Search

Ribosomal frameshifting (RFS) at the slippery site of SARS-CoV-2 RNA is essential for the biosynthesis of the viral replication machinery. It requires the formation of a pseudoknot (PK) structure near the slippery site and can be inhibited by PK-disrupting oligonucleotide-based antivirals. We obtained and compared three types of such antiviral candidates, namely locked nucleic acids (LNA), LNA-DNA gapmers, and G-clamp-containing phosphorothioates (CPSs) complementary to PK stems. Using optical and electrophoretic methods, we showed that stem 2-targeting oligonucleotide analogs induced PK unfolding at nanomolar concentrations, and this effect was particularly pronounced in the case of LNA. For the leading PK-unfolding LNA and CPS oligonucleotide analogs, we also demonstrated dose-dependent RSF inhibition in dual luciferase assays (DLAs). Finally, we showed that the leading oligonucleotide analogs reduced SARS-CoV-2 replication at subtoxic concentrations in the nanomolar range in two human cell lines. Our findings highlight the promise of PK targeting, illustrate the advantages and limitations of various types of DNA modifications and may promote the future development of oligonucleotide-based antivirals.

Emerging and re-emerging viruses periodically cause outbreaks and epidemics around the world, which ultimately lead to global events such as the COVID-19 pandemic. Thus, the urgent need for new antiviral drugs is obvious. Over more than a century of antiviral development, nucleoside analogs have proven to be promising agents against diversified DNA and RNA viruses. Here, we present the synthesis and evaluation of the antiviral activity of nucleoside analogs and their deglycosylated derivatives based on a hydroxybenzo[4,5]imidazo[1,2-c]pyrimidin-1(2H)-one scaffold. The antiviral activity was evaluated against a panel of structurally and phylogenetically diverse RNA and DNA viruses. The leader compound showed micromolar activity against representatives of the family Coronaviridae , including SARS-CoV-2, as well as against respiratory syncytial virus in a submicromolar range without noticeable toxicity for the host cells. Surprisingly, methylation of the aromatic hydroxyl group of the leader compound resulted in micromolar activity against the varicella-zoster virus without any significant impact on cell viability. The leader compound was shown to be a weak inhibitor of the SARS-CoV-2 RNA-dependent RNA polymerase. It also inhibited biocondensate formation important for SARS-CoV-2 replication. The active compounds may be considered as a good starting point for further structure optimization and mechanistic and preclinical studies.

The life cycle of severe acute respiratory syndrome coronavirus 2 includes several steps that are supposedly mediated by liquid-liquid phase separation (LLPS) of the viral nucleocapsid protein (N) and genomic RNA. To facilitate the rational design of LLPS-targeting therapeutics, we modeled N-RNA biomolecular condensates in vitro and analyzed their sensitivity to several small-molecule antivirals. The model condensates were obtained and visualized under physiological conditions using an optimized RNA sequence enriched with N-binding motifs. The antivirals were selected based on their presumed ability to compete with RNA for specific N sites or interfere with non-specific pi-pi/cation-pi interactions. The set of antivirals included fleximers, 5'-norcarbocyclic nucleoside analogs, and perylene-harboring nucleoside analogs as well as non-nucleoside amphiphilic and hydrophobic perylene derivatives. Most of these antivirals enhanced the formation of N-RNA condensates. Hydrophobic perylene derivatives and 5'-norcarbocyclic derivatives caused up to 50-fold and 15-fold enhancement, respectively. Molecular modeling data argue that hydrophobic compounds do not hamper specific N-RNA interactions and may promote non-specific ones. These findings shed light on the determinants of potent small-molecule modulators of viral LLPS.