Type III-A CRISPR-Cas Csm Complexes: Assembly, Periodic RNA Cleavage, DNase Activity Regulation, and Autoimmunity.

STAR*METHODS KEY RESOURCES TABLE CONTACT FOR REAGENT AND RESOURCE SHARING

Further information and requests for reagents could be directed to, and will be fulfilled by Lead Contact Dinshaw Patel ( pateld@mskcc.org ) METHODS DETAILS Protein Expression and Purification Proteins were expressed and purified as described previously ( Park et al., 2017 ) with some modifications. The full-length Thermococcus onnurineus csm genes, csm1, csm2, csm3, csm4, csm5 were synthesized and cloned into different expression vectors. csm1 and csm4 were subcloned into pRSF-Duet-1 vector (Novagen), in which csm1 was attached with N-terminal His6 tag. csm2 and csm3 were cloned into a modified pRSF-Duet-1 vector (Novagen), in which they were attached with N-terminal His6-SUMO tag following an ubiquitin-like protease (ULP1), respectively. csm5 was cloned into pCDF-Duet-1, in which csm5 was attached with C-terminal His10 tag. Csm1-Csm4 subcomplex, Csm2, Csm3 and Csm5 recombinant proteins were overexpressed in Escherichia coli BL21 (DE3) strain by induction with 0.25 mM isopropyl-β-D-1-thiogalactopyranoside (GoldBio) at 16°C for 20 hr. Cells were harvested by centrifugation and resuspended in lysis buffer (20 mM Tris-HCl, pH 8.0, 500 mM NaCl, 5% glycerol, 20 mM imidazole, 7 mM β-mercaptoethanol). The harvested cells that produced Csm1-Csm4 subcompelx were then lysed by the EmulsiFlex-C3 homogenizer (Avestin) and centrifuged at 20,000 rpm for 30 min in a JA-20 fixed angle rotor (Avanti J-E series centrifuge, Beckman Coulter). The supernatant was applied to 5 mL HisTrap Fast flow column (GE Healthcare). The protein was eluted with lysis buffer supplemented with 500 mM imidazole after washing the column with 10 column volumes of lysis buffer and 2 column volumes of lysis buffer supplemented with 40 mM imidazole. The elution fractions were further dialyzed against buffer A (20 mM Tris-HCl, pH 8.0, 100 mM NaCl, 5% glycerol, 7mM β-mercaptoethanol), respectively, and applied on 5 mL HiTrap Q Fast flow column (GE Healthcare). Proteins were eluted by a linear gradient from 100 mM to 1 M NaCl in 20 column volumes, and then concentrated in 10 kDa molecular mass cut-off concentrators (Amicon) before further purification over a Superdex 200 increase 10/300 GL column (GE Healthcare) pre-equilibrated in buffer B (20 mM Tris-HCl, pH 8.0, 250 mM NaCl, 5% glycerol, 2 mM DTT). The recombinant Csm5 was purified by the similar method as above, with a slight difference. The elution from Ni-NTA column were dialyzed against buffer A and loaded on HiTrap Heparin HP column (GE Healthcare). Proteins were eluted by a linear gradient from 100 mM to 1 M NaCl in 20 column volumes, and further purified over a Superdex 200 increase 10/300 GL column (GE Healthcare) pre-equilibrated in buffer B. The recombinant Csm2 was purified by the similar method as that for Csm1-Csm4 subcomplex purification. Elution fractions of Csm2 with His6-SUMO tag from Ni-NTA column were further dialyzed against buffer A overnight at 4°C by adding ULP1 during dialysis to remove His6-SUMO tag. The tag was separated by reloading the samples on Ni-NTA column, with the flow through loaded on 5 mL HiTrap Heparin HP column (GE Healthcare). The protein was eluted by a linear gradient from 100 mM to 1 M NaCl in 20 column volumes, and further purified over a Superdex 200 increase 