Data from: Cancer mutations rewire the RNA methylation specificity of METTL3-METTL14
Data files
Dec 23, 2024 version files 639.88 MB
-
ivm_seq_abcam_mock.zip
75.87 MB
-
ivm_seq_abcam_r298p.zip
118.11 MB
-
ivm_seq_abcam_wt.zip
88.60 MB
-
ivm_seq_sysy_d312a.zip
72.92 MB
-
ivm_seq_sysy_mock.zip
14.51 MB
-
ivm_seq_sysy_r298c.zip
15.64 MB
-
ivm_seq_sysy_r298h.zip
44.42 MB
-
ivm_seq_sysy_r298p.zip
104.09 MB
-
ivm_seq_sysy_wt.zip
66.06 MB
-
README.md
880 B
-
selex_mock.zip
13.10 MB
-
selex_r298p.zip
11.08 MB
-
selex_wt.zip
15.47 MB
Abstract
Chemical modification of RNAs is important for post-transcriptional gene regulation. The METTL3-METTL14 complex generates most N6-methyladenosine (m6A) modifications in mRNAs, and dysregulated methyltransferase expression has been linked to cancers. Here we show that a changed sequence context for m6A can promote oncogenesis. A gain-of-function missense mutation from cancer patients, METTL14R298P, increases malignant cell growth in culture and transgenic mice, without increasing global m6A levels in mRNAs. The mutant methyltransferase preferentially modifies noncanonical sites containing a GGAU motif, in vitro and in vivo. The m6A in GGAU context is detected by the YTH family of readers similarly to the canonical sites but is demethylated less efficiently by an eraser, ALKBH5. Combining the biochemical and structural data we provide a model for how the cognate RNA sequences are selected for methylation by METTL3-METTL14. Our work highlights that sequence-specific m6A deposition is important and that increased GGAU methylation can promote oncogenesis.
README: Cancer mutations rewire the RNA methylation specificity of METTL3-METTL14
https://doi.org/10.5061/dryad.37pvmcvvb
Description of the data and file structure
Raw sequencing data (.fastq.gz and .md5 files are compressed in each .zip file) include IVM-seq (abcam or sysy antibody) and SELEX experiments. The files with name containing "ivm-seq" or "selex" were derived from the corresponding experiment respectively. "ivm_seq_sysy" or "ivm_seq_abcam" indicates the parallel experiments performed by using sysy or abcam antibodies respectively. "mock, wt, r298p, r298c, r298h, d312a" indicates the experiment were performed by using no protein (mock pull-down control), wild-type METTL14 recombinant protein or corresponding METTL14 mutant protein. The details of raw sequencing data processing is included in the method section.
Methods
In vitro methylation sequencing (IVM-seq)
A degenerate DNA oligonucleotide mixture (5'-CTCCTTCTGGCATAAGAAGTNNNNNNNNNNNNNNNNNNNNCAAGCCAAGCAAGTATATAGG-3') consisting of 20-random-nucleotide (N20) and flanking constant regions was synthesized by MilliporeSigma with manual adjustment to achieve the overall equal ratio of nucleotides. An initial double-stranded (ds)DNA library was produced by a 7-cycle PCR amplification (F-primer: 5'-ATCTCGATCCCGCGAAATTAATACGACTCACTATAGGGCTCCTTCTGGCATAAGAAGT-3'; R-primer: 5'-CCTATATACTTGCTTGGCTTG-3') in a 1 mL reaction containing 20 pmol of N20 oligonucleotides as a template to preserve the designed library complexity. The dsDNA was then gel-eluted and transcribed into a randomized RNA library. The RNA library (6.25 µM) was methylated by the target recombinant methyltransferase heterodimer (0.25 µM) in a 20 µL reaction (50 mM Tris pH 7.5, 0.01% Triton X-100, 15 mM NaCl, 1 mM DTT, 1% glycerol, 5 µM S-(5′-adenosyl)-L-methionine (SAM) (Sigma A7007), 20 U SUPERase·In (Invitrogen AM2694)) at room temperature for 2 hours. The methylated RNA library was extracted by acid phenol-chloroform and precipitated with isopropyl alcohol before being reconstituted in nuclease-free water. The purified RNA library was incubated with 7 µg of antibody against m6A (Abcam ab151230 or SySy 202003) in 100 µL of low-salt binding buffer containing 50 mM Tris pH 7.5, 150 mM NaCl, 0.1% NP-40, 50 U SUPERase·In, and bound to protein A/G beads (Thermo Scientific 88802). The mock control sample was prepared by the same procedures but incubated with beads in the absence of antibody. The protein A/G beads were washed with 0.9 mL of low-salt binding buffer once and high-salt buffer (50 mM Tris pH 7.5, 500 mM NaCl, 0.1% NP-40) twice, then one more time by the low-salt binding buffer before the elution by 0.1 M glycine pH 2.5. Methylated RNA species were recovered from the elution and reverse transcribed (RT) using the RT primer (same as R-primer above) and SuperScript III (Invitrogen 18080044). Excessive RT primer was digested with ExoSAP-IT (Applied Biosystems 78250). The resulting cDNA was adapted for barcoding (adF-primer: 5'-GTCGGCAGCGTCAGATGTGTATAAGAGACAGCTCCTTCTGGCATAAGAAGT-3'; adR-primer: 5'-CGTGGGCTCGGAGATGTGTATAAGAGACAGCCTATATACTTGCTTGGCTTG-3') by 10-cycle of PCR amplification and the dsDNA was purified by PureLink PCR Micro kit (Invitrogen K310010). The standard Nextera i5 and i7 barcodes were added to the dsDNA by another 5-cycle of PCR amplification before pooling for Illumina NextSeq500 single-end 75-bp sequencing. The sequencing was performed at Next Generation Sequencing Core, Eugene McDermott Center, University of Texas Southwestern Medical Center. The 20-nt and 21-nt constant flanking sequences from the N20 oligonucleotides (5'-CTCCTTCTGGCATAAGAAGT-3'; 5'-CCTATATACTTGCTTGGCTTG-3') were used to trim the reads by using the Trim Ends module in Geneious (version 2021.2.2). The reads with the length of 20-nt after trimming were selected for motif analysis by MEME using the Differential Enrichment mode with the mock control set as background. The output position-specific probability matrices were reconstituted to logograph by the Seq2Logo (version 2.0).
SELEX
An aliquot (0.2 nmol) of randomized RNA library produced for IVM-seq experiment was incubated with equal molarity of 6x His-tag fused target protein complex immobilized on 7 µL Ni-NTA magnetic agarose beads (Thermo Scientific 78605) in the SELEX binding buffer (10 mM Tris pH 8.0, 150 mM NaCl, 10 mM MgCl2, 0.01% NP-40, 1% glycerol, 1 mM ꞵ-mercaptoethanol, 10 U SUPERase·In) at 22 °C for 30 min on a thermomixer with 15 s pulse-shaking at 1200 r.p.m. every 5 min. After 3x wash by 180 µL SELEX binding buffer, the RNA species with affinity was recovered from beads by heating at 65 °C for 5 min in 1 mM Tris pH 7.5 buffer containing 20 pmol of RT primer (same as used in IVM-seq). The elution of RNA was assembled in an RT reaction similarly as in IVM-seq. The subsequent cDNA was amplified by dsDNA amplification primers 100 µL PCR reaction to generate the dsDNA template for the next cycle of SELEX. Five SELEX cycles were performed before generating the library for high-throughput sequencing. The SELEX samples were adapted and sequenced together with IVM-seq samples. The sequence data were processed, and motifs analysis was performed identically as for IVM-seq.