fmet-trnaf: An Intermediate in Initiation Complex Formation

Transcription

1 Proc. Nat. Acad. Sci. USA Vol. 68, No. 12, pp , December 1971 A Complex Between Initiation Factor IF2, Guanosine Triphosphate, and fmet-trnaf: An Intermediate in Initiation Complex Formation (E. coli/protein biosynthesis/30s ribosomal subunits/sephadex filtration/70s ribosomal complex) ARTHUR H. LOCKWOOD, PRASANTA R. CHAKRABORTY, AND UMADAS MAITRA Division of Biology, Department of Developmental Biology and Cancer, Albert Einstein College of Medicine, Bronx, New York Communicated by B. L. Horecker, September 20, 1971 ABSTRACT Evidence is presented that suggests the formation of a complex between polypeptide chain-initiation factor IF 2, GTP, and fmet-trnaf. This complex transfers both fmet-trnaf and GTP to 30S ribosomal subunits in the presence of ApUpG and initiation factor IF 1. The resultant 30S initiation complex reacts with 50S subunits to form a 70S initiation complex. fmet-trnaf in this 70S complex reacts with puromycin to form fmet-puromycin. These results suggest that [IF 2, GTP, fmet-trnafj is an intermediate in the initiation of protein synthesis in Escherichia coli. Polypeptide chain initiation in Escherichia coli involves the binding of fmet-trnaf and mrna to the 30S ribosomal subunit. A 70S initiation complex is formed when a 50S ribosomal subunit joins the 30S complex (1). Initiation complex formation requires the participation of several protein factors, which can be obtained from ribosomes by washing them with a buffer that contains a high concentration of salt. Reports from various laboratories have described the isolation of three such initiation factors. We have previously designated these factors FI, FIII, and FII in keeping with the order of their elution from DEAE-cellulose columns (2). They appear to correspond to factor F 1, F 2, and F 3 of Ochoa's laboratory (3) and to factors A, B, and C described by Revel and coworkers (4).* Several properties of initiation factor IF 2 have been reported (5). (a) IF 2 is required, along with IF 1 and IF 3, for the GTP-dependent formation of a 70S initiation complex that includes mrna, fmet-trnaf, and the 70S ribosome. (b) Formation of the 70S complex proceeds through an intermediate 30S initiation complex. IF 2 is absolutely required for binding of fmet-trnaf to the 30S particle. This binding is stimulated by the addition of IF 1. The role of IF 3 in this step is obscure. (c) IF 2 also catalyzes the hydrolysis of GTP to GDP and Pi in a reaction dependent on both 30S and 50S ribosomal subunits. This hydrolysis is stimulated during formation of the 70S initiation complex (6). On the basis of these and other studies (7), it has been suggested that the function of IF 2 in polypeptide chain initiation is similar to that of factor EF G (translocase) in polypeptide chain elongation. Factor EF G catalyzes the GTPdependent movement of nascent peptidyl-trna from the acceptor site (A-site) to the donor (puromycin-reactive, P * A uniform nomenclature has been adopted by workers in this field as follows: IF 1 (f1 FI, A); IF 2 (f2, FIII, C); IF 3 (f3, FII, B). We will use the new notation in this paper. EF G and EF T were formerly called "G" and "T", respectively. site) (5), It is postulated that IF 2 catalyzes a GTP-dependent translocation of fmet-trnaf from a site on the 30S ribosome to the donor site on the 70S ribosome, concomitant with the addition of the 50S subunit. GTP would be hydrolyzed to provide energy for this movement. Although such a model has attractive features, IF 2 also has properties analogous to those of elongation factor EF T. Factor EF T catalyzes the binding of aminoacyl-trna to ribosomes in the presence of GTP and mrna. This proceeds through the intermediate formation of an EF T-GTP-aminoacyl trna complex (5). EF T factor also catalyzes the hydrolysis of GTP in a reaction dependent on both ribosomal subunits and stimulated