The design and synthesis of antiviral compounds were guided by computationally predicted data against highly conserved non-structural proteins (Nsps) of the SARS-CoV-2 coronavirus. Chromenephenylmethanone-1 (CPM-1), a novel biphenylpyran (BPP), was selected from a unique natural product library based on in silico docking scores to coronavirus Nsps with high specificity to the methyltransferase protein (2'-O-MTase, Nsp10-16), which is responsible for viral mRNA maturation and host innate immune response evasion. To target the 2'-O-MTase, CPM-1, along with intermediate BPP regioisomers, tetrahydrophenylmethanones (TPMs), were synthesized and structurally validated via nuclear magnetic resonance (NMR) data and DP4+ structure probability analyses. To investigate the activity of these BPPs, the following in vitro assays were conducted: SARS-CoV-2 inhibition, biochemical target validation, mutagenicity, and cytotoxicity. CPM-1 possessed notable activity against SARS-CoV-2 with 98.9% inhibition at 10 µM and an EC50 of 7.65 µM, as well as inhibition of SARS-CoV-2's 2'-O-MTase (expressed and purified) with an IC50 of 1.5 ± 0.2 µM. In addition, CPM-1 revealed no cytotoxicity (CC50 of >100 µM) or mutagenicity (no frameshift or base-pair mutations). This study demonstrates the potential of computational modeling for the discovery of natural product prototypes followed by the design and synthesis of drug leads to inhibit the SARS-CoV-2 2'-O-MTase.
Keywords: 2′-O-MTase; DP4+ NMR analyses; Nsp; SARS-CoV-2; cannabinoid; computational screen; coronavirus; in silico; methyltransferase; natural products; non-structural protein.