Strains and growth conditions
All bacterial and phage strains used in this study are listed in Supplementary Table 2. E. coli strains were routinely grown at 37 °C in Luria–Bertani (LB) medium for cloning and maintenance. Phages were propagated by infecting a culture of E. coli MG1655 at an optical density at 600 nm (OD600) of approximately 0.1–0.2 with a multiplicity of infection of 0.1. Cleared cultures were pelleted by centrifugation to remove residual bacteria and filtered through a 0.2-μm filter. Chloroform was then added to phage lysates to prevent bacterial growth. Antibiotics were used at the following concentrations (liquid; plates): carbenicillin (50 μg ml−1; 100 μg ml−1) and chloramphenicol (20 μg ml−1; 30 μg ml−1).
Plasmid construction
All plasmids are listed in Supplementary Table 3. All primers and synthesized gene sequences are listed in Supplementary Table 4.
For the pEssential-gp77 construct, the coding sequence of gene 77 from phage SECΦ27 was codon optimized for expression in E. coli, and commercially synthesized by Integrated DNA Technology as a gBlock (TZ-1) and assembled into pEssential vector amplified with TZ-2 and TZ-3 by Gibson assembly.
For the pBAD33-gp77 constructs, gp77 was PCR amplified from phage SECΦ27 using primers TZ-4 and TZ-5 and inserted into pBAD33 linearized with primers TZ-6 and TZ-7 by Gibson assembly. To add a C-terminal HA tag, primers TZ-10 and TZ-11 were used to PCR amplify pBAD33-gp77 followed by Gibson assembly.
For the pBAD33-portal protein constructs, gp8T7, gp31T3, gp21SECΦ18 or gp52 SECΦ27 was PCR amplified from the corresponding T7, T3, SECΦ18 or SECΦ27 phage using primers TZ-12 to TZ-17, or TZ-23 and TZ-24, respectively. Amplified fragments were inserted into pBAD33 linearized with TZ-6 and TZ-7 using Gibson assembly. To add a C-terminal HA tag, primers TZ-20 to TZ-22 were each used in combination with TZ-10 to PCR amplify the corresponding pBAD33-portal protein constructs followed by Gibson assembly.
For the pBAD33-razr construct, razr was PCR amplified from pLAND-razr using primers TZ-18 and TZ-19 and inserted into pBAD33 linearized with TZ-6 and TZ-7 by Gibson assembly.
For the pET-razr–His6 and pET-gp77–His6 constructs: razr–His6 or gp77–His6 was PCR amplified from pLAND-razr or pBAD33-gp77 using primers TZ-27 to TZ-30, respectively. Amplified fragments were inserted into pET vector linearized with TZ-25 and TZ-26 by Gibson assembly.
For pACYC constructs, to construct pACYC-gp77–HA, gp77–HA was PCR amplified from pBAD33-gp77–HA using primers TZ-33 and TZ-34 and inserted into pACYC linearized with primers TZ-31 and TZ-32. To construct pACYC-gp8T7–Flag, first gp8T7–HA was PCR amplified from pBAD33-gp8T7–HA using primers TZ-34 and TZ-35 and inserted into pACYC linearized with primers TZ-31 and TZ-32. To replace the C-terminal HA tag with a Flag tag, primers TZ-36 and TZ-37 were used to amplify pACYC-gp8T7–HA followed by Gibson assembly.
For the pLAND-razr constructs, pLAND-razr–Flag or pLAND-razr(ΔAGN) was constructed by PCR amplifying pLAND-razr with primers TZ-8 and TZ-9, or TZ-38 and TZ39, followed by Gibson assembly. To construct pLAND-razrKv, DNA encoding the razrKv open reading frame was codon optimized for expression in E. coli and 200 bp of the upstream region from the source organism was added for native expression. DNA was commercially synthesized by Integrated DNA Technology as a gBlock (TZ-40) and assembled into a promoter-less backbone of pLAND amplified with TZ-41 and TZ-42 by Gibson assembly. To construct pLAND-razrEc + ZFD(Kv), sequences for ZFD from RAZRKv were amplified with primers TZ-43 and TZ-44 and inserted into pLAND-razrEc amplified with TZ-45 and TZ-46 by Gibson assembly. To construct pLAND-razrKv+ZFD(Ec), sequences for ZFD from RAZREc were amplified with primers TZ-47 and TZ-48 and inserted into pLAND-razrKv amplified with TZ-49 and TZ-50 by Gibson assembly.
All mutants were constructed by site-directed mutagenesis using primers designed by Takara Bio In-Fusion design tool.
Strain construction
Plasmids described above were introduced into E. coli MG1655 by Transformation and Storage Solution (TSS) transformation or electroporation.
SECΦ27-mutant phages with gene 77 deleted were generated using a CRISPR–Cas system for targeted mutagenesis as previously described39. In brief, sequences for RNA guides to target Cas9-mediated cleavage were designed using the toolbox in Geneious Prime (v2022.0.2) and selected for targeting of gene 77 but nowhere else in the SECΦ27 genome. The guides were inserted into the pCas9 plasmid and tested for their ability to restrict SECΦ27. An efficient guide was selected, and the pCas9-guide plasmid was co-transformed into E. coli MG1655 with a high copy-number repair plasmid with sequences flanking gene 77 for homologous recombination, and a pEssential plasmid encoding a recoded version of Gp77 under an arabinose-inducible control. Single colonies of E. coli MG1655 containing all three plasmids were grown overnight in LB medium. Overnight cultures were back- diluted 1:10 in 3 ml of LB + 0.2% arabinose and grown at 37 °C for 2 h to induce Gp77 expression. The WT SECΦ27 phage was mixed with 150 μl of induced culture and 4 ml LB + 0.5% agar + 0.2% arabinose and spread on an LB + 1.2% agar + antibiotic + 0.2% arabinose plate. Plates were incubated at 25 °C overnight. Single plaques were screened by Sanger sequencing, and two clones with gene 77 deleted were propagated on strains containing pCas9-guide and pEssential-gp77(recoded) with arabinose induction.