by the components necessary for elongation of polypeptide chains (5). Weissbach and coworkers have reported that the initial binding of aminoacyl-trna from EF T-GTP-aminoacyl-tRNA may be to the 30S subunit (8) İn the studies presented below, we show that IF 2 interacts with GTP in the presence of fmet-trnaf. Evidence is presented to indicate that this complex is an intermediate in the binding of both fmet-trnaf and GTP to the 30S subunit. Addition of 50S subunits results in a 70S initiation complex, with fmet-trna positioned in the donor site. We suggest that the role of factor IF 2 in polypeptide initiation is analogous to the role of factor EF T in polypeptide elongation. Recently, Clark and coworkers also reported the interaction of IF 2 with GTP and fmet-trnaf (9). MATERIALS AND METHODS fmet-trnaf was prepared by charging pure trnafmet (obtained from Oak Ridge National Laboratories) with [3H]methionine (2500 cpm/pmol, New England Nuclear Corp.) or unlabeled methionine (from Schwarz BioResearch), in the presence of N'0-formyltetrahydrofolate, with a crude E. coli supernatant protein fraction. Sources of all reagents used in the present studies have been described (10). Preparation of Charged trna, Ribosomes, and Initiation Factors. The preparation of charged trna and salt-washed ribosomes, as well as the purification of initiation factors have been described (10). Initiation factors IF 1 and IF 3 were homogeneous by the criteria of gel electrophoresis with and without sodium dodecyl sulfate. IF 2 was about 90% pure by the same criteri All three factors were free of detectable elongation factor activity and nucleoside triphosphatase activity in the absence of ribosomes. 3122

2 Proc. Nat. Acad. Sci. USA 68 (1971) Preparation of Ribosomal Subunits. Ribosomal subunits were prepared from salt-washed ribosomes (11). Subunits were checked for cross-contamination by analytical sucrose gradient centrifugation and by measurement of poly(u)- dependent polyphenylalanine synthesis. The 30S subunit preparations were virtually free of 50S subunit contamination, whereas 50S subunit preparations contained small quantities (less than 5%) of 30S particles. The final preparations were stored at 50C, at a concentration of A260 units/ml; they were stable for at least 2 weeks. Aliquots of 30S subunits were heat-activated at 500C for 2 min immediately before use (12). Assay of IF 2 Complex Formation. Reaction mixtures (0.125 ml) contained "reaction buffer" [0.05 M Tris * HCl (ph 7.0) M NH4Cl-5 mm Mg(OAc)2-1 mm dithiothreitol]; 0.05 ma y_-32p-labeled nucleoside triphosphate ( cpm/pmol). 6 jg of initiation factor IF 2, and charged trna were added as indicated. Complex formation was assayed by Sephadex G-50 gel filtration as described by Gordon (13): After incubation at 250C for 10 min, 0.1-ml aliquots of each reaction mixture were applied to a 0.5 X 20 cm Sephadex G- 50 column equilibrated at 4VC with reaction buffer. Fractions (0.15 ml) were collected at 40C and counted in Bray's solution (14). Assay offaet-puromycin Formation. 50 nmol of puromycin were added per ml of reaction mixture and the mixtures were incubated 5 min at 370C. Formation of fmet-puromycin was measured by the ethylacetate extraction procedure of Leder and Bursztyn (15). RESULTS Stimulation of Binding of GTP to IF 2 by f1et-trnaf. In the presence of fmet-trnaf, initiation factor IF 2 forms a complex with GTP that can be detected by the appearance of labeled nucleotide in the void volume of a Sephadex G-50 column (Fig. 1). The reaction is specific for fmet-trnaf, since replacement of the fmet-trnaf with Phe-tRNA or I-i 3-2 FIG. 1. Binding of GTP to macromolecular material in the presence of factor IF 2 and fmet-trnaf. Reaction mixtures were prepared and analyzed on Sephadex G-50. Each reaction mixture contained 6Mug of IF 2 and 40 pmol of trna charged with amino acids as indicated