Cell lines and culture

Escherichia coli (E. coli) BL21 (DE3) (Novagen) and DH10Bac (Thermo Fisher Scientific) were grown in LB. Two kinds of insect cell lines, Spodoptera frugiperda (Sf9, Expression systems) cells and Trichoplusia ni (High Five, Expression systems) cells, were used for virus preparation and recombinant protein expression, respectively. Sf9 and High Five cells were grown at 27 °C and 130 rpm in SIM SF expression medium and SIM HF expression medium (Sino Biological), respectively.

Protein construction, expression, and purification

The cDNAs of full-length METTL3, METTL14, HAKAI, and partial ZC3H13 (residues 1106–1668) were subcloned into modified or unmodified pFastBac-HTb vectors, resulting in the expression of N-GST-TEV-METTL3, N-6× His-TEV-METTL14, untagged HAKAI, and N-Twin-Strep-tag-ZC3H13 (1106–1668). The cDNAs of full-length WTAP and VIRMA were subcloned into the pFastBac-Dual vector, resulting in co-expression of N-6× His-WTAP and N-6× His-VIRMA. MAC (METTL3/METTL14) or four components of the MACOM complex, HAKAI/WTAP/VIRMA/ZC3H13 (HWVZ), were co-expressed in High Five insect cells, respectively. About 60 h after P2 virus infection at 27 °C, cells were harvested by centrifugation at 4000 rpm for 10 min and cell pellets were collected for protein purification.

Cells were resuspended in Buffer A (150 mM KCl, 20 mM Tris-HCl, pH 8.0, 10% glycerol, 25 mM imidazole) with 0.5 mM PMSF and protease inhibitors, lysed by adding 0.5% Triton X-100 and shaken gently for 20 min at 4 °C. After centrifugation at 18,000 rpm for 60 min, the supernatant was collected for further purification. The METTL3/METTL14 complex was purified by tandem affinity chromatography of Ni-NTA and GST. After removal of His- and GST-tags by TEV protease digestion at 4 °C overnight, the protein solution was further purified by a Heparin HP column. The concentrated solution was loaded onto a Superdex 200 Increase 10/300 GL column (GE Healthcare cytiva) equilibrated in SEC buffer (100 mM KCl, 20 mM HEPES, pH 7.5, 10% glycerol, 1 mM DTT). Fractions of the METTL3/METTL14 complex were collected and concentrated to 20 mg/mL. The purification procedures of METTL3 and METTL14 truncations were the same as those of the METTL3/METTL14 complex. The HWVZ complex was purified by tandem affinity chromatography of Ni-NTA and STREP beads. The concentrated solution was loaded onto a Superose 6 Increase 10/300 GL column equilibrated in SEC buffer. Fractions corresponding to the HWVZ complex were collected and concentrated to ~10 mg/mL. The purification procedures of HAKAI/WTAP/VIRMA (HWV), WTAP/VIRMA/ZC3H13 (WVZ), and WTAP/VIRMA (WV) complexes were the same as those of the HWVZ complex.

The cDNAs of truncated WTAP constructs were cloned into pET28a (with a 6× His-SUMO tag) for protein expression in E. coli BL21 (DE3). When E. coli were cultured at 37 °C to an OD600 of 0.6, isopropyl β-D-1-thiogalactopyranoside (IPTG) was added to a final concentration of 0.2 mM. Then E. coli were grown for ~16 h at 18 °C. The cells were collected by centrifugation, resuspended in buffer containing 20 mM Tris-HCl, pH 8.0, 100 mM KCl, 25 mM imidazole, and lysed with high-pressure homogenization. The truncated WTAP constructs were first purified by Ni-NTA affinity chromatography. Then protease ULP1 was added to remove the 6× His-SUMO-tag, and dialysis was applied to remove imidazole. Then the sample was subjected to the second Ni-NTA chromatography, and the flow-through was collected. The concentrated solution was loaded onto a Superdex 75 HiLoad 16/60 column (GE Healthcare) equilibrated in SEC buffer. Fractions corresponding to truncated WTAP constructs were collected and concentrated to ~20 mg/mL.

