Single-cell m⁶A mapping in vivo using picoMeRIP–seq

Ethics statement

Zebrafish and mouse experiments were approved by the Animal Research Committee of the Norwegian Food Safety Authority (Forsøksdyrforvaltningens tilsyns- og søknadssystem (FOTS) IDs: 10898 and 24911). Experimental procedures conformed to the ARRIVE guidelines and were conducted in accordance with the ethical guidelines in Directive 2010/63/EU of the European Parliament on the protection of animals used for scientific purposes and Norwegian legislations.

Antibodies and tubes

The following antibodies to m6A were used in the experiments: Millipore, ABE572; New England Biolabs (NEB), E1610S; Diagenode, C15200082-50; Synaptic Systems, 202003.

The following low-binding tubes were used in the experiments: Axygen Maxymum Recovery 1.5-ml low-bind tubes (VWR.no, 525-0230); Axygen Maxymum Recovery 0.6-ml low-bind tubes (VWR.no, 525-0229); Axygen Maxymum Recovery 0.2-ml low-bind tubes (VWR.no, 732-0679).

Zebrafish zygotes, mouse liver, mES cells, mouse oocytes and embryo collection

Total RNA was extracted from 100 zebrafish zygotes using TRIzol reagent (Thermo Fisher Scientific) and eluted in 100 μl of RNase-free water. Then, 10-μl and 5-μl samples were taken for ten and five zygotes, respectively, and volumes were adjusted to 12 μl with nuclease-free water. Single zebrafish zygotes were manually picked and distributed into 12 μl of 1× lysis buffer (Takara). Finally, the samples were snap-frozen in liquid nitrogen and stored at −80 °C until further processing.

For Extended Data Fig. 7, mES cells (R1) were purchased from ATCC (SCRC-1011). Twelve microliters of 1× lysis buffer (Takara) was dispensed into each well of a 96-well PCR plate, and cells were sorted into each well according to the manufacturer’s instructions using a BD FACSMelody cell sorter (BD Biosciences). Plates were sealed with sealing films and immediately stored at −80 °C until further processing. Mettl3–/– and WT control mES cell lines were gifted from S. Geula et al., Jacob H. Hanna laboratory, Weizmann Institute of Science8. Mycoplasma testing was performed on a regular basis, and all cell lines were free of Mycoplasma.

Mice were housed in individually ventilated cages (IVC, Scanbur) with a stable light/dark cycle (7:00 to 19:00), with 55 ± 5% relative humidity at 22 ± 2 °C with free access to water and standard rodent chow diet (2018S; 58 E% carbohydrate, 18 E% fat, 24 E%; Teklad Global 18% Protein Rodent Diet, Envigo). The presence of pathogens was monitored quarterly in accordance with the Federation of European Laboratory Animal Science Association guidelines. Animals were specific pathogen free according to the Federation of European Laboratory Animal Science Association recommendations (specific pathogen-free status).

To collect GV oocytes, 8-week-old C57BL6/N females were injected with 5 U of pregnant mare serum gonadotropin (PMSG), and 48 h after PMSG injection, ovaries were dissected, and oocytes were isolated by puncturing the follicles. The procedure was performed in M2 medium supplemented with 0.2 mM 3-isobutyl-1-methylxanthine (a cyclic nucleotide phosphodiesterase inhibitor; Sigma) to prevent the oocytes from further progress to GV breakdown. The cumulus cells were gently removed by pipetting, and the oocytes were briefly exposed to acidic Tyrode’s solution (Sigma) to remove the zona pellucida, followed by three washes in M2 medium.

To collect MII oocytes, 4- to 5-week-old C57BL6/N females were injected with 5 U of PMSG followed by 5 U of human chorionic gonadotropin (hCG) 45 h after PMSG injection. The oviducts were dissected 20–22 h later and transferred to a clean dish containing M2 medium (Sigma). The oviduct ampulla was identified under a stereomicroscope to isolate MII oocytes containing the cumulus mass. Oocytes were treated with 0.3 mg ml–1 hyaluronidase dissolved in M2 medium to remove the cumulus cells and were exposed to acidic Tyrode’s solution (Sigma) for a few seconds to remove the zona pellucida. Finally, MII oocytes were washed in M2 medium.

To collect early embryos, female mice were superovulated by hormone injection (5 U of PMSG followed by 5 U of hCG 45 h later) and transferred to cages with C57BL/6N males (8 weeks old) for mating. At 27–28 h (zygote), 39–43 h (two cell), 68–70 h (eight cell) and 92–94 h (blastocyst) after hCG administration, female mice were killed by cervical dislocation. Embryos were flushed from the reproductive tract into HEPES-buffered CZB medium and transferred to acidic Tyrode’s solution (Sigma) for a few seconds to remove the zona pellucida, followed by three washes in M2 medium.

