A drug repurposing screen identifies decitabine as an HSV-1 antiviral

Abstract

Herpes simplex virus type 1 (HSV-1) is a highly prevalent human pathogen that causes a range of clinical manifestations, including oral and genital herpes, keratitis, encephalitis, and disseminated neonatal disease. Despite its significant health and economic burden, there is currently only a handful of approved antiviral drugs to treat HSV-1 infection. Acyclovir and its analogs are the first-line treatment, but resistance often arises during prolonged treatment periods, such as in immunocompromised patients. Therefore, there is a critical need to identify novel antiviral agents against HSV-1. Here, we performed a drug repurposing screen, testing the ability of 1,900 safe-in-human drugs to inhibit HSV-1 infection in vitro. The screen identified decitabine, a cytidine analog that is used to treat myelodysplastic syndromes and acute myeloid leukemia, as a potent anti-HSV-1 agent. We show that decitabine is effective in inhibiting HSV-1 infection in multiple cell types, including human keratinocytes, that it synergizes with acyclovir, and acyclovir-resistant HSV-1 is still sensitive to decitabine. We further show that decitabine causes G > C and C > G transversions across the viral genome, suggesting it exerts its antiviral activity by lethal mutagenesis, although a role for decitabine's known targets, DNA methyl-transferases, has not been ruled out.

Importance: Herpes simplex virus type 1 (HSV-1) is a prevalent human pathogen with a limited arsenal of antiviral agents, resistance to which can often develop during prolonged treatment, such as in the case of immunocompromised individuals. Development of novel antiviral agents is a costly and prolonged process, making new antivirals few and far between. Here, we employed an approach called drug repurposing to investigate the potential anti-HSV-1 activity of drugs that are known to be safe in humans, shortening the process of drug development considerably. We identified a nucleoside analog named decitabine as a potent anti-HSV-1 agent in cell culture and investigated its mechanism of action. Decitabine synergizes with the current anti herpetic acyclovir and increases the rate of mutations in the viral genome. Thus, decitabine is an attractive candidate for future studies in animal models to inform its possible application as a novel HSV-1 therapy.

Keywords: antiviral agents; decitabine; herpes simplex virus; lethal mutagenesis.

Fig 1

A drug repurposing screen for HSV-1 identified decitabine as a potential anti-herpetic. (A) One thousand nine hundred safe-in-human drugs were tested in duplicate for their ability to inhibit HSV-1 infection of A549 cells at 10 µM. Each dot is a single drug. Shown are drugs that did not perturb cell growth (>90%). Drugs are color-coded based on their result in the screen: blue = DMSO control, gray = non-hits (HSV-1 infection reduced by less than 80%), red = hits (HSV-1 infection reduced by 80% or more), and green (uninfected controls). (B) Zoom in on the bottom left corner of (A), showing the hits from the screen.

Fig 2

Decitabine inhibits HSV-1 infection in multiple cell lines. (A) Dose-response analysis of decitabine inhibition of HSV-1 in nTERT (immortalized human keratinocytes, orange), A549 (blue), and Vero (purple). Dots are individual biological repeats (n = 3), and lines are sigmoid curves fitted to the data using Matlab. IC₅₀ values are the decitabine concentration needed to achieve 50% HSV-1 infection inhibition. (B) Vero cells were infected with HSV-1 at an MOI of 0.1 and treated with 0–10 μM of decitabine for 2 days. Supernatants were collected and assayed for viable progeny by the plaque assay. Dots are individual biological repeats (n = 3), and bars represent the mean. (C) Effect of decitabine (green) and acyclovir (purple) on cell viability. Dots are individual biological repeats (n = 3), and bars represent the mean. Decitabine- and acyclovir-treated cells were normalized to DMSO-treated cells.

Fig 3

Decitabine is effective against acyclovir-resistant HSV-1 and synergizes with it. (A and B) Dose-response analysis of acyclovir (A) and decitabine (B) inhibition of parental HSV-1 (P0, black) and acyclovir-resistant HSV-1 [ACY (1–3), orange, green, and blue]. Dots are individual measurements (n = 1), and lines are sigmoid fits using Matlab. Numbers in legend are the IC₅₀ values. (C) Synergy plot of combination therapy of acyclovir and decitabine. The x and y axes are the drug concentrations, and the z-axis denotes the synergy score calculated by SynergyFinder. Plotted is the mean synergy score obtained from three independent biological repeats.

Fig 4

Experimental evolution of HSV-1 in the presence of decitabine. (A) Schematic of the experimental evolution design. (B) Dose-response analysis of the parental (P0, black) and evolved strains (DEC1-3, red, blue, and green). Dots are independent measurements (n = 3), and lines are sigmoid fits using Matlab. Numbers in legend are the IC₅₀ values. (C) A representative image of plaques formed by decitabine-resistant mutants. The large image shows the YFP-ICP4 of an entire well (bar = 100 mm), and zoomed boxes show six individual plaques at the same magnification (bar = 20 mm).

Fig 5

Decitabine-resistant HSV-1 shows high genetic variability. (A) Map of SNPs identified in the HSV-1 genome. X-axis is the genome coordinate, and y-axis is the variant frequency. Dots are individual SNPs (DEC-1 = red, DEC-2 = blue, and DEC-3 = green). (B) Total number of SNPs detected in the decitabine-resistant mutants. (C) Total number of SNPs shared for each of the two mutants. (D) Analysis of detected SNPs by mutation type.

Fig 6

Genetic analysis of plaque purified decitabine-resistant HSV-1. (A) Dose-response analysis of decitabine inhibition of parental HSV-1 (P0, black) and four plaque-purified decitabine-resistant isolates (DEC1.1 = red, DEC1.2 = green, DEC2.1 = blue, and DEC3.1 = purple). Dots are individual biological repeats (n = 3), and lines are sigmoid fits. Numbers in figure legend are IC₅₀ values. (B) Histograms of the SNP frequencies for the four isolates. (C) Position and SNP frequency of the detected SNPs across all four isolates. Dots are individual SNPs, color-coded as in (A). (D) Analysis of detected SNPs by mutation type. (E) Stacked bar plot, showing the number of SNPs per kb detected in each isolate in the 14 viral genes that were common to all isolates.