Sample preparation for cryo-EM

To prepare HWVZ or HWV complex samples for cryo-EM study, the HWVZ, HWV or HWVZ + M3/M14 complex was crosslinked by 0.5 mM bis(sulfosuccinimidyl)suberate (BS3, Thermo Fisher) for 2 h at 25 °C and quenched with 20 mM glycine. The complex was further treated by density gradient centrifugation. The gradient was generated from a light solution containing 10% (v/v) glycerol, 20 mM HEPES, pH 7.5, 100 mM KCl, 1 mM DTT, and a heavy solution containing 30% (v/v) glycerol, 20 mM HEPES, pH 7.5, 100 mM KCl, 1 mM DTT. Centrifugation was performed at 38,000 rpm in a Beckman SW41Ti swinging bucket rotor for 16 h at 4 °C. The peak fractions were collected and further dialyzed with buffer containing 100 mM KCl, 20 mM HEPES, pH 7.5, and 1 mM DTT at 4 °C overnight. The crosslinked HWVZ or HWV complex was concentrated to 1.0 mg/mL, and then applied to cryo-EM grids.

Preparation of RNA substrate

The RNA substrate ACTB-1 from the stop codon region of human ACTB mRNA containing two m6A sites and one RBM15 binding site9 (5′- GGCCCCUCCAUCGUCCACCGCAAAUGCUUCUAGGCGGACUAUGACUUAGUUGCGUUACACCCUUUCUUGACAAAACCUAACUUGCACAGAAAACA-3′) was obtained by in vitro transcription using T7 RNA polymerase. A linearized recombinant pUC19 plasmid containing the target sequence with 5′-hammerhead ribozyme sequence and free 3′-end digested by EcoRI was used as the DNA template for in vitro transcription of ACTB-1 RNA substrate. In vitro transcription was performed at 37 °C for 8 h in the reaction buffer containing 100 mM HEPES, pH 7.9, 10 mM MgCl2, 10 mM DTT, 6 mM NTP each, 2 mM Spermidine, 200 μg/mL linearized plasmid, and 100 μg/mL T7 RNA polymerase. The transcripts were purified by 8% denaturing urea PAGE, eluted from gel slices and precipitated with isopropanol. After centrifugation, the RNA precipitant was collected, washed twice with 70% ethanol and air-dried, and the RNA was dissolved in ultrapure water. FAM-labeled ACTB-1 RNA was produced by Silencer® siRNA Labeling kit-FAMTM according to the product instructions.

EMSA

An aliquot of 5 nM FAM-labeled ACTB-1 RNA was mixed with increasing concentrations of protein from 0.25 to 256 times of RNA concentration with a total of 11 gradients in 10 μL buffer containing 20 mM HEPES, pH 7.5, 100 mM KCl, 10% glycerol, 1 mM DTT and incubated at 25 °C for 20 min. Electrophoresis was performed with 8% native PAGE gels in running buffer containing 1× Tris-borate-EDTA (TBE) buffer at 4 °C for 2.5 h. Gels were scanned using a Typhoon FLA-9000 (GE Healthcare). As the complexes of RNA and MACOM cannot run into gels and were held as black deposits in the gel-loading wells due to their biochemical properties, we quantified the RNA-binding ratios by measuring the reduction of free RNA. Free RNA was quantified using ImageJ. Binding curves were fit individually using GraphPad Prism 9.0 software fitting with “One site–Specific binding with Hill slope” (GraphPad Software). Curves were normalized as the percentage of bound oligonucleotides and reported as the means ± SD of the interpolated Kd from three independent experiments.

