Ribosome-induced RNA conformational changes in a viral 3′-UTR sense and regulate translation levels

Figures

• 

  Fig. 1. Location, structure, and function of the TYMV 3′-UTR. a TYMV gRNA and sgRNA. ORFs and the 3′-UTR are labeled. MP movement protein, PP polyprotein, CP coat protein. Red: stop codons. The CP ORF is translationally silent in the gRNA. Both RNAs end with identical 3′ ends that contain the UPD + TLS domain and both are valylated (dashed box). b Secondary structure of the 3′-UTR. Red: CP ORF stop codon. Specific parts of the structure are labeled. c Schematic of TYMV TLS tRNA mimicry and translation enhancement. The 3′ end of the viral genome is aminoacylated by the host’s valine tRNA synthetase (green). This is followed by binding of eukaryotic elongation factor 1A (eEF1A, yellow) and translation enhancement by an unknown mechanism

• 

  Fig. 2. TYMV 3′-UTR context alters the rate of translation. a Luciferase (LUC) reporter design. Elements that were altered are indicated. The top reporter represents the 0 nt spacer, the middle reporter represents different spacer nucleotide lengths, and the bottom reporter represents the LUC stop codon mutant reporter (indicated by the magenta symbol). The red symbol represents the LUC stop codon and the stop codon that is inside the UPD. Reporters were capped unless indicated. b–d In vitro translation assays with: b two different spacers between the two stop codons; c two spacer lengths and the upstream stop codon mutated; d four spacer lengths with data showing the mean translation levels for n ≥ 3 replicates. Absolute differences in LUC activity between panels are due to different amounts of lysate used. Error bars represent one s.e.m. of n ≥ 3 replicates. Sequence details in Supplementary Fig. 2

• 

  Fig. 3. Structural coupling between RNA domains. a MCM with NMIA. X-axis: mutation position. Y-axis: sequence position. Domains indicated. Orange: modification patterns in a domain due to mutations within the same domain. Red and magenta: changes in one domain due to mutations in the other domain. b Hierarchical clustering of the difference of each mutant from the wild-type RNA (Supplementary Fig. 4). Blue: less reactive. Red: more reactive. Yellow and cyan: mutants U8A and G10C, respectively

• 

  Fig. 4. Functional coupling between RNA domains. a Secondary structures of the U8A and G10C mutants, colored to reflect the difference in chemical probing between mutant and WT. The new secondary structure is derived from the x-ray crystal structure (see refs. 9,36 for previously predicted secondary structures). Base shading represents NMIA probing and tick marks represent DMS or CMCT probing, colored as in Fig. 3b. b Left: in vitro valylation assay comparing mutants U8A, U24A, the double mutant U8A + U24A, G10C, and the compensatory mutant G10C + C22G to WT and a mutant with altered anticodon loop (ACmut) to inhibit valylation. Right: Translation assay comparing the U8A and G10C mutants to WT RNA reporters used to mimic the gRNA (blue) and sgRNA (red). Error bars represent one s.e.m. of n≥ 3 replicates

• 

  Fig. 5. TYMV 3′-UTR structure. a Secondary structure of the crystallized RNA. Lowercase: sequences altered for crystallization; this includes nucleotides that enhance aminoacylation in vitro71. b Ribbon representation of the structure, colors match a

• 

  Fig. 6. Structure of the UPD. a Secondary structure diagram of the UPD, colored by element (colors do not match Fig. 5) and labeled. Grey arrows indicate where bases interact with the minor grooves and non-canonical base pairs are shown. b Structure of the UPD in ribbon representation, colored to match a. Nucleotides are numbered

• 

  Fig. 7. Interdomain communication and the spine structure. a Stereoview of the linker region and surrounding structure. Yellow spheres indicate the O4′ atom to highlight the double inversion. Grey hashes: stacking. The Z-DNA-like motif is found in the syn-G27 and A26 step. b In vitro translation assays of a reporter with the interdomain linker length increased to 12 nucleotides. Data from two different spacers is included for comparison. Error bars represent one s.e.m. of n ≥ 3 replicates. c Left: A spine of continuously stacked bases extends from the UPD to the T-loop. Nucleotides in the spine are shown in stick representation (colored to match Fig. 5, G10C and U8A positions are shown). Right: Secondary structure of the TLS + UPD, with nucleotides visible to the left highlighted and colored to match. The stop codon in the UPD is boxed in black. The location of the leading edge of the ribosome when the stop codon is in the A site is shown. d Surface representation of the structure colored by relative crystallographic B-factor, with red highest and blue lowest

• 

  Fig. 8. A model for ribosome-induced programmed conformational changes. Schematic cartoon of the model of translation regulation on the sgRNA. With few translating ribosomes, the UPD sensor is mostly undisturbed, the TLS is stabilized, and translation is enhanced. When translation is high (many ribosomes), the UPD spends more time in an unfolded state, resulting in destabilization of the TLS structure and decreased translation enhancement