Molecular Microbiology (1992) 6(9). 1219-1229 Translation initiation in Escherichia coii: sequences within the ribosome-binding site Steven Ringquist, Sidney Shinedling, Doug Barrick,^ Louis Green, Jonathan Binkley, Gary D. Stormo and Larry Gold* Department of Molecular, Cellular and Developmental Biology, University of Colorado, Boulder, Colorado 80309, USA. Summary The translational roles of the Shine-Dalgarno sequence, the initiation codon, the space between them, and the second codon have been studied. The Shine-Dalgarno sequence UAAGGAGG initiated translation roughly four times more efficiently than did the shorter AAGGA sequence. Each Shine-Dal- garno sequence required a minimum distance to the initiation codon in order to drive translation; spacing, however, could be rather long. Initiation at AUG was more efficient than at GUG or UUG at each spacing examined; initiation at GUG was only slightly better than UUG. Translation was also affected by residues 3' to the initiation codon. The second codon can influ- ence the rate of initiation, with the magnitude depend- ing on the initiation codon. The data are consistent with a simple kinetic model in which a variety of rate constants contribute to the process of translation initiation. Introduction In the two decades since the elegant experiments of Steitz defined an Escherichia co//translation initiation site (Steitz, 1969; Steitz and Jakes, 1975; Steitz and Steege, 1977), many studies have sought to clarify the roles of the individual mRNA elements in setting translational yields. Investigations of translation initiation usually focus on the Shine-Dalgarno nucleotides 5' to the initiation codon (Shine and Dalgarno, 1974; Steitz, 1979), the choice of initiation codon amongst AUG, GUG, UUG, CUG, AUA. or AUU (Hartz etai, 1991; Munson etai, 1984; Stormo ef ai, 1982; Shinedling etai, 1987), other nucleotides within the roughly 34 residues that comprise the precise ribo- Received 6 December, 1991; revised 31 January. 1992; accepted 3 February, 1992. fPresent address: Department of Biochemistry. Stanford University, Stanford, California 94305, USA, *For correspondence, Tel. (301) 492 7864; Fax (301) 492 7744. some recognition site (from -20 to +13, when the first nucleotide of the initiation codon is designated 0; Scherer etai, 1980; Stormo etai, 1982), secondary structures of mRNAs that might hinder ribosome binding (de Smit and van Duin, 1990a, b; Blasi ef ai, 1989; Iserentant and Fiers, 1980; Hall ef ai, 1982; Kastelein ef ai, 1983; Looman etai, 1986; McPheeters etai, 1986; Schmidt ef al., 1987), and translational enhancer sequences (Olins and Rangwala, 1989; Schauder and McCarthy, 1989; Thanaraj and Pandit, 1989; Petersen etai, 1988; Boni ef ai, 1991). Reviews of translation initiation have occurred with regularity (Hartz, et ai, 1990; McCarthy and Gualerzi, 1990; de Smit and van Duin, 1990a; Gold, 1985;1988;Gualerzr efa/., 1988; Hershey, 1987; Gold, ef ai, 1981). In the present work we have constructed and analysed a set of translation initiation regions (TIRs) that specifi- cally test the role of the Shine-Dalgarno nucleotides, the three most abundant initiation codons, the space between them, and two abundant second codons on translation initiation. By intent our constructs seek to minimize intramolecular secondary structures. The data fit a simple kinetic model for initiation (McCarthy and Gualerzi, 1990; Gold and Stormo, 1990), and allow comparison with a more thoroughly studied kinetic model for transcriptional initiation (McClure, 1985). The data lead us to consider the untested variables associated with other mRNA ele- ments, mRNA structure, and translational repressors. Results All constructs utilize the same promoter, 5' mRNA sequence, and reporter gene (Fig. 1). The constructs con- tained as variable elements two different Shine-Dalgarno sequences designated as SO (UAAGGAGG) or sd (AAGGA), three different initiation codons (AUG, GUG, or UUG), a variable spacing of adenine residues between them, and three different second codons (CAG, AAA, or GCU). The data from the various ribosome-binding sites (RBSs) are summarized in Tables 1, 2, and 3. Expression of lacZ from each construct is listed relative to that obtained from SD8 in Tables 1 and 2 or to the appropriate initiation codon containing construct in Table 3. The data have been plotted in a variety of ways in Figs 2 to 6 in order to illustrate the relationship between the different elements within each RBS.