N1-methyl-pseudouridine is incorporated with higher fidelity than pseudouridine in synthetic RNAs

Figures

• 

  Figure 1. Uridine, ψ, and m1ψ are incorporated by T7 RNA polymerase with varying error rates during in vitro transcription. The RNA templates synthesized with T7 RNA polymerase were converted into cDNA with ProtoScript II reverse transcriptase followed by library preparation and sequencing. First-strand errors represented here stem from the combination of RNA polymerase and reverse transcriptase errors. (a) Combined first strand error rates are an average of four different RNA sequences (RNA1 and RNA2 are artificial RNA sequences that have been permutated to include every four-base combination; CLuc and BNT162b2 are functional mRNAs), and unpaired t-tests were used to calculate p-values. (b) Distribution of substitution, deletion, and insertion error percentages from four different RNA sequences. (c) Base substitution error profile observed for unmodified and modified RNA sequences are represented. Colors indicate specific substitution errors observed. Polymerase substitution errors are represented as the equivalent T7 RNA polymerase substitution (top substitution) or ProtoScript II reverse transcriptase substitution (bottom substitution).

• 

  Figure 2. rA → rU/dT → dA substitution errors observed in ψ-containing RNAs have sequence context preference and occur throughout the transcript body when reactions are performed with T7 RNA polymerase. The sequence context preference and distribution of the rA → rU/dT → dA substitution errors observed in CLuc and BNT162b2 mRNAs are represented. (a) Logo of sequence surrounding the rA → rU/dT → dA substitution sites. Seven nucleotides upstream and downstream of the substitution site were examined and for simplicity the immediate sequence upstream and downstream of the substitution site is represented. Sequence logos were built using WebLogo software41 and the numbers of sites plotted are summarized in Supplementary Table 15. (b) Substitution error distribution along the length of the transcript.

• 

  Figure 3. Uridine, ψ, and m1ψ are incorporated by SP6 RNA polymerase with varying error rates during in vitro transcription. The RNA templates synthesized with SP6 RNA polymerase were converted into cDNA with ProtoScript II reverse transcriptase followed by library preparation and sequencing. First-strand errors represented here stem from the combination of RNA polymerase and reverse transcriptase errors. (a) Combined first strand error rates are an average of three different RNA sequences (RNA1 and RNA2 are artificial RNA sequences that have been permutated to include every four-base combination; CLuc is a functional mRNA) and unpaired t-tests were used to calculate p-values. (b) Distribution of substitution, deletion, and insertion error percentages from three different RNA sequences (RNA1, RNA2, and CLuc). (c) Base substitution error profile observed for unmodified and modified RNA sequences are represented. Colors indicate specific substitution errors observed. Polymerase substitution errors are represented as the equivalent SP6 RNA polymerase substitution (top substitution) or ProtoScript II reverse transcriptase substitution (bottom substitution).

• 

  Figure 4. In vitro transcription error profile of T7 RNA polymerase can be modulated by altering ribonucleotide composition in the reaction. Uridine-depleted artificial RNA sequences and uridine-depleted CLuc mRNA were transcribed with T7 RNA polymerase and reactions were performed with either equal molar rNTPs or rNTPs proportional to the template sequence composition. (a) Combined first-strand errors are an average of four uridine depleted RNA sequences under the two different rNTP conditions. The three synthetic RNAs have uridine content of 5.5%, 6.6% and 12.3%; CLuc mRNA has a uridine content of 13.5%. Unpaired t-tests were used to calculate p-values. (b) Bioanalyzer traces of synthetic RNA with 5.5% uridine content demonstrating synthesis of full-length IVT products under both reaction conditions. (c) Base substitution error profile observed for unmodified and modified uridine-depleted artificial RNA sequences (RNA with 5.5% uridine content) under two different rNTP reaction conditions. (d) Base substitution error profile observed for unmodified and modified uridine-depleted CLuc mRNA from under the two different rNTP reaction conditions. Colors indicate specific substitution errors…

• 

  Figure 5. In vitro transcription error profile of SP6 RNA polymerase can be modulated by altering ribonucleotide composition in the reaction. Uridine-depleted artificial RNA sequence (uridine content of the sequence is 5.5%) was transcribed with SP6 RNA polymerase and reactions were performed with either equal molar rNTPs or rNTPs proportional to the template sequence composition. Base substitution error profile observed for unmodified and modified uridine-depleted RNA sequence under the two different rNTP reaction conditions. Colors indicate specific substitution errors observed.

• 

  Figure 6. In vitro transcription error profile of T7 RNA polymerase can be modulated by increasing the rATP concentration in the reaction. Artificial RNA sequences RNA1 and RNA2 were transcribed with T7 RNA polymerase and reactions were performed with excess of rATP (16 mM or 20 mM) as compared to the other rNTPs in the reaction (8 mM or 10 mM respectively). (a) Bioanalyzer traces of RNA1 synthesized with either 16 mM or 20 mM rATP demonstrating synthesis of full-length IVT products. (b) Base substitution error profile observed for unmodified and modified RNA sequences under the different rATP concentrations.