Phage spotting assays and EOP measurements
Phage spotting assays were conducted similarly to that previously described40. For phage spotting assays, 80 μl of a bacterial strain of interest was mixed with 4 ml LB + 0.5% agar and spread on an LB + 1.2% agar + antibiotic plate. Phage stocks were then serially diluted in 1× FM buffer (20 mM Tris-HCl pH 7.4, 100 mM NaCl and 10 mM MgSO4), and 2 μl of each dilution was spotted on the bacterial lawn. Plates were then incubated at 25 °C overnight before imaging. EOP was calculated by comparing the ability of the phage to form plaques on an experimental strain relative to the control strain. Experiments were replicated three times independently and representative images are shown.
For spotting phage SECΦ27 lacking gene 77, E. coli MG1655 containing pEssential-gp77(recoded) or an empty vector were grown overnight in LB medium. Overnight cultures were back diluted 1:10 in 3 ml of LB + 0.2% arabinose and grown at 37 °C for 2 h to induce Gp77 expression. Of each induced culture, 300 μl was mixed with 4 ml LB + 0.5% agar + 0.2% arabinose and spread on an LB + 1.2% agar + antibiotic + 0.2% arabinose plate. Phages were serially diluted as above and spotted on the bacterial lawn. Plates were incubated at 37 °C overnight. For spotting SECΦ27 lacking gene 77 onto strains producing Gp77 variants (L20R or V22R), E. coli MG1655 containing pBAD33-EV or pBAD33-gp77 (WT or each variant) were grown and processed as described above.
Isolation of phage escape mutants to infect RAZR
SECΦ27, T3, T7 or SECΦ18 escape mutants of RAZR were isolated by plating a population of phage onto RAZR-containing cells. 20 µl of 109–1010 plaque-forming units (PFU) per millilitre phage was mixed with 80 µl overnight culture of E. coli MG1655 pLAND-razr and the mixture was added to 4 ml of LB + 0.5% agar and spread onto LB + 1.2% agar. Plates were incubated at 25 °C overnight. Single plaques were isolated and propagated using the same strain in LB at 25 °C. Amplified phage lysates were pelleted to remove bacteria, and then plated to single plaques and propagated similarly for a second round of isolation to improve purity. Escape phages were then sequenced by Illumina sequencing as described below to identify mutations.
Phage DNA extraction and Illumina sequencing
Phage DNA extraction and sequencing were conducted as previously described40. To extract phage DNA, high titre phage lysates (more than 106 PFU µl−1) were treated with DNase I (0.001 U µl−1) and RNase A (0.05 mg ml−1) at 37 °C for 30 min. 10 mM EDTA was used to inactivate the nucleases. Lysates were then incubated with proteinase K at 50 °C for 30 min to disrupt capsids and release phage DNA. Phage DNA was isolated by ethanol precipitation. In brief, NaOAc pH 5.2 was added to 300 mM followed by 100% ethanol to a final volume fraction of 70%. Samples were incubated at −80 °C overnight, pelleted at 21,000g for 20 min and supernatant removed. Pellets were washed with 100 µl isopropanol and 200 µl 70% (v/v) ethanol, and then air dried at room temperature and resuspended in 25 µl 1× TE buffer (10 mM Tris-HCl and 0.1 mM EDTA pH 8).
To prepare Illumina sequencing libraries, 100–200 ng of genomic DNA was sheared in a Diagenode Bioruptor 300 sonicator water bath for 20 × 30 s cycles at maximum intensity. Sheared genomic DNA was purified using AmpureXP beads, followed by end repair, 3′ adenylation and adaptor ligation. Barcodes were added to both 5′ and 3′ ends by PCR with primers that anneal to the Illumina adaptors. The libraries were cleaned by Ampure XP beads using a double cut to elute fragment sizes matching the read lengths of the sequencing run. Libraries were sequenced on an Illumina MiSeq at the MIT BioMicro Center. Illumina reads were assembled to the corresponding reference genomes using Geneious Prime (v2022.0.2).
Toxicity assays on solid media
Bacterial toxicity assays were conducted similarly to that previously described40. For co-producing RAZR with Gp77 or portal proteins (Gp8T7, Gp31T3 and Gp21SECΦ18), single colonies of E. coli MG1655 harbouring pLAND-razr and pBAD33-gp77 or pBAD33-portal protein (WT or the corresponding variants) were grown for 6 h at 37 °C in LB–glucose to saturation. Of each saturated culture, 200 μl was then pelleted by centrifugation at 4,000g for 10 min, washed once in 1× PBS, and resuspended in 400 μl 1× PBS. Cultures were then serially diluted tenfold in 1× PBS and spotted on M9L plates supplemented with 0.4% glucose or 0.2% arabinose. M9L plates contain M9 medium (6.4 g l−1 Na2HPO4–7H2O, 1.5 g l−1 KH2PO4, 0.25 g l−1 NaCl and 0.5 g l−1 NH4Cl medium supplemented with 0.1% casamino acids, 0.4% glycerol, 2 mM MgSO4 and 0.1 mM CaCl2) supplemented with 5% LB (v/v). Plates were then incubated at 37 °C overnight before imaging.
To quantify colony sizes, cultures were processed as described above and plated on M9L plates supplemented with 0.4% glucose or 0.2% arabinose to single colonies. Colony sizes of 30–100 colonies on each plate were quantified by Fiji, and the ratio of colony sizes on arabinose and glucose plates were calculated. Data reported are three independent biological replicates.
For producing full-length RAZR, E. coli MG1655 containing pBAD33-razr were grown to saturation and processed as above. Cultures were plated onto 0.4% glucose and 0.2% arabinose and incubated at 37 °C overnight.
