Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2014 Nov 6;515(7525):143-6.
doi: 10.1038/nature13802. Epub 2014 Sep 5.

Pseudouridine profiling reveals regulated mRNA pseudouridylation in yeast and human cells

Thomas M Carlile  1 , Maria F Rojas-Duran  1 , Boris Zinshteyn  1 , Hakyung Shin  1 , Kristen M Bartoli  1 , Wendy V Gilbert  1
Affiliations

Pseudouridine profiling reveals regulated mRNA pseudouridylation in yeast and human cells

Thomas M Carlile et al. Nature. .

Abstract

Post-transcriptional modification of RNA nucleosides occurs in all living organisms. Pseudouridine, the most abundant modified nucleoside in non-coding RNAs, enhances the function of transfer RNA and ribosomal RNA by stabilizing the RNA structure. Messenger RNAs were not known to contain pseudouridine, but artificial pseudouridylation dramatically affects mRNA function--it changes the genetic code by facilitating non-canonical base pairing in the ribosome decoding centre. However, without evidence of naturally occurring mRNA pseudouridylation, its physiological relevance was unclear. Here we present a comprehensive analysis of pseudouridylation in Saccharomyces cerevisiae and human RNAs using Pseudo-seq, a genome-wide, single-nucleotide-resolution method for pseudouridine identification. Pseudo-seq accurately identifies known modification sites as well as many novel sites in non-coding RNAs, and reveals hundreds of pseudouridylated sites in mRNAs. Genetic analysis allowed us to assign most of the new modification sites to one of seven conserved pseudouridine synthases, Pus1-4, 6, 7 and 9. Notably, the majority of pseudouridines in mRNA are regulated in response to environmental signals, such as nutrient deprivation in yeast and serum starvation in human cells. These results suggest a mechanism for the rapid and regulated rewiring of the genetic code through inducible mRNA modifications. Our findings reveal unanticipated roles for pseudouridylation and provide a resource for identifying the targets of pseudouridine synthases implicated in human disease.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing financial interests.

