Structural basis for pegRNA-guided reverse transcription by a prime editor

Sample preparation

The PE2 (nSpCas9–engineered M-MLV RT), Cas9 (H840A) and M-MLV RTΔRNaseH (D200N/T306K/W313F/T330P) genes were PCR-amplified from pCMV-PE2 (Addgene plasmid 132775) and assembled separately into pET-based expression vectors with an N-terminal His6-SUMO-tag. The PE6a–d expression plasmids were constructed by replacing the RT gene in the PE2 expression plasmid with the synthesized PE6 RT genes (Eurofins Genomics), respectively. Mutations were introduced by a PCR-based method, and sequences were confirmed by DNA sequencing (Supplementary Table 3). After the plasmids were transformed into Escherichia coli Rosetta 2 (DE3), the E. coli cells were cultured at 37 °C until the optical density at 600 nm (OD600 nm) reached 0.8, and protein expression was induced at 20 °C for 18–20 h by the addition of 1 mM isopropyl-β-d-thiogalactopyranoside (Nacalai Tesque). The E. coli cells were collected by centrifugation, lysed by sonication in buffer A (20 mM Tris-HCl, pH 8.0, 1 M NaCl and 20 mM imidazole), and clarified by centrifugation. The clarified lysate was incubated with Ni-NTA Superflow resin (Qiagen) at 4 °C for 1 h and loaded into an Econo-Column (Bio-Rad). After the resin was washed with buffer A and buffer B (20 mM Tris-HCl, pH 8.0, 300 mM NaCl and 20 mM imidazole), the protein was eluted with buffer C (20 mM Tris-HCl, pH 8.0, 300 mM NaCl and 300 mM imidazole). The eluted protein was incubated with SUMO protease (produced in-house) at 4 °C overnight, and then loaded onto a HiTrap Heparin column (GE Healthcare) equilibrated with buffer D (20 mM Tris-HCl, pH 8.0 and 300 mM NaCl). The bound protein was eluted with a linear gradient of 0.3–2 M NaCl and further purified on a HiLoad 16/600 Superdex 200 pg column (GE Healthcare) equilibrated with buffer E (20 mM Tris-HCl, pH 8.0, 500 mM NaCl, 2 mM MgCl2 and 1 mM DTT). The peak fractions were collected and stored at −80 °C until use.

pegRNA preparation

Templates for in vitro transcription were prepared by annealing a forward T7 promoter oligonucleotide with an oligonucleotide containing the reverse complement of the T7 promoter and a pegRNA sequence (Supplementary Table 3). The in vitro transcription reaction was performed at 37 °C overnight, in 50 mM Tris-HCl, pH 8.0, 40 mM KCl, 20 mM MgCl2, 5 mM each NTP, 10 mM GMP, 5 mM DTT, 2 mM spermidine, 1 U ml−1 inorganic pyrophosphatase (Sigma), 80 µg ml−1 T7 RNA polymerase (produced in-house) and 20 nM template. The transcribed pegRNA was purified by 8% denaturing urea-PAGE, extracted from gel slices with Tris borate–EDTA buffer (Takara) and then ethanol precipitated. The pegRNA pellet was dissolved in nuclease-free water and stored at −20 °C.

In vitro prime editing assay

All in vitro prime editing reactions were performed using 5′-Cy5-labelled pre-nicked DNA substrates. These DNA substrates were annealed with three oligonucleotides (5′-Cy5-NTS, NTS-3′ and TS; 1:1:1 molar ratio for Fig. 1c and 1:1.5:1 molar ratio for the other experiments) (Supplementary Table 1) by heating to 95 °C for 2 min followed by slowly cooling to room temperature. For the pegRNA-MM, TS-MM was used in place of TS. When using untethered PE in place of PE2 in the reaction, purified dSpCas9 and purified RTΔRH were mixed at a molar ratio of 1:1 and handled like PE2 in the subsequent steps. The PE2–pegRNA complex (1.6 μM or 3.0 μM) was prepared by mixing the purified PE2 and pegRNA at 37 °C for 3 min. The binary complex (5 μl) was mixed with the 5′-Cy5-labelled pre-nicked DNA substrate (5 μl, 200 nM final concentration) and incubated at 37 °C for 10 min in PE reaction buffer (20 mM HEPES-NaOH, pH 7.5, 100 mM NaCl, 5% glycerol, 3 mM MgCl2, 0.2 mM EDTA and 5 mM DTT) supplemented with 250 μM each dNTP or U-Stall Solution (250 μM ddATP, 250 μM dTTP, 250 μM dGTP and 250 μM dCTP). The reaction was stopped by the addition of quench buffer containing EDTA (0.5 mM final concentration) and Proteinase K (60 ng). Aliquots (2 μl) were mixed with quench buffer (3 μl), and the reaction products were separated on 10% or 15% Novex PAGE TBE–urea gels (Invitrogen) and then visualized using an Amersham Imager 600 (GE Healthcare). The reverse transcription efficiencies of each group were calculated using Image J (ref. 23). In vitro prime editing experiments were performed at least three times.

