Mechanism of Action of Methotrexate Against Zika Virus

Abstract

Zika virus (ZIKV), which is associated with microcephaly in infants and Guillain-Barré syndrome, reemerged as a serious public health threat in Latin America in recent years. Previous high-throughput screening (HTS) campaigns have revealed several potential hit molecules against ZIKV, including methotrexate (MTX), which is clinically used as an anti-cancer chemotherapy and anti-rheumatoid agent. We studied the mechanism of action of MTX against ZIKV in relation to its inhibition of dihydrofolate reductase (DHFR) in vitro using Vero and human neural stem cells (hNSCs). As expected, an antiviral effect for MTX against ZIKV was observed, showing up to 10-fold decrease in virus titer during MTX treatment. We also observed that addition of leucovorin (a downstream metabolite of DHFR pathway) rescued the ZIKV replication impaired by MTX treatment in ZIKV-infected cells, explaining the antiviral effect of MTX through inhibition of DHFR. We also found that addition of adenosine to ZIKV-infected cells was able to rescue ZIKV replication inhibited by MTX, suggesting that restriction of de novo synthesis adenosine triphosphate (ATP) pools suppresses viral replication. These results confirm that the DHFR pathway can be targeted to inhibit replication of ZIKV, similar to other published results showing this effect in related flaviviruses.

Keywords: Zika virus; antivirals; dihydrofolate reductase; methotrexate; nucleotide metabolism.

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1

Inhibition of Zika virus (ZIKV replication in Vero Cells after methotrexate (MTX) treatment. ZIKV-infected Vero cells were treated with 5 µM MTX or 0.5% of DMSO as a negative control. (A) Immunofluorescence images (20X) of ZIKV-infected (H/PAN/2016/BEI-259634, NR-50210 strain (H/PAN), multiplicity of infection (MOI) 0.2) Vero cells were acquired to analyze the level of ZIKV-envelope protein after 5 µM MTX treatment at 48 h post-infection (PI). SYTOX green was used to stain nuclei of Vero cells. Scale bars represents 5 µm. The virus titer of two ZIKV strains, (B) H/PAN MOI 0.2 and (C) PRVABC59, NR-50240 strain ZIKV (PRV) MOI 0.2, were measured after 5 µM MTX treatment at three different time points, 1 h PI, 48xh PI, and 96 h PI. (D) ZIKV titer after MTX treatment in dose-response manner (8-fold, 50 µM, 6.25 µM, 0.781 µM, and 0.0977 µM) from ZIKV-infected (H/PAN MOI 0.2) Vero cells at 48 h PI. (E) Average cytoplasm fluorescence signal of ZIKV-envelope protein per nuclei from ZIKV-infected (H/PAN MOI 0.2) Vero cells with 5 µM MTX treatment was measured at 48 h PI. RFU = relative fluorescence units. (F) Immunofluorescence images (10×) of ZIKV-infected (H/PAN MOI 0.2) Vero cells with three different MTX concentrations (6-fold dilution series, 25.0 µM, 1.56 µM, 0.098 µM) were acquired to observe the reduction in ZIKV-envelope protein’s fluorescence signal at 48 h PI. The scale bar represents 50 µm. For (E) and (F), DAPI was used to stain nuclei of Vero cells. At least two independent replicates were performed. For (B) and (C), two-way ANOVA, followed by Sidak’s multiple comparisons test, were used for statistical analysis. For (D), one-way ANOVA, followed by Tukey’s multiple comparisons test, were used for statistical analysis; average viral titer at each MTX concentration was compared to that of untreated ZIKV at 48 h PI as a control. For (E), Student’s t-test was used for statistical analysis (two-tailed distribution, heteroscedastic). Error bars represent standard error of the mean (SEM). ** p ≤ 0.01, *** p ≤ 0.001, **** p ≤ 0.0001, n.s. not significant.

Figure 2

Determination of cytotoxicity of MTX. To determine the cytotoxicity of MTX, the number of Vero cells after MTX treatment was directly counted by trypan-blue staining. Then, the number of live cells were counted under a light microscope with a hemocytometer. (A) 10× bright-field images of Vero cells with 0.5 µM MTX treatment with initial cell density of 10,000 cells/100 µL. Scale bars represent 5 µm. (B) The cell density of Vero cells after 0.5 µM MTX treatment after incubation for 48 h at 37 °C and 5% CO2 (NuAir, Plymouth, MN, USA). (C) Cell viability assay using CTF reagent with initial cell density of 30,000 cells/100µL. Two MTX concentrations (5 µM and 0.5 µM) were tested. At least two independent replicates were performed. One-way ANOVA, followed by Tukey’s multiple comparisons test, were used for statistical analysis. The error bars represent the standard error of the mean (SEM). * p ≤ 0.05.

Figure 3

Rescue effect of leucovorin on the cell viability and ZIKV replication during MTX treatment. Two host cell lines, Vero and human neural stem cells (hNSCs), were used to examine the mechanism of action of MTX against ZIKV replication through the DHFR pathway. Virus titers of the two ZIKV strains, H/PAN MOI 0.2 (A) and PRV MOI 0.2 (B) on Vero cells were measured by standard plaque assay. With H/PAN infection, the ZIKV titer of the DMSO control was not significantly different from that of 5 µM MTX-treated sample in Vero cells (p = 0.11) (C) Virus titer of H/PAN MOI 0.1 on hNSCs was measured by standard plaque assay. (D) The cytotoxicity of 5 µM MTX and co-treatment of MTX with 50 µM folic acid or leucovorin on Vero cells was studied by CTG reagent. (E) The cytotoxicity of 5 µM MTX and co-treatment of MTX with 50 µM folic acid or leucovorin on hNSCs was cells was studied by CTG reagent. (F,G) 50 µM Leucovorin and water, as a vehicle control, were added to Vero cells and hNSCs to determine their effect on host cell viability using CTG reagent. RLU = relative luminescence units. At least two independent replicates were performed. One-way ANOVA, followed by Tukey’s multiple comparisons test, were used for statistical analysis. The error bars represent the standard error of the mean (SEM). ** p ≤ 0.01, *** p ≤ 0.001, **** p ≤ 0.0001, n.s. not significant.

Figure 4

Rescue effect of GAT medium on cell viability and ZIKV replication during MTX treatment. (A) Cell viability of Vero cells was studied by CTG reagent after MTX treatment with or without GAT medium. (B) GAT medium rescued the ZIKV replication from MTX treatment of the ZIKV-infected (H/PAN MOI 0.2) Vero cells. (C) Adenosine alone can save ZIKV replication during MTX treatment of the ZIKV-infected (H/PAN MOI 0.2) Vero cells. (D) Exogenous adenosine rescued cellular ATP levels, but thymidine and glycine could not rescue ATP levels during MTX treatment in Vero cells as measured by CTG assay. (E) Adenosine also rescued live-protease activity measured by CTF reagent during MTX treatment in Vero cells. (F,G) GAT media and adenosine were added to Vero cells and hNSCs to study determine their effect on host cell viability using CTG and CTF reagents. RLU = relative luminescence units. At least two independent replicates were performed. The error bars represent the standard error of the mean (SEM). One-way ANOVA, followed by Tukey’s multiple comparisons test, was used for statistical analysis. * p ≤ 0.05, **** p ≤ 0.0001, n.s. not significant.