The mouse oocytes and embryos were manually picked and sorted into 12 μl of 1× lysis buffer (Takara). The samples were snap-frozen in liquid nitrogen and stored at −80 °C until further use.

Total RNA of C57BL/6N mouse livers was extracted using TRIzol reagent. Poly(A)+ RNA was selected twice with a Dynabeads mRNA purification kit (Thermo Fisher Scientific). Identification of ribosomal RNA contamination was conducted using an Agilent 2100 Bioanalyzer according to the user manual.

Real-time qPCR

cDNA was synthesized from m6A-immunoprecipitated RNA using SuperScript VILO master mix (Thermo Fisher Scientific), and real-time qPCR was conducted using Fast SYBR Green master mix (Thermo Fisher Scientific) following the manufacturer’s protocol.

The following Pdzd8 and Rdh10 primer sequences were used for real-time qPCR.

Positive m6A control Pdzd8:

Forward primer, 5´-GTGGTTCTCTCATAGGACATAAAG-3´

Reverse primer, 5´-CAAAGCCAGTTATCAATACAGTCA-3´

Negative m6A control Rdh10:

Forward primer, 5´-AGTGTAGTGCTCTGTTGTGT-3´

Reverse primer, 5´-CGCTGATCTCAAACTGACATC-3´

To calculate the S/N ratio, the following formula was used:

$$begin{array}{l}{mathrm{S}}/{mathrm{N}}, {mathrm{ratio}}=\ left[right.2^{(C_t, {mathrm{input}},(Pdzd8, ({mathrm{corrected}}))-C_t, {mathrm{IP}}, (Pdzd8))}left.right]/left[right.2^{(C_t, {mathrm{input}},(Rdh10, ({mathrm{corrected}}))-C_t, {mathrm{IP}}, (Rdh10))}left.right].end{array}$$

Ct (cycle threshold) input (corrected) = (Ct input – log2 (10)). We subtract log2 (10) when the input represents 1/10th of the amount used for RNA immunoprecipitation. This is in order to correct for the difference in starting amount used for the input, and is only applied if using a different amount of starting material for the input as compared to the RNA IP. When the same amount of starting material is used for both the input and the RNA IP, there will be no correction for the input amount.

Single-cell picoMeRIP–seq

The following procedures were performed in a UV decontaminated LAF bench.

rRNA and DNA depletion

For single-tube rRNA and DNA depletion, we performed rRNA depletion using an NEBNext rRNA depletion kit (NEB) with some modifications to the user manual. Specifically, 3 µl of the RNA/probe master mix was added to a 12-µl sample, which was then subjected to a temperature ramp from 95 °C for 2 min to 22 °C at a rate of −0.1 °C s–1, followed by a 5-min hold at 22 °C. Next, 5 µl of the RNase H reaction mix was added to the samples and incubated at 37 °C for 30 min, after which 30 µl of DNase I digestion mix was added and incubated at 37 °C for an additional 30 min. The resulting samples were purified using 2.2× volume of RNAClean XP beads, washed twice with 80% freshly prepared ethanol and eluted with 78 μl of nuclease-free water. Finally, 2 μl of RiboLock RNase inhibitor (40 U µl–1) was added to the sample to prevent RNA degradation, resulting in a sample volume of 80 μl.

RNA fragmentation by sonication

A UP100H Ultrasonic Processor (Hielscher) with a 2-mm probe was used to sonicate the samples, using pulse settings with 0.5-s cycles and 27% power. The samples underwent n × 30 s sonication cycles, with 30 s of sonication followed by 30 s on ice for each cycle. The numbers (n) of sonication cycles used in this study for different amounts of input were optimized and can be found in Supplementary Table 1. For mouse liver samples, 10 ng and 100 pg were used to construct input libraries after sonication. RNA from pools of zygotes was used for input controls for single zebrafish zygotes (Supplementary Table 1). In the case of single mouse oocytes and early embryos, 10% of multiple oocyte/embryo RNA was removed and served as input control (Supplementary Table 1). To the samples consisting of 80 μl, 20 µl of 5× IP buffer (50 mM Tris-HCl (pH 7.5), 750 mM NaCl, 0.5% (vol/vol) NP-40 and 5 U µl–1 RiboLock RNase inhibitor) was added to make a final volume of 100 µl for sonication.