Cryo-EM sample preparation

The 51-nt pre-nicked DNA substrates for cryo-EM samples were prepared by annealing three nucleotides (5′-NTS+3nt, NTS-3′ and TS-MM; 1:1:1 molar ratio). For the pre-initiation and initiation complexes, 5′-NTS was used in place of 5′-NTS+3nt (Supplementary Table 2). The dSpCas9–RTΔRH–pegRNA–target DNA complexes were reconstituted by incubating the purified dSpCas9, RTΔRH, the 115-nt or 137-nt pegRNA-MM and the 51-nt pre-nicked DNA substrate at a molar ratio of 6:6:8:3 at 37 °C for 30 min in PE reconstitution buffer (20 mM HEPES-NaOH, pH 7.5, 100 mM NaCl, 2.5% glycerol and 2 mM MgCl2), supplemented with 250 μM ddATP (for the initiation state) or U-Stall Solution (for the other states). The dSpCas9–pegRNA–target DNA complex (the pre-initiation complex) was reconstituted similarly without RTΔRH. The reconstituted complexes were purified by size-exclusion chromatography on a Superdex 200 Increase 10/300 column (GE Healthcare) equilibrated with buffer F (20 mM Tris-HCl, pH 8.0, 150 mM NaCl, 2 mM MgCl2 and 1 mM DTT). The purified complex solution (A260 = 4.6–11) was applied to Au 300 mesh R1.2/1.3 grids (Quantifoil), which were freshly glow-discharged with 3 μl amylamine, using a Vitrobot Mark IV (FEI) at 4 °C and 100% humidity, with a waiting time of 10 s and a blotting time of 4 s. The grids were then plunge-frozen in liquid ethane cooled at liquid nitrogen temperature.

Cryo-EM data collection

Cryo-EM data for the initiation, pre-initiation and elongation (28-nt) complexes were collected using a Titan Krios G3i microscope (Thermo Fisher Scientific) and for the other complexes using a Titan Krios G4 microscope (Thermo Fisher Scientific), both running at 300 kV and equipped with a Gatan Quantum-LS Energy Filter (GIF) and a Gatan K3 Summit direct electron detector in electron counting mode (University of Tokyo). All movies were recorded at a nominal magnification of 105,000×, corresponding to a calibrated pixel size of 0.83 Å, with a total dose of approximately 50 electrons per Å2 per 48 frames. The data were automatically acquired using the EPU software (Thermo Fisher Scientific). The dose-fractionated movies of the pre-initiation and elongation (28-nt) complexes were subjected to beam-induced motion correction and dose weighting using MotionCor2 (ref. 24) in RELION v.3.1.1 (ref. 25); those of the termination and initiation complexes were processed using patch motion correction in cryoSPARC v.3.3.2 (ref. 26); and those of the elongation complex (16-nt) were handled using patch motion correction in cryoSPARC v.4.2.1. The contrast transfer function (CTF) parameters for the termination and initiation complexes, the pre-initiation and elongation (16-nt) complexes and the elongation (28-nt) complex were estimated using patch-based CTF estimation in cryoSPARC versions 3.3.2, 4.2.1 and v4.4, respectively.

Single-particle cryo-EM data processing

Data for the termination and initiation complexes were processed using cryoSPARC v3.3.2 and v4.2.1. Data for the pre-initiation and elongation (16-nt) complexes and the elongation (28-nt) complex were processed using cryoSPARC v4.2.1 and v4.4, respectively. All reported resolutions are based on the gold-standard Fourier shell correlation with a cut-off of 0.14327, and the local resolution was estimated with BlocRes28 in cryoSPARC.

