Pre-termination Transcription Complex: Structure and Function.

STAR*Methods Resource Availability Lead Contact Further information and requests for reagents and resources should be directed to and will be fulfilled by the Lead Contact, Dr. Evgeny Nudler ( evgeny.nudler@nyulangone.org ).

Materials Availability

Plasmids generated in this study are available upon request from the Lead Contact with a completed Material Transfer Agreement.

Data and Code Availability

The cryo-EM maps included in this study have been deposited in the Electron Microscopy Data Bank with accession codes: EMD-22114 and EMD-22115. The atomic coordinates have been deposited in the Protein Data Bank with accession codes: 6XAS and 6XAV.

Method Details Protein Expression and Purification

To purify wild-type E. coli RNAP, E. coli strain BL21 (DE3) was transformed with pVS10 ( Svetlov and Artsimovitch, 2015 ), and recombinant protein expression was auto-induced ( Studier, 2005 ). After 16 h at 30°C, cells were harvested by centrifugation (4,000g for 10 min at room temperature) and pellets were stored at −80°C. Cell pellets were resuspended in lysis buffer (50 mM Tris-HCl, pH 8.0, 5% (v/v) glycerol, 0.5 mM β-mercaptoethanol, 1M NaCl) supplemented with complete, EDTA-free protease inhibitor cocktail tablets (Roche Applied Science) and lysed using sonication (5-s pulses with 10-s intervals for 10 min on ice). The cell lysate was clarified by centrifugation (30,000g for 40 min at 4°C) to remove insoluble debris. The supernatant was applied to a HisTrap column (GE Healthcare) equilibrated in HisTrap Buffer A (50 mM Tris-HCl, pH 8.0, 5% (v/v) glycerol, 0.5 mM β-mercaptoethanol, 500 mM NaCl, 10 mM imidazole). The column was washed with 20 column volumes (CV) of HisTrap Buffer A. Protein was eluted with HisTrap Buffer B (50 mM Tris-HCl, pH 8.0, 5% (v/v) glycerol, 0.5 mM β-mercaptoethanol, 250 mM NaCl, 250 mM imidazole). Fractions containing recombinant RNAP eluted from the HisTrap column were diluted 5 times in Hep A buffer (50 mM Tris-HCl, pH 8.0, 5% (v/v) glycerol, 20 mM NaCl) and applied to a Heparin column (GE Healthcare) equilibrated in Hep A buffer. Protein was eluted using a linear NaCl gradient (0.02 to 1.5 M NaCl) in Hep B buffer (50 mM Tris-HCl, pH 8.0, 5% (v/v) glycerol). The RNAP-containing peak fractions were pooled, diluted 4 times in Hep A buffer and applied to a Mono Q column (GE Healthcare). RNAP was eluted using a linear NaCl gradient (0.15 to 0.5 M) in Hep B buffer. Finally, RNAP was purified over a Superose 6 Increase size-exclusion chromatography (SEC) column (GE Healthcare) that was equilibrated in 20 mM Tris-HCl, pH 7.5, 2 mM MgCl 2 , 150 mM KCl, 1 mM dithiothreitol, and the RNAP-containing peak fractions were collected, flash-frozen in liquid nitrogen and stored at −80°C. RNAP mutants were purified using the same protocol. The open reading frame of the full-length E. coli Rho protein was cloned into the pET21b vector. The plasmid was used to transform E. coli strain BL21 (DE3), and recombinant protein expression was auto-induced. After 16 h at 30°C, cells were harvested by centrifugation (4,000g for 10 min at room temperature) and pellets were stored at −80°C. Cell pellets were