BS3/EDC-mediated crosslinking mass spectrometry

The purified complexes were incubated with 0.5 mM BS3 (Thermo Fisher Scientific, 21580) in reaction buffer containing 50 mM HEPES, pH 7.5, 100 mM KCl, and 5% glycerol at 25 °C for 2 h or 5 mM EDC (Thermo Fisher Scientific, PG82073) in reaction buffer containing 50 mM HEPES, pH 7.2, 100 mM KCl, and 5% glycerol at 25 °C for 3 h. Crosslinked complexes were further purified to remove oligomers and glycerol by size exclusion chromatography. The proteins (10 μg) were precipitated and digested for 16 h at 37 °C by trypsin at an enzyme-to-substrate ratio of 1:50 (w/w). The trypsin-digested peptides were desalted and loaded on an in-house packed capillary reverse-phase C18 column (40-cm length, 100-µM ID × 360-µM OD, 1.9-µM particle size, 120-Å pore diameter) connected to an Easy LC 1200 system. The samples were analyzed with a 120-min HPLC gradient from 6% to 35% of buffer B (buffer A: 0.1% formic acid in water; buffer B: 0.1% formic acid in 80% acetonitrile) at 300 nL/min. The eluted peptides were ionized and directly introduced into a Q-Exactive mass spectrometer using a nano-spray source. Survey full-scan MS spectra (300–1800 m/z) were acquired in the Orbitrap analyzer with resolution r = 70,000 at 400 m/z. Crosslinked peptides were identified and evaluated using pLink2 software.34

The acquired MS/MS data were analyzed against a homemade database (including all the target proteins downloaded from UniProt) using pLink2. Cysteine alkylation by iodoacetamide was specified as a fixed modification with a mass shift of 57.02146 and methionine oxidation as a variable modification. Precursor mass tolerance and fragment mass tolerance were 20 p.p.m. BS3 (crosslinking sites K and protein N-terminus, xlink mass shift 138.0680796, mono-link mass shift 156.0786442) was defined as a crosslinker. Peptide length was set between 6 amino acids and 60 amino acids. Maximum missed cleavage sites were three. False discovery rate was set as 5%. We screened crosslinking sites that had at least 3 spectra with E-values ≤ 0.01 for structural analysis.

GST pull-down assay

GST pull-down assay was performed to detect protein–protein interactions using GST-tagged or untagged proteins purified from bacterial or insect cells. First, GST-tagged protein and untagged protein were mixed and incubated on ice for 30 min. Then the protein mixture was incubated with 15 μL GST beads in a total volume of 500 μL in binding buffer (100 mM KCl, 20 mM HEPES, pH 7.5, 10% glycerol, 1 mM DTT) at 4 °C for 2 h with gentle rotation. After centrifugation at 300× g for 3 min, the supernatant was removed, and the beads were washed three times using wash buffer (100 mM KCl, 20 mM HEPES, pH 7.5, 10% glycerol, 1 mM DTT, 0.1% NP-40). Another three washes were performed using binding buffer, followed by SDS-PAGE analysis.

In vitro m6A methyltransferase activity assay

In vitro methyltransferase activity assay was performed as previously published12 with several modifications. Briefly, the 50 μL reaction mixture contained 1 μM RNA substrate, 10 nM MAC, 15 nM MACOM, 30 μM d3-SAM, 100 mM KCl, 50 μM ZnCl2, 0.01% Triton X-100, 20 mM Tris, pH 7.5, 20 μg/μL BSA, 5 mM DTT, and 0.2 U/μL RRI (Recombinant RNase Inhibitor, Takara). The reaction was carried out by incubating at 37 °C for 1 h and quenched by inactivating the enzyme at 95 °C for 5 min. RNA was digested at 42 °C for 2 h with 20 mM CH3COONH4 and 1 U nuclease P1 (Wako). Afterwards, 1 μL BAP (Bacterial Alkaline Phosphatase, TOYOBO) and 6.4 μL 10× BAP Buffer were added, and the reaction was incubated at 37 °C for 2 h. The sample was diluted to 120 µL and then filtered through a 0.22-μm filter (Millipore). A 5-μL aliquot of the sample was injected into the LC/MS/MS system. The nucleosides were separated by reverse phase ultra-performance liquid chromatography on a C18 column with online MS detection using an Agilent 6460 QQQ triple-quadrupole LC mass spectrometer in positive electrospray ionization mode. The nucleosides were quantified by using nucleoside-to-base ion mass transitions of 282 to 150 (d3-m6A) and 268 to 136 (A). Quantification was performed in comparison with the standard curve obtained from pure nucleotide standards running on the same batch of samples. The ratios of d3-m6A to A and d3-m6A to probe were calculated on the basis of the calibrated concentrations.

