T-705 (favipiravir) induces lethal mutagenesis in influenza A H1N1 viruses in vitro

Abstract

Several novel anti-influenza compounds are in various phases of clinical development. One of these, T-705 (favipiravir), has a mechanism of action that is not fully understood but is suggested to target influenza virus RNA-dependent RNA polymerase. We investigated the mechanism of T-705 activity against influenza A (H1N1) viruses by applying selective drug pressure over multiple sequential passages in MDCK cells. We found that T-705 treatment did not select specific mutations in potential target proteins, including PB1, PB2, PA, and NP. Phenotypic assays based on cell viability confirmed that no T-705-resistant variants were selected. In the presence of T-705, titers of infectious virus decreased significantly (P < 0.0001) during serial passage in MDCK cells inoculated with seasonal influenza A (H1N1) viruses at a low multiplicity of infection (MOI; 0.0001 PFU/cell) or with 2009 pandemic H1N1 viruses at a high MOI (10 PFU/cell). There was no corresponding decrease in the number of viral RNA copies; therefore, specific virus infectivity (the ratio of infectious virus yield to viral RNA copy number) was reduced. Sequence analysis showed enrichment of G→A and C→T transversion mutations, increased mutation frequency, and a shift of the nucleotide profiles of individual NP gene clones under drug selection pressure. Our results demonstrate that T-705 induces a high rate of mutation that generates a nonviable viral phenotype and that lethal mutagenesis is a key antiviral mechanism of T-705. Our findings also explain the broad spectrum of activity of T-705 against viruses of multiple families.

Fig 1

Infectivity of influenza A (H1N1) viruses during serial passage with T-705 in MDCK cells. Cells were inoculated with a low dose (MOI of 0.0001 PFU/cell) of influenza A/Brisbane/59/2007 (A) and A/New Jersey/15/2007 (B) viruses and a high dose (starting MOI of 10 PFU/cell) of A/Denmark/524/2009 (C) and A/Denmark/528/2009 (D) viruses. The T-705 concentration at each passage is indicated by the blue curve. After each passage, the infectivity (log₁₀ PFU/ml) of progeny viruses was measured by plaque assay in MDCK cells. Values are the means ± standard deviations (SD) of two independent experiments. The dotted line indicates the assay's limit of detection (1.0 log₁₀ PFU/ml). Virus titers in supernatants were compared by analysis of variance (ANOVA). *, P < 0.01; **, P < 0.0001.

Fig 2

Antiviral activity and toxicity of T-705 in vitro. A549 cells were transfected with the reporter pPolI-358Luc and infected with influenza A (H1N1) viruses (MOI of 2 PFU/cell) in the presence of increasing concentrations of T-705 (0.1 to 1,000 μM). At 24 h p.i., the results were assessed using the Bright-Glo luciferase reagent (Promega Corp., Madison, WI), following the manufacturer's instructions. The mean value of the negative controls in each plate was set at 100% luminescence, and the percentage of luminescence of each compound-containing well was determined in relation to it. The 50% inhibitory concentration (IC₅₀) was determined by using the 4-parameter logistic nonlinear regression model equation in GraphPad Prism 5 software. Cell viability (CC₅₀) was determined independently in mock-inoculated cells after 24 h of incubation using the CellTiterGlow kit, following the manufacturer's instructions. Values are means ± SD of three replicates.

Fig 3

Plaque morphology of influenza A (H1N1) viruses during serial passage with T-705 in MDCK cells. Panels show plaques from cultures inoculated with low (A, C, E, and G) and high (B, D, F, and H) infectious doses of influenza virus: (A) A/Brisbane/59/2007, mock treated; (B) A/Denmark/524/2009, mock treated; (C) A/Brisbane/59/2007, T-705 treated; (D) A/Denmark/524/2009, T-705 treated; (E) A/New Jersey/15/2007, mock treated; (F) A/Denmark/528/2009, mock treated; (G) A/New Jersey/15/2007, T-705 treated; (H) A/Denmark/528/2009, T-705 treated. After the indicated serial passage, MDCK cells were inoculated with supernatant containing the specified virus and overlaid with 0.45% immunodiffusion-grade agarose (MP Biomedicals, Solon, OH) containing 1 μg/ml TPCK-treated trypsin. After 3 days of incubation at 37°C and 5% CO₂, the cells were stained with 1% crystal violet in 10% formaldehyde.

Fig 4

Mutation frequency and profile of nucleotide changes in influenza A/Denmark/524/2009 (H1N1) virus. (A) MDCK cells were mock treated or treated with T-705 (10 μM) or zanamivir (10 μM) and inoculated with influenza A/Denmark/524/2009 (H1N1) virus at an MOI of 0.001 PFU/cell. Virus was extracted from the supernatant after a 48-h incubation at 37°C. A 1,523-kb region of the NP gene was RT-PCR amplified and subcloned. Individual clones were sequenced to obtain the observed mutation frequencies. The numbers of mutations per 10,000 nucleotides differed significantly between T-705-treated and mock-treated viruses (P < 0.001). (B) Mutation profiles of influenza A/Denmark/524/2009 (H1N1) viruses grown in the presence of mock-treated, T-705-treated (10 μM), or zanamivir-treated (10 μM) MDCK cells. Bold numerals indicate the most frequent nucleotide changes.