resuspended in lysis buffer (50 mM Tris-HCl, pH 7.5, 10% (v/v) glycerol, 50 mM KCl) supplemented with complete, EDTA-free protease inhibitor cocktail tablets (Roche Applied Science) and lysed using sonication (5-s pulses with 10-s intervals for 10 min on ice). The cell lysate was clarified by centrifugation (30,000g for 40 min at 4°C) to remove insoluble debris. The supernatant was applied to an SP column (GE Healthcare) equilibrated in Buffer A (50 mM Tris-HCl, pH 7.5, 10% (v/v) glycerol, 50 mM KCl). The column was washed with 3 CV of Buffer A. Protein was eluted with Buffer B (50 mM Tris-HCl, pH 8.0, 10% (v/v) glycerol, 1 M KCl). Fractions containing recombinant Rho protein were pooled and diluted 5 times with Hep A buffer and applied to a Mono Q column (GE Healthcare) equilibrated in Hep A buffer. Protein was eluted with a linear NaCl gradient (0.02 to 0.5 M) in Hep B buffer. Finally, Rho protein was purified over a Superose 6 Increase SEC column (GE Healthcare) that was equilibrated in 10 mM Tris-HCl, pH 8.0, 300 mM NaCl, 1 mM dithiothreitol, and the peak fractions containing Rho protein were collected, flash frozen in liquid nitrogen and stored at −80°C. The open reading frame of the full-length E. coli NusA protein was cloned into the pSUMO vector, which has 6XHis tag conjugated with a N-terminal SUMO tag. The plasmid was used to transform E. coli strain BL21 (DE3), and recombinant 6XHis-SUMO-NusA was expressed as soluble protein by inducing with 0.5 mM Isopropyl β-D-thiogalactoside (Sigma) for 3 h at 37°C upon the culture reaching OD600=0.4–0.6. Cells were harvested by centrifugation (4,000g for 10 min at room temperature) and pellets were stored at −80°C. Cell pellets were resuspended in lysis buffer (50 mM Tris-HCl, pH 7.5, 10% (v/v) glycerol, 50 mM KCl) supplemented with complete, EDTA-free protease inhibitor cocktail tablets (Roche Applied Science) and lysed using sonication (5-s pulses with 10-s intervals for 10 min on ice). The cell lysate was clarified by centrifugation (30,000g for 40 min at 4°C) to remove insoluble debris. The supernatant was applied to a HisTrap column (GE Healthcare) equilibrated in HisTrap Buffer A (50 mM Tris-HCl, pH 8.0, 5% (v/v) glycerol, 0.5 mM β-mercaptoethanol, 500 mM NaCl, 10 mM imidazole). The column was washed with 20 column volumes (CV) of HisTrap Buffer A. Protein was eluted with HisTrap Buffer B (50 mM Tris-HCl, pH 8.0, 5% (v/v) glycerol, 0.5 mM β-mercaptoethanol, 250 mM NaCl, 250 mM imidazole). Fractions containing recombinant 6XHis-SUMO-NusA eluted from the HisTrap column were subjected to 6XHis-SUMO tag cleavage using SUMO protease (Invitrogen) in dialysis buffer (50 mM Tris-HCl, pH 8.0, 5% (v/v) glycerol, 0.5 mM β-mercaptoethanol, 200 mM NaCl) at 4°C for 16 h. After 6XHis-SUMO tag cleavage, protein mixture was applied to a HisTrap column (GE Healthcare) equilibrated in HisTrap Buffer A (50 mM Tris-HCl, pH 8.0, 5% (v/v) glycerol, 0.5 mM β-mercaptoethanol, 500 mM NaCl, 10 mM imidazole). Flow-through containing non-tagged NusA was pooled and applied to a Heparin column (GE Healthcare) equilibrated in Hep A buffer. Protein was eluted using a linear NaCl gradient (0.02 to 1.5 M NaCl) in Hep B buffer (50 mM Tris-HCl, pH 