Ribavirin-resistant variants of foot-and-mouth disease virus: the effect of restricted quasispecies diversity on viral virulence

Abstract

Mutagenic nucleoside analogues can be used to isolate RNA virus high-fidelity RNA-dependent RNA polymerase (RdRp) variants, the majority of which are attenuated in vivo. However, attenuated foot-and-mouth disease virus (FMDV) high-fidelity RdRp variants have not been isolated, and the correlations between RdRp fidelity and virulence remain unclear. Here, the mutagen ribavirin was used to select a ribavirin-resistant population of FMDV, and 4 amino acid substitutions (D5N, A38V, M194I, and M296V) were identified in the RdRp-coding region of the population. Through single or combined mutagenesis using a reverse genetics system, we generated direct experimental evidence that the rescued D5N, A38V, and DAMM mutants but not the M194I and M296V mutants are high-fidelity RdRp variants. Mutagen resistance assays revealed that the higher replication fidelity was associated with higher-level resistance to ribavirin. In addition, significantly attenuated fitness and virulence phenotypes were observed for the D5N, A38V, and DAMM mutants. Based on a systematic quantitative analysis of fidelity and virulence, we concluded that higher replication fidelity is associated with a more attenuated virus. These data suggest that the resulting restricted quasispecies diversity compromises the adaptability and virulence of an RNA virus population. The modulation of replication fidelity to attenuate virulence may represent a general strategy for the rational design of new types of live, attenuated vaccine strains.

Importance: The ribavirin-isolated poliovirus (PV) RdRp G64S variant, the polymerases of which were of high replication fidelity, was attenuated in vivo. It has been proposed (M. Vignuzzi, E. Wendt, and R. Andino, Nat. Med. 14:154-161, http://dx.doi.org/10.1038/nm1726) that modulation of replication fidelity is a promising approach for engineering attenuated virus vaccines. The subsequently mutagen-isolated RdRp variants also expressed the high-fidelity polymerase, but not all of them were attenuated. Few studies have shown the exact correlation between fidelity and virulence. The present study investigates the effect of restricted quasispecies diversity on viral virulence via several attenuated FMDV high-fidelity RdRp variants. Our findings may aid in the rational design of a new type of vaccine strain.

FIG 1

Isolation of ribavirin-resistant FMDV mutants. (A) Schematic of the FMDV RNA genome. The 3Dpol region is shaded in gray. A 500-bp section of the viral capsid gene was RT-PCR amplified and cloned for sequencing. Individual clones were used to determine the mutation frequencies presented throughout this study. (B) Generation of ribavirin-resistant FMDV mutants. The WT virus was serially passaged in the presence (gray) or absence (black) of 50 μM ribavirin at an MOI of 0.01. p0 is the initial WT stock titer. The mean titers of harvests ± standard deviations are shown (n = 3) (**, P < 0.01 by Student's t test). (C) Resistance of the D5N, A38V, and DAMM mutants to ribavirin. Matched titers of each mutant, the p18 population, or the WT were inoculated into triplicate wells with BHK-21 cells and 50 μM ribavirin. The mean titers ± standard deviations are shown (n = 3) (*, P < 0.05; **, P < 0.01 [determined by Student's t test]). (D and E) None of the mutants were resistant to 5-FU (D) or AZC (E). BHK-21 cells pretreated with 1,200 μM 5-FU or 1,000 μM AZC were infected with each mutant; WT FMDV and the ribavirin-selected p18 population at an MOI of 0.01 were used as controls. Within 72 h after infection, the titer of progeny virus was determined by a TCID₅₀ assay using BHK-21 cells. The mean virus titers ± standard deviations are shown (n = 3) (no significant differences were identified by Student's t test [ns]).

FIG 2

Mutation frequencies of each mutant and WT FMDV. Three independent stocks of virus were generated in ribavirin-free medium (A) and in medium containing 100 μM ribavirin (B). A mean of 70 partial P1 sequences (approximately 35,000 nucleotides per replicate) were obtained. The mean mutation frequencies (number of polymorphisms per 10,000 nt sequenced) ± standard deviations represent the averages of all replicates; the same pattern of reduced mutation frequency for each mutant compared to WT FMDV was observed for each replicate (n = 3) (*, P < 0.05; **, P < 0.01 [determined by a two-tailed Mann-Whitney U test]).

