Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Nov 22.
Published in final edited form as: Methods Enzymol. 2021 Jul 27;658:1–24. doi: 10.1016/bs.mie.2021.06.023

Locating chemical modifications in RNA sequences through ribonucleases and LC-MS based analysis

Priti Thakur 1, Scott Abernathy 1, Patrick A Limbach 1, Balasubrahmanyam Addepalli 1,1
PMCID: PMC9680040  NIHMSID: NIHMS1848807  PMID: 34517943

Abstract

Knowledge of the structural information is essential for understanding the functional details of modified RNA. Cellular non-coding RNA such as rRNA, tRNA and even viral RNAs contain a number of post-transcriptional modifications with varied degree of diversity and density. In this chapter, we discuss the use of a combination of biochemical and analytical tools such as ribonucleases and liquid chromatography coupled with mass spectrometry approaches for characterization of modified RNA. We present the protocols and alternate strategies for obtaining confident modified sequence information to facilitate the understanding of function.

Keywords: Post-transcriptional modifications, ribonucleases, LC-MS, RNA modification mapping, overlapping digestion products, oligonucleotide sequencing

1. INTRODUCTION

The canonical ribonucleosides (rA, rG, rC, and rU) of various RNA types are decorated with a number of additional chemical groups referred to as post-transcriptional modifications (PTMs)(Barbieri & Kouzarides, 2020; Boccaletto et al., 2018). These chemical functionalities installed by site-specific enzymes affect RNA stability, folding and decoding of mRNA during protein synthesis (Accornero et al., 2020). PTMs also regulate gene expression, development, cell identity, immunomodulation, and associated with human diseases (Chen et al., 2020; Haruehanroengra et al., 2020). Information related to the identity and location of modification in the sequence is critical to the understanding of structure-based function of modified RNA. Location-specific RNA modification analysis, termed as RNA modification mapping, is obtained through either high-throughput technique based on Next-Generation Sequencing (NGS) (Motorin & Helm, 2019) or mass spectrometric technologies (Jora et al., 2019).

NGS-based methods provide a readout of modification locations at single base resolution following reverse transcription (RT) of RNA to cDNA from minute sample amounts (~10 ng). Location-specific occurrence is inferred by indirect signals such as RT stops, nucleobase-specific reactivity, differential analysis, nucleotide skipping events, substitution, etc., from the NGS reads (Limbach & Paulines, 2017). NGS method are highly suitable for prediction of the locations of a single or small group of modifications (Helm & Motorin, 2017).

Liquid chromatography coupled with mass spectrometry (LC-MS) analysis provides direct and sensitive detection of every chemical modification in RNA. This platform depends on the physico-chemical properties of modified nucleoside such as altered chromatographic and mass spectrometric behavior apart from changes in mass values. Pioneered by the McCloskey group, initially, this approach establishes the identity of resident chemical modifications (Pomerantz & McCloskey, 1990). MS-based methods also allow the sequencing modified RNA through collision-induced dissociation based tandem mass spectrometry (MS/MS) (Kowalak et al., 1993).

Sequencing RNA through mass spectrometry is accomplished from either intact RNA (Top-down) or oligonucleotides (Bottom-up). While the top-down approaches are currently limited to highly purified RNA (Taucher & Breuker, 2012), the bottom-up approaches can characterize unknown modified tRNA with minimal purification/enrichment requirements (Yu et al., 2019). In this chapter, we discuss the enzymatic RNA cleavage methods that can generate sequence overlapping oligonucleotide digestion products while keeping the modification groups intact.

2. CHARACTERIZATION OF CHEMICAL MODIFICATIONS IN RNA

Knowledge of PTMs in modified RNA is obtained by two steps. First, the RNA is hydrolyzed to individual building blocks and subjected to chromatographic analysis to establish the identity of resident modifications. Techniques such as thin-layer chromatography (Grosjean et al., 2007) identify abundant modifications. A combination of liquid chromatography coupled with mass spectrometry establish the identity, census and elemental composition of unknown modification (Pomerantz & McCloskey, 1990). In the second step, location-specific information is obtained by partial digestion of RNA and LC-MS based sequencing.

2.1. Identity and census of resident modifications:

Modifying chemical groups are present on nucleobase and ribose sugar (but not on phosphodiester backbone) and they exhibit corresponding alterations in mass values (Figure 1). RNA is treated with nucleases such as nuclease P1, benzonase, phosphodiesterase to hydrolyze phosphodiester bonds followed by removal of phosphate groups by phosphatase to yield nucleosides. This mixture consisting of both modified and unmodified nucleosides are separated on chromatographic column for sensitive and accurate detection of mass-to-charge (m/z) values corresponding to the mass shifts (Figure 2) (Cai et al., 2015; Pomerantz & McCloskey, 1990). Relative changes in PTM levels in response to external stimuli are also revealed by LC-MS analysis either in a targeted (defined set) or untargeted (undefined and comprehensive) fashion as a fold change (Basanta-Sanchez et al., 2016; Su et al., 2014; Thüring et al., 2016). Relative quantification is more commonly used than absolute quantification (moles) performed with an external or internal standard.

Figure 1:

Figure 1:

Open in a new tab

Post-transcriptional nucleoside modifications in RNA. Modifying group is shown in red font. Canonical nucleosides, A, G, C, and U and mass-to-charge (m/z) values of protonated form are shown. m7G:7-methylguanosine, yW:wybutosine, m1A:1-methyladenosine, ms2i6A:2-methylthio-N6-isopentenyladenosine, s4U:4-thiouridine, cmnm5s2U:5-carboxymethylaminomethyl-2-thiouridine, m4Cm:N4,2’-O-dimethylcytidine, ac4C:N4-acetylcytidine

Figure 2.

Figure 2.

Open in a new tab

Characterization of chemical modifications in RNA through LC-MS/MS analytical platform. (A) Workflow for identification and census of nucleoside modifications. (B) Locating the positions of chemical modifications in the RNA sequence through digestion with nucleobase-specific ribonucleases and oligonucleotide sequencing. Chemical modifications are denoted by a star on the canonical nucleoside.

2.2. Locating the chemical modification in the RNA sequence:

Complete hydrolysis of RNA to nucleosides loses the position-specific information of modifications. Sufficient sequence context of modification has to be preserved for identification of location. Partial hydrolysis through chemical or enzymatic treatments can provide sequence context. Acid (Björkbom et al., 2015) or alkali (Donis-Keller et al., 1977) treatment allows random cleavage of phosphodiester backbone to generate oligonucleotides of suitable size for MS-based sequencing through collision induced dissociation (CID). These oligonucleotides could also exhibit sequence overlaps which in turn help determine the overall sequence. Limitations of chemical hydrolysis include requirement of high purity level for target RNA, poor control on the extent of cleavage, and potential degradation of the modification group under extreme pH conditions.