8.0, 5% (v/v) glycerol). The NusA-containing peak fractions were pooled, diluted 4 times in Hep A buffer and applied to a Mono Q column (GE Healthcare). NusA was eluted using a linear NaCl gradient (0.15 to 0.5 M) in Hep B buffer. Finally, NusA was purified over a Superose 6 Increase size-exclusion chromatography (SEC) column (GE Healthcare) that was equilibrated in 20 mM Tris-HCl, pH 7.5, 2 mM MgCl 2 , 150 mM KCl, 1 mM dithiothreitol, and the NusA-containing peak fractions were collected, flash-frozen in liquid nitrogen and stored at −80°C. The open reading frame of the full-length E. coli NusG protein was cloned into the pSUMO vector. The plasmid was used to transform E. coli strain BL21 (DE3), and recombinant protein expression was auto-induced. After 16 h at 30°C, cells were harvested by centrifugation (4,000g for 10 min at room temperature) and pellets were stored at −80°C. Cell pellets were resuspended in lysis buffer (50 mM Tris-HCl, pH 7.5, 10% (v/v) glycerol, 50 mM KCl) supplemented with complete, EDTA-free protease inhibitor cocktail tablets (Roche Applied Science) and lysed using sonication (5-s pulses with 10-s intervals for 10 min on ice). The cell lysate was clarified by centrifugation (30,000g for 40 min at 4°C) to remove insoluble debris. The supernatant was applied to a HisTrap column (GE Healthcare) equilibrated in HisTrap Buffer A (50 mM Tris-HCl, pH 8.0, 5% (v/v) glycerol, 0.5 mM β-mercaptoethanol, 500 mM NaCl, 10 mM imidazole). The column was washed with 20 column volumes (CV) of HisTrap Buffer A. Protein was eluted with HisTrap Buffer B (50 mM Tris-HCl, pH 8.0, 5% (v/v) glycerol, 0.5 mM β-mercaptoethanol, 250 mM NaCl, 250 mM imidazole). Fractions containing recombinant 6XHis-SUMO-NusG eluted from the HisTrap column were subjected to 6XHis-SUMO tag cleavage using SUMO protease (Invitrogen) in dialysis buffer (50 mM Tris-HCl, pH 8.0, 5% (v/v) glycerol, 0.5 mM β-mercaptoethanol, 200 mM NaCl) at 4°C for 16 h. After 6XHis-SUMO tag cleavage, protein mixture was applied to a HisTrap column (GE Healthcare) equilibrated in HisTrap Buffer A (50 mM Tris-HCl, pH 8.0, 5% (v/v) glycerol, 0.5 mM β-mercaptoethanol, 500 mM NaCl, 10 mM imidazole). Flow-through containing non-tagged NusG was pooled, diluted 4 times in Hep A buffer and applied to a Mono Q column (GE Healthcare). NusG was eluted using a linear NaCl gradient (0.15 to 0.5 M) in Hep B buffer. Finally, NusG was purified over a Superose 6 Increase size-exclusion chromatography (SEC) column (GE Healthcare) that was equilibrated in 20 mM Tris-HCl, pH 7.5, 2 mM MgCl 2 , 150 mM KCl, 1 mM dithiothreitol, and the NusG-containing peak fractions were collected, flash-frozen in liquid nitrogen and stored at −80°C. Nucleic-acid Scaffold Preparation Synthetic DNA and RNA oligonucleotides were obtained from Integrated DNA Technologies (IDT). The nucleic acids were dissolved in RNase-free deionized water at a concentration of 1 mM. To assemble the scaffold, template DNA and RNA were mixed at a 1:1 ratio, annealed by incubation at 95°C for 2 min, 75°C for 2 min, 45°C for 5 min, and then decreasing the temperature by 5°C every 2 min until reaching 25°C. The annealed template DNA:RNA hybrid was stored at −20°C until use.