FIG 3

One-step growth curves and RNA synthesis of each mutant and the WT. BHK-21 and IBRS-2 cells were infected with mutant or WT virus at an MOI of 10. The resulting virus was harvested at different times, titers were determined and expressed as a TCID₅₀ dose (A to D), and the genome copy numbers were measured (E to H) for the same samples by real-time RT-PCR. The mean values ± standard deviations are shown (n = 3) (no significant differences were identified by repeated-measures ANOVA).

FIG 4

Mutagen resistance of each mutant. BHK-21 cells (2.5 × 10⁶ cells in 6-well plates) were pretreated with ribavirin for 2 h prior to infection with each mutant or WT virus at an MOI of 0.01 (A to E) or 1 (F to H), as previously described (51). The cells were incubated with ribavirin at the indicated concentrations until all cells demonstrated CPEs, at which time the virus was collected and titers were determined by using BHK-21 monolayers. The mean titers ± standard deviations are shown (n = 3) (*, P < 0.05; **, P < 0.01 [determined by Student's t test]).

FIG 5

Competition assays comparing the relative fitnesses of each mutant and the WT virus. (A and B) Indirect assays. Each virus was mixed at a 1:1 ratio with the marked competitor, which contained the PstI restriction site, and inoculated in triplicate into BHK-21 (A) or IBRS-2 (B) cells at an MOI of 0.1. The progeny RNA was RT-PCR amplified, and restriction fragment length polymorphism assays were performed to determine the abundance of each competitor. Fitness is represented as the output-to-input ratio of the FMDV 3Dpol R84H variant to the marked competitor. A fitness value of <1 indicates that the mutant or the WT is less fit than the marked competitor. The mean values ± standard deviations are shown (n = 3) (*, P < 0.05; **, P < 0.01 [determined by Student's t test]). (C) Direct assays. The D5N, A38V, or DAMM mutant was mixed with the WT at a ratio of 9:1, 1:1, or 1:9 and inoculated into BHK-21 or IBRS-2 cells at an MOI of 0.1 for 3 passages, after which the polymerase gene flanking position 5 or 38 was sequenced. The abundance of each competitor was measured as the height of the nucleotide encoding either the mutant (A or T nucleotide) or the WT (G or C nucleotide) in the sequencing chromatograms.

FIG 6

Virulence of rescued mutants in suckling mice. A total of 19 groups of 3-day-old BALB/c suckling mice were inoculated cervicodorsally with 100 μl of each mutant or WT FMDV diluted to 0.01 TCID₅₀ (A), 0.1 TCID₅₀ (B and C), 1 TCID₅₀ (D), or 10 TCID₅₀ (E). Animal deaths were scored for up to 7 days after inoculation, and survivors were euthanized (n = 6 to 9 for each group).

FIG 7

Correlation between RdRp fidelity and viral virulence. Each rescued variant exhibited different increases in replication fidelity (red line) and different reductions in virulence compared to the WT virus (blue line).

FIG 8

Viral titer and RNA synthesis for each mutant or the WT. (A to C) Viral titer. Primary bovine cells, bovine kidney cells, or bovine thymus cells in 96-well plates were continuously inoculated with diluted mutant viruses or WT FMDV at 50 μl/well. After CPEs were observed, the viral titer was determined as the TCID₅₀ per ml. The mean values ± standard deviations are shown (n = 3) (**, P < 0.01 by Student's t test). (D and E) RNA synthesis. Primary bovine testis cells, primary bovine kidney cells, or bovine thymus cells in 12-well plates were incubated with each mutant or WT FMDV at an MOI of 0.1 for 1 h, washed 3 times, and supplemented with DMEM. The infected cell cultures were harvested at specific time points. Cellular RNA was extracted, and real-time PCR amplification was performed. The mean values ± standard deviations are shown (n = 3) (***, P < 0.001 by repeated-measures ANOVA).

FIG 9

Schematic diagram of viral quasispecies diversity affecting viral virulence. Suckling mice were infected with WT virus or the R84H, A38V, or DAMM mutant virus (from bottom to top). Compared with the WT virus, the mutants had different degrees of enhanced replication fidelity, resulting in different restrictions in genetic diversity. Although all mice received the same viral titers of FMDV, only those infected with diverse quasispecies, such as the WT or R84H populations, developed the disease. Subpopulations within the diverse quasispecies cooperated with each other to facilitate their transmission in mice.