Quantitative analysis of m6A RNA modification by LC-MS

Summary N6-adenosine methylation (m6A) of messenger RNA (mRNA) plays key regulatory roles in gene expression. Accurate measurement of m6A levels is thus critical to understand its dynamic changes in various biological settings. Here, we provide a protocol to quantitate the levels of adenosine and m6A in cellular mRNAs. Using nuclease and phosphatase, we digest mRNA into nucleosides, which are subsequently quantified using liquid chromatography mass spectrometry. For complete details on the use and execution of this protocol, please refer to Cho et al. (2021).

SUMMARY N 6 -adenosine methylation (m 6 A) of messenger RNA (mRNA) plays key regulatory roles in gene expression. Accurate measurement of m 6 A levels is thus critical to understand its dynamic changes in various biological settings. Here, we provide a protocol to quantitate the levels of adenosine and m 6 A in cellular mRNAs. Using nuclease and phosphatase, we digest mRNA into nucleosides, which are subsequently quantified using liquid chromatography mass spectrometry. For complete details on the use and execution of this protocol, please refer to Cho et al. (2021).

BEFORE YOU BEGIN
This protocol requires preparation of several buffers and enzymes beforehand. To prevent degradation of RNA samples, it is necessary to follow general precautions for RNA experiments including preparation of RNase-free plastic wares and wiping working surfaces with RNase inactivating agents. To avoid contamination of buffers with RNase, we recommend that the users purchase RNase-free buffers (list provided in the key resources table). Using these raw materials, prepare working solutions and enzyme mixtures as described in the Materials and Equipment Section. The quality of mass spectrometry reagents (e.g., organic solvents, water) is also critical to reduce contamination of RNA and nucleotides from external sources.

HEK293E cell culture
Timing: 3 days The protocol was used to measure m 6 A levels in HEK293E cell line but can be adapted for any cells and tissue samples. Prepare enough number of cells and tissues to isolate >50 mg total RNA.

MATERIALS AND EQUIPMENT
Below is the list of reagents that need to be prepared before experiments. Use raw materials in the key resources table or other available reagents with similar grade (i.e., RNase-free materials). Individual procedures take 10-30 min. The buffers and reconstituted enzymes are good for use for 3-6 months.

PureLink RNA Mini Kit
The kit provides PureLink Lysis Buffer, Wash Buffer I, Wash Buffer II, and Spin cartridges for total RNA isolation. Reconstitute the Lysis Buffer with beta-mercaptoethanol and Wash Buffer II with ethanol according to the manufacturer's protocol.

Zymo RNA Clean & Concentrator Kit
The kit provides Zymo RNA Binding Buffer, RNA Prep Buffer, RNA Wash Buffer, and Spin cartridges for removal of salts from RNA samples. Reconstitute the RNA Wash Buffer with ethanol according to the manufacturer's protocol.

Oligo(dT) Binding Buffer
After mixing below components, store the buffer at 4 C.

Oligo(dT) wash buffer
After mixing below components, store the buffer at 4 C.

Reconstitution of nuclease P1
Prepare 2 unit/mL nuclease P1 stock by dissolving nuclease P1 powder in nuclease P1 reconstitution buffer. Aliquot into 10 mL and store at À20 C.

Nuclease P1 reconstitution buffer
After mixing below components, store the buffer at 4 C.

Reagent Stock concentration Final concentration Amount
Tris-HCl (pH7.5) 1 M 20 mM 1 mL Reconstitution of alkaline phosphatase Prepare 2 unit/mL alkaline phosphatase stock by dissolving alkaline phosphatase powder in alkaline phosphatase reconstitution buffer. Aliquot into 10 mL and store at À20 C.

Alkaline phosphatase reconstitution buffer
After mixing below components, store the buffer at 4 C.

M ammonium bicarbonate
Dissolve 158 mg ammonium bicarbonate in 1 mL PCR grade water. Filter through 0.2 mM PES filter using 1 mL syringe. Prepare fresh ammonium bicarbonate solution on the day of experiment.

