Binding of Glycyl-tRNA synthetase to Mengovirus RNA stimulates translation

Figures

• 

  Figure 1.. GARS binding elements (GBEs) in Picornavirus RNA genomes. (A) The Poliovirus RNA genome with the 5′UTR structure (stem-loop structures numbered with roman numerals) and blowups showing the stem-loop domain (dom) II and the apical part of dom V. The GBE in dom V [22] is in dark blue, and the putative second GBE in dom II [41, 68] is in light blue. The color code shows base pair probability (bpp) values for bases predicted to pair, but single-strand probabilities for bases predicted to be unpaired according to Vienna RNAalifold output style, as detailed in the methods section. (B) The Mengovirus genome with the GBE in the 3′UTR (dark blue) and the GBE in the 5′UTR (light blue), with blowups of the apical part of dom 3 in the 5′UTR and the 3′UTR, both of Cardiovirus clade 1 (which includes Mengovirus).

• 

  Figure 2.. Replication of the Mengovirus genome depends on the 3′UTR GARS binding element. (A) The Mengovirus (MV) genome with the HiBiT tag inserted near the N-terminus of 2B. (B) Plaque assays after infection of HeLa cells with Mengovirus contained in the medium diluted supernatant of HeLa cells transfected with the wild-type (wt) or mutated (mt) 3′UTR GBE. (C) Quantification of plaque numbers after infection of wt or mt MV in HeLa cells in a time course from 24 to 96 h post-infection (hpi). (D) Quantification of plaque sizes from the experiment in (C). (E) HiBiT expression from MV genomes in HeLa cells as well as in SK-N-AS and Neuro-2A neuroblastoma cells at 24 h post-transfection (hpt). n: number of biological replicates; significances with two-sided P-values: * (P < 0.05), ** (P < 0.01), *** (P < 0.001) and **** (P < 0.0001); n.s. = not significant (P > 0.05).

• 

  Figure 3.. Expression of full-length Mengovirus genomes at early time points. (A) The Mengovirus genome with the HiBiT tag and the 3D polymerase (P) with the 3′UTR GBE in either wild-type (+) or mutant (-) version. (B) HiBiT expression from full-length MV RNA genomes after lipofectamine transfection in HeLa cells at the indicated hours post-transfection (hpt). MV genomes contained the 3′UTR GARS binding element in the wild-type (GBE+) or mutated (GBE-) form and the 3D polymerase (P) in its active wild-type form (P+) or inactivated (P-). Expression with the GBE + P + construct at 3 h was set to 1. Below the graph, n(P+) indicates the number of experiments for the P + constructs for each time point, and p(P+) indicates the significance with two-sided p-values comparing GBE + P + and GBE-P + constructs. Significances were calculated with GraphPad´s nonparametric Wilcoxon signed-rank test, with * (P < 0.05), ** (P < 0.01), *** (P < 0.001) and **** (P < 0.0001); n.s. = not significant (P > 0.05). (C) blowup from (B) at early time points. (D) HiBiT expression from full-length MV RNA genomes as in (B) but using electroporation transfection. Expression with the GBE + P + construct at 0.5 h was set…

• 

  Figure 4.. The GARS binding element affects translation efficiency. Two different translation reporter RNA systems were used. In (A), the translation reporter RNA contains the MV IRES driving expression of a very small ORF consisting of a short 11 AS linker and the 11 AS HiBiT tag, followed by the MV 3′UTR with the GBE in wild-type (wt) or mutated (mt) form. (B) HiBiT expression in HeLa cells transfected with the small HiBiT ORF reporter RNA constructs shown in (A). In (C), the ORF from (A) was replaced by the ORF for the secretable Nanoluciferase (secNLuc, grey box), and the GBEs in the 5′UTR and the 3′UTR were mutated individually or in combination. (D) HiBiT expression in the medium of HeLa cells transfected with the secNLuc reporter RNA constructs shown in (C). (E) Reporter RNA stability assays in the HeLa cells of the experiments shown in (D). Reporter RNA abundance was analyzed by RT-qPCR. (F) HiBiT expression normalized to reporter RNA stability. Data points with the same colour correspond to results obtained within the same experiment.

• 

  Figure 5.. Physical binding of the GARS protein to the GARS binding elements. (A) The GARS protein with a SNAP fusion tag was used for the binding assays. The protein and its derivatives were expressed in HeLa cells after transfection of corresponding expression plasmids, and the cell lysates were then used as a source for the GARS fusion proteins. In the SNAP-GARS fusion protein, the anticodon binding domain (ABD) was either present (wt) or deleted (ΔABD). (B) The RNA of the apical part of the Poliovirus (PV) IRES domain V with the GBE in wild-type (wt) or mutated (mt) form. (C) The RNA of the Mengovirus (MV) 3′UTR with the GBE in wild-type (wt) or mutated (mt) form. (D) Binding assays. 32P-labeled radioactive RNA of the PV IRES domain V or the MV 3′UTR with the GBE in wt or mt form were added to HeLa cell lysates made after transfection of plasmids expressing SNAP-GARS fusion proteins. 32P-labeled RNA binding to the SNAP-GARS protein was pulled down using SNAP beads, and radioactivity in RNA was measured. (E) 3′-end-biotinylated MV reporter RNAs containing the HiBiT tag and the GBE in the Mengovirus 5′UTR or the 3′UTR in wt or mt form or in combination. (F) NeutrAvidin bead affinity…

