Fig. 2: DNA-binding protein roadblocks and NusG affected post-termination sliding and repetitive transcription.

a In this representative recording, the shaded regions indicate pauses at the roadblock during post-termination sliding followed by re-initiation ① or continued sliding past the roadblock to the promoter (recapture) ②. b Heat maps indicate probabilities associated with locations at which post-termination complexes started and stopped sliding under opposing force in buffer without (upper) or with (lower) LacI. Sliding primarily began at the terminator ( ~ -0.4 μm) and ended at promoter ( ~ 0 μm) (green box), unless LacI was present to block sliding and induce terminator-to-LacI binding site (red box) and LacI binding site-to-promoter (blue box) sliding events. c Adding LacI to the buffer increased the probability of roadblocking and re-initiation at the roadblock. d Although NusG diminished the probability of sliding to promoters, it did not change the ratio of re-initiation,  ~ 0.1 (yellow/red). e A minor population of RNAPs, less than 10%, exhibited multiple cycles of transcription. f Sub-saturating levels of NTPs (100 μM and 20 μM) reduced average end-to-end transcription rates, but not pause-free rates, of both initial and re-initiating transcriptions, compared to a saturating NTP level (1000 μM). The red center line denotes the 50th percentile, while the blue box contains the 25th to 75th percentiles of datasets. The black whiskers mark the 5th and 95th percentiles, and values beyond these upper and lower bounds are considered outliers, marked with red crosses. N: number of events; P values from two-sided two sample t-test are shown in figures (red: p < 0.1; black: p > 0.1).