M HCl
Dilute 100 mL 6 M HCl in 500 mL PCR grade water. Store at 4 C.

Preparation of m 6 A and adenosine standards
Dissolve 1 mg of m 6 A or adenosine powder in 1 mL of 75% acetonitrile (acetonitrile: water, 75:25, v/v). Dilute each standard as 1 mg/L. Mix m 6 A and adenosine standard solutions with the same volume (1:1) to make a standard solution containing 500 mg/L of both m 6 A and adenosine. Aliquot this standard solution mix into 100 mL and store at À80 C. To make standard calibration curves, make serial dilution of standards at 0.5, 1, 2, 5, 10, 20 and 50 mg/L. Then, obtain y = ax equation (a is constant) with ion counts (y) and standard concentrations (x) using linear regression.

LC-MS setting
Thermo Q Exactiveä Plus Hybrid Quadrupole-Orbitrapä Mass Spectrometer coupled with Vanquish UHPLC system was used. LC-MS system was controlled by Xcalibur software (Thermo). Metabolite separation was conducted by Xbridge BEH amide column (150 3 2.1 mm, 3 mm particle size). LC gradient was generated using LC solvents A and B (Table 1). Autosampler temperature was set at 4 C and the column temperature was set at 25 C. MS analysis was performed with a full-scan Nuclease-free water n/a n/a 7.35 mL Total n/a n/a 10 mL In this step, total RNA is isolated from the cells using PureLink RNA Mini Kit. Prepare the buffers in the kit according to the manufacturer's protocol before starting the experiment.
1. Sample harvest and homogenization a. Remove medium from cells and rinse with 13 PBS (e.g., 5 mL for 60 mm plates). b. Add 350 mL of PureLink Lysis buffer to the plate. Scrape the cell lysate thoroughly using a cell scraper. c. Transfer the viscous liquid into a new 1.5 mL tube.
Pause point: Samples can be frozen at À80 C. d. Homogenize the sample with a 23G syringe needle. Repeat the suction-release step 5-10 times.
Note: Try not to generate too many bubbles during homogenization (samples can overflow the tubes). a. Add 350 mL of 70% ethanol to sample (Sample: 70% ethanol = 1:1) and vortex. b. Transfer 700 mL of the sample into the PureLink Spin cartridge and centrifuge for 15 s at 12,000 3 g at 25 C. Discard the flow through. c. Add 700 mL of PureLink Wash Buffer I and centrifuge for 15 s at 12,000 3 g at 25 C. Discard the flow through. d. Add 500 mL of PureLink Wash Buffer II and centrifuge for 15 s at 12,000 3 g at 25 C. Discard the flow through. Repeat the step twice. e. Centrifuge the column for 2 min at 12,000 3 g to ensure complete removal of the wash buffer. f. Transfer the column to a new 1.5 mL tube.

Purification of total RNA
Note: Leave the column on the tube for 5 min to evaporate any residual ethanol from the wash buffer. g. Add 50 mL nuclease-free water directly to the column matrix and incubate for 5 min. h. Centrifuge for 2 min at 12,000 3 g at 25 C. The flow through contains total RNA.
Pause point: Samples can be frozen at À80 C.
3. Measure RNA concentration using Nanodrop with absorbance at 260 nm.

Purification of mRNA using Oligo(dT) beads
Timing: 3 h In this step, polyadenylated [poly(A)] mRNA is isolated from total RNA using oligo(dT) beads. Except heat block and ice incubation steps, all procedures are performed at 25 C. When not in the reaction (i.e., while preparing beads or kits), RNA samples should be kept on ice. Before starting the experiment, bring the oligo(dT) Binding and oligo (  Note: When handling multiple samples, stagger steps 6a-6d to decrease differences in bead incubation times among the samples. e. Incubate on the magnetic rack and remove the supernatant. f. Repeat steps 6d and 6e. g. To discard the wash buffer completely, centrifuge at 200 3 g for 10 s at 25 C.
Note: Do not centrifugate the beads at speeds higher than 200 3 g. Place the tube in a metal rack and remove the residual wash buffer.

