Twenty-one RNA minihelices, resembling the coaxially stacked acceptor- /T-stems and T-loop found along the top of a tRNA's three-dimensional structure, were synthesized and used as substrates for ATP/CTP:tRNA nucleotidyltransferases from Escherichia coli and Saccharomyces cerevisiae. The sequence of nucleotides in the loop varied at positions corresponding to residues 56, 57 and 58 in the T-loop of a tRNA. All minihelices were substrates for both enzymes, and the identity of bases in the loop affected the interaction. In general, RNAs with purines in the loop were better substrates than those with pyrimidines, although no single base identity absolutely determined the effectiveness of the RNA as substrate. RNAs lacking bases near the 5'-end were good substrates for the E. coli enzyme, but were poor substrates for that from yeast. The apparent Km values for selected minihelices were 2-3 times that for natural tRNA, and values for apparent Vmax were lowered 5-10-fold.

Recognition of tRNA and tRNA-like substrates by the enzyme ATP/CTP:tRNA nucleotidyltransferase requires chemically intact nucleotides within the T-loop, especially at positions 57 and 58, which are invariant purines among naturally occurring tRNAs. To test the effects of base substitutions at these positions, which are distant from the site of catalysis, we synthesized mutant tRNA(Glu) molecules. These in vitro-synthesized RNAs also contained an extra 33 bases at the 5' end and lacked post-transcriptionally modified bases. The precursor tRNAs were used as substrates for nucleotidyltransferases from Escherichia coli and yeast. Substitution of cytidines at either position 57 or 58 had dramatic inhibitory effects on recognition by both enzymes, including raising the apparent Km and lowering the apparent Vmax.; substitution of an adenosine at position 57 or a uridine at position 58 inhibited the reaction only slightly by comparison. Our results demonstrate that the identities of nucleotides at positions 57 and 58 are relevant to recognition by nucleotidyltransferase, and that a purine is required at position 57. The extra bases at the 5' end and the lack of post-transcriptionally modified bases did not substantially inhibit interaction with the enzyme, as judged by the wild-type precursor tRNA(Glu) acting as an effective substrate for both enzymes in the presence of equal concentrations of appropriate tRNA substrates isolated from E. coli.