An intra-family conserved high-order RNA structure within the M ORF is important for arterivirus subgenomic RNA accumulation and infectious virus production

Fig 1. Synonymous site conservation analysis of PRRSV sequences. (A) Full genome analyses for PRRSV-1 (upper) and PRRSV-2 (lower) sequence alignments. The red line shows the probability that the observed conservation could occur under a null model of neutral evolution at synonymous sites, whereas the brown line depicts the ratio of the observed number of substitutions to the number expected under the null model. Note that similar plots for PRRSV-2 (albeit with different input sequences) have been presented previously in Firth (17) and Cook et al. (20). (B) Expanded views of the M and N ORF regions. Positions of the predicted RNA structures are indicated with colored bars: yellow – stem-loop 1, SL1; orange – stem-loop 2, SL2; and green – 5′ and 3′ fragments of the extended stem-loop, extSL.

Fig 2. RNA structure prediction of the synonymously conserved region within the PRRSV M ORF. RNA structure predictions for the 3′ half of the M ORF and the beginning of the N ORF regions for the PRRSV-1 (14323–14557 nt, DQ489311.1) and PRRSV-2 (14666–14900 nt, KC469618.1) sequences. Minimum free energy structures were predicted with RNAstructure (22). The SL1, SL2, and extSL duplex regions are annotated, as well as an additional conserved region (CR), the transcription regulatory sequence for the N protein sgRNA 7 (TRS7), an additional sgRNA 7 TRS in PRRSV-2 (TRS7.1), and the N ORF initiation codon.

Fig 3. Analysis of the SL1, SL2, and extSL duplexes in PRRSV-1 and PRRSV-2 genome sequences. Each unique sequence is shown, with the number at the left indicating how many times that unique sequence occurs among the 461 or 1518 sequences. One PRRSV-1 extSL sequence was omitted, as it contained an ambiguous nucleotide code (viz. "R"). The base pairings predicted in NC_001961 (PRRSV-2) are highlighted in yellow (SL1), orange (SL2), or green (extSL), except that G:U base pairs are indicated by highlighting either the G or the U in cyan. Compensatory substitutions in NC_043487 (PRRSV-1) are highlighted in pink. Other compensatory substitutions present in various sequences are highlighted in green or purple. Conserved single-nucleotide paired bulges present in extSL and in PRRSV-1 SL2 are indicated with arrows. Unpaired nucleotides are not highlighted.

Fig 4. Conservation of SL1, SL2, and extSL across genera Betaarterivirus and Nuarterivirus. (A) The SL1-SL2-extSL region of NCBI RefSeqs in the Betaarterivirus/Nuarterivirus clade with predicted structures annotated using a color scheme similar to that used in Fig. 3. M ORF codons are separated by spaces. (B) RNAz analysis of an alignment of the six RefSeqs NC_001961 (PRRSV-2), NC_028963 (rat arterivirus 1), NC_040535 (rodent arterivirus), NC_043487 (PRRSV-1), NC_048210 (rodent arterivirus), and NC_075972 (rodent arterivirus). The y-axis plots 1 − P, where P is the RNAz SVM RNA-class probability statistic. Bars correspond to the midpoints of 120 nt windows, mapped from alignment coordinates (RNAz output) back to NC_001961 (PRRSV-2) coordinates.

Fig 5. Structure probing of the PRRSV M RNA. RNAs derived by transcription of pcDNA3.1/PRRSV-1 (A and B) and pcDNA3.1/PRRSV-2 (C and D) with SP6 RNA polymerase was 5' end-labeled with γ-32P-ATP and subjected to limited RNase or chemical cleavage using structure-specific probes. Sites of cleavage were identified by comparison with a ladder of bands created by limited alkaline hydrolysis of the RNA (OH-) and the position of known RNase U2 and T1 cuts, determined empirically. Products were analyzed on 6% (A and C) or 10% (B and D) acrylamide/7M urea gels. Enzymatic structure probing was done with RNases T1, U2, and V1. Uniquely cleaved nucleotides were identified by their absence in untreated control lanes (0). The number of units of enzyme added to each reaction is indicated. Chemical structure probing was done with imidazole (4 h, I). The water lane (0) represents RNA, which was dissolved in water, incubated for 4 h, and processed in parallel to the imidazole-treated sample. R represents an aliquot of the purified RNA loaded directly onto the gel without incubation in a reaction buffer. Panel B is a combined picture from two different gels, with probing for RNase T1 from one gel and…