UV crosslinking mass spectrometry of s4U-ACTB-1 and m6A writer complex

The purified m6A writer complex was incubated with s4U-ACTB-1 for 30 min on ice to form the stable complex. Then the sample was UV-crosslinked on ice with 3 J/cm2 at 365 nm wavelength with a UVP-crosslinker (analytik jena). Subsequently, the sample was treated with RNase I/RNase A/RNase T1 mixture at 1 U/μL concentration for 1 h at 37 °C. Proteins were precipitated with acetone. The protein pellet was dried by using a SpeedVac for 1−2 min. The pellet was subsequently dissolved in 8 M urea, 100 mM Tris-HCl, pH 8.5. TCEP (final concentration is 5 mM) (Thermo Scientific) and Iodoacetamide (final concentration is 10 mM) (Sigma) for reduction and alkylation were added to the solution and incubated at room temperature for 30 min, respectively. The protein mixture was diluted four times and digested overnight with trypsin at 1:50 (w/w) (Promega). The digested peptide solutions were desalted using a MonoSpinTM C18 column (GL Science, Tokyo, Japan) and dried with a SpeedVac. The peptide mixture was analyzed by a homemade 30 cm-long pulled-tip analytical column (75 μm ID packed with ReproSil-Pur C18-AQ 1.9 μm resin, Dr. Maisch GmbH, Germany), the column was then placed in-line with an Easy-nLC 1200 nano HPLC (Thermo Fisher Scientific) for mass spectrometry analysis. The analytical column temperature was set at 55 °C during the experiments. The mobile phase and elution gradient used for peptide separation were as follows: 0.1% formic acid in water as buffer A and 0.1% formic acid in 80% acetonitrile as buffer B, 0–1 min, 5%–8% B; 1–114 min, 8%–35% B; 114–115 min, 35%–50% B; 115–116 min, 50%–100% B; 116–120 min, 100% B. The flow rate was set as 300 nL/min.

Data-dependent tandem mass spectrometry (MS/MS) analysis was performed with an Orbitrap Eclipse mass spectrometer (Thermo Fisher Scientific). Peptides eluted from the LC column were directly electrosprayed into the mass spectrometer with the application of a distal 2-kV spray voltage. The cycle time was set to 3 s. Full scan resolution was set to 120,000 and MS/MS scan resolution was set to 30,000 with isolation window of 1.6 m/z. The dynamic exclusion settings used were as follows: charge exclusion, 1, and > 7; exclude isotopes, on; and exclusion duration, 30 s. MS scan functions and LC solvent gradients were controlled by the Xcalibur data system (Thermo Fisher Scientific).

The acquired MS/MS data were analyzed by RNPxl35 using 10 ppm for MS1. The s4U-crosslinking peptides (mass adduct is same as U-H2O). Raw data were searched against a FASTA database consisting of the protein sequences of the proteins in the complexes using RNPxl node combined in PD software with 0.01 FDR. The result is filtered by score (> 5) and the MS2 spectrum. MS2 spectra are annotated in TOPPView.36

Cryo-EM data acquisition

The samples of HWVZ, HWV, and HWVZ + M3/14 complexes were diluted at a final concentration of ~1 mg/mL. Three microliters of the samples were applied onto glow-discharged 200-mesh R2/1 Quantifoil copper grids. The grids were blotted for 4 s and rapidly cryocooled in liquid ethane using a Vitrobot Mark IV (Thermo Fisher Scientific) at 4 °C and 100% humidity. The samples were imaged in a Titan Krios cryo-electron microscope (Thermo Fisher Scientific) operated at 300 kV with a GIF energy filter (Gatan) at a magnification of 105,000× (corresponding to a calibrated sampling of 0.82 Å per pixel) for three samples. Micrographs were recorded by EPU software (Thermo Fisher Scientific) with a Gatan K3 Summit direct electron detector, where each image was composed of 30 individual frames with an exposure time of 2.5 s and an exposure rate of 22.2 electrons/s/Å2. A total of 12,745 movie stacks for HWVZ, 7642 movie stacks for HWV, and 14,168 movie stacks for HWVZ + M3/14 were collected.