below. (0-O), fmet-trnaf; (A-A), MettRNAf; (X-X) Met-tRNAm; (0-0) Phe-tRNA; (U-fl), f-met-trnaf, replace IF 2 with heated IF 2 (2 min, 100'C); (0---(D) omit trna. 0 2 z )0 AI Recognition of fmet-trnaf by IF FIG. 2. Nucleotide specificity for 'ternary complex'. Reaction mixtures containing 6Mgg of IF 2 and 30 pmol of fmet-trnaf were prepared and subsequently analyzed on Sephadex G-50 columns; various nucleotides replaced GTP in the reaction mixture as indicated below. The concentration of each nucleotide added was 0.05 mm. (0-0) [y--3%p]gtp; (X-X) [y-"2p]atp or [y-3 2P]- CTP or [yy-32p]utp. (A-A) [3H]GDP (780 cpm/pmol); (-4) [3H]GDP, omit fmet-trnaf. Met-tRNAm results in an amount of GTP bound to IF 2 equal to that observed in the absence of the initiator trna. The formylation of Met-tRNAf is not absolutely required for complex formation; nonformylated Met-tRNAf stimulates the binding of GTP to IF 2 with about half the efficiency of fmet-trnaf. Omission of IF 2, or its replacement with heatinactivated IF 2, eliminates the peak of excluded GTP (Fig. 1). To demonstrate that the entire GTP molecule is present in the complex, GTP labeled with equimolar amounts of 3H and y_32p was used to measure complex formation. Both 3H and 32p were detected in the complex, in equimolar amounts (data not shown). These observations indicate that GTP is bound to high molecular weight material (presumably IF 2) in a reaction requiring IF 2 and fmet-trnaf. Since the Sephadex G-50 procedure does not resolve free fmet-trnaf from complexed material, these experiments (Fig. 1) do not directly demonstrate the presence of fmettrnaf in a [IF 2, GTP, fmet-trnaf] complex. Therefore, we attempted to detect such a complex by analysis on Sephadex G-100, where complexed fmet-trnaf would be separated from unreacted fmet-trnaf. Such an approach has been effective in demonstrating the binding of aminoacyl-trna to T factor and GTP (5). However, when a reaction mixture prepared as described in the legend to Fig. 1 was analyzed on Sephadex G-100, we failed to detect fmet-trnaf in the excluded volume. Also, the amount of GTP excluded (1.2 pmol) was the same as the amount of GTP excluded from reaction mixtures lacking fmet-trnaf. We interpret these results as an indication that a ternary complex is formed between IF 2, GTP, and fmet-trnaf, but that the equilibrium strongly favors the dissociation. Therefore, the continued presence of the intiator trna would be necessary to maintain the integrity of the ternary complex. Influence of Various Nucleotides and Components of Initiation on the Formation of the IF 2 Complex. GTP is the only nucleoside triphosphate that binds to IF 2 in the presence of fmet-trnaf (Fig. 2). ATP, CTP, and UTP do not bind. GDP is bound to IF 2, but with much less efficiency than GTP

3 3124 Biochemistry: Lockwood et al. (9 a- H rl _Q0. I FIG. 3. Binding of GTP to initiation factor IF 2. Each mixture contained initiation factors as follows (0-O), no addition; (-- -0) 6Ag IF2; (A-A)6AgIF1; (X- X),5,gIF3; (o- ) 6,ug IF 2, replace [ y-32p] GTP with [I H] GDP. (Fig. 2). Furthermore, this binding is not dependent on the presence of fmet-trnaf. The amount of GTP bound to IF 2 in the presence of fmettrna, is not influenced by the addition of IF 1, IF 3, or ApUpG codon to the reaction mixture, separately or in combination (data not shown). f~~~~~ c 0.4 / 0.24 Minutes FIG. 4. Interaction of isolated [IF 2, GTP, fmet-trnaf] with 30S ribosomal subunits. Two large-scale reaction mixtures (1.25 ml each) were prepared. Each contained reaction buffer, 200 pmole of ['H]fMet-tRNAf (1800 cpm/pmol), and 60 Mg of IF 2. In one case 0.05 mm ['y32p] GTP (400 cpm/pmol) was also added. After incubation at 250C for 10 min, each mixture was filtered through a 1.5 X 30 cm Sephadex G-50 column. Equal volumes (2.0 ml) of the excluded material (containing IF 2, about 60 pmol of GTP, and 200 pmol of ['H]fMet-tRNAf) and a solution containing reaction buffer, 10 A260 units 