10/300 GL column (GE Healthcare) pre-equilibrated in buffer B. The recombinant Csm3 was purified by the similar method as that for Csm2 purification, with a slight modification. After dialysis against buffer A with addition of ULP1 to remove His6-SUMO tag. Flow through fractions of Csm3 without His6-SUMO tag from Ni-NTA column were loaded on 5 mL HiTrap Q Fast flow column (GE Healthcare). The protein was eluted by a linear gradient from 100 mM to 1 M NaCl in 20 column volumes, and further purified over a Superdex 200 increase 10/300 GL column (GE Healthcare) pre-equilibrated in buffer B. All mutants were generated by site-directed mutagenesis, and purified by the same methods as above. In Vitro Assembly of Csm crRNA Binary, Csm crRNA -Target RNA and Csm crRNA -Target RNA anti-tag Ternary Complexes crRNAs were synthesized by Integrated DNA Technologies. Based on previous proposed Csm1 1 2 1 3 3 4 1 5 1 stoichiometry ( Park et al., 2017 ), purified Csm1-Csm4 heterodimer, Csm2, Csm3, Csm5 and crRNA were mixed at a molar ration of 1: 1.5: 4: 1.5: 1.3 and then incubate at 60°C for 20 min. The complex was further purified by gel filtration chromatography on a Superdex 200 increase 10/300 GL column pre-equilibrated in buffer B. Fractions contain the Csm crRNA complex were pooled and reloaded on Superdex 200 increase 10/300 GL pre-equilibrated in buffer C (20 mM Tris-HCl, pH 8.8, 250 mM NaCl, 2 mM DTT). We also assembled Csm1-Csm4 heterodimer, Csm2, Csm3, Csm5 and crRNA by mixing at a molar ration of 1: 5: 6: 1.5: 1.3 following the same method as described above. To assemble Csm crRNA -target RNA ternary complex, we mutated Asp36 of Csm3 into alanine to avoid target RNA cleavage. To prevent target RNA-activated non-specific DNase activities, His14 and Asp15 of Csm1 were mutated into Ala and Asn, respectively. Purified Csm1 H14A/D15N -Csm4 heterodimer, Csm2, Csm3 D36A , Csm5 and crRNA were mixed at a molar ration of 1: 1.5: 4: 1.5: 1.3 and then incubate at 60°C for 20 min. Then the target RNA was added into the mixer at a molar ratio of 1.5: 1. After incubation at 60°C for 20 min, the mixture was loaded on a Superdex 200 increase 10/300 GL column pre-equilibrated in buffer B. Fractions contain the Csm crRNA -target RNA complex were pooled and reloaded on Superdex 200 increase 10/300 GL pre-equilibrated in buffer C. Csm crRNA -target RNA ternary anti-tag complex was assembled by the same method as that for Csm crRNA -target RNA ternary complex. we only mutated Asp36 of Csm3 into alanine to avoid target RNA cleavage, and used apoform Csm1 in this complex. All mutants were assembled by similar methods as above. Cryo-EM Sample Preparation and Data Acquisition 3.0 μl of ~0.5 mg/ml purified Csm crRNA binary, Csm crRNA -target RNA and Csm crRNA -target RNA anti-tag ternary complexes were applied onto glow-discharged UltrAuFoil 300 mesh R1.2/1.3 grids (Quantifoil), respectively. Grids were blotted for 2s at ~100% humidity and flash frozen in liquid ethane using an FEI Vitrobot Mark IV. Images were collected on FEI Titan Krios electron microscope operated at an acceleration voltage of 300 kV with a Gatan K2 Summit detector with a 1.089 Å pixel size and 8.0 electrons per pixel per second. The defocus range was set from −0.8 μm to 2.5 μm. Dose-fractionated images were recorded with a per-frame exposure time of 200 ms and a dose of ~1.349 electrons per Å 2 per frame. Total accumulated dose was ~54 electrons per Å 2 .

