Epidemics caused by various viral infections, such as AIDS, Zika fever, Dengue fever, or Ebola, are a constant threat to communities of all sizes . The COVID-19 pandemic, caused by the newly emerged severe acute respiratory syndrome coronavirus type 2 (SARS-CoV-2), has put an enormous pressure on the healthcare system worldwide and called for immediate action in prevention and treatment, which in turn required the discovery of new effective therapeutic options. It seems to be clear that the widespread use of vaccines is able to stop the acute phase of the pandemic. However, antiviral therapy for COVID-19 is indispensable in case of vaccine failure, virus mutation or suppressed immunity of some patients .

SARS-CoV-2 is part of the Coronaviridae family, a group of enveloped +ssRNA viruses. The genome can directly act as a viral messenger RNA and encodes essential enzymes for replication . Inhibiting these nonstructural proteins that are part of the replication complex has already shown great success in antiviral therapy .

The viral RNA-dependent RNA polymerase (RdRp) is encoded in all RNA viruses and plays a crucial role in viral RNA replication. In the proteome of SARS-CoV-2, the catalytic subunit nsp12, expressed together with the cofactors nsp7 and nsp8, constitutes the RdRp . RdRp is usually targeted by nucleotide analogue inhibitors (NAIs) . This class of antivirals can inhibit the replication by acting as a delayed chain terminator or by causing genetic corruption in the viral RNA and includes the first FDA-approved antiviral drugs in the therapy of COVID-19 patients, remdesivir and molnupiravir . The usability of NAIs may largely depend on the metabolic activation, and they also compete with the intracellular pool of natural nucleoside triphosphates (NTPs). Nonnucleotide analogue inhibitors (NNAIs) do not face these challenges as they bind to both active but also allosteric sites of the RdRp, and therefore they represent a promising NAI alternative .

Since the beginning of the pandemic, a variety of heterocyclic small molecules – either of natural or synthetic origin – was reported as promising inhibitors of the SARS-CoV-2 RdRp . However, compounds with a sufficient combination of high potency and suitable pharmacokinetic properties are still scarce. Recently, many studies have been focusing on drug repurposing or screening libraries of already approved biologically active compounds . This approach might represent a very promising strategy in the case of targeting the coronaviral RdRp due to the highly conserved structure of the polymerase, not only across the CoV group but also in other RNA viruses . A great example of this phenomenon is remdesivir, which was originally developed as a therapeutic agent against Ebola virus .

HeE1-2Tyr (1) was originally identified by Tarantino et al. as a potent inhibitor of RdRp from all members of the genus Orthoflavivirus and was crystallized in complex with the RdRp from DENV-3 . In 2021, our group reported this compound to also exhibit inhibitory activity against feline infectious peritonitis virus (FIPV) and SARS-CoV-2 RdRp and to hinder viral replication in cell-based antiviral assays . That study highlighted the beneficial role of the tyrosine residue and the indispensable role of the C-2 substitution.

Figure 1: Structure of HeE1-2Tyr (1) and of the derivatives synthesized in this work.

In this work, replacing the sulfur atom with a (bio)isosteric oxygen atom yielded two novel structural analogues, whilst our effort towards more simple molecules led to a series of pyridone derivatives. Out of these, 3a had already been synthesized by a different approach . However, this study presents a novel and notably simpler synthetic route. Furthermore, as part of the systematic truncation of the core, we synthesized thiazolopyridone and thiadiazolopyridone derivatives because molecules based on these cores have already shown promising antimicrobial activities .

Scheme 1: Synthetic pathway to benzoxazole analogue 2 of HeE1-2Tyr (1). Reagents and conditions: a) TMSN3, TfOH, DCM, rt, overnight, b) BnBr, NaH, DMF, 0 °C to rt, 1 h, c) (R = Et) diethyl carbonate, LiHMDS, THF, −78 to 0 °C, 1 h, d) (R = allyl) allyl chloroformate, LiHMDS, THF, −78 to 0 °C, 1 h, e) POCl3, DMF, 90 °C, 30 min, f) (BnCO)2O, 100 °C, 1.5 h, g) methanesulfonic acid, DCM, 0 °C to rt, 4 h, h) CyOH, PPh3, DIAD, 1,4-dioxane, 50 °C, overnight, i) (R = Et) NaOH, H2O, EtOH, 75 °C, 3 h, j) (R = allyl) Et3SiH, PPh3, Pd(OAc)2, ACN, rt, k) H–ʟ-Tyr–OMe, HOBt, EDCI, TEA, DCM, DMF, rt, 12 h, l) LiOH⋅H2O, H2O, 1,4-dioxane, rt, 45 min. Cy = cyclohexyl.

Scheme 2: Synthetic pathway to pyridone derivatives 3a–c of HeE1-2Tyr (1). Reagents and conditions: a) MeI, K2CO3, DMF, rt, 2.5 h, b) LiOH⋅H2O, H2O, 1,4-dioxane, rt, 15 min, c) H–ʟ-Tyr–OMe, HOBt, EDCI, TEA, DCM, DMF, rt, 12 h, d) ArB(OH)2, Pd(dppf)Cl2⋅CH2Cl2, Cs2CO3, DMF, H2O, 80 °C, overnight.

Scheme 3: Synthetic pathway to thiazolopyridone derivatives 4a,b of HeE1-2Tyr (1). Reagents and conditions: a) 2-bromo-2-phenylacetyl chloride, TEA, THF, rt, 1 h, b) ethyl acrylate, toluene, 110 °C, 24 h, c) NaOH, H2O, MeOH, 70 °C, 2 h, d) H–ʟ-Tyr–OMe, HOBt, EDCI, TEA, DCM, DMF, rt, 12 h, e) LiOH⋅H2O, H2O, 1,4-dioxane, rt, 2 h.

Figure 2: Analysis of inhibitory activity against SARS-CoV-2 RdRp using primer extension assay. A) Gel-based polymerase assay with a constant concentration of fluorescently labeled template/primer RNA (0.5 µM) and the polymerase complex (nsp7, nsp8 3 µM and nsp12 1 µM), along with increasing concentrations of compounds as indicated at the top. Reactions were initiated by adding 10 µM NTPs and run for 1 h at 30 °C. The reactions were stopped by adding stop buffer, and the products were separated on a 20% denaturing gel. B) Graphical representation of the inhibitory activity of selected compounds evaluated from the gels obtained in the primer extension assay. The percentage of inhibition (against control) was plotted against the logarithm of the concentration of compounds. The results were fitted to sigmoidal dose–response curves. C) The IC50 values were determined using the GraphPad algorithm (IC50 value of compound 1 was published by Dejmek et al. ), the BEI was calculated using the function pIC50 [mol/L]/MW [kDa].

In this study, novel analogues of the antiviral HeE1-2Tyr (1) were synthesized and evaluated with respect to the in vitro inhibitory activity towards SARS-CoV-2 RdRp. To obtain the benzoxazole analogue, a new synthetic strategy avoiding base-mediated hydrolysis was successfully applied. For the simplified structural derivatives, the applied routes were optimized for maximal efficacy of the synthetic work. Regarding the inhibitory activity, six of the novel compounds showed inhibition in the fluorescence-based primer extension assay. The two simplified molecules were the most promising inhibitors, with an IC50 value below 90 µM, and the compounds 3a–c and 4a,b exerted stronger BEIs than 1. The obtained results provide important information about the structural requirements for the heterocyclic inhibitors based on HeE1-2Tyr (1), which we will use in the design of further generations of such antivirals.