Extended Data Fig. 1: Scaffold assay for Rho-dependent termination: additional data.

(a) Top, synthetic nucleic-acid scaffolds containing complementary nontemplate and template DNA strands (colours as in Fig. 1b). Bottom, RNA-release data assessing Rho-dependent termination in complexes containing NusG (left), NusG-N (center), or no NusG (right). Asterisks, released RNA products that exhibit increased levels relative to control reactions lacking Rho (lanes 6 and 7 of each subpanel). Termination is detected by an increase in released RNA products relative to control reactions lacking Rho. Assays were performed twice with consistent results. (b) As in (a), except that synthetic nucleic-acid scaffolds contain noncomplementary nontemplate and template DNA strands. The PBS ligands λtR1 rut and dC75 support efficient termination under these conditions (lanes 1 and 2 of each gel panel), but the shorter PBS ligands dC5 and dC15 do not (lanes 3 and 4 of each gel subpanel). It has been shown previously that dC15 and dC5 exhibit lower binding affinities for Rho than dC75, exhibiting half-maximally saturating concentrations 30-fold and 150-fold, respectively, the half-maximally saturating concentrations for dC75²¹. The PBS ligand λtR1 rut supports efficient and immediate termination under these conditions; with λtR1 rut, approximately 100 % of RNA is released before RNA extension (lane 1 of each gel panel). The PBS ligand dC75 supports efficient but less immediate termination under these conditions; with dC75, in the experiments of b, approximately 100% of RNA is released before RNA extension (lane 1 of each gel panel), but, in the experiments of a, approximately 30% of RNA is released before RNA extension, and approximately 70% of RNA is released only after RNA extension by 1 nt or 2 nt (lane 2 of each gel panel). The results indicate that both λtR1 rut and dC75 are effective PBS ligands and that λtR1 rut is a more effective PBS ligand than dC75 (see Extended Data Fig. 2).