30S subunits, 0.45 A20o units of ApUpG, and 10 ug of IF 1 per ml were combined at 0 C. 0.5-ml aliquots were removed at various time intervals, diluted 20-fold, and assayed for binding by Millipore filtration (20). Values obtained in the absence of 30S subunits averaged 0.01 pmol, and have been subtracted. *-_, complete ternary complex; X-X, IF 2, fmet-trna excluded from Sephadex G-50; A-A, same as X-X, plus 60 pmol of GTP at 0 min; O-O, same as X-X, plus 10 nmol of GTP at 0 time. Interaction of IF 2 with GTP and GDP in the Absence of trna. Initiation factor IF 2 also interacts with GTP in the absence of fmet-trnai, although the extent of binding is much less (6 to 7-fold) than in the presence of fmet-trnaf (Fig. 3). As noted above, IF 2 also binds GDP, but with less efficiency than GTP. Neither IF 1 nor IF 3 binds GTP or GDP. Although not shown, other nucleoside triphosphates do not react with IF 2 to yield detectable complexes. Formation of a binary complex between IF 2 and GTP has been reported 16,17). Interaction of [IF 2, GTP, fmet-trnaf] with the SOS Ribosomal Subunit. Since formation of a 30S initiation complex is an obligatory intermediate step in the formation of 70S initiation complex (1), we have examined the interaction of the [IF 2, GTP, fmet-trnaf ] complex with the 30S ribosomal subunit. To prepare the IF 2 complexes, initiation factor IF 2 was incubated for 10 min at 250C with [3H]fMet-tRNAf in the presence and absence of [y-82p]gtp. Each reaction mixture was then passed through a Sephadex G-50 column. The peak of radioactivity in the excluded volume was collected. Aliquots of this material were incubated at 0C with 30S ribosomal subunits, ApUpG, and initiation factor IF 1. The rate of transfer of fmet-trnaf from the "ternary complex" to the 30S ribosome was measured by the retention of fmet-trnaf on Millipore filters (Fig. 4). Only the radioactivity from [3H]- fmet-trnaf that had been transferred to ribosomes would be retained on the filter. For greater clarity in describing the following experiments, species enclosed in brackets refer to those present in the excluded volume from Sephadex G-50. The complex [IF 2, [_-32P]GTP, [3H]fMet-tRNAf] rapidly transfers [3H]fMet-tRNAf to 30S ribosomal subunits in the presence of ApUpG and IF 1 (Fig. 4). In contrast, no binding of [3H]fMet-tRNAf in excess of the control value is observed when [IF 2, [3H]fMet-tRNAfI] is incubated with 30S subunits. Furthermere, addition at zero time of an amount of GTP (60 pmol) equal to that present in [IF 2, GTP, [3H]fMet-tRNAf] is also insufficient to promote extensive transfer of [3]fMettRNA, from [IF 2, [3H]fMet-tRNAf] to 30S subunits. When a larger amount of GTP (10 nmol) t is added at zero time, substantial transfer of fmet-trnaf from [IF 2, [3H ]- fmet-trnaf] to the 30S subunit is observed. Under these conditions, the rate of transfer is much slower than that observed with [IF 2, GTP, [8H ]fmet-trnaf ]. Although not shown in Fig. 4, binding of fmet-trnaf to 30S particles from the "ternary complex" is almost completely dependent on the presence of ApUpG and is decreased about 40% by the omission of IF 1. These results suggest that the ternary complex [IF 2, GTP, fmet-trnaf] is a precursor for the transfer of fmet-trnaf to the 30S initiation complex. During formation of the 30S initiation Proc. Nat. Acad. Sci. USA 68 (1971) complex from its separate components, GTP as well as fmet-trnaf is bound to the 30S subunit (18). Therefore, we asked whether the GTP in the "ternary complex" also was transferred to the 30S subunit. In the experiments presented above, [3H]fMettRNAf was retained on Millipore filters in the presence of 30S subunits, but [y-32p]gtp was not retained. It was possible that [7y-32P]GTP is loosely bound to the 30S subunit and t An amount of GTP just sufficient to sustain the same extent of fmet-trnaf binding to 30S subunits as observed with [IF 2, GTP, fmet-trnaf].