Antibody–bead incubation

Before use, Dynabeads (Invitrogen) were washed by taking 20 μl of beads and washing them twice with 1× IP buffer (200 µl of 5× IP buffer supplemented with 800 µl of nuclease-free water) by vortexing, quickly centrifuging on a MiniGalaxy and placing on a magnetic rack before discarding the supernatant. In a separate tube, the antibody was diluted by taking 4 µl of anti-m6A, 16 µl of 5× IP buffer and 60 µl of nuclease-free water and mixing by gentle vortexing. The antibody-containing solution was added to the washed beads, and the antibody–beads were incubated overnight with head-over-tail rotation on a HulaMixer at 4 °C (40 r.p.m.). After conducting antibody testing and comparison experiments, anti-m6A from Millipore (ABE572) was selected for use in all subsequent experiments.

IP and washes

Antibody-coated beads were captured on the tube wall in a magnetic rack. The supernatant from the antibody–bead incubation was discarded. Antibody-coated beads were washed twice with 200 μl of 1× IP buffer by vortexing (four times for 5 s each) to remove unbound antibodies that would otherwise compete for binding to the epitope. At the end of the second wash, the antibody-coated beads were transferred to 0.2-ml PCR tubes. From 200 μl, a volume of 10 μl of homogenous antibody-coated bead solution was transferred to each PCR tube. The tubes were quickly centrifuged on a MiniGalaxy and placed in a magnetic rack for at least 2 min or until the solution became clear. After removing the supernatant, 100 μl of sonicated sample RNA solution was added to each antibody–bead-containing tube, and the samples were incubated with head-over-tail rotation on a HulaMixer at 4 °C for 2 h (40 r.p.m.). Tubes were quickly centrifuged on a MiniGalaxy and placed in a magnetic rack. The supernatant was removed, and the RNA–antibody–bead complexes were washed four times in the following solutions, quickly spun and placed in a magnetic rack in between washes: washed once with ice-cold medium-stringency RIPA buffer (10 mM Tris-HCl (pH 8.0), 300 mM NaCl, 1 mM EDTA, 0.5 mM EGTA, 1% (vol/vol) Triton X-100, 0.2% (vol/vol) SDS and 0.1% (vol/vol) sodium deoxycholate), washed twice with ice-cold high-stringency RIPA buffer (10 mM Tris-HCl (pH 8.0), 350 mM NaCl, 1 mM EDTA, 0.5 mM EGTA, 1% (vol/vol) Triton X-100, 0.23% (vol/vol) SDS and 0.1% (vol/vol) sodium deoxycholate) and washed once with ice-cold medium-stringency RIPA buffer. After the four washes, tubes were quickly spun and placed in a magnetic rack, and the supernatant was discarded. The RNA–antibody–bead complexes were then resuspended in 100 μl of 1× IP buffer and incubated for 5 min. The samples were then quickly spun and placed in a magnetic rack, and the supernatant was removed. The RNA–antibody–bead complexes were resuspended in 147.9 μl of elution buffer (5 mM Tris-HCl (pH 7.5), 1 mM EDTA, 0.05 % (vol/vol) SDS and 1 U µl–1 RiboLock RNase inhibitor). Proteinase K (2.1 μl; NEB) was added to each tube, and tubes were then incubated on a Thermomixer at 1,200 r.p.m. and 55 °C for 1.5 h. After incubation, the tubes were briefly centrifuged and incubated further on a Thermomixer at 80 °C for 20 min to inactivate the Proteinase K. The samples were then placed in a magnetic rack for 2–3 min, and the supernatant containing the m6A-immunoprecipitated RNA was transferred to a new 1.5-ml low-binding tube. The remaining beads were resuspended again in 147.9 μl of elution buffer, and 2.1 μl of Proteinase K was added. The samples were placed immediately in a Thermomixer at 1,200 r.p.m. and 55 °C for 5 min, followed by inactivation of Proteinase K on a Thermomixer at 80 °C for 20 min. The tubes were then placed back in a magnetic rack for 2–3 min, and the supernatant was collected and pooled with the first supernatant in the same 1.5-ml low-binding tube to recover as much of the m6A-immunoprecipitated RNA as possible, resulting in a total volume of about 300 μl.

Ethanol precipitation

For both input and immunoprecipitated RNA samples, nuclease-free water was added to each tube to result in a final volume of 400 μl. Next, 40 μl of 3 M sodium acetate (pH 5.2; Thermo Fisher Scientific) and 10 μl of linear acrylamide 5 mg μl–1 (Thermo Fisher Scientific) were added, followed by 1,000 μl of ice-cold 100% ethanol. The samples were vigorously vortexed without centrifugation or spinning and immediately placed at –80 °C for at least 2 h or overnight until completely frozen. Samples were recovered from –80 °C and allowed to briefly thaw on ice, and it was visually confirmed that all samples had thawed before starting centrifugation. The samples were centrifuged at 20,000g at 4 °C for 15 min, and the supernatant was carefully removed without disturbing the visible pellet. The pellet was then washed twice with 1 ml of ice-cold 75% ethanol. For washes, 75% ethanol was added, and the tube was gently vortexed until the pellet detached from the bottom; centrifugation was repeated as described above. After the last wash, as much as possible of the supernatant was removed, the tube lid was left open until all ethanol had evaporated, and the dried pellet was resuspended in 7 μl of nuclease-free water.

