Binding of Met-tRNA, and GTP to Homogeneous Initiation Factor MP

Homogeneous initiation factor MP forms a stable complex with Met-tRNA, which binds to nitrocellulose filters in the absence of ribosomal subunits. Complex formation is rapid at 0” and the rate of reaction is stimulated 20-fold by GTP when freshly prepared initiation factor MP is used. Under optimal assay conditions, a 1:l:l stoichiometry for initiation factor MP, GTP, and Met-tRNA, is indicated, based on a molecular weight for initiation factor MP of 180,000. Kinetic analysis of ternary complex formation suggests an ordered reaction sequence with binding of GTP followed by binding of Met-tRNA,. However, binding of GTP appears to produce an unstable state which leads to rapid inactivation of initiation factor MP in the absence of Met-tRNA,. Formation of a stable binary complex of initiation factor MP and Met-tRNA, occurs in the absence of GTP. The binary complex cannot subsequently bind GTP. While storage of initiation factor MP at 0” for several weeks has no effect on the rate or extent of Met-tRNA, binding in the presence of GTP, the rate of binary complex formation is increased lo-fold. The binary and ternary complexes appear to bind to 40 S ribosomal subunits with equal efficiency. extracts


Homogeneous
initiation factor MP forms a stable complex with Met-tRNA, which binds to nitrocellulose filters in the absence of ribosomal subunits. Complex formation is rapid at 0" and the rate of reaction is stimulated 20-fold by GTP when freshly prepared initiation factor MP is used. Under optimal assay conditions, a 1:l:l stoichiometry for initiation factor MP, GTP, and Met-tRNA, is indicated, based on a molecular weight for initiation factor MP of 180,000. Kinetic analysis of ternary complex formation suggests an ordered reaction sequence with binding of GTP followed by binding of Met-tRNA,.
However, binding of GTP appears to produce an unstable state which leads to rapid inactivation of initiation factor MP in the absence of Met-tRNA,. Formation of a stable binary complex of initiation factor MP and Met-tRNA, occurs in the absence of GTP. The binary complex cannot subsequently bind GTP. While storage of initiation factor MP at 0" for several weeks has no effect on the rate or extent of Met-tRNA, binding in the presence of GTP, the rate of binary complex formation is increased lo-fold. The binary and ternary complexes appear to bind to 40 S ribosomal subunits with equal efficiency.

Several laboratories
have partially purified two distinct eucaryotic protein factors from 0.5 M KC1 ribosomal extracts which appear to be required for initiation complex formation (I-12). IF-Ml,' recently purified to homogeneity (2), can bind initiator Met-tRNA, to the 40'S ribosomal subunit in the presence of template (AUG or mRNA) and IF-M2B (13). In contrast, a second factor which we call IF-MP forms a stable ternary complex with Met-tRNA, and GTP. This ternary complex appears to bind to 40 S ribosomal subunits in the absence of messenger RNA or to nitrocellulose filters in the absence of ribosomal subunits (9)(10)(11)(12)(13)(14)(15)(16)(17)(18). Partially purified fractions containing this factor have also been shown: (a) to reverse inhibition of globin mRNA translation produced by hemin deficiency (19) and double-stranded RNA (19,20); (b) to bind to synthetic poly (A) and natural mRNA (18); and (c) to be required for formation of the 80 S initiation complex (9-12, 17, 21-23). In this report, optimal conditions required for kinetic studies of Met-tRNA, binding by homogeneous IF-MP have been determined (see accompanying paper (24) for purification and characterization of IF-MP). These studies show that the ternary complex, composed of equimolar IF-MP .GTP .Met-tRNA, based on a molecular weight of 180,000 for IF-MP, is formed by an ordered sequential mechanism. In the absence of GTP, however, IF-MP can form a stable binary complex with Met-tRNA, which binds to 40 S ribosomal subunits with equal efficiency as the ternary complex. MgZ+ (data not shown). In contrast, inclusion of MgZ+ in buffer used to wash the nitrocellulose filters after binding of the ternary (or binary) complexes appears to reduce nonspecific  binding  observed  with nonhomogeneous  preparations  of  IF-MP. With IF-MP partially  purified through phosphocellulose chromatography  (13), both GTP-dependent  and GTPindependent Met-tRNA, binding are reduced an equal amount; as a result, dependency of the binding assay on GTP appears to be increased 4-fold ( Fig. 1). Although it is important to distinguish between ternary complex formation and subsequent retention on nitrocellulose filters, with homogeneous IF-MP neither process is affected by Mg2+ concentrations as high as 10 mM.