Enzymatic digestion and modification assignment:

Sequence context of the modification is preserved by enzymatic treatment with nucleobase-specific ribonucleases. Such oligonucleotide digestion products vary in their length depending on the position of recognized nucleobase in the sequence. Selective RNA cleavage offers advantage of restricting the compositional value decreasing the number of potential base compositions for a given mass measurement (Kowalak et al., 1993). Here, the modifications are mapped to the specific sites by monitoring the mass shift/increase of oligonucleotide molecular ion and its sequence informative gas phase fragment ions during tandem mass spectrometry methods (Figure 3). The interpreted oligonucleotide sequences are matched against the target RNA sequence to assign the locations of modifications.

Figure 3:

Figure 3:

Open in a new tab

Collisional induced dissociation of phosphodiester backbone in an oligonucleotide. Cleavage of P-O bond followed by C-O bond are predominant in RNA which leads to generation of c and y, a-B, and w complementary fragment ions, respectively.

Challenges and potential solutions of RNA modification mapping:

Length and uniqueness of oligonucleotide digestion products determine the number of matches on the target sequence. While shorter RNA targets and longer oligonucleotides have fewer number of matches, longer RNA targets will have multiple matches and fewer unique digestion products. In general, nucleotide composition, recognized nucleobase distribution, cleavage patterns and kinetics of ribonuclease determine the uniqueness and the length of oligonucleotides. Sequences rich in recognized nucleobase will yield smaller products leading to multiple matches, assignment ambiguities and uncertainties. Therefore, use of more than ribonuclease is common in RNA modification mapping experiments for confident assignment of modification groups. For example, ribonuclease T1 cleaves RNA at 3’-end of guanosine (rG); if the sequence is rich in rG, the digestion products would mostly be monomers, dimers, and trimer nucleotides, which cannot be uniquely placed to the target RNA sequence. In such cases, digestion products generated by RNase A (Mandal et al., 2010) or U2 (Houser et al., 2015; Solivio et al., 2018) or cytidine-specific cusativin (Addepalli et al., 2017) or uridine-specific MC1 (Addepalli et al., 2015) enzymes can fill the gaps left by RNase T1 coverage.

Knowledge of ribonuclease cleavage behavior:

Understanding the cleavage patterns of modified RNA by a given ribonuclease is critical for efficient and successful characterization of modified oligonucleotide sequences. In general, Ribonuclease A cleaves RNA at 3’-end of pyrimidines, U2 prefers purines, cusativin cleaves 3’-end of cytidine and MC1 cleaves at 5’-end of uridine. The cleavage patterns changes if the nucleobase is modified. For example, ribose methylation prohibits the RNA cleavage. Similarly, hypermodifications also inhibit the cleavage (Addepalli et al., 2015; Addepalli & Limbach, 2016; Wong et al., 2013). Cusativin does not cleave the phosphodiester bond between two cytidines leading to longer digestion products (Addepalli et al., 2017) (Figure 2).

Accurate knowledge of the cleavage behavior any ribonuclease also allows confident assignment of enzymatic nucleoside modifications during characterization (Addepalli & Limbach, 2016; Thakur et al., 2020; Yu et al., 2019). Confirmation by more than one ribonuclease digestion product allows independent corroboration of the modified sequence. Use of complementary nucleobase-specific ribonuclease enzymes have the high potential to generate overlapping digestion products that increase sequence coverage of a complex tRNA mixture or a larger RNA molecule such as ribosomal RNA (Thakur et al., 2020). Such a technology also has the capability to locate the oxidative changes to RNA sequences (Sun et al., 2020).

Alternatively, enzymes with broad specificity could also be used for RNA modification mapping in various situations. They include pyrimidine-specific RNase A which is suitable for characterization of modifications in purine-rich regions of RNA (Mandal et al., 2010). RNase U2 is a purine selective enzyme that would yield both 3’-linear phosphate and 2’,3’-cyclic phosphate products (Egami et al., 1980). Treatment with bacteriophage lambda protein phosphatase can convert the cyclic phosphate to linear phosphate (Houser et al., 2015).

Kinetically slower enzymes with limited specificity can yield long, overlapping oligonucleotide digestion products improving the sequence coverage. One such example is a nonspecific RNase U2-E49A mutant enzyme. This enzyme is robust to resist temperature changes (37-65 °C) generating digestion products with overlapping features to provide 100% sequence coverage with a highly purified tRNA species (Solivio et al., 2018).

Data analysis tools:

Interpretation of LC-MS/MS spectral data into sequence information is done either manually or in an automated fashion. The MS/MS spectra contain sequence informative fragment ions generated from oligonucleotide anions of specific m/z value following collision-induced dissociation. Such fragment ions occur as series with a common 5’-end (c1, c2, c3….cn) or 3’-end (y1, y2, y3….yn). Mass shift at specific fragment ion position and continuation in the series denote the position of modification in the sequence. Such an exercise is not a high throughput as it requires assignment of individual precursor m/z value to oligonucleotide and assignment of fragment ion series in the MS/MS spectra. Throughput may be increased by employing the data interpretation tools that include as Simple Oligonucleotide Sequencer (Rozenski & McCloskey, 2002), Ariadne (Nakayama et al., 2009), OMA/OPA (Nyakas et al., 2013), RoboOligo (Sample et al., 2015), RNAModMapper (Peter A. Lobue et al., 2019; Yu et al., 2017), and NucleicAcidSearchEgine (NASE) (Wein et al., 2020). A majority of these programs are standalone and require installation in the local computer for genomic RNA database searches. A few of them such as NASE (Wein et al., 2020) also offer false discovery rate and automated data analysis workflow features.

