Genes encode instructions to make proteins and other molecules. To issue an instruction, a gene is first used as a template to make molecules of ribonucleic acid (called mRNAs for short) in a process called transcription. An enzyme called RNA polymerase – which comprises several protein subunits that all work together – is responsible for making the mRNA molecules. Occasionally, this enzyme makes mistakes that lead to small changes in the instruction that is produced. These mistakes are rare, but because cells make thousands of mRNAs, a single human cell can make 10-100 transcription errors per second.

It has been difficult to study how often RNA polymerase makes mistakes and what effect these mistakes have on organisms because the techniques available for research are labour-intensive and technically challenging. Here, Lucas Carey demonstrates that it is possible to use a technique called RNA sequencing to study the accuracy of RNA polymerase in human and yeast cells.

The experiments show that altering the levels of the different subunits of RNA polymerase in cells can change how many mistakes are made during transcription. This suggests that cells may be able regulate number of mistakes by controlling the production of specific subunits. Carey found that the severity of the mistakes made by RNA polymerase depends on where the mistake is in the mRNA. For example, errors in specific parts of the mRNA can alter how the whole instruction is edited later, while others might make only a tiny change to the protein encoded by the gene. Carey also found evidence that the instructions encoded by genes may have evolved in such a way to minimise the effect of any errors on their roles in cells.

RNA sequencing is less labour-intensive than other methods used to study the accuracy of RNA polymerase and is already used to address other research questions on a wide variety of different organisms. Therefore, Carey’s findings will make it easier to study what genes or environmental factors influence the number of errors made during transcription. A major challenge for the future is to find out if the mistakes made by RNA polymerase can lead to cancer and other human diseases.

The information that determines protein sequence is stored in the genome, but that information must be transcribed by RNA polymerase and translated by the ribosome before reaching its final form. DNA polymerase error rates have been well characterized in a variety of species and environmental conditions, and are low – on the order of one mutation per 10⁸–10¹⁰ bases per generation (Lynch, 2011; Lang and Murray, 2008; Zhu et al., 2014). In contrast, RNA polymerase errors are uniquely positioned to generate phenotypic diversity. Error rates are high (10^-6–10^-5) (Gout et al., 2013; Lynch, 2010; Shaw et al., 2002; de Mercoyrol et al., 1992), and each mRNA molecule is translated into 2000–4000 molecules of protein (Schwanhäusser et al., 2011; Futcher et al., 1999), resulting in the amplification of any errors. Likewise, because many RNAs are present at an average of less than one molecule per cell in microbes (Pelechano et al., 2010) and in embryonic stem cells (Islam et al., 2011), an RNA with an error may be the only RNA for that gene; all newly translated protein will contain this error. Despite the fact that transient errors can result in altered phenotypes (Gordon et al., 2013, 2015), the genetics and environmental factors that affect RNA polymerase fidelity are poorly understood. This is because current methods for measuring polymerase fidelity are technically challenging (Gout et al., 2013), require specialized organism-specific genetic constructs (Irvin et al., 2014), and can only measure error rates at specific loci (Imashimizu et al., 2013).

To overcome these obstacles I developed MORPhEUS (Measurement Of RNA Polymerase Errors Using Sequencing), which enables measurement of differential RNA polymerase fidelity using existing RNA-seq data (Figure 1). The input is a set of RNA-seq fastq files and a reference genome, and the output is the error rate at each position in the genome. I find that RNA polymerase errors result in intron retention and that cellular mRNA quality control may reduce the effective RNA polymerase error rate. Moreover, my analyses suggest that the expression level of the RPB9 Pol II subunits Rpb9 and Dst1 (TFIIS) determines RNA polymerase fidelity in vivo. Because it can be run on any existing RNA-seq data, MORPhEUS enables the exploration of a previously unexplored source of biological diversity in microbes and mammals.

