase is inhibited to a far greater extent than either IF-2 binding or fMet-tRNA binding.
The results are consistent with a model in which L7/L12 are considered to play a more direct role in GTP hydrolysis than in the actual binding of IF-2.
In the absence of L7/L12 IF-2 still binds, but with reduced efficiency; however, even though bound, its catalytic function in GTP hydrolysis is significantly impaired.
A nonessential role for the L7/L12 proteins in the GTP hydrolysis reactions of EF-G and EF-Tu has been suggested recently by Hamel and Nakamoto (5). They reported that L7/ L12 deficient particles, low in GTPase activity, were stimulated by making the reaction mixtures 20yo in methanol.
Their results indicate that the L7/L12 proteins do not themselves supply the amino acid residues responsible for the catalytic site of GTP hydrolysis.
Instead the proteins were interpreted to play a role in maintaining a conformation of the 70 S ribosome necessary for efficient GTP hydrolysis and factor binding.
However, Highland et al. (30) using antibodies to single pure 50 S proteins demonstrated that only anti-L7 and anti-112 blocked the binding of EF-G.
This, in addition to the demonstration by Acharya by guest on March 17, 2020 http://www.jbc.org/ Downloaded from et al. (8) that EF-G can be crosslinked to L7/L12, suggests a direct interaction of L7/L12 in the binding of EF-G. The present results show that IF-2, like the two elongation factors EF-G and EF-Tu, shares a requirement for L7/L12 for its GTPase activity.
We can infer that L7/L12 constitutes part of the IF-2 binding site, but is more directly involved in the GTPase activity.
It seems likely that the binding site of IF-2 is identical with or overlaps the EF-G and EF-Tu sites, which may be nearly identical.
Experiments are in progress to determine whether IF-2, like EF-G, can be cross-linked to proteins L7 and L12, and whether EF-G and EF-Tu compete with IF-2 for a common binding site.