3. TOOLS TO CHARACTERIZE MODIFIED RNA SEQUENCE

3.1. Equipment (available through multiple vendors)

  1. Low-pressure chromatography columns

  2. Low-pressure chromatography system with UV detection

  3. Incubator cum shaker for cell culture on solid media (plates) and liquid media

  4. UV/visible spectrophotometer

  5. Analytical balance

  6. Bath sonicator

  7. Pipettes ranging from 10 μL to 10 mL

  8. Autoclave

  9. Sterile petri plates

  10. Erlenmeyer flasks (500 mL to 2 L capacity)

  11. Test tubes (20-25 mL capacity) with caps

  12. High speed centrifuge (x 20,000 g) with capacity for 1 L and 50 mL centrifuge tubes

  13. Microfuge (x14,000 g)

  14. Protein concentrator or disposable ultrafiltration centrifugal devices with 3 kDa MWCO

  15. Dry bath (30-70 °C)

  16. Speedvac or lyophilizer

  17. High Pressure Liquid Chromatography or Ultra High-Pressure Liquid Chromatography System with auto sampler

  18. Mass spectrometer with capabilities for tandem mass spectra (MS/MS)

  19. Total recovery clear glass vials w/bonded PTFE/silicone septum

  20. Tube racks

  21. Sterile micropipette tips and boxes

  22. Biohazardous waste disposal

  23. pH meter

  24. Polyacrylamide gel electrophoresis system with appropriate power supply system

3.2. Chemicals

  1. Bactotryptone (BD Biosciences, San Jose CA, USA)

  2. Sodium chloride (Thermo Fisher Scientific, Waltham, MA, USA)

  3. Yeast extract (BD Biosciences, San Jose CA, USA)

  4. Bactoagar (BD Biosciences, San Jose CA, USA)

  5. Trizma base (Thermo Fisher Scientific, Waltham, MA, USA)

  6. Hydrochloric acid (Thermo Fisher Scientific, Waltham, MA, USA)

  7. Potassium chloride (Thermo Fisher Scientific, Waltham, MA, USA)

  8. Magnesium chloride (Thermo Fisher Scientific, Waltham, MA, USA)

  9. Antibiotics ampicillin, chloramphenicol (Gold Biotechnology, St. Louis, Missouri, USA)

  10. Imidazole (Sigma-Aldrich, St. Louis, Missouri, USA)

  11. EDTA (Thermo Fisher Scientific, Waltham, MA, USA)

  12. Sodium hydroxide (Thermo Fisher Scientific, Waltham, MA, USA)

  13. Isopropyl β-D-1-thiogalactopyranoside (IPTG) (Gold Biotechnology, St. Louis, Missouri, USA)

  14. His-tag protein purification kit (MilliporeSigma, Burlington, MA, USA)

  15. Bradford reagent (Thermo Fisher Scientific, Waltham, MA, USA)

  16. Denaturing polyacrylamide gels (Thermo Fisher Scientific, Waltham, MA, USA)

  17. Gel staining solution and system (Thermo Fisher Scientific, Waltham, MA, USA)

  18. Protein size standards (Thermo Fisher Scientific, Waltham, MA, USA)

  19. Ammonium acetate (Thermo Fisher Scientific, Waltham, MA, USA)

  20. Zinc chloride (Thermo Fisher Scientific, Waltham, MA, USA)

  21. Betaine (Sigma-Aldrich, St. Louis, Missouri, USA)

  22. Nuclease P1 (Sigma-Aldrich, St. Louis, Missouri, USA)

  23. Phosphodiesterase I (Worthington Biochemical Corporation, Lakewood, NJ, USA)

  24. FastAP phosphatase (Thermo Fisher Scientific, Waltham, MA, USA)

  25. Nucleoside standards (rA, rG, rC, rU or 5-methylcytidine (m5C)) (Sigma-Aldrich, St. Louis, Missouri, USA)

  26. Single tRNA standard such as yeast tRNAPhe (Sigma-Aldrich, St. Louis, Missouri, USA)

  27. RP-18 column with embedded polar groups such as Acquity UPLC HSS T3 column with 1.8 μm particle size, 1.0 mm X 100 mm (Waters, Milford MA, USA)

  28. Triethylamine (Sigma-Aldrich, , St. Louis, Missouri, USA)

  29. Hexafluoroisopropanol (Sigma-Aldrich, , St. Louis, Missouri, USA)

  30. Xbridge BEH C18 XP column (Waters, Milford MA, USA)

  31. Ribonuclease T1 (Worthington Biochemical Corporation, Lakewood, NJ, USA)

  32. RNase A (Sigma-Aldrich, , St. Louis, Missouri, USA)

  33. Ribonuclease U2 (Abcam, Cambridge, MA, USA)

  34. Bacteriophage lambda protein phosphatase (New England Biolabs, Ipswich, MA, USA)

  35. Phenol saturated with sodium citrate buffer (pH 5.5) (Sigma-Aldrich, , St. Louis, Missouri, USA)

  36. Chloroform (Sigma-Aldrich, St. Louis, Missouri, USA)

  37. Bacterial strain with pET-22b-Cusativin

  38. Bacterial strain with pET-22b-MC1

  39. Ultrapure type I water (as per American Society for Testing and Materials standard)

  40. Methanol (LC-MS grade) (Thermo Fisher Scientific, Waltham, MA, USA)

  41. Acetonitrile (LC-MS grade) (Thermo Fisher Scientific, Waltham, MA, USA)

  42. Isopropanol (LC-MS grade) (Thermo Fisher Scientific, Waltham, MA, USA)

  43. Calibration solution specific to the mass spectrometer (Thermo Fisher Scientific, Waltham, MA, USA)

  44. Oligo d(T) standards d(T)3, d(T)5, d(T)15 (IDT DNA Technologies, Coralville, IA, USA)

4. PROTOCOLS

4.1. RNA hydrolysis to nucleosides

RNA hydrolysis conditions are optimized to avoid modification artifacts. Details of precautions during RNA and buffer preparation, modification standard generation for calibration curve are provided elsewhere (Jora et al., 2021; Thüring et al., 2016). Basic steps of RNA hydrolysis conditions are listed here.

1.1 All the sample preparation steps have to be done with ultrapure water and devoid of any alkali salts. All enzyme solutions, reagents are aliquoted in 10-20 μL portions and store at −20 °C.

1.2 Nuclease P1 buffer: After filtering through a 0.22 μm filter, mix 9 parts of 250 mM ammonium acetate solution with 1 part of ZnCl2 solution, both initially adjusted to pH 5 with acetic acid.

1.3 Dissolve nuclease P1 in ultrapure water to get 0.3 U/μL.

1.4 Snake venom phosphodiesterase I is diluted to 0.1 U/μL.

1.5. Dilute Fast Alkaline phosphatase to 1 U/ μL

1.6 RNA preparation: RNA with alkali salts is reprecipitated with 1/3 volume of 7.5 M ammonium acetate and 2.5-3 volumes of absolute ethanol at −20 °C overnight. The RNA is pelleted by centrifugation at 12,000xg for 15 min. Decant the supernatant, add 0.5-1 mL of 75% ethanol, vortex at medium speed and centrifuge it to pellet the RNA. The pellet is resuspended in ultrapure water after air drying.