Library preparation and sequencing

With modifications to the manufacturer’s protocol, as described earlier, the SMART-Seq stranded kit (Takara, 634442) was used to construct sequencing libraries. For the fragmented input or immunoprecipitated RNA, we performed the protocol without the fragmentation step. After the first PCR amplification and following AMPure bead purification, we resuspended the beads by adding 46.5 µl of nuclease-free water and skipped the ribosomal cDNA depletion protocol in Section D. We then incubated the samples at room temperature for 5 min to allow time for rehydration and recovered 46 µl of supernatant from each sample. We continued following the protocol until completion. The libraries were assessed for quantity using KAPA library quantification kits (Roche), and size distribution was assessed using TapeStation D1000 ScreenTape (Agilent Technologies). In combination, this information provided for good estimation of pooling at equimolar ratios. The pooled libraries were sequenced on a NovaSeq system (Illumina) with 50-base pair (bp) paired-end mode.

Spike-in controls

Spike-in control RNAs and qPCR primers are from the EpiMark N6-methyladenosine enrichment kit. Before adding the spike-in control RNAs to an RNA sample for picoMeRIP, each control RNA was diluted to 0.001 fmol μl–1, and 1 μl of the diluted control RNA was added. For the picoMeRIP–qPCR titration experiment, each control RNA was diluted to 1 fmol in 100 μl. The two control RNAs were then mixed together at the indicated ratio used for picoMeRIP–qPCR (Extended Data Fig. 4a).

Western blotting

Western blotting was performed as previously described32. Total proteins were extracted using RIPA lysis buffer (Thermo Scientific, 89900) containing protease inhibitor cocktail (Sigma-Aldrich, P8340) and phenylmethylsulfonyl fluoride (Sigma-Aldrich, P7626). Protein samples were denatured and resolved by Bolt Bis-Tris Plus gels (Invitrogen). Separated proteins were transferred onto nitrocellulose membranes and detected with primary antibodies to METTL3 (Abcam, ab195352) and GAPDH (Abcam, ab125247). The following secondary antibodies were used: donkey anti-mouse horseradish peroxidase (HRP; Abcam, ab6820) and donkey anti-rabbit (HRP; Abcam, ab6802). Blots were developed by enhanced chemiluminescence (Thermo Fisher Scientific, 32209 and 34095) and scanned with a Bio-Rad ChemiDoc XRS+ system.

Sequencing data processing, m6A peak identification and motif analysis

The code used for quality check, alignment and filtering of sequencing reads, identification of m6A peaks and m6A consensus motifs and abundance estimation of gene transcripts is available at GitHub (https://github.com/Augroup/MeRipBox).

Quality of raw sequencing reads was assessed using FastQC (v0.11.8; https://www.bioinformatics.babraham.ac.uk/projects/fastqc/) with the default parameters. After trimming sequencing adapters and low-quality bases with Cutadapt (v1.8.1)33 with the parameter ‘-q 20,20 -m 20 –max-n 0.01 –trim-n’, the clean read pairs were mapped to the reference genomes (mm10 for mouse and danRer11 for zebrafish) and reference sequences (for the two spike-in RNA controls, obtained from the manual for the EpiMark N6-methyladenosine enrichment kit) using HISAT2 (v2.1.0)34 with the parameter ‘-5 8 –no-mixed –no-discordant’. Multiply mapped read pairs (that is, more than one genomic locus per read pair as reported by HISAT2) were discarded. We further filtered out PCR duplicates by using SAMtools (v1.9)35 and read mates that overlapped with the genomic coordinates of ribosomal RNAs by using BEDTools (v2.28.0)36. These uniquely aligned, deduplicated and non-rRNA reads were used for m6A peak calling.

We identified m6A peaks using a model-based method called MACS (v2.1.2)26 with the mode ‘callpeak’, the parameter ‘–keep-dup all -B –nomodel –call-summits’ and estimated transcriptome sizes of ‘-gsize 242010196’ for mouse and ‘-gsize 117608789’ for zebrafish. The statistical significance cutoff for the identified m6A peaks was a q value of
>>> Read full article>>>
Copyright for syndicated content belongs to the linked Source : Nature.com – https://www.nature.com/articles/s41587-023-01831-7