Pharmacological Elevation of Cellular Dihydrosphingomyelin Provides a Novel Antiviral Strategy against West Nile Virus Infection

Abstract

The flavivirus life cycle is strictly dependent on cellular lipid metabolism. Polyphenols like gallic acid and its derivatives are promising lead compounds for new therapeutic agents as they can exert multiple pharmacological activities, including the alteration of lipid metabolism. The evaluation of our collection of polyphenols against West Nile virus (WNV), a representative medically relevant flavivirus, led to the identification of N,N'-(dodecane-1,12-diyl)bis(3,4,5-trihydroxybenzamide) and its 2,3,4-trihydroxybenzamide regioisomer as selective antivirals with low cytotoxicity and high antiviral activity (half-maximal effective concentrations [EC₅₀s] of 2.2 and 0.24 μM, respectively, in Vero cells; EC₅₀s of 2.2 and 1.9 μM, respectively, in SH-SY5Y cells). These polyphenols also inhibited the multiplication of other flaviviruses, namely, Usutu, dengue, and Zika viruses, exhibiting lower antiviral or negligible antiviral activity against other RNA viruses. The mechanism underlying their antiviral activity against WNV involved the alteration of sphingolipid metabolism. These compounds inhibited ceramide desaturase (Des1), promoting the accumulation of dihydrosphingomyelin (dhSM), a minor component of cellular sphingolipids with important roles in membrane properties. The addition of exogenous dhSM or Des1 blockage by using the reference inhibitor GT-11 {N-[(1R,2S)-2-hydroxy-1-hydroxymethyl-2-(2-tridecyl-1-cyclopropenyl)ethyl]octanamide} confirmed the involvement of this pathway in WNV infection. These results unveil the potential of novel antiviral strategies based on the modulation of the cellular levels of dhSM and Des1 activity for the control of flavivirus infection.

Keywords: West Nile virus; antiviral; antiviral agents; flavivirus; polyphenol; polyphenols; sphingolipid.

FIG 1

Antiviral activity of synthetic polyphenols against WNV in Vero cells. (A) Compounds used in this study. (B) Antiviral activity against WNV and cytotoxicity of polyphenols. The virus yield in the supernatant of infected Vero cells (MOI of 1 PFU/cell) was determined at 24 h p.i. Polyphenols were added 1 h prior to infection and maintained throughout the rest of the assay. The cytotoxicity of the compounds was measured by the determination of the cellular ATP concentration in uninfected samples. The dashed lines denote a 50% reduction. (C) Effect of polyphenols on the proliferation of Vero cells. Vero cells plated at a low density (<50% confluence) were incubated for 24 h in the presence of the compounds, and the cell number was determined. The dashed lines denote a 50% reduction. (D) Effect of the MOI on the antiviral activity of polyphenols. Vero cells were treated and infected at different MOIs as described above for panel A. Each dot denotes a single biological replicate. Data are expressed as means ± SD (n = 2 to 4).

FIG 2

Antiviral activity of synthetic polyphenols against WNV in SH-SY5Y cells. The virus yield in the supernatant of infected SH-SY5Y cells (MOI of 1 PFU/cell) was determined at 24 h p.i. Polyphenols were added 1 h prior to infection and maintained throughout the rest of the assay. The cytotoxicity of the compounds was measured by the determination of the cellular ATP concentration in uninfected samples. The dashed lines denote a 50% reduction. Data are expressed as means ± SD (n = 2 to 4).

FIG 3

Antiviral spectrum of AL-088 and AL-274. The virus yield in the supernatant of infected Vero cells (MOI of 1 PFU/cell) was determined at 24 h p.i. for USUV, ZIKV, VSV, and CVB5 and at 48 h p.i. in the case of DENV-2. Polyphenols were added 1 h prior to infection and maintained throughout the rest of the assay. The cytotoxicity of the compounds was measured by the determination of the cellular ATP concentration in uninfected samples at 48 h for DENV-2 and 24 h for the rest of the viruses. The dashed lines denote a 50% reduction. Data are expressed as means ± SD (n = 2 to 4).