Co-immunoprecipitation analysis
Co-immunoprecipitation experiments were conducted similar to those previously described40. For co-producing RAZR with Gp77 or portal proteins (Gp8T7, Gp31T3 and Gp21SECΦ18), E. coli MG1655 containing pLAND-razr–Flag (WT or mutant variants) and pBAD33-gp77–HA (WT or mutant variants) or pBAD33-portal protein–HA were grown overnight in M9–glucose. Overnight cultures were back diluted to OD600 = 0.05 in 50 ml of M9 (no glucose) and grown to OD600 of approximately 0.3 at 37 °C. Cells were induced with 0.2% arabinose for 30 min at 37 °C, then OD600 was measured and cells were pelleted at 4,000g for 10 min at 4 °C. Supernatant was removed and cells were resuspended in 800 μl lysis buffer (25 mM Tris-HCl pH 8.0, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100 and 5% glycerol) supplemented with protease inhibitor (Roche), 1 μl ml−1 Ready-Lyse Lysozyme Solution (Lucigen) and 1 μl ml−1 benzonase nuclease (Sigma). Samples were lysed by two freeze–thaw cycles, and lysates were normalized by OD600. Lysates were pelleted at 21,000g for 10 min at 4 °C, and 750 μl of supernatant were incubated with pre-washed anti-Flag magnetic agarose beads (Pierce) for 1 h at 4 °C with end-over-end rotation. Beads were then washed three times with 500 μl lysis buffer. Laemmli sample buffer (1×; Bio-Rad) supplemented with 2-mercaptoethanol was added to beads directly to elute proteins. Samples were boiled at 95 °C and analysed by 4–20% SDS–PAGE and transferred to a 0.2-μm PVDF membrane. Anti-Flag and anti-HA antibodies (#14793 and #3724, Cell Signaling Technology) were used at a final concentration of 1:1,000, and SuperSignal West Femto Maximum Sensitivity Substrate (Thermo Fisher) was used to develop the blots. Blots were imaged by a ChemiDoc Imaging system (Bio-Rad). Images shown are representatives of two or three independent biological replicates.
Protein expression and purifications
To produce His6-tagged RAZR (WT or mutant variants) and His6-tagged Gp77 (WT or mutant variants), E. coli BL21(DE3) cells were transformed with pET-razr–His6 or pET-gp77–His6 and grown in LB medium to OD600 of 0.5. Protein expression was induced by addition of 0.2 mM IPTG, and cells were grown overnight at 16 °C. The culture was centrifuged at 4,000g for 10 min at 4 °C, and cell pellet was resuspended in lysis buffer (50 mM Tris-HCl pH 8.0, 150 mM NaCl, 2 mM MgCl2, 10 μM ZnCl2 and 1 mM dithiothreitol) supplemented with 0.4 mM PMSF and 10 μg ml−1 lysozyme. Cells were disrupted using sonication with amplitude 50 and 3 min total process time (10 s on and 20 s off; Qsonica) and glycerol was added to the lysate at final 10% concentration after sonication. The supernatant was separated from the pellet by centrifugation (15,000 rpm for 30 min, JA-25.50 rotor (Beckman Coulter)). The clarified supernatant was loaded onto a gravity-flow chromatography column (Bio-Rad) packed with 2 ml Ni-NTA agarose resin (Qiagen) pre-equilibrated with 15 ml lysis buffer. The resin was washed with 10 column volumes of wash buffer 1 (50 mM Tris-HCl pH 8.0, 500 mM NaCl, 2 mM MgCl2, 10 μM ZnCl2, 10 mM imidazole, 10% glycerol and 1 mM dithiothreitol), and then with 10 column volumes of wash buffer 2 (50 mM Tris-HCl pH 8.0, 150 mM NaCl, 2 mM MgCl2, 10 μM ZnCl2, 50 mM imidazole, 10% glycerol and 1 mM dithiothreitol). The proteins were eluted in 4 ml elution buffer (50 mM Tris-HCl pH 8.0, 150 mM NaCl, 10 μM ZnCl2, 250 mM imidazole, 10% glycerol and 1 mM dithiothreitol). To remove remaining contaminants, the eluted protein sample was loaded onto a size-exclusion chromatography Superose 6 Increase 10/300 GL column (Cytiva) pre-equilibrated in the size-exclusion chromatography buffer (50 mM Tris-HCl pH 8.0, 150 mM NaCl, 10 μM ZnCl2, 5% glycerol and 1 mM dithiothreitol). The purity of the protein samples were assessed spectrophotometrically and by SDS–PAGE. To reconstitute the WT complex in vitro, purified RAZR–His6 and Gp77–His6 were mixed at a 1:1 molar ratio using concentrations calculated by Bradford assay and loaded onto a Superose 6 Increase 10/300 GL column (Cytiva). The fractions containing the complex of interest were pooled, concentrated and used for structural determination.
For the RAZR(H154A)–Gp77 complex, E. coli BL21(DE3) cells were transformed with pET-razr(H154A)–His6 and pACYC-gp77–HA. Protein expression and purification were conducted as described above.
For the RAZR(H154A)–Gp8T7 complex, E. coli BL21(DE3) cells were transformed with pET-razr(H154A)–His6 and pACYC-gp8T7–Flag. Protein expression and purification with Ni-NTA agarose resin were conducted as described above. Eluted protein sample was mixed with 2 ml anti-Flag M2 affinity gel (Millipore Sigma) pre-equilibrated with the size-exclusion chromatography buffer, and incubated overnight at 4 °C. The mixture was loaded onto a gravity-flow chromatography column (Bio-Rad) and washed three times with 3 ml size-exclusion chromatography buffer. One bed volume (2 ml) of 1 mg ml−1 Flag peptide (APExBIO) was added to the resin, incubated for 30 min at 4 °C, and proteins were eluted by gravity. Elution was repeated three times, pooled and concentrated, and then loaded onto a Superose 6 Increase 10/300 GL column (Cytiva) for further purification.
For analytical size-exclusion chromatography, purified RAZR–His6, Gp77–His6 or the co-purified RAZR(H154A)–His6–Gp77–HA complex was loaded onto a Superose 6 Increase 3.2/300 analytical size-exclusion chromatography column (Cytiva) pre-equilibrated in the size-exclusion chromatography buffer. Each fraction corresponding to the peak of RAZR(H154A)–His6–Gp77–HA was analysed by SDS–PAGE, transferred to a PVDF membrane and blotted with anti-HA (Cell Signaling Technology) and anti-His (Invitrogen) antibody as described below.