1.7 Ten μg of standard tRNA is diluted in 20 μL of water and exposed to 95 °C for 5 min followed by instant cooling in ice bucket for at least 2 min.

1.8 Add 2.2 μL of nuclease P1 buffer to the cooled RNA, 0.3 U of nuclease P1, 0.1 U of phosphodiesterase I, and 1 U of FastAP, mix gently. Incubate at 37 °C for 5 h (Jora et al., 2021).

1.9 Dry the mixture in a speedvac under cold conditions for storage at −20 °C.

2.0 Dried samples are resuspended in mobile phase A just before the LC-MS.

4.2. LC-MS based identification of modifications

Reversed phase liquid chromatography in combination with diode array or UV detector or mass spectrometry is the method of choice for ribonucleoside characterization. The LC system can be a binary or quaternary pump capable of generating solvent gradient for analyte separation. The autosampler enables injection of multiple samples to avoid manual errors. The column could be any RP-18 column with polar embedded groups to allow binding and separation based on hydrophobicity.

1.1 The LC method uses 5.3 mM ammonium acetate in water (pH 4.5) as mobile phase A (MPA) and acetonitrile/water (40:60 v:v) mixture with 5.3 mM ammonium acetate as mobile phase B (MPB).

1.2 For an Acquity UPLC HSS T3, 1.8 μm particle size, 1.0 X 100 mm column, the solvent gradient contains a hold of 7.6 min at 0% MPB, increase it to 2% at 15.7 min, 3% at 19.2 min, 5% at 25.7 min, 25% at 29.5 min, 50% at 32.3 min, 75% at 36.4 min, 99% at 39.6 min with a hold of 7.4 min, and returning to 0% MPB at 46.9 min for equilibration for 18 min at flow rate of 100 μL min−1 at 30 °C (Jora et al., 2021).

1.3 Two to five μg of RNA or 25 ng (~100 pmol) of ribonucleoside is sufficient for unambiguous detection and differentiation.

1.4 The column eluent is forced through a capillary while subjecting electric field with simultaneous solvent evaporation to enable electrospray ionization (ESI) at the entrance of mass spectrometer.

1.5 Data acquisition entails establishment of a tune file to get an optimal signal for ribonucleoside molecular ion in positive polarity. Such a tune file is prepared through infusion of nucleoside standard such as canonical nucleosides or 5-methylcytidine.

1.6 Orbitrap based mass analyzers acquire data under a defined scan range of masses (220-900 m/z) at resolution of 120,000, AGC (automatic gain control) of 2.0 x 105, and injection time of 100 milli seconds (ms). Data-dependent top speed MS/MS (3 scans/cycle) is acquired at resolution of 15,000, AGC of 5.0 x 104 and injection time of 200 ms.

1.7 Either collision induced dissociation (CID) or higher-energy collisional dissociation (HCD) can be used for characterization of modification on nucleobase through MS/MS events.

1.8 Qual browser feature of Xcalibur is used for qualitative analysis for Orbitrap data

1.9 Quantitative analysis is performed with triple quad mass spectrometer through optimized selected reaction monitoring (SRM) transitions involving a defined m/z of ribonucleoside molecular ion and its fragment ion forming a pair (Jora et al., 2021; Thüring et al., 2016).

4.3. Oligonucleotide generation for location-specific information

RNA modification mapping procedures provide the location-specific information. While some ribonucleases are commercially available, others are purified from overexpression strains of bacteria. Enzymes prepared with ammonium salts generally do not require further purification.

4.3.1. Nucleobase-specific ribonucleases

Ribonuclease T1:

RNase T1 cleaves RNA at 3’-end of guanosine and is widely used for RNA modification mapping. Alkali salts, Tris-buffers, and other preservative chemicals used in the vendor supplied formats of ribonuclease T1 interfere with LC-MS data acquisition. They are eliminated by protein precipitation.

1.1 Add cold acetone (2 parts cold acetone for 1 part of ribonuclease T1) to the protein solution, invert 3-4 times and precipitate overnight at −20 °C.

1.2 The precipitated protein is pelleted by centrifugation for 10 min at 12,000xg and the supernatant decanted. Add 1 mL of 75% cold acetone to the pellet, vortex gently at medium speed, centrifuge for 10 min at 12,000xg and decant the supernatant.

1.3 Air dry the pellet briefly and dissolve in 0.5 mL of sterile water.

1.4 The dissolved protein is purified through a sep-pak C18 cartridge (Waters) as per the vendor’s guidelines. It involves conditioning with 100% acetonitrile (1 mL), water (1 mL) followed by loading of enzyme in a dropwise fashion.

1.5 The cartridge is washed with water (1 mL) and the protein eluted with 75% acetonitrile.

1.6 The eluted protein is aliquoted to 500 U and dried in speedvac for long term storage at 4 °C.

1.7 The dried protein is resuspended in 220 mM ammonium acetate (pH 5.5-6.0) solution to obtain 25-50 U μL−1.

1.8 About 25-30 pmol (~0.5 μg) of single tRNA (yeast tRNAPhe) is denatured at 95 °C for 5 min followed by snap cooling at 4 °C

1.9 Add 10 μL of 220 mM ammonium acetate and T1 (25 U ) per μg RNA and incubate at 37 °C for 2 h.

1.10 The digest is dried in speed vac and resuspend in Mobile Phase A just before LC-MS injection. A representative example for LC-MS based sequencing of a T1 digestion product from anticodon region of yeast tRNAPhe is shown in Figure 4A.

Figure 4:

Figure 4:

Open in a new tab

MS-based sequencing of ribonuclease digestion products from yeast tRNAPhe. (A) G-specific ribonuclease T1 and (B) C-specific cusativin products

Ribonuclease cusativin:

Cusativin is not commercially available. The enzyme is overexpressed from a recombinant protein expression plasmid (pET 22b-Cus) in E. coli as His-tag fusion protein and purified from a nickel sulfate column (Addepalli et al., 2017).

1.1 Streak BL21(DE3) strain of E. coli bearing pET-22b-Cus on LB-agar-ampicillin (50 μg mL−1) plate and grow overnight at 37 °C to obtain single colonies.