Fig 6. Design of mutagenesis for the high-order RNA structure in the M ORF of PRRSV-1. Synonymous mutations were introduced in stem-loop1 (A), stem-loop2 (B), and extSL (C). Nucleotides forming stem1, stem2, and extSL are highlighted in yellow, orange, and green, respectively. Parentheses or brackets are used to indicate predicted base pairings in WT. The free energy of each structure after mutation is shown on the right side of each sequence.

Fig 7. Mutations disrupting the conserved RNA structure within the PRRSV-1 M ORF decrease the level of recombinant virus production. (A-B) BHK-21 cells transfected with in vitro-transcribed gRNA and recombinant viruses were harvested at 18 hpt (A) and 36 hpt (B). Virus titers were quantified on MARC-145 cells using the TCID50 method. Each sample was tested in triplicate. The titers of mutant viruses were compared with that of WT virus by one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. *: P < 0.05, **: P < 0.01, ***: P < 0.001, ****: P < 0.0001. (C-E) Pearson correlation between the estimated free energy of structures and viral titers for stem 1 mutants (C), stem 2 mutants (D), and extSL mutants (E). The titers of the mutant viruses at 36 hpt were plotted against the corresponding free energy of the mutated stem structures. A negative linear regression is depicted as a black line. The Pearson correlation coefficients (R) are shown in the corresponding plots.

Fig 8. Effect of M ORF stem-loop mutations on the accumulation levels of (+) and (−) genomic and subgenomic RNA 2, 6, and 7. (A) Diagram of arteriviral RNA transcription and replication. (B-M) In vitro transcribed gRNAs of WT or mutant viruses were transfected into BHK-21 cells. Viral RNAs were quantified at 18 hpt. The level of viral RNAs was normalized to the host housekeeping gene (TBP) first before comparing it with other viral RNAs. Each sample was tested in triplicate. (B) Ratio of mutant (+) gRNA to WT (+) gRNA. (C-E) Ratio of (+) sgRNA 2/6/7 to (+) gRNA for each virus. (F) Ratio of mutant (-) gRNA to WT (-) gRNA. (G-I) Ratio of (-) sgRNA 2/6/7 to (-) gRNA for each virus. (J) Ratio of mutant (-) gRNA to WT (+) gRNA. (K-M) Ratio of (-) sgRNA 2/6/7 to (+) gRNA for each virus. Statistical differences were evaluated by one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. *: P < 0.05, **: P < 0.01, ***: P < 0.001, ****: P < 0.0001.

Fig 9. Correlations between relative accumulation of sgRNAs and infectious virus production or thermodynamic stability of M ORF structures. (A) Virus titer at 36 hpt compared with mutant (+) gRNA to WT (+) gRNA ratio. (B-D) Virus titer at 36 hpt compared with (+) sgRNA 2 to (+) gRNA ratio (B), (+) sgRNA 6 to (+) gRNA ratio (C), and (+) sgRNA 7 to (+) gRNA ratio (D). (E) Virus titer at 36 hpt compared with mean deviation of sgRNA ratio from WT. (F-H) Mean deviation of sgRNA ratio from WT compared with the free energy of stem 1 mutants (F), stem 2 mutants (G), or extSL mutants (H). Data fitting linear regressions are depicted as black lines. The corresponding Pearson correlation coefficient value (R) is shown for each plot.