In Vitro Discovery of a Therapeutic Lead for HFMD From a Library Screen of Rocaglates/Aglains

Abstract

The lack of effective antiviral treatments for enteroviruses, including human enterovirus A71 (EV-A71), have resulted in an immense global healthcare burden associated with hand-foot-and-mouth disease (HFMD). Rocaglates and aglains belong to a family of compounds produced by Aglaia genus plants. Since the initial discovery of rocaglates in 1982, various rocaglates and aglains have been synthesized and extensively studied mainly as anticancer agents. Here, we report the discovery of a novel aglain derivative as a potential EV-A71 inhibitor. From an immunofluorescence-based phenotypic screen of a library of 296 rocaglate and aglain derivatives, we identified a lead aglain which effectively suppressed EV-A71 replication by 2.3 log fold at a non-cytotoxic concentration, with a host cell CC₅₀ of 21.78 µM, an EV-A71 infection EC₅₀ of 3.57 µM, and a selectivity index of 6.1. Further validation revealed inhibition of EV-A71 across multiple human cell types and a pan-enterovirus inhibitory spectrum against other enteroviruses. Subsequent mechanistic investigation revealed interference with EV-A71 intracellular post-entry events including viral RNA transcription and translation. Findings from this study have established a strong foundation for development of aglain scaffolds as much needed antiviral agents for HFMD, paving the way for future medicinal chemistry optimization and in vivo studies.

Keywords: aglain; antiviral agents; enterovirus; hand, foot, and mouth disease; natural products; pentafluorophenyl.

Figure 1

Chemical structures of (A) naturally occurring rocaglates (1 and 2), and (B) synthetic rocaglates (3 and 4); (C) Proposed biosynthetic transformation of aglain ketone precursors 5/ent‐5 to rocaglates such as 6 (atoms and rings are labelled according to the typical convention for rocaglates). Enantioselective total synthesis and bioactivity profiles have so far established that bioactive aglains and bioactive rocaglates are derived from aglain precursors that are enantiomeric to one another. (D) Chemical structures of naturally occurring aglains 8‐12.

Figure 2

Summary of EV‐A71 activity screen. For the cytotoxicity screen, NSC‐34 cells were treated with 10 µM of each compound and incubated for 12 h before cell viability evaluation using the alamarBlue Cell Viability Reagent (Thermo Fisher Scientific, Waltham, MA, USA). Fluorescence intensity in each treatment and control group was measured at 570 nm excitation wavelength and 600 nm emission wavelength using the Infinite 200 series microplate reader (Tecan, Männedorf, Switzerland). For the antiviral screen, NSC‐34 cells were infected with EV‐A71 (MOI 1) and incubated for 1 h, followed by treatment with 10 µM of respective compounds for 12 h within a similar incubation environment. Methanol‐fixed cells were probed with anti‐EV‐A71 VP2 1° antibodies (MAB979; Merck Millipore, Burlington, VT, USA) and anti‐mouse FITC 2° antibodies (Merck Millipore), before being visualized using the Operetta High‐Content Imaging System. Image analyses were performed with the Harmony High‐Content Imaging and Analysis Software (PerkinElmer, Waltham, MA, USA). Resulting images were captured with DAPI and FITC fluorescence filters from a pre‐determined central locus of each well and processed with the Cell Profiler Software, which generates quantitative readings on the total number of cells as well as the infected cell population within each well via signals from the DAPI‐stained nuclei and FITC‐stained EV‐A71 VP2, respectively. (A) Scatter plot of host cell viability (X‐axis) and antiviral activity (Y‐axis) with data points colored by scaffold type. All tested rocaglates with > 2 log reduction were unacceptably toxic to host cells, including exemplar rocaglates 3, 4 and 6. Only one aglain hit (CMLD012723, 13) met the established hit criteria of > 2 log reduction in viral titre with > 90% cell viability. (B) Chemical structure of aglain hit compound 13.

Figure 3

ORTEP of the single crystal X‐ray structure of aglain 13. Thermal ellipsoids are shown at the 50% probability level. Diffraction data were collected on a Bruker D8 Venture, and the ORTEP was generated using Mercury.