Elution
a. Add 50 mL of Oligo(dT) Elution Buffer to the beads. Mix well by pipetting. b. To elute mRNA from the beads, heat the samples at 75 C for 2 min. c. Immediately place the tube on the magnetic rack and incubate until the solution is clear. d. Transfer the supernatant (i.e., eluted mRNAs) to a new 1.5 mL tube.
Pause point: Samples can be frozen at À80 C.
Pause point: Samples can be frozen at À80 C.
9. Conduct RNA clean-up using Zymo RNA Clean & Concentrator kit to remove residual salts from mRNA samples for the m 6 A processing step. a. Add 100 mL Zymo RNA Binding Buffer to 50 mL mRNA sample and mix (RNA Binding Buffer: mRNA sample = 2:1).
Note: To decrease variations in the isolated mRNA amount among the samples, use same amount of total RNA as a starting material (e.g., Adjust total RNA amount as 50 mg across all samples). b. Add 150 mL of 100% ethanol and mix (mRNA-RNA Binding Buffer: 100% ethanol = 1:1) c. Transfer the sample to the Zymo Spin cartridge. d. Centrifuge at 12,000 3 g for 30 s at 25 C. Discard the flow through. e. Add 400 mL Zymo RNA Prep Buffer to the column and centrifuge at 12,000 3 g for 30 s at 25 C.
Discard the flow through. f. Add 700 mL Zymo RNA Wash Buffer to the column and centrifuge at 12,000 3 g for 30 s at 25 C. Discard the flow through. g. Add 400 mL Zymo RNA Wash Buffer to the column and centrifuge for 2 min to completely remove the wash buffer. h. Transfer the column carefully into a new 1.5 mL tube.
Note: Leave the column on the tube for 5 min to evaporate any residual ethanol from the wash buffer. i. Add 15 mL nuclease-free water directly to the column matrix and incubate for 5 min. j. Centrifuge at 16,000 3 g for 30 s at 25 C. The flow through contains purified mRNA.

OPEN ACCESS
Note: This is to have a similar proportion of organic solvent in the LC-MS sample with the starting LC mobile phase (75% acetonitrile). b. Centrifuge the samples at 16,000 3 g for 10 min at 4 C to precipitate any insoluble parts. c. Carefully transfer 40 mL of supernatant to a new LC-MS vial.
Note: Do not touch the pellet.
15. Inject 3 mL of samples to the LC-MS system with the setting parameters described above.
Note: The 3 mL sample now contains $4.5 ng mRNA if the m 6 A processing was performed with 200 ng mRNA as a starting material.
16. Run m 6 A and adenosine standards in the same LC-MS setting with the samples.

EXPECTED OUTCOMES
Under suggested conditions, m 6 A and adenosine are eluted at 1.65 min and 1.86 min, respectively ( Figure 1 and Table 3).

QUANTIFICATION AND STATISTICAL ANALYSIS
1. Convert LC-MS raw data files to mzXML using Proteowizard software.  Note: MAVEN software (https://resources.elucidata.io/elmaven) or other software can be used to use mzXML file for peak visualization and quantitation.
2. Export ion counts of m 6 A and adenosine for each sample.
3. Calculate the concentration of m 6 A and adenosine using standard calibration curves (Figure 2).