For the termination complex, 1,112,419 particles were selected using a Topaz picking model from the 4,363 motion-corrected and dose-weighted micrographs, and extracted at a pixel size of 3.32 Å. These particles were subjected to two rounds of two-dimensional (2D) classification to separate 671,078 promising particles from junk particles. Then, 500,000 particles were randomly selected from each particle set, and subsequently used for ab initio reconstruction to generate good initial and junk maps. All of the extracted particles were further curated by three rounds of heterogeneous refinement with two good initial and two junk maps, while updating the two good reference maps. The 248,187 particles in the best class were re-extracted at a pixel size of 1.30 Å and subsequently refined using non-uniform refinement29 with optimization of the CTF value, resulting in the 3.00-Å overall map. Particle subtraction was performed on the refined particles using a mask around the Cas9–pegRNA scaffold region, and the signal-subtracted particles were used for local refinement (rotation search extent 5 deg, shift search extent 2 Å, initial lowpass resolution 8 Å) with a local mask around the RTΔRH, resulting in the 3.48-Å local map. Finally, the overall and local maps were merged into the final composite map, using the vop maximum command in UCSF ChimeraX30.

For the initiation complex, 2,532,892 particles were chosen using a Topaz picking model from the 5,266 motion-corrected and dose-weighted micrographs, and extracted at a pixel size of 3.32 Å, as described above. These particles were subjected to two rounds of 2D classification to select 1,607,568 promising particles, which were further curated through three rounds of heterogeneous refinement, as described above. The 656,084 particles in the best class were re-extracted at a pixel size of 1.30 Å and then subjected to 3D classification (five classes, target resolution = 4 Å, PCA initialization mode) with a focus mask around the RTΔRH. The 118,125 particles in the best class were refined using non-uniform refinement, resulting in the 3.12-Å overall map. To further improve the local resolution around the RTΔRH, particle subtraction and local refinement were performed as described above, resulting in the 4.10-Å local map around the RTΔRH. Finally, the overall and local maps were merged into the final composite map, using the vop maximum command in UCSF ChimeraX.

For the pre-initiation complex, 3,357,907 particles were selected using a Topaz picking model from the 8,154 motion-corrected and dose-weighted micrographs, and extracted at a pixel size of 3.32 Å. These particles were subjected to two rounds of 2D classification to select 1,382,881 promising particles, which were further curated through three rounds of heterogeneous refinement in a similar manner to the procedure used for the termination complex. The 976,259 particles in the good classes were re-extracted with a pixel size of 1.30 Å and subsequently refined using non-uniform refinement, resulting in a 3.11-Å map, in which the density for the PBS–NTS heteroduplex was, however, unresolved. Therefore, the aligned particles were subjected to 3D classification (five classes, target resolution = 5 Å, PCA initialization mode) with a focus mask around the position of the PBS–NTS heteroduplex in the initiation complex. The 197,777 particles in the best class were refined using non-uniform refinement with optimization of the CTF value, resulting in the final 3.22-Å overall map.

For the elongation complex (16-nt), 3,208,543 particles were chosen using a Topaz picking model from the 7,932 motion-corrected and dose-weighted micrographs, and extracted at a pixel size of 3.32 Å. These particles were subjected to two rounds of 2D classification to select 2,262,020 promising particles, which were further curated through three rounds of heterogeneous refinement in a similar manner to the procedure used for the termination complex. The 924,985 particles in the two good classes were re-extracted at a pixel size of 1.30 Å and subjected to 3D classification (four classes, target resolution = 6 Å, PCA initialization mode) with a focus mask around the RTΔRH. The 133,711 particles in the best class were then refined with a manually generated solvent mask just before non-uniform refinement with optimization of the CTF value, resulting in the 3.10-Å overall map. To further improve the local resolution around the RTΔRH, particle subtraction and local refinement were performed as described for the termination complex, resulting in the 6.06-Å local map around the RTΔRH. Finally, the overall and local maps were merged into the final composite map, using the vop maximum command in UCSF ChimeraX.

