The amino acid substitution affects cellular response to mistranslation

Abstract

Mistranslation, the misincorporation of an amino acid not specified by the "standard" genetic code, occurs in all organisms. tRNA variants that increase mistranslation arise spontaneously and engineered tRNAs can achieve mistranslation frequencies approaching 10% in yeast and bacteria. Interestingly, human genomes contain tRNA variants with the potential to mistranslate. Cells cope with increased mistranslation through multiple mechanisms, though high levels cause proteotoxic stress. The goal of this study was to compare the genetic interactions and the impact on transcriptome and cellular growth of two tRNA variants that mistranslate at a similar frequency but create different amino acid substitutions in Saccharomyces cerevisiae. One tRNA variant inserts alanine at proline codons whereas the other inserts serine for arginine. Both tRNAs decreased growth rate, with the effect being greater for arginine to serine than for proline to alanine. The tRNA that substituted serine for arginine resulted in a heat shock response. In contrast, heat shock response was minimal for proline to alanine substitution. Further demonstrating the significance of the amino acid substitution, transcriptome analysis identified unique up- and down-regulated genes in response to each mistranslating tRNA. Number and extent of negative synthetic genetic interactions also differed depending upon type of mistranslation. Based on the unique responses observed for these mistranslating tRNAs, we predict that the potential of mistranslation to exacerbate diseases caused by proteotoxic stress depends on the tRNA variant. Furthermore, based on their unique transcriptomes and genetic interactions, different naturally occurring mistranslating tRNAs have the potential to negatively influence specific diseases.

Keywords: S. cerevisiae; mistranslation; synthetic genetic array; tRNA variants; transcriptomics.

Figure 1

tRNA variants that mistranslate. (A) Secondary structure of ${\text{tRNA}}_{\mathrm{G}3\text{:}\mathrm{U}70}^{\text{Pro}}$, which mistranslates alanine at proline codons, and ${\text{tRNA}}_{\text{UCU},\mathrm{G}26\mathrm{A}}^{\text{Ser}}$, which mistranslates serine at arginine codons. Nucleotides colored in red denote differences compared to the wild-type tRNA^Pro or tRNA^Ser, respectively. (B) Mass spectrometry analysis of the cellular proteome was performed on a control strain with no additional tRNAs (CY8611) or strains expressing mistranslating tRNA variants ${\text{tRNA}}_{\mathrm{G}3\text{:}\mathrm{U}70}^{\text{Pro}}$ (CY8612) or ${\text{tRNA}}_{\text{UCU},\mathrm{G}26\mathrm{A}}^{\text{Ser}}$ (CY8614). Each point represents one biological replicate. Each strain expressing a mistranslating tRNA had statistically higher frequency of mistranslation compared to the control strain (P ≤ 0.05; Welch’s t-test).

Figure 2

Phenotypic characterization of mistranslating strains. (A) Growth rates for either a control strain with no additional tRNA (CY8611) or strains expressing mistranslating tRNA variants ${\text{tRNA}}_{\mathrm{G}3\text{:}\mathrm{U}70}^{\text{Pro}}$ (CY8612) or ${\text{tRNA}}_{\text{UCU},\mathrm{G}26\mathrm{A}}^{\text{Ser}}$ (CY8614) were determined from growth curves of the strains diluted to an OD₆₀₀ ∼ 0.1 in various media and grown for 24 hours. (YPD—YP with 2% glucose, YP + Galactose—YP with 2% galactose, SC—synthetic complete with ammonium sulfate as the nitrogen source and 2% glucose or 2% galactose as the carbon source, minimal—medium containing ammonium sulfate as the nitrogen source, 2% glucose and adenine, histidine, leucine, lysine, and methionine, SC + MSG—synthetic complete media with monosodium glutamate and glucose as nitrogen and carbon sources). Doubling time was calculated with the R package “growthcurver” (Sprouffske and Wagner 2016). Each point represents one biological replicate. All comparisons within a growth condition are statistically different (Bonferroni corrected P ≤ 0.05; Welch’s t-test), except where indicated (ns = not significant). (B) Strains described in A were transformed with a GFP reporter transcribed from a promoter containing heat shock response elements, grown to saturation in media lacking uracil, diluted 1:100 in the same media and grown for 18 hours at 30°. Cell densities were normalized and fluorescence measured. Each point represents one biological replicate. Statistical comparisons were made between strains expressing a variant tRNA and the control strain (ns = not significant, ** P ≤ 0.005; Welch’s t-test).

Figure 3

Negative genetic interaction networks of the mistranslating tRNAs. Genetic interaction network of temperature sensitive alleles that have negative genetic interactions with ${\text{tRNA}}_{\mathrm{G}3\text{:}\mathrm{U}70}^{\text{Pro}}$ (yellow) and ${\text{tRNA}}_{\text{UCU},\mathrm{G}26\mathrm{A}}^{\text{Ser}}$ (blue). Nodes represent alleles and edges represent negative genetic interactions.

Figure 4

Transcriptome analysis of strains expressing mistranslating tRNA variants. (A) PCA of centered log ratio normalized reads from BY4742 expressing either ${\text{tRNA}}_{\mathrm{G}3\text{:}\mathrm{U}70}^{\text{Pro}}$ (Pro→Ala), ${\text{tRNA}}_{\text{UCU},\mathrm{G}26\mathrm{A}}^{\text{Ser}}$ (Arg→Ser) or an empty vector (WT). Each point represents one biological replicate (n = 3). (B) Heatmap of hierarchical clustered differentially expressed genes (P ≤ 0.05) in response to mistranslation. Fold-change for each gene is the average from three biological replicates. Upregulated genes are colored yellow while downregulated genes are colored blue. Significantly enriched GO terms within each cluster are annotated. (C) Venn diagram of upregulated and downregulated genes (P ≤ 0.05) in yeast strain BY4742 transformed with a centromeric plasmid expressing ${\text{tRNA}}_{\mathrm{G}3\text{:}\mathrm{U}70}^{\text{Pro}}$ (Pro→Ala) or ${\text{tRNA}}_{\text{UCU},\mathrm{G}26\mathrm{A}}^{\text{Ser}}$ (Arg→Ser) as compared to BY4742 transformed with empty vector. (D) Volcano plot highlighting in yellow differentially expressed genes with greater than twofold changes in ${\text{tRNA}}_{\mathrm{G}3\text{:}\mathrm{U}70}^{\text{Pro}}$ relative to the control strain. Points with gene name labels have the largest fold-change relative to control. (E) Volcano plot highlighting in blue differentially expressed genes with greater than twofold changes in ${\text{tRNA}}_{\text{UCU},\mathrm{G}26\mathrm{A}}^{\text{Ser}}$ as in D.