Cryo-EM sample preparation
For the co-purified RAZR(H154A)–His6–Gp77–HA complex, before vitrification, 2.5 µl of the complex (2.6 mg ml−1) was placed on 200-mesh Quantifoil 2/1 copper grids (0.5-s incubation time), which had been glow-discharged for 60 s in an easiGlow glow discharger (Pelco) at 25 mA and were blotted using a FEI Vitrobot Mark IV instrument for 4 s with a blot force of +4 (6 °C; 95% relative humidity).
For the in vitro-reconstituted complex of RAZR–His6 and Gp77–His6, 2.5 µl of the sample (0.5 mg ml−1) was placed on 200-mesh Quantifoil 2/1 copper grids using the same parameters as mentioned above.
For the RAZR–His6, 2.5 µl of the sample (0.75 mg ml−1) was placed on 200-mesh Quantifoil 2/1 copper grids using the same parameters as mentioned above.
For the Gp77–His6, 2.5 µl of the sample (1 mg ml−1) was placed on 200-mesh Quantifoil 2/1 copper grids using the same parameters as mentioned above, except with a 5-s incubation time of the sample on the grid.
For the co-purified complex of Gp8T7–Flag and RAZR(H154A)–His6, 2.5 µl of the complex (0.6 mg ml−1) was applied to 200-mesh Quantifoil 2/1 copper grids with a 2-nm carbon support. The grids were glow discharged for 20 s in an easiGlow glow discharger (Pelco) at 25 mA. The sample was incubated on the grid for 10 s, then blotted using a FEI Vitrobot Mark IV instrument for 3.5 s with a blot force of +4 (6 °C; 95% relative humidity).
The protein–antibody complex was formed by incubating the co-purified RAZR(H154A)–His6–Gp77–HA complex (2.6 mg ml−1) with either 0.5 mg ml−1 His-tag antibody (MA1-2135, Invitrogen) or 0.25 mg ml−1 Strep-tag antibody (MA5-37747, Invitrogen) for 1 h at 4 °C before grid freezing, or with buffer only as a control. All concentrations are reported as final. Samples were prepared by applying 2.5 μl of the mixture onto 200-mesh Quantifoil 2/1 copper grids. The grids were glow-discharged using a GloQube Plus (MiTeGen) at 25 mA for 60 s. Sample-loaded grids were then blotted for 4 s with a blot force of +4 at 6 °C and 100% relative humidity using a FEI Vitrobot Mark IV instrument (Thermo Scientific).
Cryo-EM data collection
For the co-purified complex of Gp77 and RAZR(H154A), a total of 12,965 movies without stage tilt (0° tilt) and 8,911 movies with a 30° stage tilt were collected using EPU (v2.12.1; Thermo Fisher Scientific) on a Titan Krios G3i microscope (Thermo Fisher Scientific). The microscope was operated at an acceleration voltage of 300 kV with a magnification of ×130,000, and data were recorded in super-resolution mode on a K3 detector (Gatan) at a pixel size of 0.65 Å (pre-binned by 2). Each movie consisted of 40 frames and was collected within a defocus range of −0.25 to −1.75 µm for the 0° tilt data and −0.75 to −2.5 µm for the 30° tilt data. The total electron dose per specimen was 47.62 e− Å−2 for the 0° tilt data and 47.96 e− Å−2 for the 30° tilt data.
For in vitro-reconstituted complex of RAZR–His6 and Gp77–His6, a total of 11,152 movies were collected using EPU (v2.12.1; Thermo Fisher Scientific) on a Titan Krios G3i microscope (Thermo Fisher Scientific). The microscope was operated at an acceleration voltage of 300 kV, with a magnification of ×130,000. Data were recorded in super-resolution mode on a K3 detector (Gatan) at a pixel size of 0.65 Å (pre-binned by 2). Each movie consisted of 40 frames and was collected within a defocus range of −0.5 to −2.0 µm. The total electron dose per specimen was 45.7 e− Å−2.
For Gp77–His6, 16,605 movies were collected using EPU (v2.12.1; Thermo Fisher Scientific) on a Titan Krios G3i microscope (Thermo Fisher Scientific). The microscope was operated at an acceleration voltage of 300 kV, with a magnification of ×130,000. Data were recorded in super-resolution mode on a K3 detector (Gatan) at a pixel size of 0.65 Å (pre-binned by 2). Each movie consisted of 40 frames and was collected within a defocus range of −0.75 to −2.25 µm. The total electron dose per specimen was 46.2 e− Å−2.
For RAZR–His6, 19 micrographs were collected using EPU (v2.12.1; Thermo Fisher Scientific) on a Talos Arctica G2 microscope (Thermo Fisher Scientific). The microscope was operated at an acceleration voltage of 200 kV, with a magnification of ×73,000. Data were recorded on a Falcon 3EC detector at a pixel size of 1.95 Å and a defocus of −5 µm during sample screening.
For the co-purified complex of Gp8T7–Flag and RAZR(H154A)–His6, a total of 4,139 movies were collected using EPU (v2.12.1; Thermo Fisher Scientific) on a Titan Krios G3i microscope (Thermo Fisher Scientific). The microscope was operated at an acceleration voltage of 300 kV, with a magnification of ×130,000. Data were recorded in super-resolution mode on a K3 detector (Gatan) at a pixel size of 0.65 Å (pre-binned by 2). Each movie consisted of 40 frames and was collected within a defocus range of −0.75 to −2.25 µm. The total electron dose per specimen was 48.53 e− Å−2.
For RAZR(H154A)–His6–Gp77 samples incubated with either anti-His antibody or buffer-only control, 2,000 movies were collected for each condition using EPU (v.3.11.0) on a Titan Krios G3 operated at 300 kV. For the RAZR(H154A)–His6–Gp77 samples incubated with anti-Strep tag II antibody, 1,000 movies were collected under otherwise identical conditions. Data were acquired at a nominal magnification of ×130,000 with multiple images per hole and recorded on a Falcon 4 detector, with an effective pixel size of 0.776 Å. Each movie consisted of 50 frames, with a defocus range of −0.4 to −2 μm and a total exposure of 52.48 e− Å−2.
