Transcription terminators are sequences in DNA that stop the progress of RNA polymerase and thereby reduce the transcription of downstream genes. Antiterminators increase downstream gene expression by modifying polymerase so that it no longer recognizes or responds to terminators. Nascent transcripts encoded by the two cis acting antitermination sites (""""""""put"""""""" sites) of bacteriophage HK022 promote such a modification: an E. coli RNA polymerase molecule that transcribes a put site acquires an increased ability to transcribe through downstream transcription terminators. Other polymerase molecules in the cell are unaffected. Antitermination requires persistent association of nascent put RNA with the polymerase. Characterization of RNAP mutants that specifically block put antitermination helped to identify sites of interaction between put RNA and the elongation complex. These mutations alter a Zinc binding domain (the """"""""Zn finger""""""""), a surface exposed loop of the beta' subunit. The tip of the Zn finger contains two pairs of invariant cysteines that ligate a Zinc atom. The cysteines flank a moderately well conserved segment of 13 amino acids that is rich in basic residues. Replacement of individual residues or deletion of the fingertip decreased put-dependent antitermination. Some mutants, including the deletion, were unable to antiterminate, but others retained residual activity. In some cases the residual activity could be greatly increased by a change in the put sequence. Such allele specificity suggests that the fingertip interacts directly with put RNA, a suggestion supported by the observation that a completely defective mutant was unable to bind put RNA. Surprisingly, a larger deletion that removed the entire finger partially restored antitermination. Our working model is that put RNA interacts with the Zn finger and then with a second site in RNAP. Interaction with the zinc finger opens access to the second site, but amputation of the Zn finger obviates the need for the initial interaction by unmasking the second site. The location of the second site is unknown. We suggest that put RNA reduces termination by increasing the elongation rate. The elongation complex pauses at sequences that are rich in U residues, and such U-rich sequences are essential components of """"""""intrinsic"""""""" terminators. Put suppresses such pausing, and it appears to do so by inhibiting RNAP """"""""backtracking"""""""". When the elongation complex is halted at the end of a U-rich sequence, it moves backwards on the template with consequent disengagement of the 3' RNA end from the active site, and concerted retreat of the RNA:DNA hybrid region from the 3'-end. Backtracking removes the 3' end of the nascent transcript from the active site and exposes upstream phosphodiester bonds to hydrolysis activated by the GreB protein. It also delays resumption of elongation until the 3' end is restored to the active site by forward sliding. We found that the beta'-Y75N mutation, which prevents put RNA binding, enhanced GreB cleavage and also delayed resumption of elongation when RNAP was artificially arrested at a U-rich pause site by a Lac repressor roadblock and restarted by addition of IPTG. We propose that put RNA antiterminates by favoring the posttranslocational state of the elongation complex, the state in which the catalytic site of the enzyme is engaged with the next incoming templated substrate.