Preparation of the PTC for Cryo-EM Purified E. coli

RNAP was mixed with template DNA:RNA hybrid at a molar ratio of 1:1.3 and incubated for 30 min at 30°C. Non-template DNA was added at a molar ratio of 3:1 and incubated for 20 min. To remove excess nucleic acid, the complexes were run over a Superose 6 Increase SEC column (GE Healthcare) that was equilibrated in 20 mM Tris-HCl, pH 8.0, 2 mM MgCl 2 , 50 mM KCl, 1 mM dithiothreitol. The peak fractions containing the elongation complex (EC) were pooled and sequentially mixed with NusG at a molar ratio of 1:3 and NusA at a molar ratio of 1:2, followed by SEC over a Superose 6 Increase column (GE Healthcare) that was equilibrated in 20 mM Tris-HCl, pH 8.0, 2 mM MgCl 2 , 50 mM KCl, 1 mM dithiothreitol to remove excess NusG and NusA. Purified EC–NusG–NusA complexes were then mixed with purified Rho at a molar ratio of 1:8. To stabilize and further purify the PTC complex, freshly formed PTC was purified and crosslinked using the GraFix method ( Kastner et al., 2008 ). The gradient solution contained 20 mM HEPES-KOH, pH 8.0, 50 mM KCl, 2 mM MgCl 2 , 10–30% (v/v) glycerol and 0–0.08% (v/v) gutaraldehyde. The samples were centrifuged at 36,000 rpm for 18 h at 4°C, using a Beckman-Coulter SW41 Ti swinging-bucket rotor. Fractions containing the PTC were pooled and dialyzed against buffer without glycerol and glutaraldehyde. Cryo-EM Grid Preparation UltrAuFoil (Quantifoil) R-1.2/1.3 Au 300 mesh grids were glow-discharged for 1 min. After applying 3.5 μl of sample, grids were blotted for 0.5–1 s with a blotting force of 0 and vitrified in liquid ethane using a Vitrobot Mark IV (FEI) with 100% humidity at 22°C.