Figure 4

(A) Expanded hit criteria and near‐neighbour identification for non‐cytotoxic aglains reveals additional compounds of interest 10, 11 and 14‐17. (B) Chemical structures of compounds 13‐18, with key structural differences from screening hit 13 highlighted in red.

Figure 5

Validation of cytotoxicity and antiviral profiles of selected aglain derivatives. (A) NSC‐34 cells were treated with various concentrations of respective compounds, in the presence or absence of EV‐A71 (MOI 1) infection. Resulting cell viability and virus yield from respective experimental setups were determined via alamarBlue and plaque assays, respectively. EV‐A71 inhibition potency and cytotoxicity of each compound was tabulated whereby EC₅₀ and CC₅₀ values were computed using GraphPad Prism 9 software. dose–response evaluation of cytotoxicity and antiviral properties of compound 13 against EV‐A71 in (B) RD (CC₅₀ = 11.17 µM; EC₅₀ = 0.43 µM; SI = 25.98), (C) SH‐SY5Y (CC₅₀ = 10.99 µM; EC₅₀ = 5.43 µM; SI = 2.02) or (D) primary human brain microvascular endothelial cells (HBMECs) (CC₅₀ = 89.44 µM; EC₅₀ = 32.78 µM; SI = 2.73). For (B), (C) and (D), virus titres were represented in bars and shown on the left‐hand y‐axis, whereas cell viabilities were presented in lines and illustrated on the right‐hand y‐axis. In SH‐SY5Y cells, potential pan‐enterovirus spectrum of compound 13 was investigated against (E) CV‐A6 (EC₅₀ = 0.51 µM; SI = 21.55) or (F) CV‐A16 (EC₅₀ = 0.80 µM; SI = 13.74). Every assay was performed in triplicates with error bars representing the standard deviation from the mean of triplicates. 0.1% DMSO served as the mock control group of respective assays.

Figure 6

Temporal‐based evaluation of compound 13's antiviral activity against EV‐A71. (A) Time‐of‐addition (TOA) and time‐of‐removal (TOR) of compound 13 (8 μM) at specific time points across the EV‐A71 replication cycle in SH‐SY5Y cells. For TOA, virus‐containing media was aspirated and replaced with compound 13 (8 μM) at 0, 2, 4, 6, 8, 10, 12 or 18 h postinfection (h.p.i.). For TOR, infected SH‐SY5Y cells were first incubated with compound 13 (8 μM), and media was decanted and replaced with fresh media at similar time points as TOA. At 24 h.p.i., supernatant from each treatment and control group was collected for virus quantification via plaque assay. Separately, SH‐SY5Y cells were infected with EV‐A71 (MOI 1) and treated with specific concentrations of compound 13 according to previously published experimental schemes for (B) pre‐treatment, (C) co‐treatment and (D) entry‐bypass assays [27]. At 24 h.p.i., supernatant was collected and used for virus yield measurement via plaque assay. Every assay was performed in triplicates with error bars representing standard deviation from the mean of triplicates. 0.1% DMSO served as the mock control group of respective assays.

Figure 7

Evaluation of compound 13's inhibitory effects against EV‐A71 RNA transcription and translation. Lysates of EV‐A71 (MOI 1)‐infected SH‐SY5Y cells treated with specific concentrations of compound 13 were used for (A) qRT‐PCR‐based viral strand‐specific RNA quantification or (B) immunoblotting measurement of specific protein levels. (C) Replicon‐based EV‐A71 transcription and translation machinery analyses in the presence or absence of compound 13 treatment. Guanidine hydrochloride (GuHCl), an eukaryotic RNA transcription inhibitor [38], and cycloheximide (CHX), an inhibitor of the elongation phase during RNA translation [39], were used as positive controls of respective assays. 0.1% DMSO served as the mock control group of respective assays. Luminescence readings generated by SH‐SY5Y cells transfected with either the established replication‐competent or replication‐defective replicons [27] were measured using a microplate reader and normalized against the DMSO control. Error bars represent the standard deviation from the mean of triplicates.