1.2 Inoculate 10 mL of LB broth media supplemented with ampicillin (50 μg mL−1) with a single colony and grow overnight on an incubator-shaker at 37 °C and 200 rpm.

1.3 Use the above starter culture to inoculate 1 L of fresh LB-ampicillin broth media and grow at 37 °C in an incubator-shaker. Check the optical density (OD600 nm) of the culture at 1-2 h intervals.

1.4 When the OD600 reaches 0.6-0.7, add IPTG (1 M stock. 600 μL) to 0.6 mM and grow for another 3h.

1.5 Harvest cells through centrifugation at 5,000xg for 15 min and decant the spent media to biohazardous waste.

1.6 Resuspend the cells in lysozyme buffer (25 mM Tris-HCl pH 7.5, 60 mM KCl, 10 mM MgCl2) in minimal volume (~10 mL) and store at −80 °C.

1.7 Lyse the cells by adding lysozyme solution to cell suspension (1 mg mL−1) and incubate on ice for 30-40 min.

1.8 Centrifuge the lysate at 15,000xg for 15 min and decant the clear solution into a clean tube.

1.9 Protein purification by column chromatography is preferred. Either His-tag protein purification kits or Ni-NTA agarose can be used as per manufacturer’s guidelines.

1.10 We find elution buffer rather than strip buffer most suitable for optimal protein elution.

1.11 The purified protein is concentrated through ultrafiltration using 3000 MWCO filter and the buffer exchanged with 220 mM ammonium acetate pH 4.5

1.12 Protein concentration is estimated by Bradford reagent using BSA as standard.

1.13 Purity of the ribonuclease (~25 kDa) is ascertained by denaturing SDS-PAGE gels.

1.14 The protein is aliquoted into small volumes and stored at −20 °C. Addition of glycerol (10%) or betaine (0.5 M) increases the shelf life.

1.15 The ratio of protein to RNA in a digest is empirically determined for optimal digestion.

1.16 To denatured RNA, add 10 μL of 220 mM ammonium acetate, and cusativin to incubate at 62 °C for 90 min. Typically, 0.8-1.0 μg of purified preparation is required for 1 μg RNA.

1.17 The digest is dried in speedvac and stored at −20 °C until the LC-MS analysis. A representative cusativin digestion product of tRNAPhe is shown in Figure 4B.

Ribonuclease MC1:

The uridine-specific MC1 ribonuclease requires purification from recombinant E. coli Rosetta cells that contain pET-22b-MC1 plasmid (Addepalli et al., 2015).

1.1 The recombinant E. coli cells are cultured in LB media supplemented with ampicillin (50 μg mL−1) and chloramphenicol (32 μg mL−1). Rest of the protein expression and purification procedures are identical to the cusativin enzyme.

1.2 The purified protein exhibits a molecular weight of ~23 kDa on denaturing SDS-PAGE.

1.3 Just like cusativin, optimal ratio of RNA to protein is empirically estimated for optimal digestion of RNA. Typically, equal amount of protein and RNA ( 1:1 w:w ratio) provide optimal digestion when incubated at 37 °C for 2 h.

4.3.2. Nucleoside-preferential ribonucleases (alternate approach)

Ribonuclease A:

This RNase actively cleaves RNA at every pyrimidine residue. It also contaminates LC system thereby interfering with the analysis of subsequent LC-MS injections. Therefore, it requires removal of enzyme.

1.2 To 10 μg of denatured RNA, add 20 μL of 220 mM ammonium acetate and 0.1 U of ribonuclease A and incubate at 37 °C for 2 h.

1.3 Make up the volume to 100 μL with ammonium acetate solution

1.4 Add 1 volume (100 μL) of phenol (saturated with sodium citrate pH 5.5) and chloroform mixture (1:1 v:v), vortex it and centrifuge at 12,000xg for 10 min. Carefully aspirate the aqueous phase and transfer it to a clean tube.

1.5 Add 1 volume of chloroform, vortex, and centrifuge to collect the aqueous phase.

1.6 The aqueous phase is dried in a speedvac for storage at −20 °C.

RNase U2:

RNase U2 cleaves RNA at 3’-end of purine nucleotides and is available commercially.

1.1 RNA to protein ratio recommended by the vendor is to be followed. RNA and U2 are mixed with 20 μL of 220 mM ammonium acetate and incubated at 65 °C for 30 min.

1.2 Digestion by U2 generates a mixture of oligonucleotides with 2’,3’-cyclic phosphate and 3’-linear phosphate.

1.3 Reconstitute the dried RNase U2 digest (10 μg) in 6.5 μL of water and 2.5 μL of DTT (0.1 M), 2,5 μL of Tris-HCl (1 M, pH 7.6), 12.5 μL of MnCl2 (1 mM) and 1 μL of bacteriophage lambda protein phosphatase (80 U) and incubate at 37 °C for 3 h to convert cyclic phosphate to linear one (Houser et al., 2015) .

1.4 If needed, the digest is extracted with phenol-chloroform and dried in speedvac for storage.

4.4.3. Non-specific ribonucleases

Another alternative approach is to use a ribonuclease that exhibit limited cleavage specificity but show high tendency to undercut the RNA. One such enzyme is ribonuclease U2 E49A mutant that has a substitution of alanine at position 49 in place of glutamate. This enzyme generates long and overlapping digestion products with 100% sequence coverage of a purified RNA species in a single digest.

1.1 This protein is produced from a overexpression plasmid pET22b-U2 E49A from Rosetta strain of E. coli (Solivio et al., 2018).

1.2 The cell culture media is supplemented with ampicillin and chloramphenicol antibiotics for cell growth. The protein expression and purification protocols are identical to the one described for cusativin purification.

1.3 The purified protein exhibits ~11 kDa band and is stored in aliquots at −20 °C.

1.4 Typically, ~5 ng of purified RNase U2 E49A mutant protein is sufficient to digest 5 μg RNA.

1.5 Add 10 μL of 220 mM ammonium acetate solution to the denatured RNA followed by enzyme.

1.6 Incubate the mixture with a linear temperature gradient from 28 °C to 65 °C for 7-10 min. Dry the sample in speedvac and store at −20 °C until LC-MS injection.

1.7 Such conditions provided near 100% sequence coverage when tested against Saccharomyces cerevisiae tRNAPhe (Solivio et al., 2018). However, its utility for RNA mixtures is yet to be demonstrated.

