The C. crescentus GcrA-TACD dataset was collected on a FEI Titan Krios equipped at 300 keV with a K2 Summit direct electron detector (Gatan). The images were recorded using Serial EM software at a nominal magnification of 22 500 (1.0 Å/ pixel) in counting mode. Each image movie of 32 frames was collected by exposure of 7.6 s to give a total electron exposure of 60.8 electrons/Å2 (flux: 8 electrons/pixel/s). The 1886 images were collected with defocus range from −1.2 to −2.2 μm. Frames of individual movies were aligned using MotionCor2, and contrast-transfer-function estimations were performed using CTFFIND4. Image processing was performed with RELION 3.0. A total of 865 129 particles were auto picked and extracted from the dataset by using the 2D classified templates generated from previous C. crescentus GcrA TACU. The particles were subjected to 2D classification (N = 50, iterations 25) and the good 2D classes were subjected to 3D classification (N = 4, iterations = 25) using a 40 Å low-pass-filtered cryo-EM structure of C. crescentus GcrA TACU as the initial model. The 3D class, which contains 109 869 particles and shows clear feature of C. crescentus RNA polymerase, GcrA and promoter DNA were selected and subjected to auto refinement, CTF refinement, particle polishing, a second round of auto refinement, and postprocess, resulting in the final map. The Gold-standard Fourier-shell-correlation analysis indicated a nominal resolution of 3.79 Å at 0.143 FSC cutoff.
The structure of C. crescentus GcrA TACU was fit into the cryo-EM map. The iterative cycles of model building in Coot (Ramachandran, trans peptide, planar peptide restraints applied) and refinement in Phenix were performed.
The structure of C. crescentus GcrA TACU was fit into the cryo-EM map. The iterative cycles of model building in Coot (Ramachandran, trans peptide, planar peptide restraints applied) and refinement in Phenix were performed.