Image Processing

For the Csm crRNA -target RNA dataset, motion correction was performed with MotionCor2 ( Zheng et al., 2017 ). Contrast transfer function parameters were estimated by Ctffind4 ( Rohou and Grigorieff, 2015 ). All other steps of image processing were performed by RELION 2.1 ( Scheres, 2012 ). Templates for automated particle selection were generated from 2D-averages of ~2,000 manually picked particles. Automated particle selection resulted in 755,189 particles from 1,697 images. After two rounds of 2D classification, a total of 558,984 particles were selected for 3D classification using the initial model generated by RELION as reference. Particles corresponding to the best class with the highest-resolution features were selected and subjected to the second round of 3D classification. Two of 3D classes with good secondary structural features and the corresponding 109,902 particles were polished using RELION particle polishing, yielding an electron microscopy map with stoichiometry 1 1 2 1 3 2 4 1 5 1 with a resolution of 3.1 Å after 3D auto-refinement ( Figure S3 ). Another Class and corresponding 19,111 particles were also polished and refined to get an electron microscopy map with stoichiometry 1 1 2 2 3 3 4 1 5 1 with a resolution of 3.8 Å ( Figure S3 ). The resolution was improved further to 3.6 Å by focused classification together with following 3D refinement and particle polishing. The dataset for Csm crRNA binary complex was processed by the same procedure as above. Briefly, 628,985 particles were autopicked from 1,455 images, 129,536 particles were selected for the final 3D reconstruction after two rounds of 2D and 3D classification, resulting in a Csm crRNA binary complex map with an overall resolution of 3.0 Å ( Figures S6C and S6D ). The dataset for Csm crRNA -target RNA anti-tag ternary complex was processed by the same procedure as outlined above. Briefly, 757,313 particles were autopicked from 1,346 images, 57,170 particles were selected for the final 3D reconstruction after two rounds of 2D and 3D classification, resulting in a Csm crRNA -target RNA anti-tag ternary complex map with an overall resolution of 3.1 Å ( Figures S6E and S6F ). All resolutions were estimated using RELION ‘post-processing’ by applying a soft mask around the protein density and the Fourier shell correlation (FSC) = 0.143 criterion. Local resolution estimates were calculated from two half data maps using ResMap ( Kucukelbir et al., 2014 ). Further details related to data processing and refinement are summarized in Table 1 and Figures S3 and S6 .

Atomic Model Building and Refinement

For the Csm crRNA -RNA ternary complex, the initial models of Csm1, Csm3 and Csm4 were generated by docking the crystal structure of T. onnurineus Csm1 (PDB: 4UW2), homologous M. jannaschii Csm3 and Csm4 ( Mj Csm3-Csm4, PDB: 4QTS) into the cryo-EM density map using UCSF Chimera ( Pettersen et al., 2004 ), respectively. Models for Csm2 and Csm5 were first generated by SWISS-MODEL online server. Then all docked models were manually rebuilt in COOT ( Emsley et al., 2010 ) to fit the density map. Other parts of the complex were built based on the bulky side chains to register the sequence. Residues in parts of Csm2 and Csm5 were modeled as poly-alanines due to the poor side-chain densities, densities for residues 134–169 in Csm5 is not sufficient to define their sequence order confidently. For the Csm crRNA binary, Csm crRNA -target RNA with stoichiometry 1 1 2 2 3 3 4 1 5 1 and Csm crRNA -target RNA anti-tag ternary complex, the structure of Csm crRNA -target RNA ternary complex was docked into the cryo-EM density map using UCSF Chimera ( Pettersen et al., 2004 ) and then manually rebuilt in COOT ( Emsley et al., 2010 ). Densities for Csm5 in the Csm crRNA -target RNA with stoichiometry 1 1 2 2 3 3 4 1 5 1 are not well traceable, through the existing EM density allow us to fit the atomic model of Csm5 into this position at the head of the complex. Also, densities for both Csm2 and Csm5 in the Csm crRNA -target RNA anti-tag are not well defined. Based on the existing EM density, we docked the respective atomic models into the map. All models were refined against summed maps using phenix.real_space_refine ( Adams et al., 2010 ) by applying geometric and secondary structure restraints. All figures were prepared by PyMol ( http://www.pymol.org ) or Chimera ( Pettersen et al., 2004 ). The statistics for data collection and model refinement are shown in Table 1 . Crystallization, Data Collection, and Structure Determination Crystals of Csm1-Csm4 subcomplex were grown at 20°C using hanging-drop vapor diffusion by mixing 1 μl protein solution with 1 μl reservoir solution containing 0.1 M phosphate-citrate pH 4.2, 5% PEG3000, 25% 1,2-propanediol, and 10% glycerol. Data were collected at 100 K at the Advanced Photo Source (APS) at the Argonne National Laboratory and processed by HKL suite ( Otwinowski and Minor, 1997 ). Structure was solved by the molecular replacement method using PHENIX. The refinement was performed with REFMAC5 ( Murshudov et al., 1997 ) and COOT ( Emsley et al., 2010 ). The statistics of the diffraction data are summarized in Table S1. All structure figures were prepared with PyMOL ( http://www.pymol.org/ ).