Under these assay conditions the extent of GTP dependent Met-tRNA, binding increases linearly as a function of IF-MP over the range 0.1 to 1 pg of IF-MP/SO ~1 of binding assay (Fig.  2). Similar results are obtained at 5 mM Mg*+ in the presence Preparation of Unfractionated Reticulocyte Aminoacyl-tRNA-Unfractionated reticulocyte tRNA was acylated with 16 "C-amino-acids and four '2C-amino-acids using crude rabbit reticulocyte aminoacyl-tRNA synthetase prepared as described previously (26). .Met-tRNA, were determined. The requirement for potassium is optimal over a broad concentration range, 40 to 80 mM (data not shown). Precipitation of IF-MP occurred below 20 mM KCl. The pH optimum is 7.15 (data not shown). However, since other assay systems developed for the analysis of protein synthesis initiation (28) show pH optima at 7.5, a pH of 7.5 has been used in this study even though binding efficiency is reduced by approximately 15 to 20%. The concentration of GTP required for maximal extent of binding decreases with increasing IF-MP concentration (data not shown). Since substrate inhibition is not observed, a GTP concentration which saturates at all input levels of IF-MP, 10 FM, is used in all binding assays, Exogenous Mga+ is not required and has little effect on ternary complex formation from 0 to 10 mM  Fig. 3A. At 0", GTP-dependent binding is essentially complete by 1 min. In the absence of GTP, a much slower rate of Met-tRNA, binding is obtained. Although linear for the first 5 min of incubation, the extent of the GTP-independent Met-tRNA, binding is approximately 50 to 60% of the GTP-dependent Met-tRNA, binding. Both the ternary IF-MP*GTP.Met-tRNA, and the binary IF-MP.Met-tRNAr complexes are stable for up to 2 hours at 0" under these assay conditions and no deacylation is detected (data not shown). To measure initial rates for kinetic studies, GTP-dependent Met-tRNA, binding is determined at 15-s incubation at 0", while GTP-independent binding is monitored at 5 min. The maximal extents of GTP-dependent and GTP-independent Met-tRNA, binding are obtained at 5 and 15 min, respectively. While the extent of Met-tRNA, binding at 30 or 37" was similar to that obtained at 0", initial rates of complex formation were too rapid for accurate determination by the nitrocellulose filtration technique used in this study (data not shown).
Titration of IF-MP Binding with Met-tRNA,-Titration of the extent of IF-MP binding with Met-tRNA, in the presence or absence of GTP (Fig. 3B) shows that saturation is achieved at 30 pmol/50-ccl reaction (600 nM). Storage of IF-MP at 0" for 2 weeks or freezing has no effect on the extent (or rate) of Met-tRNA, binding in the presence of GTP; however, both the rate and extent of GTP-independent binding are doubled. As the result, GTP dependence of Met-tRNA, binding decreases. From data presented in Fig. 3A, GTP dependence is also greatly affected by the length of incubation.
In contrast to a 4-fold dependency obtained at 4 min, a 20-fold dependency of Kinetic Parameters of Met-tRNA, Binding by ZF-MP-Initial velocity data of GTP-dependent Met-tRNA, binding by IF-MP were analyzed by double reciprocal plots (Fig. 4, A 6). At the earliest time point (10 s) GTP binding is maximal and quickly decays to a constant value; in the presence of Met-tRNA,, the plateau value was 1.8 to 2.3 pmol above the level observed in the absence of Met-tRNA,.
In contrast, at the earliest time point (10 s) the level of Met-tRNA, bound to IF-MP was minimal.
Met-tRNA, binding quickly reached a plateau value (3.5 pmol) in the presence of GTP, while in the absence of GTP the binding of Met-tRNA, remained linear for the length of the assay. The amount of GTP-dependent Met-tRNA, bound to IF-MP at 3 min was about 2.6 pmol. This level of GTP-dependent Met-tRNA, binding is slightly greater than the level of Met-tRNA,-dependent GTP binding (2.1 pmol), but again is suggestive of a 1:l stoichiometry.
Several other observations can be made from these data ( Since there is no hydrolysis of GTP under the conditions of complex formation (data not shown), all three forms of GTP should bind in the same manner. Therefore, it appears that [8-aH]GTP gives an artificially high value; however, we have not been able to verify this experimentally.
Inactivation of IF-MP-Although IF-MP can bind Met-tRNA, by GTP-dependent and GTP-independent mechanisms, stable GTP binding by IF-MP is strictly Met-tRNA,dependent. As shown in Fig. 7, preincubation of IF-MP with GTP in the absence of Met-tRNAI rapidly inactivates the ability of IF-MP to form a ternary complex upon subsequent addition of Met-tRNA,.
To determine whether such inactivation results from decreased formation of the ternary complex or retention of the complex to nitrocellulose filters, or both, IF-MP was preincubated with GTP in the presence or absence of 40 S ribosomal subunits prepared from salt-washed polysomes (Table II). If IF-MP is preincubated with GTP in the absence of 40 S subunits, inactivation of GTP-dependent Met-tRNA, binding is obtained with subsequent binding conducted in the presence or absence of 40 S subunits (Table II,  Following formation of these complexes as described earlier, aliquots were directly analyzed on nitrocellulose filters to determine the amount of complex formed; 40 S ribosomal subunits were then added to the remainder to determine the extent of complex binding to these. Binding of extent of ternary complex formation. This inhibition is uncompetitive with respect to Met-tRNA, (Fig. 4A) and noncompetitive with respect to GTP (Fig. 4B). While the agent responsible has not been identified, this pattern of inhibition suggests that formation of the ternary complex proceeds by an ordered mechanism (29)(30)(31), with compulsory binding of GTP preceding binding of Met-tRNA,.
It is possible that the differences in the Met-tRNA,s may reside in the source of the tRNA either as species specific (rat uersus rabbit) or commercial supplier. The level of both GTP and Met-tRNA, binding appears to increase proportionally to the amount of co-ligand bound (Fig. 5) (Fig. 8B). The 40 S subunit complex contains equal molar quantities of Met-tRNA, and GTP. Under the conditions used only 10% of the ternary complex is recovered in the 40 S region. In the absence of GTP, 3.3 pmol of binary complex are formed and 0.3 pmol is bound to 40 S subunits (Fig. 8C). Bound Met-tRNA, as assayed on nitrocellulose filters appears to be transferred with equal efficiency (approximately 10%) to 40 S ribosomal subunits whether initially present as ternary complexes with GTP or as a binary complex. Subsequent addition of GTP to IF-MP incubated with Met-tRNA, results in additional binding of 0.12 pmol of Met-tRNA, to 40 S subunits with the total bound, 0.42 pmol, comparable to that obtained with ternary complex formation (Fig. 8D). However, GTP is incorporated only to the extent that additional Met-tRNA, is bound, 0.13 pmol (Fig. 80)   IF-MP .Met-tRNAr binary complex, therefore, does not appear to be converted to a ternary complex containing GTP. Such data would support the compulsory ordered mechanism and 1:l stoichiometry of GTP and Met-tRNA, binding indicated by kinetic studies of ternary complex formation.
It should be noted that no exogenous template (AUG or mRNA) was added in these studies. Addition of AUG to a level of 0.3 A,,, unit per 50 ~1 of reaction mixture shows no additional stimulation of 40 S complex formation as assayed by sucrose gradients (data not shown). DISCUSSION

IF-MP, homogeneous
by criteria presented in the accompanying paper (24), rapidly forms a stable complex with the initiator Met-tRNA, (and to some extent Met-tRNA, also), but not other aminoacyl-tRNA species (Table I). The ternary complex IF-MP .GTP .Met-tRNAr binds to nitrocellulose filters in the absence of ribosomal subunits. Under optimal assay conditions 5 pmol of Met-tRNA, and 4 pmol of GTP are bound per I.cg of IF-MP (Fig. 3B). Based on a molecular weight of contrast to prolonged incubation at 25 or 37" (10, 22), homogeneous IF-MP maximally binds Met-tRNA, in the presence of GTP by 1 to 2 min at 0". Initial rates of binding are obtained only at 15 s or less. Optimal GTP and Met-tRNA, concentrations for ternary complex formation with IF-MP are also 1 to 2 orders of magnitude lower than conditions used by other laboratories (9,10,12,22). Cashion and Stanley (11) report that maximal binding with their IF-l is obtained only when Met-tRNA, is present in a lOOO-fold excess. In contrast, 1.7 c(g of IF-MP (15 pmol of monomer) is saturated with 7.5 pmol of Met-tRNA, out of a total of 30 pmol (25%) binding). Since GTP and Met-tRNA, are required for a maximal extent of binding, and the rate of ternary complex formation is dependent on the concentration of IF-MP, differences between homogeneous IF-MP and other functionally identical Met-tRNA, binding factors may reflect molecular heterogeneity or inhibitory components present in these latter preparations, or both.
The ratio of GTP-dependent to GTP-independent Met-tRNA, binding has been presented in several papers as evidence of extensive purification (9)(10)(11)(12). Data presented here demonstrate that IF-MP can form stable complexes with Met-tRNA, in the presence or absence of GTP (Fig. 3) The formation of stable binary and ternary complexes of IF-MP is summarized in Fig. 9. Ternary complex formation proceeds by the ordered addition of GTP and Met-tRNA,.
The subsequent interaction of ternary complex and 40 S subunits leads to the transfer of equal molar quantities of Met-tRNA, and GTP to the 40 S subunit (Fig. S), a reaction neither stimulated by, nor requiring, exogenous ApUpG (32). A curious feature of this pathway is inactivation of IF-MP by GTP in the absence of Met-tRNA, (Fig. 7), which occurs without GTP hydrolysis. This inactivation is prevented by the presence of 40 S subunits (Table II), suggesting the possibility that in uiuo the first step in this pathway might be the binding of IF-MP to the 40 S subunit.
The first step in the formation of a binary complex appears to be the conversion of IF-MP to IF-MP* (an altered state of the native molecule). This conversion is a slow process ( Based on studies of (a) inactivation of IF-MP by GTP (Fig. 7), (b) binding of radioactive GTP to  IF-MP (Fig. 5), and (c) 40 S.Met-tRNA, complex formation (Fig. 8), it seems likely that IF-MP* does not interact with GTP either as free IF-MP* or as a binary complex. The binary complex of IF-MP* and Met-tRNA, is identical to the ternary complex IF-MP .GTP .Met-tRNA, in terms of effectiveness of transfer of Met-tRNAr to 40 S subunits (Fig. 8). The functional efficiency of the binary and ternary complexes in peptide bond formation is currently being investigated.
The above observations on the binary and ternary complex pathways are suggestive evidence for an allosteric role for GTP. Additional evidence for this is that at no time during the reaction is there any appreciable GTP hydrolysis, a fact supported by the ready substitution of GMP-P(CH,)P for GTP (data not shown; see also Refs. 9 and 22). Attempts to study the function of GTP after the formation of a 40 S .Met-tRNAr .GTP complex have not yet been successful since one of the initiation factors (IF-MBA) required for 80 S complex formation is capable of hydrolyzing GTP in the presence of 40 S and 60 S subunits (33).
The homologous prokaryotic factor IF-2 also forms a stable ternary complex with initiator tRNA @Met-tRNA,) which binds first to the small ribosomal subunit during initiation complex assembly (34)(35)(36)(37). In contrast to IF-MP, 5 mM Mg'; is required for ternary complex formation with IF-2 (38). Although no hydrolysis of GTP occurs during ternary complex formation, IF-2 does catalyze GTP hydrolysis in the presence of both ribosomal subunits and appears to promote joining of the 30 and 50 S ribosomal subunits (34,(38)(39)(40).
In addition to forming a ternary complex with Met-tRNA, and GTP, IF-MP prepared in this laboratory has been shown: (a) to catalyze methionyl-puromycin formation with both natural and artificial mRNAs (13, 24); (b) to bind (as a ternary complex) to the 40 S ribosomal subunit (see Refs. 32 and 41, and "Results"); (c) to reverse hemin-deficient inhibition of translation (19,20); (d) to bind to poly(A), globin mRNA, R17 RNA, and double-stranded RNA (18,20); and (e) to reverse inhibition of translation by double-stranded RNA and oxidized glutathione (41). Whether all of these activities are related to a common mechanism of action of IF-MP during establishment of the initiation complex is not known at present, but such observations may be relevant to the possible regulation of protein synthesis at the translational level.