Cryo-EM pre-processing and particle picking
Data processing was performed in cryoSPARC41 (v4.5.3 and v4.7.0) using default parameters unless otherwise noted. For data #1 (0° tilt data) of the co-purified complex of Gp77–RAZR(H154A) (Supplementary Fig. 2), 12,965 raw movies were pre-processed using ‘Patch Motion Correction’ and ‘Patch CTF Estimation’. Visual inspection revealed that the particles predominantly adopted ‘top’ views. In the cryoSPARC live session, 14 2D classes were selected and used for ‘Template Picker’ (particle diameter of 170 Å). Particles were extracted (box size of 520 × 520 pixels, Fourier cropped to 256). After two rounds of 2D classification, 1,015,530 particles were selected and extracted (box size of 720 × 720 pixels, Fourier cropped to 360), followed by three additional rounds of 2D classification. A final stack of 793,658 particles was selected.
For data #2 (30° tilt data) of the co-purified complex of Gp77 and RAZR(H154A) (Supplementary Fig. 2), 8,911 raw movies were pre-processed using Patch Motion Correction and Patch CTF Estimation. Two 2D classes from the 0° tilt data (representing particles with one and three rings) were selected for use in the ‘Template Picker’ (particle diameter of 250 Å). Particles were extracted (box size of 800 × 800 pixels, Fourier cropped to 360). After two rounds of 2D classification, 33,400 particles were selected as a preliminary stack.
For the in vitro-reconstituted Gp77–RAZR complex (Supplementary Fig. 3), 11,152 raw movies were pre-processed with Patch Motion Correction and Patch CTF Estimation. A total of 1,796 particles were manually picked using the Manual Picker utility. After 2D classification, two classes corresponding to top and side views (1,470 particles) were selected and used for training with the ‘Topaz Train’ 42 in cryoSPARC, followed by ‘Topaz Extract’. Particles (n = 195,741) were extracted (box size of 720 × 720 pixels, Fourier cropped to 360), followed by three rounds of 2D classification. A final stack of 42,191 particles was selected as a preliminary stack.
Ab initio reconstruction, global refinement and model building
For the co-purified complex of Gp77 and RAZR(H154A), ab initio reconstruction was performed using two classes and C1 symmetry. On the basis of 2D class averages, the presence of higher-order symmetry (C12 or C24) was evident in both the inner and the outer rings. Homogeneous refinement was carried out using C24 symmetry, followed by heterogeneous refinement (C24) using two classes. Particles (n = 238,489) were selected and further refined using homogeneous refinement (C24).
We suspected that there might be a discrepancy in symmetry within the inner and outer ring, therefore homogenous refinement was tested with C12 symmetry. Duplicate particles were removed using the ‘Remove Duplicates’ utility, and the data were subjected to non-uniform refinement (C6 symmetry).
Particles (n = 33,400) from the 30° tilt dataset were used for ab initio reconstruction (C1), followed by homogeneous refinement (C1), and another round of homogeneous refinement using C12 symmetry. These particles were then extracted (box size of 720 × 720 pixels, Fourier cropped to 360) and combined with the 0° tilt data for further homogeneous refinement (C1). After heterogeneous refinement (C1), 234,639 particles were selected for non-uniform refinement with C12 symmetry.
A mask containing both the inner and the outer rings was generated using ChimeraX43 (threshold of 0.2, soft pad 15) for local refinement. The particles were then subjected to global and local CTF refinement, followed by another round of local refinement and the application of the ‘remove duplicates’ utility. Subsequently, another round of local refinement and remove duplicates was performed, followed by both global and local CTF refinement. Finally, the particles underwent a final round of local refinement (C12 symmetry) using the previous mask, resulting in a GSFSC (gold-standard Fourier shell correlation) of approximately 3.4 Å map with a sphericity score of 0.79 (out of 1; Extended Data Fig. 4a–c).
For the in vitro-reconstituted Gp77–RAZR complex, ab initio reconstruction was first performed using a single class with C12 symmetry, followed by non-uniform refinement (C12). Particles were then re-extracted (box size of 1,000 × 1,000 pixels, Fourier cropped to 440) and subjected to an additional round of 2D classification. A second ab initio reconstruction was carried out using two classes with the ‘ab initio reconstruction, high symmetry’ utility (C12 symmetry). Particles corresponding to the volume representing the intact complex were selected for non-uniform refinement (C12). After global and local CTF refinement, another round of non-uniform refinement (C12) was performed. To further improve the model, symmetry expansion (C12) was applied, followed by 3D classification with a mask containing one copy of Gp77 and two copies of RAZR, without pose refinement (10 classes). A class exhibiting well-defined density for the full complex was selected and used as the initial model for a subsequent non-uniform refinement of the original particle stack (before symmetry expansion), following local CTF refinement. This procedure yielded a final map at approximately 3.4 Å resolution (GSFSC) (Extended Data Fig. 4d–f and Supplementary Fig. 3a) with a sphericity score of 0.94 (out of 1). To improve the resolution of Gp77, signal subtraction was performed using two masks: one containing only the inner ring and another containing only the outer rings. To prepare particles for signal subtraction, local refinement was carried out with a mask applied to the outer ring to improve alignment of particles on the outer ring features. Signal subtraction was then performed using the outer ring mask. The resulting signal-subtracted particle images were used for homogeneous reconstruction with a mask applied to the central ring, followed by local refinement with C24 symmetry. This yielded a Gp77-only map at a 3.1 Å GSFSC resolution (Extended Data Fig. 5a–c and Supplementary Fig. 3b) with a sphericity score of 0.94 (out of 1).
To further improve the resolution of the RAZR ZFD (central ring), focused refinement was performed on the final particle stack (28,190) using a mask that included only the central ring, yielding a focused-refined map at a GSFSC resolution of 3.5 Å (Extended Data Fig. 5d–f and Supplementary Fig. 3c) with a sphericity score of 0.75 (out of 1). Applying the same procedure to the outermost ring did not result in improved resolution.
