Termination of transcription by Escherichia coli RNA polymerase: influence of secondary structure of RNA transcripts on rho-independent and rho-dependent termination

Abstract

The effect of RNA secondary structure on rho-independent and rho-dependent termination of transcription of T3 DNA by Escherichia coli RNA polymerase has been studied by incorporating, into nascent transcripts, base analogs that lead to altered base-pairing properties. A guanine --> hypoxanthine substitution, with attendant weakening of secondary structure, abolished the rho-independent termination at 20% of the genome; in contrast, replacement of cytosine with 5-bromocytosine, which forms stronger pairs with guanine, enhanced termination at this site. rho-Independent termination was not altered by replacing uracil with 5-bromouracil. There are two major rho-dependent termination sites on the T3 DNA-at 8 and 15%. The termination activity of rho in this system also depended on RNA secondary structure. The incorporation of 5-bromouracil instead of uracil into RNA did not alter the site specificity of rho action but rho was rendered inactive when cytosine was replaced by 5-bromocytosine. In contrast, replacement of GTP with ITP in the reaction increased rho-dependent inhibition of RNA synthesis, caused production of heterogeneous-sized transcripts, and stimulated rho-mediated ATP hydrolysis. The rho-associated ATPase activity, in the presence of isolated T3 RNA, was also stimulated by inosine substitution. Furthermore, the temperature-sensitive rho isolated from rho 15 mutant of E. coli, which does not terminate transcription in the presence of the common rNTPs, was active when GTP was replaced with ITP. These results suggest that strongly paired G.C-rich regions in RNA stem-loop structures or RNA.DNA hybrids are essential for rho-independent termination, whereas rho-dependent termination requires weakly paired cytosine residues for its action.