Counteracting quasispecies adaptability: extinction of a ribavirin-resistant virus mutant by an alternative mutagenic treatment

Abstract

Background: Lethal mutagenesis, or virus extinction promoted by mutagen-induced elevation of mutation rates of viruses, may meet with the problem of selection of mutagen-resistant variants, as extensively documented for standard, non-mutagenic antiviral inhibitors. Previously, we characterized a mutant of foot-and-mouth disease virus that included in its RNA-dependent RNA polymerase replacement M296I that decreased the sensitivity of the virus to the mutagenic nucleoside analogue ribavirin.

Methodology and principal findings: Replacement M296I in the viral polymerase impedes the extinction of the mutant foot-and-mouth disease virus by elevated concentrations of ribavirin. In contrast, wild type virus was extinguished by the same ribavirin treatment and, interestingly, no mutants resistant to ribavirin were selected from the wild type populations. Decreases of infectivity and viral load of the ribavirin-resistant M296I mutant were attained with a combination of the mutagen 5-fluorouracil and the non-mutagenic inhibitor guanidine hydrocloride. However, extinction was achieved with a sequential treatment, first with ribavirin, and then with a minimal dose of 5-fluorouracil in combination with guanidine hydrochloride. Both, wild type and ribavirin-resistant mutant M296I exhibited equal sensitivity to this combination, indicating that replacement M296I in the polymerase did not confer a significant cross-resistance to 5-fluorouracil. We discuss these results in relation to antiviral designs based on lethal mutagenesis.

Conclusions: (i) When dominant in the population, a mutation that confers partial resistance to a mutagenic agent can jeopardize virus extinction by elevated doses of the same mutagen. (ii) A wild type virus, subjected to identical high mutagenic treatment, need not select a mutagen-resistant variant, and the population can be extinguished. (iii) Extinction of the mutagen-resistant variant can be achieved by a sequential treatment of a high dose of the same mutagen, followed by a combination of another mutagen with an antiviral inhibitor.

Figure 1. Absence of extinction of FMDV M296I after treatment with ribavirin.

BHK-21 cells (2×10⁶ cells) were initially infected with the indicated viruses (WT, wild type FMDV; M296I, FMDV with substitution M296I in 3D) at a multiplicity of infection (MOI) of about 0.3 PFU/cell. In succesive passages, the same number of cells were infected with the supernatant from the previous passage. Viruses and the absence or presence of ribavirin (R) (5 mM) are indicated in the boxes at the top of panels A and B; (A) Infectivity of FMDV WT in the cell culture supernatants. Values are the mean±SD (error bars) of triplicate determinations. The discontinuous line indicates the limit of detection of FMDV infectivity. (B) Infectivity, viral RNA levels and specific infectivity in the supernatants of BHK-21 cells infected with FMDV M296I. Titrations and quantifications of FMDV RNA were carried out in triplicate; standard deviations are given. Viral RNA was extracted and quantified by real time PCR. The discontinuous lines indicate the limit of detection of infectivity and viral RNA. Values of specific infectivity (PFUs/RNA molecules) were calculated from the infectivities and viral RNA levels given in the two previous panels of (B). Procedures for titration of virus infectivity and quantification of FMDV RNA, as well as the positive and negative controls included in the assays, are described in Materials and Methods.

Figure 2. Response of FMDV wild type and mutant M296I to 5-fluorouracil-guanidine combination treatment.

FMDV wild type (WT) or mutant M296I were serially passaged in the absence or presence of R (5 mM), as described in the legend for Fig. 1, and as indicated in the boxes at the top. At passages 2, 5, 8 and 10, the viral populations were subjected to additional passages in the presence of a combination of 5-fluorouracil (FU, 200 µg/ml) and guanidinium hydrochloride (GU, 4 mM) (combination FUG-200); p indicates passage number, either in the absence of any drug (panels A and B), or in the presence of 5 mM R (panel C), prior to the FUG-200 treatment. Virus titers and amounts of FMDV RNA in the cell culture supernatants were determined at each passage, as indicated in each panel. Discontinuous lines indicate the limit of detection of infectivity or FMDV RNA. Titrations and quantifications of FMDV were carried out in triplicate; standard deviations are given. Viral RNA was amplified by RT-PCR as a test for virus extinction, as described in Materials and Methods [bottom panels: M, molecular size markers (Hind III-digested Ø29 DNA; fragment size in base pairs is indicated on the right); C−, negative control, amplification without RNA; C+, positive RT-PCR amplification control, M, C− and C+ were run for each analysis in A, B, C, but included here in A for simplicity]. Procedures for drug treatments, titration of infectivity, quantification of FMDV RNA by real time PCR, and for RT-PCR amplification are detailed in Materials and Methods.

Figure 3. Response of FMDV wild type and mutant M296I to 5-fluorouracil-guanidine treatment, with increased mutagen dose.

The experimental design, symbols, drugs, procedures and controls are the same as described in the legend for Fig. 2, except that the FU concentration used was 500 µg/ml FU [FU (500 µg/ml) and GU (4 mM), abbreviated as FUG-500 in the boxes on the top]. Note the absence of infectivity and of an RT-PCR amplification band upon replication of M296I first in the presence of R (5 mM) (2 to 8 passages, but not 10 passages), followed by treatment with FU-500. Extinction of M296I (panel C) was further ascertained by the absence of infectivity or RT-PCR amplifiable viral materials, after three passages of the cells that harbored the putatively extinguished FMDV, in the absence of any drug, as justified in Materials and Methods. Titrations and quantifications of FMDV were carried out in triplicate; standard deviations are given.

Figure 4. Response of FMDV wild type and mutant M296I to 5-fluorouracil-guanidine combination treatment.

Unpassaged FMDV wild type (WT) or mutant M296I was adjusted to a number of PFU/ml comprised between 5.5×10⁵ and 1.5×10² which correspond to the maximum titer (passage 10 in Fig. 3C) and the minimum titer (passage 2 in Fig. 3C) attained by M296I in the presence of R. Subsequently, populations were subjected to passages in the presence of 5-fluorouracil (FU, 500 µg/ml) and guanidinium hydrochloride (GU, 4 mM) (combination FUG-500). The virus (WT, panel A; M296I, panel B) and drug treatment are indicated in the boxes at the top; ‘p2’, ‘p5’, ‘p8’ and ‘p10’ indicate that the number of initial PFU/ml of FMDV WT and M296I was adjusted to the same titer attained by M296I in the presence of R at passage 2, 5, 8 and 10, respectively (data of Fig. 3C). Virus titers and amounts of FMDV RNA in the cell culture supernatants were determined at each passage, as indicated in each panel; titrations and quantifications of FMDV were carried out in triplicate; standard deviations (not given) never exceeded 30% of the mean. Viral RNA was amplified by RT-PCR as a test for virus extinction, as described in Materials and Methods [bottom panels: M, molecular size markers (Hind III-digested Ø29 DNA; fragment size in base pairs is indicated on the right); C−, negative control, amplification without RNA; C+, positive RT-PCR amplification control, M, C− and C+ were run for each analysis in A and, B, but included here only in A for simplicity]. Procedures for drug treatments, titration of infectivity, quantification of FMDV RNA by real time PCR, and for RT-PCR amplification are detailed in Materials and Methods.