• 

  Figure 6.. Mutation of the GARS binding elements in the Mengovirus 5′UTR and 3′UTR impairs the interaction with proteins. Cells were transiently transfected with 3′-end-biotinylated short reporter RNAs with 5′UTR and 3′UTR GBE either wild-type (wt) or mutated (mt) individually or in combination (see Fig. 4A). After cell lysis, RNA constructs were affinity purified on NeutrAvidin beads, and the associated proteins were identified by LC-MS/MS. (A–C) Violin plots show total numbers (below violins) and log2 protein intensities (y-axis) of all identified proteins across five or four biological replicates (A), while the other panels show the distribution of values from protein sets with significantly altered RNA binding compared to the wild RNA based on P-value threshold of < 0.1 (B) or < 0.05 (C) (Student’s t-test).

• 

  Figure 7.. Mengovirus 3′UTR GBE mutation impairs the interaction with protein networks involved in RNA metabolism and translation. (A) Visualization of protein-protein-interaction (PPI) networks of the 166 factors whose association differs between wild type and mutant RNA (Fig. 6B). Grey lines visualize physical PPIs based on experimental evidence according to the STRING data base, version 12.0. The network components were further categorized for their biological function by overrepresentation analysis. Blue and red colors of the network nodes mark components of two highly enriched pathway terms involved in RNA metabolism and translation based on the databases WikiPathways, STRING Clusters and KEGG (as shown in Supplementary Fig. S4). Translation factors that are included in the displayed network and that have known physical interactions within the network are marked in orange for initiation factors and in green for elongation factors. (B) The five top pathways, of which No. 1 and 2 are shown in the above network, plus translation factors. Underlying data are shown in Supplementary Fig. S4.

• 

  Figure 8.. Mutation of the 3′UTR GBE reduces the binding of many proteins. Volcano plots of changes in protein intensities from the data shown in Fig. 6, obtained with pull-downs with the Mengovirus reporter RNA shown in Fig. 4A, and variations in the protein intensities obtained with the GBE mutant RNAs in relation to the wild-type reporter RNA. Negative Log2 difference values mean less protein binding to the mutant RNA compared with the WT RNA; positive values mean more protein binding. (A) Only the 5′UTR GBE mutated, (B) only the 3′UTR GBE mutated, and (C) both 5′ and 3′UTR GBEs mutated. GARS, as well as selected protein groups, are marked in color.

• 

  Figure 9.. Reduced binding of selected functional protein groups to Mengovirus RNA with a mutated GARS binding element. Relative abundance of proteins bound to 3′-end-biotinylated short Mengovirus reporter RNA with the wild-type (wt) or mutated (mt) GBE in the 5′UTR and in the 3′UTR (see Fig. 4A) after NeutrAvidin bead affinity purification and mass spectrometry analysis. The data shown in Fig. 6 were selected for functional protein groups like in Fig. 8. Each data point shows the relative abundance of a specific protein binding to the 5′UTR or 3′UTR mutant GBE RNA or the 5′-3′-UTR double mutation, while the mean value includes all proteins in the respective group. Binding to the wild-type RNA was normalized to 1.0. Significances were calculated with GraphPad´s nonparametric Wilcoxon signed-rank test, with * (P < 0.05), ** (P < 0.01), *** (P < 0.001) and **** (P < 0.0001); n.s. = not significant (P > 0.05). (A) Proteins of the small ribosomal 40S subunit. (B) Proteins of the large ribosomal 60S subunit. (C) Heterogeneous nuclear ribonucleoproteins (hnRNP proteins). (D) Translation initiation factors. (E) Translation elongation factors. “n” indicates the number of different proteins used for…

• 

  Figure 10.. Binding of GARS and translation machinery components to single UTR GBEs. (A) Short 3′-end-biotinylated MV reporter RNAs containing the 5′UTR alone with the GBE in wt or mt form. (B) Short RNA of the Mengovirus 3′UTR alone with the GBE wt or mt. (C) Short RNA of the TMEV 3′UTR alone with the GBE wt or mt. (D) NeutrAvidin bead affinitiy purification and Western Blot of GARS using the above 3′-biotinylated RNAs (upper part); GAPDH detection was used as control (lower part). 97.5% of the total lysate were used for the affinity purification (pull-down, left panel), and 2.5% of the total lysate were used as input lysate control (right panel). (E–H) Relative abundance of proteins binding to 3′-end-biotinylated short RNAs of the apical loop of IRES domain I (with the GBE wt or mt) as shown in (A) and on the left side of Fig. 1B, of the Mengovirus 3′UTR (with the GBE wt or mt) as shown in (B) or the right side of Fig. 1B. Thereby, (E) shows proteins of the small ribosomal 40S subunit, (F) of the 60S subunit, (G) translation initiation factors and (H) translation elongation factors. As in Fig. 9, “n” indicates the number of different protein species used for calculation of mt/wt ratios,…