The elemental mechanism of transcriptional pausing

We only have a few minor comments:

"the bypass fraction was shifted from ~22% at 10 μM GTP to ~7% at 100 μM GTP". Why is the bypass fraction lower at higher GTP here? This is inconsistent with Figure 1—figure supplement 1C that bypass fraction increases at higher GTP.

We thank the reviewers for catching this error. Bypass indeed increases at higher GTP, as shown in Figure 1G also. The sentence in question contained a typo reversing the GTP concentrations. It is now corrected in the revised manuscript to read "the bypass fraction was shifted from ~22% at 100 µM GTP to ~7% at 10 µM GTP".

"upstream fork-junction (usFJ" should be "upstream fork-junction (usFJ)".

We thank the reviewers for catching this error and have made the suggested correction.

Subsection “Hybrid translocation is not rate-limiting for elemental pause escape”, first paragraph; please change to "(Figures 3D-F)", as Figure 3F is not cited in the manuscript.

We thank the reviewers for catching this error and have made the suggested correction.

Subsection “Hybrid translocation is not rate-limiting for elemental pause escape”, first paragraph: the reference should be (Malinen et al., 2012) instead.

We thank the reviewers for catching this error and have made the suggested correction.

"control EC (Figure 4C; (Malinen et al., 2014)." should be "control EC (Figure 4C; Malinen et al., 2014)."

We thank the reviewers for catching this error and have made the suggested correction.

"(compare to control EC; Figure 4C, D, E)" it shall be Figure 5C, D, E instead?

We thank the reviewers for catching this error and have made the suggested correction.

Subsection “Clamp loosening but not extensive clamp opening accompanies elemental pausing”, first paragraph: "Figure 5D". should be Figure 6D instead.

We thank the reviewers for catching this error and have made the suggested correction.

"apparent H-bonds of R542 in fork loop 2 to the template dC that may hinder its translocation into the active site". Could the authors comment whether R452 can interact with other +1 template bases, is this interaction sequence-specific?

We appreciate this suggestion from the reviewers. The low resolution of the ePEC cryo-EM structure (5.5 Å) makes discussion of details of the Arg-dC interaction speculative (we note this in the Results section). The interaction is potentially specific for the dC–dG as contact to the nontemplate G (H-bonds to O6 and N7) are possible, and this pattern is G-specific. The Arg contact to the WC face of template dC seen in the Murakami crystal structure or an initiation (5voi) also is cytosine-specific. Thus, it is possible that R542 may inhibit translocation by more than one type contact specific to a +1 dC in the template strand. Because of structural uncertainties, we have limited the description of the possible contacts to addition of a single sentence: “Notably the apparent R542 H-bonding contacts to the +1 dC–dG bp as well as to an unpaired but pretranslocated +1 dC in an initiation complex (Murakami et al., 2017) are not feasible for other +1 bases, possibly suggesting multiple ways R542 could inhibit template DNA translocation in an ePEC.”. We thank the reviewers for suggesting this addition, which should help readers better put our results into context. We also have edited the description of the R542 interaction in the results to refer to the dC–dG bp instead of simply to template dC and we have modified Figure 7B to include depiction of the potential H-bond to N7 (2.9 Å in 6bjs).

"15% PAG" should be "15% PAGE".

We appreciate that PAGE (polyacrylamide gel electrophoresis) is the more commonly recognized abbreviation, although in this instance PAG was an abbreviation for polyacrylamide gel not including electrophoresis. Since all other usages of this abbreviation in the manuscript are PAGE, we revised the sentence to now read “RNAs in each quenched reaction sample were separated by PAGE (15%; 19:1 acrylamide:bis-acrylamide)…”