4.4. LC-MS analysis for sequencing the oligonucleotides.

Comprehensive characterization of the nucleotide sequences of oligonucleotide digestion products is accomplished by reversed-phase liquid chromatography or hydrophilic interaction liquid chromatography (P. A. Lobue et al., 2019) coupled with mass spectrometry. For separation of oligonucleotide mixtures using C18 reversed phase column, the negative charge on phosphate backbone must be masked with an alkylamine such as triethylamine (TEA). Besides acting as an ion pairing reagent, this reagent molecule makes the oligonucleotide more hydrophobic and allows chromatographic retention. However, mass analysis requires TEA removal and is facilitated by the addition of second organic modifier, such as 1,1,1,3,3,3-hexafluoro-2-isoproyl alcohol (HFIP). The low boiling point of HFIP allows rapid evaporation from droplet leading to increased pH and TEA dissociation from oligonucleotide anions that are sampled by the mass spectrometer.

1.1 Mobile phase A (MPA) consists of 8 mM TEA and 200 mM HFIP in water. Add HFIP to water and mix it with a stir bar. While stirring add TEA drop-by-drop to the HFIP solution and adjust the pH to be within the range of 7.5 to 7.8.

1.2 Mobile phase B (MPB) is prepared in water: methanol (1:1 v:v) mixture with 200 mM HFIP and 8 mM TEA.

1.3 Just before LC-MS injection, the dried RNA digest is dissolved in mobile phase A (2-7 μL, depending on the column diameter) and transferred to the sample injection vial.

1.4 Typically, the oligonucleotide sample is injected into the column such as Xbridge BEH C18 column (Waters) at 8% B.

1.5 After an initial hold at 8% B for 1.5 min, MPB is increased to 65% at 32 min, 95% at 39 min, and switched to 8% at 41 min for re-equilibration.

1.6 The column is generally maintained around 50 °C temperature, and the mobile phase flow rate depends on the column diameter and manufacturer’s guidelines.

1.7 The column flow is directed to the mass spectrometer through a capillary for electrospray ionization (ESI).

1.8 Just like nucleoside analysis, a tune file is established for optimal detection of oligonucleotides using synthetic d(T)5, d(T)10 or d(T)15 standards. Acquisition is typically made in negative polarity with a scan range of m/z 400-2000.

1.9 Electrospray ionization conditions depend on the type of mass spectrometer. For a Synapt G2-S Q-TOF (quadrupole-time of flight) instrument (Waters), typical setting includes a source voltage of 2.7 kV, sampling cone 30 V, source and desolvation temperatures of 120 °C and 350 °C, respectively.

1.10 The cone and desolvation gas flow rates are 10 and 800 L h−1.

1.11 Data acquisition is done in sensitivity (V)-mode with 1.0 s scan time with three data-dependent MS/MS scans for the three topmost molecular ions. Fragmentation is done through collision-induced dissociation.

1.12 These parameters could vary for other instruments such as Orbitrap based Thermo Scientific instruments (Peter A. Lobue et al., 2019)

4.5. Data analysis

The primary limitations of RNA modification mapping during oligonucleotide sequencing include lack of throughput during data acquisition and data interpretation. While the former is because of sample complexity and instrument limitations, the latter is due to the laborious data interpretation (assigning fragment ions to each MS/MS spectrum). To alleviate this problem, computational tools have been developed including Mongo Oligo Mass calculator (http:// http://rna.rega.kuleuven.be/masspec/mongo.htm), RNAModMapper (RAMM) (Yu et al., 2017), and recently NASE (Wein et al., 2020).

1.1 Data analysis of the known modified oligonucleotides is typically performed with the instrument-specific software for detection (qualitative analysis) and computing the quantum of signal as peak area (quantitative analysis).

1.2 Mongo Oligo mass calculator can compute the theoretical m/z values of oligonucleotide digestion products of target RNA. It also provides a list of fragment ions formed from oligonucleotide anions of defined m/z value. These sequence informative fragment ions share features corresponding to either 5’-end (c1, c2…cn or a-B2, a-B3…a-Bn) or 3’-end (y1, y2…yn or w1, w2…wn) depending on the position of cleavage, thus enabling the reconstruction of the sequence. Scoring 70-80% of the predicted fragment ions in the MS/MS spectra provides assurance of the presence of modified oligonucleotide.

1.3 Identification of modification and its location is deciphered by the corresponding mass shift of the oligonucleotide molecular ion as well as specific fragment ion and its series, which is a low throughput process.

1.4 In case the location-specific modification information is unknown or multiple modifications have to be mapped, computational tools are available for automated detection. One tool, RoboOligo allows both automated de novo and manual analysis of the MS/MS spectra of modified oligonucleotides with a high diversity of modifications (Sample et al., 2015)

1.5 Another tool, RNAModMapper (RAMM) interprets CID data of oligonucleotides and the interpreted sequences are mapped to the RNA sequences thus enabling high throughput RNA modification mapping by LC-MS/MS (Peter A. Lobue et al., 2019; Yu et al., 2017). Both RoboOligo and RAMM adopts instrument-independent analytical approaches where the instrument-specific raw data files are converted to mgf (mascot generic format) file for subsequent data processing.

1.6 RAMM uses RNA sequences (either modified or unmodified) as FASTA file for annotation.

1.7 In the mapping window, the enzyme used for sample preparation, precursor and product ion mass tolerances, mass type (average or monoisotopic), processing parameters such as probability or P-score (weighted by relative abundance differences of fragment ions) and dot product (quantitative measure of similarity between the observed and reconstructed spectra) are defined.

1.8 The RNase digestion products of RNA sequences of FASTA file are generated in silico and compared against the experimental MS/MS data of oligonucleotides and scored.

1.9 The interpreted oligonucleotides are mapped onto the original sequences and modifications annotated in a location-specific manner.

1.10 RAMM facilitates data interpretation in fixed (for known modified RNA sequences) and variable position mapping (for unknown sequences). LC-MS platform and data acquisition-specific sample analysis parameter thresholds for MS/MS spectra interpretation and annotation are detailed elsewhere (Peter A. Lobue et al., 2019).

1.11 Annotating the interpreted modified oligonucleotides of each ribonuclease to the target sequence shows the overlaps and reconstruction of modified RNA sequence (Thakur et al., 2020). An example of such representation of all ribonuclease digestion products for yeast tRNAPhe is shown in Figure 5. Although these computation tools speed up the mapping experiments, they cannot completely eliminate the manual review of mapping data.