For atomic model generation, only the in vitro-reconstituted complex maps were used, as these showed reduced anisotropy compared with the co-purified map. The Gp77-focused map enabled docking of residues 1–125 into the inner ring using ‘phenix.local_em_fitting’ in ChimeraX, followed by manual adjustments in Coot and subsequent refinement in Phenix. The final modelled residue indicated that the remaining C-terminal domain—predicted to form three α-helices followed by a long unstructured region—is flexibly positioned at the top of the assembly, corresponding to a low-resolution density forming an additional ring-like feature (Extended Data Fig. 6b). The accuracy of the Gp77 model (1–125) was supported by well-resolved side-chain densities for bulky residues (Extended Data Fig. 6a; map–model correlation coefficient (mask) of 0.71; Q score44 (global/expected) of 0.48/0.56).
The central ring was modelled using an AlphaFold prediction of the RAZR ZFD, fitted with phenix.local_em_fitting45 and refined through manual rebuilding in Coot and Phenix. A continuous density bridging the ZFD and HEPN domains clearly delineated the interdomain loop, suggesting that the HEPN domain occupies the outermost ring, which was resolved at substantially lower resolution relative to the Gp77 core and the RAZR ZFD domain (Extended Data Fig. 4). Docking of the AlphaFold-predicted HEPN domain into this region enabled modelling of RAZR residues 3–179. The remaining portion of the HEPN domain, which could not be confidently placed, is predicted to adopt a helix–loop–helix motif that may flexibly position between subunits. After placement of the AlphaFold-predicted HEPN domain, the entire model was further refined in Phenix to reduce steric clashes and improve overall geometry (map–model CC (mask) of 0.75; overall quality control score (global/expected) of 0.41/0.48; RAZR(ZFD) Q score of 0.42; and RAZR(HEPN) Q score of 0.13).
The 270 Å diameter of the ring reported for the in vitro-reconstituted RAZR–Gp77 complex was a distance measured between the Cα atoms of Q140 in chains OA and OM, and the diameter of approximately 170 Å in Gp8T7 was measured between Cα Q201 of chains A and G (PDB ID: 7EY8) that is almost identical to that of the Gp77 ring in our cryo-EM structure (measured between Cα S31 of chains IA and IM).
Model building was performed using ChimeraX43 (v1.6), Coot46 (v0.9.4) and Phenix47 (v1.21.2-5419). Final maps were sharpened using cryoSPARC. Local resolution was estimated using MonoRes48 within cryoSPARC; angular Fourier shell correlations were calculated using the 3DFSC server; and Q scores were calculated using a ChimeraX Q score plugin44.
Mass photometry
Purified RAZR(H154A)–His6–Gp77–HA or Gp77–His6 samples were diluted to 50 nM (as a complex) in a buffer containing 50 mM Tris-HCl pH 8.0, 150 mM NaCl, 10 μM ZnCl2 and 1 mM dithiothreitol. Of each protein, 2 μl was added to 18 μl of buffer in the well, and measured using a Refeyn TwoMP mass photometer (Refeyn) with a data acquisition time of 60 s. Data were acquired by AcquireMP and analysed by DiscoverMP software. The recorded events were fitted to Gaussian distributions, and masses were calculated by applying calibrations performed with BSA (66 kDa) and thyroglobulin (660 kDa) in the same buffer. Each sample was measured independently two times as replicates.
Western blot of RAZR expression levels
Single colonies of E. coli MG1655 pLAND-razr–Flag (WT or mutant variants) were grown overnight in LB. Overnight cultures were back diluted to OD600 = 0.05 in 5 ml fresh LB and grown to OD600 = 0.3 at 37 °C. OD600 was measured, and 3 ml of cells was pelleted at 6,000g for 10 min with OD600 normalized. Supernatant was removed and pellets were resuspended in 1× Laemmli sample buffer (Bio-Rad) supplemented with 2-mercaptoethanol. Samples were then boiled at 95 °C for 15 min and analysed by 4–20% SDS–PAGE and transferred to a 0.2-μm PVDF membrane. Anti-Flag antibody (#14793, Cell Signaling Technology) was used at a final concentration of 1:1,000, and SuperSignal West Femto Maximum Sensitivity Substrate (Thermo Fisher) was used to develop the blots. Blots were imaged by a ChemiDoc Imaging system (Bio-Rad). Blots were stained with Coomassie stain and imaged as loading control. The image shown is a representative of two independent biological replicates.
Incorporation assays
Incorporation assays were performed similarly to those previously described40. For co-producing RAZR and Gp77, single colonies of E. coli MG1655 containing pLAND-razr and pBAD33-gp77 or corresponding empty vectors were grown overnight in M9–glucose. Overnight cultures were back diluted to OD600 = 0.05 in 25 ml M9–glucose and grown to OD600 of approximately 0.3 at 37 °C. Cells were pelleted at 4,000g for 5 min at 4 °C and washed once with M9 (no glucose), and then back diluted to OD600 = 0.1 in 15 ml M9 (no glucose) and recovered for 45 min at 37 °C. At the beginning of the experiment, cells were induced with 0.2% arabinose. At the indicated time points (0, 10, 20, 30 and 40 min), OD600 was measured and an aliquot of 250 μl of cells was transferred to microcentrifuge tube containing [methyl-3H]-thymidine (Revvity; 40 μCi ml−1) for replication measurements, [5,6-3H]-uridine (Revvity; 4 μCi ml−1) for transcription measurements or EasyTag EXPRESS-35S Protein Labeling Mix, [35S] (Revvity; 22 μCi ml−1) for translation measurements. Tubes were incubated at 37 °C for 2 min, then quenched by addition of non-radioactive thymidine (1.5 mM), uridine (1.5 mM) or cysteine and methionine (15 mM each) and incubated for an additional 2 min. Samples were then added to ice-cold trichloroacetic acid (TCA) (10% w/v) and incubated at least 30 min on ice to allow for precipitation. Resulting samples were vacuum filtered onto a glass microfibre filter (1820-024, Whatman) that had been pre-wetted with 5% w/v TCA. Filters were washed with 35× volume of 5% w/v TCA, then with 5× volume of 100% ethanol. Air-dried filters were placed in tubes with scintillation fluid and measured in a scintillation counter (PerkinElmer). Counts per million was normalized to OD600 and percent incorporation at each time point was calculated by normalizing to T = 0. Data reported are the individual data points from two (transcription or translation) or four (replication) independent biological replicates.