LIMITATIONS
This protocol details quantitative measurement of m 6 A modification in mRNAs using LC-MS. While this protocol is straightforward and easy to follow, it has some limitations.
First, to measure m 6 A modification of mRNAs, we purified mRNA from total RNA. However, contamination of abundant RNA species such as ribosomal RNA (rRNA) can occur. To measure m 6 A levels specifically from the mRNA m 6 A modification sequence (GA*C; A* is methylated adenosine), the users can adopt RNase T1-based assays such as 2D thin-layer chromatography (TLC) (Bodi and Fray, 2017). In the TLC assay, mRNAs are processed with RNase T1 (specifically cleaves after G) followed by 32 P labeling of nucleotides, which enables specific labeling of m 6 A from mRNAs.
Second, this protocol quantitates m 6 A levels from a total pool of mRNAs and cannot distinguish differential m 6 A modification levels in individual genes. This requires site-specific m 6 A detection using qPCR or TLC. Transcriptome-wide m 6 A sequencing methods have also been developed by several groups (reviewed in Zaccara et al., 2019).
Finally, while this protocol provides an optimized LC-MS condition for efficient measurement of m 6 A and adenosine in mRNA, users can adjust RNA purification and mass spectrometry methods to quantitate other modifications in various RNA species and DNA (Su et al., 2014;Thü ring et al., 2017;Wei et al., 2018;Wein et al., 2020). Comprehensive analysis of nucleotide chemical modifications using LC-MS technology will provide valuable tools and resources in the field of transcriptomics, genomics, and metabolomics.

TROUBLESHOOTING
Problem 1 Low yield of mRNA (Related step: purification of mRNA using Oligo(dT) beads).

Potential solution
Since mRNA is only 1-5% of total RNA, preparation of enough amount of total RNA is key to get enough amount of mRNA (e.g., 30-100 mg total RNA as a starting material). Also, use nucleic acid low-bind tubes and low retention pipette tips to minimize loss of mRNAs during purification. We recommend calculating mRNA purification yield using the amount of starting material (total RNA, step 3) and final mRNA product (step 10; consider that the yield of Zymo RNA Clean & Concentrator Kit is 70-80%). To decrease differences in mRNA yield among the samples, stagger the 5 min Oligo(dT) bead-RNA incubation step when handling several samples. Randomization of the sample order during the reaction also helps to decrease the incubation time differences caused by the sample order (i.e., randomization of sample order prevents Sample #1, 2, 3 being incubated longer with the beads than Sample #22, 23, 24).

Problem 2
Contamination of other RNA species (Related step: purification of mRNA using Oligo(dT) beads).

Problem 3
Background nucleoside signal (Related step: processing of mRNA samples for m 6 A analysis).

Potential solution
Due to the nucleic acids contaminated from the environment and reagents, background m 6 A and adenosine signals can be detected. To prevent this, we recommend using PCR grade (i.e., nucleic acid-free) water during the m 6 A processing step and subtracting the background signals detected in the water-only sample. Also, conduct m 6 A processing step in a clean chemical fume hood.

Problem 4
Confirmation of m 6 A peak (Related steps: LC-MS analysis of m 6 A and expected outcomes).

Potential solution
Some samples may show m 6 A isomers as shown in the Figure 1A (left panel). To avoid mis-annotation of peaks, we recommend running m 6 A standards in parallel with the samples to obtain accurate retention time. MS/MS profile can be used to confirm correct m 6 A peak ( Figure 1A, right panel).

Problem 5
Alternative reagents and equipment (Related step: key resources table).

Potential solution
In the key resources table, we provided catalog numbers of the reagents and equipment that this protocol used, which may not be available in other circumstances. Users can use any reagents with equivalent grade (e.g., RNase-free reagents for RNA isolation and processing steps; HPLCgrade reagents for LC-MS). Regarding the equipment, (1) Oligo(dT)-RNA sample can be mixed using orbital shakers or rotators instead of Thermomixer; (2) Shaking nuclease P1 and alkaline phosphatase reactions in Thermomixer is optional (i.e., reaction can be performed in a regular heat block without agitation); (3) for mass spectrometry analysis of nucleoside samples, any type of high-sensitivity tandem mass spectrometers such as triple quadrupole, quadrupole-time of flight, and quadrupole-orbitrap can be used.

RESOURCE AVAILABILITY
Lead contact Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Gina Lee (ginalee@uci.edu).

Materials availability
This study did not generate new unique materials, reagents, or cell lines.

Data and code availability
The published article includes all datasets generated and analyzed during this study.