Figure 5:

Figure 5:

Open in a new tab

Overlapping pattern of oligonucleotide digestion products of nucleobase-specific ribonucleases (T1, cusativin (Cus) and MC1) annotated to the modified yeast tRNAPhe sequence. The position of phosphodiester cleavage is denoted by the italicized and underlined nucleobase.

5. Summary

Locating RNA modification in RNA sequence is critical for understanding the biochemical function. The type of strategy to be adopted depends on the number of modifications to be mapped, nucleotide composition of RNA, ribonuclease specificity and capabilities of analytical platform. While NGS-based high approaches characterize simple modifications one or a few at a time, LC-MS/MS approaches are sensitive to every chemical modification in a comprehensive and unbiased manner. While digestion by just one ribonuclease may be sufficient to understand the modification status of known modification location (Wong et al., 2013), analysis of the RNA digests of multiple enzymes would be required for characterization of unknown modified RNA sequences. The overlapping pattern of the oligonucleotide digestion products of multiple enzymes provide confident reconstruction and characterization. If the target is a single RNA sequence use of a nonspecific enzyme can be sufficient. MS/MS data interpretation requires computational tools for automated mapping but cannot completely eliminate the manual review.

Acknowledgement

This work is supported by the National Institutes of Health (NIH GM058843), Rieveschl Eminent Scholar Endowment and University of Cincinnati.