RNA Cleavage Assay

In vitro RNA cleavage assay was performed by incubating 50 nM Csm crRNA complex and 50 nM 5’-FAM labeled target RNA in 50 μL reaction buffer composed of 20 mM Tris–HCl, pH 8.0, 100 mM KCl, 100 mM NaCl, 5 mM MnCl 2 and 3% glycerol at 55°C for 15 min, unless stated otherwise. Reactions were stopped by adding 2 × formamide loading buffer (90% formamide, 0.025% bromophenol blue, 0.025% xylene cyanol, 0.025% SDS, 5 mM DTT) followed by heating at 95°C for 5 min. Samples were analyzed on 15% polyacrylamide denaturing 8 M Urea gel and visualized by Typhoon 9500 scanner. ΦX174 Virion ssDNA Cleavage Assay The reactions were performed by mixing 5 nM ΦX174 Virion ssDNA and 200 nM Csm crRNA complex with or without 5 nM of various target RNAs in reaction buffer (30 mM Tris–HCl, pH 8.0, 100 mM KCl, 100 mM NaCl and 3% glycerol) supplemented with 5 mM MnCl 2 at 55°C for 5 min, unless stated otherwise. Reaction were quenched by adding 2 × loading buffer (4 mM Tris-HCl (pH 7.6), 0.01% bromophenol blue, 0.01% xylene cyanol FF, 20% glycerol, 20 mM EDTA), followed by incubation for 5 min at 95°C. The samples were separated on a 1% agarose gel, stained with GelRed (Biotium) for visualization. 90-nt ssDNA Cleavage Assay The Csm crRNA complex (10 nM) was mixed with or without various RNA substrates (10 nM) in reaction buffer (30 mM Tris–HCl, pH 8.0, 100 mM KCl, 100 mM NaCl and 3% glycerol). Radiolabeled ssDNA substrate was then added to a final concentration of 100 nM. The reaction was initiated with 5 mM MnCl 2 and incubated at 55°C for 20 minutes. The reaction was then quenched with 2 × formamide buffer and heated to 95°C for 10 minutes. Samples were separated on a 12% Tris/Borate/EDTA gel with 7 M Urea and visualized by phosphorimaging.

QUANTIFICATION AND STATISTICAL ANALYSIS

In vitro cleavage experiments were repeated at least three times, and representative results were shown.