4 Proc. Nat. Acad. Sci. USA 68 (1971) unstable to Millipore filtration. We therefore investigated this problem by the use of Sephadex G-100 analysis to detect GTP binding to the 30S subunit. Isolated [IF 2, ['y-32p]gtp, [3H]fMet-tRNAf] was incubated with 30S subunits, ApUpG, and IF 1. The reaction mixture was then analyzed on a Sephadex G-100 column. Fig. 5 shows that GTP, as well as filaet-trnaf, from the "ternary complex" is transferred to the 30S subunit. Furthermore, fmet-trnaf and GTP are present on the 30S ribosome in about equimolar amounts. This result suggests that [IF 2, GTP, fmet-trnaf] is a precursor for the binding of GTP to the 30S initiation complex. I Participation of [IF 2, GTP, fmet-trnaf ] in Formation of a Functional 70S Complex. Since [IF 2, GTP, fmet-trnaf) is capable of causing the binding fmet-trna, and GTP to the 30S ribosome in a reaction requiring ApUpG, it was of interest to determine whether the 30S complex formed in this manner was a precursor of a functional (puromycin-reactive) 70S complex. In this experiment (Table 1), the 30S complex formed after incubation of the "ternary complex" with 30S subunits, ApUpG, and IF 1 was isolated on Sephadex G-100. When 50S ribosomal subunits were added, all the fmettrnaf initially bound to 30S subunits became reactive with puromycin. This indicates that fmet-trnaf had been so positioned in the donor site of the 70S initiation complex that it could donate fmet into peptide linkage. DISCUSSION Initiation factor IF 2 is required for the binding of fmet-trnaf to 30S ribosomal subunits in the presence of GTP and ApUpG. IF 1 stimulates this binding about 2-fold, while IF 3 is not required (11). We have now observed that initiation factor IF 2 forms a complex with GTP in the presence of fmet-trnaf that can be detected by Sephadex G-50 gel filtration. Since unreacted fmet-trnaf eluted from Sephadex G-50 in the same position (excluded material) as the presumed "ternary complex", Sephadex G-50 analysis did not prove that fmettrnaf, itself, is present in a complex with GTP and IF 2. We attempted to directly demonstrate the presence of fmettrnaf in such a complex by Sephadex G-100 analysis, a technique that allows separation of complexed and free fmettrnaf. In this case we observed no complex formation. This suggests to us that the reaction between IF 2, GTP, and flmettrnaf to form a [IF 2, GTP, fmet-trnaf] complex may in fact occur, but that the equilibrium of the reaction favors the dissociated components; once unreacted fmet-trnaf is separated from the "ternary complex" in Sephadex G-100 columns, the complex would readily dissociate to its constituents. This rapid dissociation may also explain the need for large quantities of IF 2 to detect complex formation. In most of our experiments, we have used 6 ug of factor IF 2 (equivalent to 68 pmol of factor if we assume that IF 2 is 90% pure and its molecular weight is 80,000). The amount of GTP bound in the "ternary complex" in the presence of fmet-trnaf (6 to 8 pmol) was an amount far less than expected for a stoichiometric interaction between factor IF 2 and GTP. Even in react The possibility remains that GTP might first dissociate from the "ternary complex", and then reassociate with the 30S subunit. To exclude this case, the transfer reaction should, ideally, be performed in the presence of exogenous, unlabeled GTP. However, such an experiment could not be performed since GTP bound to the 30S subunit was exchangeable with exogenous, unlabled GTP. 0.4 )02 Recognition of fmet-trnaf by IF FRACTION NUMBER