Figure 1 with 2 supplements see all

Download asset Open asset

Technical errors from reverse transcription and sequencing, and biological errors from RNA polymerase look identical (single-nucleotide differences from the reference genome). Therefore, a major challenge in identifying single-nucleotide polymorphisms (SNPs) and in measuring changes in polymerase fidelity is the reduction of technical errors (Kleinman and Majewski, 2012; Pickrell et al., 2012; Li et al., 2011) (Figure 1). First, I map full-length (untrimmed) reads to the genome and discard reads with indels, with more than two mismatches, that map to multiple locations in the genome, and that do not map end to end along the full length of the read. Next, I trim the ends of the mapped reads, as alignments are of lower quality along the ends, and the mismatch rate is higher, especially at splice junctions. I also discard any cycles within the run with abnormally high error rates, and bases with low Illumina quality scores (Figure 1—figure supplement 1). Finally, using the remaining bases, I count the number of matches and mismatches to the reference genome at each position in the genome. I discard positions with identical mismatches that are present more than once, as these are likely due to subclonal DNA polymorphisms or sequences that Illumina miscalls in a systematic manner (Meacham et al., 2011) (Figure 1—figure supplement 2). The result is a set of mismatches, many of which are technical errors and some of which are RNA polymerase errors. In order to determine if RNA-seq mismatches are due to RNA polymerase errors, it is necessary to identify sequence locations in which RNA polymerase errors are expected to have a measurable effect, or situations in which RNA polymerase fidelity is expected to vary.

I reasoned that RNA polymerase errors that alter positions necessary for splicing should result in intron retention, while sequencing errors should not affect the final structure of the mRNA (Figure 2a). However, mutations in the donor and acceptor splice sites also result in decreased expression (Jung et al., 2015), and therefore are difficult to measure using RNA-seq. Therefore, I used chromatin-associated and nuclear RNA from Hela and Huh7 cells (Dhir et al., 2015), and extracted all reads that span an exon–intron junction for introns with canonical GT and AG splice sites, and measured the RNA-seq mismatch rate at each position. I find that errors at the G and U in the 5’ donor site and at the A in the acceptor site are significantly enriched relative to errors at other positions (Figure 2b), and to errors in exonic trinucleotides at splicing motifs in the human genome (Figure 2—figure supplement 1) suggesting that RNA polymerase mismatches can result in changes in transcript isoforms. The ability of RNA polymerase errors to significantly affect splicing has been proposed (Fox-Walsh and Hertel, 2009) but never previously measured.

Figure 2 with 2 supplements see all

Download asset Open asset

RPB9 is known to be involved in RNA polymerase fidelity in vitro and in vivo (Irvin et al., 2014; Knippa and Peterson, 2013). Therefore, I reasoned that cell lines expressing low levels of RPB9 would have higher RNA polymerase error rates. Consistent with this, I find that RPB9 expression varies eightfold across the ENCODE cell lines, and this expression variation is correlated with the RNA-seq error rate (Figure 2c, Figure 2—figure supplement 2). This suggests that low RPB9 expression may cause decreased polymerase fidelity in vivo.

In addition, export of mRNAs from the nucleus involves a quality-control mechanism that checks if mRNAs are fully spliced and have properly formed 5’ and 3’ ends (Lykke-Andersen, 2001). I hypothesized that mRNA export may involve a quality control that removes mRNAs with errors. I used the ENCODE dataset in which nuclear and cytoplasmic poly-A + mRNAs were sequenced; thus I can compare nuclear and cytoplasmic fractions from the same cell line grown in the same conditions and processed in the same manner. I find that the nuclear fraction has a higher RNA polymerase error rate than does the cytoplasmic fraction (Figure 2c,d), suggesting that either that nuclear RNA-seq has a higher technical error rate or that the cell has mechanisms for reducing the effective polymerase error rate by preventing the export of mRNAs that contain errors.

