Skip to main content
. 2021 Aug 13;12:4913. doi: 10.1038/s41467-021-24969-x

Fig. 2. Nsun2 ablation depletes tRNAGly causing decreased expression of glycine-rich synaptic proteins.

Fig. 2

a Left, tRNA sequencing of mouse forebrain in Nsun2 KO and WT (n = 3/genotype). Note decreased expression in 10/10 tRNAGly isodecoders (green) among 162 total isodecoders (gray). Isodecoders that are differentially expressed (adj. P < 0.05) between KO and WT have black outlined circles. Right, 11 isodecoders were significantly altered by Nsun2 ablation (FDR P < 0.05) with 7/9 downregulated isodecoders belonging to the tRNAGly family. Data are presented as mean values +/− SEM. b Unbiased proteomic screen identified 1488 proteins significantly altered (FDR adj. P < 0.05) after Nsun2 ablation (635 decreased, 853 increased; n = 5–6/genotype). c Left, gene ontology analysis using StringDB identified the top ten significant KEGG pathways downregulated in Nsun2 KO including those involved in synaptic functioning and neurotransmission, while one of the top ten significant KEGG pathways upregulated was the family of aminoacyl tRNA ligases. Right, box plot for glycine content in three subgroups of proteins, n.s (not significant; 4332 proteins; median: 6.403, IQR: 2.811, whiskers represent min (0) and max (35.93)), decreased (635 proteins; median: 6.849, IQR: 2.7, whiskers represent min (0.73) and max (24.36)), and increased (853 proteins; median: 6.234, IQR: 2.852, whiskers represent min (0) and max (27.54)) proteins in Nsun2 KO cortex vs. WT. (two-tailed, Mann–Whitney test; ***P < 0.0001, *P = 0.046). d Proteins produced by genes with higher glycine codon content exhibited a significant decrease after Nsun2 ablation. Left, LogTER (top, red) corresponding to glycine codon content enrichment scores (bottom, yellow-blue). Heatmap represents gene clusters from the top 50% of glycine codon-enriched genes, with yellow representing most enriched and blue representing least enriched. The red and dark-blue borders indicate the statistical significance of the calculated hypergeometric P values for that cluster after Bonferroni’s correction, obtained from the z scores associated with the mutual information (MI) values calculated with 10,000 randomization tests (adj. P < 0.05). e Schematic of translation at the ribosome and locations of the A, P, and E sites. f Top, RiboSeq data for A-site occupancy (KO vs. WT) (top left) and relative ribosomal dwell time (KO/WT) (top right) grouped by amino acid. Bottom, ribosome A-site occupancy shown for all 64 codons. Note the significant increase in DT and A-site occupancy for Gly (GGN) codons and specifically GGA codons (n = 2 KO/2 WT). Source data are available as a Source Data file.