Figures

Extended Data Figure 1
Extended Data Figure 1. Detection of specific snoRNA target sites by Pseudo-seq
Pseudo-seq was performed on wild-type (n=4), snr37Δ (n=2), snr81Δ (n=2), snr43Δ (n=2), and snr49Δ (n=2) yeast strains. Cultures were harvested at high density (a,b) or log phase (c,d). Ψs dependent on the deleted snoRNA are indicated in red. CMC-dependent peaks of reads are indicated with dashed red lines. Traces are representative of indicated number of biological replicates. a) Pseudo-seq reads in RDN25-1 (chrXII:452221–452270) showing SNR37-dependence of 25S-Ψ2944. b) Pseudo-seq reads in RDN25-1 (chrXII:454111–454160, left), and U2 snRNA (LSR1, chrII:681791–681840, right) showing SNR81-dependence of 25S-Ψ1052 and U2-Ψ42. c) Pseudo-seq reads in RDN58-1 (chrXII:455466–455515) showing SNR43-dependence of 5.8S-Ψ73. SNR43-dependent 25S-Ψ960 was not consistently detected in wild type due to an overlapping CMC-independent RT stop. d) Pseudo-seq reads in RDN18-1 (chrXII:457361–457610) showing SNR49-dependence of 18S-Ψ302, 18S-Ψ211, and 18S-Ψ120. 25S-Ψ990 was also detected as SNR49-dependent (data not shown).
Extended Data Figure 2
Extended Data Figure 2. Technical variations of Pseudo-seq give similar results
a-d) MetaPsi plots (left), and ROC curves (right) for various library prep conditions n=1 for each condition. CMC-treated samples (solid), and mock-treated samples (dashed) are indicated. a) Comparison of AMV-RT (orange), and SuperScript® III (blue) (0.2M CMC; 115–130 nt, 100–115 nt fragments respectively). b) Comparison of 115–130 nt (orange), and 130–145 nt (blue) RNA fragment sizes (AMV-RT; 0.2M CMC). c) Comparison of 0.2M CMC (blue), and 0.4M CMC (orange) (AMV-RT; 115–130 nt RNA). d) Comparison of shorter (orange) and longer (blue) truncated RT fragment sizes (AMV-RT; 115–130 nt RNA; 0.2M CMC).
Extended Data Figure 3
Extended Data Figure 3. Identification of pseudouridines in lowly expressed genes using multiple replicates
a) Growth curves for wild-type yeast were grown in YPD. An OD600 of 12 is indicated by the horizontal dotted line. b,c) Pseudo-seq was performed on polyA+ RNA isolated from high density wild type yeast strains. CMC-dependent peaks of reads are indicated with dashed red lines. b) Pseudo-seq reads from n=4 biological replicates in a) CDC39 (chrIII:286226–286445, 12.3 avg rpkms), and c) IQG1 (chrXVI:90655–90955, 12.4 avg rpkms) showing CDC39-Ψ6223 and IQG1-Ψ4367, respectively.
Extended Data Figure 4
Extended Data Figure 4. Codons affected by mRNA pseudouridylation
Pseudouridylation of mRNA preferentially affects GUA codons. Numbers of pseudouridines observed at the first (dark blue), second (blue), and third positions (light blue) of each codon are indicated.
Extended Data Figure 5
Extended Data Figure 5. Expression levels minimally affect identification of yeast mRNAs displaying regulated pseudouridylation
A plot of log-transformed average rpkms in high density versus log phase yeast for all coding genes with a Ψ identified by Pseudo-seq n=4 biological replicates for each condition. All genes (gray), genes with a high density induced Ψ (blue), and genes with a log phase induced Ψ (red) are indicated.
Extended Data Figure 6
Extended Data Figure 6. Inducible Pseudouridylation of ncRNAs
a,b) Pseudo-seq was performed on wild-type (n=4), snr81Δ (n=2), pus1Δ (n=2), and pus7Δ (n=2) yeast strains grown to high density. CMC dependent peaks of reads are indicated with a dashed red line. Traces are representative indicated number of biological replicates. a) Pseudo-seq reads in U2 snRNA (LSR1; chrII:681751–681790, left; chrII:681769–681818, right) showing SNR81-dependence of U2-Ψ93, and PUS1-dependence of U2-Ψ56. Both are dependent on growth to high density. b) Pseudo-seq reads in U3a snoRNA (SNR17A, chrXV:780461–780560) showing snR17A-Ψ369 (PUS7-dependent), snR17A-Ψ380, snR17A-Ψ391, and snR17A-Ψ425 (PUS1-dependent). c) Summaries of the numbers of Ψs called in ncRNAs by Pseudo-seq. Indicated are constitutive Ψs (top), and inducible Ψs (bottom).
Extended Data Figure 7
Extended Data Figure 7. Analysis of potential snoRNA targets
a–d) Pseudo-seq was performed on wild-type yeast in log phase, or grown to high density. Reads from n=4 biological replicate libraries for each condition were pooled. b-d) Indicated are the predicted snoRNA target site (black, dashed), and the expected peak of CMC-dependent reads (black, dotted). a,b) Results of analysis on sets of random Us. a) A histogram of the differences (+CMC – −CMC) in mean normalized reads at the +1 peak position for 10,000 randomizations for high density (orange) and log phase (blue). The normalized read values for the computationally predicted Ψs in exponential and high density samples are indicated by arrows. b) An averaged metaPsi plot for all randomizations. c,d) +CMC (c), and −CMC (d) MetaPsi plots for computationally predicted Ψs separated by base pairing. Sites with 8 or more (red), 9 or more (blue), and 10 or more (orange) base pairs are indicated. Data for high density (left), and log phase (right) are indicated. e) Pseudo-seq reads for computationally predicted Ψs, CAT2 (chrXII:193995–19450, left), and AIM6 (chrIV:31135–31550 right). Traces are representative of at least six replicates.
Extended Data Figure 8
Extended Data Figure 8. Mechanisms of Pus-dependent pseudouridylation
a,b) Summaries of the PUS-dependence of called Ψs using higher stringency cutoffs (10/14 libraries) (a), and lower stringency cutoffs (9/14 libraries) (b). c,f) CMC dependent peaks of reads are indicated with dashed red lines. Traces are representative of n=4 (wild type), and n=2 (pusΔ) biological replicates. Pseudo-seq reads for RPL14A (a, chrXI:431901–432200) and PDI1 (d, chrIII:49401–48760) showing PUS1- and PUS7-dependency respectively. Both are dependent on growth to high density. d,e,g,h) WebLogo 3.3 was used to generate motifs for PUS1 (d), PUS2 (e), PUS7 (g), and PUS4 (h).