For the elongation complex (28-nt), 4,851,974 particles were selected using a Topaz picking model from the 9,872 motion-corrected and dose-weighted micrographs, and extracted at a pixel size of 3.32 Å. These particles were subjected to two rounds of 2D classification to select 2,800,847 promising particles, which were further curated through three rounds of heterogeneous refinement in a similar manner to the procedure used for the termination complex. The 702,552 particles in the best class were re-extracted at a pixel size of 1.15 Å and subjected to 3D classification (six classes, target resolution = 4 Å, PCA initialization mode) with a focus mask around the RTΔRH and the RNA–DNA heteroduplex along with Cas9. The 104,057 particles in the best class were refined using non-uniform refinement with optimization of the CTF value, resulting in the final 3.19-Å overall map. To further improve the local resolution around the RTΔRH, particle subtraction and local refinement were performed as described for the termination complex, resulting in the 4.54-Å local map around the RTΔRH. Finally, the overall and local maps were merged into the final composite map, using the vop maximum command in UCSF ChimeraX.

Model building and validation

The model of the termination complex was built using the cryo-EM structure of the SpCas9–sgRNA–target DNA complex in the checkpoint state (PDB 7Z4L; ref. 12) and the crystal structure of apo-M-MLV RT (PDB 4MH8; ref. 17) as the reference models, followed by manual model building using Coot (ref. 31) against the final density map sharpened using DeepEMhancer. The models of the other complexes were built using the model of the termination complex as the reference, followed by manual model building using Coot and ISOLDE (ref. 32) against the final density map sharpened using DeepEMhancer or local-resolution filtering in cryoSPARC. All models were refined using phenix.real_space_refine v.1.20.1 (ref. 33) with secondary structure and base pair restraints. The structure validation was performed using MolProbity in the PHENIX package34. The EMRinger score35 and 3DFSC sphericity36 were calculated by PHENIX and by the 3DFSC Processing Server (https://3dfsc.salk.edu/upload/info/), respectively. The statistics of the 3D reconstruction and model refinement are summarized in Extended Data Table 1. The cryo-EM density map figures were generated using UCSF ChimeraX. Molecular graphics figures were prepared using UCSF ChimeraX and CueMol (http://www.cuemol.org).

Mammalian prime editing assay

HEK293FT cells were purchased from Thermo Fisher Scientific (R70007) and maintained in DMEM-GlutaMAX (Thermo Fisher Scientific, 10569044) with 1× penicillin–streptomycin (Thermo Fisher Scientific, 15140122) and 10% FBS (VWR, 97068-085) at 37 °C with 5% CO2. The cells were seeded at a density of 2 × 104 cells per well in 96-well plates for transfection. Transfections were performed using Lipofectamine 3000 (Thermo Fisher Scientific, L3000015) when cells reached around 90% confluency. In total, 200 ng plasmids, including 150 ng PE plasmid with 50 ng pegRNA plasmid for PE2, or 135 ng PE plasmid and 50 ng pegRNA with 15 ng sgRNA plasmid for PE3, were transfected into each well. Three wells were transfected for each condition. Three days after transfection, genomic DNA was extracted using 50 µl QuickExtract DNA extraction solution (Lucigen, QE09050) by cycling at 65 °C for 15 min, 68 °C for 15 min and 95 °C for 10 min. Two rounds of PCR were conducted to amplify target sites with NEBNext High-Fidelity 2× PCR Master Mix (NEB, M0541L). For the first round of PCR, 2.5 µl of cell lysate was used as the template in 10-µl PCR reactions under the following thermal cycling conditions: one cycle, 98 °C, 30 s; 12 cycles, 98 °C, 10 s, 69 °C, 20 s, 72 °C, 30 s; one cycle, 72 °C, 2 min; 4 °C hold. For the second round of PCR, 1 µl of PCR product from the first round was used as the template in 10-µl PCR reactions under the following thermal cycling conditions: one cycle, 98 °C, 30 s; 18 cycles, 98 °C, 10 s, 63 °C, 20 s, 72 °C, 30 s; one cycle, 72 °C, 5 min; 4 °C hold. All amplicons were sequenced using a MiSeq Reagent Kit v.2, 300-cycle (Illumina, MS-102-2002). The prime editing efficiency was quantified using the published CRISPResso2 pipeline37.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.


Source link

Total
0
Shares
Leave a Reply

Your email address will not be published. Required fields are marked *

Related Posts