Cell-free translation
Experiments with PURExpress in vitro protein synthesis kit (E6800, NEB) were performed as per the manufacturer’s instructions. All reactions were supplemented with 0.8 U µl−1 RNase Inhibitor Murine (M0314S, NEB). Purified His6-tagged RAZR protein and purified His6-tagged Gp77 protein were added to the 15 µl reaction at a final concentration of 1 µM each (as monomers). A template plasmid encoding the control protein DHFR was used at 5 ng µl−1. The reactions were incubated at 37 °C for 2 h, and 2 µl of each reaction was mixed with 10 µl of 1× Laemmli sample buffer (Bio-Rad) supplemented with 2-mercaptoethanol. The mixtures were boiled for 5 min at 95 °C and analysed by 12% SDS–PAGE. The gels were stained with Coomassie stain and imaged by a ChemiDoc Imaging system (Bio-Rad). Images shown are representatives of three independent biological replicates.
RNA extraction from cells expressing RAZR and Gp77
Single colonies of E. coli MG1655 containing pLAND-razr and pBAD33-gp77 or the corresponding empty vector were grown overnight in M9–glucose. Overnight cultures were back diluted to OD600 = 0.05 in 20 ml M9–glucose and grown to OD600 of approximately 0.3 at 37 °C. Cells were pelleted at 4,000g for 5 min at 4 °C and washed once with M9 (no glucose), and then back diluted to OD600 = 0.1 in 15 ml M9 (no glucose) and recovered for 45 min at 37 °C. At the beginning of the experiment, cells were induced with 0.2% arabinose. At the indicated time points (0, 30 and 60 min), cells were harvested by adding 900 µl of culture to 100 µl of stop solution (95% ethanol and 5% acid-buffered phenol) on ice and spinning at 13,000g for 30 s. Supernatants were removed, and pellets were flash frozen in liquid nitrogen. To extract RNAs, 400 µl of TRIzol (Invitrogen) preheated to 65 °C was added to each pellet. Resuspended pellets were incubated at 65 °C for 10 min at 2,000 rpm in a thermomixer, flash frozen in liquid nitrogen for 10 min and thawed to room temperature. Samples were spun at 21,000g for 5 min at 4 °C, and supernatants were mixed with 400 µl of ethanol. RNAs were purified using the Direct-zol RNA Miniprep kit (Zymo Research) per the manufacturer’s instructions with on-column DNase treatment. RNA yield was measured by a Nanodrop spectrophotometer. Of each purified RNA, 80 ng was analysed by a Novex 15% TBE–urea gel (Invitrogen) in 1× TBE buffer and stained with SYBR Gold (Invitrogen). For visualizing rRNAs, 1 µg of purified RNAs were analysed by 1% agarose gel in 1× TAE buffer supplemented with 1% bleach and ethidium bromide. The gels were imaged by a ChemiDoc Imaging system (Bio-Rad). Images shown are representatives of two independent biological replicates.
RNA extraction following phage infection
Single colonies of E. coli MG1655 containing pLAND-razr or an empty vector were grown overnight in LB medium. Overnight cultures were back diluted to OD600 = 0.05 in 25 ml LB and grown to OD600 of approximately 0.2 at 25 °C. At the beginning of the experiment, cells were infected with phage T7 at multiplicity of infection = 10. At the indicated time points (0, 10, 20, 30 and 40 min), 500 µl of cells were mixed with 500 µl of preheated lysis buffer containing 2% SDS and 4 mM ETDA pH 8.0. Samples were incubated at 100 °C for 5 min, flash frozen in liquid nitrogen and thawed to room temperature. Of acid-buffered phenol (pH 4.5, Sigma) preheated to 67 °C, 1 ml was added to each sample and incubated at 67 °C for 2 min. Samples were pelleted at 21,000g for 10 min, and the aqueous layer was collected and repeated with another hot phenol extraction. A third round of extraction was done with 1 ml acid-buffered phenol chloroform (pH 4.5, Ambion). RNA was precipitated with 1× volume isopropanol, 0.1× volume 3 M NaOAc pH 5.5 and 0.01× volume GlycoBlue on ice for 1 h. RNAs were pelleted at 21,000g for 30 min at 4 °C. Pellets were washed twice with 500 µl of ice-cold 70% ethanol, air dried and resuspended in 90 µl nuclease-free water. To remove DNA, samples were treated with 10 µl of 10× Turbo DNase buffer (Invitrogen) and 2 µl Turbo DNase I (Invitrogen) for 20 min at 37 °C. An additional 2 µl of Turbo DNase I was added, and samples were incubated for another 20 min at 37 °C. Samples were then mixed with 96 µl of water, extracted with 200 µl of acid-buffered phenol chloroform (pH 4.5, Ambion) and precipitated for 1 h at −20 °C with 3× volume ice-cold ethanol, 0.1× volume 3 M NaOAc pH 5.5 and 0.01× volume GlycoBlue. RNAs were pelleted, washed and resuspended as described above. RNAs were analysed by a Novex 15% TBE–urea gel and a 1% agarose gel with 1% bleach as described above.
In vitro cleavage assays
For tRNA cleavage assays, purified RAZR–His6 and Gp77–His6 were added to a 5 µl reaction at a final concentration of 1.2 µM each (as monomers) and mixed with 180 ng of extracted bulk E. coli tRNAs in cleavage buffer (50 mM Tris-HCl pH 8.0, 150 mM NaCl, 30 mM KCl, 7 mM MgCl2, 10 μM ZnCl2 and 1 mM dithiothreitol). The co-purified RAZR(H154A)–His6–Gp77–HA complex or the in vitro-reconstituted complex RAZR–His6–Gp77–His6 were each added at a final concentration of 2.5 µM (as monomers). After incubation at 37 °C for 1 h, 2.5 µl of each reaction was mixed with 2.5 µl of Novex TBE–urea sample buffer (Invitrogen) and analysed by a Novex 15% TBE–urea gel (Invitrogen) in 1× TBE buffer and stained with SYBR Gold stain.
