Quantitative analysis of rRNA modifications using stable isotope labeling and mass spectrometry - PubMed (original) (raw)

. 2014 Feb 5;136(5):2058-69.

doi: 10.1021/ja412084b. Epub 2014 Jan 27.

Affiliations

Quantitative analysis of rRNA modifications using stable isotope labeling and mass spectrometry

Anna M Popova et al. J Am Chem Soc. 2014.

Abstract

Post-transcriptional RNA modifications that are introduced during the multistep ribosome biogenesis process are essential for protein synthesis. The current lack of a comprehensive method for a fast quantitative analysis of rRNA modifications significantly limits our understanding of how individual modification steps are coordinated during biogenesis inside the cell. Here, an LC-MS approach has been developed and successfully applied for quantitative monitoring of 29 out of 36 modified residues in the 16S and 23S rRNA from Escherichia coli . An isotope labeling strategy is described for efficient identification of ribose and base methylations, and a novel metabolic labeling approach is presented to allow identification of MS-silent pseudouridine modifications. The method was used to measure relative abundances of modified residues in incomplete ribosomal subunits compared to a mature (15)N-labeled rRNA standard, and a number of modifications in both 16S and 23S rRNA were present in substoichiometric amounts in the preribosomal particles. The RNA modification levels correlate well with previously obtained profiles for the ribosomal proteins, suggesting that RNA is modified in a schedule comparable to the association of the ribosomal proteins. Importantly, this study establishes an efficient workflow for a global monitoring of ribosomal modifications that will contribute to a better understanding of mechanisms of RNA modifications and their impact on intracellular processes in the future.

PubMed Disclaimer

Figures

Figure 1

Figure 1

The qMS workflow for analysis of rRNA modifications. In a typical rRNA modification inventory experiment, rRNA isolated from a sample of interest (14N, red) is mixed with 15N-labeled external standard (blue), containing mature 16S or 23S rRNA. After ribonuclease cleavage, the mixture is submitted to LC-MS analysis. Pairs of 14N and 15N peaks are detected, and their masses are used for assignment. Experimental peaks are fitted to their theoretical isotope distributions, and obtained amplitudes are used to calculate the relative amounts of rRNA modifications present in the 14N-sample.

Figure 2

Figure 2

LC-MS separation and data fitting. Results of the control experiment in which 14N- and 15N-labeled and individually purified 16S RNA were mixed in a 1:1 molar ratio and digested with ribonuclease T1. (A) Low-resolution contour plot of the LC-MS run, showing pairs of the co-eluting 14N/15N rRNA fragments. Data were collected using negative ionization mode. (B) High-resolution LC-MS peak profiles (box 1), MS isotope distributions (red dots), and their least-squares fits (green traces) for a representative 16S fragment (box1 in A). (C) Ambiguity of peak identification as a function of the mass tolerance parameter (ppm). MS peaks were matched against the 16S theoretical digest (described in D), and the fraction of experimental peaks assignable to more than one rRNA fragment was calculated. Peak identification was carried out using m/z values for 14N-labeled fragments only (black); m/z for both 14N- and 15N-labeled fragments and assuming that fragments should elute within 0.1 min of each other (red); using 14N and 15N m/z and charge state (z) of the two species (blue). (D) Excerpt of the RNase T1 theoretical digest containing predicted 16S RNA fragments and their monoisotopic m/z values in the ‘vicinity’ of (m62A)(m62A)CCUG (gray box). Digest includes RNA species with charges 1–4, with 0–2 missed cleavages and either linear or cyclic (>p) phosphate at 3′ terminus. List is sorted by 14N m/z values. m – is a methyl group, >p – cyclic phosphate (otherwise linear), and * marks compositionally nonunique RNA fragments included as a single entry. (E) Histogram of RNA level values calculated for all 16S rRNA fragments identified in the control experiment.

