N1-Methylpseudouridine and pseudouridine modifications modulate mRNA decoding during translation

Abstract

The ribosome utilizes hydrogen bonding between mRNA codons and aminoacyl-tRNAs to ensure rapid and accurate protein production. Chemical modification of mRNA nucleobases can adjust the strength and pattern of this hydrogen bonding to alter protein synthesis. We investigate how the N1-methylpseudouridine (m¹Ψ) modification, commonly incorporated into therapeutic and vaccine mRNA sequences, influences the speed and fidelity of translation. We find that m¹Ψ does not substantially change the rate constants for amino acid addition by cognate tRNAs or termination by release factors. However, we also find that m¹Ψ can subtly modulate the fidelity of amino acid incorporation in a codon-position and tRNA dependent manner in vitro and in human cells. Our computational modeling shows that altered energetics of mRNA:tRNA interactions largely account for the context dependence of the low levels of miscoding we observe on Ψ and m¹Ψ containing codons. The outcome of translation on modified mRNA bases is thus governed by the sequence context in which they occur.

Fig. 1. Cognate amino acid addition is modestly increased on UUm¹Ѱ, but not m¹Ѱ UU or Um¹ѰU, codons.

A The chemical structures of the nucleobases we investigated. B The formation of ^fMet-Phe (MF) dipeptide as a function of time by E. coli ribosomes containing ³⁵S-^fMet-tRNA^fMet bound to an AUG start codon in the P site, and unmodified (black circles – UUU, n = 23 reactions divided between 2 experiments) or modified (blue squares - m¹ΨUU (n = 36 reactions conducted in 3 experiments), green diamonds - Um¹ΨU (n = 45 reactions over 4 experiments), red triangles - UUm¹Ψ (n = 44 reactions/3 experiments)) codons in the A site. C The K_½ curve for RF1. Fitted k_obs values (n = 12 independent measurements of k_obs) for RF1-catalyzed ³⁵S-^fMet release on UAA (circles) or m¹ΨAA (squares) are displayed as a function of [RF1]. D The K_½ curve for RF2. Fitted k_obs values (n = 11 independent measurements of k_obs) for RF2-catalyzed ³⁵S-^fMet release on an UAA (black circles) or m¹ΨAA (blue squares) are displayed as a function of [RF2]. Error bars in (C) and (D) indicate the standard error of the fitted value of k_obs. Source data are provided as Source Data file.

Fig. 2. m1Ψ impacts amino acid selectivity in vitro and in HEK293 cells.

A Representative eTLC displaying dipeptide products from translation reactions performed with 70 S initiation complexes (ICs) containing an unmodified UUU or m¹ΨUU codon in the A site and total E. coli tRNA aminoacylated with a single amino acid (aa-TC). Relative to ICs containing a UUU codon in the A site, higher levels of miscoded MI and MV dipeptide products and lower levels of MS were generated from m¹ΨUU containing ICs. B Summary of amino acid substitutions observed by mass spectrometry in a luciferase peptide incorporated on m¹Ψ-containing mRNAs translated in 293H cells (Supplementary Table 8). Source data are provided as Source Data file.

Fig. 3. Ѱ and m¹Ѱ impact the rates of the ribosome reacting with near-cognate tRNAs in a sequence context dependent manner.

Plots of dipeptide formation as a function of time (seconds). Miscoding reactions were performed with E. coli ICs containing ³⁵S-^fMet-tRNA^fMet bound to an AUG start codon in the P site, and unmodified (black circles-UUU) or modified (blue squares-Ψ/m¹ΨUU, green diamonds-UΨ/m¹ΨU, red triangles-UUΨ/m¹Ψ) codons in the A site. Purified ICs were reacted with TCs containing (A) Ile-tRNA^Ile(GAU), (B) Leu-tRNA^Leu(CAG)), or (C) Ser-tRNA^Ser(UGA). At least 30 independent time points were collected for each IC in experiments conducted over three or more separate days. D The fitted rate constants (k_obs) for isoleucine, leucine, and serine misincorporation on Ψ- and m¹Ψ- modified codons relative to the fitted rate constants for isoleucine, leucine, and serine misincorporation on a UUU codon. Each ratio has an n of 1 (since each is computed by dividing k_obs,modified/k_obs,UUU) and error bars were calculated by propagating the 95% confidence intervals of each fitted rate constant. Source data are provided as Source Data file.

Fig. 4. Changes in the energetics of mRNA:tRNA interactions correlate with observed differences in Phe and Ile incorporation on Ψ− and m¹Ψ− containing codons.

A Summary of data in Supplementary Tables 4 and 5 displaying how a Ψ and m¹Ψ impact the rate constants for the reaction of near-cognate tRNAs. B, C Summary of MM data. Gray bars reflect the change in energy for interactions between a modified mRNA position (Ψ/m¹Ψ) and the base paring (B) tRNA^Phe(GAA) or (C) tRNA^Ile(GAU) residue (ΔE_bp) relative to an unmodified mRNA U. Black bars reflect the total change in energy (ΔE_{ΣΨ/m1Ψ:X-Y}) for interactions between a modified mRNA position (Ψ/m¹Ψ) and three (B) tRNA^Phe(GAA) or (C) tRNA^Ile(GAU) residues (the base paired nucleotide, and nucleotides 5′ and 3′ the bp) relative to an unmodified mRNA U. D Molecular model of an unmodified sequence coding for a Phe UUU codon and a tRNA^Ile(GAU). The hypermodification t⁶A37 is also shown on the tRNA.