Fig. 7: VACV INR controls 5′ end formation of viral mRNAs.
From: Vaccinia virus mRNAs containing long 5′-poly(A)-leaders lack a canonical 5′-methylguanosine cap
A 5′ RACE analysis of EGFP reporter mRNAs transcribed from the intermediate wild-type G8R promoter (G8RP) or its point-mutated version (G8RPM). Single-nucleotide A/C substitution in G8R INR (empty arrow, underlined in red) resulted in the reduction in the number of added nontemplated adenosine residues and dramatic shortening of the 5′ poly(A) leaders. The upper sequence corresponds to the viral template DNA. TSS and INR are marked with a filled black arrow and a black underline, respectively. The sequences below represent individual sequenced cDNA clones. The 5′ untranslated regions are shown up to the ATG translation start codon. Nucleotides identical to the viral template DNA are labeled in red. Nucleotides added in a nontemplated manner are labeled in green. The guanosine residues corresponding to the 5′ mRNA cap are marked in black. B Statistical representation of panel (A), including results from 5′ RACE analysis of natural viral G8R mRNAs (G8RN; Supplementary Fig. S2 and Supplementary Table S2) for comparison. The general description of the box plot is the same as that in Fig. 2. **** indicates padj < 0.0001. A nonparametric Kruskal-Wallis test, followed by the Dunn post hoc test with Benjamini-Hochberg FDR p-value adjustment revealed statistically significant differences in the 5′ poly(A) leader length between G8RPM and G8RN mRNAs (padj = 0.000037) and G8RPM G8RP mRNAs (padj = 0.000037). C Single A/C substitution in G8R INR leads to a significant reduction in the 5′ poly(A) leader occurrence and its length. Bars represent the percentages of 5′ polyadenylated transcripts among all mRNAs transcribed from the G8RN, G8RP, and G8RPM promoters. Error bars depict 95% confidence intervals; **** indicates p ≤ 0.0001. D Single A/C substitution in G8R INR increases the occurrence of 5′ capped transcripts. Bars represent the percentages of 5′ capped mRNAs transcribed from the G8R, G8RP, and G8RPM promoters. Error bars depict 95% confidence intervals; * indicates p < 0.05. In total, 61 sequences were used for the analyses depicted in panels (B–D) (Supplementary Tables S7, S8). Two-sided Fisher’s exact test showed statistically significant association of point-mutated INR with low occurrence of poly(A) leaders (G8RPM x G8RN, p = 0.0001; G8RPM x G8RP, p = 0.0001) and higher occurrence of m7G cap (G8RPM x G8RP, p = 0.0248) in G8RPM mRNA.