Figure 3

Figure 3

Metabolic labeling approaches for rRNA modifications analysis. CD3-methionine (A) and 5,6-D-uracil (B) labeling results in characteristic mass shifts for methylated and pseudouridinylated rRNA fragments. Mass spectra for 23S: 1915-(m3Ψ)AΨ-1917 fragment are shown. 23S 15N-labeled spike was prepared by growing cells in 15N-ammonium sulfate (red); in 15N-ammonium sulfate with CD3-14N-methionine (blue) or in 15N-ammonium sulfate with 5,6-D-14N-uracil (cyan) added to the M9 medium. Δ_N_ is the number of 15N-labeled nitrogen atoms. Isotope distributions were fitted using 99.3% of 15N isotope enrichment (red); 99.3% of 15N and 98.5% of D (cyan), as was determined empirically. Using CD3-14N-methionine, some amount of the methionine amino group was scrambled leading to a reduced fraction of 15N labeling, which was adjusted to 98.0% for 15N and 99.3% for D (blue).

Figure 4

Figure 4

Inventory of 16S RNA modifications in wild-type E. coli ribosomes. Relative abundances of the 16S modifications with respect to the external standard are shown for the fractions collected across the 30S peak. RNA modification levels (_f_mod) were normalized to amounts of unmodified rRNA fragments measured in each fraction (_f_total). (A) Three distinct groups of rRNA modifications are shown in red, yellow, and cyan. Data for fragments reporting on m2G(1516) and m62A(1518), m62A(1519) abundances (Table 2) were combined. (B) Schematic of the 16S secondary structure with its structural domains colored to map groups of RNA modifications.

Figure 5

Figure 5

Inventory of 23S RNA modifications in wild-type E. coli ribosomes. Normalized RNA modification levels (_f_mod/_f_total) across the 50S and 70S peaks. Two distinct groups of 23S modifications are colored red and cyan. For residues that are physically linked due to digestion: m1G(745), Ψ(746), m5U(747); Ψ(1911), m3Ψ(1915), Ψ(1917); Cm(2498), ho5C(2501); and m2A(2503), Ψ(2504) (Table 2), inventory data were combined. RNA levels calculated using individual nucleolytic fragments corresponding to these residues were found to be within the measurement error, and their averaged values are shown.

Figure 6

Figure 6

Inventory analysis of rRNA modifications and ribosomal proteins. Relative levels of RNA modifications and ribosomal proteins for the subset of the small (A) and large (B) ribosomal subunit components are shown as a heat map. RNA and protein data were obtained from two separate experiments, ribosome sedimentation traces were aligned, and data linearly interpolated to account for the difference in a number of collected fractions. Protein levels have been previously reported. Here, values for the intermediate (S7/L5) and late (S2/L16) binding proteins were normalized to those of the primary binders (S4 or L24).

References

    1. Machnicka M. A.; Milanowska K.; Osman Oglou O.; Purta E.; Kurkowska M.; Olchowik A.; Januszewski W.; Kalinowski S.; Dunin-Horkawicz S.; Rother K. M.; Helm M.; Bujnicki J. M.; Grosjean H. Nucleic Acids Res. 2013, 41, D262–267. - PMC - PubMed
    1. Sergiev P.; Golovina A.; Prokhorova I.; Sergeeva O.; Osterman I.; Nesterchuk M.; Burakovsky D.; Bogdanov A.; Dontsova O. In Ribosomes; Rodnina M., Wintermeyer W., Green R., Eds.; Springer: Vienna: 2011, p 97–110.
    1. Higa-Nakamine S.; Suzuki T.; Uechi T.; Chakraborty A.; Nakajima Y.; Nakamura M.; Hirano N.; Kenmochi N. Nucleic Acids Res. 2012, 40, 391–398. - PMC - PubMed
    1. Cavaillé J.; Buiting K.; Kiefmann M.; Lalande M.; Brannan C. I.; Horsthemke B.; Bachellerie J. P.; Brosius J.; Hüttenhofer A. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 14311–14316. - PMC - PubMed
    1. Ruggero D.; Grisendi S.; Piazza F.; Rego E.; Mari F.; Rao P. H.; Cordon-Cardo C.; Pandolfi P. P. Science 2003, 299, 259–262. - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources