7URG

cryo-EM structure of ribonucleotide reductase from Synechococcus phage S-CBP4 bound with TTP

ELECTRON MICROSCOPY

---

Sample
Ribonucleotide reductase from Synechoccus phage S-CBP4 bound with TTP

Specimen Preparation
Sample Aggregation State	PARTICLE
Vitrification Instrument	FEI VITROBOT MARK IV
Cryogen Name	ETHANE
Sample Vitrification Details	blot for 4 seconds before plunging

3D Reconstruction
Reconstruction Method	SINGLE PARTICLE
Number of Particles	107885
Reported Resolution (Å)	3.46
Resolution Method	FSC 0.143 CUT-OFF
Other Details
Refinement Type
Symmetry Type	POINT

Map-Model Fitting and Refinement
Id	1
Refinement Space	REAL
Refinement Protocol	FLEXIBLE FIT
Refinement Target
Overall B Value	74.37
Fitting Procedure
Details	The sequence for the ribonucleotide reductase from Synechococcus phage S-CBP4 was retrieved from UniProt with accession number M1PRZ0. The sequence was used as input for AlphaFold2 prediction with the five default model parameters and a template date cutoff of 2020-05-14. As the five models were largely identical in the core region and differing only in the location of the C-terminal tail, the structure predicted with the first model parameter was used in the subsequent process. The predicted structure was first processed and docked into the unsharpened map in phenix. The 25 N-terminal residues and 45 C-terminal residues were then manually removed due to lack of cryo-EM density, and residues 26-426 were retained in the model. We observed unmodeled density at the specificity site, and based on solution composition, we modeled a TTP molecule. The TTP molecule with magnesium ion from the crystal structure of Bacillus subtilis RNR (pdb: 6mt9) was extracted and rigid body fit into the unmodeled density in Coot. The combined model was refined with the unsharpened and sharpened maps using phenix.real_space_refine, with a constraint applied on the magnesium ion coordinated by the triphosphate in TTP according to the original configuration. Residue and loop conformations in the resulting structure were manually adjusted in Coot to maximize fit to map and input for an additional round of real-space refinement in phenix with an additional restraint for the disulfide bond between C30 and C196. Due to poor density of the magnesium ion, it was removed when deposited into PDB.

Data Acquisition
Detector Type	GATAN K3 BIOQUANTUM (6k x 4k)
Electron Dose (electrons/Å**2)	50

Imaging Experiment	1
Date of Experiment
Temperature (Kelvin)
Microscope Model	TFS TALOS
Minimum Defocus (nm)	600
Maximum Defocus (nm)	2000
Minimum Tilt Angle (degrees)
Maximum Tilt Angle (degrees)
Nominal CS	2.7
Imaging Mode	BRIGHT FIELD
Specimen Holder Model	OTHER
Nominal Magnification	79000
Calibrated Magnification
Source	FIELD EMISSION GUN
Acceleration Voltage (kV)	200
Imaging Details	Data was collected on a Thermo Fisher Talos Arcica Cryo-TEM with a Gatan K3 camera and BioQuantum energy filter.

EM Software
Task	Software Package	Version
PARTICLE SELECTION	cryoSPARC	3.3.1
IMAGE ACQUISITION	SerialEM	3.8
CTF CORRECTION	cryoSPARC	3.3.1
MODEL FITTING	PHENIX	1.20.1
INITIAL EULER ASSIGNMENT	cryoSPARC	3.3.1
FINAL EULER ASSIGNMENT	cryoSPARC	3.3.1
RECONSTRUCTION	cryoSPARC	3.3.1
MODEL REFINEMENT	PHENIX	1.20.1

Image Processing
CTF Correction Type	CTF Correction Details	Number of Particles Selected	Particle Selection Details
PHASE FLIPPING AND AMPLITUDE CORRECTION		581884	46 high quality micrographs were then selected, from which the blob picker routine was used to pick particles. The resulting 99k particles were extracted and subjected to 2D classification, and the top four unique 2D classes were selected and used as templates for template picking on the entire dataset. Due to the large variance in ice conditions in many of our micrographs, masks were manually defined for every micrograph, and particle picks outside the ideal ice region were excluded.