Single-particle image processing and 3D reconstruction

All micrographs were first imported into Relion37 for image processing. The motion-correction was performed using MotionCor238 and the contrast transfer function (CTF) was determined using CTFFIND4.39 Then the micrographs with “rlnCtfMaxResolution < 5” were selected using the “subset selection” option in Relion. All particles were autopicked using the NeuralNet option in EMAN2.40 Then, particle coordinates were imported to Relion, where the poor 2D class averages were removed by several rounds of 2D classification. The initial maps were built and classified using the ab initio 3D reconstruction option in cryoSPARC41 without any symmetry applied. The 3D homogeneous refinements, local and global CTF refinements, and non-uniform refinements were performed using the selected 395,916 particles for the HWVZ complex, 282,821 particles for the HWV complex, and 199,741 particles for the HWVZ + M3/14 complex, resulting in ~3.0-Å, ~3.0-Å, and ~4.4-Å resolution maps, respectively. The resolution for the final maps was estimated with the 0.143 criterion of the Fourier shell correlation curve. Resolution map was calculated in cryoSPARC using the “Local Resolution Estimation” option. The figures were prepared using UCSF Chimera42 or UCSF ChimeraX43 (see more information in Supplementary information, Figs. S1–S3, S7 and Table S1).

Model building

Model building was first performed based on the 3.0-Å resolution cryo-EM map of the HWVZ complex. As none of these molecules has structural information, de novo model building was conducted. Phenix.map_to_model9 was first used to generate the initial model. Coot44 was then used to confirm the amino acid sequence registration of the initial model and assign amino acids to the cryo-EM density regions that were not resolved by phenix.map_to_model. Notably, the assignment of amino acid sequence by Coot was based on bulky residues (Trp, Lys, Arg, Phe, and Tyr). An atomic model composed of WTAP (residues 64–247), VIRMA (residues 334–1585), and ZC3H13 (residues 1492–1643) was obtained. The resulting model was refined using phenix.real_space_refine.45

To build the atomic model for the HWV complex, the final model of the HWVZ complex without the ZC3H13 subunit was fitted into the 3.0-Å resolution cryo-EM map of the HWV complex, followed by the optimization using phenix.real_space_refine and Coot. The model quality was evaluated by MolProbity25 and Q-scores.27 Statistics of the map reconstruction and model optimization are summarized in Supplementary information, Table S1. PDBsum structure bioinformatics28 was used to identify the key residues involved in interactions between subunits in our structures. All figures were made using UCSF Chimera or UCSF Chimera X.

Quantification and statistical analysis

For the quantification of the binding affinities of the HWVZ, WVZ, HWV, and WV complexes with FAM-labeled ACTB-1 RNA (Fig. 1d), measurements were carried out using ImageJ. GraphPad Prism was used to perform the statistical analysis of measuring results. Each data point represents the average of three independent experiments. Error bars represent SD. For the quantification of ACTB-1 RNA N6-adenosine methylation activity (Fig. 1g), GraphPad Prism was used to perform the statistical analysis of measuring results. Each data point represents the average of two independent experiments. Error bars represent SD.

Negative staining analysis

The purified crosslinked WVZ complex was applied to Superose 6 Increase 10/300 GL column (GE Healthcare) equilibrated with the gel filtration buffer. The complex was diluted to 35 ng/μL with SEC buffer. Five microliters of sample were applied to the glow-discharged 200 mesh carbon-coated copper grids. The samples were stained using 0.75% uranyl formate and air-dried. Data were collected on a Talos L120C transmission electron microscope equipped with a 4K × 4K CETA CCD camera (Thermo Fisher Scientific). Images were recorded at a nominal magnification of 92,000×, corresponding to a pixel size of 1.55 Å. CTF parameters of each micrograph were determined using CTFFIND4.39 Particles were picked and subjected to two rounds of 2D classification, followed by de novo 3D model generation, 3D classification, and auto-refinement in Relion.37