DATA AVAILABILITY

The atomic coordinates have been deposited in the Protein Data Bank with accession number PDB: 6MUU (Csm1 1 2 1 3 2 4 1 5 1 crRNA ), PDB: 6MUR (Csm1 1 2 1 3 2 4 1 5 1 crRNA -target RNA ternary complex), PDB: 6MUS (Csm1 1 2 2 3 3 4 1 5 1 crRNA -target RNA ternary complex), PDB: 6MUT (Csm1 1 2 1 3 2 4 1 5 1 crRNA -target RNA anti-tag ternary complex), and PDB: 6MUA (Csm1-Csm4 subcomplex). The cryo-EM density maps have been deposited in the Electron Microscopy Data Bank under accession number EMDB: 9256 (Csm1 1 2 1 3 2 4 1 5 1 crRNA ), PDB: 9253 (Csm1 1 2 1 3 2 4 1 5 1 crRNA -target RNA ternary complex), EMDB: 9254 (Csm1 1 2 2 3 3 4 1 5 1 crRNA -target RNA ternary complex), and EMDB: 9255 (Csm1 1 2 1 3 2 4 1 5 1 crRNA -target RNA anti-tag ternary complex

METHODS DETAILS Protein Expression and Purification

Proteins were expressed and purified as described previously ( Park et al., 2017 ) with some modifications. The full-length Thermococcus onnurineus csm genes, csm1, csm2, csm3, csm4, csm5 were synthesized and cloned into different expression vectors. csm1 and csm4 were subcloned into pRSF-Duet-1 vector (Novagen), in which csm1 was attached with N-terminal His6 tag. csm2 and csm3 were cloned into a modified pRSF-Duet-1 vector (Novagen), in which they were attached with N-terminal His6-SUMO tag following an ubiquitin-like protease (ULP1), respectively. csm5 was cloned into pCDF-Duet-1, in which csm5 was attached with C-terminal His10 tag. Csm1-Csm4 subcomplex, Csm2, Csm3 and Csm5 recombinant proteins were overexpressed in Escherichia coli BL21 (DE3) strain by induction with 0.25 mM isopropyl-β-D-1-thiogalactopyranoside (GoldBio) at 16°C for 20 hr. Cells were harvested by centrifugation and resuspended in lysis buffer (20 mM Tris-HCl, pH 8.0, 500 mM NaCl, 5% glycerol, 20 mM imidazole, 7 mM β-mercaptoethanol). The harvested cells that produced Csm1-Csm4 subcompelx were then lysed by the EmulsiFlex-C3 homogenizer (Avestin) and centrifuged at 20,000 rpm for 30 min in a JA-20 fixed angle rotor (Avanti J-E series centrifuge, Beckman Coulter). The supernatant was applied to 5 mL HisTrap Fast flow column (GE Healthcare). The protein was eluted with lysis buffer supplemented with 500 mM imidazole after washing the column with 10 column volumes of lysis buffer and 2 column volumes of lysis buffer supplemented with 40 mM imidazole. The elution fractions were further dialyzed against buffer A (20 mM Tris-HCl, pH 8.0, 100 mM NaCl, 5% glycerol, 7mM β-mercaptoethanol), respectively, and applied on 5 mL HiTrap Q Fast flow column (GE Healthcare). Proteins were eluted by a linear gradient from 100 mM to 1 M NaCl in 20 column volumes, and then concentrated in 10 kDa molecular mass cut-off concentrators (Amicon) before further purification over a Superdex 200 increase 10/300 GL column (GE Healthcare) pre-equilibrated in buffer B (20 mM Tris-HCl, pH 8.0, 250 mM NaCl, 5% glycerol, 2 mM DTT). The recombinant Csm5 was purified by the similar method as above, with a slight difference. The elution from Ni-NTA column were dialyzed against buffer A and loaded on HiTrap Heparin HP column (GE Healthcare). Proteins were eluted by a linear gradient from 100 mM to 1 M NaCl in 20 column volumes, and further purified over a Superdex 200 increase 10/300 GL column (GE Healthcare) pre-equilibrated in buffer B. The recombinant Csm2 was purified by the similar method as that for Csm1-Csm4 subcomplex purification. Elution fractions of Csm2 with His6-SUMO tag from Ni-NTA column were further dialyzed against buffer A overnight at 4°C by adding ULP1 during dialysis to remove His6-SUMO tag. The tag was separated by reloading the samples on Ni-NTA