For rRNA cleavage assays, purified RAZR–His6 and Gp77–His6 were added to a 15 µl reaction at a final concentration of 1.2 µM each and mixed with E. coli 70S ribosomes (NEB) at a final concentration of 0.44 µM in cleavage buffer. The co-purified RAZR(H154A)–His6–Gp77–HA complex or the in vitro-reconstituted complex RAZR–His6–Gp77–His6 were each added at a final concentration of 2.5 µM (as monomers). After incubation at 37 °C for 1 h, RNAs were extracted by acid-buffered phenol chloroform and precipitated as described above. Of purified RNAs, 1 µg were analysed by 1% agarose gel in 1× TAE buffer supplemented with 1% bleach and ethidium bromide.
For cleavage assays of RNA or DNA oligos (single stranded or double stranded), purified RAZR–His6 and Gp77–His6 were added to a 15 µl reaction at a final concentration of 1.2 µM each (as monomers) and mixed with 500 ng each corresponding oligo in cleavage buffer. After incubation at 37 °C for 1 h, 3 µl of each reaction was analysed by a Novex 15% TBE–urea gel (Invitrogen) as described above. Oligo sequences are listed in Supplementary Table 4 (TZ-51 to TZ-54).
For mRNA cleavage assays, mRNA substrates were in vitro transcribed using MEGAscript T7 transcription kit (Invitrogen) per the manufacturer’s instructions. In brief, each coding sequence was PCR amplified from genomic DNA of E. coli MG1655 with a T7 promoter introduced directly upstream of it using primers TZ-55 to TZ-70. Amplified DNA was purified with DNA clean & concentrator kit (Zymo Research) and used as template. In vitro transcription reactions were incubated at 37 °C for 4 h, and treated with 1 µl Turbo DNase at 37 °C for 15 min. Transcribed mRNAs were purified by phenol-chloroform extraction and ethanol precipitation as described above. For cleavage assays, purified RAZR–His6 and Gp77–His6 were added to a 15 µl reaction at a final concentration of 1.2 µM each (as monomers) and mixed with 2 µg of each mRNA substrate in cleavage buffer. After incubation at 37 °C for 1 h, 0.75 µl of each reaction was analysed by a Novex 15% TBE–urea gel (Invitrogen) as described above.
RNA-seq sample preparation and analysis
For co-producing Gp77 and RAZR, RNAs were extracted as described above from E. coli MG1655 cells containing pLAND-razr and pBAD33-gp77 or the corresponding empty vector, after inducing with arabinose for 0, 10 and 30 min. RNAs were purified using the Direct-zol RNA Miniprep kit (Zymo Research) per the manufacturer’s instructions and eluted in 90 µl nuclease-free water. To remove DNA, samples were treated with 10 µl of 10× Turbo DNase buffer (Invitrogen) and 2 µl Turbo DNase I (Invitrogen) for 20 min at 37 °C. An additional 2 µl of Turbo DNase I was added, and samples were incubated for another 20 min at 37 °C. Samples were then mixed with 96 µl of water, extracted with 200 µl of acid-buffered phenol chloroform (pH 4.5, Ambion), and precipitated for 1 h at −20 °C with 3× volume ice-cold ethanol, 0.1× volume 3 M NaOAc pH 5.5 and 0.01× volume GlycoBlue. RNAs were pelleted, washed and resuspended as described above. RNA quality was assessed by a Novex 6% TBE–urea gel, and yield was measured by a Nanodrop spectrophotometer. For RNA-seq during T7 infection, RNAs were extracted and processed as described above.
To prepare libraries for RNA-seq, rRNA was removed using a developed E. coli rRNA depletion kit49, with 1.7 µg of total RNAs as input. rRNA-depleted samples were further purified using RNA Clean and Concentrator kit (Zymo Research). Libraries were generated using NEBNext Ultra II RNA Library Prep Kit for Illumina (NEB) following the manufacturer’s instructions for purified mRNA or rRNA-depleted RNA. Libraries were sequenced on an Illumina NextSeq at the MIT BioMicro Center. Two biological replicates were harvested and sequenced independently.
RNA-seq data were analysed similarly to that previously described30. FASTQ files for each sample were trimmed using cutadapt (v1.15)50 and then mapped to the MG1655 genome (NC_00913.2) and the T7 genome (V01146), or the consensus map of rRNA loci as previously described30 using bowtie2 (v2.3.4.1)51 with the following arguments: –D 20, –I 40, –X 300, –R 3, –N 0, –L 20, –i S,1,0.50. Sam files generated from bowtie2 mapping were converted to bam files using samtools (v1.7)52, and then converted to numpy arrays using the genomearray3 Python library30. Gene names and coding regions were extracted from NCBI annotations. For the T7 transcriptome, mRNA positions were extracted based on the T7 promoter and terminator positions from NCBI annotations, similar to that previously described53. For analysis of fragment density across the transcriptome, one count was added for all positions between and including the 5′ and 3′ ends of the reads. To correct for variability in sequencing depth, counts at each position were divided by a sample size factor as previously described for normalization30. In brief, counts in each coding region were summed for each sample, and the geometric mean of these sums was calculated to yield a reference sample. The total counts in each coding region were then normalized by the reference sample, and the median of these ratios was taken as the size factor for that sample. The cleavage ratio at each nucleotide position was calculated as the log2 transformed + Gp77:empty vector ratio (for co-producing Gp77 and RAZR) or the log2 transformed + RAZR:empty vector ratio (for phage infection), and the average of two replicates were taken. Any regions in the empty vector sample that had fewer counts than the expression cut-off of 64 counts (for co-producing Gp77 and RAZR) or 32 counts (for phage infection) were discarded, and minimum cleavage ratio was taken for each coding region.
Homology search and conservation analysis
Homologues of RAZR were identified by ConSurf54 with default settings to search the UniRef90 database, using AlphaFold-predicted RAZR structure as input. Homologues (n = 150) were used to generate the multiple sequence alignment by MAFFT. Conservation scores were calculated using the Bayesian method and default settings.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
