PHAGE f2 RNA-DIRECTED BINDING OF FORMYLMETHIONYL-TRNA TO RIBOSOMES AND THE ROLE OF 30 S RIBOSOMAL SUBUNITS IN INITIATION OF PROTEIN SYNTHESIS

A number of studies indicate that N-formylmethionyl-tRNA (F-met-tRNAF)' is involved in initiation of protein synthesis in E. coli, and that the triplet for Fmet-tRNAF, AUG (and possibly GUG and others), is the signal for chain initiation.2-7 The most widely entertained hypothesis of the mechanism of chain initiation is that 70S ribosomes bind with a high affinity to the initiator codon in mRNA, and that the resultant ribosome mRNA complex binds the F-met-tRNAF selectively either to the peptidyl site8 or to the aminoacyl site9 on the ribosome. Some experiments using synthetic short oligonucleotides as messengers have supported this model.7' 10 However, no good evidence for it has been obtained with natural mRNA. Since RNA from RNA phage, a natural mRNA, directs synthesis of phage coat protein in a cell-free system1' and this in vitro product has formylmethionine at the N-terminal end,4' 5 we would expect the phage RNA to stimulate the binding of F-met-tRNAF to the ribosome by virtue of the presence of an initiator codon in the phage RNA molecule. Our experiments using the conventional ribosome preparation, consisting mainly of 70S ribosomes, failed to show such specific binding. However, as will be reported in this paper, we have found that F-met-tRNAF binding directed by phage RNA messenger takes place on 30S ribosomal subunits (30S particles') in the absence of the 50S subunits. 50S particles inhibit this specific binding of F-met-tRNAF, although they stimulate the binding of other aminoacyl-tRNA directed by their respective codons. We propose that the first step in protein synthesis is formation of an initiation complex consisting of the 30S particle, mRNA, and F-met-tRNAF. The 50S particle next joins this complex, followed by the binding of a second aminoacyl-tRNA and subsequent peptide bond formation. The present findings give a reasonable explanation for the existence of two basic subunits, the 30S and 50S subunits, in the ribosomes. Materials and Methods.-Ribosomes were prepared from E. coli strain Q13 and the 30S and 50S particles were purified by repeated sucrose gradient centrifugations as described previously.'2'13 Both preparations were almost completely pure, when judged by sucrose gradient sedimentation analysis. When they were analyzed for their functional ability in poly U-directed polyphenylalanine synthesis, the 30S particle preparation proved to be completely free from the 50S particle, but the 50S particle preparation was found to contain some particles (about 3-6%) which have the 30S function. Both ribosomal particle preparations were suspended in TMA buffer (10-2 M Tris-HCl, pH 7.8, 10-2 M MgCl2, 3 X 10-2M NH4Cl, 6 X 10-3 M mercaptoethanol). "Washed" 30S particles used in some experiments (Table 1, expt. 3) were prepared as follows. The 30S particles purified as above were suspended in TMA buffer containing 1 M NH4C1 and kept for several hours in the cold. The particles were recovered by centrifugation at 65,000 rpm for 3 hr. The pellet ("washed" 30S particles) was suspended in TMA and dialyzed against TMA. "Initiation factor"'44-17 was prepared as described by Ghosh et al.'8 RNA phage f2 was grown and purified according to the method described by Loeb and Zinder.19 RNA was prepared from the phage f2 by phenol extraction followed by alcohol precipitation and