column, with the flow through loaded on 5 mL HiTrap Heparin HP column (GE Healthcare). The protein was eluted by a linear gradient from 100 mM to 1 M NaCl in 20 column volumes, and further purified over a Superdex 200 increase 10/300 GL column (GE Healthcare) pre-equilibrated in buffer B. The recombinant Csm3 was purified by the similar method as that for Csm2 purification, with a slight modification. After dialysis against buffer A with addition of ULP1 to remove His6-SUMO tag. Flow through fractions of Csm3 without His6-SUMO tag from Ni-NTA column were loaded on 5 mL HiTrap Q Fast flow column (GE Healthcare). The protein was eluted by a linear gradient from 100 mM to 1 M NaCl in 20 column volumes, and further purified over a Superdex 200 increase 10/300 GL column (GE Healthcare) pre-equilibrated in buffer B. All mutants were generated by site-directed mutagenesis, and purified by the same methods as above. In Vitro Assembly of Csm crRNA Binary, Csm crRNA -Target RNA and Csm crRNA -Target RNA anti-tag Ternary Complexes crRNAs were synthesized by Integrated DNA Technologies. Based on previous proposed Csm1 1 2 1 3 3 4 1 5 1 stoichiometry ( Park et al., 2017 ), purified Csm1-Csm4 heterodimer, Csm2, Csm3, Csm5 and crRNA were mixed at a molar ration of 1: 1.5: 4: 1.5: 1.3 and then incubate at 60°C for 20 min. The complex was further purified by gel filtration chromatography on a Superdex 200 increase 10/300 GL column pre-equilibrated in buffer B. Fractions contain the Csm crRNA complex were pooled and reloaded on Superdex 200 increase 10/300 GL pre-equilibrated in buffer C (20 mM Tris-HCl, pH 8.8, 250 mM NaCl, 2 mM DTT). We also assembled Csm1-Csm4 heterodimer, Csm2, Csm3, Csm5 and crRNA by mixing at a molar ration of 1: 5: 6: 1.5: 1.3 following the same method as described above. To assemble Csm crRNA -target RNA ternary complex, we mutated Asp36 of Csm3 into alanine to avoid target RNA cleavage. To prevent target RNA-activated non-specific DNase activities, His14 and Asp15 of Csm1 were mutated into Ala and Asn, respectively. Purified Csm1 H14A/D15N -Csm4 heterodimer, Csm2, Csm3 D36A , Csm5 and crRNA were mixed at a molar ration of 1: 1.5: 4: 1.5: 1.3 and then incubate at 60°C for 20 min. Then the target RNA was added into the mixer at a molar ratio of 1.5: 1. After incubation at 60°C for 20 min, the mixture was loaded on a Superdex 200 increase 10/300 GL column pre-equilibrated in buffer B. Fractions contain the Csm crRNA -target RNA complex were pooled and reloaded on Superdex 200 increase 10/300 GL pre-equilibrated in buffer C. Csm crRNA -target RNA ternary anti-tag complex was assembled by the same method as that for Csm crRNA -target RNA ternary complex. we only mutated Asp36 of Csm3 into alanine to avoid target RNA cleavage, and used apoform Csm1 in this complex. All mutants were assembled by similar methods as above. Cryo-EM Sample Preparation and Data Acquisition 3.0 μl of ~0.5 mg/ml purified Csm crRNA binary, Csm crRNA -target RNA and Csm crRNA -target RNA anti-tag ternary complexes were applied onto glow-discharged UltrAuFoil 300 mesh R1.2/1.3 grids (Quantifoil), respectively. Grids were blotted for 2s at ~100% humidity and flash frozen in liquid ethane using an FEI Vitrobot Mark IV. Images were collected on FEI Titan Krios electron microscope operated at an acceleration voltage of 300 kV with a Gatan K2 Summit detector with a 1.089 Å pixel size and 8.0 electrons per pixel per second. The defocus range was set from −0.8 μm to 2.5 μm. Dose-fractionated images were recorded with a per-frame exposure time of 200 ms and a dose of ~1.349 electrons per Å 2 per frame. Total accumulated dose was ~54 electrons per Å 2 .

