Inhibition of Polyamine Biosynthesis Is a Broad-Spectrum Strategy against RNA Viruses

Abstract

RNA viruses present an extraordinary threat to human health, given their sudden and unpredictable appearance, the potential for rapid spread among the human population, and their ability to evolve resistance to antiviral therapies. The recent emergence of chikungunya virus, Zika virus, and Ebola virus highlights the struggles to contain outbreaks. A significant hurdle is the availability of antivirals to treat the infected or protect at-risk populations. While several compounds show promise in vitro and in vivo, these outbreaks underscore the need to accelerate drug discovery. The replication of several viruses has been described to rely on host polyamines, small and abundant positively charged molecules found in the cell. Here, we describe the antiviral effects of two molecules that alter polyamine levels: difluoromethylornithine (DFMO; also called eflornithine), which is a suicide inhibitor of ornithine decarboxylase 1 (ODC1), and diethylnorspermine (DENSpm), an activator of spermidine/spermine N¹-acetyltransferase (SAT1). We show that reducing polyamine levels has a negative effect on diverse RNA viruses, including several viruses involved in recent outbreaks, in vitro and in vivo These findings highlight the importance of the polyamine biosynthetic pathway to viral replication, as well as its potential as a target in the development of further antivirals or currently available molecules, such as DFMO.

Importance: RNA viruses present a significant hazard to human health, and combatting these viruses requires the exploration of new avenues for targeting viral replication. Polyamines, small positively charged molecules within the cell, have been demonstrated to facilitate infection for a few different viruses. Our study demonstrates that diverse RNA viruses rely on the polyamine pathway for replication and highlights polyamine biosynthesis as a promising drug target.

FIG 1

DFMO limits replication of diverse RNA viruses. (A) Diagram presenting the polyamine biosynthetic pathway with pertinent enzymes and inhibitors presented in this study. (B to D) Vero-E6 cells were treated with 500 μM DFMO or different concentrations as indicated for 4 days prior to infection with Coxsackievirus B3 (CVB3) at an MOI of 0.1. At the time of infection, exogenous polyamines were added as indicated. Titers were determined at 24 hpi. (E) Cell counts (open circles, left axis) and viability (closed circles, right axis) were measured on uninfected Vero-E6 cells treated with escalating doses of DFMO for 5 days. Dashed lines indicate levels corresponding to the results for untreated cells. (F to H) BHK-21 cells were treated as described in the legends to panels B to D and subsequently infected with vesicular stomatitis virus (VSV) at an MOI of 0.1 for 16 h. Polyamines were added at the time of infection. (I and J) BHK-21 cells were infected with Sindbis virus (SINV) at the indicated times (I) and treated with the indicated DFMO doses for 24 h (J) at a multiplicity of infection (MOI) of 0.1. (K) Similar to the description in the legend to panel I, cells were infected with Zika virus (ZIKV) at an MOI of 0.1 after pretreatment with 500 μM DFMO in BHK-21 cells, and viral titers determined over the indicated time course. (L) Cell counts and viability were measured on uninfected BHK-21 cells treated with escalating doses of DFMO for 5 days. Statistical significance was determined using Student's t test (n ≥ 3). *, P < 0.05; **, P < 0.01; ***, P < 0.001. Error bars represent 1 standard error of the mean.

FIG 2

Polyamine depletion restricts the replication of diverse families of viruses. Virus titers were determined following a 4-day 500 μM DFMO treatment and rescue with exogenous polyamines, added at the time of infection. (A) Middle East respiratory syndrome coronavirus (MERS-CoV [MCoV]) at 24 hpi in Vero81 cells; (B) poliovirus (PV) at 24 hpi in HeLa cells; (C) Enterovirus A71 (EV-A71) at 24 hpi in HeLa cells; (D) Yellow fever virus (YFV) at 96 hpi in BHK-21; (E) Japanese encephalitis virus (JEV) at 24 hpi in BHK-21 cells; (F) Dengue virus-1 (DENV1) at 48 hpi in BHK-21 cells; (G) Caribbean isolate of chikungunya virus (CHIKV) at 24 hpi in BHK-21 cells; (H) Rift Valley fever virus (RVFV) at 24 hpi in BHK-21 cells; (I) rabies virus (RABV) at 24 hpi in primary cortical neurons. MERS-CoV, CVB3, JEV, WNV, VSV, and RVFV titers were measured by plaque assay; PV and EV-A71 titers were measured by TCID₅₀; DENV1 and YFV titers were determined by fluorescent focus assay; and RABV infection was determined by quantitation of the signal intensity of somatic inclusions. Statistical significance versus the results for the untreated control was determined using one-tailed Student's t test; n = 3 replicates for all viruses except DENV (n = 8), ZIKV (n = 9), MERS (n = 6), and RABV (n = 6). *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001. Bars and error bars represent mean results ± 1 standard deviation.

