Safe, high-throughput screening of natural compounds of MERS-CoV entry inhibitors using a pseudovirus expressing MERS-CoV spike protein

Sir,

Middle East respiratory syndrome coronavirus (MERS-CoV) belongs to the Coronaviridae family of enveloped, positive-sense, single-stranded RNA viruses [1]. MERS-CoV is mainly endemic in the Middle East but can spread outside this region, as shown by the 2015 outbreak in South Korea in which there were 38 deaths [2]. MERS-CoV remains a threat, with 250 cases and 77 deaths reported in 2017; however, there are no effective antiviral reagents.

The spike (S) protein of MERS-CoV is the main determinant of virus entry as it plays an important role in both dipeptidyl peptidase 4 receptor binding and virus-cell membrane fusion [3]. Here, we produced a pseudovirus expressing the S protein of MERS-CoV (MERS-PV) and firefly luciferase. Using MERS-PV, we safely screened 502 compounds derived from natural products for their ability to block MERS-CoV entry in a BSL-2 laboratory. We then verified their antiviral activity against authentic MERS-CoV in a BSL-3 laboratory using an isolate from a patient infected with MERS during the 2015 Korean outbreak.

We used the MERS-PV assay to screen for compounds showing ≥50% inhibition to identify those with entry-blocking effects. We then conducted a secondary screen for compounds showing <20% inhibition of pseudotyped vesicular stomatitis virus (VSV-PV) to verify the specificity of the effect. Cell viability assays were performed to exclude cytotoxic compounds and those with >90% viability were retained (Supplementary methods). The effects were reconfirmed at five concentrations (0.25–4 µg/mL). Three compounds, dihydrotanshinone, E-64-C, and E-64-D, met the screening criteria at a low concentration (1 µg/mL; Fig. 1 ).

Fig. 1.

Dose-response confirmation of the three selected natural compounds. The relative infectivities of MERS-PV (filled circles) and VSV-PV (filled squares) in the presence of the three hit compounds (a) dihydrotanshinone, (b) E-64-C, and (c) E-64-D were dose-dependently evaluated (at 0.25 to 4 µg/mL). The viability of cells (filled triangles) in the presence of the hit compounds at the same concentrations was also determined. The experiment was repeated three times, and data are presented as mean (%) and standard deviation. One-way ANOVA with Tukey's post-test was used to compare the variables. *P < 0.05; **P < 0.01; ***P < 0.001; ns, not significant.

We then evaluated the inhibitory mechanisms of the hit compounds (at 1 µg/mL) in pre- and post-attachment assays. In the pre-attachment assay, the compounds were incubated with MERS-PV for 2 h, then mixed with host cells. In the post-attachment assay, MERS-PV was mixed with host cells, and the compounds were added 2 h later. While the relative infectivity of MERS-PV treated with hit compounds was approximately 80% in the post-attachment assay, the relative infectivity in the pre-attachment and during attachment assays was significantly less (≤50%; Supplementary Fig. 1), which indicates that the hit compounds may block virus entry. However, this assay could not evaluate post-attachment events, such as intracellular processes and viral RNA replication.

The three selected compounds were also evaluated for inhibitory effects on MERS-CoV to assess their potential as antiviral agents in pre- and post-attachment assays in a BSL-3 laboratory at Chungbuk National University, as approved by the Korean Centers for Disease Control and Prevention (KCDC-14-3-07). Although E-64-C did not show a consistent reduction in infection in the pre-attachment assay, dihydrotanshinone and E-64-D showed antiviral activities in the pre-attachment assay (dihydrotanshinone, 6.5 to 5.5 Log TCID₅₀/mL at 2 µg/mL; E-64-D, 6.5 to 5.1 Log TCID₅₀/mL at 0.5 µg/mL; Supplementary Fig. 2a). E-64-D irreversibly and selectively inhibits cysteine proteases, such as papain, calpain, and cathepsin, by modifying the thiol group to the thioester form [4]. The critical factors for MERS-CoV entry include host cell proteases, such as the pH-dependent endosomal cysteine proteases cathepsin B and L [3]. We showed that E-64-D reduced the titer in the pre-attachment assay, which could be explained by inhibition of viral entry; hence, the antiviral effect might be associated with cathepsin inhibition.

In the post-attachment assay, E-64-C and E-64-D did not show antiviral effects on MERS-CoV at <4 µg/mL. Interestingly, dihydrotanshinone showed a decimal reduction at 0.5 µg/mL, and excellent antiviral effects at ≥2 µg/mL, with a reduction in titer from 6.5 Log to 1.8 Log TCID₅₀/mL (Supplementary Fig. 2B). Dihydrotanshinone is a major lipophilic compound extracted from the root of Salvia miltiorrhiza Bunge (known as Dansam in Korean) that is commonly used in traditional Asian medicine. Tanshinones derived from S. miltiorrhiza showed specific inhibitory activity against SARS-CoV 3CL^pro and PL^pro proteases, which are coronavirus cysteine proteases involved in initiation of viral replication [5]. Therefore, the inhibitory effect of dihydrotanshinone in the post-attachment assay could be associated with inhibition of coronavirus replication. This study demonstrated that dihydrotanshinone had inhibitory effects against viral entry in the MERS-CoV and MERS-PV assays. Therefore, dihydrotanshinone may have dual inhibitory effects, blocking virus entry and replication in association with viral proteases. However, further studies, including mechanistic analyses of dihydrotanshinone, are needed.

In conclusion, the results of this study indicate that a high-throughput assay using MERS-PV is useful for screening antiviral compounds against MERS-CoV. Dihydrotanshinone warrants further studies as a candidate for prophylaxis or treatment of MERS-CoV infection.