Image Processing

For the Csm crRNA -target RNA dataset, motion correction was performed with MotionCor2 ( Zheng et al., 2017 ). Contrast transfer function parameters were estimated by Ctffind4 ( Rohou and Grigorieff, 2015 ). All other steps of image processing were performed by RELION 2.1 ( Scheres, 2012 ). Templates for automated particle selection were generated from 2D-averages of ~2,000 manually picked particles. Automated particle selection resulted in 755,189 particles from 1,697 images. After two rounds of 2D classification, a total of 558,984 particles were selected for 3D classification using the initial model generated by RELION as reference. Particles corresponding to the best class with the highest-resolution features were selected and subjected to the second round of 3D classification. Two of 3D classes with good secondary structural features and the corresponding 109,902 particles were polished using RELION particle polishing, yielding an electron microscopy map with stoichiometry 1 1 2 1 3 2 4 1 5 1 with a resolution of 3.1 Å after 3D auto-refinement ( Figure S3 ). Another Class and corresponding 19,111 particles were also polished and refined to get an electron microscopy map with stoichiometry 1 1 2 2 3 3 4 1 5 1 with a resolution of 3.8 Å ( Figure S3 ). The resolution was improved further to 3.6 Å by focused classification together with following 3D refinement and particle polishing. The dataset for Csm crRNA binary complex was processed by the same procedure as above. Briefly, 628,985 particles were autopicked from 1,455 images, 129,536 particles were selected for the final 3D reconstruction after two rounds of 2D and 3D classification, resulting in a Csm crRNA binary complex map with an overall resolution of 3.0 Å ( Figures S6C and S6D ). The dataset for Csm crRNA -target RNA anti-tag ternary complex was processed by the same procedure as outlined above. Briefly, 757,313 particles were autopicked from 1,346 images, 57,170 particles were selected for the final 3D reconstruction after two rounds of 2D and 3D classification, resulting in a Csm crRNA -target RNA anti-tag ternary complex map with an overall resolution of 3.1 Å ( Figures S6E and S6F ). All resolutions were estimated using RELION ‘post-processing’ by applying a soft mask around the protein density and the Fourier shell correlation (FSC) = 0.143 criterion. Local resolution estimates were calculated from two half data maps using ResMap ( Kucukelbir et al., 2014 ). Further details related to data processing and refinement are summarized in Table 1 and Figures S3 and S6 .