Rpb9 and Dst1 are known to be involved in RNA polymerase fidelity in vitro, yet there is conflicting evidence as to the role of Dst1 in vivo(Shaw et al., 2002; Irvin et al., 2014; Knippa and Peterson, 2013; Nesser et al., 2006; Walmacq et al., 2009; Kireeva et al., 2008). Part of these conflicts may result from the fact that the only available assays for RNA polymerase fidelity are special reporter strains that rely on DNA sequences known to increase the frequency of RNA polymerase errors. While I found that RPB9 expression correlates with RNA-seq error rates in mammalian cells, correlation is not causation. Furthermore, differences in RNA levels do not necessitate differences in stoichiometry among the subunits in active Pol II complexes. In order to determine if differential expression of RPB9 or DST1 are causative for differences in RNA polymerase fidelity in vivo, I constructed two yeast strains in which I can alter the expression of either RPB9 or DST1 using β-estradiol and a synthetic transcription factor that has no effect on growth rate or the expression of any other genes (Mcisaac et al., 2014, 2013). I grew these two strains (Z₃EV_pr-RPB9 and Z₃EV_pr-DST1) in different concentrations of β-estradiol and performed RNA-seq. I find that cells expressing low levels of RPB9 have high RNA polymerase error rates (Figure 3a). Likewise, cells with low DST1 have high error rates (Figure 3a). The increase in errors rate is not a property of cells defective for transcription elongation (Figure 3—figure supplement 1). The increase in error rates due to mutations in Rpb9 and Dst1 have not been robustly measured, however, there are some rough numbers. Here, the measured increase in error rate is 13%, while the measured effect of Rpb9 deletion in vitro is fivefold (Walmacq et al., 2009) and in vivo following reverse transcription is 30% (Nesser et al., 2006). If 2% of the observed mismatches are due to RNA polymerase errors, a fivefold increase in polymerase error rate results in a 10% increase in measured mismatch frequency; this is consistent with RNA polymerase fidelity of 10^-6–10^-5 and overall RNA-seq error rates of 10^-4. Note that in our assay cells still express low levels or RPB9, and we therefore expect the increase in error rate to be lower, suggesting that RNA polymerase errors constitute 5–10% of the measured mismatches. Our ability to genetically control the expression of DST1 and RPB9, and measure changes in RNA-seq error rates is consistent with MORPhEUS measuring RNA polymerase fidelity. In addition, we observe more single-nucleotide insertions in the RNA-seq data from the high error rate samples, suggesting that depletion of RPB9 and DST1 results in increased insertions in transcripts, but not increased deletions (Figure 3—figure supplement 2). Finally, genetic reduction in RNA polymerase fidelity results in increased intron retention, consistent with RNA polymerase errors causing reduced splicing efficiency (Figure 3b).

Figure 3 with 2 supplements see all

Download asset Open asset

A unique advantage of MORPhEUS is that it measures thousands of RNA polymerase errors across the entire transcriptome in a single experiment, and thus enables he complete characterization of the mutation spectrum and biases of RNA polymerase. I asked how altered RPB9 and DST1 expression levels affect each type of single-nucleotide change. I find that, with decreasing polymerase fidelity, transitions increase more than transversions, and that C→U errors are the most common (Figure 3c). This result, along with other sequencing based results (Gout et al., 2013), have shown that DNA and RNA polymerase have broadly similar error profiles (Zhu et al., 2014); it will be interesting to see if all polymerases share the same mutation spectra, and if this is due to deamination of the template base, or is a structural property of the polymerase itself. Interestingly, I find that coding sequences have evolved so that errors are less likely to produce in-frame stop codons than out-of-frame stop codons, suggesting that natural selection may act to minimize the effect of polymerase errors (Figure 4).

Figure 4

Download asset Open asset

Here I have presented proof that relative changes in RNA polymerase error rates can be measured using standard Illumina RNA-seq data. Consistent with previous work in vivo and in vitro, I find that depletion of RPB9 or Dst1 results in higher RNA polymerase error rates. Furthermore, I find that expression of RPB9 negatively correlates with RNA-seq error rates in human cell lines, suggesting that differential expression of RPB9 may regulate RNA polymerase fidelity in vivo in humans. In addition, consistent with the errors detected by MORPhEUS being due to RNA polymerase and not technical errors, in reads spanning an exon–intron junction, the measured error rate is higher at the 5’ donor splice site, suggesting that RNA polymerase errors result in intron retention. Because it can be run on existing RNA-seq data, I expect MORPhEUS to enable many future discoveries regarding both the molecular determinants of RNA polymerase error rates and the relationship between RNA polymerase fidelity and phenotype.

