Rho-dependent termination of transcription. I. Identification and characterization of termination sites for transcription from the bacteriophage lambda PR promoter.

We have conducted a detailed investigation of in vitro transcription from the bacteriophage lambda PR promoter in order to examine various aspects of the mechanism of rho-dependent termination. In these studies, we have focused particularly on nucleotide sequence specificity, both at the termini and at potential rho-binding sites on the mRNA, and on the relationships between elongation, pausing, and termination. Rho-terminated transcripts from restriction fragment templates have been analyzed by polyacrylamide gel electrophoresis, and termination efficiencies have been established by densitometry of autoradiographs. Termination sites on the template have been located by comparing the electrophoretic mobilities of terminated transcripts with those of transcripts of known length that have been artificially terminated by the incorporation of 3'-O-methyl nucleotides. We have identified five discrete rho-dependent termination sites located between 290 and 450 base pairs downstream from the lambda PR promoter. These rho-dependent 3'-termini are somewhat heterogeneous in details of sequence and potential RNA secondary structure, but all possess features that appear to be characteristic of RNA polymerase elongation pausing sites (Morgan, W. D., Bear, D. G., and von Hippel, P. H. (1983) J. Biol. Chem. 258, 9565-9574). The efficiency of termination at individual sites ranges from 20 to 70% under the usual in vitro transcription conditions; termination is inhibited by increasing the monovalent salt concentration. Lowering nucleoside triphosphate substrate concentrations increases termination efficiency at some sites located 290 or more base pairs downstream from PR, but does not enhance termination at sites closer to PR. The substitution of inosine for guanosine residues in the transcript, which decreases the stability of the RNA-DNA hybrid and of secondary structure in the nascent mRNA, results in strong rho-dependent termination at several new sites located 100 to 260 base pairs downstream from PR. In Morgan et al. (cited above), data on RNA polymerase elongation pausing as a function of reaction conditions are correlated with these termination results, and a general model for rho-dependent termination is discussed.