Atomic Model Building and Refinement

For the Csm crRNA -RNA ternary complex, the initial models of Csm1, Csm3 and Csm4 were generated by docking the crystal structure of T. onnurineus Csm1 (PDB: 4UW2), homologous M. jannaschii Csm3 and Csm4 ( Mj Csm3-Csm4, PDB: 4QTS) into the cryo-EM density map using UCSF Chimera ( Pettersen et al., 2004 ), respectively. Models for Csm2 and Csm5 were first generated by SWISS-MODEL online server. Then all docked models were manually rebuilt in COOT ( Emsley et al., 2010 ) to fit the density map. Other parts of the complex were built based on the bulky side chains to register the sequence. Residues in parts of Csm2 and Csm5 were modeled as poly-alanines due to the poor side-chain densities, densities for residues 134–169 in Csm5 is not sufficient to define their sequence order confidently. For the Csm crRNA binary, Csm crRNA -target RNA with stoichiometry 1 1 2 2 3 3 4 1 5 1 and Csm crRNA -target RNA anti-tag ternary complex, the structure of Csm crRNA -target RNA ternary complex was docked into the cryo-EM density map using UCSF Chimera ( Pettersen et al., 2004 ) and then manually rebuilt in COOT ( Emsley et al., 2010 ). Densities for Csm5 in the Csm crRNA -target RNA with stoichiometry 1 1 2 2 3 3 4 1 5 1 are not well traceable, through the existing EM density allow us to fit the atomic model of Csm5 into this position at the head of the complex. Also, densities for both Csm2 and Csm5 in the Csm crRNA -target RNA anti-tag are not well defined. Based on the existing EM density, we docked the respective atomic models into the map. All models were refined against summed maps using phenix.real_space_refine ( Adams et al., 2010 ) by applying geometric and secondary structure restraints. All figures were prepared by PyMol ( http://www.pymol.org ) or Chimera ( Pettersen et al., 2004 ). The statistics for data collection and model refinement are shown in Table 1 . Crystallization, Data Collection, and Structure Determination Crystals of Csm1-Csm4 subcomplex were grown at 20°C using hanging-drop vapor diffusion by mixing 1 μl protein solution with 1 μl reservoir solution containing 0.1 M phosphate-citrate pH 4.2, 5% PEG3000, 25% 1,2-propanediol, and 10% glycerol. Data were collected at 100 K at the Advanced Photo Source (APS) at the Argonne National Laboratory and processed by HKL suite ( Otwinowski and Minor, 1997 ). Structure was solved by the molecular replacement method using PHENIX. The refinement was performed with REFMAC5 ( Murshudov et al., 1997 ) and COOT ( Emsley et al., 2010 ). The statistics of the diffraction data are summarized in Table S1. All structure figures were prepared with PyMOL ( http://www.pymol.org/ ).

RNA Cleavage Assay

In vitro RNA cleavage assay was performed by incubating 50 nM Csm crRNA complex and 50 nM 5’-FAM labeled target RNA in 50 μL reaction buffer composed of 20 mM Tris–HCl, pH 8.0, 100 mM KCl, 100 mM NaCl, 5 mM MnCl 2 and 3% glycerol at 55°C for 15 min, unless stated otherwise. Reactions were stopped by adding 2 × formamide loading buffer (90% formamide, 0.025% bromophenol blue, 0.025% xylene cyanol, 0.025% SDS, 5 mM DTT) followed by heating at 95°C for 5 min. Samples were analyzed on 15% polyacrylamide denaturing 8 M Urea gel and visualized by Typhoon 9500 scanner. ΦX174 Virion ssDNA Cleavage Assay The reactions were performed by mixing 5 nM ΦX174 Virion ssDNA and 200 nM Csm crRNA complex with or without 5 nM of various target RNAs in reaction buffer (30 mM Tris–HCl, pH 8.0, 100 mM KCl, 100 mM NaCl and 3% glycerol) supplemented with 5 mM MnCl 2 at 55°C for 5 min, unless stated otherwise. Reaction were quenched by adding 2 × loading buffer (4 mM Tris-HCl (pH 7.6), 0.01% bromophenol blue, 0.01% xylene cyanol FF, 20% glycerol, 20 mM EDTA), followed by incubation for 5 min at 95°C. The samples were separated on a 1% agarose gel, stained with GelRed (Biotium) for visualization. 90-nt ssDNA Cleavage Assay The Csm crRNA complex (10 nM) was mixed with or without various RNA substrates (10 nM) in reaction buffer (30 mM Tris–HCl, pH 8.0, 100 mM KCl, 100 mM NaCl and 3% glycerol). Radiolabeled ssDNA substrate was then added to a final concentration of 100 nM. The reaction was initiated with 5 mM MnCl 2 and incubated at 55°C for 20 minutes. The reaction was then quenched with 2 × formamide buffer and heated to 95°C for 10 minutes. Samples were separated on a 12% Tris/Borate/EDTA gel with 7 M Urea and visualized by phosphorimaging.

Supplementary Material 1