Cryo-EM Data Acquisition and Processing

Grids were loaded into a Titan Krios electron microscope (FEI) operated at 300 kV and equipped with a Gatan K3 Summit direct electron detector (PTC60 dataset) or a Gatan K2 Summit direct electron detector (PTC18 dataset). Images of PTC60 were recorded in super-resolution mode with a pixel size of 0.539 Å and a defocus range of 0.8 – 1.8 µm, using a total dose of 50 electrons/Å 2 fractionated over 50 frames. Images of PTC18 were recorded in counting mode with a pixel size of 1.048 Å and a defocus range of 1.0 – 2.5 µm, using a total dose of 68 electrons/Å 2 fractionated over 50 frames. Collected micrographs were drift-corrected and dose-weighted in MotionCor2 ( Zheng et al., 2017 ), and the contrast transfer function (CTF) parameters were estimated using CTFFIND4 ( Rohou and Grigorieff, 2015 ). Approximately 10,000 particles were manually picked and subjected to 2D classification in RELION-3 ( Zivanov et al., 2018 ), which was used for all subsequent image processing. Projection averages of the most populated 10 classes were used as templates for automated particle picking in RELION-3 ( Scheres, 2015 ). Picked particles were manually inspected, then subjected to two subsequent rounds of 2D classification. Poorly populated classes were removed after each round, resulting in datasets of 2,865,384 particles for PTC60 and 1,182,362 particles for PTC18. To better separate images of fully-assembled PTC from those of EC and Rho ring by themselves, the particle images were subjected to supervised 3D classification with alignment, giving as reference models the EC–Rho complex (initial model obtained in RELION-3), EC (PDB: 6ALF) and Rho hexamer (PDB: 1PVO). Classes representing EC and Rho by themselves were removed, resulting in datasets of 551,397 PTC60 particles and 176,340 PTC18 particles. The PTC60 particles were 3D autorefined and subjected to 3D classification with alignment into 8 classes using the refined map and alignment angles. The best class that showed the highest resolution both in the EC and Rho sub-regions contained 82,394 particles (15% of the starting dataset), which were autorefined, post-processed and subjected to two cycles of CTF refinement and particle polishing in RELION-3, yielding the final density map at a nominal resolution of 3.1 Å. Local resolution calculations were performed using RELION-3. Multibody refinement was performed continuing on from the final 3D autorefinement, defining the EC density as Body 1 and the Rho–NusA density as Body 2. The corresponding masks were generated using the consensus map low-pass filtered to 30-Å resolution to define the boundary with the solvent region and the atomic models to define the boundaries between the two bodies. Soft-edges with a width of 8 Å were applied to the boundaries of the masks, resulting in a slight overlap of the two bodies. Motions corresponding to the first three eigenvectors are shown in Supplemental Movie S1 . The PTC18 particles were 3D autorefined and subjected to 3D classification with alignment into 8 classes using the refined map and alignment angles The class that showed the clearest features for the EC and Rho sub-regions contained 15,681 particles (12% of the starting dataset), which were autorefined, post-processed and subjected to two cycles of CTF refinement and particle polishing in RELION-3, yielding the “overall” density map at a nominal resolution of 7.9 Å. Multibody refinement was performed continuing on from the final 3D auto-refinement as described for PTC60. Motions corresponding to the first three eigenvectors are shown in Supplemental Movie S2 . The 176,340 particles after 3D autorefinement were also used for focused 3D classifications with alignment into 8 classes for the EC–NusG density and for the Rho–NusA density. For the EC–NusG density, the two classes with the highest-resolved features were combined and subjected to a second round of focused 3D classifications with alignment into 6 classes. The two classes with the highest-resolved features were combined and the 50,610 particles (29% of the starting dataset) were autorefined, post-processed and subjected to two cycles of CTF refinement and particle polishing in RELION-3, yielding the final EC–NusG density map at a nominal resolution of 4.0 Å. For the Rho–NusA density, the class showing the clearest features contained 35,760 particles (20% of the starting dataset), which were autorefined, post-processed and subjected to two cycles of CTF refinement and particle polishing in RELION-3, yielding the final Rho–NusA density map at a nominal resolution of 7.9 Å. The final EC–NusG and Rho–NusA maps were fitted into the overall PTC18 map to generate the “composite” map using the fit-in-map function in Chimera. Local resolution calculations were performed using RELION-3.