FIG 3

DENSpm inhibits RNA virus replication. (A and B) Time course of infection of Vero-E6 cells treated with 10 μM DENSpm for 16 h prior to infection with CVB3 (A) or VSV (B) at an MOI of 0.1 or 0.01, respectively. Titers were determined at the times indicated. (C to E) Vero-E6 cells were pretreated with escalating doses of DENSpm for 16 h prior to infection with CVB3 at an MOI of 0.1 (C), VSV at an MOI of 0.01 (D), or CHIKV at an MOI of 0.1 (E). Titers were determined at 24 hpi (40 h posttreatment). (F) Cell counts (open circles, left axis) and viability (closed circles, right axis) were measured for uninfected Vero-E6 cells treated with escalating doses of DENSpm for 40 h. Dashed lines indicate levels corresponding to the results for untreated cells. (G) Western blot analysis of SAT1 and GAPDH (top two panels) and thin-layer chromatographic analysis of polyamines (bottom panel; polyamines labeled for spermidine [Spd], spermine [Spm], and putrescine [Put]) were performed on uninfected samples run, measured at 40 h posttreatment. Statistical significance was determined using Student's t test (n ≥ 3). *, P < 0.05; **, P < 0.01; ***, P < 0.001. Error bars represent 1 standard error of the mean.

FIG 4

Polyamine depletion blocks RNA virus replication in combination with IFN-β. (A to C) BHK-21 cells were not treated or pretreated with 500 μM DFMO for 4 days and with 1,000 U IFN-β for 4 h, as indicated, prior to infection with CVB3 at an MOI of 0.1 (A), VSV at an MOI of 0.01 (B), and CHIKV at an MOI of 0.1 (C). Titers were determined at 24 hpi. (D to F) Vero-E6 cells were not treated or treated with 10 μM DENSpm for 16 h and 1,000 U IFN-β for 4 h, as indicated, prior to infection with CVB3 (D), VSV (E), and CHIKV (F). Statistical significance was determined using Student's t test (n ≥ 3). *, P < 0.05; **, P < 0.01; ***, P < 0.001. Error bars represent 1 standard error of the mean.

FIG 5

Polyamine depletion limits viral replication in a time-dependent manner. BHK-21 cells were treated with 10 μM DENSpm at the indicated times prior to and after infection with CVB3 at an MOI of 0.1 (A), VSV at an MOI of 0.01 (B), and CHIKV at an MOI of 0.1 (C) for 24 h. (D) Cell counts (open circles, left axis) and viability (closed circles, right axis) were measured on uninfected Vero-E6 cells treated with DENSpm for different times as described in the legend to panels A to C. Dashed lines indicate levels corresponding to the results for untreated cells. (E to I) Confluent monolayers of cells were treated with DFMO and a polyamine mixture (PA) at the time of infection with CVB3 (E), VSV (F), CHIKV (G), ZIKV (H), and SINV (I) in a plaque-reduction assay. Plaques were developed after 4 days (H), 3 days (E, G, and I), or 2 days (F). (J) Representative plaques from CVB3, CHIKV, and ZIKV, treated as indicated. Statistical significance was determined using Student's t test (n ≥ 3). *, P < 0.05; **, P < 0.01; ***, P < 0.001. Error bars represent 1 standard error of the mean.

FIG 6

Polyamine depletion limits viral replication in vivo. (A to F) Mice were provided 1% DFMO in drinking water ad libitum for 7 days prior to infection with 1,000 TCID₅₀ of CVB3 intraperitoneally. At 3 dpi, viral titers in the spleen (A), pancreas (B), heart (C), lung (D), liver (E), and kidney (F) were determined by plaque assay. (G to I) As described in the legend to panels A to F, mice were provided DFMO for 7 days prior to infection by footpad injection of 10⁵ PFU of CHIKV. At 24 hpi, organs were harvested and titers in the foot (G), muscle (H), and serum (I) were determined by plaque assay. Statistical significance was determined using Student's t test. *, P < 0.05; **, P < 0.01. Symbols show the results for individual mice, horizontal lines represent the mean values for the groups, and error bars represent 1 standard error of the mean.