Extended Data Figure 9
Extended Data Figure 9. Positive controls for human RNA Pseudo-seq
a) Pseudo-seq reads for RDN28S5 (1516–1765) containing five known Ψs (28S-Ψ1536, 28S-Ψ1582, 28S-Ψ1677, 28S-Ψ1683, and 28S-Ψ1744). CMC-dependent peaks of reads are indicated with dashed red lines. Traces are representative of n=5 biological replicates. b) A metaPsi plot of mean normalized reads (left axis) for +CMC libraries (orange), and −CMC libraries (blue). The number of Ψs at each position in the metaPsi window is indicated (black, right axis). c) A ROC curve of the Pseudo-seq signal for all known Ψs in rRNA and snRNA.
Extended Data Figure 10
Extended Data Figure 10. New pseudouridines in human RNAs
a) A Venn diagram showing the overlap of mRNA pseudouridiylation events between plus serum and serum starved HeLa cells. b) A plot of log-transformed average rpkms in serum-starved versus serum-fed HeLa for all coding genes with a Ψ identified by Pseudo-seq. All genes (gray), genes with a Ψ induced in plus serum cells (blue), and genes with a Ψ induced in serum-starved cells (red) are indicated. c) Pseudo-seq reads for RDN18S5 (184–411) (top, left), RDN18S5 (1015–1210) (top, right), RDN28S5 (2713–3108) (bottom, left), and RDN28S5 (4461–4618) (bottom, right). CMC-dependent peaks of reads are indicated with dashed red lines, and highlighted Ψs are indicated by red boxes. Traces are representative of n=4 biological replicates.
Figure 1
Figure 1. Genome-wide pseudouridine sequencing with single nucleotide resolution
a) A schematic of Pseudo-seq library preparation. b) A genome browser view of Pseudo-seq reads mapping to a 200-nt region of RDN25-1 (chrXII:452168–452367) containing six known Ψs, generated from pooled reads for n =12 technical replicates from wild-type log phase yeast cultures. Peaks of Ψ-dependent reads are indicated with dashed red lines. c) A metaPsi plot of mean normalized reads (left axis) for +CMC (orange) and −CMC (blue) libraries. The number of Ψs at each position in the metaPsi window is indicated (black, right axis). CMC-dependent RT stops are found 1 nt 3′ of known Ψs. d) A ROC curve of the Pseudo-seq signal for all known Ψs in rRNA and U2 snRNA.
Figure 2
Figure 2. Yeast mRNAs and ncRNAs are inducibly pseudouridylated
a,c) CMC-dependent peaks of reads are indicated with a dashed red line. The median Pseudo-seq peak heights in each condition are given ±SD, negative peak values occur when the reads in the −CMC library exceed those in the +CMC library. Traces are representative of four wild-type biological replicates. a) Pseudo-seq reads in RPS28B (chrXII:673163–673336), MRPS12 (chrXIV:694489–694736), and CDC33 (chrXV:50560–60875). b) Summary of locations of Ψs within mRNA features. c) Pseudo-seq reads in U5 snRNA (snR7-L, chrVII:939458–939671), RNase MRP RNA (NME1, chrXIV:585585–585925), and an H/ACA snoRNA (snR37, chrX:228090–228475). d) Summary of novel Ψs identified in ncRNA.
Figure 3
Figure 3. Mechanisms of mRNA pseudouridylation
a) MetaPsi plot of mean normalized reads for computationally predicted snoRNA-dependent targets in mRNA from high density cultures (orange), log phase cultures (blue), +CMC (solid), and −CMC (dashed). Indicated are the predicted snoRNA target site (black, dashed), and the expected peak of CMC-dependent reads (black, dotted). Reads were pooled from four wild-type biological replicate libraries. b) Summary of Ψs identified by Pseudo-seq as PUS-dependent. The few PUS6- and PUS9-dependent Ψs are not shown. The locations of Ψs within mRNAs and the distribution of Ψs among ncRNA types are indicated. c,d,f,g) Sequence motifs surrounding PUS1- (c), PUS2- (d), PUS7- (f), and PUS4-dependent (g) Ψs in mRNAs, generated by WebLogo 3.3. d) PUS7-dependent Pseudo-seq reads in MRPL36 (chrII:484301–484400).
Figure 4
Figure 4. Regulated pseudouridylation of human RNAs
Pseudo-seq was performed on HeLa cells grown in the presence or absence of serum for 24 hr. a,b) CMC-dependent peaks of reads are indicated with a dashed red line. The median Pseudo-seq peak heights in each condition are given ±SD. Traces are representative of n=4 (−serum), and n=5 (+serum) biological replicates. Genome browser views represent spliced transcripts. a) Pseudo-seq reads from RPL19 (12–460), and ATP5E (154–437). b) Summary of locations of Ψs within mRNA features. c) Pseudo-seq reads from MALAT1 (5081–5636) and RN7SK (142–307). d) Summary of novel Ψs identified in ncRNA.
See this image and copyright information in PMC

Comment in

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Davis FF, Allen FW. Ribonucleic acids from yeast which contain a fifth nucleotide. J Biol Chem. 1957;227:907–915. - PubMed
    1. Arnez JG, Steitz TA. Crystal structure of unmodified tRNA(Gln) complexed with glutaminyl-tRNA synthetase and ATP suggests a possible role for pseudo-uridines in stabilization of RNA structure. Biochemistry. 1994;33:7560–7567. - PubMed
    1. Charette M, Gray MW. Pseudouridine in RNA: what, where, how, and why. IUBMB Life. 2000;49:341–351. - PubMed
    1. Davis DR, Poulter CD. 1H-15N NMR studies of Escherichia coli tRNA(Phe) from hisT mutants: a structural role for pseudouridine. Biochemistry. 1991;30:4223–4231. - PubMed
    1. Davis DR, Veltri CA, Nielsen L. An RNA model system for investigation of pseudouridine stabilization of the codon-anticodon interaction in tRNALys, tRNAHis and tRNATyr. J Biomol Struct Dyn. 1998;15:1121–1132. - PubMed

Publication types

Associated data