Model Building and Refinement

To build an initial model for PTC60, the atomic models of EC (PDB: 6ALH; ( Kang et al., 2017 ), Rho (PDB: 1PVO; ( Skordalakes and Berger, 2003 ), and the NusA NTD and S1 domain (PDB: 5LM7; ( Said et al., 2017 ) were fit into the cryo-EM map using Chimera ( Pettersen et al., 2004 ). The same structures plus the atomic model of NusG-NTD (PDB: 6C6U; ( Kang et al., 2018 ) were fit into the cryo-EM map of PTC18. These initial models were real-space refined in PHENIX ( Adams et al., 2010 ). The subunits in RNAP, the 6 monomers in the Rho ring, NusA-NTD, S1 and the nucleic acids were first refined as rigid bodies and were subsequently refined with secondary-structure restraints. Residues at the 5’ end of the single-stranded RNA were built de novo in Coot ( Emsley and Cowtan, 2004 ), and real-space refined with the previously refined model. To build the model for PTC18, we first placed the atomic models of the components into the overall map to obtain the global architecture of PTC18 and then refined the models against the composite map. Both models were then visually inspected, and outliers and problematic regions were fixed manually in Coot. The final refinement statistics are summarized in Table S2 . RNAP Mutant Strains and Plasmid Construction E. coli strains with polypeptide loop/domain deletions were constructed by using the lambda Red recombineering method together with CRISPR-Cas9 counterselection ( Reisch and Prather, 2015 ). Oligonucleotides (~80-mers) used for recombineering were designed to target the lagging strand of replicating DNA with upstream and downstream homology to the area of deletion. To bridge the gap between the points of the deletion, sequences encoding for one to three glycine residues were introduced in place of the deleted regions. Three phosphorothioated bases were used at the 5’end of the oligos to reduce its degradation rate in vivo . pKDsg (pSg-xxx) derivatives used for counterselection were constructed by circular polymerase extension cloning using primers with overlapping 20-bp protospacer sequences that corresponded to the fragment adjacent to appropriate PAM site (5’-NGG-3’) in the deleted parental sequences. E. coli MDS42 strains were first transformed with the pCas9cr4 plasmid and subsequently transformed with the sgRNA encoding plasmid (pSg-xxx). Cells that possessed both plasmids were grown in Super Optimal Broth with spectimomycin (Sp, 50 mg/l) and chloramphenicol (Cm, 30 mg/l) at 30°C. When OD 600 reached ~0.5, lambda Red was induced with 1.2% (w/v) L-arabinose, and cells were grown for another 20 min. Then, an oligonucleotide for recombineering was electroporated into the cells. After 2 h of recovery, the cells were plated on Luria broth (LB) with Sp, Cm and anhydrotetracycline (aTc, 100 ng/l) and incubated overnight at 30°C to select for survivors of the CRISPR/Cas9 selection. Colonies were screened with specific primers and the corresponding chromosome regions were verified by sequencing. To eliminate the pSg-xxx plasmid, cells were incubated in LB for 12 h at 37°C and streaked on LB plates. Individual colonies were selected and assessed for the loss of Sp resistance. The next pSg-xxx plasmid was used in a subsequent iteration to make another deletion. The pKDsg-15a plasmid was used to cure the pCas9cr plasmid that targeted the p15a origin of replication of pCas9cr. Upon transformation of pKDsg-15a into cells that contained pCas9cr, the cells were recovered in SOC (Super Optimal broth with Catabolite repression) for 2 h at 30°C, then aTc (100 ng/l) was added and incubated for an additional 2 h before plating on LB with Sp and aTc. The pKDsg-15a plasmid was cured by growth at 37°C. To create plasmid pVS10-ΔRpoB483–491/I9, a fragment of the rpoB gene with two deletions (Δβ 483–491 and I9) was amplified from genomic DNA of the SP1176 strain and was inserted into the BbvCI and SbfI sites of the pVS10 plasmid to replace the corresponding region of the wild-type gene. mRNA Purification, Reverse Transcription and qRT-PCR The pVE-RUT81-GFP plasmid was transformed into E. coli strain MDS42( Cardinale et al., 2008 ) and corresponding mutant strains that carried chromosomal RNAP mutations. Single colonies from each of the strains were inoculated from fresh LB plates into 3 ml of LB medium and grown overnight at 30°C with 50 mg/ml kanamycin (Km) with shaking. The next day (~20 h), 30 ml of fresh LB with 50 mg/ml Km were inoculated with 0.3 ml of the overnight cultures and grown in 250-ml flasks at 30°C with shaking to an OD 600 of ~0.3. For the chromosome-based Rho-dependent terminator reporter ( dsbB -RhoT), wild-type E. coli MDS 42 strain and the corresponding mutant strain that carried the chromosomal RNAP mutation were grown in LB (without Km). 10-ml aliquots were taken from each culture and transferred into 15-ml Falcon tubes before centrifugation at 5,000g for 5 min at 4°C. The supernatant was discarded and mRNA was purified using the MasterPure Complete DNA and RNA purification kit (Lucigen) according to the manufacturer’s instructions, except that DNAse I treatment was conducted for 30 min at 37°C and was supplemented with 500 units of ExoIII and 25 units of Sau3AI (both from New England Biolabs). RNA was re-dissolved in 50 ml TE buffer (10 mM Tris-HCl, pH 8.0, 0.1 mM EDTA) to a concentration of 1–2 mg/ml and the volume was adjusted to 500 ng/ml with TE buffer. cDNA was produced from 1 μg RNA using QuantiTect reverse transcription kit (Qiagen) according to the manufacturer’s instructions. For strand specificity, 0.7 μM primers of pVE_B reverse, GFP_3 reverse, dsbB D1 reverse and dsbB D2 reverse were used instead of the random primers provided in the kit. qPCR was performed using a QuantStudio 7 Flex real-time qPCR machine (Applied biosystems) from 5 ng DNA in one well (20 μl per well) in triplicates using pVE_B forward/reverse, GFP_3 forward/reverse, dsbB D1 forward/reverse and dsbB D2 forward/reverse primer pairs (5 ng/ml each) at an annealing temperature of 60°C. The sequences of the target region of the pVE-RUT81-GFP plasmid used for qPCR and the primers are shown in Table S5 . The pVE-RUT81-GFP plasmid is a derivative of the pMW-3110 plasmid ( Sedlyarova et al., 2017 ) with the 5’ UTR of the GFP gene replaced as shown in Table S5 .