FIG. 5. Gel filtration analysis of 30S initiation complex formed by reaction of [IF 2, [y-"2p]gtp, [3H]fMet-tRNAf] with 30 subunits. [IF 2, [_yj 2P] GTP, [3H]fMet-tRNA] was isolated as in Fig. 4, except that the specific activity of [y-y2p]gtp was 2500 cpm/ pmol. Aliquots (0.6 ml) of isolated complex and of the solution of 30S subunits, IF 1, and ApUpG were combined as described in Fig. 4. After 10 min at 0WC, the reaction mixture was analyzed on a Sephadex G-100 column. (0-4) [3H]fMet-tRNAf; (0-0) [7-8 2P] GTP. tion mixtures where fmet-trnaf was present in large excess and IF 2 was limiting, not more than pmol of GTP was bound to macromolecular material per pmol of IF 2 added (data not shown). The possibility also ekists that the actual species involved in complex formation is a complex of IF 2 with another factor, analogous to the TsTu complex that is active in polypeptide elongation. This might account for the inability of 75% of our factor preparation to participate in complex formation. Unformylated Met-tRNAf also appears to complex with IF 2 and GTP, although less efficiently than fmet-trnaf. Further work is necessary to clarify this lack of specificity. The possibility that an [IF 2, GTP, fmet-trnaf I complex is indeed formed as a first intermediate is strengthened substantially by our observations that isolated [IF 2, GTP, fmettrnaf] complex transfers fmet-trnaf to the 30S ribosome in the presence of IF 1 and ApUpG more rapidly than when the TABLE 1. Properties of the 30S initiation complex formed via [IF 2, GTPfMet-tRNAf ]: formation offmet-puromycin upon interaction with 50S subunits [3H]fMet-tRNAf [3H]fMet-puromycin Bound formed Additions (pmol) (pmol) 30S complex S complex + 50S subunits (IF2, GTP, [3 H]fMet-tRNAf) was prepared and isolated as in Fig ml (containing 150 pmol of [3H]fMet-tRNAf) was added to an equal volume of the 30S subunit, ApUpG, and IF 1 solution described in Fig. 4. After 10 min at 0C, the mixture was filtered through a column of Sephadex G-100 to isolate the 30S initiation complex free of unbound [3H]fMet-tRNAf. To 0.5-ml aliquots of this excluded material (2.6 ml), 50S subunits (5 A260 units) and 50 nmol of puromycin per ml were added (where indicated). Binding of fmet-trnaf to ribosomes was measured by Millipore filtration; fmet-puromycin formation was assayed as described in Methods.

5 3126 Biochemistry: Lockwood et al. separate components are incubated with 30S ribosomes, ApUpG, and IF 1. In fact, incubation of [IF 2, fmet-trnaf] with an amount of GTP equal to that present in the isolated "ternary complex" does not result in binding of fmet-trna, to 30S ribosomal particles. The 30S initiation complex formed via [IF 2, GTPfMet-tRNAf] is a precursor of the 70S initiation complex, in that fmet-trnaf bound in the complex reactswith puromycin to form fmet-puromycin upon the addition of 50S subunits. These observations are consistent with the scheme shown below for the function of IF 2 in 30S initiation complex formation: IF 2 + GTP + fmet-trnaf T. [IF 2, GTP, fmet-trnaf] (1) [IF 2, GTP, fmet-trnaf] + 30S + ApUpG + IF 1-30S initiation complex [IF 1, IF 2, 30S, GTP, fmet-trnaf, ApUpG ] (2) There is evidence for the presence of IF 1 in the 30S complex (19). Reaction (1) might proceed via the intermediate formation of a IF 2-GTP binary complex, which would then react with trnaf to form the ternary complex. Evidence for formation of such a binary complex has been documented before and confirmed in the present communication (16, 17). However, the extent of GTP binding to initiation factor