FIG 4

AL-088 and AL-274 inhibit WNV multiplication and infectivity. (A and B) Synthetic polyphenols inhibit the entry of WNV RVPs. Vero cells were treated with AL-088 (A) or AL-274 (B) for 24 h and then infected with RVPs. The number of cells infected with RVPs was determined at 48 h p.i. by flow cytometry (n = 3). (C and D) Synthetic polyphenols inhibit WNV multiplication when added at postentry steps. Vero cells were infected with WNV (MOI of 1 PFU/cell), and AL-088 (C) or AL-274 (D) was added at 1 or 3 h p.i. The virus yield was determined at 24 h p.i. (n = 6). (E) Gallic acid derivatives inhibit the accumulation of dsRNA intermediates and viral proteins. Vero cells were infected with WNV, treated with the polyphenols at 3 h p.i., fixed, and processed for immunofluorescence to detect dsRNA or WNV E at 24 h p.i. Viral antigens are displayed in green, and cell nuclei are shown in blue. (F and G) Analysis of the specific infectivity of the viral progeny released from cells treated with AL-088 (F) or AL-274 (G) as described above for panels A and B (n = 3 to 4). For panels A to D, F, and G, data are expressed as means ± SD. Each dot denotes a single biological replicate. **, P < 0.01; ***, P < 0.001 (by ANOVA and Student’s t test using Bonferroni’s correction).

FIG 5

AL-088 and AL-274 elevate dihydrosphingomyelin levels. (A) Comparison of the lipidomes of Vero cells treated with 10 μM polyphenols (24 h) or the vehicle by the PCA method. Each dot denotes a single biological replicate. The 95% confidence regions for each group of samples are colored (n = 6). (B) Relative abundances of ceramide (Cer), dihydroceramide (dhCer), sphingomyelin (SM), dihydrosphingomyelin (dhSMs), hexosylceramide (HexCer), lactosylceramide (LacCer), phosphatidylcholine (PC), lysophosphatidylcholine (LPC), phosphatidylethanolamine (PE), diacylglycerol (DAG), triacylglycerol (TAG), cholesterol (CHOL), and cholesteryl ester (CE) in samples treated with AL-088, AL-274, or the vehicle. Each symbol denotes a single biological replicate (n = 6). **, P < 0.01; ***, P < 0.001 (by multiple t tests applying Sidak-Bonferroni correction). (C) Volcano plot displaying the lipid species significantly altered in cells treated with AL-088 and AL-274. Discontinuous lines indicate an FDR-adjusted P value of 0.5 and a log₂ fold change of 2. Each point corresponds to the mean value obtained for a single lipid species (n = 6). (D) Heat map displaying the relative abundances of dhSM species in samples treated with the vehicle, AL-088, or AL-274. Each column denotes a single biological replicate (n = 6). (E) Box-and-whisker plots showing the dhSM/SM ratios for each molecular species analyzed. Each symbol denotes a single biological replicate (n = 6). **, P < 0.01; ***, P < 0.001 (by multiple t tests applying Sidak-Bonferroni correction).

FIG 6

AL-088 and AL-274 inhibit ceramide desaturase activity. (A) Scheme of dihydrosphingolipid synthesis. Ceramide desaturase (Des1) is highlighted in blue. (B) Effect of AL-088 and AL-288 on the conversion of dhCerC6NBD to CerC6NBD in intact T98G cell cultures. Cells were treated with 10 μM each compound. Data are expressed as means ± SD. Each symbol denotes a single biological replicate (n = 3). ***, P < 0.001 (by ANOVA and Student’s t test using Bonferroni’s correction). (C) Dose-dependent inhibition of Des1 activity in T98G cell lysates by AL-088 and AL-274. Des1 activity was measured as described above for panel B. Data are expressed as means ± SD (n = 3).

FIG 7

Dihydrosphingomyelin inhibits WNV infection. (A) Antiviral activity of dhSM against WNV. Vero cells were loaded with dhSM for 24 h and infected with WNV (MOI of 1 PFU/cell), and the virus yield in the supernatant was determined at 24 h p.i. The cytotoxicity of the compounds was measured by the determination of the cellular ATP concentration in uninfected samples. Data are expressed as means ± SD (n = 5). Each dot denotes a single biological replicate. **, P < 0.01; ***, P < 0.001 (by ANOVA and Student’s t test using Bonferroni’s correction). (B) Antiviral activity of the Des1 inhibitor GT-11 against WNV. Vero cells were treated with GT-11 and infected as described above for panel A. Data are expressed as means ± SD (n = 4 to 5). Each dot denotes a single biological replicate. **, P < 0.01; ***, P < 0.001 (by ANOVA and Student’s t test using Bonferroni’s correction).