In-vitro Transcription Termination Assay

Elongation complexes were formed with 75 nM T7A1-Trpt1 DNA ( Table S5 ) and 100 nM wild-type and mutant RNAP core with an equimolar concentration of σ70 in 100 μl of TB50 buffer (40 mM Tris-HCl, pH 8.0, 10 mM MgCl 2, 50 mM NaCl, 0.003% (v/v) Igepal-60, 5 mM β-mercaptoethanol with 40 unit of RNasin (Promega)). Transcription was initiated with 10 μM AUC primer and 25 μM of ATP and GTP for 5 min at 22°C. 1 μl of CTP [ ɑ−32 P] 3000Ci/nmol, (0.33 nM) was added for 5 min at 22°C. Where indicated NusA (100 nM) was added and incubated for 5 min at 22°C ( Figure 5F ). Rho (50, 25, and 12.5 nM) and NusG (1 μM) were added with the NTPs and the reactions were chased with 1 mM ATP and 200 μM of the other NTPs at 37°C for 5 min. The termination efficiency of the mutant RNAP (RpoB Δ483–491/I9) was compared with that of wild-type RNAP. Reactions were terminated by adding 2X STOP buffer and were heated for 5 min at 95°C and resolved on a 6% urea-PAGE for 20 min at 50 W, and dried and exposed to a phosphor imager screen. Note: For the titration of Rho with RNAP ( Figure 3I ), 100 nM wild-type RNAP and 100 nM NusA were added before the final chase along with NTPs, NusG, Rho, and NTPs. Rho-dependent Transcription Termination “ in trans ” Template 1 ( Table S5 ) was produced by PCR-directed mutagenesis ( Nudler et al., 1995 ) using Phusion DNA polymerase (New England Biolabs) and synthetic DNA oligos (IDT). It has the identical T7A1 promoter and initial transcribed sequence up to position +10 (counting from the +1 start of the transcription) followed by a modified sequence as shown in Table S5 . The template DNA was purified from a 2% agarose gel using a Qiagen Gel Extraction Kit according to the manufacturer’s instructions and diluted in TE buffer to a concentration ~1 pmol/μl. His6-tagged RNAP, NusA, NusG and Rho were purified as described previously. Rut81 synthetic RNA oligo was purchased from IDT and diluted in TE buffer up to 1 μM concentration. 5 to 10 pmol RNAP were mixed with 2X molar excess of DNA in 20 μl of TB50 (40 mM Tris-HCl, pH 8.0, 10 mM MgCl 2 , 50 mM NaCl, 0.003% (v/v) Igepal-60) for 5 min at 37°C, followed by addition of ApUpC (10 μM), GTP and ATP (25 μM each) for 5 min. Next, 15 μl TB50-equilibrated Ni-NTA-agarose beads (Qiagen) were added and the slurry was shaken for 5 min at room temperature in the presence of 1.5 mg/ml heparin. The sample was washed twice with 1 ml of TB1000 (as TB50 but with 1 M NaCl), incubated for 1 min at room temperature and washed 3 times with 1 ml of TB100 (as TB50 but with 100 mM NaCl). To produce EC14, ATP (5 μM) and 2 μl of CTP [ ɑ−32 P] 3000Ci/nmol (0.33 nM) were added for 5 min at room temperature followed by CTP (5 μM) for another 2 min. Beads were washed 4 times with 1 ml of TB100. EC14 was walked to position 29, 40, or 50. The walking reaction was performed at room temperature as described previously ( Nudler et al., 1994 ) using limited NTP sets (5 μM). Samples of the corresponding ECs were divided into equal aliquots and NusG (up to 1 μM) or/and NusA (up to 1 μM) or/and RUT81 RNA (up to 0.1 μM, Table S5 ) were added for 5 min at 22° C as indicated in each figure. To measure Rho-dependent termination, the samples were mixed with either 100 nM Rho hexamer pre-mixed with ATP (up to 1 mM final concentration) or mock solution containing ATP and TB100 instead of Rho. After incubation at 22°C for the indicated time, 10-μL aliquots were withdrawn and chased with 100 μM CTP, UTP, GTP and 1 mM ATP for 5 min at 37°C. The reactions were stopped by the addition of an equal volume of Stop Buffer (SB) (1X TBE (Tris/Borate/EDTA), 20 mM EDTA; 8 M urea, 0.025% (v/v) xylenthianol, 0.025% (w/v) bromophenol blue) and the RNA products were separated by 15% sequencing PAGE. To measure the termination efficiency, the beads were not chased with NTPs but instead were washed once with 1 ml of TB1000 and once with 1 ml of TB100. The volume was readjusted to 10 μl and the samples were quenched by the addition of an equal volume of SB. The gels were exposed to a phosphor-screen and the screen was scanned using an Amersham Typhoon Scanner (GE Healthcare). The data were analyzed using Image Quant software (GE Healthcare). In-vivo and in-vitro XLMS (Covalent Crosslink Mapping by Mass Spectrometry) and Data Analysis Strain construction: rpoC : 10XHis and rho : FLAG 3 . strains were constructed by introducing coding sequences for 10X H and DYKDDDDKDYKDDDDKDYKDDDDK after codons 1407 ( rpoC ) and 419 ( rho ), respectively, into parental E. coli strain MG1655 by means of lambda Red-mediated gene replacement ( Datsenko and Wanner, 2000 ). Successful construction was confirmed in each case by genomic sequencing and whole-cell enumerative proteomics. In-vitro crosslinking: Crosslinkers BS3 and DSS (Thermo Fisher Scientific) were dissolved in LC-MS grade H 2 0 and oxygen-depleted anhydrous DMSO (ZerO2, Millipore Sigma), respectively, at a concentration of 50 mM. Crosslinker was added to the target complex prepared in NHS-ester non-reactive buffer (50 mM HEPES, pH 7.5, 500 mM NaCl, 2 mM MgSO 4 , 1 mM TCEP) to a final concentration of 100–500 µM. Reactions were performed at 25°C in disposable inert cuvettes (UVette, Eppendorf), and monitored by continuous looped dynamic light scattering measurements of polydispersity (P d 0.5% of total were included as the variable modifications in pLink2 search parameters. pLink2 results were filtered for FDR (