RNA polymerase clamp movement aids dissociation from DNA but is not required for RNA release at intrinsic terminators

ABSTRACT

In bacteria, disassembly of elongating transcription complexes (ECs) can occur at intrinsic terminators in a 2-3 nucleotide window after transcription of multiple kilobase pairs of DNA. Intrinsic terminators trigger pausing on weak RNA-DNA hybrids followed by formation of a strong, GC-rich stem-loop in the RNA exit channel of RNA polymerase (RNAP), inactivating nucleotide addition and inducing dissociation of RNA and RNAP from DNA. Although the movements of RNA and DNA during intrinsic termination have been studied extensively leading to multiple models, the effects of RNAP conformational changes remain less well-defined. RNAP contains a clamp domain that closes around the nucleic-acid scaffold during transcription initiation and can be displaced by either swiveling or opening motions. Clamp opening is proposed to promote termination by releasing RNAP-nucleic acid contacts. We developed a cysteine-crosslinking assay to constrain clamp movements and study effects on intrinsic termination. We found that biasing the clamp into different conformations perturbed termination efficiency, but that perturbations were due primarily to changes in elongation rate, not the competing rate at which ECs commit to termination. After commitment, however, inhibiting clamp movements slowed release of DNA but not of RNA from the EC. We also found that restricting trigger-loop movements with the RNAP inhibitor microcin J25 prior to commitment inhibits termination, in agreement with a recently proposed multistate-multipath model of intrinsic termination. Together our results support views that termination commitment and DNA release are separate steps and that RNAP may remain associated with DNA after termination.