1. 1. Carey LB

   (2015) PRJNA289596

   Publicly available at the NCBI BioProject database (Accession no: PRJNA289596).

   http://www.ncbi.nlm.nih.gov/bioproject/289596

1. 1. The ENCODE Consortium

   (2008) Home sapiens (human)

   Publicly available at the NCBI BioProject database (Accession no: PRJNA30709).

   http://www.ncbi.nlm.nih.gov/bioproject/30709

1. 1. Dhir A
   2. Dhir S
   3. Proudfoot NJ
   4. Jopling CL

   (2015) Microprocessor mediates transcriptional termination of long noncoding RNA transcripts hosting microRNAs

   Nature Structural & Molecular Biology 22:319–327.

   https://doi.org/10.1038/nsmb.2982

   • Google Scholar
2. 1. ENCODE Project Consortium

   (2012) An integrated encyclopedia of DNA elements in the human genome

   Nature 489:57–74.

   https://doi.org/10.1038/nature11247

   • Google Scholar
3. 1. Fox-Walsh KL
   2. Hertel KJ

   (2009) Splice-site pairing is an intrinsically high fidelity process

   Proceedings of the National Academy of Sciences of the United States of America 106:1766–1771.

   https://doi.org/10.1073/pnas.0813128106

   • Google Scholar
4. 1. Futcher B
   2. Latter GI
   3. Monardo P
   4. Mclaughlin CS
   5. Garrels JI

   (1999)

   A sampling of the yeast proteome mol

   Cellular Biology 19:7357–7368.

   • Google Scholar
5. 1. Gordon AJE
   2. Satory D
   3. Halliday JA
   4. Herman C
   5. Casadesús J

   (2013) Heritable change caused by transient transcription errors

   PLoS Genetics 9:e1003595.

   https://doi.org/10.1371/journal.pgen.1003595

   • Google Scholar
6. 1. Gordon AJE
   2. Satory D
   3. Halliday JA
   4. Herman C
   5. Herman
   6. Lost In Transcription C

   (2015) Lost in transcription: transient errors in information transfer

   Current Opinion in Microbiology 24:80–87.

   https://doi.org/10.1016/j.mib.2015.01.010

   • Google Scholar
7. 1. Gout J-F
   2. Thomas WK
   3. Smith Z
   4. Okamoto K
   5. Lynch M

   (2013) Large-scale detection of in vivo transcription errors

   Proceedings of the National Academy of Sciences of the United States of America 110:18584–18589.

   https://doi.org/10.1073/pnas.1309843110

   • Google Scholar
8. 1. Hereford LM
   2. Rosbash M

   (1977) Number and distribution of polyadenylated RNA sequences in yeast

   Cell 10:453–462.

   https://doi.org/10.1016/0092-8674(77)90032-0

   • Google Scholar
9. 1. Imashimizu M
   2. Oshima T
   3. Lubkowska L
   4. Kashlev M

   (2013) Direct assessment of transcription fidelity by high-resolution RNA sequencing

   Nucleic Acids Research 41:9090–9104.

   https://doi.org/10.1093/nar/gkt698

   • Google Scholar
10. 1. Irvin JD
    2. Kireeva ML
    3. Gotte DR
    4. Shafer BK
    5. Huang I
    6. Kashlev M
    7. Strathern JN
    8. Landick R

    (2014) A genetic assay for transcription errors reveals multilayer control of RNA polymerase II fidelity

    PLoS Genetics 10:e1004532.

    https://doi.org/10.1371/journal.pgen.1004532

    • Google Scholar
11. 1. Islam S
    2. Kjallquist U
    3. Moliner A
    4. Zajac P
    5. Fan J-B
    6. Lonnerberg P
    7. Linnarsson S

    (2011) Characterization of the single-cell transcriptional landscape by highly multiplex RNA-seq

    Genome Research 21:1160–1167.

    https://doi.org/10.1101/gr.110882.110

    • Google Scholar
12. 1. Jung H
    2. Lee D
    3. Lee J
    4. Park D
    5. Kim YJ
    6. Park W-Y
    7. Hong D
    8. Park PJ
    9. Lee E

    (2015) Intron retention is a widespread mechanism of tumor-suppressor inactivation

    Nature Genetics 47:1242–1248.

    https://doi.org/10.1038/ng.3414

    • Google Scholar
13. 1. Karolchik D

    (2004) The UCSC table browser data retrieval tool

    Nucleic Acids Research 32:493D–496.

    https://doi.org/10.1093/nar/gkh103

    • Google Scholar
14. 1. Kireeva ML
    2. Nedialkov YA
    3. Cremona GH
    4. Purtov YA
    5. Lubkowska L
    6. Malagon F
    7. Burton ZF
    8. Strathern JN
    9. Kashlev M