References

  1. Accornero F, Ross RL, & Alfonzo JD (2020). From canonical to modified nucleotides: balancing translation and metabolism. Critical Reviews in Biochemistry and Molecular Biology, 55(6), 525–540. 10.1080/10409238.2020.1818685 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Addepalli B, Lesner NP, & Limbach PA (2015). Detection of RNA nucleoside modifications with the uridine-specific ribonuclease MC1 from Momordica charantia. Rna, 21(10), 1746–1756. 10.1261/rna.052472.115 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Addepalli B, & Limbach PA (2016). Pseudouridine in the Anticodon of Escherichia coli tRNATyr(QPsiA) Is Catalyzed by the Dual Specificity Enzyme RluF. J Biol Chem, 291(42), 22327–22337. 10.1074/jbc.M116.747865 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Addepalli B, Venus S, Thakur P, & Limbach PA (2017). Novel ribonuclease activity of cusativin from Cucumis sativus for mapping nucleoside modifications in RNA. Anal Bioanal Chem, 409(24), 5645–5654. 10.1007/s00216-017-0500-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Barbieri I, & Kouzarides T (2020). Role of RNA modifications in cancer. Nat Rev Cancer, 20(6), 303–322. 10.1038/s41568-020-0253-2 [DOI] [PubMed] [Google Scholar]
  6. Basanta-Sanchez M, Temple S, Ansari SA, D’Amico A, & Agris PF (2016). Attomole quantification and global profile of RNA modifications: Epitranscriptome of human neural stem cells. Nucleic acids research, 44(3), e26. 10.1093/nar/gkv971 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Björkbom A, Lelyveld VS, Zhang S, Zhang W, Tam CP, Blain JC, & Szostak JW (2015). Bidirectional Direct Sequencing of Noncanonical RNA by Two-Dimensional Analysis of Mass Chromatograms. J Am Chem Soc, 137(45), 14430–14438. 10.1021/jacs.5b09438 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Boccaletto P, Machnicka MA, Purta E, Piatkowski P, Baginski B, Wirecki TK, de Crecy-Lagard V, Ross R, Limbach PA, Kotter A, Helm M, & Bujnicki JM (2018). MODOMICS: a database of RNA modification pathways. 2017 update. Nucleic acids research, 46(D1), D303–D307. 10.1093/nar/gkx1030 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cai WM, Chionh YH, Hia F, Gu C, Kellner S, McBee ME, Ng CS, Pang YL, Prestwich EG, Lim KS, Babu IR, Begley TJ, & Dedon PC (2015). A Platform for Discovery and Quantification of Modified Ribonucleosides in RNA: Application to Stress-Induced Reprogramming of tRNA Modifications. Methods Enzymol, 560, 29–71. 10.1016/bs.mie.2015.03.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Chen K, Song B, Tang Y, Wei Z, Xu Q, Su J, de Magalhães JP, Rigden DJ, & Meng J (2020). RMDisease: a database of genetic variants that affect RNA modifications, with implications for epitranscriptome pathogenesis. Nucleic acids research. 10.1093/nar/gkaa790 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Donis-Keller H, Maxam AM, & Gilbert W (1977). Mapping adenines, guanines, and pyrimidines in RNA. Nucleic acids research, 4(8), 2527–2538. 10.1093/nar/4.8.2527 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Egami F, Oshima T, & Uchida T (1980). Specific Interaction of Base-Specific Nucleases with Nucleosides and Nucleotides. In Chapeville F & Haenni AL (Eds.), Chemical Recognition in Biology (pp. 250–277). Springer Berlin; Heidelberg. 10.1007/978-3-642-81503-4_21 [DOI] [PubMed] [Google Scholar]
  13. Grosjean H, Droogmans L, Roovers M, & Keith G (2007). Detection of Enzymatic Activity of Transfer RNA Modification Enzymes Using Radiolabeled tRNA Substrates. In Methods Enzymol (Vol. 425, pp. 55,57–101). [DOI] [PubMed] [Google Scholar]
  14. Haruehanroengra P, Zheng YY, Zhou Y, Huang Y, & Sheng J (2020). RNA modifications and cancer. RNA Biol, 1–16. 10.1080/15476286.2020.1722449 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Helm M, & Motorin Y (2017). Detecting RNA modifications in the epitranscriptome: predict and validate [Review]. Nat Rev Genet, 18(5), 275–291. 10.1038/nrg.2016.169 [DOI] [PubMed] [Google Scholar]
  16. Houser WM, Butterer A, Addepalli B, & Limbach PA (2015). Combining recombinant ribonuclease U2 and protein phosphatase for RNA modification mapping by liquid chromatography-mass spectrometry. Anal Biochem, 478, 52–58. 10.1016/j.ab.2015.03.016 [DOI] [PubMed] [Google Scholar]
  17. Jora M, Borland K, Abernathy S, Zhao R, Kelley M, Kellner S, Addepalli B, & Limbach PA (2021). Chemical Amination/Imination of Carbonothiolated Nucleosides During RNA Hydrolysis. Angewandte Chemie (International ed. in English), 60, 3961–3966. 10.1002/anie.202010793 [DOI] [PubMed] [Google Scholar]
  18. Jora M, Lobue PA, Ross RL, Williams B, & Addepalli B (2019). Detection of ribonucleoside modifications by liquid chromatography coupled with mass spectrometry. Biochim Biophys Acta Gene Regul Mech, 1862(3), 280–290. 10.1016/j.bbagrm.2018.10.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kowalak JA, Pomerantz SC, Crain PF, & McCloskey JA (1993). A novel method for the determination of post-transcriptional modification in RNA by mass spectrometry. Nucleic acids research, 21(19), 4577–4585. 10.1093/nar/21.19.4577 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Limbach PA, & Paulines MJ (2017). Going global: the new era of mapping modifications in RNA. Wiley Interdiscip Rev RNA, 8(1), e1367. 10.1002/wrna.1367 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lobue PA, Jora M, Addepalli B, & Limbach PA (2019). Oligonucleotide analysis by hydrophilic interaction liquid chromatography-mass spectrometry in the absence of ion-pair reagents. J Chromatogr A, 1595, 39–48. 10.1016/j.chroma.2019.02.016 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lobue PA, Yu N, Jora M, Abernathy S, & Limbach PA (2019). Improved application of RNAModMapper – An RNA modification mapping software tool – For analysis of liquid chromatography tandem mass spectrometry (LC-MS/MS) data. Methods, 156, 128–138. 10.1016/j.ymeth.2018.10.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Mandal D, Köhrer C, Su D, Russell SP, Krivos K, Castleberry CM, Blum P, Limbach PA, Söll D, & RajBhandary UL (2010). Agmatidine, a modified cytidine in the anticodon of archaeal tRNA(Ile), base pairs with adenosine but not with guanosine. Proc Natl Acad Sci U S A, 107(7), 2872–2877. 10.1073/pnas.0914869107 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Motorin Y, & Helm M (2019). Methods for RNA Modification Mapping Using Deep Sequencing: Established and New Emerging Technologies. Genes, 10(1), 35. https://www.mdpi.com/2073-4425/10/1/35 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Nakayama H, Akiyama M, Taoka M, Yamauchi Y, Nobe Y, Ishikawa H, Takahashi N, & Isobe T (2009). Ariadne: a database search engine for identification and chemical analysis of RNA using tandem mass spectrometry data. Nucleic acids research, 37(6), e47. 10.1093/nar/gkp099 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Nyakas A, Blum LC, Stucki SR, Reymond JL, & Schürch S (2013). OMA and OPA--software-supported mass spectra analysis of native and modified nucleic acids. J Am Soc Mass Spectrom, 24(2), 249–256. 10.1007/s13361-012-0529-1 [DOI] [PubMed] [Google Scholar]
  27. Pomerantz SC, & McCloskey JA (1990). Analysis of RNA hydrolyzates by liquid chromatography-mass spectrometry. Methods Enzymol, 193, 796–824. 10.1016/0076-6879(90)93452-q [DOI] [PubMed] [Google Scholar]
  28. Rozenski J, & McCloskey JA (2002). SOS: a simple interactive program for ab initio oligonucleotide sequencing by mass spectrometry. J Am Soc Mass Spectrom, 13(3), 200–203. 10.1016/s1044-0305(01)00354-3 [DOI] [PubMed] [Google Scholar]
  29. Sample PJ, Gaston KW, Alfonzo JD, & Limbach PA (2015). RoboOligo: software for mass spectrometry data to support manual and de novo sequencing of post-transcriptionally modified ribonucleic acids. Nucleic acids research, 43(10), e64. 10.1093/nar/gkv145 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Solivio B, Yu N, Addepalli B, & Limbach PA (2018). Improving RNA modification mapping sequence coverage by LC-MS through a nonspecific RNase U2-E49A mutant. Anal Chim Acta, 1036, 73–79. 10.1016/j.aca.2018.08.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Su D, Chan CT, Gu C, Lim KS, Chionh YH, McBee ME, Russell BS, Babu IR, Begley TJ, & Dedon PC (2014). Quantitative analysis of ribonucleoside modifications in tRNA by HPLC-coupled mass spectrometry. Nat Protoc, 9(4), 828–841. 10.1038/nprot.2014.047 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Sun C, Limbach PA, & Addepalli B (2020). Characterization of UVA-Induced Alterations to Transfer RNA Sequences. Biomolecules, 10(11). 10.3390/biom10111527 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Taucher M, & Breuker K (2012). Characterization of modified RNA by top-down mass spectrometry. Angewandte Chemie (International ed. in English), 51(45), 11289–11292. 10.1002/anie.201206232 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Thakur P, Estevez M, Lobue PA, Limbach PA, & Addepalli B (2020). Improved RNA modification mapping of cellular non-coding RNAs using C- and U-specific RNases. Analyst, 145(3), 816–827. 10.1039/c9an02111f [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Thüring K, Schmid K, Keller P, & Helm M (2016). Analysis of RNA modifications by liquid chromatography-tandem mass spectrometry. Methods, 107, 48–56. 10.1016/j.ymeth.2016.03.019 [DOI] [PubMed] [Google Scholar]
  36. Wein S, Andrews B, Sachsenberg T, Santos-Rosa H, Kohlbacher O, Kouzarides T, Garcia BA, & Weisser H (2020). A computational platform for high-throughput analysis of RNA sequences and modifications by mass spectrometry. Nat Commun, 11(1), 926. 10.1038/s41467-020-14665-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Wong SY, Javid B, Addepalli B, Piszczek G, Strader MB, Limbach PA, & Barry CE (2013). Functional role of methylation of G518 of the 16S rRNA 530 loop by GidB in Mycobacterium tuberculosis. Antimicrobial Agents and Chemotherapy, 57(12), 6311–6318. 10.1128/AAC.00905-13 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Yu N, Jora M, Solivio B, Thakur P, Acevedo-Rocha CG, Randau L, de Crecy-Lagard V, Addepalli B, & Limbach PA (2019). tRNA Modification Profiles and Codon-Decoding Strategies in Methanocaldococcus jannaschii. J Bacteriol, 201(9), e00690–00618. 10.1128/jb.00690-18 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Yu N, Lobue PA, Cao X, & Limbach PA (2017). RNAModMapper: RNA Modification Mapping Software for Analysis of Liquid Chromatography Tandem Mass Spectrometry Data. Anal Chem, 89(20), 10744–10752. 10.1021/acs.analchem.7b01780 [DOI] [PubMed] [Google Scholar]

RESOURCES