Mutagenesis Mapping of RNA Structures within the Foot-and-Mouth Disease Virus Genome Reveals Functional Elements Localized in the Polymerase (3D pol )-Encoding Region

Figures

• 

  FIG 1. Extent of the conserved RNA secondary structures within the FMDV genome. (A) A scan of pairwise distance (in gray) and mean folding energy difference (MFED) (in red) prepared using SSE v1.4 software and 118 FMDV sequences representing all seven serotypes. The mean values for successive 400-nt fragments across the genome are plotted (where each 400-nt segment overlapped its neighbors by 380 nt). The light red shading represents error bars showing standard deviations from the mean for each MFED data point. The y axis shows genome nucleotide position. (B) MFED values for the same FMDV genomic sequences but grouped into Eurasian (A, Asia 1, C, and O serotypes) (dark blue) or SAT (SAT 1 to 3 serotypes) (light blue) clusters. (C) A dot plot graphical representation of RNA structures that were conserved across all seven FMDV serotypes. The x axis and y axis represent FMDV genome positions, with each dot representing a single pairing between two nucleotides, one with its position marked on the x axis and the other one with its position marked on the y axis. The three pale blue arrows indicate the locations of the S-fragment, cre plus IRES, and SL1 plus SL2 structures on the dot plot…

• 

  FIG 2. Comparison of average MFED values for wild-type (WT) and CDLR-scrambled sequences. Mean folding energy difference (MFED) for the regions encoding nonstructural proteins (nsps) of 118 FMDV field isolates representing all seven serotypes (blue dots), WT ptGFP replicon used in this study (red dot), and CDLR scrambled sequences (yellow dots). Among the latter is the CDLR scrambled sequence used in this study to generate replicon mutants (purple dot). To obtain CDLR-scrabbled sequences, the sequence of the regions encoding the nsps of the WT replicon was permuted 50 times by codon shuffling to minimize RNA secondary structure, while preserving protein coding, native dinucleotide frequencies, and codon usage.

• 

  FIG 3. Replication of CDLR replicons within BHK-21 and MDBK cells. (A) Schematic representation of CDLR replicons. CDLR-permuted sequences were first inserted into a subclone encoding the nonstructural proteins (nsps) of the genome (Δ1D-polyA) before cloning into the WT ptGFP replicon. (B) CDLR replicon insert sizes and number of mutations within each region. Regions were chosen based on unique restriction site usage within the regions encoding nsps. Mutations were introduced as described in Materials and Methods. (C) IncuCyte data represent the average cell (green object) GFP intensity (also called green calibrated unit [GCU]) per well at 8 h posttransfection. Results are the means of three independent experiments plus standard errors (error bars). Significant differences between WT ptGFP and CDLR replicons were determined (***, P < 0.001). The replication-incompetent 3Dpol active site mutant (GDD → GNN) ptGFP-3DpolGNN was used as a negative control.

• 

  FIG 4. Schematic representation of predicted conserved RNA structures located at the 3′-terminal end of the region encoding 3Dpol. (A) Schematic representation of conserved (in all FMDV serotypes) RNA secondary structures located at the 3′-terminal end of the region encoding 3Dpol (i.e., 3D3). Conserved putative stem-loops (ORF-SL49 to ORF-SL53) are shown, where two stem-loops located in the 3′ UTR described before (SL1 and SL2) act as the control of the computational prediction. Nucleotide positions which form conserved pairing were color coded according to number of pairing types (red for 1 and blue for 5) and conservation of a pairing (dark shades for nucleotide pairing occurred in all FMDV isolates to light shades for lack of nucleotide pairing in two FMDV isolates). Positions colored in light gray show lack of pairing for three or more FMDV isolates. Black circular outline indicates nucleotide position where a substitution resulted in an alternative pairing (see the legend in the top left corner of panel A for detail). Unstructured regions are represented as dark gray lines and are not drawn to scale. Table S3 in the supplemental material specifies details represented graphically…

• 

  FIG 5. Disruption of the predicted RNA secondary structures by silent mutagenesis. The conserved stem-loops identified in the 3′-terminal end of the region encoding 3Dpol (i.e., 3D3) of FMDV were predicted individually by Mfold for the WT ptGFP replicon. Predicted WT stem-loops were mutated to cause the highest possible disruption or change to the RNA structure without affecting neighboring stem-loops while keeping the same amino acid sequence and dinucleotide ratio (i.e., CpG and UpA). Predicted WT and mutated stem-loops visualized using the Forna web server are shown. Nucleotides highlighted in green represent mutated positions, while red brackets represent positions of the hairpin loop in the WT structures and their altered position in the disrupted structures after mutagenesis.

• 

  FIG 6. Effect of individual stem-loop (ORF-SL49, -50, -51, -52, and -53) mutagenesis on replication of the FMDV replicon. (A) Schematic representation of FMDV replicon constructs containing stem-loop mutations (individually, ORF-SL49mut, ORF-SL50mut, ORF-SL51mut, ORF-SL52mut, and ORF-SL53mut; in combination, ORF-SL49-50,52-53mut). Please see Results for detailed description of mutants. Sequence inserts containing stem-loop mutations were cloned directly into the ptGFP replicon. (B) IncuCyte data represent the average cell (green object) GFP intensity per well at 8 h posttransfection within BHK-21 and MDBK cells. Results are the means of three independent experiments plus standard errors. Significant differences between WT ptGFP and ORF-SLmut replicons were determined (*, P < 0.05; ***, P < 0.001).

• 

  FIG 7. Effect of combined mutagenesis of stem-loops ORF-SL51, ORF-SL52, and ORF-SL53 on replication of the FMDV replicon. (A) Schematic representation of FMDV replicon constructs containing combined stem-loop mutations (ORF-SL51,52mut, ORF-SL51,53mut, and ORF-SL51-53mut). Sequence inserts containing stem-loop mutations were cloned directly into the ptGFP replicon. (B) IncuCyte data represent the average cell (green object) GFP intensity per well at 8 h posttransfection. Results are the means of three independent experiments plus standard errors. Significant differences between WT ptGFP and ORF-SLmut replicons were determined (***, P < 0.001).