    (2008) Transient reversal of RNA polymerase II active site closing controls fidelity of transcription elongation

    Molecular Cell 30:557–566.

    https://doi.org/10.1016/j.molcel.2008.04.017

    • Google Scholar
15. 1. Kleinman CL
    2. Majewski J

    (2012) Comment on "widespread RNA and DNA sequence differences in the human transcriptome"

    Science 335:1302.

    https://doi.org/10.1126/science.1209658

    • Google Scholar
16. 1. Knippa K
    2. Peterson DO

    (2013) Fidelity of RNA polymerase II transcription: role of Rbp9 in error detection and proofreading

    Biochemistry 52:7807–7817.

    https://doi.org/10.1021/bi4009566

    • Google Scholar
17. 1. Lang GI
    2. Murray AW

    (2008) Estimating the per-base-pair mutation rate in the yeast saccharomyces cerevisiae

    Genetics 178:67–82.

    https://doi.org/10.1534/genetics.107.071506

    • Google Scholar
18. 1. Li H
    2. Handsaker B
    3. Wysoker A
    4. Fennell T
    5. Ruan J
    6. Homer N
    7. Marth G
    8. Abecasis G
    9. Durbin R
    10. 1000 Genome Project Data Processing Subgroup
    11. 1000 Genome Project Data Processing Subgroup

    (2009) The sequence Alignment/Map format and SAMtools

    Bioinformatics 25:2078–2079.

    https://doi.org/10.1093/bioinformatics/btp352

    • Google Scholar
19. 1. Li H
    2. Durbin R

    (2009) Fast and accurate short read alignment with burrows-wheeler transform

    Bioinformatics 25:1754–1760.

    https://doi.org/10.1093/bioinformatics/btp324

    • Google Scholar
20. 1. Li M
    2. Wang IX
    3. Li Y
    4. Bruzel A
    5. Richards AL
    6. Toung JM
    7. Cheung VG

    (2011) Widespread RNA and DNA sequence differences in the human transcriptome

    Science 333:53–58.

    https://doi.org/10.1126/science.1207018

    • Google Scholar
21. 1. Lykke-Andersen J

    (2001) MRNA quality control: marking the message for life or death

    Current Biology 11:R88–R91.

    https://doi.org/10.1016/S0960-9822(01)00036-7

    • Google Scholar
22. 1. Lynch M

    (2010) Evolution of the mutation rate

    Trends in Genetics 26:345–352.

    https://doi.org/10.1016/j.tig.2010.05.003

    • Google Scholar
23. 1. Lynch M

    (2011) The lower bound to the evolution of mutation rates

    Genome Biology and Evolution 3:1107–1118.

    https://doi.org/10.1093/gbe/evr066

    • Google Scholar
24. 1. McIsaac RS
    2. Oakes BL
    3. Botstein D
    4. Noyes MB

    (2013) Rapid synthesis and screening of chemically activated transcription factors with GFP-based reporters

    Journal of Visualized Experiments e51153.

    https://doi.org/10.3791/51153

    • Google Scholar
25. 1. McIsaac RS
    2. Oakes BL
    3. Wang X
    4. Dummit KA
    5. Botstein D
    6. Noyes MB

    (2013) Synthetic gene expression perturbation systems with rapid, tunable, single-gene specificity in yeast

    Nucleic Acids Research 41:e57.

    https://doi.org/10.1093/nar/gks1313

    • Google Scholar
26. 1. McIsaac RS
    2. Gibney PA
    3. Chandran SS
    4. Benjamin KR
    5. Botstein D

    (2014) Synthetic biology tools for programming gene expression without nutritional perturbations in saccharomyces cerevisiae

    Nucleic Acids Research 42:e48.

    https://doi.org/10.1093/nar/gkt1402

    • Google Scholar
27. 1. Meacham F
    2. Boffelli D
    3. Dhahbi J
    4. Martin DIK
    5. Singer M
    6. Pachter L

    (2011) Identification and correction of systematic error in high-throughput sequence data

    BMC Bioinformatics 12:451.

    https://doi.org/10.1186/1471-2105-12-451

    • Google Scholar
28. 1. de Mercoyrol L
    2. Corda Y
    3. Job C
    4. Job D

    (1992) Accuracy of wheat-germ RNA polymerase II. general enzymatic properties and effect of template conformational transition from right-handed b-DNA to left-handed z-DNA