IF 2 is very small ( pmol of GTP bound per pmol of IF 2). The possibility exists that such binding of GTP in the absence of fmet-trnaf is due to the presence of a contaminating GTP-binding protein in the factor preparation and that formation of the GTP-protein complex observed is unrelated to initiation of protein synthesis. We have reported the presence, in initiation factor preparations, of a GTPbinding protein that is unrelated to any of the three initiation factors (2). We have presented evidence that shows that [IF 2, GTP, fmet-trnaf] transfers GTP, as well as fmet-trnaf, to 30S ribosomes. This result is in accord with the recent observation of Thach that, under conditions of 30S initiation complex formation, GTP, as well as fmet-trnaf, becomes bound to the 30S subunit (18). The ability of IF 2 to complex with GTP and fmet-trnaf, and of the resulting "ternary complex" to transfer fmettrnaf and GTP to the 30S ribosome, is analogous to the function of elongation factor EF T, which promotes binding of aminoacyl-trna to ribosomes through the formation of an Proc. Nat. Acad. Sci. USA 68 (1971) intermediate EF T-GTP-aminoacyl-tRNA complex (5). The analogy is particularly striking in view of a recent report that EF T-GTP-aminoacyl-tRNA also interacts with the 30S ribosomal subunit (8). NOTE ADDED IN PROOF After this work was communicated, Groner and Revel (21) reported isolation of an "IF 2. IF 3" fraction that binds GTF and fmet-trnaf. The relation between their "IF 2- IF 3" fraction and our factor IF 2 remains to be elucidated. This research was supported by grants from The National Institutes of Health, the American Heart Association, and the Life Insurance Medical Research Fund. A. H. L. is a predoctoral trainee of the NIH; U. M. is an established investigator of the American Heart Association. 1. Evidence for this mechanism has been reviewed by Lengyel, P., and D. S611, Bacteriol. Rev., 33, 264 (1969). 2. Dubnoff, J. S., and U. Maitra, Cold Spring Harbor Symp. Quant. Biol., 34, 301 (1969). 3. Iwasaki, H., S. Sabol, A. J. Wahba, and S. Ochoa, Arch. Biochem. Biophys., 125, 542 (1968). 4. Revel, M., G. Brawerman, J. C. Lelong, and F. Gros, Nature, 219, 1016 (1968). 5. Cold Spring Harbor Symp. Quant. Biol., 34 (1969). 6. Kolakofsky, D., K. F. Dewy, J. W. B. Hershey, and R. E. Thach, Proc. Nat. Acad. Sci. USA, 61, 1066 (1968). 7. Kolakofsky, D., T. Ohta, and R. E. Thach, Nature, 220, 244 (1968). 8. Brot. N, B. Redfield, and H. Weissbach, Biochem. Biophys. Res. Commun., 41, 1388 (1970). 9. Rudland, P. S., W. A. Whybrow, and B. F. C. Clark, Nature, 231, 76 (1971). 10. Dubnoff, J., and U. Maitra, Methods Enzymol., 20, 219 (1971). 11. Dubnoff, J. S., Ph.D. thesis, Albert Einstein College of Medicine, Yeshiva University (1971). 12. Grunberg-Manago, M., B. F. C. Clark, M. Revel, P. S. Rudland, and J. Dondon, J. Mol. Biol., 40, 33 (1969). 13. Gordon, J., Proc. Nat. Acad. Sci. USA, 58, 1574 (1967). 14. Bray, G. A., Anal. Biochem., 1, 279 (1960). 15. Leder, P., and H. Bursztyn, Biochem. Biophys. Res. Commun., 25, 233 (1966). 16. Mazumder, R., Y. B. Chae, and S. Ochoa, Proc. Nat. Acad. Sci. USA, 63, 98 (1969). 17. Lelong J. C., M. Grunberg-Manago, J. Dondon, D. Gros, and F. Gros, Nature, 226, 505 (1970). 18. Thach, S. S., and R. E. Thach, Nature, 229, 219 (1971). 19. Thach, R. E., J. W. B. Hershey, D. Kolakofsky, K. F. Dewey and E. Renold-O'Donnell, Cold Spring Harbor Symp. Quant. Biol., 34, 277 (1969). 20. Nirenberg, M., and P. Leder, Science, 145, 1399 (1964). 21. Grouer, Y. and M. Revel, Eur. J. Biochem. 22, 144 (1971).