    European Journal of Biochemistry 206:49–58.

    https://doi.org/10.1111/j.1432-1033.1992.tb16900.x

    • Google Scholar
29. 1. Nesser NK
    2. Peterson DO
    3. Hawley DK

    (2006) RNA polymerase II subunit Rpb9 is important for transcriptional fidelity in vivo

    Proceedings of the National Academy of Sciences of the United States of America 103:3268–3273.

    https://doi.org/10.1073/pnas.0511330103

    • Google Scholar
30. 1. Pelechano V
    2. Chávez S
    3. Pérez-Ortín JE
    4. Santos J

    (2010) A complete set of nascent transcription rates for yeast genes

    PLoS One 5:e15442.

    https://doi.org/10.1371/journal.pone.0015442

    • Google Scholar
31. 1. Pickrell JK
    2. Gilad Y
    3. Pritchard JK

    (2012) Comment on "widespread RNA and DNA sequence differences in the human transcriptome"

    Science 335:1302.

    https://doi.org/10.1126/science.1210484

    • Google Scholar
32. 1. Quinlan AR
    2. Hall IM

    (2010) BEDTools: a flexible suite of utilities for comparing genomic features

    Bioinformatics 26:841–842.

    https://doi.org/10.1093/bioinformatics/btq033

    • Google Scholar
33. 1. Schwanhäusser B
    2. Busse D
    3. Li N
    4. Dittmar G
    5. Schuchhardt J
    6. Wolf J
    7. Chen W
    8. Selbach M

    (2011) Global quantification of mammalian gene expression control

    Nature 473:337–342.

    https://doi.org/10.1038/nature10098

    • Google Scholar
34. 1. Shaw RJ
    2. Bonawitz ND
    3. Reines D

    (2002) Use of an in vivo reporter assay to test for transcriptional and translational fidelity in yeast

    Journal of Biological Chemistry 277:24420–24426.

    https://doi.org/10.1074/jbc.M202059200

    • Google Scholar
35. 1. Tilgner H
    2. Knowles DG
    3. Johnson R
    4. Davis CA
    5. Chakrabortty S
    6. Djebali S
    7. Curado J
    8. Snyder M
    9. Gingeras TR
    10. Guigo R

    (2012) Deep sequencing of subcellular RNA fractions shows splicing to be predominantly co-transcriptional in the human genome but inefficient for lncRNAs

    Genome Research 22:1616–1625.

    https://doi.org/10.1101/gr.134445.111

    • Google Scholar
36. 1. Walmacq C
    2. Kireeva ML
    3. Irvin J
    4. Nedialkov Y
    5. Lubkowska L
    6. Malagon F
    7. Strathern JN
    8. Kashlev M

    (2009) Rpb9 subunit controls transcription fidelity by delaying NTP sequestration in RNA polymerase II

    Journal of Biological Chemistry 284:19601–19612.

    https://doi.org/10.1074/jbc.M109.006908

    • Google Scholar
37. 1. Walmacq C
    2. Kireeva ML
    3. Irvin J
    4. Nedialkov Y
    5. Lubkowska L
    6. Malagon F
    7. Strathern JN
    8. Kashlev M

    (2009) Rpb9 subunit controls transcription fidelity by delaying NTP sequestration in RNA polymerase II

    Journal of Biological Chemistry 284:19601–19612.

    https://doi.org/10.1074/jbc.M109.006908

    • Google Scholar
38. 1. Zhu YO
    2. Siegal ML
    3. Hall DW
    4. Petrov DA

    (2014) Precise estimates of mutation rate and spectrum in yeast

    Proceedings of the National Academy of Sciences of the United States of America 111:E2310.

    https://doi.org/10.1073/pnas.1323011111

    • Google Scholar

Author details

1. Lucas B Carey

   Department of Experimental and Health Sciences, Universitat Pompeu Fabra, Barcelona, Spain

   Contribution

   LBC, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article, Contributed unpublished essential data or reagents

   For correspondence

   lucas.carey@upf.edu

   Competing interests

   The author declares that no competing interests exist.

• 10,492

  views

• 502

  downloads

• 42

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.